Advances in Clinical Chemistry Volume 12

ADVANCES IN CLINICAL CHEMISTRY VOLUME 12 This Page Intentionally Left Blank Advances in CLINICAL CHEMISTRY Edited ...

Author: Oscar Bodansky | Corbet P. Stewart

12 downloads 1157 Views 25MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

ADVANCES IN CLINICAL CHEMISTRY VOLUME 12

This Page Intentionally Left Blank

Advances in

CLINICAL CHEMISTRY Edited by

OSCAR BODANSKY Division of Biochemistry, Sloan-Kettering Institute for Cancer Research, N e w York, N e w York

C. P. STEWART Formerly Deportment of Edinburgh,

Clinical

Royal

Chemistry,

Infirmary,

VOLUME 12

University of

Edinburgh,

0

Scotland

1969

A C A D E M I C PRESS NEW Y O R K A N D LONDON

COPYRIQHT @ 1969, BY ACADEMIC WESS,INC. ALL RIQHTS RESERVED

N O PART O F THIS BOOK MAY BE REPRODUCED IN ANY FORM,

BY PHOTOSTAT, MICROFILM, RETRIEVAL SYSTEM, OR ANY OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM THE PUBLISHERS.

ACADEMIC PRESS, INC. 111 Fifth Avenue, New York, New York 10003

United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. Berkeley Square House, London W l X 6BA

LIBRARY O F CONQRESS C A T A L W CARD

PRINTED

NUMBER: 58-12341

IN THE UNITED STATES OF AMERICA

LIST

OF CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors’ contributions begin.

IANE. BUSH (57), Department of Physiology, Medical School of Virginia, Richmond, Virginia SIR DAVIDCUTHBERTSON (1) , Department of Pathological Biochemistry, Glasgow University and Royal Infirmary, Glasgow, Scotland

PAULAJABLONSKI (309), Department of Biochemistry, Alfred Hospital, M e 1bourne, Australia

ROBERTD. LEEPER(387), Endocrine Section, Memorial Hospital for Cancer and Allied Diseases, Sloan-Kettering Institute for Cancer Research, New York, New York FREDERICK L. MITCHELL (141), Division of Clinical Chemistry, Medical Researoh Council Clinical Research Centre, Northwick Park, Harrow, Middlesex, England

J. A. OWEN (309), Department of Surgery, Monash University, Alfred Hospital, Melbourne, Australia CEDBICH. L. SHACKLETON (141), Clinical Endocrinology Unit, Edinburgh, Scotland

HAROLD V. S ~ E (217), T D e p a r h e n t of Forensic Medicine, University of Edinburgh, Edinburgh, Scotland

W. J. TILSMNE ( l ) ,Department of Pathological Biochemistry, Glasgow University and Royal Infirmary, Glasgow, Scotland

V

This Page Intentionally Left Blank

PREFACE Since its inception in 1958 the serial publication Advances in Clinical Chemistry has had a dual purpose: the description of reliable diagnostic and prognostic procedures and the elucidation of the fundamental biochemical abnormalities that underlie disease. As with so many other branches of science, both of these aspects have undergone remarkable developments during the past dozen years, and the Editors have striven to have the Advances reflect these developments. New methods for diagnosis and prognosis of disease no longer consist of relatively simple, discrete procedures, but frequently comprise the transposition of a whole research technology to the everyday needs of the clinical chemistry laboratory. Such transposition requires a careful and critical appraisal of the feasibility of the technology in its altered setting of such factors as accuracy, precision, sensitivity, and specificity, and, if possible, of automation. These features are exemplified in the present volume by Street’s review on The Use of Gas-Liquid Chromatography in Clinical Chemistry and by Bush’s review on Determination of Estrogens, Androgens, Progesterone, and Related Steroids in Human Plasma and Urine. Studies on the biochemical mechanisms that underlie disease are represented in the present volume by several reviews. T h a t by Cuthbertson and Tilstone spans forty years of investigation by the senior author in a field that is of considerable current interest. The review on The Investigation of Steroid Metabolism in Early Infancy critically summarizes a vast amount of recent work that has been made possible not only by the availability of new methods but by the ingenious utilization of biological materials obtainable from the infant a t birth and in the neonatal period. As in the past, it is a great pleasure to thank our contributors and publisher for their excellent cooperation in making this volume possible. September, 1969 OSCARBODANSKY

c. P. &ZWART* *Present addres*y: 17 Orchard Road South, Edinburgh, Scotland. vii

This Page Intentionally Left Blank

CONTE NTS . .

. .

. .

. .

. .

. .

. .

. .

.

CONTENTS OF PREVIOUS VOLUMES .

.

.

.

.

.

.

.

.

.

.

xi

. .

. .

. .

1 2

LIST OF CONTRIBUTORS PREFACE

.

.

.

.

. .

. .

.

.

V

vii

Metabolism during the Postinjury Period D . P . CUTHBERTSON A N D W J . TILSTONE

.

1. 2. 3. 4.

Introduction . . . . . . . . . . The Nature of the Immediate Changes . . . Early Postinjury Changes . . . . . . Nature of the PostShock Delayed Metabolic Response 5 . Time Factor in Multiple Injuries . . . . . 6 Nutritional Aspects . . . . . . . . 7. Environmental Factors . . . . . . . 8. Control of the Metabolic Response to Injury . . 9. Summary . . . . . . . . . . References . . . . . . . . . . . Note Added in Proof . . . . . . . . .

.

. .

. .

.

.

.

.

.

3

.

.

.

.

.

. .

. .

. .

. .

. .

19 23

.

.

.

.

.

. . .

. . .

. . .

. . .

. . .

.

.

.

.

.

23 26 32 41 42 55

Determination of Estrogens. Androgens. Progesterone. and Related Steroids in Human Plasma and Urine IANE. BUSH

1. Introduction

.

.

.

.

2 . The Assessment of Methods

. . . . . . . . . . . . . . . . . . . . . .

3 . Advances in General Techniques 4. Specific Examples . . . 5 . The Future . . . . . References . . . . . .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. . . .

. . . .

. . . .

. . . .

. . . .

57 83 111 120 129 130

The Investigation of Steroid Metabolism in Early Infancy FREDERICK L . MITCHELLAND CEDRICH . L SHACKLETON

.

1. 2. 3. 4. 5. 6.

7.

Introduction . . . . . . . . . . . . . . Nomenclature . . . . . . . . . . . . . Methodology . . . . . . . . . . . . . . The Influence of Steroid Metabolic Pathways Used in Utero . . Steroid Assays on Umbilical Cord Blood . . . . . . . 17-0x0 Steroids in Infant Blood and Urine . . . . . . . The Urinary Excretion, Blood Levels. and Production of Ca Steroids

ix

. .

.

. . .

.

142 143 143 156 170 174 177

X

CONTENTS

8. The Urinary Excretion and Blood Levels of 3p-Hydr0~y-A~ Steroids . 9. The Urinary Excretion of Estrogens . . . . . . . . . 10. Steroid Conjugation . . . . . . . . . . . . . 11 Abnormalities in Steroid Production . . . . . . . . . 12 The Control of Steroid Production in Infancy . . . . . . . 13. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References

. .

181 186 186 189 197 200 201

The Use of Gas-liquid Chromatography in Clinical Chemistry

HAROLD V . STREET

1. Introduction . . . . . . . . . . 2 . Basic Requirements of Gas-Liquid Chromatography 3. Column Preparation in Gas-Liquid Chromatography 4. Applications of GtLs-Liquid Chromatography . . References . . . . . . . . . . .

.

.

.

.

.

. .

. .

. .

. .

. .

.

.

.

.

.

.

.

.

.

.

217 218 221 237 297

The Clinical Chemistry of Bromsulfophthalein and Other Cholephilic Dyes PAWLA JABLONSKI A N D J . A . OWEN

1. 2. 3. 4.

Introduction . . . . . . . . . . . . Historical . . . . . . . . . . . . Nature of Cholephilic Dyes . . . . . . . . . . . . . Transfer of BSP from Plasma to Bile . 5 . Dye Tests Used in Clinical Investigations . . . . . 6 . ElTects of Physiological Factors on Dye Uptake and Secretion 7. Effects of Drugs on Dye Uptake and Excretion . . . 8 Effect of Disease on Dye Uptake and Excretion . . . 9. Analytical Methods . . . . . . . . . . 10. The Future of Dye Tests in Clinical Medicine . . . . Review Articles . . . . . . . . . . . . . . . . . . . . . . . . . References

.

.

.

.

310

.

.

.

.

.

.

314 317 327 342 344 352 357 361 363 364

.

.

. 311

.

.

.

.

.

.

.

.

.

. . .

. . . .

. . . . . . . .

Recent Advances in the Biochemistry of Thyroid Regulation ROBERT D . LEEPER

1. Introduction . . . . . 2. Thyrotropin-Releasing Factor . 3. Thyroid-Stimulating Hormones 4. Thyroid-Binding Proteins . . 5 . Calcitonin . . . . . References . . . . . . AUTHORINDEX

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. . . . . . . . . . . . . . .

SUBJECTINDEX .

.

.

.

.

.

.

.

.

.

.

.

.

.

387 390 393 404 411 414 425

.

459

CONTENTS

OF PREVIOUS VOLUMES

Volume 1

Plasma Iron W . N . M . Ramsay The Assessment of the Tubular Function of the Kidneys Bertil Josephson and Jan Ek Protein-Bound Iodine Albert L. Chaney Blood Plasma Levels of Radioactive Iodine-131 in the Diagnosis of Hyperthyroidism Solomon Silver Determination of Individual Adrenocortical Steroids R . Neher The 5-Hydroxyindoles C. E . Dalgliesh Paper Electrophoresis of Proteins and Protein-Bound Substances in Clinical Investigations J . A. Owen Composition of the Body Fluids in Childhood Bertil Josephson The Clinical Significance of Alterations in Transaminase Activities of Serum and Other Body Fluids Felk Wro'blewski Author Index-Subject

Index

Volume 2

Paper Electrophoresis : Principles and Techniques H . Peeters Blood Ammonia Samuel P . Bessman Idiopathic Hypercalcemia of Infancy John 0. Forfar and S. L. Tompsett xi

xii

CONTENTS O F PREVIOUS VOLUMES

Amino Aciduria E . J . Bigwood, R. Crokaert, E . Schram, P . Soupart, and H . ViS Bile Pigments in Jaundice Barbara H . Billing Automation Walton H . Marsh Author Index-Subject Index Volume 3

Infrared Absorption Analysis of Tissue Constituents, Particularly Tissue Lipids Henrp P . Schwarz The Chemical Basis of Kernicterus Irwin M . Arias Flocculation Tests and Their Application to the Study of Liver Disease John G . Reinhold The Determination and Significance of the Natural Estrogens J . B. Brown Folic Acid, Its Analogs and Antagonists Ronald H . Girdwood Physiology and Pathology of Vitamin B,, Absorption, Distribution, and Excretion Ralph Grasbeck Author Index-Subject Index Volume 4

Flame Photometry I . MacIntyre The Nonglucose Melliturias James B. Sidbury, Jr. Organic Acids in Blood and Urine Jo Nordmann and Roger Nordmann Ascorbic Acid in Man and Animals W . Eugene Knox and M . N . D.Goswami

...

CONTENTS OF PREVIOUS VOLUMES

Xlll

Immunoelectrophoresis: Methods, Interpretation, Results C . Wunderly Biochemical Aspects of Parathyroid Function and of Hyperparathyroidism B . E . C . Nordin Ultramicro Methods P. Reinouts van Haga and J . de Wael Author Index-Subject

Index

Volume 5

Inherited Metabolic Disorders : Galactosemia L. I . Woolf The Malabsorption Syndrome, with Special Reference to the Effects of Wheat Gluten A . C . Frazer Peptides in Human Urine B . Skarbyfiski and M . Sarnecka-Keller Haptoglobins C.-B. Laurel1 and C . Gronvall Microbiological Assay Methods for Vitamins Herman Baker and Harry Sobotka Dehydrogenases: Glucose-6-Phosphate Dehydrogenase, 6-Phosphogluconate Dehydrogenase, Glutathione Reductase, Methemoglobin Reductase, Polyol Dehydrogenase F. H . Bruss and P . H . Werners Author Index-Subj ect Index-Index lative Topical Index-Vols. 1-5

of Contributors-Vols. 1-5-Cumu-

Volume 6

Micromethods for Measuring Acid-Base Values of Blood Poul Astrup and 0. Siggaard-Andersen Magnesium C . P . Stewart and S. C . Frazer

xiv

CONTENTS O F PREVIOUS VOLUMES

Enzymatic Determinations of Glucose Alfred H . Free Inherited Metabolic Disorders: Errors of Phenylalanine and Tyrosine Metabolism L. I . Woolf Normal and Abnormal Human Hemoglobins Titus H . J . Huisman Author Index-Subject

Index

Volume 7

Principles and Applications of Atomic Absorption Spectroscopy Alfred Zettner Aspects of Disorders of the Kynurenine Pathway of Tryptophan Metabolism in Man Luigi Musajo and Carlo A . Benassi The Clinical Biochemistry of the Muscular Dystrophies W . H . S. Thomscm Mucopolysaccharides in Disease J . S. Brimawmbe and M . Stacey Proteins, Mucosubstances, and Biologically Active Components of Gastric Secretion George B. Jerzy Glass Fractionation of Macromolecular Components of Human Gastric Juice by Electrophoresis, Chromatography, and Other Physicochemical Methods George B . Jerzy Glass Author Index-Subj ect Index

Volume 8

Copper Metabolism Andrew Sass-Kortsak Hyperbaric Oxygenation Sheldon F. Gottlieb

CONTENTS O F PREVIOUS VOLUMES

XV

Determination of Hemoglobin and Its Derivatives E . J . van Kampen and W . G. Zijlstra Blood-Coagulation Factor VIII : Genetics, Physiological Control, and Bioassay G . I . C. Ingram Albumin and “Total Globulin” Fractions of Blood Derek Watson Author I n d e x 4 u b j e c t Index

Volume 9

Effect of Injury on Plasma Proteins J . A . Owen Progress and Problems in the Immunodiagnosis of Helminthic Infections Everett L. Schiller Isoenzymes A . L. Latner Abnormalities in the Metabolism of Sulfur-Containing Amino Acids Stanley Berlow Blood Hydrogen Ion : Terminology, Physiology, and Clinical Applications T . P. Whitehead Laboratory Diagnosis of Glycogen Diseases Kurt Steinit z Author Index-Subject

Index

Volume 10

Calcitonin and Thyrocalcitonin David Webster and Samuel C. Frazer Automated Techniques in Lipid Chemistry Gerald Kessler Quality Control in Routine Clinical Chemistry L. G . Whitby, F. L. Mitchell, and D. W . Moss

xvi

CONTENTS O F PREVIOUS VOLU1\.IES

Metabolism of Oxypurines in Man M . Earl Balis The Technique and Significance of Hydroxyproline Measurement in Man E . Carwile LeRoy lsoenzymes of Human Alkaline Phosphatase William H . Fishman and Nimai K . Ghosh Author Index-Subject

Index

Volume 11

Enzymatic Defects in the Sphingolipidoses Roscoe 0. Brady Genetically Determined Polymorphisms of Erythrocyte Enzymes in Man D . A. Hopkinson Biochemistry of Functional Neural Crest Tumors Leiv R . Gjessing Biochemical and Clinical Aspects of the Porphyrias Richard D . Levere and Attallah Kappas Premortal Clinical Biochemical Changes John E s ben Kirk Intracellular pH J. S. Robson, J . M . Bone, and Anne T . Lambie 5’-Nucleotidase Oscar Bodansky and Morton K . Schwartz Author Index-Subj ect Index-Cumulative

Topical Index-Vols.

1-1 1

METABOLISM DURING THE POSTINJURY PERIOD Sir David Cuthbertson and W. J. Tilstone Department of Pathological Biochemistry, Glasgow University and Royal Infirmary, Glasgow, Scotland

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. The Nature of the Immediate Changes.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Vascular.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Hormonal.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Early Postinjury Changes.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Renal Changes.. . ................................ 3.2. Early Biochemical .................... 3.3. Hemoglobin and Effect of Transfusion.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Hypercoagulability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. PlasmaProteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6. Reticuloendothelial System (RE) in Response to Different Types of Injury.. . . . . . . . . . . . . . ........... 4. Nature of the Post-shock Delayed Me ................. 4.1. Significance of Increased Catabolites after Injury. . . . . . . . . . . . . . . . . 5. Time Factor in Multiple Injuries.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Nutritional Aspects. . . . . . . . . . . . . . . . . . . . . ................... 6.1. Effects of Diminished Food Intake o .. 6.2. Transfusion and Nourishment of the Injured.. ...................... 7. Environmental Factors.. . . .... ............... 7.1. Introduction. . . . . . . . .... 7.2. Environmental Stress. . . . . . . . . .............. 7.3. Interaction of Environment and Stress.. . .............. 8. Control of the Metabolic Response to 8.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......................... 8.2. The Endocrine Response to Injury. 8.3. The Permissive Role of Cortisol. . . . . . . . . . . . . . . 8.4. Initiating Pathways.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. ........ .............. .............. Note Added in Proof.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.

1 2

4

13 15 16

23 23 23 25 26 26 28

32 34 39

Introduction

The nature and severity of an injury, whether arising by accident, necrotizing agent, radiation, or by elective surgery, condition the character of both the immediate and subsequent biochemical changes, that occur not only locally but also generally. The subject’s nutritive state a t the time, particularly in respect to labile protein reserves and glycogen, also affects the scale of these changes, as does restitution of the lost blood or plasma. Finally, the environmental temperature a t the time and during 1

2

D. P. CUTHBERTSON AND W. J. TILSTONE

the days immediately following the injury or operation likewise conditions the issue: the relative humidity may be linked with this. Although these changes have been spasmodically studied for many years and as early as 1794 John Hunter (H18) had realized that there is a circumstance concerning accidental injury which differs from disease, namely, that the injury done produces not only the disposition but the means of cure, nevertheless, the nature of these changes and their interpretation still constitute a fundamental and important area of biological research as accidents increase and surgeons become more venturesome (C33). It is very obvious that physical injury results in damage to and loss of body protein locally a t the site of injury, either by physical separation of tissue from the rest of the body or by destruction of it in situ. At this time there is initially a depression in vitality (“ebb phase”) which may be transitory or more manifest and there may be some degree of primary or “neurogenic” shock. The situation may deteriorate, and through loss of blood or plasma oligemia may ensue. For defense and in order to initiate repair of the damaged tissues, the injury is generally followed by an increase in metabolic reactions (“flow phase”) with components-both local and general-which together constitute the inflammatory response and beginning of repair. It is very obvious that many of the consequences of injury imperil, or are potentially capable of imperiling, survival apart altogether from the purely local damage done. Such are hemorrhage, renal failure, or cerebral fat embolism; but many, such as those concerned with preservation of homeostasis are defense mechanisms, eg., the vasoconstriction of skin and muscle which reduces bleeding, and hemodilution which helps to offset blood loss. Of the chemical constituents of the body which are lost or damaged, the proteins of the tissues suffer most, but considerable electrolyte changes take place and the tricarboxylic acid cycle is deranged. Clinically i t is necessary to recognize those effects which should be counteracted and, in particular, the pathological sequelae which need to be prevented or allayed. The concept of irreversible status requires constant revision as knowledge becomes greater and surgery more venturesome. It is convenient to divide the changes into early and late. Together they may be considered to constitute the inflammatory response. 2.

The Nature of the Immediate Changes

2.1.

VASCULAR

I n this phase we are concerned with both the local tissue changes, especially the vascular changes, and the changes in metabolism including

METABOLISM DURING THE POSTIN JURY PERIOD

3

alterations in hormonal activity. These are largely interdependent (C27, C33). A few minutes after injury there is a fleeting arteriolar contraction with coagulation of escaping blood. Then there follows dilatation of blood vessels inducing redness and heat, and in some measure this constitutes the beginning of the defensive pattern of homeostasis involving inflammation and repair. Later there is swelling, pain, and sometimes loss of function. Shortly after these initial changes there occurs an increase in polymorphonuclear leukocytes and platelets but fall in eosinophiles and lymphocytes. After severe injury there is increased coagulability of blood with decreased suspension stability of red cells which tend to clump. 2.2. HORMONAL Local production or liberation of hormones must act as mediators in these reactions. The vascular response to mild injury is biphasic and the initial change is in the small venules (M12) ; later there is increase in capillary permeability. Various local hormones, such as histamine, 5-hydroxytryptamine, and bradykinin are released, and these mediators could be responsible for the venular change, but blocking them does not prevent exudation. Systemically there is increased secretion of adrenaline (epinephrine) (W5), cortisol (J2), aldosterone (Zl), and ACTH (see Section 8.2). Stimulation of the sympathetic system seems to be the primary trigger in setting off a burst of catecholamines and the cortical steroids, the latter mainly mediated through the hypothalamic-pituitary axis, and which bring about much of what constitutes the primary reaction to the emergency. There is also hyperglycemia. Much of this “nervous” stimulation can apparently be damped down by premedication and anesthesia in the case of elective operations, and by fainting. There may be a fall in gastric acid-pepsin secretion, gaseous distension of the proximal colon and decreased gastric mobility. Cerebral hypoxia may lead to confusion. 3.

Early Postinjury Changes

Beginning within a few hours after moderate to severe injury, and generally complete in 24 hours, is a phase of local increased permeability and exudation with consequent swelling as the inflammatory reaction sets in. Polymorphs and monocytes migrate in the exudate which coagulates. There may be thrombosis of vessels in the vicinity of the damage which tends to block movement from and to the zone of damage. Unless the loss of circulatory fluid is rapidly blocked or made good, the patient may pass into a more serious state of shock by lack of adequate cardiac

4

D. P. CUTHBERTSON AND W. J. TILSTONE

output due to diminished return to the right heart. This period of ebbing vitality is associated with a progressive fall in body temperature and oxygen utilization. There may be oligemia or normal volemia during this period depending on blood loss. There may be circulatory strain a t various sites; for example, the gut may sometimes show evidence of transient shedding of the mucosa. There is oliguria and appetite may fail, the liver may be so affected by anoxemia that there is inhibition of deamination, and there may be increased concentration of amino acids in the blood; this will be further considered. Increase in capillary permeability is the second stage of the biphasic permeability response (M12). Systemically, the organism is now in the “ebb” phase of the metabolic response. Plasma sodium and chloride may fall, and there is an accumulation of water, plasma protein, and sodium a t the site of injury. There begins a marked increase of fibrous tissue a t many sites (R3).

3.1. RENALCHANGES Regional vasoconstriction arises from oligemia, and hypotension results when vasoconstriction cannot balance the reduced cardiac output. The outcome is diminished transport of oxygen: oxygen consumption is reduced. This is, however, beneficial, a t least for a time, because survival becomes feasible after an amount of blood loss which would otherwise be fatal (58, S9). It may, if extreme and prolonged, result in renal tubular necrosis and increase of heart load; and experimentally, it has been involved in the pathogenesis of irreversible hemorrhagic shock (L8). Renal vasoconstriction may result in severe degeneration or necrosis of the renal tubules. Acute renal failure may arise and associated with it the condition of “traumatic uremia.” This takes oliguric and nonoliguric forms (S13a) in both of which the azotemia persists for days or weeks, becomes considerable and is not ameliorated by infusion therapy. The underlying change in both is a fall in glomerular filtration rate. The distinguishing feature in patients with acute renal failure is the persistence of a low filtration rate, the cause of which remains obscure. Tubular necrosis may contribute, but it cannot account for the essential change, nor can changes, in total renal flow (M16). Although intrarenal cortical ischemia may be important experimentally (T5), this is uncertain in man. The volume of urine is determined both by the fall in glomerular filtration rate and by the reduced tubular reabsorption of water. The balance determines whether renal failure, if it occurs, is oliguric or nonoliguric. Retention of sodium and water help to conserve body water and

METABOLISM DURING THE POSTIN JURY PFZIOD

5

salt in the face of losses in the injured area and through hemodilution may reduce oligemia. BIOCHEMICAL CHANGES 3.2. EARLY I n general, i t is difficult to determine whether the biochemical changes seen a t this stage represent a primary metabolic change toward homeostasis or are secondary to a decrease in circulating blood volume and in blood pressure. Stoner and his colleagues (S16) have shown that although the biochemical changes in skin and kidney are mainly due to changes in blood supply this is not the case in liver, brain, and uninjured muscle. 3.2.1. Lipemia and Fat Embolism While lipemia occurs after injury it has nothing to do with fat emboli, for it occurs after hemorrhage, burning, or local cold injury, all conditions unassociated with fat embolism. Sevitt points out that there is no evidence that preexisting lipemia in man (e.g., diabetes) predisposes to fat embolism after injury, nor does digestive lipemia in rabbits increase the degree of fat embolism after injury, nor does digestive lipemia in rabbits increase the degree of fat embolism after injury, nor apparently does digestive lipemia in rabbits increase the degree of fat embolism (P3, 58,S9). Sevitt has suggested that traumatic lipemia with rise in free fatty acids in particular, and less markedly of cholesterol, neutral fat, and phospholipids, may be considered as part of the general metabolic response to injury. Pulmonary fat embolism is found a t necropsy in 90100% of patients who have died shortly after fracture-possibly a protective trapping mechanism in that few fat emboli will reach the systemic circulation. 3.2.2. Early Changes in Nitrogen Metabolism A rise in blood nonprotein nonurea nitrogen was observed following severe injury in battle casualties in World War I by Duval and Grigaut (D13), who related the increase to the degree of tissue damage, though Cuthbertson (C19, C22) did not find much elevation in blood levels in fracture cases a t the time of the maximum excretion of urinary nitrogen. Grant and Reeve (G19) in their detailed study of battle casualties in World War I1 noted that a rise in blood urea in the first 48 hours of injury was common in those who had initially lost more than 40% of their blood, irrespective of the size of the injury, and the greater the blood loss the higher in general was the urea. Two factors contributed: (1) diminished renal function in the early period after injury, and (2) increased formation of urea. Where blood loss was over 5Q% there was

6

D. P. CUTHBERTSON AND W. J. 'MLSTONE

little secretion of urine for 8-10 hours after injury, and the blood urea concentrations reached could be used as a rough method of measuring urea production. At this period a raised blood urea seems to depend on blood loss rather than on quantity of damaged tissue. I n patients with much loss of blood, urea usually reached its peak within 2 W hours and thereafter fell as renal function improved. Where the liver is affected through severe blood loss, deamination may be affected and accumulation of amino acids in the blood arises ( V l ) . Indeed, the liver of the severely shocked animal may make a net contribution of amino acids to the plasma rather than effect their normal rate of removal (R12). It has further been observed in man that the serum amino acids were usually lower 3 t o 4 hours after operation than before it, except in the case of isoleucine, which rose (55).There was an increased urinary output of all of eleven amino acids examined. Eades et al. (El) have reported the excretion of peptides in the urine of injured subjects, and increased excretion of certain enzymes have also been found (see Section 3.2.5.). I n these circumstances i t is not surprising that infused amino acids are poorly tolerated (B21). The liver defect is presumably one of deamination, since the low blood concentrations are not accompanied by any accumulation of ammonia in the liver of the shocked animal (W13). However, the observations of Gillette et al. (G7, G8) on the effect on goats of massive wounding indicate that both in untreated and penicillintreated animals increases in blood nonprotein-nitrogen, urea, creatine, creatinine, and plasma magnesium occur, with creatine showing relatively the greatest increase, but no rise in amino acids in the blood was found by them. Unwounded animals which had received an intravenous lethal dose of the endotoxin of Escherichia coli showed similar, but lesser, effects. The plasma Na+ and K+ remained practically constant in all groups. Some 48-72 hours after severe burns in man there have been reported increases in the plasma undetermined nitrogen (Tl, W l ) . The examination of 177 battle casualties by Green et al. (G21) also confirmed the absence of a rise in plasma amino nitrogen. Nevertheless, there are reports of amino aciduria, and Nardi ( N l ) has reported this following burns and other forms of trauma. It consisted of an increase in both the nonessential amino acids normally found in the urine and the essential ones, threonine, valine, leucine, isoleucine, lysine, and methionine, not normally detected. Chytil (C9) suggests that the urinary loss of amino acids in rats arising from an injection of turpentine is the main factor causing the negative nitrogen balance in this form of injury, possibly from liver damage. Definite traces of heat-coagulable protein frequently are excreted, notably during the time of maximum nitrogen excretion (C21).

METABOLISM DURING THE POSTINJURY PERIOD

7

It has been suggested that blood lost into the area of trauma may be the source of the nitrogen lost ( D l l ) , but the fact that the increased urinary nitrogen loss disappears a t highish environmental temperature and also on a protein-free diet argues against this view, and further the introduction of blood of equivalent nitrogen content to that excreted in the urine over and above that which would have been normal on the diets used, did not lead to this scale of loss (C5, C27, M19). During the early stages of the generally depressed vitality following receipt of a moderate to serious wound a t ordinary environmental temperature there may be a relative or absolute anuria (C21). This is less apparent a t higher environmental temperatures (C40). 3.2.3. Early Changes in Carbohydrate Metabolism and in the Tricarboxylic Acid Cycle

Green et al. (G21) noted a tendency for an inverse relationship to exist between the hyperglycemia which is a prominent feature some 10 hours after wounding, its degree varying with the severity of the injury, and the plasma amino acid nitrogen level. Stoner et al. (S17) have described the early period of depressed energy metabolism following injury and its associated biochemical changes. As their model they have taken the rat injured by a 4-hour period of bilateral hindlimb ischemia. About 1-5 hours after removal of the tourniquets and before failure of oxygen transport to the tissues, the rat is in a sufficiently steady state for metabolic experiments. The timing usually corresponds with the period of maximum hyperglycemia, but the conversion of tissue glycogen to blood glucose mainly occurs before the start of the experiments. When the tourniquets are removed, body temperature falls owing to a decrease in rate of heat production (515). While the rate of utilization of glucose by rat tissues is only reduced by 10% a t this stage of the injury, glucose aerobic oxidation is more severely affected, being reduced by 30%. This is about the same as the reduction in total O2 consumption of the r a t a t this time. Unlike glucose, the oxidation of pyruvate is inhibited by some 70% after injury. That this early depression of general vitality occurs has long been recognized (D5, MlO), and was earlier shown by Henderson et al. (H8) in patients with burns to amount to a decrease in oxygen consumption of 45-6076. When p y r ~ v a t e - ~is~ C injected into an injured rat, the label does not pass downward to CO, and the amino acids glutamate and aspartate associated with the tricarboxylic acid cycle, as i t does in the controls, and a greater proportion of it migrates upward toward glucose. I n consequence, there is a greater labeling of the blood glucose of the injured

8

D. P . CUTHBERTSON AND

W.

J . TILSTONE

rat despite the fact that gluconeogenesis is not increased a t this time. These findings link glycolysis to the tricarboxylic acid cycle and the reactions concerned in gluconeogenesis. Stoner and his colleagues believe that the depression in heat production shortly after injury is through interference with the tricarboxylic acid cycle a t or near the citrate synthetase stage, either through inhibition of this enzyme or through lack of acetyl-CoA. The liver, which is a major site of heat production, would be predominantly affected by the change. Interference with the first stage of the tricarboxylic acid cycle should by deprivation, lead to a fall in the concentration of compounds such as citrate, glutamate, and aspartate and also to the accumulation of pyruvate. These changes have been observed in the liver of the injured rat. Liver mitochondria isolated from injured rats, even from those on the point of death still show normal behavior in vitro (A2). If an uninjured rat is moved from an environmental temperature of 20°C to one of 3”C, there is a rapid increase in oxidation and rise in core temperature as measured in the colon, and this persists during cold exposure. This response, however, is not seen when injured rats are exposed to 3°C. I n these animals exposure to cold increases the rate of fall in body temperature. It is not necessary to have a fatal injury to produce this altered response. At 24-48 hours after an injury, when the rat can still maintain its body temperature in an ambient temperature of 2OoC, it may be unable to do so when exposed to 3°C (517). Although Stoner et al. (S17) found adrenaline (epinephrine) and noradrenaline (norepinephrine) to be inactive in relation to temperature response (see also Section 7) , isoprenaline and dichloroisoprenaline both increased the temperature of injured rats when given just before removal of the tourniquets in these experiments, or up to a t least 1.5 hours afterward. Although dichloroisoprenaline is a p-adrenergic blockade agent, it also has a powerful agonist action. These authors have pointed out that the rise in temperature is due to an increase in the rate of heat production, and they are working on the idea that the substrate is nonesterified fatty acid, and that a large excess of this can overcome the interference with oxidative metabolism. They pointed out that in their particular experiments the injured animal gained nothing from the increase in the rate of heat production. The survival time was usually shortened by a rise in body temperature in these early hours following injury when the natural reaction is that of “ebb” rather than “flow.” No effect of injury on the removal from the blood stream and tissue distribution of intravenously injected albumin-bound palmitate-l-14C has been noted, but its oxidation in organs associated with small bicarbonate pools appeared t o be depressed to about the same extent as pyruvate (817).

METABOLISM DURING THE POSTINJURY PERIOD

9

As palmitate enters the tricarboxylic acid cycle via acetyl-CoA, their results with labeled palmitate suggest that the inhibited step is the oxidation of palmitate and pyruvate after their pathways have joined. M r l z (M15) has shown that animals are more resistant to different types of shock-drum, burn, hemorrhagic-the higher their liver and muscle glycogen levels. In animals with low glycogen level severe hypoglycemia develops under certain conditions following the injury. Several drugs prevent the hypoglycemic reaction in shock and increase simultaneously the resistance of the animals. Insulin, however, does not make the animals less resistant despite lowering of blood sugar levels, and padrenergic blocking agents though permitting higher muscle glycogen levels do not increase resistance. The term “traumatic diabetes” has been used to describe the temporary alteration of carbohydrate metabolism characterized by hyperglycemia and glycosuria following injury. Howard et al. (H14) showed that glycogenolysis and gluconeogenesis are increased in proportion to the extent of the injury, and the tolerance of the body to both oral and intravenous loads of glucose is also diminished after injury (H6). Johnston et al. (55) have found that moderate injury raises the level of fasting blood glucose and reduces the tolerance t o intravenous glucose for up to 72 hours. Plasma levels of insulin and growth hormone are raised after injury. Corticotropin produces changes in carbohydrate metabolism similar to, but greater than, those induced by injury, but it has no effect on plasma insulin or growth hormone levels (see also Section 8.2). The hyperglycemia may be transformed into an acute diabetic state, as has been noted in burns where burn stress has been recorded (E6, R7). 3.2.4. Electrolytes and Cellular Injury

The classical signs of inflammation are present in degree varying with the injury and with the time interval. These involve permeability changes. I n starvation or other wasting conditions there occurs a loss of K+ paralleling the excretion of the other products which result from the wasting of the tissues. Gamble et al. (Gl) observed that during the first few days of fasting there occurred a loss of Na’ which was in excess of the amount contained in the autolyzed tissue and was presumably due to the loss of extracellular fluid. After the first 6 days of a 15-day fast the K/Na ratio in the urine corresponded to that in muscle tissue. The processes a t work in the injured animal fed an adequate diet differ in that there is either no parallel urinary loss of Na or occasionally a slight fall (C37), but there may be definite early retention in man (B17, M14, W15, W19). Retention of chloride and sodium has long been known

10

D. P. CUTHBERTSON AND W. J. TILSTONE

to occur in fevers, but Wilson e t al. (W19) were the first to point out that a considerable fall in the level of serum sodium may occur in man after extensive burns.. It was found that it could be restored rapidly to a normal level by deoxycorticosterone acetate (W18). In cats they found that after scalding under Nembutal anesthesia the level of the serum sodium in arterial blood and in cerebrospinal fluid steadily declined, while the sodium in the red cells tended to increase. Substantially the same changes followed when shock was induced by hammering the thigh muscles. There is retention of body sodium with a concomitant loss of body potassium (Ul, W15, W17). A K+ loss occurs and for a period may be relatively greater than that of N in the fractured rat in relation to the composition of the injured tissues (C37), and in this respect the reaction is somewhat similar to that in fever. For example, an analysis of the quadriceps of rats adjacent to a fractured femur and the contralateral uninjured muscle group revealed a relatively greater loss of K than N, both histochemically and chemically. The loss of K+ in the urine was greater than could be accounted for by such local changes. To a lesser extent this also holds for creatine. It is of interest that the addition of extra dietary carbohydrate t o rats on a diet some 6% short of adequacy did not affect the excretion of either Na’ or creatinine following fracture (C37). Walker et al. (W4) reported a fall in urinary calcium following operations on soft tissues of a “severe” character, and this correlated well with a fall in sodium output. On the other hand, urinary magnesium tends t o behave like nitrogen and show an increased excretion during the second to fifth or sixth days following injury (W3). 3.2.4.1. Acid Base Balance. Trauma from whatever cause may strain control of the acid-base balance by increasing production of acids, especially where there is tissue anoxia, by diminishing power of lungs and kidneys to eliminate them, and by lowering of buffering capacity through anemia and hypoproteinemia [for review, see Walker (W2) 1. Walker pointed out the possibility of massive transfusions of citrated blood producing in the postinjury period a metabolic alkalosis due to the metabolism of citrate, leaving sodium to combine with base. When fluids must be given intravenously there are two schools of thought with respect to sodium: (1) that in general an intake of not more than 80 mEq of sodium is adequate; (2) and more recent, that up to 3.0 liters of Ringer-acetate solution should be given during operation, depending on the severity of the procedure (S10). The control of potassium loss in the urine following injury is not a problem if oral feeding is resumed shortly after operation, but if the patient has t o be maintained on intravenous fluids for several days, and

METABOLISM DURING THE POSTINJURY PERIOD

11

these fluids contain no potassium, a deficit of this will develop. This may not be serious in itself, but if a t the same time there are additional losses of potassium from the bowel, as a result of ileus or some such complication, then a significant deficit may develop. It may be of the order of '2W-300 mEq K+. It can be corrected after the first 36-48 hours following injury by addition of some 80 mEq K+ each 24 hours. This should prevent an unnecessary postoperative complication (L3) (see also Section 6.2). 3.2.5. Enzyme Changes I n studies of lymph from injured and noninjured limbs, Lewis (L6) found that after 60°C burns applied to the hind limb of the anesthetized cat the K+ remained intracellular. Even after burning a t 80°C there was only a small increase of K+ in the lymph, and this could be accounted for by the hemolysis which occurred and which resulted in a slight rise in blood K+ as a direct result of cell damage. It may be that the release of K+ is associated with changes which also affect mitochondria1 enzymes. It is only after massive cell breakdown as caused by freezing the limb solid that considerable amounts of K+ appear in the lymph drainage. It has been pointed out that since K+ remains intracellular following injuries which result in the leakage of intracellular enzymes of high molecular weight, this affords further evidence that changes of the cell permeability are specific (L6). Slater et al. (S12a) have shown that in spite of the leakage of these enzymes of high molecular weight, NAD+ and NADH, which are mainly cytoplasmic in liver, remained firmly within the liver cells. 3.2.5.1. Lymph. Where a tissue is damaged, one of the first signs of injury is a leakage of fluid and protein from the vessels into the interstitial fluid. During severe acute injury there is an escape of intracellular enzymes from the injured tissue into the circulating blood (Ll, R11, T3). This is part of the local component of the inflammatory response. It is not clear whether the enzymes come from completely broken-down cells or escape from cells in which, though still functional, the membrane permeability has been increased by the injury. Wroblewski et al. (W22) have already suggested that in some types of hepatocellular injury the enzymes escape from the cells by means of a metabolic or excreting process. Lewis (L6) has described the biochemical changes which occur in lymph, which is the drainage of the interstitial spaces, coming from the hind limbs of anesthetized cats injured by ischemia or burning. After burning a t 60°C there was a significant increase in the concentrations of the cytoplasmic enzymes glutamic-oxaloacetic transaminase (GOT) and lactic dehydrogenase (LDH) as a result of the increased permeability of

12

D. P . CUTHBERTSON AND W. J. TILSTONE

the cell membrane; when the limb was burned a t 80°C, there was a marked increase not only in the cytoplasmic enzymes, but also in the mitochondrial enzyme glutamic-pyruvic transaminase (GPT) Thus with the stronger burn the permeability of the intracellular mitochondrial membrane was also increased. Not until the most severe injury of all, i.e., freezing the limb solid, was there an increase in the concentration of the lysosomal enzymes acid phosphatase and p-glucuronidase in the lymph. Very mild injury, such as hot water a t 50°C or ischemia for 1 hour, produced no increase in protein or enzyme concentration in the lymph, although after the latter there was an increase in lymph flow. It would appear th at lysosomes take little or no part in the tissue reaction to injury even after the most extreme tissue breakdown. This finding is not in agreement with the original hypothesis of de Duve (D6, D7) that rupture of lysosomes could initiate cell injury, but it is consistent with the findings of Slater and Greenbaum ( S l l , S12). It was subsequently reported that, in the rabbit, burns a t 80°C induced rises in all three groups of enzymes, namely, LDH and GOT, the cytoplasmic GPT, and the lysosomal enzymes acid phosphatase and p-glucuronidase (L7). No change occurred a t the time in the contralateral limb. The rabbit seems more sensitive in these respects than the cat. The finding that prolonged ischemia causes such a small amount of cellular injury was somewhat unexpected, since findings in other tissues such as liver (D8) and heart (W21) indicate that ischemia causes considerable leakage of lysosomal and other intracellular enzymes. The absolute amount of the enzymes released into the circulation from the heart during cardiac infarction is considerable, since high serum levels are maintained for several days, although injection of LDH raises the level for only a few hours. It has been suggested (H5) that the release is part of an acute syndrome rather than a release from necrotic tissue. Lewis' findings (L7) make it clear that, a t least during the acute phase of the reaction to injury, the intracellular enzymes escape from the cells of the injured tissue. The possibility is not excluded that some substance is released from skin cells especially in the case of skin burns, which might mediate the escape of the intracellular enzymes from other cells in the lower-lying tissues which might suffer direct damage. Lewis points out that during the experiments the blood levels of these enzymes remained very low when the concentrations in the lymph had increased many times. The enzyme levels in the lymph draining an injured area are therefore a much more accurate indicator of tissue injury than the blood enzyme levels. Hydroxycortisone and salicylate had no effect on these changes despite being antiinflammatory agents (L7). Are these changes in enzyme concentration in lymph due rather to

.

METABOLISM DURING THE POSTIN JURY PERIOD

13

increased synthesis than to just leakage? The overall picture is building up that with mild injury there is increase in the enzymes building up energy supplies. Histamine and bradykinin when infused did not produce these changes. The concentration of protein in the lymph also increases after cellular injury except in the case of ischemia. 3.2.5.2. Blood and Tissue 3.2.5.2.1. Serum isozymes following different types of injuries. The multimolecular forms of enzymes, called isozymes, may be identified using combinations of suitable gel electrophoresis and histochemistry. The changes in naphthylamidase and alkaline phosphatase following burns most likely indicate an alteration of liver metabolism and may persist for several weeks (A5). A rise in plasma alkaline phosphatase, but of kidney origin, can be found after limb ischemia in the rabbit. 3.2.5.2.2. Plasma renin. Collins and Hamilton (C11) many years ago reported that severe hemorrhage in dogs caused an increased plasma renin act,ivity. It has now been shown that operation and hemorrhage in man cause renin release, and it is suggested that this is mediated through a spinal reflex (B15). The role of the renin-angiotensin system is not clear, but it seems to be activated during operation in man in relation to adjusting the blood volume to the blood vessel capacity. Because of its marked vasoconstriction, angiotensin may stimulate the release of aldosterone which causes sodium retention and slows adjustment of blood volume. I n shock the excessive peripheral vasoconstriction may be caused by circulating angiotensin and not, as proposed by Nickerson (N5), by catecholamines alone. Secretion of ACTH may possibly stimulate renin secretion. Laparotomy in the rabbit per se causes a marked increase in plasma renin activity. Such change is in contrast to the results in rabbits subjected only to blood sampling and anesthesia (M2). Increased plasma activity could stimulate the secretion of aldosterone leading to postoperative sodium retention (L4, M14). 3.2.5.2.3. Urine: Uropepsin. The increase in adrenal activity is accompanied by a rise in gastric acidity, pepsin concentration, and urinary uropepsin excretion, regardless of the presence of the vagus nerve or gastric antrum (D10, E2). The increase in uropepsin excretion closely parallels the increase in 17-hydroxycorticoid output. The maximum gastric and adrenal responses appear on the second to fourth day (G20). 3.3. HEMOGLOBIN AND EFFECT OF TRANSFUSION

I n limb injuries, after blood loss there is usually a fairly rapid dilution of blood, first with protein-free or protein-poor fluid, but later chiefly

14

D. P. CUTHBERTSON AND W. J. TILSTONE

with fluid containing protein in nearly the same concentration a s the original plasma. During the first 12 hours after injury the average dilution is roughly proportional to the hemorrhage. In fit men a hemoglobin concentration above 14 g/100 ml indicates a previous blood loss of less than 40% of the blood volume; and a hemoglobin concentration below 12 g/100 ml, a previous blood loss of more than 40%. Transfusions of dilute plasma or of blood rarely alter the plasma protein concentration by more than k 0 . 5 g/100 ml. The greater part of the diluent in both rapidly leaves the circulation (G19). In cases of trauma, Flear and Clarke (F2) found that the nitrogen released by the breakdown of transfused red cells is not excreted but is available t o the rest of the body. The experiments of Whipple, Millcr, and Robscheit-Robbins (W12) indicate that when undernourished anemic dogs are fed protein it is preferentially utilized for synthesis of hemoglobin. It seems likely that this is the case in the anemia of trauma. Flear and Clarke (F2) have obtained the impression of improved convalescence after major injury when blood transfusions adequate to prevent the anemia are given. When this has been achieved, and in the absence of major complications, they have not seen the lengthy early stages of convalescence described by Moore and Ball (M14). Such clinical impressions suggest that when there has been generous replacement of blood loss incurred by injury and its treatment, the metabolic picture may be modified. They have not usually found the retention of sodium, chloride, and water in the transfused cases after the first 24 hours, though in nontransfused cases the amounts and period over which they were retained were of the same order as those reported by others to follow surgery. Flear and Clarke (F2) found the loss of nitrogen to occur in both transfused and nontransfused cases, but such losses were greater in the nontransfused group, and the excretion of nitrogen suggested that the posttraumatic increase in catabolism was postponed by transfusion. Their untransfused patients showed approximately the same negative balance as the senior author had earlier found after fracture (C22). Their data also suggest that in all patients the metabolism of potassium was influenced by factors other than the concurrent metabolism of nitrogen. The differences between the transfused and nontransfused cases observed by Flear and Clarke (F2) appeared to be related to the adequacy of the circulating red cell mass and the total blood volume during the early stages after injury and its treatment. They suggest that blood loss and an adequate circulating blood volume are major etiologic factors of posttraumatic metabolic changes. The concept of a fixed pattern of response to injury involves an oversimplified approach and may lead to

METABOLISM DURING THE POSTINJURY PERIOD

15

an incorrect attitude to treatment. With adequate transfusion, these observers believe that one need not be concerned about the loss of potassium or overloading with sodium and water. The characteristic later change in hemoglobin concentration is a progressive fall until about the third or fourth day after injury. The rate and extent are much influenced by the amount of blood lost and the quantities of red cells and plasma transferred. Significant quantities of red cells are rarely added by the patient before the fifth day, and often not till later. Red cells may remain low, often in the region of 50% of predicted normal. Limb and skeletal injuries show no evidence of hemoconcentration a t any stage of their course. It has been found that untransfused casualties with abdominal injuries not perforating the intestines exhibit changes in initial hemoglobin and plasma protein levels comparable with those shown by patients with limb injuries who have lost similar amounts of blood; that is, they show hemodilution (G19). Patients with intestinal penetration may show some hemoconcentration. Evidence is insufficient to determine whether, after hemorrhage, the factors tending t o cause hemoconcentration ever slow down the processes causing dilution. I n some instances reported by observers they did not. The anemia of injury with decrease in hemoglobin, elevation of erythrocyte sedimentation rate, and a loss of erythrocytes from the effective blood volume have been the subject of recent studies by Gelin and his collaborators (Gla-G4). Furthermore, immediately consequent on the injury the white cell count is usually elevated, and the thrombocyte count is decreased (R4). Serum iron and total iron-binding capacity fall after major surgery and trauma, especially when blood has been lost, and serum transferrin also falls (Fl, N7). These changes may be interpreted teleologically as a deviation of iron to the marrow for hemoglobin synthesis, just as serum iron falls when erythrocyte formation is rapid, or possibly th a t iron is taken up by the reticuloendothelial system after its release from hemoglobin in the normal metabolic cycle and is not passed into the blood. Its level is influenced by the red blood cell mass. Studies with iron chelates have shown that a greatly increased amount of iron is available for chelation by the drug desferrioxamine during recovery from posttraumatic anemia even when the serum iron is low (01). 3.4. HYPERCOAGULABILITY

Speeded clotting and activated fibrinolysis dominate the first few hours after injury but are soon followed by a rebound to normal, or even

16

D. P. CUTHBERTSON AND W . J . TILSTONE

slower, clotting and prolonged, that is inhibited, fibrinolysis. Nevertheless, the platelet count continues to fall for 1 3 days. Obviously the early accelerated clotting is an emergency mechanism against continued hemorrhage and supplements the vasoconstrictor effects of adrenaline on torn vessels. Speeded clot lysis can be regarded as a protection against excessive thrombosis in vivo; and its subsequent inhibition as a mechanism guarding against fibrinolysis and other proteolytic effects of plasmin in the blood. Speeded clotting after hemorrhage is probably due to thromboplastin or a thrombotic accelerator in the blood (T6) ; this suggests that microthrombi form in the circulation and are most numerous when hemorrhage is severe.

PROTEINS 3.5. PLASMA Owen (02) has recently reviewed the literature describing plasma protein changes following injury. Some of the salient features will be considered here. It has long been known that the level of fibrinogen in the blood plasma is very variable and that acute inflammations are invariably associated with high values: further, that in contrast to the other proteins a deficit of it can be restored in a few hours. Research indicated the general nature of the reaction to injury by heat, cold, necrotizing agents, bone fracture (C8) or by sulfur and nitrogen mustard vesicants (G10) to be the same, namely, a decrease in serum albumin and an increase in aglobulin. The electrophoretic pattern exhibited sharp spikes in the and p areas which were interpreted as an indication of the appearance of abnormal proteins in the serum. Similar changes in the electrophoretic patterns had been known to occur where there was inflammation or tissue destruction. (Y

3.5.1. Albumin Since these earlier observations were made the whole technique of plasma protein estimation has considerably improved and the earlier findings substantially confirmed, namely, that after major surgery in man or after accidental injury or burning the plasma albumin falls (B20, D4, L2, M1, M5, P5, R6, S13a). The fall, according to Owen (02),averages about 0.8 g/lOO ml (about 20% of the normal plasma albumin level), but decreases of twice this amount have been found in individual patients with the depression maximal between 4 and 10 days and sometimes normal values may not be regained for several weeks. The fall is particularly marked after extensive burns. Davis e t al. ( D l ) have noted an

METABOLISM DURING THE POSTINJURY PEBIOD

17

increase in the rate of plasma albumin turnover in burned and injured patients with a maximum period of breakdown (2-3 times normal) coinciding with the peak of urinary nitrogen excretion. Albumin is lost into the exudate from burned areas in substantial amounts. 3.5.2.

Globulins

3.5.2.1. a-Globulins. An increase in plasma al- and a,-globulins occurs after surgery and after burns and in the case of animals is much reduced if the animals are first starved (B12). Ceruloplasmin rises steadily after injury, and raised levels may be found a fortnight after surgery ( M l ) . Haptoglobulin rises until about the fourth day, with peak between 3 and 7 days, then st a rk to fall (02). A slow moving a,-globulin-variously designated az-acute phase (AP) or a-2-glycoprotein (GP)-which is not normally detectable in the plasma of patients and experimental animals, appears in acute inflammatory states and after injury in general. It appears in the plasma 12-18 hours after the injury and remains detectable for 7-10 days or longer. Adrenalectomy prevents its appearance after injury. It can be restored in the adrenalectomized rat by corticosterone or cortisol. [For review see (B10, 02,WS,W9).] There is a linearity of response to graded cortisol treatment in traumatized adrenalectomized animals (B11). The adrenal cortical control of the appearance of slow as-globulin is incomplete in the 2-day postpartum maternal rat, because the globulin can be demonstrated in most adrenalectomized animals in this condition. Adrenal cortical control over the appearance of the globulin is not simply in the postpartum maternal animal, because the percentage of adrenalectomized rats with demonstrable slow a,-globulin can be increased by the administration of corticosterone (H7), 3.5.2.2. p-Globulins, Including Transferrin. I n man there is no consistent change in total /3-globulin in plasma following surgery (H12, 22) or after burns (D9, L2). However, an increase apparently occurs after experimental injury in goats (G7) and rats (N4). Plasma transferrin, which constitutes about 30% of the p-globulin, decreases after surgery, and as a rule its concentration varies with that of plasma albumin (02)) but exceptions occur in iron deficiency, in pregnancy, and in some acute conditions, An increase in plasma transferrin has been noted after hemorrhage in man but is not apparent until 10-20 days after the loss of blood. 3.5.2.3. y-Globulins. No consistent change occurs in plasma 7-globulins immediately after surgery (B7).However, an increase in plasma 7-globuIin has been observed in man after burns (B20, D2, P4). An increase occurs in all types of infection (02). Barr et al. (B5) report th a t con-

18

D. P. CUTHBERTSON AND W. J. TILSTONE

centrations of serum y-globulin 2-4 weeks after burning were significantly lower than in patients not exposed to warm dry air. 3.5.3. Fibrinogen Plasma fibrinogen rises following stress in the form of tissue injury, hemorrhage, cortisone, Nembutal anesthesia, and probably also anxiety (C34a, G11, G15, H4, M1, W6, W9). Peak values may be twice the preoperative level or more, and are reached on days 2-8. Turpentine abcesses in animals, which normally cause an increase in plasma fibrinogen, fail to do so if the liver is poisoned (F5). Likewise in man the fibrinogen response to infection is reduced or absent in patients with extensive liver damage (B14, F4, G16, H3). On the other hand, in experimental animals moderate liver damage is itself a stimulus to fibrinogen production (F5, S6) .

Mucoproteins and Protein-Bound Carbohydrates There is a temporary rise in mucoprotein and in protein-bound carbohydrate and hexosamine (BSa, B13, N4, P5). The response is greatly reduced if the liver is first poisoned (R8, WlOa). They rise in a great many infiammatory conditions (02).

3.5.4.

3.5.5. Metabolism of Plasma Proteins The acute effects of injury on plasma protein turnover level are best explained mainly in terms of an increase in metabolic turnover of the protein concerned and the appearance of an abnormal wglobulin. All the proteins acutely affected are formed in the liver, which is stimulated to increase protein synthesis through the action of some mediator generated by injury. In those which rise, namely, haptoglobulin and fibrinogen, increased synthesis must, at least temporarily, exceed increased catabolism. In the case of proteins whose plasma levels fall (e.g., albumin and transferrin) , increased catabolism must exceed increase in synthesis. According to Owen (02) net transfer of protein from the intravascular to the extravascular compartment may contribute to the fall in plasma albumin and transferrin, but in most injuries this is likely to be a minor factor. I n burns, however, physical loss of protein from plasma, affecting in particular low molecular weight proteins, becomes a factor. RETICULOENDCWHELIAL SYSTEM (RE) IN RESPONSE TO DIFFEBENT TYPES OF INJUBY Considerable caution must still be exercised in allotting a definite role to the RE system in the defense of the body against physical trauma though there is some evidence which strongly suggests that the greater 3.6.

METABOLISM DURING THE POSTINJURY PERIOD

19

the degree of traumatic shock, the greater are both the magnitude and duration of early phagocytic depression (A4).

4.

Nature of the Post-shock Delayed Metabolic Response

This corresponds to the ‘(flow” period of returning metabolic activity following the early “ebb” period of depressed metabolism resulting from the injury, whether accidental or elective, and can be considered as a further phase of the inflammatory reaction. It has sometimes been described as the “delayed response,” but this is not a sufficiently accurate description for it connotes that something has been held back, rather than being part of a cyclical response. Our interest in understanding this reaction and the contributions of our colleagues to it extend over a period of some 40 years and have recently been reviewed by us (C19, C20,C B C27, C33, C34). I n brief, injury through accident, radiation, necrotizing agent or by elective surgery results in the immediate loss or destruction of a portion of the body’s protein content. I n addition, following moderate to severe trauma there generally occurs in the otherwise healthy and well-nourished person a t normal environmental temperature a marked rise in urinary nitrogen, sulfur, phosphorus, potassium, magnesium, and creatine accompanied hy increased basal oxygen consumption. These catabolic changes reach a maximum in 4-8 days after injury and are paralleled by increased pulse and respiration rates and by a slight degree of fevertraumatic fever-and this is part of the general component of the inflammatory response. The rise in urinary nitrogen excretion is mainly as urea, that of sulfur as sulfate and phosphorus as phosphate. There is an early retention of sodium. Comparable disturbances of nitrogen metabolism following injury have been found in animals as divergent as man, sheep, earthworms, and crabs. I n man and the rat, injury by fracture of a long bone does not affect the digestibility of dietary protein, a t least, over the periods studied (C5, C27). The role of the endocrines is discussed in Section 8.2, The source of the nitrogen and sulfur lost by man and experimental animals has never been clearly defined, but because of its magnitude and the N: S and N: K ratios of the excess in the urine it must come mainly from muscle ((321, C22, (337). Munro (M17) has postulated this labile protein in man as 300-400 g, but very much larger amounts are frequently lost as a consequence of bony injury (C21). There is also evidence that in the previously depleted subject there is little response ( A l ) , and this clinical evidence agrees with the earlier findings on animals (M4, M19, M20).

20

D. P. CUTHBEBTSON AND W. J. TILSTONE

Subsequent operations and manipulations may cause further disturbance. Even after a month there may be a slight negative balance, but normally there is a slow merging into an anabolic phase (52). I n burns loss of tissue plays a large part in the initial nitrogen depletion and this is only slowly made good: nevertheless, there is a marked catabolic response (B19, C15, H11, T l ) . More nitrogen may be eliminated in the course of 10 days following severe bony injury than is present in the whole of the liver ((321). The discharge, however, involves the liver, and we have earlier in this paper drawn attention to the effect of anoxemia on the capacity of the liver to deal with urea formation. T ha t disuse atrophy of muscIe, as might arise by fixation of a leg in a splint or by pinning, may be a factor in causing a loss of body nitrogen following injury has been found to be so, but this is smaller than that arising from the actual damage done (ClS, C19, $4). Although studies on day 7 postfracture revealed no changes in the nitrogen content of the liver of fractured rats (F3), a significant loss on day 5 has been reported (W20). As the level of food intake by patients is increased to levels quite beyond their apparent requirements a t rest in bed, the excess of urinary nitrogen and other catabolites over that anticipated from the level of intake is diminished, but does not usually drop below the level of intake during the days of maximum excretion. Neither does increasing the protein intake per se-even by large amounts-completely overwhelm the response to moderate to severe injury a t normal environmental temperature. Here the N-saving effect of surplus food may play a part in reducing loss (C28, C35, C36). The observations of Wilkinson (W14), Clark (ClO) , and others, demonstrate th at catabolic response is additional t o the changes attributable to reduced intake. The observations of Rossiter (R10) and Keys et al. ( K l ) have provided information concerning the pattern of changes in blood volume and proteins which follow partial starvation and the recovery process. Although Cairnie et al. ( C l ) found no increase in heat production or in nitrogen excretion following fracture in the protein-depleted rat indicative t ha t f a t was not being mobilized in the absence of labile protein, nevertheless some involvement of lipid metabolism seems t o occur. Birke et al. (B8) have reported high levels of circulating free fatty acids and reduced levels of cholesterol, with no change in triglyceride level in severely burned patients. Continuous infusion with noradrenaline will produce this. A search of the earlier literature on the effect of injuries ranging in severity from simple hemorrhage to the seriously wounded on the metabolic response, indicates the extent of an underlying basic pattern.

METABOLISM DURING THE POSTINJURY P m O D

21

Malcolm (M7) in his study of the Physwlogv of Death from Traumatic Fever noted that there might be evidence of increased nitrogen metabolism after a major operation although the body temperature remained low. During the concluding years of World War I, Wertheimer et al. (W11) observed that, after the shock period in battle casualties had passed, the urinary excretion or urea and ammonia remained a t a high level for some time and could amount to 17 g of urea nitrogen. This phenomenon of increased protein catabolism following trauma was later confirmed by many workers and in many forms of injury (B5, B18, C2, C10, H14-Hl6, K2, L5, L9a, M4, M14, M15, M19, M20, N2, N3, W1, W16). Actual tissue damage is not necessary. Gontzea et al. (G13a) have shown that painful stimuli can initiate protein catabolism which cannot be accounted for by tissue loss. As to the degree of injury necessary to elicit a response, Moore (M13) found no significant disturbance of nitrogen or potassium balance after a single venous hemorrhage of 770 ml in a normal volunteer, though there followed obvious sodium retention. H e also recorded no significant eosinophil or hydroxycorticoid disturbance following a sudden hemorrhage of this order. Ether anesthesia and pentothal-nitrous oxide-curare anesthesia per se likewise caused no systematic alterations in the adrenaline/noradrenaline values in man (M13). I n general, the more severe the injury the more vigorous the catabolic response, provided the patient when injured had not been deprived of his labile protein reserve by previous illness or starvation (C27). 4.1.

SIGNIFICANCE OF INCREASED CATABOLITES AFTER INJURY

Levenson and Watkin (L5) have reported that their I5N experiments indicate that both protein catabolism and anabolism are accelerated after injury: catabolism more than anabolism. Kinney (K2) surveyed studies on infections in rats, chemical abcess formation in dogs, and burns in rats and in man and came to the same conclusion. As already mentioned, a rise in the basal consumption of oxygen follows fractures and operations on bone in man, and this parallels the nitrogen and sulfur excretions in the urine (CZl). A 15-2076 increase in heat production followed fracture of one femur in the rat, and this closely paralleled the nitrogen excretion. It was essentially due to an increase in the resting rate. I n the protein-depleted animal, however, this did not occur ( Cl ) . On calculating the heat production that would arise from oxidation of protein equivalent to the extra nitrogen excreted in the urine of rats with such a fracture the value obtained fitted almost exactly the increase in heat production actually found. When the oxygen consumption required to oxidize protein, equivalent to the extra nitrogen elimi-

22

D. P. CUTHBERTSON AND W. J. TILSTONE

nated following fracture trauma in the 1930 series of cases reported by Cuthbertson in 1932, was similarly calculated, the results provided again little or no margin, a t least during the first 10 days, for fat to be oxidized in significantly greater amount than normally. Miksche and Caldwell (M11) have now reported that rats with double the injury reported by Cairnie et al. ( C l ) show a significant increase in heat production ( 1 6 2 0 % ) and in average body temperature, when housed a t 28”. From relatively contemporaneous analysis the former calculate that the heat equivalent of the urinary nitrogen excreted accounted for only 1 7 4 6 % of the actual increment and th a t the major portion of this heat increment was accounted for by the van’t Hoff effect (Ql0 effect) secondary to an elevation of the average body temperature. Cuthbertson (C21) had earlier noted in man that the calculated van’t Hoff effect actually equated the observed increase in BMR in man. Miksche and Caldwell (M11) have recently reported th a t after thyroidectomy rats with such injuries show no significant changes in heat production or average body temperature. Caldwell et al. (C4) had earlier reported t ha t thyroidectomy did not prevent the metabolic response to injury in burned rats and an earlier review by Moore (M13) of the literature found the metabolic data on the thyroid in relation to trauma difficult to interpret. H e concluded that, on the whole, it would seem that the thyroid partakes in the general alteration in endocrine and metabolic activity but without the early, massive, and systematic nature characteristic of adrenocorticol and antidiuretic alterations. I n their multiple but brief measurements of “basal” oxygen consumption by burned patients, beginning soon after trauma and extending through convalescence, Cope et al. (C13) found increments in metabolic rates to plus 3040% above normal early after injury. The oxygen consumption rate gradually declined but continued to be elevated in many patients for as long as 2 months. Although Cope e t al. considered that fever probably played a part, the temperature rises were irregular and did not correlate with the metabolic rate. Following major elective operations there was usually a rise of onIy 10-15% in basal oxygen consumption lasting from 3 to 5 days with a return to preoperative levels by the end of 1 week. Kinney (K2) found th at the resting metabolic rate rises after operation on man by about 10-15% though the total energy expenditure may remain essentially unchanged because on account of the injury there is a considerable reduction in voluntary muscular activity. With an overwhelming injury like an 85% burn, a progressive metabolic deterioration may occur. Kinney considers fractures as presenting a midway position regarding oxygen uptake.

MIDTABOLISM DURING THE POSTINJURY PERIOD

5.

23

Time Factor in Multiple Injuries

The gradation of the trauma components, the order of infliction, and the timing between them are three vital features of importance in considering the systemic effect of combined injuries. Regarding timing three alternatives exist, provided the components differ in severity: (1) the initial damage is small; (2) the initial damage is large; (3) the agents “hit” simultaneously. Under certain conditions the small trauma will adapt the organism, and temporarily increase the resistance, by means of unspecific stimulation. I n the case of 2 and 3 the combined effect will be additive or strongly synergistic depending on timing (53). According to Koslowski and Messerschmidt (K3a) , the time factor seems to have more influence on the prognosis of combined injuries involving irradiation than the dose of X-rays applied to cause irradiation damage. If the surgery was performed after radiation, the overall mortality with various forms of trauma in mice was found to increase up to 90% whereas the reversal of this procedure did not change the radiation mortality (27%) and wound healing remained either unchanged, or a significant decrease of the overall mortality sometimes occurred. The same tendency has been noted by others (SlS). This effect of irradiation on subsequent healing has been termed a “radiological traumatic shock syndrome” (L2a). If irradiation and other form of damage is produced on the same day the effects show synergism (R5). The ratio of methylcobalamin to total vitamin B,, derivatives of extractable B,, has been determined in liver from mice who were subjected to different types of injury (mechanical trauma, burns, and ionizing radiation) inflicted separately or in various combinations. A decrease in methylcobalamin was observed paralleling the severity of the damage. There may thus be a decreased synthesis of methycobalamin or a n increased catabolism or leakage from the liver-or combination of these causes. The method used did not determine the nonextractable cobalamin, so that a disappearance into a nonextractable form could have been the cause (L9). 6.

Nutritional Aspects

6.1. EFFECTS OF DIMINISHED FOOD INTAKE ON THE METABOLIC RESPONSE As described above, it is now fairly certain that the source of most of the additional nitrogen lost in the urine in the “flow” phase of the metabolic response to injury is labile body protein, and not liver or gut

24

D. P. CUTHBERTSON AND W. J. TILSTONE

protein (C21, F3) and labile body protein is a nutritionally dependent protein store (M17), although turnover of muscle protein is, if anything, decreased after injury (L5).Rats fed a protein-free diet for some time before being given fracture of one femur will not show any protein catabolic response (Cl, M20) and also groups of rats fed to different levels of protein intake each showed a protein catabolic response proportional t o the preinjury level of intake, even when the dietary regime thereafter was changed simultaneous to fracture (M19). The data associating the levels of urinary nitrogen after injury with the level of protein nutrition (and also basal heat production) seems to us to be firm evidence of the existence of a “protein metabolic response to injury” largely independent of qualitative or quantitative changes in food intake consequent on trauma. I n any case, in the animal work outlined above, changes in voluntary intake of food throughout the immediate preinjury and all of the postinjury period were negligible, so no “starvation component” existed, and Campbell and Cuthbertson (C5) have shown that there are no changes in protein digestibility in the rat consequent on injury, and fecal nitrogen output in man is little affected (C22). Indeed, the r a t has proved t o be a very suitable model for injury studies in respect t o comparison with the responses of humans (C37, C40). However, it must be recorded th at some workers feel that in the human subject the period of anorexia which usually follows injury is responsible for the negative-nitrogen balance of the postinjury period (Al, D11, D12). There are a number of unsatisfactory aspects about much of this work. We have already made mention of the considerable volume of work with well-nourished animals on a constant pre- and postinjury food intake, which show a definite and marked loss of nitrogen in the urine. Also the exact nature of the response seems to have escaped some workers. The loss of nitrogen in the urine consequent on a diminished intake is a modification of that to be found following starvation or protein-free intakes (ClO, L10, W14), that is, situations where the urinary nitrogens (which are direct measures of negative nitrogen balance in these circumstances) are maximal to start with and then decay to endogenous levels. Following trauma the pattern of excretion of urinary nitrogen is different. Starting from fairly normal levels on the first day or two following injury, urinary nitrogen rises to a peak 4-8 days postinjury and then gradually falls back to normal, sometimes with a late secondary rise. Not only does this pattern indicate that we are dealing with other than the consequences of diminished food intake, but there is also an associated and parallel rise and fall in oxygen consumption and heat loss which is characteristic of this metabolic response to injury and is quantitatively similar t o the increased protein catabolism (Cl).

METABOLISM DURING THE POSTINJURY PERIOD

25

It may be that gastrointestinal surgery, which is a common example of injury used by those who argue for the overriding importance of food intake in causing the nitrogen loss, is not a suitable model for injury studies, although the present writers have previously remarked that almost any injury will give rise to a qualitatively similar response (C33). It is now some time since Cuthbertson (C22) demonstrated th a t the negative nitrogen balance following injury could be largely counteracted by feeding large quantities of protein associated with a high-calorie diet. However, the pattern of rising urinary nitrogen was not thereby affected. Much of the work of Abbott and colleagues and of Dudley mentioned above has been essentially nutritional, and these workers have confirmed that nitrogen balance may be maintained in the postinjury period by provision of adequate energy and nitrogen, intravenously if necessary, as has Peaston (Pl).However, Moore and Ball (M14) and Cuthbertson and Tilstone (C30) have pointed out that artificial feeding in the first week or so following injury is unnecessary except in cases where normal oral nutrition is impossible, as in comatose subjects, or where a lengthy period of artificial feeding is foreseen. In cases requiring artificial feeding, tube feeding is to be preferred, but if necessary, as in cases of paralytic ileus, adequate parenteral nutrition is possible for a time, using glucose or fructose, ethanol and fat emulsions for energy, and protein hydrolyzates for essential amino acids and additional nitrogen (C30). I n the stages of the body’s response to trauma which occur before or after the “flow” phase, we are still ignorant of the role of many of the nutrients. During the “ebb” phase Threlfall and Stoner (T2) have reported that fed rats have a greater resistance to ischemic injury than have fasted rats; tissue repair, although susceptible to dietary levels of ascorbic acid and trace minerals, does not appear to be much affected by levels of protein energy intake so long as frank malnutrition is avoided. We have shown that wound healing in the r a t is not affected by a moderate reduction in food (C29). Infectious diseases, however, can be influenced by dietary protein (S7).Kovach (K6) has recorded th a t rats fasted for 2 days were more susceptible to tumbling trauma or tourniquet shock than fed controls, but rats fasted for 4 days were more resistant than controls. 6.2. TRANSFUSION AND NOURISHMENT OF THE INJURED

This topic has recently been reviewed by us (C30). There is, virtually no problem with the patient who has not lost much tissue-and blood is a tissue-and who has been in a reasonably good nutritional state before trauma and whose appetite is good: nor should there be any worry if this is lacking for a few days. A well-balanced normal diet of 40-50 kcal/kg with some 12% of the total calories present as mixed proteins

26

D. P. CUTHBERTSON AND W. J. TILSTONE

and without any essential amino acid defect, is adequate; 1415% is possibly better for the more seriously injured, for those confined to bed, and for children. It is essential to ensure that the demands for energy supply are met though a short period of partial starvation should not jeopardize the outcome. Where a patient is unable to ingest, digest, or absorb sufficient food over a period which may jeopardize such full recovery as the injury will permit, the need obviously arises to provide nourishment by routes other than oral; the alternatives are tube feeding of homogenized total diets, either by the nasogastric route or through some form of “ostomy”; or by intravenous alimentation. Gastrostomy or jejunostomy is quite unnecessary for most illnesses of relatively short duration. But first the circulating blood volume must be restored. Intravenous feeding with protein hydrolyzates, fat emulsion, carbohydrate, minerals, vitamins, etc., is now relatively safe but should be reserved for conditions where the absorptive ability of the gut is seriously impaired. Probably less effective are mixtures of pure amino acids, fat carbohydrate, etc., for most of the D-isomers in them are an extra load. Such preparations are best reserved for those so seriously impaired in their capacity to digest and absorb suEcient food over a period as to jeopardize recovery, and particularly where there is paralytic ileus, induced or accidental. 7.

Environmental Factors

7.1. INTRODUCTION Many of the metabolic changes described above are changes in energy metabolism, and energy metabolism in uninjured animals is in turn influenced by environmental conditions, in particular, temperature, humidity, and wind velocity. Furthermore, environmental conditions may themselves act as noxious stimuli as in sunburn or frostbite. We have already mentioned some examples of the interaction of environment, specifically environmental temperature, in the body’s response to injury, and a more detailed account follows. 7.2. ENVIRONMENTAL STRESS Although a considerable amount of work has been reported on the effect of environmental conditions on bodily function, particularly with reference to thermoregulation (G5) and although there is a considerable literature on the body’s response to injury (e.g., see bibliography of this review), little work has been done on the interaction of environmental conditions and injury. However, studies on the effects of environmental stress are of relevance to the problem of the metabolic response to injury.

METABOLISM DURING THE POSTINJUBY PEBIOD

27

Hale et al. (H2) have studied the effect of long-term exposure to 3"5"C, 24"-26"C, or 34"-35°C on tissue growth in the rat. Only the middle temperature range is within the normal thermoregulatory range of environmental temperature for the rat. They found kidney, liver, spleen, and testes weights to be linearly and inversely related to environmental temperature, while thymus weight was linearly and directly proportional to environmental temperature. Adrenal weight was the same a t the lower and middle ranges, but decreased a t 34"-35°C. Heart weight was minimal and body weight maximal a t 24"-26°C. Heroux and Gridgeman (H10) compared tissue weights in rats adapted to 30°C or to 6"C, and found that cold adaptation had no effect on brain, genitalia, and lung weights, but reduced growth of muscle, pelt, fat, skeleton, spleen, and thymus. Liver, gut, kidney, heart, and adrenals were hypertrophied a t 6°C. Barnett and colleagues (Bl-B4) have studied the effect of exposure to -3°C or breeding at -3°C on the mouse. Oxygen consumption is increased and the insulation of skin plus hair increased in the cold, although cold-exposed mice have lighter-weight skins than the controls. First-generation mice born and bred a t -3°C have less body fat, nitrogen, and collagen and more body water than controls, and are of lighter body weight. The fortieth generation or so of mice born and bred a t -3°C are the same body weight as controls and do not have less body fat, nitrogen, nor collagen than controls. The principal metabolic response to cold exposure is increased heat production, with increased food and oxygen consumption, although nest-building is not unimportant. On theoretical grounds, one might expect cold exposure to result in increased tissue weight in liver, kidney, and gut, secondary to increased food consumption. Adrenal hyperplasia would not be surprising, and increased insulation of the pelt would be expected. This is apparently achieved mainly by increased fur growth, and the lighter-weight skin reported by Barnett on cold exposure may be due t o energy demands preventing peripheral fat deposition. Energy demands will also tend to keep down body weight, although the consequent increase in the body surface area to body mass ratio is disadvantageous for heat loss. At high environmental temperatures, food intake and oxygen consumption will fall, as will voluntary activity. Thus a reversal of the effect secondary to food intake noted for cold exposure may be expected. Whether or not adrenal activation occurs is of interest, since one tends to consider environmental temperatures other than those normally found in the individual's natural habitat to constitute a stress, but all environmental temperatures below body surface temperature will impose an obligatory heat loss on the animal, and may thus be considered stressful.

28

D. P. CUTHBEXtTSON AND W. J. TILSTONE

Body surface temperature in the rat is about 34OC, depending on skin blood flow, environmental temperature, and the interpretation of ((surface." The results of Hale e t d. (H2) that adrenal weight is lowest a t 34'35°C would support the above, insofar as adrenal weight may or may not reflect adrenal activity. Exposure to both cold (K3) and heat (34°C) (K4, K5) will lead to stimulation of corticosteroid secretion within 24 hours, but we have shown that exposure to 30°C results in plasma corticosterone levels below those of controls at 20°C in the rat (T4). Confirmation of some link between the stress of exposure to extremes of temperature and the stress of physical injury is given by the work of Blatt and Kerkay (B9), who found plasma protein changes similar to those described above as following physical injury when men were exposed for 3 weeks to temperatures of less than 10°C. Exposure to heat of almost 5O"C, however, produced very little change in plasma protein patterns, only a slight fall in globulin being noted. OF ENVIRONMENT AND STRESS 7.3. INTERACTION

The present writers interpret ('stress" as any situation giving rise to activation of the hypothalamic-pituitary-adrenal axis. It is wise to define the activating agent when it is known. Stress may thus be psychological or follow nonpsychological injury. One stress model is th a t of the subject's ability to swim without exhaustion, and, although no physical injury is necessarily involved, this model is of interest. Interaction between environmental temperature and swimming stress has been observed by Beaton and Feleki (B6) who found that the time for exhaustion to occur increased linearly with water temperature for swimming rats fed to a high protein or high carbohydrate diet. At 30°C the rats could swim for more than 150 minutes without exhaustion. This may appear t o be related to the obligatory energy demands (ie., heat loss) of the environment, but Costill e t al. (C14) found that when young men were made to swim for 20 minutes in water a t 17.4", 26.&", or 33.1"C their energy requirements were not affected by water temperature, but the extra expenditure involved in swimming is not a considerable proportion of the total. However, core temperature increase during exercise was positively related to water temperature. A powerful model for producing shock, that is, a period of depressed oxygen consumption, usually associated with some circulatory deficiency resulting in decreased oxygen transport, is limb ischemia produced by applying tourniquets or clamps to one or more limbs of an experimental animal. Allen (A3) has shown that immersion of the ligated limb in water at

METABOLISM DURING THE POSTINJURY PERIOD

29

temperatures a few degrees above normal room temperature greatly enhances the effects of tourniquet shock. On the other hand, reductions of temperature greatly reduce the dangers of ligation. Stoner (S14) studied the interaction between environmental temperature, cold acclimatization, and ischemic shock in the rat. At an environmental temperature of 18"-2OoC, rats show a fall in liver and brain temperature during the period (4 hours) of ischemia. Cold-acclimatized rats showed no changes in liver and brain temperature in this period. I n both groups liver blood flow fell toward the end of the period of ischemia, and consequent on this, arterial blood pressure fell, remaining a t about the critical level for autoregulation of hepatic blood flow until the rat was moribund. Raising the environmental temperature accelerated the fall in blood pressure and hepatic blood flow after limb ischemia, but the liver temperature rose. Lowering air temperature accentuated the fall in tissue temperature without altering the changes in liver blood flow after limb ischemia. The oxygen saturation of the portal vein blood fell after limb ischemia. Total oxygen consumption was unchanged during limb ischemia but fell when the tourniquets were removed a t 20°C and a t 30°C air temperature. Most of the oxygen consumed by an animal is used by its muscle, and in the presence of a fall in local temperature, oxygen supply to ischemic muscle is adequate for normal metabolic demands, a s evidenced by the persistence of a normal nucleotide pattern. However, a t an air temperature of 30°C, the fall in muscle temperature is prevented, and muscle chemistry then indicates inadequate oxygen supply. Haist (Hl) using ischemic shock in the rat as his injury model has shown that a n environmental temperature of 15°C during the period of ischemia is optimal for survival. Lowering the temperature to 9.5"C or raising i t to 40°C resulted in failure of the animal to recover from the ischemia. The animals were housed a t 27°C after the somewhat lengthy (10 hours) period of bilateral hindlimb ischemia. If the ischemia is carried out a t 27°C and the rats are exposed to environmental temperatures of 9.5"40°C after clamp release, then 24"-25.5"C is the optimal temperature range for survival, the mean survival time falling as temperature is raised or lowered from this. Survival a t relative humidity SZ-42% was the same as that a t relative humidity 7 1 4 2 % over the full temperature range studied. Haist further showed that exposure to environmental temperatures in the range SOo-32"C in the postischemic period did not affect hematocrit nor blood nonprotein nitrogen, both showing a uniform rise on clamp release. Blood inorganic phosphate also rose after clamp release, but here the rise was inversely related to environmental temperature. Liver gly-

30

D. P. CUTHBERTSON AND W. J . TILSTONE

cogen fell on clamp release, the fall being roughly constant a t all environmental temperatures, although the absolute values in controls and postischemic animals were approximately directly proportional to environmental temperature. Blood sugar levels tended t o mirror the survival results, being highest a t 24"C, and lowest 10" and 32°C. Tolerance for intravenously injected glucose wm not greatly different in shocked animals at 20°C or 1O"C, but glucose utilization was significantly greater in controls a t 10°C. Other workers have studied the effect of environmental temperature on the protein catabolic response to injury. The earliest report is that of You et al. ( Y 2 ) , who found that rats with burns lost more nitrogen in the urine a t an ambient temperature of 1.5"Cthan did similar animals a t room temperature. The increased urinary nitrogen response in the cold was not dependent on thyroid or adrenal stimulation. Caldwell and colleagues (C2, C3, M11) have investigated energy metabolism in rats with burns a t environmental temperatures of 20°, 28", and 30°C. I n experiments with a moderately restricted dietary regime, injury produced no increased oxygen consumption, heat loss nor urinary nitrogen excretion when a t an environmental temperature of 30°C. Deaths consequent on burning were also reduced. Ad libitum feeding tended to diminish the general response a t the lower temperatures. I n man, Barr et a2. (B5) have reported very favorable results in treating burns by blowing dry air a t 32°C over the patients. Oxygen consumption measurements indicated a reduction in the anticipated phase of increased general metabolism. The patients' general nutrition remained good. I n cases of long bone injury in man and the rat, Cuthbertson and his colleagues have shown that the protein metabolic response to injury is considerably reduced by exposure to 30°C (C4a, C5, C40); short-term thermoregulatory adaptation did not seem to be involved. As well as helping in the clinical treatment of patients with burns, the above work on the effect of exposure to 30°C offers an exciting experimental tool for investigation of the nature of the general metabolic response to injury. The work of Allen (A3), Haist ( H l ) , and Stoner (S14) using experimental animaIs indicates that, with an injury involving marked hemodynamic changes, including local sequestering of fluid, i t would be best to keep the patient a t normal room temperatures, and that exposure to warm air should be avoided; yet, as we have noted above, patients with burns respond favorably. Caldwell (C2) originally suggested that the diminution of the protein catabolic response to burns by rats kept at 30°C was due to environ-

METABOLISM DURING THE POSTINJURY PERIOD

31

mental compensation for the obligatory heat loss in fluid evaporation from the burn surface, but such an explanation is not applicable to bone injury, as Campbell and Cuthbertson have discussed (C5). Furthermore, if heat loss due to water evaporation from the burned surface were of critical importance, then warm humid air should be better than dry air, but all available evidence indicates that warm humid air should be avoided both for satisfactory clinical outcome and general comfort. Caldwell later suggested that the minimal thyroid activity in rats a t 30°C was an important factor since partitioned heat loss measurements indicated that the lower overall heat loss of rats with burns kept a t 30°C compared to rats with burns a t 28OC was due entirely to a lower dry heat loss. I n rats a t normal room temperature, he suggests that there is a thyroid-mediated increase in average body temperature on injury, resulting in increased heat production consequent on the van% Hoff effect of temperature. This Q,, effect is estimated to be responsible for about 50% of the total increased heat production consequent on injury. Direct measurement of thyroid activity after injury would indicate that there is insufficient increase to account for a significant proportion of the additional energy metabolism (G12, 53). It is by no means certain that the thyroid plays a causal role here, and Caldwell (C4) earlier reported work suggesting that a thyroid response is unimportant. The present authors (C30)have suggested that, in the rat, 30°C is probably about the highest environmental temperature which the rat can withstand without having to increase heat production due to hyperthermia ( G 5 ) , and that any nonobligatory heat production will be avoided. We thus conclude that, in spite of its widespread occurrence, the “flow” phase of the response to injury is not of great importance and is avoided in the face of a requirement for minimal heat production. Whether such an explanation is valid for humans, where behavioral activity such as selection of clothing is important in regulating heat loss, we do not know. Our first attempts a t using exposure to 30°C as an experimental tool involved experiments on wound healing. If the “flow” phase of the metabolic response to injury is of value to the organism, then it might be reasonable to assume that preventing the increased metabolism may prejudice wound healing. However, healing of skin wounds in the rat is accelerated by exposure of the animal to 30°C (C29). The reason for this is not clear. It may be due to increased peripheral blood flow (E5) or to the slight increase in skin temperature which results on exposure to 3OoC as compared to 20°C (C29). Heroux (H9) has shown that there is a linear relationship between the logarithm of induction rate and tissue temperature for various epithelia. I n the rat, the order of epithelial

32

D. P. CUTHBERTSON AND W. J. TILSTONE

mitotic activity is: duodenum > back > foot >ear. Furthermore, although most biological reactions have a Qlo of 2-3 (B16), Heroux found a Qlo of 10 for mitotic rate in rat epithelium. Experimental evidence of the importance of local temperature in skin wound healing has been given by work in which skin-graft donor sites in the human thigh were heated. A temperature of a t least 26°C was necessary for reepithelialieation to occur, and repair was faster as site temperature was raised to about 40"C, after which a sharp fall occurred, with necrosis at 44°C (G9). Thus our results on skin-wound healing, although interesting in themselves, could not be interpreted vis ii vis the metabolic response to injury, and so we decided to study healing in a nonsurface nonepithelial tissue, namely bone. Bone is a very difficult tissue in which to study repair, but using a combination of histology and radioisotope tracer studies, we concluded that bone healing was probably not affected by exposure to 30°C (C31, C32). One of the salient features of stress is an increased secretion of cortisol or corticosterone from very shortly after the moment of injury and lasting for 3 days or more, although plasma levels rise to a peak of about three times basal and return to normal within a few hours (52). We have found that, in the rat, increased corticosterone secretion, as measured by plasma levels 1 hour after injury, due to long bone fracture is not affected by housing the animals a t 30°C before and after injury (T4). Thus,in summary, the effects of environmental temperature on homeothermic animals include producing changes in surface temperature, possible changes in average body temperature, changes in heat-producing biochemical reactions, and changes in circulation. Studies on the responses of injured animals indicate that the toxic effects of ischemic shock are potentiated a t temperatures above or below 24"-25"C in the postinjury period. The heaIing of burns and other skin lesions, but probably not of deep tissue, is accelerated by exposure to 30°C. When the injured organism (man or rat) is transferred to an environment a t 30"-32"C the protein catabolic component of the ((flow" phase is absent or reduced. The corticosteroid rise in response to injury is not affected by exposing the animal t o 30°C. 8.

Control of the Metabolic Response to Injury

8.1. INTRODUCTION Injury, by physical or chemical agents, is a threat to homeostasis. It will produce a temporary upset in the otherwise well-controlled functioning of many major physiological and biochemical systems, and unless adequate regeneration of the damaged tissue is effected, may result in

33

METABOLISM DURING THE POSTINJURY PERIOD

regulator I-- --------1Mediator

1

I

I_ _ _ _ _ _ _ _ _ _ _ 1

FIG.1. Schematic representation of pathways involved in homeostatic regulation. some degree of permanent malfunction in the injured area. Some metabolic functions remain altered for several months after injury, in spite of apparently full recovery in other ways, as for example in bone mineral dynamics after fracture (W10). A general picture of the results of a traumatic stimulus is shown in Fig. 1. The stimulus acts on the organism via a receptor (or receptors), which then stimulates some effector which in turn activates the homeostatic regulator via its own transmitter. The homeostatic regulator may be activated more directly by a forward transmitter. The concept of “homeostatic regulator” includes compensatory changes in its target tissue(s) or organ(s) via some mediator, and holds equally for immediate changes such as the “fight or flight” effects of adrenaline and nonadrenaline, and for long-term changes involved in regeneration of tissue and function. An example, in another context, is the stimulus of raised blood pressure which will act on the stretch-sensitive baroreceptors in the carotid sinus and thence via the direct neural link (receptor transmitter) in the glossopharyngeal nerve to the effector in the medulla of the brain. The effector output is stimulation of the cardioinhibitory center there, which acts on the homeostatic regulator-the heart-via the effector transmitter, which is the vagus in this case. The effective homeostatic regulator is heart rate, the mediator being acetylcholine liberated by vagal stimulation, and slowing the beating of the target organ, the heart. The “short-cut” action of the forward transmitter is seen in, for ex-

34

D. P. CUTHBERTSON AND W. J. TILSTONE

ample, the flexion reflex, where we may consider the complete nervous arc which is not controlled by higher centers as the forward transmitter acting between the pain receptors and flexor muscle target organ, the act of withdrawal of the limb being the homeostatic regulator.

RESPONSE TO 8.2. T H E ENDOCRINE

INJURY

The various transmitters and mediators in this scheme-and the metabolic response to injury involves a multiplicity of effectors and homeostatic regulators-are neural or hormonal; the Iatter including “local hormones” such as bradykinin. Hormones may also be involved as effectors. One of the salient features of “stress,” whether or not accompanied by physical injury, is an increased secretion of the hormones of the adrenal cortex. Much work has been done on this field since the early studies of Browne ( B B ) on the excretion of “cortin” in the urine after trauma. I n man, free plasma cortisol levels rise after a number of stresses (Y1) and, in the case of surgery, reach a peak of about five times the basal secretion about 5-6 hours after the operation returning to basal levels in another 5 hours or so. The increased plasma levels reflect increased secretion of cortisol and an increased removal half-time from plasma probably due to depressed liver blood flow (52,54, Y l ) . Urinary excretion of cortisol metabolites, usually measured as 17hydroxycorticoids, is increased for 3-4 days after surgery, rising from about 3 mg/day basal t o about 18 mg/day 3 days after operation (M13, P2). It is thus evident that cortisol secretion is increased for a few days after physicaI injury, and th at plasma levels are elevated for the first few hours of this time, probably because of deficiencies in metabolism of the hormone. The early increases in plasma cortisol have been found to parallel plasma levels of ACTH independently assayed (C12). Posttraumatic secretion of aldosterone has also been measured. Increased quantities were found to be present during surgery, and in the urine subsequently (D3, H17, Z l ) . Also, the aldosterone antagonist spironolactone will abolish postoperative sodium retention without affecting potassium secretion, illustrating the dependence of sodium displacements after surgery on the hormone (52, MS) . Holzbauer (H13) has investigated the part played by ACTH in stimulating aldosterone secretion during operative stress. She found a fall in cortisol and in aldosterone in adrenal venous blood up to 2 hours after hypophysectomy. The fall in aldosterone was much less marked than that of cortisol. Both cortisol and aldosterone secretion showed a linear log dose-response relation to infusion of ACTH, although the response of cortisol reached a plateau after injection of 0.3 mU/min of ACTH

METABOLISM DURING THE POSTIN J U R Y PERIOD

35

per kilogram of body weight, whereas aldosterone secretion was linear up to 3.0 mU/min/kg. Operative stress caused secretion between 0.03 and 0.3 mU/min/kg. It would appear that ACTH plays an important part in the response of cortisol secretion to physical injury, but that aldosterone secretion is more liable to be influenced by other factors, including the renin-angiotensin system (SlOa). Jouan and Samperez (J6) have reported that 5hydroxytryptamine has a specific action in stimulating aldosterone secretion. One of the posterior pituitary hormones-antidiuretic hormone (ADH, vasopressin)-may play a part in the metabolic response to injury. The reduced blood volume and increased plasma electrolyte concentration found after trauma may be expected to stimulate ADH secretion and the oliguria or anuria often found to follow injury would suggest this is so (52) * Another tropic hormone of the anterior pituitary which has been investigated in the response to injury is growth hormone (somatotropic hormone, STH). Emotional stress or injection of pyrogen will stimulate STH secretion in men. The response is associated with adrenocorticol activity but not hypoglycemia, and is not suppressed by induced hyperglycemia (G22). Although STH is not necessary for the protein catabolic phase of the metabolic response t o injury to occur (hypophysectomy has no effect) low levels of the hormone have been found in the pituitaries of patients who died a few days after surgery. This may be an attempt to circumvent protein loss (G6). Gemzell (G6), and Cuthbertson et ul. ((338) have shown th at STH administration will prevent the loss of body nitrogen consequent on injury, but it will not accelerate healing of superficial skin wounds in the rat ((39). The role of the catecholamines, adrenaline, and noradrenaline, in the “fight or flight” response is well known and the possible connection between enhanced sympathetic and adrenal medullary activity and the vascular changes, tachycardia and early hyperglycemia after injury is apparent. An increased urinary excretion of catecholamines is known to occur for several days following injury (C16, G13). Analysis of adrenal vein blood has shown that anesthesia, hemorrhage, and injury can all stimulate adrenaline production (W5). Coward and Dunlop (C17) found four catecholamines to be excreted in increased quantities after surgery, viz., metanephrine, normetanephrine, N-methylmetanephrine, and 3-methoxytyramine, The N-methylmetanephrine probably reflects increased adrenaline formation, and the 3-methoxytyramine levels probably represent the increased homovanillic acid secretion after stress. Of the metanephrine and normetanephrine, the lat-

36

D. P. CUTHBERTSON AND W. J. TILSTONE

ter increase is more marked, suggesting active participation of the sympathetic nervous system as well as the adrenal medulla. Moore (M13) could find no consistent change in plasma a.drenaline or noradrenaline levels after trauma. Adrenocorticoid and catecholamine activities are interrelated. Adrenaline and noradrenaline are required in the injured for the sensitization of many tissues to cortisol. Both medullary and cortical hormones are required for the maximal vascular response to blood loss, and adrenaline may be involved in stimulating ACTH release ( R l ) . Adrenaline a t a dose of 80 p g can activate purine metabolism and markedly increase total nitrogen excretion in rats. This action is not blocked by ergot alkaloids, and the opposite effect is produced by acetylcholine (G17, G18). Adrenaline and noradrenaline both have an action in promoting chemical thermogenesis. Spoelstra (S13) has shown this to occur after low doses of drug and to be apparently independent of the changes in cardiac output and hyperglycemia. Sandor (Sl) has investigated the effect of adrenaline on plasma proteins in the guinea pig. He found a rapid lowering of plasma albumin, with a frequent concomitant increase in a-globulin levels. Although stimulation of ACTH may be involved he believes this to be largely a direct metabolic effect of the injected adrenaline. Glucose metabolism in the “ebb” phase of the response to traumaa raised blood glucose and reduced tissue glucose utilization-are similar to those expected in insulin deficiency. However, direct measurement of plasma insulin by an immunological method shows plasma insulin levels to be increased after moderately severe surgery (R9). 8.3. THEPERMISSIVE ROLEOF CORTISOL

From the above outline of some of the endocrine changes consequent on physical injury, including the discussion of thyroid activity in Section 7.3, the overriding impression which one has is that of a blanket stimulation of endocrine secretions. Apart from catecholamine secretion, it is not obvious what role increased endocrine activity would have in homeostasis after injury. In adrenalectomized dogs subjected to hemorrhagic injury, cortisol and aldosterone will each cause an expansion of extracellular fluid volume by mobilization of cell water within 6 hours of administration. Cortisol, but not aldosterone, also will cause an expansion of plasma volume which can be attributed to its effects on the peripheral vasculature. Spironolactone will prevent the shift of water from the body cell mass to the interstitial fluid (M3). The above experimental evidence clearly im-

METABOLISM DURING THE POSTINJURY PERIOD

37

plicates a causal role of two other hormones, cortisol and aldosterone, in some of the metabolic consequences of trauma, but, unlike the catecholamines, the value of the resulting metabolic changes is not clear. The role of the adrenal gland in various responses to various stresses has been widely studied. John (Jl) found that sound stress for a period of 1.5 hours produced marked lymphocytopenia in normal but not adrenalectomized rats. Estimating the normal daily in vivo production of corticoid to be equivalent to 0.8 pg of cortisone acetate per kilogram body weight injected subcutaneously as an aqueous suspension, Johns noted that injection of 4.0 mg of cortisone acetate per 100 g body weight into adrenalectomized rats resulted in lymphocytopenia. Injection of doses of cortisone acetate more equivalent to the normal daily secretion of corticoid in intact animals did not produce lymphocytopenia in adrenalectomized rats, but did restore the lymphocytopenic response to sound stress. Weimer and colleagues have studied the response of the az-acute phase globulins of rat serum to trauma, and the interaction of adrenal function with this response (W7-W9). The response of the various plasma protein fractions to injury in the rat can be differentiated according to the effects of repeated challenge by turpentine injection. The a2-acute phase globulins will rise progressively with each challenge, giving a very rapid rise then fall on each occasion. Fibrinogen behaves similarly. The seromucoid fraction did not fall rapidly after injection of turpentine and gave a maximal response on each occasion of challenge. The presence of functioning adrenals does not have the same effect on the response of all the plasma protein fractions. Adrenalectomy will lead to a reduced response of the az-acute phase globulins and also of the seromucoid fraction to tissue injury, but is without effect on the fibrinogen response. Treating adrenalectomized rats with 0.15 mg/100 g body weight of cortisol will partially restore the response of the cr2-acutephase globulins to turpentine injury; a dose of 0.3 mg/100 g gave complete restoration of the response, but doses of 0.6 mg/100 g or 3.0 mg/100 g were inhibiting. Also, administration of cortisol to intact animals was inhibiting for the w a c u t e phase globulin response to challenge. Cortisol will increase gluconeogenesis, that is, formation of glucose from protein with loss of the nitrogenous moiety in the urine. A 25 mg subcutaneous implant of cortisone gave a response similar t o that obtained after fracture in the rat when urinary nitrogen was examined (C7). Rats with 50-mg implants excreted about twice as much nitrogen in the urine as those with a 25-mg implant and rats with a 25-mg implant fracture of one femur excreted about the Same amount of nitrogen as the

+

38

D. P. CUTHBERTSON AND W. J. TILSTONE

sum of that in rats with a 25-mg implant alone plus that in rats with fracture alone. Saline-maintained adrenalectomized rats showed no urinary nitrogen increase after fracture, but adrenalectomized animals receiving maintenance doses of cortisone produced the characteristic response to fracture. A more recent paper by Cuthbertson and co-workers (C6) investigated the interaction between oral doses of p-methasone and fracture of the femur in the rat. Heat production was also measured. As far as urinary nitrogen was concerned both treatments produced a characteristic increased output-that from the drug showing a rapid rise and fall whereas fracture gradually reaches a peak 3-4 days after treatment. Urinary nitrogen excretion in rats given fracture and drug together produced a pattern which was the sum of the individual treatments. Measurements of heat production showed that only the fracture treatment caused an increase, and the authors suggested that after fracture there is gluconeogenesis and subsequent glucose oxidation from the deaminated amino acids, whereas after steroid the deaminated moiety is excreted probably largely unchanged in the urine. Nims and Thurber (N6) found that urinary nitrogen excretion by the rat is increased after whole-body X-irradiation. Adrenalectomy suppresses the response, but administration of adrenocortical extract restored it. Weights (dry weight and wet weight) of thymus and spleen fell and t ha t of liver rose after whole-body X-irradiation of intact rats. Adrenalectomy produced only a slight qualitative change in the response of organ weight. The authors concluded that after X-rays, tissues are broken down to release nitrogenous fractions greater than amino acids or simple polypeptides, and the breakdown products are transported to the liver. This process is not dependent on functioning adrenals. I n intact animals the liver then metabolizes the polypeptides. I n adrenalectomized animals there is no breakdown of the polypeptides by the liver. Distinction between the effects of injury on protein metabolism and the effects of corticoid on protein metabolism has been made by Munro (M17, M18) who has found a marked gain in liver nitrogen content following cortisone administration to the rat, but not after femur fracture. H e also notes that cortisone has a constant action on nitrogen balance a t all levels of protein intake whereas the effect of injury is obliterated by prior protein depletion. Evidence that corticoids do not play a causal role in a t least the protein catabolic response to injury has been provided by experiments in which plasma corticosterone levels have been measured in rats housed a t 20°C or a t 3OoC, and in which a fracture of femur was induced under halothane anesthesia, or the animals given anesthesia alone, or no treatment (T4).

METABOLISM DURING THE POSTINJURY PERIOD

39

As we have already noted, exposure to an ambient temperature of 30°C will largely inhibit the protein catabolic response to injury. Rats a t both temperatures showed a maximal response t o fracture of femur when plasma corticosterone was measured 1 hour postinjury, in spite of lower basal levels in the rats a t 30°C. It is interesting to note that there is increased secretion of corticoid after trauma in the face of elevated plasma levels. Some confusion over the mechanisms controlling plasma corticoid has arisen because of this, and two control mechanisms have been proposed, a closed-loop control with negative feedback, in nonstress conditions; and an open-loop control with no feedback, in stress. However, Yates and Urquhart (Yl) point out the most likely control system is a closed-loop, negative feedback proportional control, with a variable set point, the set point being raised in stress. Why the set point should be raised (or the type of control changed) in stress when all the evidence indicates a noncausal role for corticoid in the various other responses to stress is a very interesting and as yet unanswered question. 8.4. INITIATING PATHWAYS

It is very difficult to plan experiments to study pathways involved in control of the metabolic response to injury since some surgical or a t least stressful preparation of the experimental animal before administration of the injury is almost always necessary. Early work by Gordon (G14) reported that severe scalds could cause ascorbic depletion of the rat adrenal even when the scalded limb was completely deafferentated. This suggests adrenal stimulation or ACTH release due to a humoral agent. We have already mentioned (Section 8.2) that 5-hydroxytryptamine may have a specific action in stimulating aldosterone secretion (J6) , and Johnston (52) has postulated that other local hormones released after injury, such as histamine and bradykinin, may lead to direct adrenal stimulation. Most evidence presently available, however, implicates a direct neural link from the site of injury to the CNS. Much of this, work has been done by Egdahl and colleagues, and he has recently reviewed their contribution (E3, E4). Egdahl’s model is a dog with a n isolated hind limb leaving only a n artery, vein, and the sciatic nerve intact. The dogs are surgically prepared for adrenal venous blood collection without subsequently disturbing the animal. A burn injury to the isolated limb will provoke a rise in adrenal venous skeroid content only if the nerve is intact. Severing vascular contact is without effect. Using leg fracture in the rat as an injury model, Matsuda and colleagues (M9) have shown a direct neural link to be necessary for induc-

40

D. P. CUTHBERTSON AND W. J . TILSTONE

tion of ACTH release by the stress. No increased adrenal secretion is produced by fracture if the spinal cord is transected proximal to the injured limb, but the adrenal response is unimpaired if the cord transection is distal to this. The afferent connections to the hypothalamus appear to pass through the dorsal mesencephalon. A large hypothalamic or thalamic peninsula in which the hypothalamic connections to the dorsal mesencephalon are maintained permits a normal adrenocorticoid response. Adrenal secretion was high in all animals anesthetized with ether and was not further elevated by leg break. Ether probably caused release of corticotropin-releasing factor from the median eminence. Laparotomy for adrenal vein cannulation and subsequent blood collection were carried out rapidly. I n the injured animals about 11 minutes elapsed between fracture and laparotomy. Adrenal venous blood from uninjured controls always gave lower corticosterone levels than that from rats with fractures in spite of the stress of laparotomy. This may have been due to a time factor, since maximal corticosterone secretion following intravenous injection of ACTH in hypohysectomized rats is not seen until about 10 minutes later. Using electric shock as a stressor, Redgate (R2) directly measured ACTH release in adrenalectomized rats. Adrenalectomy was performed 12 days before stressing. Forepaw stimulation and hindpaw stimulation were equally effective in producing a 6-fold increase in circulating ACTH. Spinal cord transection a t thoracic-2 inhibited stimulation of ACTH release consequent on hindpaw electric shock, but had no effect on ACTH secretion following forepaw shock. The ability to secrete ACTH following stress is thus demonstrated to be unaffected by the surgical procedures of the experiment and to be dependent on a direct neural link between the injury site and the central nervous system. Direct hormonal stimulation of the adrenals was, of course, not discounted by this work. The evidence for a direct neural link causing ACTH release following chemotoxic stress is not too clear. Makara et al. (M6) using thoracic-2 spinal cord transection in the rat has shown that formalin injection into a limb caused ACTH secretion by a mechanism involving nervous pathways, but capsaicin injections could stimulate ACTH release even when injected to sites distal to the cord transection. They also noted that repeated formalin injection did not produce tachyphylaxis, whereas capsaicin did, but only for ACTH release due to capsaicin. Repeated injections of capsaicin did not affect the ability of subsequent injections of formalin to stimulate ACTH secretion. I n conclusion, injury leads to a blanket stimulation of endocrine secretion. This increased activity, with a few notable exceptions, does not appear t o play a causal role in subsequent metabolic responses to trauma

METABOLISM DURING THE POSTINJURY PERIOD

41

--only a permissive role in certain circumstances. The initial pathways involved in provoking endocrine activity are largely but not exclusively neural in nature. 9.

Summary

The nature and the severity of a physical injury, whether arising by accident, radiation, or elective surgery, conditions the character of both the immediate and subsequent biochemical changes that occur both locally and generally. The subject’s nutritive state both a t the time of injury and in the early postinjury period, particularly in respect to labile protein reserves and glycogen, also affects the scale of those changes, as does restitution of lost blood or plasma. There is first an “ebb” and then Ylow” of metabolic activity in those who recover from a moderate to severe injury. Finally, environmental temperature a t the time of injury and during the days immediately following the injury also affects the issue. It is probably in this last area that the most interesting recent observations have been made. Studies on energy exchange, and on protein as the constituent of the body most involved, reveal an interesting fundamental difference in homeostatic response when the energy demands of the injured are minimal as in the zone of thermal neutrality. When a patient is maintained a t an ambient temperature close to his superior critical temperature, above which hyperthermy begins, then not only will his energy needs be minimal, but he will also resist unnecessary endogenous processes which result in heat production. I n an atmosphere of 30°C and with a relative humidity of 1 5 4 5 % the increased urinary loss of nitrogen and presumably of other nitrogenous and sulfur-containing catabolites of protein metabolism, which has been termed the “protein catabolic response to injury, is largely eliminated and the related increase in resting heat production is correspondingly reduced. The organism also compensates for the extra load by decreased food intake a t the higher temperature. Whatever the reason for the additional protein catabolism that usually follows physical injury it is clearly not sufficiently pressing to be obligatory in patients a t high environmental temperatures, and a conservative approach to nutritional therapy for those who are not suffering from defects in alimentation is indicated. The response of the adrenal cortex to trauma is not inhibited a t an environmental temperature of 30°C. Experimentally there is evidence of accelerated healing of superficial tissues a t 30°C and no untoward effects have been seen clinically. Indeed, warm dry air assists in the healing of burned patients. Obviously patients previously depleted in protein through malnourishment or undernourishment though a t a disadvantage when reserves

42

D. P. CUTHBERTSON AND W. J. TILSTONE

have to be called on for healing, should be assisted by hospitalization a t such higher temperatures, providing they are not suffering from injury to the brain or have other conditions countermanding such treatment. The immediately early and later effects of the response to trauma in experimental animals and in patients housed a t normal temperatures and which are all part of the “inflammatory response,” and the relation of these to such changes in the level of endocrine secretions as occur, are described. The time factor in the response to multiple injuries is also considered. ACKNOWLEDQMENT The senior author (D.P.C.) wishes to express his indebtedness to the Mediral Research Council for a personal Research Grant, during the tenure of which this joint paper was prepared.

REFERENCES A l . Abbott, W. E., and Albertsen, K., The effect of starvation, infection and injury on the metabolic processes and body composition. Ann. N . Y . A d . Sn’. 110 941-964 (1963). A2. Aldridge, W. N., and Stoner, H. B., The behaviour of liver mitochondria isolated from rats with differentbody temperatures after limb ischaemia or after injection of 3,5-dinitro-o-cresol. Biochem. J . 74, 148-154 (1960). A3. Allen, F. M., Resistance of peripheral tissues to asphyxia at various temperatures. Surg., Gynecol. Obstet. 67, 746-751 (1938). A4. Altura, B. M., and Hershey, S. G., Structure-activity relationships of neurohypophyseal polypeptides in the micro-circulation. I n “Combined Injuries and Shock” (B. Schildt and L. Thorbn, eds.), Intermedes Proceedings, pp. 185-194. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. A5. Arturson, G., Serum isozyme variations following different types of injuries. I n “Combined Injuries and Shock” (B. Schildt and L. Thorbn, eds.), Intermedes Proceedings, pp. 133-138. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. B1. Barnett, S. A., Endothermy and ectothermy in mice at -3°C. J . Exptl. Biol. 33, 124-133 (1956). B2. Barnett, S. A., The skin and hair of mice living at a low environmental temperature. Quart. J. Exptl. Physiol. 44, 35-42 (1959). B3. Barnett, S. A., Coleman, E. M., and Manly, B. M., Oxygen consumption and body fat of mice living at -3°C. Quart. J . Exptl. Physiol.44,43-51 (1959). B4. Barnett, S. A., and Widdowson, E. M., Organ weights and body composition in mice bred for many generations at -3°C. Proc.Roy. SOC.B162,502-516 (1965). B5. Barr, P. O., Birke, G., Liljedahl, S-O., and Plantin, LO.,Oxygen consumption and water loss during treatment of burns with warm dry air. Lancet I, 164-168 (1968). B6. Beaton, J. R., and Feleki, V., Effect of diet and water temperature on exhaustion time of swimming rats. Can. J . Physiol. Phamzacol. 46, 360-363 (1967). B7. Belfrage, S., Plasma protein pattern in course of acute infective disease. Acta Med. Scand. Suppl. 395 (1963). B8. Birke, G., Carlsson, L. A., and Liljedahl, S-O., Lipid metabolism and trauma. 3. Plasma lipids and lipoproteins in burns. A& Med. Scand. 178,337-350 (1965).

METABOLISM DURING THE POSTIN JURY PERIOD

43

B8a. Bjornesjo, K. B., Werner, I., and Odin, L., The influence of surgery on serum and urine hexosamine, serum mucoprotein, glutamic-oxaloacetic transaminase (GOT) and C-reactive protein. Scand. J . Clin. & Lab. Invest. 11, 238-244 (1959). B9. Blatt, W. F., and Kerkay, J., The effect of repeated heat and cold exposure on serum protein composition in man. Can. J. Physiol. Pharmacol. 46, 511-516 (1967). BlO. Bogden. A. E., and Gray, J. H., Glycoprotein synthesis and steroids. 1. Relationship to trauma, cortisol administration and LU-Z-GPsynthesis. Endocrinology 82, 1077-1084 (1968). B11. Bogden, A. E., Gray, J. H., and Rigiero, C. S., Glycoprotein synthesis and steroids. 111. An immunoassay for the anti-inflammatory a c t i d y of glucocorticoids. Endocrinology 82, 1093-1097 (1968). B12. Bossak, E. T., Wnng, C., and Adlersberg, D., Effect of cortisone on plasma Exptl. Biol. globulins in the dog. Studies by paper electrophoresis. Proc. SOC. Med. 88, 634-636 (195.5). B13. Bottinger, C. E., and Eklund, A. E., Studies on serum glycoprotein aft,er surgical operations. Acta Chir. Scand. 118, 349-352 (1960). B14. Bowman, H. S., Bcqnired fibrinogenopenia. Am. J . Med. 24, 967-973 (1958). B15. BoioviE, J., Castenfors, J., Kayser, L., and Liljedahl, S-O., Plasma renin activity in patients during and after surgical intervention. I n “Combined Injuries and Shock” (B. Schildt and L. Thorbn, eds.), Intermedes Proceedings, pp. 143-160. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. B16. Brody, S., “Bioenergetics and Growth,” Chapter 11and 13. Reinhold, New York, 1945. B17. Brooks, D. K., Williams, W. G., Manley, R. W., and Whitman, P., Osmolar and electrolyte changes in haemorrhagic shock. Hypertonic solutions in the prevention of tissue damage. Lancet I, 521-527 (1963). B18. Browne, J. S. L., I n “Conference on Bone and Wound Healing, Second Meeting,” pp. 43-47. Josiah Macy, Jr. Found., New York, 1942. B19. Browne, J. S. L., Schenker, V., and Stevenson, J. A. E., Some metabolic aspects of damage and convalescence. J. Clin. Invest. 23, 932 (1944). B20. Bull, J. P., Nitrogen balance after injuries. Proc. Nutr. SOC.(Engl. Scot.) 17, 114119 (1958). B21. Burt, R. L., The utilization of amino acid nitrogen following trauma. J. Lab. Clin. Med. 44, 702-709 (1954). C1. Cairnie, A. B., Campbell, R. M., Cuthbertson, D. P., and Pullar, J. D., The heat production consequent on injury. Brit. J . Exptl. Patho2. 38, 504-511 (1957). C2. Caldwell, F. T., Metabolic response to thermal trauma. 11. Nutritions1 studies on rats a t two temperatures. Ann. Surg. 166, 119-126 (1962). C3. Caldwell, F. T., Hammel, H. T., and Dolan, F., A calorimeter for simultaneous determination of heat production and heat loss in the rat. J. Appl. Physiol. 21, 1665-1671 (1966). C4. Caldwell, F. T., Osterholm, J. L., Eower, N. D., and Moyer, C. A., Metabolic response to thermal trauma of normal and thyroprivic ratv at three environmental temperatures. Ann. Surg. 160, 976-988 (1959). C4a. Campbell, R. M., and Cuthbertson, D. P., Effect of environmental temperature on the metabolic response to injury. Nature 210, 206-208 (1966). (25. Campbell, R. M., and Cuthbertson, D. P., The effect of environmental temperature on the metabolic response to injury. Quart. J. Exptl. Physiol. 62, l l P 1 2 9 (1967).

44

D. P. CUTHBEBTGON AND W. J. TILSTONE

C6. Campbell, R. M., Cuthbertson, D. P., and Pullar, J. D., The effects of betamethasone and fracture on nitrogen metabolism. Quart. J . Ezptl. Phy&oZ. 49, 141-150 (1964). C7. Campbell, R. M., Sharp, G., Boyne, A. W., and Cuthbertson, D. P., Cortisone and the metabolic rcsponse to injury. Brit. J . Exptl. Pathol. 36, 566-576 (1954). C8. Chanutin, A., and Gjessing, E. C., Electrophoretic analyses of sera of injured dogs. J. Biol. Chem. 166, 421-428 (19461. C9. Chytil, F., Injury and protein metabolism. Cesk. Gastroenterol.Vyziva 10,244-248 (1956). C10. Clark, R. G., Surgical nutrition. Brit. J . Surg. 64, 445-459 (1967). C11. Collins, D. A., and Hamilton, A. S., Changes in renin-angiotensin system in haemorrhagic shock. Am. J. Physwl. 140, 499 (1944). C12. Cooper, C. E., and Nelson, D. H., ACTH levels in plasma in preoperative and surgically stressed patients. J . Clin. Invest. 41, 1599-1605 (1962). C13. Cope, O., Nathanson, I. T., Rourke, G. M., and Wilson, H., Symposium on the management of Coconut Grove burns at Massachussetts General Hospital: Metabolic observations. Ann. Surg. 117, 937-958 (1943). C14. Costill, D. L., Cahill, P. J., and Eddy, D., Metabolic response to submaximal exercise in three water temperatures. J. A p p l . Phgsiol. 22, 628-632 (1967). C15. Co Tui, Wright, A. M., Mulholland, J. H., Barcham, I., and Breed, E. S., The nutritional care of cases of extensive burns with special reference to the use of oral amino acids (Amigen) in 3 cases. Ann. Surg. 119, 81fr823 (1944). C16. Coward, R. F., and Smith, P., Recovery of metanephrines from urine and their estimation by paper chromatography. Clin. Chim. A& 14, 672-678 (1966). C17. Coward, R. F., and Smith, P., Excretion of metanephrines in post-operative stress. Clin. Chim. Acta 14, 832-833 (1966). C18. Cuthbertson, D. P., The influence of prolonged rest on metabolism. Biocliem. J. 23, 1328-1345 (1929). CIS. Cuthbertson, D. P., The disturbance of metabolism produced by long bone and non-bony injury, with notes on certain abnormal conditions of bone. Biochem. J . 24, 1244-1263 (1930). C20. Cuthbertson, D. P., The distribution of nitrogen and sulphur in the urine during conditions of increased catabolism. Biochem. J . 26, 236-244 (1931). C21. Cuthbertson, D. P., Observations on disturbance of metabolism produced by injury to the limbs. Quart. J . Med. 26, 233-246 (1932). C22. Cuthbertson, D. P., Further observations on the disturbance of metabolism caused by injury with particular reference to the dietary requirements of fracture cases. Brit.J . Surg. 23, 505-520 (1936). C23. Cuthbertson, D. P., Post-shock metabolic response. Lancet. I, 433-437 (1942). C24. Cuthbertson, D. P., Injury and its effects on protein metabolism. Proc. 11th Intern. Cong. Pure Appl. C h . ,London, 1947 Vol. 111, pp. 461465. Hepworth & Co., London, 1950. C25. Cuthbertson, D. P., The metabolic changes consequent on injury. 2. Tierernuhr. Futtermittellc. 12, 259-314 (1957). C26. Cuthbertson, D. P., The disturbance of protein metabolism following physical injury. In. “The Biochemical Response to Injury” (H. B. Stoner and C. J. Threlfall, eds.), pp. 193-216. Blackwell, Oxford, 1960. C27. Cuthbertson, D. P., Physical injury and its effect on protein metabolism. Zn “Mammalian Protein Metabolism” (H. N. Munro and J. B. Allison, eds.), Vol. 2, pp. 373-414. Academic Press, New York, 1964. C28. Cuthbertson, D. P., and Munro, H. N., A study of the effectof overfeeding on

METABOLISM DURING THE POSTINJURY PERIOD

45

the protein metabolism of man. 3. The protein-saving effect of carbohydrate and fat when superimposed on a diet adequate for maintenance. Bwchem. J . 31, 694-705 (1937). C29. Cuthbertson, D. P., and Tilstone, W. J., Effect of environmental temperature on the closure of full-thickness skin wounds in the rat. Quart. J. Exptl. Physiol. 62, 249-257 (1967). C30. Cuthbertson, D. P., and Tilstone, W. J., Nutrition of the injured. Am. J . Clin. NUB. 21, 911-922 (1968). C31. Cuthbertson, D. P., and Tilstone, W. J., The effect of environmental temperature on healing of bone lesions in the rat. I. Effect of environmental temperature on mineral metabolism. Quart. J. Ezptl. Physiol. 63, 422-427 (1968). C32. Cuthbertson, D. P., and Tilstone, W. J., The effect of environmental temperature on healing of bone lesions in the rat. 11. The effect of bone injury on mineral metabolism at 20°C and 30°C. Quart. J. Expll. Physiol. 63, 428-436 (1968). C33. Cuthbertson, D. P., and Tilstone, W. J., Metabolic changes following different types of injuries and their relation to diet and environmental temperature. I n “Combined Injuries and Shock” (B. Schildt and L. Thorbn, eds.), Intermedes Proceedings, pp. 99-114. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. C34. Cuthbertson, D. P., and Tilstone, W. J., The metabolic and hormonal response of the body to injury. In “Pathology of the Human Adrenal Gland” (T. Symington, ed.), Part 1. Livingstone, Edinburgh and London, 1969 (in press). C34a. Cuthbertson, D. P., and Tompsett, S. L., Note on the effect of injury on the level of the plasma proteins. Brit. J. Ezptl. Pathol. 16, 471475 (1935). C35. Cuthbertson, D. P., McCutcheon, A., and Munro, H. N., A study of the effect of overfeeding on the protein metabolism of man. 1. The effect of superimposing raw and boiled milks on a diet adequate for maintenance. Biochem. J. 31, 681-688 (1937). C36. Cuthbertson, D. P., McCutcheon, A., and Munro, H. N., A study of the effect of overfeeding on the protein metabolism of man. 2. The superimposition on a diet adequate for maintenance of beef (or soya flour) plus lactose plus butter equivalent in protein, carbohydrate and fat content to a litre of milk. Biochem. J. 31, 689-693 (1937). C37. Cuthbertson, D. P., McGirr, J. L., and Robertson, J. S. M., The effect of fracture of bone on the metabolism of the rat. Quart. J. Exptl. Physiol. 29, 18-25 (1939). C38. Cuthbertson, D. P., Shaw, G. B., and Young, F. G., The anterior pituitary extract on the metabolic response of the rat to injury. J . Endomhol. 2, 468-474 (1940-1941). C39. Cuthbertson, D. P., Shaw, G. B., and Young, F. G., The anterior pituitary gland and protein metabolism. 3. The influence of anterior pituitary extract on the rate of wound healing. J . Endocrinol. 2, 475-478 (1940-1941). ( 3 0 . Cuthbertson, D. P., Smith, C. M., and Tilstone, W. J., The effect of transfer to a warm environment (30°C) on the metabolic response to injury. Brit. J . Surg. 66, 513-516 (1968). Dl. Davies, J. W. L., Ricketts, C. R., and Bull, J. P., Studies of plasma protein breakdown. 1. Albumin in burned and injured patients. Clin. Sci. 23, 411423 (1962). D2. Davies, J. W. L., Ricketts, C. R., and Bull, J. P., Studies of plasma protein metabolism. 11. Pooled a-globulin in burned and injured patients. Clin. Sn‘. 24, 371-382 (1963). D3. Davis, J. O., Urquhart, J., Higgins, J. T., Rubin, E. C., and Hartcroft, P. M.,

46

D. P. CUTHBERTSON AND W. J. TILSTONE

Hypersecretion of aldosterone in dogs with a chronic aortic-caval fistula and high output heart failure. Circulation Res. 14, 471-485 (1964). D4. Deadrich, R. E,, Morico, J. I., Logans, E. R., Tansig, F., Hammel, R. P., and Landatil, S. F., Serum protein changes in thermal trauma. Clin. Chem. 2, 266 (1956). D5. Declasse (1834). Quoted by Cuthbertson, D. P., Post-shock metabolic response. Lancet I, 433-437 (1942). D6. de Duve, C., Lysosomes a new group of cytoplasmic particles. I n “Subcellular Particles” (T. Hayaishi, ed.), p. 128. Ronald Press, New York, 1959. D7. de Duve, C., Lysosomes and chemotherapy. I n “Biological Approaches to Cancer Chemotherapy” (R. J. C. Harris, ed.), p. 101-112. Academic Press, New York, 1961. D8. de Duve, C., and Beaufay, H., Tissue fractionation studies. 10. Influence of ischaemia on the state of some bound enzymes in rat liver, Biochem. J . 73, 610-616 (1959). D9. Deitrick, J. E., Whedon, G. D., and Short, E., Effects of immobilization upon various metabolic and physiologic functions of normal man. Am. J . Med. 4, 3-13 (1948). D10. Drye, J. C., Schoen, A. M., and Schuster, G., Gastric secretion in the immediate post-operativc period. A . M . A . Arch. Surg. 67, 469474 (1953). Dl1. Dudley, H. A. F., The neuro-endocrine response to injury. J . Roy. CoZZ. Surg. Edinburgh 4, 137-146 (1959). D12. Dudley, H. A. F., Surgical convalescence. J . Roy. CoZl. Surg. Edinburgh 13, 1-11 (1968). D13. Duval, P., and Grigaut, A., L’intoxication par les plaies de guerre. La ret4ntion azot6e des bless&. Compt. Rend. Soc. BioZ. 81,873-884 (1918). El. Eades, C. H., Pollack, R. L., and Hardy, J. D., Thermal burns in man. IX. Urinary amino acid patterns. J. Clin. Invest. 34, 1756-1759 (1953). EL’. Eastcott, H. H. G., Fawcett, J. K., and Rob, C. G., Some factors affecting uropepsinogen excretion in man. Lancet I, 1068-1070 (1953). E3. Egdahl, R. H., Pituitary-adrenal response following trauma to the isolated leg. Surgery 46, 9-21 (1959). E4. Egdahl, R. H., Peck, L., and Mack, E., Pituitary adrenal activation following different types of trauma. In “Combined Injuries and Shock” (B. Schildt and L. Thorh, eds.), Intermedes Proceedings, pp. 91-98. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. E5. Erici, I., En experimentall sbrlakningsstudie (An experimental study of wound healing). Svensk Kir. Foren. Forh. pp. 34-35 (1954). E6. Evans, E., and Butterfield, W. J. H., The stress response in the severely burnedan interim report. Ann. Surg. 134, 588-613 (1951). F1. Feldthusen, V., and Larsen, N. A., Serum iron after coronary occlusion and traumatic injuries. Acta Med. S c a d 160, 53-62 (1954). F2. Flear, C. T. G., and Clarke, R., The influence of blood loss and blood transfusion upon the changes in the metabolism of water, electrolytes and nitrogen following civilian trauma. Clin. Sci. 14, 575-599 (1955). F3. Fleck, A., and Munro, H. N., Protein metabolism after injury. Metab., Clin. Ezptl. 12, 783-789 (1963). F4. Foster, D. P., A clinical study of blood fibrin with observations in normal persons, pregnant women and in pneumonia and liver disease. A . M . A. Arch. Internal Med. 34, 301-321 (1924).

METABOLISM DURING THE POSTINJURY PERIOD

47

F5. Foster, D. P., and Whipple, G. H., Blood fibrin studies. Am. J . Physiol. 68, 365-372 (1921-1922). GI. Gamble, J. L., Ross, G. S., and Tisdall, E. F., The metabolism of fixed base during fasting. J. Biol. Chem. 67, 633-695 (1923). Gla. Gelin, L-E., Intravascular aggregation and capillary flow. Acta Chir.S a n d . 113, 463-465 (1957). G2. Gelin, L-E., The significance of intravascular aggregation following injury. Bull. SOC.Intern. Chir. Bruxelles 18, 4 1 9 (1959). G3. Gelin, L-E., Rheologicdisturbance and the use of low viscosity dextran in surgery. Rev. Surg., 19, 385-400 (1962). G4. Gelin, L-E., Sijlvell, L., and Zederfeldt, B., Plasma volume expanding effect of low viscous dextran and macrodex. Acta Chir. Scand. 122, 309-323 (1961). G5. Gelineo, S., Organ systems in adaptation: the temperature regulating system. I n “Handbook of Physiology” (Am. Physiol. SOC.,J. Field, ed.), Sect. 3, Vol. I, pp. 259-282, Williams & Wilkins, Baltimore, Maryland, 1964. G6. Gemzell, C . A., In “Symposium on Protein Metabolism: Influence of Growth Hormone, Anabolic Steroids, and Nutrition in Health and Disease” (F. Gross, ed.), Discussion, p. 297. Springer, Berlin, 1962. G7. Gillette, R. W., Mansberger, A. R., Oppenheimer, J. H., Caldwell, B. F., and Neff, E. G., A new preparation for the study of experimental shock from massive wounds. VI. Paper electrophoretic studies of the serum proteins. Surgery 43, 747-757 (1958). G8. Gillette, R. mi., Mansberger, A. R., Johnston, C. E., and Kookootsedes, G. J., A new preparation for the study of experimental shock from massive wounds. V. Changes in some serum electrolytes and nitrogenous fraction values. Surgery 43, 740-746 (1958). G9. Gimbel, N. A., and Farris, W., Influence of surface temperature upon epithelialization rate of split thickness skin donor sites. Arch. Surg. 92, 554-557 (1966). GlO. Gjessing, E. C., and Chanutin, A., Electrophoretic analyses of sera after treating dogs with 8-chloroethyl vesicants. J . Biol. C h m . 166, 413-420 (1946). G11. Godal, H. C., Quantitative and qualitative changes in fibrinogen following major surgical operations. Acta Med. Scund. 171, 687-694 (1962). G12. Goldenberg, I. S., Lutwack, L., Rosenbaum, P. T., and Hayes, M. A., Thyroid activity during operation. Surg., Gynecol. Obstet. 102, 129-133 (1956). G13. Goodall, M., Stone, C., and Wanes, B. W., Urinary output of adrenaline and noradrenaline in severe thermal burns. Ann. Surg. 146, 479-487 (1957). G13a. Gontzea, I., Sutzesco, P., and Gontzea, L. Effect of painful (algogenic) stimuli on the N balance. J . Physiol. (Paris), 64, 637-646 (1962). G14. Gordon, M. L., The evaluation of afferent nervous impulses on the adrenal cortical response to trauma. Endocrinology 47, 347-350 (1950). G15. Govaerts, P., and de Harven, J., Variations de la teneur du sang en fibrinoghne aprb les traumatismes aseptiques et l a interventions chirurgicales. Compt. Rend. Soc. Biol. 99, 1051-1053 (1928). G16. Gram, H. C., The results of a new method for determining the fibrin percentage in blood and plasma. Actu Med. Scund. 66, 107-121 (1922). G17. Gransitsas, A. N., Effect of adrenaline on nitrogen excretion in normal rats. Am. J . Physiol. 198, 603-604 (1960). G18. Gransitsas, A. N., Effect of acetylcholine on nitrogen excretion in normal rats. Am. J . Physiol. 198, 811-813 (1960). G19. Grant, R. T., and Reeve, E. B., Observations on the general effects of injury in man. Med. Res. Council, Spec. Rept. Ser. 277 (1951).

48

D. P. CUTHBERTSON AND W. J. TILSTONE

G20. Gray, S. J., and Ramsey, C. G., Adrenal influences upon the stomach and the gastric responses to stress. Recent Progr. Hormone Res. 13, 583-613 (1957). G21. Green, H. N., Stoner, H. B., Whiteley, H. J., and Eglin, D., The effect of trauma on the chemical composition of the blood and tissues of man. Clin. Sci. 8, 65-87 (1949). G22. Greenwood, F. C., and Landon, J., Growth hormone secretion in response to stress in man. Nature 210, 540-541 (1966). H1. Haist, R. E., Influence of environment on the metabolic response to injury. I n “The Biochemical Response to Injury” (H. B. Stoner and C. J. Threlfall, eds.), pp. 313-340. Blackwell, Oxford, 1960. H2. Hale, H. B., Mefferd, R. B., Vawter, G., Foerster, G. E., and Criscuolo, D., Influence of long-term exposure to adverse environments on organ weights and histology. Am. J . Physiol. 196, 520-524 (1959). H3. Ham, T. H., and Curtis, F. C., Plasma fibrinogen response in man; influence of the nutritional state, induced hyperpyrexia, infections and disease and liver damage. Medicine 17,413-445 (1938). H4. Hardaway, R. M., Johnson, D. G., Houchin, D. M., Jenkins, E. B., Burns, J. W., and Jackson, D. R., Influence of stress on fibrinogen. J . Trauma 4, 673-676 (1964). H5. Haws, W. H., and Leppelmann, H. J., Reactive changes of enzyme activities in serum and liver as symptoms of “acute syndrome.” Ann. N . Y. Acad. Sci. 76, Art. 1, 250-259 (1958). H6. Hayes, M. A., and Brandt, R., Carbohydrate metabolism in the immediate post-operative period. Surgery 32, 819-827 (1952). H7. Heim, W. G., and Ellenson, S. R., Adrenal cortical control of the appearance of rat slow alpha-globulin. Nature 213, 1260-1261 (1967). H8. Henderson, Y., Prince, A. L., and Haggard, H. W., Observations to surgical shock: A preliminary note. J . Am. Med. Assoc. 69,965 (1917). H9. Heroux, O., Mitotic rate in the epidermis of warm- and cold-acclimatized rats. C U ~J . Biochm. PhysioZ. 38, 135-142 (1960). HlO. Heroux, O., and Gridgeman, N. T., The effect of cold acclimatization on the size of organs and tissues of the rat, with special reference to modes of expression of results. Can. J . Biochem. Physiol. 36, 209-216 (1958). H11. Hirschiield, J. W., Williams, W. H., Abbott, W. E., Heller, C. G., and Pilling, M. A., Significanceof the nitrogen loss in the exudate from surface burns. Surgery 16, 76G773 (1944). H12. Hoch-Ligetti, E., Irvine, K., and Sprinkle, E. P., Investigations of serum protein patterns in patients undergoing operation. Proc. Soc. Exptl. Biol. Med. 84, 707-710 (1953). H13. Holabauer, M., The part played by ACTH in determining the rate of aldosterone secretion during operative stress. J . PhysioZ. (Lolzdon)172, 138-149 (1964). H14. Howard, J. M., Olney, J. M., Frawley, J. P., Paterson, R. E., Smith, L. H., Davis, J. H., Guerra, S., and Dibrell, W. H., Studies of adrenal function in combat and wounded soldiers. A study in the Korean Theatre. Ann. Surg. 141, 314-320 (1955). HIS. Howard, J. E., Parson, W., Stein, K. E., Eisenberg, H., and Reidt, V., Studies on fracture convalescence; nitrogen metabolism after fracture and skeletal operations in healthy males. Bull. Johns Hopkins Hosp. 76, 156-168 (1944). H16. Howard, J. E., Winternitz, J., Parson, W., Bigham, R. S., and Eisenberg, H., Studies on fracture convalescence: influence of diet on posetraumatic nitrogen

METABOLISM DURING THE POSTINJURY PERIOD

H17.

HIS. J1. 52. 53. 54. J5.

J6.

K1. K2.

K3.

K3a.

K4. K5. K6.

L1. L2.

L2a.

49

deficit exhibited by fracture patients. Bull. Johns Hopkins Hosp. 76, 209-224 (1944). Hume, D. M., Bell, C. C., and Bartler, F. M., Direct measurement of adrenal secretion during operative trauma and convalescence. Surgery 62, 174-186 (1962). Hunter, J., “A Treatise on Blood, Idammation and Gunshot Wounds.” Nicol, London, 1794. Johns, M. W., Leucocyte response to sound stress in rats: Role of the adrenal gland. J . Pathol. Bacteriol. 93, 681-685 (1967). Johnston, 1. D. A., Endocrine aspects of the metabolic response to surgical operation. Ann. Roy. Coll. Surgeons Engl. 36, 270-286 (1964). Johnston, I. D. A., and Bell, T. K., The effect of surgical operation on thyroid function. Proc. Roy. SOC.Med. 68, 1017-1020 (1965). Johnston, I. D. A., and Welbourn, R.B., Metabolic and endocrine response to trauma: A review. Rev. Surg. 22, 9-16 (1965). Johnston, I. D. A., Ross, H., Welborn, T. A., and Wright, A. D., The effect of trauma on glucose tolerance and the serum levels of insulin and growth hormone. In “Combined Injuries and Shock” (B. Schildt and L. ThorQn,eds.), Intermedes Proceedings, pp. 127-132. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. Jouan, P., and Samperez, S., Specific action of STH on the secretion of aldosterone in vitro. Ann. Endocrinol. (Paris)26, 70-75 (1964). Keys, A,, Taylor, H. L., Mickelsen, O., and Henschel, A., Famine edema and mechanism of its formation. Science 103, 669-670 (1946). Kinney, J. M., Protein metabolism in human pathological states: The effect of injury upon human protein metabolism. In “Symposium on Protein Metabolism: Influence of Growth Hormone, Anabolic Steroids, and Nutrition in Health and Disease” (F. Gross, ed.), pp. 275-296. Springer, Berlin, 1962. Knigge, K. M., Neuroendocrine mechanisms influencing ACTH and TSH secretion and their role in cold acclimatization. Federation Proc. 19, No. 4, Part 11, Suppl. 5, 45-56 (1960). Koslowski, L., and Messerschmidt, O., The role of the time factor in the combined injury syndrome. In “Combined Injuries and Shock” (B. Schildt and L. Tho+ eds.), Intermedes Proceedings, pp. 4 1 4 6 . Swedish Res. Inst. Natl. Defence, Stockholm, 1968. Kotby, S., and Johnson, H. D., Rat adrenal cortical activity during exposure to a high (34°C)ambient temperature. Life Sci. 6, 1121-1132 (1967). Kotby, S., Johnson, H. D., and Kibler, H. H., Plasma corticosterone response to elevated environmental temperature (34°C) and related physiological activities as influenced by age. Life Sci. 6, 709-720 (1967). Kovach, A. G. B., I n “The Biochemical Response to Injury” (H. B. Stoner and C. J. Threlfall, eds.), Discussion, p. 339. Blackwell, Oxford, 1960. La Due, J. S., Wrobleswski, F., and Karmen, A., Serum glutamic oxaloacetic transaminase activity in human acute transmural myocardial infarction. Science 120, 497-499 (1954). Lanchantin, G. F., and Deadrich, R. E., Serum changes in thermal trauma. 1. Electrophoretic analysis a t pH 8.6. J . Clin. Invest. 37, 1736-1745 (1958). Langendorff, H., and Messerschmidt, O., Radiation injury combined with open wounds. In “Combined Injuries and Shock” (B. Schildt and L. ThorBn, eds.). Intermedes Proceedings, pp- 4 1 4 6 . Swedish Res. Inst. Natl. Defence, Stockholm; 1968.

50

D. P. CUTHBERTSON AND W. J. TILSTONE

L3. Le Quesne, L. P., Fluid and electrolyte balance. Brit. J . Surg. Lister Centenary Number, 64, 449-452 (1967). L4. Le Quesne, L. P., and Pewis, A. A., Post-operative water and sodium retention. Lancet I, 153-158 (1953). L5. Levenson, S. M., and Watkin, D. M., Protein requirements in injury and certain acute and chronic diseases. Federation Proc. 18, 1155-1190 (1959). L6. Lewis, G. P., Intracellular enzymes in local lymph as a measure of cellular injury. J . Physiol. (London) 191, 5914307 (1967). L7. Lewis, G. P., Changes which occur in local lymph during tissue injury. Lecture at Royal Collegeof Physicians and Surgeons, Glasgow, under auspices of Glasgow Post Graduate Medical Board, March, 1968. L8. Lillehei, R. C., Longerbeam, J. K., Block, J. H., and Mannax, W. G., The nature of irreversible shock. Experimental and clinical observations. Ann. Surg. 160, 682-708 (1964). L9. Lindstrand, K., and Schildt, B., The effect of different types of injuries on the methylcobalamine level in mice. I n “Combined Injuries and Shock” (B. Schildt and L. Thorh, eds.), Intermedes Proceedings, pp. 151-155. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. L9a. Lucido, J., Metabolic and blood chemical changes in severe burns. Ann. Surg. 3, 640-644 (1940). LIO. Lusk, G., “The Science of Nutrition,” 4th ed. Saunders, Philadelphia, Pennsylvania, 1931. MI. McCathie, M., Owen, J. A., and MacPherson, A. I. S., Effect of surgery on plasma proteins. Scot. Med. J. 11, 83-88 (1966). M2. McKenzie, J. K., Ryan, J. W., and Lee, M. R., Effect of laparotomy on plasma renin activity in the rabbit. Nature 216, 542-543 (1967). M3. McNeill, I. F., Dixon, J. P., and Moore, F. D., The effects of haemorrhage and hormones on the partition of body water. 111. The effects of haemorrhage and corticosteroids on fluid partition in the adrenalectomized dog. J . Surg. Res. 3, 344-353 (1963). M4. Madden, S. C., and Clay, W. A., Protein metabolism and protein reserves during acute sterile inflammation. J . Exptl. Med. 82, 65-78 (1945). M5. Magalini, S. I., Dell’amore, M., Beltocchi, G., and Hecht-Lucan, G., Comparac tive study of the behaviour of serum proteins, glycoproteins, and phospholipids in the surgical gynecologicpatient. Am. J. Obstet. Gynewl. 83, 1319-1327 (1962). M6. Makara, G. B., Stark, E., and Mihaly, K., Sites at which formalin and capsaicin act to stimulate corticotrophin secretion. Can. J . Physiol. Pharmacol. 46,669-674 (1967). M7. Malcolm, J. D., “The Physiology of Death from Traumatic Fever.” Churchill, London, 1893. M8. Marks, L. J., Chute, R., O’Sullivan, J. V. I., and Giovannello, T. J., Observations on the role of the adrenal in electrolyte response to surgery. Metab. Clin. Exptl. 10, 610-620 (1961). M9. Matsuda, K., Duyck, C., Kendall, J. W., and Greer, M. A., Pathways by which traumatic stress and ether induce increased ACTH release in the rat. Endocrinology 74, 981-985 (1964). M10. Meltser, S. J., The nature of shock. A . M . A . Arch. Internal Med. 1, 571-588 (1908). M11. Miksche, L. W., and Caldwell, F. T., The influence of fever, protein metabolism and thyroid function on energy balance following bilateral fracture of the femur in the rat. Ann. Surg. 64, 455-459 (1967).

METABOLISM DURING THE POSTINJURY PERIOD

51

M12. Miles,A. A., Current notions about acute inflammation. I n “Wound Healing” (C. Illingworth, ed.), pp. 3-15. Churchill, London, 1966. M13. Moore, F. D., Endocrine changes after anesthesia, surgery and unaasthetized trauma in man. Recent Progr. H o m e Res. 13, 511-576 (1957). M14. Moore, F. D., and Ball, M. R., “The Metabolic Response to Surgery.” Thomas, Springfield, Illinois, 1952. M15. M r h , M., Carbohydrate metabolism and resistance to shock. I n “Combined Injuries and Shock” (B. Schildt and L. ThorBn, eds.), Intermedes Proceedings, pp. 121-125. Swedish Res. IMt. Natl. Defence, Stockholm, 1968. M16. Munch, O., “Renal Circulation in Acute Renal Failure.” Oxford Univ. Press, London and New York, 1958. M17. Munro, H. N., General aspects of the regulation of protein metabolism by diet and by hormones. In “Mammalian Protein Metabolism” (H. N. Munro and J. B. Allison, eds.), Vol. 1, pp. 381481. Academic Press, New York, 1964. M18. Munro, H. N., Nutritional factors influencing the metabolic response to injury. I n “Wound Healing” (C. Illingworth, ed.), pp. 171-179. Churchill, London, 1966. M19. Munro, H. N., and Chalmers, M. I., Fracture metabolism a t different levels of protein intake. Brit. J . Exptl. Pathol. 26, 396-404 (1945). M20. Munro, H. N., and Cuthbertson, D. P., Response of proteinmetabolism to injury. Biochem. J . 37, xii (1943). N1. Nardi, G. L., “Essential” and “non-essential” amino acids in urine of severely burned patients. J . Clin. Invest. 33, 847-854 (1954). N2. Needham, A. E., Nitrogen excretion in Carcinides Maenas (Pennant) during the early stages of regeneration. J . Embryol. Exptl. Morphol. 3, 189-212 (1955). N3. Needham, A. E., The pattern of nitrogen excretion during regeneration in oligochaetes. J . Exptl. 2001.138, 369-430 (1958). N4. Neuhaus, 0. W., Balegno, H. F., and Chandler, A. M., Biochemical significance of serum glyooproteins. 1. Changes in rat serum following surgery. Proc. SOC. Exptl. Biol. Med. 107, 960-964 (1961). N5. Nickerson, M., I n “Shock: Pathogenesis and Therapy” (K. D. Bock, ed.), Ciba Found. Symp., pp. 356-370. Springer, Berlin, 1962. N6. Nims, L. F., and Thurber, R. E., Whole body X-irradiation, nitrogen excretion and the adrenal gland. Endocrinology 70, 589-594 (1962). N7. Nylander, G., Iron metabolism in fractures. Acta Endocrinol. 20, 148-156 (1955). 01. O’Shaughnessy, M. C., Brunstrom, G. M., and Fielding, J., Iron chelation in haematomas a t fracture sites. J . Clin. Pathol. 19, 364-367 (1966). 02. Owen, J. A., Effect of injury on plasma proteins. Advan. Clin. Chem. 9, 141 (1967). P1. Peaston, M. J. T., Parenteral nutrition in serious illness Brit. J . Hosp. Med. 2, 708-711 (1968). P2. Pekkarinen, A., The effect of operations and physical injury on the adrenal glands and the vegetative nervous system in man. In “The Biochemical Response to Injury” (H. B. Stoner and C. J. Threlfall, eds.), pp. 217-268. Blackwell, Oxford, 1960. P3. Peltier, L. F., Fat embolism. The failure of lipaemia to potentiate the degree of fat embolism accompanying fractures of the femur in rabbits. Surgerv 38, 720-722 (1955). P4. Prendergast, J. J., Fernichel, B. L., and Daly, B. M., Albumin and globulin changea in burns as demonstrated by electrophoresis. A . M . A . Arch. Surg. 64, 733-740 (1952). P5. Probst, V., Schumacker, G., and Miller, E., Uber die Abhangigkeit “normaler”

52

D. P. CUTHBERTSON AND W. J. TILSTONE

postoperativer Serumeiweissverinderungen von der Schwere des operativen (1958). Gewebstraumas. Medizinische pp. 3-6 R1. Ramey, E. R., and Goldstein, M. S., The adrenal cortex and the sympathetic nervous system. Physiol. Rev. 37, 155-195 (1957). R2. Redgate, E. S., Spinal cord and ACTH release in adrenalectomized rats following electrical stimulation. Endinocrinology 70, 263-266 (1962). R3. Reynolds, B. L., and Strickland, W. M., Formation of an anatomical and pathophysiological membrane in interstitial substance after acute injury. In “Combined Injuries and Shock” (B. Schildt and L. Tho&, eds.), Intermedes Proceedings, pp. 157-161. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. R4. Reynolds, B. L., and Strickland, W. M., Increase in erythrocyte fractions of peripheral blood after acute injury through the use of water solutions equivalent to protein free plasma. In “Combined Injuries and Shock” (B. Schildt and L. Thorh, eds.), Intermedes Proceedings, pp. 281-289. Swedish Res. Inst. NatI. Defence, Stockholm, 1968. R5. Richmond, D. R., Jones, R. K., and White, C. S., The effects of blast and ionizing radiations in rats. In “Combined Injuries and Shock” (B. Schildt and L. ThorBn, eds.), Intermedes Proceedings, pp. 67-74. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. R6. Robinson, R., Serum protein changes following spinal cord injuries. Proc. Roy Soc. Med. 47, 1109-1113 (1954). R7. Rosenberg, S. A., Brief, D. K., Kinney, J. M., Herrera, M. G., Wilson, R. E., and Moore, F. D., The syndrome of dehydration coma and severe hyperglycaemia without ketosis in patients convalescing from burns. New Engl. J. Med. 272, 931-938 (1965). R8. Rosenbund, B., The role of the liver in stress hyperhexosaminemia. Sand. J. Clin. & Lab. Invest. 10, 219-220 (1958). R9. Ross, H., Johnson, I. D. A., Welborn, T. A., and Wright, A. D., Effect of abdominal operation on glucose tolerance and serum levels of insulin, growth hormone and hydrocortisone. Lancet I, 563-566 (1966). RlO. Rossiter, R. J., Pattern of recovery in protein deficiency. Nature 168, 304 (1946). R11. Rudolph, L. A., Dutton, R., and Schaefer, J. A., Glutamic-oxaloacetic transaminase levels in experimental tissue damage. J . Clin. Invest. 34, 960 (1955). R12. Russell, J. A., and Long, C. N. H., Amino-nitrogen in liver and muscle of rats in shock after haemorrhage. Am. J. Physiol. 147, 175-180 (1946). Sl. Sandor, G., “Serum Proteins in Health and Disease.” Chapman & Hall, London, 1966. S2. Schenker, V., Stevenson, J. A. E., and Browne, J. S. L., The characteristic pattern of changes in nitrogen metabolism after trauma. Fedemtion Proc. 6,91-92 (1946). S3. Schildt, B., and Thorh, L., Experimental and clinical aspects of combined injuries. In “Combined Injuries and Shock” (B. Schildt and L. ThorBn, eds.), Intermedes Proceedings, pp. 3-15. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. S4. Schsnheyder, F., Heilskov, N. S. C., and Olesen, K., Isotopic studies on mechanism of negative nitrogen balance produced by immobilization. Scund. J . Clin. & Lab. Invest. 6, 178-188 (1954). S5. Schreier, K., and Karch, H. L., uber den Einfluss von chirurgischen Eingriffen auf den Aminosauren-Stoffwechsel (mit einer Untersuchung der Phenylhydrozinpositiven Corticoide in Urin). Arch. Klin. Chir. 280, 516-535 (1954-1955). S6. Schultz, E. W., Nicholes, J. K., and Schaefer, J. H., Studies on blood fibrin: Its

IbhTABOLISM DURING THE POSTINJURY PERIOD

$3

quantitative determination, normal fibrin values and factors which influence the quantity of blood fibrin. Am. J. Pathol. 1, 101-113 (1925). S7. Scrimshaw, N. S., Protein deficiency and infective disease. I n “Mammalian Protein Metabolism” (H. N. Munro and J. B. Allison, eds.), Vol. 2, pp. 569592. Academic Press, New York, 1964. S8. Sevitt, S., Treatment of severe haemorrhage. Prescribers’ J . 7, 68-77 (1967). S9. Sevitt, S., The boundaries between physiology, pathology and irreversibility after injury. Lancet 11, 1203-1210 (1968). S10. Shires, T., and Jackson, D. E., Poshperative salt tolerance. Arch. Surg. 84, 703-706 (1962). SlOa. Slater, J. D. H., Barbour, B. H., Henderson, H. H., Casper, A. G. T., and Bartter, F. C., Influence of the pituitary and the renin-angiotensin system on the secretion of aldosterone, cortisol and corticosterone. J . Clin. Invest. 42, 1504-1520 (1963). S l l . Slater, T. F., Greenbaum, A. L., and Wang, D. Y., Ciba Found. Symp., LYSOsomes p. 311 (1963). S12. Slater, T. F., and Greenbaum, A. L., Changes in lysosomal enzymes in acute experimental liver injury. Biochem. J . 96, 484-491 (1965). S12a. Slater, T. F., Strauli, V. D., and Sawyer, B. C., Changes in liver nucleotide concentrations in experimental liver injury. 1. Carbon tetrachloride poisoning. Biochem. J. 93, 260-266 (1964). S13. Spoelstra, A. J. G., Studies on the calorigenic effect of adrenaline and noradrenaline. J. Physiol. (Paris) 66, 677-696 (1963). S13a. Sterling, K., Liosky, S. T., and Freedman, L. J., Disappearance curve of intravenously administered 1131 tagged albumin in the post-operative injury reaction. Metab., Clin.Exptl. 4, 343-350 (1955). 514. Stoner, H. B., Studies on the mechanism of shock. The quantitative aspects of glycogen metabolism after limb ischaemia in the rat. Brit. J . Exptl. Pathol. 39, 635-651 (1958). S15. Stoner, H. B., and Pullar, J. D., Studies on the mechanism of shock: heat loss after limb ischaemia injury. Brit. J. Exptl. Pathol. 44, 586-592 (1963). S16. Stoner, H. B., and Threlfall, C. J. The effect of limb ischaemia on carbohydrate dist.ribution and energy transformation. I n “The Biochemical Response to Injury” (H. B. Stoner and C. J. Threlfall, eds.), pp. 105-128. Blackwell, Oxford, 1960. 517. Stoner, H. B., Health, D. F., Threlfall, C. J., and Ashby, M. M., The depression of energy metabolism in the early stage of the response to injury. I n “Combined Injuries and Shock” (B. Schildt and L. ThorBn, eds.), Intermedes Proceedings, pp. 115-119. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. S18. Stromberg, L. R., Woodward, K. T., Mahim, D. T., and Donati, R. M., Combined surgical and radiation injury. 11. Effect of bone marrow shielding on wound healing in X-irradiated rats. I n “Combined Injuries and Shock” (B. Schildt and L. ThorBn, eds.), Intermedes Proceedings, pp. 35-39. Swedish Res. Inst. Natl. Defence, Stockholm, 1968. TI. Taylor, F. H. L., Levenson, S. M., Davidson, C. S., and Adam, M. A., Abnormal nitrogen metabolism in patients with thermal burns. New Engl. J . Med. 229, 855-859 (1943). T2. Threlfall, C. J., and Stoner, H. B., Carbohydrate metabolism in ischaemic shock. Quart. J. Exptl. Physiol. 39, 1-9 (1954). T3. Ticktin, H. E., Ostrow, B. H., and Evans, T. M., Serum glutamic oxalacetic transaminase in trauma. Clin. Res. Proc. 4, 102-103 (1956).

