Advances in Applied Microbiology, Volume 27

ADVANCES IN

Applied Microbiology VOLUME 27

CONTRIBUTORS TO THIS VOLUME

A. M. Cundell Robert J. Fleischaker John C. Godfrey

A. Hurst

L. S. McClung Vedpal Singh Malik Anthony J. Sinskey James C. Weaver

ADVANCES IN

Applied Microbiology Edited by D. PERLMAN School of Pharmacy The University of Wisconsin Madison, Wisconsin

and Allen I. Laskin Exxon Research and Engineering Company Linden, New Jersey

VOLUME 27

@

1981

ACADEMIC PRESS A Subsidiary of Harcourt Brace Jovanovich, Publishers

New York London Toronto Sydney San Francisco

COPYRIGHT @ 1981, BY ACADEMIC PRESS, INC. ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM O R 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 l 7 D X

LIBRARY OF CONGRESS CATALOG CARDNUMBER:59-13823

ISBN 0-12-002627-9 PRINTED IN THE UNITED STATES OF AMERICA 81828384

9 8 7 6 5 4 3 2 1

CONTENTS LJST

OF

CONTRIBUTORS., .............................

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

ix

Recombinant DNA Technology

VEDPALSINGHMALIK I . Introduction ........................................................

I1. Plasmid DNA Purification

111.

.

IV V. VI . VII . VIII .

............................................ Isolation of Genes ................................................... Making Recombinant DNA Molecules in Vitro .......................... Transformation ...................................................... Host-Vector Systems ................................................ UtiEity of New Technology ............................................ A Postscript ........................................................ References .........................................................

2 3 4 22 31 32 47 64 67

Nisin

A . HURST I . Introduction ........................................................ I1. Biology ............................................................

.

111

IV . V.

Chemistry .......................................................... Biosynthesis ........................................................ Use of Nisin ........................................................ References .........................................................

85 87 98 105 108 119

The Coumermycins: Developments in the Late 970s JOHN

C . GODFREY

......... ......... 111. Biological Developments ............................................. IV. Conclusion ......................................................... References ......................................................... I . Introduction .............................................. I1. Chemical Developments. 1972-1979 .........................

V

125 125 128 133 133

vi

CONTENTS

Instrumentation for Process Control in Cell Culture

ROBERTJ . FLEISCHAKER. JAMES C. WEAVER.AND ANTHONYJ . SINSKEY I . Introduction

........................................................ ........................ Microcarriers ....................................................... Generalized Model of Cell Growth ..................................... Evaluation of Instrumental Methods ................................... Temperature ........................................................ Mixing and Viscosity ................................................. pH ......................... ................................ Ionic Strength ......................................................

I1. Comparison between Animal and Microbial Cells

I11. IV. V. VI . VII . VIII . IX. X. Dialyzable Components: Carbohydrates. Organic Acids. Amino Acids. Salts. and Other Nutrient Components ...................................... XI * Oxygen ............................................................ XI1. Carbon Dioxide ..................................................... XI11. NAD+-NADH ...................................................... XIV. Oxidation-Reduction Potential ......................................... xv. ATP .......................... ................................. XVI. Cytofluoremetry ..................................................... XVII . Conclusion ......................................................... References ...................... ..............................

137 138 139 140 142 145 146 148 149 152 153 156 159 162 163 163 165 165

Rapid Counting Methods for Coliform Bacteria

A . M . CUNDELL I. I1. 111. IV. V.

Introduction .................................................... Criteria for a Rapid Coliform-Counting Procedure ........................ Review of Rapid Coliform-Counting Procedures ......................... Potential for Future Development ..................................... Summary ........................................................... References .........................................................

169 170 171 179 180 180

Training in Microbiology at Indiana University-Bloomington

L. S . MCCLUNG ........................................................

I. Introduction I1 The Early Period .................................................... 111 The Modem Period ..................................................

. .

185 187 187

CONTENTS

vii

IV . Faculty and Research ................................................ V . Physical Facilities ................................................... Appendix: List of Ph.D. Recipients. 1949-1979 .......................... References .........................................................

191 197 201 205

INDEX .................................................................... CONTENTS OF PREVIOUS VOLUMES ...........................................

207 211

This Page Intentionally Left Blank

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

A. M. CUNDELL,Flow Laboratories, Znc., 25 Lumber Road, Roslyn, New York 11576 (169) ROBERTJ . FLEISCHAKER, Department of Nutrition and Food Science, Massachusetts Znstitute of Technology, Cambridge, Massachusetts 02139 (137)

C. GODFREY,Godfrey Science and Design, Znc., 282 Soundview Avenue, White Plains, New York 10606 (125)

JOHN

A. HURST, Microbiology Research Division, Health Protection Branch, Health and Welfare Canada, Ottawa, Ontario, Canada KIA O D (85)

L. S. MCCLUNG, Department of Biology, Zndiana University, Bloomington, Zndiana 47405 (185) VEDPALSINGHMALIK,The Upjohn Company, Kalamanoo, Michigan 49008 (1) ANTHONYJ. SINSKEY,Department of Nutrition and Food Science, Massachusetts Znstitute of Technology, Cambridge, Massachusetts 02139 (137) C. WEAVER,Haruard-MZT Division of Health Sciences and Technology, Cambridge, Massachusetts 02139 (137)

JAMES

ix

This Page Intentionally Left Blank

Recombinant DNA Technology

VEDPAL SINGH MALIK The Upjohn Company, Kalamazoo, Michigan

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Isolation of Genes

,

. .. .. ... .. .

A. Selection for Function.

....

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

............. B. Ligation of Restriction Fragments V. Transformation. . .

,.........

C. Streptomycetes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Yeast . . . . . F. Mammalian Cells.. . . . . . ..................... VII. Utility of New Technology. . . A. Merodiploids Constructio ............ B. Operon Fusions . . . . . . . . . . . . . C. Directed Secretion of Proteins . . . . . . . . . . . . . . . . . . . . . . D. I n Vitro Localized Mutagenesis . . . . . . . . . . . . . . . . . . . . . . . E. Genome Organization . . . . .... F. Origin (On) of Replication. . . .............. .............. G. Human Proteins.. . . . . . . . . . . ............. H. Vaccines . . . . . . . . . . . I. Improvement of Industrial .......,......... J. Vanishing Organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . K. Resource Recovery and Waste Disposal. . . . . . . . . . . . . . . . L. Improved Crops.. . . . . . . . . . VIII. A Postscript . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . .

.

.. . .

.

.

2 3 4 5 9 13 18 19 22 23 2.3 27 29 31 32 34 38 39

43 44 45 47 48 48 50 50 51 53 54

55 56 61 61 61 64

67

1 ADVANCES IN APPLIED MICROBIOLOGY. VOLUME 27 Copyright @ I 1881 by Academic Press. Inc. Au rights of reproduction in any bnn reserved. ISBN 0-13-0026e7-9

I

2

VEDPAL SINGH MALIK

I. Introduction Where ignorance of local manners-and languagehas led me astray, I ask due pardon from the local proprietors. J. D. B u ’ L ~ ~(1966) K

.

Read not to contradict and confute, nor to believe and take for granted. . but to weigh and consider. FRANCIS BACON(1597)

This article is addressed to those biologists who have knowledge of molecular genetics but would like to be familiar with the literature dealing with the rapidly advancing forefront of recombinant DNA technology. Genetic recombination that results from the breaking and rejoining of DNA molecules is subject to strong taxonomic constraints in viuo. Exchange of genetic material rarely occurs between unrelated organisms, and even similar bacterial species may not exchange chromosomal genes. Plasmids and viruses can incorporate small pieces of host DNA in their own chromosomes and act as vectors for transference of these genes to closely related bacterial species. These vectors bearing chromosomal genes can either become integrated into the chromosome of the recipient cell or exist as independently replicating units in the cytoplasm. A certain degree of DNA sequence homology is a prerequisite for genetic recombination in viuo. However, recombination between DNA molecules in uitro is not subject to these taxonomic restrictions. Consequently, genetic material from organisms of quite diverse origins with little DNA sequence homology can now be attached into an appropriate vector and introduced into a host cell for propagation and expression. The possibility of cloning DNA of higher organisms into a genetically well-characterized haploid microbe such as Escherichiu coli is an important recent development in applied microbiology, This new technology has paved the way for the propagation of lines of organisms that are genetically alike and that all contain identical recombinant DNA molecules. A variety of novel genetic combinations in microorganisms can thus be created. This new technology also makes it possible to isolate defined genes and to prepare large quantities of homogeneous fractions of DNA from organisms whose genetics and biochemistry are not well known. Microorganisms that have been used to benefit man for centuries can now be genetically modified by implanting foreign genes into them. Furthermore, mammalian proteins and other useful products can be synthesized through the use of microorganisms genetically modified by the implantation of foreign genes. Ultimately, this new technology will surely accelerate the understanding of the regulatory biology of the organisms that have thus far been refractory to in uitro methods of genetic analysis (Dawid and Wahli, 1979;Sinsheimer, 1977; Setlow and Hollaender, 1979; Maniatis, 1980). Recombinant DNA technology generally involves several steps:

RECOMBINANT DNA TECHNOLOGY

3

1. DNA fragments coding for proteins of interest are synthesized chemically or isolated from an organism. 2. These DNA fragments are inserted in a restriction endonuclease cleavage site of the vector that does not inactivate any gene required for the vector’s maintenance and selective marker. 3. The recombinant DNA molecules are then introduced into a host to replicate using the replication origin of the vector. 4. Recipient host cells that have acquired the recombinant DNA are selected. Selection pressure is applied to enrich bacteria with a selectable marker. 5. Desired clones are then characterized to ensure that they maintain true copies of the DNA segment that was originally cloned.

II. Plasmid DNA Purification Plasmids are widely used as vectors and methods of their isolation can vary depending upon the host organism. In E. coli, many methods are used for isolating plasmid DNA free from chromosomal DNA. Plasmid-harboring E. cold is grown in any rich medium such as Typticase Soy broth or a minimal salt solution supplemented with casamino acids. In certain plasmid isolations, such as ColE1 derivatives, where protein synthesis is not required to initiate plasmid DNA replication, 300 &ml of chloramphenicol is added in the log phase of growth. This selectively inhibits chromosome duplication and allows amplification of plasmid DNA so that 45%of total cellular DNA is plasmid (Clewell, 1972). Cells are centrifuged and suspended in 50 mM Tris buffer containing 25%sucrose. Lysozyme in the presence of EDTA is added to hydrolyze the peptidoglycan layer of the cell wall. Protoplasts thus formed are then lysed by addition of the anionic detergent sodium dodecyl sulfate so that the chromosome remains attached to the cell envelope. Sodium chloride is added to 1 M concentration to precipitate chromosomal DNA and cell debris. After centrifugation, the supernatant containing plasmid DNA is removed (Guerry et al., 1973).Usually, phenol or pronase and ribonuclease are used to remove protein and RNA. The plasmid DNA is subsequently precipitated with ethanol or with polyethylene glycol (Johnson and Gunsalus, 1977).The DNA is then purified by CsCl density-gradient centrifugation in the presence of ethidium bromide (Martens and Clayton, 1977). Other procedures are now available for isolating plasmids from E. coli (Ohlsson et al., 1978; Colman et al., 1979), pseudomonads (Johnson and Gunsalus, 1977; Palchaudhuri and Chakrabarty, 1976; Hansen and Olsen, 1978); Bacillus subtilis (Gryczan and Dubnau, 1978); Serratia murcescem (Timmis and Winkler, 1973); Streptococcus (Clewell et al., 1974); Streptomyces (Bibb et al., 1978; Malik, 1977, 1978; Schrempf et d.,1975; Yagisawa et al., 1978), Agrobacterium (Currier and Nester, 1976); yeast

4

VEDPAL SINGH MALIK

(Zakian et al., 1979);and other eukaryotic cells (Shoyab and Sen, 1978). It is extremely important that DNA be free from contaminating nucleases and RNA species. Sometimes DNA has to be purified by polyacrylamide gel electrophoresis, sucrose gradient centrifugation, or column chromatography before it can be hrther used (Hardies and Wells, 1976, 1979; Hardies et al., 1979).

Ill. Isolation of Genes Chromosomal location of DNA sequences of a gene affects the ease with which the genes are isolated. On the basis of phenotypic effect and organization in the chromosomes, genes can be subdivided into the following groups: 1. Simple genes. Their product is a single polypeptide or single RNA (e.g., ribosomal RNA, tRNA, insulin, histone, ovalbumin, or monomeric enzymes). Such genes are the easiest ones to isolate. 2. Complex genes. Many multimeric proteins are aggregates of several nonidentical polypeptide subunits. Some require cytochromes and specific cofactors for their biological activity. All of these genes coding for the various parts of the hnctional enzyme aggregate may not be clustered together on the chromosome, e.g., the genes that code for the tryptophan synthetase, fatty acid synthetase, mixed-function oxidase, hydroxylases, and enzymes that oxidize drugs but require cytochrome P-450. 3. Operons. Many genes specifying degradative and biosynthetic pathways are clustered adjacent to each other on the chromosome, and these gene clusters are called operons. Examples are lac operon and histidine operon (Miller and Reznikoff, 1978). 4. Regulon. Certain genes specifying a biosynthetic function such as arginine biosynthesis are organized in several minioperons widely scattered on chromosomes. However, all such genes are regulated coordinately and have been called regulons. 5. Multiple regulons. Many products, e.g., antibiotics, are assembled from precursors derived from several different pathways. For example, L-cysteine, L-valine, and L-a-aminoadipate are involved in the biosynthesis of P-lactam antibiotics (Malik, 1980). Multiple regulons participating in the synthesis and regulation of the biosynthesis of such complex molecules are probably scattered all over the chromosome. Isolation of all genes involved in antibiotic synthesis can therefore be a difficult task. Since genes are composed of DNA stretches, to isolate genes one must first isolate DNA. Purified cytoplasmic organelles such as mitochondria and chloroplasts can be used to obtain their DNA. However, isolation of specific genes from total DNA is difficult and many properties of genes are used to advantage to isolate the genes. Human genome contains about lo' globin gene-sized DNA sequences, of which approximately 60% occur in one or a

RECOMBINANT DNA TECHNOLOGY

5

few copies. To increase the probability of success in cloning a particular gene from such a highly complex genome, it helps to enrich DNA sequences before cloning into a vector. A combination of several methods could yield an enrichment of several thousandfold. Genes located on plasmids or phages are greatly enriched by purifying the plasmid or-phage DNA. Many genes involved in aromatic and sterol metabolism (Chakrabarty, 1978), heavy metal transformations, and drug resistance are located on plasmids. Genes for the insecticide toxin produced by Bacillus thuringiemis may be located on a bacteriophage (Perlak et al., 1979; Templeton et al., 1979). After preparations that have been enriched for a desired gene, recombinant DNA technology is used to isolate the genes from such preparations.

A.

SELECTION FOR

FUNCTION

Many genes can be selected by their function if mutants of a host organism lacking that function are available. Cohen and Chang (1973)were the first to take advantage of the fact that certain genes determining resistance to antibiotics were located on plasmids. These genes were enriched by purd$ng plasmid DNA. Their isolation was therefore made easier by selecting clones that harbored genes responsible for resistance to antibiotics. These authors reported the in vitro construction of DNA molecules that combined genetic information from two different E. coli plasmids carrying drug-resistance markers. These recombined plasmids were inserted by transformation into drug-sensitive E . coli. The transformant clones were selected by directly plating cells on nutrient agar plates supplemented with antibiotics. On such media, only cells that acquired recombinant plasmids survived since the host was killed by antibiotic. Cohen and Chang were also the first to insert the penicillinase gene from a gram-positive bacterium Staphylococcus plasmid into an E. coli plasmid. The penicillinase gene was expressed and maintained when the chimeric plasmid was introduced into E. coli. A thymidylate synthetase gene from the B. subtilis bacteriophage 43T has also been cloned and expressed in E. coli. Phage $3T DNA was digested with Eco RI and ligated to an EcoRI-digested pMB9 plasmid of E. coli. This recombinant DNA was used to transform a thy- E. coli. The desired transformants had a thy+ tet phenotype (Ehrlich et al., 1976). These recombinant plasmids replicated in E. coli, utilizing the origin of DNA replication of the E. coli vector plasmid. Such experiments did not tell whether genes of gram-positive organisms were being transcribed in E. coli from a promoter of the vector or from their own promoter. Base sequences of regulatory elements, e.g., promoters, could be very different in different organisms. The level of gene expression in a given organism could vary depending upon the efficiency of interaction between regulatory sequences and corresponding protein molecules.

6

VEDPAL SINCH MALIK

1 . Genome Library

Techniques for the construction and screening of a complete library of clones for any genome is a practical starting point for ampllfying specific genes (Wensink et al., 1974;Benton and Davis, 1977; Sternberg et al., 1977; Maniatis et al., 1978; Dodgson et al., 1979; Gergen et al., 1979).A complete library of genome fragments is a set of independent clones that, statistically, contains the entire genome among the recombinant DNA molecules (Clarke and Carbon, 1976). Complete sets of DNAs can be maintained if no segment is disruptive to the vector or host. If every DNA segment is transformed and maintained with equal efficiency, then the probability (P) of the presence of a desired DNA sequence in a genome bank is as follows:

P

= 1 - [(l

-+)4"

where x = length of desired segment of DNA, L = average length of DNA fragment cloned, F = fraction of genome represented by an average fragment, and N = number of transformant colonies containing cloned DNA. Clarke and Carbon calculated that 720 transformed colonies will give a probability of 0.90 that one E. coli gene was cloned with an average molecular weight of 8.5 X lo6. However, the probability increases to 0.99 with a genome bank of twice this size. A library of soybean genome (Breiner et al., 1979) and chicken genome has been constructed in the vector A charon 4A (Maniatis et al., 1978). Chicken DNA was partialky digested with restriction enzymes HaeIII and AZuI. Fragments of 14-22 kb were purified by sucrose gradient fractionation. This DNA was methylated by EcoRI methylase, ligated to synthetic dodecameric linkers, digested with EcoRI restriction enzyme, and refractionated on sucrose gradients. These fragments were then ligated to purified charon 4A left and right arms joined at the A cohesive end site. The ligated concatemeric DNA was treated with the in uitro packaging extracts. Packaged phages were purified by cesium chloride centrifugation. These phages were amplified by growth on plates at subconfluent densities. Several independent globin gene-containing recombinants were selected from the chicken genome library by screening with adult and embryonic globin cDNA as a probe (Dodgson et al., 1979).

2 . Screening of Genome Library a. Auxotrophic Complementation. Many genes have been identified by transforming auxotrophs with recombinant molecules. For making genomic banks in E. coli, the recipient is usually F+. This allows transfer of recom-

RECOMBINANT DNA TECHNOLOGY

7

binant plasmids to other E. coli auxotrophs (Miller, 1972) and provides a rapid method for identifying a recombinant plasmid containing a specific gene from a genome bank of randomly cloned genome fragments (Clarke and Carbon, 1976; Latrou et al., 1980). Whenever applicable, this procedure is extremely rapid, since a large number of colonies from a genome bank can be screened with little effort. Many individual colonies from the genome library can be grown in a grid on a master plate and replicated to a lawn of recipient auxotroph cells on the appropriate selective media. However, the only genes selected by this method are those for which a bacterial mutation exists and that are expressed in the bacterial host. The expression of the desired gene can vary depending on the vector, cloning site, and host cell; and such a selection may not produce the gene of interest even though the source of a gene is a bacterium related to the host organism. Methods have been developed for the selection of DNA sequences that complement metabolic mutations and structural defects such as those involved in mutant flagellar genes (Silverman and Simon, 1977; Schell and Wilson, 1979; Walz et al., 1978; Moore and James, 1979; Vapnek et al., 1977; Chinault and Carbon, 1979; Gardner, 1979; Struhl et al., 1976). Recombinant molecules carrying specific genes have been selected directly by using appropriate E. coli auxotrophs as host. Restrictionless strains of E. coli that are transformed with high efficiency are usually used as recipients. Genes that complement trp, gal, lop, l a d , ara, thyA, leu, bio, and many others have been identified in this manner (Clarke and Carbon, 1976; Hershfield et aZ., 1974; Berg et al., 1976; Collins et aZ., 1976; Cohen et al., 1978). A Saccharomyces cerevisiae DNA segment of 4700 base pairs containing the galactokinase gene (gal-])has been cloned in pBR322 (Hind111 site) and maintained in an E. coli strain that carries a deletion in its own galactokinase gene (galK). The yeast gene was shown to be present (1) by complementation of the E. co2i galactokinase deletion, (2) by hybridization of the cloned DNA fragment to the restriction enzyme digest of total yeast DNA, and (3) by demonstration of yeast galactokinase, a monomeric protein, in cell-free extracts of the E. coli harboring the plasmic that contains the yeast galactokinase gene. The yeast galactokinase activity in E. coli extracts is 0.7%of the bacterial galactokinase activity present in wild-type E. coli cells fully induced with fucose. E. coli cells lacking a functional galactokinase gene but harboring the yeast galactokinase gene grow slowly in minimal medium containing galactose as the sole carbon source. Such cells have a generation time of 14.3 hours (Citron et al., 1979). The replication of pBR322 is under relaxed control and 50-100 copies per cell of the hybrid plasmid containing the yeast galactokinasegene could be expected. These multiple copies of the gene per cell should allow production of galactbkinase enzyme in sufficient

8

VEDPAL SINGH MALIK

amount for good growth of the host, considering that turnover numbers in vivo of the E . coli and yeast galactokinase enzymes are similar. The yeast galactokinase gene is expressed in E. coli very poorly. This decreased expression of yeast gene in E. coli could be due to inefficient transcription, translation, posttranscriptional or posttranslational modification of galactokinase, or a combination of all. St. John and Davis (1979) have also reported the cloning in E. coli of yeast galactose-inducible sequences selected by another technique. The similarity between sequences cloned by St. John and Davis (1979) and Citron et al. (1979) cannot be assessed by the data given. Saccharomyces cerevisiae genes coding for P-galactosidase (Dickson and Martin, 1978) and several genes involved in other biosynthetic pathways (Struhl et al., 1976; Carbon et al., 1977; Ratzkin and Carbon, 1977; Struhl and Davis, 1977; Clark and Carbon, 1978; Bach et al., 1979) are expressed when cloned in E. coli. Several yeast genes, for example, leu-2, his-3, ura-3, trpG, and adh-2, have been isolated by complementation test (Williamson et al., 1980). Yeast cells carrying an ampR gene of E . coli plasmid ligated to a yeast replicon produce penicillinase (Chevallier and Aigle, 1979).

b. In Situ Hybridization. Colonies or phage plaques with the desired gene can be identified in situ by hybridization to a radioactively labeled RNA probe complementary to the desired gene (Grunstein and Hogness, 1975). This procedure is used when the desired clones cannot be assayed for the function of the gene of interest. Under such circumstances, DNA sequences for the gene of interest are detected by taking advantage of the complimentarity of the RNA or DNA probe. For each screening of the library, petri dishes (150 mm) with up to 10 plaque-forming units (PFU) or colonies per plate can be used. From each petri plate, viral plaque DNA or colonies are fued to the nitrocellulose filters and lysed, and the DNA is denaturated by alkali treatment (Benton and Davis, 1977). Colonies or plaques that hybridize with 32P-labeledprobe RNA or cDNA of the genes of interest are located by autoradiography or fluorography. Duplicate positive plaques or clones can be purified through several further platings at low plaque density. Colony hybridization has been used widely. Similar procedures have been developed for SV40 plaques (Villarreal and Berg, 1977), and have been used to isolate Drosophila tRNA genes (Dunn et al., 1979) and ribosomal DNA of B. subtilis (Moran and Bott, 1979), soybean (Breiner et al., 1979), and Dictyostelium (Maizels, 1976) c. In Situ Immunoassays. Radioisotopes such as lz5Ihave been used for many years as labels for monitoring the distribution of reagents in immunological assay systems. Nonradioactive labels such as bacteriophages, enzymes, stable free radicals, and fluorescent and chemiluminescent groups

RECOMBINANT DNA TECHNOLOGY

9

have also been used (Simpson et al., 1979; O’Sullivan et al., 1979; Harris et al., 1979). Radioimmunoassays have been used to identlfy genes for which E. coli mutants do not exist (Henning et al., 1979). The method depends on the presence of the gene product and requires expression of the cloned gene in the host. Both viral and phage vectors can be used, since the technique is applicable to phage plaques as well as colonies. Broome and Gilbert (1978) developed a very sensitive solid-phase radioimmunoassay capable of detecting picograms of specific antigens. This pioneering methodology was first used to select E. coli clones that synthesized rat proinsulin (Villa-Komaroff et al., 1978). Such assay can be used to screen clones for the presence and expression of specific gene sequences. Because of the extreme sensitivity of this assay, it can detect even the very low level of expression resulting in the synthesis of only a few molecules per cell. Hybridoma technology can be used to make specific monoclonal antibody (F. Melchers et al., 1978). The combination of radioimmunoassayswith gel electrophoresis can be used to characterize immunoreactive proteins (Secher and Berke, 1980). Enzymes involved in the biosynthesis of peptide antibiotics have been purified to homogeneity (Kleinkauf and Koischvitz, 1980). Such pure enzymes could now be used to make specific antibodies needed for a radioimmunoassay to isolate the genes involved in peptide antibiotic synthesis.

d . Sib Selection. This method can be used to isolate clones from a genome library. The whole genome library is divided into pools, each containing many different clones. Each pool is tested by the pertinent assay procedure for the presence of a desired gene. That pool containing the gene of interest is subdivided and retested until the individual clone is identified. Sea urchin histone genes and human interferon genes have been isolated in this way (Kedes et al., 1975; Bylinsky, 1980; Gilbert and Villa-Komaroff, 1980). B. PHYSICAL DIFFERENCES 1 . Density

If the gene differs significantly in base composition from the total cellular DNA and is present in multiple copies per genome, then the repeated sequences band in CsCl gradient as a satellite in addition to the main chromosomal band. Ag+ or Hg+ increases such density differences. Ribosomal RNA genes of Xenopus laeuis were isolated on the basis of density differences and were the first eukaryotic genes to be cloned in E. coli (Morrow et al., 1974).

10

VEDPAL SINCH MALIK

Amplified rRNA genes of Tetrahymenu are free from chromosomes in the macronucleus of Tetrahymenu (Din and Engberg, 1979) and can be easily separated on sucrose gradient or by agarose gel electrophoresis. An EcoRI-generated fragment of plasmid pSC122 was purified by equilibrium centrifugation in cesium chloride (Timmis et al., 1975). This fragment coding for ampicillin resistance had a buoyant density of 1.692 gm/cm3 whereas the remaining DNA fragment banded at a buoyant density of 1.710 gm/cm3. Firtell et al. (1976) enriched recombinant DNA molecules of Dictyosteliurn dwcoideum (22% GC) and pSClOl (50% GC) by fractionation on a CsCl density gradient. The region between 30 and 40% GC yielded about 85%hybrid molecules, whereas the initial population contained 2 4 % hybrids. The approach used by Firtell et aZ. (1976)could be used to enrich recombinant molecules with pieces of Streptmyces genome (70% GC) if they could be ligated with homopolymer (dGidC) tailing into a vector with low GC content. 2 . Size Many enrichment procedures utilize information about the size of the DNA fragment required for the desired gene. a. Gel Electrophoresis. Restriction enzyme-generated fragments are separated according to size by gel electrophoresis (Southern, 1979a). Bands containing a gene can be designated (Helling et al., 1974)when the fragment order of a simple genome such as that of a plasmid or phage is known (Table I). For organisms with a high efficiency of transformation, the bands of endonuclease-digested DNA can be isolated after electrophoresis and correlated with specific genes after transformation (Young et al., 1977; Brown and Carlton, 1980; Mazza and Zalizzi, 1978). Gel electrophoresis is routinely used for determining the fragment composition of recombinant DNA molecules constructed by ligation of cohesive ends generated by restriction enzymes. Digestion by the same restriction enzyme regenerates the original DNA fragments and such fragments are easily identified by gel electrophoresis. As little as 10 pg of DNA per band can be visualized by staining the gel with ethidium bromide, which fluoresces under UV light (Helling et al., 1974). DNA can be extracted from these resolved bands and used for further analysis with various restriction enzymes, nucleotide sequencing, or transformation. Agarose gel electrophoresis is best suited for characterizing fragments above 1.5 kb. For smaller fragments, polyacrylamide gels are used. Standards of A and 4x174 DNA already cleaved with restriction enzymes are commercially available

11

RECOMBINANT DNA TECHNOLOGY

TABLE I FRAGMENT SIZES' PRODUCED BY b S T l U C r I O N

ENDONUCLEASE DIGESTION OF A AND 4x174 GENOMES

A Fragment number

Hind111

EcoRI

1 2 3 4 5 6 7 8 9 10 11

23,720 9,460 6,670 4,260 2,250 1,960 590 100

21,800 7,550 5,930 5,540 4,800 3,380

4x174 BamHI

HaeIII

HaeII

TaqI

17,400 7,300 6,760 6,470 5,540 5,510

1,353 1,078 872 603 310 281 271 234 194 118 72

2,314 1,565 783 269 185 123 93

2,914 1,175 404 327 231 141 87

54

54

33 20

'In base pairs. and can be used as markers for determining size of unknown DNA fragments. Digestion of E. coli DNA with EcoRI restriction enzyme yields about 500-600 fragments differing in molecular weight, base composition, and sequence. When this digest is examined by agarose gel electrophoresis, a featureless pattern is obtained. However, Fischer and Lerman (1979) have developed a two-dimensional electrophoretic method that resolves the large number of components of an EcoRI-digested E. coli chromosome into 350 discernible spots. The significance of this technique is demonstrated by the detection of four spots unique to a digest of E. coli lysogenized by A. These four spots were not observed in uninfected E. coli, but two of these spots were due to A phage DNA alone. Creighton (1979)used a similar procedure in a study of protein denaturation. Rapid and sensitive procedures for detecting plasmids by agarose gel electrophoresis have been developed (LeBlanc and Lee, 1979).Plasmids present in one copy per cell with a molecular weight of up to 1.5 x 108 can be identified. One colony is sufficient to obtain a fragment pattern of a recombinant plasmid after restriction endonuclease cleavage (Eckhardt, 1978; Klein et al., 1980). Southern (1975)developed the technology for studying gene arrangement directly. High-molecular-weight DNA is cleaved with a restriction enzyme and separated into fragments by agarose gel electrophoresis. The DNA is

12

VEDPAL SINGH MALIK

then denatured and transferred from the gel onto a cellulose nitrate strip. The cellulose nitrate strip is hrther baked to ensure binding of DNA to the paper and is then hybridized with a specific labeled RNA, DNA, or cDNA probe. The probe hybridizes only to its complementary partner and can be localized by autoradiography. The size of the gene sequence localized as a radioactive band can be determined by comparing with DNA standards electrophoresed on the same dilute agarose gel. Using the Southern hybridization technique, physical maps of restriction endonuclease cleavage sites in several structural genes and in the DNA region flanking the genes have been constructed. Comparison of the size of the restriction fragments of structural genes and cDNA has revealed that many genes are not contiguous in the DNA but have noncoding inserts. Such noncoding inserted sequences have been found in the rabbit P-globin gene, chick ovalbumin (Kourilisky and Chambon, 1978; Catterall et al., 1979), mouse P-globin (Tilghman et al., 1978), mouse immunoglobulin light chain (Brack and Tonegawa, 1977), Drosophiln ribosomal genes (Glover and Hogness, 1977), and yeast mitochondria1 tRNA (Goodman et al., 1977). Southern (1979b) has designed a gene machine in which DNA is electrophoresed from the outside to the center of a 2% agarose cylinder. DNA is collected on a dialysis membrane when it reaches the center and is eluted at intervals. The high capacity of the gene machine makes it an ideal first fractionation device. Approximately 200300 fractions are separated, of which only a few contain the desired gene fragment. In this manner, up to 50-fold enrichment of a specific gene sequence can be achieved.

b. Electron Microscopy. The most accurate and reliable method for determining the size of DNA is electron microscopy. Heteroduplex analysis is used to build the map of A DNA digested with EcoRI. Such maps of plasmid and phage genomes as well as other DNA segments are routine these days and can be confirmed by comparing the electrophoretic pattern of DNA fragments obtained by restriction cleavage of recombinant molecules differing by deletion or by the addition of DNA from other genomes. New methods for visualizing RNA-DNA hybrids by electron microscopy permit precise localization of individual genes (Thomas et al., 1976; Miwa el al., 1979). c. Genetic Methods. DNA coding sequences for yeast tyrosine transfer RNA have been located on eight digerent EcoRI restriction fragments. There is significant natural variation among various yeast strains in the sizes of EcoRI and Hind111fragmentsthat hybridize with tyrosine tRNA. Restriction fragments of various sizes hybridizing with the same tRNA genes occupy genetically homologous chromosomal sites. These size variants of EcoRI-HindIII

RECOMBINANT DNA TECHNOLOGY

13

restriction fragment behave as Mendelian traits. Olson et al. (1979) associated six of the eight tyrosine tRNA loci with specific restriction fragments by establishing genetic linkage between the suppressor tRNA and a size variant of the respective restriction fragment. Even though all eight genes code for suppressor tyrosine tRNA, pronounced differences in phenotype exist between different suppressors. Now, cloned DNA sequences can be correlated with the corresponding genetic loci by transforming into appropriate strains of yeast and testing for functional expression of the suppressor. Restriction fragment size variants have been used to analyze inheritance of plant and animal mitochondrial DNA (Levings and Pring, 1976), rDNA, and 2 pm plasmid sequences of yeast (Petes, 1979; Livingston, 1977; Fonty et al., 1978), and Drosophila mitochondrial DNA (Shaw and Langley, 1979). Widespread Occurrence of restriction fragment size variants can have an impact on molecular genetics similar to that of electrophoretic variants on proteins and may speed up the genetics of eukaryotes (Birky, 1978). C.

COMPLEMENTARY RNA

1 . cDNA

Many eukaryotic genes are split. The mRNA coding sequences (exons)are interrupted by intervening sequences (introns) of unknown function. These split genes are transcribed into an RNA precursor that matures into mRNA by a splicing mechanism (Abelson, 1979). Since split genes are not found in bacteria, they are not expressed in E. coli (Puijalon and Kourilsky, 1979). Therefore, cDNA made in vitro has been used to get expression of several mammalian proteins in E. coli. This method of gene isolation is possible only if mRNA species that code for a certain protein can be sufficiently enriched. Several methods for isolating biologically active ribonucleic acid from various sources have been described (Chirgwin et al., 1979; Mansson et al., 1979; Payvar and Schimke, 1979; Brandt and Ingversen, 1978). Total RNA containing poly(A) tails is sometimes translated in a cell-free system and the polysomes precipitated by the purified antibody. The enriched RNA is further purified by affinity chromatography on an oligo(dT)column. RNA-dependent DNA polymerase from avian myeloblastosis virus is used for the synthesis of complementary DNA (cDNA) strand from RNA. This enzyme is also called reverse transcriptase and generates a hairpin structure with a short double-stranded region at the 3’-terminal of cDNA. Escherichia coli DNA polymerase I is used to synthesize a second DNA strand of this self-priming cDNA with a 3’-terminal hook. The hairpin loop connecting the two complementary

14

VEDPAL SINGH MALIK

strands is digested by S1 nuclease and the RNA is hydrolyzed by alkali treatment. The desired species of cDNA is identified by restriction enzyme analysis and size determination of the DNA fragment as determined by gel electrophoresis. The cDNA can be cloned into a vector by homopolymer tailing, blunt-end ligation, or utilization of linkers. After restriction endonuclease digestion, a distinct fragment pattern is obtained from cDNA synthesized from an enriched mRNA species. This effect has made it possible to clone rat insulin cDNA (Ullrich et al., 1977), human growth hormone (Goeddel et al., 1979a; Roskam and Rougeon, 1979), human chorionic gonadotropin (Shine et al., 1977), and rat growth hormone (Seeburg et al., 1978b). Ullrich et al. (1977) developed ingenious methods for obtaining DNA sequences coding for insulin. Purified rat insulin mRNA, which could have been used as a probe, was not available at that time. Total polyadenylated mRNA was used to make cDNA from specialized pancreatic cells that synthesize insulin. Restriction endonuclease digestion with Hue111 of the cDNA produced major fragments of 180 and 80 nucleotides. These two fragments were present in an undigested species of cDNA of about 450 nucleotides. This 450-nucleotide-long cDNA was joined to Hind111 synthetic linkers before insertion into a plasmid vector. The recyclization of the vector PMBQ was prevented by removing phosphate groups with alkaline phosphatase. In this way, only plasmids that annealed to cDNA with terminal 5'-phosphates were ligated by ligase. Hue111 digests of plasmid DNA from transformed cells were used to identify clones that carried 450, 180, and 80 nucleotide fragments characteristic of the major species of cDNA. Direct sequencing of cDNA showed that it carried information for coding known amino acid sequences for rat insulin. The recombinant plasmids isolated by Ullrich et al. (1977)contained the whole of the proinsulin coding sequence and part of the N-terminal prepeptide. Even though the recombinant plasmid-bearing strain did not produce any insulin, these pioneering experiments opened the way for obtaining gene sequences that utilized the size of cDNA and size of its characteristic restriction fragments to select the correct DNA sequence that codes for the polypeptide of interest. Recently, Bell et al. (1979) have isolated and sequenced a cDNA clone encoding human preproinsulin. Double-stranded cDNA, prepared from unfractionated mRNA isolated from human insulinoma, was inserted into PstI endonuclease site of P-lactamase gene of pBR322 using poly(dG)-poly(dC) homopolymer extensions. This DNA was used to transform E. coli, and 525 tetracycline-resistant transformants were obtained. These were replicaplated and screened for insulin sequences by colony hybridization using nick-translated cloned rat preproinsulin I cDNA as a probe. Autoradiography revealed that 2 out of 525 colonies hybridized with the probe. Both

RECOMBINANT DNA TECHNOLOGY

15

colonies were ampicillin-sensitive and contained inserts of about 250 and 500 base pairs. Since the two sequences were in different reading frames, fused P-lactamase human preproinsulin polypeptide was not synthesized by either one of these clones. However, isolation of these human preproinsulin sequences is a step toward microbial production of human insulin. Fiddes and Goodman (1979) isolated sequenced and cloned cDNA for the a-subunit of human chorionic gonadotropin (HCG). Polyadenylated mRNA from first trimester placenta was enriched for human chorionic gonadotropin-specific mRNA. These RNA samples were translated in a wheat germ cell-freetranslation system with [ 35SJmethionine, and the synthesized proteins were characterized on an SDS-15% polyacrylamide gel after imnoprecipitation by an antiserum against human chorionic gonadotropin. This polyadenylated RNA was used to synthesize cDNA, which was restricted with AZuI or HpaII. With both enzymes, a specific band pattern variable with the RNA used to make cDNA was observed. The 340-base pair cDNA fragment obtained from full-term mRNA is present in human chorionic somatomammotropin cDNA and is present only in minute amounts in first trimester cDNA (Shine et al., 1977). However, AZuI digestion of first trimester cDNA yields segments of 230 and 80 base pairs as major fragments that correspond to either cx- or P-HCG cDNA. The full length cDNA in the size range of 600-750 base pairs was ligated to synthetic HindIII linkers and then cloned into the HindIII site of pBR322. DNA from 22 transformants was analyzed by restriction enzyme cleavage. One recombinant had a 620-base pair-long Hind111 insert, which corresponds to the major cDNA species and contains AluI and HpaII fragments similar to those observed in the first trimester cDNA. Sequencing of this cDNA showed that the entire coding region, including the 24 amino acid presequence and most of the untranslated region, was present in the fragment. In eukaryotes, a precursor polypeptide is generally the immediate product of mRNA translation. The precursor polypeptide is often processed to produce a functional product. Precise posttranslational processing will probably not be accomplished by E. coli. Therefore, in vitro manipulation of primary translational products may have to be carried out for producing certain eukaryotic proteins in bacteria. Nakanishi et al. (1979) have reported the nucleotide sequence of a 1091base pair cloned cDNA insert. This fragment encodes bovine corticotropin-&lipotropin precursor mRNA. DNA sequencing has been reduced to a routine exercise owing to the advances in methodology (Maxam and Gilbert, 1977; Sanger et al., 1977a,b; Weissman, 1979) and has been used to confirm the identity of many DNA sequences that code for eukaryotic proteins. Nucleotide sequences for insulin, rat growth hormone, and human somatomammotropin have been identifed by direct sequencing of

16

VEDPAL SINGH MALIK

cloned DNA segments and comparison of the nucleotide sequence with that predicted by the amino acid sequence of its protein product (Ullrich et aZ., 1977; Seeburg et al., 1978b; Shine et al., 1977). Recombinant DNA technology has made possible the examination of DNA from eukaryotes in genetically and biochemically well-characterized microbes such as E . coli. Genetic and biochemical analyses that were possible only for E. coli genes are now possible for many cloned genes. Fritsch et al. (1979) have already taken initial steps toward localizing sequences that are involved in controlling the switch from fetal to adult fi-globin-like gene expression in man. They used complementary DNA and genomic DNA fragments as probes in Southern hybridization experiments to map the end points of deletions in DNA from individuals with hereditary persistence of fetal hemoglobin and GB-thalassemia. The possibilities of cloning eukaryotic genes in an organism amenable to genetic analysis may allow selection for metabolic mutants of eukaryotic genes and the testing of the regulatory effect and function of the mutant genes. 2 . Complementary RNA as a Probe Development of gene-specific probes allows the selection of the recombinants containing the sequences of interest. A gene-specific probe functions by specific hybridization; for 'example, ribosomal RNA labeled to a high specific activity with lZsI hybridizes to ribosomal genes and is therefore a specific probe. Ribosomal RNA has been iodinated (Commerford, 1971)and successfully used as a probe to detect ribosomal gene sequences in mouse (Amheim, 1979), Dictyostelium discoideum (Cockburn et al., 1976), Bacillus thuringiensis (Klier et al., 1979; Taylor, 1979), Euglena gracilis (Graf et al., 1980). New affinity chromatography, using cDNA containing recombinant plasmid DNA as probe and reverse phase chromatography, allows enrichment so that single mammalian genes together with adjacent sequences can be cloned (McFarland and Borer, 1978; Miura et al., 1979; Persson et al., 1979; Sodja and Davidson, 1978; Shih and Martin, 1974; Taylor, 1979). Affinity columns of nucleic acids bound to cellulose can be used to fish out complementary nucleic acid strands. Column effluent denaturing and recycling through the column yields high efficiency of hybridization. Column-bound RNA can be used as a probe to isolate a complementary DNA, which can then be used to isolate the other complementary DNA strand. Annealing of separately purified complementary strands produces DNA duplexes (Woo et al., 1977a). The 3'-poly(A) tail on most eukaryotic mRNAs can be hybridized to column-bound poly(U) or poly(aT). The enzymes A exonuclease and E. coli exonuclease I11 are used to digest DNA from the 5'-end and 3'-end, respec-

RECOMBINANT DNA TECHNOLOGY

17

tively, exposing the beginning and end of a coding strand. The middle of a gene can be hydrolyzed by pancreatic deoxyribonuclease and exonuclease 111. These single-stranded DNA segments are hybridized with mRNA and isolated. A specific DNA fragment hybridized with mercuriated RNA binds to sulfhydryl agarose. It can be eluted with mercaptoethanol bound to POly(U) agarose and subsequently eluted with ribonuclease (Dale and Ward, 1975; Longacre and Mach, 1979; Longacre et al., 1977; Brown and Balmain, 1979; Beebee and Carty, 1980). Ricciardi et al. (1979) have documented methods for isolating specific mRNAs and for mapping them to the complementary DNA sequences. Total cytoplasmic RNA is hybridized to DNA restriction fragments that have been immobilized on nitrocellulose filters after fractionation on agarose gels. The RNA species that are so selected by Southern hybridization are eluted and their products identified by translation in a cell-free system. This procedure permits preparative purification of mRNAs that are further defined by their ability to direct the synthesis of specific polypeptides (Zubay, 1977). Liu et al. (1979) have developed a microinjection method for detecting translation products from small amounts of mRNA. They were able to detect the synthesis of human fibroblast interferon thymidine kinase, hypoxanthine phosphoribosyl transferase, adenine phosphoribosyl transferase, and propionyl-CoA carboxylase in response to mRNA injected into eukaryotic cells. Using this method, these authors followed the purification of interferon mRNA sequences on sucrose density gradients. This method should aid in isolation of those mRNA sequences for which a biological, immunological, autoradiographic, or selective assay exists. This sensitive assay requires extremely small amounts of material and time. 3. Hybridization Arrest of Complementary RNA Translation In uitro cell-free protein-synthesizing systems using wheat germ embryos can be used to translate mRNA into proteins. The proteins are identified by urea polyacrylamide gel electrophoresis and specific antibody reaction. A bank of clones containing cDNA is generated. Their recombinant plasmid DNA is sonicated, denatured, and hybridized to the enriched mRNA mixture to be translated. The recombinant plasmid DNA that will arrest the translation of one of the mRNAs into the desired protein is further characterized with respect to its DNA sequences and can be further used as a probe for purifying more mRNA or DNA. When a specific mRNA cannot be purified sufficiently for it to be used as a probe, hybridization arrest of RNA translation can be employed to select specific DNA sequences (Patterson et al., 1977). Kronenberg et al. (1979) inserted a DNA copy of mRNA coding for bovine preparathyroid hormone into the PstI site of plasmid pBR322 by

18

VEDPAL SINCH MALIK

using the poly(dG)-poly(dC) homopolymer extension tails. Recombinant plasmids coding for preparathyroid hormone were identified by the ability of the specific plasmid DNA to arrest the translation of preparathyroid hormone mRNA. 4 . R-loop Hybridization

RNA can hybridize to double-stranded DNA in the presence of formamide by displacing the identical DNA strand (Thomas et al., 1976). The threestranded complex so formed is called an R-loop and can be separated by CsCl density gradient centrifugation. Using this method in combination with preparative agarose gel electrophoresis, Tonegawa et aZ. (1977) obtained a 350-fold enrichment of the immunoglobin gene coding for a mouse A chain variable region. This enriched DNA was cloned into a phage vector. Cech and Rio (1979) used R-loop hybridization and electron microscopy to map RNA transcription products of the extrachromosomal ribosomal RNA genes of Tetrahymena thermophila. In order to obtain an efficient rate of RNA-DNA hybridization, the reaction's conditions must be near to or above the melting temperature of the DNA sequences that will hybridize with RNA. White and Rosbash (1979) have constructed a vector containing a sequence high in G C content. This vector with a high strand-separation melting temperature has been designed for gene isolation by forming R-loops. When heated to a temperature at which the majority of DNA sequences are fully denatured, recombinant molecules stay together, owing to high G + C content of the vector. However, RNA is annealed to denatured regions and is retained in the form of an R-loop when the temperature is lowered. DNA molecules containing R-loops can be further fractionated by chromatography on poly(U)Sepharose columns, or on anti-RNA-DNA antibody columns (Rudkin and Stollar, 1977; Wittig and Wittig, 1979).

+

D. SYNTHETIC NUCLEOTIDE PRIMERS When poly(A)-RNA isolated from rat insulinoma is translated in a wheat germ cell-free system, about 25% of the synthesized protein is immunoprecipitated with anti-insulin sera (Villa-Komaroffet aZ., 1978; P. T. Chan et al., 1979). However, restriction endonuclease Hue111 analysis of the dscDNA prepared from total poly(A)-RNA suggests that not more than 2% of the cDNA represents insulin-coding sequences. To overcome the difficulties of obtaining bacterial clones containing full-length insulin cDNA inserts, P. T. Chan et al. (1979)constructed a highly specific cDNA probe corresponding in sequence to the 5'-region of rat insulin mRNA. This was achieved by using a specific deoxydecanucleotide as a primer for the reverse transcription

RECOMBINANT DNA TECHNOLOGY

19

of total rat insulinoma poly(A)-RNA. The synthetic decanucleotide primer d(C-C-T-C-C-A-G-C-A-G)corresponded in sequence to the region coding for amino acids 11-13 plus the first nucleotide of the codon for residue 14 of the insulin B chain as documented by Ullrich et al. (1977).Colony hybridization with decanucleotide-primed cDNA was used to identify seven clones coding for rat insulin I mRNA and nine to rat insulin I1 mRNA. Mevarech et al. (1979)have used a similar approach to isolate an mRNA species that is present at less than 1%concentration in poly(A)-containing tissue fractions. They utilized synthetic nucleotide primers to detect and selectively transcribe gastrin mRNA into cDNA (Noyes et al., 1979). Short oligonucleotide primer could be synthesized because of the knowledge of the amino acid sequence of gastrin. This method has wide applications and provides a new strategy for identifying important genes whose mRNA is less abundant. The method is applicable also for isolating chromosomal genes.

E. CHEMICAL SYNTHESISOF GENES Work on the chemical-enzymatic synthesis of bihelical DNA began in the middle 1960s (Khorana, 1978). The total synthesis of the biologically functional tyrosine suppressor transfer RNA gene of E . coli is an achievement of unparalleled importance in the history of biology and chemistry (Khorana et al., 1976). The total synthesis of the 207 base pair-long DNA included the control elements, as well as the EcoRI restriction endonuclease-specific sequences at both ends (Ryan et al., 1979a). This synthetic gene had amber suppressor activity in vivo when introduced into a suitable E. coli host after being inserted into a double amber mutant of bacteriophage A (Khorana, 1979). During the last decade, Narang and his associates have developed the triester method for the unambiguous chemical synthesis of short deoxyribooligonucleotideswith defined sequences of bases (Narang et al., 1979; Itakura et al., 1973). The simplicity, efficiency, and speed of this method have already enabled the synthesis of many biologically useful nucleotide sequences such as lac operator DNA (Bahl et al., 1977, 1978),linkers containing restriction sites of endonucleases (Bahl et al., 1976a), somatostatin gene (Itakura et al., 1977), tridecadeoxyoligonucleotideinhibitor of Rous sarcoma virus (Zamecnik and Stephenson, 1978), and human insulin gene (Hsiung et al., 1979; Goeddel et al., 1979a; Sung et al., 1979).These results indeed establish the possibility of gene synthesis if the amino acid sequence of the protein product is known and the size of the gene is not extremely large. The mammalian hormone somatostatin, which could be useful in treating acute pancreatitis and insulin-dependent diabetes, consists of 14 amino

20

VEDPAL SINGH MALIK

EcoR1,HwX pwify 77 bp fragment from gal

m

FIG.1. Construction of a plasmid (pHGH107) for the bacterial expression of human growth hormone. pHGH3 (10 pg) was cleaved with EcoRI and HaeIII restriction endonucleases and the 77 base pair (bp) fragment containing coding sequences for HGH amino acids 1-23 was isolated from an 8% polyacrylamide gel. The plasmic pHGH31 (-5 pg) was cleaved with HaeIII. The 551 bp HGH sequence and a comigrating 540 bp HaeIII fragment of pBR322 were, purified by gel electrophoresis. Subsequent treatment with XmaI cleaved only the HGH sequence, removing 39 base pairs from the 3’-noncoding region. The resulting 512 bp fragment was purified from the 540 bp pBR322 HaeIII piece by electrophoresis on a 6% polyacrylamide gel. Next, 0.3 p g of the 77 bp EcoRI-HaeIII fragment and 0.3 p g of the 512 bp HahIII-Xmul fragment were polymerized with T4DNA ligase in a 16 p1 reaction for 14 hours at 4°C. The mixture was heated at 70°C for 5 minntes to inactivate the ligase, then treated with EcoRI (to

RECOMBINANT DNA TECHNOLOGY

21

acids. Somatostatin is produced in the hypothalamus at the base of the brain and inhibits the secretion of several hormones including growth hormone, insulin, and glucagon. Somatostatin was originally obtained in milligram quantities by extracting the brain tissue of a half million sheep. The synthesis of milligram quantities of somatostatin by a few liters of E. coli implanted with synthetic gene was the first demonstration of the utility of recombinant DNA technology for producing human peptides in E. coli (Itakura et al., 1977). The DNA sequence for somatostatin could be chemically synthesized because of the knowledge of the amino acid sequence of somatostatin. This synthetic somatostatin gene was inserted at the carboxy-terminal end of the P-galactosidase gene of a plasmid vector. The carrier P-galactosidase was purified from E. coli cells carrying the synthetic somatostatin gene. Somatostatin was recovered from the carrier protein by treating the hybrid molecules with cyanogen bromide, which cleaves the peptide bond at a methionine residue. The residue has been inserted preceding the first amino acid in the somatostatin peptide. The tertiary structure of Pgalactosidase-somatostatin-fused protein prevented proteolytic degradation of the somatostatin. Achievement of the microbial synthesis of a polypeptide with the size and immunological properties characteristic of mature human growth hormone is a fine example of the sophistication that recombinant DNA technology has already reached (Goeddel et al., 1979a). Human growth hormone (HGH) consists of 191 amino acids. A precursor consisting of a signal peptide attached to the N-terminus of growth hormone is produced in the anterior lobe of the pituitary. Genentech scientists used the knowledge of the restriction endonuclease map and sequence of cDNA of HGH to tailor a bacterial plasmid that directs synthesis of mature HGH in an E. coli cell (Fig. 1). cleave fragments that had dimerized through their EcoRI sites) and with SmaI (to cleave X m I dimers), yielding a 591 bp fragment with an EcoRI “sticky”end and a SmaI “blunt” end. After purification on a 6% polyacrylamide gel, approximately 30 ng of this fragment were obtained. The expression plasmid pGH6, containing tandem lac UV-5 promoters, was treated successively with HindIII, nuclease S1, and EcoRI; then purified by gel electrophoresis, which removes most of the promoter sequence of the tcRgene, but leaves the structural gene intact. A portion (So ng) of the resulting vector, which had one EcoRI “sticky” end and one “blunt” end, was ligated to 10 ng of the 591 bp HGH DNA. The ligation mixture was used to transform x’’’~. Colonies were selected for growth on tetracycline (12.5 gmlml). Since the tcR promoter is no longer functional, tetracycline resistance is dependent upon transcription from the lac promoters reading through the HGH gene sequence into the tcRgene. Approximately 400 transformants were obtained. Colony screening by the filter hybridization procedure identified 12 colonies containing HGH sequences. The plasmids isolated from three of these colonies gave the restriction patterns expected for the correctly assembled gene when cleaved with HaeIII, PouII, and PstI. The DNA sequence of one clone, pHGH107, was determined. Reproduced with permission from Goeddel et al. (197913).

22

VEDPAL SINGH MALIK

Sequence coding for amino acids 23 and 24 as well as the 3'-noncoding region of HGH contains Hue111 sites. Restriction digestion of HGH cDNA with HaeIII yields a 551-base pair fragment, which codes for amino acids 24-191 of HGH. Genentech scientists constructed DNA sequences coding for HGH by fusing a 551-base pair cDNA fragment with a chemically synthesized DNA fragment containing an ATG initiation codon and a coding sequence for amino acids 1-23 of HGH. This synthetic-natural hybrid gene was transcribed in E. coli from a lac promoter (see Fig. 1).Since most E. coli proteins do not contain N-terminal methionine, the formyl-methionine was removed, yielding 2.4 mglliter of human growth hormone that was stable and could be recovered by breaking cells open. Martial et al. (1979) also cloned and sequenced cDNA containing the entire coding and some of the noncoding portion of HGH mRNA. The desired mRNA was isolated from acromegalic tumors. The polyadenylated RNA from tumors that had the most HGH mRNA, as determined by translation assay in the wheat germ protein synthesizing system, were used to synthesize cDNA. This cDNA, highly enriched in full-length copies of HGH mRNA as judged by molecular weight and restriction enzyme analysis, was inserted in pBR322. For growth hormone expression, the cDNA was placed under the control of t r p regulatory elements by inserting it into a vector that contains the regulatory region, the first structural gene (WE)and 15%of the second gene (trpD)of the E . coli trp operon. A fusion protein containing the amino-terminal region of the trpD protein, amino acids coded by the 5'untranslated portion of HGH mRNA, the 26 amino acids of the signal peptide, and all of the amino acids of the growth hormone constitute 3%of total bacterial protein when cells are derepressed for trp transcription (Martial et al., 1979). Several fused proteins such as P-lactamase rat proinsulin (Villa-Komaroff et al., 1978), P-lactamase rat pregrowth hormone (Seeburg et al., 1978a), and P-galactosidase-chickenovalbumin (Fraser and Bruce, 1978) have been expressed by cloning cDNA. The fused proteins are still recognized by specific antibody but they have not yet been processed to yield natural product. However, cloning of A repressor (Backman et al., 1976)and mouse dihydrofolate reductase (Chang et al., 1978; Alt et al., 1978) yielded natural gene product. These proteins are probably initiated at the initiation codon of the mRNA derived from their specific genes.

IV. Making Recombinant DNA Molecules in Vitro In all the productions of nature or of art, what already exists potentially is brought into being only by what exists actually.

