International Review of Cytology: A Survey of Cell Biology, Volume 52

INTERNATIONAL REVIEW OF CYTOLOGY VOLUME 52 ADVISORY EDITORS H. W. BEAMS ARNOLD MITTELMAN HOWARD A. BERN DONALD G...

Author: Geoffrey H Bourne | James Frederick Danielli

10 downloads 767 Views 24MB 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

INTERNATIONAL

REVIEW OF CYTOLOGY VOLUME 52

ADVISORY EDITORS

H. W. BEAMS

ARNOLD MITTELMAN

HOWARD A. BERN

DONALD G. MURPHY

W. BERNHARD

ROBERT G. E. MURRAY

GARY G. BORISY

ANDREAS OKSCHE

ROBERT W. BRIGGS

VLADIMIR R. PANTIC

STANLEY COHEN

DARRYL C. REANNEY

RENE COUTEAUX

LIONEL I. REBHUN

MARIE A. DI BERARDINO

JEAN-PAUL REVEL

N. B. EVERETT

WILFRED STEIN

CHARLES J. FLICKINGER

ELTON STUBBLEFIELD

M. NELLY GOLARZ DE BOURNE

HEWSON SWIFT

K. KUROSUMI

DENNIS L. TAYLOR

MARIAN0 LA VIA

TADASHI UTAKOJI

ROY WIDDUS GIUSEPPE MILLONIG ALEXANDER L. YUDIN

INTERNATIONAL

Review of Cytology E D I T E D BY

G. H. BOURNE

J. F. DANIELLI

Yerkes Regional Primate Research Center Emory University Atlanta, Georgia

Worcester Polytechnic Institute Worcester, Massachusetts

ASSISTANT EDITOR K. W. JEON De)Jartment o j z o o b g y University of Tennessee Knoxville, Tennessee

VOLUME 52

ACADEMIC PRESS New York

San Francisco London

A Subsidiary of Harcourt Brace Jooanooich, Publishers

1978

COPYRIGHT @ 1978, BY ACADEMIC PRESS, INC. ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS, INC.

111 Fifth Avenue, New York, New York 10003

United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. 24/28 Oval Road, London N W 1 7 D X

LIBRARY OF CONGRESS CATALOG CARD NUMBER:52-5203 ISBN 0-12-364352-X PRINTED IN THE UNITED STATES OF AMERICA

Contents LIST OF CONTRIBUTORS

.

.

.

.

.

.

.

.

.

.

.

vii

Cytophysiology of Thyroid Parafollicular Cells ELADIO A . NUNEZ AND MICHAEL D . GERSHON I . Introduction . . . . . . . . . . . . . . I1 . Histochemical Studies of Mammalian Thyroid Parafollicular Cells 111. Electron Microscope Studies of Mammalian Thyroid Parafollicular Cells . . . . . . . . . . . . . IV.5-HT . . . . . . . . . . . V. Tryptophyl Peptides . VI . Medullary Carcinoma of the Thyroid . . . . . . . . . . VII . Parafollicular Cells in Nonmalignant Human Diseases . VIII . Conclusion . . . . . . . . . . . . . References . . . . . . . . . . . . .

1 6 23 54 63 63 66 67 68

Cytophysiology of the Amphibian Thyroid Gland through Larval Development and Metamorphosis ELIANE REGARD

I . Introduction . . . . I1 . Iodide Pathways . . . 111. Thyroglobulin Biosynthesis IV. Hypophyseal Regulation . V. Conclusions . . . . References . . . .

. .

. .

. .

. .

. .

. .

. .

. .

. .

.

.

.

.

.

.

.

.

.

. .

. .

. .

. .

. .

. .

. .

. .

. .

.

.

.

.

.

.

.

.

.

.

81 87 100 104 113 115

The Macrophage as a Secretory Cell ROY C . PAGE. PHILIP DAVIES. I. I1 . I11. IV . V.

AND

A . C . ALLISON

Introduction . . . . . . . . . Regulation of Stem-Cell Growth . . . . Substances Affecting Fibroblast Growth and Activity Antimicrobial Substances . . . . . . Substances Activating or Regulating Host Defense against Bacteria, Viruses. and Tumor Cells . . . . . VI . Prostaglandins and Cyclic Nucleotides VII . Cytotoxic Substances . . . . . . VI I I . Hydrolytic Enzymes . . . . . . . References . . . . . . . . . V

.

.

.

. .

. .

. .

. .

. .

. . .

. . .

. . .

. . .

119 120 121 124

. .

125 132 134 137 151

vi

CONTENTS

Biogenesis of the Photochemical Apparatus TIMOTHY TREFFRY I . Introduction . . . . . . I1. Biogenesis of Chlorophyll . . . I11 Biogenesis of Chloroplast Membranes IV . Development of Photochemical Activity V. Concluding Remarks . . . . References . . . . . . Note Added in Proof . . . .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

.

.

.

.

.

.

.

.

.

159 160 173 185 190 191 196

I . Introduction . . . . . . . . . . . . I1. Methods . . . . . . . . . . . . 111. Characterization and Distribution of Extrusomes . . . . IV. Fine Structure. Extrusion Mechanism. Function. and Origin of the Different Types of Extrusomes . . . . . . . . V. Conclusions . . . . . . . . . . . . References . . . . . . . . . . . .

. . .

197 198 199

.

202 267 268

.

.

.

.

.

.

.

.

.

.

.

. . .

. . .

Extrusive Organelles in Protists KLAUS

HAUSMANN

.

.

Lectins JAY c. BROWNAND

I . Introduction . . I1 . Lectin Biochemistry

.

.

.

c. HUNT

RICHARD

.

.

.

.

.

.

.

.

.

.

.

.

. . I11 . Lectin-Induced Lymphocyte Mitogenesis . . . . . IV . Selective Agglutination of Transformed Cells . V. Interaction of Lectins with Cells Infected by Nononcogenic Viruses . . . . . VI . Interaction of Lectins with Developing Cells . . . . . . VII . Biochemistry of Cell Surface Lectin Receptors VIII . Lectin Toxicity . . . . . . . . . . . . IX . The Biological Role of Lectins . . . . . . . . . References . . . . . . . . . . . . .

277 278 292 297 319 322 326 330 333 336

SUBJECTINDEX . . . . CONTENTSOF PREVIOUSVOLUMES

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

351 355

.

.

.

. .

. .

. .

. .

List of Contributors Numbers in parentheses indicate the pages on which the authors' contributions begin.

A. C. ALLISON(119), Division of Cell Pathology, Clinical Research Centre, Northwick Park, Harrow, Middlesex, Engluncl JAY

C. BROWN(277), Department of Microbiology, University of Virginia School of Medicine, Charlottesville, Virginia

PHILIPDAVIES(119),The Merck Institute for Therapeutic Research, Rahway, New Jersey MICHAELD. GERSHON(I),Department of Anatomy, Columbia University, College of Physicians and Surgeons, New York, New York

KLAUS HAUSMANN(197), Lehrstuhl f u r Zellenlehre, Universitat Heidelberg, Heidelberg, West Germany RICHARDC . HUNT (277),Department of Biochemistry, Oxford University, Oxford, England ELADIOA. NUNEZ(l), Department of Anatomy, Columbia University, College of Physicians and Surgeons, New York, New York ROY C. PAGE(119),Department of Pathology and Periodontics and the Center for Research i n Oral Biology, University of Washington, Seattle, Washington

ELIANEREGARD(81),Universite' Paris-Sud, Laboratoire de BiologieVerte'bre's, Orsay, France TIMOTHYTREFFRY(159), Department of Biochemistry, The University, Sheffield, United Kingdom

vii

This Page Intentionally Left Blank

Cytophysiology of Thyroid Parafollicular Cells ELADIOA. NUNEZ AND MICHAEL D. GERSHON Department of Anatomy, Columbia University, College of Physicians and Surgeons, New York, New York

I. Introduction

.

.

.

.

.

.

11. Histochemical Studies of Mammalian Thyroid

111.

IV. V. VI.

VII. VIII.

.

.

. . . . . . Parafollicular Cells . A. Identification . . . . . . . B. Location and Morphology . . . . . C. Distribution and Number . . . . . D. Origin of Parafollicular Cells . . . . E. Histochemical Studies of Hibernators and . . . . . Antler Development . F. Congenital Osteopetrosis in Mice . . . G . Experimental Studies of Parafollicular Cells . Electron Microscope Studies of Mammalian Thyroid Parafollicular Cells . . . . . . . A. Normal Tissue . . . . . . . B. Experimental Studies . . . . . . . 5-HT . . . . . . . . . . Tryptophyl Peptides . . . . . . . Medullary Carcinoma of the Thyroid . . . A. Histochemical Evidence . . . . . B. Biochemical Evidence . . . . . . C. Electron Microscope Evidence . . . . Parafollicular Cells in Nonmalignant Human Diseases Conclusion . . . . . . . . . References . . . . . . . . .

.

1

. . . . .

6 6 12 12

. . .

16 17 18

.

23 23 53 54 63 63

. . . .

.

. . . . . .

15

64 64 65 66 67

68

I. Introduction

The thyroid gland of the mammal is a bilobed structure located at the base of the neck on either side of the trachea. The most important function of the thyroid gland is the synthesis, storage, and secretion into the blood of two iodinated amino acid hormones, L-thyroxine Thyroid hor(tetraiodo-L-thyronine) and 3,5,3'-triiodo-~-thyronine. mones (thyroxine and triiodothyronine) are required for normal growth and development and for normal metabolic activity. Thyroid hormones act by accelerating general and specific metabolic processes of the body, leading to an increase in oxygen consumption and heat production (Tata, 1964).The importance of the thyroid gland was 1

2

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

known as early as 1895, when Murray demonstrated the therapeutic nature of extracts of the thyroid glands of animals in the treatment of hypothyroidism. The crystallization of thyroxine was achieved in 1919 (Kendall, 1919), and its chemical structure was determined in 1926 (Harington, 1926).The existence of triiodothyronine was first reported by Gross and Pitt-Rivers in 1952. Studies by Oppenheimer and his associates (1972) and others (Hervas et al., 1976) strongly indicate that L-thyroxine may be a prohormone and that triiodothyronine is actually the only hormone of the thyroid gland that stimulates metabolism. The mammalian thyroid gland is also responsible for the elaboration, storage, and secretion of a second type of hormone, calcitonin, which lowers the calcium concentration of blood. Calcitonin was discovered by Copp and his associates in 1962. By perfusing the thyroidparathyroid complex of a dog with hypercalcemic blood, they induced the release of a factor, designated calcitonin, which lowered the concentration of calcium in the blood of a recipient animal. In 1963-1964, Hirsch et al. (1964) and Foster et al. (1964b) demonstrated that the thyroid gland, and not the parathyroid as first proposed by Copp et a2. (1962),was the source of calcitonin. Materials extracted from thyroid tissue of the dog, rat, rabbit, ox, pig, goat, sheep, and human have all been effective in lowering the concentration of calcium in the blood when injected into a recipient rat (Hirsch and Munson, 1969). Isolation and purification of the hormone has been carried out in four mammalian species, human, pig, cow, and sheep (Brewer and Ronan, 1969; Neheret al., 1968; Potts et al., 1968; Raulais et al., 1974).The hormone from all four species is a polypeptide with a molecular weight of about 3200, and it contains 32 amino acids in a single chain. Purified polypeptides are effective in lowering blood calcium in the rat. However, they differ in potency, and this difference has been attributed to the large variation in the amino acid sequence of the four molecules (Hirsch and Munson, 1969). Human calcitonin has been synthesized in the laboratory (Sieber et al., 1968). In lower vertebrates (birds, reptiles, amphibia, and fishes), calcitonin is found not in the thyroid gland but in the ultimobranchial body (Birov, 1971; Copp et al., 1967; Cutler et al., 1974; O’Dor et al., 1969; Tauber, 1967). The ultimobranchial body is a separate and distinct gland. Avian and fish calcitonin has been purified and, like the hormone in mammals, consists of 32 amino acids (Potts et al., 1970). The major action of calcitonin is on bone. It prevents the removal of calcium from the skeleton (Aliapoulios et al., 1966a; Hirsch, 1967; Wallach et al., 1967). Calcitonin exerts this action by inhibiting all forms of bone resorption, both osteoclasia (Friedman and Raisz, 1965)

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

3

and osteocytic osteolysis (Whalen et ul., 1975). If excess calcitonin reaches the circulation during development, the resultant inhibition of bone resorption results in the retention of cartilage (Whalen et ul., 1975).Indications are that calcitonin does not enhance bone formation or alter the number of osteoblasts (Rasmussen and Bordier, 1974). However, it also acts on the kidney, where it decreases tubular resorption of calcium (Ogata and Kimura, 1973). These effects of calcitonin have been made use of. Synthetic human calcitonin has been demonstrated to have a very positive therapeutic effect in the treatment of adult Paget’s disease (Doyle et ul., 1974; Haymovits et al., 1975) and familial hyperphosphatasemia (Honvith et al., 1976; Woodhouse et al., 1972), diseases characterized by excessive resorption of bone. However, calcitonin may also play a role in the mediation of some diseases. There are indications that it is involved in the pathogenesis of certain metabolic diseases such as pseudohypoparathyroidism (Aliapoulios and Rose, 1970; Aliapoulios et al., 1966b), hyperparathyroidism (Tashjian and Voekel, 1967), osteopetrosis in mice (Walker, 1966b), and parturient paresis in cattle (Young and Capen, 1968; Mayer et al., 1975). But the physiological significance of calcitonin in the mammal is still not clear, since neither humans nor rats become hypercalcemic following total thyroidectomy (Editorial, 1973). Despite this, suggestions that it has a role in the adult have been put forward. It has been proposed that calcitonin is physiologically active in protecting animals against hypercalcemia during the period of calcium absorption following a meal (Munson et d.,1974). Other suggestions are that calcitonin is required in liver regeneration (MacManus et al., 1975) and that it is important in protecting the skeleton from excessive resorption throughout life (MacIntyre, 1971). Microscopically, the mammalian thyroid gland is composed of a large number of closed, oval to spherical sacs referred to as follicles. Each follicle is lined by a single layer of cuboidal epithelial cells, called follicular cells, which surround a central lumen. The lumen is filled with a semifluid material which appears amorphous, the colloid. There are approximately 2 to 3 x lo7 follicles in the human thyroid gland, and the individual follicles vary in size from 50 to 100 pm (Rhodin, 1974).A basement membrane is clearly evident around each follicle, and a rich capillary network is present in the interfollicular tissue. Every follicular cell borders a capillary, and many sympathetic nerve fibers accompany the blood vessels (Rhodin, 1974). The lining follicular cells are responsible for the formation of thyroid hormone. They synthesize a complex glycoprotein, thyroglobulin, which they release into the follicular lumen by exocytosis (Nadler

4

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

et al., 1964). They also trap iodine and transport it from the blood to the follicular lumen, where thyroglobulin is iodinated. The iodinated thyroglobulin molecules are then stored in the lumen of the follicle in the form of colloid. Through the action of apical pseudopods (Ekholm and Smeds, 1966; Wetzel et d., 1965), or invagination of the apical plasma membrane (Nunez et al., 1972a; Seljelid et al., 1971), the stored iodinated thyroglobulin is reabsorbed by the lining follicular cells in the form of colloid droplets. The intracellular colloid droplets eventually fuse with lysosomes, and the thyroglobulin undergoes degradation. This results in the liberation of thyroid hormone and secretion into the blood of thyroxine and triiodothyronine (Wetzel et al., 1965). The secretion of thyroid hormone is under the control of a negative feedback mechanism regulated by the hypothalamus. The hypothalamus secretes thyrotropin-releasing hormone (TRH) which stimulates the secretion of thyrotropin or thyroid-stimulating hormone (TSH) from the anterior pituitary. TSH then acts on the follicular cells, increasing thyroid hormone release. In turn, the output of TRH, and thus TSH, is governed by the circulating level of thyroid hormone, a decrease in the latter increasing the discharge of TRH and a rise diminishing it (Bowers et al., 1970). At the ultrastructural level, follicular cells are characterized by many evenly spaced microvilli projecting into the colloid from the apical surface. Beneath the surface plasma membrane, a heterogeneous population of tiny vesicles is found. They are considered as representing transport vesicles carrying thyroglobulin or thyroperoxidase to the follicular lumen (Nadler et al., 1964; Novikoff et al., 1974).The nucleus is typically near the base of the cell but has no distinctive, features. The cytoplasm contains elaborate rough-surfaced endoplasmic reticulum which consists of many widely dilated profiles. A well-de+ veloped Golgi apparatus is usually found lying between the nucleus and the lumen of the follicle. Mitochondria are scattered throughout the cell. At least four types of rounded structures are present in follicular cells. They are: (1) colloid droplets which are particularly conspicuous after the administration of exogenous TSH, (2) membrane-limited dense granules which stain for acid phosphatase, (3)phagolysosomes which are bodies that arise from the fusion of colloid droplets and lysosomal dense granules, and (4) autophagic vacuoles. Apical pseudopods are not usually observed in normally active follicular cells. They are, however, especially numerous after TSH stimulation. A basement membrane borders the cell. For a more extensive discussion of the cytophysiology and ultrastructure of thyroid follicular cells, see the reviews by Fujita (1975) and Panti6 (1974).

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

5

In addition to follicular cells, the mammalian thyroid gland contains a second type of epithelial cell. These cells have been given various names but are generally referred to as parafollicular cells. In contrast to follicular cells, parafollicular cells are relatively uncommon in the thyroid gland. As early as 1876, they were reported to be morphologically distinct from follicular cells. At that time, Baber (1876) reported that in the canine thyroid gland a second epithelial cell type, described as parenchymatous, differed markedly in size and shape from the follicular cells among which it lay. Since then, in addition to being described as parafollicular by Nonidez in 1931, they have been called protoplasm-rich cells (Hurtle, 1894), ovoid cells (Bensley, 1914), interfollicular cells (Takagi, 1922), clear cells (Zechel, 1933), light cells (Stux et d., 1961),neurohormonal cells (Sunder-Plassmann, 1939), mitochondria-rich cells (Seecof, 1927), acidophilic cells (Hamperl, 1937), macrothyrocytes (Idelman, 1962; Kroon, 1958), argyrophilic cells (Sandritter and Klein, 1954), gray cells (Godwin, 1937), stem cells (Ponse, 1951), and most recently, C cells (Pearse, 1966a). Nonidez, in his excellent histochemical studies of the dog thyroid gland in the early 1930s (1931, 1932, 1933), provided the first evidence that parafollicular cells differ functionally from follicular cells. From observations of tissue impregnated by the silver nitrate method of Cajd, he noted that parafollicular cells, but not follicular cells, contained brown to black cytoplasmic granules. In the early 195Os, Sandritter and Klein (1954), Sandritter et d . (1956),and others (Dejardin, 1955; Dumont, 1956a; Gabe, 1959) demonstrated the existence of still other cytochemical differences between parafollicular and follicular cells in such species as the dog, rabbit, and guinea pig. In 1960 it was further reported that parafollicular cells of the thyroid gland of the rat, unlike follicular cells, do not respond to TSH administration (Isler et al., 1960). However, despite these findings and suggestions that parafollicular cells are secretory in nature (Altman, 1940; Arimitsu, 1937; Bensley, 1914; Ohkubo, 1934; Sato, 1959; Sugiyama, 1954), the general consensus during this period, 1876- 1962, was that parafollicular cells were nonsecretory (Grossi and Servide, 1961; Tashiro, 1962; Waller, 1960) and represented a stage in the life cycle of follicular cells (Saito, 1956; Voitkevitch, 1963), that they might be nerve cells (Sunder-Plassmann, 1939), or that they might be ectopic parathyroid cells (Getzowa, 1907). Even as late as 1954 some workers denied the existence of parafollicular cells as a separate cell type within the thyroid gland, stating that they were artifacts of the procedures used in their demonstration (Ehrenbrand, 1954; Ludwig, 1953,1954). Following the discovery of calcitonin in 1962 by Copp et al., there was immediate interest in the possibility that parafollicular cells

6

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

might be the cellular source of the polypeptide. Foster et al. (1964a) examined the canine thyroid gland after perfusion with a hypercalcemic solution, noted the apparent degranulation of a cell identified as a parafollicular cell, and suggested that parafollicular cells were responsible for the production of calcitonin. With the use of immunofluorescence methods, the specific localization of calcitonin in parafollicular cells was soon demonstrated by Bussolati and Pearse (1967) and Kracht and associates (1968a). This, then, was the beginning of a period of great interest in the morphology, function, and cytophysiology of the parafollicular cells of the mammalian thyroid gland, which has extended into the present. This interest has resulted in a large number of publications during the last dozen years. We have attempted to review these investigations and, on the basis of this review, to propose a role for the parafollicular cell in mammalian physiology. However, it is not the aim of this report to review the literature concerning the hormone calcitonin, since several excellent reviews have already been published on this subject (Haymovits and Rosen, 1972; Hirsch and Munson, 1969; Queener and Bell, 1975). 11. Histochemical Studies of Mammalian Thyroid

Parafollicular Cells A. IDENTIFICATION Parafollicular cells of the thyroid gland have been distinguished from follicular cells and other thyroid elements by various histochemical procedures. The techniques described in the following discussion have most often been used during the past decade to identify parafollicular cells. A short discussion of each of these methods follows, and criticisms of the techniques are listed in Table I .

1. Cujd Silver Nitrate As early as 1931, Nonidez demonstrated that the silver nitrate method of Cajd could distinguish parafollicular cells from follicular cells. More recently, further modifications of this method, used by Fitzgerald (1964), DeGrandi (1970), Sawicki and Bajko (1974), and Blahser and Kraus (1972), have simplified the procedure and demonstrated that silver nitrate impregnation is a reproducible and sensitive method for parafollicular cell identification in most, but not all, mammalian species. Silver nitrate has been used to demonstrate the presence of parafollicular cells in the thyroid glands of dogs, asses, sheep, rabbits, cats, rats, wolves, pigs, hamsters, and guinea pigs (Biddulph,

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

7

TABLE I HISTOCHEMICAL PROCEDURES EMPLOYEDTO IDENTIFYPARAFOLLICULAR CELLS

CRITICISM OF

Technique Cajal silver nitrate Masked metachromasia Colloidal iron Nonspecific esterases

a-Glycerophosphate dehydrogenase Acid phosphatase Amine precursor uptake Immunohistochemistry

Source of criticism Kracht et al. (1970); Lietz (1971); Sawicki and Bajko (1974); Velickq (1970) Blahoiova et al. (1974); DeGrandi (1970);Ljungberg (1970a); Maunder and Rost (1972); Roszkiewicz (1974a) Sawicki (1975); VelickL (1970) Mietkiewski et al. (1974); Mikhailov (1972); Sawicki (1975) Birov (1971); Lietz (1971); Mietkiewski et al. (1973a); Roszkiewicz (1974a); Sawicki (1975); Stachura (1971b); Zabel (1973) Beskid and Rosciszewska (1968); Mietkiewski et al. (1974); Mikhailov (1972); Roszkiewicz (1974a); Sawicki (1975) Lietz (1971); McMillan et al. (1974) Kalina et al. (1970)

1968; DeGrandi, 1970; Nonidez, 1931; Pearse, 1968c; Sawicki and Bajko, 1974; Van Dyke, 1945; Young et al., 1968). It is not known whether the argyrophilic properties of parafollicular cells are due to lipid interaction or to the interaction of the secretory granules with silver nitrate (Lietz, 1971; Roediger, 1973a). 2. Masked Metachromasia Masked metachromasia can be demonstrated in parafollicular cells by staining with a metachromatic basic dye such as toluidine blue, azure A, or methylene blue after a Feulgen-type hydrolysis of suitably fixed material (Lietz and Zippel, 1969a; Sawicki, 1971; Solcia and Sampietro, 1965a). Similar results have been obtained with lead hematoxylin (Sawicki, 1975). Masked metachromasia of parafollicular cells is believed to be due to the presence in their cytoplasm of secretory granules containing a polypeptide with a high concentration of acidic groups and a random-coil conformation (Bussolati et al., 1969c; Maunder and Rost, 1972; Solcia e t al., 1968a,b). The phenomenon of masked metachromasia was first noted in other polypeptide-containing cells by Manocchio in 1964. Solcia and Sampietro (1965a,b) were the first to develop this method for the demonstration of parafollicular cells. They stained the cells with toluidine blue or azure A after treatment with hot, dilute hydrochloric acid. According to them, the treat-

8

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

ment with hydrochloric acid eliminates diffuse metachromasia in the tissue and increases the metachromasia of the parafollicular cells. Recent modifications of the method include a fluorescence technique involving the acridine dye coriphosphine 0 (Bussolati et al., 1969~). Other workers have employed thiazine dyes such as gentian violet B (Guglielmone, 1971)and l79-dimethylmethy1eneblue (Petk6, 1974~). Reipforcement of masked metachromasia in parafollicular cells has been claimed in the treatment of tissue with potassium ferricyanide and ammonium heptamolybdate, supposedly to decrease dehydration of the metachromasia (Petk6, 1974b). The silver nitrate method has been reported to be intensely positive in parafollicular cells of the thyroid gland of the rat (Gittes et al., 1968; Kameda, 1968,1970). However, investigators using masked metachromasia have found that in the rat, in contrast to other species, this method shows cells only faintly (Roszkiewicz, 1974a). This may be due to the small number of secretory granules in the cytoplasm of rat parafollicular cells (Roszkiewicz, 1974a; Stachura, 1971a; Williams, 1966),but it may also indicate that in the rat masked metachromasia is simply less sensitive than other histochemical procedures. It is interesting to note that in certain mammals, such as hibernators (Olivereau, 1970; Olivereau and Fontaine, 1970), as well as some nonhibernators (Lietz and Donath, 1970a; Ljungberg, 1970a),parafollicular cells stain well with such dyes as toluidine blue, cresyl fast violet, and hematoxylin-eosin even without prior acid hydrolysis, whereas in most nonhibernating species the reaction of unhydrolyzed tissue is very weak (Lietz and Donath, 1970a; Hedhammer et al., 1974). 3. Colloidal Iron The reaction with colloidal iron has been reported to show a fairly high degree of selectivity in the demonstration of parafollicular cells in the thyroid glands of some mammals (Lietz and Zippel, 19f39a; Roszkiewicz, 1974a).According to Lietz and Zippel (1969a), colloidal iron is similar to masked metachromasia and indicates the presence of acidic groups in the cytoplasm of parafollicular cells. However, it has been reported that in the guinea pig colloidal iron is much less selective than masked metachromasia for parafollicular cells, since a strong reaction with colloidal iron occurs in both parafollicular cells and follicular cells (Sawicki, 1975). 4. Enzymic Reactions As early as 1955- 1956, cholinesterase (acetyl-or butyrylcholinesterase) was reported (Dejardin, 1955; Dumont, 1956b) to be a particularly good indicator of parafollicular cells of the mammalian thyroid

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

9

gland (Pepler and Pearse, 1967). For example, parafollicular cells exhibit a strong cholinesterase reaction in pigs, dogs, rabbits, guinea pigs, and rats (Carvalheira and Pearse, 1967a,b; Gauguin et al., 1973; Mietkiewski et al., 1973a; Roediger, 1973b; Roszkiewicz, 1974a).The role of cholinesterase in parafollicular cell activity is not clear (Roszkiewicz, 1974a), however, its involvement in the synthesis of the membranes surrounding parafollicular cell granules and in the secretion of granules has been postulated (Pearse and Welsch, 1969; Stachura, 1971a). Pearse (1966a) and others (Mietkiewski et al., 1973a; Sawicki, 1975) used reactions for nonspecific esterases and acid phosphatase as selective markers for parafollicular cells. Unfortunately, it has also been reported that both esterase and acid phosphatase are much less selective in identifying parafollicular cells than other histochemical methods (Roszkiewicz, 1974a; Sawicki, 1975). The very strongly positive reaction with a-glycerophosphate dehydrogenase noted by Foster et al. (1964a) in parafollicular cells of the dog thyroid was subsequently regarded by Pearse (1966a) and others (Beskid and Rosciszewska, 1968; Roediger, 197313; Sawicki, 1975) as a good parafollicular cell marker. However, studies on such rodents as rats (Birov, 1971; Stachura, 1971b; Roszkiewicz, 1974a), rabbits (Mietkiewski et al., 1973a), and guinea pigs (Birov, 1971) have indicated that this reaction is useless for distinguishing parafollicular cells from follicular cells. Reactions with other enzymes such as adenosine triphosphate (Ansari, 1967; Lietz and Zippel, 1969b; Mikhailov, 1972; Rother, 1970) and alkaline phosphatase (Mikhailov, 1972), as well as several other oxidoreductases (Beskid et al., 1968; Mietkiewski et al., 1974; Mikhailov, 1972; Pearse, 1969; Rother, 1970), have also been studied. However, their reliability and reproducibility as specific indicators of parafollicular cells have yet to be established. Parafollicular cells have been reported to contain high a-glycerophosphate menadione reductase activity (Pearse, 1969; Rojo-Ortega et al., 1971), which is considered by some (Pearse, 1969) to be a manifestation of high lipid turnover. This implies a possible involvement of phospholipids in hormone secretion. In summary, it seems clear that enzymic markers are not noncontroversial indicators of parafollicular cells. Of the available methods, cholinesterase seems clearly the most reliable, even if its significance to the economy of the parafollicular cell remains to be established.

5. Amine Precursor Uptake The ability of mammalian parafollicular cells to take up and convert exogenous L-5-hydroxytryptophan (5-HTP) to serotonin (5hydroxy-

10

ELADIO A. NUNEZ AND MICHAEL

D.

GERSHON

tryptamine, 5-HT), or ~-3,4-dihydroxyphenylalanine (L-dopa)to dopamine (DA), has been employed by many workers to identify parafollicular cells in the thyroid gland (Dahlstrom and Ericson, 1972; Englund et al., 1972; Falck et al., 1964; Gershon and Nunez, 1970; Larson et al., 1966; Owman and Sundler, 1968; Pearse, 1966a,b; Ritzen et al., 1965; Tjalve and Slanina, 1971). Both of these amino acids, 5-HTP and L-dopa, are decarboxylated in the cytoplasm of parafollicular cells to give the corresponding amines, 5-HT and DA, respectively (Ericson, 1972b; HBkanson et al., 1971a; Owman and Sundler, 1968). The amines are then stored in the cytoplasm and can be converted into fluorescent compounds in tissue that has been freeze-dried and exposed to formaldehyde vapor of appropriate relative humidity. This permits detection of the amines in parafollicular cells by fluorescence microscopy (Gershon and Nunez, 1970; Owman and Sundler, 1968; Pearse, 1966a). Ritzen et al. (1965) and Gershon and Ross (1966a,b) documented this process by injecting tritium-labeled L-dopa or 5-HTP and localizing labeled amine in parafollicular cells by autoradiography. Thus, by using either the histochemical fluorescence method or light microscope autoradiography, the distribution of parafollicular cells in the mammalian thyroid gland has been shown in many species including the dog (Gershonet al., 1971; Pearse, 1966b),mouse (Gershon and ROSS, 196613; Larson et al., 1966; Tjalve and Slanina, 1971), rat (Dahlstrom and Ericson, 1972), and bat (Gershon and Nunez, 1970). Pearse ( 1 9 6 6 ~1968a) ~ has classified cells that share the ability to produce a low-molecular-weight protein or polypeptide and the ability to take up and decarboxylate exogenous 5-HTP or L-dopa in a group referred to as the amine precursor uptake and decarboxylation (APUD) series of endocrine cells. Thyroid parafollicular cells, as well as the cells of the ultimobranchial body, are considered to belong to this APUD series. Other cell types in the series include pituitary corticotrophs and melanotrophs, pancreatic endocrine cells, and gastrointestinal enteroendocrine cells (Pearse, 1966a). All these, like parafollicular cells, are likely to have as at least one of their functions the synthesis and secretion of polypeptides. 6. Immunohistochemical Methods

Specific demonstration of calcitonin in parafollicular cells has been obtained by the use of fluorescein-labeled antibody to calcitonin (Bussolati and Pearse, 1967; Bussolati et al., 1969a; Kalina et al., 1970; Kracht et al., 1968a,b) and unlabeled antibody-immunoperoxidase bridge techniques (LiVolsi et al., 1973; McMillan et al., 1974; Peng, 1975; Peng et al., 1975; Tashjian et al., 1974; Wolfe and Tashjian,

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

11

1974).Moreover, the use of these methods for the identification of calcitonin-containing parafollicular cells has clearly shown that parafollicular cells can easily be distinguished from follicular cells in the thyroid gland of the human and other species. A comparison of routine histochemical methods, such as those described above, with the immunoperoxidase technique shows that the immunohistochemical method is more sensitive and more specific in the demonstration of parafollicular cells in the normal thyroid gland (DeLellis et al., 1974). 7. Miscellaneous Methods The staining of parafollicular cells with strong dyes such as pseudoisocyanin and aldehyde fuchsin after oxidation of the tissue with performic acid (Dorrenhaus et al., 1971; Lietz and Zippel, 1969b) has been claimed to be specific. However, these dyes have not been widely used, and thus it is difficult to judge their reliability and potential in the identification of parafollicular cells. Also, the use of other histochemical methods, such as histophotometry, to determine the calcitonin content and granulation of parafollicular cells has yet to be proved reproducible and reliable (Lietz et al., 1969; Rouais et al., 1973). Foster et al. (1964a) and others (Harcourt-Webster and Stott, 1966; Seecof, 1927) referred to parafollicular cells as mitochondria-rich cells and hoped to distinguish them by their mitochondria1 content. Electron microscope studies (see Section III,A,l) and histochemical studies (Mikhailov, 1972; Pearse, 1968a) of thyroid tissue have clearly shown that parafollicular cells are not particularly rich in mitochondria. In the human thyroid gland, however, there are cells, referred to as Ashkinay, oxyphilic, Hurtle, or oncocytic cells, which are in fact rich in mitochondria as revealed electron microscopically and histochemically (Hamperl, 1962; Raikhlin and Smirnova, 1970; Roth et al., 1962; Tremblay and Pearse, 1960). These cells are clearly different from parafollicular cells (Roediger, 1975). Their function in the human thyroid gland is not known, but it has been thought by some workers that they are dystrophic persisting follicular cells (Friedman, 1949; Roediger, 1975). Others, however, have suggested, on the basis of enzyme histochemical studies, that they have an independent function (Kraevski et al., 1974). In summary, many methods have been applied for the demonstration of parafollicular cells. Of these, the best in terms of sensitivity and specificity appear to be amine precursor uptake and immunohistochemistry. The Cajhl silver nitrate method, masked metachromasia, and cholinesterase histochemistry seem to be acceptable second

12

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

choices if circumstances do not permit use of the more specific procedures. B. LOCATIONAND MORPHOLOGY Parafollicular cells are nonpolarized cells and occupy as many as three positions in relation to follicles. They are found in intrafollicular (between follicular cells), parafollicular (between the follicular cells and basement membrane of the follicle), and interfollicular (between follicles but surrounded by the basement membrane) locations in the thyroid glands of mammals. The interstitial position of parafollicular cells reported in some light microscope studies (Carvalheira and Pearse, 1967a; Gabe and Martoja, 1969; Salzer, 1971), namely, outside the basement membrane, has not been upheld by electron microscope examination of thyroid glands (Lietz, 1971). In some species, such as the human (Pearse, 1966a), cat (Kameda, 1971b), and dog (Kameda, 1971a; Roediger, 1973b; Teitelbaum et al., 1970), groups of parafollicular cells aggregate, forming parafollicular cell follicles. Parafollicular cells in ordinary thyroid follicles are easily distinguished from follicular cells by their greater size and their position away from the colloid. In the normal rat, for example, the cuboidal follicular cells are approximately 5-7 pm in width and about 2 pm in height and are located in direct contact with the colloid. However, parafollicular cells are usually Iocated in the basal region of the follicle, wedged between follicular cells away from the colloid. Parafollicular cells usually appear oval and are two to three times as large as follicular cells. Histochemically, parafollicular cells do not react as intensely as follicular cells with cationic dyes. Thus they usually exhibit a very fine cytoplasmic texture. In contrast to follicular cells, parafollicular cells do not concentrate radioactive iodide (Azzali, 1968). Histochemically, two types of parafollicular cells, a clear cell and a dark cell, have been described in various species (Jordan et d.,1973; Kiyama et al., 1968; Saito and Shibata, 1957; Sugiyama, 1950).Whether these types correspond to a real difference in cell physiology is not known.

c.

DISTFUBUTIONAND NUMBER Parafollicular cells tend to be present in small numbers and to be unevenly distributed in the thyroid glands of most mammals. In the human, where numerous histochemical studies have been carried out (Beskid et al., 1970; Hachmeister et al., 1969; PagBs, 1956; Rosciszewska and Beskid, 1969; Solcia et al., 1970), parafollicular cells are extremely few in number, comprising about 0.1% of the epithelial cell

CYTOPHYSIOLOCY OF THYROID PARAFOLLICULAR CELLS

13

mass of the normal adult gland (Sugiyama, 1967; Tashjian et al., 1974). Although several studies have reported a uniform distribution of parafollicular cells in the adult thyroid gland of the human (Englund et al., 1972; Kalina et al., 1970),the overwhelming histochemical (McMillan et al., 1974; Solcia et al., 1970; Tashjian et al., 1974; Wolfe et al., 1974) and biochemical (Tashjian et al., 1974; Wolfe and Tashjian, 1974; Wolfe et al., 1974) evidence clearly demonstrates that parafollicular cells and calcitonin are concentrated in the central region of the middle third of each lobe. The number of parafollicular cells in the adult thyroid in most other mammals has also been found to be relatively small. Parafollicular cells have been estimated as comprising 1% of the epithelial cell mass in mice (Marks, 1969; Marks and Walker, 1969; Walker, 1966a), 1% in the pig (Young et al., 1968), 2% in the rabbit (Lupulescu, 1972), 13% in the guinea pig (Sawicki, 1975), and variously as 1% (Stux et al., 1961),2% (Rohr and Hasler, 1968), or 6% (Peng, 1975) in the rat. However, a much higher figure, 30 parafollicular cells for every 100 follicular cells, has been reported but not replicated (Kameda, 1968). Thus, despite intraspecies differences, sampling variation, functional differences in parafollicular cell activity, and the relative nonspecificity of some of the histochemical procedures used to estimate parafollicular cell number, there is generally good agreement as to the small number of parafollicular cells in the majority of mammalian thyroid glands. Parafollicular cells appear to be more numerous in the thyroid glands of rodents than in those of primates (Liek and Zippel, 1969a; Solcia et al., 1970). There is some species variation in parafollicular cell distribution. In the rat (Kameda, 1971b; Peng et al., 1975; Petk6, 1974a; Stux et al., 1961), mouse (Kameda, 1971b; Walker, 1966a; Welsch et al., 1969), dog (Hedhammar et al., 1974), guinea pig (Welsch et al., 1969), and rabbit (Kameda, 1971b; Velickf, 1971a), as in the human, the greatest number of parafollicular cells is found in the middle of the gland, whereas in the cat and lion (Kameda, 1971b) they are restricted to the upper two-thirds of the gland. In the tree shrew, parafollicular cells are confined to a zone surrounding the internal parathyroids (Welsch et al., 1969). Although parafollicular cells account for only a relatively small percentage of the thyroid cell population in most mammals, they are widespread in the thyroid glands of hibernators (Biddulph and Maibenco, 1972; Gabe and Martoja, 1969; Nunez et d.,196713; Olivereau, 1970; Velickf and Titlback, 1972). Moreover, under appropriate experimental conditions the number of parafollicular cells may increase

14

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

and become a much higher percentage of the total number of thyroid cells than normal. For example, exposure to cold has been reported to increase mitosis of parafollicular cells (Kaissling and Bucher, 1973, 1974) and induce hyperplasia of parafollicular cells in the thyroid gland of the rat (Krstii. and Bucher, 1970, 1971). It has also been reported that parafollicular cells are more numerous in transplanted thyroid glands than in glands left in situ (Pantii. et al., 1970). The proportion of parafollicular cells in the thyroid gland appears to change during the course of development. More parafollicular cells have been found in adult glands than in the glands of young guinea pigs (Sawicki, 1975; Solcia and Sampietro, 1968), newborn (Petk6 et al., 1976) and young rats (Lietz, 1971; Penget al., 1976; Petk6, 1974a), and newborn rabbits (Solcia and Sampietro, 1968). It has also been reported that in rats of advanced age there may be hyperplasia of parafollicular cells, changing with time into so-called gamma tumors (Lietz, 1971; Lietz et al., 1969). In contrast to the situation in these lower animals, the fetal and neonatal human thyroid contains not a smaller but a much higher proportion of parafollicular cells than the adult gland (Pearse, 1968c; Wolfe et al., 1974, 1975). This explains biochemical data showing that the neonatal human thyroid gland has 10 times the concentration of calcitonin in the adult human gland (Wolfe et al., 1974,1975).Again, the different results observed in animals and humans may be due either to species differences or to the nonspecificity of some of the histochemical methods used. Argyrophilia, metachromasia, and cholinesterase staining were used in the animal studies, however, the absence of argyrophilia and metachromask (Petk6, 1974a; Petk6 et al., 1976) and low cholinesterase activity (Welsch, 1971) in the parafollicular cells of young animals may be due to functional differences in parafollicular cell properties rather than to fewer cells. In this regard, there has been considerable recent interest in the possibility that calcitonin is physiologically important in the newborn. The importance of parafollicular cells during early development is discussed in greater detail in Section III,A,4. Histochemical, but not immunohistochemical, studies have reported the existence of cells like parafollicular cells in the parathyroid gland and in the thymus (Carvalheira and Pearse, 1967a; Dubois and Dumont, 1966; Kameda, 1971b; Kracht et al., 1970; Rasmussen, 1969; Solcia and Sampietro, 1968). However, more recent studies employing iinmunohistochemical methods have failed to demonstrate the existence of calcitonin-containing cells in the parathyroid gland or in any organ outside the thyroid (Kalina et al., 1970; LiVolsi et al., 1973; McMillan et al., 1974; Peng et al., 1975; Wolfe et al., 1974).

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

D.

ORIGIN OF PARAFOLLICULAR

15

CELLS

It has long been recognized that the thyroid gland of the mammal has a double origin (Fisher and Dussault, 1974; Wollman and Nkve, 1971). The embryonic thyroid is derived from a midline outpouching of the endoderm of the floor of the primitive buccal cavity, which migrates caudally to form the isthmus and the lobes. In most mammals there are also lateral anlagen of the thyroid gland derived from the ultimobranchial portions of the third (fourth in the rat) pharyngeal pouch; these develop at about the time the median anlagen appears (Fisher and Dussault, 1974; Klapper, 1946; Rogers, 1927). During development the ultimobranchial body is incorporated into each of the lobes of the growing gland and may contribute 10- 15%of the thyroid tissue (Fisher and Dussault, 1974; Rogers, 1929; Sugiyama, 1942; Zuckerkandl, 1903). There is now strong evidence that the ultimobranchial body is the source of parafollicular cells. Other suggestions as to the possible origin of parafollicular cells have been proposed in the literature. They have been considered derivations, or transformations, from follicular cells (Allara, 1954; Calvert and Isler, 1970; Gabe, 1961; Ito et al., 1963; Mamina, 1962; Raymond, 1932; Sarkar and Isler, 1963; Thomas, 1934; Voitkevitch, 1963; Yoshimura et al., 1962; Young and Leblond, 1963), thyroglossal duct cells (Dumont, 1956b, 1958), or argyrophilic connective tissue cells which migrate in between the follicles (Dempsey, 1954), or Feyrter system endocrine cells (Ponse, 1951). The ultimobranchial origin of parafollicular cells was first proposed by Godwin in 1937. He found that parafollicular cells in the dog are distributed mainly in the part of the thyroid supposedly developed from the ultimobranchial body. Subsequent histological studies in such species as the mouse (Sato et al., 1966), rat (Rother, 1970; Van Dyke, 1944), sheep (Jordan et al., 1973; Van Dyke, 1945), human (Lietz et al., 1971; Sugiyama, 1969; Sugiyamaet al., 1969), rabbit (Dumont, 1956a), and hamster (Sato, 1959; Takagi et al., 1974) were in agreement with Godwin’s conclusion. However, the strongest confirmatory evidence has been the histochemical studies of Pearse and Carvalheira (1967), Carvalheira and Pearse (1968), Fontaine (1974),and Kirkeby et al. (1973).Using as tracers amine precursor uptake and cholinesterase staining, they followed the migration of parafollicular cells from the ultimobranchial body into the lobes of the growing thyroid. Further proof of a biological basis for separate origins of the parafollicular and follicular cells of the thyroid gland has come from studies of the ultimobranchial body of lower verte-

16

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

brates. In mammals, the ultimobranchial body, as just described, is completely incorporated into the thyroid gland during development. In contrast, in lower vertebrates, it reinaiiis distinct and separate from the thyroid gland. It has clearly been established that the ultimobranchial body of fish, amphibians, and birds is the source of calcitonin in these species (Copp et aZ., 1967; Cutler et d . , 1974; O’Dor et c i l . , 1969; Tauber, 1967). LeDouarin (1974), LeDouarin and LeLiBvre (1971), LeDouarin et cil. (1974), and Polak et nZ. (1974), using both histochemical and iniiiiunoliistoclieiiiical nietliods, demonstrated that the avian glandular cells of the ultiinobrancliial body have a neuroectodennal origin. A siinilar conclusion has been reached for inaininaliaii ultiinobranchial cells by Pearse and Polak (1971). If this is correct, the parafollicular cells of the thyroid gland join the ultiinobranchial glandular cells, pheochroniocytes of the adrenal medulla, and carotid body cells in having a coninion iieuroectoderinal precursor in the neural crest (Pearse and Takor, 1976). This supports the view that all peptide-secreting endocrine cells may be of neural crest origin (PagBs, 1974; Pearse and Polak, 1974; Pearse and Takor, 1976; Weichert, 1970). It may help explain why at any given time each of these cells has the capacity to develop into a wide variety of peptide-producing endocrine tumors (Birkenhager et aZ., 1976; Pearse, 1969; Tischler et nl., 1976; Weichert, 1970). E. HISTOCHEMICAL STUDIESOF HIBERNATORS AND ANTLER DEVELOPMENT It is well established tliat inaininalian hibernators develop sequentially hypocalceinia and hypercalceinia (Azzali, 1967, 1968; Biorck et ul., 1956; Hayinovits et al., 1976; Riedesel, 1957), in addition to loss of bone (Haller and Ziininy, 1976; Kayser and Frank, 1963; Whalen et [ i l . , 1972), during the hibernating phase of their yearly life cycle. Because of a possible association between parafollicular cell activity and such changes in calcium metabolism, a considerable number of studies has been done on the subject. Most of thein were carried out at the ultrastructural level and are reported in Section 111,A75. Histochemical studies, however, have also provided valuable infonnatioii regarding the relationship between parafollicular cells and the cycle of hibernation. For example, in addition to the demonstration that the thyroid glands of hibernators contain large numbers of parafollicular cells (Bensley, 1914; Gabe and Martoja, 1969; Olivereau, 1970), it has been shown that in the hamster and garden dormouse

CYTOPHYSIOLOCY OF THYROID PARAFOLLICULAR CELLS

17

(Eliomys quercinus) parafollicular cells exhibit striking cyclic changes. Prior to hibernation, parafollicular cells are abundant and large, but during the course of the hibernating period they decrease in number and size (Gabe and Martoja, 1969; Ouvrard-Pascaud et al., 1976) and undergo cytoplasmic vacuolization (Olivereau, 1970; Olivereau and Fontaine, 1970). Olivereau (1970), on the basis of histochemical staining, states that degranulation of parafollicular cells occurs during hibernation. At arousal, however, regranulation takes place (Olivereau, 1970), and parafollicular cells again become abundant (Gabe and Martoja, 1969; Ouvrard-Pascaud et al., 1976). It has also been reported that parafollicular cells are abundant in the thyroid gland of the deer (Pantic, 1967).In a study of the thyroid gland of Belje deers and roebucks at different ages and during all seasons, it has been suggested, from observations of cyclic histological changes, that parafollicular cells may have an active role in deer antler development, and also in the quality of the antler (Pantic, 1967; Panti6 and Stosic, 1966). It was also suggested that calcitonin facilitates calcium deposition in antlers. However, no physiological or biochemical data as yet exist to support this hypothesis.

F. CONGENITAL OSTEOPETROSIS IN MICE Osteopetrosis is a disorder of the skeletal system in which the absence of bony remodeling results in the accumulation of heavy but fragile and abnormal bone. This results in total or partial obliteration of marrow cavities. In mice, the condition is inherited as an autosomal recessive; three inbred strains, gray-lethal, microophthalmic, and osteosclerotic, develop the disease (Walker, 1972). Such mice have been shown to have 4 to 10 times as many parafollicular cells as their normal littermates (Marks, 1969; Marks and Walker, 1969). In these osteopetrotic animals the repeated injection of large amounts of parathormone fails to raise the low level of blood calcium but increases still further the size of the parafollicular cell population (Walker, 1966b; Marks and Walker, 1969). It has been postulated that the influence of excessive amounts of exogenous parathormone is consistently offset by the effect of calcitonin of endogenous (parafollicular cell) origin (Walker, 1966b). In experimentally induced osteopetrosis, there is also hyperplasia of the parafollicular cells (Marks, 1969). Moreover, experimental osteopetrosis does not develop in thyroidectomized mice in response to treatment with parathormone (Marks, 1969).Thus the increased number of parafollicular cells is not a secondary effect due to parathormone secretion. On the basis of such evidence it has

18

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

been proposed that congenital osteopetrosis in mice arises as the result of an overproduction of calcitonin. A second hypothesis is that parafollicular cells are the source of yet another hormone, an osteoblast-stimulating factor (Marks, 1969; Walker, 1966a). STUDIES OF PARAFOLLICULAR CELLS G. EXPERIMENTAL

1. Calcium

It has been well established biochemically that hypercalcemia (induced by the administration of calcium, parathormone, or vitamin D) is the natural stimulus for secretion of calcitonin (Hirsch and Munson, 1969). Histologically, the effect of hypercalcemia on parafollicular cells has also been well documented. Acute hypercalcemia causes hypertrophy (Kracht et al., 1968a; Pearse, 1970; Roszkiewicz, 1974a),diminishes the argyrophilia, metachromasia, and enzymic staining (Ansari, 1967; Mietkiewski et al., 197313; Roszkiewicz, 1974a, Swarup and Das, 1974; Velick?, 1971b), and increases the mitotic activity (Kameda, 1970; Kiyama et al., 1968; Mietkiewski et al., 1973b; Swarup and Das, 1974) and nuclear size (Wuttke et al., 1971; Zabel and Peil, 1974) of parafollicular cells in such species as the rat, guinea pig, and mongoose. Chronic hypercalcemia lasting from 2 to 8 weeks increases acid phosphatase activity (Ansari, 1967; Beskid and Rosciszewska, 1968), diminishes mitotic activity (Lietz et al., 1969), induces a proliferation of parafollicular cells (Zabel, 1976), and in some cases may even lead to the formation of adenoma-like nodules of these cells or to their degeneration (Kameda, 1968, 1974b; Roszkiewicz, 1974a,b; Roszkiewicz and Roszkiewicz, 1974). The effect of hypocalcemia on parafollicular cells, however, is less well established. Gittes et al. (1968) have reported that the calcitonin content of the thyroids of rats exposed to prolonged hypocalcemia is markedly decreased. Several histological studies have demonstrated the effect of hypocalcemia on parafollicular cells. It has been reported that increased mitotic activity (Biddulph and Maibenco, 1972; Lietz and Donath, 1970a; Lietz et al., 1969), increased argyrophilia (Ansari, 1967), increased acid phosphatase (Ansari, 1967), and hypertrophy and hyperplasia of parafollicular cells occurs during hypocalcemia (Biddulph and Maibenco, 1972). Beskid and Rosciszewska (1968) have reported that hypocalcemia brings about intense vacuolization of parafollicular cells. If this is so, hypocalcemia may inhibit the release of calcitonin from parafollicular cells. However, there are no chemical data to confirm this.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

2. Loio-locliiie-lnrlzccedGoiier

r iiir l

19

Toxic Goiter

Large numbers of parafollicular cells have been found in goiters of rats fed a low-iodine diet for almost a year (Axelrod and Leblond, 1955; Isler, 1959; Peng et ul., 1975). Such tissue has also been shown to contain a much higher concentration of calcitonin than nonnal rat thyroid tissue (Peng et d . , 1975). It has also been demonstrated that human toxic goiters (Englund et n l . , 1972; Nilson, 1972) contain a significantly larger number of parafollicular cells than tissue from patients with atoxic goiters (Englund et ( i l . , 1972). Since thyrotoxicosis has been associated with a tendency toward hypercalcemia and osteoporosis in patients with tlie disease (Adams et al., 1967), it has been suggested that the large numbers of parafollicular cells in goitrous tissue reflect an increased demand placed on the thyroid of these individuals for tlie production of calcitonin (Englund et ul., 1972). Chronic hypercalcemia induced in animals may represent an analogous situation.

3. Mngnesiuni Calcium is recognized as the principal regulator of parathonnone (Habener and Potts, 1976) and calcitonin (Hirsch and Munson, 1969) secretion. Biochemically, it has been shown that hypermagnesemia influences parathormone (Habener and Potts, 1976) and calcitonin (Gitelman et al., 1968) secretion a s well, but is substantially less potent than calcium. Histologically, however, hypomagnesemia, and not hyperinagnesemia, has been found to affect the morphology of parafollicular cells. Hypennagneseinia has been shown to cause marked hypertrophy and hyperplasia of parafollicular cells in the thyroid gland of the rat and dog (Rojo-Ortegaet nl., 1971; Stachura and Pearse, 1970).In addition, hypoinagneseinia also causes a reduction in cholinesterase staining, dopamine storage, and masked metachromasia (Rojo-Ortega et al., 1971; Stachura and Pearse, 1970) and increases aglycerophosphate menadione reductase activity (Rojo-Ortega et d., 1971) in parafollicular cells. The reason for the difference between the biochemical findings and the histochelnical results is not clear at this time. However, the possibility tliat the magnesium ion may be directly involved in regulation of the secretory activity of parafollicular cells as postulated by Stachura and Pearse (1970) should not be overlooked at this time. In addition to calcium and magnesium ions, glucagon (Avioli et al., 1969), catecholamines (Broulik and Pacovskq, 1973; Bruce and Care, 1969; Fisher et al., 1973),pentagastrin (Cooper et al., 1971), pancreo-

20

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

zymin (Cooper et al., 1972), dibutyryl cAMP (Feinblatt and Raisz, 1971), and prostaglandins E2(Roos et al., 1974) all increase the secretion of calcitonin by the thyroid. The significance of this action being shared by so many diverse agents is still unclear. However, it has been proposed that many of these activators influence parafollicular cells through the activation of beta adrenoceptors (Broulik and Pacovskq, 1973). This would in turn activate adenylate cyclase, ultimately resulting in an increased level of cAMP and initiating secretion of calcitonin from the thyroid parafollicular cells. cAMP would thus be seen as a common pathway for stimulus-secretion coupling. 4. Endocrine Associations

a. Pituitary. The relationship between parafollicular cells of the thyroid gland and other endocrine glands is still enigmatic. Published reports are often contradictory. In the case of the pituitary gland, hypophysectomy (Chan and Bhlanger, 1968; Pantii. and Kalusevic, 1974; Sarkar, 1961; Thompson et al., 1962; Yoshimura et al., 1962) and the administration of growth hormone (Sarkar, 1961; Thompson et al., 1962) have both been reported to increase the number of parafollicular cells in the thyroid gland of the rat by as much as a factor of 3. However, hypophysectomy in dogs (Kameda, 197413) and rats (Matsuzawa, 1966; Saito and Shibata, 1957), and the injection of TSH (Isler et al., 1960; Saito and Shibata, 1957; Sarkar, 1961; Stux et al., 1961),ACTH, prolactin, oxytocin, or vasopressin (Sarkar, 1961) into rats, have been reported not to alter the histological picture of parafollicular cells. Biochemically, TSH seems not to change the calcitonin concentration in the thyroid gland (Aliapoulios and Munson, 1965).However, Yoshimura et al. (1962) have reported that parafollicular cells disappear from the thyroid following the persistent administration of TSH. In addition, Less (1970) has reported that higher TSH levels are associated with increased numbers of parafollicular cells and that lower TSH levels are associated with decreased numbers of parafollicular cells. Certainly no conclusions are justified at this time concerning what effect, if any, pituitary hormones have on parafollicular cells or the secretion of their hormone. b. Thyroid. It has been reported that the long-term administration of calcitonin to rabbits markedly decreases radioactive iodide (1311) uptake into the thyroid gland and plasma (Lupulescu and Stebner, 1974) and induces hyperplastic goiter (Lupulescu, 1972).It is thus possible that a functional interrelationship may exist between follicular and parafollicular cells in the thyroid gland. However, biochemically and

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

21

histologically, the effect of thyroid hormones (thyroxine and triiodothyronine) on the activity of parafollicular cells has not been extensively studied, and again contradictory results are found in the literature. Sarkar (1961) reported that treatment with thyroxine doubles the number of parafollicular cells in the thyroid gland of the rat, whereas Yasumura et al. (1967) observed a significant decrease in calcitonin in the thyroid gland of the rat following the administration of thyroxine. Wuttke et al. (1971) observed that triiodothyronine appears to reduce the size of the nuclei of parafollicular cells. Yoshimura and associates (1972) have reported a method for the isolation of follicular cells and parafollicular cells from rats pretreated with thyroxine. They claim that the addition of calcium or TSH to cultures of isolated parafollicular cells induces the formation of typical new follicles. Thus they feel they have induced parafollicular cells to transform into typical follicular cells. Yoshimura et al. (1962) have described such a transformation in viva They reported that, following the long-term administration of thyroxine, parafollicular cells give rise to new typical follicles which replace atrophic old follicles. This transformation of parafollicular cells into typical follicles has not been confirmed by other investigators. The emergence of typical follicles from cultures thought to contain only parafollicular cells could have been due to selection by the experimental conditions of residual follicular cells that were always present but had gone undetected prior to the addition of calcium or TSH. Nevertheless, this suggestion that parafollicular cells serve as precursors of typical follicles should be ruled in or out b y future research. c. Vitamin D.Parafollicular cells have also been implicated in the distribution of vitamin D. Using the technique of whole-body autoradiography, Dencker and Tjalve (1973) found that there was an accumulation of radioactivity in parafollicular cells 8 hours after the injection of radioactive vitamin D,. They suggested that this localization represents a link in the complicated regulation of calcium homeostasis, however, the significance of their observation has not yet been determined. d. Pined Gland. There may also be an association between parafollicular cells and the pineal gland. Csaba and Borith (1974) have reported that pinealectomy of the rat results in a striking increase in the number of parafollicular cells. The extent of the proliferation is such that, following removal of the pineal, parafollicular cells outnumber follicular cells in the thyroid gland. If this is the case for all mammals, it is possible that the pineal may have a tonic inhibitor effect on the proliferation of parafollicular cells (Csaba and BorAth, 1974). This

22

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

would explain the increase in parafollicular cell number brought about by pinealectomy. In summary, although there are hints of possible associations between parafollicular cell activity and the activity of other endocrine glands, the evidence is far too tenuous at this time to permit one to draw firm conclusions regarding the nature or importance of these associations.

5. Drugs There is some biochemical and histological evidence that the antithyroid drugs, thiouracil and methylthiouracil, affect parafollicular cell function. Care et al. (1966) showed that these drugs reduce the hypocalcemic response (to calcitonin secretion) following hypercalcemic perfusion of the thyroid. They concluded that the drugs block the release of calcitonin from the thyroid. Histological studies have provided confirmatory data. Marked hyperplasia of parafollicular cells has been observed in the thyroid glands of the guinea pig (Velick?, 1971c), pig (Young et al., 1968), and dog (Kameda, 1974a) following chronic administration of thiouracil or methylthiouracil. A direct effect of these drugs on parafollicular cell activity could thus account for an earlier observation by Nunez et al. (1967a) that the epithelium of thiouracil-induced thyroid tumors transplanted into a normal rat contained approximately 20% parafollicular cells. In contrast, the thyroid gland of the recipient animals was relatively free of these cells (< 1% of the follicular cell population). Although it has been suggested that the antithyroid drugs act on a cellular peroxidase, as in follicular cells, such a system has yet to be demonstrated in parafollicular cells (Kameda, 1974a). It has also been observed that cells with a distribution corresponding to that of parafollicular cells in the thyroid gland of the mouse accumulate nicotine (Slanina and Tjalve, 1973) and atropine (Slanina and Tjalve, 1972) following the injection of these drugs in radioactive form. It has been suggested that, since parafollicular cells possess uptake mechanisms for amino acids that are precursors of biogenic amines, the incorporation of these drugs into the cells reflects a common transport mechanism (Slanina and Tjalve, 1972,1973). However, it has also been shown that catecholamines increase the secretion of calcitonin from parafollicular cells (Bruce and Care, 1969; Fisher et al., 1973).Nicotine is known to alter the turnover rate of catecholamines in the nervous system (Slanina and Tjalve, 1973). It is impossible at this time to do more than speculate as to the meaning of the uptake of these two compounds by parafollicular cells.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

23

111. Electron Microscope Studies of Mammalian Thyroid Parafollicular Cells

A.

NORMALTISSUE

1. Adult Fine Structure In the last dozen years a great many studies of the fine structure of the mammalian thyroid parafollicular cell have been carried out. These studies, and the species examined, are listed in Table 11. Although there are some minor differences among species, such studies have clearly demonstrated that the parafollicular cells of all mammals so far examined are almost identical in fine structure. This fine structure is clearly different from that of the thyroid follicular cell. Electron microscopy has shown that parafollicular cells are found in close apposition to follicular cells and within the follicular basement membrane. They usually occur singly or in small groups of three or four. However, occasionally, large clusters are found forming irregular sheets and masses containing between 10 and 20 cells (Ekholm and Ericson, 1968; Nunez et al., 1967b; Welsch et al., 1969).When found singly, parafollicular cells are mostly round to oval, whereas in groups they appear polyhedral. Although direct contact of parafollicular cells with the follicular colloid has been reported by some workers (Dorrenhaus et al., 1971; Rost and Rost, 1975; Stoeckel and Porte, 1967; Tashiro, 1964), the overwhelming majority of electron microscope investigations have failed to find such a contact in the normal state. A thin strand of follicular cell cytoplasm probably always separates parafollicular cells from the colloid. The most salient ultrastructural feature of the parafollicular cell is the numerous round secretory granules that fill extensive areas of the cell’s cytoplasm (Plate I, Fig. 1).In early studies, when tissues were fixed with osinic acid alone, such granules appeared as empty vesicles (Lucian0 and Reale, 1964; Trolldenier, 1967; Young and Leblond, 1963).In subsequent studies of glutaraldehyde-fixed tissue, the secretory granules were found to contain a homogeneous or finely granular electron-dense core, sometimes separated by a light rim from the smooth membrane limiting the granule. Secretory granules have been found to be rather constant in size and to have diameters ranging from 50 to 150 nm in the mouse (Walker, 1966a), from 100 to 240 nm in the rat (Christov et al., 1972; Ekholm and Ericson, 1968; Murakami, 1970), from 100 to 200 nm in the rabbit, pig, and hedgehog (Lupulescu, 1972; Pearse and Welsch, 1968), and from 100 to 300 nm in the

24

ELADIO A. NUNEZ AND MICHAEL D. GERSHON TABLE I1 ELECTRONMICROSCOPE STUDIES OF PARAFOLLICULAR CELLS OF THE MAMMALIANTHYROID GLAND Species

Normal adult animals Armadillo Bat

Cat cow

Dog

Dolphin Garden dormouse Guinea pig Hamster Hedgehog Human Monkey Mouse

Opossum Pig Rabbit Rat

References Azzali (1966, 1968) Azzali (1964, 1967, 1968); Gershon and Nunez (1973, 1976); Haymovits et al. (1976); Nunez et al. (1967b, l968,1969,1970a,b); Nunez and Becker (1970); Nunez and Gershon (1972);Velickf and Titlback (1972, 1974) HHkanson et al. (1973); Velickf (1971a); Welsch and Buchheim (1972) Black et al. (1973b); Capen and Young (1967a, 1969); Yarrington et al. (1976);Young and Capen (1968,1971) Dorrenhaus et al. (1971): Kalina and Pearse (1971); Kameda (1973, 1974a,b); Nunez et al. (1974); Pearse (1966a); Rost and Rost (1975); Shively and Epling (1969); Tashiro (1962, 1964); Teitelbaum et al. (1970, 1972) Young and Harrison (1969) Martoja (1974) Lietz (1971); Velick? (1970); Welsch et al. (1969) Azzali (1964, 1968); Biddulph and Maibenm (1972); Stoeckel and Porte (1967); Stoeckel et al. (1967) P a r s e and Welsch (1968) Braunstein and Stephens (1968); Teitelbaum et al. (1971) Aoi (1966) Azzali (1968);Bauer and Murad (1964); Ericson (1970, 1972a); HHkanson et al. (1971b); Kracht et al. (1968a); Lietz (1971); Melander et al. (1971a); Nanba and Fujita (1969); Walker (1966a,b); Welsch et al. (1968) Azzali (1964, 1966, 1968); Fortney (1973) Bauer and Teitelbaum (1966); DeGrandi et al. (1971); Fetter and Capen (1970);Nunez et al. (1976);Trolldenier (1967); Young et al. (1968) Lupulescu (1972); Welsch et al. (1969) Azzali (1962); BlahosovP et al. (1974); Blaser and Schnorr (1972); Boulet and Pellelier (1974); Bucher and KrstiC (1974);Cameron (1968); Chan and BBlanger (1968);Csaba and BorPth (1974); Dahlstrom and Ericson (1972); Diaz-Flores et al. (1972); Dorrenhaus et al. (1971); Ekholm and Ericson (1968); Ericson (1968); Gauguin et al. (1973); Gonticas et al. (1968); Krstik (1969);Krsti6 et al. (1974); Lietz (1970, 1971); Lietz et al. (1969); Lucian0 and Reale (1964); Martoja (1974); Matsuzawa (1966); Matsuzawa and Kurosumi (1967); Moura et al. (1971); Murakami (1970); Pantie et a:. (Continued)

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR

CELLS

25

TABLE I1 (Continued) Species

Sheep Sloth (three-toed) Tree shrew Woodchuck Fetal-neonatal studies Cat cow Dog Guinea pig Human Mouse Opossum Pig Rat

Sheep I n oitro studies Dog Rat

References (1970); Rohr and Hasler (1968); Stachura and Pearse (1970); Stoeckel and Porte (1972); Sundstrom (1971); Velickq (1970); Wassermann (1976); Weisbrode and Capen (1974); Wissig (1962); Yoshimura et al. (1962); Young and Leblond (1963) Atack et al. (1972); Collins and Barlet (1972); Etcheveny and Zieher (1968);Velick? (1971a) Azzali (1964, 1968) Pearse (1968a); Welsch et al. (1968) Krupp et al. (1976) Welsch (1971, 1972) Lindberg and Talanti (1971) Gershon et al. (1971); Nunez and Gershon (1976b); Welsch (1971, 1972) Welsch (1971) Chan and Conen (1971); Lietz (1971); Lietz et al. (1971); M a d e (1970) Sato et al. (1966); Treilhou-LaHille and Beaumont (1975) Azzali (1966, 1968) DeCrandi et al. (1971) Calvert (1972, 1974); Chan (1972); Christov et al. (1972); Ishikawa (1965); Ker (1972); Welsch (1971, 1972) Jordan et al. (1973) Bussolati and Monga (1970); Bussolati et al. (1969b, 1970) Roth et al. (1974)

dog and sheep (Atack et al., 1972; Kameda, 1973; Rost and Rost, 1975). Secretory granules tend to concentrate in the cytoplasmic areas closest to the follicular basement membrane (Ekholm and Ericson, 1968; Lietz, 1971; Nunez et al., 1967b; Young and Capen, 1971). However, in spite of the very close apposition of mature secretory granules to the plasma membrane, signs of exocytosis, such as invaginations (omega figures), along the course of the plasma membrane have rarely been encountered in micrographs of adult cells. This may indicate that adult cells are not very active in secretion. In the adult thyroid gland, parafollicular cells filled with secretory granules predominate. These are considered to be in a storage phase of the secretory cycle (Young and Capen, 1971).

26

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

27

Parafollicular cells are also distinguished by the presence of a welldeveloped Golgi complex (Plate 11, Fig. 2). The Golgi complex typically has three to five groups of slightly dilated sacculesand many vesicles, vacuoles, and small prosecretory granules. A common finding in such species as the mouse (Lietz, 1971; Welsch et al., 1969), dog (Kameda, 1973; Shively and Epling, 1969), rat (Dorrenhaus et al., 1971; Ekholm and Ericson, 1968), active bat (Nunez et al., 1967b), and tree shrew (Welsch et al., 1969) is that the vesicles and saccules of the Golgi complex contain varying amounts of granular material suggestive of a stage in the formation of prosecretory granules. Centrioles and multivesicular bodies have also been found in the Golgi zone. Prominent cilia are occasionally present (Plate 11, Fig. 2). The rough endoplasmic reticulum usually appears in the form of tubular profiles scattered throughout the cytoplasm but is occasionally arranged in parallel stacks or whorls of concentric lamellae (Plate I, Fig. 1).In some species, such as the dog (Kameda, 1973) and rat (Dorrenhaus et al., 1971), some parafollicular cells display prominent dilated granular endoplasmic reticulum. In these cells the intracisternal space contains a pale, granular substance. In the dog such cells are packed with secretory granules (Kameda, 1973), whereas in the rat (Dorrenhaus et al., 1971) similar cells are found to contain many fewer granules. Multiple cisternae of rough endoplasmic reticulum have been observed in rat parafollicular cells (Wassermann, 1976). Parafollicular cells contain moderate numbers of mitochondria which are mostly round to elongate in shape. The mitochondria often exhibit branching, and their cristae are numerous and randomly oriented in relation to the mitochondria1 axis. Mitochondria display no particular orientation or concentration in the cell. A single, large, round or oval to irregularly shaped nucleus, sometimes with deeply indented nuclear membranes, is generally located in the center of the cell. The chromatin appears finely dispersed and exhibits an irregular density at the nuclear periphery. Usually, a single, large nucleolus is found. Multivesicular bodies, autophagic vacuoles, and/or iysosomal dense structures, about 0.6-1.0 pm in diameter, are encountered in the cell PLATE I. FIG. 1. Electron micrograph showing a group of parafollicular cells (PC) of the thyroid gland from an adult bat captured in June. Extensive areas of the cytoplasm of the parafollicular cells are filled with dense secretory granules which are mostly round. The cells are within the basement membrane (arrowheads)and are separated from the colloid (C) by lining follicular cells (FC). ER, Rough endoplasmic reticulum; I, intercellular space. x 15,000.

28

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

29

PLATE 111. FIG.3. Parafollicular cell from a 2-year-old dog (beagle). The cells often exhibit images (arrow) indicating autophagic removal of secretory granules. x 35,000.

cytoplasm (Plate 111, Fig. 3) but occur sparsely (Dorrenhaus et al., 1971; Lietz, 1971; Luciano and Reale, 1964; Young and Harrison, 1969). In the rat, it has been estimated that each adult parafollicular cell contains from one to four lysosomes (Ekholm and Ericson, 1968). Luciano and Reales (1964) have reported, however, that the number of lysosomes in parafollicular cells can increase in older animals. Microtubules and 6.5-nm filaments are characteristic components of adult parafollicular cells. Microtubules are particularly numerous near the Golgi complex and in the peripheral areas of the cells (Bussolati and Monga, 1970; Welsch et al., 1969). There are more microtuPLATE 11. FIG.2. Portions of a follicular cell (FC) and a parafollicular cell (PC) in the thyroid gland of a I-year-old basenji dog. A well-developed Golgi complex (G) with numerous prosecretory granules is prominent in the cytoplasm of the parafollicular cell. A cilium (CI) protrudes into the neighboring follicular cell. Note the large number of large, dense lysosomal bodies (L) in the cytoplasm of the follicular cell. Also note that some of the secretory granules in the cytoplasmic matrix of the parafollicular cell do not exhibit a dense core (arrowhead). This is probably due to a fixation artifact. x 30,000.

30

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

bules in parafollicular cells than in follicular cells (Sandborn et al., 1964). Filaments occasionally form dense bundles of parallel units and occasionally touch the membranes of secretory granules (Lietz, 1971). They also insert into desmosomes when these are present. Numerous free ribosomes, often arranged as rosettes of polysomes, are scattered throughout the cytoplasmic matrix. It has been claimed that the number of free ribosomes varies with the secretory state of the cells (Bussolati et al., 1970; Lietz, 1970). In a few species, such as the active bat (Nunez et aZ., 1967b), the plasma membrane shows many contortions and complex interdigitations with adjoining cells, but in most species, including the rat (Ekholm and Ericson, 1968),it has a fairly straight course with only a few irregularities and bulges. In some cases, long, fingerlike projections extending between adjacent parafollicular cells are encountered and suggest the formation of intercellular canaliculi. Desmosomes are present between apposed parafollicular cell membranes (Lietz, 1970; Matsuzawa, 1966; Rost and Rost, 1975) and have also been found between adjoining follicular and parafollicular cells (Lucian0 and Reale, 1964; Nunez and Becker, 1970). Intercellular channels or canaliculi, distinguished by abundant desmosomes, have been reported between parafollicular and follicular cells (Young, 1966; Young and Harrison, 1969). Young (1966) has suggested that the presence of desmosomes between these epithelial cells of different embryological origin, follicular and parafollicular cells, can be explained by both cells participating in the formation of such intercellular channels. The role of intercellular channels in parafollicular cell function is not yet clear. However, in some secretory cells, such as the chief cells of the parathyroid gland (Nunez et al., 1972b), they are considered to represent sites for the discharge of hormonal material. Electron microscope studies have shown that parafollicular cells are closely associated with capillaries and axons (Plate IV, Fig. 4). The majority of these axons are located in the interfollicular space. However, recent studies have clearly shown that axons are also found within the follicular basement membrane (Nunez and Gershon, 197613; Tice and Creveling, 1975; Young and Harrison, 1969),often intimately associated with parafollicular cells. Such associations between parafollicular cells and bundles of axons have been seen in both fetal (Nunez and Gershon, 1976b) and adult (Young and Harrison, 1969) thyroid glands. It has generally been accepted that the mammalian thyroid gland receives an autonomic nerve supply innervating the blood vessels (Cunliffe, 1961;Nonidez, 1935). However, in

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

31

PLATE IV. FIG.4. Electron microscope autoradiograph of bat thyroid incubated with n~repinephrine-~H (9 x lo-' M ) . Silver grains (arrows)are found only over adrenergic nerves. No grains are found over parafollicular cells (PC) or follicular epithelium (FC). I, Intercellular space; V, blood vessel. x 15,000.

the case of the ultimobranchial body in lower vertebrates, there is evidence that the secretory activity of the ultimobranchial glandular cells may actually be under sympathetic nerve control (Robertson, 1967). In mammals, parafollicular cells are intimately associated with nerve terminals that can be labeled with radioactive norepinephrine (Plate IV, Fig. 4) and so are adrenergic (Gershon and Nunez, 1976; Melander et al., 1974a). Catecholamines increase parafollicular cell secretion (Broulik and Pacovskjr, 1973; Bruce and Care, 1969; Fisher et aZ., 1973). Therefore the possibility that parafollicular cell function is under the influence of the sympathetic nervous system warrants further investigation. Dark and light parafollicular cells have been observed in electron micrographs of such species as the bat (Nunez and Gould, 1967), tree shrew (Welsch et aZ., 1969),dog (Tashiro, 1964),mouse (Welsch et al., 1969),and guinea pig (Welsch et al., 1969). However, the nature of the

32

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

PLATE V. FIG.5. Electron micrograph showing portion of a parafollicular cell follicle (PC) found in the thyroid gland of a 1-year-old dog (beagle). Arrowheads point to microvilli extending from the parafollicular cell into the lumen (L) of the follicle. x 17,000.

differences in density seen in electron micrographs is open to question. It may represent an artifact of fixation. Ultimobranchial follicles or cysts composed of a mixture of different cells, including parafollicular cells, are occasionally found in the thyroid glands in such species as the rat (Krupp and Frink, 1974), mouse (Calvert and Isler, 1970), and dog (Kameda, 1973). However, rare follicles composed solely of parafollicular cells have also been found in the thyroid gland of the adult dog (Nunez and Gershon, 1976b; Teitelbaum et d., 1970).Parafollicular cells lining such follicles exhibit microvilli which protrude into the luminal cavity (Plate V, Fig. 5). Moreover, tight junctions at the apical portion of the lateral surface and desmosomes join the lining parafollicular cells. Some of the parafollicular cells forming such follicles exhibit an ultrastructure similar to that of developing fetal parafollicular cells (Nunez and Gershon, 1976b). The significance of these parafollicular follicles is unknown. Teitelbaum et aZ. (1970)

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

33

have speculated that the lumina of such follicles may function as storage depots for calcitonin. Another view is that they represent rests of the ultimobranchial bodies which remain as atypical follicles throughout postnatal life (Kameda, 1973). 2. Calcitonin Localization There is now firm evidence that calcitonin is stored intracellulary in the secretory granules of parafollicular cells. The studies that provided this evidence are outlined below. a. Cell Fractionation Studies. Studies employing differential ultracentrifugation of thyroid homogenates of the pig (Bauer and Teitelbaum, 1966; Cooper and Tashjian, 1966), rat (Cooper and Tashjian, 1966), sheep (Atack et al., 1972; Oeltgen and L’Heureux, 1975), and cow (Cooper and Tashjian, 1966)have all demonstrated that bioassayable calcitonin activity is mainly concentrated in the 100,000 g particulate fraction. Electron microscope examination of this fraction in two species, the sheep (Atack et al., 1972) and pig (Bauer and Teitelbaum, 1966), has shown that the fraction is rich in membrane-limited densecored granules which are indistinguishable from the secretory granules found in the cytoplasm of parafollicular cells of intact animals. It was shown b y Greenberg et al. (1971) that such granule-containing particulate fractions, obtained from pig thyroid homogenates, released calcitonin when diluted in isotonic buffer. These investigators, furthermore, concluded that the release of calcitonin on dilution in isotonic buffer is consistent with the hypothesis that there is an intracellular release of hormone from storage granules on physiological stimulation of parafollicular cells. However, this evidence is open to alternative interpretation and, as discussed in Section III,A,2,b there is considerable reason to believe that exocytosis, not intracellular release, is the physiological mechanism mediating calcitonin secretion. b. Hypercalcemic Stimulation. The primary secretory stimulus for the parafollicular cell is elevated blood calcium (Hirsch and Munson, 1969). Electron microscope studies of acute hypercalcemia, induced either by perfusion of the thyroid with hypercalcemic blood or by the injection of calcium into intact animals (Christov et al., 1972; Foster et al., 1964a; Matsuzawa, 1966; Matsuzawa and Kurosumi, 1967), causes in all cases marked degranulation of parafollicular cells by 2 hours after hypercalcemic stimulation (see Plate VI, Figs. 6 and 7).A similar observation was reported when the calcium concentration in the incubating solution of thyroid tissue maintained in organ culture was elevated (Bussolati and Monga, 1970; Bussolati et ol., 1969b). However,

34

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

35

in the pig, for reasons not yet fully understood, degranulation of parafollicular cells does not occur even after hypercalcemic stimulation for 1hour (Young et al., 1968). In contrast, in the rat, as early as 30 minutes after injection of calcium there is already considerable parafollicular cell degranulation (Lietz, 1971).This difference in the early response to stimulation may have been due to a real species difference, but it may have been simply related to the amount of calcium used by the experimenters. Regranulation of stimulated parafollicular cells has been reported after injection of ethylenediamine tetraacetic acid (EDTA) to lower blood calcium (Christov et al., 1972). Chronic hypercalcemia, induced either by the injection of large doses of vitamin D or the administration of calcium or parathormone for 3-30 days, results in degranulation of parafollicular cells in such species as the cow (Black et al., 1973b; Capen and Young, 1969),cat (HBkanson et al., 1973), hamster (Biddulph and Maibenco, 1972), mouse (Melander et al., 1971a), and rat (Boulet and Pellelier, 1974; Cameron, 1968; Dorrenhaus et al., 1971; Ericson, 1968; Lietz, 1970; Lietz et al., 1969; Murakami, 1970; Rohr and Hasler, 1968; Stoeckel and Porte, 1972; Weisbrode and Capen, 1974). This prolonged stimulation is rather specific with regard to the secretory granules and is not a toxic effect since no other components of parafollicular cells show degenerative changes. Moreover, follicular cells, as well as other thyroid cells, remain normal in all respects following chronic hypercalcemia. A striking increase in free ribosomes and in elements of the rough endoplasmic reticulum and Golgi complex of parafollicular cells has been noted following such chronic stimulation (Biddulph and Maibenco, 1972; Capen and Young, 1969; Ericson, 1968; Murakami, 1970; Nanba and Fujita, 1969).This may indicate that a compensatory increase in the parafollicular cell’s machinery for protein synthesis occurs under conditions leading to chronic release of calcitonin. In several cases in which thyroid calcitonin was measured during stimulation, a striking decrease in the calcitonin content of the gland ~~

PLATE VI. FIGS.6 and 7. Electron micrograph of a parafollicular cell (PC) from an active bat captured in late May immediately after the highly active period preceding arousal from hibernation. At this time, parafollicular cells are again characterized by the transient appearance of intracisternal dense granules (large arrows). It is interesting to note that the injection of calcium chloride (1mgkg) at this time does not significantly affect the fine structure of the parafollicular cells. In contrast, when active bats caught in late June are injected with calcium chloride, many of the parafollicular cells, as depicted in Fig. 7, undergo partial or total degranulation and only a few small secretory granules (small arrows) remain in the cytoplasm. FC, Follicular cells; C, colloid. x 20,000; x 15,000.

36

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

paralleled the degranulation of parafollicular cells (Capen and Young, 1969; HBkanson et al., 1973). This observation clearly supports the view that the granules are the intracellular storage vesicles for calcitonin. The results of an ultrastructural study on parafollicular cells of young growing dogs fed a high-calcium diet for 60 weeks are also consistent with this conclusion. It was found that parafollicular cells examined at about midperiod were characterized by an abundance of cytoplasmic microtubules and microfilaments, often intimately associated with secretory granules (Nunez et al., 1974). Microtubules and microfilaments are thought by some investigators to play an important role in the transport and secretion of hormone in such endocrine cells as thyroid follicular cells (Williams, 1976), parathyroid chief cells (Nunez et al., 1976), and pancreatic endocrine cells (Malaisse et al., 1971).Thus their abundance in parafollicular cells undergoing such prolonged stimulation may be related to the sustained secretion of calcitonin (Nunez et al., 1974). Bussolati and Monga (1970) also noted a striking increase in the number of microtubules in the cytoplasm of parafollicular cells of a dog thyroid maintained in culture and exposed to a high amount of calcium. Signs of exocytosis of secretory granules can rarely be demonstrated ultrastructurally in chronically stimulated parafollicular cells. This may indicate that the step during exocytosis following fusion of the vesicle membrane with the plasma membrane, when the granule is open to the outside but still recognizable, is exceedingly brief. Thus, if exocytosis is a short-lived, evanescent event, it may be difficult to “freeze” it in a recognizable form during fixation. This is also true of the adrenal medulla in most species, where exocytosis is known to occur but is difficult to observe (Smith et al., 1973),but not of such cells as islet cells (Lacy, 1967) or anterior pituitary cells (Farquhar, 1971), where signs of granule extrusion are rather common following stimulation. However, stimulated parafollicular cells often exhibit pale vesicles and fading secretory granules with low density. Lietz (1971) and others (Fortney, 1973), not attributing these images to poor fixation, have proposed that, in addition to exocytosis, an intracellular mechanism for release of the granule contents is operative. This view leaves unexplained the question of how free intracellular calcitonin passes through the plasma membrane and escapes from the parent parafollicular cell. It probably will be necessary to obtain specific evidence that exocytosis is the mechanism of secretion of calcitonin by parafollicular cells. In light of the ultrastructural observations made to date, reliance on analogy with all the other cells once thought to secrete by another mechanism and now known to secrete by exocy-

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

37

tosis is not sufficient. Still, it should be noted that there are situations in which ultrastructural evidence for exocytosis b y parafollicular cells can readily be obtained. These include the thyroid of the newborn dog (Nunez and Gershon, 1976b) (see Plate VII, Fig. 9) and the thyroid of the bat on arousal from hibernation (see Plate VII, Figs. 8 and 10). It may be that, where exocytosis is extensive, the probability of observing it increases and overcomes the difficulty imposed by the rapidity of the event. c. Ultrastructural Zmmunocytochemical Studies. Using cytochrome-c-labeled antibody and peroxidase-labeled antibody, DeGrandi et al. (1971) and Kalina and Pearse (1971), respectively, examined the ultrastructural localization of calcitonin in pig and dog thyroid glands. In both cases, it was found only in the parafollicular cells. Moreover, in these cells, it was associated mainly with the secretory granules. However, reaction product was also found throughout the cytoplasm. No reaction product was found in the Golgi saccules, mitochondria, and nuclei of adult parafollicular cells. Kalina and Pearse (1971), however, also found calcitonin in the cisternae of the rough-surfaced endoplasmic reticulum. DeGrandi et al. (1971) suggest that thq labeling of the cytoplasm may be due to leakage of the polypeptide from the granules. However, as proposed by Kalina and Pearse (1971), the reaction product found in the cytoplasm may be artifactual and simply reflect the limitations of the techniques used. Nevertheless, the fact that the majority of secretory granules in both studies were associated with calcitonin is consistent with the conclusions drawn by others from cell fractionation and studies of the morphological effects of hypercalcemic stimulation. It is therefore safe to conclude that the specific granules represent the intracellular storage depots of calcitonin.

3. Cytochemical Studies a. Enzyme Localization i. Cholinesterase. The localization of cholinesterase has been examined ultrastructurally in the rabbit by Welsch et al. (1968), in the rat by Welsch and Pearse (1969), Welsch (1972), and Gauguin et al. (1973), and in the cow by Young and Capen (1971). In all these animals, cholinesterase activity was found in the cisternae of the rough endoplasmic reticulum, in the Golgi complex, in and around secretory granules, and in the plasma membrane. In addition, Welsch and Pearse (1969) reported that stimulated parafollicular cells lose much of their cholinesterase activity, except in the plasma membrane where it is most concentrated.

38

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

PLATE VII. FIG.8. Dense secretory granules (arrow),indistinguishablefrom those in the cytoplasm of parafollicular cells, are found in the intercellularspaces (I) between thyroid follicles in bats captured during the active arousal period. x30,OOO. FIG.9.

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

39

In contrast to the above studies, Lietz (1971) reported that cholinesterase was not found in the Golgi complex and was very seldom present in the cisternae of the endoplasmic reticulum in parafollicular cells of the thyroid glands of the rat, guinea pig, mouse, and rabbit. He did, however, confirm its localization in the plasma membrane. ii. Acid phosphatase. Lietz (1971) reported that, except in the mouse, there is only slight acid phosphatase activity in the parafollicular cells of most mammals. Acid phosphatase reaction product was found in lysosomes and in the Golgi complex. Welsch (1971) and Dorrenhaus et al. (1971) also found acid phosphatase activity in lysosomes and in the Golgi complex of rat parafollicular cells. Moreover, Welsch (1971)reported that much stronger acid phosphatase activity occurs in the Golgi complex of young animals. iii. Aryl sulfatase and alkaline phosphatase. Lietz (1971) reported that, as would be expected, the localization of aryl sulfatase in parafollicular cells appears to be identical to that of acid phosphatase. Calvert (1974), in a study of alkaline phosphatase activity in the rat thyroid gland, reported that parafollicular cells of embryonic and neonatal rat thyroids did not contain this enzyme. Unfortunately, he did not examine the thyroid glands of adult animals. iv. Peroxidase. The localization of peroxidase was studied by Boulet and Pellelier (1974) in parafollicular cells of the rat thyroid gland. They reported that peroxidase content is markedly enhanced in parafollicular cells acutely stimulated by calcium or chronically stimulated by vitamin D,. Peroxidase was found in the smooth vesicles located in the Golgi region and at the periphery of the cytoplasm (perhaps in the plasma membrane), as well as in lysosomes. v. Nucleosidephosphatase. Lietz (1971), using adenosine-and inosine-5’-diphosphatase as substrates, found reaction product in the cisternae of the Golgi complex. With adenosine-Striphosphataseas substrate, distinct activity was found in the outer plasma membrane in parafollicular cells of the rat, mouse, guinea pig, and rabbit. vi. Summary. The significance of the above enzymes in the function of parafollicular cells is still not clear. However, several views Signs of exocytosis of secretory granules (arrow)are also commonly found in the thyroid gland of neonatal puppies immediately after birth. I, intercellular space. x60,OOO. From Nunez and Gershon (1976b). FIG. 10. Electron micrograph of the thyroid gland of an active bat caught in early May, just after the active arousal period. At this time, many secretory-like granules (arrows) are still found in the intercellular spaces (I) of thyroid follicles. Occasionally, such granules appear to be engulfed (arrowheads) by macrophagiclike cells. FC. Follicular cell. x30,OOO.

40

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

have been proposed. Lietz (1971) has concluded that cholinesterase, acid phosphatase, and aryl sulfatase are enzymes involved primarily in the catabolism of cell particles. He feels they do not have a specific role in calcitonin synthesis and release. However, an opposite view has been proposed by Welsch and Pearse (1969), Welsch et al. (1968), and Young and Capen (1971). They suggest that cholinesterase, and presumably acid phosphatase and aryl sulfatase, participate in control of the synthesis and release of calcitonin. A similar view has been put forward for peroxidase by Boulet and Pellelier (1974). They suggest that, following exocytosis of secretory granules, granular membranes are retrieved and partially digested by peroxidase-containing enzymes. The participation of lysosomes in the control of secretion of secretory granules has also been suggested by Rohr and Hasler (1968). b. Nonenzymic Cytochemical Studies i. Masked metachromasia. Rost and Rost (1975) examined the phenomenon of masked metachromasia electron microscopically. They reported that hydrolysis with acid removes the limiting membrane but does not alter the dense core of the secretory granules. They concluded that masked metachromasia is most likely the result of an interaction between the stain and the polypeptide material stored in the core of the secretory granules. ii. Argyrophilic reaction. Vassallo et al. (1971) and Hakanson et al. (1971b) examined the argyrophilic reaction in ultrathin fixed sections of mouse and dog thyroid glands. Ultrastructurally, it was found that silver particles localized at the margin and in the substance of secretory granules in both mouse and dog parafollicular cells. The results of the electron microscope observations are therefore not inconsistent with the view that disulfide bonds, either of a calcitonin-membrane combination product or of calcitonin itself, are responsible for silver attachment (Roediger, 1975). iii. Calcium localization. Martoja (1974), in an electron microscope study of the site of calcium localization in mammalian thyroid glands, reported that calcium densely coated the secretory granules of parafollicular cells. Calcium deposits were also found in the nucleus of parafollicular cells, in intercellular spaces, and along basal laminae. The significance of calcium deposits in the secretory granules, however, is not yet known. 4. Developmental Studies Electron microscope studies of parafollicular cells of the developing thyroid gland in various species, including the human (Lietz et al., 1971), dog (Nunez and Gershon, 1976a,b), opossum (Azzali,

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

4I

1968), cow (Lindberg and Talanti, 1971), mouse (Treilhou-LaHille and Beaumont, 1975), and rat (Calvert, 1972; Chan, 1972; Ker, 1972; Stoeckel and Porte, 1970; Welsch, 1971), have shown (1) that developing parafollicular cells clearly differ from developing follicular cells and (2) the embryonic development of parafollicular cells and their ,secretory granules occurs relatively early in gestation. Secretory granules first begin to develop in association with the Golgi complex and, as development proceeds, become more widely distributed in the cytoplasm. Secretory granules of fetal parafollicular cells are indistinguishable from those of adult parafollicular cells. The development of parafollicular cells also appears to be associated with an inverse relationship between the extent of rough-surfaced endoplasmic reticulum and the number of secretory granules in the parafollicular cells. Early developing parafollicular cells seem to have much rough endoplasmic reticulum and few secretory granules (Plate VIII, Fig. l l ) , while more mature cells have less granular endoplasmic reticulum but many more granules (Azzali, 1964; Chan and Conen, 1971; Nunez and Gershon, 1976b; Treilhou-LaHille and Beaumont, 1975). By the end of fetal life, the parafollicular cells resemble those of adult animals. If the presence of secretory granules in the cytoplasm is taken as evidence, the findings of developmental studies indicate that parafollicular cells synthesize and store calcitonin prior to birth. Nuclear bodies in fetal parafollicular cells have been seen and probably reflect cellular hyperactivity at this time (Nunez and Gershon, 1976b). The presence of large numbers of secretory granules in fetal parafollicular cells has led some workers to suggest that secretion of calcitonin occurs during intrauterine life (Chan and Conen, 1971; Jordan et al., 1973; Lindberg and Talanti, 1971; Welsch, 1971). However, in the dog ultrastructural findings indicate that secretion of calcitonin begins not during fetal life but immediately after birth (Nunez and Gershon, 197613). For example, just before birth, large, dense granules, 1-2 pm in diameter and resembling the cores of calcitonin granules, are found in the cisternal spaces of the rough endoplasmic reticulum of parafollicular cells (Plate IX, Fig. 12). Such intracisternal granules, however, are not found in earlier developing parafollicular cells (Nunez and Gershon, 1976b), nor have they been observed in parafollicular cells of the thyroid glands of the neonate or adult dog (Dorrenhaus et al., 1971; Kameda, 1973; Kameda, 1974a,b; Nunez et al., 1974; Nunez and Gershon, 1976b; Shively and Epling, 1969; Tashiro, 1962, 1964; Teitelbaum et al., 1972). However, similar intracisternal granules have been reported in a variety of other types of cells in association with cellular hypofunction (Watari, 1974). The occurrence of intracisternal

42

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

43

PLATE IX. FIG. 12. Electron micrograph of a parafollicular cell of a beagle fetus about 3 days prior to the expected date of birth. At this time, prominent dense, large granules often fill the cistemae of the rough endoplasmic reticulum (arrows). Also, apparent stages of condensation of these granules from a less granular material are seen (arrowheads). x 30,000.

granules in cells may reflect a situation in which synthesis of protein

by attached ribosomes of the endoplasmic reticulum and its consequent transfer to the cisternal space outstrips the capacity of the cells to transport it from the rough endoplasmic reticulum to the Golgi complex for packaging into prosecretory granules (Nunez and Gershon, 1972). Thus the phenomenon of protein condensation into granules within the endoplasmic reticulum of fetal dog parafollicular cells just before birth may be due to cells’ synthesizing calcitonin which they cannot at this time secrete. In addition, just after birth and during the ..

~~

~

~

~

PLATE VIII. FIG.11. Parafollicular cell (PC) of a beagle fetus about 3 weeks prior to the expected date of birth. The cell contains many mature secretory granules (arrowhead), abundant endoplasmic reticulum (ER), and a well-developed Golgi complex (G). Many microtubules (arrows) are also scattered in the cytoplasmic matrix. FC, Follicular cell. x 22,000.

44

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

early neonatal period, ultrastructural evidence for exocytosis of granules by parafollicular cells is abundant, in contrast to the absence of such evidence before birth (Nunez and Gershon, 1976b). This latter observation particularly is consistent with the hypothesis that parafollicular cells are inactive prior to birth but are highly active at birth (Nunez and Gershon, 1976b). Also consistent with the view that parafollicular cells secrete actively during the neonatal period is the observation by Welsch (1972) that juvenile animals contain fewer secretory granules in their parafollicular cells than adult animals. Welsch has interpreted this finding to indicate a higher turnover of calcitonin in young animals. Moreover, the observation that adult parafollicular cells contain more lysosomes than juvenile parafollicular cells (Welsch, 1971,1972) suggests that there is a slower rate of secretion in older animals. Lysosomes may be required to remove secretory granules as they accumulate in the cytoplasm of older cells. The above hypothesis, that parafollicular cells secrete actively during the neonatal period, is compatible with recent clinical observations in humans. Biochemical studies have shown that tissue and blood levels of calcitonin are high at birth but fall rapidly thereafter (Bergman et al., 1973; Garel et al., 1974; Samaan et al., 1975; Wolfe et al., 1975). Moreover, larger numbers of parafollicular cells have been found in human neonatal thyroid than in adult thyroid (Pearse, 196th; Wolfe et al., 1974, 1975), which may account for the higher levels of calcitonin in the blood and thyroid glands of neonates. Although the exact role of calcitonin in early development is still unclear, several roles for calcitonin in neonatal life have been proposed. These include regulation of nutrient absorption from milk rich in lipids (Garel et al., 1975), maintenance of normocalcemia in the face of rapid growth (Swaminathan et al., 1972), and regulation of osteocytic chondrolysis during bone modeling (Whalen et al., 1975). Parenthetically, a light microscope report, describing a study using semithin sections (0.5-1.0 pm in thickness), states that parafollicular cells of maternal animals also become degranulated during lactation (Blahosovii et al., 1974). 5. Hibernation Histochemical studies have clearly demonstrated that parafollicular cells undergo cyclic changes during the yearly life cycle of hibernating animals. Electron microscope studies have defined the intracellular changes that occur during the annual cycle and have also aided in explaining the significance of parafollicular cells in the life cycle of the hibernator.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

45

Electron microscope studies of hibernators such as the bat (Azzali, 1968; Nunez et al., 196713; VelickG and Titlback, 1972), hamster (Stoeckel et al., 1967), and hedgehog (Pearse and Welsch, 1968) have shown that animals captured in the active phase of their life cycle, June to September, exhibit a fine structure similar to that of other mammalian species, such as the rat (Ekholm and Ericson, 1968) and dog (Nunez et al., 1974), indicating that the cells are in a normal state of activity. However, examination of the fine structure of parafollicular cells of animals entering the prehibernation phase of the life cycle, September to November, has shown that parafollicular cells begin to exhibit striking intracellular changes suggestive of a slowdown in function (Nunez et al., 1970a,b; Pearse and Welsch, 1968). At this time, there is a progressive accumulation of secretory granules in the cytoplasm (Nunez et al., 1970a; Pearse and Welsch, 1968), and in the hedgehog many of the granules fuse together to form large, vesiculated bodies (Pearse and Welsch, 1968). Groups of mitochondria often form large clusters (Pearse and Welsch, 1968). In addition, another striking change is the appearance of large, dense granules, 1-5 pm in diameter, within the cisternal space of the rough-surfaced endoplasmic reticulum (Nunez et aZ., 1968, 1970a) (Plate X, Fig. 13, and Plate XI, Figs. 14 and 15).The presence of intracisternal granules in the cytoplasm of parafollicular cells at this time, as in parafollicular cells of fetal puppies just prior to birth (Nunez and Gershon, 1976b), may be indicative of a slowing of intracellular transport of secretory material from the granular endoplasmic reticulum to the Golgi complex. This indicates a decrease in the secretion of calcitonin and, if this hypothesis is correct, would account for the accumulation of secretory granules in the cytoplasm of parafollicular cells. Consistent with a slowdown in parafollicular cell activity at this time is the transient occurrence of large, slender, paracrystalloid bodies in the cytoplasm of parafollicular cells (Nunez et al., 1970a,b) (Plate XII, Figs. 16 and 17). Similar paracrystalloid bodies have been found in several other endocrine cells. They have also been found in thyroid follicular cells (Nunez et al., 1975; Yoshimura and Iris, 1961), chief cells of the parathyroid gland (Nunez et aZ., 1976), and cells of the anterior pituitary (Shiino and Rennels, 1974). Identical structures can also be induced pharmacologically in follicular cells of the thyroid gland of the rat (Nunez et al., 1975). It has been proposed that such paracrystalloids may universally accompany a rapid shutdown of cellular activity (Nunez et al., 1975). If so, their occurrence just prior to hibernation in parafollicular cells of the thyroid gland of the bat is consistent with the view that parafollicular cell activity slows down at this time. The ap-

46

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

47

pearance in the cytoplasm of parafollicular cells at this time of many atypical secretory granules characterized by a spermlike shape has also been interpreted to represent a slowdown in intracellular activity (Nunez et al., 1970b). With the start of hibernation parafollicular cells undergo further changes which indicate continued progressive inactivation of the cells (Azzali, 1967; Nunez et al., 1969,1970a; Pearse and Welsch, 1968).At the beginning of hibernation they undergo degradation with a loss of density of the preformed secretory granules (Azzali, 1968; Nunez et al., 1967b, 1969; Velick? and Titlback, 1972). Many granules exhibit short segments of membranes and myelinlike figures (Nunez et al., 196713) (Plate XIII, Fig. 18 and inset). Eventually all the secretory granules are removed from the cytoplasm, probably by lysosomal digestion (Nunez et al., 1969) (Plate XIV, Fig. 19).The large intracisternal granules also disappear from the cytoplasm of parafollicular cells at the start of hibernation (Nunez et al., 1970a). In other cells, such as pituitary thyrotrophs, Farquhar (1971) has reported that intracisternal granules undergo lysosomal removal. This may also be the case in the thyroid parafollicular cell. As bats enter hibemation, images are obtained (see Plate XI, Fig. 15) indicating the apparent dissolution and/or lysosomal removal of the large, dense granules within the cisternal space. Near the end of hibernation, parafollicular cells form new secretory granules; just before arousal from hibernation fully granulated parafollicular cells are encountered in the thyroid gland (Azzali, 1968; Nunez et al., 1969; Velick? and Titlback, 1972), but there are no ultrastructural signs of exocytosis of secretory granules. However, during the transient but very active arousal period, signs of exocytosis of secretory granules are common (Plate VII, Figs. 8 and 10).It is interesting to note that cells occasionally exhibit large, dense, intracisternal granules just after arousal (Plate VI, Fig. 6). Such intracisternal granules also appear transiently in the late May- July period (Velick? and Titlback, 1974). The electron microscope studies thus indicate that parafollicular cell activity slows down prior to hibernation, ceases during hibernation, and resumes at arousal. The results of biochemical studies of calcitonin levels in the thyroid glands of hibernators seem consistent with this view. It has been shown that the amount of calcitonin in the PLATE X. FIG.13. Electron micrograph of bat parafollicular cells just pnor to the start of hibernation. At this time, October, the secretory granules accumulate and pack the cytoplasmic matrix. Also, in some cells dense material accumulates within the cistemae of the rough endoplasmic reticulum (arrowheads),and such profiles fill the cell’s cytoplasm. x 15,000.