54

D. P. CUTHBERTSON AND W. J. TILSTONE

T4. Tilstone, W. J., and Roach, P. J., Plasma corticosterone in the rat in relation to trauma and environmental temperature. Quart. J . Exptl. Physiol. 64, 341-345 (1969). T5. Trueta, J., Barclay, A. E., Daniel, P., Franklin, K. J., and Pritchard, X.I. M. L., “Studies of the Renal Circulation.” Oxford Univ. Press, London and New York, 1947. T6. Turpini, R., and Stefanine, M., The nature and mechanism of the hemostatic breakdown in the course of experimental hemorrhagic shock. J. Clin. Invest. 38, 53-65 (1959). U1. Upjohn, H. L., and Levenson, S. M., Some metabolic and nutritional changes associated with injury. A . M . A . Arch. Internal Med. 101, 537-550 (1958). V1. Van Slyke, D. D., Phillips, R. A., Hamilton, P. P., Archibald, R. M., Dole, V. P., and Emerson, K., Jr., Effect of shock on kidney. Trans. Assoc. Am. Physicians 68, 119-128 (1944). W l . Walker, J., Changes in the non-protein fractions of the plasma nitrogen following extensive thermal burns. Am. J. Med. 209, 413-414 (1945-1946). W2. Walker, W. F., Acid-base balance. Brit. J . Surg. 64, 452-455 (1967). W3. Walker, W. F., Fleming, L. W., and Stewart, W. K., Urinary magnesium excretion in surgical patients. Brit. J . Surg. 66, 466-469 (1968). W4. Walker, W. F., Watt, A., Morgan, H. G., and McCowan, M. A. A., The effect of operations of varying severity upon calcium and phosphorus metabolism in the elderly. Brit. J . Surg. 61, 783-790 (1964). W5. Walker, W. F., Zileli, M. S., Reutter, F. mi., Schoemaker, W. C., Friend, D., and Moore, F. D., Adrenal medullary secretion in haemorrhagjc shock. Am. J . Physiol. 197, 773-780 (1959). W6. Warren, R., Amdur, M. O., Balko, J., and Baker, D. V., Post-operative alterations in the coagulation mechanism of the blood. A . M . A . Arch. Surg. 61, 419432 (1950). W7. Weimer, H. E., and Benjamin, D. C., Influence of adrenal cortex on synthesis of CWAP globulin of rat serum. Proc. SOC.Exptl. Biol. Med. 122, 1112-1114 (1966). W8. Weimer, H. E., and Coggshall, V., Divergent responses of serum glycoprotein fractions to tissue injury in adrenalectomized rats. Can. J . Physiol. Phurmacol. 46, 767-776 (1967). W9. Weimer, H. E., and Humelbaugh, C., The effects of periodic challenge on the response of m-AP globulin and other acute phase reactants of rat serum to tissue injury. Can. J . Physiol. Pharmacol. 46, 241-248 (1967). W10. Wendeberg, B., Mineral studies of fractures of the tibia in man studied with external countingof Srg5.Acta Orthopaed. Scand. S u p p l . 62, 1-79 (1967). WlOa. Werner, I., On the regeneration of serum polysaccharide and serum proteins in normal and intoxicated rabbits. Acta Physiol. Scand. 19, 27-39 (1949). W11. Wertheimer, Fabre, and Clogne, Bull. Mem. SOC.Chir. Paris 46, 9 (1919); loc cit. Cannon, W. B., “Traumatic Shock,” p. 88. Appleton, New York, 1923. W12. Whipple, G. H., Miller, L. L., and Robscheit-Robbins, F. S., Raiding of body tissue protein to form plasma protein and hemoglobin. J . Exptl. Med. 86,277-286 (1947). W13. Wilhelmi, A. E., Russell, J. A., Engel, M. G., and Long, C. N. H., Some aspects of the nitrogen metabolism of liver tissue and rats in hemorrhagic shock. Am. J . Physiol. 144, 674-682 (1945). W14. Wilkinson, A. W., Starvation and operation. Lancet 11, 783-784 (1961). W15. Wilkinson, A. W., Billing, B. H., Nagy, G., and Stewart, C. P., Excretion of chloride and sodium after surgical operations. Lancet I, 640-644 (1949). W16. Wilkinson, A. W., Billing, B. H., Nagy, G., and Stewart, C. P., Nitrogen metab-

METABOLISM DURING THE POSTINJURY PERIOD

W17.

W18.

w19.

w20.

w21. w22. Y1. Y2.

z1. 22.

55

olism after surgical operations: Use of protein hydrolysate. Lancet I, 533-537 (1950). . . Wilkinson, A. W., Billing, B. H., Nagy, G., and Stewart, C. P., Excretion of potassium after partial gastrectomy. Lancet 11, 135-137 (1950). Wilson, W. C., and Stewart, C. P., Changes in blood chemistry after burning injuries and in other grave surgical conditions, with some references to treatment by desoxycorticosterone acetate. Trans. Med. Chir. Soc. Edinburgh pp. 153-173 (1938-1939). Wilson, W. C., MacGregor, A. R., and Stewart, C. P., Clinical course and pathology of burns and scalds under modern methods of treatment. Brit. J. Surg. 26, 826-865 (1938). Wray, J. P., Changes in liver composition following fracture of the tibia in the rat. J. Trauma 7, 811-817 (1968). Wroblewski, F., and La Due, J. S., Lactic dehydrogenase activity in blood. Proc. SOC.Exptl. Biol. Med. 90, 210-213 (1955). Wroblewski, F., Friend, C., Nydick, I., Ruegsegger, P., and La Due, J. S., The mechanism and significance of alterations in serum glutamic pyruvic transaminase in liver and heart,. Clin. Res. Proc. 4, 102 (1956). Ya.tes, F. E., and Urquhart, J., Control of plasma concentrations of adrenocortical hormones. Physiol. Rev. 42, 359-443 (1962). You, S. S., You, R. W., and Sellars, E. A., Effect of thyroidectomy, adrenalectomy and burning on the urinary nitrogen excretion of the rat maintained in a cold environment. Endom.nology 47, 156-161 (1950). Zimmermann, B., Casey, J. H., Block, H. S., Bickel, E. V., and Govrick, K., Excretion of aldosterone by the post-operative patient. Surg. Forum, Proc. 6, 3-6 (1955). Zwicker, M., Post-operative serum Kufferspiegel-veranderungen.Klin. Wochchr. 37, 933-939 (1959).

NOTEADDB

IN

PROOF

Since this paper went to press a noteworthy publication by J. W. L. Davies, S.-0. Liljedahl, and G. Birke has appeared in Injury, 1, 43 (1969). Nineteen patients with burns of varying sizes were treated in a warm (32"C), dry environment and another group of 41 with similar sized burns in a cooler (22'C), more moist environment. Those in the warm, dry environment showed smaller losses of body weight while in the hospital, with significant reductions in basal metabolic rate and the rate of catabolism of plasma albumin and y-G-globulin, smaller amounts of protein in the extravascular space, increased serum albumin concentrations, a greater rate of synthesis of albumin, and a lower rate of synthesis of y-G-globulin. Less protein was lost in the exudate when the patients were exposed in the warm, dry environment and their clinical condition was often better than those in the normal hospital environment.

This Page Intentionally Left Blank

DETERMINATION OF ESTROGENS, ANDROGENS, PROGESTERONE, AND RELATED STEROIDS IN HUMAN PLASMA AND URINE Ian E. Bush Department of Physiology, Medical School of Virginia, Richmond, Virginia 1.

Introduction .............. ,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57 57 60 67 76 80 2. The Assessment of Methods.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 2.1. The Development of Techniques for Steroids. . . . . . . . . . . . . . . . . . . . . . . . 83 2.2. Specificity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 2.3. Accuracy, Precision, and Sensitivity. .......................... 90 2.4. Operational and Logical Design.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 2.5. Diagnostic Relevance.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 2.6. Return on Investment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 3. Advances in General Techniques.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 3.1. Techniques of Extraction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 3.2. Techniques of Separation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 3.3. Techniques of Detection and Measurement.. . . . . . . . . . . . . . . . . . . . . 116 118 3.4. Techniques of Automation and Data Analysis.. . . . . . . . . . . . . . . . . . . . . . 4. Specific Examples.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Estrogens and Related Steroids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 4.3. Progesterone and Related Steroids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 4.4. Testosterone and Related Steroids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 5. The Future.. . . . . . . .............................................. 129 References. ............ .............................................. 130 1.1. General Considerations. ........... .......................... 1.2. Major Steroids of Blood and Urine.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3. Secretion and Metabolism of Steroid Sex Hormones.. . . . . . . . . . . . . . . . . 1.4. Major Disorders of Secretion and Metabolism.. . . . . . . . . . . . . . . . . . . . . . 1.5. Parameters of Clinical Significance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

1.

Introduction

1.1. GENERALCONSIDERATIONS Taxidermy in modern times, with borax mothproofing aa its latest and greatest step forward, can be a fascinating hobby well able to fill in many spare hours with undreamed-of satisfaction. (P9,p. VII)

The measurement of the steroid hormones and their metabolites in tissues and body fluids has long been a major technical problem to endocrinologists and biochemists, and even more of a problem to clinical chemists. Steroid biochemistry has been somewhat esoteric and to some 57

58

IAN E. BUSH

extent has remained so despite the tremendous increase in work in this field in the last fifteen years. I n large part this has been due to the great technical difficulty of measuring the often minute concentrations of steroids commonly found in the blood and urine of man and other species. Indeed, i t can reasonably be suggested that the study of steroid endocrinology and biochemistry has been almost completely dependent upon the laborious development of analytical techniques, often to the exclusion of other considerations. This has naturally led to a situation in which large areas of methodology and technique have been closed to nonspecialists and quite obviously unsuited to general use in a laboratory of clinical chemistry. At the present time we are probably a t the threshold of a major expansion of the use of methods for measuring steroids in routine clinical chemistry, and a number of methods which were confined to the research laboratory five years ago are now quite widely available in the larger hospitals and in some of the more sophisticated commercial service laboratories. However, these methods are extremely expensive compared with other methods in clinical chemistry, and it is still difficult for the clinical chemist who is not a steroid specialist to obtain a balanced appraisal of the field. The increasing sophistication of clinical endocrinology has lead to a greatly increased pressure upon the clinical chemist to provide more and more relatively difficult techniques in the laboratory, and the choice of techniques-and even more of e q u i p m e n t is fraught with an uncertainty which is perhaps made worse rather than better by the ernbarras de richesse to be found in the literature on the subject. Nor is the aspiring clinical chemist likely to get much help if he takes the trouble to seek out one of the specialists in the field, unless he is very lucky. The specialists have been banging their shins, if not their heads, against seemingly intractable problems of sensitivity, specificity, and recovery for the greater part of their professional lives, and are only too likely to emphasize the complexity and difficulty of analytical methods for steroids. It is only natural that they often become a little impatient with the visiting general clinical chemist whose main interest is in something simple and quick. The latter will often return to his home ground with the impression that steroids in blood and urine are entities in such vanishingly small concentrations and of so desperately unstable a nature that only a genius can actually catch one in a test tube. If, as is so often the case, he has also been told that the few simpler methods he currently uses in his routine laboratory are hopelessly nonspecific and largely useless for any but the crudest diagnostic purposes, this impression will be superimposed upon a mood which is not conducive to strictly rational thought. In this, however, he may be luckier than his colleague

DEjTERMINATION OF STEROIDS IN PLASMA AND URINE

59

who is given to understand that with $6sooO-sO,O0O of the “right” equipment and half the time of a good technician he should be under way with half a dozen reliable modern methods in three or four months. I n view of this situation I shall attempt to write this article mainly from the point of view of the general clinical chemist. As is so often the case in real life, the true situation is described neither by one of the two extreme views often presented in this field, nor by a hazy compromise between them, but by specific parts of each. The measurement of steroids in body fluids and tissues is indeed an exceptionally difficult task in many cases, and in some, still a task for minor geniuses only. However, this is by no means true of all cases and there are ways in which the competent nonspecialist can achieve his aims of introducing into his routine operations new and improved techniques for measuring steroids. He should be tolerant and understanding when the specialist tends to emphasize the complexity of the problem, and a t the same time refuse to be overawed by him. I shall adopt one or two principles as axiomatic in what follows. First, it is not reasonable to lay down any ideal standards of performance of an analytical method considered in isolation. An evaluation of an analytical method is a t best incomplete-and a t worst quite misleading-if it does not take account of the purpose of that method. This “purpose” is mainly defined by the nature of the sample; the minimum significant change in the variable that is to be measurable; the minimum number of samples that must be analyzed per unit time; and the maximum allowable interval between the collection of the sample and the presentation of the estimate of the measurement. Second, the maximum quality of performance that is economically feasible should be aimed a t in any method used in a routine laboratory; in other words, all economically reasonable steps to increase one’s margin of safety over the minimum acceptable standards should be adopted. Third, all the steps of an analytical procedure and all the criteria determining the nature of those steps must be considered as a whole when designing or evaluating an analytical method. T h a t is, an analytical method is just another example of an “operation” in the modern sense and must be assessed not only in terms of its physical and chemical technicalities, but also in the terms of systems analysis or operations research. I believe that these principles are always implicit and often recognized in the work of practicing clinical chemists: they are naturally of secondary consequence to most specialists, in any field, since the specialists are mainly devoted to cracking the technical problems. Finally, this article is designed to be selective in its scope. I n an age of plethora in all media, I believe it is more valuable to be comprehensible than to be comprehensive. This aim is made all the more reasonable by

60

IAN E. BUSH

the considerable number of specialist books and reviews on this topicsome of them excellentwhich have been published in recent years (E2, G8, G9, L8, R3, T5).

1.2. MAJORSTEROIDS OF HUMAN BLOOD AND URINE In general the hard-scaled kinds of fishes are to be preferred for mounting, i.e. the sunfishes, perch, pickerel, muskellunge, gars, . . Those kinds with much fat make poor taxidermy specimens. (P9, P. 27)

.

This review is concerned with methods for the measurement of the steroid sex hormones and their metabolites. While not strictly concerned with adrenocortical steroids or with synthetic analogs of the steroid hormones, i t will deal briefly with some of them where necessary. This is unavoidable for three main reasons. I n the first place some steroids recognized as major or minor sex hormones are secreted by the adrenal cortex, and the gonads are capable of secreting steroids of adrenocortical type under certain conditions. Second, estimations of the adrenocortical glucocorticoids is essential in the differential diagnosis of most disorders of steroid sex hormones. Third, synthetic steroids are of great importance as oral contraceptives and therapeutic or diagnostic drugs: they may need to be measured, or may interfere with the measurement of naturally secreted steroids or their metabolites. The steroid nucleus with its numbered carbon skeleton is shown in Fig. 1A. By convention the hydrogen atoms are not shown, and the number attached to any carbon atom is obtained by subtracting the number of carbon-carbon bonds to it from 4, the valency of carbon. I n particular i t should be noted that the carbon atoms 18 and 19 are normally the centers of methyl groups often referred to as the “18-” and “19-methyl groups” although this is, strictly speaking, inaccurate. I n Fig. 1B are shown the structures of the main steroid sex hormones of the three classes currently recognized. I n each case there is good evidence t ha t the compound is the main secretory product of its class. In Fig. 1C are shown three typical major metabolites of the first two classes. In these structures another convention is apparent. The nucleus is considered as lying more or less in the plane of the paper: bonds directed toward the reader or above the paper are drawn as dashed lines and are designated “arr;those away from the reader or below the paper are drawn as solid lines and designated “j3.” An excellent account of these and other stereochemical features of the steroids will be found in Klyne’s monograph (K9). The steroids in Figs. 1B and 1C are hydrophobic lipids and are called “free” steroids to distinguish them from their water-soluble conjugates,

DETERMINATION OF STEROIDS IN PLASMA AND URINE

61

CH3

1

&

0

c19

Testosterone

Etiocholanolone

o&oHo&

c11

Progesterone

Androsterone

Estradiol C38

Pregnanolone

(c)

FIQ. 1. Important steroid structures. (A) The steroid nucleus and numbering. (B) Major sex hormones. (C)Typical metabolites.

some examples of which are shown in Fig. 2. The predominant steroid conjugates of human urine are /3-glycosides of glucuronic acid and esters of sulfuric acid. There is some confusion in the nomenclature of the former. For a long time, and in most of the literature prior to about 1956, they were simply called “glucuronides,” and many people stick to this

62

JAN E. BUSH

0

I

OH

booDehydroepiandrosterone sulfate

Pregnanediol glucuronide

Testosterone glucuronide

FIG.2. Typical steroid conjugates.

term colloquially or even lapse into i t in print. It is now commoner to call them “glucosiduronates” or “glucuronosides,” the latter being disapproved, but both names aiming to emphasize the glycosidic linkage between the steroid and the uronic acid, and distinguishing it from the ester linkage of glucuronic acid seen in the conjugates of some hydroxylic substances in urine (W6, p. 289). More recently they have been described as 3-/3-~-glucopyranosiduronates (B9) to remind readers th a t the uronic acid radical is in the pyranose rather than the furanose form. Since the earlier term has been declared acceptable by the American Endocrine Society and is shorter it will be used in what follows. It is probable that other conjugates of steroid hormones and their metabolites will eventually be found to exist in human urine, either as normal or unusual constituents. A conjugate of 17a-estradiol with pglucuronic acid and N-acetylglucosamine has been isolated recently (J3) ; and a n unusual compound, not a conjugate in the usual sense, of urea and

DETERMINATION OF STEROIDS IN PLASMA AND URINE

63

3cu,ll/?-dihydroxyandrost-4-en-17-onehas been isolated by Fukushima (F6). The existence of the major proportion of urinary steroids in the form of conjugates has long been one of the greatest technical difficulties in this branch of analytical chemistry. The physical state of these steroids in the blood is of the greatest interest to the endocrinologist and may greatly affect the significance attached t o the values of steroid concentrations which are obtained by the usual methods. All steroids in plasma are to some extent associated with, or “bound” to, the plasma proteins. It is convenient to distinguish three common types of binding. In the first there exists a more or less highly specific binding protein for the free steroid hormone under consideration. Usually the dissociation constant of the steroid-protein complex is very low and the concentration of the protein, or more strictly of its binding sites, is low. Under these circumstances, 80-95% of the total steroid in the plasma is in the form of this complex a t “normal” concentrations of steroid, and the rest is loosely associated with plasma albumin or other proteins, or in free solution. The best example of this type is cortisol and its plasma carrier, transcortin (see B28, K4, M8), but a seemingly specific testosterone-binding globulin has recently been found in the plasma of pregnant women (K3, M3). The case of progesterone is more complex since it is quite strongly bound by transcortin but becomes part of a complex chemical equilibrium in competition with cortisol and corticosterone (T3). In the second type the free steroid is more loosely bound to albumin and other plasma proteins. Again an equilibrium between the steroidprotein complex and the free steroid and protein is established. One important difference from the first case, however, is that the concentration of binding sites is usually very large. Thus over quite a wide range of changing total steroid concentration the proportions of free and bound steroid remain the same. This is in contrast with the first type in which the low concentration of the carrier protein and the low dissociation constant of the steroid-protein complex leads to an equilibrium in which most of the binding sites of the protein are occupied a t normal steroid concentrations. Under these circumstances the proportion of unbound steroid and of steroid loosely bound to albumin increases as the total steroid concentration rises. This is important physiologically because there is considerable evidence supporting the generally held hypothesis that the concentration of the hormone in the extravascular tissue fluid, and hence its biological actions, are determined by the concentration of the free steroid in the blood plasma (Fl, K4, T3). Similarly the distribution of the steroid between the plasma and the blood cells is affected by such binding, and influences to some extent one’s

64

IAN E. BUSH

choice of plasma or whole blood as the appropriate sample to extract when measuring steroid concentrations (e.g., F1) . Despite valid objections to the use of plasma when dealing with adrenal or gonadal venous blood, the use of plasma if separated soon after collection is the general method of choice. Serum is rarely, if ever, used. The third type occurs only with the conjugates of steroids. The two commonest forms, glucuronides and sulfates, exist largely as anions a t physiological p H values, and it is well known that plasma albumin contains a large number of binding sites with a high affinity for anionic groups of all kinds. This binding is probably far stronger with the steroid sulfates than with the glucuronides and is of interest for two major reasons. I n the first place such binding will affect the distribution and excretion of such compounds to a considerable degree. I n the second, it is sufficiently strong to be a potential difficulty in securing the complete extraction or hydrolysis of steroid conjugates in plasma and some urine specimens. In contrast, neither of the two previous types of protein binding seriously affects the extractability of free steroids from plasma when the usual solvents are used in the usual volume ratios. I n normal urine the major part of the steroids present are in free solution as the anions or salts of the conjugates. Very small quantities of some steroids are present as free steroids. In the presence of the normal small amounts of protein and particulate debris in urine, it is probable that these relatively minute quantities of such steroids are partly or largely associated with, or adsorbed on, such particles in a nonspecific manner because of their very low solubility in pure aqueous solution. This is not known to hinder their extraction with organic solvents or with adsorbents. However, one will expect to encounter quite serious losses if the urine is in prolonged contact with certain types of plastic tubing or bottles during or after its collection, or if heavy sediments are not broken up and mixed well before a fraction of a urine sample is drawn off for analysis. Representative values for the concentrations in plasma of the major steroids under review here are given in Table 1. These values bridge what is a t present an important boundary between relatively easy and rather difficult problems of sensitivity with current techniques. Thus, progesterone a t concentrations of 0.3-1.0 pg/100 ml in the luteal phase of the menstrual cycle and testosterone in male plasma a t concentrations of 0.5-0.8 pg/loO ml can be measured by a wide variety of existing methods. On the other hand, the plasma concentrations of estrogens in nonpregnant women, of testosterone in women, and of progesterone in the follicular phase of the menstrual cycle are in a range which can be reached only by a small number of rather difficult techniques at present.

U

TABLE 1

PLASMA CONCENTRATIONS OF MAJORSTEROIDSEX HORMONES AND RELATED STEROIDS~.~ Female Hormone

Male 0.8 k 0.07 (R2)

Testosterone

0.06 k 0.004 (H6)

Androst-4-ene-3,17-dione Progesterone

-

Estradiol

0.0021

0,0064 k 0.00075 (Bl)

Estrone

0

F = Follicular phase; IH

* Values pgg/lOO ml.

+ 0.00026 (Bl)

=

F

L

0.041 f 0.003 (B2) 0.074 k 0.010, IH (B2) 0.117 k 0.019, PCO (B2) 0.140 k 0.008 (H6) 0.032 k 0.025 (54) 1.45 zk 0.11 (J4) 2.48 k 0.73 (54) (Preg., 5 wk) 1.67 k 0.74 (54) (Preg., 9 wk) 12.5 f 2.8 (54) (Preg., 32 wk) 12.5 k 2.8 (54) (Preg., term) 0.0064 +_ 0.0015 (B1) 0.0196 k 0.0016 (Bl) 0.398 (Bl) (Preg., 16 wk) 0.0057 f 0.00073 (Bl) 0.011 f 0.00057 (Bl) 0.112 (Bl) (Preg., 16 wk)

idiopathic hirsutism; L = Luteal phase; PCO = polycystic ovaries.

n

xs

z58 z

0 q u,

e

m

Fd

2 U

u1

2 t+

iP

rn

z

;P

5 E! 3

66

IAN E. BUSH

I n Table 2 the major metabolites of the steroid sex hormones and some important related steroids are listed. Their relationships to their precursors are in some cases controversial a t present and will be discussed in the next section of this article. Suffice i t to say a t the moment th a t simple relationships between the excretion rates of some of these metabolites and the secretion rates of their precursors are not easy to establish. On the other hand, it is fair to add that some of the complexities implied by recent work on the metabolism of testosterone and androstenedione, and on the metabolic activity of the placenta and fetus in pregnancy should not be allowed to obscure the relative simplicity of many aspects of the metabolism of progesterone and of estrogens. Once again the boundary between relatively easy problems of measurement and rather difficult ones is bridged by the range of values encountered. Pregnanediol glucuronide in women, estrogens in pregnant women, and the six major 17-ketosteroids in both sexes are excreted a t rates between 0.25 and 4.0 mg in 24 hours and are within the range of easy measurement by a large number of existing techniques. On the other hand, the measurement of estrogens in nonpregnant women, and of TABLE 2 PRINCIPAL WELL-ESTABLISHED METABOLITES OF STEROID SEX HORMONES AND RELATED PRODUCTS

Metabolites

Approximate range of excretion rates (mg/% hr)

1

Androsterone Etiocholanolone Dehydroepiandrosterone @HA)

2-6 0.1-10.0

11-Ketoetiocholanolone 118-Hydroxyandrosterone 118-Hydroxyetiocholanolone

0.4-2.0 0.2-1.0 0.1-0.5

Pregnane-3a, 17a;20a-triol Pregnane-3a, 17a-diol-20-one Pregnane-3a,2Oa-diol Pregnan-3a-ol-20-one Estrone Estradiol-17 Estriol 2-Hy droxyestrone ZMethoxyestrone

DHA-SOa, testosterone, androstenedione DHA-SO,, 17a-hydroxy-Aspregnenolone

I

o.3-1?a .Oa 0.50-6. ?a O"

Probable precursors

I

5

0.001-0.010" 0.001-0.008" 0.001-0.0145 0.003-0.020" 0.002-0.015"

Cortisol, 118-hydroxyandrostenedione 17a-hydrooxyprogesterone, 17~-hydroxy-As-pregnenolone Progesterone, A6-pregnenolone Estradiol Estrone DHAS04 (? fetus) -

~~

a Nonpregnant women only. Ranges uncertain due to menstrual variation, small numbers of subjects, or great variety of methods.

DETEBMINATION OF STEROIDS IN PLASMA AND URINE

67

many steroids in postmenopausal women or hypophysectomized patients is extremely difficult, At the present time it is generally recognized that a completely satisfactory quantitative description of the metabolic fate of the major steroid sex hormones is not yet available. On the other hand, it is not reasonable to exaggerate this gap in our knowledge to a n extent which casts too much doubt on the clinical value of measurements of the known metabolites of these hormones.

1.3. SECRETION AND METABOLISM OF STEROID SEX HORMONES Here innumerable birds and small mammals were stuffed. (P9,p. VIII) The most important and useful generalization that has been established by the research of the last fifteen years is that the biosynthesis of all the steroid hormones involves a large number of pathways common to all steroid-secreting tissues. Having a common embryological origin in the intermediate cell mass, it is perhaps not surprising that the adrenal cortex and the endocrine tissues of the gonads should have a basic stock of similar enzymes operating via similar biosynthetic pathways. This “common currency” of steroid biosynthesis also provides a satisfactory basis for the understanding of the many possible alterations which are seen in disease. Progesterone is a central intermediate metabolite of all the biosynthetic pathways found in the gonads and the adrenal cortex (Fig. 3). I n the production of the CISandrogens and the CIS estrogens, 17a-hydroxyprogesterone is almost equally important, although a second pathway to testosterone by the direct conversion of progesterone to testosterone acetate is probably possible. Although complete quantitative evaluation of the role of these reactions in determining the characteristic secretory products of the various steroid-secreting tissues is not yet available, it is generally believed, and there is a considerable body of evidence suggesting, that these characteristics are determined by the amounts of crucial enzymes present in the different tissues. The nature of the genetic and other regulatory mechanisms determining these enzyme concentrations remains t o be worked out. Careful analysis of the biosynthetic properties of the membrana granulosa, the theca interna, and luteinized cells of the Graafian follicle and corpus luteum have revealed some of the factors determining the secretory products of these structures, but the role of the ovarian interstitial tissue is not completely established. There is much evidence suggesting that the latter tissue may secrete the major portion of ovarian androgens in women and in virilizing syndromes accompanying ovarian disease (G11, L4,Mla, M5).

I

Acetate 6

t Cholesterol

& HO

HO

HO

DHA'

CH, OH

& I / Pregnenolone

2

L CH,

I

DOC

Corticosterone'

CHIOH

CHzOH

I

& o

o

CH, OH

I

o&Lo&

I

c=o

___t

0 Progesterone

1

cortisol*

CH3

&

/

A=o

0

17 01 -Hydroxyprogesterone

Lo&

L o &

-

1 I p - Hydroxy androstenedione.

Androstenedione*

7\

Estradiol*

Testosterone.

FIG.3. Major pathways in the biosynthesis of adrenocortical steroids in cells

DETERMINATION OF STEROIDS IN PLASMA AND URINB

69

I n the last five years much more work has been done on what might for the moment fairly be called the minor or secondary secretory products of the steroid-secreting endocrine organs. They are potentially important in the assessment of the possible role of steroid sex hormones in hormonedependent cancers of the female reproductive organs, and of the origin of such products in postmenopausal women. Since much of this information has been obtained in subprimate species, its relevance to human endocrinology is uncertain. Thus, relatively large amounts of progesterone have been identified in the adrenal venous blood of cows, and small amounts have been found in the same site in women, sheep, and pigs (%I), In the dog a number of C,, and C,, steroids have been identified (H3) in adrenal venous blood. Similarly 20a-hydroxypregn-4-en-3-one is found in ovarian venous blood (M4, M5), and androstenedione has been shown to be the major secretion of the testis in immature bulls and to become a minor product in the mature animal due to the increasing secretion of testosterone (L6). One of the major changes in outlook in the last five years has resulted from the original discovery by Baulieu (B3) that dehydroepiandrosterone (DHA) sulfate was secreted by several human adrenocortical carcinomas. Wieland et al. (W4), using the percutaneous adrenal cannulation technique of Bush and Cranston (B36), were subsequently able to obtain strong direct evidence that this substance and androstenedione were secreted by the normal human adrenal cortex, their secretion being stimulated by corticotropin (ACTH) . Lieberman and Gurpide (L5, V3), using isotopic methods, have obtained evidence suggesting that the normal human adrenal cortex secretes in 24 hours about 24 mg of this steroid conjugate. Sizable amounts of testosterone, 17a-hydroxyprogesterone,and androstenedione have been identified in human ovarian venous blood in

responsive to ACTH. The exact order of reactions 3, 4, 6, and 6 is not certain, and several orderings may be possible. The pathway for ddosterone is not shown. It is probable (eg., B2) that the pathways leading to the formation of androstenedione, testosterone, and estrogens are at least partly in cells not responsive to ACTH in the manner of the major eella of the zona fasciculata. DHA is released into the blood stream as the sulfate: sulfation occurs at an unknown part of the sequence. Estradiol is suggested as a secretion product although i t is not known how much or what estrogen is normally secreted. Normally secreted products are asterisked. All the components shown may be secreted in abnormal states. Based mainly on Hechter, Dorfman, Talalay, and many others, for which see Bush (B28). Key to reactions (enzymes) : 0,via mevalonate, isopentenyl phosphate, squalene ; 1, hydroxylase, desmolase (20,221 ; 9, As-3/3-hydroxysteroid dehydrogenase, 5(4)isomerase ; 3, 21-hydroxylaae ; 4, 17a-hydroxylase ; 6, llp-hydroxylase ; 6, desmolase (17,20) ; 7, 17/3-hydroxysteroid dehydrogenase ; 8, 19-hydroxylase ; 9, aromatization.

70

IAN E. BUSH Reaction

[ 1 ] 4,s

- Hydrogenases

[2] Hydroxysteroid Dehydrogenases

Product

Substrate [group I

A6 -3-Ketone group

501-H or 5P-H dihydrosteroid

Ketone group [redl aldehyde group [red] 1 ry hydroxyl [ox] 2 ry hydroxyl [ox]

2 ry hydroxyl 1 ry hydroxyl aldehyde ketone

[Known or probable for following hydroxyl groups and related carbonyl groups: 301-, 30-, 6P-, 1 IP-, I&-, 16/3-, 17a-, 17p-, 18-, 19-, 20a-, 20p-, 211 [3] 17,20-Desmolase

C2,-17a, 20-diols

17-ketosteroid

[4] Hydroxylation

nucleus or side chain

hydroxysteroid

[With estrogens: 2-, 6-, 15, 16 With others: 2a-, 6p-, 601-, 7a-, 1 5-, 16-, 2 1 I [5 I Methylation

phenolic hydroxyl [estrogensI

methyl ether

[6] Conjugation

Unhindered hydroxyl groups 13, 17, 20?, 21, ?others]

6-glycoside or ester

FIG.4. Major eliminative reactions of steroid hormones and their metabolites. Hydroxylation reactions of biosynthetic tissues are not included, although they may be partly responsible for some of the observed metabolic transformations of exogenously administered steroids.

women suffering from ovarian hyperthecosis (Mla, M4, M5; see also B2). The biological significance of these secondary products in normal states is as yet unknown. In particular, the possible hormonal function of DHA-sulfate is an intriguing mystery in view of its large plasma concentration and secretion rates and its negligible androgenic activity. Its role as a provocative factor in acute porphyria has recently been suggested, and it has been shown to increase the activity of hepatic aminolevulinate synthetase (G6). The metabolism of the secreted steroid hormones is seemingly complex but can readily be understood in terms of a relatively small set of common biochemical transformations (B28).These major reactions are summarized in Fig. 4. Although the stereochemical properties and large number of positions open t o metabolic “attack” give rise to a tremendous number and variety of metabolites through these reactions, they are in fact typical “detoxication” reactions common to a large number of nonsteroidal substances. For many years the metabolism of steroid hormones has been regarded largely as an irreversible detoxicating activity carried out mainly by the

DETERMINATION OF STEROIDS I N PLASMA AND URINE

71

liver. It was believed that in most cases conjugation of a metabolite lead to its excretion without further significant metabolism. Minor exceptions to this idea were made in the case of some steroid sulfates because of the known moderate biological activity of estrone 3-sulfate1 the low renal clearance of steroid sulfates (see, e.g., B14, B15), and the known reversibility of sulfate formation. I n the last five years this idea has been modified fairly extensively in the light of new discoveries. I n the first place the discovery of the secretion of DHA-sulfate by the adrenal cortex in apparently major quantities suggested a hitherto completely unsuspected role of steroid sulfates in steroid biosynthesis and metabolism. Second, a number of steroid metabolites conjugated with two molecules of glucuronic or sulfuric acid were discovered, particularly in pregnancy urine, amniotic fluid, and in perfused placentofetal preparations. Finally, isotopic studies revealed several cases in which administered doubly labeled steroid conjugates were metabolized in the steroid part of the molecule without hydrolysis of the conjugate (e.g., D4, R5; cf. B5). These findings have lead some authors to imply that the older idea of the role of steroid metabolism and conjugation is now quite outmoded, and that an extensive recirculation and interactive metabolism of steroid metabolites and their conjugates takes place of potentially considerable physiological significance. The pyrogenic properties of intramuscularly administered etiochiolanolone and many other 5/3 (H)-3a-hydroxysteroids (K2) have also been cited in support of this notion, particularly after Bondy et al. (B12) and others suggested that patients suffering from idiopathic recurrent fever had abnormally high concentrations of ;neonjugated etiochiolanolone in their plasma. This extreme view seems to be unjustified on many grounds and may give rise to unnecessary confusion, so that some attempt t o present a more balanced picture seems to be in order. Many saturated Cz,-steroids similar to, or identical with natural steroid metabolites possess anesthetic or depressant properties (W8) and a large range of 58(H)-3a-hydroxysteroids cause fever on intramuscular injection (G5). Similarly androsterone and some other steroid metabolites are potent androgens (B28) , and the work of Baulieu et al. (B4) and of Bruchovsky and Wilson (B23) strongly suggests that the local conversion of testosterone to androgenically active reduced metabolites by some target tissues may be of great biological significance. Again, estriol, which is regarded as being a metabolite of estradiol or estrone, or a placental or fetal product in pregnancy (D4, G13, L2), is a weak “impeded” estrogen (H7). However, these facts take on a less spectacular aspect when their quantitative significance in normal human physiology is examined and when further

72

IAN E. BUSH

recent findings are taken into consideration. Among these considerations are the following. It is no longer possible to support the idea that 5P(H)-3a-hydroxysteroids are endogenous pyrogens in man, or, in particular, th a t free etiocholanolone in the plasma is the cause of a sizable number of cases of idiopathic recurrent fever. Using a highly specific and sensitive method for measuring free etiocholanolone in plasma, George et aL. (G4a) examined cases of such fevers and found that the minute concentrations of free etiocholanolone in their plasma, while slightly higher than in normal subjects in some cases, were within the normal range in the majority (35 _+ 4 ng/100 ml). Furthermore there was no correlation between the concentration of etiocholanolone in any of the patients with febrile and nonfebrile episodes of their disease. They and other workers have now convincingly shown that the pyrogenic effects of etiocholanolone in normal subjects are due to a purely local inflammatory reaction and are not caused by intravenous administration unless local inflammation occurs in the vein used for the injection. I n reviewing their results, these authors have shown that previous workers had not obtained sufficient evidence to justify the earlier claims for the existence of naturally occurring “etiocholanolone fever,” and they suggest that no such clinical entity exists. It is hard to disagree. The idea that saturated steroid metabolites may play a role in normal circumstances in regulating cerebral function is equally devoid of quantitative support a t present, although more specific actions on the brain of the steroid hormones themselves are well established and a growing field of interest (e.g., Ll). Thus, i t is reasonable to suppose a t present that the plasma concentrations of other free saturated steroid metabolites are similar to those of free etiocholanolone determined by George e t aL. (G4a), namely in the range 30-50 ng/100 ml. I n contrast, Bush (B26) has measured the concentrations of the free (saturated) metabolites of 21-hydroxy-5P-pregnan-3,2Q-dione-21 -hemisuccinate (Viadril) , which are most probably responsible for the anesthetic action of this agent, a t the onset and offset of anesthesia and found them to be in the range 1 W 2 0 0 pg/lOO ml. Again it should be remembered that the central depressant effect of progesterone and other steroids first discovered by Selye in 1941 (see W8) was only convincingly demonstrated with massive doses in partially hepatectomixed animals. With the possible exception of progesterone in late pregnancy (plasma concentration in the range 10-60 pg/lOO ml (S3), it seems highly unlikely that the generalized central depressant actions possessed by many steroids when given in large doses have any biological significance in man under normal conditions. Nor is there yet any demonstration that steroid hormones or their metabolites

DETERMINATION OF STEROIDS IN PLASMA AND URINE

73

are responsible for nonspecific central depressant actions in any rtbnormal states. Another factor to be taken into account is the enterohepatic circulation of steroids and their metabolism by the flora of the alimentary canal (Wl, W2). Long suspected as a complicating factor in this field, its major quantitative significance, a t least in the rat, has now been demonstrated convincingly by Sjiivall, Gustafsson, and their colleagues using “gemfree” animals (E4, G14, G15). Its quantitative significance in man remains to be established, but the biliary excretion of many steroid metabolites in man is well established (e.g., A l l A3, J2), and the partial hydrolysis of some steroid glucuronides by endogenous or bacterial pglucuronidase in occasional samples of urine has also been demonstrated. Although the existence of significant metabolism of steroids by the alimentary flora could give rise to some form of commensal metabolic interdependence, two facts seem to be of overriding importance a t present. First, the net fluxes of most of the reactions leading from the secreted steroid hormones to their conjugated urinary metabolites are overwhelmingly greater than the fluxes of the reverse reactions. Second, the concentrations of free steroid metabolites in plasma are vanishingly small, and only a minor fraction of them are known to possess any biological actions a t all except in large or local concentrations. At present, therefore, i t does not seem justifiable to abandon the earlier general concept that the majority of the reactions undergone by steroid hormones after they have been secreted function largely as metabolic pathways leading to the formation of relatively rapidly excreted inactive end products. The close relationship between these pathways, in general and often in detail, and those for the elimination of other classes of endogenous and exogenous substances, perhaps needs reemphasis (B7, B28, W6). The latter point is in fact emphasized most eloquently by another major discovery of recent years, which has a direct bearing on the metabolism of steroids. This is that a large number of hepatic enzymes responsible for the metabolism and elimination of drugs and other exogenous substances appear to be grouped together in the smooth endoplasmic reticulum ((33). The activities and amounts of these enzymes in the liver are increased in concerted fashion, as a group, by a wide variety of agents, some of the most striking effects being seen with barbiturates, alcohol, certain tranquilizers, and pesticides. These enzymes include many hydroxylases, nitroreductase, and some dehydrogenases, and are all dependent upon NADP and cytochrome P-450 for electron transport. It is now clear that administration of barbiturates greatly increases the proportion of certain steroids which are eliniinated as hydroxylated metabolites. Striking

74

IAN 1. BUSH

effects of barbiturates in increasing the excretion of 6~-hydroxycortisol in guinea pig and human urine have been discovered (see C3, B24). Similarly SjGvall and his group (S6) have shown that low doses of ethanol produce an immediate and profound effect upon the ketone: hydroxyl ratio of several steroid redox-pairs. I n view of the extremely frequent consumption of barbiturates, tranquilizers, and alcohol both in and out of hospitals, the interactions of these agents with steroid metabolizing enzymes deserve careful study and may well require the reexamination of many inferences drawn from past work. Isotopic studies of steroid metabolism have been frequent since about 1954. The classical isotope dilution method was first introduced in the steroid field, using urinary metabolites, by Pearlman et al. (P5). I n 1957 Pearlman (P4) pioneered the method of measuring the steady-state clearance from the blood of an isotopically labeled steroid administered by constant intravenous infusion. If a steady stat,e is achieved, a knowledge of the rate of infusion of the isotopic steroid and its specific radioactivity, and measurement of the specific radioactivity of the steroid in the blood, yields the net flux of endogenous steroid entering the blood, which can usually be equated with its secretion rate. The same concepts were employed by Peterson (P5a, P7) in the investigation of cortisol metabolism. Subsequently Tait (T1) introduced the metabolic clearance rate (MCR) as a combination of Pearlman’s approach with Homer Smith’s well-known concept of renal clearance. I n this treatment the clearance rate of a hormone from the blood is expressed as the (conceptual) volume of blood which would contain the amount of steroid “cleared” from the (actual) blood per unit time. At the same time, Gurpide et al. (G12) introduced the concept of fractional interconversion rates or p values. This is used unfortunately to define two sorts of change. One is the fractional rate of transfer of a substance from one compartment or “pool” to another. The other is the fractional rate of metabolic conversion of one substance to another. Further refinements and complications of these concepts are described in the original papers, in the review by Tait and Burstein (T3), and in Horton and Tait (H6). These later concepts follow the traditional practices of chemical physiology in providing parameters which facilitate the calculation of metabolic exchanges. Those outside the steroid field may well regret the seemingly unnecessary departure of these treatments from the wellestablished usages of chemical kinetics and thermodynamics, and physiologists and endocrinologists will find some of the jargon unnecessary, confusing, and often conflicting (see, e.g., T4, p. 403). However, Gurpide’s

DETERMINATION OF STEROIDS IN PLASMA AND UBINE

75

derivation of the p parameter is a valuable achievement in the facilitation of the analysis of isotopic experiments in this field. The value of these more complex approaches to the secretion and metabolism of steroids in clinical chemistry remains to be assessed. At present their use of complex techniques, or in some cases of potentially dangerous isotopes, confines them to the research laboratory. However, some of the results achieved by these methods are of importance to the clinical chemist and will be reviewed briefly. One of the commonest problems laid at the door of the clinical chemist involved in endocrinology is to estimate the secretion rate, or some parameter thereof, of a steroid hormone. Is it subnormal, in the normal range, or abnormally large? Any procedure not using the metabolic blood clearance technique assumes either a simple correlation between the plasma concentration and the secretion rate of the steroid in question, or a simple correlation between its secretion rate and the excretion rate of a known urinary metabolite. In the case of the classical urinary isotopic dilution method the urinary metabolite chosen must be a unique metabolite of the steroid whose secretion is to be estimated. The more modern isotopic methods have been of great value in demonstrating the inaccuracy of earlier simplified assumptions that particular urinary steroids were unique, or nearly unique, metabolites of particular secretory products. An excellent introduction to this occasionally difficult field is the paper by Lipsett et al. (L12) showing that a large fraction of urinary testosterone glucuronide in women is derived from the intrahepatic conversion of androstenedione to testosterone, which is in a metabolic “pool” that is not exchanged with plasma testosterone. Testosterone glucuronide is thus not a unique metabolite of plasma testosterone and cannot be used as a means for estimating testosterone secretion in women by the urinary isotopic dilution method. I n men this fraction is too small to invalidate the method. On the other hand, it cannot yet be claimed that the newer methods have produced a sizable increase in the scope and precision of our knowledge of the majority of the kinetic aspects of steroid secretion and metabolism. Despite the diligent work of many people, the errors (e.g., T4), and particularly the statistical treatment, of the often complex compartmental models that are used are of uncertain magnitude (e.g., B8). The assumptions governing such models are often of doubtful validity but of critical importance (F7, T3). Another major difficulty of these, and indeed of most, methods for the assessment of the kinetics of steroid secretion and metabolism in man is that the time required for the infusion and sampling of plasma is relatively long-in the case of progesterone assessed by the steady-state

method, excessively so (P4). These times are long not only when compared with the known time-course of sizable changes in the secretion rates of many steroid hormones, but in some cases long even when compared with the slow diurnal variation that is seen with adrenocortical secretion. The time required for the complete or 95% dilution method is even greater, varying from 12 to 72 hours. Once again one could reasonably advise the clinical chemist to keep in touch with this type of approach to the study of steroid secretion and metabolism but to avoid being inhibited by it. A tremendous amount of useful clinical chemistry in this field can in effect be carried out, and justifiably, by the judicious selection of relatively simple tests using nonisotopic methods. 1.4.

MAJORDISORDERS OF SECRETION AND METABOLISM

. . . some kinds have heads too large for peeling the skin over . . . (P9,p. 423) There is no completely satisfactory way of classifying the disorders of the secretion and metabolism of steroid sex hormones. The morphological pathologist and the chemical pathologist will tend to use different classifications. The endocrine disorders involving steroid sex hormones follow very similar basic lines of pathophysiology to those of the thyroid gland but are more complicated because of the existence of a t least five recognizably distinct types of steroid-secreting tissue in mammals. Since all these tissues have a common embryological origin and a large number of crucial enzymes in common, i t is possible, and in most cases known to happen, that the inappropriate secretion of a steroid hormone of almost any type can arise from a disorder of a minimum of five functionally distinct tissues in the gonads or adrenals. In turn these disorders may arise by disease of the tissues themselves or by a disorder in the secretions of a minimum of three anterior pituitary hormones, either from the pituitary itself or from certain malignant tissues (L10). Finally the picture is complicated by the diurnal and other sorts of variation of adrenocortical activity and the changing secretory activity of the ovary with the menstrual cycle, pregnancy, and age. I n such a complex field it is well for the clinical chemist to try to achieve a clear idea of exactly how his laboratory tests can contribute to the activities of his clinical colleagues. It is fair to observe that in most problems of differential diagnosis in this field steroid measurements provide strong corroborative evidence for a decision rather than conclusive and critical evidence. The latter can usually be obtained only if the measurements are part of a sophisticated and lengthy investigation. However, such corroborative evidence is of more than trivial value in

DETERMINATION OF STEROIDS IN PLASMA AND URINE

77

this field, and besides has a useful research function in building up experience and in occasionally providing the opportunity for new discoveries. Excessive secretion of androgens can occur in both sexes. I n infants, children, and young adolescents the commonest cause is a biochemical defect in a n otherwise normal adrenal cortex. The result is the adrenogenital syndrome whose signs and symptoms depend on the severity and nature of the biochemical lesion and the time a t which the disease is noticed. The fundamental error of metabolism was first elucidated by Bartter and Albright and their colleagues. The detailed biochemical studies of Bongiovanni’s group and the clinical and biochemical work of Wilkins and his colleagues have given us an extremely complete picture of this disease (e.g., B10, B13, B15, G 3 ) . The severe congenital form of the disease causes congenital pseudohermaphroditism in which the enlarged clitoris or pseudo-hypospadic penis often causes a misidentification of the sex of the newborn baby. I n boys macrogenitosomia praecox results with its characteristic picture of precocious puberty in the presence of normally infantile testes. I n childhood and adolescence, the excessive concentrations of androgens cause premature fusion of the epiphyses and a characteristic, short-limbed, stubby appearance. The syndrome is due to the combination of the enzymatic defect producing a deficient proportion of cortisol and corticosterone in the secreted products of the adrenal cortex, and the normal response to this deficiency of the system regulating the secretion of ACTH. I n the absence of the negative feedback of a normal concentration of plasma cortisol, ACTH secretion is increased. Since the action of ACTH on the adrenal cortex is to stimulate a very early step in the biosynthetic pathway (Fig. 3), the production of all products subsequent to this step is increased. The commonest enzymatic deficiency is a relatively severe or total absence of steroid 21-hydroxylase1 but minor or severe deficiencies in the 11P-hydroxylase, and in A5-3p-hydroxysteroid dehydrogenase also occur. As with any metabolic defect of this sort, products of the reactions prior to the deficient step reach abnormally high steady-state concentrations, and their conversion by side branches of the main biosynthetic pathway to usually minor or trace metabolites may become spectacularly prominent. The diagnosis of this condition is usually fairly clear-cut in early life, and there is no situation in which the clinical chemist can offer more valuable and clear-cut aid to the endocrinologist. All these young patients are a t risk from sudden death in adrenal insufficiency, and early diagnosis and treatment are essential if stunting of growth and disturbance of sexual function in later life are to be avoided.

78

IAN E. BUSH

T he effects of excessive androgen secretion in late adolescence and adult life are much more various and less easily analyzed. The largest group of cases are young women suffering from hirsutism, acne, and varying degrees of menstrual irregularity or with complaints of infertility. Many patients of this type are obese and can be categorized in terms of the original criteria for the Stein-Leventhal syndrome (see B2, H5, M l a , M5). Many others lack the full clinical picture but are clearly closely related to this syndrome because enlarged polycystic ovaries are discovered and confirmed by surgical treatment. Finally there is a sizable group who lack the ovarian lesion but who are often severely virilized with or without serious menstrual disorders. It is now generally agreed, as the result of increasing numbers of such cases who have been carefully studied by more recent sophisticated analytical methods for steroids, that a wide variety of adrenocortical enzymatic defects of almost all degrees of severity can be found in many of these patients. Although some of them may be mild, or very occasionally quite severe cases of congenital adrenogenital syndrome not discovered in early life, the majority have been free of symptoms until puberty or some years thereafter. The metabolic defect that is discovered is probably of the same general type that is seen in the congenital or prepubertal adrenogenital syndrome. The first suggestion of such “transitional” forms of the adrenogenital syndrome was made by Kappas et al. ( K l ) . An interesting example of two cases in this category and a useful review of the literature up to 1965 is given by Gabrilove et al. ( G l ) . A larger group of similar cases who suffered from hirsutism and acne, with an approximately 50% incidence of serious menstrual irregularity, but who also had no detectable defect in cortisol secretion were first demonstrated by Nugent e t al. These patients have been shown to have a distinctly elevated excretion of testosterone glucuoronide (M3) and of plasma testosterone concentration (B2, H5). They were distinguished by the absence of any other major abnormality of steroid metabolism (although total urinary 17-ketosteroids excretion were in the high normal range or up to twice the upper limit of normal) and by the fact that the excess testosterone appeared to arise from a tissue sensitive to ACTH. Thus, small doses of Dexamethasone or of other active glucocorticoids will usually suppress the plasma concentrations and excretion rates of testosterone to normal and achieve useful or even striking therapeutic results (B2, G11, H5, M3). This is another group of patients whose fate can be critically dependent upon reliable chemical findings. Many clinicians are loth to embark on adrenal suppression therapy in the absence of more severe symptoms or of gross abnormalities in the standard “group” tests (see Section

80

IAN E. BUSH

Mahesh and Greenblatt (Mla) , however have shown th a t large amounts of androstenedione and DHA are secreted into the ovarian venous blood of patients with polycystic ovaries. Although DHA is a poor precursor of plasma testosterone, androstenedione, probably mostly secreted by the adrenal cortex, gives rise to approximately two-thirds of the plasma testosterone in normal women (H6, T4,W4). The biological significance of androgens or their precursors secreted by the ovary remains doubtful a t present, both in normal women and in the polycystic ovary syndrome, because of the lack of complete and consistent information. On the other hand, the sizable number of observations which fail to show a definite role of adrenal androgens in the polycystic ovary syndrome suggests that ovarian androgen secretion may well turn out to be of major importance in a sizable fraction of cases.

1.5. PARAMETERS OF CLINICAL SIGNIFICANCE To sex a bird, slit through the ribs on one side, and look for the the testes or ovaries near the front end of the pelvis. (P9,p. 60)

The preceding review of recent work on the major disorders of steroid sex hormone secretion, although lengthy enough, is far from comprehensive. Even so, many of the topics discussed so far are a t present too complex and sometimes too controversial to be of immediate application in routine clinical chemistry. Even though the boundary between research and application is a diffuse one in this field, it is reasonable to suggest that the major task of clinical chemists in this field is to define those sorts of measurements which come nearest to providing really crucial diagnostic aids to the clinician. While the specialist rightly seeks to achieve the maximum precision in determining the source and fate of steroid sex hormones in normal and diseased states, the clinical chemist has to seek the most economical means of achieving answers to rather different questions. Thus, the “source” of a steroid in blood is, to the specialist, a functional definition in terms of compartments, pools, and tissues with certain biosynthetic and hormone-responsive characteristics. The identification of these with anatomically recognized structures is inferential and, although usually reliable, is often not crucial to the argument. To the clinical chemist, on the other hand the “source” is more often an anatomical structure which must be identified a s a place for the surgeon to start, or not to start, cutting. The objective clinical chemist will realize without too much humiliation that this identification is more often made by a good X-radiograph or by vaginal palpation than by a chemical investigation. I n what follows I will try to provide a balanced guide to those areas of investigation which seem to offer the best returns on investment of time, effort, and equipment. The two biggest methodological questions facing the clinical chemist

DETERMINATION OF STEROIDS IN PLASMA AND URINE

79

2.3) for urinary or plasma steroids. Since many of these patients seek medical help only after several years of fruitless cosmetic treatments, often after previous rebuffs from doctors, loss of hair and full restoration of skin texture under suppression treatment is slow, and neither patient nor clinician should be allowed to be discouraged from persisting in the treatment for several years. A clear-cut demonstration of an abnormality in plasma or urinary testosterone levels is of crucial importance in establishing a confident diagnosis capable of justifying the therapy. It should be noted however that Casey and Nabarro (Cl) failed to find any significant elevation of plasma testosterone concentrations in 40 clinically similar cases of idiopathic hirsutism. The exact nature and pathogenesis of the Stein-Leventhal syndrome is still unknown, and considerable doubt has often been cast on the validity of this syndrome as a genuine clinical entity. Useful descriptions of recent work with isotopic methods and good reviews of other recent literature in that field are given by Horton (H5) and Bardin et al. (B2). A wider survey of this field is given by Greenblatt (G11). Two major trends can be discerned in the work of the last fifteen years. First, as the result of Dorfman’s original suggestion that attention be focused on pIasma testosterone in the study of virilism and hirsutism, more and more studies have demonstrated increased concentrations of testosterone in plasma, and increased secretion rates (or MCR’s) of testosterone. Second, there has been a steadily increasing number of observations demonstrating a close relationship between this disease and the postpubertal form of the adrenogenital syndrome. As far as the definition of the disease is concerned, ovarian polycystic disease with hyperthecosis has clearly emerged as the single unifying pathological feature from the original findings of Stein and Leventhal. More recently, a fair number of cases have been observed in which the excessive testosterone concentrations in plasma or other abnormalities of androgen metabolism have been rapidly reduced to normal levels by the administration of normal maintenance doses of glucocorticoids although this has not been observed by others (cf. B2). This is now regarded as fairly clear-cut evidence that the source of plasma testosterone or of its precursors must have been the adrenal cortex or some presently unknown ACTH-responsive extraadrenal tissue. A reasonable suspicion arises that an excessive secretion of androgens by the adrenal cortex may be a major causal factor in the polycystic ovary syndrome. Thus, the actual secretion of testosterone by the human adrenal cortex has been demonstrated in some cases by direct sampling of adrenal venous blood. Similarily polycystic ovaries have been reported in several clear-cut cases (i.e., severe and not “transitional”) of the adrenogenital syndrome.

DETERMINATION OF STEROIDS IN PLASMA AND URINE

81

are: (1) whether to invest heavily in well-established but only moderately specific “group” methods for measuring steroids or to develop and use the more specific methods for individual steroids; and (2) to assess the relative value of measurements in blood plasma and urine. These two questions are partly interdependent and are also dependent upon the methods chosen for measuring cortisol and other adrenocortical steroids outside the class of sex hormones and their metabolites. Thus, each clinical chemist will want to reach an optimal strategy based on the size and budget of his laboratory and on many other local factors. It is sometimes valuable to use a good commercial service laboratory for certain types of measurement whose inclusion in the routine armamentarium of the laboratory would prevent the introduction of other more desirable types of measurement. In any event he will wish to minimize the number of different basic techniques and types of equipment to be used unless he can afford to allocate a very sizable number of staff to steroid work. At the outset it is fair to suggest that the more specific single-steroid methods need a minimum of one full-time senior worker for each method to carry out a reasonable number of measurements per week on a routine basis. Even the less specific “group” methods need a large fraction of the time of a technician of superior quality for the reliable handling of the average work load of a large hospital. Quite apart from the actual performance of the method, a lot of backup is required in terms of purification of solvents and reagents even for the simplest methods. In order to provide a reliable battery of several tests of steroid functions on a routine basis, it is usually necessary to have three or more technicians and one staff member as a minimum team devoted almost entirely to steroid work. The relative value of blood and urinary measurements is a long-standing source of controversy. Arguments about the relative “fundamental” value of such measurements in endocrine research are rarely relevant to routine clinical work. The first priority for the clinical chemist must be the value of the chosen measurements as a discriminant in a differential diagnosis. Despite fashionable implications to the contrary, there are few disorders involving steroid sex hormones in which properly selected urinary measurements are not as valuable as the alternative measurements in blood. More important, there are no steroids of wellestablished diagnostic value which cannot be measured in human urine by “group” methods or relatively simple, more specific methods. I n contrast, a large number of diagnostically important steroids in plasma cannot be measured on a routine basis by the simpler methods, although the situation has improved for some of them with the introduction of isotopic-displacement methods. None of the steroid sex hormones can

82

IAN D. BUSH

be measured in plasma with optimal specificity without using a procedure containing a minimum of one chromatographic separation. The major advantage of measurements on blood is that direct extraction without prior hydrolysis of conjugates is all that is usually needed, and that collection of the sample is rapid and volumetrically reliable. The latter is of importance, but not essential, in allowing the rapid performance of adrenocortical suppression and stimulation tests which are now needed in the differential diagnosis of virilism in women. Despite the recent recognition that few urinary steroids have unique circulating hormonal precursors, the few that are unique or approximately unique in clinically important conditions are of considerable diagnostic value. They include the major fractions of urinary cortisol (C5,cf. F7), pregnanediol (P4, P5) , 3,3’,4,5-tetrahydro-ll-deoxycortisol, tetrahydrocortisol (THF), and tetrahydrocortisone (THE) ; and the cortols and cortolones (cf. F7); the pH 1-labile conjugate of aldosterone (T3), and (F3). While probably in most cases ll-ketopregnane-3a,l7~,2Ot~-triol many others result from several, or an unknown mixture of, precursors, their urinary excretion rates in disease or during adrenocortical suppression or stimulation tests are of great use in differential diagnosis. The fact that plasma concentrations are more closely related to the effective biological significance of the hormones is not only complicated by protein-binding (K4, T3) , but usually redundant in routine clinical chemistry-the signs and symptoms of abnormal biological activity are already sufficiently obvious to have brought the patient to a doctor. The point of these remarks is not to decry the great value of modern sophisticated methods of steroid analysis in research and thus in the elucidation of the chemical pathogenesis of many disorders involving steroid sex hormones. I n general the time is ripe for an expansion of the intelligent use of modern methods in routine clinical chemistry, and an extension of the scope of steroid research to a wider range of clinics. However, i t is fair to point out that much more could be done i m m d i ately by the intelligent routine application of some relatively simple techniques to blood as well as urine, and that the popularity or relevance of a particular type of technique in basic research is not necessarily the best guide to its usefulness in clinical chemistry. A common problem in the assessment of the milder disturbances of steroid sex hormones is the rather large range of normal values for most steroid measurements given in the literature. There is now enough information in the literature to suggest that these ranges could be reduced by taking into account the weight, age, thyroid function, and psychological state of the patient. The collection of “normal” control values should be done so as to take this into account. In some cases the identification of particular steroid metabolites in

DETERMINATION OF STEROIDS I N PLASMA AND U RIN E

83

urine may be of great value, and the question arises of how feasible this is outside the research laboratory. Thanks to the great volume of research in this field over the last fifteen years, it is usually possible to do this quite simply without satisfying all the stringent criteria that were necessarily required in the original research. The use of simple chromatograms on readily available thin-layer plates or paper with carefully selected methods of detection will often suffice for the identification 15-20%) of certain abnormal and approximate visual measurement (t metabolites, which are often present in relatively large quantities. Thus, for instance, the enormous excretion rates of DHA-sulfate in many cases of adrenocortical carcinomata (e.g., B3, L10) could be detected reliably in just over an hour (15 minutes working time or less) with a variety of simple procedures using 0.2 ml of urine or less. Similar although slower techniques could also be used for the detection of pregnanetriol and its ll-oxygenated derivatives in the common form of the adrenogenital syndrome, and for the demonstration of urinary THS in the rarer form involving llp-hydroxylase deficiency (GI, W7). Although this viewpoint is somewhat heterodox a t present, a little reflection will, I believe, show that it is a rational and justifiable one. The only serious trap lies in possible confusions with drugs or certain foodstuffs. However the latter are still more likely to interfere with group methods [see Gray et al. (G10) for a valuable survey of these substances] than with methods including a chromatographic step, and must always be guarded against. It will be remembered that the recognition and diagnosis of the aminoacidurias was revolutionized by the application of such techniques in the late 1940’s by Dent and that similar methods are now in routine use throughout the world for the early detection of phenylketonuria and other gross metabolic disorders. It goes without saying that whatever methods of measurements are used, stimulation and suppression tests using ACTH, prednisone, or dexamethasone, and in some cases chorionic gonadrotropin, other pituitary tropic hormones and steroid sex hormones, are an essential part of any chemical diagnostic procedures in this field. 2.

The Assessment of Methods

2.1. THEDEVELOPMENT OF TECHNIQUES

FOR

STEROIDS

If I could then have known borax mothproofing, as I have developed and perfected it during the past thirty years, my early experience would have been completely happy. As it was, arsenical methods came near to being my finish. (P9,p. VIII)

The decision that a steroid hormone or metabolite is worth measuring is derived originally from the discovery and investigation of a biological

84

IAN E. BUSH

phenomenon. While the modern philosophy of the undersirability of biological assays in endocrinology is a very reasonable and fruitful one, it is perhaps useful a t the present time to drop a hint about the consequences of complete adherence to this doctrine. Thus, a t a time when most investigators in this field were concentrating on the development of newly available physical and chemical methods for measuring steroid hormones in minute amounts, Simpson and Tait discovered aldosterone by the imaginative combination of paper chromatography with a biological assay ( G l l a ) . The study of the prostaglandins is still heavily dependent on biological assays. After the early stages of biological assay, the methods for measuring steroids have all gone through certain characteristic phases. A brief sketch is useful in placing the more recent developments in a balanced context. At first, attempts were made to use simple physical properties of characteristic groups in the steroid molecule. Because of the low concentrations of steroids in most situations these relatively crude methods usually suffered from low recoveries and a more or less serious lack of specificity. Another difficulty was the early necessity of using rather drastic chemical methods for the hydrolysis of urinary conjugates. The numerous artifacts caused by this technique and also by the then common use of adsorption chromatography on alumina lead to a quite reasonable obsession with the dangers of instability and artifact production. I n more recent years the extensive use of infrared spectroscopy, isotopically labeled steroids, high resolution chromatography, and mass spectroscopy has revealed the frequency with which supposedly “pure” steroids extracted from natural sources are contaminated with known or unknown compounds. The newcomer to this field may find it difficult to get a balanced view from the current literature. It will sometimes seem to him as though a mass spectrometer is the minimal equipment needed to venture beyond the routine performance of an estimation of total 17-ketosteroids. Once again my advice would be, “Don’t let yourself be ‘bowled over’ by all this.” With relatively cheap modern techniques it is still a t present more difficult to make a reliable measurement of urinary or plasma cortisol than of plasma glucose-but not that much more difficult. There is plenty of scope for the enterprising clinical chemist to take advantage of these opportunities. Nor, as we shall see, should he accept on trust the common implication that the objective specifications of the more complex recent methods are so much better than his own more modest procedures. After a close look a t some of the complex methods, he may well agree with a sceptical visitor to a recent fashion show who remarked, “All that glitters is not brass.”

DETERMINATION OF STEROIDS IN PLASMA AND UBINE

85

2.2. SPECIFICITY Lay the specimen on an old blanket t o keep it clean while skinning; a litter of forest rubbish results in a miserable job. (P9,p. 4)

One of the greatest bugaboos of the study of steroids is the problem of specificity. The majority of “group” methods for plasma or urinary steroids, and the simpler methods estimating one or two steroids in plasma without a chromatographic step (e.g., simple fluorometric methods for cortisol) are moderately nonspecific. Even the deservedly popular Nelson and Samuels method (N4) for 17-hydroxycorticosteroids in plasma, which includes a crude adsorption chromatographic step, probably overestimates the real value by 15-30% a t normal levels. With the profusion of new drugs in wide use, especially soporifics, tranquilizers, and contraceptives, the specificity of any method needs continuous review and checking (G10). However, specificity is essentially a question of probabilities, and more use could be made of quantitative approaches to its evaluation (B27, B32, B33a). Even though no exact magnitude can be assigned to the concept of specificity, valid preferences can be established by the careful use of inequalities and known parameters of separation methods. Too many discussions of this topic are woefully inadequate because their authors use inappropriate parameters and functions of them. Some are theoretically sounder but fail to take account of well-known empirical problems such as the dangers of impurities in solvents and reagents. Finally, i t should be obvious that the specificity claimed for a method of measurement is critically dependent upon the nature and source of the sample. The severity of the problem with steroids is simply due to the very large number of steroids which actually exist or may possibly exist in nature, and the relatively low concentration of most of the interesting ones in blood, urine, and tissues. To this must be added the large number of organic substances with closely similar physical properties and many similar chemical features. I have emphasized elsewhere that almost all chemical analytical methods depend for their specificity on two classes of procedural steps, namely, purification steps, and a final signalgenerating step (B33, B33a). I n some cases the signal-generating step depends on a chemical process which also enables an additional purification step to be achieved. Examples of this are the extraction of the Zimmerman chromogen prior to colorimetry (P6) and the esterification of hydroxysteroids with isotopic acetic anhydride followed by rechromatography of the labeled ester. Because of its probabilistic nature the overall specificity of a method

86

IAN E. BUSH

must be some product function of the specificities achieved by each of the steps of a procedure. This has an important bearing on the operational characteristics and design of procedures where specificity is a major problem. Since this has not been explicitly recognized by the majority of workers in this field it is worth a little attention here. It can be shown rigorously that the separation of a substance to be measured from an undesired contaminant is an additive function of the chemical characteristics of the two substances when the distribution isotherms between the two phases are linear. Although the exact function cannot always be obtained because of vicinal effects, this fact is of great value because the role of a t least four major types of interaction between the substances and the partitioning phases can be quantitatively assessed. Thus it can be shown quite rigorously that a succession of partitions (either chromatographic, countercurrent, or simple extraction and washing) will always have a better chance of improving the specificity of the purification stages of a procedures than the use of a lesser number of steps devised empirically so as to minimize empirically known sources of interference. This fundamental consideration enters into the design of a procedure a t all stages from simple extraction through to the choice of phases to be used in liquid-liquid (LLC) or gas-liquid (GLC) chromatography (B33a). This is very relevant, for instance, in the oft-discussed problem of the relative value of selective, nonselective, or “mixed” stationary phases in GLC (see, e.g., L8, pp. 265,286, and 292; E2, KlO, T7, T9). The empirical selection of a mixed phase can undoubtedly achieve an improved separation of particular pairs of steroids, but as a one-step solution i t is a bad strategy because it offers no guarantee against the risk of reduced separation from other steroids or nonsteroid contaminants. This can often outweigh its obvious operational advantages. A similar criticism applies to many existing methods for the measurement of specific steroids by isotopic labeling. Because of the extreme difficulty of eliminating added impurities present in or formed during the use of radioactively labeled reagents of very high specific activity (Gla, R1, R2), most of these methods concentrate on an extensive sequence of chromatographic separations after the labeling step. The use of a sequence is desirable, but with few exceptions (e.g., G l a ) these methods usually use sequences of solvent systems, either paper LLC or TLC, in which it is already known that the solute-solvent interactions are quantitatively quite similar except for the factor of molar volume (B33a). Such sequences are far less efficient than is possible with existing solvent systems. The relative effectiveness of different chromatographic methods of

DEITERMINATION OF STEROIDS IN PLASMA AND URINI

87

separation in separating closely related compounds is also obscured by most discussions in the literature. The view is often expressed that particular sorts of method are greatly superior to others in this and other characteristics. Global statements of this sort are invariably meaningless without reference to the particular tasks of separation that are involved. The efficiency of separation of a chromatographic system, and thus the maximum specificity that can be achieved by one pass through it, is determined simply by a product function of the number of theoretical plates achieved in the system and the difference in mobilities of the two substances to be separated. The appropriate function of mobility for simple additive manipulation is the R, value in thin-media chromatograms, and the logarithm of the corrected retention volume for effluent column chromatograms of whatever type. A full review of this treatment has been given previously (B33a) and an excellent recent account of the relationships between paper LLC and GLC is given by Thomas (T7). As would be expected the separation factors expressed in AR, values (which have the great value of enabling a direct quantitative comparison between different types of chromatography) are lower in GLC than in LLC, probably largely because of the different temperatures employed. Yet, the efficiency of resolution of GLC is, for many pairs of substances, still greater than the majority of commonly used LLC systems because of the much larger number of theoretical plates that can be achieved in GLC in practice. Thus, the average 3 X overrun (16 hours) paper chromatogram achieves approximately 1600, the average single-length TLC chromatogram (1 hour) approximately 800, and a good 6-foot GLC column with SE-30 packing may achieve 3500 (if packed by an expert at retention times of ca. 30 minutes). An excellent discussion of these basic factors in GLC is given by Horning et al. (H4; also E2, p. 1). It is seen that the general superiority of GLC for this class of substances is not so much in the resolution t o be expected per system but in resolution achieved per unit time. The relative efficiency of these different sorts of chromatography is seriously affected in practice by nonideal behavior, by the formation of artifacts, and by limitations in the choice of phases, which are particularly severe in GLC. Now, the superiority of the usual GLC machines in number of theoretical plates per system (approximately 2-fold) compared with conveniently overrun (overnight) systems using LLC on paper is approximately canceled by the reduction in useful AR, values in GLC mentioned above. In more detail (B33a, T7) the AR, values of ketone groups, hydroxyl groups, or their trimethylsilylether analogs, range from one-sixth to just under one-half the values for the comparable groups in common

88

IAN E. BUSH

LLC systems for steroids. A GLC system therefore must achieve a number of theoretical plates approximately six times the number achieved in low-temperature LLC systems to achieve rough parity of performance in terms of the resolution of hydroxysteroids per system with currently available stationary phases, and twice the number for steroids differing by number of ketone groups. Such values can be achieved with expertly packed long GLC columns, Golay GLC columns, and with runs of 1-2 hours. Careful inspection of the literature will reveal that such conditions are not often used in published methods for steroids, despite the empirical recommendations of Horning and other pioneers in this field of technique. Confusion on this matter is probably largely due to the common preference of running single-length LLC chromatograms on paper (3-4 hours) despite specific recommendations that much better results can, expectedly, be obtained by 2- to 3-fold overrun chromatograms which are conveniently achieved overnight with 48-60 cm long strips (B26, B27) or in 2-5 hours on small strips (B34). Thus, it is no surprise to find on careful inspection of the literature that some separations of closely related steroids which are relatively easily achieved by LLC are also easy with TLC or adsorption chromatography on columns but are quite impossible on a 6000-plate GLC system using several stationary phases [e.g., androsterone and etiocholanolone, using NGA (E2, H4) 1. Others, like the separation of testosterone and 17-epitestosterone, are difficult with both GLC and TLC but easy with LLC. Finally a number of straight-chain geometrical isomers, and nonpolar steroids differing only by single unconjugated double bonds are easily separated on GLC but difficult or impossible with LLC, and difficult on T L C unless AgN0,-impregnated plates are used. The point of this discussion is not to disparage one or other method, but to suggest that the balance of comment involving value judgments in the literature of the last five or six years has been somewhat one-sided and often has little foundation in fact. The general clinical chemist needs a t least a warning to this effect since his decision on what methods to adopt may be seriously and expensively affected by these matters. The specificity provided by the different chromatographic methods used in modern steroid techniques is considerable by all the principal popular methods when properly used and is by far the major factor in the specificity of all currently popular procedures. However, for the majority of problems of separation that arise, GLC as commonly used in the steroid field is only moderately better as a whole than any other form of chromatography when judged quantitatively. By the use of appropriate conditions the general advantage of GLC over LLC or T L C could probably

DETERMINATION OF STEROIDS IN PLASMA AND URINE

89

be increased by one or two orders of magnitude without serious technical difficulties, but the majority of published methods use conditions in which the resolution achieved is usually no better, and often somewhat worse, than standard 16-hour overrun paper chromatograms although decidedly better than single-length (3-4 hour) chromatograms on paper and T L C plates. The specificity derived from the signal-generating step of popular procedures for steroid hormones varies widely from method to method. Thus fluorescence or color reactions in which the mechanism of reaction and the stoichiometry are known, provide a far more certain basis for assessing the risk of interference from expected or unexpected contaminants than the use of reactions of unknown mechanism. In most cases however, this is not an area which can be treated quantitatively with any degree of confidence, and careful empirical checking is necessary with a wide range of the samples for which the technique is designed. The many types of fluorescence reaction with concentrated strong acids are of largely unknown mechanism and are subject to interference by quenching as well as by enhancement due to impurities. The most sensitive modern methods for steroids nearly all use some form of labeling technique to achieve a highly sensitive signal-generating step. The specificity of the labeling reaction itself is fairly low but the formation of a derivative provides an additional increment of specificity in the chromatographic step or steps which usually follow the introduction of the label. The four commonest methods are the formation of 35S-thiosemicarbazones of ketosteroids (Rl, R2) , of 3H-labeled acetates of hydroxysteroids (see B18), and of chloroacetates (B22, V1, V2) or various polyfluorinated esters (C2, V4) of hydroxysteroids. The latter two methods provide halogenated derivatives giving exceptionally strong signals with the electron-capture detector in GLC. Unlike the almost nonspecific but most commonly used flame-ionization detector in GLC, this detector is highly specific for halogen atoms and certain polyunsaturated groups. Despite the relatively unspecific reaction used in these methods the specificity and sensitivity of the signal-generating step is greatly increased. The use of other microchemical conversion reactions is very common in this sort of work as a part of the purification stages of many procedures (e.g., G l a ) . The use of such reactions for the formation of heat-stable derivatives in GLC, which is essential for many classes of steroid, also provides an improvement in the purification steps of the overall procedure. Some of the most elegant methods of this type involve the enzymatic conversion of a nonlabeling group to one which can be labeled. Thus progesterone from a first chromatogram can be purified by acetyla-

90

IAN

E. BUSH

tion of suspected hydroxylic contaminants, rechromatographed, and the eluate reduced enzymatically to 20P-hydroxypregn-4-en-3-one whose BOP-hydroxyl group can be esterified with 3H-labeled acetic anhydride for isotopic detection after LLC, or chloroacetic anhydride for electroncapture detection in GLC. An additional enhancement of specificity is obtained by the use of the fairly specific enzyme for the reduction. Other aspects of the use of microchemical reactions have been considered extensively elsewhere (B27, B32, E2). For the examination of novel sampIes and the initial identification and detection of steroids in complex mixtures, more elaborate techniques are required. Infrared spectra are now less sensitive and less valuable in practice than the use of the mass spectrometer. The direct coupling of the latter to GLC columns pioneered by Ryhage has provided the most powerful and chemically specific research tool now available for steroid research (E2, R6). Although LKB (Sweden) now supplies a good commercial model of the apparatus, this technique is too expensive and needs too much technical backup to be applicable to routine work outside the larger medical centers. Even so, the full exploitation of this technique outside of basic research will probably require the extensive use of automation and electronic data processing. Although this review of the problem of specificity in methods for measuring steroids is far from complete, I have devoted a considerable space to it because it remains probably the single most important feature to be assessed when considering the value of a particular procedure or the selection of techniques and equipment. Although my assessment of some features will be challenged by some workers, I believe that the general clinical chemist will find it a useful guide to further reading, and that in any event the literature quoted can be relied upon to correct any unintentional errors of bias that I may have made. Although an objective clinical chemist will want to achieve the best specificity he can by the most economical means, he will not take too seriously suggestions that his estimates of blood glucose or urinary creatinine are unacceptable without the support of a mass spectrum. 2.3. ACCURACY, PRECISION,AND SENSITIVITY Pose the head and neck on its side in several positions and record each attitude with a contact outline. (P9,p. 4)

The accuracy of an analytical method is defined as the inverse of the difference between an estimate and the true value that would be obtained by a n ideal method. It is heavily dependent upon the specificity of the method. The precision is defined as a function of the agreement between repeated estimates carried out on fractions of the same sample, and is

DETERMINATION OF STEROIDS I N PLASMA AND URINE

91

the main determinant of the sensitivity of the method. The sensitivity is usually considered as the minimum value which can be distinguished significantly from a zero value. For significance a t the 95% level the sensitivity is thus twice the standard error of the method at that level. I n 1961 Braunsberg and James (B17) gave an exhaustive account of the statistical evaluation of methods for estimating adrenocortical steroids and pointed out that very few published methods up to that time had been evaluated in a manner which would pass muster by the orthodox criteria of general analytical chemistry. I n this, however, the steroid field is not much worse than other areas of analytical biochemistry, and it is only relatively recently that a serious move to improve precision and quality control in clinical chemistry has gained acceptance. An excellent general review is given by Whitby at al. (W3). I n this section I shall cover only certain points which seem to be specifically important in steroid work. The reporting of the statistical characteristics of methods for steroids varies considerably. It is now usual to try to obtain an estimate of the coefficient of variation of a method. This is the standard error of the calibration curve, treated as a linear regression, divided by the value of the estimate of the mean. A problem arises here because of the lack of general agreement about the practical application of linear regression analysis to calibration curves in analytical chemistry. The standard method is based on a model which assumes a symmetrical increase in the error of the estimate about the mean of the independent variable. The standard error of the estimate when the method is applied in practice (in contrast to that of the population of values used to calculate the regression), is thus not indicated by the parallel lines sometimes drawn either side of the calibration curve but by two hyperbolas (Fig. 5). If in fact one measures the standard deviations of sets of estimates a t values of the independent variable corresponding to A , B, and C in Fig. 5, i t is common experience that the standard deviation a t the higher value C is approximately equal to that a t the intermediate value B and that both are much less than the standard deviation a t the lowest value A. This suggests a realistic error envelope of type shown in Fig. 5. Common sense suggests that the reason for this is that the overall errors of analytical procedures are not symmetrical. In many cases it is known experimentally, for instance, that measuring instruments have a larger error a t low values and that certain losses are not due to first-order processes but have a zero-order component-that is, a constant, irreducible fraction which becomes proportionately very large with very small quantities. A very good discussion of these problems is given by Williams (W5,

IAN E. BUSH a

FIQ.5. Classical linear regression plot. R is the regression line for a calibration curve; a and b, the true 1% fiducial limits for future estimates of y. The bars show the probable standard deviations of measurements of y at x = A , B, and C. c, d = the probable form of the actual 1%fiducial limits of a calibration curve in analytical chemistry when working near the limits of sensitivity.

p. 90) but there is as yet no authoritative analysis of what sort of weighted model should be used. A practical approximation adopted by Tait and his colleagues emphasizes this fact by characterizing the precision of a method not only by its coefficient of variation in the middle region of its range of normal application, but also by the standard error a t this point expressed directly in units of mass. This seems to be a very valuable practice a t present since it emphasizes the limitations of the method in an immediately recognizable fashion €or all levels of its application. A major improvement in the precision of the more complex methods has been achieved by the now almost standard practice of using an internal standard of radioactively labeled steroid. Some features of this practice need discussion. It is axiomatic that the mass of added tracer must be small compared with the mass to be estimated in the sample, and that the precision of addition and measurement of the internal standard be as great as possible. Also, the tracer affords a correction for losses only in the stages of the procedure following its addition. Whcre

DETERMINATION OF STEROIDS IN PLASMA AND U R I N l

93

the signal-generating step is achieved by labeling with isotopic reagents, the two isotopes must be different and all of the final sample is counted in a two-channel radioactive counter. If GLC is used, a small part of the sample is usually taken before the GLC step and used for the counting of radioactivity, because the collection of specific peaks in GLC is technically rather cumbersome a t present. There are, however, some extremely useful methods in which an internal radioactive standard is not used, and it is important to try to assess both the improvement in precision that has been achieved, and the main sources of the errors th at have been reduced by this method. The use of labeled internal standards is expensive in materials, equipment, and time, and should not be adopted unless its advantages can be assessed and found to be worthwhile (B27). In recent years the procedure has been improved greatly, largely by the increasing availability of isotopic steroids and reagents of high specific activity. On the other hand, new and disturbing information has raised new doubts about this technique so that an objective appraisal is still rather inconclusive. There are some very decided advantages of this technique, but they are not applicable indiscriminately to all types of method. It seems likely th at the three major advantages of this method are as follows. First, it allows the occasional serious loss of material due to technical error or accident to be corrected with confidence. This is of tremendous logistic importance in a difficult field because it makes the use of duplicate or triplicate estimates unnecwsary, and reduces very greatly the anxiety of technicians, which can be considerable in this arduous field. Second, it eliminates or a t least greatly reduces the mnlinear effects of disproportionate losses with very small quantities. Third it greatly reduces, and in theory eliminates, the more or less vandoom errors introduced by losses during the many steps that are usually involved in such procedures. All these considerations are critically dependent on the absence of any significant isotope effect, and their relative weights are obviously closely related to the type of technique that is involved. Probably the main value of this technique in the modern complex methods is in the third factor. Thus, the elimination of steroid, and even more of nonsteroid, impurities from the final estimate is crucial a t the low levels which are involved so that most procedures using isotopic labeling or GLC involve two or more microchemical reactions (one is a minimum for isotopic reagent methods) and three or more chromatograms with transfer operations before each one. Recovery by elution from TLC or paper chromatograms is often poor (50-700/0 recovery) unless very wide margins are taken around the zone to be eluted (B26), and

94

IAN E. BUSH

it is desirable in all types of chromatogram to take as fine a “cut” as possible to reduce the inclusion of impurities with similar mobilities. This latter factor makes the use of an internal radioactive sample essential with isotopic labeling methods. Thus the overall recovery of progesterone in the double-isotope method of Riondel et al. is only about 3% (Rl). For some years, however, it was known or strongly suspected that a significant isotope effect could often be seen with tritiated, as distinct from I4C or 12C, steroids in their mobilities on Celite LLC partition columns (“2).Sizable and consistent differences in their mobilities on paper LLC chromatogram have now been convincingly demonstrated by Gold and Crigler (G5). As they point out, the omission of any part of a chromatographic zone from elution will produce a significant error in the measurement of specific activity or isotopic ratios. The intentional and, in the absence of an isotope effect, extremely sensible earlier procedure of eluting less than the whole of a chromatographic zone during the steps of a double-isotope procedure must now be regarded as a potentially serious source of errors with tritiated steroids. On the other hand, it is not yet known whether the same effect is found with 12C-steroids labeled with tritium-containing reagents. Thus values obtained with 14C-steroids as internal standards and tritiated acetic anhydride for labeling are probably reliable. Similarly the high losses in the 35S-thiosemicarbazone methods of Riondel et al. (Rl, R2) were attributed largely to losses in the microchemical reactions with progesterone, but poor elution of the thiosemicarbazones from paper and silica gel plates was also regarded as a major factor with testosterone. Since 1,2-3H-labeled internal standards are used in both these methods, the elution of the zones with large safety margins would seem to be desirable. An unfortunate aspect of this problem is that the full power of the double-isotope methods in achieving really high sensitivities is reached only when very small quantities of internal standard, and hence of very high specific activity, can be employed. This is now economical with tritiated steroids, but very expensive with I4C-labeled steroids of comparable specific activities (B18). The same considerations apply to the majority of ultrasensitive GLC methods in which labeled internal standards are used as controls on recovery, because not only do they usually involve a similar series of preliminary subfractionation chromatograms but, if the internal standard is counted in the effluent (as yet a rare technique), f a r more serious isotope effects are to be expected. Indeed many examples are now known of almost complete separation of isotopically labeled substances from the naturally occurring compounds. I n GLC the losses on the column itself and variations in detector sensitivity can be corrected very adequately, but not in a theoretically abso-

DETERMINATION OF STEROIDS IN PLASMA AND URINEI

95

lutely satisfactory way, by using closely related nonisotopic steroids as internal standards. This is undoubtedly of value since it also corrects for the often large transfer errors on injection of the sample which are only partly overcome by solid-sample injection equipment. If methods are used in which the overall recovery is 70% or above, and if the possibility of accidental but undetected very large losses is reduced by using a minimum of steps and transfer operations, or by automation, it is questionable whether in practice the precision will be greatly increased by current techniques of using internal radioactive standards. I n the case of many urinary steroids, conjugated internal standards would be required for full control of the procedure and are not yet generally available. Finally the technique itself is not error-free, even in the absence of isotope effects, and the usual techniques employed so far often involve very small volumes of volatile solvents in which transfer errors can be large unless very superior manual technique is used. It is not fair to end this discussion, in which so many difficulties in the use of radioactive internal standards have been dwelt upon, without admitting that the position taken here is controversial. It should be obvious that the use of internal radioactive controls is absolutely essential for all existing ultrasensitive methods using radioactively labeled reagents, and highly desirable with the majority of existing GLC methods working in the same range. However, neither the real gains nor the often unrecognized limitations of this approach should be obscured by uncritical advocacy. The working precision attained by existing methods for steroids is not always easily discovered. I n this the field is not much worse than many others in biochemical analysis or clinical chemistry. I n part it is due to the fact that in many areas of steroid technique we have only just got out of a period in which almost any method was better than none. Partly owing to the stimulating critical review of Braunsberg and James (B17) and partly to increasing confidence, it is now more usual to find some attempt a t an assessment of the precision of a new method over the working range intended for it. However, many of these estimates of precision are based on rather limited numbers of replicate determinations (usually 6-10) or a series of duplicate estimations during routine work. These estimates of precision are probably considerably overoptimistic as an indication of what is obtained during ordinary routine use, and as yet it is not usual for steroid research laboratories to employ intensive methods of quality control. The results of double-blind trials or quality control are now very, even painfully, familiar to general clinical chemists. It is well recognized that a routine performance of a test of precision on a dozen samples over a day or two tends to be much more carefully

96

IAN E. BUSH

performed than routine work. A significant improvement in results is also seen with blind trials-that is to say, the inclusion of surrogate or duplicate samples in a coded batch in which the identity of the test samples is unknown, but their existence is known or suspected (W3). The hard facts really emerge only when quality control is practiced as a routine with an absolutely foolproof system of coded samples. Precision assessed in the manner conventional in most steroid laboratories should be regarded only as an essential preliminary in the setting-up of a new method. Some estimates of the precision of various typical modern procedures are given in Table 3. General clinical chemists may well be surprised a t some of the values when compared with objective experience in other areas. Thus it is well known that the estimation of serum bilirubin has a coefficient of variation (C.V.) of around &15% in most laboratories, and that with creatinine a value of +8-10% is fairly normal (W3). I n a recent blind (but not double-blind) interlaboratory trial of a standardized method for 17-ketogenic steroids and one for 17-ketosteroids, Gray et al. (G10) obtained C.V.’s varying between &4% and 2 1 4 % for both methods. Six of the ten laboratories cooperating in the trial had “special steroid experience” and were asked to obtain duplicate estimates of “any thirty” routine specimens of urine. These results are for two well-established and relatively simple procedures. In a comparison of orthodox, blind, and double-blind estimates of precision by the duplicate method, Roberts e t al. (R4) examined the precision of a modification of the systematic method for nine common urinary 17-ketosteroids and reducing steroids of Bush and Willoughby (B29, B36, B37). The two samples of the duplicates in the double-blind trial were submitted, unknown to the technical staff, on separate days up to one month apart. This method is semiautomated, involving a conventional type of extraction procedure followed by photometric scanning of paper chromatograms. Recovery values and estimates of precision by open trials were in the conventionally acceptable range. The precision of the standard methods for 17-ketogenic steroids and 17-ketosteroids were similar in the conventional tests to those found in other laboratories and deteriorated to the levels observed by Gray e t al. (G10) in the doubleblind samples. The precision of the estimates of the nine individual steroids in the double-blind trials varied from C.V.’s of 2 1 0 % with the majority to A20747 with the steroids of smaller quantity (0.2-0.4 mg/22 hours). An assessment of the sources of these errors among the many steps involved has not been made, but a relatively rapid and unsophisticated method of triangulation was used to obtain peak areas on the chromatogram records.

TABLE 3 PRECISIONS CLAIMED WITH VARIOUS MORE COMPLEX STEROID METHODS Substance (source)'

Basis of method, final (subfractionation)

Testosterone (U) Testosterone (P) Testosterone (P) Progesterone (P)

GLC GLC GLC GLC

Progesterone (P) Progesterone (P) Pregnanediol (U)

GLC (various) GLC (TLC) GLC (TLC)

Estriol (U) Estriol (U) Estrogens (U) Progesterone (P) Progesterone (P)

GLC GLC GLC (Ah03column) Isotope displacement Isotope displacement (TLC)

Progesterone (P)

Isotopic labeling (double isotope) Isotopic labeling (double isotope) GLC (paper LLC)

Testosterone (P) 17-Ketosteroids (U) 17-Ketosteroids (U) a

(TLC) (paper LLC) (TLC) (paper LLC)

Photometric scanning (paper LLC)

(P) . . = -Dlasma: IU) . = urine. I

.

Method of assessment Replicates (open) Replicates (open) Replicates (open) Replicates (open) (part of procedure tested) Various Replicates (open) Addition of known quantities (open) Replicates (open) Replicates (open) Addition of known amounts Replicates Addition of known amounts to HZO (open) Replicates (open) Replicates (open) Comparison with established method, part of procedure (open) Replicates of urine samples or pools (double-blind)

Precision as coefficient of variance

Reference

+7% +9% f 30% k4.776

L9, L9, L9, L9,

+5% to *25% +5 to +20% -14.5%

E2, p. 180 L9, p. 153 L9, p. 135

+2%

L9, p. 215 L9, p. 229 L9, p. 195

f4.876 +3-5% +8 .5 % to k25% f 1 2 % @ 0.5pg +38%@ 0.2pg +7.5%

p. p. p. p.

19 35 23 143

54 N1 R1

R2

*5.5%

L9, p. 1

+ l o % to

+25%

R4

98

IAN E. BUSH

Although one would like to see the precision of these methods improved, it should be realized that most diagnostic problems in this field involve elevations in steroid levels in plasma or urine of the order of 2- to 10fold and often of several hundredfold. By many criteria these levels of precision vary from not quite, to reasonably, adequate (W3). Despite the lack of quality control the satisfactory precision achieved by most modern methods for individual steroids is indicated by more recent studies in which they have been used for measurements of steroids providing rather sharp discrimination of normals from diseased patients. There is also reasonable concordance between normal values achieved by different workers for many steroid sex hormones and their metabolites. Also, in these studies the better GLC and double-isotope LLC methods are in general agreement (E2,T5). Until the more complex ultrasensitive methods have been subjected to double-blind quality control their effective precision in routine use should be assessed with caution. It is highly unlikely that any steroid method whatever in existence today will have a C.V. of less than *lo% in routine use in more than 30% of laboratories. [In the study of Gray et al. (GlO), the highest C.V. in the 17-ketogenic steroid test, one of the simplest tests considered here, was *14% and was returned by one of the six laboratories with “special steroid experience”!] Claims of C.V.’s of k 5 % of less for the more complex methods should be reviewed with extreme caution. Thus van der Molen points out (E2, p. 171) that triangulation or planimetry of GLC peak areas has an error &3-5% in his experience. Again the widely used method of Moore and Stein for amino acids has a C.V. of &3% for pure synthetic amino acid mixtures. When such relatively large errors are observed by some of the best workers in their fields in the final steps of such methods, using pure synthetic mixtures of substances, it is quixotic to claim more than moderate precision with current methods when analyzing natural mixtures using many additional preliminary steps of intrinsically far lower reliability. This is quite similar to the situation in other areas of clinical chemistry. Sensitivity is closely related to the specificity and precision of a method. As an extremely important aspect of any method for plasma steroids, and of many for urinary steroids, it has been much discussed. I n terms of the best levels of signal-to-noise ratio obtained with pure substances under ideal conditions, the mass spectrometer attached directly to a good GLC column is probably the most sensitive analytical method currently available. In terms of steroids in natural sources the most sensitive methods a t present established are those using the isotopicdisplacement (protein-binding) technique (0.1-0.5 ng) and eIectron-

D&PERMINATlON O F STEROIbS fN PLASMA ANB URINE

99

capture detectors with GLC (0.5-5 ng) . Double-isotope methods using paper LLC and TLC have an effective sensitivity of 0.2-10 ng, but this is highly dependent on the substance to be estimated, the overall yield of the typically complex procedure, and the availability of both substance and reagents with high specific activity. I n the highly sensitive range of methods (signal-to-noise > 20 a t > 20 ng), fluorescence methods abound with a sensitivity at least the equal of the more common GLC methods using flame-ionization detectors. A good survey of many of these earlier methods will be found in the book by Gray and Bacharach (G9) and in Braunsberg and James’ review (B17). Preeminent among them are the methods of Brown and Ittrich (B20,B21, 12, S2), Finkelstein (F2, F3, F4), and Preedy (P10) for estrogens, and of Eik-Nes and his colleagues for androgens, progesterone, and related steroids (02). I n the moderately sensitive range (>SO0 ng) a large number of both fluorometric and colorimetric methods are suitable and quite competitive with GLC using the flame-ionization detector. However, for compounds with few reactive substituents, GLC with the flame-ionization detector is obviously the method of choice (e.g., E4, G14, G15, H4, Pl). The best clue to the sensitivity of a given method in practice is obtained by discovering the volume of plasma or urine that is routinely used in the recommended overall method, and calculating the quantity of steroid being measured in the sample (see also 2.4). This exercise often yields surprising results.

2.4. OPERATIONAL AND LOGICAL DESIGN Profesional taxidermy is apt to develop into a grind, With its aesthetic values growing dulled in the rush of required production. (P9, p. VII)

The preceding sections have dealt with the chemical and physical properties of the analytical methods under consideration and, except where absolutely necessary, have dealt with them without any consideration of the operational design and characteristics of those methods. While it is perfectly proper to neglect operational factors within reason when attempting to achieve an aim in fundamental research it is not possible for the clinical chemist to do so. The application of systems analysis to analytical chemistry has been limited in the past, but has now become of major importance because of the increasing need for tests involving complex procedures and expensive equipment. For the modern clinical chemist analytical chemistry is as much a problem of management of his resources as of chemical technique. A systems analysis is concerned with optimizing the yield and/or costs

100

IAN E. BUSH

of an operation or set of operations and the design of systems to achieve this aim. It depends on a clear definition of aims or “product” (output, yield) and of resources available. While it does not necessarily concern itself with technical improvements per se it will often yield a very useful guide t o areas where the rewards of technical progress will be greatest. At first sight this approach seems to be one of applied common sense plus a little statistics: the results, however, are often surprising. The conclusions of some earlier approaches to this question (B25a, B29, B33a) need relatively little modification today. I n this article I shall focus on examples which may be of use to the clinical chemist in his d a c u l t choice of strategies to be adopted in the steroid field. Obsolescence is now an important factor in these considerations and will be given due weight. First i t should be recognized that each major area of technique in the steroid field is still a difficult one. None of the more complex methods in the highly sensitive or ultrasensitive range can be run on a part-time basis. A large number of items of know-how and manual skill are required and a minimum of six-month’s training will be needed for an above-average technician without previous experience of steroid work if GLC or double-isotope methods are used. Less will be required for most of the isotope-displacement (protein-binding) assays, although I know of one laboratory staffed by outstanding workers in the GLC field who have had difliculty. Some impression of the difficulties that may arise can be gained from reading the discussions in the symposia edited by Lipsett (LS),Paulsen (P3), and Grant (G8).Both GLC and doubleisotope methods are still prone to serious stretches of down-time due to unknown impurities interfering when new batches or solvents or reagents are used, or to equipment failures. One of the most expert teams in the double-isotope field has been out of operation for up to six months a t a time because of problems of this sort. Another in the GLC field had to try out about 40 electron-capture detectors before one of adequate sensitivity could be selected. Similar problems can arise with some fluorescence methods, particularly with different batches of the sulfuric acid used in the less specific methods-here, however, the volumes used are small so that a good batch, once found, can be reserved and provide good results for long periods. All the methods for individual steroids involve chromatographic separation steps, and this is likely to continue for a long time. Because of the advantages in specificity to be obtained by using a sequence of selectively designed purification steps (see Section 2.2)) the use of multiple chromatograms is common and likely to remain so. It is to be noted that this is a major factor in both the lag-time and work-time of

DETERMINATION OF STEBOIDS IN PLASMA AND URINE

101

methods using GLC or double-isotope LLC and TLC. When we talk of GLC in the highly sensitive range and below, we in fact mean GLC preceded by anything from 2 to 5 subfractionation chromatograms on paper (LLC), silica gel plates (TLC), or columns (LLC or alumina). This is because of the almost total lack of specificity of the signalgenerating step with the flame-ionization detector, the low specificity of the labeling reagents in electron-capture GLC and double-isotope methods, and the high level of unidentified radioactive impurities in the latter. One of the main problems of the general clinical chemist today is to guess a t the needs and methods of tomorrow rather than just those of today. Of the range of methods and equipment now available, which ones will be preferable in five year’s time? Where and how can he get into the act without serious risks of obsolescence? For the ultrasensitive range it is clear that almost all examples of the three major approaches-double-isotope, GLC, and isotope-displacement methods-involve preliminary paper LLC or TLC chromatography. Both the latter techniques are relatively cheap to set up and have a low risk of obsolescence. They represent a minimal essential investment. They can be used on their own for many types of steroid in both the high and moderate sensitivity ranges with precisions that are acceptable for most diagnostic tasks and many research projects. For many purposes TLC is faster overall than paper LLC. For very large rates of sample acceptance, the advantages in speed of TLC become much smaller or vanish altogether, and its extra cost becomes a very large factor. The next factor to be considered, which is relatively unlikely to alter much for some time, is the awkward conflict between the gain in specificity achieved by sequential chromatographic steps, and the great cost in time, and danger of losses, a t each transfer step (elution and redeposition of chromatographic zones). Although marginal improvements can be expected from technical advances, the situation is unlikely to change much. It leads one to suggest that methods incorporating greater specificity into the signal-generating step of a method, and thus reducing the need for sequential purification steps, are to be preferred where possible. I n terms of available techniques this suggests a strong preference for isotope-displacement or the more specific fluorescence methods. Thus it is notable that all the current ultrasensitive isotope-displacement techniques use one, or a t most two, T L C or paper LLC chromatograms before the signal-generating step. A special rapid method for progesterone uses none (54). I n the moderately sensitive range, a number of good fluorescence methods use no chromatograms or merely one or two. I n terms of future developments it is highly likely that a t least moderate improvements in specificity will be achieved in GLC by the development of more

102

IAN E. BUSH

selective detectors or labeling reactions, and that a major improvement will be achieved by the use of combined GLC-mass spectrometry using on-line computer analysis with or without nonradioactive isotope labeling (R6). It is, however, equally likely that comparable improvements involving far lower cost will be made in the chemistry and instrumentation for fluorescence methods. Another factor to consider is the actual and potential scope of these different methods, i.e., the number of different types of substance amenable to analysis. Within the steroid field, GLC with or without mass spectrometry is supreme in this respect precisely because of its lack of specificity, i.e., its lack of dependence on the presence of reactive substituents for the signal-generating step. For the analysis of hydrocarbons, fatty acids, and other moderately to highly volatile substances, GLC is also far and away the best method. On the other hand, it is not a s yet to be recommended for most other classes of substances of biological and medical interest. LLC, TLC, and thin-media electrophoresis, using fluorescence or colorimetric methods, vary from poor to useless for volatile or unreactive substances, but overall have a far wider range of applications than GLC. From this point of view, however, the isotopicdisplacement methods are weak because new binding agents have to be isolated, purified, and tested for each new substance or small group of substances to be measured. Unless the preparation of the necessary wide range of selective binding agents is taken up by commercial suppliers on a large scale, this method is likely to remain of relatively narrow scope. Double-isotope methods are similarly limited, but for quite different reasons-namely, that the final purification stages are rarely adequate for handling more than two compounds a t a time (e.g., Gla , H6). The logistics of the several broad types of methods for steroids have been discussed previously (B29, B33a). For the large-scale analysis of many components in complex mixtures the fast direct photometry of thin-media chromatograms (paper, coated or loaded paper, fiberglass sheets, TLC, plastic-supported TLC, etc.) is still far and away the most productive method although a t present not applicable to the ultrasensitive range of problems. For the estimation of single steroids in the ultrasensitive range, the isotopic-displacement methods are supreme. Thus testosterone, progesterone, and one or two related steroids can now be estimated in plasma a t the rate of 20 or more per technician per day (54, K3, M3, M7, N3, N4, N6, S8). At the other end of the scale are the excessively difficult double-isotope methods for estrogens (Bl) with which one technician with skilled supervision can accomplish 3-5 estimations per week. For ultrasensitive, and some highly sensitive methods, for steroids not yet measured by isotopic displacement methods, reasonable produc-

DETERMINATION OF STEROIDS IN PLASMA ANb URINE

103

tivity can be achieved with GLC. Such methods are decidedly more productive than comparable double-isotopic methods where these have been developed. For most endocrine studies the lag time for small batches of samples is not a serious factor in choosing a procedure. Actual lag time, however, can be a serious problem if the load of sample acceptance for a laborious method becomes too large for available staff and equipment. Thus with the majority of installations, no more than 10-20 GLC chromatograms can be run per working day, and a t this rate it will be hard to analyze mixtures of more than 4-6 steroids per run. With most methods the necessary prior subfractionations usually reduce the number of measurable compounds per run to 3 or less. I n one or two cases lag time is crucial. The most interesting example involves the need for rapid measurement of total urinary estrogens in cases of infertility being treated with gonadotropins (G4). I n order to avoid multiple pregnancy, treatment must be stopped within 6-12 hours of the attainment of a critical level of urinary estrogen. No attempt is made to obtain 24-hour samples, and the estimation is carried out by a modification of the rapid method of Brown. One technician can carry out the method and calculate the results for 12 samples in 3-3.5 hours. Similar though less stringent requirements must be met in the estimation of plasma or urinary estriol in attempts to forestall threatened abortion. At this point it is usual to suggest that no one method will suffice to cover all the tasks to be carried out and that “a variety of techniques must be used.” While true in the literal sense, this needs further definition because of the moderately high costs of the equipment involved and the very high costa of time and personnel, both directly in carrying out the complex manual procedures, and indirectly in the considerable amount of technical backup that is needed for maintenance of equipment and the preparation of pure solvents and reagents. The dangers of obsolescence are considerable in the areas both of equipment and of investment in the training of staff. The latter especially should not be underestimated. The strategy to be adopted depends on the size and scope of the effort that is contemplated. Before attempting to make specific suggestions (see Section 2.6) we must return to the question of the diagnostic tasks required of a modern laboratory for steroid chemistry. 2.5. DIAGNOSTIC RELEVANCE Using this outfit, I turned out a lot of creditable work, though many a modern lad might feel that such a layout would cramp his style. (P9,p. 1)

From the point of view of clinical work in this area, the methods chosen should be those that meet most economically and reliably those requests

104

IAN E. BUSH

of clinicians in which a chemical measurement, or set of them, provides the most crucial diagnostic information. Although not considered in detail in this article, the estimation of glucocorticoids is essential in a large number of cases. Because of the wide variety of methods that are adequate for most diagnostic tasks of this type, attention will be focused on those areas in which only a few methods can be used. The differentiation of normal from abnormal concentrations of pIasma testosterone in women requires the use of either electron-capture GLC, isotopic-displacement methods, double-isotope methods, or fluorometric methods. Evidence to date suggests that the isotopic-displacement method is preferable (K3, M3). Existing fluorometric methods are adequate in analytical characteristics, but much more laborious (F2). This differentiation is crucial in cases of simple, provisionally idiopathic, hirsutism (cf. however, C l ) . I n many cases, however, an elevated excretion of urinary testosterone glucuronide would be equally satisfactory despite the uncertainty of its metabolic precursors (B2, 11).This measurement is also desirable in cases of provisionally idiopathic amenorrhea. For the investigation of primary amenorrhea, measurement of urinary estrogens may be required. The number of measurements needed excludes most methods despite their adequate sensitivity and precision. Brown’s Kober or Ittrich fluorescence methods (B20, B21, 52) are the best established and probably preferable, although others are probably adequate but slower. I n the absence of signs of the adrenogenital syndrome, plasma progesterone or urinary pregnanediol measurements may be desirable in the investigation of some types of amenorrhea, of anovular cycles, and irregular menstruation. The most satisfactory method for plasma progesterone as a routine in these circumstances is the isotope-displacement method, although a variety of GLC methods, and fluorescence methods with TLC or LLC would be adequate (54,M7,N3, N4). With a good method for plasma progesterone, urinary pregnanediol measurements would be less necessary. Under these circumstances, however, the best established modern methods for pregnanediol are those using GLC (E2, L8,L l l ) , although TLC and LLC paper methods using fluorescence are also adequate for many purposes (e.g., S9). Other problems of differential diagnosis in which chemical measurements are of crucial value all involve moderately sensitive, or a t most, highly sensitive methods, with the exception of the acute obstetric probl e m mentioned previously. I n many of these problems a battery of simple and well-established colorimetric and fluorometric methods not involving chromatography could reasonably be used. Thus cortisol could be measured by a variant of the Silber and Busch method (see B17) or an isotope-displacement method (M7,N6) ; 17-ketosteroids and 17-keto-

DETERMINATION OF STEROIDS IN PLASMA AND URINE

105

genic steroids by the standard methods (see G10, P6) ; pregnanetriol by benzene-water partition followed by the method for 17-ketogenic steroids using periodate (W7) ; and DHA-sulfate by the Pettenkoffer reaction (B16). A greater certainty of chemical identification could be achieved if desired by a wide variety of simpler chromatographic methods used as checks on the above. It is reasonable, however, to demand that a modern clinical chemistry laboratory be in a position to carry out more than the bare minimum of essential, crucial diagnostic tests in a field in which much clinical research remains to be done. The introduction of more sophisticated methods is not only desirable in yielding results possibly leading to new discoveries, but also because it provides extra information which improves the reliability both of the chemical findings and the differential diagnosis. It also enables the confirmation of clinical-chemical correlations obtained by other workers. Thus one cannot agree with the suggestion of Whitby e t al. (W3) that the more complex steroid methods are completely outside the scope of routine use. A more balanced position would be to suggest that the research of the last ten years has created a definite need for the introduction of more sophisticated steroid methods into routine use, but that the methods to be adopted should not necessarily reflect the preferences of existing workers in basic research if the cost and risk of obsolescence are excessive. 2.6.

RETURNON

INVEsTMENT

At the start before buying costly additions to his tool kit, the beginner should try to produce good work with a simple home outfit. (P9,p. VIII)

The main concern of this article is to try to assess the current state of the art in steroid analytical chemstry for the clinical chemist who has not previously embarked on the more complex modern methods. I hope it will be understood that in trying to do this for the present situation, and for the immediate future where obsolescence could be very costly, it is often necessary to adopt criteria which differ from those that are appropriate for the specialized research laboratories already established in the field. I n casting doubt upon the value to clinical chemistry of several general methods which are popular in research a t the present time, there is no intention to disparage the tremendous amount of earnest effort and sometimes great elegance which has gone into the development of these methods. I n large part it is precisely because of the excellence of much of the research done with them that it is now possible to suggest other lines which seem most likely to be valuable to the clinical chemist in the near future.

106

IAN E. BUSH

The first and most important investment is in the training and supervision of people. While no exact job specifications can be provided, a descriptive assessment of this can be given. First, it seems unlikely that one can go beyond the simpler methods without having a t least one professionally qualified leader in the laboratory whose main, although not exclusive, interest is in the chemical and technical problems of steroid work. Laboratories without this minimum requisite have sometimes managed to set up one of the more complex modern methods under the direction of specialist-trained superior technical staff. However, the range of methods covered by these personnel was very small and downtime was large because of their lack of know-how for coping with unanticipated problems. The scientific leader of the steroid group will be hard to find. It should be realized that most of the leaders in the development and current use of these methods are either experienced researchoriented M.D.’s a t the consultant level, full time research Ph.D.’s or M.D.’s usually a t the full professor level, or bright young M.D.’s or Ph.D.’s who have spent 1-3 years in postdoctoral training with a leading group of workers. At this point it seems that two possible policies can be distinguished which could be the basis of further discussion. One is to accept the extreme view of some specialists that the steroid group can be a success only if someone of the caliber of the present leaders of the field is recruited t o lead it. The other is to go for a good man with general experience who is willing to undergo any special training that may be needed. Larger laboratories would be well advised to consider the first policy seriously. It would, however, entail the establishment of a recognized steroid division with a minimum of two additional workers with a Ph.D. or M.D., and 3 to 6 positions for technicians, graduate students, or visiting workers. A minimum of 2000 sq. ft. of space and a generous capital budget would be required. Considerable autonomy and scope for research activity would have to be granted. Several centers in the United States and the United Kingdom now have laboratories which have almost grown t o this state of affairs, having started as small research groups ten or more years ago. All of them are attached to departments of medicine, surgery, pediatrics, or obstetrics and gynecology and are not integral parts of the clinical chemistry department. These groups maintain a healthy mixture of research, development, and service functions. Nearly all the ones I know specialize in certain areas and are too small to cover the full range of desirable modern methods in parallel a t the same time. I know of only one center in which large-scale use of modern steroid methods is under way in the clinical chemistry department itself. The department is part of a teaching hospital of about 1000 beds and is

DETERMINATION OF STEROIDS IN PLASMA AND URINE

107

moderately large and well-equipped for general clinical chemistry. There are three senior leaders with excellent research ability, and approximately one-half of the entire technical staff and general facilities are devoted full-time to steroid work. The main techniques used are GLC and mass spectrometry. Considerable support from outside research grants is obtained. Even if a separate well-financed steroid division is established, it may be difficult t o find the right man to head it. Such men are relatively rare, and if they are good they are likely to be well established and more interested in their own research. I n view of this, it is worth considering the alternate policy. It will probably be preferable in most cases to seek an interested clinical chemist or analytical biochemist of high quality rather than a man of lesser ability with specific experience of steroid work. Experience with some form of biological work involving organic chemistry is very desirable-e.g., food technology, pharmaceuticals, natural products analysis, pulp and paper technology, marine biochemistry. A minimum of six months’ experience of postdoctoral research training in a good group dealing with a reasonable range of steroid techniques will be necessary. For a really good man this will be adequate and a stretch of one or more years is probably to be avoided: if the man is any good he will either be lost t o his teacher’s research group or be in danger of getting compulsively hooked to one or the other of the more complex techniques. After this period the training of technical staff must be undertaken. For a t least one or two particular techniques outside the specific training of the senior worker, it will usually be desirable to send one aboveaverage technician to a center specializing in the method to be adopted. Costs of this policy can be roughly compared with the costs of the alternatives. I n my experience a good technician learns about as much in this field by working alongside one “skilled in the art?’ for 2-3 weeks, as will be learned from 4-6 full-day consultation visits by an expert, or from 6-18 months trial and error without any outside help. I n several cases I have known of errors which had either completely held up or seriously hindered the work of a laboratory in this field for as much as three years, which were realized by an intelligent technician within the first 2 hours of a week’s stay in an expert’s laboratory. The same information might have been missed, however, if the setting had merely been a conventional 1-day visit. Six-months’ frustration on some technical difficulty before deciding to seek outside help is a very common experience of newcomers to this field. Training of this sort must be obtained by work o n the job, and care is needed to ensure that the technician really has carried out every step of a given procedure and has also had as much

108

IAN E. BUSH

contact as possible with problems of technical backup. The technician should return with a list giving the supplier and catalog number of every piece of equipment used, and every reagent used. We can now try to assess the material requirements of such a group. It is not profitable to attempt a completely quantitative analysis, but simple counting and ranking methods can be used to assess the relative merits of different sorts of approaches. The main factors to be taken into account are: current diagnostic value of current methodology; current research value of current methodology ; scope for expanding current applications and integration with future technical developments ; cost of equipment investment; cost of down-time and servicing of typical equipment; sample acceptance rates manageable; cost of rapid obsolescence (less than 5-7 years), and the personnel investment mentioned above. Most of these factors are interactive to a sizable degree. Paper LLC and adsorption TLC are basic parts of all major complex methods in current use and likely to remain so. Capital cost is small, risk of obsolescence is small, and training investment is minimal. One technician should specialize in it, and all others should be familiar with it. Simple columns mainly for adsorption chromatography are necessary for the Brown method and are potentially or actually useful for many methods. The simpler techniques are easily mastered, but for some, considerable know-how is needed, most of which can be gleaned from Brown’s papers and symposia discussions (E2, G8,L8, P3). The choice of signal-generating methods to concentrate upon-a small group must concentrate to some extent-is best approached by tabulating the important procedures not possible for technical reasons with any given method. This is done in Table 4,where attention is mainly focused on well-established or thoroughly worked out methods, but reasonably feasible alternatives for the near future are given some weight. When productivity and training investment are taken into account, the isotopedisplacement methods win hands down if measurements in the ultrasensitive ranges of testosterone and progesterone in plasma are desired. On the other hand, the crucial diagnostic value of plasma testosterone estimates (ultrasensitive) is confined to the relatively small group of women with provisionally idiopathic hirsutism in which urinary testosterone glucuronide (highly sensitive range) is probably as good a discriminant (N5a, cf. V4). It may be objected that I have been a little harsh on fluorometric methods in that Short and Levit (53) and Heap (Hl) have devised methods capable of detecting 5 5 0 ng of progesterone; and Brown (B21) and Preedy and Aitken (PlO), methods capable of detecting even smaller quantities of estrogens. The application of these methods to samples of human plasma, however, is too limited as yet to

DETERMINATION OF STEROIDS IN PLASMA AND URINE

109

TABLE 4 SURVEY OF LIMITATIONS OF MAJORSIQNAL-GENERATING TECHNIQUES Technique

Currently NOT fully developed for :

Estimate of probable extent of feasible advances in near future

Plasma and urinary estrogens, Estrogens-good ; aldosterone-fair; minor aldosterone, and major and increase in scope-good; minor metabolites major increase in scopepoor Poor Flame ionization GLC All ultrasensitive methods Electron-capture GLC Plasma estrogens, aldosterone, Fair most metabolites Fair Most major and minor Isotopic-labeling; LLC and TLC metabolites Good Photometric (flucAll ultrasensitive methods rescence, absorption) except estrogens Isotope-displacement methods

warrant recognition as ‘‘fully developed.” Both electron-capture-GLC or isotopic-displacement methods are significantly more sensitive and have been far more extensively used. The isotopic labeling methods may be subject to some technical improvement in the future, but with heavy running costs and low productivity, they seem a t present to carry a large risk of obsolescence. Because of the current importance of measuring estrogens in relatively large numbers of samples, often with very short lag times, a good fluorometer seems to be an essential instrument and, for a minimal investment, might be the best single instrument to buy. For a more than minimal investment, this would still be the first instrument to buy, but a good scintillation counter would be the next. The point is that for current methods in the highly sensitive and ultrasensitive ranges, most methods involve overall recoveries well below SO%, so that an internal radioactive standard is probably essential. This equipment is thus not only the core of the preferred isotopic-displacement method, and of the less favored double-isotope method, but a desirable or necessary underpinning for all the main techniques in Table 4. A further advantage is that this instrument has enormous potential scope outside the steroid field and a low risk of major obsolescence. As the only heavy capital cost in the isotopic-labeling method, the group would be well placed to capitalize on any major technical advances in isotopic-labeling techniques. With these two instruments, the group could encompass the whole range of steroid measurements of crucial diagnostic significance, and

110

IAN E. BUSH

also with the isotopic-displacement method, be capable of all of the three most difEcult ultrasensitive measurements of plasma steroids, namely estrogens, progesterone, and testosterone. The only serious gap in their repertoire would be in the measurement and identification of specific steroid metabolites in large numbers per sample a t moderate to high levels of sensitivity. Although the development of GLC methods for obtaining analyses of complex patterns of urinary steroids and of other materials (Dl, E2, H4, R6) is a fascinating endeavor and may well be achieved using GLC-mass spectrometry and electronic data processing, this method is expensive and far from fully developed. This gap could be filled a t present either by TLC or paper LLC, or by current GLC methods with flame-ionization detection. There is considerable doubt a t present whether current flame-ionization-GLC techniques are, in the best hands, more reliable and precise than much older methods using thin-media chromatography. Indeed one can reasonably suggest that the use of qualitative TLC or paper LLC is so much more economical of time and effort that it should be adopted by our hypothetical group for the time being. As suggested earlier, this approach is capable of answering most of the additional questions that will need answers in routine diagnosis or clinical research work. No extra capital investment is required, and operating costs are small. It also holds open the possibility of exploiting future advances in methods using rapid photometric scanning and other photometric methods using these media for separation. As pointed out elsewhere, these methods seem to be readily transferable to previously inexperienced workers so that investment in training is small (B33a). Nevertheless, it is surprising, as Engel and Finkelstein have pointed out, that the considerable sensitivity of fluorescence reactions has received so little attention in recent years (LS, pp. 263 and 264). At present, this suggested nucleus of methods and equipment seems to be the optimum ‘(mix” for a small steroid division within a routine department of clinical chemistry. A further step into GLC or doubleisotope methods should be taken only if an addition to the group of a t least one professional and two technicians is contemplated. This holds independently of capital cost. A reasonable range of GLC methods needs the full-time service of a t least this number of staff. Even then, this step should be taken with caution and after a careful assessment of the real return. Thus if we consider the, by and large, excellent electron-capture-GLC method of van der Molen and Groen (V2), the two extra technicians could achieve 40 determinations of plasma testosterone and progesterone per week. (No time allowed for technical backup work.) The same two technicians using Johansson’s rapid modification (54)

DETERMINATION OF STEROIDS IN PLASMA AND URINE

111

of Murphy’s original isotopic displacement methods (M7) could achieve40 determinations of progesterone per day. Both methods require a good scintillation counter for measurements of radioactivity, and Murphy’s needs small TLC equipment. The GLC method would undoubtedly need a considerable amount of professional supervision and a GLC apparatus costing approximately $5000. The isotope-displacement assay would need no extra supervision and no extra major equipment. The cost of materials would be similar and so would the down-time of the scintillation counter. The GLC method would suffer from the additional down-time of the GLC apparatus. The latter on occasion can be so large that with the time needed for trial and development of new methods, testing of new reference standards and reagents, calibration, and so on, the effective time available for steady routine use can be down to 50%. A routine lab embarking on a heavy load of routine use would probably need two or three GLC machines to avoid the risk of disruptions of its schedule of sample acceptance. With current methods, 8 to 12 runs can be made per day per machine if enough staff exist t o do the preliminary extraction and paper or T L C subfractionation chromatograms. I n summary, I have tried to assess some of the major problems facing a general clinical chemist in deciding which routes to take in bringing some of the modern methods for steroids into the routine laboratory. I n doing so, I have considered two possible approaches and have risked formulating a definite policy for one of them. Emphasis has been laid on the operational characteristics of these methods because this aspect of the problem is so frequently overlooked or relegated to small print. 3.

Advances in General Techniques

3.1.

TECHNIQUES

OF

EXTRACTION

Do not leave skins lying endlessly in borax solution. (P9,p. 89)

The first step of nearly all methods for steroids involves extraction of lipid-soluble materials from plasma or urine and evaporation of the organic solvent. I n the case of urine the water-soluble conjugates are usually hydrolyzed or solvolyzed prior to this step. There are a thousand different ways of carrying this out, but their empirical details are of little interest. A few general principles and the most useful specific examples of novel techniques will be discussed here. The classical method of liquid-liquid extraction was devised in the days when pure solvents and large pieces of laboratory glassware were very expensive. Great use therefore was made of the binomial principle showing that for a given volume of extractant, and a given partition coefficient of the substance to be extracted, the maximum yield was ob-

112

IAN E. BUSH

tained by subdividing the extractant as much as was feasible and extracting the sample repetitively with small volumes of extractant. Thus, before the 1950’s extractions of steroids were commonly carried out by recipes calling for 3 to 6 extractions of the aqueous phase with onefourth to one-half its volume of extractant. I n the early part of 1950, G. F. Marrian pointed out a t a meeting that emulsification during the extraction of urine with organic solvents was greatly reduced in frequency by using two or more volumes of solvent per volume of urine. Since then most methods have usually involved 2-3 extractions with 1-4 volumes of organic solvent. It has long been recognized that the liquid-liquid extraction of steroids and the concentration by evaporation of the organic solvent extract represent a tremendous fraction of the working time and a large fraction of the total or lag time of the average method for measuring steroids. It is a pity therefore, that the majority of extraction procedures in the literature appear to have been devised without more than an intuitive or empirical appraisal of the optimum solvent and procedure to be used. The general approach seems to be to adopt a convenient, volatile solvent which is readily available and use a safe excess, in volume or number of extractions, to minimize the effects of accidental carelessness. Dichloromethane (i.e., methylene chloride), diethyl ether (ether) , chloroform, and less often ethyl acetate are the most frequently used. Two main factors are overlooked in this intuitive approach. First, with a favorable partition coefficient, 2-5 volumes of extractant secure such a good recovery with one extraction that the use of repetitive slightly smaller volumes gains only a marginal increase in recove? a t very considerable expense of time and effort. As Peterson (P6, P6a) first pointed out, one extraction of an aqueous solution of cortisol with 5 volumes of dichloromethane should, and in fact does, secure a recovery of 98% of the cortisol in the aqueous sample. Because of a preference for an upper layer extractant, my own work has usually employed one 1:2 by volume, for extraction with 5-6 volumes of ethyl acetat-ther, this class of steroids. I n their double-isotope method for plasma testosterone and 17-ketosteroids1 Gandy and Peterson (Gla) used one extraction with 9 volumes of dichloromethaneethyl acetate, 1 :1 by volume. Calculation will quickly show that with a partition coefficient of 10 in favor of the extractant, two extractions with three volumes will secure a recovery of 99.8% as compared with 98.4% using one extraction with 6 volumes. While it can be shown that with a random error of performance of the extractions the double extraction procedure will reduce the overall error in the recovery value, this is trivial compared with the extra labor in the extraction of many samples. If an internal radioactive

DETERMINATION O F STEROIDS IN PLASMA AND URINE

113

standard is used, the achievement of such a marginal improvement in recovery is completely valueless. While the partition coefficients between water and extractant cannot be used directly for plasma, the difference is usually small and can be measured. A second feature, worthy of more attention, is the more selective use of solvents to extract only what is desired and a minimum of impurities (e.g., B25, 54, M7, P6, P6a, S4, S 5 ) . While such selective extractions, or one-step partitions, will never be as effective as the subsequent chromatographic steps of purification, they may well, in the early stages of a procedure, reduce very considerably the total mass of the extract and of widely different steroids from the one to be measured, thus permitting smaller and more convenient chromatographic systems to be used, and better resolutions to be obtained, in the first chromatographic step. It is probable that this principle should be applied cautiously with plasma when strongly binding proteins are known to be present (note, however, 54, M7). It is, however, of great significance in urine, where, for instance, ether is commonly used for a wide variety of steroids. Thus methods for testosterone, 17-ketosteroids1 and probably pregnanediol should probably benefit greatly by using benzene, or hexane-benzene approximately 1:1 as extractant. This would be of particular benefit in GLC methods. Similarly in our laboratory we have always extracted urine with ether or ether-ethyl acetate before hydrolysis, solvolysis, or conjugate extraction, when attempting to measure small amounts of specific, conjugated steroids. A surprising amount of freely extractable material can be removed from many urine samples in this way and thus eliminated from the extract of the urine made after hydrolysis. The use of resins or adsorbents for extracting steroids from urine deserves more attention. Our earlier method using the weak anion exchange resin “Decolorite” (B35) was discarded in 1960 when the suppliers changed the process of manufacture. Recently Bradlow (B16a) has achieved good recoveries of steroid conjugates from urine with the nonionic lipophilic resin Amberlite XAD-2 (Rohm and Haas). These methods, using simple percolation through columns of the resin, are great savers of time and labor and deserve more attention. A useful review is given by Pasqualini (PZ). One of the most interesting recent developments has been the application of “liquid ion-exchangers” to the extraction of steroid conjugates from urine (M2) and bile (H2). The chromatographic use of these reagents by Mattox (M2) is one of the most useful applications of the R N theory in recent years, and may enable some revolutionary modifications of the operational characteristics of methods for steroids in urine.

114

IAN E. BUSH

The increased speed and productivity of methods using these reagents will undoubtedly be of the greatest significance in many applications of steroid methods in the routine laboratory. The operational characteristics of extraction procedures can often be greatly improved by close attention to the vessels and volumes used, choice of upper or lower layer extractants, and the means by which phase transfers are achieved. Most methods for plasma, and many for urine, nowadays employ stirring or shaking in standard-taper glassstoppered test tubes in preference to the orthodox separating funnels of old. Horizontal shakers for large numbers of samples have been used by Townsend and James (T10) in a method using one extraction with dichloromethane. This method, however, is subject to leakage around the stoppers unless heavy pressures and a very good set of matched tubes and stoppers are used. This problem is overcome by the use of special separating funnels which have been described elsewhere (B30). A similar device has been used by Brown (B21a) in which the extract of estrogens for his fast method is transferred directly to an attached fluorimeter tube. Batches of 12 or more extracts can be shaken in parallel. The introduction of simple modifications of conventional apparatus of this sort can often secure surprisingly large reductions in working time, lag time, or both, and are well worth considering whenever a new method is set up.

3.2. TECHNIQUES OF SEPARATION The long-winded papier-mlchi: formulas of bygone times are largely obsolete, along with arsenical mothproofing and hay stuffing. (P9,p. 88)

Apart from the use of liquid ion-exchange reagents mentioned above, there have been few striking advances in this area which are likely to be of much application in steroid work in the immediate future. The paper of Dalgliesh et al. (Dl) gives a fascinating glimpse of the possible developments in computerized GLC-mass spectrometry, but involve a large and expensive system needing further development before its potentialities can be fully realized. Work on the exploration of new types of stationary phases for GLC seems to have reached a plateau. An interesting review of recent work with Golay capillary GLC is given by Desty (D3). It should be emphasized that the use of GLC of any kind is critically dependent for its precision as a quantitative measurement upon the technique of introducing the sample (see, e.g., S2a, and others in G8). It seems to be generally agreed that the gauze type, or other types, of solid sample injection techniques are essential for the best results (E2, G8, K10).

DETERMINATION OF STEROIDS IN PLASMA AND URINE

115

The importance of transfer operations in multiple chromatographic steps has been emphasized above. They are very costly in time directly, and also indirectly because of the large losses that are usually entailed, particularly in isotopic-labeling methods (see, e.g., B1, R1, R2). I n this connection one would emphasize the value of simple elution and “running-up” methods for the transfer of fractions or extracts to and from chromatograms (B27, B29, B33a, G7). Another time-saving procedure of considerable value in our experience is the technique of pretreatment of paper chromatograms with ethermethanol-water mixtures (B34, T8). This is effectively a hydration step, not an “impregnation”with stationary phase. Equilibration of LLC paper chromatograms, using volatile stationary phases, which normally needs a minimum of 3 hours a t 25”C, can be cut to a few minutes. By using small tanks and strips, many simple separations can be achieved in times comparable to standard T L C methods for steroids. The time saved by this technique in more complex methods using standard length chromatograms can also be of great value (e.g., B18). The general opinion is still that one of the major advantages of T L C over paper LLC is the low values for “blanks” of various sorts when zones are eluted from preliminary chromatograms. While the empirical findings cannot be challanged, the implication that this is a general and invariable phenomenon can. There are many examples in which eluates from paper LLC chromatograms show satisfactory blank values, and one suspects that too little attention is given to selection of solvents, methods of selecting and washing filter paper, and details of technique (e.g., H l a , L3, L7). As pointed out before, the simpler and shorter the method of washing paper, the better; and the use of masks and a no-touch quasi-surgical technique for handling the paper once it has been washed is essential (B11, B27, B29, L3). Prolonged washing procedures and solvent extractions may introduce more “dirt” than they remove. The problem of impurities in solvents and reagents, especially with GLC and fluorescence methods, receives frequent mention. It is probably a sizable part of the problems usually attributed to “paper blank,” although this fact may not emerge from incomplete tests. An example from my own laboratory may indicate just how serious, indeed completely ruinous, this can be with an otherwise sound method. I n developing a flame-ionization-GLC method for plasma progesterone a t moderate to highly sensitive levels, J. McCracken found that one preliminary TLC or LLC paper chromatogram was sometimes quite adequate to obtain a clean peak on GLC. More often the peak was useless due to massive amounts of material tailing back from the front zone. A t first it seemed that this was material from the paper. Careful tests showed that the material was in fact contained in the methanol used for eluting the

116

IAN E. BUSH

progesterone zone from the paper chromatograms. Ten different grades of methanol had to be tested before a satisfactory one was found. When found, it turned out to be the second-best grade offered by its manufacturer-the extra cost of the “best” grade was matched only by the extra dirt i t contained. Microchemical modification reactions are now an essential and frequent part of practically all separation procedures. A variety of derivatives are essential or useful for many steroid separations on GLC, and a full survey will be found in available monographs (E2, G8, H4, K10) ; one of the most elegant advances in this area stems from the use of enzymatic reactions for the reduction of progesterone to its 20p-dihydro derivative (see V I ) . This not only provides a characteristic change in chromatographic properties and a hydroxyl group available for convenient labeling reactions, but also introduces the considerable chemical specificity of an enzymatic reaction into the factors determining the overall specificity of the method. Improved resolution by using very long columns has been obtained, using adsorption chromatography, by Vestergaard, who has exploited U-shaped columns and capillary Teflon tubing (V8, V9) for compactness. A large number of adsorption systems for TLC plates have been worked out by Lisboa (L13-L16).

3.3. TECHNIQUES OF DETECTION AND MEASUREMENT Clean all meat from the leg bones but do not disarticulate the joints. (P9,p. 75)

One of the most striking advances of recent years has been the introduction of the electron-capture detector for GLC by Lovelock (L17). This is extraordinarily sensitive to halogen-containing molecules and has enabled the introduction of several promising labeling methods for steroids. The most successful and well established is the use of chloroacetic anhydride to form chloroacetate esters of several hydroxylic steroids (B22, V1) . Polyfluoro labels in the form of heptafluorobutyrate groups give promise of great sensitivity, but completely satisfactory methods have yet to be established (cf. V4). This technique is subject to a large number of highly technical details which may cause changes in sensitivity of 104-fold or more. The introduction of newer and more stable isotopes as a source of /I-particles is likely to secure greater stability and, possibly, sensitivity in the near future. Meanwhile, the most important technical improvement for its use with steroids has been the introduction of alternating current or pulsed voltages to the detector (L8, p. 266). It is well recognized that the development of more selective detectors

DETERMINATION OF STEROIDS IN PLASMA AND URINE

117

for GLC could produce great improvements in its value for the analysis of high-boiling biochemicals of all kinds. Although a phosphate-selective detector has recently been introduced, there has been relatively little advance in this area despite the considerable effort of the last ten years. A large number of promising techniques exist, but their transformation to stable reliable devices has yet to be achieved. Nevertheless the electron-capture detector is a major advance in that a wide variety of halogen-labeling methods should be feasible in addition to those already in use. I n this connection, the recent introduction of chloro- and bromotrimethylsilyl ethers promises to be very useful (E2, T6). Direct monitoring of radioactivity in GLC effluents is still under development (e.g., N3). Sensitivity is limited by the necessarily short period of counting which is possible. Photometric methods have also advanced. Several new fluorometric instruments have appeared, many of them with attachments for direct scanning of paper chromatograms. Although none of the commercially available chromatogram scanners are fully satisfactory for the reasons given earlier (B29) , the considerable sensitivity possible with differential scanners has been clearly demonstrated by the elegant differential absorptiometric scanner of Salganicoff et al. (S2a). I n my own laboratory we have concentrated on a differential fluorometric instrument. Sensitivities seem to be extremely high and interference from contaminants in plasma very small if specific rather than nonspecific fluorescence reactions are used. This instrument is already quite satisfactory for the highly sensitive range of measurement of all A4-3-ketosteroids1 having a signal-to-noise ratio of 20 with quantities of 3 20 ng (B33a, cf. P11). Wide use of photometric scanning for high-precision measurements of steroids will depend on the introduction of a commercially available instrument with satisfactory specifications. This is probable in the next six months. Meanwhile it should be noted that many fluorometric methods, whet,her by direct photometric scanning or by more conventional techniques, are strictly competitive with all other existing methods in the highly sensitive range. Their future successful extension to the ultrasensitive range is not unlikely. Even with conventional scanners on the market a t present, some surprisingly effective techniques have been devised (e.g., B25a, C4, E3, M6). Although these methods are still slightly less precise as signal-generating steps than GLC or scintillation counters, and more dependent on careful manual technique than is possible with a more sophisticated instrument, their precision is good enough for a very large majority of diagnostic purposes. The general opinion is still doubtful of the value of direct photometric scanning of paper or T L C chromatograms as a general

118

IAN E. BUSH

technique. This opinion, however, is mainly based on the use of deficient instruments and techniques. Until a good commercial instrument becomes available, the general clinical chemist should be advised t o adopt a “waitand-see” attitude.

3.4. TECHNIQUES OF AUTOMATION AND DATA ANALYSIS Glass eyes, once strictly a European product, are now made even in America. (P9, p. VIII)

This field is advancing rapidly and only a limited part of its scope can be ascertained from generally available publications (e.g., S6a gives an excellent review). One major line of work is in the development of large hospital computer systems in which diagnostic aids, monitoring of therapy, and record-keeping and retrieval are the main functions. The other is in the on-line or real-time processing of data obtained from chemical analytical apparatus or from physiological recording equipment in intensive-care units. Several systems have been developed commercially for laboratories of general clinical chemistry, but the overall automation of more complex analytical methods is still under development. Probably the single advance with the greatest significance for steroid work in the last five years has been the development of automatic sample-changers, printers, tape punches, and data-processors for scintillation counters. Simple computer programs for handling the paper tape output of this equipment are numerous and easily written. Many are available from the manufacturers of isotope-counting equipment. Several systems for analyzing chromatogram records have been devised in recent years (see B a a , K7). They are all still in the stages of further development, and it is not possible a t present to recommend any particular one with confidence for exclusive use. The IBM 1800 gas chromatograph system (I.B.M., Inc.) and the DEC Gaschrom-8 system (Digital Equipment Corp.) are designed to monitor 2@40 GLC columns simultaneously. Fairly complete data-processing is achieved. At the moment there seem to be no published tests of the analytical and operating characteristics of these systems. Inspection of program listings and supplier’s brochures suggests that the numerical methods used may have some serious weaknesses. Until such systems have been validated or converted to satisfactorily validated systems by expert users they cannot be recommended as worthwhile investments. However, computer analysis of mass spectra and complex GLC records is progressing rapidly (e.g., PACE and QUAD systems; Electronic Associates, Inc.) . WeIlestablished general-purpose systems are not yet available for the general user, and considerable development is still needed in this field.

DETERMINATION OF STEROIDS IN PLAsMA AND URINE

119

Automatic peak integrators of various sorts are in fairly wide use for monitoring the effluents of IEC, LLC, and GLC columns (DISC integrator, Perkin-Elmer, Infotronics) While these can give excellent results with very clean and drift-free backgrounds and have some arrangements for eliminating noise and drift, their performance is well below acceptable standards of precision outside the fields of ion-exchange columns and unusually good GLC separations. The best-developed computer programs that have been published are nearly all for medium-size to very large computers. It is probably helpful here t o note that the use of the larger and better-known machines for most purposes of analytical chemistry is rapidly and progressively becoming less attractive because of the tremendous development of smaller machines in the last two years. Thus, quite sophisticated and complete systems for analytical laboratories have recently been exhibited using computers with 4096- or 8192-word core memories in configurations costing only $25,000-90,000. This is largely because of more skillful programming and because these computers were developed from lines specifically designed to handle large amounts of data from scientific measuring equipment and industrial machines. The cost of interfacing some of the better-known large computers to a small set of analytical instruments is sometimes more than the entire small computer that would suffice. The cost of expanding the interface to accommodate an additional measuring instrument may be more than 10-fold greater than with the small machine. The frequent advantages of the large machine in systems programs is shrinking rapidly with the increase in user-experience of the smaller machines. Finally, the expansion of small computer systems to encompass more complex tasks is cheap and relatively simple. The larger configurations are now in fact recognized to be some of the best in existence for complex time-sharing systems. Further details can be obtained from the books and symposia quoted, but a lot of the “lore” of this field is still unpublished. The policy to be adopted for automating steroid analysis or any other type of complex biochemical analysis needs careful development and the gathering of a lot of data from current users. Automation of the procedures for analyzing steroids is difficult because of the present lack of a really efficient liquid-liquid extractor. Townsend and James (T10) have published an ingenious method for adapting a n Autoanalyzer or similar system to the estimation of cortisol in plasma by their earlier fluorescence method. The semiautomated method is capable of making 20 estimations of plasma cortisol per hour using 2 ml samples of plasma. Automation of adsorption chromatography on columns has been achieved by Vestergaard using a system capable of handling 12 columns

.

120

IAN E. BUSH

in parallel. Although used specifically for the main 17-ketosteroids and 17-ketogenic steroids a t moderately sensitive levels it could be applied to many other procedures (V5, V6, V7, V10). As discussed above, the precision of automated methods becomes evident only when tested by double-blind quality control (R4,W3). Although Vestergaard’s method appears to be good in this respect, it is interesting and perhaps cautionary that Townsend and James claimed a precision no better than fl.O pg/lOO ml for their automated method for plasma cortisol, i.e., coefficient of variation of about +-8% of the mean. This was much better (one-third to one-half) when sequential rather than random duplicates were assessed (T10). Zak and Epstein ( Z l ) , have automated the Zimmerman reaction for 17-ketosteroids by the development of a novel aqueous reagent using Hyamine-KOH as base. This paper well repays a close study. A useful device to have for paper LLC is an automatic solvent dispenser. Thus runs taking between 8 and 16 hours can be started by dispensing the mobile phase into the troughs during the night. The run is then completed a t the start of work next day. A recent example is described by Nerenberg e t al. (N5). 4.

4.1.

Specific Examples

INTRODUCTION

While this writer does not favour the indiscriminate slaughter of turtles and tortoises for taxidermy use, occasionally a fine specimen does turn up as a temptation. (P9, p. 83)

Since the main aim of this article is to give a review which will, hopefully, be a useful guide to the nonspecialist presently trying t o assess the future introduction of advanced methods of steroid analysis into the routine laboratory, no attempt will be made to provide a comprehensive and detailed summary of steroid methodology in the last five years. Some excellent books, monographs, and symposia have been published in the last two or three years, so that a specialist’s survey of the field would be redundant. Instead I shall pick one or two examples of methods which seem best to illustrate and support the general ideas expressed in earlier sections. Enough references will be given to enable the clinical chemist to pursue these matters for himself in more detail if he desires; by and large, references will be confined to the main and most useful papers in each area, so that he will not be smothered. The complete coverage of this field as a specialist review would require the citation of about 15013 references.