ARIST~TLE(FOURTHCENTURY BC)

RECOMBINANT DNA TECHNOLOGY

23

A. HOMOPOLYMER TAILING

In 1970, H. Gobind Khorana and co-workers discovered that DNA ligase produced by E. co2i virus T4catalyzes end-to-end linkage of DNA (Lehman, 1974). Khorana’s laboratory exploited T4 ligase to link short synthetic sequences of nucleotides with single-strand projections into longer DNA pieces. Now using C02+as a cofactor, flush-ended DNA or DNA with 5’sticky ends works as a primer for terminal transferase (Roychoudhury et al., 1976a,b). DNA of various sources can also be treated with A exonuclease to remove a small number of nucleotides from the 5’-end. Deoxynucleotidyl transferase is then used for stepwise addition of a series of identical nucleotides at 3’-OH ends of a DNA strand. A block of identical nucleotides is added to one population of DNA and a block of complementary nucleotides is added to DNA from another source. The two preparations of DNA are then annealed by hydrogen bonding and covalently linked by the combined action of exonuclease 111, DNA polymerase I, and DNA ligase. If the homopolymer tails are longer than 40 nucleotides, the joint is stable, so the hybrid molecule can be transformed without in vitro ligation. This method will join any two species of DNA irrespective of their size and base sequences of their ends, and has been successfully used for cloning DNA (Jackson et aZ., 1972; Lobban and Kaiser, 1973; Bollum, 1974, 1978; Clarke and Carbon, 1976; Wensink et al., 1974). DNA joining is only intermolecular, since intramolecular circularization is prevented by the homopolymer tailing. The original starting material can be recovered with altered ends only (Hofstetter et aZ., 1976; Roberts, 1978; Backman et al., 1976). However, addition of poly(dG) tails to SmaI- or PstI-cleared DNA regenerates the recognition sequence of the enzyme originally used. Single-strand breaks incurred during DNA preparation can also be a starting point for terminal nucleotidyl transferase, and cloning of such DNA could yield deletions, additions, or replacement at the break point. B. LIGATIONOF RESTRICTIONFRAGMENTS Bacteria possess a unique combination of the two enzyme activities to prevent replication of invading DNA. The cell is able to methylate its own DNA at specific adenine sites. The specific action of cellular enzymes recognizes and degrades any DNA that is not modified by a species-specific methylation pattern. The restriction endonucleases cleave the foreign DNA at specific base sequences that are protected by methylation in the host DNA. These enzymes are widely distributed in bacteria and are of two kinds. Class I enzymes are nonspecific in cleavage and require ATP, magnesium,

24

VEDPAL SINGH MALIK

and S -adenosylmethionine as cofactors for activity. They recognize specific sites on DNA but make a limited number of double-stranded random cuts away from recognition sites. The state of methylation at the restriction site determines the subsequent enzyme action (Smith, 1979). Most class I1 restriction endonucleases recognize and cleave a specific sequence of base pairs in DNA duplex and require only magnesium. These are used for DNA sequencing, isolation, physical mapping, cloning, and structural analysis of highly repetitive satellite DNAs and eukaryotic genomes (Fonty et aZ., 1979). These enzymes have also been used to study site-specific interaction of DNA with protein molecules such as RNA and DNA polymerase, transcriptional repressors, and activators (Nathans, 1979). A majority of enzymes classified as restriction 11 endonucleases are not associated with the restriction-modification system (Roberts, 1976) and many enzymes from taxonomically different organisms recognize the same sequence. Roberts (1976) coined the name of isoschizomers for such enzymes. However, isolation of isoschizomers is no reflection of genetic relatedness since similar isoschizomers occur in gram-negative Hemophilus and gram-positive Cwynebacterium. Several enzymes cleave both double- and single-stranded DNA. It is possible that such enzymes recognize and cleave small DNA duplex regions generated by secondary loops within second-stranded DNA. Some enzymes produce fragments with 5’-single-strand extensions (HpaII, HinI, EcoRI, EcoRII), 3’-single-strand extensions (PstI, HhaI), and flush ends (HaeIII). Enzymes of general use are produced commercially (Table 11). Another method of making cohesive termini for joining DNA molecules depends on the ability of restriction endonucleases to cleave DNA into complementary-ended fragments. The cleavage is at or near an axis of two fold rotational symmetry; a palindrome, where nucleotide sequence on one DNA strand is read 5’ to 3’, is the same as the sequence on the complementary strand read in the 3’ to 5’ direction. Because of the symmetrical palindromic arrangement of nucleotides in the vicinity of the cleavage point, the restriction enzymes produce DNA molecules with cohesive ends. During DNA isolation, most DNA is sheared randomly to molecular weights ranging from 1to 40 x lo6. However, by controlled random shearing of large DNA, fragments of a size suitable for cloning can be obtained (Adam and Zimm, 1977). In this case, a large number of restriction endonucleases cleave DNA at defined points and generate a unique set of discrete fragments that are ideal far cloning. If the restriction site is within the gene, then partially digested DNA or digestion in presence of distamycin can be used (Nosikov et al., 1976). A different restriction enzyme that does not cleave within the gene is another choice. Hybrid molecules cloned in this

TABLE I1

SOME USEFUL RESTRICTIONENZYMES Restriction endonuclease

Sequence recognized

Microorganism

AluI

5‘-A GJC T-3‘ 3‘-T C t G A-5’

Arthrobacter luteus

BamHI

5 ‘ d G A T C C-3’ 3’-C C T A GtG-5’

BociUus amybliquefaciens H

BglI

5’. . G c c N N N NJN G G c . . 3’ 3‘. . C G G NtN N N N C C G . . 5‘ 5 f . .A ~ A cT T . . 3’ 3‘. T C T AGTA.. 5’

BglII

5 ’ d A A T T cat 3’-C T T A AtG-5’

EcoRI

BaciUus globigii B. gbbigii E. coli RY 13 canying plasmid RI

EcoRII

5’- C C A G G-3’ 3‘-G G T C C t - 5 ‘

E. coli RK 22

HaeII

5’-G G k C-3’ 3’-C C t G G-5’

Hemophilus aegyptius

HaeIII

5’-Pu G C G CIPy-3’ 3’-PytC G C G Pu-5’

H. aegyptius

Hincll

5’. . G T PyiPu A C . . 3’ 3’..CAPu?PyTG..5’

Hemophilus influenzae &

Hind111

5 ’ . . A J A G C T T . . 3’ 3’. . T T C G A t A . . 5’

H. influenzae R,, com-’Q

Hinfl

5t.. c..3‘ 3’..C TNAfG..5’

H. influenzae Rf

5’.. G A C G C N N N N N

1 . . . . . . . . . 3‘

HemPhilus gallinamm

3‘..C T G C G N N N N N

N N N N N t...5’

HgaI

ANT

(Continued )

TABLE I1 (Continued ) Restriction endonuclease

Sequence recognized

Microorganism

HhaI

5‘-G c GJC . . 3’ 3‘-CtG c c . .5’

Henwphilus hemolyhcus

HhaII

5’.. G J A N T C . . 3 ‘ 3’..C TNAtG..5’

H. hemolyticus

HpaI

5’-G T TJA A c-3’ 3’-C A AtT T G 5 ’

Henwphilus parainfluenure

HpaII

5t-CJC G G-3’ 3 ‘ 4 G CtC-5’

H. parainjluenzae

MboI

5’.. 3.C A T C . . 3’ 3‘..C T A Gt..5’

MorareUo boois

MboII PstI

S’.. G A A C A N N N N N N N 3’..CTTCTNNNNNNN

N J . . 3’ . ..5‘

5’-cT G c A J G - ~ ~

M. boois

Prwidencia stuartii

3 ’ 9 A C G T C5’ 5’.. GJT C G A C . . 3‘ 3 ’ . . C A C G T t G . . 5‘

Streptomyces albus

CCCJGGG

Smatiu wcescens S,

5‘. . T k G A . . 3’ 3‘. . A G C t T . . 5’

Thennus aquatinrs

XbaI

5 ’ - T k T A G A-3’ 3‘-A G A T CtT-5‘

Xanthrrmonas badrii

XhOI

5’-C.l?. C G A C 3 ’ 3’-G A G C T t C 5 ’

Xanthomonas hulcicolu

SalI

RECOMBINANT DNA TECHNOLOGY

27

way can be digested later by the same enzyme to recover the original DNA segment. Enzymes that make staggered cuts are especially useful. The enzyme DNA ligase is used to covalently link the 3’-OH terminus of one strand to the 5’-phosphate end of a second strand. Thus the enzyme can repair singlestrand breaks such as those present in hydrogen-bonded sticky ends of two DNA preparations that have been cleaved with the same restriction enzyme. DNA ligase from E. coli requires NAD as cofactor, while DNA ligase from T4 and higher organisms utilizes ATP as cofactor. In addition to repairing single-stranded breaks as does E. coli enzyme, T4 ligase can link RNA and DNA strands aligned on a complementary DNA template. Another enzyme RNA ligase joins two single-stranded DNA, or two RNA molecules, or RNA molecule to single-stranded DNA molecules (Sugino et al., 1977). During the course of the ligation reaction, most often the vector anneals back to the original form, which decreases the percentage of real recombinant molecules formed. However, advantage can be taken of the identical cohesive termini produced by BglII and BamHI cleavage. When two various DNAs are ligated in the presence of these two enzymes, 100% of the recombinant molecules are obtained (Blakesley, 1980; Rao and Rogers, 1978). Some vectors contain sites where insertion of foreign DNA leads to loss of antibiotic resistance. The kanamycin determinant of Pkc, is inactivated by insertion at the BglII site. Similarly, pBR322 contains several sites where insertion of foreign DNA inactivates genes for resistance to ampicillin or tetracycline (Boliver et al., 1977; Sutcliff, 1978a). Alkaline phosphatase treatment of the vector prevents self-ligation and consequently increases the proportion of recombinant plasmids (Ullrich et al., 1977). Linkage of two different DNA molecules is favored by a high concentration of DNA molecules. Therefore, after ligation of DNA segments at high concentrations of DNA, the mixture is sometimes diluted so that circularization occurs preferentially (Goff and Berg, 1978, 1979). However, .the rate of circularization decreases as the length of the DNA segment increases. Owing to this effect and to high transformation efficiency with small plasmids, short DNA segments are cloned more efficiently than long ones.

C. BLUNT-END LIGATION Sgaramella et al. (1970) discovered that T4ligase enzyme can join (bluntended) fully base-paired DNA duplexes. Even though blunt-end joining occurs at a much slower rate compared to sealing of single-stranded nicks, it is of considerable importance for joining DNA molecules that lack cohesive ends (Mottes et al., 1977). Any DNA segments obtained by shear, by using two different restriction enzymes or by using restriction enzymes that pro-

28

VEDPAL SINGH MALIK

duce blunt ends, can be joined without adding homopolymer tails (Sgaramella, 1971). DNA polymerase can be used to fill out non-flush-ended DNA duplexes. Blunt-end ligation of longer molecules obtained by shearing DNA at random rather than at specific sequences offers the possibility of cloning intact genes, since the DNA should appear on some fragment. Chemically synthesized DNA sequences can also be linked by this procedure, where it is necessary not to add additional bases, e.g., in cloning of promoters or specific restriction enzyme recognition sequence. However, blunt-end ligation does not allow the recovery of the cloned fragment after the recombinant DNA has been propagated. Backman et al. (1976) used flush-end fusion techniques involving no sequence overlap to construct a plasmid carrying the repressor gene of bacteriophage A fused to the promoter of lac operon. In some cases, the abutting of DNA fragments generated by different restriction endonucleases creates a sequence at the junction that is cleaved by one of the restriction endonuclease (Table 111). These findings expand the spectrum of utility of TABLE 111

USE OF

THE

Restriction endonuclease

EcoRl

FILLING-INMETHOD WITH VAFUOUSRESTRICTION ENDONUCLEASES~,~

Recognition sequence

GJAATTC

Requirement for site regeneration 5‘C

Subsite recognition sequence

Subsite

J AATT

EcoRJ

TTAAf

CTTAAtC EcoRII

JCCAGG

None

JCCACG

EcoRII

GCTCC f Hind111

A~AGCTT

GCTCCt 5‘T

Alu I

GJCATCC CCTACtG

5’C

MboI

AJGATCT

5’T

AGJCT TCtCA

TTCCAtA

Barn1 BgZII

JGATC CTAGf

MboI

JGATC

CTAC?

TCTACtA ~~~

~

Several endonncleases that produce 5’-protruding ends are listed with the DNA sequence they recognize. After filling in these ends with DNA polymerase I, the original recognition sequence can be regenerated by joining the filled-in end to a flush-ended DN molecule that meets the requirement listed in the third column. When joined to a flush-ended fragment not meeting that requirement, the filled-in end generates a subspecificity, described in the fourth and fifth columns. The top line of each recognition sequence reads from the 5’- to the 3’end; the bottom line reads in the opposite direction. Reproduced with permission from Backman et al. (1976).

RECOMBINANT DNA TECHNOLOGY

29

endonucleases and allow recovery of fragments that have been inserted by blunt-end ligation. D. DNA LINKERS

1. Blunt-End Linkers Restriction site DNA linkers for manipulating DNAs that lack cohesive ends (sheared DNA, synthetic DNA, cDNA) are commercially available. The decanucleotide d(C-C-G-G-A-T-C-C-G-G) contains recognition sequences of BamHI, HpaII, and HapII. When joined to blunt-ended DNA by polynucleotide ligase, this linker adds the recognition sequences of these restriction enzymes to both ends of the DNA fragment, which can be inserted into a vehicle by using the restriction enzyme cohesive termini. If the DNA fragment is not flush-ended, it can be first digested with the Aspergillus nuclease S1 to produce even ends (Ghangas and Wu, 1975), since this one linker can be used with several restriction enzymes. An enzyme that does not cleave the DNA fragment to be cloned can be used. The high number of G-C base pairs in this self-complementary restriction site linker gives double-stranded DNA great stability. However, a large DNA fragment is likely to contain many restriction sites and therefore cannot be cloned this way. The improved phosphotriester method has been used (Bahl et al., 197613; Scheller et aZ., 1977) to synthesize the linker sequence for BamHI d(C-C-G-G-A-T-C-C-G-G),EcoRI (C-C-G-A-A-T-T-C-G-G),and HindIII.

2. Preformed Adaptors Such linkers do not require cleavage with restriction enzymes for generating cohesive ends and have a blunt end and a single-stranded cohesive end corresponding to one of the restriction enzyme recognition sites. The adaptor will join to DNA only at the blunt end if the cohesive end lacks 5’phosphate. The DNA joined to the adaptor is phosphorylated before ligating to the vector. Adaptors with a six-nucleotide-longduplex region have been synthesized to contain the recognition sequence for XmaI and S m I and cohesive ends for BamHI (Norris et al., 1979).

3. Conversion Adaptor When a DNA fragment obtained by digestion with one restriction enzyme is inserted into a different restriction site in the vector, conversion adaptors are used. For example, using such an adaptor, an EcoRI-cleaved piece of DNA can be inserted into the BamHI site of the vector. There are two methods of preparing conversion adaptor: (1) blunt-end ligation of two different types of preformed adaptors and (2) synthesis of two

30

VEDPAL SINGH MALIK

different single-stranded decadeoxynucleotides, each containing a sequence of four nucleotides at the 5’-end corresponding to the central part of the recognition sequence of two restriction enzymes, and a six-nucleotide-long d(C-C-C-G-G-G) self-complementary sequence at the 3’-end. Annealing these two decadeoxynucleotides causes a conversion linker to form with a BamHI sticky terminus on one side and an EcoRI, 5’-sticky terminus on the other (Narang et al., 1977). This adaptor can be used to clone a DNA segment that has an EcoRI recognition sequence at one end and a BamHI sequence at the other. Such an adaptor was ligated to the lac operator fragment carrying EcoRI cohesive ends. The lac operator DNA, now car?ing BamHI sticky ends, was ligated to BamHI-cleaved pBR322 plasmid DNA and transformed into E . coli (Marians et al., 1976; Bahl et al., 1976b). The transformed cells were plated on nutrient agar plates containing X-gal (5-bromo-4-chloro-3indolyl-P-D-galactoside). The transformants produced a large amount of 4-galactosidase, which hydrolyzed X-gal and released blue dye 5-bromo-4-chloro indigo. Escherichia coli transformants containing lac operator ligated to the multicopy plasmid pBR322 have all lac repressor bound to the cloned operator and thereby derepress lac operon. 4. Single-stranded Adaptors

Such adaptors can be used to clone a DNA fragment with 3’-protruding ends (PstI-digested DNA into the BamHI, or EcoRI-digested vector with 5’-protruding ends). These linkers are single-stranded decadeoxynucleotides containing a five nucleotide sequence complementary to the 3’-protruding end and the other pentanucleotide sequence complementary to the 5’cohesive end. Two such linkers, the EcoRI-PstI (5’-A-A-T-T-C-C-T-G-CA-3’) and the EcoRI-Hue11 (5’-A-A-T-T-C-G-G-C-G-C-3‘)adaptors, have been synthesized by the triester method. Bahl et al. (1976b) ligated BamHI decamer to a 21-base pair-long synthetic lac operator. Duplex DNA cleavage of this large fragment with BamHI produced a lac operator flanked by BamHI-generated cohesive ends. This operator was then inserted into plasmid pMB9 in the Bum HI restriction site (Rodriquez et al., 1976). Sadler et al. (1977) synthesized single-stranded fragments of the lac operator that, upon annealing, produced a central piece of lac operator flanked by singlestrand sequences similar to those yielded by EcoRI endonuclease digestion. This synthetic lac operator was then spliced into the EcoRI site of pMB9 and transformed into E. coli. About 10% of the tetracycline-resistant transformants had functional lac operator. Functional lac operator has been cloned independently by five groups (Marians et al., 1976; Bahl et al., 1976b; Heyneker et al., 1976; Sadler et al., 1977; and Scheller et al., 1977). Linker molecules have to be so designed that, after linking, triplets are

RECOMBINANT DNA TECHNOLOGY

31

introduced on either side of the DNA to be cloned. This would leave the reading frame for protein synthesis unchanged. A variety of linkers may be available commercially in the near future. These may offer promoters, operators, new restriction sites, ribosome binding sites, chain initiation, and termination sites flanked by restriction enzyme recognition sequences.

V. Transformation Transformation is the process of introducing in uitro-constructed recombinant DNA molecules into the cytoplasm of an organism (venema, 1979). Even though transformation is not common among most bacteria, it occurs naturally in Streptococcus, Azotobacter, Pneunwcoccus, Bacillus, Henwphilus, Neisseria, Acinetobacter, and Thermoactinomyces vulgaris (Deddish and Ravin, 1979; Page and Tigerstorm, 1979; Sisco and Smith, 1979; Morrison and Mannarelli, 1979; Morrison and Baker, 1979; Fuchs and Dobrzanski, 1978). Mandel and Higa (1970) discovered that treatment of E . coli cells with calcium chloride enables them to take up viral DNA. Cohen et aZ. (1972) and Cosloy and Oishi (1973) found that a similar treatment made E . coli cells permeable to purified R factor DNA. This method has been critical for the use of plasmids as cloning vectors and has been used with SalmneZZa (Lederberg and Cohen, 1974), Enterobacter (Taketa, 1972; Momson, 1977), Staphylococcus (Lofdahl et aZ., 1978a,b), and Pseudomonas (Chakrabarty, 1976). Taketa (1972) found that Ca2+-inducedtransformation of E . coli allowed uptake of RNA and single- and double-stranded DNA. Heterologous duplex DNA but not RNA or single-stranded DNA inhibited the transformation by duplex viral DNA. Ca2+could not be substituted for any other cation, but Mg2+ doubled the number of Ca2+-inducedtransformants. Zn2+ severely inhibited transformation. Cells from the exponential growth phase were best for transformation. Different growth media had no effect on transformation, but temperature did. Efficiency of transformation was decreased if cells were grown below 32°C. No transformation occurred if the Ca2+treatment was performed at 3TC, suggesting that lower temperature was necessary for the Ca2+-mediatedbinding of DNA to cell surfxe. Most DNA molecules can transform cells successfully, but only rare cells are competent to take up DNA. Individual competent cells can be transformed by more than one plasmid simultaneously, which allows indirect selection of a cryptic plasmid if transformation is carried out with a mixture of a cryptic plasmid and a small selectable plasmid. High cotransduction fiequencies (80%)with a small cryptic plasmid and pSCl01 have been obtained

32

VEDPAL SINGH MALIK

(Kretschmer et al., 1975). A temperature-sensitive mutant (pSC205) that is eliminated from the cotransformed cells leaves the cryptic plasmic behind for further experimentation. The nature of transforming nucleic acid in calcium-treated E . coli has been studied by Strike et al. (1979). Many alterations of the initial transformation procedure of Higa and Mandel have yielded very high transformation rates (Norgard et al., 1978; Kushner, 1978). Mutants lacking exonuclease and restriction endonucleases are used as recipients to prevent the degradation of incoming foreign DNA. Mutants altered in the cell envelope are also transformed at a high rate. Prolonged incubation of E . coli cells in calcium chloride improves the competence, and 2 x lo' transformants per microgram of pBR322 DNA can be obtained routinely (Dagert and Ehrlich, 1979; Molholt and Doskocil, 1978; Norgard

et al., 1979). Liposomes can be used for introducing specific DNA sequences into cells. Fraley et al. (1979) describe the entrapment of plasmic pBR322 within liposomes and then transfer to E . coli by standard transformation procedure at a frequency of 1% of the control. This frequency was unaffected by the addition of DNAase to the incubation mixture. Procedures for high-frequency transformation of protoplasts of B . subtilis (Tanaka, 1979; Chang and Cohen, 1979; Mottes et al., 1979), Streptomyces (Bibb et al., 1978; Thompson et al., 1980; Suarez and Chater, 1980), Spirillum lipoferum (Mishra et al., 1979), Neurospura crassa (Case et al., 1979; Mishra, 1979), and yeast (Hinnen et al., 1978; Gafner and Philippsen, 1980; Petes, 1980) have now been developed and can be routinely used to introduce recombinant DNA molecules into these organisms. Even though procedures for transformation of industrial strains may differ, base line information is now available to make a start in the direction of introducing foreign genes into commercially important microbes.

VI. Host-Vector Systems Most segments of DNA do not have an inherent capacity for selfreplication and should be linked to DNA sequences that are self-replicating. Genetic signals required for replication are host-specific. If foreign DNA is to be replicated in an organism, it should be ligated to a suitable DNA fragment (vector)that can replicate in that organism. The recombinant DNA molecules consisting of the desired foreign DNA and the vector can be introduced into the host, and cells that have received the DNA chimeras can then be identified. The yield and purity of cloned genes and efficiency of ligation and transformation are greater with small vectors. However, origin of replication, gene insertion sites, and genes useful for selection in trans-

RECOMBINANT DNA TECHNOLOGY

33

formation determine the minimum size of the vector. For easy recovery of the recombinants, an insertion site in a vector gene whose function is easily identifiable is a good attribute (Fig. 2). Then clones containing a plasmid with inserted DNA that inactivates the gene can be selected easily from clones with wild plasmid phenotype. Another desirable property of the vector is its role as a promoter site for initiating transcription across the inserted fragment. Foreign genes can be transcribed from this vector promoter if their normal promoters are not recognized by the host transcription machinery. Promoters can be followed by the ribosome binding sites, initiation codon, and then a restriction site to insert the DNA for obtaining expression.

nRPQ33

FIG.2. Plasmid pBR322. pBR322 is obtained from pBR313 and has a molecular weight of 2.6 x los. pBR322 contains a single PstI restriction site, located within the ampicillin-resistant gene. There are also single sites for the restriction endonucleases EcoRI, HindIII, BamHI, and SalI, all located within the tetracycline-resistant gene. Reproduced with permission from Bethesda Research Laboratories, Inc., Rockville, Maryland.

34

VEDPAL SINGH MALIK

A. Escherichia coli 1. Plasmids Cohen et al. (1973) were the first to construct a small plasmid-cloning vector (pSC101). It carried the DNA sequences essential for replication in E. coli, a selectable gene conferring tetracycline resistance and a single site for EcoRI that was outside regions determining replication of the vector and selectable gene of tetracycline. Plasmid pSClOl was used to clone ribosomal genes from Xenopus laevis (Morrow et al., 1974), histone genes from sea urchin (Kedes et al., 1975), and mouse mitochondrial DNA (Chang et al., 1975). Now, many derivatives (pBM9, pBR322) of plasmid ColE1 have been developed as vectors for cloning foreign DNA into E. coli (Bastia, 1978;Tait and Boyer, 1978; Sutcliffe, 1978b; Helinski, 1979).These plasmid vectors are present as 3 0 5 0 copies per E. coli cell under normal conditions of growth. Their DNA can be amplified to about 45% of total cellular DNA if the E. coli is grown in the presence of chloramphenicol. Consequently, very high yields of both plasmid DNA and any DNA segment inserted into the plasmid can be obtained (Chakrabarty, 1978; Norgard et al., 1979). Rao and Rogers (1978) constructed another plasmid vector, pKC7, by replacement of the HindIII-BamHI fragment of pBR322 with 1.8 kb HindIII-BamHI Tn5 segment derived from A. This vector normally replicates by the plasmid system, but amplification of the vector is induced by thermal induction of the A replication system. This vector confers resistance to ampicillin (100 mg/ml) and kanamycin (100mg/ml) but not to tetracycline. Plasmid vector pKC7 contains several new unique restriction sites in addition to most that are in pBR322 (Rao and Rogers, 1979). Depending on the orientation of insertion of DNA segments in the single BglII site, both kanamycin-sensitive and kanamycin-resistant clones are produced. Similar results have been obtained when DNA is inserted in the HindIII site of pBR322 (Widera et al., 1978). Some pBR322 recombinants carrying DNA inserts in the HindIII site are resistant to tetracycline and others are sensitive, because the BgIII site in pKC7 and HindIII site in pBR322 are in the beginning of the promoter for the kanamycin and tetracyline genes, respectively. Backman et al. (1976) constructed plasmid vectors that carry the bacteriophage A c1 gene and confer immunity to A infection on host bacteria. Using the same plasmid pKB158 selection for tetracycline resistance was about twice as efficient as selection for immunity. Bernard et al. (1979) have constructed a plasmid vector (pHUB4) into which restriction fragments can be inserted so that they are transcribed from

RECOMBINANT DNA TECHNOLOGY

35

the A PL promoters, and the A N gene product stimulates reading through transcription terminators downstream from A PL. A hundredfold increase relative to the wild-type E. coli in w aporepressor level was obtained when a 1.3 kb BamHI restriction fragment containing entire trpR gene was cloned in pHUB4. Uhlin et a2. (1979) derived a vector that has lost the control of its replication above 35°C. This vector replicates without copy number control and its DNA could amount to 75% of the total cellular DNA. During the vector amplification, growth and protein synthesis continue at a normal rate. Final gene dosage decreases with increasing weight of the recombinant plasmid. Degree of amplification is limited by the total amount of plasmid DNA rather than the number of plasmid copies. P-Lactamase was amplified up to 400fold when cloned into this vector. Construction of E. coli plasmid vectors that can be used to isolate promoter active DNA fragments has been reported (An and Friesen, 1979; West et al., 1979). These “promoter-probe” plasmids select for promotercontaining DNA fragments by insertional activation of the tetracycline resistance gene that has an inactive promoter.

2. Cosmids Cosmids are plasmids usable as gene-cloning vectors in an in vitro packaging system by coliphage A (Collins, 1979; Collins and Hahn, 1978; Collins and Bruning, 1978). These plasmids carry the A bacteriophage COS site, which allows the plasmid, including large inserted DNA fragments, to be packaged as pseudobacteriophage A particles. Such particles adsorb to recipient E. coli and introduce the large recombinant DNA into the host. Original cosmids are very small and therefore are not packaged as bacteriophage A particles, which allows selection of recombinant molecules of a certain size only. Since the average size of the cloned fragment is comparatively large (10 megadaltons), fewer recipient clones are screened for the desired gene without a selective marker (Mayer et al., 1979).

3. Bacteriophage A Phage A (Fig. 3) has been well characterized physically, genetically, and chemically (Szybalski and Szybalski, 1979). Most A vectors are unable to lysogenize, since they have deletions for att+, int+, or c 1genes essential for lysogeny. Bacteriophage A unable to establish lysogeny can be converted into the plasmid state at a low frequency. Furthermore, A deleted for int+ and att+ genes can lysogenize E. coli at secondary sites at low frequency. However, about one-third of phage A can be replaced without any impairment of the lytic cycle. The upper limit of DNA that can be packaged into the phage head is about 105% of the normal A genome. The lower limit is

36

VEDPAL SINGH MALIK

c0.L

LEFT ARM

m Nu1

J

MATURE PHAGE

0.a'

b mqion

attP

5'G

RIGHT ARM oosR

N cI OP

R

r!~'

3'

CO.

.

c!. .o.ttL b.0'

N

CI

OP

R

cos Nu1

J

b region

attR I....

PROPHAGE

bio I .

o'b'

FIG.3. A simplified genetic and physical map of the bacteriophage h genome. The heavy lines represent the complementary 1 arid r DNA strands. The genome is presented in the form of an open circle obtained by bending downward the left and right termini of the linear A DNA molecule (drafted above the circle), according to the convention. Strand 1 is transcribed leftward (counterclockwise) and has a 5 ' 4 at its left cohesive terminus, m. Strand r is transcribed rightward (clockwise) and has a 5'-A at its right cohesive terminus, n'.Strand r displays a higher density in the CsC1-poly(U,G)gradient and a lower density in the alkaline CsCl gradient than strand 1. The shaded arrows represent the various leftward and rightward transcripts, always placed next to the complementary coding strand. The prophage map (drawn at the bottom) is a circular permutation of the linear map of mature DNA (drawn at the top). The RNA polymerase binding sites are represented by small black circles near the center of the drawing. The approximate positions of the ribosomal binding sites (short radial bars near the center) were determined only for the exo-m' portion of A. Reproduced with permission from Szybalski and Szybalski (1979).

RECOMBINANT DNA TECHNOLOGY

37

about 80%. Therefore, a A vector with deletion of 20% of its own DNA can accept an insertion amounting to 25% of the total length of A DNA, which is about 50,000 base pairs. Phage A vectors usually lack a restriction endonuclease target in the essential part of the genome and have one restriction site in the dispensable region. Certain A vectors have two restriction sites in the nonessential region. This, then, permits insertion of foreign DNA by replacement of A DNA. Plaque type of these recombinant phage is altered which can be visually screened (Murray et al., 1977). Foreign DNA can also be inserted into A vectors so that its expression is from the very efficient A leftward promoter PL. In such recombinants, phage A N gene product allows RNA polymerase to override transcriptional stop signals. Regulatory mutants in A gene Q can be used to prevent cell lysis and expression of the late A genes. Following infection of E. coli, a A phage carrying tryptophan synthetase gene, constructed by such reasoning, produced tryptophan synthetase as a very high proportion of total cell protein (Moir and Bremmar, cited in Murray, 1976; Crawford and Stauffer, 1980; Guarente et al., 1980). Phage A vectors in which the inserted DNA fragment is transcribed from A PL or lac promoter have been constructed by Blattner et al. (1977).These A vectors are named after Charon, the mythological boatman of the River Styx. Some of these phages (charons 8 and 9) can be used to clone up to 25,000 base pairs of DNA fragments produced by EcoRI, Hind111 and PstI. Other charon phages have restriction sites for two enzymes and can be used to insert DNA digested with pairs of enzymes, which prevents self-cyclization of heterogenously terminated vectors and fragments. Some charon vectors can be converted into viable phage only by incorporating foreign DNA, and formation of a plaque suggests successful cloning. Following insertion, charons 6 and 7 produce clear rather than turbid plaques. Charon 12 is unable to plaque on pol A strains of E. coli if foreign DNA is inserted into the A red gene. When DNA is inserted into 2acZ gene of A lac-5 derivative charons (2 and 16), colorless rather than colored plaques are produced on lac-indicator bacteria. Several other A vectors have been constructed (Skalka, 1978; Enquist et al., 1976; Donoghue and Sharp, 1978). 4 . Filamentous Single-Stranded Phages

Since the length of the virus particle is determined by the length of virus DNA, these viruses can accommodate almost any size of inserted DNA. Such recombinant viruses may or may not be defective. Furthermore, the rapid DNA-sequencing technique of Sanger and Coulson (1975) utilizes single-separated DNA strands. A new class of cloning vector, the singlestranded bacteriophage M13, can be used as a ready source of single-strand

38

VEDPAL SINGH MALIK

DNA of target DNA sequence (Denhardt et al., 1979). Barnes (1979) has constructed a M 13-histidine transducing cloning phage, known as M 13Hog, which has a single EcoRI site and amarker (hisD)for auxotrophic selection. Gronenborn and Messing (1978) have constructed an M13 lac cloning vehicle that has a single EcoRI site in the ffih codon of the P-galactosidase gene. Two hundred copies per cell of the replicative form of this vector occur as double-stranded DNA (Buchel et al., 1980). Cells never lyse but excrete many copies of phage single-stranded DNA. DNA sequences inserted in the EcoRI site are transcribed from lac promoter and translated when in phase. If inserted DNA has its own promoter, then fusion products to the N-terminal 130 residue of P-galactosidase are produced. These can be assayed by in vitro complementation with inactive mutant A-M15 P-galactosidase (Langley et al., 1975).

B. Bacillus subtilis Bacillus species produce a variety of different enzymes and antibiotics and are therefore important in the fermentation industry (Yoneda et al., 1979). Unlike E . coli, Bacillus lack endotoxins in their architecture. Genetic manipulation of Bacillus can be achieved by high-frequency transformation. Several generalized transducing phages of B . subtilis are also known. It has about 200 genes well mapped on its chromosome (Hoch, 1976). Plasmids

Several Bacillus species harbor small cryptic plasmids, (Lovett and Burdick, 1973; Bernhard et al., 1978; Bingham et al., 1979; Gonzalez and Carlton, 1980). Furthermore, many drug resistance plasmids of Staphylococcus aureus can replicate and express their drug resistance genes in B . subtilis (Ehrlich, 1977, 1978; Iordanescu, 1975; Gryczan et al., 1978; Gryczan and Dubnau, 1978; Chi et al., 1978; Canosi et al., 1978; Murphy and Novick, 1979). Development of the S . aureus plasmid as vectors now allows cloning of DNA sequences in B . subtilis. Specific segments of B . subtilis chromosome can thus be maintained in an extrachromosomal state. This will allow gene purification and complementation of regions of B. subtilis chromosomes not amenable previously to such analysis (Haldenwangand Losick, 1979; Segall and Losick, 1977). A colony bank of E . coli containing hybrid plasmids representative of the B . subtilis genome has been constructed (Rapoport et al., 1979). The initial cloning of B . subtilis chromosomal fragments in B . subtilis used 3-megadalton neomycin-resistance plasmid pUBllO as vector. Keggins et al. (1979) used plasmid pUBllO as a vector to clone fragments of Bacillus

RECOMBINANT DNA TECHNOLOGY

39

DNA that complement trp mutations in B. subtilis. The levels of the plasmid-specified trp enzymes in B. subtilis were higher than the repressed level of chromosomally specified trp enzymes. Certain cloned trp fragments had a single HindIII restriction site. Insertion of HindIII-generated DNA sequences into such plasmids abolished trpC complementing activity and reduced by tenfold the plasmid-specified trpF product. These results suggest that tryptophan cluster in B. subtilis is transcribed as a unit in the direction of trpE trpA. Gene cloning was used to determine polarity of an operon by Crabeel et al. (1980). C. STREPTOMYCETES A change must sometimes be made to things one is not accustomed to.

HIPPOCRATES (FIFTHCENTURY, BC)

Antibiotics are the best known examples of secondary metabolites produced by streptomycetes. Since some of them are toxic to their producers, the producers have evolved mechanisms that protect them against this toxic effect. The formation of secondary metabolites is usually repressed during logarithmic growth and is depressed during the stationary phase. In organisms that produce secondary metabolites, certain primary pathways are not well regulated and the overall metabolism lacks coordination, which results in the accumulation of intermediates and end products of primary metabolism. To prevent this accumulation, many regulatory mechanisms have evolved among microorganisms. For example, E. coli switches off its operons if the concentration of intercellular metabolites exceeds certain levels. Industrial organisms, such as streptomycetes, lack coordinated regultion of metabolism but have evolved an alternative strategy. They developed subsidiary pathways to dispose of intracellularly accumulated toxic levels of metabolites. Genes determining such subsidiary pathways are superimposed on growth and their expression is overridden by a growth-linked regulatory mechanism (Malik, 1979a). It is only during the suboptimal or stationary growth phase when the growth-linked overriding regulatory mechanism is abolished that the expression of genes involved in the antibiotic synthesis commences. The molecular nature of this overriding mechanism and the induction process involved in antibiotic synthesis are not understood. In spite of the production of many metabolites of commercial value, the genetics and biochemistry of streptomycetes have been poorly studied. However, the development of a cloning system into streptomycetes could increase the understanding of the regulatory biology of these industrially important organisms. Once the regulatory mechanisms that control antibiotic production are understood, this knowledge can be used to derepress

40

VEDPAL SINGH MALIK

biosynthetic pathways determining antibiotic biosynthesis instead of the present trial-and-error approach by media manipulation. As a result, many potential producers of antibiotics (and other secondary metabolites) are discarded as nonproducers because they were grown in media that do not allow transcription of genes involved in antibiotic production. If regulatory elements, e.g., operators, involved in pathways coding for antibiotic synthesis can be cloned, then these elements can be introduced back into the parent organism on a multicopy plasmid so that the repressor that prevents transcription of antibiotic-synthesizing genes will be neutralized and the corresponding genes will be expressed. Such operators can also be ligated to ribosomal RNA genes and inserted into the chromosome at many locations, since many copies of ribosomal RNA genes are present per haploid genome. Ribosomal RNA genes provide the DNA homology necessary for recombination and integration into the chromosome. In B . subtilis and E . coli, ribosomal genes are near the origin of replication and are amplified during the growth phase, when many replicating forks of chromosome are present. Many genes that regulate growth are present in the vicinity of the ribosomal genes. Since antibiotic production is inversely related to growth rate, study of the genes of the ribosomal cluster could provide clues to the regulation of antibiotic synthesis. It is, therefore, very timely that the gene clusters that determine ribosome synthesis in streptomycetes be isolated, sequenced, and further studied. 1 . Plasmids

Any streptomycete plasmid or actinophage may probably be used as a vector to clone genes in streptomycetes. Unfortunately, all streptomycete plasmids isolated to date are cryptic, and hence have no selectable marker. Once a streptomycete gene has been identified, it can be prepared in large amounts and used to transform streptomycetes after ligation to DNA fragments of actinophage or plasmid that can replicate in streptomycete. Streptomyces erythreus contains a ribosomal RNA methylase that may be another marker conferring resistance to erythromycin, lincomycin, and streptogramin B type antibiotics. Eventually, such a gene can be used as a selective marker to isolate the origin of replication of a streptomycete plasmid or phage. The origin of replication of actinophage or streptomycete plasmid ligated to the ribosomal RNA genes and the DNA sequence-complementing auxotrophs may be a suitable vector for cloning genes in streptomycetes. Numerous cryptic plasmids have been isolated from streptomyces (Malik, 1977, 1979b; Hayakawa et al., 1979; Nojiri et al., 1980; Okanishi et al., 1980). The SCP2 is a self-transmissible plasmid of the 20 X 106daltons that promotes chromosomal recombination in Streptomyces coelicolor and regulates production of and resistance to antibacterial substances (Troost et al.,

RECOMBINANT DNA TECHNOLOGY

41

1979).This plasmid can replicate stably in Streptmyces parvulus and Streptomyces lividans. The latter also harbors several plasmids (SLP,-SLP6) that have a molecular weight of more than 6.25 x lo6. Streptomyces lividans plasmids are primed with host genome sequences of various lengths. Purified DNA of SCP2* and SLP can be introduced in the protoplasts of plasmidless strains of S . coelicolor, S . lividans, and S . parvulus ATCC 12434. The transformants are detected after segregation of polyethylene glycol-treated protoplasts (Ochi et al., 1979; Bibb et al., 1978). Plasmidcontaining clones produce a zone of growth inhibition due to antibiotic production active against plasmid-free cells, a trait exhibited by SCP2* and SLP, into S . coelicolor, S . parvulus, and S . lividans (Troost et al., 1979). Foreign DNA sequences can now be cloned in these industrially important organisms. Plasmids (SLP1-SLP6) have BamHI and PstI restriction sites where foreign DNA can be inserted without inactivating the lethal zygosis property or replication functions (Suarez and Chater, 1980; Thompson, et al., 1980). With the knowledge of such restriction sites, cloning of foreign genes into streptomycetes using these plasmids as vectors can now proceed. Several vectors that can be used to clone genes in streptomyces have recently been developed. Fragments of S . coelicolor genome carrying methylenomycin resistance (M. Bibb, personal communication) and various prototrophic alleles (D. A. Hopwood, personal communication), have already been cloned, using SCP2* as a vector and S . coelicolor as a host. Such clones are highly unstable but can be stabilized if a fragment of SCP2* that carries a specific plasmid segregation function is included in the recombinant plasmid. In another series of experiments, Hopwood and his co-workers shotgun-cloned antibiotic resistance genes from Streptomyces j-adiae and Streptomyces azureus into S . lividans. The S . lividans plasmid SLP1.2 was used as a vector to clone neomycin-3-phosphotranseferase gene and neomycin-3-acetyltransferase gene from S , fradiae into S . lividuns. Using the same vector, thiostrepton resistance and erythromycin resistance from S . azureus and S . erythreus have also been cloned in S . lividans. These antibiotic resistance markers can now be inserted into other vectors or combined to build vectors with multiple resistances to generate a genetransporting vehicle analogous to E . coli-cloning vector pBR322.

2 . Actinophages Various actinophages that grow on Streptomyces have been described (Chater and Carter, 1978, 1979; Hranueli et al., 1979; Klaus et al., 1978, 1979; Sladkova et al., 1979; Lomovskaya et al., 1980).Vectors with repressor gene of actinophages ligated to streptomyces plasmid could be worth constructing. Transformants could be selected by virtue of their resistance to

42

VEDPAL SINGH MALIK

lysis by actinophages, and commercial large-scale fermentations with Streptomyces harboring such plasmid-actinophage repressor vectors could be resistant to lysis by actinophages. The temperature actinophage, (31, has been developed as a suitable Streptomyces cloning vector (K. Chater, personal communication). Like all known Streptomyces temperate phages in size and morphology, this actinophage resembles E . coli phage A and contains sticky ends. A 20-gene linkage map has been constructed by using temperature-sensitive plaque morphology and host range mutants. Heteroduplex analysis, denaturation, and restriction enzyme digestion have been used to align the linkage map with the physical map of the linear 39 kb DNA of C31. Repressor gene (C) is located in the middle of the genome and many genes for essential functions are located to the left of the C gene. Actinophages that are resistant to chelating agents have deletions in the center of the genome-removing repressor gene or in the rightmost 25% of the molecule-removing phage attachment site to the chromosome. Partial EcoRI digests of E . coli plasmid pBR322 and actinophage deletion mutant C31Cts 23 were ligated and transfected in S. lividans. Transfectants were selected by hybridizing clones to radioactive pBR322 DNA. The resulting C31-pBR322 recombinant molecule replicates as a phage in Streptomyces and as a stable plasmid in E . coli. Further modifications have produced a vector that has single restriction sites for BamHI and PstI and can be used to clone up to 7 kb of foreign DNA. A C31-pBR322 hybrid is restricted by Streptomyces albus P and does not form plaques on R-m+ S. albus G mutants if part of pBR322 is inserted in C31. The latter effect may be used to select deletion mutants. Ampicillin- and tetracycline-resistancegenes of pBR322 may be expressed in Streptomyces if inserted in such a manner that they are transcribed from an actinophage promoter. Chloramphenicol acetyl transferase gene from E . coli has also been cloned in S. Zividam using S . lividans plasmid SLP1.2 as a vector. This enteric bacterial gene is expressed in S. Ziuidans and its transcription may initiate from the promoter of Streptomyces vector plasmid (M. Bibb, personal communication). These cloned antibiotic resistance genes provide a good selective marker for cloning DNA sequences in Streptomyces. The convenient assay of chloramphenicol acetyltransferase and P-lactamase can be further used to understand the molecular biology of Streptomyces promoters. As a consequence, the time has finally arrived when systematic study of the molecular biology of Streptomyces can be exploited to generate hybrid antibiotics, to introduce useful genes for utilization of cheap carbon sources (e.g., cellulose degradation), and for production of valuable human metabolites (e.g., insulin, interferon). Multicopy vectors with highly efficient constitutive promoters will soon be available to amplify and express genes of

RECOMBINANT DNA TECHNOLOGY

43

interest in Streptomyces, making them very attractive for producing valuable metabolites. Certain thermophilic Streptomyces that grow to a very high cell mass on cheap carbon sources are another promising host for producing valuable metabolites (e.g., amino acids, organic acids, vitamins, enzymes, antibiotics, coloring pigments) at a reduced cost.

D. YEAST Several hybrid vectors that replicate in both E. coli and yeast are now available (Broach et al., 1979; Farabaugh and Fink, 1980). Hinnen et al. (1978) were the first to conclusively demonstrate transformation of yeast leu-2 - double auxotroph. It is a low-frequency system giving 2-102 transformants per microgram of DNA. These workers used DNA of recombinant plasmid pYE (Zeu-2)10 that carried the yeast Zeu-2 gene. This yeast leu-2 gene was identified by complementation of an E. coli leu-2 auxotroph. Plasmid pYE (leu-2)10 did not autonomously replicate but integrated in yeast chromosome at several sites including the leu-2 locus. Leu-2 gene fused to a yeast plasmid and to an E . coli plasmid provides a second type of vector. Such a hybrid plasmid replicates in both E. coli and yeast (Beggs, 1978).This vector is stably inherited when carrying cloned leu-2 DNA and transforming yeast to leu+ at frequencies of 104-105transformants per microgram of DNA (Haber et al., 1980; Shine et al., 1980; Struhl et al., 1979; Hicks et al., 1979a; Gerbaud et al., 1979). Storms et al. (1979)have constructed chimeric plasmids with a unique restriction endonuclease site that transform both Saccharomyces cerevisiae and E . coli with a high frequency. These plasmids carry part or all of the yeast 2 pm circle linked to E . coli plasmid pBR322 and yeast his3 gene or yeast leu-2 gene. Plasmids carrying a 2.4 megadalton EcoRI fragment of 2 pM circle transform yeast with 2- to 10-fold higher frequency than those carrying the 1.5-megadalton EcoRI fragment of the 2 p M circle. A clone bank of yeast genes has been screened and chimeric plasmids that complement the ura-3, tyr-I, and met-2 auxotrophs of yeast have been identified. A third yeast transformation system is based on segments of yeast chromosomal DNA that carry yeast replication origins. Hsiao and Carbon (1979) transformed yeast arg-4 deletion mutant to prototrophy with hybrid ColE1 plasmid containing yeast urg-4 gene. The transforming plasmid pYE (arg-4) replicates autonomously in both yeast and E. coli. About lo3transformants per microgram of pYE (arg-4)DNA were obtained, which is a yield three orders of magnitude higher than that obtained by Hinnen et al. (1978; see also Gafher and Philippsen, 1980; Petes, 1980). The yeast DNA fragments in pYE (arg-4)and yep-7 probably contain the origin of DNA replication, since they autonomously replicate in yeast.

44

VEDPAL SINGH MALIK

Another plasmid containing yeast trp-l gene (yep-7 and pLC544) replicates autonomously in tryptophan auxotrophs of yeast (Struhl et al., 1979; Kingsman et al., 1979). The replicon of these two plasmids can be isolated on a small DNA fragment and ligated to an E. coli plasmid along with a selective marker for yeast cloning. Such a hybrid plasmid can be shuttled between yeast and E. coli. Yeast transformants utilizing nutritional markers such as arginine are genetically unstable and have to be maintained on minimum media lacking the auxotrophic requirement. Since most media for producing industrial metabolites are quite complex, vectors that integrate into chromosome or that would not be lost under nonselective conditions may be preferable for industrial purposes. Yeast ribosomal DNA is on chromosome XI1 between asp-5 and ura-4 as determined by mitotic linkage (Szostak and Wu, 1979). The leu-2+ gene, inserted into the rDNA locus, was used as a genetic marker for mapping rDNA that would be otherwise hard to score. This is analogous to the use of antibiotic resistance transposons in mapping bacterial genes. Yeast rDNA consists of about 140 copies of a tandemly repeated 9-kb segment. When half of the yeast rDNA repeat unit is ligated to pBR322 carrying the yeast leu-2 gene, the frequency of yeast transformation increases up to 200-fold, in proportion to the increased amount of homologous repetitive rDNA available for integration. When the other half of the repeat unit is inserted into a plasmid, the transformation frequency increases by a factor of lo4, but the transformants are highly unstable. This segment of rDNA contains a yeast origin of replication (Szostak and Wu, 1979).

E. PLANTCELLS Crowngall is a neoplastic disease of plants caused by the gram-negative bacterium Agrobacterium (Moore et al., 1979). Agrobacterium harbors Ti plasmids, which are responsible for tumor induction in plants. Modified Ti plasmids may be used as vectors for introducing genes into plants that can be propagated vegetatively (Aerts et al., 1979; G . Melchers et al., 1978). When Agrobacterium infects plants, a part of the Ti plasmid called tDNA is transferred to plant cells. tDNA contains genes for opine synthesis and infection. It consists of three functional units. The central unit is present in all Ti plasmids (Thomashow et al., 1980). These plasmids are large, with molecular weights ranging from 150 to 230 kb (Drummond, 1979). They are extensively cleaved by restriction enzymes. As a result, unnecessary segments and restriction sites of these plasmids must be deleted before they can be useful as vectors. Use of an enzyme such as HphI, which cleaves at some distance from its recognition site, is another

RECOMBINANT DNA TECHNOLOGY

45

possibility for inserting DNA sequences in Ti plasmid. The unique sticky ends generated at each cut allow accurate plasmid assembly. The DNA could be inserted into an isolated tDNA fragment before the whole vector is assembled. Schell and associates avoided manipulating DNA in vitro altogether (Schell et al., 1979). They obtained transposition of streptomycin resistance (Tn7) into tDNA of a nopaline plasmid before introducing it into tobacco genome. It should be possible to construct a plasmid that would have an insertion sequence at either end of the DNA sequence to be cloned and then the whole unit could be transposed into the tDNA.

F. MAMMALIANCELLS Even though the first demonstration of recombinant DNA methodology was the introduction of galactose operon of E . coli into simian virus 40 (Jackson et al., 1972), the cloning of genes in higher eukaryotes has been slow. This is owing to the primitive state of knowledge of the genetics of both mammalian cells and their viruses. Tightly blocked mutants and selectable markers are few, and methods for cell-to-cell genetic exchange are poorly developed. In mammalian cells, there are no equivalents of F’ factors, plasmids, and bacteriophage X for E . coli. The mammalian genome contains a thousandfold more DNA than E . coli, and is therefore very complex. This tremendous complexity of the mammalian genome and the lack of strong selection pressure makes it hard to isolate specific genes by selecting for their functional expression. On top of all this, tumor viruses have been the most promising vectors, and development of these genomes as vectors has been significantly impeded by the National Institutes of Health (NIH) guidelines on recombinant DNA technology. Simian virus 40 DNA has been used as a vector for cloning many prokaryotic genes into mammalian cells. However, linking the origin of replication of SV40 to the thymidine kinase gene of Herpes simplex virus type I or the dihydrolate reductase gene derived from the methotrexate resistant AT-3000 line of murine sarcoma 180 cells could yield a suitable vector, one that would grow as a plasmid in permissive cells and have a selectable marker. By joining the SV40 vector to some of the repeated sequences such as histone genes, ribosomal RNA genes, or satellite DNA, homology of the vector with the cellular DNA could be increased. Such a vector would be useful for introducing new genetic information into mammalian chromosomes. Viruses serve as natural vectors for intercellular transmission of DNA. As do transducing bacteriophages, some mammalian viruses recombine with host chromosomal DNA and incorporate host DNA in their chromosomes. Such transducing particles, when they carry their origin of replication, can

46

VEDPAL SINGH MALIK

be propagated in the presence of helper viruses that complement the missing function of the transducing particle. Infection of nonpermissive mouse cells by a hybrid molecule, SV40 joined to phage h DNA, yielded transformants containing h and SV40 sequences (Muzyezka, 1979). Transformed clones were positive for SV40 T-antigen. The DNA linked to the transforming segment of SV40 DNA was integrated into the chromosome, and cyclization of the hybrid linear DNA preceded integration. Purchio et al. (1979) inserted the ornithine transcarbamylase gene of E. co2i into the late region of SV40 DNA. This hybrid molecule was propagated in monkey kidney cells. Hamer and Leder (1979) infected cultured monkey kidney cells with a recombinant plasmid in which the complete genomic mouse p-maj globin gene replaced the SV40 late gene region. Transcription from the SV40 late region promoter generates the globincoding sequence. Production of a substantial amount of mouse p-maj globin by such infected cells suggests that the mouse globin signals for RNA splicing, polyadenylation, and translation are recognized in the monkey cells. Therefore, the SV4O-monkey cell system can be used to clone and study many eukaryotic genes (Berg, 1979). There are several advantages of using SV40 as a vector: 1. The small size of SV40 makes localized biochemical mutagenesis easy. 2. Replication of SV40 is very efficient. Late in infection, up to 100,000 copies of SV40 per infected monkey cell are present, which facilitates detection of gene products. 3. Strong viral late region promoter can be used to direct the transcription of the inserted foreign DNA. However, not more than 2500 base pairs of foreign DNA can be inserted into the late region of SV40 if it is to be packaged into virions. Propagation of hybrid plasmid in persistently infected monkey cells can be used to bypass this limitation. Tissue cultures of mammalian cells have been transformed with DNA coding for selectable biochemical markers (Wigler et al., 1977; Minsen et al., 1978).The thymidine kinase (TK) gene from Herpes simplex virus type 1 can transform mouse L cells deficient in thymidine kinase, yielding clones with TK+ phenotype. Genes that do not code for a selectable marker can be introduced in these cells by cotransformation with the TK gene (Wigler et al., 1979). Large DNA fragments can also be first joined with the Herpes simplex virus TK gene and then transformed into mouse thymidine kinase negative cells. Mouse thymidine kinase negative L cells have been transformed with cloned rabbit chromosomal P-globin gene linked to the cloned

RECOMBINANT DNA TECHNOLOGY

47

TK gene of Herpes simplex virus type 2 (Mantei et al., 1979; Wigler et al., 1979). Most TK+ cells contained rabbit P-globin DNA and produced up to 2000 copies of rabbit P-globin RNA per cell. The human genes for thymidine kinase (TK), galactokinase, and pro collagen type 1 located on chromosome have been transferred to cultured (TK-) mouse cells (Perucho et al., 1980; Klobutcher et al., 1980; Wigler et al., 1980; Franche and Taggart, 1980). Catterall et al. (1979) transformed mouse cells with a chimeric plasmid containing the natural genes for chicken ovalbumin and the viral TK gene. The entire 12 kb chicken DNA containing the ovalbumin gene integrated into mouse cellular DNA and was expressed into the production of ovalbumin. The site of integration of chicken ovalbumin into mouse cell chromosomes could be important for expression of chicken ovalbumin gene, since not all ovalbumin gene-containing mouse transformants synthesize ovalbumin. Similarity between the junction sequences in the genes for ovalbumin and P-globin suggests that ovalbumin gene transcript should be correctly spliced in mouse L cells (Konkel et al., 1978). By improving the efficiency with which the incoming DNA is integrated into the cellular DNA, the efficiency of transformation is increased. When restriction enzyme digests of the TK+-transformedmouse cell DNA are used to retransform TK - mouse cells, the transformation frequency is significantly higher than with the 3.4 kb restriction enzyme fragment carrying the TK gene isolated from HSV-I DNA. Integration is increased owing to homology of the mouse cell DNA bracketing thymidine kinase gene (Minson et al., 1978; Pellicer et al., 1978). VII. Utility of New Technology Les combinaisons de la nature sont a la fois plus simples et plus variees que celles de notre imagination. L. PASTEUR(1880)

The understanding of the structure, organization, and function of genes of E. coli and its phages has progressed far beyond the knowledge of other organisms. This is owing to the availability of techniques of genetic manipulations using F' episomes, phage A, and many translocatable elements. Small segments of DNA can be moved to make merodiploids for any part of the chromosome, which allows the analysis of operons and their regulation by study of the interaction between different alleles of the same regulatory genes and controlling elements in the same cell (Beckwith and Zipfer, 1970; Stent and Calender, 1978; Miller and Reznikoff, 1978; Calos and Miller, 1980). Study of the chemistry, physiology, and genetics of lactose metabolism in E. coli has been easily elucidated owing to a simple enzyme

48

VEDPAL SINGH MALIK

assay. Induction with the gratuitous inducers and availability of noninducing substrates such as X-G have been used to distinguish constitutive from inducible strains.

A. MERODIPLOIDSCONSTRUCTION COMPLEMENTATION TEST

AND

Knowledge of the genetics and biochemistry of industrially important microorganisms is much less advanced than it is in E. coli. Even though many phages infect microorganisms of industrial importance, knowledge of the regulatory biology of these phages is primitive compared with that of coli phages. Also, little is known about the biology of plasmids in such microorganisms. Gene amplification and in vivo merodiploid formation has not yet been achieved by specialized transducing phages or by plasmids for strains of organisms that are being used in industry. Recombinant DNA technology can now be used to ligate small pieces of genomes of industrially important organisms to appropriate vectors that replicate in such organisms as well as in E. coli, which allows application of complementation test for understanding regulatory mechanisms. Such vectors that replicate in E. coli, as well as in some other organism such as B . subtilis or yeast, are already being used to shuttle genes between two different host organisms. B. OPERONFUSIONS Another application of this technology emerges because fusion of any two pieces of DNA is now possible. Genes of any organism can now be fused to the well-studied E. coli promoters with the expectation of efficient expression (Franklin, 1978). Elegant experiments from Beckwith's laboratory have provided the pioneering methodology for in vivo fusing of the P-galactosidase genes into foreign regulatory elements (Bassford et al., 1979; Bassford and Beckwith, 1979; Brickman et al., 1979). Such gene fusions allow exploitation of a simple P-galactosidase assay to probe complex regulatory systems (Cassadaban, 1976). Similarly recombinant DNA technology could be used to create fusions in which the P-galactosidase gene is controlled by promoters of other genes that are functional in industrial organisms. The type of genetic analysis that has been used for lac and phage A repressors could now be applied to other proteins. Point mutations were used to define the regions of the repressors responsible for binding to operator, and to inducer, and aggregation. Thus, it was shown that the operator binding site of lac repressor is restricted to 59 N-terminal residues and the first 92 N-termind residues are sufficient to bind X repressor to A operator

RECOMBINANT DNA TECHNOLOGY

49

(Miller and Reznikoff, 1978). Negative dominance can be used as a tool to study function of any protein that carries two different enzyme activities on one polypeptide chain such as threonine-deaminasekhreoninekinase. Mutants lacking one of the activities can show which end of the polypeptide chain is responsible for each activity. Nonsense fragments of lac repressor are destroyed effectively by proteolytic enzymes. Chimeras that add different pieces of repressor to active P-galactosidase can be isolated and many regions of repressor can be added to P-galactosidase without destroying its enzymic activity. Some chimeras are active both as repressor and P-galactosidase. In one of them the repressor lost the last 4 residues and P-galactosidase, the first 24. It may be that the first 24 amino acids of P-galactosidase can be easily replaced by any protein sequence that does not interfere with the folding and aggregation of P-galactosidase. DNA segments coding for N-terminal regions of A repressor and A-cro have been fused in vitro to other genes. The importance of N-terminal region for DNA binding is supported by the fact that repressor or cro activity is recovered in vivo (Ptashne et al., 1980). The knowledge, gained from repressor P-galactosidase fusions, that certain peptides that are degraded in E. coli can be stabilized by fusing them to high-molecular-weight proteins has already been exploited to make somatostatin hormone in E. coli. Somatostatin was found to be stable in E. coli when it was fused to P-galactosidase protein (Itakura et al., 1977). Similarly, insulin is not destroyed when it is fused to the P-lactamase gene (Villa-Komaroff et al., 1978). Both insulin and somatostatin were fused to the carboxylterminal ends of protein and were transcribed from the promoters of the E. coli gene. Backman et al. (1976)constructed a plasmid in which the A repressor gene was fused to the promoter of the lac operon. The plasmid bore two copies of lac promoter, and synthesis of A repressor in strains bearing this fused plasmid was regulated by lac repressor. Since several translational stop signals immediately precede C gene, repressor synthesized in these strains was not a fused product initiated at the lac2 translational start point. About 3 to 35 times more A repressor was produced as compared to single lysogens. The source of variation in repressor level is not known. These A repressor levels are 10- to 25-fold lower than expected on the basis of the known strength of lac promoter. This inefficient expression could be due to frequent attenuation of transcription beginning at the lac promoter, since A repressor could bind to the A operator, which is situated between the repressor structural gene and the lac promoter. Strains carrying 20 copies of recombinant plasmids in which level of A repressor is controlled by its own promoter produce about five times more repressor than does a single lysogen. These strains exemplify autogenous regulation of repressor synthesis.

50

VEDPAL SINGH MALIK

c. DIRECTEDSECRETION OF PROTEINS To gain insight into the mechanisms by which proteins are secreted from the cytoplasm to the cell membranes, genetic recombination was used to replace the amino-terminus of P-galactosidase (a cytoplasmic enzyme) with the amino-terminal sequence of membrane proteins such as the maltosebinding protein (Silhavy et al., 1976) or phage A receptor lam B protein (Silhavy et aZ., 1977). Owing to the lack of convenient assay for secretory proteins, P-galactosidase was used as a reporter enzyme. In both cases, P-galactosidase activity became associated with membranes, which suggested that addition of a signal sequence to a nonsecreted protein will cause it to be secreted (Davis and Tai, 1980). On the basis of these results, Villa-Komaroff et al. (1978) fused in uitro rat insulin gene cDNA to the P-lactamase gene of pBR322. The hybrid insulin P-lactamase protein was found to be secreted, which protected the insulin from proteolytic degradation by cellular enzymes. Using a very sensitive radioimmunoassay, these workers detected about 100 molecules of insulin per cell. The insertion of a viral DNA sequence coding for surface antigen into a carrier protein gene could result in the secretion of a fused protein that could be used as a vaccine even though no correct virus product is synthesized (Enouye and Halegou, 1980; Koshland and Botstein, 1980). Various genes that determine the synthesis of enzymes secreted by industrial organisms might be fused to appropriate vectors. For example, yeast invertase gene could first be inserted in a vector (Hackel and Khan, 1978; Lampen, 1978). Genes directing synthesis of peptides could then be inserted in this vector in such a manner that the amino-terminal end of invertasa facilitates the secretion of fused protein. Invertase would also be helpful to the yeast while growing on molasses, a cheap industrial carbon source (G. Melchers et al., 1978). D. In Vitro LOCALIZED MUTAGENESIS Recombinant DNA technology can be used to isolate, amplify, and determine the structure of any DNA segment. The functional significance of a specific region of DNA can be explored by generating localized modifcation in the DNA molecule and then studying the resulting effects on the biological properties of the molecule in uiuo or in uitro (Humayun and Chambers, 1979; Gillam and Smith, 1979a,b). I n uitro mutagenesis offers the possibility of mutating isolated purified DNA sequences to obtain enzymes with altered properties (Muller et al., 1978; Paterson and Clarke, 1979; Hibner and Albert, 1980). Analogs of peptide hormones could also be produced by their genes that have been altered in predetermined base sequences. DNA sequences of regulatory elements (operator, promoter, attenuator) can be sub-

RECOMBINANT DNA TECHNOLOGY

51

jected to in uitro localized mutagenesis, and mutated sequences can be used to study how corresponding proteins interact with them (Taniguchi et al., 1979). Such studies have provided insight into the types of interaction between lac operator and repressor (Siebenlist et al., 1980) and promoter and RNA polymerase (Taylor and Burgess, 1979; Johnstrud, 1978; Taniguchi et al., 1979; S . J. Chan et al., 1979). Scherer and Davis (1979) have developed a general method for replacing Sacchromyces cereuisiae chromosome segments with altered DNA sequences. Any isolated gene, deleted or altered by in uitro mutagenesis, can be propagated by attaching it to a vector with a selective chromosomal marker. Upon transformation, the composite DNA molecule, consisting of vector, selective marker, and altered DNA, recombines with the homologous DNA sequence in the chromosome, yielding a direct nontandem duplication separated by the vector and the marker DNA. The vector and marker DNA are lost owing to the second recombinational event on the other side of the altered sequence relative to the first. This replaces the altered sequence with the wild-type DNA segment of the untransformed cell. Homologous sequences on both sides of the mutant DNA sequences must be present in the transforming DNA. The composite molecule does not have to replicate autonomously in the recipient host where altered DNA sequences are to be replaced. An internal deletion mutant of his3 gene and a transposition of a galactose-inducible region into chromosome XV were generated by using the ura-3 gene as a selective marker. Scherer and Davis’ method has general application and relies on the ability to transform cells to achieve integration by homologous recombination and on the availability of selective markers as cloned DNA fragments. The expected phenotype of the organism after insertion of foreign DNA is not known in advance. The desired strain can be identified by restriction enzyme analysis of the clones with the selective marker, which allows deletion or alteration of genes with no known functions. If a good, nonreverting mutation for the selective marker is not available, such a mutation can be transplated into the organism of interest after its creation in uitro. This technique could help the manipulation of the chromosomes of industrially important organisms, since it can provide well-characterized sequence variants of any genetic locus.