48

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHY SIOLOGY O F THYROID PARAFOLLICULAR CELLS

49

PLATE XII. FIG. 16. Electron micrograph of a parafollicular cell from a bat caught in October. At this time, prominent crystalloid inclusion bodies (arrow) are found i n the cytoplasmic matrix. x 22,000. FIG. 17. At a higher magnification, such intracellular crystalloid bodies are characterized by many parallel membranes which run along the axis the full length of the body. x 60,000.

thyroid gland of the bat is very high at the start of hibernation and then undergoes a steady decline until it approaches the active level by the end of the hibernating period (Haymovits et al., 1976).A high concentration of calcitonin has also been found in the thyroid glands of hibernating ground squirrels (Mussacchia and Kenny, 1973). It has been ~

PLATE XI. FIG.14. Electron micrograph of a parafollicular cell (PC) from a bat caught in early October, illustrating the compact dense material filling the cisternal space of the rough endoplasmic reticulum (arrows). Whorls of rough endoplasmic reticulum (ER) and many small, dense, secretory granules fill out the rest of the cell profile. FC, Follicular cell; C, colloid. X20,OOO. FIG.15. Electron micrograph of a parafollicular cell from a bat caught in late October, just prior to the start of hibernation. At this time, the dense material in the cisternal space of the rough endoplasmic reticulum often appears to be undergoing dissolution, and autophagic-like figures sometimes enclose the ribosome-dotted membrane (arrow). Note that the small, dense, secretory granules exhibit a normal dense core. x 23,000.

50

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

51

suggested that the initial elevation of calcitonin in the thyroid gland at

the start of hibernation reflects retention of calcitonin within the parafollicular cells and a relative lack of parafollicular cell secretion (Haymovits et al., 1976).The subsequent slow decline in thyroid calcitonin level during hibernation probably reflects the intracellular removal of parafollicular cell secretory granules by autophagy. The reported hypocalcemia of hibernation (Haymovits et al., 1976; Riedesel, 1957) may be responsible for the apparent inactivity of parafollicular cells on the one hand, and on the other hand it may be responsible for the reported hyperactivity of parathyroid chief cells (Nunez et al., 197213) during hibernation. Thus secretion of parathormone, at a time when parafollicular cells are inactive, and at a time when there is no dietary supply of calcium, probably accounts for the increased skeletal resorption that occurs during hibernation (Haller and Zimmy, 1976; Whalen et al., 1972). Removal of calcium from the skeleton is probably required in order to maintain calcium homeostasis during this period of the life cycle. In summary, we postulate that parafollicular cells and parathyroid chief cells have reciprocal roles and act synergistically in the regulation of calcium metabolism during the seasonal life cycle of the mammalian hibernator. 6. Parturient Paresis There is evidence that calcitonin plays a major role in the pathogenesis of parturient paresis, a spontaneous metabolic disease of cattle. Parturient paresis is an afebrile, rapidly developing disease. It is characterized by the development of severe hypocalcemia and hypophosphatemia near the time of parturition and, in later stages, paralysis and coma. The disease is fatal if the affected cows are not treated with calcium solutions (Capen and Young, 1967a). Biochemical studies have reported that high circulating levels of plasma calcitonin occur near the time of parturition (Black and Capen, 1973), as do reduced levels of calcitonin in the thyroid glands of pregnant cows (Capen and Young, 196713; Young et al., 1972). Electron microscope studies by Capen and Young (1967a) and Young and Capen (1968) of parafollicular cells in thyroid glands from cows with parturient paresis showed PLATE XIII. FIG.18. Parafollicularcells (PC) from the thyroid gland of a hibemating bat obtained in late November. The cells are now characterized by the presence of large numbers of partially or totally degranulated secretory granules. Such granules (arrows) often exhibit membranous segments. The inset depicts several degranulated secretory granules at a higher magnification. x 19,OOO.

52

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

53

that parafollicular cells had discharged their secretory granules. On the basis ofthese studies, it has been concluded that the thyroid gland of diseased cows abruptly releases calcitonin at the time of parturition. If this is so, an abrupt release of calcitonin at this time is probably limited to dairy cows with parturient paresis, since in the ewe biochemical and electron microscope studies indicate that parafollicular cells are normal in appearance and the calcitonin level in the thyroid does not change at parturition (Collins and Barlet, 1972). It has been suggested that the release of a significant amount of calcitonin from parafollicular cells near parturition, rather than being an effect, is related to the cause of the hypocalcemia of parturient paresis (Capen and Young, 1967a). B. EXPERIMENTAL STUDIES In addition to hypercalcemic stimulation, the effects of several other agents on the fine structure of parafollicular cells have also been studied. However, the results of inany of these studies have not been without controversy. For example, Tashiro (1964) has reported that the administration of TSH does not alter the fine structure of parafollicular cells of the thyroid gland of the rat. In contrast, Bauer and Murad (1964) and Azzali (1964)have stated that, in the mouse and bat, respectively, TSH induces marked ultrastructural changes in parafollicular cells. Moreover, according to Bauer and Murad (1964),TSH administration causes the secretory granules to lose their dense contents. This results in the appearance of spherical, empty, membranelimited sacs. These studies should be repeated under more modern conditions of tissue preparation to eliminate the possibility that the changes resulted from fixation artifacts. Hypophysectoiny and diet-induced hypocalcemia have also been alternatively reported as having no effect at all on the fine structure of parafollicular cells (Kameda, 1974b; Matsuzawa, 1966; Matsuzawa and Kurosumi, 1967; Nunez et al., 1976) or as inducing striking (stimulatory) changes in parafollicular cells (Cameron, 1968; Moura et al., 1971; Young and Capen, 1971). Similarly, the administration of calcitonin to such species as the rabbit and rat has been reported to increase the number and size (Lupulescu, 1972) or to decrease the number (Bucher and PLATE XIV. FIG.19. Partially degranulated parafollicular cell (PC) from a hibernating bat caught in midhibernation (January).At this time, cells are characterized by the presence of prominent agranular whorls of cytoplasmic membranes (arrows), which enclose a inass of cytoplasmic material. FC, Follicular cell; C, colloid. x23,OOO. From Nunez et crl. (1969).

54

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

Krstib, 1974; Krstib et al., 1974) of secretory granules in the cytoplasm of parafollicular cells. Because of such equivocal results, some investigators have questioned the validity of such electron microscope studies (Lietz, 1971). It is possible that such different results are due to species differences or dosage variation. However, the difficulty may lie in the scarcity of parafollicular cells in most mammalian thyroid glands. This limits the number of cells that are encountered during the course of an electron microscope investigation. Other agents that have also been studied include sodium fluoride (Sundstrom, 1971),glucagon (Teitelbaum et al., 1972), cortisone (Gonticas et al., 1968), propylthiouracil (Moura et al., 1971), and disodium ethane-1-hydroxyl-1,l-diphosphonate (EHDP), a potent inhibitor of bone resorption (Yarrington et al., 1976). Only propylthiouracil and cortisone were found to have an effect on parafollicular cells and, in both cases, the drugs caused decreased granularity of the cells. The lack of an effect of glucagon on parafollicular cells may have been related to the dosage of drug used. In comparing the effect of glucagon to other secretogogues of calcitonin, such as gastrin and pentagastrin, Munson et aZ. (1974)reported that a much larger dose of glucagon still induced only a rather small and uncertain release of calcitonin from the thyroid gland of the pig.

IV. 5-HT In addition to the hormones thyroxine, triiodothyronine, and calcitonin, some mammalian thyroid glands also contain relatively large amounts of the biogenic amine, 5-HT (Erspamer, 1966). In mice and rats, 5-HT is stored in mast cells (Melander and Sundler, 1972; Nunez and Gershon, 1973). These mast cells have specialized storage granules which, by their uniform dense matrix, can be distinguished from the typical mast cell granules of other species (Nunez and Gershon, 1973). Since these specialized granules, but not typical mast cell granules, both contain and take up 5-HT, it seems likely that their structure reflects their involvement in 5-HT mechanisms. Some investigators have noted apparent responses of thyroid mast cells to TSH and have speculated on their involvement in mediating the release of thyroid hormones (Clayton and Szego, 1967; Melander et al., 1971b). In other mammalian species, such as the sheep, horse, and goat, 5HT has been observed by histofluorescence in parafollicular cells (Falck and Owman, 1968; Falck et al., 1964; Solcia and Sampietro, 1968). Recently, dopamine has been found in bovine parafollicular

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

55

cells (Melander et al., 1973), and both catecholamine and 5-HT in the parafollicular cells of callithricid primates (Machado, 1976). In sheep, parafollicular cell granules have been concentrated and partially purified by ultracentrifugation (Atack et al., 1972). These granules appear to contain both 5-HT and calcitonin. Ultrastructural observations on sheep thyroid support the view that both hormones are stored in parafollicular cells in the same subcellular storage organelles, the small, dense secretory granules. These are subsequently referred to as calcitonin granules. Ultrastructural cytochemistry, using the chromaffin reactivity of thyroid exposed to formaldehyde, reveals the presence of an endogenous indoleamine in the calcitonin granules of sheep parafollicular cells (Etcheverry and Zieher, 1968). Other mammals, with the exception of bats, to be discussed later, generally do not have high concentrations of thyroidal 5-HT. In these animals, 5-HT cannot be demonstrated in parafollicular cells by histofluorescence (Nunez and Gershon, 1973).This may be due to the relative insensitivity of the Falck-Hillarp technique for indoleamines. This technique is much less than its sensitivity for catecholamines (Fuxe and Jonsson, 1967). However, it is curious that, while endogenous levels of 5-HT in the parafollicular cells of many mammalian species are too low to be detected by histofluorescence, these cells in all species take up exogenously administered 5-HTP and convert it to 5-HT (see Section II,A,5). The plasma membrane of parafollicular cells does not permit these cells to take up 5-HT itself (Barith and Csaba, 1974; Gershon and Nunez, 1970, 1976). Thus parafollicular cells of all species must contain aromatic L-amino acid decarboxylase to convert 5-HTP to 5-HT. Since 5-HTP does not circulate in the blood (Gershon, 1978) and the cells do not take up 5-HT, the actual presence of the amine in the parafollicular cells depends on the activity in the cells of the enzyme tryptophan hydroxylase. This enzyme catalyzes the synthesis of 5-HT from L-tryptophan, but has not yet been studied in parafollicular cells. There is evidence that 5-HT synthesis may be prominent early in the development of parafollicular cells in species which, as adults, have low levels of thyroidal 5-HT. This is true of the dog, in which the concentration of 5-HT is high prior to birth and the amine is detectable in fetal parafollicular cells by histofluorescence (Gershon et al., 1971). The 5-HT level falls within a week of birth (Gershon et ul., 1971). At birth the thyroidal 5-HT concentration is about 3.5 pg/gm in beagles and 7pg/gm in basenjis and falls to adult values of 0.3 pg/gm in beagles and 0.6 pg/gm in basenjis. 5-HT cannot be detected by histofluorescence after its concentration has fallen.

56

ELADIO A. NUNEZ AND MICHAEL D. CERSHON

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

57

In animals synthesizing 5-HT from exogenously administered 5HTP in parafollicular cells, the amine, like the endogenous 5 H T of sheep parafollicular cells, appears to be localized in the calcitonin granules. This localization has been detected by Kalina and Pearse (1970), using ultrastructural examination of thyroid exposed to dichromate. It has also been confirmed by Ericson (1970, 1972a), using autoradiography. Interestingly, the location of 5-HT-H3 in the mouse parafollicular cell following the injection of 5-HTP-3H is the same in calcitonin granules from 5 minutes to 8 hours. Further confirmation of the location of 5-HT and calcitonin in the same granules comes from the physiological studies of Dahlstrom and Ericson (1972). They reported that, following the induction of hypercalcemia with vitamin D2, degranulation of parafollicular cells and a parallel loss of fluorescence of 5-HT occur in cells previously loaded by administration of exogenous 5-HTP. Bats are another group of mammals having a great quantity of thyroidal 5-HT (Gershon and Nunez, 1970; Nunez and Gershon, 1972). T h e thyroid gland of active Myotis lucifugus contains more than 4 pglgm of 5-HT. In bats mast cells do not contain 5-HT, but parafollicular cells d o (Gershon and Nunez, 1970). Bat parafollicular cells display a yellow fluorophore after treatment with formaldehyde vapor, which resembles 5-HT (Gershon and Nunez, 1970). In bats, the injection of tritiated 5-HTP leads to the labeling of parafollicular cell granules (Plate XV, Fig. 20) (Gershon and Nunez, 1973; Nunez and Gershon, 1972). However, the injection of 5-HT-3H does not label parafollicular cells at all (Gershon and Nunez, 1976) (Plate XVI, Fig. 21). The labeling of parafollicular cell granules after the injection of 5-HTP-3H is confined to the periphery of the granules, that is, the radioactive granules behave as hollow disks in autoradiography. Labeling from 10 minutes to 24 hours is confined to calcitonin granules (Nunez and Gershon, 1972). The bat is an excellent model for the investigation of serotonergic mechanisms in the thyroid, because experimental advantage can be taken of the phenomenon of hibernation. The bat fortunately contains more parafollicular cells per unit of thyroid mass than are found in the thyroids of most other mammals. Parafollicular cell morphology, and PLATE XV. FIG.20. Electron micrograph of active bat thyroid gland, 10 minutes after the injection of 5-HTP-H3. The silver grains (arrows) are located over the small, dense secretory granules of the parafollicular cell (PC).FC, Follicular cell; I, intercellular space. Inset illustrates large intracisternal granules. Such granules are never labeled. Note the silver grains over the small secretory granules. ~20,000.From Nunez and Gershon (1972).

58

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

59

probably activity as well, is synchronized during the yearly cycle of activity and hibernation (see Section 111,A75). Bats captured during the prehibernating period of their life cycle have been used for the study of parafollicular cell serotonergic mechanisms. This period has been used because at this time parafollicular cells are characterized by an extensive accumulation of granules of two types, calcitonin granules, and intracisternal (within rough endoplasmic reticulum) granules. They also contain a well-developed machinery for protein synthesis and export (Nunez et ul., 1967b). The calcitonin granules are labeled following injection of 5-HTP-3H, but intracisternal granules are not (Nunez and Gershon, 1972) (Plate XV, Fig. 20, inset). This observation has led to the postulate that parafollicular cell granules derive from the Golgi apparatus the ability to take up 5 H T (Nunez and Gershon, 1972). The reasoning behind this postulate is as follows. If one assumes protein synthesis for export by parafollicular cells is similar to that in other secretory cells such as those of the exocrine pancreas (Jamieson and Palade, 1967a,b), it is likely that the polypeptide chain destined for secretion is synthesized on attached ribosomes and transferred to the cisternal space of the rough endoplasmic reticulum. Intracisternal granules probably represent this material, accumulating in the lumen of the rough endoplasmic reticulum because of interference with transport to the Golgi apparatus. Calcitonin granules, surrounded b y a smooth membrane, probably represent a post-Golgi stage in the intracellular transport of this material. Since aromatic Lamino acid decarboxylase is a soluble enzyme, 5-HT must be synthesized from 5-HTP in the cytosol. Its early and persistent localization in calcitonin granules (Ericson, 1972a; Nunez and Gershon, 1972) indicates that these granules must take up the amine. Since the enzyme that catabolizes 5-HT, monoamine oxidase, coexists in parafollicular cells with the amine (Larson et ul., 1966; Nunez and Gershon, 1972; Tjalve, 1971), this uptake must be necessary to preserve the 5-HT, since free amine in the cytosol would be catabolized. The fact that intracisternal granules do not take up 5-HT indicates that this uptake must be associated with post-Golgi structures. In light of the observations outlined above, indicating that calcitonin and 5-HT are stored in the same subcellular granules in parafolPLATE XVI. FIG.21. Electron microscope autoradiograph of bat thyroid gland incubated with 5HT-3H (9 x lo-' M ) . Silver grains (arrows)are routinely found within the follicular basement membrane (arrowheads), mostly over follicular cells (FC) and luminal colloid (C). Silver grains are never found over parafollicular cells (PC). V, Blood vessel; I, intercellular space. x 13,000.

60

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

licular cells, it is not surprising that drugs such as reserpine, known to affect storage of biogenic amines (Shore, 1962), also affect these granules. Reserpine, given acutely or chronically to bats, causes calcitonin granules to change shape and also alters the morphology of the granular matrix. The granules are normally nearly spherical prolate elipsoids with a uniformly electron-opaque inner matrix (Gershon and Nunez, 1973). After the injection of reserpine the axial ratio of the ellipsoidal granules increases greatly, and a faint internal striation parallel to the long axis of the granules becomes apparent (see Plate XVII, Fig. 22). This effect of resperine is probably not simply related to the loss of 5-HT that follows administration of the drug. Reserpine depletes 5-HT by interfering with its storage in granules, leading to intracellular release of the amine with subsequent metabolism by monoamine oxidase (Carlsson, 1966).5-HT in the thyroid can also be depleted by administering an inhibitor of 5-HT biosynthesis, para-

PLATE XVII. FIG.22. Electron micrograph of a bat parafollicular cell 24 hours after the injection of reserpine (5 mg/kg). Note that reserpine induced striking changes in the shape of the secretory granules (arrowheads),causing many to change from a sphere to a prolate ellipsoid. x 30,000.

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

61

chlorophenylalanine (PCPA). When PCPA is given to bats in amounts adequate to deplete 5-HT as much as doses of reserpine that affect calcitonin granules, there is no corresponding change in granular matrix or shape (Gershon and Nunez, 1973).Thus the effect of reserpine may reflect the drug’s action on 5-HT binding by granules but is not simply indicative of loss of the amine itself. The action of reserpine on calcitonin level or storage has not yet been determined. Reserpine blocks labeling of parafollicular cell granules by 5-HT-3H following the injection of 5-HTP-3H into bats (Gershon and Nunez, 1973) or mice (Ericson, 1972a), however, PCPA does not (Gershon and Nunez, 1973). Further indication of the probable storage of the two hormones in the same granules comes from studies on the changes in concentration of 5-HT and calcitonin in the bat thyroid as a function of the yearly cycle of activity and hibernation (Haymovits et al., 1976).The concentrations of both hormones rise and fall together. Early in hibernation both peak to about three times the prehibernating level. This is followed by a steady decline until arousal. The rise early in hibernation may reflect retention or lack of secretion of hormone, and the decline late in hibernation may be due to autophagy of calcitonin granules (see Section III,A,5), The role of 5-HT in the thyroid has not yet been determined. Melander and Sundler (1972) and Melander and associates (1972,1974b) showed that 5-HT activates follicular cells. The amine causes these cells to extend pseudopods into the colloid and take in colloid by endocytosis, initiating the secretion of thyroid hormone. Clayton and Szego (1967) postulated that 5-HT is an intermediate in the acute activation of follicular cells by TSH. In rats and mice, mast cells may be the source of 5-HT. In sheep, horses, goats, and bats, if not in other species, parafollicular cells are probably the only 5-HT source (Falck and Owman, 1968; Gershon and Nunez, 1970; Solcia and Sampietro, 1968). It is unlikely that thyroidal 5-HT acts on a distant target organ as calcitonin does. There are too many mechanisms to prevent the circulation of free, active 5-HT. These mechanisms include uptake by platelets, uptake by pulmonary endothelial cells, and uptake and metabolism by the liver (Gershon, 1978). Thus the action of thyroidal 5HT, if there is any action, must be local. The storage of 5-HT and calcitonin together in the same subcellular granules seems to ensure that both hormones will be released together by exocytosis on presentation of the appropriate stimulus. There is evidence indicating that this occurs. Following the injection of 5-HTP-3H to label 5-HT-3H in parafollic-

62

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

ular cells, no label appears over follicular cells for 24 hours. After this time grains begin to appear in the follicular cells (Plate XVIII, Fig. 23), suggesting that these cells take up 5 H T released from parafollicular cells, although follicular cells are unable to synthesize the amine itself (Nunez and Gershon, 1972). If thyroids are loaded with 5-HT-3H and then incubated in media containing a greater-than-normal concentration of calcium, the calcium induces a release of 5-HT-3H (Kanarek et al., 1976). Thus the normal stimulus for calcitonin secretion, calcium, is also an excellent stimulus for 5-HT secretion. This result is what would be predicted if both hormones were released simultaneously by exocytosis. These observations have led to the hypothesis that 5-HT is a parafollicular-to-follicularcell messenger. The amine may thus provide a link between blood calcium concentration and metabolic regulation via thyroid hormone. If 5-HT plays this role, the ability of follicular cells to take up 5-HT (see Plate XVI, Fig. 21) (Ger-

PLATE XVIII. FIG.23. Electron micrograph of an active bat thyroid 21 hours after the injection of 5-HTP-3H. Autoradiographic labeling (arrows) is now found routinely over follicular cells (FC) and colloid (C). Grains are still also found over parafollicular cells (PC). ~20,000.From Nunez and Gershon (1972).

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

63

shon and Nunez, 1976) may be related to the action of 5-HT. If 5 H T acts on receptors on the surface of follicular cells, the uptake of 5-HT by the cells may be a mechanism for inactivating the amine once it has reached and combined with its receptor. Alternatively, 5-HT may have to enter follicular cells to activate them. If so, the uptake mechanism may provide a means of entry for 5-HT into these cells. The administration of thiouracil and calcitonin has been reported to interfere with the disappearance of exogenous 5-HT from the parafollicular cells of the mouse. It has been suggested that this effect may be due to a direct link with reduced release of calcitonin from these cells (Owman and Sundler, 1968). If 5-HT and calcitonin are stored in the same granules, this link of S H T secretion to calcitonin should be expected. The administration of exogenous 5-HTP in a dosage of 25 mg/kg was shown by Welsch and associates (1968) to increase the volume of the rough-surfaced endoplasmic reticulum and the density of the nucleus of guinea pig parafollicular cells. These investigators concluded that such changes indicate that the parafollicular cells undergo a period of enhanced protein synthesis in response to stimulation by injected amine precursor.

V. Tryptophyl Peptides It has also been reported that parafollicular cells of the thyroid glands of the cat and pig exhibit a strong greenish-yellow fluorescence following treatment of the thyroid with formaldehyde and ozone (Hikanson et al., 1972). Microspectrofluorometric and chemical analysis indicates that the fluorogenic compound may be a tryptophyl peptide (Hikanson et al., 1972). The significance of tryptophyl peptides in parafollicular cell function is still not clear.

VI. Medullary Carcinoma of the Thyroid Medullary carcinoma of the thyroid gland is a human tumor comprising between 5 and 10% of all thyroid carcinomas (Tashjian et al., 1974).There are two types of occurrence of the tumor. One is sporadic or spontaneous, in which it is usually unilateral. The other is familial, in which the neoplasm is inherited as an autosomal dominant trait (Tashjian et al., 1974).Familial tumors are usually bilateral and multicentric. Medullary carcinoma of the thyroid is often associated with tumors of other endocrine organs such as pheochromocytoma (Sipple, 1961). The syndrome of multiple endocrine tumors associated with medullary carcinoma of the thyroid is termed Sipple’s syndrome

64

ELADIO A. NUNEZ AND MICHAEL D . GERSHON

(Keiser et al., 1973). It has been proposed that each of the tumors represents a clone derived from a single malignant cell (Baylin et al., 1976). Histologically, medullary carcinoma of the thyroid is composed of nests and cords of epithelial cells usually surrounded by an amyloidcontaining stroma (Hazard et al., 1955). The tumor cells typically exhibit a solid homogeneous pattern without papillary or follicular differentiation (Gordon et al., 1973). Amyloid deposition is whorled or forms an aggregated pattern (Hazard et al., 1955).On the basis of comparative histological morphology, Williams (1966) and Williams et al. (1966) concluded that medullary carcinoma of the human thyroid gland originates through neoplastic transformation of the parafollicular cells. Firm confirmatory evidence, outlined in the following discussion, has now clearly shown that medullary carcinoma of the human thyroid is indeed a tumor of the parafollicular cells in which there is excessive secretion of calcitonin. A. HISTOCHEMICAL EVIDENCE Tumor cells of medullary carcinoma of the human thyroid, like normal parafollicular cells, exhibit masked metachromasia and argyrophilia (DeLellis and Balogh, 1973). Moreover, calcitonin has been shown immunocytochemically to be present in the cytoplasm of most, but not all, of the tumor cells (Bussolati et al., 1969a; Wolfe et al., 1973).Ljungberg (1970b,c, 1972) and Ljungberg and Dymeling (1972) have reported that, in addition to the principal round “parafollicular” tumor cells, medullary carcinomas of the thyroid also contain a second type of cell. These additional cells are spider-shaped and are characterized by positive argentaffin and chromaffin reactions. Parafollicular cell hyperplasia has been found in several family members of patients having familial medullary carcinoma (Wolfe et al.,1973). This hyperplasia of parafollicular cells is not associated with hyperparathyroidism. It is thought to be an early change and to precede the development of medullary carcinoma.

B. BIOCHEMICALEVIDENCE The calcitonin content of tissue from medullary carcinoma of the thyroid is highly elevated, 100 to 30,000 times higher than that found in normal adult thyroid tissue (Dub6 et al., 1969; Meyer and AbdelBari, 1968; Tashjian et al., 1974; Voelkel et al., 1973). Moreover, patients with clinically active medullary carcinoma have a high concentration of calcitonin in their serum (Bussolati et al., 1969a; Keiser et aZ., 1973; Melvin et al., 1972), and tumor parafollicular cell secretion

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

65

can be stimulated by calcium infusion (Melvin et al., 1972). In this regard, measurement of serum calcitonin either under basal conditions or during calcium infusion has been shown to be the best diagnostic test for the tumor (Keiser et al., 1973; Melvin et al., 1972). Moreover, medullary carcinoma of the thyroid not only produces calcitonin in vivo but also in vitro (Gautvik and Tashjian, 1974; Grimley et al., 1969; Roos et al., 1975; Tashjian and Melvin, 1968). For example, tumor tissue in organ culture secretes calcitonin for up to several weeks and also responds with calcitonin secretion to increased extracellular calcium (Gautvik and Tashjian, 1974). As might have been predicted from studies of parafollicular cells of other species (see Section IV), medullary carcinoma cells contain and elaborate 5-HT (Beskid and Lorenc, 1970; Grimley et al., 1969; Ljungberg et al., 1967; Moertel et al., 1965).

C. ELECTRONMICROSCOPE EVIDENCE The tumor cells of medullary carcinomas are ultrastructurally similar to mammalian calcitonin-secreting parafollicular cells and, like normal parafollicular cells, are characterized by the presence of numerous calcitonin granules in the cytoplasmic matrix (Braunstein et al., 1968; Gonzalez-Licea et al., 1968; Grimley et al., 1969; Lietz and Donath, 1970b; Meyer, 1968; Meyer and Abdel-Bari, 1968; Tateiski et al., 1972; Thliveris et al., 1976). In addition, examination of argyrophilia electron microscopically has shown the localization of a dense precipitation of silver over the calcitonin granules of tumor cells (Tateiski et al., 1972). A second type of tumor cell has also been reported by Bordi et al. (1972) at the ultrastructural level, This type of cell differs from tumor cells that resemble parafollicular cells in containing much larger and more dense granules, and may correspond to the spider-shaped argentaffin cell described above. The presence of a second type of tumor cell is not at all surprising, in view of the fact that medullary carcinomas of the thyroid can secrete several humoral agents. For example, in addition to calcitonin and serotonin, it has been reported that medullary carcinomas of the thyroid secrete prostaglandins (Levin et al., 1973; Williams et al., 1968), histamine (Baylin et al., 1970), ACTH (Donahower et al., 1968), and possibly epinephrine and norepinephrine (Voelkel et al., 1973). Medullary carcinoma of the thyroid also occurs in old rats (Boorman, 1976; Boorman et al., 1972) and aged bulls (Capen and Black, 1974; Wilkie and Krook, 1970). The ultrastructural demonstration of tumor cells loaded with calcitonin granules in the rat (Boorman, 1976)

66

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

and bull (Black et al., 1973a),in addition to the observation that these neoplastic cells contain immunoreactive calcitonin (Bollman and Pearse, 1974),shows that the animal tumors resemble the corresponding human tumor and support the hypothesis that the parafollicular cell is the cell of origin of medullary carcinoma of the thyroid. It is not yet certain, however, that the amyloid of medullary carcinoma is derived from medullary tumor cells. There is histochemical and electron microscope evidence that it is. Ultrastructurally, fibrils similar to those of amyloid have been found within the cytoplasm of medullary carcinoma cells (Amouroux et al., 1970; Gonzalez-Licea et al., 1968; Huang and McLeish, 1968; Meyer, 1968; Meyer et al., 1973) and, on the basis of histochemical reactions, Pearse et al. (1972) and others (Albores-Saavedra et al., 1964; Beskid, 1964) have proposed that the formation of amyloid results from the production of a precursor substance (a procalcitonin molecule) within the neoplastic parafollicular cell. However, there are also strong opinions that amyloid is produced by stromal cells such as fibroblasts (Grimley et al., 1969; Lietz and Donath, 1970b; Thliveris et al., 1976). Perhaps both neoplastic parafollicular cells and stromal cells are involved in amyloid formation (Bordi et al., 1972).

VII. Parafollicular Cells in Nonmalignant Human Diseases It has been reported that toxic goiter (see Section II,G,2) and chronic hypercalcemia of primary hyperparathyroidism (Donnay et al., 1974; Kracht et al., 1970; Ljungberg and Dymeling, 1972)produce parafollicular cell hyperplasia. Moreover, Ljungberg and Dymeling (1972)have proposed that chronic hypercalcemia is an etiological factor in human parafollicular cell neoplasm, including some forms of medullary carcinoma. Several reports of parafollicular cell adenomata in the thyroid have also appeared in the literature (Beskid et al., 1971; Beskid and Kobuszewska-Faryna, 1972; Zaridze, 1974). However, according to Roediger (1975), the features of these cases do not conform to those of true thyroid adenomata, showing no evidence of encapsulation or compression of the adjacent thyroid. It is also interesting to note that examination of the thyroid glands of infants with congenital osteopetrosis has so far failed to demonstrate increased numbers of parafollicular cells in the thyroid gland (Morrow et al., 1967; Solcia et al., 1968a).Thus it is possible that the postulated role of parafollicular cells in the pathogenesis of osteopetrosis in mice (see Section I1,F) is not necessarily the same as in the human. However, it is also possible

CYTOPHYSIOLOGY O F THYROID PARAFOLLICULAR CELLS

67

that inadequate sampling of tissue was responsible for the observed differences. The entire thyroid gland of humans with osteopetrosis has not yet been examined quantitatively for parafollicular cells. In summary, although the exact role of parafollicular cells in human thyroid disease is still unclear, there is now sufficient evidence that warrants further investigation of these unique cells in various thyroid disease processes. VIJI.

Conclusion

Work done to date on parafollicular cells supports the following conclusions about these cells.

1. Parafollicular cells are clearly different functionally, morphologically, and developmentally from follicular cells. 2. Parafollicular cells lie within the follicular basement membrane, and so, together with follicular cells, constitute a “follicular apparatus.” 3. Parafollicular cells synthesize, store, and secrete calcitonin. 4. Hypercalcemia is a physiological stimulus for these cells to secrete. 5. In at least some species parafollicular cells also store biogenic amines, particularly 5-HT. When present, 5-HT is stored in the same specific storage granules as calcitonin. 6. Medullary carcinoma of the thyroid is a tumor of the parafollicular cells. Many experimental studies have tried to define the role played by parafollicular cells in the normal economy of the body. This role and that of calcitonin remain unclear. Suggested roles include an involvement in the control of bone modeling and calcium homeostasis during development, prevention of hypercalcemia, and maintenance of blood calcium during hibernation in hibernating animals. Similarly, the reason, if there is one, for the close relationship between follicular and parafollicular cells has not been found. The presence of biogenic amines in some parafollicular cells has led to the hypothesis that parafollicular cells communicate with follicular cells by means of chemical signals. The storage of 5-HT with calcitonin ensures that both hormones will be released together, on appropriate stimulation, by exocytosis. One, calcitonin, may be destined to act on a distant target, bone; the other, 5-HT, may be destined to act locally on follicular cells. This hypothesis remains to be established.

68

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

In summary, much is now known about what parafollicular cells contain and about what they may do. Future investigations should move in the direction of telling us what they do. The availability of many reliable light and electron microscope methods of demonstrating parafollicular cells should lead to rapid disappearance of the cloud of mystery that has surrounded these cells since their discovery. ACKNOWLEDGMENTS The authors express their gratitude to Mr. John Patrikes for his expert technical assistance. This study was supported by United States Public Health Service grants AM19743 and NS12969.

REFERENCES Adams, P. H., Jowsey, J., Kelly, P. J., Riggs, B. L., Kenny, V. R., and Jones, J. D. (1967). Q.]. Med. 36, 1. Albores-Saavedra, J., Rose, G. G., Ibanez, M. L., Russel, W. O., Grey, C. E., and Dmochowski, L. (1964). Lab. Inuest. 13,77. Aliapoulios, M. A., and Munson, P. L. (1965).Surg. Forum 16, 55. Aliapoulios, M. A., and Rose, E. H. (1970). Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 295. Aliapoulios, M. A., Goldhaber, P., and Munson, P. L. (1966a). Science 151,330. Aliapoulios, M. A., Voelkel, F. F., and Munson, P. L. (1966b). J . Clin. Endocrinol. Metab. 26,897. A h a , E. (1954). Monit. Zool. ltal. 62, Suppl., 46. Altman, H. W. (1940).Beitr. Pathol. Anat. Allg. Pathol. 104,420. Amouroux, J., Bernier, J. J., Ramboud, J. C., Dupuy-Coin, A. M., and Kalifat, S. R. (1970).Z . Krebsforsch. 74, 122. Ansari, P. M. (1967). Beitr. Pathol. Anat. Allg. Pathol. 135,411. Aoi, T. (1966). Okajmas Folia Anat. Jpn. 42,63. Arimitsu, K. (1937). Norm. Befund. Kaibo 2. 10,233. Atack, C. V., Ericson, L. E., and Melander, A. (1972).]. Ultrastruct. Res. 41,484. Avioli, L., Birge, S. J., Scott, S., and Shieber, W. (1969).Am. 1. Physiol. 216,939. Axelrod, A. A., and Leblond, C. P. (1955). Cancer 8,329. Azzali, G. (1962), Boll. SOC. ltal. Biol. Sper. 38, 1319. Azzali, G. (1964). Ann. Endocrinol. 25, Suppl. 8. Azzali, G. (1966). Ateneo Parmense, Acta Bio-Med. 37, 569. Azzali, G. (1967). Ateneo Parmense, Acta Bio-Med. 38, 5. Azzali, G. (1968). Calcitonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 152. Baber, E. C. (1876). Philos. Trans. R . SOC. London, Ser. B 166,557. BarPth, P., and Csaba, Gy. (1974).Acta Morphol. Acad. Sci. Hung. 22,327. Bauer, W. C., and Murad, T. M. (1964). Lab. Inuest. 13,942. Bauer, W. C., and Teitelbaum, S. L. (1966).Lab. Invest. 15,323. Baylin, S . B., Beaven, M. A., Engleman, K., and Sjoerdsma, A. (1970). N. Engl. 1. Med. 283, 1239.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR

CELLS

69

Baylin, S. B., Gann, D. S., and Hsu, S. H. (1976). Science 193,321. Bensley, R. R. (1914).Anut. Rec. 8,431. Bergman, L., Kjellrner, I., and Selstarn, V. (1973). B i d . Neonute 24, 151. Beskid, M. (1964).Foliu Histochem. Cytochem. 2,261. Beskid, M., and Kobuszewska-Faryna, M. (1972). Folis Histochem. Cyctochem. 10,31. Beskid, M., and Lorenc, R. (1970).Actu Histochem. 38, 74. Beskid, M., and Rosciszewska, A. (1968). Folia Histochem. Cytochem. 6,461. Beskid, M., Lewicki, Z . , and Rosciszewska, A. (1968). Pol. Arch. Med. Wewn. 41,587. Beskid, M., Lewicki, Z., and Rosciszewska, A.'(1970).Pol. Arch. Med. Wewn. 44, 13. Beskid, M., Lorenc, R., and Rosciszewska, A. (1971).]. Puthol. 103, 1. Biddulph, D. M. (1968).Anat. Rec. 160,316. Biddulph, D. M., and Maibenco, H. C. (1972). Anut. Rec. 173,25. Biorck, G., Johansson, B., and Veige, S. (1956).Actu Physiol. Scund. 37,281. Birkenhager, J. C., Upton, G. V., Seldenrath, H. J., Krieger, D. T., and Tashjian, A. H., Jr. (1976).Actu Endocrinol. (Copenhagen) 83,280. Birov, V. V. (1971).Cytohgiu 13, 1178. Black, H. E., and Capen, C. C. (1973).Horn. Metub. Res. 5, 297. Black, H. E., Capen, C. C., and Young, D. M. (1973a). Cancer 32.865. Black, H. E., Capen, C. C., Yarrington, J. T., and Rowland, G. N. (1973b). Lob. Invest. 29.437. BlahoSovB, A., NeradilovQ, A., Velickq, J., Titlbach, M., Marsikova, and Reisenauer, R. (1974). Endokrinologie 63, 122. Blahser, U. S., and Kraus, G. (1972). Mikroskopie 28,288. Blihser, U. S., and Schnorr, B. (1972).2. Zellforsch. Mikrosk. Anut. 134, 13. Bollrnan, R., and Pearse, A. G. E. (1974). Virchows Arch. B 15, 95. Boorman, G. A. (1976).A m . ] . Puthol. 83,237. Boorman, G. A., van Noord, M. J., and Hollander, C. F. (1972). Arch. Pathol. 94,35. Bordi, C., Anversa, P., and Vitali-Mazza, L. (1972). Virchows Arch. A 357, 145. Boulet, L. P., and Pellelier, G. (1974).Union Mecl. Can. 103,440. Bowers, C. Y., Schally, A. V., Enzmann, F., Bdler, J., and Folkers, K. (1970).Endocrinology 86, 1143. Braunstein, H., and Stephens, C. L. (1968). Arch. Puthol. 86, 659. Braunstein, H., Stephens, C. L., and Gibson, R. L. (1968).Arch. Pathol. 85,306. Brewer, H. B., Jr., and Ronan, R. (1969).Proc. Natl. Acad. Sci. U.S.A. 63, 940. Broulik, P. D., and Pacovskq, V. (1973).Endocrinol. Exp. 7, 183. Bruce, B. J. B., and Care, A. D. (1969).]. Endocrinol. 46, XI. Bucher, O., and Krstii., R. (1974). Verh. Anut. Ges., Versamml., 68th, 1974 p. 799. Bussolati, G., and Monga, G. (1970). Experientia 26,881. Bussolati, G., and Pearse, A. G. E. (1967).J. Endocrinol. 37,205. Bussolati, G., Foster, G. V., Clark, M. B., and Pearse, A. G. E. (1969a).Virchows Arch. B 2,234. Bussolati, G., Navone, R., Gasparri, G., and Monga, G. (1969b). Experientia 25,641. Bussolati, G., Rost, F. W. D., and Pearse, A. G. E. ( 1 9 6 9 ~ )Histochem. . J . 1,517. Bussolati, G., Monga, G., Navone, R., and Gasparri, G. (1970). Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 240. Calvert, R. (1972).Am. ]. Anat. 133,269. Calvert, R. (1974).Anut. Rec. 180,663. Calvert, R., and Isler, H. (1970).Anat. Rec. 168,23. Cameron, D. A. ( 1968).Puruthyroid H o r n . Thyroculcitonin (Calcitonin),Proc. Paruthyroid Conf, 37-4 1967 Excerpta Med. Int. Congr. Ser. No. 159, p. 437.

70

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

Capen, C. C., and Black, H. E. (1974). Am. J . Pathol. 74,377. Capen, C. C., and Young, D. M. (1967a).Lab. Invest. 17,717. Capen, C. C., and Young, D. M. (1967b). Science 157,205. Capen, C. C., and Young, D. M. (1969). Am. J . Pathol. 57,365. Care, A. D., Duncan, J., and Webster, D. (1966).J.Endocrinol. 35, xxvii-xxviii. Carlsson, A. (1966). Handb. Erp. Pharmakol. 19,529. Carvalheira, A. F., and Pearse, A. G. E. (1967a).Histochemie 7, 175. Carvalheira, A. F., and Pearse, A. G. E. (196713). J . R. Microsc. SOC. 86,20. Carvalheira, A. F., and Pearse, A. G. E. (1968). Calcitonin.Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 122. Chan, A. S. (1972). Anat. Rec. 172,287. Chan, A. S., and BBlanger, L. F. (1968). Calcitonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 192. Chan, A. S., and Conen, P. E. (1971). Lab. Invest. 25,249. Christov, K., Bollman, R., and Thomas, C. (1972). Beitr. Pathol. 147,341. Clayton, J. A., and Szego, C. M. (1967). Endocrinology 80,689. Collins, J. P., and Barlet, J. P. (1972). Z . Zellforsch. Mikrosk. Anat. 126,363. Cooper, C. W., and Tashjian, A. H., Jr. (1966). Endocrinology 79,819. Cooper, C. W., Schwesinger, W. H., Mahgoub,A. M., and Ontjes, D. A. (1971). Science 172, 1238. Cooper, C. W., Schwesinger, W. H., Mahgoub, A. M., and Munson, P. L. (1972).Endocrinology 91, 1079. Copp, D. H., Cameron, E. C., Cheney, B. A., Davidson, A. G. F., and Henze, K. G. (1962). Endocrinology 70,638. Copp, D. H., Cockcroft, D. W., and Kueh, Y. (1967). Science 158,924. Csaba, G., and BorAth, P. (1974). Acta Anat. 88, 137. Cunliffe, W. J. (1961).Acta Anat. 146, 135. Cutler, G. B., Habener, J. F., and Potts, J. T., Jr. (1974).Gen. Comp. Endocrinol. 24,183. Dahlstrom, A., and Ericson, L. E. (1972). Z . Zellforsch. Mikrosk. Anat. 128,406. DeGrandi, P. B. (1970). Virchows Arch. B 6, 137. DeGrandi, P. B., Kraehenbuhl, J. P., and Campiche, M. A. (1971).J . Cell Biol. 50,446. Dejardin, M. (1955). C . R. Seances SOC. Biol. Ses Fil. 149,621. DeLellis, R. A., and Balogh, K. (1973). Am. J . Pathol. 72, 119. DeLellis, R. A., Wolfe, H. J., and Ucci, A. A. (1974). Clin. J . Pathol. 74,23a. Dempsey, E. W. (1954).In “Histology” (R. 0.Greep, ed.), p. 810. McGraw-Hill (Blakiston), New York. Dencker, L., and Tjalve, H. (1973). Experientia 29,719. Dim-Flores, L., Mora, 0. A., and Marin, B. (1972). Erperientia 28,946. Donahower, G. F., Schumacher, 0. P., and Hazard, J. B. (1968). J . Clin. Endocrinol. Metab. 28, 1199. Donnay, M., Cosyn, J. P., and Hustin, J. (1974).Ann. Anat. Pathol. 19,163. Dorrenhaus, A., Kohler, H., and Luciano, L. (1971).Z . Zellforsch. Mikrosk. Anat. 112, 247. Doyle, F. H., Pennock, J,, Greenberg, P. B., Joplin, G. F., and MacIntyre, I. (1974). Endocrinol. Proc. lnt. Symp., 4th, 1973 p. 425. DubB, W. J., Bell, G. O., and Aliapoulios, M. A. (1969).Arch. Intern. Med. 123,423. Dubois, P., and Dumont, L. (1966). C . R. Seances SOC. Biol. Ses Fil. 160,2331. Dumont, L. (1956a).Ann. Endocrinol. 17,700. Dumont, L. (1956b). C . R. Assoc. Anat. 3,480.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

71

Dumont, L. (1958).C. R. Seances SOC. Biol. Ses Fil. 152,790. Editorial. (1973). Lancet 2, 771. Ehrenbrand, F. (1954).2. Mikrosk.-Anat. Forsch. 60,337. Ekholm, R., and Ericson, L. E. (1968).J . Ultrastruct. Res. 23, 378. Ekholm, R., and Smeds, S. (1966).J. Ultrastruct. Res. 16,71. Englund, N. E., Nilsson, G., Owman, C., and Sundler, F. (1972).J . Clin. Endocrinol. Metab. 35, 90. Ericson, L. E. (1968).J. Ultrastruct. Res. 24,145. Ericson, L. E. (1970).J. Ultrastruct. Res. 31, 162. Ericson, L. E. (1972a).J. Ultrastruct. Res. 41,467. Ericson, L. E. (1972b).Z. Zellforsch. Mikrosk. Anat. 126, 182. Erspamer, V. (1966).Handb. Exp. Pharmakol. 19,245. Etcheverry, J. G., and Zieher, L. M. (1968).Erperientia 24,593. Falck, B., and Owman, C. (1968).Ado. Pharmacol. 6a, 211. Falck, B., Larson, B., Mecklenburg, C. V., Rosengren, E., and Svenaeus, K. (1964).Acta Physiol. Scand. 62,491. Farquhar, M. G. (1971).Mem. SOC. Endocrinol. 19, 109. Feinblatt, J. D., and Raisz, L. G. (1971).Endocrinology 88,797. Fetter, A. W., and Capen, C. C. (1970).Pathol. Vet. 7,171. Fisher, D. A., and Dussault, J. H. (1974).Handb. Physiol., Sect. 7: Endocrinol. 3,21. Fisher, J. A., Blum, J. W., and Binswanger, U. (1973).Clin. Res. 21,623. Fitzgerald, M. J. T. (1964).Q. J. Microsc. Sci. [N.S.] 105,359. Fontaine, J. (1974).Arch. Anat. Microsc. Morphol. Erp. 63,217. Fortney, J. A. (1973).Anat. Rec. 175,321. Foster, G. V., MacIntyre, I., and Pearse, A. G. E. (1964a).Nature (London) 203, 1029. Foster, G. V., Baghdiantz, A., Kumar, M. A., Slack, E., Soliman, H. A., and MacIntyre, I. (1964b).Nature (London)203, 1303. Friedman, J., and Raisz, L. G. (1965). Science 150, 1465. Friedman, N. (1949).J. Clin. Endocrinol. Metab. 9,874. Fujita, H . (1975). Int. Reo. Cytol. 40, 197. Fuxe, K., and Jonsson, G. (1967).Histochemie 11, 161. Gabe, M. (1959).Actu Anat. 38,332. Gabe, M. (1961).Acta Anat. 47,34. Gabe, M., and Martoja, M. M. (1969).Arch. Anat. Microsc. Morphol. E x p . 58, 105. Garel, J. M., Care, A. D., and Barlet, J. P. (1974).J . Endocrinol. 62,497. Garel, J. M., Barlet, J. P., and Kervran, A. (1975).Am. J . Physiol. 229,669. Gauguin, J., Rbmert, P., and Kirkeby, S. (1973).Histochemie 34,97. Gautvik, K . M., and Tashjian, A. H., Jr. (1974).H o r n . Metab. Res. 6, 70. Gershon, M. D. (1978).Handb. New. Syst. 1 (in press). Gershon, M. D., and Nunez, E. A. (1970).Endocrinology 86, 160. Gershon, M. D., and Nunez, E. A. (1973).J. Cell Biol. 56, 676. Gershon, M. D., and Nunez, E. A. (1976).Mol. Cell. Endocrinol. 5, 164. Gershon, M. D., and Ross, L. L. (1966a).J . Physiol. (London) 186,451. Gershon, M. D., and Ross, L. L. (1966b).J . Physiol. (London) 186,477. Gershon, M. D., Belshaw, B. E., and Nunez, E. A. (1971).Am. J . Anat. 132,5. Getzowa, S . (1907).Virchows Arch. Pathol. Anat. Physiol. 188, 181. Gitelman, H. J., Kukolj, S., and Welt, L. G. (1968).J. Clin. Invest. 47, 118. Gittes, R. F., Toverud, S. U., and Cooper, C. W. (1968).Endocrinology 82,83. Godwin, M. C. (1937).Am. J . Anat. 60,299

72

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

Gonticas, S. C., Braams, W. G., and Doorenbos, H. (1968). Lancet 1,923. Gonzalez-Licea, A., Hartmann, W. H., and Yardley, J. H. (1968).Am. J . Clin. Pathol. 49, 512. Gordon, P. R., Huvos, A. G., and Strong, E. W. (1973). Cancer 31,915. Greenberg, P. B., Martin, T. J., and Melick, R. A. (1971). Nature (London)233,201. Grimley, P. M., Deftos, L. J., Weeks, J. R., and Robson, A. S. ( 1 9 6 9 ) Natl. ~ Cancer Znst. 42,663. Gross, J., and Pitt-Rivers, R. (1952). Lancet 1,439. Grossi, A., and Servide, E. (1961). Biol. Lat. 13,253. Guglielmone, R. (1971). Mikroskopie 27,342. Habener, J. F., and Potts, J. T., Jr. (1976).Endocrinology 98, 197. Hachmeister, U., Kracht, J., and Zimmermann, H. D. (1969).Acta Endocrinol. (Copenhagen), Suppl. 138, 180. Hlkanson, R., Owman, C., and Sundler, F. (1971a). Biochem. Pharmacol. 20,2187. Hikanson, R., Owman, C., Sporrong, B., and Sundler, F. (1971b). Histochemie 27,226. Hlkanson, R., Owman, C., and Sundler, F. (1972).J.Histochem. Cytochem. 20,205. Hlkanson, R., Melander, A., Owman, C., and Sundler, F. (1973).Histochemie 36,89. Haller, A. C., and Zimmy, M. L. (1976). Anat. Rec. 184,421. Hamperl, H. (1937). Virchows Arch. A 298,327. Hamperl, H. (1962). Virchows Arch. Pathol. Anat. Physiol. 335,452. Harcourt-Webster, J., and Stott, N. (1966).J . Pathol. Bacteriol. 92,291. Harington, C. R. (1926). Biochem. J . 20,300. Haymovits, A., and Rosen, J. F. (1972). Ado. Metab. Disord. 6, 177. Haymovits, A., Wright, M., Ling, A,, Hobitz, H., and Pryog, B. (1975).Clin. Res. 23,322. Haymovits, A., Gershon, M. D., and Nunez, E. A. (1976).Proc. Soc. E x p . B i d . Med. 153, 383. Hazard, J. B., Hawk, W. A., and Crile, G., Jr. ( 1 9 5 5 ) ~Clin. . Endocrinol. Metab. 19,152. Hedhammar, A,, Wu, F. M., Krook, L., Schryver, H. F., Lahunta, A., Whalen, J. P., Kallfelz, F. A., Nunez, E. A., Hintz, H. F., Sheffy, B. E., and Ryan, G. D. (1974). Cornell Vet. 64, Suppl. 5. Hervas, F., Morreale d e Escobar, G., and Escobar del Rey, F. (1976).Endocrinology 98, 77. Hirsch, P. F. (1967). Endocrinology 80,539. Hirsch, P. F., and Munson, P. L. (1969). Physiol. Rev. 49, 548. Hirsch, P. F., Voelkel, E. F., and Munson, P. L. (1964). Science 146,412. Honvith, M., Nunez, E. A., Krook, L., Viteri, F., Torun, B., Mena, E., MacIntyre, I., and Whalen, J. P. (1976). Clin. Endocrinol. 5, Suppl., 341. Huang, S . N., and McLeish, W. A. (1968). Cancer 21,302. Hurtle, K. (1894). Pfluegers Arch. Gesamte Physiol. Menschen Tiere 56, 1. Idelman, C. (1962). C . R. Assoc. Anat. 118,781. Ishikawa, K. (1965).Okajimas Folia Anat. Jpn. 41,313. Isler, H. (1959).J . Natl. Cancer Znst. 23,675. Isler, H., Thompson, B., Tonkin, R., and Sarkar, S. K. (1960). Anat. Rec. 136,338. lto, T., Takahashi, Y., and Koboyashi, K. (1963). Arch. Histol. J p n . 24, 113. Jamieson, J. D., and Palade, G. E. (1967a).J . Cell Biol. 34, 577. Jamieson, J. D., and Palade, G. E. (1967b).J . Cell Biol. 34,597. Jordan, R. K., McFarlane, B., and Scothorne, R. J. (1973).J.Anat. 114, 115. Kaissling, B., and Bucher, 0. (1973). Z. Zellforsch. Mikrosk. Anat. 146,417. Kaissling, B., and Bucher, 0. (1974). Verh. Anat. Ges., Versamml., 68th, 1974 p. 817. Kalina, M., and Pearse, A. G. E. (1970). Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 262.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

73

Kalina, M., and Pearse, A. G. E. (1971). Histochemie 26, 1. Kalina, M., Foster, G. V., Clark, M. B., and Pearse, A. G. E. (1970). Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 268. Kameda, Y. (1968).Arch. Histol. Jpn. 30,83. Kameda, Y. (1970). Arch. Histol. Jpn. 32, 179. Kameda, Y. (1971a).Arch. Histol. Jpn. 33, 115. Kameda, Y. (1971b). Arch. Histol. Jpn. 33,283. Kameda, Y. (1973).Arch. Histol.Jpn. 36,89. Kameda, Y. (1974a).Arch. Histol. Jpn. 36,205. Kameda, Y. (1974b). Arch. Histol. Jpn. 37, 225. Kanarek, D. A., Nunez, E. A., and Gershon, M. D. (1976). Endocr. Soc., 58th Annu. Meet. p. 367. Kayser, C., and Frank, R. M. (1963).Arch. Oral Biol. 8, 703. Keiser, H. R., Beaven, M. A., Doppman, J., Wells, S . , and Buja, L. M. (1973).Ann. Intern. Med. 78, 561. Kendall, E. C. (1919).J. Biol. Chem. 39, 125. Ker. J. K. (1972).J. Anat. 111,333. Kirkeby, S., RQmert, P., and Gauguin, J. (1973). Histochemie 37,243. Kiyama, H., Totsuka, S., and Yoshimura, F. (1968).Endocrinol. Jpn. 15,439. Klapper, C. E. (1946).Am. J . Anat. 78, 139. Kracht, J., Hachmeister, U., Breustedt, H. J., Bonicke, J., and Lenke, M. (1968a).Calci. tonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 143. Kracht, J., Hachmeister, U., Breustedt, H. J., and Lenke, M. (1968b).Endokrinologie 52, 395. Kracht, J., Hachmeister, U., and Christ, U. (1970). Calcitonin, Proc. Int. Symp., 2nd 1969 p. 274. Kraevski, N. A., Raikhlin, N. T., Mikhailov, I. G., Smimova, E. A., and Zaridze, D. G (1974).Neoplasma 21,83. Kroon, D. B. (1958). Acta Anat. 33,76. Krstii., R. (1969). Z . Anat. Entwicklungsgesch. 129,353.. Krstii., R., and Bucher, 0. (1970).C. R. Assoc. Anat. 149, 1084. Krstii., R., and Bucher, 0. (1971).Anat. Anz. 128, Erg. H. 95. Krstii., R., Bucher, O., and Kazimierczak,J. (1974).Z . Anat. Entwicklungsgesch. 144,19. Krupp, P. P., and Frink, R. (1974).Anat. Rec. 178,396. Krupp, P. P., Young, R. A,, and Frink, R. (1976).Anat. Rec. 184,454. Lacy, P. E. (1967). N . Engl. J. Med. 276, 187. Larson, B., Owman, C., and Sundler, F. (1966). Endocrinology 78, 1109. LeDouarin, N. (1974). Med. Biol. 52, 281. LeDouarin, N., and LeLiBvre, C. (1971). C . R. Assoc. Anat. 152,558. LeDouarin, N., Fontaine, J., and LeLiBvre, C. (1974).Histochemistry 38,297. Less, E. (1970). Endokrinologie 56, 73. Levin, D. L., Perlia, C., and Tashjian, A. H., Jr. (1973). Pediatrics 52, 192. Lietz, H. (1970). Virchows Arch. A 350, 136. Lietz, H. (1971). Curr. Top. Pathol. 55, 109. Lietz, H., and Donath, K. (1970a).Calcitonin, Proc. I n t . Symp., 2nd, 1969 p. 227. Lietz, H., and Donath, K. (1970b). Virchows Arch. A 350,261. Lietz, H., and Zippel, H. (1969a).Z. Zellforsch. Mikrosk. Anat. 102,85. Lietz, H., and Zippel, H. (196913).Verh. Dtsch. Ges. Pathol. 53,431. Lietz, H., Schmahling, H. U., and Zippel, H. (1969). Virchows Arch. A 348,290. Lietz, H., Wohler, J., and Pomp, H. (1971).Z. Zellforsch. Mikrosk. Anat. 113,94.

74

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

Lindberg, L. A., and Talanti, S. (1971).Acta Vet. Scand. 12, 560. LiVolsi, V. A., Feind, C. R., LoGerfo, P., and Tashjian, A. H., Jr. (1973).J . Clin. Endocrinol. Metab. 37, 550. Ljungberg, 0. (1970a).Acta Pathol. Microbiol. Scand., Sect. A 78, 618. Ljungberg, 0. (1970b).Virchows Arch. A 349,312. Ljungberg, 0.( 1 9 7 0 ~ )Acta . Pathol. Microbiol. Scand., Sect. A 78,621. Ljungberg, 0. (1972).Acta Pathol. Microbiol. Scand., Sect. A 80, 589. Ljungberg, O., and Dymeling, J. F. (1972).Acta Pathol. Microbiol. Scand., Sect. A 80, 577. Ljungberg, O., Cederquist, E., and von Studnitz, W. (1967).Br. Med. J . 1,279. Luciano, L., and Reale, E. (1964).2. Zellforsch. Mikrosk. Anat. 64, 751. Ludwig, K. S. (1953).Acta Anat. 19,28. Ludwig, K. S. (1954).Acta Anat. 20,36. Lupulescu, A. (1972). Endocrinology 90, 1046. Lupulescu, A,, and Stebner, F. (1974). Proc. SOC.Exp. Biol. Med. 146,56. Machado, C. R. S. (1976).Histochemistry 48, 187. MacIntyre, I. (1971). Mem. Soc. Endocrinol. 19, 122. MacManus, J. P., Youdale, T., and Braceland, B. M. (1975). Horn. Metab. Res. 7,83. McMillan, P. J,, Hooker, W. M., and Deftos, L. J. (1974).Am. J . Anat. 140,73. Malaisse, W. J., Malaise-Lagae, F., Walker, M. O., and Lacy, P. (1971). Diabetes 20, 257. Mamina, V. V. (1962). Probl. Endokrinol. 8,23. Manocchio, I . (1964).Arch. Vet. Ital. 15,3. Marks, S . C., Jr. (1969).J.Bone Joint Surg., Am. Vol. 51,875. Marks, S. C., Jr., and Walker, D. G. (1969). Am. J . Anat. 126,299. Martoja, M. (1974). Calcif. Tissue Res. 14, 67. Matsuzawa, T. (1966).Arch. Histol. Jpn. 27, 521. Matsuzawa, T., and Kurosumi, K. (1967). Nature (London)213,927. Maunder, C., and Rost, F. W. D. (1972). Histochem. J . 4, 145. Maude, E. (1970). Verh. Dtsch. Ces. Pathol. 54,681. Mayer, G. P., Blum, J. W., and Deftos, L. J. (1975). Endocrinology 96, 1478. Melander, A,, and Sundler, F. (1972). Endocrinology 90,802. Melander, A,, Owman, C., and Sundler, F. (1971a).Histochemie 25,32. Melander, A., Owman, C., and Sundler, F. (1971b). Endocrinology 89,528. Melander, A., Nilsson, F., and Sundler, F. (1972). Endocrinology 90, 194. Melander, A., Sundler, F., and Westgreen, V. (1973). Endocrinology 93,193. Melander, A., Ericson, L. E., Ljundgren, J. G., Norberg, K. A., Persson, B., Sundler, F., Tibblin, S., and Westgren, U. (1974a).J.Clin. Endocrinol. Metab. 39,713. Melander, A., Ericson, L. E., Sundler, F., and Ingbar, S. H. (197413).Endocrinology 94, 959. Melvin, K. E. W., Tashjian, A. H., Jr., and Miller, H. H. (1972). Recent Prog. Horn. Res. 28,399. Meyer, J. S. (1968).Cancer 21,406. Meyer, J. S., and Abdel-Ban, W. (1968). N . Engl. J . Med. 278,523. Meyer, J. S., Hutton, W. E., and Kenny, A. D. (1973). Cancer 31,433. Mietkiewski, K., Zabel, M., and Linke, K. (1973a).Folia Morphol. (Warsaw)32,245. Mietkiewski, K., Lubaszyk, A,, and Zabel, M. (197313). Endokrynol. Pol. 24, 57. Mietkiewski, K., Zabel, M., and Linke, K. (1974).Folia Histochem. Cytochem. 12,247. Mikhailov, G. (1972). Byull. Eksp. Biol. Med. 72, 1454. Moertel, C. G., Beahrs, 0. H., Woolner, L. B., and Tajce, G. M. (1965).N. Eng2.J. Med. 273,244.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

75

Morrow, G., Barness, L. W., and Fast, A. (1967).Am. J . Dis. Child. 114, 161. Moura, A. M., Dupuy, B., Baghdiantz, A., and Blanquet, P. (1971).C. R. Hebd. Seances Acad. Sci. 272, 1145. Munson, P. L., Cooper, C. W., Gray, T. K., Peng, T. C., Toverud, S . U., Harper, C., and Ontjes, D. A. (1974). Endocrinol., Proc. Znt. Symp., 4th, 1973 p. 131. Murakami, K. (1970).Arch. Histol. Jpn. 32, 155. Mussacchia, X. J., and Kenny, A. D. (1973). Endocrinology 92,249A. Nadler, N. J., Young, B. A., Leblond, C. P., and Mitmaker, B. (1964).Endocrinology 74, 463. Nanba, H., and Fujita, H. (1969).Arch. Histol. Jpn. 30, 283. Neher, R., Riniker, B., Rittel, W., and Zuher, H. (1968).Helo. Chim. Acta 51, 1900. Nilson, G. (1972).Acta Med. Scand. 191,249. Nonidez, J. F. (1931).Am. J. Anat. 49,479. Nonidez, J. F. (1932).Anat. Rec. 53,339. Nonidez, J. F. (1933).Anat.Rec. 56, 131. Nonidez, J. F. (1935).Am. J . Anat. 57, 135. Novikoff, A. B., Novikoff, P. M., Ma, M., Shin, W. Y., and Quintana, N. (1974).Adu.Cytophannacol. 2,349. Nunez, E. A., and Becker, D. V. (1970). Am. J . Anat. 129,369. Nunez, E. A., and Gershon, M. D. (1972). Endocrinology 90,1008. Nunez, E. A., and Gershon, M. D. (1973). Endocrinology 92, 152. Nunez, E. A., and Gershon, M. D. (1976a).Anat. Rec. 184, 133. Nunez, E. A., and Gershon, M. D. (1976b). Am. J . Anat. 147,375. Nunez, E. A., and Gould, R. P. (1967).J.Anat. 101,616. Nunez, E. A., Money, W. L., and Becker, D. V. (1967a).Endocrinology 80,629. Nunez, E. A., Could, R. P., Hamilton, D. W., Hayward, J. S., and Holt, S. J. (1967b).J. Cell Sci. 2, 401. Nunez, E. A., Could, R. P., and Holt, S. J. (1968).Calcitonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 203. Nunez, E. A., Could, R. P., and Holt, S. J . (1969).J. Cell Sci. 5, 531. Nunez, E. A., Gould, R. P., and Holt, S. J. (1970a).J . Cell Sci. 6,821. Nunez, E. A., Could, R. P., and Holt, S. J. (1970b). Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 252. Nunez, E. A., Belshaw, B. B., and Gershon, M. D. (1972a).Am. J . Anat. 133,463. Nunez, E. A., Krook, L., and Whalen, J. P. (1972b). Am. J . Anat. 134,459. Nunez, E. A., Hedhammer, A., Wu, F. M., Whalen, J. P., and Krook, L. (1974). Lab. Znuest. 31,96. Nunez, E. A., Greif, R. L., and Gershon, M. D. (1975). Lab. Inuest. 33,352. Nunez, E . A., Krook, L., and Whalen, J. P. (1976). Cell Tissue Res. 168,373. O’Dor, R. K., Parks, C. O., and Copp, D. H. (1969). Comp. Biochem. Physiol. 29,295. Oeltgen, P. R., and L’Heureux, M. V. (1975).Honn. Metab. Res. 7,88. Ogata, E., and Kimura, S. (1973). In “The Kidney and Calcium” (T. Fujita, J. Ono, and E. Ogata, eds.), p. 45. Ishiyaku Publishers, Inc., Tokyo. Ohkubo, K. (1934).Jpn. J . Med. Sci. 1 5,47. Olivereau, M. (1970).2. Zellforsch. Mikrosk. Anat. 107,374. Olivereau, M., and Fontaine, M. M. (1970). C. R. Hebd. Seances Acad. Sci. 270, 155. Oppenheimer, J. H., Schwartz, H. L., and Surks, M. I. (1972).J . Clin. Inuest. 51,2493. Ouvrard-Pascaud, N., Vendrely, E., and DaLage, C. (1976). C. R. Hebd. Seances Acad. Sci. 282, 575. Owman, C., and Sundler, F. (1968). Calcitonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 110.

76

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

PagBs, A. (1956). Biol. Med. 45,414. Pages, A. (1974).Ann. Anat. Pathol. 19,335. Pantii., V. (1967). Z. Zellforsch. Mikrosk. Anat. 81,487. Pantib, V . (1974). Znt. Reu. Cytol. 38, 153. Pantii., V., and Kalusevic, S. (1974).Acta Anat. 90,369. Pantii., V., and StoSii., N. (1966).Acta Anat. 63, 580. Pantii., V., Pavlovic-Hournac, M. R., and Radivojevic, D. V. (1970).Acta Anat. 75,553. Pearse, A. G . E. (1966a). Proc. R. Soc. London, Ser. B 164,478. Pearse, A. G . E. (196613).Nature (London) 211,598. Pearse, A. G. E. (1966~). Vet. Rec. 79,587. Pearse, A. G. E. (1968a). Proc. R. Soc. London, Ser. B 170,71. Pearse, A. G. E. (1968b).Adu. Phannacol. 6, Part B, 156. Pearse, A. G . E. ( 1 9 6 8 ~ )Calcitonin, . Proc. S y m p . Thyrocalcitonin C Cells, 1967 p. 98. Pearse, A. G. E. (1969).J.Histochem. Cytochem. 17,303. Pearse, A. G. E. (1970). Calcitonin, Proc. l n t . S y m p . , 2nd, 1969 p . 125. Pearse, A. G. E., and Carvalheira, A. F. (1967).Nature (London)214,929. Pearse, A. G . E., and Polak, J. M. (1971). Histochemie 27,96. Pearse, A. G. E., and Polak, J. M. (1974).Med. Biol. 52,3. Pearse, A. G. E., and Takor, T. T. (1976).Clin. Endocrinol. 5, Suppl., 229. Pearse, A. G. E., and Welsch, U. (1968).Z . Zellforsch. Mikrosk. Anat. 92,596. Pearse, A. G . E., and Welsch, U. (1969).Histochemie 17, 1. Pearse, A. G. E., Ewen, S. W. B., and Polak, J. M. (1972). Virchows Arch. B 10, 93. Peng, T. C. (1975). Fed. Proc., Fed. Am. SOC. E x p . Biol. 34, Part 1, 761. Peng, T. C., Cooper, C. W., Petrusz, P., and Volpert, E. M. (1975).Endocrinology 97, 1537. Peng, T. C., Cooper, C. W., and Garner, S. C. (1976).Proc. Soc. E x p . Biol. Med. 153,268. Pepler, W. J., and Pearse, A. G. E. (1967). Br. J. Exp. Pathol. 38,221. Petk6, M. (1974a).Acta Morphol. Acad. Sci. Hung. 22,213. Petk6, M. (197413).Verh. Anat. Ges., Versamml., 68th 1974 p. 837. Petk6, M. ( 1 9 7 4 ~ )Stain . Technol. 49,65. Petk6, M., Rig6, Gy., and Varga, Z. (1976). Cell Tissue Res. 166,541. Polak, J. M., Pearse, A. G. E., LeLikvre, C., Fontaine, J., and LeDouarin, N. M. (1974). Histochemistry 40,209. Ponse, K. (1951). Ann. Endocrinol. 12,266. Potts, T. J., Jr., Niall, H. D., Keutmann, H. T., Brewer, B., and Deftos, L. J. (1968).Proc. Natl. Acad. Sci. U.S.A. 59, 1968. Potts, T. J., Jr., Deftos, L. J., and Parsons, J. A. (1970).Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 56. Queener, S . F., and Bell, N. H. (1975). Metab., Clin. E x p . 24, 1975. Raikhlin, N. T., and Smimova, E. A. (1970). Folia Histochem. Cytochem. 8,231. Rasmussen, H., and Bordier, P. (1974).In “The Physiological and Cellular Basis of Metabolic Bone Disease” (H. Rasmussen and P. Bordier, eds.), p. 164. Williams & Wilh s , Baltimore, Maryland Rasmussen, P. (1969). Nord. Med. 81,553. Raulais, D., Merle, M., and Milhaud, G. (1974).Endocrinol.,Proc. Znt. Symp., 4th, 1973 p. 177. Raymond, N. (1932). Anat. Rec. 53,355. Rhodin, A. G. (1974). “Histology: A Text and Atlas,” p. 442. Oxford Univ. Press, London and New York. Riedesel, M. L. (1957). Trans. Kans. Acad. Sci. 60,99.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

77

Ritzen, M., Hammarstrom, L., and Ullberg, S. (1965).Biochem. Pharmacol. 14,313. Robertson, D. R. (1967).Z . Zellforsch. Mikrosk. Anat. 78,328. Roediger, W. E. W. (1973a).S. Afr. J . Med. Sci. 38, 17. Roediger, W. E. W. (1973b).J. Anat. 115,255. Roediger, W. E. W. (1975).Cancer 36, 1758. Rogers, W. H. (1927).Am. J. Anat. 38,349. Rogers, W. H. (1929).Am. J. Anat. 44, 283. Rohr, H. P., and Hasler, K. (1968).Experientia 2.4, 152. Rojo-Ortega, J. M., Brecht, H. M., and Genest, J. (1971).Virchows Arch. B 7, 81. Roos, B. A., Bundy, L. L., Bailey, R., and Deftos, L. J. (1974).Endocrinology 95, 1142. Roos, B. A., Bundy, L. L., Miller, E. A., and Deftos, L. J. (1975).Endocrinology 97,39. Rosciszewska, A., and Beskid, M. (1969).Endokrynol. Pol. 20, 567. Rost, M. C. M., and Rost, F. W. D. (1975).Histochem. J . 7,307. Roszkiewicz, J. (1974a).Folia Morphol. (Warsaw)33, 177. Roszkiewicz, J. (197413).Folia Histochem. Cytochem. 12,355. Roszkiewia, J., and Roszkiewicz, A. (1974).Folia Morphol. (Warsaw) 33,247. Roth, S . I., Olsen, F., and Hanson, L. (1962).Lab. Invest. 11, 933. Roth, S. I., Feinblatt, J. D., and Raisz, L. G. (1974).Am. J. Pathol. 75, 27. Rother, P. (1970).Endokrinologie 56, 92. Rouais, F., Baquez, C., Dupuy, B., and Blanquet, P. (1973).C. R. Seances Soc. B i d . Ses. F i l . 167, 1591. Saito, K. (1956).Kitakanto Igaku 6, 511. Saito, K., and Shibata, K. (1957).Endocrinol. Jpn. 4,253. Salzer, G. M. (1971).Acta Endocrinol. (Copenhagen),Suppl. 157,5. Samaan, N. A., Anderson, G. D., and Adams-Mayne, N. E. (1975).A m . ] . Obstet. Gynecol. 121,622. Sandborn, E., Koen, P. F., McNabb, J. D., and Moore, G. (1964).J. Ultrastruct. Res. 11, 123. Sandritter, W., and Klein, V. K. H. (1954). Frankf. Z . Pathol. 65,204. Sandritter, W., Kummer, E., Pillat, G., and Rowe, L. (1956).Klin. Wochenschr. 34,871. Sarkar, S. K. (1961).Anat. Rec. 139,270. Sarkar, S. K., and Isler, H. (1963).Endocrinology 73, 199. Sato, T. (1959).Okajimas Folia Anat. Jpn. 33,225. Sato, T., Ishikawa, K., Aoi, T., Kitoh, J., and Sugiyama, S. (1966).Okajimas Folia Anat. Jpn. 42, 91. Sawicki, B. (1971).Folia Morphol. (Warsaw) 30,404. Sawicki, B. (1975).Acta Theriol. 20,281. Sawicki, B., and Bajko, K. (1974).Folia Morphol. (Warsaw)33,45. Seecof, D. P. (1927).Am. J. Pathol. 3,365. Seljelid, R., Reith, A., and Nakkem, K. F. (1971).Lab. Invest. 23,595. Shiino, M., and Rennels, E. G . (1974).Am. J. Anat. 139, 135. Shively, J. N., and Epling, G. P. (1969).Am. 1.Vet. Res. 30,219. Shore, P. A. (1962).Phamacol. Reu. 14,531. Sieber, P., Brugger, M., Kamber, B., Riniker, B., and Rittel, W. (1968).Helu. Chim. Acta 51,2057. Sipple, J. H. (1961).Am. J. Med. 31, 163. Slanina, P., and Tjalve, H. (1972).Acta Physiol. Scand. 84,254. Slanina, P., and Tjalve, H. (1973).J. Endocrinol. 58,21. Smith, U., Smith, D. S., Winkler, H., and Ryan, J. W. (1973). Science 179,79. Solcia, E., and Sampietro, R. (1965a).2. Zellforsch. Mikrosk. Anat. 68,689.

78

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

Solcia, E., and Sampietro, R. (196513).Riu. Zstochim. Norm. Patol. 9,265. Solcia, E., and Sampietro, R. (1968).Calcitonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p. 127. Solcia, E., Rondini, G., and Capella, C. (1968a).Helu. Paediatr. Acta 23,650. Solcia, E., Vasallo, G., and Capella, C. (1968b). Stain Technol. 43, 257. Solcia, E., Capella, C., Sampietro, R., and Vasallo, G. (1970). Calcitonin, Proc. Znt. Symp., 2nd, 1969 p. 220. Stachura, J . (1971a). Folia Med. Crakoo. 13,375. Stachura, J . (1971b). Patol. Pol. 22,287. Stachura, J., and Pearse, A. G. E. (1970). Virchows Arch. B 5, 173. Stoeckel, M. E., and Porte, A. (1967). C . R . Seances SOC. Biol. Ses Fil. 161,2040. Stoeckel, M. E., and Porte, A. (1970).Z. Zellforsch. Mikrosk. Anat. 106,251. Stoeckel, M. E., and Porte, A, (1972).Z. Zellforsch. Mikrosk. Anat. 128,347. Stoeckel, M. E., Porte, A., and Canguilhem, B. (1967). C. R. Hebd. Seances Acad. Sci. 264,2490. Stux, M., Thompson, B., Isler, H., and Leblond, C. P. (1961).Endocrinology 68, 292. Sugiyama, S. (1942).Jpn.J . Med. Sci. 19, 153. Sugiyama, S. (1950). Okajimas Folia Anat. Jpn. 23,57. Sugiyama, S. (1954).Anat. Rec. 120,363. Sugiyama, S. (1967).Ergeb. Anat. Entwicklungsgesch. 39, No. 3. Sugiyama, S. (1969).Acta Endocrinol. (Copenhagen) 138, Suppl., 179. Sugiyama, S., Kitoh, J., and Yokota, S. (1969). Okajimas Folia Anat. J p n . 46,59. Sunder-Plassmann, P. (1939).Dtsch. Z. Chir. 252,210. Sundstrom, B. (1971).Acta Pathol. Microbial. Scand. 79,407. Swaminathan, R., Bates, R. F. L., and Care, A. D. (1972).J. Endocrinol. 54,525. Swarup, K., and Das, V. K. (1974).Arch. Histol. J p n . 37, 121. Takagi, I., Yamada, K., and Sugiyama, S. (1974).Folia Anat. Jpn. 50,359. Takagi, K. (1922). Folia Anat. Jpn. 1, 69. Tashiro, M. (1962).Acta Anat. Nippon. 37, 63. Tashiro, M. (1964). Okajimas Folia Anat. J p n . 39, 191. Tashjian, A. H., Jr., and Melvin, K. E. W. (1968).N . Engl. J . Med. 279,279. Tashjian, A. H., Jr., and Voekel, E. F. (1967).J. Clin. Endocrinol. Metab. 27, 1353. Tashjian, A. H., Jr., Wolfe, H. J., and Voekel, E. F. (1974).Am. J . Med. 56,840. Tata, J. R. (1964).Adu. Metab. Disord. 1, 153. Tateiski, R., Takahashi, Y., and Noguchi, A. (1972).Cancer 30,755. Tauber, S. D. (1967). Proc. Natl. Acad. Sci. U.S.A. 58, 1684. Teitelbaum, S. L., Moore, K. E., and Shieber, W. (1970).Anat. Rec. 168,69. Teitelbaum, S. L., Moore, K. E., and Shieber, W. (1971).Nature (London)230,33. Teitelbaum, S . L., Shieber, W., and Moore, K. E. (1972). Cytobios 5,43. Thliveris, J. A., Dub&,W. J., and Banerjee, R. (1976).Virchows Arch. A 369,249. Thomas, F. (1934).Arch. Biol. 45, 189. Thompson, B., Isler, H., and Sarkar, S . K. (1962).Endocrinology 70,786. Tice, L. W., and Creveling, C. R. (1975). Endocrinology 97, 1123. Tischler, A. S., Dichter, M. A., Biales, B., BeLelli, R. A., and Wolfe, H. (1976).Science 192,902. Tjalve, H. (1971). Acta Physiol. Scand., Suppl. 360,5. Tjalve, H., and Slanina, P. (1971).Z. Zellforsch. Mikrosk. Anat. 113,83. Treilhou-LaHille, F., and Beaumont, A. (1975).J . Ultrastruct. Res. 50,387. Tremblay, A., and Pearse, A. G. E. (1960).J . Pathol. Bacteriol. 80,353. Trolldenier, H. (1967).Z. Mikrosk.-Anat. Forsch. 77,29.

CYTOPHYSIOLOGY OF THYROID PARAFOLLICULAR CELLS

79

Van Dyke, J. H. (1944).Anat. Rec. 88,369. Van Dyke, J. H. (1945). A m . J. Anat. 76,201. Vassallo, G., Capella, C., and Solcia, E. (1971).Stain Technol. 46, 7. Velickf, J. (1970).Folia Morphol. (Prague) 18,389. Velickf, J. (1971a). Folia Morphol. (Prague) 19,97. Velickq, J. (1971b). Folia Morphol. (Prague) 19, 105. Velickf, J. ( 1 9 7 1 ~ )Folia . Morphol. (Prague) 19, 125. Velickf, J., and Titlback, M. (1972). Folia Morphol. (Prague)20,406. Velickf, J., and Titlback, M. (1974). Z. Mikrosk.-Anat. Forsch. 88, 1069. Voelkel, E. F., Tashjian, A. H., Jr., Davidoff, F. F., Cohen, R. B., Perlia, C. P.,and Wurtman, P. J. (1973).J . Clin. Endocrinol. Metab. 37,297. Voitkevitch, A. A. (1963).Arkh. Anat., Gistol. Embriol. 44, 52. Walker, D. G. (1966a). Z. Zellforsch. Mikrosk. Anat. 72, 100. Walker, D. G . (196613).Endocrinology 79,836. Walker, D. G. (1972).Endocrinology 91, 916. Wallach, S., Chausmer, A., and Mittleman, R. (1967).Endocrinology 80, 61. Waller, U. (1960). Acta Endocrinol. (Copenhagen) 35,334. Wassermann, D. (1976). Cell Tissue Res. 170,89. Watari, N. (1974).J. Electron Microsc. 23, 255. Weichert, R. F. (1970). Am. J . Med. 49,232. Weisbrode, S. E., and Capen, C. C. (1974). Virchows Arch. B 16,231. Welsch, U. (1971).Mem. SOC. Endocrinol. 19,557. Welsch, U . (1972).Ergeb. Anat. Entwicklungsgesch. 46, 7. Welsch, U., and Buchheim, W. (1972). Endocrinology, Proc. Znt. S y m p . , 3rd, 1971, p. 291. Welsch, U., and Pearse, A. G . E. (1969).Histochemie 17, 1. Welsch, U., Flitney, F. W., and Pearse, A. G . E. (1968). Calcitonin, Proc. S y m p . Thyrocalcitonin C Cells, 1967 p. 167. Welsch, U., Flitney, F. W., and Pearse, A. G. E. (1969).J.Microsc. (Oxford)89,83. Wetzel, B. K., Spicer, S. S., and Wollman, S. H. (1965).J . Cell Biol. 25, 593. Whalen, J. P., Krook, L., and Nunez, E. A. (1972).Anat. Rec. 172, 97. Whalen, J. P., Krook, L., MacIntyre, I., and Nunez, E. A. (1975).J.Endocrinol. 66,207. Wilkie, B. N., and Krook, L. (1970). Pathol. Vet. 7, 126. Williams, E. D. (1966).J.Clin. Pathol. 19, 114. Williams, E. D., Brown, C. L., and Doniach, I. (1966).J.Clin. Pathol. 19, 103. Williams, E. D., Karim, S. M. M., and Sandler, M. (1968).Lancet 1,22. Williams, J. A. (1976).Endocrinology 98, 1351. Wissig, S. L. (1962).Electron Microsc., Proc. Znt. Congr. 5th, 1962 Vol. 2, Artic. WW-1. Wolfe, H. J., and Tashjian, A. H., Jr. (1974). Endocrinol., Proc. Int. Symp., 4th, 1973 p . 323. Wolfe, H. J., Melvin, K. E. W., Cervi-Skinner, S. J., Al-Saadi, A. A., Juliar, J. E., Jackson, C. E., and Tashjian, A. H., Jr. (1973).N. Engl. J . Med. 289,437. Wolfe, H. J., Voelkel, E. F., and Tashjian, A. H., Jr. (1974).J.Clin. Endocrinol. Metab. 38,688. Wolfe, H . J., DeLellis, R. A.,Voelkel, E. F., andTashjian, A. H., Jr. (1975).J.Clin. Endocrinol. Metab. 41, 1076. Wollman, S. H., and NBve, P. (1971). Recent Prog. Horn. Res. 27,213. Woodhouse, N. J. Y., Fisher, M. T., Sigurdsson, G., Joplin, G. E., and MacIntyre, I. (1972). Br. Med. J . 4,267. Wuttke, H., Kessler, F. J., and Jungck, A. (1971).Z. Gesamte Erp. Med. Einschl. Exp. Chir. 155,20.

80

ELADIO A. NUNEZ AND MICHAEL D. GERSHON

Yamngton, J. T., Capen, C. C., Black, H. E., Re, R., Potts, J. T., Jr., and Ceho, W. B. (1976).Am. J . Pathol. 83,569. Yasumura, S., Burk, M., Chausmer, M., Mittleman, R., and Wallach, S. (1967).Endocrinology 81,256. Yoshirnura, F., and Iris, M.(1961). Z . Zellforsch. Mikrosk. Anat. 55,204. Yoshimura, F., Tonetsu, T., and Nakamura, M. (1962). Endocrinol. Jpn. 9,284. Yoshimura, F., Yachi, H., Ishikawa, H., Ohtsuka, Y., and Kiguchi, Y. (1972).Endocrinol. Jpn. 19, 151. Young, B. A. (1966).J . Anat. 100,895. Young, B. A., and Harrison, R. J. (1969).2. Zellforsch. Mikrosk. Anat. 96,222. Young, B. A., and Leblond, C. P. (1963). Endocrinology 73,669. Young, B. A., Care, A. D., and Duncan, T. (1968).J.Anat. 102,275. Young, D. M., and Capen, C. C. (1968). Calcitonin, Proc. Symp. Thyrocalcitonin C Cells, 1967 p, 181. Young, D. M., and Capen, C. C. (1970).Calcitonin, Proc. Int. Syrnp., 2nd, 1969 p. 141. Young, D. M., and Capen, C. C. (1971). Virchows Arch. B 8,288. Young, D. M., Capen, C. C., and Black, H. E. (1972). Endocrinology 90,276. Zabel, M. (1973). Folia Morphol. (Warsaw) 32,85. Zabel, M. (1976). Endokrinologie 67,315. Zabel, M., and Peil, J . (1974). Morphol. Jahrb. 120,854. Zaridze, D. C. (1974). Arkh. Patol. 36,45. Zechel, C. (1933).Anat.Rec. 56, 119. Zuckerkandl, E. (1903).Anat. Hefte, Wiesb., 21, 1.

Cytophysiology of the Amphibian Thyroid Gland through Larval Development and Metamorphosis ELIANEREGARD Universitt? Paris-Sud, Lboratoire de Biologie-VertSbrSs, Orsay, France

I. Introduction . . . . . . . . . . Movements Directed toward the Colloid . . . . Migrations Directed toward the Interfollicular Space . . . . . . . . . . . 11. Iodide Pathways . . . . . . . . . A. Iodide Oxidation Processes . . . . . . B. Qualitative and Quantitative Variations in the Iodine Content in the Colloid-Relative Intensity of Transepithelial Iodine Flow . . . . . . C. Effects of Hypophysectomy . . . . . . 111. Thyroglobulin Biosynthesis . . . . . . . A. Relative Measurement of Thyroglobulin Synthesis B. Ultrastructural Features Related to Thvroelobulin Synthesis . . . . . . . . . . C. Effects of Hypophysectomy . . . . . . IV. Hypophyseal Regulation . . . . . . . A. Sensitivity of Thyroid Gland to Exogeneous Ovine . . Prolactin, in Larvae at Metamorphic Climax B. Response of the Thyroid Gland of Hypophysectomized . . . . . . . . LarvaetoTSH C. Characteristics of the Thyroid Gland in Hypcphysectomized Larvae Treated with Prolactin or with Both Prolactin and TSH in Various Ratios . . . . . . . D. Hypophyseal Regulation of Thyroid Activity Based on a Bihormonal Model and Neotenic Development E. Production of the Two Hormones in Different Ratios . V. Conclusions . . . . . . . . . . References . . . . . . . . . . I.

81 83 83 87 87 92 99 100 102 103 104 105 108

109 112 112 113 115

I. Introduction In anuran amphibians, functional differentiation of the thyroid starts at an early stage of embryonic life when the anlage consists of only a thickening of the pharyngeal epithelium. The cells of this anlage can already incorporate iodide into small, soluble proteins (Flickinger, 1964; Hanaoka et al., 1973). Functional maturation of the gland is achieved when it separates into two distinct lobes at the time of hatching. The ability to synthesize thyroglobulin appears at this time, 81

82

ELIANE REGARD

and the colloid fills the lumen of the first follicles (Hanaoka et uZ., 1973). Different cytological and functional criteria, such as the height of the cells (Coleman et d.,1968; Etkin, 1968; Fox and Turner, 1967; Seed and Goldberg, 1965), the electric resistance of the epithelium (Gorbman and Ueda, 1963), and the iodide uptake (Flickinger, 1964; Gorbman, 1964; Kaye, 1961; Leloup and Fontaine, 1960; Neuenschwander, 1972; S a x h et uZ., 195713; Dodd and Dodd, 1976), define three other stages in the evolution of the thyroid. Each of them corresponds to a period of larval life or metamorphosis. A resting period, with high transepithelial resistance, characterizes the thyroid during premetamorphosis. This is a period of rapid growth for the tadpole, accompanied by few morphological changes. This first period is followed by an hyperactive one, during which both iodide uptake and epithelial height increase and transepithelial resistance decreases. This active phase corresponds on the one hand to prometamorphosis with a reduced body growth rate and accelerated morphological changes, and on the other hand to the climax period of rapid differentiation and absence of growth. Subsequently, a new equilibrium, characterized by high transepithelial resistance and a decrease in cell height, is established. This third phase occurs at postclimax, which is the beginning of adult life. In neotenic development, which is frequent in urodeles, the thyroid evolves differently, and its physiology is poorly known (Larsen, 1968). Generally, the gland of the axolotl is presumed to be in an hypofunctional state (Blount, 1950; Huxley, 1920). Although the diversity of thyroid physiological states is well established in amphibians through larval development and metamorphosis, correlations between cytological and functional characteristics have hardly been studied. No great difficulties should be encountered in such studies, since the fundamental aspects of the main processes that take place in the thyroid cells, demonstrated in mammals, are common to all vertebrates (Barrington, 1964; Leloup and Fontaine, 1960). There are four such processes: utilization of iodine, thyroglobulin synthesis, storage of iodinated thyroglobulin in the colloid, and finally secretion of hormones. The organelles involved in the different sequences are known (Fig. 1).The colloid is located at the apical pole of the cells, filling the follicular lumen, while the entry of metabolites (iodide and various precursors) and the exit of hormones take place at the basal pole, because of the bipolar characteristics of the thyroid cells. The mechanisms that regulate these two opposite flows are now examined.

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

83

MOVEMENTSDIRECTEDTOWARD THE COLLOID(FIG.1) After its uptake by an active process, iodide is oxidized during its migration toward the apical pole of the cell. The oxidative reaction occurs prior to its incorporation into the thyroglobulin molecules (Edelhoch, 1965) and involves a peroxidase (thyroperoxidase) (Alexander and Corcoran, 1962; Hosoya, 1968; Taurog, 1970), the activity of which can be localized in different cellular organelles (Fig. 2) (Graham and Karnovsky, 1966; Hosoya et al., 1971; Nakai and Fujita, 1970; Novikoff, 1971; Shin et al., 1971; Strum and Karnovsky, 1970) and on the apical microvilli (Tice and Wollman, 1972). Synthesized in the ergastoplasm, thyroglobulin also migrates in the direction of the apical pole, where it is observed in secretory microvesicles (Haddad et al., 1971; Leblond and Gross, 1948; Nadler et al., 1964; Seljelid, 1967) which open into the colloid. The sedimentation coefficient of this specific protein of vertebrate thyroid tissue is 19s.There are no important differences in the sedimentation coefficients of thyroglobulins of the various classes of vertebrates (Brisson, 1972; Hoshino and Ui, 1970; Rosenberg and Cavalieri, 1971). Thyroglobulin acts as the substrate in the iodinating process, which presumably occurs on the apical microvilli (Fujita, 1969; Stein and Gross, 1964; Salvatore et al., 1965). After this first series of migrations, iodinated thyroglobulin molecules make up most of the colloid. MIGRATIONS DIRECTEDTOWARD THE INTERFOLLICULAR SPACE(FIG.1) Colloid resorption is due to pinocytosis and produces intracellular colloid droplets which fuse with lysosomes (Ekholm and Smeds, 1966; Wissig, 1964). The hydrolysis of thyroglobulin, giving rise to monoiodotyrosine (MIT) and diiodotyrosine (DIT), may take place during this resorption, followed by thyroid hormone secretion [triiodothyronine (TJ and tetraiodothyronine (T,)] in the follicular space (Deiss and Peake, 1968; Wollman, 1969; Wissig, 1960). Moreover, the existence of extreme functional states in the amphibian thyroid poses the question of their control. Many studies have shown that the thyroid of hypophysectomized larvae is in a resting state and that the function of the gland depends largely on the activity of the hypothalamohypophyseal complex. A complicated system of interactions is progressively established during development. Etkin called this plurihormonal system the thyrostat system, suggesting that the level of T, secretion depends on the thyroid-stimulating hormone (TSH) and prolactin plasmic levels. An antagonism between the

84

ELIANE REGARD COLLOID iodinated Tg.

1-

pZiiGiodination] f

synthesis

hydrolysis

a

INTERFOLLICULAR SPACE

FIG.1. The specific reactions of thyroid tissue (a) and their intracellular localization in Xenopus thyroid cells (b). aa, Amino acids; C, colloid; cd, colloid droplet; G, Golgi apparatus; I-, iodide; ly, lysosome; mv, microvesicle; M, mitochondria; N, nucleus; RER, rough endoplasmic reticulum; BC, blood capillary; IS, interfollicular space; Tg, thyroglobulin; MIT, monoiodotyrosine; DIT, diiodotyrosine; T,, triiodothyronine; T4, thyroxine.

levels of these two hormones is plausible in amphibian larvae, as the administration of prolactin generally delays metamorphosis (Etkin, 1968; Etkin et al., 1969; Etkin and Gona, 1967a,b; Gona, 1967; Gona et al., 1970; Dodd and Dodd, 1976). The cytophysiological studies on the amphibian thyroid gland re-

FIG. lb. See legend on facing page.

ported deal with anurans at different stages of development and metamorphosis and with urodeles exhibiting neotenic development. This article is divided into three main parts. The first discusses iodide utilization and the changes in iodine flow through the follicular epithelium of glands in different functional states. The second deals with correlations between the evolution of the ability to synthesize thyroglobulin and the adaptations of the organelles involved in this synthesis. The possibility of modulation of amphibian thyroid activity by

86

ELIANE REGARD

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

87

prolactin and TSH, and its relevance in the interpretation of special physiological states observed in neotenic urodeles, are considered in the third part.

11. Iodide Pathways In amphibians as well as in other vertebrates, the ability of the thyroid to use iodide is a good criterion for judging its functional differentiation. This step occurs just before thyroglobulin synthesis (Hanaoka et uZ., 1973).The ability to concentrate iodide is usually called uptake. Its variations have been studied b y comparing the radioactivity of glands labeled with lZ5Ior lS1I.Several works have demonstrated that in the anuran thyroid the uptake of radioiodide increases during the larval growth phase, reaches a maximum during metamorphosis, and decreases at the beginning of adult life (Flickinger, 1964; Gorbman, 1964; Hanaoka, 1966; Kaye, 1961; Neuenschwander, 1972; S a x b et aZ., 1957a,b; Dodd and Dodd, 1976). The corresponding curve is given in Fig. 3 (Nataf et al., 1975). In neotenic axolotls the uptake of iodide is always very low (Mauchamp and Nunez, 1968; Taurog, 1974). Nevertheless, in all cases most of the iodide trapped is incorporated into thyroid proteins. These different observations suggest variations in peroxidase activity, since the oxidation of iodide represents a preliminary step in its incorporation into the thyroglobulin molecules. They also reflect variations in the amount of iodine contained in the gland, located essentially in the colloid. The variations in this amount indicate the difference between iodine uptake and iodine secretion through the follicular epithelium.

A. IODIDEOXIDATIONPROCESSES

The changes in iodide oxidation processes can be indirectly demonstrated by the changes in the intracellular distribution of peroxidase, the responsible enzyme (Alexander and Corcoran, 1962; Hosoya, 1968; Taurog, 1970).

FIG.2. Cytochernical localization of peroxidase activity in the thyroid gland ofXenopus larva at the end of prornetarnorphosis.Characteristic black osmium deposits are observed in many structures: perinuclear space; REG: ergastoplasrnic formations; G: Golgi saccules; MV: apical rnicrovesicles. Other abbreviations: C: colloid; CD: lysosome; CV: colloid droplet; M: mitochondria; N: nucleus.

88

ELIANE REGARD

%

bdide uplake

2

1.5

I

0.5

0

/

PREMETAMORPHOSIS

iPRoKTAMORPHOSJS

I

CLIMAX

j POSTCLIMAX

slage

I

I

FIG.3. Curve of radioiodide uptake in the thyroid of Xenopus larvae incubated 72 hours in a bath containing lZsI(16 pCi/ml). The solid line indicates the intracellular localization of peroxidase activity during the corresponding larval period.

1. Evolution of Peroxidase Activity in Xenopus Thyroid Gland The cellular distribution of peroxidase activity in the thyroid cells ofXenopus changes throughout larval life and metamorphosis (Regard and Mauchamp, 1973a,b). This activity can be identified locally, at the end of premetamorphosis, and the follicular epithelium, homogeneous from one follicle to the next, is rather flat (7 pm). It is observed at the following sites: the nuclear envelope, several ergastoplasmic formations, some Golgi saccules, and apical microvesicles (Fig. 3 ) . These different organelles are few in number during this period. Enzymic activity is more widely distributed during prometamorphosis, when the height of the epithelium reaches, on the average, 12 p m . It is generalized at climax to all the parallel ergastoplasmic formations in the infranuclear region and is also present in several Golgi saccules and in many apical microvesicles (Fig. 3). The height of the cells is

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

89

close to 20 pm. During postclimax, on the contrary, the height of the epithelium decreases (10 pm), and peroxidase activity is observed only in some ergastoplasmic formations, the development of which is not so important as during climax, and in some apical microvesicles (Fig. 3). It was observed that during the different stages studied the intensity of the enzymic reaction may vary from one follicle to the next. These results gave two kinds of information related to the cytophysiology of the gland:

1. The different localizations of peroxidase activity suggest that the enzyme migrates from its synthesis site (nuclear envelope or ergastoplasm) toward its site of utilization (apical microvilli) through secretory microvesicles (Novikoff et al., 1971; Regard and Mauchamp, 197313; Strum and Karnovsky, 1970). However, the processes involved in the transfer of enzyme from the apical microvesicles to the microvilli are not known. The presence of the enzyme at all stages of development suggests that most of the iodide fixed by the thyroid gland is incorporated into proteins (Nataf et al., 1975; Regard and Mauchamp, 1973a,b). 2. The variations in the distribution of peroxidase activity suggest that the intensity of iodine uptake changes from one follicle to the next. During climax, enzymic activity appears to be the most intense and the most widely distributed. This observation agrees with an especially high total iodine uptake throughout this period. 2. Special Distribution of Peroxidase Activity in the Neotenic Axolotl

The cytological and cytochemical characteristics of the thyroid gland in the neotenic axolotl show a decreased ability to use iodine (Mauchamp and Regard, 1971; Regard 1974). Morphologically, thyroid follicles are very irregular. They consist of confluent structures (Fig. 4). The epithelium of these formations exhibits heterogeneity which is both inter- and intrafollicular (Fig. 4). Follicles bordered with high cells, which are rich in ergastoplasm and are active, are adjacent to follicles with low cells, which have a high nucleocytoplasmic ratio and are inactive. These observations have also been reported for other species of neotenic urodeles (Gabrion and Sentein, 1972; Larsen, 1968). The follicular epithelium itself is heterogeneous (Figs. 4 and 5). It is formed of cells of different heights, and the transition from high cells to low cells occurs through intermediate cubic cells (Fig. 5). Studies on the localization of peroxidase activity show that the mor-

90

ELIANE REGARD

FIG.4. (1-4) Serial sections of the thyroid gland of the axolotl showing confluent follicular structures (a and b, c and d, e and f, g and h). (5) Cellular heterogeneity of the follicular epithelium.

phological heterogeneity reflects the functional heterogeneity of the tissue. The distribution of enzymic activity is variable along the follicular epithelium (Regard, 1974).In high cells the activity is visualized at the sites described in other amphibians (Fig. 5 ) , such as the perinuclear space, the ergastoplasm, the Golgi stalk, and the apical microvesicles. The reaction product of enzyme activity is also observed as a thin, black osmium deposit on the apical microvilli, the presumed site of thyroglobulin iodination. In cubic cells, the reaction product is restricted to some microvesicles of the supranuclear region and is very uncommon in flat cells (Fig. 5).Enzymic activity seems to be present especially in high cells, where it is the most widely distributed. It is possible that the iodide flows that pass through the different types of cells do not have the

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

IDistribution of

91

A peroxidase activity

FIG. 5. Morphological (distribution of organelles) and functional (distribution of peroxidase activity) heterogeneity of the three cell types observed in the thyroid gland of the axolotl: A, high cell, B, intermediate cell; C, flat cell.

same intensity. They may be very low in cells where peroxidase activity is the most restricted (inactive cells). The presence of a significant proportion of inactive cells suggests that iodine uptake is very low in the thyroid of neotenic larvae. This is in agreement with the low radioiodine uptake in the gland (Mauchamp and Nunez, 1968; Taurog,

1974). Iodine uptake seems to follow the same sequence occurring in the amphibian thyroid as well as in that of higher vertebrates such as mammals. Moreover, in the rat fetus, peroxidase activity is detected on the nineteenth day of pregnancy, when the gland becomes functional. At this stage iodinated thyroglobulin can be measured and thyroid hormones identified (Jost, 1961; Geloso, 1967; Nataf and Sfez, 1967; Strum et al., 1971). The ability to synthesize peroxidase develops in correlation with the ability to take up iodide. Moreover, the morphological and functional heterogeneity of the thyroid of the axolotl more likely resembles that observed in mammals. In mammals the gland has small and large follicles. It has been demonstrated that the rate of iodine uptake and renewal is a function of follicle size (Nadler

92

ELIANE REGARD

et al., 1954; Simon and Morel, 1960; Triantaphyllidis and Verne,

1963).

B. QUALITATIVE AND QUANTITATIVE VARIATIONS

IN THE

IODINECONTENTOF THE COLLOID-RELATIVE INTENSITY OF TRANSEPITHELIAL IODINE FLOW The colloid, especially the portion containing iodinated thyroglobulin, represents in vertebrates the storage form of hormone in the gland (Constantinescu, 1972).