DETERMINATION OF STEROIDS IN PLASMA AND URINE

121

4.2. ESTROGENS AND RELATED STEROIDS The best method for urinary estrogens of clinical importance is that devised by Brown (B20, B21). This depends on the modification of the Kober-Ittrich color reaction to provide an extremely sensitive fluorescence reaction (see also S2). Acid hydrolysis (15 minutes) is followed by either extraction (2 1 volume) ; evaporation to dryness; a partition between petroleum ether-benzene and water (estriol) and 1.6% NaOH (estradiol, estrone). The two solutions are methylated (10 minutes) and extracted with ether; the ether is evaporated, and the residues are run on alumina columns. I n the original procedure the less specific and sensitive form of the Kober reaction required more extensive purification; one technician could complete four estimations in 1.5-2 days. The new procedure enables four determinations to be made every $6 hours. An ultrarapid version of this method for specialized obstetric use has been devised by Gemzell and his group and will be published shortly. By this method, twelve samples can be measured in 3 hours by one technician. The limit of sensitivity by the latest version (B21) is 0.05-0.1 ng of individual estrogen per sample, equivalent to excretion rates of about 6 ng/24 hours. While this method has not been applied to plasma and requires a good deal of know-how and manual skill, it is at present, in one of its forms, a “must” for any laboratory venturing into the more difficult areas of steroid methodology (see P3). Quite apart from the fact that current electron-capture-GLC or isotopic-labeling methods achieve little or no greater sensitivity in practice, they are considerably more laborious, the latter requiring up to 1 week for the analysis of 3-5 samples. For the measurement of the large amounts of estriol in pregnancy urine, many methods are suitable (Hla, J1, 01, 52). The Kober-Ittrich fluorescence reaction can be used, as Ittrich originally showed (I2), directly on small volumes of urine without any chromatographic procedure. For those wishing to use GLC methods, the well-designed methods of Touchstone are well validated and reliable. Useful summaries of these methods are given in the symposium edited by Paulsen (P3, appendix). One of the most useful innovations in this field is the gel-filtration procedure of Beling (A2, B6). This provides an excellent although rather slow “clean-up” step as a preliminary to many subsequent techniques. A recent paper by Brown e t al. (B21) supplants some of the material in these books and provides a totally new procedure. An excellent but extremely laborious isotopic labeling method has been described by Baird e t al. (Bl) and is a good example of the complexities of such methods in this area.

x

122

IAN E. BUSH

One can reasonably hope that rapid ultrasensitive methods based on the isotope-displacement method will soon be available for the principal estrogens. A promising method has been published by Korenman (K11). The reliability of this assay and the ease of preparation of the uterine “macromolecule” used in this assay remain to be established. The excellent review by Adlercreutz and Luukkainen (A4) gives many details of the above methods and of those devised for the measurement of the numerous recently discovered hydroxylated metabolites of the classical estrogens. Attempts to measure the latter substances routinely are not advisable for nonspecialists a t present. AND RELATED STEROIDS 4.3. PR~GESTERONE

A good review of previous methods for this important group of steroids is given by Short (S3), and an outstandingly good account of GLC methods by van der Molen (Vl; see also G8). A good summary of van der Molen’s method is given in Lipsett (L8, pp. 153-168) based on his original paper with Groen (V2). This is an example of an electron-capture-GLC method with good analytical characteristics, which illustrates several general principles and problems of current technique in the steroid field. is added to the A very small tracer quantity of proge~terone-7-~H plasma sample (10 ml) which is then made slightly alkaline with NaOH. The latter step stabilizes lipoproteins and prevents the extraction of excessive amounts of triglyceride, phospholipid, and cholesterol (S3). The plasma is extracted with ether (6 X 15 ml) and the combined extracts are washed with water (2 x 5 ml). After evaporation to dryness the extract is transferred to a TLC plate, which is run with benzeneethyl acetate, 2:l (ca. 60 minutes). The progesterone zone is eluted, and the eluate is evaporated. Buffer solution (1 ml), NADPH, and 20phydroxysteroid dehydrogenase are added, and the mixture is incubated for 120 minutes a t 37OC to form the reduction product 20P-hydroxypregn-4-en-3-one. This solution is now extracted (4 X 1 ml) and the combined extracts are evaporated and carefully dried in a desiccator (2-3 hours). The dry residue is now esterified with chloroacetic anhydride and pyridine (overnight), and the chloroacetate of 20p-hydroxypregn-4-en%one is extracted. This product is now run on a second TLC plate (ca. 60 minutes), and the appropriate zone is located by a marker steroid, scraped off, and eluted. The eluate is now dissolved in 1 ml of methanol containing 10-40 ng of testosterone chloracetate as an internal standard to correct for losses in transfer to the GLC column. One-tenth of this solution is pipetted into a counting vial, and the tritium is counted to provide a correction for recovery in all the preceding steps. The remain-

DETERMINATION OF STEROIDS IN PLASMA A N D URINE

123

ing nine-tenths of the final fraction is then evaporated, dissolved in 1 M O p1 of benzene, and injected into a %foot XE-60column with an electroncapture detector using a 15O-psec pulse. The chloroacetate is eluted in about 25 minutes and has a relative retention time of 29.7 ( x cholestane) . The testosterone-chloroacetate serves as internal standard and has a relative retention time of 21.4. A valuable feature of this method is that, by simple variations of the technique, 20p- and 20a-hydroxypregn-4-en%one themselves and testosterone can be measured in the same plasma sample (see also B22). This method is 5-10 times as sensitive overall as the best comparable flame-ionization method. The use of heptafluorobutyrate in place of chloroacetate as an electron-capturing labeling radical achieves a further 5-fold increase in sensitivity but has some problems associated with it (Vl, V4). The useful sensitivity of this method is 2-5 ng with a precision (C.V.) of &25% in this range. In the range 1&100 ng, the C.V. is -+10-15%. Despite its elegance and relative simplicity compared with many others, this method is very laborious, involving 12 liquid-liquid partitions in the form of extractions or washes; the addition of two internal standards ; three transfers to chromatograms; two microchemical reactions, and numerous evaporations of organic solvents. However, 20 plasma samples can be analyzed per week for both progesterone and testosterone. Among its suboptimal features are: the use of ether instead of a more selective extractant; an unnecessarily large number of partitions for extractions; and a difficult addition of the second internal standards using very small volumes of volatile organic solvents. However, this method is well established and tested. I n comparison it should be noted that Heap’s fluorometric method ( H l a ) is very nearly as sensitive (ca. 5 ng) and precise, and approximately equally laborious. By comparison, Murphy’s isotope-displacement method (54, M7, N2N4) uses less than one-tenth the volume of plasma, an excellent selection of solvent allowing 1-step extractions, has about 20 times the sensitivity, a similar precision, uses one TLC chromatographic step, and allows about 20 samples to be measured per day by one technician. This method has been modified by Johansson (54) so that 0.25 ml of plasma or less can be used and the TLC chromatogram eliminated. An application of the earlier method by Knobil’s group has shown its adequacy for studying the complete menstrual cycle (N3, N4). Strott and Lipsett (SS) have modified the method so as to enable the simultaneous measurement of 17a-hydroxyprogesterone simultaneously with progesterone and used it for a complete and penetrating study of these two plasma steroids throughout the menstrual cycle in women (L8). I n evaluating the various methods for plasma progesterone in the ultra-

124

IAN E. BUSH

sensitive (follicular phase) and highly sensitive (luteal phase) ranges for their possible use in the routine laboratory, it is impossible to mistake the superiority of these isotope-displacement or competitive binding methods. The proof of the pudding is in the eating. Within a few years of the first publication of Murphy’s methods (M7), half a dozen laboratories have got underway and several have published first-class studies using these methods. Promising or well-developed methods for other classes of steroids using other binding agents have also been published in this period (see 4.1, 4.3). Johansson’s modification of Murphy’s method is a n impressive example of both the gains (and the possible dangers) of careful attention to the selection of analytically specific and operationally rapid extraction techniques (54). I n a careful discussion of specificity, he points out that 14% of 17a-hydroxyprogesterone will be extracted by his solvent. This probably means that his progesterone estimates may be increased 20% or more over their true values a t and just before the onset of ovulation (L9). Since, however, the test is of clinical interest mainly in determining adequate function of the corpus luteum this is not a serious drawback. This degree of specificity may not satisfy many steroid specialists, but can reasonably be defended, in that specificity is not in practice an absolute property of a method but a probabilistic one, which in this case has been carefully and quantitatively assessed. I n fact it is interesting to note here that in nonvirilized women the specificity of this method is in question for only 1-2 days in the menstrual cycle. The resulting speed of the method is potentially of the greatest value in routine gynecological or acute obstetric problems. The present dependence of both Murphy and Johannson’s methods on particular lots of petroleum ether is, however, a weakness and unnecessary. Thus, the property most probably determining the suitability of particular lots of petroleum ether for this selective extraction is either the proportion of aromatic hydrocarbons, and/or the proportion of longer chain or cyclic aliphatic hydrocarbons. It would probably be easy and very valuable to compare suitable and unsuitable batches of solvent by GLC (one of its most valuable and appropriate spheres of application!) and use the results to formulate a synthetic mixture of pentane, cyclohexane, and toluene or benzene for the extractant in this method. This would achieve a standardization enabling other laboratories to be far more confident of success with this method. For some ten years the most useful method for measuring urinary pregnanediol was that of Klopper et al. (K8). The principal isomer excreted in human urine is the 5 8 ( H )-3 ~ ~ ,20 ~ - d io Much l. smaller quantities of the 5a(H) epimer and of the 5m(H)-3/3-OH bis-epimer have been

DETERMINATION OF STEROIDS IN PLASMA AND URINE

125

identified in pregnancy urine. A complete separation of all the eight possible isomers of this compound is a laborious task. Even using GLC, acetylation or other microchemical reactions are needed to obtain a complete characterization. Nor does this usually take the form of a final complete resolution of these isomers, but rather the complete separation of some members of the set on one column, and of others on another. Probably the most exhaustive method is that of Gardiner and Horning (G2; see also Vl). There is a t present no evidence that the resolution of even the three commonest isomers in urine is in fact of any diagnostic significance. However, the measurement of the combined mixture of isomers has been of great clinical value. A t the moment there are probably around fifty papers describing GLC methods for pregnanediol in human urine (Vl, G8). The main advantage of GLC is that with its high sensitivity, very small samples of urine can be used. Van der Molen has shown in a careful comparison of one method of GLC (using pregnanediol diacetate) with the H,SO -ch romogen reaction applied to Klopper’s method of isolation, that the two methods are entirely comparable in precision and accuracy for excretion rates over 1-2 mg per 24 hours. For lower rates, he suggests with good reason that the GLC values may be more accurate because of their higher specificity (Vl) . Since the Klopper method is undoubtedly inadequate for the lower ranges of pregnanediol excretion rates, and needs a fair degree of skill and know-how for it to be reliable, it is fair to suggest that new methods are needed. On the other hand, none of the GLC methods is obviously superior, either operationally or analytically, and some are grossly inferior, to earlier absorptiometric or fluorometric methods. Taking into account the relatively low diagnostic value of this measurement now that good methods for plasma progesterone are available, there seems to be little reason to try to set up GLC methods for this substance in a routine laboratory. If a GLC apparatus is already available, the method of Lipsett and Kirschner would be a good one to choose because it includes the valuable measurement of pregnanetriol (L11; see also 55). However, a small steroid group should probably concentrate on plasma progesterone measurements or simpler methods for pregnanediol (e.g., S9) until improved methods for urinary pregnanediol are developed. I n the case of pregnanetriol it is almost essential to establish a reliable method because of the importance of this steroid in the differential diagnosis of virilism, hirsutism, and the adrenogenital syndrome. Here again, however, one would suspect that plasma 17~hydroxyprogesteronemeasurements will probably become an equally good discriminant for this differential diagnosis. Since this can be measured by the isotope-displace-

126

IAN E. BUSH

ment method (SS),its diagnostic value should be investigated as soon as possible. Meanwhile, the elevation of urinary excretion of pregnanetriol and ll-oxygenated pregnanetrioh seen in most cases of the adrenogenital syndrome, and its suppression with glucocorticoids, is so clear-cut that simple “qualitative” methods are in part sufficient (see B15, D2, F3). This statement too may cause raised eyebrows; but an objective assessment of errors and costs can support it. Thus, De Courcy’s trichloroacetic acid reagent (D2) is capable of convenient application to the detection of pregnanetriol in the nonketonic fraction from one-sixteenth of a 24hour collection from men (100-150 ml) on simple LLC paper chromatograms (a TLC method could no doubt also be devised for this). As she pointed out, the same quantity can be obtained from one-thousandth of a 24-hour collection (1.5-2 mI) of urine from some patients with adrenal hyperplasia. Quantitative estimation by the same reaction after elution from the paper or TLC chromatogram could also be used either by one of the many variations of the micro-17-ketogenic steroid methods or by the same fluorescence reaction (e.g., W7). Further confirmation of identity could be obtained by using a small fraction of the latter after the periodate oxidation for a paper LLC, or TLC, chromatogram and application of the Zimmerman reaction to demonstrate the formation of etiocholanolone (B32). Even quantitative assessment of pregnanetriol by visual comparison with standards would be adequate using carefully coded samples, and preferably duplicates. It is agreed from objective evaluations in a wide variety of fields that this method has a C.V. of +15 to *20% (B27). In the presence of 5- to 100-fold elevations of this steroid in the adrenogenital syndrome, the visual method would meet the usual requirements for a clinically valid discriminant (W3). These arguments could be simplified with justice to the suggestion that a standard qualitative test for pregnanetriol in urine be used in which De Courcy’s method is applied to 10 ml of urine and one-fifth of the nonketonic fraction is used for a preliminary paper LLC chromatogram. The remaining four-fifths of the fraction would be used, if desired, for confirmation of identity or more precise quantitative measurement in the event of a positive finding in the first chromatogram. Such a method is incomparably faster and simpler than any others, and there is little or no objective evidence that it is significantly less precise for clinical purposes than other published methods.

TESTOSTERONE AND RELATED STEROIDS The most complete account of an isotope-displacement method for the ultrasensitive range of testosterone measurements in plasma is that of 4.4.

DETERMINATION OF STEROIDS IN PLASMA AND URINE

127

Mayes and Nugent (M3). This method involves adding testosteronel,2-3H as internal standard and displaced steroid to a 2-ml sample of alkalinized plasma (females) or 0.2 ml (males) diluted with water to 2 ml; extraction with dichloromethane (1 x 35 ml) and two washes with water (4 ml). The residue from evaporation of this extract is run on a paper LLC chromatogram (14 hours), eluted and run through a microalumina column in one step, the eluate run on a TLC plate (ca. 40 minutes), and the zone eluted via an alumina-silica gel column. The eluate is then mixed with 0.15 ml of dilute testosterone-binding protein (prepared from human plasma, and stood a t room temperature for 2 hours. A fraction (one-third) is then removed for counting the internal standard (i.e., total test~sterone-l,Z-~H recovered), and the remainder is treated with 0.1 ml of saturated ammonium sulfate. After 5 minutes the tubes are centrifuged for 10 minutes and 0.1 ml of supernatant is taken for counting the unbound testosterone. The precision determined on 14 samples from a pool of female plasma was k 2 . 6 ng/100 ml, giving a C.V. of ~ 5 . 3 %In . men the C.V. was t 3 . 8 % . Overall recovery of testosterone was 64.2%. Accuracy was determined by adding known amounts of testosterone to a plasma sample containing no testosterone taken from a woman with proved adrenal and ovarian insufficiency. This method is moderately laborious, and the paper is a useful one to read because of its careful description of numerous technical difficulties which have to be avoided. The alternative method of Kato and Horton e t al. (K3) is much simpler and faster, but supposedly not quite so sensitive. Full publication of similar methods by Murphy and Fritz and Knobil (see M3, M7) should be awaited before venturing on the rather complex method of Mayes and Nugent for routine use. Meanwhile, the simpler method of Kato et al. (K3), which is adequate for male plasma and for determining many cases in which values in female plasma are elevated, is probably worth setting up for males and for urinary estimations. The best-established electron-capture-GLC method for plasma testosterone is that of Brownie et al. (B22) and is very similar in principle to that already described for progesterone (V2). Another is described by Kirschner and Coffman (K6). The latter paper uses the enzymatic conversion of androstenedione to testosterone and the use of heptafluorobutyrate radicals in place of the chloroacetate group used originally by Brownie et al. (B22; see also V4). These methods cannot be recommended for routine use a t present in preference to the isotope-displacement methods. The latter are rather more complex than those for progesterone because the chromatographic separation procedure has to be more extensive for testosterone. Thus,

128

IAN E. BUSH

unlike progesterone, testosterone falls in a class containing a lot of other common steroids which compete for the binding protein and have similar chromatographic properties. Their complete elimination from the final step is crucial to this method. Isotopic-labeling methods cannot be recommended for routine use either. However, the papers of Tait’s group (including one on aldosterone) and that of Gandy and Peterson are well worth studying, since they give by far the best accounts available of such methods, and convey a very complete picture of the principles and great laboriousness which they entail (Bl, Gla, R1, R2). The measurement of 17-ketosteroids is an important test for the investigation of practically all disorders in this field even though the proportion of 11-deoxy-17-ketosteroidsderived from testosterone is less then originally supposed. GLC methods cannot be recommended for routine use although in many ways the measurements of the six major 17-ketosteroids by flame-ionization techniques is a good deal easier than that of many other steroids in blood and urine. A very good account of this field is given by Knights and by Thomas and Walton (K10, T6). The detection of the trimethylsilyl ethers at highly sensitive levels with flame-ionization-GLC enables very small volumes of urine to be used (10-50 PI), and the “noise” from “impurities” is relatively small. This is a case in which the speed and sensitivity of GLC can be used to obtain great operational advantages by its intelligent application to an analytically easy task; and also one in which there are strong reasons for using the simplest possible preliminary extraction and chromatographic steps. Very few published methods are in fact optimal from this latter point of view. The fact that an optimum procedure sufficiently trouble-free for routine use by nonspecialists has not yet been found is easily ascertained by a reading of the better specialist discussions in symposia (GS, LS). The main criteria of diagnostic importance in this area are the distinction of extremely large from only moderately increased excretion rates of 17-ketosteroids ; the demonstration of their amenability to alteration by hormonal suppression or stimulation tests ; and the specific demonstration of excessively high excretion rates of DHA. A combination of older colorimetric procedures, the Zimmerman reaction (P6) and the Pettenkoffer reaction (B16), are quite adequate for this purpose in most cases. As suggested earlier, the specific demonstration of DHA in human urine is easily achieved by a wide variety of simple T L C and LLC paper chromatograms. Alterations of genuine value in routine differential diagnosis fall in the class that can be recognized by qualitative tests using that greatly underestimated instrument the human eye. Observer bias must, as always, be eliminated by coding the samples.

DETERMINATION OF STEROIDS IN PLASMA AND URINE

129

If greater precision is required, the zones from paper LLC or TLC, chromatograms can be eluted and measured by the micro-Zimmerman reaction (e.g., W7). There is no objective evidence that the precisions to be obtained are any worse than those by the more sophisticated modern methods for comparable degrees of care and attention to detail. 5.

The Future

It is possible that five to ten years from now some large centers, massively equipped with mass spectrometers, a large computer, and many GLC machines, will exist in which small drops of blood or urine are introduced by nurses or doctors directly, or in capillary tubes, into ports leading from the ward to the clinical chemistry laboratory. The latter will be staffed almost entirely with systems analysts, computer programmers, and physicists. Smaller centers may be equipped with several chemical analytical systems based on small desk-top computers costing $3000 or so. Until that day of their eclipse clinical chemists should take heart from the fact that venereologists and E.N.T. men are still around twenty years after the prediction of their disappearance, and that the manifold complexities of chemistry are unlikely to be mastered quite so easily or completely by automated physical methods. If they can make due allowances for the usual enthusiasms of expert protagonists of new technologies they should be able a t the present time to bring a large number of more difficult classes of steroid measurement into their routine repertoire. The difficulties of doing so should not be underestimated, but, equally important, unnecessary and expensive difficulties should not be embraced just for the sake of getting into the act. There is indeed some ripe fruit ready to be plucked, but it must be carefully selected. When surveying this complex field it is hard to see the forest for the trees. It is salutary to note that, in another field of expertise some time after the end of a t least a century’s use of arsenic, it was possible for our guiding taxidermist to write as recently as 1945: “Borax-solution immersion followed by dry borax dusting and fluffing of skins brings to the art of taxidermy its greatest boon-moth-proofing that will not kill the operator.” Until the steroid research worker has discovered the “borax” for this field, the clinical chemist must remain cautious and patient. ACKNOWLEDGMENTS

The author is grateful to many workers for reprints and private communications over the last ten yews, but especially t o Dr. J. Sjavall, Dr. R. Ryhage, Dr. G . H. Thomas, Dr. R. B. Bulbrook, D. B. Thomas, Professor Carl Gemrell, Dr. E. Johansson, Dr. H. van der Molen, Dr. J. F. Tait, Dr. C. Lloyd, Dr. I. Weliky, for information in advance of or supplementing publication.

130

IAN E. BUSH

REFERENCES” Al. Adlercreutz, H., A comparison between conjugated estrogens in late pregnancy blood and bile. Acta Med. Scand. S u p p l . 412, 133-139 (1964). A2. Adlercreutz, H., Studies on the hydrolysis of gel-filtered urinary oestrogens. Acta Endocrinol. 67, 49-68 (1968). A3. Adlercreutz, H., and Luukkainen, T., Biochemical and clinical aspects of the enterohepatic circulation of oestrogens. Acta Endocrinol. Suppl . 124, 101-140 (1967). A4. Adlercreutz, H., and Luukkainen, T., Gas phase chromatographic methods for estrogens. Biological fluids. In “Gas Phase Chromatography of Steroids” (K. B. Eik-Nes and E. C. Homing, eds.) pp. 72-149. Springer, Berlin, 1968. *B1. Baird, D. T., A method for the measurement of estrone and estradiol-17p in peripheral human blood and other biological fluids using SS pipsylchloride. J . Clin. Endocrinol. 28, 244-258 (1968). *B2. Bardin, C. W., Hembree, W. C., and Lipsett, M. B., Suppression of testosterone and androstenedione production rates with dexamethasone in women with idiopathic hirsutism and polycystic ovaries. J. Clin. Endocrinol. 28, 1300-1306 (1963). B3. Baulieu, E. E., Studies of conjugated 17-ketosteroidsin a case of adrenal tumor. J . Clin. Endm’nol. 22, 501-510 (1962). B4. Baulieu, E. E., Lasnitzky, I., and Robel, P., Metabolism of testosterone and action of metabolites on prostate glands grown in organ culture. Nutwe 219, 1155-1156 (1968). B5. Baulieu, E. E., and Michaud, G., Mdtabolisme des ester-sulfates d’androst4rone et d’btiocholanolone. Bull. Soe. Chim. Biol. 43, 885-895 (1961). B6. Beling, C. G., Gel filtration of conjugated urinary oestrogens and its application in clinical assays. Acta Endocrinol. S u p p l . 79 (1963). B7. Berliner, D. L., and Dougherty, T. F., Hepatic and extrahepatic regulation of corticosteroids. Pharmacol. Rev. 13, 329-359 (1961). B8. Berman, M., Compartmental analysis in kinetics. In “Computers in Biomedical Research” (R. W. Stacy and B. Waxman, eds.), Vol. 2, pp. 173-201. Academic Press, New York, 1965. B9. Bernstein, S., Dusza, J. P., and Joseph, E. P., “Physical Properties of Steroid Conjugates.” Springer-Verlag, New York, 1968 B10. Birke, G., Diczfalusy, E., Plantin, LO.,Robbe, H., and Westman, A., Familial congenital hyperplasia of the adrenal cortex. Acta Endocrinol. 29, 55-69 (1958). B11. Bondy, P. K., and Upton, G. V., Simultaneous determination of cortisol and corticosterone in human plasma by quantitative paper chromatography. Proc. SOC.Exptl. Biol. Med. 94, 585-589 (1957). B12. Bondy, P. K., Cohn, G. L., Hermann, W., and Crispell, K. R., The possible relationship of etiocholanoloneto periodic fever. Yale J . Biol. Med. 30,495 (1958). B13. Bongiovanni, A. M., and Eberlein, W. R., Critical analysis of methods for measurements of pregnane-3a,17a,20a-diol in human urine. Anal. Chem. 30, 388-393 (1958). B14. Bongiovanni, A. M., and Eberlein, W. R., The renal clearance of neutral 17ketosteroids in man. J. Clin. Endocrinol. Metab. 17, 238-249 (1957). B15. Bongiovanni, A. M., and Clayton, G. W., Jr., Simplified method for estimation of

* Particularly valuable reviews, and symposia, or papers which are exceptionally valuable examples of particular types of techniques are asterisked.

DETERMINATION OF STEROIDS IN PLASMA AND URINE

B16. B16a. *B17. *B18. B19. *B20. *B21. B21a. *B22.

B23.

B24. B25, B25a. B26. B27. B28. *B29. B30. B31. B32. B33.

131

ll-oxygenated neutral 17-ketosteroids in urine of individuals with adrenocortical hyperplasia. Proc. SOC.Exptl. Biol. filed. 86, 428-429 (1954). Borrell, S., A simplified procedure for determination of urinary dehydroepiandrosterone. J . Clin. Endocrinol. 27, 1321-1327 (1961). Bradlow, H. L., Extraction of steroid conjugates with a neutral resin. Steroids 11, 265-272 (1968). Braunsberg, H., and James, V. H. T., The determination of cortisol and corticosterone in blood: A review. J . Clin Endocrinol. Metab. 12, 1146-1188 (1961). Brodie, A. M., Shimizu, N., Tait, S. A. S., and Tait, J. F., A method for the measurement of aldosterone in peripheral plasma using 3H-acetic anhydride. J . Clin. Endocrinol. 27, 997-1011 (1967). Brooks, C. J. W., and Watson, J., Quantitative thin-layer chromatography of trimethylsilyl ethers of hydroxylic steroids. J . Chromatog. 396-404 (1967). Brown, J. B., A chemical method for the determination of oestriol, oestrone and oestradiol in human urine Biochem. J . 60, 185-193 (1955). Brown, J. B., MacNaughtan, C., Smith, M. A., and Smyth, B., Further observations on the Kober colour and Ittrich fluorescence reactions in the measurement of oestriol, oestrone, and oestradiol. J . Endocrinol. 40, 175-188 (1968). Brown, J. B., Macleod, S. C., MacNaughtan, C., Smith, M. A., and Smyth, B., A rapid method for measuring oestrogens in human urine using a semiautomatic extractor. J. Endocrinol. 42, 5-15 (1968). Brownie, A. C., van der Molen, H. J., Nishizawa, E. E., and Eik-Nes, K. B., Determination of testosterone in human peripheral blood using gas-liquid chromatography with electron-capture detection. J . Clin. Endocrinol. 24, 10911102 (1964). Bruchovsky, N., and Wilson, J. D., The conversion of testosterone to 5a-androstan-17a-ol-3-one by rat prostate i n vivo and in uitro. J . Biol. C h m . 243, 2012-2021 (1968). Burstein, S., Determination of initial rates of cortisol 2a-and Ga-hydroxylation by hepatic microsomal preparations in guinea pigs: Effect of phenobarbital in two genetic types. Endocrinology 82, 547-554 (1968). Burstein, S., Extraction of Cz106and CZ,O, steroids from aqueous media. Science 124, 1030 (1956). Burstein, S., and Kimball, H. L., Quantitative paper chromatography of Cz1051and Cp106corticosteroids from Guinea pig urine extracts. Anal. Biochem. 4, 132-142 (1962). Bush, I. E., Quantitative estimation of steroids by direct scanning of paper chromatograms. Memo. SOC.Endocrinol. 8, 24-39 (1960). Bush, I. E., “The Chromatography of Steroids.” Pergamon Press, Oxford, 1961. Bush, I. E., Chemical and biological factors in the activity of adrenocortical steroids. Phamnacol. Rev. 14, 317-445 (1962). Bush, I. E., Advances in direct scanning of paper chromatograms for quantitative estimations. Methods Biochem. Anal. 11, 149-209 (1963). Bush, I. E., New design of separating funnel for multiple liquid-liquid extraction. Res. Develop. 19, 44-45 (1963). Bush, I. E., The metabolism and mode of action of a steroid anesthetic. Proc. 6th Pan Am. Congr. Endorrinol. Excerpta Med. Intern. Congr. Ser. No. 99, Abstr. 198, Excerpta Med. Found., Amsterdam, 1965. Bush, I. E., Applications of the R,w treatment in chromatographic analysis Methods Biochem. Anal. 13, 357-438 (1965). Bush, I. E., Automation of steroids analysis. Science 164, 77-83 (1966).

132

IAN E. BUSH

B33a. Bush, I. E., Automation of the analysis of urinary steroids using quantitative paper chromatography and a small laboratory digital computer. J. Clin. Chem. 14, 49 1-5 12 (1968). *B34. Bush, I. E., and Crowshaw, K., A technique for rapid paper chromatography. J. Chromatog. 19, 114-129 (1965). B35. Bush, I. E., and Gale, M., Extraction and fractionation of steroid conjugates. Biochem. J . 67, 29 (1957). B36. Bush, I. E., and Mahesh, V. B., Adrenocortical function with sudden onset of hirsutism. J. Endocrinol. 18, 1-25 (1959). B37. Bush, I. E., and Wdloughby, M., The excretion of allotetrahydrocortisol in human urine. Biochemical J. 67, 689-700 (1957). C1. Casey, J. M., and Nabarro, J. D. N., Plasma testosterone in idiopathic hirsutism and the changes produced by adrenal and ovarian stimulation and suppression. J. Clin. Endocrinol. 27, 1431-1435 (1967). C2. Clark, S. J., and Wotiz, K., Separation and detection of nanogram amounts of steroids. Steroids 2, 535-540 (1963). *C3. Conney, A. H., Pharmacological implications of microsomal enzyme induction. Pharmawl. Rev. 19, 317-366 (1967). C4. Conrad, S., Mahesh, V., and Herrmann, W., Quantitative paper chromatography of conjugated 17-ketosteroids in plasma. J . Clin. Invest. 40, 947-935 (1961). *C5. Cope, C. L., and Black, E. G., The behavior of 14C-cortisol and estimation of cortisol production rate in man. Clin. Sci. 17, 147-163 (1958). D l . Dalgliesh, C. E., Homing, E. C., Homing, M. G., Knox, K. L., and Yarger, K., A gas-liquid-chromatographic procedure for separating a wide range of metabolites occurring in urine or tissue extracts. Biochem. J. 101, 792-810 (1966). D2. DeCourcy, C., A trichloracetic acid reagent for the detection of pregnane3a:17a:2O-triols on paper chromatograms. J. Endocrinol. 14, 164-170 (1956). D3. Desty, D. H., Characteristics of capillary columns. Mem. SOC.Endocrinol. 16, 7-24 (1967). D4. Diczfalusy, E., I n vivo biogenesis and metabolism of estrogens in the foetoplacental unit. Proc. and Intern. Congr. Endocriml. London, 1964 Vol. 2, p. 732. Excerpta Med. Found., Amsterdam, 1965. El. Eaborn, C., and Walton, D. R. M., Preparation and gas chromatography of steroid bromomethyldimethylsilyl ethers. Chem. & Ind. (London) p. 827 (1967). *E2. Eik-Nes, K. B., and Homing, E. C., eds., “Gas Phase Chromatography of Steroids.” Springer, Berlin, 1968. E3. Epstein, E., and Zak, B., A new staining technique for ketosteroid chromatograms. J . Clin. Endocrinol. Metab. 23, 355-357 (1963). E4. Eriksson, H., Gustafsson, J-A., and Sjovall, J., Steroids in germ-free and conventional rats. European J . Biochem. 6, 219-226 (1968). F1. Farese, R. V., and Plager, J. E., The i n vitro red blood cell uptake of C14-cortisol; studies of plasma protein binding of cortisoI in normal and abnormal states. J . Clin. Invest. 41, 53-60 (1962). F2. Finkelstein, M., Forchielli, E., and Dorfman, R. I., Estimation of testosterone in human plasma. J. Clin. Endocrinol. Metab. 21, 98-101 (1961). F3. Finkelstein, M., and Goldberg, S., A test for quantitative and qualitative estimation of pregnane-3a, 17a,20a-diol-ll-one, in urine and its significance in adrenal disturbances. J . Clin. Endocrinol. 17, 1063-1070 (1957). F4. Finkelstein, M., Jewelewicz, R., Klein, O., Pfeiffer, V., and Sodmoriah-Beck, S., Fluorometric analysis of urinary oestrogens and clinical application. In “Prenatal Care,” pp. 12-27. Noordhoff , Groningen, The Netherlands, 1960. F5. Frankel, A. I., Cook, B., Graber, J. W., and Nalbandov, A. V., Determination of

DETERMINATION OF STEROIDS IN PLASMA AND URINE

133

corticosterone in plasma by fluorometric techniques. Endomhology 80, 181-194 (1967). F6. Fukushima, D. K., The origin of 3a-ureido-1lj3-hydroxy-A4-androsterone 17-one. J. Biol. Chem. 241, 2490 (1966). F7. Fukushima, D. K., Bradlow, H. L., Hellman, L., and Gallagher, T. F., On cortisol production rate. J. Clin. Endocrinol. 28, 1618-1622 (1968). *G1. Gabrilove, J. L., Sharma, D. C., and Dorfman, R. I., Adrenocortical 11b-hydroxylase deficiency and virilism first manifest in the adult woman. New Engl. J. Med. 272, 1189-1196 (1965). *Gla. Gandy, M., and Peterson, R. E., Measurement of testosterone and 17-ketosteroids in plasma by the double-isotope dilution derivative technique. J. Clin. Endocrinol. 28, 949-977 (1968). G2. Gardiner, W. L., and Homing, E. C., Gas-liquid chromatographic separation of CIS and C21 human urinary steroids by a new procedure. Biochim. Biophys. A C ~ 116, U 524-526 (1966). G3. Gardner, L. I., Adrenocortical metabolism of the fetus, infant and child. Review article. Pediatrics 17, 897-924 (1956). G4. Gemzell, C., Personal communication (1969). G4a. George, J. M., Wolff, S. M., Diller, E. D., and Bartter, F. C., Recurrent fever of unknown etiology: failure to demonstrate association between fever and plasma unconjugated etiocholanolone. J. Clin. Invest. in press. *G5. Gold, N. I., and Crigler, J. F., Jr., Effect of tritium labelling of cortisol on its physical-chemical propertie3 and on its metabolism in man. J. Clin. Endom’nol. Metab. 26, 133-142 (1966). G6. Goldberg, A., Moore, M. R., Beattie, A. D., Hall, P. E., McCallum, J., and Grant, J. K., Excessive urinary excretion of certain porphyrinogenic steroids in human acut.e intermittent porphyia, Lancet I, 115-118 (1969). G7. Goldfien, A., Yannone, M. E., McComas, D. B., and Poraga, C., The application of gas-liquid chromatography to the measurement of plasma progesterone. In “Gas Chromatography of Steroids in Biological Fluids” (M. B. Lipsett, ed.). Plenum Press, New York, 1965. *G8. Grant, J. K., The gas-liquid chromatography of steroids. Mem.SOC.Endocrinol. 16, 294 (1967). *G9. Gray, C. H., and Bacharach, A. L., eds., “Hormones in Blood.” Academic Press, New York, 1961. *G10. Gray, C. H., Baron, D. N., Brooks, R. V., and James, V. H. T., A critical appraisal of a method of estimating urinary 17-oxosteroids and total 17-oxogenicsteroids. Lancet I, 124-127 (1969). G11. Greenblatt, R. B., “The Hirsute Woman.” Thomas, Springfield, Illinois, 1963. Glla. Grundy, H. M., Simpson, S. A., and Tait, J. F., Isolation of a highly active mineralocorticoidfrom beef adrenal extract. Nature 169, 795-796 (1952). G12. Gurpide, E., Mann, J., and Lieberman, S., Analysis of open systems of multiple pools by administration of tracers at a constant rate or as a single dose as illustrated by problems involving steroid hormones. J . Clin.Endocrinol. 23, 1155-1176 ( 1963). G13. Gurpide, E., Schivers, J., Welch, M. T., Vande Wiele, R. L., and Lieberman, S., Fetal and maternal metabolism of estradiol during pregnancy. J. Clin. Endom’nol. 26,1355-1365 (1966). G14. Gustafsson, J-A., Steroid metabolism in germfree and conventional rats. Opuscula Med.Suppl.8 (1968). G15. Gustafsson, B. E., Gustafason, J-A., and Sjovall, J., Steroids in germfree and conventional rats. 2. Identification of 3a,16a-dihydroxy-5~-pregnan-2O-oneand

134

IAN E. BUSH

related compounds in faeces from germfree rats. European J. Biochem. 4, 56&573 (1968). H1. Heap, R. B., A fluorescence assay of progesterone. J. Endocrinol. 30, 293-305 (1964). Hla. Heys, R. F., Scott, J. S., Oakey, R. E., and Stitch, S. R., Urinary oestrogen in late pregnancy. Oestriol excretion as a guide to impending foetal death before term. Lancet I, 328-331 (1968). H2. Hofmann, A. F., Efficient extraction of bile acid conjugates with tetraheptylammonium chloride, a liquid ion exchanger. J. Lipid Res. 8, 55-58 (1967). H3. Holzbauer, M., and Newport, H. M., Quantitative estimation of 17a-hydroxypregn-4-ene-3,2O-dione (17cuOH-progesterone) in adrenal venous blood and adrenal glands. J . Physiol. (London) 198, 91-102 (1968). *H4. Horning, E. C., Vanden Heuvel, W. J. A., and Creech, B. G., Separation and determination of steroids by gas chromatography. Methods Biochem. Anal. 11, 69-148 (1963). *H5. Horton, R., and Neisler, J., Plasma androgens in patients with the polycystic ovary syndrome. J. Clin. Endocrinol. 28, 479-484 (1968). H6. Horton, R., and Tait, J. F., Androstenedione production and interconversionrates measured in peripheral blood and studies on t,he possible site of its conversion to testosterone. J . Clin. Invest. 46, 301-313 (1966). H7. Huggins, C., and Jensen, E. V., The depression of estrone-induced uterine growth by phenolic estrogens with oxygenatedfunctions a t positions 6 or 16: The impeded oestrogens. J . Exptl. Med. 102, 335-346 (1955). 11. Ismail, A. A. A., and Harkness, R. A., A method for the estimation of urinary testosterone. Bwchem. J . 99, 717-725 (1966). 12. Ittrich, G., Eine neue Methode zur chemischen Bestimmung der oestrogenen Hormone in Harn. Z . Physiol. Chem. 312, 1-14 (1958). J1. Jaffe, S. H., and Levitz, M., An accurate method for the determination of urinary estriol in pregnancy. Am. J. Obstet. Cynecol. 98,992-995 (1967). 52. Jirku, H., Kogsaneler, V., and Levitz, M., 15a-hydroxyestrone “Sulfate”: A biliary metabolite of estrone sulfate in the non-pregnant females. Biochim. Biophys. Acta 137, 588-591 (1967). 53. Jirju, M., and Layne, D. S., The formation of estradiol-3-glucuronoside-17a-Nacetylglucosaminide by rabbit liver homogenate. Biochemistry 4, 2126-2131 (1965). *J4. Johansson, E. D. B., Progesterone levels in peripheral plasma during the normal human menstrual cycle measured by a rapid competitive protein binding technique. Acta Endoerinol. (1969) (in press). 55. Jungmann, R. A., Calvary, E., and Schweppe, J. S., Quantitative analysis of urinary 1l-deoxy-17-ketosteroids and pregnanediol by gas-liquid chromatography. J . Clin. Endom‘nol. Metab. 27, 355-364 (1967). K1. Kappas, A., Pearson, 0. H., West, C. D., and Gallagher, T. F., A study of idiopathic hirsutism; a transitional adrenal abnormality. J . Clin. Endocrinol. Metab. 16, 517-528 (1956). K2. Kappas, A., H e h n , L., Fukushima, D. K., and Gallagher, T. F., The pyrogenic effect of etiocholanolone. J. Clin. Endocrinol. Metub. 17, 451-453 (1957). K3. Kato, T., and Horton, R., A rapid method for the estimation of testosterone in female plasma. Steroids 12, 637-639 (1968). K4. Keller, N., Sendelbeck, L. R., Richardson, V. I., Moore, C., and Yaks, F. E., Protein binding of corticosteroids in undiluted rat plasma. Endom.nology 179, 884-906 (1966).

DETERMINATION OF STEROIDS IN PLASMA AND URINE

135

K5. Kellie, A. E., and Smith, E. R., Characterization of 17-oxosteroidsby derivatives formed in the Zimmerman reaction. Nature 178, 323 (1956). K6. Kirschner, M. A., and Coffman, G. D., Measurement of plasma testosterone and Acandrostenedione using electron-capture gas-liquid chromatography. J . Clin. Endocrinol. 28, 1347-1355 (1968). K7. Kittinger, G. W., Quantitative gas chromatography of adrenal cortical steroids. Steroids 11, 47-71 (1968). K8. Klopper, A., Mitchie, E. A., and Brown, T. B., A method for the determination of urinary pregnanediol. J . Endocrinol. 21, 209-219 (1955). *K9. Klyne, W., “Chemistry of the Steroids.” Methuen, London, 1957. *K10. Knights, B. A., Gas chromatography of steroids. I n “Steroid Hormone Analysis” (H. Carstensen, ed.), pp. 367-406. Marcel Dekker, New York, 1967. Kll. Korenman, S. G., Radio-ligand binding assay of specific estrogens using a soluble uterine macromolecule. J . Clin. Endocrinol. 28, 127-130 (1968). L1. Lapin, I. P., and Oxenkrug, G. F., Intensification of the central serotoninergic processes as a possible determinant of the thymoleptic effect. Lancet I, 132-136 (1969). L2, Levitz, M., Dancis, J., Goebelsmann, U., and Diczfalusy, E., The metabolism and disposition of estriol and estriol conjugates of the fetal-placental unit in the second trimester. Proc. 2nd Intern. Congr. Hormonal Steroids, Milan, 1966 pp. 646-652. Excerpta Med. Found., Amsterdam, 1966. L3. Lewis, B., Preparation of paper for quantitative chromatography of corticosteroids. Biochim. Biophys. Acta 20, 396 (1956). L4, Leymarie, P., and Savard, K., Steroid hormone formation in the human ovary. VI. Evidence for two pathways of synthesis of androgens in the compartment. J . Clin. Endom’nol. 28, 1547-1554 (1968). L5. Lieberman, S., and Gurpide, E., Isotopic dilution methods for the estimation of the rates of secretion of steroid hormone. I n “Steroid Dynamics” (G. Pincus, J. F. Tait, and T. Nakao, eds.), pp. 531-547. Academic Press, New York, 1966. L6. Lindner, H. R., Androgens in spermatic vein blood of farm animals. J . Endocrinol. 20, v-vi (1960). L7. Lingren, W. E., Reeck, G., and Marson, R., Contamination of paper chromatograms by scissors. J . Chromatog. 24, 269 (1966). *L8. Lipsett, M. B., ed., “Gas Chromatography of Steroids in Biological Fluids.” Plenum Press, New York, 1965. L9. Lipsett, M. B., Personal communication (1968). L10. Lipsett, M. B., Hormonal syndromes associated with neoplasia. Advan. Metab. Disorders 3, 112-146 (1968). L11. Lipsett, G. B., and Kirschner, M. A., Analysis of urinary pregnanediol and pregnanediol by gas-liquid chromatography. I n “Gas Chromatography of Steroids in Biological Fluids” (M. B. Lipsett, ed.), pp. 135-142 Plenum Press, New York, 1965. *L12. Lipsett, M. B., Korenman, S. G., Wilson, H., and Bardin, C. W., The effects of metabolic transformations of androgens. I n “Steroid Dynamics” (G. Pincus, J. F. Tait, and T. Nakao, eds.), pp. 117-129. Academic Press, New York, 1966. L13. Lisboa, B. P., Separation and characterisation of A4-3-ketosteroidsof the pregnaneseries by means of thin-layer chromatography. Acta Endocrinol. 43,47-66 (1963). L14. Lisboa, B. P., Separation of 21-deoxy A4-3-oxo-steroidsof the pregnane series by thin-layer chromatographic procedures. Steroids 8, 139-343 (1966). L15. Lisboa, B. P., Formation and separation of Girard hydrazones on thin-layer chromatography by elatographic techniques. J . Chromatog. 24, 475-477 (1966).

136

IAN E. BUSH

L16. Lisboa, B. P., and Diczfalusy, E., Separation and characterisation of steroid oestrogens by means of thin-layer chromatography. Acta Endocrinol. 40, 60-81 (1962). L17. Lovelock, J. E., A sensitive detector for gas chromatography. J. C h r m t o g . 1, 35-46 (1958). MI. Mahesh, V. B., and Greenblatt, R. B., Physiology and pathogenesis of the SteinLeventhal syndrome. Nature 191, 888-890 (1961). Mla. Mahesh, V. B., and Greenblatt, R. B., Steroid secretions of the normal and polycystic ovary. Recent Progr. Homnone Res. 20, 341-394 (1964). M2. Mattox, V. R., and Goodrich, J. E., Use of liquid ion exchange agents for chromatography of steroidal glucuronides. Federation Proc. 27, 822 (1968). *M3. Mayes, D., and Nugent, C. A., Determination of plasma testosterone by the use of competitive protein binding. J. Clin. Endocrinol. 28, 1169-1176 (1968). M4. Mikhail, G., Sex steroids in blood. Clin. Obstet. Gynecol. 10, 29-39 (1967). M5. Mikhail, G., Zander, J., and Allen, W. M., Steroids in human ovarian vein blood. J . Clin. Endocrinol. 23, 1267-1270 (1963). M6. Moss, M. S., and Rylance, H. J., A thin-layer chromatography study on the metabolism of prednisolone in the horse. J. Endominol. 37, 129-137 (1967). *M7. Murphy, B. E. P., Clinical evaluation of urinary cortisol determinations by competitive protein-binding radioassay. J. Clin. Endocrinol. Metab. 28, 343-348 (1968). M8. Murray, D., Cortisol binding to plasma proteins in man in health, stress and a t at death. J. Endocrinol. 39, 571-591 (1967). *N1. Neill, J. D., Johansson, E. D. B., Datta, J. K., and Knobil, E., Relationship between the plasma levels of luteinizing hormone and progesterone during the normal menstrual cycle. J. Clin. Endocrinol. Metub. 27, 1167-1173 (1967). N2. Neill, J. D., Johansson, E. D. B., and Knobil, E., Levels of progest,erone in peripheral plasma during the menstrual cycle of the rhesus monkey. Endocrinology 81, 1161-1164 (1967). N3. Nelson, D. C., Ressler, P. C., Jr., and Hawes, R. C., Performance of an instrument for simultaneous gas chromatographic and radioactivity analysis. Anal. Chem. 36, 1575-1579 (1963). N4. Nelson, D. H., and Samuels, L. T., A method for the determination of 17-hydroxycorticosteroids in blood : 17-hydroxycorticosteronein the peripheral circulation. J . Clin. Endocrinol. 12, 519-526 (1952). N5. Nerenberg, C. A., Sharma, D. C., and Dorfman, R. I., An automatic device for adding solvents to chromatography tanks. Anal. Biochem. 20, 284-298 (1967). N5a. Nichols, T., Nugent, C. A., and Tyler, F. H., Glucocorticoid suppression of urinary testosterone excretion in patients with idiopathic hirsutism. J. Clin. Endocrinol. 26, 79-86 (1966). *N6. Nugent, C. A., and Mayes, D. M. (with technical assistance of S. Morrett) Plasma corticosteroids determined by use of corticosteroid-binding globulin and dextran-coated charcoal. J . Clin. Endocrinol. Metab. 26, 1116-1122 (1966). 01. Oakey, R. E., Stitch, S. R., Heys, R. F., and Scott, J. S., Practicability and cost of oestriol assays for saving babies in a maternity hospital. Lancet I, 331-332 (1968). 02. Oertel, G. W., Beckmann, I. N., and Eik-Nes, K. B., The measurement of pregnane301,20a-diol and its isomers in human plasma. Arch. Biochem. Biophys. 86, 148-153 (1960). P1. Pal, S. B., A fluorimetric determination of unconjugated cortisol in human urine after paper chromatography. A d a Endocrinol. 61,415-428 (1966).

DETERMINATION OF STEROIDS IN PLASMA AND URINE

137

P2. Pasqualini, J. R., In “Steroid Hormone Analysis” (H. Carstensen, ed.), pp. 412414. Marcel Dekker, New York, 1967. *P3. Paulsen, C. A., ed., “Estrogen Assays in Clinical Medicine.” Univ. of Washington Press, Seattle, Washington, 1965. P4. Pearlman, W. H., Circulating steroid hormone levels in relation to steroid hormone production. Ciba Found. Colloq. Endrocrinol. 11, 233-251 (1957). P5. Pearlman, W. H., Pearlman, M. R. J., and Rakoff, A. E., Estrogen metabolism in human pregnancy; a study with the aid of deuterium. J . Biol. Chem. 209, 803-812 (1954). P5a. Peterson, R. E., The miscible pool and turnover rate of adrenocortical steroids in man. Recent Progr. Hormone Res. 16, 231-274 (1959). *P6. Peterson, R. E., Determination of urinary neutral 17-ketosteroids. I n “Standard Methods of Clinical Chemistry” (D. Seligson, ed.), Vol. 4, pp. 151-162. Academic Press, New York, 1963. *P6a. Peterson, R. E., Icarrer, A., and Guerra, S. L., Evaluation of the Silber-Porter procedure for the determination of plasma hydrocortisone. Anal. Chem. 29, 144-149 (1957). P7. Peterson, R. E., and Wyngaarden, J. B., The miscible pool and turnover rate of hydrocortisone in man. J . Clin. Invest. 36, 552-561 (1956). P8. Pincus, G., Nakm, T., and Tait, J. F., eds., “Steroid Dynamics.” Academic Press, New York, 1966. *P9. Pray, L. L., “Taxidermy.” Macmillan, New York, 1945. P10. Preedy, J. N., and Aitken, E. M., The determination of estrone, estradiol-li’j3, and estriol in urine and plasma with column partition chromatography. J. Bid. Chem. 236, 1300-1311 (1961). P11. Puro, H., Epstein, E., and Zak, B., Adaptation of a manual fluorometer for continuous paper-strip scanning and recording. Chemist-Analyst 64, 54-55 (1965). *R1. Riondel, A., Tait, J. F., Tait, S. A. S., Gut, M., and Little, B., Estimation of progesterone in human peripheral blood using 3%-thiosemicarbazide. J. Clin. Endocrinol. 23, 620-628 (1963). *R2. Riondel, A., Tait, J. F., Gut, M., Tait, S. A. S., Joachim, E., and Little, B., Estimation of testosterone in human peripheral blood using aKS-thiosemicarbazide. J . Clin. Endocrinol. 23, 620-628 (1963). R3. Rivarola, M. A., and Migeon, C. J., Determination of testosterone and androst4-ene-3,17-dione concentration in human plasma. Steroids 7, 1103-1171 (1966). R4. Roberts, J. B., Bush, I. E., and Gibree, N. B., Urinary steroids analysis: A comprehensive procedure. Anal. Biochem. 23, 378-390 (1968). R5. Roberts, K. D., Bandi, L., Calvin, H. I., Drucker, W. D., and Lieberman, S., Evidence that steroid sulfates serve as biosynthetic intermediates. IV. Conversion of cholestrol sulfate in vivo to urinary C19 and C21 steroidal sulfates. Biochemistry 3, 1983-1988 (1964). R6. Ryhage, R., Use of a m a s spectrometer as a detector and analyzer for effluents emerging from high temperature gas-liquid chromatography columns. Anal. Chem. 36, 759-764 (1964). S1. Salganicoff, L., Kraybill, M., Mayer, D., and Legallais, V., Sensitive dual wavelength densitometer with suppression of the support introduced background optical density. J . Chromatog. 26, 434-441 (1967). 52. Salokangas, R. A. A., and Bulbrook, R. D., The determination of small quantities of urinary oestrone, oestradiol-17p and oestriol using Ittrich’s extraction method. J . Endorrinol. 22, 47-58 (1961).

138

IAN E. BUSH

S2a. Sans, M. C., Handling pl. and sub-pl. quantities of solutions. Memo. SOC.Endocrinol. 16, 27-35 (1967). S3. Short, R. V., Progesterone. In “Hormones in Blood” (C. H. Gray and A. L. Bucharach, eds.), pp. 379. Academic Press, New York, 1961. S4. Silber, R. H., Bush, R. D., and Oslapas, R., Practical procedure for estimation of corticosterone and hydrocortisone. Clin. Chem. 4, 278-285 (1958). *S5. Silber, R. H., and Porter, C. C., Determination of 17,2l-dihydroxy-20-ketosteroids in urine and plasma. Methods Biochem. Anal. 4, 139-169 (1957). S6. Sjovall, J., Personal communication (1969) (in preparation). S6a. Stacy, R. W., and Waxman, B. D., eds., “Computers in Biomedical Research,” Vols. 1 and 2. Academic Press, New York, 1965. 57. Starnes, W. R., Rhodes, A. H., and Lindsay, R. H., Thin-layer chromatography of 17-ketosteroid 2,4-dinitrophenylhydrazones. J . Clin. Endominol. Metab. 26, 1245-1250 (1966). S7a. Stein, I. F., and Leventhal, M. L., Amenorrhea associated with bilateral poloycystic ovaries. Am. J . Obstet. +necol. 29, 181-191 (1935). 58. Strott, C . A., and Lipsett, M. B., Measurement of 17a-hydroxprogesterone in human plasma. J . Clin. Endocrinol. 28, 1426-1430 (1968). *S9. Sulimovici, S., Lunenfeld, B., and Shelesnyak, M. C., A practical method for estimation of urinary pregnanediol and allopegnanediol using thin layer chromatography. Acta Endoerinol. 49, 97-106 (1965). *T1. Tait, J. F., The use of isotopic steroids for the measurement of production rates in vivo. J . Clin. Endocrinol. 23, 1285-1297 (1963). T2. Tait, J. F., Personal communication (1964). *T3. Tait, J. F., and Burstein, S., In vivo studies of steroid dynamics in man. I n “The Hormones” (G. Pincus, K. V. Thimann, and E. B. Astwood, eds.), Vol. 5, pp. 441-557. Academic Press, New York, 1964. T4. Tait, J. F., and Horton, R., The in vivo estimation of blood production and interconversion rates of anderstenedione and testosterone and the calculation of their secretion rates. In “Steroid Dynamics” (G. Pincus, J. F. Tait, and T. Nakao, eds.), pp. 393-425. Academic Press, New York, 1966. T5. Tamm, J., ed., “Testosterone,” Thieme, Stuttgart, 1968. T6. Thomas, B. S., and Walton, D. R. M., Chloromethyldimethylsilyl ethers in the routine assay of urinary ll-deoxy-17-oxosteroids by gas-liquid chromatography. J . Endocrinol. 41, 203-211 (1968). T7. Thomas, G. H., The RM approach to the structural analyses of steroid metabolites. Memo. SOC.Endominol. 16, 129-140 (1967). T8. Tomisek, A. J., and Wedeles, P., Water content of paper as a variable in paper chromatography. J . Chromatog. 33, 35-37 (1968). T9. Touchstone, J. C., Wu, C-H., Nikolski, A., and Murawec, T., Retention behavior of steroids in gas chromatography with a series of combination columns. J . Chromatog. 29, 235-238 (1967). T10. Townsend, J., and James, V. H. T., A semi-automated fluorimetric procedure for the determination of plasma corticosteroids. Steroids 12,497-511 (1968). *V1. van der Molen, H. J., Gas phase chromatography of progesterone and related steroids. “Gas Phase Chromatography of Steroid,” (K. B. Eik-Nes and E. C. Horning, eds.), pp. 190-235. Springer, Berlin, 1968. V2. van der Molen, H. J., and Groen, D., Determination of progesterone in human peripheral blood using gas-liquid chromatography with electron-capture detection. J . Clin. Endocrinol. 26, 1625-1639 (1965).

DETERMINATION OF STEROIDS IN PLASMA AND URINE)

139

V3. Vande Wiele, R. L., MacDonald, P. C., Gurpide, E., and Lieberman, S.,?Studiea on the secretion and interconversion of androgens. Recent Progr. Hormone Res. 19, 175-310 (1963). V4. Vermeulen, A., Determination of testosterone in plasma by electron-capture detection of the heptafluorobutyrate. In “Testosterone” (J. Tamm, ed.), pp. 13-16. Thieme, Stuttgart, 1968. *V5. Vestergaard, P., Technique for simultaneous multiple-column gradient elution chromatography. J . Chromatog. 3, 560-569 (1960). *V6. Vestergaard, P., A multicollector: A fraction collector for the simultaneous collection of fractions from a number of chromatographic columns. J. Chromatog. 3, 554-559 (1960). V7. Vestergaard, P., and Sayegh, J. F., Semi-automated assays for urinary 17ketogenic steroids; a comparison of bismuthate and periodate oxidation methods. Clin. Chim. Acta 14, 247-262 (1966). V8. Vestergaard, P., and Sayegh, J. F., Capillary teflon columns for adsorption and partition chromatography. J. Chromatog. 24, 422-426 (1966). V9. Vestergaard, P., Sayegh, J. F., and Witherell, C. S., U-shaped columns for adsorption and partition chromatography. J . Chromatog. 24, 417-421 (1966). V10. Vestergaard, P., Witherell, C., and Piti, T., Magazine Fed Fraction Collector for multi-column liquid chromatography. J . Chromatog. 31, 337-344 (1967). W l . Wade, A. P., Slater, J. D. H., Kellie, A. E., and Holliday, M. E., Urinary excretion of 17-KS following rectal infusion of cortisol. J . Clin. Endocrinol. 19, 444-458 (1959). W2. Wettstein, A., Conversion of steroids by micro-organisms. Ezperientia 11,465-479 (1955). *W3. Whitby, L. G., Mitchell, F. L., and Moss, D. W., Quality control in routine chemical chemistry. Aduan. Clin. Chem. 10, 66-156 (1967). W4. Wieland, R. G., DeCourcy, C., Levy, R. P., Zala, A. P., and Hirschmann, K., C,gOt steroids and some of their precursors in blood from normal human adrenals. J . CZin. Invest. 44, 159-168 (1965). W5. Williams, E. J., “Regression Analysis.” Wiley, New York, 1959. W6. Williams, T., “Detoxication Mechanisms.” Chapman & Hall, London, 1959. W7. Wilson, H., and Fairbanks, R. A., Micro method for the detection and assay of 17-hydroxy~-glycols.Arch. Biochem. Biophys. 64, 457-466 (1955). steroid W8. Witzel, H., Uber anaesthetisch wirksame steroide. 2.Vitamin-, Hormon- Fermentforsch. 10, 46 (1959). *Z1. Zak, B., and Epstein, E., Automation of the photometric determination of 17-ketosteroids. Chemist-Analyst 62, 45-47 (1963).

This Page Intentionally Left Blank

THE INVESTIGATION OF STEROID METABOLISM I N EARLY INFANCY Frederick 1. Mitchell and Cedric H. 1. Shackleton Division of Clinical Chemistry, Medical Research Council Clinical Research Centre, Northwick Park, Harrow, Middlesex, England, and Clinical Endocrinology Unit, Edinburgh, Scotland 1. Introduction.. . . . . . . . . . . . . ................................. 2. Nomenclature.. . . . . . . . . . . . . ................................ 3. Methodology.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 3.1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . ............. ............................. 3.2. Methods of Hydrolysis. 3.3. Assays of Groups of Steroids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Assays of Individual Steroids. . . . . . . . . . . . . . . . . 4. The Influence of Steroid Metabolic Pathways Used in 4.1. Metabolism of C19Steroids and Estrogens. . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Metabolism of CZISteroids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . ...................... 4.3. Steroid Content of Amniotic Fluid 4.4. Conclusions.. ............ .............................. 5. Steroid Assays on Umbilical Cord ............................... 5.1. 3@-Hydroxy-A6Steroids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Progesterone and Its Metabo ............................ ............................ 5.3. Cortisol and Its Metabolites. 5.4. Testosterone and the 17-0x0 Steroids.. . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5. Estrogens. ..... .......................................... 6. 17-0x0 Steroids in Infa lood and Urine.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. The Urinary Excretion, Blood Levels, and Production of Cn Steroids.. . . . . . 7.1. Urinary C ZSteroids ~ Assayed by Group Methods.. . . . . . . . . . . . . . . . . 7.2. Urinary Excretion of Individual CZISteroids.. . . . . . . . . . . . . . . . . . . . . 7.3. Blood Levels, Production Rates, and Metabolism of Cortisol, Cortisone, and Corticosterone. . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. The Urinary Excretion and Blood Levels of 3@-Hydroxy-A6Steroids.. . . . . . . 9. The Urinary Excretion of Estrogens.. . . . . . . . . . . . . . . . . . . . . . . . . . . 10. Steroid Conjugation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Abnormalities in Steroid Production. . . . . . . . . . . . . . . . ....... 11.1. Congenital Adrenal Hyperplasia. . . . . . . . . .................. 11.2. Congenital Adrenal Hypoplasia ...................... 11.3. Defects in the Production of Al 12. The Control of Steroid Production in Infancy.. . . . . . . . . . . . . . . . . . . . . . . . . . 13. Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

142 143 143 147 148

157 160 168 169 170 170 171 172 172 174 174 177 177 178 180 180 181

189 191 196 197

142

FREDERICK L. MITCHELL AND CEDRIC H . L. SHACKLETON

1.

Introduction

Striking anatomical changes take place in the adrenal glands of the human infant during the first few days and weeks of life (B9, M6, 526, S31) while they are adapting from the type of steroid production necessary in utero to that required for independent life, and during this period considerable qualitative and quantitative changes occur in the urinary steroid output. For many years it was thought that the placenta acted as an independent endocrine gland during pregnancy, and only comparatively recently was it realized that the fetus is also considerably involved in steroid production and catabolism, certain mechanisms for the metabolism of steroids being present in the fetus and not in the placenta, and vice versa. The complex steroid metabolic interrelationships which exist between the fetus, the placenta, the maternal endocrine glands and the mother as a whole have been discussed in a previous review (M16) and will be covered only insofar as they affect an understanding of the situation in early infancy. At birth three populations of steroids may be considered to exist in the infant: (1) those arising from the metabolism of the large amounts of estrogen and progesterone received from the placenta ; (2) 3p-hydroxyA5 steroids produced in utero as precursors of estrogen and possibly progesterone; and (3) cortisol and other steroids normally found in adulthood. From all three groups, steroids or their metabolites might be expected to be excreted in the urine of the newborn infant. I n the case of group 1, such compounds would be excreted after birth in rapidly decreasing quantity since their source has been eliminated (D10). The steroids in group 2 are produced by enzyme systems which operated for special purposes in utero, but production does not cease a t birth and the urinary excretion may actually increase during the first month (R6) when steroids with the 3P-hydroxy-A5 configuration predominate quantitatively and qualitatively over all the others (S9). Of the adult-type steroids considered in group 3, cortisol is produced in near normal amount (when related to surface area) throughout infancy (B12), corticosterone production (related to surface area) is more than three times that of the adult (L17), and the ll-deoxy-17-oxosteroids (DHA, etiocholanolone, and androsterone) normally present in adults, are almost completely absent in infant urine (C6). Thus, in infants, it is in the presence of a complex and changing normal pattern that any abnormality of steroid metabolism must be identified. Three media are available in reasonable quantity for investigation: (1) a sample of amniotic fluid may be collected a t birth, and since this contains steroids excreted in urine by the fetus, its contents may give an

STEROID METABOLISM I N EARJAY INFANCY

143

indication of steroids produced and excreted in utero; (2) approximately 50 ml of umbilical vein blood and 15 ml of umbilical artery blood may be obtained a t birth from the severed umbilical cord, and from this may be obtained an accurate measure of the type and quantity of steroids sent to, and obtained from, the placenta by the fetus; (3) 24hour specimens of urine may be collected after birth and assays on these will show any changing pattern of urinary steroid excretion taking place. It is not usually practicable to assay steroids in the quantities of blood which can reasonably be removed from a n infant. Steroids have been assayed in meconium and it has recently been shown that feces collected from infants up to 6 months after birth contain relatively large amounts (G7): it is nat yet known, however, whether the measurement of steroids in meconium or feces will be of diagnostic significance. Most of the knowledge available has been obtained from work done on the three media mentioned, plus incubation and perfusion studies using fetal and placental material from therapeutic abortions carried out before week 24 of pregnancy. Experimentation on infants must, for ethical reasons, be strictly limited to procedures that produce no harm or inconvenience, and for this reason full advantage has been taken of treatments given for therapeutic purposes, for example, exchange blood transfusions or the administration of corticotropin and gonadotropin for various disorders. Much information has also been obtained by the investigation of anencephalic monsters both in utero and during their limited life. I n the fetoplacental unit and in early infancy, steroid metabolism is dominated by many steroids which are relatively unimportant later in life (for list, see M16), and many steroidlike compounds present a t this time remain to be identified. This review is concerned with the normal steroid metabolism which is to be expected in the newborn period and in infancy, techniques which may be used in its investigation, and the clinical significance of any abnormalities. It is important to stress that the overall understanding of the situation is still incomplete and the clinical significance of many of the aspects is not known. 2.

Nomenclature

Throughout the text, full names have been used for steroids not referred to repetitively, but for the more common compounds mentioned, the trivial names in the table on pp. 146146 are used throughout.

3. Methodology 3.1. INTRODUCTION Until comparatively recently, the examination of infant urine for steroids of adrenal origin was done mainly by means of group steroid

Trivial name Adrenosterone Aldosterone Allo-tetrahydrocorticosterone Allo-THE A110-THF Androstenediol (17a) Androstenediol (178) 16-0x0-androstenediol Androstenedione 118-OH-androstenedione Androstenetriol Androsterone 118-OH-androsterone 11-0x0-androsterone Corticosterone 18-OH-corticosterone Cortisol 16a-OH-cortisol 68-OH-cor tisol Cortisone 6p-OH-cortol 68-OH-cortolone DHA 16a-OH-DHA 168-OH-DHA 11-Deoxycorticosterone 11-Deoxycortisol 20a-Dihydroprogesterone

Systematic name androsM-ene-3,11,17-trione 11~,21-dihydroxypregn-4-ene-3-20-dion-18-al 3a, 11~,2l-trihydroxy-5a-pregnan-2O-one 3a117a121-trihydroxy-5a-pregnane-11,20-dione 3a,11~,17a,2l-tetrahydroxy-5a-pregnan-20-one 3j3,17a-dihydroxyandrost5-ene 3p117j3-dihydroxyandrost5-ene 3p,17p-dihydroxyandrost5-en-16-one androsM-ene-3,17-dione 1l~-hydro~yandrost-4-ene-3~17-dione 3p116a,178-trihydroxyandrost-5-ene 3a-hydroxy-5a-andros tan-17-one 3a,1l~-dihydroxy-5a-androstan-17-one 3a-hydroxy-5a-androstane-ll,l7-dione llj3,21-dihydroxypregn-4-ene-3,20-dione 11p,18,21-trihydroxypregn-4-ene-3,2O-&one 118,17a,2l-trihydroxypregn-4-ene-3,20-dione 118,16a117~,2l-tetrahydroxypregn4ene-3,2O-dione 68, 118,17a,2l-tetrahydroxypregn4-ene-3,20-dione 17a,21-dihydroxypregn-4-ene-3,1l12O-trione 3a168, 118,17a,208,21-hexahydroxy-58-pregnane 3a,6p,17a,20~,2l-pentahydroxy-5~-pregnan-ll-one 3j3-hydroxyandrostr5-en-17-one 3~,16a-dihydroxyandrost5-en-17-one 3~,16~-dihydroxyandrost5-en-17-one 2 l-hydro~ypregn-4-ene-3~20-dione 17a,2l-dihydroxypregn-4-ene-3,20-dione 20a-hydroxypregn-4-en-3-one

F

9

3

H

0

z

20p-Dihydroprogesterone 17a,21-Dihydroxypregnenolone 16-Epies t riol Rpitestosterone Estetrol Estradiol 16-0x0-es tradiol Estriol Estrone 16a-OH-es trone 16p-OH-estrone E tiocholanolone 1Ip-OH-etiocholanolone 11-0x0-etiocholanolone

3p-HSD 17-OHCS 17-08 Pregnanediol Pregnanetriol 11-0x0-pregnanetriol Pregnanolone Pregnenolone 16a-OH-pregnenolone 17a-OH-pregnenolone 21-OH-pregnenolone Progesterone I&-OH-proges terone l7a-OH-progesterone 6p-OH-progesterone Testosterone

2Op-hydroxypregri4-en-3-one 3p,1701,2l-trihydroxypregn-5-en-20-one 3,16p,17p-trihydroxyestra-l,3,5(10)-triene 17a-hydroxyandrost-4-en-3-one 3,1501,16a,17p-tetrahydrosyestra-1,3,5(10)-triene 3,17p-dihydroxyestra-l,3,5(lO)-triene 3,17p-dihydroxyestra-1,3,5(10)-trien-16-one 3,16aY, 17p-trihydroxyestra-1,3,5(lO)-triene 3-hydroxyestra-1,3,5( 10)-trien-l7-one 3,16a-dihydroxyestra-l,3,5 (10)-trien-l'l-one 3,16p-dihydroxyestra-l,3,5(10)-trien-17-one 3a-hydrosy-K~-androstan-17-one 3a, 118-dihy droxy-58-androstan- 17-one 301-hydroxy-5p-androstan-ll,l'i-dione 3p-hydroxysteroid dehydrogenase plus epimerase 17-hydroxycortical steroids 17-0x0 steroids 3ay,20a-dihydroxy-5p-pregriane 3a,1i'a,20a-trihydroxy-5p-pregnarle 3a,17a,20a-trihydrosy-5p-pregnari-ll-one 3a-hydroxy-6p-pregnan-20-one 3p-hydroxypregn-5-en-20-one 3p, 16a-dihydroxypregn-5-e1i-2O-one 3p, 17a-dihydroxypregn-5-en-20-or1e 3p,21-dihydroxypreg11-5-en-20-one pregn-4-ene-3,20-dione 16a-hydroxypregn4-ene-3,20-dione 17a-hydroxypregn-4-ene-3,20-dione 6p-hydroxypregn-4-ene-3,20-dione 17pbydroxyandrost -4-en-3-one

.

(Continued)

d

e

c.L

+

UI

Trivial n a m e Tetrahydroaldosterone Tetrahydrororticosterolie Tetrahydro-11-deoxycorticosterone Tetrahydro-11-dehydrocorticosterone Tetrahydro-11-deoxycortisol

THE THF THS 1 lp, 17a,21-Trihydroxypregnenolone

Systematic n a m e 3a,1lp,21-trihydroxy-6p-pregnan-20-on-18-al

3a, 1 lg,21-trihydroxy-5p-pregnan-20-one 3a,21-dihydroxy-T,p-pregnan-20-one 3a,21-dihydroxy-.5p-pregnane-l1,20-dione 3a,17a,21-trihydroxy-5p-pregnan-20-one 3a,17a,2l-trihydroxy-:ip-pregnane-l1,20-dione 3a,1lp,17a,21-tetrahydroxy-5p-pregnan-20-one 3n, 17a,2l-trihydroxy-rip-pregnan-20-0ne 3p,1 lp, 17a,21-tetrahydroxypregn-5-en-20-one

d

m

U

E

0

x

r

STEROID METABOLISM I N EARLY INFANCY

147

assays designed to measure the principal end products of adrenal steroid metabolism in adults. When, however, special fractionation techniques were developed (B18, R2, R3, S9), it was found that major qualitative and quantitative differences exist in the steroid content of infant and adult urine (B16, C6, M16, R2, R3, R6, S9, S11,513), and the compounds measured by group assay techniques originally designed for investigations on adults are very different when these are applied t o infants. Considerable caution must therefore be applied in the interpretation of any results. Similarly the steroid profile of umbilical cord blood is very unlike that in adult blood and is dominated by large quantities of steroids with the 3p-hydroxy-A5 configuration and by estrogens and progesterone and their metabolites. Many of the compounds involved are unique to this period of life, and it has been necessary to develop special techniques for their measurement (C12, '213, E4, S9, SlO, S11, S20, 24). It is therefore evident t h a t in addition to special compounds requiring new techniques for their assay, all methods designed for steroid investigations in adults must be reassessed before being applied to infants; not only is the composition of steroids in group assays different, but nonspecific chromogen interference in these assays can be considerable (BlO, B12, M10, W5). Also, techniques for the measurement of single compounds such as pregnandiol and pregnanetriol in adult urine may be completely invalid for such assays in the urine of infants, where the contaminants may be very different from those which the techniques were specially designed to remove from adult urine. I n this section methods of hydrolysis and the difficulties involved in group assays will be discussed, and techniques will be mentioned which are now available for the measurement of single steroids either individually or collectively in profile analysis. 3.2. METHODSOF HYDROLYSIS Drayer and Giroud (D16) have shown that, although the proportion of steroids excreted in urine free is considerably increased in infants, sulfate conjugation is also increased, and for the 17-OHCS, sulfation is a t least as important as conjugation with glucuronic acid (see Section 10). The important 3P-hydroxy-A5 steroids in umbilical cord blood and infant urine are also mainly present as sulfates (E2, EX, M l , R6, S20) ; only very small portions of D H A and ~ ~ w O H - D H Afor , instance, have been found free in blood (S21). Because of the importance of sulfate conjugation, special care must be taken, if enzyme hydrolysis is used alone, to ensure the presence of the specific sulfatases required. For instance, a t least one steroid present in infant urine is diconjugated and cannot be completely hydrolyzed by the enzymes in the crop fluid of

148

FREDERICK

L. MITCHELL AND CEDRIC

H.

L. SHACKLETON

Helix pomatia ( S l l ) (Section 10; Fig. 5 ) . Alternative modes of conjugation other than sulfation and with glucuronic acid have been indicated (MIO, 527). Most assay techniques depend upon hydrolysis of any steroid conjugates before extraction. The use of hot acid is normally applicable only to the assay of 17-0s (either present in urine as such or produced by bismuthate or periodate oxidation), the three major estrogens, and pregnanediol, and even with these compounds major qualitative changes can occur (C9). If other steroids are to be preserved intact, milder methods involving either solvolysis (for sulfates) or the action of enzymes must be used. Solvolysis is particularly important for infant urine because of the preponderance of sulfate conjugation. Complete hydrolysis of all steroid conjugates in infant urine cannot a t present be guaranteed since the nature of many is not known, but a comprehensive system to cover all known conjugates can be devised. The visceral hump enzyme preparation of Patella vulgata is a richer source of p-glucuronidase than is the crop fluid of Helix pomatia, but the reverse is true for 3p-hydroxyh5 steroid sulfatase. The Helix pomatia enzyme preparation also contains other steroid sulfatases (LIO) and both preparations contain enzymes other than sulfatases and p-glucuronidase (D14, M21,Ol)-an advantage for the hydrolysis of unknown conjugates. To utilize the features of both preparations t o advantage, Birchall et al. (B18) used them sequentially, incubating once for 24 hours with the Helix pomatia preparation followed by a second 24-hour incubation with the Patella vulgata preparation. Shackleton et al. (S10) used only the Helix pomatia enzyme, but i t was found t h a t a two-stage incubation was necessary to achieve maximal hydrolysis of cortisol metabolites. Because of the long incubation periods, it is necessary to suppress bacterial action by adding penicillin and streptomycin. Care must be taken initially to remove enzyme inhibitors from the urine. Phosphate and sulfate ions markedly inhibit the sulfatase activity of both the Helix pomatia and Patella vulgata preparations (H8, 55,R16, 530) and may be removed satisfactorily by precipitation with barium chloride a t p H 11.5 (55, 529). This procedure also removes the specific p-glucuronidase inhibitor, saccharo-1 :4-lactone (M4). The action of the sulfatases in the Helix pomatia enzyme is insufficient to achieve complete hydrolysis a t least of the 3P-hydroxy-A5 steroids in umbilical cord blood and infant urine, and a further stage involving a solvolytic procedure (B40) is essential (S9, S11). 3.3. ASSAYSOF GROUPSOF STEROIDS

It will be seen from Fig. 1 t h a t the absorption spectra of the Zimmermann chromogens in urine from infants up to 2.5 years of age vary

STEROID METABOLISM IN EARLY INFANCY

i i

149

DHA

0.1

01

1

1

360 400

I

I

I

'

440 400

I

'

I

'

520 560 Wavelength (mp)

I

'

1

600

FIG.1. Absorption spectra of Zimmermann chromogens obtained with 2 ml of urine from: a, a pool from infants 1-3 days old; 0, a pool from infants 6 days old; 0, a 6-month-old infant; 0 ,a pool from infants 1-2.5 years old; A,a pool of adult urine. From Birchall and Mitchell (B15).

considerably, and a t no time closely resemble the adult pattern, which is not greatly different from t h a t of pure DHA (B15). The absorption curves were obtained using a modification of the Zimmermann reaction specially selected to reduce interference from nonspecific chromogen, but even so, this specificity of measurement would appear to be very low. An indication of the specificity has been obtained by using the Allen correction formula (A4) to measure the preponderance of specific Zimmermann chromogen over material giving linear light absorption in the region 480-560 mp. The values were calculated from actual, a , and corrected, b, extinction values a t 520 mp, the percentage of specific chromogen being given by b/a x 100. The findings for 17-0s given in Table 1 show that for infants up to 7 months old, as much as 95.5% of the absorption a t 520 mp can be due to nonspecific chromogen. The percentage is less for the assay of 17-OHCS (A7) (Table 1 ) )possibly owing t o the removal of some impurity by the oxidation and reduction procedures involved. However, one finding of 85.4% is shown. Bertrand et al. (B12) investigated the assay of plasma 17-OHCS as Porter-Silber chromogens by the method of Nelson and Samuels (N3) and found t h i ~ i r m i gT ~ I (

150

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

TABLE 1 P R O P O R T I O N O F SPECIFIC T O NONSPECIFIC ZIMMERMANN C H R O M O G E N S IN THE

URINEOF ADULTSAND INFANTS'"

Specific chromogen expressed as a percentage of the total

17-0s

17-OHCS

lJrine samples

Mean

Range

Mean

Adult, male and female Infant 1-3 days 6th day 4-7.5months 1-2.5 years

72.4 (39)& 43.8 (9) 31.1 (16) 33.9 (15) 53.6 (3)

33.7-99.3 32.8-47.6 19.4-40.4 4.5-56..5 47.2-57.8

71.1 (10) 56.6 (5) 69.4 (16) 58.5 (10) 82.8 (5)

Range

63.2-78.2 61 .349.7 39.8-1O6.Zc 14.6-97.2 76.6-89.2

Reproduced from Birchall and Mitchell ( B E ) . The number of replicate assays is given in parentheses. c The apparent anomaly of having higher corrected than uncorrected values resulting in a percentage greater than 100% may be explained by the nonspecific chromogens having nonlinear extinction/wavelength curves; this invalidates the application of the Allen correction formula. a

b

first 2 weeks after birth the conditions required for the application of the Allen correction formula are not fulfilled. Birchall and Mitchell (B15) have investigated various modifications of the techniques for the assay of 17-0s and 17-OHCS t o improve tlic specificity for newborn and infant urine (B15). Suggestions for improvement include the use of aqueous-ethanolic potassium hydroxide in the Zimmermann reagent, a color development period of 3 hours at O"C, and extraction of the color with ethylene dichloride. Even though the suggested technique gives improved specificity, the application must have limited value until more is known on the exact nature, origin, and importance of the substances measured. 3.4. ASSAYSOF INDIVIDUAL STEROIDS

Inevitably the study of steroid metabolism in infancy has had t o be delayed until suitable techniques were available. Crystallographic techniques initially used for isolating steroids from adult and animal material were not applicable since considerable amounts of tissue or urine would be required. The early solvent partition and adsorption techniques involving countercurrent distribution and column chromatography could have been applied, but since considerable labor would have been involved and no gross difference from the adult pattern of steroid metabolism was suspected in infants, little work was done until the availability of paper and thin-layer chromatography greatly improved the feasibility of study.

STEROID MEPABOLISM IK EARLY INFANCY

151

When paper chromatographic techniques designed for work on adult urine, such as those of Bush and Mahesh (B41), were applied to infant urine, considerable modifications were required mainly to deal with the increased amount of material which interfered both with chromatography and the specificity of color reactions (B18). Thin-layer silica-gel chromatography was found to be less susceptible to this interference, and with its use little preliminary purification is necessary (S9, SlO) . In the paper chromatographic technique developed for infant urine by Birchall et al. (B18) for the assay of blue tetrazolium and Zimmermann staining compounds, two sequential chromatographic separations were used for low and medium polarity steroids, and three for those of high polarity. Assay was by direct staining on the paper followed by densitometric scanning. Ulstrom et al. (U3) used paper chromatography with elution for the measurement of Porter-Silber reacting steroids.

A

B

t

I 2 3 4 5 6 7 FIG.2. A typical 20 X 20 em thin-layer chromatography plate for the assay of the following compounds in infant urine: A, pregnenolone; B, DHA; C , unknown; D, 21-OH-pregnenolone ; E, androstenediol ; F, 16-0x0-androstenediol; G, 1 6 ~ 0 H DHA ; H, 16p-OH-DHA ; I, 16~-OH-pregnenolone.The position of androstenetriol (not assaycd) is shonn at J. Samples of 1601-OH-DHA were run in positions f (1.25 pg), 5 (3.75 pg), and 7 (5.0 p g ) . Samples (2.5 fig) of all the steroids measured (except for the unknown and 16P-OH-DHA) were run in position 3. Duplicate extracts of urines collected from the same baby on days 1, 2, 3, 4, 5, and 6 of life were run in positions 2, 4, 6, 8, 9, and 10. The plate was sprayed with antimony trichloride then heated. From Shackleton and Mitchell (S9).

152

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

I

2

3

4

5

6

7

a

9

FIG.3. Part of a 20 X 20 cm thin-layer chromatography plate for the assay in urine of: A, androsterone; B, DHA; C , etiocholanolone ; D, ll-oxo-androsterone; E, ll-oxo-etiocholanolone ; F, llp-OH-androsterone ; G, llp-OH-etiocholanolone. Standard compounds have been run in positions 1, S, 6, and 8. Duplicate extracts from normal adult wines have been run in positions 2, 4,and 5 ; from a normal day-old infant in position 7 ; and from an infant with the adrenogenital syndrome in position 9. The plate was sprayed with Zimmermann reagent, then heated. From Shackleton e t al. (SIO).

For the separation of certain 3P-hydroxy-Aj steroids in infant urinc, Reynolds (R2, R3, R6) has used one or two paper chromatographic separations followed by elution and acetylation with further paper and silica-gel column chromatography. All these procedures are excessively long for clinical purposes, and techniques involving elution before colorimetry are not suited to the assay of more than a small number of steroids, as others tend to be too close together on the chromatogram for accurate detection and delineation using a parallel run of standards before elution. For multiple steroid assay, a more suitable technique has been developed employing thin-layer chromatography followed by direct staining of the chromatogram and quantitative densitometric scanning (S9, S10). This has the advantage of rapidity, and it is possible t o assess the

153

STEROID METABOLISM IN EARLY INFANCY

chromatograms visually if accurate quantitation is not necessary. The precision of quantitation by scanning is adequate for diagnostic and many clinical research purposes (the standard deviation averages _t8.1% for 17 types of steroid assayed), and steroids staining with antimony trichloride, blue tetrazolium, and Zimmermann reagents may be measured. Typical chromatograms stained with each reagent are shown in Figs. 2-4. Thin-layer chromatography may be followed by trimethylsilyl ether formation and subsequent gas-liquid chromatography ( C l l ) . T o elucidate the urinary steroid excretion pattern in various cases of the adrenogenital syndrome Bongiovanni (B28) used a long but excellent technique involving digitonin, and Girard separations, multiple chromatography on silica-gel and paper, and a variety of colorimetric techniques for detection and assay. A situation similar to that for urine applies for the assay of the 3phydroxy-A5 steroids in umbilical cord blood. Thc amount of blood

A

B

C

D E F G

I

2

3

4

5

6

7

a

9

FIG.4. Part of a 20 X 20 em thin-layer chromatography plate for the assay in urine of: B, THS; C, allo-THE; E, THE; F, allo-THF; G, T H F . Standard compounds have been run in positions 2, 4, 6, and 8. Cortisol is shown in position D and is measurable only if present in abnormally increased amount. The compounds shown in position A are mainly CIS steroids with an a-ketol group. Duplicate extracts from normal adult urines have been run in positions 1 and 3; from a case of adrenal carcinoma in position 5 ; from a normal infant in position 7; and from an infant wlth adrenogenital syndrome in position 9. The plate was sprayed with blue tetrazolium reagent, then heated. From Shackleton et d.(SIO).

154

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

Cholesterol

Preg nenolone DHA

17a -OH- pregnenolone

30, 20uDihydroxy-pregn- 5-ene

Androstenediol

2 I-OH- pregnenolone 16a-OH

- DHA 3p, l7a, 2 0 a - Tri hydroxy pregn- 5- ene

16 a-OH-pregnenolone

Androstenetriol

Plasma 2.5 ml enzyme hydrolyzed

Plasma 1.0 ml solvolyzed after enzyme

FIG.5. Thin-layer sihra-gel chromatography of an extract of umbilical cord plasma after enzyme hydrolysis and a further extract obtalned after solvolysis. The chromatogram was developed once in the system benzene-ethyl alcohol (95:5 v/v) and twice in the system cyclohexane-ethyl acetate (50:50 v/v) . Color development was by spraying with antimony trlchloride with subsequent heating.

normally available and the high concentration of the compounds being measured allows the use of a colorimetric assay, and again separation by paper or thin-layer chromatography with direct staining and densitometry (E2, S9, S l l , S20) would appear to be the technique of choice for routine clinical and quantitative research purposes. A typical separation is shown in Fig. 5 . Shackleton et al. (Sll) found t h a t fatty impurities present in plasma extracts could prevent chromatographic separation

STEROID METABOLISM IN EARLY INFANCY

155

on thin-layer plates. The impurities could not be removed after hydrolysis since their polarity was very similar to that of the steroids being separated. Removal was achieved by performing the separation on a Sephadex LH-20 column while the steroids were still in conjugated form. This technique has the added advantage of separating free, mono-, and diconjugated steroids if this is desired (S22a). Column chromatography on alumina has been successfully used by Col&s et al. (C12, C13) for the measurement of selected 3P-hydroxy-As steroids. Zander et al. (24) used paper chromatography, with elution, to detect and measure mainly progesterone and its metabolites and for the separation of steroids purely for identification purposes. Eberlein (E4) has used column, paper, thin-layer, and gas chromatography with Girard and digitonin separations. Several workers (e.g., B18, E2, S20) have corrected for losses during laboratory procedures by measuring the recovery of initially added isotopically labeled steroids. Gas chromatography has been applied by Horning et al. (HlO), and their technique, involving the formation of methoxime-trimethylsilyl ethers to stabilize the molecule, shows great promise for profile analysis if problems involving the quantitative formation of the double derivatives can be solved. For the assay of estriol, estrone, and estradiol in umbilical cord blood (A3, D8, 11, M3, R17) and in infant urine (DlO) methods designed for adult material have been used, and there appears to be no reason to doubt the validity of the results obtained. I n all cases the methods depended upon column chromatography for separation, with various modifications of the Kober reaction or Auorimetry, for assay. Special techniques have been used for the measurement of individual steroids. For blood progesterone Harbert et al. (H4) obtained an initial separation on an alumina column, but final purification and assay was achieved by further alumina chromatography after preparing the hisdinitrophenylhydrazone. The method of Short (Sl6) involving paper chromatography and assay by ultraviolet absorption, with Allen corrections for impurity, has been used by Greig et al. (G5) and Aitken et al. (A3). Cortisol has been measured in urine and blood by fluorimetry and isotope dilution (U3, U4), and in blood by double isotope dilution derivative techniques together with cortisone (H9) and also with cortisone, 11-deoxycortisol, and corticosterone (B38). A double isotope derivative technique has also been used for the assay of pregnenolone ((316). Attempts have been made to measure pregnanediol in infant urine (e.g., CS) by the method of Klopper et al. (K9), which is widely used for adults. However, because of the very low specificity of the final color developed, and the unknown effect of steroids and other compounds

156

FREDERICK L. MITCHELL AND CEDRIC H . L. SHACKLETON

present in infant and not in adult urine, for which the technique was designed, any results obtained must be treated with caution. The same difficulty is applicable to similar techniques used for the assay of pregnanetriol (B29, H5, L9, S28). Cortisol and corticosterone production rates have been measured by isotope dilution by means of techniques developed for adults (B12, KI, L17). There appears t o be no reason why these should not be satisfactory.

4.

The Influence of Steroid Metabolic Pathways Used in Ufero'

The fetus can synthesize the steroid nucleus (B22, B23, R9, S24), and in this respect the fetal zone of the adrenal cortex in utero is very active. Its production is considerable and owing to a low activity of the enzyme 3p-HSD (B23, C8, G2, G3, V6), is largely composed of 3p-hydroxy-A5 steroids. There is evidence that 3P-HSD is a t least to some extent substrate specific ( B l , V5), and histochemical findings have shown that 3pHSD enzymes specific for certain substrates may be present and active in the fetal zonc of the adrenal gland ( C l ) . The level of some of the 3P-hydroxy-A5 steroids in fetal blood is higher than that of any other known steroid (cholesterol apart) in human blood a t any time. Simmer and co-workers (S20) have indicated that about 75 mg of DHA are produced every 24 hours in late pregnancy and possibly even more 16aOH-DHA (D9, M16). Certain of the steroids formed, notably DHA, are used by the placenta for estrogen production, but the role of many others remains to be elucidated. The distribution, between the placenta and fetus, of the activity of some of the major enzymes involved in steroid synthesis is shown in Table 2. After birth the lack of 3p-HSD, the aromatizing enzymes, and sulfotransferases is not compensated for by the placenta, with the result that (a) only relatively small quantities of 3-0x0-A4 steroids are formed, but despite this, cortisol production is adequate (see Section 7.3) ; (b) thc newborn infant cannot convert its large supply of estrogen precursors to estrogen; and (c) it cannot desulfurylate its steroids, which hare been sulfated by the considerable and widely distributed activity of sulfotransferases (B24, W3). The mechanism of control of stcroid production by the fetus is not understood ; corticotropin stimulation in the newborn acts preferentially on cortisol production (see Section 12) and presumably if acting alone in utero, would not stimulate the production of large quantities of 3/3hydroxy-A5 steroids without producing a considerable amount of cortisol. 'For reviews, see Diczfalusy et al. (D9) and Mitchell (Ml6).

157

STEROID METABOLISM IN EARLY INFANCY

TABLE 2 THE PLACENTA A N D FETUS

ENZYME ACTIVITIESIN Enzyme system

Placenta

Fetus

3p-HSD Aromatization Sulfatases Sulfotransferases (aryl and 3p) Hydroxylases (llp, 1601, 1701,21) 17,20-Desmolase Steroid synthesis from acetate

High High High Low Low Low Low

Traces

+"

Low High High High High

a Only GI, 3-Ox0-a~steroids are aromatized. It is presumed that the lack of 38-HSD prevents the aromatization of 3p-hydroxy-A5 steroids (M2).

The adrenal gland is relatively 10-20 times larger than that in the adult and is mainly composed of the fetal zone. After birth involution commences and by day 28 of life the size has decreased by between fourfifths and nine-tenths (P13, T l ) . It would not be expected that such a large amount of highly productive tissue would cease activity immediately after birth, and in fact production of steroids of the pattern found in the fetus does not cease until after the sixth month of life (R3, R6, Fig. 7). The concentration of the 3P-hydroxy-A5 steroids in urine (Fig. 7) and blood (E4) and of the 17-0s in blood (E4), appears roughly to parallel the involution of the fetal zone of the adrenal (E4). The fetus receives estrogen and progesterone in considerable quantity from the placenta; for example, Zander (21) has estimated that approximately the same amount of progesterone goes to the fetus as to the mother. It is not known whether this large dose of potent hormone is received b y the fetus by design and is required for essential purposes, or whether by an accident of nature the hormones are received in such large amounts as to necessitate the development of special processes for detoxication. Whatever the reason, steroid metabolism in the fetus is very different from that in the adult, and some of the differences persist up to 4 years after birth (B3). The metabolism of the CIS and C,, steroids and estrogens in the fetus will be considered separately, and a separate subsection on the steroid content of amniotic fluid has been included. 4.1. METABOLISM OF C,, STEROIDS AND ESTROGENS

4.1.l. S/3-Hydroxy-A5 Steroids The production, transport, and metabolism of the C,, steroids in the fetoplacental unit appears to be of very great importance. ColBs et al.

158

FREDERICK L. MIT CHE L L AND CEDRIC H. L . SHACKLETON

ThBLE 3

COMPAR~SON OF STEROID CONCENTRATIONS (pg/100 ml) IN PLASMAFROM MATERNAL PERIPHERAL VEIN (PV), UMBILICAL CORDARTERY(CA), A N D CORDVEIN (CV) AT DELIVERY

THE

PV ( d l 0 0 ml)

CA

cv

Steroid

( d l 0 0 ml)

(wg/100 ml)

Es trone Est radio1 Estriol Progesterone 17~1-OH-proges terone 20a-Dihydroproges terone 20pDihydroprogesterone Total DHA DHA sulfate Total 16a-OH-DHA ~GcI-OH-DHAsulfate

3 . 5 (17)"rh 1 . 3 (17)' 9 . 0 (17p 14 (62)b 5 . 4 (4)d 32 (12)e 100 (20)' 7 (12)" <10 (16)'

1 . 9 (16)" 0 . 4 (16)a 103 (16)a 62 (24)b 3.3c 2.7" 1.4c 81 (12)a 162 (20)g 147 (12)e 110 (16)'

2 . 8 (16p 0 . 6 (16p 102 (16)c 89 (37)b 0.6~ 1.W 0.3 68 (12)' 130 ( 2 0 ) ~ 114 (12)" 87 (16)'

Maner et al. (M3). Greig (G4). Zander (81). Short (S17). ColSs et al. (C13). f Easterling et al. (E2). 0 Simmer et al. (S20). The number of cases investigated is shown in parentheses. When results by more than one group of workers have been published only the latest are given.

(C13) (Table 3) have demonstrated a fall in the level of both DHA and 16a-OH-DHA, from umbilical cord arterial, to umbilical cord venous, to maternal blood, showing that their origin was in the fetus. That the source is actually the fetal adrenal gland has been indicated by markedly decreased concentrations in the blood of anencephalic fetuses and in 6 cases in which the mother received large doses of prednisolone or cortisol throughout pregnancy (Cl2, E2, S20). Hypoplasia or atrophy of the fetal adrenal in both conditions has been described (L2, O2), and hypoplasia was proved in several of the cases reported. The levels of sulfated DH A and 16a-OH-DHA in fetal blood have been found to vary inversely with the amount of corticoid given to the mother (S22). The aromatization of C,, 3P-hydr0~y-A.~steroids received by the placenta from the fetus is the major source of the estrogens in the later stages of pregnancy (M16). Their utilization by the placenta, together with estrogen production will be seen from the arteriovenous differences given in Tables 3 and 4.

I59

STEROID METABOLISM IN EARLY INFANCY

TABLE 4 3fl-HYDROXY-A5 STEROIDS

IN

ARTERIAL

AND

ArterialE (A) (rg/100 mU

Steroid Pregnenolone DHA 1701-OH-Pregnenolone 2 1-OH-Pregnenolone 16a-OH-DHA (+16-0x0-androstenediol) 168-OH-DHA 16a-OH-Pregnenolone Androstenediol (1701 178)

+

91 80 56 62 348

(30)d

(19) (25) (34) (164)

27 (22) 178 (89) 123 (70)

VENOUS UMBILICAL CORD Venousb (V) (Crg/100 ml) 64 63 42 56 277

(34)d

(21) (24) (31) (138)

21 (15) 141 (83) 112

PLASMAa

Significance A-V of differencec irg/100 ml) (%) 27 17 14 6 71

2.0 0.5 0.2

6 37 I1

3.0

NS 0.6

NS 0.2

a From Shackleton et al. (S12). Presented a t the International Symposium on Faetoplacental [‘nit, Milan, Italy, 1968. Average of 20 assays. c Since the difference is most likely to be nearer to a constant ratio than a constant difference, Stndent’s t test for significance has been applied to the logarithmic values of the results. NS = not significant. d Standard deviations are given in parentheses.

4.1.2. Testicular Formation Unlike the low activity of 3P-HSD found in the fetal zone of the adrenal, intense activity has been detected in the Leydig cells of the fetal testis. This rises to a maximum between weeks 12 and 16 (B31, C8, G2, J8, N5), thereafter declining. At the sanie time there is rapid growth in the interstitial tissue, which reaches a maximum amount never again attained during the further life of the human testis (h’5). During this period the testes actively produce compounds including androgens (J9) causing, in the male, regressive changes in the Mullerian ducts and masculinization of the external genitals. Without suitable androgenic stimulation a t this time every fetus would develop as a female and production of the A4 steroidal androgens requires 3P-HSD activity. At birth testicular activity has decreased considerably, not to be revived until puberty. 4.1.3. Estrogens The special type of estrogen catabolism present in infants is presurnably designed t o deal in utero with the large amounts of estrogens received by the fetus from the placenta. Barr et al. (B3) found t h a t in 11 infants aged 2-36 months only 14% of a dose of exogenous estriol could be recovered as such from the urine, compared with approximately 80%

160

FREDERICK I,. MITCHELL AND CEDRIC H. L. SHACKLETON

recovered in adults (B39). The adult type of estrogen metabolism was not fully achieved until the fourth year of life. The metabolism of estradiol also differs in the newborn from that later in life; extensive hydroxylation occurs, and this pattern of metabolism may be one of the mechanisms by which the fetus is protected against its high biological potency. It will be seen in Section 9 that during the first few days of life large amounts of estriol are excreted, but very little estrone and estradiol, and after the administration of estradiol to newborn infants and infants aged up to 10 months, only trace amounts of it and estrone are found in the urine (B3, D l l ) , contrasting with considerable quantities of estrone and estradiol recovered from the urine of children 2.5-9 years of age treated similarly (B3). Hagen e t al. (H2) administered estradiol 4-14C to infants, aged 42-43 days, with multiple malformations; they recovered less than 5% of the radioactivity in estrone and estradiol, 20% in estriol, 16a-OH-estrone, and 16-0x0-estradiol, and 16% in a highly polar product. That this type of metabolism obtains in the fetus in utero has been shown by the work of Gurpide e t al. (G6) and Schwers e t al. (S7), who reported the major product of the metabolism of estradiol by the fetus to be a tetrol, possibly made exclusively by the fetus. The compound is probably the same as the highly polar radioactive metabolite reported by Hagen e t al. (H2). It has been synthesized (F 2 ) , identified as estetrol (Z5), and shown to be a normal constituent of newborn and pregnancy urine and meconium (26). The fetus rapidly conjugates estrogen as early as the third month of gestation. The conjugation mechanisms are widely distributed ; sulfurylation predominates in the lungs and the liver, but estriol glucosiduronate has been shown to be a major conjugate in the intestine (D12). 4.2. METABOLISM OF C,, STEROIDS One of the first indications that, the steroid metabolism in the fetus differs from t h a t in the adult was the finding by Appleby and Norymberski (A6) and Schuller (S6), in 1957, that toward the end of pregnancy the urinary excretion of 21-deoxyketols rises to a level of up to 3 mg/24 hours above the normal excretion of about 0.16 mg/24 hours in nonpregnancy. Appleby and Norymberski (A6) also showed in 4 patients that after intrauterine death the excrction fell sharply to normal levels while the excretion of pregnanediol remained within the normal range for pregnancy. The adrenogenital syndrome is the only other known human condition in which the excretion is raised, and the cause is then due t o a relative deficiency of 11- and 21-hydroxylase Ieading to the production of such compounds as 17a-OH-progesterone instead of cortisol (Section 11).The findings would seem to indicate that the pIacenta is not

STEROID METABOLISM I N EARLY INFANCY

161

the source of the 21-deoxyketols, and the maternal adrenal would also appear t o be eliminated because administration of cortisol did not markedly decrease the amount excreted as it does in the adrenogenital syndrome. The fetus was therefore suspected of being the source, and though this was contraindicated by the finding of low levels of 21deoxyketols in newborn infant urine, the measurements were not considered to be reliable since the technique of assay was not entirely satisfactory when applied to infant urine. It is probable t h a t the increased amounts of pregnanetriol (H5), corticosterone, and 17-deoxycorticosteroids (E8, N6) and of Pettenkoffer chromogens (N2), also found in maternal urine, emanate from the fetoplacental unit if not from the fetus itself (M16). The fetus contains a large amount of pregnenolone, it metabolizes considerable quantities of progesterone, and its metabolism of cortisol differs from the adult. Each of these aspects will be dealt with in turn. 4.2.1. Pregnenolone

It is probable t h a t the fetus is not significantly implicated in the synthesis of progesterone in pregnancy since various workers have shown that urinary pregnanediol ( a characteristic metabolite of progesterone excretion by the mother) can remain high or actually rise after the death or removal of a fetus, with the placenta remaining intact (A5, A6, C3, K8; for review, see MIS). There is also little difference from normal in the pregnanediol excretion of women with anencephalic monsters ( F 6 ) where the fetal adrenal zone is absent. Hellig et al. (H6) have concluded from experiments with radioactive cholesterol administered to the mother that progesterone in late pregnancy is made almost entirely from maternal cholesterol. Pion et ul. (P12) have shown, however, t h a t approximately 15% of pregnenolone perfused through mid-term placentas in sihi was converted to progesterone, and a similar high conversion has been obtained from pregnenolone sulfate perfused through term placentas (PI, P11). There is a considerable quantity of pregnenolone in fetal blood; as the arterial and venous levels are 91 pg/lOO ml, arterial; 64 pg/IOO ml, venous (Table 4; see also Cl6) , it seems likely t h a t fetal pregnenolone is used by the placenta possibly for progesterone production, though the major precursor is almost certainly cholesterol (JI, M l 6 ) . Large quantities of pregnenolone are presumably formed in the placenta during the synthesis of progesterone. If any of this is secreted into the fetoplacental circulation i t would be expected to be present in the free form since sulfotransferase activity in the placenta is low (Table 2) ; however very little, if any, is free (Sl2). The fetal adrenal in vivo (week 13 of gestation) can convert pregneno-

162

FREDERICK L. MITCHELL AND CEDRIC H . L. SHACKLETON

lone into DHA and 17a-OH-pregnenolone, and the liver can form the 20a- and 16a-hydroxy derivatives (12). Using in vitro incubation techniques with newborn adrenal homogenates, Villee and Loring (V3) have obtained conversions into a variety of steroids, notably cortisol, corticosterone, and androstenedione. The use for the large quantity of pregnenolone produced by the fetus in utero is therefore uncertain, but production continues into infancy and is then reflected in the considerable urinary excretion of 16a-OH-pregnenolone (see Tables 13 and 14, and Fig. 6 ) . 4.2.2. Progesterone

It is evident from the differing concentrations of progesterone in the umbilical cord artery and vein (Table 3) that the fetus metabolizes some 30 pg of progesterone in every 100 ml of plasma passing through its circulation. The peripheral tissues of the fetus show a remarkable ability to metabolize progesterone to a variety of products (S24). Zander and Solth (23) pointed out in 1953 t h a t progesterone metabolism in the fetus differs markedly froin that in the adult, where the major urinary metabolite is pregnanediol. If an adult type of metabolism obtained, it would be expected t h a t the large quantity of progesterone received by the fetus in utero would be cleared in the first few days of infancy, largely as pregnanediol; however, little (L4, R14) or no (C6, 23) pregnanediol has been found in infant urine although it has been isolated from this source ( N l ) . Even after a loading dose of 20 iiig of progesterone, Zander and Solth (23) could not detect pregnanediol in the urine of newborn infants. From work on the metabolism of cortisol, it is evident that the infant is deficient in A-ring reducing enzymes (Section 7.3), and this may explain the absence or partial absence of pregnanediol. It is perhaps fortunate t h a t pregnanediol is not present in quantity in early life because, both free and as the glucosiduronate, it has been shown to be a potent inhibitor of the conjugation of bilirubin with glucuronic acid (B14). This is, in infants, a rate-limiting step in the excretion of bilirubin. Major metabolites of progesterone are indicated from the arteriovenous differences shown in Table 3 as being 17a-OH-progesterone, 20a-dihydroprogesterone, and 20P-dihydroprogesteronr. Zander ( Z l ) has provided additional proof by injecting radioactive progesterone into the umbilical cord vein during a laparotomy operation for sterilization and subsequently identifying, among the steroids in the fetal tissue, radioactivity in the three steroids previously mentioned and also in 6P-OH-progesterone. It has been proposed by Wiener and Allen (W4) t h a t the forma-

STEROID METABOLISM I N EARLY INFANCY

163

tion of 20cr-dihydroprogesterone may operate a feedback control mechanism for the production of progesterone from pregnenolone by the placenta, since the enzyme system responsible for the conversion is noncompetitively inhibited by 20a-dihydroprogesterone, the Ki of the inhibition being of the same order of magnitude as is the placental inhibitor concentration. I n considering the catabolism of progesterone by the fetoplacental unit, however, the action of the placenta itself cannot be ignored since Kitchin et al. (K2) have shown that 20a-dihydroprogesterone is a major product when placentas are perfused in situ with labeled progesterone. The adrenal gland is probably one of the major sites of progesterone metabolism in the fetus since autoradiographic studies on previable fetuses perfused with radioactive progesterone have shown a striking accumulation of radioactivity in the fetal adrenals (B8), and a considerable amount of in vitro incubation and perfusion work done on fetal adrenal tissue has shown t ha t i t is capable of converting progesterone into a variety of steroidal products (for review, see M16) of which the following are worthy of note : cortisol, corticosterone, 16a-OH-progesterone, 20a-dihydroprogesterone, 16a-OH-cortisol, deoxy17~~-OH-progesterone, corticosterone, androstenedione, llp-OH-androstenedione, and pregnanoloneLT h at fetal liver can in fact form pregnanediol was shown by Solomon and co-workers (S25), who perfused fetuses of gestational age 17-21 weeks with progesterone-4-14C and isolated in the form of pregnanediol in the liver approximately 25% of the activity perfused; 20a-dihydroprogesterone and pregnanolone in the liver were also radioactive. It thus appears t h a t the adrenal gland of the fetus can utilize circulating progesterone for the production of corticoids while the function of the liver is to produce reduced metabolites. Zander (22) has shown by both in vitro and in vivo experiments that much of the 20a-dihydroprogesterone produced by the fetus from placental progesterone may be reconverted to progesterone by the placenta, and thus a placental-fetal circulation is possible of some of the progesterone produced by the placenta. During the first few days of life, therefore, infant urine will contain a wide variety of steroidal metabolic products of the progesterone which the fetus received from the placenta; these should be cleared by the sixth day. The high concentration of progesterone in fetal blood may be responsible for the low activity of SP-HSD in the fetus and in the infant. Progesterone occupies a key position in the normal chain of steroid synthesis (Fig. S ) , and i t might be expected from the known effects of increases in product concentration upon the activities of other enzymes that inhibition

164

FREDERICK L . MIT CHE L L AND CEDRIC H. L. SHACKLETOX

of the conversion of pregnenolone into progesterone would occur. Confirmatory evidence for this suppressive action of progesterone has recently been provided by D. B. Villee (V4) using adrenal cultures. If this explanation is to be accepted, however, overriding influences must be present in the placenta and in the fetal testes during weeks 12-15 of gestation (Section 4.1.2.), where intense activity of 3p-HSD is present. 4.2.3. Cortisol and Corticosterone Cortisol can cross the placental barrier (L13, M l l , M14), but i t is not known to what extent, if any, the fetus draws upon a maternal supply. Leyssac (L14), after infusion of cortisol into the mother, demonstrated an average gradient of 203: 111:79 pg/loO ml from maternal t o umbilical vein t o fetal heart plasma. Likewise from experiments carried out by Migeon e t al. ( M l l ) , it is evident t h a t cortisol or its metabolites can be transferred from fetus to mother, probably via the placenta. It would appear from the levels of cortisol in maternal plasma (52.4 pg/lOo ml) and umbilical cord plasma (7.8 pg/lOO ml) shown in Table 5, that a gradient exists from maternal to fetal blood. The free cortisol levels, however, probably do not show such a wide difference since most of the cortisol in maternal blood is bound to the cortisol-binding protein transcortin, the concentration of which, through the effect of estrogen ( W l ) TABLE 5 hIEAN BLOODSTEROID CONCENTRATIONS I N NORMAL YOUNG WOMEN, I N BLOOD,A N D I N INFANTS DURING EXCHANGE TRANSFUSION"

Sample source Young womenb (aged 18-25 years) Beginning of transfusionb (approx. 3 hours after birth) End of transfusionb Donor plasmab Cord bloo& Cord plasmad Maternal bloodc Maternal plasmad

Cortisol (pg/100 ml)

11-DeoxyCortiCortisone cortisol costerone (pg/100 ml) (pg/100 mi) (pg/100 ml)

7 4 (39p

1.8 (23)

1.3 (17)

2 . 5 (19)

7 2 (3)

4 . 7 (5)

1.9 (5)

3 . 2 (5)

4.3 ( 5 ) 1.5 (5) 13.4(9) 13.6 (12) 4 . 4 (9) 5 . 2 (12)

1.4 ( 6 ) 0.8 (5) -

2 . 5 (5) 1.4 (5)

14 6 9 7 47

5 (5) 0 (5) 2 (9) 8 (12)

4 (9) 52 4 (12)

Reproduced from Mitchell (M16). Data from BULISet al. (B43). Data from Bro-Rasmussen et al. (B37). d Data from Hillman and Giroud (H9). The number of cases investigated is shown in parentheses. a

b

8

C OR D

-

-

STEROID METABOLISM I N EARLY INFANCY

165

rises two to three times during pregnancy, and over 99% of the plasma cortisol is then protein bound (M7). The level of plasma cortisol in the fetus a t term is almost the same as in the nonpregnant woman, and transcortin levels are low (D2, D7, S2), indicating t h a t for some reason the effect of the high concentration of estrogen is slight or absent. Sandberg and Slaunwhite (S2) have shown that a high percentage of cortisol in cord blood is unbound (36% compared with 16% in normal adults) and the binding is probably due to albumin, not to transcortin (M15, 52). Since protein-bound cortisol is probabl? inactive (53, S23, W l ) , the activity of cortisol in the infant a t birth must be relatively high and, contrary to the position which might be inferred from measurements of total blood cortisol, the level of free “active” cortisol in fetal blood is probably higher than in maternal. For this reason Mills (M15) has suggested t h a t the mother steals cortisol from the fetus during most of fetal life, and it is not until the maternal adrenal cortex is strongly stimulated by the stress of birth that the high concentration of transcortin is saturated and the hormone can escape into the fetal circulation. If this does occur, the levels of cortisol found in cord blood may not represent the normal concentration in the fetoplacental circulation (Section 5.3; Table 5). From Table 5 it will be seen that the cortisol: cortisone ratio of 0.7: 1 in mixed arterial and venous umbilical cord blood is a complete reversal of the ratio of 11:1 in maternal blood. This change could be produced by the placenta because blood “milked” from a cord a t delivery is predominantly venous. James (53) has measured the two steroids in samples of pooled venous and pooled arterial plasma; although he found the same levels (7 ,ug/100 ml) in each for cortisol, the level of cortisone was lower in the artery (10.5 pg/loO ml) than in the vein (14.0 pg/100 ml), indicating a supply of cortisone from the placenta. The considerable activity of llp-dehydrogenase, possibly with a low substrate specificity, which has been shown to be present in placental tissue (03, P2) might be the cause of the normal cortisol :cortisone equilibrium shifting toward cortisone. The fetus a t mid-pregnancy does not appear to be capable of converting cortisone to cortisol ( P 3 ) . The placenta may thus provide the fetus with a supply of “inactive” cortisone which can act as a reserve supply of cortisol for use after birth. It is also possible t h a t cortisone is not completely “inactive” and the increased amount in fetal blood may serve some special function. It has been shown t h a t the fetal adrenal can convert progesterone and pregnenolone into cortisol in substantial yield (Sections 4.2.1 and 4.2.2), and, notably, Solomon (S25), after perfusion of fetuses with progesteronc-4-*‘CC,has isolated radioactive corticosteronc and cortisol from the

166

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

adrenal tissue, indicating that circulating progesterone can be thus converted. The fetus, however, does also possess an independent means of production as indicated by Sachazkaja ( S l ) , who showed t h a t fetal adrenal tissue from fetuses 9-11 weeks old could produce cortisol from endogenous substrates during in vitro incubation in the presence of corticotropin. To achieve the synthesis of cortisol, and possibly to a lesser extent corticosterone, it seems probable t h a t the fetus and newborn infant can utilize a pathway which bypasses progesterone. The low activity of 3pHSD in the fetus and in infancy which has already been mentioned in Section 4 would be expected t o cause an inhibition in the normal pathway of cortisol synthesis a t the stage of conversion of pregnenolonc into progesterone. Increased use of secondary pathways for corticoid synthesis via 21-OH-pregnenolone or 17a-OH-pregnenolone would then occur, with 3P-HSD (possibly substrate specific) acting a t a later stage subsequent to hydroxylation a t C-17, C-21, and C-11 for the synthesis of cortisol or corticosterone. The possibility of these pathways has been indicated by several workers (for review, see D15), and it seems likely t h a t i t is the major mode of synthesis of corticosteroids, a t least in guinea pigs, swine, and rats ( Y l ) . TWO alternative pathways for androgen synthesis have been known for some years, one proceeding via pregnenolone, progesterone, 17a-OHprogesterone, and androstenedione and the other via pregnenolone, 17aOH-pregnenolone, DHA, and androstenedione, bypassing progesterone and 17a-OH-progesterone (for review, see D15). Evidence that the A5 pathway exists in man for corticosteroid synthesis, even though it may not normally be used to any extent, has been provided by the isolation of small quantities of 21-OH-pregnenolone from human urine after corticotropin stimulation (D13, P4) and of polyhydroxy A5-pregnene compounds shown to be present in the urine of infants with a deficiency of 3P-HSD. That the pathway may be of major importance in the fetus and newborn is indicated by the identification and measurement of 21OH-pregnenolone in moderate quantity in the urine of newborn infants (B16, C7) (Tables 13 and 14; Section 8) , and Eberlein (E4) has partially identified in cord blood 17a,21-dihydroxypregnenolone and 11/3,17a,21trihydroxypregnenolone, two steroids which might be expected to be on the A5 synthetic pathway to cortisol. Finally, in v i t r o work involving the incubation of newborn adrenal homogenates with pregnenolone and progesterone has shown t h a t cortisol and androstenedione may be produced from pregnenolone without necessarily involving progesterone as an intermediate (V4). I n a similar experiment using adrenal slices from a premature infant, Klein and Giroud (K5) showed a preferential path-

STEROID METABOLISM I N EARLY INFAKCY

167

way for cortisol synthesis via 17a-OH-pregnenolone; 21-OH-pregnenolone, and 17a,21-dihydroxypregnenolone were also produced but in low yields. Pasqualini et al. (P7) have recently shown that the adrenals of mid-term human fetuses can convert 17a,21-dihydroxypregnenoloneinto cortisol. The control of cortisol production in the fetus is not fully understood, but it must be somewhat complex for the following reasons: ( a ) The fetus could be supplied by the transfer of free cortisol across the placenta from the mother and presumably need not have a supply of its own. (b) The fetal adrenal could synthesize cortisol from progesterone circulating in high concentration, but if this pathway was used exclusively, an alternative would have to be available a t birth. (c) The low activity of 3p-HSD in the fetal adrenal would cause a block if the normal mode of corticoid synthesis in the adult were wed. The difficulty raised under (a) is probably overcome by there being a constant gradient of free cortisol from fetus to mother and conscqucntly, as previously explained, a steady loss to, instead of a gain from, the mother. The lack of SP-HSD may be circumvented by the use of a synthetic pathway, but there appears t o be no clear explanation why under (b) there is not a considerable production (out of corticotropin control) of cortisol from circulatory progesterone. Work done on the measurement of corticotropin in the fetus has so far not provided direct information on its use for steroid production control, but adrenals, normal for the newborn in size and morphology, have been produced in anencephalic monsters by corticotropin stimulation (L2), and the administration of corticoids to pregnant women before and during delivery leads to a decrease of D H A and 16a-OH-DHA in cord plasma (S22). This circumstantial evidence indicates t h a t corticotropin may be involved as a controlling factor. This question is further discussed in Section 12. corticosterone production appears to be relatively greater in the newborn than in the adult (Section 7.3). The reason for this is not known, but several mechanisms are possible. Klein and Giroud (K4) have shown that adrenal slices from newborn infants can actively form the (2-21 ester sulfates of corticosterone and several other A4-3,20-dionepregnenes. Further hydroxylation was not then observed, and it is possible t h a t in V ~ V Othe activity of this specific sulfokinase limits synthesis. A further explanation for a greater synthesis of corticosterone may be in a preferential hydroxylatjon of pregnenolone a t C-21 over hydroxylation a t C-17, or hydroxylation a t C-17 may be so rapidly followed by the action of 17,2O-desmolase producing the large quantities of DHA made in utero, that a relatively smaller proportion of 17a-OH-pregnenolone remains for subsequent hydroxylation a t C-11 and C-21.

168

FREDERICK L. MIT CHE L L AND CEDRIC H. L. SHACKLETON

TABLE 6

STEROIDCONTENT OF NORMAL AMNIOTIC FLUIDIN LATEPREGNAKCY Steroid Cortisol Cortisone 17-0s 17-OHCS Pregnanetriol 6p-OH-cortisol 6p-OH-cortol 6p-OH-cortolone Estrone Estradiol Estriol 21-OH-pregiienolonec Androstenediol (mainly 1701)~ 16-0x0-androstenediop 16a-OH-DHh 16p-OH-DHAc 16cr-OH-pregnenolonec a

Concentration ( d l 0 0 mC

References

2 . 6 k 0.32a 1 . 3 A 0.14y (10) 3.8 (4) 2.4 3.1 (4) 4.3 (6) 0.8(6) 3 . 1 (6) 0 . 4 4 (10) 0 . 3 1 (10) 79 (10) 19.8k 16.5d (28) 10.9 k 27.1d(28) 18.9 k 2 0 . 9 (28) 4 0 . 8 f 54.0d (28) 19.3 f 20.0d(28) 12.0 f 11.2d (28)

Standard error of the mean. The number of cases studied is given in parentheses. Results obtained in the author’s laboratory. Standard deviation.

4.3. STEROID CONTENT OF AMNIOTIC FLUID

The fetal kidney can produce urine as early as the fourth month of gestation (K3) and probably excretes freely from a t least week 20 of pregnancy ( J 6 ) . It therefore seems reasonable to conclude t h a t the steroid content of the liquor reflects fetal production. The concentration of various steioids assayed in late pregnancy is given in Table 6 . Dynamic transfer is continually taking place between the amniotic fluid, the fetus, and the mother ( M l l ) , and it is thus difficult to estimate excretion rates into the fluid from any assay of its contents. Lambert and Pennington (Ll ) report abnormally lorn7 levels of 6,8-OH-cortisoll Gp-OH-cortol, and 6P-OH-cortolone in the amniotic fluid of anencephalic fetuses. I n similar cases low levels of pregnanetriol and 17-0s have also been found ( 5 7 ) , and high levels of both 17-0s and pregnanetriol have been reported a t 40 weeks’ gestation in a case of adrenogenital syndrome in the fetus (57). There is no difference in the cortisol or cortisone content when the mother has diabetes (B2), but 6P-OH-cortisol is reduced ( L l ) . Estriol glucosiduronate appears to be selectively retained in the amniotic fluid, probably because glucosiduronates are poorly transferred

STEROID METABOLISM IN EARLY INFANCY

1200-1

169

Adult male urine

-

E

0

P \ Is,

I

Umbilical cord plasma

FIG.6. The content of the major 3P-hydroxy-A' steroids in adult male urine, infant urine, term amniotic fluid, and plasma obtained from the umbilical cord a t birth.

and hydrolyzed by the membranes and the placenta [concentration 32 pg/lOo ml compared with 20 pg/lOO ml for estriol-free, sulfated, and as the sulfoglucosiduronate (L12) 1. The 3P-hydroxy-As steroid content of amniotic fluid (S12) is compared with that in adult and infant urine and umbilical cord blood in Fig. 6. Schindler and co-workers have assayed several 3P-hydroxy-As steroids and others in amniotic fluid for both normal and abnormal pregnancies (S4, S5).

4.4. CONCLUSIONS Many interrelated mechanisms exist for the synthesis and metabolism of steroids in the fetoplacental unit, and after birth t h a t part of the metabolism which takes place in the fetus may be expected to continue in the newborn infant for a varying period of time. These effects may be summarized as follow,s: a. The concentration of progesterone in the blood going from the

170

FREDERICK L. MITCHELL AND CEDRIC H. L . SHACKLETON

placenta to the fetus is some 5-6 times, and of estriol 10 times, that found in the maternal circulation. It is not known whether this high concentration of potent hormone is essential for the well-being of the fetus or whether i t is received by accident, but the fetus does handle progesterone and estrogen differently from the adult and this difference persists (for estrogen a t least) in infants up to the fourth year of life. b. The fetus has a low activity of 3p-HSD and a high activity of 16ahydroxylase; this situation appears to persist almost up to the sixth month of infancy, and consequently the major quantity of steroids excreted during this period have the 3p-hydroxy-as configuration and are 16-hydroxylated. c. Quantitatively the major function of the adrenal gland in the fetus is t o produce estrogen (and possibly progesterone) precursors which are subsequently converted to the active hormones by the placenta. The production of these compounds continues in decreasing amount until the sixth month of infant life, and initially their 16-hydroxylated metabolites dominate the steroid pattern in infant urine. d. It seems probable t h a t cortisol is synthesized in the fetus and in early infancy by a pathway involving the hydroxylation of As steroids, and i t is possible t h a t this mode of synthesis remains dominant in infancy until adult levels of activity of 3P-HSD are assumed. 5. Steroid

Assays on Umbilical Cord Blood

By studying the steroid content of arterial and venous umbilical cord blood, it is possible to obtain a considerable amount of information on steroid metabolism in both the fetus and the placenta. Quantitative assays can indicate uptake or production by either body, and as the blood is relatively easily obtained in quantity it also provides a ready source of material from which new compounds can be extracted and identified. I n abnormal conditions a study of its steroid contcnt can provide early information on steroid production by the infant adrenal a t birth. The groups of steroids present will be discusscd under the following headings : 3P-hydroxy-ns steroids, progesteroiic and progesterone inetabolites, cortisol and its metabolites, testosterone and the 17-0S1 and the estrogens.

5.1.

3p-HYDROXY-A5

STEROIDS

The major group of steroids found in cord blood has the 3p-hydroxy-A5 configuration. The steroids present are largely sulfate esters (C12, C13, C16, E2, E4, M12, S5, S12, S20). I n addition to those listed in Table 4 and shown on the thin-layer chromatogram in Fig. 5, the following have been characterized or tentatively identified : androstenetriol (S12) ; 3,8,17~~,20~~-trihydroxypregn-5-ene (E4) ; 3P,20a-dihydroxypregn-5-ene

171

STEROID METABOLISM I N EARLY INFANCY

(E4) ; 3P,ZOp-dihydroxypregn-5-ene (E4) ; 3p,20~,21 -trihydroxypregn-5ene (E4) ; 3b,17a,21-trihydroxypregn-5-en-20-one(E4) ; 3P,ll,8,17a,21tetrahydroxypregn-5-en-20-one (E4) ; 3P,17a,20~,21-tetrahydroxypregn5-ene (E4). From the differences in the arteriovenous concentration given in Table 4, it will be seen t ha t for all the steroids listed there is a net production by the fetus and uptake by the placenta. The utilization by the placenta of certain of the steroids mentioned has been discussed in Sections 4.1 and 4.2. Fetal D H A sulfate and 16a-OH-DHA sulfate, for instance, provide the major source of estrogen in pregnancy, the aromatization taking place in the placenta. Pregnenolone may be used in the placental production of progesterone, and 16a-OH-pregnenolone may have a role in the synthesis of 16a-OH-DHA, since a desmolase enzyme system for 16a-OHpregnenolone has recently been demonstrated (S14). As in infant urine, androstenediol is present mainly as the 17a epimer (Section 8). It will be noticed from Table 4 that the lowest arteriovenous concentration differences are shown for 21-OH-pregnenolone, lGp-OH-DHA, and androstenediol (mainly 1 7 ~ ~It) .is possible that their low uptake by the placenta is due to their being diconjugated (Table 15), and i t is possible t h a t though the placenta can readily hydrolyze 3p-sulfates i t may have difficulty hydrolyzing (P6), and therefore removing from circulation, steroids sulfated a t other positions. Anencephalic monsters have low umbilical cord plasma levels of DHA sulfate and 16a-OH-DHA sulfate (Table 7). This is associated with an abnormally low excretion of estrogen by the mother (Cl2, F5, F7, M9). 5.2.

PROGESTERONE AND ITSMETABOLITES

It will be seen from the arteriovenous differences given in Table 3 that a considerable amount of progesterone produced by the placenta is TABLE 7 UMBIIJCAI,CORDPLASMA LEVELSOF DHA SULFATE A N D 1601-OH-DHA SULFATE IN ANENCEPHALIC FETUSES Reference East,erling el al. (E2)

C:olBs and Heinrichs (C12) a

DHA sulfate (pg/100 ml mixed plasma)

16a-Hydroxy-DHA sulfate (pg/lOO ml mixed plasma)

0-25 (5)a

Q-118 (4)

Arterial plasma

Venous plasma

Arterial plasma

Venous plasma

6 . 7 (3)

6 . 2 (3)

1 4 . 4 (3)

8 . 1 (3)

The number of cases studied is given in parentheses.

172

FREDERICK L. MITCHELL AND CEDRIC H. L. SH A C K LETO S

metabolized by the fetus. The concentrations of oldy three of the metabolites are given in the table, but the overall metabolism has been discussed in Section 4.2.2. 5.3. CORTISOL A N D ITSMETABOLITES

Cortisol and cortisone have been identified and assayed in cord blood by several groups (B37, H9, 53, U4, Tablc 5), and the implications of the results have been discussed in Section 4.2.3. Little information is available on the concentration in abnormal infants. Cathro and Coyle (C4) found a low level of cortisol in an infant with congenital adrenal hypoplasia (3.3 pg/lOO ml) , and James (J3) measured the concentration of cortisol and cortisone in the cord blood of two infants born to mothers with acute adrenal insufficiency. The compounds must have been formed almost entirely by the fetus since negligible quantities of cortisol were found in the maternal blood. Thc low levels of cortisol (2.5 Fg/lOO ml) and cortisone (4 pg/lOO ml) found for these fetuses is difficult t o explain unless they represent the true levels of the steroids in the fetoplacental circulation, which are normally "boosted" by transfer from the mother during the stress of childbirth (see Section 4.2.3). 5.4. TESTOSTEROKE AND THE 17-0x0 STEROIDS

Early in pregnancy the testes of male fetuses produce androgen (see Section 4.1.2). Concentrations of testosterone and androstenedione found in cord plasma by Rivarola et al. (RlO) and Mizuno e t al. (MlS) using TABLE 8 CONCENTRATION OF TESTOSTERONE A N D ASDROSTENEDIOXE I N UMBILICAL CORD PLASMA ~

Umbilical vein

~

Umbilical artery -

I3 eference Rivarola et ul. (R10) Males Females Mizuno et a / . (M18) Males Females Beck (B7) Males Females

Aridrdstenedioiie Testosterone A4ndrostenedioiie Testosterone (mpg/100 ml) (mpg/100 ml) (mpg/100 ml) (mpg/100mi)

138 k 68" (5)b 53 k 30 ( 3 ) 112 k 49 (4) 39 f 14 (4) 12,5 k 20 (5) 86 t 8 (-5)

54 k 20 (5) 7% k 47 (.5)

-

940 (4) 1080 (4)

-

Standard deviation. Number of infants studied is given in parentheses.

-

-

-

-

113 k 16 (5) 43 12 (Ti) 91 t 2.5 (.i) 57 & 22 (5) -

-

8540 (4) 2830 (4)

173

STEROID METABOLISM I N EARLY INFANCY

TABLE 9 PLASMA TESTOSTERONE AND ANDROSTENEDIONE IN INFANTS DURING EXCHANGE TRANSFUSION FOR RHESUSINCOMPATIBILITY' Androstenedione (mpg/100 ml) Males 438 (l)b 236 (2) 32 ( 3 ) 306 (4) .54 (4) 20 ( 5 ) 54c (5) Mean values: 163 a

Females Iindetectable ( < 10) (2) Undetectable ( < 10) (2) Undetectable ( < 10) (4) (a 34 < 16

Testosterone (mwg/100 ml) Males

Females

106 (1)

58 (2) 36 (3) 326 (4) 117 (4) Undetectable (5) 5gC ( 5 ) 100

Mizuno et nl. (M18). Age of infants, in days, is given in parentheses. This figure was obtained from a pool of specimens from 3 healthy male babies.

isotope dilution techniques are given in Table 8. The results obtained by the two groups of workers agree well, and for testosterone, no significant difference was found between male and female infants, though androstenedione was present in greater amount in the plasma from male infants and Mizuno et al. (M18) have shown t h a t this difference is statistically significant. The differences in the concentration of androstenedione and testosterone in the vein and artery were not significant, but it could be that a more extensive study may show a significantly higher concentration of testosterone in the vein than in the artery. This would then indicate its synthesis in the placenta. The findings of Beck (B7) are also given in Table 8 and the discrepancy between his results and those of the other two groups of workers is possibly due to the nonspecificity of the technique used. Mizuno et al. (M18) have determined the concentration of testosterone and androstenedione in plasma from infants undergoing exchange transfusion for rhesus incompatibility. From their results (Table 9 ) , i t is evident t h a t during the first few days of life, male infants have higher levels of testosterone and androstenedione than female infants. However, since these babies were not normal, it is possible that their androgen metabolism differed from t h a t of normal infants. The concentration of testosterone in maternal peripheral plasma increases during pregnancy (M18, RlO), this increase being associated with an increased proteinbinding of testosterone (RlO). The level in maternal plasma is not related to the sex of the fetus (MlS, R10). Adrenosterone [0.2 pg/lOO ml (24) ] and 1lp-OH-androstenedionc [0.3

174

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

pg/lOO ml (Z4)] have also been detected in mixed arterial and venous cord blood, and androsterone [0.2 pg/lOO ml (E4)] and etiocholanolone [1.2 pg/lOO ml (E4)] have been tentatively identified in infants with congenital adrenal hyperplasia. 5.5. ESTROGENS

It will be seen from Table 3 that lower levels of estrone and estradiol, but higher levels of estriol, are found in cord blood compared with maternal blood. Estrone and estradiol can be formed from DHA-sulf ate reaching the placenta from the fetus (B25) or the mother (B4, B25, S18, W2). These steroids are then secreted by the placenta into the maternal and fetal compartments. Estrone and estradiol reaching the fetus by the umbilical vein can be 16a-hydroxylated to form estriol, but most of the estriol is probably formed by placental arornatization of the large amounts of 16a-OH-DHA sulfate synthesized by the fetus (D6). Estriol is rapidly catabolized by infants (see Section 4.1.3), but if this is so for the fetus, the similar concentration in arterial and venous cord blood is difficult to explain, though a small amount of the estriol in the arterial blood may have been formed by the fetus from estrone and estradiol supplied by the placenta. 6. 1 7 - 0 x 0 Steroids in Infant Blood and Urine

Eberlein (E4) found that t h e level of total 1 7 - 0 s in infant blood rises after delivery and remains elevated during the first week of life, thereafter declining. The level rises in response to medical or surgical stress. When the group assay of 17-08 using the Zimmermann reaction was applied to infant urine, the results showed a high concentration (56 mg/liter) during the first few days after birth, but by day 14 i t had fallen to approxiniately 4 mg/liter (PlO). From findings such as these, it was thought t h a t infants after birth go through a period of hypoadrenalism. Infant urine, however, contains considerable quantities of nonspecific chromogen which interferes with the colorimetry in steroid assays (see Section 3.3) ; also, when the Zimmermann staining steroids and steroidlike compounds were separated by paper chromatography, they were found to be quantitatively and qualitatively very different from those in adult urine (B18, C6). The 11-deoxy-17-08, DHA, etiocholanolone, and androsterone, which predominate in the group in adults are present in the infant in only trace amoiints if they are detectable a t all (B17, C6), and between five and eight chromogens are present with a polarity lower than that of androsterone. Their rate of excretion falls off rapidly during the first few days after birth to almost undetectable levels on day 6, and their excretion is almost doubled in premature infants (C5).

175

STEROID METABOLISM IN EARLY INFANCY

TABLE 10

EXCRETION OF ll-Ox~-17-OS A N D ONE UNKNOWN COMPOUND IN THE URINEOP ADULT FEMALES, INFANTS 1-3 DAYSOLD, A N D ONE NORMAL INFANT OVER THE FIRST 4 POSTNATAL DAYS^ Compound

adult female"

Infant," 1-3 days

Infant, day 1

1lp-OH-etiocholanolorie 1lp-OH-androsterone

370d 610

Tinknown 11-0x0-etiocholanolone 11-0x0-androsterone

650 180

13 18 29 17 13

4 41 24 24

-e

Infant, Infant, day2 day3

52 30 30

-

4 6 9

Infant, day4

3 12 -

D a t a from Cathro et al. (C6). Average of 10 estimations on a pool of 24-hour urine specimens from 6 women aged 18-36 years. c Average of estimations on pooled 24-hour nrine specimens from 74 full-term and 5 premature infants and on individual 24-hour specimens from 31 infants. d Values are expressed as micrograms per 24 hours. e A dash denotes that the level was too low for measurement. a

It is generally agreed that up to 7 years of age D H A is either undetectable in urine (B6, F4, M13, P8, S9) or present only in small amount (C6, L6, V2). Both DHA (fetal life apart) and androsterone seems to be associated only with the reproductive years of life in both sexes since Migeon et al. (M13) found the levels in plasma to be low or zero, from shortly after birth until the age of 5-7 years when a steady rise took place until 20 years. After 30 years the levels declined until after the menopause in women, and in men over 65 years, zero or near zero levels were again found. The urinary excretion pattern of the 11-oxy-17-0s during the first few days bears more resemblance t o t h a t found in adults (Table l o ) , but again paper chromatography shows other compounds with similar polarity to be present (C6). The rate of excretion of a major unknown substance is given in Table 10, and it will be seen that the urinary content of this and of the known compounds rapidly decline over the first few days. The results shown were determined by scanning paper chromatograms stained with Zimmermann reagent. Identification of the named steroids was not complete and consisted of a comparison of R f values in various solvent systems before and after acetylation of the compounds. Figure 3 shows examples of the Zimmermann staining urinary steroids in chromatograms of extracts of adult and infant urine and of urine from an infant with a form of the adrenogenital syndrome. Table 11 shows t h a t the excretion of the major adult type 17-0s is very low a t 6 months, but by 3-5 years of age substantial amounts are present, and the increase in excretion continues until puberty.

176

FREDERICK L. MIT CHE L L AND CEDRIC H. L. SHACKLETON

TABLE 11

EXCRETION OF IND~VIDUAL 17-0s FROM 6 MONTHS TO 7 YEARSOF AGE" 5-7 Yearsa 13 67b 0 45 94 0 43 167 378 45 133 11 87 172 -

59 306 522 17 44h -

a

Values are expressed as micrograms per square meter of body area/24 hours. on the validity of the result were expressed by the original authors.

* Reservations

It is evident that during the first few postnatal days quantitative and qualitative changes in the excretion of the 1 7 - 0 s are rapidly taking place and initially only a small proportion of the urinary Zimmermann chromogen excretion is made up of the type of 1 7 - 0 s normally present in the aduIt. The early postulation that the rapid fall indicated a period of adrenal hypofunction is certainly not correct for all steroids ; cortisol production, for instance, shows no reduction (Section 7.3), but i t does show t h a t sudden changes concurrent with birth are taking place which affect the production and fate of the steroids and steroidlike hormones concerned. Even for adults, the biological significance of the 17-0s in the urine, except as metabolites of the corticoids and testosterone, is not understood. It is possible t h a t with the rapid growth which takes place in the fetus and the newborn, their anabolic properties are more important than their action as androgens, and in this respect it is worthy of note t h a t after perfusion of a fetus with androstenedione Mancuso e t al. (M2) found t h a t 1 Ip-OH-androstenedione was a major metabolitc both in the adrenals and in the perfusate. Meyer et al. (M8) have also

STEROID METABOLISM IN EARLY INFANCY

177

shown that of eight naturally occurring androgens tested, only ll/3-OHandrostenedione showed an anabolic: androgenic ratio greater than testosterone. 11p-OH-androstenedione is catabolized to adrenosterone and the four urinary 11-oxy-17-0S shown in Table 10. A prominent spot with the R f value of adrenosterone was found by Cathro et al. ( 0 3 ) after paper chromatography of both infant and maternal urine extracts. The level of DHA in cord blood is very high and the lack of significant quantities of the compound in infant urine may be explained by 16ahydroxylation being the dominant metabolic pathway (see Section 8 for the excretion of l6a-hydroxy-As steroids). In adults DHA is the principal precursor of androsterone and etiocholanolone, and they are its main metabolites (R11, V l ) . 7.

The

Urinary Excretion, Blood Levels, and Production

of Czl Steroids

C,, STEROIDS ASSAYEDBY GROUPMETHODS 7.1. URINARY The urinary excretion of 17-OHCS measured as Porter-Silber chromogens is normally approximately 0.5 mg/24 hours during the first 8 days of life ((317). For premature infants the figure is somewhat lower during their first week. Ulstrom e t al. (U2) obtained normal values of approximately 0.1 mg/24 hour (0.6 mg/m2/24 hours) for the first day of life, increasing to approximately 0.5 mg/24 hours (2.0 mg/m2/24 hours) by day 7. Comparable values for adults were 3.1 mg/m2/24 hours,. During the early days after birth the pattern of urinary corticoid excretion is very different from that found in the normal adult, and this must be taken into account in the interpretation of group assays. An increased proportion of polar metabolites is formed from cortisol (U3) ; these are poorly extracted by the methylene dichloride normally used for extracting Porter-Silber chromogens, and ethyl acetate is required (R8). The finding of low levels, increasing after a few days, may thus reflect an alteration in metabolism toward the adult type rather than an increasing cortisol production. For children up to 12 months of age, the techniques developed by Norymberski and his colleagues for measuring urinary 17-oxogenic steroids and 17-OHCS (A7, G1, N7) can include the measurement of a high proportion of nonspecific chromogens (B15), The levels measured average less than 4 mg/24 hours up to 6 years of age (ClO, N7, L l l ) , and as for the measurement of Porter-Silber chromogens, the makeup of the groups of steroids assayed is probably very different from that found in adults (C6). An increased excretion of corticosterone (Sections 4.2.3 and 7.3) is reflected in a high ratio of 17-deoxycorticosteroids [assayed by the

178

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

method of Exley et al. (E8 )] to 17-OHCS. This ratio is 3:l in infants compared to 1:3 in adults (Exley and Norymberski, quoted in C6). In view of the above findings, criteria which are used in the interpretation of group assays in infants must be different from those which are normally applied t o adults. 7.2.

URINARY EXCRETION OF INDIVIDUAL CZ1STEROIDS

By means of a rigorous preliminary purification procedure, the blue tetrazolium staining steroids in infant urine have been separated by paper chromatography (B18, C6 ); more recently a similar effect has been obtained using thin-layer chromatography, but in this case only a very simple preliminary wash of the extract is required (SlO). With both techniques the pattern of excretion has been found t o be entirely different from that obtained for normal adults. Blue tetrazolium is not only reduced by an a-ketol group in the side chains of CZ1 steroids but the blue formazan is also produced by C19 steroids with a reducing group in ring D, and since substitution a t position 16 (forming a-ketols with substituents a t position 17) is a feature of many steroids in infant urine, their presence complicates the detection of C,, steroids by blue tetrazolium. Figure 4 shows typical patterns formed from extracts of adult and infant urine and of urine from an infant with a form of the adrenogenital syndrome. The lower polarity compounds in the case of the normal infant and that of the adrenogenital syndrome are almost certainly a-ketolic C,, steroids. The most polar compounds present in normal infant urine have not been identified and do not necessarily correspond to standards with similar R f values. By means of the long paper chromatographic technique (B18), accurate Rf values for the bands have been determined, and these, together with the amounts excreted and the Rf values of standard steroids, are shown in Table 12. It is evident that although THF and THE are the most abundant reducing steroids in the urine of adults, in infants neither is the dominant reducing steroidlike compound. This finding is in accord with the studies of Bertrand et al. (B12), which showed that in infants ~ Crecovered in only 10% of the radioactivity injected as ~ o r t i s o l - 4 - ~ is urine as THE, THF, and allo-THE compared with 25% (F8) and 36% (F3) found in adults. An explanation could lie in the inadequacy of Aring reduction in the newborn (see Section 7.3). I n the blue tetrazolium-stained paper chromatograms, 21-OH-pregnenolone has been identified as the major compound in the medium polarity fraction (B16). The unknown more polar compounds may be grouped approximately in three areas of concentration: one more polar

179

STEROID METABOLISM IN EARLY INFAKCY

TABLE 12

EXCRETION OF Crl STEROIDS AND OTHER UNKNOWN BLUE TETRAZOLIUM REDUCING COMPOUNDS IN INFANTS~.~

URINARY

Infant urine Standard compound Allo-dihydrocorticosterone

Rf 0.157

rg/24 hours 9.7 8

5

6fl-OH-cortisol 6p-OH-cortisone

THF

0.0017 (4) 0.016 (48) 0.067 (48)

THE

0.113 (52)

Cortisol Cortisone Tetrahydrocorticosterone Dihydrocortisone

0.156 (130) 0.29 (107) 0.39 (7) 0.43 (17)

Corticosterone

0.61 (30)

6 5 28 9 10 8 19 44 63 30 79 56 39 26 9 25 160 106 36

Rf 0.13 0.16 0.18 0.22 0.24 0.31 0.33 0.34 0.38 0,0019 (25) 0.016 (114) 0.064 (104) 0.081 (32) 0.103 (102) 0.113 (80) 0.14 (84) 0.156 (114) 0.29 (34) 0.36 (67) 0.39-0.45 (44) 0.50-0.58 (44) 0.61 (65)

After Cathro et al. (C6). The first eight compounds were separated in Bush paper chromatography system LB 21/80 and the remainder in Bush system T/75. Identification was by Rfvalue only. Quantitation was by averaging the results from urine collected during the first 3 postnatal days from a t least 50 infants. Number of individual measurements from which each average Rf value was obtained is given in parentheses. a

b

than THF, one in the THF-THE area, and one between cortisone and corticosterone (C6). Eagle (El) was the first t o suggest that, in the infant, corticosterone is excreted in greater amount than cortisol (Sections 7.1 and 7.3).Possible explanations have been discussed in Section 4.2.3. The highly polar fraction is difficult to investigate because the necessary use of ethyl acetate to attain complete extraction gives extracts that also contain large quantities of highly polar impurities. Ulstrom e t al. (U3) however, have identified 6P-OH-cortisol and have indicated that this and other very polar metabolites are more important pathways of cortisol disposition in the newborn than later in life. This was confirmed by Reynolds e t al. (R8). )

180

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

The excretion of the Cz, 3/3-hydroxy-As steroids pregnenolone, 21-0.Hpregnenolone, and 16a-0H-pregnenolone will be dealt with in Section 8. 7.3.

BLOODLEVELS, PRODUCTION RATES,AND METABOLISM OF CORTISOL,CORTISONE, AND CORTICOSTERONE

Blood levels of 17-OHCS have been assayed by the method of Nelson and Samuels (N3) by two groups of workers (B12, K7), who found values of approximately 20 pg/lOO ml in cord blood, falling to 0.9 (14 assays, K7) and 8.5 (36 assays, B12) ,ug/lOo ml over the first 5 days of life, little change then taking place over the next 2.5 months (B12). The average levels found for adults were approximately 13pg/loO ml. However, when cortisol is assayed by a specific double isotope derivative dilution technique, the level in cord blood is similar to that found in the blood of normal nonpregnant women (Table 5 ) , and there is no marked fall during the first 5 months (H9, U4). The cortisone content of cord blood is considerably greater than th at found in nonpregnant women (Table 5). It falls rapidly after birth, but a low cortisol: cortisone ratio persists in the infant for a t least the first 2 weeks after birth (B43). Possible explanations for this have been discussed in Section 4.2.3. Cortisol production rates have also been measured, and when corrected for surface area little change is shown during the first 2.5 months (B12, K l ) , though on average the levels are approximately 1.5 times those in adults. The corticosterone production rate during the first 2 months when corrected for surface area is more than three times that in the adult (L6). I n the newborn, the half-life of THE is comparable to that in adults (B32), but the half-life of cortisol is considerably prolonged (B32, C18, M10, R8). This can be caused in adults by estrogen therapy, but an explanation on this basis is unlikely for infants because the effect continues after the fifth day of life, when estrogen received during uterine life has been eliminated (D10). An increase in protein-binding w,hich also might be an explanation, and which occurs in the mother in pregnancy, does not take place in the fetus (Section 4.2.3) ; another discrepancy is that the half-life of the tetrahydro compounds is increased by estrogen therapy (Wl). The most probable explanation is that in the infant there is an inadequacy of A-ring reducing enzymes similar to that occurring during cirrhosis of the liver in adults (B32, M10, R8). I n the newborn, the activity of several hepatic enzymes is reduced (D17). 7.4. CONCLUSIONS Early studies with the group assays of urinary and blood steroids indicated that the newborn infant may go through a period of adrenal

181

STEROID METABOLISM I N EARLY INFANCY

hypofunction (see Sections 6, 7.1, and 7.3), and certainly major changes were to be expected, considering the extensive morphological changes which take place in the adrenal a t this time, and considering that steroid metabolism in general is changing dramatically from that required in utero to a very different system required for independent life. However, specific measurements of cortisol in blood, together with its production rate, have shown that there is no deficiency; indeed, the production relative to adults is somewhat higher. During the first fortnight after birth the conditions required for the application of the Allen correction formula, used in the group assay method of Nelson and Samuels (N3), are not fulfilled (U4) , and this could account for the low values obtained. Until more is known of the compounds assayed by group techniques in infant urine and blood, considerable caution must be exercised in the interpretation of any results obtained. 8.

The Urinary Excretion and Blood levels of 30-Hydroxy-A5 Steroids

I n the infant, unlike the adult, the major CISsteroids have the A5 configuration. Table 13 compares the excretion of the more important 3phydroxy-A5 steroids in day-old infants with that in adult males, and Fig. 6 shows the comparison in histogram form. One unknown A5 steroidlike compound is also included. The excretion of such a large quantity and variety of A5 steroids is undoubtedly related to the low activity of 3p-HSD in the fetus (see Section 4) and must persist during early infancy. The results shown in Table 14 and Fig. 2 indicate that on the whole TABLE 13 OF 3b-HYDROXY-A6 STEROIDS" URINARY EXCRETION

Steroid

Day-old infant average of 9 (pg/m2/24 hours)

DHA -b Unknown 880 Androstenediol (17a) 710 Androstenediol (176) 16-0x0-androstenediol 1800 16a-OH-DHA 2480 168-OH-DHA 2400 16a-OH-pregnenolone 2380 Androstenetriol 2120 3@,17~~,20a-trihydroxypregn-5-ene -

Adult male average of 4 C.g/m9/24 hours) 1000 -

i

370 150 745

-

300 340

Reproduced from Mitchell (M16). A dash denotes presence only in barely detectable or undetectable quantity.

TABLE 14 URINARY EXCRETION^

O F 3B-HYDROXY-As

STEROIDS I N

EARLY INFANCYb

m

Age in days

Unknown 21-OH-pregnenolone

+ 17p)

Androstenediol (1701

16-0x0-androstenediol 16a-OH-DHA 16B-OH-DHA 16-OH-pregnenolone Androstenetriol Total

483 (36-2357) 210 (15-471) 172 ( <20-840) 903 (87-2420) 962 (72-2240) 666 (94-1520) 827 (52-1520) 640 (84-1620) 4863

555 (285-1608) 236 (40-492) 191 ( <20-800) 1060 (80-3000) 1026 (170-2560) 808 (200-1444) 900 (80-2260) 626 (112-1220) 5402

616 (374-1221) 279 (128-590) 208 ( <20-880) 1502 (285-3260) 1304 (410-3180) 780 (200-1740) 1167 (200-3020) 968 (250-1600) 6824

Average excretion (with range) expressed as micrograms per 24 hours. a Reproduced from Shackleton and Mitchell (S9). 6

The number of infants is given in parentheses.

683 (320-1690) 311 (104-690) 144 (<20-220) 1518 (660-2080) 1057 (539-1565) 7 19 (350-1393) 1298 (910-2424) 996 (820-1300) 6726

429 (212-895) 193 (42435) 47 ( <20-120) 1031 (680-1680) 770 (900-1140) 460 (176-878) 1037 (264-2080) 759 (383-1100) 4726

450 (150-675) 165 (28-320) 37 ( <20-140) 899 (340-1480) 855 (NO-1800)

406 (100-800) 1114 (280-2680) 621 (404-820) 4547

n M tc

2 Q m

P

n

w

STEROID METABOLISM I N EARLY INFANCY

183

the compounds measured are being synthesized by the infant in fairly constant, or even increasing, amounts during the first 6 days of life, though the variation in excretion of individual steroids by different individuals is considerable (C11, S9). Reynolds (R5,R7) has shown that there is on average an increase in the excretion of 16a-0H-pregnenolone1 16a-OH-DHA, and androstenetriol by premature infants for a t least 40 days. It is thus evident that the ~1~ compounds are not formed only in utero, as they would then be excreted in rapidly decreasing amount. Reynolds (R3) also measured 16a-OH-pregnenolone and 16a-OH-DHA in the urine of children up to 12 years of age and could detect none after the first 6 months of age. Urinary excretion rates for the major 3phydroxy-A5 steroids during the first 6 months are shown in Fig. 7. Evidence that the A5 steroids are of adrenal origin has been supplied by Reynolds (R2) , who showed that the urinary excretions of 16a-OH-DHA and 16-oxo-androstenediol were 0.3-1.4 mg/24 hours in a newborn female with the C-21 hydroxylase deficiency type of congenital adrenal hyperplasia. These fell to undetectable amounts with dexamethasone suppression therapy. A very low excretion of three 3p-hydroxy-A5 steroids was found by the authors in the case of a baby shown subsequently a t autopsy to have little adrenal tissue (see Section 11). Further circumstantial evidence is given when abnormalities are considered in Section 11 and also the effect of corticotropin in Section 12. Of the compounds measured, androstenediol, 16-oxo-androstenediol, lGa-OH-DHA, and androstenetriol may be considered to be estrogen precursors (or the metabolites of estrogen precursors) produced initially by the fetus for aromatization in the placenta. l6a-0H-pregnenolone is not an immediate precursor of estrogen but is possibly simply a catabolic product of the quantitatively very important steroid, pregnenolone (Section 4.2.1). 21-0H-pregnenolone is excreted in small quantity (150 pg/24 hours or less) by pregnant women (C7, P4, P5), but has been detected in adult urine only after the administration of corticotropin (D13). The significance of the presence of the compound in infant urine has been discussed in Section 4.2.3. I n infant urine and also in umbilical cord blood, the 17a epimer of androstenediol predominates approximately 9 times over its 17p counterpart whereas in adults the 17p epimer is dominant ( S ll) . The reason for this is not readily apparent. One possibility is that both epimers are produced and 16a-hydroxylase selectively acts on 17p-androstenediol producing androstenetriol. The authors have recently found that such a selective hydroxylation does take place during the in vitro incubation of fetal liver slices. The effect may be enhanced in u f e r o by the placenta

184

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

a 16 a-OH- pregnenolone

270240-

0

16a-OH- DHA

210-

2 180-

8

0

150-

\

rn 120-

a.

0

'

9060-

300

0

0

0

0 0

I

I

I

I

I

I

I

0 I

08 I

PI

I

I

I

0

8 P

I

I

? I

i

21-OH-pregnenolone

o Androstenediol

01 (780)

180-

3

120-

\

s

:;;I

604020-

0

Unknown

0

16-0x0 -0ndrostenediol

A I6P-OHDHA

2

160,g 140-

A

o Androstenetriol

0

0

b 2

A

a". A 1

B b 0

e

0

.E:

8 8

1 A 1

0

e

Age (days)

FIQ.7. The urinary excretion of the major 3p-hydroxy-A' steroids and one unknown compound during the first 6 months after birth. Assays were performed on urines from different individuals, each infant being investigated once only.