E. GENOMEORGANIZATION

1. Gene Mapping Restriction mapping is increasingly being used for gene mapping. The positions of about 350 genes on human chromosomes have been localized and the major part of the human gene map may be constructed in the next

52

VEDPAL SINGH MALIK

few decades. Availability of cloned probes will speed up chromosome mapping (Little et al., 1980; Orkin et al., 1980). Labeled probes add speed and resolution to restriction mapping, an advantage over in situ hybridization. Using labeled probes of complementary RNA prepared from cloned sea urchin genes, five human histone genes have been localized to the long arm of chromosome VII (Gusella et al., 1979). Scott et al. (1979)and Haigh et al. (1980)have precisely localized p-globin gene on the short arm of human chromosome XI. DNA was isolated from hybrid clones that contain Chinese hamster genome together with a section of human chromosome XI. The DNA was treated with restriction endonucleases and fragments resolved on agarose gels. The separated fragments were transferred onto nitrocellulose filters and hybridized to a radioactively labeled human P-globin DNA probe. The pattern of hybridization as revealed by autoradiography was used to conclude that the gene for P-globin is carried on a section of the short arm of human chromosome XI (Scott et al., 1979). 2. Gene Structure

a. Intervening Sequences. The availability of homogeneous purified chromosomal DNA fragments has allowed the determination of their exact base sequence. This information has resulted in the unexpected discovery of intervening sequences (introns) within many genes of higher organisms (eukaryotes)and their cytoplasmic organelles (mitochondria). The diversity, complexity, and size of genes with introns was unexpected (Fedoroff, 1979; Abelson, 1979; Darnell, 1978; Schechter et al., 1979; Long and Dawid, 1980; Crick, 1978; Cory et al., 1980). Including its intervening sequences, the gene that codes for the enzyme dihydrofolate reductase in mammalian cells is about one million nucleotides long. About 10% of these nucleotides code for amino acids in the enzyme. The chick ovomucoid gene contains seven intervening sequences (Catterall et al., 1979; O’Hare et al., 1979). The entire gene is transcribed into a large RNA transcript that is subsequently spliced to yield mature mRNA. The mRNA coding portion (exons) of the chicken conalbumin genes is in 17 pieces (Cochet et al., 1979).

b. Overlapping Genes. The total genomes of several small viruses have now been sequenced. Analysis of these sequences has confirmed that certain genes overlap and some pieces of DNA direct synthesis of more than one protein (Fiddes and Godson, 1979; Shaw and Murialdo, 1980; Yang and Wu, 1979). Overlapping genes have gone undetected for a long time, but DNA sequencing now can aid in their discovery. c. Promoters in the Middle of the Gene. Promoters are sequences of nucleotides in a genome where RNA polymerase first binds before transcrib-

RECOMBINANT DNA TECHNOLOGY

53

ing a gene into mRNA. In bacteria, such RNA polymerase binding sites are at the beginning of a gene. However, sequencing of the DNA of several higher organisms suggests that promoters might be in the middle of the gene (Korn et al., 1979; Moss and Birnsteil, 1979; Bogenhagen et al., 1980; Hentschel et al., 1980).

d . Regulation by Inversion of Promoter and Chromosomal Rearrangement. Salmonella possess two distinct flagella antigens. In general, however, only one of the antigens is expressed at a given time. This phenomenon, called phase variation, has recently been understood through the application of recombinant DNA technology. Transition from expression of one antigen to another results from an inversion in a region containing a promoter-like DNA sequence that controls initiation of gene transcription. In one orientation, transcription occurs. However, inversion of the sequence produces reverse orientation of the promoter contained within that region, so that transcription of genes that direct synthesis of flagellin subunits is interrupted (Silverman and Simon, 1977; Zieg et al., 1978).The mechanism for regulating the inversion of the promoter DNA sequence is not known. Transposition of structural genes as a developmental regulatory mechanism may be involved in the life cycle of homothallic bakers’ yeast (Kushner et al., 1979). Yeast haploid cells of one mating type (a or a) produce progeny of the opposite mating type during mitotic growth. A diploid phase (da)is established by conjugation between siblings of opposite mating types. Switches between mating types occur by transposition of a genetic element that carries genes for the mating type. By using a copy of the Q mating-type gene as a probe for Southern hybridization (Hicks et al., 1979b), the conclusion was reached that the mating type interconversion process in yeast involves genetic rearrangement. Silent copies of the mating type genes are located at the H M L and H M R loci on chromosome 111. Translocation of the replicas of these silent sequences to the mating-type locus affects the switch between mating types. Genome rearrangement, e.g., inversion and translocation, may be involved in the regulation of functions that are not essential for survival of the organism and are subjected to catabolite repression (Williams et al., 1979). Enzymes involved in antibiotic production by Streptomyces may be controlled by such a regulatory system. F. ORIGIN(Om) OF DNA REPLICATION Plasmids have been constructed that contain DNA replication origin (ori) of B . subtilis (Seiki et al., 1979; Niaudet and Ehrlich, 1979), E . coli (Sugimoto et al., 1979), phage G4 (Kaguni aad Ray, 1979), Salmonella typhimurium (Zyskind et al., 1979), yeast (Stinchomb et al., 1979), and

54

VEDPAL SINGH MALIK

Drosophila mitochondrial DNA (Sugino, 1979). A nonreplicating EcoRI fragment conferring Km resistance was used for selecting two different EcoRI fragments containing the replication origin from the F factor and the S. typhimurium chromosome. Owing to incompatibility with the E. co2i chromosome, E. coli and S. typhimurium origin plasmids are not stably maintained in E. coli in the absence of selective pressure. The stability increases with decreased plasmid size, perhaps because genes whose products are detrimental to cell growth have been eliminated. This possibility is supported by the observation that a disruption of the nucleotide sequence at the PstI site in the uncB gene, a site outside the S. typhimurium Ori region, increases plasmid stability by preventing the production of functional uncB gene product (Zyskind et al., 1979). Recombinant DNA technology is being used for the analysis of plasmid structure and function (Timmis et al., 1978a,b). Segments of any plasmid can be inserted in ColE1. The hybrid plasmids are used to transform Poly(A) E. coli that is deficient in DNA polymerase I. Since the ColE1 plasmid component requires DNA polymerase I for replication, only those recombinant plasmids that have inserted a fragment with origin of replication will replicate in Poly(A) Escherichia. This approach has been used to construct hybrid plasmid replicons that harbor the origin of replication of pSCl01 and ColEl plasmids. An autononously replicating 850-base pair sequence has been isolated from yeast (Stinchomb et al., 1979).This sequence has no detectable homology to other yeast sequences. Such replicator-containingplasmids are used for cloning foreign genes and for understanding DNA replication. G. HUMAN PROTEINS The only hope of obtaining proteins such as human interferon, human insulin, and human growth hormone in good amounts is recombinant DNA technology. The microbial production of some human proteins of importance to medicine has already been achieved using recombinant DNA technology (Bylinsky, 1980; Secher and Berke, 1980; Goeddel et al., 1979b; Shine et al., 1980; Talmadge et al., 1980; Vamvakopoulos et aZ., 1980). Recombinant DNA technology has added a new dimension to applied microbiology and has already been used to produce ovalbumin (The Upjohn Company), human interferon (Genentech Inc., Biogen Inc.), human insulin (Eli Lilly and Co., Genentech Inc., Biogen Inc.), human growth hormone, somatostatin, thymosin, and many other products in E. coli (Genentech Inc.). An outstanding example of the benefit to be derived from the application

RECOMBINANT DNA TECHNOLOGY

55

of the new technology would be the production of the highly desirable interferon. Fibroblast and leukocyte interferon genes have been cloned and expressed in E. coli by Genentech scientists. Induced leukocytes were used to make cDNA. Amino acid sequence of tryptic peptide of interferon was used to synthesize mRNA probe. Transformants fall into six classes as analyzed by restriction mapping and nucleotide sequencing. One class that may be similar to interferon was independently cloned by C. Weismann collaborating with the Biogen Laboratory. Two hundred and fifty million units of leukocyte interferon per liter of E. coli culture were produced. Nonglycosylated interferon appears to be biologically active. Cloning of the human insulin gene in E. coli and commitment to commercialize microbially produced human insulin by Eli Lilly and Company further supports the notion that industry is doing its best to insure that the results of research are reaching the market place. H. VACCINES Hepatitis B virus infects man and can cause severe complications such as cirrhosis, hepatitis, and hepatocarcinoma. One hundred and twenty million chronic carriers of hepatitis virus have been estimated worldwide. This virus cannot be propagated in cell cultures, and human serum is the only viral source. However, cloning of this virus in E. coli now allows the purification and determination of the complete nucleotide sequence of hepatitis B virus genome (Gilbert et al., 1979). Since high titers of hepatitis B virus are not produced in tissue culture cells, Valenzuela et al. (1979) used recombinant DNA technology to obtain large quantities of viral genome. Hepatitis B virus DNA was cloned in E. coli using plasmid pBR322 as vector. Restriction endonuclease analysis of full-length hepatitis virus genome and the nucleotide sequence of an 892-base pair fragment encoding for hepatitis B surface antigen has been reported. The portion of the gene coding for surface antigen has no intervening sequences. The DNA sequence suggests that the surface antigen consists of 226 amino acids of MW 25.398. This information combined with the availability of large amount of viral genome will help in the development of antiviral vaccines. Segments of viral genome may be used to produce appropriate viral antigens in alternative hosts like Saccharomyces cerevisiae or B . subtilis. The core antigen gene is expressed in E. coli and, when injected into rabbits, the antigen produced by E. coli induces antibody that reacts with human serum core antigen (Pasek et al., 1979a,b). Recombinant DNA technology can be used to identify the genes that code for a surface protein of foot and mouth disease virus. Such genes can be introduced in the appropriate host to produce the protein that may be an effective vaccine. Emtage et al. (1980)

56

VEDPAL SINGH MALIK

have inserted a DNA sequence from fowl plaque virus into an E. coZi plasmid. The recombinant plasmid directs the synthesis of a protein that is 0.756 of total cell protein and reacts with antisera to FPV hemagglutinin. Transcription of the inserted fragment is from a tryptophan promoter.

I.

IMPROVEMENT OF INDUSTFUAL ORGANISMS

There was the door to which I found no key. There was the veil through which I might not see. OMARKHAYYAM (TWELFTH CENTURY)

For commercial production of recombinant DNA products, some organisms may prove better than E. colt Groomingof organisms already being used in the fermentation industry could be highly beneficial. Even though experiments with industrial organisms have not received as much fanfare as have the unraveling of secrets of nature using E. coli, industrial organisms harbor metabolic pathways that are specific to them. For example, pathways that lead to the biosynthesis of antibiotics, gibberellins, and alkaloids exist in only a few microorganisms. Regulation of general metabolism in such organisms is not well coordinated. The lack of fine regulation of primary cellular metabolism leads to accumulation of many branch point intermediates and end products. These elevated levels of intermediates are the precursors of many commercial metabolites (Malik, 197%) that are produced in significant quantities. For example, penicillin and glutamic acid are accumulated in concentrations of up to 50 and 120 gm/liter of growth medium, respectively. Efficient and economical processes for the industrial production of these microbial metabolites can now utilize the tools of recombinant DNA technology and make even further improvement.

1. Gene Ampl@cation From what has already been demonstrated, you can plainly see the impossibility of increasing the size of structures to vast dimensions either in art or in nature; likewise the impossibility of building ships, palaces or temples of enormous size in such a way that their oars, yards, beams, iron bolts, and, in short, all their other parts will hold together; nor can nature produce trees of extraordinary size because the branches would break down under their own weight, so also it would be impossible to build up the bony structures of men, horses, or other animals so as to hold together and perform their normal functions if these animals were to be increased enormously in height; for this increase in height can be accomplished only by employing a material which is harder and stronger than usual, or by enlarging the size of the bones, thus changing their shape until the form and appearance of the animals suggest a monstrosity. This is perhaps what our wise poet had in mind, when he says, in describing a huge giant: “Impossible it is to reckon his height So beyond measure is his size” CALILEO CALILEI

(FIFTEENTHCENTURY)

RECOMBINANT DNA TECHNOLOGY

57

Many molecules such as insulin, interferon, and growth hormone are active in very minute amounts. Such molecules that are not needed in thousands of pounds and can be sold at a good price may very well be commercially produced in E. coli (Shepard et al., 1979). Many multicopy plasmid and phage vectors can be used to ampllfy genes in E. coli (Malik, 1979d; Collins, 1977). Foreign genes on a multicopy plasmid are easy to isolate and purify in relatively large amounts. The products of such genes can be made in significant quantities if they are inserted in such a manner that their transcription is initiated by a strong promoter (Backman and Ptashne, 1978; Twigg and Sherratt, 1980; Gelfand et al., 1978; Teather et al., 1978). Gene cloning is an effective means of amplifying the number of copies of a specific gene (Malik, 1979d). A pertinent demonstration of this is the construction of a hybrid bacteriophage A into which an E . coli gene coding for DNA ligase has been inserted. Upon infection of E. coli with this chimera, a 500-fold increase in concentration of DNA ligase over that produced by uninfected E. coli is generated. DNA ligase accounts for about 5% of the total E. coli protein and is easy to purlfy in large amounts from such strains (Panasenko et al., 1977). DNA polymerase I was amplified 100-fold following infection of E. coli with X polA QS phage (Kelley et al., 1977). Enhancement of yields of any protein could be achieved by fusing the respective gene to an eficient promoter and inserting it into a multicopy plasmid containing heatinducible A regulatory system. A copy number mutant of plasmid ColE1 exists that produces 210 plasmid copies per E. coli chromosome (Shepard et al., 1979).Tenfold amplification of the lactose carrier protein in E. coli using a recombinant plasmid carrying the y gene of the lac operon has been achieved (Teather et al., 1978). Weisblum and co-workers (1979a,b) have isolated a high copy number mutant of the S . aureus plasmid pE194 that is replicated and maintained in B. subtilis. It is not known if this plasmid will be maintained in other grampositive organisms such as Streptomyces. The mutant plasmid is present up to 100 copies per cell and determines resistance to macrolides, lincosamide, and streptogramin type B antibiotics. This plasmid could be used to clone and ampllfy genes in B. subtilis, which is a source of many commercial enzymes. Yeast and many streptomycetes are known to harbor plasmids. The size of these plasmids can be decreased by available techniques. Decreasing the size of the plasmid sometimes elevates the copy number. 2. Novel Genes

With the increasing prices of media constituents for microbial fermentations, naturally abundant substrates such as cellulose might someday replace expensive carbon sources. Genes for the efficient utilization of cellulose and other substrates are not present in all organisms. Such genes

58

VEDPAL SINGH MALIK

might be isolated and cloned into organisms of commercial interest. Using such genetically tailored microbes that utilize cheap media constituents, more and more natural products might be made by fermentation, thereby replacing their expensive chemical synthesis. Single genes that direct the synthesis of enzymes involved in the modification of functional groups may be isolated. Such genes (for example hydroxylation, chlorination, amination, acetylation, adenylation, phosphorylation, and glycosylation) can be introduced into industrial organisms to obtain altered metabolites. Genes determining the synthesis of a chlorinating enzyme, for example, could be isolated from chloramphenicol-producing Streptomyces venezulae and introduced in corynecin-producing Corynebacterium. Corynecin is analogous to chloramphenicol but lacks chlorine. If the gene is expressed, Corynebacterium possessing the chlorinating enzyme may produce chloramphenicol if it is resistant to excreted chloramphenicol and does not inactivate it.

3. Novel Partial Metabolic Pathways Recombinant DNA technology can be used to transfer genes for partial metabolic pathways to organisms that do not have genes for such pathways.

a. Catabolic Pathway. Many organisms possess catabolic pathways that are usehl in microbial transformation programs. Sometimes mutants blocked in a catabolic pathway are used to accumulate valuable intermediates. Genes for a portion of a catabolic pathway that direct transformation of substrate to the desired intermediate can now be isolated and introduced into a new organism. This could allow accumulation of the desired metabolite and avoid undesirable additional transformations. This strategy might have immediate application in the biotransformations of steroids (Atrat et al., 1979a,b; Tenneson et al., 1979).

b. Anabolic Pathways. Many secondary metabolites, e.g., antibiotics, alkaloids, and gibberellins, consist of subunits that are present in several unique biologically active compounds produced by unrelated organisms (Table IV). Genes for some of the unusual moieties, e.g., sugars with side chains, could be isolated and introduced into a suitable organism to obtain hybrid molecules. Fusion of various macrolide and aminoglycoside antibiotics may now be possible and may yield a superior antibiotic. Fusing two natural products with no biological activity could also yield an active metabolite. In Penicillium chysogenum, a transacylase exchanges the L-a-amino adipyl side chain of isopenicillin N for nonpolar-substituted acetic acid esters. This transacylase gene could be cloned and subjected to in vitro mutagenesis. A mutant gene may be obtained that will direct the synthesis of

59

RECOMBINANT DNA TECHNOLOGY TABLE IV SUBUNITSCOMMON TO SEVERALANTIBIOTICS Antibiotics that share the unit