1. Nature and Ratio of Dischargeable Iodinated Amino Acids Chromatographic analysis of gland hydrolyzates labeled with lZ5Ior 1311 allows their dischargeable iodinated amino acid composition to be determined. The method used does not permit iodinated amino acids extracted from cells (hormones and precursors) to be distinguished from those released from the colloid after the hydrolysis of iodinated thyroglobulin. It is generally admitted that the iodinated amino acids extracted from the cells are negligible (Simon, 1964), so the results express only the variations of the iodinated amino acids contained in the follicular lumens. The distribution observed inXenopus (Natafet al., 1975) and in neotenic axolotls (Mauchamp and Nunez, 1968; Taurog, 1974) reflects the existence of different functional states. After 72 hours of labeling the percentages of MIT, DIT, TS, and T, discharged from the Xenopus thyroid gland during normal development and metamorphosis are close to those found in young rats (Vigouroux, 1974) and adult rats at isotopic equilibrium (Simon, 1964) (Table I).This observation suggests that in amphibians the mechanism of hormonogenesis is comparable to that occurring in mammals: iodotyrosine formation (MIT and DIT), which by coupling produces iodothyronines (T, and T3) (PittRivers, 1962; Simon, 1964).The fact that thyroid iodide distribution of iodotyrosines is identical throughout premetamorphosis up to the climax (MIT/DIT< 1) suggests that hormonal synthesis processes change greatly during this period. A high proportion of thyroid hormones (T, and T3)is always present, exceeding 20%during metamorphosis. During postclimax, however, this percentage decreases to 6%. Whatever its functional state, the Xenopus thyroid seems to be able to synthesize thyroid hormones. The situation is different in the axolotl (Table I), where MIT/DIT> 1 and where the proportion of T, and T, is always very low, not exceeding 2% (Mauchamp and Nunez, 1968; Taurog, 1974). This last result suggests a defect in iodotyrosine cou-

93

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

TABLE I DISTRIBUTIONOF LABELEDIODINATED AMINO ACID IN THE THYROIDGLANDOF Xenopus THROUGH LARVALDEVELOPMENT AS COMPARED TO THOSE FOR THE AXOLOTL AND THE RAT

Xenop us In oioo 72-hour bath Premetamorphosis Prometamorphosis Climax Postclimax H ypophysectomy Axolotl In uioo 48-hour injection* In uivo isotopic equilibrium bathb In oivo 48-hour injectionc Rat Adult, isotopic equilibriumd Hypophysectomye

MIT/DIT T,

+ T3

I-

Front

6.40 5.84 4.15 6.32 50.27

0.76 1.12 1.05 1.06 0.27

1

-

-

1.3

1.8

-

-

52.4

0.5

2.7

-

-

41.6

0.46 1.15

18.4 2

4.19 3.4

2.85 0.3

Origin

MIT

DIT

4.48 6 5.50 5.10 2.5

22.06 20.18 19.35 32.04 27.55

51.18 47.98 45.10 47.96 21.75

0.56 0.43 0.43 0.68 1.03

-

39

53

0.74

-

54

39

-

26.2

8.44 7.7

19.3 39

44

11.14 18.66 23.92 6.66 2.5

Values represent percentage of the chromatogram. Mauchamp and Nunez (1968). Taurog (1974). Simon (1964). Taurog (1958).

pling and consequently a decreased ability to synthesize hormones. This hypothesis is confirmed by a recent study in which no trace of T, was found in the blood of these animals (Taurog et al., 1974).

2. Quantitative Distribution of Stable Zodine The kind of analysis we have presented does not give results dealing with the quantitative distribution of iodine in the colloid, and autoradiographic studies on amphibians have not given any information in this respect (Gabrion and Sentein, 1972; Kaye, 1961; Neuenschwander, 1972). The use of an x-ray microprobe permits determination of the punctate concentration of iodine incorporated into colloid proteins (ZPC)in a constant analyzed volume estimated to be 10 pm3 (Regard, 1975).The relative value is given in impulses per second, and in Xenopus the quantitative distribution of iodine is variable, even within a follicle and from one follicle to the next. Heterogeneity is

94

ELIANE REGARD

especially evident during prometamorphosis, as the IPC value may triple, going from 10 to 30 impulses per second in a single follicle section (Fig. 6). During climax, most of the values are near 30 impulses per second (Fig. 6), while during postclimax the numerical distribution of the follicles according to their ZPC again becomes heterogeneous. The averages are between 5 and 20 impulses per second (Fig. 7). The variable density of the spots observed in the scanning picture of iodine reflects the inter- and intrafollicular heterogeneity of iodine distribution during prometamorphosis (Fig. 6) and postclimax (Fig. 7). A homogeneous distribution characterizes the different follicle sections at climax (Fig. 6). Colloid analysis of axolotl follicular structures shows a high ZPC, reaching generally 30 impulses per second. These values are in-

PROMETAMORPHOSIS

CLIMAX

30 -

b

20 -

I, I , I, 3

40

IPC Irnp/s

FIG.6. (a and b) Distribution of the ZPC in the colloid in a single follicle. A line is drawn for each value of the ZPC, and its length is proportional to the number of points with this value. (c and d) Distribution of iodine ('*'I) in a section of gland. Scanning electron microscope associated with an x-ray spectrometer.

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

95

POSTCLIMAX

a

LLA&

0 -0

10

20

Aver. Imp /s

HYPOPHYSECTOMY

, nb foll.

b

FIG. 7. Numerical distribution of the follicles (nb foll.) according to their average IPC (impulses per second) (a and b) and x-ray scanning picture of the follicular lumen at postclimax (c) and after hypophysectomy (d), showing the distribution of iodine.

dicated in Table 11. It is clear that the distribution of ZPC values is independent of the size of the follicle sections. The diversity of thyroid functional states in amphibians is evinced with respect to quantitative distribution of iodine only at some stages of development. This phenomenon may be typical of lower vertebrates. An extreme case is observed in a chimerical halocephalid fish. Large amounts of iodine are observed in voluminous phosphocalcic spherical crystals located in the follicular lumen, while it is not detected in the surrounding colloid (Martoja and Vu, 1971). In the rat, the distribution of iodine in colloids studied either by autoradiog-

96

ELIANE REGARD TABLE I1 AVERAGEIPc

IN DIFFERENT TYPESOF THYROID GLANDFOLLICLES AND IN THE WHOLE GLANDOF THE NEOTENIC AXOLOTL

Number and size of follicles per section 8 Large 8 Medium 12 Small

Average IPC (impulses/second)

Average IPC in the whole gland (impulses/second)

35.4 31.5 29.5

32.8

f

0.8

raphy (Simon and Droz, 1973a,b) or by x-ray spectrometry (Robison and Davis, 1969) is homogeneous and independent of follicle size. Iodine punctual concentration is at most multiplied by 2 in a single section of gland.

3. Measurement of Zodine Compartment of the Colloid The amount of iodine contained in the colloid, which we call the iodine compartment of the colloid Q, is the sum of the amounts of the element q present in the various follicular lumens of the gland (Fig. 8). Such a measure can be determined from the average ZPC per animal (I=) and the volume of the colloid. In Xenopus, during premetamorphosis, as early as stage 55 (Nieuwkoop and Faber, 1956) planimetric measurements indicate that colloid occupies 50 5% of the volume V of the gland. The iodine compartment of the gland is expressed by

*

Q (impulses per second)

=

Z X

x (V/2)

The results in relative units are listed in Table 111.

a

b

FIG. 8. Iodine flow and iodine compartment of the colloid. (a) A section of gland with five follicles, in which follicular epithelium limits the colloid.f, andf, are the elementary transepithelial iodine flows. The amounts of iodine contained in the colloid are, q l , q2,q3,q4, and q5. (b) The iodine compartment of the gland Q is the sum of the different compartments q I , q,, q3,q4, and q5.F , and F z are the sum of the elementary flows ( f I , f Z , . . .).

97

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

TABLE I11 MEASURE OF

THE IODINE COMPARTMENT OF THE COLLOID THROUGHOUT

LARVALLIFE Premetamorphosis

Xenopus stages Average ZPC (impulses/second) Average volume of gland (pm3) Average volume of the colloid (pm3) Iodide comparhnent of the colloid (impulses/ second)

Beginning

End

Prometamolphosis

Climax

Postclimax

1.9

8.2

14.8

23.1

13.7

0.28 x 105

8 x 1oJ

37 x lW

5 6 x 1oJ

47x 1oJ

0.14 x 1oJ

4 x 1oJ

18.5 x 1oJ

28 x 1oJ

23.5 x 106

0.27 x l(r 33 x l(r

274 x l(r

647 x l(r

322 x l(r

Between premetamorphosis and climax the thyroid grows, and the accumulation of iodine in the colloid characterizes a state of expansion (in comparison to the end of premetamorphosis, the factor of variation is 20). This result is contradictory to those of some investigators who, upon observation of collapsed follicles, proposed that an emptying phenomenon of the colloid occurs during climax (Sax& et a1 ., 1957b). In the days following metamorphosis contraction occurs; the mean volume of the gland remains mostly unchanged, but the iodine compartment of the colloid becomes twice as small as during climax. In the neotenic axolotl, where the colloid occupies 60 2 5% of the volume of the gland, the IPC values remain high. In this case the glands can store large amounts of iodine.

4. Relative Intensity of Transepithelial Iodine Flow The variations in the colloid iodine compartment reflect the difference between iodine uptake F , and iodine secretion F , (Fig. 8).They represent the sum of the flows,f, and fi,that cross each cell. The difference between the two flows determines the direction of the variation in the colloid iodine compartment. In Xenopus, where Q increases until climax, iodine uptake is greater than iodine secretion. It is also known that the protein-bound iodine (PBI) value triples between premetamorphosis and climax (Saxkn et al., 1957a,b; Just, 1972). Moreover, our recent results concerning serum T, and T, in spontaneously metamorphosing tadpoles of Rana catesbeiana show a gradual rise in the two hormones from premetamorphosis to the beginning of climax, followed by a sharp rise

98

ELIANE REGARD

during climax. During this period the average values for serum T, reach 0.55 pg/lOO ml, and for T,, 78 ng/100 ml (Regard and Taurog, 1977). We deduce from these two observations that iodine intake is itself increasing (Fig. 9). During postclimax the decrease in Q implies that F , is higher than F , . But the PBI value and the level of thyroid hormone decrease simultaneously. Consequently, during the first days following metamorphosis the ability of the epithelium to take up

PREMETAMORPHOSIS

PROMETAMORPHOSIS

CLIMAX

[Ti==F'

Q3

POSTCLIMAX

FIG.9. Evolution of the iodine compartment of the colloid (right) and a representation of transepithelial flows (left)for each period of larval development. The circles representing compartment Q,the length of arrows symbolizing the flows ( F , and F2),the height of the cells are proportional to the intensity of the corresponding physiological processes.

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

99

iodine decreases considerably. The peroxidase activity permits this conclusion, as it becomes more difficult to detect in the cells. The significant accumulation of iodine in the thyroid of the axolotl, where iodine uptake is presumed to be slow, does not appear contradictory, since the gland lacks secretory activity (Taurog, 1974).The involvement of iodine throughout neotenic development is limited to iodine uptake and storage in the colloid.

c.

EFFECTSO F HYPOPHYSECTOMY

The functional state of the thyroid gland of amphibians is not excluded from hypophyseal control. Histological and physiological arguments favor a decrease in thyroxine secretion following surgery. Metamorphosis stops, and the follicular epithelium flattens (Regard, 1975; Regard and Mauchamp, 1971).The processes of iodine utilization are themselves modified, as the ability to take up iodine decreases considerably. The radioactivity of the glands represents only 0.05% of the radioactivity of the incubation bath, the operated Xenopus larvae remaining in the bath 3,5, or 10 days ( E . Regard, unpublished data). 1. Qualitative and Quantitative Analysis of the Iodine Content of the Colloid The amount of dischargeable iodinated amino acids in the glands is highly modified in hypophysectomizedXenopus, as shown in Table I. The MIT/DIT ratio becomes greater than 1, and the amount of T, and T, decreases from 18 to 2.5%.Values in the same range are found in the hypophysectomized rat (Taurog et al., 1958) (Table I). The agreement between these results suggests that hypophyseal control is exerted over the ability to synthesize hormones. Moreover, the distribution of iodine remains heterogeneous from one follicle to another and goes from an average range of 10 to 30 impulses per second to a range of 8 to 16 impulses per second (Fig. 7). The low values of the ZPC reflect a decrease in the iodine compartment of the colloid, the volume of the gland remaining almost the same after hypophysectomy. This is a consequence of hypophyseal control exerted on iodine uptake.

2. Relative Intensity of Transepithelial Flow The decrease in the iodine compartment of the colloid expresses a variation in transepithelial flow. Chromatographic analysis of hydrolyzates ofXenopus thyroid glands placed in water containing 1251for 3,5, or 10 days reveals 50%iodide. This especially high value shows a de-

100

ELIANE REGARD

crease in the ability of the gland to incorporate the element organically. The percentage of radioiodide fixed by the glands being very low, it is evident that hypophysectomy results in a decrease in iodide uptake. However, the presence of a high percentage of labeled iodide may indicate labilization in the binding of iodide to proteins, and consequently significant deiodination during preparation of the thyroid for chromatography (Taurog, 1963). Moreover, the absence of localized peroxidase activity after preparation explains why the iodide taken up by the glands is incorporated into proteins in a low ratio (Regard and Mauchamp, 1973b). Finally, the interruption of metamorphosis suggests a decrease in iodine secretion. The variations in the iodine compartment of the colloid demonstrate the hypophyseal control exerted over transepithelial iodine flow. This control is also exerted over the synthesis and/or activation of peroxidase. The same kind of effect is also observed in mammals, as in hypophysectomized rats iodine uptake decreases (Nunez, 1970; Taurog, 1970; Taurog et al., 1958) and no peroxidase activity is observed in thyroid cells from hypofunctioning glands (Tice, 1974; Dempsey and Peterson, 1955). The types of hypophyseal control exerted over iodine metabolism are analyzed in Section 111.

111. Thyroglobulin Biosynthesis Thyroglobulin molecules, which act as the substrate in halogenation, constitute the essential part of the colloid (Edelhoch, 1965).The possibility of an expanded or a contracted state of the iodine compartment of the colloid may explain the variability in the capacity of the follicular epithelium to synthesize thyroglobulin. SYNTHESIS A. RELATIVE MEASUREMENT OF THYROGLOBULIN After 6 hours of incubation with t y r o ~ i n e - ~the H sedimentation profiles of the soluble synthesized proteins show a peak corresponding to thyroglobulin. As in other vertebrates (Brisson, 1972; Hoshino and Ui, 1970) the sedimentation coefficient in Xenopus and in the axolotl is close to 19s (Mauchamp and Nunez, 1968; Regard and Mauchamp, 1971).Analogous values have been found in other anurans and in urodeles (Peyrot et al., 1973).Based on the incubation time used in the experiments, the 12s subunit is present in a very low concentration, and a peak of light molecules is also observed (3-8s).

101

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

1. Evolution of Thyroglobulin Synthesis throughout Larval Development and Metamorphosis During premetamorphosis the relative amount of thyroglobulin synthesized (12 and 19s)represents 10% of the total soluble proteins. This percentage increases, reaching a maximum of 73% at the end of prometamorphosis and at the beginning of climax. A decrease in the relative amount of thyroglobulin synthesized appears during postclimax. Nevertheless it still represents 40% of the total soluble proteins. The curve of the variations in this capacity for synthesis is shown in Fig. 10. The existence of different functional states throughout larval development and metamorphosis is expressed by a variable ability to synthesize thyroglobulin.

,oo

%

50

1

Thyroglobulin

0

-

04 54

55

. .

56

57

I

5059606i62 63 6465 66+

PREMETAMORPHOSIS PROMETAMORPHOSIS CLIMAX

!

stage

POSTCLIMAX

FIG.10. Diagrammatic representation of thyroid cells and of the curve corresponding to the percentage of thyroglobulin present in soluble proteins newly synthesized by the gland after 9 hours of incubation with tritiated tyrosine, at different stages of development of the Xenopus larva.

102

ELIANE REGARD

2. Intensity of Thyroglobulin Synthesis during Neotenic Development

The percentage of thyroglobulin synthesized after 6 hours of incubation is relatively high and represents 50-60% of the total soluble proteins (Mauchamp and Nunez, 1968).A knowledge of ultrastructural characteristics may allow the interpretation of such results. B. ULTRASTRUCTURAL FEATURES RELATED TO THYROCLOBULIN SYNTHESIS The measured thyroglobulin synthesis is a global approximation resulting from the summation, of the activity of each thyroid cell. In Xenopus, it reflects the activity of a homogeneous population of cells (Regard and Mauchamp, 1971) and, in the neotenic axolotl, the activity of a heterogeneous population (Mauchamp and Regard, 1971; Regard, 1974).

1. Homogeneous Cell Populations Throughout the larval life of Xenopus, the variations in f'ollicular epithelium height are accompanied by structural changes. The ergastoplasm, which is the site of protein synthesis, is especially modified. During premetamorphosis rough endoplasmic reticulum is constituted of vesicles located in the supranuclear area of the cells. During climax the infranuclear area is occupied by ergastoplasmic cisternae seen in a parallel arrangement. During postclimax the development of these formations is more restricted. These different aspects are shown in Fig. 10. They suggest an increase in protein synthesis until climax and a decrease in this synthesis during postclimax. There is a correlation between the ability to synthesize thyroglobulin, which is maximal at climax, and the development of ergastoplasmic structures. The adaptation of cells to synthesize increasing amounts of thyroglobulin involves two kinds of reactions: (1) nonspecific reactionssynthesis of the elements of the reticulum (phospholipids, membrane proteins, proteins, and rRNA); and (2) specific reactions-thyroglobulin synthesis, requiring the synthesis of a specific mRNA (Vassart,

1972). 2. Heterogeneous Cell Populations We have already described the morphological manifestations of the fluctuations in cell activity in the thyroid of the axolotl. Ergastoplasmic formations are well represented in the high cells and seldom in the flat cells (Regard, 1974). The presence of many inactive cells is not

103

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

compatible with active thyroglobulin synthesis. The presence of active cells explains the high proportion of thyroglobulin synthesized by the glands under our experimental conditions (Fig. 5).

c.

EFFECTSOF HYPOPHYSECTOMY

A hypophysectomy performed at the end of prometamorphosis in Xenopus produces extensive changes in thyroid cells. The structural characteristics are similar to those observed during premetamorphosis. In all the cells, the ergastoplasmic cisternae, numerous at the time of the operation, disappear 10 days later. A few ergastoplasmic vesicles remain (Fig. 11).Under these conditions the amount of neosynthesized thyroglobulin represents only 10% of the total soluble proteins. Fifteen days after hypophysectomy thyroglobulin is no longer found in the soluble protein. 100

50

0

10

20

30

40

Fr

FIG.11. Diagrammatic representationofthyroid cells and ofthe curves showing the relative amount of thyroglobulin newly synthesized by the glands after hyphophyseo tomy. (a) Xenopus. (b) Axolotl.

104

ELIANE REGARD

In the axolotl, hypophysectomy results in the disappearance of active cells. Then the epithelium presents an homogeneous appearance. Its very flat cells contain some ergastoplasmic vesicles and look like the flat cells described in normal animals (Fig. 11). The relative amount of thyroglobulin synthesized represents 10-25% of the total soluble protein 15 days after the operation (Fig. 11). The close correlation that exists between the ability to synthesize thyroglobulin and the development of ergastoplasmic formations is under the control of the hypophysis. Such control exists in higher vertebrates also. After hypophysectomy, the epithelium of the rat thyroid flattens and contains few ergastoplasmic formations (Dempsey and Peterson, 1955). The amount of thyroglobulin synthesized decreases by about 60% compared to that in the controls (Rappaport, 1970). In conclusion, the studies described in Sections I1 and I11 indicate that the hypophyseal control of thyroid function in amphibians determines the variations in intensity of iodine flow as well as of thyroglobulin synthesis. These variations are accompanied by corresponding ultrastructural adaptations. However, in the neotenic axolotl, in a given gland, the ultrastructural adaptations are various and reflect different functional states. In spite of the hypophyseal control, which is applied at the level of the follicular epithelium, fluctuations in cellular activity are observed. The well-known activating part played by TSH is not questioned. TSH cells have been identified in amphibians (Etkin, 1968; van Oordt, 1974),and the hypophysis of the axolotl is not an exception (Connelly, 1973; Hauser-Gunsbourg et al., 1973). But in the axolotl it has been suggested that the low rate of TSH secretion explains the neotenic state and the fluctuations in cellular activity; the hormonal message received by some cells would spread along the follicular epithelium with attenuation (Regard and Mauchamp, 1973b; Taurog, 1974).This interpretation is questionable, since thyroid activity is in fact regulated by two hypophyseal hormones: TSH and prolactin. Arguments supporting this biohormonal regulation in amphibians are given in Section IV.

lV. Hypophyseal Regulation An antagonism between prolactin and TSH in amphibians is suggested by the interruption of metamorphosis in anuran larvae treated with prolactin (Etkin, 1968; Etkin et aZ., 1969; Etkin and Gona, 1967a,b; Gona, 1967; Gona et aZ., 1970; Dodd and Dodd, 1976). No proof of the possible effect of prolactin on thyroid structure and function has been given, except in a study indicating that the hormone may interfere with iodide uptake mechanisms (Gona, 1968). Such an

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

105

effect should be added to the different activities influenced by prolactin in vertebrates: osmoregulation (fishes), hydrotropism and growth (amphibians),molt cycle (reptiles), functioning of the gonads and mucous glands of the crop (birds), and functioning of corpus luteuni and mammary glands (mammals) (Bern and Nicoll, 1968). Knowledge of the different kinds of controls exerted b y the hypophysis on thyroid function (Sections I. and 11) guided the research related to the nature of the regulation of thyroid activity in these amphibians. The effects of hormone treatment on the ultrastructure are studied, as well as the ability to use iodine shown by the thyroid of Xenopus tadpoles at climax or of larvae hypophysectomized at the end of prometamorphosis (Regard and Hourdry, 1975). The experiments lasted 6 days, during which the animals received three injections of prolactin [42 international milliunits (ImU)each], TSH (6ImU each), or a mixture of prolactin and TSH.

A.

SENSITIVITY OF THYROID GLAND TO EXOGENOUS OVINE PROLACTIN IN LARVAE AT METAMORPHIC CLIMAX

1. Ultrastructural Characteristics The involution of ergastoplasmic formations, well developed at the beginning of the treatment, result in a great decrease in the height of the cells, which does not exceed 5 pm (Fig. 12). Some ergastoplasmic vesicles, however, remain surrounded by many free ribosomes. Such changes reflect compared to the controls, a decrease in thyroid activity. The regression of ergastoplasmic formations also suggests a decrease in protein synthesis, especially of thyroglobulin.

2 . Ability to Use Iodine The effects of the treatment on the ability to take up and incorporate iodide organically have been investigated in vitro and are summarized in Table IV. When morphological transformations of the larvae are slowed down, iodide uptake is four times weaker than in the controls. It is only two times as weak when the climax is only slightly delayed, as caudal regression does not occur in larvae. In the various cases most of the iodine is incorporated into proteins. Nevertheless, the distribution of peroxidase activity is limited to some vesicles and to some Golgi saccules. These two kinds of results show that the ability to use iodine is less important than in the controls and also suggest a decrease in iodine uptake. The concentration of iodine in the colloid is high, the ZPC value often exceeding the values registered throughout climax (Table V). The high frequency of iodine spots observed on the scanning pictures of iodine shows that the ZPC values have a nar-

106

ELIANE REGARD

A-XENOPUS LARVA AT CLIMAX treatment Prolactin 42 ImU/in]ection

RA = 2985 f 5 2 corn /lobe

IR = 32. 2 5 0.5 Imp/s

8 - HYPOPHYSECTOMIZED XENOPUS LARVA treatment TSH 6 IrnU/injection

RA = 5800 f 15 cpm/lobe

= 32 f 0.6 Imp/s treatment TSH (6 ImUl + Prolactin (42 1mU)

RA = 4081 f 5 8 cpm /lobe

treatmenlTSH (6 ImU) + Prolactin ( 8 4 ImU)

RA = 705 5 2 0 cpm/lobe

lm = 45 f 0.7 Imp/s

FIG. 12. Effects of hormonal treatment on the structure of the thyroid cells, on the localization of peroxidase activity (dark organelles in the drawings), and on the iodine distribution in sections of glands (a, b, c). RA:radioactivity of the glands after 3 hours of incubation with lz5I. ZPC, Average IPC for the whole animal in a given experimental assay.

TABLEIV EFFECTSOF PROLACTIN, TSH,

AND A

OF

MIXTUREOF PROLACTIN AND TSH

RADIOIODIDE BY

THE

ON THE in Vitro UPTAKE THYROIDGLANDOF Xenopus LARVAE=

AND

ORGANIC INCORPORATION

Stage 58, hypophysectomized Stage 57, Stage 57, Initial stage and early, no late, no treatment treatment treatment Stage at the end 58 of the experiment Total iodine 1897 63 (3,W Protein bound 1864 87 iodine Iodide 3821

* *

Stage 63, no treatment

59-60

66

6043 (L4)

892 (194)

-

6 ImU TSH

42 ImU prolactin

6 ImU TSH plus 42 ImU prolactin

6 ImU TSH plus 84 ImU prolactin

58

60

58

58-60

58-60

35 (1.2)

-

5800 2 15 (2,2) 5089 2 15

40 f 8 (3,4) 31 2 6

4081 2 58 (3,4) 3781 f 75

705 2 20 (4,Z)

6002 27

-

50523

523

262213

6625

No Stages 59 and 60,42 ImU prolactin treatment

60

62

63

1590 f 45 2985 2 52 2693 f 132 (3,2) (592) (%4) 1376 f 60 2942 2 59 2130 * 127 62 f 7

11 k 4

120 * 16

Each treatment included three injections. Glands were incubated for 3 hours at 29°C in 0.2 ml Wolff and Quimby amphibian medium (Gibco) containing and 1.75 pCi/ml '"I. Results are expressed in counts per minute per lobe. The number of incubations and the number of lobes used per incubation are given in parentheses.

(lo-' M )lz'I

108

ELIANE REGARD

TABLE V EFFECTOF PROLACTINAND/OR TSH TREATMENT ON THE IPc IN THE COLLOID Stage 58, hypophysectomized

Tadpole stages and treatments Number of animals Iodine punctate concentration (impulses/second)

Average

Stages 58-60, 42 ImU prolactin

No treatment

6 ImU TSH

6 ImU TSH plus 84 ImU prolactin

5

2

3

3

25.6 f 0.8 29.5 f 0.7 36.3 2 1.3 45.6 f 1.6 26.0 -e 0.5 32.2 2 0.6

10.4 f 0.4 10.7 2 0.5

19.7 f 0.5 43.5 f 1.6 36.4 f 0.9

-

-

39.0 f 0.8 52.9 2 1.5 42.7 2 0.9

10.6

32.7 2 0.9

45.0 f 0.7

-

-

row range. Because of the weak intensity of the flow of iodide toward the colloid, the retention of iodine observed can be explained only by the analogous decrease in the flow in the direction of the interfollicular space. The morphological and functional changes following treatment with prolactin are numerous. All the results, summarized in Fig. 12, suggest a decrease in secretory activity in the gland corresponding to the delay in metamorphosis events as indicated by the persistence of the tail in larvae treated with prolactin. It is, however, difficult to connect these observations with an effect due to prolactin itself. Injection of the hormone at a period when the hypophysis is especially active may result in complex feedback phenomena, leading secondarily to a rest phase in gland activity. Because of this problem the other parts of the study were performed on hypophysectomized larvae. B. RESPONSE OF THE THYROID GLANDOF HYPOPHYSECTOMIZED LARVAETO TSH

1. Ultrastructural Adaptations The height of the follicular epithelium of larvae operated on at the end of prometamorphosis increases, from 7 to 25 pm. Ergastoplasmic formations are well developed and can be observed even in the supranuclear area; cellular polarity apparently is lost under these conditions (Fig. 12).This phenomenon shows the limits of intracellular regulations in a system no longer in equilibrium. Under the influence of TSH amounts certainly higher than normal, only the activating pro-

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

109

cess occurs, as the hypophyseal retroinhibitory phenomenon cannot take place (Etkin, 1968). The development of ergastoplasmic formations is a morphological expression of thyrotropic stimulation.

2 . Restoration of the Ability to Use Zodine In operated animals the injection of TSH restores the capacity for iodide uptake and organic incorporation. The results of the in vitro study are reported in Table IV. The values observed are in the same range as those observed in larvae at climax. Moreover, thyrotropic treatment induces the reappearance of peroxidase activity, which is again localized in the perinuclear space, the ergastoplasmic cisternae, some Golgi saccules, and many microvesicles close to the apical membrane (Fig. 12). These two observations suggest the restoration of increased iodide flow under the influence of thyrotropic stimulation. In addition, the average value of the ZPC reaches and exceeds that found in the animals at climax (Table V). Scanning pictures of iodine show a homogeneous distribution of the element in the different follicular lumens. The thyrotropic treatment causes an increase in the iodine compartment of the colloid, compatible with the increase in iodide uptake. The recovery of metamorphosis events suggests that the secretion of iodine is intense, but less than the uptake. In hypophysectomized Xenopus larvae, as in other vertebrates, the thyrotropic treatment modifies the structural characters of the thyroid and induces an acceleration of various metabolic processes (Nkve and Dumont, 1970; Setoguti, 1973a,b; Tice, 1974; Wagar et al., 1973; Wollman, 1969).Once more we emphasize that the variations in the iodine compartment of the colloid are only a reflection of the hypophyseal control exerted on transepithelial iodine uptake F and secretion F , (Section I). It appears that TSH causes an increase in F , and F , in such a way that F,> F , and that Q increases. The effect of prolactin, on the contrary, causes a decrease in F , and F 2 , F , remaining higher than F,, so that Q increases also. Knowledge of the iodide compartment of the thyroid alone does not give evidence for the hormonal effects involved.

,

C. CHARACTERISTICS OF THE THYROIDGLANDIN HYPOPHYSECTOMIZED LARVAETREATEDWITH PROLACTIN OR WITH BOTH PROLACTIN AND TSH I N VARIOUS RATIOS The observed responses depend largely on the amount of prolactin (42 or 84 ImU) contained in the mixture of prolactin and TSH, as the concentrations of TSH remain the same (6 ImU) (Fig. 12).

110

ELIANE REGARD

1. Mixture Cmtaining 42 I m U of Prolactin and 6 ImU of T S H Cells initially resting show different reactivation characteristics. The follicular epithelium appears to be formed with high cells which reach about 20 pm in height. The ergastoplasm invades the intranuclear area, and its numerous cisternae are seen in a parallel arrangement (Fig. 12).

2. Mixture Containing 84 ImU of Prolactin and 6 ImU of T S H a. Ultrastructural Adaptations. In this case, the height of the epithelium remains lower than 10 pm. The infranuclear region is reduced and is occupied by ergastoplasmic formations. Thus the structural adaptation of thyroid cells to a hormonal mixture containing TSH and 42 ImU of prolactin is a response of the gland to thyrotropic stimulation (Fig. 12). However, the mixture of TSH and 84 ImU of prolactin produces a low level of thyroid activity close to that observed in larvae treated at climax with prolactin only (Fig. 12). b. Effects on Iodine Metabolism. Prolactin alone (42 ImU) does not result in changes in the iodine uptake of the glands, which remains very low, as in hypophysectomized, untreated control larvae. The results of the study in vitro in the different experimental cases are listed in Table IV. Note that the mixture ofTSH and (42 ImU) of prolactin stimulates iodine uptake. The values observed are similar to those obtained after treatment with TSH only. In contrast, the hormonal mixture containing 84 ImU of prolactin results in a low iodine uptake. The values observed are three times lower than in larvae at prometamorphosis. Under these conditions the stimulating effect induced by thyrotropin on iodine uptake in vitro is masked. Whatever the treatment, most of the iodide taken up is incorporated into proteins, while the amount of free iodide is low (Table IV). Peroxidase activity, involved in the iodide oxidative pathway, can be localized in the thyroid cells of larvae treated with the two mixtures. However, this activity is especially intense in numerous microvesicles close to the apical membrane when the mixture of TSH and 42 ImU of prolactin is used (Fig. 12). In the other case (TSH and 84 ImU of prolactin) peroxidase activity is less widely distributed in the cells. The reaction product is observed only in some perinuclear spaces, some ergastoplasmic cavities, and apical microvesicles (Fig. 12). Cells without activity are numerous. These results show, indirectly, the existence of more-or-less intense iodine flow depending on the hormonal mixture used. The more-or-

111

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

less important distribution of peroxidase activity suggests that this enzyme is the limiting factor in iodine incorporation (Tice, 1974). Table V summarizes the effects of treatment with the mixture of TSH and (84ImU) of prolactin on the quantitative distribution of iodine in the colloid. Higher than the average value determined in hypophysectomized larvae, the ZPC also exceeds the values usually reached in larvae at climax or in hypophysectomized larvae treated with TSH. The scanning picture of iodine in sections of glands shows a homogeneous distribution of iodine in the different follicles. The increase in the iodine compartment of the colloid reflects the control exerted by the hormonal mixture on the flow of iodine through the epithelium. c. How Prolactin Acts. The morphological and functional adaptations of thyroid cells in Xenopus larvae after the different hormonal treatments are summarized in Table VI. Two facts appear: (1)Prolactin can decrease the activity of thyroid cells in metamorphosing larvae. (2) Prolactin can mask the stimulating effect of TSH on the follicular epithelium of hypophysectomized larvae. Such conclusions suggest that the physiological state of the gland may be modulated by the simultaneous production of the two hormones in various ratios, as proposed by Etkin (1968). Two possibilities can be suggested to explain how prolactin acts. It may decrease thyroid activity directly or indirectly, competing with TABLE VI HORMONALTREATMENTS ON THE STRUCTURE OF THYROID CELLS AND ON SOME ASPECTSOF IODINE METABOLISM

INFLUENCE OF

Stage 58, hypophysectomized

Tadpole stages and treatments Ergastoplasm Dense bodies Colloid vesicles In vitro uptake and organic incorporation of radioiodide Peroxidase activity Iodine punctate concentration of the colloid

Stages 58-60, 42 ImU prolactin

6 1mU TSH plus 42 ImU prolactin

6 ImU TSH plus 84 ImU prolactin

-

+++ ++ ++

+ +

+

++ ++

-

-

++

All treatments included three injections. -, Rare, weak;

+ +, frequent, high; + + +, very frequent, very high.

-

+

+++ +, infrequent, average;

112

ELIANE REGARD

thyrotropin. Prolactin may act too, in a direct or indirect way, at the peripheral level on T, receptor sites. This peripheral antithyroid effect would produce secondarily a resting state in the gland. The validity of such an interpretation is suggested by several studies. On the one hand prolactin can inhibit the events of metamorphosis induced by T, (Blatt et ol., 196Y), and on the other hand it exerts in vitro a stabilizing action on lysosomal membranes in tail or gut cells. Consequently the degenerative processes usually induced by thyroid hormones are delayed (Giunta et ol., 1972). D. HYPOPHYSEAL REGULATION OF THYROID ACTIVITY BASEDON A BIHORMONALMODEL AND NEOTENIC DEVELOPMENT The thyroid of the axolotl, which has numerous inactive cells, a very high iodine compartment of the colloid, and low iodine flow is similar to the thyroid of Xenopus larvae treated with prolactin at climax. How can the hypothesis of a bihormonal regulation of thyroid activity be applied to the axolotl and explain neoteny? Many arguments favor such an hypothesis. The dosage of prolactin does not reveal any difference between larval neotenic axolotls and ambystoma (Norris et al., 1973). Moreover, a recent study suggests a very low level of TSH secretion in the axolotl, metamorphosis being induced by repeated injections of TSH (Taurog, 1974). The secretion of both prolactin and TSH in such a ratio that the plasmic concentration of each hormone does not change may explain neoteny. The morphological heterogeneity of the follicular epithelium would reflect the antagonism (direct or indirect) between the two hormones. E. PRODUCTION OF THE TWO HORMONES IN DIFFERENTRATIOS The modulation of thyroid function by prolactin and TSH poses the question of the control of these two hypophyseal secretions. Many investigators have proposed the regulation of prolactin by a prolactininhibiting factor (PIF) which appears at the climax of metamorphosis, while the thyrotropin-releasing hormone (TRH) maintains a high secretion of TSH (Etkin, 1968). Another point of view is that the production of TSH may be inhibited by a thyrotropin-inhibiting factor (TIF) (Rosenkilde, 1972). With a very sensitive radioimmunological method, TRH has been identified in the hypothalamic and extrahypothalamic area of the brain of the neotenic axolotl, salamander, and frog (Taurog et al., 1974). The amounts of TRH observed in the frog and the neotenic axolotl are high, while paradoxically the secretion of

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

113

TSH is presumably low. TRH may regulate the secretion of hypophyseal hormone other than TSH, such as prolactin. Recent works show that in mammals TRH is responsible for TSH as well as prolactin discharge. Studies were done on rats (Tashjan et al., 1971), ewes (Kann et al., 1973), sheep (Debeljuk et al., 1973), and humans (Bowers et al., 1971; Repetoff et al., 1974; Snyder et al., 1973). In sheep previous treatment with T, inhibits the increase in serum prolactin concentration usually induced by TRH injections (Debeljuk et al., 1973). Analogous results are obtained in humans (Snyder et al., 1973). The regulation of both TSH and prolactin production by TRH appears to be very sensitive in mammals, at the level of circulating thyroid hormones. The control of TRH by simultaneous prolactin and TSH production in a variable ratio is plausible in amphibians as long as the T I F or PIF has not been identified. It is difficult to give, for this class of vertebrates, the real curves of variation in TRH level and of the prolactin/TSH ratio throughout development. Nevertheless, the great sensitivity of immunological dosages should allow this problem to be solved and the diagram proposed by Etkin (1968) to be completed.

V. Conclusions It is generally admitted that throughout phylogenesis the mode of action of thyroid hormones changes rather than thyroid function itself (Barrington, 1964). The functional unit is represented either by endostyle cells in protochordates or thyroid follicles in vertebrates (Constantinescu, 1972). This interpretation assumes the specialization of target tissues and an adaptative relationship between thyroid hormones and effectors; in this respect metamorphosis if a good example. No evidence of the thyroid function itself is observed; only simple changes in the intensity of the main characteristics of the tissue are seen because of early differentiation of the gland in amphibians. From premetamorphosis to climax, the transition from a resting stage to a hyperactive stage takes place through a series of structural and metabolic adaptations. A close parallelism appears between the development of ergastoplasmic formations and an increase in thyroglobulin synthesis. The development of the ergastoplasm is the most important at climax, when thyroglobulin synthesis reaches a maximum. At the same time the distribution of peroxidase activity, the enzyme responsible for iodide oxidation, undergoes large variations; difficult to localize at premetamorphosis this activity is especially important at climax. Such activity favors the organic incorporation of

114

ELIANE REGARD

iodide, which increases during this period. This is confirmed by an increase in the iodine compartment of the colloid. From premetamorphosis to climax iodide uptake increases and remains higher than iodine secretion. During postclimax the activity of the gland reaches an equilibrium state. The ergastoplasmic formations regress at the same time that the percentage of thyroglobulin synthesized decreases. Although the size of the gland changes slightly, the amount of iodine observed in the colloid is twice as low as at climax. The secretion of thyroid iodine is then greater than the uptake, and the ability to fix iodide decreases considerably. In neotenic urodeles, the morphological heterogeneity of the gland reflects a high functional heterogeneity. High, active cells in which ergastoplasm is well represented are located next to flat, inactive cells with a high nucleocytoplasmic ratio. Though thyroglobulin synthesis is high, iodination is low. The result is a storage of thyroglobulin and iodine, since electrophoresis of the serum does not reveal any trace of T,. The thyroid function itself does not seem to change throughout phylogenesis, while the regulation of the activity of the gland reveals a certain evolution. In protochordates the neuroendocrine correlations are in these respects poorly known (Barrington, 1964). In vertebrates thyroid function is under hypophyseal control, and thyrotropin seems to play a dominant role in the activating processes (Constantinescu, 1972).We have indeed observed that hypophysectomy in Xenopus or the axolotl produces an hypofunctional state characterized by decreased thyroglobulin synthesis and iodide utilization. In amphibians the very different functional states of the thyroid gland, though controlled by the hypophysis, result in a special regulatory pathway in which, besides TSH, prolactin interferes. The effects of the latter are variable, depending on the class of vertebrate (Bern and Nicoll, 1968). As expected, the treatment of larvae with prolactin (a 42-ImU injection) delays metamorphosis. The regression of ergastoplasmic formations proves the decrease in its utilization. But iodine storage is observed in the colloid. These results allow speculation that the secretory activity of thyroid is reduced, as indirectly proved by the decline in metamorphosis. The effects of a mixture of 6 ImU of TSH and either 42 or 84 ImU of prolactin administered to hypophysectomized Xenopus larvae demonstrates more precisely the sensitivity of thyroid cells to prolactin. The mixture of TSH and 42 ImU of prolactin restores thyroid activity, as revealed by the structural characteristics of the cells (ergastoplasmic formations are well developed) or by io-

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

115

dine metabolism (intense iodide uptake and widely distributed peroxidase activity). These modifications are correlated with thyrotropic stimulation of the gland. The mixture of TSH and 84 ImU of prolactin produces, on the contrary, relatively low thyroid activity. The ultrastructural (few ergastoplasmic formations) and metabolic (low iodide uptake and limited distribution of peroxidase activity) characteristics give evidence of this phenomenon. These results show that the physiological state of the amphibian thyroid is modulated by both hormones, the effect on the gland depending on the preponderance of one of the two hormones. Consequently the physiological state of the gland throughout larval development determining metamorphosis or neoteny may depend on the prolactin/TSH ratio in the plasma. The different information obtained for amphibians and mammals suggests that hypophyseal regulating processes have evolved throughout phylogenesis. Thyroid regulation by prolactin and TSH consequently result in the appearance of extreme physiological states in the former. In mammals TSH is the regulator of thyroid function, although a hypothalamic factor, TRH, enhances the simultaneous secretion of prolactin and TSH in different ratios. In other classes of vertebrates the regulation of thyroid activity by prolactin and TSH has seldom been studied (Bern and Nicoll, 1968).More data are necessary to understand how the endocrine interrelationships in hypothalamic pituitary axis have evolved from one class of vertebrates to another. REFERENCES

Alexander, N. M., and Corcoran, B. J. (1962).J.Biol. Chem. 237,243-248. Banington, E. J. W. (1964).“Hormones and Evolution.” English Universities, London. Bern, H. A., and Nicoll, C. S. (1968).Recent Prog. Homn. Res. 24,681-720. Blatt, L. M., Slickers, K. A., and Kim, K. H. (1969). Endocrinology 85, 1213-1215. Blount, R. F. (1950).J.Exp. Zool. 113, 717-739. Bowers, C. Y., Friesen, H. G., Hwang, P., Guyda, H. J., and Folkers, K. (1971).Biochem. Biophys. Res. Commun. 45, 1033-1041. Brisson, A. (1972). These Doct. Sci. Nut., Paris 6 , AO-CNRS-7963. Coleman, R., Evennett, P. J., and Dodd, J. M. (1968). Gen. Comp. Endocrinol. 10,3446. Connelly, T. G. (1973). Gen. Comp. Endocrinol. 20, 236-255. Constantinescu, E. (1972). Rez;. Roum. Embryol. Cytol. 9, 205-232. Debeljuk, L., Arimura, A , , Redding, T., and Schally, A. V. (1973).Proc. SOC. Exp. Biol. Med. 142,421-423. Deiss, W. P., and Peake, R. L. (1968). Ann. Intern. Med. 69, 881-890. Dempsey, E. W., and Peterson, R. R. (1955). Endocrinology 56,46-58. Dodd, M. H. I., and Dodd, J . M. (1976).I n “Physiology of the Amphibian” (B. Lofts, ed.), Vol. 3, pp. 467-576. Academic Press, New York.

116

ELIANE REGARD

Edelhoch, H. (1965). Recent Prog. Horm. Res. 21, 1-31. Ekholm, R., and Smeds, S. (1966).J.Ultrastruct. Res. 16, 71-82. Etkin, W. (1968).In “Metamorphosis, a Problem in Development Biology” (W. Etkin and L. I. Gilbert, eds.), pp. 313-348. Appleton, New York. Etkin, W., and Gona, A. G. (1967a). Life Sci. 6,703-707. Etkin, W., and Gona, A. G. (1967b).J.Exp. Zool. 165,249-258. Etkin, W., Derby, A., and Gona, A. G. (1969). Cen. Comp. Endocrinol., Suppl. 2,253259. Flickinger, R. A. (1964). Gen. Comp. Endocrinol. 4,285-289. Fox, H., and Turner, S. C. (1967).Arch. Biol. 78,61-90. Fujita, H. (1969). Virchows Arch. B 2,265-279. Gabrion, J., and Sentein, P. (1972). C. R. Skances SOC. Biol. Ses Fil. 166, 142-146. Geloso, J. P. (1967). Ann. Endocrinol. 28, 1 bis. Giunta, C., Campantico, E., Vietti, M., and Guastalia, A. (1972). Gen. Comp. Endocrinol. 18,568-571. Gona, A. G. (1967). Endocrinology 81,748-754. Gona, A. G. (1968).Anat. Rec. 160,355. Gona, A. G., Pearlman, T., and Etkin, W. (1970).J.Endocrinol. 48,585-590. Gorbman, A. (1964). In “Physiology of the Amphibia” (J. A. Moore, ed.), pp. 371-425. Academic Press, New York. Gorbman, A., and Ueda, K. (1963). Gen. Comp. Endocrinol. 3,308-311. Graham, R. C., and Kamovsky, M. J. (1966).J.Histochem. Cytochem. 14,291-302. Haddad, A., Smith, M. D., Herscovics, A., Nadler, N. J., and Leblond, C. P. (1971).J . Cell Biol. 49, 856-882. Hanaoka, Y. (1966).J.Fac. Sci., Hokkaido Unio. 16, 106-112. Hanaoka, Y., Koya, S. M., Kondo, Y., Kobayashi, Y.,and Yamamoto, K. (1973). Gen. Comp. Endocn’nol. 21,410-423. Hauser-Gunsbourg, N., Doerr Schott, J., and Dubois, M. P. (1973). Z . Zellforsch. Mikrosk. Anat. 142,539-548. Hoshino, T., and Ui, N. (1970). Endocrinol. Jpn. 17,521-533. Hosoya, T. (1968). Gunma Symp. Endocrinol. 5,219-237. Hosoya, T., Matsukawa, S., and Nagai, I. (1971). Biochemistry 10,3086-3093. Huxley, J. (1920).Nature (London) 104,435. Jost, A. (1961). Colloq. Tireoide, Inst. Biofis., Rio deJaneiro pp. 81-109. Just, J. (1972). Physiol. Zool. 45, 143-152. Kann, G., Habert, R., and Denamur, R. (1973).C. R. Hebd. Sdances Acad. Sci. 276,13211324. Kaye, N. (1961). Gen. Comp. Endocrinol. 1,l-19. Larsen, J. H., Jr. (1968).J . Ultrastruct. Res. 24,190-200. Leblond, C. P., and Gross, J. (1948). Endocrinology 43,306-324. Leloup, J., and Fontaine, M. (1960).Ann. N.Y. Acad. Sci. 86,316-353. Martoja, M., and Vu, T. T. (1971).J.Microsc. (Paris) 12,41-50. Mauchamp, J., and Nunez, J. (1968). Cen. Comp. Endocrinol. 10,47-55. Mauchamp, J.,and Regard, E. (1971). C. R. Skances SOC. Biol. Ses Fil. 165, 229-232. Nadler, N. J., Leblond, C. P., and Bogoroch, R. (1954). Endocrinology 54, 154-172. Nadler, N. J., Young, R. A., Leblond, C. P., and Mitmaker, B. (1964).Endocrinology 74, 339-354. Nakai, Y., and Fujita, H. (1970).Z. Zellforsch. Mikrosk. Anat. 107,104-110. Nataf, B., and Sfez, M. (1967).Ann. Endocrinol. 28, 1-79.

AMPHIBIAN THYROID GLAND CYTOPHYSIOLOGY

117

Nataf, B., Regard, E., and Hourdry, J. (1975). C. R. Hebd. Skances Acad. Sci. 280,657660. Neuenschwander, P. (1972). Z. Zellforsch. Mikrosk. Anat. 130,553-574. N&ve, P., and Dumont, J. E. (1970). Z. Zellforsch. Mikrosk. Anat. 103,61-74. Nieuwkoop, P. D., and Faber, J. (1956). Normal table ofXenopus laevis (Daudin). NorthHolland Publ., Amsterdam. Nonis, D. O., Jones, R. E., and Criley, B. B. (1973). Gen. Cornp. Endocrinol. 20,437442. Novikoff, A. R . (1971). Electron Microsc., Proc. Znt. Congr., i'th, 1970 pp. 565-566. Novikoff, A. B., Beard, M. E., Albala, A., Sheib, B., Quintana, N., and Biempica, L. (1971).J. Microsc. (Paris) 12,381-404. Nunez, J. (1970). Rapport final DGRST. Contrat. 6600268. Peyrot, A., Andreoletti, G. E,, Pons, G., Crotta, L., and Vottero, C. (1973). Gen. Comp. Endocrinol. 21,461-467. Pitt-Rivers, R. (1962). Biochem. J . 82, 108-111. Rappaport, L. (1970). These Doct. Sci., Paris, AO-CNRS 4854. Regard, E. (1974). Gen. Comp. Endocrinol. 23, 199-211. Regard, E. (1975). Thhse Doct. Sci. Nut., Paris 11, AO-CNRS 11825. Regard, E., and Hourdry, J. (1975).J . Microsc. Biol. Cell. 22,39-54. Regard, E., and Mauchamp, J. (1971).J. Ultrastruct.Res. 37,664-678. Regard, E., and Mauchamp, J. (1973a). C. A. Hebd. Skances Acad. Sci. 276, 121-123. Regard, E., and Mauchamp, J. (1973b).J . Microsc. (Paris) 18,291-306. Regard, E., and Taurog, A. (1977). Proc. 59th Annu. Meet. Am. Endocr. Soc., Chicago. Regard, E., Hourdry, J., and Boumati, P. (1975). Electron Microsc., Proc. Znt. Congr., 8th, 1974 Vol. 2, pp. 88-89. Repetoff, S., Fang, V. S., Rapoport, B., and Friesen, H. G. (1974).J . Clin. Endocrinol. Metab. 38,450-457. Robison, W. L., and Davis, D. (1969).J.Cell Biol. 43, 115-121. Rosenberg, L. L., and Cavalieri, R. R. (1971). Endocrinology 8 9 , 4 3 9 4 4 . Rosenkilde, P. (1972). Gen. Comp. Endocrinol., Suppl. 3 , 3 2 4 0 . Salvatore, G., Sena, L., Viscidi, E., and Salvatore, M. (1965). Curr. Top. Thyroid Res., Proc. Int. Thyroid Conf., 5th, 1965 pp. 193-206. Sax&, L., SaxBn, E., Toivonen, S., and Salimaki, K. (1957a). Endocrinology 61,35-44. SaxBn, L., SaxBn, E., Toivonen, S., and Salimaki, K. (1957b). Ann. Zool. SOC.Zool. Bot. Fenn. Vanamo 18,l-44. Seed, R. W., and Goldberg, I. H. (1965).J.Biol. Chem. 240,764-773. Seljelid, R. (1967).J. Ultrastruct.Res. 17, 195-219. Setoguti, H. (1973a).Z. Zellforsch. Mikrosk. Anat. 137, 177-194. Setoguti, H. (197313).Z. Zellforsch. Mikrosk. Anat. 137, 195-208. Shin, W. Y., Ma, M., Quintana, N., and Novikoff, A. B. (1971). Electron Microsc., Proc. Int. Congr., 7th, 1970 pp. 565-566. Simon, C. (1964). Thise Doct., Rapp. C.E.A. R 2590. Simon, C., and Droz, B. (1973a).J.Physiol. (Paris),66,51-64. Simon, C., and Droz, B. (1973b).J . Physiol. (Paris) 66,65-77. Simon, C., and Morel, F. (1960). Int. J . Appl. Radiat. Zsot. 8,35-45. Snyder, P. J., Jacobs, L. S., Utiger, R. D., and Daughaday, W. H. (1973).J. Clin. Inoest. 52,2324-2329. Stein, O., and Gross, J. (1964).J . Endocrinol. 75,787-798. Strum, J . M., and Karnovsky, M. J. (1970).J.Cell Biol. 44,655-666.

118

ELIANE REGARD

Strum, J. M., Wicken, J., Stanbury, J. R., and Karnovsky, M. J. (1971).J.Cell B i d . 51, 163- 175. Tashjan, A. H., Barowsky, N. J., and Jensen, D. K. (1971). Biochem. Biophys. Res. Commun. 43,516-523. Taurog, A. (1963). Endocrinology 73,57-62. Taurog, A. (1970). Recent Prog. Horn. Res. 26, 189-247. Taurog, A. (1974). Gen. Comp. Endocrinol. 24,257-266. Taurog, A., Tong, W., and Chaikoff, I. L. (1958).Endocrinology 62,646-663. Taurog, A., Oliver, C., Eskay, R. L., Porter, J. C., and McKenzie, J. M. (1974). Gen. Comp. Endocrinol. 24,267-279. Tice, L. W. (1974).Endocrinology 95,421-453. Tice, L. W., and Wollman, S . H. (1972). Lab. Inoest. 26, 1, 63-73. Triantaphyllidis, E., and Verne, J. (1963).Ann. Endocrinol. 24,39-37. van Oordt, P. G. W. J. (1974).In “Physiology of the Amphibians” (B. Lofts, ed.), Vol. 2, pp. 52-106. Academic Press, New York. Vassart, G. (1972). FEBS Lett. 22,53-56. Vigouroux, E. (1974). ThBse Doct. Sci. Nut., Paris 6, AO-CNRS 10057. Wagar, G., Ekholm, R., and Bjorkman, U. (1973). Acta Endocrinol. (Copenhagen) 72, 453-463. Wissig, S . L. (196O).J.Biophys. Biochem. Cytol. 7,419-432. Wissig, S. L. (1964).In “The Thyroid Gland” (R. Pitt-Rivers and W. Trotter, eds.), Vol.1, pp. 32-70. Butteworth, London. Wollman, S. H. (1969). Front. Biol. 14B, 483-512.

The Macrophage as a Secretory Cell ROY C. PAGE Department of Pathology and Periodontics and the Center f o r Resenrch in Oral Biology, University of Washington, Seattle, Washington

PHILIP DAVIES The Merck Institute f o r Therapeutic Research, Rahway, New Jersey AND

A. C. ALLISON Division of Cell Pathology, Clinical Research Centre, Northwick Park, Harrow, Middlesex, England

I. Introduction . . . . . . . . 11. Regulation of Stem-Cell Growth . . . . 111. Substances Affecting Fibroblast Growth and Activity IV. Antimicrobial Substances . . . . . . A. Interferon . . . . . . . . B. Antibacterial Factors . . . . . . V. Substances Activating or Regulating Host Defense against Bacteria, Viruses, and Tumor Cells . . A. Factors Mecting T-Lymphocyte Activity . . B. Factors Mecting B-Lymphocyte Activity . . C. Complement. . . . . . . . D. Pyrogen . . . . . . . . . E. PMN Chemotactic Agent . . . . . VI. Prostaglandins and Cyclic Nucleotides . . . VII. Cytotoxic Substances . . . . . . . VIII. Hydrolytic Enzymes . . . . . . . A. Acid Hydrolases . . . . . . . B. Neutral Proteinases . . . . . . C. Consequences of Enzyme Release by Macrophages References . . . . . . . . .

. .

.

119 120 121

.

124 124

.

125

. .

125 125 128 129 131 132 132 134 137 138 146 150 151

.

. .

. . . . .

. . . .

I. Introduction The macrophage was one of the first cell types to be established successfully in primary culture (Carrel and Ebling, 1926),and it has been used extensively for the study of a wide variety of biologicrl phenomena (see Gordon and Cohn review, 1973).Since the time of Metchni119

120

ROY C. PAGE ET AL.

koff, in the early nineteen hundreds, macrophages have been considered an important part of normal host defense systems, predominantly because of their capacity to ingest and destroy microorganisms, foreign substances, and damaged tissue components (Metchnikoff, 1905).Recently, it has become apparent that macrophages participate in a much broader range of activities than previously suspected. In addition to the classically described functions of tissue debridement and phagocytosis and killing of microorganisms, macrophages also participate in numerous other activities including (1) stimulation of leukocyte stem-cell growth; (2)production and release of numerous humoral substances which are active in various host defense mechanisms, including complement, pyrogen, interferon, and factor(s) chemotactic toward neutrophils; (3) interaction with immunocompetent lymphoid cells during humoral and cell-mediated immune responses and in some immunopathological processes; (4) regulation of the replicative and synthetic activities of fibroblasts; (5)destruction of malignant cells; (6) production and secretion of substantial quantities of hydrolytic enzymes including acid hydrolases, neutral proteinases, and lysozyme; and (7) production and release of various prostaglandins and cyclic nucleotides. The capacity of mononuclear phagocytes to undergo marked morphological changes, both in vivo (Ebert and Florey, 1939) and during culture in vitro (Lewis, 1925; Camel and Ebeling, 1926; Weiss and Fawcett, 1953) is well established. Cohn and his colleagues determined optimal conditions for the culture of mouse peritoneal macrophages and carried out extensive studies on the properties of the cells in vitro (for reviews, see Cohn, 1968, 1970; Gordon and Cohn, 1973). Macrophages vary considerably with regard to their levels of cellular protein and enzymes, depending on their site of origin and the stimuli to which they have been exposed (Cohn and Wiener, 1963; Cohn and Benson, 1965; Saito and Suter, 1965).Although macrophages undergo mitosis in vitro only rarely, they synthesize large amounts of protein. The levels of cellular constituents are very responsive to the composition of the extracellular environment and to the uptake of substances by pinocytosis and phagocytosis. That mononuclear phagocytes release biologically important substances in response to environmental factors now seems unquestionable, although in many instances these substances have not been identified or characterized biochemically.

11. Regulation of Stem-Cell Growth Mice given endotoxin show proliferation of pluripotent stem cells in the bone marrow (Eaves and Bruce, 1974) and a transient elevation

THE MACROPHAGE AS A SECRETORY CELL

121

in serum levels of a substance that has the capacity to support growth of marrow cells in vitro (Metcalf, 1971; Quesenberry et al., 1972; Chervenik, 1972). This substance, referred to as colony-stimulating activity (CSA), has also been detected in urine, in medium conditioned by the culture of various cells and tissues, and in extracts of several organs and tissues. The cell source of the substance appears to be the monocyte or macrophage. When human blood monocytes are used to condition medium, or as a feeder layer, colonies of granulocytes and mononuclear cells can be established from both human and murine bone marrow (Chervenik and LoBuglio, 1972). Lymphocytes do not support colony formation, and neutrophils are actually inhibitory. Cultures of unstimulated human blood monocytes produce detectable quantities of CSA, and production is enhanced two- to threefold by stimulation of the cells with either endotoxin or polyinosinic-polycytidylic acid (Ruscetti and Chervenik, 1974). Maximal stimulation of CSA production was observed with 0.01 and 1.0 puglml of these substances, respectively. Further evidence that the monocvte and macrophage are sources of the CSA found in vivo was provided by Eaves and Bruce (1974), who showed that murine peritoneal macrophages maintained in vitro in the presence of endotoxin produce large quantities of CSA. The kinetics of production by these cells in vitro closely parallels the appearance of CSA in the serum of endotoxin-treated animals, and both activities decrease in a similar manner when serially diluted. CSA purified about 100,000-foldis a glycoprotein that migrates electrophoretically between a,-globulin and albumin and has a molecular weight of between 45,000 and 60,000 (Stanley et al., 1975).There is some evidence that macrophages may also produce an inhibitor of granulocyte and monocyte production (Ichikawa et al., 1967).

111. Substances Meeting Fibroblast Growth and Activity Interactions between macrophages and fibroblasts may play an important role in normal and aberrant wound healing, acute and chronic inflammation, and diseases characterized by fibrosis such as pneumoconioses. The nature of these interactions is not understood, and their elucidation has begun only recently. Pulmonary lesions resulting from the inhalation of toxic particles and dusts are made up almost totally of macrophages, fibroblasts, and fibrotic connective tissue substances. Toxic particles and dusts are taken up and reside within the alveolar macrophages. These lesions have served as models for investigation of the role of macrophages in

122

ROY C. PAGE ET AL.

pathological fibroblastic activity. Several factors derived from macrophages or their supernatant culture fluids, which affect fibroblast synthetic activities, have been described. Heppleston and Styles (1967) found a factor in cultures of macrophages incubated with quartz dust, which enhanced hydroxyproline production by fibroblasts derived from chick embryo 3- to 15-fold relative to control fibroblast cultures. However, this result could not be confirmed by Harington et al. (1973), who found a 6 1 4 1 % inhibition of collagen production b y hamster fibroblast cultures to which suspensions of frozen and thawed silica-treated hamster macrophages had been added. When the same system was used, silica-treated rat macrophages had no effect on the synthetic activity of chick fibroblasts, while there was a decrease in the amount of collagen formed by hamster fibroblasts. Kilroe-Smith et a2. (1973)exposed guinea pigs to quartz dust or to the intratracheal injection of quartz, after which the alveolar macrophages were harvested. The sonically disintegrated cells yielded an insoluble factor which, when injected subcutaneously in guinea pigs, caused granuloma formation. Thus, whether or not quartz- and silica-treated macrophages elaborate fibrogenic factor(s) remains unresolved. I n more recent experiments, the supernatant culture fluids from rat peritoneal macrophages obtained by lavage, and cells induced by prior injection of thioglycollate or paraffin, were incubated in vitro in the presence and absence of silica, and the fluids subsequently analyzed for factors affecting collagen synthesis (Aalto et al., 1976). The assay system was based on the incorporation and conversion of radiolabeled proline into hydroxyproline by slices of sponge granulation tissue incubated in the macrophage culture fluids. Supernatant from silica-containing macrophage cultures significantly enhanced collagen synthesis by the tissue slices. Medium from cultures of thioglycollate- or paraffin-induced macrophages incubated without silica were also stimulatory, and the addition of silica to these cultures did not enhance the level of stimulatory activity further. Viable cells were not necessary; the fibrogenic substance was also detected in preparations of cells ruptured by freeze-thawing and subsequently exposed to quartz. When supernatant fluids from the various cell cultures were tested on fibroblast cultures from embryonic chick tendon, they exerted no effect on collagen synthesis. In fact, macrophage cultures incubated without silica significantly inhibited the synthesis of DNA, RNA, collagen, and protein by these cells. Monocytes appear in large numbers in clean, experimentally induced wounds; their appearance is followed by the immigration of large numbers of fibroblasts and collagen production. The interaction

THE MACROPHAGE AS A SECRETORY CELL

123

of these cells in healing wounds has been studied by Leibovich and Ross (1975). Experimental wounds essentially free of monocytes and macrophages were created by the local injection of antimacrophage serum at the wound site in guinea pigs made monocytopenic b y the systemic injection of hydrocortisone. In these wounds, fibroblasts, which usually appear 3 days after wounding, did not appear until the fifth day, and their rate of proliferation was retarded. In addition, wound debridement did not occur. The. same investigators (Leibovich and Ross, 1976) examined culture fluids of peritoneal macrophages maintained in culture for factors affecting fibroblast activities. The macrophage culture medium activated DNA synthesis and cell growth in quiescent fibroblasts. The level of enhancing substance reached a maximum during the first 6 hours of macrophage culture, and this level was not significantly affected by phagocytosis of latex particles or of zymosan. The level of enhancing activity was reduced significantly by repeated freezing and thawing of the cells, demonstrating that viable cells were necessary. The activity, which was not detected in cultures of lymph node cells, could not be dialyzed and was heatstable at 56°C for 30 minutes. Calderon et al. (1974) described a factor released from mouse peritoneal macrophages maintained in culture, which inhibits DNA synthesis of several cell types including 3T3 fibroblasts. The substance could be absorbed by cells, was resistant to digestion by trypsin and phosphodiesterase, and was dialyzable. In subsequent experiments the same investigators (Calderon and Unanue, 1975), using mouse peritoneal exudate cells induced with proteose-peptone, found both inhibitors and stimulators of DNA synthesis. While the nondialyzed culture medium inhibited DNA synthesis, after dialysis it stimulated DNA synthesis b y spleen cells activated with phytohemagglutinin (PHA). There are also indications that fibroblast products may influence some macrophage functions and that a type of feedback mechanism may be operative between these two cell types. Mauel and Defendi (1971) reported that macrophages, which usually do not divide in culture, when grown in medium previously conditioned by mouse fibroblasts, synthesize DNA within 24-48 hours following exposure. More recently, it was shown that culture medium from L cells can induce DNA synthesis by macrophages previously activated and maintained in serum-containing medium (Cifone et al., 1975).Ascites fluid from mice carrying an Ehrlich ascites tumor and injected with complete Freund’s adjuvant also exhibits activity. Macrophages can also be induced to proliferate in vitro either by peritoneal lavage fluid

124

ROY C. PAGE ET AL.

from mice previously stimulated with dextran of molecular weight 40,000 (Adolphe et al., 1975) or by exudates obtained from diffusion chambers previously filled with synthetic culture medium and inserted surgically into the peritoneal cavity of mice for 4 days (Wynne et al., 1975).Mitogenic factors from fibroblasts may account for the incorporation of thymidine into DNA b y a small proportion of macrophages in vivo and by macrophages at sites of chronic inflammation (Mariano and Spector, 1974).

IV. Antimicrobial Substances A. INTERFERON The important role macrophages play in host defense against viruses has been reviewed in detail elsewhere (Allison, 1974). It is significant that these cells secrete interferon. Early reports that peripheral blood leukocytes and cells from tissues mediating reticuloendothelial cell functions produced interferon (Gresser, 1961; Kono and Ho, 1965) were followed by more definitive reports of its production by mouse peritoneal (Glasgow and Habel, 1963; Hirsch et al., 1970) and rabbit alveolar macrophages (Acton and Myrvik, 1966).Smith and Wagner (1967a) studied in some detail the production of interferon by rabbit peritoneal macrophages induced by glycogen. Cells infected with Newcastle disease virus showed production and secretion of interferon within 2 hours of virus adsorption, with maximal synthesis occurring between 2 and 6 hours afterward. Actinomycin D inhibited the synthesis of interferon if added within 4 hours of virus adsorption; addition at a later time had no effect. This was attributed to the formation of stable mRNA during the initial 4-hour period after addition of the virus. Puromycin also inhibited interferon synthesis, showing that its production in macrophages followed the transcriptional and translational pathways described in other cells. Macrophages not exposed to virus were also shown to produce interferon, but only about one-hundredth of the amount obtained after the exposure of cells to virus. The production of interferon by cells not exposed to virus may have been stimulated by the presence in the culture medium of endotoxin which is known to induce interferon secretion by macrophages. These workers also compared the physical properties of macrophage interferon with those of interferon produced by rabbit kidney cells infected with Newcastle disease virus (Smith and Wagner, 1967b). It was found that, in addition to a product with a molecular weight of approximately 45,000 produced by both rabbit kidney cells and macrophages, the lat-

THE MACROPHAGE AS A SECRETORY CELL

125

ter cells secreted interferon of molecular weight 37,000 in the absence of viral induction. Macrophages also synthesized interferon of polydisperse molecular weights after stimulation with endotoxin. Although it has been shown that interferon is active against viruses, intracellular parasites, and tumor cells, its contribution to the role played by macrophages in host defense against these agents remains to be established. Macrophages have been shown to augment the production of interferon by both T and B lymphocytes exposed to mitogens. These studies have been well reviewed by Epstein (1976). It is of interest that she describes the production of an interferon antagonist by freshly explanted human peripheral blood monocytes but not by more mature macrophages maintained in culture for 1 week. B. ANTIBACTERIALFACTORS Macrophages are generally considered to participate in host defense against bacterial infection through their capacity to phagocytize, kill, and digest microorganisms, especially if the organisms have been opsonized. However, Bast et al. (1974) showed that additional mechanisms may be involved and that secretory activity may be important. Macrophages from bacillus Calmette-Gukrin (BCG)-immune guinea pig peritoneum, incubated in vitro with purified protein derivative (PPD), released a listericidal factor into the culture medium. Considerably less material was released by nonimmune cells incubated with antigen and by immune cells incubated without antigen. The factor, which was not heat-sensitive at 60°C for 30 minutes and not lost through dialysis, was not found in cultures of fibroblasts or hepatoma cells.

V. Substances Activating or Regulating Host Defense against Bacteria, Viruses, and Tumor Cells Evidence is rapidly accumulating that macrophages participate in host defense systems not only through their ability to phagocytize and kill microorganisms and eliminate damaged or abnormal cells and connective tissue constituents, but also through a much broader range of activities including secretion of substances that activate or regulate other host defense mechanisms.

A. FACTORS AFFECTINGT-LYMPHOCYTE ACTIVITY Macrophages play an essential role in both the induction and expression of immune responses (Unanue, 1972; Rosenthal, 1975). Immunological response to certain antigens requires the presence of

126

ROY C. PAGE ET AL.

macrophages as well as B and T lymphocytes. Cell cluster formation facilitates the presentation of antigens to T lymphocytes by macrophages (Mosier, 1967; Miller and Avrameas, 1971; Schmidtke and Unanue, 1971), and soluble factors can substitute for the macrophages in these systems (Shortman and Palmer, 1971; Hoffmann and Dutton,

1971). Gery and his colleagues (1971, 1972) showed that, under certain conditions, supernatants from human peripheral leukocytes or syngeneic splenocytes stimulate the incorporation of tritiated thymidine into mouse thymocytes and splenocytes. The release of a factor, termed by these investigators lymphocyte-activating factor (LAF), is increased by treating the producing cells with endotoxin or PHA. LAF was not active in whole spleen cell preparations. However, fractions of spleen cells prepared on discontinuous bovine serum albumin gradients and containing lymphocytes, but not macrophages, were stimulated b y LAF. Furthermore, spleen cells recovered from animals 1 month following lethal irradiation and reconstituted with thymocytes exhibited activity, but similar preparations from animals reconstituted with bone marrow cells or with spleen cells stimulated with endotoxin were not responsive to this factor. B lymphocytes did not appear to be target cells for LAF. Identification of the macrophage as the source of LAF (Gery and Waksman, 1972) was based on the separation of adherent cells, using plastic surfaces and nylon columns. In mice, adherent cells from bone marrow and spleen produced this factor. In general, the cells were found to produce some LAF in the absence of any stimulus, but more LAF was produced by exposing the cells to PHA or concanavalin A (Con A), and most was produced by exposing them to endotoxin. LAF was also produced by monocytes from human peripheral blood and adherent cells from rabbit spleens. Gery and Wiener (1975) showed that the production of LAF by mouse peritoneal macrophages was not increased by stimuli known to cause an increase in cellular levels of lysosomal enzymes. Culture medium containing 25% newborn calf serum was used in vitro as a stimulus for acid hydrolase production, while an intraperitoneal injection of thioglycollate broth was needed to induce cells with high levels of LAF. Gery and Handschumacher (1974) characterized LAF obtained from human leukocytes and mouse peritoneal macrophages as a nondialyzable substance with a molecular weight of approximately 15,000, which is partially destroyed by heating at 60°C but which is stable over a pH range of 5.5 to 10.5. Yoshinaga et al. (1975) have also reported that supernatants from murine macrophages induced with sodium caseinate enhance the response of thymocytes to PHA.

T HE MACROPHAGE AS A SECRETORY CELL

127

The observations of Gery et al. have been recently extended by Calderon and his colleagues (Calderon and Unanue, 1975; Calderon et al., 1975; Unanue et al., 1976). They found a 20-fold enhancement in thymidine uptake b y syngeneic and allogeneic thymocytes incubated in supernatants from mouse peritoneal macrophages induced with proteose-peptone. Such activity was not obtained with splenic lymphocytes. The stimulatory activity resides in a fraction eluting from Sephadex G-100 columns corresponding to a molecular weight ranging from 15,000 to 21,000 (Calderon et al., 1975).This material is resistant to treatment with trypsin, RNase, carboxypeptidase, papain, and neuraminidase, but is completely destroyed b y chymotrypsin and pepsin. Further studies by Unanue et a,?. (1976) showed that the release of LAF from macrophages can be stimulated by phagocytic stimuli such as latex particles, Listeria vaccine, and antibody-coated lymphocytes, and also by endotoxin and beryllium sulfate. Most of the activity is released within the first 48 hours of challenge by the various stimulants. Macrophages from mice challenged intraperitoneally with endotoxin or thioglycollate broth, agents that stimulate other macrophage activities, do not release LAF when cultured in uitro. Moreover, such cells are refractory to the effect of stimuli that cause secretion of LAF by proteose-peptone-induced macrophages. Cultures from mice infected with Listeria monocytogenes release large quantities of this substance. In this system, the release of activity appears to be associated with the presence of activated T lymphocytes, although the lymphocytes clearly do not produce the stimulating factor. This possibility was confirmed by the observation that the addition of stimulated T lymphocytes to unstimulated macrophages led to the production of LAF. Thus it seems likely that macrophages activated with lymphokines may produce LAF in addition to their other secretory activities. Diamanstein and Ulmer (1976) showed that murine peritoneal macrophages can be stimulated to produce LAF by exposure to guanosine nucleotides. The stimulatory effects of 3 ’ , 5’-guanosine monophosphate are seen at a concentration of 5 mM and are dependent on the synthesis of RNA and protein. Further evidence for the importance of soluble factor(s) with properties similar to those of LAF has come from the studies of Rosenstreich et al. (1976). Using cultures in which macrophages and lymphocytes were separated in double-chambered vessels of the Marbrook type, these investigators demonstrated the dependence of the mitogenic response of guinea pig lymph node T lymphocytes on soluble factors from macrophages. The macrophage factor(s) could not be replaced by the reducing agent, 2-mercaptoethanol.

128

ROY

C.

PAGE ET AL.

The significance of LAF remains to be established. Clearly, it may play a role in expanding populations of T lymphocytes, which would facilitate the expression of their various functions. It is not clear whether different subpopulations of T lymphocytes respond to LAF differently; this is a problem for future investigation. Erb and Feldmann (1975a,b) provided evidence that mouse peritoneal macrophages secrete two products which support a he1per-Tcell function. Supernatants obtained from macrophages incubated with antigen for several days generate helper-cell activity when added, in the absence of antigen or macrophages, to T cells with identical I-A and H-2 complex regions. Erb and Feldmann (1975b) suggest that the specificity of the response is dictated by an antigenic fragment associated with the macrophage supernatant designated genetically related macrophage factor (GRF). Another activity was induced in macrophages incubated in the absence of antigen and, as its activity in generating helper-cell activity was not seen to be genetically restricted, has been termed nonspecific macrophage factor (NMF). N M F facilitated the generation of helper-cell activity in the presence of an insoluble antigen (keyhole limpet hemocyanin bound to Sepharose) but not of soluble antigens such as keyhole limpet hemocyanin and amino acid polymers. Differences in the kinetics of release of the two factors were observed. More CRF was released during the third day in culture, while more NMF was released during the second day in culture. B. FACTORS AFFECTING B-LYMPHOCYTE ACTIVITY Mouse peritoneal macrophages (Schrader, 1973; Wood and Gaul, 1974; Calderon et al., 1975; Unanue et al., 1976), and also human monocytes (Wood and Gaul, 1974; Wood and Cameron, 1975, 1976; Wood et al., 1976), produce factors that enhance antibody production by B lymphocytes. Evidence that this factor(s) acts directly on B cells has been derived from studies on lymphocytes from athymic nude mice and on T-lymphocyte-depleted spleen cell preparations. Wood and his colleagues conducted detailed studies of the B-lymphocyte-activating factor (BAF) released by human monocytes. They showed that, although BAF activity is not found in human peripheral blood monocytes at the time of their isolation, its production and release can be enhanced by maintaining these cells in culture. Endotoxin, PHA, and the polyene antibiotic mycostatin induce BAF release, but phagocytic stimuli such as latex particles and antigen-antibody complexes are not effective in this respect (Wood and Cameron, 1975). The release of BAF occurred after a brief lag period and

THE MACROPHAGE AS A SECRETORY CELL

129

was completed approximately 48 hours after adding the stimulus. This behavior differentiated BAF from LAF and from colony stimulating factor (CSF). Also, it was differentiated from plasminogen activator by its retention of activity on B lymphocytes in plasminogen-free serum. Wood et al. (1976) showed that BAF elutes from Sephadex G-75 in the position of spherical molecules with a molecular weight of 18,000. C. COMPLEMENT Recent advances in technology have allowed the study, in tissues and cells maintained in vitro, of the synthesis of several biologically active components of the complement system (Table I). These studies have been the subject of an excellent review by Colten (1976). Stecher (1970) showed that peritoneal and alveolar macrophages from several species synthesized and secreted proteins which reacted with specific anti-& and anti-& antisera. Biological activity of the respective complement components, namely, C3 and C4,was not demonstrated. More recently, Lai et al. (1975) have also provided evidence that human macrophages secrete Cl,, C3, and C4.Synthesized products were detected by autoradiography of precipitin lines obtained by interaction with specific antisera. Bentley et al. (1976) demonstrated the synthesis and secretion of immunochemically detectable C3 by mouse peritoneal macrophages.

TABLE I SECRETION OF COMPLEMENT COMPONENTS BY MACROPHAGES Complement component

c1, c2 c3 c4

c5 C6 Factor B

Tissue and species of origin of macrophage Monkey peritoneal and lung Guinea pig peritoneal Human peripheral blood Human peripheral blood Human peripheral blood Mouse, rabbit, monkey and human peritoneal Monkey, human, and rat peritoneal Guinea pig peritoneal Human fetal tissues Rabbit liver Mouse peritoneal

References Stecher (1970) Wyatt et al. (1972) Einstein et (11. (1975) Einstein et al. (1976) Lai et al. (1975) Stecher (1970) Stecher (1970) Littleton et al. (1970); Wyatt et al. (1972) Kohler (1973) Rother et al. (1968) Bentley et al. (1976)

130

ROY C. PAGE ET AL.

Convincing evidence is available that macrophages also synthesize and secrete C2. Wyatt et al. (1972)demonstrated the production of hemolytically active C2 by guinea pig peritoneal exudate cells, while Colten (1972) showed that C2 secretion by human fetal liver cells was associated with a population of large, mononuclear cells. The most detailed study of C2 synthesis and secretion b y mononuclear phagocytes is that made by Einstein et ul. (1976) on human peripheral blood monocytes maintained in tissue culture over long periods. They found that biologically active C2 was secreted into the culture medium over a period of several months. This secretion was reversibly inhibited by low concentrations of cycloheximide, as was the incorporation of 14Clabeled amino acids into the biologically active protein. Secretion increased with time in culture, being approximately three times as much during the eighth week as during the second week of culture. They also observed a lag period before the onset of secretion; this may be a function of the maturation of the monocytes in culture. As expected, monocytes from homozygous CZdeficient humans secreted no C2 in vitro over an 8-week culture period (Einstein et al., 1975). The synthesis of C2 and C4 by guinea pig peritoneal macrophages is inhibited by certain carcinogens (Colten and Borsos, 1974), while other noncarcinogenic agents such as turpentine and complete Freund’s adjuvant injected intramuscularly induce an increase in C2 and C4 secretion by cultured peritoneal macrophages from these animals (Colten, 1976). Bentley et (11. (1976) demonstrated the synthesis of factor B, a component of the alternative pathway of complement activation, by mouse periotoneal macrophages. Amino acid precursors were incorporated, and the protein secreted in a functionally active form. Synthesis and secretion were inhibited by cycloheximide. Production continued at a relatively constant rate over an experimental period of 4 days. The biological activity of factor B was demonstrated by its capacity to deplete the hemolytic activity of purified guinea pig C3 in the presence of factor D. Synthesis was inhibited by cycloheximide, while addition of amino-14Cacids to the macrophage culture medium resulted in synthesis of radioactive factor B. The mechanisms controlling the synthesis and secretion of complement by macrophages are unclear. Both Stecher (1970) and Colten (1976) showed that serum stimulates C2 and C4 synthesis by macrophages, while phagocytosis of heat-killed pneumococci results in up to 10-fold increases in the synthesis of C2 and C4 by guinea pig peritoneal macrophages (Colten, 1976).Stecher (1970) found that a small amount of hydrocortisone substituted for serum is a requirement for the production of C3 by rat but not by mouse macrophages.

THE MACROPHAGE AS A SECRETORY CELL

131

In many of the instances discussed here the contribution made by the macrophage to the overall synthesis of various complement components may not be a major one. However, such secretion may be of great significance at sites of inflammation where activated complement components may initiate chemotaxis, cell proliferation, or lysis and secretion of cellular products. D. PYROGEN In a variety of mammalian species, including man, fever is caused by the direct action of pyrogen on the hypothalamic thermoregulatory center (Atkins and Bodel, 1974).Endogenous pyrogen was first noted in large quantities in sterile peritoneal exudates of rabbit polymorphonuclear leukocytes (PMNs) (Bennett and Beeson, 1953; King and Wood, 1958), and subsequently in rabbit peritoneal macrophages (Hahn et al., 1967; Atkins et al., 1967) and in human monocytes (Bodel and Atkins, 1967). The production and release of pyrogen from mononuclear phagocytes and from tissues such as lung, spleen, and lymph nodes is greatly increased by phagocytic stimuli (Bodel and Atkins, 1967; Atkins et al., 1967). Pyrogen is also released from the alveolar macrophages of BCG-stimulated rabbits (Atkins et al., 1967). The release of pyrogen is an active process dependent on temperature and requiring protein synthesis (Atkins et al., 1967). Serum antibody and complement are not essential for its release. Human peripheral blood monocytes do not show a spontaneous release of pyrogen, even after 5 days in culture (Bodel, 1974).However, endotoxin exposure and phagocytosis induce pyrogen production, which begins within 4 hours and continues for at least 24 hours. Both the production and secretion of pyrogen are inhibited by puromycin. Dinarello et al. (197413) distinguished between pyrogens from human PMNs and mononuclear phagocytes on the basis of their physical characteristics and biological activity. Monocyte pyrogen is a larger and more acidic protein than that from PMNs. It has a molecular weight of 38,000 and an isoelectric point of 5.1. Its activity is destroyed during incubation with PMNs. Fever induced by monocyte pyrogen was found to be slower in onset than fever induced by P M N pyrogen, and defervescence was also slower. Monocytes produce 20 times as much pyrogen as an equivalent number of PMNs. The separation of these two pyrogens should allow a better understanding of the role played by endogenous pyrogen secreted by mononuclear phagocytes in clinical fever. The cellular mechanisms by which endogenous pyrogen synthesis is initiated remain unclear. Bodell (1976) showed that low concentraand 1 x lo-' M ) initiate the synthesis tions of colchicine (2.5 x

132

ROY C. PAGE ET AL.

and secretion of pyrogen by human peripheral blood monocytes but not by rabbit leukocytes. These observations are difficult to reconcile with those of the antipyretic effect of colchicine in familial Mediterranean fever (Dinarello et al., 1974a). E. PMN CHEMOTACTIC AGENT Stevenson (1974) demonstrated that human blood monocytes cultured in the presence of hydrocortisone release a factor that stimulates the migration of PMNs in vitro. This factor was not produced b y comparable cultures of steroid-treated lymphoid cells.

VI. Prostaglandins and Cyclic Nucleotides Prostaglandins (PGs) are a family of chemically similar fatty acids of molecular weight 300 to 400, which were first described by von Euler (1935). They are produced from prostanoic acid by a membranebound enzyme complex known as prostaglandin synthetase (see review by Samuelsson and Paoletti, 1976). PGs are generally classified as local or cellular hormones because they are produced and liberated locally, have a very short range of biological activity and, with the exception of prostaglandin A , can be inactivated very rapidly (Ramwell and Shaw, 1970; Piper and Vane, 1971; Samuelsson et al., 1971). PGs appear to be important channels of intercellular communication; they have a remarkably broad range of biological activity both in normal cell and tissue function and in many pathological processes. PGs have been implicated both as mediators and modulators of numerous biological processes, including acute and chronic inflammation, normal and pathological immune reactions, pathological connective tissue alterations, and bone resorption. The capacity to produce PGs is considered almost ubiquitous (Hinman, 1972),although definitive data on this point are not available. PG production has been demonstrated in cultures of peripheral blood lymphocytes (Ferraris and DeRubertis, 1974), fibrosarcoma cells (Levine et al., 1972), fibroblasts (Hamprecht et al., 1973), PMNs (Higgs and Youlton, 1972; Movat et al., 1971; Velo et al., 1973), eosinophils (Hubscher, 1975), and macrophages (Myatt et al., 1975; Gordon et al.,

1976). Analysis of various exudates provides circumstantial evidence that leukocytes, including macrophages, may synthesize PGs. Velo et al. (1973) studied PG levels in peritoneal exudates induced in rats by injecting carrageenan. As the lesions evolved, the ratio of PGEz to PGF, rose markedly and, after 48 hours, began to fall. These changes

THE MACROPHAGE AS A SECRETORY CELL

133

correlated with changes in the appearance and morphology of the mononuclear cells. The close correlation between prostaglandin levels and the levels of acid hydrolases in carrageenan-induced inflammation led Anderson e t al. (1971) to suggest that lysosomal hydrolases may stimulate prostaglandin synthesis at sites of inflammation. For example, phospholipases may generate arachidonic acid, a precursor of PGs. Ohuchi et al. (1976) detected PGE and PGF,, in carrageenan granulomas, lesions in which macrophages are the predominant cell type. The joint fluid of rabbits with ovalbumin-induced arthritis contains large quantities of PGE along with high concentrations of lysosomal enzymes. The PG levels can be reduced by the administration of indomethacin (Blackman et al., 1974). Organ cultures of rheumatoid synovia produce PG, and this production is inhibited by indomethacin (Robinson et al., 1975) and by corticosteroids (Kantrowitz et al., 1975). Cultures of cells derived from human rheumatoid synovium by treatment with enzymes have been studied by Dayer et al. (1976). These cultures contain fibroblasts and mononuclear cells, 10-25% of which exhibit properties of macrophages; they have Fc receptors, make lysozyme, and phagocytize avidly. Cultures containing these cells produce large quantities of PGs and collagenase. Within about 7 days of culture, the macrophage-like cell population decreases in size and there is an accompanying decrease in phagocytosis and PG production, although collagenase production continues. PG production, but not collagenase secretion, was inhibited by the addition of indomethacin to the cultures. Direct evidence has been provided that macrophages harvested from used intrauterine devices and macrophages from guinea pig peritoneum exposed to antigen and maintained in vitro synthesize prostaglandins (Myatt et al., 1975; Gordon et al., 1976). Macrophages appear to have the capacity to synthesize and secrete cyclic nucleotides, and this activity may be related to PG levels. Casein-induced rat peritoneal macrophages maintained in culture and exposed to PG produce and release large amounts of CAMP (Gemsa et al., 1975). The levels of CAMPbegin to rise intracellularly immediately following PG exposure and reach maximal levels within about 5 minutes; levels in the culture medium rise in a linear manner for at least 1 hour, during which the extracellular levels surpass the intracellular levels. PGE,, PGE,, PGAI, and PGA, are effective, in order of decreasing potency, while PGF, is ineffective. Exposing adherent macrophages cultured in serum-containing medium to lymphokines causes an increase in adenylate cyclase activity. This increase may be

134

ROY C. PAGE ET AL.

mediated by PGs, since it is inhibited by indomethacin (RemoldO’Donald and Remold, 1974). In addition, indomethacin alone increases adenylate cyclase activity when added to macrophages, and this drug largely obliterates the stimulation caused by exposure to lymphokines. Ruptured guinea pig macrophage preparations bind PGEl and PGE2,as well as catecholamines, and subsequently exhibit enhanced adenylate cyclase activity (Remold-O’Donald, 1974). The catecholamine site can be blocked by propranolol, but this inhibitor does not affect the PG-binding sites. Preincubation of macrophages in PG prevents subsequent stimulation of the preparation by PG. Furthermore, incubation of intact cells in the presence of PG prevents the elevation in adenylate cyclase activity usually induced by exposure to serum-containing medium (Remold-O’Donald, 1974). Thus the same hormones that stimulate adenylate cyclase activity under some conditions and at some concentrations may cause the reverse effects at other concentrations and under other conditions. The ability of macrophages to synthesize and release PGs and cyclic nucleotides may permit these cells to interact and to regulate the activities of other cells. It is conceivable that some of the roles of macrophages in acute and chronic inflammation, in wound healing, and in the immune response are a result of this capacity.

VII. Cytotoxic Substances Macrophages are prominent infiltrating cells in tumors of many types, and a major role for macrophages in tumor cell rejection phenomena has been postulated (Gorer, 1956). Although this hypothesis has been under intensive investigation for over 20 years, their exact role in homograft rejection and in tumor cell killing remains unresolved. Several facts have been established. Peritoneal macrophages from mice previously immunized with certain tumor cells have the capacity to kill the tumor cells specifically in vitro (Bennett et ul., 1964; Granger and Weiser, 1964,1966; Tsoi and Weiser, 1968; Rabinovitch, 1970; Evans and Alexander, 1970; Lohmann-Matthes et ul., 1972; Chambers and Weiser, 1972; Piper and McIvor, 1975). Phagocytosis does not appear to be involved. Allogeneic malignant cells and normal syngeneic cells are not killed by the activated cells, and normal nonactivated macrophages do not exhibit cytotoxicity. Administration of a nonspecific stimulant such as endotoxin or BCG protects experimental animals, at least in part, against malignancy induced by the injection of tumor cells (Bober et ul., 1976), and macrophages exhibiting nonspecific tumor cell killing in vitro can be ob-

THE MACROPHAGE AS A SECRETORY CELL

135

tained from animals previously injected with endotoxin, complete Freund’s adjuvant, or BCG, or animals infected b y various persistent intracellular parasites such as Toxoplasma, Besnoitia jellisoni, and L. monocytogenes (Alexander and Evans, 1971; Hibbs et al., 1972a,b). Furthermore, macrophages can be “armed” or “activated” for nonspecific killing of tumor cells by exposure in vitro to an array of substances including endotoxin, polynucleotides, pyran copolymer, and lymphokines (Evans and Alexander, 1970; Keller and Jones, 1971; Keller et al., 1971; Hibbs et al., 197213; Keller and Hess, 1972; Keller, 1972, 1973, 1974; Holtermann et al., 1972; Cleveland et al., 1974; Kaplan et al., 1974; Currie and Basham, 1975). Interestingly, inflammatory stimulants (e.g., thioglycollate) that in some cases induce enzyme secretion by macrophages do not produce cells with a cytotoxic capacity (Hibbs et al., 1972a,b). Macrophages become cytotoxic when exposed either to hyperimmune spleen cells in the presence of antigen, or to supernatants from immune lymphocytes incubated in the presence of sensitizing antigen (Evans and Alexander, 1972; Meltzer and Bartlett, 1972; Piessens et al., 1975). However, macrophages from athymic nude mice are as effective in cytotoxic reactions as cells from normal control animals (Keller, 1974; Meltzer, 1976), indicating that the presence of thymus-derived lymphocytes is not essential. Keller (1973, 1974) studied the kinetics of the cytostatic and cytolytic activity of nonspecifically activated macrophages on syngeneic tumor cells. Decreased synthesis of DNA is seen within 3-4 hours of exposure of the tumor cells to the activated macrophages. Subsequently, the cells shrink; by 10-12 hours their numbers have decreased, and by 24-48 hours they are all gone. If any survive, they are killed by the addition of more macrophages. The role of macrophage secretory activity in cytotoxicity reactions remains unclear. Most of the tumor cell killing systems involving macrophages appear to require cell-to-cell contact and membrane activity. However, several soluble factors have been detected. Lysosoma1 enzymes may be involved. In experiments done by Hibbs and his co-workers (1972a,b; Hibbs, 1974), endotoxin-activated mouse macrophages distinguished between untransfonned and transformed mouse fibroblasts, presumably because of the latter’s abnormal growth and antigeniccomposition, and selectively killed the transformed cells. Complement and serum factors did not appear to participate. Using a similar system, but with macrophages obtained from BCG-infected animals in which the secondary lysosomes had been labeled with dextran sulfate, a nontoxic, nondigestible substance which is detectable microscopically by its metachromasia after treatment with toluidine

136

ROY C . PAGE ET AL.

blue, they demonstrated a transfer of lysosomal contents from the activated macrophages to the cytoplasm of susceptible target cells. The target cells subsequently underwent lysis. Agents preventing exocytosis of macrophage lysosomes (e.g., hydrocortisone) and agents inhibiting lysosomal activity (e.g., trypan blue) interfered with the cytotoxicity. The macrophage appears to be the primary cell involved in the rejection of sarcoma ascites allografts (Tsoi and Weiser, 1968; Baker et d., 1962), and a soluble cytotoxic factor is involved. Alloimmune peritoneal macrophages can specifically destroy the sarcoma cells in vitro (Granger and Weiser, 1964, 1966). A comparable observation was made by Lohmann-Matthes et al. (1972), using mastocytoma-immune C57BL/6 macrophages. While cell contact was required, phagocytosis was not involved. Weiser et a2. (1969) demonstrated that alloimmune macrophages release a heat-labile growth-inhibiting factor (GIF) exhibiting immunological specificity when placed on cultures of the sensitizing cells. Subsequently, this factor was shown to be a cytotoxin (McIvor and Weiser, 1971),and its name was changed to specific macrophage cytotoxin (SMC). The specificity of SMC seems to be dependent on the transplantation antigens of the immunizing allograft tissue. This factor, which has a molecular weight of about 100,000, is produced during the first 2 hours of interaction between alloimmune macrophages and target cells (Piper and McIvor, 1975). SMC has also been tested for activity against Con-A-stimulated spleen cells from A/Jax and C57BW6 mice. In cultures from A/Jax, the animal from which the tumor originated, this factor stimulated incorporation of precursor into RNA but markedly inhibited incorporation of thymidine into DNA; it exerted no effect on cultures from C57BL/6 mice (Bell and McIvor, 1976). Melsom and Seljelid (1973) showed that mouse peritoneal macrophages activated only by overnight incubation in vitro on glass or plastic surfaces cause lysis of both allogeneic and syngeneic erythrocytes. The reaction was enhanced by the addition of Con A to the cultures, although prior immunization of the animals did not affect activity. Phagocytosis was not involved in the reaction, but cell contact appeared to be necessary. In subsequent experiments (Melsom et al., 1974), a macrophage cytotoxic factor (MCF) was detected in cultures prepared in serum-free medium to which 2-mercaptoethanol had been added. This factor is extremely labile and is not detected under ordinary cell culture conditions. It has a low molecular weight and is heat-stable at 60°C for 30 minutes. MCF is not present in freshly seeded cultures. Rifamycin, an antibiotic that irreversibly blocks RNA

THE MACROPHAGE AS A SECRETORY CELL

137

polymerase in bacterial but not mammalian cells, blocks the cytotoxic effect of MCF on erythrocytes (Melsom and Seljelid, 1974). A cytotoxic factor was also detected by Currie and Basham (1975) in cultures of endotoxin-activated macrophages. This factor exhibited no immunological specificity comparable to that described by Granger and Weiser (1964, 1966); it was cytotoxic only to malignant cells and did not affect normal cells. Cells not exposed to endotoxin exhibited cytotoxic activity, but this was greatly enhanced by 25 pglml of endotoxin. More activity was found in serum-free than in serum-containing cultures, indicating that serum may contain an inhibitor. I n contrast to the factor described by Melsom et aZ. (1974), this factor lysed neither syngeneic nor allogeneic erythrocytes.

VIII. Hydrolytic Enzymes The synthesis and secretion of various hydrolytic enzymes by macrophages participating in acute and chronic inflammatory responses are probably essential factors in the chronicity of, and tissue damage produced by, such lesions. The characteristics of substances inducing enzyme secretion, the identity and nature of the substances released, and the conditions under which the cells secrete have been investigated intensively in recent years (see reviews by Gordon and Cohn, 1973; Allison and Davies, 1974; Page et aZ., 1974b; Davies and Allison, 1976a,b). Stimulation of macrophages maintained in vitro by any of a large array of substances may lead to substantial enhancement of the production of various enzymes. Some of these enzymes, such as lactate dehydrogenase which is a cytoplasmic enzyme participating in energy production, are retained within the cell, while others, such as acid hydrolases, neutral proteinases, and lysozyme, are released from otherwise viable cells. In general, substances provoking chronic inflammation induce a selective release of hydrolytic enzymes, while several inert or readily digestible substances do not (Axline and Cohn, 1970; Davies et al., 1974a). Some of the same stimuli that induce the secretion of acid hydrolases also cause the release of neutral proteinases (Reich, 1975; Pantalone and Page, 1977), although these can also be released by some stimuli considered noninflammatory (Gordon et al., 197413; Werb and Gordon, 1975a,b). The timing and extent of acid hydrolase and neutral proteinase release differ markedly. Acid hydrolase is released in large amounts shortly after contact with an appropriate stimulus, while the release of neutral proteinase is usually delayed for at least 24 hours but, once initiated, continues over a very long time.

138

ROY C. PAGE ET AL.

Lysozyme appears to be a constitutive product of macrophages, being secreted continuously and in approximately similar amounts by cells maintained under widely varying culture conditions (Gordon et al., 1974a).

A. ACID HYDROLASES The synthetic and secretory activity of macrophages occurs in response to the uptake of specific substances by pinocytosis and phagocytosis or to the binding of a molecule such as immunoglobulin or complement to the cell surface. Macrophages cultured in media containing high concentrations of heat-inactivated newborn calf serum show high rates of pinocytosis and exhibit striking increases in the levels of various cellular constituents, especially lysosomal enzymes (Cohn, 1966), although under such conditions these enzymes are not released. The stimulation of pinocytotic activity by newborn calf serum is due to a macroglobulin in the serum, which is active against certain surface components of murine macrophages (Cohn and Parks, 1967). Macrophages also carry out a form of pinocytosis in which the vesicles formed cannot be seen in the light microscope. This process, referred to as micropinocytosis, differs from macropinocytosis in that it cannot be inhibited by cytochalasin B (Allison et al., 1971; Wills et al., 1972) and inhibitors of phosphorylation (Allison and Davies, 1975). Although the association remains to be definitively proved, pinocytotic activity appears to trigger the elevation of cellular lysosomal enzyme levels. Elevation of acid hydrolase activity and extracellular release of the enzymes may result from the phagocytosis of some types of particulate nondigestible substances. Phagocytosis of inert carbon particles and latex beads does not appear to cause an elevation in cellular enzyme levels (Axline and Cohn, 1970; Davies et d., 1974a), while other nondigestible materials such as sucrose (Cohn and Ehrenreich, 1969),and digestible particles such as sheep red blood cells and aggregated yglobulin (Weissmann et al., 1971), evoke marked increases in the cellular levels of lysosomal and other cell enzymes but not in their extracellular release. However, phagocytosis of zymosan by macrophages causes the release of acid hydrolases without a change in cell viability (Weissmann et aZ., 1971). It is notable that zymosan contains glucans with marked inflammatory activity (McCandless, 1965), although the relationship of these substances to enzyme release remains unclear. Because macrophages are consistently present in chronic inflammatory lesions of both immunological and nonimmunological origin and probably play a major role in tissue destruction, several investigators

THE MACROPHAGE AS A SECRETORY CELL

139

have studied the role of substances known to elicit chronic inflammatory lesions in vivo for their effectiveness in causing acid hydrolase release from macrophages maintained in vitro. Acid hydrolases are released rapidly, frequently within 4-6 hours and in large quantities, sometimes up to 80% of the total enzyme within the cells, after macrophages are exposed to any of a large number of inflammation-inducing substances. The release is not a general response to phagocytosis of particulate materials, since only potentially pathogenic substances are effective; biologically unreactive substances do not induce release. The secreted enzymes derive almost certainly from a preformed store contained in the lysosomes. However, enzyme production de novo may also increase, as suggested by the elevated levels of acid hydrolases in cultures maintained for 72 hours in the presence of stimuli (Page et al., 1973; Davies et al., 1974a). Enzyme release continues throughout the experimental period, and the responding cells remain viable, as determined by several different criteria including dye exclusion, retention of cytoplasmic enzymes and a specific esterase, and continued cell adherence. The enzymes released appear to include all the acid hydrolases, such as cathepsins, glycosidases, acid phosphatase, aryl sulfatase, and others. These enzymes have the potential to participate in the initiation and perpetuation of inflammation through their capacity to degrade connective tissue substances, to activate the zymogens of several humoral mediators of inflammation, including those of the complement, coagulation, and kinin systems, and to modify cell function by limited hydrolysis of certain plasma membrane constituents.

1. Bacterial Substances

The association of group-A streptococci with several diseases, including rheumatic fever, is well established (Wannamaker and Matsen, 1972; Ginsburg, 1972). After single injections at various sites, a type-specific C-mucopolysaccharide-peptidoglycan complex from group-A streptococcal cell walls (PPG) causes chronic inflammatory lesions in which macrophages are the predominant cells (Schwab et al., 1959; Ginsburg, 1972; Page et al., 1974a). The cell wall material is resistant to the action of lysosomal enzymes (Ayoub and McCarty, 1968) and persists for many months at sites of chronic inflammation. These granulomatous lesions do not appear to have an immunopathological basis but are due to the direct effects of the cell wall material on macrophages, as shown in studies with cultures of these cells exposed to PPG (Davies et al., 1974a; Page et al., 1974a). After exposure to PPG, macrophages undergo rapid and marked

ROY C. PAGE ET AL.