STEROID METABOLISM I N EARLY IKFANCY

185

selectively aromatizing the 17P epimer to form estradiol. The conversion of DHA sulfate to androstenediol has been indicated in adults (R12) and in the fetus (B24), but it is not known what is the ratio of 17a- to 17P-androstenediol produced. An important aspect of the relative amounts of circulating 17a and 17P epimers arises if androstenediol is an important intermediate in the synthesis of testosterone, as has been suggested may be the case (B5, Hl). An excessive amount of 17a-androstenediol may conceivably lead to an excessive production of epitestosterone relative to testosterone. The recent identification of 16P-OH-DHA as a major compound in infant urine (S13, Tables 13 and 14) plus the fact that under certain conditions it spontaneously isomerizes to the more stable 16-oxo-androstenediol, raises the question to what extent 16-oxo-androstenediol, also reported as a major component in infant urine (R2, S9, Tables 13 and 14) and tentatively identified in umbilical cord blood (C12), has been produced as an artifact during extraction and purification. An explanation for the presence of 16p-OH-DHA in quantity is not immediately apparent. Similarly to the formation of estriol in the placenta from 16aOH-DHA, it may form 16-epiestriol, 16/3-OH-estrone or its isomer 16oxo-estradiol, though the urinary output of these compounds in pregnancy is not great, being respectively 0.8, 0.7, and 1.1 mg/24 hours (B34). The unknown compound shown in Tables 13 and 14 has not been identified, but certain of its properties have been determined (S8). The 3 P - h y d r o x y - ~steroid ~ content of infant blood has been found by Eberlein (E4) to rise rapidly after delivery and decline after the first week. A similar rise and fall was noted for the 17-05 (Section 6 ) - Initially the .A5 content was highest, but the situation was reversed after 7 days. Eberlein suggested that the effect might be explained by involution of the fetal zone of the adrenal concomitant with a maturing liver function. The production of 3p-hydroxy-A5 steroids evidently proceeds in quantity for many weeks after birth, and it will be important to discover whether this is by design, the compounds being needed for some purpose, or whether they are produced incidentally by enzyme systems designed solely for use in utero and still functional after birth. The high activity of l6a-hydroxylase similarly could be a vestigial remnant of its necessary function in utero concerned with the production of estriol. It may be involved with the important problem of electrolyte balance at birth and in early independent life, since natriuretic properties have been attributed to 16a-hydroxylated C& steroids (H7), and a patient with congenital adrenal hyperplasia of the salt-losing type has been shown to produce excessive quantities of 16a-OH-progesterone (54).

186

FREDERICK L. MITCHELL AND CEDRIC H . L. SHACKLETON

9.

The Urinary Excretion of Estrogens

It has long been known that certain physiological and histological changes in the newborn are due to estrogenic stimulation or estrogen withdrawal, the hormones having been received from the placenta (H3). It is generally accepted that after birth until puberty estrogen production is very low, but during the first few days the infant has to clear from its system the considerable quantity of estrogen received from the placenta. It has been shown in Section 4.1.3 that the fetus and infant metabolize estrogen differently from the adult. Estradiol is rapidly hydroxylated and estriol is also dealt with more completely. This type of metabolism is reflected in the urinary estrogen excretion pattern in infancy. I n an investigation on approximately 150 healthy boys, Diczfalusy e t al. (D10) found that the average urinary content of estrone in micrograms per liter decreased from 6.6 on the second day of life to zero or1 the fifth day; estradiol was undetectable on all days. Estriol decreased from 7160 on the second day to 60 on the fifth day, and i t was believed to have disappeared completely from the urine by 2 weeks. Thus the rate of fall in the concentration of estriol in infant urine roughly parallels that in the mother post partum but estrone and estradiol are still present in maternal urine when they are undetectable in the urine of infants. A new estrogen, estetrol, which is believed to be produced solely by the fetus and newborn infant, has recently been found to be a normal constituent of newborn urine (26) (Section 4.1.3). 10.

Steroid Conjugation

Lathe and Walker (L3) have suggested that glucuronyl transferase (responsible for the formation of bilirubin glucosiduronate) is deficient in human newborn liver, leading to a limitation in the ability of the newborn to excrete bilirubin. Similarly there is a limitation in the formation of the glucosiduronates of THE and T H F (the first stage in the major pathway of cortisol metabolism to identifiable compounds in the adult), and an inverse relationship has been demonstrated between the maximal serum bilirubin concentration attained in infants and the percentage of radioactivity injected as cortisol-4-*"C and excreted as the glucosiduronate ( A l ) . Only 1-4% of cortisol administered to infants is recovered as urinary steroid glucosiduronate compared with 17% in an adult (PQ,R8), and ~ Crecovered as radioactivity in the only 15% of injected ~ o r t i s o l - 4 - ~ is glucosiduronate fraction compared with 46% in adults (B12). The effect is even more pronounced in premature infants ( A l ) . These experiments all indicate a reduced glucosiduronate formation, and other methods of

STEROID METABOLISM IN EARLY INFANCY

187

steroid excretion must have increased importance. When the newborn infant is loaded with cortisol, more than 50% of the total urinary phenylhydrazine-reacting steroids extractable after P-glucosiduronidase hydrolysis are unconjugated (R8) whereas the comparable figure for adults is only 5% (S19). This could be due either to saturation of the glucosiduronate-forming mechanisms or an increase in the conversion of cortisol to highly polar steroids which are normally excreted in the free form. Bertrand et a2. (B11) have provided evidence for a maturation of the glucosiduronate-forming mechanism over the first 40 days of life; initially the level of plasma 17-OHCS conjugated with glucuronic acid is very low and shows no increase with corticotropin even though the increase of free 17-OHCS is normal. By day 42 both the resting level of the glucosiduronate-conjugated fraction and the increase after corticotropin are normal. A deficiency in A-ring reduction is also evident in the newborn, since though THE has a normal half-life in infant blood (B32), that of cortisol is considerably prolonged (C18, R8). This discrepancy cannot be explained either by increased protein-binding (which does not occur in infancy) or by the effect of estrogen, since this increases the half-life of THF (Wl), but may be due to an inadequacy of A-ring reducing enzymes (B32, M10, R8) similar to that postulated as occurring during cirrhosis of the liver. It is thus likely that there is a deficiency both in the mechanism responsible for A-ring reduction of cortisol and in glucosiduronate conjugation. The latter is not a limiting factor in the rate of cortisol metabolism and excretion because sulfation probably takes its place (see below). Both deficiencies disappear with maturation. The production and excretion of highly polar steroidlike material (C6), including 6/3-OH-cortisol (U3), mainly in the free form (R8)may account to some extent for the low recoveries of glucosiduronate conjugated metabolites of administered cortisol. It could be due to the action of estrogen (R13, W l ) , but there is no change in the ratio of highly polar to less polar Porter-Silber chromogens during the first 6 days of life (U3) and excess estrogens have largely been eliminated by day 5. Though the proportion of steroids excreted free is considerably increased in infants, sulfate conjugation is of even greater importance. Of the total urinary hydroxysteroids, the amount excreted sulfated is as great as, or greater than, the free or P-glucosiduronidase hydrolyzable fraction, and most of the urinary cortisol, cortisone, and corticosterone is excreted as ester sulfates (D16). In prematurity, the proportion may be even higher: Ducharme e t al. (D18) found that 87, 69, and 88% respectively, were freed by solvolysis. The 3P-hydroxy-A5 steroids are quantitatively and qualitatively the most important group of steroids in

188

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

infant urine and cord blood (E4,R6, SS), and the majority are present as the 3P-sulfates (E4). It is possible that such sulfate conjugation is of importance for their biosynthetic role (L15). Certainly the distribution of hydrolyzing enzymes can affect the utilization by tissues of steroid conjugates; for instance, the virtual absence of sulfatase in the fetus probably restricts its metabolism of the 3P-hydroxy-s15 steroids while the abundance of this enzyme in the placenta means that there they can be freely utilized. Similarly the transplacental passage of estrogen sulfate is twice that of the glucosiduronate, presumably due to the hydrolysis by the placenta of the sulfate but not of the glucosiduronate. Unconjugated estrogen is transferred very much more rapidly (L12). Steroid conjugation in the fetus could be of major importance by inactivating horomones which are present in high concentration in the fetoplacental circulation and impeding transfer to the mother. TABLE 15 STEROIDS IN UMBILICAL CORDPLASMA MONO-AND DICONJIJGATEDO

PROPORTION OF INDIVIDUAL 3j3-HYDROXY-A6

Percent conjugated Steroid Pregnenolone

DHA 17a-OH-pregnenolone 21-OH-pregnenolone 16-0x0-androstenediol 16a-OH-DHA 16j3-OH-DHA 16a-OH-pregnenolone Androstenediol (17a) Androstenetriol

Mono-

Di-

100 100 100 0 50 90 0 100 0 60

0 0 0 100 50 10 100 0 100 40

From Shackleton et al. (S12). Presented at the International Symposium on Fetoplacental Unit, Milan, Italy, 1968. (I

At least three, 3P-hydroxy-A5 steroids have been shown to be present in urine in diconjugated form; of these 16P-OH-DHA and 21-OH-pregnenolone are hydrolyzed by the Helix pomatia enzyme, but androstenediol ( 1 7 ~ ~is) freed only by solvolysis (Sll, Fig. 5 ) . The proportions of the individual 3P-hydroxy-A5 steroids found in umbilical cord plasma monoand diconjugated are shown in Table 15. Very little is free, but Simmer and co-workers (S21) have established a concentration of free 16a-OHDHA of approximately 0.4 pg/lOO ml.

189

STEROID METABOLISM IN EARLY INFANCY

1 1.

Abnormalities in Steroid Production

The congenital abnormalities of steroid production have been extensively discussed in previous reviews (B30, B31, V7). Visser (V7) has also covered in infants and children other disorders which lead to abnormal steroid production. Table 16 shows the major points of interest in patients with these abnormalities; they will be discussed in the order: congenital adrenal hyperplasia, congenital adrenal hypoplasia, and hypoaldosteronism.

-

Accumulates m adrenal ( L i p i d adrenal hyperplasia)

HO Cholesterol Deamolaee defect- - -.

-

I--

7%

@

7

16a-OH-Pregnenolone 17a-OW-Pregnenolone-DHA-l6~-OH-D~

~

21 -OH-Pregnenolone

HO Pregnenolone

y.

CH,OH

Pregnanetriol

c0

l7o-OH-Progesterone

/

Progesterone 2l-Hydroxylase _ _ _ - - _ defect

Deoxycorticosterone_cTetrahYdrodeoxycorticosterone

I

Tetrahydro- 11 deoxycortisol

,

11-Deoxycortisol 1 Ig-Hydroxylase - - defect

Cortisol

--I <

Tetrahydracorticosterone

corticosterone

18-Oxidation

--H

&lo-tetrahydrocorticosterone Fh;trahydm-ll-dehydrocorticosterone

Aldoste-nne -Tetrahvdroaldosterone

FIQ.8. Pathways for the synthesis of cortisol, corticosterone, and aldosterone, showing the enzyme defects that may occur.

TABLE 16 FORMS OF ADRENAL MALFUNCTION IN INFANTS ~~

Malfunction

Urinary steroids

Deficiency

Congenital adrenal 21-Hydroxylase with and hyperplasia without salt loss

38-Hydrox y deh y drogenase 110-Hydroxylase Cholesterol desmolase

Congenital adrenal All steroid-forming hypoplasia mechanisms Hypoaldosteronism 18-Oxidation mechanism

Delayed maturation of zona glomerulosa

High 17-05 excretion High 3@-hydroxy-A6steroid excretion (infants) High pregnanetriol excretion High testosterone excretion High 3/3-hydroxy-A6-steroid excretion High tetrahydro-11-deoxycortisol and tetrahydro-lldeoxycorticosterone excretion No information on steroids in urine (cholesterol accumulates in adrenal glands) Not known (low maternal estriol excretion prior to delivery) Abnormally high excretion of tetrahydrocorticosterone, allotetrahydrocorticosterone, and tetrahvdro-lldehydrocorticosterone Normal steroid excretion (other than aldosterone) and metabolites

-

Virilization

Aldosterone production rate

Blood pressure

+

Normal or low

Normal or low

LOW

Low

-

High

-

Low

Low

-

Low

-

Low

+ +

R

2 X

x

r m

m

9 0

P

-

Low increasing to normal after first months or years of life

-

3

0

Z

STEROID METABOLISM I N EARLY I N FA N C Y

191

11.1. CONGENITAL ADRENAL HYPERPLASIA This can be due to defects in several enzyme systems required for the formation of cortisol (Fig. S ) , but in all cases the inadequate level of cortisol in the blood causes corticotropin to be secreted by the pituitary in increasing amount; this in turn causes hyperplasia of the adrenal glands with overproduction of those steroids not affected by the enzyme block. The most common deficiency is in 21-hydroxylase, but deficiencies have also been reported in infants and children of 3p-HSD, 1lp-hydroxylase, and cholesterol desmolase, and in older infants of 17a-hydroxylase W4) * 11.1.l. d l -Hydroxylase Deficiency Deficiency in this enzyme is predominant in nine-tenths of all cases of congenital adrenal hyperplasia. Adrenal androgens are not affected by the block and though sometimes virilization becomes apparent in females only later in life, generally infants are virilized a t birth (V7). For this, testosterone production by the fetus early in pregnancy is probably responsible; it is excreted in large amounts by patients with a 21-hydroxylase defect. Degenhart et al. (D5) reported testosterone production rates of 342-11,400 pg/24 hours by 11 infants and children with the defect, the normal production rate being less than 500 pg/24 hours. It is possible that some of this testosterone is produced by the peripheral metabolism of DHA and androstenedione and it may be excreted before having passed into the circulation in biologically active form. It is likely, however, that some “active” testosterone is produced, and this may be sufficient to effect the virilization of fetuses with this disorder, since adrenocortical activity in the fetus begins as early as the third month of pregnancy, before the complete differentiation of the genitalia. 17a-OH-progesterone is the compound immediately prior t o the enzyme block in the chain of synthesis of cortisol (Fig. 8 ) , and large quantities of i t are produced. Before excretion it is largely reduced to pregnanetriol (B26, B42, F9, 52). Bongiovanni et al. (BS1) reported considerable amounts of 17a-OH-progesterone and no detectable 21-hydroxylase activity in the adrenal glands of affected individuals. The excretion of 170s is increased, probably because of excessive production of DHA and androstenedione rather than through the metabolism of C,, steroids (B36, F9). An elevated excretion of ll-deoxytetrahydrocortisol has been reported (B19, D4). I n approximately one-third of patients with a 21-hydroxylase defect, there is an excessive excretion of sodium chloride with abnormally high serum potassium and low serum sodium concentrations. After treatment with cort,isol and deoxycorticosterone acetate (or equivalent therapeutic

192

FREDERICK L . MITCHELL AND CEDRIC H. L . SHACKLETON

compounds) the serum electrolytes rapidly return to normal and the elevated excretion of the 17-05 is reduced. Since aldosterone is principally responsible for controlling the levels of serum electrolytes, Degenhart and co-workers (D4) have studied its secretion rate in 9 infants with 21-hydroxylase deficiency, three with the “salt-losing” form of the disorder. They found it was normal (60-125 pg/24 hours), for the infants not displaying excessive salt loss, but extremely low for the salt losers (< 10 pg/24 hours). In normal infants and children and in patients with simple virilizing adrenal hyperplasia, reduction of dietary sodium results in a 2-fold to 3-fold increase, both in the production rate of aldosterone (D4) , and the concentration of aldosterone metabolites in the urine (B31). Infants with 21-hydroxylase deficiency leading to salt loss however, show insignificant increases in aldosterone production on salt deprivation (B31, D4). These findings indicate that adrenal hyperplasia with salt loss iz due to a deficiency in the production of aldosterone. It might be expected that the production of aldosterone (a 21-hydroxylated steroid) should always be reduced together with cortisol in cases of 21-hydroxylase deficiency. However, it is possible that different 21hydroxylase enzymes are required for the production of the two compounds (D4) , and a deficiency of the cortisol-producing enzyme may not always be accompanied by a lack of that used in the synthesis of aldosterone. The 21-hydroxylases acting on progesterone and 17a-OHprogesterone have been shown to have different physical constants (S15), and deficiency of a 21-hydroxylase acting on 17a-OH-progesterone may not affect the synthesis of 17-deoxy steroids, e.g., corticosterone and aldosterone. Alternatively, if only the one 21-hydroxylating system exists, deficiency may be more severe in cases showing salt loss. Evidence that the last theory may be correct has been supplied by Eberlein (E3) and Mattox e t al. (M5), who showed that salt losers have Iower IeveIs of cortisol metabolites than do non salt losers. I n older children and adults with the disorder, pregnanetriol is the major urinary steroid metabolite, but several other 21-deoxy steroids have also been identified (B30) including 3a1l7a-dihydroxy-5p-pregnan20-one (L16) , 11-oxo-pregnanetriol (Fl), and 3a,l7a-dihydroxy-5/3-pregnane-11,2Q-dione (D3). For infants, the situation is different; it is not always possible to detect pregnanetriol (E5), and the disorder must be identified against a steroid metabolism which is in any case very different from that normally found in the adult. Pregnanetriol, for instance, has not been detected in normal infant urine whereas it is a normal component for adults. Since the major steroids excreted by newborn infants have the 3phydroxy-A‘ structure (SS), and the excretion of these compounds is in-

STEROID METABOLISM I N EARLY INFANCY

193

creased when corticotropin is administered to infants (L7), it may be expected that greatly increased amounts would be produced by infants with a 21-hydroxylase defect. Eberlein (E5) has reported large quantities of DHA, pregnenolone, and 16a-OH-pregnenolone in plasma obtained from newborn infants with the disorder, and it has been shown by Reynolds (R4) that patients deficient in the enzyme excrete greater than normal amounts of 1h-OH-DHA and 16a-OH-pregnenolone; administration of corticotropin to these infants resulted in increased excretion of these compounds. A 7-day-old male infant with 21-hydroxylase deficiency studied by the authors2 excreted 35 mg of 3P-hydroxy-A5 steroids, the average value for normal infants on the sixth day of life being 4.5 mg. The compounds measured were only those 3P-hydroxy-A5 steroids which have so far been identified in the urine; the absolute total excretion may therefore be considerably greater than the figure given. For this infant, the levels of plasma electrolytes did not show any abnormality until the eighth day of life, a finding not unusual in salt-losing adrenal hyperplasia, since cases have been reported where the aldosterone secretion rate was shown to be below normal but salt-losing symptoms did not develop until postnatal days 7-9 (BSl, K10). Such newborn infants are probably in negative sodium balance for the first week of life, with an aldosterone secretion rate below normal yet sufficiently high to prevent symptoms becoming apparent earlier. Further investigation of the infant typifies difficulties which may be encountered in such cases. Urinary steroid assays were carried out by the method of Shackleton and Mitchell (S9) on the days shown in Fig. 9. Due to the deterioration of the condition of the infant, cortisone and fluorocortisol treatment was started on the ninth day of life. The considerable excretion of the 3P-hydroxy-A5 steroids on day 7 was largely made up of 16a-OH-DHA, 16-oxo-androstenediol, and l&-OH-pregnenolone. It will be seen from Fig. 9 th at the excretion of 21-OH-pregnenolone was also considerable (average normal excretion on postnatal day 6, 170 pg/24 hours), and it was considered from this finding that a deficiency of 21-hydroxylase was unlikely. The anomaly might be explained, as mentioned earlier, by there being different 21-hydroxylase enzymes, in this case for 17a-OH-progesterone and pregnenolone. However, on day 8, the 21-OH-pregnenolone had decreased considerably and substantial amounts of pregnanetriol had appeared in the urine, suggesting a progressive deficiency of a 21-hydroxylating enzyme. It is possible that the ratio of 16n-OH-pregnenolone to 21-OH-preg'The authors are grateful to Dr. J. W. Farquhar for permission to publish results which were obtaincd with his help and collaboration.

194

FREDEBICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

I5-

ro-

5-

0-

L

e

'

8

Il-0x0-ondrosterone and II-0x0- etiocholonolone

l-l

IlIL

78

12 13 . 15

#;c

12 13

15

7-

3 ' 6-

r"

16 a-OH-DHA

2.5

-

Unknown

2.0 -

1.5 -

1.0-

-

0.5

78

1.5

12 13

15

21-OH- Pregnenolone

1.51

Pregnonetriol

0.5

78

12 13 15

78

12 13 15

Age (days)

FIG.9. The urinary excretion of certain steroids in an infant with a deficiency of 21-hydroxylase. The filled-in columns indicate the steroid excretion while the infant was receiving therapy with cortisone and fluorocortisol. 21-OH-pregnenolone and pregnanetriol were identified by masa spectrometry. The unidentified compound reacts with blue tetrazolium and has a polarity similar to 21-OH-pregnenolone.

nenolone may be a useful parameter for the diagnosis of 21-hydroxylase deficiency. This ratio was 11:1 on day 7 and increased to 30: 1 on day 8 (normal average value for infants, 7 : l ) . The value of this ratio will be very limited if 21-hydroxylase is a t all substrate specific. T o confirm the diagnosis, the infant was reinvestigated a t the age of 14 months when, after stimulation with corticotropin for 14 days while on maintenance therapy with dexamethasone and fIuorocortiso1, the

STEROID METABOLISM I N EARLY INFANCY

195

urinary excretion of pregnanetriol was 47 mg/24 hours with an excretion of T H E and THF (for method of assay, see S10) of only 0.9 and 0.2 mg/24 hours, respectively. Il-Oxo-pregnanetriol, 3Pl16a-dihydroxy-5Ppregnan-20-one and 3a,l7a-dihydroxy-5p-prcgnan-20-one were also identified in the urine by gas chromatography-mass spectroscopy. 11.1.2. Sp-HSD Deficiency This deficiency is rare and was first described by Bongiovanni in 1961 (B27) and later in greater detail (B31). It is always associated with salt loss, and, as treatment with corticosteroids has little beneficial effect, it is almost invariably fatal. Due to the severity of the enzyme block in the production of androgens by the testes in early embryonic life, incomplete function of the genital structures is frequent in male infants, whereas females may be slightly virilized, possibly due to excessive production of the weak androgen DHA. The enzyme 3p-HSD has been shown to be absent from adrenal and testicular tissues obtained from infants with this disorder (G3). A high excretion of 21-OH-pregnenolone has been reported by Cathro et al. ((37) in an infant with the disorder (450 pg/24 hours, normal average 50 pg). The presence of this compound in abnormal amount cannot, however, be used for diagnosis since the infant with a 21-hydroxylase defect, previously described, excreted an even greater quantity on postnatal day 7 (Fig. 9). Bongiovanni et al. (B31) has isolated the following steroids from infants with the disorder: (1) DHA; (2) 16a-OH-DHA; (3) 16a-OHpregnenolone; (4) 17a-OH-pregnenolone; (5) 3p,17a,20a-trihydroxypregn-5-ene; (6) 3p,17a,20P,21-tetrahydroxypregn-5-ene.Since very large amounts of 3p-hydroxy-As steroids are excreted by normal infants and infants with a 21-hydroxylase defect, the finding of large amounts of these steroids in urine obtained from an abnormal infant is insufficient evidence for a diagnosis of 3P-HSD deficiency. However, compounds 1, 4, 5, and 6 mentioned above were not detected in elevated amount in urine excreted by the infant previously mentioned with 21-hydroxylase deficiency, and the measurement of these compounds may be useful in the diagnosis of 3P-HSD deficiency. 11.1.3. llp-Hydroxylase Deficiency The first thorough investigation of the deficiency was made by Eberkin and Bongiovanni (E6). It is associated with virilization and hypertension, and low levels of aldosterone production have been reported (K10). ll-Deoxycortisol is found in excess in plasma and its tetrahydro derivative is the major urinary steroid ; urinary tetrahydro-ll-deoxy-

196

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

corticosterone is elevated, but pregnanetriol only moderately so. The major urinary 17-05 is etiocholanolone. 11.1.4. Lipoid Adrenal Hyperplasia The disorder is due to a deficiency in the cholesterol side chain-splitting enzyme responsible for converting cholesterol to pregnenolone (P14). The adrenal cortex is hyperplastic and yellow due to accumulated cholesterol. Adrenal insutliciency is indicated clinically and incomplete masculinization of male fetuses is due to the inadequacy of testosterone production by the testes in early fetal life. A mild form of the disorder has been described by Camacho e t al. (C2). 11.2. CONGENITAL ADRENAL HYPOPLASIA Acute adrenal insufficiency in infancy may be due to congenital adrenal hypoplasia. This may be in association with hypoplasia or absence of the pituitary gland (B20, B35, E7, M19, R1) or as isolated congenital adrenal hypoplasia (B33, C4, M17, R15). Cathro and Coyle (C4) have described the second pregnancy of a woman who had already given birth to an infant with adrenal hypoplasia. The first infant died of respiratory failure 18 hours after birth and postmortem examination revealed very small adrenal glands but no other detectable endocrine abnormality. During the latter part of the second pregnancy, the maternal estriol excretion was low (1 mg/24 hours), an expected finding since l&-OH-DHA produced by the fetus is an important precursor of estriol (D6, D9) (Section 4.1.1.). Evidence for adrenocortical insufficiency in the infant was obtained from the low level of cortisol in the umbilical cord blood. The infant was satisfactorily treated a t birth with steroid therapy. The first indication of possible adrenal hypoplasia in an infant studied by the authors was the finding of an abnormally low level of estriol in plasma obtained from the mother prior t o delivery. The urinary excretion of lGa-OH-DHA, 16-0x0-androstenediol, and 1&-OH-pregnenolone was only about one-tenth that expected for normal newborn infants. This infant died of respiratory failure 30 hours after birth; postmortem examination showed very small adrenal glands. Adrenal cortical hypoplasia in two male siblings, who both died in the neonatal period, has been reported by Boyd and McDonald (B33). Mitchell and Rhaney (M17) also observed two male siblings with adrenal hypoplasia, one died a t the age of 2 months and the second was treated successfully with fluorocortisol.

STEROID METABOLISM IN EARLY INFANCY

11.3. DEFECTS IN

THE

197

PRODUCTION OF ALDOSTERONE

Deficiency in the 18-oxidation of corticosterone was first described by Visser and Cost (V8) in three related infants with salt-losing syndrome. A defect in the conversion of corticosterone to aldosterone is not associated with adrenal hyperplasia since cortisol production is not affected a,nd no excess corticotropin is produced. The patients studied by Visser and Cost (V8) had low serum sodium concentrations with high serum potassium. The total excretion of 17-05 and cortisol metabolites was normal. However, no aldosterone was detected in the urine and abnormally large amounts of 17-deoxy CZ1steroids were excreted (corticosterone, 11-dehydrocorticosterone, tetrahydrocorticosterone, allotetrahydrocorticosterone, and tetrahydro-11-dehydrocorticosterone). Ulick and co-workers ( U l ) have also studied an infant with the salt-losing syndrome associated with defective production of aldosterone. These workers found increased production of corticosterone, 18-OH-corticosterone and 11-deoxycorticosterone and postulated that the defect was in the oxidation of 18-OH-corticosterone to aldosterone. The excess production of 11-deoxycorticosterone and corticosterone by infants with an 18-oxidation defect is probably due to an excess production of angiotensin, the stimulant for aldosterone production (M20). Renin, produced by the juxtaglomerular cells in the kidneys, initiates the formation of angiotensin, and a loss of salt causes this substance t o be produced in excessive amounts. Evidence for this has been obtained by Ulick et al. ( U l ) , who found an increased amount of renin in the plasma of their patients during sodium depletion experiments. A defect in the synthesis of aldosterone in infancy may not be permanent. Royer et al. (R18) reported two siblings with a salt-losing syndrome who responded well to 11-deoxycorticosterone acetate ; when one of the children was reinvestigated after 5 years, it was found that the aldosterone secretion rate was normal. Two infants studied by Russell and co-worker (R19) had salt loss in infancy and were treated with salt and 11-deoxycorticosterone acetate, but it was possible to discontinue the treatment of one infant a t the age of 12 months and the other after 3% years. It has been suggested by Visser (V7) that two types of hypoaldosteronism exist in infancy: permanent hypoaldosteronism due to a deficiency in the 18-oxidation of corticosterone, and transient hypoaldosteronism caused by delayed maturation of the zona glomerulosa after birth. 12.

The Control of Steroid Production in Infancy

There is little doubt that the infant adrenal responds to corticotropin stimulation and stress with an increased production of cortisol, though

198

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

Klein and Hanson (K6) found a reduced eosinophile cell count response in the first week of life compared with that found during the second week, and Colle and Ulstrom (C14) and Colle et al. (C15) could demonstrate no increase in urinary Porter-Silber chromogens or cortisol in response to corticotropin or surgical stress. However, using the method of Nelson and Samuels (N3), Bertrand et al. (B12) showed a good response to corticotropin by newborn and older infants, and the blood cortisol level, measured by double isotope dilution has been found to rise markedly during the stress of exchange transfusion of infants (B43). The rise was not due to donor blood, as its cortisol and corticotropin content was low. Cathro e t al. (C5) found an above normal level of cortisol metabolites in the urine of some stressed babies and of one infant subjected to exchange blood transfusion. Hillman and Giroud (H9) obtained a good response in blood cortisol and cortisone levels (assayed by double isotope dilution techniques) in 9 babies aged 16-74 hours, after the injection of corticotropin, and the corticotropin stimulation of newborn infants has also been shown to cause an increase greater than 600% in the urinary excretion of cortisol (Table 17). The use of modern techniques has shown that soon after birth the infant adrenal is fully capable of responding to a stimulus of stress, presumably operating through a functional hypothalamic-pituitary-adrenal axis, and during early life cortisol is produced a t a rate relatively higher than that in normal adult life (Section 7.3). The inability of TABLE 17 URINARY STEROIDEXCRETION IN 5 NEWBORN INFANTS AVERAGED FOR THE FIRST 3 POSTNATAL DAYSWITH No TREATMENT, AND FOR DAYS4, 5, A N D 6, DURING WHICH CORTICOTROPIN WAS ADMINISTERED"

Steroid 2 1-OH-pregnenolone Androstenediol (1701) 16-0x0-androstenediol 16a-OH-DHA 16a-OH-pregnenolone Androstenetriol Cortisol Tetrahydro compounds*

Average, days 1, 2, 3 ( d 2 4 hr) 280

<340 1850 1370 1710 1060 < 170
Average, days 4, 5, 6, on corticotropin ( d 2 4 hr) 510 380 2940 3430 4860 1410 > 1250 > 1200

Percent increase 82

> 11 59 150 185 32 >600 > 1000

Lauritzen et al. (L7). A group of unidentified blue tetrazolium staining compounds with the chromatographic mobility of THE and THF.

STEROID METABOLISM I N EARLY INFANCY

199

earlier workers to demonstrate any response to stimuli may be due to the type of assay procedures used. Work mentioned in Section 4.2.3. has indicated that corticotropin is involved in cortisol production in the fetus, but three reasons are there given why the control of production may be complex. At birth two of the three conditions mentioned suffer dramatic change; i.e., cortisol can no longer be obtained across the placental barrier from the maternal circulation, and the abundant supply of progesterone from which cortisol might be synthesized is cut off. To maintain control in the face of changes such as these, the mechanism must surely be robust. The control of estrogen and progesterone production by the placenta in pregnancy is not understood but since the precursors of the estrogens at least are found in the fetus, it is likely that the primary control of their production resides there. In early independent life with the absence of the placenta these precursors are excreted in the urine either unchanged, 16a-hydroxylated or otherwise metabolized, and it is reasonable to assume that their production proceeds by the same mechanisms as in utero. EIarly work with the administration of corticotropin to infants showed an increase in the excretion of DHA (L5, L6), and recently, with the same stimulation a range of urinary steroids has been measured (L7). The results, summarized in Table 17, show a relatively small increase (occasionally a decrease) in the 3 P - h y d r 0 ~ y - n steroids, ~ with a very considerable increase in cortisol and its metabolites. Allowing for the peculiarities in the catabolism of cortisol by infants, and if urinary excretion is taken as a measure of production, it is evident that to produce the normal pattern of steroids found in infant urine and also in utero (a large amount of A5 steroid with a much smaller quantity of cortisol metabolites), a different or more complex mechanism of control must obtain. It is of interest to compare the increases in the 3P-hydroxy-A5 steroids under corticotropin stimulation with those found by various workers for the major group of urinary steroids not emanating to any extent from cortisol, i.e., the 17-0s. The results found for this group vary considerably, the average increases being from 35 to 350%, but on the whole the rate of excretion of the 17-0sin later life seems to be about as sensitive to corticotropin stimulation as are the 3P-hydroxy-A5 steroids in infancy. It is thus possible that corticotropin has some influence on the production of estrogens and possibly progesterone in pregnancy. To support this, a considerable increase has been reported in the urinary excretion of estriol by women in late pregnancy after the infusion of corticotropin whereas no such increase was obtained in cases of intrauterine fetal death ( D l ) . In addition, when corticoids are given to the mother there is a marked

200

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

lowering in the level of DHA and 16a-OH-DHA in cord blood (S22), indicating (if the production of these steroids is under the control of corticotropin) that the administered steroids have crossed the placenta and have suppressed the production of corticotropin in the fetus. Corticotropin has been shown to stimulate steroidogenesis by third-trimester fetal adrenals (B21). It has been suggested (E4, S9) that the explanation for the large quantity of A5 steroids produced after birth lies in their incidental formation by corticotropin stimulation, during a somewhat inefficient synthesis of cortisol via hydroxylated A5 intermediates (see Section 4.2.3). Some other explanation must be found, however, since under corticotropin stimulation the efficiency of cortisol, relative to A5 steroid production is apparently considerably improved. Chorionic gonadotropin (HCG) may be involved since it has recently been shown (Table 18) that its administration for therapeutic purposes to newborn male infants with undescended testicles produces a selective increase in the excretion of DHA with no apparent effect upon the excretion of other steroids. It is possible that the DHA is produced by stimulation of the testes. A total of 7 babies was studied, but stimulation was obained only in the 4 whose results are given in Table 18. 13.

Concluding Remarks

During the early weeks and months of human life after birth, steroid metabolism and excretion are rapidly changing, and the results of assays on blood and urine obtained using techniques designed for work with TABLE 18

URINARYSTEROIDEXCRETION IN 4 NEWBORN INFANTS, AVERAGED FOR POSTNATAL DAYS1-3

WITH N O TREATMENT, AND FOR DAYS4, 5, AND 6, DURlNG WHICHHUMANCHORIONIC GONADOTROPIN WAS ~ M I N I S T E R E D ~

Steroid Pregnenolone

DHA 21-OH-pregnenolone Androstenediol 16-Oxc-androstenediol I&-OH-DHA 166-OH-DHA 16a-OH-pregnenolone Androstenetriol 0

Lauritzen et al. (L8).

Average, days 1, 2, 3 (pg/24 hours)

Average, days 4, 5, 6, on HCG ( d 2 4 hours)

<24 <38

<24

156 95 545 645 380 953 378

74 1 152 105 600 730 298 1028 325

Percent increase 0

> 1850 -3 11 10 13 - 22 8 - 14

STEROID METABOLISM Ih' EARLY INFANCY

201

adults must be interpreted with considerable caution. Group assays such as those for 17-08 and 17-OHCS must be used only on the understanding that the contents of the groups are very different for infants. Special methodology is available and is especially valuable in the investigation of infants suspected of having a defect in steroid synthesis. Over the first 6 months of life, two patterns of change can be distinguished: for 4 or 5 days the newborn infant is disposing of the large amounts of steroids received in utero from the placenta, then until the sixth month, enzyme systems are still functional for the production of steroids associated with uterine existence. After the sixth month the childhood pattern of steroid metabolism has emerged, though for the estrogens a t least, the type of catabolism found in the newborn is still present. Knowledge of steroid metabolism during early life is far from complete, and i t is highly probable that some of the steroids present have a definite role in infant survival, and are not produced solely by the vestigial remains of a system of metabolism designed for uterine life. It is likely that further research will indicate new clinical applications for the investigation of already identified steroids, unidentified compounds, or the profiles of both, in amniotic fluid, umbilical cord blood, and infant urine.

REFERENCES Al. Aarskog, D., Stea, K. F., and Thorsen, T., Cortisol metabolism in the newborn. Actu Endocrinol. 46, 286-296 (1964). 112. Abramovich, D. R., and Wade, A. P., The origin and levels of 17-oxosteroids and 17-hydroxycorticoids in liquor amnii. Excerpta Med. Intern. Congr. Ser. 170, 28-29 (1968). A3. Aitken, E. H., Preedy, J. R. K., Eton, B., and Short, R. V., Estrogen and progesterone levels in fetal and maternal plasma at parturition. Lancet 11, 1096-1099 (1958). A4. Allen, W. M., A simple method for analysing complicated absorption curves, of use in the colorimetric determination of urinary steroids. J. Clin. Endocrinol. Metub. 10, 71-83 (1950). A5. Allen, W. M., In discussion of Hunt, A. B., and McConahey, W. M., Pregnancy associated with diseases of the adrenal glands. Am. J. Obstet. Gynecol. 66, 970-987 (1953). A6. Appleby, J. I., and Norymberski, J. K., The urinary excretion of 17-hydroxycorticosteroids in human pregnancy. J . Endom'nol. 16, 310-319 (1957). A7. Appleby, J. I., Gibson, G., Norymberski, J. K., and Stubbs, R. D., Indirect analysis of corticosteroids. 1. The determination of 17-hydroxycorticoids. Bwchem. J. 60, 453-467 (1955). B1. Baillie, A. H., and Griffiths, K., 38-Hydroxysteroid dehydrogenase in the fetal mouse Leydig cell. J. Endocrinol. 31, 63-66 (1964). B2. Baird, C. W., and Bush, I. E., Cortisol and cortisone content of amniotic fluid from diabetic and non-diabetic women. Acta Endocrinol. 34, 97-104 (1960).

202

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

B3. Barr, M., Diczfalusy, E., and Tillinger, K. G., Studies on estrogen metabolism in infants and children. Acta Endocrinol. 37, 241-252 (1961). B4. Baulieu, E.-E., and Dray, F., Conversion of Ha-dehydroisoandrosterone sulfate to Ha-estrogensin normal pregnant women. J . Clin. Endominol. Metab. 23, 1298-1301 (1963). B5. Baulieu, E.-E., and Robel, P., Androst-5-ene 3B, 17&diol-17or-3Hto testosterone17a-3H and 5or and 5P-androstane3a,l7p-di01-17a-~H in vivo. Steroids 2, 111-118 (1963). B6. Beas, F., Zurbriigg, R. P., Cam, J., and Gardner, L. I., Urinary CI9 steroids in normal children and adults. J . Clin. Endocrinol. Metab. 22, 1090-1094 (1962). B7. Beck, J. C., In discussion of Jost, A., The function of the foetal adrenal cortex. Ciba Found. Study Group 27, 22-27 (1967). B8. Bengtsson, G., Ullberg, S., Wiquist, N., and Diczfalusy, E., Autoradiographic studies on previable human foetuses perfused with radioactive steroids. Acta Endocrinol. 46, 544-551 (1964). B9. Benner, M. C., Studies on the involution of the fetal cortex of the adrenal glands. Am. J . Pathol. 16, 787-798 (1940). B10. Bergstrand, C. G., Adrenocortical function in the newborn and its disorders. Nord. Med. 64, 1030-1033 (1960). B11. Bertrand, J., Gilly, R., and Loras, B., Neonatal adrenal function. Plasma 17hydroxycorticosteroids in 102 infants from birth to 42 days of age. Ezcerpta Med. Intern. Congr. Ser. 61,279 (1962). B12. Bertrand, J., Loras, B., Gilly, R., and Cautenet, B., Contribution a ]’etude de la skcrktion et du metabolisme du cortisol chez le nouveau-rk et le nourrisson de moins de trois mois. Pathol. Biol., Semaine Hop. [ N . S.]11, 997-1022 (1963). B13. Bertrand, J., Cautenet, B., Ollagnon, C., ROUX,H., Loras, B., Forest, M., and de Peretti, E., Etude de la chromatographie sur papier des 17-C.S. urinaires chez le nourrisson, l’enfant, l’adolescent et l’adulte. Ann. Endocrinol. (Paris) 27, 3 7 4 6 (1966). B14. Bevan, B. R., Holton, J. B., and Lathe, G. H., The effect of pregnanediol and pregnanediol glucuronide on bilirubin Conjugation by rat liver slices. Clin. Sci. 29, 353-361 (1965). B15. Birchall, K., and Mitchell, F. L., Investigation of the Zimmermann reaction when used for the assay of groups of steroids in infants. Clin. Chim. Acta 12, 125-136 (1965). B 16. Birchall, K., and Mitchell, F. L., The identification of 21-hydroxypregnenolone in the urine of newborn infants. Steroids 6,427-436 (1965). B17. Birchall, K., Cathro, D. M., Forsyth, C. C., and Mitchell, F. L., Separation and estimation of adrenal steroids in the urine of newborn infants. Lancet I, 26-28 (1961). B18. Birchall, K., Cathro, D. M., Forsyth, C. C., and Mitchell, F. L., A method for the separation and estimation of neutral steroids in the urine of newborn infants. J . Endorrinol. 27, 31-51 (1963). B19. Birke, G., Diczfalusy, E., Plantin, L. O., Robbe, H., and Westman, A., Familial congenital hyperplasia of the adrenal cortex. Acta Endocrinol. 29, 55-69 (1958). B20. Blizzard, R. M., and Alberts, M., Hypopituitarism, hypoadrenalism, and hypogonadism in the newborn infant. J . Pediat. 48, 782-792 (1956). B21. Bloch, E., In vitro steroid synthesis by adrenals and gonads during mammalian fetal development. Ezcerpta Med. Intern. Congr. Ser. 111, 64-65 (1966). B22. Bloch, E., and Benirschke, K., Synthesis in vitro of steroids by human fetal adrenal gland slices. J . B i d . Chem. 234, 1085-1089 (1959).

STEROID METABOLISM I N EARLY INFANCY

203

B23. Bloch, E., and Benirschke, K., Steroidogenic capacity of foetal adrenals in vitro. In “The Human Adrenal Cortex” (A. R. Currie, T. Symington, and J. K. Grant, eds.), pp. 589-595. Livingstone, Edinburgh and London, 1962. 1324. Boltb, E., Wiquist, N., and Dicsfalusy, E., Metabolism of dehydroepiandrosterone and dehydroepiandrosterone sulphate by the human foetus a t midpregnancy. Acta Endocrinol. 62, 583-597 (1966). 1325. Boltk, E., Mancuso, S., Eriksson, G., Wiquist, N., and Dicsfalusy, E., Studies on the aromatisation of neutral steroids in pregnant women. A d a Endocrinol. 46, 576-599 (1964). B26. Bongiovanni, A. M., The detection of pregnanediol and pregnanetriol in the urine of patients with adrenal hyperplasia. Suppression with cortisone. (A preliminary report.) Bull. Johns Hopkins Hasp. 92, 244-251 (1953). B27. Bongiovanni, A. M., Unusual steroid pattern in congenital adrenal hyperplasia: Deficiency of 36-hydroxydehydrogenase. Clin. Endocrinol. 21, 86&862 (1961). B28. Bongiovanni, A. M., The adrenogenital syndrome with deficiency of 3P-hydroxysteroid dehydrogenase. J . Clin. Invest. 41, 2086-2092 (1962). B29. Bongiovanni, A. M., and Eberlein, W. R., Critical analysis of methods for measurement of 3a,17a,2Oa-pregnanetriol in human urine. Anal. Chem. 30, 388-393 (1958). B30. Bongiovanni, A. M., and Root, A. W., The adrenogenital syndrome. New Engl. J. Med. 268, 1283-1289, 1342-1351, and 1391-1399 (1963). B31. Bongiovanni, A. M., Eberlein, W. R., Goldman, A. S., and New, M., Disorders of adrenal steroid biogenesis. Recent Progr. Homnone Res. 23, 375-449 (1967). B32. Bongiovanni, A. M., Eberlein, W. R., Westphal, M., and Boggs, T., Prolonged turnover rate of hydrocortisone in the newborn infant. J . Clin. Endocrinol. Metab. 18, 1127-1130 (1958). B33. Boyd, J. F., and MacDonald, A. M., Adrenal cortical hypoplasia in siblings Arch. Disease Childhood 36, 561-568 (1960). B34. Breuer, H., Occurrence and determination of the newer estrogens in human urine. In “Research on Steroids” (C. Cassano, ed.), Vol. 1, pp. 133-148. I1 Pensiero ScientXco, Rome, 1964. B35. Brewer, D. B., Congenital absence of the pituitary gland and its consequences. J . Pathol. Bacterial. 73, 59-67 (1957). B36. Brooks, R. V., The metabolism of 17~~-hydroxyprogesterone in congenital adrenal hyperplasia. J . Endocrinol. 21, 277-286 (1960). B37. Bro-Rasmussen, F., Buus, O., and Trolle, D., Ratio cortisone/cortisol in mother and infant at birth. Acta Endocrinol. 40, 579-583 (1962). B38. Bro-Rasmussen, F., BUUS,O., Lundwall, F., and Trolle, D., Variations in plamsa cortisol during pregnancy, delivery and the puerperium. Acta Endocrinol. 40, 571-578 (1962). B39. Brown, J. B., The relationship between urinary oestrogens and oestrogens produced in the body. J . Endocrinol. 16, 202-212 (1957). B40. Burstein, S., and Lieberman, S., Hydrolysis of ketosteroid hydrogen sulfates by solvolysis procedures. J . Bwl. Chem. 233, 331-335 (1958). B41. Bush, 1. E., and Mahesh, V. B., Adrenocorticol hyperfunction with sudden onset of hirsutism. J . Endocrinol. 18, 1-25 (1959). B42. Butler, G. C., and Marrian, G. F., The isolation of pregnane -3, 17, 20-trio1 from the urine of women showing the adrenogenital syndrome. J . Bid. Chem. 119, 565-572 (1937). B43. Buus, O., Bro-Rasmussen, F., Trolle, D., and Lundwall, F., Response of the

204

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

adrenal cortex during exchange transfusion and maintenance of the low cortisol/ cortisone ratio in the neonatal period. Acta Endocrinol. 62, 7-16 (1966). C1. Calman, K. C., Baillie, A. H., Ferguson, M. M., and McK. Hart, D., Hydroxysteroid dehydrogenases in the human adrenal cortex. J . Endocrinol. 34, 439-446 (1966). C2. Camacho, A. M., Kowarski, A., Migeon, C. J., and Brough, A. J., Congenital adrenal hyperplasia due to a deficiency of one of the enzymes involved in the biosynthesis of pregnenolone. J . Clin. Endocrinol. 28, 153-161 (1968). C3. Cmsmer, O., Hormone production of the isolated human placenta. Acta Endocrinol. Suppl. 46 (1959). C4. Cathro, D. M., and Coyle, M. G., Adrenocortical function in newborn infants of low birth weight. Ezcerpta Med. Intern. Congr. Ser. 132, 688-694 (1966). C5. Cathro, D. M., and Forsyth, C. C., Excretion of corticosteroids by infants of diabetic and pre-diabetic mothers. Arch. Disease Childhood 40, 583-592 (1965). C6. Cathro, D. M., Birchall, K., Mitchell, F. L., and Forsyth, C. C., The excretion of neutral steroids in the urine of newborn infants. J . Endocrinol. 27,53-75 (1963). C7. Cathro, D. M., Birchall, K., Mitchell, F. L., and Forsyth, C. C., 3j3:21-Dihydroxypregn-5ene-20-one in urine of normal newborn infants and in third day urine of child with deficiency of 3j3-hydroxysteroid-dehydrogenase. Arch. Disease Childhood 40, 251-260 (1965). C8. Cavallero, C., and Magrini, U., Histochemical studies on 36-hydroxysteroiddehydrogenase and other enzymes in foetal steroid-secreting structures. Ezcerpta Med. Intern. Cmgr. Ser. 111, 63-64 (1966). C9. Cawley, L. P., Musser, B. O., Faucette, W., Beckloff, S., and Learned, H., Evaluation of hydrolytic products of 17-ketosteroids by means of gas-liquid chromatography. Clin. Chem. 11, 1009-1018 (1965). C10. Clayton, B. E., Edwards, R. W. H., and Renwick, A. G. C., Adrenal function in children. Arch. Disease Childhood 38, 49-53 (1963). C11. Cleary, R. E., and Pion, R. J., Urinary excretion of 16a-hydroxydehydroepiandrosterone and 16-keto-androstenediol during the early neonatal period. J . Clin. Endorrinol. Metab. 28, 372-378 (1968). C12. Col&s, A., and Heinrichs, W. L., Pettenkofer chromogens in the maternal and fetal circulations: Anencephalic pregnancies, Cesarean sections, and tentative identification of 3j3,17j3-dihydroxyandrost-5-en-16-onein umbilical cord blood. Steroids 6, 753-764 (1966). C13. Col&s, A., Heinrichs, W. L., and Tatum, H. J., Pettenkofer chromogens in the maternal and fetal circulations: Detection of 3j3,16a-dihydroxyandrost5-en-17-one in umbilical cord blood. Steroids 3, 417434 (1964). C14. Colle, E., and Ulstrom, R. A., Urinary excretion of corticosteroids in infants: The response to surgery and corticotropin (ACTH). Am. J . Diseases Children 98, 574-575 (1959). C15. Colle, E., Ulstrom, R. A., Burley, J., and Gunville, R., Adrenocortical steroid metabolism in newborn infants. 111. Urinary excretion of 17-Hydroxycorticosteroids following major surgery and stimulation with adrenocorticotropin. J . Clin. Endocrinol. Metab. 20, 1215-1224 (1960). C16. Conrad, S. H., Pion, R. J., and Kitchin, J. D., Pregnenolone sulfate in human pregnancy plasma. J . Clin. Endocrinol. Metab. 27, 114-119 (1967). C17. Cranny, R. L., and Cranny, C. L., The urinary excretion of 17-Hydroxycorticosteroids by premature infants. Am. J . Diseases Children 99, 344-348 (1960). C18. Cranny, R. L., Kirschvink, J. F., and Kelly, V. C., The half-life of hydrocortisone in normal newborn infants. Am. J . Diseases Children 99, 437-443 (1960).

STEROID METABOLISM IN EARLY INFANCY

205

D1. Diissler, C.-G., Der Einfluss von corticotrophin auf die &trogenausscheidung in der Schwangerschaft und bei intrauterinem Fruchttod. Acta Endocrinol. SS, 401-402 (1966). D2. Daughaday, W. H., Steroid protein interactions. Physiol. Rev. 39,885-902 (1959). D3. David, R. R., Bergada, C., and Migeon, C . J., Isolation, identification and measurement of 3a, 17cu-dihydroxypregnane-ll,2O-dione in congenital adrenal hyperplasia. J . Clin. Endom’nol. Metab. 26, 322-326 (1965). D4. Degenhart, H. J., Visser, H. K. A., Wilmink, R., and Croughs, W. Aldosterone and cortisol secretion rates in infants and children with congenital adrenal hyperplasia suggesting different 21-hydroxylation defects in “salt losers” and in “nonsalt losers.” Acta Endocrinol. 48, 587-601 (1965). D5. Degenhart, H. J., Visser, H. K. A., Wilmink, R., and Frankena, L., Production and excretion of testosterone in children with congenital adrenal hyperplasia and precocious puberty. A d a Endocrinol. Suppl. 100, 51 (1965). D6. Dell’Acqua, S., Mancuso, S., Benagiano, G., Wiquist, N., and Diczfalusy, E., Androgen metabolism in the human foebplacental unit at mid-term. Excerpta Med. Intern. Congr. Ser. 111, 60 (1966). D7. De Moor, P., Heirwegh, K., Heremans, J. F., and Declerck-Raskin, M., Protein binding of corticoids studied by gel filtration. J . Clin. Invest. 41, 816-827 (1962). D8. Diczfalusy, E., and Magnusson, A.-M., Tissue concentration of estrone, estradiol176, and estriol in the human fetus. A d a Endocrinol. 28, 169-185 (1958). D9. Diczfalusy, E., Pion, R., and Schwers, J., Steroid biogenwis and metabolism in the human feto-placental unit a t midpregnancy. Arch. Anat. Microswp. Morphol. Exptl. 64, 67-83 (1965). DIO. Diczfalusy, E., Tillinger, K.-G., and Westman, A., Studies on estrogen metabolism in newborn boys. 1. Excretion of estrone, estradiol-17p and estriol during the first few days of life. Acta Endocrinol. 26, 303-312 (1957). D 11. Diczfalusy, E., Tillinger, K.-G., and Westman, A., Studies on estrogen metabolism in new-born boys. 11. A study of the metabolism of exogenous estradiol-178. Acta E n M n o l . 26, 313-321 (1957). D12. Diczfalusy, E., Cassmer, O., Alonso, C., and de Miguel, M., Estrogen metabolism in the human fetus and newborn. Recent Progr. Hormone Res. 17, 147-206 (1961). D13. Dobriner, K., and Lieberman, S., The metabolism of steroid hormones in humans. Ciba Found. Colloq. Endocrinol. 2, 381-417 (1952). D14. Dodgson, K. S., and Spencer, B., A preliminary account of the glycosulfatase of Littorina littorea. Biochem. J . 67, 310-315 (1954). D15. Dorfman, R. I., and Sharma, D. C., An outline of the biosynthesis of corticosteroids and androgens. Steroids 6 , 229-236 (1965). D16. Drayer, N. M., and Giroud, C. J. P., Corticosteroid sulfates in the urine of the human neonate. Steroids 6, 289-317 (1965). D17. Driscoll, S. G., and Hsia, D. Y.-Y., The development of enzyme systems during early infancy. Pediatrics 22, 785-845 (1958). D18. Ducharme, J. R., Leboeuf, G., and Sandor, T., C21 steroid secretion in the premature infant. Excerpta Med. Intern. Congr. Ser. 111, 293 (1966). El. Eagle, J. F., Rept. 13th M . and R . Pediat. Res. Conf., Syracuse, New York, 1954 In discussion, p. 26. M. & R. Lab., Columbus, Ohio, 1955. E2. Easterling, W. E., Jr., Simmer, H. H., Dignam, W. J., Frankland, M. V., and Naftolin, F., Dehydroepiandrosterone sulfate, 16a-hydroxydehydroepiandrosterone and 16a-hydroxydehydroepiandrosteronesulfate in maternal and fetal blood of pregnancies with anencephalic and normal fetuses. Steroids 8, 157-178 (19f36).

206

FREDERICK L. MITCHELL AND CEDRIC H . L. SHACKLETON

E3. Eberlein, W. R., The salt-losing form of congenital adrenal hyperplasia. Paediatrim 21, 667-672 (1958). E4. Eberlein, W. R., Steroids and sterols in umbilical cord blood. J . Clin. Endocrinol. Metab. 26, 1101-1118 (1965). E5. Eberlein, W. R., Fetal adrenal function in congenital adrenal hyperplasia. Ezcerpta Med. Intern. Congr. Ser. 111, 66 (1966). E6. Eberlein, W. R., and Bongiovanni, A. M., Plasma and urinary corticosteroids in the hypertensive form of congenital adrenal hyperplasia. J . Biol. C h m . 223, 85-94 (1956). E7. Ehrljch, R. M., Ectopic and hypoplastic pituitary with adrenal hypoplwia. J . Pediat. 61, 377-384 (1957). E8. Exley, D., Ingall, S. C., Norymberski, J. K., and Woods, G. F., Indirect analysis of corticosteroids. 5) The determination of 17-Deoxycorticosteroids. Bwchem. J . 81, 428-434 (1961). F1. Finkelstein, M., von Euw, J., and Reichstein, T., Isolierung von 3a,17,20aTrioxy-pregnanon-( 11) aus pathologischem menschlichen Ham. Helv. Chim. Acta 36, 1266-1277 (1953). lSa,17pF2. Fishman, J., and Guzik, H., Synthesis of estra-l,3,5(10)-trien-3,15a, tetrol. A new metabolite of estradiol. Tetrahedron Letters 30, 2929-2932 (1967). F3. Flood, C., Layne, D. S., Ramscharan, S., Rossipal, E., Tait, J. F., and Tait, S. A. S., An investigation of the urinary metabolites and secretion rates of aldosterone and cortisol in man and a description of methods for their measurement. A d a Endocrinol. 36, 237-264 (1961). F4. Francis, F. E., Shen, N.-H. C., and Kinsella, R. A., Jr., Enteric excretion of metabolites of steroid hormones in the human. ii. Isolation of As-androsten3p-OL-17-one from meconium. J . Biol. Chem. 236, 1957-1959 (1960). F5. Frandsen, V. A., The site of production of oestrogenic hormones in human pregnancy. European Rev. Endocrinol. I, 227-246 (1965). F6. Frandsen, V. A., and Stakemann, G., Production of oestrogenic hormones in human pregnancy. Acta Endoerinol. 38, 383-391 (1961). F7. Frandsen, V. A., and Stakemann, G., The site of production of oestrogenic hormones in human pregnancy. 111 Further observations on the hormone excretion in pregnancy with anencephalic foetus. Acta Endocrinol. 47, 265-276 (1964). F8. Fukushima, D. K., Bradlow, H. L., Hellman, L., Zumoff, B., and Gallagher, T. F., Metabolic transformation of hydrocortisone-4-C14in normal men. J . Biol. Chem. 236, 2246-2252 (1960). F9. Fukushima, D. K., Bradlow, H. L., Hellman, L., Zumoff, B., and Gallagher, T. F., Study of 17-Hydroxyprogesterone-4-C14 in man. J . Clin. Endocrinol. Metab. 21, 765-778 (1961). GI. Gibson, G., and Norymberski, J. K., A note on the rapid assay of 17-ketogenic steroids in urine. Ann. Rheumatic Diseases 13, 59 (1954). G2. Goldman, A. S., Yakovac, W. C., and Bongiovanni, A. M., Development of activity of 3p-hydroxysteroid dehydrogenase in human fetal tissues and in two anencephalic newborns. J . Clin. Endocrinol. Metab. 26,14-22 (1966). G3. Goldman, A. S., Bongiovanni, A. M., Yakovac, W. C., and Prader, A., Study of A6,3p-hydroxysteroid dehydrogenase in normal hyperplastic and neoplastic adrenal cortical tissue. J . Clin. Endocrind Metab. 24, 894909 (1964). G4. Greig, M., A study of progesterone metabolism in pregnancy. Ph. D. Thesis, University of St. Andrews, Scotland (1965). G5. Greig, M., Coyle, M. G., Cooper, W., and Walker, J., Plasma progesterone in

STEROID METABOLISM IN EARLY INFANCY

207

the mother and fetus in the second half of human pregnancy. J . Obstet. Gynecol. Brit. Commonwealth 69, 772-776 (1962). G6. Gurpide, E., Schwers, J., Welch, M. T., Vande Wiele, R. L., and Lieberman, S., Fetal and maternal metabolism of estradiol during pregnancy. J. Clin. Endoerinol. Metab. 26, 1355-1365 (1966). G7. Gustafsson, J.-i., Shackleton, C. H. L., and Sjovall, J., Faecal excretion of steroids by infants during the first year of life. Proc. 7th Acta Endocrinol. Congr., Stockholm, 1969. HI. Hagen, A. A., and Eik-Nes, K. B., Testosterone biosynthesis in the canine in vivo following infusion of l7a-hydroxy (4-14C)progesterone and (17a-aH) dehydroepiandrosterone via the spermatic artery. Biochim. Biophys. Acta 90, 593-599 (1964). H2. Hagen, A. A., Barr, M., and Diczfalusy, E., Metabolism of 17fl-estradi014-14C in early infancy. Acta Endocrinol. 49, 207-220 (1965). H3. Halban, J., Schwangerschaftsreaktionen der fotalen Organe und ihre puerperale Involution. 2.Geburtshilfe Gynaekol. 63, 191-234 (1904). H4. Harbert, G. M., McGaughey, H. S., Gcoggin, W. A., and Thorton, W. N., Concentration of progesterone in newborn and maternal circulations a t delivery. Obstet. Gynecol. 23, 413426 (1964). H5. Harkness, R. A., and Love, D. N., Studies on the estimation of urinary pregnanetriol during pregnancy and childhood. A d a Endocrinol. 61, 526-534 (1966). H6. Hellig, H., Lefebvre, Y., Gattereau, D., and Bolt6, E., Placental progesterone synthesis in humans. Ezcerpta Med. Intern. Cmgr. Ser. 170, 8 (1968). H7. Hems, B. A., Adrenocortical steroids (synthesis of natural compounds and some analogues). Brit. Med. Bull. 18, 93-98 (1962). H8. Henry, R., and Thevenet, M., Action du suc digestif d’Helix pomatia L. sur les diffbrents st6roides conjugu6s urinaires. Bull. SOC.Chim. Biol. 34, 897-899 (1952). H9. Hillman, D. A., and Giroud, C . J. P., Plasma cortisone and cortisol levels at birth and during the neonatal period. J . Clin. Endocrinol. Metab. 26, 243-248 (1965). H10. Homing, E. C., Gardiner, W. L., and Brooks, C. J. W., Gas chromatographic and gas chromatographic-mass spectrometric procedures for the study of human steroid hormones and their metabolites. Ez~erptaMed. Intern. Congr. Ser. 132, 197-201 (1967). 11. Ittrich, G., Eine methode zur chemischen Bestimmung von Ostrogenen Hormonen in Blut, Milch und Colestrurn. 2. Physiol. Chem. 320, 103-110 (1960). 12. Iwamiya, M., Tweedie, F. J., and Solomon, S., Quoted by Solomon, S., Formation and metabolism of neutral steroids in the human placenta and fetus. J . Clin. Endocrinol. Metab. 26, 762-772 (1966). JL. Jaffe, R. B., and Peterson, E. P., I n situ placental perfusion with cholesterol7a-3H. Steroids 8, 695-710 (1966). 52. Jailer, J. W., Gold, J. J., Vande Wiele, R., and Lieberman, S. 17a-Hydroxyprogesterone and 2l-desoxyhydrocortisone; their metabolism and possible role in congenital adrenal virilism. J. Clin. Invest. 34, 1639-1646 (1955). 53. James, V. H. T., Corticosteroid secretion by human placenta and foetus. European J . Steroids 1, 1-14 (1966). 54. Janoski, A. H., Roginsky, M. S., Christy, N. P., and Kelly, W. G., Production of 16a-hydroxy steroids in the salelosing form of congenital adrenal hyperplasia (CAH). Ezcerpta Med. Intern. Congr. Ser. 111, 297 (1966). 55. Jayle, M. F., and Baulieu, E. E., Donn6es experimentales sup l’hydrolyse des 17-c6tost&oides conjugub urinaires par les diastases du suc digestif d’Helk pomatia. Bull. SOC.Chim. Bwl. 36, 1391-1406 (1954).

208

FREDERICK L. MITCHELL AND CEDRIC H . L. SHACKLETON

J6. Jeffcoate, T. N. A., and Scott, J. S., Polyhydramnios and oligohydramnios. Can. Med. Assoc. J . 80, 77-86 (1959). 57. Jeffcoate, T. N. A., Fliegner, J. R. H., Russel, S. H., Davis, J. C., and Wade, A. P., Diagnosis of the adrenogenital syndrome before birth. Lancet 11, 553-555 (1965). 58. JirBsek, J. E., The relationship between the structure of the testis and differentiation of the external genitalia and phenotype in man. Ciba Found. Colloq. Endocrinol. 16, 3-27 (1967). J9. Jost, A., Testicular structure in hypogonadal states (remarks concerning the testes of male pseudohermaphrodites). Exmpta Med. Intern. Congr. Ser. 83, 1291-1292 (1964). K1. Kenny, F., Malvaux, P., Preeyasombat, C., Spaulding, J., and Migeon, C. J., Cortisol production rate in the perinatal period: Normal newborns and abnormal conditions. Excerptu Med. Intern. Congr. Ser. 111, 175 (1966). K2. Kitchin, J. D., Pion, R. J., and Conrad, S. H., Metabolism of progesterone by the term human placenta perfused in situ. Steroids 9, 263-274 (1967). K3. Kjellberg, S. R., and Rudhe, U., Fetal renal secretion and its significance in congenital deformities of ureters and urethra. A& Radiol. 31, 243-249 (1949). K4. Klein, G. P., and Giroud, C. J. P., Sulfation of corticosteroids by the adrenal of the human newborn. Steroids 6, 765-772 (1963). K5. Klein, G. P., and Giroud, C. J. P., Incorporation of 7craH-pregnenolone and 4-14C-progesteroneinto corticosteroids and some of their ester sulfates by human newborn adrenal slices. Steroids 9, 113-134 (1967). K6. Klein, R., and Hanson, J., Adrenocortical function in the newborn infant as measured by adrenocorticotropic hormone-Eosinophil response. J . Pediut. 6, 192-196 (1950). K7. Klein, R., Fortunato, J., and Papadatos, C., Free blood corticoids in the newborn infant. J . Clin. Invest. 33, 35-37 (1954). K8. Klopper, A., MacNaughton, M. A., and Michie, E. A., Oestriol and pregnanediol excretion during pregnancy and labour. XIV. J. Endocrinol. 22, XIV-XV (1961). K9. Klopper, A., Michie, E. A., and Brown, J. B., Method for the determination of urinary pregnanediol. J . Endocrinol. 12, 209-219 (1955). K10. Kowarski, A., Finkelstein, J. W., Spaulding, J. S., Holman, G. H., and Migeon, C. J., Mdosterone secretion rate in congenital adrenal hyperplasia. A discussion of the theories on the pathogenesis of the salt-losing form of the syndrome. J . Clin. Invest. 44, 1505-1513 (1965). L1. Lambert, M., and Pennington, G. W., The estimation of polar steroids in liquor amnii. J . Endocrinol. 32, 287-293 (1965). L2. Lanman, J. T., An interpretation of human foetal adrenal structure and function. I n “The H m n Adrenal Cortex” (A. R. Currie, T. Symington, and J. K. Grant, eds.), pp. 547-558. Livingstone, Edinburgh and London, 1962. L3. Lathe, G. H., and Walker, M., An enzyme defect in human neonatal jaundice and in Gunn’s strain of jaundiced rats. Biochem. J . 67, 9P (1957). L4. Lauritzen, C., Physiological effects, concentration and metabolism of chorionic gonadotropins and steroid hormones in the newborn. Excerpta Med. Intern. Congr. Ser. 111, 294 (1966). L5. Lauritzen, C., Action of HCG and steroids on physiological processes in the neonatal period. In “Research on Steroids” (C. Cassano, ed.), Vol. 2, pp. 109-119. I1 Pensiero Scientifico, Rome, 1966. L6. Lauritzen, C., and Lehmann, W.-D., Ausscheidung von Dehydroepiandrosteron in Neugeborenenhorn. Stimulienung durch Choriongonadotropin und ACTH. Arch. -kol. 200, 699-709 (1965).

STEROID METABOLISM IN EARLY INFANCY

209

L7. Lauritzen, C., Shackleton, C. H. L., and Mitchell, F. L., The effect of exogenous corticotropin on steroid excretion in the newborn. Acta Endocrinol. 68, 655-663 (1968). L8. Lauritzen, C., Shackleton, C. H. L., and Mitchell, F. L., The effect of exogenous human chorionic gonadotropin on steroid excretion in the newborn. A d a En&crinol. 61, 83-88 (1969). L9. Leon, Y. A., and Bulbrook, R. D., A modification of a method for the determination of urinary pregnanetriol. J . Endocrinol. 20, 236-245 (1960). L10. Leon, Y. A., Bulbrook, R. D., and Corner, E. D. S., Steroid sulfatase, arylsulfatase and p-glucuronidase in the molluscs. Biochem. J. 76, 612-617 (1960). L l l . Levell, M. J., Mitchell, F. L., Paine, C. G., and Jordan, A., The clinical value of urinary 17-ketogenic steroid determinations. J. Clin. Pathol. 10, 72-79 (1957). L12. Levitz, M., Conjugation and transfer of fetal-placental steroid hormones. J. Clin. Endocrinol. Metab. 26, 773-777 (1966). lJ3. Leyssac, P., Hydrocortisone in foetal plasma following intravenous administratior of hydrocortisone to the mother. Part I. With special reference to the binding of hydrocortisone by plasma proteins. Acta Obstet. Gynewl. Scand. 40, 174-180 (1961). L14. Leyssac, P., Hydrocortisone in foetal plasma following intravenous administration of hydrocortisone to the mother. Part 11. With special reference to foetal elimination of hydrocortisone. A& Obstet. Gynewl.Scand. 40, 181-186 (1961). 1’15. Lieberman, S., Steroid sulfates as biosynthetic intermediates. Excerpta Med. Intern. Congr. Ser. 132, 22-36 (1967). L16. Lieberman, S., and Dobriner, K., The isolation of pregnanediol-3a, 17-one-20 from human urine. J . Biol. C h . 161, 269-278 (1945). L17. Loras, B., Cautenet, B., Do, F., Forest, M., de Peretti, E., Saez, J. M., and Bertrand, J., Corticosterone secretion rate in infants and children before and during salt depletion. Excerpta Med. Intern. Congr. Ser. 111, 175 (1966). M1. Magendantz, H. G., and Ryan, K. J., Isolation of an estriol precunor, 16nhydroxydehydroepiandrosterone,from human umbilical sera. J. Clin. Endocrinol. Metab. 24, 1155-1162 (1964). M2. Mancuso, S., Benagiano, G., Dell’Acqua, S., Shapiro, M., Wiquist, N., and Diczfalusy, E., Aromatization and hydroxylation products formed by previable fetuses perfused with androstenedione and testosterone. Acta Endocrinol. 67, 208-227 1968). M3. Maner, F. D., Saffan, B. D., Wiggins, R. A., Thompson, J. D., and Preedy, J. R. K., Interrelationship of estrogen concentrations in the maternal circulation, fetal circulation and maternal urine in late pregnancy. J. Clin. Endocrinol. Metab. 23, 445-458 (1963). M4. Marsh, C. A., Inhibition of 8-glucuronidase by endogenous saccharate during hydrolysis of urinary conjugates. Nature 194, 974-975 (1962). M5. Mattox, V. R., Hayles, A. B., Salassa, R. M., and Dion, F. R., Urinary Steroid patterns and loss of salt in congenital adrenal hyperplasia. J . Clin. Endocrinol. 24, 517-527 (1964). M6. McIntosh, A. D., The human fetal adrenal: A morphological and histochemical study with comment on the problem of function. Scot. Med. J . 6, 242-249 (1960). M7. Meyer, C. J., Increases of free cortisol in plasmas of patients with toxemia of pregnancy. Steroids Suppl. 1, 199-205 (1965). M8. Meyer, C. J., Krahenbuhl, C., and Desaulles, P. A., Etude comparative des effects sexuels des androgknes naturels. Acta Endocrinol. 43, 27-46 (1963).

210

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

M9. Michie, E. A., Oestrogen levels in urine and amniotic fluid in pregnancy with live anencephalic foetus and the effect of intra-amniotic injection of sodium dehydroepiandrosterone sulphate on these levels, Acta Endocrinol. 61, 535-542 (1966). M10. Migeon, C. J., Cortisol production and metabolism in the neonate. J . Pediat. 66, 280-295 (1959). M11. Migeon, C. J., Bertrand, J., and Gemaell, C. A., The transplacental passage of various steroid hormones in mid-pregnancy. Recent Progr. Hormne Res. 17, 207-248 (1961). M12. Migeon, C. J., Keller, A. R., and Holmstrom, E. G., Dehydroepiandrosterone, androsterone and 17-hydroxycorticosteroid levels in maternal and cord plasma in cases of vaginal delivery. Bull. Johns Hopkins Hosp. 97,4 1 5 4 2 1 (1955). M13. Migeon, C. J., Keller, A. R., Lawrence, B., and Shepard, T. H., Dehydroepiandrosterone and androsterone levels in human plasma. Effect of age and sex; day to day and diurnal variations. J . Clin. Endocrinol. Metab. 17, 1051-1062 ( 1957). M14. Migeon, C. J., Bertrand, J., Wall, P. E., Stempfel, R. S., and Prystowsky, H., Metabolism and placental transmission of hydrocortisone during pregnancy near term. Am. J . Diseases Children 94, 502-504 (1957). Ml5. Mills, 1. H., in discussion of Prader, A., The adrenal cortex in childhood and adolescence. Ciba Found. Study Group 27, 38 (1967). M 16. Mitchell, F. L., Steroid metabolism in the fetoplacental unit and in early childhood. Vitamins Hormones 26, 191-269 (1967). M17. Mitchell, R. G., and Rhaney, K., Congenital adrenal hypoplasia in siblings. Lancet I, 488492 (1959). M18. Mizuno, M., Lobotsky, J., Lloyd, C. W., Kobayashi, T., and Murasawa, Y., Plasma androstenedione and testosterone during pregnancy and in the newborn. J . Clin. Endocrinol. Metab. 28, 1133-1142 (1968). M19. Mosier, H. D., Hypoplasia of the pituitary and adrenal cortex (report of occurrence in twin siblings and autopsy findings). J . Pediat. 48, 633-639 (1956). M20. Mulrow, P. J., and Ganong, W. F., The role of the renin-angiotensin system in the regulation of aldosterone secretion in the dog and man. In “Aldosterone” (E. E. Baulieu and P. Rohel, eds.), pp. 265-278. Blackwell, Oxford, 1964. M21. Myers, F. L., and Northcote, D. H., A survey of the enzymes from the gastrointestinal tract of Helix pomatia. J . Ezptl. Biol. 36, 639-648 (1958). N1. Nakajima, T., Staemmler, H.J., and Lipp, C., Untersuchungen Uber die Pregnandiolausschcidung Neugeborener Knaben. Klin. Wochschr. 38, 389-391 (1960). N2. Natoli, A., Nicolosi, G., and Magliocca, R., Relations between the adrenal cortex and pregnancy: Urinary elimination of Zimmermann and Pettenkofer reacting ketosteroids during various months of pregnancy. Quaderni Clin. Ostet. Ginecol. 12, 31-44 (1957). N3. Nelson, D. H., and Samuels, L. T., A method for the determination of 17-hydroxycorticosteroids in blood : 17-hydroxycorticosterone in the peripheral circulation. J . Clin. Endocrinol. Metab. 12, 519-526 (1952). N4. New, M. I., and Peterson, R. E., A new form of congenital adrenal hyperplasia. J. Clin. Endocrinol. Metab. 27, 300-305 (1967). N5. Niemi, M., Ikonen, M., and Hervonen, A., Histochemistry and fine structure of the interstitial tissue in the human foetal testis. Ciba Found. Colloq. Endocrinol. 16, 31-52 (1967). N6. Norymberski, J. K., Methods of group corticosteroid estimation in urine. 11. In “The Adrenal Cortex” (G. K. McGowan and M. Sandler, eds.), pp. 88-109. Pitman, New York, 1961.

STEROID METABOLISM IN EARLY INFANCY

21 1

N7. Norymberski, J. K., Stubbs, R. D., and West, H. F., Assessment of adrenocortical activity by assay of 17-ketogenic steroids in urine. Lancet I, 1276-1281 (1953). 01. Oertel, G. W., and Eik-Nes, K. B., Dehydroepiandrosterone in blood. Cleavage of dehydroepiandrosterone conjugates by phosphatases. A eta Endocrinol. 28,293-299 (1958). 02. Oppenheimer, E. H., Lesions in the adrenals of an infant following maternal corticosteroid therapy. Bull. Johns Hopkins Hosp. 114, 146-151 (1964). 03. Osinski, P. A., Steroid 11/3-01 dehydrogenase in human placenta. Nature 187, 777 (1960). P1. Palmer, R., Eriksson, G., Wiquist, N., and Diczfalusy, E., Studies on the metabolism of C-21 steroids in the human foeto-placental unit. Ada Endocrinol. 62, 598-606 (1966). P2. Pasqualini, J. R., and Diczfalusy, E., Metabolism of corticosterone sulphate in the human foeto-placental unit a t mid-pregnancy. Excerpta Med. Intern. Congr. Ser. 111, 288 (1966). P3. Pasqualini, J. R., and Diczfalusy, E., Cortisol metabolism in the human foetoplacental unit a t mid-pregnancy. Excerpta Med. Intern. Congr. Ser. 170, 30 (1968). P4. Pasqualini, J. R., and Jayle, M. F., Identification of 3&21-diiydroxy-5-pregnen2O-one disulfate in human urine. J . Clin. Invest. 41, 981-987 (1962). P5. Pasqualini, J. R., Dutter, F., and Jayle, M. F., Mode de conjugaison du pregna 5 e n e 3/3,21-diol-20-one e t du pregna 5-ene 3/3,17&?1-triol 20-one chez l’homme. Identification des ester-sulfates correspondants. Biochim. Bwphys. A cta 69, 331-336 (1963). P6. Pasqualini, J. R., Cedard, L., Nguyen, B. L., and Alsatt, E., Differences in the activity of human term placenta sulphatasw for steroid ester sulphates. Biochim. Biophys. Acta 139, 177-179 (1967). P7. Pasqualini, J. R., Lowy, J., Wiquist, N., and Diczfalusy, E., Biosynthesis of cortisol from 3/3,17a,2l-trihydroxypregn-5-en-20-one by the intact human foetus at midpregnancy. Biochim. Biophys. Acta 162, 648-650 (1968). P8. Paulsen, E. P., Sobel, E. M., and Shafran, M. S., Urinary steroid metabolites in children. 1.Individual 17-ketosteroids in children with normal sexual development. J . Clin. Endocrinol. Metab. 26, 329-339 (1966). P9. Peterson, R. E., Wyngaarden, J . B., Guerra, S. L., Brodie, B. B., and Brunim, J. J., The physiological disposition and metabolic fate of hydrocortisone in man. J . Clin. Invest. 34, 1779-1794 (1955). E’10. Philipp, E., and Soetbeer, M., Die Ausscheidung der 17-Keto-steroide im Harm des Neugeborenen (ein Beitrag zur Funktion der Fetalen Nebenniere). Med. Welt 20, 301-304 (1951). I’ll. Pion, R. J., Conrad, S. K., and Wolf, B. J., Pregnenolone sulfate-an efficient precursor for the placental production of progesterone. J . Clin. Endocrinol. Melab. 26, 225-226 (1966). P12. Pion, R. J., Jaffe, R., Eriksson, G., Wiquist, N., and Diczfalusy, E., Studies in the metabolism of C-21 steroids in the human foeto-placental unit. Acta Endocrinol. 48, 234-248 (1965). P13. Potter, E. L., “Pathology of the Fetus and Infant,” p. 326. Year Book Publ., Chicago, Illinois, 1961. P14. Prader, A., and Siebenmann, R. E., Nebenniereninsuffizienz bei kongenitaler Lipoidhyperplasie der Nebennieren. Helv. Puediut. Acta 12, 569-595 (1957). R1. Reid, J. D., Congenital absence of the pituitary gland. J . Pediat. 56, 65S-664 (1960).

212

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

R2. Reynolds, J. W., The isolation of 16-ketoandrostenediol from the urine of a newborn infant). Steroids 3, 77-83 (1964). R3. Reynolds, J. W., Excretion of two A6-3B-OH1 16a-hydroxysteroids by normal infants and children. J . Clin. Endocrinol. Metab. 26, 416-423 (1965). R4. Reynolds, J. W., The excretion of two A~-3~-OH,16a-hydroxysteroids by patients with congenital adrenal hyperplasia. Pediatrics 36, 583-591 (1965). R5. Reynolds, J. W., Studies of neonatal adrenal steroid metabolism. Ezcerpta Med. Intern. Cmgr. Ser. 111, 152 (1966). R6. Reynolds, J. W., Excretion of l6a-hydroxysteroids by premature infants. J . Clin. Endocrinol. Metub. 26, 1251-1256 (1966). R7. Reynolds, J. W., The identification and quantification of AK-androstenetriol isolated from the urine of premature infants. Steroids 8, 719-727 (1966). Rg. Reynolds, J. W., Colle, E., and Ulstrom, R. A., Adrenocortical steroid metabolism in newborn infants. V. Physiologic disposition of exogenous cortisol loads in the early neonatal period. J . Clin. Endocrinol. Metab. 22, 245-254 (1962). R9. Rice, B. F., Johanson, C. A., and Sternberg, W. H., Formation of steroid hormones from Acetate-1-14C by a human fetal testis preparation grown in organ culture. Steroids 7 , 79-90 (1966). R10. Rivarola, M. A., Forest, M. G., and Migeon, C. J., Testosterone, androstenedione and dehydroepiandrosterone in plasma during pregnancy and a t delivery: Concentration and protein binding. J . Clin. Endocrinol. Metab. 28, 3 4 4 0 (1968). R11. Roberts, K. D., Vande Wiele, R. L., and Lieberman, S., The conversion in vivo of dehydroisoandrosterone sulfate to androsterone and etiocholanolone glucuronidates. J . Biol. Chem. 236, 2213-2215 (1961). R12. Roberts, K. D., Bandi, L., Calvin, H. I., Drucker, W. D., and Lieberman, S., Evidence that steroid sulfates serve as biosynthetic intermediates. IV. Conversion of cholesterol sulfate in viva to urinary Clo and CZlsteroidal sulfates. Biochemistry 3, 1983-1988 (1964). R13. Robertson, M. E., Stiefel, M., and Laidlaw, J. C., The influence of estrogen on the secretion, disposition and biologic activity of cortisol. J . Clin. Endocrinol. Metab. 19, 1381-1398 (1959). R14. Roby, C . C., Ober, W. B., and Drorbaugh, J. E., Pregnanediol excretion in the urine of newborn male infants. Pediatrics 17, 877-881 (1956). R15. RoseIli, A., and Barbosa, L. T., Congenital hypoplasia of the adrenal glands. Report of two cases in sisters with necropsy. Pediatrics 36, 70-75 (1965). R16. Roy, A. B., The steroid sulfatase of Patella vulgata. Biochem. J . 62, 41-50 (1956). R17. Roy, E. J., The concentration of estrogens in maternal and fetal blood obtained a t Caesarian section, and the effect of hospitalization on maternal blood estrogen levels. J . Obstet. Gynaecol. Brit. Commonwealth 69, 196-202 (1962). R18. Royer, P., Lestradet, H., de Menibus, C. H., and Vermeil, G., Hypoaldostkronisme familial chronique B dkbut nkonatal. Ann. Pediat., Semaine Hop. 8, 133-138 (1961). R19. Russell, A., Levin, B., Sinclair, L., and Oberholrer, V. G., A reversible salt-wasting syndrome of the newborn and infant. Possible infantile hypoaldosteronism. Arch. Disease Childhood 38, 313-325 (1963). S1. Sachaekaja, T., Inffuence of adrenocorticotropic hormone on the biosynthesis of A4-3-unsaturated steroids in the adrenals of human foetuses. Excerpta Med. Intern. Cmgr. Ser. 111, 287-288 (1966). 52. Sandberg, A. A., and Slaunwhite, W. R., Jr., Transcortin: A corticosteroid binding protein of plasma. 11. Levels in various conditions and the effects of estrogens. J . Clin. Invest. 38, 1290-1297_(1959).

STEROID METABOLISM IN EARLY INFANCY

213

S3. Sandberg, A. A., and Slaunwhite, R. R., Parameters affecting steroid metabolism in pregnancy. Excerpta Med. Intern. Congr. Ser. 111, 68-69 (1966). S4. Schindler, A. E., and Ratanasopa, V., Profile of steroids in amniotic fluid of normal and complicated pregnancies. Ada Endoerinol. 69, 239-248 (1968). 55. Schindler, A. E., and Siiteri, P. K., Isolation and quantitation of steroids from normal human amniotic fluid. J . Clin. Endocrinol. Metab. 28, 1189-1198 (1968). S6. Schuller, E., Corticosteroid-Auscheidung in der Normden Graviditat. Acta Endocrinol. 26, 345-364 (1957). S7. Schwers, J., Gurpide, E., Vande Wiele, R. L., and Lieberman, S., Urinary metabolites of estradiol and estriol administered intr%amniotically. J . Clin. Endocrinol. Metub. 27, 1403-1408 (1967). S8. Shackleton, C. H. L., Steroid metabolism in utero and in the neonatal period. Ph.D. Thesis, University of Edinburgh, Scotland (1969). S9. Shackleton, C. H. L., and Mitchell, F. L., The measurement of 38-hydroxy-As steroids in human fetal blood, amniotic fluid, infant urine and adult urine. Steroids 10, 359-385 (1967). [Note publisher’s errors, corrections given in Steroids 11, 415 (1968).] S10. Shackleton, C. H. L., Charro-Salgado, A. L., and Mitchell, F. L., Urinary neutral steroid profile analysis in adults and infants. Clin. Chim. Acta 21, 105-118 (1968). S11. Shackleton, C. H. L., Livingstone, J. R. B., and Mitchell, F. L., The conjugated 17-hydroxy epimers of A~-androstene-3@-17diol in infant and adult urine and umbilical cord plasma. Steroids 11, 299-311 (1968). 512. Shackleton, C. H. L., Livingstone, J. R. B., Michie, E. A., and Mitchell, F. L., Arterio-venous blood and amniotic fluid levels of individual 38-hydroxy-As steroids. Excerpta Med. Intern. Congr. Ser. 170, 28 (1968). 513. Shackleton, C. H. L., Kelly, R. W., Adhikary, P. M., Brooks, C. J. W., Harkness, R. A., Sykes, P. J., and Mitchell, F. L., The identification, measurement of a new steroid, 16~-hydroxydehydroepiandrosterone in infant urine. Steroids 12, 705-718 (1969). S14. Shahwan, M. M., Oakey, R. E., and Stitch, S. R., A new pathway of oestriol biosynthesis in pregnancy. Excerpta Med. Intern. Congr. Ser. 167, 172 (1968). 515. Sharma, D. C., and Dorfman, R. I., Studies on in vivo C-21 hydroxylation of progesterone and 17a-hydroxyprogesterone.Federalion Proc. 22, 530 (1963). 816. Short, R. V., The chemical determination of progesterone in peripheral blood. J. Endocrinol. 16, 415-425 (1958). S17. Short, R. V., Progesterone. I n ‘LHormonmin Blood” (C. H. Gray and A. L. Bacharach, eds.), p. 379. Academic Press, New York, 1961. 818. Siiteri, P. K., and MacDonald, P. C., The utilization of circulating dehydroisoandrosterone sulfate for estrogen synthesis during human pregnancy. Steroids 2, 713-730 (1963). 819. Siber, R. H., Estimation of hydrocortisone secretion. Method of calculation from ufinary excretion d a b . Clin. Chem. 1, 234-240 (1955). 820. Simmer, H. H., Easterling, W. E., Jr., Pion, R. J., and Dignam, W. J., Identification of dehydroepiandrosterone sulfate in fetal blood and quantification of this hormone in cord arterial, cord venous and maternal peripheral blood in normal pregnancies at term. Steroids 4, 125-135 (1964). 821. Simmer, H. H., Frankland, M. V., Greipel, M., and Zlotnik, G., Isolation and identification of unconjugated dehydroepiandrosteroneand 16a-hydroxydehydroepiandrosterone from pooled cord blood. Steroids 11, 13-26 (1968). 822. Simmer, H. H., Dignam, W. J., Easterling, W. E., Jr., Frankland, M. V., and

214

FREDERICK L. MITCHELL AND CEDRIC H. L. SHACKLETON

Naftolin, F., Neutral Clp-steroids and steroid sulfates in hnman pregnancy. Steroids 8, 179-193 (1966). S22a. Sjovall, J., and Vihko, R., Analysis of solvolyzable steroids in human plasma by combined gas chromatography-mass spectrometry. Acta Endocrinol. 67, 247-260 (1968). 823. Slaunwhite, W. R., Jr., and Sandberg, A. A., Transcortin: A corticosteroidbinding protein of plasma. J . Clin. Invest. 38, 3&1-391 (1959). 524. Solomon, S., Formation and metabolism of neutral steroids in the human placenta and fetus. J . Clin. Endocrinol. Metab. 26, 762-772 (1966). S25. Solomon, S., Bird, C. E., Wilson, R., Wiquist, N., and Diczfalusy, E., Progesterone metabolism in the fetalplacental unit. Excerpta M e d . Intern. Congr. Ser. 83, 721-726 (1964). S26. Starkel, S., and Wegrzynowski, L., Beitrag zur Hitologie der Nebenniere bei Feten und Kindern. Arch. Anat. Physiol. pp. 214-236 (1910). 527. Stempfel, R. S., I n discussion following “Metabolism and placental transmission of hydrocortisone during pregnancy, near term,” by Migeon, C. J., Bertrand, J., Wall, P. E., Stempfel, R. S., and Prystovsky, H., Am. J . Diseases Children 94, 502-504 (1957). 528. Stern, M. I., A method for the determination of 5p-pregnane-3a:17a:2Oa-triol in urine. J. Endocrinol. 16, 180-188 (1957). S29. Stitch, S. R., and Halkerston, I. D. K., Hydrolysis of steroid conjugates in urine. Biochem. J . 63, 710-715 (1956). S30. Stitch, S. R., Halkerston, I. D. K., and Hillman, J., Sulfatase and p-glucuronidase activity of molluscan extracts. Biochem. J. 63, 705-710 (1956). S31. Swinyard, C. A., Growth of suprarenal glands. Anat. Record 87, 141-150 (1943). TI. Tahka, H., On the weight and structure of the adrenal glands and the factors affecting them, in children of 0-2 years. Acta Paediat. Suppl. 40, 81 (1951). U1. Ulick, S., Gautier, E., Vetter, K. K., Markello, J. R., Yaffe, S., and Lowe, C. U., An aldosterone biosynthetic defect in a salt-losing disorder. J . Clin. Endocrinol. Metab. 24, 669-672. (1964). U2. Uktrom, R. A., Colle, E., Burley, J., and Gunville, R., Adrenocortical steroid metabolism in newborn infants. I. Urinary excretion of free and conjugated 17-hydroxycorticosteroids in normal full-term infants. J . Clin. Endocrinol. Metab. 20, 1066-1079 (1960). U3. Ulstrom, R. A., Colle, E., Burley, J., and Gunville, R., Adrenocortical steroid metabolism in newborn infants. 11. Urinary excretion of 6p-hydroxycortisol and other polar metabolites. J. Clin. Endocrinol. Metab. 20, 1080-1094 (1960). U4. Ulstrom, R. A., Colle, E., Reynolds, J. W., and Burley, J., Adrenocortical steroid metabolism in newborn infants. IV. Plasma concentrations of cortisol in the early neonatal period. J . Clin. Endocrinol. Metab. 21, 414-425 (1961). Vl. Vande Wiele, R. L., MacDonald, P. C., Gurpide, E., and Lieberman, S., Studies on the secretion and interconversion of the androgens. Recent Progr. Hormone Res. 19, 275-310 (1963). V2. Vestergaard, P., Urinary excretion of individual 17-ketosteroids in children. Acta Endodnol. 49, 436-442 (1965). V3. Villee, C. A., and Loring, J. M., Synthesis of steroids in the newborn human adrenal in vitro. J . Clin. Endocrinol. Metab. 26, 307-314 (1965). V4. Villee, D. B., The role of progesterone in the development of adrenal enzymes. Excerpta Med.Intern. Congr. Ser. 111, 65-66 (1966). V5. Villee, D. B., and Driscoll, S. G., Pregnenolone and progesterone metabolism in human adrenals from twin female fetuses. Endocrinology 77, 602-608 (1965).

STEROID METABOLISM IN EARLY INFANCY

215

V6. Villee, D. B., Engel, L. L., and Villee, C. A., Steroid hydroxylation in human fetal adrends. Endocrinology 66, 465-474 (1959). V7. Visser, H. K. A., The adrenal cortex in childhood. Part 2. Pathological aspects. Arch. Disease Childhood 41, 113-136 (1966). V8. Visser, H. K. A+, and Cost, W. S., C21-Corticosteroid excretion pattern in a familiar salblosing syndrome. A new enzyme defect in the biosynthesis of aldoskrone. Acta Endocrinol. Suppl. 89, 31-32 (1964). W1. Wallace, E. Z., and Carter, A. C., Studies on the mechanism of the plasma 17hydroxycorticosteroid elevation induced in man by estrogens. J. Clin. Inoest. 39, 601-605 (1960). W2. Warren, J. C., and Timberlake, C. E., Biosynthesis of estrogens in pregnancy: Precursor role of plasma dehydroisoandrosterone. Obstet. Gynecol. 23, 689498 (1964). W3. Wengle, B., Distribution of some steroid sulphokinases in foetal human tissues. Acta Endocrinol. 62, 607-618 (1966). W4. Wiener, M., and Allen, S. H. G., Inhibition of placental steroid synthesis by steroid metabolites: Possible feedback control. Steroids 9, 567-582 (1967). W5. Wood, M. E., and Gray, C . H., The urinary excretion of neutral 17-ketosteroids in childhood. J . Endocrinol. 6, 111-119 (1949). Y1. Yudaev, N. A., and Druzhinina, K. V., The role of pregnenolone and progesterone in the biosynthesis of corticosteroids. Excerpta Med. Intern. COngT. Ser. 111, 248 (1966). Z1. Zander, J., Relationship between progesterone production in the human placenta and the fetus. Ciba Found. Study Group 9, 32-39 (1961). 22. Zander, J., Progesterone and its metabolites in the placental-foetal unit. Excerpta Med. Intern. Congr. Ser. 83, 715-720 (1964). 23. Zander, J., and Solth, K., Die Ausscheidung der Czl-Steroide Bei Neugeborenen. Klin. Wochschr. 31, 317-321 (1953). 24. Zander, J., von Munstermann, A.-M., and Runnebaum, B., Steroide im plasma von menschlichem plazentablut (nabelschnurplasma). A cta Endocrinol. 41, 507-520 (1962). 25. Zucconi, G., Lisboa, B. P., Simonitsch, E., Roth, L., Hagen, A. A., and Diczfalusy, E., Isolation of l5a-hydroxy-estriol from pregnancy urine and from the urine of newborn infants. Acta Endocrinol. 66, 413-423 (1967). Z6. Zucconi, G., Simonitsch, E., Lisboa, B. P., Roth, L., Hagen, A. A., and Dicnfalusy, E., A new estrogen metabolite formed by the human fetus and newborn. Excerpta Med. Intern. Congr. Ser. 111, 291 (1966).

This Page Intentionally Left Blank

THE USE OF GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY Harold V. Street Department of Forensic Medicine, University of Edinburgh, Edinburgh, Scotland 1. Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................ 2. Basic Requirements of G iquid Chromatography. 2.1. Carrier Gas. . . . . . . .......................................... 2.2. Sample Introduction. ................................. ..... 2.3. Columns. ............................. ....................... 2.4. Detectors.. . . . . . . ...................................... 3. Column Preparation in d Chromatography.. . . . . . . . . . . . . . . . . . . . . 3.1. Solid Supports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............. 3.2. Liquid Phases.. . . ...................................... 3.3. Deactivation of D us Supports and Preparation of Columns. 4. Applications of Gas-Liqui matography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Steroid Hormones. ....................................... 4.2. Catecholamines, T nes, and Other Biological Amines. . . . . . . . . . 4.3. Amino Acids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Carbohydrates.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Toxicological Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Miscellaneous Applications.. . . . . . . . . . . . .

.

R.eferences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.

217 218 219 220 220 221 22 1 22 1 224 225 237 237 249 256 268 278 290 297

Introduction

At the present time, gas chromatography (GC) has a rather limited function in its application to clinical chemistry. There are two main reasons for this limitation. Many of the substances which are required to be determined in clinical chemistry are ionic in nature and are, therefore, not sufficiently volatile, a t the temperatures a t which present gas chromatographs operate, to allow of their analysis. by GC. I n certain cases it is, of course, possible to form volatile derivatives, e.g., in the case of amino acids. Furthermore, the volatility of nonionic compounds whose molecular weight is greater than about 500 is low enough to cause difficulties to be encountered, mainly because of the relatively high temperatures and carrier gas flow rates necessary for the compounds to elute in a reasonable time. However, it is interesting to note that Subbaram (515) claims to have separated the palmitic and stearic esters of cholesterol (molecular weights 624 and 652, respectively) using temperatures up to 217

218

HAROLD V. STREET

340°C. Even metalloporphyrins have been separated in a high pressure (1830 psi) column by Klesper et al. (K7) using dichlorodifluoromethane as carrier gas, although, in this case, the compounds were not eluted but were located in the column after the “run.” Thus it may be that ultimately the limiting factor may be not the size of the molecule, but rather the stability (in the carrier gas) of the compounds to be analyzed a t very high pressures and high temperatures. Even so, gas-liquid chromatography (GLC) is proving to be an increasingly important analytical tool in clinical chemistry although it is probably true to say that many clinical chemistry laboratories are not a t present making as much use as they could of the technique. Perhaps one of the reasons for this stems from the fact that commercially prepared columns do not always perform adequately the function which their manufacturers claim. There is also a certain amount of ignorance on the part of some users who, for example, purchase a packed column, fit this into their apparatus, and find it not to function properly. They most probably have not realized that it is necessary to “optimize” the conditions of the detector to suit the carrier-gas flow rate. But they then condemn the use of GLC for that particular purpose, and this condemnation spreads by geometric progression to workers in other laboratories. I n other instances, many workers do not realize the paramount importance of the part which the forces of adsorption play in GLC. In the writer’s opinion, these workers are unfortunately retarding to quite a great extent the rate of development of the use of GLC in clinical chemistry laboratories. The purpose of this chapter is to show how GLC can be used in the clinical chemistry laboratory and to mention some of the most recent work in this field. Several techniques are described in detail, and particular attention is paid to column preparation. For a comprehensive review of recent developments (up to the beginning of 1968) in gas chromatography, the reader is referred to the publications of Juvet and Dal Nogare (D2, D3, J9, JlO). 2.

Basic Requirements of Gas-Liquid Chromatography

A schematic diagram of the basic requirements of gas chromatography is shown in Fig. 1, which is reproduced from the chapter by Goldbaum e t at. ( G 5 ) . The mixture to be separated is introduced into a stream of carrier gas and is thereby swept into the column, where separation of the components of the mixture is effected. The emerging components are detected, and the results are presented graphically on a recorder a s detector response us. time. Ideally, Gaussian peaks are presented by the recorder. The area

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

219

FIG.1. Schematic diagram of gas chromatography. Reproduced from Goldbaum e t al. (G5) with permission. under these peaks is proportional to the amount of component in the detector over that time interval. Mechanical or electronic integrators are available to measure these areas. Alternatively, the peak height (which is proportional to the area of a symmetrical peak) may be used for quantitative evaluation, but significant errors may be introduced by this method if the peaks are not symmetrical. Unsymmetrical peaks may be encountered when attempting to separate polar compounds. 2.1. CARRIER GAS The type of detector to be employed determines the nature of the carrier gas which may be used. Argon is used with the argon ionization detector. Helium is used with flame-ionization, thermal conductivity, thermionic emission, and cross-section detectors. Hydrogen may be used in thermal conductivity detectors to give maximum sensitivity. Probably the commonest and cheapest carrier gas is nitrogen, which can be used with flame-ionization, electron capture, thermal conductivity, and crosssection detectors. Argon-methane mixtures may be used with electron capture detectors. The manufacturer’s instructions regarding flow rates of carrier gas and the “optimization” of gas flows to, say, the flame ionization detector, are usually satisfactory, but the following points, while seemingly rather obvious, are nevertheless very important if the apparatus is to function properly. a. The pressure of the carrier gas as indicated on the cylinder gauge must be sufficiently high to allow the flow controller to function properly. b. The flow rate of the carrier gas should be measured whenever a new carrier gas cylinder is fitted. C. The column must be tight enough not to allow any gas leaks. This is especially important a t high input pressures with tightly packed metal

columns.

220

HAROLD V. STREBT

d. The hydrogen and air flows to the flame-ionization detector must be “optimized” whenever the carrier gas flow rate is altered.

2.2. SAMPLE INTRODUCTION For introduction into the gas chromatograph, the sample may be solid, liquid, or gas. Solids are introduced by the capsule or metal-grid technique or by means of specially constructed syringes. Liquids are injected with microliter syringes through a septum in the apparatus. For gases, special airtight syringes or systems of sampling valves are used. Solids may be dissolved in a suitable solvent, which is then injected as a liquid.

2.3. COLUMNS

It is often stated that the column is the heart of the gas chromatograph. This is indeed true, and it is column preparation (which is discussed in greater detail in another section) that largely determines whether ‘‘good” or “bad” results are obtained-this is especially true in the gas chromatographic analysis of polar compounds, The column, of glass or metal tubing, usually measures 6-12 feet in length and from 1/8 to 1/4 inch in diameter. Open tubular columns (capillary columns) and packed capillary columns have been described in a very useful publication by Ettre (E4).These types of columns will not be considered in this chapter. The “packed column” is filled with a support material which usually consists of a diatomaceous earth or glass beads. This support, which must have a large surface area, is coated with a liquid phase. The different types of supports and liquid phases are described in a subsequent section. The mixture of compounds to be separated is carried, after introduction into the gas chromatograph, by the carrier gas along the column. Here, the compounds (in the vapor phase) are “partitioned” between the liquid and vapor phases and, depending on their partition coefficients (and certain adsorption forces, e.g., van der Waals, hydrogen bonding) are resolved to a greater or lesser degree. These resolved (or separated) compounds are presented to the detector, which presents an electronic signal to the recorder (via an electrometer), which in turn displays the separated components of the original mixture as a series of (ideally) Gaussian curves. These are actually graphs of detector response versus time. The area under each of these Gaussian curves is proportional to the amount of substance passing through the detector over the time interval required for presentation of the particular curve. The time interval measured from the time of injection of the sample to the time a t which the curve maximum occurs is called the retention time ( t , ). These curves are referred to by workers in the field of gas chromatography as peaks. A useful note on the preliminary recommendations on nomencla-

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

221

ture and presentation of data in gas chromatography has been prepared by a special group of workers of the International Union of Pure and Applied Chemistry under the Chairmanship of Dr. D. Ambrose (A6) of the National Chemical Laboratory, Teddington, Middlesex, England. 2.4. DETECTORS There are two main types of detectors in common use. These are the thermal conductors and the ionization detectors. I n general these detectors have been fully described in previous reports (B27, K1, L10, L11). Other types of detector have been described recently. An interesting development in this connection is the use of a flame photometric detector (J11, Zl) for the determination of transition metal halides, chelates, and organic compounds. Braman (B18) has described flame emission and dual flame emission-flame ionization detectors for use in conjunction with gas chromatography. Detection sensitivity was in the microgram range. An improvement in the electron-capture detector was reported by Simmonds e t al. (S6), who replaced the usual tritium source by a nickel-63 source. This modification permits safe operation up to 400°C and yet maintains the general performance features of the parallel plate detector. A microwave emission detector has been devised (M14) using argonhelium carrier gas which, it is claimed, has improved sensitivity and selectivity for several organophosphorus pesticides plus p,p’-DDT, lindane, and 2-iodobutane. A novel type of detector has been proposed, (V5a) which achieves chemical selectivity by making use of the principles involved in polarography. Very recently, an extensive study has been made by Price e t al. (P10) of the application of a photoionization detector in gas chromatographic systems. Possible applications of this type of detector suggested by these workers include studies of air pollution. Chromatograms are shown of air samples containing 5 ppm of petroleum ether using a sample size of only 1 ml. 3.

C o l u m n P r e p a r a t i o n in Gas-Liquid C h r o m a t o g r a p h y

3.1. SOLID SUPPORTS

I n 1963, Ottenstein (03) produced a very useful paper dealing with most of the important aspects of column support materials for use in gas chromatography. A more recent review by Palframan and Walker ( P l ) covers the diatomaceous supports, fluorocarbon supports, and various “specialist” supports including the porous polymer beads such a s Porapak, Polypak, and Chromosorb 102. The Johns-Manville Products Corporation, Celite Division (New York) issue very useful leaflets describing their various Chromosorb products, e.g., Chromosorbs A, G, P, T, W, and

222

HAROLD V. STREET

102. These leaflets give details of relevant physical properties of the supports, coating procedures, mesh sizes, etc. 3.1.1. White Diatomaceous Supports

These are prepared from flux-calcined diatomaceous earth, also called diatomite, diatomaceous silica, or kieselguhr. This earth is composed of the skeletons of diatoms, which are single-celled algae. It is excavated from deposits found in various parts of the world. The calcining is carried out by mixing the earth with a little sodium carbonate (called the flux), and raising the temperature to above 900°C. Various white diatomaceous earths are available commercially. The more common of these are Anakrom U, Celite 545, Chromosorb G (very robust material), Chromosorb W, Gas-Chrom CL, Gas-Chrom P, and Gas-Chrom S. 3.1.2. Pink Diatomaceous Supports

These are prepared by crushing the diatomaceous earth, pressing it into brick form, and then calcining it above 900°C. The characteristic pink color is probably due to the formation of an iron oxide from complete silicates. I n the case of the white diatomite, the absence of pink color is probably due to the reaction of the sodium carbonate with iron oxide to form a colorless complex sodium iron silicate. A number of pink diatomaceous supports are commercially available; the more common of these are Anakrom P, Chromosorb P, Diatomite S, and Gas-Chrom R. 3.1.3. Fluorocarbon Supports

The most inert of all the support materials are the fluoroethylene polymers (Teflon-6, Fluoropak-80), and the chlorotrifluoroethylene polymer (Kel-F). I n a comparison of various support materials, Sawyer and Barr (S3) found Teflon (Fluoropak-80) to be the least efficient. However, because of the inertness of fluorocarbon supports, they have proved useful in the analysis of aqueous samples (L2). 3.1.4. Porous Polymer Beads

These are the styrene-divinyl copolymer type (H10) and include Porapak, Polypak, and Chromosorb 10.2. Five different Porapaks are available: P, Q, R, S, and T, which differ in degree of cross-linking of styrene with ethylvinyl benzene. These polymers give sharp, symmetrical peaks and low retention volumes for water, alcohols and glycols, while less polar compounds are retarded (B28, H11). The beads are stable up to about 250°C. As they are generally used without liquid phase, there is no “bleed” from the column. Figure 2 shows the excellent results obtained

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

0 1 2 3 4 5 Time [minutes)

L

I

6

7

223

FIG.2. Separation of organic compounds from water. Conditions: 115”C, 60 ml of HZ per minute, column 6 foot X K6 inch, ethylvinylbenzene-divinylbenzene polymer. Reproduced from Hollis (H10) with permission.

by Hollis (H10) for the separation of a number of volatile organic compounds from water, using an ethylvinylbenzenedivinylbenzene polymer. Unfortunately, the capacity of the porous polymer columns is low, i.e., they are easily overloaded. Hollis (H10) found that the peak shape began to distort with about 0.2 pl per component (see example shown in Fig. 2 ) , although the peak shape remained fairly good up to 1 p1. However, a rapid recovery occurs following “flooding” with water, due to the nonsorptive nature of the material. This is particularly useful in analysis of trace constituents. 3.1.5. Other Solid Supports

Among other solid supports which have been proposed are glass beads (C2, D9, G3, L9, OZ),modified glass beads (D9, G4, K4, K5), the naturally occurring sterrasters (from marine sponges) (W3), Tide detergent iD7, D8, P9), and graphitized carbon on glass beads (H2). Recently, Vidal-Madjar and Guiochon (V10, V11) have pointed out that the full potential of gas-solid chromatography has not yet been revealed. They have produced some very interesting results by what they call gas-solid chromatography on organic crystals. Good resolution of a mixture of polynuclear aromatic hydrocarbons was obtained with copper phthalocyanine coated on graphitized carbon black. Palframan and Walker (P1) feel that the technological development of diatomaceous earths is probably approaching a limit. They suggest

224

HAROLD V. STREET

that, while there is room for further advances in the use of porous layer glass beads, the ideal support may ultimately be found in the development of solids of the cross-linked polystyrene type. At present, however, it appears that the best support materials are the diatomaceous earths. 3.2.

LIQUID

PHASES

It is probable that no more than half a dozen different liquid phases are needed to undertake resolution of any mixture of substances by GLC. But the number of liquid phases which have been suggested, many of which are available commercially, is almost legion. Table 1 contains a list of liquid phases which are currently available from Perkin-Elmer Ltd., and Table 2 is a list of polyester liquid phases which are available from the same suppliers. It is claimed that the polyesters listed in Table 2 have a thermal stability which is superior to that of the polyesters contained in Table 1. Probably the most widely used liquid phases are the silicones. These will be considered in more detail because many workers, even those who are “experienced” in the art of gas chromatography, seem to hold the idea that one and only one particular methyl silicone, for example, will effect resolution of the components of a particular mixture. While this may hold in a few isolated instances, it is certainly not a correct generalization. All methyl silicones have the following basic structure :

The greater the value of n , the higher the molecular weight and the greater the viscosity. As n increases, the compounds change from oils to gums. Cross-linking produces silicone rubbers, but these are not used as liquid phases in GLC. When the chain length is such that the molecular weight is of the order of 10,000, the vapor pressure of the gum is negligible a t 350°C, which is not far from the decomposition temperature. The “bleed” which is observed from columns containing these silicone liquid phases is generally due to contamination of the gum with lower molecular weight silicones (i.e., molecular weight ca. 1000-2000). Furthermore, if the support material has not been deactivated properly, this may lead to decomposition of the gum, thus producing a “bleed” of lower molecular weight compounds. The commonest methyl silicones in use as liquid phases are DC-200, SE-30, SF-96, JXR, OV-1, and UCL-45. These phases should all give approximately equivalent GLC resolution of any particular mixture of coinpounda. Yet i t is often claimed that one 01’other of

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

225

them is superior to the rest, in some particular respect. Horning (H13), for instance, states th at his samples of OV-1 were somewhat more thermostable than his SE-30 samples. The present writer would suggest that any such differences are due either to impurities in the SE-30 or to differences in the activity of the support material in the compared columns-a to a combination of both these factors. Other silicone liquid phases are available in which the methyl groups are replaced t o various extents by other groups. Enhanced selectivity may be obtained by chemical modification of the silicone polymer by the introduction of a polyester which, unfortunately, reduces the thermal breakdown temperature. Ethylene glycol succinate is commonly used as the polyester for these modifications. The silicone part may also be substituted by phenyl or cyanoethyl, etc. Table 3 shows some of these liquid phases which are available commercially. As regards the differences between the various liquid phases, these are largely due to differences in polarity of the liquids. The term “polarity” should be used only in a relative sense in comparing the various phases. The methyl silicones are less polar than the substituted phenyl silicones, which in turn are less polar than QF-1 and the Carbowax (polyethylene glycol) phases. It is often stated that the methyl silicones are nonpolar, but one has only to construct a molecular model of part of these types of polymer to see that they do not fit this description. Rohrschneider ( R l ) has recently examined a t length the relationship which exists between polarity as applied to gas-liquid chromatography, and solvent polarity. He considers the possibilities of defining polarity and using it to predict retention times. O F DIATOMACEOUS SUPPORTS 3.3. DEACTIVATION PREPARATION OF COLUMNS

AND

3.3.1. General Considerations The preparation, scope, and limitation of columns for GLC analysis, especially of submicrogram amounts of polar compounds, have been discussed by McMartin and Street (M10). These workers pointed out that the problem of adsorption of solute by the support material is much greater in the submicrogram region than when dealing with tens of micrograms of solute. The deactivation, by which is meant the reduction of the forces of adsorption between support and solute, is of extreme importance if one is to obtain reproducible chromatograms of compounds (however slight their polarity may be) when dealing with amounts of solute less than 1 pg. As an alternative to support deactivation, one may consider modifying the solute itself so that its polarity is reduced. Indeed, in some cases (e.g., the amino acids) the solute is too polar, without

226

HAROLD V. STREET

TABLE 1 LIQUIDPHASESO ~~

MRT" Material Acetonyl acetone (2,5-hexanedione) Antaxox CO-990 Apiezon L Apiezon M Apiezon N Atpet 80 (sorbitan monooleate) 'I,,&Benzoquinoline Benzyldiphenyl Bis-2-methoxyethyl adipate (BMEA) Bis[2-(2-methoxyethoxy)ethylJether (BMEEE) m-Bis(m-phenoxyphenoxy)benzene (MBM) 1,4-Butanediol succinate polyester (BDS) Carbowax 200 (polyethylene glycol) Carbowax 400 (polyethylene glycol) Carbowax 400 dioleate Carbowax 600 (polyethylene glycol) Carbowax 750 (polyethylene glycol) Carbowax 1000 (polyethylene glycol) Carbowax 1000 monostearate Carbowax 1500 (polyethylene glycol) Carbowax 1540 (polyethylene glycol) Carbowax 4000 (polyethylene glycol) Carbowax 6000 (polyethylene glycol) Carbowax 20 M (polyethylene glycol) Castorwax (hydrogenated castor oil) Celanese C-9 ester 2-Cyanoethyl methyl silicone (XE-60) Diisodecyl phthalate Di-n-decyl phthalate Diethylene glycol Diethylene glycol adipate polyester (LAC-IR-296) Diethylene glycol adipate cross-linked with pentaerythritol (LAC-2R446) Diethylene glycol succinate polyester (LAC-3R-728) Di-2-ethylhexyl sebacate Digol (diethylene glycol) Diglycerol Dimer acid (Empol 1022) Dimethyl formamide Dimethyl sulfolane Dimethyl sulfoxide Dinonyl phthalate Dipropylene glycol Emulphor-0 (Mulgofen ON-870)

("C)

Solventc

20 225 250 250 250 150 100 100 50 220 270 70 100 100 125 125 150 150 150 150 200 200

A

220

200 200 230

160 125

75 200 180 190 150 75 100 150 50 50 30 150 75 175

C H H H H A H H C B C DM DM DM DM DM DM DM DM DM DM DM C C 4 h

H H M A A

A H M M T M C C C C C

GAS-LIQUID

227

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

TABLE 1 (Continued) Material Kpikote 1001 (polyepoxy resin) Flexol8N8 Fluorene Fluorolube Fluorosilicone oil FS-1265 (QF-1) Hallcomid &I-18 (dimethylstearamide) n.-Hexadecane 1,2,3,4,5,6-Hexakis(2-cyanoethoxy)hexane Hexamethyldisilaaane (HMDS) Hexamethylphosphoramide (HMPA) 2,5-Hexanedione Igepal CO-990 [nonylphenoxypoly (ethyleneoxy)ethanol] LAC-IR-296 (diethylene glycol adipate polyester) LAC-2R-440 (diethylene glycol adipate cross-linked with pentaerythritol) LAC-3R-728 (diethylene glycol succinate polyester) LAC-6R-860 (lJ4-butanediol succinate) Marlophen 87 (heptaglycol monoisononylphenyl ether) Nonyl phenol p,p'-Oxydipropionitrile Polyethylene glycol-see Carbowax Polyethylene glycol succinate ester Polyoxyethylene sorbitan monostearate (Tween 60) Polyoxyethylene sorbitan monooleate (Tween 80) Polyphenyl ether OS124 Polypropylene glycol--see Ucon fluid Propylene carbonate Reoplex 400,polypropylene glycol adipate Sebacic acid Silicone fluid FS-1265 (trifluoropropyl) Silicone fluid MS200/50cSt (methyl) Silicone fluid MS200/100cSt (methyl) Silicone fluid MS200/12500cSt (methyl) Silicone fluid MS550 (methyl phenyl) Silicone fluid MS710 (methyl phenyl) Silicone fluid SF-96 (methyl) Silicone OV-1 Silicone OV-17 Silicone grease Silicone gum rubber E-301 (equivalent to SE-30) Silicone gum rubber E-302 (equivalent to SE-31) Silicone gum rubber SE-52 (phenylmethyl) Silicone gum rubber XE-60 (cyanoethylmethyl) Squalene Tetraethylene glycol dimethyl ether

Solvent' 225 175 150 75 225 180 50 50

-

50 20 220 200 200 190 200 150 125 70

A A A A A C H T T C A C A A

A C A DM DM

190 150 160 200

A C

50 200 200 225 200 200 200 200 220 250 350 350 250 250 250 250 230 160 60

A

C A

C M A

H H H H H H H H C H H H A

H A

(Continued)

228

HAROLD V. STREET

TABLE 1 (Continued) Material Tetraisobutylene N,N,N',N'-Tetrakis(%hydroxypropyl)ethylenediamine (Quadrol) Tricresyl phosphate Triethanolamine Trimer acid (Empol 1040) Trimethylol-propane tripelargonate 1,2,3-Tris-(2-cyanoethoxy)propane Tritolyl phosphate Trixylenyl phosphate Tween 60 (polyoxyethylene sorbitan monostearate) Tween 80 (polyoxyethylene sorbitan monooleate) Ucon fluid 50-HB-2000(polypropylene glycol) Ucon fluid LB-550-X (polypropylene glycol) Versamid 900 (polyamide resin) Versamid 930 (polyamide resin) Versamid 940 (polyamide resin) (I

MRT6 ("C)

So1vent

20 150

DM C

125

C C T C C C DM C C H H CB CB CB

75 200 200 175 125 125 150 160 200 150 250 250 250

Available from Perkin-Elmer Ltd.

* MRT = maximum recommended temperature ("C). A, acetone; B, benzene; C, chloroform; CB, chloroform-n-butanol (2:1); DM, dichloromethane; H, hexane; M, methanol; T,toluene.

modification, to be analyzed a t all by GLC. A common example of solute modification of this type is seen in the formation of the alkyl esters of fatty acids prior to GLC analysis. However, even after modification, the compound in many cases may still be fairly polar. This leaves us then with our attention focused on the support material. An account of the surface properties of silica powders has been given by Hockey (H7). Basila (B3) has studied the surface chemistry of silicas. In the diatomaceous earths, it is generally considered that the main forces responsible for adsorption of solutes are the weak van der Waals forces and the stronger forces due to hydrogen bonding. The van der Waals forces are neutralized by all liquid phases and so need not be considered here. In the diatomaceous support there will be both

I I

-Si--OH

(1)

and

I I

I

-Si+Si-

I (11)

groups and, as Palframan and Walker (PI) have pointed out, these will give rise to two types of hydrogen bond, one where the Group (I) functions as the proton donor in the hydrogen bond and the other where the Group (11) functions as the proton acceptor. From this, it would seem

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

TABLE 2 HIQH EFFICIENCY POLYESTERS USED Material

MRTa (“C)

Hutane-1’,4-diol adipate Butane-1,Pdiol succinate (Craig) Cyclohexane dimethanol adipate Cyclohexane dimethanol succinate Diethylene glycol adipate Diethylene glycol succinate Ethylene glycol adipate Ethylene glycol isophthalate Ethylene glycol phthalate Ethylene glycol sebacate Ethylene glycol succinate Ethylene glycol terephthalate 2-Ethylhexane-l,&diol adipate 2-Ethylhexane-1,3-diol succinate Neopentyl glycol adipate Neopentyl glycol sebacate Neopentyl glycol succinate Pentane-1,5-diol adipate Pentan+1,54ol succinate Phenyldiethanolamine succinate Trimethylene glycol adipate

200 200 250 250 200 190 190 190 190 180 180 190 180 180 200 200 200 180 180 200 180