Subunit ~

D-Oleandrose D-Desosamine L-Mycarose 2-Deoxystreptamine a-,P-Diaminopropionic acid P-Methyl-L-lanthionine Butyric acid Ribostamycin P-Lactam ring

~~~~~

~~

~~

Oleandomycin, oleandrin Macrolide antibiotics (methymycins, erythromycins, picromycin, oleandomycin, narbomycin, megalomicin) Megnamycins, leucomycins, erythromycin, megalomicin A, tylosin, niddamycin, spiramycins, relomycin Aminoglycosides (neomycin, kanamycin A, gentamicin, butirosin) Edeine A, viomycin Nisin, subtilin Butirosin, 2-amino-4,4-dichlorobutyricacid Butirosin, neomycin Penicillins, cephalosporins, cephamycin, nocardicin, thienamycin, wildfue toxin

an enzyme that could exchange the D-a-aminoadipyl side chain of cephalosporins for nonpolar-substituted acetic acid esters. If the suitable mutant gene is introduced and expressed in the appropriate p-lactam producing host, cephalosporins with nonpolar-substituted acetic acid side chains may be produced. Such side chain precursors and their analogs have to be added to the growth medium of the tailored microbe. Hybridization among P-lactam-producing streptomycetes does, indeed, offer many attractive possibilities. Expression of Streptomyces genes determining synthesis of hydroxylases and methylases in cephalosporin might allow cephamycin production utilizing highly efficient strains and technology already in use in fermentation industry. Benzylpenicillin acylase of E. coli hydrolyses benzyl penicillin to 6-aminopenicillanic acid and phenylacetic acid (Gang and Shaikh, 1976). At low pH, it catalyzes the synthesis of new penicillins from 6-aminopenicillanicacid and different side chain precursors. Several semisynthetic penicillins that are sold commercially have already been prepared by this method. Mayer et al. (1979)used the cosmid-packaging method of cloning to isolate the gene that contains the penicillin acylase segment from the total E. coli genome without positive selection pressure. This fragment was further shortened by cleavage with endonucleases and cloned into a plasmid. E. coli strains carrying the penicillin-acylase gene produced high constitutive levels of enzyme. This gene can be brought under the control of stronger promoters to obtain increased enzyme levels. The penicillin-acylase gene can also be subjected to site-directed in uitro mutagenesis to alter the specificity of the enzyme. The benzylpenicillin-acylase gene as well as in uitro-isolated

60

VEDPAL SINGH MALIK

mutant genes of these commercially important enzymes could be ligated to appropriate promoters and then introduced into penicillin-producing organisms. If the gene is expressed, then 6-aminopenicillanic acid instead of penicillin could be produced. A novel two step process for the conversion of cephalosporin C into 7-aminocephalosporanicacid has been reported. D-Amino acid oxidase of the yeast Trigonopsis uariabilis, converts cephalosporin C into glutaryl-7aminocephalosporanicacid, which is further deacylated by a mutant of Comamonus. If the genes determining the synthesis of these two enzymes could be introduced and expressed in the cephalosporin C-producing organisms, direct production of 7-ACA may be possible. Alternatively, novel penicillins might be produced in one step by introducing penicillin acylase of a different specificity into the penicillin producers and incorporating the appropriate side chains of penicillin in the fermentation medium. Many enzymes may be used in the industrial hydrolysis of cephalosporin and penicillins and acylation of 7-aminocephalosporanic acid as well as 6aminopenicillanic acid (Brodelius, 1979).The genes for these enzymes can be isolated and planted in appropriate organisms to make the process of preparing penicillin and cephalosporin analogs economical (Serizawa et al., 1979). Many streptomycetes produce several antibiotics of different chemical structure. These antibiotics are sometimes produced during various phases of growth. Self-cloning in streptomyces could result in increased levels of antibiotic production. Furthermore, clones that harbor amplified operators involved in antibiotic synthesis could neutralize the repressors, resulting in the constitutive production of antibiotics and thus altering the relationship between various growth phases and antibiotic production. Alteration of this relationship could also result in production of novel coupling of antibiotic moieties that were never linked in the parent culture because various moieties were not produced during the same phase of growth. A genome library of any streptomycete harboring various segments of its chromosome ligated to a multicopy plasmid or ribosomal RNA genes could be generated and screened for antibiotics that have not been seen before owing to lack of production or low yield. Thus self-cloning, which is essentially creation of merodiploids using recombinant DNA technology, offers also the possibility of doing complementation tests for gene functions and regulatory studies. Some of the DNA sequences could be ligated to a vector whose replication is temperature-sensitive, and then amplified during any growth phase by manipulating growth temperature. The realization of all the aforementioned possibilities hinges upon the development of vectors with selectable markers that could be used to transport, maintain, and faithfully replicate various segments of the Streptomyces

RECOMBINANT DNA TECHNOLOGY

61

genome into novel Streptomyces that produce many secondary metabolites of intricate chemical structures.

J. VANISHINGORGANISMS Every year many species of plants and animals vanish and a source of genetic information is lost from nature. Making banks of genes of endangered species in E. coli and then storing them may offer a way to salvage the genes of organisms facing extinction.

K. RESOURCERECOVERYAND WASTEDISPOSAL Microbes can be used to recover valuable minerals such as uranium and silver from low grade ores (Manchee, 1979; Charley and Bull, 1979). Many bacteria harbor plasmids that cany genes for transformation of metals, aromatics, and other hydrocarbons (Chakrabarty, 1976). Since they are located on plasmids, such genes can be easily isolated and then introduced into organisms like Azotobacter. Azotobacter can fN nitrogen and utilize the excessive carbon present in municipal sewage. The microbially treated sewage can be used as fertilizer (Domino, 1979). Use of microorganisms as a source of protein has been widely advocated (Kharatyan, 1978). Genes directing the synthesis of proteins that are enriched in essential amino acids can be incorporated into organisms to improve the nutritional value of their protein. Such tailored organisms, grown on sewage and other garbage material, can be used as feed for poultry and cattle. Genes responsible for dehalogenase can be introduced into organisms that grow in sewage polluted with halogenated chemicals (Slater et al., 1979; Weightman et al., 1979). Soil microflora play a very important role in soil fertility. Many microbes could be improved so that they could easily establish themselves in plant rhizospheres (Hirota et al., 1978; Newman, 1978). For example, most of the phosphorus in the soil is fixed in insoluble form. Many phosphorussolubilizing organisms such as Bacillus megatherium var phosphaticum could potentially be modified to enhance their activity, resulting in increased crop yield (Malik, 1964).

L. IMPROVED CROPS Recombinant DNA technology can be used to mod* and characterize the genetic makeup of plants (Seshadri and Ranjekar, 1979; Nasyrov, 1978; Hoffmann, 1980; Gleba and Hoffmann, 1980). Many plant proteins are defi-

62

VEDPAL SINGH MALIK

cient in amino acids that are essential for human health. Genes determining the synthesis of a superabundant protein that is rich in essential amino acids could be introduced to improve the quality of plant protein. It also might be possible to move genes for resistance to pests, herbicides, microbial toxins, and nitrogen fuation. If nitrogen-fuing enzyme genes (dflwere to be introduced into a plant such as corn, their expression would have to be subjected to eukaryotic control unless they were planted in mitochondria or chloroplast genome. Also, some way would have to be found to protect nitrogenase from oxygen poisoning (Brill, 1979; Mortenson, 1978; Mortenson and Thorneley, 1979; Shanmugam and Valentine, 1975). Some strains of Rhizobium form nodules only on a very narrow range of plants, while other promiscuous Rhizobium form nodules on many different plants, such as cowpea, soybean, peanut, and mungbean (Goodchild, 1977; Beringer et al., 1979). Rhizobium also forms nodules on a nonleguminous plant called Trema. Many nitrogen-fuing trees, such as Alnus, possess nodules formed by an actinomyces-Frankia (Callaham et al., 1978). These bacteria may be tailored for nodule formation on nonlegume plants (Neyra and Dobereiner, 1977). Nitrogen fuation capability could also be conferred on many free-living bacteria that are abundant in soil and plant rhizospheres. Their excretion products and cellular material upon death could enrich soil. One of the limiting factors in symbiotic nitrogen fixation is the availability of energy rich bonds (ATP)that are synthesized during photosynthesis in the green part of the plant. By increasing the efficiency of the photosynthetic process, the symbiotic nitrogen fixation may be simultaneously enhanced. Resistance to plant pests and diseases is often conferred by a single gene known as a major gene (Russell, 1978). Resistance based on a single major gene or a few such genes becomes ineffective against new races of pathogens and pests after a few years of its introduction into a crop plant (Simons, 1979; Lacy and Leary, 1979). Therefore, novel strategies are needed for obtaining crops resistant to pests and pathogens. Breeding of resistant varieties is one solution to crop diseases. An integrated approach involving resistant cultivars, presticides, fungicides, herbicides, biological control, and agronomical methods is usually the most effective approach. The photosynthetic enzyme ribulose bisphosphate carboxylase/oxygenase is probably the most abundant protein in nature. This highly soluble protein occurs in the stromal compartment of chloroplast and represents up to 65% of the total soluble protein in leafextracts. A similar enzyme constitutes 40% of the soluble protein in the photosynthetic autotroph Rhodospirillum rubrum. It catalyzes the carbon dioxide-fixing step in photosynthesis. C02

-

+ ~-ribulose-l,5-bisphosphate + H 2 0 Mg'+

2,3-phospho-~-glycerate

RECOMBINANT DNA TECHNOLOGY

63

This enzyme catalyzes another reaction involving molecular oxygen, which is the first step in the process of photorespiration (Heber and Krause, 1980). 0 2

+ D-ribulose-l,5-bisphosphateMg2+ + 3-phospho-~-glycerate+ 2-phosphoglycolate

This important oligomeric enzyme consists of two subunits. The cistron that codes for the large subunit is located on chloroplast DNA. The small subunit is encoded in the nuclear genome and is synthesized as a high-molecularweight precursor. While entering the chloroplast, this precursor is processed so that it can combine with the large subunit in the stroma of the chloroplast to yield holoenzyme. Both microbial and plant ribulose bisphosphate carboxylase enzymes are very inefficient. At the saturating level of carbon dioxide, the turnover number for the carboxylase activity is lo00 moles of carbon dioxide fHed per mole of enzyme per minute. In nature, where the substrate concentration is limiting, the turnover number drops to 200. Another metabolic constraint on plant productivity is the oxygenase activity of this enzyme that produces glycolate, which is lost from the plant via photorespiration. Alteration of the activity in favor of carboxylase at the expense of oxygenase could enhance crop yield. No chemical that would stimulate carboxylase activity or inhibit oxygenase activity under conditions of agricultural production is known. Ribulose bisphosphate carboxylase from tetraploid varieties of rye grass has a higher affinity for carbon dioxide than does the same enzyme from diploid varieties. Diploid leaves have a higher photorespiration rate than do tetraploid leaves. Because of the tremendous importance of decreasing photorespiration in agriculture, many wild plants should be screened for this enzyme with desired properties. Naturally occurring genes coding for efficient ribulose bisphosphate carboxylase could be isolated and introduced into plants of agronomic value. Cloning of the large subunit gene from chlamydomonasand maize and the small subunit gene from pea have already been achieved. I n uitro mutated versions of cloned gene can be put back into plant cells when such technology becomes available. Mutant genes that determine the synthesis of proteins stored in seeds could be selected to increase protein lysine content. Synthetic nucleotide sequences that will code for essential amino acids can also be fused to natural genes and inserted into plants. The nutritional value of plant protein could be improved this way (Ellis, 1979). Biomass could be used as a source of energy if photosynthetic efficiency could be increased, as basic processes of photosynthesis are fully understood (Kjaergaard, 1978).They can be mimicked to provide energy-like hydrogen, fixed carbon, and nitrogen (Ladha and Kumar, 978). The red photosynthetic bacterium Thiocapsa roseopersicina contains a thermostable hydrogenase

64

VEDPAL SINGH MALIK

that has a temperature optimum of 75°C and is stable in air. These hydrogenases can be used in biophotolytic systems that evolve hydrogen gas (I. N. Gogotov, personal communication). Application of recombinant DNA technology can speed up the genetic analysis of such useful organisms. A promiscuous sex factor has already been used to mobilize segments of the Rhodopseudomonas capsulata chromosome that codes for photopigment synthesis, reaction centers, and tryptophan synthetase (B. Marrs, personal communication). These R- prime plasmids can now be isolated and the photo pigment genes can then be cloned for use in studies of gene structure, function, and regulation. R. Haselkorn (personal communication) identified fragments of cyanobacterial DNA containing nif genes by hybridization with recombinant plasmids containing defined segments of the nif region of Klebsiella DNA (Puhler et al., 1979) and found a 10 kb fragment that contains DNA sequences complementary to Klebsiella nitrogenase genes. Plasmids of several Cyanobacteria can be used to mobilize chromosomal pieces in these organisms. Nuti et al. (1979) have documented hybridization between restriction fragments of plasmids from Rhizobium leguminosarum and nitrogenase gene DNA sequences from Klebsiella pneumoniae. These results suggest that some of the Rhizobium nif genes are plasmid-borne (Nuti et al., 1979). Genes required for nodulation are also carried on plasmids that are widespread among Rhizobium species (Zurkowski and Lorkiewicz, 1979). Klebsiella nif DNA-encoding nitrogenase can now be used to clone Rhizobium nif genes. A combined molecular and' genetic analysis of Rhizobium nif genes will now facilitate the study of symbiotic nitrogen fixation (Andersen, 1980).

VIII. A Postscript Lastly, the innermost causes and reasons for the creations of these various elemental beings is hidden from our understanding and knowledge. But at the end, all things shall be revealed to us, the greatest and the least, and.the reayn for all things. THEOPHRASTUS VON HOHENHEIM (PARACELSUS) (SIXTEENTH CENTURY)

Recombinant DNA technology has already uncovered many mysteries, for example, overlapping genes and intervening sequences. Several polypeptides of human origin have already been produced in E. coli and their commercialization in the near future is possible. The yield of other single gene products such as enzymes can definitely be improved by this approach and the technology could have impact on all microbial processes. Owing to the availability of many nucleic acid-modifying enzymes (Table V), rapid progress in gene isolation and gene sequencing makes possible exper-

TABLE V DNMRNA MODIFYING ENZYMES”

Application

Enzyme

Other intrinsic activities

Molecular weight

Most efficient substrate

Cofactors Mg2+ Mg2+

Nick translation Sequencing of DNA

3’,5’-Exonuclease 3’-Exonuclease

109,000 76,000

Synthesis of second strand of single-stranded DNA DNA synthesis

3’-Exonuclease

110,000

None

3’-Extensions

None

110,000220,OOO 32,460

cDNA copies of mRNA

RNase H

160,OOO

Poly(rA) oligo(dT)

Mgz+

Ligation

None

Poly(dAdT)

T4 DNA kinase

”P labeling of 5‘-end of RNMDNA

3’-Phosphatase

60,00068,000 140,000

Mg2+ Mn Mg2+

E . coli RNA Pol.

RNA copy of DNA

None

E . coli Pol. 1 E . coli large fragment of Pol. 1 T4 DNA Pol. Calf thymus a-DNA Pol. Terminal deoxynucleotidyl transferase AMV reverse transcriptase T4 DNA ligase

Poly(dA-dT) Activated calf thymus DNA Heat-denatured calf thymus DNA Poly(dT)oligo(rA)

Mgzi MgZi Mg2+ co 2+

Micrococcal nuclease-treated CNA Native calf thymus DNA

’+

Mg2+

(Continued)

TABLE V (Continued )

Enzyme

Other intrinsic activities

Application

3’-Phosphatase RNase H apurinic endonuclease None

Exo 111

Removal of 3’- and 5’-mononuclee tides of DNA

S, -Nuclease

Single-stranded-specificnuclease

Tobacco acid pyrophosphatase Bacterial alkaline phosphatase

Removes the cap structure from 5’end of mRNA Dephosphorylation of 5‘-RNA/ DNA

RNase Phy I

Sequence analysis of RNA

None

RNase H ( E . coli)

Degradation of RNA in RNA:DNA hybrid; removal of poly(A) from mRNA site-specificcleavage of RNA Analysis of RNA processing

None

RNase 111 (E. coU)

a Reproduced with permission of Bethesda Research

None None

None

Laboratories, Inc., RockviUe, Maryland.

Molecular weight

32,00036,000 270,000280,000 m000 86,000 89,000 25,000

50

Most efficient substrate

Cofactors

Sonicated H-E. coli DNA

Mg2+

Heatdenatured DNA Hydrolysis of ATP

ZnZ+

Hydrolysis of ATP

None

Cytosinedeficient RNA RNA:DNA hybrid

None

Poly(A,U)

Mkf+

None

Me+

RECOMBINANT DNA TECHNOLOGY

67

iments that could not be imagined before. Almost any piece of DNA can now be ligated to ribosomal (Schwarz and Kossel, 1980) or any other genes that should integrate into the chromosome of the host. The replication origin of many self-replicating genomes have already been well-characterized and provide another type of vector that can be used to autonomously replicate genes of interest. In the near future, E. coli will continue to be the Romeo of the molecular biology theater for unraveling the molecular basis of life. However, commercial aspects of the technology will definitely be realized in the years ahead by commitment of application of this new technology to the improvement of organisms currently being used in industry. Hybrid organisms can be obtained also by protoplast fusion. However, fusion of taxonomically different strains may produce very unstable strains owing to poorer genome homology than is desirable for recombination. This will further add to the already prevalent instability of industrial strains (Malik, 1979a). Furthermore, protoplast fusion does not offer the possibility of introducing and ampllfyingpredefined DNA sequences with known biological functions in an organism and is therefore not a viable substitute for recombinant DNA technology. REFERENCES Abelson, J. (1979). Annu. Rev. Biochem. 48, 1035. Adam, R. E., and Zimm, B. H. (1977). Nucleic Acids Res. 4, 1513. Aerts, M., Jacobs, M., Hemalsteens, J . P., Montagu, M. V., and Schell, J. (1979). Plant Sci. Lett. 17, 43. Alt, F. W., Kellems, R. E., Bertins, J. R., and Schimke, R. T. (1978). J . B i d . Chem. 253, 1357-1370. An, G., and Friesen, J . D. (1979). J . Bacteriol. 140, 400. Andersen, K. (1980). Trends Biochern. Sci. 5, (2) 35. Arnheim, N. (1979). Gene. 7, 83. Atrat, P., Deppmeyer, V., and Horhold, C. (1979a). Z. Allg. Mikrobiol. 19, 315. Atrat, P., Deppmeyer, V., Groh, H., and Horhold, C. (197913). Z. AZZg. Mikrobiol. 19, 375. Bach, M. L., Lacroute, F., and Botstein, D. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 386. Backman, K., and Ptashne, M. (1978). Cell 13, 65. Backman, K., Ptashne, M., and Gilbert, W. (1976). Proc. Natl. Acad. Sci. U.S.A. 73, 41744178. Bahl, C. P., Wu, R., Stawinski, J., and Narang, S. A. (1976a). Gene 1, 81. Bahl, C. P., Wu, R., Itakura, K., Katagiri, N., andNarang, S. A. (197613).Proc. Natl. Acad. Sci. U.S.A. 73, 91. Bahl, C . P., Wu, R., Stawinski, J., and Narang, S. A. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 966.

Bahl, C . P., Wu, R., Brousseau, R., Sood, A. K., Hsiung, H. M., and Narang, S. A. (1978). Biochem. Biophys. Res. Commun. 81, 95. Baldacci, P., Royal, A., Cami, B., Perrin, F.,Krust, A., Garapin, A., and Kourilsky, P. (1979). Nucleic Acids Res. 6, 2683.

68

VEDPAL SINCH MALIK

Barnes, W. M. (1070). Gene 5, 127. Barrell, B. G., Bankier, A. T., and Drouin, J, (1070). Nature (London)282, 180. Bassford, P., and Beckwith, J. R. (1070). Nature (London) 277, 538. Bassford, P., Silhavy, T. J., and Beckwith, J. (1070).J . Bacterfol. 130, 10. Bastia, D. (1078). J . Mol. Btol. 124, 601. Bauer, W. R. (1078). Annu. Reo. Biophys. Bioeng. 1, 287. Bazetoux, S., Jouanin, L., and Huguet, T. (1078). Nuckic Ad& Res. 5, 751. Beck, C. F. (1070). Proc. Natl. Acad. Sci. U.S.A. 76, 2376. Beckwith, J. R., and Zipfer, D., eds. (1070). “The Lactose Operon.” Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Beebee, T. J. C., and Carty, D. S. (1080). Anal. Biocha. 101,7. Beggs, J. D. (1078). Nature (London)275, 164. Behrens, B., Luder, G., Behncke, M.,Trautner, T. A,, and Ganesan, A. T. (1070). Mol. Gen. Genet. 175,351. Bell, G. I., Swain, W. F., Pictet, R., Cordell, B., Goodman, H. M.,and Rutter, W. J. (1070). Nature (London) 282,525. Belliard, G., Vedel, F.,and Pelletier, G. (1070). Nature (London)281, 401. Bennett, G. N., and Yanofsky, C. (1078).J . Mol. Biol. 121, 170. Benton, W. D., and Davis, R. W. (1077). Science 196, 180. Benzinger, R. (1078). Mtcrobiol. Reo.42, 194. Beremand, M. N., and Blumenthal, T. (1070). Cell 18, 257. Berg, P. (1070). CRC Crft. Reo. Biochem. 7, 75. Berg, P. E., Gayda, R., Avni, H., Zehnbauer, B., and Markovitz, A. (1076). Proc. Natl. Acad. Sci. U.S.A. 73, 607. Beringer, J. E., Brewin, N., Johnston, A. W. B., Schulman, H. M.,and Hopwood, D. A. (1070). Proc. R. SOC.London 204, 251. Berman, M. L., and Beckwith, J. (1970).J . Mol. Biol. 10, 303. Bernard, H. U.,Remaut, E., Hershfield, M.V., Das, H., Helinski, D. R., Yanofsky, C., and Franklin, N. E. (1070). Gene 5, 50. Bernhard, K., Schrempf, H., and Goebel, W. (1978). 1.Bncteriol. 133, 807. Beyreuther, K., and Gronenborn, B. (1076). Mol. Gen. Genet. 147, 115. Beyreuther, K., Adler, K., Geisler, N., and Klemm, A. (1073). Proc. Natl. Acad. Sci. U.S.A. 70,3576. Bibb, M.J., Ward, J., and Hopwood, D. A. (1078). Nature (London)274,308. Bingham, A. H. A., Bruton, C. J., and Aktinson, T. (1070). J . Gen. Mtcrobiol. 114,401. Birky, C. W. (1078). Annu. Reo. Genet. 12, 471. Blakesley, R. (1080). Focus 2, (6), 1. Bulletin of Bethesda Research Laboratories, Inc. Blattner, F. R., Williams, W. G., Blechl, A. E.,Denniston-Thompson, K., Faber, H. E., Furlong, L. A., Grunwald, D. J., Kiefer, D. O., Moore, D. D., Schumm, J. W., Sheldon, E. L.,and Smithies, 0. (1077). Sdence 196, 161. Bogenhagen, D. F., Sakonju, S., and Brown, D. D. (1060). Cell 10, 27. Boliver, F., Rodriquez, R. L., Green, P. J., Betlach, C., Heyneker, H. L., Boyer, H. W., Crosa, J. H., and Falkow, S. (1077). Gene 2,95. Bollum, F. J, (1074). Ado. Enzymol. 47, 347. Bollum, F. J. (1078). “Mechanisms in the Control of Neoplasia” (P. Chandra, ed.), p. 157. Plenum, New York. Bortrand, K., Korn, L.,Lee, F., Platt, T., Squires, C. L., Squires, C., and Yankofsky, C. (1957). Science 180,227. Bostian, K. A., Lee, R. C., and Halvorson, H. 0. (1070). Anal. Biochem. 05, 174.

RECOMBINANT DNA TECHNOLOGY

69

Boyer, H. W., Crosa, J. H., and Fakow, S. (1977).Gene 2, 95. Brack, C., and Tonegawa, S. (19n). Proc. Natl. Acad. Sci. U.S.A. 74, 5652. Brammar, W.J. (1977).Bfochem. SOC.Trans. 5, 1633. Brandt, A., and Ingversen, J. (1978).Carlsberg Ree. Commun. 43, 451. Breiner, A., Gusella, J. F., Keys, C., and Hbusman, D. E. (1979).Gene 1, 317. Brickman, E., Silhavy, T. J., Bassford, P. J., Shuman, H. A., and Beckwith, J. R. (1979).J . Bacteriol. 139, 13. Brill, W. J. (1979).Am. Sci. 67,458. Broach, J. R., Strathem, J. N., and Hicks, J. N. (1979).Gene 8, 121. Brodelius, P. (1979).Adu. Biochem. Eng. 10, 75. Broome, A., and Gilbert, W. (1978).Proc. Natl. Acad. Sci. U.S.A. 75, 2746. Brown, B. J., and Carlton, B. C. (1980).J . Bacteriol. 142, 508. Brown, K. D., Bennett, G . N., Lee, F., Schweingruber, E., andyanofsky, C. (1978).J . Mol. B b l . 121, 153. Brown, T. D. K., and Balmain, A. (1979).Nucleic Acids Res. 7,2357. Buchel, D. E., Gronenborn, B., and Muuer-Hill, B. (1980).Nature (London)283, 541. Bull, A. T., Ellwood, D. C.,and Radedge, C. (eds.). “Microbial Technology: Current State, Future Prospects.” Cambridge Univ. Press, London and New York. Burrell, C. J.. Mackay, P., Greenaway, P. J., Hofschneider, P. H., and Murray, K. (1979). Nature (London)270, 43. Bylinsky, G. (1980).Fortune, June 16,p. 144. Cdaham, D.,Tredic, P. D., andTorrey, J. G. (1978).Sdence 199, 899. Cdos, M. P., and Miller, J. H. (1980).CeU 20, 579. Cami, B., and Kourilsky, P. (1978).Nucbic Adds Res. 5, 2373. Canosi, U.,Morelli, G., and Trautner, T. A. (1978).Mol. Gen. Genet. 166,259. Carbon, J., Ratzkin, B., Clarke, L.,and Richardson, D. (1977).Miomi Winter Symp. 12,59. Case, M. E., Schweizer, M., Kushner, S. R., and Giles, N. H. (1979).Proc. Natl. Acad. Sci. U.S.A. 76,5259. Cassadaban, M. J. (1976).J . Mol. Biol. 104, 557. Casse, F., Boucher, C., Julliott, J. S., Michel, M., and Denarie, J. (1979).J . Gen. Microbiol. 113, 229. Catterall, J. F., Stein, J. P., Lai,E. C., Woo,S. L. C., Dugaiczyk, A,, Mace, M. L., Means, A. R.,and O’Malley, B. W. (1979).Nature (London)278, 323. Cech, T. R., and Rio, D. C. (1979).Proc. Natl. Acad. Sd. U.S.A. 76,5051. Chakrabarty, A. L. (1976).Annu. Reu. Genet. 10, 7. Chakrabarty, A. M. (1978).“Genetic Engineering.” CRC Press, Boca Raton, Florida. Chan, P. T., Lebowitz, J., and Bastia, D. (1979).Nucbic A& Res. 7, 1247. Chan, S. J., Noyes, B. E., Aganval, K. L., and Stein, R. (1979). Proc. Natl. Acad. Sd. U.S.A. 76,5036. Chandler, P. M., Rimkus, D., and Davidson, N. (1979).Anal. Bfochem. 99,200. Chang, A. C. Y., Lansman, R. A., Clayton, D. A,, and Cohen, S. N. (1975).CeU 6, 231. Chang, A. C. Y., Nunberg, J. H., Kauhan, R. J., Erlich, H. A., Schimke, R. T., andCohen, S. N. (1978).Nature (London)275,617. Chang, S., and Cohen, S. N. (1979).Mol. Gen. Genet. 168, 111. Charley, R. C., and Bull, A. T. (1979).Arch. Microbiol. 123,239. Chater, K. F., and Carter, A. T. (1978).J . Cen. Microbiol. 109, 181. Chater, K. F., and Carter, A. T. (1979)./. Gen. Microbiol. 115,431. Chevallier, M. R., and Aigle, M. (1979).FEBS Lett. 108, 179. Chi, N. W.,Ehrlich, S. D., and Lederberg, J. (1978).J . Bacteriol. 133, 816.

70

VEDPAL SINGH MALIK

Chinault, A. C., and Carbon, J. (1979). Gene 5 , 111. Chirgwin, J. M., Przybyla, A. E., MacDonald, R.J., and Rutter, W. J. (1979). Biochemistry 18, 5294. Chou, J., Casadaban, M. J., Lemaux, P. G., and Cohen, S. N. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4020. Citron, B. A., Feiss, M., and Donelson, J. E. (1979). Gene 6, 251. Clarke, L., and Carbon, J. (1976). Cell 9, 91. Clarke, L., and Carbon, J. (1978). J . Mol. Biol. 120, 527. Clewell, D. B. (1972). J . Bacteriol. 110, 667. Clewell, D. B., Yagi, Y., Dunny, G. M., and Schultz, S. K. (1974).J . Bacteriol. 117, 383. Cochet, M., Gannon, F., Hen, R., Maroteaux, L., Perrin, F., and Chambon, P. (1979). Nature (London) 282,567. Cockburn, A. F., Newkirk, M. J., and Firtel, R. A. (1976). Cell 9, 605. Cohen, G., Zimmer, Z., Gurevich, R., and Yanofsky, S. (1978). Mol. Gen. Genet. 166, 305. Cohen, S. N. (1975). Sci. Am. 233,24. Cohen, S. N., and Cbang, A. C. Y. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 1293. Cohen, S. N., Chang, A. C. Y., and Hsu, L. (1972). Proc. Natl. Acad. Sci. U.S.A. 69, 2110. Cohen, S. N., Chang, A. C. Y., Boyer, H. W., and Helling, R. B. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 3240. Collins, C. J., Jackson, D. A,, and deVries, F. A. J. (1976). Proc. Natl. Acad. Sci. U.S.A. 73, 3838. Collins, J. (1977). C u r . Top. Microbiol. Zmmunol. 78, 122. Collins, J. (1979). Gene 6, 29. Collins, J., and Bruning, H. J. (1978). Gene 4, 85. Collins, J., and Hahn, B. (1978). Proc. Natl. Acad. Sci. U.S.A. 9, 4242. Colman, A., Byers, M. J., Primrose, S. B., and Lyons, A. (1979). Eur. J. Biochem. 91, 303. Commerford, S. L. (1971). Biochemistry 16, 1993. Cooper, J. A., and Moss, B. (1979). Nuleic Acids Res. 6, 3599. Cordell, B., Bell, G., Tischer, E., Denoto, F. M., Ullrich, A., Pictet, R.,Rutter, W. J., and Goodman, H. M. (1979). Cell 18, 533. Cory, S., Jackson, J., and Adams, J. M. (1980). Nature (London) 285, 450. Cosloy, S. D., and Oishi, M. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 84. Coutelle, Ch., Ioannou, P., and Williamson, R. (1978). Gene 3, 117. Crabeel, M., Charlier, D., Weyens, G., Feller, A., Pierard, A., and GlansdorE, N. (1980). 1. Bacteriol. 143, 921. Crawford, G. P., and Stauffer, D. V. (1980). Annu. Reo. Biochem. 49, 163. Crea, R.,Kraszewski, A., Hirose, T., and Itakura, K. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 57655769. Creighton, T. E. (1979). J . Mol. Biol. 128, 235. Crick, F. H. C. (1978). Eur. J. Biochem. 83, 1. Currier, T. C., and Nester, E. W. (1976). Anal. Biochem. 76, 431. Dagert, M., and Ehrlich, S. D. (1979). Gene 6,23. Dale, R. M. K., and Ward, D. C. (1975). Biochemistry 14, 2458. Dale, R. M. K., Martin, E., Livingston, D. C., and Ward, D. C. (1975). Biochemistry 14,2447. Darnell, J . E. (1978). Science 202, 1257. Davis, B. D. (1977). Jpn. J. Antibiot. 30, 262. Davis, S. B. D., and Tai: P. C. (1980). Nature (London) 283,433. Dawid, I. B., and Wahli, W. (1979). Dew. Biol. 68, 305-328. Debarbouille, M., and Schwartz, M. (1979). J . Mol. Biol. 132, 521. Deddish, P. A., and Ravin, A. W. (1979). J . Bacteriol. 138, 94.

RECOMBINANT DNA TECHNOLOGY

71

Delisle, A. L., Nauman, R. K., and Minah, G. E. (1978). Infect. Immun. 20, 303. Denhardt, D. T., Dressler, D., and Ray, D. S. (1979). “The Single-Stranded DNA Phages. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. DeVries, F. A. J., Collins, C. J., and Jackson, D. A. (1976). Biochim. Biophys. Acta 435,213. Dickson, R. C., and Martin, J. S. (1978). Cell 15, 123. Dimitriadis, G. J. (1978). Nucleic Acids Res. 5 , 1381. Dimitriadis, G. J. (1979). Nucleic Acids Res. 6, 2697. Din, N., and Engberg, J. (1979). J . Mol. B i d . 134, 555. Dodgson, J. B., Strommer, J., and Engel, J. D. (1979). Cell 17, 879. Domino, F. A. (1979).“Energy from Solid Waste.” Noyes Data Corporation, Park Ridge, New Jersey. Donoghue, D. J., and Sharp, P. A. (1978). J. Bacteriol. 138, 1192. Driesel, A. J., Crouse, E. J.. Gordon, K., Bohnert, H. J., and Herrmann, R. G. (1979). Gene 6, 285. Drurnmond, M. (1979). Nature (London) 281,343. Dudits, D., Hadlaczky, G . , Bajszar, G. Y.,Koncz, C., Lazar, G., and Horvath, G . (1979). Plant Sci. Lett. 15, 101. Dugaiczyk, A., Boyer, H. W., and Goodman, H. M. (1975). 1.Mol. Biol. 96, 171. Dunn, R., Delaney, A. D., Gillam, I. C., Hayashi, S., Tener, G. M., Grigliatti, T., Misra, V., Spurr, M. G., Taylor, D. M., and Miller, R. C. (1979). Gene 1, 197. Eckhardt, T. (1978). Plasmid 1, 584. Ehrlich, S. D. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 1680. Ehrlich, S. D. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 1433. Ehrlich, S. D., Burstyn-Pettegrew, H., Stroynowski, I., and Lederberg, J. (1976). Proc. Natl. Acad. Sci. U.S.A. 72, 4145. Ellis, J. G., Kerr, A., Tempe, J., and Petit, A. (1979). Mol. Gen. Genet. 173, 263. Ellis, R. J. (1979). Trends Biochem. Sci. 11, 241. Emtage, J. S., Tacon, W. C. A., Catling, H., Jenkins, B . , Porter, A. G . , and Carey, N. H. (1980). Nature (London) 283, 171. Enouye, M., and Halegou, S. (1980). CRC Crit. Reo. Biochem. 7, 339. Enquist, L., Tiemeier, D., Leder, P., Weisberg, R., and Sternberg, N. (1976). Nature (London) 259,596. Evans, G . A., and Rosenfeld, M. G. (1979). J . Biol. Chem. 245, 8023. Farabaugh, P. J., and Fink, G. R. (1980). Nature (London) 286, 352. Farrell, R., Gunsalus, I. C., Crawford, I. P., Johnson, J. B., and Ito, J. (1978). Biochem. Biophys. Res. Commun. 82, 411. Fedoroff, N. V. (1979). Cell 16, 697. Fiddes, J . C., and Godson, G . N. (1979). J . Mol. Biol. 133, 19. Fiddes, J . C., and Goodman, H. M. (1979). Nature (London) 281, 351. Fiddes, J. C., Seeburg, P. H., Denoto, F. M., Hallewell, R. A., Baxter, J. D., andGoodman, H. M. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4294. Firtell, R. A., Cockburn, A., Frankel, G., and Hershfield, V. (1976). J . Mol. Biol. 102, 831. Fischer, S. G., and Lerman, L. S. (1979). Cell 16, 191. Fischer, R. A., and Turner, N. C. (1978). Annu. Reo. Microbial. 29, 277. Flavell, R. A., Bernards, R., Kooter, J. M., deBoer, E., Little, P. F. R., Annison, G . , and Williamson, R. (1979). Nucleic Acids Res. 6, 2749. Fonty, G., Goursot, R., Wilkie, D., and Bernardi, G. (1978). J . Mol. Biol. 119, 213. Fonty, G., Culard, F., Baldacci, G., Goursot, R., Prunell, A,, and Bernardi, G. (1979).J . Mol. Biol. 134,493. Fraley, R. T., Fornari, C. S., and Kaplan, S. (1979). Proc. Natl. Acad. Sci. U.S.A. 66,3348.

72

VEDPAL SINGH MALIK

Franche, U., and Taggart, R. T. (1980). Proc. Natl. Acad. Sci. U.S.A. 77,3595. Franklin, N. C. (1978). Annu. Reo. Genet. 12, 193. Fraser, T. H., and Bruce, B. J. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 996. Fritsch, E. F., Lawn, R., and Maniatis, T. (1979). Nature (London) 279,598. Fuchs, P. G., and Dobmnski, W. T. (1978). J . Bacterbl. 136, 295. Fujii, T., Shibuya, T., and Matsumoto, K. (1979). C h a . SOC.Spec. Publ. Gafner, J., and Philippsen, P. (1980). Nature (London) 286, 414. Gang, D. M., and Shaikh, K. (1976). Biochim. Biophys. Acta 425, 110. Gantotti, B. V.,Patil, S. S., and Mandel, M. (1979). Mol. Gen. Genet. 174, 101. Garapin, F., Chousterman, S., Horodniceanu, F., Kourilsky, P., and Garapin, A. C. (1979). Proc. Natl. Acad. Sci. U.S.A. 76,3755. Gardner, J. F. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 1706. Gelfand, D. H., Shepard, H. M., O’Farrell, P. H., and Polisky, B. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 5869. Gerbaud, C., Fournier, P., Blanc, H., Aigle, M., Heslt, H., and Guernequ, M.’(1979). Gene 5, 233.

Gergen, J. P., Stem, R. H., and Wensink, P. C. (1979). Nucleic A& Res. 7, 2115. Ghangas, G., and Wu, R. (1975). J . B w l . Chem. 18, 387. Gilbert, F., Mandart, E., Fitoussi, F., Tiollais, P., and Charnay, P. (1979). Nature (London) 281,646.

Gilbert, W., and Maxam, A. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 3581. Gilbert, W., and Villa-Komaroff, L. (1980). Sd.Am., May. Gillam, S., and Smith, M. (1979a). Gene 8, 81. Gillam, S., and Smith, M. (1979b). Gene 8, 99. Gillam, S., Jahnke, P.,Astell, C., Phillips, S., Hutchison, C. A., and Smith, M. (1979). Nucleic AcidP Res. 6, 2973. Glaser, G., Enquist, L., and Cashel, M. (1977). Gene 2, 159. Gleba, Y. Y., and H o h a n n , F. (1980). Planta 149, 112. Glover, D. M., and Hogness, D. S. (1977). Cell 10, 167. Goeddel, D. V., Kleid, D. G., Bolivar, F., Heyeneker, H. L.,Yansura, D. G., Crea, R., Hirose, T., Kraszewski, A., Itakura, K., and Riggs, A. D. (1979a). Proc. Natl. Acad. Sd. U.S.A. 76, 106. Goeddel, D. V., Heyneker, H., Hozumi, T., Arentzen, R.,Itakura, K., Yansura, D. G., Ross, M. J., Miozzari, G., Crea, R.,and Seeburg, P. H. (1979b). Nature (London) 281, 544. Goff, S. P., and Berg, P. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 1763. Goff, S. P., and Berg, P. (1979). J . Mol. Biol. 133, 359. Gonzalez, J. M., and Carlton, B. C. (1980). Plasmid 3, 171. Goodchild, D. J. (1977) Znt. Reo. Cytol. 6, 235-288. Goodman, H. M., Olson, M. V.,and Hall, B. D. (1977). Proc. Natl. Acad. Sci. U.S.A. 74,5453. Graf, L., Kossel, H.,and Stutz, E. (1980). Nature (London) 286, fN8. Graham, M. Y., and Weisblum, B. (1979). J . Bacterlol. 13, 1464. Graham, S., Yoneda, Y., and Young, F. E. (1979). Gene 7,69. Gromkova, R., and Goodgal, S. (1979). Biochem. Biophys. Res. Commun. 88, 1428. Gronenborn, B., and Messing, J. (1978). Nature (London) 272,375. Gross, D. C., Vidaver, A. K., and Klucas, R. V. (1979). J . Cen. Microbtol. 114, 257. Grossmann, M. K., and Zimmermann, F. K. (1979). Mol. Gen. Genet. 175, 223. Grunstein, M., and Hogness, D. S. (1975). Roc. Natl. Acad. Sci. U.S.A. 72, 3961-3965. Gryczan, T. J., and Dubnau, D. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 1428. Gryczan, T. J., Contente, S., and Dubnau, D. (1978). J . Bocteriol. 134, 318. Guarente, L., Lauer, G., Roberts, T. M., and Ptashne, M. (1980). Cell 20,543.

RECOMBINANT DNA TECHNOLOGY

13

Gueny, P., LeBlanc, D. J., and Fakows (1973). J . Bacteriol. 116, 1064. Gunsalus, R. P., Zurawski, G., and Yanofsky, C. (1979). J . Bacteriol. 140, 106. Curskaya, G. V., Grokhovsky, S. L., Zhuze, A. L., and Gottiks, B. P. (1978). Biochim. Biophys. Acta. 563,336. Gusella, J., Breiner, A,, Kaot, P. J . , Jones, C., Puck, T. T., Keys, C., Orkin, S., and Housman, D. (1979). Proc. Natl. A d . Sci. U.S.A. 76, 5239. Haber, J., Mascioli, D. W., and Rogers, D. T. (1980). Cell 20, 519. Hackel, R. A., and Khan, N. A. (1978). Mol. Gen. Genet. 164, 295. Haigh, L. S., Anderson, W. F., and Francke (1980). Nature (London) 263,683. Haldenwang, W. G., and Losick, R. (1979). Nature (London) 282,256. Hamer, D. H., and Leder, P. (1979). Nature (London) 281,35. Hansen, J., and Olsen, R. (1978). J . Bacteriol. 135, 227. Hardies, S. C., and Wells, R. D. (1976). Proc. Natl. Acad. Sci. U.S.A. 73,3117. Hardies, S. C., and Wells, R. D. (1979). Gene 7, 1. Hardies, S. C., Patient, R. K., Klein, R. D., Ho, F., Reznikoff, W. S., and Wells, R. D. (1979). J . Biol. Chem. 254, 5527. Harpold, M. M., Dobner, P. R.,Evans, R. M., and Bancroft, F. C. (1978). N u c k Acids Res. 5, 2039. Harris, C. C., Yolken, R. H., Krokan, H., and Hsu, I. C. (1979). Proc. Natl. Acad. Sci. U.S.A. 76,5336. Hautala, J. A,, Schweizer, M., Giles, N. H., andKushner, S. R.(1979). Mol. Gen. Genet. 172, 83. Hayakawa, T., Otake, N., Yonehara, H., Tanaka, T., and Sakaguchi, K. (1979)./. Antibiot. 32, 1348. Heber, V., and Krause, G. H. (1980). Trends Biochem. Sci. 5 (2), 32. Hedden, P., MacMillan, J., and Phinney, B. 0. (1978). Annu. Reo. Plant Physiol. 29, 149. Helinski. D. R. (1979). CRC Crit. Reo. Biochem. 7, 83. Helling, R. B., Goodman, H. M., and Boyer, H. W. (1974). J . Virology 14, 1235. Henning, V., Royer, H., Teether, R. M., Hindennach, I., and Hollenberg, C. P. (1979). Proc. Natl. Acad. Sci. U.S.A. 76,4360. Hentschel, C., Irminger, J. C., Bucher, P., and Birnstein, M. (1980). Nature (London) 285, 147-151. HersMeld, V., Boyer, H. W., Yanofsky, C., Lovett, M. A., and Helinski, D. R. (1974). Proc. Natl. Acad. Sci. U.S.A. 71, 3455. Herson, D., Schmidt, A., Seal, S., Marcus, A., and Doting, L. (1979). J . Biol. Chem. 254, 8245. ‘Heyneker, H. L., Shine, J., Goodman, H. W., Boyer, H., Rosenberg, J., Dickerson, R. E., Narang, S. A., Itakura, K., Lin, S., and Riggs, A. D. (1976). Nature (London) 263, 748. Hibner, U.,and Albert, 8. M. (1980). Nature (London) 285, 300. Hicks, J. B.,Hinnen, A., and Fink, G. R. (1979a). Cold Spring Harbor Symp. Quant. Biol. 38, 1305. Hicks, J. B., Strathem, J. N., and Klar, A. J. S. (1979b). Nature (London) 282,478. Hinnen, A., Hicks, J. B., and Fink, J . R. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 1929. Hirota, Y., Fujii, T., Sano, Y., and Iyama, S. (1978). Nature (London) 276, 416. Hoch, J. A. (1976). Ado. Genet. 18, 69. Hoffmann, F. (1980). Natutwissenschafen 67,301. Hofstetter, H., Schambock,A., VandenBerg, J., and Weissmann, C. (1976). Biochim. Biophys. Acta. 454, 587. Hollaender, A. ed. (197‘7). “Genetic Engineering for Nitrogen Fixation.” Plenum, New York. Hooykaas, P. J. J., Roobol, C., and Schilpemrt, R. A. (1979). J . Gen. Mhobiol. 110, QQ.

74

VEDPAL SINGH MALIK

Horinouchi, S., Vozumi, T., Hoshino, T., Ozaki, A., Nakajima, S., Beppu, T., and Arima, K. (1978).Mol. Gen. Genet. 157, 175-182. Howley, P. M., Israel, M. A., Law, M., and Martin, M. A. (1979).J . B i d . Chem. 254,4876. Hranueli, D., Pigac, J., and Vesligaj, M. (1979).J . Gen. Microbiol. 114,295. Hsiao, C., and Carbon, J. (1979).Proc. Natl. Acad. Sci. U.S.A. 76,3829. Hsiung, H. M., Brousseau, R., Michniewicz, J., and Narang, S. A. (1979).Nucleic Acids Res. 6 ,

1371. Hull, R., Covey, S. N., Stanley, J., and Davies, J. W. (1979).N u c k Acids Res. 1, 669. Humayun, M. Z., and Chambers, R. (1979).Nature (London) 278, 524. Humphries, P., Old, IT..,Coggins, L. W., Meshane, T., Watson, C., and Paul, J. (1978). Nucleic Acids Res. 5, 905. Iordanewu, S. (1975).J . Bacteriol. 124,597. Itakura, K., Bahl, C. P., Katagiri, N., Michniewicz, J. J., Wightman, R. H., and Narang, S. A. (1973).Can. J. Chem. 51,3649. Itakura, K., Hirose, T., Crea, R., Rigs, A. D., Heyneker, H. L., Boliver, F., and Boyer, H. W. (1977).Science 198, 1056. Jackson, D. A., Symons, R. H., and Berg, P. (1972).Proc. Natl. Acad. Sci. U.S.A. 69,29042909. Johnson, J. B., and Gunsalus, I. C. (1977).Biochem. Biophys. Res. Commun. 75, 13. Johnston, A. W. B., Setchell, S. M., and Berringer (1978).J . Gen. Microbiol. 104,209. Johnstrud, L. (1978).Proc. Natl. Acad. Sci. U.S.A. 75,5314. Jovin, T. M., Geisler, N., and Weber, K. (1977).Nature (London) 269, 668. Kaguni, J., and Ray, D. S. (1979).J . Mol. B i d . 135,863. Kania, J., and Brown, D. T. (1976).Proc. Natl. Acad. Sci. U.S.A. 73,3529. Kania, J., and Muller-Hill, B. (1977).Eur. J. Biochem. 79,381. Katagiri, N., Itakura, K., and Narang, S. A. (1975).J . Am. Chem. Soc. 97, 7332. Kato, K., Umeda, Y., Suzuki, F., Hayashi, D., and Koska, A. (1979).Clin. C h m . Winston Salem N.C. 25, 1306. Kedes, L. H., Chang, A. C. Y., Houseman, D., and Cohen, S. N. (1975).Nature (London)255, 533. Keggins, K. M., Duvall, E. J., and Lovett, P. S. (1978a).J . Bacteriol. 134,514. Keggins, K. M., Lovett, P. S., and Duvall, E. J. (197813).Proc. Natl. Acad. Sci. U.S.A. 75, 1423. Keggins, K. M., Lovett, P. S., Marrero, R., and Hoch, S. (1979).J . Bacteriol. 139, 1001. Kelley, W. S., Chalmers, K., and Murray, N. E. (1977).Proc. Natl. Acad. Sci. U.S.A.74,5632. Kemp, D.J., Cory, S., and Adams, J. M. (1979).Proc. Natl. Acad. Sci. U.S.A. 76,4627. Kenerley, M. E.,Morgan, E. A., Post, L., Lindahl, L., and Nomura, M. (1977).J . Bacteriol. 132, 931-949. Kharatyan, S. G . (1978).Annu. Reo. Microbiol. 32, 301. Khorana, H. (1978).Biourg. C h m . 7, 351. Khorana, H. G . (1979).Science 203,614. Khorana, H. G.,Aganval, K. L., Bresmer, P., Buchi, H., Caruthers, M. H., Cashion, P. J., Fridkin, M., Jay, E., Kleppe, K., Kleppe, R.,Kumar, A., Loewen, P. C., Miller, R. C., Minamoto, K., Panet, A,, RajBhandary, U. L., Ramamoorthy, B., Sekiya, T., Takeya, T., and van de Sande, J. H. (1976).J . Biol. C h a . 251,565. Kingsman, A. J., Clarke, L., Motrimer, R. K., and Carbon, J. (1979).Gene 1, 141. Kjaergaard, L. (1978).Biotechnol. Bioeng. 20, 1691. Klaus, S . , Suss, F., Juch, C., and Hauck, A. (1978).2. All. Mikrobiol. 18,375. Klaus, S., Triebel, H., Hartmann, M.,Walter, A., Walter, F., Zopel, P., Bar, H., and Hradecna, 2.(1979).Mol. Gen. Genet. 172,312.

RECOMBINANT DNA TECHNOLOGY

75

Klein, B., and Murray, K. (1979). J. Mol. Biol. 133, 289. Klein, R. D., Selsing, E., and Wells, R. D. (1980). Plasmid 3, 88. Kleinkauf, H., and Koischvitz, H. (1980). In “Multifunctional Proteins” (H. Bisswanger and E. Schmincke-Ott, eds.), p. 217. Wiley, New York. Klier, A. F., Kunst, F., and Rapoport, G. (1979). Nucleic Acids Res. 1, 997. Klobutcher, L. A., and Ruddle, F. H. (1979). Nature (London)280, 657. Klobutcher, L. A., Miller, C. L., and Ruddle, F. H. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 3610. Knopf, U. C. (1979). Sub-cell. Biochem. 6, 143. Kochert, G. (1978). Annu. Reo. Plant Physiol. 29, 461. Koekman, B. P., Ooms, G., Klapwijk, P. M., and Schilperoort, R. A. (1979). Plasmid 2,347. Komeda, Y.,and Iino, T. (1979). J . Bacteriol. 139, 721. Konkel, D. A., Tilghman, S. M., and Leder, P. (1978). Cell 150, 1125. Kom, L. J., Birkenmeier, E. H., and Brown, D. D. (1979). Nuckic Acids Res. 7, 947. Koshland, D., and Botstein, D. (1980). Cell 20, 749. Kourilsky, P., and Chambon, P. (1978). Trends Biochern. Sci. 2, 244. Kretschmer, P. J . , Chang, A. C. Y., and Cohen, S. N. (1975). J . Bacteriol. 124, 225. Kreuzaler, F., Ragg, H., Heller, W., Tesch, R.,Witt, I., Hammer, D., and Hahlbrock, K. (1979). Bur. J . Biochern. 99, 89. Kronenberg, H. M., McDevitt, B. E., Majzoube, J. A . , Nathans, J., Sharp, P. A., Potts, J . T., and Rich, A. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4981. Kushner, P. J., Blair, L. C., and Herskowitz, I. (1979). Proc. Natl. Acad. Sci. U.S.A. 76,5264. Kushner, S. R. (1978). Genet. Eng. 17, 178. Lacy, G. H., and Leary, J. V. (1979). Annu. Reo. Phytopathol. 17, 181. Ladha, J. K., and Kumar, H. D. (1978). B i d . Reo. 53, 355. Lampen, J. 0. (1978). Syrnp. Soc. Gen. Microbiol. 28, 283. Langley, K. E., Villarejo, M. R., Fowler, A. V., Zamenhof, P. J., and Zabin, I. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 1254. Latrou, K., Tsitilou, S. G., Goldsmith, M. R.,and Kafatos, F. C. (1980). Cell 20, 659. Lau, P. B., Yu, S. H., Spring, T. G., and Gray, H. B. (1979). Biochirn. Biophys. Acta. 560,313. LeBlanc, D. J., and Lee, L. N. (1979). J . Bacteriol. 140, 1112. Lederberg, E. M., and Cohen, S. (1974). J . Bacteriol. 119, 1072. Lehman, I. R. (1974). Science 186, 790. Levings, C. S., and Pring, D. R. (1976). Science 193, 156. Little, P., Curtis, P., Coutelle, Ch., Van Den Berg, J., Dalgleish, R.,Malcolm, S., Courtney, M., Westaway, D., and Williamson, R. (1978). Nature (London) 273, 640. Little, P. F. R., Annison, G., Darling, S., Williamson, R., Camba, L., and Modell, B. (1980). Nature (London) 285, 144. Liu, C. P., Slate, D. L., Gravel, R.,and Ruddle, F. H. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4503. Livingston, D. M. (1977). Genetics 86, 73. Livingston, D., and Hahne, S. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 3727. Lobban, P. E., and Kaiser, A. D. (1973). J . Mol. Biol. 78,453. Locker, J. (1979). Anal. B i o c h a . 98, 358. Lofdahl, S., Sjostrom, J. E., and Philipson, L. (1978a). Gene 3, 161. Lofdahl, S., Sjostrom, J. E., and Philipson, L. (1978b). Gene 3, 149. Lomax, M. I., Helling, R. B., Hecker, L. I., Schwartzbach, D. S., and Barnett, W. E. (1977). Science 196, 202. Lomedico, P., Rosenthal, N., Efstratiadis, A., Gilbert, W., Kolodner, R., andTizard, R. (1979). CeU 18, 545.

76

VEDPAL SINGH MALIK

Lomovskaya, N. D., Chater, K. F., and Mkrtumian, N. M. (1980). Mfcrobfol. Reu. 44,206. Long, E. O., and Dawid, I. V. (1980). Annu. Reu. Biochem. 49,727. Longacre, S., and Mach, B. (1979). Nuclefc Acids Res. 6, 1241. Longacre, S., Rougeon, F., and Mach, B. (1977). Experenttu 33,823. Losick, R.,and Chamberlin, M., eds. (1976). “RNA Polymerase.” Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Lovett, P. S.,and Burdick, B. D. (1973). Biochem. Biophys. Res. Cmmun. 54,365. Lovett, P. S., Duvall, E. J., and Keggins, K. M. (1976). J . Bacterfol. 127, 817. Lurquin, P. F. (1979). Nuclefc Acids Res. 6, 3773. McFarland, G. D., and Borer, P. N. (1979). Nuclefc A& Res. 7, 1067. McReynolds, L., O’MaUey, B. W., Nisbet, A. D., Fothergill, J. E., Givol, D., Fields, S., Robertson, M., and Brownlee, G. G.(1978). Nature (London) 273,723. Maizels, N. (1976). CeU 9, 431. Malik, V. S. (1964). “The Effect of Phosphobacterin and an Indian Culture on the Uptake of Nitrogen and Phosphorus by Sorghum Crop under Different Fertilizer Treatments.” Indian Agriculture Research Institute, New Delhi. Malik, V. S. (1972). Ado. Appl. Mfcrobiol. 15, 297. Malik, V. S. (1977). J . Antibiot. 30, 897. Malik, V. S. (1978). Nature (London)274, 844. Malik, V. S. (1979a). Ado. Genet. 20, 37. Malik, V. S. (1979b). Ado. Appl. Microbiol. 25,75. Malik, V. S. (197%). A.S.M. News45, 25. Malik, V. S. (1979d). A.S.M. N m 4 5 , 602. Malik, V. S. (1980). Trends Biochem. Sd. 37, 431. Malik,V. S., and Reusser, R. (1979). Plasmid 2, 627. Malik, V. S., and Vining, L. C. (1970). Can. J . Mfcrobtol. 16, 173. Malik, V. S.,and Vining, L. C. (1971). Can. J . Mfcrobiol. 17, 1287. Malik, V. S., and Vining, L. C. (1972a). Can.J . Microbfol. 18, 137. Malik, V. S., and Vining, L. C. (1972b). Can. J . Mfcrobiol. 18,583. Malnoe, P., Rochaix, J . D., Chua, N. H., and Spahr, P. F. (1979). J . Mol. Biol. 133,417. Manchee, R. (1979). Trends Btochem. Sci. 4, 77. Mandel, M., and Higa, A. (1970). J . Mol. Biol. 93, 159. Maniatis, T. (1980). I n “Cell Biology: A Comprehensive Treatise” (L. Coldstein and D. M. Prescott, eds.), Vol. 3, p. 606.Academic Press, New York. Maniatis, T., Sim, G. K., Efstratiadis, A., and Kafatos, F. C. (1976). Cell 8, 163. Maniatis, T., Hardison, R. C., Lacy, E., Lauer, J., O’Connell, C., Quon, D., Simm, G.K., and Efstratiadis, A. (1978). CeU 15, 687. Mansson, P. E., Carter, D. B.,Silverberg,A. B., Tuly, D. B., and Harris,S.E. (1979). Nucletc Acids Res. 7, 1553. Mantei, N., Boll, W., and Weissmann, C. (1979). Nature (London) 281, 40. Marians, K., Wu, R., Stawinski, J., Hozumi, T., and Narang, S. A. (1976). Nature (London) 263,744. Martens, P. A., and Clayton, D. A. (1977). Nuclefc Acids Res. 4, 1393. Martial, J. A., Hdewell, R. A., Baxter, J. D., and Coodman, H. M. (1979). Sdence 205,602. Marton, L., Wullems, G. J., Molendijk, L., and Schilperoort, R. A. (1979). Nature (London) 277, 129. Manola, D. L., and Bartholomew, D. P. (1979). Sdence 205,555. Maxam, A., and Gilbert, W. (1977). Froc. Natl. Acad. Sct. U.S.A. 74,560. Mayer, H., Collins, J., and Wagner, F. (1979). I n “The Environmental, Medical, and Commerc d Importance of Plasmids.” Elsevier-North Holland, Amsterdam.

RECOMBINANT DNA TECHNOLOGY

77

Mama, G., and Zalizzi, A. (1978). Microbbbgica 1, 111. Melchers, F., Potter, M., and Warner, N. L. (1978). Cum Top. Microbiol. Zmmunol. 81, 187. Melchers, G., Sacristan, M. D., and Holder, A. A. (1978). Carlsberg Res. C m m u n . 43,203. Mercer, A. A., and Loutit, J. S. (1979). J . Bacteriol. 140, 37. Mercereau-Puijalon, 0.. and Kourilsky, P. (1979). Nature (London)279,647. Mercereau-Puijalon, O., Royal, A., Cami, B., Garapin, A., Krust, A., Cannon, F., and Kourilsky, P. (1978). Nature (London) 275, 505. Mertz, J. E., and Davis, R. W. (1972). Proc. Natl. Acad. Sci. U.S.A. 69, 3370. Messing, J., Gronenborn, B.,Mullerhill, B., and Hofschneider, P. H. (1977). Proc. Natl. Acad. S c i . U.S.A. 74, 642. Mevarech, M., Noyes, B. E., and A g d , K. L. (1979). J . Biol. Chem. 245, 7472. Meyerink, J. H., Klootwijk, J., Planta, R. J., Ende, A., and Bruggen, F. (1979). Nucleic Acidp Res. 7, 69. Miller, J. H. (1972). “Experiments in Molecular Genetics,” p. 398. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Miller, J. H., and Reznikoff, W. S. (1978). “The operon.” Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Miller, L. (1980). Anal. Biochem. 100, 166. Mills, D. R., and Kramer, F. R. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 2232. Minson, A. C., Wildy, P., Buchan, A., and Darby, G. (1978). Cell 13, 581-587. Miozzari, G. F., and Yanofsky, C. (1978). Nature (London) 276,684. Mishra, A. K., Roy, P., and Bhattacharya, S. (1979). J . Bacteriol. 137, 1425-1427. Mishra, N. C. (1979). J . Gen. Microbiol. 113, 255. Miura, K., Law, S. W. T., Nishi, S., and Tamaoki, T. (1979). J . Biol. Chem. 254,55153521. Miwa, T., Takanami, M., and Yamagishi, H. (1979). Gene 6, 319-330. Molholt, B., and Doskocil, J. (1978). Biochm. Biophys. Res. Commun. 82, 477. Moore, L., Warren, G., and Strobel, G. (1979). Plasmid 2, 617. Moore, S. K., and James, E. (1979). Gene 5, 159. Moran, C. P., and Bott, K. F. (1979). J . Bacteriol. 140, 99-105. Morrison, D. A. (1977). J . Bacteriol. 132, 3494.51. Morrison, D. A., and Baker, M. F. (1979). Nature (London) 282, 215-217. Morrison, D. A,, and Mannarelli, B. (1979). /. Bacteriol. 146, 655-665. Morrow, J. F., Cohen, S. N.,Chang, A. C. Y., Boyer, H. W., Goodman, H. M., and Helling, R. B. (1974). Proc. Natl. Acad. Sci. U.S.A. 71, 1743-1747. Mortenson, L. E. (1978). Curr. Top. Cell. Regul. 13, 179-232. Mortenson, L. E., and Thorneley, R. (1979). Annu. Rev. Biochm. 48, 387. Moss, T., and Birnsteil, M. L. (1979). Nuckic Acids Res. 6, 3733. Mottes, M.,Morandi, C., Cremaschi, S., and Sgaramella, V. (1977). Nuckic Acids Res. 4, (7) 2467-2476. Mottes, M., Grandi, M., Sgaramella, V., Canosi, V., and Morelli, G. (1979). Mol. Gen. Genet. 174, 281-286. Muller, W., Weher, H., Meyer, F., and Weissmann, C. (1978). J . Mol. B i d . 124, 343. Muller-Hill, B. (1975). Prog. Biophys. Mol. Biol. 30, 227. Mufier-Hill, B., Heidecker, G., and Kania, J. (1976). Proc. John Znnes Symp. 3r4 1976. p. 167. Muuer-Hill, B., Gronenborn, B., Kania, J., Schlotmann, M., and Beyreuther, K. (1977). In “Nucleic Acid Protein Recognition” (H. J. Vogel, V. Bryson, and J . 0. Lampen, eds.) p. 219. Academic Press, New York. Mulligan, R. C., Howard, B. H., and Berg, P. (1979). Nature (London) 277, 108-114. Murphy, E., and Novick, R. P. (1979). Mol. Gen. Genet. 175, 19-30. Murray, K. (1976). Endeaoow 35, 129.

78

VEDPAL SINGH MALIK

Murray, M. G., Palmer, J . D., Cuellar, R. E., andThompson, W. F. (1979).Biochemistry 18,

5259. Murray, N. E., and Murray, K. (1974).Nature (London) 251, 476. Murray, N. E., Brammer, W. J., and Murray, K. (1977). Mol. Gen. Genet. 150, 53. Murray, N. E., Bruce, S. A,, and Murray, K. (1979).J . Mol. Biol. 132,493-505. Muzyczka, N. (1979).Gene 6, 107-122. Mylroie, J. R.,Friello, D. A., and Chakrabarty, A. M. (1978).Biochem. Biophys. Res. Commun. 82, 281. Nakamura, K., and Inouye, M. (1979).Cell 18, 1109. Nakanishi, S.,Inoue, A,, Kita, T., Nakamura, M., Chang, A. C. Y., Cohen, S. N., and Numa, S. (1979).Nature (London) 278,423427. Narang, S. A., Bahl, C. P., and Wu, R. (1977).Can. J . Biochem. 55, 1125. Narang, S. A,, Hsiung, H. M., and Brousseau, R. (1979).In “Methods in Enzymology” (R. Wu, ed.), Vol. 68, pp. 90-98. Nasyrov, Y. S. (1978).Annu. Reo. Plant Physiol. 29, 215. Nathans, D. (1979).Science 206, 891. Newman, E. I. (1978).B i d . Reo. 53, 511. Neyra, C. A., and Dobereiner, J . (1977).Ado. Agron. 29, 9. Niaudet, B., and Ehrlich, S. D. (1979).Plasmid 2, 4858. Nojiri, C., Watanabe, H., Katsumata, K., Yamada, Y., Murakami, T., and Kumata, Y. (1980).J . Antibiot. 33, 118. Norgard, M. V., Keem, K., and Monahan, J. J. (1978).Gene 3, 279. Norgard, M. V.,Emigholz, K., and Monahan, J. J. (1979).J . Bacteriol. 138,270-272. Norris, K. E., Iserentant, D., Contreras, R., and Fiers, W. (1979).Gene 1, 355. Nosikov, V. V., Braga, E. A., Karlishev, A. V., Zhuze, A. L., and Polyanovsky, 0. L. (1976). Nucleic Acids Res. 3, 2293. Noyes, B. E., Mevarech, M., Stein, R., and Agarwal, K. L. (1979).Proc. Natl. Acad. Sci. U.S.A. 76, 1770. Nuti, M. P., Lepidi, A. A., Prakash, R. K., Schilperoort, R. A,, and Cannon, F. C. (1979). Nature (London) 282,533. Ochi, K., Hitchcock, M. J. M., and Katz, E. (1979).J. Bacteriol. 139, 984-992. Ogata, R. T., and Gilbert, W. (1979).J . Mol. Biol. 132, 709-728. O’Hare, K., Breathnach, R., Benoist, C., and Chambon, P. (1979).Nucleic Acids Res. 7 (2),

305-320. Ohlsson, R., Hentschel, C. C., and Williams, J . G. (1978).Nucleic Acids Res. 5, 583. Okanishi, M., Manome, T., and Umezawa, H. (1980).J . Antibiot. 33, 88. Olson, M. V., Loughney, K., and Hall, B. D. (1979).J. Mol. Biol. 132,387410. Ooyen, A. V., Berg, J., Mantel, N., and Weissman, C. (1979).Science 206, 337. Orkin, S. H., Kolodner, R., Michelson, A., and Husson, R. (1980).Proc. Natl. Acad. Sci. U.S.A. 77,3558. O’SuUivan, M. J., Bridges, J. W., and Marks, V. (1979).Ann. Clin. Biochem. 16, 221. Page, W. J.. and Tigerstorm, M. (1979).J . Bacteriol. 139, 1058-1061. Palchaudhuri, S.,and Chakrabarty, A. (1976).J . Bacteriol. 126, 410416. Panasenko, S. M., Cameron, J . R., Davis, R. W., and Lehman, I. R. (1977).Science 196,

188. Panayatotos, N., and Wells, R. D. (1979).J . Biol. Chem. 254, 55555561. Pasek, M., Goto, T., Gilbert, W., Zink, B., Schaller, H., MacKay, P., Leadbetter, G., and Murray, K. (1979a).Nature (London) 282, 55. Pasek, M., Goto, T., Gilbert, W., Zink, B., Schaller, H., MacKay, P., Leadbetter, G., and Murray, K. (1979b).Nature (London) 282,575. Paterson, A., and Clarke, P. H. (1979).1. Gen. Microbiol. 114, 75-86.

RECOMBINANT DNA TECHNOLOGY

79

Paterson, B. M., Roberts, B. E., and Kuff, E. W. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 4370. Payvar, F., and Schimke, R. T. (1979). Eur. J. Biochem. 101, 271. Pellicer, A., Wigler, M., Axel, R., and Silverstein, S. (1978). Cell 14, 133-141. Pemberton, J. M., and Fisher, P. R. (1977). Nature (London) 268, 732-733. Perlak, F. J., Mendelsohn, C. L., and Thome, C. B. (1979).J . Bacteriol. 146, 699. Pemn, F., Cochet, M., Gerlinger, P., Cami, B., LePennec, J. P., and Chambon, P. (1979). Nucleic Acids Res. 6, 2731-2748. Persson, H., Perricaudet, M., Tolun, A , , Philipson, L., and Petersson, V. (1979). J . Biol. Chem. 245, 7999-8003. Perucho, M., Hanahan, D., Lipsich, L., and Wigler, M. (1980). Nature (London) 285, 207. Peterson, B. M., Roberts, B. E., and Kuff, E. L. (1977). Proc. Natl. Acad. Sci. U.S.A. 74,4370. Petes, J. D. (1980). Annu. Reu. Biochem. 49, 845. Petes, T. D. (1979). Proc. Natl. Aca. Sci. U.S.A. 76, 410414. Petes, T. D., and Botstein, D. (1977). Proc. Natl. Sci. U.S.A. 74, 50915095. Pfahl, M., Stockter, C., and Gronenbom, B. (1974). Genetics 76, 669. Polisky, B., Greene, P., Garfin, D. E., McCarthy, B. J., Goodman, H. M., and Boyer, H. S. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 3310. Poonian, M. S., Schlabach, A. J., and Weissbach, A. (1971). Biochemistry, 10, 424. Porter, A. G., Barber, C., Carey, N. H., Hallewell, R. A,, Threlfall, G . , and Emtage, J. S. (1979). Nature (London) 282, 471. Ptashne, M., Jeffrey, A., Johnson, A. D., Maurer, R., Meyer, B. J.. Pabo, C. O., Roberts, T. M . , and Sauer, R. T. (1980). Cell 19, 1. Puhler, A., Burkardt, H. J., and Klipp, W. (1979). Mol. Gen. Genet. 176, 17. Puijalon, 0. M., and Kourilsky, P. (1979). Nature (London) 279, 647-649. Purchio, A., James, E., and Fareed, G. C. (1979). Gene 7, 15. Radloff, R., Bauer, W., and Vinograd, J. (1967). Proc. Natl. Acad. Sci. U.S.A. 57, 1514-1520. Rambach, A., and Tiollais, T. (1974). Proc. Natl. Acad. Sci. U.S.A. 71, 3927. Rao, R. N., and Rogers, S. G. (1978). Gene 3, 247-263. Rao, R. N., and Rogers, S. G. (1979). Gene 7, 79-82. Rapoport, G., Klier, A,, Billault, A , , Fargette, F., and Dedonder, R. (1979). M o ~ .Gen. Genet. 176, 239. Ratzkin, B., and Carbon, J . (1977). Proc. Natl. Acad. Sci. U.S.A. 74,487. Reddy, V. B., Thimmappaya, B., Dhar, R., Subramanian, W., Zain, B. S., Pan, J., Ghosh, P. K., Celma, M. S., and Weissman, S. M. (1978). Science 200, 494. Ricciatdi, R. P., Miller, J. S., andRoberts, B. E. (1979).Proc. Natl. Acad. Sci. U.S.A. 76,4927. Rigby, P. W. J., Dieckman, M., Rhodes, C., and Berg, P. (1977).J. Mol. Bid. 113,237-251. Roberts, J. L., Seeburg, P. H., Shine, J., Herbert, E., Baxter, J. D., and Goodman, H. M. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 2153. Roberts, R. J. (1976). CRC Crit. Reo. Biochem. 4, 123. Roberts, R. J. (1978). Gene 4, 183. Roberts, T. M., Kacich, R., and Ptashne, M. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 760, 1979. Rodriquez, R. L., Bolivar, F., Goodman, H. M., Boyer, H. W., and Betlach, M. C. (1976). In “Molecular Mechanisms in the Control of Gene Expression” (D. P. Nierlich, W. J. Rutter, and C. F. Fox, eds.), pp. 471-477. Academic Press, New York. Rodriquez, R. L., West, R. W., Heyneker, H. L., Boliver, F., and Boyer, H. W. (1979). Nucleic Acids Res. 6, 3267. Roskam, W. G., and Rougeon, F. (1979). Nucleic Acids Res. 7,305-320. Roychoudhury, R., Jay, E., and Wu, R. (1976a). Nucleic Acids Res. 3, 101. Roychoudhury, R., Jay, E., and Wu, R. (1976b). Nucleic Acids Res. 3, 863.

80

VEDPAL SINCH MALIK

Rudkin, G. T., and Stollar, B. D. (1977). Nature (London) 265,472. Russell, G. E.(1978).“Plant Breeding for Disease Resistance.” Butterworth, London. Ryan, M. J., Brown, E.L.,Sekiya, T., Kupper, H., and Khorana, H. G. (197Qa).J . Biol. Chem. 2!54 (3),5871. Ryan, M. J., Belagaje, R.,Brown, E. L., Fritz, H., and Khorana, H. G . (1979b)./. B i d . C h a . 254,10,944. Sadler, J. R., Tecklenburg, M., Betz, J. L.,Goeddel, D. V., Yansura, D. G., and Caruthers, M. H. (1977). Gene 1,305. St. John, T. P.,and Davis, R. W. (1979).Cell 16,443. Sanger, F., and Coulson, A. R. (1975).J. Mol. Biol. 94,441. Sanger, F., Air, G. M., Barrell, B. G., Brown, N. L., Coulson, A. R.,Fiddes, J. C., Hutchinson, C. A., IH, Slocombe, P. M., and Smith, M. (1Q77a).Nature (London) 265, 687. Sanger, F., Nicklen, S., and Coulson, A. R. (1977b).Proc. Natl. Acad. Sci. U.S.A. 74,5463. Sarthy, A., Fowler, A., Zabin, I., and Beckwith, J. R. (1979).J . Bacterfol. 139,832-839. Sauer, R. T., Pabo, C. O., Meyer, B. J., Ptashne, M., and Backman, K. C. (1979).Nature (London) 279,3964. Schechter, I., Wolf, O., Zemell, R.,and Burstein, V. (1979).Fed. Proc. Fed. Am. SOC. Erp. Biol. 38, 1839-1846. Schell, M. A., and Wilson, D. B. (1979).Gene 5, 291. Scheller, R. H., Thomas, T., Lee, A., Niles, W., Klein, R., Britton, R., and Davidson, E. (1977).Sdence 196,197. Scherer, S., and Davis, R. W. (1979).Proc. Natl. Acad. Sd.U.S.A. 76, 4951. Schmitz, A., Schmeissner, U., and Miller, J. H. (1976).J . Biol. C h m . 251,3359. Schrempf, H.,Bujard, H., Hopwood, D. A., and Goebel, W. (1975).1.Bacterlol. 121,416. Schroeder, J. L.,and Blattner, F. R. (1978).Gene 4, 167-174. Schwan, Z.,and Kossel, H. (leS0).Nature (London) 283, 739. Schweitzer, S. M., and Matzura, H. (1977).Mol. Gen. Genet. 155,213. Scott, A. F., Philips, J. A., and Migeon, B. (1979).Proc. Natl. Acad. Sd. U.S.A. 76, 4563. Secher, D., and Berke, D. C. (1980).Nature (London) 285, 446. Seeburg, P. H., Shine, J., Martial, J. A., Baxter, J. D., and Goodman, H. M. (1977).Nature (tondon) 270,486. Seeburg, P. H., Shine, J., Martial, J. A., Ivarie, R. D., Morris, J. A., Ullrich, A., Baxter, J. D., and Goodman, H. M. (lQ78a).Nature (London) 276,795. Seeburg, P. H., Shine, J., Martial, J. A., Ivarie, R. D., Morris, J. A., Ullrich, A,, Baxter, J. D., and Goodman, H. M. (1978b).Sdence 276,5690. Segall, J., and Losick, R. (1977). Cell 11, 751. Seiki, M., Ogasawara, N., and YoShikawa, H. (1979).Nature (London) 281, 699. Serizawa, N.,Nakagawa, K., Kamimura, S., Miyadera, T., and Arai, M. (1979).J . Antibiot. 32, 1016. Seshadri, M., and Ranjekar, P. K. (1979).ZndianJ. Biochm. Btophys. 16,15. Setlow, 1. K., and Hollaender, A. (1979).“Genetic Engineering.” Plenum, New York. Sgaramella, V. (1972).Proc. Natl. A d . Sci. U.S.A. 69,3389. Sgaramella, V., VandeSande, J. H., and Khorana, H. G. (1970).Proc. Natl. Acad. Sd.U.S.A. 67, 1468. Shanmugam, K. T., and Valentine, R. C. (1975).Science 187,919. Shanmugam, K . T..O’Gara, F., Anderson, K., andvalentine, R. C. (1978).Annu. Reo. Plant Physiol. 29,263. Shaw, D. M.,and Langley, C. H. (1979).Nature (London)281,696-699. Shaw, J. E.,and Murialdo, H. (1980).Nature (London)283,30. Shenk, T.(1978).Cell 13, 791.

RECOMBINANT DNA TECHNOLOGY

81

Shepard, H. M., Gelfand, D. H., and Polisky, B. (1979). Cell 18, 267-275. Shih, T. Y.,and Martin, M. A. (1974). Biochemistry 13,3411. Shine, J., Seeburg, P. H., Martial, J. A., Baxter, J. D., and Goodman, H. M. (1977). Nature (London) 270,494. Shine, J., Fettes, I., Lan, N. C. Y., Roberts, J. L., and Baxter, J. D. (1980).Nature 285,456. Shoyab, M., and Sen, A. (1978).J . Biol. Chem. 253, 6654. Siebenlist, U., Simpson, R. B., and Gilbert, W. (1980). Cell 20,269. Silhavy, J. J., Casadaban, M. J., Shuman, J. A., and Beckwith, J. R. (1976). Proc. Natl. Acad. Sci. U.S.A. 73,342.3. Silhavy, T. J., Shuman, H. A., Beckwith, J., and Schwartz, M. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 5411. Silverman, M., and Simon, M. I. (1977). Annu. Reo. Microbiol. 31, 397419. Simons, M. D. (1979). Annu. Reo. Phylopathol. 17, 75. Simpson, J. S. A., Campbell, A. K.,Ryall, M. E. T., and Woodhead, J. S. (1979). Nature (London) 279,646-647. Sinsheimer, R. L. (1977). Annu. Reo. Biochem. 46,415. Sisco, K. L.,and Smith, H. 0. (1979). Proc. Natl. Acad. Sci. U.S.A. 76,972-977. Sjostrorn, J. E., Lindberg, M., and Philipson, L. (1972).J . Bacteriol. 109, 285. Sjostrom, J., Lofdahl, S., and Philipson, L. (1979). Plasmid 2, 529-535. Skalka, A. (1978). Gene 3,39. Skalka, A., and Shapiro, L. (1976). Gene 1 , G . Sladkova, 1. A., Chinenova, T. A., Lomovskaya, N. D., and Mkrtumian, N. M. (1979). Genetika 15, 1953. Slater, J. H., Lovatt, D., Weighhnan, A. J., Senior, E., and Bull, A. T. (1979). J . Cen. Microbiol. 114, 125. Smith, H. 0. (1979). Science 205,455461. Sodja, A., and Davidson, N. (1978). Nuclejc Acids Ree. 5, 385. Soni, G. L., and Bhatia, I. S. (1979). ZndianJ. Biochem. Biophys. 16, 223. Sood,A. K., and N m g , S. A. (1977). Nucleic Acids Ree. 4,2757. Southern, E. M. (1975). J. Mol. Biol. 98, 503. Southern, E. M. (1979a). Anal. Biochem. 100,304. Southern, E. M. (1979b). Anal. Biochem. 100,319. Stawinski, J., Hozumi, T., and Narang, S. A. (1976). Can. J. Chem. 54,670. Stawinski, J,, Hozumi, T., Narang, S. A,, Bahl, C. P., and Wu, R.(1977). Nuckic AddsRes. 4,

353. Stent, G. S.,and Calender, R. (1978). “Molecular Genetics.” Freeman, San Francisco. Sternberg, N., Tremeier, D., and Enquist, L. (1977). Gene 1, 255. Stinchomb, D. T., Struhl, K., and Davis, R. W. (1979). Nature (London)282,39. Storms, R. K., McNeil, J. B., Khandekar, P. S.,An, G., Parker, J., and Friesen. (1979).J . Bacteriol. 140,73. Strike, P., Humphreys, G. O., and Roberts, R. J. (1979).J . Bacteriol. 138, 1033. Struhl, K., and Davis, R. W. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 51555259. Struhl, K.,Cameron, J. R., and Davis, R. W. (1976). Proc. Natl. Acad. Sci. U.S.A. 73, 1471. Struhl, K.,Stinchcomb, D. T., Scherer, S., and Davis, R. W. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 1035-1039. Suarez, J. E., and Chater, K. F. (1980). Nature (London)286, 527. Sugimoto, K.,Oka, A., Sugisaki, H., Takanami, M., Nishimura, A., Yasuda, S.,and Hirota, Y. (1979). Proc. Natl. Acad. Sd. U.S.A. 76, 575579. Sugino, A. (1979). Biochem. Biophys. Res. Commun. 91, 1321.

82

VEDPAL SINGH MALIK

Sugino, A., Goodman, H. A., Heyneker, H. L., Shine, J., Boyer, H. W., andcoyzarelli, N. R. (1977). J . Biol. Chem. 252, 2987. Sung, W. L.,Hsiung, H. M., Brousseau, R., Micniewicz, J., Wu, R., and Narang, S. A. (1979). Nucleic Acids Res. 7, 2199. SutcliEe, J. G. (1978a). Cold Spring Harbor Symp. Quant. Biol. 43, 77. SutcliEe, J. G. (1978b). Nucleic Acids Res. 5, 2721. Szostak, J. W., and Wu, R. (1979). Plasmid 2, 536. Szybalski, E., and Szybalski, W. (1979). Gene 7,217. Tait, R. C., and Boyer, H. W. (1978). Mol. Gen. Genet. 164, 285. Taketa, A. (1972).J . Biochem. 72, 973. Talmadge, K., Stahl, S., and Gilbert, W. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 3369. Tanaka, T. (1979). Mol. Gen. Genet. 175,235. Tanaka, T., and Weisblum, B. (1975).J. Bacteriol. 121, 354. Taniguchi, T., O’Neill, M., and Crombrugghe, B. D. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 5090. Taylor, J . M. (1979). Annu. Rev. Biochem. 48, 681. Taylor, W. E., and Burgess, R. R. (1979). Gene 6,331365. Teather, R. M., Muller-Hill, B., Abrutsch, U.,Aichele, G., and Overath, T. (1978). Mol. Gen. Genet. 159, 239-248. Templeton, G. E., Tebeest, D. O., and Smith, R. J. (1979). Annu. Reo. Phytopathol. 17,301. Tenneson, M. E., Bilton, R. F., Drasar, B. S., and Mason, A. N. (1979). FEBS Lett. 102,311. Thayer, R. E. (1979). Anal. Biochem. 98, 60. Thomas, M., and Davis, R. W. (1975). J . Mol. B i d . 91, 315. Thomas, M., Cameron, J. R., and Davis, R. W. (1974). Proc. Natl. Acad. Sci. U.S.A. 71,4579. Thomas, M., White, R. L., and Davis, R. W. (1976). Proc. Natl. Acad. Sci. U.S.A. 73,2294. Thomashow, M. F . , Panagopoulos, C. S., Gordon, M. P., and Nester, E. W. (1980). Nature (London) 283, 794. Thompson, C. J., Ward, J. M., and Hopwood, D. A. (1980). Nature (London) 286, 525. Thompson, J. P., and Skerman, V. D. (1979).“Azotobacteraceae.” Academic Press, New York. Tiemeier, D., Enquist, L., and Leder, P. (1976). Nature (London)263, 526. Tiemeier, D. C., Tilghman, S. M., and Leder, P. (1977). Gene 2, 173. Tilghman, S. M., Tiemeier, D. C., Peterlin, B. M., Sullivan, M., Maize], J. V., and Leder, P. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 775. Timko, J., and Zelinka, J. (1977).J. Antibiot. 30, 644-648. Timmis, K., and Winkler, V. (1973). J . Bactm’ol. 113, 508-509. Timmis, K. N., Cabello, F., and Cohen, S. N. (1974). Proc. Natl. Acad. Sci. U.S.A. 71,4556. Timmis, K., Cabello, F., and Cohen, S. N. (1975). Proc. Nad. Acad. Sci. U.S.A. 72, 2242. Timmis, K. N., Cabello, F., andCohen, S. N. (1978a). Mol. Gen. Genet. 162, 121-138. Timmis, K., Cabello, F., and Cohen, S. N. (1978b). Prog. Mol. Subcell. B i d . 6, 1. Timothy, D. H., Levings, C. S., Pring, D. R., Conde, M. F., and Kermicle, J. L. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4220. Tonegawa, S., Brack, C., Hozain, N., and Schuller, R. (1977).Proc. Natl. Acad. Sci. U.S.A. 74, 3518. Trinick, M. J . (1973). Nature (London)244, 459. Troost, T. R., Danilenko, V. N., and hmovskaya, N. D. (1979).J . Bacteriol. 140, 359. Tu, C. D., Jay, E., Bahl, C. P., and Wu, R. (1976). Anal. Biochem. 74, 73. Twigg, A. J . , and Sherratt, D. (1980). Nature (London) 283, 216. Ucker, D. S., and Signer, E. R. (1978).J . Bacteriol. 136, 1197-1200. Uhlin, B. E., Molin, S., Gustafsson, P., and Nordstrom, K. (1979). Gene 6, 91. Ullrich, A., Shine, J., Chirgwin, J., Pictet, R., Tischer, E., Rutter, W. J., and Goodman, H. M. (1977). Science 196, 1313.

RECOMBINANT DNA TECHNOLOGY

83