140

morphological changes which persist for many days. These include a two- to fourfold increase in cell size, an increase in the number of lysosomes, as shown by vital staining with acridine orange, and an increase in ruffled membrane activity. Exposure to doses of cell wall material in the range of 1-15 pg/ml for 72 hours results in marked increases in cellular levels of nonlysosomal enzymes, such as lactate dehydrogenase and leucine-2-naphthylamidase.There are also smaller increases in cellular levels of lysosomal enzymes such as P-glucuronidase. At these relatively low concentrations of PPG, there is no detectable release of enzymes from the cells into the culture medium. However, doses ranging from 15 to 50 pg/ml induce the release of lysosoma1 enzymes from viable macrophages in a dose- and time-dependent manner (Fig. 1).The release proceeds rapidly, and much of the enzyme is redistributed within 4-6 hours of exposure of the cells to the cell wall material (Davies et al., 1974a). The exposure of mononuclear phagocytes maintained in culture to endotoxin results in a variety of responses which are not completely understood. Cells exposed in vitro to concentrations near 1 pg/ml of medium exhibit enhanced levels of pinocytosis and phagocytosis (Bennett and Cohn, 1966),and mouse macrophages treated with endotoxin from Salmonella abortus equi exhibit increased cell size and

-

0 PPG Conc. pg/rnl

FIG. 1. The dose-dependent selective release of P-glucuronidase from mouse pentoneal macrophages exposed to increasing doses of PPG for 24 hours. Solid circles, total activity in the culture; triangles, cellular level of P-glucuronidase; open circles, amount of enzyme released into culture medium. (Reproduced with permission from Davies et d., 1974a.)

THE MACROPHAGE AS A SECRETORY CELL

141

granularity with the elevation of cellular acid phosphatase levels (Cohn et al., 1966). Furthermore, endotoxin-treated cells are more effective in killing bacteria and various tumor cells than are normal untreated control cells (see Section VII). The morphological and biochemical effects of the exposure of mouse peritoneal mononuclear phagocytes maintained in culture to endotoxin differ from those seen in cultures containing inert particles or streptococcal cell walls (Allison et al., 1973). Exposed cells exhibit enhanced spreading on the surface of the culture vessel and increased granularity soon after exposure. However, these are short-term effects which disappear during the first 24 hours of incubation. In cultures of cells exposed to 10 pg of endotoxin per milliliter of medium for 72 hours, the levels of lactate dehydrogenase and leucine-2-naphthylamidase increase about twofold over control values. Dental plaque is a mixed bacterial growth found on the sheltered areas of teeth. It contains numerous microorganisms of various types. There is substantial evidence that plaque is the primary etiological agent involved in the pathogenesis of chronic periodontitis. Plaque sterilized by irradiation causes changes very similar to those seen when streptococcal cell wall material is added to cultures of macrophages (Page et al., 1973). Plaque in doses of 5-50 pg/ml causes a rapid and massive redistribution of acid hydrolases from the cells into the culture medium. Although no net increase in acid hydrolase levels was detected in cultures exposed to plaque for 24 hours, there were significant increases in the levels of P-glucuronidase in cultures exposed to concentrations of plaque greater than 10 &ml for 72 hours. This indicates that the continuous depletion of intracellular lysosomal enzymes may serve as a stimulus for the replacement of enzymes released into the culture medium. These changes also occur in the absence of any detectable cell death. The plaque constituents responsible for this effect are not known, although subsequent studies (Page et al., 1974b) have shown that sterile homogenates of Actinomyces uiscosus have effects similar to those of plaque. Actinomyces viscosus is a gram-positive anaerobic filamentous microorganism which has the capacity to induce severe chronic inflammatory periodontal disease in monoinfected germ-free rodents. It is also found in high concentrations in human periodontal pockets. 2. Toxic Substances of Nonbacterial Origin Carrageenan, a mixture of sulfated D-galaCtOSe and 3,6-dehydro-~galactose, induces chronic inflammation in experimental animals (Robertson and Schwartz, 1953; Benitz and Hall, 1959; Davies et d.,

142

ROY C. PAGE ET AL.

1975). Preparations of carrageenan that cause chronic inflammation also induce the selective release of acid hydrolases from macrophages in culture in a dose- and time-dependent manner (Allison and Davies, 1975). In contrast to the effects of bacterial substances, enzyme release is delayed approximately 10 hours. This delay may be accounted for by the ingestion of carrageenan, which is soluble, by pinocytosis, a less rapid process than phagocytosis. There is a strong positive correlation between the ability to induce chronic inflammation in uivo and the stimulation of selective hydrolase release in uitro. Structure-activity relationships show that h-carrageenan, a linear polymer of Dgalactose with a-1,3 linkages and sulfated at C-4, has the greatest activity. K-Carrageenan, containing alternating units of 4-sulfated D-galactose and 3,6-anhydro-~-galactosewith a-1,3 linkages and @1,4 linkages possesses lower inflammatory activity. The pneumoconioses are a family of chronic inflammatory diseases caused by the inhalation of toxic particles (Spencer, 1968). In these diseases, which are characterized by fibrosis and a predominance of mononuclear cells, macrophage secretory activity appears to play an important role. Toxic particles entering the pulmonary alveoli are taken up by macrophages (Allison, 1968). Failure of the macrophages to digest or remove the particles from the lungs results in the inflammatory pathogenic process. Toxic particles such as silica and asbestos have cytotoxic effects on macrophages maintained in uitro (Allison et al., 1966; Allison, 1971).Asbestos is less cytotoxic than silica, but as little as 2 pg/ml of chrysotile asbestos fibers induces the selective release of lysosomal enzymes from cultured macrophages (Davies et al., 1974b). This release is dose- and time-dependent (Fig. 2); up to 80% of the enzyme can be released within 12 hours under favorable conditions. Enzyme release is accompanied by marked increases in cellular levels of cytoplasmic lactate dehydrogenase, although this enzyme is not released into the culture medium, indicating that the cells remain viable. Amphibole crocidolite asbestos is less potent than chrysotile in inducing lysosomal enzyme release, while two other amphiboles, amosite and anthophyllite, are intermediate between chrysotile and crocidolite in activity.

3. Products of Lymphoid Cells Many chronic inflammatory lesions are the result of immunological responses and are dominated by lymphocytes and macrophages. Thus substances derived from lymphoid cells have been evaluated for their effects on macrophages. The stimulation of cultured lymphocytes by mitogen or of previously sensitized lymphocytes by antigen results in

143

THE MACROPHAGE AS A SECRETORY CELL

I

I

I

I

I

1

4

8

12

16

20

24

Time In Hours

FIG. 2. The tiine-dependent release of P-galactosidase from mouse peritoneal macrophages exposed to chrysotile asbestos (50 &ml). Solid circles, percentage of total enzyme activity in medium of asbestos-treated cells; open circles, enzyme in medium of control culhires not exposed to asbestos. (Reproduced with permission from Davies et NI., 1974b.)

the release of lymphokines (Bloom and Bennett, 1966; Bennett and Bloom, 1968; Granger and Kolb, 1968; Pick et al., 1969; Pick and Turk, 1972). Some of these substances induce remarkable morphological and functional changes in macrophages, such as enlargement and decreased mobility (David, 1966), greater adherence to culture vessels (Mooney and Waksman, 1970), increased metabolic activity (Nathan et al., 1971; Nath et al., 1973), and enhanced microbicidal activity (Fowles et d.,1973; Krahenbuhl et aZ., 1973; Magliulo et al., 1973). Pantalone and Page (1975) studied lysosome hydrolase secretion b y lymphokine-stimulated macrophages. The cells contained more lactate dehydrogenase and leucine-2-naphthylamidase, as well as acid hydrolases, than the cells of control cultures. While the first two enzymes were retained within the cell cytoplasm, acid hydrolases

144

ROY C. PAGE ET AL.

were secreted in large quantities by otherwise intact and viable cells. Maximal stimulation of enzyme release, amounting to between 70 and 80% of the total in the culture for P-glucuronidase and N-acetyl-P-Dglucosaminidase, occurred when the supernatant obtained from approximately 0.5 x lo6lymphoid cells was added to 2 to 4 x lo6macrophages (Fig. 3). The synthetic and secretory activities induced in macrophages by lyniphokines differ remarkably in time course, magnitude, and pattern from those exhibited by cells exposed to various other substances. For example, large amounts of acid phosphatase activity are released within 15-30 minutes by cells exposed to streptococcal cell walls (Davies et al., 1974a), while there is a lag phase of 36 hours in the release of this enzyme from lymphokine-treated cells (Pantalone and Page, 1975). 0-Glucuronidase release induced by the cell walls is essentially complete within 4 hours, while the same enzyme is released from lymphokine-activated cells throughout a culture period of 72 hours in an almost linear manner. Although definitive experiments along these lines were not made, data relating to time course, dose response, and quantity indicate that secretory activity in lymphokineactivated cells may be coupled with synthetic activity. This does not seem to be the case with cells activated by streptococcal cell wall

* FIG. 3. The selective release of P-glucuronidase from mouse peritoneal macrophages exposed for 48 hours to supernatants from human peripheral blood lymphocytes cultured in the presence or absence of PHA for 72 hours. Clear bars, cells. Stippled bars, culture medium. (Reproduced with permission from Pantalone and Page, 1975.)

T H E MACROPHAGE AS A SECRETORY CELL

145

preparations or by certain other bacterial substances (see review by Page et al., 197413). Whether or not the lymphokine responsible for inducing the synthetic and secretory activities observed is the same as macrophage migration inhibition factor (MIF) or macrophage activating factor (MAF) (David, 1966; Nathan et al., 1971) remains unknown. However, electrophoretic migration of the active substance in sodium dodecyl sulfate (SDS)-acrylamide gels indicates a molecular weight of 40,000 to 50,000, a size range comparable to that of MIF (Remold et al., 1970). Tissue macrophages and human peripheral blood monocytes possess F c receptors with the capacity to bind immune complexes containing either IgG or IgM (see review by Shevach et al., 1973). In human cells, the site displays specificity for human IgG subclasses IgGl and IgG3 (Huber and Wagner, 1974). Immune complexes of horseradish peroxidase and specific antibody are taken into mouse peritoneal macrophage secondary lysosomes and degraded (Steinman and Cohn, 1972). Macrophages exposed to complexes formed at antigen antibody equivalence from bovine serum albumin and rabbit antibody exhibit selective release of acid hydrolase activity from the cells into the culture medium (Cardella et al., 1974). The release, which is caused neither by antigen nor antiserum alone, increases with exposure time over a period of at least 24 hours and is dose-dependent. The combination of a substance that alone is effective in inducing lysosomal hydrolase release from macrophages, such as streptococcal cell wall material, with specific antibody may enhance its effectiveness in inducing enzyme release (Page et al., 1974a). Immune complexes are deposited at sites of chronic inflammation in several common diseases, including rheumatoid arthritis and certain types of glomerulonephritis, and the intradermal injection of complexes into rats results in a lesion characterized by the presence of many macrophages containing residual immune complexes (Spector and Heesom, 1969). Thus the release of hydrolytic enzymes by macrophages phagocytizing complexes may play a central role in the pathogenesis of these diseases.

4. Complement-Derived Substances In looking for a common factor in the variety of agents that induce chronic inflammation in vivo and hydrolase secretion b y macrophages i n vitro, attention has been drawn to the fact that all of them activate complement b y the alternate pathway. This is true of carrageenans and dextran sulfates (Burger et al., 1975), zymosan, moldy hay dust

146

ROY C. PAGE ET AL.

containing Micropolysporum fueni (Edwards et ul., 1976), and streptococcal cell walls(D.Bitter-Suermann,personal communication, 1976). An early event in such activation is cleavage of the complement component C3 to a smaller fragment C3a and a larger fragment C3b. The highly purified C3a component, when incubated with mouse peritoneal macrophages in culture, causes the release of both lysosoma1 hydrolases and lactate dehydrogenase, indicating cell death (Schorlemmer et al., 1976). In contrast, when highly purified C3b is incubated with the macrophages, a selective dose- and time-dependent release of several glycosidases, but not of lactate dehydrogenase, is observed. Similar results have been obtained with guinea pig peritoneal macrophages, using highly purified guinea pig C3a and C3b (Schorlemmer and Allison, 1976). Supernatants from macrophages stimulated by asbestos, but not those from unstimulated macrophages, cleave C3. Thus generation of C3b appears to be a common and important mechanism for switching on hydrolase secretion by macrophages. The secreted enzymes can themselves cleave C3, thereby generating further C3b, so that an amplification system is set in motion. This may well play an important role in chronic inflammation. The various substances that induce chronic inflammation have two common features: They are not readily biodegradable, so they persist in macrophages, and they have the polymeric acidic structural requirements for activation of complement by the alternative pathway.

B. NEUTRALPROTEINASES In addition to acid hydrolases, stimulated macrophages may synthesize and secrete several other enzymes which operate at neutral pH and play an important role in both the destructive and reparative phases of the chronic inflammatory process. Although cell stimulation by some substances, such as lymphokines, may cause the release of both acid hydrolases and neutral proteinases, other stimulators may activate only one of these processes. Exactly which stimulants activate which secretory processes and the time course of this activation remain to be elucidated.

1. Collagenase Collagenase plays a major role in both normal tissue turnover and remodeling and in the destruction of collagen, which occurs in most acute and chronic inflammatory reactions. PMNs contain considerable amounts of collagenase (Lazarus, 1973), and these cells are a likely source of this enzyme in acute inflammatory processes. However, collagen breakdown is most prominent in chronic inflammation where

T H E MACROPHAGE AS A SECRETORY CELL

147

relatively few PMNs are present and the lesions are dominated by macrophages. Alveolar macrophages (Senior et al., 1972; Robertson et al., 1973) and Kupffer cells (Fujiwara et al., 1973) contain collagenase, and peritoneal macrophages maintained in culture in the presence of endotoxin or lymphokine synthesize and secrete substantial quantities of this enzyme (Wahl et al., 1974, 1975). Stimulation of the cells appears to induce production and secretion of collagenase; activity is not found in freshly isolated cells or in 5-day cultures of nonstimulated cells following detergent lysis or fi,eeze-thawing. Production begins only after a 24-hour lag period following stimulation, and the appearance of activity is prevented by exposure of the cells to cycloheximide. The capacity to stimulate collagenase production and secretion by macrophages resides in the lipid A fraction; only very low levels of activity reside in the glycolipid and lipid-free polysaccharide fractions. Peritoneal macrophages obtained from mice to which intraperitoneal thioglycollate broth has been administered 4 days previously secrete considerable amounts of collagenase over a period of at least 7 days (Werb and Gordon, 1975a). However, macrophages from unstimulated mice secrete only barely detectable amounts of this enzyme. These cells, however, secrete some collagenase after the ingestion of latex particles or dextran sulfate, although in considerably smaller amounts than are obtained from cells taken from thioglycollate-stimulated animals. The failure of unstimulated macrophages to secrete this enzyme is not due to the simultaneous production of inhibitors of collagenase. Horwitz and Crystal (1976) demonstrated the secretion of collagenase by pulmonary alveolar macrophages. The rate of secretion is greatly increased if donor animals have been stimulated previously with BCG. T h e secretion of collagenase b y macrophages from unstimulated animals can be increased severalfold by phagocytic stimuli such as silica, latex, and Mycobacterium butyricum. Most of the collagenase is secreted in a latent form, since treatment of supernatants of cultured cells with trypsin increases the yield of collagenase by u p to nine times the amount found in untreated supernatants. T h e enzyme was shown to possess the properties associated with mammalian collagenase and to cleave collagen types I, 11, and 111. Birkedal-Hansen et al. (1976) showed that collagenase obtained from rabbit pulmonary alveolar macrophages subjected to gel chromatography has several molecular-weight values ranging from 45,000 to 165,000. Chromatography with salt concentrations exceeding 1.0 M yielded a species of collagenase with a molecular weight of 45,000,

148

ROY C. PAGE ET AL.

suggesting that the higher-molecular-weight species correspond to polymeric forms of this enzyme. The mechanism b y which lymphokines stimulate macrophages to produce and secrete collagenase remains unknown. It is clear that the continuous presence of lymphokines is not essential. I n the experiments of Pantalone and Page (1977), in which macrophages were exposed to lymphokine for only 24 hours following which lymphokinefree medium without serum was used, elevated protein synthesis and collagenase production continued for 4 days. Whether or not the stimulating substance becomes bound to the macrophage and whether or not the observed macrophage stimulation is a reversible process remains unresolved. In these experiments, the lymphokine preparation did not induce cell mitosis. Macrophage-derived collagenase resembles other mammalian collagenases in most respects. Its activity is inhibited by serum, EDTA, and a,-macroglobulin, and it cleaves the native collagen molecule into three-quarter and one-quarter portions. Whether the peritoneal macrophage-derived enzyme is produced in a precursor or an active form is not clear, although the latter seems likely, since culture supernatants do not require prior trypsin activation for the demonstration of activity. The amount of collagenase activity produced in vitro by macrophages is considerably less than that obtained from synovial cells maintained in culture (Werb and Reynolds, 1974). Macrophages from thioglycollate-stimulated mice produced 0.11 unit of collagenase per day per milligram of cell protein, while rabbit synovial cells produced 1.8 units of enzyme per day per milligram of cell protein. Gordon and Werb (1976) have reported that 5 x lop6M colchicine stimulates the secretion of collagenase by thioglycollate-induced mouse peritoneal macrophages by more than 100%.

2. Elustuse Janoff and Basch (1971) described the presence of elastase in lysates of alveolar macrophages, and Werb and Gordon (1975b) studied the secretion of this enzyme by thioglycollate-induced macrophages. Secretion is time-dependent, continuing for at least 12 days, and sensitive to inhibitors of protein synthesis. As with collagenase, macrophages from unstimulated mice secrete very little elastase; such cells, however, secrete some enzyme after phagocytosis of latex beads. The enzyme secreted by macrophages differs significantly from enzymes with elastinolytic activity purified from pancreas and PMNs. Macrophage elastase is more restricted in its substrate specificity, showing

T H E MACROPHAGE AS A SECRETORY CELL

149

no activity toward synthetic substrates such as N-acetyl-L-alanyl-Lalanyl-L-alanine-p-nitroanilide or benzyloxycarbonyl-~-alanine-2naphthol ester. It is also resistant to inhibitors of other elastases such as active site-directed alanyl chloromethyl ketones and turkey ovomucoid. This enzyme shows a marked affinity for elastin complexed to SDS; its cationic nature makes such a complex a suitable substrate. The secretion of elastase (Gordon and Werb, 1976) by macrophages can be stimulated by both colchicine and cytochalasin B. Fivefold increases in elastase secretion are observed in macrophages exposed to 4 x M colchicine for 24 hours, while two- to threefold increases are observed in cells exposed to cytochalasin B (4 pg/ml).

3. Plasminogen Activator The enzymic cleavage of plasminogen to plasmin is of considerable relevance in inflammatory processes. In addition to degrading fibrin, the final product of blood coagulation, plasmin also has the capacity to split C3. The split products have the capacity to activate B lymphocytes and induce hydrolytic enzyme release by macrophages. In addition, plasmin activates Hageman factor, the initial component of the coagulation cascade, which in turn can convert kallikreinogen to kallikrein, leading to the generation of kinins, extremely potent mediators of acute inflammation (Cochrane et aZ., 1974). Interest in plasminogen activator has recently been stimulated by observations that it is released by mouse embryo fibroblasts following transformation by oncogenic viruses (Unkeless et al., 1973), whereas untransformed cells show minimal release of this enzyme. Recent studies by Unkeless et al. (1974) and Gordon et al. (1974b) have shown that macrophages stimulated in certain ways secrete large quantities of plasminogen activator. While unstimulated peritoneal macrophages from mice do not appear to synthesize this enzyme, cells induced with thioglycollate broth and subsequently maintained in culture secrete large amounts. Peritoneal cells obtained b y lavage and stimulated in vitro with endotoxin, calf serum, BCG, or mineral oil produce only small amounts. However, if cells stimulated in vivo by these agents are exposed to phagocytic stimuli such as latex beads in vitro, their secretion of plasminogen activator is greatly enhanced. Nondigestible substrates appear to be more efficient in stimulating this kind of activity than digestible substrates such as M . lysodeikticus, heat-aggregated y-globulin, and immune complexes. The production and secretion of plasminogen activator by stimulated macrophages continue for periods of up to 9 days following stimulation.

150

ROY C. PAGE ET AL.

4. Lysozyme Lysozyme is a low-molecular-weight, cationic protein which hydrolyzes N-acetylmuramic-P- 1,4-N-acetylglucosamine linkages in bacterial cell walls. High concentrations of lysozyme are found in both PMNs and in rabbit alveolar macrophages. In PMNs the enzyme is largely particulate, being found in both azurophilic and specific granules. The early studies of Cohn and Wiener (1963) showed that BCGinduced rabbit alveolar macrophages contain lysozyme and that a portion of the activity is released from the cells during phagocytosis. Large quantities of this enzyme are found both in the serum and in the urine of humans and experimental animals suffering from monocytic leukemia. Mouse peritoneal macrophages and human peripheral blood monocytes synthesize and secrete substantial amounts of lysozyme when maintained in vitro (Gordon et al., 1974a).This enzyme is a major secretory product of both macrophages and monocytes, since its daily production represents 0.5-2.5% of the total cell protein. Secretion begins after a lag period and continues for as long as 17 days. Macrophages secrete the equivalent of their cellular pool of lysozyme in a period of 5-8 hours; secretion can be almost completely abolished by an inhibitor of protein synthesis such as cycloheximide. This enzyme is secreted under a wide variety of conditions, and secretion is not affected to any great extent by the presence or absence of serum, by phagocytosis, or by the presence of a substrate such as M . lysodeikticus. Lysozyme secretion seems to be restricted to macrophages, since other cells such as epithelioid cell lines, fibroblasts, and lymphocytes do not produce it. PMNs contain intracellular stores of this enzyme in their lysosomes, although there is no evidence of secretion of new enzyme by these cells. At present there is little to suggest that lysozyme secreted by macrophages influences the course of chronic inflammatory reactions.

c.

CONSEQUENCES O F ENZYMERELEASE BY

MACROPHAGES

The capacity of lysosoinal enzymes to degrade natural substrates has been amply documented (Barrett, 1969; Aronson and de Duve, 1968; Coffey and de Duve, 1968; Fowler and de Duve, 1969). Studies with specific antisera to cathepsin D have demonstrated the activity of secreted enzyme (Dingle et aZ,, 1971) in organ culture and intralysosoma1 cathepsin D (Dingle et al., 1973) in cultured macrophages. The activity of collagenase, elastase, and plasminogen activator on

THE MACROPHAGE AS A SECRETORY CELL

151

their respective natural substrates has been extensively discussed elsewhere (Lazarus, 1973; Janoff et al., 1975; Reich, 1975), and the reader is referred to these reviews for further information. Elastase is active against a wide range of connective tissue substrates besides elastin, including cartilage proteoglycan (Janoff, 1972). Similarly, plasmin generated by plasminogen activator is effective against a variety of natural substrates. Activation of the clotting, kinin, complement, and fibrinolytic systems are all features of the inflammatory process. In each case, proteinases play an important part in the activation process (Greenbaum, 1975; Miiller-Eberhard, 1975; Reich, 1975). Consequently, proteinases secreted by phagocytic cells at sites of inflammation may initiate the activity of one or more of the cascade systems mentioned above. Chang et al. (1972) showed that an acid proteinase from PMNs or macrophages, probably cathepsin D, can generate polypeptides with kinin activity from plasma leukokininogen. The cleavage of C3 and C5 by plasmin or lysosomal enzymes (Snyderman et al., 1972; Goldstein and Weissman, 1974; Ward and Hill, 1970) results in the formation of cleavage products with chemotatic, anaphylotoxic, and lysosomal enzyme-mobilizing activity. It is also notable that lysosomal enzymes are able to degrade certain inflammatory mediators; for example, aryl sulfatase from eosinophils degrades SRS-A released by mast cells at sites of inflammation (Wasserman et al., 1975). ACKNOWLEDGMENT We are grateful to Dr. David Wood for his critical reading of the manuscript. REFERENCES Aalto, M., Potila, M., and Kulonen, E. (1976). E x p . Cell Res. 97, 193. Acton, J. D., and Myrvik, Q. N. (1966).J.Bucteriol 91, 2300. Adolphe, M., Fontagne, J., Pelletier, M., and Giroud, J . P. (1975).Noturc (London)253, 637. Alexander, P., and Evans, R. (1971). Nature (London),New Biol. 232,76. Allison, A. C. (1968). Sci. Basis Med. p. 18. Allison, A. C. (1971).Arch. Intern. Med. 128, 131. Allison, A. C. (1974). Transplant. Rev. 19,3. Allison, A. C., and Davies, P. (1974). In “Future Trends in Inflammation” (C. P. Velo, D. A. Willoughby, and J . P. Giroud, eds.), p. 449. Piccin, Padua and London. Allison, A. C., and Davies, P. (1975).In “Mononuclear Phagocytes in Immunity, Infection and Pathology” (R. van Furth, ed.), p. 487. Blackwell, Oxford. Allison, A. C., Harington, J. C., and Birbeck, M. (1966)./. E x p . Med. 124, 141. Allison, A. C., Davies, P., and De Petris, S. (1971).Nature (London),New Biol. 232,153.

152

ROY C. PAGE ET AL.

Allison, A. C., Davies, P., and Page, R. C. (1973).J.Infect. Dis. 128, Suppl. S212. Anderson, A. J,, Brocklehurst, W. E., and Willis, A. J. (1971).Pharmacol. Res. Commun. 3, 13. Aronson, N. N., Jr., and de Duve, C. (1968).J.B i d . Chem. 243,4564. Atkins, E., and Bodel, P. T. (1974). In “The Inflammatory Process” (B. W. Zweifach, L. Grant, and R. T. McCluskey, eds.), 2nd ed., Vol. 3, p. 467. Academic Press, New York. Atkins, E., Bodel, P., and Francis, L. (1967).J.E x p . Med. 126,357. Axline, S. G., and Cohn, Z. A. (1970).J.E x p . Med. 131, 1239. Ayoub, E. M., and McCarty, M. (1968).]. E x p . Med. 127,833. Baker, P., Weiser, R. S., Jutila, J., Evans, C. A., and Blandeau, R. J. (1962).Ann. N.Y. Acad. Sci. 101,46. Barrett, A. J. (1969). LrJsosomes Biol. Pathol. 2,245. Bast, R. C. Jr., Cleveland, R. P., Littman, B. H., Zbar, B., and Rapp, H. J. (1974).Cell. Immunol. 10,248. Bell, R. B., and McIvor, K. L. (1976). Cell. Immunol. 22, 19. Benitz, K. F., and Hall, L. M. (1959).Proc. SOC. E x p . Biol. Med. 102,442. Bennett, B., Old, L. J., and Boyse, E. A. (1964). Transplantation 2, 183. Bennett, B., and Bloom, B. R. (1968).Proc. Natl. Acad. Sci. U.S.A. 59,756. Bennett, I. L., Jr., and Beeson, P. B. (1953).J.E x p . Med. 98,493. Bennett, W. E., and Cohn, Z. A. (1966).J.E x p . Med. 123, 145. Bentley, C., Bitter-Suermann, D., Hadding, U., and Brade, V. (1976).Eur. J. Immunol. 6,393. Birkedal-Hansen, H., Cobb, C. M., Taylor, R. E., and Fullmer, H. M. (1976).Biochim. Biophys. Acta 438,273. Blackham, A., Farmer, J. B., Radziwonik, H., and Westwick, J. (1974). Br.J. Pharmacol. 51,35. Bloom, B. R., and Bennett, D. (1966).Science 153,80. Bober, L. A,, Kranepool, M. J. and Hollander, V. P. (1976).Cancer Res. 36,927. Bodel, P. M. (1974).J.E x p . Med. 140,954. Bodel, P. M. (1976).J. E x p . M i d . 143, 1015. Bodel, P. M. and Atkins, E. (1967).N . Engl. J. Med. 276, 1002. Burger, R., Hadding, U., Schorlemmer, H. U., Brade, V., and Bitter-Suermann, D. (1975).Immunology 29, 549. Calderon, J., and Unanue, E. R. (1975).Nature (London) 253,359. Calderon, J., Williams, R. T. and Unanue, E. R. (1974).Proc. Natl. Acad. Sci. U.S.A.71, 4273. Calderon, J.. Kiely, J. M., Lefko, J. L., and Unanue, E. R. ( 1 9 7 5 ) ~E.x p . Med. 142, 151. Cardella, C. J., Davies, P., and Allison, A. C. (1974).Nature (London) 247,46. Carrel, A,, and Ebeling, A. H. (1926).J.E x p . Med. 44,285. Chambers, V. E., and Weiser, R. S. (1972). Cancer Res. 32,413. Chang, J., Freer, R., Stella, R., and Greenbaum, L. M. (1972). Biochem. Pharmacol. 21, 3095. Chervenik, P. A. (1972).J.Lab. Clin. Med. 79, 1014. Chervenik, P. A., and LoBuglio, A. F. (1972). Science 178, 164. Cifone, M., Mocarelli, P., and Defendi, V. (1975). E x p . Cell Res. 96,96. Cleveland, R. P., Meltzer, M. S., and Zbar, B. (1974).J. Natl. Cancer Znst. 52, 1887. Cochrane, C. G., Revak, S . D., Wuepper, K. D., Johnston, A., Morrison, D. C., and Ulevitch, R. (1974).Adu. Biosci. 12,237. Coffey, J. W., and de Duve, C. (1968).J.B i d . Chem. 243,3255.

T H E MACROPHAGE AS A SECRETORY CELL

153

Cohn, Z. A. (1966).J. E x p . Med. 124,557. Cohn, Z. A. (1968).Adv. Zmmunol. 9, 163-214. Cohn, Z . A. (1970).In “Mononuclear Phagocytes” (R. van Furth, ed.), p. 121. S. A. Davis and Co., Philadelphia. Cohn, Z. A., and Benson, B. (1965).J. E x p . Med. 121, 153. Cohn, Z . A., and Ehrenreich, B. A. (1969).J.E x p . Med. 129,201. Cohn, Z. A., and Parks, E. (1967).J. E x p . Med. 125, 1091. Cohn, Z. A., and Wiener, E . (1963).J. E x p . Med. 118, 991. Cohn, Z. A., Hirsch, J. C., and Fedorko, M. (1966).J. Exp. Med. 123,747. Colten, H. R. (1972).J.Clin. Znoest. 51,725. Colten, H. R. (1976).Ado. Zmrnunol. 22,67. Colten, H. R., and Borsos, T. (1974).J. Zmmunol. 112, 1107. Currie, G. A., and Basham, C. (1975).J.E x p . Med. 142, 1600. David, J. R. (1966). Proc. Natl. Acud. Sci. U.S.A. 56, 72. Davies, P., and Allison, A. C. (1976a). In “Immunobiology of the Macrophage” (D. S. Nelson, ed.), p. 427. Academic Press, New York. Davies, P., and Allison, A. C. (1976b). Agents Actions 6,60. Davies, P., Page, R. C., and Allison, A. C. (1974a).J.Exp. Med. 139, 1262. Davies, P., Allison, A. C., Ackerman, J., Butterfield, A., and Williams, S. (197413).Nature (London)251,423. Davies, P., Allison, A. C., Dym, M., and Cardella, C. J. (1975). In “Proceedings of the WHO Symposium of Infectian and Immunology in the Rheumatic Diseases” (D. C. Dumonde, ed.), p. 365. Blackwell, Oxford. Dayer, J. M., Krane, S . M., Graham, R., Russell, G., and Robinson, D. R. (1976). Proc. Natl. Acad. Sci. U.S.A. 73,945. Diamanstein, T., and Ulmer, A. (1976).Zmmunology 30,741. Dinarello, C. A,, Wolff, S. M., Goldfinger, S. E., Dale, D. C., and Alling, D. W. (1974a). N . Engl. J. Med. 291,934. Dinarello, C. A., Goldin, N. P., and Wolff, S. M. (1974b).J. E x p . Med. 139, 1369. Dingle, J. T., Barrett, A. J., and Weston, P. D. (1971). Biochem. J. 123, 1. Dingle, J. T., Poole, A. R., Lazarus, G., and Barrett, A. J. (1973).J.E x p . Med. 137, 1124. Eaves, A. C., and Bruce, W. R. (1974). Cell Tissue Kinet, 7 , 19. Ebert, R. H., and Florey, H. W. (1939). Br. J. E x p . Pathol. 20,342. Edwards, J. H., Wagner, J. C., and Seal, R. M. E. (1976).Clin. Allergy 6, 155. Einstein, L. P., Alper, C. A., Bloch, K. J., Herrin, J. T., Rosen, F. S., David, J. R., and Colten, H. R. (1975). N . Engl. J. Med. 292, 1169. Einstein, L. P., Schneeberger, E. E., and Colten, H. R. (1976).J.E x p . Med. 143, 114. Epstein, L. B. (1976). In “Immunobiology of the Macrophage” (D. S. Nelson, ed.), p. 201. Academic Press, New York. Erb, P., and Feldmann, M. (1975a).Nature (London) 254,352. Erb, P., and Feldmann, M. (197513).Eur. J. Zmmunol. 5,759. Evans, R., and Alexander, P. (1970). Nature (London) 228,620. Evans, R., and Alexander, P. (1972). Nature (London) 236,168. Ferraris, V. A., and DeRubertis, F. R. (1974).J. Clin. Znuest. 54,378. Fowler, S . , and d e Duve, C. (1969).J. Biol. Chem. 244,471. Fowles, R. E., Fajardo, I. M., Leibowitch, J. L., and David, J. R. (1973).J.Exp. Med. 138, 952. Fujiwara, K., Sakai, T., Oda, T., and Igarashi, S. (1973).Biochem. Biophys. Res. Commun. 54,531. Gemsa, D., Stegmann, L., Menzel, J., and Till, G. (1975).J.Zmmunol. 114, 1422.

154

ROY C. PAGE ET AL.

Gery, I., and Handschumacher, R. E. (1974). Cell. Immunol. 11, 162. Gery, I., and Waksman, B. H. (1972).J.E x p . Med. 136, 143. Gery, I., and Wiener, E. (1975).J.Retioculoendothel. SOC. 18,252. Gery, I., Gershorn, R. K., and Waksman, B. H. (1971).J.Zrnmunol. 107, 1778. Gery, I., Gershorn, R. K., and Waksman, B. H. (1972).J. E x p . Med. 136, 128. Ginsburg, I. (1972).J.Infect. Dis. 126,294 and 419. Glasgow, L. A,, and Habel, K. (1963).J.E x p . Med. 117, 149. Goldstein, I., and Weissmann, G. (1974).J.Immunol. 113, 1583. Gordon, D., Bray, M. A., and Morley, J. (1976). Nature (London) 262,401. Gordon, S., and Cohn, Z. A. (1973). Int. Reu. Cytol. 36, 171. Gordon, S., and Werb, Z. (1976).Proc. Natl. Acad. Sci. U.S.A. 73,872. Gordon, S., Todd, J., and Cohn, Z. A. (1974a).J.E x p . Med. 139, 1228. Gordon, S., Unkeless, J. C., and Cohn, Z. A. (1974b).J.E x p . Med. 140,995. Corer, P. A. (1956).Adu. Cancer Res. 4, 149. Granger, G. A., and Kolb, W. P. (1963).J.Immunol. 101, 111. Granger, G. A,, and Weiser, R. S. (1964). Science 145, 1427. Granger, G. A., and Weiser, R. S. (1966).Science 151, 97. Greenbaum, L. M. (1975).In “Proteases and Biological Control” (E. Reich, D. B. Rifkin, and E. Shaw, eds.), p. 223. Cold Spring Harbor Lab.. Cold Spring Harbor, New York. Gresser, I. (1961).Proc. SOC. E x p . Biol. Med. 108, 799. Hahn, H. H., Char, D. C., Postel, W. B., and Wood, W. B. (1967).J.E x p . Med. 126,385. Hamprecht, B., Jaffee, B. M., and Philpott, G. W. (1973).FEBS Lett. 36, 193. Harington, J. S., Ritchie, M., King, P. C., and Miller, K. (1973).J.Pathol. 109,21. Heppleston, A. G., and Styles, J . A. (1967). Nature (London)214,521. Hibbs, J . B., Lambert, L. H., and Remington, J . S. (1972a). Science 177,998. Hibbs, J . B., Lambert, L. H., and Remington, J . S. (1972b).Nature (London)New B i d . 235,48. Higgs, G. A., and Youlton, L. J . F. (1972). Br. J. Pharmacol. 44,330. Hinman, J. W. (1972).Annu. Reo. Biochem. 41, 161. Hirsch, M. S., Zisman, B., and Allison, A. C. (1970).J . Immunol. 104, 1160. Hoffmann, M., and Dutton, R. W. (1971). Science 172, 1047. Holtermann, 0. A., Casale, G. P., and Klein, E. (1972).J. Med. 3,305. Horwitz, A. L., and Crystal, R. G. (1976). Biochem. Biophys. Res. Commun. 69,296. Huber, H., and Wagner, M. (1974). In “Activation of Macrophages” (W. H. Wagner and H. Hahn, eds.), p. 54, Excerpta Med. Found., Amsterdam. Hubscher, T. (1975).J. Immunol. 144, 1389. Ichikawa, Y., Pluznik, D. H., and Sachs, L (1967).Proc. Natl. Acad. Sci. U.S.A.58,1480. Janoff, A. (1972).Am. J. Pathol. 68, 579. Janoff, A., and Basch, R. S. (1971). Proc. SOC. E x p . B i d . Med. 136, 1054. Janoff, A., Blondin, J., Sandhaus, R. A,, Mosser, A., and Malamud, C. (1975).I n “Proteases and Biological Control” (E. Reich, D. B. Rifkin, and E. Shaw, eds.), p. 603. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Kantrowitz, F., Robinson, D. R., and McGuire, M. B. (1975).Nature (London)258,737. Kaplan, A. M., Morahan, P. S., and Regelson, W. (1974).J.Natl. Cancer Inst. 52, 1919. Keller, R. (1972). Schweiz. Med. Wochenschr. 102, 1148. Keller, R. (1973).J. E x p . Med. 138,625. Keller, R. (1974). Immunology 27, 1974. Keller, R., and Hess, M. W. (1972). Br. J. E x p . Pathol. 53, 570. Keller, R., and Jones, V. E. (1971). Lancet 2,847. Keller, R., Ogilvie, B. M., and Simpson, D. M. (1971). Lancet 1,678.

THE MACROPHAGE AS A SECRETORY CELL

155

Kilroe-Smith, T. A., Webster, I., Van Drimmelen, M., and Marasas, L. (1973). Enoiron. Res. 6,298. King, M. K., and Wood, W. B. (1958).J.E x p . Med. 107,279. Kohler, P. F. (1973).J.Clin. Znoest. 52,617. Kono, Y., and Ho, M. (1965). Virology 25, 162. Krahenbuhl, J. L., Rosenberg, L. T., and Remmington, J. S. ( 1 9 7 3 ) ~Immunol. . 111,992. Lai, A., Fat, R. F. M., and Van Furth, R. (1975). Zmmunology 28,359. Lazarus, G . S. (1973). Lysosomes Biol. Pathol. 3,338. Leibovich, S. J., and Ross, R. (1975).Am. J . Pathol. 78,71. Leibovich, S. J., and Ross, R. (1976).Am. J . Pathol. 84, 501. Levine, L., Kinkle, P. M., Voelkel, E. F., and Tashian, A. H. (1972).Biochem. Biophys. Res. Commun. 47,888. Lewis, M. R. (1925).Johns Hopkins Hosp. Bull. 36,361. Littleton, C., Kessler, D., and Burkholder, T. M. (1970). Zmmunology 18,693. Lohmann-Matthes, M. L., Schipper, H., and Fischer, H. (1972).Eur. J . Zmmunol. 2,45. McCandless, E. L. (1965).Ann. N . Y. Acad. Sci. 118,867. McIvor, K. L., and Weiser, R. S. (1971).Zmmunology 20,315. Magliulo, E., De Feo, V., Stripe, A., Riva, C., and Scevola, D. (1973). Clin. E x p . Zmmunol. 14,371. Mariano, M., and Spector, W. G. (1974).J.Pathol. 113, 1. Mauel, J., and Defendi, V. (1971).Exp. Cell Res. 65,377. Melsom, H., and Seljelid, R. (1973).J.E x p . Med. 137,807. Melsom, H., and Seljelid, R. (1974).Exp. Cell Res. 86,415. Melsom, H., Keamy, G . , Gruca, S., and Seljelid, R. (1974).J.E x p . Med. 140, 1085. Meltzer, M. S. (1976). Cell. Immunol. 22, 176. Meltzer, M. S., and Bartlett, G. L. (1972).J.Natl. Cancer Znst. 49, 1439. Metcalf, D. (1971). Zmmunology 21,427. Metchnikoff, E. (1905). “Immunity in Infective Diseases” (F. G. Binnie, trans].). Cambridge Univ. Press, London, and New York. Miller, H. R. P., and Avrameas, S. (1971).Nature (London)New Biol. 229, 184. Mooney, J. J., and Waksman, B. H. (1970).J.Immunol. 105, 1138. Mosier, D. E. (1967). Science 158, 1573. Movat, H. Z., Macmorine, D. R. L., and Takeuchi, Y. (1971). Znt. Arch. Allergy Appl. Zmmunol. 40,218. Miiller-Eberhard, H. J. (1975).In “Proteases and Biological Control” (E. Reich, D. B. Rifkin, and E. Shaw, eds.), p. 229. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Myatt, L., Bray, M. A., Gordon, D., and Morley, J. (1975). Nature (London)257,227. Nath, I., Poulter, L. W., and Turk, J. L. (1973). Clin. Exp. Zmmunol. 13,455. Nathan, C. F., Karnovsky, M. L., and David, J. R. (1971).J.Erp. Med. 133, 1356. Ohuchi, K., Sato, H., and Tsuoufuji, S. (1976). Biochim. Biophys. Acta 424,439. Page, R. C., Davies, P., and Allison, A. C. (1973). Arch. Oral Biol. 18, 1481. Page, R. C., Davies, P., and Allison, A. C. (1974a). Lab. Inuest. 30,568. Page, R. C., Davies, P., and Allison, A. C. (1974b).J. Reticuloendothel. Soc. 15,413. Pantalone, R. M., and Page, R. C. (1975).Proc. Natl. Acad. Sci. U.S.A.72,2091. Pantalone, R. M., and Page, R. C. (1977).J.Reticuloendothel SOC. 21,341. Pick, E., and Turk, J. L. (1972). Clin. Exp. Zmmunol. 10, 1. Pick, E., Krejci, J., Cech, K., and Turk, J. L. (1969). Immunology 17, 741. Piessens, W. F., Churchill, W. H., and David, J. R. (1975).J.Immunol. 144,293. Piper, C. E., and McIvor, K. L. (1975). Cell. Zmmunol. 17,423.

156

ROY C. PAGE ET AL.

Piper, P. J., and Vane, J. R. (1971). Nature (London) 223,29. Quesenberry, P. J., Morley, A., Stohlman, F., Jr., Rickard, K., Howard, D., and Smith, M. (1972). N . Engl. J . Med. 286, 227. Rabinovitch, M. (1970).In “Mononuclear Phagocytes” (R. van Furth, ed.), p. 299. S. A. Davis and Co., Philadelphia. Ramwell, P. W., and Shaw, J. E. (1970). Recent Prog. H o r n . Res. 26, 139. Reich, E. (1975).Zn “Proteases and Biological Control” (E. Reich, D. B. Rifkin, and E. Shaw, eds.), p. 333. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Remold, H. G., Katz, A. B., Haber, E., and David, J. R. (1970). Cell. Immunol. 1, 133. Remold-O’Donald, E. (1974).J. Biol. Chem. 249,3615. Remold-O’Donald, E., and Remold, H. G. (1974).J.Biol. Chem. 249.3622. Robertson, P. B., Shru, K. W., Vail, M. S., Taylor, R. E., and Fullmer, H. M. (1973). J. Dent. Res. 52, 189. Robertson, W. Van B., and Schwartz, B. (1953).J. Biol. Chem. 201,689. Robinson, D. R., Tashjian, A. H., and Levine, L. (1975).J.Clin. Znoest. 56, 1181. Rosenstreich, D. L., Farrar, J. J., and Doughtery, S. (1976).J. Zmmunol. 116, 131. Rosenthal, A. S., ed. (1975). “Immune Recognition.” Academic Press, New York. Rother, U., Thorbecke, G. J., Stecher, V. J., Hurlimann, J., and Rother, K. Immunol. 143549, 1968. Ruscetti, F. W., and Chervenik, P. A. (1974).J. Lab. Clin. Med. 83,64. Saito, K., and Suter, E. (1965).J . E x p . Med. 121, 739. Samuelsson, B., and Paoletti, R. (1976). “Advances in Prostaglandin and Thromboxane Research,” Vols. 1 and 2. Raven, New York. Sanuelsson, B., Granstrom, E., Greene, K., and Hamberg, M. (1971). Ann. N.Y. Acad. Sci. 180, 138. Schmidtke, J., and Unanue, E. R. (1971). Nature (London),New Biol. 233,84. Schorlemmer, H.-U., and Allison, A. C. (1976).Immunology 31,781. Schorlemmer, H.-U., Allison, A. C., and Davies, P. (1976). Nature (London) 261,48. Schrader, J. W. (1973).J.E x p . Med. 138, 1466. Schwab, J. J., Cromartie, W. J., and Robertson, B. S. (1959).J.E x p . Med. 109,43. Senior, R. M., Bielefeld, D. R., and Jeffrey, J. J. (1972).Clin. Res. 20,83. Shevach, E. M., Jaffe, E. S., and Green, I. (1973). Transplant. Reo. 16,3. Shortman, K., and Palmer, J. (1971). Cell. Zmmunol. 2,399. Smith, T. J,, and Wagner, R. R. (1967a).J.E x p . Med. 125,559. Smith, T. J., and Wagner, R. R. (1967b).J . E x p . Med. 125,579. Snyderman, R., Shin, H. S., and Dannenberg, A. M. (1972).J.Zmmunol. 109,896. Spector, W. G., and Heesom, N. (1969).J.Pathol. 98,31. Spencer, H. (1968). “Pathology of the Lung,” 2nd ed., p. 434. Pergamon, Oxford. Stanley, E. R., Hansen, G., Woodcock, J., and Metcalf, D. (1975). Fed. Proc., Fed. Am. SOC.E x p . Biol. 34,2272. Stecher, V. J. (1970). In “Mononuclear Phagocytes” (R. van Furth, ed.), p. 133. S. A. Davis and Co., Philadelphia. Steinman, R. M., and Cohn, Z. A. (1972).J.Cell Biol. 55,616. Stevenson, R. D. (1974). Clin. E x p . Zmmunol. 17, 601. Tsoi, M. S., and Weiser, R. S. (1968).J.Natl. Cancer Znst. 40,23. Unanue, E. R. (1972). Adv. Immunol. 15,95. Unanue, E. R., Kiely, J. M., and Calderon, J. (1976).J. E x p . Med. 144, 155. Unkeless, J. C., Tobia, A., Ossowski, L., Quigley, J. P., Rifkin, D. B., and Reich, E. (1973).J. E x p . Med. 127,85. Unkeless, J. C., Gordon, S., and Reich, E. (1974).J.E x p . Med. 139,834.

T H E MACROPHAGE AS A SECRETORY CELL

157

Velo, G. P., Dunn, C. J., Giroud, J. P., Timsit, J., and Willoughby, D. A. (1973).J.Puthol. 111, 149. von Euler, S. (1935). Klin. Wochenschr. 14, 1182. Wahl, L. M., Wahl, S. M., Mergenhagen, S. E., and Martin, G. R. (1974). Proc. N n t l . Acnd. Sci., U S A . 71,3598. Wahl, L. M., Wahl, S. M., Mergenhagen, S. E., and Martin, G. R. (1975). Science 187, 261. Wannamaker, L. W., and Matsen, J. M., eds. (1972). “Streptococci and Streptococcal Diseases.” Academic Press, New York. Ward, P. A., and Hill, J. H. (1970).J.Zmmunol. 104,535. Wasserman, S. I., C o e d , E. J., and Austen, K. F. (1975).J.Zmmunol. 114, 645. Weiser, R. S., Heise, E., McIvor, K., Han, S., and Granger, G. (1969).In “Cellular Recognition” (R. T. Smith and R. A. Good, eds.), p. 215. Appleton, New York. Weiss, L. P., and Fawcett, D. W. (1953).J.Histochem. Cytochem. 1,47. Weissmann, G., Dukor, P., and Zurier, R. B. (1971).Nnture (London),New B i d . 231, 131. Werb, Z., and Gordon, S. (1975a).J . Erp. Med. 142,346. Werb, Z., and Gordon, S. (1975b).J.Erp. Med. 142,361. Werb, Z., and Reynolds, J. J. (1974).J.Erp. Med. 140, 1482. Wills, E . D., Davies, P., Allison, A. C., and Haswell, A. D. (1972).Nature (London), New Biol. 240, 58. Wood, D. D., and Cameron, P. M. (1975).J.Zmmunol. 144, 1094. Wood, D. D., and Cameron, P. M. (1976). Cell. Zmmunol. 21, 133. Wood, D. D., and Gaul, S. L. (1974).J.Zmmunol. 113,925. Wood, D. D., Cameron, P. M., Poe, M. T., and Morris, C. A. (1976). Cell. Zmmunol. 21, 188. Wyatt, H. V., Colten, H. R., and Borsos, T. J. (1372). Immunology 108, 1609. Wynne, K. M., Spector, W. G., and Willoughby, D. A. (1975).Nature (London)253,636. Yoshinaga, M., Nakamura, S., and Hayashi, H. (1975).J . Zmmunol. 115, 533.

This Page Intentionally Left Blank

Biogenesis of the Photochemical Apparatus TIMOTHYTREFFRY Depurtment of Biochemistry, The Unioemity, Sheffield, United Kingdom

I. Introduction . . . . . . . . . A. Scope of This Article . . . . . . B. Background . . . . . . . . 11. Biogenesis of Chlorophyll . . . . . . A. Formation of 6-Aminolevulinic acid, . . . B. Formation of Protochlorophyllide . . . . C. Spectroscopic Studies of ProtochlorophyllideChlorophyll Transformations in Vivo . . . D. Formation of Chlorophyll a . . . . . E. Formation of Chlorophyll b . . . . . 111. Biogenesis of Chloroplast Membranes . . . A. Chloroplast Development in the Green Leaf . B. Structural Changes in the Etiochloroplast . . C. Developmental Changes in the Surface Properties of Etiochloroplast Membranes . . . . D. Developmental Changes in Etiochloroplast Membrane Proteins . . . . . . E. Formation of Chlorophyll-Protein Complexes . F. Biosynthesis of Lipids . . . . . . IV. Development of Photochemical Activity . . . A. Components of the Photochemical Apparatus . B. Development of the Photosystems . . . . C. Development of Other Photochemical Activities . V. Concluding Remarks . . . . . . . References . . . . . . . . . Note Added in Proof' . . . . . . .

. . .

.

. .

. .

. . . .

159 159 160 160 160 163 165 170 172 173 173 173 177

. . . . . . . . . .

179 180 183 185 186 187 189 190 191 196

I. Introduction A.

SCOPE OF THIS ARTICLE

In recent years several reviews dealing with general (Kirk, 1970; Arntzen and Briantis, 1975) or particular (Rosinski and Rosen, 1972; Bishop, 1974; Anderson, 1975a; Thornber, 1975; Leech, 1976) aspects of chloroplast structure and function have appeared. This article concerns the biogenesis of the photochemical apparatus, regarding it mainly as a component of the chloroplast membrane, We restrict coverage almost entirely to work on higher plants. Within this area we deal mainly with topics in which we have a particular interest and neglect others. We have sought to be in part critical and in part speculative, rather than encyclopedic. 159

160

TIMOTHY TREFFRY

B. BACKGROUND The study of chloroplast development and the biogenesis of photochemical activity in higher plants has been based almost entirely on the phenomenon of etiolation. One of the terminal steps in the biosynthesis of chlorophyll is the reduction of protochlorophyllide to chlorophyllide (Fig. 1).In gymnosperms this can take place in the dark, but in higher plants it requires light. The photoreceptor appears to be protochlorophyllide itself. If embryos from wheat are transplanted into pine megagametophytes the resulting wheat seedlings form chlorophyll in the dark (Bogdanovib, 1973), but normally seedlings of higher plants germinated and grown in darkness do not contain chlorophyll. They do accumulate some protochlorophyllide which appears to limit further synthesis. Protochlorophyllide is concentrated in fluorescent centers in the plastids of etiolated leaves, which are called etioplasts (rather than chloroplasts, as they lack chlorophyll). Electron micrographs show that most of the internal membrane of etioplasts forms a highly regular structure, the (PLB), prolamellar body which is presumably the fluorescent center observed in light microscopy. When etiolated plants are illuminated, protochlorophyllide is reduced to chlorophyllide which is esterified with phytol to give chlorophyll a, and the synthesis of further chlorophyll is stimulated. The PLB loses its regular structure, and its membranes give rise to sheets of lamellae; further membrane synthesis is stimulated, and normal chloroplast structure is established. An excellent review of these processes has been provided by Rosinski and Rosen (1972) and is outlined in Fig. 1. The apparent ease with which chloroplast development can be observed in etiolated seedlings following illumination has led to a substantial body of work. Much of this was done with seedlings that had suffered prolonged etiolation and showed a pronounced lag in development following illumination. More recent results, referred to elsewhere in this article, indicate that, when etiolation is reduced to a convenient minimum and high humidity is maintained during illumination, there is no lag phase in chlorophyll synthesis and photochemical activities can be measured after a few minutes. 11. Biogenesis of Chlorophyll A. FORMATION OF 6-AMINOLEVULINIC ACID 8-Aminolevulinic acid (ALA) is widely recognized as the first committed metabolite in the pathway leading to porphyrin biosynthesis.

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

161

Pide I I

!

.

comple

.

FIG. 1. A diagrammatic representation of stages in the development of the etiochloroplast, in the synthesis of chlorophyll, and in the development of the photochemical apparatus. For discussion and references see the text. (a) The proplastid has two compartments, the stroma (stippled) and the intermembrane space. ( b ) The stroma membranes perforate and curve, producing the PLB (c) which may remain stable for a period after illumination. ( d ) The PLB breaks down to give perforated thylakoids. (e)An initial stage in the production of grana. (f) A granum, partition zones (double hatching), and unstacked membranes (single hatching). Diagrams are not to scale, but magnification increases from (a) to (f). Biosynthesis of protochlorophyllide. (1)Glutamate, (2) a-oxoglutarate, (3) dioxovalerate, (4) A M ; Proto, Protophorphyrin; Pide, protochlorophyllide; Cide, chlorophyllide; Cyll, chlorophyll. The terminal steps may take place in association with the apoprotein. Biosynthesis of the photochemical apparatus. Several photoacts are indicated, as protochlorophyllide molecules in the complex must be reduced individually. A conformational change in the complex is also indicated. PSU represents the nonchlorophyll components of the photochemical apparatus and is an intrinsic component of the etioplast membrane.

Until recently it was assumed that the biosynthesis of the chlorin ring of chlorophyll was basically similar to that of the porphyrin moiety of heme. Succinate

+ glycine

A L A synthetase

AM

ALA dehydratase

porphobilinogen (PBG)

162

TIMOTHY TREFFRY

It has been suggested (Granick, 1959; Bogorad et al., 1968) that ALA synthetase is an important regulatory enzyme, that its synthesis is stimulated by light, and that its rapid turnover prevents substantial accumulation of protochlorophyllide from occurring in the dark. If this were so, ALA synthetase in the pathway leading to chlorophyll synthesis would behave in a way very different from that in the pathway leading to heme, where such effects have not been reported. However, it now appears that ALA synthetase is not associated with chlorophyll synthesis, so that the need to explain different behaviors of the same enzyme in these two pathways does not arise. Although Wellburn and Wellburn (1971) found that isolated etioplasts produced ~hlorophyll-'~C when supplied with glycine-14Cand concluded that ALA synthetase was active in etioplasts, they now believe their data to be spurious. The use of levulinic acid (LA), a potent inhibitor of ALA dehydratase in the presence of which ALA accumulates, permits a more direct approach to the study of ALA synthesis, and Wellburn (1975b) was unable to detect synthetase activity in etiochloroplasts (developing chloroplasts found in etiolated leaves after illumination) using this technique. In retrospect, the finding of Nadler and Granick (1970) that succinate and glycine did not stimulate the greening of etiolated barley, although ALA did, may also have been indicative of the noninvolvement of ALA synthetase in greening. Beale and Castelfranco (1974) showed that, in the presence of LA, glutamate, glutamine, and a-oxoglutarate are considerably more effective precursors of ALA than glycine. In later work, using specifically labeled precursors and degrading the ALA produced to establish its labeling pattern (Beale et aZ., 1975), it was shown that the carbon skeleton of glutamate is incorporated directly into ALA. The suggested pathway is: Glutamate

-

a-oxoglutarate

-

a,&dioxovalerate

-

ALA

Similar experiments carried out concurrently by Meller et al. (1975) have confirmed this result. Fluhr and Hare1 (1975) and Wellburn (1975b) showed that succinyl-coenzyme-A synthetase does not appear to be associated with etioplasts and that its activity is not increased by greening. By examining the incorporation of label from citric-1,5-14Cacid into chlorophyll in greening Avena, Wellburn (1975a) also provided support for the glutamate pathway. Although control of the formation of ALA can no longer be ascribed to ALA synthetase, control of the formation of ALA must still be recognized as a key factor in the control of chlorophyll biosynthesis.

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

163

When actively greening leaves (etiolated maize illuminated for 4 hours) are returned to the dark in the presence of LA (Fluhr et al., 1974), accumulation of ALA, which is rapid initially, ceases after 6 hours. The half-life for the enzyme system producing ALA appears to be 60-80 minutes. In experiments with Chlorella, Beale (1971) has suggested that turnover of the ALA-synthesizing system, which has a half-life of 30 minutes, can account for the entire requirement for protein synthesis associated with greening. When Chlorella are treated with LA, synthesis of chlorophyll following a subsequent treatment with cycloheximide mirrors the fall in accumulated ALA. The experiments of Fluhr et al., however, indicate that controls at a level other than ALA synthesis also operate. When actively greening maize is returned to the dark, protochlorophyllide accumulation ceases more rapidly than ALA accumulation in leaves treated with LA, indicating that control is primarily due to feedback inhibition rather than to turnover of the ALA-synthesizing system. Although the time course of protochlorophyllide accumulation in the dark is slowed by LA, the ultimate level of protochlorophyllide attained is similar and considerably less than could be produced by the level of ALA accumulated concomitantly. The high level of ALA accumulating in the presence of LA may be due in part to the inhibition of back-reactions in the ALA-synthesizing system, which is presumably reversible, as well as to its inhibition of ALA dehydratase.

B. FORMATION OF PROTOCHL~ROPHYLLIDE Several workers have established that, when etiolated leaves are supplied with ALA in the dark, levels of protochlorophyllide accumulate that greatly exceed the normal level. Also, labeled chlorophyll is formed in isolated etioplasts supplied with ALA-14C (Wellburn and Wellburn, 1971). Thus an anabolic capacity [ALA + (proto)chlorophyll(ide)] exists. Nadler and Granick (1970) suggest that all the enzymes required for this portion of the pathway are nonlimiting, at least during the first 12 hours of greening in etiolated barley. Schneider (1973) reports that tobacco tissue cultures and etiolated maize seedlings differ as to whether ALA dehydratase and related enzymes increase during greening. Hampp and Ziegler (1975)report an increase in ALA dehydratase activity in isolated Avena etioplasts, which is enhanced by white light or 660-nm light and is insensitive to cycloheximide. No response is obtained in 731-nm light or in the presence of chloramphenicol. The responses observed, however, are about 20%,whereas a manyfold increase in activity occurs on about the eighth day of etiolation even in the absence of treatment.

164

TIMOTHY TREFFRY

Sundqvist et aZ. (1975)also report an increase in dehydratase activity during the greening of etiolated wheat. The precise steps between PBG and protochlorophyllide and their enzymology have yet to be described (for a review, see Rebeiz and Castelfranco, 1973),and we rely rather heavily in this area on studies of heme biosynthesis which may or may not be directly relevant. The terminal steps in the biosynthesis of protochlorophyllide have also been the subject of a series of articles by Ellsworth and co-workers (Ellsworth and Lawrence, 1973; Ellsworth and Hsing, 1973, 1974; Ellsworth et aZ., 1974),who recently reported the biosynthesis of protochlorophyllide a in a cell-free extract of etiolated wheat (Ellsworth and Hervish, 1975). In much of the literature, the term “protochlorophyll” is used when nonesterified protochlorophyllide (magnesium vinyl pheoporphyrin as) is the predominant form or when the procedures used fail to distinguish between the esterified and nonesterified forms, Similar confusion often exists between “chlorophyllide” and “chlorophyll” in reference to the pigment first formed following illumination. In this discussion “(proto)chlorophyllide” refers to the acid form, and “(proto)chlorophyll” to the corresponding pigments after esterification with phytol. Where these forms are not rigorously identified, it is assumed that protochlorophyllide predominates before illumination and chlorophyllide immediately after illumination. At some stage during the formation of protochlorophyllide the pigment becomes associated with a protein, the complex being known as protochlorophyllide holochrome. For a review of early work in this area, see Boardman (1966).The apoprotein is similar to, but antigenically distinct from, carboxydismutase (fraction I protein). It has a molecular weight of 600,000 per chromophore (Boardman, 1966) and behaves in vivo as if 5 to 25 chromophores make up a unit of activity (see Section 11,C72). Protochlorophyllide holochrome is associated with the membranes of the PLB (Boardman and Anderson, 1964; Kahn, 1968) and, at least in etiolated beans, can be considered an extrinsic protein (using the terminology of Singer and Nicolson, 1972), as it is readily extractable with aqueous buffers of low ionic strength. Protochlorophyllide holochrome preparations have also been obtained from beans using Triton X-100 (Schoper and Siegelman, 1968) and saponin (Henningsen and Kahn, 1971),having molecular weights of 300,000 and 99,000, respectively. Henningsen and Kahn studied more closely a saponin protochlorophyllide holochrome complex from barley. It has a molecular weight of 63,000 for one chromophore and a

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

165

red absorption maximum at 644 nm and is fully photoreducible to chlorophyllide holochrome having a red absorption maximum at 678 nm. At room temperature, but not at 0°C (Henningsen et d.,1974), this absorption maximum shifts to 670 nm, and the apparent molecular weight of the complex drops to 29,000. However, the saponin chlorophyllide holochrome complex isolated from etiolated leaves 15 minutes after illumination has an apparent molecular weight of 100,000. The photoreduction of protochlorophyllide is accompanied by a shift in its red absorption maximum to longer wavelengths and can therefore be studied by direct spectroscopic examination of intact etiolated leaves, isolated etioplasts, or isolated holochrome preparations (Thorne and Boardman, 1971). Protochlorophyllide is not photoreducible when removed from the apoprotein, nor are all holochrome preparations or all the protochlorophyllide in intact leaves photoactive. C. SPECTROSCOPIC STUDIESOF PROTOCHLOROPHYLLIDEin Vivo CHLOROPHYLL TRANSFORMATIONS

1. Spectral Forms of Protochlorophyllide The development of techniques that allow absorption spectra of intact leaves to be recorded (Shibata, 1957), and of spectrophotometers specifically designed for recording spectra of scattering samples, has permitted an enormous number of experiments to be performed which simply record the absorption maxima and various shifts in absorption maxima of chlorophylls in vivo. Interpretation of these data in molecular terms lags far behind the recording of the phenomena. Most commonly spectra are obtained in double-beam instruments by addition to the reference beam of a nonpigmented material with scattering properties similar to those of the sample, or a material whose scattering properties greatly exceed those of the sample is interposed in both the sample and the reference beam. It is extremely difficult to produce identical optical properties in both the sample and the reference beam, and therefore the apparent absorption maximum of the sample is displaced accordingly and cannot be stated with great precision. This is especially true of etiolated systems in which the concentration of pigment is low. In experiments with intact leaves of etiolated seedlings it has been shown that at least two forms of protochlorophyllide are produced. One with an absorption maximum near 635 nm, protochlorophyllide 635 (P635, sometimes referred to as P63,), and the other with a red ab-

166

TIMOTHY TREFFRY

sorption band at 650 nm (P650). is directly photoactive, but P635 is not (Shibata, 1957; Murray and Klein 1971). Exogenous ALA appears to lead to the formation of P635, but in leaves incubated with ALA and given repeated flashes of light high levels of chlorophyllide are produced. P635 is apparently converted to P650 in experiments of this type (Sundqvist, 1969; Murray and Klein, 1971; Gassman, 1973b). Granick and Gassman (1970) believe that the change P630 + Peso is due to the formation of a complex between the pigment and the photoenzyme (protochlorophyllide holochrome?) and a reductant. This reductant is apparently able to reduce dichlorophenolindophenol (DCIP) (Becker, 1974). Recently Griffiths (1975) and Brodersen (1976) showed that, in isolated etioplasts of both barley and maize incubated with NADPH, inactive P,, is converted to active P650. Horton and Leech (1972) showed that ATP inhibits the degradative P650 + P635 shift that occurs in isolated etioplasts. Gassman (1973a) showed that active P, can be converted reversibly to inactive P,, by hydrogen sulfide. Kahn et al. (1970) identified three forms of protochlorophyllide in etiolated beans. Pm8,which fluoresces at 630 nm and is inactive, and P, and P, which fluoresce at 655 nm and are both photoconvertible. Dujardin and Sironval (1970) point out that protochlorophyllide in solution absorbs at 628 nm and suggest that P628 is inactive because the chromophore is detached from, or only weakly attached to, the apoprotein. Bovey et al. (1974) showed that the nonconvertible form in etiolated beans absorbing at 628 nm is protochlorophyll and distinguish it chromatographically from the Pa36 formed in response to A M . They imply that this form in turn has a Rf different from that of native P636 but do not present their results. P650 is converted to chlorophyllide 684 ( c 6 8 4 ) on illumination (see Section II,C,2), but freezing and thawing (Butler and Briggs, 1966; Henningsen, 1970), warming (Henningsen, 1970), or treatment with glutaraldehyde (Treffry, 1968) produces P630-635 which forms C6,O on illumination. This is a situation different from that obtaining in unmanipulated systems where P, and P650 are both convertible and give rise to C,. Kahn and Nielsen (1974) attempted to preferentially convert P,, or PW but were unable to do so and concluded that PS3,and P650 either are a single species ordered in such a way as to cause splitting of the red absorption band or are two species in dynamic equilibrium. Virgin and French (1973) suggest that P,, and P650 arise from a common precursor, and that the latter is directly convertible but that P,, changes to P, before conversion. Dujardin and Sironval (1970) regard Pa as a monomeric form and P650 as a form of protochlorophyl-

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

167

lide holochrome in which the chromophores are sufficiently close to produce a red shift in absorption (Boardman, 1967). 2. Shifts in Red Absorption Maxima lmmediately following 11lumination

Over the last 20 years the catalog of spectral changes that occur in etiolated leaves at various times following illumination has become increasingly complex (Fig. 2). a. Shifts in Absorption between 650 and 684 nm. In the original experiments of Shibata (1957), the absorption spectra of etiolated bean leaves were recorded before and after 1 minute of illumination, and the red absorption maximum which had been at 650 nm in the etiolated leaves was found to have shifted to 684 nm. Illumination for 10 seconds gave the same result, but in view of later results the time interval between illumination and the recording of the absorption spectrum was presumably somewhat longer than this. Shibata, 1957

- -

Granick and Gassman, 1970

678 ----+ 684 ---+ 673

635 ----+ 650 Thorne and Boardrnan, 1971

650

668

-

678

----+

684 ---+ 673

684

--:-+

673 ----+ 677

Bauer and Siegelrnan, 1972

650

678

_-_-_ + + C-

red

Virgin and French, 1973

630

----+

650

IoWyield h i g h yield

Ogawa et al., 1974

-

,672, --* ,67211

684 -----* 672111

650 670 +678 ----+ 684 ----+ 673 ----+ 677 FIG. 2. Shifts in absorption maxima (in nanometers) in etiolated leaves following illumination. Room temperature except where indicated. Solid arrow, light-induced; dashed arrow, light-independent =, shift with an isobestic point; #, shift without an isobestic point.

168

TIMOTHY TREFFRY

By obtaining spectra within 30 seconds of a flash illumination many workers (Gassman et al., 1968; Sironval et al., 1968; Bonner, 1969; Granick and Gassman, 1970) have detected an intermediate species absorbing at 678 nm. The 650 + 678 shift is rapid, requires light, and can occur at - 70°C. It is followed by a dark (non-light-requiring) shift, 678 + 684, which is slower and temperature-sensitive and has an isosbestic point indicating direct conversion without intermediates. Bauer and Siegelman (1972)also reported the 678 intermediate. They showed that the 678 + 684 dark shift could be reversed by red light in a process that lacked an isosbestic point. This shift was restored during a subsequent dark period. The system behaved reversibly for several cycles, diminishing as 684 was lost through the 684 + 673 dark shift. Using light flashes, or low light intensity, Litvin and Belyaeva (1968; 1971),resolved the 650 + 678 shift into two light-requiring steps with an intermediate fluorescing at 675 nm. A similar intermediate absorbing near 670 nm has been reported after conversion of only part of the available P6, (Thorne 1971), or when leaves are plunged into liquid nitrogen immediately after illumination (Ogawa et al., 1974). Thorne detected this intermediate by its excitation and emission maxima E&674. This intermediate is stable in the dark for at least 5 hours and only on further illumination is it replaced by E678F687. An interesting feature of this work is that after fractional conversion of P65, energy transfer can be detected between the residual P65, and each of the longer-wavelength absorbing forms identified (Fig. 2). Transfer from to E678F687 decreases with time, indicating that the latter changes its location. Sironval and co-workers (Sironval et al., 1968; Sironval and Brouers, 1970) report only the intermediate absorbing at 676 nm and claim an isosbestic point in the F657 + F688 (i.e., 650 + 676) shift. Although this may be taken as evidence against the intermediate reported by Thorne, an absolutely sharp isosbestic point is difficult to obtain and the possibility of an intermediate present at a low concentration and emitting near the isosbestic point (675 nm) (cf. cannot be excluded. Thorne’s E6&,74) Sironval and Kuyper (1972) claim to distinguish phototransformation of protochlorophyllide from its reduction (separable as a dark step) and also reduction of the first molecule in a protochlorophyllide complex from reductions of the remaining molecules. They regard the intermediate absorbing at 668 nm as being due to a mixed population of phototransformed protochlorophyllide and photoreduced protochlorophyllide (chlorophyllide) within a single transfer unit. The time course of photoreduction is sigmoid. The model presented by SironVal and Kuyper thus shows interesting parallels with cooperative ef-

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

169

fects in oligomeric enzymes where a change occurring in one subunit affects the probability of the same change occurring in an adjacent subunit (Koshland et al., 1966). Virgin and French (1973) detected only one intermediate in the 650 + 684 nm shift (see Fig. 2), a species absorbing at 672 nm designated 6721to distinguish it from another species 67211formed by lowintensity illumination and the 672111 formed as a result of the Shibata shift (see Section II,C,2,b). 672,, may be analogous to the E 6 6 8 F 6 7 4 described by Thorne (1971), but Thorne rejects the possibility that this species is the terminal product of a parallel photoact rather than an intermediate in the biosynthetic sequence. For a recent description of forms of protochlorophyllide, chlorophyllide, and chlorophyll produced before and after illumination, see Klockare and Sundqvist (1977). b. The Shibata Shift. Over a period of minutes (the time is increased at low temperature) or in leaves illuminated after prolonged etiolation, C,,, is replaced by C,,,. This “blue” shift was first observed by Shibata (1957) and is often referred to as the Shibata shift. Spectroscopic observations during this shift show a sharp isosbestic point (at 676 nm), which has been taken to indicate a simple conversion of one form of chlorophyll(ide) a into another. By rapidly freezing leaves at various times during the shift and examining their spectra at low temperature, Virgin and French (1973) showed that the resulting spectra consist of four components rather than two. These components, having red absorption maxima at 662, 670, 677, and 683 nm, were previously identified by spectral analysis of mature chloroplasts, as four universal forms of chlorophyll a (French et aZ., 1972). Virgin and French therefore ascribe the Shibata shift to a developmental change in the proportion of these components. Shibata ( 1973) developed a dual-wavelength spectrophotometer suitable for scanning thin-layer plates and, by freezing leaves at various times after illumination and extracting their pigments, examined chromatographically the chlorophylls present before and after the Shibata shift (Ogawa et al., 1974).The pigment present in leaves with an absorption maximum at 684 nm is chlorophyllide. Chlorophyll a appears 10 minutes after a 1-minute illumination but is preceded by an intermediate, X, with a slightly higher Rf in the methanol-dichloromethane-water system of Schneider (1966). The absorption maximum at 673 nm after the Shibata shift is due to a mixture of chlorophyll a and X. Two hours after illumination, when the absorption maximum in vivo is at 677 nm, only chlorophyll a is present. This association of the Shibata shift with esterification is controversial. Akoyunoglou and Michalopoulos (1971) have indicated that these processes, although

170

TIMOTHY TFWFFRY

broadly associated, have a different time course, while Henningsen and Thorne (1974) suggest that the time courses of the two processes are identical and are similarly affected by temperature. These workers point out that, although the E6,p6, --* E,&67, shift is associated with esterification, this event alone cannot explain the shift since chlorophyllide a and chlorophyll a have almost identical spectroscopic characteristics. Esterification presumably permits or requires a change in the environment of the pigment. As a result of the shift the quantum efficiency of the fluorescence doubles. The tl12for this effect is about twice that for esterification and the spectral shift. D. FORMATION O F CHLOROPHYLL a

Chlorophyll a differs from chlorophyllide in that it has a large lipophillic phytol derivative attached by an ester bond. The way in which esterification is achieved in vivo is controversial. The enzymic removal of phytol from chlorophyll is accomplished by chlorophyllase (chlorophyll-chlorophyllidohydrolase,EC 3.1.1.14). Water-soluble preparations of the enzyme have been made (Bacon and Holden, 1970; Ogura, 1972), and greater yields are obtained using organic solvents (Holden, 1961; Shimizu and Tamaki, 1972) or detergents (McFeeters et al., 1970; Ogura, 1972). Optimum hydrolytic activity is achieved in 40% acetone, but this may be required to solubilize the substrate (chlorophyll or pheophytin) (Bacon and Holden, 1970). When chlorophyll is added to a preparation of chlorophyllase, partition chromatography indicates that hydrolytic activity is predominant. However, repeated extractions with petroleum ether (Treffry, 1964), or, better, the formation of a paraffin emulsion in the incubation medium (T. Treffry, unpublished) forces complete reversal of the reaction, and chlorophyll is produced. Presumably, sequestration of chlorophyll into the lipid (or petrdleum ether) phase reverses the normal in vitro reaction. It is tempting to suggest a parallel to the in vivo situation in which the chromophore of relatively water-soluble protochlorophyllide holochrome is phytolated during the formation of insoluble membrane-bound chlorophyll. Shimizu and Tamaki (1963) showed that the hydrolytic activity of chlorophyllase in acetone extracts of tobacco can also be reversed by raising the level of phytol in the incubation medium. Holden (1963), however, was unable to produce this effect. In addition to their hydrolytic and esterification capacities chlorophyllase preparations also possess transesterification activity (Holden, 1963). Ellsworth (1972) showed by gel filtration of a chlorophyllase

BIOGENESIS OF T H E PHOTOCHEMICAL APPARATUS

171

preparation that these activities can be separately distributed in three different fractions. The enzyme responsible for transesterification reactions has a markedly lower molecular weight than the hydrolyzing or esterifying moieties. In higher plants the terminal stages of chlorophyll biosynthesis are generally taken to be: Protochlorophyllide

2

chlorophyllide

phyto!

chlorophyll a

However, some protochlorophyll is present in etiolated seedlings, and the ratio of protochlorophyllide to protochlorophyll may range from 0.15 to 9 in different species (Godnev et al., 1963). Controversy exists as to whether or not protochlorophyll is photoactive (Godnev et al., 1963, 1968). Bovey et al. (1974) have clearly determined protochlorophyll to be nonconvertible in etiolated beans. According to Smith (1960) and Rebeiz and Castelfranco (1973), protochlorophyll can be esterified with a range of other prenols including phytol. Jones and Ellsworth (1969) reported a protochlorophyllase in darkgrown wheat, but in later work (Ellsworth and Nowak, 1973) no esterification activity was observed in protochlorophyllase preparations with either phytol or farnesol as substrate. The protochlorophyll present in etiolated barley, at approximately one-tenth the concentration of protochlorophyllide, is esterified exclusively with protochlorophyllide geranyl geraniol ester (Liljenberg, 1974). A portion of this ester appears to form chlorophyllide geranyl geraniol ester on illumination. In most studies of the esterification of chlorophyllide the proportions of chlorophyllide and chlorophyll are determined by partition methods which fail to distinguish between pigments esterified with various prenols (Boardman, 1967; Treffry, 1970; Akoyunoglou and Michalopoulos 1971; Henningsen and Thorne, 1974). Granick (1967) suggested that phytol pyrophosphate may be an intermediate in the esterification of chlorophyllide with phytol. Phyto1 pyrophosphate appears to stimulate esterification reactions with phytol, but this may be a result of its detergent properties which make phytol more soluble (Watts and Kekwick, 1974). The presence of Triton X-100 inhibits the hydrolytic activity of chlorophyllase (McFeeters et al., 1971). Both these results may be due to a stimulation of esterification, detergent micelles having the same effect as that of oil emulsions described previously. Wellburn (1970) showed that neither phytol pyrophosphate nor geranyl geraniol pyrophosphate was esterified with chlorophyllide by a chlorophyllase preparation which, confirming the work of Chiba et

172

TIMOTHY TREFFRY

al. (1967), esterified methylchlorophyllide but not chlorophyllide with phytol. Esterification of methylchlorophyllide with geranyl geraniol was also demonstrated, but these activities are transesterifications rather than the de novo formation of an ester bond. In a later investigation Wellburn ( 1976) found chlorophyllide geranyl geraniol ester present to the extent of 0.03%of the chlorophyll in emerging leaves ofAesculus hippocastanum. This investigator provides a useful discussion of possible terminal steps in chlorophyll formation, suggesting that chlorophyllide is esterified with geranyl geraniol which is later partly saturated to give chlorophyll. There is really no firm evidence for this, although Wellburn (personal communication) has pointed out that chlorophyllide geranyl geraniol ester behaves chromatographically like an unknown intermediate formed during the Shibata shift (Ogawa et al., 1974). Because chlorophyllase is active in organic solvents and, in addition to its hydrolytic and esterification potential, also mediates transesterification reactions (Holden, 1963), the trace amount of chlorophyllide geranyl geraniol ester detected could have been produced artifactually during isolation. It may be difficult to produce clear evidence without an in vitro study using purified materials. The detailed study required, involving as it does a hydrophobic enzyme whose substrates and products vary widely in surface activity, is certain to be difficult (see Note Added in Proof).

E. FORMATION OF CHLOROPHYLL b The biosynthesis of chlorophyll b has been considered in an excellent review by Shlyk (1971),in which the sequential and common precursor hypotheses for the formation of chlorophylls a and b are considered. Shlyk believes the main evidence is that chlorophyll b is formed from chlorophyll a. Ellsworth et al. (1970) showed that substantial amounts of chlorophyll b- 14Cmay be formed from uniformly labeled chlorophyll a- 14Cby bean leaf homogenates. Certainly, in the greening of etiolated seedlings it is chlorophyll a that appears first. Chlorophyll b is detectible, to the extent of 1 molecule per 300 molecules, in etiolated peas after 10 minutes of illumination (Thorne and Boardman, 1971). Chlorophyll-b formation is very rapid, the ah ratio dropping to 15 in the next hour. This occurs while the total chlorophyll content remains constant. As net synthesis commences, chlorophyll b appears to accumulate preferentially, and the a/b ratio drops to about 3. This process is reversed in seedlings returned to the dark. Evidence for the photodependence of chlorophyll-b formation is provided by Argyroudi-Akoyunoglou and Akoyunoglou ( 1970).

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

173

111. Biogenesis of Chloroplast Membranes Like chlorophyll biosynthesis, the biogenesis of chloroplast membranes in higher plants has mainly been studied using etiolated seedlings. The perhaps more normal situation of development in the green leaf has proved more intractable; no developmental stages can be isolated, and any changes must be measured against a background of existing structures and activities.

A. CHLOROPLAST DEVELOPMENT IN THE GREENLEAF In the expanding leaf of a young plant under natural or artificial illumination chloroplasts can be seen to grow and divide by fission (Stetler and Laetsch, 1969; Possingham and Saurer, 1969). This process undoubtedly requires that membrane biosynthesis take place, but we are not aware of any detailed research into the manner in which it occurs. The effects of light quality (Possingham, 1973)and an association with DNA synthesis, both chloroplastic and nuclear (Rose et UZ., 1975), have been reported. Division has also been observed in preparations of isolated chloroplasts (Ridley and Leech, 1969), although in these experiments there was no evidence of membrane biogenesis.

B. STRUCTURAL CHANGESIN

THE

ETIOCHLOROPLAST

Structural changes in the developing etioplast and during chloroplast development following the illumination of etiolated leaves have been well documented in work summarized in an excellent review by Rosinski and Rosen (1972) and need only be outlined here.

1. Development of the PLB Elaboration of the PLB proceeds with prolonged growth in darkness and appears to parallel the accumulation of protochlorophyllide (Weier and Brown, 1970). This process continues even during prolonged etiolation, when the proportion of the nonconvertible form of protochlorophyllide is increasing. 2. Trunsformation of the PLB When plants are illuminated, transformation of the PLB takes place. This process appears to be linked with the phytolation of chlorophyllide rather than the photoreduction of protochlorophyllide, which precedes it (Treffry, 1970). During transformation the PLB first loses its characteristic structural regularity and forms sheets of “perforated thylakoids” (Gunning and Jagoe, 1967). The action spectrum for transformation (Henning-

174

TIMOTHY TREFFRY

sen, 1967) appears to be similar to the absorption spectrum of protophyllide. However, Bradbeer (1974b) reports that far-red light (700800 nm) causes transformation, while blue light (450 nm) does not. Far-red light would not be expected to photoreduce protophyllide, whereas blue light, unless the protophyllide is shielded by carotenoids, should do so. The membrane area in the PLB can account for all the thylakoid membranes observed during the first few hours of greening in white light (Gunning and Jagoe, 1967; Bradbeer et al., 1974a). During the treatment of etiolated beans with far-red light, however (Bradbeer et al., 1974b), the volume of the PLB per plastid decreases, as does the thylakoid area. Even the initial structural responses of etioplasts to light are complex, and the concept of a simple change in membrane configuration is difficult to reconcile with these results. 3. Development of Thylakoids and the Formation of Grana

Further development of chloroplast membranes also requires light, and the type and degree of development can be controled by light intensity, quality, and duration. These effects are explained to some extent by the importance of light in providing carbon compounds after photosynthesis has become established (Klein and Neuman, 1966; Dowdell and Dodge, 1970). Phytochrome and other light effects have been reported (reviews, Virgin, 1973; Sundqvist et al., 1976). When etiolated peas are exposed to low intensities of red or white light, membrane development is greatly impaired, PLBs enlarge and recrystallize (Henningsen and Boynton, 1970), and the protochlorophyllide content may increase to a higher steady-state level than that previously found in the dark (Treffry, 1973). At higher intensities of red light etiochloroplasts are found which contain grana and stroma thylakoids and regular PLBs. Such plastids do not contain protochlorophyllide but have a relatively high proportion of chlorophyllide. Plastids of similar structure are found in dark-grown pine seedlings which contain chlorophyll but lack protochlorophyllide and chlorophyllide (Michel-Wolwertz and Bronchart, 1974). Nonesterified forms of chlorophyll are therefore not an invariable characteristic of regular PLBs as suggested (Treffry, 1973). Plastids containing regular PLBs are also found in immature regions of light-grown corn leaves (Leech et al., 1973). Sironval and his associates have produced a considerable body of work on the effect of light on the development of etiochloroplasts and present the useful concept (Sironval, 1975) that chloroplast development may be induced to pass through a variety of steady states which

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

175

can be produced or modified by light treatment. The etioplast is the steady state produced in the dark. Intermittent illumination produces a chloroplast with unstacked thylakoids. Continuous illumination of adequate intensity and suitable quality produces a normal chloroplast. These states are to some extent interchangeable. Sironval’s results with etiolated beans are summarized in Table I. This ability to produce particular structural states has been of value in attempts to relate structure and function. Species differences as well as environmentally induced changes in the structured steps in plastid development have also been analyzed (Whatley, 1977). Other workers, notably von Wettstein (von Wettstein et aZ., 1974), have selected mutants that show variations in plastid structure and response to illumination, but aberrant states produced environmentally in normal plants are perhaps more attractive for research purposes. The rate of grana formation in etiolated leaves following illumination has been studied by many workers. Membrane growth is believed to take place by intersusception of new materials among existing components, rather than by the de no00 synthesis of new areas. In the developing chloroplast this concept is supported by the work of Lafleche et uZ. (1972), who supplied etiolated maize leaves with ALA-3H and, having shown chromatographically that prior to illumination label was incorporated into protochlorophyllide and after 24 hours of illumination into chlorophyll a and b, illustrated by electron microscope autoradiography that radioactivity was associated first with the PLB and later occurred randomly in both grana and stroma lamellae. Surprisingly, these workers also used these data to claim that not only the chlorin ring (which was labeled by this procedure) but also the protein moiety of the protochlorophyllide holochrome (which would not be expected to be labeled) was similarly redistributed. Such data may indicate, however, that as membranes grow their original components are spread and diluted with new material. In a study of grana formation in tobacco, Stetler and Laetsch (1969) have reported that grana are formed by a localized infolding and appression of contiguous areas of a stroma thylakoid as evidenced b y a Y-shaped granal profile frequently observed in sections. Some structural modification must subsequently occur if grana are free to separate in low-molarity buffers (Izawa and Good, 1966). In avocado and sunflower leaves grana membranes appear to arise from the membrane surrounding plastid inclusions (Cran and Possingham, 1974). In the mature chloroplast, stroma membranes and grana membranes differ in their chlorophyll ah ratio and photochemical activity and ap-

176

TIMOTHY TREFFRY

pear (Sane et al., 1970) to be separated by digitonin (Anderson and Boardman, 1966). Akoyunoglou and Argyroudi-Akoyunoglou (1974) showed that chloroplast membranes that have been dissociated and TABLE I BE PRODUCED IN TREATMENTS'

STRUCTURAL STATES THATCAN VARIOUS LIGHT

Conditions for preparation Chlorophyll-b content Plastid state* inside the leaf Etioplast

Darkness

Ptc

Flash regime (1-msec flash every 15 minutes) Sensitized ptc Flash regime followed by 6 minutes of weak continuous white light Far-red light Continuous PtC far-red light Induced ptc Flash regime followed by 6-10 minutes of intense continuous light Bound in pair, 2 minutes of PtC light every 100 minutes, or 2 minutes of light every 120 minutes Chloroplast Normal light continuously

Membrane organization

PLASTIDS BY

Oxygen evolution on turning intense continuous light on

First shuctural changes in intense continuous light

0

PLB

0

LOW

Primary thylakoids

0

Disaggregation of PLB Pair formation

Low

?

+

Pair formation

LOW

Primary thylakoids Bound in pair, primary thylakoids

+

Pair formation

+

Evolution to mna

LOW

Bound in pair, primary thylakoids

+

Evolution to grana

Normal

Grana stacks

+

LOW

From Sironval, 1975. ptc, Primary thylakoid chloroplast.

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

177

reassociated by variations in ionic strength can be fractionated with digitonin to produce the same range of chlorophyll a/b ratios and distribution of photochemical activity obtained in undissociated specimens. This indicates that the reassociation is specific and reestablishes connections between lamellar regions previously linked in the native grana. Anderson (1975a) provides a detailed account of the ionic and molecular phenomena associated with membrane stacking in the mature chloroplast and also of studies with mutants. Since the preparation of her review, Staehelin (1975) has shown that stacking results in the regular association of recognizable particles on the surfaces in apposition. It has also been reported (Miller and Staehelin, 1976) that particles associated with coupling factor activity (In phosphorylation) are specifically restricted to regions of the lamellae that do not stack. If stacking is associated with specific membrane components, these components must be produced during development in discrete regions of the thylakoids or aggregate at particular sites after diffusion within the plane of the membrane.

C. DEVELOPMENTAL CHANGESIN THE SURFACE PROPERTIES OF ETIOCHLOROPLAST MEMBRANES During the development of the PLB in the dark, its transformation, and the subsequent development of the grana and stroma thylakoids of the mature chloroplast in the light, the plastid membranes pass through several conformations. First they form flattened sheets, then a regular array of interconnected tubules (the P L B ) , then perforated sheets, and later separated (stroma) or tightly appressed (grana) thylakoids. Scriven (1976) recently produced a three-dimensional analysis of a hypothetical two-phase system in which both phases are continuous, which may arise in oil-water mixtures and which is remarkably similar to models of the PLB. The configuration of the PLB membrane resembles that of a cubic array of spheres. This spherical configuration may arise because an inner surface membrane component is attracted to a central focus (a ribosome? Gunning, 1965), or because some outer surface component must be sequestered away from the etioplast stroma. Hewitt (1977)has suggested that binding of polyvalent ligands produce curved membrane surfaces. Certainly, during development in darkness membranes having a flat profile are replaced by membranes having a curved profile. On illumination this situation is reversed. The subsequent replacement of free membrane sheets

178

TIMOTHY TREFFRY

with thylakoids, which are tightly appressed over a large portion of their surface (grana formation), must also be the result of a change in the membrane surface (Section III,B,3). The development by Albertsson (1970) of analytical procedures sensitive to variations in the surface properties of cell membranes permits these apparent changes in surface properties to be identified. The method is based on the fact that aqueous solutions of poly(ethy1ene glycol) (PEG) and dextran beyond a critical concentration (ca. 5% wlw) form two-phase systems in which samples can be repeatedly partitioned by a countercurrent distribution system in which the PEGrich upper layer forms the mobile phase and the dextran-rich lower layer the stationary phase. The distribution of any polyelectrolyte (including membranes) in the countercurrent distribution train is determined by its intrinsic affinity for the phases. If a potential difference exists between the phases, the net charge on the polyelectrolyte also affects the partition. We have examined developmental changes in the surface properties of etiochloroplast membranes using the polymer two-phase system of Larsson et al. (1971). When membrane vesicles prepared from etioplasts of peas or oats are investigated in this way, they are found to be of two main classes. About half, having a lower concentration of protochlorophyllide, have a higher affinity for the lower phase. The remainder, with a higher protochlorophyllide level, have a higher affinity for the upper phase. The latter show increased affinity for the PEG phase immediately after illumination. This shift is delayed at 0°C (Treffry, 1974), or after prolonged etiolation (Treffry, 1975). During greening this shift is reversed, and membranes show an increasing affinity for the lower phase. This change, although not apparent at 30 minutes, occurs in membranes from leaves illuminated for 2 hours at room temperature or for 30 minutes at room temperature followed by 90 minutes in darkness at 0°C. These results are obtained with phase systems having a net negative charge in the lower phase. When the ionic composition of the phase system is modified to eliminate the interfacial potential, the distribution is changed and it can be concluded that the apparent higher affinity for the upper phase in some etioplast membranes is largely due to their net negative charge. This increases initially after illumination (the cold-suppressed first shift) but later decreases (the temperature-independent second shift). Changes in hydrophobicity are less marked. A small proportion of etioplast membranes is more hydrophobic than the remainder, but this characteristic is lost progressively in all developmental stages examined.

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

179

D. DEVELOPMENTAL CHANGESIN ETIOPLAST MEMBRANEPROTEINS Anderson ( l975b) has summarized current knowledge of chloroplast membrane proteins and proposed a partial structural model in terms of the fluid mosaic concept (Singer and Nicolson, 1972).A fairly complete picture of the peptide composition of chloroplast membranes has emerged (Anderson and Levine, 1974; Henriques et al., 1975), but the developmental situation has been less well studied. Cobb and Wellburn ( 1973) identified and attempted to characterize and quantify etiochloroplast peptides following the illumination of etiolated Avena. All the peptides found in the dark were also present after 24 hours of illumination, but the proportion of those of higher molecular weight increased markedly. Major changes, including the appearance of peptides not found initially, occurred only during the latter half of the illumination period. These investigators emphasize that, during the early period, when major structural changes are occurring, peptide composition is remarkably constant. Unfortunately, their data refer to total plastid peptides, that is, peptides in the stroma and membrane, and, insofar as the proportion of intact etiochloroplasts in their preparations almost certainly varies with developmental stage, these results are difficult to interpret. In a cleverly designed experiment to investigate light-induced synthesis of membrane peptides in the early stages following illumination of etiolated maize, Kaveh and Hare1 (1973) supplied leucine-14C to leaflets which were to be illuminated and l e ~ c i n e - ~to H leaflets which were to remain in the dark. Triton-solubilized lamellae of plastids isolated from leaves freeze-dried after equal periods in light and dark were combined, and the 14CPH ratios in peptides separated by Sephadex chromatography examined. Only two peptides showed substantial light induction which occurred after 1 hour of illumination but not after 0, 3, or 6 hours of illumination. Cobb and Wellburn (1974) report that a transient increase in protein content occurs in both the envelope and stroma lamellae of etioplasts from dark-grown oats illuminated for 30 minutes. This does not take place in isolated etioplasts, and it is suggested that the increase occurring only in vivo may be due to an influx of protein from the cytoplasm. In later work (Cobb and Wellbum, 1976) protein synthesis, rather than protein content, was considered in a series of pulse-chase experiments in which m e t h i ~ n i n e - ~was ~ s supplied to etiolated leaflets in

180

TIMOTHY TREFFRY

the dark and its subsequent distribution in etioplast fractions compared after further treatment in light or dark with nonradioactive methionine. A light-stimulated transient incorporation occurs in the thylakoid fraction (PLBs?) at 30 minutes, in the supernatent (stroma?) at 1 hour, and in the envelopes at 2 hours. Only in the thylakoid fraction did incorporation continue subsequently. A component of the increase found in the envelopes at 2 hours appears to be the small unit of fraction I protein. In these experiments, however, it is difficult to interpret the apparent changes in the intraplastid distribution of radioactivity, as the proportion of label outside the etioplasts at each stage is unknown. Nielsen (1975) compared membrane peptides by a nondetergent method in etiolated, mutant, and normal barley. Peptides of 100,000, 34,000,41,000,39,000,70,000,and 63,000 molecular weight appeared sequentially during illumination. The time of their appearance was very sensitive to the period of etiolation and was substantially delayed in 7-day-old seedlings in comparison with 6-day-old seedlings. Such sensitivity makes the work of different laboratories extremely difficult to compare. OF CHLOROPHYLL-PROTEIN COMPLEXES E. FORMATION

1. Problems of Membrane Solubilization and the Reality of Chlorophyll-Protein Complexes A considerable amount of effort has in recent years been devoted to the examination of chlorophyll-protein complexes resulting from the solubilization of chloroplast membranes with detergent. In the interpretation of such work too little attention has been paid to the complexity of the solubilization process. This is now being examined in some detail for other membrane systems, with the use of a variety of detergents (Helenius and Simons, 1975; Helenius et al., 1976; Egan et al., 1976), but has yet to be considered in chloroplast studies. The relative concentrations of detergent and membrane are critically important. This is commonly defined as moles of detergent per kilogram of lipid but, as membranes contain a considerable quantity of hydrophobic proteins which bind detergent molecules, the concentration of these proteins is also presumably important although unspecified. When workers using chloroplast membranes define solubilization conditions, detergent/chlorophyll ratios are usually quoted. This is not satisfactory in the developmental situation, in which chlorophyll content does not bear a constant relationship to the amount of membrane present.

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

18 1

As increasing levels of detergents are added to membrane suspensions, (1) binding, (2) lysis, (3) solubilization yielding lipoproteindetergent complexes, and finally (4)formation of lipid detergent micelles and delipidation of membrane proteins occur. The treatment of chloroplast membranes with detergent can yield proteins, chlorophyll-protein complexes, and protein-free chlorophyll. Two major chlorophyll-protein complexes are recognized. One (complex I) has a molecular weight of 110,000 and contains chlorophyll a and P,,, [the reaction center chlorophyll for photosystem I (PS I)]; the other, with a molecular weight of 32,000, contains chlorophylls a and b and was associated with photosystem I1 (PS 11)in early work. It is now believed to be a light-harvesting complex. For reviews of this work see Thornber (1975) and Anderson (1975a). A major problem in these studies is whether the chlorophyll-protein complexes that have been isolated actually exist in vivo. This has not been firmly established, and there are some indications that they do not fully reflect the in vivo situation. Biomembranes are an arrangement of amphipathic molecules produced in an aqueous environment. Hydrophobic regions are driven together as they are excluded from the aqueous phase whose structure they would otherwise perturb (Tanford, 1973). The addition of detergent micelles to this system introduces a further hydrophobic phase. The insertion of detergent monomers into the membrane disrupts it in the manner described earlier, and relationships established originally may not persist. Certainly, chlorophylls can be associated with membrane proteins in vivo (Anderson, 1975b) but, insofar as the conformation of membrane proteins is altered by detergent, this alters the binding relationships. Moreover, lipids and chlorophylls must be partitioned between hydrophobic regions of the detergent-protein complex and detergent micelles. The resulting partition presumably depends on the concentration and nature of the components of the system. It seems unlikely therefore that relationships established in vivo would persist unchanged. Scott and Gregory (1975)compared the circular dichroic (CD) spectra of chloroplast membranes with those of membrane fractions and complex I and complex 11. They showed that a combination of spectra from detergent-solubilized fractions cannot provide all the characteristics of the CD spectrum of the native membranes, indicating that the arrangement of chlorophylls in the detergent-solubilized complexes is modified. Only for bacteriochlorophyll (which is water-soluble) has the structure of a chlorophyll-protein complex been established. Here the

182

TIMOTHY TREFFRY

chlorophyll molecules are confined entirely within a globular protein (Fenna and Mathews, 1975) and do not conform to the peripheral arrangement proposed by Anderson (1975b) for chlorophyll in chlorophyll-protein complexes in higher plants. The recent theoretical treatment of Colbow and Danyluk (1976) shows that energy transfer in photosynthesis can be adequately explained in terms of current evidence on the basis of a model in which specific chlorophyll complexes are not required. 2. Formation of Chlorophyll-Protein Complexes in Etiochloroplasts As described in Section III,E,l, the extent to which membranes are broken down and the degree of delipidation of membrane proteins are critically dependent on the detergent/lipid (hydrophobic site) ratio. This is extremely difficult to define in the developmental situation in which the composition and amount of membrane are changing. In etiolated maize, Guignery et al. (1974) observed that at least some protochlorophyllide is bound to membrane peptides (with molecular weights of 21,000 and 29,000) after sodium dodecyl sulfate (SDS) solubilization and electrophoresis, but most of the pigment appears to be peptide-free after detergent treatment. After 48 hours of illumination some chlorophyll is also associated with peptides of molecular weight 25,000 and 70,000, but some remains bound to the 21,000- and 29,000-molecular-weight peptides. These workers suggest, although the evidence for this is unclear, that the latter chlorophyll arises from protochlorophyllide formed in the dark, while the former is synthesized de novo in the light. The appearance of the chlorophyll-protein complexes described by Thornber has also been investigated in the greening situation. In an examination of the jack bean, Alberte et al. (1972) report that complex I1 appears after 2 hours of illumination and complex I after 6 hours of illumination. Hiller et al. (1973) showed that beans grown under intermittent illumination, which inhibits grana formation, show PS I1 activity but lack complex 11. As complex I1 is no longer associated with PS I1 (Thornber and Highkin, 1974), this is not surprising. They also report that complex I1 appears if the dark period between light flashes is shortened, but that there is little change in peptide composition, indicating that complex I1 is formed by the addition of chlorophyll to a preexisting peptide. This conclusion is qualified in a later article (Genge et al., 1974). Scott and Gregory (1975) characterized the CD spectra of complex I, complex 11, and “free pigment” in mature chloroplasts. In greening maize dichroism of these three types can be

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

183

found after 5 hours of illumination (Demeter et al., 1976). Beyond this period the proportion of the signal associated with free pigment increases. The intense positive component at 684 nm found in mature chloroplasts only appeared at 10 hours.