229

AS LlQUlD PHASES“

Solvent” C C A C A

A A C

C C C C C C C C C C C A A

~~~~

0

Available from Perkin-Elmer Ltd. MRT = maximum recommended temperature (“C). A, acetone; C, chloroform.

that even when the (I) groups have been “silanized,” there still remains the possibility of adsorption of certain types of compounds, i.e., those compounds which can denote a proton t o the (11) group, e.g., amines, alcohols, and water. According to Ottenstein ( 0 3 ) , the hydrogen bond formed from the Group (11) is much stronger than that formed from the Group (I). I n a later paper, Ottenstein ( 0 4 ) has pointed out that the tail of the peak in peak tailing is related to the second derivative of the adsorption isotherm, and as the amount of sample introduced into the gas chromatograph is gradually decreased, one sees successively lower portions of the isotherm. Figure 3, which is taken from Ottenstein’s ( 0 4 ) paper, illustrates beautifully the effect of adsorption on retention time as sample size is reduced. Obviously, it would be almost impossible to even start to attempt to identify a compound from its retention time under these circumstances. Various ways of reducing this adsorption effect have been proposed. Acid-washing of support material was described by James and Martin (53) in their original paper on gas chromatography. The function of

13

w

0

TABLE 3 TYPES OF SILICONES USED AS LIQUIDPHASES Substituent group Type of liquid phase Methyl silicone Substituted silicone

Silicone polyestern

0

I n silicone

I n polyester

Percent of substituent group

-

SE-30, JXR, DC-200, SF-96, OV-1, ov-101, ucL-45

Phenyl Phenyl Phenyl Chlorophenyl Vinyl Nitrile Nitrile Nitrile Trifluoro Methyl Methyl Phenyl Phenyl Nitrile Nitrile

Ethylene glycol succinate.

Name of liquid phase

Methyl silicone Methyl silicone Phenyl silicone Phenyl silicone Nitrile silicone Nitrile silicone

5% phew1 25% phenyl 50% phenyl 10% p-chlorophenyl ca. 0.5% vinyl 5% cyanoethyl 25% cyanoethyl 50% cyanoethyl 50% trifluoropropyl

SE-52 DC-550 DC-710, OV-17 F-60 UCW-96, UCW-98 XF-1105 XE-60 XF-1150 QF-1; LSX-34295

5% methyl silicone 30% methyl silicone 5% phenyl silicone 30% phenyl silicone 5% cyanoethylmethyl silicone 30% cyanoethylmethyl silicone

EGSS-X EGSS-Y EGSP-A EGSP-Z ECNSS-S ECNSS-M

P 8 5 -4

m

;3

Be

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

231

A

Time

-

FIG.3. A family of peaks with increasing retention time is observed with decreasing sample size. The entire tailing edge of the peak (AB) is related to the second derivative of the adsorption isotherm. Reproduced from Ottenstein ( 0 4 ) with permission.

acid-washing was stated to be the removal of metallic oxides. Acidwashing has often been used in conjunction with other procedures for improving support material, and this makes it difficult to assess the importance of acid-washing. Recently, however, Ottenstein ( 0 4 ) has tackled this problem and has reached the conclusion that, using an alcohol test sample, acid washing, base washing, or a combination of both of these treatments are ineffective in themselves in reducing tailing. H e found that, possibly, such treatments might increase tailing. However, he did find that silanized acid-washed supports showed less tailing than silanized nonacid-washed supports. This work confirms the findings of McMartin and Street (MlO), who used Chromosorb W and dichlorodiinethylsilane (DDS). Treatment of supports with DDS was described by Homing et al. (H16), who exposed their support material, which had been washed with concentrated hydrochloric acid solution, to DDS vapor. Holmes and Stack (H12) found the conditions for adequate treatment with DDS to be critical. They recommended a detailed procedure in which acid-washed support is treated with a very dilute solution of DDS. Hexamethyldisilazane (HMDS) treatment of support materials was first described by Bohemen et al. (B16), who believed that acid-washing was unimportant

232

HAROLD V. STREET

and that their treatment, which involved refluxing support material with HMDS in petroleum ether, was superior to that involving DDS. Brochmann-Hanssen and Svendsen (B20, B21) used a column packed with acid-washed, HMDS-treated support for analysis of barbiturates and sympathomimetic amines. Improvement of columns by injection of HMDS has been described by Atkinson and Tuey (A9). HMDS and DDS are believed to react with the silanol (I) groups on the surface of the support material and thus remove adsorbing sites. It is assumed that the following reactions take place:

With DDS: OH

OH

I

I

I

I

-Si-0-Si-

1 I

-Si-0-Si-

OH I

i

‘s/

+

+

h c \s i/C% 0’ ‘ 0

c1

c1

I

L

-Si-0-Si-

SC’

‘CH,

c1

c1

‘s/ / \

Hsc

I

1

+

2 H C1

I

7%

h C- S i- Cl ___c

I

+ HC1

I

-Si-0-SiI I

C h

Here the C1 attached to Si is removed subsequently by washing with methanol. The reaction is as follows:

7%

1

H,c-Si-Cl I 0

I

-Si-0-Si-

f CH,OH

-

I

-Si-0-Si-

I

1

I

p H,C-SiI 0

0-CH, 4-

HCl

I

I

(1)

With HMDS: OH -ki-

I

OH I

0- Si-

’

&G H,C- Si-NH,C’

7%

/c& Si- CH, ‘CH,

qC-Si-CH, I b 0 I

--Si-O-Si-

I

7%

C&--Si-C& I

0 I

f NH,

I

Using DDS, we are therefore left with a number of methoxy derivatives (formula I, above); and when HMDS is used, a number of unreacted (I) groups will remain since, presumably, the distance between pairs of -OH groups (on the surface) which react with HMDS is critical. The presence of a variable number of methoxy groups and unreacted (I) groups may account for the variation in degree of adsorption which

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

233

is observed between one column and another, even though all the columns have, apparently, been prepared in exactly the same manner. The general unsatisfactory nature of columns prepared by these methods of “silanization” for use with submicrogram amounts of polar compounds stimulated McMartin and Street (M10) to investigate other ways of reducing adsorption. Starting with acid-washing, they found th a t Chromosorb W which was rapidly washed with concentrated hydrochloric acid and not allowed to stand with the acid gave results which indicated that the degree of adsorption was about the same as was obtained with Chromosorb W which had had prolonged contact with acid. Washing with aqua regia and with concentrated nitric acid gave results which showed no improvement over washing with concentrated hydrochloric acid. Injection of HMDS or DDS into columns containing these acid-washed Chromosorb W preparations caused a marked reduction in adsorption. Silanization of the acid-washed powder resulted in less tailing than when silanized nonacid washed diatomaceous earth was used. This finding has since been confirmed by Ottenstein ( 0 4 ) . It was found (M10) that treatment with HMDS using a procedure similar to that described by Bohemen et al. (B16) reduced the amount of adsorption, but that this effect was lost when the column was heated t o 260”. When methanol rinsing was omitted after DDS treatment, it was found (M10) that the DDS treatment was not effective in reducing adsorption. This is presumably due t o the production of (I) groups by hydrolysis of

1

-Si-CI

I

(111)

groups. It was further discovered (M10) that DDS treatment was more effective with damp support than with dry support material. During these investigations, it was observed th at support material treated with DDS shows less adsorption after heating. This led to investigations into the effect of heat on acid-washed Chromosorb W which had been coated with methyl-substituted polysiloxane SE-30. These investigations showed that adsorption could be reduced by heating the coated diatomaceous earth to about 370” to 420” in the absence of oxygen, and have resulted in the types of columns described by McMartin and Street (MlO) and Street (S11). It should be noted that these columns are not silanieed by DDS, HMDS, or other “silanizing agent,’’ but are probably deactivated a t approximately 370” by breakdown products of the silicone gum rubber. We have also found that improved results are obtained by first heating the empty steel column in air to a temperature of about 800’ (in a furnace) for 18 hours.

234

HAROLD V. STRFiET

Coated

Sintered glass 6k

I I I

I

I

I

Furnace 370' 1.5 hours

I

I I I I I I

I FIG.4. Preparation of column packing. Heat treatment-silanization a t 370°C. Reproduced from Street (512)with permission.

3.3.2. Preparation of Column Packing 3.3.2.1. Method of Street (S12) The first stage consists of washing and coating the earth. About 250 ml of Chromosorb G (100-120 mesh) is washed several times with about 1-liter portions of concentrated hydrochloric acid and the powder is then boiled in the acid in a large conical flask for 10 minutes. The powder is rinsed several times with concentrated hydrochloric acid and then with water until the supernatant liquid is neutral to a p H paper. The suspension of the powder in the water is then boiled for 10 minutes, rinsed several times with about 10 liters of water, the ((fines" decanted after each rinsing and excess water removed by vacuum filtration. The powder is placed in flat glass dishes and dried, with frequent stirring, on a boiling-water bath. Of this washed Chromosorb G, 60 ml is put into a 400-ml beaker, 200 ml of toluene is added, and the suspension is stirred thoroughly with a glass rod. The powder is allowed to settle, and the ('fines" are decanted. The washing with toluene is then repeated, and as much toluene is removed as is possible by decantation. Then 50 ml of toluene is added, followed by 100 ml of 10% water-saturated SE-52 solution, and the mixture is thoroughly stirred. Excess toluene is removed by vacuum filtration and the coated powder is dried in four separate portions with stirring on a hot plate. The second stage consists of the heat treatment of the coated diatomaceous earth. The SE-52-coated Chromosorb G is placed in a Pyrex glass tube (see Fig. 4) 2.5 cm in diameter and 40 cm long and fitted a t one end with a sintered glass disk. Oxygen-free nitrogen is passed (30 ml per minute) through the powder in the glass tube first a t room temperature for 5 minutes and then while the tube is heated in a furnace a t 370" for 1.5 hour. At the end of this period, the tube is removed from the furnace and allowed t o cool down to room temperature with the nitrogen flowing. "he

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

235

powder is removed by suction in three separate fractions. These will be referred to as F1, F2, and F3 where fraction F1 is nearest to the nitrogen input. Fraction F1 is then packed into a stainless steel column and the column is heated a t 370” in a stream of oxygen-free nitrogen (50 ml per minute) for 18 hours. The column is then emptied, packed with fraction F3, fitted with a short precolumn, and heated a t 370” in oxygen-free nitrogen (30 ml per minute) for 1 hour. This packed column is then ready for use. A possible explanation of the results obtained with our procedure is that the water in the water-saturated SE-52 solution may be required for hydrolysis of the silicone polymer, and the breakdown products so produced, effectively “silanize” the Chromosorb G a t 370”. It is also possible that a similar process may account for the silicone polymer treatment of the oxidized steel column. Here again, a thermal breakdown product of SE-52 may react with the metal oxide on the inner surface of the metal column to produce a layer which is less polar than the oxide itself. This might then lead t o less adsorption and/or less destruction of the compounds being analyzed. Such a reaction might also account for the improvement in peak shape when stainless steel columns are heated in air prior to packing. The oxide or partial oxide so formed may facilitate subsequent reaction with the thermal decomposition product of the silicone polymer. It is important to note, however, that this thermal breakdown product must be formed in the absence of free oxygen. We have also found that steel injector blocks can be treated in a similar manner by injecting a solution of SE-30 or SE-52 into the injector from which the column has been disconnected. The injector should be a t about 380” to 400”,and of course nitrogen should be flowing through the block. Metal systems prepared in this way have proved effective in reducing catalytic destruction of solute. For example, we have found it possible g in an allto chromatograph testosterone in amounts down to 5 X metal system with a flame-ionization detector. Columns prepared in the above manner are stable up to a t least 320”; they display relatively little “bleeding” even a t this temperature ; and they are suitable for the analysis of relatively nonpolar compounds, such as the methyl esters of fatty acids, to more polar compounds of high molecular weight (which are generally regarded as being “difficult” to analyze by GLC), such as morphine, even in the submicrogram region. 3.3.2.2. Method of Homing et al. (H16), as Described b y Anders and Mannering (A7) Sieve the diatomaceous earth to 80/100 mesh. Suspend 50 g of the sieved support in 400 ml of concentrated hydrochloric acid and allow to stand overnight. Remove the acid by decantation, and gently suspend the

236

HAROLD

V. STREET

support in distilled water. After the support particles have settled, remove the water, containing finely divided particles, by decantation. Wash the support three times by decantation, transfer to a Buchner funnel, and wash with several liters of distilled water. Then dry the support in an oven a t 100-120". Suspend 25 g of dry support material in 100 ml of a 5% (v/v) solution of dimethyldichlorosilane (DDS) in toluene in a sidearm filter flask. Apply a vacuum (from an aspirator) for a few minutes with shaking, to dislodge bubbles. Allow the pressure to return to atmospheric, then transfer the support material to a Buchner funnel and wash with 100 ml of toluene followed by 1 liter of anhydrous methanol. Dry the silanized support in an oven a t 100-120°. Using an appropriate solvent, prepare a 2 4 % (w/v) solution of liquid phase. Suspend 25 g of the dry silanized support material in 100 ml of the liquid phase solution and hold the flask under vacuum (from an aspirator) for 5 minutes with gentle shaking to dislodge bubbles. Release the vacuum and allow the suspension to stand a t atmospheric pressure for 5 minutes. Transfer the slurry to a Buchner funnel and filter under vacuum. Discontinue filtration while the cake solid is still damp. Spread the coated support material in a baking dish and allow to dry in air for a few hours. Then dry the material in an oven at 100-120" and pack into the column tubing. It is recommended that, after the column has been packed, it should be 'Lconditioned" overnight at a temperature greater than its anticipated use, with carrier gas flowing. 3.3.3. Procedure for Packing Columns

I n the case of steel columns, a sintered steel disk is pressed tightly into one end of the tubing, which has previously been fitted with the appropriate (usually swagelock) fittings to enable it to be fitted into the oven of the gas chromatograph. This disk-end of the column is connected to a water pump, and gentle suction is applied. The open end of the column is held uppermost, and the coated support material is poured into the tube through a small filter funnel attached by clean rubber tubing. The column is tapped gently until no more powder will enter. For a 6 foot by % inch, id., column this process takes about 5 minutes. The open end of the column is then plugged with glass wool which has previously been silanized by immersion in a 1% (v/v) solution of DDS in petroleum ether, and dried. The procedure for packing glass columns is similar to that for steel columns except that a glass wool plug is used a t each end of the tube. Sintered Teflon plugs are also available commercially.

GAS-LIQUID

4.

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

237

Applications of Gas-liquid Chromatography

4.1.

STEROID HORMONES

This section includes the androgens, the estrogens, progesterone, the corticosteroids, and some of the metabolites of these hormones. 4.1.1.

General Considerations

Kuksis (K13) has given an extensive account of the gas chromatography of bile acids and steroids covering the literature up to 1966. His section on the quantitative estimation of estrogens, 17-ketosteroids, testosterone, progestogens, and corticosteroids is of particular interest to clinical chemists. More recently, Eik-Nes and Horning (E2) have produced an excellent monograph on the gas-phase chromatography of steroids. As Brooks (B25) has pointed out, urinary steroid estimations may be classified according to the amount of purification necessary prior to GLC. For example, in the estimation of pregnanediol in pregnancy a fairly simple extraction and clean-up of the hydrolyzed urine is all that is needed before application of GLC, whereas the analysis of estrogens in nonpregnancy urine requires extensive purification procedures prior to GLC. Estimation of pregnanetriol in nonpregnancy urine provides a case which is intermediate between these two examples. I n the analysis of blood, the high sensitivity of the flame ionization and electron capture detectors has proved valuable for estimating steroids a t the submicrogram level. The recent work of Exley (E5, E6) using heptafluorobutyrate derivatives has resulted in greatly increased sensitivity with the electron-capture detector. This worker has shown that by this method as little as 40 ppg of peripheral plasma testosterone can be detected. With 10 ml of plasma the method is su5ciently sensitive to allow accurate determination of testosterone in human female plasma. It is claimed that the method is 20 times as sensitive as electron-capture techniques with the monochloroacetate derivative. The high order of accuracy is achieved by the use of a radioactive label and an internal standard for GLC, and by obtaining several gas chromatograms from the same plasma sample. The collection of carbon-14 and tritium-labeled steroids in GLC has been described by Kliman and Briefer (K8) with application to the analysis of testosterone in human plasma. Using the phenylmethylsiloxane polymer, OV-17, as liquid phase, Horning (H13) has achieved complete separation of testosterone and epitestosterone, both as free steroids and as trimethylsilyl derivatives. Figure 5 illustrates these separations.

238

HAROLD V. STREET

Test Epitest 1 O/o OV-17 T P Io/min

(2000)

i I

30 Time (min)

I

40

FIQ.5. Separation of testosterone (T), epitestosterone (E), testosterone TMSi ether (T-D), and epitestosterone TMSi ether (E-D) with an OV-17 column. Conditions: 12-foot, 1% OV-17 (on 100-120 mesh Gas-Chrom P) column; temperature programmed at 1" per minute from ZOO"; injector, 260"; detector, 300"; flame detector. The separation of the free steroids is due to a difference in selective retention with an OV-17 phase. The secondary 17P-OH group of testosterone is less sterically hindered than the secondary 17a-OH group of epitestosterone, and this leads to selective retention of testosterone compared with epitestosterone. The TMSi ethers are eluted in the same order, but the separation is based on a difference in molecular shape, just as observed for OV-1. The derivatives (E-D, T-D) are eluted before the free steroids (E, T) as a consequence of the selective retention effect of the phase for hydroxyl groups. Reproduced from Horning (H13) with permission.

Vanden Heuvel and Horning (V5) have discussed a t length the conditions for the separation of steroids by GLC. Studies of analytical separations of human steroids and steroid glucuronides have been made by Horning et al. (H14), who have investigated the formation of trimethylsilyl derivatives of steroids with bistrimethylsilyl acetamide. This reagent has been found to be an effective silylating reagent when used alone, but it may also be used with trimethlychlorosilane a s a catalyst. The use of haloalkylsilyl ether steroid derivatives in GLC analysis has also been described by Thomas and Walton (Tl).They have shown that the chloromethyldimethylsilyl ethers of testosterone and dehydroepiandrosterone remain stable under conditions of thin-layer chromatography on silica gel, and that the chloromethyldimethylsilyl ether derivatives are suitable for GLC analysis of 11-deoxy-17-oxosteroids.

239

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

TABLE 4 GAS CHROMATOGRAPH~C t, VALUESFOR SOMESTEROIDSILYLETHERS”’) Steroid

TMSEc

CDMSEc

BDMSEc

Androsterone Etiocholanolone Dehydroepiandrosterone Epitestosterone Testosterone Cholestane

0.88 1.00 1.19 2.00 2.35 1.00

2.75 3.13 3.75 5.62 7.03

4.10 4.80 5.80 9.50 (1.65)d 12.10 (1.95)d ~

~

~~

Reproduced from Eaborn, Walton, and Thomas (El) with permission. 6 Column: 5 foot 1% XE-60 on Gaschrome Q (100-120 mesh) at 210”. Gas flow: Nz at 50 ml min-l. c TMSE, trimethylsilyl ethers; CDMSE, chlorodimethylsilyl ethers; BDMSE, bromodimethylsilyl ethers. dColumn: 5 foot 2% JXR on Gaschrome Q (100-120 mesh) at 220”. Gas flow: Nz at 50 ml min-l. a

Eaborn et al. (El) have prepared steroid bromomethyldimethylsilyl ethers and studied their gas chromatographic behavior. The ethers are prepared by mixing equal volumes of a 0.5 M solution (in hexane) of bromomethyl dimethylchlorosilane and a 0.5 M solution (in hexane) of diethylamine. The mixture is centrifuged, and 0.2 ml of supernatant liquid is allowed to react with 20 pg of steroid for 3 hours. Excess volatile reagents are removed by vacuum desiccation. The dry residue is dissolved in hexane and injected into the gas chromatograph (Pye model 104 fitted with 63Ni electron-capture detector). The results are shown in Table 4. It was found ( E l ) th at the steroid bromomethyldimethylsilyl ethers gave a marked increase in sensitivity, when compared with the corresponding monochloroacetates, approaching tha t of heptafluorobutyrates. In a study of the gas-liquid chromatography of dimethylsilyl (DMS), trimethylsilyl (TMS) , and chloromethyldimethysilyl (CMDMS) ethers of steroids, Vanden Heuvel (Vl) has investigated the mechanism of silyl ether formation and the effect of trimethylsilation upon detector response. Two stationary phases of different partitioning properties were used to examine the GLC behavior of three pairs of epimeric hydroxysteroids and their DMS, TMS, and CMDMS ethers. In general it was found that stereochemical differences a t C-3 produced greater differences in retention times than those a t C-17, and that improved resolution was obtained with derivative formation. However, free hydroxyl groups a t C-20 lead to greater epimer differentiation than did the derivatives. CMDMS ethers showed increased retention times on both “polar” and “nonpolar” liquid phases. This was in contrast to DMS and TMS derivatives which gave shorter retention times on “polar” phases. It was

240

HAROLD V. STREET

Minutes

FIG. 6. Gas chromatographic patterns of pure standards; A = aldosterone ylactone; B = ddadiene. Column 6 foot X 3 mm i.d., glass; Gas-Chrom Q (1W120 mesh) coated with 3% SE-30; 240°C; helium 40 ml per minute. Reproduced from Bravo and Travis (B19) with permission.

claimed (Vl) that the ability to characterize polyhydroxy steroids is enhanced by using mixtures of different silylating reagents, thereby giving rise to multiple derivatives. This technique is similar to the “peakshift” technique described by Anders and Mannering (A7), who used mixtures of acetic and propionic anhydrides to form homogeneous and heterogeneous derivatives of morphine. Darcey and Evenson (D4) have compared GLC and the Zimmerman reaction in the analysis of total 17-ketosteroids in urine. Urinary 17-ketosteroid GLC patterns have been studied by Cawley e t al. (C5), who noted the presence of several peaks with retention times shorter than those of 17-ketosteroids. A comparison of different methods of hydrolysis has been made by Curtius and Muller (ClS) in their study of the GLC of 17-ketosteroids and progesterone metabolites of urine. The simultaneous determination by GLC of pregnanediol and pregnanolone in urinary extracts has been reported by Guarnieri and Barry (G8). GLC was found to be the method of choice in a study by Barry et al. (B2) of four methods for the quantitative analysis of pregnanediol in urine. I n GLC studies of acetylated corticosteroids and related 20-oxopregnane derivatives, Brooks (B23) has shown that complete acetylation of the side chain contributes to the stability of the compounds during

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

241

TABLE 5 URINARY ALDOSTERONE EXCRETION VALUESIN CERTAIN CONDITIONS IN MAN^ Urinary aldosterone excret~ion1jig/24 hours) Type of patient

Gas chromatography

Double isotope derivative assay ~

Uncomplicated essential hypertensives on high-salt diet for 4 days A.C. B.S. V.B. T.D. E.L. Malignant hypertensive on Low-salt diet High-salt diet Primary aldosteronism Before surgery

4.70 6.00 4.00 5.75 4.50

-

23.56

24.30 18.00

16.00

97.10 60.00

After surgery

5.45 8.69

Adrenalectomized

0.00

-

a

b

~~

-

96.00 59.00 6.00 8.00

-

Reproduced from Bravo and Travis (B19) with permission. Subtotal adrenalectomy.

chromatography. Patti et al. (P5) have shown that there are many advantages in using GLC compared to paper chromatography for the determination of 17-hydroxycorticoids in urine and blood. Wotiz and Chattoraj ( W l l ) studied the specificity, accuracy, sensitivity, adaptability, and the sources of error in the analysis of estrogens using thin-layer and gas-liquid chromatography. Their method permits the determination of as little as 0.2 pg of estrogen per 24 hours. Free estrone in blood plasma has been determined by Attal, Hendeles, and Eik-Nes (A10) using GLC with electron-capture detection. GLC was used by Ruchelman (R3) in solubility studies of estrone in organic solvents. A procedure involving GLC for the quantitative estimation of aldosterone in urine has been described by Bravo and Travis (B19). After prelirninary purification and paper chromatography, the aldosterone is oxidized with periodic acid to aldosterone 7-lactone which is extracted and subjected to GLC, using aldadiene as an internal standard. Figure 6 shows the GLC peak characteristics obtained for aldosterone y-lactone and aldadiene. The results obtained by this procedure are compared (B19) in Table 5 with those obtained by a double isotope derivative assay (K9).

242 4.1.2.

HAROLD V. STREET

Analytical Procedures

This section presents detailed procedures for the GLC analysis of selected steroids or groups of steroids. Of particular interest is the approach t o the analysis of urinary steroids by van Kampen and Hoek (H8, V6) of The Netherlands, who have pointed out that a lot of time is taken up in the purification and preparation of derivatives. They have presented a simple method for obtaining information about the total steroid spectrum in urine. These spectra provide important information regarding the quantitative excretion of total 17-ketosteroids, 17-ketogenic steroids, etiocholanolone, and pregnanediol, and have made it possible to characterize several endocrine disorders. A more lengthy but more comprehensive procedure involving preparation of derivatives has been described by Dalgliesh et al. ( D l ) , who studied the application of GLC to the separation and identification on the same chromatogram of a wide range of metabolites occurring in urine or tissue extract. 4.1.2.1.

Urinary Steroid GLC Spectra

Method of van Kampen and Hoek (H8, V6) Apparatus: F. & M. 402 Bio-medical gas chromatograph equipped with a flame ionization detector. Column: 3.8% UC-W 98 on Chromosorb S, 80-100 mesh packed into two 4-foot U-shaped glass columns (0.4 cm diameter) used in parallel. Conditions: Nitrogen or helium, 48 ml/min ; injector and detector temperature 260"; column temperatures ZOO" for 2 minutes then 2"/minute 8 7

% z

6 5

in

$ 4

3

2 I 10

20 30 Minutes FIQ.7. Urinary steroid spectrum in Cushing's syndrome. Upper curve: attenuation 10 X 4. Lower curve: attenuation 10 X 16; K = 17-ketosteroid excretion (mg/24 hours) ; H = ketogenio steroid excretion (mg/24 hours) ; Peaks: 1, oxyandrostane; 2, eticholanolone; 3, androsterone and dehydroisoandrosterone; 4, ll-oxy-17-ketosteroids; 6, p-cortol compounds; 6, cholesterol. Reproduced from Hoek and van %pen (HS) with permiasion.

0

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

GAS-LIQUID

243

87-

K 75

6-

H

2 5g 4-

16

0 P

3-

2II

0

I

I

I0

I

I

20 Minutes

I

4

I

30

FIQ.8. Urinary steroid spectrum in a case of Cushing syndrome after bilateral adrenalectomy. Substitution therapy 20 mg of hydrocortisone daily. Letters and peak numbers as in Fig. 7. Reproduced from Hoek and van Kampen (HS) with permission.

to 250" and kept a t 250' for 5 minutes. Attenuation usually 10 X 4 was adapted to the height of the signal output. Procedure: Urine 8 ml, is heated in a boiling-water bath with 3 ml of concentrated hydrochloric acid for 10 minutes. After rapid cooling, the sample is extracted with 10 ml of distilled 1,2-dichloroethane for 15 minutes. The extract is washed once with 2.5 ml of water. Then 2.5 ml of the washed dichloroethane extract is evaporated to dryness on a water bath. The dry residue is dissolved in 0.2 ml of ether, and 5 pl of this solution is injected into the GLC apparatus using a 10-pl Hamilton syringe. Results obtained with this procedure have been described (H8) in the following cases : Cushing's syndrome, adrenogenital syndrome, hypopituitarism, mammary carcinoma, Stein-Levinthal syndrome, and phaeochromocytoma. Figures 7 and 8 show the urinary steroid spectra obtained in Cushing's syndrome before and after bilateral adrenalectomy. I n a personal communication to the writer, van Kampen and Hoek have emphasized the two great problems involved in the determination of steroid excretion in man. First, there is uncertainty about the fate of steroids in (pathological) body fluids, e.g., losses via the skin, or transformation of degradation products. Secondly, difficulties are encountered in the isolation of steroids from urinary conjugates. Even in enzyme hydrolysis, losses of up to 50% can be found, e.g., in the determination of pregnanediol when the time of hydrolysis is prolonged from 30 to 180 minutes. It is for these reasons that van Kampen and Hoek (H8) have chosen not a single steroid determination, but a whole spectrum with a multi-

244

HAROLD V. STREET

plicity of parameters to describe endocrine disorders. These steroid GLC spectra have proved to be very useful.

4.1.2.2. Urinary Estrogens Method of Wotiz and Chattoraj (W11) Apparatus: A commercial gas chromatograph equipped with a flame ionization detector. Column: 3% SE-30 on 80-100 mesh Diatoport S packed into 4-footlong stainless steel tubes ( q g inch i.d.). The columns are specially prepared as follows: The 80-100 mesh silanized Diatoport S is washed several times with 50% aqueous ethanol and dried in an oven a t 110°C for 2 hours. To a 0.6% (w/v) solution of SE-30 in dichloromethane is added the prepared Diatoport S. This mixture is evaporated on a hot plate with stirring. After this preparation has been packed into the steel tube, the column is fitted into the gas chromatograph and “cured” by heating for not less than 2 hours (usually overnight) at 300°C without any gas flowing. The temperature is then reduced to 260”C, and carrier gas (nitrogen) is passed for a t least 24 hours. Conditions: Nitrogen, 60 ml/min ; column temperature 228°C; injector temperature 250°C. Procedure: Acid hydrolysis of the urine, extraction, and separation into phenolic and nonphenolic fractions is carried out according to the method of Brown (B26). The dried phenolic residue is subjected to preliminary thin-layer chromatography and areas corresponding to estrone, estradiol17,8, and estriol are eluted with ethanol. The eluates are centrifuged and the supernatant ethanol is evaporated to dryness. The residue is dissolved in a mixture of 5 volumes of acetic anhydride and 1 volume of pyridine and is kept at 68°C for 1 hour. To the acetylated mixture, 5 ml of water is added while the mixture is stirred thoroughly with a glass rod. After transfer to a separating funnel, the sample is extracted once with 10 ml and then twice with 5 ml of light petroleum ether. The combined petroleum extracts are washed with 5 ml of 8% sodium bicarbonate solution and then with 3-ml portions of water until the washings are neutral. The solvent is evaporated to dryness and the residue is transferred with petroleum ether to 2-ml tubes. The residue is dissolved in 50-100 p1 of acetone and 2 5 pl of this solution is injected into the gas chromatograph. The procedure is made quantitative by comparing the peak height of the unknown to that obtained for a known concentration of its authentic standard. Further thin-layer chromatography was found necessary (W11) in

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

245

some instances. This was chiefly used to separate the ring D-a-ketols and estradiol from neutral 17-ketosteroids. The technique was found to be sensitive to a range of 0.1 to 0.2 pg of individual estrogens per 24-hour collection of urine. This resulted from a lower practical limit of detection of 0.02 pg of steroid. 4.1.2.3.

Urinary Pregnanediol and Pregnanolone

Method of Guarnieri and Barry (G8) Apparatus: Barber-Coleman 5000 series chromatograph equipped with a flame ionization detector. Column: Mixture of 1% SE-30 plus 3% QF-1 on 100-110 mesh Anakrom ABS previously washed with methanol and methylene chloride; 0.9 meter X 6 mm glass column. Conditions: Nitrogen, 50-150 ml/minute ; column temperature 210'220" ; injector and detector temperature 260'. Quantitative results were obtained by measuring peak areas (peak height times width a t halfpeak height). Procedure: A 100-ml aliquot part of a urine sample is adjusted to pH 4.54.8 and treated with 6.0 ml of Ketodase (Warner-Chilcott; form of beef liver P-glucuronidase) and 10 rnl of acetate buffer (prepared from 3 volumes of 1.66 M sodium acetate solution and 2 volumes of 1 M acetic acid, and adjusted to pH 4.5-4.8). The mixture is incubated a t 37OC for 24 hours and then extracted with two 50-ml portions of chloroform. The combined extracts are washed successively with 50 ml of 0.1 N sodium hydroxide solution and 150 ml of water, and then dried over anhydrous sodium sulfate. The chloroform is removed in vacuo and the residue made up to a suitable volume (usually 0.50 or 1.00 ml) for GLC analysis using a (1 + 1) mixture of methylene chloride and methanol as solvent. When smaller urine volumes are used, the reagents are adjusted proportionally. Samples of 0.5-5.0 pl of standards containing 0.5-2.0 mg each of pregnanediol and pregnanolone per milliliter of methanol-chloroform (1:1) solution are used to relate the peak height to the amount injected. The detector is calibrated daily using fresh standards. Using this method, Guarnieri and Barry (G8) found approximately 98 & 57% recovery of both steroids when 2.0 mg of pregnanediol and 2.0 mg of pregnanolone were added to blank specimens; 0.2 pg of either steroid could easily be detected. It was found that determinations involving hydrochloric acid hydrolysis and toluene extraction resulted in a 10-30% loss of both steroids. I n a personal communication, Dr. R. D. Barry states that he is of the opinion that one need not prepare a silyl ether, ester, or other derivative for these steroids. 4.1.2.3.1.

246

HAROLD V. STREET

4.1.2.3.2. Method of Hammond and Leach (H3)

Apparatus: Pye Panchromatograph equipped with a flame ionization detector. CoZumn: 3% XE-60 on $0-100 or 100-120 mesh acid-washed and silanized Chromosorb W; glass columns 1.5 m x 4 mm i.d. silanized before being packed. Conditions: Argon, 45 ml/min; hydrogen, 50 ml/min; air 250 ml/ minute; column temperature 210"; detector a t oven temperature; oncolumn injection. Procedure: Of a 24-hour sample of urine, 25 ml is brought to boiling in a 100-ml flask fitted with a reflux condenser; 4 ml of concentrated hydrochloric acid and 20 ml of toluene are poured down the condenser into the flask. Boiling is continued for a €urther 15 minutes. After cooling, 20 ml of toluene is added and the lower aqueous layer is removed by suction, care being taken not to remove any emulsion a t the interface. Then 5 ml of 1 N sodium hydroxide solution is added and the mixture is shaken vigorously. The aqueous layer is removed by suction, and the toluene is washed with 10 ml of water. The water is removed as completely a s possible. A few "anti-bump" beads are added, and the toluene is boiled on a hot plate until its volume is reduced to about 10 ml. The toluene extract is transferred to a small beaker using two 2-ml portions of fresh toluene to rinse out the flask, and then evaporated carefully to dryness on a hot plate. Using three 2 ml portions of acetone, the extract is transferred to a small test tube. The acetone is evaporated on a boiling water bath. After the outside of the tube has been dried, the tube is placed in an oven a t 100°-llOoC for a few minutes to remove the last traces of water. If derivatives are not to be prepared immediately, the tubes are stored in a desiccator. I t is of paramount importance to exclude traces of water. The residue is dissolved in 0.7 ml of redistilled tetrahydrofuran (THF) . Hexamethyldisilazane, 0.2 ml, and trimethylchlorosilane, 0.1 ml, are added and the mixture is agitated vigorously. The tube is then capped with tinfoil and stored in a desiccator for 3 hours at room temperature or it may be left overnight. Then 2 pl of this mixture is injected (using a 10 fil Hamilton syringe which has been rinsed out with T H F ) into the gas chromatograph. If desired, an internal standard of cholesterol may be used. The average recoveries of pregnanediol by this method were greater than 90%; those of pregnanolone about 87%. It is interesting to note that Hammond and Leach (H3) found that for clinical purposes the extent of degradation of these steroids during the hydrolysis was insignificant. This is in contrast to the findings of Guarnieri and Barry (G&),q.v.

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

247

FIG.9. Special reaction tube, with conical shape, permitting close extraction of aqueous layer by needle aspirator. Capacity 225 ml. Reproduced from Ruchelman and Cole (R4)with permission.

The presence of androsterone may cause some interference with the determination of pregnanediol in nonpregnancy and early pregnancy urine, when XE-60 is used as the liquid phase. Leach and Hammond (L3) removed this source of interference by using borohydride reduction. These workers also found that this technique was useful in providing confirmat o r y evidence of identity of the other steroids which may be found in toluene extracts of acid-hydrolyzed urine. The reduction is effected by adding 200 mg of potassium borohydride to 25 ml of urine. After standing a t room temperature for 3 hours, the urine is hydrolyzed with HCl and after extraction and derivative formation is chromatographed as described above. 4.1.2.4.

Urinary 17-Ketosteroids

Method of Ruchelman and Cole (R4) Apparatus: Barber-Coleman 5000 series chromatograph equipped with a flame ionization detector. Column: 2% neopentyl glycol succinate (NGS) on 80-100 mesh Chromosorb W, acid-washed DMCS; or 3.3% XE-60 on Anakrom ABS (90-100 mesh) ; 6 foot X 5 mm and 6 foot x 4 mm i.d. glass U-tubes. Conditions: Nitrogen, 100 ml/min; column temperatures 215°C (for (NGS) and 232°C (for XE-60) ; injector temperature 245'; detector temperature 260°C ; attenuation 5 and sensitivity 1OOO. Procedure: Ten milliliters of 40% (by volume) sulfuric acid is added to 50 ml of urine contained in a special glass tube (see Fig. 9) and the mixture is hydrolyzed for 1 hour a t 80°C in a water bath. Seventy milliliters of diethyl ether is added to the cooled mixture.

248

HAROLD V. STREET

,

P-20H

I '

A

DEA ' E P-30H 11-OH-A

II-OH-E

b

Minutes

I

0

I

I

6

I

I

12

I

I 18

I

I

24

I

J

30

,

,

36

,

,

42

Minutes

FIQ.10. GLC of trimethylsilyl derivatives of synthetic steroid mixture. (A) Procedure using 2% neopentyl glycol succinate on Chromosorb W, acid-washed, dimethyldichlorosilane treated, 80-100 mesh, in 0 foot X 5 mm i.d. glass column at 215" with 10 psi nitrogen. (B) Procedure using 33% xE-60 on Anakrom ABS (90100 mesh), in 6 foot X 5 mm i.d. glass column, at 232", with 5 psi nitrogen. Reproduced from Ruchelman and Cole (R4) with permission.

The reaction tubes are shaken for 15 minutes, and the lower layer is then aspirated and discarded. The ether extract is washed with 70 ml of N NaOH by shaking for 15 minutes. The aqueous phase is removed and discarded. About 60 NaOH pellets are added to the ether extract, which is then shaken for 15 minutes. This, mixture is filtered through about 1 inch of purified sea sand resting on a glass wool plug in a small funnel, rinsed with fresh ether, and evaporated to dryness in a 250-ml conical flask. The residue is dissolved in 2 ml of redistilled tetrahydrofuran (THF) containing 10 mg per 100 ml of coprostan-3a-01 (COP) as internal standard. (Blowout pipettes must not be used because moisture is detrimental to the formation of trimethyl siIyI ethers.) The solution is transferred by capillary pipette to a 15-ml conical centrifuge tube, and the flask is rinsed out with 0.5 ml of fresh THF. Hexamethyldisilazane (0.3 ml) and trimethylchlorosilane (0.1 ml) are added. The tube is then stoppered, thoroughly mixed, and allowed to stand at room temperature overnight. The mixture is taken just to dryness by placing the tube in a 56°C bath and using a stream of dry nitrogen. Immediately prior to injection,

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

249

0.1 ml of THF is added and thoroughly mixed with the residue. The tube is centrifuged for about 2 minutes to pack the NH,Cl precipitate. Two to 4 pl of the supernatant liquid are then injected into the gas chromatograph. The COP internal standard compensates for any loss incurred during the procedure and during GLC. Results obtained by this procedure are illustrated in Fig. 10. 4.2. CATECHOLAMINES, TRYPTAMINES, AND OTHER BIOLOGICAL AMINES 4.2.1. General Considerations

Catecholamines and amines of the histamine and tryptophan group are highly polar compounds and hence have low volatility. It is preferable, as is the case with amino acids, q.v., therefore, to form derivatives to obtain satisfactory gas-liquid chromatograms. The main problem in the analysis of urine for these amines lies not in the gas chromatographic aspect of the procedure, but rather in their isolation from the urine prior to GLC, where there may be as little as 1 pg of amine per 100 ml of urine. It appears that this problem of extracting the amines into a sufEciently small volume of suitable solvent prior to GLC has not yet been solved satisfactorily. The fact that GLC can be used successfully to separate catecholamines through their derivatives, when starting with pure substances should provide a stimulus for increased research activity In this field. Horning and co-workers (B24, C3, H15, V3) have made a very extensive study of the application of gas-liquid chromatography to the analysis of catecholamines, tryptamine-related indole bases, and other amines. Prior to chromatography, derivatives of the amines were prepared. Formation of these derivatives involved acylation, condensation with acetone, and silylation. Their data showed quantitative relationships using standard mixtures containing 0.2 pg to 2 p g of catecholamines and other amines of the phenylalkylamines, imidazole, and indole types. The quantitative determination of 3-methoxy-4-hydroxyphenylethyleneglycol by gasliquid chromatography has been discussed briefly by Wilk et al. (W6). The so-called VMA (3-methoxy-4-hydroxymandelic acid) has been assayed quantitatively by gas-liquid chromatography in a method described by Wilk et al. (W7). Horning et al. (H17) carried out a GLC study and estimation of several urinary aromatic acids including p-hyclroxyphenylacetic acid, 5-hydroxyindoleacetic acid, and VMA. The derivatives used were the trimethylsilyl ether methyl ester and methyl ester of acids occurring in urine. Clarke et al. (C7, CS) have shown that dopamine (3-hydroxytyramine) can be determined a t the nanogram level using trifluoroacetate derivatives and electron-capture detection.

250

HAROLD V. STREET

In a study of the gas chromatography of several indole derivatives, Brook et al. (B22) demonstrated that the N-trifluoroacetyl methyl esters could be used for analysis in the 50-nanogram region when flame ionization was used and in the 50-picogram region with an electron-capture detector. A new method for the separation of catecholamines by GLC has been presented by Horning et al. (H18), who reinvestigated the problem of derivative formation and separation using N-trimethylsilylimidazole and bistrimethylsilyl acetamide as silylating reagents on 5% OV-1 and 5% OV-17 columns. Very recently Vanden Heuvel (V2) has illustrated the value of combined GLC-mass spectrometry in complementing GLC techniques in establishing reaction conditions appropriate for the preparation of derivatives suitable for the GLC of compounds of biological interest, with special reference to several “biological” amines. 4.2.2.

Analytical Procedures

As mentioned above, extraction of many dihydroxyamines from aqueous solutions is not entirely satisfactory. This should be borne in mind when attempting to adapt the following procedures to a particular problem

.

4.2.2.1.

Catecholamines and Related Compounds

Method of Brooks and Horning (B24) Apparatus: Barber-Colman Model 10 or an E.I.R. instrument fitted with argon ionization detector. Column: Glass U-tubes, 6 foot x 4 mm. i.d. Gas Chrom P, 80-100 or 100-120 mesh, acid-washed dichlorodimethyl-silane treated. Two types of packing used: 10% neopentyl glycol succinate (NGS) and a twocomponent phase consisting of 7% silicone oil F-60 with 1% ethylene glycol succinate-phenyl methylsiloxane copolymer (EGSS-Z) . ConditionHydroxyphenylethylamine Tyramine

Homovanillylamine

1 d

L

L

C

Octopamine

Normetanephrine

J

FIG.11. Separation of acetylated phenylethylamines at 198°C with an F-60-Z phase. Reproduced from Brooks and Horning (B24)with permission.

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

251

Minutes

FIG.12. Separation of acetylated amines at 216°C with an F-60-Z phase; PHOPE = P-hydroxyphenylethylamine; Tyr = tyramine ; 3-MeO-TYR = 3-methoxytyramine ; OCTOP = octopamine ; DOP = dopamine; TRYPT = tryptamine ; M E T = metanephrine; NORMET = normetanephrine; EPI = epinephrine; NOREPI = norepinephrine. Reproduced from Brooks and Horning (B24) with permission.

ing was a t 210-225" in a stream of argon for a t least 24 hours to give a satisfactory baseline. Conditions: Argon; column temperatures, 170", 198O, and 216'. Injector and detector were 30"-50" above column temperature. Procedure: Amines, 1O-lOOO pug, were acetylated by treatment with 10-20 pl of redistilled acetic anhydride and 10-20 p1 of pyridine (distilled over potassium hydroxide). Excess reagents were removed in a vacuum desiccator. Methyl esters of amine acids were prepared by refluxing in methanolic hydrogen chloride. N-Acetylation of these esters was by treatment with acetic anhydride and pyridine. Esterification of other acids was carried out with diazomethane in ether. The products were dissolved in ethyl acetate, and volumes of 0.1 to 2 p1 were injected into the gas chromatograph. Amounts of sample injected were about 0.1 pg for rapidly eluted compounds and 1 pg for those with long retention times. I n some cases, acetone condensation products of the primary amines were chromatographed. Catecholamines, however, underwent decomposition under these conditions. Figures 11-14 show some of the results obtained with this procedure. 4.2.2.2. Tryptamine-Related Amines 4.2.2.2.1. Method of Horning et al. (H15) Apparatus: Barber-Colman and E.I.R. (Models 10, 5000, and AU-8) gas chromatographs fitted with hydrogen flame ionization detector.

252

HAROLD V. STREET

Minutes

FIQ.13. Separation of amines including histamine and N-dimethylhistamine as acetyl derivatives a t 198°C with m F-60-Z phase; PHO-PE = P-hydroxyphenylethylamine ; ME-HIST = N-dimethylhistamine ; TY R = tyramine ; HIST = histamine; 3-MeO-TYR = 3-methoxytyramine; OCTOP = octopamine; TRYFT = tryptamine ; M E T = metanephrine; NORMET = normetanephrine. Reproduced from Brooks and Horning (B24) with permission.

al ln 0 9 ' a

E L

al

T?

8al

a:

Minutes

FIQ.14. Separation of several aromatic acids as methyl esters (A) before and (B) after acetylation, with a 7% FdO/l% Z at 170°C. VAN = vanillic acid; VER = veratric acid; HOMOVAN = homovanillic acid; HIPP = hippuric acid; VMA = 4hydroxy-3-methoxymandelic acid ; Ac = acetylated. Reproduced from Brooks and Horning (B24) with permidon.

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

253

Column: Glass-U-tubes, 6 foot X 4 mm i.d. Gas Chrom P; 80-100 or 100-120 mesh, acid-washed, dichlorodimethylsilane treated. Liquid phases were (1) F-60-Z : a mixture of 7% F-60 (a methyl-p-chlorophenylsiloxane polymer) and 1% EGSS-Z (a copolymer of ethylene glycol, succinic acid, and a methylphenylsiloxane monomer) ; and (2) 10% neopentyl glycol succinate polyester (NGS) ; (3) 1.25%SE-30; (4) 3% DC-710; (5) 0.6% JXR (methylsiloxane polymer) and 0.2% cyclohexanedimethanol succinate polyester (CHDMS) . Conditions: Argon; column temperatures (see captions to figures) ; injector 30" above column temperature ; detector, 250O. Procedures: (i) Silylation. Hexamethyldisilazane was used to prepare trimethylsilyl derivatives of the amines according to the method of Luukkainen et al. (L13). Horning et al. (H15) used acetone or tetrahydrofuran (THF) as solvent. However, the present writer believes that acetone should be avoided here owing to the possibility of eneamine formation. (ii) Formation of Schiff bases. The amines were condensed with acetone by dissolving the amines in acetone and injecting the mixture, without isolation, into the gas chromatograph. (iii) Acyl derivatives. Ten milligrams of the amine (or hydrochloride) was added to 0.3 ml of acetonitrile; 0.1 ml of anhydride (acetic or propionic) and 0.1 ml of pyridine were then added, followed by refluxing. The mixtures were injected directly into the gas chromatograph. In the case of pentafluoropropionyl derivatives, direct injection of reaction mixtures gave erratic results. This was thought to be due to the presence of pentafluoropropionic acid. However, by adding sodium bicarbonate followed by extraction with ethyl acetate, washing of the extract, and drying with anhydrous magnesium sulfate, satisfactory results were obtained. Results obtained with these procedures are shown in Figs. 15 and 16. 4.2.2.2.2. Method of Brook et al. (B22) Apparatus: F. & M. Model 400 fitted w,ith a hydrogen flame ionization detector (FID) and F. & M. Model 700 equipped with a pulsed voltage electron-capture detector (ECD) . Columns: Glass, 6 or 8 foot x 0.25 inch i.d., packed with acid-washed silane-treated Chromosorb W (60-80 mesh) coated with 15% SE-30. Conditions: For FID-Helium, 60 ml/minute; injector port, 240"; column 190" ; detector 240". For ECD-95% argon plus 5% methane, 60 ml/minute; injector port, 230"; column, 180"; detector, 190". Procedures: (i) Esterification b y diazomethane. Mix 1 ml of an ethereal indole acid standard (1 mg/ml) with 3 ml of an ethereal-diazomethane solution. CAUTION! Diazomethane is explosive and poisonous. After

254

HAROLD V. STREET

,

-.A 6-OH-DMT-TMSi

Minutes FIG.15. Separation of tryptamine-related indole bases. The compounds are N,Ndimethyltryptamine (DMT), eneamine (acetone condensation product) from tryptamine (TRYPTSB), 7-trimethylsilyloxy-N,N-dimethyltryptamine (7-OH-DMTTMSi), 4-trimethylsilyloxy-N,N-dimethyItryptamine (4-OH-DMT-TMSi), 5-trimethylsilyloxy-N,N-dimethyltryptamine (5-OH-DMT-TMSi), &trimethylsilyloxy-N, N-dimethyltryptamine (6-OH-DMT-TMSi), and the eneamine (acetone condensation product) from 5-trimethylsilyloxytryptamine (5-OH-TRYPT-TMSi-SB) . Conditions : 7% F40, 1% EGSS-8, on 100-120-mesh Gas Chrom P, 182"C, 18 psi; argon ionization detection system. Reproduced from Horning et al. (H15), with permission.

15 minutes, evaporate the mixture to dryness with a stream of nitrogen (in a fume cupboard). Dissolve the residue in an appropriate volume of ethyl acetate and inject an aliquot part of this solution into the gas chromatograph. Results obtained by this procedure are shown in Fig. 17, using the flame-ionization detector.

Norepinephrine

Min

15

I0

5

0

I

I

I

I

2250 2050 '185~ 165O FIG.16. Separation of acetyl derivatives of epinephrine, norepinephrine, metanephrine, and normetanephrine with JXR-CHDMS. Conditions: 0.6% JXR and 02% CHDMS on Gas-Chrom P (80-100 mesh) ; 18 psi nitrogen. Reproduced from Horning e t al. (H15) with permission. Oc

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

255

Retention time, minutes

FIG.17. GLC of a mixture of indole derivatives (1 pg of each derivative injected). For conditions see text. IAN = indole-3-acetonitrile; IAA-ME = methyl ester of indole3-acetic acid; IAA-EE = ethylindole %acetate; ICA-ME = methyl ester of indole-3-carboxylic acid ; IPA-ME = methyl ester of indole-3-propionic acid ; IBA-ME = methyl ester of indole-3-butyric acid ; 5-OH-IAA-ME = 5-hydroxyindole-3-acetic acid (methyl ester). Reproduced from Brook e t al. (B22) with permission. (ii) Trifluoroacetylution of indoles. Place 1 ml of a 1 mg/ml indole solution in a 13-ml graduated tube and evaporate to dryness in vucuo. (Note: In the case of the indole acids, it is the methyl ester which is trifiuoroacetylated.) Dissolve the residue in 3 ml of anhydrous ethyl acetate. Add 0.5 ml of trifluoroacetic anhydride (TFA) and a small amount of sodium sulfate. Cap the tube. Agitate with magnetic stirrer for 1 hour a t room temperature. Then remove excess TFA under reduced pressure. Dissolve the residue in 1 ml of ethyl acetate. Inject an appropriate aliquot part of this solution into the gas chromatograph. Figure 18 shows some results obtained with this procedure, using the electroncapture detector.

256

HAROLD V. STREET

W

W

z

5

I

I

I

I

I

i

2 3 4 5 6 7 8 9 1011 1213141516 Retention time, minutes

I

I

I

.I

1

1

I

I

I

1

1

FIG.18. GLC of a mixture of trifluoroacetylated indole derivatives (0.5 ng of each derivative injected). For conditions see text. F = trifluoroacetyl derivatives; other letters as in Fig. 17. Reproduced from Brook et al. (B22) with permission.

4.3. AMINOACIDS

4.3.1.

General Considerations

An extensive review of the literature dealing with the GLC of amino acids up to the year 1966 has been presented by Weinstein (W4). The section presented here discusses some of the more important work on the GLC of amino acids which has appeared since Weinstein's review. Because of the extremely polar nature of their functional groups, amino acids cannot be successfully analyzed by GLC without carrying out some modification of one or both of these groups. As long ago as 1956, Hunter et al. (H19) devised a gas chromatographic procedure for the determination of amino acids following oxidation. They made use of the fact that a-amino acids react quantitatively with ninhydrin (triketohydrindene hydrate) to form aldehydes containing one carbon atom less than the original acids. The resulting aldehydes were collected in a trap cooled with solid carbon dioxide-acetone or liquid nitrogen and subsequently subjected to gas chromatography using a 10-foot column filled with a silicone-Celite mixture. Column temperature was 69", and the flow rate of the helium carrier gas was 23 ml/min. In this way, thew

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

257

workers were able to separate successfully a mixture consisting of 3methylbutanal and 2-methylbutanal which had been prepared by the ninhydrin oxidation of equal weights of leucine and isoleucine, respectively. I n 1960, Zlatkis et al. (22) described a direct gas chromatographic method for determining amino acids that yield volatile aldehydes on oxidation. The aldehydes were chromatographically separated, catalytically cracked, and analyzed in a thermal conductivity cell. Conversions of the amino acids to methyl-a-hydroxy esters (L8, W l ) and methyl a-chloro esters (M5)have been described, and Bier and Teitelbaum (B13) have chromatographed the volatile amines produced by decarboxylation of amino acids. More recently, attention has been focused on the use of the N-acylated esters of amino acids, e.g., n-amyl N-acetyl (57) , n-butyl N-trifluoroacetyl (23), n-propyl N-acetyl (G6), n-amyl N-trifluoroacetyl (B14) , and methyl N-trifluoroacetyl (C13, H1) . Silyl derivatives have also been prepared (R5). I n a study of the quantitative aspects of gas chromatography of amino acids, Lamkin and Gehrke (Ll) made a comparison of methyl N-trifluoroacetyl, methyl N-acetyl, n-butyl N-trifluoroacetyl, and n-butyl N-acetyl esters of each of valine, phenylalanine, glutamic acid, and lysine. These four amino acids were selected so that most of the functional groups present in protein amino acids were represented. The conclusion reached was that the n-butyl N-trifluoroacetyl derivatives had advantages over the others investigated. Lamkin and Gehrke obtained single chromatographic peaks for all the common protein amino acids except tryptophan and arginine, and also showed that these derivatives gave superior results from a quantitative aspect. Tryptophan gave two peaks but could be converted into a single derivative by longer acylation. It was felt that further investigation was needed in the case of arginine. 4.3.1.1. Preparation of n-Butyl N-Trifluoroacetyl Esters (Ll). The amino acid mixture (60 mg) was placed in a 125-ml flat-bottom flask, and 10 ml of anhydrous methanol containing 1.20 + 0.10 mEq of anhydrous HC1 per milliliter were added. The flask was then stoppered with a ground-glass stopper; the solution was stirred on a magnetic stirrer for 30 minutes a t room temperature; and the methanol was removed by vacuum distillation a t 60" +- 1°C. Ten milliliters of l-butanol containing 1.20 -F- 0.10 mEq of anhydrous HCl per milliliter, were added; the solution was heated for 180 minutes with magnetic stirring in an oil bath a t 90" k 3"C, and the butanol was removed by vacuum distillation a t 60" + 1°C. The n-butyl ester hydrochlorides were then trifluoroacetylated by adding 5.00 ml of methylene chloride and 0.50 ml of trifluoroacetic anhydride and stirring (magnetically) a t room temperature

258

HAROLD V. STREET

for 120 minutes. The trifluoroacetic anhydride and solvent were removed by vacuum distillation a t room temperature, and the n-butyl N-trifluoroacetyl esters were dissolved in anhydrous chloroform prior to gas chromatography on 1% (w/w) neopentyl glycol succinate on 60- to 80-mesh Gas-Chrom A packed into a 100 cm x 0.3 cm i.d. borosilicate glass column. Flow rate was 38 ml of N, per minute. Column temperature was 67°C for 6 minutes, then programmed a t 3.3"Cper minute to 218°C. Five microliters of a 2-ml chloroform solution containing derivatives prepared from the amino acid mixture were injected directly onto the chromatographic column without the use of a flash heater. I n the examples given by Lamkin and Gehrke ( L l ) , the 2 ml of chloroform contained derivatives prepared from 10 mg of an amino acid, i.e., an amount of derivative equivalent to 25 pg of amino acid was injected on to the column. Using this (Ll) preparation procedure, Stefanovic and Walker (SS) studied the effect of stationary phase-support ratio on the gas-chromatographic separation of trifluoroacetylamino acid butyl esters using ethylene glycol adipate polyester columns. They found that, for certain amino acids, the elution pattern was a function of the amount of liquid phase on the column packing. When 0.5% ethylene glycol adipate was used, sharp well-defined peaks were obtained and all the amino acids listed in Table 6 except cysteine and methionine were completely sepaTABLE 6 RELATIVE RETENTION DATAFOR AMINOACIDDERIVATIVES" Amino acid

0.5% EGA

1.0% EGA

Alanine Valine Glycine Isoleucine Leucine Proline Threonine Serine Cysteine Methionine Phenylalanine Aspartic acid Glutamic acid Tyrosine 0mithine Lysine Tryptophan

0.285 0.347 0.379 0.420 0.470 0.499 0.548 0.627 0.733 0.751 0.814 0.856 1 .ooo

0.311 0.359 0.430 0.430 0.481 0.539 0.539 0.621 0.732 0.787 0.849 0.849 1.000

0

1.102

1.088

I.191

1.199 1.268 1.316

1.264 1.316

Reproduced from Stefanovic and Walker (S9) with permission.

Ab X 1000

26 12 51

10 11 40 -9 -6 -1 36 35 -7 - 14 8 4 0

* Relative retention time on 1.0% EGA minus relative retention time on 0.5% EGA.

GAS-LIQUID

259

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

P

?!

Pso

10

I 20

I 30

Time (min)

FIO.19. GLC separation of trifluoroacetylamino acid butyl esters on 0.65% ethylene glycol adipate. Reproduced from Stefanovic and Walker (S9) with permission.

rated. These last two acids were completely resolved using 0.7% or higher levels of liquid phase. However, this increase in liquid phase resulted in a shift of the glycine peak toward the isoleucine peak. A t the level of 1% liquid phase, glycine and isoleucine were eluted together, as were the proline-threonine and phenylalanine-aspartic acid pairs, while a t the 2% level the order of elution of the amino acids was altered. It was considered (S9) that changes in elution pattern were probably due to interaction of the polar liquid phase with polar groups on the amino acid derivatives, and that a similar effect would be observed with most other polar packings. These workers (S9) felt that the separation of the trifluoroacetylamino acid butyl esters on B single stationary phase was a feasible proposition. Figure 19 shows the results they obtained for a mixture of n-butyl N-trifluoroacetyl esters of 17 amino acids using a liquid phase of 0.65% ethylene glycol adipate. Low responses were obtained for arginine, histidine, and cystine, but this problem is still being investigated. Gee (Gl) considered that the many hours required for preparation of the butyl esters of amino acids made for too lengthy a procedure; accordingly, she devised an improved method for making the methyl esters of amino acids in 30 minutes. This was followed by gas chromatography of the N-trifluoroacetates of these methyl esters using dimethyl dodecanedioate as an internal standard. Five milligrams of each amino acid together with 5 mg of dodecanedioic acid were refluxed with 10 ml of HCl in methanol (prepared by bubbling anhydrous HCI into a chilled con-

260

HAROLD

V.

STREET

tainer of methanol) and 0.1 ml of thionyl chloride for 30 minutes. Excess esterification reagents were removed by rotary evaporation using a water pump. The methyl esters were then refluxed with 2 ml of trifluoroacetic anhydride for 10 minutes. Any hydroxyl and amino groups present were thereby trifluoroacetylated. Aliquot portions (2-4 PI) of this mixture were injected directly into the gas chromatograph. A l-meter, %-inch o.d., 0.075-inch i.d. aluminum column was used packed with 5% neopentyl glycol succinate on Chromosorb W, HMDS, 100/200 mesh. Carrier gas (helium) flow was 50 ml per minute; injector and detector were held a t 250' ; the column was programmed from 64 to 210" a t 4" per minute. The peak shapes shown in this paper ( G l ) display an appreciable amount of tailing and are not as symmetrical as those shown in the article by Stefanovic and Walker (SS). It would seem that if one 'were considering using a gas chromatographic method for the qualitative and quantitative estimation of amino acids in body fluids, a good starting point would be the n-butyl N-trifluoroacetyl ester procedure (S9), although good results have been obtained very recently using the n-propyl N-acetyl esters ((211) and the alkylidine and alkyl amino acid esters

(D5). 4.3.2. Analytical Procedures

The first of the procedures described below requires only about 1 hour for esterification and analysis, and amounts of amino acids down to mole can be estimated satisfactorily. The second procedure is taken from a comparative study of the use of bis (trimethylsilyl) acetamide and bis (trimethylsilyl) trifluoroacetamide as silylating reagents. 4.3.2.1. Analysis Using Ohe n-Propyl N-Acetyl Esters

Method of Coulter and Hann (C11) Apparatus: Perkin-Elmer 881 gas chromatograph, dual column, equipped with flame ionization detector. Column: All-glass 106 cm x 3 mm i.d. 80-100 mesh Chromosorb GHP. Equal amounts of coated Chromosorb were mixed together before packing. Half was coated with 0.7% Carbowax 6000 and half with 0.7% Carbowax plus 0.05% tetracyanopentaerythritol. Conditions: Nitrogen, 30 ml/min column temperature programmed 100-240° a t G"/minute started immediately after the solvent front; detector temperature, 30" above column temperature ; injector temperature, 240". Propnol-HCI: Propano] is dried by refluxing with calcium hydride, and is then distilled. Only the middle fraction is used. Dry hydrochloric

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

261

_c

Propanol-HCI

nitrogen

U

--I

Steam t

FIG.20. Apparatus used for preparation of derivatives. Amino acid solution is placed in tube A and propanol-HC1 in the funnel. Dry nitrogen is passed through the apparatus as required. Reproduced from Coulter and Hann ((311) with permission.

acid is bubbled into the propanol until its concentration is 8 M . Immediately after preparation, the propanol-HC1 is transferred to the funnel shown in Fig. 20. HC1 gas may be prepared by the action of concentrated sulfuric acid on fused ammonium chloride. The HCI gas is then dried by passing through two washes of sulfuric acid made 103% with oleum. Procedure: The apparatus shown in Fig. 20 is used for esterification. About 0.1 ml of amino acid solution or protein hydrolysate is placed in tube A (Quickfit MF 24/0, capacity ca. 2 ml) and quickly dried in a vigorous stream of dry nitrogen (dried by passing through a molecular sieve grade 4A, Union Carbide). A convenient amount for preparation of standards is 10-8 mole of each amino acid but to mole is satisfactory for unknowns. After drying, the nitrogen is shut off; ca. 0.4

262

HAROLD V. STREET

ml of propanol-HC1 is run in, incubated a t 100" for 10 minutes (using a steam jacket), and then evaporated with nitrogen. This propylation step is repeated. The tube containing the dry ester hydrochlorides is then attached to a similar apparatus not carrying a funnel, to remove residual HCl gas, and 0.4 ml of a freshly prepared mixture of purified pyridine and acetic anhydride (4: 1, v/v) is added to the tube. Acetylation is complete in 5 minutes a t room temperature. Excess reagents are carefully evaporated in the apparatus used for removal of HC1. The oily esters are dissolved in dry ethyl acetate; 1 pl of solution containing about 5 pmoles/ml (i.e., 5 X mole) is injected into the gas chromatograph. It is interesting to note that the authors of this procedure (C11) found that, when the syringe needle was retained in the injection port, destruction of ornithine, lysine, and tryptophan occurred. The extent of the reduction was proportional to the time the needle was retained. Retention for 8 seconds resulted in 90% destruction of these three amino acids, and under these circumstances tyrosine was also reduced, They recommended, therefore, that injections should be made as rapidly and in as standard a manner as possible. The present writer finds it difEcult to accept that more destruction of some amino acid derivatives occurs the longer the syringe needle is retained in the injection port. In a well-designed injection port (and the injection port on most commercial instruments is well designed), the e

GI ASP

Met

I

Thr

Leu I

0

, I

Isothermal

Programmed temperature 6'C/min

FIG.21. Gas-liquid chromatogram showing the elution of 17 amino acids as the n-propyl N-acetyl esters. 5 X lo-' mole of each in 1pl was injected with an attenuation X 200. For conditions see text. Reproduced from Coulter and H a m (C11) with permission.

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

263

compounds being chromatographed are volatilized and swept out of the injection port very rapidly. If this were not so, there would be considerable peak broadening. So that 1 or 2 seconds after the syringe plunger has been pressed home, there should be little or no compound remaining in the injection port. And it is difficult to see how leaving a syringe needle in the injection port for several minutes causes destruction of a substance which has already left the port and is on its way along the column. Perhaps a more probable explanation is that there was a leak of carrier gas between needle and septum when the authors made their observations. Figure 21 shows the chromatogram obtained w,hen 17 amino acids are subjected to the above procedure. Retention times and relative responses of n-propyl N-acetyl amino acids are given in Table 7. For complete esterification of all amino acids, it is essential that absolutely dry reagents be used. Arginine and histidine also present some difficulties, but it is suggested that by conversion of arginine to ornithine (using arginase) and histidine to aspartic acid (by ozonolysis) , these two amino acids can be satisfactorily analyzed quantitatively by the above procedure. TABLE 7 RETENTION TIMESAND RELATIVE RESPONSES OF n - h o p n N-ACETYLAMINOACIDS" Amino acid

Retention time (min-sec)

Relative molar response; serine = 1.00

Alanine Valine Isoleuciiie Leucine Glycine Proline Threonine Serine Aspartic acid Histidine (as aspartic acid) Methionine Cysteine Phenyldanine Ely droxyproline Glutamic acid Tyrosine Ornithine Lysine Methionine sulfone Tryptophan

4-48 5-06 6-12 6-24 6-51 848 9-36 11-00 12-06 12-06 13-00 1348 14-12 14-24 14-36 20-12 21-06 22-00 23-00 2&45

0.89 1.18 1.33 1.49 0.78 1.34 1.43 1 .oo 1.69 1.69 1.50 0.37 2.45 1.54 1.84 1.85 1.69 1.76 1.30 1.30

5Reproduced from Coulter and Hann (C11) with permission.

264

HAROLD V. STREET

4.3.2.2. Analysis of Sulfur-Containing Amino Acids

Method of Shahrokhi and Gehrke (54) This paper (54) compares the two silylating reagents bis (trimethylsilyl) acetamide (BSA), first described by Klebe et al. (K6), and bis(trimethylsilyl) trifluoroacetamide (BSTFA) , for the preparation of volatile trimethylsilyl (TMS) derivatives of 12 sulfur-containing amino acids. BSTFA was recommended as the reagent of choice for taurine, cysteic acid, homocystine, djenkolic acid, ethionine, methionine sulfone, ~-2-thiolhistidine~ cysteine, and cystine. For S-methyl-L-cysteine, methionine sulfoxide, and methionine, BSA was used as silylating reagent. Apparatus: F. & M. model 300 gas chromatograph, attached to a column oven and detector module of an F. & M. model 400, fitted with an F. & M. model 1609 flame ionization attachment. Column: 0.5% SE-30 on 60-mesh acid-washed dimethylchlorosilanetreated Chromosorb G, packed into 1.0 meter x 3.5 mm borosilicate glass column. Conditions: Nitrogen, 40 ml/minute ; air, 450 ml/min ; hydrogen, 36 ml/minute ; column temperature, 75"-20O0C a t 4.6"/minute; chart speed 1/3 inch/min. Procedure: Place 10 mg (ca. 0.04-0.08 mmole) of sulfur-containing amino acid and 10 mg of phenanthrene (as internal standard) in a 16 X 75 mm culture tube (screw cap with Teflon liner). Add 0.5 ml (ca. 2.5 mmoles) of BSA (or 0.5 ml ca. 3 mmoles, of BSTFA) and 1.5 ml of acetonitrile. Screw the cap on the tube and immerse just to the top of the liquid level in an oil bath a t 150° for 5 minutes. Allow to cool and inject 5 pl into the gas chromatograph. The chromatograms shown in Figs. 22 and 23 illustrate the results obtained by this procedure. The authors claim that the minimum detectable limit a t a 2: 1 signal-to-noise ratio with respect t o the flame ionization detector is about 5 ng of sulfur-containing amino acid injected on the column. 4.3.2.3. Analysis of Thyroid Amino Acid Homones

Method of Hansen (H4) Using bis (trimethylsilyl) acetamide (BSA) volatile trimethylsilyl (TMS) derivatives of the active components 3,5,3',5'-tetraiodothyronine (thyroxin, T4)and 3,5,3'-triiodothyronine (T,) have been prepared as well as TMS derivatives of the nonphysiologically active components 3,3',5'-triiodothyronine (T3'), 3,5-diiodothyronine (T,), and 3,5-diiodotyrosine (DIT).Separation and quantitative estimation of these iodinated amino acids is achieved by gas-liquid chromatography. The method is

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

265

S-Methyl- L -cysteine

\

Iysteine

al u)

Cvsteine 'acid

C 0

a. u)

Homocystine Cystine

B

I

150

I

0

1I

175

200

1

1

I

10

20

30

OC Min

Methionine sulfoxide

FIGS. 22 ( t o p ) and 23 (bottom). Gas-liquid chromatograms of the trimethylsilyl derivatives of sulfur-containing amino acids. Column: 05% w/w SE-30 on acidwashed, dichlorodimethylsilane-treated Chromosorb G (80-100 mesh) ; 1 meter X 3.5 mm i.d. glass; initial temperature, 75", programmed a t 4.6" per minute; Nz flow 40 ml per minute. Injected mixture contained ca. 5 pg of each amino acid. Reproduced from Shahrokhi and Gehrke (54)with permission.

being used to investigate the iodinated amino acids in hydrolyzates resulting from hydrolysis of thyroid preparations. Apparatus: Barber-Colman Model 5000 fitted with both flame ionization and electron-capture detectors. Column: 2% SE-33 on 100/200 mesh Chrom Q packed into a 2 foot X 3 mm i.d. glass column. Conditions: Nitrogen, 50 ml/min; splitting device used to channel 95% column effluent to flame ionization detector and 5% to electron-

266

HAROLD V. STREET

I.S.

DIT W v)

c

0

% E! L

W

e 8

W (r

100 -

-I L

I

I

0

3

6

I

I

I

I

9 12 Time, minutes I

I

I

15

18

I

150 (IO"/minI 200 250 280 Column temperature

OC

-

FIG.24. GLC of a standard mixture of thyroid hormones. See text for conditions and abbreviations. Ty = tyrosine. Reproduced from Hansen (H4) with permission.

capture detector. Injector temperature, 270" ; detector temperature, 275' ; column programmed from 150 to 280" a t 10" per minute. Procedure: Amino acid standards containing 6 pmoles per 5 ml in 5% (v/v) ammoniscal methanol are used. The internal standard (13.) is 25 mg of squalene per 50 ml in methanol. A suitable aliquot part of amino acid solution is evaporated to dryness in a 2-dram vial. Then 100 pl of reagent mixture (prepared by mixing 5 ml of tetrahydrofuran, 2 ml of BSA, and 5 drops of chlorotrimethylsilane in a 2-dram vial with a foil-lined screw cap) is added. The vial is sealed with a foil-lined screw cap and warmed on the front edge of a water bath (in a fume cupboard) for 1 minute. Appropriate aliquot parts of this mixture are then injected into the gas chromatograph.

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

267

5-Methyl- L -cysteine Zysteine

W u)

Cysteine acid

C 0

a u)

J

B

150

I

J

0

fl I

Homocystine Cystine , , acid ~ i c

175

200

1

I

I

10

20

30

OC Min

Methionine sulfoxide Ethionine

Methionine

Methianine

L-2-Thiolhistidine

W

I

I

i!,5 0

I00

I

125 10

150, 20

I

175

290C' 30 Min

FIGS.22 ( t o p ) and 23 (bottom). Gas-liquid chromatograms of the trimethylsilyl derivatives of sulfur-containing amino acids. Column: 0.5% w/w SE-30 on acidwashed, dichlorodimethylsilane-treated Chromosorb G (80-100 mesh) ; 1 meter X 3.5 mm i.d. glass; initial temperature, 75", programmed a t 4.6" per minute; Nz flow 40 ml per minute. Injected mixture contained ca. 5 pg of each amino acid. Reproduced from Shahrokhi and Gehrke (54)with permission.

being used to investigate the iodinated amino acids in hydrolyzates ypQ,,1+ino

r l m l x r o ; o r.F +L.7mm:-l

fvnm

0'

-..--.-.-..L:---

I

I

10

20

Minutes

FIG. 25. Gas chromatograph of the urine (85 pg creatinine equivalent) of a patient with homocystinuria. Urine was treated as described in text. 5% QF-1 on Gas-Chrom P (80-100mesh), a t 140" ; argon, 46 ml per minute. This peak corresponds to 150 pg of homocystine per milligram of creatinine. Reproduced from Creer and Williams (G7) with permission.

268

HAROLD V. STREET

Expected retention time of homocystine derivative

I

I

I

0’

10 20 Minutes FIG.26. Gas chromatograph of the urine of a “normal” control. Conditions aa in Fig. 25. Reproduced from Greer and Williams (G7) with permission.

Procedure: Pass 10 ml of urine through a cation-exchange resin column (Dowex 50-W-X4, 20-50 mesh, in the hydrogen form) and wash the column with 30 ml of distilled water. Elute the amino acids with 20 ml of concentrated ammonia solution. Evaporate this eluate to dryness in vacuo. To the residue add 1 ml of trifluoroacetic anhydride and 0.5 ml of trifluoroacetic acid, and allow the mixture to stand a t room temperature for 20 minutes. Remove excess reagents with a stream of air and treat the oily residue with an ethereal solution of distilled diazomethane (DANGER! Poisonous-explosive) for 15 minutes. Remove excess diaeomethane (CAUTION!) with a stream of air. Dissolve the residue in methanol and inject a suitable aliquot part into the gas chromatograph. Figures 25 and 26 show results obtained by this procedure. 4.4. CARBOHYDRATES

General Considerations The presence of the hydroxyl groups in carbohydrates confers to these molecules a polarity which is sufficiently high to render the vapor pressures of this class of compounds far too low for them to be analyzed

4.4.1.

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

269

directly by GLC. Derivatives of the carbohydrates must be formed to provide sdicient volatility for GLC analysis. Some of the procedures described for this purpose include alkylation (usually methylation), acetylation, benzoylation, benaylation, and trimethylsilylation (A8, B5, F1, H6, 58, M4, S7, 517). I n 1964, Bayer and Widder (B4) produced a very interesting paper in which they showed that GLC could be used to demonstrate the existTABLE 8 RELATIVERETENTIONS OF ISOMERIC SUGARMETHYLETHERS"~~ Substance

Relative retention

Arabinose 2,3,5-Trimethyl-~methyI-~-arabinoside 2,3,5-Trimethyl-j3-methyl-~-arabinoside 2,3,4,5-Tetramethyl-al-~-arabinose 2,3,4-Trimethyl-j3-methyl-~-arabinoside 2,3,4-Trimethyl-~methyl-~-arabinoside Glucose

1.60 2.04 2.50 3.05 3.20 2.32 2.80 3.41 3.88 4.25 4.87 5.25 6.61 7.01 8.03 9.87 10.84 14.42

2,3,4,5,6-Pentamethyl-al-~-glucose 2,3,4,6-Tetramethyl-j3-methyl-~-glucoside 2,3,5-Trhethyl-~-glucosana (1,4)-6 (1,6) 2,3,4,6-Tetramethyl-a-methyl-~-glucoside 2,3,5,6-Tetramethyl-@-methyl-~-glucoside 2,3,5,6-Tetramethyl-~-methyl-~-glucoside 2,3,4-Trimethyl-~-glucosan-cu (I,5)-j3(1,6) 2,3,4-Trimethyl-j3-methyl-~-glucopyranoside 2,3,6-Trimethyl-j3-methyl-D-glucof uranoside 2,3,6-Trimethyl-a-methyl-~-glucofuranoside 2,3,4-Trimethyl-~~-methyl-~-glucopyranoside 2,3,6-Trimethyl-@-methyl-~-glucopyranoside 2,3,6-Trimethyl-a-methyI-~-glucopyranoside Galactose 2,3,4,5, 6-pent amethyl-al-D-galactose

2.71 3.02 3.81 4.58 4.76 5.30

2,3,4,5-Tetramethyl-@-methyl-~-galactoside 2,3,5,6-Tetramethyl-j3-methyl-~-galactoside 2,3,4,6-Tetramethyl-methyl-~-galactoside 2,3,5,6-Tetramethyla-rnethyl-~-galactoside 2,3,4,5-Tetramethyl-cu-methyl-~-galactoside Fructose I ,3,4,5,6-Pentamethyl-keto-~-fructose 1,3,4,5-Tetramethyl-cu-methyl-~-f ructoside 1,3,4,6-Tetramethyl-cu-methyl-~-fructoside 1,3,4,6-Tetramethyl-@-rnethyl-~-f ructoside L,3,4,5-Tetramethyl-j3-methyl-~-fmctoside ~

2.16 2.75 3.05 3.83 6.75 ~

~~~

~

Reproduced from Bayer and Widder (B4)with permission. * Retentions were measured relative to ethyl ether of succinic acid; for conditions see text. "

270

HAROLD V. STREET

ence of carbonyl forms of sugars in solution. Glycosidification was carried out with methanolic hydrochloric acid. The resulting glycosides were then methylated and subjected t o GLC. The chromatograms were compared with those obtained by direct methylation of the sugars, thus taking advantage of the fact that mutarotation of the methylglycosides is not possible. The sugars studied were fructose, galactose, glucose, and arabinose. Table 8 lists the relative retention times of the isomeric sugar methyl esters. For GLC, Bayer and Widder used a 2-meter column (0.4 cm id.) filled with 20% polyethylene glycol 1500 on kieselguhr (0.2-0.3 mm). They found that liquid phases of Apiezon or polyesters were not as satisfactory as the polyglycols. Sweeley and Walker (S16) carried out simultaneous GLC analyses of glucose, galactose, galactosamine, and sialic acid as found in neutral glycolipides and gangliosides. Treatment with anhydrous dilute methanolic hydrochloric acid was used to form methyl glycosides from the hexoses. Under these conditions, N-acetylhexosamine was only partially converted to the free amino sugar; and the 2-0-methyl ketal of methyl neuraminate was formed from the neuraminic acid. These products were then silylated with a mixture of hexamethyldisilazane and trimethylchlorosilane in pyridine (2 : 1 : 10, v/v), and the resulting trimethylsilyl derivatives were subjected to GLC a t 160° on acid-washed, silaniaed Gas Chrom S

IIIII I

3

Time, minutes

FIG.27. Gas-liquid chromatogram of the separation of trimethylsilyl ethers of sugars on hexamethyldisilaeane-treated Chromosorb W (80-100 mesh) coated with 15% (w/w) Carbowax 20M. Column: 12 foot X y4 inch copper coil; 170"; helium 100 ml per minute ; 4 pl of 0.2% solution. Peaks : 1, ribose ; 2, ribose ; 3, a-mannose ; 4, y-galactose; 6, a-galactose; 6, internal standard ; Y, a-glucose; 8, p-mannose; 9, p-galactose; 10, p-glucose. Reproduced from Sawardeker and Sloneker (S2) with permission.

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

271

(100-120 mesh) coated with 2.5% (w/w) SE-30. This method, which includes details of the procedure for isolation of glycolipids and gangliosides from human brain, was used for both qualitative and quantitative determination of glucose, galactose, and neuraminic acid in glycolipids. These workers (S16) also have concluded that GLC is probably the most satisfactory method available for distinguishing qualitatively between glucosamine, galactosamine, and mannosamine. Trimethylsilylation was also used by Alexander and Garbutt (A5) for the determination of D-glucose by GLC. It was found that D-sorbitol provided a satisfactory internal standard. The retention times for the trimethylsilyl derivatives of these sugars were in the order a-D-glucose. D-sorbitol, P-D-glucose on a 3% SE-52 column. The support material was Chromosorb P (100-120 mesh) ; column temperature was 180". A carrier gas flow rate (argon or nitrogen) of 2 S 3 0 ml per minute gave the most efficient performance with a hydrogen flame ionization detector. I n 1965, Sawardeker and Sloneker (S2) reported on the effectiveness of a liquid phase of Carbowax 20 M for the GLC resolution of trimethylsilyl ethers of mixtures of monosaccharides. Figure 27 shows the chromatagram they obtained under these conditions. 4.4.2. Analytical Procedures 4.4.2.1.

Plasma Galactose and Glucose

Method of Copenhaver ((210) This method for the simultaneous quantitative and qualitative determination of plasma or blood galactose and glucose eliminates the necessity to remove glucose if galactose is to be estimated. Apparatus: Perkin-Elmer model 801 gas chromatograph fitted with a hydrogen ionization detector. Column: Dual glass columns, 6 foot X 2 mm i.d. packed with hexamethyldisilazane-treated Chromosorb W (80-100 mesh) and coated with 25% (w/w) Carbowax 20 XI. Columns were conditioned overnight a t 225" with carrier gas flowing. Conditions: Helium, 30 psi and 30 ml/min; injector temperature, 220" ; column temperature, 145O. Procedure: Starting with 0.2 ml of plasma, prepare a protein-free filtrate by the method of Somogyi (S8). Copenhaver (CIO) has pointed out that the choice of a method for deproteinization which yields an alkaline filtrate is important because of the finding of Wells et al. (W5) that acidic filtrates inhibit trimethylsilylation. Lyophilize 1 ml of protein-free filtrate (or 1 ml of galactose and glucose standard). Add 0.5 ml of anhydrous pyridine, 100 pl of hexamethyl-

272

HAROLD V. STREET

W v)

c

0 0.

In

E! L

0 +

0 W

+

aJ

n

I

I

5 Minutes

I

I

FIG.28. Gas chromatogram of trimethylsilyl derivatives of an equilibrated mixture of galactose, glucose, and methyl a-D-mannopyranoside (2:2: 1, w/w). The numbered peaks refer to (1) methyl a-n-mannopyranoside ; (2) a possible furanose form of galactose ; (3) a-D-galactopyranose ; ( 4 ) a-n-g]ucopyranose ; (6) /3-D-galactopyranose ; ( 6 ) P-D-glucopyranose. For GLC conditions see text. Reproduced from Copenhaver (ClO) with permission. disilazane and 50 ,ul of trimethylchlorosilane. Shake and centrifuge the mixture. Inject an aliquot part of the clear supernatant liquid directly into the gas chromatograph. GLC results obtained with this procedure are illustrated in Fig. 28. For galactose (50 mg/100 ml) added to Somogyi filtrates, ten replicate analyses showed (C10) a mean recovery of 100.2 with a standard deviation of &1.2. For endogenous glucose, ten analyses gave a mean recovery of 83.6 mg/100 ml (standard deviation, +1.3) compared w,ith the value of 86.6 mg/100 ml when determined by an automated ferricyanide method. Recoveries of added sugars are shown in Tables 9 and 10.

GAS-LIQUID CHROMATOGRAl’WY I N CLINICAL CHEMISTRY

273

TABLE 9 RECOVERY OF ADDEDQUANTITIES OF GALACTOSE TO SOMOGYI FILTRATE OBTAINED FROM 1 ML OF PLASMA^.) Galactose ~

Sample No.