Upcroft, P., Skolnick, H., Upcroft, J. A., Solomon, D., Khoury, G., Hamer, D. H., and Fareed, G. C. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 2117-2121. Valenzuela, P., Gray, P., Guiroga, M., zaldivar, J., Goodman, H. M., and Rutter, W. J. (1979). Nature (London) 280, 815-819. Vamvakopoulos, N. C., Monohan, J. J., and Kourides, I. A. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 3149. Vapnek, D., Hautala, J. A,, Jacobson, J. W., Giles, N. H., and Kushner, S. R. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 35083512. Venema, G. (1979). Ado. Microb. Physiol. 19, 245. Verdiere, J., and Petrochilo, E. (1979). Mol. Gen. Genet. 175, 209. Villa-Komaroff,L., Efstratiadis, A., Broome, S., Lomedico, P., Tizard, R., Naber, S. P., Chick, W. L., and Gilbert, W. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 37273731. Villarreal, L. P., and Berg, P. (1977). Science 196, 183. Volovitch, M., Drugeon, G., Dumas, J. P., Haenni, A. L., and Yot, P. (1979). Eur. J . Biochem. 100,245. Wade, M., and Albersheim, P. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4433. Wahl, G. M., Padgett, R. A., and Stark, G. R. (1979a).J . BioE. Chem. 254, 867923689. Wahl, G. M., Stem, M., andstark, G.R. (197913). Proc. Natl. Acad. Sci. U.S.A. 76,36833687. Walker, G. C., Uhlenbec, 0. C., Bedows, E., and Gumport, R. I. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 122. Walz, A., Ratzkin, B., and Carbon, J. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 6172. Warren, G., and Sherratt, D. (1978). Mol. Gen. Genet. 161, 39. Weare, N. M. (1978). Biochim. Biophys. Acta 502, 486. Weightman, A. J., Slater, J. H., and Bull, A. T. (1979). FEMS Microbiol. Lett. 6, 231. Weinard, U . , Bruschke, C., and Feix, G. (1979). Nucleic Acids Res. 68, 2707-2716. Weisblum, B., Grahm, M. Y., Gryczan, T. J., and Dubnau, D. (1979a). J . Bacteriol. 137, 6354543. Weisblum, B., Holder, S. B., and Halling, S. M. (1979b).J . Bacteriol. 138, 990-998. Weiss, B., and Milcarek, C. (1974). Methods Enzymol. 29, 180-193. Weissman, S. M. (1979). Anal. Biochem. 98, 243. Weisz, P. B., and Marshall, J. F. (1979). Science 206, 24-30. Wensink, P. C., Finnegan, D. J., Donelson, J. E., and Hogness, D. S. (1974). Cell 3,315325. West, R. W., Neve, R. L., and Rodriquez, R. L. (1979). Gene 7,271. White, R. L., and Rosbash, M. (1979). Gene 7, 97. Whitman, W. B., Martin, M. N., and Tabita, F. R. (1979).J . Biol. C h m . 254, 10,329. Wickens, M. P., Buell, G. N., and Schimke, R. T. (1978).J . Biol. Chem. 253, 2483. Wickner, R. B. (1979). Plasmid 2,303322. Widera, G . , Gautier, F., Lindenmaier, W., and Collins, J. (1978). Mol. Gen. Genet. 163, 301305. Widmer, H. J., Jaggi, R. B., Weber, R., Rflel, G. V., and Muellener, D. (1979). Eur. J . Biochem. 99, 23. Wieslander, L. (1979). Anal. Biochem. 98, 305. Wigler, M., Silverstein, S., Lee, L. S., Pellicer, A., Cheng, Y. C., a n d h e l , R. (1977). Cell 11, 223-232. Wigler, M., Sweet, R., Sim, G . K., Wold, B., Pellicer, A., Lacy, E., Maniatis, R., Silverstein, S., and Axel, R. (1979). Cell 16, 777-785. Wigler, M., Perucho, M., Kurtz, D., Dane, L., Pellicer, A., Axel, R., and Silverstein, N. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 3567. Wilkie, N. M., Clements, J. B., Boll, W., Mantei, N., Lonsdale, D., and Weissmann, C. (1979). Nucbic Acids Res. 1, 859. Williams, R. O., Young, J. R.,and Majiwa, P. A. 0. (1979). Nature (London) 282, 847.

84

VEDPAL SINGH MALIK

Williamson, V. M., Bennetzen, J , , Young, E. T., Nasmyth, K., and Hall, B. D. (1980). Nature (London) 283,214.

Willshaw, G. A,, Smith, H. R.,and Anelerson, E. S. (-1979).J . Gen. Mbobiol. 114, 15-26. Wiseman, A. (1977). “Topics in Enzyme and Fermentation Biotechnology.” Ellis H o w d Ltd., Chichester, England. Wittig, B., and Wittig, S. (1979). Biochem. Biophys. Res. Commun. 01, 554. Wittig, B., Wittig, S.,and Grunz, H. (1979). Nucleic A& Res. 6, 3759. Woo, S. L. C., Chandra, T., Means, A. R.,and O’Malley, B. (1977a). Biochemisty 16,5670. Woo, S. L. C., Monahan, J. J., and O’Malley, B. W. (1977b). J . Biol. Chem. 252,5789. Wovcha, M. G., Antosz, F. J., Knight, J. C., Kominek, L. A., and Pyke, T.R.(1978). Biochim. Biophys. Acta 531, 308. Wovcha, M. G., Brooks, K.,and Kominek, L. A. (1979). Biochim. Biophys. Acta 574,471479. Wu, R. (1978). Annu. Reo. Biochem. 47, 607. Wu, R.,and Bahl, C. P. (1978). Prog. Nucleic Acid. Res. Mol. Biol. 21, 101-141. Wu, R.,Bahl, C. P., and Narang, S. A. (1978). Currents Top. Cell. Regul. 13, 137-178. Yagisawa, M., Huang, R. T. S.,and Davies, J . E. (1978). J . Antibiot. 31, 809. Yalow, R. S. (1978). Science 200, 1236-1245. Yang, R. C. A,, and Wu, R. (1979). Science 206,456-462. Yoneda, Y., Graham, S., and Young, F. E. (1979). Gene 7,5148. Young, F. E., Duncan, C., and Wilson, G. A. (1977). I n “Recombinant Molecules” (R.F. Berrs and E. G. Bassett, eds.), p. 33. Raven, New York. Zakian, V. A., Brewer, B. J., and Fangman, W. L. (1979). Cell 17,923-834. Zamecnik, P. C., and Stephenson, M. L. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 280. Zasloff, M., Ginder, G. D., and Felsenfeld, G. (1978). Nucleic A& Rer. 5, 1139-1152. Zieg, J.. Silverman, M., Hilmm, M., and Simon, M. (1978). In “The Operon” 0. H. Miller and W. S. Reznikoff, eds.). Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Zubay, G. (1977). Annu. Reo. Genet. 7,267-287. Zurkowski, W., and Lorkiewicz, Z. (1979). Arch. Microbid. 123, 195. Zyskind, J. W., Deen, L. T., and Smith, D. W. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 30973101.

A. HURST Microbiology Research Division, Health Protection Branch, Health and Welfare Canada, Ottawa, Ontario, Canada I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Biology ............................... A. Cult , and Media.. ....................... 87 B. Genetic Control of Nisin Synthesis.. . . . . . . . . . . . . . . . . . . 89 90 C. Function of Nisin to the Producer Organism . . . . . . . . . . . D. Mode of Action 111. Chemistry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 A. Historical.. . . . . . . . . . . . . . . . . . . . . . . . . . 98 B. Units of Measurement and Metho . . . . . . 98 C. Preparation of Nisin D. Occurrence of Sever . . . . . . . . . . . . . . . . . 101 G. Enzymic Inactivation of Nisin ........................ IV. Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Use ofNisin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Introduction

E. Fully Heat-Processed Foods . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

104 105 108

119

I. Introduction Nisin has been acknowledged for about 40 years and continues to be of interest to food microbiologists. Nisin is a polypeptide that, together with acids and peroxides, is among the biological inhibitors produced by the lactic streptococci. These organisms belong to serological group N; the name was coined by Mattick and Hirsch (1947) from the letters N Inhibitory Substance. A number of reviews have been published, some emphasizing the applied aspects and others the basic microbiology and chemistry (Berridge, 1953; Hawley, 1957, 1962; Marth, 1966; Jarvis and Morisetti, 1969; Hurst, 1972, 1978; Lipinska, 1977). This article represents an attempt to produce a comprehensive account covering all these aspects, but, because of the long history and widely scattered references, a coverage of all the literature has not been achieved. Hence, some work has been wholly or partially omitted but no slight is intended by this selection. 85 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 27 Copyright @ 1981 by Academic Press, Inc. All rights of reproduction In any form reserved. ISBN 0-12-002627-9

86

A. HURST

The chemistry and biosynthesis of nisin offer interesting and d8icult challenges. A pronisin has been postulated but not isolated; a substance that could be pronisin has been isolated from non-nisin-producing streptococci, and enzyme(s)that convert this into nisin have been found but not yet purified or studied. The function of nisin for the producer cell also poses interesting questions. In the case of Bacillus peptide antibiotics, a function in the growth cycle of the organism can readily be hypothesized because of the overt differentiation that occurs among spore-formers. This process is not so obvious among gram-positive cocci, but it is possible that a more cryptic cellular differentiation does take place. One can speculate that nisin and other antibiotics, as well as peptides not recognized as antibiotics (Hurst, 1967), may play a role in this aging or differentiation process. Subtilin produced by Bacillus subtilus and nisin produced by Streptococcus lactis are extremely unusual yet very similar substances (Gross et al., 1969). It is astonishing that such different microorganisms should produce such similar substances. Nisin-like substances are commonly produced by the lactic streptococci (Oxford, 1944; Hurst, 1967). Indeed, nisin-like substances may be widely distributed within the whole genus Streptococcus. For example, an antibiotic-like substance, streptococcin A-F22, was described by Tagg and Wannamaker (1978). This substance is either an antibiotic or a bacteriocin and is formed by the clearly pathogenic member of this genus, Streptococcus p yogenes. The production of bacteriocins by gram-positive bacteria has been reviewed (Tagg et al., 1976). Among antibiotics, nisin has the unique function of being used as a food preservative. There are several reasons for this: It is not used medicinally or in animal feed, and, when eaten, it is nontoxic. Since it is a polypeptide, any residues remaining in foods are digested. In spite of these factors, the application of nisin to food preservation has been slow and is probably still in its infancy. An interesting use of nisin-producing starters is being developed by Lipinska (1977)to replace the use of nitrate in cheese making. As of December 1979, its use was not allowed in North America. In 1966, the United States Food and Drug Administration (FDA)rejected applications for the use of nisin in canned foods, probably for the good reason that some essential information about it was missing, which appears still to be the case. As pointed out elsewhere in this review, the information on cross-resistance is inadequate and perhaps other tests, e.g., tests for mutagenicity, should now be done. Also, more recently, a greater awareness has developed of the risks of feeding antibiotics to farm animals (Anonymous, 1972a). It is a mistake to lump together everything that is called an antibiotic. Nisin is evidently very different from the antibiotics used in medical and veterinary practice and it has not been used for purposes of growth promotion. Progress in the area of food

NISIN

87

preservation will be greatly hampered as long as the popular confusion remains over all substances called antibiotics being kept out of foods. It might be better to call nisin a biological food preservative and class it together with lactic or acetic acid and hydrogen peroxide, the other antibacterial products of the lactic streptococci. Nisin has advantages in food preservation that merit further technological and fundamental research. In the developed countries it can be used as an adjunct to heat-processing. Some evidence exists that spores become “injured” and more heat-sensitive when nisin is present, and spores that survive heating may become more nisin-sensitive, Provided that residual nisin remains in the food after heating, the development of these spores can be prevented. Products thus treated could attain “commercial sterility” (any viable spores that they contained could not outgrow and develop) with less heat treatment than is currently necessary. The consumer would benefit from products that, due to milder heating, would have improved nutritional value, flavor, texture, and appearance, and the savings in energy would be an important consideration to the processor. Authorities charged with regulating the use of food additives have been concerned that such substances might be used to hide sloppy manufacturing practices. However, this type of abuse is not likely with nisin because of its narrow antibacterial spectrum. The effectiveness of nisin is related to the bacterial load, and there is evidence that it is ineffective when poor quality starting materials are used (Gibbs and Hurst, 1964). Developing countries often require imported dried milk products to supplement low-protein diets. However, the water supply of such countries is often unsafe, and dried milk products reconstituted with contaminated water may do more harm than good. Under such circumstances, it would appear sensible to establish reconstituting plants, in which the product would be heat-treated before distribution. The addition of nisin to this milk powder might permit diminished heating with a concomitant gain in nutritional value. The nisin remaining in the product might prolong its shelf-life in areas where refrigeration is not available. The process just described is already being followed in some countries in the Middle East.

11. Biology A. CULTURES, STRAINS AND MEDIA Immediately after their discovery, the “inhibitory streptococci” were known to belong to the group of lactic streptococci. The organism used by Oxford (1944) appears to have been Steptococcus cremoris. Detailed taxonomic tests were first done in 1951by Hirsch and Grimsted. They used

88

A. HURST

serological and physiological tests and investigated 21 nisin-producing strains, all of which were S. Zactis. The only remarkable feature of their results is that 20 of the 21 strains were sucrose fermentors, a property of this species normally regarded as variable (Deibel and Seeley, 1974). The connection, if any, between sucrose fermentation and nisin synthesis has not been followed up. Baribo and Foster (1951)found that both S. Zactis and S . cremoris were able to produce antibiotics. Nisin is formed in milk or in the usual complex organic media in which these fastidious organisms can grow (the culture requires incubation at 2030°C without aeration). Hirsch (1951b)found that in a complex glucose meat extract medium, the usual yield of nisin was about 80 International Units/ milliliter (IUlml). Acidity was the cause for cessation of growth, so that periodic neutralization approximately doubled the yield. Glucose then became limiting; increasing the glucose from 1to 2.5% in the medium, combined with periodic neutralization of the acid, increased the yield from 160 to 460 IU/ml. Pantothenate then became limiting and with this supplement, 620 IU/ml were attained. Replacement of periodic neutralization by a three component buffer system further increased the yield to about 1700 IU/ml. The maximum buffer capacity of this medium was at about pH 5.9-6.1, which thus appeared to be the optimal for nisin synthesis. Increased nisin synthesis always went hand in hand with increased cell mass (Hirsch, 1951b). That maximal nisin synthesis was related to maximal biomass formation was confirmed by Egorov et al. (1971). They optimized the concentrations of various medium ingredients such as corn extract, yeast-autolysate, caseinhydrolysate, glucose, phosphates, and mineral salts and obtained maximum yields of about 2000 IU/ml. Later they reported that thymine, adenine, and hypoxanthine stimulated nisin synthesis (Egorov et al., 1976). However, their own results show no greater yield of the product than that reported by them in 1971. Hirsch (1951b) found no effect that could be ascribed to purines and pyrimidine bases. A minimal synthetic medium was reported by Kozak and Dobrzanski (1977). The minimal requirements for growth were also the minimal requirements for nisin synthesis. About three-four times less nisin was produced in this than in a complex medium. The medium consisted of nine amino acids, four B group vitamins, glucose, and mineral salts. They confirmed that purines and pyrimidines had no effect on nisin synthesis. However, adding four extra amino acids to the minimal medium (serine, proline, cysteine, and cystine) increased nisin synthesis to levels comparable to those found in complex media. Unfortunately, Kozak and Dobrzanski (1977) express their results in ratios of either IU/OD or IU/colony-forming units (CFU) so that a direct comparison of their published results with those of others is impossible.

89

NISIN

Selection of strains for nisin production was mentioned by Mattick and Hirsch (1947). They observed a large strain-to-strain variation and a decline in the inhibitory activity in milk as the culture became “used to” broth. In broth, nisin production appeared to be stable for at least 2 years. Cheeseman and Berridge (1957) describe the selection of a strain from a culture freeze-dried 10 years previously. The yield of nisin from the culture declined with repeated subculturing and it appeared that the strain was becoming nisin sensitive. Cheeseman and Berridge (1957) selected tolerant forms by using a 10%inoculum into 500 IU of added nisirdml of medium and after overnight incubation, plated on agar medium containing 2000 IU/ml. A colony was picked that produced 700 IU/ml, whereas the original culture formed only 250 IU/ml. The variable effect of the number of serial transfers and other early strain selection experiments is also discussed by Hirsch

(1951b).

Lipinska (1977) summarizes experiments concerning the use of industrial wastes for growing the nisin producing streptococci. These wastes include dairy and pancreatin-hydrolyzed wastes from insulin production. The actual yields attained are not mentioned. It appears that with the present strains, 2000 IU/ml is about the maximum that can be attained. The stability of the nisin preparations of English, Polish, and Soviet manufacturers was compared by Goudkov et al. (1973)who found that the English preparation was somewhat more stable than the other two. B. GENETICCONTROLOF NISIN SYNTHESIS Mutants with increased nisin yield following irradiation treatment were reported by Lipinska (1977). Kalra et aZ. (1973) reported the isolation of mutants that formed approximately double the amount of nisin of the parent organism. They used UV irradiation and colonies were isolated when there was 1 4 . 1 %survival. They quote earlier work by others in which similar treatments resulted in 10-fold increases in nisin synthesis. It would seem that more experiments of this kind are needed. Plasmid DNA is known in S. Zactis and 4-6 different sizes have been reported in the strains examined (Larsen and McKay, 1978). Kozak et al. (1973) reported preliminary experiments suggesting that nisin synthesis might be plasmid controlled. They examined 12 untreated strains of S . Zactis and picked 8081 colonies, of which 3 were Nis- (0.004%).After treatment with proflavin, 11,337 colonies were isolated and examined, of which 200 were Nis- (2%);after ethidium bromide treatment, 6392 colonies were isolated and 29 were Nis- (0.5%).This increase in Nis- cultures was mainly owing to one strain. Because proflavine and ethidium bromide treatments are classically believed to “cure” bacteria from plasmids, there was some evidence that nisin synthesis may be plasmid controlled. ’

90

A. HURST

One strain that readily responded to curing also contained more spontaneously formed Nis- clones (35 of 4248 examined). This property was stable: Passaging 120 cultures 10 times through broth did not result in any Nis+ cultures. Cultures cured of a plasmid-controlled property should not be able to regain this property by means of a mutation. Kozak et aZ. (1974) examined two cured cultures after treatment with nitrosoguanine. They examined 30,730 colonies from one of these cultures and 20,115 colonies from the other, but no Nis+ colonies were found. This finding supports the notion that nisin is plasmid regulated. Further experiments were reported in which attempts were made to characterize the plasmid DNA by centrifuging in CsCl gradients (Fuchs et al., 1975). However, no conclusive results were reported: Among eight nisin-producing S. Zactis, two were found not to contain covalently closed circular (ccc) DNA. Two further strains that had been cured of nisin were tested. Compared to the parent strain, only one had lost ccc DNA while the other one had not. OF NISIN TO THE PRODUCER ORGANISM C. FUNCTION

1 . As a Secondary Metabolite

It is possible that nisin has no function for the organism producing it. In batch cultures, it is formed when more than half of the cell mass has been synthesized (Hurst, 1966b). At that time, it is possible that the cell reacts to the exhaustion of a limiting but essential nutrient. Woodruff (1966) suggested that instead of “switching-off’altogether, cells turn to the synthesis of antibiotics as an alternative, which represents a means of using fully formed enzyme systems and available substrates. Hence, antibiotic synthesis may be regarded as an alternative to breakdown of cellular control mechanisms and a means of maintaining a fully functional machinery while awaiting a new source of the limiting metabolite. However, this view seems inappropriate to nisin synthesis, which is an energy-requiring process, and it is energy in particular that may be limiting toward the end of the growth phase. In addition, the enzyme(s) converting pronisin to nisin is not present all the time; on the contrary, it appears to be made only for a brief period toward the end of the growth phase (Hurst and Paterson, 1971). 2. As a Dominance Factor A second view of the function of nisin is that it ensures dominance when its producer organism is in competition with other lactic streptococci. This argument concerns the contest between S . Zactis and S. cremorkr. The natural habitat of these organisms remains uncertain (Sandine et al., 1972),

91

NISIN

F t it is known that they are the organisms that naturally sour raw milk. The argument put forward by Hirsch (1952)can be summarized briefly as follows: "Spilt milk or "milk in a container" is a relatively recent substrate dating back only to the domestication of milk-producing animals. Two species converging on this new substrate are struggling for dominance; hence, they produce antibiotics aimed at each other. Thus, S . lactis produces nisin and S. cremoris is the organism most sensitive to it. Similarly, S . cremoris produces diplococcin and the organism most sensitive to it is S. lactis (Hirsch and Grinsted, 1951).Also in favor of this argument is the fact that lactose fermentation-a basic need for growth in a milk substrate-is plasmid rather than chromosomally controlled (Larsen and McKay, 1978). In industrial practice, strains of lactic acid bacteria are frequently used in mixtures. These strains can be distinguished by phage-typing. After a few transfers, one of the strains becomes dominant and, since the work of Hoyle and Nichols (1948), antibiotic production has been accepted as one of the important causes of strain dominance. Although other causes for dominance have been suggested, the question has remained with us (Reddy et al., 1971). 3. As a Cellular Growth Regulator A third way of looking at the function of nisin is that it has a regulatory action in the growth cycle of the producer organism. Biosynthesis of nisin in batch cultures starts after about 50% of the mass of the organism has been formed (Fig. 1)(Hurst, 1966b). If nisin is added to the medium before the start of its synthesis, it inhibits and lyses the producer organism (Hurst and 200

150

100 50

-

z3 2

0

Hours

FIG. 1. Biosynthetic activities of Streptococcus Zactb growing in batch culture. Dry weight (C--.-); DNA (0-); nisin (e-. .-), protein (0---), RNA, (El--). After Hurst, 1966b.

92

A. HURST

I. I.

0. 0. $ 0 0 0 lD

G

0.

>-

t 0, v) Z

w

n-1 o a

$ 0 0 (

MINUTES

FIG. 2. A nisin-producing Streptococcus Zactis grown overnight in a complex organic medium was reinoculated into the same medium. Tube 1 had no added nisin. Tubes 2-5 had nisin added (- 500 IU/ml). Tube 2 nisin added before inoculation, tube 3 after incubation for 30 min, tube 4 after incubation for 105 min, and tube 5 after beginning of nisin synthesis at 120 min. After Hurst and Kruse, 1972.

Kruse, 1972). If nisin is added to the medium before inoculation, growth is slightly delayed but then the organism grows at the same rate as the control and attains the same mass (Fig. 2). These results suggest that the growth of the producer organism commences with the inactivation of nisin. To test this possibility, a culture controlled to pH 6.8 was grown to stationary phase (Hurst, 1968). At this pH, nisin remained cell-associated and its concentration increased from 0 (at the time of inoculation) to 1.3%of the dry weight (at stationary phase). At this point, the cells were used as a 1%inoculum to start a new fermentation. The cells were reisolated and nisin was then only 0.13% of the dry weight of these cells (Fig. 3). A nisin-destroying enzyme(s) could not be demonstrated and it was suggested that uptake of calcium was linked to the disappearance of nisin

93

NISIN

12.0

3g

-

I: 10.0 e 3 9.0 I .-CI :. 8.0 11.0

.Q

00

1.0

1

I

I

I

,

Min

FIG. 3. Apparent destruction of nisin by the producer organism. Streptococcus lactis was grown for the time shown in a hatch fermentor at pH 6.8. At time 0 the culture was reinoculated into fresh medium of the same composition. After Hurst, 1968.

(Hurst and Lazarus, 1968). Because uptake of calcium could be expected to alter the optical properties of cell suspensions, this effect was studied (Hurst, 1969). On subculture, the optical density of cell suspension declined. The reaction required salts (e.g., calcium) and glucose, and was temperature dependent. It could be poisoned by p-mercuribenzoate and iodoacetate (Hurst and Kruse, 1970). Electron microscopic observations of stationary phase cells suspended in ammonium molybdate showed clear structures of wall, membranes, adhesions of wall to membrane, septae, and mesosomes. However, 4 minutes after subculture, ammonium molybdate no longer revealed any structures, the interior of the cells being stained solidly black. Presumably, in this short interval, the cells had become permeable to the stain (Hurst and Stubbs, 1969). The events that occur at subculture can be summarized as follows: An optical reaction occurs, indicating a change in cell size. These cells become permeable to ammonium molybdate, incorporate 45Ca2+,and lose bioassayable nisin. Omission of glucose or addition of iodoacetate stops the optical reaction and the 45Ca2+uptake, and nisin is not inactivated. This effect suggests that nisin inactivation results from configurational change of the cell (Hurst and Kruse, 1970).

94

A. HURST

A small survey of “wild’ strains of S. lactis showed that, although most of them produced nisin (35 of 40), some did not (5 of 40). Hence, it was dficult to see how nisin could have a regulatory function if some strains did not produce it (Hurst, 1967). However, these strains produced a basic peptide that had no inhibitory properties, yet was so similar to nisin that it could not be separated from nisin by Sephadex chromatography and behaved as does nisinl in polyacrylamide gel electrophoresis’. It was further found that, in batch cultures, nisin accumulated as growth came to an end. The amount of cell-associated nisin was also strongly correlated with the length of the lag-phase (Hurst and Dring, 1968). As a result, there is speculative evidence that nisin and other basic peptides (and they are largely unknown because they are not antibiotics) may have a function related to the beginning and the ending of growth in batch cultures. A regulatory function concerned with the sporulation of the producer organism has been demonstrated for gramicidin (Mukherjee and Paulus, 1977).

D. MODE OF ACTION 1 . Antibacterial Spectrum Nisin has a narrow spectrum, and it does not inhibit gram-negative organisms, yeasts, or fungi. Neisseria are an exception to this general statement, the three strains reported by Mattick and Hirsch (1947) being quite sensitive. Gram-positive organisms are sensitive to nisin. The activity of nisin is not affected by the presence of milk or serum. Streptococci and the closely allied S. cremoris are especially sensitive and in this respect, nisin resembles a bacteriocin rather than an antibiotic (Tagg and Wannamaker, 1978). Staphylococci are also sensitive (Gowans et al., 1952). Mattick and Hirsch (1947) tested six strains of Mycobacterium tuberculosis and listed them as sensitive organisms. Nevertheless, these strains were about 100 times more resistant than streptococci.

2 . Effect of Nisin on S p o r e f m s The early literature on the effect of nisin on sporeformers has been reviewed by Campbell and O’Brien (1955) and later by Boone (1966). In 1947, Mattick and Hirsch reported that bacilli and clostridia were sensitive to nisin. Andersen et al. (1953) reported that nisin was effective against Clostridium botulfnum. Lewis et al. (1954) tested the complementary action of heat and antibiotics. Nisin was among 67 antibiotics they tested ‘The basic peptides from one of these strains were used to generate an antibiotic that resembled nisin; this is described in Section IV.

NISIN

95

against Clostridium sporogenes (PA 3679); it caused a 55% decrease in the D value (decimal reduction time in the count due to the applied heat). O’Brien et al. (1956) obtained similar results using only subtilin and nisin but a wide range of microorganisms: C . botulinum (PA 3679), C . thermosaccharolyticum, Bacillus coagulans, and B . stearothermophilus. Campbell et al. (1960)used various foods and processing conditions to extend these findings. They confirmed the preservative effectiveness of nisin combined with mild heat. These workers suggested that the reduction in the D value was due to carry-over of nisin into the recovery medium. Campbell and O’Brien (1955) showed that about 0.03 ppm in carry-over was sufficient to inhibit the development of severely heat-treated spores of clostridia or bacilli. Campbell and Sniff (1959a) using B . coagulans showed that the heated spores were not killed by nisin because the counts could be restored to that of the control when trypsin was incorporated into the plating medium. Thorpe (1960)obtained similar results with B . stearothermophilus and suggested that the reduction in heat resistance was due to the adsorption of nisin on the spores. Tramer (1964) also found that trypsin reversed the heat effect of nisin, but he found no evidence for adsorption to spores. Nisin could be diluted out on plating. On the other hand, Hawley (1962) reported that nisin cannot be washed off C. botulinum spores, nor is it clear why trypsin, which does not inactivate nisin (see Section 111, G), should reverse its action. The evidence presented by Hawley (1962) strongly suggests that heated C. botulinum types A and B became injured. At that time, spore injury was not known (for review of injury see Adams, 1978) and Hawley was unable to express his results in modem terms. Nevertheless, he noted that spores of nisin-resistant organisms “will germinate and grow profusely in the presence of high concentrations of nisin if they have not previously been exposed to a sub-lethal heat treatment” (Hawley, 1962). Direct evidence for heated spores becoming increasingly nisin sensitive was provided by Heinemann et al. (1965).They used heat-damaged spores of four species and in every case observed a dramatic increase in nisin sensitivity as the severity of the heating was increased. The results of Heinemann et al. (1965) leave no doubt that heat injury in bacilli and clostridia is expressed by increased sensitivity to nisin. Hirsch and Grinsted (1954) reported that nisin was sporicidal for Clostridium butyricum but not for C. sporogenes and C . bij-ermentans. Ramseier (1960)confirmed the sporicidal effect of nisin on C . butyricum. He suggested that the spores remained unaffected, but once germinated, they were killed by nisin. Campbell and Sniff (1959a) thought that nisin was sporostatic and that it prevented the outgrowth of spores of B. coagulans. Hitchins et al. (1963) and Gould (1964), using spores of bacilli, reported phase darkening and loss of heat resistance in the presence of nisin but they

96

A. HURST

also found prevention of further outgrowth steps. Whether spores of bacilli and clostridia behave in the same way remains to be established.

3. Factors Affecting Effectiveness Ramseier (1960) emphasized the numerous factors that affect nisin activity. The important ones were size and age of inoculum, composition of medium, and p H of the solution in which nisin was dissolved prior to its addition to the broth. The best solvent tested was 0.02 N HCI. At the other extreme, in buffer at pH 10, 80%of the nisin was inactivated within 2 hours. Campbell and Sniff (1959b)used tomato juice adjusted to different pH values and confirmed the increased effectiveness of nisin at low pH. At p H 5.3, all 31 strains of B . coagulans were inhibited by 200 IU of nisin but none were inhibited at p H 7.2 by 560 IU.

4 . Molecular Basis for Nisin Action Although Tramer (1964) reported that nisin is not adsorbed to spores, the weight of the evidence supports the contrary view. Adsorption to the microbe appears to be the first step required for nisin to act (Hirsch, 1954; Ramseier, 1960; Thorpe, 1960). Hirsch (1954), working with Streptococcus agalactiae, reported that 150 IU/ml killed the organism in 10-15 min. However, when culturing was done in broth containing charcoal, 20-30 min were required. This difference is most easily explained by the assumption that a nisin-bacterium complex that can be disrupted by charcoal is formed first; after 2 0 3 0 min, either the reaction becomes irreversible or the cells have been killed. Ramseier (1960) also reported that nisin was strongly adsorbed by sensitive vegetative cells of C . butyricum. Cells that had been made resistant did not adsorb nisin. He also showed that the adsorption by sensitive cells was highly pH-dependent, peaking at pH 6.5. Taking adsorption at p H 6.5 as 100%,69%was adsorbed at pH 5.5 and only 43% at p H 4.5. Ramseier (1960) further reported that treatment of sensitive cells with nisin caused the release of cytoplasmic material and lysis. Similar results were obtained by Hurst and Kruse (1972) when they added nisin to the producer organism in its nisin-sensitive growth phase (see Fig. 2). Ramseier (1960)also reported that anionic soaps neutralized nisin and concluded that it acted as a cationic surface-active detergent. Sack (1952)had earlier reported that the closely allied antibiotic subtilin acted similarly. Nisin may also have other effects. Gross and Morel1 (1967)used it successfully as an antimalarial in mice. They reasoned that the unsaturated amino acids in the molecule would compete for sulfhydryl-containing enzymes, such as coenzyme A. Malarial parasites are known to be sensitive to deficiency in coenzyme A.

97

NISIN

In cell-free systems, nisin-inhibited peptidoglycan synthesis catalyzed by enzymes from B . stearothermophilus or Escherichia coli (Linnett and Strominger, 1973). These authors used 40 puglml (about 1600 IU/ml) to cause a 50%inhibition of peptidogylcan synthesis. This concentration is about loo0 times more than the minimal inhibitory concentration (MIC) for B . stearothemnophilus. It therefore seems unlikely that inhibition of peptidoglycan synthesis could be a primary site of nisin action.

5 . Resistance to Nisin Nisin was bactericidal against S . agalactiae, a group B streptococcus (Hirsch and Mattick, 1949). More than 99% of the inoculum was killed first but the surviving organisms could outgrow in the presence of nisin. These cultures became nisin resistant (Hirsch, 1950). The relationship between the developed resistance and the amount of nisin in the culture medium is shown in Fig. 4. It can be seen that after only one subculture in medium containing 100 IU/ml, the resulting culture displayed a 40-fold increase in resistance. Carlson and Bauer (1957) trained S. pyogenes, B . subtilis, and StaphyZococcus aureus to nisin resistance. The MIC of the latter organism

$3 [ ,

0

, I0

20

,

,

,

,

,

,

,

,

30 40 50 60 70 80 ORGANISM GROWN IN NlSlN IU/rnl

90

100

FIG.4. Streptococcus agalactiae was grown in broth containing the nisin concentrations shown. From each, three colonies were isolated and their nisin resistance tested; their averaged response is shown. Based on the data of Table 2, Hirsch, 1950.

98

A. HURST

increased from 2.5 to 2000 IU/ml after 17 “passages.” The nisin-resistant staphylococcus did not produce a soluble nisinase, but water extracts of acetone-dried cells had a nisin-inactivating property. After 7 “passages” in nisin-free broth, the resistant staphylococcus reverted to its original MIC of 2.5 IU/ml. Ramseier (1960) also used the “training” procedure to obtain resistant C. butyricum. The resistance of the organism increased from 128 to 4096 W m l in the course of 12 passages. Culturing in nisin-free broth resulted in a decrease in resistance from 4096 to 1024 IU/ml in 17 passages. Subculturing for another 14 passages did not result in further increase in sensitivity. The staphylococcus trained by Carlson and Bauer (1957) to resist nisin to 2OOO IU/ml had unchanged sensitivity to penicillin, streptomycin, terramycin, and tetracycline. Carlson and Bauer also tested 14 strains of staphylococci isolated from clinical sources. All were penicillin resistant and some were resistant to as many as four additional antibiotics. Of the 14 penicillin-resistant strains, eight were also nisin resistant. The enzyme nisinase is formed by some lactic acid bacteria and bacilli; this is further discussed in Section 111, G.

111. Chemistry A. HISTORICAL By 1933, Whitehead was aware that nisin was a peptide. The first purified material was obtained from acidified culture fluids by Mattick and Hirsch (1947). It was concentrated into a chloroform gel by the method of Sevag et al. (1938), dissolved in acid alcohol, and finally precipitated at neutral pH. The resulting preparation probably contained only 2 4 % nisin, but Berridge (1947) used it as a starting material to prepare crystalline nisin. The activity of the crystalline material suggested that it was almost pure nisin (Berridge, 1949). Falconer (1952) caused nisin to remain cell associated by maintaining the pH of the fermentation at about 6. After the cells had been collected, nisin was extracted with solvents. Nisin could also be concentrated from culture fluids by frothing, which achieved about a 10-fold increase in concentration of the antibiotic activity (Hirsch, 1951b).

B. UNITS

OF

MEASUREMENTAND METHODS OF ASSAY

1 . Dejlnition of Units Several units have been defined in the past, but more recently, Tramer and Fowler (1964) defined a Reading Unit (RU) in terms of the activity

NISIN

99

exhibited by 1pg of a standard batch of commercial nisin.2 The term Reading Unit was chosen because much of the original work on nisin was done at the University of Reading in England. This standard was accepted by the World Health Organization (WHO) (1969), so that Reading Units are now referred to as International Units (IU). The approximate activity of 1 pg of pure nisin is 40 IU.

2. Assay As early as 1934, Cox recommended the use of methylene blue reduction for the detection of “inhibitory streptococci” in milk. In this test, starter organisms in milk rapidly reduce methylene blue to the colorless compound. If the milk contains nisin-producing streptococci, the time required for reduction is increased. The detection in raw milk of inhibitory streptococci, most being apparent nisin producers, was also described by Chevalier et aZ. (1957).A rich agar medium capable of growing the thermophilic LactobaciZZus Zactis was inoculated simultaneously with the milk under test. Plates were first incubated for 1 day at 20°C, when colonies developed of the nisin-producing organism but not the thermophilic lactobacillus. This procedure was followed by incubation at 30°C. Nisin-producing colonies were then surrounded by clear zones of inhibition. Hirsch (1951b) described a quantitative version of the methylene blue assay and also a more complex assay, the lag-phase method. Streptococcus agaZactiae was the test organism and within the range of 5-50 IU/ml, the length of the lag was linearly related to the concentration of nisin. Growth was estimated by the lowering of pH. A turbidimetric assay was developed by Berridge and Barrett (1952), and was also used in a modified form by Hurst (1966a). This assay method can be used with concentrations of 0.1-1 pg/ml ( 4 4 0 IU/ml). The test organism was S. cremoris, which was grown for 2%-3 hours in broth containing different concentrations of nisin, growth then being stopped with an organic mercurial disinfectant. A standard was then plotted relating OD at 600 nm to nisin concentration. The test required clear solutions but, because of its speed, sterile precautions were unnecessary. A tube method using‘vertical diffusion was suggested by Beach (1952). None of the methods mentioned so far used the conventional horizontal agar diffusion techniques: Nisin usually occurs in a polymeric form (MW a 7000) and therefore, does not readily diffuse through agar. Mocquot and Lefebre (1956) attempted to overcome this limitation by incorporating Tween 80 in the agar medium. The plates were refrigerated overnight to permit nisin to ‘Standardized nisin preparations can be obtained from the manufacturers Aplin and Barrett Ltd., Technical Service Dept., 11 North Street, Beaminster, Dorset DT8 3DZ, England.

100

A. HURST

diffuse into the agar, and then incubated to enable growth of the test organism and to develop the zones of inhibition. Tramer and Fowler (1964) used 1% Tween 20, which eliminates the need for prediffusion and hence permitted a simple diffusion assay of nisin.'This assay was especially useful for the estimation of nisin in foods, because such samples may be cloudy or contain particulate matter. The test organism for this assay is Micrococcus flauus, which grows profusely after overnight incubation at 30°C. The size of the zones of inhibition is linearly related to nisin concentration, in the range of 0.5-10 IU/ml. Many foods contain substances that interfere with the assay either by binding the nisin or by influencing the zones of inhibition so that direct reading off a standard curve can introduce serious error. However, Tramer and Fowler (1964) described the preparation of suitable controls to overcome these problems. Assay techniques not suitable for routine use were described for special applications by a number of authors. For example, Lewis et al. (1954) wanted to test the combined effect of moist heat with antibiotics. They aimed to screen antibiotics that might be useful for canning. They found that subtilin and nisin were the only antibiotics that accelerated death due to heat. Stumbo et al. (1964) developed a micromethod for testing traces of residual nisin that may be present in canned foods after heat processing or after long storage. Their estimation was based on counting colonies that developed and they could estimate as little as 0.005 IU/ml (0.000125 ppm on the standard curve). Ingram (1969) working on the biosynthesis of nisin used incorporation of radioactive amino acids to assay nisin and elucidate the mechanism of its cellular synthesis.

C. PREPARATION OF NISIN Cheeseman and Berridge (1957)described detailed schemes for obtaining nisin of high purity and in good yield. The culture was grown in a highly buffered complex medium (Hirsch, 1951b), and both the cells and the culture fluid were used. Briefly, the method consisted of extraction of nisin with n-propanol containing NaCl followed by a number of fractional precipitations. The concentrated material was then further purified by preparative countercurrent. Similar methods appear to have been used by Gross and Morel1 (1967) for their structural studies of nisin. Bailey and Hurst (1971)used cells of the producer organism grown at pH 6.7 so that they retained the synthesized nisin. The cells were mechanically broken and then acid-extracted in the cold. The extract was first chromatographed on a resin column, acetone-precipitated, and finally purified by pH gradient elution from CM cellulose columns. Wilimowska-Pelc et al. (1976) described the purification of nisin from commercial products. In these products, the biological activity of nisin is

101

NISIN

standardized by dilution with either skim milk powder or lactose and salt. Nisin was fractionated on columns of BioGel P-10 followed by ion-exchange chromatography on CM-Sephadex G-25. The activity of this material was about 40 times higher than that of the starting material, despite which it gave three bands on electrophoresis. The authors suggest that pure nisin contains several biologically active peptides as originally suggested by Hirsch (1951a) and Berridge et al. (1952). D. OCCURRENCE OF SEVERAL NISINPEPTIDES During the early 1950s, several lines of biological evidence pointed to the notion that nisin was not a unique peptide but consisted of a mixture of peptides with differing biological activities. Hirsch (1951a) and Hirsch and Grinsted (1951) used different strains of antibiotic-producing streptococci both for the production of this antibiotic and as test organisms. Thus, the sensitivity of the different producers to different antibiotics was determined and it was suggested that four different antibiotic types existed. Serum or milk inactivated these substances to different extents, providing additional evidence for the distinctness of these antibiotics. Although produced by different strains, the antibiotics seemed to be slightly modified forms of nisin. Their composition was not known and no new names were proposed for them. Bavin et al. (1952) worked with nisin made from a single strain and, by countercurrent distribution, found two active peaks. Similarly, Berridge et al. (1952) resolved a single batch of nisin into four active polypeptides that they called nisin A, B, C , and D. Nisin C and D had only about one fifth the biological activity of A and B. The authors also pointed out that nisin contained the unusual sulfur amino acid lanthionine, and that in this and other respects, it resembled the antibiotic subtilin produced by a strain of B . subtilis (Dimick et al., 1947). Jarvis and Farr (1971)described five nisins: A, B, C , D, and E. Nisins A, B, C, and E were sensitive to nisinase from Bacillus cereus but nisin D was not. Therefore, it is clear that the name “nisin” describes a family of closely related peptide antibiotics. The more recent work by Gross and co-workers on structure and molecular weight was carried out with nisin A (Jarvis et al., 1968).

E.

COMPOSITION, STRUCTURE, AND

MOLECULAR

WEIGHT A product of low molecular weight was slowly released from nisin by mild acid hydrolysis or after treatment with cyanogen bromide. This product could be isolated by gel chromatography on columns of Sephadex G-25

102

A. HURST

COOH

FIG.5. The structure of nisin. ABA = aminobutyric acid, DHA = dehydroalanine, DHB = dehydrobutyrine 0-methyldehydroalanine); Ala-S-Ala = Ianthionine, and Ala-S-ABA = /3-methyIlanthionine. After Gross, 1977.

(Gross and Morell, 1967). The only amino acid that this fraction contained was lysine. Hydrolysate of the dinitrophenylated product and optical densities at 570 and 440 nm indicated that N" lysine was substituted with pyruvate so that the original fragment was probably dehydroalanyllysine. Further tests confirmed this hypothesis (Gross and Morell, 1967). The fragment was assigned to the carboxyl terminal of nisin. Later experiments by Gross and Morell (1971) assigned isoleucine to the amino terminal of the molecule. These were startling findings, since previously neither carboxyl nor amino end groups could be found (Cheeseman and Berridge, 1957). Several unsaturated a,P amino acids were established in the nisin molecule (Gross and Morell, 1968), and a novel cyclic structure of the sulfide bridges of nisin was worked out (Gross and Morell, 1970). This work led to the complete elucidation of the structure of nisin as shown in Fig. 5 (Gross and Morell, 1971). Nisin contains the a,P-unsaturated amino acids dehydroalanine (residues 5 and 33) and dehydrobutyrine (residue 2). It contains five internal ring structures of disulfide bridges. One of these is ala-S-ala (residues 3-7), which is lanthionine, and the other four are P-methyllanthionine linking residues 8-11, 13-19, 23-26, and 25-28. Hence, each molecule of nisin contains one residue of lanthionine and four residues of P-methyllanthionine, which accounts for the high sulfur content of nisin. The probable biosynthetic

FIG. 6. The structure of an a,p-unsaturated amino acid. (When R = H then it is dehydroalanine; when R = CHS then it is dehydrobutyrine.) After Gross, 1977.

103

NISIN

kH

I

kH

I FIG.7. Crosslinkages resulting from the P-addition of nucleophiles to a$-unsaturated amino acids. X may be S and can lead to either lanthionine or P-methyllanthionineformation depending on whether R = H or CHI. Alternatively, X may be NH and can lead to peptide bond formation. Slightly modified from Gross, 1977.

mechanism whereby these unusual amino acids are formed is discussed in Section IV. The a-carbon atom of the first amino acid in lanthionine and p- methyllanthionine is always in the D configuration. However, the configuration of the @-carbonatoms of P-methyllanthionine was not worked out till somewhat later and was found to be in the L configuration (Morell and Gross, 1973). These double amino acids thus occur in the meso configuration, containing one alanine in the D and the other in the L configuration. X-Ray structure analysis and measurements of torsion angles confirmed this result (Knox and Keck, 1973). In later work, the same authors also report the C-S bond distance (1.83 di) and the C-S-C angle (103") (Knox and Keck, 1975). a$-Unsaturated amino acids are reactive and Fig. 6 shows the formula of either dehydroalanine or dehydrobutyrine. Either readily forms a crosslinkage with S, to form lanthionine or methyllanthionine, or with NH, to form a peptide bond (Fig. 7). The peptide bond formation leads to the possibility of polymer formation, i.e., multiple forms of the monomeric form. This reactivity of the a,P-unsaturated acids of nisin is the probable explanation for the controversy that has surrounded the molecular weight of the antibiotic. Cheeseman and Berridge (1957) reported a molecular weight of 7000 for nisin A. They determined the amino acid composition and the diffusion coefficient from electrophoresis in a Tiselius apparatus. They did not detect amino or carboxy end groups. Confirmation of this value was obtained by Ingram et al. (1967), who found two molecular species by polyacrylamide-gel electrophoresis corresponding to molecular weights of 7000 and 14,000. However, Gross and Morell (1967), who used partial dinitrophenylation and countercurrent distribution, obtained a molecular weight of 3500. Jarvis et al. (1968)confirmed the molecular weight results of Gross and Morell (1967).The dimers and tetramers of nisin are stable: Jarvis et al. (1968)obtained a mean molecular weight of 6800 after treatment with 4 M guanidine hydrochloride, 8 M urea, or 80% ethanol in 0.02 M HCI.

F. PROPERTIES OF NISIN The structure and molecular weight of nisin were discussed in the previous section. Nisin contains no aromatic amino acids so that it has no absor-

104

A. HURST

bance at 260 or 280 nm (Bailey and Hurst, 1971). The hydrochloride contains 5.6% of its weight as sulfur (Hall, 1966). The solubility of nisin depends on the pH of the solution. At pH 2.5, its solubility is 12%,decreasing to 4%at pH 5.0, and being practically insoluble at neutral and alkaline pH values. Nisin is a basic polypeptide that migrates to the cathode on electrophoresis. The solubility properties and electrophoretic behavior suggest that the isoelectric point of nisin might be in the alkaline range. The stability of nisin is related to its solubility. In dilute HCI solutions at pH 2.5 or less, solutions can be boiled without loss of activity, and even autoclaving does not cause “serious loss of activity” (Hall, 1966). Above pH 7, irreversible inactivation occurs even at room temperature. Hall (1966) states that nisin has a higher antibiotic activity when dissolved in 50 M HCI than when assayed from solutions in distilled water. It may be that the tendency to polymerize is increased by increasing pH, but the relationship between biological activity and polymeric forms of nisin has not been established. Tramer (1964) reported that nisin was stable to autoclaving to 115.6”Cat pH 2, but that 40%of the activity was lost at pH 5 and more than 90% was lost at pH 6.8. Large molecules such as milk or broth had a protective effect so that the degree of inactivation may be less drastic in foods than in buffer. G. ENZYMIC INACTIVATION OF NISIN Nisin is not inactivated by all the proteolytic digestive enzymes; crystalline enzymes have to be used for these tests to ensure their specificity. Trypsin, elastase, carboxypeptidase A, pepsin, and erepsin were without effect (Jarvisand Mahoney, 1969). Both pancreatin and a-chymotrypsin inactivated nisin, and Jarvis and Mahoney (1969) presented evidence to show that the nisin-inactivating component of pancreatin was a-chymotrypsin. According to Gross (1975), a-chymotrypsin breaks the peptide bond between residues 31 and 32 (His-Val, see Fig. 5). The occurrence of a nisinase from Lactobacillus plantarum was reported by Kooy (1952) and was also reported to occur in other lactic acid bacteria (Galesloot, 1956; Lipinska, 1977). The nisinase from Streptococcus thermophilus was partially purified and its specificity studied (Ahfax and Chevalier, 1962):It inactivated nisin but not penicillin, aureomycin, bacitracin, polymyxin, or subtilin. Nisinase produced by members of the aerobic sporeforming genus Bacillus is also interesting. Gould and Hurst (1962) and Gould (1964) classified spores into two categories: M spores, which appeared to open their sporecoats by mechanical rupture (e.g., B . subtilis), and the L spores, which

105

NISIN

TABLE I INHIBITIONOF Bacillus SPOREDEVELOPMENT BY NISIN"

M spores

MICb

L spores

MICb

Subtilis (12 strains) Coagulans Lichenijimis Atterirnus

5 5 10 2

Cereus (4 strains) Breois Polymira

>100

>100

>loo

"After Gould and Hurst (1962). MIC: minimal inhibitory concentration (in IU/ml) to prevent the development of a spore inoculum on agar medium.

appeared to open their sporecoats by lysis (e.g., B. cereus). Table I shows that the M spores are appreciably more nisin sensitive than L spores. Jarvis (1967) confirmed these results. However, it was originally thought that the spore lytic enzyme might also be a nisinase: Jarvis (1967) showed that the antinisin enzyme was distinct from the enzyme(s) involved in spore germination and that there was no quantitative relation between the production of the antinisin activity and the resistance of vegetative forms. Unlike the nisinase from S. themophilus, the B . cereus nisinase of Jarvis (1967) inactivated subtilin but had no effect on polymyxin, gramicidin or bacitracin. Further studies by Jarvis showed that the nisinase from several species of Bacillus reduced the C-terminal dehydroalanyllysine of nisin to alanyllysine. The enzyme was active on nisins A, B, C, and E, but not on nisin D. It was suggested that nisinase was a dehydropeptide reductase (Jarvis, 1970; Jarvis and Farr, 1971). Because nisinase appears to be quite specific, it may be useful for identifying the antibiotic in foods (Jarvis and Morisetti, 1969). Carlson and Bauer (1957) reported that penicillinase also inactivated nisin when used at about a 1000-fold excess.

IV. Biosynthesis Nisin is synthesized when cells of S. Zactis (strain NCDO 4973) are incubated either in a complex growing medium or in a reaction mixture (Hurst, 1966a). This reaction mixture contained amino acids, salts, glucose, growth factors, and buffer. Biochemical activity of the cells was followed by the incorporation of radioactive tracers and nisin by bioassay. Penicillin and mitomycin, antibiotics to which the organism was sensitive, had no effect on nisin synthesis. Actinomycin D inhibited RNA synthesis immediately and nisin synthesis after a delay of 60 min. Synthesis was most sensitive to 3National Collection of Dairy Organisms, Shinfield, Reading, England.

106

A. HURST

inhibitors of protein synthesis, e.g., chloramphenicol, puromycin, and tetracycline. The immediate inhibition of nisin synthesis by chloramphenicol is shown in Fig. 8. Nisin synthesis was more sensitive than protein synthesis (Fig. 9). Ingram (1969) used the reaction mixture of Hurst (1966a) to confirm these findings. Instead of bioassay, he used incorporation of radioactive amino acids into material that behaved as does nisin on purification. This material was oxidized with performic acid, which converted lanthionine to lanthionine sulfone. It was not resolved by high voltage electrophoresis from p-methylanthionine sulfone, though it was clearly separated from cysteic acid. When a culture of S. lactis was incubated with [3H]threonine and [ 35 Slcysteine, radioactivity was incorporated into the lanthioninelp-methyl lanthionine spot. The 3H/35S ratios were calculated and there was close agreement between the observed and theoretical ratios. A year later, Ingram (1970) extended these observations using polyacrylanide disc-gel electrophoresis to resolve the “lanthionine peptides” from S. lactis. Essentially, the previously published results were confirmed. Lanthionine and P-methyllanthionine were shown to be formed from cysteine and serine or cysteine and threonine (Fig. lo), which accounts for the fact that when these labeled amino acids were added to the reaction mixture, they were recovered in nisin, although not one of these amino acids is a constituent of nisin (for the structure of nisin, see Fig. 5). Lanthionine and P-methyllanthionine are two of the three nonprotein

:I 2

Time

(a 1

FIG.8. The effect of chloramphenicol (20 &ml) on nisin synthesis and incorporation of [U-’C]-~-glutamicacid by Streptococcus lactis. Radioactivity (-), nisin (---), control (0); chloramphenicol added at time zero (O),at 50 min (0--).After Hurst, 1966a.

107

NISIN

0

2

4

6

0

10

12

14

16

Chloramphenicol(pg /ml )

FIG.9. Protein and nisin synthesisby Streptococcus kwtis in the presence of various mncentrations of chloramphenicol, as a percent of control without antibiotic. Protein (G), nisin (0--).AAer Hurst, 1966a.

amino acids that occur in nisin. The third, dehydroalanine arises from the dehydration of serine (Gross, 1977). The presence of nonprotein amino acids is a common feature of peptide antibiotics. It led Bodanszky and Perlman (1964) to suggest that these amino acids are made by nonribosomal mechanisms, a suggestion which has since been amply confirmed with other peptide antibiotics. They considered that an invariant genetic code does not permit synthesis containing unusual amino acids. However, the synthesis of nisin was sensitive to chloramphenicol and puromycin, inhibitors that af€ect protein synthesis at the ribosomal level. It is most easy to explain these results by assuming the existence of a pronisin that contains only protein amino acids and is synthesized by a ribosomal mechanism. The pronisin is then enzymically modified to the antibiotic. The work of Hurst and Paterson (1971) extended and confirmed these suggestions. Hurst (1967) had earlier isolated a nonantibiotic-producing S . Zactis that produced a peptide that had properties resembling nisin but was not an antibiotic. Preparations of this peptide incubated with cell extract made from the nisin-producing S. Zactis generated an antibiotic that resembled nisin (Hurst and Paterson, 1971).

108

A. HURST

- NHCHCO - - NHCH,CO -

Ribosomal from

-

synthesis

cysteine,

serine

k%OH

SH

and threonine

-NH.YH.CO-

SH

CHfHOH FH2 C ~ & H . N H -COCH.NH-

-

-NH.FH.CO-

-NH.FH*CO-

y42

y

p

S

s

CHjFH I -CO.C H,NH

yI

2

c2

- -COCH.NH-

2

-NH’$COCH2

SH CHjCH II

-CO.C,NH-

2 {:

-COCH.N H-

FIG. 10. Proposed mechanism for the synthesis of nisin based on Ingram, 1970.

The nisin-generating enzyme(s) was made only briefly during the exponential growth phase. Its specific activity declined sharply in stationary phase. It was readily obtained by blending cells with microbeads for only 45 sec, a treatment that affected neither the phase contrast appearance nor the Gram reaction of the cells, suggesting that the enzyme(s) was located on the cell surface. Also, White and Hurst (1968)used physical and chemical means of cell disintegration that showed that nisin was located on the outside of the producer organism. These results are consistent with the notion that the conversion of pronisin into nisin occurs in the outer layers of the cell. It is suggested that the synthesis of pronisin is the step sensitive to inhibitors of protein synthesis. Precursor proteins are well known and accepted concepts in eukaryotic biochemistry. Insulin, which nisin resembles by molecular weight, is now known to be synthesized not only as a proinsulin but also as a preproinsulin (Steiner, 1979).

V. Use of Nisin A. INTRODUCTION Inhibitory streptococci were first thought of as a nuisance in cheesemaking in New Zealand (Whitehead, 1933; Whitehead and Riddet, 1938) and later in Britain (Meanwell, 1943).They were observed in certain batches of milk in which “starter” development was slow, resulting in faulty cheese. Souring by “inhibitory streptococci” was found to destroy the bovine tubercle organism present in the milk, an action not observed in milk soured by the usual starter organisms (Mattick and Hirsch, 1946). This effect directed attention to the possible therapeutic application of nisin. Nisin appeared to

NISIN

1 09

be promising in the treatment of experimental tuberculosis in guinea pigs (Hirsch and Mattick, 1949). Suspensions of nisin in oil were successfully used for udder infusion of cows suffering from streptococcal and staphylococcal bovine mastitis (Taylor et al., 1949). However, the latter reports also emphasized the local irritant effect of nisin. Gowans et al. (1952)confirmed this; it is an effect due to the low solubility of nisin at physiological p H values. The material precipitates at the site of injection and could cause edema, necrosis, and peeling. In addition, at the time of these trials, only impure material was available in limited quantities. Later, when more abundant supplies became available, other antibiotics (e.g., penicillin and streptomycin) were already firmly established therapeutic agents. The possibility of using inhibitory streptococci for food preservation was first suggested by the results of Hirsch et al. (1951). They used a nisinproducing “starter” and successfully prevented gas production by clostridia in cheese. Gas formation by lactate-utilizing clostridia can be a serious defect of Swiss types of cheese (e.g., GruyBre), causing heavy economic loss. This successful application was soon followed by a similar application to processed Swiss-type cheese. McClintock et al. (1952) added to cheese-melt milk soured by nisin-producing streptococci in order to control clostridial blowing: The use of milk soured by nisin-producing streptococci, however, presented a number of industrial problems and Hawley (1955) recommended instead the addition of a nisin-containing skim milk powder. This remains one of the most important applications of nisin. B. TOXICITYAND LEGALASPECTSOF THE USE OF NISIN FOR FOODPRESERVATION 1. Toxicity

There appears to be general agreement that when eaten, nisin is nontoxic. Various concentrations of nisin may occur in milk and cheese as a result of chance contamination with nisin-producing streptococci (Anon, 1959). Frazer et al. (1962)emphasized that nisin has been consumed by people for their normal life spans without apparent ill effects. While this does not prove that nisin is harmless, especially if used in a different context, it does indicate, at least, that nisin has low toxicity. Toxicological evidence was reported in Japan by Hara et al. (1962);they investigated oral administration to kittens and rats and found the LDS0to be similar to that of common salt, i.e., about 7 gm/kg. In Britain, the toxicity of nisin was extensively studied by Frazer et al. (1962).They tested both acute toxicity and the effects of long-term feeding at levels lo00 times greater than what might be consumed naturally.

110

A. HURST

They concluded that nisin was a safe substance for use in food at the level of treatment proposed. Lipinska (1977) quotes a number of studies in the USSR. These, as well as the British tests, concern diverse properties such as carcinogenicity, survival, reproduction, blood chemistry, kidney function, and effect of stress factors. All the results confirm the nontoxicity of nisin. Shtenberg and Ignat’ev (1970) extended these studies to nisin alone and in combination with other preservatives, e.g., sorbic acid. They confirmed the lack of toxicity of nisin. Combinations of preservatives were in general more toxic than the single compounds (e.g. benzoic acidhodium bisulfite) but the reverse was true with the sorbic acidhisin mixture. 2. Legislation

The use of nisin for food preservation was first permitted in Britain (Anon, 1959). More detailed regulations were published later (Anonymous,197213).Its use was permitted without limitation on amount or the need to declare its presence on the label, but was restricted to canned foods that had received a minimal C . botulinum heat-treatment ( F o = 3 instead of the usually applied F , = 9) or that have a pH of less than 4.5, and to cheese and cream. In 1969, the Joint Food Additives Organization/WHOExperts Committee on food additives gave nisin international acceptance (WHO, 1969). They recommended that use of nisin as a food “should be considered acceptable, the unconditional Average Daily Intake being 033,OOO unitskg of body weight.” (There are about 40 x 106 units in 1 gm of pure nisin. Up to 400 units/gm of food are usually recommended for food preservation, i.e., about 10 ppm. A total of 33,OOOunits, the Average Daily Intake, is just less than 1 mg. A 70 kg person may thus consume about 58 mg of pure nisin daily. If the preservative was in processed cheese, 58 kg would have to be consumed daily to reach the maximum daily intake of nisin.) Fowler and McCann (1972) listed 20 countries that permitted nisin in foods. According to Fowler (personal communication), 27 countries permitted its use in 1979 (Table 11). It is interesting to note that Poland is not listed in Table 11, although much of the recent work on nisin comes from that country. Nisin is not allowed in foods in the United States and Canada. Jarvis and Morisetti (1969) reported that an application for the use of nisin in canned foods was rejected in 1966 by the FDA, who asked for more extensive tests before reconsidering this decision,

3 . Cross-Resistance between Nisin and Medically Important Antibiotics Whether organisms exposed to nisin can become resistant to it is discussed in Section 11, D, 5. What concerns us at this point is to learn whether

111

NISIN

TABLE I1 COUNTFUES PERMI~TINCTHE USE OF NISI" Abu Dhabi

Permitted in pasteurized milk, flavored milk, long-life milk, processed cheese, cheese, other dairy products, and canned foods. No limit.

Australia

Permitted in all states for use in cheese, tomato purke, canned tomato pulp, canned tomato paste, and canned fruit, in any case in which the pH of these articles is below 4.5. There are no limits on quantity. Must be declared, not by name, but in the form PRESERVATIVE ADDED, in bold-faced sans serif capital letters of not less than 8 point measurement as prescribed by Regulation 7 under the Pure Food Act. In addition, a federal recommendation to permit the use of nisin in canned soups has been ratified in some states and is awaiting acceptance in others.

Bahrain

Permitted in pasteurized milk, flavored milk, long-life milk, processed cheese, other dairy products, and canned foods. No limit.

Belgium

Permitted in cheese. 2.5 mg/kg.

Bolivia

Nisin permitted, since it is not on the list of prohibited preservatives.

Costa Rica

Permitted in cheese products. No limit.

Cyprus

Permitted in same products as in United Kingdom.

Czechoslovakia

Nisin (200 unitdgm) is allowed in cheese.

Denmark

Permitted in processed cheese. 200 mg/kg (with special permission for the State Dairy & Eggs Board).

Dubai

Permitted in pasteurized milk, flavored milk, long-life milk, processed cheese, cheese, other dairy products, and canned foods. No limit.

Finland

Permitted in processed cheese. 100 ppm in the finished product.

France

Permitted in processed cheese. No limit.

India

Permitted in cheese and processed cheese.

Irish Republic

Permitted in processed cheese. 100 mglkg.

Israel

Permitted in cheese, not including soft white cheeses. No limit.

IdY

Permitted in cheese; maximum concentration 500,000 unitdkg. Permitted in canned vegetables and cream desserts. Maximum concentration 100,OOO unitdkg of finished product.

Kuwait

Permitted in processed cheese. No limit.

Malta

Approved for cheese, clotted cream, canned food, or any food prepared from cheese, clotted cream, or canned food. No limit.

Mexico

Permitted in all foods without limit.

Portugal

Permitted in processed cheese. Maximum 100 mglkg. (Continued)

112

A. HURST

TABLE I1 (Continued) Qatar

Permitted in pasteurized milk, flavored milk, long-life milk, processed cheese, cheese, other dairy products, and canned foods. No limit.

Singapore

Permitted in cheese and canned foods that have been sufficiently heat processed to destroy spores of Clostridium botulinum.

South Africa

Permitted in processed cheese preparations, processed spreadable cheese, and some other cheeses. Limit 500 IU/gm.

Spain

Permitted in processed cheese. 100 mgkg.

Sweden

permitted in sterile condensed milk and cream [Group l(e)], cheese [Group l(f)]. No limit.

United Kingdom

Permitted in cheese, clotted cream, and canned food (defined as “food in an hermetically sealed container which has been sufficiently heat processed to destroy any Clostridium botulinum in that food or container, or which has a pH of less than 4.5”).No limit. Must be declared on label in the form “Contains Permitted Preservative,” but does not have to be identified specifically.

USSR

Permitted in certain canned fruits and vegetables.

“Based on personal communication from G. G. Fowler, November 1979.

resistance to nisin also induces resistance to therapeutically important antibiotics; data on this question virtually do not exist. The only available information is the work of Carlson and Bauer (1957) done with only one strain of Staph. aureus. This strain was trained to become nisin resistant and had unchanged sensitivity to penicillin, streptomycin, Terramycin, and tetracycline. More of this kind of information is needed, particularly because methods used in bacterial genetics have developed so much since 1957. Only when more complete data are available will it be possible to arrive at a full evaluation of the suitability of nisin as a food preservative. However, it is reasonable to suppose that as long as nisin is not used therapeutically or as an animal feed additive, there is little likelihood that it will induce multiple antibiotic resistance. Nevertheless, few data support this notion. Lipinska (1977)states that industrially used, nisin-resistant lactic acid bacteria showed no cross-resistance to medically important antibiotics (penicillin, streptomycin, erythromycin, chloramphenicol, tetramycin, and aureomycin). Nisin is used in flavored milk drinks such as chocolate milk (Fowler and McCann, 1971). The possibility that this might affect the oral flora was considered by several groups of workers. Claypool et al. (1966) and Cowell et al. (1971) used volunteers to drink chocolate flavored milk; then they examined saliva samples. Neither a change in the flora nor nisin-resistant organisms were found. Johnson et al. (1978) used monkeys fed for 5 months with a nisin-containing diet (lo00 IU/gm). They examined the dental plaque

NISIN

113