F. BIOSYNTHESISOF LIPIDS 1. In Chloroplasts In an investigation of the incorporation of acetate-l-14Cinto lipid by isolated barley chloroplasts (Hawke and Stumpf, 1965), only 0.5% of the label was incorporated. In later work at the same laboratory (Appleqvist et al., 1968) 16% incorporation was achieved, and the extent of incorporation correlated with the proportion of intact (class I, Spencer and Unt, 1965) chloroplasts in the preparation. This relationship was confirmed (Stumpf and Boardman, 1970) when preparations of spinach chloroplasts were used in which 90% (30% after incubation) were intact and up to 60% of the acetate supplied was incorporated into lipid. Maximal incorporation required CoA, magnesium, carbon dioxide, NADH, NADPH, phosphate, high-intensity illumination, and a trace of Triton X-100. The detergent was not sufficient to disrupt the membranes but apparently increased their permeability to cofactors. The requirement for light could be replaced by ATP. A higher proportion of unsaturated fatty acids was produced when detergent was omitted. In studies of chloroplast development, Leech and her co-workers made considerable use of maize seedlings (Leech et al., 1972, 1973). In common with other grasses, cell division in maize leaves occurs at the leaf base, and examination of a developing leaf from base to apex reveals a developmental sequence from proplastids, to developing and dividing plastids, to mature chloroplasts. In work with leaf pieces, not isolated chloroplasts (Hawke et al., 1974a), it was found that in the lower third of the leaf, where plastids are markedly immature in terms of ultrastructural development, acetate was incorporated mainly into saturated fatty acids with 20 or more carbon atoms. This incorporation was largely independent of light intensity. In the upper two-thirds and especially near the tip of the leaf, there was more incorporation into the more typical chloroplast fatty acids 16 :0,18: 1, and 18:2. This incorporation was strongly dependent on light intensity. In these experiments only acetate, bicarbonate, and phosphate were supplied, so that photosynthetic competence was required for reducing equivalents and ATP. Similar experiments with plastids isolated from particular regions of

184

TIMOTHY TREFFRY

the maize leaf (Hawke et al., 1974b) showed the greatest incorporation of acetate, particularly into monoenoic fatty acids, where the synthesis of grana was most rapid (6-8 cm from the base of a 12-cm leaf). No marked qualitative changes in lipid composition were found during chloroplast development (Leese and Leech, 1976), and the only major quantitative change, an increase in phosphatidyl glycerol, was greatest when grana were being formed and paralleled chlorophyll synthesis. 2 . I n Etiochloroplasts

In a study of the lipid composition of etiolated pea seedlings after various periods of illumination, Tremolieres and Lepage (1971)found that mono- and digalactosyl diglycerides increased over 72 hours at the expense of phospholipid. By far the most abundant phospholipids at all stages were phosphatidylcholines in which linoleic and linolenic acids predominated. Kannangara et al. (1971), using “consistently good” (the proportion of intact plastids was not quantified) preparations of etiochloroplasts from barley, found only light-stimulated incorporation of acetate into lipid when etiolated leaves were illuminated for at least 6 hours before plastid isolation. It was noted that, although light-stimulated acetate incorporation increased substantially as the period of preillumination was increased, acetyl CoA carboxylase activity showed a considerable decline. The incubation medium included ATP, CoA, and bicarbonate but, as in the developing nonetiolated maize leaf (Hawke et al., 1974a),photosynthetic competence was presumably required to provide reducing equivalents. The importance of photosynthesis in lipid biosynthesis in etiochloroplasts was confirmed by Panter and Boardman (1973), who recorded light-stimulated lipid synthesis in etiochloroplasts isolated from etiolated peas after only 2 hours of illumination. This was inhibited by dichlorophenylmethyl urea (DCMU), which inhibits photosynthetic electron transport. Hill activity (see Section IV) is not apparent in this system until leaves have been illuminated for at least 3 hours and, although cytochromes associated with PS I are active after 30 minutes, NADP reduction cannot be measured until 6 hours. These workers provide data suggesting that a capacity for noncyclic phosphorylation is required to support lipid biosynthesis in etiochloroplasts. They admit that NADPH may also be required, as the amount needed to support the rates of synthesis obtained were well below the sensitivity of the assay system used and may have been produced photosynthetically after 2 hours of illumination.

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

185

IV. Development of Photochemical Activity When studying the development of photochemical activity, it is important to consider what is actually being measured. A generally accepted scheme for photosynthetic reactions can be represented as: Electron transport

$0, Hv ADP ATP hv

t.lechoii

Carbon fixation CO, + ATP

How

n

-

ATP

+ NADPH

ADP C[HOH]

Thus a measurement of oxygen evolution coupled to the reduction of NADP is a measure of the joint activity of both photosystems. PS I1 can be measured either by oxygen evolution in the presence of an artificial electron acceptor receiving electrons from the intersystem electron transport chain (commonly oxidized indolphenols or ferricyanide), or by ignoring oxygen evolution and measuring the reduction of the electron acceptor. Diphenyl carbizide (DPC) appears to bypass water as an electron donor to PS I1 and may indicate PS I1 activity where there is no capacity for oxygen evolution. Electron flow through PS I1 is inhibited by DCMU and is often referred to as Hill activity. PS I activity is commonly measured by providing a reduced electron donor N,N,N’,N’-tetramethyl-p-phenylenediamine dihydrochloride and a system for its regeneration [e.g., (TMPD)-ascorbate] and a high-potential electron acceptor, methyl viologen. The latter is autooxidized. Thus PS I activity can be measured in terms of oxygen consumption. DCMU is added to prevent electron flow from PS 11. PS I activity can also be measured b y estimating NADP reduction in the presence of TMPD and ascorbate, but this requires the electron transport chain PS I + NADP to be intact. Methods of assessing photosynthetic electron transport are detailed by Trebst (1972) and Bearden and Malkin (1975). Evidence is beginning to accumulate (Bishop, 1974) that the popular Hill oxidants DCIP and ferricyanide are primarily reduced by PS I rather than PS I1 and that such reduction coupled to oxygen evolution is a measure, not of PS I1 alone, but of joint PS-11-PS-I activity.

186

TIMOTHY TREFFRY

Activity of the photosystems can be assessed by measurements of photophosphorylation (Plesnicar and Bendall, 1973). Cyclic photophosphorylation or, more often, phenazine methosulfate-mediated pseudocyclic photophosphorylation, is a PS I activity. Noncyclic (DCMU-sensitive)photophosphorylation is indicative of linked PS-IIPS-I activity. Even in mature chloroplasts, however, the interpretation of phosphorylation experiments is controversial, especially where artificial redox systems are involved, and these difficulties are enhanced in studies of etiochloroplasts. Carbon dioxide fixation can also be measured in greening leaves (Luettge et al., 1974), but again, if leaves or leaf slices are used, difficulties in interpretation arise not only in separating real and apparent fixation and respiration but also in estimating photoinduced respiration. Photosynthetic electron transport, in common with electron transport in oxidative systems, is associated with the generation of proton and ion gradients. Attempts have also been made to measure these gradients in developing systems (Luettge et al., 1974). The presence of cytochromes and P700 (a form of chlorophyll a acting as the primary electron acceptor in PS I) can be assessed by comparing the oxidized and reduced spectra of suitable preparations. The presence of a component detected in this way does not necessarily indicate that it plays an active role in electron transport. The Hill-Bendall model of electron flow in photosynthesis has provided an extremely useful basis for conceiving and interpreting experiments but, insofar as it indicates a rigid separation of the photosystems and a strictly sequential movement of electrons between components, it has perhaps outlived its usefulness. Variations have been suggested (Boehme and Trebst, 1969; Arnon et al., 1971) but have not gained widespread support. More recently, however, concepts of a single light-harvesting complex serving both photosystems (Seeley, 1973) with multiple reaction centers linked to electron carrier pools (Boardman et al., 1975; Haehnel, 1976), and the suggestion that the direction of electron flow may vary with the redox state of the components (Heber et al., 1976), have introduced a much more flexible and dynamic view which may stimulate and modify research in photosynthesis as fundamentally as the fluid mosaic model has altered the conception and interpretation of biomembrane studies.

A. COMPONENTS OF THE PHOTOCHEMICAL &PARATUS Most of the components of the photosynthetic electron transport chain appear to be present in the etioplast.

BIOGENESIS O F THE PHOTOCHEMICAL APPARATUS

187

Etioplasts contain cytochrome f, cytochrome B,, and the low- (but not the high-) potential form of cytochrome 559 (Whatley et al., 1972; Phung Nhu Hung et al., 1972; Plesnicar and Bendall, 1973; Henningsen and Boardman, 1973; Strasser and Cox, 1974). High-potential cytochrome 559 appears to be derived from some of the low-potential form following illumination (Plesnicar and Bendall, 1973). Plastoquinone production may be stimulated by light, but substantial amounts can be present in the etioplast (Lichtenthaler and Becker, 1972). Plastocyanin (Plesnicar and Bendal, 1973),ferridoxin, and ferridoxin-NADP reductase are also present (Whatley et al., 1972). Etioplasts lack chlorophyll a (and therefore P,oo)and chlorophyll b but contain abundant carotenoids. In considering the properties of etioplasts it should be remembered that almost none of this work was performed on material grown in total darkness. Manipulations are usually carried out in dim green light, and Strasser and Sironval(l973) showed that low-intensity green light may have pronounced developmental effects (see Section IV,B,l).

B. DEVELOPMENT OF THE PHOTOSYSTEMS Bishop (1974)and Arntzen and Briantis (1975) have provided recent reviews of the considerable amount of research carried out on the sequence of appearance of photochemical activities when etiolated seedlings are illuminated.

1. Development under Continuous Illumination There is wide agreement that, when etiolated seedlings are illuminated, photochemical activities associated with PS I are detectible before those associated with PS I1 (Arntzen and Briantis, 1975). PS-I-mediated electron transport from ascorbate to methyl viologen can be measured in etiochloroplasts isolated from etiolated barley leaves illuminated for 5 minutes (Egneus et al., 1972) or 15 minutes (Henningsen and Boardman, 1973).Photooxidation of cytochrome f i n the presence of DCMU, an index of PS I activity, can be demonstrated after 3 minutes of illumination followed by a 2.5-hour dark period (Hiller and Boardman, 1972). The action spectrum for this effect (Hiller and Boardman, 1971) shows a maximum at 682 nm with no shoulder at 700 nm. Hiller and Boardman (1971) also reported that cytochrome f was not oxidized by 703-nm light after 90 minutes of preillumination (when 675-nm light was effective). After 2Y2 hours of illumination, however, light at 703 nm was more effective than light at 675 nm. These findings support the suggestion (Whatley et al., 1972) that P,,, does not

~

188

TIMOTHY TREFFRY

become active in PS I until later in development. Alberte et al. (1973) detected P700in greening jack beans only after 6 hours of illumination when the chlorophyll/P700ratio was 35 and 2% of the chlorophyll present was associated with complex I (Sectio? 111,EJ). The earliest reported occurrence of PS I1 activity (oxygen evolution from intact leaves) takes place after 30 minutes of illumination of etiolated barley (Henningsen and Boardman, 1973), but the fact that DCMU is required in the demonstration of cytochrome-f oxidation in leaves illuminated for 3 minutes and kept for 2Yz hours in darkness indicates that electron transport from PS I1 occurs under these conditions. A combination of illumination and time, in excess of that needed for chlorophyll formation, seems to be required for the development of PS 11. Low intensities (Ogawa and Shibata, 1973) or light flashes (see Section IV,B,2) produce plants that show only PS I activity. An effect of light quality on the development of Hill activity has also been reported (Harnischfeger et al., 1974). Plastids from wheat seedlings grown under blue light showed twice as much Hill activity, per milligram of chlorophyll, as those from leaves grown under red light. Light intensities were chosen that produced similar amounts of chlorophyll. 2. Development under lntermittent Illumination When etiolated seedlings are exposed to intermittent flashes of light, in addition to the structural effects described earlier (Section 11,B73),etiochloroplasts are produced that show PS I activity but not oxygen evolution (Sironval et al., 1969). They, however, can oxidize DPC (Remy, 1973),indicating that the reaction center of PS I is active. Intermittent light appears to produce plastids in which the reaction center of PS I1 is intact but in which water cannot act as an electron donor. Such plastids lack grana and complex I1 (Argyroudi-Akoyunoglou et al., 1971) but apparently fix carbon dioxide (Akoyunoglou and Argyroudi-Akoyunoglou, 1972) by the C-4 pathway (Hatch and Slack, 1970). In continuous illumination, or if the flashes are longer or more frequent, normal structure, oxygen evolution, and carbon dioxide fixation by the C-3 pathway develop. The rapid completion of PS I1 under these circumstances, which may precede observable structural changes, has been dramatically demonstrated by Strasser and Sironval(l972). Phaseolus vulgaris was grown under a flash-illumination regime (1 msec of light plus 15 minutes of dark) for 14 days and showed no oxygen evolution. Leaf disks

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

189

were then exposed to light while in contact with an oxygen electrode. Disks from leaves that had received 700 flashes began to show net oxygen evolution after 3 minutes. Leaves that had received only 100 flashes responded more slowly, while disks from normal green leaves evolved oxygen immediately. When the same technique was used in a later experiment (Strasser and Sironval, 1973), it was found that, if flashed leaves were exposed to low-intensity (100 ergs cm-* sec-I) green light for 8 minutes before testing, the response time was considerably shortened. Remy (1973) showed that in etiolated wheat seedlings oxygen evolution is established in leaves exposed to a l-msec flash every 5 seconds (following the 1 msec per 15 minutes growth conditions). Half the maximum response was obtained after 15 flashes.

c.

DEVELOPMENT OF OTHER PHOTOCHEMICAL ACTIVITIES

Cyclic photophosphorylation (with PMS) was detected in etiochloroplasts of barley illuminated only after isolation (Plesnicar and Bendall, 1972) or isolated “after only a brief period of illumination” (Plesnicar and Bendall, 1973). Noncyclic photophosphorylation was apparent only after leaves had been illuminated for 2?h hours. Carbon dioxide fixation has been observed in etiolated barley after 1-2 hours of illumination (Egneus et al., 1972). However, substantial dark carbon dioxide fixation also occurs in the etiolated barley leaf (Robertson and Laetsch, 1974),and fixation is stimulated by light only after 2 hours of illumination. Luettge et al. (1971) have pointed out that in the early stages of greening the ratio of carbon dioxide fixation to oxygen evolution (in leaf slices) is 0.15 and approaches 1.0 only as the plants become fully green. They suggest that this may indicate (as claimed by Alberte et al., 1972) that PS I1 develops before PS I, but their evidence seems rather indirect. Tamas et al. (1970) showed that the dark fixation of carbon dioxide in etiolated barley is into malate, aspartate, and glutamate. This is also the case after 2-4 hours of illumination but shows substantial light dependence. After 6 hours of illumination fixation is predominantly into Calvin cycle intermediates. Light intensity and temperature play a part in the development of a capacity for carbon dioxide fixation. Etiolated wheat reached compensation (where fixation balances respiration) after 3’/2hours of illumination at 8000 ft-c and 25”C, but required longer time periods at lower light intensities and/or temperatures (Wolf, 1971).

190

TIMOTHY TREFFRY

V. Concluding Remarks The most significant recent change in our understanding of chloroplast development has occurred, perhaps surprisingly, as a result of studies of chlorophyll biosynthesis. ALA for chlorophyll synthesis is now believed to arise not from succinate and glycine (as in heme biosynthesis), but from a four-carbon precursor, most probably a-oxoglutarate. The details of this pathway are now being investigated, and a new approach to the subsequent stages (the use of LA as an inhibitor of ALA dehydratase) is permitting better understanding of the control of protochlorophyllide synthesis. The considerable body of work on spectral shifts during the terminal stages of chlorophyll synthesis remains largely devoid of any detailed interpretation at the molecular level. Twenty years after its discovery even the Shibata shift is not fully understood, but recent work on the cofactor requirements for protochlorophyllide activation and kinetic and chemical analysis of the changes occurring immediately after illumination is promising. Work on detergent-solubilized membranes and the resulting chlorophyll-protein complexes lacks a sound theoretical base and, especially with regard to the developing chloroplast, is still largely uninterpretable. It is scarcely surprising that active protein and lipid synthesis occurs during chloroplast biogenesis and that it occurs primarily during a period when membrane proliferation is obvious at the ultrastructural level. More significantly however, the onset of photochemical activity in etiolated leaves following illumination precedes any significant protein or lipid synthesis. Studies on the development of photochemical activity have been somewhat clouded by changing views on the site of action of commonly used redox compounds, but it is reasonable to conclude that in young etiolated barley seedlings, illuminated under conditions of high humidity, photochemical activity can be detected very rapidly. Cyclic photophosphorylation can be measured more or less instantly. Other photochemical activities associated with PS I appear after a few minutes, and PS I1 (oxygen evolution) after 30 minutes. The only other changes occurring with similar rapidity are the photoreduction of protochlorophyllide, phytolation of at least some of the chlorophyllide formed, and some unspecified changes in the immediate environment of the chlorin ring, as indicated by the spectral shifts and energy transfer phenomena observed during this period. Ultrastruc-

BIOGENESIS OF THE PHOTOCHEMICAL APPARATUS

191

tural changes other than PLB transformation have not been firmly established during this time. Etioplasts appear to contain all the components necessary for photosynthetic electron transport except chlorophyll. All that may be required on illumination to produce at least a functional PS I is completion of the terminal steps of chlorophyll synthesis. This can perhaps be inferred from the fact that among gymnosperms, which can synthesize chlorophyll in the dark, seedlings of pine show PS I activity immediately on illumination (Oku et d.,1974), and in spruce both photosystems become active immediately (Oku et al., 1975). Experiments indicating that photosystems in higher plants did not become active for several hours after illumination lent support to the idea of a multistep assembly process. More recent indications of a very rapid onset of photochemical activity do not necessarily invalidate this concept. Recovery from etiolation may not be analogous to a normal course of development. It may well be that active or potentially active photosystems exist in the embryonic leaves of developing seeds. Certainly these are green in beans and peas. Chlorophyll appears to be lost during seed maturation. If the etiolated seedling contains a latent photosynthetic apparatus, in studying biogenesis we need to distinguish between the completion of preexisting systems and development de nooo. This remains to be achieved. ACKNOWLEDGMENTS Many workers have been most generous in providing reprints, preprints, and unpublished data; some have helped in the preparation of the text; to all I am most grateful.

REFERENCES Akoyunoglou, G., and Michalopoulos, G . (1971).Physiol. Plant. 25,324. Akoyunoglou, G., and Argyroudi-Akoyunoglou, J. H. (1972). Proc. Int. Congr. Photosynth., 2nd, 1971 p. 2427. Akoyunoglou, G., and Argyroudi-Akoyunoglou, J. (1974).FEBS Lett. 42, 135. Alberte, R. S.,Thornber, J. P., and Naylor, A. W.(1972).J. Erp. Bot. 23, 1060. Alberte, R. S.,Thornber, J. P., and Naylor, A. W.(1973).Proc. Natl. Acad. Sci. U S A . 70,

134.

Albertsson, P. A. (1970).Ado. Protein Chem. 24,309. Anderson, J . M. (1975a).Biochim. Biophys. Acta 416, 191. Anderson, J. M. (1975b).Nature (London) 253,536. Anderson, J. M.,and Boardman, N. K. (1966).Biochim. Biophys. Acta 112,403. Anderson, J. M., and Levine, R. P. (1974).Biochim. Biophys. Acta 333,378. Appelqvist, L. A,, Stumpf, P. K., and von Wettstein, D. (1968).Plant Physiol. 43, 163.

192

TIMOTHY TREFFRY

Argyroudi-Akoyunoglou, J . H., and Akoyunoglou, G. (1970).Plani Physiol. 46,247. Argyroudi-Akoyunoglou, J. H., Feleki, Z., and Akoyunoglou, G. (1971). Biochem. Biophys. Res. Commun. 45,606. Arnon, D. I., Knaff, D. B., McSwain, B. D., Chain, R. K., and Tsujimoto, H. Y. (1971). Photochem. Photobiol. 14,397. Arntzen C. J., and Briantis, J . M: (1975).In “Bioenergetics of Photosynthesis” (Govindjee, ed.), p. 51. Academic Press, New York. Bacon, M. F., and Holden, M. (1970). Phytochemistry 9, 115. Bauer, S.. and Siegelman, H. W. (1972). FEBS Lett. 2 4 3 5 2 . Beale, S . I. (1971).Plant Physiol. 48,316. Beale, S. I., and Castelfranco, P. A. (1974). Plant Physiol. 53,297. Beale, S. I., Gough, S. P., and Granick, S. (1975). Proc. Natl. Acad. Sci. U.S.A.72,2719. Bearden, A. J., and Malkin, R. (1975). Q. Rev. Biophys. 7, 131. Becker, A. F. (1974). Photosynthetica 8,235. Bishop, D. G. (1974). Photochem. Photobiol. 20,281. Boardman, N. K. (1966). I n “The Chlorophylls” (L. Vernon and G. R. Seeley, eds.), p. 437. Academic Press, New York. Boardman, N. K. (1967).In “Harvesting the Sun: Photosynthesis in Plant Life” (A. San Pietro, F. A. Greer, and T. J. Army, eds.), p. 211. Academic Press, New York. Boardman, N. K., and Anderson, J. M. (1964).Aust. J . Biol. Sci. 17,86. Boardman, N. K., Bjorkman, O., Anderson, J. M., Goodchild, D. J., and Thorne, S. W. (1975). Proc. Znt. Congr. Photosynth., 3rd, 1974 p. 1809. Boehme, H. S., and Trebst, A. (1969). Biochim. Biophys. Acta 180, 137. Bogdanovii., M. (1973). Physiol. Plant. 29, 19. Bogorad, L., Laber, L., and Gassman, M. (1968). In “Comparative Biochemistry and Biophysics of Photosynthesis” (K. Shibatu et al., eds.), p. 299. Univ. Park Press, State College, Pennsylvania. Bonner, B. A. (1969).Plant Physiol. 44, 73. Bovey, F., Ogawa, T., and Shibata, K. (1974). Plant Cell Physiol. 15, 1133. Bradbeer, J. W., Gyldenholm, A. O., Ireland, H. M. M., Smith, J. W., Rest, J., and Edge, H. J. W. (1974a).New Phytol. 73,271. Bradbeer, J. W., Gyldenholm, A. O., Smith, J. W., Rest, J., and Edge, H. J. W. (1974b). New Phytol. 73,281. Brodersen, P. (1976).Photosynthetica 10,33. Butler, W. L., and Briggs, W. R. (1966). Biochim. Biophys. Acta 112,45. Chiba, Y., Aiga, I., Idemori, M., Sato, Y., Matsushita, K., and Sasa, T. (1967). Plant Cell Physiol. 8,623. Cobb, A. H., and Wellburn, A. R. (1973).Planta 114, 131. Cobb, A. H., and Wellburn, A. R. (1974). Planta 121, 273. Cobb, A. H., and Wellburn, A. R. (1976). Planta 129, 127. Colbow, K., and Danyluk, R. P. (1976). Biochim. Biophys. Acta 440, 107. Cran, D. G., and Possingham, J. V. (1974).Ann. Bot. (London)38,843. Demeter, S., Mustardy, L., Machowicz, E., and Gregory, R. P. F. (1976). Biochem. J . 156,469. Dowdell, R. J., and Dodge, A. D. (1970). Planta 94,282. Dujardin, E., and Sironval, C. (1970). Photosynthetica 4, 129. Egan, R. W., Jones, M. A., and Lehninger, A. L. (1976).J. B i d . Chem. 251,4442. EgnBus, H., Reftel, S., and SelldBn, G. (1972).Physiol. Plant. 27,48. Ellsworth, R. K. (1972). Photosynthetica 6,276. Ellsworth, R. K., and Hervish, P. V. (1975).Photosythetica 9, 125.

BIOGENESIS O F T H E PHOTOCHEMICAL APPARATUS

193

Ellsworth, R. K., and Hsing, A. S. (1973).Biochim. Biophys. Acta 313, 119. Ellsworth, R. K., and Hsing, A. S. (1974).Photosynthetica 8,228. Ellsworth, R. K., and Lawrence, D. G. (1973).Photosynthetica 7,73. Ellsworth, R. K., and Nowak, C. A. (1973).Photosynthetica 6,32. Ellsworth, R. K., Perkins, H. J., Detwiller, J. P., and Liu, K. (1970).Biochim. Biophys. Acta, 223,275. Ellsworth, R. K., Dullaghan, J. P., and St. Pierre, M. E. (1974).Photosynthetica 8,375. Fenna, R. E., and Mathews, B. W. (1975).Nature (London)258,573. Fluhr, R., and Harel, E. (1975).Phytochemistry 14,2157. Fluhr, R., Harel, E., Klein, S., and Ne’eman, E. (1975).Proc. Znt. Congr. Photosynth., 3rd, 1974 p. 2097. French, C. S., Brown J. S., and Lawrence, M. C. (1972).Plant Physiol. 49,421. Gassman, M. (1973a).Plant Physiol. 51, 139. Gassman, M. (1973b).Plant Physiol. 52, 590. Gassman, M., Granick, S., and Mauzerall, D. (1968).Biochem. Biophys. Res. Commun. 32,295. Genge, S., Pilger, D., and Hiller, R. G. (1974).Biochim. Biophys. Acta 347,22. Godnev, T. N., Rotfarb, R. M., and Akulovich, N. K. (1963).Photochem. Photobiol. 2, 119. Godnev, T. N., Galaktionov, S. G., and Raskin, V. I. (1968).Dokl. Akad. Nauk S S S R 181, 237. Granick, S. (1959).Plant Physiol. 34, xviii. Granick, S. (1967).In “The Biochemistry of Chloroplasts” (T. W. Goodwin, ed.), Vol. 2, p. 401. Academic Press, New York. Granick, S., and Gassman, M. (1970).Plant Physiol. 45,201. Griffiths, W. T. (1975).Biochem. J . 142,623. Guignery, G., Luzatti, A., and Duranton, J. (1974).Planta 115,227. Gunning, B. E. S. (1965).Protoplasm 40, 111. Gunning, B. E. S., and Jagoe, M. P. (1967). I n “The Biochemistry of Chloroplasts” (T. W. Goodwin, ed.), Vol. 2, p. 655. Academic Press, New York. Haehnel, W. (1976).Biochim. Biophys. Acta 423,499. Hampp, R., and Ziegler, H. (1975).Planta 124,255. Harnischfeger, G., Trehame, K., and Feierabend, J. (1974).Plant Sci. Lett. 3, 61. Hatch, M. D., and Slack, C. R. (1970).Annu. Reu. Plant Physiol. 21, 141. Hawke, J. C., and Stumpf, P. K. (1965).J.Biol. Chem. 240,4746. Hawke, J. C., Rumsby, M. G., and Leech, R. M. (1974a).Plant Physiol. 53,555. Hawke, J. C., Rumsby, M. G., and Leech, R. M. (1974b).Phytochemistry 13,403. Heber, U . , Boardman, N. K., and Anderson, J. M. (1976).Biochim. Biophys. Acta 423, 275. Helenius, A., and Simons, K. (1975).Biochim. Biophys. Acta 415,29. Helenius, A., Fries, E., Garoff, H., and Simons, K. (1976).Biochim. Biophys. Acta 436, 319. Henningsen, K. W. (1967).In “The Biochemistry of Chloroplasts” (T. W. Goodwin, ed.), Vol. 2, p. 453. Academic Press, New York. Henningsen, K. W. (1970).J . Cell Sci. 7,587. Henningsen, K. W., and Boardman, N. K. (1973).Plant Physiol. 51, 1117. Henningsen, K. W., and Boynton, J. E. (1970).J.Cell Biol. 44, 290. Henningsen, K. W., and Kahn, A. (1971).Plant Physiol. 47,685. Henningsen, K. W., and Thome, S. W. (1974).Physiol. Plant. 30,82. Henningsen, K. W., Thorne, S. W., and Boardman, N. K. (1974).Plant Physiol. 53,419.

194

TIMOTHY TREFFRY

Henriques, F., Vaughan, W., and Park, R. (1975). Plant Physiol. 55,338. Hewitt, J. A. (1977).J . Theor. Biol. 64,455. Hiller, R. G., and Boardman, N. K. (1971). Biochim. Biophys. Acta 253,449. Hiller, R. G., and Boardman, N. K. (1972).Plant Physiol. 50, 183. Hiller, R. G., Pilger, D., and Genge, S. (1973).Plant Sci. Lett. 1,81. Holden, M. (1961). Biochem. J . 78,359. Holden, M. (1963).Photochem. Photobiol. 2, 175. Horton, P., and Leech, R. M. (1975). Plant Physiol. 56, 113. Izawa, S., and Good, N. E. (1966). Plant Physiol. 41, 554. Jones, C. B., and Ellsworth, R. K. (1969).Plant Physiol. 44, 1478. Kahn, A. (1968). Plant Physiol. 45,1769. Kahn, A., and Nielsen, 0. F. (1974). Biochirn. Biophys. Acta 333,409. Kahn, A., Boardman, N. K., and Thome, S. W. (1970).J. Mol. Biol. 48,85. Kannangara, C. G., Henningsen, K. W., Stumpf, P. K., Appelqvist, C. A., and von Wettstein, D. (1971). Plant Physiol. 48,526. Kaveh, D., and Harel, E. (1973). Plant Physiol. 51, 671. Kirk, J. T. 0. (1970).Ann. Rev. Plant Physiol. 21, 11. Klein, S., and Neuman, J. (1966). Plant Cell Physiol. 7, 115. Klockare, B., and Sundqvist, C. (1977). Photosynthetica (in press). Koshland, D. E. Jr., Nemethy, G., and Filmer, D. (1966).Biochemistry 5, 365. Lafleche, D., Bove, J. M., and Duranton, J. (1972)./. Ultrastruct. Res. 40,205. Larsson, C., Collin, C., and Albertsson, P. A. (1971). Biochim. Biophys. Acta 2A5,425. Leech, R. M. (1976).In “Perspectives in Experimental Biology” (N. Sunderland, ed.), Vol. 2, p. 145. Pergamon Press, New York. Leech, R. M., Rumsby, M. G., Thomson, W. W., Crosby, W., and Wood, P. (1972).Photosynth., Two Centuries Its Discovery Joseph Priestley, Proc. Int. Congr. Photosynth. Res., 2nd, 1971 p. 2479. Leech, R. M., Rumsby, M. G., and Thomson, W. W. (1973).Plant Physiol. 52,240. Leese, B. M., and Leech, R. M. (1976). Plant Physiol. 57,789. Lichtenthaler, H. K., and Becker, K. (1972). Proc. Int. Congr. Photosynth., Znd, 1971 p. 2451. Liljenberg, C. (1974). Physiol. Plant. 32,208. Litvin, F. F., and Belyaeva, 0. B. (1968). Biokhimiya 33,928. Litvin, F. F., and Belyaeva, 0. B. (1971). Photosynthetica 5,200. Luettge, U . , Kramer, D., and Ball, E. (1974). 2. Planzenphysiol. 71,6. McFeeters, R. F., Chichester, C. O., and Whitaker, J. R. (1971). Plant Physiol. 47,609. Meller, E., Belkin, S., and Harel, E. (1975).Phytochernistry 14,2388. Michel-Wolweriz, M. R. and Bronchart, R. (1974). Plant Sci. Lett., 2,45. Miller, K. R., and Staehelin, L. A. (1976).J.Cell Biol., 68,30. Murray, A. E., and Klein, A. E. (1971). Plant Physiol. 41,383. Nadler, K., and Granich, S. (1970). Plant Physiol. 46,240. Nielsen, N. C. (1975).Eur. J . Biochern. 50,611. Ogawa, T., and Shibata, K. (1973). Physiol. Plant. 29, 112. Ogawa, T., Bovey, F., Inoue, Y., and Shibata, K. (1975). Proc. Znt. Congr. Photosynth., 3rd, 1974 p. 1829. Ogura, N. (1972). Plant Cell Physiol. 13, 971. O h , T., Sugahara, K., and Tomita, G. (1974). Plant Cell Physiol. 15, 175. Oku, T., Hayashi, H., and Tomita, G. (1975). Plant Cell Physiol. 16, 101. Panter, R. A., and Boardman, N. K. (1973)./. Lipid Res. 14,664. Phung Nhu Hung, S., Remy, R., and Moyse, A. (1972). Proc. Znt. Congr. Photosynth., 2nd, 1971 p. 2407.

BIOGENESIS OF T H E PHOTOCHEMICAL APPARATUS

195

Plesnicar, M., and Bendall, D. S. (1972). Proc. Int. Congr. Photosynth., 2nd, 1971 P. 2367. Piesnicar, M., and Bendall, D. S. (1973). Biochem. J . 136,803. Possingham, J . V. (1973).J. Erp. Bot. 24, 1247. Possingham, J. V., and Saurer, W. (1969). Planta 86, 186. Rebeiz, C. A., and Castelfranco, P. A. (1973).Annu. Reo. Plant Physiol. 24, 129. Remy, R. (1973). Photochem. Photobiol. 18, 409. Ridley, S. M., and Leech, R. M. (1969). Nature (London)227,463. Robertson, D., and Laetsch, W. M. (1974).Plant Physiol. 54, 148. Rose, R. J., Cran, D. G., and Possingham, J. V. (1975).J . Cell Sci. 17,27. Rosinski, J., and Rosen, W. G. (1972). Q. Reu. Biol. 47, 160. Sane, P. V., Goodchild, D. J., and Park, R. B. (1970).Biochim. Biophys. Acta, 216,162. Schneider, H. A. W. (1966).J.Chromatogr. 21,448. Schneider, H. A. W. (1973).Enzyme 16, 108. Schoper, P., and Siegelman, H. W. (1968). Plant Physiol. 43,990. Scott, B., and Gregory, R. P. F. (1975).Biochem. J . 149,341. Scriven, L. E. (1976). Nature (London)263, 123. Seeley, G. R. (1973).J. Theor. Biol. 40, 189. Shibata, K. (1957).J . Biochem. (Tokyo)44, 147. Shibata, K. (1973). Biochim. Biophys. Acta 304, 249. Shimizu, S., and Tamaki, E. (1962).Bot. Mag. 75,462. Shimizu, S., and Tamaki, E. (1963).Arch. Biochem. Biophys. 102, 152. Shlyk, A. A. (1971).Annu. Reu. Plant Physiol. 22, 169. Singer, S. J., and Nicolson G. L. (1972). Science 175, 720. Sironval, C. (1975). Proc. Int. Congr. Photosynth., 3rd, 1974 p. 2153. Sironval, C., and Brouers, M. (1970). Photosynthetica 4,38. Sironval, C., and Kuyper, Y. (1972). Photosynthetica 6,254. Sironval, C., Brouers, M., Michel, J. M., and Kuyper, Y. (1968).Photosynthetica 2,268. Sironval, C., Michel, J. M., Bronchart, R., and Englert Dujardin, E. (1969).Prog. Photosynth. Res., Proc. Int. Congr. [ l s t ] ,1968 Vol. 1, p. 47. Smith, J. H. C. (1960).In “Comparative Biochemistry of Photoreactive Systems” (M. B. Allen, ed.), p. 257. Academic Press, New York. Spencer, D., and Unt, H. (1965). Aust,J. Biol. Sci. 18,197. Staehelin, L. A. (1975). Biochim. Biophys. Acta 408, 1. Stetler, D. A., and Laetsch, W. M. (1969).Am. J . Bot. 56,260. Strasser, R. J., and Cox, R. P. (1974). Plant Sci. Lett. 3, 19. Strasser, R. J., and Sironval, C. (1972). FEBS Lett. 28, 56. Strasser, R. J., and Sironval, C. (1973). FEBS Lett. 29,286. Stumpf, P. K., and Boardman, N. K. (1970).J. Biol. Chem. 245,2579. Sundqvist, C. (1969). Physiol. Plant. 22, 147. Sundqvist, C., Odenguard, B., and Person, G. (1975).Plant Sci. Lett. 4,89. Sundqvist, C., Bjorn, L. O., and Virgin, H. (1976).Results Probl. Cell Dqjer. 4 (in press). Tamas, I. A., Yemm, E. W., and Bidwell, R. G. S. (1970). Can. J . Bot. 48,2313. Tanford, C . (1973). “The Hydrophobic Effect.” Wiley, New York. Thornber, J. P. (1975).Annu. Reu. Plant Physiol. 26, 127. Thornber, J. P., and Highkin, H. (1974). Eur. J . Biochem. 41, 106. Thorne, S. W. (1971). Biochim. Biophys. Acta 226, 113. Thome, S. W., and Boardman, N. K. (1971).Plant Physiol. 47,252. Thome, S. W., and Boardman, N. K. (1972). Biochim. Biophys. Acta 267, 104. Trebst, A. (1972). In “Methods in Enzymology” (A. San Pietro, ed.), Vol. 24, Part B, p. 146. Academic Press, New York.

196

TIMOTHY TREFFRY

Treffry, T. (1964). Plant Physiol. 39, xxxiii. Treffry, T. (1968). Planta 85,376. Treffry, T. (1970).Planta 91,279. Treffry, T. (1973).J.E x p . Bot. 24, 185. Treffry, T. (1974). Eur. J . Biochem. 43,349. Treffry, T. (1975).Planta 126, 11. Tremolieres, A., and Lepage, M. (1971). Plant Physiol. 47,392. Virgin, H. I. (1973). In “Phytochrome” (K. Mitrakos and W. Shropshire, eds.), p. 371. Academic Press, New York. Virgin, H. I., and French, C. S. (1973). Physiol. Plant. 28,350. von Wettstein, D., Kahn, A., Nielsen, 0. F., and Cough, S. (1974). Science 184,800. Watts, R . B., and Kekwick, R. C. 0. (1974). Arch. Biochem. Biophys. 160,469. Weier, T. E., and Brown, D. L. (1970).Am. J . Bot. 57,267. Wellbum, A. R., (1970). Phytochemistry 9,2311. Wellburn, A. R. (1975a). Phytochemistry 14,699. Wellburn, A. R. (1975b).Phytochemistry 14, 1171. Wellburn, A. R. (1976). Biochem. Physiol. Planz. 169,265. Wellburn, A. R., and Wellburn, F. A: M. (1971). Biochem. Biophys. Res. Commun. 45, 747. Whatley, J. M. (1977). New Phytol. 78,407. Whatley, F. R., Gregory, P., Haslett, B. G., and Bradbeer, J. W. (1972).Proc. Znt. Congr. Photosynth., 2nd, 1971 p. 2375. Wolf, F. T. (1971).Z. Pflanzenphysiol. 64, 124. NOTEADDED IN PROOF The recent work of S. Schoch, U. Lempert, and W. Riidiger [“On the Last Steps of Chlorophyll Biosynthesis. Intermediates between Chlorophyllide and Phytol-containing Chlorophyll.” Z. Pflanzenphysiol. 83,427-435 (1977)l indicates clearly that chlorophyllide is not esterified directly with phytol as has been suggested in this chapter. When etiolated oat seedlings are illuminated the chlorophyllide formed is first esterified with geranyl geraniol, which is subsequently saturated.

Extrusive Organelles in Protists' KLAUS HAUSMANN Lehrstuhl fur Zellenlehre, Universitat Heidelberg, Heidelberg, West Germany I. Introduction . . . . . . , . . A. Definition of Extrusive Organelles . . . B. Historical Background . . . . . . 11. Methods . . . . . . . . . A. Preparation of Material . . . . . . B. Light Microscopy . . . . . . . C. Electron Microscopy . . . . . . 111. Characterization and Distribution of Extrusomes IV. Fine Structure, Extrusion Merhanism, Function, and Origin of the Different Types of Extrusomes . . A. Spindle Trichocysts . . . . . . B. Mucocysts (Including Muciferous Bodies, . . Kinetocysts, and Clathrocysts) . . . . C. Toxicysts (Including Cyrtocysts, Pexicysts, and Haptocysts) . . . . . . . . D. Rhabdocysts . . . . . . . . E. Ejectisomes (Taeniobolocysts) . . . . F. Discobolocysts . . . . . . . G . Nematocysts (Cnidocysts) . . . . . H. Polar Filament of Cnidosporidians . . . I. Varia . . . . . . . . . V. Conclusions . . . . . . . . . References . . . . . . . . .

.

197 197 198 198 198 199 199 199

. .

202 202

.

224

.

243 259 260 263 265 265 267 267 268

. . . .

. . . .

I. Introduction A. DEFINITIONOF EXTRUSIVEORGANELLES Extrusive organelles are membrane-bounded structures of protists, usually located in the cortical cytoplasm of these cells. Although they have different type-specific structures and functions, they all exhibit one general characteristic: They are readily discharged when subjected to a wide range of stimuli (mechanical, electrical, and chemical). During the transition from the resting state to the ejected form, which in several cases takes place within milliseconds, the organelles undergo characteristic morphological changes. In the following dis-

This article is dedicated to Professors R. Hovasse, Fr. Kriiger, and K. E. WohlfarthBottermann.

197

198

KLAUS HAUSMANN

cussion the term “extrusome” is used as a general name for all kinds of extrusive organelles (according to Grell, 1973).

B. HISTORICALBACKGROUND As far as can be determined, the first observation of extruded extrusomes was made in Paramecium by Ellis in 1769; however, these organelles were first described in detail by Allman in 1855. Later investigations were made by Kolsch (1902), Maier (19O3),Schuberg (1905), Mitrophanow (19O5), and Khainsky (1911). In 1914 Tonniges reviewed the literature on extrusomes in an article on the nature of these organelles in the ciliate Frontonia (Tonniges, 1914). Following Saunders’ 1925 studies, Kriiger (1929, 1930, 1931a,b, 1934a,b) initiated his extensive work on extrusomes using the darkfield light microscope; the results are summarized in his 1936 review article. Development of the electron microscope provided new opportunities for studying extrusomes, and since its introduction our knowledge of these organelles has rapidly increased (for reviews, see Hausmann, 1972a; Hovasse, 1965a, 1969; Hovasse and Mignot, 1975). Today many different kinds of extrusomes are known. In some cases electron microscope studies have shown the macromolecular arrangement of filamentous elements making up special parts of several extrusomes. In addition, the extrusion mechanism of some of these organelles has been revealed by electron microscope visualization of different stages of expulsion interrupted in progress. Many remaining unanswered questions are also discussed in this report. 11. Methods

A. PREPARATION OF MATERIAL One of the easiest ways to prepare extrusomes for light microscope examination is to press the protists between a microscope slide and a cover glass. By this procedure the cells are destroyed, and ejected extrusomes become visible. However, a triggering of the extrusion by cover glass pressure is not always possible because of the small size of the cells (e.g., flagellates), or not desirable if special methods are required (e.g., negative staining or shadow casting for electron microscopy). Fumes of formaldehyde or similar solutions also produce extrusion (Kriiger, 1931a, 1934a, 1950). If an increased quantity of inhibited extrusomes is required, one can use a solution of 5% potassium ferrocyanide (Kriiger, 1930), which is added to the cells prior to their destruction or irritation.

EXTRUSIVE ORGANELLES IN PROTISTS

199

For special problems it is possible to induce extrusion of all the extrusomes and still keep the cells alive. Geranium juice is reported to be such a stimulant (Schuster et al., 1967), and the application of a moderate electric shock has also been used in some cases (Hausmann and Allen, 1976; Wohlfarth-Bottermann, 1953; Yusa, 1963, 1965).

B. LIGHT MICROSCOPY It is not necessary to employ special staining methods for light microscope examinations of extrusomes if one makes use of the properties of the phase-contrast microscope, the anoptral-contrast microscope, and the differential-interference-contrast microscope. Kriiger (1936) has reviewed the different light microscope staining methods for extrusomes. C. ELECTRONMICROSCOPY Basically all the conventional electron microscope techniques can be employed for the study of extrusomes, although it is frequently useful to change or modify the concentration of the fixative, the kind of buffer system, or the pH of the solution to prevent an inopportune extrusion. It is known that some buffers trigger the discharge of extrusomes (Allen and Wolf, 1974), and different negative-staining procedures have been reported to influence the fine structure of these organelles (Hausmann, 1973a; Hausmann and Stockem, 1973). For other methods (freeze-fracture, shadow casting, scanning electron microscopy, and so on) see Allen and Hausmann (1976), Bachmann et al. (1972), Hausmann (1974), Jakus and Hall (1M6), Marczalek and Small (1969), Plattner et al. (1973b), and Small and Marczalek (1969).

111. Characterization and Distribution of Extrusomes Extrusomes are listed alphabetically in Table I. Their morphology in the resting and ejected state is described and information on their mode of ejection is provided in the table. In 1936, Chadefaud attempted to use extrusomes as a characteristic for taxonomic purposes. But the presence of these organelles does not seem to be correlated with other characteristics used to classify protists. Kriiier (1936) has pointed out, however, that extrusomes in ciliates are restricted mainly to holotrichs. The distribution of the various types of extrusomes in protists is summarized in Table 11.

KLAUS HAUSMANN

200

TABLE I CHARACTERIZATION OF EXTRUSOMES Name

Resting state

Discobolocyst (Fig. 46)

Spherical body with a diskshaped ring at one pole

Ejectisome (taeniobolocyst) (Fig. 44)

Spirally wound ribbon, usually bipartite

Haptocyst (missile-like body or phialocyst) (Fig. 42)

Bottlelike organelle with a complex internal structure composed of several components Compound organelle consisting of a central element enveloped by a ringlike jacket Saclike vesicle filled with an unordered material Polyhedral paracrystalline body

Kinetocyst (Fig. 29)

Muciferous body (Fig. 26)

Mucocyst (Figs. 20 and 21)

Nematocyst (cnidocyst) (Fig. 48) Rhabdocyst (Fig. 43)

Spindle-shaped capsule with a coiled tube Stick-shaped organelle

Spindle trichocyst

Spindle-shaped or

Ejected state

Mode of ejection

Solid ring with the same dimensions as in the resting state furnished with a long tail Tubelike; smaller and longer than the resting state

Unknown (stretching?)

Partly everted; excretion of poisonous material

Sudden unrolling of the ribbon and formation of a tube by rerolling laterally Unknown

The central element lies in front of the open jacket

Unknown

Amorphous mucilage

Secretion of the material through a pellicular pore

Polyhedral paracrystalline body whose diameter and length are multiples of those of the resting state Capsule with an everted tube

Secretion lasting several seconds; unfolding of a preexisting network of filaments

Tubelike structure with the same length and plus or minus the same diameter as the resting form Thread-shaped

Eversion of the tube Telescopic expulsion

Sudden (lasting

(Continued)

20 1

EXTRUSIVE ORGANELLES IN PROTISTS TABLE I

(Continued)

Name

Resting state

Ejected state

Mode of ejection

(acontobolocyst) (Figs. 6 and 7)

rhomboid paracrystalline body; sometimes furnished with a specially constructed tip

paracrystalline filament with plus or minus the same diameter as the resting state

less than a second) unfolding of a preexisting three-dimensional network of protein filaments

Toxicyst (including pexicyst and cyrtocyst) (Figs. 36 and 37)

Inverted tube inside a capsule

Polar filament (of cnidosporidians)”

4 coiled tube in-

side the spore

Capsule with an everted tube of the same width and length as in the resting state; excretion of poisonous material Everted tube in contact with the spore

Sudden eversion and/or telescopic expulsion of a tube

Evagination of the tube

This structure is not an extrusomesensu stricto, since it is not a membrane-bounded organelle (see Section &A). TABLE I1 EXAMPLESOF THE OCCURRENCE OF EXTRUSOMES IN THE DIFFERENT TAXONOMIC CATEGORIES OF PROTISTS‘ Systematic group

Type of extrusome

Flagellata

Discobolocyst Ejectisome Muciferous body Mucocyst Nematocyst Spindle trichocyst Toxicyst Kinetocyst Haptocyst Mucocyst Rhabdocyst Spindle trichocyst Toxicyst Polar filament

Rhizopoda Sporozoa Ciliata

Cnidosporidia

Example

Ochromonas tuberculatus Chilomonas paramecium Euglena spirogyra Euglena splendens Pol ykrikos schwartzi Oryrrhis marina Colponema loxodes Acanthocystis aculeata Acineta tuberosa Tetrahymena pyrifomis Kentrophoros latum Paramecium caudatum Didinium nasutum Nosema bombycis

Some protists possess several different types of extrusomes; for example, Didinium I.”

-..---.,”+.. ....,I

c-4

-.,..

tr

202

KLAUS HAUSMANN

IV. Fine Structure, Extrusion Mechanism, Function, and Origin of the Different Types of Extrusomes The choice of extrusomes and order in which they are discussed are based on their similarities to each other and on the extent to which they have been investigated. A. SPINDLETRICHOCYSTS 1. Resting Spindle Trichocysts of Ciliates

In ciliates, resting spindle trichocysts are located in the cortex of the cell (Fig. 1).In Paramecium there are about 6000 to 8000 trichocysts in each cell, and they are located at predictable positions within the pellicular system (Fig. 2) (Allen, 1971; Ehret and de Haller, 1963; Ehret and McArdle, 1974; Ehret and Powers, 1959; Grimstone, 1961; Hufnagel, 1969; Jacobson, 1931; Jurand and Selman, 1969; Khainsky, 1911; Klein, 1952; Pitelka, 1963, 1965; Plattner et al., 1973a; Schneider and Wohlfarth-Bottermann, 1964; Schuberg, 1905; Sedar and Porter, 1955; Stewart and Muir, 1963; Wichterman, 1953). In Paramecium the spindle-shaped or rhomboid trichocyst bodies (tb in Fig. 6a), furnished with a specially constructed tip (tt in Fig. 6a), regularly alternate with single or paired basal bodies (Fig. 2a and b). Their tips (tt in Fig. 2c) pass through the space between adjacent alveoli (a in Fig. 2c), where they closely underlie the plasma membrane. In freeze-fracture replicas the P face of the plasma membrane displays regular ringlike aggregates of membrane-intercalated particles located directly above the trichocyst tips (Figs. 3a and b; arrows in Fig. 4a). [In this article the freeze-etching nomenclature proposed by Branton et al. (1975) is used.] These aggregates consist of an outermost single or double ring, 300 nm in diameter (a-type granules), either a concentric middle ring or a diffuse zone of particles, 180 nm in diameter (b-type granules), and finally a central rosette, 80 nm in diameter (c-type granules) (Figs. 3a and b; arrows in Fig. 4a) (Janisch, 1972; Plattner et al., 1973a; Satir, 1974b; Satir and Satir, 1974). The trichocyst membrane is also connected to the alveolar membrane by particles (e-type granules) (e in Figs. 3b and c, and 4b). Plattner et al. (1973a) interpret these particle accumulations to be membrane-to-membrane attachment sites, providing the trichocyst tip with a solid anchorage in the plasma membrane and a connection with the alveoli. This explanation, which is supported by isolation experiments (Anderer and Hausmann, 1977), seems to be plausible, since the expulsion, which takes place within milliseconds (Pitelka, 1963),

EXTRUSIVE ORGANELLES IN PROTISTS

203

FIG. 1. Diagram (a) and electron micrograph (b) of the general organization of P . cuudutum. The resting spindle trichocysts (tr) are located in the cortex of the cell. ma, Macronucleus; mi, micronucleus; cv, contractile vacuole; fv, food vacuole; pe, peristome; ve, vestibulum. (b) x800. (a) After Grell, 1973.

must expose the orifice of the opened trichocyst membrane to extremely high forces which could tear the trichocyst membrane or pull it out of the cell if it were not firmly anchored to the surrounding pellicle (Allen and Hausmann, 1976). According to D. R. Pitelka (personal communication) the function of the particles in the plasma membrane-at least the c-type granulesmay be, in addition to providing attachment sites, to prevent inopportune fusion of the plasma membrane with the trichocyst membranean idea also proposed independently by Bardele (1976a) for the kinetocysts in heliozoan axopods (see Section IV,B,8).

204

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES I N PROTISTS

205

:I .'I :I :I

FIG.3. Diagram of the distribution of membrane-intercalated particles in the cortex of Paramecium after freeze-fracture (a and c), and thin sections (b). In the thin sections the particles are located inside the membranes rather than in between as shown. (b) is drawn in this way to show clearly the position of the particles. a, b, c, and e , Different aggregates of particles associated with the plasma membrane (pm), the trichocyst membrane (tm), and the alveolar membrane (am). al, Alveolus; d, depression; EF-tm, external face of the trichocyst membrane; p, particle; PF-pm, protoplasmic face of the plasma membrane; PF-tm, protoplasmic face of the trichocyst membrane; tb, trichocyst body; tc, tubular collar; tt, trichocyst tip. (c) From Allen and Hausmann, 1976.

However, other workers have postulated that similarly arranged particles in the plasma membrane of Tetruhyrnena, calledfusion rosettes (Fig. 22g), are necessary for fusion of the plasma membrane with the extrusome membrane in this ciliate (Satir et al., 1972, 1973) and may be a general feature of other membrane fusions (Satir, 1975; Satir and Satir, 1974). As indicated by Beisson et al. (1976),the true function of the particles in the plasma membrane could be discovered by examining paraFIG. 2. Electron micrograph (a) and simplified diagrams (b and c) of the pellicular system of P. caudatum. a, Alveolus; ci, cilium; kf, kinetodesmal fiber; pmt, posterior microtubules; tb, trichocyst body; tmt, transverse microtubules; tt, trichocyst tip. (a) x 40,000. (a) From Hausmann and Allen, 1976.

206

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

207

mecia with mutations that affect the trichocysts, such as those reported by Jurand and Saxena (1974), Ruiz et al. (1976), and Pollack (1974). Since the membrane junctions in Paramecium resemble to some extent metazoan gap junctions (for reviews, see McNutt and Weinstein, 1973; Staehelin, 1974), which are generally considered sites of selective permeability for intercellular communication or coupling, Plattner et al. (1975)have analyzed, using the technique of high-resolution freeze-fracture and tracer experiments (microperoxidase, lanthanum, and cytochrome c), the morphology and functional role of the ringlike aggregates of intercalated particles in the plasma membrane ofParamecium. They showed that in none of the techniques did low-molecular-weight electron microscope tracers penetrate the membranes. Therefore the assumption of Satir and Satir (1974), that extrusome discharge can be triggered by an osmotic shift via transmembranous canals, has not yet been confirmed experimentally. However, these results cannot rule out the possibility that the particles operate as active transmembranous carriers of ions, for example, Ca2+, which can then induce exocytosis. [Plattner (1974) induced trichocyst discharge artificially b y using ionophoretic Ca2+ injections.] However, special electron microscope methods for detecting Ca2+ have never shown corresponding electron-dense deposits in the region of the membrane-intercalated particles of type a, b, or c (Plattner and Fuchs, 1975).This is not surprising, since it is likely that exocytosis is triggered from inside rather than from outside the cell. However, Ca2+may play a more prominent role in the actual fusion of the trichocyst membrane with the plasma membrane than in triggering discharge, since it is known to be important in other exocytic processes (Allison and Davies, 1974; Douglas, 1974; Poste and Allison, 1973). Recent localizations of Ca2+-accumulatingsites in Paramecium (Fisher et al., 1976; Plattner and Fuchs, 1975) support this possibility. Except at the distal part of its tip, the trichocyst has a membrane that shows the normal characteristics of biomembranes following freeze-fracture preparation (Figs. 3c and 4b). The P face possesses randomly distributed 10-nm-diameter particles (p in Fig. 3c; PF-tm in FIG.4. (a) In freeze-fracturepreparations of the P face of the plasma membrane (PFpm) particle rings (arrows) between the cilia (arrowheads)indicate the position of the underlying trichocyst tips. (b) The P face of the trichocyst membrane (PF-tm) possesses randomly distributed particles, whereas the E face (EF-tm) shows corresponding depressions. Most of the distal part of the tip is free of particles (bracket) except at the extreme end of the tip where one observes a conglomeration of particles (e).(a and b) x 27,000. (a) From R. D. Allen; (b) from Allen and Hausmann, 1976.

208

KLAUS HAUSMANN

FIG.5. Schematic three-dimensional reconstruction of the detailed structure of part of the trichocyst tip and ensheathing structures (1 to IV), cut away to expose the different layers. tc, Tubular collar. From Bannister, 1972.

Fig. 4b) which are present in a frequency of about 800 particles per square micrometer. The E face shows a corresponding number of depressions (d in Fig. 3c; EF-tm in Fig. 4b). The membrane surrounding the distal third of the trichocyst tip is encased in a tubular collar (tc in Figs. 3b and c, and 5). This external specialization of the trichocyst membrane is complemented by a differentiation in the adjacent intramembranous region; both the P face and the E face are free of particles and depressions and are smooth (tc in Fig. 3c; bracket in Fig. 4b). The particles that were in this area during trichocyst development (Allen and Hausmann, 1976) are now concentrated within the trichocyst membrane at its distal tip (e in Figs. 3c and 4b). Ultrastructurally the undischarged trichocyst consists of several different components (Bannister, 1972; Hausmann et al., 1972b):

1. The crystalline matrix of the trichocyst tip (I in Fig. 5; Fig. 6b) and the body, which reveals a periodic, 7-nm striation (Fig. 6c and d). 2. A meshlike sheath surrounding the body of the organelle, which is best seen in isolated trichocysts (Anderer and Hausmann, 1977) (arrowheads in Fig. 6d).

EXTRUSIVE ORGANELLES I N PROTISTS

209

3. An inner sheath surrounding the tip, consisting of four helically arranged envelopes with a square net substructure (I1 in Fig. 5). 4. An outer sheath around the tip composed of tubular structures (I11 and IV in Fig. 5). 5. A membranous trichocyst sac which has an apical region surrounded by a cylinder of tubular structures joined to each other by dense material, that is, the tubular collar (tc in Figs. 3c and 5).

2 . Discharged Spindle Trichocysts of Ciliates The discharged spindle trichocysts of Paramecium caudatum, which measure 25-35 pm in length and 0.5-0.6 pm in width (Fig. 6e

FIG.6. Resting (a-c) and ejected trichocysts of P. caudaturn (d-h). The trichocyst is a bipartite structure consisting of a tip (tt) and a body (tb). In the electron microscope the tip (b)as well as the body (c and d) of the resting stage show a regular pattern. The ejected form is elongated by a factor of 8 (e).The shaft discloses electron-dense (d) and electron-light (1) cross striations. The striae are due to a regular arrangement of fila. x95,OOO. (c and d ) x80,OOO. (e) ments (points and asterisks) (h). (a and e) ~ 8 0 0(b) ~ 4 5 0 0 (f . and g ) x 110,000. (f) From Hausmann, 1971.

2 10

KLAUS HAUSMANN

and f), are characterized by transverse striations with a period of 55 nm (Fig. 6f and g). Electron-dense (d in Fig. 6g) and electron-lucent (1 in Fig. 6g) bands make up this periodic structure (Bannister, 1972; Beyersdorfer and Dragesco, 1952b; Ehret and McArdle, 1974; Hausmann, l971,1973b,d; Hausmann and Stockem, 1973; Hausmann et al., 1972a; Jakus, 1945; Jakus and Hall, 1946; Knoch and Konig, 1951; Kriiger and Wohlfarth-Bottermann, 1952; Nemetschek et al., 1953; Pease, 1947; Wohlfarth-Bottermann, 1950, 1953; Wohlfarth-Bottermann and Pfefferkorn, 1952). The same or a similar structure can be observed in spindle trichocysts of other species of ciliates (Beyers-

FIG. 7. The various fine-structural features of the discharged trichocyst shaft (a-d) can he explained by a three-dimensional model (a‘-d‘). Depending on the plane of view different patterns can be seen. Note the marks at the bottom of the model. (a-d) X 210,000. From Hausmann et al., 1972a.

EXTRUSIVE ORGANELLES IN PROTISTS

211

dorfer and Dragesco, 1952a; Bretschneider, 1950; Dragesco, 1968; Hausmann, 1973b; Kawakami and Yagiu, 1960; Kriiger et al., 1952; Luporini and Magagnini, 1970; Nilsson, 1969; Potts, 1955; Rouiller and Faur6-Fremiet, 1957). The alternation of electron-dense and electron-transparent striae is caused by a regular three-dimensional arrangement of thin filaments (Fig. 6h) constituting the structural elements of the trichocyst shaft at the macromolecular level (Bannister, 1972; Hausmann et al., 1972a). The filaments are connected in a very regular manner (Figs. 6h and 7). One of the striations is always the same (0in Fig. 6h), while the other alternates in its fine structure (asterisk in Fig. 6h). However, it can be shown by means of a three-dimensional model of the trichocyst shaft that these periods always have the same structure; the bands are just twisted against each other at an angle of 90" (Fig. 7) (Hausmann et al., 1972a). 3. Comparison of Resting and Ejected Spindle Trichocysts of Ciliates On comparing the morphological features of resting and ejected trichocysts of Paramecium the following facts emerge:

1. The resting trichocyst is 3-4 pm long; the ejected form is 25-35 pm long and is therefore about eight times longer than the resting trichocyst. 2. Both resting and ejected trichocysts have about 500 periodic cross-striations. 3. The cross-striations of the resting trichocyst have a periodicity of 7 nm (Fig. 8a'); that of the striations in the ejected trichocyst is 55 nm (Fig. 8b'). Besides an eightfold increase in the length of the trichocyst, the width of the periodicity increases by the same factor of 8. 4. The resting trichocyst has a homogeneous cross-striation (Figs. 6c and d, and 8a'), whereas the striae of the ejected form are divided into several subperiods (Fig. 6h, 7, and 8b').

4. Inhibited Spindle Trichocysts of Ciliates The question arises, what is the mechanism of trichocyst expulsion which takes only a few milliseconds to complete (Jahn and Bovee, 1967; Pitelka, 1963)? The extrusion process cannot be analyzed in the light microscope in oivo, because of the rapidity of the discharge. Even high-speed microcinematography was unable to show the details of expulsion (Miller et al., 1968; Pitelka, 1963). Nevertheless, Kriiger (1930) studied the discharge indirectly. He inhibited the ex-

212

KLAUS HAUSMANN

a'

b'

FIG. 8. Schematic comparison of resting (a) and ejected trichocysts of Paramec:ium

(b). After discharge the homogeneous cross-striationof the resting stageI (a') widens and becomes divided into subperiods (b'). From Hausmann et al., 1972a.

pulsion by using a solution of 5% potassium ferrocyanide. The various trichocyst forms resulting from this treatment can be used to reconstruct the possible course of the discharge (Fig. 9a-f) (Hausmann, 1971). However, the light microscope shows only elongation of the trichocyst and is not able to provide more detailed information. Electron micrographs of these specimens disclose that the essential process of discharge is a rearrangement (recrystallization) of filaments (Figs. 9g-i, 10, and 13a) from a highly ordered structure with a period of 7 nm (Fig. 6c and d; rt in Fig. 10) to another highly ordered structure with a period of 55 nm (Figs. 6f-h, and 7; dt in Fig. 10). Therefore trichocyst expulsion is due to the rapid stretching and unfolding of a network of filaments which are initially packed very close together and are preformed in the resting stage (rt in Fig. 10; Fig. 13a). Such a model has been proposed by Schmidt in 1939, postulating that

EXTRUSIVE ORGANELLES IN PROTISTS

213

an unfolding of filamentous molecules is the basic event in trichocyst discharge; his conclusions were based on polarizing microscope observations. What happens at the molecular level during the expulsion remains to be explained. There is almost no information regarding this problem, with the exception of observations that ATP inhibits the discharge (Anderer and Hausmann, 1977; Hoffmann-Berling, 1960, 1961), whereas bivalent ions and many different stimuli may trigger the expulsion. The reported cytochemical localization of ATPase activity in trichocysts during extrusion (Kawakami, 1971) needs to be verified.

5. Spindle Trichocysts of Microthoracidae The holotrichous ciliate Pseudomicrothorax dubius, family Microthoracidae, is a rather rare protozoan with an uncertain systematic position (Corliss, 1958; Thompson and Corliss, 1958). Like other members of this family, for example, Drepanomonas dentata (Hausmann, 1973d; Hausmann and Mignot, 1975; Prelle, 1968) and Leptopharynx costatus (Prelle and Aguesse, 1968), this ciliate possesses a rather different type of extrusome (tr in Fig. 11)(FaurBFremiet and Andre, 1967; Peck, 1971, 1974); its spindle trichocyst has a quadripartite tip structure (tr in Fig. 11; Fig. 12). The shaft of this trichocyst is structurally and cytochemically similar to the spindle trichocyst of Paramecium (Hausmann and Mignot, 1975); a periodically striped structure is seen in both the resting and elongated stages with periods of 12 and 50 nm, respectively (Fig. 12a and c). But the four apical tubes of the tip are the most striking feature of this extrusome (t in Fig. 12a and b). This tip is a dynamic structure (Figs. 12e-g, and 13b). During or following elongation of the shaft the tubes spread and at the same time excrete a very electron-dense material (arrows in Fig. 12g) which can be seen, even in the light microscope, as four drops at the very tip of the trichocyst (5in Fig. 13b)(Hausmann and Mignot, 1975; Pknard, 1922). The significance of these tubes has not yet been clarified. A problem, especially for the classification of ciliates, is the appearance of similar trichocysts in the scuticociliate Ctedoctema (P. Didier, personal communication). 6. Spindle Trichocysts of Flagellates Flagellates are known to possess extrusomes (for reviews, see Cachon et ul., 1974, 1975; Dodge, 1973; Hall, 1946). Many dinoflagellates have spindle trichocysts, for example, species of Amphidinium (Dodge and Crawford, 1968; Dragesco and Hollande, 1965), species of

214

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

2 15

FIG.10. This partially inhibited spindle trichocyst ofPurumecium demonstrates the mode of ejection in which the resting body (rt) is transformed into the discharged shaft (dt) by a sudden expansion. tt, Trichocyst tip. x 25,000. From Hausmann, 1973d. FIG.9. Inhibited spindle trichocysts of Paramecium in light micrographs (a-f) and electron micrographs (g-i). The different stages found in the light microscope are represented in the electron microscope as a gradual disorganization of the network of the trichocyst filaments. (a-f) x 2100. (g) x 30,000. ( h and i) x 8000. From Hausmann et ul., 1972b.

216

KLAUS HAUSMANN

FIG. 11. Localization of compound trichocysts (tr and arrows) in the pellicular system of the ciliate P . dubius. a, Alveolus; ci, cilium; ep, epiplasm; epr, epiplasmic ridge; mt, microtubule; ps, parasomal sac.

Blastodinium (Soyer, 1970b), Gonyaulax polyedra (Bouck and Sweeney, 1966; Sweeney and Bouck, 1966), species of Gymnodinium (Dragesco and Hollande, 1965; Hausmann, 1973d), Nematodinium armatum (Mornin and Francis, 1967),Noctiluca miliaris (Soyer, 1968, 1969, 1970a,b), Oodinium cyprinodontum (Lom and Lawler, 1973), Oxyrrhis marina (Dodge and Crawford, 1971; Dragesco, 195213; Dragesco and Hollande, 1965; Hausmann, 1973a,d), Peridinium westii (Messer and Ben Shaul, 1969, 1971),Prorocentrmm micans (Dragesco and Hollande, 1965; Sweeney and Bouck, 1966), Strippsiella sweeneyi (Sweeney and Bouck, 1966),Thecadinium kofoidi (Dragesco and Hollande, 1965),Warnovinia pulchra (Greuet, 1969),and Woloczynskia micra (Leadbeater and Dodge, 1966). In the resting stage the trichocyst tip of 0. marina assumes a tubular configuration (tt in Fig. 14a); the material of the shaft shows a striation of 8 nm (tb Fig. 14a). The elongated form is characterized by a cross-striation with a period of 68 nm, which is subdivided into 16- to 17-nm subperiods (Fig. 14d). The smallest structural components of

EXTRUSIVE ORGANELLES IN PROTISTS

217

the shaft are l-nm filaments. This striated pattern is a general characteristic of ejected trichocysts of dinoflagellates, for example, 0. marina (Fig. 14d) and Gymnodinium fuscum (Fig. 14e). The mechanism of discharge of potassium ferrocyanide-treated trichocysts is a sudden change from one paracrystalline state to another (Fig. 14b and c ) ,like the unfolding and elongating process of the Paramecium trichocyst (see Section IV,A,4). A less common and not yet understood spindle trichocyst is found in Gonyostomum semen (Hall, 1946; Mignot and Hovasse, 1974- 1975). Further investigation will be required before its structure and mechanism of extrusion can be determined. 7. Chemical Composition of Spindle Trichocysts Determination of the isoelectric point (IEP) of spindle trichocysts was made by Jakus (1945) and by Wohlfarth-Bottermann and Schwantes (1952). An IEP of pH 4.1 was found for both resting and ejected trichocysts. The chemical composition of spindle trichocysts has been studied by Pollack and Steers (1973) and by Steers et al. (1969). It was shown that trichocysts, isolated for biochemical experiments, are composed of proteins containing no detectable carbohydrate or nucleic acid moieties. When analyzed by acrylamide disk gel electrophoresis in gels containing sodium dodecyl sulfate, two components are detected. These two forms are estimated to have molecular weights of 17,000 and 36,000, respectively. These investigators suggest the name “trichynin” for this structural protein. The trichocysts of Paramecium have frequently been compared with collagen, because of some structural similarities (Beyersdorfer, 1951; Jakus, 1945; Nemetschek et al., 1953; Wohlfarth-Bottermann and Pfefferkom, 1952). However, amino acid analysis indicates that the Paramecium trichocyst is not collagenous (Steers et al., 1969). Moreover, it is significant that a similar noncollagenous amino acid profile has also been reported for the trichocyst protein of the flagellate Peridinium (Messer and Ben Shaul, 1971). Cytochemical experiments have also added to our understanding of the chemical composition of the Paramecium trichocyst. In contrast to the results from biochemical studies a fine staining for glycoprotein material was found in the periphery of the in situ trichocyst body, that is, the meshlike sheath (see Section IV,A,l) (Esthve, 1974). A similar staining is also found in the compound trichocysts of D . dentata (Hausmann and Mignot, 1975). Further studies should clarify the significance of this sheath.

218

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

219

FIG.13. Comparison of the stretching mechanism of the Parumecium trichocyst (aJ4) and the stretching (b,l-4)and spreading process (b,5) of the Microthoracidae extrusome (b). From Hausmann and Mignot, 1975.

8 . Membrane Behavior of Trichocyst Vacuoles during and after Trichocyst Discharge in Ciliates For trichocyst discharge to occur the trichocyst membranes must fuse with the plasma membrane (tmand pm in Fig. 15a and e). The first visible step in the ejection process is the formation of an extremely narrow opening of less than 40 nm diameter (Hausmann and Allen, 1976). This opening is smaller than the diameter of the rosette of particles (c type) found on the P face of the adjacent plasma membrane (Figs. 3a and 4a). As ejection proceeds, this opening widens until it attains the diameter of the tubular collar (Fig. 15e) (Hausmann and Allen, 1976; Plattner et al., 1973a).The tubular collar may reinforce and protect the membrane at the distal tip of the trichocyst (Allen and Hausmann, 1976).As the trichocyst passes through the tubular collar, as through a nozzle, it is transformed into an elongated structure with a 55-nm periodicity (Fig. 15a and e). Immediately after expulsion the membrane of the vacuole is again separated from the plasma membrane and released back into the cytoplasm (Fig. 15b and e) (de Haller and ten Heggeler, 1969; Hausmann and Allen, 1976; Pitelka, 1965; Plattner, 1976). This process takes place very rapidly, in FIG. 12. Compound trichocyst of Microthoracidae in electron micrographs. The trichocyst consists of a body (tb) and four tubes (t). After ejection the elongated shaft (d) shows a periodic cross-striation (c). The tubes spread (e-g) and excrete an electron-opaque material (arrows). (a) ~22,000.(b) x 34,OOO. (c) ~54,000.(d) X3500. (e-g) x 9500. (a) From R. K. Peck; (d-g) from Hausmann and Mignot, 1975.

220

KLAUS HAUSMANN

FIG. 14. Spindle trichocysts of the dinoflagellates 0. marina (a-d) and G .fuscum (e).(a) Resting stage. (b and c) During expulsion. (d and e) Ejected stage. tb, Trichocyst

body; tt, trichocyst tip. (a and b) x 35,000.(c) x 25,000.(d and e) x 200,000. (a-d) From Hausmann, 1973a.

less than a second. The membrane then vesiculates into small units (Fig. 15c-e), which can no longer be distinguished from vesicles of the same dimensions that normally exist within the cell's cytoplasm. The entire process is completed within 5-10 minutes. By this mechanism of trichocyst discharge the complex pellicular

EXTRUSIVE ORGANELLES IN PROTISTS

22 1

FIG.15. Membrane behavior of the trichocyst vacuole in Paramecium. During the discharge the trichocyst membrane and plasma membrane fuse (a). Afterward the trichocyst vacuole is pinched off (b) and broken down into small vesicles (c and d). (e) Schematic representation of this process. am, Alveolar membrane; pm, plasma membrane; tb, trichocyst body; tc, tubular collar; tm, trichocyst membrane; tt, trichocyst tip. (a-d) x 30,000. From Hausmann and Allen, 1976.

system of the ciliate (Allen, 1971) is not disturbed. A similar situation has been reported for another exocytic event known to occur in ciliates: The membranes of defecation vacuoles are not incorporated into the plasmalemma but are pinched off into the cytoplasm after exocytosis (Allen and Wolf, 1974).

9. Origin and Cycle of Spindle Trichocysts In those cells that have been investigated (Ehret and de Haller, 1963; Selman and Jurand, 1970; Yusa, 1963, 1965) the spindle trichocysts of ciliates originate from the endoplasmic reticulum. The,first stage seen is a membrane-limited endoplasmic vesicle enclosing a homogeneous mass (asterisk in Fig. 16a and b) which is itself a condensation of closely packed, linearly oriented fibrous elements (0in Fig. 16a and b; 1in Fig. 18).This paracrystalline core is the first definitive sign of the presumptive trichocyst. The form of the juvenile trichocysts, recognizable as a system of closely packed, multilayered, fibrous sheets, becomes increasingly evident (2 and 3 in Fig. 18).Matu-

222

KLAUS HAUSMANN

FIG. 16. Early stages of trichocyst development in P. cuudatum. A paracrystalline structure (0)lies embedded in an amorphous matrix (asterisk). (a and b) x 25,000.

ration of the trichocyst includes formation of the cap, elaboration of the characteristic periodic striation of the body, and development of the tubular collar (4 in Fig. 18). Development of the compound trichocysts in P. dubius shows a variation. Obviously the four arms are composed of dictyosomederived globules which consolidate to form ultimately the rodlike tubes of the trichocysts (1-5 in Fig. 17) (Hausmann, 1977a). The flagellate's spindle trichocysts are reported to originate from swollen vesicles which arise in the vicinity of the Golgi apparatus (Bouck and Sweeney, 1966; Leadbeater and Dodge, 1966). The succeeding steps in their development are quite similar to those for ciliate trichocysts. After trichocyst discharge (5 in Fig. 18) the trichocyst membrane remains inside the ciliate (6 in Fig. 18) and is transformed into small vesicles (7 in Fig. 18) which probably return to the endoplasmic reticulum (er in Fig. 18) or are engulfed by autolysosomes (Fig. 18).

FIG. 17. Development of' compound trichocysts. Dictyosoma-derived vesicles fuse with a vacuole (1). The contents of these vesicles are the material for the rodlike tubes of the mature trichocyst (5). 2-4, different developmental stages. From Hausmann,

1977a.

EXTRUSIVE ORGANELLES IN PROTISTS

223

FIG. 18. Scheme of the spindle trichocyst cycle in Paramecium. Vesicles with a paracrystalline core (l),presumably derived from the endoplasmic reticulum (er), mature to normal-shaped trichocysts (2-4) which then become fixed to the pellicle a t special sites (4). During expulsion the trichocyst membrane and plasmalemma fuse (5);the vacuole is then pinched off (6)and broken down into small vesicles (7) which may then be incorporated into the endoplasmic reticulum or into autophagosomes.

At present the manner in which young trichocysts find their way to their destinations in the complex architecture of the pellicle is not yet clear. Recently, however, Peck (1977a,b) reported that specific proteins in the epiplasm of P . dubius are localized at the trichocyst insertion sites. These proteins can be selectively dissolved and identified electrophoreticall y.

10. Function of Spindle Trichocysts Originally spindle trichocysts were thought to be organelles of defense. For example, Didinium, when attacking Paramecium, is sometimes forced back mechanically by a massive discharge of trichocysts (Mast, 1909).But Didinium, feeding only on Paramecium, can survive and multiply very well. However, some less specialized enemies are probably chased away by trichocyst discharge. A second possibility is that trichocysts are used by Paramecium to adhere to surfaces (Saunders, 1925). But this idea cannot explain the abundance of trichocysts in a single cell (6000 to 8000). Wohlfarth-Bottennann

224

KLAUS HAUSMANN

(1950,1953) has suggested an osmoregulatory function for the spindle trichocysts ofParumecium in which the trichocysts may operate as ion carriers. However, this concept was not confirmed when more sensitive experimental techniques became available, for example, electron probe x-ray analysis. The function proposed for dinoflagellate trichocysts by Ukeles and Sweeney (1969) seems to be of secondary rather than primary importance. Ejected trichocysts are said to block the mouth orifice of the bivalve mollusc larvae Crussostrea virginicu, which prevents them from taking in food. B.

MUCOCYSTS (INCLUDINGMUCIFEROUS BODIES, KINETOCYSTS, AND CLATHROCYSTS)

1. Resting Mucocysts of Ciliates In ciliates mucocysts are found in the resting stage, like spindle trichocysts, in predictable positions in the well-organized cortical system, for example, in Tetruhymena (mu in Fig. 19)(Allen, 1967; Elliott, 1973; Nilsson, 1976; Pitelka, 1961; Satir et al., 1973; Wunderlich and

FIG.19. Schematic representation of the regular position of the mucocysts (mu) in the pellicular system of T . pyriformis. For a description of the other details see the original article. From Allen, 1967.

EXTRUSIVE ORGANELLES I N PROTISTS

225

Speth, 1972).The appearance of these organelles is very similar in the different species of ciliates; they are membrane-limited, oval bodies (Fig. 20a) with paracrystalline contents (Fig. 20b-d), and they underlie the plasmalemma (Bohatier, 1970, 1972; Bohatier and Detcheva, 1973; Cheissin and Mosevich, 1962; Dragesco et al., 1965; Foissner and Simonsberger, 1975; Grain, 1968a,b, 1970; Grain and Golinska, 1969; Grasse and Mugard, 1961; Hausmann, 1972c; Kink, 1973; Kovaleva, 1974; de Puytorac, 1964b; de Puytorac and Grain, 1968; de Puytorac and Kattar, 1969; Rieder, 1971; Rodrigues d e Santa Rosa and Didier, 1975; Roque et al., 1965, 1967; Rosati Raffaelli, 1970; Tokuyasu and Scherbaum, 1965; Wessenberg and Antipa, 1968; Yagiu and Shigenaka, 1958a,b; Zebrun et ul., 1967). The organelles at times reveal an elaborate fine structure, particularly after the use of certain fixatives (Fig. 20d) (Hausmann, 1973c; Williams and Luft, 1968). Only rarely are other structures, besides the paracrystalline body, found in mucocysts, for example, tubules and bundles of filaments (Didier and Detcheva, 1974). In a few cases two or more morphologically discernable types of mucocysts have been described for the same cell (Grain, 1970; Kovaleva, 1974; Rodrigues de Santa Rosa, 1974; de Puytorac and Rodrigues de Santa Rosa, 1975). For example, Loxophyllum meleagris possesses normal mucocysts (Fig. 20a, c, and d), as well as conocysts (Hausmann, 1977b; Rodrigues de Santa Rosa, 1974).The significance of this variation has not been explained. The extrusive pigment granules (pigmentocysts)reported for Blepharisma (Inabaet d ,1958; Kennedy, 1965; Rao, 1963),Stentor (Tartar, 1961),Loxodes (Mashansky et al., 1963), and Truchelonema (Kovaleva, 1974; Kovaleva and Raikov, 1972; Raikov and Kovaleva, 1968) represent, in all likelihood, a special type of mucocyst. 2. Ejected Mucocysts of Ciliates Mucocysts have in general a specific form after ejection, as well as during the resting stage (Fig. 21a-c) (Bresslau, 1923; Hausmann, 1972c,d, 1973c,d; Hausmann and Bohatier, 1977; Hayashi, 1974; Kriiger, 1934a; Wohlfarth-Bottennann and Pfefferkorn, 1953a,b), and this is not a mucigenic, amorphous mass as previously reported (Satir et al., 1973). They are composed of filaments connected in a three-dimensional network (Fig. 21d'). This structure has been observed following negative staining (Fig. 21a and d-g), shadow casting (Fig. 21b), and thin-sectioning (Fig. 21c) (Hausmann and Stockem, 1973). In negatively stained specimens different network patterns can be visualized: (1)an unordered network (Fig. 21d and d'), (2) a pattern of

226

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES I N PROTISTS

227

rhombic structures (Fig. 21e and e’), (3) a lattice of hexagonal elements (Fig. 21f and f‘), and (4) a latticework composed of tetragonal units (Fig. 21g and g’). All these different patterns can be reduced to one three-dimensional structure (Fig. 21d’-g’), since the various patterns depend on the plane of view of the ejected mucocyst network (Hausmann, 1972d). 3 . Chemical Composition of Ciliate Mucocysts The chemical composition of mucocyst material was first studied by Alexander (1968). H e demonstrated that the mucocysts of Tetrahymenu are rich in acidic residues and contain two proteolipids and one protein. Moreover, the presence of acid mucopolysaccharides in mucocysts has been demonstrated b y cytochemical methods (Grain, 1968b). In a more recent article Hayashi (1974) described the influence of pH variations on discharged mucocysts, as well as the effect of different temperatures and changes in ionic strength. Hayashi also investigated the effect of some chemicals (EDTA, EGTA, ATP, GTP, calcium chloride, dithiothreitol, mercaptoethanol, colchicine, and vinblastine), and several enzymes (DNase, RNase, phosphodiesterase, lecithinase, lipase, lysozyme, pronase, a-chymotrypsin, and trypsin) on the secreted organelles. Hayashi’s work shows that isolated, discharged mucocysts consist mainly of proteins which can be solubilized under mild conditions (40°C for 30 minutes at pH 7.0).

4. Membrane Behavior during Mucocyst Discharge in Ciliates In order for the mucocyst contents to be discharged the mucocyst membrane (mum in Fig. 22a) must first fuse with .the plasmalemma (pm in Fig. 22a), an event that can be traced in thin sections (Fig. 22ac). During this process the material within the mucocyst’s vacuole expands and becomes less electron-opaque (Fig. 22c). ~

~

FIG.20. Thin sections of resting mucocysts of L. meleagris (a and d), T. pyriformis (b), and Phacodinium metchnikoffi (c). The paracrystalline bodies (mub) are membrane-bounded (mum) and lie below the plasma membrane (pm). Following the use of special fixatives more details of the periodic structure become visible (d).Conocysts of L. meleagris are present, in the resting stage (e-g) and in negatively stained preparations, as ejected organelles (h-k). This type of extrusome consists of an axial rod (+) enveloped by a conical jacket (e, f, and g). After ejection the jacket seems to be amorphous (h), whereas the rod exhibits a distinctive fine structure (i, k). (a) x50,OOO. (b) x77,OOO.(c) x74,000.(d) ~275,OOO.(e-g) x70,000.(h) x35,000.(iandk) ~53,OOo. (e-k) from Haus(b) From R. D. Allen; ( c ) from P. Didier; (d) from Hausmann, 1973~; mann, 197%.

228

KLAUS HAUSMANN

FIG. 21. Ejected mucocysts of L. meleagris show definite forms (a) which can b e shown following negative staining (a), shadow casting (b), and thin-sectioning (c). The various network patterns of filamentous structure (d-g) are explained by means of a three-dimensional model (d‘-g’), The form of the pattern depends on the plane in which the mucocyst is viewed [compare (d-g) with d’-g’l. (a) ~ 6 0 0 0 (b . and c) ~40,000.(d-g) x 75,000. (b-g’) From Hausmann, 1972d, 1973c.

EXTRUSIVE ORGANELLES IN PROTISTS

229

The freeze-fracture technique was employed by Satir et al. (1972, 1973) to reveal details of the process of fusion of the plasma membrane with the mucocyst membrane in Tetrahymena pyriformis. The undischarged mature mucocyst finds its way to a special site on the plasmalemma. A rosette of 15-nm-diameter particles forms within the plasma membrane (Fig. 22d; arrow in Fig. 22g) when the mucocyst reaches a critical distance from it. Adjacent to this site, within the mucocyst membrane, is an annulus of ll-nm particles, whose inner edge comes to encircle the rosette in the plasma membrane (Figs. 22e and 23a). During discharge the cytoplasm between the two matching membrane sites is squeezed out, and the membranes fuse (Fig. 22g). Steps in membrane reorganization can be reconstructed from changes in the appearance of the rosette in the fracture faces. First, a depression within the rosette, the fusion pocket, forms in the plasma membrane (Figs. 22h and 23b). The rosette particles spread at the lip of the pocket as the pocket deepens and enlarges from 60 to 200 nm (Figs. 22i and k,and 23c). The annulus particles of the mucocyst membrane then become visible at the open lip, indicating completed fusion of the P faces of the mucocyst and the plasma membrane (Fig. 22m and n). While the contents of the mucocyst are released, the edges of these faces join so that the unit membrane runs uninterruptedly from the plasma membrane around the lip and into the pocket of the mucocyst (Fig. 23d). What happens to the mucocyst membrane after the discharge of its contents? The hypothesis proposed by Satir (1974a,b, 1975) and Satir et al. (1976)postulates that in Tetrahymena at least the lipid portion of this membrane is incorporated into the plasma membrane. The protein portion may be excluded. Satir and Satir (1974) calculated that the membrane surface area of all the mucocysts of one cell is sufficient to produce the plasma membrane of a daughter individual. Thus it is concluded that mucocysts may be a source of new plasma membrane during cell growth and division. And yet, how the growth of other intimately associated pellicular membranes, for example, the alveoli, is accomplished was not discussed. Experimental work will be required before this hypothesis concerning the fate of mucocyst membranes in Tetrahymena can be confirmed. It is also possible that a similar mode of membrane uptake or reengulfment exists, as is evident for the trichocyst membranes in Paramecium (see Section IV,A,8). The recent micrographs of Nilsson (1976) support this alternative explanation, that is, a breakdown of the mucocyst membrane of Tetrahymena into small vesicles and their subsequent engulfment. In Litonotus duplostriatus, a ciliate whose plasma membrane has a well-developed

230

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

231

surface coat, the mucocyst membranes do not have such a coat during the secretion process (Hausmann and Mocikat, 1976), and consequently they are not likely to become part of the plasma membrane.

5. Mechanism of Mucocyst Discharge The mechanism of mucocyst secretion is somewhat similar to the ejection of spindle trichocysts. One paracrystalline form gives rise to another paracrystalline form. In both mucocysts and spindle trichocysts threadlike structures are the subunits that make u p a three-dimensional network. The differences are that mucocyst discharge takes several seconds, while spindle trichocyst discharge lasts less than a second; that during discharge the mucocyst increases in all dimensions (length, width, and depth), while spindle trichocysts increase only in length; and that the chemical composition of the mucocyst seems to be less homogeneous than that of the spindle trichocyst. The mechanism of mucocyst secretion is, however, more similar to that of spindle trichocyst ejection than to any other type of extrusomal discharge. Mucocyst secretion results from the stretching and unfolding of a preexisting network of filaments, as does spindle trichocyst ejection. 6. The Origin of Mucocysts in Ciliates The origin of mucocysts has been studied by several investigators (Dragesco et al., 1965; Grain, 196813; Kovaleva and Raikov, 1972; Rieder, 1971; Tokuyasu and Scherbaum, 1965; Wessenberg and Antipa, 1968). Except for Grain (1968b), who proposed an elaborate cycle of mucocyst development in Balantidium, the various workers unaniFIG. 22. Mucocyst secretion seen in thin sections (a-c) and freeze-fracture preparations (d-n). The thin sections ofhrophyllvrn show the fusion (arrows) of the mucocyst membrane (mum) and plasma membrane (pm). The content is lost (c). Freeze-fracture preparations of mucocyst secretion in Tetrahymena have been interpreted in the following way. Prior to discharge (d) the resting mucocyst lies directly below a rosette of particles within the plasma membrane (arrowhead). In the vesicular membrane a matching site, the annulus, is present (e).The P faces of the plasma and mucocyst membranes fuse (f). Note the presence of the annulus at the joined region (arrowhead). In freeze-fracture micrographs the fusion process can be followed best on the P face of the plasma membrane. (g) Rosettes of particles are found before fusion. (h) A depression then appears within the rosette. (i) The depression widens and the particles of the rosette separate at the edge of the pocket. (k) Further separation of the rosette particles takes place, and the pocket deepens. (1 and m) The particles of the annulus become visible. (n) Different pocket depths can sometimes be seen within a single fracture. (a-c) x 60,000. (d) x 96,OOO. (e-n) x 72,000. (a-c) From Hausmann, 1973c; (d-f) from Satir and Satir, 1974; (g-n) from Satir et al., 1973.

FIG.23. Diagram of membrane fusion during mucocyst discharge. (a) A cross section of the plasma membrane (PM), which contains large rosette particles and an underlying mucocyst membrane (Mc) containing rows of smaller annulus particles just before fusion. Wavy lines indicate the fracture plane along which the membrane will split into A and B halves. (b) Fusion begins when a hinge is formed between the surface complex, corresponding to the rosette and the inner edge of the annulus. (c)After the rosette particles spread, evolution of the annulus takes place, and the fusion of the A faces (known now as the P faces) is completed. (d) Reorganization occurs around the fusion lip. The fracture line now passes through the expanding mucocyst content. From Satir et al., 1973.

EXTRUSIVE ORGANELLES I N PROTISTS

233

mously describe one developmental pathway. A small vesicle is postulated to arise from the endoplasmic reticulum (er in Fig. 24), whose contents are first granular and then crystalloid (1in Fig. 24). The vesicle grows to the normal size of a mucocyst (2 and 3 in Fig. 24) and then finds its way toward its final location in the pellicular system (4 in Fig. 24). After discharge (5in Fig. 24) the empty vacuole remains for a time fused with the plasmalemma (6 in Fig. 24). The fate of the mucocyst membrane is unknown, or at least controversial.

7. Mucocysts of Flagellates Different flagellates are known to have mucocysts (for reviews, see Cachon et al., 1974-1975; Hall, 1946; Hovasse, 1969). In some cases their fine structure is similar to that of the mucocysts of ciliates, but they are usually different. Very often their contents seem to be granular to amorphous, and sometimes honeycomblike or fibrillar. In the chloromonadine Chattonella subsala the mucocysts (mu in Fig. 25a) arise from the Golgi apparatus (Ga in Fig. 25a). After discharge, a large, hollow tube is formed by irregularly connected fila-

FIG. 24. Diagram of the development of mucocysts in Tetruhymenu. Originating from the endoplasmic reticulum (er), vesicles with a paracrystalline core (1)increase in size and assume the normal morphology of mucocysts (2,3, and 4). After discharge (5) the mucocyst membrane remains attached to the plasma membrane for some time (6). Further steps in the development of this membrane are unknown.

234

KLAUS HAUSMANN

FIG.25. The mucocysts of the chloromonadine C. subsala (mu) originate from the Golgi apparatus (Ga). After the discharge, filaments form a hollow tube (arrow). n, Nucleus. (a) x 20,000. (b) x 6000. From Mignot, 1976.

EXTRUSIVE ORGANELLES IN PROTISTS

235

FIG.26. Muciferous bodies of euglenoids. (a) Distigrnu proteus. (b) Colacium mucronuturn. ( c )Euglenu stellutu. From Mignot, 1966.

ments (Fig. 25b) (Mignot, 1976). A similar situation is found in the mucocysts of Euglena splendens (Hausmann and Mignot, 1977). In euglenoids, besides mucocysts, muciferous (mucigenic) bodies occur (for reviews, see Dodge, 1973; Leedale, 1967). These bodies, whose structure varies in different species (Fig. 26), are presumed to secrete mucilage continuously. The mucus is probably necessary for the movement of the cell. Special pores in the pellicle (Arnott and Walne, 1967; Mignot, 1965a, 1966, 1967a,b; Schwelitz et al., 1970; Sommer, 1965) allow the continuous secretion of mucus. Some correlation may exist between the presence of muciferous bodies and mucocysts and the palmella stage of flagellates (Hovasse and Mignot, 1975). Very little is known about the chemical nature of the mucocysts and muciferous bodies of flagellates. However, mucin isolated from cultures of Euglena gracilis var. bacillaris has been found to have a carbohydrate content of at least 82%. Chromatographic analysis of acid hydrolyzates of the mucin has shown the presence of glucose, galactose, mannose, xylose, fucose, rhamnose, and small amounts of unidentified slow-moving sugars (Barras and Stone, 1968). Several amino acids have also been detected in the acid hydrolyzates of the mucin, indicating the presence of a relatively small amount of protein in these preparations; whether this is part of the mucin structure is not yet known (Barras and Stone, 1968). A unique kind of extrusome, which may be a mucocyst, has been described in the euglenoid flagellate Entosiphon sulcatum (Mignot, 1963,1967b; Mignot and Hovasse, 1973). A similar type is reported in Zsonema nigricans (Schuster et al., 1968). In the resting stage these organelles are thick-walled tubes (Fig. 27a-c) surrounded by a mem-

236

KLAUS HAUSMANN

FIG. 27. Extrusomes of the euglenoid flagellate E . sulcatum. In the resting state tubelike structures (a and b) are surrounded by a membrane (m). The ejected stages (d and e) reveal a regular pattern of two kinds of filaments (fi and fi). (a) x 12,500. (b) x 25,000. (c and d) x 60,000. (e) x 75,000. (a, b) From Mignot, 1966; (c-e) from Mignot and Hovasse, 1973.

EXTRUSIVE ORGANELLES IN PROTISTS

237

brane (m in Fig. 27c), but after discharge the wall becomes thinner (Figs. 27d and 2%). Two sets of filaments make up the hollow tube formed at discharge (fi and f2 in Fig. 27e; Fig. 28b). These two kinds of filaments do not seem to have the same chemical composition, since they show different affinities for stains for polysaccharides; the fi filaments reveal a heavy precipitation, whereas the f2 filaments are devoid of stain. Both are unaffected by enzymic treatment (Mignot and Hovasse, 1973). No explanation has been pregented for these results. The discharge of this type of extrusome is characterized as a sudden unfolding and stretching of a preexisting pattern, similar to the ejection of spindle trichocysts in ciliates (see Section IV,A,4). Expansion occurs in length and width by factors of 3 and 1.5, respectively (Fig. 28a).

I -

FIG.28. Mode of ejection of the extrusome in E . sulcatum. The organelle increases in size, expanding in length by a factor of 3 and in width by a factor of 1.5 (a). During this process ordered filaments become visible (b). (b) After Hovasse and Mignot, 1975.

238

KLAUS HAUSMANN

8. Kinetocysts of Rhizopods Kinetocysts are particles in the axopods of centrohelidians (Fig. 29ac) (Bardele, 1969, 1971, 197213, 1975, 1976a; Hovasse, 1965c; Troyer and Hauser, 1977), which perform discontinuous, jerky, bidirectional movements within the axopods at a velocity of 1-5 pm per second. A

FIG.29. Kinetocysts of the pseudoheliozoan Clathrulina elegans-(a), and the centrohelidians Acanthocystis rnyriospina (b), and Heterophrys sp. (c). The organelle is bounded by a membrane (km) and consists of a central element enveloped by a ringlike jacket. Characteristically they have a position between the plasma membrane of the filopods (pm) and filopodial microtubules (mt).'(d) In Acanthocystis en'nnceoides the P face of the plasma membrane (PF-pm) displays numerous kinetocyst attachment domains (arrowheads). (a) x 30,000. (b) x 65,000. (d) x 60,000. (a, b) From Bardele, 1972b; (c) from Troyer and Hauser, 1978; (d) from Bardele, 1976a.

EXTRUSIVE ORGANELLES IN PROTISTS

239

kinetocyst is a complex polar organelle surrounded by a membrane (km in Fig. 29a and c). It contains an electron-opaque bipartite central element enclosed in a jacket of less electron-opaque material (Fig. 29c). The organelles are situated between bundles of microtubules (mt in Fig. 29a and b) and the plasma membrane (pm in Fig. 27b and c) (Bardele, 1972b; Troyer and Hauser, 1978). The dense globules, which give rise to the characteristic bumpy appearance of the filopods of the ameboid stage of Chrysamoeba radians (Hibberd, 1971), apparently represent a type of kinetocyst. The mucilage vesicles in the ameboflagellate Gyromitus limax (Swale and Belcher, 1975) may be a variation of the kinetocyst, too. In the presence of formaldehyde fumes, the centrohelidian kinetocyst is discharged, and negative staining reveals that the jacket has been expelled. The material of the jacket seems to function as a propellant for the central element which is observed lying in front of the burst jacket. Bardele (1976a) characterizes the kinetocyst as a special type of compound motile mucocyst which most likely engages in trapping food. In fact, centrohelidian axopods are sticky and have an immobilizing effect on certain other protozoa. Freeze-fracture studies of these extrusomes, initially performed by Davidson (1974), showed that the attachment site of kinetocysts to the plasmalemma bears an array of particles (Fig. 29d) (Bardele, 1976a; Davidson, 1976) somewhat similar to the rosette found at the mucocyst attachment site in Tetrahymena (see Section IV,B,4). However, the function of this particle array in centrohelidians, called an attachment domain (Bardele, 1976a), was interpreted quite differently from the proposed function of the fusion rosette of mucocysts. Although the attachment domain of the kinetocyst defines the site where the fusion of the organelle’s membrane with the plasma membrane will occur upon an adequate stimulus, the particular particle arrangement is interpreted as a membrane differentiation which may prevent the organelle from being discharged at the wrong time (Bardele, 1976a), rather than to prepare for its expulsion, as proposed for Tetrahymena (Satir et al., 1973). Since it has been claimed that membrane fusion requires areas of relatively high fluidity (Poste and Allison, 1973), a fusion-delaying function of particles seems to be obvious; the attachment domain probably represents a very stable, nonfluid membrane area. 9. Function of Mucocysts In investigating mucocyst function, Bresslau (1921a,b, 1923, 1924) discovered a previously unknown type of behavior in ciliates. Following treatment with several dyes (trypaflavin, neutral red, methylene

240

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

24 1

blue, or cresyl blue), ciliates, in particular Colpidium colpodu, secreted a gelatinous capsule. At a later time the protozoan can leave the capsule (Fig. 30a, for Tetrahyrnena), especially after it has been returned to its normal culture medium. Moreover, Bresslau (1924, 1928) found that the capsules were composed of small, secreted, fused rods, Tektinstabchen. He felt that these rods, which in fact were discharged mucocysts, were somehow related to trichocysts, and he concluded that trichocysts are responsible for encystment, a conclusion which was at least partly correct. Later Schneider (1930) systematically examined 92 species of ciliated protozoans for Tektin. He showed the presence of this material in many ciliates which are able to encyst and are known today to have mucocysts . Light and electron microscope studies, performed in a manner similar to the experiments of Bresslau, have now been made with Tetruhymena (Tiedtke, 1976).The method used to trigger the secretion of capsules is to expose the cells to a 0.01% alcian blue solution which is reported to free cells of all mucocysts. Ciliates, at the time they leave the capsule (Fig. 30a), have almost no mucocysts (Fig. 30b). The capsule stains with the same dye that stains discharged mucocysts (Nilsson and Behnke, 1971) and is composed of filamentous material resembling similar material in ejected mucocysts (Fig. 30b and c). The filaments are connected in an irregular manner (Fig. 30d). This experimentally induced capsule is not a cyst wall, sensu stricto, but probably a precursor of the cyst wall. Generally mucocysts are thought to play an important role as organelles, providing the raw material for formation of the cyst wall (Cheissin and Mosevich, 1962; Dragesco et al., 1965; Kawakami and Yagiu, 1963a,b,c, 1964a,b,c; Grass6 and Mugard, 1961; Repak and Pfister, 1967; Rieder, 1971; Roque et ul., 1965; Zebrun et ul., 1967). In this context an observation of some special metazoan cells is interesting. Cortical granules with a paracrystalline fine structure similar to that of mucocysts (Afzelius, 1956) are present in sea urchin eggs. They are composed of polysaccharides as well as proteins (Monn6 and Harde, 1951; Monnk and Slautterback, 1950). These granules are FIG.30. Capsule shedding in Tetrahyrnena. After treatment with 0.01%alcian blue the ciliates form a capsule which they can leave at a later time (a). Then the cells are devoid of mucocysts (b). The capsule is built from discharged mucocysts (c). A higher magnification [compared with the rectangle in (c)] reveals the filamentous nature of the capsule [(d); see also (b)]. (a) x 1100. (b) ~33,000.(c) ~ 3 7 5 0 .(d) ~ 5 0 , 0 0 0 From . Tiedtke, 1976.

242

KLAUS HAUSMANN

ejected and become part of the fertilization membrane (Lallier, 1977; Schatten and Mazia, 1976; Vacquier, 1976), and in this way they resemble mucocysts which are discharged to become part of the cyst wall. In Didinium nasutum, a ciliate that readily encysts, just before cyst formation special extrusomes appear in the cytoplasm, the clathrocysts (cl in Fig. 31) (Holt and Chapman, 1971; Rieder, 1970). These organelles are most probably highly modified mucocysts. They supply

FIG.31. Clathrocysts (cl) during an early stage of encystment inD. nasutum. x 7900. From Holt and Chapman, 1971.

EXTRUSIVE ORGANELLES I N PROTISTS

243

the largest portion of material for the multilayered cyst wall which consists of polysaccharides, proteins, and lipids (Rieder, 1973). Normal mucocysts supply a smaller portion of cyst wall material. Other investigators have proposed different functions for mucocysts (Chapman-Andresen and Nilsson, 1968; Nilsson, 1972,1976). In Tetrahymena mucous material (mucocysts) has been shown, by using alcian blue, to have binding properties similar to those of the ameba mucous coat (Nilsson and Behnke, 1971). In amebas such mucous material is capable of concentrating solutes, proteins, and inorganic cations up to 10 times or more of that in the external medium. Alcian blue binds to the mucous coat of amebas and is taken up by endocytosis (Chapman-Andresen, 1972). In Tetrahymena, the dye-mucus complex also appears to be engulfed by food vacuoles, a finding interpreted to be an indication that the mucus derived from extruded mucocysts may be involved in the feeding process. Other cytological evidence suggests that not only alcian blue but also components of the growth medium are adsorbed by extruded mucocysts and in this way are ingested by food vacuoles (Nilsson, 1976).Thus, in Tetruhymena as in Amoeba,the mucous substances have been interpreted to be implicated in endocytic uptake of nutrients. However, the extruded mucocyst’s ability to adsorb and concentrate material may not be of much physiological significance to the ciliate, since normally the discharged mucocyst, unlike the glycocalyx of the ameba, is completely separated from the cell and probably quickly lost into the surrounding medium. The ingestion of stained and unstained mucocysts may be affected by the experimental conditions. However, mucocysts in several flagellates seem to be involved mechanically in food capture. For example, the dinoflagellate N . miliaris has a tentacle for grasping food. This tentacle is bordered with numerous mucocysts which make it sticky and allow it to hold the food particles (Soyer, 1969, 1970a,b).This function is similar to that of the kinetocysts in centrohelidian axopods (see Section IV,B,8). C. TOXICYSTS(INCLUDING CYRTOCYSTS, PEXICYSTS, AND HAPTOCYSTS)

1. Definition of Toxicysts, Cyrtocysts, Pexicysts, and Haptocysts Toxicyst was the original term for the kind of extrusome in ciliates that is extruded as a tubelike structure and which secretes poisonous material (Kriiger, 1931b, 1934a, 1936).