Added

1 2 3 4 5

6 7 0

~~~

Recovery

(Pg)

Found (PP)

1000.0 750.0 500.0 250.0 125.0 50.0 0.0

995.0 720.0 501.0 253.0 123.0 56.5 0.0

99.5 96.0 100.2 101.2 98.4 113.0

(%)

-

Reproduced from Copenhaver (ClO) with permission.

* Each value is the average of duplicate analyses.

4.4.2.2. Neutral Sugars in Glycoproteins

Method of Lehnhardt and Winzler (L5) This method involves hydrolysis of the protein sample with a cation exchange resin in the hydrogen form, reduction of the sugars to the corresponding alcohols, acetylation with acetic anhydride in pyridine, and gas chromatography of the acetylated alcohols. Apparatus: Microtek model 220 gas chromatograph equipped with flame ionization detectors. The injector port was modified by the use of a glass insert (8.4 cm x 1 cm 0.13. and x 0.2 cm i.d.) to reduce the dead volume of the vaporizing chamber. TABLE 10 RECOVERY OF ADDEDQUANTITIES OF GLUCOSE TO SOMOGYI FILTRATE OBTAINED FROM 1 ML OF PLASMAO.~ Glucose Sample No.

Added

Found

Recovery

Gcg)

(rg)

( %)

1 2

200 400

866 1056 1265

95.0 99.0

3

-

Reproduced from Copenhaver ((310) with permission. Each value is the average of duplicate analysis. Endogenous glucose was determined to be 860 p g / m l by an automated ferricyanide method. b

274

HAROLD V. STREET

x

Column: Glass U-tube 1.83 meters 4 mm i.d. containing by weight 0.75% HiEFF-lBP, 0.25% EGSS-X, and 0.1% 144-B (phenyldiethanolamine) on Gas Chrom Q (60-80 mesh). Conditions: Nitrogen, 40 ml/min; injector temperature, 225"; detector temperature, 325" ; column temperature, initial 160" programmed at 1.3" per minute to 210'; hydrogen (to detector) 50 ml/min; air (to detector burner), 189 ml/min; air (to detector purge), 189 ml/min; chart speed, 30 inches per hour. Procedure: Dry the glycoprotein sample, 0.1-3 mg, containing about 0.1 &mole of neutral sugar, in a 6 x 50 mm culture tube under reduced pressure in a centrifuged biodryer (VirTis Research Equiment, Model 10-310). Then store the sample in a desiccator until ready to proceed with hydrolysis. Dissolve the dried sample in 50 p1 of water and 50 pl of a 40% w/v suspension of Dowex 50 X 2 (H') 200400 mesh resin in 0.02 N hydrochloric acid. Seal the sample tube with a silicone-rubber septum, mix the contents of the tube, and place in a steam bath for 40 hours. Remove the sample from the hydrolysis bath, allow to cool, and inject through the septum into the sample 50 pl of an internal standard solution containing 0.0277-0.2771 pmole of arabitol, depending on the carbohydrate content. Mix the sample thoroughly, and centrifuge for 2 minutes a t 2800 rpm to sediment the sample from the septum and the walls of the tube. Remove the septum, add 0.4 ml of water to the tube, and mix thoroughly. Using a disposable pipette, transfer the mixture to a column containing Dowex 1-X8 (HC0,-) 200-400 mesh resin, and allow to elute into a 12-ml siliconized conical centrifuge tube. This column is prepared in a 6-inch disposable pipette with a glass wool plug by the addition of 50 pl of a 20% w/v suspension of Dowex 1-XS (HC0,-) 200-400 mesh resin. A glass wool plug is used to separate the Dowex 1-X8 (HC0,-) resin from the Dowex 50-X2 (H+)resin used in the hydrolysis. Wash the hydrolysis tube twice with 0.3 ml of water, and pass the washings through the column. Then pass 1 ml of a 50% v/v aqueous solution of methanol through the column. Concentrate the combined eluates in the biodryer under reduced pressure. Add 100 pl of water to the conical centrifuge tube to dissolve the residue and then transfer the solution to a 5 x 50 mm polyethylene disposable centrifuge tube using a 100 pl Lang-Levy pipette. Wash the centrifuge tube with two 100 p1 portions of water and transfer the washings to the 5 x 50 mm tube. Then take the sample to dryness in the biodryer under reduced pressure. Dissolve the sample in 50 pl of water and add 50 p1 of 0.22 M sodium borohydride (NaBH,). Allow the mixture to stand at room temperature for 1 hour. Add 50 pl of glacial acetic acid, to decompose excess NaBH4, and take the sample to dryness in the biodryer.

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

275

Remove the borate, as the volatile trimethylborate, by the addition of three 100 pl portions of a 1:lOOO v/v HC1-methanol solution with concentration to dryness in the biodryer under reduced pressure after each addition. At this stage keep the sample overnight in a desiccator to ensure dryness. To acetylate the sample, add 50 pl of pyridine (purified by refluxing for 1 hour with ninhydrin, distilling with protection from moisture, and storing over potassium hydroxide until used) and 50 J.J of acetic anhydride, and place the tube in a steam bath (at 100") for 15 minutes. Remove the tube; mix, and replace in the steam bath for an additional 15 minutes. Cool the mixture and inject 1-10 pl aliquot parts directly into the gas chromatograph. Lehnhardt and Winzler (L5) state that, at this stage, the samples can be kept for a t least 3 months without decomposition. Figure 29 shows the gas chromatogram of the alditol acetates produced by this procedure from a standard mixture of the parent carbohydrates. This standard mixture consisted of erythrose, 2-deoxyribose1 rhamnose, fucose, ribose, arabinose, xylose, 2-deoxyglucose, 2-deoxygalactose, mannose, galactose, and glucose. I n this system, 2-deoxyglucose and 2-deoxygalactose appear as a single peak, and glucose and galactose are not completely resolved. Single peaks will also be produced by those sugars which produce identical alcohols upon reduction, e.g., arabinose and lyxose. The authors have applied their procedure to a number of glycoproteins. Figure 30 shows the results they obtained with orosomucoid, canine

Minutes FIQ. 29. Gasliquid chromatogram of the alditol acetates produced from a standard mixture of the parent carbohydrates. For GLC conditions see text. Reproduced from Lehnhardt and Winzler (L5)with permission.

A

% c 0

D $!

6 + Q) u

+

2

Minutes

8

Minutes

C

FIQ. 30. Gas-liquid chromatograms of the alditol acetates of the neutral sugars released after hydrolysis of various glycoproteins. Arabitol was added as the internal standard. (A) Orosomucoid; (B) canine submaxillary m u c h ; ( C ) a glycopeptide released by the action of trypsin from intact human erythrocytes. For GLC conditions see text. Reproduced from Lehnhardt and Winzler (L5) with permissiop.

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

277

submaxillary mucin, and a glycopeptide released by trypsin from intart human erythrocytes. The reproducibility of this method with three different mixed (known) standards was found to be 100 =k 3.8%. Reproducibility of the method when starting with ten separate orosomucoid solutions was within 3% for each of the sugars. 4.4.2.3.

Urine Sugars

Method of Welk et al. (W5) With this method, glucose and other sugars (e.g., galactose, fructose, pentose, mannitol, inositol) are determined by GLC of their trimethylsilyl ethers. Urea interferes with the silylation step but is conveniently removed by prior incubation of the urine with urease. A mixed-bed ionexchange resin is used to deionize the resulting solution. Apparatus: Barber-Colman Model 10 gas chromatograph equipped with an Argon ionization detector (A-4183). Column: 6 foot x y4 inch glass column packed with 15% ethyleneglycol succinate polyester (EGS) on Chromosorb W (80-100 mesh). Conditions: Argon, inlet pressure 15 psi; column temperature 158”. Procedure: Mix 1 ml of urine with 3 ml of water in a test tube and

1

I

10

20

Minutes FIG.31. Gas chromatogram of the trimethylsilyl ethers of sugars in “normal” human urine. Peaks: 1, unknown; 2, fructose; 3, unknown; 4, a-n-glucose; 5, mannitol; 6, unknown; 7 , P-n-glucose; 8, myoinositol. Urine from a “normal” patient undergoing a mannitol clearance test. GLC conditions as in text. Reproduced from Wells and Weber (W5) with permision.

278

HAROLD V. STREET

place the tube in boiling-water bath for 5 minutes to destroy enzyme activity. Cool the tube to 45”, add 10-20 mg of urease powder, and incubate a t 45” for 30 minutes. Place the tube in boiling water for 5 minutes to inactivate the urease, and pour the mixture through a column of 2 g of a mixed-bed ion-exchange resin packed in a suitable small glass tube. Collect the effluent in a 50-ml centrifuge tube. Remove the water, in vacuo, a t 50”. Immediately add to the residue 1 ml of silylating reagent. (This reagent, which keeps 2-3 days if stored in a tightly stoppered container, is prepared by mixing together 17 ml of pyridine, 2 ml of hexamethyldisilazane, and 1 ml of trimethylchlorosilane.) Instant and complete reaction occurs a t room temperature. Inject 1-3 pl of the mixture directly into the gas chromatograph using a 10-pl syringe. To prepare a standard, evaporate 3 ml of standard sugar solution (containing 1 mg of sugar per milliliter of saturated aqueous benzoic acid solution) to dryness a t 50°, in vacuo. Immediately, add 3 ml of silylating reagent and inject a suitable aliquot part of the mixture into the gas chromatograph. By comparing the areas (area = height of peak multiplied by width of the peak a t half its height) of “test” and “standard” peaks the method is made quantitative. The chromatogram obtained from “normal” urine treated according to the above procedure is shown in Fig. 31. I n this figure, mannitol is present because the urine was collected from a patient tested for mannitol clearance. ANALYSIS 4.5. TOXICOLOGICAL 4.5.1. General Considerabions An excellent chapter on the application of gas chromatography to toxicology has been produced recently by Anders and Mannering (A7), which supplements the first chapter on this subject in this series (G5). Kazyak and Knoblock (K3) have also discussed the problems of application of gas chromatography to analytical toxicology. I n a study of the gas-liquid chromatography of submicrogram amounts of drugs, McMartin and Street (M10, M11) and Street (S11) have described the preparation, scope, and limitations of columns; the analysis of barbiturates and related drugs; and the analysis of alkaloids, in biological media. More recently, Street (S12) has outlined some of the forensic problems associated with the application of gas-liquid chromatography to the analysis of amines and alkaloids. I n toxicological analysis, the main problem is one of identification, since the toxicologist may not know initially what the compound is that

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

279

he is required to estimate. It is in this connection that gas-liquid chromatography provides a very powerful weapon in the armory of the practicing toxicologist, and this is no less true in the case of the clinical chemist who is generally dealing with a more limited field of toxicological analysis. Theoretically, gas chromatography should be capable of being usred to screen biological material for the presence or absence of any “foreign” substance of a nonionic nature. Even in the case of ionic substances, i t should still be possible to use GLC provided th a t the ionic nature can be modified (perhaps by derivative formation, decarboxylation, etc.) to a compound which is less polar than the parent substance, and which in its modified form shows an increase in vapor pressure which is sufficiently high to enable a reasonable number of molecules to enter the vapor phase a t a “reasonable” temperature. Having used GLC t o identify the toxic substance, it is not usually difficult to estimate the amount of such substance quantitatively, either by GLC or other means. However, there are, a t present, certain problems, associated with the general use of GLC as a screening procedure in toxicological analysis. These problems include thermal instability of certain compounds, difficulty in forming derivatives in some cases, etc. But, in general, many of the problems are associated with the extraction procedures, not with the GLC itself. It is the writer’s belief that GLC,in association with mass spectrometry, and with derivative formation within and outside of the gas chromatograph, will ultimately provide the answer to the majority of toxicological analytical problems. 4.5.2. Analytical Procedures

I n choosing the following procedures, particularly in the case of the basic drugs, the general aim has been to select compounds which are “difficult” t o analyze by means other than GLC or which might on first consideration be thought to be “difficult” to analyze using GLC. This selection has been made within the general framework based on increasing molecular weight, i.e., from “volatile” compounds, such as ethanol, to so-called “nonvolatile” substances, such as morphine. I n the submicrogram region, GLC may well provide the only suitable method of analysis. It is hoped that the following procedures will be of help to the clinical chemist who is starting t o use GLC for toxicological analysis. It is not intended to be a comprehensive coverage for forensic toxicologists. 4.5.2.1. Ethanol. Numerous GLC methods have been devised for the analysis of ethanol in biological fluids. Four main types of procedure may be distinguished, namely, those preceded by distillation (F2, M7), those employing extraction (Cl,L14),those using direct injection of sample or diluted sample (B17, C16, D6, MI, M6,P4, S14),and those

280

HAROLD V. STREET

in which an aliquot part of the gas phase in equilibrium with the sample (or treated sample) is injected into the gas chromatograph (C15, M2, W2).

Method of Curry et al. (C16). Apparaks: Pye 104 chromatograph equipped with a flame ionization detector, and fitted to a Kent Chromalog integrator (set a t 1 mV range on a “200” attenuator setting of detector amplifier; 0.5% cut o f f ) . Column: 10% PEG400 on 100-120-mesh Celite, 5-foot column. Conditions: Argon, 150 ml/min; injector, column and detector, all a t 85”. Proceduye: I n a small stoppered vial, dilute 20 pl of blood or urine sample with 200 pl of aqueous n-propanol (0.25 ml of n-propanol per liter of water t o give ca. 20 mg of n-propanol per 100 ml) using a Griffin and George type 221 hemoglobin-type “Diluspence.” I n a separate vial, dilute 20 p1 of ethanol standard (containing 200 mg of ethanol per 100 ml) with 200 pl of the aqueous n-propanol solution. Make several injections of 1-pl aliquot parts of “test” and “standard” diluted solutions using a Hamilton 10-pl syringe. 4.5.2.1.1.

Calculation. Blood ethanol

R X 200 mg/100 ml R,

= t-

area under ethanol curve of ‘(test” injection area under propanol curve of “test” injection area under ethanol curve of “standard” injection R, = area under propanol curve of “standard” injection With this method, chloroform is not separated from ethanol. However, the authors state that the detector response is such that 100 ml of chloroform per 100 ml gives a response which is equivalent to only 15 mg of ethanol per 100 ml and that in ‘%urgical’’anesthesia, blood chloroform levels are about 15 mg/lOO ml. 4.5.2.2. Barbiturates. I n 1960, Janak (54) attempted to identify barbiturates by applying gas chromatography to the products of pyrolysis (at 800”) of the drugs. Nelson and Kirk (N3) described a similar procedure and presented unique patterns for 22 barbiturates. Although such techniques may be suitable when only one barbiturate is present, it is difficult to see how the complicated pattern obtained when several barbituratcs are present could be correctly interpreted. Cook et al. (C9) mere unable to separate barbiturates as such, but, by reaction overnight with diazomethane, they obtained good separation of the resulting diinethyl derivatives. Such a method will not resolve barbiturates which are already N-methylated from their lower nonmethylated homolog.

Rt

=

G.4S-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

281

Parker and Kirk (P2), in 1961, attempted the GLC separation of barbiturates as the free acids, using a solid support of acid-washed firebrick (apparently unsilanized) They found that the retention time varied with sample size. I n a later article, Parker et al. (P3) described the use of acid-washed Chromosorb W coated with either 5% SE-30 or 1% Carbowax 20 M in an attempt to provide a gas chromatographic screening procedure for barbiturates, alkaloids, sympathomimetic amines, and tranquilizers. Their conditions probably represent a compromise, with the result that their column was not necessarily the best that could have been prepared for resolution of mixtures containing only barbiturates. An attempt t o prevent tailing of barbiturates has been made by Cieplinski (C6), who incorporated dimer acid (0.75%) with SE-30 (1.5%) as the liquid phase. In this paper (C6) he also reported results with a liquid phase of trimer acid (0.75%) and neopentylglycol adipate (3%).However, it has been found in my laboratory th a t these columns may function satisfactorily a t 180°, but when they are heated a t 250" for 1 hour and then cooled a t 180" their performance a t 180" is quite inferior to that before the 250" treatment. Obviously, this would preclude continual temperature programming of such columns, up to 250" or so. Vanden Heuvel eb al. (V4) have briefly discussed the GLC detection of barbiturates in their more general article dealing with drugs and drug metabolites. They found that the barbiturates showed some tailing with a liquid phase of QF-1 and concluded that more than one column was necessary for satisfactory resolution of mixtures. The tristearin SE-30 column described by McMartin and Street (M11) for the analysis of barbiturates gives good results provided the column is not used above 245'. An improved column preparation has been described by Street (S12), which may be used not only for barbiturates, but for brucine and strychnine resolutions a t temperatures up to 320".

.

Method of Street1 Apparatus: Perkin-Elmer model 800 gas chromatograph fitted with a flame ionization detector. Column: 6 foot X 1/8 inch i.d. stainless steel tube packed with SE-52 on Chromosorb G. Column and support were prepared as described in Section 3.3 of this chapter. Conditions: Nitrogen, 30 ml/min. Injector temperature, 260"; detector temperature 260"; column temperature programmed a t 5" per minute from 140" to 250". 'This is an unpublished method which combines parts of methods described in various papers which have emanated from the author's laboratory from time to time (see M11, S12, 513).

282

HAROLD V. STREET

Procedure: To 65 ml of water contained in a 40 mm X 240 mm test tube, add 5 ml of blood. Mix and allow to stand for a t least 2 minutes for the cells to lake. Add 15 ml of 10% sodium tungstate (NaZWO4.2 HzO) solution, mix, and then run in 15 ml of 2% sulfuric acid solution (20 ml of conc. HzSO, per liter). Stopper the tube and shake the mixture until it changes from red to reddish-brown or brown. This takes about 1 minute. Remove the stopper and immerse the tube for 3 minutes in a bath containing vigorously boiling water. The contents of the tube must be beneath the water surface. Remove the tube and filter the hot mixture through a Whatman No. 1 paper into a 100-ml measuring cylinder. Cool the filtrate and measure its volume (usually about 90 ml). Extract the cooled filtrate by shaking with 100 ml of washed ether (“anesthetic ether” shaken with portions of 10% NaOH until no further color is extracted, then washed free of alkali). Allow the liquids to separate and run off the aqueous layer. Shake the ether phase with two 10-ml portions of phosphate buffer (0.5 M , p H 7.3) and discard the aqueous layers. Extract the organic phase with two separate 3-ml portions of 2.5% sodium hydroxide solution. Run the alkaline extracts immediately into 5 ml of 10% sulfuric acid solution (100 ml conc. H,SO, per liter) contained in a clean separating funnel. Shake this acidified solution with 25 ml of washed ether. Discard the aqueous phase and dry the ether layer by shaking with about 2 g of anhydrous sodium sulfate. Pour the ether extract into a 25-ml measuring cylinder and divide into two separate portions in 50-ml glass-stoppered flasks. Into one of the flasks place 4/10, and into the other flask 6/10, of the ether extract. Evaporate the ether in each flask to dryness on a boiling-water bath. Remove the flasks from the bath immediately the ether has evaporated and place them on a cold slab of metal t o cool. The flask containing 4/10 of extract is used for ultraviolet spectrophotometric assay. To the residue in the flask containing 6/10 of extract, add 100 yl of ethanol. Stopper the flask and rotate it to ensure complete dissolution of the residue. Inject an aliquot part (usually 2 or 3 pl) of this solution into the gas chromatograph. Figure 32 shows the results obtained by this procedure using a postmortem sample of blood which was found to contain amobarbital and secobarbital. The chromatogram of a mixture of nine “common” barbiturates is illustrated in Fig. 33. 4.5.2.3.

Basic Drugs and Morphine

4.5.2.3.1. Urine: Method of Street ( S l l ) Apparatus and Column: As described under barbiturates (Section 4.5.2.2.).

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

283

Conditions: As described under barbiturates, but with column temperatures chosen according to Table 11. Detector a t same temperature as column ; Injector about 50" above column temperature. Procedure: Acidify 5 ml of urine by adding 0.1 ml of 2 N sulfuric acid solution. Shake with 30 ml of washed ether (see below). Discard the separated organic phase. Add 0.2 ml of 2 N sodium hydroxide solution

I

6

I

4 Minutes

FIG.32. Gas chromatogram of an extract of blood from a case of suspected overdose of drugs. Total blood barbiturates (by ultraviolet spectrophotometry) = 0.7 mg per 100 ml. Amount of amobarbital+secobarbital present in the injected aliquot part of the residual solution = 0.9 pg, in 5 p1 of ethanol. For column preparation see text. Column temperature programmed from 150" a t 5" per minute, Atteeuqtion X 100. Perkin-Elmer Model 800 gas chromatograph.

284

HAROLD V. STREET

B A

I

m c 0

a

e

-

H

L

0

I

0 m +

m

n

4

I

8

I

12 Minutes

I

I

16

20

FIG. 33. Gas chromatogram of a standard mixture of nine barbiturates. Two micrograms of each barbiturate was injected in 4 p l of CHCL. For column preparation see text. Column temperature programmed from 160" at 4" per minute. Attenuation X 160. A = Barbital; B = diallylbarbituric acid; C = butethal; D = amobarbital; E = pentobarbital; F = secobarbital; G = hexobarbital; H = cyclobarbital; I = heptabarbital. F. I%M. Model 810 gas chromatograph.

(to give pH not less than 10) and extract with two separate 30-ml portions of washed ether. Keep aqueous phase (AP) for extraction of morphine (see below). Shake the combined ether extract with about 1 g of purified anhydrous sodium sulfate (purified by washing with ethanol, drying to remove ethanol, and then heating a t 600" for 20 hours). Carefully evaporate the dried ether extract to dryness in a 15-ml glass-stoppered conical centrifuge tube a t about 20" using a stream of dry nitrogen. If the analysis is not to include the more volatile basic drugs, the ether may be carefully

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

285

removed by holding the tube in the steam from a boiling-water bath. Cool the tube. Dissolve the residue in 100 p1 of ethanol and inject 5 pl of the resulting solution into the gas chromatograph. T o the aqueous phase AP (see above) add 0.1 ml of 2 N sulfuric acid solution followed by 1 ml of a saturated solution of purified sodium hydrogen carbonate (purified by washing with liberal amounts of purified ether and then used when all ether has evaporated-CAUTION! Danger of fire from large volumes of ether). Then extract with two separate 30-ml portions of washed ether. Dry the ether extract by shaking with about 1 g of purified anhydrous sodium sulfate, and evaporate carefully to dryness in a 15-ml glass-stoppered centrifuge tube held in the steam from a boiling-water bath. Dissolve the residue in 100 pl of ethanol and inject 5 pl of this solution into the gas chromatograph. Results obtained with this procedure are illustrated in Figs. 34-36. It was found that to obtain “peak-free” blanks it was essential to purify the Na,SO, and NaHCO, as described above. It was also found to

Strychnine

\

Retention time (min)

FIG.34. A chromatogram of an ether extract (see text) of 5 ml of urine containing 1 p g each of strychnine and brucine per milliliter of urine. Residue was dissolved in 100 pl of ethanol; 5 p1 of this solution was injected into thr gas chromatograph. Column temperature 300” ; attenuation X 20. Reproduced from Street (Sll) with permission.

TABLE 11 RETENTION TIME(IN MINUTES)OF ALKALOIDS SUBJECTED TO GAS-LIQUID CHROMATOGRAPHY AT VARIOUS COLUMN TEMPERATURES"sb Column temperature ("C) i\llialoitl

Mol. wt.

Amphetamine Phentermine Methylamphetamine Mephentermine Chlorphentermine Ephedrined Phenmetrazine Hordenined Meclophenoxate Metaraminole Pheniramine Metyrapone Orphenadrine Aribine Chlorpheniramine Adiphenine

135 149 149 163 184 165 177 165 242 167 240 250 269 182 275 311

130"

140"

160"

180"

205"

215"

235"

270"

290"

310"

E8 b 4

1.5 2.25 2.3 2.65 3.2 4.4 -

-

-

1.25 1.3 1.7 2.3 9.1 10.0

-

1.2 1.6 2.85 3.5

-

4.25 -

2.65 2.85 3.2 3.9 6.85

2.1 3.5

Nortriptylinec Chlorcyclizine Diazepam Chlorpromaeine Pentaquinec Acetylpromaeine Quinine Papaverine Proclorperaeined Oct averine Prolixinee Strychninec Brucinec

263 301 285 319 30 1 326 324 339 374 397 42 1 334 394

3.7 3.9 6.6 7.9 9.3 13.5 16.4 18.4

-

33.2

-

-

2.4 2.75 2.95 4.3

5.05 5.15 7.6 8.3 8.8

-

4.5

-

0

3.1 5.7

5

3 G,

Reproduced from Street (Sll) with permission. m7iththe exceptions noted, the amount of alkaloid injected was 0.1 pg in 1 pl of solution a t attenuation X 20. This is not the minimum detectable amount (MDA) but represents the amount necessary to produce a peak height of a t least 1 inch a t attenuation X 20,which is a practical attenuation limit for the extraction procedures described (see text). MDA a t attenuation X 20 = 0.2 pg. MDA a t attenuation X 20 = 0.3 pg. 8 MDA a t attenuation X 20 = 0.5 pg. a

L-d

2 2 Q

E

3

288

HAROLD V. STREET

-a x

c

0 .-+ 0

t a,

c

0

tn aJ 0 In Q

?! L 0)

E 8 a, U

8 4 0 Retention time (min)

FIG.35. A chromatogram of an ether extract (see text) of 5 ml of urine containing 0.5 pg each of imipramine, promazine, chlorpromazine, and mepazine per milliliter of urine. Residue was dissolved in 100 p l of ethanol; 5 p1 of this solution was injected into the gas chromatograph. Column temperature; 240” ; attenuation X 20. Reproduced from Street (S11) with permission.

be necessary t o purify the ether in the following manner: Shake “anesthetic” ether (diethyl ether) with 10% NaOH. Use this ether to wash a fresh solution of 10% NaOH. Use the purified alkali to wash a fresh batch of untreated “anesthetic” ether. Shake this washed ether with several portions of water until the washings are neutral. 4.5.2.3.2. Blood. In the following description of the analysis of blood, the procedure is given for basic alkaloids only. If the morphine fraction is required, a procedure similar to that described for urine (q.v.) may be applied after preliminary treatment. The first part of the extraction procedure for blood is based on Curry’s (C14) modification of the method described by Dubost and Pascal (D11) for analysis of phenothiaaine derivatives. The gas-chromatographic part of the method is taken from papers of Street (Sll, S12). As Curry has pointed out (C6), the treatment with hot hydrochloric acid also gives increased yields with alkaloids other than the phenothiazines. This is probably due to liberation of more protein-bound drug. It is suggested, therefore, that, provided the drug is

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

289

stable under such conditions, the hot HC1 treatment may be used for analysis of all alkaloids in protein-containing fluids and tissues. In the following procedures, it may be advisable to use sodium carbonate in place of potassium hydroxide in those cases. where the drug may be sensitive to a high local concentration of hydroxyl ions. Procedure: To 5 ml of blood in a boiling tube add 1 ml of water and 4 ml of concentrated hydrochloric acid solution. Place the tube in a boiling-water bath for 5 minutes and then cool in ice water. Add about 6 ml of ether-washed 60% aqueous potassium hydroxide solution, making certain that the mixture is a t least p H 10. Shake with two separate 30-ml portions of washed ether. Wash the combined ether extracts with 5 ml of 2.5% ether-washed sodium hydroxide solution and then with two separate 5-ml portions of water. Shake the separated, washed ether extract with 5 ml of 0.1 N sulfuric acid solution. At this stage, this acid extract may be subjected (after bubbling nitrogen through the solution to remove dissolved ether) to ultraviolet spectrophotometry. To the acid extract, add 60% ether-washed potassium hydroxide solu-

7 Morphine

6 (u

I

c

0 .c 0

a

c al c 0 d

8c 0 In a

?! L. 0)

E

0

B

L ,

I2 8 4 Retention time (min) FIG.36. A chromatogram of an ether extract (see text) of 2.5 1111 of urine containing 7 5 pg each of morphine and nalorphine per milliliter of urine. Residue was dissolved in 100 p1 of ethanol; 5 p1 of this solution were injected into the gas chromatograph. Column temperature ; 240" ; attenuation X 20. Reproduced from Y t i eet (S11)with permission.

290

HAROLD V. STREET

21 10

Retention time. minutes

FIG.37. Gas chromatogram of a synthetic mixture of C2 to Clt fatty acids as the free acids. Aluminum column, 2 meters X y4 inch; 0.25% Carbowax and 0.4% isophthalic acid on 200-p glass microbeads (acid-washed) . Temperature-programmed from 90" to 175" at 3.3" per minute. Sample 02 pl; flow rate, 80 ml of helium per minute. Identification of peaks: 1, acetic; 2, propionic; 3, isobutyric; 4, butyric; 6, isopentanoic ; 6, pentanoic ; 7, hexanoic; 8, 2-ethylhexanoic ; 9, octanoic ; 10, nonanoic; 11, decanoic; 12, undecanoic ; 19, undecenoic. Reproduced from Nikelly (N4) with permission.

tion drop by drop to give a p H of about 10 units. Then extract with 30 ml of washed ether. Dry the ether phase with purified anhydrous sodium sulfate. Then proceed as described under urine (Section 4.5.2.3.1.), i.e., evaporate to dryness, dissolve the residue in ethanol, and inject a suitable aliquot part of the resulting solution into the gas chromatograph. For quantitative results an internal standard may be included in the ethanol.

APPLICATIONS 4.6. MISCELLANEOUS 4.6.1.

F a t t y Acids and Other Carboxylic Acids

The classical example in this field is, of course, the original work of James and Martin (J3),who separated anhydrous volatile fatty acids on columns of Celite impregnated with a mixture of silicone oil and stearic acid. Separation by GLC of volatile organic acids in the presence of water, using a liquid phase of diethylene glycol adipate polymers, has been described (H20). Various derivatives and methods of preparing derivatives of the fatty acids have been suggested, such as formation of alkyl esters (E3, G2, L12, M9, M12, M13, S5, V7) and 2-chloroethanol esters (01).Achman and Burgher ( A l ) found that in the GLC estima-

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

291

tion of volatile fatty acids in aqueous media, the addition of formic acid vapor to the carrier gas produced improved results. The main problem in the GLC analysis of fatty acids stems from the presence of the polar COOH group. This leads to adsorption which is seen on the chromatogram as reduced peak height and as tailing. Although perhaps the most obvious way to overcome this problem is by ester formation, much attention has been focused on the modification of the columns to reduce adsorption, and so permit GLC of the free fatty acids. For example, Nikelly (N4) was able to separate free fatty acid homologs and isomers from C, to CI8 (inclusive) on columns made by coating dilute-acid-washed glass beads (200 p) with 0.25% Carbowax 20 M and 0.4% isophthalic acid. The results obtained (N4) for synthetic mixtures of fatty acids are shown in Fig. 37 (for C, to Cll) and in Fig. 38 (for to CIEd. Carbowax columns containing terephthalic acid (B29) have also been proposed. However, most promising results for the volatile fatty acids in aqueous solution have been obtained by Mahadevan and Stenroos (M3), who used Porapak Q coated with phosphoric acid. These workers were interested in the volatile fatty acids present in the breath and serum of patients with hepatic coma, and therefore required a rapid and reliable method for estimating these acids in aqueous solution. They found that, with untreated Porapak Q, the acids failed to emerge from the column but that symmetrical peaks were obtained by coating the Porapak Q with phosphoric acid. The amount of phosphoric acid needed to remove the binding sites from the Porapak Q was found to be dependent upon

c,,

0

Ic n 0

% $!

L

a,

E 8

2 15

30

45

60

Retention time, minutes

FIG.38. Gas chromatogram of a synthetic mixture of CISt o CIS fatty acids as the free acids. Column as in Fig. 37. Flow rate of helium 70 ml per minute; temperature 180"; sample 0.2 pl of 50% acetone solution. Identification of peaks: 1, palmitic; 2, palmitoleic; 3, stearic; 4, linoleic; 6, linolenic. Reproduced from Nikelly (N4) with permission.

292

HAROLD V. STREET

Minu tes

FIG.39. Gas chromatographic separation of short-chain fatty acids (as the free acids) in aqueous solution. Column: aluminum 8 foot X I/s inch; Porapak Q,(150200 mesh) 4% (w/w) phosphoric acid. Column temperature, 220"; flash heater,

+

275" ; flame ionization detector, 280" ; flow rate, 35 ml helium per minute; sample volume, 1 pl. Identification of peaks: A, acetic; B, propionic; C, isobutyric; D, butyric; E , isovaleric; F, valeric. Reproduced from Mahadevan and Stenroos (M3) with permission.

the mesh size of the material. For 80-100 mesh material, 2% phosphoric acid was sufEcient whereas for a mesh size of 150-200, it was necessary to use 4% phosphoric acid. Figure 39 shows the chromatogram obtained (M3) for several short-chain fatty acids in aqueous solution. 4.6.2. Pyrolysis Pyrolysis of samples in conjunction with gas chromatography was originally used for examining nonvolatile material. I n 1960 a micropyrolytic-gas chromatographic technique was devised (L4) for the characterization of organophosphorus compounds. Pyrolysis was effected in the inlet stream of a gas chromatograph, and the volatile pyrolytic products emerging from the column were collected individually and subjected to mass or infrared spectrometry. A hot-wire pyrolysis unit has been described by Vassallo (VS) for the "pyrolytic analysis" and characterization of organic polymers. The products were examined directly by gas chromatography or mass spectrometry. The application of pyrolysis to the identification of barbituric acid derivatives was studied by Nelson and Kirk (N3). Their pyrolysis unit is shown in Fig. 40. Cornparison was made of the free acids, the sodium salts, and mixtures of free acid with anhydrous potassium carbonate. The pyrolysis products of 27 substituted barbituric acids were studied using a Pye argon gas chromatograph. While this technique may have its uses when only a

GAS-LIQUID CHROMATOGRAPHY I N CLINICAL CHEMISTRY

293

single drug is present, it is difficult to see how the procedure could be used successfully to identify a mixture of say three or four barbituric acid derivatives. I n an interesting article on the pyrolysis-gas chromatography of a number of simple organic molecules, Wolf and Rosie (W10) have tabulated some of their results according to the functional group present in the original compound. The size of the pyrolysis tube which they used was large enough to bring almost the whole of the sample to the pyrolysis temperature. They conclude that when using pyrolysis as an analytical tool, best results may be obtained by using only partial breakdown in the 700”-900”C range. These conditions would also minimize secondary reactions. The pyrolysis of tetraalkylammonium salts of aliphatic and aromatic acids has been investigated by Downing (D10) and Bailey ( B l ) . It was found ( B l ) that in order to obtain complete conversion of the tetramethylammonium salts of organic acids to their methyl esters, it was necessary to pack loosely the injection port of the GLC apparatus with glass wool. It was also found (Bl) that the injection port must be maintained between 360’ and 400°C and that when the sample is being injected, minimum penetration of the septum by the needle is necessary. The salts were prepared from the organic acids by titration with a methanolic solution of the appropriate tetraalkylammonium hydroxide. I n a paper with the very modest title “Electrical discharge pyrolyzer for gas chromatography,” Sternberg and Litle (S10) have described, among other things, the pyrolysis of mesoporphyrin IX. Their electrical discharge technique, used a t a temperature of about 600°C, provides valuable qualitative information because large characteristic fragments remain intact, and yet produces sufficient breakdown products to permit c,haracterization.

C FIG.40. Pyrolysis unit. A, Foil heating cup; B, wires to variable voltage emf; (7, ground joint connection to top of chromatographic column. Reproduced from Nelson and Kirk (N3) with permission.

294

HAROLD V. STREET

During their studies of the biochemical role of peroxidation, Blondin et al. (B15) developed a method for distinguishing between steroidal peroxides and other sterols. By using a flash heater a t 350°C, it was found that steroidal 5,8-peroxides decomposed to give a number of peaks, whereas a single peak was obtained with most sterols under the same conditions. The use of pyrolysis in peak identification in gas chromatography has been reviewed recently by Beroza and Coad (B9) and by Perry (P7). Improvements in the design and pyrolysis units for gas chromatography have been discussed by Levy (L6).

4.6.3. Catalytic Hydrogenation in Carbon-Skeleton Chromatography Carbon-skeleton chromatography was reported and subsequently developed by Beroza and co-workers (BCB8, B10-B12). This technique involves catalytic hydrogenation of substances a t elevated temperatures to produce hydrocarbons, which are then separated by GLC and identified by retention times and, sometimes, by other means (e.g., mass spectrometry). Under certain conditions, the hydrocarbon produced is the carbon skeleton of the original substance. Such a process is obviously of great value in providing an additional parameter in the problem of determination of chemical structure. Beroza (B9) has reviewed the developments in this field up to 1966. His procedure uses a precolumn containing a catalyst of 1% palladium on Gas Chrom P (60-80 mesh) which is maintained a t an elevated temperature (300’). This precolumn is connected directly with a flame ionization detector. Ten to 20 micrograms of material were usually injected into the precolumn, but valid results were also obtained with as little as 0.5 pg. An extension of Beroza’s work has been made by Adhikary and Harkness (A4) in their study of the carbon skeletons obtained by thermal catalytic hydrogenation of steroids and sterols. These workers found that each steroid gave rise to two peaks, corresponding to the 5a- and 5pforms. As the temperature of the catalyst was raised, the proportion of the more stable 5a-form of the hydrocarbon increased. By keeping the catalyst temperature a t 17O0-20O0 the ratio of 5p- to 5a-forms was about unity, thus enabling more useful information about the structure of the original material to be obtained than when only the 5a-form is p r e s e n t as is the case with too high temperatures. It was also found (A4) that with diatomaceous earth supports, neither palladium nor platinum was a satisfactory catalyst for steroid work. This was presumably due to cracking of the hydrocarbons under the influence of the “active sites” on the support. Best results were obtained with a highly active catalyst 011 “siliconized” glass beads (40mesh). With this type of support, platinum

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

295

was found to be superior to palladium. The glass tube containing the catalyst was also found to require “siliconization.” The reduction products were collected in chloroform contained in a tube cooled in ice, and were subsequently subjected to GLC. Figure 41 shows the construction of the apparatus used for the catalytic reduction. The catalyst was prepared as follows: Place 50 g of cleaned “siliconized” glass beads (40mesh) in 100 ml of 1% (w/v) ethanolic solution of chloroplatinic acid. Remove the solvent by evaporation on a boilingwater bath with the aid of a stream of air. Mix the beads and solution continuously during this evaporation process. Then dry the mixture in an oven a t 105’ overnight. This procedure gives about 1 3 % platinum (w/w) coated on the glass beads. When kept in a desiccator, the catalyst remains active for a t least 8 months. I n a personal communication to the present writer, Harkness (H5) pointed out that a number of steroid drugs with hydrocarbon skeletons which are not found in the naturally occurring steroids have been reduced and the products separated by GLC (A3). One application of such a technique in clinical chemistry is the detection of steroid drugs which in vivo are completely altered metabolically without changing their carbon skeletons. For example, after the administration to normal men of therapeutic doses of the anabolic steroid 17a-methyl-17p-hydroxyandrosta-1,4-dien-3-one (Dianabol; Ciba) none of the unaltered drug could be detected in urine. A further difficulty in detecting their administration Asbestos tape Heating tape I Aluminum cylinder

I I 07 0 and 0 joint

Thermometer

Beaker contoining ice

y Gas purifying bottle

T

On and off valve

FIG.41. Cross-sectional view of apparatus used for catalytic reduction to carbon skeleton. Catalyst tube measures 21.7 cm X 0.7 cm 0.d. Reproduced from Adhikary and Harkneas (A4) with permission.

296

HAROLD V. STREET

is that blood levels of such drugs are likely to be low and to fall rapidly. Although a metabolite of Dianabol is excreted in urine (R2),this metabolite cannot be obtained commercially. This metabolite and others can be reduced by the above technique to a product which has the same retention time as the product from the drug itself. Thus the drug and its metabolites are probably reduced to their carbon skeleton, which is a distinctive common feature of both the drug and its metabolites (A2). From the evidence at present available, the hydrocarbon skeletons of steroid drugs are rarely altered metabolically. Chromatographic evidence, however, provides tentative identification which may need to be supplemented by information from other physical methods. Harkness (H5) has concluded that the carbon skeleton of a steroid drug can be used as a sensitive method of detecting its administration despite the alteration of that drug to unknown metabolites. 4.6.4.

Other Uses of Gas Chromatography

Table 12 lists a number of miscellaneous applications of GLC which are of interest t o clinical chemists. TABLE 12

MISCELLANEOUS GLC APPLICATIONS Substance or technique Vitamin A Vitamins of B6 group Urinary methylmatonic acid (indicative of vitamin BI2deficiency) Vitamin C Vitamins D1 and Dt Vitamin E and tocopherols showing vitamin E activity Vitamin K1 Oxygen and carbon dioxide in blood Oxygen and carbon monoxide in blood Oxygen and nitrogen in blood Lactic acid in blood Acetylcholine Acetylsalicylic acid Automated GLC/infrared analysis Cardiac glycosidea Cholesterol in serum Glutethimide Pesticides Phenolic acids and alcohols in urine Urinary and tissue metabolites including phenols, alcohols, fatty acids, di- and tricarboxylic acids, keto acids, steroids, glucuronidw, etc.

Reference

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

297

ACKNOWLEDGMENTS Thanks are due t o the editors of the following journals who kindly gave me permission t o reproduce the relevant figures and/or tables from their publications : Analytical Biochemistry, Analytical Chemistry (copyright American Chemical Society), Chemistry and Industry, Clinical Chemistry, Clinica Chimica Acta, Journal of Chromatography, Journal of Gas Chromatography, Journal of Laboratory and Clinical Medicine. Permission to reproduce material from their publications is also gratefully acknowledged from Springer-Verlag (Heidelberg), Professor E. C. Horning, Dr. D. M. Ottenstein, and Dr. R. H. Travis. I am grateful to Mrs. G. M. Hamilton for typing the manuscript of this chapter.

REFERENCES A l . Ackman, R. G., and Burgher, R. D., Quantitative gas liquid chromatographic estimation of volatile fatty acids in aqueous media. Anal. Chem. 36, 647-652 (1963). A2. Adhikary, P. M., and Harkness, R. A., Catalytic hydrogenation, etc. Personal communication (1968). A3. Adhikary, P. M., and Harkness, R. A., High-temperature catalytic reduction of steroids. Biochem. J . 106, 40P (1967). A4. Adhikary, P. M., and Harkness, R. A., Determination of the carbon skeletons of microgram amounts of steroids and sterols by gas chromatography after their high temperature catalytic reduction. Anal. Chem. 41, 470-476 (1969). A5. Alexander, R. J., and Garbutt, J. T., Use of sorbitol as internal standard in determination of D-glucose by gas liquid chromatography. Anal. Chem. 37,303-305 (1965). A6. Ambrose, D., Preliminary recommendations on nomenclature and presentation of data in gas chromatography. Pure Appl. Chem. 1, 177-186 (1960). A7. Anders, M. W., and Mannering, G. J., Application of gas chromatography t o toxicology. Progr. Chem. Tozicol. 3, 121-218 (1967). A8. Aspinall, G. O., Gas-liquid partition chromatography of methylated and partially methylated methyl glycosides. J . Chcm. Soc. pp. 1676-1680 (1963). A9. Atkinson, A. P., and Tuey, G. A. P., In situ treatment of Kieselguhr for gas chromatography. Nature 199, 482483 (1963). A10. Attal, J., Hendeles, S. M., and Eik-Nes, K. B., Determination of free estrone in blood plasma by gas-phase chromatography with electron capture detection. Anal. Bicchem. 20, 394409 (1967). A l l . Avioli, L. V., and Lee, S. W., Detection of nanogram quantities of vitamin D by gas-liquid chromatography. Anal. Biochem. 16, 193-199 (1966). B1. Bailey, J. J., Determination of aliphatic and aromatic acids by pyrolysis of their tetramethylammonium salts. Anal. Chem. 39, 148,51489 (1967). B2. Barry, R . D., Guarnieri, M., Besch, P. K., Ring, W., and Besch, N. F., Comparison of four methods of determining urinary pregnanediol. Anal. Chsm. 38, 983-986 (1966). €33. Basila, M. R., Hydrogen bonding interaction between adsorbate molecules and surface hydroxyl groups on silica. J . Chem. Phys. 36, 1151-1158 (1961). B4. Bayer, E., and Widder, R., GZMchromatographic determination of the carbonyl forms of sugars. Anal. Chem. 36, 1452-1455 (1964). B5. Bentley, R., Sweeley, C. C., Makita, M., and Wells, W. W., Gas chromatography of sugars and other polyhydroxy compounds. Biochem. Riophys. Res. Commun. 11, 14-18 (1963).

298

HAROLD V. STREET

B6. Beroza, M., Ultramicrodetermination of chemical structure of organic compounds by gas chromatography. Nature 196, 768-769 (1962). B7. Beroza, M., Determination of the chemical structure of microgram amounts of organic compounds by gas rhromatography. Anal. Chem. 34, 1801-1811 (1962). B8. Beroza, M., Alkaloids from tripterygium wilfordii hook. The chemical structure of wilfordic and hydroxywilfordic acids. J . Org. C h . 28, 3562-3564 (1963). B9. Beroza, M., and Coad, R. A., Reaction gas chromatography. J . Gus Chromatog. 4, 199-216 (1966). B10. Beroza, M., and Sarmiento, R., Determination of the carbon skeleton and other structural features of organic compounds by gas chromatography. Anal. Chem. 36, 1353-1357 (1963). B11. Beroza, M., and Sarmiento, R., Carbon skeleton chromatography using hot-wire thermal conductivity detection. Anal. Chem. 36, 1744-1750 (1964). B12. Beroza, M., and Sarmiento, R., New catalyst and technique for analyzing fatty acids, their esters, and long-chain compoundsby carbon-skeleton chromatography. Anal. Chem. 37, 1040-1041 (1965). B13. Bier, M., and Teitelbaum, P., Gas chromatography in amino acid analysis. Ann. N . Y . A d . Sn’. 72, 641-648 (1958). B14. Blau, K., and Darbre, A., Preparation of volatile derivatives of amino acids for gas chromatography. Biochem. J . 88, 8P (1963). B15. Blondin, G. A., Kulkarni, B. D., John, J. P., van Aller, R. T., Russell, P. T., and Nes, W. R., Identification of steroidal 5,8-peroxidesby gas-liquid Chromatography. Anal. Chem. 39, 36-40 (1967). B16. Bohemen, J., Langer, S. H., Perrett, R. H., and Purnell, J. H., A study of the adsorptive properties of firebrick in relation to its use as a solid support in gasliquid chromatography. J . Chem. Soc. pp. 2444-2451 (1960). B17. Bonnichsen, R., and Linturi, M., Gas chromatographic determination of some volatile compounds in urine. Acta Chem. Scand. 16, 1289-1290 (1962). B18. Braman, R. S., Flame emission and dual flame emission-flame ionization detectors for gas chromatography. Anal. C h . 38, 734-742 (1966). B19. Bravo, E. L., and Travis, R. H., A quantitative gas-liquid chromatographic method for aldosterone in human urine. J . Lab. Clin. Med. 70, 831-840 (1967). B20. Brochmann-Hanssen, A., and Svendsen, A. B., Separation and identification of barbiturates of some related compounds by means of gas-liquid chromatography. J . Pharm. Sci. 61, 31&321 (1962). B21. Brochmann-Hanssen, A., and Svendsen, A. B., Separation and identification of sympathomimetic amines by gas-liquid chromatography. J . Pharm. Sci. 61, 938-941 (1962). B22. Brook, J. L., Biggs, R. M., St. John, P. A., and Anthony, D. S., Gas chromatography of several indole derivatives. Anal. Biochem. 18, 453-458 (1967). B23. Brooks, C. J. W., Studies of acetylated corticosteroidsand related 2O-oxopregnane derivatives by gas liquid chromatography. Anal. Chem. 37, 636-641 (1965). B24. Brooks, C. J. W., and Horning, E. C., Gas chromatographic studies of catecholamines, tryptamines and other biological aminw. Part I. Catecholamines and related compounds. Anal. Chem. 36, 1540-1545 (1964). B25. Brooks, R. V., Fractionation of urinary steroids prior to gas chromatography. I n “The Gas Liquid Chromatography of Steroids” (J. K. Grant, ed.), pp. 143-154. Cambridge Univ. Press, London and New York, 1967. B26. Brown, J. B., A chemical method for the determination of oestriol, oestrone and oestradiol in human urine. Biochem. J. 60, 185-193 (1955). B27. Burchfield, H. P., and Storrs, E. E., “Biochemical Applications of Gas Chromatography.” Academic Press, New York, 1962.

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

299

B28. Burger, J. D., Gas chromatographic behaviour of various types of organic compounds on Porapak Q. J . Gas Chromatog. 6, 177-178 (1968). B29. Byars, B., and Jordan, G., Efficient packed column for free fatty acids. J . Gas Chromatog. 2, 304-305 (1964). C1. Cadman, W. J., and Johns, T., Gas chromatographic determination of ethanol and other volatiles from blood. Pittsburgh Conf. Anal. Chem. Appl. Spectry., 1958. C2. Callear, A. B., and Cvetanovits, R. J., The application of gas-liquid partition chromatography to problems in chemical kinetics. Can. J . Chem. 33, 1256-1267 (1955). C3. Capella, P., and Horning, E. C., Separation and identification of derivatives of biologic amines by gas-liquid chromatography. Anal. Chem. 38, 316-321 (1966). C4. Cawley, L. P., Musser, B. O., Campbell, S., and Faucette, W., Analysis of total serum cholesterol by means of gas liquid chromatography. Am. J . Clin. Pathol. 39, 450-455 (1963). C5. Cawley, L. P., Musser, B. O., Faucette, W., Beckloff, S., and Learned, H., Evaluation of hydrolytic products of 17-ketosteroids by means of gas-liquid chromatography. Clin. Chem. 11, 1009-1018 (1965). C6. Cieplinski, E. W., Prevention of peak tailing in the direct gas chromatographic analysis of barbiturates. Anal. Chem. 36, 256-257 (1963). C7. Clarke, D. D., Wilk, S., Gitlow, S. E., and Franklin, M. J., Gas chromatographic determination of dopamine at the nanogram level. J . Gas Chromatog. 6, 307-310 (1967). C8. Clarke, D. D., Wilk, S., Gitlow, S. E., and Franklin, M. J., Gas chromatographic determination of dopamine a t the nanogram level. I n “Advances in Gas Chromatography” (A. Zlatkis, ed.), pp. 145-148. Preston Tech. Abstr. Co., Evanston, Illinois, 1967. C9. Cook, J. G. H., Riley, C., Nunn, R. F., and Budgen, D. E., Gas chromatography of methyl derivatives of some barbiturates. J . Chromatog. 6, 182-185 (1961). ClO. Copenhaver, J. H., Quantitative analysis of plasma galactose and glucose by gas-liquid chromatography. Anal. B i o c h . 17, 76-83 (1966). C11. Coulter, J. R., and Hann, C. S., A practical quantitative gas chromatographic analysis of amino acids using the n-propyl N-acetyl esters. J . Chromatog. 36, 42-49 (1968). C12. Crippen, R. C., and Freimuth, H. C., Determination of aspirin by gas chromatography. Anal. Chem. 36, 273-275 (1964). C13. Cruickshank, P. A., and Sheehan, J. C., Gas chromatographic analysis of amino acids as N-trifluoroacetylamino acid methyl esters. Anal. Chem. 36, 1191-1 197 (1964). C14. Curry, A. S., “Poison Detection in Human Organs.” Thomas, Springfield, Illinois, 1963. C15. Curry, A. S., Hurst, G., Kent, N. R., and Powell, H., Rapid screening of blood samples for volatile poisons by gas chromatography. Nature 196, 603-604 (1962). C16. Curry, A. S., Walker, G. W., and Simpson, G. S., Determination of ethanol in blood by gas chromatography. Analyst 91, 742-743 (1966). C17. Curtius, H. C., Buergi, W., and Keller, K., Gas chromatographic determination of serum cholesterol. 2. Klin. Chem. 4, 38-42 (1966). C18. Curtius, H. C., and Muller, M., Gas-liquid chromatography of 17-ketosteroids and progesterone metabolites of urine: Comparison of different methods of hydrolysis. J. Chromatog. 30, 410427 (1967). D1. Dalgliesh, C. E., Horning, E. C., Horning, M. G., Knox, K. L., and Yarger, K., A gas-liquid-chromatographicprocedure for separating a wide range of metabolites occurring in urine or tissue extracts. Biochem. J . 101, 792-810 (1966).

300

HAROLD V. STREET

D2. Dal Nogare, S., and Juvet, R. S., Gas chromatography. Anal. Chem. 34,35R-47R (1962). D3. Dal Nogare, S., and Juvet, R. S., Gas chromatography. Anal. Chem. 38, 61R-79R (1966). D4. Darcey, B., and Evenson, M. A., Gas liquid chromatographic fractionation of the 17-ketosteroids for normal adults. Clin. Chem. 12, 533-534 (1966). D5. Davis, J. W., and Furst, A., Preparation of alkylidine and alkyl amino acid esters for the gas chromatographic analysis of amino acids. Anal. Chem. 40, 1910-1912 (1968). D6. Davis, R. A., The determination of ethanol in blood or tissue by gas chromatography. J . Forensic Sci. 11, 205-213 (1966). D7. Decora, A. W., and Dinneen, G. U., Gas-liquid chromatography of pyridines using a new solid support. Anal. Chem.32, 164-169 (1960). D8. Desty, D. H., and Harbourn, C. L. A., Evaluation of a Commercial alkyl aryl sulfonate detergent as a column packing for gas chromatography. Anal. Chem. 31, 1965-1970 (1959). D9. Dewar, R. A., and Maier, V. E., High efficiency glass bead columns for gas chromatography. J . Chromatog. 11, 295-300 (1963). D10. Downing, D. T., Analysis of aqueous solutions of organic acids by pyrolysis of their tetramethylammonium salts in the gas chromatograph. Anal. Chem. 39, 218-221 (1967). D11. Dubost, P., and Pascal, S., Dosage du largactil dans les liquides biologiques. Ann. P h a m . Franc. 11, 615-619 (1953). D12. Dunagin, P. E., Jr., and Olson, J. A., Gas liquid chromatography of retinol (Vitamin A) derivatives. Anal. Chem. 36, 756-759 (1964). El. Eaborn, C., Walton, D. R. M., and Thomas, B. S., Preparation and gas chromatography of steroid bromomethyldimethylsilyl ethers. Chem. & Ind. (London) p. 827 (1967). E2. Eik-Nes, K. B., and Homing, E. C., eds., “Gas Phase Chromatography of Steroids.” Springer, Berlin, 1968. E3. Estes, F. L., and Bachmann, R. C., Gas chromatographic determination of pymvic and lactic acids and Krebs cycle components. Esterification and recovery. Anal. Chem. 38, 1178-1182 (1966). E4. Ettre, L. S., “Open Tubular Columns in Gas Chromatography.’’ Plenum Press, New York, 1965. E5. Exley, D., The ultramicro detection of steroids using gas liquid chromatography with electron capture detection. I n “The Gas Liquid Chromatography of Steroids” (J. K. Grant, ed.), pp. 117-128. Cambridge Univ. Press, London and New York, 1967. E6. Exley, D., Ultramicro determination of plasma testosterone by electron-capture detection of testosterone diheptafluorobutyrrtte. Biochem. J . 107, 285-292 (1968). F1. Ferrier, R. J., and Singleton, M. F., The trimethylsilation of xylose and the nuclear magnetic resonance spectra of the major products. Tetrahedron 18, 11431148 (1962). F2. Fox, J. E., Gas chromatographic analysis of alcohol and certain other volatiles in biological material for forensic purposes. Proc. Soc. Ezptl. Biol. Med. 97, 236-237 (1958). G1. Gee, M., Preparation of methyl esters of amino acids for gas chromatography using dimethyl dodecanedioate as an internal standard. Anal. Chem. 39,1677-1679 (1967).

GAS-LIQUID

CHROMATOGRAPHY I N CLINICAL CHEMISTRY

301

G2. Gehrke, C. W., and Goerlitz, D. F., Quantitative preparation of methyl esters of fatty acids for gas chromatography. A n d . Chem. 36, 76-80 (1963). G3. Giddings, J. L., Liquid distribution on gas chromatographic support. Relationship to plate height. Anal. Chem. 34, 458-465 (1962). G4. Golay, M. J. E., Brief report on gas chromatographic theory. In “Gas Chromatography 1960” (R. P. W. Scott, ed.), pp. 139-144. Butterworth, London and Washington, D. C., 1960. G5. Goldbaum, L. R., Schloegel, E. L., and Dominguez, A. M., Application of Gas chromatography to toxicology. Progr. Chem. Toxicol. 1, 11-52 (1963). G6. Graff, J., Wein, J. P., and Winitz, M., Quantitative determination of amino acids by gas liquid chromatography. Federation Proc. 22, 244 (1963). G7. Greer, M., and Williams, C. M., Diagnosis of homocystinuria by gas chromatography. Anal. Biochem. 19, 40-45 (1967). G8. Guarnieri, M., and Barry, R. D., Simultaneous determination of pregnanediol and pregnanolone in urinary extracts by gas chromatography. Clin. Chem. 14, 35-37 (1968). H1. Hagen, P., and Black, W., A method for the quantitative determination of the composition of a mixture of 19 amino-acids by gas chromatography of their N-trifluoroacetyl methyl esters. Federation Proc. 23, 371 (1964). H2. Halasz, I., and Horvath, C., Micro beads coated with a porous thin layer as column packing in gas chromatography. Some properties of graphited carbon black as stationary phase. Anal. Chem. 36, 1178-1186 (1964). H3. Hammond, K. B., and Leach, H., A routine method for the determination of pregnanediol and pregnanolone by gas chromatography. Clin. Chim. Acta 16, 145-148 (1967). H4. Hansen, L. B., Gwliquid chromatographic separation of thyroid hormones. Anal. Chem. 40, 1587-1589 (1968). H5. Harkness, R. A., Personal communication (1968). H6. Hedgeley, E. G., and Overend, W. G., Trimethylsilyl derivatives of carbohydrates. Chem. & Ind. (London) pp. 378-380 (1960). H7. Hockey, J. A., Surface properties of silica powders. Chem. & Ind. (London) pp. 57-63 (1965). H8. Hoek, W., andvan Kampen, E. J., Urinary steroid spectra by gas chromatography. Clin. Chim. A d a 19, 371-381 (1968). H9. Hoffman, N. E., and Barboriak, J. J., Gas chromatographic determination of urinary methylmalonic acid. Anal. Biochem. 18, 10-17 (1967). 1310. Hollis, 0. L., Separation of gaseous mixtures using porous polyaromatic polymer beads. Anal. Chem. 38, 309-316 (1966). H11. Hollis, 0. L., “Advances in Gas Chromatography, 1965” (A. Zlatkis and S. Ettre, eds.). Preston Tech. Abstr. Co. Evanston, Illinois, 1966. 1312. Holmes, W. L., and Stack, E., Gas chromatography of squalene, sterols and bile acid methyl esters. Biochim. Biophys. A d a 66, 163-165 (1962). 1113. Horning, E. C., Gas phase analytical methods for the study of steroid hormones and their metabolites. In “Gas Phase Chromatography of Steroids” (K. B. Eik-Nes and E. C. Horning, eds.), pp. 1-71. Springer, Berlin, 1968. H14. Homing, E. C., Horning, M. G., Ikekawa, N., Chambaz, E. M., Jaakonmaki, P. I., and Brooks, C. J. W., Studies of analytical separations of human steroids and steroid conjugates. In “Advances in Gas Chromatography” (A. Zlatkis, ed.), pp. 122-134. Preston Tech. Abstr. Co., Evanston, Illinois, 1967. €115. Homing, E. C., Horning, M. G., Vanden Heuvel, W. J. A., Knox, K. L.,

302

HAROLD V. STREET

Holmstedt, B., and Brooks, C. J. W., Part 11. Tryptamine-related amines and catecholamines. Anal. Chem. 36, 1546-1549 (1964). H16. Horning, E. C., Moscatelli, E. A., and Sweeley, C. C., Polyester liquid phases in gas-liquid chromatography. Chem. & Ind. (London) pp. 751-752 (1959). H17. Horning, M. G., Knox, K. L., Dalgleish, C. E., and Horning, E. C., Gas-liquid chromatographic study and estimation of several urinary aromatic acids. Anal. Biochem. 17, 244-257 (1966). H18. Horning, M. G., Moss, A. M., and Horning, E. C., A new method for the separation of the catecholamines by gas-liquid chromatography. Biochim. Bwphys. Acta 148, 597-600 (1967). H19. Hunter, I. R., Dimick, K. P., and Corse, J. W., Determination of amino acids by ninhydrin oxidation and gas chromatography. Chem. & Ind. (London) pp. 294-295 (1956). H20. Hunter, I. R., Ortegren, V. H., and Pence, J. W., Gas chromatographic separation of volatile organic acids in presence of water. Anal. Chem. 32, 682-684 (1960). J1. Jaakonmaki, P. I., and Stouffer, J. E., Gas chromatographic analysis with electron capture detection of thyroid hormones. I n “Advances in Gas Chromatography” (A. Zlatkis, ed.), pp. 149-152. Preston Tech. Abstr. Co., Evanston, Illinois, 1967. 52. Jain, N. C., Fontan, C. R., and Kirk, P. L., Simplifiedgas chromatographic analysis of pesticides from blood. J . Pharm. Pharmacol. 17, 362-367 (1965). 53. James, A. T., and Martin, A. J. P., Gas-liquid partition chromatography: The separation and micro-estimation of volatile fatty acids from formic acid to dodecanoic acid. Biochem. J . 60, 679-690 (1952). 54. Janak, J., Identification of the structure of nonvolatile organic substances by gas chromatography of pyrolytic products. Nature 186,684-686 (1960). 55. Jelliffe, R. W., and Blankenhorn, D. H., Gas chromatography of digitoxigenin and digoxigenin. J . Chromatog. 12, 268-270 (1963). J6. Jenden, D. J., Hanin, I., and Lamb, S. I., Gas chromatographic microestimation of acetylcholine and related compounds. Anal. Chem. 40, 125-128 (1968). 57. Johnson, D. E., Scott, S. J., and Meister, A., Gas liquid chromatography of amino acid derivatives. Anal. Chem. 33, 669-673 (1961). J8. Jones, H. G., Jones, J. K. N., and Perry, M. B., The gas liquid partition chromatography of carbohydrate derivatives. 111. The separation of amino glycose derivatives and of carbohydrate, acetal and ketal derivatives. Can. J . Chem. 40, 1559-1563 (1962). J9. Juvet, R. S., and Dal Nogare, S., Gas chromatography. Anal. Chem. 36, 36R-51R (1964). J10. Juvet, R. S., and Dal Nogare, S., Gas chromatography. Anal. Chem. 40, 33R-50R (1968). Jll. Juvet, R. S., Jr., and Durbin, R. P., Characterization of flame photometric detector for gas chromatography. Anal. Chem. 38, 565-569 (1966). K1. Kaiser, R., “Gas Phase Chromatography, Vol. 1: Gas Chromatography.” Butterworth, London and Washington, D. C., 1963. K2. Karoum, F., Ruthven, C. R. J., and Sandler, M., Gas chromatographic measurement of phenolic acids and alcohols in human urine. Clin. Chim. A d a 20, 427-437 (1968). K3. Kasyak, L., and Knoblock, E. C., Application of gas chromatography to analytical toxicology. Anal. Chem. 36, 1448-1452 (1963). K4. Kirkland, J. J., Porous thin-layer modified glass bead supports for gas liquid chromatography. Anal. Chem. 37, 1458-1461 (1965). K5. Kirkland, J. J., Some modified gas chromatographic adsorbents and supports.

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

I n “Gas Chromatography Proc. Symp. 5, 1964” (A. Goldup,

303

ed.), pp. 285-300.

Inst. Petrol., London, 1965. K6. Klebe, J. F., Finkbeiner, H., and White, D. M., Silylations with bis(trimethylsily1)acetamide, a highly reactive silyl donor. J. Am. Chem. SOC.88, 3390-3395 (1966). K7. Klesper, E., Corwin, A. H., and Turner, D. A., High pressure gas chromatography above critical temperatures. J . Org. Chem. 27, 700-701 (1962). K8. Kliman, B., and Briefer, C., Jr., Collection of carbon-14 and tritium labelled steroids in gas liquid chromatographywith application to the analysis of testosterone in human plasma. In “The Gas Liquid Chromatography of Steroids” (J. K. Grant, ed.), pp. 229-240. Cambridge Univ. Press, London and New York, 1967. K9. Kliman, B., and Peterson, R. E., Double isotope derivative assay of aldosterone in biological extracts. J . Biol. Chem. 236, 1639-1648 (1960). K10. Kofler, M., Sommer, P. F., Bollinger, H. R., Schmidle, B., and Vecchi, M., Physicochemical properties and assay of the Tocopherols. Vitamins Homumes 30, 407439 (1962). K11. Korytnyk, W., Fricke, G., and Paul, B., Gm chromatography of compounds in the vitamin Bs group. A n d . Biochem. 17, 66-75 (1966). K12. Korzun, B. P., Brody, S. M., Keegan, P. G., Luders, R. C., and Rehm, C. R. Rapid chromatographic method for the identification and estimation of glutethimide (Doriden) in blood. J . Lab. Clin. Med. 68, 333-339 (1966). K13. Kuksis, A., Newer developments in determination of bile acids and steroids by gas chromatography. Methods Biochem. Anal. 14, 325454 (1966). L1. Lamkin, W. M., and Gehrke, C. W., Quantitative gas chromatography of amino acids. Preparation of wbutyl N-trifluoroacetyl esters. A n d . Chem. 37, 383-389 (1965). L2. Landault, C., and Gurochen, G., Etude de l’utilisation du tkflon comme support en chromatographiegaz-liquide. Application ZI la separation de corps trbs polaires. J . Chromatog. 9, 133-146 (1962). L3. Leach, H., and Hammond, K. B., Gas chromatography of some reduction products of urinary steroids. Clin. Chim. A d a 21, 23-28 (1968). L4. Legate, C. E., and Burnham, H. D., Micropyrolytic-gas chromatographic technique for the analysis of organic phosphates and thiophosphates. Anal. Chem.32, 1042-1045 (1960). n glycoL5. Lehnhardt, W. F., and Winder, R. J., Determination of neutral sugars i proteins by gas-liquid chromatography. J . Chromatog. 34. 471-479 (1968). L6. Levy, R. L., Trends and advances in design of pyrolysis units for gas chromatography. J. Gus Chromdog. 6, 107-113 (1967). L7. Libby, D. A., Prosser, A. R., and Sheppard, A. J., Gas-liquid chromatographic method for the determination of fat-soluble vitamins. IV. Application to vitamin K1. J. Assoc. Ofic. Anal. Chemists 60, 806-809 (1967). L8. Liberti, A., and Cartoni, G. P., Analysis of essential oils by gas chromatography. In ‘‘Gas Chromatography” (D. H. Desty, ed.), pp. 321-342. Butterworth, London and Washington, D. C., 1958. L9. Littlewood, A. B., “Gas Chromatography.” Academic Press, New York, 1962. LlO. Lovelock, J. E., A sensitive detector for gas chromatography. J . Chromatog. 1, 35-46 (1958). L11. Lovelock, J. E., Ionization methods for the analysis of gssee and vapours. A n d . Chem. 33, 162-178 (1961). L12. Luke, H. H., Freeman, T. E., and IGer, L. B., Identification of the methyl esters of the stable Krebs cycle acids by gas liquid chromatography. Anal. Chem. S6, 1916-1918 (1963).

304

HAROLD V. STREET

L13. Luukkainen, T., Vanden Heuvel, W. J. A., Haahti, E. 0. A., and Homing, E. C., ;Gas chromatographic behaviour of trimethylsilyl ethers of steroids. Biochim. Biophys. A& 62, 599-601 (1961). L14. Lyons, H., and Bard, J., Gas chromatographic determination of lower alcohols in biologic samples. Clin. Chem. 10, 429-432 (1964). MI. Machata, G., Die Routineuntersuchung der Blutalkoholkonzentration mit den Gaschromatographen. Mihochim. Ada pp. 691-700 (1962). M2. Machata, G., tfber die gaschromatographkche Blutalkoholbestimmung. Analyse der Dampfphase. Milcrochim. Acta pp. 262-271 (1964). M3. Mahadevan, V., and Stenroos, L., Quantitative analysis of volatile fatty acids in aqueous solution by gas chromatography. Anal. Chem. 39, 1652-1654 (1967). M4. Makita, M., and Wells, W. W., Quantitative analysis of fecal bile acids by gasliquid chromatography. Anal. Bwchem. 6, 523-530 (1963). M5. Mdamed, N., and Renard, M., Analyse de m6lange d’acides aminb par chromatographie gazeuse. J . Chromatog. 4, 339-346 (1960). M6. Mather, A., and h i m o s , A., Evaluation of gas-liquid chromatography in assays for blood volatiles. Clin. Chem. 11, 1023-1035 (1965). M7. McCord, W. M., and Gadsden, R. M., The identification and determination of alcohols in blood by gas chromatography. J . Gas Chromatog. 2, 38-39 (1964). M8. McCredie, H. M., and Jose, A. D., Analysis of blood carbon monoxide and oxygen (1967). by gas chromatography. J. Appl. Physiol. 22, 86-66 M9. McKeown, G. G., and Read, S. I., Esterification and gas chromatography of some acids of the tricarboxylic acid cycle. Anal. Chem. 37, 1780-1781 (1965). M10. McMartin, C., and Street, H. V., Gwliquid chromatography of submicrogram amounts of drugs. I. Preparation, scope, and limitation of columns. J . Chromatog. 22, 274-285 (1966). M11. McMartin, C., and Street, H. V., Gas-liquid chromatography of submicrogram amounts of drugs. 11. Analysis of barbiturates and related drugs in biological media. J . ChrMnatog. 23, 232-241 (1966). M12. Metcalfe, L. D., and Schmitz, A. A., The rapid preparation of fatty acid esters for gas chromatographic analysis. Anal. Chem. 33, 363-364 (1961). M13. Metcalfe, L. D., Schmitz, A. A., and Pelka, J. R., Rapid preparation of fatty acid esters from lipids for gas chromatographic analysis. Anal. Chem. 38, 514-515 (1966). M14. Moye, H. A., An improved microwave emission gas chromatography detector for pesticide residue analysis. Anal. Chem. 39, 1441-1445 (1967). Nl. Nair, P. P., Bucana, C., de Leon, S., and Turner, D. A., Gas chromatographic studies of vitamins DZand D,. Anal. Chem. 37, 631-636 (1965). N2. Natelson, S., and Stellate, R. L., Apparatus for extraction of gases for injection into the gas chromatograph. Application to oxygen and nitrogen in jet fuel and blood. Anal. Chem. 36, 847-851 (1963). N3. Nelson, D. F., and Kirk, P. L., Identification of substituted barbituric acids by gas chromatography of their pyrolysis products. Anal. Chem. 34, 899-903 (1962). N4. Nikelly, J. G., Gas chromatography of free fatty acids. Anal. Chem. 86, 22442248 (1964). N5. Nikelly, J. G., Gas chromatographic determination of acetylsalicylic acid. Anal. Chem. 36, 2248-2250 (1964). 01. Oette, K., and Ahrens, E. H., Quantitative gas-liquid chromatography of shortchain fatty acids as 2-chloroethanol esters. Anal. Chem. 33, 1847-1851 (1961). 02. Ohline, R. W., and Jojola, R., Etched glass beads as gas chromatographic support. A n d . Chem. 36, 1681-1682 (1964).

GAS-LIQUID

CHROMATOGRAPHY IN CLINICAL CHEMISTRY

305

03. Ottenstein, D. M., Column support materials for use in gas chromatography. J . G ~ ChTGmdOg. s 1, 11-23 (1963). 04. Ottenstein, D. M., Comparison of support deactivation in gas chromatography, J . Gas Chromatog. 6, 129-134 (1968). P1. Palframan, J. F., and Walker, E. A., Techniques in gas chromatography. Part I. Choice of solid supports-a review. Analyst 92, 71-82 (1967). P2. Parker, K. D., and Kirk, P. L., Separation and identification of barbiturates by gas chromatography. Anal. Chem. 33, 1378-1381 (1961). P3. Parker, K. D., Fontan, C. R., and Kirk, P. L., Rapid gas chromatographic method for screening of toxicological extracts for alkaloids, barbiturates, sympathomimetic amines, and tranquilisers. And. Chem. 36, 356-359 (1963). P4. Parker, K. D., Fontan, C. R., Yee, J. L., and Kirk, P. L., Gas chromatographic determination of ethyl alcohol in blood for medicolegal purposes. Separation of other volatiles from blood or aqueous solution. Anal. Chem. 34, 1234-1236 (1962). P5. Patti, A. A., Frawley, T. F., and Stein, A. A., Analysis of 17-hydroxy-corticoids in urine and blood by gas-liquid chromatography. Am. J. Clin. Pathol. 46, 142-147 (1966). P6. Pendry, A., Automated GLC/IR analysis. Spectrowision 19, 5-8 (1968). P7. Perry, S. G., Peak identification in gas chromatography. Chromatog. Rev. 9, 12-13 (1967). P8. Pillsbury, H. C , Sheppard, A. J., and Libby, D. A., Gas-liquid chromatographic method for the determination of fat-soluble vitamins. V. Application to pharmaceuticals containing vitamin E. J . Assoc. Ofic. Anal. Chemists 60, 809-814 (1967). P9. Porcaro, P. J., and Johnston, V. D., Primary alcohols determined by gas chromatography. Anal. Chem. 33, 361-362 (1961). P10. Price, J. G. W., Fenimore, D. C., Simmonds, P. G., and Zlatkis, A., Design and operation of a photoionization detector for gas chromatography. Anal. Chem. 40, 541-547 (1968). R1. Rohrschneider, L., The polarity of stationary liquid phases in gas chromatography. I n “Advances in Chromatography” (J. C. Giddings and R. A. Keller, eds.), pp. 333-363. Arnold, London, 1967. R2. Rongone, E. L., and Segdoff, A., In vivo metabolism of 17~-methyl-A‘-testosterone. Steroids 1, 179-184 (1963). R3. Ruchelman, M. W., Solubility studies of estrone in organic solvents using gasliquid Chromatography. Anal. Biochem. 19, 9t3-108 (1967). R4. Ruchelman, M. W., and Cole, V. W., Analysis of urinary 17-ketosteroids by gasliquid chromatography. Clin. Chem. 12, 771-788 (1966). R5. Ruhlmann, K., and Giesecke, W., Gas chromatographie silylierter aminosauren. Angm. Chem. 73, 113 (1961). S1. Savory, J., and Kaplan, A., A gas chromatographic method for the determination of lactic acid in blood. Clin. Chem. 12, 559-569 (1966). S2. Sawardeker, J. S., and Sloneker, J. H., Quantitative determination of monosaccharides by gas-liquid chromatography. Anal. Chem. 37, 945-947 (1965). 53. Sawyer, D. T., and Barr, J. K., Evaluation of support materials for use in gas chromatography. Anal. Chem. 34, 1518-1520 (1962). S4. Shahrokhi, F., and Gehrke, C. W., Quantitative gas-liquid chromatography of sulphur-containing amino acids. J. Chromatog. 36, 31-41 (1968). 55. Simmonds, P. G., Pettitt, B. C., and Zlatkis, A., Esterification, identification, and gas chromatographic analysis of Krebs cycle keto acids. Anal. Chem. 39, 163-167 (1967).

306

HAROLD V. STREET

S6. Simmonds, P. G., Fenimore, D. C., Pettitt, B. C., Lovelock, J. E., and Zlatkis, A., Design of a nickel-63 electron absorption detector and analytical significance of high temperature operation. Anal. Chem. 39, 1428-1433 (1967). S7. Smith, B., and Carlsson, O., Gas chromatographic analysis of polyhydric organic compounds. Acta Chem. Scund. 17,455-460 (1963). S8. Somogyi, M., A method for the preparation of blood filtrates for the determination of sugar. J . B i d . Chem. 86, 655-663 (1930). Sg. Stefanovic, M., and Walker, B. L., Effect of stationary phase-support ratio on the gas chromatographic separation of trifluoroacetylamino acid butyl esters. Anal. Chem. 39, 710-713 (1967). S10. Sternberg, J. C., and Litle, R. L., Electrical discharge pyrolyzer for gas chromatography. Anal. Chem. 38, 321-330 (1966). S11. Street, H. V., Gas-liquid chromatography of submicrogram amounts of drugs. 111. Analysis of alkaloids in biological media. J . Chramutag. 29, 68-79 (1967). S12. Street, H. V., Forensic problems in the gas chromatography of amines and alkaloids. J . Chromatog. 37, 162-171 (1968). S13. Street, H. V., and McMartin, C., Quantitative estimation and identification of barbiturates in blood in emergency cases. Nature 199, 456-459 (1963). S14. Sturner, W. Q., and Coumbis, R. J., The quantitation of ethyl alcohol in vitreous humor and blood by gas chromatography. Am. J . Clin. Pathol. 46,349-351 (1966). 515. Subbaram, M. R., Separation of saturated and unsaturated fatty acid esters of cholesterol by gas-liquid chromatography. J . ChrMnatog. 16, 79-80 (1964). S16. Sweeley, C. C., and Walker, B., Determination of carbohydrates in glycolipides and gangliosides by gas chromatography. Anal. Chem. 36, 1461-1466 (1964). S17. Sweeley, C. C., Bentley, R., Makita, M., and Wells, W. W., Gas-liquid chromatography of trimethylsilyl derivatives of sugars and related substances. J . Am. Chem. SOC.86, 2497-2507 (1963). TI. Thomas, B. S., and Walton, D. R. M., The use of haloalkylsilyl ether steroid derivatives for’ analysis by .gas liquid .chromatography. I n “The Gas Liquid Chromatography of Steroids” (J. K. Grant, ed.), pp. 199-209. Cambridge Univ. Press, London and New York, 1967. v1. Vanden Heuvel, W. J. A., The gas-liquid chromatography of dimethylsiiyl trimethylsilyl and chloromethyldimethylsilyl ethers of steroids. Mechanism of silyl ether formation and effect of trimethylsilylation upon detector response. J . Chromatog. 27, 85-95 (1967). v2. Vanden Heuvel, W. J. A., Gas-liquid chromatography of the trimethyhilyl derivatives of several amines of biological interest. J . Chromatog. 36, 354-358 (1968). v3. Vanden Heuvel, W. J. A., Gardiner, W. L., and Horning, E. C., Characterization and separation of amines by gas chromatography. A n d . Chem. 36, 1550-1560 (1964). v4. Vanden Heuvel, W. J. A., Haahti, E. 0. A., and Horning, E. C., Gas chromatographic separations of drugs and drug metabolites. Clin. Chem. 8, 351-359 (1962). v5. Vanden Heuvel, W. J. A., and Horning, E. C., Conditions for the separation of steroids by gas liquid chromatography. I n “The Gas Liquid Chromatography of Steroids” (J. K. Grant, ed.), pp. 39-68. Cambridge Univ. Press, London and New York, 1967. V5a. van D u p e , R. P., and Aikens, D. A., A chemically selective polarographic detector for gas chromatography. Anal. Chem. 40, 254256 (1968).

GAS-LIQUID CHROMATOGRAPHY IN CLINICAL CHEMISTRY

307

V6. Van Kampen, E. J., and Hoek, W., Urinary steroid spectrum by gas chromatography. Clin. Chim. Acta 16, 442444 (1967). V7. Van Wijngaarden, D., Modified rapid preparation of fatty acid esters from lipids for gas chromatographic analysis. Anal. Chem. 39, 848-849 (1967). V8. Vassallo, D. A., Pyrolysis techniques. Anal. Chem. 33, 1823-1825 (1961). V9. Vecchi, M., and Kaiser, K., The gas chromatographic determination of ascorbic acid in the form of its trimethylsilyl ether derivative. J . Chromatog. 26, 22-29 (1967). V10. Vidal-Madjar, C., and Guiochon, G., No. 185. Chromatographie gaz-solide. Etude d’un support absorbant contenant une forte densit6 superfkielle de charges n6gatives: La benzoph6none solide. Bull. SOC.Chim. France No. 3, 1096-1100 (1966). V11. Vidal-Madjar, C., and Guiochon, G., Gas-solid chromatography on organic crystals. Nature 216, 1372-1373 (1967). W1. Wagner, J., and Rausch, G. Z., Gas-chromatographie trennung und bestimmung von aminosaiiren. 2. Anal. Chem. 194, 350-356 (1963). W2. Wallace, J. E., and Dahl, E. V., Rapid vapor phase method for determining ethanol in blood and urine by gas chromatography. Am. J . Clin. Pathol. 46, 152-154 (1966). W3. Webb, J. L., Smith, V. E., and Wells, H. W., Use of sponge spicules as a column support in gas chromatography. J . Gas Chromatog. 3, 384-387 (1965). W4. Weinstein, B., Separation and determination of amino acids and peptides by gas-liquid chromatography. Methods Biochem. Anal. 14, 203-323 (1966). W5. Wells, W. W., Chin, T., and Weber, B., Quantitative analysis of serum and urine sugars by gas chromatography. Clin. Chim. Acta 10, 352-359 (1964). W6. Wilk, S., Gitlow, S. E., Mendlowitz, M., and Clarke, D. D., Quantitative determination of 3-methoxy, 4-hydroxyphenyl ethylene glycol (NMG) by gas-liquid chromatography (G.L.C.). Federation Proc. 26, 736 (1966). W7. Wilk, S., Gitlow, S. E., Franklin, M. J., Carr, H. E., Mendlowitz, M., and Clarke, D. D., A quantitative assay for vanillylmandelic acid (VMA) by gas-liquid chromatography. Anal. Biochem. 13, 544-551 (1965). W8. Williams, C. M., Gas chromatography of the dihydroxybenzoic acids. Anal. Biochem. 11, 224-229 (1965). W9. Winsten, S., A simplified method for the analysis of glutethimide (Doriden) by gas liquid chromatography. Clin. Chem. 12, 536 (1966). WlO. Wolf, T., and Rosie, D. M., The pyrolysis-gas chromatography of simple organic molecules. Anal. Chem. 39, 725-729 (1967). W11. Wotiz, H. H., and Chattoraj, S. C., Determination of estrogens in low and high titer urines using thin layer and gas liquid chromatography. Anal. Chem. 36, 1466-1472 (1964). Y1. Yee, H. Y., Calibration for blood gas determination by gas chromatography. Anal. Chem. 37, 924-925 (1965). 21. Zado, F. M., and Juvet, R. S., Jr., A new selective-nonselective flame photometric detector for gas chromatography. Anal. Chem. 38, 569-573 (1966). 22. Zlatkis, A., Or6, J. F., and Kimball, A. P., Direct amino acid analysis by gas chromatography. Anal. Chem. 32, 16%164 (1960). 23. Zomzely, C., Marco, G., and Emery, E., Gas chromatography of the n-butyl-Ntrifluoroacetyl derivatives of amino acids. Anal. Chem. 34, 1414-1417 (1962).

This Page Intentionally Left Blank

THE CLINICAL CHEMISTRY OF BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

. .

Paula Jablonski and. J A Owen Department

of Biochemistry. Alfred Hospital. and Department of Surgery.

Monash University. Alfred Hospital. Melbourne. Australia 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Historical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.

.

5

6.

.

7

310 311 ..................................... 314 ..................................... 314 ..................................... 316 317 lic Dyes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transfer of BSP from Plasma to Bile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 318 4.1. Cells Involved in Hepatic Uptake of BSP . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Binding of BSP to Tissue Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 . . . . . . . . . . 321 4.3. Metabolism of BSP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Metabolism of Other Dyes . . . . . . . . . . . . . . . . . . . . . . 4.5. Effect of Other Dyes on BSP Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . 324 4.6. Relation of Plasma, Bile, and Lymph to Dye Transfer . . . . . . 4.7. Intracellular Relationships. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.8. Extrahepatic Uptake of Dye . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dye Teats Used in Clinical Investigations . . . . . . . . . . . . . . . . . . . . . . . . 5.1. The Standard BSP Retention Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 5.2. Parameters Defining BSP Disappearance from Plasma . . . . . . . . . . . . . 330 334 5.3. Parameters of BSP Uptake by the Liver . . . . . . . . . . . . . . . . . . . . . . . . . . ............................ 336 5.4. Measurement of Biliary Exc 5.5. Urinary Excretion Teats ..... ............................ 336 ............................. 337 5.6. Tests Using Other Cholephili 338 5.7. Estimation of Hepatic Blood Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 5.8. Measurement of Body Fluid Spaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9. Use of Radioactive Cholephilic Dyes . . . . . . . . . . . . . . . . . . . . . . . 341 .................................... 341 5.10. Toxic Effects of Dye ... Effects of Physiological Factors on Dye Uptake and Secretion... . . 342 342 6.1. Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343 6.2. Sex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343 6.3. Pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 6.4. Feeding, Exercise, and Posture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 6.5. Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 Effects of Drugs on Dye Uptake and Excretion. . . . . . . . . . . . . . . . . . . . . . . . . . 7.1. Cholecystographic Agents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 7.2. Probenecid ... .._. ............................................ 345 346 7.3. Flavaspidic Acid and Related Compounds. . . . . . . . . . . . . . . . . . . . . . . . 346 7.4. Icterogenin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 7.5. Antibiotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

310

PAULA JABLONSKI AND J. A. OWEN

7.6. Phenothiaaines. . . . . . . ..................................... 7.7. .......................................... 7.8. .......................... 7.9. Benziodarone. . . . . . . . 7.10. Ethanol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.11. Anesthetics and Hypnotics.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.12. Drugs that Affect the .......................... 7.13. Anti-Inflammatory Drugs.. . . . . ........... 7.14. Phenindione . . . . . . . . . . . . . . . . . . . . . . . . 7.15. Miscellaneous Enzyme Inhibitors. .................... 7.16. Compounds that Caus ............................. 7.17. Steroids. . . . . . . . . . . . . . . . 7.18. Bile Salts.. . . . . . . . . . . 8. Effect of Disease on Dye Uptake and Excretion.. . . . . . . . . . . . . . . . . . . . . . . . 8.1. Hepatobiliary Diseases.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2. Dye Tests in Jaundice.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3. Acute Abdominal Conditions. ......................... 8.4. Septicemia and Other Infections. .......................... 8.5. Circulatory Disorders. . . . . . . . . . . . . . . . . . . . .............

347 348 348 348 349 349 349 350 352 352 355 356

8.8. Renal Disease.. . . . . . . . . . . . . . . . . . . . . . 9. Analytical Methods.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1. BSP.. . . . . . . .................................... 9.4. 9.5.

RoseBengal.. . . . . . . . . . . . . Phenol Dibromophthalein D

References. . . . . . . . . . . . .

...........

1.

........................... ............

..........................

360 361 364

Introduction

The term cholephil has been proposed (H4) for endogenous and exogenous substances which are removed from the circulation by the liver and concentrated in the bile. Among the cholephils are bromsulfophthalein (BSP) and other dyes which have been used in the study of liver function in patients and experimental animals for over half a century. The uptake of cholephilic dyes, indeed, is widely held to be the most sensitive index of hepatobiliary function. The purpose of this review is to provide an account of the clinical chemistry of the cholephilic dyes. Many different dyes are concentrated in bile but in clinical and laboratory studies the use of BSP has far exceeded that of other dyes. For this reason, much of this review is devoted to metabolism of BSP and to the various tests which involve this dye.

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

2.

311

Historical

The history of cholephilic dyes begins in 1866, ten years after Perkins had produced the first synthetic organic dye, aniline purple (P4). Chrzonszczewsky (C20), investigating the ability of a number of vital stains to display the anatomical arrangement of bile canaliculi, found that aniline red (fuchsin) and indigo carmine were taken up by both liver and kidney tissue whereas carmine, Berlin blue, and aniline blue were taken up by kidney tissue only. Aniline red, however, was very toxic. Indigo carmine was not visible in the liver or kidney cells, but bile ducts, gallbladder, bile, and urine were brightly colored. I n 1909 Abel and Rowntree (A2), studying the action of phenol tetrachlorophthalein as a purgative, found that it was absorbed from the colon and excreted in bile, and from these studies the suggestion arose that dyes might be used to evaluate liver function (R25). Phenol tetrachlorophthalein was administered intravenously (R25), and the amount excreted in feces was determined. Later, to avoid difficulties inherent in the analysis of feces, dye excretion was determined in duodenal aspirate (All M9). However, the use of the dye to assess liver function was not widely accepted until Rosenthal (R17) suggested measuring the rate of disappearance of the dye from plasma. Rosenthal (R18), using partial hepatectomy, showed that this measurement of disappearance rate was related t o liver mass. The demonstration (B29, R18) that the rate of disappearance of phenol tetrachlorophthalein from the circulation formed a convenient index of liver function led to the suitability of many other dyes being tested (A4). The main criteria of suitability were outlined by Delprat (D4), who suggested that the dye should be (a) nontoxic, (b) excreted mainly or exclusively by the liver, (c) crystalloid, presumably so that it would not be taken up by the reticuloendothelial cells, and (d) retained in the circulation long enough after injection to allow blood samples to be taken at convenient intervals. Delprat (D4, D5) found that rose bengal, an iodinated fluorescein derivative, fitted these criteria. Using the same criteria, Rosenthal and White (R22) tested many halogenated phthaleins and demonstrated that phenol tetrabromophthalein disulfonate (BSP) was most suitable. Tests using either BSP (R23) or rose bengal (K6) were carried out, and patients with liver disease were found to retain the dye in the circulation longer than control patients. Clinical tests were also carried out with other dyes. The elimination of indigo carmine and azofuchsin S was markedly reduced in liver disease (A4, R16) but the converse was found with methylene blue (R21). Azorubin S was found to be removed from the plasma by the liver, but in

312

PAULA JABLONSKI A N D J . A. OWEN

patients with liver disease some was excreted in the urine (T2). Congo red gave varied results in liver disease but its uptake from plasma appeared to be dependent also on the state of the reticuloendothelial system (W19). I n general, tests with these dyes were found to be less suitable than tests using BSP or rose bengal because of (a) toxicity (R15), (b) apparent insensitivity to liver dysfunction, or (c) extrahepatic uptake. Only indocyanine green (Cl6, H27), a tricarbocyanine dye used initially for blood flow studies (F5), has shown any promise as an alternative to BSP or rose bengal. For many years, the results of dye tests were expressed simply as the percentage of dye remaining in the circulation (percent retention) a t a stated time after the injection or as the percentage removed between two stated times. About 1950, however, attempts to obtain more precise information on liver function led to the use of parameters, such as the fractional disappearance rate, the percentage disappearance rate, or the plasma dye half-time, for expression of dye-test results (see Table 1). Later still, parameters expressing discrete liver functions, such as maximal uptake, liver storage capacity or biliary transport maximum were introduced (see Section 5.3). More recently, serial collection of blood samples has been made for complex mathematical analyses of plasma dye decay curves (see Section 5.2) to provide further information on the different processes involved in dye excretion. Availability of isotopically T10) has led to methods of measuring dye uptake by labeled dyes (T3, external counting techniques and to radioscanning procedures (see Section 5.9) for providing information on the size and shape of the liver and on the location of discrete pathological lesions. Another development has been the study of the conjugates formed when BSP is taken up by the parenchymal cells of the liver (B46, C5, C6, G23). A portion of these regurgitate back into the plasma, and measurement of the absolute or relative amounts of conjugates offers an alternative approach to the clinical study of hepatobiliary function (see Section 9.2). It is of historical interest that cholecystographic agents were developed from the findings (A2) that led to the development of the BSP test. Knowing that many substances in bile are concentrated in the gallbladder, Graham and Cole (G18) tested tetrachloro-, tetrabromo-, and tetraiodophenolphthalein as cholecystographic agents and obtained radiological visualization of the gallbladder. At the time the bromo compound was found to be the most satisfactory, but later investigators used tetraiodophenolphthalein administered by mouth (G19, P2) or by intravenous injection (G19). Phenol tetraiodophthalein was also used (G19) but

t3

s P

TABLE 1 PARAMETERS DESCRIBING UPTAEE OF DYESFROM

m

PLASMA"

d

u

Preferred name Fractional disappearance rate (FDR)

Synonyms

Preferred symbol

Fractional clearance (G14, L l l ) , clearance coefficient (K4)

a

Formula In pt - In p2 t z - tl

Comment Ingelfinger et al. (11) have used the approximation 1 - d, where d = e-a (see Section 5.2.2). Lewis (L10) has used the approximation ( p i - p 2 ) / ( p t p 2 ) 4 (see Section 5.2.2)

Logarithmic disappearance

Clearance coefficient (L2),

rate Percentage disappearance rate (PDR) Half-life

elimination constant (C3, N2) Fractional clearance (Fl)

Plasma clearance these are based on the equation p , =

01'

-

logp1 - logp:

'd

! Fr E

Z

9

S

8

3 a

100a or 230d

-

0

-

tz -

tl

t:

0.693/a or 0.301/a'

-

CBSP

01.

=

0

-

-

or p t

-l

v,

p ~ l O - ~ a' ~= ; 2 . 3 ~ ~(see ' Section 5.2.2).

V , = plasma volume

r M v

E

i tl

.e M

u1

314

PAULA JABLONSKI AND J. A. OWEN

both iodophthaleins have since been replaced by other contrast agents (S43) . 3.

Nature of Cholephilic Dyes

3.1. INVESTIGATION OF DYES The term cholephilic is used in this review to describe dyes which are excreted mainly in the bile. Many different dyes have been examined in this respect (see reviews A4, C12). However, in spite of these studies, relatively little information is available on the factors which cause a dye to be excreted by the liver rather than by the kidney or by some other organ. Such information is potentially valuable because of the light it sheds on the mechanisms operating in the hepatobiliary system. None of the three dyes used in clinical studies-BSP, indocyanine green, or rose bengal-is ideal, and search for better alternatives also requires this information. 3.1.1. Phthlein Dyes This is the group of dyes which has been most thoroughly investigated. Phenolphthalein itself is mainly excreted in the urine as a glucuronide (533) , but halogenated derivatives are rapidly removed from the plasma by the liver and excreted in the bile. Various phenol tetrahalogen phthaleins (e.g., BSP) , tetrahalogen phenolphthaleins, and halogenated fluoresceins (e.g., rose bengal) were tested (R22), and all were found to be excreted mainly in the bile; only traces appeared in the urine. The chemical structures of some such dyes are given in Fig. 1. Other halogenated phthaleins, such as thymol blue and bromocresol purple, are likewise excreted mainly in the bile, with some excretion in the urine (C12). The presence of halogen atoms in these phthaleins is important because nonhalogenated phthaleins, such as phenol red, are excreted mainly in the urine (A4, S30a) though a little appears in the bile (R23, M6). Dyes of the fluorescein series with less than six halogen atoms per molecule are excreted approximately equally by the liver and kidney whereas those with six or more halogen atoms are excreted almost exclusively in the bile (K6). 3.1.2. Other Dyes Dyes have many diverse structures since chromophore groups may be attached t o all aromatic molecules. A comprehensive review of hepatic dye uptake and biliary dye excretion has been published (C12). ilsimple classification, based on the anionic or cationic nature of the dye and its chromophore group, is applied, but such a classification may not be

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

w1

c1

315

coo-

@I-

Br*Br

Br

0

I

I

so;

OH

0

OY

so; (C)

0D

C

a

O

H

FIG.1. The structure of some phthalein dyes. (A) Phenol tetrachlorophthalein; (B) tetraiodophenolphthalein; (C)phenol tetrabromophthalein disulfonate (BSP); (D) phenol 3,6-dibromophthalein disulfonate ; (E) phenol sulfonephthalein (phenol red) ; (F) tetraiodotetrachlorofluorescein (rose bengal). valid in a biological system. Although a great many dyes have been studied in the whole animal and in the perfused frog liver (A4, C12), many types of dyes have not been tested. Nitro dyes (e.g., Martius yellow, naphthol yellow) are excreted in bile and urine whereas the nitroso dye, Alsatian green, is found in the urine only. Monoazo dyes are rapidly removed from the circulation but are eliminated in the urine rather than in bile (A4). Azorubine S and Ponceau 3R,however, are excreted mainly in the bile. Diazo dyes tend to accumulate in tissues, but there is a slow excretion of these dyes in the bile (Tl)and in the urine (W20). However, Palatine black 4B and benzo-

316

PAULA JABLONSHI AND J . A. OWEN

I

FIQ.2. Structure of indocyanine green.

purpurine lOB, two diazo dyes, are rapidly excreted in the bile (T2). Triazo dyes do not appear in either the bile or urine. Both basic and acidic triphenylmethane dyes are excreted in bile. A number of basic dyes, e.g., trypaflavine (H16) and aniline blue (T7), are not excreted in bile, and dyes with a great many sulfonic acid groups (e.g., Ponceau 6R, azofuchsine V, and azofuchsine N), which are highly hydrophilic, are apparently not excreted by the liver or by the kidney (H18). Colloidal dyes, with the exception of Congo red, and strongly lipidsoluble dyes do not appear in bile (H17). Indocyanine green (see Fig. 2), a tricarbocyanine dye, is rapidly and almost exclusively excreted by the liver and has been used in clinical studies (see Section 5.6.1). The uptake and concentration of 200 different dyes was tested using perfused frog livers (H18, H19) in an attempt to correlate chemical structure with the ability of the liver to take up, excrete, and concentrate the dye. Although acidic dyes were concentrated to a higher level than basic dyes, no correlation with other chemical properties could be found. The speed of dye passage through the liver also could not be related to the chemical nature of the dye or to its diffusibility in gelatin or agar. 3.2. PROTEIN BINDING Early workers showed that BSP and rose bengal were protein bound (R18a, R19). Subsequently, salting out, electrophoresis, and gel filtration techniques have indicated that this association is primarily with the “albumin” fraction (B3, B17). More recently it has been reported that a considerable proportion (30-60%) of BSP is bound to al-lipoprotein (B2). This protein fraction has an even higher affinity for indocyanine green, 90% of which is bound. However these studies on the binding of BSP to protein are not definitive because nonphysiological and nonequilibrium conditions were used.

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

317

A great many other dyes have also been found to be bound to protein. In particular, the binding of azorubin (B17, W5-W7) naphthol yellow (B17), phenol red (B27, M25), rose bengal (E4), and indocyanine green (C16) has been investigated. Whole serum does not bind phenol red below a critical albumin concentration of 1.5 g%. Above this critical value, dye is bound in direct proportion to the amount of albumin in excess of the critical value. In jaundice the critical value is elevated. These results were taken to indicate that some binding sites in whole serum are occupied by, for example, bilirubin, bile acids, or fatty acids (B27). Bile salts reduce the binding of BSP ( A l l , A14, R18a). This effect may be due to the detergent properties of bile salts because Teepol has been found to have a similar effect (A14). Binding of rose bengal was not affected by sodium dehydrocholate or by BSP (E4). The rate a t which dyes disappear from the blood may depend on the dissociation constants of dye-plasma protein and dye-cellular protein complexes. The excretion route also may be influenced by the affinity of a dye for plasma proteins. In a study (G20) of four related diazo dyesEvans blue (ditoluidine) , trypan blue (ditoluidine) , Niagara sky blue (dianisidine) , and Niagara sky blue 6B (dianisidine) A ia n is id in e dyes disappeared a t a faster rate from the circulation than ditoluidine dyes. The position of the four sulfonic acid groups in these dyes is also important: shifting the groups from the 2,4 to the 3,6 positions increases the disappearance rate. These relative disappearance rates could be related to the protein binding of the dyes; Evans blue, which disappears a t a slow rate, is most firmly bound whereas Niagara sky blue, which disappears most quickly, is least firmly bound (R2).

3.3. CHARACTERISTICS OF CHOLEPHILIC DYE~ From the available data, it appears that cholephilic dyes must be water soluble and noncolloidal. Anionic dyes, regardless of their chemical structure, are excreted by the liver more readily than cationic dyes (H19). Many dyes are bound to plasma albumin, and it has been suggested that those which are more strongly bound are secreted by the liver whereas those which are more loosely bound are excreted by the kidney (B17). Thus, a t a concentration in plasma of 1 mg/100 ml, about 20% of phenol red is not protein bound whereas less than 1% of BSP is dialyzable. 4.

Transfer of BSP from Plasma to Bile

The transfer of BSP from plasma to bile involves three distinct phases: (a) hepatic uptake of the dye from plasma by a membrane transport

318

PAULA JABLONSEI AND J . A. O W N

system, (b) conjugation of BSP, (c) biliary excretion of conjugated and free BSP. 4.1. CELLSINVOLVED IN HEPATIC UPTAKE OF BSP I n a complex organ with a great variety of functions, it is often difficult to determine which cells are involved in any one function. Histological evidence has indicated that liver parenchymal cells are involved in the uptake of 35S-labeled BSP (K17), fluorescent rose bengal (M15), uranin (G17, H3), and indocyanine green (H27), whereas earlier studies indirectly implicated Kupffer cells in the uptake of BSP. These earlier results of inhibition of BSP uptake were obtained by “blockade” of the reticuloendothelial system with India ink, with saccharated iron oxide, or by splenectomy (C3, H12, K11, M26, S6, S25). It is probable that the observed inhibition by India ink was not related to reticuloendothelial blockade as this material has been found to contain a hepatotoxic agent (V9) and to suppress carbohydrate metabolism as well as BSP uptake (A9). It has also been suggested that the effect of such blockade is to cause swelling of Kupffer cells, which then obstruct sinusoids and so reduce blood flow and BSP uptake (M17). Splenectomy may also reduce total blood flow through the liver and so inhibit BSP uptake. It has been suggested that the epithelial cells lining the small bile ducts may play a n important, even exclusive role in the uptake and secretion of BSP (A10, A l l , A13). The small bile ducts are enmeshed by a vascular plexus which is largely supplied by the hepatic artery. In testing the hypothesis that an excretory function is served by this peribiliary plexus, experiments were devised to measure separately, but simultaneously, the rates of secretion into bile of BSP taken up from either the portal vein or the hepatic artery (B51). Simultaneous injections of 35S-labeled BSP into one blood vessel and nonradioactive BSP into the other were made, and biliary excretion rates and hepatic extraction efficiency were determined. The ratio of hepatic arterial to portal venous flow is somewhere between 1 :3 and 1 :5; thus distribution differences would be large enough to reveal any real differences in excretory activity. The results obtained indicated that BSP excretion was not influenced by the route of injection and that BSP extraction efficiency was insignificantly higher in the hepatic artery (53%) than in portal vein (47%). It was concluded (B51) that the peribiliary arterial plexus had no special role in BSP secretion into bile, and it was proposed that this plexus may have a function in the modification of bile composition by exchange of certain constituents among bile, blood, and lymph. Substances actively taken up from plasma by the cells pass through

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

319

an extracellular space. The nature of this space in the case of BSP uptake by hepatic parenchyma is not known. Andrews and del Rio Lozano (A15) have suggested that the sinusoidal endothelium can dissociate BSP from albumin and the dye diffuses into the perisinusoidal space, where it is either taken up by the parenchymal cells or diffuses into lymph. This conclusion was based on studies which involved simultaneous injection of BSP and I3lI-labeled albumin into anesthetized animals and measurement of BSP and radioactivity in plasma and lymph. Maximum dye concentration in lymph was attained after 15 minutes whereas maximum radioactivity in lymph occurred after 50 minutes. These results, however, do not eliminate the possibility that dissociation of the dye-albumin complex occurs rapidly a t the parenchymal-cell surface or within the cell.

PROTEINS 4.2. BINDINGOF BSP M TISSUE The movement of organic anions such as BSP from blood to bile involves transfer across the sinusoidal and bile canalicular regions of the membrane of hepatic parenchymal cells. The initial rate of removal of BSP from the blood is not related directly to the dose injected (G15) ; there is a systematic decrease in this rate with increasing doses of the dye. It was proposed (G15) that the transfer of the dye from blood into liver can be described in a manner similar to that of Michaelis-Menten (M22) for enzyme kinetics, i.e., dy%

+ C*dye

ka

lh

9

C + dye*

kr

+C

where C represents a carrier, dye, and dyel represent dye in plasma and in liver, respectively, and dye.C represents the carrier-bound dye. If p = concentration of dye in plasma, 1 = concentration of dye in liver, c = concentration of carrier, f = concentration of carrier-bound dye, k,, kz, k, = rate constants, Kd = dissociation constant for carrier complex

If

T

= removal rate of dye from blood to cell

r

=

k a .f

When all the dye is carrier bound, the maximal rate of removal (rmu) is attained

320

PAULA JABLONSKI AND J. A. OWEN

f = c

.. _1r -- -1

Tmav

Ka + -'P rma

This proposal is substantiated by graphs of the reciprocal of the initial removal rate against the reciprocal of the injected dose which are linear (G15). It seems probable that tissue proteins are involved in this process in the specific binding of the dye. The excretion of the dye from liver to bile has a maximal excretion rate which is also consistent with a carriermediated process. Liver cell membranes isolated and incubated in buffered solutions of BSP (C36), bound 160 2 pg BSP per milligram of membrane protein. BSP binding was maximal a t 500 pg of BSP per milliliter of incubation mixture and a t p H values below 7. Maximum binding was completed in 3 minutes and unaffected by the addition of K+, Ca2+,Mg2+,or C1-. Taurocholate, bilirubin, and uranin had no effect on binding, but flavaspidic acid (see Section 7.3), iodipamide methylglucamine (see Section 7.1), and indocyanine green were inhibitory. Liver slices when incubated in buffered BSP solutions readily take up the dye (B48). BSP uptake was unaffected by temperature in the range of 22'49°C or by p H in the range 5-8.5. Addition of crystalline bovine serum albumin or whole human serum to the incubation medium reduced the percentage of BSP extracted by the liver. Metabolic inhibitors had no effect on the uptake of BSP by the liver. Since slices from other tissues (e.g., heart, kidney, muscle) also extracted BSP under these conditions, it was suggested that the reaction was the result of a physicochemical rather than that of a physiological phenomenon resulting from the relative dissociation constants of the BSP-albumin and BSP-intracellular protein complexes (B48). The shape of the saturation curves obtained by repeated transfer of the slices through media with or without protein indicated that the dissociation constant for the BSP-intracellular protein complex was larger than that for the BSP-plasma albumin complex (B48), a conclusion also reached by others (A15, B7). It has been suggested (B3) that the albumin-bound dye may be directly involved in cellular uptake. Electron microscopy has shown that the sinusoidal epithelium may be permeable to plasma proteins (F2), and thus the BSP-plasma albumin may come into direct contact with the parenchymal cell membrane. Rapid transfer of the dye from albumin to the carrier protein is then fe:isihle. This eschaiigc may be facilitated or made unidirectiorial by the conjugation of the dye since it has becn

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

321

shown that the affinity of albumin for BSP is a t least 100 times greater than for its glutathione conjugate (B3). Similarly albumin has a greater affinity for bilirubin than for its glucuronide (LlOa). 4.3.

METABOLISM OF BSP

4.3.1. Nature of BSP Metabolites Chromatographic analysis (C5, D1, K18, M20, M29) of the dye excreted in rat bile after the administration of BSP revealed that a major proportion of this dye is different from the parent dye. Two or three metabolites were found (C5, K15, K18). Subsequently, on examination of the bile of different species, up to four different metabolites (W16) have been demonstrated. Treatment with P-glucuronidase had no effect on the metabolites; this indicated that they were not glucuronides of BSP. Thus, the metabolism of BSP is different from that of bilirubin (K18, M20). The composition of the major metabolite was investigated by Grodski and co-workers (C5, G23). Chromatography after hydrolysis with 6 M HCI or with 6 M HBr a t 120°C for 18 hours revealed glycine, glutamic acid, and, possibly, cysteine. Hydrogenolysis with Raney nickel yielded alanine, which could have originated by the desulfuration of cysteine. Staining of chromatograms with a stain specific for mercaptide bonds was positive. Synthetic complexes of BSP with cysteine or glutathione corresponded chromatographically with the metabolites (G24). It was therefore concluded that BSP was conjugated with glutathione and cysteine. This was subsequently confirmed by administration of 35Slabeled cystine and glutathione (C34) and of I4C-labeled glutamic acid, glutamine, or glycine (C28) to rats and the detection of the radioactive label in all BSP metabolites. It was shown that the conjugation was effected by the formation of a thioether linkage (C31), but the exact point of attachment of the thiol group to BSP is still debatable. There is good evidence that neither sulfonate group is removed and, since the absorption spectra of parent dye and the metabolites are similar (M30), i t is unlikely that the basic phenolphthalein structure is altered. It seems likely that the conjugation of BSP with glutathione involves the removal of one bromine atom from BSP. Several workers (C31, 55) have demonstrated the release of bromide during in vitro conjugation of BSP by liver homogenates; the bromide released was directly related to the number of moles of conjugate formed. However, others have been unable to find a change in the bromine content of the metabolites (K18). The other metabolites are also mercaptides (C32,55) and are probably conjugates of cysteine, cysteinylglycine, and cysteinylglutamate. The presence of a diglutathione conjugate has also been reported (55). The relative proportion of these metabolites and of the free dye in bile

322

PAULA JABLONSKI AND J . A. OWEN

appears to vary in different species (K18, W16) and under conditions of bile stasis (B46, H14, K19).

4.3.2. Conjugation of BSP Conjugation of BSP is probably catalyzed by S-aryl glutathione transferase, an enzyme which is present in high concentration in the liver (B35, B53). This enzyme is involved in the detoxification of monohalogenobenzenes and related compounds and in their excretion in urine as mercapturic acids. The excretion of mercapturic acids was first noted in 1879 (B14, J2), but the involvement of glutathione in the process was suggested only in 1959 (B13) after it was found that the glutathione level in rat livers decreased on administration of compounds that form mercapturic acids. Subsequently, glutathione conjugates of p-chlorobenzyl chloride (B53), 1,2-dihydro-2-hydroxy-l-naphthalene(B34) and 2-hydroxy-l,2,3,4-tetrahydro-l-naphthalene(B36) were isolated after incubation of liver slices with p-chlorobenzyl chloride, naphthalene, and 1,2dihydronaphthalene, respectively. Aromatic amines also form glutathione complexes and are excreted as mercapturic acids (B39). I n most cases activation of the molecule is necessary before conjugation is feasible; e.g., naphthalene is first converted to trans-1,2,3,4-tetrahydro-l,2-dihydroxy naphthalene before conjugation. Halogenobenzenes do not require activation, and conjugation usually results in the release of halide, eg., 3,4-dichloronitrobenzene glutathione + S- (2-chloro-4-nitrophenyl) gluHCI (B53). The enzyme is highly specific for glutathione; tathione cysteinylglycine can be substituted, in some cases, but conjugation proceeds much more slowly. Oxidized glutathione, cysteine, cystine, N acetylcysteine, taurine, methionine, glycine, glutamic acid, and alanine do not take part in conjugation (B57). 4.3.2.1. Subcellular Site of Conjugation. Subcellular fractionation has indicated that the conjugation of BSP occurs in the soluble fraction ((32) and that the S-aryl glutathione transferase is also localized in this fraction (B53). 4.3.2.2. Chemical Conjugation of BSP. Many halogenated compounds undergo nonenzymatic conjugation with glutathione. Booth et al. (B35) showed that BSP, benzyl chloride, bromoethane, bromopropane, ethyl could be methanesulfonate and 1,2-epoxy-1,2,3,4-tetrahydronaphthalene conjugated chemically with glutathione in the p H range 7-8. However, the reaction could be accelerated by the addition of liver homogenate or a soluble fraction of it. In this context, it is noteworthy that most of liver glutathione is localized in the soluble fraction (E2),and that this may influence studies of the distribution of conjugating activity. 4.3.2.3. Formation of Other Metabolites. By injection of individual

+

+

BBOMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

323

metabolites, it was possible to establish the following chain of consecutive transformation (B46, K19) : BSP -+ glutathionyl-BSP -+ cysteinyl-BSP

Thus, i t is likely that the other metabolites are products of hydrolysis of the major conjugate. This hydrolysis may be catalyzed by glutathionase; this step is known to occur in liver (58, N5a). A transfer reaction is also possible, e.g. (B52) : Glutathionyl-BSP

+ cysteine -+

cysteinyl-BSP

+ glutathione

Biliary stasis changes the excretory pattern of the metabolites (B46, H14, K19, W16). I n particular, the excretion of cysteinyl-BSP and that of the other metabolites is increased a t the expense of the glutathione conjugate. This effect is related to the duration of stasis, and it has been suggested that it is due to hydrolysis of the glutathionyl compound in the biliary tree (K19). The longer the transit time through the biliary tree, the higher the proportion of hydrolyzed compounds in the bile. 4.3.2.4. Factors Affecting Conjugation of BSP 4.3.2.4.1. Level of glutathione in the liver. Reduction of the level of glutathione by feeding rats a protein-free diet for 2 days markedly reduced conjugation and excretion of BSP (C29). The level of liver glutathione is low in neonatal animals, and this accounts in part for the reduced conjugation of the dye by the liver in these animals (C29). However, the conjugation enzyme is also low in neonatal animals and matures slowly; in rats it takes 5-7 weeks to attain the adult level (C29, C33, G11). 4.3.2.4.2. Level of ATP. When the nucleotide was depleted by administration of ethionine (or ethionine plus dinitrophenol) to 25% of the ATP level in the control, no effect was observed on BSP conjugation. This suggests either that 25% of ATP is enough or else that ATP is not involved (S7). The latter seems more likely since no requirement for ATP has been shown in experiments in vitro. 4.4. METABOLISM OF OTHERD m The metabolism of other dyes has not been studied comprehensively. 4.4.1. Dyes Related to BSP

Phenol tetrabromophthalein is conjugated with glutathione to a small extent with a concomitant release of bromide (C32). Another conjugate is formed which, on hydrolysis with P-glucuronidase, yields the glutathione conjugate (54). Phenol tetrabromophthalein monosulfonate forms similar conjugates which suggests that the glucuronide is formed with

324

PAULA JABLONSKI AND J. A. OWEN

the phenolic hydroxyl group and that BSP does not form glucuronides because of sulfonate groups in positions ortho to this hydroxyl group. Phenol tetrabromophthalein tetrasulfonate is removed from plasma much more slowly than BSP, appearance in bile is delayed, and no conjugation is detectable (54). This compound is not conjugated by the enzyme in vitro (54). Phenol dibromodihydroxyphthalein sulfonate is excreted as both glucuronide and glutathionyl complex. Phenol dibromophthalein disulfonate, however, is not conjugated (H20,53). A diiodo-substituted BSP is also metabolized; two or three metabolites can be detected as well as free BSP (C28, 57, M7), indicating that deiodination occurs as well as conjugation. There is some confusion in the literature over the exact nature of rose bengal used in liver function studies. The name rose bengal is generally accepted to mean the sodium or potassium salt of tetraiodotetrachlorofluorescein (M11, M19). However, other closely related compounds have been given the same name ( M l l j : Kerr and co-workers (K6), who initially described the rose bengal test stated that they used diiodotetrachlorofluorescein whereas others (S2, 530) use the term to refer to tetraiodotetrabromofluorescein. McNalty (M8) defines rose bengal as tetraiododichlorofluorescein. Evidence has been obtained that some of the preparations used in liver function studies were mixtures (B22, L14). No conjugation of rose bengal has been detected although some deiodination occura (B15, Kn, R1) . It is probable that deiodination of aromatic iodocompounds is due to the high activity of the liver deiodinase, an enzyme responsible for the deiodination of thyroxine and triiodothyronine (B14a). We have found that bromophenol blue is chemically unaltered by passage through the liver into bile whereas phenolphthalein is excreted as glucuronide (C32) and phenol sulfonephthalein is not conjugated (D6). 4.4.2. Indocyanine Green Indocyanine green is excreted in bile unconjugated (L5j . 4.5.

EFFECT OF OTHERDYESON BSP TRANSFER

Simultaneous injection of phenol dibromophthalein disulfonate and BSP resulted in a reduced excretion of both compounds, but the total excretion was approximately the same as after a dose of the dibromo derivative (H20). BSP is taken up preferentially by the liver (C23, F1, P9j when administered together with rose bengal, but its excretion is slower (B49).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

325

Others have noted that BSP has a far greater inhibitory effect on rose bengal removal (C24, Ml6 ) than the converse effect of rose bengal on BSP removal. However, on a mole-to-mole basis, the BSP dose in all these experiments was two to four times higher than that of rose bengal. Indocyanine green also increased the retention of BSP (H28, W9). There is evidence of molecular competitive inhibition between indocyanine green, BSP, rose bengal, and bilirubin (H28). 4.6. RELATION OF PLASMA, BILE,AND LYMPHTO DYE TRANS^

There is a large production of hepatic lymph which appears to arise by passage of fluid from the perisinusoidal space, from bile duct area and in the portal tract area (A12). It has been suggested that blood, lymph, and bile enter into an equilibrium in the portal tract (Review article, 2). The lymphatics could also act as a transport system between the liver lobule and the biliary epithelium ( A l l ) . A detailed study of the relative concentrations of BSP and its metabolites in lymph, bile, and plasma indicated that the composition of lymph differs from that of bile and plasma (K20). Thus, it seems unlikely that lymphatics were the transport system for BSP from liver cells to bile. During biliary stasis there is a rapid rise in lymphatic BSP levels (B32, G12) and the relative concentrations of BSP metabolites approach those found in bile (K20). On ligation of the bile duct the level of bilirubin and bile acids in lymph also rises (G12) as does the plasma level of BSP (B10) and rose bengal (B22). Micropuncture samples of portal venule blood a t the very end of the portal canals during continuous infusion of BSP had higher concentrations of BSP than samples from the portal vein a t the hilum of the liver. After biliary stasis this ratio continued to rise (K21, N3). These results, together with the indication that bile secretion continues during biliary stasis (B9), suggest that the biliary system acts as an open tube with pores along its length. Obstruction does not alter the excretion rate but causes leakage to occur all along this conduit into an extracellular, extravascular space. This leakage begins in the sinusoidal-bile canalicular region and extends along the biliary tree to lymphatics and, thus, to the blood stream. This process may be selective, involving certain biliary constituents only. It is also possible that the peribiliary arterial plexus may be involved in the transfer of BSP from bile to plasma and lymph. I n contrast, indocyanine green concentration in lymph approximates that iri plasma (H28) indicating that this dye does not leak out of the biliary system or that equilibration occurs between plasma and lymph in the case of indocyanine green, possibly because i t is not conjugated.

326

PAULA JABLONSBI AND J. A. OWEN

4.7.

INTRACELLULAR RELATIONSHIPS

The dye removed from the plasma by the hepatic cells may be metabolized before excretion into bile as in the case of BSP, or it may be excreted without metabolism, as in the case of rose bengal and indocyanine green. When BSP is injected intravenously there is a time lapse before it reaches maximum concentration in the bile (C2, C7, C8, C34, R10). This has been attributed to storage of BSP in the liver and possibly to its conjugation, although some of the delay is due to the dead space of the biliary system in analogy to the delay time of the urinary tract (C18, M23). It is difficult to ascertain whether conjugation of BSP is essential for excretion because free BSP is present in bile. We have found that initially after a single injection, only a small percentage of the biliary dye is conjugated ; this percentage increases to almost 100%. However, when dye is continuously infused, the biliary level of unconjugated dye remains constant. These results indicate a relationship between the excretion of free dye and its level in the cells; under conditions where the free dye is not replenished there is a rapid depletion by conjugation and biliary excretion and, thus, the level of free dye in the bile continually diminishes. The maximum excretion rate of BSP (free plus conjugated) is not apparently affected by the percentage of free dye in the bile which may vary from 20 to 60%. It is, thus, possible that the carrier for biliary excretion of BSP does not discriminate between the different forms of the dye and that conjugation is not essential for this transfer. However, the mechanism for the hepatic uptake of BSP from plasma is specific for the free dye; the conjugated dye is removed from plasma a t a much slower rate than the free dye (K16). Conjugation may be important in the storage of the dye. I n general, the hepatic storage of BSP and other dyes may be considered to be the result of the disproportionate capacities of the two transport mechanisms (G16), that is, the capacity for dye uptake exceeds the maximal capacity for dye excretion. However, the high capacity of the uptake mechanism may, in part, be due to a rapid conversion of B$P to its conjugate and the low affinity of the uptake carrier for this form of the dye. 4.8. EXTRAHEPATIC UPTAKE OF DYE Data obtained from the study of dye distribution in hepatectomized dogs has indicated that about 50% of BSP leaves the plasma within 3 hours (C26, H22). Bengmark et al. (B16) estimated the half-life of indocyanine green in hepatectomized dogs to be 4 hours, although it is not clear whether this finding was from extrapolation or interpolation. Both dyes are bound to albumin and, as approximately 60% of body albumin is extravascular (S36), a fall in plasma dye concentration is to be ex-

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

327

pected from distribution of dye molecules among total body albumin. Passage of dye out of the plasma presumably involves unbound dye for plasma albumin mixes throughout its volume of distribution much more slowly, but binding will recur in the extravascular compartment. HOWever, this finding does not necessarily mean that extrahepatic disposal of dye significantly affects the plasma disappearance rate after a single dose, a t least in normal persons, for removal of dye by the liver is so rapid (see Table 3) that there is relatively little time for extravascular diffusion compared with the time ( 3 4 hours) allowed in the studies with hepatectomized animals. 5.

Dye Tests Used in Clinical Investigations

5.1. THESTANDARD BSP RETENTIONTEST This is the test most frequently used: A dose of BSP is given by intravenous injection, and the percentage of the dose remaining in the plasma a t a given time after the injection is determined. The dose of dye originally proposed was 2 mg/kg of BSP as its sodium salt (Ft23). Later workers (H8, M1, M14,02), however, found that a dose of 5 mg/kg was more useful in demonstrating mild liver dysfunction, and over the past 20 years this dose (5 mg/kg) has tended to become standard. A few investigators have recommended higher doses, such as 7.5 mg/kg (H14, T9) or even 20 mg/kg (ClO), but a return to the smaller dose (2 mg/ kg) has also been advocated (L15) on the grounds that there is significant reabsorption from the gut of BSP excreted in the bile during the course of the test with a dose of 5 mg/kg. Others (05) have concluded that even with the larger dose such enterohepatic circulation is negligible. The dose of BSP is generally determined on a basis of body weight in order to reduce the spread of results in any one category of persons. However, dosage on the basis of lean body mass (B57) or of surface area (A7, A8, 11) appears to be more advantageous. The plasma concentration a t zero-time obtained by extrapolation had a coefficient of variation of 18% when the dose was based on body weight and only 8% when it was based on lean body mass (B57). I n one instance, a patient of 90 kg and a surface area of 1.82 m2 had an abnormal BSP result (26% retention) when given a dose of 5 mg/kg, but a normal result (3% retention) with a dose of 150 mg/m2 (A8). The results of the BSP retention test are expressed in terms of the percentage of BSP remaining in the circulation a t a given time after the injection. Most investigators have taken a single sample a t a given time in the period from 30 to 60 minutes after the injection, with 45 minutes the most popular choice. Robinson (R14), however, took blood samples

328

PAULA JABLONSKI AND J . A. OWEN

only at 2.5 and 7.25 minutes. Tests with multiple samples have been termed fractional BSP tests (D7). A slide rule for easy calculation of BSP retention values has been devised by Gindler (G6). The actual measurement made is the concentration of BSP in plasma, and it is usually assumed that the plasma volume is equal to 50 ml per kilogram of body weight. An initial dose of 5 mg/kg is taken to give a zero-time concentration of 10 mg/100 ml so that the percentage of BSP remaining in the plasma is obtained by multiplying the plasma concentration a t the given time by 10. Some have preferred to take multiple blood samples and to determine zero-time plasma concentrations by extrapolation (Ll). Some values obtained for the initial volume of distribution of BSP are listed in Table 2 along with similar data for other substances. Since the initial volume of distribution is, in effect, the plasma volume, it is TABLE 2 INITIALVOLUME OF DISTRIBUTION OF DYESAND OTHER SUBSTANCES IN HEALTHY ADULTS Initial volume of distribution

W/kd

Substance

BSP

Males Females

48.4 46.4 4 0 . 0 f 8.7 40.9 f 6 . 2 40, go*" 43.5,' 47.P

Rose bengal

Males

42.4 f 6 . 2 38.6 f 6 . 6

Females

Indocyanine green Evans blue

Vital red 1311-albumin

a b

53.6"eb

44.1 f 5 . 9 45.4 f 5 . 4 47.5 40.7 f 4.4 43.2 43.3 f 6 . 0 41.2 42.0 53. 6"sb

Result derived from reported data. Assuming a mean body weight of 65 kg.

Ref ererices

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

329

evident from the data presented in Table 2 that, in health a t least, calculation of plasma volume on the basis of 50 ml per kilogram of body weight leads to an underestimate of the “zero-time” plasma concentration and to an overestimate of BSP retention. In disease, the ratio of plasma volume to body weight (ml/kg) is not always as it is in health. A mean plasma volume of 63 (56-71) ml/kg was found in 6 patients with chronic liver disease who had a mean BSP retention of 3.3 (0-6) %, after a dose of 5 mg/kg (M17). The mean zerotime plasma concentration was found by extrapolation to be 8.1 mg/100 ml instead of 10.0 mg/100 ml as intended. When the BSP retention values were corrected for underestimation of plasma volume, a mean value of 15 (8-25%) was obtained. However, the procedure used in this correction was not disclosed. Errors in estimating the initial volume of distribution are not removed by basing the dose on surface area. Thus, Ingelfinger et al. (11) gave doses based on surface area and obtained zero-time plasma values within 10% of those expected in only 45 out of 50 healthy persons and only 41 out of 58 patients. Other factors involved in predicting plasma volume have been reviewed elsewhere (12, 04). Normal values for BSP retention are given in Table 3. It should be noted that the normal range selected by different investigators has included a variable percentage (90-100%) of normal persons. TABLE 3 BSP RETENTION IN NORMALS” Dose (mg/kg)

Time after injection (min)

BSP retention*

(%I

References

~

2

5

20

20 18 15 20

0 0

25 30 30 30 4.5

0 10 6 0 4

45 45

10

120

5 0

5 2.5

(CW

It should be noted that different investigators have defined their normal ranges in terms of different percentages (90-100%) of normal persons. Upper limit of normal range.

330

PAULA JABLONSKI AND J. A. OWEN

mg/100 ml

mg/100 ml

I0.C

I

5.c

I .c

0.I

1 20 40 60 801

FIG. 3. BSP disappearance curves after single injection (5 mg/kg) to a normal subject. Concentrations are expressed as milligrams per 100 ml of plasma. (A) Semilogarithmic graph. (B) Linear graph of the results of Winkler (W24).

5.2.

PARAMETER^ DEFINING BSP DISAPPEARANCE FROM PLASMA

5.2.1. Interpretation of BS? Disappearance Curves

When BSP is injected intravenously in normal people, the curve obtained by plotting the plasma BSP concentration ( p ) against time can be represented by the sum of two exponential curves, as shown in Fig. 3. The equation p t = p , (e-at be+) is accepted as a description of these curves, where po is the concentration a t zero time and p t is the concentration a t time t ; a,p , and b are constants. The plasma disappearance curves of rose bengal (B4, L4), and indocyanine green (L5, L8, 537, W9) are similar. Attempts have been made to explain the shape of the plasma disappearance curve of these dyes in terms of physiological processes. Evans (E7) considered a model system comprising three compartments, the plasma,

+

Extrahepatic space

Plasma space

Liver

FIG.4. Kinetics of dye disappearance from plasma: Model proposed by Evans (E7).

BROMSULFOPHTHALEfN AND OTHER CHOLEPHILIC DYES

331

TABLE 4 DISAPPEARANCE RATESOF D m s FROM PLASMA IN NORMALS‘ Dye

BSP

B

ff

0.168’’ 0.119 f 0.012 0.154b 0.188* 0.131 0.145 f 0.018 0.140 0.113 f 0.018 0.138 f 0.04 0.117

-

0.115* Rose bengal Indocyanine green

References

0.089 0.048 f 0.006 0.044‘‘

0.15 0.259 f 0.082 0.185 f 0.031 0.180 & 0.021 0.232 0.186 f 0.028

a is the exponential coefficient for the first phase of the plasma disappearance curve; B is the coefficient for the second phase (see Section 5.2.1). Data are expressed as means k standard deviation (where reported). ’’ Data have been standardized for comparison.

the extrahepatic compartment, and the liver (Fig. 4). Passage of dye from plasma to the extravascular compartment was considered to be reversible, but transfer from plasma to the liver was considered irreversible. This model, though providing a satisfactory explanation of the shape of the curve, fails to account for the reduced slope of the first phase which occurs in liver disease ( D 7 , I l ) . Alternative three-compartment models comprising plasma, liver, and bile have been proposed (B12, R10, T12, W25). These models involve reversible transfer of dye from plasma to liver but irreversible transfer from liver to bile (Fig. 5 ) ; the first phase of the curve is referable to the uptake of dye by the liver, and the second to the excretion of dye into bile (Fl, H25, H26). A more complex four-compartment system has been proposed by Brauer (B45)based on data obtained from experiments involving simultaneous injection of radioactive BSP and unlabeled BSP and by others (M24). Some values for the exponential coefficients (a#) in normals are given in Table 4. The value in individual patients has been found to vary with dose (N10, W24). (Y

332

PAULA JABLONSKI AND J . A. OWEN

Plasma space

Liver space

Bile

FIQ.5. Kinetics of dye disappearance from plasma: Model proposed by Winkler (W24-W26), Richards (R9, RlO), and Barber-Riley (B12).

I n persons with impaired hepatobiliary function, the plasma disappearance curve of dye may differ from that observed in normal persons. Sometimes the initial fall of BSP is much slower, so that a plot of the plasma concentration against time is almost linear and the second phase cannot clearly be distinguished. More commonly the curve in the second phase is flattened or shows a temporary rise in BSP concentration which is believed t o be due to the passage of BSP or of its metabolites back into the plasma from the liver. 5.2.2. Fractional Disappearance Rate (FDR)

The use of kinetic parameters to define BSP disappearance from plasma has been subject to a serious nomenclature problem. The nomenclature used in this review along with some of the synonyms from the literature, has been listed in Table 1. The first exponential phase of the plasma curve can be represented. p , = pOe-at (1) where p , is the concentration a t time zero and p , is the concentration a t time t. The same curve can be represented p t = pol0-dt

The parameter a represents the fraction of dye removed from the circulation in unit time, which we term the fractional disappearance rate (FDR) (see Table 1) ; 100 times gives the percentage of circulating BSP leaving the plasma in unit time, that is, the percentage disappearance rate (PDR) An estimate of fi can be obtained graphically by plotting p against time on semilog paper. However, this involves collection of several blood samples within minutes of the injection, and it is usually more convenient to calculate FDR (or PDR) without determining p o . For example, if pl and p , are concentrations a t time tl and t2, respectively, (Y

.

pl = p 0 c a t 1 p z = pOe-atl

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

333

therefore pz =

ple-a(tY-tl)

and u =

111pi - 111 pz - 2.3 (log pi - log pz) 22

- tl

tz

- tl

similarly =

log pl - log pz ( t z - tl)

The equation p2 = p1dtst1 has been used to express the rate of fall of plasma BSP after a single injection. Comparison of this equation with Eq. (2) shows that d is identical with e-a. Paradoxically, Ingelfinger et al. (11) use the expression 1-d as an estimate of the FDR ( a ) . However, ass-0 1 - e-a

and so the approximation is valid. An alternative approximation of expression a = (Pl

+ a!

a

was obtained (L11) by the

- pz)/(p1pa)l’*

where ( p l p z )lI2 is used as an estimate of the mean plasma BSP in the time interval between p l and pa. 5.2.3. Plasma BSP Clearance The plasma clearance (C,) of a substance is the volume of plasma just containing the amount of substance removed in unit time. The plasma clearance of BSP (Fl, L11, L12) is the fractional disappearance rate ( a ) multiplied by the plasma volume (Vp), thus: CBSP

= uvp

Problems in measuring plasma volume or in its estimation have been discussed (see Section 5.1). The determination of V , may be avoided by infusing BSP a t the rate, I , giving constant plasma concentration, peq. In such circumstances,

Z

CHSP

=-

Pt 9

The plasma clearance of BSP multiplied by the plasma concentration indicates the rate of loss of BSP (mg/minute) from the plasma. Some values for plasma clearance in normals are listed in Table 5. The

334

PAULA JABLONSKI AND J. A. OWEN

TABLE

5

PLASMA CLF,ARANCE OF DYESIN NORMALS

References

BSP

5.1 f 0.12 5.0 6.5 f 0 . 9 O

Indocyanine green

7.5 f 2.0

(WW

Assuming a body weight of 65 kg.

half-life of BSP in plasma is also used as a parameter of BSP disappearance (V10, V12).

Other Parameters of the Plasma Disappearance Curve A few investigators (T7, T8) have advocated use of the second exponential coefficient ( p ) as well as, or instead of, a parameter from the first phase. The ratio of the two exponential coefficients ( c Y / ~ )has been found to provide additional information (T7, T8). A more detailed mathematical analysis of the plasma curve, in accordance with a threecompartment model system, has been carried out to obtain estimates of the rate of dye uptake by liver cells, the rate a t which dye returns to the plasma from the liver, and the rate a t which dye is transferred to the bile (C21, R10). Calloway and Merrill ( C l ) have expressed BSP disappearance rates as the slope of log plasma BSP plotted against log time. The validity of this approach remains to be established. 5.2.4.

OF BSP UPTAKEBY 5.3. PARAMETERS

THE

LIVER

5.3.1. Maximum Rate of Uptake of BSP by Liver (Lm)

The rate of uptake of BSP by the liver depends on the plasma concentration, and measurement of the maximum rate therefore requires infusion of BSP at a sufficiently high rate to “saturate” the BSP transport carrier (see Section 4.2). If i t is assumed that there is negligible uptake of BSP from the plasma except by the kidney and the liver, the amount taken up by the liver, L,, in unit time equals the rate of loss from the plasma, R , less the rate of loss from the urine, U , i.e.

L,=R-U The net amount lost from the plasma, R , equals the rate of infusion of BSP, I, less the rate a t which BSP is accumulating in the plasma, i.e.,

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

335

I - Ap/At * V ,

R

=

=

I - U - Ap/At. V,

Thus

L,

5.3.2. BSP Storage Capacity of the Liver ( S ) and Maximum Rate of Transport to Bile (T,)

The parameter L, equals the sum of the hepatic storage and the biliary excretion rates. More commonly the latter two parameters have been determined separately (W10, W13). The hepatic storage capacity, S, is a measure of the amount of BSP stored in the liver in relation to the plasma concentration. The parameter S is usually expressed in terms of milligrams (in the liver) per milligram per 100 ml (in the plasma) and is determined by infusing BSP a t two different rates, both greater than the maximum rate a t which the liver can transport BSP from the plasma to the bile. Because the infusion rate exceeds the rate of biliary excretion, the plasma concentration rises as in the determination of L,. The calculation of the hepatic storage capacity, S, requires the assumption that the rate of transfer of BSP from the liver to the bile has an upper limit, T, (W12). The rate a t which BSP is stored is given by the expression S.Ap/At. The rate a t which BSP is lost from the plasma equals this rate of storage plus the maximum rate of biliary excretion, thus

R

=

S ’ Ap/At

+T,

(3)

Thus, if I is the rate of infusion and U the rate of urinary excretion

+ U + Ap/At V , substituting for R , I = (S + V,)Ap/At + T, + U. I =R

*

Measurements are made a t two infusion rates giving two values for

I , Ap/At and U . V , is constant, and therefore S and T,,, can be determined. Reference to Eq. (3) shows that S is not a storage capacity in the ordinary sense. It has the dimension of volume and represents the volume of plasma which would contain the amount of BSP stored in the liver. In this light, the storage parameter S (milliliters of plasma) is analogous in a static sense to the classical clearance parameter C (ml/minute of plasma). Estimates of the rate of hepatic uptake and the rate of transfer to bile have also been obtained from detailed analysis of the plasma disappear-

336

PAULA JABLONSKI AND J . A. OWEN

TABLE 6 FOR BSP HEPATIC UPTAKE PARAMETERS

a

IN

NORMALP

Data are expressed as means and standard deviations.

ance curve, R (R10) (see Section 5.2.4). Values obtained for L,,, S, and T, in normals are listed in Table 6. I n determining S and T,, usually no allowance has been made for regurgitation of conjugated BSP. We have observed that up to 45% of BSP in the plasma of patients during such measurements is conjugated. Clearly regurgitation of this magnitude must affect interpretation of changes in these parameters in disease. 5.4. MEASUREMENT OF BILIARY EXCRETION I n health, BSP appears in bile 5-20 minutes after injection and continues to be excreted for 5-6 hours (C4). Measurement of the appearance time of BSP in the bile obtained by duodenal intubation (C7, C8,S14), or from a T-tube after choledochotomy (N130), has been suggested as a test of liver function. Normal persons and patients with hepatitis had appearance times which never exceeded 20 minutes whereas longer times were characteristic of patients with biliary obstruction (C7). Measurement of the amount of BSP in bile coIlected by duodena1 intubation was found to be a more sensitive index of liver function in convalescent hepatitis patients than various other biochemical tests (W31). TABLE 7

Dose

0

URINARY EXCRETION OF BSP

IN N o R U L s "

Urinary excretion (% of dose)

Time

Data are expressed as means and range (where reported).

References

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

337

5.5. URINARY EXCRETION TESTS A small proportion of BSP given to healthy persons a s a single injection is excreted in the urine (Table 7 ) . Larger proportions are excreted by persons who have increased BSP retention (e.g., from hepatobiliary dysfunction). At any given time after injection, the plasma BSP in such persons is higher than in normals and presumably allows more BSP to be excreted by the kidney (11, N7, R23, W22). The maximum rate of urinary excretion occurs 20 minutes after the injection. Assuming that the renal clearance of BSP is independent of the plasma concentration and that the plasma concentration after a standard dose falls exponentially over the period of excretion into the urine with a decay constant a, the amount of BSP ( U ) excreted by the kidney during the time t is given by equation

where CK,Bis the renal clearance. On integration

u;=CR,B . po(l - p ) a

For large values of t CK.B

= a'

Ulp,

(4)

For the standard dose, a = 0.15, po = 10 and 1%of the dose is excreted in the urine (Table 7). Substituting these values, Eq. (4) gives a value of about 5 ml/minute for CK,,+ This figure reflects the protein-binding of BSP (see Section 4.2). There is evidence that less than 2% of the BSP in plasma after a standard dose is unbound (G15). Allowance for this gives a minimal clearance for BSP of 250 ml/min, which suggests that either free BSP or its conjugate or both are secreted by the renal tubules. Benemid, which is known to depress renal tubular activity (M5), reduces the renal excretion of BSP (B28) which supports this view. The percentage of conjugated BSP in urine is usually higher than in serum. This could reflect the reduced affinity of albumin for the conjugate (B3) or the conjugate could have a higher renal clearance. Alternatively, the conjugate may be synthesized in the kidney, which has some S-aryl glutathione transferase activity (B35, G25). 5.6. T m ~ USING s OTHERCHOLEPHILIC DYE~ 5.6.1. Indocyanine Green

An indocyanine green retention test similar to the BSP retention test (C16) gave a mean retention value in health of 3.976, 20 minutes after a

338

PAULA JABLONSKI AND J . A. OWEN

dose of 0.5 mg/kg. The initial volume of distribution is similar to that of BSP (Table 2). Most investigators who have used indocyanine green to assess liver function have measured its F D R or PDR (Cl6, H28, M17). Some values for F D R in normals are listed in Table 4. After intravenous injection, the plasma decay curve of indocyanine green is more closely represented by a single exponentiaI curve than that of BSP (C16, H14). There is no detectable indocyanine green in the urine even in patients with hepatic disease, which may be due to the instability of the dye in the absence of protein (L5), and no demonstrable enterohepatic circulation (L5). 5.6.2.

Rose Bengal

The use of rose bengal for the clinical evaluation of liver function was first proposed by Delprat (D4). Usually blood samples are collected a t 2 minutes and 8 minutes. The amount of dye present a t 8 minutes expressed as a percentage of the dye a t 2 minutes is taken as a n index of liver function. Normally not more than 50% of dye remains at 8 minutes (D5, K6, 531, 532). The use of multiple blood samples has been tested (S2), but this offers little advantage over collection of two samples. The FDR (a) of rose bengal has also been used as a n index of dye uptake by the liver (52). More recent studies with 1311-labeled rose bengal have demonstrated a normal plasma decay curve similar to that of BSP with a mean F D R (a) of 0.15 (L4).

5.6.3. Phenol 5,6-Dibromophthalein (Phenol Dibromophthalein Disulfonat e ) This close relative of BSP has been used in experimental studies only. Studies in animals (B38, H2, J3, K9) have shown that, unlike BSP, it is excreted by the liver without conjugation. The plasma disappearance curve of phenol 3,6-dibromophthalein was unaffected by benziodarone, a compound which reduces BSP clearance probably by inhibiting conjugation (B38). I n rats, the parameters S and T, for this compound have been found to be similar to those for BSP (K9). However, a protein-free diet reduced the T, of BSP only. The clinical value of tests using phenol 3,g-dibromophthalein is yet to be determined. 5.7. ESTIMATION OF HEPATIC BLOOD FLOW The estimation of hepatic blood flow in man or in experimental animals has been based on the Fick principle (using either uptake of substances

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

339

such as dyes, ethanol, or galactose or release of substances such as urea), indicator dilution procedures, calorimetric techniques, and the direct collection of blood from the hepatic vein. Only those involving dyes are considered in this review. Information on other procedures is to be found elsewhere (H9, W l ) . 5.7.1. Ficlc Principle Hepatic blood flow is obtained from the hepatic plasma flow and the packed cell volume. The hepatic plasma flow F , is obtained from the expression

FL = L/(PHA- PHV) where p H Ais the mean dye concentration in hepatic arterial plasma and pHvis the mean dye concentration in hepatic venous plasma, and L is the rate of uptake of dye by the liver. I n practice, estimation of hepatic blood flow usually involves the infusion of dye, but measurements have also been made using a single injection of dye. I n the former method, a dye is infused a t a constant rate after a priming dose until the plasma concentration is steady. Hepatic venous blood is obtained by catheterization. The rate of uptake of dye by the liver is usually taken to be the rate of infusion of dye without allowing for urinary excretion. BSP (B42, C9, S22, W27, W30), indocyanine green (L9, R3, W18, W27, W28), and isotopically labeled rose bengal (L4,W27) have been used in this way. I n the single injection procedure (B5),the plasma clearance is determined (see Section 5.2.3) and the rate of uptake by the liver ( A ) is obtained from the relation

L

= C e s ~PHA .

where PHA is the mean plasma arterial BSP. This relation does not allow for urinary loss of dye. Plasma leaving the liver is not completely cleared of dye. The extent of clearance is expressed in the extraction ratio (Ex)where

EX =

(PHA

-

PAV)/~HA

The extraction ratio for a particular dye tends to vary inversely with the plasma concentration (C9). There is also a tendency for the extraction ratio to vary inversely with the hepatic plasma flow. This has been reported after portacaval shunt operations in man (B43) and after hemorrhage or pyrogen administration in dogs (H7a). The validity of this method has been discussed a number of times (B42, S20, 522, W l ) . An important source of error is the assumption that

340

PAULA JABLONSKI AND J . A. OWEN

the rate of uptake of dye by the liver equals the rate of infusion of dye in the infusion method or the rate of loss from the plasma in the injection method. This source of error is considered to be less a t higher plasma concentrations than a t lower plasma concentration and to be greater with BSP than with indocyanine green (L4, W27). Some values for hepatic blood flow in normals are given in Table 8. TABLE 8 SOME ESTIMATES OF HEPATIC BLOODFLOWIN NORMALS OBTAINED USINGDYES ~

a b

~~

Dye

Method

Liters/min

BSP BSP BSP BSP BSP IGa IG

Infusion lnfusion Infusion Infusion Injection Injection Infusion

1. 43 -

-

Liters/ 1.73 m2/min* 1.50 1.45 1.08 1.50

-

1.21 0.83

References (B42) (524) (L5) (~30) (W23) (WW (L5)

Indocyanine green. Calculated from original data.

5.7.2. Indicator Dilution Method

I n this procedure, which is primarily for use with experimental animals, hepatic blood flow is calculated from the dye dilution curves recorded from the hepatic vein after injection of dye into the portal vein near its entrance to the liver. Hepatic blood flow in dogs has been measured with this technique using indocyanine green and correcting for uptake of dye by the liver (M34). It would seem advantageous in this procedure to use a dye that was not taken up by the liver. 5.8. MEASUREMENT OF BODY FLUIDSPACES 5.8.1. Plasma Volume The possibility of measuring plasma volume using BSP has been described earlier (see Section 5.1). More commonly, *811-albumin (B44, C16, 12, Z6) or the dyes Evans blue (A7, B44, G22, SlO) or vital red (G22, R2, S44) have been employed, but rose bengal (B4, S30), and indocyanine green (Cl6) have also been used. 5.8.2. Ascitic Fluid Volume

BSP has been used to measure ascitic fluid volumes (B3a). A standard dose of BSP is injected intraperitoneally, and a sample of ascitic fluid is

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

341

collected after 2 hours to allow for mixing throughout the peritoneal space. The ascitic fluid volume (V,) is obtained as follows: dose of BSP injected (mg) v* = concentration of RSP in ascitic fluid (mg/ml) Values for ascitic fluid volume obtained by this means agree with those obtained using 1311-albumin (S22),

5.9. USE OF RADIOACTIVE CHOLEPHILIC DYES BSP has been labeled with 35S in the sulfonic acid groups (B50).It has also been labeled with l3II (57, TlO), but the site of iodination is uncertain; by analogy with the iodination of (unlabeled) rose bengal it has been assumed to occur in positions ortho to the hydroxyl group (J7), but the possibility that some iodine is exchanged for bromine has not been excluded. 35S-labeled BSP has been used for studying the distribution of BSP within tissues (B50, G15) and in investigating the maturation of liver function in chick embryos (B45). It has also been used, in conjuction with unlabeled dye to show that the rate of hepatic uptake of the dye injected into the hepatic artery is the same as that after an injection into the portal vein (B51) (see Section 4.1). Injection of ?t-labeled BSP with and without unlabeled dye has been used to study the effect of dosage on dye uptake (T10). Rose bengal has been labeled with 1311by substitution for lZ7I((27, T3). After injection the rate of decay measured by external counting over the head closely parallels the rate of decay in the blood (N8, N9). There is also good correlation between the retention of rose bengal (measured by external counting) and the retention of BSP measured by the standard procedure (N9). External counting over the liver gives curves which reflect uptake and excretion (T3) of dye. Although the use of isotopically labeled cholephilic dyes has provided information about the behavior of nonlabeled dye (E5) , it has been more used in radioscanning as a means of determining the size, shape, and position of the liver and of localizing areas showing functional changes, e.g., tumor metastases. The principles and practice of radioscanning are outside the scope of this article, and the reader is referred to other sources of information (F3, L7). 5.10. TOXIC EFFECTS OF DYE With doses up to 7.5 mg/kg given as a single injection, systemic reactions t o BSP are rare. Transient headaches, feeling of faintness and chills 45-60 minutes after injection of BSP (5 mg/kg) have been reported