of these monkeys and reported a significant decrease in the potentially cariogenic streptpcoccal population, but they reported neither nisin resistance of the bacterial population nor the incidence of caries. Gibbs and Hurst (1964)treated sides of pork manufactured to bacon with low levels of nisin (up to 10 ppm). They then isolated nisin-resistant micrococci and lactobacilli from the stored bacon. It is not clear from this work whether these resistant organisms were selected or “trained. ” However, in this context, the crucial factor is the cross-resistance of the organisms with medically important antibiotics and this attribute was not tested for. It is evident from this discussion that there is need for further information on the cross-resistance between nisin and medically important antibiotics.

4 . Nisin and Growth Stimulation Low level antibiotic feeding to farm animals results in growth stimulation but is currently frowned upon because it may induce multiple antibiotic resistance in some components of the gut flora. At first, owing largely to British work, it was thought that nisin was not a growth stimulant. Coates et al. (1951) fed nisin to chicks and Barber et al. (1952) fed milk soured by nisin-producing streptococci to pigs, both without obtaining growth promotion. It was therefore never used as a feedstuff supplement. It was surmised that nisin, being a polypeptide, was being digested; Barber et al. (1952) found no evidence that the streptococci were being established in the gut. That nisin could be destroyed by digestive enzymes was later confirmed by Jarvis and Mahoney (1969). The lack of growth stimulation by nisin of experimental farm animals was in turn a consideration that led to the British approval of nisin as a food additive. On the other hand, more recent results show that nisin may have growthpromoting properties. Increased body weights were reported by Shtenberg and Ignat’ev (1970) in toxicological tests with laboratory animals. Lipinska (1977) summarizes four feeding experiments in Poland and Czechoslovakia involving about 100 calves and 100 piglets. She states that the growthpromoting properties of nisin were clearly demonstrated by these tests, although there was no difference in the intestinal flora between the treated and control groups. At present, nisin is the only antibiotic that is used in food preservation. Because of the dangers of producing nisin-resistant organisms, the application to foods is incompatible with application as a feedstuffs additive.

c. USE OF NISIN-PRODUCING STARTERS IN HARDCHEESE It is generally recognized that clostridial spores potentially capable of causing spoilage occur in most English-type cheese (Cheddar) but are unable

114

A. HURST

to develop, probably because of the combined effects of acidity, salt, and inhibiting flora also present in the cheese (Goudkov and Sharpe, 1966). On the other hand, in the interior of Swiss and Dutch cheeses (Emmenthal, Edam), there is less salt and pH may be higher, so that more favorable conditions exist for the development of these organisms. Nisin is then expected to inhibit the undesirable clostridia but not the flavor-producing lactobacilli and the “eye”-formingpropionic acid bacteria. Such results were reported in early trials with Edam cheese (Kooy and Pette, 1952; Kooy, 1952) and in other trials (Lipinska, 1956; Lipinska et al., 1962). Others, however, reported abnormal or slow ripening of cheese (Winkler and Frohlich, 1958; Galesloot and Pette, 1957; Pulay, 1956). The general conclusions of these workers were that clostridial fermentations were controlled but frequent defects of cheese quality arose owing to the high sensitivity of the desirable flora to nisin. Two other observations contributed to the difficulty of using nisin-producing starters: the high sensitivity of these organisms to bacteriophage attack, and the discovery of nisinase-producing organisms in milk (see Section 111,G). Attempts were then made to use the nisin-producting organisms side by side with nisin-resistant starters. This work is excellently reviewed by Lipinska (1977). Attempts were made to “train” lactic streptococci, S . thwmophilus, various lactobacilli, and propionic acid bacteria to grow in the presence of nisin, but so far these attempts appear to have been only partially successful. Resistance is not permanent, the resistant organisms produce nisinase, or the organisms cease to grow satisfactorily in milk. Nevertheless, Lipinska (1977) reported successful experiments in Poland and in the USSR that were carried out with a combination of nisin-resistant and nisin-producing starters. Edam cheese was made with milk naturally or artifcially infected with clostridial spores. The butyric fermentation was controlled even in cheese of low salt content and the resulting product was said to be of first quality. Similar results were reported for Tilsit cheese, based on large pilot-scale trials using 120 L batches. Lipinska (1977), quoting her own work and that of others, also reported industrial trials with Gouda cheese carried out in Poland and the USSR in 1971 and 1974. Application of nisin-producing starters in combination with the nisin-resistant starters resulted in 90% of the cheeses being of high quality, while 59% of the normally made cheeses were blown. This work probably deserves more attention than it has received at present. Most of the references given by Lipinska are in Polish and Russian journals and trade magazines not easily available to or understood by an English readership. Western manufacturing techniques use nitrate to control butyric fermentation, which is probably reduced by bacterial action to nitrite, the effective inhibitor. Nitrite is a well-known potential carcinogen and its use in foods is

NISIN

115

being discouraged. The work reported by Lipinska offers an interesting alternative to nitrate in cheese-making. There are no recommended nisin levels in milk destined for cheesemaking by lactic and other fermentations. Lipinska (1977) states that good quality cheeses made with nisin-producing and nisin-resistant starters contained 400-1000 IU/gm. Nisin preparations are seldom added to the milk before cheese-making because most of it is lost in the whey. A different application of nisin to cheese-making was described by Jones (1974). When acidification by cheese-starter organisms is slowed down by phage attack, there is a risk that enterotoxigenic staphylococci might develop. Jones found that active nisin-producing starters controlled the development of Staph. aureus. However, when there was starter failure, added nisin preparations failed to control Staph. aureus and enterotoxin A synthesis occurred. Butter fat was shown to diminish the effectiveness of nisin.

D. MILDLY HEAT-PROCESSED FOODS 1 . Processed Cheese, Cheese Spreads, and Gelatin

Following the early successful control in processed cheese of butyric fermentation by nisin (McClintock et al., 1952), numerous authors reported similar findings. These results were reviewed excellently by Hawley (1955), Jarvis and Morisetti (1969), and Lipinska (1977). The application of nisin to edible gelatin was suggested by Eastoe and Long (1959). The development of CZostidium perfringens (6.6 X lo3 spores/ml) in this product was completely prevented by 40 IU/ml, and as little as 8 IU/ml gave noticeable inhibition. A dairy spread containing 3550% milk fat and 8-13% solids that were not fat was prepared by Goel et aZ. (1969). The preparation was stabilized with high levels of nisin (1465 ~ p mand ) ~it had a shelf-life of 5-6 weeks at about 5°C. Later work has shown that high levels of butter fat diminish the effectiveness of nisin (Jones, 1974). Nisin preparations for use in cheese processing have standardized potency by dilution with skim milk powder, lactose, andlor salt. These preparations are known to be made in Poland, the USSR, and Britain. The trade name of the British preparation is Nisaplin, and Fowler (1979) suggested that a reasonable level of usage in processed cheese was 500 mg/kg. (Assuming that Nisaplin contains only 2% nisin, this level corresponds to about 10 ppm in ‘Although this is not clear firom the article, the concentrationprobably refers to a preparation containingabout 2.5%pure nisin; the concentrationin terms of the pure antibiotic was probably of the order of 35 ppm.

116

A. HURST

the finished product.) The concentration chosen depends on several factors. First, the temperature of the melt is important; at 85-95”C, about 15% of nisin activity is destroyed. Second, the minimal inhibitory concentration depends on the spore load; for example, lo4 spores of C. spwogeneslml require 0.3 mg of Nisaplidml, but 10’ sporedm1 require only one-tenth of that concentration. Third, the storage temperature affects the levels used. Finally, nisin activity slowly disappears during storage; the initial levels must be so chosen that sporostatic amounts are still present at the end of the forecasted storage life of the product (Fowler, 1979).

2 . In Semipreserved Meats The use of nisin for the preservation of mildly heat-processed meat products has not been an unqualified success. Reports are contradictory regarding its usefulness. Much of this work was done in the period 1955-1965 and has been reviewed by Hawley (1957, 1962), Marth (1966), Boone (1966), and Jarvis and Morisetti (1969). Many of these reviews are very detailed (e.g., Marth, 1966), but they tend to deal mainly with American and British papers. Most recently, Lipinska (1977)has added new data of Eastern European origin. The reader is also referred to Section II,D,2, on the mode of action of nisin, particularly its effect on spores. Briefly, nisin was never claimed to be a “cure-all” for all preservation problems. Early optimistic claims were made for its use in meat products, but these claims have not been fulfilled. At first, it was thought that nisin might be used in canned ham, tongue, meat, and fish-based sandwich spreads. However, in Britain, its use was not approved for semipreserved meat products because it was felt that its effectiveness was not proven (Anonymous, 1959). Other countries followed suit, and nisin is generally approved in dairy products and canning but not in meat preservation. More recently, Caserio et al. (1979) have shown that nisin can be a very useful preservative in Italian delicatessen sausages, Mortadella, Wurstel, and cooked ham. These sausages and ham normally contain salt and 150 ppm nitrite and are slightly acid (pH 5.6-6.0). All these products received an unspecified cook, and were then packed in Cryovac under vacuum and stored at either 2” or 20°C. The authors note that these products are highly perishable because they contain a high proportion of free water. Their distribution to regions far from the factory is almost impossible during the summer months. They found that the lower level of nisin (150 IU/gm) had little or no value in extending shelf-life but 200 IU/gm offered great advantages (Table 111). In conclusion, it does seem logical that nisin should be tried further in semipreserved meat products. The purpose would be either to extend shelflife or to reduce the undesirably high nitrite levels in use at present. The

117

NISIN

TABLE 111 EFFECTOF NISIN ON THE ORCANOLEPTIC PROPERTIES OF CUREDMEAT PRODUCTS STORED AT 20°C“ Control Product Mortadella Wurstel Ham

2 C

Disagreeable Acid smell, bitter taste Unacceptable

With nisinb

3

2

3

Disagreeable Not acceptable

Acceptable Acceptable

Less acceptabled

Acceptable

Acceptable

Less acceptabled

“Composed from the data of Caserio et al. (1979). ux)IUlgm of product. ‘Time in months. Early signs of deterioration detectable.

,

spoilage organisms in these products are generally lactic acid bacteria, which should be nisin susceptible. Nisin is effective against C . botulinum (Andersen et aZ., 1953; Denny et al., 1961). Indeed it is possible that the previous failures of nisin in cured cooked meats could be attributed to inadequate concentration, and further tests using more nisin would appear to be reasonable.

E. FULLYHEAT-PROCESSED FOODS

1 . Chocolate Milk In low acid foods (i.e., those with pH > 4.5) nisin may be used to reduce the processing to that required to destroy C . botulinum. This organism is not as heat resistant as other spoilage bacteria. An F , value5 of 3.0 is generally accepted as a minimal botulinum cook and processing severities greatly in excess of this value are used in practice to ensure freedom from spoilage. Heinemann et al. (1965) reported several trials that clearly showed the preservative value of nisin in chocolate milk. As little as 80 IU of nisidml (about 2 ppm) completely prevented spoilage in 792 cans, at room temperature for 6 months after treatment to an Fa value of 3.0. Similar results were obtained without nisin but at an Fa value of 11.0. In a subsequent paper from the same group (Claypool et al., 1966), it was argued that nisin inactivation by digestive enzymes would not affect intestinal bacteria, so the only reason against its use might be its effect on the oral flora. However, after consumpS F , value is the time in minutes equivalent to 250°F at the center of the can.

118

A. HURST

tion of chocolate milk, nisin disappeared from the oral cavity in 1 min. It was thought that this rapid disappearance would not give time for resistant organisms to develop. This claim was essentially confirmed by Cowell et al. (1971), who fed high levels of nisin (25,000 I U of nisidday) for 14 days to groups of humans. The treated and control groups were analyzed for viable and nisin-resistant bacteria but no significant differences were found. Notwithstanding, nisin is not permitted in the United States and Canada; it is widely used in flavored milk drinks elsewhere (Fowler, 1979). Because of the reduced heating time, such milks are likely to have improved flavor and nutritional qualities (Gregory et al., 1964). 2 . Dairy Products for Third World Countries

Shehata et al. (1976) in Egypt successfully used low levels of nisin as an adjunct in the production of sterile whole bufFalo milk or chocolate milk. They were able to reduce the heat processing by about 80% and reported satisfactory storage for 21 days at 37°C. Similar results were reported from India (Wajid and Kalra, 1976). Nisin extended the shelf-life of sterilized milk to 60 days whereas the controls spoiled in 3-7 days. Nisin was also used in kheer, an Indian dessert obtained from milk, sugar, and rice (Sukumar et al., 1976).This product stored at 37°C had a shelf-life of 2-3 days and, with nisin (200 IU/gm), 8-10 days. The manufacturers of nisin believe that there could be an important application in those parts of the world that have little or no fresh milk but have plants for the production of reconstituted milk products (Fowler and McCann, 1972). To maintain flavor and nutritional quality, further heat treatment should be reduced; nisin could then help to control surviving bacterial spores. Although this seems a good idea, it should be stressed that nisin is a narrow spectrum antibiotic suitable only for extending the shelf-life of the product. Good manufacturing practice and safe water supply constitute a sine qua non to ensure safe products. The suggested procedure of reduced heat treatment plus nisin appears to be appreciably superior to making powdered dairy products available for reconstitution in areas where the water supply is unsafe. 3. Canned Foods

The stability of nisin increases with decreasing pH and it can be used successfully in high acid foods to control gram-positive spoilage flora. The highly heat-resistant flat-sour organism B. stearothemnophillus is very sensitive to nisin, as is C . thennosaccharolyticurn. Gillespy (1957) carried out trials with canned beans in tomato sauce inoculated with the latter organism and found no spoilage when nisin was used at 200 IU/ml. With nonacid foods, cans should receive a heat treatment to ensure the

NISIN

119

destruction of C. botulinum ( F , = 3). The purpose of nisin, then, is to reduce the energy consumption of the process, to improve nutritional value, and to improve appearance and texture. Numerous such applications have been described, and nisin appears to have application in the canning industries of European countries such as the USSR, Poland, and Hungary. Nisin has been used in carrot puree and mushrooms (Vas, 1964; Heinemann et al., 1965); peas (Kiss et al., 1968); soups (Bardsley, 1962); and other foods (Lipinska, 1977; Fowler, 1979). As an example, the paper by Kiss et al. (1968) describes industrial-scale experiments on the heat processing of Lincoln peas filled in jars of about 1 kg capacity. They used nisin at 200 IU/ml and were able to reduce the normal heat processing F o = 7.4 (equivalent to 20 min at 118OC) by 60%(equivalent to 10 min at 111-113°C). Only partial control of spoilage was reported by Heinemann et al. (1965) in chicken chow mein and cream-style corn deliberately contaminated with spores of the anaerobic spoilage organism PA 3679 inoculated at levels lo3lo4 times greater than occur naturally. Gibbs and Hurst (1964) reported similar experimental results with a soup made from heavily contaminated and unwashed lentils. These last two citations emphasize that nisin cannot be used to mask poor quality raw materials (Jarvis and Morisetti, 1969). REFERENCES Adams, D. M. (1978). Ado. Appl. Microbiol. 23, 245-261. Alifax, R., and Chevalier, R. (1962). J. Dairy Res. 29, 233-240. Andersen, A. A., Michener, H. D., and Olcott, H. S. (1953). Antibiot. Chemother. (Washington, D . C . ) 3, 521-526. ' Anonymous (1959). Food Standards Committee Report on Preservatives in Food, p. 49. HMSO, London. Anonymous (1966). Food Chem. News 8 (Aug. 22), 25. Anonymous (1972a). Report to the Commissioner of the Food and Drug Administration by the FDA Task Force on the Use of Antibiotics in Animal Feeds. FDA, Rockville, Maryland. Anonymous (1972b).Food Additives and Contaminants Committee Report on the Review of the Preservatives in Food Regulations 1962. HMSO London. Bailey, F. J., and Hurst, A. (1971). Can. J. Microbiol. 17, 61-67. Barber, R. S., Braude, R., and Hirsch, A. (1952). Nature (London) 169, 200-201. Bardsley, A. (1962). Food Technol. Aust. 14, 532537, 606-611. Baribo, L. E., and Foster, E. M. (1951). J . Dairy Sci. 34, 1136-1144. Bavin, E. M., Beach, A. S., Falconer, R., and Friedmann, R. (1952). Lancet (i), 127-129. Beach, A. S. (1952).I . Gen. Microbiol. 6, 60-63. Bemdge, N. J. (1947). Lancet (ii), 7-8. Berridge, N. J. (1949). Biochem. J. 45, 486493. Berridge, N. J. (1953). Chem. Ind. 1953, 1158-1161. Berridge, N. J., and Barrett, J. (1952). J . Cen. Mimobiol. 6, 14-20. Berridge, N. J., Newton, G. G. F., and Abraham, E. P. (1952). Biochem. J. 52, 529535. Bodanszky, M., and Perlman, D. (1964). Nature (London) 204,840-844.

120

A. HURST

Boone, P. (1966).Food Manuf. 41,49-51. Campbell, L. L., and O’Brien, R. T. (1955).Food Technol. (Chicago) 9,461465. Campbell, L. L., and Sniff, E. E. (l959a).J . Bacteriol. 77, 766-770. Campbell, L. L., and Sniff, E. E. (1959b).Appl. Microbiol. 7,298-291. Campbell, L. L., and Sniff, E. E., and O’Brien, R. T. (1960)Food Technol., 13,462. Carlson, S.,and Bauer, H. M. (1957).Arch. Hyg. Bakteriol. 141,445459. Caserio, G., Ciampella, A,, Gennari, M., and Barluzzi, A. M. (1979).Industria alimntari, No. 1, 1-12. Cheeseman, G. C., and Berridge, N. J. (1957).Biochem. J . 65,603408. Chevalier, R., Fournaud, J., Lefebre, E., and Mocquot, G. (1957).Ann. Technol. Agric. 2, 117-137. Claypool, L., Heinemann, B., Voris, L., and Stumbo, C.R. (1966).J . Dairy Sci. 49,314316. Coates, M. E., Harrison, G. F., Kon, S. K., Mann, M. E., and Rose, C. D. (1951).Biochem. J. 48, XII. Cowell, N. D., Allen, A. R., and Jarvis, B. (1971).J . Appl. Bacteriol. 34, 787-791. COX,G. A. (1934).N . Z . J . Agric. 49,231-234. Deibel, R. H., and Seeley, H. W., Jr. (1974).In “Bergey’s Manual of Determinative Bacteriology” (R. E. Buchanan and N. E. Gibbons, eds.), p. 507.Williams and Wilkins, Baltimore, Maryland. Denny, C. B., Sharpe, L. E., and Bohrer, C. W. (1961).Appl. Microbiol. 9, 108-110. Dimick, K. P., Alderton, G., Lewis, J. C., Lightbody, H. D., and Fevold, H. L. (1947).Arch. Biochem. 15, 1-11. Eastoe, J. E., and Long, J. E. (1959).J . Appl. Bacteriol. 22, 1-7. Egorov, N. S.,Baranova, I. P., and Kozlova, Yu. I. (1971).Mikrobiologiya 40, 860464. Egorov, N. S., Baranova, I. P., and Kozlova, Yu, I. (1976).Mikrobiologiya 45, 87-90. Falconer, R. (1952).British Patent No. 683423. Fowler, G. G. (1979).Food Manuf. 54,57-59. Fowler, G. G., and McCann, B. (1971).Aust. J . Dairy Technol. 26, June, 4446. Fowler, G. G., and McCann, B. (1972).Food Industries of South Af;lca 25,49-50,55. Frazer, A. C., Sharratt, M., and Hickmann, J. R. (1962).J . Sci. Food A@. 13,3242. Fuchs, P. G., Zajdel, J., and Dobrzanski, W. T. (1975).J . Gen. Microbiol. 88, 189-192. Galesloot, Th.E. (1956).Ned. Melk Zuioeltijdschr. 10, 143-155. Galesloot, Th. E., and Pette, J. W. (1957).Ned. Melk Zuiueltijdschr. 11, 144-151. Gibbs, B. M., and Hurst, A. (1964).In “Microbial lnhibitors in Foods” 0. Molin, ed.), pp. 151-165.Almqvist and Wiksell, Stockholm. Gillespy, T. G. (1957).Fruit and Vegetable Canning and Quick Freezing Research Association, Leatlet No. 3. Chipping Campden, United Kingdom. Goel, M. C., Calbert, H. E., and Marth, E. H. (1969).J. Mllk Food Technol. 32,312318. Goudkov, A. V., and Sharpe, M. E. (1966).J . Dairy Res. 33, 139-149. Goudkov, A. V., Trofimova, T. I., Dolidze, G. G., Lyubimova, L. A., Sileva, M. N., and Blagushnia, R. F. (1973).Antibiotiki 18, 162-165. Gould, G. W. (1964).In “Microbial Inhibitors in F o o d (N. Molin, ed.), pp. 17-24.Almqvist and Wiksell, Stockholm. Gould, G. W., and Hurst, A. (1962).Proc. 8th Int. Conf. Microbiol., Abstr. A2-11. Gowans, J. L., Smith, N., and Florey, H. W. (1952).Brit. J. P h a m c o l . 7, 438449. Gregory, M. E., Henry, K. M., and Kon, S. K. (1964).J . Dairy Res. 31, 113-119. Gross, E. (1975).In “Peptides: Chemistry, Structure and Biology.” Proceedings of the Fourth American Peptide Symposium. (R. Walter and J. Meienhofer, eds.), pp. 3142.Ann Arbor Sci., Ann Arbor, Michigan.

NISIN

121

Gross, E. (1977).In “Protein Cross-linking-B” (M. Friedman, ed.), pp. 131-153. Plenum, New York. Cross, E., and Morell, J. L. (1967).J . Am. Chem. Soc. 89, 2791-2792. Gross, E.,and Morell, J. L. (1968).FEBS Lett. 2, 61-64. Cross, E.,and Morell, J. L. (1970).J . Am. Chem. Soc. 92,2919-2920. Gross, E.,and Morell, J. L. (1971).J . Am. Chem. Soc. 93,4634-4635. Gross, E.,Morell, J. L., and Craig, L. C. (1969).Proc. Natl. Acad. Sci. U.S.A. 62,952-956. Ha& R. H. (1966).Process Biochem. 1, 461434. Ham, S.,Keido, Y.,Koji, M., Tokako, T., Tatsumori, K., Masachiro, S., Ziro, I., and Takeshi, S. (1962). Department of Pharmacology Report, Tokyo Medical College. Quoted after Caserio et al., 1979. Hawley, H. B. (1955).J . Appl. Bacteriol. 18,388395. Hawley, H. B. (1957).Food Manuf. 32,370376,430434. Hawley, H. B. (1962).In “Antibiotics in Agriculture” (M. Wnodbine, ed.), pp. 272-288.Buttenvorth, London. Heinemann, B., and Williams, R. (1966).J . Dairy Sci., 49,312313. Heinemann, B.,Voris, L., and Stumbo, C. R. (1965).Food Technol. 19, 592-596. Hirsch, A. (1950).J . Gen. Microbiol. 4, 70-83. Hirsch, A. (1951a).Nature (London) 167, 1031-1032. Hirsch, A. (1951b).J . Cen. Microbiol. 5, 208-221. Hirsch, A. (1952).J . Dairy Res. 19, 291-293. Hirsch, A. (1954).J . Appl. Bacteriol. 17, 108-115. Hirsch, A., and Grinsted, E. (1951).J . Dairy Res. 18, 198-204. Hirsch, A., and Crinsted, E. (1954).J . Dairy Res. 21, 101-110. Hirsch, A., and Mattick, A. T. R. (1949).Lancet (ii), 190-197. Hirsch, A., Crinsted, E., Chapman, H. R., and Mattick, A. T. R. (1951).J . Daisy Res. 18, 205-206. Hitchins, A. D., Gould, G. W., and Hurst, A. (1963).J . Gen. Microbiol. 30, 445-453. Hoyle, M., and Nichols, A. A. (1948).J . Dairy Res. 15, 398408. Hurst, A. (1966a). J . Gen. Microbiol. 44, 209-220. Hurst, A. (1966b). J . Cen. Microbiol. 45, 503413. Hurst, A. (1967).Nature (London) 214, 1232-1234. Hurst, A. (1968).Nature (London) 219, 403-404. Hurst, A. (1969).J . Bacteriol. 97, 1062-1068. Hurst, A. (1972).J . Milk Food Technol. 35,418-423. Hurst, A. (1978).In “Streptococci” (F. A. Skinner and L. B. Quesnel, eds.), pp. 297314. Academic Press, London. Hurst, A,, and Dring, G. J. (1968).J . Cen. Microbial. 50, 383390. Hurst, A., and Kruse, H. (1970).Can. J . Microbiol. 16, 1205-1211. Hurst, A., and Kruse, H. (1972).Antimicrob. Agents Chemother. 1, 277-279. Hurst, A., and Lazarus, W. (1968).Nature (London) 219,404-405. Hurst, A., and Paterson, C. M. (1971).Can. J . Microbid. 17, 1379-1384. Hurst, A., and Stubbs, J. M. (1969).J . Bacteriol. 97, 1466-1479. Ingram, L.(1969).Biochim. Biophys, Acta 184, 216-21b. Ingram, L.(1970).Biochim. Biophys. Acta 334,263-265. Ingram, L.,Tombs, M. P., and Hurst, A. (1967).Anal. Biochem. 20, 24-29. Jarvis, B. (1967).J. Cen. Microbiol. 47,3348. Jarvis, B. (1970).Biochem. J . 119, 56P. Jarvis, B.,and Farr, J. (1971).Biochim. Biophys. Acta 227, 232-240.

122

A. HURST

Jarvis, B., and Mahoney, R. R. (1969). J. Dairy Sci. 52, 1448-1450. Jarvis, B., and Morisetti, M. D. (1969).Int. Biodeterior. Bull. 5, 39-61. Jarvis, B., Jeffcoat, J., and Cheeseman, G. C. (1968). Biochim. Biophys. Acta 168, 153-155. Johnson, I. H., Hayday, H., and Colman, G. (1978). J. Appl. Bacteriol. 45, 99-109. Jones, L. W. (1974). Can. J . Microbiol. 20, 1257-1260. Kalra, M. S., Kuila, R. K., and Ranganathan, B. (1973). Erperientia 29, 624-625. Kiss, I., Kiss, K. N., Farkas, J., Fabri, I., and Vas, K. (1968). Elelmiszertudomuny 2, 51-57. Knox, J. R.,and Keck, P. C. (1973). Biochem. Biophys. Res. Commun. 53, 567-571. Knox, J. R., and Keck, P. C. (1975). Acta Cystallogr. Sect. B 31, 2698-2700. Kooy, J. S. (1952). Ned. Melk Zuioeltijdschr. 6, 323-330. Kooy, J. S., and Pette, J. W. (1952). Ned. Melk Zuioeltzj’dschr. 6, 317-322. Kozak, W., and Dobrzanski, W. T. (1977). Acta. Microbiol. Pol. 26, 361-368. Kozak, W., Rajchert-Trzpil, M., and Dobnanski, W. T. (1973). Bull. Acad. Pol. Sci. 21, 811-817. Kozak, W., Rajchert-Tnpil, M., and Dobrzanski, W. T. (1974). J . Gen. Microbiol. 83,295302. Larsen, L. D., and McKay, L. L. (1978). Appl. Enuiron. Mimobiol. 36, 944-952. Lewis, J. C., Michener, H. D., Stumbo, C. R., and Titus, D. S. (1954).J . Agric. Food Chem. 2, 298302. Linnett, P. E., and Strominger, J. L. (1973). Antimicrob. Agents Chemother. 4, 231-236. Lipinska, E. (1956). Acta. Mimobiol. Pol. 5, 271-275. Lipinska, E. (1977). In “Antibiotics and Antibiosis in Agriculture” (M. Woodbine, ed.), pp. 103-130. Buttenvorth, London. Lipinska, E., Strzalkowska, M., Goettlich, W., and Soltys, W. (1962). Proc. 16th Int. Dairy Congr. 2,849-860. McClintock, M., Serres, L., Manzolf, J. J . , Hirsch, A., and Mocquot, G. (1952). J . Dairy Res. 19, 187-193. Marth, E. H. (1966). Residue Reo. 12, 65-161. Mattick, A. T. R., and Hirsch, A. (1946). Lancet (i),417418. Mattick, A. T. R., and Hirscb, A. (1947). Lancet (ii), 5-7. Meanwell, L. J. (1943). Proc. Soc. Agric. Bacteriol. 19, 21. Mocquot, G., and Lefebre, E. (1956). J . Appl. Bacteriol. 19, 322-323. Morell, J. L., and Gross, E. (1973). J . Am. Chem. Soc. 95, 6480481. Mukherjee, P. K., and Paulus, H. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 780-7&1. O’Brien, R. T., Titus, D. S., Devlin, K. A., Stumbo, C. R.,and Lewis, J. C. (1956). Food Technol. 10, 352355. Oxford, A. E. (1944). Biochem. J. 38, 178-182. Pulay, G. (1956). Proc. 14th Int. Dairy Congr. 2, 432-440. Ramseier, H. R. (1960). Arch. Mikrobiol. 37, 57-94. Reddy, M. S., Vedamuthu, E. R., Washam, C. J., and Reinbold, G. W. (1971). J . Milk Food Technol. 34,236-240. Sack, L. E. (1952). Antibiot. Chemother. (Washington, D . C . ) 2, 79-85. Sandine, W. E., Radich, P. C., and Elliker, P. R. (1972).J . Milk Food Technol. 35,176-185. Sevag, M. G . , Lackman, D. B., and Smolens, J. (1938). J. Biol. Chem. 124, 425436. Shehata, A. E., Khalafalla, S. M., Magdoub, M. N. I., and Hofi, A. A. (1976). Egypt. J. Dairy Sci. 484, 3742. Shtenherg, A. I., and Ignat’ev, A. D. (1970). Food Cosmet. Toricol. 8, 369-380. Steiner, D. F. (1979). Nature (London) 279, 674-675. Stumbo, C. R., Voris, L., Skaggs, B. G., and Heinemann, B. (1964).J. Food Sci. 29,859461. Sukumar, D., Thompkinson, D. K., Gahlot, D. P., and Mathur, 0. N. (1976). Indian]. Dairy Sci. 29, 316318.

NISIN

123

Tagg, J . R.,and Wannamaker, L. W. (1978).Antimimob. Agents Chemother. 14,3139. Tagg, J . R.,Dajani, A. S., and Wannamaker, L. W. (1976).Bacteriol. Reo. 40, 722-756. Taylor, J . I., Hirsch, A,, and Mattick, A. T. R. (1949).Vet. Rec. 61, 197-198. Thorpe, R. H. (1960).J . Appl. Bacteriol. 23, 136-143. Tramer, J. (1964).In "Microbial Inhibitors in Food" (N. Molin, ed.), pp. 25-33.Almqvist and Wiksell, Stockholm. Tramer, J . , and Fowler, G. G. (1964).J . Sci. Food Agric. 15, 522528. Vas, K. (1964).Dtsch. Lebensm.-Rundsch. 60,63-67. Wajid, H. R. A,, and Kalra, M. S. (1976).J. Food Sci. Technol. 13,6-8. White, R. J., and Hurst, A. (1968).J . Gen. Mimobiol. 53, 171-179. Whitehead, H. R. (1933).Biochem. J . 27, 1793-1800. Whitehead, H. R.,and Riddet, W. (1938).N. 2. J . Agric. 46, 225-229. World Health Organization (1969).Specifications for Identity and Purity of Some Antibiotics. WHOlFood Add.(69.34,pp. 5347. Wilimowska-Pelc, A., Olichwier, Z., Malicka-Blaszkiewicz, M., and Mejbaum-Katzenellenbgen, W. (1976).Acta Microbiol. Pol. 25, 71-77. Winkler, S., and Frohlich, M. (1958).Milchwiss. Ber. 8, 87-96. Woodruff, H. B. (1966).Symp. SOC. Gen. Mimobiol. 16,2246.

This Page Intentionally Left Blank

The Coumermycins: Developments in the Late 1970s ~ O H NC.

GODFREY

Godfrey Science and Design, lnc., White Plains, New York

........................................... ...................... 111. Biological Developments ................................ Iv. Conclusion ............................................ References ............................................ I. Introduction

11. Chemical Developments, 1972-1979

125 125 128 1% 133

1. Introduction A fairly recent review (Godfrey and Price, 1972) has provided a comprehensive summary of all information bearing upon the structure-activity relationships (SARs) in coumermycins. The review included not only the natural coumermycins and their relative RP 18,631, clorobiocin, but also the large number of semisynthetic coumermycins that had been prepared by that time. The same material was republished without change in the 1977 monograph “Structure- Activity Relationships among the Semisynthetic Antibiotics” (Godfrey and Price, 1977). The coverage of literature bearing upon SARs through 1972 was sufficiently thorough that such information will not be presented in this article. In preparation for the present article, the entire coumermycin literature, both publications and patents, was assembled. All known coumermycin publications and patents are listed in the References. Since these include all articles referenced in the cited review, not all of the entries in the References will be cited in the text of the present article. The Reference Section is presented in its entirety in the interest of listing conveniently all of the patents and publications on coumermycin through 1979. After 1972, there were only two patents and one publication until 1976. In that year appeared the first reports of modern studies on the mechanism by which coumermycin interferes with DNA replication. Since 1976, interest in this activity of coumermycin has increased significantly, with four papers in 1977, ten in 1978, and seven in 1979.

II. Chemical Developments, 1972-1979 In a careful investigation of the steps in the biosynthesis of coumermycin, Scannell and Kong (1969)proved that 8-aminolevulinic acid is the precursor for the central 3-methylpyrrole-2,4-dicarboxylicacid of the coumermycins 125 ADVANCES IN APPLIED MICROBIOIKGY. VOLUME 27 C o m g h t @I lMd by Academic Press, Inc. AU rights of reproduction in any form remewed. ISBN O-l%KI2627-8

126

JOHN C. CODFREY

decorboxylotlon HO&-CHz I ,CHpCHpCOzH YHP O=C 1 ,C=O ,CHz H$ HZN-CY

oxidation

+

t

I

4 -aminoievulinic

oxidative deamlnatlon

acid

1

H3cDC0

HOpC

H

FIG. 1. Biosynthesis of 3-methylpyrrole-2.4-dicarboxylicacid (Scannell and Kong, 1969).

Glucose

I

hexokinase

HO

J,

tyrosine oxidative cyclization

Glucose 6-phosphate V

Glucose I-phosphate nucleotidal t ransferaso

H ...

Uridine diphosphoglucose

1

I- epimerize

methylation

E%Ga+a

FIG. 2. Steps in the biosynthesis of the noviosylcoumarin moiety, “NC” (Kominek and Sebek, 1974).

COUMERMYCINS IN THE

1970s

127

(Fig. 1). These workers also proved that proline is the precursor of the terminal pyrrole-2-carboxylic esters. Esterification of the novioses with pyrrole-2-carboxylic acid leads to coumermycin A 2, while 5-methylpyrrole2-carboxylic acid gives coumermycin A l . They did not establish whether methylation of the pyrrole nucleus occurs before or after esterification. The study of Scannell and Kong (1969) complements that of Kominek and Sebek (1974),in which the sequence of the major steps in the biosynthesis of the coumermycins was elucidated. The noviosylcoumarin moiety “NC” is elaborated from glucose, as shown in Fig. 2. Two noviosylcoumarins are condensed with one 3-methylpyrrole-2-carboxylicacid (in undetermined sequence) to form the despyrroyl intermediate. The latter is then esterified, either with pyrrole-2-carboxylic acid (coumermycin A,) or the 5-methylpyrrole-2-carboxylicacid (coumermycin A1)(Fig. 3). Although it is still not known whether methylation of the pyrrole-2-carboxylic acid occurs before or after esterification, it is known that this methyl group, as well as those of noviose (5-methyl and 4-0-methyl) and coumarin (8-methyl) derive from methionine. In light of the early discovery (Claridge et al., 1966) of the strong influence of cobalt on direction of the biosynthesis toward coumermycin Al, it is apparent that pyrrole methylation is more highly dependent

d-

-c

N

H

R = CH,,

Coumrrmycln A,

R=H,

Coumrrmycin

4

FIG.3. Final steps in biosynthesis of the coumermycins (Kominek and Sebek, 1974).

128

JOHN C. GODFREY

upon the presence of adequate vitamin B12 than are the other methylations. Once the structure of coumermycin Al had been established (Kawaguchi et al., 1965c,d),all publications assumed that the noviose-coumarin linkage was best represented as the a-anomer, as in Fig. 3. Although the NMR spectrum of coumermycin Al was consistent with the a-anomeric structure, the identity was not completely unambiguous. The correctness of the assumption was finally established by an X-ray dfiaction study by Wick et al. (1976). The placement of the chlorine in the coumarin-8 position and the methyl in the 5 position of the pyrrole of the novobiocidcoumermycin hybrid clorobiocin, RP 18,631, completed the structure of this interesting antibiotic (Dolak, 1973; see Fig. 4.) An excellent review of the isolation and purification procedures for clorobiocin and for coumermycin Al was published by Berger and Batcho (1978).

111. Biological Developments A highly signifcant discovery concerning the mode of action of coumermycin Al was reported by Gellert et al., (197613)in December 1976. In the previous month, Gellert et al. (1976a) had reported the discovery of an ATP-dependent enzyme, which they called DNA gyrase. The existence of this enzyme was inferred from the observations that (1)integrative recombination of phage A DNA in a cell-free system from Escherichia coli required a negatively supercoiled DNA substrate, and (2) the substrate could be replaced by relaxed closed-circular DNA only if the DNA was first incubated with an E. coli fraction and ATP. Purification and partial characterization of the enzyme were reported at that time. This discovery was followed by the finding of the DNA gyrase inhibition by coumermycin A (Gellert et al., 1976b). So far as is known at this time, this activity is almost unique to

-.*

*+p a‘

f

0

OH

0

Clorobiocin

FIG.4. The structure of clorobiocin (RP 18,631) showing its close relationship both to the cournermycinsand to novobiocin.

COUMERMYCINSIN THE

1970s

129

coumermycin A,, being shared solely by novobiocin which is, however, much less active. Clues to the existence of DNA gyrase had existed for some time. It had been known for a number of years that coumermycin A, (and novobiocin) inhibits nucleic acid synthesis in intact, sensitive organisms. It was shown more recently by Staudenbauer (1975) and by Ryan (1976a,b) that these antibiotics strongly and specifically inhibit semiconservative DNA replication. Whereas replication in toluenized E. co2i cells was 50% inhibited by coumermycin A, at 0.035 pglml, inhibition of transcription required 0.600 pg/ml. When a coumermycin A,-resistant E. coli was selected and investigated under the same conditions, replication was 50% inhibited at 0.140 &ml, but RNA synthesis was not affected by much higher concentrations of the antibiotic. In spite of these marked effects on E. coZi, Ryan’s evidence suggested that DNA was not degraded by coumermycin A,, and, in fact, a number of standard biochemical analyses showed no interaction between DNA and coumermycin A 1 . Although Ryan suggested a number of plausible explanations for these facts, the nearly simultaneous announcements of the discovery of DNA gyrase and its inhibition by coumermycin A , (Gellert, 1976b) provided the most likely explanations of Staudenbauer and Ryan’s observations. The observations relating DNA and RNA inhibition to an interaction of coumermycin Al (and novobiocin, at higher concentrations) with DNA gyrase were briefly reviewed by Cozzarelli (1977). e stated succinctly that “it appears that DNA gyrase is the target enzyme [ or coumermycin A,] and that negative supercoiling is required for both DNA replication and transcription.” In 1977, Gellert et al. reported their results from an investigation of the inhibition of DNA gyrase by nalidixic and oxolinic acids. They found that the enzyme is inhibited by both of these acids, but that oxolinic acid is much more potent than nalidixic acid. The locus of action of oxolinic acid is entirely different from that of coumermycin A,. They found no cross-resistance between oxolinic acid inhibl’tionand coumermycin Al inhibition of E. coli DNA gyrase. They hrther discovered that in the absence of ATP and with a high concentration (6 m M )of magnesium ion, DNA gyrase will transform supercoiled DNA to the relaxed form-in exact opposition to what DNA gyrase usually does. This relaxing activity was totally insensitive to the coumermycin family of antibiotics, but was inhibited by 50 pg/ml of oxolinic acid, providing more evidence of the different loci of action of the two agents. It was proposed that the nickingdosing activity of DNA gyrase is carried by a subunit sensitive to oxolinic acid (but not to coumermycin), and that this nickingdosing activity is different from that of the known w-protein. Support for the hypothesis that coumermycin A, (and novobiocin, at higher

B

130

JOHN C. GODFREY

concentrations)interacts with DNA gyrase at a point different from the locus of action of nalidixic and oxolinic acids is found in the work of Higgins et al. (1978), who showed that the coumarin antibiotic inhibits supertwisting but not nicking and closing, and that nalidixic and oxolinic acids do inhibit nicking and closing and induce site-specific covalent attachment of the gyrase complex to DNA, as well as inhibit supertwisting. Further evidence for the presence of two different subunits in E. coli DNA gyrase was provided by the finding of Hays and Bohma (1978a) that coumermycin Al blocks repair and recombination of a phage ADNA by inhibition of a DNA gyrase subunit that is different from the subunit that is blocked by oxolinic acid. Evidence has been presented by Hansen and Meyenburg (1979) suggesting that the coumermycin Al- sensitive B subunit of E. cold DNA gyrase is a protein having a molecular weight of 91,000. The treatment of a mixture of ColE1 DNA and DNA gyrase with oxolinic acid renders the DNA susceptible to double-strand scission by protease K, resulting in the full length linear form (Gellert et al., 1977). The gyraseinduced cleavage was shown to be different from that produced by the known relaxation complexes. On the basis of evidence then available, the authors proposed that DNA gyrase has a coumermycin-sensitive subunit that carries out the energy-transducing role of coupling ATP hydrolysis to produce a (negative)twisting stress on the DNA. In order to introduce negative superhelical turns into DNA, the gyrase must partition the closed-circular DNA into positive and negative supercoiled regions and carry out a nicking-closing reaction with the (+) region to relieve the positive winding stress so that it can reclose with a negative twist. This activity is energized by ATP . The work of Itoh and Tomizawa (1977)on the replication of phage T7 DNA proved that DNA gyrase also has a role in the replication of linear doublestranded DNA. In their T7-infected E. coli system, inhibition of DNA gyrase was shown by a 50% reduction in thymidine incorporation into T7 DNA at a coumermycin concentration of about 9 pglml. Using the yield of P-galactosidase from synthesis directed by linear DNA fragments from E. coli as an indicator, Yang et al. (1979) produced further evidence that transcription from a linear template is sensitive to inhibition of DNA gyrase. McCarthy (1979)found that the DNA-delay mutants of phage T4 require a functional host (E. coli) gyrase for growth and efficient initiation of replicative growing points in DNA. Both coumermycin Al and novobiocin inhibit phage T4 DNA synthesis and phage production via interference with the host’s DNA gyrase. Coumermycin A, was from three to seven times as active as novobiocin in his experimental systems. He noted that genetic experiments involving derivatives of E. coli K12 can be done more easily with

COUMERMYCINS IN THE

1970s

13 1

coumermycin A, than with novobiocin because E. coli K12 strains possess a permeability barrier to novobiocin, but not to coumermycin Al. Taylor and Levine (1979)have shown that a mutant H group plasmid uses E. coli DNA gyrase for replication. This was demonstrated by the use of coumermycin A, or novobiocin to induce resistant mutants. In their study, coumermycin A was about 10 times as active as novobiocin and this difference was thought to be the result of much better penetration of bacterial cells by coumermycin A,. Drlica and Snyder (1978) showed that coumermycin A relaxes chromosomal DNA supercoiling in E . coli because it inhibits DNA gyrase. They isolated folded chromosomes from coumermycin A I-treated cells and deduced that this chromosomal DNA is largely intact and capable of forming positive supertwists at high ethidium bromide concentrations. The reduction of chromosome supercoiling was found to parallel the rapid decline in DNA synthesis after coumermycin A, treatment, in agreement with the hypothesis that chromosome replication depends on DNA gyrase and negative supercoiling. Falco et al. (1978) reported that in uivo transcription by the N 4 virion RNA polymerase is inhibited by coumermycin Al. It is inferred therefrom that the E. coli DNA gyrase is required for N4 transcription. Chao (1978) has presented evidence for an interaction between the nalidixic acid target and the coumermycin Al target of E. coli DNA gyrase. He found a clear (though not strong) synergy between the effects of nalidixic acid and novobiocin on the growth of E . coli KL161, and increased sensitivity to novobiocin on the part of a nalidixic acid resistant mutant, KL166. This unusual collateral sensitivity is taken as evidence for an interaction between the two separate drug targets with DNA gyrase. It has been recently reported that coumermycin A l does block the repair of UV-irradiated DNA (Hays and Bohma, 1978a). Although these authors point out that this finding is in apparent conflict with the report of Staudenbauer (1975) that DNA gyrase inhibitors efficiently block semiconservative DNA replication but do not block DNA repair, they suggest that the difference may lie in their criterion for repair, which is the restoration of DNA infectivity. Coumermycin Al has been found to be much more effective than novobiocin in the inhibition of phage promoter-dependent transcription of the trp operon in 480 p t r p (Smith et al. 1978). DNA gyrase was found to be involved in this transcription, It was suggested that the greater inhibition by coumermycin A, could be due to the greater permeability of E. coli to this antibiotic. Between its discovery in 1976 and early 1979, DNA gyrase was found only in bacteria and viruses. Knowing that its discovery in mammalian cells as

132

JOHN C . GODFREY

well would have broad implications for mammalian genetic manipulations and possibly for antitumor research and therapy, a number of workers set out to find it there. Castora and Simpson (1979a,b)were the first to provide strong evidence that rat liver mitochondria do in fact contain DNA gyrase. Their sucrose and cesium chloride gradient separations of DNA fractions obtained after incubation of rat liver mitochondria with coumermycin A], novobiocin, nalidixic acid, and oxolinic acid showed that fragments having the density associated with relaxed circular DNA were produced. Electron microscopy confirmed that the fractions so identified were almost entirely relaxed circular DNA. Isolation of the postulated DNA gyrase is in progress. In their rat liver mitochondria system, coumermycin Al was 25 times as active as novobiocin in its ability to inhibit the DNA gyrase. The best description of the subunit structure and ATPase activity of highly purified DNA gyrase is that published by Gellert and co-workers (Mizuuchi et al. 1978). DNA gyrase from E. coli was purified to near homogeneity and was then found to consist of two subunits of MW 90,OOOand 100,OOO. They are present in equimolar amounts. The subunits are identified as the products of two genes, one of which determines resistance to coumermycin Al (and novobiocin), while the other determines resistance to nalidixic and oxolinic acids. These have been named, respectively, the cou gene and the nalA gene. The cou subunit was found to have an ATP binding site and the

A

B

C

D

FIG.5. Possible model for the catalysis of DNA supercoiling by DNA gyrase. ATP-driven translocation of DNA past the bound enzyme (A), combined with rotation of the helix to keep the same structural features facing the enzyme, partitions the DNA into overwound (+) and underwound (-) loops (B). The overwound loop is then relaxed by the nicking-closing activity (C). Removd of enzyme liberates a negatively supercoiled DNA molecule (D).

COUMERMYCINS IN THE

1970s

133

DNAdependent hydrolysis of ATP has been shown to be blocked by novobiocin (and therefore also by coumermycin Al). The nu& subunit is the one specifically involved in the energy transduction step of the negative supercoiling reaction. The authors propose a model for the action of DNA gyrase (Fig. 5). A model that is similar although somewhat less detailed in its treatment of the concept has been proposed by Liu and Wang (1978).

IV. Conclusion The steps in the biosynthesis of coumermycin Al have been completely defined, with the possible exception of the timing of the methylation of the coumarin and the pyrrole-2-carboxylate moieties. Mode of action studies have led to the discovery that coumermycin Al is a specific inhibitor of the ATP-dependent enzyme DNA gyrase. Coumermycin A, is active at very low concentrations, preventing DNA gyrase from carrying out negative supercoiling of double-stranded circular DNA. Because coumermycin Al does not act directly upon DNA itself, the uncoiled DNA is recoverable without damage and may be recoiled with fresh, uninhibited DNA gyrase to form the negative superhelix, or with ethidium bromide to form the positive supercoil. In the absence of ATP, DNA gyrase has been found to carry out uncoiling of negatively supercoiled DNA. Coumermycin Al may then be added to prevent recoiling, and the DNA so obtained is rendered susceptible to agents that would otherwise be denied access to it. Data currently available suggest that the mechanisms of interaction of coumermycin Al and novobiocin with DNA gyrase(s)are very similar. However, novobiocin is in general between 1/3 and 1/100 as active as coumermycin (weight basis), and there is evidence that in E. coli K12, at least, this difference is due to a permeability barrier to novobiocin that does not prevent penetration of coumermycin Al. REFERENCES Batcho, A. D., Berger, J . , Furlenmeier, A., Keller, O., Pecherer, B., Schocher, A. J . , Spiegelberg, H., and Vaterlaus, B. P. (1969) Swiss patent 472,428, June 30. Batcho. A. D., Berger, J., Furlenmeier, A., Keller, O., Pecherer, B., Schocher, A. J., Spiegelberg, H., and Vaterlaus, B. P. (1972). U.S. patent 3,706,729, Dec. 19. Berger, J. (1973). U.S. patent 3,723,254, March 27. Berger, j., and Batcho, A. D. (1978).I . Chromatogr. Library 15 (Antibiotics isol., sep., and purif.), 10158. Berger, J., and Mamsich, W. L. (1972). U.S.patent 3,662,063, May 9. Berger, J., Schocher, A. J., Batcho, A. D., Pecherer, B., Keller, O., Maricq, J., Karr, A. E., Vaterlaus, B. P., Furlenmeier, A., and Speigelberg, H. (1965). Antimimob. Agents Chemother.. 778-785.

134

JOHN C. GODFREY

Bristol-Banyu Research Institute, Ltd. (1964). Belgian patent 641,209, June 12. Bristol-Myers Co. (1969) French patent 1,583,964, Dec. 12. Castora, F. J., and Simpson, M. V. (1979a). Fed. Proc.,Fed. Am. SOC. Exp. Biol. 38(3), Part 1 779. Castora, F. J., and Simpson, M. V. (1979h).J. B i d . Chem. 254(22), 11193-11195. Chao, L. (1978). Nature (London) 271,385386. Claridge, C. A., and Gourevitch, A. (1968) U.S. patent 3,403,078, Sept. 24. Claridge, C. A., Rossamano, V. Z., Buono, N. S., Gourevitch, A., and Lein, J. (1966). Appl. Microbiol. 14 (2), 280-283. Cleeland, R., Beskid, G., and Grunberg, E. (1970). Znfect.Zmmun. 2(4) 371375. Collins, A., and Johnson, R. (1979). Nuckic Acids Res. 7 (5), 1311-19. Cozzarelli, N. R. (1977). Ann Reo. Biochern. 46, 641-668. Cron, M. J., Ragan, E. A., and Hooper, 1. R. (1969) U.S. patent 3,547,903, Dec. 15. Cron, M. J., Godfrey, J. C., Hooper, I. R., Keil, J. G., Nettleton, D. E., Price, K. E., and Schmitz, H. (1970). Prog. Antimicrob. Anticancer Chemother., Proc. Znt. Congr. Chemother., 6th, 1969, Vol. 2, pp. 1069-1082. Devine, L. F., and Hagerman, C. R. (1970b)Appl. Microbiol. 19(2), 329334. Devine, L. F., Johnson, D. P., Hagerman, C. R., Pierce, W. E., Rhode, S . L., 111, and Peckinpaugh, R. 0.(1970a). Am. J. Med. Sci. 260, 165-170. Dolak, L. (1973).J. Antibiot. 26(3), 121-125. Drlica, K., and Snyder, M. (1978). Appl. Microbiol. 120(2), 145-154. Duma, R. J., and Warner, J. F. (1969).Appl. Microbiol. 18(3), 404-405. Falco, S . C., Zivin, R., and Rothman-Denes, L. B. (1978). Proc. Natl. Acad. Sci. U.S.A. 75(7), 32203224. Fedorko, J . , Katz, S., and Allnoch, H. (1969). Appl. Microbiol. 18(5), 869-873. Furlenmeier, A . , Schocher, A. J., Spiegelherg, H., Vaterlaus, B. P., Batcho, A. D., Berger, J., Keller, O., and Pecherer, B. (1969). Swiss patent 467,799, March 14. Gellert, M., Mizuuchi, K., O’Dea, M. H., and Nash, H. (1976a). Proc. Natl. Acad. Sci. U.S.A. 73(11), 38723876, Gellert, M., O’Dea, M. H., Itoh, T., andTomizawa, J. (1976h). Proc. Natl. Acad. Sci. U.S.A. 73(12), 447447 8. Gellert, M., Mizuuchi, K., O’Dea, M . H., Itoh, T., andTomizawa, J. (1977). Proc. Natl. Acad. Sci. U.S.A. 74(11), 47724776. Godfrey, J. C., and Price, K. E. (1972). Ado. Appl. Microbiol. 15, 231-296. Godfrey, J . C., and Price, K. E. (1977). “Structure-Activity Relationships among the Semisynthetic Antibiotics.” (D. Perlman, ed.), pp. 653-717. Academic Press, New York. Grigg, G. W. (1978). Mutat. Res. 47 161-181. Grunberg, E., and Bennett, M . (1965). Antimicrob. Agents Chemother., 786-788. Grunberg, E., Cleeland, R., and Titsworth, E. (1966). Antirnicrob. Agents Chemother., 397398. Hansen, F. G., and von Meyenburg, K. (1979). Mol. Gen. Genet. 175, 135-144. Hays, J. B. and Bohma, S . (1978a) Proc. Natl. Acad. Sci. U.S.A. 75(9), 4125-4129. Hays, J., and Bohma, S . (1978h). J. Supranwl. Struct.. 7(Suppl. 2.). 54. Higgins, N. P., Kreuzer, K. N., Morrison, A., Sugino, A., Peebles, C. L., Brown, P., and Cozzarelli, N. R. (1978).J. Suprumol. Struct.. 7(Suppl. 2) 54. Hoeprich, P. D. (1967). Antimicrob. Agents Chemother., 697-704. Hoeprich, P. D. (1971).J. Infect. Dis. 123(2), 125-133. F. Hoffmann-LaRoche and Co. (1965) Belgian patent 665,237, Dec. 19. F. Hoffman-LaRoche.and Co. (1966) Netherlands Application 6,507,401, Dec. 12. F. Hoffman-LaRocheand Co. (1967)Coumermycin compositions and processes for the manufacture thereof. SA 6615644. Feb. 27.

COUMERMYCINS IN THE

1970s

135

F. Hoffman-LaRoche and Co. (1972). British patent 1,267,282, March 15. Issaq, H. J., Ban-, E. W. Wei, T., Meyers, C., and Aszalos, A . (1977). J. Chromatog. 133(2), 291301. Itoh, T., and Tomizawa, J. (1977). Nature (London)270, 7840. Kaplan, S. A. (1970). J. Pham. Sci. 59(3), 309-313. Karr, A. E. (1969) German Offenlegungsschrift 1,905,328, Sept. 11. Kawaguchi, H., Okanishi, M., and Miyaki, T. (1965a) U.S. patent 3,201,386, Aug. 17. Kawaguchi, H., Tsukiura, H., Okanishi, M., Miyaki, T., Ohmori, T., Fujisawa, K., and Koshiyama, H. (1965b). J. Antibiot. Ser. A 18(1) 1-10. Kawaguchi, H., Naito, T., and Tsukiura, H. (1965~).J. Antibiot. Ser. A 18(5) 11-25. Kawaguchi, H., Miyaki, T., and Tsukiura, H. (1965d)J. Antibiot. Ser.A 18(5), 220-222. Keil, J . G., and Hooper, I. R. (1969a). U.S. patent 3,428,623, Feb. 18. Keil, J. G., and Hooper, I. R. (196913). U.S. patent 3,454,548, July 8. Keil, J. G., and Hooper, I. R. (1969~).German Offenlegungsschrift 1,921,012, Nov. 13. Keil, J. G. Hooper, I. R., Cron, M. J., Schmitz, H., Nettleton, D. E., and Godfrey, J. C. (1968a). Antimicrob. Agents Chemother., 120-127. Keil, J. G . , Hooper, I. R., Cron, M. J., O’Herron, F. A., Ragan, E. A., Rousche, M. A., Schmitz, H., Schreiber, R. H., and Godfrey, J. C. (1968b). I. Antibfot. 21(9), 551556. Keil, J. G., Hooper, I. R., Schreiber, R. H., Swanson, C. L., and Godfrey, J. C. (1969). Antimicrob. Agents Chemother., 200-208. Keil, J. G., Godfrey, J. C., Cron, M. J., Hooper, I. R., Nettleton, D. E., Price, K. E., and Schmitz, H. (1971). Pure Appl. Chem. 28(4), 571-601. Kominek, L. A., and Sebek, 0. K. (1974). Deu. Znd. Microbiol. 15 60-69. Liu, L. F., and Wang, J. C . (1978). Proc. Natl. Acad. Sci. U . S . A . 75(5), 2098-2102. Maehr, H., Scannell, J. P., and Zeitz, V. (1969)German Offenlegungsschrift 1,912,016, Oct. 2. Maehr, H., Scannell, J. P., and Zeitz, V. (1970) U.S. patent 3,530,114, Sept. 22. McCarthy, D. (1979). J. Mol. Biol. 127, 265-83. Michaeli, D., Meyers, B., and Weinstein, L. (1969a), J. Infect. Dis. 120(4), 488490. Michaeli, D., Meyers, B. T., and Weinstein, L. (196913). Antimicrob. Agents Chemother., 463467. Michaeli, D., Molavi. A. Mirelman, D., Hanoch, A., and Weinstein, L. (1970). Antimimob. Agents Chemother., 95-99. Mizuuchi, K., O’Dea, M. H., and Gellert, M. (1978). Proc. Natl. Acad. Sci. U.S.A. 7!5(12) 59605963. Nettleton, D. E. (1968). U.S. patent 3,380,994, April 30. Newmark, H. L. (1970a) German Offenlegungsschrift 1,917,859, Jan. 15. Newmark, H. L. (1970b) U.S. patent 3,519,712, July 7. Newmark, H. L., and Berger, J., (1970). J. Phann. Sci., 59(9) 1246-1248. Newmark, H. L., Berger, J., and Carstensen, T. (1970). J. Pham. Sci. 59(9), 1249-1251. Ninet, L., Benazet, F., Charpentie, Y.,Dubost, M., Florent, J . , Mancy, D., PreudHomme, J., Threlfall, T. L., Vuillemin, B., et al. (1972). C . R. Hebd. Seances Acad. Sci. S e r . C , 275(8), 455458. Price, K. E., Chisholm, D. R., Leitner, F., and Misiek, M. (1969). Antimicrob. Agents Chemother., 209-218. Price, K. E., Chisholm, D. R., Godfrey, J. C., Misiek, M., and Gourevitch, A. (1970). Appl. Microbol. 19(1), 14-26. Ryan, M. J. (1976a). Diss. Abstr. Znt. B . 37(1). 203-204. Ryan, M. J. (1976b). Biochemistry 15(17), 37693782. Scannell, J. (1967). Antimimob. Agents Chemother. 470474. Scannell, J. P., and Kong, Y. L. (1969). Antimicrob. Agents Chemother., 139-143. Schmitz, H., and DeVault, R. L. (1970). U.S. patent 3,547,902, Dec. 15.

136

JOHN C. GODFREY

Schmitz, H., and Godfrey, J . C. (1970). J . Antibiot. 23(10), 497501. Schmitz, H., DeVault, R. L., McDonnell, C. D., and Godfrey, J. C. (1968).J . Antibiot. 21(10), 603-610. Smith, C . L., Kubo, M., and Imamoto, F. (1978). Nature (Landon) 275, 420-423. Societe des Usines Chimiques Rhone-Poulenc. (1970). French Medical 7,884, May 4. (Chem. Abst. 76, 149607e). Staudenbauer, W. L. (1975). /. Mol. Biol. 96, 201-205. Tamaki, S., Sato, T., Sugino, Y.,Matsuzawa, H., and Matsuhashi, M. (1970). Prog. Antimimob. Anticancer Chenwther., Proc. Znt. Congr. Chempther. 6th 1969 Vol. 1, 405409. Taylor, D. E., and Levine, J. G. (1979). Mol. Gen. Genet. 174, 127-133. Umezawa, H., Hamada, M., Takita, T., and Naganawa, H. (1971) Japanese patent 71 15, 616, April 27. Whitaker, W. D. (1968a) British patent 1,111,511, May 1. Whitaker, W. D. (1968b) British patent 1,114,470, May 22. Whitaker, W. D. (1968c) British patent 1,114,469, May 22. Whitaker, W. D. (1968d) British patent 1,114,468, May 22. Wick, A. E., Blount, J. F., and Leimgruber, W. (1976). Tetrahedron, 32(17), 2057-2065. Yang, H. L., Heller, K., Gelled, M., andzubay, G. (1979). Proc. Natl. Acad. Sci. U.S.A. 76(7) 330448.

Instrumentation for Process Control in Cell Culture ROBERTJ. FLEISCHAKER, JAMES c. WEAVER,*

AND A”l’H0NY

J. SINSKEY

Department of Nutrition and Food Science, Massachusetts Znstitute of Technology, Cambridge, Massachusetts, and *Haruard-MZT Division of Health Sciences and Technology, Cambridge, Massachusetts

I. Introduction . . . . . . . . . . . . . . . . . . . .

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

11. Comparison between Animal and Microbial Cells . . . . . 111. Microcarriers . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

IV. Generalized Model of Cell Growth

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

B. Cell-Cell Interactions . . . . . . . . . . . . . . . . . . . . V. Evaluation of Instrumental Methods . . . . . . . . . . . .

............ ............ VIII. p H . . . . . . . . . . . . . . . Acids, Salts, and Other Nutrient Components.. XI. Oxygen ..................................... X I . Carbon Dioxide . . ..........

XV. ATP

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

XVII. Conclusion References

...........

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

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

137 138 139 140 141 141 142 142 144 144 145 146 148 149 152 153 156 159 162 163 163 165 165

I. Introduction The increasing use of instrumentation in the fermentation industry has led to significant advances in the understanding of microbial processes and their optimization (Wang et al., 1979; Ryu and Humphrey, 1973). In contrast, the use of similar instrumentation for the study and control of cell culture has been rare, even though cell culture is used increasingly for the production of vaccines and cell products. The adaptation and development of better instrumentation for environmental control and monitoring of cell function can contribute to the improvement of cell yields and productivity. This article will review the instruments and measurement techniques used by the 137 ADVANCES IN APPLIED MICROBIOLOGY. VOLUME $7 Copyright @ 1881 by Academic Reu. Inc. Au rights of reproduction in any form reserved. ISBN 0-1edoese7-9

138

ROBERT J. FLEISCHAKER ET AL.

fermentation industry, and assess the potential of these and emerging techniques for use in cell culture. An important first step in the development of an instrument system is the selection of the quantities to be measured. The choice will generally depend upon (1) the overall aim of the project and (2) the nature of the biological system that is utilized. In this article we will be concerned primarily with determining cell mass and, to a lesser extent, with metabolic activity. The system to be considered is the growth of anchorage-dependent animal cells [e.g., WI-38 cells, a normal human diploid fibroblast cell strain derived from fetal lung tissue (Slonim, 1974)] in microcarrier culture. Instrumental methods can be classified into two general categories: offline and on-line. Off-line methods involve removal of discrete samples for subsequent treatment and analysis. Typically, off-line methods, such as wet chemical or immunological methods, involve more processing than on-line methods. Although off-line methods are becoming more automated and rapid, they are still relatively slow and cumbersome by comparison to online methods. In contrast, on-line methods involve measurements performed directly on or within the biological system. These measurements are made either without any sample removal or utilizing a continuous (generally small) sample withdrawal. Owing to the real time nature of on-line methods, direct interfacing to a computer is usually possible. This review is concerned primarily with on-line methods of analysis, in view of the favorable outlook for continued improvement in cost and performance of inexpensive microprocessors and computers.

II. Comparison between Animal and Microbial Cells Although animal and microbial cells are subject to the same hndamental laws, there are a number of essential differences between them, as summarized in Table I. From the viewpoint of measurement, the most significant differences are the lower metabolic activities per cell mass of animal cells and their lower cell densities (seldom more than 1 gmlliter of dry cell weight). As a consequence, during a given period of time, the effects that animal cells have upon their environment or the demands they make on it are considerably less than those of microbial cells, so that there is comparatively less to measure. In addition, growth of animal cells is restricted to a narrow range of environmental conditions, and strict control of these conditions greatly enhances growth. For example, Ceccarini and Eagle (1971)showed that diploid fibroblasts at controlled pH showed significantly better growth than under variable pH conditions (COP-bicarbonatebuffer system). Similarly, Kilburn and Webb (1968)showed that if dissolved oxygen was maintained within the range of 40-100 mm Hg, cell growth and peak cell densities were maximal.

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

COMPARISON OF

TABLE I MICROBIAL AND ANIMALCELLS

Microbial Size (diameter) Metabolic regulation Nutritional spectrum Growth rate Environment

139

1-10 pm Internal Wide range of substrates Doubling time typically 0.5-2 hours Wide range of tolerance

Others

Animal

10-100 p m Internal and hormonal Very fastidious nature Doubling time typically 12-60 hours Narrow range of tolerance Limited population age of normal cells Lack of protective cell wall

In the case of microbial cells, there is considerable flexibility in the composition of the media. For example, Wang et al. (1977) used NHIOH as the sole nitrogen source, the addition of which was regulated by pH control. Consequently, measurements of base addition greatly facilitated the estimation of cell mass. Such flexibility does not exist with cell culture, where the medium consists of a complex mixture of vitamins, minerals, and amino acids supplemented with animal sera. These nutritional requirements are complex and difficult to determine. Only recently has it been possible to grow cells in a completely defined medium (Hayashi and Sato, 1976). Work in Sato’s laboratory has shown that serum serves principally as a source of hormones, although which hormones and in what concentrations appears to vary with cell type. Hence, for a given cell line, it is difficult to change the composition of the medium significantly. In particular, several researchers have noted the instability of glutamine in cultures giving rise to the formation of ammonia (Higuchi, 1970; Tritsch and Moore, 1962; GriEths and Pirt, 1967), thus requiring the addition of supplemental glutamine. In this case, both the loss of glutamine and the formation of ammonia are undesirable, yet efforts toward replacement of glutamine with glutamic acid have had only mixed results.

I II. Microcarriers The cultivation of anchorage-dependent mammalian cells is one of the main problems in the large-scale production of cellular products and virus vaccines. A primary difficulty is one of providing large, accessible surfaces for cell growth. A number of techniques have been proposed as means of

140

ROBERT J. FLEISCHAKER ET AL.

solving this problem, including the roller bottle (Ubertini et al., 1960), multiplate propagators (Schleicher and Weiss, 1968), spiral plastic film bottles (House et al., 1972), artificial capillary propagators (Knazek et al., 1972), and membrane tubing reels (Jensen, 1977). In general, however, these systems suffer &om the following shortcomings: 1. limited potential for scale up, 2. difficulty of taking cell samples, 3. limited potential for monitoring the system, and 4. difficulty of maintaining homogeneous environmental conditions throughout the culture. In an effort to overcome these limitations, van Wezel (1967, 1973, 1977) and van Wezel et al. (1978) examined the use of the commercial ionexchange resin, DEAE-Sephadex A50 for large-scale cultivation of anchorage-dependent cells. When suspended in culture media by gentle stirring, the Sephadex beads provided a large surface for attachment and growth by normal diploid cells. However, a toxicity was observed at bead concentrations exceeding about 1 gm of DEAE-Sephadex per liter, as indicated by increased inoculum loss and diminished capacity for cell growth. Complete death of the inoculum was observed at bead concentrations exceeding 5 gm/liter. Our laboratories found that this toxicity could be largely eliminated by reducing the amount of exchange capacity of Sephadex beads (Levine et al., 1977a,b; Levine, 1977). Using beads prepared with an exchange capacity of 2.0 meq/gm dextran (approximately 40% of the exchange capacity of the commercial preparation), normal human embryonic lung fibroblasts (HEL 299) showed good growth and proliferation at Sephadex concentrations up to 5 gm/liter dextran. The potential of microcarriers for scale-up is illustrated by the fact that 1 liter of culture with a DEAE-Sephadex concentration of 5 gmfliter has enough available surface area (27,000 cm2)to replace approximately 55 roller bottles (490cm2 each). Experiments by Giard et al. (1977, 1979) indicate that cells grown on beads are normal with respect to product formation. Their work shows that microcarrier-grown cells can produce viruses and peptide products (interferon) at per cell yields typical of cells grown by more conventional techniques.

IV. Generalized Model of Cell Growth Most of the measurements made using on-line sensors provide indirect estimates of cell growth and metabolism. As cells grow and divide, they

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

141

produce changes in their liquid environment, which can be measured and used to monitor cell processes. Thus, for example, measurement of the rate of oxygen utilization provides a direct indication of metabolic activity, which can be used to estimate cell mass. In order to devise a system for measuring cell growth, an understanding of the underlying processes of growth is essential.

A. CELL-ENVIRONMENT INTERACTIONS In the course of growth and metabolism, the cells consume nutrients from the medium and secrete back into it a number of metabolic by-products, as shown in Fig. 1. Animal cells are very sensitive to the concentration of these nutrients, since they have very “leaky” membranes. Hence, if the nutrients in the medium fall below a certain concentration, nutrients will diffuse outward from the cells. This is of particular importance with respect to certain growth factors produced by the cells themselves. As a direct consequence, the cloning of animal cells is typically very inefficient and, in addition, losses in cell inocula incurred during passaging increase with lower cell densities.

B. CELL-CELLINTERACTIONS Different types of animal cells are capable of influencing one another by hormonal secretions. However, since cell culture usually deals with growth of a single cell type, this interaction is not often encountered. On the other

cow RG.1. Generalized model of animal cell metabolism.

142

ROBERT J. FLEISCHAKER ET AL.

hand, density-dependent interactions (which were previously termed “contact inhibition”)are frequently observed; that is, for a given medium composition, cell growth stops at a given surface coverage. This phenomenon has been shown by Folkman and Moscona (1978) to reflect the fact that DNA synthesis is strongly dependent on the curvature of the cell membrane; i.e., as the cell becomes more spherical, DNA synthesis declines. This interaction has been viewed by some as an explanation for the anchoragedependency of these cells. It should be noted that this phenomenon puts a limit on the minimum diameter acceptable for microcarriers. Marcoudas (1972) determined this limit to be about 55 p m for glass beads using baby hamster kidney cells (BHK21 C13).

V. Evaluation of Instrumental Methods In order to establish a basis for evaluating and comparing the performance of instrumental techniques, a number of terms need to be introduced and defined. Most important are the concepts of response time, gain, specificity, signal-to-noise ratio, and drift. A. RESPONSETIMEAND GAIN The fundamental process is the translation of some microscopic (or physiological) event into an electric signal, which is then recorded by either a computer or a stripchart recorder. Such a translation is called a transduction, and the associated devices are referred to as transducers. An example of a transducer is the pH electrode, which converts hydrogen ion activity (the microscopic event) into an electric potential that is subsequently recorded. Similarly, the Clark oxygen electrode produces an electric current proportional to the oxygen partial pressure in the surrounding fluid. The exact relationship between the microscopic activity and the electric signal can be very complicated, but for our purposes the transduction can be adequately described in a “black box” fashion, as shown in Fig. 2A. The ability to describe the transfer function (G) mathematically is very valuable, particularly from the viewpoint of process control (D’azzo and Houpis, 1975). This is because many process controls involve feedback loops, the design of which requires a knowledge of the transfer functions, in order to ensure stable control operations. Transfer functions can be characterized by the concepts of response time and gain, which can be evaluated by following the system response to a step change (e.g., Cobbold, 1974). Given a step change at the microscopic level (as shown in Fig. 2B), the instrument will show a change (typically slower than the input function) from some value eoto a value el, where eoand e l are

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

143

A input

transfer function

physiological event

f* 0

E

output

electric signal

I l ~ q / I

soYde,-4)

-t, to

4

time

*