244

KLAUS HAUSMANN

Cyrtocysts, described in Didinium (Wessenberg and Antipa, 1968), are basically similar to prototype toxicysts but appear strongly curved. Pexicysts, also found in Didinium (Wessenberg and Antipa, 1968), resemble toxicysts but are said to be organelles for the fixation and fastening of prey. Haptocysts (Bardele and Grell,' 1967) (missile-like bodies; Rudzinska, 1965) (phialocysts; Batisse, 1967) are extrusomes which are nearly always found in suctorians. Probably the only exception to this is the ciliate Cyathodinium which has haptocysts within its endosprits (Paulin and Corliss, 1969). Unlike toxicysts, they have a bottle-shaped structure. They join the suctorian tentacle to the prey. 2. Distribution and Location of Toxicysts in Ciliates Numerous ciliates, which are mainly, but not exclusively, Gymnostomatidae, are reported to have toxicysts, for example, Actinobolina (Holt and Corliss, 1973; Holt et al., 1973; Kriiger, 1936; Moody, 1912; Wenrich, 1929),Acropisthium (Bohatier and Detcheva, 1973),Ancistrocoma (de Puytorac, 1969),Chaenea (Dragesco, 1952a, 1962; FaurBFremiet and Ganier, 1969), Coleps (Kruger, 1936),Didinium (Dragesco, 1952a, 1962; Krtiger, 1936; Rieder, 1968a,b, 1971; Schwartz, 1965; Wessenberg and Antipa, 1968,1970; Yagiu and Shigenaka, 1965),Dileptus (Dragesco, 1952a, 1962; Dragesco and MBtain, 1948; Dragesco et al., 1965; Dumont, 1961; Golinska, 1974; Golinska and Grain, 1969; Grain and Golinska, 1969; Hausmann and Bohatier, 1978; Hayes, 1938; Kink, 1973; Kriiger, 1936; Metzner, 1933; Miller, 1968; Studitsky, 1930; Visscher, 1923), Enchelys (Dragesco, 1962; Kriiger, 1936), Helicoprorodon (de Puytorac and Kattar, 1969), Hemiophrys (Kriiger, 1936), Holophrya (Kriiger, 1936), Homalozoon (Kriiger, 1936), Lacrymaria (Bohatier, 1970, 1972; Dragesco, 1962; Kriiger, 1936), Lagynophrya (Grain, 1970), Legendrea (Kahl, 1926; PBnard, 1914), Litonotus (Bohatier and Njine, 1973; Dragesco, 1952a, 1962; Kruger, 1936), Loxophyllum (Dragesco, 1952a; Fritzsche, 1911; Hausmann, 1972b; Hausmann and Hausmann, 1973; Hausmann and Wohlfarth-Bottermann, 1973; Kriiger, 1931b, 1936; Peschkowsky, 1931; de Puytorac and Rodrigues de Santa Rosa, 1975), Monodinium (Rodrigues de Santa Rosa and Didier, 1975), Prorodon (Dragesco, 1952a; Hausmann and Wohlfarth-Bottermann, 1973; Kriiger, 1934a, 1936; de Puytorac, 1964a; Wohlfarth-Bottermann and Pfefferkom, 1953a,b), Pseudoprorodon (Dragesco, 1952a; Kattar, 1972; Kriiger, 1936),Spathidium (Dragesco, 1952a; Kriiger, 1936; Moody, 1912),and Trachelophyllum (Kriiger, 1936). Toxicysts may be distributed randomly in the cortex of the cil-

EXTRUSIVE ORGANELLES IN PROTISTS

245

iate, as for example in Prorodon teres (Hausmann and WohlfarthBottermann, 1973), but more commonly they are found in quite specific parts of the cell or in special regions of the cortex. The proboscis is frequently well equipped with toxicysts, for example, in Didinium, Dileptus, Helicoprorodon, Lacrymaria, Lagynophrya, and Litonotus. Legendrea (Fig. 32a) and Actinobolina (Fig. 32b) have more unique locations for their toxicysts. These ciliates have retractable tentacles

-10

a --+

L-

I

i

-

C

\

FIG. 32. The tentacle-bearing ciliates Legendrea (a) and Actinobolina (b). Legendrea is drawn with retracted (1)and extended tentacles (2).A toxicyst is located at the tip of the tentacles ofActinobolina (to); the tentacle has an ordered fine structure (d).a, Alveolus; cv, contractile vacuole; mt, microtubules; n, nucleus; os, mouth; t, tentacle. (a) After PBnard, 1914; (b) after Wenrich, 1929.

246

KLAUS HAUSMANN

(1and 2 in Fig. 32a). Inside the end of each tentacle, one (Fig. 32b-d) to several toxicysts are situated (Fig. 32a). The protozoans float through the water with extended tentacles and catch prey organisms by using these toxicysts. The tentacles of Actinobolina smalii have an elaborate ultrastructure (Fig. 32d) (Holt and Corliss, 1973;'Holt et al., 1973). The extended tentacle averages 80-100 pm in length (Fig. 32c) and is limited by a unit membrane subtended by alveoli (a in Fig. 32d). Transverse sections through the tentacles display rings of microtubules (mt in Fig. 32d). In longitudinal view these microtubules are seen to penetrate into the endoplasm. Toward the tip the lumen of the rings is occupied by a toxicyst (to Fig. 32c and d). Loxophyllum meleagris shows another peculiarity regarding the position of toxicysts. This ciliate has protuberances (Fig. 33 a-c, arrows) filled with numerous toxicysts (Figs. 33d and e, and 34a) (Hausmann and Hausmann, 1973) of two kinds, a long, light type (tol in Fig. 34a) and a short, dark type (to, in Fig. 34a). It is possible that these two types are comparable to the two kinds of toxicysts inside the noselike proboscis of Didinium, the toxicysts sensu strict0 (to in Fig. 34b) and the pexicysts (p in Fig. 34b) (Wessen-

FIG.33. The protuberances in L. meleagris (arrows) disclose in thin sections numerous toxicysts (d and e). The dotted line in (d) and (e) indicates the plane of section through the protuberance. (a-c) x 125. (d and e) x 6500. From Hausmann and Hausmann, 1973.

EXTRUSIVE ORGANELLES IN PROTISTS

247

FIG. 34. Schematic comparison of the fine-structural features of the toxicyst protuberances ofLoxophyllurn (a) and the proboscis ofDidiniurn (b). f, Filaments; m, mucocyst; p, pexicyst; to, toxicyst. (b) From Wessenberg and Antipa, 1968.

248

KLAUS HAUSMANN

berg and Antipa, 1968; Yagiu and Shigenaka, 1965).Surprisingly, the toxicyst-containing warts of Loxophyllum have a fine structure quite similar to that of the proboscis of Didinium (compare Fig. 34a and b), except that the cytostome of Loxophyllum is situated directly opposite the warts, whereas it is the proboscis of Didinium itself that contains the cytostome.

FIG. 35. Schematics of the involvement of toxicysts in the prey-catching process of Didinium (a). The ciliate attacks a Purumeciurn (b) by firing pexicysts (p) and finally toxicysts (t). ci, Cilium; tr, spindle trichocysts. (a) From Wessenberg and Antipa, 1968; (b) after Mast, 1909; (c, d) from Wessenberg and Antipa, 1970.

EXTRUSIVE ORGANELLES I N PROTISTS

249

3. Function of Toxicysts in Ciliates In an excellent study the function of the toxicysts of Didinium were demonstrated on the electron microscope level (Wessenberg and Antipa, 1970) (Fig. 35).Didinium (Fig. 35a) is known to feed on Parumecium (Mast, 1909) (Fig. 3%). Feeding starts with the attachment of a Didinium to a Paramecium (Fig. 35b). A few milliseconds after contact is made between predator and prey the pexicysts (p in Fig. 35c) are discharged and become attached to the surface of the Puramecium. Just as the discharging toxicysts (t in Fig. 35c) begin to penetrate into the Paramecium, it in turn fires its spindle trichocysts (tr in Fig. 35c). A short while later the toxicysts have all been discharged, probably killing the prey (Fig. 35d). The feeding behavior of Didinium, which is easy to observe in the light microscope, convincingly demonstrates the function of the toxicysts: catching and killing prey. Similar observations, which can also b e demonstrated with microcinematography (Dragesco, 1960; Dragesco and MBtain, 1948; Grell, 1964), have been reported for many other ciliates (for reviews, see Dragesco, 1962; Dogiel, 1%5; Grell, 1973; Hall, 1953; Kudo, 1971; Pitelka, 1963; Sleigh, 1973; Westphal, 1974). Studies on the chemical nature of toxicysts have shown for seven species of ciliated protozoans that they are rich in acid phosphatase (FaurbFremiet, 1962, 1967). In this context the apparent similarities in structure, effect, and mode of function between the toxicysts of protozoa and the nematocysts of cnidaria should be mentioned (Brown, 1973). 4. Resting and Ejected Toxicysts in Ciliates Structurally the resting toxicysts are very complicated organelles which contain fully developed structures inside a capsule which can later be everted (Fig. 36). For example, tubes (arrowheads in Fig. 36b-e) lying within the capsule (c in Fig. 36b, c, and e) of both kinds of toxicysts in Loxophyllurn (all arrows in Fig. 36a) are filled with an ordered material which is probably the poison (0in Fig. 36c and e). Furthermore, an amorphous substance is detectable in the vicinity of the capsule wall (asterisk in Fig. 36b-e). During ejection the shorter type of toxicyst, found in Loxophyllum, everts a tube which is the same length as the capsule (Fig. 37a). Therefore in the resting state one tube with the same length as the capsule itself is housed within the capsule (arrowheads in Fig. 36b and c). The long type ejects a tube whose length is twice that of the

250

KLAUS HAUSMANN

FIG.36. Resting toxicysts of Loxophyllurn. (a) Between the two types of toxicysts (different arrows) mucocysts (mu) are seen. Higher magnifications of cross sections (b and d) and longitudinal views of negatively stained material (c and e) reveal the different internal structures (0, arrowheads, arrows, asterisks). c, Capsule. (a) x52,OOO. (b-e) x 120,000. From Hausmann and Wohlfarth-Bottermann, 1973.

capsule (Fig. 37b). So, in a cross section of the resting stage of this toxicyst, the one tube is detectable twice (arrowheads in Fig. 36d; arrowheads and arrows in Fig. 36e). The Prorodon toxicyst is even more complex. In its resting form the lumen of its tube, which is twice as long as the capsule, contains another tubule which is the length of the capsule (Fig. 39d).

EXTRUSIVE ORGANELLES IN PROTISTS

25 1

FIG.37. Ejected toxicysts ofloxophyllurn.(a)The shorter type secretes what appears to he a sticky material from a tube ( t ) .(h)The longer kind ejects a tube (t) which is twice the length of the capsule (c). (a) x 11,500. (h) x 5500. From Hausmann and WohlfarthBottermann, 1973.

5. Mode of Ejection of Toxicysts in Ciliates Two types of ejection mechanisms are found in toxicysts:

1.Telescopic discharge of a tube, as reported for the pexicysts ofDidinium (Fig. 3 9 4 (Rieder, 1971; Wessenberg and Antipa, 1970). 2. Discharge of a tubule via evagination, as shown for the toxicysts of Loxophyllum (Figs. 38a-c, and 39b and c) (Hausmann and Wohlfarth-Bottennann, 1973).

252

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

253

The ejection of the toxicysts ofProrodon and Dileptus (long type) is a combination of both mechanisms (Fig. 39d) (Hausmann and Bohatier, 1978; Hausmann and Wohlfarth-Botterman, 1973; Wohlfarth-Bottermann and Pfefferkorn, 1953a,b). During or near the end of toxicyst expulsion a sticky (Fig. 37a) and/or poisonous material is secreted by the tubules (Figs. 38d and e, and 39) (Hausmann and Wohlfarth-Bottermann, 1973). Kriiger (1934a, 1936), using the darkfield microscope, postulated different categories of toxicysts, basing his scheme on the ratio of the length of the different parts of the ejected extrusomes: capsule (l),tubule (2), and Fadenendstuck (end tube) (3). Three classes were described:

1. Capsule/tubule ratio 1: 1 (Fig. 39a and b). Examples are the pexicysts of Didinium and the short toxicysts of Loxophyllum and Di1ep tus. 2. Capsule/tubule ratio 1:2 (Fig. 39c). Examples are the long toxicysts of Loxophyllum (it has been adequately demonstrated that the seemingly bipartite everted tube is indeed one structure: Hausmann and Wohlfarth-Bottermann, 1973). 3. Capsule/tubule/end tube ratio 1:2 : 1 (Fig. 39d). Examples are the toxicysts of Prorodon and Dileptus (long type). Any deviations from these ratios, if they exist, have not been reported. The motive forces for the expulsion of the tubes are still unknown. One can speculate that the material in the vicinity of the capsule wall (asterisk in Fig. 36b-e) can serve such a function. However, since the chemical nature of this material is unknown, assigning a function to it is premature at this time. 6. Origin of Toxicysts All reports on the origin of toxicysts show that they are initially synthesized by the endoplasmic reticulum (Bohatier and Detcheva, 1973; Dragesco et al., 1965; Rieder, 1968a, 1971; Wessenberg and Antipa, 1968). After a process of differentiation and maturation (Fig. 40)

FIG.38. Tube ejection (t) occurs essentially through evagination [arrows and arrowheads in (a-c)]. After the discharge, filamentous and/or amorphous material is secreted (d and e). The toxicyst-containing warts of Lorophyllurn have a brushlike appearance after toxicyst discharge [arrows in (f and g)] (c) x 50,000. (d and e) x 35,000. (f) x 300. (9) x 1500.(c-e) From Hausmann and Wohlfarth-Bottermann,1973.

254

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

255

FIG.40. Different stages in the development of toxicysts in Litonotus (a and b) and Loxophyllum (c and d). (a, b, and c) x 18,000. (d) x 16,000.

they migrate to their predestined positions in the cortex of the protozoan. A recent study based on cytochemical methods reports D N A in the toxicysts of the ciliate Homalozoon vermiculare (Gautier and Fakan, 1974). If this is true, it would disagree with the principle of compartmentalization of the cell (Schnepf, 1966a,b), which states that active D N A can be located only in the nucleocytoplasmic matrix (Schnepf, FIG. 39. Schematic summary of the four ways in which toxicysts are known to discharge. (a) Telescopic discharge of a tube (t) which is the same length as the capsule (c) (example: Didinium, pexicyst). (b) Evagination of a tubule the same length as the capsule (examples: Loxophyllum and Dileptus, short toxicysts). (c) Evagination of a tube twice as long as the capsule (example: Loxophyllum, long toxicyst). (d) Evagination of a tube twice as long as the capsule combined with a telescopic discharge of a tubule which is the same length as the capsule (examples: Prorodon, Dileptus, long toxicyst).

256

KLAUS HAUSMANN

1977) and cannot be enclosed in vesicles, particularly those that will be excreted from the cell. 7. Structure and Function of Toxicysts in Flagellates Only one flagellate is known to possess toxicysts, the phagotrophic Colponema loxodes (Mignot and Brugerolle, 1975; Mignot and Hovasse, 1974-1975). The resting toxicysts are spindlelike structures which lie in an oval vacuole below the plasma membrane (arrows in Fig. 41a) and are composed o f a capsule (c in Fig. 41b) surrounding a

FIG.41. The toxicysts of the phagotrophic flagellate C . lorodes lie below the plasma membrane [arrows in (a)]. The main structures of the toxicysts are a tube (t) and a capsule (c). During food capture the toxicysts are discharged and a tube is evaginated [arrow in (c)]. p, Prey. (a) x 14,000. (b) ~80,000. (c) x 27,000. (a) From J.-P. Mignot; (b, c) from Mignot and Brugerolle, 1975.

EXTRUSIVE ORGANELLES IN PROTISTS

257

tube (t in Fig. 41b). This tube can be discharged, as is seen in a feeding flagellate (arrow in Fig. 41c). Although connections between toxicysts and prey have never been seen (p in Fig. 41c), the involvement of these toxicysts in food catching is obvious.

8. Structure and Function of Haptocysts in Suctorians The knob of many suctorian tentacles is equipped with a highly complex organelle, the haptocyst (missile-like body or phialocyst) (ha in Fig. 42a, b, and d) (Bardele, 1970, 1972a, 1974; Bardele and Grell, 1967; Batisse, 1965,1966,1967,1972; Curry and Butler, 1976; Hauser, 1970; Hauser and van Eys, 1976; Hitchen and Butler, 1974; Jurand and Bomford, 1965; Lom and Kozloff, 1967; Rudzinska, 1965, 1970, 1973; Spoon et al., 1976; Tucker and Mackie, 1975).The knob is covered by the plasma membrane (pm in Fig. 42d). The protoplasmic fracture face of this membrane (PF-pm in Fig. 42c) contains randomly distributed intramembranous particles with an average density of 2300/pm2. Highly ordered membrane domains can be seen where the haptocysts are attached to the plasma membrane (arrowhead in Fig. 42c). A rosette of 12-nm-diameter particles with a large particle in the center is found above every haptocyst. The 60-nm-diameter rosette is surrounded b y an annular area, 40 nm wide and free of particles. According to Bardele (1976b) the rosette and annulus can be regarded as a multifunctional membrane differentiation having both receptor and effector functions, that is, recognition of prey, as well as involvement in rearrangement of the membrane components to allow an orderly discharge of the haptocysts. Bardele (1976b) stresses that the general order and apparent rigidity of particle arrangements found at the attachment sites of various extrusomes (e.g., spindle trichocysts, mucocysts, kinetocysts, and haptocysts) may be important in preventing their untimely discharge (see Section IV,B,8). Haptocysts are discharged on contact with the prey and, by puncturing the pellicle of the prey, give rise to a firm connection between the two cells (arrows in Fig. 42e). As compared with that of undischarged haptocysts (Fig. 42d), the altered structure of discharged organelles (Fig. 42e) indicates that part of their contents has been injected into the prey. The complex structure of the haptocyst, with its several parts, suggests the presence of several enzymes which may be responsible for puncturing the prey’s pellicle, abolishing ciliary motion, and producing local solubilization of the prey’s cytoplasm (Bardele, 1969, 1974; Batisse, 1967; Evans, 1953; Rudzinska, 1973). Moreover, a very important function of the haptocysts is their role in the “fusion” of the

258

KLAUS HAUSMANN

EXTRUSIVE ORGANELLES IN PROTISTS

259

plasma membranes of predator and prey (Bardele, 1974). The discharge of haptocysts may occur within milliseconds. At this point other organelles, characteristically found in the suctorian’s knob, should also be mentioned-the solenocysts (Hitchen and Butler, 1973) or osmiophilic granules (Bardele and Grell, 1967; Hauser, 1970). It has been suggested that these particles, containing electron-opaque, ordered material (Bardele and Grell, 1967; Batisse, 1968) may represent membrane precursors used to form the vacuole around the newly ingested food. Whether or not these organelles fit into the category of extrusomes is not clear. However, Hitchen and Butler (1973) interpret their electron micrographs as showing a blebbing activity of the solenocysts at the tip of the feeding tentacle of Choanophrya. After discharge the haptocysts must be replaced. Large areas showing numerous forming haptocysts were found in the cytoplasma of the cell (Fig. 42f) (Bardele, 1970). Mature haptocysts move upward through the tentacle, where they find their way to certain regularly distributed anchoring sites in the membrane of the knob, probably in a manner similar to the way other types of extrusomes find their anchoring sites.

D. RHABDOCYSTS In lower marine ciliates, for example, in Trachelonema sulcata and Tracheloraphis dogieli, a special extrusome is found, the rhabdocyst (Raikov, 1971-1972, 1974; Raikov et al., 1975).The internal structure of a rhabdocyst consists of an apex (a in Fig. 43), a tube (t in Fig. 43), a dark band (db in Fig. 43), and a vesicular basal part (bp in Fig. 43). The organelle is covered by a membrane (m in Fig. 43). The mechanism of ejection resembles to some extent that of toxicysts (Fig. 39a), which involves the telescopic discharge of a tube (2 and 3 in Fig. 43). The process is thought to be initiated by a swelling of the basal vesicle. The pressure thus produced can be transferred to the dark band at the end of the tube which is ejected like an arrow being expelled from a blowpipe (I. B. Raikov, personal communicaFIG.42. Haptocysts ofAcineta (a, d, e, and f ) and Discophrja (b and c). In the knob of the tentacle, haptocysts (ha) underlie the plasma membrane (pm). In freeze-fracture preparations the P face of the plasma membrane (PF-pm) shows rings of particles over each haptocyst (arrowhead). A tentacular necklace [arrows in (c)] composed of tightly packed particles indicates the end of the alveoli. The haptocysts make contact between the prey and the tentacle [arrows in (e)]. Generative areas of haptocysts exist within the cell (f). (a) x 17,500. (b) x 15,000. (c and f) ~ 4 5 , 0 0 0 .(d) ~ 2 6 , 0 0 0 (e) . ~ 2 4 , 0 0 0 .(a, e) From Bardele and Grell, 1967; (c) from C. F. Bardele; (d, f ) from Bardele, 1970.

260

KLAUS HAUSMANN

L_ _ _ _ _ _ _ _ _ _ _ J FIG.43. Schematic interpretation of the mode of extrusion of rhabdocysts. A tube (t) with a specially constructed apex (a) and a dark band (db) near the proximal end (1) is ejected (2) by swelling of the proximal part (bp) of the extrusome (3).The organelle is covered by a membrane (m).

tion). The morphology of the tube changes little during the extrusion; it becomes only slightly longer and thinner (3 in Fig. 43). The function of the rhabdocyst is entirely unknown.

E.

EJECTISOMES

(TAENIOBOLOCYSTS)

1. Structure of Resting and Discharged Ejectisomes Ejectisomes are types of extrusomes which occur exclusively in flagellates (for reviews, see Dodge, 1974; Hall, 1946; Hovasse, 1965a, 1969; Hovasse and Mignot, 1975). They are present in most Cryptophyceae (Anderson, 1962; Dodge, 1969; Dragesco, 1951; Hovasse 1965b; Joyon, 1963; Kriiger, 1934b; Mignot, 196513; Mignot et al., 1968,1970; Schuster, 1968,1970; Wehrmeyer, 1970) and in a few Prasinophyceae (Manton, 1969; Norris and Pearson, 1975). Undischarged ejectisomes are situated mainly toward the anterior end of the cell adjacent to the gullet (Fig. 44a and b). In some organisms small ejectisomes are also located around the periphery of the

EXTRUSIVE ORGANELLES IN PROTISTS

26 1

FIG. 44. Ejectisomes. In Chilomonas paramecium the ejectisomes are situated adjacent to the gullet (T).The resting ejectisomes are bipartite structures (c) composed of a tape coiled into tight spirals (d). After ejection the extrusome is a needlelike structure with a sharply bent tip [arrows in (e and f)]. The mechanism of discharge is an unrolling of the tape [arrows in (g)]. AM, Amphosome; B, basal body; C, cortical area; CV,contractile vacuole; F, flagellum; N, nucleus; NCL, nucleolus; P, inpocketing of plasma membrane; PB, parabasal body; PM, paramylum bodies; R, rhizoplast; T, trichocysts (ejectisomes); V, vestibulum. (b) X37,OOO. (c) ~ 2 5 , 0 0 0 (d . and f ) x40,OOO. (e) x 6500; (9) X 50,000. (a) From Anderson, 1962; (b-d) from Hovasse et al., 1967; (g) from Mignot et, al., 1970.

cell, adjacent to grooves in the periplast (Dodge, 1969; Mignot, 1965b). The internal structure of ejectisomes consists of two parts: a large cylinder composed of a tapelike structure coiled into a tight spiral (of about 50 turns), which surrounds a narrow, cone-shaped canal

262

KLAUS H A U S M A N N

(Fig. 44c and d), and a second part of similarly coiled material (arrow in Fig. 44c), which is connected to the first structure (Fig. 45a). These extrusomes have a proteinaceous nature (Mignot and Hovasse, 1973). The discharged ejectisome, first investigated by Anderson (1962), is a hollow tube (Figs. 44e and 45c-e) with a sharply bent tip (arrows in Fig. 44e and f; Fig. 45b and e), with the exception that in Prasinophyceae ejectisomes do not have a bipartite structure but are single tubes after discharge. 2. Mode of Function and Regeneration of Ejectisomes The possible mechanisms by which ejectisomes form a tubular structure after discharge have been reviewed by Hovasse et al. (1967). The generally accepted theory of the expulsion of ejectisome is shown in Fig. 45. The coiled extrusome is unrolled (arrow in Fig. 44a; Fig. 45b), and a hollow tube is formed as the tape rolls up laterally (Fig. 45c and d ) (Mignot et al., 1970; Hovasse, 1965a; Hovasse and Mignot, 1975; Hovasse et al., 1967). The development of ejectisomes has been studied by several inves-

FIG. 45. Schematic reconstruction of the discharge of ejectisomes. The resting stage is a coiled ribbon (a) which unrolls during ejection (b) and forms a hollow tube by rolling up laterally ( c and d). This results in a needlelike structure (e).(a, c-e) From Hov a s e et d.,1967; (b) after Hovasse and Mignot, 1975.

EXTRUSIVE ORGANELLES IN PROTISTS

263

tigators (Anderson, 1962; Schuster, 1970; Wehrmeyer, 1970). It has been shown that vesicles derived from the Golgi apparatus contain an extrusomal tape of only few turns. During further maturation the number of turns increases, and a bipartite structure is then formed. 3. Comparison of Ejectisomes with the R Bodies of the Kappa Particles of Paramecium Ejectisomes are curious and almost unique structures. The closest comparable structures known seem to be the R bodies of the kappa particles (symbiotic bacteria) found in Paramecium aurelia (Hovasse, 1965a; Hovasse et al., 1967; Preer et al., Soldo, 1974).The R body is a ribbon which in its normal compact form is wound into a tight roll consisting of several turns, just like an ejectisome. On heating to 60°C or treating with sodium dodecyl sulfate or phosphotungstate, the rolled ribbon of a stock-1039 R body suddenly and irreversibly unrolls into a long tape. The ribbon of a stock-51 R body unrolls only in response to lowering the pH below 6. In this case the process is reversible and, when the pH is raised above 7, rerolling occurs (Preer et al., 1966). Preer et al. (1974) stated that the difficulty in imagining how convergent evolution could have produced such bizarre and similar structures in such widely different organisms is matched only by the problem of imagining how an apparently bacterial structure can be phylogenetically related to an algal structure.

F. DISCOBOLOCYSTS Discobolocysts are restricted to flagellates, especially those of the order Chrysophyceae. The species known to have this type of extrusome are Chromulina georgesiana (Bourrelly, 1957),Cyclonexis annularis (Hovasse, 1948, 1949), Ochromonas crenata (Conrad, 1926; Klebs, 1893; Kalina, 1964), Ochromonas hovassei (Bourrelly, 1957), and Ochromonas tuberculatus (Hibberd, 1970). In the resting stage the organelles in 0. tuberculatus are almost spherical (Fig. 46b; 4 in Fig. 47). The part containing a disk with approximately 20 radial canals and a central hole (d in Fig. 46b; Fig. 46c and d) protrudes above the general level of the cell surface (di in Fig. 46a). The extrusome, bounded by a unit membrane (arrows in Fig. 46b) is filled with a fibrous, reticulate material (asterisk in Fig. 4%) in the space not occupied by the apical disk. After ejection the disk of the discobolocyst seems to be unaltered in fine structure (d in Fig. 46e), whereas the rest of the organelle is transformed into a long tail (t in Fig. 46e; 5 in Fig. 47). The tail consists of an unordered, fibrous material.

264

KLAUS HAUSMANN

FIG.46. Discobolocysts of the Chrysophycea Ochromonas tuberculatus. The organelles protrude above the general level of the cell surface (di). The organelles, consisting of a disk (d) and an amorphous portion (asterisk),are enveloped by a membrane (arrows). After discharge the extrusomes show a long tail (t) composed of unordered filaments. ch, Chloroplast; cv, contractile vacuole; Ga, Golgi apparatus; lv, leukosine vesicle; nu, nucleus. (a) x 7500. (b) x 20,000. (c and d ) x 10,000. (e) x 8000. From Hibberd, 1970.

The ontogeny of discobolocysts starts with balloon-like vesicles derived from the Golgi apparatus (1in Fig. 47). These vesicles, filled with reticulate contents (2 in Fig. 47), increase in size to 250-500 nm in diameter (3 in Fig. 47) and then migrate between the plastids (3 in Fig. 47) to the surface of the cell where they develop into typical discobolocysts (4 in Fig. 47). When development is complete, the fully

EXTRUSIVE ORGANELLES IN PROTISTS

265

FIG.47. Cycle of discobolocysts. Vesicles (l),originating from the Golgi apparatus (Ca) and filled with a reticulate material ( 2 ) ,increase in size to become large vacuoles (3). These migrate between the chloroplasts (ch) to the plasma membrane, where they develop into their typical form (4), ready to be discharged (5).

formed organelles presumably move to positions at the periphery of the cell (4 in Fig. 47), since typical discoboloysts are always seen in the cytoplasm between the plastid and the cell membrane. The function of these organelles is unknown. G. NEMATOCYSTS(CNIDOCYSTS) Nematocysts, extrusomes found in dinoflagellates such as PoZykrikos (Chatton and Grass6, 1929; Faur6-Fremiet, 1913; Hovasse, 1963) and Nematodinium (Hovasse, 1951b; Mornin and Francis, 1967), are basically small capsules filled with a coiled tube (n in Fig. 48a; 1 in Fig. 48b). This tube is furnished at the tip with a stylet and can be discharged by a process of evagination (Fig. 48b) (Chatton, 1914). The organelle may have a defensive function. It was first suggested that these extrusomes are self-duplicating organelles and should consequently have a fairly complicated developmental cycle (Chatton and Hovasse, 1944; Hovasse, 1951a,b). However, recent electron microscope observations do not support this idea (Greuet, 1971,1972; Greuet and Hovasse, 1977).They are formed by a normal developmental process which is complicated only because of the morphological complexity of the mature cnidocyst. OF CNIDOSPORIDIANS H. POLARFILAMENT The polar filament of cnidosporidians is not an extrusome by definition (see Section 1,A). It is mentioned in this article because of its extrusive properties.

266

KLAUS HAUSMANN

I

'

a

FIG.48. Polykrikos schwartzi with nematocysts (cnidocysts) (n).The coiled tube inside a capsule (1)is ejected by evagination (2 and 3). K, Nucleus; p, parabasal body; t, trichocyst. (a) After Chatton and GrassB, 1929.

Myxosporidians and microsporidians are parasites with complicated life cycles. They infect their host as small ameboid organisms which emerge from ingested spores and multiply within the host. Eventually they differentiate to form characteristic spores which contain ameboid sporoplasm. These spores are provided with a tubelike filament, the polar filament, which can be extruded from the spore by a process of sudden evagination. The extruded tube retains morphological contact

EXTRUSIVE ORGANELLES IN PROTISTS

267

with the spore, making it possible for the sporoplasm to infect new cells. In microsporidians the motive force for the extrusion of the tube is located in two special organelles, the polaroplast and the posterior vacuole. These parts of the spore are thought to be capable of swelling, which generates a sudden pressure and causes evagination of the tube (for review, see VQvra, 1977). I. VAFUA Other extrusomes have also been reported in protozoa, but they cannot be classified, since they have seldom been seen or have been poorly described, for example, in Discotricha papillifera (Tuffrau, 1954) and Urostyla cristata (Fig. 49) (Jerka-Dziadosz, 1964, 1965, 1967, 1970; Suganuma, 1973; Weinke, 1972). However, there are some very well-known protozoan organelles which may be extruded but for which proof of extrusion is still lacking, for example, in Blepharisrna (Giese, 1973), Petalotricha (Laval, 1971, 1972), and Strombidium (FaurBFremiet and Ganier, 1970). Further detailed studies will be necessary to fill this gap.

V. Conclusions This article has attempted to show the structural and functional diversity of extrusomes in protozoa. Some have a wide distribution in flagellates as well as in ciliates (spindle trichocysts, mucocysts, and toxicysts); others are restricted to small systematic groups, for example, the discobolocysts of the flagellate order Chrysophyceae or the cnidocysts of a few species of dinoflagellates. Surprisingly, some enigmatic similarity exists between extrusomes and organelles of systematically widely differing organisms such as ejectisomes of algae and

FIG.49. This unique extrusome of the ciliate Urostyla cristata cannot be placed in one of the categories of extrusomes, since not enough is known about its method of expulsion. x 26,000. From Suganuma, 1973.

268

KLAUS HAUSMANN

the R bodies of the bacterial symbionts of Paramecium or the toxicysts of ciliates and the nematocysts of cnidarians. The physiological function of extrusomes is evident only in a few cases, for example, toxicysts are involved in food capture and mucocysts probably in encystment; the significance of most extrusomes is still obscure. Even the function of the well-known spindle trichocyst in Paramecium has not been clarified. Therefore further studies on extrusomes should focus on investigating the physiology of these organelles rather than on cataloging further morphological variations. ACKNOWLEDGMENTS

Without the help of my wife, Dr. E. Hausmann, this article would not have been written. I thank Dr. Richard D. Allen, University of Hawaii at Honolulu, for his help in translating the article and for his critical review of the manuscript. Collaborations and discussions with the following scientists are appreciated (the names of those who kindly provided illustrations are followed by an asterisk): R. D. Allen*, R. Anderer, G. Antipa, C. F. Bardele*, J. Bohatier, J. 0. Corliss, G. Deichgraber, P. Didier*, K. G. Grell, G. de Haller, M. Hauser*, J. Hibberd*, H. Hoffmann-Berling, P. Holt*, R. Hovasse, N. Hiilsmann, Fr. Kriiger, J. Lom, J. -P. Mignot*, D. J. Patterson, R. K. Peck*, D. Pitelka, P. de Puytorac, J. Raikov, B. Satir*, E. Schnepf, Y. Suganuma*, A. Tiedtke*, D. Troyer*, J. Vavra [I gratefully acknowledge Dr. VQvra’skindness in allowing me to see the unpublished manuscript for his article, “The Fine Structure of Microsporidia,” in The Microsporidia (J. V5vra and V. Sprague, Eds.) to be published by Plenum Press, New York], H. Wessenberg*, and K. E. Wohlfarth-Bottermann. I am indebted for technical assistance to P. Batta, B. Koeppen, D. Laupp, K. -L. Medved, A. Riiskens, M. Sauernheimer, and M. Ueno. Financial support was given in part by the Deutsche Forschungsgemeinschaft, BonnBad Godesberg, West Germany. REFERENCES Afzelius, B. A. (1956). Erp. Cell Res. 10,257. Alexander, J. B. (1968). E x p . Cell Res. 49,425. Allen, R. D. (1967).J . Protozool. 14,553. Allen, R. D. (1971).J . Cell Biol. 49, 1. Allen, R. D., and Hausmann, K. (1976).J.Ultrustruct. Res. 54,224. Allen, R. D., and Wolf, R. W. (1974).J.Cell Sci. 14,611. Allison, A. C., and Davies, P. (1974).In “Transport at the Cellular Level” (M. A. Sleigh and D. H. Jennings, eds.), p. 419. Cambridge Univ. Press, London and New York. Allman, G. J. (1855). Q. J. Microsc. Sci. [N.S.] 3, 177. Anderer, R., and Hausmann, K. (1977),J . Ultrastruct. Res. 60,21. Anderson, E. (1962).J . Protozool. 9,380. Arnott, H. J., and Walne, P. L. (1967).Protoplasma 64,330. Bachmann, L., Schmitt, L., and Plattner, H. (1972).Proc. Eur. Congr. Electron Microsc., 244.

EXTRUSIVE ORGANELLES IN PROTISTS

269

Bannister, L. H. (1972).]. Cell Sci. 11, 899. Bardele, C. F. (1969).Z . Naturforsch., Teil B 24, 362. Bardele, C. F. (1970).]. Protozool. 17, 51. Bardele, C. F. (1971). 29th Annu. Proc. A m . Electron Microsc. SOC. p. 334. Bardele, C. F. (1972a).Z. Zellforsch. Mikrosk. Anat. 126, 116. Bardele, C. F. (1972b). Z. Zellforsch. Mikrosk. Anat. 130,219. Bardele, C. F. (1974).I n “Transport at the Cellular Level” (M. A. Sleigh and D. H. Jennings, eds.), P. 191. Cambridge Univ. Press, London and New York. Bardele, C. F. (1975). Cell Tissue Res. 161,85. Bardele, C. F. (1976a). Z . Naturforsch. Teil C 31, 190. Bardele, C. F. (197613).J . Protozool. 23, Suppl., 32A. Bardele, C. F., and Grell, K. G. (1967).2. Zellforsch. Mikrosk. Anat. 80, 108. Barras, D. R., and Stone, B. -4. (1968).In “The Biology ofEuglena” (D. E. Buetow, ed.), Vol. 2, p. 159. Academic Press, New York. Batisse, A. (1965). C.R. Hebd. Seances Acad. Sci. 261, 5629. Batisse, A. (1966). C.R. Hebd. Seances Acad. Sci. 262,771. Batisse, A. (1967). C.R. Hebd. Seances Acad. Sci. 265, 972. Batisse, A. (1968).Protistologica 4,271. Batisse, A. (1972).Protistologica 8,477. Beisson, J., Lefort-Tran, M., Pouphile, M., Rossignol, M., and Satir, B. (1976).J . Cell Biol. 69, 126. Beyersdorfer, K. (1951). Z. Naturforsch., Teil B 6, 57. Beyersdorfer, K., and Dragesco, J. (1952a). Congr. Microsc. Electron. (Paris) p. 655. Beyersdorfer, K., and Dragesco, J. (1952b). Congr. Microsc. Electron. (Paris)p. 661. Bohatier, J. (1970). Protistologica 6,331. Bohatier, J. (1972). Protistologica 8,439. Bohatier, J., and Detcheva, R. (1973). C.R. Seances SOC. Biol. Ses Fil. 167, 972. Bohatier, J., and Njine, T. (1973).Protistologica 9,359. Bouck, B. C., and Sweeney, B. M. (1966).Protoplasma 61,205. Bourrelly, P. (1957).Reu. Algol., Mem. Hors Ser. 1, 1. Branton, D., Bullivant, St., Gilula, N. B., Karnovsky, M. J., Moor, H., Muhlethaler, K., Northcote, D. H., Packer, L., Satir, B., Satir, P., Speth, V . , Staehelin, L. A., Steere, R. S., and Weinstein, R. S. (1975). Science 190, 54. Bresslau, E. (1921a). Naturwissenschaften 9,57. Bresslau, E. (1921b).Verh. Dtsch. Zool. Ges. 26,35. Bresslau, E. (1923). Mikrokosmos 16,97. Bresslau, E . (1924). Verh. Dtsch. 2001.Ges. 29, 91. Bresslau, E. (1928).Arb. Staatsinst. Exp. Ther. 21, 26. Bretschneider, L. H. (1950).Mikroskopie 5,257. Brown, F. A., Jr. (1973). In “Comparative Animal Physiology” (C. L. Prosser, ed.), p. 909. Saunders, Philadelphia, Pennsylvania. Cachon, J., Cachon, M., and Greuet, C. (1974-1975).Ann. Stn. Biol. Besse-en-Chandesse 9, 177. Chadefaud, M. (1936). Ann. Protistenkd. 5,323. Chapman-Andresen, C. (1972).In “The Biology of Amoeba” (K. W. Jeon, ed.), p. 319. Academic Press, New York. Chapman-Andresen, C., and Nilsson, J. R. (1968). C.R. Trau. Lab. Carlsberg 36,405. Chatton, E. (1914).Arch. Zool. Exp. Gen. 54, 157. Chatton, E., and G r a d , P.-P. (1929). C.R. Seances Sac. Biol. Ses Fil. 100,281. Chatton, E., and Hovasse, R. (1944).C.R. Hebd. Seances Acad. Sci. 218,60.

270

KLAUS HAUSMANN

Cheissin, E. M., and Mosevich, T. N. (1962).Arch. Protistenkd. 106, 181. Conrad, W. (1926).Arch. Protistenkd. 72, 538. Corliss, J. 0. (1958).J. Protozool. 5, 184. Curry, A., and Butler, R. D. (1976).J. Ultrastruct. Res. 56, 164. Davidson, L. (1974). Ph.D. Dissertation, University of California, Berkeley. Davidson, L. (1976). Cell Tissue Res. 170,353. de Haller, G., and ten Heggeler, B. (1969).Protistologica 5, 115. de Puytorac, P. (1964a).Acta Protzool. 2, 147. de Puytorac, P. (1964b). C.R. Seances Soc. Biol. Ses F i l . 158,526. de Puytorac, P. (1969). C.R. Hebd. Seances Acad. Sci. 268,820. de Puytorac, P., and Grain, J. (1968).Protistologica 4,405. de Puytorac, P., and Kattar, M. R. (1969).Protistologica 5, 549. de Puytorac, P., and Rodrigues de Santa Rosa, M. (1975). Protistologica 11,379. Didier, P., and Detcheva, R. (1974). Protistologica 11, 159. Dodge, J . D. (1969).Arch. Mikrobiol. 69,266. Dodge, J . D. (1973). “The Fine Structure of Algal Cells.” Academic Press, New York. Dodge, J. D., and Crawford, R. M. (1968).Protistologica 4,231. Dodge, J. D., and Crawford, R. M. (1971).Protistologica 7,295. Dogiel, V. A. (1965). “General Protozoology,” 2nd ed. Oxford Univ. Press (Claredon), London and New York. Douglas, W. W. (1974). Biochem. Soc. Symp. 39, 1. Ihagesco, J . (1951). Bull. Microsc. Appl. 1, 9. Dragesco, J. (1952a).Bull. Microsc. Appl. 2, 92. Dragesco, J . (1952b). Bull. Microsc. Appl. 2, 148. Dragesco, J . (1960).Inst. Wiss. Film Goettingen Film W439. Dragesco, J . (1962). Bull. Biol. Fr. Belg. 96, 123. Dragesco, J. (1968).Protistologica 4, 157. Dragesco, J., and Hollande, A. (1965). C.R. Hebd. Seances Acad. Sci. 260,2073. Dragesco, J., and Mktain, C. (1948). Bull. Soc. Zool. Fr. 73, 62. Dragesco, J., Auderset, G., and Baumann, M. (1965).Protistologica 1,81. Dumont, J. N. (1961).J. Protozool. 8,392. Ehret, C. F., and de Haller, G. (1963).J.Ultrastruct. Res. 6, Suppl., 1. Ehret, C. F., and McArdle, E. W. (1974). In “Paramecium-A Current Survey” (W. J. van Wagtendonk, ed.), p. 263. Elsevier, Amsterdam. Ehret, C. F., and Powers, E. L. (1959).Int. Reu. Cytol. 8, 97. Elliot, A. M. (1973). In “Biology of Tetrahymena” (A. M. EIIiott, ed.), p. 76. Dowden, Hutchinson & Ross, Stroudsburg. Ellis, J . (1769).Philos. Trans. R. SOC. London 69, 138. EstBve, 1.-C. (1974).Protistologica 10,371. Evans, F. R. (1953).Trans. Am. Microsc. Soc. 72, 171. Faurk-Fremiet, E. (1913).Bull. Soc. Zool. Fr. 38,289. Faure-Fremiet, E. (1962). C.R. Hebd. Seances Acad. Sci. 254,2691. FaurC-Fremiet, E. (1967). Chem. Zool. 1,21. FaurC-Frerniet, E., and AndrB, J. (1967).J. Protozool. 14,464. FaurB-Fremiet, E., and Ganier, M.-C. (1969).Protistologica 5,353. Faurk-Fremiet, E., and Ganier, M.-C. (1970).Protistologica 6,207. Fisher, G., Kaneshiro, E. S., and Peters, P. D. (1976).J . Cell Biol. 69,429. Foissner, W., and Simonsberger, P. (1975).Protoplasma 86,65. Fritzsche, R. (1911).Arch. Hydrobiol. 6, 99. Gautier, A., and Fakan, J. (1974).J. Microsc. (Paris) 21, 197.

EXTRUSIVE ORGANELLES IN PROTISTS

271

Giese, A. C. (1973). “Blepharisma-The Biology of a Light-Sensitive Protozoan,” 1st ed. Stanford Univ. Press, Stanford, California. Golinska, K. (1974).Acta Protozool. 12,289. Golinska, K., and Grain, J. (1969). Protistologica 5, 447. Grain, J. (1968a).C.R. Hebd. Seances Acad. Sci. 266, 1511. Grain, J. (1968b).J. Microsc. (Paris) 7,993. Grain, J. (1970). Protistologica 6, 37. Grain, J., and Golinska, K. (1969). ProtistoZogica 5, 269. Grasse, P.-P., and Mugard, H. (1961).C.R. Hebd. Seances Acad. Sci. 253,31. Grell, K . G. (1964). Znst. Wiss. Film Gdettingen Film C882. Grell, K. G. (1973). “Protozoology.” Springer-Verlag, Berlin and New York. Greuet, C. (1969). Ph.D. Dissertation, University of Nice. Greuet, C. (1971).Protistologica 8,345. Greuet, C. (1972). C.R. Hebd. Seances Acad. Sci. 275, 1239. Greuet, C., and Hovasse, R. (1977). Protistologica 13, 145. Grimstone, A. V. (1961). Biol. Rev. Cambridge Philos. Soc. 36,97. Hall, R. P. (1946).Bot. Reu. 12, 515. Hall, R. P. (1953). “Protozoology.” Prentice-Hall, Englewood Cliffs, New Jersey. Hauser, M. (1970).Z. Zellforsch. Mikrosk. Anat. 106,584. Hauser, M., and van Eys, H. (1976).J. Cell Sci. 20, 589. Hausmann, E., and Hausmann, K. (1973).Protistologicu 9, 139. Hausmann, K. (1971). Mikrokosmos 60,322. Hausmann, K. (1972a).Mikrokosmos 61, 114. Hausmann, K. (197213).Mikrokosmos 61, 134. Cytobiologie 5,468. Hausmann, K. (1972~). Hausmann, K . (1972d). Protistologica 8,401. Hausmann, K . (1973a).Helgol. Wiss. Meeresunters. 25,39. Hausmann, K . (1973b).J. Microsc. (Paris) 17, 199. Protistologica 9,235. Hausmann, K. (1973~). Hausmann, K . (1973d).Ann. Stan. Biol. Besse-en-Chandesse 8,331. Hausmann, K . (1974).Microsc. Acta 76, 113. Hausmann, K. (1977a).Protoplasma 92,263. Hausmann, K. (197713).Arch. Protistenkd. 119, 233. Hausmann, K., and Allen, R. D. (197q.J. Cell Biol. 69,313. Hausmann, K., and Bohatier, J. (1978).Ann. Stun. B i d . Besse-en-Chandesse (in press). Hausmann, K., and Mignot, J.-P. (1975).Protoplasma 83, 61. Hausmann, K., and Mignot, J.-P. (1977).Protistologica 13,213. Hausmann, K., and Mocikat, K.-H. (1976).Cytobiologie 13,469. Hausmann, K., and Stockem, W. (1973).Microsc. Acta 74, 110. Hausmann, K., and Wohlfarth-Bottermann, K. E. (1973). Z. Zellforsch. Mikrosk. Anat. 140, 235. Hausmann, K., Stockem, W., and Wohlfarth-Bottermann, K. E. (1972a).Cytobiologie 5, 208. Hausmann, K., Stockem, W., and Wohlfarth-Bottermann, K. E. (1972b).Cytobiologie 5, 228. Hayashi, M. (1974).Cytobiologie 9,460. Hayes, M. L. (1938).Truns. A m . Microsc. Soc. 57, 11. Hibberd, D. J. (1970). Br. Phycol. J . 5, 119. Hibberd, D. J. (1971).Br. Phrycol. J. 6,207. Hitchen, E. T., and Butler, R. D. (1973).Z. Zellforsch. Mikrosk. Anat. 144, 37.

272

KLAUS HAUSMANN

Hitchen, E. T., and Butler, R. D. (1974).J. Ultrastruct. Res. 46,279. Hoffmann-Berling, H. (1960).Comp. Biochem. 2,341. Hoffmann-Berling, H. (1961).Ergeb. Physiol., B i d . Chem. E x p . Pharmakol. 51,98. Holt, P. A., and Chapman, G. B. (1971).J. Protozool. 18,604. Holt, P. A., and Corliss, J. 0. (1973).].Cell B i d . 58,213. Holt, P. A., Lynn, D. H., and Corliss, J. 0. (1973).Protistologica 9,521. Hovasse, R. (1948).C.R. Hebd. Seances Acad. Sci. 226, 1038. Hovasse, R. (1949).Botaniste 34,243. Hovasse, R. (1951a). Arch. Zool. E x p . Gen. 87,299. Hovasse, R. (1951b).Arch. Zool. E x p . Gen. 88, 149. Hovasse, R. (1963).Arch. Zool. E x p . Gen. 102, 189. Hovasse, R. (1965a).Protoplasmatologia 3, F, 1. Hovasse, R. (196%). C.R. Hebd. Seances Acad. Sci. 261,2947. Hovasse, R. (1965~). Protistologica 1,81. Hovasse, R. (1969).Ann. Stn. Biol. Besse-en-Chandesse 4,245. Hovasse, R.,and Mignot, J.-P. (1975).Annee. Biol. 14,397, Hovasse, R.,Mignot, J.-P., and Joyon, L. (1967).Protistologica 3,241. Hufnagel, L.A. (1969).J . Cell B i d . 40,779. Inaba, F.,Nakamura, R., and Yamaguchi, S. (1958).Cytologia 23,72. Jacobson, . (1931). Arch. Protistenkd. 75,31. Jahn, T. L., and Bovee, E. C. (1967).In “Research in Protozoology” (T.-T. Chen, ed.), Vol. 1, p. 87.Pergamon, Oxford. Jakus, M. A. (1945).].E r p . Zool. 100,457. Jakus, M. A,, and Hall, C. E. (1946).B i d . Bull. 91, 141. Janisch, R. (1972).J. Protozool. 19,470. Jerka-Dziadosz, M. (1964). Acta Protozool. 2, 123. Jerka-Dziadosz, M. (1965). Acta Protozool. 3, 133. Jerka-Dziadosz, M. (1967).Acta Protozool. 5, 59. Jerka-Dziadosz, M. (1970). Acta Protozool. 7,505. Joyon, L. (1963).Ann. Fac. Sci. Uniu.Clermont 22, 1. Jurand, A., and Bomford, R. (1965).J . Microsc. (Paris) 4, 509. Jurand, A., and Saxena, D. M. (1974).Acta Protozool. 12,307. Jurand, A,, and Selman, G. G . (1969).“The Anatomy of Paramecium aurelia,” MacmilIan, New York. Kahl, A. (1926).Arch. Protistenkd. 55, 197. Kalina, T.(1964).Acta Uniu. Carol. Biol. 2, 149. Kattar, M. R. (1972).Protistologica 8, 135. Kawakami, H. (1971).J . Sci. Hiroshima Uniu.,Ser. B , Diu. 1 23, 121. Kawakami, H., and Yagiu, R. (1960).Bull. B i d . Soc. Hiroshima Uniu. 27, 11. Kawakami, H., and Yagiu, R. (1963a). Zool. Mag, 72,89. Kawakami, H., and Yagiu, R. (1963b).Zool. Mag. 72, 146. Kawakami, H., and Yagiu, R. (1963~). Zool. Mag. 72,224. Kawakami, H., and Yagiu, R. (1964a).Zool. Mag. 73,33. Kawakami, H., and Yagiu, R. (1964b).Zool. Mag. 73, 78. Kawakami, H.,and Yagiu, R. (1964~). Zool. Mag. 73, 112. Kennedy, J. R. (1965).J . Protozool. 12, 542. Khainsky, A. (1911).Arch. Protistenkd. 21, 1. Klein, B. M. (1952).Mikrokosmos 41,267. Kink, J. (1973). Acta Protozool. 12, 173. Klebs, G . (1893).Z. Wiss. Zool. 55,353. Knoch, M., and Konig, H. (1951).Naturwissenschaften 38, 273.

EXTRUSIVE ORGANELLES IN PROTISTS

273

Kolsch, K. (1902).Zool. Jahrb., Abt. Anat. Ontog. Tiere 16,273. Kovaleva, V. G. (1974). Tsitologiya 16,217. Kovaleva, V. G., and Raikov, I. B. (1972). Protistolgica 8,413. Kriiger, F. (1929). Verh. Dtsch. 2001.Ges. 32,267. Kriiger, F. (1930).Arch. Protistenkd. 72,91. Kriiger, F . (1931a).Arch. Protistenkd. 74,207. Kriiger, F. (1931b).Verh. Dtsch. Zool. Ges. 34, 140. Kriiger, F. (1934a).Arch. Protistenkd. 83,275. Kriiger, F. (1934b).Arch. Protistenkd. 83,321. Kriiger, F. (1936).Zoologica 34, 1. Kriiger, F. (1950).Mikrokosmos 40,36. Kriiger, F., and Wohlfarth-Bottermann, K. E. (1952). Mikroskopie 7, 121. Kriiger, F., Wohlfarth-Bottermann, K. E., and Pfefferkorn, G. (1952). Z. Naturforsch. Teil B 7,407. Kudo, R. R. (1971). “Protozoology,” 6th ed. Thomas, Springfield, Illinois. Lallier, R. (1977).Experientia 33, 1263. Laval, M. (1971). C. R. Hebd. Seances Acad. Sci. 273, 1383. Laval, M. (1972).Protistologica 8,369. Leadbeater, B., and Dodge, D. (1966). Br. Phycol. Bull. 3, 1. Leedale, G. F. (1967). “Euglenoid Flagellates.” Prentice-Hall, Englewood Cliffs, New Jersey. Lom, J., and Kozloff, E. N. (1967).J. Cell Biol. 33,355. Lorn, J., and Lawler, A. R. (1973).Protistologica 9,293. Luporini, P., and Magagnini, G. (1970).Protistologica 6, 113. McNutt, N. S., and Weinstein, R. S. (1973). Prog. Biophys. Mol. B i d . 26,47. Maier, H . (1903).Arch. Protistenkd. 2,73. Manton, I. (1969).Oesterr. Bot. 2. 116,378. Marczalek, D. S., and Small, E. B. (1969).Proc. 2nd Annu. SEM Symp. Mashansky, V. F., Seravin, L. N., and Vinnichenko, L. N. (1963).Acta Protozool. 1, 403. Mast, S. 0. (1909).Biol. Bull. 16, 91. Messer, G., and Ben-Shad, Y. (1969).J . Protozool. 16,272. Messer, G., and Ben-Shad, Y. (1971).J. Ultrastruct. Res. 37,M. Metzner, J. (1933).Science 78,341. Mignot, J.-P. (1963).C. R. Hebd. Seances Acad. Sci. 257,2430. Mignot, J.-P. (1965a).Protistologica 1, 5. Mignot, J.-P. (196513).J . Microsc. (Paris) 4,239. Mignot, J.-P. (1966).Protistologica 2,51. Mignot, J.-P. (1967a).Protistologica 3,5. Mignot, J.-P. (1967b).Ann. Stn. B i d . Besse-en-Chandesse 2, 161. Mignot, J.-P. (1976). Protistologica 12,279. Mignot, J.-P., and Brugerolle, G. (1975).Protistologica 11,429. Mignot, J.-P., and Hovasse, R. (1973). Protistologica 9,373. Mignot, J.-P., and Hovasse, R. (1974-1975). Ann. Stn. Biol. Besse-en-Chandesse 9,201. Mignot, J.-P., Joyon, L., and Pringsheim, E. G. (1968). Protistologica 4,493. Mignot, J.-P., Hovasse, R., and Joyon, L. (1970).J. Microsc. (Paris) 9, 127. Miller, D. M., Jahn, T. L., and Fonseca, J. R. (1968).J . Protozool. 15,493. Miller, S . (I968).J.Protozool. 15,313. Mitrophanow, P. (1905).Arch. Protistenkd. 5, 78. MonnB, L., and Harde, S . (1951).Ark. 2001.1,31. Monn6, L., and Slautterback, D. B. (1950).Exp. Cell Res. 1,477.

274

KLAUS HAUSMANN

Moody, J. E. (1912).J. Molphol. 23,349. Mornin, L., and Francis, D. (1967).J . Microsc. (Paris)6,759. Nemetschek, T., Hofmann, U., and Wohlfarth-Bottemann, K. E. (1953).Z. Naturforsch. Teil B 8, 383. Nilsson, J. (1969). C. R. Trau. Lab. Carlsberg 37,49. Nilsson, J. (1972).C . R. Trau. Lab. Carlsberg 39,83. Nilsson, J . (1976).C. R. Trau. Lab. Carlsberg 40,215. Nilsson, J., and Behnke, 0. (1971).J. Ultrastruct. Res. 36, 542. Norris, R. E., and Pearson, B. R. (1975).Arch. Protistenkd. 117, 192. Paulin, J. J., and Corliss, J. 0. (1969).J. Protozool. 16,216. Pease, D. C. (1947).J. Cell. Comp. Physiol. 29,91. Peck, R. K. (1971). Ph.D. Dissertation, University of Illinois, Urbana. Peck, R. K. (1974).Protistologica 10,333. Peck, R. K. (1977a).Fifth Int. Congr. Protozool., New York, p. 294. Peck, R. K. (197%). J. Cell Sci. 25, 367. PBnard, E. (1914). Rev. Suisse Zool. 22,407. PBnard, E. (1922). “Etudes sur les infusoires deau douce.” Georg, Geneve. Peschkowsky, L. (1931).Arch. Protistenkd. 73, 179. Pitelka, D. R. (1961).J . Protozool. 8, 75. Pitelka, D. R. (1963).“Protozoa as Cells. Electron Microscopic Structure of Protozoa.” Pergamon, Oxford. Pitelka, D. R. (1965).J. Microsc. (Paris) 4,373. Plattner, H. (1974).Nature (London) 252,722. Plattner, H., and Fuchs, S. (1975).Histochemistry 45, 23. Plattner, H. (1976).E x p . Cell Res. 103,437. Plattner, H., Miller, F., and Bachmann, L. (1973a).J. Cell Sci. 13,687. Plattner, H., Schmitt-Fumian, W. W., and Bachmann, L. (1973b).SOC, Fr. Microsc. Electron., p. 81. Plattner, H., Wolfram, D., Bachmann, L., and Wachter, E. (1975).Histochemistry 45,l. Pollack, S. (1974)J. Protozool. 21,352. Pollack, S., and Steers, E., Jr. (1973).E x p . Cell Res. 78, 186. Poste, C., and Allison, A. C. (1973). Biochim. Biophys. Acta 300,421. Potts, B. P. (1955).Biochim. Biophys. Acta 16,464. Preer, J. R., Jr., Hufnagel, L. A., and Preer, L. B. (1966).J. Ultrastruct. Res. 15, 131. Preer, J. R., Jr., Preer, L. B., and Jurand, A. (1974). Bacteriol. Reu. 38, 113. Prelle, A. (1968).J.Protozool. 15, 517. Prelle, A., and Aguesse, P. (1968).Bull. SOC. Zool. Fr. 93,479. Raikov, I . B. (1971-1972). Ann. Stn. Biol. Besse-en-Chandesse 6-7,21. Raikov, I . B. (1974). Tsitologiya 16, 626. Raikov, I. B., and Kovaleva, V. G. (1968).Acta Protozool. 6,309. Raikov, I . B., Gerassimova-Matejeva,Z. P., and de Puytorac, P. (1975).Acta Protozool. 14, 17. Rao, P. A. V. S. (1963).J.Protozool. 10,204. Repak, A. J., and Pfister, R. M. (1967). Trans. Am. Microsc. SOC. 86,417. Rieder, N. (1968a).A. Naturfor.sch., Teil B 23,569. Rieder, N. (1968b).Z. Naturforsch., Teil B 23, 569. Rieder, N. (1970).Z. Naturforsch., Teil B 25, 1494. Rieder, N. (1971).Formaet Functio 4, 46. Rieder, N. (1973).Arch. Protistenkd. 115, 125. Rodrigues de Santa Rosa, M. (1974). C. R. Seances SOC. Biol. Ses Fil. 168, 1349. Rodrigues de Santa Rosa, M., and Didier, P. (1975). Protistologica 11,469.

EXTRUSIVE ORGANELLES IN PROTISTS

275

Roque, M., de Puytorac, P. and Savoie, A. (1965). Arch. Zool. E x p . Gen. 105,309. Roque, M., de Puytorac, P., and Lom, J. (1967).Protistologica 3,79. Rosati Raffaelli, G. (1970). Arch. Anat. Microsr. Morphol. Exp. 59,221. Rouiller, C., and Faure-Fremiet, E. (1957). Bull. Microsc. Appl. 7, 135. Rudzinska, M. A. (1965).J. Cell Biol. 25,459. Rudzinska, M. A. (1970).J. Protozool. 17,626. Rudzinska, M. A. (1973). BioScience 23,87. Ruiz, F., Adoutte, A., Rossignol, M., and Beisson, J. (1976). Genet. Res. 27, 109. Satir, B. (1974a).J. Supramol. Struct. 2,529. Satir, B. (1974b).In “Transport at the Cellular Level” (M. A. Sleigh and D. H. Jennings, eds.), p. 399. Cambridge Univ. Press, London and New York. Satir, B. (1975). Sci. Am. 233, 28. Satir, B., Schooley, C., and Satir, P. (1972).Nature (London) 2 3 5 , s . Satir, B., Schooley, C., and Satir, P. (1973).J . Cell B i d . 56, 153. Satir, B., Sale, W. S., and Satir, P. (1976). E x p . Cell Res. 97,83. Satir, P., and Satir, B. (1974).In “Control of Proliferation in Animal Cells” (B. Clarkson and R. Baserga, eds.), p. 233. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Saunders, J. T. (1925). Proc. Cambridge Philos. S O C . , Biol. Sci. 1, 249. Schatten, G., and Mazia, P. (1976).J. Supramol. Struct. 5, 343. Schmidt, W. J. (1939).Arch. Protistenkd. 92, 527. Schneider, L., and Wohlfarth-Bottermann, K. E. (1964).Stud. Gen. 17,95. Schneider, W. (1930).Arch. Protistenkd. 72,482. Schnepf, E. (1966a). Biol. Rundsch. 4,259. Schnepf, E. (196613)In “Probleme der biologischen Reduplikation” (P. Sitte, ed.), p. 372. Springer-Verlag, Berlin and New York. Schnepf, E. (1977).In “Biophysik” (W. Hoppe et al., eds.), p. 1. Springer-Verlag, Berlin and New York. Schuberg, A. (1905).Arch. Protistenkd. 6,61. Schuster, F. L. (1968).Exp. Cell Res. 49,277. Schuster, F. L. (1970).J.Protozool. 17,521. Schuster, F. L., Prazak, B., and Ehret, C. F. (1967).J.Protozool. 14,483. Schuster, F. L., Goldstein, S., and Hershenov, B. (1968). Protistologica 4, 141. Schwartz, V. (1965).Z . Naturforsch., Teil B 20,383. Schwelitz, F. D., Evans, W. R., Mollenhauer, H. H., and Dilley, R. A. (1970). Protoplasma 69,341. Sedar, A. W., and Porter, K. R. (1955).J.Biophys. Biochem. Cytol. 1,583. Selman, G . G., and Jurand, A. (1970).J . Gen. Microbiol. 60,365. Sleigh, M. A. (1973). “The Biology of Protozoa.” Arnold, London. Small, E. B., and Marczalek, D. S. (1969). Science 163, 1064. Soldo, A. T. (1974).In “Paramecium-A Current Survey” (W. J. van Wagtendonk, ed.), p. 377. Elsevier, Amsterdam. Sommer, J. R. (1965).J . Cell Biol. 24,253. Soyer, M. 0. (1968). Vie Milieu 19,305. Soyer, M. 0. (1969). Protistologica 5,327. Soyer, M. 0. (1970a).Z . Zellforsch. Mikrosk. Anat. 104,29. Soyer, M. 0. (1970b). Z . ZelZforsch. Mikrosk. Anat. 105,350. Spoon, D. M., Chapman, G. B., Cheng, R. S., and Zane, S. F. (1976).Trans. Am. Micros. S O C . 95,443. Staehelin, L. A. (1974). Int. Reu. Cytol. 39, 191. Steers, E., Beisson, J., and Marchesi, V. T. (1969).Exp. Cell Res. 57,392.

276

KLAUS HAUSMANN

Stewart, J. M., and Muir, A. R. (1963).Q. J . Microsc. Sci. [N.S.] 104, 129. Studitsky, A. (1930). Arch. Protistenkd. 70, 155. Suganuma, Y. (1973).J. Electron Microsc. 22,347. Swale, E. M. F., and Belcher, J. H. (1975). Arch. Protistenkd. 117,20. Sweeney, B. M., and Bouck, G. B. (1966).In “Bioluminescence in Progress” (F. H. Johnson and Y. Haneda, eds.), p. 331. Princeton Univ. Press, Princeton, New Jersey. Tartar, V. (1961). “The Biology of Stentor.” Pergamon, Oxford. Thompson, J. C., and Corliss, J. 0. (1958).J. Protozool. 5, 175. Tiedtke, A. (1976). Naturwissenschaften 63,93. Tokuyasu, K., and Scherbaum, 0. H. (1965).J. Cell Biol. 27,67. Tonniges, C. (1914).Arch. Protistenkd. 32,298. Troyer, D., and Hauser, M. (1978). Cytobiologie (in press). Tucker, J. B., and Mackie, J. B. (1975). Tissue G Cell 7,601. Tuffrau, M. (1954).J. Protozool. 1, 183. Ukeles, R., and Sweeney, B. M. (1969). Limnol. Oceanogr. 14,403. Vacquier, V. D. (1976).J. Supramol. Struct. 5,27. Vavra, J. (1978). In “The Microsporidia” ( J . Vavra and V. Sprague, eds.). Plenum, New York. Visscher, J. P. (1923). Biol. Bull. 45, 113. Wehrmeyer, W. (1970). Protoplusmu 70,295. Weinke, K. A. (1972). Anat. Rec. 172,423. Wenrich, D. H. (1929). Biol. Bull. 56,390. Wessenberg, H., and Antipa, G. (1968).Protistologica 4,427. Wessenberg, H., and Antipa, G. (1970).J. Protozool. 17,250. Westphal, A. (1974). “Spezielle Zoologie, 1: Protozoen.” Ulmer, Stuttgart. Wichterman, R. (1953).“The Biology of Paramecium.” McGraw-Hill (Blakiston), New York. Williams, N. E., and Luft, J. H. (1968).J.Ultrastruct. Res. 25, 271. Wohlfarth-Bottermann, K. E. (1950). Naturwissenschaften 37,562. Wohlfarth-Bottermann, K. E. (1953). Arch. Protistenkd. 98, 169. Wohlfarth-Bottermann, K. E., and Pfefferkorn, G . (1952). Umschau 52, 114. Wohlfarth-Bottermann, K. E., and Pfefferkorn, G. (1953a). Z . Wiss. Mikrosk. Mikrosk. Tech. 61,239. Wohlfarth-Bottermann, K. E., and Pfefferkorn, G. (1953b). Umschau 53,366. Wohlfarth-Bottermann, K. E., and Schwantes, H. 0. (1952). Z . Nuturforsch., Teil B 7, 489. Wunderlich, F., and Speth, V. (1972).J. Ultrastruct. Res. 41,258. Yagiu, R., and Shigenaka, Y. (1958a).J. Electron Microsc. 6,38. Yagiu, R., and Shigenaka, Y. (195813).Zool. Mag. 67, 106. Yagiu, R., and Shigenaka, Y. (1965).J. Protozool. 12,363. Yusa, A. (1963).J. Protozool. 10,252. Yusa, A. (1965).J. Protozool. 12,51. Zebrun, W., Corliss, J. 0, and Lom, J. (1967). Trans. Am. Microsc. Soc. 86,28.

Lectins C. BROWN

JAY

Department of Microbiology, University of Virginia School of Medicine, Charlottesville, Virginia AND

RICHARDC.HUNT Department of Biochemistry, Oxford University, Oxford, England I. Introduction

.

.

11. Lectin Biochemistry

111.

IV.

V. VI .

VII. VIII. IX.

.

.

. .

.

. .

. . .

. . . . . . .

A. Blood Croup Specificity . B. Lectin Purification . . . . C. Lectin Structure . . . . Lectin-Induced Lymphocyte Mitogenesis A. Nature o f t h e Mitogenic Response . B. Mechanism of Lectin-Induced Mitogenesis Selective Agglutination of Transformed Cells A. Correlation of Lectin Agglutinability with . . . . . Transformation . B. Lectin Agglutination and Tumorigenicity C. Transient Agglutinability of Untransformed D. Mechanism of Agglutination . . . Interaction of Lectins with Cells Infected by NononcogenicViruses . . . . . Interaction of Lectins with Developing Cells A. Eggs and Embryos .- . . . . B. Male Germinal Cells . . . . Biochemistry of Cell Surface Lectin Receptors Lectin Toxicity . . . . . . The Biological Role of Lectins . . . References . . . . . . .

.

. . . .

. . .

.

. .

.

. . . .

. .

. .

. . .

.

.

.

.

.

.

. . .

297 305 305 309

.

. .

319 322 322 325 326 330 333 336

.

. .

. . . .

.

Cells

.

. .

.

. . .

277 278 278 280 288 292 292 292 297

.

. . .

. .

.

. .

. . .

I. Introduction Lectins are proteins that can bind noncovalently to specific carbohydrate groups without modifying them chemically. Binding is reversible, and all lectins have more than one specific carbohydratecombining site. No enzymic activity has as yet been associated with any purified lectin molecule. The presence of more than one carbohydrate-combining site allows individual lectin molecules to serve as cross-linking agents, and in fact lectins were first identified in extracts of plant seeds found to contain soluble factors that agglutinate red 277

278

JAY C. BROWN AND RICHARD C. HUNT

blood cells. The factors responsible for hemagglutination were called phytohemagglutinins (from the Greek phyton, meaning “plant”), and one still encounters this term. The word “lectin” was first proposed by Boyd (1954) to take account of the fact that similar hemagglutinating factors can also be isolated from animal sources. Since individual plant or animal species ordinarily contain only one type of lectin, lectins are named with reference to their species of origin. For example, one speaks of wheat germ agglutinin (WGA) or soybean lectin, and we adhere to this system of nomenclature in this article. Cell biologists have made the most conspicuous use of lectins as probes in studies of cell surface structure and function. Lectins are bound specifically to carbohydrate-containing groups on the cell surface, and biochemical or microscopic methods are employed to examine the consequences of such binding. Striking results are often observed. For instance, some lectins induce mitosis in resting lymphocytes (see Section 111),some agglutinate neoplastic, transformed cells but not their normal, noncancerous counterparts (see Section IV), and many have been employed to demonstrate significant changes in cell surface architecture following virus infection (see Section V) and during development (see Section VI). One lectin, concanavalin A (Con A), has even been employed as the basis of a method for physically isolating the plasma membranes of Neurospora crassa (Scarborough, 1975). In this article we describe the ways lectins have been used to study animal cell surfaces, and we attempt to evaluate the results obtained. Special attention is devoted to studies published since this subject was last reviewed in the Znternational Review of CyOther recent reviews which have covered tology (Nicolson, 1974~). some of the same material include those by Toms and Western (1971), Sharon and Lis (1972), Lis and Sharon (1973b), Burger (1973), Rapin and Burger (1974), Nicolson (1976a,b), and Sharon (1977). Collections of papers edited by Cohen (1974) and by Chowdhury and Weiss (1975) are also relevant. Even brief attention to this literature will convince the reader that the potential value of lectins as structural and functional membrane probes is only just beginning to be realized. 11. Lectin Biochemistry

A. BLOOD GROUPSPECIFICITY The hemagglutinating activity of plant extracts was first observed by Stillmark (see Boyd, 1963), who studied Ricinus communis (castor bean) agglutinin (RCA). Much later, Landsteiner (see Boyd, 1963) noted that crude lectin preparations did not agglutinate all red cells to

LE CTINS

279

the same extent; agglutination was found to depend on the species and blood type of the donor. Similarly, Renkonen (1948)observed that extracts of Vicia cracca agglutinated human blood type-A cells more strongly than cells of groups B and 0. These original observations were quickly and thoroughly extended in an effort to identify lectins suitable for use in blood typing. For example, Boyd and Reguera (1949) extracted 262 types of seeds from 63 different plant families and tested the extracts for hemagglutinating activity; 191 did not agglutinate, 46 showed agglutination that was not specific to any blood group, and 25 agglutinated cells of only a particular blood group. An even more extensive study was carried out by Allen and Brilliantine (1969).Material from 2663 plants was prepared and tested for hemagglutination; 1635 extracts were inactive, 711 were nonspecific agglutinins, 227 lysed red cells of all types, and 90 caused specific agglutination. Studies on the blood group specificity of lectins and their use in blood typing has been thoroughly reviewed (Makela, 1957; Bird, 1959a; Boyd, 1963).Table I lists some of the lectins found to be blood group-specific. The differential agglutination of red cells based on blood type clearly indicates that different lectins must bind to different chemical groups on the red cell surface. The identity of the group recognized by a given lectin is ordinarily determined by measuring the ability of simple monosaccharides or oligosaccharides to inhibit lectin-induced hemagglutination. Inhibition is usually found to be quite specific for a particular monosaccharide. For example, Morgan and Watkins (1959) showed that agglutination by lectins specific for blood group A was inhibited b y N-acetyl-Dglucosamine, while those specific for blood group 0 were inhibited by a-methyl-L-fucose. Similar studies have been carried out to determine the saccharide-binding specificity of most lectin preparations, and some of the results are given in Table 11. Although the most commonly studied lectins have been extracted from plants, particularly from the seeds of Leguminoseae, many other sources have been found to contain lectins. These include plant roots (Saint-Paul, 1961; Allen and Neuberger, 1973),fungi (Elo et al., 1951; Tktry et al., 1954; Saint-Paul, 1961; Sage and Connett, 1969), and the hemolymph of a variety of invertebrates including the horseshoe crab (Cohen et al., 1965, 1974; Marchalonis and Edelman, 1968), sea hare (Pauley et al., 1971), snail (Cohen et al., 1965; Prokop et al., 1968), lobster (Cornick and Stewart, 1973), earthworm (Cooper et al., 1974), and oyster (Tripp, 1966, 1974). Other sources include the ova of fish (Prokop et al., 1967,1968;Pardoe and Uhlenbruck, 1970),slime molds (Rosen et al., 1973, 1974, 1975; Simpson et al., 1974),the electric eel (Teichberg et at., 1975), toad eggs (Wyrick et al., 1974), rabbit liver

JAY C. BROWN AND RICHARD C. HUNT

280

TABLE I BLOOD GROUP-SPECIFIC LECTINS Blood grow A

A+B

B

Lectin source Clitocybe nebularis Crotalaria aegyptiaca Dolichos bijorus (horse gram) Hyptis suaveolens Lathyrius sylvestris Phaseolus lunatus (lima bean) Vicia cracca Bandeiraea simplicifolia Calpurina aurea Coronilla varia Crotalaria mucronata Crotalaria striata Sophora japonica Euonymus spp. Marasmius oreades Polyporus fomentarius

0

M N

Vitis aestivalis Cytisus sessilifolius Laburnum alpinum Lotus tetragonolobus Ononsis spinosa Ulex europeus (I and 11) Xylaria polymorpha Iberia amava Vicia graminia

Reference TBtry et al. (1954) Bachrach et a!. (1957) Bird (1951, 1952) Bird (1959b, 1960) Cazal and Lalaurie (1952) Boyd and Reguera (1949) Koulumies (1949) Makela and Makela (1956) Makela (1957) Kriipe and Braun (1952) Ottensooser and Sat0 (1963) Makela and Makela (1956) Kriipe and Braun (1952) Potapov (1968, 1970); Ottensooser et al. (1968) Elo et al. (1951) Saint-Paul (1961); Gillespie and Gold (1960) Boyd and Reguera (1949) Renkonen (1948) Renkonen (1948) Renkonen (1948) Saint-Paul (1961); Herzog (1959) Cazal and Lalaurie (1952) Saint-Paul (1961) Boyd (1963) Saint-Paul (1961)

(Lunney and Ashwell, 1976), eel serum (Bezkorovainy et al., 1971; Desai and Springer, 1973),and the albumin gland of mollusks (Prokop et al., 1965; Kriipe and Pieper, 1966; Ishiyama et al., 1974; Hammarstrom, 1974).

B. LECTINPURIFICATION Many of the crude extracts found to contain hemagglutinins have now been fractionated to yield purified lectin preparations. In some cases purification has been accomplished by traditional, methods of protein chemistry, although quite often more modern techniques of affinity chromatography have been employed. The latter procedures exploit the binding of lectins to specific saccharide groups, and they

TABLE I1 BIOCHEMICALPROPERTIES OF SOME IMPORTANT LECXINS

Source Plant lectins Abrus pecatorius

Molecular weight

Number of subunits

60,000

2

Nontoxic

126,000

4

Peanut lectin

110,000

Lectin

Glycoprotein

Inhibitor

References'

Abrin Yes

DGalactose

1

Yes

DGalactose

2

4

30,& (A) and 35,000 (B) 2 x 33,800 and 2 x 32,200 27,500

No

Dgalactose

3

114,000

4

28,500

Yes

Dgalactose

4

Con A

102,000

4

25,500

No

5

Sunn hemp lectin Horse gram lectin Soybean lectin

120,000

-

Yes

DGlucose, Dmannose DGalactose

111,000

4

110,000

Toxic

Arachis hypogaea Bandeiraea simplicifolia Canaualia ensiformis Crotolaria juncea Dolichos bifirus Glycine mar

Subunit molecular weight

-

6

Yes

DgalNAc

7

4

26,000 and 26,500 30,000

Yes

DgalNAc, Dgalactose DMannose, Dglucose

8

Lens cvlinaris

LCA-A

48,000

2

24,500

Yes

LCA-B A B C

48,000 120,000 58,000 117,000

2 4 2 4

24500

Yes

Lotus tetragonolobus

27,000

-

-

L-Fucose L-Fucose L-Fucose

9

10 (Continued)

TABLE I1 (Continued)

Source Phaseolus lunatus

Lectin Lima bean lectin 1 2

Phaseolus uulgaris

Phytolacca americana Pisum satioum

Ricinus communis

PHA 1 2 Pokeweed niitogen Pea lectin

Castor bean lectin RCAI RCAn

Solanum tuberosum Suphora japonica

Potato lectin

Subunit molecular weight

Molecular weight

Number of subunits

180,000 or 269,000" 90,000or 138,000"

-

31,000

Yes

DgalNAc

11

128,000 128,OOO 32,oocI

4 4

29,000 33,000

Yes Yes Yes

DgalNAc DgalNAc

12

8000 and x 15,000 or 10,OOO and x 18,000

Yes

DMannose, Dglucose DMannose, Dglucose

Yes

Yes

-

2x 2 2x 2

-

4 9 , W or

4

55,000

4

120,000

4

60,000

2

120,000

4

2 x 29,500 and 2 x 37,000 1 x 29,500 and 1 x 34,000 46,000

132,000135,000

4

33,000

Glycoprotein

Yes

Yes Yes

Inhibitor

-

References'

13

14

DGalactose, DgalNAc DGalactose, DgalNAc DglcNAc

15

Dgalactose, DgalNAc

17

16

Triticum uul-

WGA

34,000

2

17,000

Yes or no"

garis

Ulex europaeus Wistaria floribunda Animal lectins Anguilla rostrata Helix pomatia Limulus pol yphemus

DglcNAc, NANAb

18

L-Fucose Di-N-acetylchitobiose DgalNAc Di-N-acetylchitobiose

19 20

Gorse lectin

I

Yes Yes

11

170,000 170,000

WFH WFM

136,000 67,000

4 2

35,000 32,000

Yes Yes

Eel antihuman blood group protein Snail hemagglutinin Horseshoe crab agglutinin

123,000

3

40.000

Yes

L-Fucose

22

79,000

6

13,000

Yes

23

-

22,500

DgalNAc, DglcNAc -

400,000

21

Disagreement between reports. N-Acetylneuraminic acid. References: (1)McPherson and Rich, 1973; Olsnes et al., 1974a, 1975 (2) Olsnes et al., 1974a; Wei et al., 1975 (3)Lotan et al., 1975b (4) Hayes and Goldstein, 1974,1975 (5)Goldstein et al., 1965; Goldstein and So, 1965;Agrawal and Goldstein, 1968a,h; Ahe et al., 1971; Entlicher et al., 1971; McKenzie et al., 1972 (6)Ersson et al., 1973 (7) Etzler and Kabat, 1970; Carter and Etzler, 1975a,b (8)Lis et al., 1966a, 1970; Lotan et al., 1974,1975a (9) Entlicheret al., 1969; Howardet al., 1971; Stein et al., 1971(10) Springer and Williamson, 1962; Yarivet al., 1973; Pereira and Kabat, 1974 (11)Could and Scheinberg, 1970; Galbraith and Goldstein, 1970 (12) Allen et al., 1969; Kornfeld and Kornfeld, 1970; Allan and Crumpton, 1971 (13)Reisfeld et al., 1967 (14) Entlicher et al., 1969, 1970; Trowbridge, 1974; Marik et al., 1974; Van Wauwe et al., 1975 (15) Nicolson and Blaustein, 1972; Nicolson et al., 1974; Olsnes et al., 1974a, 1975 (16) Allen and Neuberger, 1973 (17) Poreiz, 1973; Poretz et al., 1974 (18)LeVine et al., 1972; Nagata and Burger, 1972, 1974; Allen et al., 1973; Greenaway and LeVine, 1973 (19) Matsumoto and Osawa, 1969,1970 (20) Toyoshima et al., 1971; Toyoshima and Osawa, 1975 (21) Marchalonis and Edelman, 1968 (22) Springer et al., 1965 (23) Hammarstrom and Kabat, 1969; Hammarstrom, 1973.

284

JAY C. BROWN AND RICHARD C. HUNT

may result in a one-step procedure yielding a homogeneous protein. For example, Con A can be purified in this way by taking advantage of its specific binding to Sephadex, a glucose polymer. The lectin can be eluted from Sephadex with glucose (Agrawal and Goldstein, 1967) or by lowering the pH (Olson and Liener, 1967). Similarly, galactosebinding lectins (such as those from R. communis, Momordia charantia, Abrus pecatorius, and Crotalaria juncea) can be purified by affinity chromatography on agarose gels (Nicolson and Blaustein, 1972; Tomita et al., 1972b; Ersson et al., 1973). Affinity columns for other lectins require attachment of a specific receptor to an insoluble matrix. This has been achieved by cyanogen bromide activation of Sepharose (Cuatrecasas, 1970), or by copolymerization of the receptor with the N-carboxy anhydride of L-leucine (Tsuyuki et al., 1956).The former method was employed, for example, to insolubilize ovomucoid for the purification of WGA (LeVine et al., 1972), and the latter to insolubilize hog A + H blood group substance for the preparation of Dolichos bijlorus lectin (Kaplan and Kabat, 1966; Etzler and Kabat, 1970; Etzler, 1973).Table I11 lists some of the lectins that have been purified and indicates the method of purification employed. Difficulties encountered in lectin purification have been connected with the fact that lectins can occur in different polymeric states. For example, Con A exists as a dimer in solution below pH 5.6 and as a tetramer above pH 5.6 (Kalb and Lustig, 1968; Abe et al., 1971; McKenzie et al., 1972) Lectin properties can also differ as a result of the presence of ions (Agrawal and Goldstein, 1968a; Uchida and Matsumoto, 1972) or sugars (Agrawal and Goldstein, 196813).Also, lectin purification has occasionally been complicated by the presence in extracts of more than one form of the same lectin. Such isolectins have identical saccharide-binding specificities and similar, but not identical, structures. Isolectins have now been identified in the case of D. biflorms lectin (Carter and Etzler, 1975a), lentil lectin (Tichi et al., 1970; Howard et al., 1971), lima bean lectin (Gould and Scheinberg, 1970),red kidney bean agglutinin (Yachnin and Svenson, 1972; Miller et al., 1973; Pusztai and Watt, 1974),soybean lectin (Lis et al., 196613; Catsimpoolas and Meyer, 1969), and WGA (Allen and Neuberger, 1973; Rice and Etzler, 1975).Much less frequently a single biological source may be found to yield multiple lectins having different structures and saccharide-binding specificities. The most prominent example of this situation is the gorse (Ulex europeus) lectins UEA, and UEA,; UEAl binds fucose, while UEAn binds di-N-acetyldichitobiose (D-glcNAc), (Matsumoto and Osawa, 1969, 1970). I

TABLE 111 LECTIN PURIFICATION Lectin and source

Method of purification

Abrin (Abrus pecotorius)

(NH,),SO, precipitation; affinity chromatography on Sepharose; elute with galactose Extract jack bean meal with NaCI; (NH,),SO, precipitation; affinity chromatography on Sephadex G-50; elute lectin with glucose or lower pH Affinity chromatography on Sepharose; ellite with lactose Alcohol precipitation Affinity chromatography on polyleucyl-hog blood group A + H substance; elute with N-acetylglucosamine Extract with NaCI; (NH,),SO, precipitation UEA: CM-cellulose, Sephadex G-200, Biogel P-200. UEA 11: DEAE cellulose; electrophoresis; biogel P-200 Affinity chromatography. Lectin I on fucose coupled to starch gel; elute with glycine-HC1. Lectin I1 on tri-N-acetylchitotriose-starch gel; elute with glycine-HC1 buffer Extract in NaCI; DEAE-cellulose chromatography Affinity chromatography on Sephadex; elute with glucose

Con A (Conouolia enisformis, jack bean)

Crotolnriu junceo lectin

Dolichos bijlorus lectin (horse gram)

Gorse lectins (Ulex europeus): UEA I binds fucose; UEA I1 binds (DglcNAc),

Lentil lectin (Lens culinuris)

Lima bean lectin (Phaseolus lunutus)

Extract with NaCI; (NH,),SO, precipitation; affinity chromatography on polyleucyl type-A blood group substance; elute with N-acetylglucosamine

Reference Tomita et oZ. (1972b)

Agrawal and Goldstein (1965, 1967, 1973); Olson and Liener (1967) Ersson et al. (1973)

Bird (1959a) Kaplan and Kabat (1966); Etzler and Kabat (1970); Etzler (1973) Matsumoto and Osawa (1969, 1970) Osawa and Matsumoto (1973)

Matsumoto and Osawa (1972)

Howard and Sage (1969); Sage and Green (1973) Entlicher et ul. (1970); Tichi et al. (1970); Toyoshima et al. (1970); Howard et al. ( 1971) Galbraith and Goldstein (1970, 1973)

(Continued )

286

JAY C. BROWN AND RICHARD C. HUNT

TABLE 111 (Continued) Lectin and source

Lotus tetragonolobus lectin

Mormordia charantia lectin

Mushroom lectin (Agaricus bisporus) Pea lectin (Pisum sativum) Potato lectin (Solanum tuberosum)

PHA (Phaseolus uulgaris)

RCA

Robin (Robinia pseudoacacia, block locust)

Method of purification Extract in phosphate buffer; (NH,),SO, precipitation; gel filtration on Biogel A0.5 or Biogel P-300 Extract with phosphate-buffered saline; affinity chromatography on Sepharosefucose; elute with fucose (NH4),S04precipitation; affinity chromatography on Sepharose 4B; elute with galactose Extract in NaCl solution; DEAE-cellulose, Sephadex G-100, phosphocellulose chromatography (NH,),SO, precipitation; affinity chromatography on Sephadex G-150; elute with glycine-HC1 (NH,),SO, precipitation; DEAE-cellulose, CM-cellulose, Sephadex G-100, SP-Sephadex chromatography Ethanol precipitation; (NH,),SO, precipitation SE-Sephadex chromatography; Sephadex-G150 filtration; phosphocellulose chromatography Affinity chromatography on Sepharose-th yroglobulin; elute with glycine-HC1 (NH,),SO, precipitation; affinity chromatography on agarose; elute with galactose or lactose. Separates into two fractions on Sephadex G-200, called RCA, and RCA,, Extract in phosphate-buffered saline; (NH4)*S04 precipitation; DEAE-cellulose and CM-cellulose chromatography

Reference Could and Scheinberg (1970) Blumberg et al. (1972); Yariv et al. (1973) Tomita et al. (1972b)

Sage and Connett (1969); Presant and Komfeld (1972) Entlicher et al. (1970)

Allen and Neuberger (1973)

Rigas and Osgood (1955) Kornfeld et al. (1973)

Matsumoto and Osawa (1972) Nicolson and Blaustein (1972); Tomita et al. (1972b)

Bourrillon and Font (1968)

287

LECTINS TABLE 111 (Continued) Lectin and source

Method of purification

Sophora japonica lectin

Extract in phosphate-buffered saline; ethanol precipitation; affinity chromatography on polyleucylhog gastric mucin; elute with galactose Extract in water; (NH,),SO, precipitation; dialyze against ethanol; chromatography on calcium phosphate and DEAE-cellulose Extract in NaCl solution; (NH4),S04precipitation; affinity chromatography on Sepharose-galactose; elute with galactose Affinity chromatography on Sepharose -bl ood group-A substance; elute with acetate buffer Extract wheat germ with NaCI; (NH,),SO, precipitation; affinity chromatography on ovomucoid-Sepharose; elute with 0.1 N acetic acid Affinity chromatography on chitin column; elute with 0.05 N HCI Affinity chromatography on N-acetylglucosamine-Sepharose; elute with N-acetylglucosamine or 0.1 N acetic acid Extract in NaCI; (NH,),SO, precipitation; SE-Sephadex chromatography; Sepharose-6B filtration

Soybean lectin (Clycine max)

Vicia cracca lectin

WGA (Triticum uulgaris)

Wistaria floribunda lectin

Reference Poretz (1972); Poretz et al. (1974)

Liener and Pallansch (1952); Liener (1953); Wada et al. (1958);Lis et al. (1966a); Lis and Sharon (1973a) Gordon et al. (1972, 1973)

Sundberg et QZ. (1970)

LeVine et al. (1972); Marchesi (1973)

Bloch and Burger (1974) Lotan et al. (1973a); Shaper et al. (1973)

Toyoshima et aZ. (1971); Osawa and Toyoshima (1972)

288

JAY C. BROWN AND RICHARD C. HUNT

C. LECTINSTRUCTURE

1. Concanavalin A Although many lectins have been chemically purified, relatively few have been subjected to detailed structural analysis. By far the best studied lectin structure is that of Con A. Chemical studies of Con A have shown that it is a tetramer of identical protomers, each of which has a molecular weight of 25,500. The protomer polypeptide chain consists of 237 amino acid residues whose sequence has been determined (Wang et al., 1975a; Cunningham et al., 1975);Con A contains no covalently bound carbohydrate. Individual Con-A protomers contain binding sites for two metal ions, one for Mn2+and one for Ca2+, plus one saccharide-binding site. Mn2+must be bound before Ca2+, and both metal ions are required for optimal saccharide binding (Yariv et al., 1968). In solution the Con-A tetramer dissociates into two identical dimers below pH 5.6 (Becker et al., 1971; Hardman and Ainsworth, 1972a), and it self-associates in a time-dependent fashion to form high-molecular-weight aggregates above neutral pH (McKenzie et al., 1972). Studies on the specificity of the Con-A saccharide-binding site have revealed that Con A binds specifically to mannosyl and also to glucosyl groups at the nonreducing ends of oligosaccharide chains (Goldstein et al., 1965, 1974; Poretz and Goldstein, 1970). Certain internal mannosyl groups can also be accommodated. Con A can be induced to form crystals suitable for structural studies using x-ray diffraction techniques, and two laboratories have now completed this analysis (Hardman and Ainsworth, 1972b; Becker et al., 1975; Reeke et al., 1975). Their results are in good overall agreement. The Con-A monomer (protomer) that forms the crystallographic asymmetric unit has been found to be a compact, dome-shaped structure having a height of 43 A and a cross section of 39 x 40 A. Two such protomers are joined at their bases to form a dimer having roughly the shape of a prolate ellipsoid of revolution (84 x 40 x 39 A) in which the protomers are related by a twofold axis of rotation. Tetrameric Con A is formed by the association of two dimers to create a roughly tetrahedral structure in which the protomers are related by three twofold crystallographic axes; the tetramer therefore has D z symmetry, as shown in Fig. 1. The dimers referred to above, which exist in solution at low pH, are formed between subunits I and I1 or I11 and IV but not, for example, between I and I11 or I and IV (Reeke et al., 1975), as shown in Fig. 1. The most conspicuous feature of the Con-A polypeptide is the large

LECTINS

289

FIG. 1. Schematic representation of the Con-A tetramer. The manganese and calcium sites are indicated by Mn and Ca, respectively. The saccharide-binding site is indicated by S and the major cleft by I. Adapted from Reeke et al. (1975).

amount of p structure found in this molecule; almost 55% of the amino acid residues are involved in two regions of an antiparallel pleated sheet. One of these regions, the “back” pleated sheet, consists of 64 amino acid residues arranged in six antiparallel chains found in the back of the molecule, as iIlustrated in Fig. 2. Except for a slight curl at the top, this region is relatively flat. It forms the back of the major cleft to the lower right in Fig. 2 and contains most of the amino acid residues involved in intersubunit interactions. Details of the intersubunit contacts are given by Reeke et al. (1975).The second major region of /3 structure, the “front” pleated sheet, consists of 57 amino acid residues arranged in seven antiparallel chains which lie on top of the back pleated sheet with their long axes oriented at approximately a 45” angle to those of the back sheet. The front pleated sheet is twisted through an angle of about go”, and it forms the upper portion of the major cleft which lies roughly between the back and front pleated sheets and opens at the lower right in Fig. 2. The amino- and carboxyterminal amino acids are found to the right of the front pleated sheet, and the metal ion-binding sites are just above it. The Mn2+ and Ca2+ ions are found 4.6 A apart in the upper portion of the molecule, and

290

JAY C. BROWN AND RICHARD C. HUNT

FIG. 2. Stereo view of the Con-A protomer with the back region of the p structure darkened. The back B structure forms a vertical plane with a curl at the top left and a small bent portion at the far left (Becker et al., 1975).

each is surrounded by an octahedral coordination shell consisting of four ligands from the protein and two water molecules. A drawing of these sites is given by Becker et al. (1975). The saccharide-binding site is perhaps the most interesting and yet controversial feature of the Con-A structure. Con-A crystals are found to have a deep, narrow cavity formed by portions of the back and front pleated sheets which opens at the lower right in Fig. 2; it extends 18 A into the core of the molecule and is roughly 5 A wide and 7.5 A high. This cavity is easily large enough to accommodate the a-D-mannose and a-Dglucose groups for which Con A is specific, and Becker et al. (1971) originally proposed that this cavity is the place where saccharides are ordinarily bound. Studies on Con-A-saccharide interactions in solution, however, have not supported this view. For example, Brewer et al. (1973) used pulsed 13Cnuclear magnetic resonance (NMR) spectroscopy to study the interaction of 13C-enricheda-methylD-glucopyranoside (a-MG) with Con A. The techniques employed allowed these workers to calculate the distance between the transition metal (MnZ+)site and individual carbon atoms in bound a-MG. Their results showed clearly that a-MG was too close to Mn2+for the sugar to be bound in the major cleft. For example, the C-3 and C-4 carbon atoms ofa-MG were found to be at a mean distance of no more than 10A from Mn2+,while they would have to be more than 20 A from Mn2+if aMG were bound in the major cleft.

LECTINS

29 1

Recent x-ray crystallographic studies of Con A containing specifically bound saccharides have been in overall agreement with the NMR studies. Methyl-a-D-mannopyranoside (Hardman and Ainsworth, 1976) and an iodine-containing analog (2-deoxy-2-iodomethyla-Dmannopyranoside) of a-D-mannose (Becker et al., 1976) were found to bind to Con A at a shallow pocket near the metal ion sites but slightly to the left and in front of them, as shown in Fig. 1. Since the metal ions are required to stabilize this site, the results neatly explain why demetallized Con A has no specific saccharide-binding site. 2. Other Lectins It is clear that our extensive knowledge of Con-A structure is due primarily to the growth of suitable crystals of this lectin and to their analysis by x-ray diffraction techniques. Crystals that promise to be equally useful for structural studies have been prepared from two other lectins, WGA (Wright, 1974; Wright et al., 1974) and favin (Wang et al., 1974),and analysis of these crystals is currently in progress. In the case of WGA an electron-density map has been obtained at 2.2 A resolution by the isomorphous replacement method (C. Wright, personal communication). This map has confirmed previous studies on WGA in solution, which showed that the overall WGA molecule is a dimer of two identical subunits each of which has a molecular weight of about 17,500 (Nagata and Burger, 1974; Rice and Etzler, 1974). In the crystal these two monomers appear in close association with each other across an exact twofold axis. Each monomer contains an assembly of four spatially distinct but structurally similar domains of approximately 41 amino acids each. The domains are related to each other in such a way as to maximize the exposure of amino acid side chains to the solvent; nearly 75%of the WGA side chains have access to the surrounding solvent. Four intradomain disulfide bonds bridge the polypeptide chain within each domain, and this accounts for the unusually high content of cysteine in WGA (approximately 32 of 164 amino acid residues). The WGA molecule contains no regular secondary structure such as an a-helix or a P-pleated sheet. However, each domain contains one irregular a-helical turn and numerous hydrogen bonds between neighboring regions of backbone. The specific binding sites for [(D-gkNAc),], of which there are two per WGA monomer, have recently been identified (Wright, 1977). Studies on the structure of favin were undertaken because it has the same saccharide specificity as Con A. These studies may clarify some of the unresolved questions about Con-A mitogenesis, and it will be of interest to compare the saccharide-binding sites of these two lectins.

292

JAY C. BROWN AND RICHARD C. HUNT

Much relevant structural information about other well-studied lectins can be found in the review by Lis and Sharon (1973b).

111. Lectin-Induced Lymphocyte Mitogenesis A. NATURE OF

THE

MITOGENIC RESPONSE

One of the most remarkable effects of certain lectins is their ability to stimulate a mitogenic response in normal lymphocytes. This property was first discovered in 1960 by Nowell, working with human lymphocytes and red kidney bean agglutinin (PHA). When lymphocytes are treated with a mitogenic lectin, they first undergo “blast” transformation; they increase progressively in size, endocytosis is stimulated, cytoplasmic vacuoles appear, and many metabolic processes are stimulated including DNA, RNA, protein, and lipid synthesis. Certain ion transport activities also increase. Cells ultimately divide and continue to do so as long as the lectin is present. After many cell divisions in the presence of the lectin, cell division becomes independent of the lectin and cells can be cultured indefinitely. This strategy has been extensively employed for establishing continuously growing lines of lymphoid cells from many sources, and it is also routinely employed to expand blood lymphocyte populations for karyotyping purposes. Not all lectins are mitogenic, and it is difficult to predict in advance whether a lectin will be mitogenic or not. The mitogenicity of several well-known lectins is indicated in Table IV. Studies of Wands et al. (1976) have suggested that the multiple valence of Con A is required for its mitogenicity, (but see Fraser et al., 1976)and this may be true of other mitogenic lectins as well. Whereas some lectins are mitogenic for both T and B lymphocytes, others are mitogenic for T lymphocytes only. No known lectin is mitogenic for B lymphocytes only, although other structures such as bacterial lipopolysaccharides have this property. Some instances of specific mitogenicity are indicated in Table IV along with the concentrations of lectin required for optimal mitogenesis. In contrast to their soluble form, insolubilized or aggregated T-cell-specific lectins including Con A and PHA may be mitogenic for B as well as for T cells (Andersson et al., 1972; Greaves and Bauminger, 1972).

B. MECHANISM OF LECTIN-INDUCED MITOGENESIS The mechanism of lectin-induced mitogenesis has now been widely studied, because of its similarity to specific antigen-mediated

293

LECTINS TABLE IV MITOGENICITY OF SELECTED LECTINS ~

Lectin Mitogenic Con A Lentil lectin Lima bean lectin Pokeweed mitogen PHA Ulex europeus agglutinin I Vicia faba agglutinin (favin)

~~

Mitogenic concentration (pg/ml)

Specificity

5 50

T cells only

5 5 -

T and B cells T cells only T and B cells Unknown

-

Nonmitogenic Dolichos bijloms lectin RCAI and RCAI, Soybean lectin WCA

lymphocyte activation and because it offers a convenient model system for studying the regulation of cell growth and proliferation. PHA and Con A are the lectins most frequently examined, and they are used in conjunction with peripheral blood lymphocytes from human or other mammalian sources; the uptake of radioactively labeled thymidine into DNA is measured as an overall index of the mitogenic response. When sensitive lymphocyte cultures are treated with a lectin, they reach and maintain a maximal rate of DNA synthesis between 36 and 72 hours after the lectin is applied. This experimental system has now been quite thoroughly studied in an effort to understand the chain of molecular events involved in the induction of mitogenesis. The system has also been adapted for clinical use to identify patients with impaired immunological function.