342

PAULA JABLONSKI AND J. A. O W E N

in a small percentage of individuals. However, doses of 20 mg/kg were followed by pain in the lumbar region 8-16 minutes after the dose in 15% of healthy volunteers (B57). Thrombophlebitis has been observed after intravenous injection with an incidence of 0.6% (S21), although it seems doubtful that thrombophlebitis is related specifically to the use of BSP. Rosenthal and White (R23) found that doses of 100 mg/kg killed 50% of dogs; one dog died with a dose of 50 mg/kg. However, others (L5) have found that BSP is well tolerated in man in doses of up to 50 mg/kg given over 3 hours. About 25 severe systemic reactions including 9 deaths to standard (5 mg/kg) or smaller doses of BSP have been reported up to 1966 (C14, K3, P9). At least three of the fatalities occurred in patients who received BSP for the first time and had no previous history of allergy. Unfortunately, the irritant properties of BSP make intradermal testing impractical (M31). Cautious injection of BSP is widely recommended, but of unestablished efficacy. Indocyanine green is less irritating on injection (L4)than BSP and has not apparently been found to cause severe systemic reaction. However, the number of patients who have to date received indocyanine green is much smaller than the number who have received BSP. Rose bengal is not toxic to animals in doses of 5 mg/kg, which is 2-3 times the usual dose. Severe reactions have not been reported. However, rose bengal, like other triphenylmethane dyes, exhibits photodynamic activity (S9), so that patients receiving it may show photosensitivity and blood containing it may undergo some hemolysis when exposed to sunlight. 6. Effects of Physiological Factors on Dye Uptake and Secretion

6.1. AGE

Most investigators have found that the removal of BSP from plasma is slower in newborn infants than in adults ( H l l , K4, M28, Y l ) , but one study failed to reveal a significant difference (Sl). The use of a smaller dose (2 mg/kg) may have been responsible for this. Ability to eliminate BSP increases with the age of the infant but is still less than in adults a t the end of the first year of life (K4). I n older children, the excretion of BSP is slightly less than in adults when computed in terms of body weight or surface area, but there is no difference when it is related to length or to the basal metabolic rate (V11). Studies in premature infants have demonstrated a slower removal of BSP than in adults ( 0 1 ) . Tests made during the first 24 hours show a slow, almost linear, disappearance of dye from plasma; later the plasma disappearance curve shows two phases as found in adults. Age since birth

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

343

is more important than weight in determining the removal rate of BSP, so that a newborn full-term infant generally shows a slower removal of BSP than a premature infant of some days or weeks. Reduced BSP removal in neonates is probably due to an insufficient rate of excretion into the bile because hepatic uptake and conjugation are adequate (V7, V8). However, changes in BSP handling with age in neonates has been attributed to circulatory changes ( 0 1 ) . No correlation has been found between plasma bilirubin and BSP retention in the neonatal period (M28, Y l ) . I n the higher age groups (over 50 years), there is again increase in BSP retention ((31). A standard dose (5 mg/kg) in a person of this age tends to give higher initial plasma concentrations than in young adults. This has been attributed to circulatory differences (Cl) although the mean plasma volume does not vary with age in adults (SlO).

6.2.

SEX

No significant sex difference has .been found in the results of the standard BSP retention test (R14) or in plasma disappearance parameters (T6) in spite of the greater plasma volume found in males (B4, R14). However, the liver parameters S and T, are higher in men than in women (P11). A tendency for higher BSP retention values to occur toward the end of the menstrual cycle in women taking oral contraceptives has been noted (A6), but no such effect can be demonstrated in control subjects. The percentage disappearance rate of indocyanine green is slightly higher in normal males (H28). 6.3. PREGNANCY BSP retention is increased in the latter half of normal pregnancy (B56, C19, T6). The storage parameter S is increased (C30, K1, S29), but the transport parameter T, has been found to be normal (Kl) or reduced (C30). Detailed kinetic studies (T6) have shown (a) a slight increase in the rate of uptake of BSP from plasma; (b) a 2-fold increase in the return of dye t o the plasma; (c) a reduction by half to two-thirds in the rate of elimination into the bile, and (d) an alteration in the proportion of dye lost from the liver cells per minute to the plasma and bile, respectively. These changes disappear early in the puerperium and are probably secondary to the raised estrogen level (T6) (see also Section 7.17). Except for elevation of plasma alkaline phosphatase by enzyme released from the placenta (B24, W17), other biochemical indices of liver function are usually normal in pregnancy though very occasionally pregnancy may be associated with cholestatic jaundice (K2, 528, S47), possibly conditioned by a genetic factor (H21). BSP mas not detected in 27 specimens of cord blood collected 7-90

344

PAULA JABLONSKI AND J. A. OWEN

minutes after BSP was given to the mother or in 18 specimens of liquor amnii collected 3-96 minutes after BSP was given to the mother (T6).

6.4. FEEDING, EXERCISE, AND

POSTURE

Although it is customary to carry out BSP tests on fasting patients, eating has no appreciable effect on the percentage disappearance rate (V6). Diet may influence the parameters of BSP removal in humans since protein-deficient diets in rats reduce BSP removal (C29, K9). Strenuous exercise reduces the percentage removal rate of BSP to a variable extent (M17). Assumption of the upright position after lying down reduces the hepatic blood flow (C37) ; although the latter change does not necessarily affect the excretion of BSP (see Section 5.7)) it seems sensible to perform the test with the patient in the supine position. Fainting reduces the rate of clearance of BSP from the plasma (R14), possibly by increasing the plasma volume. 6.5. TEMPERATURE

Hypothermia reduces plasma BSP removal rate in rats (B47, V2, V4) and in dogs (B58). The biliary transport maximum (T,) was found (R13) to be reduced by hypothermia induced by pentobarbital which on its own depresses BSP removal (see Section 7.11) so that the effect cannot be attributed unequivocally to cold. However, direct cooling of dogs to 27O-28'C by extracorporeal circulation through a cooling bath reduced bile flow and biliary excretion of BSP although the biliary BSP concentration was unchanged (S40). Schoenfield et al. (517) reported that fever induced by pyrogens does not influence the 60-minute BSP retention or the parameters S and T,. Others (B40,B41, H13) found that pyrogen reduces BSP uptake. Since pyrogens increase the hepatic blood flow (B40), they must presumably reduce the extraction ratio. Direct warming of dogs to 42°C had little or no effect on the bile flow or on the biliary excretion of BSP (540).

7. Effects of Drugs on Dye Uptake and Excretion Drugs and other chemicals may affect the disposal of dyes a t any of the steps involved in their transfer from plasma to bile, by: 1. Competition for binding sites on plasma proteins; it is probable that protein-bound dye is more readily taken up than free dye. 2. Competition for binding sites on intracellular proteins. 3. Competitive or noncompetitive inhibition of uptake or excretory systems. 4. Alteration of cellular metabolism in general, leading to depletion of

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

345

necessary cofactors, e.g., glutathione. Alteration of the permeability of cellular membranes. 5. Inhibition of conjugating enzyme to an extent that could influence storage and/or excretion of the dye. 6. Morphological changes of the biliary system (e.g., intra- or extrahepatic cholestasis) or of the cell (e.g., necrosis). Drugs could also influence dye transfer nonspecifically by: 7. Changing the hepatic blood flow. 8. Increasing the concentration of a compound (e.g., bilirubin, bile acids), which influences BSP retention. Such an effect could be obtained through an increased enterohepatic circulation of the compound (L15). These nonspecific effects are excluded in the following account of some drugs known to influence BSP retention.

7.1. CHOLECYSTOGRAPHIC AGENTS Bunamiodyl (B19, S26) and iodopanoic acid (S17) cause an elevation of BSP retention. These compounds, and the related iodinated aryl compound iodipamide, competitively inhibit the enzyme S-aryl glutathione transferase (B38). However, these compounds also affect the excretion of bilirubin (B21, B33), which is excreted as a glucuronide, of indocyanine green (H4), which is not conjugated by the liver, and of conjugated bilirubin (B21). Bunamiodyl and iodipamide reduce the uptake of BSP by liver slices (B6, B7) and by the isolated perfused liver (B19). It has been suggested that these compounds compete with BSP, bilirubin, and indocyanine green for cellular protein-binding sites (R9) and that their major site of action is an excretory mechanism (H4) which is common to all these structurally different compounds. It should be noted that some iodinated phthalein dyes, e.g., tetraiodophenolphthalein and phenol tetraiodophthalein have been used in cholecystography. These compounds are very similar to BSP structurally and probably competitively inhibit uptake and excretion of the dye. 7.2.

PROBENECID (di-n-propylsulfamylbenzoic acid, Benemid)

Probenecid reduces the hepatic clearance of BSP (B23, B25) as well as inhibiting the renal clearance of penicillin, phenol red, and PAHI (M5). Probenecid is also a choleretic (G7) ; the concentration of both BSP and bilirubin in bile are reduced, but the excretion rate of BSP only is affected (S39). There is an increased reflux of dye from the liver to plasma (G7, S39). The effect of the drug as a choleretic and a s an inhibitor of BSP PAH = p-amino hippurate.

346

PAULA JABLONSKI AND J. A. OWEN

excretion is dose dependent (S39). Probenecid also lowers the uptake of BSP by rat liver slices (B6). ACIDAND RELATED COMPOUNDS 7.3. FLAVASPIDIC The use of an extract of male fern (Dryopteris felix-mas) for treatment of tapeworm infestation frequently leads to an increased retention of BSP and a transient hyperbilirubinemia (N11). This extract contains a large number of phloroglucinol derivatives (S45), some of which inhibit BSP uptake, e.g., flavaspidic acid, flavaspidic acid-N-methyl glucaminate, p-aspidin, methylene-bis-norflavispidic acid, and filixic acid, whereas phloretin, phlorizin, or phloroglucinol itself have no effect (N12). Rose bengal and bilirubin retention are also elevated, but the effect on indocyanine green retention is less marked (N12). Flavaspidic acid and these related compounds may interfere with hepatic uptake of BSP, rose bengal, and bilirubin by competition for intracellular binding sites (H2, N12). However, flavaspidic acid inhibits S-aryl glutathione transferase (B38) and UDP-glucuronyl transferase (H5). Since these drugs also affect transfer of substances which are not conjugated, such as rose bengal (N12) and conjugated bilirubin (H2), it is unlikely that their primary action is due to inhibition of conjugating enzymes. It is probable that the effects in vitro on the conjugating enzymes were obtained at unphysiological levels of the drug. Since flavaspidic acid does not affect the excretion of BSP or of rose bengal (N12) into bile and there is no accumulation of conjugated BSP or bilirubin in the plasma ( N l l ) , the effect of these compounds is specifically on the uptake mechanism. Phlorizin has been reported to have a marked effect on biliary excretion of BSP and bilirubin; it usually increases bile flow, but excretion of both compounds is reduced (J6, S40, V3). The action of this compound may be nonspecific, since it is believed that the drug inhibits the oxidation of all substrates of the tricarboxylic acid cycle (L16) by alteration of the permeability of cell membranes (K5). Phlorizin is also excreted in bile (S34), and its choleretic action is possibly due to the osmotic effect of excretion of this organic anion. 7.4. ICTEROGENIN Icterogenin is a triterpenic acid, isolated from Lippia rehmanii, which causes jaundice in sheep (R12). This compound reduces excretion of bilirubin, conjugated bilirubin, BSP, and phylloerythrin in rabbits (H7) and, more specifically, the maximal transport rate of BSP and bilirubin in man (A16). On histological examination, livers from animals that have received icterogenin show changes in the number and location of lyso-

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

347

somes and in the bile canaliculi (A16), the structures which are probably involved in bile secretion. Liver cell necrosis was not evident. These functional and morphological changes suggest that icterogenin interferes with the excretory mechanism. Since excretion of a great variety of compounds is affected, it is probable that t,he effect of the compound is nonspecific and possibly due to cholestasis.

7.5. ANTIBIOTICS Novobiocin has been associated with hyperbilirubinemia in the newborn (S46),and it has been shown to depress the excretion of BSP (B21a) and indocyanine green (H4, H5) as well as the maximal excretion rate for bilirubin (A3). Novobiocin also inhibits the excretion of bilirubin by liver slices (H5). Novobiocin was found to inhibit the glucuronyl transferase (H4), but this could be secondary to an accumulation of the conjugated bilirubin after inhibition of the excretory mechanism. Novobiocin, in contrast to other agents, also inhibits hepatic uptake of dyes and this may be its main effect (B21a). Rifamycin increases the retention of bilirubin by inhibiting its excretion rate. This antibiotic has no effect on conjugation ( A 3 ) . It is probable that rifamycin would increase plasma retention of BSP, but this has not been tested. It has been postulated that only those compounds which are readily excreted in bile are inhibitory of the BSP and bilirubin excretory mechanisms (H4). Both antibiotics have high biliary excretion rates (B55)and both are concentrated to a considerable degree in bile. Erythromycin, another antibiotic with high biliary excretion (B55)has been implicated in a transient cholestasis (P10). It would be of interest therefore to test the effect of other antibiotics which have high biliary excretion rates, e.g., penicillin. 7.6. PHENOTHIAZINES

Chlorpromazine decreases BSP excretion in rats (C22) and in man (DS). However, in mice, the biliary excretion of BSP is not reduced by this drug (El). Chlorpromazine does not inhibit the clearance of either bilirubin or indocyanine green (H5), but its action and that of other phenothiazines (e.g., promazine, trifluoroperazine, prochlorperazine, and pecazine) is complex. These drugs, cause a cholestatic reaction with a transient cholestatic hepatitis (P10, W4, Z l ) , with an incidence of approximately 1 %. Thus, the effect on BSP retention may be a reflection of the cholestasis induced by these drugs. Their action may also be due to increased membrane permeability (H5).

348

PAULA JABLONSKI AND J.

A.

OWEN

7.7. DINITROPHENOL AND RELATED COMPOUNDS 2,4-Dinitrophenol produces a prolonged choleresis (S39), but the BSP excretion rate remains the same or is slightly elevated (540). This compound also causes a marked pyrexia, but choleresis is not due to hyperthermia since no change in bile flow occurs when animals are warmed to 42OC (S40). A wide range of related compounds have been tested (P12) ; phenols and mononitrophenols had no effect; isomeric dinitrophenols had less effect; picric acid and 2,4-dinitrophenetole were as effective. 2,4-Dinitrophenol is known to uncouple oxidative phosphorylation, but it is difficult to correlate a decrease in ATP with an increased bile flow. Moreover, the 3,5 isomer which is a more potent uncoupler (B59) has a smaller choleretic effect. I n the isolated rat liver 2,4-dinitrophenol causes a decrease in bile flow, the concentration of bilirubin and of BSP in the bile rises slightly (B25, B26), but total excretion rate is reduced.

7.8. ETHIONINE Ethionine ingestion by rats resulted in an increased bile flow and a reduced excretion of BSP. After prolonged treatment, doses of BSP otherwise nontoxic were apparently lethal. Histological examination indicated proliferation of ductules (B11). The effect of ethionine on BSP excretion is possibly due to reduced levels of ATP. However, hepatic glutathione levels are also reduced (H23).

7.9. BJKNZIODARONE Benaiodarone, a coronary vasodilator, causes in vitro inhibition of the BSP conjugating system. This drug increases the retention of BSP in plasma but has no effect on the retention of phenol dibromophthalein disulfonate, which is not conjugated. Benziodarone probably acts specifically on the conjugating enzyme since it has' been shown to inhibit other conjugations with glutathione (B38).

7.10. ETHANOL Increased retention of BSP is common in alcoholic patients (see Section 8.1.2). Generally this is considered to be due to histological changes in the liver. However the excretion of BSP, in particular as conjugated BSP, is reduced in rats treated with one dose of ethanol (L13). Uptake and storage are not affected. When conjugated BSP was administered no reduction of excretion occurred after ethanol injection. Thus, retention of

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

349

BSP in alcoholic patients may result, in part, from impairment of the conjugation process (L13). 7.11. ANESTHETICS AND HYPNOTICS

Anesthetics such as chloroform or ether cause increased BSP retention in dogs (R20) but a mixture of nitrous oxide and ethylene has no effect. Thiopental and cyclopropane cause a fall in hepatic blood flow but the BSP extraction ratio tends to increase ( H l ) . Sodium amytal has been found to reduce BSP excretion in man (V6), and pentobarbital decreases indocyanine green excretion in dogs (H28). It should be noted that many investigators have used short-acting barbiturates for induction of anesthesia but have not allowed for possible effects of these drugs in the interpretation of their findings. Chloral increases bile flow (G26) but both bilirubin and indocyanine green excretion are reduced (H4). The mode of action of these agents on dye excretion is not clear; it is possibly a nonspecific effect on the liver cell. 7.12. DRUGS THAT AFFECTTHE SPHINCTER OF ODDI

Morphine and certain other central analgesics have a contractile action on the sphincter of Oddi (M18). Increased biliary pressure arrests the (‘secondary phase” of BSP disappearance from the circulation ; the initial rapid phase is unaffected. This effect is not demonstrable in patients with intact gallbladders since the gallbladder has some capacity t o regulate pressure. Codeine phosphate has a similar effect to morphine in cholecystectomized patients, causing biliary colic and elevated BSP retention. Pholcodine and 1- (2,12-benzylphenoxyl-1-methylethyl) piperidine has no such effect. Papaverine exerts a counter-effect, relieving colic and BSP stasis (H26). Pethidine and amidone have a similar effect to morphine (B60). 7.13. ANTI-INFLAMMATORY DRUGS Ibufenac reduces hepatic clearance of both indocyanine green and bilirubin (H5). This drug also inhibits glucuronyl transferase (H5) and X-aryl glutathione transferase (B38). In wivo, phenylbutazone depresses bilirubin excretion, but not that of BSP; in vitro, the enzymes conjugating bilirubin and BSP, respectively, are both inhibited. There is no effect on the clearance of indocyanine green (H5). 7.14. PHENINDIONE

This anticoagulant drug causes inhibition of bilirubin clearance and conjugation (H5) as well as inhibition of S-aryl glutathione transferase (B38). The action of this drug on dye excretion has not been reported.

350

PAULA JABLONSKI AND J. A. OWEN

7.15. MISCELLANEOUS ENZYME INHIBITORS Potassium cyanide considerably reduces bile flow (B25). The concentrations of BSP and bilirubin in bile are not diminished (B25), and total excretion of BSP and bilirubin is similar to that of controls (V3). Sodium fluoride reduces hepatic blood flow and increases BSP retention. Excretion into bile is unaffected (B25). Monoiodoacetic acid initially produces a choleresis which is followed by an irreversible inhibition of bile flow with a reduced excretion of BSP and bilirubin (B25). Acetazolamide does not influence biliary output or excretion of BSP and bilirubin (B25). 7.16. COMPOUNDS THAT CAUSEHEPATIC CELLINJURY Carbon tetrachloride administered 5 hours before BSP had no effect on biliary excretion, but the drug injected 25 hours before BSP caused a marked reduction in biliary excretion of the dye. There was also an impaired hepatic blood flow and reduced extraction by the liver ((22, W32). Hepatic cell injury induced by CCl,, dimethyl nitrosamine, or thioacetamide is accompanied by a reduced secretion of biliary components whereas ductular proliferation is observed after chronic administration of a-naphthoisothiocyanate which also exerts a hydrocholeretic effect (G8).All these compounds tend to increase plasma retention of BSP. 7.17.

STEROIDS

Norethandrolone, methyltestosterone, and other 17a-alkyI substituted testosterone-like steroids increase the plasma retention of BSP in man (A16, H6, K14, L6) and sometimes cause hyperbilirubinemia (A16, A17) and elevation of alkaline phosphatase (H6). However, the retention of indocyanine green is not affected by norethandrolone (L6). Natural estrogens in very high dosage also increase the plasma retention of BSP in man (M33) and in rats (G2). Steroidal oral contraceptives in therapeutic dosage have a similar effect in a proportion of women (A6, L l ) . I n the child-bearing age, these rarely produce other biochemical signs of liver dysfunction (R7, R8), but in postmenopausal women the effect of oral contraceptives on liver function tends to be more frequent and more severe (E3, P3). Detailed studies have shown that norethandrolone affects both the uptake and excretory phases of BSP disposal (L6) and the plasma contains more conjugated BSP (58). In high dosage (100-150 mg daily), this drug reduces the storage capacity of the liver for BSP a s well as the , but a t lower dosage (20 mg daily) the stormaximal transport rate (L6)

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

351

age capacity is unaffected. Steroidal oral contraceptives also reduce the maximal transport rate, but hepatic storage capacity is unimpaired ; BSP is retained in the plasma largely as the conjugate (K12). The findings are similar to what is present in the last trimester of pregnancy (see Section 6.3). The precise mechanism of action of steroids on dye excretion is not defined. Biliary stasis can be demonstrated histologically after administration of norethandrolone (S3, S4, W2), methyltestosterone (W3), and other 17a-alkyl substituted testosterone-like steroids (S3). Atrophy of microvilli in the bile canaliculi has also been found (53) suggesting that some secretory dysfunction is probable, but the effect on microvilli is not a consistent finding (K12, S8). There is no effect on the conjugation of BSP (A16, B38) though the conjugation of bilirubin by liver slices is inhibited (H5). Another possible mechanism for the action of drugs is on protein-binding in plasma or in cells. Reduction of protein binding, for example, could reduce uptake of dye. The effect of steroids on dye excretion cannot be correlated with their anabolic, progestational, androgenic, or estrogenic properties (D3, G2, K14). There are conflicting findings on the relation between chemical structure and effect; these are possibly due t o species variation or to differences in dosage (G2, K14). Many investigators testing steroids have given doses much larger than those used in therapy. There has also been a suggestion (H21) that a genetic factor may be involved in the incidence of hepatic dysfunction with oral contraceptive therapy and in pregnancy. However, i t appears that an alkyl substitution on C17 is usually a requirement and that a keto group on C3 is more effective than a hydroxyl group (D3). The presence of a phenolic A ring also enhances activity ( G 2 ) . There is a suggestion that inactive steroids may potentiate the effect of others; thus progesterone which is without effect increases the activity of estrogens ( N l ) . 7.18. BILE SALTS Bile salt transport is important as a determinant of bile flow (W14, W15), and administration of bile salts causes a prompt increase in flow ((3, M17, 534, W8). Administration of high doses of sodium dehydrocholate (B49, C3, M16, M27), or of sodium taurocholate (03, W15) decreases the clearance of BSP from plasma. Both uptake and excretion rates appear to be affected. The plasma retention of indocyanine green (H28), uranin (H3), and bilirubin (C3) were also elevated on administration of sodium dehydrocholate. This inhibitory effect has been ascribed to competition for the uptake carrier (H28) and the carrier sites on the excretory membrane

352

PAULA JABLONSKI AND J . A. OWEN

(03). Bile salts can also displace BSP from plasma albumin and hepatic intracellular protein (A14), which may contribute to the reduced uptake of the dye. BSP can reduce the transport maximal rate for bile acids (03), an indication that the mechanism is competitive. On the other hand, administration of low doses of bile salts ( < 2 pmole/kg/min) increased the maximal excretory rate of BSP as well as increasing the bile flow; biliary concentration of BSP was unaltered (03). It has been suggested that the limiting factor in the maximal transfer of BSP from the cell to the bile canaliculus is the concentration of the dye in the canaliculus a t the excretory membrane (03). Since bile salts are probably secreted a t the same anatomical site as BSP (A14, C15) and produce an increased excretion of water a t that site, the local concentration of BSP can be maintained only by an increase in the rate of BSP excretion. Choleretics, e.g., secretin, which initiate a secretion of water a t some other site, do not effect the excretion of BSP, and biliary concentrations of the dye are reduced (03, W l l ) . 8.

Effect of Disease on Dye Uptake and Excretion

The number of published reports of clinical investigations in which dye tests have been used is very large. This section is intended as a summary of the effect of disease on dye uptake and excretion rather than as an exhaustive list of clinical studies involving dyes.

DISEASES 8.1. HEPATOBILIARY Acute Hepatitis I n acute infective hepatitis, there is almost invariably a decreased rate of removal of BSP (B37, D7, M21, N5, W31, Z5), rose bengal (K6, S41), and indocyanine green (L5, L8, H27). The BSP plasma disappearance curve shows a reduced slope in the first phase and a flat second phase (D7, 11). Abnormal dye handling is one of the most consistent biochemical findings in acute hepatitis; it occurs, as a rule, before bilirubin can be detected in the urine, before the serum bilirubin is elevated or the flocculation reactions become abnormal (N5). There are conflicting reports, however, of the relative sensitivities of different tests in the recovery phase. One group of patients tended to retain abnormal flocculation reactions and elevated l-minute serum bilirubin values longer than abnormal BSP values (N5). Other investigators have found decreased BSP removal rates in patients who had had infective hepatitis but who no longer showed other clinical or biochemical evidence of the disease (11, D7, W32). Abnormal BSP retention persists for longer than abnormal indocyanine green disposal (115). 8.1.1.

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

353

The BSP retention test has been used in studies on the factors in recovery from infective hepatitis. Thus the normal ability to handle BSP returns more quickly with cortisone therapy (H24). Ability to remove BSP improves more rapidly on a high protein diet (C13). I n infective hepatitis, a small percentage of patients relapse, with return of persistently abnormal biochemical values, but BSP retention values obtained early in convalescence are not helpful in predicting which of the patients will relapse (C13). I n acute hepatitis from toxic agents, such as carbon tetrachloride, there is likewise increased BSP retention (D7, D10). I n infectious mononucleosis, BSP retention is occasionally increased, but flocculation reactions (B18) are abnormal more frequently. 8.1.2. Chronic Hepatitis and Cirrhosis This is the group of patients in which dye metabolism has been most studied. There is variable impairment of dye handling in patients classified as postnecrotic cirrhosis (K10) , chronic active hepatitis (M4), or simply as cirrhosis (D7, 25). Increased BSP retention commonly also occurs in bilharzial hepatic fibrosis ( B l ) . A positive correlation has been observed between the degree of BSP retention in cirrhosis and the degree of fibrosis revealed by biopsy (M32, S12), but not between the plasma albumin level and the degree of fibrosis (S12). The mean disappearance rate of BSP is slightly reduced in alcoholics without other evidence of liver dysfunction (G14) and is significantly reduced when there is other evidence of liver dysfunction. It has been found ((211, (217) that BSP values in alcoholics correlate better with the degree of histological change in the liver than any of the other biochemical tests studied. Alcohol has an acute depressant effect on BSP excretion in experimental animals (see Section 7.10). 8.1.3. Portacaval Operations

Portacaval anastomosis decreases BSP disposal (B l, B42, L3). There was a fall in the hepatic blood flow in all 12 patients who underwent portacaval anastomoses for portal hypertension, but the BSP extraction ratio was increased (B42).Formation of Eck fistulae in dogs reduces BSP clearance (527). Portacaval transposition does not affect the hepatic blood flow or the extraction ratio (S35). 8.1.4. Biliary Obstruction I n extrahepatic biliary obstruction with jaundice, reduced dye clearance is usual (D7, M20). However, there is frequently increased BSP

354

PAULA JABLONSKI AND J. A. OWEN

retention when partial biliary obstruction is insufficient to cause jaundice (B54, C38). I n obstetric cholestasis, there is reduced ability of the liver to take up and excrete BSP (K2), and similar dysfunction occurs t o a lesser extent in pruritus of pregnancy (K2). I n biliary cirrhosis there is reduced dye clearance which tends to fluctuate much less than the serum bilirubin level (A5).

8.1.5. Carcinomatosis of the Liver This condition is often accompanied by increased BSP retention in the absence of jaundice (P5, 529). A similar finding was noted earlier with phenol tetrachlorophthalein (G20). I n patients with gastrointestinal carcinoma, there is a correlation between the BSP retention values and the presence of metastases in the liver (P5, T4),

Hereditary Hyperbilirubinemia I n Gilbert’s disease and in the Crigler-Najjar syndrome the disposal of BSP from plasma is usually normal (B20, MlO), but in the Rotor syndrome, there is BSP retention (R24). I n the Dubin-Johnson syndrome, BSP retention is usually increased (B61, D9, D11, S16). 8.1.6.

8.2. DYETESTS IN JAUNDICE Retention of dyes, such as BSP, is increased in acute and chronic hepatitis and in intra- and extrahepatic biliary obstruction. Gross retention (>60%) occurs more commonly in patients with hepatocellular jaundice than with obstructive jaundice (R5, S42), but there is considerable overlap between the two groups and retention tests are consequently of little value in distinguishing between hepatocellular and obstructive jaundice or in assessing hepatocellular function in the presence of biliary obstruction. Measurements of the fractional disappearance rate (a),the hepatic storage capacity (8), or the maximal biliary excretion ( Tm)have likewise not proved helpful in discriminating between the two conditions (S17). Attempts have been made to allow for the effect of biliary obstruction by the use of a correction factor based on the l-minute serum bilirubin concentration as a measure of the degree of obstruction (23). However BSP retention values corrected in this way were found to be less valuable in the differential diagnosis of jaundice than the results of the cephalin flocculation test or the thymol turbidity test (M21). Tura et al. (T11, T12), using indices derived from both phases of the plasma disappearance curve, have reported significant statistical differ-

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

355

ences between the two diagnostic groups, but there w,as again too much overlap for the procedure to be of practical use. On the other hand, a diagnostic chart, obtained by plotting values of the first exponential coefficient (a)against values of the second exponential coefficient ( p ), has been found to be of some value in distinguishing between several types of jaundiced patients (T7). Separate measurement of free and conjugated BSP has given statistically different results in hepatocellular and obstructive jaundice. I n one series of results (H14), the percentage of conjugated BSP in the plasma 4&60 minutes after an injection, was 77 (29-100)% in patients with extrahepatic biliary obstruction and 94 (76100) % in patients with hepatocellular jaundice. Although there was overlap, 8 out of 10 patients with biliary obstruction had values less than 88% whereas 8 out of 9 patients with hepatocellular jaundice had values equal to or greater than 88%. Similar trends have been noted by others (C6, S15), but all have found some overlap. Estimation of the amount of BSP, or its degree of conjugation, in the urine has not proved of practical value in the differential diagnosis of D7, H14). jaundice (C6, Studies with indocyanine green (C16) and rose bengal (C25, N8, N9) have similarly failed to reveal an index that would discriminate between hepatocellular and obstructive jaundice.

8.3. ACUTEABDOMINAL CONDITIONS BSP retention is usually increased in acute cholecystitis, but only occasionally is i t increased in acute peritonitis and other acute abdominal conditions (B60, W21). The mechanism in these conditions is either partial biliary obstruction or alteration in hepatic blood flow. The effect of morphine in causing BSP retention by constriction of the sphincter of Oddi is relevant (see Section 7.12). Burnett (B60) found increased BSP retention ranging from 5 to 2076 in 5 out of 6 patients with minor ailments given morphine, all of whom had shown normal BSP retention on the previous day. 8.4. SEPTICEMIA AND OTHERINFECTIONS BSP retention accompanied by other biochemical abnormalities but without jaundice is common in septicemia (N4). Findings in patients with intrahepatic infection are the same as in those with extrahepatic infection (B54). Reduced BSP handling is common also in malaria (F6, K13, M3). Injections of pyrogens such as typhoid-paratyphoid A and B vaccine (TAB) in man (M3) or bacterial endotoxin in rabbits (H15) cause increased BSP retention (see Section 6.5).

356

PAULA JABLONSKI AND J . A. OWEN

The reduced clearance of BSP in infective conditions may be an effect of pyrexia. However, the finding that a high proportion of patients with mild chronic ill-health secondary to chronic infection have reduced BSP disappearance rates (5%) and that direct warming of dogs to 42" has no effect (S40) suggests that toxins themselves may have an effect. 8.5. CIRCULATORY DISORDERS Increased BSP retention is commonly present in congestive heart failure (B30, E6, F4, 11, 531). Results correlate well with the clinical severity of the condition, and abnormal retention may be present in the absence of jaundice or any other biochemical abnormality. 8.6. INJURY There is reduced BSP clearance for 10-20 days after major operations (C35). Eight patients rendered paraplegic by spinal cord injuries of 2 week's duration had a mean BSP retention of 28 (0-40) % a t 60 minutes (C35) ; none were jaundiced. There was no correlation with the degree of catabolism assessed on nitrogen balance measurements (C35), The changes were considered to be due to a reduction in hepatic blood flow, but this does not necessarily cause a fall in BSP clearance (see Section 5.7). 8.7. METABOLIC AND ENDOCRINE DISORDERS 8.7.1. Obesity Obesity is associated with increased BSP retention when the standard test is used; 20 patients who were grossly (50-910/0) overweight all had increased BSP retention, viz. 15 (6-33) % (22). Although biochemical and biopsy evidence of liver dysfunction was found in 50% of the group, it is likely that the dose of BSP (5 mg/kg) was an important contributing factor because it was almost certainly higher relative to liver mass than in normal persons (see Section 5.1). 8.7.2. Nutritional Deficiencies An effect of diet on BSP excretion has been observed in rats given a protein-free diet for 2 days (C29). Reduction in BSP excretion was accompanied by a decrease in glutathione content of the liver and a fall in conjugating enzyme activity which was restored by supplementing the protein-free diet with cystine. Patients with kwashiorkor have increased BSP retention (K7, K8) ;

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

357

those whose BSP values remain abnormal a few days after treatment starts have a poor prognosis.

8.7.3. Diabetes Mellitus BSP retention and other evidence of hepatic dysfunction is not infrequent in this condition (B3, L10). Hemochromatosis Biochemical evidence of liver dysfunction is usually absent (S22), but a reduced BSP clearance is present in some patients (M17).

8.7.4.

Toxemia of Pregnancy BSP retention commonly increased in toxemia of pregnancy (B56, C19). I n a group of 15 patients, the BSP retention had a mean value (excluding a result of 2%) of 20 (10-37)%. The conjugated BSP in plasma a t 45 minutes comprised 48 (24-80) ”/. of the total (B56).

8.7.5.

8.7.6. Hypopituitarism I n 8 pstients with growth hormone deficiency the capacity of the liver parenchymal cells to excrete BSP into bile was reduced (Sll).

8.8. RENALDISEASE The effect of renal disease on dye removal rates has been little studied. BSP bound t o albumin escapes into the urine in patients with proteinuria and, where this is gross, the rate of disappearance of dye from plasma is slightly increased (V10). Further investigations in this field would seem worthwhile since liver disease, particularly severe liver disease, is often associated with renal dysfunction. 9.

Analytical Methods

9.1. BSP Methods for estimating BSP in plasma (or serum), bile or urine are based on the purple color which develops when solutions containing BSP are made alkaline. This change results from ionization, with internal molecular rearrangements, to form a quininoid structure. The pK of the reaction is about 8.5, and peak absorbance in protein-free solutions is a t 580-585 mp (S18). In the presence of plasma or serum, however, the peak is shifted to a higher wavelength (SlS). This is due to absorption of dye on protein molecules. The position of the absorption peak in protein solution varies with the pH of the solution (T3a). It has been reported

358

PAULA JABLONSKI AND J . A. OWEN

that the addition of heparin to blood containing BSP does not affect the absorption spectrum but administration of heparin to patients affects the spectrum in plasma samples obtained later (T3a). The main problems in determining BSP in biological fluids are background absorbance due to other pigments, such as bilirubin or hemoglobin, or to lipemic turbidity, and the effect of protein-binding on the absorption spectrum.

Colorimetry at One Wavelength A simple method which has widely been used is to dilute the plasma, or serum, in 0.1 M NaOH and measure the color produced (V5).To allow for background absorbance, a blank reading is obtained after diluting the material with 0.1 M HC1 to convert the BSP into the colorless form. An alternative procedure, which has been used (S28) to allow for background absorbance, consists of analyzing a sample taken before administration of BSP as a blank, but this will not allow for a variable amount of hemolysis in collecting blood specimens. It may, however, be necessary to analyze a preinjection sample if a dye test is being carried out a short time after a previous dye test to allow for any BSP remaining in the plasma. This is particularly important in the study of patients with marked BSP retention for, in such patients, measurable amounts of dye may remain in the plasma for days (G5, J l ) . Addition of NaOH to plasma usually gives a p H high enough to cause conversion of any hemoglobin in plasma to alkaline hematin which absorbs a t 580 m p so that the blank reading obtained after acidification is no longer a true blank. For this reason, Seligson et at. (S19)used an alkaline buffer giving a p H 10.5 to produce color without affecting any hemoglobin present and an acid buffer giving a final pH 7.0 which likewise does not affect hemoglobin for blank readings. p-Toluenesulfonate, which is competitively bound to albumin, was added to the buffers so that proteinbinding of the dye was minimized by a high p-toluenesu1fonate:BSP molar ratio. With Seligson’s method, plasma bilirubin in concentrations of up to 20 mg/100 ml does not interfere, but the method is unreliable if hemolysis is anything more than slight or if the sample is appreciably lipemic (H10). 9.1.1.

9.1.2. Colorimetry at Two or More Wavelengths Gaebler ( G l ) suggested measuring absorbance a t two wavelengths (580 mp and 620 q p ) to correct for background without the need of measuring a blank after acidification. However, Gaebler’s method was found to be invalid in the presence of hemoglobin (W29). Reinhold (R6) suggested reading a t three wavelengths (420 mp, 580 mp, and 660 mp).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

359

9.1.3. Protein Precipitation Methods Precipitation of plasma or serum proteins has been tried as a means of removing interference due to hemoglobin or turbidity. Various protein precipitants have been tried, such as acetone, ethanol, trichloroacetic acid ( I l ) , or ethyl phosphate (Z5), but recovery of added BSP has generally been found to be poor. How,ever, an acetone extraction procedure devised by Henry et al. (H10) gives good recoveries although it requires standard solutions made up in acetone because the latter depresses the absorbance of the dye.

9.1.4. Automatic Analysis The method of Seligson et al. (Sl8) has been adapted for use with the AutoAnalyser ( Vl) . Samples are first analyzed using an alkaline buffer, and blank values are then obtained by rerunning the sample using an acid buffer. An alternative method using an AutoAnalyser employs a dialyzer to eliminate hemoglobin and turbidity so that only a single run is required (G3). Use of a dialyzer also allows continuous sampling from patients, but a correction factor is required because of a n effect of whole blood on dialysis. 9.2. BSP CONJUGATES

The absorption spectrum of conjugated BSP in the visible region is closely similar to that of BSP, with an absorbance peak a t 580 mp in protein-free solution shifting to a higher wavelength on the addition of protein (S18). The ultraviolet absorption spectrum for the conjugates differs from that of parent dye in that there is an absorption peak a t 340 mp. This is the basis of measurement of the rate of formation of conjugate from free BSP in vitro (G10). The methods which have been used for the measurement of unconjugated BSP are in general applicable to the measurement of conjugated BSP. However biological fluids containing BSP conjugates nearly always also contain unconjugated BSP so that a separation process has to be applied before colorimetry. The various procedures which have been used to detect and measure conjugated BSP are listed in Table 9. Detection has been mainly by means of paper or thin-layer chromatography whereas measurement has involved these procedures and also column chromatography on alumina or molecular separation on Sephadex. However, one method (P7) is based on the differential solubility of BSP and its conjugates.

360

PAULA JABLONSKI AND J . A, OWEN

TABLE 9 METHODSFOR DETECTINQ AND DETERMININQ BSP METABOLITES Method

References

Paper chromatography

Thin-layer chromatography Alumina chromatography

Sephadex chromatography Solvent extraction

Studies in normals using these methods have shown that only traces of conjugated BSP appear in plasma after a standard dose (C5, P7, 515) ; in urine the percentage of conjugated BSP is relatively high (C6, D1, H14). I n bile, obtained by T-tube drainage from patients showing no evidence of liver dysfunction, BSP is mainly conjugated (M30).

9.3. INDOCYANINE GREEN Indocyanine green in plasma is measured colorirnetrically a t 800-850 mp. The absorbance peak of indocyanine green solution in water is a t 780 mp; the shift in plasma is due to albumin binding. Indocyanine green in aqueous solutions is not stable in the absence of protein (L5), and for this reason plasma or albumin solution should be added to bile (W9) or urine (C16) before measurements are carried out. Hemoglobin and bilirubin interfere minimally with the estimation (H27). With turbid samples, the procedure devised by Gaebler ( G l ) for use with BSP has been used. This involves measurement a t 680 and 800 mp. I n analyzing intestinal lymph, ethanol has been used to precipitate protein and dissolve fat. The dye was then measured a t 780 mp (H28).

9.4. ROSE BENGAL Rose bengal is colored a t the p H of plasma with an absorption peak a t 545 mp (530). However, the absorption spectrum is affected by absorption of dye on albumin so that, in diluted plasma, the absorption peak is a t 560 mp (S2). Hemolysis, which rose bengal tends itself to produce (S9),interferes.

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

361

Allowance for mild hemolysis can be made by subtracting a fraction of the absorbance a t 400 mp, a t which wavelength rose bengal has negligible absorption (S2).

9.5. PHENOL DIBROMOPHTHALEIN DISULFONATE This dye, like BSP, is purple in alkaline solution. Concentrations in plasma have been measured by an adaptation of Gaebler’s method (GI ) for BSP (H20). 10.

The Future of Dye Tests in Clinical Medicine

The wealth of reports in the literature describing the use of BSP and other dyes is testimony to the value of dye tests in clinical and physiological investigations. Comparative studies in clinical situations have shown that dye tests are almost invariably more sensitive to liver damage than are other biochemical tests and that there is good correlation between the results of dye tests and histological findings in patients who have minimal clinical evidence of liver dysfunction. There are a number of factors to be considered in using the result of a dye test as a measure of parenchymal liver cell function. The rate of disappearance of cholephilic dyes from plasma depends on the plasma flow to the liver and on the patency of the biliary system as well a s on the efficiency of parenchymal cells, and the influence of these factors must be considered in interpreting the results of dye tests. I n a patient or in an experimental animal, the plasma flow to the liver has to be considered in relation to the plasma volume, because only a portion of the plasma circulates through the liver in a single cycle. Thus, increase in plasma volume without change in hepatic plasma flow or in the extraction ratio will on its own slow down the removal of BSP from the circulation. This has been suggested (G4) a s a factor operating in late pregnancy, which is characterized by increased dye retention, and possibly i t acts also in fainting which reduces BSP uptake (R14).The tendency for the efficiency of parenchymal cells (as reflected in the extraction ratio) to vary inversely with the hepatic plasma flow has been described earlier. The patency of the biliary tract influences the regurgitation of BSP (conjugated and/or free) into the circulation. There is also the effect of reabsorption of dye which has reached the gut via a patent biliary tract, although this appears to be of smaller magnitude. As we have noted throughout this review, BSP has been used far more frequently than any of the other cholephilic dyes. However, during the last 10 years or so, indocyanine green has been used in liver function

362

PAULA JABLONSKI AND J . A. OWEN

studies in a number of investigations, and it is pertinent to consider whether this dye has any advantages over BSP. Early work suggested th at the plasma disappearance curve of indocyanine green was essentially linear, but more recent studies have shown that the slope of the curve becomes progressively less as time elapses. There were also early claims that extrahepatic disposal of indocyanine green is less than that of BSP. However, studies on hepatectomized dogs have shown that indocyanine green disappears from the circulation about as rapidly as BSP. Chemically, indocyanine green is much less stable than BSP; i t is also less soluble in aqueous media, which makes large doses cumbersome, and it is more expensive. However, indocyanine green is less irritating when injected into tissues, and severe systemic reactions have not been reported. In clinical medicine the real issue is whether tests using indocyanine green give more information about patients than tests using BSP, and there are insufficient data to answer this question. For no clearly defined reason, more investigators currently use BSP than indocyanine green. The majority of investigations in patients have involved measurement of dye retention only. The retention test, a t least with BSP, fails to take into account the shape of the plasma decay curve. Thus, in some patients the plasma dye concentration falls more slowly than normal during the initial phase but still reaches a normal level a t 45 minutes. The plasma disappearance curve of indocyanine green more closely approximates a single-phase exponential curve during the period of the test, and if it was considered desirable to collect only one sample it would be preferable to use indocyanine green. The optimal dose of dye to be administered in a retention test cannot be defined until further comparative studies have been done. It would be sensible meanwhile to use the standard dose (5 mg/kg) if only to obtain results comparable with the majority of those obtained during the past 20 years. As yet the advantages of kinetic parameters, such as the fractional disappearance rate, over dye retention values in the investigation of individual patients has not been clearly demonstrated. Such tests require a t least two blood samples taken during the first phase of the disappearance curve, and a further two samples if the second phase is to be defined also, but we would predict that the advantages of such tests will become evident as experience with the interpretation of multiple parameters accrues. There is also a problem in defining the second phase of the dye disappearance curve in some patients because of the low plasma dye concentrations a t this stage, but this could be overcome by the development of more sensitive analytical methods. The measurement of hepatic parameters such as the liver cell dye

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

363

storage capacity or the maximal biliary transfer capacity have mostly been carried out in studies using the infusion of dye a t two different rates. It is unlikely that procedures involving infusion will gain a place among tests used daily, but if it proves possible to derive by mathematical analysis hepatic parameters from single injections of dye, the use of such parameters would be more acceptable. However, the use of such parameters in clinical practice requires demonstration (S23) that they provide useful information about individual patients not otherwise obtainable. The relative merits of tests using radioactive dyes and tests using unlabeled dye have also to be evaluated. In many fields of clinical investigation the use of isotopes has proved to be advantageous to both the patient and the investigator. The use of radioactive dye, for example, with simultaneous external counting over the head and over the liver, offers the possibility of differentiating between removal of the dye from the blood by liver cells and it5 flow into the intestine via the biliary passages. The use of radioactive dye would make it possible to follow plasma disappearance curves for much longer than with chemical analysis. Further, injection of minute quantities of radioactive dye in place of the normal amounts of unlabeled dye would reduce the toxic hazards from dye although severe reactions to dye are very rare. We would venture to suggest that technical improvement in the use of isotopes in everyday medicine will eventually cause the standard dye test to be replaced by tests using radioactive substances. The place of dye tests in the investigation of the jaundiced patient is another aspect on which further studies are required. Biliary obstruction itself causes dye to be retained in plasma so that in a jaundiced patient it is not known to what extent a high dye retention value is due to biliary obstruction or to hepatocellular impairment. It is possible, however, that mathematical analysis of plasma dye disappearance curves may allow the two components to be separately assessed. With BSP, measurement of dye conjugates in plasma offer another means of separating the effects of biliary obstruction from those of hepatocellular failure, although results to date have not been promising in the assessment of individual patients.

REVIEW ARTICLES 1. Brauer, R. W.,Mechanisms of bile secretion. Gastroenterology 34, 1021-1028 (1958). 2. Brauer, R. W., Mechanisms of bile secretion. J . A m . Med. Assoc. 169, 1462-1477 ( 1959). 3. Carbone, J. V., and Grodsky, G. M., Metabolism of sulfobromophthalein sodium (BSP).Recent Advan. Gastroenterol. 38, 659-660 (1959).

364

PAULA JABLONSKI AND J . A. OWEN

4. Hanzon, V., Dye secretion and dye uptake by the liver. In “Liver Function” (R. W. Brauer, ed.), pp. 281-288. Waverly Press, Baltimore, Maryland, 1958. 6. Roman, W., and Hecker, R., The toxicity of oral contraceptives. A critical review of literature. Med. J . Australia ii, 682-688 (1968).

REFERENCES Al. Aaron, A. H., Beck, E. C., and Schneider, W. S., The phenol tetrachlorophthalein test for liver function. J . Am. Med. Assoc. 77, 1631-1634 (1921). A2. Abel, J. J., and Rowntree, L. G., On the pharmacological action of some phthaleins and their derivatives, with especial reference to their behavior as purgatives. J . Phamnacol. Exptl. Therap. 1, 231-264 (1909). A3. Acocella, G., and Billing, B. H., Effect of drugs on the hepatic transport of bilirubin. I n “Therapeutic Agents and the Liver’’ (N. McIntyre and S. Sherlock, eds.), pp. 1-8. Blackwell, Oxford, 1965. A4. Adler, A., Die Leber als Excretionsorgan. I n “Handbuch der normalen und pathologische Physiologie” (E. Bethe, ed.), Vol. 4, p. 769. Springer, Berlin, 1929. A5. Ahrem, E. H., Payne, M. A., Kunkel, H. G., Eisenmenger, W. J., and Blondheim, S. H., Primary bdiary cirrhosis. Medicine 29, 299-364 (1950). A6. Allan, J. S., and Tyler, E. T., Biochemical finding in long-term oral contraceptive usage. I. Liver function studies. Fertility Sterility 18, 112-123 (1967). A7. Allen, T. H., Symposium on nutrition: Prediction of blood volume and adiposity in man from body weight and cube of heights. Metab. Clin. Exptl. 6, 328-345 (1956). A8. Alliot, M., and Paraf, A., Lexploration fonctionelle du foie par 1’6preuve de la bromsulfonephthal6ine. Presse Med. 69, 501-503 (1951). A9. Althausen, T . L., Blumquist, B. E., and Whedon, E. F., The influence on carbohydrate metabolism of experimentally induced hepatic changes. IV. Block of the reticuloendothelial system. Am. J . Digest. Diseuses 2, 532-546 (1936). 810. Andrews, W. H. H., Excretory function of the liver. Lancet 11, 166-169 (1955). A l l . Andrews, W. H. H., Bile duct cells and bile secretion. I n “Liver Function” (R. W. Brauer, ed.), pp. 241-245. Waverly Press, Baltimore, Maryland, 1958. A12. Andrews, W. H. H., Hepatic circulation with special reference to the relationship of the bile duct and the lymphatic system. I n “The Biliary System” (W. Taylor, ed.), pp. 79-86. Blackwell, Oxford, 1965. A13. Andrews, W. H. H., Maegraith, B. G., and Richards, T. G., The effect upon bromsulphthalein extraction of the rate and distribution of blood flow in the perfused canine liver. J . Physwl. (London) 131, 669-677 (1958). 814. Andrews, W. H. H., and Richards, T. G., The activity of bile salts and certain detergents on the hepatic storage and protein-binding of sulphobromophthalein. Quart. J. Exptl. Physiol. 46, 275-283 (1960). A15. Andrews, W. H. H., and del Rio Lozano, I., Two postulated functions of hepatic endothelium. J . Physiol. (London)16S, 14P (1962). A16. Arias, I. M., Effects of a plant acid (icterogenin) and certain anabolic steroids on hepatic metabolism of bilirubin and sulfobromophthalein (BSP). Ann. N . Y . Acad. Sci. 104, 1014-1025 (1963). A17. Armstrong, P. B., Function in the developing liver. Federation Proc. 6, 70-71 (1947). B1. Badawi, H. S., Nomeir, A. M., and Zaher, R. A. S., Factors influencing bromsulphthalein retention and its extraction in bilharzial hepatic fibrosis. Gut 3,31&321 (1962).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

365

B2. Baker, K. J., Binding of BSP and indocyanine green by plasma w lipoproteins. Proe. SOC.Exptl. Bwl. Med. 122, 957-963 (1966). B3. Baker, K. J., and Bradley, S. E., Binding of BSP by plasma albumin. Its role in hepatic BSP extraction. J. Clin. Invest. 46, 281-287 (1966). B3a. Baker, L., Puestow, R. C., Kruger, S., and Last, J. H., Estimation of ascitic fluid volumes. J . Lab. Clin. Med. 39, 30-37 (1952). B4. Balkisson, B., Sheltun, T. G., and Spellman, M. W., The rapid determination of plasma volume with simultaneous estimation of excretory liver function in patients by the rose bengd dye (tetraiodo tetrabromo fluorescein) method. Ann. S w g . 147,505-514 (1958). B5. Banasgah, E. F., Stekiel, W. J., Grace, R. A., and Smith, J. S., Estimation of hepatic blood flow using a single injection dye clearance method. Am. J . Physiol. 198,877-880 (1960). B6. Barber-Riley, G., Competitive effect of probenecid on bromsulphthalein uptake in rat and dog liver slices. J. Physiol. (London) 163, 22P (1960). B7. Barber-Riley, G., Proportion of bromsulphthalein protein-bound with incubated rat liver slices. Nature 194, 184-185 (1962). B8. Barber-Riley, G., Measurement of the capacity of the biliary tree in rats. Am. J. Physiol. 206, 1122-1126 (1963). B9. Barber-Riley, G., The rat biliary tree during short periods of obstruction of the common duct. Am. J. Physiol. 206, 1127-1131 (1963). B10. Barber-Riley, G., Measurement of the capacity of the biliary tree. In “The Biliary System” (W. Taylor, ed.), pp. 89-95. Blackwell, Oxford, 1965. B l l . Barber-Riley, G., Biliary capacity in rats following ethionine ingestion. Am. J. Physiol. 214, 133-138 (1968). B12. Barber-Riley, G., Goetzee, A. E., Richards, T. G., and Thomson, J. Y., Transfer of bromsulphthalein from the plasma to the bile in man. Clin. Sci. 20, 149-159 (1961). B13. Barnes, M. M., James, S. P., and Wood, P. B., Formation of mercapturic acids. 1. Formation of mercapturic acid and glutathione levels in tissues. Biochem. J . 71, 680-690 (1959). B14. Baumrtnn, E., and Preusse, C., Uber Bromphenyl mercaptur Saure. Ber. Deut. C h . Ges. 12, 806-810 (1879). B14a. Becker, D. V., and Prudden, J. F., The metabolism of laWabelled thyroxine, tri-iodothyronine and di-iodotyrosine by an isolated perfused rabbit liver. Endocrinology 64, 136-148 (1959). B15. Begon, F., and Benhamou, J. P., Etude cinbtique du transfert hbpatocytaire de la bromesulfonephtd6ine. Clin. Chim. A d a 18, 227-239 (1967). B16. Bengmark, S., Almersjii, O., Engevix, L., Hafstrom, L. O., and Nilsson, N. J., Changes in indocyanine elimination after total hepatectomy and liver transplantation. Recent Advan. Gastroenterol. 3, 404-406 (1966). B17. Bennhold, H., Ott, H., and Wiech, M., Uber den Bindugsunterschied lebergangiger und nierenglingiger Substansen an die Serum Eiweiskiirper. D a d . Med. Woehsehs. 76, 11-15 (1950). B18. Berk, J. E., Shay, H., Ritter, J. A., and Siplet, H., Infectious mononucleosis and infectious hepatitis. Studies bearing on certain resemblances and differences. Gastroenterobgy 11, 658-671 (1948). B19. Berthelot, P., and Billing, B. H., Effect of bunamiodyl on hepatic uptake of sulfobromophthaleinin the rat. Am. J. Physwl. 211, 395-399 (1966). B20. Billing, B. H., Defects in bile pigment metabolism causing jaundice. Actu Med. Okayama 16, 185-197 (1961).

366

PAULA JABLONSKI AND J. A. OWEN

B21. Billing, B. H., Maggiore, Q., and Cartter, M. A., Hepatic transport of birubin. Ann. N . Y . Acad. Sci. 111, 319-324 (1963). B21a. Billing, B. H., Maggiore, Q. S., and Goulii, G., The action of cholecystographic contrast media and novobiocin on the hepatic transport of bilirubin. In “The Biliary System” (W. Taylor, ed.), pp. 215-223. Blackwell, Oxford, 1965. B22. Biorck, G., Gardell, S., Carlberger, G., and Meurman, L., Excretion of rose bengal in bile. Nature 186, 847-848 (1960). B23. Birkenhager, W. H., De Verdwijningskurne van Bromsulfoleine ter Bepaling van de Leverfunctie. Ned. Tijdschr. Geneesk. 100, 1550 (1956), cited by Nosslin and Morgan (N12). B24. Birkett, D. J., Done, J., Neale, F. O., and Posen, S., Serum alkaline phosphatase in pregnancy. An immunological study. Brit. Med. J. I, 1210-1216 (1966). B25. Bizard, G., Enzyme inhibitors and biliary secretion. I n “The Biliary System” (W. Taylor, ed.), pp. 315-321. Blackwell, Oxford, 1965. B26. Bizard, G., Vanlerenberghe, J., Robelet, A., Milbled, G., and Gverrin, G., Action of 2,4-dinitrophenol on choleresisin the rat. J . Physwl. (Paris) 62, 21-22 (1960). B27. Blondheim, S. H., Relationship between albumin concentration of serum and its dye-binding capacity. J . Lab. Clin. Med. 46, 740-747 (1955). B28. Blondheim, S. H., Effect of probenecid on excretion of bromsulphthalein. J. A p p . Physiol. 7, 529-532 (1955). B29. Bloom, W., and Rosenau, W. H., A simple method for the determination of phenoltetrachlorphthalein in blood serum. J . Am. Med. Assoc. 82, 547 (1924). B30. Blumberg, N., and Schloss, E., The effect of circulatory failure on the bromsulphthalein test in liver disease. Am. J. Med. Sci. 213, 470474 (1947). B31. Bogoch, A., Casselman, W. G. B., Kaplan, A., and Bockus, H., Studies of hepatic function in diabetes mellitus, portal cirrhosis and other liver diseases. Am. J . Med. 18, 354-384 (1955). B32. Bollman, J. L., Bile pigments in blood, lymph and bile in the course of experimental stasis. In “Liver Function” (R. W. Brauer, ed.), pp. 291-297. Waverly Press, Baltimore, Maryland, 1959. B33. Bolt, R. J., Dillon, S., and Pollard, H. M., Interference with bilirubin excretion by a gall bladder dye (Bunamiodyl). New Eng. J . Med. 266, 1043-1045 (1961). B34. Booth, J., Boyland, E., and Si,P., Metabolism of polycyclic compounds. 15. The conversion of naphthalene to the glutathione derivative by rat-liver slice. Biochem. J . 74, 117-122 (1960). B35. Booth, J., Boyland, E., and Sims, P., An enzyme from rat liver catalysing conjugations with glutathione. Biochem. J . 79, 516-524 (1961). B36. Booth, J., Boyland, E., Sato, T., and Sims, P., Metabolism of polycyclic compounds. 17. Reaction of 1:2-dihydronaphthalene and 1:2-epoxy-l:2:3:4 tetrahydronaphthalene with glutathione catalysed by tissue preparations. Bwchem. J. 77, 182-186 (1960). B37. Borg, A., and Julsrud, A. C., Leverfunksjonsprerverved akutt hepatitt. Nord. Med. 67, 64-68 (1957). B38. Boyland, E., and Grover, P. L., The relationship between hepatic glutathione conjugation and BSP excretion and the effect of therapeutic agents. Clin. Chim. A d a 16, 205-213 (1967). B39. Boyland, E., Manson, D., and Nevy, R., The biochemistry of aromatic amine. 9. Mercapturic acids as metabolites of aniline and 2-naphthylamine. Biochem. J . 86, 263-270 (1963). B40. Bradley, S. E., Variation in hepatic blood flow in man during health and disease. New Eng. J . Med. 240, 456461; also correction 240, 691 (1947).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHlLIC DYES

367

B41. Bradley, S. E., and Conan, N. J., Estimated hepatic blood flow and bromsulphthalein in normal man during pyrogenic reaction. J. Clin. Invest. 26,1175-1192 (1947). B42. Bradley, S. E., Ingelfinger, F. J., Bradley, G. P., and Curry, J. J., The estimation of hepatic blood flow in man. J . Clin. Invest. 24, 890-898 (1945). B43. Bradley, S. E., MacPherson, A. I., Gammeltofti, A., and Blakemore, A. H., Effect of portacaval anastomosis on hepatic oxygen extraction and estimated hepatic blood flow in patients with cirrhosis. J. Clin. Invest. 30, 630-634 (1951). B44. Brady, L. W., Cooper, D. Y . , Colodzin, M., King, E. R., and Williams, R., Blood volume studies in normal humans. Surg. Gynecol.Obstet. 97,25-32 (1953). B45. Brauer, R. W., Julian, L. M., and Krebs, J. S.,Maturation of liver function in the chick embryo as explored with 3~S-sulfobromophthalein;vascular factors, biliary secretion and conjugation. Ann. N . Y . Acad. Sci. 111, 136-156 (1963). B46. Brauer, R. W., and Krebs, J. S., La conjugaision de la BSP. Etude recentes sur la biochimie de ce phenombne et Bur cons6quencea pour 1’6liination de la BSP par le foie. Rev. Intern. Hepatol. 12, 379406 (1962). B47. Brauer, R. W., Leong, G. F., and Holloway, R. J., Mechanics of bile secretion; effect of perfusion pressure and temperature on bile flow and bile secretion pressure. Am. J . Physiol. 177, 103-112 (1954). B48. Brauer, R. W., and Pessotti, R. L., The removal of bromsulphthalein from blood plasma by the liver of the rat. J . Pharmucol. Exptl. Tkerap. 97, 358-370 (1949). B49. Brauer, R. W., and Pessotti, R. L., Hepatic uptake and biliary excretion of bromsulphthalein in the dog. Am. J . Physiol. 162,565-574 (1950). B50. Brauer, R. W., Pessotti, R. L., and Krebs, J. S.,Distribution and excretion of %-labelled bromsulphthalein, administered to dogs by continuous infusion. J . Clin. Invest. 34, 3 5 4 3 (1955). B51. Brauer, R. W., Shill, 0. S., and Krebs, J. S., Studies concerning functional differences between liver regions supplied by hepatic artery and portal vein. J . Clin. Invest. 38, 2202-2210 (1959). €352. Bray, H. G., Franklin, T. J., and James, S. P., The formation of mercapturic acids. 2. The possible role of glutathionase. Biochem. J. 71, 690-695 (1959). B53. Bray, H. G., James, S. P., and Thorpe, W., The metabolism of 2:4, 2:5 and 3:4 dichloronitrobenzenein the rabbit. Biochem. J . 66, 483490 (1957). B54. Brem, J. M., The use of hepatic function tests in the diagnosis of amebic abscess of the liver. Am. J . Med. Sn’. 229, 135-137 (1955). B55. Brette, R., Lambert, R., and Truckot, R., Biliary excretion of antibiotics. In “The Biliary System” (W. Taylor, ed.), pp. 148428. Blackwell, Oxford, 1965. B56. Brewer, T. H., and Hjelte, V., Bromsulfophthaleinin toxaemia of late pregnancy. Am. J . Obstet. Gynecol. 92, 1114-1116 (1965). B57. Broholt, J., Standardization of the bromsulphthaIein test with body composition. Scund. J. Clin. Lab. Invest. 19,67-71 (1967). B58. Brokaw, R., and Penrod, K., Bromsulphthalein removal mtes during hyperthermia in the dog. Am. J. Physiol. 168, 365-368 (1945). B59. Burke, J. F., and Whitehouse, M. W., Concerning the differences in uncoupling activity of isomeric dinitrophenols. Bwchem. Pharmacol. 16,209-211 (1967). B60. Burnett, W., BSP retention test in acute cholecystitis and in some other acute intra-abdominal conditions. Lancet 11, 488490 (1954). B61. Butt, H. R., Foulk, W. T., Baggenstoss, A. H., and Dickson, E. R., Hyperbiliiubinaemias of unknown cause: With a preliminary report on a large family with Dubin-Johnson syndrome. I n “The Biliary System” (W. Taylor, ed.), pp. 619-636. Blackwell, Oxford, 1965.

368

PAULA JABLONSKI AND J. A. OWEN

C1. Calloway, N. O., and Merrill, R. S., The ageing adult liver. I. Bromsulphthalein and bilirubin clearance. J . Am. Geriat. SOC.13, 594-598 (1965). C2. Cantarow, A., and Wirts, C. W., Studies of the excretion of bromsulphthalein in the bile. Proc. SOC.Exptl. Bwl. Med. 47, 252 (1941). C3. Cantarow, A., and Wirts, C. W., The effect of dog’s bile, certain bile acids and indian ink on biliiubinemis and the excretion of bromsulphthalein. Am. J. Digest. Diseases 10, 261-266 (1943). C4. Cantarow, A., Wirts, C. W., Snape, W. J., and Miller, L. L., Excretion of bilirubin and bromsulphthalein in bile. Am. J. Physwl. 164, 211-219 (1948). C5. Carbone, J. V., Grodsky, G. M., and Fanska, R., Chemical and clinical studies of bromsulphthalein metabolites. J. CZin. Invest. 38, 994 (1959). C6. Carbone, J. V., Grodsky, G. M., and Hjelte, V., Effect of hepatic dysfunction on circulating levels of sulfobromophthalein (BSP) and its metabolites. J . Clin. Invest. 38, 1989-1995 (1959). C7. Caroli, J., and Charbonnier, A., Diagnostic aphoristique des ictbres par rCtention &ape biologique. Rev. Med-Chif. Ma1 Foie 39, 5-24 (1954). C8. Caroli, J., and Tanasoglu, Y., Le temps d’apparition de la bromosulfonepthalCne dans le bile. Nouveau test pour le diagnostic des icarea incompletspar retention et blocages anicgrique de 1s voie principale. S w i n e Hop. Paris 29, 591-600 (1953). C9. Castenfors, H., Eliasch, H., and Hultman, E., Splanchnic blood flow and oxygen consumption estimated in man by the bromsulphthalein method with special reference to infiuence of peripheral dye level. Smnd. J. Clin. Invest. 12, 158-171 (1960). C10. Castenfors, H., and Hultman, E., Elimination of a single high dose of bromsulphthalein (BSP) used as a test of liver function. Scad. J. Clin. Lab. Invest. 14 (Suppl. 64), 45 (1962). C11. Cates, H. B., Relation of liver function to cirrhosis of the liver and to alcoholism. Comparison of results of liver function test with degree of organic change in cirrhosis of liver and with results of such tests in persons with alcoholism but without cirrhosis of liver. A . M . A . Arch. Internal Med. 67, 383-398 (1941). C12. Caujolle, F., and Meynier, D., L’6tape biliare des traversees chimique dans l’organisme. Rev. Intern. Hepatol. 6, 2944 (1946). C13. Chalmers, T. G., et al., The treatment of acute infectious hepatitis. Controlled studies of the effects of diet, rest and physical reconditioning on the acute course of the disease and on the incidence of relapses and residued abnormalities. J. Clin. Invest. 34, 1167-1235 (1955). C14. Chambers, W. N., and Moister, F. C., Bromsulphthalein reaction. Am. J . Med. 6 , 308-310 (1948). C15. Chenderovitch, J., Stop-flow analysis of bile secretion. Am. J. Physwl. 214, 86-93 (1968). C16. Cherrick, G. R., Stein, S. W., Leevy, C. M., and Davidson, C. S., Indocyanine green: Observations on its physical properties, plasma decay and hepatic extraction. J . Clin. Invest. 39,592400 (1960). C17. Childs, A. W., Kivel, R. M., and Lieberman, A., Effect of ethyl alcohol on hepatic circulation of sulfobromophthalein, clearance and hepatic glutamic-oxalocetic transaminase production in man. Gastroenterology 46, 176-181 (1963). C18. Chinard, F. P., Time relationships of certain renal processes. Federation Proc. 11, 197 (1952). C19. Christhilf, S. M., Bonsnes, R. W., and Barad, B., Liver function during pregnancy and the puerperium as measured by the cephalin-cholesterol flocculation, the

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

369

thymolturbidity and the bromsulphthalein test. Am. J . Obstet. Gynecol. 69, 1100-1104 (1950). C20. Chrzonszczewsky, N., Zur Anatomie und Physiologie der Leber. Arch. Pathol. A d . Physiol. 36,153-161 (1866). C21. Clarkson, M. J., and Richards, T. G., Steady state plasma clearance of bromsulphthalein and indocyanine green measured by single injection. Res. Vet. Sci. 8, 454-462 (1967). C22. Clodi, H., and Schnak, H., Uber die Beeinfiussung der Bromosulfophthalein Ausscheidung durch die Galle bei Ratten nach langerdauerner Chlorpromazinangabe Vein. Klin. Wochschr. 73, 801-807, (1960). C23. Cohen, E. S., Althausen, T. L., and Giansiracusa, J. E., Studies on bromsulphthalein excretion. iv. Variation in the mechanism of hepatic excretory function when challenged with different dyes. Gastroenterology 30, 232-243 (1956). C24. Cohen, E. S., Giansiracusa, J. E., and Althausen, T. L., Studies on bromsulphthalien excretion; simultaneous perforniance of bromsulphthalien and rose bengal excretion tests in individuals with normal hepatic function. Gastroenterology 36, 237-242 (1953). C25. Cohen, E. S., Giansiracusa, J. E., Strait, L.A,, Althausen, T. L., and Karg, S., Studies on bromsulphthalien excretion, 11. Method for simultaneous bromsulphthalien and rose bengal excretion tests. Gastroenterology 62, 232-236 (1953). C26. Cohn, C., Livine, R., and Streicher, D., Rate of removal of intravenously injected bromsulphthalein by liver and extrahepatic tissues of the dog. Am. J. Physwl. 160, 299-303 (1947). C27. Cohn, E. M., Zagerman, J., Sklaroff, D. M., and Tumen, H. G., The use of radioactive rose bengal as a test in the differential diagnosis of jaundice. Am. J . Gastroenterology 38, 621-628 (1957). C28. Combes, B., Biliary excretion of sulfobromophthalein sodium in the rat as a conjugate of glycine and glutamic acid. J. Clin. Invest. 38, 1426-1433 (1959). C29. Combes, B., The importance of conjugation with glutathione on sulfobromophthalein transport from blood to bile. J . Clin. Invest. 44,1214 (1962). C30. Combes, B., Shibata, H., Adams, R., Mitchell, R. D., and Trammell, V., Alterations in sulfobromophthalein sodium removal mechanism from blood during normal pregnancy. J. Clin. Invest. 43, 1431-1442 (1963). C31. Combes, B., and Stakelum, G. S., Conjugation of bromsulphophthalein with glutathione in thioether linkage in the rat. J . Clin.Invest. 39, 1214-1222 (1960). C32. Combes, B., and Stakelum, G. S., A liver enzyme that conjugates sulfobromophthalein sodium with glutathione. J. Clin. Invest. 41, 981-988 (1961). C33. Combes, B., and Stakelum, G. S., Maturation of sulfobromophthalein sodiumglutathione conjugating system in rat livers. J. Clin. Invest. 41, 750-757 (1962). C34. Cook, D. L., Lawler, C. A., and Green, D. M., Studies on the effect of hydrocholeretic agents on hepatic excretory mechanisms in dogs. J. Pharmacol. Exptl. Therap. 293-299 (1954). C35. Cooper, I. S., Ryneanon, E. H., and McCarty, C. S., Impairment of liver function subsequent to injury to the spinal cord. J . Lab. Clin. Med. 38, 689-692 (1951). C36. Cornelius, C. E., Ben-Ezzer, J., and Arias, I. M., Binding of sulfobromophthalein sodium (BSP) and other organic anions by isolated hepatic cell plasma membranes in vitro. Proc. Soc. Exptl. Bwl. Med. 134, 665-667 (1967). '237. Culbertson, J. W., Wilkins, R. W., Ingelfinger, F. J., and Bradley, S. E., The effect of the upright position upon hepatic blood flow in normotensive and hypertensive subjects. J. Clin. Invest. 30, 305-311 (1951).

370

PAULA JABLONSKI AND J. A. OWEN

C38. Culver, P. J., McDermott, W. V., and Jones, C. M., Diagnostic value of selective interference with certain excretory processess of the liver. Gastroenterology 33, 169-171 (1957). D1. De Fraiture, W. H., Heemstra, H., Vegeter, J. J. M., and Mandema, E., Chromatographic separation of different bromsulphthalein metabolites in urine and bile. Acta Med. S c a d . 166, 153-156 (1959). D2. Delcourt, A., Delcourt, R., and Domb, A., Etude de la clearance de la bromsulfophtal6ine chea l’homme normal et dans les affections h6patiques. Acta Clin. Belg. 10, 70-71 (1952). D3. De Lorimer, A. A., Gordan, G. S., Lowe, R. C., and Carbone, J. V., Methyltestosterone, related steroids and liver function. Arch. Internal Med. 116,289-294 (1965). D4. Delprat, G. D., Studies in liver function. Rose bengal elimination from the blood as influenced by liver injury. A . M . A . Arch. Internal Med. 32, 401410 (1923). D5. Delprat, G. D., Epstein, N. N., and Kerr, W. J., A new liver function test. The elimination of rose bengal when injected into the circulation of human subjects. A . M . A . Arch. Internal Med. 34, 537-541 (1924). D6. Despopoulos, A., and Gross, C. W., Hepatic conjugation and excretion of phenolsulphonphthaleins. Federation Proc. 19, 184 (1960). D7. Deutsch, E., A fractional bromsulphthalein test to determine liver damage in the non-jaundiced patient. New Eng. J . Med. 226, 171-175 (1941). D8. Dickes, R., Schenker, V., and Deutsch, C., Serial liver function and blood studies in patients receiving chlorpromazine. New EngZ. J . Med. 266, 1-7 (1957). D9. Dollinger, R., and Branborg, L. L., Late elevation in serum bromsulphthalein in Dubin-Johnson syndrome. A comparative case study. Am. J . Digest. Diseases 12, 413417 (1967). DlO. Drill, V. A., and Ivy, A. O., Comparative value of bromsulphthalein, serum phosphatase, prothrombin time and intravenous galactose tolerance tests in detecting hepatic damage produced by carbon tetrachloride. J . Clin. Invest. 23, 209-216 (1944). D11. Dubin, I. N., and Johnson, F. A, Chronic idiopathic jaundice with unidentified pigment in liver cells. A new clinic-pathologic entity with a report of 12 cases. Medicine 33, 155-197 (1954). El. Eckhardt, E. T., and Plaa, G. L., The effect of phenothiazine derivatives on the disappearance of sulphobromophthalein from mouse plasma. J . Phurmaw2. Ezptl. Therap. 138, 387-391 (1962). E2. Edwards, S., and Westerfield, W. W., Blood and liver glutathione during protein deprivation. Proc. SOC.Exptl. Bwl. Med. 79, 1475-1503 (1955). E3. Eisalo, A., Jiirvinen, P. A., and Luukkainen, T., Hepatic impairment during intake of contraceptive pills. Clinical tests with post menopausal women. Brit. Med. J . 11, 426-427 (1964). E4. Englert, E., Albumin binding and hepatic uptake of anionic dyes. Clin. Res. 7 , 289 (1959). E5. Englert, E., Jr., Burrows, J. A., and Ingelfinger, F. J., Differential analysis of the stages of hepatic excretory function with gamma emitting isotopes. J . Lab. Clin. Med. 66, 181-192 (1953). E6. Evans, J. M., Zimmerman, H. J., Wilmer, J. G., Thomas, L. J., and Etheridge, C. B., Altered liver function of chronic congestive heart failure. Am. J . Med. 13, 704-712 (1952). E7. Evans, R. L., Liver function and bromsulphophthalein disappearance. Science 117, 718-719 (1953).

BBOMSULFOPHTHALEIN AND OTHEB CHOLEPHILIC DYES

371

F1. Fauvert, R. E., The concept of hepatic clearance. Gastroenterology 37, 603416 (1955). F2. Fawcett, D. W., Observations on the cytology and electron microscopy of hepatic cells. J . Natl. Cancer Inst. 16, 1475-1503 (1955). F3. Fee, D. A., and Fedorck, S. O., Clinical value of liver photoscanning. New. Engl. J. Med. 262, 123-127 (1960). F4. Felder, L., Mund, A., and Parker, J. G., Liver function tests in chronic congestive heart failure. Circulation 2, 286-297 (1950). F5. Fox, I. J., Broker, L. G. S., Heseltine, D. W., and Wood, E. H., A tricarbocyanine dye for continuous recording of dilution curves in whole blood independent of variation in oxygen saturation. Proc. StuffMeetings Mayo Clinic 32, 47&486 (1957). F6. Fredricks, M. G., and Hoffbauer, F. W., A study of hepatic function in malaria. J. Am. Med. Assoc. 128, 495-498 (1945). G1. Gaebler, 0. H., The determination of bromsulphthalein in normal, turbid, haemolized or icteric serum. Am. J . Clin. Pathol. 16, 452455 (1945). G2. Gallagher, T. F., Mueller, M. N., and Kappas, A., Studies on the mechanism and structural specifity of the oestrogen effect on BSP metabolism. Trans. Assoc. Am. Physicians 78, 187-195 (1965). G3. Galli, A., Jeanmaire, J., Choisy, H., and Schuller, E., Etude de 1’6preuve B la bromsulphal6ine par l’analyse en contenu. Ann. Biol. Clin. (Paris) 19, 759-767 (1961). G4. Gambino, S. R., Liver disease in pregnancy. Med. J . Australia I, 373 (1968). G5. Giges, B., Prolonged retention of bromsulphthalein in patients with regurgitation jaundice. J. Lab. Clin. Med. 38, 210-212 (1951). G6. Gindler, E. M., Slide rule for time conversion of bromsulphthalein retention studies. Clin. Chem. 12, 506-508 (1966). G7. Goetzee, A. E., Richards, T. G., and Tindall, V. R., Experimental changes in liver function induced by probenecid. Clin. Sci. 19, 63-78 (1960). G8. Goldfarb, S., Singer, E. J., and Popper, H., Biliary ductules and bile secretion. J . Lab. Clin. Med. 62, 608-615 (1963). G9. Goldstein, J., and Combes, B., The effect of steroids on the activity of the enzyme that catalyses bromsulphthalein conjugation with glutathione. J. Lab. Clin. Med. 67, 830-835 (1966). G10. Goldstein, J., and Combes, B., Spectrophotometric assay of the liver enzyme that catalyses bromsulphthalein conjugation with glutathione. J. Lab. Clin. Med. 67, 863-872 (1966). G11. Goldstein, J., Schenker, S., and Combes, B., Sulfobromophthalein sodium conjugation and excretion in neonatal guinea pigs. Am. J. Physiol. 208, 573-577 (1965). G12. Gomalez-Oddone, M. V., Bilirubin, bromsulphthalein, bile acids, alkaline phosphatase and cholesterol of thoracic duct lymph in experimental regurgitation jaundice. Proc. SOC.Exptl. Biol. Med. 63, 144-147 (1946). G13. Goodman, R. D., Bromsulphthalein clearance. A quantitative clinical test of liver function. J . Lab. Clin. Med. 40, 531-536 (1952). G14. Goodman, R. D., and Kingsley, G. R., Sulfobromophthalein clearance test. J . Am. Med. Assoc. 163, 462-466 (1953). (215. Goresky, C. A., Initial distribution and rate of uptake of BSP in liver. Am. J . Physiol. 207, 13-26 (1964). G16. Goresky, C. A., The hepatic uptake and secretion of sulfobromophthalein and bilirubin. J . Can. Med. Assoc. 92, 851-857 (1965).

372

PAULA JABLONSKI AND J. A. OWEN

G17. Grafflin, A. L., Excretion of fluorescein by the liver under normal and abnormal conditions, observed in nitro with the fluorescence microscope. Am. J . Anat. 81, 63-116 (1947). 618. Graham, E. A., and Cole, W. H., Roentgenologic examination of the gallbladder. J . Am. Med. Assoc. 82, 613-614 (1924). G19. Graham, E. A., Cole, W. H., Moore, S., and Copher, G. H., Cholecystographyoral administration of sodium tetraiodo-phenolphthalein.J . A m , Med. Assoc. 86, 935-955 (1925). G20. Greene, C. H., McVicar, C. S., Walters, W., and Rowntree, L. G., Diseases of the liver. IV. Functional tests in cases of carcinoma of the liver and biliary tract. A . M . A . Arch. Internal Med. 36, 542-560 (1925). G21. Gregersen, M. I., and Rawson, R. A., The disappearance of T-1824 and structurally related dyes from the blood stream. Am. J. Physiol. 138, 695-707 (1943). G22. Gregersen, M. I., and Rawson, R. A., Blood volume. Physiol. Rev. 99, 307-342 (1962). G23. Grodsky, G. M., Carbone, J. V., and Fanska, R., Identification of metabolites of sulfobromophthalein. J . Clin. Invest. 38, 1981-1988 (1959). G24. Grodsky, G. M., Carbone, J. V., and Fanska, R., Biosynthesis of bromsulphophthalein mercaptide with glutathione. Proc. SOC.Ezptl. Biol. Med. 106, 526-530 (1961). G25. Grover, P. L., and Sims, P., Conjugation with glutathiondistribution of glutathione s-aryl transferase in vertebrates. Bwchem. J. 90, 603-606 (1964). G26. Gubar, A. V., Uch. Zap., Mosk. Med. Inst. 12, 249 (1957), cited in Chem. Abstr. 63, 22484 (1958). H1. Habif, D. V., Papper, E. M., and Fitepatrick, H. F., The renal and hepatic blood flow, glomerular filtration rate and urinary output of electrolytes during cyclopropane, ether and thiopental anaesthesia, and the immediate poshperative period. Surgery 30, 241-255 (1951). H2. Hammaker, L., and Schmid, R., Interference with bile pigment uptake in the liver by flavaspidic acid. Gustroenlerology 63, 31-37 (1967). H3. Hanzon, V., Liver cell secretion under normal and pathological conditions studies by fluorescence microscopyon living rats. A d a Physiol. Scand. 28, Supplied. 101 (1952). H4. Hargreaves, T.,and Lathe, G. H., Inhibitory aspects of bile secretion. Nature 200, 1172-1176 (1963). H5. Hargreaves, T., and Lathe, G. H., Drugs affecting biliary secretion. In “Therapeutic Agents and the Liver” (N. McIntyre and S. Sherlock, eds.), pp. 9-16. Blackwell, Oxford, 1965. H6. Heaney, R. P., and Whedon, G. D., Impairment of hepatic bromsulphthalein clearance by two 17-substituted testosterones. J . Lab. Clin. Med. 62, 169-175 (1955). H7. Heikel, T., Knight, B. C., Rimington, C., Ritchie, H. D., and Williams, A. J., Studies on biiary excretion in the rabbit. Proc. Roy. SOC.(London) B163, 47-79 (1961). H7a. Heinman, H., Smythe, C., and McMarks, P. A., The effectof hemorrhage and the pyrogen reaction on hepatic circulation in dogs. J. CZin. Invest. 31,637-645 (1952). H8. Helm, J. D., and Machella, T. E., The significance of dosage and time factors on the value of the bromsulphthalein test for liver function. Am. J. Digest. Diseases 9, 141-143 (1942). H9. Henning, N., and Demling, L., Blood flow in the liver. In “Progress in Liver

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

373

Disease” (H. Popper and F. Schaffner, eds.), Vol. I, pp. 162-173. Grune & Statton, New York, 1961. H10. Henry, R. J., Chiamori, N., and Ware, A. G., Determination of bromsulphthalein in serum by means of acetone precipitate of protein. Am. J . Clin. Pathol. 32, 201-203 (1959). H11. Herlitz, C. W., Rosenthal and White’s Leberfunktionsprobe (Bromsulphaleinprobe) bei Kindern unter einem Jahr und besonders bei Ikterus neonatorum. A d a Paediat. 6 , 214-224 (1926). H12. Herlitz, C. W., Contributions to knowledge of function of the liver and reticule endothelial system particularly in infectious disease of children. Acta Paediat. 12, Suppl. 1-80 (1931). H13. Hicks, M. M., Holt, H. P., Guerrant, J. L., and Leavell, B. S., The effect of spontaneous and artificially induced fever on liver function. J . Clin. Invest. 27, 580-588 (1958). i m ,F. E., Foulk, W. T., and Bollman, J. L., Column chromatography of H14. H bile, serum and urine, after intravenous administration of sulfobromophthalein. Gastroenterology 38, 194-202 (1960). H15. Hirsch, R. L., McKay, D. G., Travers, R. I., and Shraly, R. K., Hyperlipidemia fatty liver and bromsulphthalein retention in rabbits injected intravenously with endotoxins. J . Lipzd Res. 6, 563-568 (1964). H16. Hirt, A., Ansorge, J., and Markstahler, H., Luminescenznikroskopische Untersuchungen an der lebenden Frosch- und Rattenleber. 2.Anat. EntwicklungsgesChichte 109, 1-32 (1939). H17. Hiiber, R., Correlation between molecular configuration of organic compounds and their transfer in living cells. Symp. Quunt. BioZ. 8, 40-48 (1940). H18. Hober, R., Studies concerning the nature of the secretory activity of the isolated Ringer-perfused frog liver. 1. Differential secretion of pairs of dyestuffs. J . Gen. Physiol. 23, 185-190 (1940). H19. Hober, R., and Titajew, A., tfber die Sekretionsarbeit der Leber vom Frosch. Arch. Ges. Physiol. 223, 180-194 (1929). H20. Hoch, J., and Combes, B., Biliary excretion of phenol 3:6 dibromophthalein disulfonate in rats fed a protein-free diet. Proc. SOC.Exptl. Bwl. Med. 127,789-792 (1968). H21. Hohbach, R. T., and Sanders, H. W., A unique response to synthetic oestrogen progestational hormone in a familial case of recurrent intrahepatic cholestasis of pregnancy. Gastroenterology 48, 822-826 (1965). H22. Horvath, S. M., Hutt, B. K., Knapp, D. W., and Werner, A. Y., Studies on bromsulphthalein and cardiac output in the hepatecomized dog. J. Physiol. (London) 119, 129-138 (1953). H23. HEU,J. M., Buchanan, P. J., Anilane, J., and Anthony, W. L., Hepatic glutathione concentrations linked to ethionine toxicity in rats. Biochem. J . 106, 639-644 (1968). H24. Huber, T. E., and Wiley, A. T., Cortisone in the treatment of infectious hepatitis. Ann. Internal Med. [N. S.] 42, 1011-1025 (1955). H25. Hultman, E., and Castenfon, H., Some aspects of the bromsulphthalein disappearance curve on giving single doses. Scand. J . Clin. Lab. Invest. 10, Suppl. 31, 278-279 (1957). H26. Hultman, E., and Castenfors, H., An indirect method for detecting increases in biliary pressure. Scand. J . Clin. & Lab. Invest. 14, 277-284 (1962). H27. Hunton, D. B., Bollman, J. L., and Hoffman, H. N., Studies of hepatic function with indocyanine green. Gastroenierology 39, 713-726 (1960).

374

PAULA JABLONSKI AND J. A. OWEN

H28. Hunton, D. B., Rollman, J. L., and Hoffman, H. N., Plasma removal of indocyaninegreen and bromsulphalein, effect of dosage and blocking agents. J . Clin. Invest. 40, 1648-1655 (1961). 11. Ingelfinger, F. J., Bradley, S. E., Mendeloff,A. I., and Kramer, P., Studies with bromsulphalein. 1. Its disappearance from blood after a single intravenous injection. Gastroenterology 11, 646-657 (1948). 12. Inkley, S. R.,Brooks, L., and Krieger, H., A study of method for the prediction of plasma volume. J . Lab. Clin. Med. 46, 841-850 (1957). 13. Israel, H. L., and Reinhold, J. G., Detection of cirrhosis and other diseases of the liver by laboratory tests. J . Lab. Clin. Med. 23, 588-592 (1938). J1. Jablonski, P., Douglas, M., Gordon, E., Owen, J. A., and Watts, J. McK., Unpublished observations (1969). 52. Jaff6, M., Uber die nach Einfuhrung von Brombenzol und Chlorbenzol im Organismus entstcherden schwefelhaltigen Sauren. Ber. Deut. Chem. Ges. 12 1092-1098 (1879). 53. Javitt, N. B., Phenol 3,6-Dibromophthalein Disulfonate, A new compound for the study of liver disease. Proc. SOC.Exptl. Biol. Med. 117, 254-257 (1964). 54. Javitt, N. B., Hepatic glutathione and glucuronide conjugation. Zn “Therapeutic Agents and the Liver” (N. McIntyre and S. Sherlock, eds.), pp. 17-30. Blackwell, Oxford, 1965. 55. Javitt, N. B., Wheeler, H. O., Baker, K. J., Ramos, 0. L., and Bradley, S. E., The intrahepatic conjugation of sulphobromophthalein and glutathione in the dog. J . Clin. Znmst. 39, 1570-1577 (1960). J6. Jenner, J. A., and Smyth, D. H., Excretion of phlorrhizin by liver. J . Physiol. (L~ndon)137, 18P-19P (1957). 57. Jirsa, M., and Hykes, P., Preparation, properties and metabolism of iodinated bromsulphthalein. Nature 211, 645-646 (1966). 58. Johnson, M. K., A distinct enzyme of rat liver and kidney coupling glutathione with some aliphatic halogen compounds. Biochem. J . 87, 9P-10P (1963). K1. Kater, R. M. H., Harrison, D. D., and Mist&, S. P., Alterations in sulfobromophthalein sodium removal from blood in patients with pruritus of pregnancy. Gastroenterology 63, 941-946 (1967). K2. Kater, R. M. H., and Mistilis, S. P., Obstetric cholestasis and pruritus of pregnancy. Med. J . Australia I, 638-640 (1967). K3. Katz, W. J., and Scarf, M., Bromosulphthalein reactions. Report of a fatal case and a review of the literature. Am. J . Med. Sci. 248, 545-549 (1964). K4. Kauhtio, J., Bromsulphalein liver function test in severe infectious infantile gastroenteritis, experimental studies on functional capacity of liver in infants. Ann. Med. Exptl. Biol. Fenniae (Helsinki) 28, Suppl. 5, 1-115 (1950). K5. Keller, D. M., and Lotspeich, W. D., Some effects of phlorizin on the metabolism of mitochondria. J. Biol. Chem. 234, 987-990 (1959). K6. Kerr, W. J., Delprat, G. D., Epstein, N. N., and Dunievitz, M., The rose bengal test for liver function. J. Am. Med. Assoc. 86, 942-946 (1925). K7. Kinnear, A. A., and Pretorious, P. J., Liver function in kwashiorkor. Brit. Med. J . I, 1528-1530 (1956). K8. Kinnear, A. A., and Pretorius, P. J., Influence of high-protein diet on liver function tests with kwashiorker. Brit. Med. J . 11, 389-391 (1957); also S. African Med. J . 31, 174-175 (1957). K9. Klaasen, C. D., and Plaa, G. L., Hepatic disposition of phenol dibromophthalein disulfonate and sulphobromophthalein. Am. J . Physiol. 216, 971-976 (1968). K10. Klatskin, G., Subacute hepatic necrosis and post-necrotic cirrhosis due to anicteric infections with the hepatitis virus. Am. J . Med. 26, 333-358 (1958).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

375

K11. Klein, R. I., and Levinsohn, S. A., Removal of bromsulphalein from the blood stream by the reticuloendothelial system. Proc. Soc. Exptl. Biol.Med. 31, 179-181 (1933). K12. Kleiner, G. J., Kresch, L., and Arias, 1. M., Studies of hepatic excretory function. 11. The effect of norethynodrel and mestranol on bromsulphthalein sodium metabolism in women of childbearing age. New. Engl. J. Med. 273, 420-423 (1965). K13. Kopp, I., and Solomon, H. C., Liver function in therapeutic malaria. Am. J. Med. Sci. 206, 90-97 (1943). K14. Kory, R. C., Bradley, M. H., Watson, R. N., Callahan, R., and Peters, B. J., A 6-month evaluation of an anabolic drug, norethandrolone, in underweight persons. 11. BSP retention and liver function. Am. J. Med. 26, 243-248 (1959). K15. Krebs, J. S., Metabolism of BSP and its effect on plasma clearance of BSP. In “Liver Function” (R. W. Brauer, ed.), pp. 302-309. Waverly Press, Baltimore, Maryland, 1958. K16. Krebs, J. S., Relation of metabolism of sulfobromophthalein sodium to its blood clearance in the rat. Am. J. Physiol. 197, 292-296 (1959). K17. Krebs, J. S., and Brauer, R. W., Uptake of bromsulphthalein by the liver of the rat. 11. Studies with radioactive bromsulphthalein (BSP). Federation Proc. 8, 310 (1949). K18. Krebs, J. S., and Brauer, R. W., Metabolism of sulfobromophthalein sodium in the rat. Am. J . Physiol. 194, 37-43 (1958). K19. Krebs, J. S., and Brauer, R. W., Modification of BSP metabolism by bile stasis and liver injury. Am. J. Physiol. 198, 774-778 (1960). K20. Krebs, J. S., Brauer, R. W., Bollman, J. L., and Leong, G. F., Formation of hepatic lymph. Am. J . Physiol. 207, 877-882 (1964). K21. Krebs, J. S., Nakata, K., and Brauer, R. W., (1964). Cited by Brauer, R. W., In “The Biliary System” (W. Taylor, ed.), pp. 41-68. Blackwell, Oxford, 1965. K22. Kubin, R. H., Grodsky, G. M., and Carbone, J. V., Investigation of rose bengal conjugation. Proc. SOC.Exptl. Biol.Med. 104, 640-653 (1960). L1. Larsson-Cohn, U., The two hour sulfobromophthalein retention test and the transaminase activity during oral contractive therapy. Am. J. Obstet. Gynecol. 98, 188-193 (1967). L2. Lavern, G. D., Cole, W. H., Keeton, R., Gephardt, M. C., and Dyniewicz, J. M., Bromsulfalein clearance. J . Lab. Clin. Med. 34, 965-972 (1949). L3. Laws, J. F., and Eversen, T. C., The significance of decreased dye clearance by the liver following portacaval anastomosis. J. Lab. CZin. Med. 37, 515-519 (1951). L4. Leevy, C. M., Dye extraction by the liver. In “Progress in Liver Disease” (H. Popper and F. Schaffner, eds.), Vol. 1, pp. 174-186. Grune & Stratton, New York, 1961. L5. Leevy, C. M., Bender, J., Silverberg, M., and Naylor, J., Physiology of dye extraction by theliver: Comparative studies of sulfobromophthalein and indocyanine green. Ann. N . Y . A d . Sci. 111, 161-175 (1963). L6. Leevy, C. M., Cherrick, G. R., and Davidson, C. S., Observations on norethandrolone-induced abnormalities in plasma decay of sulfobromophthalein and indocyanine green. J. Lab. Clin. Med. 67, 918-926 (1961). L7. Levy, C. M., and Greenberg, J., Radioisotope scanning as a guide to liver biopsy. Am. J . Med. Sci. 233, 28 (1957). L8. Leevy, C. M., Longueville, J., Paumartner, G., and Howard, M. M., Indocyanine green clearance as a test for hepatic function. J . Am. Med. Assoc. 200, 236-240 (1967).

376

PAULA JABLONSKI AND J . A. OWEN

L9. Leevy, C. M., Mendenhall, C. L., Lesko, W., and Howard, G., Estimation of hepatic blood flow with indocyanine green. J . Clin. Znvest. 41, 1109-1178 (1967). L10. Leevy, C. M., Ryan, C. M., and Fineberg, J. C., Diabetes mellitus and liver dysfunction; etiological and therapeutic considerations. Am. J . Med. 8, 290-299 (1950). LlOa. Lester, R., I n “The Biliary System” (W. Taylor, ed.), p. 227. Blackwell, Oxford, 1965. L11. Lewis, A. E., The concept of hepatic clearance. Am. J . Clin. Pathol. 18, 789-795 (1948). L12. Lewis, A. E., Investigation of hepatic function by clearance techniques. Am. J . Physiol. 163, 54-61 (1960). L13. Lieberman, A. H., and Chdds, A. W., Effect of ethanol on hepatic metabolism of sulfobromophthalein. Am. J . Physiol. 213, 353-357 (1967). L14. Lima, F. W., and Pieroni, R. R., Paper chromatographic separation of components of rose bengal labelled with 1’31. Nature 184, 1065 (1957). L15. Lorber, S. H., and Shay, H., Enterohepatic circulation of bromsulphalein. I. Studies on man with special reference to clinical BSP test. Gastroenterology 20, 262-271 (1952). L16. Lotspeich, W. D., and Keller, D. M., A study of some effects of phlorkin on the metabolism of kidney tissue in vitro. J . B i d . Chem. 222, 843-853 (1950). M1. MacDonald, D., Some observations on the disappearance of bromsulphthalein dye from the blood; its relation to liver function (preliminary report). Can. Med. Assoc. J . 39, 556-560 (1938). M2. MacDonald, D., A practical and clinical test for liver reserve. Surg., Gynecol. Obstet. 69, 70-82 (1939). M3. Machella, T. E., Relationship of bromsulphalein retention to the fever of natural P . falciparum malaria. Am. J . Med. Sci. 213, 81-86 (1947). M4. Mackay, I. R., Chronic hepatitis. Edect of prolonged suppressive treat.ment and comparison of azathioprine with prednisolone. Quart. J . Med. [N. S.] 37, 379-392 (1968). MFi. McKinney, 8. E., Peck, H. M., Bochey, B., Byham, B., and Beyer, I<. H., Benemid, p(di-n-propylsulfamy1)-benroic acid. J . Pharmacol. Exptl. Therap. 102, 208-213 (1951). M6. McLeod, G. M., French, A. B., Good, C. J., and Wright, F. S., Gastrointestinal absorption and biliary excretion of phenolsulphonephthalein (phenol red) in man. J . Lab. Clin.Med. 71, 193-200 (1968). M7. McMillion, C. R., and Dunning, H. A. B., A chromatographic technique for the identification of fluorescein and phenolphthalein derivatives. J . Am. Pharm. ASSOC., Sci.Ed. 48, 249-251 (1959). M8. McNalty, A. A. (ed.), “The British Medical Dictionary.” Caxton Publ. Co., London, 1963. M9. McNeil, H. L., Quantitative estimation of phenol-tetrachlorophthalein excreted in fresh bile in disease of the liver. J . l a b . Clin. Med. 1, 822-825 (1916). M10. Mandema, E., De Fraiture, W. H., Nieweg, H. O., and Arends, A., Familial chronic idiopathic jaundice (Dubin-Sprinz Disease), with a note on bromsulphthdein metaboliim in this disease. Am. J . Med. 28, 42-50 (1960). M11. Martindale, W., “The Extra Pharmacopoeia,” Vol. 11, p. 1376. Pharm. Press, London, 1955. M12. Mason, M. F., Hawley, G., and Smith, A., Application of the saturation principle

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

377

to the estimation of functional hepatic mass in normal dogs. Am. J . Physiol. 162, 42-47 (1945). M13. Mateer, J. G., Raltz, J. I., Marion, D. F., and MacMillan, J. RI., Liver function tests; general evaluation of liver function tests, and appraisal of comparative sensitivity and reliability of newer tests, with particular emphasis on cephalincholesterol flocculation test, intravenous hippuric acid test and improved bromsulphthalein test with new normal standard. J. Am. Med. Assoc. 121,723-728 (1943). M14. Mateer, J. G., Balts, J. I., Comanduras, P. D., Steelf, P. D., Brouwer, S., and Yagle, E. M., Further advances in liver function tests and the value of a therapeutic test in facilitating the earlier diagnosis and treatment of liver impairment. Gastroenterology 8, 52-70 (1947). M15. Mendeloff, A. I., Fluorescence of intravenously administered rose bengal appears only in hepatic polygonal cells. Proc. SOC.Exptl. Biol. Med. 70, 556-558 (1949). M16. Mendeloff, A. I., Kramer, P., Ingelfinger, F. J., and Bradley, S. E., Studies with bromsulphalein. 11. Factors altering its disappearance from the blood after single intravenous injection. Gastroenterology 13, 222-234 (1949). M17. Mendenhall, C. L., and Leevy, C. M., False negative bromsulfalein tests. New Engl. J. Med. 264, 431-433 (196lj. M18. Mercier, J., Jacquenoud, P., and Sebastier, M. R., Etude de quelques propri6t6s pharmacodynamiques de la N-allyl-normorphine. Anesthesie Analgesie 12, 256-278 (1955). M19. “Merck Index, 8th ed. Merck & Co., Rahway, New Jersey, 1968. M20. Meltzer, J, I., Wheeler, H. O., and Cranston, W. I., Metabolism of sulfobromophthalein sodium (BSP) in dog and man. Proc. SOC.Exptl. Biol. Med. 100, 174-179 (1959). M21. Metzler, C., Hoffbauer, F. W., and Benson, E., The ‘‘corrected bromsulfalein test.” A comparison with cephalin-cholesterol flocculation test and the thymol turbidity reaction in jaundiced patients. J. Lab. Clin. Med. 47, 519-528 (1956). M22. Michaeli, L., and Menten, M. L., Die Kinetic der Invertinwirkung. Biochem. 2. 49, 333-369 (1913). M23. Michie, A. J., and Michie, C. R., Attainment of equilibrium between plasma and urine with reference to memurement of renal clearances. J . Urol. 66, 518-526 (1951). M24. Mikulecky, M., Detailed study of the kinetics of plasma bromsulphthalein. New experimental findings and theoretical considerations. Clin. Chim. A& 21, 43-48 (1968). M25. Miller, G. H., Davis, M. E., King, A. G., and Huggins, G. B., Serum proteins in pregnancy. Thermal coagulation and the binding of anions. J . Lab. Clin. Med. 37, 538-543 (1951). M26. Mills, M. A., and Dragstedt, C. A., Bromsulphalein dye retention test as a measure of the functional activity of the reticuloendothelial system. Proc. SOC. Exptl. Biol. Med. 34, 228-231 (1936). M27. Mills, M. A., and Dragstedt, C. A., Removal of intravenously injected bromsulphthalein from bloodstream of the dog; comparison of removal of intravenously injected bilirubin and that of bromsulfalein. A . M . A . Arch. Internal Med. 62, 216-221 (1938). M28. Mollison, R. L., and Cutbush, H., Bromsulphtlialein excretion in the new-born. Arch. Disease Childhood 24, 7-11 (1949). M29. Monroe, L. S., and Kittinger, A., The mechanism of bromsulfalein excretion. Am. J. Gastroenterol. 31, 634-640 (1959).

378

PAULA JABLONSKI A N D J . A. OWEN

M30. Monroe, L. S., and Kittinger, A., The biliary dynamics of the metabolites of sulfobromophthaleinsodium (BSP) in man. J . Lab. Clin. Med. 68,468476 (1961). M31. Morey, G., Gabzuda, G. J., and Scudamore, H. H., Sensitization to bromsulphalein. Gastroenterology 13, 246-249 (1945). M32. Moyer, J. H., and Wurl, 0. A., Liver biopsy: Correlation with clinical and biochemical observations. A m . J . Med. Sci. 221, 28-37 (1951). M33. Mueller, M. N., and Kappas, A., Estrogen pharmacology. I. The influence of estradiol and estriol on hepatic disposal of sulfobromophthalein (BSP) in man. J . Clin. Invest. 43, 1905-1914 (1964). M34. Murray, J. F., and Nebel, L., Measurement of hepatic blood volume in dogs by an indicator dilution technique. Clin. Res. 7 , 290 (1959). M35. Myers, J. D., Observations on the excretion of bromsulfalein. J . Clin. Invest. 28, 801 (1949). N l . Nadasdi, M., Ockner, R. K., and Davidson, C. S., Oral contraceptives. New. Engl. J . Med. 276, 1327-1328 (1967). N2. Nadeau, G., A simplified bromsulphthalein clearance tast. A m . J . Clin. Pathol. 24, 740-746 (1954). N3. Nakata, K., Leong, G. F., and Brauer, R. W., Direct measurement of blood pressures in the minute vessels of the liver. A m . J . Physiol. 199, 1181-1188 (1960). N4. Neale, G., Caughey, D. E., Mollin, D. L., and Booth, C. O., Effects of intrahepatic and extrahepatic infection on liver functions. Brit. Med. J . I, 382-387 (1966). N5. Neefe, J. R., and Reinhold, J. G., Laboratory aids in the diagnosis and management of infectious (epidemic) hepatitis; analysis of results obtained by studies on 34 volunteers during early and convalescent stages of induced hepatitis. Gastroenterology 7 , 397413 (1946). N5a. Neubeck, C. E., and Smythe, C. V., Glutathione. Arch. Biochem. 4 443 (1944). N6. Neumayr, A., Parzer, O., and Vetter, H., Zur Problematik der Bromsulphaleinclearances als Leberfunktionsprufung. Deut. Med. Wochschr. 79, 1039-1041 (1954). N7. Norcross, J. W., White, R. M., and Bradley, R. F., The bromsulfalein liver function tests. BSP test with special reference to renal excretion. Am. J . Med. Sci. 221, 137-139 (1951). N8. Nordyke, R. A., Diagnosis of distal common bile duct obstruction with radioiodinated rose bengal. Clin. Res. 7 , 295 (1959). N9. Nordyke, R. A., and Blahd, W. H., Blood disappearance of radioactive rose bengal-rapid simple test of liver function. J . Am. Med. Assoc. 170, 1159-1164 (1959). NlO. Nosslin, B., Influence of dosage level on disappearance rate of bromsulphthalein (BSP). Scand. J . Clin. Lab. Invest. Suppl. 31, 295 (1958). N11. Nosslin, B., Bromsulphthalein retention and jaundice due to unconjugated bilirubin following treatment with male fern extract. Scand. J . Clin. Lab. Invest. Suppl. 16, 69 (1963). N12. Nosslin, B., and Morgan, E. H., The effectof phloroglucinol derivatives from male fern on dye excretion by the liver in the rabbit and the rat. J . Lab. Clin. Mad. 66, 891-902 (1965). 0 1 . Obrinsky, W., Denley, M. L., and Brauer, R. W., Sulfobromophthdein sodium extraction test as a measure of liver function in premature infants. Pediatrics 9, 421438 (1952). 0 2 . O’Leary, P. A., Greene, C. H., and Rowntree, L. C., Diseases of the liver. VIII.

BEOMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

03. 04. 05.

PI. P2. P3. P4. P5.

P6. P7. P8. P9. P10. P11. P12. R1. R2. R3. R4. H5.

379

The various types of syphilis of the liver with reference to tests for hepatic function. A . M . A . Arch. Internal Med. 44, 155-193 (1929). O’MBille, E. R. L., Richards, T. G., and Short, A. H., Factors determining the maximal rate of organic anion secretion by the liver and further evidence on the hepatic site of action of secretin. J . Physiol. (London) 186, 429-438 (1966). Overall, J. E., Williams, C. M., and Hill, C., A note concerning sources of variance in determination of human plasma volume. J . Lab. Clin. Med. 64, 186-191 (1959). Owen, C. A., The effect of entrohepatic circulation on the bromsulphthalein test of hepatic function. J . Lab. Clin. Med. 38, 583-584 (1951). Pagliardi, E., and Giangradni, E., Studies on hepatic sulfobromophthalein removal during infusion. CZin. Chem. 7 , 246-255 (1961). Palefski, I. O., Visualization of the gall-bladder with sodium tetrabromophenolphthalein by oral or intraduodenal administration. Med. J . & Record 121,474482 (1925). Palva, I. P., and Mustala, 0. O., Oral contraceptives and liver damage. Brit. Med. J . 11, 688-689 (1964). Partington, J. R., “A Short History of Chemistry,” p. 317. Macmillan, New York, 1938. Paulson, M., and Wyler, C. I., Multiple tests of hepatic function with gastroenteric malignancy; the value of bromsulphthalein, hippuric acid and tbe van den Bergh reaction in detecting hepatic metastases, with an evaluation of normality of hippuric acid test. Ann. Internal Med. [N. S.] 16, 872.478 (1942). Phdp, J. R., Grodsky, G. M., and Carbone, J. V., Mercaptide conjugation in the uptake and secretion of sulfobromophthalein. Am. J . Physwl. 200,545-547 11961). Plaa, G. L., Sparks, R. D., Abide, J. K., and Hunter, F. M., Clinical usefulness of BSP and its metabolic products. Preliminary observations. Southern Med. J . 64, 102&1030 (1961). Plotner, K., Kuhn, H. W., and Wirsching, W., Die quantitative Bestimmung der renalen Bromsulphthalein Ausscheidung als Leberfunktionsprobe. Deut. Med. Wochschr. 83, 2051-2055 (1958). Popper, H. P., and Schaffner, F., “Liver Structure and Function.” Chapter 12, pp. 80-87. McGraw Hill, New York, 1957. Popper, H., Hepatic drug reactions simulating viral hepatitis. I n “Therapeutic Agents and the Liver” (N. McIntyre and S. Sherlock, eds.), pp. 81-84. Blackwell, Oxford, 1965. Preisig, R., Williams, R., Sweeting, J., and Bradley, S. E., Changes in sulfobromphthalein transport and storage by the liver during viral hepatitis in man. Am. J . Med. 40, 170-183 (1966). Pugh, P. M., and Stone, S. L., The effect of 2,4-dinitrophenol and related compounds on bile secretion. J . Physiol. (London) 198, 3 9 4 9 (1968). Raban, P., and Jirsa, M., Metabolism of rose bengal. Nature 196, 1110-1111 (1962). Rawson, R. A., Binding of T-1824 and structurally related diazo dyes by plasma proteins. Am. J . Physiol. 138, 708-717 (1943). Reemtsma, K., Hottinger, G. C., D e Craff, A. C., and Creech, O., Studies of hepatic function and blood flow using indocyanine green. Clin. Res. Proc. 8, 77 (1960). Reemtsma, K., Hottinger, G. C., De Graff, A. C., and Creech, O., The estimation of hepatic blood flow using indocyanine green. Surg. Gynecol. Obstet. 110, 353-356 (1960). Reich, J. S., and Davis, W. D., The differential value of bromsulfalein retention

380

PAULA JABLONSEI AND J . A. OWEN

testa in acute hepatitis and obstructive jaundice. Am. J . Digest. Diseases 6, 770-775 (1960). R6. Reinhold, J. G., Bromsulphthalein testa. In “Medical and Public Health Laboratory Methods” (J. S. S i m o n and D. G. Gentzkou, eds.), pp. 100-102. Lea & Febiger, Philadelphia, Pennsylvania, 1955. R7. Rice-Wray, E., Oral contraceptives and liver damage. Brit. Med. J . 11, 1011 (1964). R8. Rice-Wray, E., Schulz-Contreras, M., Guerrero, I., and Aranda-Rosell, A., Long term administration of norethindrone in fertility control. J . Am. Med. Assoc. 180, 355-361 (1962). R9. Richards, T. G., Plasma concentration of bromosulphaleinafter single intravenous injection in normal and abnormal subjects. In “The Biliary System” (W. Taylor, ed.), pp. 567-579. Blackwell, Oxford, 1965. RlO. Richards, T. G., Tindall, V. R., and Young, A., A modification of the bromsulphthalein liver function test to predict the dye content of the liver and bile. Clin. Sci. 18, 499-511 (1959). R11. Rimington, C., Biliary secretion of porphyrins. In “The Biliary System” (W. Taylor, ed.), pp. 325-331. Blackwell, Oxford, 1965. R12. Rimington, C., Quin, J. I., and Roets, G. C. S., Onderstepoort. J . Vet. Sci. Animal Ind. 9, 225 (1935), cited by Rimington (R11). R13. Roberts, R. J., Klaassen, C. D., and Plaa, G. L., Maximum biliary excretion of bilirubin and sulfobromophthalein during anesthesia-induced alteration of rectal temperature. Proc. Soc. Exptl. Biol. Med. 126, 313-316 (1967). R14. Robinson, G. L., A study of liver function and plasma volume in chronic rheumatism by means of phenol tetrabromo-phthalein sodium sulphonate. Ann. Rheumatic Dbeases 3, 208-221 (1942-1943). R15. Rosenau, W. H., Dangers in the use of certain halogenated phthaleins as functional tests. J. Am. Med. Assoc. 86, 2017-2020 (1925). R16. Rosenberg, D. H., and Soskin, S., The azorubin S test of liver function; an evaluation, with a comparative study of the bromsulphthalein and hippuric acid tests. Ann. Internal Med. [N. S.] 13, 1644-1654 (1940). R17. Rosenthal, S. M., An improved method for using phenol tetrachlorophthalein as a liver function test. J . Pharmacol. Exptl. Therap. 19, 385-391 (1922). R18. Rosenthal, S. M., A new method for testing liver function with phenol tetrachlorophthalein. IV. Proc. SOC.Exptl. Biol. Med. 21, 73-75 (1923). R18a. Rosenthal, S. M., Liberation of absorbed substances from proteins. J . Pharmacol. Exptl. Therap. 26, 449-457 (1925). R19. Rosenthal, S. M., Studies upon the combining power of proteins with rose bengal. J. Pharmacol. Exptl. Therap. 29, 521-532 (1926). R20. Rosenthal, S. M., and Bourne, W., The effectof anaesthetics on hepatic function. J . Am. Med. Assoc. 90, 377-379 (1928). R21. Rosenthal, F., and von Falkenhausen, M., Untersuchungen iiber die Moglichkeit einer Funktionspriifung der Leber mit gallenfahigen Farbstoff en (Chromocholoskopie). Kliri. Wochschr. 44, 1293-1295 (1921). R22. Rosenthal, S. M., and White, E. C., Studies in hepatic function. VI. A. The pharmacological behavior of certain phthalein dyes. B. The value of selected phthalein compounds in estimation of hepatic function. J . Pharmawl. Exptl. Therap. 24, 265-288 (1924). R23. Rosenthal, S. M., and White, E. C., Clinical application of the bromsulphaleirl test for hepatic function. J. Am. M e d . Assoc. 84, 1112-1114 (1925).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

381

R24. Rotor, A. B., Manahan, L., and Florentin, A., Familial non-hemolytic jaundice with direct van den Bergh reaction. A& Med. Philippina 6, 37-47 (1966). R25. Rowntree, L. G., Hurvite, S. H., and Bloomfield, A. L., An experimental and clinical study of the value of phenol tetrachloro-phthalein as a test for hepatic function. Bull. Johns Hopkins Hosp. 24, 327-342 (1913). S l . Salmon, G. W., and Richman, E. L., Liver function in the new-born infant. J . Pediat. 26, 522-533 (1943). S2. Sapirstein, L. A., and Simpson, J. A. M., Plasma clearance of rose bengal. (Tetraiodotetrabroduorescein.) Am. J . Physiol. 182, 337-346 (1955). 53. Schaffner, F., Effect of anabolic steroids in man. I n “Therapeutic Agents and the Liver” (N. Mclntyre and S. Sherlock, eds.), pp. 99-113. Blackwell, Oxford, 1965. 54. Schdner, F., and Popper, H., Cholestasis produced by the administration of norethandrolone. Am. J . Med. 26, 249-254 (1955). S6. Schellong, F., and Eisler, B., Experimentelle Beitrage eur Funktionsprufung der Leber und des reticuloendothelialen Apparates mit Farbstoffen: Der klinische Wert der Leberfunktionspriifung mit Tetrachlorphenolphthalein. 2. Ges Ezptl. Med. 68, 738-756 (1928). 57. Schenker, S., and Combes, B., Role of hepatic adenosine triphosphate in BSP transport and metabolism i n vitro. Am. J . Physiol. 212, 295-300 (1967). S8. Scherb, J., Kirschner, M., and Arias, I., Studies of hepatic secretory function. The effect of 17 a-ethyl-19-nortestosterone on sulfobromophthaleinsodium (BSP) metabolism in man. J . Clin. Invest. 42, 404-408 (1963). S9. Schmidt, C. L. A., and Norman, G. J., Further studies on eosin haemolysis. J . Gen Physiol. 4, 681-786 (1922). SlO. Schmidt, L. A., Iob, V., Flotte, C. T., Hodgson, P. E., and McMath, M., Blood volume changes in the aged. Surgery 40, 938-944 (1956). S l l . Schmidt, M. L., Garner, L. M., and Arias, I. M., Studies of hepatic excretory function. 111. Effect of hypopituitarism on the hepatic excretion of sulfobromophthalein sodium in man. Gastroenterology 62, 998-1002 (1967). 512. Schneider, E. M., Berman, J. R., Gall, E. A., and Schiff, L., Needle biopsy of the liver. 1V. Relationship of clinical and laboratory findings and histologic studies in 100 cases of pelvic cancer. Am. J . Med. 16, 207-215 (1953). 513. Schoenfield, L. J., Sulfobromophthalein transport and metabolism. Gastroenterology 48, 530-533 (1965). 514. Schoenfield, L. J., Duodenal drainage of sulfobromophthalein (BSP) in hepatobiliary disease. Gastroenterology 61, 59-64 (1966). 515. Schoenfield, L. J., Foulk, W. T., and Butt, H. R., Studies of sulfobromophthalein sodium (BSP) metabolism in man. I. In normal subjects and patients with hepatic disease. J . Clin. Invest. 43, 1409-1418 (1964). S16. Schoenfield, L. J., McGill, D. B., Hunton, D. B., Foulk, W. T., and Butt, H. R., Studies of chronic idiopathic jaundice (Dubin-Johnson syndrome). I. Demonstration of hepatic excretory defect. Gastroenterology 44, 101-111 (1963). S17. Schoedeld, L. J., McGill, D. B., and Foulk, W. T., Studies of sulfobromophthalein sodium (BSP) metabolism in man. 111. Demonstration of a transport maximum (Tm)for biliary excretion of BSP. J . Clin. Invest. 43, 1424-1432 (1964). 518. Seligson, D., Marino, J., and Dodson, E., Determination of sulfobromophthalein in serum. Clin. Chem. 3, 638-645 (1957). S19. Seligson, D., and Marino, J., Sulfobromophthalein (BSP). I n “Standard Methods of clinical Chemistry” (D. Seligson, ed.), Vol. 2, pp. 186-206. Academic Press, New York, 1958.

382

PAULA JABLONSKI AND J . A. OWEN

520. Selkurt, E. E., Validity of the bromosulphthalein (BSP) method for estimating hepatic blood flow. A m . J . Physiol. 176, 461-467 (1953). 521. Serby, A. M., and Block, L., Liver function as determined by bromsulphthalein in 76 cases. Am. J . Med. Sci. 176, 367-378 (1928). S22. Sherlock, S., “Disease of the Liver and Biliary System.” Blackwell, Oxford, 1959. S23. Sherlock, S., Biliary secretory failure. I n “The Biliary System” (W. Taylor, ed.), pp. 675-678. Blackwell, Oxford, 1965. 524. Sherlock, S., Bearn, A. G., Billing, B. H., and Paterson, J. C. S., Splanchnic blood flow in man by the bromsulphthalein method: The relation of the peripheral plasma bromsulphthalein level to the calculated flow. J . Lab. Clin. Med. 36, 923-932 (1950). S25. Shore, M. I., and Zilversmit, D. B., Effect of india ink on bromsulfalein excretion, phagocytosis and circulation in liver. Am. J . Physiol. 177, 436440 (1954). 526. Shotton, D., Carpenter, M., and Rinehart, W. B., Bromsulfalein retention due to administration of a gall-bladder dye (Bunamiodyl). New. Engl. J . Med. 264, 550-552 (1961). 527. Silen, W., Harper, H. A., DeReimer, R. H., and McCorkle, H. J., Comparative studies of liver function, in dogs with Eck fistula or portacaval transposition. I n “Liver Function” (R. W. Brauer, ed.), pp. 394406. Waverly Press, Baltimore, Maryland, 1958. S28. Simcock, M. J., and Forster, F. M. C., Pregnancy is cholestatic. Med. J . 9ustralia 11, 971-973 (1967). 529. Smons, R. L., Hepatic tests in metastatic carcinoma of liver, review of 73 cases proven at necropsy. Am. J . Med. Sci. 228, 312-316 (1954). S30. Simpson, A. M., Earow, L., and Sapirstein, L. A., Measurement of plasma volume with rose bengal. Am. J . Physiol. 177, 319-324 (1954). S30a. Smith, H. W., Finkelstein, N., Alininosa, L., Crawford, B., and Graber, M., The renal clearance of substituted hippuric acid derivatives and other aromatic acids in dog and mean. J . Clin. Invest. 24, 388-404 (1945). 531. Snell, A. M., and Magath, T. G., The use and interpretation of tests for liver function clinical review. J . Am. Med. Assoc. 110, 167-174 (1938). S32. Soffer, L. J., Present day studies of liver function tests. Medicine 14, 185-284 (1935). S33. Sperber, I., Secretion of organic anions in formation of urine and bile. Pharmacol. Rev. 11, 109-134 (1959). 534. Sperber, I., Biliary secretion of organic anions and its influence on bile flow. I n “The Biliary System” (W. Taylor, ed.), pp. 45Ck457. Blackwell, Oxford, 1965. S35. Staral, T. E., Scanlon, W. A., Shoemaker, W. C., Thornton, F. H., Wendel, R. M., and Schlachter, L., Studies of hepatic blood flow and the rate of bromsulfalien clearance in dogs with portacaval transposition. Surgery 62, 660-663 (1962). S36. Steinfeld, J. L., Difference in daily albumin synthesis between normal men and women as measured with 1311-labelled albumin. J . Lab. Clin. Med. 66, 904-911 (1960). S37. Stekiel, W. J., Kampine, J. P., BanaBaak, E. F., and Smith, J. J., Hepatic clearance of indocyanine in the dog. A m . J . Physiol. 198, 881-885 (1966). 538. Stiles, M. H., Stiles, M. T., and Kolb, A. M., Bromosulphalein retention in low grade chronic illness. J . Lab. Clin. Med. 28, 180-187 (1942). S39. Stone, S. L., The effect of benemid on the hepatic extraction and biliary excretion of bromosulphalein. I n “Liver Function” (R. W. Breuer, ed.), pp. 298-301. Waverly Press, Baltimore, Maryland, 1958.

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

383

540. Stone, S. L., Energy requirements for bile secretion. I n “The Biliary System” (W. Taylor, ed.), pp. 277-292. Blackwell Press, Oxford, 1965. 541. Stowe, W. P., Rose bengal excretion and hippuric acid synthesis tests of liver function. J . Lab. Clin. Med. 24, 866-868 (1939). 542. Strade, M. A., Dotti, L. B., and Ilka, S. J., A comparative study of the colloidal red test in liver diseases with special reference to the zinc turbidity, thymol turbidity and bromosulphthalein tests. Gastroenterology 12,934-998 (1949). 543. Strain, W. H., and Greenlaw, R. H., Cholecystopaques. Med. Radiography Phot. 40,SUPPI., 47-61 (1964). 544. Sunderman, F. W., and Austin, J. H., The measurement of serum volume. Am. J . Physwl. 117, 474-486 (1936). 545. Sundman, V., and Sundman, J., Bacterial growth inhibiting effect of some phloroglucinol derivatives. A d a Pathol. Microbiol. Scand. 63, 345-355 (1961). S46. Sutherland, J. M., and Keller, W. H., Novobiocin and neonatal hyperbilirubinaemia. Am. J . Diseases Children 101,447-450 (1961). S47. Svanberg, A., and Ohlsson, S., Recurrent jaundice of pregnancy. A clinical study of twenty-two cases. Am. J . Med. 27, 40-52 (1959). TI. Tada, M., Cited by Adler (A4). T2. Tada, Y., and Nakashima, K., A new dye for test of liver function and biliary tract function. J . Am. Med. Assoc. 83, 1292-1296 (1924). T3. Taplin, G. V., Meredith, 0. M., and Kade, H., The radioactive (W-tagged) rose bengal uptake excretion test for liver function, using external gamm*ray scintillation counting techniques. J . Lab. Clin. Med. 46, 656-678 (1955). T3a. Therkelsen, A. J., Effect of serum on spectral absorption of bromosulphthalein after intravenous injections of heparin. S a n d . J . Clin. & Lab. Invest. 9, 156-159 (1957). T4. Thomas, L. J., and Zimmerman, H. J., The pattern of abnormality of h e r function in metastatic carcinoma. J . Lab. Clin. Med. 39, 882-887 (1952). T5. Thomas, M. C., and Plaa, G. L., A method for the simultaneous determination of sulfobromophthalein sodium and metabolite in blood. Am. J . Clin. Pathol. 84, 488-492 (1960). T6. Tindall, V. R., and Beazley, J. M., An assessment of changes in liver function during normal pregnancy-using a modified bromosulphalein test. J . Obstet. Gynaecol. Brit. Commonwealth 72, 717-737 (1967). T7. Toulet, J., Albot, G., Lunel, J., and Davezac, J. F., Interpretation correlative des deux clearances successives d’bpuration sanguine de la bromesulfonephtal6he. Recent Advan. Gastroaterol. 3, 407-416 (1966). T8. Toulet, J., Davezac, J. F., Lunel, J., and Albot, G., Etude des 6lbments qui interviennent dans les modifications physio-pathologiques de l’bpuration sanguine de la bromesulphonephtal6ine. Presse Zed. 41, 1854-1864 (1965). T9. Tovey, J. E., Bromsulphalein clearance test of liver function. Clin. Chim. A d a 16, 149-154 (1967). TlO. Tubis, M., Nordyke, R. A., Posnick, E., and Blahd, W. H., The preparation and use of l*lI-labelled sulfolromophthalein in liver function testing. J . Nucl. Med. 2, 282-288 (1961). T11. Tura, S., Pieragnoli, E., and Dalmonte, P. R., The d i e r e n t i d diagnosis of icterus by bromsulphthalein. Recent. Advan. Gastroenterol. 3, 399402 (1966). T12. Tura, S., Pieragnoli, E., Dalmonte, P. R., and Campanacci, D., Die Kinetik des Bromsulphaleins. Acta Hepato-Splenol. 11, 1-18 (1964). V1. Vanden Bossche, H., Automatic determination of bromosulfophthalein in serum. Clin. Chim. Acta 9, 310-313 (1964).

384

PAULA JABLONSKI AND J. A. OWEN

V2. Vanlerenberghe, J., The effects of hypothermia on biliary function. I n “The Biliary System” (W. Taylor, ed.), pp. 263-276. Blackwell, Oxford, 1965. V3. Vanlerenberghe, J., Milbled, G., and Guerin, T., A study of the secretory mechanism of choleresis. Pathol. Bwl. Semaine Hop. [N. S.] 8, 1231-1234 (1960). V4. Vanlerenberghe, J., Trupin-Bar, N., Guislain, R., and Bel, C., Action de l’hypothermie sur la fonction biliare. J . Physiol. (Paris) 69, 111-138 (1967). V5. Varley, H., In “Practical Clinical Biochemistry,” p. 383. Heinemann, London, 1967. V6. Verschure, J. C. M., Clinical use of measurements of clearance and maximum capacity of the liver. Acta Med. Scana. 142,409-419 (1952). V7. Vest, M. F., Conjugation of sulfobromophthalein in newborn infants and children. J . Clin. Invest. 41, 1013-1020 (1962). V8. Vest, M. F., and Rossier, It., Detoxification in the newborn. The ability of the newborn infant to form conjugates with glucuronic acid, glycine, acetate and glutathione. Ann. N . Y . Acad. Sci. 111, 183-198 (1967). V9. Victor, J., Van Buren, J. R., and Smith, H. P., Studies on vital staining. IV. India ink and brilliant vital red. Importance of considering liver excretion in the study of “blockade of R. E. S.” J . Exptl. Med. 61, 531-548 (1930). V10. Vink, C. L. J., The half-life concept in the determination of functional capacity of the liver. Clin. Chim. Acta 4, 583-589 (1959). V l l . Vink, C. L. J., and Kroes, A. A., Liver function and age. Clin. Chim. Acta 4, 674-682 (1959). V12. Vink, C. L. J., and Kroes, A. A., The half-life of galactose and bromosulphalein in pathological conditions. Clin. Chim. Acta 4, 883-888 (1959). W1. Waldenstein, S. S., and Arcilla, R. A., Measurement of hepatic blood flow by clearance methods. Am. J . Digest. Diseases 3, 137-158 (1958). W2. Watson, R. N., Bradley, M. H., Callahan, R., Peters, B. J., and Kory, R. C., A six month evaluation of an anabolic drug norethandrolone in underweight persons. I. Weight gain. Am. J . Med. 26, 238-248 (1955). W3. Werner, S. C., Hanger, F. M., and Kritzler, R. A., Jaundice during methyl testosterone therapy. Am. J . Med. 8, 325-331 (1950). W4. Werther, J. L., and Korelitz, B. I., Chlorpromazine jaundice: Analysis of twentytwo cases. Am. J . Med. 22, 351-366 (1957). W5. Westphal, U., de Armond, R., Priest, S. G., and Stets, J. F., Azorubin-binding capacity and protein composition of serum of rats subjected to tourniquet shock and to treatment with carbon tetrachloride. J . Clin. Invest. 31, 1064-1068 (1952). W6. Westphal, U., and Gedigk, P., Azorubin-binding capacity of normal and pathological sera. Proc. SOC.Exptl. Biol. Med. 76, 83&843 (1951). W7. Westphal, U., Priest, S. G., and Stets, J. F., Protein composition and azorubinbinding capacity of serum of rabbits subjected to tourniquet shock. 4 m . J . Phvsiol. 173, 305-311 (1953). W8. Wheeler, H. O., The function of the biliary tract. Progr. Liver Diseases 2, 15-25 (1965). W9. Wheeler, H. O., Cranston, W. I., and Meltzer, J. I., Hepatic uptake and biliary excretion of indocyanine green in dogs. Proc. Soe. Exptl. Biol. Med. 99, 11-14 (1958). W10. Wheeler, H. O., Epstein, R. M., Robinson, R. R., and h e l l , E. S., Hepatic storage and excretion of sulfobromophthalein sodium in dog. J . Clin. Invest. 39, 236-247 (1960). W11. Wheeler, H. O., and Mancusi-Ungaro, P. L., Role of bile ducts during secretin choleresis. Am. .J. I’hysiol. 210, 1153-1159 (1966).

BROMSULFOPHTHALEIN AND OTHER CHOLEPHILIC DYES

385

W12. Wheeler, H. O., Meltzer, J. I., Epstein, R. M., and Bradley, S. E., Hepatic storage and biliary transport of bromsulphalein in dog and man. J . Clin. Invest. 37, 942 (1958). W13. Wheeler, H. O., Meltzer, J. I., and Bradley, S. E., Biliary transport and hepatic storage of sulfobromophthalein sodium in the unaenesthetized dog, in normal man and in patients with liver disease. J . Clin. Invest. 39, 1131-1144 (1960). W14. Wheeler, H. O., Pier, C., Mancusi-Ugaro, P. L., and Whitlock, R. T., Bile salt transport in the dog. J. Clin. Invest. 39, 1039-1040 (1960). W15. Wheeler, H. O., and Ramos, 0. L., Determinants of the flow and composition of bile in the unaenesthetized dog during constant infusion of sodium taurocholate. J . Clin. Invest. 39, 151-170 (1960). W16. Whelan, F. J., and Plaa, G. L., The application of thin layer chromatography to sulfobromophthalein metabolism studies. J . Toxicol. Appl. Pharm. 6, 457-463 (1963). W17. Whetstone, H. I., La Motta, R. V., Middlebrook, L., Tennant, R., and White, B. V., Studies of cholinesterase activity. IV. Liver function in pregnancy. Values of certain standard liver function tests in normal pregnancy. Am. J . Obstet. Gynecol. 76, 480-488 (1958). W18. Wiegand, B. D., Ketterer, S. G., and Rapaport, E., The use of indocyanine green for the evaluation of hepatic function and blood flow in man. Am. J . Digest. Diseases 6, 427436 (1960). W19. Wilensky, L. S., Zur Lehre von funktionellen Diagnostik des Reticuloendothel apparates. 2. Ges. Exptl. Med. 64, 257-270 (1920). W20. Williams, W. L., Vital staining of damaged liver cells, reactions to acid azo dyes. A n d . Record 101, 133-147 (1948). W21. Wilson, W. C., Blood volume in surgical disorders. Edinburgh Med. J . 67, 30-43 (1950). W22. Winkler, K., Urinary elimination of bromsulfalein in man. Scand. J . Clin. & Lab. Invest. 13, 44-49 (1961). W23. Winkler, K., The arterio-hepatic venous difference of bromosulphalein after a single intravenous injection and its use for the determination of the hepatic blood flow in man. Scand. J . Clin. & Lab. Invest. 16, 635-645 (1964). W24. Winkler, K., The kinetics of elimination of bromosulphalein in man after single injections and continuous infusion. I n “The Biliary System” (W. Taylor, ed.), pp. 551-566. Blackwell, Oxford, 1965. W25. Winkler, K., and Gram, C., Models for description of bromsulfalein elimination curves in man after single intravenous injections. Acta Med. Scand. 169, 263-272 (1961). W26. Winkler, K., and Gram, C., Kinetics of bromsulphalein elimination during continuous infusion in man. Acta Med. Scand. 178,439452 (1965). W27. Winkler, K., Larsen, J. A., Munkner, T., and Tygstrup, N., Determination of the hepatic blood flow in man by simultaneous use of five test substances measured in two parts of the liver. Scand. J . Clin. & Lab. Invest. 17,423-432 (1967). W28. Winkler, K., and Tygstrup, N., Determination of hepatic blood flow in man by cardio green. Scand. J. Clin. & Lab. Invest. 12, 353-356 (1960). W29. Winkler, K., Tygstrup, N., and Munkner, T., A study of Gaebler’s method for determination of bromosulphalein in plasma. Scand. J . Clin. & Lab. Invest. 12, 357-361 (1960). W30. Winkler, K., and Tygstrup, N., Methodologic study of the bromosulphalein method for hepatic bloodflow determinations. Scand. J . Clin. & Lab. Invest. 67, 413-422 (1965).

386

PAULA JABLONSKI AND J. A. OWEN

W31. Wirts, C. W., and Bradford, B. K., The biliary excretion of bromsulphthalein as a test of liver function in a group of patients following hepatitis or serum jaundice. J . Clin. Invest. 27, 600-608 (1948). W32. Wirts, C. W., and Cantarow, A., A study of the excretion of bromsulphthalein in bile. Am. J . Digest. Diseases 9, 101-106 (1942). Y1. Yudkin, S., and Gellii, S. S., Liver function in newborn infants with special reference to excretion of bromosulphalein. Arch. Disease Childhood 24, 12-14 (1949). Z1. Zatuchni, J., and Miller, G., Jaundice during chlorpromazine therapy. New. Engl. J . Med. 261, 1003-1006 (19%). 22. Zelman, S., The liver in obesity. A . M . A . Arch. Zntenzal Med. 90, 141-156 (1952). 23. Zieve, L., Hawson, M., and Bill, E., Studies of liver function tests. 11. Derivation of a correction allowing use of the bromsulfalein test in jaundiced patients. J . Lab. Clin. Med. 37, 40-50 (1951). 24. Zieve, L., and Hill, E., An evaluation of factors influencing the discriminative effectiveness of a group of liver function tests. 11. Normal limits of eleven representative hepatic tests. Gastroenterology 28, 766-784 (1955). 25. Zieve, L., Hill, E., and Nesbitt, S., Studies of liver function tests. I. Combined intravenous bromsulphalein-hippuric acid-galactose test. J . Lab. Clin. Med. 36, 705-709 (1950). Z6. Zipf, R. E., Webber, J. M., and Grove, G. R., A comparison of routine plasma volume method using radioidinated human serum albumin and Evan’s blue (T-1824) J . Lab. Clin. Med. 46, 800-805 (1955).

RECENT ADVANCES IN THE BIOCHEMISTRY OF THYROID REGULATION1

.

Robert D Leeper Endocrine Section. Memorial Hospital for Cancer and Allied Diseases. Sloan-Kettering Institute for Cancer Research. N e w York. New York

.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1.1. Outline of the Feedback Loop in Thyroid Hormone Regulation . . . . . . . 1.2. Format ................... ..................... 2 Thyrotropin-Releasing Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Isolation and Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Physiology and Biochemistry of TRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. TRF in Man . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Thyroid-Stimulating Hormones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Properties of TSH ... ......................... 3.3. Control of TSH Synthesis and Secretion. . . . . . . . . . . . . .. 3.4. Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Mechanism of Action of TSH .. ................................ 3.6. TSH in Thyroid Disease....... .................... 3.7. Long-Acting Thyroid Stimulator (LATS) ........................... 4 Thyroid-Binding Proteins ......... .................. 4.1. Introduction .............................. 4.2. Characteristics of Thyroid-Binding Proteins ..................... 4.3. Measurement of Thyroxine-Binding Proteins ....................... 4.4. Free Thyroxine ........ .................... 4.5. The Role of Thyroxine-Binding Proteins in Thyroxine Transport and Turnover .......................................... 4.6. Causes of Abnormalities in Thyroxine-Binding Capacity .............. 5 Calcitonin........................................................... 5.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Physiology of Calcitonin .. ................................... 5.4. Calcitonin in Man ............. .... ................... References........ ... .................. .......... 1

.

.

.

.

1

.

387 388 390 390 390 390 391 391 392 393 393 393 394 395 399 401 401 404 404 404 405 407 407 409 411 411 411 412 413 414

Introduction

Advances in the knowledge of thyroid physiology and biochemistry have been ao rapid and numerous in the past half decade that no single chapter could cover the field. The author must therefore apologize for obvious gaps in the discussion which follows. The topics to be discussed 'This work was supported in part by NCI grant CA 08748 387

.

388

ROBERT D. LEEPER

are arbitrary selections, since an attempt to cover all areas of thyroid research would constitute an injustice to each of the areas. I n the short span of six years, a new hormone has been identified in the thyroid gland, its structure elucidated, and a successful synthesis has been performed. The rapid identification and synthesis of the calcium-lowering principle (thyrocalcitonin) is only an example of the advances being made on several fronts in the thyroid field. Other advances, t o name a few, have been the partial purification and characterization of a hypothalamic thyrotropin-releasing factor (TRF) and the further purification of thyroid-stimulating hormone (TSH). Very early effects of TSH have been measured. The immunoglobulin nature of the long-acting thyroid stimulator (LATS), thought by many to be the principle which maintains hyperfunction in Graves’ disease, has been established. Rapid advances have been made in establishing the homeostatic role which various thyroid binding proteins in the blood play in maintaining free thyroxine levels. The development of methods to measure free thyroxine in the blood has been an important advance since it is likely that free thyroxine is the active moiety of blood thyroxine. The following pages will include discussion of various aspects of thyroid regulatory processes and a discussion of thyrocalcitonin. THE FEEDBACK Loop IN THYROID HORMONE REGULATION

1.1. OUTLINEOF

I n order to provide a framework for subsequent discussion, a brief outline of the currently accepted concept of the control of thyroid function is presented a t this point. The evidence to justify this brief synopsis will be presented in the subsequent sections. Figure 1 presents this concept graphically. Thyrotropin-releasing factor (TRF) is produced in the hypothalamus and arrives a t the pituitary via the hypophyseal portal blood system. TRF mediates the release of thyrotropin from the anterior hypophysis. The thyrotropin release is inhibited by thyroxine, presumably free thyroxine, and the inhibition is dose dependent. Thus excessive levels of thyroxine depress thyrotropin release, and lower thyroxine levels result in increased thyrotropin release from the anterior pituitary. TSH, in turn, stimulates thyroid hormone synthesis and secretion by the thyroid gland. Thyroxine and triiodothyronine are bound to specific binding proteins in the blood. The amounts and binding constants of the specific thyroid-binding proteins, together with the rate of thyroid hormone release from the thyroid, determine the amount of free thyroxine in the blood. Free thyroxine levels are determined not only by the rate of

BIOCHEMISTRY OF THYROID REGULATION

TRF

389

Y l A M ?

(Thyrotropin releasing factor)

"eL:

.,

~

"^

--

TISSUES

FIQ.1. Schematic outline of factors controlling thyroid hormone levels in blood and tissue.

hormone release from the gland, but also by the rate of removal of free thyroxine from the blood compartment by binding proteins and processes in tissues. Rapid alterations in the level of the specific thyroxine-binding proteins also affect free thyroxine levels. Total thyroid hormone levels in the blood are thus a function of the binding capacity of thyroid binding proteins in the blood, and total thyroid hormone levels do not necessarily reflect the true metabolic state. Compounds or conditions which alter the amount or binding affinity of the specific thyroid-binding proteins may alter the total thyroid hormone content of the blood without necessarily changing free thyroxine levels. To complete the loop then, free thyroxine controls the rate of release of thyrotropin. The proportionate influence which each of the above control points has in overall thyroid hormone hemostasis is of course still controversial, but the above system serves well as a point of departure for subsequent discussion.

390

ROBERT D. LEEPER

1.2. FORMAT

A brief background will be given and pertinent reviews quoted in each of the succeeding sections. Methods of isolation and extraction of various hormonal substances, will not be described, but references to such procedures will be given. An attempt will be made to confine the discussion to human and clinical studies, but, gaps will be filled with animal data where necessary. Assays will be discussed in principle and evaluated as they apply to clinical situations or investigation. Three general references are listed (C2, G4, G5). 2.

Thyrotropin-Releasing Factor

2.1. INTBODUCTION

Experiments reported by Harris (H6) in 1955 clearly foretold that the function of the pituitary gland was regulated by the hypothalamus. Harris and other investigators showed that if the ventromedian nuclei and the paraventricular nuclei in the hypothalamus were destroyed, thyroid function was depressed. Greer (G3) described experiments in which rats, maintained on propylthiouracil (a goitrogen) , did not develop goiters if specific hypothalamic areas were destroyed. These studies indicated that the pituitary could not respond normally to decreased thyroid hormone levels in the absence of certain hypothalamic stimuli. Subsequent studies have shown that substances can be isolated from the hypothalamus of man as well as many other species, which release TSH from the pituitary both in vitro and in vivo. Although the hypothalamus is required for normal thyroid gland function, little is known of the factors that control the formation and release of TRF. The scarcity of TRF has precluded many investigations in man, and most of the following discussion refers to animal experiments except where indicated. Guillemin has recently reviewed this subject (G7).

2.2. ISOLATION AND CHARACTEBIZATION Guillemin et al. (G9)reported th at they had extracted and purified a substance from the hypothalamus of sheep which stimulated thyroid function in the rat. Others showed that crude hypothalamic extracts would stimulate thyroid function in mice and rabbits, and Schreiber (SS) reported that a crude extract of bovine hypothalamus would release TSH from r a t pituitaries in vitro. Schally et al. (53) obtained a more purified TRF factor from 100,OOO porcine hypothalami. TRF is apparently a small molecular weight substance and, although originally thought to be a protein (G8, S2), more recent evidence suggests that

BIOCHEMISTRY OF THYROID REGULATION

391

it may not be a simple polypeptide (G7). The material is inactivated by incubation in serum but is not destroyed by proteolytic enzymes (G7, S3). TRF has not been found in the peripheral blood of man. It has been identified in hypophyseal portal blood of rats after electrical stimulation of the hypothalamus ( A l l ) . 2.3. ASSAY All assays require hypothalamic extracts; these can be crude or purified. Tests can be performed either in viva or in vitro. Assays depend upon the ability of T R F to release TSH from the pituitary, and the TSH is measured by in vivo or in vitro methods. 2.3.1. In Vivo Assay (R4) Mice are pretreated with l3II to label the thyroid, 1 pg thyroxine, and codeine prior to injection of the material to be tested. The test material must be free of TSH, vasopressin, and MSH, which exert effects directly on the thyroid (R4). Codeine is given to “sensitize” (R4) the pituitary, and the small dose of thyroxine is given to partially suppress endogenous TSH secretion and establish basal conditions. Blood I3lI levels obtained 2 hours after injection are proportional to the amount of T R F injected (R4). Plasma TSH can be measured directly by use of the McKenzie assay ( M l l ) , but this requires another group of animals. Pituitary TSH depletion can be measured directly by biossay of TSH in the pituitary of treated mice (S3). TSH assays will be discussed in Section 3.4. 2.3.2.

In Vztro Assays (B14, G10, S6, Sll, S15)

Pituitary glands are incubated in tissue culture medium and TSH measured in the medium after introduction of the TRF. The presence of calcium in the medium is necessary (V1). TSH is measured either by bioassay or more recently by radioimmunoassay (W2). In vitro methods require only a few rat pituitary glands, and incubation times are short. TSH release into the medium is a linear function of the log dose (G10). Crude hypothalamic extracts can be assayed in this way, since contaminating vasopressin does not stimulate TSH release from the pituitary (R2). This latter point has been recently questioned, however, by Krass et al. (K5). Vasopressin may stimulate TSH release in vitro. 2.4. PHYSIOLOGY AND BIOCHEMISTRY OF T R F

TRF is formed in the hypothalamus and reaches the pituitary via the hypophyseal portal system. The mechanisms whereby T R F secretion or release is controlled are not known. Electrical stimulation of the hypothalamus causes increased amounts of TaF to appear in the hy-

392

ROBERT D. LEEPER

pophyseal portal system ( A l l ) . Although increased aniouiits of TRF are found in the hypothalamus of hypothyroid rats ( S l l ) , T R F levels in the hypothalamus are not depressed by thyroid hormone administration. Studies using both in vivo and in vitro T R F assay systems indicate that thyroxine exerts its regulatory effect on TSH a t the level of the pituitary. I n vitro, low concentrations of thyroxine will partially inhibit the effect of TRF (W2), and higher concentrations (in the order of 10 &mi) will completely inhibit the T R F effect (W2). Such concentrations are extremely high when it is considered that free thyroxine conmole. This discentration in blood cannot be much higher than 1 X crepancy between in vitro concentrations and in vivo doses is usually seen in in vitro studies using thyroxine. As more T R F is added to in vitro or in vivo assay systems, more thyroxine is required to block TSH release, thus suggesting a competitive effect. In vivo administration of thyroid hormone requires lesser amounts of hormone to inhibit TSH release (B15, R2, V2). Actinomycin D and puromycin (B12, B13, R3) block thyroxine inhibition of TSH release after T R F stimulation but do not block the release phenomenon. This observation suggesb that the ability of thyroxine to block T R F depends on protein synthesis. There are conflicting reports about the ability of cycloheximide to block the T4 effect (B13, W2). Ouabain had no effect in one study, and oligomycin and 2,4-DNP had partial effects (W2). Injection of mammalian T R F into tadpoles did not elicit a metamorphic response (E7). Since thyroid hormone causes metamorphosis in the tadpole, this study indicates species specificity or a different neurophypophyseal relationship in the tadpole. Sakiz and Guillemin (Sl) reported that T R F increased TSH secretion and decreased blood cortisol, suggesting inverse relationships between ACTH and TSH production. It will be of interest to study this inverse relationship between ACTH and TSH production in man, and it is now more feasible with the newer radioimmunoassays.

2.5. T R F

IN

MAN

The scarcity of T R F has precluded many studies in man. T R F has not been found in blood. Bowers et al. (B16) reported that the administration of 300 pg of purified porcine T R F to man increased TSH in blood within 3 minutes. Maximum response was reached a t 30 minutes, and the response lasted 2 hours. TSH was measured by bioassay and a radioimmunoassay. Cretins were used as subjects to eliminate nonspecific thyroid responses. Specific thyroid syndromes involving the hypothalamus in man have not been established as yet, but such syndromes will undoubtedly be described in the future.

BIOCHEMISTRY O F THYROID REGULATION

3.

393

Thyroid-Stimulating Hormones

3.1. INTRODUCTION

I n 1922, Smith and Smith (S12) found a substance in bovine pituitary glands which activated the thyroid of hypophysectomized tadpoles. For many years this finding was the basis for the assay of what has come to be called thyroid-stimulating hormone (TSH) . The mechanism by which TSH stimulates the thyroid to organify iodine and secrete thyroxine has received much attention in recent years. Since the initial discovery of TSH, other thyrotropic substances have been described. Some thyrotropin-like substances do not originate in the thyroid. One such substance, long-acting thyroid stimulator (LATS), a thyrotropin found in the serum of patients with Graves’ disease, has been the object of intense investigation. LATS, first described by Adams and Purves ( A ll A2), apparently does not arise in the pituitary. The immune globulin nature of LATS and the ability of anti-immune globulin antibody to inhibit its activity have prompted many workers to suggest that the hyperthyroidism of Graves’ disease is an autoimmune phenomenon. The isolation of highly purified human TSH by Condliffe ((36) stimulated comparative studies of the properties and antigenicity of human TSH (h-TSH) and bovine TSH (b-TSH) . The isolation of h-TSH made possible the development of sensitive radioimmunoassays. Radioimmunoassays have been useful in studying TSH levels in human blood in health and disease. The findings that TSH stimulated the conversion of g l ~ c o s e - l - ~to~ C 14C0, in the thyroid and the incorporation of 32Pinto thyroidal phospholipids have become valuable tools in studying the mechanism of action of TSH. Even more recently, studies by Pastan (P4) and others indicate that the earliest effects of TSH may occur a t the level of the cell membrane. The initial effects of TSH occur in the presence of inhibitors of protein synthesis, suggesting that such effects are not dependent on new protein formation. McKenzie (M8) and Condliffe and Robbins (C7) have recently reviewed the subject of TSH. Additional recent and valuable information is contained in the Proceedings of the Third International Congress of Endocrinology (G4) and the Proceedings of the 5th International Thyroid Conference in Rome (1965) (C2). OF TSH 3.2. PROPERTIES

Human TSH has been assigned a molecular weight of approximately 28,000 (C7), which is similar to the molecular weight of bovine TSH (F4).

394

ROBEBT D. LEEPER

Highly purified material gives a single line of identity by Ouchterlony gel diffusion (U2). Unabsorbed h-TSH antibody contains binding sites for gonadotropins and luteinizing hormone (02). Purified h-TSH is separated into several bands by gel electrophoresis (C7), some of which have activity. Human TSH is found just behind albumin when blood is filtered on a Sephadex G-200 column (Ul). Human TSH can also be separated from LATS by gel filtration (M5). The assays to be discussed in the following sections are based on bovine TSH (standards), and most studies of the biochemical effects of TSH have involved the use of b-TSH. International standards are available for both b-TSH and h-TSH assays, and these may be obtained from the British Medical Research Council. The response to a particular TSH varies with the type of assay (Bl) . Therefore, when TSH is assayed, dose response curves must be based on a reference standard. Species variation is valuable in identifying differences between TSH and other TSH-like material. LATS, for example, does not stimulate thyroid function in the chick (L3). Immunological cross reactivity between TSH’s from different species is of some importance since several reports are based on the ability of antibodies to the TSH of one species to inhibit the biological or biochemical effect of TSH from another species. Cross reactivity also has pertinence in the development of immunoassays. Highly purified h-TSH and b-TSH showed no cross reactivity by gel diffusion methods and radioimmunological precipitation techniques (U2). However, in several bioassay procedures limited cross reactivity has been shown between anti-b-TSH antibodies (L4, M9, U2) and human TSH. A radioimmune assay described by Lemarchand-Beraud (L2) is based on the use of anti-b-TSH antibodies developed in rabbits. Recently cross reactivity has been reported between h-TSH and antibodies to porcine TSH (F9). Radioimmunoassay procedures were used in these studies. It would appear that there may be two groups of antigenic sites on human TSH, one specific and one shared in common with b-TSH and p-TSH. 3.3. CONTROL OF TSH SYNTHESIS AND SECRETION The mechanism by which TRF controls the release of TSH from the pituitary is discussed in Section 2.4. It has been known for several years that the administration of thyroxine (TJor triiodothyronine (T3) would inhibit the uptake of radioiodine by the thyroids of normal individuals. The administered hormone presumably acted by decreasing the secretion of TSH by the pituitary. This action is the basis of the T, suppression test described by Werner ( W l ) . T, is given to the patient for 7 days. I n the normal individual, l S 1 I uptake by the thyroid will be

BIOCHEMISTRY OF THYROID REGULATION

395

decreased by 50% or more, as compared to pre-T, uptake. The uptake of l3II by the thyroids of patients with Graves’ disease is not inhibited by the same amount of T,. This is important evidence that Graves’ disease is not caused by excess suppressible TSH secretion. Little is known about the synthesis of TSH. Direct measurement of serum TSH by radioimmunoassay techniques (R6, U l ) or in Gtro or in vivo bioassays (A3, K3) showed that TSH in blood decreased when thyroid hormone was given to hypothyroid persons. Free thyroxine levels were more directly related to the changes in TSH concentration (R6). The half-life of TSH in blood as measured by the use of I3lI-labeled h-TSH is about 30-60 minutes (02, 03). 3.4. ASSAY With each report of a new biochemical effect of TSH on the thyroid gland, a new assay for TSH has been developed. Many of the assays have withstood the test of time. It is not within the scope of this chapter to discuss all the assays which have been developed over the years. Rather we will discuss recent progress in this field and describe those methods that are particularly useful in clinical situations. Kirkham’s review (K3) is helpful on the subject of TSH assays. Methods currently in use involve either bioassay procedures or immunological techniques. The bioassay methods can be classified as gravimetric, biochemical, or histiometric, depending on which end point is measured. Bioassays are performed either in vivo or in witro. Table 1 shows the characteristics and sensitivity of some of these assays and the amount of TSH found in blood by each method. It is important in any assay that the dose response curve of the unknown be similar to the response curve of the TSH control. As in most bioassays, there are a number of problems that interfere with sensitivity and specificity. Among these problems are the amount of material that can be used, nonspecific substances in blood that interfere with the parameter being measured, species differences, and toxic substances (A4). Biological variability requires complex mathematical analysis in evaluating data derived from some of the in vivo tests. Despite these drawbacks, much of our information on the biological effects of TSH has been gained from data derived from these assays. The radioimmunoassays developed in the past few years will undoubtedly give more precision to the measurement of TSH in body fluids. A brief description of some assays follows. Units will be expressed in terms of the b-TSH International Standard. 3.4.1. Bwassays in Vivo 3.4.1.1. Increase in Protein-Bound Radiowdine in Plasma. The general method is undoubtedly the most widely used TSH assay a t this

TABLE 1 SOMECHARACTERISTICS OF TSH ASSAYAND TSH LEVELSIN BLOOD^ Lower limit of sensitivity Method Bioassay in vivo Colloid droplet 1311 release in mouse l*lI depletion Bioassay in vitro Weight change

Reference

D2 M4 B3 B2

13'1 depletion

K3

lalI discharge

El

Animal

(mu)

Guinea pig Mouse Chick

0.02 0.025 0.5

Bovine thyroid slice Guinea pig thyroid slice Guineapig thyroid slice

0.008 0.007 0.05

Normal

(mU/ml) 0.25-0.5 0.166

-

-

02

-

0,002

< O . 01-0.05

Immunoassay

L2

-

-

0.38