FIG.2. (A) Transducer represented in “black box” form. The relationship between the output signal of the transducer and events occurring at the physiological level are described by the transfer function G . (B) Typical transducer output response to a step change in the input.

steady-state values. The response time ( t J is defined as the time required for the output signal to change from the initial value to 90% of the final value, following the step change at the microscopic level. The reason that the response time is defined in terms of the 90% value is that small errors in the steady-state value lead only to small errors in evaluating the time required to reach the 90% value, whereas these same errors can lead to significant errors in estimating the time required to reach the steady state values. The gain is simply e (output) divided by x (input) and may or may not be constant over the range in question. A plot of gain vs x represents the calibration curve for the transducer. The term “sensitivity,” which is closely related to gain, refers to change in e divided by change in x . The acceptable values for these measurements will depend upon the particular system in question. In the case of animal cells grown in culture, the doubling time is about 24 hours, with a typical experiment lasting 150hours. Thus, a response time of 30 minutes or less would be adequate to resolve the doubling time to within 5%. Most sensors by themselves have response

144

ROBERT J. FLEISCHAKER ET AL.

times on the order of seconds to minutes, but the overall response time of the system can be considerably longer. Delays can be caused by flow through tubing and the time required for homogeneous mixing in a large vessel. Additionally, if the information needed is a rate of change, 10-30 minutes may be required with animal cell systems to obtain a sufficiently accurate measurement, even though the response time of the transducer is in the order of seconds.

B. SIGNAL-TO-NOISE RATIO Another concept to be defined in relation to the transducer is the signalto-noise (S/N) ratio. In the example shown (see Fig. 2B), the signal is measured as the difference, el - eo, and the noise is measured as the root mean square (rms) of the fluctuations about the time average of signal el. Signal noise can generally be segregated into low- and high-frequency components. The high-frequency component consists of those fluctuations with frequencies higher than l/tr, and the low-frequency component consists of fluctuations with frequencies lower than l/tp The high-frequency signal noise often causes the most serious limitation to the overall accuracy of measurement and is the dominant component of the noise in the S/N ratio. The detection limit of the instrument is the smallest quantity that can be measured above the system noise at the physiological level. Typically, one wants to operate with an S/N ratio of 10 or greater.

C. DRIFT The low-frequency portion of signal noise contains components referred to as signal drift, the magnitude of which determines how frequently an instrument must be recalibrated. For many instruments, such as an external gas analyzer, recalibration does not represent a serious problem. However, for a number of other instruments, such as dissolved oxygen probes, recalibration may be difficult or impractical, and it is important to determine beforehand whether such drift is going to cause serious problems over the course of the experiment. In some instances, the use of an internal standard, differential measurement, or a pulsed addition will satisfactorily recalibrate a transducer in situ. In summary, a transducer that converts a biological activity into an electric signal can be characterized by its response time, specscity, gain or sensitivity, S/N ratio, ease of recalibration, and the frequency with which it must be recalibrated. Two other important considerations are sterility (that is, whether the instrument can be sterilized if it comes in contact with the culture broth) and cost, which will not be discussed in this article.

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

145

VI. Temperature Monitoring and control of temperature are important for a number of reasons. First, animal cells can grow only in a narrow temperature range, and second, since the transfer functions of many sensors are strongly dependent upon temperature, errors can result if the temperature is neither controlled nor corrected for. Hence, the ability to maintain the temperature of thP culture within a narrow range is essential for cell growth and for reducing measurement errors. In general, biological systems are not efficient in converting their substrates into cell mass. It is estimated that Escherichia coli is capable of utilizing only about 50% of the available free energy in the conversion of glucose to cell mass, with the other 50% being liberated as heat (Wang et al., 1976). Animal cells produce even less cell mass in relation to heat generated, owing to their proportionately higher maintenance requirements. Hence, calorimetry might be used to monitor cell growth. There are essentially three methods for measuring heat produced: measurement of cooling requirements, flow calorimetry (Eriksson and Holme, 1973), and dynamic calorimetry (Cooney et al., 1969; Mou and Cooney, 1976). The measurement of cooling requirements is easily adapted to very large fermentors. However, in laboratory fermentors, other methods need to be used, in view of the volumes and increased relative heat losses (Swartz and Cooney, 1978). Flow calorimetry involves the continual removal into an insulated chamber of a stream of culture from the fermentation vessel and the measurement of the temperature rise in a small flow cell (Fig. 3). The temperature rise, AT, depends upon the cell mass concentration and the residence time of the flow cell. The dynamic technique involves periodic disconnection of the temperature control and measurement with a sensitive thermistor of inwlate d block

culture V I M d

/2/

coils

residence time, T

temperature, TI

a temperature, Tz

FIG. 3. Assembly for flow calorimetry.

146

ROBERT J. FLEISCHAKER ET AL.

the heat generated in the fermentor. The heat accumulated in the fermentor is calculated as Qacc = K T, where Qacc is the rate of accumulation, K is a lumped parameter representing the heat capacity of the system, and T is the rate of change of temperature. The heat released during growth (QH) can and the heat lost then be determined by evaluating the heat of agitation (Q=) to the surroundings (Qsurr),so that Qn =

Qacc

- Q ae +

Qsum

The heat losses due to evaporation and differences of humidity in the air blown into the fermentor can be neglected, if the air is saturated with water at the operating temperature of the fermentor. Both flow calorimetry and dynamic calorimetry have distinct advantages and disadvantages. A flow calorimeter can be used to measure heat evolution independent of culture volume; contributions due to stirring; the addition of gas, alkali, or nutrients; and inadequate insulation of the fermentor, whereas these parameters limit the accuracy of the dynamic technique. However, flow calorimetry is believed to be inadequate for use with filamentous organisms or non-Newtonian fluids; organisms that tend to form wall growths; and cultures above a moderate cell density, owing to the potential for depleting oxygen or other limiting nutrients within the flow cell. To evaluate the potential of these thermal methods, an estimation of the heat produced in culture by animal cells is required. Given a culture of 10' WI-38 cells/ml (achievable with microcarriers), with an expected oxygen consumption of 0.15 nmole Odliter-hour (Mclimans et al., 1962), and heat conversion of 0.11 kcaVmmole O 2 (Cooney et al., 1969), then a heat production of about 0.017 kcaliter-hour is expected. In our laboratory, in a wellinsulated 10-liter fermentor (1-in.-thick aircraft Fiberglas with aluminum foil on the outside), temperature decreased at a rate of 1.l"C/hour when it was agitated at 60 rpm at 37°C with no aeration. The rate of heat loss was five times greater without the insulation. A flow microcalorimeter supplying such resolution and sensitivity would encounter two serious problems: that of pumping a bead suspension to the flow cell and that of beads settling in the flow cell in the absence of stirring. Hence, it appears that a special adaptation of calorimetry is needed to measure cell growth in microcarrier culture. However, for suspension culture the flow microcalorimeter should be directly applicable, if the problem of wall growth in the flow cell can be eliminated.

VII. Mixing and Viscosity

The interactions involved in the rheology of a culture fluid can be complex (Charles, 1978). Adequate mixing is required to ensure uniform environ-

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

147

mental conditions and to prevent beads from settling. Since the cells are attached to a surface, the diffusional layers controlling the maximal rates of nutrient uptake will be determined by the velocity of the liquid relative to the bead. However, the effect of a high-velocity gradient (shear stress) is not well understood. Augenstein et al. (1971) have also shown that animal cells are relatively tolerant of momentary exposure to relatively high shear stress. However, the long-term effect of viscous shear forces on cell growth is currently not well understood. Viscosity changes in culture media have long been advocated as a possible means of measuring cell growth. However, this approach has not yet been fully developed, as measurement of viscosity in a non-Newtonian fluid is complicated, and to date there is not enough research in the area. The technical aspects of the rheological properties of microbial cultures have been extensively reviewed by Charles (1978). Perley et al. (1979) attempted to measure yeast concentration by means of a continuous flow viscometer, which has the advantages of being easy to operate and inexpensive. In principle, viscosity is proportional to the pressure drop for a constant flow across a capillary tube. However, owing to changing morphology of the yeast and other problems, the measurement of viscosity was found to be of little value, except at high cell concentrations. The flow viscometer is probably not well suited for microcarrier culture on account of the problem of bead settling in the capillary tube. However, the method may be well suited for monitoring growth in suspension cultures. Alternatively, viscosity can be determined from measurements of the power requirements of the impeller system in the fermentor. In the laminar flow region for an ungased, Newtonian fluid, the power number will be inversely proportional to the Reynolds number. Hence, given a constant geometry, the power uptake is related to the viscosity (p)and the impeller speed (n)as

P

a nzp

True power measurement can be determined by means of a strain gauge on the impeller shaft. The microcarrier system consists of cells growing on spherical beads with the system being nongased (in any case, periods of nongasing pose no real problems) and the broth viscosity expected to be Newtonian. As the cells grow on the beads, their effective diameter increases and viscosity is expected to rise. Given an original effective bead diameter of 168 p m , as the cells grow to confluence over this surface, bead diameter can be expected to increase by approximately 10 pm. Using the Einstein equation for the modeled viscosity of suspended spheres (Einstein, 1906), where pc is the viscosity of cell suspension, pRis the viscosity of the supernatant, and C is

148

ROBERT J. FLEISCHAKER ET AL.

the volume fraction of the suspended particles. The ratio p J p R is expected to change from 1.22 to 1.27 for cells confluent on beads at a concentration of 5 gm of beadsfliter. Alternatively, using the correlation determined by Perley et al. (1979) to relate yeast concentration to viscosity, a change of 1.25 to 1.33 would be expected in p J p R . In principle, cell growth on microcarriers should alter the viscosity, but only very slightly (about 4%), and sensitive measurements of that change may prove difficult. Additionally, two other methods might in principle be used to monitor.cell growth on the basis of a changing effective bead diameter; these are settling time and scattered light. The latter technique is not particularly well suited for use with particle sizes greater than 10 pm in diameter. The most serious limitation to the use of any of the three techniques-viscosity, scattered light, and settling time-is that the distribution of bead sizes (168 2 15 pm) is larger than the change expected to result from the growth of a layer of cells about the beads (approximately 10 pm). The measurement suffers from a low signal-to-noise ratio and the accuracy with which one is able to predict growth is verv poor.

VIII. pH pH is an important parameter to control, since many biochemical processes are sensitive functions of the ionization state of the molecular species concerned. Such processes include the rate of enzyme-catalyzed reactions, protein conformation, ion-binding properties of proteins, and others. pH is commonly measured with a combined glass-reference electrode. Steam sterilizable electrodes are commercially available, and, with proper electronics and temperature control, the pH can be measured to an accuracy of 0.01 pH units. Work by Ceccarini and Eagle (1971) indicates that the absolute value of the pH is less critical than the ability to maintain a constant value during the growth of normal cells. By controlling pH, a cell density twice the usual value can be achieved. Although the individual pH optima vary with cell type, most of the normal fibroblast cells grow fairly well at 7.4. Work by Fodge and Rubin (1975) indicates that 90% of the change in pH is due to the production of CO and lactate, which is interesting since, if the pH is kept constant, the amount of base added will give a direct and fairly good measurement of the amount of acid formed. In the study of the growth of chicken embryo fibroblasts (CEFs) in our laboratory, we found that the amount of glucose consumed is proportional to the amount of lactic acid

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

149

produced and correlates well with resultant cell numbers. Hence, a measurement of base added can give a reasonable estimation of the cell mass, and an estimation of the growth rate of the cells can be made by calculating the rate of base addition. Measurements of the CO and glucose concentrations would further refine these estimates. Our studies found CEFs to produce up to 10 m M lactic acid, which means that for a total volume of 5 liters, approximately 200 ml of 0.05 M sodium hydroxide would be required over the course of the experiment to control pH. Base addition could best be monitored by means of a load cell to measure the weight of the contents of the bottle from which base is added. Load cells have two basic designs: position-spring transducers and strain gauges. The position-spring transducers are less expensive but also less accurate. By means of a spring, the weight of an object is translated into a vertical displacement, which is measured by the position transducer. The nonlinearity and hysteresis of the spring are the major limitations on the accuracy of the device, approximately 2% with extensive calibration (Swartz and Cooney, 1978). Strain gauges give results that are reproducible to 0.01%, display nonlinearities of 0.03%, and are essentially insensitive to changes in the ambient temperature (technical information from BLH Electronics, Inc., Waltham, Massachusetts). In summary, maintenance of a constant pH is beneficial for cell growth. In addition, growth and growth rates can be estimated by quantitating the base addition used to control pH over the course of the experiment. The measurement of base addition is simple, direct, and quite accurate.

IX. Ionic Strength Animal cells are also sensitive to the ionic strength of their environment. However, the changes in ionic strength occurring over the normal course of growth are not large enough to be physiologically significant. Such changes are the result of the removal of charged species from the medium or the conversion by the organism of noncharged substances to charged compounds, such as glucose to lactate. These small changes can be directly measured electrically and provide a basis for a continuous measurement of the accumulation of metabolites. Hence, a number of investigators (Hardy et al., 1977; Kagen et al., 1977; Zafari and Martin, 1977; Richards et al., 1978; Cady, 1975; Ur and Brown, 1975) have proposed using electrical conductance measurements to monitor cell metabolism. This is a relatively simple procedure, which can be rapid, sensitive, and widely applied. A conductivity measurement is usually performed using alternating current conditions of 1-10 kHz rather than direct current and utilizing two platinum electrodes in contact with the broth (Fig. 4) The .impedance of a

150

ROBERT J. FLEISCHAKER ET A L

A

platinum electrodes

0.11

I

I

I

I

20

40

60

80

Hours

FIG.4. (A) Conductivity cell. (B) Series electrical equivalent of conductivity cell. (C) Growth of a transformed human fibroblast (SVSO) and change in conductance (volts).

measurement cell containing bacteria culture or tissue culture medium can

be crudely modeled as either a series or a parallel resistor-apacitor circuit (Fig. 4B). The capacitance is believed to arise at the liquid electrode interface (Richards et al., 1978), whereas the resistance is due mainly to the ionic

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

151

content of the liquid. In determinations of conductivity, certain precautions need to be observed. The conductance should be measured in an AC circuit to minimize electrode polarization (Schwan, 1968; Geddes, 1972; Cobbold, 1974). Furthermore, the voltage drop across the electrodes should be kept below 1V to minimize electrolysis of the liquid, and the temperature should be carefully controlled, since conductivity is a significant function of temperature (about 2%PC). By way of comparison, a 1% change in conductance is expected to occur over the entire culturation due to the growth of the organisms. Richards et al. (1978) argue that the resistive component of impedance provides a more appropriate measurement of the bacterial growth than total impedance, since this component is a measure of the ionic content of the medium. They were also able to quantitate the resistive component of total impedance, using sophisticated electronics that would lock onto that portion of the output signal in phase with the input signal. In cell culture, an estimate of the expected change in conductance can be made if it is assumed that changes in the ionic content of the medium result only from production of lactic acid and ammonia (assumed to have final concentrations of20 and 3 mM, respectively). Conductance ( L )is defined for a slab of conducting material to be

L = Ufl' where A is surface area of the slab (cm'), 2 ' is the thickness of the slab (cm), and K is a parameter dependent on the ion species involved and the concentration (ohm-' cm-I). K is defined as

K = C(h+ + A-) where C is concentration (equivalentdml), and h+ and h- are constants dependent on the species involved. The A+ A- for NH40H is 274 cm2 equiv-' ohm-' (Castellan, 1964), and for sodium lactate, about 92 cm2 equiv-' ohm-', assuming lactate to behave approximately as acetate. Given 1' to be 7 cm and A to be 0.18 cm2, the total conductance change would be 1 x ohm-'; therefore, a 1.3%change in conductance is expected. In principle, conductance measurement can be used to determine inoculum size, lag time, and population-doubling times. The method is conceptually simple but requires moderately sophisticated electronics and a very well-controlled temperature environment. Animal cells in culture appear to produce charged compounds in amounts sufficient to be detected by the instrumentation described for work with bacteria. Preliminary results in our laboratory have shown the changes in conductance of the medium to be a good indicator of cell growth. Figure 4C shows the growth of SV8Os (a transformed human fibroblast) and the change in conductance as measured

+

152

ROBERT J, FLEISCHAKER ET AL.

in volts on our instruments. A primary limitation of this method will be the stability of the medium. The decomposition of the medium (glutamine to ammonia, for example) could produce significant background drift, requiring a differential measurement. A major criticism of the technique is that it is not certain what chemical species is being measured, since conductance i s affected by many factors.

X. Dialyzable Components: Carbohydrates, Organic Acids, Amino Acids, Salts, and Other Nutrient Components In order to grow and/or maintain its metabolic activity, an organism must utilize a number of nutrients from the environment. Examples of these nutrients include glucose, amino acids, certain organic acids, minerals, and vitamins. Quantitating the uptake rates of these nutrients provides useful information for monitoring cell growth and metabolism. Glucose is the most commonly examined of these nutrients. A number of enzyme probes have been constructed to measure glucose concentration (Gray et al., 1977; Guilbault, 1976). The major problem with these devices is that they cannot be steam-sterilized. Alternatively, a dialysis membrane probe can be used to remove glucose from the broth to an external glucose analyzer (Zabriskie and Humphrey, 1978a)(Fig. 5). In principle, the liquid flows across one side of a dialysis membrane, and the dialyzable compound in the bulk solution diffuses across the membrane into the liquid stream (Fig. 6). The concentration of each compound leaving the dialysis probe (S) is related to its concentration in the bulk fluid (Sb), the liquid velocity (V), the dimensions of the probe (W, D,L), and the permeability of dialysis Wobe reagents buffers

7

4

overflow

I%.

5. Assembly for dialysis probe.

1

External Analyzer

I1

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

153

FIG. 6. Diffusion of solute across membrane into channel of dialysis probe.

the membrane to the dialyzable compound ( U ) , as given by the following equation: ln(1 - a)= -(UL/D)(l/V),

where a = S/Sb

The response time of the measurement will be related to the liquid velocity, but is typically about 5 minutes. The accuracy and linearity of the measurement will depend largely upon the downstream transducer and the quality of the pumps. To expand upon the usefulness of a measurement of glucose, we have found in our studies with CEFs that the change in cell mass is proportional to glucose consumption. Hence, a knowledge of glucose concentration can provide a basis for estimating cell numbers and growth rates. In addition, a number of control strategies are suggested, such as replacement of the medium as glucose is depleted or keeping the glucose concentration of the medium constant by addition of concentrated glucose. The dialysis probe is a promising development in the area of instrumentation for cell culture. First, it provides a means for on-line use of transducers that cannot be steam-sterilized (such as the enzyme electrodes). Second, because of the transducer’s external location, recalibration and other adjustments are easily made. Finally, the dialysis probe in conjunction with the appropriate transducers has the potential for measuring a wide spectrum of substances, including carbohydrates, organic acids, amino acids, vitamins, ammonia, mineral salts, and a number of growth factors. XI. Oxygen Oxygen is another basic parameter in biological processes. Of particular importance are the effect of oxygen concentration on the growth and metabolism of the cells, and the problem of transporting oxygen from the gas

154

ROBERT J. FLEISCHAKER ET AL.

phase to the cells. Kilburn and Webb (1968)have shown with mouse LS cells that optimal growth occurred when the oxygen partial pressure (P,) was controlled within the range of 40-100 mm Hg. At Poz levels below this range, metabolism became more anaerobic, as indicated by the increased production of organic acids and elevated levels of lactic dehydrogenase. Conversely, an oxygen toxicity was observed at Po2 levels above this range. Mclimans et al. (1962, 1968) have reported on the importance in cell culture of adequate oxygen transfer for the establishment and growth of primary hepatic cells. In very detailed work, the depth of the liquid overlay in monolayer culture was related to overall gas exchange kinetics. In microbiological systems, the oxygen demand for cells growing in submerged culture can generally be provided by sparging the gas at the bottom of the vessel. The spinning impeller breaks up the air into tiny bubbles, which on average are evenly distributed throughout the vessel. The objective is to maximize the interfacial area between the gas and the liquid and to minimize the distance that the oxygen must diffuse from the gas to the cell. A large amount of data has been accumulated, correlating vessel geometry, impeller speed, gas flow, and broth rheology with mass transfer rates (Wang et al., 1979). However, it is not clear that sparging can provide adequate mass transfer for cell growth in microcarrier culture. In the first place, the medillm contains a number of serum components that would lead to serious foaming problems. Moreover, since animal cell membranes are not covered with a rigid cell wall, many of the antifoaming agents used in microbial fermentations cannot be employed. Second, for reasons not well understood, the presence of gas bubbles causes pronounced damage to cells in cell culture (Kilburn and Webb, 1968). Similar damage (platelet losses and hemolysis) by air bubbles in blood oxygenation is well documented (Mortensen, 1977). Although there is some evidence that increasing the serum concentration will minimize this damage (Kilburn and Webb, 1968), this possibility has not been well investigated nor does it promise to be realistic, since the problem of foaming will remain. Surface oxygenation, which has been used in the laboratory, would not appear to be suitable for large-scale work, since the surface area per unit volume needed to support adequate mass transfer is not available. However, membrane oxygenation similar to that used in blood oxygenation (Kolobow et ul., 1975) is able to supply the oxygen demands for large-scale microcarrier culture. Membrane oxygenators function by providing a very large surface area for gas exchange. Quantification of oxygen demand can be used to estimate cell growth and metabolism. The oxygen uptake rate can be measured by means of off-gas analysis or dynamic assessment. The off-gas technique involves measuring

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

155

the gas flow rate, the inlet oxygen concentration, and the outlet concentration. The oxygen uptake rate is equal to the product of the flow rate and the difference of gas concentrations (see Fig. 7). The usefulness of this method depends on both the sensitivity of the oxygen analyzer and the uptake rate itself. For example, the paramagnetic analyzer is capable of measuring a 2% change in the oxygen concentration with 1%accuracy (Snedecor, 1977). For a 5-liter culture of WI-38 cells growing at a density of 106cells/milliliter, the corresponding oxygen demand would be 0.75 mmole Odhour (16.8 ml 0 4 hour). For a 2% change in exit gas concentration, the flow rate through the system would be 840 mlhour. However, the major problem with the paramagnetic analyzer is that it is quite sensitive to the variations in flow for low gas-flow rates. In order to maintain 1% accuracy in the oxygen uptake rate, the flow to the analyzer must be kept constant to within 0.2% (Swartz and Cooney, 1978). Alternatively, the oxygen uptake rate can be assessed using the dynamic

-*-

OUR =

A 10-

0.1

.09

98-

3z,

7-

0 6-

% -T 45 )E

25

3-

- 2f

9

11

1.01

30

60

90 Hours

120

150

180

0

FIG.7. (A) Mass balance of oxygen into and out of a fermentor. (B) Growth of a human diploid fibroblast and the oxygen uptake rates (OUR) as measured by the dynamic technique.

156

ROBERT J. FLEISCHAI(ER ET AL.

technique. In this procedure, the air flow to the culture is stopped temporarily, and the change in dissolved oxygen is measured over time. For lo6 WI-38 cells/milliliter, the dissolved oxygen concentration would be expected to change from 50% to 25% saturation in about 25 minutes. In our laboratory, we have shown an excellent correlation between cells grown in microcarrier culture and the oxygen uptake rate as measured by this technique. Figure 7B shows the growth of a human diploid fibroblast and the oxygen uptake rate as measured by the dynamic technique. The response time for a fermentation quality dissolved oxygen probe is about 45 seconds. These probes are accurate to 0.3% and feature a drift of less than 2Wweek (technical data, Instrumentation Laboratories, Lexington, Massachusetts). Finally, the mass spectrometer (MS) can be used to measure dissolved oxygen and gas phase oxygen simultaneously with other gases. Although the use of MS is not yet widespread in fermentation monitoring (Reuss et aZ., 1975; Weaver and Abrams, 1979), it has been used successfully in many medical applications, particularly blood gas (Woldering et al., 1966; Dyken, 1972). In summary, it appears that animal cells grow best at dissolved oxygen concentrations close to that found in blood. Supplying the oxygen demand for large-scaleculture cannot be achieved by surface oxygenation. Moreover, sparging gas into the culture can be damaging to the cells and cause serious foaming problems. Hence, it would appear that membrane oxygenation must be used for adequate gas exchange. Finally, the cell mass can be indirectly quantitated by measuring the oxygen uptake rate.

XII. Carbon Dioxide Measuring the production of carbon dioxide promises to be another useful means of monitoring and controlling fermentation processes. In a number of instances, cell mass has been correlated with cumulative COOproduction (Mou, 1979; Mou and Cooney, 1976), and others have used COz production as a means to close carbon mass balances and estimate growth rates and cell mass (Wang et al., 1976). Carbon dioxide is the end product of a number of metabolic processes, and the total amount produced is dependent upon the activity of the individual pathways, as well as the total amount of cell mass. The respiratory quotient (RQ) is defined here and serves as a valuable indication of the cell's COz production rate O2 uptake rate metabolism. A number of workers have controlled the growth of yeast by limiting the glucose addition on the basis of RQ values (Nagai et aZ., 1976; RQ =

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

157

Wang, 1977). Excess glucose was observed to induce an anaerobic state in yeasts and to form ethanol instead of cell mass, the change in metabolism being concurrent with a sudden increase in RQ. Efficient conversion of glucose to cell mass was effected by limiting the rate of glucose addition to maintain the RQ at a value of about 1. Carbon dioxide plays an additional role in animal cells, since the C02bicarbonate system is one of the three important buffers in the blood of vertibrates. Hence, it is of interest that in the development of BME, Eagle (1971) found the C02-bicarbonate system optimal for the growth of cells. Additionally, a class of nongaseous buffers has been developed for cells in mass culture (Eagle, 1971). However, colonies will grow from single cells with high efficiency only if carbon dioxide is also present. There are essentially two methods to measure C 0 2 production: off-gas analysis and dynamic assessment. The off-gas analysis instruments typically measure the infrared absorbance. They are considerably less sensitive to gas flow rates than the paramagnetic oxygen analyzers and have much faster response times. Whether the flow of gas through the system will give adequate resolution depends upon the concentration of C02 in the inlet gas stream, assuming the pH to be buffered in part by a C02-bicarbonate system. Suppose we have lo6WI-38 cells per milliliter and the RQ = 1, then 5 liters of culture should produce 16.8 ml CO.Jhour. If the Pcoz in the inlet gas stream is 2% and the desired Pco, in the outlet stream is 2.2%,then the flow would be 8.4 literhour. Instruments with adequate resolution in this range are available. However, an important problem associated with C02 off-gas analysis is that the partial pressure of C 0 2in the off-gas is not in equilibrium with the concentration of dissolved COP(Reuss et al., 1975).This is owing to the fact that desorption of COBfrom the liquid is a relatively slow process, presumably due to slow conversion of bicarbonate to carbonic acid. The method of off-gas analysis is valuable for determining an integrated production of COP.With increasing flow rates, the measurements of instantaneous COPproduction rates become more reliable. The techniques of dynamic assessment have a distinct advantage in the case of instantaneous C 0 2 production rate. The methodology is essentially the same as for measurement of the oxygen uptake rate: the gas flow is momentarily interrupted and the accumulation of COOin the broth is measured. There are basically two steam-sterilizable instrumental methods available for measuring COP:the tubing method and the mass spectrometer. The tubing method (Phillip and Johnsen, 1961; Kilburn and Webb, 1968; Self et al., 1968; Kilburn et al., 1969) involves passing a stream of inert carrier gas (e.g., nitrogen or helium) through a coil of permeable tubing (e.g., silicone rubber or polypropylene), and then measuring the dissolved gas, which enters the tubing by diffusion-permeation and is swept to a C02

158

ROBERT J. FLEISCHAKER ET AL.

transducer. The system has a response time of about 2-10 minutes, depending on the flow rate and the permeability of the tubing. Although it has found little use in microbial systems, it is well suited for work with animal cells. Using a 3 m length of silicone rubber tubing (1.0 and 2.0 mm OD), Kilburn and Webb (1968)found the CO, content of carrier gas to be a linear function of the partial pressure of COPexternal to the tubing. At 38 mm Hg, the C 0 2 content of the carrier gas (with a flow rate of 55 muminute) was 0.355%. The tubing method can also be used to measure other dissolved gases by placing other gas transducers downstream in the carrier gas (Fig. 8A). A mass spectrometer (MS) can serve as a C 0 2 transducer in the tubing method, or it can also be used continuously to measure dissolved COP.In the probe configuration (Fig. 8B), using a macroscopic probe (Reuss et al., 1975), the mass spectrometer system consists of a specially constructed membrane-covered inlet system introduced directly into the culture and coupled to the instrument by means of a stainless-steel tube. The silicone rubber membrane interfaces the culture broth to the vacuum of the MS, allowing the low-molecular-weight compounds to enter the MS by diffusion across the membrane. In principle, the MS system is very similar to the tubing method. Its advantages are that faster response times are obtained and that a large number of different compounds can be measured simultaneously with one A inert aas

FIG.8. Assemblies for measurement of dissolved carbon dioxide by (A) tubing method and (B) mass spectrometer.

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

159

instrument. The MS has been used for some time in blood gas analysis (Woldering et al., 1966), and more recently in fermentation monitoring of dissolved C 0 2 , dissolved oxygen, and methanol concentrations (Reuss et al., 1975). The technique provides good chemical concentration resolution and responds linearly to dissolved gas and dissolved volatile compounds over a wide range of concentrations (e.g., 5 X to 1 M for dissolved ethanol) (Reuss et al., 1976; Weaver et al., 1978). Generally, MS cannot be used to examine a number of substances simultaneously without additional separation techniques (such as gas chromatography). The reason for this is that mass spectra tend to be complicated and examination of several compounds simultaneously can result in significant overlap, seriously compromising the quantitation of the individual species. The nightmare that would result from injecting the culture broth directly into the MS is in practice avoided by the selectivity of the semipermeable membrane. The membrane allows only the dissolved gases and highly volatile organics to diffuse into the vacuum chamber, and thus greatly simplifies the interpretation of the spectra. Recent work by Weaver and Abrams (1979) has shown that selective expression of volatile acids and bases is made possible by controlling the pH at the MS interface. Hence, overlap from organic acids and bases can in principle be accounted for. Nonetheless, neutral organic compounds are not removed and the presence of several similar acids or bases may prove to be troublesome. In summary, measurements of the production of CO, can prove useful for estimating cell mass and growth rates. The measurement is complicated by slow rates of COBdesorption (when using off-gas analysis) and the presence of CO, in the inlet stream used for the C0,-bicarbonate buffer system. The dynamic method of analysis for COP production appears to be the most promising for measuring CO, production rates. Additionally, the respiration quotient, which can be determined by combining the C 0 2 production rate with the oxygen uptake rate, is a useful indication of cell metabolism.

XIII. NAD+-NADH It is possible to measure the concentration of cell mass on the basis of total culture fluorescence (Fig. 9), since the intracellular reduced nicotinamide nucleotides fluoresce at 460 nm when cells are irradiated by 360-nm light (Zabriskie and Humphrey, 1978b; Harrison and Harmes, 1972; Harrison and Chance, 1970; Ristroph et al., 1977). The fluorescence property has been used extensively by biochemists in studying the response of NAD-NADH levels to environmental changes. Zabriskie and Humphrey (1978b)and Harrison and Chance (1970) have taken this approach for estimation of biomass concentration in fermentation culture. They found that culture fluorescence

160

ROBERT J. FLEISCHAKER ET AL.

FIG. 9. Assembly for measurement of culture fluorescence.

was a function of biomass concentration, together with a number of environmental factors, such as dissolved oxygen concentration, energy substrate concentration, temperature, and pH. When these environmental conditions were carefully controlled, the log of the fluorescence was linearly related to the log of the biomass. Fluorescence increased dramatically with temperature (2.5WoC),which again mandates the requirement for accurate temperature control. However, the fluorescence decreased with increasing pH, an effect thought to be due to the ionic state of the fluorophores or some quenching species. For los WI-38 cells/milliliter, the dry cell weight is about 0.6 gm/liter. If the fluorescence on a per weight basis is the same as that measured for yeast, we would expect a fluorescence signal with an SIN ratio of 50, as measured using the instrument of Zabriskie. Media used to support microbial growth often have a background fluorescence (Ristroph et al., 1977). While it is not clear how much background signal is to be expected from media used for cell culture, these numbers do not seem unreasonable. The use of this technique with microcarrier culture has two added advantages. One is the possibility of allowing the beads to settle and determining the background signal at any time, the other is the possibility of making the measurement without interference from air bubbles, which apparently can cause significant instability in the signal. Although a suitable fluorescence instrument is not readily available, it would not be complicated to construct (Mayer et al., 1969). However, it would encounter one problem in long-term use, which is the drift in the lamp intensity over time. It would appear essential in construction of such an

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

161

II

A

AY

diolysis membrane

Substrate

y

_ _ _ - - _H'

NAD

'@&O

Enzyme

Product

A anode

cathode

FIG. 10. (A) Electrode for estimation of biomass (see also Matsunaga et al., 1978). (B) Idealized process occurring at electrode surface.

instrument to provide for measurement of the intensity of the excitation beam and to determine the ratio of the output signal to the intensity of the lamp, in order to normalize the signal. Matsunaga et al. (1979)have reported on an electrode system for estimating cell mass directly (Fig. 1OA). The system consists of two electrodes, which are basically identical except that one is covered with a dialysis membrane (the reference electrode). Each electrode has a platinum anode and a silver peroxide cathode. Phosphate buffer (1.0 M , pH 7.0) was used as the catholyte, and an anion-exchange membrane was used as a separator. The electrode is steam-sterilizable, and the output current is proportional to the cell number. In principle, the electrode operates by accepting electrons directly from NADH-FADH2 groups located on the cell surface (Fig. 10B) (Matsunaga et al., 1978). By virtue of the dialysis membrane the reference electrode excludes the cells but not the oxidizable groups in solution, and thus the background signal can be determined. Only viable cells seem to contribute to the electrode signal. Since the cells themselves must be transported to the electrode surface in order to transfer electrons, the response time of this electrode should be governed by the transport rate of cells or by the diffusion

162

ROBERT J . FLEISCHAKER ET AL.

of substrate materials to cells in the vicinity of the electrode surface. Hence, a slow response is expected, and in fact the electrodes have a response time of about 15 minutes. With cells bound to microcarriers, still longer response times might be anticipated. In their work with a yeast suspension, Matsunaga et a2. (1979) found the signal to be sensitive to temperature (about 2%/”C)and pH (about 16%per pH unit). However, with adequate control the relative error in these signals can be reduced to less than 1%. The electrode showed a linear response with bacteria and yeast over concentration ranges of lo8- 4 x 10’ and 10’- 4 x lo8, respectively. It is not known whether this rnethood will work for monitoring growth on microcarrier culture. If the response is assumed to be proportional to cell mass, we can estimate the bacterium culture to vary over the range of 0.1- 4 gmfliter. Growth of WI-38 cells on microcarriers would be expected to vary over the range of 0.1- 0.6 gm/liter, so that mass is of the same order of magnitude. On the other hand, surfwe area of the minimum number of observable bacteria or yeasts can be estimated as about 7 cm2/milliliter.The surface area of the beads is about 30 cm2/milliliter.Hence, one would also hope to be able to measure the growth of cells on beads on the basis of surface area. In summary, measurements of NADH, whether done intracellularly by fluorescence or on the cell surface by means of an electrode system, are worth considering as a means of measuring cell mass on microcarriers.

XIV. Oxidation-Reduction Potential

The use of oxidation-reduction potential (ORP) has been advocated by a number of investigators for use in the control and monitoring of animal cell culture (Toth, 1977). The argument is that measurements of changes in the availability of electrons in the culture fluid would be valuable, owing to the importance of molecular charge conditions for the regulation of animal cell membrane activity. However, there are two major problems involved in making these sorts of measurements with the present redox probe. The first is that under aerobic conditions, the ORP is principally (about 90%)a measurement of the dissolved oxygen concentration (Lengyel and Nyiri, 1965). However, a correction might be applied, if an independent dissolved oxygen measurement is made. The second problem is more fundamental, in that it is likely the cells are capable of adjusting their intracellular ORP independently of the ORP of bulk environment. In summary, it is felt that the information obtained with the present ORP electrode can be more easily obtained with dissolved oxygen probes. Moreover, if one is interested in measuring the ORP within the cell, the fluorescence technique for measuring the NADH concentration is more effective.

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

163

XV. ATP Adenosine triphosphate (ATP) functions as currency for the energy available in living cells. Consequently, the measurement of intracellular ATP concentrations gives a valuable indication of the metabolic state of the organism. For an actively growing culture, the total amount of ATP can be related to total cell mass. However, there is a difficulty encountered in measuring the metabolic intermediates and coenzymes on account of their rapid decay. The turnover time for the complete pool of ATP in a growing cell is about 1second. Hence, measuring these compounds requires a means of sampling and quenching the culture within a small fraction of a second. Harrison and Harmes (1972) described the operation of a pneumatic ball valve rapid sampler, which would allow the operator to collect a 10-ml sample of culture in a fraction of a second. Cells thus collected can be quenched with phosphoric acid and then refrigerated for several hours without loss of ATP. Measurement of the amount of ATP can be easily accomplished by use of the biometer luminescence method (Trauberman, 1969).

XVI. Cytofluoremetry In introducing this section, a general comment concerning interpretation of microbiological data seems in order. If every cell were uniform in size and in composition, the interpretation of measurements would be greatly simplified. However, cells show a great deal of variability over the course of their growth cycle with respect to composition, in particular DNA, RNA, and protein contents. Cells also show variability in relation to the environment, as changes in temperature, pH, and media components can alter cell size, shape, composition, and metabolism. What then does one examine and how does one interpret the data? The answers will depend upon the problem being analyzed. For example, in the production of single-cell protein, one would be most interested in measurements of total cell protein. However, most situations are not so well defined and the user must determine which measured parameters can best characterize his process. In general, the tendency (especially with microorganisms) is to treat the system as a homogeneous population, uniform in composition and activity. Such treatment is especially valuable in the evaluation of bulk thermodynamic constants such as heat or gram of cells produced per mole of glucose consumed. Variations in metabolism that occur over the growth cycle of an organism are “smoothed out” by such methods of analysis. On the other hand, analysis of population distributions are very useful for understanding those processes in which the organisms of interest undergo various forms of differentiation (e.g., the production of antibiotics by molds).

164

ROBERT J. FLEISCHAKER ET AL.

Under such circumstances one is interested in quantitating the size of a specific population and the influence of the environment on it. The cytofluoremeter has rapidly gained prominence as a means to determine population distributions (Horan and Wheeles, 1977). The instrument operates by transporting a suspension of cells past a laser beam via a laminar sheath-flow technique (Fig. 11). The sheath-flow techique confines the cells to the center of the flow stream and, by adjusting the cell concentration, the cells are illuminated one at a time. Present day electronics and mechanics allow analysis of as many as 3 X lo4cells/second. By means of various dyes, previously taken up by the cells, different cellular components can be detected by their fluorescence. Cells have been characterized by means of their size, nuclear-to-cellular diameter ratio, DNA content, RNA content, protein content, and content of certain specific enzymes. Several of these determinations can be evaluated simultaneously. The use of cytofluoremetry could be quite valuable in the analysis of cells on beads, provided the cells can be readily removed.

FIG. 11. Assembly for cytofluoremetry.

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

165

XVII. Conclusion The development of microcarrier cultivation by anchorage-dependent cells promises an improvement in both scale of operation and ease of maintaining homogeneous environmental conditions in cell culture. One of the more subtle features of the microcarrier system is the potential for careful monitoring of the growth of animal cells on a surface and for quantitating the effect of the environment on the growth and production of various products. Much of the instrumentation which is useful for obtaining kinetic data on animal cells has been used for some time in work with microbial systems. However, because of the differences in metabolism in animal and microbial cells, a number of pronounced modifications in the measurement of these biochemical events is anticipated. Many of the techniques used with microbial cells may not be sufficiently sensitive for use with animal cells, and most will have to be somewhat modified. Nonetheless, the application of instrumentation to cell culture promises a rapid advance in our understanding of the physiology of animal cells and the influence of the environment upon their metabolic processes. ACKNOWLEDGMENT This work was supported in part by Training Grant 1 T32 CA09.258-01 from the National Cancer Institute.

REFERENCES Augenstein, D. C., Sinskey, A. J., and Wang, D. I. C. (1971). Biotechnol. Bioeng. 13,4@-418. Cady, P. (1975). In “New Approaches to the Identification of Microorganisms” (G. Heden and T. Ileni, eds.), pp. 74-99. Wiley, London. Castellan, G. W. (1964). In “Phpical Chemistry,” pp. 581-593. Addison-Wesley, Reading, Massachusetts. Ceccarini, C., and Eagle, H. (1971). Proc. Natl. Acad. Sci. U.S.A. 68, 229-233. Charles, M. (1978). Ado. Biochem. Eng. 8, 1-62. Cobbold, R. S . C. (1974). “Transducers for Biomedical Measurements. Principle and Applications.” Wiley, New York. Cooney, C. L., Wang, D. I. C., and Mateles, R. I. (1969). Biotechnol. Bioeng. 11, 269-281. D’azzo, J. J., and Houpis, C. H. (1975). “Linear Control System Analysis and Design, Conventional and Modem.” McCraw-Hill, New York. Dyken, M. L. (1972). Stroke 3, 279-285. Eagle, H. (1971). Science 174,500-503. Einstein, A. (1906). Ann. Phys. (Leipzig) 19, 289307. Eriksson, R., and Holme, T. (1973). Biotechnol. Bioeng. Symp. 4, 581-590. Fodge, D. W., and Rubin, H. (1975).J . Cell Physiol. 85, 6 3 5 4 2 . Folkman, J., and Moscona, A. (1978). Nature (London) 273, 345-149.

166

ROBERT J. FLEISCHAKER ET AL.

Geddes, L. A. (1972). “Electrodes and the Measurement of Bioelectric Events,” WileyInterscience, New York. Giard, D., Thilly, W. G., Wang, D. I. C., and Levine, D. (1977). Appl. Enoiron. Microbiol. 34, 668-672. Giard, D., Loeh, D. H., Thilly, W. G., Wang, D. I. C., and Levine, D. (1979). Biotechnol. Bioeng. 21, 433442. Gray, D. N., Keyes, M. H., and Watson, B. (1977). Anal. Chem 49, 1067A-1078A. Griffiths, J . B., and Pirt, S. J. (1967). Proc. R. Soc. London, Ser. B 168, 421438. Guilhault, G. G. (1976). “Handbook of Enzymatic Methods of Analysis.” Marcel Dekker, New York. Hardy, D., Kraeger, S., Dufour, S., and Cady, P. (1977).Appl. Enoiron. Microbiol. 34,14-17. Harrison, D. E. F., and Chance, B. (1970). Appl. Microbiol. 19, 446450. Harrison, D. E. F., and Harmes, C. S. (1972). Process Biochem., 7, April, 13-16. Hayashi, I., and Sato, G. (1976). Nature (London) 259, 132-134. Higuchi, K. (1970). J. Cell Physiol. 75, 65-72. Horan, P. K., and Wheeles, L. L. (1977). Science 198, 149-157. House, W., Shearer, M., and Marcoudas, G. (1972). E x p . Cell Res. 71, 293-296. Jensen, M. D. (1977). In “Cell Culture and Its Application” (R.T. Acton and J. D. Lynn, eds.), pp. 589-602. Academic Press, New York. Kagen, R. L., Schuette, W. H., Zierdt, C. H., and MacLowry, J. D. (1977).J . Clin. Microbiol. 5, 51-57. Kilburn, D. G., and Webb, F. C. (1968). Biotechnol. Bioeng. 10, 801-814. Kilburn, D. G., Lilly, M., Self, D., and Webb, F. (1969). J . Cell Sci. 4, 2 5 3 7 . Kolobow, T., Stool, E., Sacks, K., and Vurek, G. G. (1975).J. Thwac. Cardiooasc. Surg. 69, 947-953. Knazek, R. A., Kohler, P., and Dedrick, R. (1972). Science 178, 65-67. Lengyel, Z. L., and Nyiri, L. (1965). Biotechnol. Bioeng. 7, 91-100. Levine, D. L. (1977). Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, Massachusetts. Levine, D. L., Wang, D. I. C., andThilly, W. G. (1977a). In “Cell Culture and Its Application” (R.T. Acton and J. D. Lynn, eds.), pp. 191-216. Academic Press, New York. Levine, D. L., Wong, J., Thilly, W. G., and Wang, D. I. C. (197%). Somtic Cell Res. 3, 149-155. Mclimans, W. F., Blumenson, L., and Tunnah, K. (1962). Biotechnol. Bioeng. 10, 741-763. Mclimans, W. F., Crouse, E., Tunnah, K., and Moore, G. (1968). Biotechnol. Bioeng. 10, 725-740. Marcoudas, N. G. (1972). E x p . Cell Res. 74, 337-342. Matsunaga, T., Karube, I., and Suzuki, S. (1978). Anal. Chim. Acta 98, 2 5 3 0 . Matsunaga, T., Karube, I., and Suzuki, S. (1979). Appl. Enoiron. Microbiol. 37, 117-121. Mayer, D. H., Williamson, J., and Legallais, V. (1969). Chem. Znstrum. 1, 383389. Mortensen, J. D. (1977). “Evaluation ofASAlO Blood Damage Test,” Vol. 1. Report TR171-008. Mou, D. G . (1979). Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, Massachusetts. Mou, D. G., and Cooney, C. L. (1976). Biotechnol. Bioeng. 18, 1371-1392. Nagai, S., Nishizawa, Y., and Yamagata, T. (1976). Fifth International Fermentation Symposium, Abstract, p. 30. Perley, C. R.,Swartz, J.. and Cooney, C. L. (1979). Biotechnol. Bioeng. 21, 519-523. Phillip, D. H . , and M. J. Johnson. (1961).J . Biochem. Microbiol. Technol. Eng. 3, 261-275. Reuss, M., Piehl, H., and Wagner, F. (1975). Eur. J. Appl. Microbiol. 1, 323325.

INSTRUMENTATION FOR CULTURE PROCESS CONTROL

167

Reuss, M., Piehl, H., and Wagner, F. (1976). Fijlh International Fermentation Symposium, Abstract, p. 25. Richards, J. C. S., Jason, A. C., Hobbs, G., Gibson, D. M., andchristie, R. H. (1978).J . Phys. E. 11,560568. Ristroph, D. L., Watteeuw, C. M., Armiger, W. B., and Humphrey, A. E. (1977). Hakko Kogaku Zasshi 55,599-608. Ryu, D., and Humphrey, A. (1973).I . Appl. Chem. Biotechnol. 23, 283-295. Schwan, H. P. (1968). Ann. N.Y. Acad. Sci. 148, 191-209. Schleicher, J. B., and Weiss, R. E. (1968). Biotechnol. Bioeng. 10, 617-624. Self, D. A., Kilburn, D., and Lilly, M. (1968). Biotechnol. Bioeng. 10, 815-828. Slonim, D. (1974).J . Biol. Stand. 2, 103-110. Snedecor, B. (1977). M.S. Thesis, Massachusetts Institute of Technology, Cambridge, Massachusetts. Swartz, J. R.,and Cooney, C. L. (1978). Process Biochem. 13, February, 3-7. Toth, G. M. (1977). In “Cell Culture and Its Application” (R.T. Acton and J. D. Lynn, eds.), pp. 617-635. Academic Press, New York. Trauberman, L. (1969). Food Eng. 41 (August), 94. Tritsch, G. L., and Moore, G . E. (1962). E s p . Cell Res. 28, 360-364. Ubertini, B., Nardelli, L., Santero, G., and Panina, G. (1960).J . Biochem. Microbiol. Technol. Eng. 2, 327342. Ur, S., and Brown, D. F. J. (1975). In “New Approaches to the Identification of Microorganisms” (G. Heden and T. Ileni, eds.), pp. 61-71. Wiley, London. van Wezel, A. L. (1967). Nature (London) 216, 64-65. van Wezel, A. L. (1973). I n “Methods and Applications ofTissue Culture” (P. F. Kruse and M. K. Patterson, eds.), pp. 372-377. Academic Press, New York. van Wezel, A. L. (1977). Dev. Biol. Stand. 37, 143-147. van Wezel, A. L., and van der Velden-deGroot, C. A. M. (1978). Process Biochem. 13, March, 6-8. Wang, H. (1977). Ph. D. Thesis, Massachusetts Institute of Technology, Cambridge, Massachusetts. Wang, H., Mou, D. G., and Swartz, J. (1976). Biotechnol. Bioeng. 18, 1811-1814. Wang, H., Cooney, C. L., and Wang, D. I. C. (1977). Biotechnol. Bioeng. 19, 69-86. Wang, D. I. C., Cooney, C. L., Demain, A. L., Dunnill, P., Humphrey, A. E., and filly, M. (1979). In “Fermentation and Enzyme Technology,” pp. 187-193. Wiley-Interscience, New York. Weaver, J. C., and Abrams, J. H. (1979). Rso. Sci. Instrum. 50, 478481. Weaver, J. C., Reames, F. M., DeAlleaume, L., Perley, C. R., andcooney, C. L. (1978). Enz. Eng. 4,403404. Woldering, S., Owens, G., and Woolford, D. (1966). Science 153, 885-887. Zabriskie, D. W., and Humphrey, A. E. (1978a). Biotechnol. Bioeng. 20, 1295-1301. Zabriskie, D. W., and Humphrey, A. E. (197813). Appl. Enoiron. Microbiol. 35, 337-343. Zafari, Y., and Martin, W. J. (1977).J . Clin. Microbiol. 5, 545-547.

This Page Intentionally Left Blank

Rapid Counting Methods for Coliform Bacteria A. M. CUNDELL Flow Laboratories, lnc., Roslyn, New York

I. Introduction ........................................... 11. Criteria for a Rapid Colifonn-Counting Procedure .......... 111. Review of Rapid Coliform-CountingProcedures ............ A. Radiometric Methods ............................... B. Serological Methods ................................ C. Electrochemical Methods .................. D. Enzymic Methods ........................ E. Chromatographic Methods ........................... F. Chemiluminescent Methods ......................... IV. Potential for Future Development ........................ V. Summary ....................... ................... References ............................................

169 170 171 171 172 178 178 179 180 180

I. Introduction The ability to detect fecal contamination of water and foodstuffs is a prerequisite to the maintenance of public health. The bacterium Escherichia coli is universally preferred to disease-causing organisms as an indicator of possible fecal pollution because (1)disease-causing bacteria may be emitted from the lower intestines of infected humans and animals intermittently or continuously, in small numbers; (2) diseasecausing bacteria may be fastidious in their growth requirement and hence, are not easy to culture in the laboratory, making their detection difficult; (3) E. coli is present in the human intestine in greater numbers than is any other bacterium (with the exception of the strict anaerobe Bacteriodes fiagilis) (Leclerc et at., 1977), can be readily cultivated, and is therefore a suitable candidate as a bacterial indicator of fecal matter; and (4) E. coli survives longer than do most pathogenic organisms in water, but does not multiply under these conditions (after Anonymous, 1976). The aims of developing new technologies for the rapid enumeration of coliform bacteria are to reduce the time needed to count coliform from 24 hours to less than 1, to standardize the procedure so that it will be more acceptable to the scientific and regulatory communities, to increase the number of coliform determinations a single technician can process in a working day from 15-30 to more than 100, and to ensure selectivity for the differentiation of total and fecal coliforms. The detection levels required to monitor potable water and sewage effluent are 2 and 200 coliforms per 100 169 ADVANCES IN APPLIED MICROBIOUX?I, VOLUME 27 copylight @ 1881 by AaLienlic Ren. rm. All rights of mpmductioo in my form reserved. ISBN O-l%002&7-9

170

A. M. CUNDELL

ml, respectively. Detecting coliform bacteria at this density is like detecting the proverbial needle in the haystack. Two standard methods are prescribed by the American Public Health Association (APHA, 1977), namely, the multiple-tube fermentation technique and the membrane filtration technique. These procedures allow for the presumptive enumeration of coliform bacteria after 24 hours of incubation at 35°C on selective media. An alternative to the development of additional procedures requiring special instrumentation for the rapid counting of coliform bacteria has been advanced by Reasoner and his co-workers (Reasoner et al., 1979). The membrane filtration technique for fecal coliforms was modified by using a lightly buffered lactose medium combined with a double pH indicator system to detect fecal coliform colonies on a membrane after 7 hours of incubation at 41.5"C. Although this approach is useful, it does not meet the criteria for a rapid coliform enumeration procedure laid down in this review article.

II. Criteria for a Rapid Coliform-Counting Procedure The criteria for an ideal rapid coliform-countingprocedure might include: (1) specificity for fecal coliforms, e.g., E. coli; (2) test completion in less than an hour; (3) results comparable to those achieved by the present standard methods; (4) good reproducibility and accuracy; (5) ability to detect a single coliform; (6) amenability to automation; (7) equipment modestly priced; (8) technical simplicity; and (9) potential for field use or on-line monitoring. A number of procedures are described in the literature that have potential as rapid coliform-counting techniques. These procedures are relatively meritorious but are all somewhat less than the ideal. Methods discussed in this review article can be described as (1) radiometric, (2) serological, (3) electrochemical, (4) enzymic, (5) chromatographic, and (6) chemiluminescent. Attempts will be made to assess the methodologies in terms of selectivity for fecal coliforms, detection limits, and rapidity. Before each procedure is examined, a more detailed discussion of the criteria for an ideal rapid coliform-counting procedure would be useful. Fecal coliform bacteria are operationally defined as gram-negative, nonspore-forming, rod-shaped bacteria that ferment lactose with gas production at 445°C within 24 hours of incubation. Epidemiological studies have established that there is a strong correlation between the presence of waterborne pathogens and the presence of bacteria of lower intestinal origin, e.g., E. coli. The presence of coliforms within a water sample is indicative of fecal contamination and the potential presence of pathogens. The acceptability of the standard methods for counting coliform bacteria is well established and is recognized by the scientific and regulatory communities. In the presumptive

RAPID COUNTING METHODS FOR COLIFORM BACTERIA

171

test for the presence of coliforms, a single bacterium may initiate growth in the lactose broth culture tubes and divide through a number of successive generations, so that sufficient bacteria are present within the broth to register as turbidity and gas production within the incubation period. In effect, an extended incubation time allows growth to occur, concentrating the coliform cell density for a visual analysis of the end result, e.g., gas production within an inverted Durham tube. The author believes that for rapidity of detection, the use of an enzyme system, specific to coliform bacteria, as the basis of a counting procedure that permits the organism to be counted without the need for growth or concentration, appears to be a sound approach. To take advantage of the rapidity of the potential coliform-counting methods, automation of both the sample and data handling is necessary. Mechanical devices are now available to automate both continuous flow and discrete analysis and inexpensive data processors can be purchased. The price of specialized equipment for automated coliform analyses will probably be costly and hence, attractive only to state, municipal, and public watertesting laboratories that process significant numbers of coliform determinations. As with all automated analytical equipment, the availability of the device would tend to create an extra demand among potential users. The transfer of routine analyses from the hands of laboratory workers to machines would be welcomed because of increasing labor costs. Technicians would be freed from more routine tasks to concentrate on increasing the quality of service that a water-testing laboratory could deliver. A potential for field use would be invaluable for the emergency monitoring of potable water distribution systems, and on-line monitoring would warn of the presence of fecal contamination in industrial process water or the noncompliance of discharge effluent to environmental regulations.

Ill. Review of Rapid Coliform-Counting Procedures A. RADIOMETRIC METHODS The fermentation of radioactively labeled substrates such as lactose should provide a sensitive measure of coliform growth in an inoculated broth culture. A number of research workers have investigated the use of radiometric methods for counting coliforms (Levin et al., 1956, 1961; Pugsley et al., 1973; Waters, 1972; Bachrach and Bachrach, 1974; Reasoner and Geldreich, 1978). The common approach was to collect a water sample, inoculate a broth culture containing a radioactively labeled fermentable sugar or related metabolite, and incubate at an elevated temperature, i.e., 35-44.5"C. The radioactive 14C02released during fermentation is trapped and detected by a

172

A. M. CUNDELL

liquid scintillation counter, ionization chamber, or electrometer. The 'C02 activity is corrected for the uninoculated control and plotted against the number of coliform bacteria. Problems encountered with these radiometric assays include abiological release of 14C02, unavailability of radioactively labeled lactose, and differences in physiological state of field-collected coliforms. Stressed coliforms usually enter an extended lag phase before respiration and division is initiated. Damage owing to stress such as chlorination, osmotic stock, and temperature changes can occur to the bacterial cell membrane and especially the active transport system (Camper and McFeters, 1979). Repair mechanisms have been postulated that enable the damaged organisms to reenter the growth phase. Typically, coliform bacteria repair better in nutrient-rich media than in selective media containing a limited range of substrates and dyestuffs that exert the selectivity to enumerated coliforms. However, selectivity is required to differentiate noncoliforms from total and fecal coliforms. Two examples of the radiometric method that illustrate the promise of the procedure to the enumeration of coliform bacteria will be discussed. Waters (1972), using a commercially available radiometric respirometer (BACTEC, Johnston Laboratories Inc., Cockeysville, Maryland), found eight common bacterial species, including E. coli, that could be detected at low cell density by incubating the organism with a I4C-labeledglucose broth and measuring the I4CO2liberated. The detection time for a particular culture was proportional to the logarithm of the initial inoculum cell concentration. As few as 10 E. coZi cells could be detected within 8 hours. Bachrach and Bachrach (1974) extended this technique to selectively detect coliforms in water. They found that the growth of E. coli on lactose was slow and that the induction of the key enzyme P-D-gdactosidase by isopropyl-P-D-thiogalactopyranoside (IPTG) markedly increased the release of 14C02from the radioactive lactose. If a fitration step was included in the procedure, as few as 1-10 E . coli could be detected within a 6-hour incubation period.

B. SEROLOGICAL METHODS The fluorescent antibody (FA) technique would appear to offer potential for use in a rapid coliform-countingprocedure. For example, Danielsson and Laurel1 (1965) detected coliform bacteria in water samples by collecting the organism on a nonfluorescent membrane filter where they were combined with fluorescein isothiocyanate (F1TC)-labeled antiserum. Examination of the filter surface with an incident-light fluorescence microscope lead to the detection of the fluorescent bacterial cells. Other researchers have developed this approach. For example, considerable success has been achieved

RAPID COUNTING METHODS FOR COLIFORM BACTERIA

173

in applying the FA technique to the rapid identification of enteropathogenic

E. coli responsible for diarrheas especially in children (Whither et al., 1958; Chadwick, 1966). Unfortunately, the multiplicity of serotypes of E . coZi, e.g., 157 0 antigens, 93 K antigens and 52 H antigens (Edwards and Ewing, 1972) make a fluorescent antibody procedure for counting coliforms impractical unless a universal polyserum is fouhd for E. coli that exhibits little or no cross-reactivity toward other enteric organisms. Guthrie and his co-workers (Abshire and Guthrie, 1973; Guthrie and Reeder, 1969) used an FA technique to detect fecal coliform pollution in water. After the water sample was filtered, the filter was incubated in lactose broth for 2-3 hours and the presence of E. coli within the broth culture was confirmed by staining a smear of the culture on a slide with the fluorescent antibody. In this way, the presence of fecal coliforms within multiple-fermentation tubes is confirmed before gas production is detected. More recently, an automated FA procedure was developed to count enteropathogenic E. coli concentrated on a membrane filter (Drapeau and Laurence, 1977). A pulsed dye laser was the excitation light source and the response from the fluorescing bacterial cells was detected by a photomultiplier tube. As the filter was scanned with the laser, the impulses from the cells were recorded electronically so that the technique was quantitative. This approach to the enumeration of fecal coliforms in wastewater should be further investigated. The Limulus amebocyte lysate assay for endotoxin may be applicable to the rapid assessment of water quality. The lysate clots within 2 hours when exposed to the endotoxin portion of gram-negative bacterial cell membranes. Unfortunately, the selectivity of the procedure embraces gram-negativebacteria and is not exclusive to coliforms. In the so-called fm-clot method, serial dilutions of the water sample are incubated with equal amounts of lysate for 1 hour at 37°C with a circulating water bath. At the end of the incubation period, the tubes are gently inverted and the lowest dilution that forms a firm clot is noted (jorgensen and Smith, 1973). A spectrophotometric determination (absorption at 360 nm), developed by Dr. S.W. Watson of the Woods Hole Oceanographic Institute and described by Evans et al. (1978) appears to be more convenient than the firm-clot method. Evans et al. found excellent correlations between total coliform numbers and the spectrophotometrically determined bound endotoxin for as few as 100 organisms per milliliter. In contrast, Jorgensen et aZ. (1979) reported that the Limulus endotoxin assay was an inadequate measure of both total plate counts and total coliform number in treated final sewage effluent. The inability of the assay to distinguish between viable and nonviable bacterial cells was probably responsible for the poor showing of the assay. Preliminary results suggest that the determination of lipopolysaccharide