1. Lectin Binding to the Lymphocyte Cell Surface It is clear that the first step in mitogenesis is binding of the lectin to carbohydrate-containing groups (glycoproteins or glycolipids) on the cell surface; if the lectin is prevented from binding, no mitogenesis occurs. However, Stobo et al. (1972) and Inbar et al. (1973b)showed that relatively few (6-25%) of the potential Con-A-binding sites on the surface of rat or mouse lymphocytes need to be occupied to induce an optimal mitogenic response. Furthermore, there is now good evidence (Anderson et al., 1972; Greaves and Bauminger, 1972) that some lectins, including Con A, PHA, and pokeweed mitogen, can in-

294

JAY C. BROWN A N D RICHARD C. HUNT

duce mitogenesis without entering the cell. Lectins covalently attached to large, insoluble supports such as plastic culture dishes or Sepharose beads are found to be mitogenic under conditions in which only a negligible amount of the lectin is lost from the support. This result strongly suggests that plasma membrane events may be of crucial importance in the initial phases of mitogenesis and draws attention to the need for detailed studies of the cell surface receptors for mitogenic lectins. Progress in this area is discussed in Section VII. The differential response of T and B lymphocytes to Con A and PHA cannot be explained by a difference in the total amount of lectin bound to the cell surface (Greaves et aZ., 1972), since both mouse T and B cells bind the same amount (approximately 10’ molecules) of Con A (Stobo et al., 1972). The difference must involve events that occur after lectin binding. However, migration of lectin-lectin receptor complexes on the cell surface to form large “patches” or a “cap” does not appear to be required for mitogenesis (Edelman, 1974). 2. Commitment to Mitogenesis Gunther et al. (1974) intensively studied the time course with which mouse lymphocytes become committed to DNA synthesis after they have bound Con A. Sensitive lymphocytes are exposed to a mitogenic concentration of Con A for various periods of time and then washed free of lectin with a-methyl-smannopyranoside. The mitogenic response is determined by measuring the total thyn~idine-~H uptake between 48 and 72 hours after the addition of lectin and by auH the same period. toradiography of cells exposed to t h ~ m i d i n e - ~for The results show that lymphocyte cultures must be exposed to Con A for 20 hours before a maximal mitogenic response is observed; shorter periods of exposure result in less total DNA synthesis (Novogrodsky and Katchalski, 1971; Lindahl-Kiessling, 1972). Cell autoradiographic experiments have demonstrated that this is due to the fact that lymphocytes are heterogeneous with respect to the time of exposure to Con A required to induce them to become committed to mitogenesis. That is, different lymphocytes become committed to DNA synthesis at different times during the 20-hour interval of exposure to Con A. Once a cell has become committed to mitogenesis, however, DNA is synthesized at a rate independent of the amount of time the cell was exposed to Con A. The interpretation of these results is not entirely clear. It is possible, for example, that resting lymphocytes in the GI phase of the cell cycle are actually undergoing a secondary or subcell cycle within GI and that they are sensitive to Con-A mitogenesis only during a portion

LECTINS

295

of this subcell cycle. On the other hand, it may be that Con-A-treated lymphocytes accumulate a metabolite required for mitogenesis at different rates and that commitment to DNA synthesis results when a critical concentration of this (unknown) metabolite is reached. Further experimentation will be required to settle this issue, but it is clear that the heterogeneity of lymphocyte response to Con A must be considered in attempts to interpret the biochemical results described in the following discussion.

3. Biochemical Changes Accompanying Mitogenesis Numerous biochemical studies have been undertaken in an attempt to account for the crucial molecular events that follow lectin binding to lymphocytes and which result in mitogenesis. Although the problem has not yet been solved, much relevant information is now available. For example, one of the earliest changes detectable in human peripheral blood lymphocytes treated with Con A or PHA is a dramatic increase in the intracellular concentration of 3',5'-cyclic GMP (cGMP). The concentration increases 10- to 50-fold in the first 30 minutes following cell exposure to mitogen and then returns to the original level after approximately an hour (Hadden et al., 1972; Watson, 1975).No corresponding change is observed in the intracellular concentration of 3',5'-cyclic AMP (CAMP)and, in fact, CAMP has been found to inhibit mitogenesis (Hadden et al., 1970; Watson, 1975).The mitogenic effects of cGMP (Watson, 1975) and of phorbol myristate acetate (Goldberg et al., 1973; Mastro and Mueller, 1974; Wang et al., 1975b), which increases the intracellular concentration of cGMP, suggest that the transitory rise in cGMP concentration may be functionally involved in mitogenesis. If so, then cGMP must serve as a signal or trigger to induce subsequent events, because its level is not elevated throughout the mitogenic response. A significant transitory increase in the fluidity of human lymphocyte membranes following the treatment of cells with Con A or PHA has been found to follow the same time course as the rise in intracellular cGMP concentration (Barnett et d . , 1974), and these two phenomena may be related. A dramatic increase in the metabolic turnover of phosphatidylinositol, but not other membrane phospholipids, has also been shown to take place within the first 30 minutes after stimulation of human lymphocytes with PHA (Fisher and Mueller, 1971). Quastel and Kaplan (1970a,b) carried out experiments which indicate that an increased intracellular Concentration of K+ may be necessary for mitogenesis. A two- to threefold increase in the rate of K+ influx is observed beginning approximately 1 hour after treatment of

296

JAY C. BROWN AND RICHARD C. HUNT

human lymphocytes with PHA. No corresponding increase in the rate of K+ efflux is observed, so the intracellular level of K+ probably rises slightly. A similar increase in K+ influx is observed in quiescent 3T3 cells stimulated to divide by the addition of serum (Rozengurt and Heppel, 1975). Increased K+ transport is found to be due to an increased number of K+ carriers on the lymphocyte cell surface rather than to an increased rate of functioning by a fixed number of carriers (Wright et al., 1973).The increased rate of K+ influx is probably due to an increased number of membrane-associated Na,K-ATPase molecules (Dahl and Hokin, 1974), since ouabain, a specific inhibitor of the Na,K-ATPase, blocks K+ uptake. Ouabain is also a potent inhibitor of mitogenesis; it can act in a reversible fashion at any stage of the mitogenic response. High concentrations of K+, however, can overcome all effects of ouabain (Quastel and Kaplan, 1970a). This argues strongly that continued functioning of the Na,K-ATPase is required throughout the mitogenic response. Significant evidence also exists suggesting that increased Ca2+ transport may be functionally involved in lectin-induced mitogenesis. A twofold increase in the rate of Ca2+uptake is observed within an hour after PHA stimulation of human lymphocytes, and this increase persists throughout the period of blast transformation and DNA synthesis (Allwood et al., 1971; Whitney and Sutherland, 1972,1973). No mitogenesis is observed in the absence of Ca2+.The increased rate of Ca2+uptake is found to be due to an increased affinity of membrane carriers for Ca2+(Whitney and Sutherland, 1973). A functional role for Ca2+in the induction of mitogenesis is suggested by the fact that the specific Ca2+ionophore A23187 can stimulate blast transformation and mitosis; it also potentiates PHA mitogenesis (Maino et al., 1974; Luckasen et al., 1974).Although sensitive lymphocytes must be exposed to the ionophore for a considerable length of time (several hours), a low M) of ionophore can induce blastogenesis in a subconcentration ( stantial fraction (50%) of lymphocytes. The requirement for Ca2+ in mitogenesis may be related to the proposed involvement of intracellular filamentous elements. Both microtubules and microfilaments have been implicated in mitogenesis by the observation that both vinca alkaloids (Edelman, 1974),which disrupt microtubule structure, and cytochalasin B (Yoshinagaet al., 1972), which inhibits microfilament formation, are potent inhibitors of lectin-induced mitogenesis. Lectin stimulation of lymphocytes leads to an increased uptake of several other cellular metabolites including glucose (Peters and Hausen, 1971)and uridine (Cooper, 1972).Compared to K+ and Ca2+,however, stimulation of these transport activities occurs relatively late (10

LECTINS TABLE V METABOLIC CHANGESOBSERVED DURING THE LECTIN-INDUCED MITOGENESIS OF LYMPHOCYTES Biochemical response

Greatest effect observed

Increased cGMP con20 minutes centration Increased membrane flu- 30 minutes idity Increased K+ transport 2 hours

Required continuously

No No Yes

Increased lipid turnover

2 hours

Increased uridine and glucose concentration Increased Ca2+concentration

10-20 hours

Yes

48 hours

Yes

48 hours

Yes

DNA synthesis

Probably

Reference Hadden et al. (1972); Watson (1975) Barnett et al. (1974) Quastel and Kaplan (1970a,b) Fisher and Mueller (1971) Cooper (1972); Peters and Hausen (1971) Whitney and Sutherland (1973); Allwood et al. (1971) Gunther et al. (1974)

to 20 hours) after the exposure of cells to lectin. This suggests that increased glucose and uridine uptake are probably secondary consequences of earlier mitogenic events and not directly involved in the induction of commitment to blastogenesis. A summary of some of the metabolic changes observed in lectin-stimulated lymphocytes is given in Table V. It is clear from these results that we are a long way from the goal of understanding the complete sequence of molecular events, which begins with the binding of lectin to the cell surface and ends with DNA replication and cell division. However, a most promising start has been made on this project, and there is every reason to believe that further experimental effort will not be in vain.

IV. Selective Agglutination of Transformed Cells A.

CORRELATIONOF LECTINAGGLUTINABILITY WITH TRANSFORMATION

1. Situations in which a Positive Correlation Exists Although lectins were originally identified by their ability to agglutinate erythrocytes, most are found to clump other animal cell types as well. In fact, a great many cell types are agglutinable by lectins, and

298

JAY C. BROWN AND RICHARD C. HUNT

this property has been thoroughly studied in experiments designed to probe molecular events taking place at the cell surface. These studies have revealed striking changes in lectin agglutinability accompanying neoplastic transformation, infection of cells with some nononcogenic viruses, and certain developmental processes. By far the best studied of these phenomena is the change in lectin agglutinability that accompanies neoplastic transformation; transformed cells are usually found to be agglutinated at a much lower concentration of lectin than is required for the corresponding normal, untransformed cell type. This difference has been extensively studied in an effort to identify the particular features of transformed cells that render them more agglutinable than normal cells. Studies on this topic have been the subject of previous reviews by Burger (1970a, 1971a,b), Rapin and Burger (1974), Nicolson ( 1 9 7 4 ~1976b), ~ and Poste (1975). Selective lectin agglutination of transformed cells was first observed by Aub et al. (1963), who found that, when a crude preparation of wheat germ lipase was added to suspensions of isolated mouse lymphoma or ascites tumor cells, the cells agglutinated, while normal, untransformed cells remained dispersed. The properties of the agglutination reaction suggested that the active factor in producing aggregation was not the lipase itself, but a contaminant in the preparation. This contaminant was later purified and called WGA (Burger and Goldberg, 1967). In the relatively short period since these original observations were made, many lectins have been examined and found to agglutinate selectively the transformed, but not the normal, cells from a variety of sources, including cells of both fibroblastic and lymphoid origin. References to some of the relevant literature are given in Tables VI and VII. These studies have shown clearly that the preferential agglutination of transformed cells by lectins is independent of the means by which transformation was achieved. For example, whereas normal mouse, rat, and hamster fibroblasts show no agglutination at 500 pg/ml Con A, considerable agglutination is observed after transformation with SV-40 virus, or polyoma virus, x rays, or chemical carcinogens such as dimethylnitrosamine and benzopyrene (Inbar and Sachs, 1969a; Inbar et al., 1972a; Weber, 1973) at Con-A concentrations as low as 1 pg/ml. Furthermore, early studies with normal and transformed spleen cells, and with normal and regenerating liver cells, indicated that agglutination by WGA is not the result of cellular alterations associated with rapid growth, but rather with the acquisition of transformation-specific functions (Aub et al., 1965). More recent studies have served to confirm the correlation between

TABLE VI ENHANCED LECTIN-MEDIATED AGGLUTINATION

Lectin Con A

Cell type Mouse 3T3

Mouse 3T3

Mouse 3T3 Hamster BHK

Chick embryo fibroblasts

Hamster fibroblasts

OF

TRANSFORMED CELLS

Transforming agent or transformed cell type

References

SV-40 virus

Inbar and Sachs (1969a,b); Ben-Bassat et al. (1970); AmdtJovin and Berg (1971); Cline and Livingston (1971); Nicolson (1971); Ozanne and Sambrook (1971a,b); Culp and Black (1972); Tomita et al. (1972a); Van der Noorda et al. (1972); Yin et al. (1972); Noonan et al. (1973a); Van Nest and Grimes (1974); Poste et al. (1975~); Poste and Nicolson (1976) Eckhart et al. (1971); Polyoma virus Inbar et al. (1971); DePetris et al. (1973); Noonan and Burger (1973a); Sakiyama and Robbins (1973) Van Nest and Grimes Murine leukemia (1974) virus Amdt-Jovin and Berg Polyoma virus (1971); Cline and Livingston (1971); Ozanne and Sarnbrook (1971a); Poste (1972); Weber (1973); Nicolson et al. (1975a); Poste et al. (1975b) Biquard and Vigier RSV (1972); Burger and Martin (1972); Kapeller and Doljanski (1972); Lehman and Sheppard (1972) Ben-Bassat et al. (1971); Various agents inInbar et al. (1972a); cluding polyoma Poste (1972) viruses and SV-40 and RSV (Continued)

300

JAY C. BROWN AND RICHARD C. HUNT

TABLE VI (Continued)

Lectin Con A

Cell type

Transforming agent or transformed cell type

Mouse 3T3

SV-40 virus

Inbar and Sachs (1969a) DeSalle et al. (1972); Ben-Bassat et al. (1974, 1976) Borek et al. (1973); Becker (1974); Nakamura and Terayama (1975) Salzberg and Green (1972); Vesely et al. (1972); Noonan e t al. (1976) Tomita et al. (1972a)

Rat hepatocytes Mouse 3T3 Rat and human fibroblast-type cells Mouse 3T3

Hepatoma SV40 virus RSV

Borek et al. (1973) Tomita et al. (1972a) Veselg et al. (1972)

Lymphocytes

Leukemic cells

Hepatocytes

Hepatoma cells

Sarcoma and leuVarious fibrokosis viruses blast-type cells

Lentil lectin and PHA Pea lectin

RCA

Soybean lectin

WGA

References

Hamster BHK Mouse 3T3 Hamster and rat fibroblast-type cells Mouse 3T3

Mouse 3T3

Tomita et al. (1972a); Nicolson (1973a); Nicolson and Lacorbiere (1973) Polyoma virus Nicolson e t al. (1975a) Inbar et al. (1971) S V 4 0 virus Sela et al. (1970, 1971); Various agents ineluding SV40 and Inbar et al. (1972a) polyoma viruses and RSV SV-40 virus Burger (1969, 1970a); Pollack and Burger (1969); Inbar et al. (1971); Ozanne and Sambrook (1971a); Sheppard (1971); Sivak and Wolman (1972); Tomita et al. (1972a) Polyoma virus Benjamin and Burger (1970); Biddle et al. (1970); Eckhart e t al. (1971);Inbar et al. (1972a); Sakiyama and Robbins (1973); Weber (1973) SV40 virus

30 1

LECTINS

TABLE VI (Continued)

Lectin

Cell type

WGA

Transforming agent or transformed cell type

Hamster BHK

Polyoma virus

Chick embryo fibroblasts

RSV

Lymphocytes Hepatocytes

Leukemic cells Hepatoma cells

References Hakomori and Murakami (1968);Burger (1969, 1970a); Ozanne and Sambrook (1971a); Nicolson et al. (1975a) Burger and Martin (1972); Kapeller and Doljanski (1972); Lehman and Sheppard (1972) Aub et al. (1965) Borek et al. (1973)

transformation and sensitivity to lectin-mediated agglutination. For example, Pollack and Burger (1969) showed a good correlation between saturation density (a measure of relative malignancy) reached by mouse 3T3 cells or their transformed derivatives growing in tissue culture and the WGA concentration causing agglutination. Using SV-40 virus-transformed 3T3 cells, 3T12 (spontaneously transformed) TABLE VII LECTIN AGGLUTINATION OF MOUSE 3T3 AND SV40-TRANSFORMED 3T3 CELLS ~~

Lowest lectin concentration for agglutination (pg/ml)

Lectin

3T3

Con A"

1000

Lentil lectinb Pea lectinb PHAb RCAb Soybeanlectin" WGA"

2000 2000 500 250 500 500

a

Purified lectin. Crude extract.

Trypsinized 3T3 62 500 250 62.5 31.25 100 6

SV-40 3T3

Trypsinized SV40 3T3

62

-

500 250 31.25 31.25 20 12

250 250 7.8 7.8 -

-

Reference Ozanne and Sambrook (1971a,b) Tomita et al. (1972a) Tomita et al. (1972a) Tomita et al. (1972a) Tomita et al. (1972a) Sela et al. (1970) Ozanne and Sambrook (1971a,b)

302

JAY C. BROWN AND RICHARD C. HUNT

cells, 3T3E cells (which lack thymidine kinase) and F1-SV101 cells (“flat” revertants of SV-40-transformed 3T3 cells), these investigators showed that the WGA concentration required for agglutination falls as the cell saturation density increases. Weber (1973) and Inbar et al. (1972a) also observed an inverse relationship between the concentration of lectin required for agglutination and the saturation density to which many normal and transformed cell lines grow in culture. Nontransforming mutants of polyoma virus fail to produce the same increase in lectin-mediated agglutination observed after infection of 3T3 cells with transforming viruses (Benjamin and Burger, 1970).Further, when a mutant of polyoma virus temperature-sensitive for producing the transformed cell phenotype (ts 3) was used to infect BALB/3T3 or BHK cells, it was found that agglutinability by WGA or Con A occurred only at the permissive temperature for transformation (Eckhart et al., 1971). Renger and Basilico (1972) and Noonan et al. (1973a) obtained similar results using a temperature-sensitive mutant of SV-40-transformed3T3 cells. Similar studies have been carried out with a mutant of Rous sarcoma virus (T5) which transforms chick embryo cells in such a way that cells express the transformed phenotype at 36°C but not at 41°C. These cells exhibit enhanced agglutinability by Con A and WGA at the low but not at the high temperature (Burger and Martin, 1972; Biquard and Vigier, 1972). However, the correlation of agglutinability and some morphological aspects of transformation does not always exist in this system. For example, when cells are shifted from 41” to 36”C, it is possible to inhibit expression of the morphological aspects of transformation by blocking protein synthesis with puromycin or cycloheximide. This treatment, however, does not prevent the increase in lectin agglutinability (Biquard, 1973) associated with transformation. Further evidence supporting the correlation between transformation and lectin agglutinability has been obtained from studies of mutant cell lines called revertants, which are derived from transformed cells but which lack many of the properties of the parent transformed cells. Inbar et al. (1969) examined the properties of revertant polyoma virus-transformed hamster embryo cells that retained the polyomaspecific nuclear (T) antigen. They found a partial or complete loss of Con-A agglutinability. Flat revertants of SV-40-transformed 3T3 cells have been produced by treating SV-3T3 cells with 5-fluoro-2’-deoxyuridine, a compound that selectively kills actively dividing cells and which is employed to select for cells subject to contact or densitydependent inhibition of growth (Pollack et al., 1970; Culp et al., 1971). Revertants of this type were found to have lost their trans-

LECTINS

303

formed phenotype; they had regained sensitivity to growth control, the normal 3T3 cell morphology had reappeared, and the cells had lost their agglutinability at low lectin concentrations. At the same time, these cells had retained the viral genome, as demonstrated by the fact that SV-40 T antigen was still expressed and SV-40 virus could be recovered from the revertants by fusing them with permissive msnkey cells. Similarly, revertants of SV-40-transformed 3T3 cells have been selected by growing these cells in the presence of WGAand Con A (Ozanne and Sambrook, 1971b; Culp and Black, 1972; Ozanne, 1973). Although these revertants were found to retain the SV-40 genome, they acquired many of the properties of untransformed cells including loss of sensitivity to agglutination by Con A and WGA. The treatment of transformed fibroblasts with dibutyryl cAMP has now been extensively studied as a method for phenotypically reverting these cells to an apparently normal state. The randomly oriented, multilayered transformed cell cultures are converted by dibutyryl CAMP into elongated, parallel, fibroblast-like cells which have regained density-dependent inhibition of growth and other characteristics of untransformed cells. These changes take place over a period of hours in culture, and they are completely reversible when dibutyryl cAMP is removed (Hsie and Puck, 1971; Sheppard, 1971,1972; Johnson et al., 1971). In oncornavirus-transformed mouse cells, a fall in Con A-mediated agglutinability accompanies this phenotypic reversion to the untransformed state (Hsie et al., 1971; Kurth and Bauer,

1973). 2. Exceptions to the Rule Although there are now many instances in which cell transformation correlates well with agglutinability at low lectin concentration, this is not always the case. There are many normal cell types that are agglutinated at low lectin concentration, including erythrocytes, cells of embryonic origin (see Section VI), sperm (see Section VI), rat lung cells, monkey kidney cells (Sivak and Wolman, 1972), mouse spleen cells, and mouse bone marrow cells (Liske and Franks, 1968). Moreover, there are instances in which normal cell lines are more susceptible to lectin-induced agglutination than the transformed cells derived from them. For example, normal rat liver cells are agglutinable by lentil lectin, but rat hepatoma cells are not (Borek et d., 1973),and Con-A agglutinates mouse lymphocytes to a greater extent than L1210 leukemic cells (Burger, 1973). Con A also agglutinates malignant mouse mammary cells to a lesser extent than nonmalignant variants (Hozumi

304

JAY C. BROWN AND RICHARD C. HUNT

et al., 1972). In a survey of one nonneoplastic and five neoplastic cell lines, Gantt et al. (1969) found that the nonneoplastic and two of the five neoplastic lines agglutinated at the same WGA concentration. Similarly, Ukena et al. (1976) identified several established lines of mouse and hamster cells in which Con-A agglutinability did not correlate with the ability of cells to escape from density-dependent inhibition of growth in culture. No correlation was observed between transformation and RCA or Con-A agglutinability among several different types of human lymphoid cells (Glimelius et al., 1974, 1975; Maca, 1976),and a similar lack of correlation between Con-A or WGA agglutinability and transformation of several hamster cell lines has been reported (Berman, 1975). Thus, in spite of the frequent association of lectin agglutinability and transformation, the two do not always go together; agglutinability is not an invariant feature of the transformed cell phenotype. Indeed, from Table VI it can be seen that most investigators who have concluded that transformed cells are more agglutinable than normal cells have utilized only a few normal cell types, predominant among which are 3T3, BHK, and CEF cells. Furthermore, in most cases one of three transforming viruses (SV-40,polyoma, or RSV) has been employed. It is possible, therefore, that the correlation of agglutinability with transformation would become much less clear if a wider range of cells were investigated. In monolayer cultures, transformed fibroblasts are found to differ in their lectin agglutinability, depending upon whether cultures are sparse or confluent. For example, Inbar et al. (1971) and Ben-Bassat et al. (1971) found that 1 day after subculturing transformed hamster cells were not agglutinable b y Con A. Maximum agglutination was attained after 4 days of growth, when cultures had become confluent. Noonan and Burger (1973b) made similar observations with polyomatransformed 3T3 cells. Agglutination of rat c6 glioma cells is also density-dependent; as cells grow to a higher density in culture, their sensitivity to Con-A-mediated agglutination increases. This density-dependent agglutination appears to be related to the anchorage of cells to a solid surface, since cells grown in suspension do not show the same effect. It has been proposed that the extracellular matrix produced by c6 cells may be involved in this phenomenon (Skehan and Friedman, 1975). In the case of spontaneously transformed mouse 3T6 cells, however, there is a decrease in Con-A-mediated agglutinability when the cells reach saturation density (Goto et al., 1972).Agents that decrease the saturation density in culture also decrease the Con-Amediated agglutinability. Similar results were obtained with 3T3 and SV-40-transformed3T3 cells when both Con A and RCA agglutinability were examined as a function of cell density (Nicolson and Lacor-

LECTINS

305

biere, 1973; Nicolson, 1974a). It was noted that, while RCA agglutinability decreases in 3T3 cells at confluence, more RCA is bound at this time. B. LECTINAGGLUTINATION AND TUMOFUGENICITY The generally satisfactory correlation between lectin agglutinability and transformed cell growth in tissue culture has motivated several studies designed to determine whether lectin agglutinability also correlates well with cell tumorigenicity in vivo. In some cases a positive correlation has been observed. For example, DeMicco and Berebbi (1972) examined Chinese hamster ovary (CHO) fibroblast-like cells and found that the most tumorigenic cells were also best agglutinated by Con A. Similar studies with mouse 3T3 cells showed that Con-A agglutinability correlates well with tumorigenicity but does not predict which cells will form progressively growing tumors which will eventually kill the host (Van Nest and Grimes, 1974). In other cases, however, no correlation is observed between lectin agglutinability and tumorigenicity. Sakiyama and Robbins (1973) and Berman (1975) found no correlation between Con A or WGA agglutinability and the ability to form tumors in vivo among several different hamster cell lines including N1L 2 and BHK-21. Dent and Hillcoat (1972) and Gantt et al. (1969) made similar observations with mouse lymphoid cells, while Glimelius et al. (1975) showed that neoplastic character does not correlate with Con A or RCA agglutinability in several human lymphoid cell types. In fact, two studies, one involving mouse mammary tumor cells (Hozumi et al., 1972) and the other using mouse lymphoid cells (Smets and Broekhuysen-Davies, 1972), have indicated that transplantability of tumor cells in vivo is correlated with a low rather than a high sensitivity to agglutination by Con A; highly transplantable tumor cells were weakly agglutinable, while weakly transplantable or nontransplantable cells were agglutinable at low Con-A concentrations. It is clear therefore that, insofar as a correlation exists, it is between lectin agglutinability and the expression of transformed cell growth properties in culture. Agglutination does not correlate well with the ability to escape immunological attack and form a progressively growing tumor in uivo.

c.

TRANSIENT AGGLUTINABILITY OF UNTRANSFORMED CELLS

1. Agglutinability of Protease-Treated Cells Although untransformed cells are ordinarily resistant to agglutination by lectins, two situations exist in which normal cells become tran-

306

JAY C. BROWN AND RICHARD C. HUNT

siently agglutinable. One is when they are treated with proteolytic enzymes. Brief proteolysis renders normal cells as agglutinable as transformed cells with WGA (Burger, 1969; Fox et al., 1971), Con A (Inbar and Sachs, 1969a), RCA (Nicolson and Blaustein, 1972), and soybean agglutinin (Sela et al., 1970,1971), but it does not ordinarily affect the agglutinability of transformed cells. Representative results for Con A and trypsinized 3T3 cells are shown in Table VII. Only a very gentle proteolytic digestion is required to produce increased agglutinability in untransformed fibroblast-type cells. For instance, Burger (1969) found that untransformed BHK cells became as agglutinable with WGA as polyoma virus-transformed BHK cells after treatment with 25 pg/ml trypsin for 5 minutes. Similar results are observed with other cell types (Rapin and Burger, 1974). Trypsin treatment of revertants (Inbar et al., 1969) and temperature-sensitive mutants of transformed cells at the nonpermissive temperature (Eckhart et al., 1971) renders them as agglutinable as the corresponding transformed cell type. The changes in the normal cell surface produced by proteolytic digestion, however, are not permanent. Trypsinized normal cells revert to their nonagglutinable state after a few hours of growth in tissue culture. The similarity, as judged by lectin agglutinability, of transformed cells and normal cells treated with trypsin has motivated a wide variety of studies on the molecular details of how trypsin affects normal cell surfaces. These studies have indicated that trypsin treatment increases the agglutinability of untransformed fibroblasts without changing the total number of Con A or WGA binding sites available on the cell surface (Rosenblith et al., 1973; Cuatrecasas, 1973a; Inbar et al., 1971; Nicolson, 1973a). This is consistent with the fact, as discussed below, that normal and transformed fibroblast-type cells do not usually differ in the number of binding sites for Con A or WGA (Ozanne and Sambrook, 1971a; Arndt-Jovin and Berg, 1971; Cline and Livingston, 1971; Inbar et al., 1971; Phillips et al., 1974; Trowbridge and Hilborn, 1974; Nicolson et al., 1975a; but compare Noonan and Burger, 1973a,b). However, proteolytic treatment of normal cells does increase the ability of Con A and WGA receptors to move laterally in the plane of the plasma membrane (Nicolson, 1972; Huet and H e n berg, 1973; Rosenblith et al., 1973; Garrido et al., 1974) to the point where the receptors of treated normal cells come to exhibit the same high degree of lateral mobility found in transformed cells. Trypsin treatment of normal cells also results in a loss of intracellular actin cables (Pollack and Rifkin, 1975) and in the digestion of a high-molecular-weight cell surface protein called LETS (large, external transformation-sensitive glycoprotein; Hynes, 1974). Both of these structures

LECTINS

307

are found in normal but not in transformed fibroblasts. Any of these factors, or a combination of them, may be responsible for the increased agglutinability of trypsinized compared to untreated normal cells. It is attractive in the present context, however, to speculate that the generation of extracellular proteolytic activity may be functionally involved in the natural expression of the transformed cell phenotype. Transformed cell surfaces may actually come to resemble those of trypsinized normal cells, because transformed cells themselves synthesize and secrete a proteolytic enzyme which has the same effect on the cell surface as exogeneously added trypsin. According to this hypothesis, a crucial step in the transformation process would be the induction of a proteolytic enzyme activity that could be expressed extracellularly. This view is supported by the fact that proteolytic treatment causes untransformed cells to lose their sensitivity to densitydependent inhibition of growth in culture (Burger, 1970b; Sefton and Rubin, 1970) and therefore to resemble transformed cells in this key property. Furthermore, many transformed fibroblasts are found to express an extracellular proteolytic activity (Unkeless et al., 1973; Goldberg, 1974). This proteolytic function is generated by a cellular plasminogen activator interacting with serum plasminogen to produce plasmin (Unkeless et al., 1974), a potent trypsin-like proteolytic enzyme. Functioning of this system appears to be required for expression of certain aspects of transformed cell growth (Ossowski et al., 1974; Schnebli, 1974; Weber, 1975; but compare Chou et al., 1974; McIlhinney and Hogan, 1974). It is therefore possible that added trypsin converts normal fibroblasts phenotypically to the transformed state by mimicking the effect of a proteolytic system normally involved in the expression of transformed cell behavior. Nicolson (1973a) observed that treatment of both normal and SV-40 virus-transformed 3T3 cells with neuraminidase increases their sensitivity to agglutination by RCA in much the same way that trypsin treatment increases agglutinability; similar results were obtained with normal and polyoma virus-transformed BHK cells (Nicolson et al., 1975a). At the same time, binding of RCA-lZ5Iincreased two- to threefold in both cell types. These observations are best explained by assuming that the removal of terminal sialic acids with neuraminidase exposes subterminal galactose groups which are then available for binding to RCA. This interpretation is supported by the observation that neuraminidase treatment is required before many cell surface glycoproteins will react with galactose oxidase (Gahmberg and Hakomori, 1973; Hunt and Brown, 1974). Unlike RCA-mediated agglutination, the agglutination of transformed cells by WGA (Burger and Gold-

308

JAY C. BROWN AND RICHARD C. HUNT

berg, 1967; Kapeller and Doljanski, 1972; Greenaway and Levine, 1973)and Con A (Nicolson et al., 1975a) is reduced by neuraminidase treatment of the cell surface. Associated with these changes is a decrease in WGA binding but no change in Con-A binding (Nicolson et al., 1975a). 2. Agglutinability during Mitosis A second situation in which untransformed cells become transiently agglutinable with lectins is during the process of cell division (the M phase of the cell cycle). For example, Shoham and Sachs (1974a,b) found that normal hamster fibroblasts become agglutinable with low concentrations of Con A and WGA specifically during mitosis, but that the cells revert to the nonagglutinable state during interphase. A similar situation applies to 3T3 (Collard et al., 1975) and BHK (Glick and Buck, 1973) cells that are found to be agglutinable only in M phase. Transformed cells also vary in their sensitivity to lectin agglutination during mitosis, but some become more, and others less, agglutinable. For example, hamster fibroblasts transformed by RSV or by dimethylnitrosamine are agglutinated by Con A or WGA during interphase but not during M phase (Shoham and Sachs, 1974a,b), while SV-40-transformed 3T3 cells (Collard et al., 1975) and Epstein-Barr virus-transformed human lymphocytes (Smets, 1973) are most agglutinable during M phase. As in the case of trypsinized compared to untreated normal fibroblasts, it is not yet known why normal cells in mitosis are more agglutinable than similar cells in interphase. Some studies have shown that mitotic normal cells bind more Con A and WGA than interphase cells when the lectin is present at a very low concentration (Fox et al., 1971; Shoham and Sachs, 1972, 1974b; Noonan and Burger, 1973a; Noonan et al., 1973b). At a high lectin concentration, however, cells in all phases of the cycle bind similar amounts of lectin (Shoham and Sachs, 1974b). It is not likely therefore that increased lectin agglutinability during mitosis can be explained by increased lectin binding. Increased agglutinability of normal cells in M phase is, however, consistent with many other observations suggesting that the surfaces of normal cells in mitosis bear an intriguing similarity to transformed cell surfaces. For instance, Glick and Buck (1973) found that the set of fucose-containing glycopeptides solubilized by trypsin treatment of mitotic BHK cells more closely resembled the set of glycopeptides obtained from transformed cells than the set obtained from interphase normal cells. Also, a study (Garrido, 1975) of Con-A and WGA receptor mobility in CHO cells revealed that, whereas these receptors are relatively immobile during interphase, they display the

LECTINS

309

same high degree of lateral mobility during mitosis that one finds for these receptors in transformed cells. D. MECHANISM OF AGGLUTINATION

1. Role o f t h e Lectin The striking difference in lectin agglutinability between normal and transformed cells has motivated many studies on the molecular basis of this phenomenon. It is quite reasonable to suppose that, if one could rationalize the selective agglutinability of transformed cells, one would at the same time learn something significant about the role of the cell surface in aspects of transformed cell behavior such as uncontrolled growth, invasiveness, metastasis, and resistance to immunological attack. One expects that agglutination of cells by a multivalent lectin would involve binding of the lectin to carbohydrate groups on adjacent cells. This should result in the cross-linking or aggregation of cells, and the experimental studies carried out so far are fully consistent with this view. For example, when cells agglutinated with fenitin-labeled Con A are examined in the electron microscope, ferritin-Con-A molecules are found between agglutinated cells in the regions where their plasma membranes are attached (DePetris et al., 1973). Also, when the number of carbohydrate-combining sites (the valence) of Con A is reduced by succinylation (Gunther et al., 1973; Trowbridge and Hilborn, 1974) or by treatment with chymotrypsin (Steinberg and Gepner, 1973), its ability to agglutinate a wide variety of cells is also significantly reduced. Conversely, when the functional valence of soybean lectin is increased by cross-linking native molecules with glutaraldehyde to produce dimers, trimers, and tetramers, its ability to agglutinate erythrocytes and lymphocytes is increased 10fold or more (Lotan et al., 197313).The role of the saccharide-combining site is indicated b y the fact that the specific, but not other, saccharide haptens can inhibit cell agglutination by WGA (Burger, 1969) or Con A (Inbar and Sachs, 1969a).In fact, agglutination of trypsinized or SV-40-transformed 3T3 cells by Con A can be reversed by a-MG if the hapten is provided shortly (within approximately 5 minutes) after agglutination takes place (Burger, 1970a; Rottmann et al., 1974). If a lectin is to agglutinate cells by serving as a cross-link or bridge between two adjacent cells, it must do so by binding to carbohydratecontaining structures on the surface of both cells. Cell surface glycoproteins and glycolipids are most likely to serve as lectin receptors, and all cells sensitive to lectin agglutination possess receptors for the agglutinating lectin. For example, transformed fibroblasts have be-

310

JAY C. BROWN AND RICHARD C. HUNT

tween 1 x lo7and 4 x lo7binding sites for Con A and a similar number of sites for WGA. In the case of fibroblast and fibroblast-like cells, however, even normal, untransformed cells, which are resistant to lectin agglutination, still contain a significant number of lectin receptors. In fact, in spite of one group’s contrary results (Noonan and Burger, 1973a,b), there is now general agreement that normal cells do not differ significantly from transformed or trypsinized normal cells in the number of receptors for Con A, WGA, or RCA (Ozanne and Sambrook, 1971a; Cline and Livingston, 1971; Arndt-Jovin and Berg, 1971; Inbar et al., 1971; Nicolson and Lacorbiere, 1973; Phillips et al., 1974; Trowbridge and Hilborn, 1974; Nicolson et al., 1975a). Considerable controversy was involved in establishing this point, as previously described by Nicolson (1974~). It is now clear, however, that one cannot account for the difference in agglutinability between normal and transformed cells by assuming that untransformed cells are lacking or depleted in cell surface lectin-binding sites. A more complicated explanation must apply. 2. Receptor Distribution and Mobility It was suggested quite early in the development of this field that the difference in agglutinability between normal and transformed cells may be due not so much to the total number of lectin receptors as to the way they are arranged on the cell surface. In particular, it was suggested that, if transformed cell lectin receptors were organized into small islands or clusters while normal cell receptors were evenly distributed over the cell surface, and if cells could be cross-linked only in the regions of receptor clusters where multiple cross-links could be formed, one would be able to account for the selective agglutination of transformed cells (Burger, 1970a; Sela et al., 1971). Experimental support for this model was provided by Nicolson (1971), who examined electron micrographs of ferritin-labeled Con A (Con A-FT) bound to the spread membranes of 3T3 and SV-40-transformed3T3 cells. These micrographs were interpreted to support the view that Con-A receptors are aggregated into small patches or clusters on the transformed cell surface but evenly distributed in normal cells (Nicolson, 1971). Similar micrographs of cells agglutinated by Con A-FT showed that the cells were cross-linked preferentially at regions of the plasma membrane where lectin-receptor complexes were located in clusters (Nicolson, 1972). Other studies appeared to confirm these early results. For example, an electron microscope technique involving horseradish peroxidase coupled to Con A was employed to demonstrate clustering of surface Con-A receptors in adenovims-12-trans-

LECTINS

31 1

formed (Rowlatt et al., 1973) and SV-40-transformed (Bretton et al., 1972; Garrido et al., 1974) hamster cells, Ehrlich ascites cells (Roth et al., 1974), chemically transformed rat liver cells (Roth, 1974; Roth et al., 1975), and erythroblasts in chickens infected with avian erythroblastosis virus (Barbarese et al., 1973),whereas Con-A receptors were found to be uniformly distributed in the normal cells from which all of these were derived. Subsequent studies have not confirmed these early results. It has now been clearly demonstrated that clustering or patching of surface lectin-binding sites in transformed cells is induced by the presence of the lectin and is not a property of the unperturbed cell membrane. Binding sites for Con A (Comoglio and Guglielmone, 1972; Inbar et al., 1973a,b; Inbar and Sachs, 1973; Nicolson, 1973b; Rosenblith et al., 1973; DePetris et al., 1973; Garrido et al., 1974; Marikovsky et al., 1974; Temmink et al., 1975; Ukena et al., 1974; Berlin, 1975), WGA (Garrido et al., 1974), and RCA (Nicolson, 1974a) were found to be evenly distributed on the cell surface when normal, trypsinized normal, or transformed fibroblasts were fixed prior to labeling with lectin, or if they were labeled with lectin at 4°C. Clustering of receptors in transformed or trypsinized normal cells was observed only when unfixed cells were exposed to lectin and incubated at 37°C. At this temperature lectin receptors of transformed cells are free to move laterally in the plane of the membrane, and they can be aggregated to form clusters by multivalent lectin molecules. The results of Inbar and Sachs (1973) illustrate this point clearly. These investigators bound a fluorescent (fluorescein isothiocyanate-labeled) derivative of Con A (F-Con A) to normal and transformed cells either with or without prior glutaraldehyde fixation. F-Con-A molecules were then localized on the cell surface by examining cells in the fluorescence microscope. The results as shown in Fig. 3 indicate that, whereas F-Con-A molecules are uniformly distributed on the surface of prefixed normal or transformed cells, they are aggregated in clusters in unfixed transformed cells. Although the Con-A receptors of most transformed or trypsinized normal fibroblasts can be aggregated into patches (for the few exceptions reported, see Martinez-Palomo et al., 1972; Francois et al., 1972), untransformed cells differ among themselves in this property. Some normal cells have Con-A receptors that can be aggregated into clusters (Smith and Revel, 1972; Roth et al., 1973; DePetris et al., 1973; Torpier and Montagnier, 1973; Huet, 1974; Huet and Bernhard, 1974; Marikovsky et al., 1974; Raff et al., 1974; Temmink and Collard, 1974; Collard et al., 1975; Temmink et al., 1975), while in other cases

312

JAY C. BROWN AND RICHARD C. HUNT

FIG. 3. Distribution of F-Con A on the surface membrane of normal and transformed cells. (A) Binding of F-Con A to lymphoma cells after fixation with 2.5% glutaraldehyde. Similar results were obtained with normal lymphocytes, normal fibroblasts, and transformed fibroblasts. The binding of F-Con A shows a completely uniform distribution on the cell surface. (B) Distribution of F-Con A of the type seen with unfixed normal fibroblasts. Little or no change from the uniform distribution is detectable. (C) Distribution of F-Con A of the type seen with unfixed transformed fibroblasts and lymphoma cells. The formation of clusters is clearly evident. (D) Cap formation in normal lymphocytes. Cells were incubated with 100 pg F-Con Nml for 15 minutes at 37"C, washed, and examined in a Leitz Ortholux microscope with transmitted ultraviolet light. x 2500.

the receptors remain dispersed at 37°C in the presence of lectin (Inbar and Sachs, 1973; Nicolson, 1973b; Rosenblith et aZ., 1973).In spite of the difference in ability to form receptor clusters, however, normal cells usually remain resistant to lectin agglutination. It is highly unlikely therefore that a patched distribution of Con-A receptors or the ability of receptors to be aggregated into patches by Con A is directly involved in the selective agglutination of transformed cells. In fact, a recent study devoted to reexamining the role of receptor mobility in the Con-A-induced agglutination of fibroblasts has confirmed the fact

LECTINS

313

that untransformed cells may differ in their ability to form receptor patches without differing in agglutinability (Ukena et al., 1976).It was concluded that the clustering of Con-A receptor sites is neither necessary nor sufficient to account for the difference in agglutinability between normal and transformed cells. It is possible, however, that, although formation of larger clusters or patches of Con-A receptors is not involved in specific agglutination, some degree of lateral receptor mobility may be required. Rutishauser and Sachs (1974,1975) tested this possibility by examining the distribution of Con-A receptors on the surface of single lymphoid cells. Cells immobilized on nylon fibers were allowed to interact with free cells in the presence or absence of lectin. It was found that Con A agglutinated two lymphoma cells and, less frequently, one lymphoma cell and one lymphocyte. Agglutination of two lymphocytes was only rarely observed. Perhaps the most interesting observation was that, while agglutination was inhibited by rixation of both the immobilized and the free cells, fixation of only one of the cells actually enhanced agglutination. This clearly indicates that receptor mobility is necessary in only one of the t.wo cells involved in a cross-link. It is possible therefore that an important factor in the formation of stable lectin cross-bridges is the ability of the receptors on at least one of the cells to come into precise alignment with the receptors on the other cell. Thus only short-range lateral mobility in the unfixed cells is necessary to stabilize agglutination. 3. Effect of Temperature and Membrane Fluidity In contrast to agglutination by either WGA or soybean lectin, which is independent of temperature (Inbar et al., 1973b; Gordon and Marquardt, 1974; c. Huet, 1974; M. Huet, 1975; Horwitz et al., 1974), Con-A-mediated agglutination is inhibited by cooling cells to 4°C. Con A binds to transformed cells at 4°C in sufficient amounts to cause agglutination if the cells are subsequently warmed to 37°C (Inbar et al., 1971, 1973b; Noonan and Burger, 1973a,b; Gordon and Marquardt, 1974; c. Huet, 1974; M. Huet, 1975; Horwitz et al., 1974),but no agglutination takes place at 4°C. Since agglutination involves the interaction of a multivalent lectin with receptors on the surfaces of two cells, any change in agglutinability on changing the temperature could result from an alteration in either the lectin or the surface membrane. The available evidence suggests that both can occur in the case of Con-A-mediated agglutination. For example, cooling from 37" to 4°C results in the conversion of Con A from a tetramer to a dimer, but this temperature change has no effect on the valence of PHA, WGA,

314

JAY C. BROWN AND RICHARD C. HUNT

or soybean lectin (Gordon and Marquardt, 1974; C. Huet, 1974; M. Huet, 1975). Since reduction of the valence of Con A by succinylation or chymotrypsin digestion (Burger and Noonan, 1970; Gunther et al., 1973) has been shown to result in decreased agglutination (Trowbridge and Hilborn, 1974), this may explain the temperature-dependence phenomenon. Cooling also results in changes in the cell membrane. Current models of membrane structure envisage a lipid bilayer containing globular proteins dispersed throughout the lipid (Singer and Nicolson, 1972; Nicolson, 1976a).At physiological temperatures, the lipids are relatively fluid and proteins are free to move laterally in the plane of the membrane. However, on cooling the lipids become less mobile, and they eventually freeze into a paracrystalline state below their transition temperature. Electron and fluorescence microscopy of Con-A-Con-A receptor complexes on the surface of many transformed cell types has demonstrated that at 4°C the receptors behave similarly to fixed cells at 37°C; that is, they remain randomly distributed over the cell surface. On warming, however, the receptors are able to migrate in the presence of Con A to form clusters of Con-A-Con-A receptor complexes (Inbar et al., 1973a; Nicolson, 197313; Rosenblith et al., 1973; Ukena et al., 1974; Berlin, 1975). This temperature-dependent increase in receptor mobility most probably results from a change in the fluidity of the lipid continuum. Support for this idea is derived from studies on the effect of altering the membrane fatty acid composition. Con-A-mediated agglutination of LM- or SV-40-transformed 3T3 cells shows a temperature-dependent transition at about 14"- 18°C; above this temperature the cells are readily agglutinated, whereas below 14"- 18°C they are not. When cells are cultured in medium supplemented with unsaturated fatty acids, there is a change in the composition of the membrane such that membrane phospholipids come to have a higher content of unsaturated fatty acyl chains. This results in a lowering of the transition temperature for the lipid bilayer itself and also for Con-A-mediated agglutination. An increase in both transition temperatures is observed when cells are grown in medium supplemented with saturated fatty acids (Horwitz et al., 1974; Rittenhouse et al., 1974). Thus changes in the fluidity of membrane lipids with temperature may account for the difference in Con-A-mediated agglutinability at 0°C compared to 37°C. It is very unlikely, however, that a difference in bilayer fluidity could account for the difference in Con-A-mediated agglutinability between normal and transformed fibroblasts. The lipid composition of normal and transformed cells is quite similar (Quigley et al., 1971,1972; Gaffney et al., 1974; Micklem

LECTINS

315

et al., 1976), and physical methods for measuring membrane fluidity have either failed to reveal a significant difference between normal and transformed cells (Robbins et al., 1974; Gaffney, 1975) or have shown greater fluidity in normal cell membranes (Fuchs et al., 1975; Edwards et al., 1976; Yau et al., 1976; but compare Barnett e t al., 1974). Furthermore, the mechanism for selective agglutination of transformed cells by lectins other than Con A must be even less dependent on membrane fluidity, since in these cases agglutination is relatively independent of temperature.

4. Role of Cytoskeletal Elements Neoplastic transformation affects both major intracellular cytoskeletal systems, the microtubules and the microfilaments. After transformation, cell surface-associated tubulin (Brinkley et al., 1975; Fine and Taylor, 1976) and actin (Wickus et al., 1975; Pollack and Rifkin, 1975; Pollack et al., 1975) are dramatically reduced in amount. Both fluorescence and electron microscopy (McNutt et al., 1971, 1973; Nicolson, 1975) show this reduction to be associated with less well-defined intracellular microtubule and microfilament structures. If these cytoskeletal systems are involved in lectin agglutinability, agents that disrupt microtubules or microfilaments should cause an increase in the agglutinability of untransformed cells. In fact, there is considerable evidence to support this expectation. Incubation of 3T3 cells with colchicine (which disrupts microtubule structure) and cytochalasin B (which affects microfilaments) together results in the clustering of Con A receptors and in enhancement of agglutinability. Agents such as colchicine and calcium ionophores, which disrupt only microtubules, have little effect on either parameter (Yin et al., 1972; Ukena et al., 1974; Poste and Nicolson, 1976). Trypsin treatment, which converts untransformed cells from the unagglutinable to the agglutinable state, is also found to disrupt the intracellular microfilament system severely in rat embryo cells (Pollack and Rifkin, 1975). Local anesthetics such as dibucaine, tetracaine, and procaine, which affect cell membrane fluidity, act on 3T3 and BHK cells in a manner similar to colchicine and cytochalasin B combined; that is, they lead to a redistribution of lectin receptors into clusters and to an increase in Con-Amediated agglutinability (Poste et al., 1975a,b,c).Although local anesthetics increase the fluidity of membrane lipids, and therefore probably allow receptors to become more mobile (see, e.g., Papahajopoulos et al., 1975), the increase in fluidity is small, and it has been proposed that the anesthetics may increase cell agglutinability by disrupting intracellular cytoskeletal elements rather than b y affecting membrane

316

JAY C. BROWN AND RICHARD C. HUNT

lipids. This site of action is supported by electron microscope studies showing that both intracellular microtubules and microfilaments are reduced in number in normal cells following incubation with local anesthetics (Nicolson et al., 1976). Compared to the situation with normal cells, the effects of microtubule- and microfilament-disrupting agents on transformed cells are much more difficult to interpret. Indeed, it would be expected that, as transformed cells appear to have a disorganized cytoskeletal system, these agents would have no further effect on agglutinability. In some cases this is so. Colchicine and cytochalasin B together have no effect on the Con-A-mediated agglutinability of SV-40-transformed 3T3 cells (Poste et al., 1975b), and colchicine alone has no effect on the agglutinability of CHO (Van Veen et al., 1976) or rat ascites tumor cells (Kaneko et al., 1973). Cytochalasin B has no effect on the agglutinability of CHO cells. However, cytochalasin B does have an inhibitory effect on the Con-A-mediated agglutinability of rat ascites tumor cells (Kaneko et al., 1973), and agglutination of several transformed cell types by Con A is altered by microtubule-disrupting agents. For example, incubation of SV-40-transformed 3T3 cells with colchicine or with calcium ionophores leads to the inhibition of Con-A-mediated agglutination (Yin et al., 1972; Poste et al., 1975b; Poste and Nicolson, 1976). Similarly, the agglutination of rat hepatoma cells is reduced by microtubule disruption (Nakamura and Terayama, 1975). LM cells treated with colchicine show either enhanced or decreased Con-A agglutinability, depending on the length of time they are exposed to the lectin (Rittenhouse et al., 1976). Morphological observations on SV-40-transformed 3T3 cells confirm that colchicine is able to alter the distribution of Con-A receptors on the surface of these cells. A clustered distribution of receptors is induced by Con A alone, and these clusters can be redistributed into a cap when cells are also incubated with colchicine. Presumably this capping is under direct control of microfilaments, because cells incubated with cytochalasin B plus colchicine are not capped (Ukena et al., 1974). Local anesthetics have no effect on the Con-A-mediated agglutination of SV-40-transformed 3T3 cells or on the distribution of Con-A receptors in these cells. Receptors remain clustered. in the presence of Con A plus anesthetics. Local anesthetics, however, reverse the inhibition of agglutination produced by colchicine (Poste et al., 1975b,c). Together these observations provide reasonable evidence that cytoskeletal systems are in fact involved in the sensitivity of cells to lectin-mediated agglutination. The overall effect of cytoskeletal involve-

LECTINS

317

merit is to prevent or inhibit agglutination. Thus one observes agglutination only in cells whose cytoskeletal systems have been disrupted either as a result of neoplastic transformation or in the presence of drugs. In fact, it may be that the effect of microtubules and microfilaments will prove to be central to our understanding of why transformed but not normal cells are agglutinable. Since both normal and transformed cells possess abundant lectin receptor sites, it is more difficult to understand why normal cells are resistant to agglutination than why transformed cells are agglutinable. Therefore it may be that, if one could determine how cytoskeletal structures are involved in the inhibition of normal cell agglutination, one would have the key to the whole question of selective transformed cell agglutinability. Microtubules and microfilaments are most likely to exert their effects on agglutination by controlling the distribution of lectin receptor molecules on the cell surface. It is reasonable to assume that this control can be exerted b y a direct interaction inside the cell between cytoskeletal structures and “transmembrane” lectin receptor glycoproteins which are exposed both outside and inside the cell. The association between glycophorin and spectrin inside the human erythrocyte provides a model for how this type of interaction may occur (Nicolson and Painter, 1973), and there is evidence in lymphocytes for an association between cell surface structures and intracellular microtubules (Edelman, 1974). The results of experiments in which transformed cells are exposed to microtubule- and microfilament-disrupting drugs strongly suggest that, in spite of the fact that the cytoskeletal systems of transformed cells are quite abnormal, some degree of cytoskeletal control over cell surface properties still exists.

5. Microvilli An idea related to the involvement of cytoskeletal elements suggests that microvilli may be the key to selective agglutination of transformed cells. If transformed cells had more microvilli than untransformed cells, and if agglutination by lectins required cross-bridges in the regions of interdigitating microvilli, or if lectin receptors were concentrated in the microvilli, it would be reasonable to expect that transformed cells would be more agglutinable than normal ones (Boyde et al., 1972; Porter et al., 1973a; Willingham and Pastan, 1975; Borek and Fenoglio, 1976; Malick and Langenbach, 1976). Some experimental studies have been interpreted to support this view. For example, the incubation of mouse L-929 cells (which have numerous surface microvilli) with CAMPcaused a reduction both in the number of microvilli and in lectin-mediated agglutinability (Willingham and

318

JAY C. BROWN AND RICHARD C. HUNT

Pastan, 1975). In 3T3 cells trypsin treatment, which enhances agglutinability, was also found to increase the number of microvilli (Follett and Goldman, 1970; Willingham and Pastan, 1975). Furthermore, some of the effects of microtubule- and microfilament-disrupting agents on agglutinability may result from the loss of microvilli caused by the disruption of cytoskeletal elements (Loor, 1973; Loor and Hagg, 1975). However, there are many significant exceptions to the proposed correlation between increased number of microvilli and enhanced agglutinability in transformed cells. Temmink and his associates showed that normal 3T3 cells and normal lymphocytes have a greater number of microvilli than their more agglutinable transformed counterparts (Collard and Temmink, 1975, 1976; Temmink et al., 1976). Similarly, in a study of BALB/c 3T3 cells and several transformed lines derived from them, no correlation was observed between microvillus formation and tumorigenicity. Lines transformed by SV-40, murine sarcoma virus, or polyoma virus were largely free of microvilli, while a “spontaneous” transformant possessed numerous microvilli (Porter et al., 1973b). It seems most unlikely therefore that an increased number of microvilli in transformed cells is the only explanation for their increased lectin agglutinability. 6. Conclusion It is fair to say at the present time that, despite a great deal of effort, we really do not know why transformed cells are generally more agglutinable by lectins than their normal counterparts. We do know of some explanations that do not apply. One cannot account for the difference between normal and transformed fibroblast agglutinability with Con A or with WGA by assuming that transformed cells have a greater total number of receptors for these lectins. Experimental studies have shown that the total number of binding sites for Con A or for WGA does not differ significantly in normal compared to transformed cells (see Section IV,D,l). Similarly, one cannot account for the difference in agglutinability by assuming that lectin receptors are distributed differently on the surfaces of normal and transformed cells; a uniform distribution of Con-A receptors is found on the unperturbed surface of both normal and transformed fibroblasts. Although a small amount of lateral receptor mobility in the plane of the membrane may be required for Con-A agglutination, it is now clear that the migration of receptors to form large clusters or patches also cannot account cleanly for the selective agglutination of transformed cells. Some nonagglutinable cells are able to form receptor patches, while some agglutinable cells are not (Ukena et al., 1976).

LECTINS

319

Elimination of these earlier hypotheses has not entirely depleted our supply of good ideas to explain the selective agglutination of transformed cells. There are several proposals that have not yet been fully tested experimentally. For example, it may be that untransformed fibroblasts are not agglutinated by lectins because of their high surface negative charge (zeta potential); electrostatic repulsion between cells may prevent them from being cross-linked noncovalently by lectins. According to this hypothesis, transformed cells, which usually have a much lower density of surface negative charge, would be agglutinated because of the reduced electrostatic charge repulsion between cells. conversion of normal cells from the unagglutinable to the agglutinable state by trypsin treatment would be explained by the fact that proteolysis simply reduces the net cell surface negative charge. Ukena and Karnovsky (1976) recently proposed that selective agglutination of transformed cells occurs because transformed fibroblasts spontaneously, in the absence of lectin, undergo an aggregation process that is not found in untransformed cells or that takes place much more slowly in untransformed cells. According to this “spontaneous adhesion” hypothesis, the role of the lectin would simply be to accelerate a preexisting transformed cell property. This proposal clearly deserves further study. The same can be said of proposals that emphasize the role of cytoskeletal elements and microvilli in selective agglutination. Studies with drugs that disrupt microtubules and microfilaments have provided a strong case for the involvement of these structures in selective lectin-mediated agglutination (see Section IV,D,4), and it would be very surprising if microvilli, by which fibroblasts ordinarily make initial contact with each other, were not also somehow involved in the agglutination process. Despite the unsatisfactory state of our knowledge at the present time, the motivation for trying to understand the molecular basis of the selective agglutination of transformed cells remains as strong as ever. It may really be that a satisfactory explanation of this phenomenon will provide important new information about the role of the cell surface in transformed cell growth. The time may have come in the development of this field, however, when it would be more fruitful to pursue basic studies on membrane structure and function than to continue to attack the issue of selective agglutination directly.

V. Interaction of Lectins with Cells Infected by Nononcogenic Viruses Although neoplastic transformation of normal cells is by far the best studied situation in which changes in lectin agglutinability accom-

320

JAY C. BROWN AND RICHARD C. HUNT

pany biological variables, such changes are not restricted to transformation. Animal cells may also undergo changes in sensitivity to lectin agglutination during the productive growth of certain nononcogenic viruses. The growth process ordinarily involves intracellular replication of virus particles and their release either with or without lysis of the host cell. In the case of infection by viruses containing a membrane envelope, virus replication may involve insertion of virusspecific proteins into the host cell plasma membrane, followed by budding of the virus particle from the cell at this region of modified membrane. Lectin-detectable changes in the 'cell surface have been observed accompanying productive infection with a wide variety of both RNA- and DNA-containing animaI viruses, as shown in Table VIII. In their studies of myxoviruses, Becht et al. (1972) consistently observed an increase in the Con-A agglutinability of cells infected with these viruses. These investigators concluded that it was insertion of TABLE VIII VIRUSES CAPABLE OF INDUCING INCREASED LECTIN AGGLUTINABILITY I N HOST CELLS DURING PRODUCTIVE INFECTION

Virus class RNA-containing viruses Orthomyxoviruses Paramyxoviruses

Arboviruses Rhabdoviruses DNA-containing viruses Pox viruses

virus

Influenza virus Fowl plague virus NDV

Sendai virus, canine distemper virus, parainfluenza a, mumps virus, measles virus, respiratoq syncitial virus Simian virus 5 Sindbis virus Vesicular stomatitis virus

References

Becht et al. (1972) Fresen and Illiger (1974) Becht et al. (1972); Poste and Reeve (1972); Bubel and Blackman (1975) Poste (1975)

Becht et al. (1972) Birdwell and Strauss (1973) Becht et al. (1972)

Herpes viruses

Herpesvirus hominis type 1

Zarling and Tevethia (1971); Bubel and Blackman (1975) Tevethia et al. (1972); Poste

Adenoviruses

Adenovirus type 2

Salzberg and Raskas (1972)

Vaccinia

(1972)

LECTINS

32 1

virus-specific glycoproteins into the cell membrane that gave rise to enhanced Con-A-mediated agglutinability. They also noted that Con-A precipitated purified fowl plague and SV-5 viruses. Other investigators have suggested that the insertion of viral components into the cell membrane may not be necessary for the surface change that results in increased agglutinability. For example, Zarling and Tevethia (1971)reported that vaccinia virus causes rabbit kidney cells to become agglutinable with Con A early (2 hours after infection) in its replicative cycle, although no virus particles are detected until much later (6 hours after infection). Herpesvirus hominis type- 1 infection of chick embryo fibroblasts also leads to an early expression of Con-A agglutinability; in this case agglutinability is observed 4 hours before the expression of a virus-specific cell surface antigen (Tevethia et al., 1972). De no00 protein synthesis is required after virus infection before this increased agglutinability can be detected. The agglutinability of cells infected by Newcastle disease virus (NDV) has been more extensively studied than any other nononcogenic virus system (Poste, 1975; Reeve et al., 1975). A useful feature of NDV has been the availability of virus strains with varying degrees of virulence. The infection of cells by virulent strains causes an increase in Con-A- or WGA-mediated agglutinability. At 4-8 hours after infection, cell agglutinability rises and, at the same time, the thickness of the cell coat decreases. The decreased thickness of the cell coat results from shedding of surface material into the medium and is accompanied by the release of lysosomal enzymes. In contrast, after infection by an avirulent strain, no change in coat thickness or agglutinability is observed (Poste and Reeve, 1972; Poste et al., 1972; Reeve et al., 1972, 1975). Although the increases in agglutinability observed after virulent virus infection could be due simply to the insertion of virus-specific glycoproteins into the plasma membrane, this is not thought to be the case; no difference can be detected in the amount of labeled Con A bound to the cell surface before or after infection with virulent or with avirulent strains of NDV (Poste and Reeve, 1974). There is also no difference in the binding at 0" and 37°C (compare Noonan and Burger, 1973a). BHK cells labeled with fluorescent Con A after infection with a virulent strain of NDV reveal another similarity to transformed cells. The surface of these cells shows patches of fluorescence at 37"C, while uninfected cells or cells infected with an avirulent strain of the virus show uniform fluorescence. Glutaraldehyde fixation or cooling to 0°C inhibits patch formation (Poste and Reeve, 1974). The increase in Con-A agglutinability induced by infection with virulent NDV is

322

JAY C. BROWN AND RICHARD C. HUNT

therefore strikingly similar to the increase observed after neoplastic transformation.

VI. Interaction of Lectins with Developing Cells A. EGGSAND EMBRYOS Eggs and embryonic cells of several organisms resemble erythrocytes and transformed cells in that they are strongly agglutinated by lectins. Fertilized eggs are frequently more agglutinable than unfertilized eggs but, as the embryo develops, its cells generally become less agglutinable. The molecular mechanism by which agglutinability is gained and lost during development has yet to be elucidated, but it is reasonable to suppose that these changes may resemble those that take place after neoplastic transformation of normal cells. In fact, transformed and embryonic cells have many properties in common, including a high rate of growth, the ability to migrate in uiuo, and the ability to invade surrounding tissues. Whether or not these similarities have a common basis at the molecular level remains to be determined. Con A does not agglutinate unfertilized eggs of the sea urchin Paracentrotus liuidus, but it does inhibit the fertilization process. It also blocks formation of the fertilization membrane and cleavage (Lallier, 1972).One day after fertilization, dissociated sea urchin embryo cells are strongly agglutinated by Con A and RCA, but the effect of the lectins diminishes during subsequent development (Oppenheimer and Odencrantz, 1972; Krach et al., 1974). A most interesting observation is that the cells of sea urchin embryos at a later stage of development (the 32- to 64-cell stage) differ among themselves in their relative agglutinability by Con A. Cells that show migratory activity are both more agglutinable and have a greater degree of Con-A-induced receptor clustering than nonmigratory cells (Roberson and Oppenheimer, 1975; Neri et al., 1975).At no developmental stage does WGA agglutinate sea urchin cells that have not first been treated with trypsin. Studies on the agglutination of chick embryonic cells have shown that receptors for different lectins behave quite differently during development. At very early stages after fertilization (e.g., in eggs immediately after fertilization or in 22-hour embryos), Con A and WGA both agglutinate chick embryo cells. WGA agglutination is lost quickly, however, and by 8 days of incubation chick embryonic liver or neural retina cells are agglutinated only by Con A or RCA. WGA agglutination can be restored by trypsin digestion of the cell surface [Moscona,

LECTINS

323

1971; Kleinschuster and Moscona, 1972; Zalik and Cook, 1976; but compare McDonough and Lilien (1975) who were unable to demonstrate agglutination of early chick embryo cells by Con A]. At later stages of embryonic development Con-A-mediated agglutination progressively decreases, until by day 20 it is not present in mechanically dissociated chick cells. Cells remain fully agglutinable by RCA up to at least 20 days, and trypsinization restores full agglutinability by WGA and Con A at all stages up to this time (Kleinschuster and Moscona, 1972). The binding of 1251-labeledCon A to chick embryo cells shows that, contrary to what would be expected, decreasing agglutinability with age is accompanied by increased lectin binding. At all stages, lectin binding to mechanically dissociated cells is greater than to cells dissociated with trypsin, even though trypsin increases agglutinability. The lack of correlation between Con-A binding and agglutinability has been interpreted to imply that a difference in receptor mobility with age may be the basis for the decrease in agglutinability. In support of this proposal is the observation that, although binding of C O ~ - A - ' ~to~ Itrypsinized 19-day-old chick cells is not affected by lowering the temperature to 4"C, agglutination is completely abolished. By analogy with observations on transformed and normal cells it might be expected that glutaraldehyde fixation of chick embryo cells would inhibit their agglutination by Con A. Instead, fixation promotes agglutination not only of mechanically dissociated cells at 37°C but also of trypsin-dissociated cells at 4°C. This puzzling result is suggested to be the consequence of receptor cross-linking by the fixative (Martinozzi and Moscona, 1975), but further research will be necessary to determine whether this is in fact the case. In addition to the changes in Con-A agglutinability and in Con-A binding observed during the development of chick embryonic cells, Con A is also found to affect DNA synthesis differently as development proceeds (Kaplowitz and Moscona, 1973; Roguet and Bourrillon, 1975a,b). As in the case of chick cells, rodent embryo cells also exhibit changes in lectin agglutinability during development. For example, unfertilized mouse eggs lacking a zona pellucida (a glycoprotein layer around the egg plasma membrane which the sperm penetrates before fusion with the egg) are not agglutinated by Con A unless they have been pretreated with protease. However, after fertilization the eggs are strongly agglutinated (Pienkowski, 1974).The difference in agglutinability in the fertilized and unfertilized eggs may result from a difference in receptor distribution. With the use of fluorescent Con A, it

324

JAY C. BROWN AND RICHARD C. HUNT

has been demonstrated that the lectin binds to both fertilized and unfertilized mouse eggs, but that the pattern of fluorescence is different; whereas unfertilized eggs bound the lectin over only part of their surface, fertilized eggs bound lectin uniformly over their whole surface (Johnson et al., 1975).As in chick embryos, agglutination is reduced as the embryo develops, so that by the blastocyst stage high Con-A concentrations fail to agglutinate mouse embryo cells (Rowinski et al., 1976). In hamster eggs, the change in lectin agglutinability after fertilization is not as pronounced as in the mouse. Fertilized zona pellucidafree eggs are only slightly more agglutinated by Con A, lentil (Lens culinaris) lectin, and WGA than unfertilized eggs. RCA and D . biflorus agglutinin (DBA) agglutinate the zona pellucida of both fertilized and unfertilized eggs, and binding of fluorescent lectins to the zona pellucida does not change during maturation. In zona-free eggs the binding of fluorescent Con A, RCA, and WGA decreases as the egg develops (Yanagimachi and Nicolson, 1976). Pretreatment of hamster eggs with WGA, RCA, or DBA prevents fertilization and reduces the sensitivity of the zona pellucida to digestion by trypsin. These lectins cause a change in the zona pellucida which results in a greater degree of light scattering when the zona is viewed by dark-field illumination. Spermatozoa do not either penetrate the WGA-treated zona or even bind to its surface. Con A, however, does not block fertilization, nor does it cause the change in light scattering (Oikawa et al., 1973,1974). It has been speculated that lectins that inhibit fertilization do so by cross-linking their receptors and thereby block digestion of the zona pellucida by sperm-borne enzymes; according to this model, they inhibit fertilization by mimicking the action of a natural lectin contained within the egg and released after fertilization. This lectin reacts with zona pellucida glycoprotein to prevent multiple fertilization (Wyrick et al., 1974). Electron microscope analysis of hamster eggs after incubation with ferritin-labeled lectins has shown that RCA, Con-A, and WGA receptors are localized in the zona pellucida and in the underlying plasma membrane. RCA and WGA receptors are distributed asymmetrically throughout the zona pellucida, with the highest concentrations at the surface. Con-A receptors are located sparsely throughout this layer (Nicolson et al., 197%). The plasma membrane receptors for the three lectins are randomly distributed in fixed cells or in cells that have been incubated with the lectin at 4°C. However, Con-A and WGA receptors form clusters at 25°C (Nicolson et al., 197513).It is attractive to

LECTINS

325

speculate that the mobility of plasma membrane receptors may be important in sperm fusion but, as yet, there is no experimental evidence to support this possibility. Lectin interactions with the cell surface have been examined in several other developmental systems including human fetal intestine (Weiser, 1972), cultured muscle (Betschart and Burger, 1975), the slime mold Dictyostelium discoideum (Weeks and Weeks, 1975), spinal ganglions of chick embryos (Treska-Ciesielski et al., 1971), ascidian eggs (Monroy et al., 1973), and neural crest cells from frog (Johnson and Smith, 1976) and from the urodele Ambystoma mexicanum (Moran, 1974). B. MALE GERMINALCELLS The early germinal cells of the seminiferous tubule (the spermatogonia) in adult animals undergo waves of proliferation, giving rise to cells which differentiate into spermatozoa. This process of proliferation and differentiation takes place in a highly coordinated, synchronous manner. The spermatogonia, which adhere tightly to one another, give rise first to spermatocytes; these differentiate into spermatids and finally spermatozoa. During differentiation the germinal cells lose their mutual adhesion, and spermatozoa are secreted into the lumen of the tubule as a suspension of individual motile cells. Since glycoproteins have been implicated in intercellular adhesion and in information transfer during differentiation, it is not surprising that they have been the subject of experimental studies in the field of spermatogenesis. Lectins have been shown to agglutinate the sperm cells of several species, including clam (Venus mercenaria and Mytilus mytilus; see Bade1 and Brilliantine, 1969), rodent (Edelman and Millette, 1971; Nicolson and Yanagimachi, 1972, 1974; Nicolson et al., 1972; Gordon et al., 1974), bull (Kashiwabara et al., 1965), and human (Uhlenbruck and Hermann, 1972). Separation of the head and tail regions of rodent sperm revealed that Con-A receptors were predominantly located in the head region, and fluorescence microscope studies showed binding to the acrosome to be most visible (Edelman and Millette, 1971). Electron microscopy of sperm incubated with ferritin-RCA revealed clusters of receptors in the postacrosomal region only. At O'C, or after glutaraldehyde fixation, no clusters were observed (Nicolson and Yanagimachi, 1974). From these data, it is possible to conclude that lectin receptors cannot move freely over the entire sperm cell surface, even though the cell is surrounded by a continuous plasma membrane. It

326

JAY C. BROWN AND RICHARD C. HUNT

may be that the more mobile receptors in the region behind the acrosome are involved in the fusion of the sperm with the egg plasma membrane, as suggested by Nicolson and Yanagimachi (1974). Changes that take place during differentiation of spermatozoa are more difficult to study, as they necessitate the isolation of different cell types in the germinal cell differentiation pathway. The rat testis has been employed as a source of cell suspensions highly enriched in secondary spermatocytes, spermatids, spermatozoa, and nongerminal interstitial cells. A major lectin-detectable difference exists between the cells of the germinal differentiation pathway and the nongerminal cells. Spermatocytes, spermatids, and spermatozoa are strongly agglutinated by either Con A or WGA, but the interstitial cells are agglutinated by neither lectin. Trypsinization of the interstitial cells increases their agglutinability by WGA but not by Con A. Nevertheless, A to the interstitial there is no difference in the binding of 1251-C~n cells and to the spermatocytes (Hunt et al., 1977).

VII. Biochemistry of Cell Surface Lectin Receptors The cell-agglutinating and mitogenic effects of lectins have stimulated a considerable amount of research into the nature of the cell surface structures recognized by lectins. Although both glycolipids (Surolia et al., 1975)and glycoproteins may serve as lectin receptors, most studies have focused on the glycoprotein components only. These studies usually begin with purified membrane fractions or with cells whose surface structures have been radioactively labeled by one of the many techniques, such as the lactoperoxidase method, now available for this purpose. Cells or membrane fractions are then solubilized and subjected to affinity chromatography on lectin-Sepharose columns; cell lectin receptors are assumed to be among those glycoproteins that are bound to the column and eluted with the appropriate haptenic saccharide. Receptor glycoproteins prepared in this way are ordinarily characterized by their mobility on sodium dodecyl sulfatepolyacrylamide gels. Further studies are ordinarily required to prove that the lectin-binding glycoproteins isolated in this way actually serve as lectin receptors in vivo on the cell surface as well as in solubilized extracts. For example, the observation that a lectin can protect its receptor from digestion when cells are treated with proteolytic enzymes has been taken as evidence for a receptor function in vivo (Brown, 1973; Taylor et al., 1974). Since lectins bind to the carbohydrate portions of their cell surface receptors, it is clear that receptors for the same lectin on different cell types need not have the same bio-

327

LECTINS

logical function. The same or similar carbohydrate groups may exist on different protein species. Furthermore, the same glycoprotein species may contain receptor sites for more than one lectin. The experimental strategy outlined above has led to the isolation and purification of lectin receptors from several different cell types. Human erythrocytes, for instance, are found to have two cell surface glycoprotein components. One of these, the major glycoprotein or glycophorin, has receptor sites for WGA, PHA, and U . europeus lectin (Marchesi et al., 1972). Recent structural studies on the glycophorin molecule (Winzler, 1969; Tomita and Marchesi, 1975) have rationalized these receptor activities. Glycophorin (MW 31,000) has been shown to be composed of a polypeptide backbone containing 131 amino acid residues, as shown in Fig. 4; a portion of the polypeptide backbone spans the erythrocyte plasma membrane. Sixteen carbohy-

( (

20

t

50

60

70

80

90

too

110

I20

I30

FIG.4. Amino acid sequence of the major human erythrocyte glycoprotein (glycophorin). The sites of attachment for 0-glycosidically linked (squares) and N-glycosidically linked (hexagon) carbohydrates are indicated (Tomita and Marchesi, 1975).

328

JAY C. BROWN AND RICHARD C. HUNT

drate side chains of two different types are attached to the N-terminal portion of the polypeptide and extend outward from the external surface of the cell. Of the 16 side chains 15 are identical; they are O-glycosidically linked to serine or threonine residues, and they have the structure shown in Fig. 5a (Thomas and Winzler, 1969). These side chains are thought to have the receptor sites for PHA (N-acetylgalactosamine, galNAc). The remaining side chain is N-glycosidically linked to asparagine and has the structure shown in Fig. 5b (Kornfeld and Komfeld, 1970). It is thought to contain the receptor sites for WGA [(glcNAc),] and for U . europeus lectin [(glcNAc),]. Since this structure has the carbohydrate groups recognized by Con A (a-linked mannose) and RCA (galactose),it is surprising that glycophorin does not serve as a receptor for these lectins as well (Marchesi et al., 1972; Fukuda and Osawa, 1973). However, Fukuda and Osawa (1973) showed that Con-A and RCA receptor sites can be exposed after removal of the 0-glycosidically linked carbohydrate groups from the glycophorin molecule by alkaline sodium borohydride treatment. This suggests that a portion of the N-glycosidically linked carbohydrate group may be masked on the surface of the glycophorin molecule. The minor human erythrocyte cell surface glycoprotein, called component I11 or component a, has been extensively purified by affinity chromatography on a column of Con A-Sepharose (Findlay, 1974). It has a molecular weight of approximately 100,000, it spans the erythrocyte membrane (Bretscher, 1973), and it has recently been shown to be a major anion channel through the erythrocyte plasma membrane (Cabantchik and Rothstein, 1974a,b; Ho and Guidotti, 1975). Affinity chromatography on lectin-Sepharose columns has also been employed to purify receptors from several other cell types, including human platelets, mouse L cells, D.discoideum, rat brain, and Torpedo (a)

$1,3

0-glycosidic Serine or

or 4

1;1,3

or 4

p-glycosidic Asparagine

Threonine

FIG.5. Structures of the 0-glycosidically linked (a) and N-glycosidically linked (b) carbohydrate groups found in the major human erythrocyte glycoprotein (glycophorin). Adapted from Thomas and Winzler (1969) and from Kornfeld and Kornfeld (1970).

329

LECTINS

californicus electric organ. A summary of the properties of these receptors is given in Table IX. The rat brain lentil lectin receptor and the Torpedo Con-A receptor have been shown to possess the major cell surface acetylcholinesterase activities of these tissues. Similar studies have shown that bovine rhodopsin can serve as a Con-A receptor in isolated rod outer segments (Steinemann and Stryer, 1973) and that the insulin receptor of rat fat cells can also serve as a receptor for WGA and for Con A (Cuatrecasas, 1973b; Cuatrecasas and Tell, 1973). Other studies have resulted in the identification of one major (MW 60,000) and three minor lentil lectin receptors from human KB cells (Butters and Hughes, 1975),lentil lectin and WGA receptors from rat brain synaptic plasma membranes (Gurd and Mahler, 1974), and Con A, WGA, and lentil lectin receptors from L1210 cells (Hourani et al., 1973; Jansons and Burger, 1973). The isolation of lymphocyte cell surface lectin receptors has been undertaken in an attempt to understand the role of these structures in lectin-induced mitogenesis. So far these studies have produced a complicated set of results. All workers have reported multiple lymphocyte receptors for the lectins they have studied. For example, Henkart and Fisher (1975) identified at least five different glycoprotein receptors for Con A on the surface of human peripheral blood lymphocytes. Three of these have molecular weights of 68,000, 53,000, and 43,000, respectively. Mouse B lymphocytes have been shown to have at least TABLE IX PURIFIEDCELL SURFACELECTIN RECEPTORS _

Cell type Human erythrocyte Human erythrocyte

Human platelet Mouse L cells Dictyostelium discoideum Torpedo californicus electric organ Rat brain Rat thymocytes

Receptor for WGA and PHA Con A

_

_

~ ~

Molecular weight 31,000 100,000

Comment Glycophorin Component 111 Anion channel

Lentil lectin

80.000

LIS extraction LIS extraction Aggregation function? Acetylcholinesterase

Con A Lentil lectin

80,000 25,000

Acetylcholinesterase Thy-1 antigen

Con A Con A Con A

100,OOO 100,000 Unknown

Reference Winzler (1969); Tomita and Marchesi (1975) Findlay (1974) Cabantchik and Rothstein (1974a.b); Ho and Guidotti (1975) Nachman et al. (1973) Hunt et al. (1975) Wilhelms et al. (1974) Taylor et al. (1974)

Wenthold et al. (1974) Letarte-Muirhead et al.

(1975)

330

JAY C. BROWN AND RICHARD C. HUNT

six different Con-A receptors (Hunt and Marchalonis, 1974), and pig lymphocytes have multiple receptors for both Con A (Allan et al., 1972) and for lentil lectin (Hayman and Crumpton, 1972). LetarteMuirhead et al. (1975) isolated a lentil lectin receptor from rat thymocytes in purified form and showed it to be identical to the serologically defined thy-1 antigen. Choi and Jenson (1974) employed a clever method to identify a Con-A receptor (MW 160,000) on the surface of chick spleen lymphocytes and, recently, R. Emmons and D. C. Benjamin (personal communication) demonstrated that rabbit thymocytes have at least three receptors for Con A (MW 200,000, 70,000, and 40,000) and one for PHA (MW 20,000). None of these studies has as yet determined which, if any, of the known receptors is involved in lectin-induced mitogenesis. Clearly, this must be regarded as a priority for future studies of lymphocyte lectin receptors. A second experimental strategy employed for the isolation and characterization of cell surface lectin receptors has been to solubilize cell surface glycopeptide fragments by treating intact cells with proteolytic enzymes. Solubilized glycopeptides are then separated from each other chemically and assayed for receptor activity by measuring their ability to inhibit lectin-induced cell agglutination. In this way one isolates glycopeptide fragments derived from the parent molecule and not the intact receptor molecule itself. The method has been successfully employed for the isolation and structural characterization of N- and of O-glycosidically linked carbohydrate chains derived from the glycophorin molecule, as shown in Fig. 5 (Thomas and Winzler, 1969; Kornfeld et al., 1971). Recent studies have resulted in the isolation of Con-A- and WGA-binding glycopeptides from Novikoff ascites hepatoma cells (Neri et al., 1974) and from AS-SOD rat ascites hepatoma cells (Smith et al., 1973). Earlier studies of this type have been adequately discussed in previous review articles (Lis and Sharon, 1973b; Nicolson, 1974~).

VIII. Lectin Toxicity In addition to their mitogenic and agglutinating properties, some lectins are found to be extremely toxic to animal cells. It has long been known, for example, that the seeds of the castor bean plant, R . communis, are poisonous. Their toxicity results from a lectin called ricin, and there has been considerable debate concerning whether this toxin and RCAII,which is isolated from castor beans by affinity chromatography (Nicolson and Blaustein, 1972; Kornfeld et al., 1974), are actually the same molecular species. Both ricin and RCA,, bind to cell

LECTINS

331

surface carbohydrates and to Sepharose (Tomita et al., 197213; Kornfeld et al., 1974; but contrast Lugnier and Dirheimer, 1973),both have hemagglutinating activity (Kabat et al., 1947; Nicolson and Blaustein, 1972; Kornfeld et al., 1974; Olsnes et al., 1974a; but contrast Ishiguro et al., 1964), and the two have similar molecular weights (Ishiguro et al., 1964; Nicolson and Blaustein, 1972).Like ricin, RCAIlis extremely toxic, and the toxicity of both RCAII and ricin is blocked by lactose (Kornfeld et al., 1974). RCAII and ricin, however, behave differently on ion-exchange chromatography and in some polyacrylamide gel systems (Lugnier and Dirheimer, 1973). The resolution of these conflicting results appears to be in a small structural difference between ricin and RCAII. Both ricin and RCAIl consist of two polypeptides, an A chain (MW 30,000) and a B chain (MW 35,000), which are linked by a disulfide bond in the intact molecule (Olsnes and Pihl, 1972b; 1973a,b). Whereas the B chain of RCAII has been shown by immunological criteria to be similar to the B chain of ricin, the A chains are clearly different (Olsnes et al., 1974b; Pappenheimer et al., 1974). Ricin and a similar toxic protein called abrin, which can be isolated from the seeds of Abrus pecatorius (McPherson and Rich, 1973; Olsnes et al., 1974a), inhibit protein biosynthesis (Lin et al., 1971, 1972; Onozaki et al., 1972, 1975; Olsnes, 1972; Olsnes and Pihl, 1972a; Montanaro et al., 1973; Sperti et al., 1973; Grollman et al., 1974; Kornfeld et al., 1974; Refsnes et al., 1974). It was originally reported that agarose-coupled RCA had toxic effects similar to those of the free lectin, and this was interpreted as showing that the toxic effect was exerted without the lectin entering the cell (Onozaki et al., 1972). However, this interpretation proved to be incorrect, probably because the lectin was able to detach from agarose beads and enter the cell. In fact, strong evidence now exists showing that ricin and abrin must enter the cell to exert their toxic effects. For example, a lag is observed between the binding of the lectin and the onset of inhibition of protein synthesis (Refsnes et al., 1974; Kornfeld et al., 1974; Nicolson et al., 1975~); this lag is of the order of 30 minutes at 37°C in Ehrlich ascites cells, whereas no lag is observed when cell lysates are employed (Refsnes et al., 1974). Various observations strongly suggest a multistep process consisting of binding to the cell surface, entry into the cell, and inhibition of protein synthesis. Lactose or antisera against the toxin can stop the inhibitory effects only during the initial part of the lag period (Refsnes et al., 1974; Nicolson, 1974b). Protein synthesis is not inhibited in whole reticulocyte cells (which have little endocytotic activity) when they are incubated with

332

JAY C. BROWN AND RICHARD C. HUNT

the toxin, but protein synthesis is inhibited in reticulocyte lysates (Refsnes et al.,.1974; Nicolson et al., 1975c; Sandviget al., 1976). Furthermore, electron microscope studies have shown that the toxin enters cells by endocytosis (Nicolson, 1974b; Gonatas et al., 1975). For the inhibition of protein synthesis to occur in a reticulocyte lysate, the A and B polypeptides of the toxin must be separated (Olsnes et al., 1976). In whole cells the B chain is thought to be involved in binding to the cell surface, since it binds to carbohydrates but is not toxic alone. The A chain, however, does not interact with carbohydrates but is toxic alone in lysates (Olsnes and Pihl, 1973a,b; Olsnes et al., 1976). The inhibition of protein synthesis by ricin and abrin occurs be-

cause these toxins stop the completion of nascent peptide chains already initiated on the ribosome (Olsnes, 1972; Olsnes and Pihl, 1973a,b). In vitro protein synthesis studies have demonstrated that ricin inhibits poly-U-directed incorporation of phenylalanine into polypeptide in an irreversible fashion. Ricin functions by interacting with the ribosome in such a way as to prevent the action of elongation factor 2, the translocase, probably by preventing it from binding to the ribosome surface (Montanaro et al., 1973; Carrasco et al., 1975). Since a single ricin molecule can inactivate many ribosomes, it is likely that ricin acts catalytically. Hybrid ribosomes consisting of control and toxin-treated subunits have been employed to localize the site of ricin action to the 60s ribosomal subunit (Sperti et al., 1973; Onozaki et al., 1975) and, within the 60s subunit, the toxin acts on an 8s complex which can be released from the ribosome by EDTA (Benson et al., 1975).This 8s complex contains the GTPase and ATPase activities associated with elongation factors 1 and 2. The toxin affects the site to which elongation factor 2 binds (Benson et al., 1975; Olsnes et al., 1975; Fernandez-Puentes et al., 1976; Sperti et al., 1976). The toxin may have a hydrolytic effect on either the ribosomal proteins or the rRNA, but polyacrylamide gel electrophoresis has failed to show any gross change in either of these components (Olsnes et al., 1975; Lugnier et al., 1976). In addition to the well-documented action of ricin and abrin on the ribosome, it has been suggested that both toxins may act indirectly to inhibit protein synthesis, by raising intracellular levels of RNase, and thereby disaggregate polysomes (Lin et al., 1972; Grollman et al., 1974). Unfortunately, the mechanism(s) b y which other lectins exert their toxic effects has not been elucidated as clearly or as elegantly as that of ricin and abrin. A few results, however, are worth noting. PHA has been found to impair DNA and RNA synthesis, but how it does so is not known (Caso, 1968; Dent, 1971; Dent and Hillcoat, 1972). Also,

LECTINS

333

little is known of how Con A and WGA exert their toxic effects, except that WGA appears to be toxic to mouse L-929 cells only during the S phase of the cell cycle-that is, when cells are actively synthesizing DNA (P. Welch and J. Brown, unpublished observations). Not only are many lectins toxic, but this toxicity is selective. Transformed cells are frequently much more sensitive than normal cells (Shoham et al., 1970; Ozanne and Sambrook, 1971b; Ozanne, 1973). For example, while 3T3 cells were found to be resistant to the toxic effects of Con A at a concentration of 1 mg/ml over a period of 8 hours, 70% of the SV-40-transformed 3T3 cells were killed (Shoham et al., 1970). Thus it is not surprising that lectins have been employed in attempts to inhibit tumor growth in vivo. The injection of hamsters with Con A following injection with transformed cells caused a significant inhibition of tumor growth. This effect was abolished by simultaneous injection of a-MG (Shoham et al., 1970).Tumor inhibition by Con A has also been observed in other systems (Lin and Bruce, 1971; Gericke et al., 1971; Inbar et aZ., 197213; Friberg et al., 1972; Ralph and Nakoinz, 1973). Unfortunately, such high concentrations of Con A usually must be employed for tumor suppression that some control animals die from its toxic effects. Other lectins that have been employed in experiments on tumor cell growth suppression include Robinia pseudoacacia lectin (Aubery et al., 1972), PHA (Datta et al., 1969; Robinson and Mekori, 1972; Ralph and Nakoinz, 1973),RCA (Nicolson and Blaustein, 1972), and ricin and abrin (Lin et al., 1970). None of these experiments has led to any therapeutic use of lectins.

IX. The Biological Role of Lectins A fascinating aspect of lectin studies, and one that has received comparatively little attention, is the biological functions lectins may perform in the organisms from which they are isolated. What does Con A do for the jack bean? A lack of experimental results in this field has not inhibited speculation, and many possible functions have been suggested. These include a protective effect, because lectins have a superficial similarity to antibodies, an involvement in the selective binding of rhizobia during the initiation of root nodules in legumes, and facilitation of sugar transport within the plant or, alternatively, immobilization of sugars within a particular part of the plant (Boyd et al., 1958; Boyd, 1963). Only the first two of these suggestions are supported by existing experimental evidence. Although lectins have some properties in common with antibodies, their comparatively narrow range of specificities

334

JAY C. BROWN AND RICHARD C. HUNT

makes a precise correlation unlikely. Nevertheless, it is possible to see how the presence of a toxin or a precipitating agent could protect a plant against attack from a variety of organisms from viruses to herbivores. It has been long known, for example, that some plants are toxic to animals, and a protective function of PHA against beetle infestation of Phaseolus vulgaris has been demonstrated. Bruchid beetles are killed by a diet of black beans (P. uulgaris),but not by cowpeas (Vigna unguiculata) which are agglutinin-free (Janzen et al., 1976). These studies are of potential significance to agriculturalists, since the selection of a bean crop free of the agglutinin so as to reduce processing costs is likely to produce a crop with no beetle resistance. The elegant experiments of Shannon and his colleagues suggest that some lectins may protect plants against the spread of viral infections. It has been observed that certain plant viruses contain surface glycoproteins and that their presence correlates with the manner in which the virus is transmitted from plant to plant. For instance, barley stripe mosaic virus (BSMV) and cowpea mosaic virus are both transmitted via the seeds of the host plant, and both have surfaceassociated glycoproteins, while three non-seed-transmitted viruses studied (including a virus closely related to BSMV) do not contain carbohydrate (Partridge et al., 1974). Thus lectins (which are usually concentrated within the seed) may confer on the plant a selective advantage by protecting it against seed-transmitted viruses. In fact, barley has been found to contain a lectin which is specific for nonacetylated amino sugars. This lectin binds to and precipitates BSMV in uitro. Furthermore, virus treated in vitro with the lectin is noninfectious (L. Shannon, personal communication). It would be most interesting to determine the degree of resistance to BSMV infection of barley mutants that lack or produce low concentrations of the lectin. Although lectins are frequently concentrated in seeds, they can also be secreted from the roots of some plants, particularly legumes. It has therefore been suggested that a lectin may function in the binding of symbiotic rhizobia to form root nodules (Hamblin and Kent, 1973) and, indeed, some specificity seems to exist. Soybean lectin combines, for example, with most strains of Rhizobium japonicum, the soybean-nodulating bacterium, but not with strains of Rhizobium that fail to nodulate soybeans (Bohlool and Schmidt, 1974).Secreted lectin interacts with the bacterial 0-antigen-containing lipopolysaccharide and, when tested with a variety of lipopolysaccharides, lectins from a range of legumes have been found to bind only the lipopolysaccharide of their symboint Rhinobium (Wolpert and Albersheim, 1976). Most of the lectins that have been extensively studied come from

LECTINS

335

plants, but proteins with similar properties are known in other organisms where their functions have aroused considerable interest. For example, surrounding the eggs of the African clawed toad Xenopus laevis is a jelly coat composed of glycoprotein (Yurewicz et al., 1975). On fertilization, the cortical granules of the egg fuse with the plasma membrane and release their contents. The granule contents, which have the properties of a lectin, bind to the components of the jelly coat to form the fertilization envelope. The precipitation of the fertilization envelope probably serves as a block to polyspermy, fertilization of the egg by more than one sperm cell (Wyrick et al., 1974). A lectinlike protein has also been isolated from rabbit liver. This protein agglutinates erythrocytes and is inhibited by several monosaccharides including N-acetyl-D-glucosamine. It has been named mammalian hepatic lectin, is a constituent of the plasma membrane of hepatocytes, and has been shown to be involved in the removal of desialylated plasma proteins from the blood (Stockert et al., 1974; Lunney and Ashwell, 1976). This lectin is also mitogenic for lymphocytes. There is a series of interesting lectinlike proteins which have been isolated from cellular slime molds (Rosen et al., 1973, 1974, 1975; Simpson et al., 1974).These proteins, called discoidin I and discoidin 11, are located on the surface of cohesive but not vegetative cells. It is therefore thought that they may function in intercellular adhesion (Chang et d., 1975; Reithennan et d., 1975; Frazier et d., 1975). Finally, Teichberg et al. (1975) identified a saccharide-binding protein called electrolectin in the electric organ of Electrophorus electricus. This lectin agglutinates trypsinized rabbit erythrocytes, and agglutination is inhibited by compounds containing P-D-galactose groups but not by a variety of other saccharides. A similar lectin activity has been identified in chick embryonic muscle cells (Den et al., 1976) and in the L6 line of rat myoblasts (Gartner and Podleski, 1975). This galactose-specific lectin is found on the cell surface in both cell types, but there are conflicting results regarding whether it is involved in the specific fusion of myoblast cells to form myotubes; 15 mM thiodigalactoside inhibits the fusion of L6 rat myoblasts, but not that of chick embryonic myoblasts. It is clear from the scarcity of our knowledge about lectin function that considerable experimentation and perhaps some more good ideas will be required before we will have a reasonable understanding of this subject. It is ironic in fact that, although lectins have been successfully employed to probe a wide variety of cell surface structures and activities, we still do not know what their normal, in vivo functions are.

336

JAY C. BROWN AND RICHARD C. HUNT ACKNOWLEDGMENTS

We thank Drs. M. Hunt, G. Reeke, C. Wright, and A. Christopher for help in preparing the manuscript and Drs. L. Shannon, R. Emmons, T. Ukena, and C. Wright for communicating results to us prior to publication. Work in the authors’ laboratories was supported by grants from the American Cancer Society, the Medical Research Council (United Kingdom), and the U.S. Public Health Service. REFERENCES Abe, Y., Iwahuchi, M., and Ishii, S.-I. (1971). Biochem. Biophys. Res. Commun. 45, 1271. Agrawal, B. B. L., and Goldstein, I. J. (1965).Biochem. J . 96, 23c. Agrawal, B. B. L., and Goldstein, I. J. (1967). Biochim. Biophya. Acta 147,262. Agrawal, B. B. L., and Goldstein, I. J. (1968a).Can. J . Biochem. 46, 1147. Agrawal, B. B. L., and Goldstein, I. J. (1968b).Arch. Biochem. Biophys. 124,218. Agrawal, B. B. L., and Goldstein, I. J. (1973). In “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 313-318. Academic Press, New York. Allan, D., and Crumpton, M. J. (1971). Biochem. Biophys. Res. Commun. 44, 1143. Allan, D., Auger, J., and Crumpton, M. J. (1972). Nature (London),New Biol. 236, 23. Allen, A. K., and Neuberger, A. (1973).Biochem. J . 135,307. Allen, A. K., Neuberger, A., and Sharon, N. (1973). Biochem. J . 131, 155. Allen, L. W., Svenson, R. H., and Yachnin, S. (1969):Proc. Natl. Acad. Sci. U.S.A. 63, 334. Allen, N. K., and Brilliantine, L. (1969).J.Zmmunol. 102, 1295. Allwood, G., Asherson, G. L., Davey, M. J., and Goodford, P. J. (1971).Zmmunology 21, 509. Anderson, J., Edelman, G. M., Moller, G., and Sjoherg, 0. (1972).Eur. J . Immunol. 2, 233. Arndt-Jovin, D. J., and Berg, P. (1971).J.Virol. 8, 716. Aub, J. C., Tieslau, C., and Lankester, A. (1963). Proc. Natl. Acad. Sci. U S A . 50,613. Aub, J. C., Sanford, B. H., and Cote, M. N. (1965).Proc. Natl. Acad. Sci. U.S.A. 54,396. Auhery, M., Font, J., and Bourrillon, R. (1972). Exp. Cell Res. 71, 59. Bachrach, U., Gurevitch, J.. and Zaitschek, D. (1957).J.Zmmunol. 78,229. Badel, P., and Brilliantine, L. (1969). Proc. Soc. E x p . Biol. Med. 130, 621. Barbarese, E., Sauerwein, H., and Simpkins, H. (1973).J.Membr. Biol. 13, 129. Barnett, R. E., Scott, R. E., Furcht, L. T., and Kersey, J. H. (1974).Nature (London)249, 465. Becht, H., Rott, R., and Klenk, H.-D. (1972).J.Gen. ViroZ. 14, 1. Becker, F. F. (1974). Proc. Natl. Acad. Sci. U.S.A.71,4307. Becker, J. W., Reeke, G. N., Jr., and Edelman, G. M. (1971).]. Biol. Chem. 246,6123. Becker, J. W., Reeke, G. N., Jr., Wang, J. L., Cunningham, B. A., and Edelman, G. M. (1975).J.B i d . Chem. 250, 1513. Becker, J. W., Reeke, G. N., Jr., Cunningham, B. A., and Edelman, G. M. (1976).Nature (London)259,406. Ben-Bassat, H., Inbar, M., and Sachs, L. (1970).Virology 40,854. Ben-Bassat, H., Inhar, M., and Sachs, L. (1971).J.Membr. Biol. 6, 183. Ben-Bassat, H., Goldhlum, N., Manny, N., and Sachs, L. (1974).Znt. J. Cancer 14,367. Ben-Bassat, H., Goldblum, N., Mitrani, S., Klein, G., and Johansson, B. (1976). Znt. J . Cancer 17,448.

LECTINS

337

Benjamin, T. L., and Burger, M. M. (1970).Proc. Natl. Acad. Sci. U.S.A. 67, 929. Benson, S., Olsnes, S., Pihl, A,, Skorve, J., and Abraham, A. K. (1975). Eur.J. Biochem. 59,573. Berlin, R. D. (1975).Ado. E x p . Med. Biol. 55, 173- 186. Berman, L. D. (1975).Int. J. Cancer 15,973. Betschart, B., and Burger, M. M. (1975).Experentia 31,734. Bezkorovainy, A., Springer, G. F., and Desai, P. R. (1971). Biochemistry 10,3761. Biddle, F., Cronin, A. P., and Sanders, F. K. (1970). Cytobios 5, 9. Biquard, J.-M. (1973).Interuirology 1, 220. Biquard, J.-M., and Vigier, P. (1972). C.R. Hebd. Seances Acad. Sci., Ser. D 274, 144. Bird, G . W. G. (1951).Curr. Sci. 20,298. Bird, G . W. G. (1952).Nature (London) 170, 674. Bird, G . W. G. (1959a). Br. Med. Bull. 15, 165. Bird, G. W. G. (1959b). Nature (London) 184, 109. Bird, G . W. G. (1960).Br. J . Haematol. 6, 151. Birdwell, C. R., and Strauss, J. H. (1973).J. Virol. 11, 502. Bloch, R., and Burger, M. M. (1974). Biochem. Biophys. Res. Commun. 58, 13. Blumberg, S., Hildesheim, J., Yariv, J., and Wilson, K. J. (1972). Biochim. Biophys. Acta 264,171. Bohlool, B. B., and Schmidt, E. L. (1974). Science 185,269. Borek, C., and Fenoglio, M. (1976). Cancer Res. 36, 1325. Borek, C., Grob, M., and Burger, M. M. (1973). E x p . Cell Res. 77,207. Bourrillon R., and Font, J. (1968). Biochim. Biophys. Acta 154,28. Boyd, W. C. (1954).In “The Proteins” (H. Neurath and K. Bailey, eds.), Vol. 2, Part B, pp. 755-839. Academic Press, New York. Boyd, W. C. (1963).Vox Sang. 8, 1. Boyd, W. C., and Reguera, R. M. (1949).J.Immunol. 62,333. Boyd, W. C., Everhart, D. L., and McMaster, M. H. (1958).J.Immunol. 81,414. Boyde, A., Weiss, R. A., and Vesely, P. (1972).Exp. Cell Res. 71,313. Bretscher, M. S. (1973). Science 181, 622. Bretton, R., Wicker, R., and Bernhard, W. (1972). Znt. J . Cancer 10,397. Brewer, C. F., Sternlicht, H., Marcus, D. M., and Grollman, A. P. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 1007. Brinkley, B. R., Fuller, G. M., and Highfield, D. P. (1975).Proc. Natl. Acad. Sci. U.S.A. 72,4981. Brown, J. C. (1973). Biochem. Biophys. Res. Commun. 51,686. Bubel, H. C., and Blackman, K. E. (1975).Proc. SOC. Exp. Biol. Med. 148, 107. Burger, M. M. (1969).Proc. Natl. Acad. Sci. U.S.A. 62, 994. Burger, M. M. (1970a). Zn “Permeability and Function of Biological Membranes” (A. Bolis, ed.), pp. 107-119. North-Holland Publ., Amsterdam. Burger, M. M. (1970b). Nature (London)227, 170. Burger, M. M. (1971a).Growth Control Cell Cult., Ciba Found. Symp., 1970pp. 45-63. Burger, M. M. (1971b). Cun. Top. Cell. Regul. 3,135. Burger, M. M. (1973).Fed. Proc., Fed. Am. SOC. E x p . Biol. 32,91. Burger, M. M., and Goldberg, A. R. (1967). Proc. Natl. Acad. Sci. U.S.A. 57,359. Burger, M. M., and Martin, G. S. (1972).Nature (London),New Biol. 237,9. Burger, M. M., and Noonan, K. D. (1970). Nature (London) 228,512. Butters, T. D., and Hughes, R. C. (1975). Biochem. J. 150,59. Cabantchik, Z. I., and Rothstein, A. (1974a).J . Membr. Biol. 15,207. Cabantchik, Z. I., and Rothstein, A; (1974b).J. Membr. Biol. 15, 227.