~~~~

~~~~~

Hyper (mU/ml)

50 0.12-0.64 -

Low in 4 cases Delayed

-

-

0.06-2.3

Comment

Concentrated sera

-

Concentrated serum

T1W M

-

E

<0.01-0.68

-

r

-

-

i

-

0.13 Males, 0.01-0.17 0.068 females -

Immunoassay

(I

HYPO (mU/ml)

P n

m

Not detected

1.0-8.8

~

Sensitivity and blood levels are expressed as milliunita compared to bovine TSH standard.

0.26

h-TSH values changed to b-TSH on basis of 10U/mg h-TSH Assay based on cross reaction of h-TSH and b-TSH antibody. Note higher values

BIOCHEMISTRY OF THYROID REGULATION

397

time. As first described by Adams and Purves (A2), 1311 was given to guinea pigs to label the thyroid gland. The pituitary was then blocked by injecting the animal with thyroxine, and 1 day later TSH or serum was injected into the ear vein. The rise in serum protein-bound lSII was measured after discharge of lSII from the thyroid. The method was used in the discovery of LATS (see below). McKenzie (M4) modified the method for use in mice, which were easier to maintain and handle. I n the current version (M11) of the McKenzie assay, mice are fed a low-iodine Remington diet for 10 days. The mice are then injected with 15 pCi of lZ5I and 10 pg of sodium thyroxine. Four days later solutions of the sample to be tested are given by intravenous injection. Volumes of 0.5 ml can be given. At 2 and 9 hours after sample injection, retroorbital blood specimens are taken with a capillary pipette and radioactivity is counted in a well-type scintillation counter. The TSH response occurs in 2 hours, and the LATS effect is measured a t 9 hours. Low levels of activity give poor discrimination in this assay and data are subjected to covariance analysis after logarithmic transformation of the raw radioactive counts (M11). The McKenzie method is widely used and, although subject to variability from laboratory t o laboratory, results obtained in each laboratory are usually internally consistent (Bl). The advantage of the McKenzie method is its simplicity, and of course its ability to detect LATS. Adams et al. (A4) have described some of the nonspecific responses in the use of this assay for TSH and LATS. They emphasize the necessity for selection of proper control solutions. 3.4.1.2. Depletion of Iodine in Chick Thyroid. The Bates (B3) assay is useful in that it can discriminate between LATS and TSH. Radioiodine is given to label the thyroid gland, and the pituitary is blocked with thyroxine. Twenty- four hours later propylthiouracil is administered to prevent uptake of 1311 recirculating after thyroxine degradation. TSH or test material is given in graded doses, and the depletion of 1311is measured by external counting as compared to control uptakes. Serum must be concentrated in this method to assay normal levels of TSH. It is important to use chicks whose thyroids take up more than 15% of the I3lI dose (B3). 3.4.2.

Bwassays in Vitro

3.4.2.1. Gravimetric Assay. I n vitro methods are more sensitive than in vivo methods, but the dose response curves are not as steep, and thus they are less precise. Baake et al. (B2) measured the weight of beef thyroid slices incubated for 24 hours in Krebs-Ringer phosphate buffer.

398

ROBERT D. LEEPER

The method is based on the fact that TSH prevents the weight loss which normally occurs if TSH is left out of the medium. 3.4.2.2. Biochemical Assays. I n vitro biochemical methods are based on the fact that TSH will release 1311 from thyroids previously labeled with I3lI by incubation of guinea pig thyroid slices with the labeled iodine (B11). E l Kabir (El) improved the method by treating the guinea pigs with a goitrogen prior to removal of the thyroids. Kirkham (K2) described a very sensitive bioassay. Guinea pigs are maintained on a goitrogenic regime for 100 days; the thyroids are then removed, diced, and incubated in a culture medium with 1311. TSH is added and another incubation is performed. After the second incubation an aliquot of the medium is removed and counted for radioactivity. The aliquot is replaced by KSCN and incubation continues for 4 hours, after which a second aliquot is counted. TSH increases the organification of iodine in this system, and thus the amount of nonorganic iodine discharged by KSCN into the medium is inversely proportional to the amount of TSH in the medium. 3.4.2.3. Histimetric Assay. TSH rapidly causes the formation of colloid droplets within the thyroid follicle, and this is the basis of one of the earlier TSH assays (D2). The method is not as precise a s others listed, but is useful to confirm the presence of TSH.

Radioimmunoassay The fundamental studies of Berson and Yalow (B8) have permitted the establishment of methods which measure minute amounts of several hormones. TSH is no exception. When highly purified preparations of h-TSH became available for use as antigens, methods were rapidly developed for the radioimmunoassay of TSH in human sera and other fluids. Radioimmunomethods depend on the ability of unlabeled antigen to displace antigen labeled with 1311 from specific antibody. Reaction mixtures are set up containing constant amounts of specific anti-TSH antibody and various concentrations of TSH of known potency or samples containing unknown amounts of TSH. A tracer amount of purified 1311 labeled h-TSH is added and the degree of dispIacement of labeled TSH from the specific antibody is quantitated by counting radioactivity after separation of free and bound label. The amount of radioactivity displaced from antibody by unlabeled TSH is proportional to the concentration of TSH in the sample or standard. Methods vary in the way unbound or free labeled TSH is separated from IabeIed TSH bound to antibody. Radioimmunoelectrophoretic techniques separate the bound and unbound label by chromatoelectrophoresis ( U l ) , Unbound label remains 3.4.3.

399

BIOCHEMISTRY OF THYROID REGULATION

a t the origin and bound label migrates with antibody. Ode11 et al. (02) precipitate antibody-bound TSH with antirabbit y-globulin antibody. Bound label can also be precipitated with ethanol-saline (01). All methods give approximately equal results and are sensitive down to about 2 10-6 units (02) (bovine) based on chick assay of the starting purified human TSH. Radioimmunoassays are undoubtedly a great advance in the measurement of TSH. An assay based on the use of antibodies to the TSH of some other species would be most helpful, but a t this time there is still considerable confusion about the degree of cross reactivity which may exist between TSH preparations obtained from various species. After proper adsorption, antibodies to relatively crude preparations of human TSH can be used, but greater quantities of purified h-TSH must be available for labeling before these methods can be introduced into general use. Unfortunately, purified h-TSH is not completely stable and loses activity on storage (C7).

x

3.5. MECHANISM OF ACTIONOF TSH The fact that TSH stimulates the trapping and organification of iodine was well known for many years, but the mechanisms by which this occurs have been difficult to elucidate. Several recent findings promise to increase our knowledge. Field et al. (Fl) made the important observation that the addition of TSH to incubating dog thyroid slices caused a n increase in the rate of oxidation of glu~ose-l-~"C to 14C02. Dumont (D8) stimulated this pathway with iodotyrosines. Recent studies suggest that this stimulation of the hexose monophosphate shunt is mediated through increased levels of cyclic 3',5'-AMP. Sutherland (S25) has postulated that many if not all of the hormones exert their effects by altering cyclic 3',5'-AMP levels within the cell. The cyclic AMP effect may occur a t the level of the cell membrane. It is not the purpose of this chapter to discuss the possible role of cyclic AMP in hormone action, and for more details the reader is referred to reviews by Sutherland (52.5). Suffice i t to say that one of the earliest effects seen with TSH is a stimulation of the cyclic AMP system. I n 1962 Klainer et al. (K4) reported that TSH increased the amount of cyclic AMP in incubated sheep thyroid particulate fractions by 2535%. Gilman and Rall (Gl) reported that TSH increased the level of cyclic AMP in incubated bovine thyroid slices. Cyclic AMP itself is not a good stimulator of COz production from glucose in some thyroid preparations. This may be due to inactivation of cyclic AMP by phosphodiesterase (P6). However, an analog of cyclic AMP, dibutyryl cyclic AMP, will enhance the conversion glucose-l-14C to l4COz ( P l ) The dibu-

.

400

ROBERT D. LEEPER

tyryl analog resists hydrolysis by phosphodiesterase (P6).Pastan (P2) reported that TSH activated adenyl cyclase in thyroid homogenates within 30 seconds. Adenyl cyclase converts ATP to cyclic AMP. There is some question (B19) as to whether the dibutyryl analog will augment the effect of suboptimal concentrations of TSH. Augmentation should occur if the same pathway is involved. Ensor and Munro (E5) have reported that cyclic A M P itself will mimic the effect of TSH on iodine release from a cultured mouse thyroid preparation. Another important finding which bears on the mechanism of action of TSH was that of Bore11 (BlO), who observed that TSH increased the turnover of phosphorus by the thyroid. This observation was pursued, and i t was found that 32Pincorporation into phospholipids was increased by TSH (F8, S9). Dibutyryl cyclic AMP also stimulates the incorporation of 32Pinto phospholipid (Pl). Since phospholipids are in part a constituent of cell membranes, the findings suggested that TSH might act on thyroid cell membranes. Pastan (P4) showed that TSH was quickly bound to thyroid cell membranes. Antibody to TSH inhibited the subsequent effects of TSH on glucose oxidation if given early, but if added later did not abolish the chain of events which had been initiated. TSH could not be removed from the membranes by repeated washing. The results suggested that the site of action of TSH was a t or near the cell membrane. Phospholipase C (Ml) also inhibited the increased glucose oxidation caused by TSH, again suggesting that an intact membrane is a requirement for TSH action. It was thought that phospholipid C was too large to enter the cells and therefore must have acted on the membrane. Whether or not the effects of TSH on phospholipid metabolism and on glucose oxidation are linked is open to question, since some evidence suggests that the two effects may be dissociated (A8, B18). Epinephrine stimulated glucose oxidation while inhibiting 32Puptake into phospholipids in incubating dog thyroid slices (A8). Recently Pastan (P3) reported that fluoride, which stimulates the formation of cyclic AMP, increased the rate of glucose oxidation and phospholipid turnover in thyroid slices. However, it did not increase colloid droplet formation in the follicles. Colloid droplet formation is a characteristic action of TSH (D2). Although TSH may indeed stimulate the hexose monophosphate shunt, its specificity must involve other pathways as well. TSH also increased pyrimidine nucleotide synthesis in the bovine thyroid (H2, L6). Hall and Tubman postulated that increases in nucleotide synthesis may be mediated by increased ribose generated from the hexose monophosphate shunt. Incorporation of labeled formate and adenine into nuclear, cytoplasmic, or whole RNA was not inhibited by puromycin a t a time when protein synthesis was inhibited as measured

BIOCHEMISTRY OF THYROID REGULATION

401

by leucine-I4C incorporation into protein (H3, L7). Thus protein synthesis was not necessary for TSH to stimulate RNA synthesis. RNA synthesis also occurs in the presence of actinomycin D and cyclohexemide. McKenzie e t al. (M12) have shown that these agents do not inhibit 1311 release from the mouse thyroid in vivo unless given 6-8 hours before TSH. Protein synthesis was decreased by 2 hours, indicating that the release of I3lIwas not dependent on new protein formation. Whether or not the primary TSH effect is a t the cell membrane is still a question which requires further study. However, if the fundamental effect of TSH is related to its earliest known actions, then we are much closer to finding the main site of action.

3.6. TSH IN THYROID DISEASE

If the feedback hypothesis of thyroid hormone regulation is correct, TSH levels in the blood of hypothyroid people should be increased. Several reports (Table 1) confirm this, including reports based on studies using radioimmunoassay techniques. TSH levels decline when patients are treated by thyroid hormone replacement (02). This does not apply when the hypothyroidism is secondary to pituitary dysfunction. The feedback hypothesis would suggest that TSH levels would be decreased or absent in hyperthyroidism unless hyperthyroidism is caused by excess TSH secretion. Reports conflict as to whether or not TSH blood levels are increased in Graves’ disease. Much of the conflict arises from the fact that the earlier reports may not have taken LATS into account. Most recent studies suggest that TSH levels are not elevated in hyperthyroidism (Table 1 ) . Since no assay to date is sensitive enough to establish a lower limit of normal for TSH in blood, i t has not been possible to determine whether TSH was absent in the blood of patients with hyperthyroidism. The finding that TSH levels are a t least normal in hyperthyroidism is important evidence that TSH itself is not the cause of thyroid overactivity. TSH-like activity has been found in the blood or tissues in some patients with choriocarcinomas (02, 0 4 ) , and hyperthyroidism has been reported in this disease. The TSH-like activity found in choriocarcinomas does not cross react immunologically with purified h-TSH (02). 3.7. LONG-ACTING THYROID STIMULATOR (LATS) The finding by Adams and Purves (Al, A2) that there was a substance in the blood of some hyperthyroid patients which gave a delayed maximal response in the guinea pig TSH assay was important, since it suggested that an abnormal thyroid stimulator was present. The finding was confirmed by McKenzie (M4) and by others. Subsequent investigation

402

ROBERT D. LEEPER

has shown that LATS cannot be found in the pituitary gland (D3, M7), and that it probably is not TSH, but rather an IgG globulin. These latter reports have prompted the speculation that Graves’ disease may be some form of autoimmune phenomenon.

3.7.1. Characteristics of LATS As defined by McKende ( M l l ) , LATS is a substance found in blood which causes a maximal 1311response a t 9 hours instead of a t 2 hours when serum is injected into appropriately prepared mice. At present LATS can be measured only by this delayed response. TSH causes a peak response a t 2 hours. It has not been possible to demonstrate LATS in human pituitaries (D3, M7). LATS persisted in a few patients with Graves’ disease after hypophysectomy (B4). Some patients remain hyperthyroid after hypophysectomy. This is indirect evidence of a nonpituitary source of LATS. Persistent thyroid function following hypophysectomy is also evidence that Graves’ disease is not caused by excess TSH. Much evidence indicates that LATS is a 7 S y-globulin. Kriss et al. (K6) were able to isolate LATS activity from the 7 S y-globulin fraction and to inhibit this activity with IgG antibody. Meek et al. (M13) showed that the H chain of the 7 S globulin contained the long-acting activity, and that papain treatment of the LATS-containing chain rendered the LATS short acting. Dorrington et al. (D4) also confirmed the 7 S character of LATS and showed that the short-acting fragment derived from proteolysis was not inhibited by anti-TSH antibody. The fragment was inhibited by anti-IgG antibody. Lymphocytes (M7) taken from appropriate LATS donors can be stimulated to produce LATS by phytohemagglutinin. The immunoglobulin nature of LATS would suggest that it might have a higher molecular weight than h-TSH, and therefore might have a longer half-life in blood. This is in fact the case. I n neonatal Graves’ disease, a self-limiting disease which apparently is transmitted by LATS from mother to fetus, LATS is found in the newborn’s blood for up to 2 months (M6). Two months is the usual duration of the disease. Sunshine et al. (S23) showed that the half-life of long-acting thyroid stimulator in neonatal Grave’s disease is about 6 days, compared t o a halflife for h-TSH of about 60 minutes (03). The origin of LATS is obscure. As stated above, it has not been possible t o detect LATS-like material in the pituitary. Studies have been done which showed that LATS is inactivated by thyroid fractions (D5, E2, K6). Other studies have shown that rabbits immunized with thyroid fractions develop LATS-like activity in the blood and increased

BIOCHEMISTI1Y OF THYROID REGULATION

403

PBI’s (M7, S14). The rabbits develop a thyroiditis, which might explain some of the increased PBI, but does not explain the presence of LATSlike material. Recently it was reported that an abnormal TSH was found in the pituitary glands of patients who had died with active Graves’ disease (K7). The TSH found in these patients gave a typical early response by bioassay, but the response was not inhibited by anti-hTSH or anti-b-TSH antibodies. The report must be confirmed and the finding pursued, for it indicates that an abnormal TSH occurs in the pituitaries of patients with Graves’ disease, which in the blood might be bound to IgG globulin. The dose response curve for the abnormal TSH in mouse assay was similar to the curve produced by normal TSH. However, P-MSH or vasopressin effects were not excluded in the study. Neither were anti-IgG antibodies tested against the thyroid stimulator. Mechanism of Action. TSH and LATS cause similar effects on glucose oxidation, 32Puptake (F2) , and iodine release (E6). The time cowse is delayed, presumably due to the molecular size of the LATS. Antihuman TSH antibody did not inhibit the effect of LATS on 32P uptake into phospholipids or on glucose oxidation. As stated above (M13) proteolytic digestion of LATS makes it a short-acting thyroid stimulator. 3.7.2. LATS in Disease LATS has not been found in patients with hyperthyroidism caused by autonomous toxic nodules (M&). Not all patients with Graves’ disease have LATS in their blood (D3, M8), but the failure to find it may be the fault of the bioassay used to detect LATS. Serum rich in LATS gives a long-lasting LATS-like response when given to individuals with no long-acting substance in the blood (A10). Although it was originally thought that LATS was associated with the ophthalmopathy of Graves’ disease, i t is now felt that there is no causal relationship (M10). Whereas thyroid tissue will inactivate LATS, retroorbital tissue does not have this property ( S 5 ) . The relationship of LATS to other autoimmune disease, including Hashimoto’s thyroiditis, is also unclear. LATS (W5) has been demonstrated in the blood of a few patients with Hashimoto’s thyroiditis and ophthalmopathy. Some of the patients did not respond normally to the T, suppression test. These findings are suggestive of, but do not prove, relationship between autoimmune thyroiditis and Graves’ disease. Whether or not Graves’ disease is autoimmune in nature remains for the future to tell. If such is the case, however, the finding will have implications in other diseases characterized by endocrine gland overactivity. The blood and pituitary gland of some patients with Graves’ disease contains a substance called exophthalmic producing substance (EPS)

404

ROBERT D. LEEPER

which causes proptosis in the fundulus. There have been conflicting reports as t o whether or not this factor can be separated from LATS and/or TSH.Dobyns et al. has recently discussed these findings (D2a). 4.

Thyroid-Binding Proteins

4.1. INTRODUCTION The form in which thyroid hormone circulated in the blood was not well understood until the discovery of a specific thyroid-binding globulin was reported by Gordon et al. (G2) and two other groups (LI, R8). I n 1958, Ingbar (12) reported that a protein which migrated ahead of albumin on electrophoresis also bound thyroxine. There was considerable controversy about the significance of the prealbumin as a physiological thyroid-binding protein until it was realized that the binding of the hormone was inhibited by the then commonly used barbital buffer. On the basis of theoretical considerations, Robbins and Rall (R9) postulated that a small free thyroxine fraction would be the governing influence on thyroid hormone turnover and metabolic action. Earlier Recant and Riggs (Rl) had suggested that free thyroxine might be important; they based this on their studies of the discrepancy between the thyroid status and the PBI in patients with nephrosis. Sterling and Hegedus (SlS) first reported the actual measurement of free thyroxine in human serum, and several improved methods have since been reported. The availability of methods to measure free thyroxine levels made i t possible to test the role that thyroxine binding and free thyroxine play in regulating the thyroxine availability to the tissues. The later studies have shown that tissue binding sites, especially in the liver, must play an additional important role in the turnover of thyroid hormones and presumably in their metabolic effects. Oppenheimer ( 0 5 ) , Robbins and Rall (RIO), and Ingbar (14) have recently reviewed the thyroidbinding globulins. OF THYROID-BINDING PROTEINS 4.2. CHARACTERISTICS Two specific proteins are responsible for binding most of the thyroid hormones in blood. Thyroid-binding globulin, the first reported, migrates in the interalpha area on electrophoresis. Seal and Doe ( N O ) reported the isolation and purification of TBG in 1964. It is an acidic glycoprotcin with a molecular weight of between 50,000 and 60,000.These workers reported an amino acid composition for their preparation. Preliminary studies in a more recent report assigned a molecular weight of 80,OOO to TBG (S16). The binding of thyroxine to TBG is specific, and some related analogs of thyroxine are bound much less strongly (16). Of

405

BIOCHEMISTRY O F THYROID REGULATION

TABLE 2 SOMBVALUES FOR BINDING PROTEINS AND FREETHYROXINE IN NORMAL SERUM

Percent total thyroxine each protein Binding capacity (pg/thyroxine 100 ml serum) Protein concentration

Albumin

TBG

TBPA

57.4 f 6 . 7

32.0 f 5 . 4

-

15.0

-

20 k 3 24.5 f 2 . 6

256 f 38

-

27.7

10.6

* 1.3

-

-

Reference 011 w3 R9 011 010

(mg/100

Free thyroxine Nanograms per 100 d

Percent total thyroxine

4.21 f 0.76 2.34 k 0.29 4.03 f 1.08

0.046 & 0.005 0.0277 f 0.0029 0.050 3 ~ 0 . 0 0 9

S17 011 I6

particular importance is the fact that triiodothyronine (Ta), the other naturally occurring hormone secreted by the thyroid, is bound much less strongly by TBG (I6), and not at all by TBPA (13). TBG concentration in blood is probably about 1.5 mg/100 ml (05). Thyroid-binding prealbumin (TBPA) has been easier to isolate and purify because it can be separated from other proteins by electrophoretic methods and because it is present in greater quantities (Table 2). Purified TBPA is immunologically pure as well as homogeneous on ultracentrifugation and electrophoresis (013). The molecular weight is about 73,000 (03, S10). TBPA has been crystallized (P8). The binding properties of TBPA for thyroxine are specific, but different from those of TBG. TBPA does not bind T a (13) but binds the propionic and acetic acid analogs of thyroxine much more strongly than it does thyroxine (R11). TBPA also binds vitamin A (A9). Albumin also binds thyroxine (RlO), but its binding affinity is considerably less than is that of TBG, and therefore albumin plays less of a role in thyroxine transport. 4.3. MEASUREMENT OF THYROXINE-BINDING PROTEINS

Serum TBPA can be measured directly since it is present in sufficient amounts and can be separated from other proteins by electrophoretic methods (010). It is not practical to measure serum TBG directly because of the small quantities involved and because separation from

406

ROBERT D. LEEPEB

albumin is difficult. A pure preparation of TBG would make the development of a radioimmune assay possible. An indirect measure of the concentration of a specific binding protein can be obtained by adding increasing amounts of thyroxine to serum which has been partitioned by electrophoretic means. Saturation, or the maximal binding capacity is a function of the number and capacity of binding sites available [l/mole (013) in the case of TBPA]. The concentration of binding sites is proportional to the amount of thyroxine necessary t o saturate the protein. Such methods have been described ( 0 1 0 ) . Reverse-flow techniques can be used to eliminate trailing albumin from the TBG (E3). The distribution of thyroxine among the various binding proteins is also of importance, and the percentage of total thyroxine attached to each species of binding protein can be obtained by paper electrophoresis of serum to which 1311thyroxine has been added ( 0 1 1 , S19). Both the maximal binding capacity and the distribution of thyroxine among the binding proteins are a function of the composition and the p H of the buffers used (13). A useful buffer is glycine-acetate, p H 8.6. Because of the artificial nature of the buffer and p H used in the determination, the distribution of thyroxine in vivo may not be accurately reflected by the electrophoretic data. In particular, the physiological role of TBPA (C4)in binding thyroxine in vivo has been questioned, but changes in the concentration of TBPA have been associated with alterations in free thyroxine ( O l l ) , thus suggesting that TBPA does bind some thyroxine in vivo. Some normal values for thyroxinebinding proteins are shown in Table 2. Woeber et at. (W3) have recently questioned the values listed in Table 2 for TBPA thyroxine binding. They used specific antibody for TBPA and reported that approximately 15% of labeled thyroxine was precipitable by antibody. This value is about one-half the previously reported values of approximately 30%. An indirect assessment of the number of free binding sites can be made by adding a competing adsorbing substance to the serum in question. The distribution of tracer amounts of I3lI-labeled thyroxine or triiodothyronine between serum TBG and the adsorbant can then be measured. This is the basis of the so-called T, resin uptake or red cell uptake tests. The most widely used methods employ triiodothyronine, since it is more weakly bound to TBG and is not bound a t all to TBPA. A resin is used (M17, S20) as the competing binding agent in many tests. Historically, red cells (H4) were first used as the competing binding agent. Both of these tests are dependent on p H and temperature, both in the test system and in the individual from whom the serum was taken Resin tests are probably to be preferred because red cell ability to adsorb Tsis dependent upon COz concentration (H5) , the presence of immature

BIOCHEMISTRY O F THYROID REGULATION

407

erythrocytes (SS), and other variables. The number of free binding sites is inversely proportional to the amount of T, bound to the adsorbant.

4.4. FREETHYROXINE Until recently it was not always possible to measure unbound thyroxine in serum directly because of the minute quantities involved. Therefore a free thyroxine index calculation was often made, since this value is proportional to the free thyroxine concentration. This index can be arrived a t by multiplying a measure of thyroxine in the blood (usually the PBI) by a value derived from the measurement of thyroxine-unoccupied binding sites of thyroxine-binding proteins. This was possible since the uptake of thyroxine or triiodothyronine by red cells or resin is inversely proportional to the number of unoccupied binding sites. As thyroxine-binding sites become increasingly saturated, the free thyroxine levels increase. An example of this method of arriving a t a free thyroxine index is the work of Clark and Horn ((3). Calculation of the free thyroxine index is valuable since the PBI determination and the resin uptake of T, are determinations that are routinely made in many clinical laboratories. The direct measurement of free thyroxine was first reported by Sterling and Hegedus in 1962 (S18). Several methods are now available. The general method involves the addition of tracer amounts of lSII-labeled thyroxine to serum and then the use of equilibrium dialysis or ultrafiltration to separate unbound thyroxine. The methods vary primarily in their approach to the problem of separating 1311-labeled contaminants from labeled thyroxine. Magnesium precipitation of the dialyzed thyroxine fraction has been one approach (517). Other approaches include the use of Sephadex as the precipitating agent, or the use of indifferent serum in the dialysis fraction as a binding agent with subsequent dialysis against Amberlite IRA-400 to remove the contaminating iodine (L5). I n a fourth method, serum is diluted, thus increasing the dialyzable fraction of thyroxine which can be precipitated with TCA (06). All these methods appear to be internally consistent although they may not be comparable to each other. The results depend on temperature, and the temperature of the reaction may not be comparable to the in vivo situation. Free thyroxine concentration is obtained by multiplication of some measure of serum thyroxine (usually PBI) by the dialyzable fraction.

PROTEINS IN 4.5. THEROLEOF THYROXINE-BINDING

THYROXINE TRANSPORT AND TURNOVER TBG and TBPA can be considered as a buffer system which maintains a steady-state equilibrium in the concentration of free thyroxine.

408

ROBERT D. LEEPER

In the absence of all binding proteins, large variations might be expected in the concentration of thyroid hormone available to tissue when changes in thyroid hormone output occurred. Robbins and Rall (R9)postulated that the absolute turnover of thyroxine would be proportional to the free thyroxine concentration. According to this hypothesis, a decrease in thyroxine-binding proteins would cause a decrease in the total extrathyroidal pool of iodine, an increased free thyroxine fraction, and an increase in fractional turnover rate of thyroxine a5 related to total body pool. Absolute thyroxine turnover, and absolute free thyroxine concentration would be normal. The converse would be true if thyroxine-binding capacity were increased. The total extrathyroidal pool of iodine would be increased, the dialyzable fraction of thyroxine would be decreased, and the fractional thyroxine turnover rate would decrease. The absolute thyroxine turnover rate and free thyroxine would remain constant. Adjustment of TSH secretion would not be necessary to maintain free thyroxine concentration since changes in binding capacity would also regulate free thyroxine levels. The delivery of thyroxine to the tissues and its metabolism and effects would be more directly reIated to the free thyroxine concentration than it would be to total serum thyroxine content. The PBI (or total serum thyroxine if measured directly) would not necessarily reflect the metabolic state of the patients in the presence of increased or decreased amounts of binding proteins. Since binding proteins are normally unsaturated, a modest increase in thyroxine would not alter free thyroxine concentrations significantly. TBG is the main binding protein in that two-thirds or more of blood thyroxine is bound to this protein because of its higher binding affinity. The amount of TBG is limited, however (1.5 mg/100 ml) ( 0 5 ) , and an additional fraction of thyroxine is bound to TBPA. TBPA binds thyroxine less firmly than does TBG, and, in addition, TBPA has a short halflife in the serum (012, 513). Rapid alterations can occur in the binding capacity of TBPA, thus causing relatively rapid changes in free thyroxine Concentration. The rapid fluctuations might be considered as a fine-tuning mechanism which can alter the levels of free thyroxine temporarily. As the free thyroxine concentration increases, thyroxine turnover increases and a new steady state ensues with normal free thyroxine levels. I n general, the above hypothesis is supported by experimental studies, but recent studies indicate that tissue-binding factors also modify thyroxine turnover and free thyroxine levels. Therefore the regulatory function theory of serum thyroxine binding has to be broadened to include tissue binding or degradation factors. The liver in particular concentrates thyroxine (C3, 08),and as milch

BIOCHEMISTRY OF THYROID REGULATION

409

as 30% of total body thyroxine may be concentrated in that organ. Changes in the ability of liver to bind thyroxine may result in an altered free thyroxine level in the absence of TBG binding capacity changes. The other natural hormone, T,, may also alter the rate of thyroxine turnover. T,, which is metabolically 4-5 times as potent as T,, is bound poorly, if a t all, to TBG in Wivo ( Z l ) . T,, however, has been reported to increase the turnover rate of thyroxine in man and the ratio of turnover to free thyroxine increased, indicating that T, stimulated the binding of thyroxine by extravascular sites (57). Phenobarbital increased thyroxine levels in the liver and increased thyroxine turnover and thyroidal function in the rat in the absence of binding changes (09). Such reports point out the role of tissue binding in regulating thyroid hormone distribution. 4.6. CAUSES OF ABNORMALITIE~ IN THYROXINE-BINDING CAPACITY 4.6.1. Genetic

Inherited decreases in TBG have been reported (R5, Tl). The serum of these patients was characterized by a low protein-bound iodine in the absence of demonstrable hypothyroidism. Tracer studies revealed decreased extrathyroidal thyroxine pools and normal absolute thyroxine turnover (C3, T2). The condition is probably transmitted as a dominant X chromosome-linked trait (N3, R5). No abnormalities in TBPA have been reported in these conditions. Several subjects have also been studied whose serum had an elevated thyroxine-binding capacity (B5,F3, 15, 52). Although high P B I levels were present, patients were euthyroid. Free thyroxine concentration was normal in these patients (I5), pool sizes were increased, and absolute thyroxine turnover was normal. 4.6.2. Estrogens

Estrogens increase the binding capacity of TBG (D6, E4),and this is the reason for the increased PBI’s found during the administration of these hormones. Pregnancy is also associated with increased PBI’s. Absolute thyroxine turnover is normal in these circumstances (D7, S17), and free thyroxine is normal or slightly decreased. Testosterone has, opposite effects (E4). 4.6.3. Drugs

Diphenylhydantoin, salicylates (07, W4) , among other drugs, displace thyroxine from binding proteins in vitro and lower the PBI in vivo. This

410

ROBERT D. LEEPER

mechanism has been invoked to explain the low PBI found in patients receiving this drug. However, effects on extravascular binding of thyroxine or its degradation have not been excluded.

Thyroid Disease In hyperthyroidism with the attendant large increase in thyroxine output, the PBI increases. There is an associated increase in free thyroxine concentration (16, 011, 517) and an increase in the thyroxine turnover rate. Total thyroxine binding capacity is reduced as a result of a combination of factors (I1,16, 011, R9).The number of available binding sites is decreased by additional thyroxine, TBPA is decreased, and in some patients the binding capacity of TBG is decreased. The converse is true in hypothyroidism (16, 011, R9). Persistent decreases in the binding of thyroxine to TBPA (B17) in patients with treated hyperthyroidism suggest extravascular factors in regulating net binding of thyroxine in blood.

4.6.4.

Nonthyroidal Disease A number of nonthyroidal illnesses may be associated with diminished protein binding (16, 011, SlS). An increased free thyroxine fraction is often seen, and the absolute concentration of TBPA may be decreased. TBPA-binding capacity is decreased. Weight loss, fever, and general debilitation are characteristic of these diseases, and decreases in TBPA (011) have been attributed to decreased synthesis and the unusually rapid (2-3 days) turnover of this protein (012,513). I n chronic diseases, thyroxine turnover is generally increased, thus maintaining a normal free thyroxine level. In some patients, however, the free thyroxine concentration as well as the dialyzable fraction remain elevated. Elevated free thyroxine levels are found most often after surgical stress (S24), pyrogen administration (012)) and the like. Increased free thyroxine levels have also been found a t the onset of acute cardiac arrhythmias (S4). Fever itself will elevate free thyroxine values (B7). The persistent free thyroxine values implicate extravascular factors in the regulation of free thyroid hormone levels. Hollander et al. (H9) have recently reported that free fatty acids increase free thyroxine levels both in vitro and after infusion of fatty acids in man. Whether or not the increased free thyroxine levels caused by fatty acids have physiological importance is not clear a t this time. The study does show that thyroxine binding may be regulated by other substances in the blood compartment. Although it seems clear that free thyroxine concentration does vary with metaboh changes, it is by no means certain that altered free 4.6.5.

BIOCHEMISTRY OF THYROID REGULATION

41 1

thyroxine levels are a necessary homeostatic adjustment to physiological and biochemical changes. It is certain that the gross basal metabolic rate measurement must be replaced by a more precise and sensitive method of measuring peripheral thyroxine effects. Neither is it clear just what role triiodothyronine (TJ plays in the human metabolic economy. Newly described methods for measuring T3( N l , 521) in blood will aid in elucidating that role. 5.

Calcitonin

5.1. INTRODUCTION Prior to 1962 it was generally accepted that the parathyroid gland exerted the only specific hormonal control over calcium hemostasis. It was thought that high blood calcium levels acted to inhibit the secretion of parathormone and that low blood calcium levels acted to increase hormone secretion. Copp ((310) first showed that another substance was responsible for lowering serum calcium in the rat, and named it calcitonin. Copp ((210) initially thought that this hormone came from the parathyroid after a series of experiments in which he perfused isolated rat thyroid and parathyroid glands with hypercalcemic blood and noted a fall in serum calcium. These findings were confirmed in 1963 by two other groups (H7, K8). Hirsch et al. (H7) suggested that the source of the calcium-lowering principle might be the thyroid, based on the observation that hot wire cautery caused a more rapid loss of calcium than did parathyroidectomy by surgery alone. The cautery, of course, also destroyed thyroid tissue. Foster (F6) confirmed the presence of calcitonin in the thyroid by perfusion experiments in the goat. Perfusion of the parathyroid alone had no hypocalcemic effect, whereas perfusion of the thyroid and parathyroid glands together with hypercalcemic blood did cause a fall in serum calcium. Autogenous extracts of the thyroid had calcium lowering properties and hypercalcemia caused increased secretion of calcitonin from the thyroid. Progress in the field has been rapid. Calcitonin has been isolated from the thyroid or the ultimobranchial body in more than twenty species (F5) including man. Calcitonin was purified, and the amino acid sequence was determined (N2, P7). The molecular weight is approximately 3600. The synthesis of calcitonin was reported within the past year (B6, R7). Recent reviews on calcitonin are available (F5, M2). 5.2. ASSAY The current bioassay of calcitonin depends on the ability of the material in question to lower the blood calcium in a fasted rat or in rats on

412

ROBERT D. LEEPER

a low calcium intake. A typical assay as modified from the original was described by Cooper et al. (C8). Five-week Holteman rats are placed on a special low-calcium, normal-phosphorus diet for 1 day. The test substance is injected subcutaneously in 3 concentrations, 1- to 3-f0ld, and blood is taken by cardiac puncture a t 70 minutes. Calcium levels in the blood are measured by the EDTA method of Copp (C9) or a semiautomated autoanalyzer method ((38). Activity is compared to a British Medical Research Council (MRC) standard. The data are treated by analysis of variance, the slope of the curve being a straight-line function of the log dose within limits. Response differences were noted among various strains of rat (C8). Sensitivity variations were seen and depended on the age of the rat and the method of administration of the test substance. Continuous infusion or intravenous infusion gave more sensitivity at low dose ranges (C8). A radioimmune assay for calcitonin has recently been developed (Dl), and this should give more precision in measuring calcitonin in blood. The radioimmunoassay was used in preliminary studies (Dl) of normal human serum; no activity was found, although the assay is mole. sensitive t o 1 x Hypocalcemic activity was found in human blood after acetone ether precipitation of protein, extraction of the precipitate with an acetic acid-ethanol-water mixture, and lyophilieation (S22). Assay in fasted rats gave levels of hypocalcemic activity of 100 microunits (MRC) of calcitonin per liter. Gel filtration (G6) was also used to concentrate human plasma hypocalcemic factor. 5.3. PHYSIOLOGY OF CALCITONIN At least two types of cells are present in the mammalian thyroid-the follicular cell, and the parafollicular cell. The parafollicular cell has been referred t o as the “C”cell by Pearse (P5) and others, because this cell is thought to secrete calcitonin. Evidence that these cells contained calcitonin was provided by immunofluorescent studies (B20). The thyroid of the bat, a hibernating species, contains large numbers of the granular parafollicular cells during activity (N4). These cells become totally degranulated during hibernation a t a time when the bat is hypocalcemic. The bat would seem to represent a good experimental animal in the study of calcitonin physiology. Regulation of calcitonin secretion is controlled by calcium levels in the blood. The early studies are in accord with this concept. Care ( C l ) showed by direct assay of venous effluent from perfused pig thyroid that calcitonin release is a function of the calcium level in the perfusate.

BIOCHEMISTRY O F THYROID REGULATION

413

Hypophysectomy (M15) did not alter extractable levels of thyrocalcitonin levels in the thyroid of treated rats as compared to controls. The known effects of thyrocalcitonin are primarily in bone. Removal of the gut had no effect on the hypocalcemic effect of calcitonin (A6), and neither did nephrectomy (HS). No change in soft tissue calcium content was seen in soft tissues to explain the hypocalcemia ( K l ) . Calcitonin apparently inhibits bone resorption and thereby decreases calcium entry into the blood. Calcitonin prevents the release of calcium from cultured bone (A5, F10). In vivo, the release of 45Ca from prelabeled bone is decreased by calcitonin ( J l ) . The bone arteriovenous difference in calcium levels is increased by calcitonin (M3). The mode of action of calcitonin is unknown. Calcitonin does not inhibit parathormone (A6, H7, T3),nor is its effect apparently mediated through RNA synthesis (T3). 5.4. CALCITONIN IN MAN Calcitonin can be isolated from the thyroid of man (A7). The concentration of calcitonin in human thyroids is normally lower than found in the rat. Medullary carcinoma of the thyroid is thought to arise from the parafollicular cell in man. Increased calcium-lowering activity was reported in the tumor tissue or blood in several patients with this cancer (C11, M14, MlS). Increased amounts of calcitonin were found in the thyroids of two patients with pseudo-hypoparathyroidism (A7, T5), and low levels as compared to normal were reported (T4) in patients with parathyroid adenomas. These latter studies are difficult to evaluate, since one does not know whether the changed calcitonin levels are a cause or an effect of the serum calcium abnormalities. Thyroidectomized patients have normal calcium levels as a rule, and this fact is an argument against a major homeostatic role for thyrocalcitonin in man. Other tissues may secrete calcitonin, however, and the thyroid gland may not be the only organ which can secrete hypocalcemic principle. MacIntyre (M2) recently reported the finding of calcitonin in thymus and parathyroid glands. Thus we have come full circle in six years, and Copp’s original theory (C10) that parathyroid glands contained calcitonin may yet be proved correct. Glucagon also has hypocalcemic effects (M18). When thyrocalcitonin was given to patients with hypercalcemia due to a variety of causes, serum calcium levels fell (B9, F7, HI). The magnitude of the effect was related to the degree of calcium elevation ( H l ) . It would appear that the use of calcitonin for treatment of acute hypercalcemia is not warranted a t this time. Studies of the long-term use of thyrocalcitonin in diseases such as osteoporosis have not yet been reported.

414

ROBERT D. LEEPER

REFERENCES Al. Adams, D. D., The presence of an abnormal thyroid stimulating hormone in the serum of some thyrotoxic patients. J . Clin. Endohnol. Metab. 18,699-712 (1958). A2. Adams, D. D., and Purves, H. D., Abnormal response in the assay of thyrotropin. PTOC.Univ. Otago Med. School 34, 11 (1956). A3. Adams, D. D., and Purves, H. D., The role of thyrotropin in hyperthyroidism and exophthalmos. Metab., Clin. Exptl. 6, 26-35 (1957). A4. Adams, D. D., Kennedy, T. H., and Purves, H. D., Non-specific responses in the assay of thyrotropin and the long-acting thyroid stimulator. Australian J . Exptl. Biol. Med. Sci. 44, 355-366 (1966). A5. Aliapoulios, M. A., Goldhaber, P., and Munson, P. L., Thyrocalcitonin inhibition of bone resorption induced by parathyroid hormone in tissue culture. Science 161, 330 (1966). A6. Aliapoulios, M. A., Savery, A., and Munson, P. L., New experiments with thyrocalcitonin. Federation PTOC.24, 322 (1965). A7. Aliapoulios, M. A., Voelkel, E. F., and Munson, P. F., Assay of human thyroid glands for thyrocalcitonin activity. J . Clzn. Endocrinol. Metab. 26, 897-901 (1966). A8. Altman, M., Oka, H., and Field, J. B., Effects of TSH, acetylcholine, epinephrine, serotonin and Synkavite on 8 2 P incorporation into phospholipids in dog thyroid slices. Biochim. Biophys. Acta 116, 586-588 (1966). A9. Alvsaker, J. O., Haugli, F. B., and Laland, S. G., Presence of Vitamin A in human tryptophan-rich prealbumin. Biochem. J. 102, 362-366 (1967). 810. Arnaud, C. D., Kneubuhler, H. A., Seiling, V. L., Wightman, B. K., and Engbring, N. H., Response of the normal human to infusions of plasma from patients with Graves’ disease. J . Clin. Invest. 44, 1287-1294 (1968). A l l . Averill, R. L., Salaman, D. F., and Worthington, W. C., Jr., Thyrotropin releasing factor in hypophyseal portal blood. Nature 211, 144-145 (1966). Bl. Bakke, J. L., Assay of human thyroid-stimulating hormone by 18 different assay laboratories using 12 different methods. J . Clin. Endocrinol. Metab. 26, 545-550 ( 1965). B2. Bakke, J. L., Heideman, M. L., Jr., Lawrence, N. L., and Wiberg, C., Bio-assay of thyrotropic hormone by weight response of bovine thyroid slices. Endocrinology 61, 352-367 (1957). B3. Bates, R. W., and Cornfield, J., An improved assay method for thyrotropin using depletion of I131 from the thyroid of day-old chicks. Endocrinology 60, 225-238 (1957). B4. Becker, D. V., and Furth, E. D., I n “Current Topics in Thyroid Research” (C. Cassano and M. Andreoli, eds.), pp. 596-602. Academic Press, New York, 1965. B5. Beierwaltes, W. H., and Robbins, J., Familial increase in thyroxine-binding sites in serum alpha globulin. J . Clin. Invest. 38, 1683-1688 (1959). B6. Bell, P. H. et al., Purification and structure of porcine calcitonin-1. J . Am. Chem. SOC.90, 2704-2706 (1968). B7. Bernstein, G., and Oppenheimer, J. H., Factors influencing concentration of free and total thyroxine in patients with non-thyroidal disease. J . Clin. Endocrinol. Metab. 26, 195-201 (1966). B8. Berson, S. A., and Yalow, R. S., Isotopic t,racers in the study of diabetes. Advun. Biol. Med. Phys. 6, 349-430 (1958). B9. Bijvoet, 0. L. M., Van der Sluys Veer, J., and Jansen, A. P., Effects of thyro-

BIOCHEMISTRY O F THYROID REGULATION

B10. B11. B12. B13. B14.

B15.

B16. B17. B18. B19. B20. C1. C2. C3.

C4. C5. C6. C7.

415

calcitonin on patients with Paget’s disease, thyrotoxicosis, or hypercalcemia. Lancet I, 876-881 (1968) Borell, U., On the transport route of the thyrotropic hormone, the occurrence of the latter in different parts on the brain and its effect on the thyrodea. Acta Med. Scand. Suppl. 161, 1-227 (1945). Bottari, P. M., and Donavan, B. T., A sensitive in vitro technique for the assay of thyrotropic hormone. J. Physiol. (Lon&) 140, 36P-37P (1958). Bowers, C. Y., Lee, K. L., and Schally, A. V., Effect of Actinomycin D on hormones that control the release of thyrotropin from the anterior pituitary glands of mice. Endocrinology 82, 303-310 (1968). Bowers, C. Y., Lee, K. L., and Schally, A. V., A study of the interaction of the thyrotropin-releasing factor and L-triiodothyronine. Effects of puromycin and cycloheximide. Endocrinology 82, 75-82 (1968). Bowers, C. Y., Redding, T. W., and Schally, A. V., Effect of thyrotropin-releasing factor (TRF) of ovine, bovine, porcine and human origin on thyrotropin release in vitro and in vivo. Endorrinology 77, 609-616 (1965). Bowers, C. Y., Schally, A. V., Reynolds, G. A., and Hawley, W. D., Interactions of L-thyroxine or btriiodothyronine and thyrotropin-releasing factor on the release and synthesis of thyrotropin from anterior pituitary gland of mice. Endocrinology 81, 741-747 (1967). Bowers, C. Y., Schally, A. V., Hawley, W. D., Gual, C., and Parlow, A., Effect of thyrotropin-releasing factor in man. J. Clin. Endocrinol. Metab. 28, 978-982 (1968). Braverman, L. E., Foster, A. E., and Ingbar, S. H., Thyroid hormone transport in the serum of patients with thyrotoxic Graves’ disease before and after treatment. J . Clin. Invest. 47, 1349-1357 (1968). Burke, G., Effects of iodotyrosines on basal and stimulated glucose oxidation and phospholipogenesis. Endocrinology 83, 495-500 (1968). Burke, G., Effects of cyclic 3’,5’-adenosine monophosphate and dibutyryl cyclic 3’,5’-adenosine monophosphate on basal and stimulated thyroid function. J. Clin. Endocrinol. Metab. 28, 1816-1823 (1968). Bussolati, G., and Pearse, A. G. E., Immunofluorescentlocalization of calcitonin “C” cells of pig and dog thyroid. J. Endocrinol. 37, 205-209 (1967). Care, A. D., Cooper, C. W., Duncan, T., and Orimo, H., A study of thyrocalcitonin secretion by direct measurement of in vivo secrction rate in pigs. Endocrinology 83, 161-169 (1968). Cassano, C., and Andreoli, M., eds., “Current Topics in Thyroid Research.” Academic Press, New York, 1965. Cavalieri, R. R., and Searle, G. L., Kinetics of distribution between plasma and liver of lalI-labeled Irthyroxine in man: Observations of subjects with normal and decreased serum thyroxine-binding globulin. J. Clin. Invest. 46, 939-949 (1966). Christensen, L. K., and Litonjua, A. D., Is thyroxin-binding by prealbumin of physiologic importance? J . Clin. Endocrinol. Metab. 21, 104-106 (1961). Clark, F., and Horn, D. B., Assessment of thyroid function by the combined use of the serum protein-bound iodine and resin uptake of 1811-triiodothyronine. J . Clin. Endomlzol. Metab. 26, 39-45 (1965). Condliffe, P. G., Purification of human thyrotropin. Endocrinology 72, 893-896 (1963). Condliffe,P. G., and Robbins, J., Pituitary thyroid-stimulating hormone and other

416

ROBERT D. LEEPER

thyroid-stimulating substances. I n “Hormones in Blood” (C. H. Gray and L. A. Bacharach, eds.), 2nd rev. ed., Vol. 1, pp. 333-381. Academic Press, New York, 1967. C8. Cooper, C. W., Hirsch, P. F., Toverud, S. U., and Munson, P. L., An improved method for the biological assay of thyrocalcitonin. Endocrinology 81, 610-616 (1967). C9. Copp, D. H., Simple and precise micromethod for EDTA titration of calcium. J . Lab. Clin. Med. 61, 1029-1037 (1963). C10. Copp, D. H., Cameron, E. C., Vheney, B. A., Davidson, A. G. F., and Henze, K. G., Evidence for calcitonin-a new hormone from the parathyroid that lowers blood calcium. Endocrinology 70, 638-649 (1962). C l l . Cunliffe, W. J. et al., Calcitonin-secreting thyroid carcinoma. Lancet 11, 63-66 (1968). D1. Deftos, L. J., Lee, M. R., and Potts, J. T., Jr., A radioimmune assay for thyrocalcitonin. Proc. Natl. Acad. Sci. U.S. 60,293-299 (1968). D2. De Robertis, E., Assay of thyrotropic hormone in human blood. J. Clin. Endocrinol. Metab. 8, 956-966 (1948). D2a. Dobyns, B. M., Rudd, A., and Liebe, D., The assay of the exopthalmus-producing substance (EPS) and the long-acting thyroid stimulator (LATS) in whole and fractionated serum of patients with progressive exopthalmos. I n “Current Topics in Thyroid Research’’ (C. Cassano and M. Andreoli, eds.), pp. 484499. Academic Press, New York, 1965. D3. Dorrington, K. J., and Munro, D. S., The long-acting thyroid stimulator. Clin. Pharmawl. Therap. 7 , 788-806 (1966). D4. Dorrington, K. J., Carniero, L., and Munro, D. S., The proteolysis of immunoglobulin-G with long-acting thyroid stimulating activity. Biochem. J . 98, 858-861 (1966). D5. Dorrington, K. J., Carniero, L., and Munro, D. S., Absorption of the long acting thyroid stimulator by human thyroid microsomes. J . Endocrinol. 34, 133-136 (1966). D6. Dowling, J. T., Freinkel, N., and Ingbar, S. H., Effect of diethyl stilbesterol on binding of thyroxine in serum. J. Clin. Endocrinol. Metab. 16, 1491-1506 (1956). D7. Dowling, J. T., Freinkel, N., and Ingbar, S. H., Effect of estrogens upon peripheral metabolism of thyroxine. J . Clin. Invest. 39, 1119-1130 (1960). D8. Dumont, J. E., Stimulation in viQo of hexose monophosphate pathway in thyroid by iodothyrosines and Synkavit. Biochim. Bioph3s. A cta 60, 506-512 (1961). El. El Kabir, D. J., Assay of thyrotropic hormone in blood. Nature 194, 688-689 (1962). E2. El Kabir, D. J., Benhamou, G., Doniach, D., and Roitt, I. M., Absorption of thyroid-stimulating globulin from thyrotoxic sera by organ homogenates. Nature 210, 319-321 (1966). E3. Elzinga, K. E., Carr, E. A., Jr., and Beierwaltes, W. H., Adaptation of standard Durrun-type cell for reverse flow electrophoresis. Am. J . Clin.Pathol. 36, 125-131 (1961). E4. Engbring, N. R., and Engstrom, W. W., Effects of estrogen and testosterone on circulating thyroid hormone. J . Clin. Endocrinol. Metab. 16, 783-796 (1959). E5. Ensor, J. M., and Munro, D. S., A comparison of the influence of thyroid-stimulating hormone and cyclic 3’,5’-adenosine monophosphate on the in uitro release of radio-iodine from the thyroid glands of mice. J . Endorrinol. 38, XXViii (1967). E6. Ensor, J. M., and Munro, D. S., A comparison of the actions of thyroid stimulating

BIOCHEMISTRY OF THYROID REGULATION

E7.

F1. F2. F3. F4.

F5. F6. F7.

F8. F9.

F10. G1. G2. G3. G4.

G5. G6.

G7. G8.

417

hormone and the long-acting thyroid stimulator in an in vitro assay. Proc. Roy. SOC.Med. 61, 652-654 (1968). Etkin, W., and Gona, A. G., Failure of mammalian thyrotropin-releasing factor preparation to elicit metamorphic responses in tadpoles. Endocrinology 82, 10671068 (1968). Field, J. B., Pastan, I., Johnson, P., and Herring, B., Stimulation in vitro of pathways of glucose oxidation in thyroid by thyroid-stimulating hormone. J. Biol. C h m . 236, 1863-1866 (1960). Field, J. B., Remer, A., Bloom, G., and Kriss, J. P., In vitro stimulation of longacting thyroid stimulator of thyroid glucose oxidation and **Pincorporation into phospholipid. J. Clin. Invest. 47, 1553-1560 (1968). Florsheim, W. H., Dowling, J. T., Meister, L., and Bodfish, R. E., Familial elevation of serum thyroxine-binding capacity. J . Clin. Endocrinol. Metab. 22, 735-740 (1962). Fontaine, Y. A., and Condliffe, P. G., Density gradient centrifugation of bovine thyroid-stimulating hormone. Biochemistry 2, 290-293 (1963). Foster, G. V., Calcitonin (Thyrocalcitonin). New Engl. J . Med. 279, 349-360 (1968). Foster, G. V., Baghdiantz, A., Kumar, M. A., Slack, E. J., Soliman, N. A., and MacIntyre, I., Thyroid origin of calcitonin. Nature 202, 1303-1305 (1964). Foster, G. V., Joplin, G. F., MacIntyre, I., Melvin, K. E. W., and Slack, E., Effect of thyrocalcitonin in man. Lancet I, 107-109 (1966). Freinkel, N., Pathways of thyroidal phosphorus metabolism: The effect of pituitary thyrotropin upon phospholipids of the sheep thyroid gland. Endocrinology 61, 448-460 (1957). Freychet, P., Study of immunological cross-reactions and specificity of human thy rotropin (HTSH) by means of radioimmuno-assay. Abstracts of brief communications. Proc. 3rd Intern. Congr. Endocrinol., Mexico City, 1968 Intern. Congr. Ser. No. 157, p. 13 Excerpta Med. Found., Amsterdam, 1968. Friedman, J., and Raisz, L. G., Thyrocalcitonin: Inhibitor of bone resorption in tissue culture. Science 160, 1465-1467 (1965). Gilman, G. A., and Rall, T. W., Studies on the relation of cyclic 3’,5’-AMP (CA) to TSH action in beef thyroid slices. Federation Proc. 26, 617 (1966). Gordon, A. H., Gross, J., O’Connor, D., and Pitt-Rivers, R., Nature of circulating thyroid-hormone plasma protein complex. Nature 169, 19-20 (1952). Greer, M. A., Studies of the influence of the central nervous system on anterior pituitary function. Recent Progr. Hormone Res. 8, 67-104 (1957). Gual, C., ed., Proc. 3rd Intern. Congr. Endocrinol., Mexico City, 1968. Excerpta Med. Found., Amsterdam, 1969. Gual, C., ed., Abstracts of brief communications. Proc. 3rd Intern. Congr. Endocrinol., Mexico City, 1968. Intern. Congr. Ser. No. 157. Excerpta Med. Found., Amsterdam, 1968. Gudmundsson, T . V., Akrawi, L., Galante, L., Kenny, A. D., Matthews, E. W., Tse, A., and Woodhouse, N. J., Calcitonin-like activity in human and rat blood. Abstracts of brief communications. Intern. Congr. Ser. No. 157, p. 8. Excerpta Med. Found., Amsterdam, 1968. Guillemin, R., The adenohypophysis and its hypothalamic control. Ann. Rev. Physiol. 20, 313-345 (1967). Guillemin, R., Sakis, E., and Ward, D. N., Further purification of TSH-releasing factor (TRF) from sheep hypothalamic tissues, with observations on the amino acid composition. Proc. SOC.Exptl. Biol. Med. 118, 1132-1 137 (1965).

418

ROBERT D. LEEPER

G9. Guillemin, R., Yamazaki, E., Jutisz, M., and Sakiz, E., Presence dans un extrait de tissus hypothalamiques d’une substance stimulant la s6cr6tion de l’hormone hypophysaire thyreotrope (TSH). Premibre purification par filtration sur gel Sephadex. Compt. Rend. 266, 1018-1020 (1962). G10. Guillemin, R., Yamazaki, E., Gard, D. A., Jutisz, M., and Sakis, E., I n vitro secretion of thyrotropin (TSH) : Stimulation by a hypothalamic peptide (TRF). Endocrinology 73, 564-572 (1963). H1. Haas, H. G., and Dumbacher, M., Calcitonin in hypercalcaemia. Lancet 11, 217-218 (1968). H2. Hall, R., and Tubman, J., Further studies on effects of thyroid stimulating hormone on thyroid nucleotide synthesis. J. Biol. Chem. 240, 3132-3135 (1965). H3. Hall, R., and Tubman, J., Effect of thyroid-stimulating hormone on ribonucleic acid synthesis in the calf thyroid in vitro in the presence of puromycin. J . Biol. Chem. 243, 1598-1602 (1968). H4. Hamolsky, M. W., Stein, M., and Freedberg, A. S., Thyroid hormone-plasma protein complex in man. 11. New in vitro method for study of uptake of labelled hormonal components by human erythrocytes. J . Clin. Endocrinol. Metab. 17, 33-43 (1957). H5. Hamolsky, M. W., Stein, M., Fischer, D. B., and Freedberg, A. S., Further studies of factors affecting the plasma protein-thyroid hormone complex. Endocrinology 68, 662-670 (1961). H6. Harris, G. W., “Neural Control of the Pituitary Gland.” Williams & Wilkins, Baltimore, Maryland, 1955. H7. Hirsch, P. F., Gauthier, G. F., and Munson, P. L., Thyroid hypocalcemic principle and recurrent laryngeal nerve injury as factors affecting response to parathyroidectomy in rats. Endocrinology 73, 244-252 (1963). H8. Hirsch, P. F., Voekel, E. F., and Munson, P. L., Thyrocalcitonin: Hypocalcemic hypophosphatemic principle of thyroid gland. Science 146, 412 (1964). H9. Hollander, C. S., Scott, R. L., Burgess, J. A.!Rabinowitz, D., Merimec, T. J., and Oppenheimer, J. H., Free fatty acids: A possible regulator of free thyroid hormone levels in man. J . Clin. Endocrinol. Metab. 27, 1221-1222 (1968). 11. Inada, M., and Sterling, K., Thyroxine transport in thyrotoxicosis and hypothyroidism. J . Clin. Invest. 46, 1442-1450 (1967). 12. Ingbar, S. H., Pre-albumin: Thyroxine binding protein of human plasma. Endocrinology 63, 256-259 (1958). 13. Ingbar, S. H., Observations concerning binding of thyroid hormones by human serum prealbumin. J . Clin. Invest. 42, 143-160 (1963). 14. Ingbar, S. H., Proc. 3rd Intern. Congr. Endomkol., Mexico City, 1368 (C. Gual, ed.). Excerpta Med. Found., Amsterdam, 1969. 15. Ingbar, S. H., Waterhouse, C., and Cushman, P., Observations on nature of underlying disorder and occurrence of associated plasma transport abnormalities in patient with idiopathic increase in plasma thyroxine-binding globulin. J . Clin. Invest. 48, 226G2271 (1964). 16. Ingbar, 8. H., Braverman, L. E., Dawber, N. A., and Lee, G. Y., New method for measuring free thyroid hormone in serum and analysis of factors that influence its concentration. J . CZin. Invest. 44, 1679-1689 (1965). J1. Johnston, C . C., Jr., and Deiss, W. P., Jr., Inhibitory effect of thyrocalcitonin on calcium release in vivo and on bone metabolism in vitro. Endocrinology 78, 11391143 (1966). 52. Jones, J. E., and Seal, U. S., X-chromosome linked inheritance of elevated thyroxine-binding globulin. J . Clin. Endocrinol. Metab. 27, 1526-1528 (1967).

BIOCHEMISTRY OF THYROID REGULATION

419

K1. Kenny, A. D., and Heiskell, C. A., Effect of crude thyrocalcitonin on calcium and phosphorus metabolism in rats. Proc. SOC.Exptl. Biol. Med. 120, 269-271 (1965). K2. Kirkham, K. E., A new bioassay technique for the measurement in uitro of thyrotropic hormone in serum and pituitary extracts. J. Endocrinol. 26, 259-269 (1962). K3. Kirkham, K. E., The detection and measurement of thyrotropic substances in health and disease. Vitamins Hormones 24, 173-266 (1966). K4. Klainer, L. M., Chi, Y. M., Friedberg, S. L., Rall, T. W., and Sutherland, E. W., Adenyl cyclase. IV. The effects of neurohormones on the formation of adenosine 3',5'-phosphate by preparations from brain and other tissues. J . Biol. Chem. 237, 1239-1243 (1962). K5. Krass, M. E., La Bella, F. S., and Vivian, S. R., Thyrotropin release in vitro: The role of metabolism in the secretory response to vasopressin, oxytocin and epinephrine. Endocrinology 82, 1183-1189 (1968). K6. Kriss, J. P., Pleshakov, V., and Chien, J. R., Isolation and identification of the long-acting thyroid stimulator and its relation to hyperthyroidism and ckcumscribed pre-tibia1 myxedema. J. Clin. Endocrinol. Metab. 24, 1005-1028 (1964). K7. Kumahara, Y., Iwatsubo, H., Miyai, K., Masui, H., Fukuchi, M., and Abe, H., Abnormal thyrotropic substance in the pituitaries of patients with Graves' disease. J . Clin.Endocrinol. Metab. 27, 333-340 (1967). K8. Kumar, M. A., Foster, G . V., and MacIntyre, I., Further evidence for calcitonin: Rapid acting hormone which lowers plasma calcium. Lancet 11, 480-482 (1963). L1. Larson, F., Deiss, W. P., and Albright, E. C., Localization of protein-bound radioactive iodine by filter paper electrophoresis. Science 116, 626 (1952). L2. Lemarchand-Beraud, T., and Vanotti, A., A radioimmunoassay for the determination of thyroid stimulating hormone. Experientia 21, 353-356 (1965). L3. Lepp, A., and Oliner, L., Failure of long-acting thyroid stimulator globulin (LATS) and serum to stimulate thyroid function in the chick. Endocrinology 80, 369-374 (1967). L4. Levy, R. P., McQuire, W. L., and Heideman, M. L., Jr., Quantitative studies of the reaction of bovine thyrotropin preparations. Proc. Soc. Exptl. Biol. Med. 110, 598-600 (1962). L5. Liewendahl, K., and Lamberg, B.-A., Free thyroxine in serum determined by dialysis and Sephadex G-25 filtration. J . Clin. Endocrinol. Metab. 26, 991-993 (1965). L6. Lindsay, R. H., Cash, A. G., and Hill, J. B., Thyrotropin stimulation of pyrimidine nucleotide synthesis in the bovine thyroid. Biochem. Biophyls. Res. Commun. 29, 850-855 (1967). L7. Lindsay, R. H., Cash, A. G., and Hill, J. B., TSH stimulation of nucleotide and RNA synthesis in bovine thyroid slices. Abstracts of brief communications. Proc. 3rd Intern. Congr. Endocrinol., Mexico City, 1968 Intern. Congr. Ser. No. 157, p. 14. Excerpta Med. Found., Amsterdam, 1968. M1. Macchia, V., and Pastan, I., Action of phospholipase C on the thyroid. Abolition of the response to thyroid-stimulating hormone. J. Biol. Chem. 242, 1864-1869 (1967). M2. MacIntyre, I., Calcitonin: Some recent advances. Proc. Srd Intern. Congr. Endocrinol., Mexico City. Excerpta Med. Found., Amsterdam, 1969. M3. MacIntyre, I., Parsons, J. A., and Robinson, C. J., Effect of thyrocalcitonin on blood-bone calcium equilibrium in perfused tibia of the cat. J . Physiol. (London) 191, 393-405 (1967).

420

ROBERT D. LEEPER

M4. McKenzie, J. M., The bioassay of thyrotropin in serum. Endocrinology 63,372-382 (1958). M5. McKensie, J. M., Fractionation of plasma containing the long acting thyroid stimulator. J . Biol. Chem. 237, 3571-3572 (1962). M6. McKensie, J. M., Neonatal Graves’ disease. J . Clin. Endocrinol. Metab. 24, 660-668 (1964). M7. McKensie, J. M., The long-acting thyroid stimulator: Its role in Graves’ disease. Recent Progr. Hormone Res. 23, 1-46 (1967). M8. McKenrie, J. M., Humoral factors in the pathogenesis of Graves’ disease. Physiol. Rw. 48, 252-310 (1968). M9. McKenzie, J. M., and Fishman, J., Effects of antiserum in bioassay of thyrotropin and thyroid activator of hyperthyroidism. Proc. SOC.Exptl. Biol. Med. 106, 126-128 (1960). M10. McKenzie, J. M., and McCullach, E. P., Observations against a causal relationship between long-acting thyroid stimulator and ophthalmopathy in Graves’ disease. J . Clin. Endocrinol. Metab. 28, 1177-1182 (1968). M11. McKenaie, J. M., and Williamson, A., Experience with the bioassay of the longacting thyroid stimulator. J . Clin. Endocrinol. Metab. 26, 518-526 (1966). M12. McKensie, J. M., Adiga, P. R., and Murthy, P. V. N., Effects of actinomycin-D, cycloheximide and puromycin on thyroid stimulation. Endocrinology 83, 11321139 (1968). M13. Meek, J. C., Jones, A. E., Lewis, V. J., and Vanderlaan, W. P., Characterization of the long-acting thyroid stimulator of Graves’ disease. Proc. Natl. Acad. Sci. 77.8. 62, 342-349 (1964). M14. Melvin, K. E. W., and Tashjian, A. H., Jr., Syndrome of excessive thyrocalcitonin produced by medullary carcinoma of the thyroid. Proc. Natl. Acad. Sci. U. S . 69, 1216-1222 (1968). M15. Milhaud, G., and Moukhtar, M. S., Hypophysectomie et thyrocalcitonine. Compt. Rend. 260, 3179-3182 (1965). M16. Milhaud, G., Tubiana, M., Parmentier, C., and Coutris, G., Epithelioma de la thyroide dcretant de la thyrocalcitonine. Compt. Rend. 266, 608-610 (1968). M17. Mitchell, M. L., Resin uptake of radiothyroxine in sera from non-pregnant and pregnant women. J . Clin. Endocrinol. Metab. 18, 1437-1439 (1958). M18. Morain, W. D., and Aliapoulios, M. A., Comparative hypocalcemic effects of thyrocalcitonin and glucagon. Abstracts of brief communications. Proc. 3rd Intern. Congr. Endocrinol., Mem’co City, 1968 Intern. Congr. Ser. No. 157, p. 6. Excerpta Med. Found., Amsterdam, 1968. N1. Nauman, J. A., Nauman, A., and Werner, S. C., Total and free triiodthyronine in human serum. J . Clin. Invest. 46, 1346-1355 (1967). N2. Neher, R., Riniker, B., Zuber, H., Rittell, W., and Kahnt, F. W., Thyrocalcitonin. 11. Struktur von Thyrocalcitonin. Helv. Chim. Acta 61, 917-924 (1968). N3. Nikolai, T. F., and Seal, U. S., X-chromosome linked inheritance of thyroxinebinding globulin deficiency. J . Clin. Endocrinol. Metab. 27, 1515-1520 (1967). N4. Nunez, E. A., Gould, R. P., Hamilton, D. W., Hayward, J. S., and Holt, S. J., Seasonal changes in the fine structure of the basal granular cells of the bat thyroid. J . Cell Sci. 2, 401-410 (1967). 01. Odell, W. D., Wilber, J. F., and Paul, W. E., Radioimmunoassay of thyrotropin in human serum. J . Clin. Endorrinol. Metab. 26, 1179-1188 (1965). 02. Odell, W. D., Wilber, J. F., and Utiger, R. D., Studies of thyrotropin physiology by means of radioimmunoassay. Recent Progr. Hormone Res. 23, 47-85 (1967).

BIOCHEMISTRY O F THYROID REGULATION

42 1

03. Odell, W. D., Wilber, J. F., and Condliffe, P. G., Estimation of the secretion rate of thyrotropin in man. J. Clin. Invest. 46, 953-959 (1967). 04. Odell, W. D., Bates, R. W., Rivlin, R. S., Lipsett, M. B., and Hertz, R., Increased thyroid function without clinical hyperthyroidism in patients with choriocarcinoma. J. Clin. Endocrinol. Metab. 23, 658-664 (1963). 05. Oppenheimer, J. H., Role of plasma proteins in the binding distribution and metabolism of the thyroid hormones. New Engl. J . Med. 278, 1153-1162 (1968). 06. Oppenheimer, J. H., and Surks, M. I., Determination of free thyroxine in human serum: theoretical and experimental analysis. J. Clin. Endorrinot. Metab. 24, 785-793 (1964). 07. Oppenheimer, J. H., and Tavernetti, R. P., Studies on thyroxine-diphenylhydantoin interaction: Effect of 5,5’-diphenylhydantoin on displacement of Lthyroxine from thyroxinebinding globulin (TBG). Endocrinology 71,496-504 (1962). 08. Oppenheimer, J. H., Bernstein, G., and Hasan, J., Estimation of rapidly exchangeable cellular thyroxine from plasma disappearance curvea of simultaneously administered thyroxinela11 and albumin-1261.J. Clin. Invest. 46, 762-777 (1967). 09. Oppenheimer, J. H., Bernstein, G., and Surks, M. I., Increased thyroxine turnover after stimulation of hepatocellular binding of thyroxine by phenobarbital. J. Clin. Invest. 47, 1399-1406 (1968). 010. Oppenheimer, J. H., Martinez, M., and Bernstein, G., Determination of madmd binding capacity and protein concentration of thyroxinebinding prealbumin in human serum. J. Lab. Clin. Med. 67, 500-509 (1966). 011. Oppenheimer, J. H., Squef, R., Surks, M. I., and Hauer, H., Binding of thyroxine by serum proteins evaluated by equilibrium dialysis and electrophoretic techniques: Alterations in non-thyroidal illness. J. Clin. Invest. 42, 1769-1782 (1963). 012. Oppenheimer, J. H., Surks, M. I., Bernstein, G., and Smith, J. C., Metabolism of iodinel31-labeled thyroxinebinding pre albumin in man. Science 149, 748-751 (1965). 013. Oppenheimer, J. H., Surks, M. I., Smith, J. C., and Squef, R., Isolation and characterization of human thyroxine-binding pre albumin. J. Bid. C h m . 240, 173-180 (1965). P1. Pastan, I., The effect of dibutyryl cyclic 3’,5’-AMP on the thyroid. Bwchem. Bwphys. Res. Commun. 26, 14-16 (1966). P2. Pastan, I., and Katzen, R., Activation of adenyl-cyclase in thyroid homogenate by thyroid stimulating hormone. Biochem. Biophys. Res. Cmmun. 29, 792-798 (1967). P3. Pastan, I., Macchia, V., and Katzen, R., Effect of fluoride on the metabolic activity of thyroid slices. Endomnology 83, 157-160 (1968). P4. Pastan, I., Roth, J., and Macchia, V., Binding of hormone to tissue: The first step in polypeptide hormone action. Proc. Natl. Acad. Sci. U.S. 66, 1802-1809 (1966). P5. Pearse, A. G. E., Cytochemistry of thyroid C cells and their relationship to calcitonin. Proc. Roy. SOC.(London) B164, 478487 (1966). P6. Posternak, T., Sutherland, E. W., and Henion, W. F., Derivatives of cyclic 3’,5’-adenosine monophosphate. Biochim. Biophys. Acta 66, 558 (1962). P7. Potts, J. T., Jr., Niall, H. D., Keutmann, H. T., Brewer, H. B., Jr., and Deftos, L. J., Amino acid sequence of porcine thyrocalcitonin. Proc. Natl. Acad. Sci. U.S. 69, 1321-1328 (1968). P8. Purdy, R. H., Woeber, K. A., Holloway, M. T., and Ingbar, S. H., Preparation of crystalline thyroxinebinding pre albumin from human plasma. Biochemistry 4, 1888-1895 (1965).

422

ROBEBT D. L m E R

R1. Recant, L., and Riggs, D. S., Thyroid function in nephrosis. J . Clin. Invest. 31, 789-797 (1952). R2. Redding, T. W., and Schally, A. V., Depletion of pituitary thyrotropic hormone by thyrotropin-releasing factor. Endocrinology 81, 918-921 (1967). R3. Redding, T. W., and Schally, A. V., I n vitro studies with thyrotropin releasing factor. Proc. Soc. Exptl. Biol. Med. 126, 320-325 (1967). R4. Redding, T. W., Bowers, C. Y., and Schally, A. V., An i n vivo assay for thyrotropin releasing factor. Endocrinology 79, 229-236 (1966). R5. Refetoff, S., and Selenkow, H. A., Familial thyroxine binding globulin deficiency in a patient with Turner’s syndrome (xo). New Engl. J. Med. 278, 1081-1087 (1968). R6. Reichlin, S., and Utiger, R., Regulation of the pituitary-thyroid axis in man: Relationship of TSH concentration to concentration of free and total thyroxine in plasma. J . Clin. Endocrinol. Metab. 27, 251-255 (1967). R7. Rittel, W., Brugger, M., Kamber, B., and Sieber, P., Thyrocalcitonin. 111. Die Synthese des Thyrocalcitonine. Helv. Chim. Acta 61,924928 (1968). R8. Robbins, J., and Rall, J. E., Zone electrophoresis in ater paper of serum I131 after radioiodine administration. Proc. Soc. Exptl. Biol. Med. 81, 530-536 (1952). R9. Robbins, J., and Rall, J. E., 11. Hormone transportation in circulation: Interactions of thyroid hormones and proteins in biological fluids. Recent Progr. Hormone Res. 13, 166-208 (1957). R10. Robbins, J., and Rall, J. E., The iodine-containing hormones. I n “Hormones in Blood” (C. H. Gray and A. L. Bacharach, eds.), pp. 383-489. Academic Press, New York, 1967. R11. Ross, J. E., and Tapley, D., Effect of various analogues on the binding of thyroxine to thyroxine-binding globulin and pre albumin. Endocrinology 79, 493-504 (1966). S l . Sakiz, E., and Guillemin, R., Inverse effects of purified hypothalanic T R F on the acute secretion of TSH and ACTH. Endocrinology 77, 797-801 (1965). 52. Schally, A. V., Bowers, C. Y., and Redding, T. W., Purification of thyrotropic hormone-releasing factor from bovine hypothalamus. Endocrinobgy 78, 726-732 (1966). S3. Schally, A. V., Bowers, C. Y., Redding, T. W., and Barrett, J. F., Isolation of thyrotropin releasing factor (TRF) from porcine hypothalamus. Biochem. Biophys. Res. Commun. 26, 165-169 (1966). S4. Schatz, D. L., Serum free thyroxine and thyroxine-binding protein studies in patients with supraventricular tachycardias. J. Clin. Endocrinol. Metab. 27, 165-172 (1967). 55. Schillinglaw, J., and Utiger, R. D., Failure of retro-orbital tissue to neutralize the biological activity of the long-acting thyroid stimulator. J . Clin. Endocrinol. Metab. 28, 1069-1070 (1968). S6. Schreiber, V., EckertovB, A., Franc, Z., KoEi, J., Rybak, M., and KmentovB, V., Effect of a fraction of bovine hypothalamic extract on the release of TSH by rat adenohypophyses in vitro. Ezperientia 17, 264-265 (1961). S7. Schussler, G. S., and Vance, V. K., Effect of thyroid suppressive doses of triiodothyronine on thyroxine turnover and on the free thyroxine factor. J . CKn. Invest. 47, 720-728 (1968). S8. Schwarts, H. L., Carter, A. C., Kydd, D. M., and Gordon, A. S., Relationship of red blood cell ‘3*I-L-triiodothyronine binding coefficient and cell maturation. 11. Effect of cell age and metalrmlic inhihitors. Endocrinology 80, 65-68 (1967). S9. Scott, T. W., Jay, S. M., and Freinkel, N., Further studies on the action of

BIOCHEMISTRY OF THYROID REGULATION

423

pituitary thyrotropin on the individual phosphatides of thyroid tissue. Endocrinology 79, 591-600 (1966). S10. Seal, U. S., and Doe, R. P., Purification, some properties and composition of corticosteroid and thyroxin binding globulins from human sera. Proc. bnd Intern. Congr. Endocrinol., London, 1964 Intern. Congr. Ser. No. 83, pp. 325-328. Excerpta Found., Med. Amsterdam, 1965. S l l . Sinha, D., and Meites, J., Effects of thyroidectomy and thyroxine on hypothalamic concentration of ‘thyrotropin releasing factor’ and pituitary content of thyrotropin in rats. Neuroendocrinology ( N . Y.) 1, 4-14 (1965). 512. Smith, P. E., and Smith, I. P., The repair and activation of the thyroid in the hypophysectomized tadpole by the parenteral administration of fresh anterior lobe of the bovine hypophysis. J . Med. Res. 43, 267-283 (1922). S13. Socolow, E. L., Woeber, K. A., Purdy, R. H., Holloway, M. T., and Ingbar, S. H., Preparation of 1131-labeled human serum prealbumin and its metabolism in normal and sick patients. J . Clin. Invest. 44, 160Ck1609 (1965). 814. Solomon, D. H., and Beall, G. N., Thyroiddimulatiug activity in the serum of immunized rabbits. 11. Nature of thyroid-stimulating material. J . Clin. Endocrinol. Metab. 28, 1496-1502 (1968). 515. Solomon, S. H., and McKenzie, J. M., Release of thyrotropin by rat pituitary gland in vitro. Endocrinology 78, 699-706 (1966). S16. Sterling, K., Hamada, S., Newman, E. S., Brenner, M. A., and Inada, M., Preparation and properties of thyroxine binding alpha-globulin (TBG). Abstracts of brief communications. Proc. 3rd Intern. Congr. Endocrinol., Mexico City, 1968 Intern. Congr. Ser. No. 157, p. 3. Excerpta Med. Found., Amsterdam, 1968. S17. Sterling, K., and Brenner, M. A., Free thyroxine in human serum: Simplified measurement with aid of magnesium precipitation. J . Clin. Invest. 46, 153-163 (1966). 518. Sterling, K., and Hegedus, A., Measurement of free thyroxine concentration in human serum. J . Clin. Invest. 41, 1031-1040 (1962). S19. Sterling, K., and Tabachnick, M., Paper electrophoretic demonstration of thyroxine-binding prealbumin fraction in serum. Endocrinology 68, 1073-1075 (1961). S20. Sterling, K., and Tabachnick, M., Resin uptake of I-’z1 triiodothyronine as test of thyroid function. J . Clin. Endocrinol. Metab. 21, 456-464 (1961). 521. Stouffer, J. E., Jaakonmiiki, P. I., and Wenger, T. J., Gas-liquid chromatographic separation of thyroid hormones. Biochim. Biuphys. Actu 127, 261-263 (1966). 522. Sturtridge, W. C., and Kumar, M. A., Hypocalcemic activity in human plasma. Abstracts of brief communications. Proc. 3rd Intern. Congr. Endocrinol., Mexico City, 1968 Intern. Congr. Ser. No. 157, p. 8. Excerpta Med. Found., Amsterdam, 1968. S23. Sunshine, P., Kusumoto, H., and Kriss, J. P., Survival time of circulating longacting thyroid stimulator in neonatal thyrotoxicosis: Implications for diagnosis and therapy of this disorder. Pediatrics 36, 869-876 (1965). S24. Surks, M. I., and Oppenheimer, J. H., Postoperative changes in concentration of thyroxinebinding prealbumin and serum free thyroxine. J . Clin. Endocrinol. Metab. 24, 794-802 (1964). S25. Sutherland, J. E., Robison, G. A., and Butcher, R. W., Some aspects of the biological role of adenosine 3’,5’-monophosphate (cyclic AMP). C i r c d a t i m 37, 279-306 (1968). T1. Tanaka, S., and Stan; P., Euthyroid man without thyroxinebinding globulin. J . Clin. Endocrinol. Metub. 19, 485-487 (1959).

424

ROBERT D. LEEPER

T2. Tanaka, S., and Starr, P., Clinical and physiological observations in patient with idiopathic decrease in thyroxinebinding globulin of plasma. J . Clin. Invest. 40, 2053-2063 (1961). T3. Tashjian, A. H., Jr., Homeostasis of plasma calcium: Effects of Actinomycin D, parathyroidectomy and thyrocalcitonin. Endocrinology 77, 375-381 (1965). T4. Tashjian, A. H., Jr., and Voelkel, E. F., Decreased thyrocalcitonin in thyroid glands from patients with hyperparathyroidism. J . Clin. Endocrinol. Metub. 27, 1353-1357 (1967). T5. Tashjian, A. H., Jr., Frantz, A. G., and Lee, J. B., Pseudohypoparathyroidism: Assays of parathyroid hormone and thyrocalcitonin. PTOC. Natl. A d . Sci. U.S. 66, 1138-1142 (1966). U1. Utiger, R. D., Radioimrnunoassay of human plasma thyrotropin. J . Clin. Invest. 44, 1277-1286 (1965). U2. Utiger, R. D., Odell, W. D., and Condliffe, P. G., Immunologic studies of purified human and bovine thyrotropin. Endocrinology 73, 359-365 (1963). V1. Vale, W., Presence of calcium ions as a requisite for the i n vitro stimulation of TSH release by hypothalamic TRF. Experientia 23, 853-855 (1967). V2. Vale, W., Burgus, R., and Guillemin, R., Competition between thyroxine and TRF at the pituitary level in the release of TSH. Proc. SOC. Exptl. Biol. Med. 126, 210-213 (1967). W1. Werner, S. C., Spooner, M., and Hamilton, H., Further evidence that hyperthyroidism (Graves’ disease) is not hyperpituitarisrn: Effects of tri-iodothyronine and sodium iodide. J . Clin. Endocrinol. Metab. 16, 715-723 (1955). W2. Wilber, J. F., and Utiger, R. D., In vitro studies on mechanism of action of thyExptl. Biol. Med. 127, 488-490 (1968). rotropin-releasing factor. Proc. SOC. W3. Woeber, K. A., and Ingbar, S. H., The contribution of thyroxine-binding prealbumin to the binding of thyroxine in human serum BS assessed by irnmunoadsorption. J . Clin. Inved. 47, 1710-1721 (1968). W4. Wolff, J., Staudgert, M. E., and Rall, J. E., Thyroxine displacement from serum proteins and depression of serum protein bound iodine by certain drugs. J . Clin. Invest. 40, 1373-1379 (1961). W5. Wyse, E. P., McConahey, W. M., Woolner, L. B., Scholz, D. A., and Kearns, T. P., Opthalmopathy with hyperthyroidism in patients with histologic Hashimoto’s thyroiditis. J . Clin. Endocrinol. Metab. 28, 1623-1629 (1968). Z1. Zaninovich, A. A., Farach, H., Ezrin, C., and Volpe, R., Lack of significant binding of Gtriiodothyronine by thyroxine-binding globulin in vivo as demonstrated by acute disappearance of lal-I-labeled triiodothyronine. J . Clin. Invest. 46, 1290-1301 (1966).

AUTHOR INDEX Numbers in parentheses are reference numbers and indicate that a n author's work is referred to, although his name is not cited in the text. Numbers in italics show the page on which the complete reference is listed.

A Aaron, A. H., 311(A1), 364 Aarskog, D., 186(A1), ,901 Abbott, W. E., 19(A1), 20(Hll), 24(A1),

4% 48

Allen, F. M., 28(A3), 30, &? Allen, S. H. G., 162, 216 Allen, T. H., 327(A7), 340(A7), 364 Allen, W. M., 67(M5), 69(M5), 70(M5), 78(M5), 136, 149(A4), 161(A4), ,901 Alliot, M., 327(A8), 364 Almersjo, O., 326(B16), 366 Alonso, C., 160(D12), 206 Alsatt, E., 171(PB), 211 Althausen, T. L., 318(A9), 324(C23), 325(C24), 355(C25), 364, 369 Altman, M., 400(A8), 414 Altura, B. M., 19(A4), 42 Alusaker, J. O., 414 Ambrose, D., 221, 297 Amdur, M . O., 18(W6), 64 Anders, M. W., 235, 240, 278, 297 Andreoli, M., 390(C2), 393(C2), 416 Andrews, W. H. H., 317(All, A14), 318(A10, All, A13), 319, 320(A15), 325(A11, A12), 352(A14), 364 Anilane, J., 348(H23), 373 Ansorge, J., 316(H16), 373 Anthony, D. S., 250(B22), 253(B22), 255(B22), 256(B22), 298 Anthony, W. L., 348(H23), 373 Appleby, J. I., 149(A7), 160, 161(A6), 177(A7), 2001 Aranda-Rosell, A., 350(R8), 380 Archibald, R. M., 6(V1), 64 Arcilla, R. A., 339(W1), 384 Arends, A., 354(M10), 376 Arias, I. M., 320(C36), 346(A16), 347(A16), 350(A16, SS), 351(A16, K12, SS), 357(S11), 364, 369, 376,

Abe, H., 403(K7), 419 Abel, J. J., 311, 312(A2), 364 Abide, J. K., 3?9(P7), 360(P7), 379 Abramovich, D. R., 201 Ackman, R. G., 290(A1), 297 Acocella, G., 347(A3), 364 Adams, D. D., 393, 395(A3, A4), 397 401, 414 Adams, M. A., 6 ( T l ) , 20(T1), 63 Adams, R., 343(C30), 369 Adhikary, P. M., 147(513), 185(S13), 213, 294, 295, 296, 657 Adiga, P. R., 401(M12), 420 Adler, A., 311(A4), 314(A4), 315(A4), 364, 383 Adlercreutz, H., 73(A1, A3), 121(A2), l22(A4), 130 Adlersberg, D., 17(B12), 43 Ahrens, E. H., 29O(Ol), 304, 354(A5), 364 Aikens, D. A., 221(V5a), 306 Aitken, E. H., 155, 201 Aitken, E. M., 99(P10), 108, 137 Akrawi, L., 412(G6), 417 Albot, G., 316(T7), 334(T7, TS), 355(T7), 383 Alberta, M., 196(B20), 202 Albertsen, K., 19(A1), 24(A1), 42 Albright, E. C., 404(L1), 419 Aldridge, W. N., 8(A2), 42 381 Alexander, R. J., 271, 997 Armstrong, P. B., 350(A17), 364 Aliapoulios, M. A., 413(A5-A7, M18), Arnaud, C. D., 403(A10), 414 414, 420 Arturson, G., 13(A5), $2 Alininosa, L., 314(530a), 38.2 Ashby, M. M., 7(S17), 8(517), 63 Allan, J. S., 343(A6), 350(A6), 3G4 Aspinall, G. O., 269(A8), 297 425

426

AUTHOR INDEX

Assimos, A., 279(M6), 304 Atkinson, A. P., 232, 297 Attal, J., 241, 297 Austin, J. H., 328(S44), 340(S44), 383 Averill, R. L., 391(All), 392(A11), 4 4 Avioli, L. V., 296(A11), 297

B Bacharach, A. L., 60(G9), 99, I S 3 Bachmann, R. C., 290(E3), 300 Badawi, H . S., 353(B1), 364 Baggenstoss, A. H., 354(B61), 367 Baghdiantz, A., 411(F6), 417 Bailey, J. J., 293(B1), 297 Baillie, A. H., 156(B1, C I ) , 201, 204 Baird, C. W., 168(B2), 201 Baird, D. T., 65(B1), 102(Bl), 115(B1), 121, 127(B1), 130 Baker, D. V., 18(W6), 64 Baker, K. J., 316(B2, B3), 320(B3), 321(B3, J5), 337(B3), 357(B3), 366, 374 Baker, L., 340(B3a), 366 Bakke, J. L., 394(B1), 396(B2), 397(B1, B2), 414 Balegno, H. F., 17(N4), 18(N4), 61 Balkisson, B., 328(B4), 330(B4), 340(B4), 343(B4), 366 Balko, J., 18(W6), 64 Ball, M. R., 9(M14), 13(M14), 14, 21(M14), 25, 51 Baltz, J. I., 327(M14), 329(M13, M14), 331(M14), 377 Banasgah, E. F., 339(B5), 366 BanagzaJs, E. F., 330(S37), 382 Bandi, L., 71(R5), 137, 185(R12), 818 Barad, B., 343(C19), 357(C19), 368 Barber-Riley, G., 320(B7), 325(B9, BlO), 332, 345(B6, B7), 346(B6), 348(B11), 366 Barboriak, J. J., 296(H9), 301 Barbosa, L. T., 196(R15), 212 Barbour, B. H., 35(S10a), 63 Barcham, I., 20(C15), 44 Barclay, A. E., 4(T5), 64 Bard, J., 279(L14), 304 Bardin, C. W., 65(B2), 69(B2), 70(B2), 75(L12), 78(B2), 79, 104(B2), 130, 135

Barnes, M. M., 322(B13), 366

Barnett, S. A., 27, 42 Baron, D. N., 83(GIO), 85(GIO), 96(G10), 98(G10), 105(G10), 13s

Barr, J. K., 222, 305 Barr, M., 157(B3), 159, 160(B3, 1x21, 202, 207 Barr, P. O., 17, 21(B5), 30, -@ Barrett, J. F., 390(53), 391(53), &2 Barry, R. D., 240, 245, 246, 297, 301 Bartler, F. M., 34(H17), 4.9 Bartter, F. C., 35(S10a), 63, 72(G4a), 133

Basila, M. R., 228, 297 Bates, R. W., 396(B3), 397, 401(04), 414, Baulieu, E.-E., 69, 71(B4, B5), 83(B3), 130, 148(J5), 174(B4), 185(B5), 20.2 Baumann, E., 322(B14), 366 Bayer, E., 269, 297 Beall, G. N., 403(514), 423 Bearn, A. G., 340(S24), 382 Beas, F., 175(B6), 202 Beaton, J. R., 28, 42 Beattie, A. D., 70(G6), 133 Beaufay, H., 12(D8), 46 Beazley, J. M., 336(T6), 343(T6), 344(T6), 360(T6), 383 Beck, E. C., 311(A1), 364 Beck, J. C., 172, 173, 202 Becker, D. V., 324(B14a), 366, 414 Beckloff, S., 148, 204, 240(C5), 299 Beckmann, I. N., 99(02), 136 Begon, F., 324(B15), 366 Beierwaltes, W. H., 406(E3), 409(B5), 414, 4l6 Bel, C., 344(V4), 384 Belfrage, S., 17(B7), 42 Beling, C. G . , 121, 130 Bell, C. C., 34(H17), 49 Bell, P. H., 411(B6), 414 Bell, T. K., 31(J3), 49 Beltocchi, G., 16(M5), 60 96 Benagiano, G., 174(D6), (D6).. 906,. 209 Bender, J., 324(L5), 330(L5), 338(L5), 340(L5), 342(L5), 352(L5), 36O(L5), 376 Ben-Ezzer, J., 320(C36), 369 Bengmark, S., 326, 366

AUTHOB INDEX

Bcngtsson, G., 163(B8), 202 Benhamou, G., 402(E2), 416 Benhamou, J. P., 324(B15), 565 Benirschke, K., 156(B22, B23), 202, 203 Benjamin, D. C., 37(W7), 54 Benner, M. C., 142(B9), 202 Bennhold, H., 316(B17), 317(B17), 365 Benson, E., 352(M21), 354(M21), 377 Bentley, R., 269(B5, 5171, 297, 806 Bergada, C., 192(D3), 205 Bergstrand, C. G., 147(B10), 202 Berk, J. E., 353(B18), 365 Berliner, D. L., 73(B7), 130 Berman, J. R., 353(S12), 381 Berman, M., 75(B8), 130 Bernstein, G., 405(010), 406(010), 408(08, 0121, 409(09), 410(B7,

427

Birkenhager, W. H., 345(B23), 366' Birkett, D. J., 343(B24), 366 Bizard, G., 345(B25), 348(B25, B26), 350(B25), 366 Bjornesjo, K. B., 18(BSa), 43 Black, E. G., 82(C5), I32 Black, W., 257(H1), 801 Blahd, W. H., 312(T10), 341(N9, TlO), 355(N9), 378, 383 Blakemore, A. H., 339(B43), 353(B43), 367

Blankenhorn, D. H., 296(J5), 308 Blatt, W. F., 28, 4.3 Blau, K., 257(B14), 898 Blizzard, R. M., 196(B20), 202 Bloch, E., 156(B22, B23), 2W(B21), g08, 203

Block, H. S., 3(Zl), 34(21), 55 ow, 4147 @1 Block, J. H., 4(L8), 50 Bernstein, S., 62(B9), 130 Block, L., 342(S21), 382 Beroza, M., 294, 298 Blondheim, S. H., 317(B27), 337(B28), Berson, S. A,, 398, 414 354(A5), 364, 366 Berthelot, P., 345(B19), 365 Bertrand, J., 142(B12, L17), 147(B12), Blondin, G. A., 294, 298 149, 156(B12, L17), 164(Mll), Bloom, G., 403(F2), 417 168(Mll), 176(B13), 178, 180(B12), Bloom, W., 311(B29), 366 Bloomfield, A. L., 311(R25), 381 186(B12), 187, 198, 202, 209, 210 Blumberg, N., 356(B30), 866 Besch, N. F., 240(B2), 297 Blumquist, B. E., 318(A9), 364 Besch, P. K., 240(B2), 297 Bochey, B., 337(M5), 345(M5), 376 Bevan, B. R., 162(B14), 202 Bockus, H., 366 Beyer, K. H., 337(M5), 345(M5), 376 Bodfish, R. E., 409(F3), 417 Bickel, E. V., 3(Z1), 34(Z1), 55 Bogden, A. E., 17(B10, B l l ) , 4.3 Bier, M., 257(B13), 298 Boggs, T., 180(B32), 187(B32), 203 Biggs, R. M., 250(B22), 253(B22), Bogoch, A., 366 255(B22), 256(B22), 298 Bohemen, J., 231, 233, 298 Bigham, R. S., 21(H16), 48 Bollinger, H. R., 296(K10), SO3 Bijvoct, 0. L. M., 413(B9), 414 Billing, B. H., 9(W15), lO(W15, W17), Rollman, J. L., 312(H27), 318(H27), 322 (H14), 323(H14), 325(H28, K20, 21(W16), 54, 65, 340(S24), 345(B19, K32), 327(H14), 331(H27), 338(H14, B21), 347(A3, B21a), 354(B20), 364, H28), 343(H28), 349(H28), 351 365, 366, 382 (H28), 352(H27), 355(H14), 360 Biorck, G., 324(B22), 325(B22), 366 (H14, H27, H28), 366, 373, 374 Birchall, K., 142(C6), 147(B16, B18, C6), 148, 149(B15), 150, 151(B18), Bolt, R. J., 345(B33), 366 155(B18), 166(B16, C7), 174(B17, Bolth, E., 156(B24), 161(H6), 174(B25), 185(B24), 203, 207 B18, C6), 177(C6), 178(B16, B18, C6), 179(C6), 183(C7), 187(C6), Bondy, P. K., 71, 115(Bll), 130 Bongiovanni, A. M., 71(B14, B15), 77 195(C7), 802, 204 (B13, B15), 126(B15), 130, 153, 156 Bird, C. E., 163(S25), 165(S25), 214 (B29, G2, G3), 159(S31), 180(B32), Birke, G., 17(B5), 20, 21(B5), 30(B5), 187(B32), 189(B30, B311, 191(B26, 43, 66, 77(B10), 130, 191(B19), 202

428

AUTHOR INDEX

B31), 192(B30, B31), 193(B31), 195 (B27, B31, E6, G3), $03, 206 Bonnichsen, R., 279(B17), g98 Bonsnes, R. W., 343(C19), 357(C19), 368 Booth, C. O., 355(N4), 378 Booth, J., 322(B34-B36), 337(B35), 360 (B35), 3G6 Borell, U., 400, 416 Borg, A., 252(B37), 366 Borrell, S., 105(B16), 128(B16), 131 Bosak, E. T., 17(B12), 43 Bottari, P. M., 398(B11), 416 Bottinger, C. E., 18(B13), 43 Bourne, W., 349(FUO), 380 Bowers, C. Y., 390(S2, S), 391(B14, R4, S3), 392(B12, B13, B15, B16), 416, 422

Bowman, H. S., 18(B14), 43 Boyd, J. F., 196(B33), 203 Boyland, E., 322(B34-B36, B39), 337 (B35), 338(B38), 345(B38), 348 (B38), 351(B38), 360(B35), 366 Boyne, A. W., 37(C7), 44 Boiovi6, J., 13(B15), 43 Bradford, B. K., 336(W31), 352(W31), 386 Bradley, G. P., 339(B42), 340(B42), 353 (B42), 367 Bradley, M. H., 350(K14), 351(K14, W2), 376, 384 Bradley, R. F., 336(N7), 337(N7), 378 Bradley, S. E., 316(B2, B3), 320(B3), 321(B3, J5), 325(M16), 327(11), 329 (Il), 331(11), 333(11), 335(W12, W13), 336(P11, W13), 337(B3, I l ) , 339(B42, B43), 340(B42), 343(P11), 344(B40, B41, C37), 351(M16), 352 (Il), 356(11), 357(B3), 359(11), 366, 367, 369, 379, 586 Bradlow, H. L., 75(F7), 82(P7), 113, lS1, 133, 178(F8), 191(F9), 206 Brady, L. W., 328(B44), 340(B44), 567 Braman, R. S., 221, 298 Branborg, L. L., 354(D9), 370 Brandt, R., 9(H6), 48 Brauer, R. W., 312(B46), 318(B51, K17), 320(B48), 321(K18), 322(B46, K18, K19), 323(B46, K19), 324(B49), 325 (2, K20, K21, N3), 331, 341(B45, B50, B51), 342(01), 343(01), 344

(B47), 351(B49), 360(K18), 363, 367, 676, 618 Braunsberg, H., 91, 95, 99, 104(B17), 131 Braverman, L. E., 404(16), 45(16), 410 (B17, 161, 416, 418 Bravo, E. L., 240, 241, 298 Bray, H. G., 322(B52), 323(B52), 367 Breed, E. S., 20(C15), 44 Brem, J. M., 354(B54), 355(B54), 367 Brenner, M. A., 404(S16), 405(S17), 407 (S17), 409(S17), 410(S17), 423 Brette, R., 347(B55), S67 Breuer, H., 205 Brewer, D. B., 196(B35), 203 Brewer, H. B., Jr., 411(P7), 491 Brewer, T. H., 343(B56), 357(B56), 367 Brief, D. K., 9(R7), 62 Briefer, C., Jr., 237, 303 Brochmann-Hanssen, A., 232, B 8 Brodie, A. M., 89(B18), 94(B18), 115 (B18), 131 Brodie, B. B., 186(P9), $11 Brody, S., 4.3 Brody, S. M., 296(K2), 303 Broholt, J., 322(B57), 327(B57), 342 (B571, 587 Brokaw, R., 344(B58), 367 Broker, L. G. S., 312(F5), 371 Brook, J. L., 250, 253, 255, 256, 298 Brooks, C. J. W., 131, 147(S13), 155 (HlO), 185(S13), 207, 213, 238(H14), 240, 249(B24), 250, 251, 252, g98, 301, 302 Brooks, D. K., 9(B17), 43 Brooks, L., 328(12), 329(12), 340(12), 374

Brooks, R. V., 83(G10), 85(G10), 96 (GlO), 98(G10), 105(G10), 133, 191 (B36), 203, 237, %98 Bro-Rasmussen, F., 155(B38), 164(B37, B43), 172(B37), 180(B43), 198(B43), 203

Brough, A. J., 196(C2), 204 Brouwer, S., 327(M14), 329(M14), 331 (MW, 377 Brown, J. B., 99, 104, 108, 114, 121, 131, 155(K9), 160(B39), 161(K8), 203, 208, 244, 298 Brown, T. B., 124(K8), 136 Browne, J. S. L., 20(B19, S2), 34, 43, 62

429

AUTHOR INDEX

Brownie, A. C., 89(B22), 116(B22), 123 (B22), 127, 131 Bruchoveky, N., 71, 131 Brugger, M., 411(R7), 428 Brunim, J. J., 186(P9), 211 Brunstrom, G. M., 15(01), 61 Bucana, C., 296(N1), 304 Buchanan, P. J., 348(H23), 373 Budgen, D. E., 280(C9), 299 Buergi, W., 296(C17), 299 Bulbrook, R. D., 99(52), 104(S2), 121 (S2), 137, 148(L10), 156(L9), 209 Bull, J. P., 16(B20, D l ) , 17(B20, D2), 4% 46 Burchfield, H. P., 221(B27), 298 Burger, J. D., 222(B28), 299 Burgess, J. A., 410(H9), 418 Burgher, R. D., 290(A1), 297 Burgus, R., 392(V2), 424 Burke, G., 400(B1S, B19), 416 Burke, J. F., 348(B59), 367 Burley, J., 151(U3), 155(U3, U4), 172 (U4), 177(U2, U3), 179(U3), 180 (U4), 181(U4), 187(U3), 198(C5), 204, 214 Burnett, W., 349(B60), 355(B60), 367 Burnham, H. D., 292(L4), 303 Burns, J. W., 18(H4), 48 Burrows, J. A., 370 Burstein, S., 63(T3), 74(B24), 75(T3), 82(T3), 100(B25a), 113(B25), 117 (B25a), 131, 138, 148(B40), 203 Burt, R. L., 6(B21), 43 Bush, I. E., 63(B28), 69(B28, B36), 70 (B28, 71(B28), 73(B28), S(B27, B32, B33, B33a), 86(B33a), 87 (B33a), SS(B26, B27, B34), 90(B27, B32), 93(B26, B27), 96(R4), 97(R4), lOO(B29, B33a), 102(B29, B33a), 110 (B33a), 113(B35), 114(B30), 115 (B27, B29), B33a, B34), ll’I(B30), 118(B33a), 120(R4), 126(B32), 131, 132, 137, 151, 168(B2), 201, 803 Bush, R. D., 113(S4), 138 Bussolati, G., 412(B20), 415 Butcher, R. W., 399(S25), 423 Butler, G. C., 192(B42), 903 Butt, H. R., 331(515), 336(S15), 354 (B61, S16), 355(S15), 360(515), 367, 581

Butterfield, W. J. H., 9(E6), 46 Buus, O., 155(B38), 164(B37, B43), 172 (B37), 180(B43), 198(B43), 203 Byars, B., 299 Byham, B., 337(M5), 345(M5), 376

C Cadman, W. J., 279(Cl), 299 Cahill, P. J., 28(C14), 4 Cairnie, A. B., 20(C1), 21(C1), 22, 24 (Cl), 43 Caldwell, B. F., 6(G7), 17(G7), 47 Caldwell, F. T., 21(C2), 22, 30, 31, @, 60 Callahan, R., 350(K14), 351(K14, W2), 376, 384 Callear, A. B., 223(C2), 299 Calloway, N. O., 334, 343(Cl), 368 Calman, K. C., 156(C1), 204 Calvary, E., 125(J5), 134 Calvin, H. I., 71(R5), 137, 185(R12), 21.2 Camacho, A. M., 196, 204 Cameron, E. C., 411(C10), 413(C10), 41 6 Campanacci, D., 331(T12), 354(T12), 383 Campbell, R. M., 7(C5), 19(C5), 2O(C1), 21(C1), 22(C1), 24(C1, C5), 30 (C4a, C5), 31, 37(C7), 38(C6), 43,

44 Campbell, S., 296(C4), 299 Cantarow, A., 318(C3), 326(C2), 336(C4, W31), 350(C2), 351(C3, W32), 352 (W32), 368, 386 Capella, P., 249(C3), 899 Cara, J., 175(B6), 202 Carbone, J. V., 312(C5, C6,G231, 321(C5, G23, G24), 324(KZ2), 351(D3), 355 360(C5, C6,G23), 363,368,570, 372, 376, 379 Care, A. D., 412, 416 Carlberger, G., 324(B22), 325(B22), 366 Carlsson, L. A., 20(B8), 42 Carlsson, O., 269(57), 306 Carniero, L., 402(D4, D5), 416 Caroli, J., 326(C7, CS), 336(C7, C8),368 Carpenter, M., 345(526), 38.9 Carr, E. A,, Jr., 406(E3), 416 Carr, H. E., 249(W7), 307 Carter, A. C., 164(W1), 165(W1), 180 (WI), 187(W1), 816, 407(58), 42.2

(a),

430

AUTHOR INDEX

Cartoni, G. P., 257(L8), 303 Cartter, M. A., 345(B21), 366 Casey, J. H., 3(21), 34(21), 66 Casey, J. M., 79, 104(Cl), 13.2 Cash, A. G., 400(L6), 401(L7), @9 Casper, A. G. T., 35(SlOa), 63 Cassano, C., 390(C2), 393(C2), 416 Casselman, W. G. B., 366 Cassmer, O., 160(D12), 161(C3), 204, 206 Castenfors, H., 327(C10), 329(C10), 331 (H25, H26), 339(C9), 349(H26), 368, 373 Castenfors, J., 13(B15), 43 Cates, H. B., 353(Cll), 368 Cathro, D. M., 142(C6), 147(B18, C6), 148(B18), 151(B18), 155(B18), 166 (C7), 172, 174(B17, B18, C5, C6), 175(C6), 177, 178(B18, C6), 179, 183 (C7), 187(C6), 195, 196, 198, ,802, 204 Caughey, D. E., 355(N4), 378 Caujolle, F., 314(C12), 315(C12), 368 Cautenet, B., 142(312, L17), 147(B12), 149(B12), 156(B12, L17), 176(B13), 178(B12), 180(B12), 186(B12), 198 (BE?), 202, 209 Cavalieri, R. R., 408(C3), 409(C3), 4i6 Cavallero, C., 156(C8), 159(C8), 904 Cawley, L. P., 148(C9), 204, 240, 296 (C4), 299 Cedard, L., 171(P6), dl1 Chalmers, M. I., 7(M19), 19(M19), 21 (M19), 24(M19), 61 Chalmers, T. G., 353(C13), 368 Chambaz, E. M., 238(H14), 301 Chambers, W. N., 342(C14), 368 Chandler, A. M., 17(N4), 18(N4), 61 Chanutin, A., 16(CS, GlO), 44, 47 Charbonnier, A., 326(C7), 336(C7), 368 Charro-Salgado, A. L., 147(S10), 148 (SlO), 151(SlO), 152(SlO), 153(S10), 178(S10), 195(SlO), 913 Chattoraj, S. C., 241, 244, SOY Chenderovitch, J., 352(C15), 368 Cherrick, G. R., 312(C16), 317(C16), 328 (C16), 331(C16), 337(C16), 338 (C16), 340(C16), 350(L6), 355(C16), 360(C16), 368, 3'76 Chi, Y. M., 399(K4), 419 Chiamori, N., 358(H10), 359(H10), 373

Chien, J. R., 402(K6), u 9 Childs, A. W., 348(L13), 349(L13), 353 ((3171, 368, 376 Chin, T., 271(W5), 277(W5), 307 Chinard, F. P., 326(C18), 368 Choisy, H., 359(G3), 371 Christensen, L. K., 406(C4), 416 Christhilf, S. M., 343(C19), 357(C19), 368 Christy, N. P., 185(J4), dM Chrzonszczewsky, N., 311(C20), 369 Chute, R., 34(M8), 60 Chytil, F., 6, 44 Cieplinski, E. W., 281, 288, 299 Clark, F., 407, 416 Clark, R. G., 21(C10), 24(C10), 44 Clark, S. J., 89(C2), 13.2 Clarke, D. D., 249(W6, W7), 299, 307 Clarke, R., 14, 46 Clarkson, M. J., 334(C21), 369 Clay, W. A., 19(M4), 21(M4), 60 Clayton, B. E., 177(C10), 204 Clayton, G. W., Jr., 71(B15), 77(B15), 126(B15), 130 Cleary, R. E., 153(Cll), 183(Cll), 204 Clodi, H., 347(C22), 369 Clogne, 21(W11), 64 Coad, R. A., 294(B9), 298 Coffman, G. D., 127, 136 Coggshall, V., 17(W8), 37(W8), 64 Cohen, E. S., 324(C23), 325(C24), 355 (C25), 369 Cohn, C., 326(C26), 369 Cohn, E. M., 341(C27), 369 Cohn, G. L., 71(B12), 130 Colds, A., 147(C12, C13), 155, 158(Cl2, C13), 170(C12, C13), 171, 185(C12), 804

Cole, V. W., 247, 248, 306 Cole, W. H., 312, 331(L2), 372, 376 Coleman, E. M., 27(B3), .@ Colle, E., 151(U3), 155(U3, U4), 172 (U4), 177(R8, U2, U3), 179(R8, U3), 180(R8, U4), 186(R8), 187(R8, US). 198, 204, 212, d l 4 Collins, D. A., 13, 44 Colodzin, M., 328(B44), 340(B44), 367 Comanduras, P. D., 327(M14), 329 (M14), 331(M14), 377 Combes, B., 321(C28, C31, C32), 322

431

AUTHOR INDEX

(C32), 323(C29, C32, C33, G11, S7), 324(C28, C32, H20), 343(C30), 344 (C29), 356(C29), 359(G10), 361 (H20), 369, 371, 373, 381 Conan, N. J., 344(B41), 367 Condliffe, P. G., 393, 394(C7, F4, U2), 395(03), 399 ((371, 402(03), 405(03), 415, 417, 421, 494 Conney, A. H., 73(C3), 74(C3), 132 Conrad, S., 132 Conrad, S. H., 155(C16), lSl(Cl6, P l l ) , 163(K2), 170(C16), 204, 208, 111 Cook, B., 132 Cook, D. L., 321(C34), 326(C34), 369 Cook, J. G. H., 280, 299 Cooper, C. E., 34(C12), 4-4 Cooper, C. W., 412(C1, CS), 416, 416 Cooper, D. Y., 328(B44), 340(B44), 367 Cooper, I. S., 356(C35), 369 Cooper, W., 155(G5), 206 Cope, C. L., 82(C5), 132 Cope, O.,22, 44 Copenhaver, J. H., 271(C10), 272(C10), 273, 299 Copher, G. H., 312(G9), 37.2 Copp, D. H., 411, 412, 416 Cornelius, C. E., 320(C36), 369 Corner, E. D. S., 148(L10), 156(L9), 209 Cornfield, J., 396(B3), 397(B3), 414 Corse, J. W., 256(H19), 302 Corwin, A. H., 218(K7), 303 Cost, W. S., 197, 215 Costill, D. L., 28, 44 Co Tui, 20(C15), 44 Coulter, J. R., 260, 261, 262(C11), 263, 299 Coumbis, R. J., 279(S14), 306 Coutris, G., 413(M16), &0 Coward, R. F., 35(C16, C17), 4.4 Coyle, M. G., 155(G5), 172, 196, 204, 306 Cranny, C. L., 177(C17), 204 Cranny, R. L., 177(C17), 18O(C18), 187 (CIS), 204 Cranston, W. I., 321(M20), 325(W9), 330(W9), 353(M20), 360(M20, W9), 377, 384 Crawford, B., 314(S30a), 382 Creech, B. G., 87(H4), 88(H4), 99(H4), llO(H41, 116(H4), 134 Creech, O., 331(R4), 379

Crigler, J. F., Jr., 71(G5), 94, 133 Crippen, R. C., 296(C12), 299 Criscuolo, D., 27(H2), 28(H2), 48 Crispell, K. R., 71(B12), 130 Croughs, W., 191(D4, D5), 192(D4), 606 Crowshaw, K., 88(B34), 115(B34), 132 Cruickshank, P. A., 257(C13), 299 Culbertson, J. W., 344(C37), 369 Culver, P. J., 354(C38), 370 CunMe, W. J., 413(Cll), 416 Curry, A. S., 279(C16), 280(C15, C16), 288, 299 Curry, J. J., 339(B42), 340(B42), 353 (B42), 367 Curtis, F. C., 18(H3), 48 Curtius, H. C., 240(C18), 296(C17), 299 Cushman, P., 409(15), 418 Cutbush, H., 342 (M28), 343(M28), 377 Cuthbertson, D. P., 2(C33), 3(C27, C33), 5, 6(C21), 7(C5, CZl, C27, C40), 9 (C37), 10(C37), 14(C22), 18(C34a), 19(C5, C19, C20, C21, C22, C23, C24, C25, C26, C27, C33, C34, C37, M19, MZO), 20(C1, ClS, C19, C21, C28, C35, C36), 21(C1, C21, C27, MZO), 22(C1, CZl), 24(C1, C5, C21, C22, C37, MN), 25(C22, C29, C30, C33), 30, 31 (C5, C29), 32(C31, C32), 35(C38, C39), 37(C7), 38, 43, 44, 45, 61 Cvetanovits, R. J., 223(C2), 299

D Dahl, E. V., 280(W2), 307 Dalgliesh, C. E., llO(Dl), 114, 132, 242, 249(H17), 296(D1), 299, SO2 Dalmonte, P. R., 331(T12), 354(T11, T121, 383 Dal Nogare, S., 218, 300, 302 Daly, B. M., 17(P4), 51 Davesac, J. F., 316(T7), 334(T7, TS), 355(T7), 383 Dancis, J., 71(LZ), 136 Daniel, P., 4(T5), 54 Darbre, A., 257(B14), 298 Darcey, B., 240, 300 D b l e r , C.-G., 199(D1), 206 Datta, J. K., 97(N1), 136 Daughaday, W. H., 165(D2), 206 David, R. R., 192(D3), 206

432

AUTHOR INDEX

Davidson, A. G. F., 411(C10), 413(C10), 416 Davidson, C. S., 6(T1), 20(T1), 63, 312 (C16), 317(C16), 328(C16), 331 (ClS), 337(C16), 338(C16), 340 (C16), 350(L6), 351(N1), 355(C16), 360(C16), 368, 375, 378 Davies, J. W. L., 16, 17(D2), 46, 66 Davis, J. C., 168(J7), 208 Davis, J. H., 9(H14), 21(H14), 48 Davis, J. O., 34(D3), 46 Davis, J. W., 300 Davis, M. E., 317(M25), 377 Davis, R. A., 279(D6), 300 Davis, W. D., 354(F%), 379 Dawber, N. A., 404(16), 405(16), 410 (161, 418 Deadrich, R. E., 16(D4, L2), 17(L2), 46, 49 de Armond, R., 317(W5), 384 Declasse, 7(D5), 46 Declerck-Raskin, M., 165(D7), 206 Decora, A. W., 223(D7), 300 DeCourcy, C., 69(W4), 80(W4), 126 (D2), 132, 139 de Duve, C., 12(D6, D7, D8), 46 De Fraiture, W. H., 321(D1), 336(D1), 354(M10), 360(D1), 370, $76 Deftos, L. J., 411(P7), 412(D1), 416, 421 Degenhart, H. J., 191(D4, D5), 192(D4), 605 De Graff, A. C., 331(R4), 379 de Harven, J., 18(G15), 47 Deiss, W. P., 404(L1), 419 Deiss, W. P., Jr., 413(J1), 418 Deitrick, J. E., 46 Delcourt, A., 328(D2), 331(D2), 370 Delcourt, R., 328(D2), 331(D2), 370 de Leon, S., 296(N1), 304 Dell’Acqua, S., 174(D6), 176(M2), 196 (D6), 205, 209 Dell’amore, M., 16(M5), 60 De Lorimer, A. A., 351(D3), 370 Delprat, G. D., 311(K6), 314(K6), 324 (K6), 338(D4, D5, K6), 352(K6), 370, 374 de Menibus, C. H., 197(R18), 212 de Miguel, M., 160(D12), 206 Demling, L., 339(H9), 372 De Moor, P., 165(D7), 206 Denley, M. L., 342(01), 343(01), 378

de Peretti, E., 142(L17), 156(L17), 176 (Bl31, 902, 209 De Reimer, R. H., 353(S27), 382 de Rio Lozano, I., 319, 320(A15), 364 De Robertis, E., 396(D2), 398(D2), 400 (D21, 416 Desaulles, P. A., 176(M8), 209 Despopoulos, A., 324(D6), 370 Desty, D. H., 114, 132, 223(D8), SO0 Deutsch, C., 347(D8), 370 Deutsch, E., 328(D7, 329(D7), 331(D7), 352(D7), 352(D7), 353(D7), 355 (D7), 370 Dewar, R. A., 223(D9), 300 Dibrell, W. H., 9(H14), 21(H14), 48 Dickes, R., 347(D8), 370 Dickson, E. R., 354(B61), 367 Diczfalusy, E., 71(D4, L2), 77(B10), 116(L16), 130, 135, 136, 142(D10), 155(D8, DlO), 156(B24, D9), 157 (B3), 159(B3), 160(B3, D11, D12, H2, 25, Z6), 161(P1, P12), 163(B8, S25), 165(P2, P3, S25), 167(P7), 168(D8), 174(B25, D6), 176(M2), 180(D10), 185(B24), 186(26), 191 (B19), 196(D6, D9), 202, 203, 206, 207, 209, 211, 214, 615 Dignam, W. J., 147(E2, S20), 154(E2, S20), 155(E2, S20), 156(S20), 158 (E2, S20, S22), 167(S22), 170(E2, S20), 171(E2), 5’05, 213 Diller, E. D., 72(G4a), 133 Dillon, S., 345(B33), 366 Dimick, K. P., 256(H19), 302 Dinneen, G. U., 223(D7), 300 Dion, F. R., 192(M5), 209 Dixon, J. P., 36(M3), 50 Do, F., 142(L17), 156(L17), 209 Dobriner, K., 166(D13), 183(D13), 192 (L16), 206, 209 Dobyns, B. M., 404, 416 Dodgson, I(.S., 148(D14), 206 Dodson, E., 357(518), 359(518), 381 Doe, R. P., 404, 405(S10), 42.3 Dolan, F., .@ Dole, V. P., 6(V1), 64 Dollinger, R., 351(D9), 370 Domb, A., 328(D2), 331(D2), 370 Dominguez, A. M., 218(G5), 219(G5), 278(G5), 301 Donati, R. M., 23(S18), 63

AUTHOR INDEX

Donavan, B. T., 398(B11), 416 Done, J., 343(B24), 366 Doniach, D., 402(E2), 416 Dorfman, R. I., 78(G1), 83(G1), 99(F2), 104(F2), 120(N5), 136, 133, 136, 166 (D15), 192(S15), 206, ,913 Dorrington, K. J., 402(D3, D4, D5), 403 (D3), 416 Dotti, L. B., 354(S42), 383 Dougherty, T. F., 73(B7), 130 Douglas, M., 358(J1), 360(J1), 374 Dowling, J. T., 409(D6, D7, F3), 416, 417 Downing, D. T., 293, 300 Dragstedt, C. A., 318(M26), 351(M27), 377 Dray, F., 174(B4), 185(B5), 202 Drayer, N. M., 147, 187(D16), 206 Drill, V. A,, 353(D10), 370 Driscoll, S. G., 156(V5), 180(D17), 606, 614 Drorbaugh, J. E., 162(R14), 612 Drucker, W. D., 71(R5), 137, 185(R12), 212 Druzhinina, K. V., 166(Y1), 216 Drye, J. C., 13(D10), 46 Dubin, I. N., 354(D11), 370 Dubost, P., 288, 300 Ducharme, J. R., 187, 206 Dudley, H. A. F., 7(Dll), 24(Dll, D12), 46 Dumbacher, M., 413(Hl), 418 Dumont, J. E., 399, 416 Dunagin, P. E., Jr., 296(D12), 300 Duncan, T., 412(C1), 416 Dunievitz, M., 311(K6), 314(K6), 324 (K6), 338(K6), 352(K6), 374 Dunning, H. A. B., 324(M7), 376 Durbin, R. P., 221(511), 306 Dussa, J. P., 62(B9), 130 Dutter, F., 183(P5), 211 Dutton, R., 11(R11), 62 Duval, P., 5, 46 Duyck, C., 39(M9), 60 Dyniewicz, J. M., 331(L2), 376

E Eaborn, C., 132, 239, 300 Eades, C. H., 6, 46 Eagle, J. F., 179, 606

433

Eastcott, H. H. G., 13(E2), 46 Easterling, W. E., Jr., 147(E2, S20), 154 (E2, S20), 155(E2, S20), 156(Su)), 158(E2, S20, 5221, 167(S22), 170 (E2, 5201, 171, 200, 206, g13 Eberlein, W. R., 71(B14), 77(B13), 147 (E4), 155, 156(B29), 157(E4), 159 (5311, 166, 170(E4), 171(E4), 174 (E4), 180(B32), 187(B32), 188(E4), 189(B31), 191(B31), 192(B31, E5), 193(B31), 195(B31), 200(E4), ,903, 206 EckertovL, A., 390(56), 391(S6), 42.8 Eckhardt, E. T., 347(E1), 370 Eddy, D., 28(C14), 44 Edwards, R. W. H., 177(C10), 604 Edwards, S., 322(E2), 3'70 Egdahl, R. H., 39, 46 Eglin, D., 6(G21), 7(G21), @ Ehrlich, R. M., 196(E7), 206 Eik-Nes, K. B., 60(E2), 86(E2), 87(E2), 88(E2), 89(B22), 90(E2), 97(E2), 98 (E2), 99, 104(E2), 108(E2), llO(E21, 114(E2), 116(B22, E2), 117(E2), 123 (B22), 127(B22), 131, 132, 136, 148 (Ol), 185(H1), 207, 211, 237, 241, 297, 300 Eisalo, A., 350(E3), 370 Eisenberg, H., 21(H15, H16), /fi Eisenmenger, W. J., 354(A5), 364 Eider, B., 318(S6), 381 Eklund, A. E., 1S(B13), 43 Ellenson, S. R., 17(H7), 4.8 Emerson, K., Jr., 6(V1), 64 Emery, E., 257(23), 307 Engbring, N. H., 403(A10), 409(E4), 414, 416 Engel, L. L., 156(V6), 616 Engel, M. G., 6(W13), 64 Engevix, L., 326(B16), 566 Englert, E., 317(E4), 370 Englert, E., Jr., 370 Engstrom, W. W., 409(E4), 416 Eliasch, H., 339(C9), 368 El Kabir, D. J., 396(E1), 398, 402(E2), 416 Elzinga, K. E., 406(E3), 416 Ensor, J. M., 400, 403(E6), 416 Epstein, E., 117(E3, Pll), 120(21), 13.9, 137, 139

434

AUTHOR INDEX

Epstein, N. N., 311 (D5, K6), 314(K6), 324(K6), 338(D5, K6), 352(K6), 370, 374 Epstein, R. M., 335(WlO, W12), 384, 385 Erici, I., 31(E5), 4G Eriksson, G., 161(Pl, P12), 174(B25), 203, 211 Eriksson, H., 73(E4), 99(E4), 132 Estes, F. L., 290(E3), 300 Etheridge, C. B., 330(E6), 356(E6), 370 Etkin, W., 392(E7), 417 Eton, B., 155(A3), 201 Ettre, L. S., 220, 300 Evans, E., 9(E6), 4G Evans, J. M., 330, 356(E6), 370 Evans, R. L., 330, 370 Evans, T. M., ll(T3), 53 Evenson, M. A., 240, 300 Eversen, T. C., 353(L3), 3Y5 Exley, D., 161(E8), 206, 237, 300 Ezrin, C., 409(21), 424 Ezrow, L., 324(530), 340(S30), 360(S30), 382

Field, J. B., 399, 400(A8), 403(F2), 414, 41 7 Fielding, J., 15(01), 51 Fineberg, J. C., 357(L10), 376 Finkbeiner, H., 264(K6), SO3 Finkelstein, J. W., 193(K1), 195(KlO), 208

Finkelstein, M., 82(F3), 99, 104(F2), 126 (F3), 132, 192(F1), 206 Finkelstein, N., 314(S30a), 382 Fischer, D. B., 406(H5), 418 Fishman, J., 160(F2), 206, 394(M9), 420 Fitzpatrick, H. F., 349(H1), 372 Flear, C. T. G., 14, 46 Fleck, A., 20(F3), 24(F3), 4G Fleming, L. W., 10(W3), 54 Fliegner, J. R. H., 168(J7), 208 Flood, C., 178(F3), 206 Florentin, A., 354(R24), 381 Florsheim, W. H., 409(F3), 417 Flotte, C. T., 340(510), 343(510), 381 Foerster, G. E., 27(H2), 28(H2), 48 Fontaine, Y. A., 394(F4), 417 Fontan, C. R., 279(P4), 281(P3), 296 (J2), 302, 305 Forchielli, E., 99(F2), 104(F2), 132 F Forest, M., 142(L17), 156(L17), 176 (B13), 202, ,909 Fahre, 21(W11), 64 Fairbanks, R. A,, 83(W7), 105(W7), 126 Forest, M. G., 172(R10), 173(R10), 212 Forster, F. M. C., 343(528), 358(528), (W7), 129(W7), 139 582 Fanska, R., 312(C5, G23), 321(C5, (223, Forsyth, C. C., 142(C6), 147(B18), 148 G24), 360(C5, G23), 368, 372 (B18), 151(B18), 155(B18), 166(C7), Farach, H., 409(21), 424 174(B17, B18, C5, CS), 175(C6), 177 Farese, R. V., 63(F1), 64(F1), 132 (C6), 178(B18, C6), 179(C6), 183 Farris, W., 32(G9), 47 (C7), 187(C6), 195(C7), 198(C5), Faucette, W., 148(C9), 204, 240(C5), 208, 204 296(C4), 299 Fauvert, R. E., 324(F1), 331(F1), 333 Fortunato, J., 180(K7), 208 Foster, A. E., 410(B17), 416 (Fl), S7f Foster, D. P., 18(F4, F5), 4G, 47 Fawcett, D. W., 320(F2), 371 Foster, G . V., 411(F5, F6), 413(F7), Fawcett, J. K., 13(E2), 46 Fedorck, S. O., 341(F3), 371 417, 419 Foulk, W. T., 322(H14), 323(H14), 327 Fee, D. A., 341(F3), 97f (H14), 331(515), 336(S15), 338 Felder, L., 356(F4), 371 (H14), 344(517), 345(517), 354 Feldthusen, V., 15(F1), 4 G (B61, S16, S17), 355(H14, S15), 360 Feleki, V., 28, 42 (H14, S15), 367, 373, 381 Fenimore, D. C., 221(P10, S6), 305, JOG Fox, I. J., 312(F5), 371 Ferguson, M. M., 156(C1), 204 Fox, J. E., 279(F2), 300 Fernichel, B. L., 17(P4), 51 Franc, Z., 390(56), 391(S6), 422 Fenier, R. J., 269(F1), 300

AUTHOR INDEX

435

Gamble, J. L., 9, 47 Francis, F. E., 175(F4), 206 Frandsen, V. A., 161(F6), 171(F5, F7), Gammeltofti, A., 339(B43), 353(B43), 206 367 Gandy, M., 86(Gla), 89(Gla), 102 Frankel, A. I., 132 (Gla), 112, 128, 133 Frankena, L., 191 (D4, D5), 192(D4), Ganong, W. F., 197(M20), 210 206 Frankland, M. V., 147(E2, S21), 154 Garbutt, J. T., 271, 297 (E2), 155(E2), 158(E2, Sn),167 Gard, D. A., 391(G10), 418 (S22), 170(E2), 171(E2), 188(521), Gardell, S., 324(B22), 325(B22), 366 Gardiner, W. L., 125, 133, 155(H10), 207, 200(S22), 205, 213 249(V3), 306 Franklin, M. J., 249(C7, CS), 299 Gardner, L. I., 77(G3), 133, 175(B6), Franklin, T. J., 323(B52), 367 202 Franklin, K. J., 4(T5), 64 Garner, L. M., 357(S11), 381 Frantz, A. G., 413(1"5), 424 Gattereau, D., 161(H6), 207 Frawley, J. P., 9(H14), 21(H14), 48 Gauthier, G. F., 411(H7), 413(H7), 413 Frawley, T. F., 241(P5), 306 Fredricks, M. G., 355(F6), 371 (H7), 418 Gautier, E., 197(U1), 214 Freedberg, A. S., 406(H4, H5), 418 Freedman, L. J., 4(S13a), 16(S13a), 63 Gedigk, P., 384 Gee, M., 259, 260(G1), 300 Freeman, T. E., 29O(L12), 303 Gehrke, C. W., 257, 258, 264, 265, 290 Freimuth, H. C., 296(C12), 299 (GZ), 301, 303, 306 Freinkel, N., 400(F8, S9), 409(D6, D7), Gelin, L-E., 15, 47 416, 417, 422 Gelineo, S., 26(G5), 31(G5), 47 French, A. B., 314(M6), 376 Gellis, S. S., 342(Y1), 343(Y1), 386 Freychet, P., 394(F9), 417 Gemzell, C., 103(G4), 133 Fricke, G., 296(K11), 303 Gemsell, C. A., 35(G6), 47, 164(Mll), Friedberg, S. L., 399(K4), 419 168(Mll), $10 Friedman, J., 413(F10), 417 George, J. M., 72, 133 Friend, C., ll(W22), 66 Gephardt, M. C., 331(L2), 376 Friend, D., 3(W5), 35(W5), 64 Giangradni, E., 336(P1), 379 Fukuchi, M., 403(K7), 4 9 Fukushima, D. K., 63, 71(K2), 75(F7), Giansiracusa, J. E., 324(C23), 325(C24), 355(C25), 369 82(F7), 133, 134, 178(Fg), 191(Fs), Gibree, N. B ., 96(R4), 97(R4), 120(R4), 206 137 Furst, A., 300 Gibson, G., 149(A7), 177(A7), $01, 206 Furth, E. D., 414 Giddings, J. L., 223(G3), 301 G Giesecke, W., 257(R5), 306 Giges, B., 358(G5), 371 Gabrilove, J. L., 78, 83(G1), 133 Gillette, R. W., 6, 17(G7), 47 Gabzuda, G. J., 342(M31), 378 Gilly, R., 142(B12), 147(B12), 149(B12), Gadsden, R. M., 279(M7), 304 156(B12), 178(B12), 180(B12), 186 Gaebler, 0. H., 358, 360, 361(G1), 371 (BlZ), 187(Bll), 198(B12), 202 Galante, L., 412(G6), 41'7 Gilman, G. A., 399, 417 Gale, M., 113(B35), 132 Gimbel, N. A., 32(G9), 47 Gall, E. A., 353(S12), 381 Gallagher, T. F., 71(K2), 75(F7), 78 Gindler, E. M., 328, 371 ( K l ) , 82(F7), 133, 134, 178(F8), 191 Giovannello, T. J., 34(M8), 60 Giroud, C. J. P., 147, 155(H9), 164, 166, (F9), 206, 350(G2), 351(G2), 371 167, 172(H9), 180(H9), 187(D16), Galli, A., 359(G3), 371 198(H9), 206, 207, 208 Gambino, S. R., 361(G4), 571

436

AUTHOR INDEX

Girpide, E., 177(V1), 914 Gitlow, S. E., 249(C7, C8, W6, W7), 809, 307 Gjessing, E. C., 16(C8, GlO), 44, 47 Godal, H. C., M G l l ) , 47 Goebelsmann, U., 71(L2), 136 Goetzee, A. E., 331(B12), 345(G7), 366, 371 Goetzee, A. F., 345(G7), 371 Golay, M. J. E., 223(G4), 301 Gold, J. J., 191(J2), 207 Gold, N. I., 71(G5), 94, 133 Goldbaum, L. R., 218, 219, 278(G5), 301 Goldberg, A., 70(G6), 133 Goldberg, S., 82(F3), 99(F3), 126(F3), 132 Goldenberg, I. S., 47 Goldfarb, S., 350(G8), 371 Goldfien, A., 115(G7), 133 Goldhaber, P., 413(A5), 414 Goldman, A. S., 156(G2, G3), 159(B31), 189(B31), 191(B31), 192(B31), 193 (B31), 195(B31, G3), 203, 206 Goldstein, J., 323(G11), 359(G10), 3'71 Gona, A. G., 392(E7), 417 Gonzales-Oddone, M. V., 325(G12), 371 Good, C. J., 314(M6), 376 Goodall, M., 35(G13), 47 Goodman, R. D., 328(G13, G14), 331 (G13), 334(G14), 353(G14), 371 Goodrich, J. E., 113(M2), 136 Gordan, G. S., 351(D3), 370 Gordon, A. R.,404, 417 Gordon, A. S., 407(S8), 423 Gordon, E., 358(J1), 360(J1), 374 Gordon, M. L., 39, 47 Goresky, C. A., 319(G15), 320(G15), 326 (G16), 337(G15), 341(G15), 371 Gould, R. P., 412(N4), 480 Goulis, G., 347(B21a), 366 Govaerts, P., 18(G15), 47 Govrick, K., 3(21), 34(21), 66 Graber, J. W., 139 Grahcr, M., 314(S30a), 382 Grace, R. A,, 339(B5), 366 Graff. J., 257(G6), 301 Grafflin, A. L., 318(G17), 372 Graham, E. A., 312, 372 Gram, C., 331(W25), 332(W25, W%), 386

Gram, H. C., 18(G16), 47 Gransitsas, A. N., 36(G17, GM), 47 Grant, J. K., 6O(G8), 70(G6), 100, 108 (GS), 114(G8), 116(G8), 122(G8), 125(G8), 128(G8), I33 Grant, R. T., 5, 14(G19), 15(G19), 47 Gray, C. H., 6O(G9), 83, 85(G10), 96, 98, 99, 105(G10), 147(W5), 133, 216 Gray, J. H., 17(B10, B l l ) , 43 Gray, S. J., 13(G20), 48 Green, C. H., 317(G20), 327(02), 354 (GZO), 372, 378 Green, D. M., 321(C34), 326(C34), 369 Green, H. N., 6, 7, 48 Greenbaurn, A. L., 12, 63 Greenberg, J., 341(L7), 376 Greenblatt, R. B., 67(Gll, Mla), 70 (Mla), 78(Gll), 79, 80, 133, 136 Greenlaw, R. H., 314(S43), 383 Greenwood, F. C., 35(G22), .bg Greer, M., 267, 268,301 Greer, M. A., 39(M9), 60, 390, 417 Gregersen, M. I., 340(G22), 378 Greig, M., 155, 158, 206 Greipel, M., 147(S21), 188(521), 813 Gridgeman, N. T., 27, 48 Griffiths, K., 156(B1), 201 Grigaut, A., 5, 46 Grodsky, G. M., 312(C5, C6, G23), 321 (C5, G23, G24), 324(K22), 355(C6), 360(C5, C6, G23), 363, 368, 3'72,376, 379 Groen, D., 89(V2), 110, 112(V2), 127 (V2), 138 Gross, C. W., 324(D6), 370 Gross, J., 404(G2), 417 Grove, G. R., 328(26), 340(26), 386 Grover, P. L., 337(G25), 338(B38), 345 (B38), 348(B38), 349(B38), 351 (B38), 366, 372 Grundy, H. M., 133 Gual, C., 390(G4, G5), 392(B16), 393 (G4), 416, 417 Guarnieri, M., 240(B2), 245, 246, 297, 301 Gubar, A. V., 349(G26), 572 Gudmundsson, T. V., 412(G6), 417 Guerin, T., 346(V3), 350(V3), 384 Guerra, S. L., 9(H14), 21(H14), @, 112 (P6a), 113(P6a), 137, 186(€'9), 211

AUTHOR INDEX

Guerrant, J. L., 344(H13), 373 Guerrero, I., 350(R8), 380 Guillemin, R,., 390(G7-G9), 391(G7), 391(G10), 392(V2), 417, 418, 42.9, 424 Guiochon, G., 223, 307 Guislain, R., 344(V4), 384 Gunville, R., 151(U3), 155(U3, U4), 177 (U2, U3), 179(U3), 187(U3), 198 (C15), 204, 214 Gurochen, G., 222(L2), 303 Gurpide, E., 69, 71(G13), 74, 133, 136, 139, 160(S7), 177(V1), 2U7, 213, 21.4 Gustafsson, B. E., 73(G14), 99(G14), 133 Gustafsson, J d . , 73(E4, G14, G15), 99 (E4, G14, G15), 1S2, 1S3, 143(G7), 807 Gut, M., 65(R2), 86(R1, R2), 89(R1, R2), 94(Rl, R2), 97(R1, R2), 115 (Rl, R2), 128(R1, R2), 137 Guzik, H., 160(F2), 206 Gverrin, G., 348(B26), 366

H Haahti, E. 0. A,, 253(L13), 30.4, 306 Haas, H. G., 413(H1), 418 Habif, D. V., 349(H1), 372 Hafstrom, L. O., 326(B16), 366 Hagen, A. A,, 160(Z5, Z6), 185(H1), 186 (Z6), 207, $16 Hagen, P., 257(H1), 301 Haggard, H. W., 7(H8), 48 Haist, R. E., 29, 30, 48 Halasz, I., 301 Halban, J., 186(H3), g07 Hale, H. B., 27, 28, 48 Halkerston, I. D. K., 148(S29, S30), 214 Hall, P. E., 70(G6), 133 Hall, R., 400(H2), 401(H3), 418 Halloway, M. T., 405(P8), 481 Ham, T. H., 18(H3), 48 Hamada, S., 404(516), 423 Hamilton, A. S., 13, .&$ Hamilton, D. W., 412(N4), &?O Hamilton, H., 394(W1), 424 Hamilton, P. P., 6(V1), 64 Hammaker, L., 338(H2), 346(H2), 372 Hammel, H. T., 43 Hammel, R. P., l6(D4), 46

437

Hammond, K. B., 223(H3), 246, 247, $01, so3 Hamolsky, M. W., 406(H4, H5), 418 Hanes, B. W., 35(G13), 47 Hanger, F. M., 351(W3), 384 Ifanin, I., 296(J6), 302 Ham, C. S., 260, 261, 262(C11), 263, $99 Hansen, L. B., 264, 266, 301 Hanson, J., 198, 808 Hanzon, V., 318(H3), 351(H3), 364, 372 Harbert, G. M., 155, 207 Harbourn, C. L. A., 223(D8), 300 Hardaway, R. M., 18(H4), 48 Hardy, J. D., 6(E1), 46 Hargreaves, T., 310(H4), 345(H4), 346 (H5), 347(H4, H5), 349(H4, H5), 351(H5), 372 Harkness, R. A., 104(11), 134, 147(S13), 156(H5), 161(H5), 185(S13), 2U7, 213, 294, 295, 296, 297, 301 Harper, H. A., 353(S27), 382 Harris, G. W., 390, 418 Harrison, D. D., 336(K1), 343(K1), 374 Hartcroft, P. M., 34(D3), 46 Hasan, J., 408(08), 421 Hauer, H., 405(011), 406(011), 410 (010, 421 Haugli, F. B., 414 Hauss, W. H., 12(H5), 48 Hawes, R. C., 102(N3), 104(N3), 117 (N3), 123(N3), 136 Hawley, G., 376 Hawley, W. D., 392(B15, B16), 416 Hawson, M., 354(23), 386 Hayes, M. A., 9(H6), 31(G12), 47, 48 Hayles, A. B., 192(M5), 209 Hayward, J. S., 412(N4), 420 Health, D. F., 7(S17), 8(S17), 63 Heaney, R. P., 331(H6), 350(H6), 372 Heap, R. B., 108, 134 Hecht-Lucan, G., 16(M5), 60 Hecker, R., 364 Hedgeley, E. G., 269(H6), 301 Heemstra, H., 321(D1), 336(D1), 360 (DU Hegedus, A., 404(S16, S18), 407(S18), 410(S18), 423 Heideman, M. L., Jr., 394(L4), 396(B2), 397(B2), 414, 419 Heikel, T., 346(H7), 372

438

AUTHOR INDEX

Heilskov, N. S. C., 20(54), 62 Heim, W. G., 17(H7), 48 Heinman, H., 339(H7a), 372 Heinrichs, W. L., 147(C12, C13), 155 (C12, C13), 158(C12, C13), 170 ((312, C13), 171, 185(C12), 204 Heirwegh, K., 165(D7), 206 Heiskell, C. A,, 419 Heller, C. G., 20(Hll), 48 Hellig, H., 161, 207 Hellman, L., 71(K2), 75(F7), 82(F7), 133, 134, 178(F8), 191(F9), 206 Helm, J. D., 327(H8), 372 Hembree, W. C., 65(B2), 69(B2), 70 (B2), 78(B2), 79(B2), 104(B2), 130 Hems, B. A., 185(H7), 207 Hendeles, S. M., 241, 297 Henderson, H. H., 35(SlOa), 63 Henderson, Y., 7, 48 Henion, W. F., 399(P6), 421 Henning, N., 339(H9), 372 Henry, R., 148(H8), 207 Henry, R. J., 358(Hl0), 359, 373 Henschel, A., 20(K1), 49 Henze, K. G., 411(ClO), 413(ClO), 416 Heremans, J. F., 165(D7), 206 Heroux, O., 27, 31, 48 Herrera, M. G., 9(R7), 62 Herring, B., 399(Fl), 417 Herlitz, C. W., 318(H12), 342(H11), 373 Hermann, W., 71 (B12), 130, 132 Hershey, S. G., 19(A4), 48 Hertz, R., 401(04), @ l Hervonen, A., 159(N5), 210 Heseltine, D. W., 312(F5), 371 Heys, R. F., 115(Hla), 121(Hla, Ol), 123(Hla), 134, 136 Hicks, M. M., 344(H13), 373 Higgins, F. E., 322(H14), 323(H14), 327 (H14), 338(H14), 355(H14), 360 (H14), 373 Higgins, J. T., 34(D3), 46 Hill, C., 329(04), 379 Hill, E., 329(24), 352(25), 353(25), 353 (251, 38G Hill, J. B., 400(L6), 401(L7), 419 Hillman, D. A., 155(H9), 164, 172(H9), 180(H9), 198(H9), 207 Hillman, J., 148(S30), 214

Hirsch, P. F., 411(H7, HS), 412(C8), 413(H7, HS), 416, 41s Hirsch, R. L., 355(H15), ST3 Hirschfield, J. W., 20(H11), 48 Hirschrnann, K., 69(W4), 8O(W4), 139 Hirt, A., 316(H16), 373 Hjelte, V., 312(C6), 343(B56), 355(C6), 357(B56), 360(c6), 367, 368 Hober, R., 316(H17-H19), 317(H19), 373

Hoch, J., 324(H20), 361(H20), 37s Hoch-Ligetti, E., 17(H12), 48 Hockey, J. A., 228, 301 Hodgson, P. E., 340(S10), 343(S10), 381 Hoek, W., 242, 243, 301, 307 Hoffbauer, F. W., 352(M21), 354(M81), 355(F6), 371, 377 Hoffman, H. N., 312(H27), 318(H27), 325(H28), 331(H27), 338(H28), 343 (H28), 349(H28), 351(H28), 352 (H27), 360(H27, H28), 373, 374 Hoffman, N. E., 296(H9), 301 Hofmann, A. F., 113(H2), l S 4 Hollander, C. S., 410, 418 Holliday, M. E., 73(W1), 139 Hollis, D. L., 222(H10, H l l ) , 223, 301 Holloway, M. T., 408(S13), 410(S13), 423

Holloway, R. J., 344(B47), 367 Holman, G. H., 193(K1), 195(K10), 208 Holmes, W. L., 231, 301 Holmstedt, B., 249(H15), 251(H15), 253 (H15), 254(H15), 302 Holmstrom, E. G., 170(M12), 210 Holt, H. P., 344(H13), 373 Holt, S. J., 412(N4), 420 Holton, J. B., 162(B14), 202 Holzbach, R. T., 343(H21), 351(H21), 313 Holzbauer, M.. 48, 69(H3), 134 Horn, D. B., 407, 415 Homing, E. C., 60(E2), 86(E2), 87, 88 (E2, H4), 90(E2), 97(E2), 98(E2), 99(H4), 104(E2), 108(E2), llO(D1, E2, H4), 114(D1, E2), 116(E2, H4), 117(E2), 125, 132, 133, 134, 155, 207, 225, 231, 235, 237, 238, 242(D1), 249, 250, 251, 252, 253(L13), 254, 296 (DI), 298, 299, 300, 301, 302, 304, 306

AUTHOR INDEX

Horning, M. G., 110(D1), 114(D1), 132, 238(H14), 242(D1), 249W15, H17), 250(H18), 251(H15), 253(H15), 254 (H15), 296(D1), 299, 301, 302 Horton, R., 63(K3), 65(H6), 74(T4), 75 (T4), 78(H5), 79, 80(86, T4), 102 (H6, K3), 104(K3), 127, 134, 138 Horvath, C., 901 Horvath, S. M., 326(H22), 373 Hottinger, G. C., 331(R4), 379 Houchin, D. M., 18(H14), 48 Howard, G., 339(L9), 376 Howard, J. E., 21(H15, H16), 48 Howard, J. M., 9, 21(H14), 48 Howard, M. M., 330(L8), 352(L8), 375 Hsia, D. Y.-Y.,180(D17), 205 Hsu, J. M., 348(H23), 373 Huber, T. E., 353(H24), 373 Huggins, C., 71(H7), 134 Huggins, G. B., 317(M25), 377 Hultman, E., 327(C10), 329(ClO), 331 (H25, H26), 339(C9), 349(H25), 368, 373

Hume, D. M., 34(H17), 49 Humelbaugh, C., 17(W9), 18(W9), 37 (Wg), 64 Hunter, F. M., 359(P7), 360(P7), 379 Hunter, I. R., 256, 290(H20), 302 Hunter, J., 2, 49 Hunton, D. B., 312(H27), 318(H27), 325 (H28), 331(H27), 338(H28), 343 (H'B), 349(H28), 351(H28), 352 (H27), 360(H27, H28), 373, 374 Hurst, G., 280(C15), 299 Hurvitz, S. H., 311(R25), 381 Hutt, B. K., 326(H22), 373 Hykes, P., 324(J7), 341(57), 374

I Ikekawa, N., 238(H14), 301 Ikonen, M., 159(N5), 910 Ilka, S. J., 354(S42), 383 Inada, M., 404(S16), 410(11), 418, 4.23 Ingall, S. C., 161(E8), 906' Ingbar, S. H., 404(14, 161, 405(13, 16, PS, W3), 406(13, W3), 408(S13), 409 (D6, D7, 151, 410(B17, 16, S13), 415, 416, 418, @I7 493, 434 Ingelfinger, F. J., 325(M16), 327(Il), 329, 331(11), 333, 337(11), 339(B42),

439

340(B42), 344(C37), 351(M16), 352 (Il), 353(B42), 356(11), 359(11), 367, 369, 370, 374, 376

Inkley, S. R., 328(12), 329(12), 340(12), 374

Iob, V., 340(S10), 343(S10), 381 Irvine, K., 17(H12), 48 Ismail, A. A. A., 104(11), I34 Israel, H. L., 329(13), S74 Ittrich, G., 99, 121, 134, 155(11), 207 Ivy, A. O., 353(D10), 370 Iwamiya, M., 162(12), 207 Iwatsubo, H., 403(K7), 419

J Jaakonmaki, P. I., 238(H14), 267, 301, 302, 411(S1), 423 Jablonski, P., 358(J1), 360(J1), 374 Jackson, D. E., lO(SlO), 5s Jackson, D. R., 18(H4), 48 Jacquenoud, P., 349(M18), 377 Jaff6, M., 322(J2), 374 Jaffe, R., 161(P12), 211 Jaffe, R. B., 16l(J1), 207 Jaffe, S. H., 121(J1), 134 Jailer, J. W., 191(J2), 207 Jain, N. C., 296(J2), 302 James, S. P., 322(B13, B53), 323(B52), 365, 367

James, A. T., 229, 290, 302 James, V. H. T., 83(G10), 85(G10), 91, 95, 96(G10), 98(G10), 99, 104(B17), 105(G10), 114, 119, 120, 131, 133, 138, 165, 172(J3), %07 Janak, J., 280, 302 Janoski, A. H., 185(54), 207 Jansen, A. P., 413(B9), 41.4 Jawinen, P. A., 350(E3), 370 Javitt, N. B., 321(J5), 323(J4), 323(J3, J4), 338(J3), 374 Jay, S. M., 400(S9), 422 Jayle, M. F., 148(J5), 166(P4), 183(P4, P5), 207, 211 Jeanmaire, J., 359(G3), 371 Jeffcoate, T. N. A., 168(J6, J7), 208 Jelliffe, R. W., 296(J5), 302 Jenden, D. J., 296(J6), 302 Jenkins, E. B., 18(H4), 48 Jenner, J. A,, 346(J6), 374 Jmsen, E. V., 71(H7), 134

440

AUTHOR INDEX

Jewelewicz, R., 132 JirLsek, J. E., 159(J8), 208 Jirju, H., 62(J3), 134 Jirku, H., 73(J2), l S 4 Jirsa, M., 324(J7, Rl), 341(J7), 374, 379 Johanson, C. A., 156(R9), 212 Johannson, E. D. B., 65(J4), 97(J4, Nl), 100(J4), 102(J4), 104(J4), 110, 113 (J4), 123(J4, NZ), 124, 134, 136 John, J. P., 294(B15), 298 Johns, M. W., 37, 49 Johns, T., 279(C1), 299 Johnson, D. E., 257(J7), SO2 Johnson, D. G., 18(H14), 48 Johnson, F. A,, 354(Dll), 370 Johnson, H. D., 28(K4, K5), 49 Johnson, I. A. D., 36(R9), 52 Johnson, M. K., 323(J8), 374 Johnson, P., 399(Fl), 417 Johnston, C. C., Jr., 413(J1), 418 Johnston, C. E., 6(G8), 47 Johnston, I. D. A., 3(J2), 9, 31(J3), 32 (JZ), 34(J2, J4), 35(J2), 39, 49 Johnston, V. D., 223(P9), 306 Jojola, R., 304 Jones, A. E., 402(M13), 403(M13), PO Jones, C. M., 354(C38), 370 Jones, H. G., 269(J8), 302 Jones, J. E., 409(J2), 418 Jones, J. K. N., 269(58), 802 Jones, R. K., 23(R5), 62 Joplin, G. F., 413(F7), 417 Jordan, A., 177(Lll), 209 Jordan, G., 299 Jose, A . D., 296(M8), 304 Josnph, E. P., 62(B9), 130 Jost, A., 159(J9), 208 Jouan, P., 35, 39(J6), 49 Julian, L. M., 331(B45), 341(B45), 367 Julsrud, A. C., 252(B37), 366 Jungmann, R. A., 125(J5), 134 Jutisz, M., 390(G9), 391(G10), dl8 Juvet, R. S., 218, 300, 302 Juvet, R. S., Jr., ZZl(Jl1, Zl), 30.2, 307

K Kade, H., 312(T3), 341(T3), 383 Kahnt, F. W., 411(N2), 4SO Kaiser, K., 296(V9), SOT Knisrr, I?., 221(K1), 302

Kamber, B., 411(R7), 422 Kampine, J. P., 330(537), 382 Kaplan, A., 296(51), 305, 366 Kappas, A., 71(K2), 78(K1), 134, 350 (G2,M22), 351(G2), 360, 371, 378 Karch, H. L., 6(55), 52 Karg, S., 355(C25), 369 Karmen, A., 11(L1), 45 Karoum, F., 296(K2), 302 Karrer, A., 112(P6a), 113(P6a), 137 Kater, R. M. H., 336(K1), 343(K1, K2), 354(K2), 374 Kato, T,, 63(K3), 102(K3), 104(K3), 127, 134 Katz, W. J., 342(K3), 374 Katznn, R., 400(P2, P3), 431 Kauhtio, J., 342(K4), 374 Kayser, L., 13(B15), 43 Kazyak, L., 278(K3), 302 Kearns, T. P., 403(W5), 424 Keegan, P. G., 296(K2), 303 Keeton, R., 331(L2), 376 Keller, A. R., 170(M12), 175(M13), 210 Keller, D. M., 346(K5, L16), 374, 376 Keller, K., 296(C17), 299 Keller, N., 63(K4), 82(K4), 134 Keller, W. H., 347(S36), 383 Kellie, A. E., 73(W1), 136, 139 Kelly, R. W., 147(513), 185(S13), d l 3 Kelly, V. C., lSO(ClS), 187(C18), 204 Kelly, W. G., 185(J4), 207 Kendall, J . W., 39(M9), 60 Kennedy, T. H., 395(A4), 397(A4), 414 Kenny, A. D., 412(G6), 417, 419 Kenny, F., 156(K1), 180(Kl), 208 Kent, N. R., 280(C15), 299 Kerkay, J., 28, 43 Kerr, W. J., 311(D5, K6), 314(K6), 324, 338(D5, K6), 352(K6), 370, 374 Ketterer, S. G., 331(W18), 334(W18), 339(W18), 340(W18), 386 Keutmann, H. T., 411(P7), 421 Keys, A., 20, 49 Kibler, H. H., 28(K5), 49 Kier, L. B., 29O(L12), 303 Kimball, A. P., 257(22), SOT Kimball, H. L., 100(B25a), 117(B25a), 131

King, A . G . , 317(M25), 377 King, E. R., 328(B44), 340(B44), 367

441

AUTHOR INDEX

Kingsley, G. R., 328(G14), 334(G14), 353(G14), 571 Kinney, J. M., 9(R7), 21(K2), 22, 49,62 Kinnear, A. A., 356(K7, KS), 374 Kinsella, R. A., Jr., 175(F4), ,906 Kirk, P. L., 279(P4), 280, 281(P2, Pa), 292, 293, 296(J2), 302, 304, 306 Kirkham, K. E., 395(K3), 396(K3), 398, 419

Kirkland, J. J., 223(K4, K5), 302 Kirschner, M., 350(S8), 351(S8), 381 Kirschner, M. A., 104(Lll), 125, 127, 136 Kirschvink, J. F., 180(C18), 187(C18), 204

Kitchin, J. D., 155(C16), 161(C16), 163, 170(C16), 904, 208 Kittinger, A,, 321(M29, M30), 336(M30), 360(M30), 377, 378 Kittinger, G. W., 118(K7), 135 Kivel, R. M., 353(C17), 368 Kjellberg, S. R., 168(K3), 208 Klaasen, C. D., 338(K9), 344(K9, R131, 574, 380

Klainer, L. M., 399, 419 Klatskin, G., 353(K10), 374 Klebe, J. F., 264, SO3 Klein, G. P., 166, 167, 208 Klcin, O., 99(F4), 132 Klein, R., 180(1<7), 198, 208 Klein, R. I., 318(Kll), 576 Kleiner, G. J., 351(K12), 375 Klesper, E., 218, SO5 Kliman, B., 237, 241(K9), 303 Klopper, A., 124, 136, 155, 16l(K8), 208 Klyne, W., 60, 136 Kmentovh, V., 390(56), 391(S6), 4.22 Knapp, D. W., 326(H22), 373 Kneubuhler, H. A., 403(A10), 414 Knigge, K. M., 28(K3), 49 Knight, B. C., 346(H7), 372 Knights, B. A., 86(K10), 114(K10), 116 (KlO), 128, 155 Knobil, E., 97(N1), 123(N2), 136 Knoblock, E. C., 278(K3), 302 Knox, K. I,., llO(Dl), 114(D1), 132, 242(D1), 249(H15), 251(H15), 253 (H15), 254(H15), 296(D1), 299, 301 Kobayashi, T., 172(M18), 173(M18), ,910 KoFi, J., 390(S6), 391(S6), 422 Kofler, M., 296(K10), 303

Kogsaneler, V., 73(J2), 13.4 Kolb, A. M., 356(S38), 582 Kookootsedes, G. J., 6(G8), 47 Kopp, I., 355(K13), 375 Korelitz, B. I., 347(W4), 384 Korenman, S. G., 75(L12), 122, 155 Kory, R. C., 350(K14), 351(K14, W2), 375, 384

Korytnyk, W., 296(K11), 303 Korzun, B. P., 296(K2), 303 Koslowski, L., 23, 49 Kotby, S., 28(K4, K5), 49 Kovach, A. G. B., 25, 49 Kowarslri, A,, 193(K1), 195(K10), 196 (c2), 204, 908 Krahenbiihl, C., 176(M8), 209 Kramer, P., 325(M16), 327(11), 329(Il), 331(11), 333(11), 337(11), 351(M16), 352(Il), 356(11), 359(11), 374, 376 Krass, M. E., 391, 419 Kraybill, M., 137 Krebs, J. S., 312(B46), 318(K17), 321 (K15, K18), 322(B46, K8, K9), 323 (B46, K19), 324(B49), 325(K20, K21), 326(K16), 331(l345), 341(B45, B50, B51), 360(K18), 367, 376 Kresch, L., 351(K12), 576 Kriegw, H., 328(12), 329(12), 340(12), 374

Kriss, J. P., 402(K6), 403(F2), 417, 419, 423

Kritzler, R. A,, 351(W3), 384 Kroes, A. A,, 334(V12), 342(Vl1), 384 Kruger, S., 340(B3a), 365 Kubin, R. H., 324(K22), 376 Kuhn, H. W., 336(P8), 379 Kuksis, A,, 237, 303 Kulkarni, B. D., 294(B15), 298 Kumahara, Y ., 403(K7), 419 Kumar, M. A,, 411(F6), 412(S22), 417, 419, 423

Kunkel, H. G., 354(A5), 364 Kusumoto, H., 4.23 Kydd, D. M., 407(S8), 492

1 La Bella, F. S., 391(K5), 419 La Due, J. S., ll(L1, W22), 12(W21), 49, 66

442

AUTHOR INDEX

Laidlaw, J . C., 187(R13), 212 Laland, S. G., 414 Lamb, S. I., 296(56), 302 Lamberg, B.-A., 407(L5), 419 Lambert, M., 168(L1), 208 Lambert, R., 347(B55), 367 Lamkin, W. M., 257, 258, 303 La Motta, R. V., 343(W17), 385 Lanchantin, G. F., 16(L2), 17(L2), 49 Landatil, S.F., 16(D4), 46 Landault, C., 222(L2), 303 Landon, J., 35(G22), 48 Langendorff, H., 23(L2a), 49 Langer, S. H., 231(B16), 233(B16), 298 Lanman, J . T., 158(L2), 167(L2), 208 Lapin, I. P., 72(L1), 135 Larsen, J. A,, 339(W27), 340(W27), 385 Larsen, N. A., 15(F1), 46 Larson, F., 404(L1), 419 Larsson-Cohn, U., 328(L1), 350(L1), 375 Lasnitzky, I., 71(B4), 130 Last, J . H., 340(B3a), 365 Lathe, G. H., 162(B14), 186, 202, 208, 310(H4), 345(H4), 346(H5), 347 (H4, H5), 349(H4, H l ) , 351(H5), 572

Lavers, G. D., 331(L2), 375 Lawler, C. A., 321(C34), 326(C34), 369 Laws, J . F., 353(L3), 375 Lauritzen, C., 162(L4), 175(L6), 180(L6), 193(L7), 198, 199(L5, L6, L7), 200, 208, 209

(M17), 350(L6), 351(M17), 352G5, LS), 355(C16), 357(L10, M17), 360 (C16, L5), 568, 3?5, 376, 377 Lefebvre, Y., 16l(H6), 207 Legallais, V., 137 Legate, C . E., 292(L4), 303 Lehmann, W.-D., 175(L6), 180(L6), 199 (L6), 208 Lehnhardt, W. F., 273(L5), 275, 276, 303 Lemarchand-Beraud, T., 394, 396(L2), 419

Leon, Y. A., 148(L10), 156(L9), 209 Leong, G. F., 325(K20, N3), 344(B47), 367, 375, 378

Lepp, A., 394(L3), 419 Leppelmann, H . J., 12(H5), 48 Le Quesne, L. P., 11(L3), 13(L4), 50 Lesko, W., 339(L9), 316 Lester, R., 321(LlOa), 376 Lestradet, H., 197(R18), 212 Levell, M. J., 177(Lll), 209 Levenson, S. M., 21(L5), 50 Leventhal, M. L., 138 Levin, B., 197(R19), 212 Levinsohn, S. A., 318(Kll), 375 Levenson, S. M., 6(T1), lO(Ul), 20 ( T I ) , 55, 54 Levitz, M., 71(L2), 73(J2), 121(J1), 134, 155, 169(L12), 188(L12), 209 Levy, R. L., 294, 303 I A ~ vR. J ~P., , 69(W4), 80(W4), 139, 394 (L4), 419 Lewis, A. E., 333(L11, L12), 376 Lewis, B., 115(L3), 155 Lewis, G. P., l l ( L 6 ) , 12, 50 Lewis, V. J., 402(M13), 403(M13), 420 Lewmarie, P., 67(L4), 155 Leyssac, P., 164(L13, L14), 209 Libby, D. A., 296(L7, PS), 303, 305 Liberti, A., 257(L8), 303 Liebe, D., 404(D2a) Lieberman, A,, 353(C17), 368 Lieberman, A. H., 348(L13), 349(L13),

Lawrence, B., 175(M13), 210 Lawrence, N. L., 396(B2), 397(B2), 414 Layne, D. S., 62(J3), 134, 178(F3), 206 Leach, H., 223(H3), 246, 247, 301, 303 Learned, €I., 148(C9), 204, 240(C5), 299 Leavell, B. S., 344(H13), 373 Leboeuf, G., 187(D18), 205 Lee, G. Y., 404(16), 405(16), 410(16), 418 Lee, J. B., 413(T5), 424 Lee, K. L., 392(B12, B13), 415 Lee, M. R., 13(M2), 50, 412(D1), 416 376 Lee, S. W., 296(A11), 297 Leevy, C. M., 312(C16), 317(C16), 318 Lieberman, S., 69, 71(G13, R5), 74(G12), 153, 135, 137, 139, 148(B40), 160(57), (M17), 324(L5), 328(C16), 329 166(D13), 177(Rll, Vl), 183(D13), (M17), 330(L4, L5, L8), 33l(Cl6, 185(R12), 188(L15), 192(L16), 203, L4, M17), 337(C16), 338(c16, L4, 205, 207, 209, 212, 213, 214 L5, M17), 339(L4, L9), 340(C16, L4, L5), 341(L7), 342(L4, L5), 344 Liewendahl, K., 407(L5), 419

443

AUTHOR INDEX

Liljedahl, S.-O., 13(B15), 17(B5), 20 (m), 21(B5), 3O(B5), 42, 43, 55 Lillehei, R. C., 4(L8), 50 Lima, F. W., 324(L14), 376 Lindner, H. R., 69(L6), 135 Lindsay, R. H., 158, 4M)(L6), 401(L7), 419

Lindstrand, K., 23(L9), 50 Lingren, W. E., 115(L7), 135 Linturi, M., 279(B17), 298 Liosky, S. T., 4(S13a), 16(S13a), 53 I,ipp, C., 162(N1), 910 Lipsett, M. B., 60(L8), 65(B2), 69(B2), 70(B2), 75, 76(L10), 78(B2), 79(B2), 83(L8), 97(L9), 100, 102(SS), 104 (B2, L8,L l l ) , 108(L8), llO(LS), 116 (LS), 122, 123(L8), 124(L9), 125, 128 (LS), 130, 135, 138, 401(04), 481 Lisboa, B. P., 116, 135, 136, 160(Z5, Z6), 186(Z6), 216 Litle, R. L., 293, 306 Litonjua, A. D., 406(C4), 415 Little, B., 65(R2), 86(Rl, RZ), 89(R1, RZ), 94(R1, RZ), 97(R1, RZ), 115 (RI, RZ), 128(R1, RZ), 137 Littlewood, A. B., 303 Livine, R., 326(C26), 369 Livingstone, J. R. B., 147(S11), 148 (Sll), 154(Sll), 159(S12), 161(S12), 169(S12), 170(S12), 183(511), 188 (Sll, S12), 213 Lloyd, C. W., 172(M18), 173(M18) Lobotsky, J., 172(M18), 173(M18), 210 Logans, E. R., 16(D4), 46 Long, C. N. H., 6(R12, W13), 62, 64 Longerbeam, J. K., 4(L8), 60 Longueville, J., 330(L8), 352(L8), 375 Loras, B., 142 ( B E , L17), 147(B12), 149 (BE?), 156(B12, L17), 176(B13), 178 (BlZ), 180(B12), 186(B12), 187 ( B l l ) , 198(B12), 202, 209 Lorber, S. H., 327(L15), 345(L15), 376 Loring, J. M., 162, 214 Lotspeich, W. D., 346(K5, L16), 374, 376

Love, D. N., 156(H5), 161(H5), g07 Lovelock, J. E., 116, 136, 22l(LlO, L11, S6). 303, 305, 306 Lowe, C. U., 197(U1), 214 Lowe, R. C., 351(D3), 370

Lowy, J., 211 Lucido, J., 2l(L9a), 50 Luden, R. C., 296(K2), $03 Luke, H. H., 29O(L12), 305 Lundwall, F., 155(B38), 164(B37, B43), 180(B43), 198(B43), 203 Lunel, J., 316(T7), 334(T7, TS), 355

(T7), 383 Lunenfeld, B., 104(59), 125(S9), 138 Lusk, G., 24(L10), 50 Lutwack, L., 31(G12), 47 Luukkainen, T., 73(A3), 122(A4), 130, 253, 304, 350(E3), Lyons, H., 279(L14), $04

M Maachia, V., 393(P4), 400(M1, P3, P4), 419, 421 McCallum, J., 70(G6), 133 McCarty, C. S., 356(C35), 369 McCathie, M., 16(MI), 17(M1), 18 (MI), 50 McComas, D. B., 115(G7), 133 McConahey, W. M., 403(W5), 424 McCord, W. M., 279(M7), 304 McCorkle, H. J., 353(S27), 382 McCowan, M. A. A., 10(W4), 54 McCredie, R. M., 296(M8), 304 McCullach, E. P., 403(M10), 420 McCutcheon, A., ZO(C35, C36), 46 McDermott, W. V., 354(C38), 370 MacDonald, A. M., 196(B33), 203 MacDonald, D., 327(M1), 329(M1, M2), 376

MacDonald, P. C., 69(V3), 139, 174 (SlS), 177(V1), 213, 214 McGaughey, H. S., 155(H4), 207 McGill, D. B., 344(S17), 345(S17), 354 S16, S17), 381 McGirr, J. L., 9(C37), 10(C37), 19 (C37), 24(C37), 46 MacGregor, A. R., 9(W19), Io(W19), 55

Machata, G., 279(M1), 280(M2), so4 Machella, T. E., 327(H8), 355(M3), 576, 376

McIntosh, A. D., 142(M6), 209 MacIntyre, I., 411(F6, MZ), 413(F7, M2, M3), 417, 419

444

AUTHOR INDEX

Mack, E., 39(E4), 46 McKay, D. G., 355(H15), 373 Mackay, I. R., 353(M4), 376 McKenzie, J. K., 13(M2), 60 McKenzie, J. M., 391(M11, S15), 393, 394(M5, M9), 396(M4), 397, 401, 402(M6, M7, M l l ) , 403(M7, M8, MlO), PO,423 McK. Hart, D., 156(C1), 604 McKinney, S. E., 337(M5), 345(M5), 376

McKeown, G. G., SO4 McLeod, G. M., 314(M6), 376 Macleod, S. C., 114(B21a), 131 McMarks, P. A., 339(H7a), 372 McMartin, C., 225, 231, 233, 278, 281 ( M l l ) , 304, 306 McMath, M., 340(S10), 343(SlO), 381 MacMillan, J. M., 329(M13), 377 McMillion, C. R., 324(M7), 376 McNalty, A. A., 324, 376 MacNaughtan, C., 99(B21), 104(B21), 108(B21), 114(B21a), 121(B21), f31 MacNaughton, M. A., 16l(K8), 608 McNeil, H. L., 311(M9), 376 McNeill, I. F., 36(M3), 60 McPherson, A. I. S., 16(M1), 17(M1), 18(M1), 50 MacPherson, A. I. S., 339(B43), 353 (B43), 367 McQuire, W. L., 394(L4), 419 McVicar, C. S., 317(G20), 354(G20), 372 Madden, S. C., 19(M4), 21(M4), 60 Maegraith, B. G., 318(A13), 364 Magalini, S. I., 16(M5), 50 Magath, T. G., 338(S31), 356(S31), 386 Magendantz, H . G., 147(M1), 209 Maggiore, Q. S., 345(B21), 347(B21a), 365, 366 Magliocca, R., 161(N2), 210 Magnusson, A.-M., 155(D8, DlO), 168 (DS), 205 Magrini, U., 156(C8), 159(C8), SO4 Mahadevan, V.. 291, 292(M3), 304 Mahesh, V. A,, 151, 203 Mahesh, V. B., 67(Mla), 68(B36), 70 ( M l a ) , 80, 132, 136 Mahim, D. T., 23(S18), 63 Maier, V. E., 223(D9), 300 Makara, G. B., 40, 60

Makita, M., 269(B5, M4, S17), 697, 304, 306

Malamed, N., 257(M5), 304 Malcolm, J. D., 21, 60 Malvaux, P., 156(K1), 180(K1), 208 Manahan, L., 354(R24), 381 Mancusi-Ungaro, P. L., 352(Wll), 384, 385

Mancuso, S., 174(B25, D6), 176, 196(D6), 205, 205, 209 Mandema, E., 321(D1), 336(D1), 354 (MlO), 360(D1), 370, 3?6 Maner, F. D., 155(M3), 158, 209 Manley, R. W., 9(B17), 43 Manly, B. M., 27(B3), 48 Mann, J., 71(G13), 74(G12), 133 Mannax, W. G., 4(L8), 60 Mmnering, G. J., 235, 240, 278, 297 Mansberger, A. R., 6(G7, GS), 17(G7), 47 Manson, D., 322(B39), 366 Marco, G., 257(23), 307 Marino, J., 357(S18), 358(S19), 359(S18), S81

Marion, D. F., 329(M13), 377 Markello, J. R., 197(U1), 21.4 Marks, L. J., 34(M8), 60 Markstahler, H., 316(H16), 373 Marrian, G. F., 191(B42), 208 Marsh, C. A., 148(M4), 209 Marson, R., 115(L7), 136 Martin, A. J. P., 229, 290, 302 Martindale, W., 324(M11), 376 Martinez, M., 405(010), 406(010), 421 Mason, M. F., 376 Masui, H., 403(K7), 419 Mateer, J. G., 327(M14), 329(M13, M14), 331(M14), 377 Mather, A., 279(M6), 304 Matsuda, K., 39, 60 Matthews, E. W., 412(G6), 417 Mattox, V. R., 113, 13’6, 192, 209 Mayer, D., 137 Mayes, D. M., 63(M3), 78(M3), 102 (M3, N6), 104(M3, N6), 127, 136 Meek, J. C., 402, 403(M13), 420 Mefferd, R. B., 27(H2), 28(H2), 48 Meister, A,, 257(J7), 302 Meistcr, L., 409(F3), 417 Mcites, J., 391(Sll), 392(S11). 493

AUTHOR INDEX

Mrltzrr, J. I., 321(M20), 325(W9), 330 (WS), 335(W12, W13), 336(W13), 353(M20), 360(M20, W9), 377, 384, 386 Meltzer, S. J., 7(M10), 50 Melvin, K. E. W., 413(F7, M14), 417,

@O Mendeloff, A. I., 318(M15), 325(M16), 327(11), 329(11), 331(11), 333(11), 337(11), 351(M16), 352(11), 356(11), 359(11), 374, 376 Mendenhall, C. L., 318(M17), 329(M17), 331(M17), 338(M17), 339(L9), 344 (M17), 351(M17), 357(M17), 376, 376 Mendlowitz, M., 249(W6, W7), 307 Menten, M. L., 319, 577 Mercier, J., 349(M18), 377 Meredith, 0. M., 312(T3), 341(T3), 383 Merimec, T. J., 410(H9), 418 Merrill, R. S., 334, 343(Cl), 368 Messerschmidt, O., 23, 49 Metcalfe, L. D., 290(M12, M13), 304 Metzler, C., 352(M21), 354(M21), 977 Meurman, L., 324(B22), 325(B22), 366 Meyer, C. J., 165(M7), 176, 209 Meynier, D., 314(C12), 315(C12), 368 Michaelis, L., 319, 977 Michaud, G., 71(B5), 130 Michie, A. J., 326(M23), 377 Michie, E. A., 155(K9), 159(S12), 161 (K8, S12), 169(S12), 170(S12), 171 (M9), 188(S12), 208, 210, 213 Michie, C. R., 326(M23), 377 Mickelsen, O., 20(K1), 49 Middlebrook, L., 343(W17), 986 Migeon, C. J., 60(R3), 137, 147(M10), 148(M10), 156(K1), 164, 168(Mll), 170(M12), 172(R10), 173(R10), 175 (M13), lSO(K1, MlO), 187(M10), 192(D3), 193(K1), 195(KlO), 196 (C2), 204, 206, 208, 210, 212 Mihaly, K., 40(M6), 60 Mikhail, G., 67(M5), 69(M4, M5), 70 (M4, M5), 78(M5), 136 Miksche, L. W., 22, 30(Mll), 60 Mikulecky, M., 331(M24), 377 Milbled, G., 346(V3), 348(B26), 350(V3), 366, 384 Miles, A. A., 3(M12), 4(M12), 61

445

Milhaud, G., 413(M15, M E ) , 420 Miller, E., 18(P5), 61 Miller, G., 347(21), 386 Miller, G. H., 317(M25), 377 Miller, L. L., 14, 64, 336(C4), 368 Mills, I. H., 165(M15), 210 Mills, M. A., 318(M26), 351(M27), 377 Mistilis, S. P., 336(K1), 343(K1, K2), 354(K2), 374 Mitchell, F. L., 91(W3), 96(W3), 98 (W3), 105(W3), 120(W3), 126(W3), 139, 142(C6, M16, S9), 143(M16), 147(B16, B18, C6, M16, s9, S10, s11, S13), 148(B18, S9, S10, 8111, 149 (B15), 150, 151(B18, S9, SlO), 152 (S9, SlO), 153(S10), 15469, S l l ) , 156(M16), 159(S12), 161(M16, 5121, 163(M16), 164, 166(B16, C7), 169 (S12), 170(512), 174(B17, B18, C6), 175(C6), 177(C6, L l l ) , 178(B16, B18, C6, SlO), 179(C6), 181, 182, 183(C7, S9, S l l ) , 185(S9, S13), 187 (C6), 188(S9, S11, S12), 192(59), 193 (L7, S9), 195(C7, SlO), 198(L7), 199(L7), 200(L8, S9), 202, 204, 209, 210, 213

Mitchell, M. L., 406(M17), 4.20 Mitchell, R. D., 343(C30), 969 Mitchell, R. G., 196(M17), 210 Mitchie, E. A., 124(K8), 156 Miyai, K., 403(K7), 4l9 Mizuno, M., 172, 173, 210 Moister, F. C., 342(C14), 568 Mollin, D. L., 355(N4), 378 Mollison, R. L., 342(M28), 343(M28), 377 Monroe, L. S., 321(M29, M30), 336 (M30), 360(M30), 377, 978 Moore, C., 63(K4), 82(K4), 134 Moore, F. D., 3(W5), 9(M14, R7), 13 (M14), 14, Zl(M13, M14), 22, 25, 34 (M13), 35(W5), 36(M3), 60, 61, 64 Moore, M. R., 70(G6), 133 Moore, S., 312(G9), 372 Morain, W. D., 413(M18), 420 Morey, G., 342(M31), 378 Morgan, E. H., 346(N12), 378 Morgan, H. G., 10(W4), 64 Morico, J. I., 16(D4), 46

446

AUTHOR INDEX

Moscatelli, E. A., 231(H16), 235(H16), 302

Mosier, H. D., 196(M19), 810 Moss, A. M., 250(H18), 302 Moss, D. W., 91(W3), 96(W3), 9S(W3), 105(W3), 120(W3), 126(W3), 139 Moss, M. S., 136 Moukhtar, M. S., 413(M15), 4.80 Moye, H. A., 2'2l(M14), 304 Moyer, C. A., 22(C4), 31(C4), 43 Moyer, J. H., 353(M32), 378 Mrhz, M., 9, 21(M15), 51 Mueller, M. N., 350(G2, M33), 351(G2), 871, 378

Mulholland, J. H., 20(C15), & Muller, M., 240(C18), 299 Mulrow, P. J., 197(M20), 210 Munch, O., 4(M16), 51 Mund, A., 356(F4), 871 Munkner, T., 339(W27), 340(W27), 358 (W29). 385 Munro, D.'S., 400, 402(D3, D4, D5), 403 (D3, E6), 416 Munro, H. N., 7(M19), 19(M17, M19, M20), 20(C28, C35, C36, F3), 21 (M19), 24(F3, M17, M19, M20), 38,

44, 45, 46, 51 Munson, P. L., 411(H7, H8), 412(H8), 413(A6, A7, H7, H8), 414, 416, 418 Murasawa, Y., 172(M18), 173(M18), 210 Murawec, T., 86(T9), 138 Murphy, B. E. P., 102(M7), 104(M7), 111, 113(M7), 123, 124, 127, 136 Murray, D., 63(M8), 136 Murray, J. F., 340(M34), 378 Murthy, P. V. N., 401(M12), 420 Musser, B. O., 148(C9), 204, 240(C5), 296(C4), 299 Mustala, 0. O., 350(P3), 879 Myers, F. L., 148(M21), 210 Myers, J. D., 378

N Nabarro, J. D. N., 79, 104(C1), 132 Nadasdi, M., 351(N1), 378 Nadeau, G., 331(N2), 378 Naftolin, F., 147(E2), 154(E2), 155(E2), 158(E2, 5 2 3 , 167(522), 170(E2), 171 (E2), 200(S22), 206, 213

Nagy, G., 9(W15), IO(W15, W17), 21 (WIG), 54, 55 Nair, P. P., 296(N1), 304 Nakajima, T., 162(N1), 210

Nakao, T., 137 Nakashima, K., 312(T2), 316(T2), 383 Nakata, K., 325(K21, N3), 875, 878 Nalbandov, A. V., 132 Nardi, G. L., 6, 51 Natelson, S., 296(N2), 804 Nathanson, I. T., 22(C13), & Natoli, A,, 161(N2), 210 Nauman, A., 411(N1), 4.820 Nauman, J. A., 411(N1), 4.90 Naylor, J., 324(L5), 330(L5), 338(L5), 340(W), 342(L5), 352(L5), 360(L5), 875

Neale, G., 355(N4), 378 Neale, F. O., 343(B24), 366 Nebel, L., 340(M34), 878 Needham, A. E., Zl(N2, N3), 51 Neefe, J. R., 352(N5), 878 Neff, E. G., 6(G7), 17(G7), 47 Neher, R., 411(N2), 4.80 Neill, J. D., 97(N1), 123(N2), 136 Neisler, J., 78(H5), 79(H5), 134 Nelson, D. C., 102(N3), 104(N3), 117 (N3), 123(N3), 136 Nelson, D. F., 280, 292, 293, 304 Nelson, D. H., 34(C12), &, 85, 102(N4), 104(N4), 123(N4), 136, 149, 180, 181, 198, 210

Nerenberg, C. A,, 120, 136 Nes, W. R., 294(B15), 298 Nesbitt, S., 352(25), 353(25), 359(25), 386

Neubeck, C. E., 323(N5a), 378 Neuhaus, 0. W., 17(N4), 18(N4), 51 Neumayr, A., 328(N6), 329(N6), 334 (N6), 378 Nevy, R., 322(B39), 366 New, M., 159(B31), 189(B31), 191(B31), 191(B31), 193(B31), 195(B31), 803 New, M. I., 191(N4), 210 Newman, E. S., 404(516), 423 Newport, H. M., 69(H3), 134 Nguyen, B. L., 171(P6), 211 Niall, H. D., 411(P7), 4.81 Nicholes, J. K., 18(S6), 52 Nichols, T., 108(N5a), 136

447

AUTHOR INDEX

Nickerson, M., 13, 51 Nicolosi, G., 161(N2), 210 Niemi, M., 159(N5), 210 Nieweg, H. O., 354(M10), 876 Nikelly, J. G., 290, 291, 296(N5), 304 Nikolai, T. F., 409(N3), 420 Nikolski, A,, 86(T9), 138 Nilsson, N. J., 326(B16), 365 Nims, L. F., 38, 51 Nishizawa, E. E., 89(B22), 116(B22), 123 (B22), 127(B22), 131 Nomeir, A. M., 353(Bl), 364 Norcross, J. W., 336(N7), 337(N7), 878 Nordyke, R. A,. 312(T10), 341(N8, N9, TlO), 355(N8, N9), 378, 388 Norman, G. J., 342(S9), 360(S9), 881 Northcote, D . H., 148(M21), 210 Norymberski, J. K., 149(A7), 160, 161 (A6, E8, N6), 177(A7, N7), 201, 206, 210, 211

Nosslin, B., 331(N10), 346(Nl1, N12),

Ollagnon, C., 176(B13), 202 Olney, J. M., 9(H14), 21(H14), 48 Olson, J. A., 296(D12), 300 O’MBille, E. R. L., 352(03), 379 Oppenheimer, E. H., 158(02), 211 Oppenheimer, J. H., 6(G7), 17(G7), 47, 404, 405(05, 010, 011, 0131, 406 (010, 011, 013), 407(06), 408(05, 08, 012), 409(07, 09), 410(B7, H9, 011, 0121, 414, 418, 421, @3 Orimo, H., 412(C1), 416 Orb, J. F., 257(22), 307 Ortegren, V. H., 290(H20), 302 O’Shaughnessy, M . C., 15(01), 51 Osinski, P. A., 165(03), 211 Oslapas, R., 113(S4), 138 Osterholm, J. L., 22(C4), 31(C4), 43 Ostrow, B. H., l l ( T 3 ) , 63 O’Sullivan, J. V. I., 34(M8), 60 Ott, H., 316(B17), 317(B17), 366 Ottenstein, D. M., 221, 229, 231, 233, 505

878

Nugent, C. A., 63(M3), 78(M3), 102 (M3, N6), 104(M3, N6), 108(N5a), 127, 136 Nunez, E. A,, 412(N4), 420 Nunn, R. F., 280(C9), 299 Nydick, I., ll(W22), 55 Nylander, G., 15(N7), 51

Overall, J. E., 329(04), 379 Overend, W. G., 269(H6), 301 Owen, C. A,, 327(05), 379 Owen, J. A., 16(MI, 02), 17(M1, 021, 18(M1, 0 2 ) , 60, 51, 358(J1), 360(51), 374

Oxenkrug, G. F., 72(L1), 135

0

P

Oakey, R. E., 115(Hla), 121(Hla, Ol), 123(Hla), 134, 136, 171(S14), 918 Ober, W. B., 162(R14), 212 Oberholzer, V. G., 197(R19), 212 Obrinsky, W., 342(01), 343(01), 378 Ockner, R. K., 351(N1), 378 O’Connor, D., 404(G2), 417 Odell, W. D., 394(02, UZ), 395(02, 0 3 ) , 396(02), 399(01, 02), 401(02, 04), 402(03), 405(03), 420, 421, 424 Odin, L., 18(B8a), 43 Oertel, G. W., 99(02), 186, 148(01), 211 Oette, K., 290(01), 304 Ohline, R. W., 304 Ohlsson, S., 343(S47), 383 Oka, H., 400(A8), 414 O’Leary, P. A., 327(02), 378 Olesen, K., 20(S4), 52 Oliner, L., 394(1,3), 419

Pagliardi, E., 336(P1), 379 Pain?, C. G., 177(Lll), 209 Pal, S. B., 99(P1), 136 Palefski, I. O., 312(P2), 3?9 Palframan, J. F., 221, 223, 228, 306 Palmer, R., 161(P1), 211 Palva, I. P., 350(P3), 379 Papadatos, C., 180(K7), 208 Papper, E. M., 349(H1), 372 Paraf, A., 327(A8), 864 Parker, J. G., 356(F4), 371 Parker, K. D., 279(P4), 281, 306 Parlow, A,, 392(B16), 415 Parmentier, C., 413(M16), 480 Parson, W., Zl(H15, H16), 48 Parsons, J. A,, 413(M3), 419 Partington, J . R., 311(P4), W9 Parzer, O., 328(N6), 329(N6), 334(N6), 378

448

AUTHOR INDEX

Pascal, S., 288, 300 Pasqualini, J. R., 113, 137, 165(P2, P3), 166(P4), 167, 171(P6), 183(P4, P5), 211

Pastan, I., 393, 39!3(F1, Pl), 400(M1, P1, P2, P3, P4), 417, 419, 421 Paterson, J. C. S., 340(524), 382 Paterson, R. E., 9(H14), 21(H14), .48 Patti, A. A,, 241, 306 Paul, B., 296(K11), 303 Paul, W. E., 399(01), 420 Paulsen, C. A., 100, 108(P3), 121(P3), 137

Paulsen, E. P., 175(P8), 176(P8), 211 Paulson, M., 354(P5), 379 Paumartner, G., 330(L8), 352(L8), 376 Payne, M. A., 354(A5), 364 Pearlman, M. R. J., 74, 82(P5), 137 Pearlman, W. H., 74, 76(P4), 82(P4, P5), 137

Pearse, A. G. E., 412(B20, P5), 416, 421 Pearson, 0. H., 78(K1), 134 Peaston, M. J. T., 25, 61 Peck, H. M., 337(M5), 345(M5), 376 Peck, L., 39(E4), 46 Pekkarinen, A., 34(P2), 61 Pelka, J. R., 290(M12, M13), 304 Peltier, L. F., 5(P3), 61 Pence, J. W., 290(H20), 302 Pendry, A., 296(P6), 305 Pennington, G. W., 168(L1), 208 Penrod, K., 344(B58), 368 Perrett, R. H., 231(B16), 233(B16), 298 Perry, M. B., 269(J8), 302 Perry, S. G., 294, 305 Pessotti, R. L., 320(B48), 324(B49), 341 (B50), 351(B49), 367 Peters, B. J., 350(K14), 351(K14, W2), 375, 384

Peterson, E. P., 161(J1), 207 Peterson, R. E., 74, 85(P6), 86(Gla), 89(Gla), 102(Gla), 105(P6), 112, 113(P6, P6a), 128(P6), 133, 137, 186(P9), 191(N4), 210, 211, 241(K9), 303

Pettitt, B. C., 221(S6), 290(S5), 305, 306 Pewis, A. A., 13(L4), 50 Pfeiffer, V., 99(F4), 132 Philipp, E., 174(P10), 211 Phillips, R . A , , 6(V1), 64

Philp, J. R., 379 Pier, C., 351(W14), 385 Pieragnoli, E., 351(T12), 354(T11, T12), 383

Pieroni, R . R., 324(L14), 376 Pilling, M. A,, 20(Hll), 48 Pillsbury, H. C., 296(P8), 306 Pincus, G., 137 Pion, R. J., 147(S20), 153(Cll), 154 (S20), 155(C16, SZO), 156(D9, 5201, 158(S20), lSl(Cl6, P11, P12), 163 (KZ), 170(C16, SZO), 183(Cll), 196 (D9), 204, 206, 208, 211, 913 Piti, T., 120(V10), 139 Pitt-Rivers, 404(G2), 417 Plaa. G . L., 321(W16), 322(W16), 323 (W16), 338(K9), 344(K9, R13), 347 ( E l ) , 359(P7), 360(P7, T5, W16), 370, 374, 379, 380, 383, 386

Plager, J. E., 63(F1), 64(F1), 132 Plantin, L.-O., 17(B5), 21(B5), 30(B5), 42, 77(B10), 130, 191(B19), 208 Pleshakov, V., 402(K6), 419 Pliitner, K., 336(P8), 379 Pollack, R. I>., 6(E1), 46 Pollard, H. M., 345(B33), 366 Popper, H. P., 324(P9), 342(P9), 347 (PlO), 350(G8), 351(54), 371, 379, 381

Poraga, C., 115(G7), 133 Porcaro, P. J., 223(P9), 306 Porter, C. C., 113(S5), 138 Posen, S., 343(B24), 366 Posnick, E., 312(T10), 341(T10), 383 Posternak, T., 399(P6), 400(P6), 4.21 Potter, E. L., 157(P13), 211 Potts, J. T., Jr., 411(P7), 412(D1), 416, 421

Powell, H., 280(C15), 299 Prader, A., 156(G3), 195(G3), 196(P14), 206, 211

Pray, L. L., 57(P9), 60(P9), 67(P9), 76 (P9), 80(P9), 83(P9), 85(P9), 90 (P9), 99(P9), 103(P9), 105(P9), 111 (P9), 114(P9), 116(P9), 118(P9), 120(P9), 137 Preedy, J. N., 99, 108, 137 Preedy, J. R. K., 155(A3, M3), 201, 209 Preeyasombat, C., 156(K1), 180(K1), 608

AUTHOR INDEX

Preisig, R., 336(P11), 343(P11), 379 Prendergast, J. J., 17(P4), 61 Pretorious, P. J., 356(K7, K8), 374 Preusse, C., 322(B14), 366 Price, J. G. W., 221, 306 Priest, S. G., 317(W5, W7), 384 Prince, A. L., 7(H8), 48 Pritchard, M. M. L., 4(T5), 64 Probst, V., 18(P5), 61 Prosser, A. R., 296(L7), 303 Prudden, J. F., 324(B14a), 366 Prystowsky, H., 210 Puestow, R. C., 340(B3a), 366 Pugh, P. M., 348(P12), 379 Pullar, J. D., 7(S15), 2O(C1), 21(C1), 22 ((311, 24(C1), 38(C6), 43, 44, 63 Purdy, R. H., 405(P8), 408(S13), 410 (S13), 421, @3 Purnell, J. H., 231(B16), 233(B16), 298 Puro, H., 117(Pll), 137 Purves, H. D., 393, 395(A3, A4), 397(A2, A4), 401, 414

449

Redgale, E. S., 40, 52 Reeck, G., 115(L7), 136 Reemtsma, K., 331(R4), 339(R3), 379 Reeve, E. B., 5, 14(G19), 15(G19), 47 Refetoff, S., 409(R5), @2 Rehm, C. R., 296(K2), 303 Reich, J. S., 354(R5), 379 Reichlin, S., 395(R6), 4286 Reichstein, T., 192(F1), 206 Reid, J. D., 196(R1), 211 Rcidt, V., 21(H15), 48 Reinhold, J. G., 329(13), 352(N4), 358, 374, 378, 380 Remer, A., 403(F2), 417 Renard, M., 257(M5), 304 Rcnwick, A. G. C., 177(C10), 204 Rcslcr, P . C., Jr., 102(N3), 104(N3), 117(N3), 123(p\'3), 136 Rcuttcr, F. W., 3(W5), 35(W5), 64 Reynolds, B. I,., 4(R3), 15(R4), 62 Reynolds, G. A., 392(B15), 416 Reynolds, J. W., 1 4 2 ( B ) , 147(R2, R3, R6), 152(R2, R3, R6), 177(R8), 179 (RS), 180(R8), 183(R2, R3, R5, R7), Q 186(RS), 187(R8), 188(R6), 193, 212 Quin, J. I., 346(R12), 380 Rhaney, K., 196(M17), 210 Rhodes, A. H., 138 R Rice, B. F., 156(R9), 212 Raban, P., 324(R1), 379 Rice-Wray, E., 350(R7, R8), 380 Rabinowitz, D., 410(H9), 418 Richards, T . G., 317(A14), 318(A13), 326 Raiss, L. G., 413(F10), 417 (RlO), 331(R10), 332(R10), 334(C21, Rakoff, A. E., 74(P5), 82(P5), 137 RlO), 336(R10), 345(G7, RS), 352 Rall, J. E., 404(RBR10), 405(R9, RlO), (A14, R9), 364, 365, 369, 371, 379, 380 408, 409(W4), 410(R9), 426, 424 Richardson, V. I., 63(K4), 82(K4), 134 Rall, T. W., 399(G1, K4), 417, 419 Richman, E. L., 342(S1), 381 Ramey, E. R., 36(R1), 62 Richmond, D. R., 23(R5), 66 Ramsey, C. G., 13(G20), 48 Ricketts, C. R., 16(D1), 17(D2), 46 Ramm, 0. L., 321(J5), 351(W15), 374, Higiero, C. S., i7(Bll), 43 386 Riggs, D. S., 404, 422 Ramscharan, S., 178(F3), 206 Illley, c . , 280(C9), 299 Rapaport, E., 331(W18), 334(W18), 339 Rimington, C., 346(H7, R12), 372, 380 (W18), 340(W18), 385 Rinehart, W. B., 345(S26), 3886 Ratanasopa, V., 169(54), 213 Ring, W., 240(B2), 697 Rausch, G. Z., 257(W1), 307 Riniker, B., 411(N2), 420 Rawson, R. A., 317(R2), 340(G22, R2), Riondel, A., 65(R2), 86(R1, RZ), 89 312, 379 ( R l , RZ), 94, 97(R1, RZ), 115(Rl, Read, S. I., 304 R2), 128(R1, RZ), 137 Recant, L., 404, 422 Ritchie, H. D., 346(H7), 372 Redding, T. W., 390(52, S3), 391(B14, Rittell, W., 411(N2, R7), 420, 422 R2, R4, S3), 392(R2, R3), 416, 422 Ritter, J. A., 353(B18), 366

450

AUTHOR INDEX

Rivarola, H. A., 60(R3), 137, 172, 173 Rossier, R.: 343(V8), 384 Rossipal, E., 178(F3), 206 (RlO), 212 Rossiter, R. J., 20, 62 Rivlin, R. S., 401(04), 421 Roach, P. J., 28(T4), 32(T4), 38(T4), 64 Roth, J., 393(P4), 400(P4), 421 Roth, L., 160(25, a), 186(Z6), 216 Rob, C. G., 13(E2), 46 Robbe, H., 77(B10), 130, 191(B19), 202 Rotor, A. B., 354(R24), 381 Robbins, J., 393, 394(C7), 399(C7), 404 Rourke, G. M., 22(C13), 44 (R8, R9, RlO), 405(R9, RlO), 408, Roux, H., 176(B13), 202 Rowntree, L. G., 311, 312(A2), 317(G20), 409(B5), 410(R9), 414, 416, 422 327(02), 354(G20), 364, 373, 378, Robel, P., 71(B4), 180, 202 581 Robelet, A., 348(B26), 366 Roberts, J. B., 96, 97(R4), 120(R4), 187 Roy, A. B., 148(R16), 212 Roberts, K. D., 71(R5), 137, 177(Rll), Roy, E. J., 155(R17), 212 Royer, P., 197, 212 185(R12), 212 Rubin, E. C., 34(D3), 45 Roberts, R. J., 344(R13), 380 Robertson, J. S. M., 9(C37), 10(C37), Ruchelman, M. W., 241, 247, 248, 305 Rudd, A., 404(D2a), 416 19(C37), 24(C37), 46 Rudhe, U., 168(K3), 208 Robertson, M. E., 187(R13), 212 Rudolph, L. A,, 11(1111), 52 Robinson, C. J., 413(M3), 419 Robinson, G. L., 327, 328(R14), 343 Ruegsegger, P., 11(W22), 55 Ruhlmann, K., 257(R5), 305 (R14), 344(R14), 361(R14), 380 Runnebaum, B., 147(24), 155(25), 173 Robinson, R., 16(R6), 62 (Z4), 174(24), 215 Robinson, R. R., 335(W10), 384 Russel, P. T., 294(B15), 298 Robison, G. A., 399(S25), 423 Russel, S. H., 168(J7), 208 Robscheit-Robbins, F. S., 14, 64 Russell, A., 197, 21% Roby, C. C., 162(R14), 212 Russell, J. A., 6(R12, W13), 62, 64 Roets, G. C . S., 346(R12), 380 Ruthven, C. R. J., 296(K2), 302 Roginsky, M. S., 185(54), 207 Ryan, C. M., 357(L10), 376 Rohrschneider, L., 225, 306 Ryan, K. J., 147(Ml), 209 Roitt, I. M., 402(E2), 416 Ryan, J. W., 13(M2), 60 Roman, W., 364 Rybak, M., 390(S6), 391(S6), 422 Rongone, E. L., 296(R2), 305 Ryhage, R., 90, 102(R6), llO(R6), 197 Root, A. W., 189(B30), 192(B30), 9203 Rylance, H. J., 136 Roselli, A., 196(R15), 218 Rosenau, W. H., 311(B29), 312(R15), Rynearson, E. H., 356(C35), 369 366, 380 S Rosenbaum, P. T., 31(G12), 47 Rosenberg, D. H., 311(Rl6), 329(R16), Sachazkaja, T., 166, 212 Saez, J. M., 142(L17), 156(L17), 209 380 Saffan, B. D., 155(M3), 158(M3), 209 Rosenberg, S. A., 9(R7), 6.8 Sakiz, E., 390(G8, G9), 391(G10), 392, Rosenbund, B., 18(R8), 62 Rosenthal, F., 311(R21), 380 417, 418, 492 Rosenthal, S. M., 311, 314(R22, R23), Salaman, D. F., 391(All), 392(A11), 414 316(R18a, R19), 317(R18a), 327 Salassa, R. M., 192(M5), 209 (R23), 329(R23), 336(R23), 337 Salganicoff, L., 137 Salokangas, R. A. A., 99(52), 104(52), (R23), 342, 349(R20), 380 121(S2), i s 7 Rosie, D. M., 293, 307 Salmon, G. W., 342(51), 381 Ross, G. S., 9(G1), 47 Samperez, S., 35, 39(J6), 49 Ross, H., 9(J5), 36(R9), 48, 62 Samuels, L. T., 85, 102(N4), 104(N4), ROSS, J. E., 422

AUTHOR INDEX

123(N4), 136, 149, 180, 181, 198, 210

Sandberg, A. A., 165(S2, S3, 5231, 212, 213, 214

Sanders, H. W., 343(H21), 351(H21), 373

Sandor, G., 36, 62 Sandor, T., 187(D18), 906 Sandler, M., 296(K2), 302 Sans, M. C., 114(S2a), 117(S2a), 138 Sapirstein, L. A., 324(S2, 5301, 338(S2), 340(S30), 360(S2, S30), 361(S2), 381, 382 Sarmiento, R., 294(B10, B11, B12), 298 Sato, T., 322(B36), 366 Savard, K., 67(L4), 136 Savery, A., 413(A6), 414 Savory, J., 296(S1), 306 Sawardeker, J . S., 270, 271, 306 Sawyer, B. C., ll(S12a), 63 Sawyer, D. T., 222, 306 Sayegh, J. F., 116(V8, V9), 120(V7), 139 Scanlon, W. A,, 353(S35), 382 Scarf, M., 342(K3), 374 Schaefer, J. A., 11(R11), 62 Schaefer, J . H., 18(S6), 62 Schaffner, F., 324(P9), 342(P9), 351(S3, S4), 379, 381 SchaUy, A. V., 390(S2, S3),391(B14, R2, R4, S3), 392(B12, B13, B15, B16, R2,

m),416, @2

Schatz, D. L., 410(S4), 422 Schellong, F., 318(S6), 381 Schenker, S., 323(Gll, S7), 371, 381 Schenker, V., 20(B19, S2), 4.3, 62, 347, (DS), 370 Scherb, J., 350(S8), 351 (SS), 381 Schiff, L., 353(S12), 381 Schildt, B., 23(L9, S3), 60, 62 Schillinglaw, J., 403(55), 42.2 Schindler, A. E., 169(S4, S5), 170(S5), 213

Schivers, J., 71(G13), f33 Schlacter, L., 353(S35), 382 Schloegel, E. L., 218(G5), 219(G5), 278 (G5), 301 Schloss, E., 356(B30), 366 Schmid, R., 338(H2), 346(H2), 372 Schmidle, B., 296(K10), 303 Schmidt, C. L. A., 342(S9), 360(59), 381

451

Schmidt, L. A., 340(S10), 343(S10), 581 Schmidt, M. L., 357(511), 381 Schmits, A. A., 290(M12, M13), 304 Schnak, H., 347(C22), 369 Schneider, E. M., 353(S12), 381 Schneider, W. S.,311(A1), 364 Schoemaker, W. C., 3(W5), 35(W5), 64 Schoen, A. M., 13(D10), 46 Schoenfield, L. J., 331(S15), 336(514, S15), 344, 345(S17), 354(S16, 5171, 355(S15), 360(S15), 381 Sch@nheyder, F., 2O(S4), 62 Scholtz, D. A., 403(W5), 424 Schreiber, V., 390, 391(S6), 422 Schreier, K., 6(S5), 62 Schuller, E., 160, 213, 359(G3), 571 Schults, E. W., 18(56),66 Schultz-Contreras, M., 350(R8), 380 Schumacker, G., 18(P5), 61 Schussler, G. S., 409(S7), 4.22 Schuster, G., 13(DlO), 46 Schwartz, H. L., 407(S8), 422 Schweppe, J. S., 125(J5), 134 Schwers, J., 156(D9), 160, 196(D9), 606, $07, 213

Scoggin, W. A., 155(H4), 207 Scott, J. S., 115(Hla), 121(Hla, O l ) , 123 (Hla), f34, 136, 168(J6), 208, 257 (J7), 302 Scott, R. L., 410(H9), 418 Scott, T. W., 400(S9), 4.22 Scrimshaw, N. S., 25(S7), 63 Scudamore, H. H., 342(M31), 378 Seal, U. S., 404, 405(S10), 409(52, N3), 418, 420, 423

Searle, G. L., 408(C3), 409(C3), 416 Sebastier, M. R., 349(M18), 377 Segaloff, A., 296(R2), 306 Seiling, V. L., 403(A10), 414 Selenkow, H. A., 409(R5), 4% Seligson, D., 357(S18), 358(S19), 359, 381

Selkurt, E. E., 339(S20), 382 Sellars, E. A,, 30(Y2), 66 Sendelbeck, L. R., 63(K4), 82(K4), 134 Serby, A. M., 342(S21), 382 Sevitt, S., 4(S8, S9), 5, 63 Shackelton, C. H. L., 142(59), 143(G7), 147(S9, S10, S l l , Sl3), 148(S9, SlO, S l l ) , 151(S9, SlO), 152(59, SlO),

452

AUTHOB INDEX

153,154(S9, Sll), 159,161(812), 169 (S12), 170(512), 175(S9), 178(510), 182, 183(S9, Sll), 185(S8, S9, 5131, lSS(S9, Sll, S12), 192(S9), 193(L7, S9), 195(S10), 198(L7), 199(L7), 200 (L8,S9), 207, 209, 213 Shafran, M. S., 175(P8), 176(P8), 211 Shahrokhi, F.,264, 265,305 Shahwan, M.M., 171(S14), 213 Shapiro, M., 176(M2), 209 Sharma, D. C.,78(G1), 83(G1), 120 (N5), 133, 136, 166(D15), 192(S15), 205, 213

Sharp, G.,37(C7), 44 Shaw, G.B.,35(C38, C39), 45 Shay, H., 327(L15), 345(L15), 353(B18), 366, 3ro

Sheehan, J. C.,257(C13), 299 Shelesnyak, M.C., 104(S9), 125(S9), 138 Sheltun, T. G., 328(B4), 330(B4), 340

(B4),343(B4), 365 Shen, N.-H. C., 175(F4), 206 Sheppard, A. J., 296(L7, PS), 303, 305 Shepard, T. H., 175(M13), 210 Sherlock, S., 339(S22), 340(S24), 341 (S22),357(S22), 363(S23), 382 Shibata, H., 343(C30), 369 Shill, 0.S., 318(B51), 341(B50), 367 Shimisu, N., 89(B18), 94(B18), 115

( B W ,131 Shires, T., lO(SlO), 53 Shoemaker, W.C., 353(S35), 382 Shore, M.I., 318(S25), 382 Short, A.

H., 352(03), 379

Short, E.,17(D9), 46 Short, R. V., 69(S3), 72(S3), 108(S3), 122,138, 155(A3), 158,201, 213 Shotton, D., 345(526), 382 Shraly, R.K., 355(H15), 373 Siebenmann, R.E., 196(P14), 211 Sieber, P.,411(R7), 422 Siiteri, P. K., 169(S5), 170(S5), 174 (SlS),213 Silber, R.H., 113(S4, a), 138,187(519), 213

Silen, W., 353(S27), 382 Silverberg, M., 324(L5), 330(L5), 338

(L5), 340(L5), 342(L5), 352(L5), 360(L5), 375 Simcock, M. J., 343(528), 358(S28), 382

Simmer, H. H., 147(E2, S20, S21), 154

(E2, SZO), 155(E2, S20), 156(520), 158(E2, S20,S22), 167(S22), 170(E2, SZO), 171(E2), 188(521), 200(s22), 605, 213

Simmonds, P. G., 221(P10), 290(S5), 305, 306

Simonitsch, E., 160(25, Z6), 186(Z6), 216 Simons, R. L.,343(S29), 354(S29), 382 Simpson, A. M., 324(530), 340(S30), 360

(S30),382 Simpson, G.S., 279(C16), 280(C16), 899 Simpson, J. A. M., 324621,338(S2), 360

(SZ),361(S2), 381 Simpson, S.A.,133 Sims, P.,322(B34-B36), 337(B35, G25),

360(B35), 366, 372 Sinclair, L.,197(R19), 212 Singer, E.J., 350(GS), 371 Singleton, M.F.,269(F1), 300 Sinha, D., 391(Sll), 392(S11), 423 Siplet, H., 353(B18), 365 Sjiivall, J., 73,74,99(E4, G15), 132, 133, 138, 143(G7), 155(S22a), 207, 214

Sltlaroff, D. M., 341(C27), 369 Slack, E.J., 411(F6), 413(F7), 417 Slater, J. D.H., 35(S10a), 53, 73(W1), 139

Slater, T. F., 11, 12,53 Slaunwhite, W.R., Jr., 165(S2, S3,S23), 212, 213, 214

Sloneker, J. H., 270,271,305 Smith, A., 376 Smith, B., 269(S7), 306 Smith, C . M., 7(C40), 24(C40), 30(C40), 45 Smith, E. R., 135 Smith, H.P.,318(V9), 384 Smith, H.W., 314(530a), 382 Smith, I. P., 393,423 Smith, J. C.,405(013), 406(013), 408

(012), 410(012), 421 Smith, J. J., 330(S37), 382 Smith, J. S., 339(B5), 365 Smith, L.H., 9(H14), 21(H14), 48 Smith, M. A., 99(B21), 104(B21), 108 (BZl), 114(B21a), 121(B21), 131 Smith, P., 35(C16, C17), 66 Smith, P. E., 393,&3

AUTHOR INDEX

Smith, V. E., 223(W3), SO7 Smyth, B., 99(B21), 104(B21), 108(B21), 114(B21a), 121(B21), 131 Smyth, D. H., 346(J6), 374 Smythe, C., 339(H7a), 372 Smythe, C. V., 323(N5a), 318 Snnpe, W. J., 336(C4), 368 Snell, A. M., 338(S31), 356(S31), 982 Snell, E. S., 335(W10), 384 Socolow, E. L., 408(S13), 410(S13), 42s Sobel, E. M., 175(P8), 176(P8), 211 Sodmoriah-Beck, S., 99(F4), 132 Soetbeer, M., 174(PlO), 211 Sovell, L., 15(G4), 47 Soffer, L. J., 338(S32), 582 Soliman, N. A., 411(F6), 417 Solomon, D. H., 403(S14), 423 Solomon, H. C., 355(K13), 376 Solomon, S., 156(S24), 162(12, S24), 163, 165, 207, 214 Solomon, S. H., 391(S15), 42s Solth, K., 162, 216 Sommer, P. F., 296(K10), SOS Somogyi, M., SO6 Soskin, S., 311(R16), 329(R16), S80 Sower, N. D., 22(C4), 31(C4), 43 Sparks, R. D., 359(P7), 360(P7), Si'9 Spaulding, J. S., 156(K1), 180(K1), 193 ( K l ) , 195(K10), 208 Spellman, M. W., 328(B4), 330(B4), 340 (B4), 343(B4), 866 Spencer, B., 148(D14), 206 Sperber, I., 314(S33), 346(S34), 351 (S34), 582 Spoelstra, A. J. G., 36, 65 Spooner, M., 394(W1), 424 Sprinkle, E. P., 17(H12), 48 Squef, R., 405(011, 013), 406(011, 013), 410(011), 421 Stack, E., 231, SO1 Stacy, R. W., 118(S6a), 138 Staemmler, H.J., 162(N1), 210 Stakelum, G. S., 321(C31, C32), 322 (C32), 323(C32, C33), 324(C32), S69 Stakemann, G., 161(F6), 171(F7), 206 Stark, E., 40(M6), 60 Starkel, S., 142(S26), 21.4 Starnes, W. R., 138 Starr, P., 409(T1, T2), 423, 494 Starzl, T. E., 353(S35), 982

Staudgert, M. E., 409(W4), 424 Steelf, P. D., 327(M14), 329(M14), (M14), 377 Stefanine, M., 16(T6), 64 Stefanovic, M., 258, 259, 260, SO6 Stein, A. A,, 241(P5), SO5 Stein, K. E., 21(H15), 48 Stein, I. F., 138 Stein, M., 406(H4, H5), 418 Stein, S. W., 312(C16), 317(C16), (C16), 331(C16), 337(C16), (C16), 340(C16), 355(C16), (ClS), 568 Steinfeld, J. L., 326(S36), 988 Stekiel, W. J., 330(S37), 339(B5),

453 331

328 338 360

566,

382

Stellate, R. L., 296(N2), 304 Stempfel, R. S., 148(S27), 810, 214 Stenroos, L., 291, 292(M3), SO4 Sterling, K., 4(S13a), 16(S13a), 5S, 404 (Sl6, S18), 405(S17), 406(S19, S20), 407(S17, S H ) , 409(S17), 410(11, 517, SlS), 418, 4% Stern, M. I., 156(S28), 214 Sternberg, J. C., 293, SO6 Sternberg, W. H., 156(R9), d l 2 Stets, J. F., 317(W5, W7), S84 Stevenson, J. A. E., 20(B19, S2), 43, 62 Stewart, C. P., 9(W15, W19), lO(W15, W17-W19), 21(W16), 64, 66 Stewart, W. K., 10(W3), 64 Stiefel, M., 187(R13), 212 Stiles, M. H., 356(S38), 382 Stiles, M. T., 356(S38), S82 Stitch, S. R., 115(Hla), 121(Hla, Ol), 123(Hla), 134, 1.96, 148(S29, S30), 171(S14), 213, 214 St. John, P. A., 250(B22), 253(B22), 255 (B22), 256(B22), ,998 Stda, K. F., 186(A1), 201 Stone, C., 35(G13), 47 Stone, S. L., 344(S40), 345(S39), 346 (S39, S40), 348(P12, S39, S40), 356 (S40), 379, 382, 583 Stoner, H . B., 5, 6(G21), 7(G21, 515, 5171, 8(A2), 25, 29, 30, 42, 48, 63 Storm, E. E., 221(B27), 298 Strade, M. A., 354(S42), 583 Strain, W. W., 314(S43), 583 Strait, L. A., 355(C25), 369

454

AUTHOR INDEX

(B18, R1, FU), 94(B18, R1, R2), 97 (Rl, R2), 115(B18, R1, RZ), 128 (Rl, RZ), 131, 137, 178(F3), 206 Tanaka, S., 409(T1, TZ), 423, 424 Tamm, J., 60(T5), 98(T5), 138 Tanasoglu, Y., 326(C8), 336(C8), 368 Tansig, F., 16(D4), 46 Tapley, D., 422 Taplin, G. V., 312(T3), 341(T3), 383 Tashjian, A. H., Jr., 413(M14, T3, T4, T5), 420, 424 Tatum, H. J., 147(C13), 155(C13), 158 (C13), 170(C13), 204 Tavernetti, R. P., 409(07), 421 Taylor, F. H. L., 6(T1), 20(T1), 63 Taylor, H. L., 20(K1), 49 Teitelbaum, P., 257, 298 Tennant, R., 343(W17), 385 Therkelsen, A. J., 357(T3a), 358(T3a), 383 Thevenet, M., 148(H8), 207 Thomas, B. S., 117(T6), 128, 138, 238, 239, 300, SO6 Thomas, G. H., 86(T7), 87, 138 Thomas, L. J., 330(E6), 354(T4), 356 (E6), S O , 383 Thomas, M. C., 360(T5), 383 Thompson, J. D., 155(M3), 158(M3), 209 Thomson, J. Y., 331(B12), 366 Tho&, L., 23(S3), 62 Thornton, F. H., 353(S35), 382 Thorpe, W., 322(B53), 367 Thorsen, T., 186(A1), 201 Thornton, W. N., 155(H4), 207 Threlfall, C. J., 5(S16), 7(S17), 8(S17), 25, 63 Thurber, R. E., 38, 61 T Ticktin, H. E., ll(T3), 63 Tabachnick, M., 406(S19, S20), 423 Tillinger, K. G., 142(D10), 155(D10), Tada, M., 315(T1), 383 157(B3), 159(B3), 160(B3), 180 Tada, Y., 3 1 2 ( n ) , 316(T2), 383 (DlO), 186(D10), 202, 205 Tiihka, H., 157(T1), 914 Tilstone, W. J., 2(C33), 3(C33), 7(C40), Tait, J. F., 63(T3), 65(H6, R2), 74, 75 19(C33, C34), 24(C40), 25(C29, C30, (T3, T4), W H 6 , T4), 82(T3), 86 C33), 28(T4), 30(C40), 31(C29), 32 (Rl, R2), 89(B18, Rl, R2), 94(B18, (C31, C32, T4), 38(T4), 46, 64 R1, R2, T2), 97(R1, R2), 102(H6), 115(B18, R1, RZ), 128(R1, R2), 131, Timberlake, C. E., 174(W2), 215 Tindall, V. R., 326(R10), 331(R10), 332 133, 134, 137, 206 (R10). 334(Rl0), 336(R10, T6), 343 Tait, S. A. S., 65(R2), 86(R1, R2), 89

Strauli, V. D., ll(S12a), 63 Street, H. V., 225, 231, 233, 234, 278, 281 (M11, 512, S13), 282, 285, 287, 288, 289,304,306 Streicher, D., 326(C26), 369 Strickland, W. M., 4(R3), 15(R14), 62 Stromberg, L. R., 23(S18), 63 Strott, C. A., lOZ(SS), 123, 1 3 Stouffer, J. E., 267, 302, 411(S21), 4.93 Stowe, W. P., 352(S41), 383 Stubbs, R. D., 149(A7), 177(A7, N7), 201, 211 Sturner, W. Q., 279(S14), 306 Sturtridge, W. C., 412(S22), 423 Subbaram, M. R., 217, 306 Sulimoviei, S., 104(S9), 125(S9), 138 Sunderman, F. W., 328(S44), 340(S44), 383 Sundman, J., 346(S45), 383 Sundman, V., 346(S45), 383 Sunshine, P., &3 Surks, M. I., 405(011, 013), 406(011, 013), 407(06), 408(012), 409(09), 410(011, 012, S24), 421, 423 Sutherland, E. W., 399(K4, P6), 400 (P6), 419, 4.21 Sutherland, J. E., 399, 423 Sutherland, J. M., 347(S36), 383 Svanberg, A., 343(S47), 383 Svendsen, A. B., 232, 998 Sweeley, C. C., 231(H16), 235(H16), 369 (B5, S17), 270, 271(S16), 297, 309, 306 Sweeting, J., 336(P11), 343(P11), 379 Swinyard, C. A., 142(S31), 214 Sykes, P. J., 147(513), 185(S13), 213

455

AUTHOR INDEX

(TS), 334(T6), 345(07), 360(T6), 371, 380, 383 Tisdall, E. F., 9(G1), 47 Titajew, A.9 316(H19), 317(H19), 373 Tomisek, A. J., 115(T8), 138 Tompsett, S. L., 18(C34a), & Touchstone, J. C., 86(T9), 138 J., 316(T7)> 334(T7, T8), 355 (T7), 383 Toverud, S. U., 412(C8), 416 Tovey, J. E., 327(T9), 331(T9), 383 Townsend, J., 114, 119, 120, 138 Trammell, V., 343(C30), 369 Travis, R. H., 240, 241, 298 Travers, R. I., 355(H15), 3?3 Trolle, D., 155(B38), 164(B37, B43), 172 (B37), 180(B43), 198(B43), 203 Truckot, R., 347(B55), 367 Trueta, J., 4(T5), 64 Trupin-Bar, N., 344(V4), 384 Tse, A., 412(G6), 417 Tubiana, M., 413(M16), 490 Tubis, M., 312(TlO), 341(T10), 583 Tubman, J., 400(H2), 401(H3), 418 Tuey, G. A. P., 232, 297 Tumen, H. G.,341(C27), 369 Tura, S., 331(T12), 354, 383 Turner, D. A.9 218(K7), 2N(N1), 303,

V Vale, W., 391(Vl), 392(V2), 424 van Aller, R. T., 294(B15), 298 Van Buren, j. R., 3l8(V9), S84 Vance, V. K., 409(S7), @2 Vanden Bossche, H., 359(Vl), 383 Vanden Heuvel, W. J. A., 87(H4), 88 (H4), 99(H4), 110(H4), 116(H4), 134, 238, 239, 240, 249(H15, V3), 250, 251(H15), 253(H15, L13), 254(H15), 301, 304, 306 Vanderlaan, W. P., 402(M13), 403(M13), 420

van der Molen, H. J., 89(B22, V1, V2), 110, 116(B22, Vl), 122(Vl, V2), 123 (B22, Vl), 125(V1), 127(B22, V2), 131, 138 Van der Sluys Ueer, J., 413(B9), 414 Vande Wiele, R. L., 69(V3), 71(G13), 133, 139, 160(S7), 177(Rll, V l ) , 207, 212, 213, 214 van Duyne, R. P., 221(V5a), 306 van Kampen, E. J., 242, 243, 301, 307 Vanlerenberghe, J., 344(V2, V4), 346 (V3), 348(B26), 350(V3), 366, 384 Vanotti, A., 394(L2), 396(L2), 419 Van Slyke, D. D., 6(V1), 64 van Wijngaarden, D., 290(V7), 307 304 Varley, H., 3580'51, 384 Turpini, R., 16(T6), 64 Vassallo, D. A., 292(V8), 307 Tweedie, F. J., 162(12), 207 Vawter, G., 27(H2), 28(H2), 48 Tyler, E. T., 343(A6), 350(A6), S64 Vecchi, M., 296(K10, V9), 303, 307 Tyler, F. H., 108(N5a), 136 Vegeter, J. J, M., 321(D1), 336(D1), 360 Tygstrup, N., 339(W27, W28, W301, 340 (DI), s7o (W27, W30), 358(W29), 386 Vermeil, G., 197(R18), 218 U Vermeulen, A., 89(V4), 108(V4), 116 (V4), 123(V4), 127(V4), 139 Ulick, S., 197, 214 Verschure, J. C. M., 336(V6), 344(V6), Ullberg, S., 163(B8), 302 349(V6), 384 U1strom> A., l5l, 155(U3, u4), 172 Vest, M. F., 343(V7, V8), 384 (U4), 177(R8), 179(RS, U3), 180 Vehrgaard, p., 116W8, v9), 12o(v5, ( ~ 8 u, 4 ) , 181(u4), 1 8 6 ( ~ ) ,187 V6, V7, VlO), 139, 175(V2), 176(V2), (RS, u3), 198, 204, 212, a14 214 Upjohn, H. L., lO(Ul), 64 Vetter, H., 328(N6), 329(N6), 334(N6), Upton, G. V., 115(Bll), 130 378 Urquhart, J., 34(D3, YO, 39, 45, 66 Vetter, K. 2i4 Utiger, R. D., 391(W2), 392(W2), 394 (02, U1, U2), 395(02, R6, Ul), 396 VheneY, B. A.9 411(CIo), 413(c10), 416 (02), 398(U1), 399(02), 401(02), 403 Victor, J., 318(V9), 384 Vidal-Madjar, C., 223, SO7 (s.9, 4.90, 4.@, 4.94 K.p

197(u1)j

456

AUTHOR INDEX

Vihko, R., 155(S22a), 214 Villee, C. A., 156(V5, V6), 162, 214, 216 Villee, D. B., 1560'5, V6), 166(V4), 214, 216

Vink, C. L. J., 331(V10), 334(V10, V12), 342(Vl1), 357(V10), 384 Visser, H . K. A., 189, 191(D4, D5, V7), 192(D4), 197, 206, 216 Vivian, S. R., 391(K5), 419 Voelkel, E. F., 411(H8), 413(A7, H8, T4), 414, 418, 424 Volpe, R., 409(21), 424 von Euw, J., 192(F1), 206 von Falkenhausen, M., 311(R21), 380 von Miinstermann, A.-M., 147(Z4), 155 (Z4), 173(24), 174(24), 216

W Wade, A. P., 73(W1), 139, 168(J7), 201, 208

Wagner, J., 257(Wl), 307 Waldenstein, S. S., 339(Wl), 384 Walker, B., 270, 271(S16), 306 Walker, B. L., 258, 259, 260, 306 Walker, E. A., 221, 223, 228, 306 Walker, G. W., 279(C16), 2SO(Cl6), 299 Walker, J., 6(W1), 21(W1), 64, 155(G5), 206

Walker, M., 186, 208 Walker, W. F., 3(W5), lO(W2-W4), 35 (W5), 64 Wall, P. E., 210 Wallace, E. Z., 164(W1), 165(W1), 180 ( W l ) , 187(W1), 216 Wallace, J. E., 280(W2), 307 Walters, W., 317(G20), 354(G20), 372 Walton, D. R. M., 117(T6), 128,132, 138, 238, 239, 300, 306 Wang, C., 17(B12), 43 Wang, D. Y . , 63 Ward, D. N., 417 Ware, A. G., 358(Hl0), 359(H10), 37.5 Warren, J. C., 174(W2), 216 Warren, R., 18(W6), 64 Waterhouse, C., 409(15), 418 Watkin, D. M., 21(L5), 60 Watson, J., 131 Watson, R. N., 350(K14), 351(K14, W2), 376, 884

Watt, A., lO(W41, 64 Watts, J. McK., 358(J1), 360(J1), 374 Waxman, B. D., 118(S6a), 138 Webb, J. L., 223(W3), 307 Webber, J. M., 328(Z6), 340(26), 386 Weber, B., 271(W5), 277(W5), SOT Wedeles, P., 115(T8), 138 Wegrzynowski, L., 142(S26), 214 Weimer, H. E., 17(W8, W9), lS(W9), 37, 64 Wein, J. P., 257(G6), 301 Weinstein, B., 256, 307 Welborn, T. A., 9(J5), 36(R9), 49, 62 Welbourn, R. B., 34(J4), 49 Welch, M. T., 71(G13), 133, 207 Wells, H. W., 223(W3), 307 Wells, W. W., 269(B5, M4, S17), 271, 277, 297, 304, 306, 307 Wendeberg, B., 33(W10), 64 Wendel, R. M., 353(S35), 882 Wenger, J. J., 411(S21), 423 Wengle, B., 156(W3), 216 Werner, A. Y., 326(H22), 373 Werner, I., 18(B8a, WlOa), 43, 64 Werner, S. C., 411(Nl), 420, 351(W3), 384, 394(W1), 424 Wertheimer, 21, 64 Werther, J. L., 347(W4), 384 West, C. D., 78(K1), 13.4 West, H. F., 177(N7), 211 Westerfield, W. W., 322(E2), 370 Westman, A., 77(B10), 180, 142(D10), 155(D10), 160(Dll), 180(D10), 186 (DlO), 206 Westman, F. L., 191(B19), 202 Westphal, M., 180(B32), 187(B32), 903 Westphal, U., 317(W5, W6, W7), 384 Wettstein, A,, 73(W2), 139 Whedon, E. F., 318(A9), 364 Whedon, G. D., 17(D9), 46, 331(H6), 350(H6), 372 Wheeler, H. O., 321(J5, M20), 325(W9), 330(W9), 335(W10, W12, W13), 336 (W13), 351(W8, W14, W15), 352 ( W l l ) , 353(M20), 360(M20, W9), 374, 377, 384, 386

Whelan, F. J., 321(W16), 322(W16), 323 (W16), 360(W16), 386 Whetstone, H. I., 343(W17), 386

AUTHOR INDEX

Whipple, G. H., 14, 18(F4), 47, 64 Whitby, L. G., 91, 96(W3), 98(W3), 105, 120(W3), 126(W3), 139 White, B. V., 343(W17), 386 White, C. S., 23(R5), 62 White, D. M., 264(K6), ,903 White, E. C., 311, 314(R22, R23), 327 (R23), 329(R23), 336(R23), 337 (R23), 342, 380 White, R. M., 336(N7), 337(N7), 378 Whitehouse, M. W., 348(B59), 367 Whiteley, H. J., 6(G21), 7(G21), 48 Whitlock, R. T., 351(W14), 386 Whitman, P., 9(B17), @ Wiberg, C., 396(B2), 397(B2), 414 Widder, R., 269, 297 Widdowson, E. M., 27(B4), 4.2 Wiech, M., 316(B17), 317(B17), 366 Wiegand, B. D., 331(W18), 334(W18), 339(W18), 340(W18), 386 Wieland, R. G., 69, 80(W4), 139 Wiener, M., 162, 216 Wiggins, R. A., 155(M3), 158(M3), 209 Wightman, B. K., 403(A10), 414 Wilber, J. F., 391(W2), 392(W2), 394 (02), 395(02, 0 3 ) , 396(02), 399 (01, 021, 401(02), 402(03), 405 (03), 4.20, @ I , 424 Wilensky, L. S., 312(W19), 386 Wiley, A. T., 353(H24), 373 Wilhelmi, A. E., 6(W13), 64 Wilk, S., 249(C7, C8), 299, 307 Wilkins, R. W., 344(C37), 369 Wilkinson, A. W., 9(W15), lO(W15, W17), 20(W14), 21(W16), 24(W14), 64, 66 Williams, A. J., 346(H7), $72 Williams, C. M., 301, 307, 329(04), 367, 368, 379 Williams, E. J., 91, 139 Williams, R., 328(B44), 336(P11), 340 (B44), 343(P11), 367, 379 Williams, T., 62(W6), 73(W6), 139 Williams, W. G., 9(B17), 43 Williams, W. H., 20(Hll), 48 Williams, W. L., 315(W20), 386 Williamson, A., 391(M11), 397(M11), 402(Mll), 420 Willoughbp, M., 132

457

Wilmer, J. G., 330(E6), 356(E6), 370 Wilmink, R., 191(D4, D5), 192(D4), ,906 Wilson, J. D., 71, 131 Wilson, H., 2Z(C13), 44, 75(L12), 83 (W7), 105(W7), 126(W7), 129(W7), 136, 139 Wilson, R., 163(S25), 165(S25), 814 Wilson, R. E., 9(R7), 68 Wilson, W. C., 9(W19), lO(W18, W19), 66, 355(W21), 386 Winitz, M., 257(G6), 301 Winkler, K., 330, 331(W24, W25), 332 (W24-W26), 337(W22), 339(W27, W28, W30), W ( W 2 3 , W27, W30), 358(W29), 3% Winsten, S., 296(W9), 307 Winternitz, J., 21(H16), 4s Winzler, R. J., 273(L5), 275, 276, 303 Wiquist, N., 156(B24), 161(P1, P12), 163 (B8, S25), 165(S25), 167(P7), 174 (B25, D6), 176(M2), 185(B24), 196 (D6), $02, 203, 206, 809, 211, 214 Wirsching, W., 336(P8), 379 Wirts, C. W., 318(C3), 326(C2), 336 (C4, W31), 350(C2), 351(C3, W31), 352 (W32), 368, 386 Witherell, C. S., 116(V9), 120(V10), 139 Witzel, H., 71(W8), 72(W8), 139 Woeber, K . A., 405(P8, W3), 406, 408 (S13), 41O(S13), 4.21, 42% 424 Wolf, B. J., 161(Pll), 211 Wolf, T., 293, 307 Wolff, J., 409(W4), 424 Wolff, S. M., 72(G4a), 133 Wood, E. H., 312(F5), 371 Wood, M. E., 147(W5), g16 Wood, P. B., 322(B13), 366 Woodhouse, N. J., 412(G6), 417 Woods, G. F., 161(E8), 206 Woodward, K. T., 23(S18), 63 Woolner, L. B., 403(W5), 424 Worthington, W. C., Jr., 391(All), 392 ( A l l ) , 414 Wotiz, H. H., 241, 244, 307 Wotiz, K., 89(C2), 132 Wray, J. P., 20(W20), 65 Wright, A. D., 9(55), 36(R9), 4.9, 6.9 Wright, A. M., 20(C15), 44 Wright, F. S., 314(M6), 376

458

AUTHOR INDEX

Wrobleswski, F., 11(L1), 12(W21), 49, 66 WU, C.-H., 86(T9),138 Wurl, 0. A., 353(M32), 378 Wyler, C. I., 354(P5), 379 Wyngaarden, J. B., 74(P7), 137, 186 (P9), 211 Wyse, E. P., 403(W5), 424

Z

Zado, F. M., 221(Z1), $07 Zagerman, J., 341((2271, 369 Zaher, R. A. S., 353(B1), 364 Zak, B., 117(E3, P l l ) , 120(21), 132, 137, 199 Zala, A. P., 69(W4), 80(W4), 139 Zander, J., 67(M5), 69(M4, M5), 70(M4, M5), 78(M5), 136, 1 4 7 ( W , 155, 157, Y 158, 162, 163, 173(24), 174(Z4), 216 Zaninovich, A. A., 409(Z1), @4 Yaffe, S., 197(U1), 214 Yagle, E. M., 327(M14), 329(M14), 331 Zatuchni, J., 347(21), 386 Zederfeldt, B., 15(G4), 47 (M14), 377 Yakovac, W. C., 156(G2, G3), 195(G3), Zelman, S., 356(Z2), 386 Zieve, L., 329(24), 352(25), 353(Z5), 206 354(Z3), 359(Z5), 386 Yalow, R. S., 398, 414 Zileli, M. S., 3(W5), 35(W5), 64 Yamazaki, E., 390(G9), 391(G10), 418 Zilversmit, D. B., 318(S25), 382 Yannone, M. E., 115(G7), 133 Yarger, K., llO(Dl), 114(D1), 132, 242 Zimmerman, B., 3(Z1), 34(Z1), 66 Zimmerman, H. J., 330(E6), 354(T4), 356 ( D l ) , 296(D1), 299 (E6), 370, 383 Yates, F. E., 34(Y1), 39, 65, 63(K4), 82 Zipf, R. E., 328(Z6), 340(Z6), 386 (K4), 134 Zlatkis, A., 221(P10, S6), 257, 290(S5), Yee, H. Y., 296(Y1), 307 505, 306, 307 Yee, J. L., 279(P4), 306 Zlotnik, G., 147(S21), 188(S21), 213 You, R. W., 30(Y2), 65 Zomsely, C., 257(Z3), SO7 You, S. S., 30, 66 Young, A., 326(R10), 33l(R10), 332 Zuber, H., 411(N2), 420 Zucconi, G., 160(Z5, Z6), 186(Z6), 216 (RlO), 334(R10), 336(R10), 380 Zumoff, B., 178(F8), 191(F9), 206 Young, F . G., 35(C38, C39), 45 Zurbriigg, R. P., 175(B6), 202 Yudaev, N. A., 166(Y1), 216 Zwicker, M., 17(Z2), 65 Yudkin, S., 342(Y1), 343(Y1), 586

SUBJECT INDEX A Adrenocortical steroids, major pathways in biotrynthesis, 68 Adrenogenital syndrome As-3P-hydroxysteroid dehydrogenase, deficiency in, 77 in late adolescence and adult life, 78 polycystic ovaries and, 79 steroid 21-hydroxylase, deficiency in, 77 steroid 11 P-hydroxylase, deficiency in, 77 Amenorrhea, primary biochemical determinations in, 104 17-ketosteroids, urinary excretion of, 104 Androsterone, urinary excretion of, 66

colorimetry in, 357 protein precipitation methods in, 357 BSP tests in acute abdominal conditions, 355 circulatory disorders, 356 diabetes mellitus, 357 nutritional deficiencies, 357 obesity, 356 postoperative injury, 356 renal disease, 357 septicemia and other infections, 355 BSP uptake by liver, maximum rate, 335 BSP uptake and excretion acute hepatitis, effect of, 352 anesthetics and hypnotics, effect of,

349

antibiotics, effect of, 347 antiinflammatory drugs, effect of, 349 B benziodarone, effect of, 348 BSP bile salts,effect of, 351 conjugation of, 322-323 in biliary obstruction, 353 excretion of, in bile, 336 in carcinomatosis, 354 in urine, 336 cholecystographic agents, effect of, 345 hepatic uptake of, 318 in chronic hepatitis and cirrhoais, 353 maximum rate of transport to bile, circulatory disorders, effect of, 354 335 dinitrophenol, effect of, 348 metabolites of, 321 ethanol, effect of, 348 plasma clearance of, 333 ethionine, effect of, 348 protein binding of, 319 flavaspidic acid, effect of, 346 storage capacity in liver, 335 icterogenin, effect of, 346 toxic effects of, 341 in obesity, 356 transfer of, 324 phenothiazines, effect of, 347 effect of other dyes on, 325 phloriain, effect of, 346 relation of plasma, bile and lymph postcaval operations, effect of, 353 to, 325 probenecid, effect of, 345 use of, in determining sphincter of Oddi, relation to, 340 ascitic fluid volume, 341 steroids, effect of, 350 plasma volume, 340 BSP uptake and secretion BSP retention test, 327-328 age, 342 disappearance of dye from plasma in, body temperature, 344 330 exercise, 344 fractional disappearance rate in, 332 feeding, 344 BSP tests posture, 344 analytical methods in, 359 pregnancy, 343 automatic analysis of, 359 sex, 343 459

460

SUBJECT INDEX

C Cholephilic dyes, 314ff. characteristics of, 317 extrahepatic uptake of, 326 metabolism of, 323 radioactive, uee of, 341 transfer of, 325 intracellular relationship involved in, 326

Estrone plasma concentration, 65 structure of, 61 urinary excretion of, 66 Etiocholanolone, urinary excretion of, 66

F Fick principle, hepatic blood flow and, 339

G

D Dye tests in clinical medicine, 327 future of, 361f. normal values for, 329 techniques for, 328

E Early postinjury changes, 3ff. acid-base balance in, 10 blood loss and hemoglobin in, 14f. carbohydrate metabolism in, 7 cellular injury in, 9 electrolytes in, 9 enzyme changes in lymph, 11 in plasma, 12f. in urine, 13 hypercoagulability in, 15 lipemia and fat embolism in, 5 nitrogen metabolism in, 5 plasma proteins in, 16ff. albumins, 16 fibrinogen, 18 globulins, 17 renal, 4 serum isozymes in, 13 tricarboxylic cycle in, 8 Environmental factors in injury, 26ff. carbohydrate metabolism, relation to, 30f. physical, 28 protein metabolism, relation to, 30f. temperature, 27 Estradiol plasma concentration, 65 structure of, 61 urinary excretion of, 66 Estrogens, methods of determination, 121

Gas-liquid chromatography, applications of, 2376. amines, biological, 24%255 catecholamines, 24S251 tryptamine-related, 251-255 amino acids, 256-268 n-butyl N-trifiuoroacetyl esten, preparation of, 257-260 n-propyl N-acetyl esters, use in analysis, 260-263 sulfur-containing, analysis of, 264 thyroid amino acid hormones, analysis of, 264-267 carbohydrates, 268-278 glycoprotein sugar, analysis for, 273277 plasma galactose and glucose, 271273 preparation of volatile derivatives, 268-271 sugar in urine, 277 carbon-skeleton determination, 294 fatty acids, 290-292 pyrolysis, 292 steroid hormones, 237-249 aldosterone, 241 separation of testosterone and related compounds, 237-239 urinary estrogens, 244 urinary pregnanediol and pregnanalone, 245-247 G w l i q u i d chromatography, basic requirements of, 218-221 carrier gas, 219 columns, 220 detectors, 221 sample, introduction of, 220 Gas-liquid chromatography, column preparation in, 221-236 column packing, 234-236

461

SUBJECT INDEX

liquid phases, 224-234 supports, 222-224 Gasliquid chromatography, toxicological analysis, 278-290 alkaloids in urine and blood, 285-290 barbiturates, 280, 2% ethanol, 279 general considerations, 278 morphine, 282

I Immediate postinjury changes, 2f. Indicator dilution method, 340 Indocyanine green, 326 analytic method for, 360 retention test, 337 toxic effects of, 342 Infants, abnormalities in steroid production, 189-197 adrenal hyperplasia, congenital, 197 adrenal malfunction, 190 aldosterone defects, 193 hydroxylase deficiencies, 191-194 lipid adrenal hyperplasia, 196 Infants, steroids in plasma and urine, 174-180 steroids, 177-181 assay by group methods, 177 corticosterone, 180 cortisol, 180 cortisone, 180 17-0x0 steroids, 174f. Injuries, multiple, time factor in, 23 Injury, control of metabolic response to, 32-41 ACTH, 34, 36, 40 aldosterone, 34 catecholamines, 35 neural pathways, 39 plasma cortisol, 34, 37 Injury, nutritional aspects, 23-26 nitrogen metabolism and, 24-25 transfusion in, 25-26

P Phenol dibromphthalein disulfonate, clearance of, 338 Phthalein dyes, 314 Post-shock, delayed mrtabolic response in, 1%22 Pregnanediol, urinary, method, 124

Progesterone excretion of, in urine, 66 methods for determination, 122-124 plasma concentration, 65 structure of, 61 Protein binding of BSP, 319 of cholephilic dyes, 316 of thyroxine, 406, 409

R Rose bengal analytic method, 360 retention test, 338 toxic effects of, 342

S Stein-Leventhal syndrome, 78 plasma concentration of testosterone, 79 secretion rate of testosterone, 79 Steroid determination in plasma and urine, 83ff. accuracy, 90 development of techniques, 83-99 gas, liquid, and combined chromatography, 86 operational and logical design in, 9W. precision, 85-99 sensitivity, 83-99 specificity, 85-90 systems analysis, 99 Steroid hormones, eliminative reactions, 70 Steroid metabolic pathways in utero, 156-170 amniotic fluid, 168 corticosterone, 165 cortisol, 164 estrogens, 159 fetal testis, steroids in, 159 fetus, enzyme activities in, 157 placental enzyme activities, 157 pregnenolone, 161 progesterone, 162 Steroid metabolism in early infancy, 141ff. methodology, 143-156 assays of groups of steroids, 14S149 assays of individual steroids, 151-156 hydrolysis of steroids, 147

462

SUBJECT INDEX

steroids in infant urine and umbilical blood, 143 Zimmerman chromogens in urine, 150 Steroid production in infancy control of, 197 corticotropin effect, 199 urinary excretion of steroids, 1% Steroids, of human blood and urine, m 7 binding of proteins, 63 conjugates of, 62 structures of, 61 Steroid sex hormones clinical considerations in determination of, 8 M disorders of metabolism of, 7-0 secretion and metabolism of, 67-76

T Testosterone concentration in plasma, 65, 104 excretion in urine, 66 methods for determination of, 126129 structure of, 61 Thyroid and calcitonin assays for calcitonh, 411 physiology of calcitonin, 412 role of calcitonin in man, 413 Thyroid hormone regulation, feedback loop in, 388 Thyroid-stimulating hormones (TSH), 393-404 bioassays in vitro, 397 in vivo, 396 levels in blood, 396 mechanism of action, 399

properties of, 393 radioimmunoassay, 398 synthesis and secretion, 394 in thyroid disease, 401 Thyroid stimulator, long acting (LATS), 401 characteristics of, 402 in disease, 403 Thyrotropin-releasing factor (TRF) assay in vitro, 391 in vivo, 391 characterization of, 390 isolation from hypothalamus, 390 physiology of, 391 Thyroxine-binding capacity drugs, effect of, 409 estrogens, effect of, 409 genetic abnormalities, 409 in nonthyroidal disease, 410 in thyroid disease, 410 Thyroxine-binding proteins characteristics of, 404 measurement of, 404-406 thyroid-binding globulin (TBG), 404 thyroid-binding prealbumin (TBPA) , 405 Thyroxine, free, 407 Thyroxine transport and turnover, 407409

U Umbilical cord blood cortisol and metabolites, 172 estrogens, 174 3P-hydroxy-A' steroids, 170 17-0x0 steroids, 172 progesterone and metabolites, 171 testosterone, 172