338

JAY C. BROWN AND RICHARD C. HUNT

Carrasco, L., Fernandez-Puentes, C., and Vazquez, D. (1975). Eur. J. Biochem. 54,499. Carter, W. G., and Etzler, M. E. (1975a).J. Biol. Chern. 250,2756. Carter, W. G., and Etzler, M. E. (1975b). Biochemistry 14,2685. Caso, L. V. (1968).Anat. Rec. 162,459. Catsimpoolas, N., and Meyer, E. W. (1969).Arch. Biochem. Biophys. 132,279. Cazal, P., and Lalaurie, M. (1952).Acta Haematol. 8,73. Chang, C.-M., Reitherman, R. W., Rosen, S. D., and Barondes, S. H. (1975). Erp. Cell Res. 95, 136. Choi, Y. S., and Jenson, J. C. (1974).J. E x p . Med. 140,597. Chou, I.-N., Black, P. H., and Roblin, R. (1974). I n “Control of Proliferation in Animal Cells” (B. Clarkson and R. Baserga, eds.), pp. 339-350. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Chowdhury, T. K., and Weiss, A. K. (1975). Ado. E x p . Med. Biol. 55. Cline, M. J., and Livingston, D. C. (1971). Nature (London),New Biol. 232, 155. Cohen, E. (1974). Ann. N.Y. Acad. Sci. 234. Cohen, E., Rowe, A. W., and Wissler, F. C. (1965). Life Sci. 4, 2009. Cohen, E., Rozenberg, M., and Massaro, E. J. (1974).Ann. N.Y. Acad. Sci. 234,28. Collard, J . G., and Temmink, J. H. M. (1975).J.Cell Sci. 19, 21. Collard, J. G., and Temmink, J. H. M. (1976).J.Cell Biol. 68, 101. Collard, J. G., Temmink, J. H. M., and Smets, L. A. (1975).Adu. E x p . Med. Biol. 55,221. Comoglio, P. M., and Guglielmone, R. (1972). FEBS Lett. 27, 256. Cooper, E. L., Lemmi, C. A. E., and Moore, T. C. (1974).Ann. N.Y. Acad. Sci. 234,34. Cooper, H . L. (1972).Transplant. Reu. 11,3. Comick, J. W., and Stewart, J. E. ( 1 9 7 3 ) ~Znoertebr. Pathol. 21,255. Cuatrecasas, P. (1970).J . B i d . Chern. 245,3059. Cuatrecasas, P. (1973a).Biochemistru 12, 1312. Cuatrecasas, P. (1973b).J.Biol. Chem. 248,3528. Cuatrecasas, P., and Tell, G. P. E. (1973).Proc. Natl. Acad. Sci. U.S.A. 70,485. Culp, L. A., and Black, P. H. (1972).J. Virol. 9, 611. Culp, L. A., Grimes, W. J., and Black, P. H. (1971).J.Cell Biol. 50,682. Cunningham, B. A., Wang, J. L., Waxdal, M. J., and Edelman, G . M. (1975).J. Biol. Chern. 250, 1503. Dahl, J. L., and Hokin, L. E. (1974).Annu. Reo. Biochem. 43, 327. Datta, S. P., Cerini, M., Chose, T., and Cerini, J. (1969). Br. J. Cancer 23,616. DeMicco, P., and Berebbi, M. (1972). Znt. J . Cancer 10,249. Den, H., Malinzak, D. A., and Rosenberg, A. (1976).Biochem. Biophys. Res. Commun. 69, 621. Dent, P. B. (1971).J.Natl. Cancer Znst. 46, 763. Dent, P. B., and Hillcoat, B. (1972).J. Natl. Cancer Inst. 49, 373. DePetris, S., Raff, M. C., and Malluci, L. (1973).Nature (London),New B i d . 244,275. Desai, P. R., and Springer, G. F. (1973).I n “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 383-388. Academic Press, New York. DeSalle, L., Munakata, N., Pauli, R. M., and Strauss, B. S. (1972). Cancer Res. 32,2463. Eckhart, W., Dulbecco, R., and Burger, M. M. (1971).Proc. Natl. Acad. Sci. U.S.A. 68, 283. Edelman, G. M. (1974).I n “Control of Proliferation in Animal Cells” (B. Clarkson and R. Baserga, eds.), pp. 357-377. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Edelman, G. M.,and Millette, C. F. (1971).Proc. Natl. Acad. Sci. U.S.A. 68,2436. Edwards, H. E., Thomas, J. K., Burleson, G. R., and Kulpa, C. F. (1976). Biochim. Biophys. Actu 448, 451.

LECTINS

339

Elo, J., Estola, E., and Malmstrom, N. (1951). Ann. Med. E x p . Biol.Fenn. 29, 297. Entlicher, G., Ticha, M., Kbstii, J. V., and Kocourek, J. (1969). Experientia 25, 17. Entlicher, G., Khstii, J. V., and Kocourek, J. (1970). Biochim. Bioph?ys.Acta 221, 272. Entlicher, G., Khstii, J . V., and Kocourek, J. (1971). Biochim. Bioph!y.s.Acta 236, 795. Ersson, B., Aspberg, K., and Porath, J. (1973).Biochim.Biophys. Acta 310,446. Etzler, M. E. (1973).I n “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 340-344. Academic Press, New York. Etzler, M. E., and Kabat, E. A. (1970). Biochemistry 9,869. Femandez-Puentes, C., Benson, S., Olsnes, S., and Pihl, A. (1976).Eur.1. Biochem. 64, 437. Findlay, J. B. C. (1974)./. Biol. Chem. 249,4398. Fine, R. E., and Taylor, L. (1976). Exp. Cell Res. 102, 162. Fisher, D. B., and Mueller, G. C. (1971).Biochim.Bioph?ys.Acta 248,434. Follett, E. A. C., and Goldman, R. D. (1970). E x p . Cell Res. 59, 124. Fox, T. O., Sheppard, J. R., and Burger, M. M. (1971). Proc. Natl. Acad. Sci. U.S.A. 68, 244. Francois, D., Tuyen, V. V., Febvre, H., and Haguenau, F. (1972). C.R. Hehd. Seances Acad. Sci., Ser. D 274, 1981. Fraser, A. R., Hemperly, J. J., Wang, J. L. and Edelman, G. M. (1976). Proc. Natl. Acad. Sci. U.S.A. 73, 790. Frazier, W. A., Rosen, S. D., Reitherman, R. W., and Barondes, S. H. (1975).J . Biol. Chem. 250,7714. Fresen, K. O., and Illiger, H. J. (1974). Arch. Gesamte Virusforsch. 44, 377. Friberg, S., Golub, S. H., Lilliehook, B., and Cochran, A. J . (1972). E x p . Cell Res. 73, 101. Fuchs, P., Parola, A., Robbins, P. W., and Blout, E. R. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 3351. Fukuda, M., and Osawa, T. (1973)./. Biol. Chem. 248,5100. Gaffney, B. J. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 664. Gaf€ney, B. J., Branton, P. E., Wickus, G. G., and Hirschberg, C. B. (1974).In “Viral Transformation and Endogenous Viruses” (A. S. Kaplan, ed.), pp. 97-115. Academic Press, New York. Gahmberg, C. G., and Hakomori, S. (1973)./. B i d . Chem. 248,4311. Galbraith, W., and Goldstein, I. J. (1970). FEBS Lett. 9, 197. Galbraith, W., and Goldstein, I. J. (1973). I n “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 318-323. Academic Press, New York. Gantt, R. R., Martin, J. R., and Evans, V. J. (1969)./. Natl. Cancer Inst. 42,369. Ganido, J. (1975).E x p . Cell Res. 94, 159. Garrido, J., Burglen, M.-J., Samolyk, D., Wicker, R., and Bernhard, W. (1974). Cancer Res. 34, 230. Gartner, T. K., and Podleski, T. R. (1975). Biochem. Bioph?ys. Res. Commun. 67,972. Gericke, D., Chandra, P., and Wacker, A. (1971).Naturwis.sen.schaften 58,458. Gillespie, E. M., and Gold, E. R. (1960). Vox Sang. 5,497. Click, M. C., and Buck, C. A. (1973).Biochemistry 12,85. Glimelius, B., Westermark, B., and PontCn, J. (1974). Znt.‘/. Cancer 14,314. Glimelius, B., Nilsson, K., and PontBn, J. (1975).Int. /. Cuncer 15, 888. Goldberg, A. R. (1974).Cell 2, 95. Goldberg, N. D., Haddox, M. K., Estensen, R., White, J. G., Lopez, C., and Hadden, J. W. (1973). I n “Cyclic AMP, Cell growth and the Immune Response” (W. Braun, L. Lichtenstein, and C. Parker, eds.), pp. 247-262. Springer-Verlag, Berlin and New York.

340

JAY C. BROWN AND RICHARD C. HUNT

Goldstein, I. J., and So, L. L. (1965).Arch. Biochem. Biophys. 111,407. Goldstein, I. J., Hollerman, C. E., and Smith, E. E. (1965).Biochemistry 4,876. Goldstein, I. J., Reichert, C. M., and Misaki, A. (1974).Ann. N.Y. Acad. Sci. 234,283. Gonatas, N. K., Steiber, A., Kim, S. U., Graham, D. I., and Avrameas, S. (1975).E x p . Cell Res. 94,426. Gordon, J. A., and Marquardt, M. D. (1974).Biochim. Biophys. Acta 332, 136. Gordon, J. A., Blumberg, S., Lis, H., and Sharon, N. (1972). F E B S Lett. 24, 193. Gordon, J. A,, Blumberg, S., Lis, H., and Sharon, N. (1973).I n “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 365-368. Academic Press, New York. Gordon, M., Dandekar, P. V., and Bartoszewicz, W. (1974).J.Reprod. Fertil. 36, 211. Goto, M., Kataoka, Y., Goto, K., Yokoyama, T., and Sato, H. (1972).Gann 63, 505. Could, N. R., and Scheinberg, S. L. (1970).Arch. Biochem. Biophys. 137, 1. Greaves, M. F., and Bauminger, S . (1972).Nature (London),New Biol. 235,67. Greaves, M. F., Bauminger, S., and Janossy, G. (1972). Clin. E x p . Imrnunol. 10, 537. Greenaway, P. J., and LeVine, D. (1973).Nature (London), New Biol. 241, 191. Grollnian, A. P., Grunfeld, C., Brewer, C. F., and Marcus, D. M. (1974). Cancer Chemother. Rep., Part I 58,491. Gunther, G. R., Wang, J. L., Yahara, I., Cunningham, B. A., and Edelman, G. M. (1973). Proc. Natl. Acnd. Sci. U.S.A. 70, 1012. Gunther, G. R., Wang, J. L., and Edelman, G. M. (1974).J . Cell Biol. 62,366. Curd, J. W., and Mahler, H. R. (1974).Biochemistq 13, 5193. Hadden, J. W., Hadden, E. M., and Middleton, E., Jr. (1970).Cell. Immunol. 1, 583. Hadden, J. W., Hadden, E. M., Haddox, M. K., and Goldberg, N. D. (1972). Proc. Natl. Acad. Sci. U.S.A. 69,3024. Hakomori, S., and Murakami, W. T. (1968).Proc. Natl. Acad. Sci. U.S.A. 59,254. Hamblin, J., and Kent, S. P. (1973).Nature (London),New B i d . 245,28. Hammarstrom, S. (1973). In “Methods in Enzymology” (V. Ginsberg, ed.), Vol. 28, Part B, pp. 368-383. Academic Press, New York. Hammarstrom, S. (1974).Ann. N.Y. Acad. Sci. 234, 183. Hammarstrom, S., and Kabat, E. A. (1969).Biochemistry 8, 2696. Hardman, K. D., and Ainsworth, C. F. (1972a).Nature (London),New Biol. 237,54. Hardman, K. D., and Ainsworth, C. F. (1972b). Biochernistr!y 11,4910. Hardman, K. D., and Ainsworth, C. F. (1976).Biochemistry 15, 1120. Hayes, C. E., and Goldstein, I. J. (1974).J . Biol. Chem. 249, 1904. Hayes, C. E., and Goldstein, I. J. (1975).J . Biol. Chem. 250, 6837. Hayman, M. J., and Crumpton, M. J. (1972).Biochern. Biophys. Res. Cornmun. 47, 923. Henkart, P. A., and Fisher, R. I. (1975).J . Zmtnunol. 114, 710. Herzog, P. (1959).Z. Itntnunitaetsforsch. 117,53. Ho, M . K., and Guidotti, G. (1975).J . Biol. Chem. 250, 675. Honvitz, A. F., Hatten, M. E., and Burger, M. M. (1974).Proc. Natl. Acad. Sci. U.S.A.71, 3115. Hourani, B. T., Chace, N. M., and Pincus, J. H. (1973).Biochim. Biophys. Acta 328,520. Howard, I. K., and Sage, H. J. (1969). Biochemistry 8, 2436. Howard, I. K., Sage, H. J., Stein, M. D., Young, N. M., Leon, M. A., and Dyckes, D. F. (1971).J.B i d . Chem. 246, 1590. Hozumi, M., Miyake, S., Mizunoe, F., Sugimura, T., h i e , R. F., Koyama, K., Tomita M., and Ukita, T. (1972). I n t . J . Cancer 9, 393. Hsie, A. W., and Puck, T. T. (1971). Proc. Nutl. Acad. Sci. U.S.A. 68, 358. Hsie, A. W., Jones, C., and Puck, T. T. (1971).Proc. Nntl. Acud. Sci. U.S.A. 68, 1648. Huet, C. (1974). I n “Electron Microscopy and Cytochemistry” (E. Wisse et al., eds.), pp. 295-298. North-Holland Publ., Amsterdam.

LECTINS

34 1

Huet, C., and Bernhard, W. (1974). Znt. J . Cancer 13,227. Huet, C . , and Herzberg, M. (1973).J. Ultrustruct. Res. 42, 186. Huet, M. (1975). Eur. J . Biochem. 59,627. Hunt, R. C., and Brown, J. C. (1974).Biochemistry 13,22. Hunt, R. C., Bullis, C. M., and Brown, J. C. (1975). Biochemistry 14, 109. Hunt, R. C., Letts, P. J., Pinteric, L., and Schachter, H. (1977). Submitted for publication. Hunt, S . M., and Marchalonis, J. J. (1974). Biochem. Biophys. Res. Commun. 61, 1227. Hynes, R. 0. (1974). Cell 1, 147. Inbar, M., and Sachs, L. (1969a).Proq Natl. Acad. Sci. U.S.A. 63, 1418. Inbar, M., and Sachs, L. (196913).Nature (London)223, 710. Inbar, M., and Sachs, L. (1973). FEBS Lett. 32, 124. Inbar, M., Rabinowitz, Z., and Sachs, L. (1969).Znt. J. Cancer 4,690. Inbar, M., Ben-Bassat, H., and Sachs, L. (1971).Proc. Natl. Acad. Sci. U.S.A. 68,2748. Inbar, M., Ben-Bassat, H., and Sachs, L. (1972a).Nature (London),New B i d . 236,3. Inbar, M., Ben-Bassat, H., and Sachs, L. (1972b). Int. J . Cancer 9, 143. Inbar, M., Huet, C., Oseroff, A. R., Ben-Bassat, H., and Sachs, L. (1973a). Biochim. Biophys. Acta 311,594. Inbar, M., Ben-Bassat, H., and Sachs, L. (197313).Exp. Cell Res. 76, 143. Ishiguro, M., Takahashi, T., Funatsu, G., Hayashi, K., and Funatsu, M. (1964).J. Biochem. (Tokyo)55,587. Ishiyama, I., Mukaida, M., and Takatsu, A. (1974). Ann. N . Y . Acad. Sci. 234, 75. Jansons, V. K., and Burger, M. M. (1973). Biochirn.Biophys. Acta 291, 127. Janzen, D. H., Juster, H. B., and Liener, I. E. (1976). Science 192,795. Johnson, G . S., Friedman, R. M., and Pastan, I. (1971). Proc. Natl. Acad. Sci. U.S.A.68, 425. Johnson, K. E., and Smith, E. P. (1976). Exp. Cell Res. 101, 63. Johnson, M. H., Eager, D., Muggleton-Harris, A,, and Grave, H. M. (1975).Nature (London) 257,321. Kabat, E. A., Heidelberger, M., and Bezer, A. E. (1947).J.B i d . Chem. 168,629. Kalb, A. J,, and Lustig, A. (1968).Biochim. Biophys. Acta 168, 366. Kaneko, I., Satoh, H., and Ukita, T. (1973). Biochem. Biophys. Res. Commun. 50, 1087. Kapeller, M., and Doljanski, F. (1972). Nature (London),New B i d . 235. 184. Kaplan, M. E., and Kabat, E. A. (1966).J.Exp. Med. 123,1061. Kaplowitz, P. B., and Moscona, A. (1973).Biochem. Biophys. Res. Commun. 55, 1326. Kashiwabara, T., Tanaka, R., and Matsumoto, T. (1965).Nature (London) 207,831. Kleinschuster, S. J., and Moscona, A. A. (1972).Exp. Cell Res. 70,397. Komfeld, R., and Komfeld, S. (1970).J. Biol. Chem. 245,2536. Kornfeld, R., Gregory, W. T., and Komfeld, S. A. (1973). I n “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 344-349. Academic Press, New York. Komfeld, S., Rogers, J., and Gregory, W. (1971).J.Biol. Chem. 246,6581. Komfeld, S., Eider, W., and Gregory, W. (1974).In “Control of Proliferation in Animal Cells’’ (B. Clarkson and R. Baserga, eds.), pp. 435-445. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Koulumies, R. (1949).Ann. Med. E x p . Biol. Fenn. 27,20. Krach, S. W., Green, A., Nicolson, G . L., and Oppenheimer, S. (1974).Exp. Cell Res. 84, 191. Kriipe, M., and Braun, C. (1952). Naturwissenschaften 39,284. Kriipe, M., and Pieper, H. (1966). Z. Zmmunitaetsforsch. 130, 296. Kurth, R., and Bauer, H. (1973).Nature (London),New Biol. 243,243. Lallier, R. (1972). Exp. Cell Res. 72, 157.

342

JAY C. BROWN AND RICHARD C. HUNT

Lehman, J. M., and Sheppard, J. R. (1972).Virology 49,339. Letarte-Muirhead, M., Barclay, N., and Williams, A. F. (1975).Biochem. J. 151, 685. LeVine, D., Kaplan, M. J., and Greenaway, P. J. (1972).Biochem. J . 129,847. Liener, I. E. (1953).J.Nutr. 49, 527. Liener, I. E., and Pallansch, M. J. (1952).J.Biol. Chern. 197,29. Lin, H., and Bruce, W. R. (1971).Proc. Am. Assoc. Cancer Res. 12, 66. Lin, J. Y., Tsemg, K. Y., Chen, C.-C., Lin, L. T., andTung, T. C. (1970).Nature (London) 227,292. Lin, J.-Y., Liu, K., Chen, C.-C., and Tung, T.-C. (1971).Cancer Res. 31,921. Lin, J.-Y., Pao, C. C., Ju, S.-T., and Tung, T.-C. (1972). Cancer Res. 32, 943. Lindahl-Kiessling, D. (1972).Erp. Cell Res. 70, 17. Lis, H., and Sharon, N. (1973a). In “Methods in Enzymology” (V. Ginsburg, ed.). Vol. 28, Part B, pp. 360-365. Academic Press, New York. Lis, H., and Sharon, N . (1973b).Annu. Reo. Biochem. 42,541. Lis, H., Sharon, N., and Katchalski, E. (1966a).J.Biol. Chem. 241. 684. Lis, H., Fridman, C., Sharon, N., and Katchalski, E. (1966b). Arch. Biochem. Biophys. 117,301. Lis, H., Sela, B. A,, Sachs, L., and Sharon, N. (1970). Biochim. Biophys. Acta 211,582. Liske, R., and Franks, D. (1968). Nature (London) 217,860. Loor, F. (1973). Exp. Cell Res. 82,415. Loor, F., and Hagg, L.-B.(1975).Eur. J. Immunol. 5,854. Lotan, R., Gussin, A. E. S., Lis, H., and Sharon, N. (1973a). Biochem. Biophys. Res. Commun. 52,656. Lotan, R., Lis, H., Rosenwasser, A., Novogrodsky, A,, and Sharon, N. (197313).Biochem. Biophys. Res. Commun. 55, 1347. Lotan, R., Siegelman, H. W., Lis, H., and Sharon, N. (1974).J.Biol. Chem. 249, 1219. Lotan, R., Debray, H., Cacan, M., Cacan, R., and Sharon, N. (1975a).J.Biol. Chem. 250, 1955. Lotan, R., Skutelsky, E., Danon, D., and Sharon, N. (1975b).J.B i d . Chem. 250,8518. Luckasen, J. R., White, J. G., and Kersey, J. H. (1974).Proc. Natl. Acad. Sci. U.S.A. 71, 5088. Lugnier, A,, and Dirheimer, G. (1973).FEES Lett. 35, 117. Lugnier, A. A. J., Dirheimer, G., Madjar, J. J., Reboud, J.-P., Gorden, J., and Howard, G. A. (1976). FEES Lett. 67,343. Lunney, J., and Ashwell, G. (1976). Proc. Natl. Acad. Sci. U.S.A. 73,341. Maca, R. D. (1976). Cancer Res. 36, 1218. McDonough, J., and Lilien, J. (1975).J.Cell Sci. 19, 357. McIlhinney, A., and Hogan, B. (1974). Biochem. Biophys. Res. Commun. 60,348. McKenzie, G. H., Sawyer, W. H., and Nichol, L. W. (1972).Biochim. Biophys. Acta 263, 283. McNutt, N. S., Culp, L. A., and Black, P. H. (1971).J.Cell B i d . 50,691. McNutt, N. S., Culp, L. A., and Black, P. H. (1973).J.Cell B i d . 56,412. McPherson, A., and Rich, A. (1973). FEES Lett. 35,257. Maino, V. C., Green, N. M., and Crumpton, M. J. (1974).Nature (London) 251,324. Makela, 0 .(1957).Ann. Med. E x p . Biol. Fenn. 35, 1. Makela, O., and Makela, P. (1956). Ann. Med. E r p . Biol. Fenn. 34,402. Malick, L. E., and Langenbach, R. (1976).J.Cell B i d . 68, 654. Marchalonis, J. J., and Edelman, G. M. (1968).J.M o l . Biol. 32,453. Marchesi, V. T. (1973).In “Methods in Enzymology” (V. Ginsburg, ed.). Vol. 28, Part B, pp. 354-356. Academic Press, New York.

LECTINS

343

Marchesi, V. T., Tillack, T. W., Jackson, R. L., Segrest, J. P., and Scott, R. E. (1972). Proc. Natl. Acad. Sci. U.S.A. 69, 1445. Marik, T., Entlicher, G., and Kocourek, J. (1974). Biochirn. Biophys. Acta 336,53. Marikovsky, Y., Inbar, M., Danon, D., and Sachs, L. (1974). Exp. Cell Res. 89,359. Martinez-Palomo, A., Wicker, R., and Bernhard, W. (1972). Znt. J. Cancer 9,676. Martinozzi, M., and Moscona, A. (1975). Exp. Cell Res. 94,253. Mastro, A. M., and Mueller, G . C. (1974). Exp. Cell Res. 88,40. Matsumoto, I., and Osawa, T. (1969).Biochim. Biophys. Acta 194, 180. Matsumoto, I., and Osawa, T. (1970).Arch. Biochem. Biophys. 140,484. Matsumoto, I., and Osawa, T. (1972). Biochem. Biophys. Res. Commun. 46, 1810. Micklem, K. J., Abra, R. M., Knutton, S., Graham, J. M., and Pastemak, C. A. (1976). Biochem. 1. 154,561. Miller, J. B., Noyes, C., Heinrikson, R., Kingdon, H., and Yachnin, S. (1973).J. Exp. Med. 138, 939. Monroy, A., Ortolani, G., O’Dell, D., and Millonig, G. (1973).Nature (London)242,409. Montanaro, L., Sperti, S., and Stilpe, F. (1973). Biochem. J . 136,677. Moran, D. (1974). Erp. Cell Res. 86,365. Morgan, W. T. J., and Watkins, W. M. (1959). Br. Med. Bull. 15, 109. Moscona, A. A. (1971). Science 171,905. Nachman, R. L., Hubbard, A., and Ferris, B. (1973).J. Biol. Chem. 248,2928. Nagata, Y., and Burger, M. M. (1972).J. Biol. Chem. 247,2248. Nagata, Y., and Burger, M. M. (1974).J. Biol. Chem. 249,3116. Nakamura, J., and Terayama, H. (1975). Proc. Natl. Acad. Sci. U.S.A. 72,498. Neri, A., Roberson, M., Connolly, D. T., and Oppenheimer, S. B. (1975).Nature (London) 258,342. Neri, G., Smith, D. F., Gilliam, E., and Walborg, E. F., Jr. (1974). Arch. Biochem. Biophys. 165, 323. Nicolson, G . L. (1971). Nature (London),New Biol. 233,244. Nicolson, G . L. (1972). Nature (London)New Biol. 239, 193. Nicolson, G. L. (1973a).J. Natl. Cancer Znst. 50, 1443. Nicolson, G . L. (1973b). Nature (London)New B i d . 243,218. Nicolson, G. L. (1974a).In “Control of Proliferation in Animal Cells” (B. Clarkson and R. Baserga, eds.), pp. 251-270. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Nicolson, G. L. (1974b).Nature (London)251,628. Znt. Reu. Cytol. 39,89. Nicolson, G . L. (1974~). Nicolson, G . L. (1975).Adu. Exp. Med. B i d . 55, 153-172. Nicolson, G . L. (1976a). Biochim. Biophys. Acta 457, 57. Nicolson, G . L. (1976b). Biochim. Biophys. Acta 458, 1. Nicolson, G . L., and Blaustein, J. (1972). Biochim. Biophys. Acta 266,543. Nicolson, G . L., and Lacorbiere, M. (1973).Proc. Natl. Acad. Sci. U.S.A. 70, 1672. Nicolson, G . L., and Painter, R. G . (1973).J . Cell Biol. 59,395. Nicolson, G . L., and Yanagimachi, R. (1972). Science 177,276. Nicolson,‘G. L., and Yanagimachi, R. (1974). Science 184, 1294. Nicolson, G. L., Lacorbiere, M., and Yanagimachi, R. (1972).Proc. SOC. Exp. Biol. Med. 141, 661. Nicolson, G . L., Blaustein, J., and Etzler, M. E. (1974). Biochemistry 13, 196. Nicolson, G . L., Lacorbiere, M., and Eckhart, W. (1975a).Biochemistry 14, 172. Nicolson, G. L., Yanagimachi, R., and Yanagimachi, H. (1975b).J. Cell Biol. 66,263. Cancer Res. 35, 144. Nicolson, G . L., Lacorbiere, M., and Hunter, T. R. (1975~).

344

JAY C. BROWN AND RICHARD C. HUNT

Nicolson, G. L., Smith, J. R., and Poste, G. (1976).J. Cell Biol. 68,395. Noonan, K. D., and Burger, M. M. (1973a).I . B i d . Chem. 248,4286. Noonan, K. D., and Burger, M. M. (197313).J. Cell Biol. 59, 134. Noonan, K. D., Renger, H. C., Basilico, C., and Burger, M. M. (1973a).Proc. Natl. Acad. Sci. U.S.A. 70,347. Noonan, K. D., LeVine, A. J., and Burger, M. M. (1973b).J. Cell Biol. 58,491. Noonan, K. D., Wright, P. J., Bouck, N., and DiMayorca, G. (1976).J. Virol. 18, 1134. Novogrodsky, A., and Katchalski, E. (1971). Biochim. Bioph!ys. Acta 228,579. Nowell, P. C. (1960). Cancer Res. 20,462. Oikawa, T., Yanagimachi, R., and Nicolson, G. L. (1973). Nature (London)241,256. Oikawa, T., Nicolson, G. L., and Yanagimachi, R. (1974). E x p . Cell Res. 83,239. Olsnes, S. (1972). Naturwissenschaften 59,497. Olsnes, S., and Pihl, A. (1972a).FEBS Lett. 20,327. Olsnes, S., and Pihl, A. (197213).Nature (London)238,459. Olsnes, S., and Pihl, A. (1973a).Biochemistry 12,3121. Olsnes, S., and Pihl, A. (1973b).Eur. I . Biochem. 35, 179. Olsnes, S., Saltvedt, E., and Pihl, A. (1974a).J. Biol. Chem. 249, 803. Olsnes, S., Pappenheimer, A. M., and Meren, R. (1974b).J. Zmmunol. 113,842. Olsnes, S., Refsnes, K., Christensen, T. B., and Pihl, A. (1975). Biochim. Biophys. Acta 405, 1. Olsnes, S., Sandvig, K., Refsnes, K., and Pihl, A. (1976).J. B i d . Chem. 251,3985. Olson, M. 0. J., and Liener, I. E. (1967). Biochemistry 6, 105. Onozaki, K., Tomita, M., Sakurai, Y.,and Ukita, T. (1972).Biochem. Biophys. Res. Commun. 48,783. Onozaki, K., Hayatsu, H., and Ukita, T. (1975).Biochim. Biophys. Acta 407,99. Oppenheimer, S. B., and Odencrantz, J. (1972).E x p . Cell Res. 73,475. Osawa, T., and Matsumoto, I. (1973).In “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 323-327. Academic Press, New York. Osawa, T., and Toyoshima, S. (1972).In “Methods of Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 328-332. Academic Press, New York. Ossowski, L., Quigley, J. P., and Reich, E. (1974)./. Biol. Chem. 249,4312. Ottensooser, F., and Sato, M. (1963).Vox Sang. 8,733. Ottensooser, F., Sato, R., and Sato, M. (1968).Transfusion (Philadelphia)8,44. Ozanne, B. (1973).J. Virol. 12,79. Ozanne, B., and Sambrook, J. (1971a).Nature (London),New Biol. 232, 156. Ozanne, B., and Sambrook, J. (1971b).In “Biology of Oncogenic Viruses” (L. G. Silvestri, ed.), 2nd Lepetit Colloq., pp. 248-257. North-Holland Publ., Amsterdam. Papahadjopoulos, D., Jacobson, K., Poste, G., and Shepherd, G. (1975). Biochim. Biophys. Acta 394,504. Pappenheimer, A. M., Olsnes, S., and Harper, A. A. (1974).J. Immunol. 113,835. Pardoe, G. I., and Uhlenbruck, G. (1970)./. Med. h b . Technol. 27,249. Partridge, J. E., Shannon, L. M., Gumpf, D. J., and Colbaugh, P. (1974).Nature (London) 247,391. Pauley, G. B., Granger, G. A., and Krassner, S. M. (1971).J. Inuertebr. Pathol. 18,207. Pereira, M. E. A., and Kabat, E. A. (1974). Biochemistry 13,3184. Peters, J. H., and Hausen, P. (1971).Eur. I . Biochem. 19, 502. Phillips, P. G., Furmanski, P., and Lubin, M. (1974).E x p . Cell Res. 86,301. Pienkowski, M. (1974). Proc. Soc. E x p . Biol. Med. 145,464. Pollack, R., and Rifkin, D. (1975). Cell 6,495. Pollack, R., Wolman, S., and Vogel, A. (1970).Nature (London)228,938.

LECTINS

345

Pollack, R., Osborn, M., and Weber, K. (1975).Proc. Natl. Acad. Sci. U.S.A. 72,994. Pollack, R. E., and Burger, M. M. (1969).Proc. Nutl. Acad. Sci. U.S.A. 62, 1074. Poretz, R. D. (1973).I n “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 349-354. Academic Press, New York. Poretz, R. D., and Goldstein, I. J. (1970).Biochemistry 9,2890. Poretz, R. D., Riss, H., Timberlake, J. W., and Chien, S.-M. (1974). Biochemistry 13, 250. Porter, K. R., Prescott, D., and Frye, J. (1973a).J.Cell Biol. 57,815. Porter, K. R., Todaro, G. J., and Fonte, V. (1973b).J . Cell B i d . 59, 633. Poste, G. (1972).E x p . Cell Res. 73, 319. Poste, G. (1975).Ado. E x p . Med. Biol. 55, 117-152. Poste, G., and Nicolson, G. L. (1976).Biochim. Bioph4.s. Acta 426, 148. Poste, G., and Reeve, P. (1972).Nature (London)New Biol. 237, 113. Poste, G., and Reeve, P. (1974).Nature (London)247,469. Poste, G . , Reeve, P., Alexander, D. J., and Terry, G. (1972).J.Gen. Virol. 17,81. Poste, G., Papahadjopoulos, D., Jacobson, K., and Vail, W. J. (1975a).Nature (London) 253,552. Poste, G., Papahadjopoulos, D., Jacobson, K., and Vail, W. J. (1975b).Biochim. Biophys. Acta 394, 520. Poste, G., Papahadjopoulos, D., and Nicolson, G. L. (1975~).Proc. Natl. Acad. Sci. U.S.A. 72,4430. Potapov, M. I. (1968).Zzu. Akad. Nauk SSSR (Leipzig) Ser. Biol. 5,712. Potapov, M. I. (1970).Folia Haematol. 93,458. Presant, C. A., and Komfeld, S. (1972).J . Biol. Chem. 247,6937. Prokop, O., Schlesinger, D., and Rackwitz, A. (1965).Z. Immunitaetsforsch. 129,402. Prokop, O., Schlesinger, D., and Geserick, G. (1967).Z. Immunitaetsforsch. 132,491. Prokop, O., Uhlenbruck, G., and Kohler, W. (1968). Vox Sang. 14,321. Pusztai, A., and Watt, W. B. (1974). Biochim. Biophys. Acta 365,57. Quastel, M. R., and Kaplan, J. G. (1970a).E x p . Cell Res. 62,407. Quastel, M. R., and Kaplan, J. G. (1970b).E x p . Cell Res. 63,230. Quigley, J. P., Rifkin, D. B., and Reich, E. (1971).Virology 46, 106. Quigley, J. P., Rifkin, D. B., and Reich, E. (1972).Virology 50,550. Raff, M. C., DePetris, S., and Mallucci, L. (1974).In “Control of Proliferation in Animal Cells” (B. Clarkson and R. Baserga, eds.), pp. 271-282. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Ralph, P., and Nakoinz, I. (1973).J.Natl. Cancer Inst. 51,883. Rapin, A. M . C., and Burger, M. M. (1974).Ado. Cancer Res. 20, 1-91. Reeke, G. N., Jr., Becker, J. W., and Edelman, G. M. (1975).J.Biol. Chem. 250,1525. Reeve, P., Poste, G., Alexander, D. J., and Pope, G. (1972).J . Gen. Virol. 15,219. Reeve, P., Poste, G., and Alexander, D. J. (1975).In “Negative Strand Viruses” (B. W. J. Mahy and R. D. Barry, eds.), Vol. 2, pp. 859-873. Academic Press, New York. Refsnes, K., Olsnes, S., and Pihl, A. (1974).]. B i d . Chem. 249,3557. Reisfeld, R. A., Borjeson, J., Chessin, L. N., and Small, P. A., Jr. (1967).Proc. Natl. Acud. Sci. U.S.A. 58,2020. Reitherman, R. W., Rosen, S. D., Frazier, W. A., and Barondes, S. H. (1975).Proc. Notl. Acad. Sci. U.S.A.72,3541. Renger, H. C., and Basilico, C. (1972).Proc. Natl. Acad. Sci. U.S.A. 69, 109. Renkonen, K. 0. (1948).Ann. Med. E x p . Biol. Fenn. 26,66. Rice, R. H., and Etzler, M. E. (1974).Biochem. Biophys. Res. Commun. 59,414. Rice, R. H., and Etzler, M. E. (1975).Biochemistq 14,4093.

346

JAY C. BROWN AND RICHARD C. HUNT

Rigas, D. A., and Osgood, E. E. (1955).J.Biol. Chem. 212, 607. Rittenhouse, H. G., Williams, R. E., Wisnieski, B., and Fox, C. F. (1974). Biochem. Biophys. Res. Commun. 58,222. Rittenhouse, H. G., Williams, R. E., and Fox, C. F. (1976).Biochem. Biophys. Res. Commun. 71, 975. Robbins, P. W., Wickus, G. G., Branton, P. E., Gaffney, B. J., Hirschberg, C. B., Fuchs, P., and Blumberg, P. M. (1974). Cold Spring Harbor Symp. Quant. Biol. 39, 11731180. Roberson, M., and Oppenheimer, S. B. (1975). Exp. Cell Res. 91,263. Robinson, E., and Mekori, T. (1972). Zsr. J . Med. Sci. 7,83. Roguet, R., and Bourrillon, R. (1975a).Biochim. Biophys. Acta 389,380. Roguet, R., and Bourrillon, R. (1975b). Biochem. J . 152,421. Rosen, S. D., Kafka, J. A., Simpson, D. L., and Barondes, S. H. (1973).Proc. Natl. Acad. Sci. U.S.A. 70, 2554. Rosen, S. D., Simpson, D. L., Rose, J. E., and Barondes, S. H. (1974).Nature (London) 252, 128. Rosen, S . D., Reitherman, R. W., and Barondes, S. H. (1975).Exp. Cell Res. 95, 159. Rosenblith, J. Z., Ukena, T. E., Yin, H. H., Berlin, R. D., and Karnovsky, M. J. (1973). Proc. Natl. Acud. Sci. U.S.A. 70, 1625. Roth, J. (1974). Int. J . Cancer 14,762. Roth, J., Meyer, H. W., Neupert, G . , and Bolck, F. (1973). Exp. Pathol. 8, 19. Roth, J., Wagner, M., and Meyer, H. W. (1974). Pathol. Eur. 9,31. Roth, J., Neupert, G., and Bolck, F. (1975).Exp. Pathol. 10, 143. Rottmann, W. L., Walther, B. T., Hellerqvist, C. G., Umbreit, J., andRoseman, S. (1974). J . Biol. Chem. 249,373. Rowinski, J., Solter, D., and Koprowski, H. (1976). Exp. Cell Res. 100,404. Rowlatt, C., Wicker, R., and Bemhard, W. (1973).Znt.J. Cancer 11,314. Rozengurt, E., and Heppel, L. A. (1975).Proc. Natl. Acad. Sci. U.S.A. 72,4492. Rutishauser, V., and Sachs, L. (1974). Proc. Natl. Acad. Sci. U.S.A. 71,2456. Rutishauser, V., and Sachs, L. (1975).J . Cell Biol. 65,247. Sage, H. J., and Connett, S. L. (1969).J.Biol. Chem. 244,4713. Sage, H. J., and Green, R. W. (1973). In “Methods in Enzymology” (V. Ginsburg, ed.), Vol. 28, Part B, pp. 332-339. Academic Press, New York. Saint-Paul, M. (1961). Transfusion (Paris) 4 , 3 . Sakiyama, H., and Robbins, P. W. (1973).Fed. Proc., Fed. Am. S O C . E x p . Biol. 32,86. Salzberg, S., and Green, M.(1972). Nature (London),New Biol. 240, 116. Salzberg, S., and Raskas, H. J . (1972). Virology 48,631. Sandvig, K., Olsnes, S., and Pihl, A. (1976).J.Biol. Chem. 251, 3977. Scarborough, G. A. (1975).J.Biol. Chem. 250, 1106. Schnebli, H. P. (1974). In “Control of Proliferation in Animal Cells” (B. Clarkson and R. Baserga, eds.), pp. 327-337. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Sefton, B. M., and Rubin, H. (1970). Nature (London) 227, 843. Sela, B.-A., Lis, H., Sharon, N., and Sachs, L. (1970).J.Membr. Biol. 3,267. Sela, B.-A., Lis, H., Sharon, N., and Sachs, L. (1971). Biochim. Biophys. Acta 249,564. Shaper, J. H., Barker, R., and Hill, R. L. (1973).Anal; Biochem. 53,564. Sharon, N., and Lis, H. (1972). Science 177, 949. Sharon, N. (1977).Scientijic American 236, 108. Sheppard, J . R. (1971).Proc. Natl. Acad. Sci. U.S.A.68, 1316. Sheppard, J. R. (1972). Nature (London),New Biol. 236, 14.

LECTINS

347

Shoham, J., and Sachs, L. (1972). Proc. Natl. Accid. Sci. U . S . A . 69, 2479. Shoham, J., and Sachs, L. (1974a). E x p . Cell Res. 85,8. Shoham, J., and Sachs, L. (1974b). I n “Control of Proliferation in Animal Cells” B. Clarkson and R. Baserga, eds.), pp. 297-304. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Shoham, J., Inhar, M., and Sachs, L. (1970).Nature (London) 227, 1244. Simpson, D. L., Rosen, S. D., and Barondes, S. H. (1974). Biochemistry 13,3487. Singer, S. J., and Nicolson, G. L. (1972).Science 175, 720. Sivak, A., and Wolman, S. R. (1972).In Vitro 8, 1. Skehan, P., and Friedman, S. J. (1975).E x p . Cell Res. 92, 350. Smets, L. A. (1973).Nature (London),New Biol. 245, 113. Smets, L. A., and Broekhuysen-Davies, J. (1972).Eur. J . Cancer 8, 541. Smith, D. F., Neri, G., and Walborg, E. F., Jr. (1973). Biochemistry 12, 2111. Smith, S. B., and Revel J.-P. (1972). Deu. B i d . 27,434. Sperti, S . , Montanaro, L., Mattioli, A,, and Stirpe, F. (1973).Biochem. J . 136,813. Sperti, S., Montanaro, L., Mattioli, A,, Testoni, G., and Stripe, F. (1976).Biochem.]. 156, 7. Springer, G. F., and Williamson, P. (1962).Biochem. J . 85,282. Springer, G. F., Takahashi, T., Desai, P. R., and Kolecki, B. J. (1965). Biochemistry 4, 2099. Stein, M. D., Howard, I. K., and Sage, H. J. (1971).Arch. Biochem. Biophys. 146,353. Steinberg, M. S., and Gepner, I. A. (1973).Nature (London),New Biol. 241,249. Steinemann, A., and Stryer, L. (1973).Biochemistry 12, 1499. Stobo, J. D., Rosenthal, A. S., and Paul, W. E. (1972).J.Zmmunol. 108, 1. Stockert, R. J., Morell, A. G., and Scheinberg, I. H. (1974). Science 186,365. Sundberg, L., Porath, J.. and Aspberg, K. (1970).Biochim. Bioph!ys. Acta 221,394. Surolia, A., Bachhawat, B. K., and Podder, S. K. (1975). Nature (London)257,802. Taylor, P., Jones, J. W., and Jacobs, N. M. (1974). M o l . Pharmacol. 10,78. Teichherg, V. I., Silman, I., Beitsch, D. D., and Resheff, G. (1975).Proc. Natl. Acad. Sci. U.S.A. 72, 1383. Temmink, J. H. M., and Collard, J. G. (1974).In “Electron Microscopy and Cytochemistry” (E. Wisse et al., eds.), pp. 299-301. North-Holland Publ., Amsterdam. Temmink, J. H. M., Collard, J. G., Spits, H., and Roos, E. (1975).Ezp. Cell Res. 92,307. Temmink, J. H. M., Collard, J. C., Roosien, J., and Van den Bosch, J. F. (1976).]. Cell Sci. 21, 563. Tktry, A,, Sutton, E., and Moullec, J. (1954). C. R. Hebd. Seances Acad. Sci. 238,277. Tevethia, S . S., Lowry, S., Rawls, W. E., Melnick, J. L., and McMillan, V. ( 1 9 7 2 ) Gen. ~ Virol. 15, 93. Thomas, D. B., and Winder, R. J. (1969). Biochem. Biophys. Res. Commun. 35, 811. TichB, M., Entlicher, G., Kbstii, J., and Kocourek, J. (1970).Biochim. Biophys. Acta 221, 282. Tomita, M., and Marchesi, V. T. (1975).Proc. Natl. Acad. Sci. U.S.A. 72,2964. Tomita, M., Kurokawa, T., Osawa, T., Sakurai, Y., and Ukita, T. (1972a).Gann 63,269. Tomita, M., Kurokawa, T., Onozaki, K., Ichiki, N., Osawa, T., and Ukita, T. (1972b). Experientia 28,84. Toms, G. C., and Western, A. (1971). In “Chemotaxonomy of the Leguminosae” (J. B. Harborne, D. Boulter, and B. L. Turner, eds.), pp. 367-462. Academic Press, New York. Torpier, G., and Montagnier, L. (1973).Int. J . Cancer 11,604. Toyoshima, S . , and Osawa, T. (1975).I . Biol. Chem. 250, 1655.

348

JAY C. BROWN AND RICHARD C. HUNT

Toyoshima, S., Osawa, T., and Tonomura, A. (1970).Biochim. Biophys. Acta 221,514. Toyoshima, S., Akiyarna, Y., Nakano, K., Tonornura, A., and Osawa, T. (1971).Biochemistry 10,4457. Treska-Ciesielski, J., Gombos, G., and Morgan, I. G. (1971).C. R. Hebd. Seances Acad. Sci., Ser. D 273, 1041. Tripp, M. R. (1966).J . Inoertebr. Pathol. 8,478. Tripp, M. R. (1974).Ann. N.Y. Acad. Sci. 234, 18. Trowbridge, I. S. (1974).J . Biol. Chem. 249,6004. Trowbridge, I. S., and Hilborn, D. A. (1974).Nature (London) 250,304. Tsuyuki, H., van Kley, H., and Stahmhn, M. A. (1956).J. Am. Chem. Soc. 78,764. Uchida, T., and Matsurnoto, T. (1972).Biochim. Biophys. Acta 257,230. Uhlenbruck, G., and Herrmann, W. P. (1972).Vox Sang. 23,444. Ukena, T. E., and Kamovsky, M. J. (1976).In “Membranes and neoplasia: New Approaches and Strategies” (V. Marchesi, ed.), pp. 261-273. Alan R. Liss, Inc., New York. Ukena, T. E., Borysenko, J. Z., Kamovsky, M. J.,and Berlin, R. D. (1974).].Cell Biol. 61,70. Ukena, T. E., Goldrnan, E., Benjamin, T. L., and Kamovsky, M. J. (1976).Cell 7,213. Unkeless, J. C.,Tobia, A., Ossowski, L., Quigley, J. P., Rifkin, D. B., and Reich, E. (1973).J . E x p . Med. 137,85. Unkeless, J. C., Dano, K., Kellerman, G. M., and Reich, E. (1974).J. Biol. Chem. 249, 4295. Van der Noordaa, J., van Haagen, A,, Wallboomers, J. M. M., and van Someren, H. (1972).J . Virol. 10,67. Van Nest, G. A., and Grimes, W. J. (1974).Cancer Res. 34, 1408. Van Veen, J., Roberts, R. M., and Noonan, K. D. (1976).J . Cell Biol. 70,204. Van Wauwe, J. P., Loontieus, F. G., and d e Bruyne, C. K. (1975).Biochim. Biophys. Acta 379,456. Vesel?, P., Entlicher, G., and Kocourek, J. (1972).Erperientia 28, 1085. Wada, S.,Pallansch, M. J., and Liener, I. E. (1958).J.Biol. Chem. 233,395. Wands, J. R., Podolsky, D. K., and Isselbacher, K. J. (1976).Proc. Natl. Acad. Sci. U.S.A. 73,2118. Wang, J. L., Becker, J. W., Reeke, G. N., Jr., and Edelman, G. M. (1974)~ Mol. Biol. 88, 259. Wang, J. L., Cunningham, B. A., Waxdal, M. J., and Edelman, G. M. (1975a).J . Biol. Chem. 250,1490. Wang, J. L., McClain, D. A., and Edelman, G. M. (1975b).Proc. Natl. Acad. Sci. U S A . 72, 1917. Watson, J. (1975).J . E r p . Med. 141,97. Weber, J. (1973).J . Cell. Physiol. 81,49. Weber, M. J. (1975).Cell 5,253. Weeks, C., and Weeks, G. (1975).E r p . Cell Res. 92,372. Wei, C . H., Koh, C., Pfuderer, P., and Einstein, J. R. (1975).J . Biol. Chem. 250,4790. Weiser, M . M. (1972).Science 177,525. Wenthold, R. J., Mahler, H. R., and Moore, W. J. (1974).J . Neurochem. 22,945. Whitney, R. B., and Sutherland, R. M. (1972).Cell. Immunol. 5, 137. Whitney, R. B., and Sutherland, R. M. (1973).Proc. Leucocyte Cult. Conf., 7th, 1972 pp. 63-74. Wickus, G., Gruenstein, E., Robbins, P. W., and Rich, A. (1975).Proc. Natl. Acad. Sci. U.S.A.72,746.

LECTINS

345

Wilhelms, O., Luderitz, O., and Westphal, 0. (1974).Eur. J. Biochem. 48,89. Willingham, M. C., and Pastin, I. (1975).Proc. Natl. Acad. Sci. U.S.A. 72, 1263. Winzler, R.J. (1969).In “Structure and Function of the Red Cell Membrane” (G. Jamieson and T. Greenwalt, eds.), pp. 157- 171.Lippincott, Philadelphia, Pennsylvania. Wolpert, J. S., and Albersheim, P. (1976).Biochem. Biophys. Res. Commun. 70,729. Wright, C. S. (1974).J. Mol. Biol. 87,835. Wright, C. S., Keith, C., Langridge, R., Nagata, Y., and Burger, M. (1974).j. Mol. Biol.

87,843.

Wright, C. S. (1977).J. Mol. Biol. 111,439. Wright, P., Quastel, M. R., and Kaplan, J. G. (1973).Proc. Levcocyte Cult. Conf., 7th, 1972 pp. 87- 104. Wyrick, R. E., Nishihara, T., and Hedrick, J. L. (1974).Proc. Natl. Acad. Sci. U S A . 71,

2067.

Yachnin, S.,and Svenson, R. (1972).Zmmunology 22,871. Yanagimachi, R., and Nicolson, G. L. (1976).E x p . Cell Res. 100,249. Yariv, J . , Kalb, A. J., and Levitzki, A. (1968).Biochim. Biophys. Acta 165,303. Yariv, J . , Kalb, A. J., and Blumberg, S. (1973).In “Methods in Enzymology” (V. Ginsburg, ed.) Vol. 28,Part B, pp. 356-365.Academic Press, New York. Yau, T. M., Buckman, T., Hale, A. H., and Weber, M.J. (1976).Biochemistry 15,3212. Yin, H. H.,Ukena, T. E., and Berlin, R. D. (1972).Science 178,867. Yoshinaga, M., Yoshinaga, A., and Waksman, B. H. (1972).Proc. Natl. Acad. Sci. U.S.A.

69,3251. Yurewicz, E. C., Oliphant, G., and Hedrick, J. L. (1975).Biochemistry 14,3101. Zalik, S. E., and Cook, G. M. W. (1976).Biochim. Biophys. Acta 419, 119. Zarling, J. M., and Tevethia, S. S. (1971).Virology 45,313.

This Page Intentionally Left Blank

Subject Index A

Complement host defense and, 129-131 Cyclic nucleotides macrophages and, 132-134 Cyrtocysts, 243-259

6-Aniinolevulinate formation of, 160-163 Antler development thyroid parafollicular cells and, 16-17

D Discobolocysts, 263-265

B

E

Blood groups lectins and, 278-280 B-l ymphocyte activity, factors affecting, 128-129

Eggs lectins and, 322-325 Ejectosomes, 260-263 Embryos lectins and, 322-325 Etiochloroplast me nib rane protein changes, 179-180 surface property changes, 177-178 structural changes, 174-177 Extrusive organelles definition, 197-198 historical background, 198 methods electron microscopy, 199 light microscopy, 199 preparation of material, 198-199 Extrusomes characterization and distribution, 199-201 fine structure, extrusion mechanism, function and origin discobolocysts, 263-265 ejectisomes, 260-263 mucocysts, 224-243 nematocysts, 265 polar filament of Cnidosporidians, 265-267 rhabdocysts, 259-260 spindle trichocysts, 202-224 toxicysts, 243-259 varia, 267

C Cell surface lectin receptors, biochemistry, 326-330 Chlorophyll biogenesis Gaminolevulinate, 160-163 chlorophyll a, 171-173 chlorophyll b, 173 protochlorophyllide, 163-165 spectroscopic studies, 165-171 Chlorophyll a formation of, 171-173 Chlorophyll b formation of, 173 Chlorophyll-protein complexes formation of, 180-183 Chloroplast development in green leaf, 174 Chloroplast membranes hiogenesis, 173-174 chlorophyll-protein complex formation, 180-183 chloroplast i n green leaf, 174 etiochloroplast structural changes, 174- 177 lipid biosynthesis, 183-184 protein changes, 179-180 surface property changes, 177-178 Clathrocysts, 224-243 Cnidocysts, 265 CnidosDoridians. polar filaments of, 265-267

F Fibroblasts substances regulating growth and activity, 121-124 351

352

SUBJECT INDEX H

Haptocysts, 243-259 Hibernators thyroid parafollicular cells, 16-17 Host defense factors activating or regulating B-lymphocyte activity, 128-129 complement, 129-131 PMN chemotactic agent, 132 pyrogen, 131-132 T-lymphocyte activity, 125-128 5-Hydroxytryptamine thyroid parafollicular cells and, 54-63 Hypophysectomy thyroglobulin biosynthesis and, 103-104 thyroid iodide pathways and, 99-100

I Interferon macrophages and, 124-125 Iodide thyroid pathways, 87-100

K Kinetocysts, 224-243

L Lectins biochemistry blood group specificity, 278-280 purification, 280-287 structure, 288-292 biological role, 333-335 cells infected with nononcogenic viruses and, 319-322 cell surface receptors, biochemistry, 326-330 mitogenesis and mechanism, 292-297 nature of response, 292 interaction with developing cells eggs and embryos, 322-325 male germinal cells, 325-326 toxicity, 330-333 transformed cell agglutinability and, 297-305 tumorigenicity and, 305

Lipid chloroplast, biosynthesis of, 183-184

M Macrophages antibacterial factors and, 125 cytotoxic substances and, 134-137 hydrolytic enzymes, 137-138 acid hydrolases, 138-146 consequences of enzyme release, 150-151 neutral proteinases, 146-150 interferon and, 124-125 prostaglandins or cyclic nucleotides and, 132-134 Male germinal cells lectins and, 325-326 Mitogenesis lectin-induced mechanism of, 292-297 nature of response, 292 Mucocysts, 224-243

N Nematocysts, 265

0 Osteopetrosis congenital, thyroid parafollicular cells and, 17-18

P Pexicysts, 243-259 Photochemical activity development, 185-186 other activities, 189 photochemical apparatus, 186-187 photosystems, 187-189 Photochemical apparatus background, 160 scope of article, 159 Polymorphonuclear cells chemotactic agent, 132 Prostaglandins macrophages and, 132-134 Protochloroph yllide formation of, 163-165 transformation to chlorophyll, 165-171

353

SUBJECT INDEX Pyrogen host defense and, 131-132

R Rhabdocysts, 259-260

5 Stem-cell growth regulation, 120-121

T Taeniobolocysts, 260-263 Thyroglobulin bios ynthesis effects of hypophysectomy, 103-104 relative measurement, 100-102 ultrastructural features, 102-103 Thyroid development movement toward colloid, 83 movement toward interfollicular space, 83-87 hypophyseal regulation, 104-105 bihormonal model, 112 production of two hormones, 112-113 response to TSH, 108-109 sensitivity to prolactin, 105-108 treatment with hormones, 109-112 iodide pathways effects of hypophysectomy, 99-100 iodide oxidation processes, 87-92 variations in colloid iodine content, 92-99 medullary carcinoma, 63-64 biochemical evidence, 64-65 electron microscopy, 65-66

histochemical evidence, 64 structure and function, 1-6 Thyroid parafollicular cells electron microscopy experimental studies, 53-54 normal tissues, 23-53 histochemical studies congenital osteopetrosis in mice, 17-18 distributiori and number, 12-14 experimental studies, 18-22 hibernators and antler development, 16-17 identificatiqn, 6-12 location and morphology, 12 origin, 15-16 5-hydroxytryptamine and, 54-63 nonmalignant human diseases and, 66-67 tryptophyl peptides and, 63 T-1ymphocyte activity, factors affecting, 125-128 Toxicysts, 243-259 Transformed cells selective agglutination lectin agglutinability and, 297-305 mechanism, 309-319 transient agglutinability of untransformed cells, 305-309 tumorigenicity and, 305 Trichocysts, spindle, 202-224 Tryptophyl peptides thyroid parafollicular cells and, 63 Tumorigenicity lectin agglutination and, 305

V Varia, 267

This Page Intentionally Left Blank

Contents of Previous Volumes Volume 1

Ascorbic Acid and Its Intracellular Localization, with Special Reference Some Historical Features in Cell Biology to Plants-J. CHAYEN -ARTHUR HUGHES Aspects of Bacteria as Cells and as OrNuclear Reproduction-C. LEONARD galliSnlS-sTUART MUDD AND EDWARD HUSKINS D. DELAMATER Enzymic Capacities and Their Relation Ion Secretion in Plants-J. F. SUTCLIFFE to Cell Nutrition in Animals-GEORGE Multienzyme Sequences in Soluble W. KIDDER Extracts-HENRY R. MAHLER The Application of Freezing and Drying The Nature and Specificity of the FeulTechniques in Cytology-L. G. E. gen Nucleal Reaction-M. A. LESSLER BELL Quantitative Histochemistry of PhosphaEnzymatic Processes in Cell Membrane taSeS-wILLIAM L. DOYLE ROSENBERG AND w. Penetration-TH. Alkaline Phosphatase of the NucleusWILBRANDT M. CHPVREMONT AND H. FIRKET Bacterial Cytology-K. A. BISSET Gustatory and Olfactory Epithelia-A. F. Protoplast Surface Enzymes and AbsorpBARADIAND G. H. BOURNE tion of Sugar-R. BROWN Growth and Differentiation of Explanted Reproduction of Bacteriophage-A. D. Tissues-P. J. GAILLARD HERSHEY Electron Microscopy of Tissue SectionsThe Folding and Unfolding of Protein A. J. DALTON Molecules as a Basis of Osmotic Work A Redox Pump for the Biological Per-R. J. GOLDACRE formance of Osmotic Work, and Its Nucleo-Cytoplasmic Relations in AmphibRelation to the Kinetics of Free Ion ian Development-G. FRANK-HAUSER Diffusion across Membranes-E. J. Structural Agents in Mitosis-M. M. CONWAY SWA" A Critical Survey of Current Approaches Factors Which Control the Staining of in Quantitative Histo- and CytochemTissue Sections with Acid and Basic istry-DAvD GLICK Dyes-MARcus SINGER Nucleo-cytoplasmic Relationships in the The Behavior of Spermatozoa in the Development of Acetabularia-J. HAMNeighborhood of Eggs-Lorn ROTHSMERLING CHILD

The Cytology of Mammalian Epidermis and Sebaceous Glands-WILLIAM MONTAGNA The Electron-Microscopic Investigation H. BFLETSCHof Tissue Sections-L.

AUTHOR INDEX-SUB

JECT INDEX

Volume 3

NEIDER

The Histochemistry GOMORI AUTHOR INDEX-SUB

Report of Conference of Tissue Culture Workers Held at Cooperstown, New J. HETHERINGTON York-D.

of

Esterases-G.

JECT INDEX

Volume 2 Quantitative Aspects of Nuclear NucleoSWIFT proteins-HEWSON

The Nutrition of Animal CeIb-CHARITY WAYMOUTH Caryometric Studies of Tissue CulturesOTTO BUCHER The Properties of Urethan Considered in Relation to Its Action on MitosisIVORCORNMAN

355

356

CONTENTS OF PREVIOUS VOLUMES

Composition and Structure of Giant Chromosomes-MAx ALFERT How Many Chromosomes in Mammalian Somatic Cells?-R. A. BEATTY The Significance of Enzyme Studies on Isolated Cell Nuclei-ALEXANDER L. DOUNCE The Use of Differential Centrifugation in the Study of Tissue EnzymesCHR. DE DUVEAND J. BERTHET Enzymatic Aspects of Embryonic Differentiation-TRYccvE GUSTAFSON Azo Dye Methods in Enzyme Histochemistry-A. G. EVERSONPEARSE Microscopic Studies in Living Mammals with Transparent Chamber Methods-Roy G. WILLIAMS The Mast Cell-G. ASBOE-HANSEN Elastic Tissue-EDwms W. DEMPSEY AND ALBERT I. LANSING The Composition of the Nerve Cell Studied with New Methods-SVENOLOEB R A T T C ~AND D HOLCERHYDEN

Evidence for a Redox Pump in the Active Transport of Cations-E. J. CONWAY AUTHOR INDEX-SUB

JECT INDEX

Volume 5

Histochemistry with Labeled Antibody -ALBERT H. COONS The Chemical Composition of the Bacterial Cell Wall-C. S. CUMMINS Theories of Enzyme Adaptation in Microorganisms-J. MANDELSTAM The Cytochondria of Cardiac and Skeletal M U S C ~ ~ - J O H W.N HARMON The Mitochondria of the NeuronWARRENANDREW The Results of Cytophotometry in the Study of the Deoxyribonucleic Acid (DNA) Content of the NucleusR. VENDRELYAND C. VENDRELY Protoplasniic Contractility in Relation to Gel Structure: Temperature-Pressure Experiments on Cytokinesis and Amoeboid Movement - DOUGLAS MARSLAND AUTHOR INDEX-SUBJECT INDEX Intracellular pH-PETER C. CALDWELL The Activity of Enzymes in Metabolism Volume 4 and Transport in the Red Cell-T. A. Cytochemical Micrurgy-M. J. KOPAC J. PRANKERD Uptake and Transfer of Macromolecules Amoebocytes-L. E. WAGGE by Cells with Special Reference to Problems of Fixation in Cytology, HisGrowth and Development-A. M. tology, and Histochemistry-M. WOLSCHECHTMAN MAN Bacterial Cytology-ALmD MARSHAK Cell Secretion: A Study of Pancreas and C. J. JUNQUEIRA Histochemistry of Bacteria-R. VENDRELY Salivary Glands-L. AND G. C. HIRSCH Recent Studies on Plant MitochondriaThe Acrosome Reaction-JEAN c. DAN DAVIDP. HACKETT Cytology of ~permatogenesis-vIs€iWA The Structure of Chloroplasts-K. NATH M~~HLETHALER The Ultrastructure of Cells, as Revealed Histochemistry of Nucleic Acids-N. B. by the Electron Microscope-FkmoF KURNICK S. SJOSTRAND Structure and Chemistry of NucleoliAUTHOR INDEX-SUB JECT INDEX W. S. VINCENT On Goblet Cells, Especially of the InVolume 6 testine of Some Mammalian SpeciesHARALDMOE Localization of Neuromuscular TEAUX

Cholinesterases at Junctions-R. Cou-

The Antigen System of Paramecium aurelia-G. H. BEALE The Chromosome Cytology of the Ascites Tumors of Rats, with Special Ref-

CONTENTS OF PREVIOUS VOLUMES

357

erence to the Concept of the Stemline The Structure and Innervation of Lamellibranch Muscle-J. BOWDEN Cell-sAJIRO MAKINO Hypothalamo-neurohypophysial NeuroThe Structure of the Golgi Apparatussecretion-J. C. SLOPER ARTHUR w. POLLISTER AND h I S C m A Cell Contact-PAUL WEIS F. POLLISTER An Analysis of the Process of Fertiliza- The Ergastoplasm: Its History, Ultrastructure, and Biochemistry-Mtion and Activation of the EggCOISE HAGUENAU A. MONROY The Role of the Electron Microscope in Anatomy of Kidney Tubules-JoHANNEs RHODIN Virus Research-ROBLEY c. WILLIAMS Structure and Innervation of the Inner The Histochemistry of PolysaccharidesEar Sensory Epithelia-Hms ENGARTHURJ. HALE STROM AND JAN WERSALL The Dynamic Cytology of the Thyroid The Isolation of Living Cells from Gland-J. GROSS Animal Tissues-L. M. RINALDINI Recent Histochemical Results of Studies on Embryos of Some Birds and Mam- AUTHOR INDEX-SUB JECT INDEX mals-ELI0 BORGHESE Carbohydrate Metabolism and Embryonic Volume 8 Determination-R. J. OCONNOR Enzymatic and Metabolic Studies on Isolated Nuclei&. SIEBERTAND R. M. S. The Structure of Cytoplasm-CHluRLEs OBERLING SMELLIE Wall Organization in Plant Cells-R. D. Recent Approaches of the Cytochemical PRESTON Study of Mammalian Tissues-GEORGE H. HOGEBOOM, EDWARD L. KUFF, AND Submicroscopic Morphology of the Synapse-humm DE ROBERTIS WALTERC. SCHNEIDER The Cell Surface of Paramecium-4. F. The Kinetics of the Penetration of NonEHRET AND E. L. POWERS electrolytes into the Mammalian ErythThe Mammalian Reticulocyte-LEAH rOCYte-fiEDA BOWYER MIRIAMLOWENSTEIN AUTHOR INDEX-SUB JECT INDEX The Physiology of ChromatophoresCUMULATIVE SUBJECT INDEX MILTONFINGERMAN (VOLUMES 1-5) The Fibrous Components of Connective Tissue with Special Reference to the Elastic Fiber-DAVID A. HALL Volume 7 Experimental Heterotopic OssificationJ. B. BRIDGES Some Biological Aspects of Experimental A Survey of Metabolic Studies on IsoRadiology: A Historical Review-F. G. lated Mammalian Nuclei-D. B. SPEAR ROODYN The Effect of Carcinogens, Hormones, Trace Elements in Cellular Functionand Vitamins on Organ Cultures-ILsE L. BERT L. VALLEE AND FREDERIC LASNITZKI HOCH Recent Advances in the Study of the Osmotic Properties of Living CellsKinetochore-A. LIMA-DE-FARIA D. A. T. DICK Autoradiographic Studies with S"-Sulfate Sodium and Potassium Movements in -D. D. DZIEWIATKOWSKI Nerve, Muscle, and Red Cells-1. M. The Structure of the Mammalian SperGLYNN matozoon-DoN W. FAWCETT Pinocytosis-H. HOLTER The L y m p h o c y t 4 . A. TROWELL AUTHOR INDEX-SUB JECT INDEX

3 58

CONTENTS OF PREVIOUS VOLUMES

Volume 11

Volume 9

The Influence of Cultural Conditions on Electron Microscopic Analysis of the F. WILKINSON Secretion Mechanism-K. KUROSUMI Bacterial Cytology-J. The Fine Structure of Insect Sense AND J. P. DUGUID OrganS-ELEANOR H. SLIFER Organizational Patterns within Chromosomes-BERWIND P. KAUFMANN, Cytology of the Developing EyeALFRED J. COULOMBRE HELEN GAY, AND MARGARETR. The Photoreceptor Structures-J. J. MCDONALD WOLKEN Enzymic Processes in Celk-JAY BOYD Use of Inhibiting Agents in Studies on BEST Fertilization Mechanisms-CHARLES B. The Adhesion of Ceh-LEONARD WEISS METZ Physiological and Pathological Changes The Growth-Duplication Cycle of the in Mitochondria1 Morphology-CH. Cell-D. M. PRESCOTT ROUILLER The Study of Drug Effects at the Cy- Histochemistry of Ossification-RoMuLo L. CABRINI B. WILSON tological Level-G. Cinematography, Indispensable Tool for Histochemistry of Lipids in OogenesisCytology-C. M. POMERAT VISHWANATH Cyto-Embryology of Echinoderms and Amphibia-KuTsuMA DAN The Cytochemistry of Nonenzyme ProteinS-RONALD R. COWDEN

AUTHOR INDEX-SUB

SECT INDEX

Volume 12

Sex Chromatin and Human ChromoL. HAMERTON somes-IoHN Chromosomal Evolution in Cell Populations-T. C. Hsu Volume 10 Chromosome Structure with Special Reference to the Role of Metal IonsThe Chemistry of Shiff's ReagentDALE M. STEFFENSEN FREDERICK H. KASTEN Electron Microscopy of Human White Spontaneous and Chemically Induced Blood Cells and Their Stem CellsKUMAR Chromosome Breaks-ARuN MARCEL BESSISAND JEAN-PAULTHIERY SHARMAAND ARCHANASHARMA In Vivo Implantation as a Technique in The Ultrastructure of the Nucleus and Skeletal B~o~o~Y-WILLIAM J. L. Nucleocytoplasmic Relations-SAUL FELTS WISCHNITZER The Nature and Stability of Nerve The Mechanics and Mechanism of CleavMyelin-J. B. FINEAN age-LEwIs WOLPERT Fertilization of Mammalian Eggs in The Growth of the Liver with Special Vitro-C. R. AUSTIN Reference to Mammals-F. DOLJANSKI Physiology of Fertilization in Fish Eggs Cytology Studies on the Affinity of the -Tom-o YAMAMOTO Carcinogenic Azo Dyes for Cyto- AUTHOR INDEX-SUB JECT INDEX plasmic COmpOnentS-YOSHIMA NAGAAUTHOR INDEX-SUB

JECT INDEX

TAN1

Epidermal Cells in Culture-A. MATOLTSY AUTHOR INDEX-SUB

JECT INDEX

CUMULATIVE SUBJECT INDEX

(VOLUMES

1-9)

GEDEON Volume 13 The Coding

Hypothesis-MARTYNAS

YEAS

Chromosome Reproduction-J. TAYLOR

HERBERT

359

CONTENTS OF PREVIOUS VOLUMES

Sequential Gene Action, Protein Synthesis, and Cellular DifferentiationREED A. FLICKINCER The Composition of the Mitochondria] Membrane in Relation to Its Structure and Function-ERIC G. BALL AND CLIFFE D. JOEL Pathways of Metabolism in Nucleate A. and Anucleate Erythrocytes-H. SCHWEICER Some Recent Developments in the Field of Alkali Cation Transport-W. WILBRANDT

Chromosome Aberrations Induced by Ionizing Radiations-H. J. EVANS Cytochemistry of Protozoa, with Particular Reference to the Golgi Apparatus and the MitochondriaVISHWANATH AND G. P. DUTTA AND Cell Renewal-FELIX BERTALANFFY CHOSENLAu AUTHOR INDEX-SUBJECT

Volume 14

INDEX

The Tissue Mast Wall-DoucLAs SMITH AUTHOR INDEX-SUB

E.

JECT INDEX

Volume 15 The Nature of Lampbrush Chromosomes -H. G. CALLAN The Intracellular Transfer of Genetic Information-J. L. SIRLIN Mechanisms of Gametic Approach in PhtS-LEONARD MACHLISAND ERIKA RAWITSCHER-KUNKEL The Cellular Basis of Morphogenesis and Sea Urchin Development-T. GUSTAFSON AND L. WOLPERT Plant Tissue Culture in Relation to Development Cytology-CAm R. PARTANEN

Regeneration of Mammalian LiverNANCYL. R. BUCHER Collagen Formation and Fibrogenesis with Special Reference to the Role of Ascorbic Acid-BEmAm S. COULD The Behavior of Mast Cells in Anaphylaxis-IVAN MOTA Lipid Absorption-ROBERT M. WOTTON

Inhibition of Cell Division: A Critical AUTHOR INDEX-SUB JECT INDEX and Experimental AnalySiS-sEYMOUR GELFANT Electron Microscopy of Plant Protoplasm Volume 16 -R. BUVAT Cytophysiology and Cytochemistry of the Ribosomal Functions Related to Protein Synthesis-Tom HULTIN Organ of Corti: A Cytochemical A. VINNIKOV Physiology and Cytology of Chloroplast Theory of Hearing-J. Formation and “Loss” in EugknuAND L. K. TITOVA M. GRENSON Connective Tissue and Serum ProteinsCell Structures and Their Significance R. E. MANCINI for Ameboid Movement-K. E. WOHLThe Biology and Chemistry of the Cell FARTH-BOTTERMAN Walls of Higher Plants, Algae, and Microbeam and Partial Cell Irradiation Fungi-D. H. NORTHCOTE -C. L. SMITH Development of Drug Resistance by Nuclear-Cytoplasmic Interaction with Staphylococci in Vitro and in VivoIonizing Radiation-M. A. LESSLER MARYBARBER Cytological and Cytochemical Effects of I n Vioo Studies of Myelinated Nerve Fibers-CARL CASKEYSPEIDEL Agents Implicated in Various Pathological Conditions: The Effect of Respiratory Tissue: Structure, Histophysiology, Cytodynamics. Part I: Viruses and of Cigarette Smoke on Review and Basic Cytomorpholomthe Cell and Its Nucleic Acid-CEcnm FELIXD. BERTALANFFY AND RUWLF LEUCHLEUCHTENBERCER TENBER GER

AUTHOR INDEX-SUB

JECT INDEX

360

CONTENTS OF PREVIOUS VOLUMES

Volume 19

Volume 17

The Growth of Plant Cell Walls-K. “Metabolic” DNA: A Cytochemical WILSON Study-H. ROELS Reproduction and Heredity in Trypano- The Significance of the Sex Chromatinsomes: A Critical Review Dealing MURRAY L. BARR Mainly with the African Species in Some Functions of the Nucleus-J. M. J. WALKER MITCHISON the Mammalian Host-P. The Blood Platelet: Electron Microscopic Synaptic Morphology on the Normal and Studies-J. F. DAVID-FERREIRA Degenerating Nervous System-E. G. The Histochemistry of MucopolysacchaGRAYAND R. W. GUILLERY rides-ROBERT c. CURRAN Neurosecretion-W. BARGMANN Respiratory Tissue Structure, Histophysiology, Cytodynamics. Part 11. Some Aspects of Muscle RegenerationE. H. BETZ, H. FmKET, AND M. New Approaches and Interpretations REZNIK -FELIX D. BERTALANFFY W. The Cells of the Adenohypophysis and The Gibberellins as Hormones-P. BRIAN Their Functional Significance-MARC Phototaxis in PlantS-WOLFGANG HAUPT HERLANT Phosphorus Metabolism in Plants-K. S. AUTHOR INDEX-SUBJECT INDEX ROWAN AUTHOR INDEX-SUBJECT

INDEX

Volume 18 The Cell of Langerhans-A.

S. BREATH-

NACH

The Structure of the Mammalian EggROBERTHADEK Cytoplasmic Inclusions in OogenesisM. D. L. SRIVASTAVA The Classification and Partial Tabulation of Enzyme Studies on Subcellular Fractions Isolated by Differential Centrifuging-D. B. ROODYN Histochemical Localization of Enzyme Activities by Substrate Film Methods: Ribonucleases, Deoxyribonucleases, Proteases, Amylase, and Hyaluronidase -R. DAOUST Cytoplasmic Deoxyribonucleic Acid-P. B. GAHANAND J. CWYEN Malignant Transformation of Cells in V~~TO-KATHERINE K. SANFORD Deuterium Isotope Effects in CytologyS . BOSE, H. I. E. FLAUMENHAFT, CRESPI,AND J. J. KATZ The Use of Heavy Metal Salts as Electron Stains-C. RICHARDZOBEL AND MICHAEL BEW AUTHOR INDEX-SUB

JECT INDEX

Volume 20 The Chemical Organization of the Plasma Membrane of Animal Cells-A. H. MADDY Subunits of Chloroplast Structure and Quantum Conversion in Photosynthesis-RoDERIc B. PARK Control of Chloroplast Structure by Light -LESTER PACKERAND PAUL-AND& SIECENTHALER The Role of Potassium and Sodium Ions as Studied in Mammalian BrainH. HILLMAN Triggering of Ovulation by Coitus in the Rat-CLAUDE h O N , GITTAASCH, AND JAQUELINE Roos Cytology and Cytophysiology of NonMelanophore Pigment Cells-JOSEPH T. BACNARA The Fine Structure and Histochemistry of Prostatic Glands in Relation to Sex BRANDES Hormones-DAvm Cerebellar Enzymology-Lucm ARVY AUTHOR INDEX-SUBJECT

INDEX

CONTENTS OF PREVIOUS VOLUMES

361

Volume 23

Volume 21

Histochemistry of Lysosomes-P. B. Transforniationlike Phenomena in Somatic Cells-J. M. OLENOV GAHAN Physiological Clocks-R. L. BRAHM- Recent Developments in the Theory of Control and Regulation of Cellular ACHARY PrOCeSSeS-ROBERT ROSEN Ciliary Movement and Coordination in Ciliates-BELA PARDUCA Contractile Properties of Protein Threads from Sea Urchin Eggs in Relation to Electromyography: Its Structural and SAKAI Cell Division-HIKorcHr Neural Basis-JoHN V. BASMAJIAN Cytochemical Studies with Acridine Electron Microscopic Morphology of Oogenesis-ARNE N~RREVANC Orange and the Influence of Dye Contaminants in the Staining of Dynamic Aspects of Phospholipids during Nucleic Acids-FmDERICK H. KASTEN Protein Secretion-LOWELL E. HOKIN Experimental Cytology of the Shoot The Golgi Apparatus: Structure and Apical Cells during Vegetative Function-H. W. BEAMSAND R. G. Growth and Flowering-A. NouKESSEL GAR~DE The Chromosomal Basis of Sex DeterNature and Origin of Perisynaptic Cells nhatiOn-KENNETH R. LEWIS AND of the Motor End Plate-T. R. SHANBERNARD JOHN THAVEERAPPA AND G. H. BOURNE AUTHOR INDEX-SUB JECT INDEX AUTHOR INDEX-SUB

JECT INDEX

Volume 24 Volume 22 Synchronous Cell DifferentiationCurrent Techniques in Biomedical ElecGEORGE M. PADILLAAND IVANL. tron Microscopy-SAUL WISCHNITZER CAMERON The Cellular Morphology of Tissue Re- Mast Cells in the Nervous SystemYNCVE OLSON pair-R. M. H. MCMINN Structural Organization and Embryonic Development Phases in Intermitosis and Differentiation-GA JANAN V. SHERBET the Preparation for Mitosis of MamAND M. S. LAKSHMI A. malian cells in VitrO-BLAGOJE NEBKOVI~ The Dynamism of Cell Division during Antimitotic Substances-Guy DEYSSON Early Cleavage Stages of the EggN. FAUTREZ-FIRLEFYN AND J. FAUTREZ The form and Function of the Sieve Lymphopoiesis in the Thymus and Other Tube: A Problem in ReconciliationP. E. WEATHERLEYAND R. P. C. Tissues: Functional Implications-N. B. EVERETT AND RUTH W. TYLER JOIXNSON ( CAFFREY ) Analysis of Antibody Staining Patterns Obtained with Striated Myofibrils in Structure and Organization of the MyoFluorescence Microscopy and Electron neural Junction-C. COERS Microscopy-FRANK A. PEPE The Ecdysial Glands of ArthropodsWILLIAMS. HERMAN Cytology of Intestinal Epithelial CellsCytokinins in Plants-B. I. SAHAISRIVAS- PETER G . TONER TAVA Liquid Junction Potentials and Their Effects on Potential Measurements in AUTHOR INDEX-SUB JECT INDEX Biology Systems-P. C. CALDWELL CUMULATIVE SUBJECT INDEX ( VOLUMES 1-21 )

AUTHOR INDEX-SUB

JECT INDEX

362

CONTENTS O F PREVIOUS VOLUMES

Volume 25

Volume 27

Cytoplasmic Control over the Nuclear Events of Cell Reproduction-NOEL DE TERRA Coordination of the Rhythm of Beat in Some Ciliary Systems-M. A. SLEIGH The Significance of the Structural and Functional Similarities of Bacteria and Mitochondria-SYLVAN NASS The Effects of Steroid Hormones on Macrophage Activity-B. VERNONROBERTS The Fine Structure of Malaria Parasites -MARIA A. RUDZINSKA The Growth of Liver Parenchymal Nuclei and Its Endocrine Regulation -RITA CARRIERE Strandedness of Chromosomes-SHELDON WOLFF Isozymes: Classification, Frequency, and Significance-CmRLEs R. SHAW The Enzymes of the Embryonic Nephron -LUCIE ARVY Protein Metabolism in Nerve Cells-B. DROZ Freeze-Etching-HANS MOOR

Wound-Healing in Higher PlantsJACQUES LIPETZ Chloroplasts as Symbiotic OrganellesDENNISL. TAYLOR The Annulate Lamellae-SAUL WISCH-

AUTHOR INDEX-SUB

Volume 26

JECT INDEX

NITZER

Gametogenesis and Egg Fertilization in Planarians-G. BENAZZI LENTATI Ultrastructure of the Mammalian Adrenal Cortex-SIMON IDELMAN The Fine Structure of the Mammalian Lymphoreticular System-Im CARR Immunoenzyme Technique: Enzymes as Markers for the Localization of Antigens and Antibodies-SmiATIs AVRAMEAS AUTHOR INDEX-SUB

JECT INDEX

Volume 28 The Cortical and Subcortical Cytoplasm Of LymnUea Egg-CHRISTIAAN P. RAVEN The Environment and Function of InE. vertebrate Nerve Cells-J. THEHERNEAND R. B. MORETON Virus Uptake, Cell Wall Regeneration, and Virus Multiplication in Isolated Plant Protoplasts-E. C. COCKING The Meiotic Behavior of the DrosophiZu OOCYte-ROBERT c. KING The Nucleus: Action of Chemical and Physical Agents-RENd S I ~ ~ R D The Origin of Bone CeIIS-MAUREEN OWEN Regeneration and Differentiation of sieve Tube Elements-wILLIAM p.

A New for the Living A Summary Of the and Recent Experimental Evidence in Its Support -GILBERT N. LING The Cell PeriPherY-LEONARD Mitochondria] DNA: Physicocheniical Properties, Replication, and Genetic jAcoBs Function-€'. BORSTAND A. M. KROON Cells, Solutes, and Growth: Salt AcMetabolism and Enucleated Cells-KoNcumulation in Plants ReexaminedRAD KECK F. C. STEWARDAND R. L. MOTT Stereological Principles for Morphometry AUTHOR INDEX-SUB JECT INDEX in Electron Microscopic CytologyEWALDR. WEIBEL Some Possible Roles for Isozyniic Substi- Volume 29 tutions during Cold Hardening in Gram Staining and Its Molecular MechPlants-D. W. A. ROBERTS anism-B. B. BISWAS,P. s. BASU,AND AUTHOR INDEX-SUB JECT INDEX M. K. PAL

363

CONTENTS OF PREVIOUS VOLUMES The Surface Coats of Animal Cells-A. MARTiNEZ-PALOMO Carbohydrates in Cell Surfaces-RICHARD J. WINZLER Differential Gene Activation in Isolated Chromosomes-MmKus LEZZI Intraribosomal Environment of the Nascent Peptide Chain-HmEKo KAJI Location and Measurement of Enzymes in Single Cells by Isotopic Methods Part I-E. A. BARNARD Location and Measurement of Enzymes in Single Cells by Isotopic Methods Part 11-G. C. BUDD Neuronal and Glial Perikarya Preparations: An Appraisal of Present Methods -PATRICIA V. JOHNSTONAND BETTY I. ROOTS Functional Electron Microscopy of the Hypothalamic Median EminenceTOKUZO MATSUI, HIDESHIKOBAYASHI, AND SUSUMIISHII Early Development in Callus CulturesMICHAELM. YEOMAN

Morphological and Histochemical Aspects of Glycoproteins at the Surface of Animal Cells-A. RAMBOURC DNA Biosynthesis-H. S. JANSZ,D. VAN DER MEI, AND G. M. ZANDVLIET Cytokinesis in Animal Cells-R. RAPPAPORT

The Control of Cell Division in Ocular Lens-C. V. HARDING,J. R. REDDAN, N. J. UNAKAR,AND M. BAGCHI The Cytokinins-HANS KENDE Cytophysiology of the Teleost Pituitary -MARTIN SAGE AND HOWARD A. BERN AUTHOR INDEX-SUB

JECT INDEX

Volume 32

Highly Repetitive Sequences of DNA in Chromosomes-W. G. FLAMM The Origin of the Wide Species Variation in Nuclear DNA Content-H. REES AND R. N. JONES Polarized Intracellular Particle Transport: AUTHOR INDEX-SUB JECT INDEX Saltatory Movements and Cytoplasmic Streaming-LIONEL I. REBHUN Volume 30 The Kinetoplast of the HemoflagellatesLARRYSIMPSON High-pressure Studies in Cell BiologyTransport across the Intestinal Mucosal ARTHUR M. ZIMMERMAN S. Cell: Hierarchies of Function-D. Micrurgical Studies with Large FreePARSONSAND C. A. R. BOYD Living Amebas-K. W. JEON AND Wound Healing and Regeneration in the J. F. DANIELLI Crab Paratelphusa hydrodromousThe Practice and Application of Electron RITA G. ADIYODI Microscope Autoradiography-J. JACOB The Use of Ferritin-Conjugated AntiScanning Electron Applications of bodies in Electron MicroscopyMicroscopy in Biology-K. E. CARR COUNCILMAN MORGAN Acid Mucopolysaccharides in Calcified Metabolic DNA in Ciliated Protozoa, Tissues-SHIN JIRO KOBAYASHI Salivary Gland Chromosomes, and AUTHOR INDEX-SUB JECT INDEX Mammalian Cells-S. R. PELC CUMULATIVE

(VOLUMES

SUBJECT INDEX

1-29)

AUTHOR INDEX-SUB

JECT INDEX

Volume 31

Volume 33

Studies on Freeze-Etching of Cell Membranes-KuRT M~HLETHALER Recent Developments in Light and Electron Microscope Radioautography -C. C. BUDD

Visualization of RNA Synthesis on Chromosomes-0. L. MILLER,JR. AND BARBARA A. HAMKALO Cell Disjunction ( “Mitosis”) in Somatic Cell Reproduction-ELAINE G. DIA-

364

CONTENTS OF PREVIOUS VOLUMES

SCOTT HOLLAND, AND PAULINE PECOM Neuronal Microtubles, Neurofilaments, and Microfilaments-RAYMOND B. WUERKERAND JOEL B. KIRKPATRICK Lymphocyte Interactions in Antibody Responses-J. F. A. P. MILLER Laser Microbeams for Partial Cell Irw. BERNS AND radiation-MIcmEL CHRISTIAN SALET Mechanisms of Virus-Induced Cell Fusion-GEORGE POSTE Freeze-Etching of Bacteria-Cmms C. REMSENAND STANLEYW. WATSON The Cytophysiology of Mammalian Adipose Celk-BERNARD G. SLAVIN CUMAKOS,

AUTHOR INDEX-SUB

JECT INDEX

Synthetic Activity of Polytene Chromosomes-HANS D. BERENDES Mechanisms of Chromosome Synapsis at Meiotic Prophase-PETER B. MOENS Structural Aspects of Ribosomes-N. NANNINGA Comparative Ultrastructure of the Cerebrospinal Fluid-Contacting NeuronsB. VIGH AND I. VIGH-TEICHMANN Maturation-Inducing Substance in Starfishes-Hmuo KANATANI The Limonium Salt Gland: A Biophysical and Structural Study-A. E. HILL AND B. S . HILL Toxic Oxygen Effects-Hmom M. SWARTZ AUTHOR INDEX-SUB

JECT INDEX

Volume 36

Volume 34

Molecular Hybridization of DNA and RNA h SitZGWOLFGANG HENNIG The Relationship between the PlasmaNITZER lemma and Plant Cell Wall-JEANThe Energy State and Structure of IsoCLAUDEROLAND lated Chloroplasts: The Oxidative Recent Advances in the Cytochemistry Reactions Involving the Water-Splitand Ultrastructure of Cytoplasmic ting Step of Photosynthesis-ROBERT Inclusions in Mastigophora and L. HEATH Opalinata (Protozoa)-G. P. DUTTA A. Transport in Neurospora-GENE Chloroplasts and Algae as Symbionts in SCARBOROUGH MO~~USCS-LEONARD MUSCATINEAND Mechanisms of Ion Transport through RICHARDW. GREENE Plant cell Membranes-EMmum The Macrophage-SAIMoN GORDONAND ERSTEIN ZANVIL A. C o r n Cell Motility: Mechanisms in ProtoDegeneration and Regeneration of Neuroplasmic Streaming and Ameboid secretory Systems-Honsr-DmTER Movement-H. KOMNICK,W. STOCDELLMANN KEM, AND K. E. WOHLEFARTHAUTHOR INDEX-SUB JECT INDEX BOTTERMANN The Gliointerstitial System of MoIIuscsGHISLAINNICAISE Volume 37 Colchicine-Sensitive Microtubles-Lm MARGULIS Units of DNA Replication in ChromoThe Submicroscopic Morphology of the Interphase Nucleus-SAUL WISCH-

AUTHOR INDEX-SUB

JECT INDEX

Volume 35 The Structure of Mammalian Chromosomes-ELTON STUBBLEFIELD

HERBERT somes of Eukaroytes-J. TAYLOR Viruses and Evolution-D. C. REANNEY Electron Microscope Studies on Spermiogenesis in Various Animal SpeciesGONPACHIRO YASVZVMI

Morphology, Histochemistry, and Bio-

CONTENTS O F PREVIOUS VOLUMES chemistry of Human Oogenesis and S. GURAYA Ovulation-SmuL Functional Morphology of the Distal Lung-KAYE H. KILBURN Comparative Studies of the Juxtaglomerular Apparatus-HIRomiw S O W E AND MIZUHOOCAWA The Ultrastructure of the Local Cellular Reaction to Neoplasia-Im C m AND J. C. E . UNDERWOOD Scanning Electron Microscopy in the Ultrastructural Analysis of the Mammalian Cerebral Ventricular SystemD. E. SCOTT, G. P. KOZLOWSKI, AND M. N. SHERIDAN AUTHOR INDEX-SUB

JECT INDEX

Volume 38

365

Nucleocytoplasmic Interactions in Development of Amphibian HybridsSTEPHENSUBTELNY The Interactions of Lectins with Animal Cell Surfaces-GARTH L. NICOLSON Structure and Function of Intercellular Junctions-L. ANDREW STAEHELIN Recent Advances in Cytochemistry and Ultrastructure of Cytoplasmic Inclusions in Ciliophora (Protozoa)-G. P. DUTTA Structure and Development of the Renal Glomerulus as Revealed by Scanning SmElectron Microscopy-FRANC0 NELLI

Recent Progress with Laser Microbeams -MICHAEL W. BERNS The Problem of Germ Cell Determinants -H. W. BEAMSAND R. G. KESSEL SUBJECT INDEX

Genetic Engineering and Life Synthesis: An Introduction to the Review by R. Widdus and C. Ault-JAms F. DANIELLI Progress in Research Related to Genetic Engineering and Life Synthesis-Roy WIDDUSAND CHARLESR. AULT The Genetics of C-Type RNA Tumor Viruses-J. A. WYKE Three-Dimensional Reconstruction from Projections: A Review of AlgorithmsRICHARD GORDON AND GABOR T. HERMAN The Cytophysiology of Thyroid CellsVLADIMIRR. PAN TI^ The Mechanisms of Neural Tube Formation-PEmw KARFUNKEL The Behavior of the XY Pair in MammPIS-ALBERTO J. SOLAFU Fine-Structural Aspects of Morphogenesis in Acetabularia-G. WERZ Cell Separation by Gradient Centrifugation-R. HARWOOD SUBJECT INDEX

Volume 39 Androgen Receptors in the Nonhistone Protein Fractions of Prostatic Chromatin-TUNG YUE WANGAND LEROY M. NYBERG

Volume 40 B-Chromosome Systems in Flowering Plants and Animal Species-R. N. JONES The Intracellular Neutral SH-Dependent Protease Associated with Inflammatory Reactions- HIDEOHAYASHI The Specificity of Pituitary Cells and Regulation of Their Activities - VLADIMIR R. PANTIC Fine Structure of the Thyroid GlandHISAO FUJITA Postnatal Gliogenesis in the Mammalian Brain- A. PRIVAT Three-Dimensional Reconstruction from Serial Sections-RANDLE W. WARE AND VINCENT LOPREST1 SUBJECT INDEX

Volume 41

The Attachment of the Bacterial Chromosome to the Cell Membrane-PAUL J. LEIBOWITZAND MOSELIO SCHAECHTER

Regulation of the Lactose Operon in Escherichia coli by CAMP-G. CARPENTER AND B. H. SELLS Regulation of Microtubules in Tetrahymena - NORMAN E. WILLIAMS

366

CONTENTS OF PREVIOUS VOLUMES

The Evolution of the Mitotic SpindleDONNAF. KUBAI Germ Plasma and the Differentiation of SUNG LlAO the Germ Cell Line-E. M. EDDY A Cell Culture Approach to the Study of Anterior Pituitary Cells- A. TIXIER- Gene Expression in Cultured Mammalian P. COX AND JAMESc . Cells-RODY VIDAL,D. GOURDJI,AND C. TOUGARD Imnirinoliistoclieinical Demonstration of KING Morphology and Cytology of the AccesNeurophysin in the Hypothalamoneusory Sex Glands in Invertebratesrohypophysial System- W. B. WATIUNS K. G. ADIYODI AND R. G. ADIYODI The Visual System of the Horseshoe SUBJECT INDEX Crab Limulus polyplaemus- WOLF H. FAHRENBACH Volume 44 SUBJECT INDEX The Nucleolar Structure- SIBDASGHOSH The Function of the Nucleolus in the Volume 42 Expression of Genetic Information: Studies with Hybrid Animal CellsRegulators of Cell Division: Endogenous E. SIDEBOTTOM AND I. I. DEAK Mitotic Inhibitors of Mammalian Phylogenetic Diversity of the Proteins Cells - BISMARCK B. LOZZIO,CARMEN Regulating Muscular Contraction B. LOZZIO,ELENAG . BAMBERGER, AND WILLIAMLEHMAN STEPHENV. LAIR Cell Size and Nuclear DNA Content in Ultrastructure of Mammalian ChromoVertebrates -HENRYKSZARSKI some Aberrations - B. R. BRINKLEYAND Ultrastructural Localization of DNA in WALTERN. HITTELMAN Ultrathin Tissue Sections - ALAIN Computer Processing of Electron MicroGAUTIER graphs: A Nonmathematical AccountCytological Basis for Permanent Vaginal P. W. HAWKES Changes in Mice Treated Neonatally Cyclic Changes in the Fine Structure of with Steroid Hormones-NOBORU the Epithelial Cells of Human EndoTAKASUGI metrium - MILDRED GORDON On the Morphogenesis of the Cell Wall The Ultrastructure of the Organ of of Staphylococci - PETERGIESBRECHT, C ~ ~ ~ ~ - R O BS.EKIMURA RT JORGWECKE,AND BERNHARDREINICKE Endocrine Cells of the Gastric MucosaENRICO SOLCIA, CARLO CAPELLA, Cyclic AMP and Cell Behavior in Cultured Cells - MARK C. WILLINGHAM GABRIELE VASSALLO, AND ROBERTO Recent Advances in the Morphology, BUFFA Histochemistry, and Biochemistry of Membrane Transport of Purine and Steroid-Synthesizing Cellular Sites in Pyrimidine Bases and Nucleosides in the Nonmammalian Vertebrate OvaryAnimal CellS-fiCHARD D. BERLIN SARDUL S. GURAYA AND JANETM. OLIVER Cellular Receptors and Mechanisms of Action of Steroid Hormones- SHUT-

SUBJECT INDEX

SUBJECT INDEX

Volume 45 Volume 43 The Evolutionary Origin of the Mitochondrion: A Nonsymbiotic ModelHENRYR. MAHLER AND RUDOLF A. RAFF Biochemical Studies of Mitochondria1 Transcription and Translation-C. SACCONE AND E. QUAGLIARIELLO

Approaches to the Analysis of Fidelity of DNA Repair in Mammalian CellsMICHAEL W. LIEBERMAN The Variable Condition of Euchromatin and Heterochromatin - FRIEDRICH BACK Effects of 5-Bromodeoxyuridine on Tumorigenicity, Immunogenicity,

CONTENTS OF PREVIOUS VOLUMES Virus Production, Plasminogen Activator, and Melanogenesis of Mouse Melanoma Cells- SELMASILAGI Mitosis in Fungi- klELVIN s. FULLEII Small Lymphocyte and Transitional Cell Populations of the Bone Marrow; Their Role in the Mediation of Immune and Hemopoietic Progenitor Cell Functions- CORNELIUS ROSE The Structure and Properties of the Cell Surfiace Coat- J. H. LUFT Uptake and Transport Activity of the Median Eminence of the Hypothalamus- K. M. KNIGGE,S. A. JOSEPH,J. R. SLADEK,M. F. NOTTER, M. MORRIS, D. K. SUNDBEHG, M. A. HOLZWAHTH, G . E. HOFFMAN,AND L. O’BRIEN

367

Chemical Nature and Systematization of Substances Regulating Animal Tissue Growth-VICTOH A. KONYSHEV Structure and Function of the Choroid Plexus and Other Sites of Cerebrospinal Fluid Formation-THOMAS H. MILHOKAT The Control of Gene Expression in Somatic Cell Hybrids-H. P. BERNHARD Precursor Cells of MechanocytesALEXANDER J. FRIEDENSTEIN SUBJECT INDEX

Volume 48

Mechanisms of Chromatin Activation and Repression- NORMANRIACLEANANII VAUCIIAN A. IIILUEH Origin and Ultrastructure of Cells i n Vitro - L. h1. FRANKS ANI) PATKICIA 1). Volume 46 WILSON Neurosecretion by Exocytosis - TOM Electrophysiology of the Neurosecretory Cell-hNJI YACI A N D SIIIZUWO CHRISTIANNORMA” IWASAKI Genetic and Morphogenetic Factors in Reparative Processes . in Mammalian Hemoglobin Synthesis during Higher Wound Healing: The Role of ContractVertebrate Development: An Approach ile Phenomena- CIULIO GAHRIANI to Cell Differentiation MechanismsVICTOH NIGON AND JACQUELINE AND DENYS XIONTANDON Smooth Endoplasmic Reticulum in Rat GODET Hepatocytes during Glycogen DeposiCytophysiology of Corpuscles of Stannius tion and Depletion - ROHEHT R. -V. G . KKISHNAMUHTHY CAHDELL,J K . Ultrastructure of Human Bone Marrow Cell Maturation- J. BRETON-GOHIUS Potential and Limitations of Enzyme Cytochemistry: Studies of the IntraAND F. REYES cellular Digestive Apparatus of Cells Evolution and Function of Calciumin Tissue Culture-M. HUNUGEN Binding Proteins- R. H. KKETSINCEH Uptake of Foreign Genetic Material by SUBJECT INDEX Plant Protoplasts- E. C. COCKING The Bursa of Fabricius and Immunoglobulin Synthesis - HRUCE CLICK Volume 47 SUBJECT INDEX

SUBJECT INDEX

Responses of Mammary Cells to Hormones-M. R. BANEHJEE Recent Advances in the Morphology, Histochemistry, and Biochemistry of Steroid-Synthesizing Cellular Sites in the Testes of Nonmammalian Vertebrates-SARDUL S. GURAYA Epithelial-Stromal Interactions in Development of the Urogenital TractGERALDR. CUNHA

Volume 49 Cyclic Nucleotides, Calcium, and Cell Division-LIONEL I . REBHUN Spontaneous and Induced Sister Chroniatid Exchanges as Revealedby the BUdRLabeling Method-HATAO KATO Stnictnral, Electrophysiological, Bio-

368

CONTENTS O F PREVIOUS VOLUMES

chemical, and Pharmacological Properties of Neuroblastoma-Glioma Cell Hybrids in Cell Culture-B. HAMPRECHT Cellular Dynamics in Invertebrate Neurosecretory Systems-ALLAN BERLIND Cytophysiology of the Avian Adrenal Medulla-ASOK GHOSH Chloride Cells and Chloride Epithelia of Aquatic Insects-H. KOMNICK Cytosomes (Yellow Pigment Granules) of Molluscs as Cell Organelles of Anoxic Energy Production--IMM ZS.-NAGY

Action of Testosterone on the Differentiation and Secretory Activity of a Target Organ: The Submaxillary Gland of the Mouse-MONIQUE CHRETIEN SUBJECT INDEX

Volume 51

Circulating Nucleic Acids in Higher Organisms-MAURICE STROUN, PHILIPPE ANKER, PIERRE MAURICE,AND PETERB. GAHAN Recent Advances in the Morphology, HisSUBJECT INDEX tochemistry, and Biochemistry of the Developing Mammalian OvarySARDULS. GURAYA Volume 50 Morphological Modulations in Helical Cell Surface Enzymes: Effects on Mitotic Muscles (Aschelminthes and Annelida) Activity and Cell Adhesion-H. BRUCE -GIULIO LANZAVECCHIA Interrelations of the Proliferation and BOSMANN Differentiation Processes during CarNew Aspects of the Ultrastructure of Frog diac Myogenesis and RegenerationRod Outer Segments-JijRGEN ROSENPAVELP. RUMYANTSEV KRANZ Mechanisms of Morphogenesis in Cell The Kurloff Cell-PETER A. REVELL Cultures-J. M. VASILEV AND I. M. Circadian Rhythms in Unicellular Organisms: An Endeavor to Explain the GELFAND Cell Polyploidy: Its Relation to Tissue Molecular Mechanism-HANS-GEORG Growth and Functions-W. YA. BRODSCHWEIGER AND MANFREDSCHWEIGER SKY AND I. v . U R W A E V A SUBJECT INDEX

A 6 8 c 9 D O E l F 2

6 3 H 4 1 5 J 6