174

A. M. CUNDELL

content in a bacterial suspension may be an alternate approach to coliform counting. The bioluminescence technique described recently (Ulitzur et al., 1979) depends on the luminescence response of the marine bacterium Beneckea harueyi to the myristic acid component of the lipopolysaccharide fraction of the cell membrane of gram-negative bacteria. Additional work is needed to demonstrate the applicability of this new procedure to the enumeration of coliform bacteria in water samples.

C. ELECTROCHEMICAL METHODS Three electrochemical methods for the enumeration of bacteria are found in the literature, They are (1) the measurement of the evolution of molecular hydrogen within a growing bacterial culture, (2) the measurement of changes in resistivity (impedance)as the bacterial culture modifies the medium, and (3) measurement of electron transfer as bacterial cells contact an electrode surface. Wilkins and his co-workers (Wilkins et al., 1974, 1978; Wilkins and Boykin, 1976) described an electrochemical technique for detecting bacteria using a platinum/calomel electrode to measure hydrogen production within bacterial cultures. They have established a linear relationship between initial cell concentration and the time at which hydrogen was detected within the medium as a voltage increase between the electrodes. Using a lactose broth incubated at 445°C as the growth medium, a single E. coli can be detected in 8 hours. An automated sensor for the detection of coliforms based on the technique developed by Wilkins and his co-workers was designed for use in the National Aeronautical and Space Administration (NASA) Water Monitoring System. Differentiation within the coliform group was possible through the fermentation of lactose at 35 (total)and 44.5"C(fecal).The sensor consisted of (1)a control switch, (2) eight incubation cells, (3) a sample reservoir and transfer system, and (4) instrumentation. Each cell is charged with a measured volume of fresh medium maintained at the required .temperature and inoculated automatically; the incubation period required for a rapid release of hydrogen is determined from the electrode signal. A cleanup procedure including flushing and sterilization with a hydrochlorite solution prepares the cells for the next determination. Such an automated sensor could float on a water body and monitor fluctuation in coliform numbers (Anonymous, 1978). Electrical impedance measurements have been successfully applied to the detection of bacteriuria. An automated continuous impedance-monitoring apparatus (Bactometer 32, Bactomatic Inc., Palo Alto, California)is available for clinical use and can detect lo5 bacterial cells within 2 hours of culture.

RAPID COUNTING METHODS FOR COLIFORM BACTERIA

175

The basis of the technique is the breakdown of nutrients such as carbohydrates that are large electrically inert molecules to large numbers of small electrically charged molecules and ions. This increase in electrical conductivity of the medium is reflected in changes in the electrical impedance. A detailed discussion of the principles of the technique has been recently published (Cady, 1978). Silverman and Munoz (1979)found that fecal coliforms growing in a selective lactose-based broth at 44.5"Cgenerated an electrical impedance change relative to a sterile control of the same medium when the inoculum multiplied to excess of 106organisms per milliliter. They found that a linear relationship existed between the initial number of coliforms and the time taken for the change in impedance to occur. A single coliform bacterium could be detected within 8-12 hours. However, this extended incubation time and the need for a selective medium that may inhibit the recovery of stressed coliforms are serious disadvantages to the technique. Recently, a dual electrode system was reported to be useful in determining bacterial cell densities in fermenters (Matsunaga et al., 1979). Two electrodes, each consisting of a platinum anode and a silver peroxide cathode, were placed in the bacterial culture. The anode of the reference electrode was shielded with a cellulose dialysis membrane. After 15 minutes in the culture broth, the current difference between the two electrodes was found to be proportional to the cell numbers. Apparently, the difference in current flow was due to the direct transfer of electrons from the bacterium Lactobacillus fmnentum on contact with the unshielded electrode. However, the detection threshold was lo8 organisms per milliliter, so this approach would not be useful in monitoring coliform bacteria in wastewater.

D. ENZYMIC METHODS As stated earlier in this article, a sound approach to the rapid detection of coliform bacteria is the use of an enzyme system (specific to coliform bacteria) to detect the coliforms. The use of this system would mean that the enumeration is not dependent on the growth of the organisms to a sufficient concentration to be detected visually as a turbid broth tube or colonies on a solid medium. Unfortunately, a limited number of enzymes are specific to any one species of bacterium. In addition, the enzyme should be constitutive and not adaptive, to limit the time needed to induce the enzyme system. Examination of the biochemical characteristics of E. coli reveals limited possibilities. The ability to ferment lactose implies the presence of the enzymes P-galactosidase and formic hydrogenlyase. Indole production, lysine decarboxylase activity, and arabinose fermentation are characteristic but not exclusive to E. coli.

176

A. M. CUNDELL

An automated procedure for the counting of coliform bacteria, especially

E. coli, has been developed by Leclerc and his co-workers (Trine1 and Leclerc, 1972; Leclerc et al., 1977) based on the presence of the constitutive enzyme glutamate decarboxylase in E. coli. Water samples are filtered and concentrated during passage through a hollow fiber ultrdtration system. The E. coli are enriched during growth in lactose broth at 41°C. After the carbon dioxide is stripped from the culture, glutamate is added, the solution is buffered at low pH to liberate the C o t produced during the enzymic reaction, and the COP is passed into a phenolphthalein solution where it is detected colorimetrically. As few as 1 coliform per 100 m1 can be detected if the lactose broth is incubated for up to 10 hours to ensure that the concentration of E. coli reaches at least 5 X lo4per milliliter. This procedure has been used to count fecal coliforms in milk and polluted water (Moran and Witter, 1976). Bacterial isolates of clinical significance have been identified by measuring enzyme profiles by a continuous flow system (Bascomb, 1978). The three classes of enzyme assayed were (1) ammonia-releasing enzymes, e.g., phenylalanine deaminase, (2) nitrophenyl-releasingenzymes, e.g., B-galactosidase and phosphatases, and (3)dehydrogenase, e.g., alanine dehydrolase. The advantages of this automated system appear to be the ability to process large numbers of cultures within 1 hour of receipt, the saving in labor, and the amenability of the system to data collection, storage, and processing. Whether any of these enzymes have potential in a coliformcounting procedure remains unexplored to date. Two rapid coliform-counting procedures based on the P-D-galactosidase activity in E. coli have been presented recently. These procedures are based on a colorimetric and fluorometric assay, respectively (Warren et al., 1978; Cundell et al., 1979a). The first procedure makes use of the chromogenic substrate o-nitrophenyl-6-D-galactoside(ONPG), which is hydrolyzed within a selective lactose broth at 44.5"C by fecal coliforms. The lag time before the development of the color of the liberated o-nitrophenol was found to be proportional to the number of E. coli present at the start of the incubation period. An examination of the curves of ONPG hydrolysis rate generated by increasing sized inocula of E. coli K-12 suggests the occurrence of simultaneous induction of P-galactosidase activity and increase in cell density. The technique was used to count fecal coliform in field samples. Coliform densities in the range of 5-1000 fecal coliform per 100 ml could be detected within 24 to 9 hours, respectively. The author developed a rapid coliform-counting procedure based on the hydrolysis of a fluorogenic substrate by the inducible enzyme PD-gdactosidase (Rotman, 1961). The method involves the induction of the

RAPID COUNTING METHODS FOR COLIFOFlM BACTERIA

177

enzyme in a coliform suspension using the powerful inducer IPTG. The suspension is dispersed onto a modified microscope slide, containing a layer of silicone oil, using a simple aerosol sprayer. Within the resultant microdroplets (5-50 p m diameter), the fluorogenic substrate fluorescein di@-Dgalactopyranoside) (FDG) at a concentration of 5 X M is hydrolyzed by the coliform bacteria when the slide is incubated at 44.5"C for 15 minutes. The liberated fluorescein accumulates within the individual oil-encapsulated microdroplets so that they can be counted with a fluorescence microscope (Excitation Filter BG 38, Barrier filter yellow 510). A standard curve of the percentage of fluorescent droplets per field of view (125X magnification) and the cell density (5 x 105-108organisms per milliliter) in the coliform suspension was established. The curve was used to count coliform bacteria in sewage effluent within 90 minutes (Cundell et al., 1979a). However, the major disadvantage of the procedure is the necessity of having a cell density above lo5 organisms per milliliter to obtain sufficient bacteria within the microdroplets so that they can be counted with the fluorescence microscope. It may be possible to concentrate dilute suspensions of coliform bacteria using (1) continuous centrifugation, (2) hollow fiber filtration, or (3) membrane filtration. Large volumes of water sample could be processed with moderate recovery rates with any of these concentration techniques but their use would by necessity increase the time taken to enumerate the coliform bacteria. Preliminary work by the author (Cundell et al., 1979b)has suggested that individual coliforms can be counted within a bacterial suspension using a flow fluorocytometer. Fluorescein can be concentrated within bacterial cells using (1)the phenomenon of fluorochramasia [Rotman and Papermaster (1966)defined this as the intracellular accumulation of fluorescein produced by the hydrolysis of the fluorogenic substrate FDG by the enzyme p~-galactosidase]or (2) the addition of a fluorescein-coupled antiserum to the enzyme p-D-gdactosidase. For example, a water sample containing coliform bacteria is collected and filtered to remove the particulates, and the enzyme is induced within the bacteria. FDG added to the cell suspension is transported into the bacterial cells by means of a specific permease system where it is subject to enzymic hydrolysis. Since fluorescein is produced more rapidly than it diffuses out of the cells, the intracellular concentration of fluorescein will increase to a sufficient level to enable the coliform bacteria to be specifically detectable by flow fluorocytometry. Although bacterial cells are at the lower threshold for detection in commercially available flow fluorocytometers, e. g., Cytofluorograf (Ortho Instruments, Westwood, Massachusetts), microorganisms can be detected by light scattering and fluorescence (Skogen-Hagenson, 1976; Hutter et al., 1978).

178

A. M. CUNDELL

E. CHROMATOGRAPHIC METHODS Gas chromatography with and without mass spectrography is a promising approach for the identification of different bacterial species. Characteristic organic components or fermentation products of the bacteria can be analyzed after (1)pyrolysis of a lyophilized bacterial sample, (2) solvent extraction of the bacterial media or disrupted bacterial cells, and (3) sampling the head space of a growing bacterial culture. The chromatograms obtained by these procedures can be considered “fingerprints” that can be used to identify the bacterial species. The disadvantages of gas chromatographic techniques have proved to be insufficient resolution of the fingerprint, lack of reproducibility owing to the degradation of the gas chromatographic column with time, the technical complexity of the technique, and its inability to process samples at a rate greater than one or two per hour (Meuzelaur et al., 1977). However, Newman and O’Brien (1975) developed a gas chromatographic procedure that determined coliform bacteria concentration from the incubation time required for ethanol to be produced in a lactose broth at 37 or 44.5”C. Using this technique, five coliforms per milliliter were detected using a 9-hour incubation period. Another approach for the detection of fecal contamination of a water sample is the analysis of that sample for the presence of the fecal sterol coprostanol. Coprostanol is the major sterol excreted by man. Cholesterol is converted to coprostanol by the anaerobic microflora of the lower intestine and the latter chemical has the potential to be a chemical indicator of recent fecal pollution (Murtaugh and Bunch, 1967; Smith and Gouron, 1969).The coprosterol output has been estimated at about 500 mg per person per day. Analytical procedures are available to detect coprostanol at a concentration as low as 0.5 ppb. A column extraction procedure is now available to concentrate fecal sterols from water samples to be analyzed by gas chromatography (Wun et al., 1978).However, recent evidence suggests that the relationship between sterol concentration and bacterial water quality parameters is not simple (Dutka et al., 1974). This lack of correlation between sterol concentration and coliform cell density in sewage treatment plant effluent, and the time that must be expended to run the chemical analyses reduce the potential of this technique for predicting levels of indicator organisms in water.

F. CHEMILUMINESCENT METHODS A number of chemical reactions that produce light can be used to detect microorganisms, including (1)the firefly luciferase ATP assay, (2) the liminol chemiluminescent bacterial iron porphyrin assay, and (3)the chemiluminescent immunoenzymatic assay. Commercial instruments are available for the measurement of bacterial ATP. The assumptions underlying this technique

RAPID COUNTING METHODS FOR COLIFORM BACTERIA

179

are that all living cells contain ATP, and the average ATP content of a single bacterium is about 2.5 x ng per organism. One thousand bacteria (10-13 g ATP) represent the detection limit and the procedure may be completed within 10 minutes (Chappelle and Levin, 1968). Most commercial photometers have a liquid-liquid injection system into their light-shielded compartment. However, Chappelle and his co-workers (Picciolo et al., 1977) evaluated a new type of photometer with a moving filter tape (Vitatect IIs, Vitatect Corp., Alexandria, Virginia) to concentrate the bacteria and reactants. Escherichia coli suspensions in the concentration range of 7 X 103-107 cells per milliliter gave linear responses within a 1-ml sample with a processing time of 2 minutes. The luminolhydrogen peroxide reaction is catalyzed by bacterial porphyrins. Automated detection procedures based on this reaction have been reported (Oleniacz et al., 1968; Picciolo et al., 1978). Oleniacz and his co-workers used an Autoanalyzer (Technicon, Tarrytown, New York) as a sampling, mixing, and incubation device, with the reactants being passed into a light-proof reaction chamber where a photometer measured the light emitted. The instrument showed a linear response for pure cultures of common bacteria in the concentration range of 103-105organisms per milliliter. An automated system developed out of the NASA Advanced Technology Transfer Program incorporated the capacity to measure total bacteria, viable bacteria, and coliform levels using the new ATP luciferase, luminol, and a hydrogen electrode procedure, respectively (Picciolo et al., 1978). A concentration step, i.e., the passage of samples through a hollow fiber fiter, was included in the system. It should be emphasized that the ATP-luciferase assay is specific to viable cells but will not differentiate coliform bacteria from other microorganisms, whereas the luminol assay can respond to soluble porphyrins and oxidizing agents that may be present in the sample, as well as to particulate bacterial porphyrins. Immunoenzymatic assays for the detection of small numbers of specific bacteria present an attractive solution to the rapid enumeration of coliforms. Halmann et al. (1977) described a chemiluminescent method for the measurement of peroxidase-labeled antibody bound to Smatia marcescens that enables as few as 100 bacteria to be detected within 1 hour. The peroxidase activity was measured by determining the number of photons emitted during the oxidation of pyrogallol with hydrogen peroxide. Again, the multiplicity of serological types among coliform bacteria prevents this approach in the absence of a universal polyserum.

IV. Potential for Future Development The literature would seem to suggest that the automation of a rapid coliform-counting procedure is possible in the near future. Although the

180

A. M. CUNDELL

problem of counting a specific bacterial population within a mixed population, where the coliform numbers may be as low as one bacterial cell per 100 ml, is formidable, a number of researchers have made considerable progress toward finding a solution to the problem of rapid enumeration of coliforms. Since the role of fecal contamination of potable water in the transmission of human disease was first understood, bacteriological parameters have been sought as indicators of water quality. The standard methods prescribed by the American Public Health Association, i.e., multiple tube fermentation and membrane filtration techniques, are recognized by the public health, water purification, and legal communities. Any new procedure would need to gain the acceptance of these communities. The acceptance of the membrane filtration procedure was slow and the conservatism of the regulatory groups is justified, considering their responsibility for human life. However, when an automated procedure is developed that has obvious advantages over existing manual procedures, the forces for acceptance of that procedure will be strong. The most important features of any acceptable technique will be its specificity for coliforms (especially fecal coliforms), its accuracy compared to the current standard methods, the sensitivity of detection of coliform bacteria, the speed in which the procedure can be completed, and the cost of the instrumentation.

Y. Summary Recent progress in the development of rapid coliform-counting procedures was reviewed. Emphasis was placed on the detection level and the rapidity of the procedures. The methods were classified as (a) radiometric, (b) serological, (c) electrochemical, (d) enzymic, (e)chromatographic, and (Q chemilumescent

.

REFERENCES Abshire, R. L., and Guthrie, R. K. (1973). Fluorescent antibody as a method for the detection of fecal pollution: Eschffichh coli as indicator organisms. Can. J . Microbid. 19,201-206. American Public Health Association (APHA) (1977). “Standard Methods for the Examination of Water and Wastewater,” 14th ed. APHA, New York. Anonymous (1976). Defmition of terms concerning coliform bacteria and recommended methods for their detection. Report of recommendations of the Microbiological Society’s Committee on coliform bacteria, June 1975. N. Z. J . Sci. 19, 215-219. Anonymous (1978). Automated electrochemical detection of colifonns. NASA Tech. Brie$. Summer, 243-244. Bachrach, U., and Bachrach, Z. (1974). Radiometric method for the detection of coliform organisms in water. Awl. Microbial. 28, 169-171.

RAPID COUNTING METHODS FOR COLIFORM BACTERIA

181

Bascomb, S. (1878).Identification of bacteria by automated continuous flow analysis. In ‘‘Automation in Microbiology” (R.S. First, ed.). Cady, P. (1878).Progress in impedance measurements in microbiology. In “Mechanizing Microbiology” (A. N. Sharpe and D. S. Clark, eds.), pp. 199-239.Thomas, Springfield, 11linois. Camper, A. K., and McFeters, G. A. (1979).Chlorine injury and enumeration of waterborne coliform bacteria. Appl. Enuiron. Microbiol. 37, 633-641. Chadwick, P. (1966).The relative sensitivity of fluorescent antibody and cultural methods in detection of small numbers of pathogenic serotypes of Escherichia coli. Am. J . Epidaiol. 84,150-155. Chappelle, E. W., and Levin, G. V. (1988).Use of frefly bioluminescent reaction for rapid detection and counting of bacteria. Biochem. Med. 2,4152. Cundell, A. M., Pisani, A. M., and Findl, E. (1979a).A rapid method for detection of coliform bacteria. Devel. Znd. Microbiol. 20,571-577. Cundell, A. M., Porter, J . E., Pisani, A. M., and Findl, E. (1979b).Development of a rapid coliform detection procedure suitable for monitoring renovated wastewater. Proc. Water Reuse Symp. 3, 1895-1899. Danielsson, D., and Laurell, G. (1965).A membrane fdter method for the demonstration of bacteria by the fluorescent antibody technique. 2. The application of the method for detection of small numbers of bacteria in water. Acta Pathol. Microbwl. S c a d . 72, 118124. Drapeau, A. J., and Laurence, R. A. (1977).Le laser a impulsions pour le denombrement en fluorescence de Escherichia coli enteropathogene sur membrane fdtrante. Water Res. 11, 937-943. Dutka, B. J., Chau, A. S. V., and Coburn, J. (1974).Relationship between bacterial indicators of water pollution and fecal sterols. Water Res. 8, 1047-1055. Edwards, P. R.,and Ewing, W. H. (1972).“Identification of Enterobacteriaceae.” Burgess, Minneapolis, Minnesota. Evans, T. M., Schillinger, J. E., and Stuart, D. G.(1978).Rapid determination ofbacteriological water quality by using Limuluu lysate. Appl. Enuiron. Microbiol. 35, 376382. Guthrie, R. K.,and Reeder, D. J. (1969).Membrane fdter-fluorescent-antibodymethod for detection and enumeration of bacteria in water. Appl. Microbiol. 17,399401. Halmann, M., Velan, B., and Sery, T. (1977).Rapid identifiation and quantitation of small number of microorganisms by a chemiluminescentimmunoreaction. Appl. Enoiron. Microbid. 34,47347. Hutter, K.J., Eipel, H. E., and Hettwer, H. (1978).Rapid determination of the purity of yeast culture by flow cytometry. Eur. J . Appl. Microbiol. 5, 109-112. Jorgensen, J. H., and Smith, R. F. (1973).Preparation, sensitivity and specificity of Limulus lysate for endotoxin assay. Awl. Microbiol. 26, 43-48. Jorgenson, J, H., Lee, J, C., Alexander, G. A., and Wolf, H. W. (1979).Comparison of Limulus assay, standard plate count and total coliform count for microbiological assessment of renovated wastewater. Appl. Enuiron. Microbiol. 37, 928-931. Leclerc, H., Mossel, D. A., Trinel, P. A., and Gavini, F. (1977).Microbiological monitoring4 new test for fecal contamination. In “Bacterial IndicatordHealth Hazards Associated with Water” (A. W.Hoodley and B. J. Dutka, eds.), pp. 21-31.American Society for Testing and Materials, Washington, D.C. Levin, G. V., Harrison, V. R.,Hess, W. C., and Gurney, H. C. (1956).A radioisotope technique for the rapid detection of coliform organism. /. Publ. Health. 46, 1405-1414. Levin, G. V., Stauss, V. L., and Hess, W. C. (1961).Rapid coliform organism determination with C14.J . Water Pollut. Control Fed. 33, 1021-1037.

182

A. M. CUNDELL

Matsunaga, T., Karube, I., and Suzuki, S. (1979). Electrode system for the determination of microbial populations. Appl. Enuiron. Microbiol. 37, 117-121. Meuzelaur, H. L. C., Kistemaker, P. G., Eshuis, W., and Engel, H. W. B. (1977). Progress in automated and computerized characterization of microorganisms by pyrolysis mass spectrometry. In “Rapid Methods and Automation in Microbiology”(H. H. Johnston and S. W. B. Newsom, eds.), pp. 230-250. Research Studies Press, Forest Grove, Oregon. Moran, J. W., and Witter, L. D. (1976). An automated rapid method for measuring fecal pollution. Water Sewage Works, May. Murtaugh, J. L., and Bunch, R. L. (1967). Sterols as a measure of fecal pollution. J. Water Pollut. Control Fed. 39, 376389. Newman, J. S., and O’Brien, R. J. (1975). Gas chromatographic presumptive test for coliform bacteria in water. Appl. Microbiol. 30, 584-588. Oleniacz, W. S., Pisano, M. A., Rosenfeld, M. H., and Elgart, R. L. (1968). Chemiluminescent method for detecting microorganisms in water. Enuiron. Sci. Technol. 2, 1030-1033. Picciolo, G. L., Thomas, R. R., Chappelle, E. W., Taylor, R. E., Jeffers, E. J., and McGany, M. A. (1977). In “Rapid Methods and Automation in Microbiology” (H. H. Johnston and S. W. B. Newsom, eds.), pp. 294-299. Research Studies Press, Forest Grove, Oregon. Picciolo, G. L., Chappelle, E. W., Thomas, R. R., and McGarry, M. A. (1978). Performance characteristics of a new photometer with a moving filter tape for luminescence assay. Appl. Enuiron. Microbiol. 34, 720-724. Pugsley, A. P., Evison, L. M., and James, A. (1973). A simple technique for the differentiation of E. coli in water examination. Water Res. 7, 1431-1437. Reasoner, D. J., and Geldreich, E. E. (1978). Rapid detection ofwater borne fecal coliforms by 14 COz release. In “Mechanizing Microbiology” (A. N. Sharpe and D. S. Clarke, eds.), pp. 120-129. Thomas, Springfield, Illinois. Reasoner, D. J., Blannon, J. C., and Geldreich, E. E. (1979). Rapid seven-hour fecal coliform test. Appl. Enuiron. Microbiol. 38, 229-236. Rotman, B. (1961). Measurement of activity of simple molecules of a j3-D-gdactosidase. Proc. Natl. Acad. Sci. U.S.A. 47, 1981-1991. Rotman, B., and Papermaster, B. W. (1966). Membrane properties ofliving mammalian cells as studied by enzymatic hydrolysis of fluoregenic esters. Proc. Natl. Acad. Sci. U.S.A. 55, 134. Silverman, M. P., and Munoz, E. F. (1979). Automated electrical impedance technique for rapid enumeration of fecal coliforms in effluents from sewage treatment plants. Appl. Enuiron. Microbiol. 37, 521-526. Skogen-Hagenson, M. J. (1976). A high efficiency flow microfluorometer: Application to bacterial fluorescence. M.S.Dissertation. Iowa State University, Ames, Iowa. Smith, L. L., and Gouron, R. E. (1969).Sterol metabolism. VI. Detection of5j3-choleston-3~-01 in polluted waters. Water Res. 3, 141-148. Trinel, P. A., and Leclerc, H. (1972). Automatisation de l’analyse bacteriogique de l’eau. I. Etude d u n nouveau test specifique de contamination fwale et des conditions optimales d e sa mise en evidence. Water Res. 6 , 1445-1458. Ulitzur, S., Yagen, B., and Rottem, S. (1979). Determination of lipopolysaccharide by a bioluminescence technique. Appl. Enuiron. Microbiol. 37, 782-784. Warren, L. S., Benoit, R. E., and Jessee, J. A. (1978). Rapid enumeration of fecal coliforms in water by a colorimetric j3-galactosidase assay. Appl. Enuiron. Microbiol. 35, 136-141. Waters, J. R. (1972). Sensitivity of the 14 COz radiometric method for bacterial detection. Appl. Microbiol. 23, 198-199. Whitaker, J., Page, R. H., Stulberg, C. S.,and Zuelzer, W. W. (1958). Rapid identification of

RAPID COUNTING METHODS FOR COLIFORM BACTERIA

183

enteropathogenic Escherichia coli 0127:B by the fluorescent antibody technique. AMA J.

Dis.Child. 95, 1-8. Wilkins, J. R., and Boykin, E. H. (1976). Electrochemical method for early detection and monitoring of coliform in water. J. Am. Water Works Assoc. 68, 257-263. Wilkins, J. R., Stoner, G . E., and Boykin, E. H. (1974). Microbial detection method based on sensing molecular hydrogen. Appl. Microbiol. 27, 949-952. Wilkins, J. R., Yound, R. N., and Boykin, E. H. (1978).Multichannel electrochemical microbial detection unit. Appl. Enoiron. Microbiol. 35, 214-215. Wun, G. K., Walker, R. W., and Litsky, W. (1978).An improved polystyrene polymeric XAD-2 resin column extraction of 5P-cholestan3P-01 from polluted water. Water Res. 15, 67-71.

This Page Intentionally Left Blank

Training in Microbiology at Indiana University-Bloomington' L. S. MCCLUNC Department of Biology, Indiana University, Bloomington, Indiana

I. Introduction

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

185

IV. Faculty and Research Appendix: List of Ph.D. Recipients, 1949-1979. . . . . . . . . . . . . ....................................... References

201 205

I. Introduction The main campus of Indiana University, the oldest state university west of the Allegheny Mountains, is located in Bloomington, Indiana, the county seat of Monroe County (Clark, 1976a, 1976b, 1977; Myers, 1951; Wylie, 1890). Although the Medical School originated on this campus, the training was transferred to Indianapolis about 1908 (Myers, 1956)and there now are located the Schools of Dentistry, Medicine, and Nursing (Indiana University Medical Center on West Michigan Street), together, in late years, with a rapidly developing general undergraduate program in cooperation with Purdue University (IUPUI campus). The latter institution, a land grant university at West Lafayette, over the years has developed strong Schools (colleges) of Agriculture and Engineering. The Indiana University system now has, in addition to the programs in Bloomington and Indianapolis, several regional campuses scattered in various population centers throughout the state. Only introductory courses in bacteriology (microbiology) have been introduced at these centers. It is surprising that a curriculum in microbiology was established in 1940 at Indiana University-Bloomington, since the campus lacked programs in agriculture, dairy and food sciences, medicine, and public health, which traditionally have given rise to strong programs in microbiology as at Wisconsin (Sarles, 1973) and in other states (Clark, 1961).Although there had been earlier sporadic offerings of a course in bacteriology on the Bloomington campus, as noted later, no attempt was made to initiate a full curriculum until 1940, with the appointment in May of the author to the faculty of the College of Arts and Sciences. 'This article is dedicated to the memory of Dr. David Perlman, who requested that it be prepared. 185 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 27 Copyright @ 1981 by Academic Press, Inc. All rights of reproduction in any form reserved. ISBN 0-12-002627-9

186

L. S. MCCLUNG

The action was due in part to the request of the Home Economics faculty for the regular offering of a one-semester course primarily for dietetics majors. The appointment charge was made not only to offer such a course but also to develop a curriculum in the College to lead to the establishment of a department comparable to those of the sister sciences of Botany (chairman, Ralph E. Cleland) and Zoology (chairman, Fernandus Payne), which were, by then, strong departments (Campaigne, 1968;Torrey, 1940). Payne, then also Dean of the Graduate School, was an influential advisor to the new President, Herman B Wells, who was attempting to build a strong faculty on the Bloomington campus. The initiation of work in bacteriology was a bold move, since the majority of the strong departments in this subject area in other United States universities were located on campuses with agriculture or medical programs. The year chosen for the introduction of offerings in bacteriology was fortunate, because it was at the beginning of the remarkable era of development of the university under the energetic leadership of the new President. Even though the imminent World War I1 years occasioned a great reduction in the total student body, the enrollment in the various courses in bacteriology, initially much higher than predicted, continued to grow. With the expansion of the total campus curricular offerings through the efforts of the new President, bacteriology (as it was then called) grew and soon reached separate departmental status. Rather than create a one-man department in 1940, by mutual consent the name of the Department of Botany was changed to Botany and Bacteriology and the initial and early faculty appointments were assigned to the combined department until July 1946, when the two units were separated and a Department of Bacteriology was created, with the author as Chairman. The budgets had been separate from the beginning, so that the departmental separation brought little change in administrative duties. The name of the department was retained until 1966, when it was changed to Department of Microbiology with the appointment of Howard Gest as Chairman to succeed McClung. In the previous year, the Departments of Bacteriology, Botany, and Zoology ( and initially Anatomy and Physiology)had “federated’ to form a Division of Biological Sciences, modeled after a similar pattern at the University of Illinois, with each department remaining an intact unit with its own chairman. The first Director of the new division was Frank Putnam, and McClung assumed (for a 3-year term) the duties of Assistant Director. The Divisional status was retained, with C.H.W. Hirs succeeding Putnam as Director after 3 years, until 1977, when the Division was replaced with a single Department of Biology and previous departmental curricula and degrees changed to programs. The Chairman of the new unified department from 1977 to 1979was John Preer, a protozologist. He was succeeded in 1979 by Gary Sojka, a microbiologist. The following account will emphasize various aspects of the

MICROBIOLOGY AT INDIANA UNIVERSITY

187

bacteriology-microbiology program during the period 1940-1966, during which the author served as Chairman, with brief notes relative to the early period and to the post-1966 changes. II. The Early Period A previous review (McClung, 1941) reveals that the first course in bacteriology on the Bloomington campus was offered by Robert E. Lyons in the Department of Chemistry, about 1896. Beginning in 1892, Lyons had studied bacteriology at the University of Heidelberg, in Wiesbaden, in Jorgenson’s laboratory in Copenhagen, and with Metchnikoff in Paris. The students in his course were primarily advanced undergraduate and graduate students in chemistry. A laboratory and media kitchen were available in Wylie Hall. The equipment was good for its time, having been purchased by Lyons in Europe for the most part. An independent study course was added in 18971898, and in 1902, a course in bacteriological chemistry. About 1900, additional instruction in bacteriology was added as a summer course at a lake station (initially, Turkey Lake and later, Winona Lake). In the fall of 1903, a separate division as a medical department was established on the Bloomington campus, but soon this was transferred to Indianapolis (Myers, 1956). Wilfred H. Manwaring, later to be a member of the faculty at Stanford University and a well-known immunologist, was in charge of the bacteriology course in the medical program, aided in 1907 by Samuel W. Famulener. Students during this early era who achieved bacteriological fame include Leo F. Rettger, later of Yale University; James N. Currie, later of the Charles Pfizer Company, where he perfected a commercial process for production of citric acid; and I. M. Lewis, who became Professor in charge of bacteriology at the University of Texas.

111. The Modern Period As noted above, the author was appointed in 1940 to the College faculty to establish a curriculum in bacteriology. As Assistant Professor, he offered two introductory courses during the first year; the first was a two-semester lecture-laboratory course for science majors, and the second, in the spring semester, was a brief course intended for home economics majors, although the majority of the enrollment consisted of majors from the other sciences. In both courses, the enrollment greatly exceeded the prediction of about 10-12 students; the major course enrolled 25 and the brief one, 59. Since the available laboratory room had only 24 stations, a third laboratory section was needed for the brief course, much to the amazement of the Dean (a professor of Latin). This third section occasioned the necessity (and the leverage) for additional faculty, and in the next year (1941-1942) John S. Sylvester, a

188

L. S. MCCLUNG

Ph.D. candidate at Wisconsin, was added and was assigned to the brief course. Since the two-semester introductory course introduced some problems in the student’s schedules, in 1941-1942, it was divided into separate courses, with the second semester titled “Applied Microbiology.” The enrollment in this introductory course continued to climb in the early years until it reached 145 in 1948, and by then the course was being offered each semester and in the summer session; it later expanded to 345 per semester. The course number changed over the years and various members of the faculty took part in its development including: Bard, Brock, Gest, Poindexter, Ramaley, Rickenberg, and particularly Konetzka. In 1970, the traditional laboratory for the course was changed to an audiotutorial system designed by Konetzka. In 1941-1942, a course in “Bacteriological Technique” was added, as was an independent “Advanced Problems” course with variable credit. The technique course was expanded and the title changed to “Diagnostic Microbiology” by Weinberg. Also initiated about this time was a “Seminar” (for advanced undergraduates and graduates) and a “Research” course (for graduate majors) for thesis research. The next major addition to the curriculum (1946-1947) was a course in general “Virology,” and this was offered by S. E. Luria, who had been appointed Instructor to succeed Sylvester upon his return to Wisconsin to complete his doctorate program. Luria was first assigned to the brief introductory course and the course in technique. Although at the time, courses in medical virology were rare at other institutions (and there were none in any undergraduate program), Luria’s broad interest and training offered a unique opportunity to present basic information on bacterial viruses as well as those of interest in animal and plant pathology. With varying instructors (Weiss, Fraser, Taylor) through the years, and for a period of time with the title changed to include the rickettsia, the course has remained in the curriculum to the present. The next major curriculum addition reflected the desire to include the van Nielian biology of bacteria type of training. With the appointment held open for a year to permit him to accept a Guggenheim Fellowship for study with Pringsheim at Cambridge, this type of training was offered by Roger Y. Stanier in 1946-1947, who also offered the first strictly graduate course, “Microbial Biochemistry.” The undergraduate course was probably initially titled “Determinative Bacteriology,” and, with varying titles, it has remained in the curriculum; other instructors include Bard, McClung, Stokes, Brock, and Hegeman. With the title “Biology of the Prokaryotes” at present, it is required for the undergraduate major. During the period when Luria and Gunsalus were on the faculty, a new course was developed by them and titled, for want of a more descriptive name, “Advanced General Bacteriology.” This course included advanced

MICROBIOLOGY AT INDIANA UNIVERSITY

189

material on bacterial morphology and cytology as well as aspects of the physiology of bacteria. Later, a portion of the material from this course was used to develop a new course, “Microbial Physiology and Biochemistry”; several of the faculty have been involved with the course, which has always been required. Introducing a required course in advanced chemistry. an innovation for a major in a program in undergraduate biology, was accomplished by making quantitative and organic chemistry prerequisites to the required physiology course. This was then the only mechanism for a requirement for a nondepartmental course. With the appointment of Emilio Weiss to succeed Luria, it was possible to introduce a course in “Immunology.” This was a course in fundamental immunology, with laboratory rather than applied serology. Weiss was succeeded by W. Dean Fraser, who taught both virology and immunology for a number of years. Later, the immunology course was dropped when Fraser assumed other duties. It has been reintroduced by J. R. Preer in 1980-1981. “Medical (or Pathogenic) Bacteriology” was introduced by Weinberg in 1954. Originally, this was a combined lecture-laboratory course, but later, the laboratory credit was separated. Konetzka, has recently developed a special laboratory for the course. The lecture course has continued to grow and now rivals the introductory course in size. The newest course to be added to the curriculum is “Environmental Microbiology,” which has been developed by Konetzka in recent years, though a course by the same title was taught by Koch for a few years. A one-credit course, “Bacteriological Literature,” was introduced by McClung as early as 1949. It has remained in the curriculum as one of the requirements for the B.S. degree and has recently been renamed “Biomedical Sciences Documentation.” This course seeks to give the student training in the efficient use of current and old library resource materials. An additional one-credit course in “History of Microbiology” by Konetzka was offered for a few years from about 1958, but has been withdrawn from the curriculum. “Mycology” has not been offered by the Department of Bacteriology (Microbiology) since an introductory (and some years advanced) course was offered by the Department of Botany (Plant Sciences). Early instructors included Marion Lohman, John R. Raper, Harold J. Brodie, Robert Johns, Stanley Dick, and, for the last several years, Michael Tansey. Similarly, a course in “Protozoology”was available in the Department of Zoology, with T.M. Sonneborn as instructor until lately, when Ruth Dippell has assumed responsibility for the course on the retirement of Sonneborn. The first person to complete the A.B. degree requirements in bacteriology was Cornelius F. Sterling; this was in 1942. By 1966, the number increased to 174, and by 1979, to 392. The latter figure includes the recipients of both the A.B. and B.S. degrees. The first B.S. degree was conferred in 1974 and

190

L . S. MCCLUNG

since that time, the majority of the undergraduates have opted for this program, which includes a more strict list of course in the major and the cognate sciences. From the beginning, many of the undergraduate major continued their training to the M.A. andlor the Ph.D. in microbiology, with a number completing the M.D. or D.D.S. degrees. A good number of the early graduates are now faculty members at various institutions scattered throughout the country from Maine to California. Those who did not continue with graduate training have had success in careers in clinical or industrial microbiology. As noted earlier, the first strictly graduate course “Microbial Biochemistry,” was introduced by Stanier in 1946. On a visiting basis, Carl Robinow had presented a graduate course in bacterial cytology in 1946. The biochemistry course remained in the curriculum for a few years, with others participating in the instruction, but was withdrawn about the time some of the faculty of the department cooperated with members €rom Botany and Zoology to offer a new, broader-based course, “Molecular Aspects of Biology.” In all versions, the laboratory was scheduled for the entire day on Saturday. The microbiology faculty has generally held that training in research rather than a proliferation of courses should dominate in graduate programs in the department. Hence, relatively few strictly graduate courses have been initiated. “Medical Microbiology” was introduced by Brock for medical and graduate microbiology majors in 1960-1961, when a special combined degree (medicine and speciality area research) program was added to the offeringson the Bloomington campus in addition to the standard M , D. curriculum on the medical school campus. With some success, a “rotating” course, “Recent Advances in Microbiology,” has been used to present certain basic advanced material on a rotating schedule (by various instructors) and also for a single topic course, taught usually by a visiting faculty member. In late years, Gest has introduced “Prokaryotic Physiology,” and White has offered “Microbiol Development.” The first M.A. degrees in bacteriology were conferred in 1943 on Raymond N. Doetsch and Allen P. Saunders. By 1966, 73 individuals had completed the M.A. requirements, and by 1979, the total was 118. One of the graduates, Robert F. Acker, who finished the Ph.D. degree elsewhere, now serves as the Executive Director of the American Society for Microbiology. The numbers of M.A. students for whom the faculty served as advisors were: Bard-3, Blumenthd-2, Brock-4, Fraser-7, Gest-2, Gunsalus-3, H e g e m a n 3 , Koch-5, Konetzka-11, Luria-5, McClung-27, Poindexter-1, Ramaley-2, Repaske-1, Rickenberg-1, Sojka-2, Stokes-1, Taylor-6, Weinberg-25, Weiss-2, White-3, and faculty in other departments-2. The first to complete the Ph.D. requirements was Raymond C. Bard, in 1949. By 1966, 31 individuals had completed the requirements, and by 1979,

MICROBIOLOGY AT INDIANA UNIVERSITY

191

the number was 73. Many of the Ph.D. graduates have assumed teaching positions on the faculties of various universities or important positions in federal research laboratories. Only one will be mentioned by name-Willis A. Wood, now Chairman of the Department of Biochemistry at Michigan State University and the 1980 President of the American Society for Microbiology. Another illustrious graduate of Indiana University, James D. Watson of DNA structure fame, technically filled the Ph.D. requirements in the Department of Zoology, although Professor Luria in Bacteriology served as the major professor and supervised the thesis work in Luria’s quarters in Kirkwood Hall. A list of those awarded the Ph.D. degree, together with the thesis title and the name of the advisor, is given in the Appendix. From the beginning, aid to graduate students in the form of teaching assistantships (now called associate instructors) was available. Since the total departmental enrollment in laboratory courses was not huge, only two such assistantships existed initially; there are now about 12 per year. This aid was insufficient to support all who were admitted to the graduate program so it became necessary to seek additional funds for this purpose. Fortunately, just as the problem was becoming critical in our program, the National Institutes of Health announced a new program of training grants. The department immediately made application; when approved, it had one of the first such programs in microbiology. Beginning in 1959 for a 5-year period, with W. Dean Fraser designated as the Principal Investigator, the program, with successive renewals, continued until the mid-l970s, when the NIH predoctoral training grant program was phased out. From the grant, funds were available to support about 12 predoctoral and one or two postdoctoral students each year. These funds also permitted the department to bring in 12 or more distinguished seminar speakers per year. A limited amount of support, often for summer fellowships for those on teaching assistantships during the academic year, was available from the Bayard Floyd Fund, a designated bequest to the Indiana University Foundation from an early major in the biological sciences.

IV. Faculty and Research Over the years, the faculty has changed and expanded in number. Rather than include a chronological list of the changes, the following list includes the major appointments, together with their teaching assignments, their educational backgrounds, and, when pertinent, their post-Indiana University (IU) appointments. The earlier faculty appointments, as well as those of the modern group, brought to the program a wealth of experience and personal achievement. At one time, all members of the faculty who were elegible to apply had held Guggenheim Fellowships. The individuals are listed here in alphabetical order.

192

L. S. McCLUNC

BARD,RAYMONDC. (B.S., 1938, College of the City of New York; M.A., 1947, Ph. D., 1949, Indiana University) At ZU: Instructor, 1949-1951; Assistant Professor, 1951-1953. Teaching: Introductory Bacteriology, Bacteriological Technique, Advanced General Bacteriology. Research: bacterial metabolism. Post-ZU: Head, Microbiology Section, Research and Development Division; Associate Director of Research and Development; Vice President and Director of Research, National Drug Company, Philadelphia (1953-1962). Director of Research, College of Dentistry; Assistant Vice President for Research; Executive Director, University of Kentucky Research Foundation; Professor of Cell Biology (1962-1967). Professor of Cell and Molecular Biology and Dean, School of Allied Health Sciences, Medical College of Georgia, Augusta, Georgia (1967-present). BLUMENTHAL, THOMAS (A. B., 1966, Antioch College; Ph. D., 1970, Johns Hopkins University) A t ZU: Assistant Professor, 1973-1977; Associate Professor, 1977-present. Teaching: Molecular Aspects of Biology, Microbial Genetics. Research: genetics of bacteria and bacterial viruses. Guggenheim Fellowship, 1980; Fogarty Fellowship, 1980.

BROCK,THOMASD. (B.S., 1949; M.S., 1950, Ph.D. 1952, Ohio State University) At ZU: Associate Professor, 1960-1964; Professor, 19641970. Teaching: Medical Microbiology, Biology of Bacteria. Research: mechanism of action of antibiotics, amino acid transport systems, bacteriocines and bacteriophages, yeast genetics, ecology of bacteria (particularly in high thermal environments). Post-ZU: E. B. Fred Professor, Department of Bacteriology, University of Wisconsin. NIH Career Development Award, 1980. FRASER,W. DEAN(B.S. Harvard University, 1938; M.S., 1939, Ph.D., 1941, University of Illinois) At ZU:Associate Professor, 1955-1960; Professor, 1960-present; Chairman, Department of Microbiology, 19701976. Teaching: Virology, Immunology, and Introductory Biology, Strategy of Life (nonmajor course). Research: morphology, molecular biology and genetics of bacterial viruses, and biology of mycoplasma. GEST, HOWARD(A.B., 1943, University of California-Los Angeles; Ph.D., 1949, Washington University) At IU: Professor and Chairman, Department of Microbiology, 1966-1970; Professor, 1970-1979; Distinguished Professor, 1979-present. Guggenheim Fellowships, 1970, 1979. Teaching: Introductory Microbiology, Biology of Bacteria Laboratory, Prokaryote Physiology, and a large group of postdoctorates. Research: comparative biochemistry of photosynthetic processes.

MICROBIOLOGY AT INDIANA UNIVERSITY

193

GUNSALUS, IRWINC. (B.S., 1935; M.S., 1937, Ph.D. 1940, Cornell University) A t ZU: Professor, 1947-1959. Teaching: Advanced General Bacteriology, Microbial Biochemistry. Research: microbial biochemistry. Post-ZU: Professor and Head of Department (1955-1966) of Biochemistry, University of Illinois. Guggenheim Fellowships, 1949, 1959, 1968. Member National Academy of Sciences. HEGEMAN, GEORGED. (A.B., 1960, Harvard University; Ph.D., 1965, University of California-Berkeley) A t ZU: Associate Professor, 19721978; Professor, 1978-present. Teaching: Biology of Prokaryotes. Research: microbial biochemistry; microbial metabolism, regulation, and genetics.

KOCH, ARTHURL. (B.S . , 1948, California Institute of Technology; Ph. D. 1951, University of Chicago). A t ZU: Professor, 1967-present. Teaching: Environmental Microbiology, Introductory Biology. Research: microbial physiology and metabolism, prokaryote evolution, biophysical aspects of micrabiology. KONETZKA, WALTER A. (A.B., 1950, M.S., 1952, Ph.D., 1954, University of Maryland) A t ZU: Assistant Professor, 1955-1959; Associate Professor, 1959-1963; Professor, 1963-present. Teaching: General Microbiology of Bacteria, Applied Microbiology, History of Bacteriology, Environmental Microbiology, Medical Microbiology Laboratory. Research: microbiology of lignans, bacterial nutrition, antimicrobial agents, magnetotatic bacteria. Distinguished Teaching Award, IU, 1965; ASM Carski Award for Teaching, 1970. Senior Teaching Awards, 1970, 1974, 1978. Post-ZU: University of Maryland-Baltimore County Campus, 1966-1968.

LURLA,SALVADOR E. (M.D., 1935, University of Torino, Italy) A t ZU: Instructor, 1943-1945: Assistant Professor, 1945-1947: Associate Professor, 1947-1950. Teaching: Bacteriological Technique, Virology. Research: genetics of bacteria and bacterial viruses. Post ZU: Professor of Bacteriology, University of Illinois, 1950-1959; Professor of Microbiology and Sedgwick Professor of Biology, Institute Professor, Massachusetts Institute of Technology, 1959-1978. From 1972, Director of Cancer Research. Nobel Prize, 1969;Lenghi Prize, Accademia dei Lincei, 1965. Honorary doctorates: Indiana University and Rutgers University, 1970; Providence College, 1972; Brown University, 1973; University of Palermo, 1973. Member National Academy of Sciences, American Philosophical Society; President, American Society for Microbiology, 1967-1968.

194

L. S. McCLUNG

MCCLUNG,LELANDS. (A.B., 1931, M.A., 1932, University of Texas; Ph.D., 1934, University of Wisconsin) At ZU: Assistant Professor, 1940-1944; Associate Professor, 194.4-1948; Professor, 1948-present; Chairman, Department of Bacteriology, 1946-1966; Assistant Director, Division of Biological Sciences, 1965-1968. Teaching: General Bacteriology, Applied Bacteriology, Determinative Bacteriology, Biomedical Sciences Documentation, Man and Microorganisms (nonmajor course). Research: taxonomy of Clostridium, Serratia, Aeromonas; ecology of clostridia; bacteriophagy of Clostridium p h i n g e n s ; food microbiology, particularly C . perfiingem as agent of food poisoning; history of microbiology. Films on introductory biology and microbiology. Director of NSF Summer Institutes in Microbiology for High School Teachers, Delegate to Binational Conferences on Biological Education (Japan, 1966; India, 1971). Guggenheim Fellowship, 1940. POINDEXTER, JEANNESTOVE (A.B., 1958, Indiana University: M.A., 1961, Ph. D., 1963, University of California-Berkeley) At ZU: Assistant Professor, 1964-1967. Teaching: Introductory Microbiology. Research: microbial ecology and taxonomy. Post-ZU: City College of New York, Public Health Research Institute of New York. POLISKY, BARRYA. (BA., 1961, University of Chicago; Ph.D. 1973, University of Colorado) At ZU: Assistant Professor, 1977-present. Teaching: Molecular Biology and Genetics. Research: expression of eukaryotic DNA in bacteria. NIH Career Development Award, 1980. RAMALEY, ROBERT F. (A.B., 1959, M.S., 1962, Ohio State University; Ph.D. 1962, University of Minnesota) At ZU: Assistant Professor of Microbiology, 1966-1972. Teaching: Introductory Microbiology, Molecular Aspects of Biology Laboratory. Research: microbial biochemistry. Post-ZU: Associate Professor, Department of Microbiology, Creighton University. REPA~KE, ROY (B.S., 1948, Western Reserve University; M.S., 1950, University of Michigan: Ph.D. 1954, University of Wisconsin) At ZU: Instructor, 1953-1955; Assistant Professor, 1955-1959. Teaching: General Bacteriology, Microbial Metabolism. Research: carbon dioxide metabolism, mechanism of oxidative metabolism. Post-ZU: Laboratory of Microbiology, National Institute of Allergy and Infectious Diseases. RICKENBERG,HOWARD V. (B. S., 1950, Cornell University; Ph. D., 1954, Yale University) At ZU: Associate Professor 1961-1963; Professor, 1963-1968. Teaching: Microbial Physiology: Molecular Aspects of Biology. Research: regulation of enzyme formation. Post-ZU: Director, Division of Molecular and Cellular Biology, National Jewish Hospital

MICROBIOLOGY AT INDIANA UNIVERSITY

195

and Research Center; Professor, University of Colorado School of Medicine. Editor, Microbiological Reviews, 1979-present. CARLFWZ (M.D., 1935, University of Hamburg) At IU: Visiting Professor, 1946-1947. Teaching: Cytology of Bacteria. Research: cytological studies on genus Bacillus. Post-IU: Faculty, University of Western Ontario.

ROBINOW,

SOJKA,GARY(A.B., 1962, Coe College: M.S., 1965, Ph.D., 1967, Purdue University) At IU: Research Associate, 1967-1969; Assistant Professor, 1969-1973; Associate Professor, 1973-1979; Professor, 1979-present; Chairman, Department of Biology, 1979-present. Teaching: Microbial Cell Biology, Physiology and Biochemistry of Microorganisms, Introductory Biology. Research: metabolic mechanisms and genetics of photoheterotrophic photosynthetic bacteria. 1U Award for Distinguished Teaching, 1977.

ROGER YATE (B.A., 1936, University of British Columbia; M.A., 1940, University of California-Los Angeles; Ph.D., 1942, Stanford University) At IU: Assistant Professor, 1946-1947. Teaching: Biology of Bacteria, Microbial Biochemistry. Research: biology of the myxobacteria, bacterial metabolism, metabolic pathways. Post-IU: University of California-Berkeley, 1947-1971; Institut Pasteur 1971-1980. Honors: Eli Lilly Award, 1950;American Academy of Arts and Sciences (Boston), 1955; Emil Christian Hansen Award, 1976; Chevalier de la Legion d’Honneur, 1977; Fellow of the Royal Society, 1978. Honorary Doctorates: Reims, 1973;Chicago, 1978. Member National Academy of Sciences, 1979; Honorary Member, American Society for Microbiology, 1979.

STANIER,

STOKES, JACOB LEO(B.S., 1934, Rutgers University; M.S.,

1936, University of Kentucky; Ph.D., 1939, Rutgers University) A t IU: Associate Professor, 1950-1953. Teaching: Determinative Bacteriology, Microbial Biochemistry. Research: microbial ecology and biochemistry. Post-ZU: Bacteriologist, Western Utilization Research Laboratory, USDA, 1953-1957; Professor and Chairman, Department of Bacteriology and Public Health, Washington State University, Pullman, Washington, i959-present. (B.S., 1939, M.S., 1940, Ph.D. 1943, University of Wisconsin) At IU: Instructor, 1941-1942. Teaching: Introductory Bacteriology. Research: bacteriophagy of butyl alcohol organisms. Post-IU: Research bacteriologist, then Director of Experimental Biology Division, Abbott Laboratories, North Chicago, Illinois. Chairman, Inter-

SYLVESTER, JOHN C.

196

L. S. McCLUNG

science Conference on Antimicrobial Agents and Chemotherapy, 1962; American Academy of Microbiology, Board of Governors, 1963-1965. TAYLOR, MILTON (B.S., 1961, Cornell University; Ph.D., 1966, Stanford University. At IU: Assistant Professor, 1967-1970; Associate Professor, 1970-1976; Professor, 1976-present. Teaching: Virology. Research: viral regression of tumors, viral RNA synthesis, molecular virology. WEINBERG, EUGENED. (B.S., 1942, University of Chicago; U.S. Army 1942-1947; M.S., 1948, Ph.D., 1950, University of Chicago) A t IU: Instructor, 1950-1953; Assistant Professor, 1953-1957; Associate Professor, 1957-1960; Professor, 1961-present. Interim Chairman, 19761977; Associate Dean, Research and Graduate Development, 19781980. Teaching: Diagnostic Microbiology, Medical Microbiology. Research: trace metal effects on antimicrobial agents, synthesis and roles of secondary metabolites, effects of metals on infectious processes. IU Award for Distinguished Teaching, 1974. WEISS, EMILIO(A.B., 1941, University of Kansas; M.S., 1942, Ph.D., 1948, University of Chicago) At IU: Assistant Professor, 1950-1953. Teaching: Virology and Immunology. Research: virology. Post-IU: Rickettsia Division, Chemical Corps Biological Laboratories, Camp Detrichu, 1953-1954; Assistant Head, Virology Division; Deputy Director, Department of Microbiology; Chairman, Department of Microbiology, Naval Medical Research Institute, 1954-present. Also Professor, Department of Preventive Medicine and Biometrics, Uniformed Services, University of the Health Sciences, 1977-present. WELSHIMER, HERBERTJ. (B.S., 1943, Ph.D. 1947, Ohio State University) At IU: Instructor, 1947-1949. Teaching: Bacteriological Techniques, Introductory Bacteriology. Research: cytology of bacteria. Post-IU: Faculty Department of Bacteriology, Medical College of Virginia. WHITE, DAVID(A.B., 1958, Ph.D., 1965, Brandeis University) A t IU: Assistant Professor, 1967-1973; Associate Professor, 1973-present. Teaching: Introductory Microbiology, Developmental Biology, Cell Biology, Bacterial Physiology. Research: developmental microbiology, bacterial physiology, biology of myxobacteria. Almost all of the faculty have been successful in securing from various sources (e.g., NSF, NIH, USDA, AEC) grants of some magnitude to support their research programs. In addition to moneys for equipment, consumable

MICROBIOLOGY AT INDIANA UNIVERSITY

197

supplies, and full time assistants or associates, many of these grants have been sufficient to support one or more graduate students who serve as research assistant. Traditionally, the teaching load for a new faculty member is purposely kept light during at least the first two semesters in order to encourage and permit the initiation of a vigorous research program. A steady flow of publications is expected of those who are considered for raise in academic rank. Space and time do not permit the listing of individual research projects. This is regretted because a number of fundamental discoveries that have opened new areas for investigation were made in the faculty research programs.

V. Physical facilities No separately designated facilities for bacteriology existed on the Bloomington campus in 1940, and no space was available in the Biology Building, which housed Botany and Zoology. To accomodate the initial phases of the new curriculum, three rooms on the west end of the third floor of the Chemistry Building were made available. By the start of the first semester, the rooms had been cleared of the former occupants but the promised changes for bacteriology had not been initiated; the first laboratory session actually took place in November. One of the rooms was sufficiently large to accommodate tables for 24 student stations, leaving in place a long chemical bench that could be used for demonstrations and for advanced students working during laboratory hours of the introductory course(s).Vertical to the remaining long wall were four smaller rooms. These were converted to 37°C and room temperature incubator rooms, a storeroom, and a media kitchen complete with autoclave, Arnold sterilizer, and a small mechanical dishwasher. The storeroom soon became inadequate and items were placed in corridor cabinets and atop an additional walk-in incubator with shaker for advanced problem and faculty research on X-ray mutants of Penicillium for penicillin production. One of the other smaller rooms had been partitioned; the front of this was used as the department (and faculty) office and the remaining portion as a faculty (McClung) laboratory, since it already had a chemical bench. Since only 6 V electricity was available in the room, some changes were necessary. The remaining room of similar size included a long chemical bench (again with 6 V electricity)and was hence designated for graduate students, and, in the second year, for a new faculty member. Very soon a portion of this room had to be used as a small enclosure for animal quarters-hardly pretentious for a future Nobel Prize winner (Luria) in his first permanent United States position. During the war years, the enrollment in bacteriology increased although it declined in the other sciences; it therefore became imperative to move to

198

L. S. McCLUNG

larger quarters since expansion within the Chemistry Building was not possible. In searching for space-any reasonably unoccupied area on the campus that could be converted to useful laboratory space-during a fourth of July (1946) sleuthing expedition (when most buildings would be unoccupied), a possible space was located, although it had several undesirable features. This was the attic fourth floor of the language building, Kirkwood Hall, in which there was no elevator, and, in the attic, no water, minimal electricy, and of course no steam line. It did have the redeeming feature of space, however-two large rooms plus two other fairly large rooms and two or three smaller ones. It was then occupied almost solely by a retired professor who used one of the rooms for office space and cluttered the remainder with plaster busts of Cicero, and sundry other discards, including the stage for the university theater. Since all other space located that day had even more objectional features, the Administration approved the choice (the department offered as first choice a large, three-story frame house, with full private knowledge that it would not be available, since it was already scheduled to be renovated for use as the first dormitory for Black women, who then were not eligible for assignment in the only Women’s Residence Quadrangle). Soon after approval by the Administration, plans were developed for the conversion of the Kirkwood attic space. One of the large rooms could accommodate the existing teaching laboratory tables (and another allowing expansion to 32 students per section)for the introductory course(s) and still leave space for a walk-in room temperature incubator (an old cold storage unit) and storage cabinets for precision glassware and chemicals. Space for 37°C incubation was created by opening up sections of under-the-eaves areas and even a small laboratory for the technician was possible, for quality testing ofwater and milk purchased by the university dormitory and union kitchens. In the Chemistry Building days, these tests had been done on a half-time basis by the half-time department secretary. Now the work load for the secretarial duties required a full-time person, so the bacteriological testing was assigned to another half-time appointee, who served as the stock culture curator. A small room at the comer of the teaching laboratory served as the media kitchen (with the autoclaves in the student laboratory). Adjacent to this room, and connecting to a smaller glassware-washing and sterilization area, was a room without windows but already provided with floor-to-ceiling (20 foot!) shelving; this was used for storage of sterile glassware and bulk supplies. The second of the larger rooms was divided by an 8-foot partition; one side was used as a laboratory for advanced courses, and the other as a research laboratory for graduate students. Office-study areas for these students were provided again by opening up under-the-eaves areas with portions being

MICROBIOLOGY AT INDIANA UNIVERSITY

199

reserved for a 37°C incubation and a “sterile” transfer area-sterile by virtue of a shielded ultraviolet light near the ceiling. When such areas were floored and desks placed against the slanted roof, it was still possible to stand in front of the desks-not ideal, but suitable. The two larger of the additional rooms were outfitted with chemical bench, hoods, 30-im-high tables, and an enclosed office cubicle, and were used by the faculty. One of the other rooms was used for several years as the animal quarters (until pressure of space moved the animals to a thenavailable quonset hut, about 1949), and another, quite a bit smaller, as an office-laboratory for an instructor. It is recalled that the conversion of this attic area for laboratory space, despite funds being available to bring in the needed electricity, water, and steam, was not without some manpower problems. This was due to the fact that the university was moving to the campus 250 trailers to serve as postwar student housing and the same workmen were needed for both projects. On a give-and-take basis, both projects were completed by the fall of 1946, and bacteriology moved to its new spacious quarters early in the fall semester. It should be mentioned that for ventilation, large fans were installed in the ceilings (about 30 feet high) of the two large rooms used for teaching; these provided the only means of temperature control during the hot summer months. For teaching, as far as possible, agar plates were poured in the early morning hours, since by the time of the afternoon laboratory session, 2 3 hours would be required for agar in petri plates to harden sufficiently to be streaked. One exception to this was provided for the laboratory work of Professor Luria; his laboratory was air-conditioned, or at least air-cooled. This laboratory also had an open under-the-eaves area for students, including James D. Watson, working with Luria. Space for the departmental secretary and an office for the chairman were provided by partitioning a section at the top of the stairs, and a small room behind this served as the chairman’s laboratory. A walk-in commercial cold storage room was placed in the area. Enrollment increases and space for additional faculty demanded expansion to the third floor by 1950. It was hence with great relief that plans were announced in 1951 for a new Biology Building that would include space for Bacteriology as well as Botany and Zoology. The new building-Jordan Hall of Biology-was dedicated and occupied in 1955. McClung was a member of the planning committee, thus using his early training in engineering. Built of Indiana limestone at a cost of $5,750,000, this five and one-half floor air-conditioned building contained 207,000 square feet, and Bacteriology was assigned sufficient modem space. This consisted mostly of the entire fourth floor (except for one wing that was used for animal quarters and shared with Zoology), the east wing of the third

200

L. S. McCLUNG

floor, and two laboratories on the fifth floor, together with a small greenhouse. The particular space assignment was chosen to permit the bacteriological laboratories to be as far away as possible from front lobby dust contamination and to allow trucking of glassware on a horizontal basis rather than vertically. Initially, three teaching laboratories were provided, one of which was soon converted to faculty use. Two of these were separated and commonly served by three walk-in constant-temperature rooms. One laboratory, used primarily for introductory courses, had 42 stations and the other, primarily for advanced courses, had 32 stations plus a double chemical bench and space for sterilizers and other equipment. The various faculty suites consisted of office, laboratory, and an adjacent-r across the hall-additional laboratory for graduate students. The service area (central media kitchen, glassware facility, and storeroom) was spacious and conveniently located. Additional storage area and specialty rooms were located in the basement. Separate smaller media kitchens and sterile glassware storage for graduate student use were provided on each floor, as were darkrooms, walk-in incubators, and cold storage. From extremely modest initial holdings in 1940, the bacteriology library had been brought to top quality by funds provided by a new university librarian who had given carte blanche order authority on the basis of “order what you need for a good library in your field.” Prior to 1955, the library for bacteriology had been combined with, although shelved separately from, the chemistry holdings in the Chemistry Ruilding. With the move to Jordan Hall, it was natural to combine the bacteriology volumes with the holdings of Botany and Zoology to make a unified Biology Library. Although she cannot be termed a “facility,”it seems an appropriate placesince she was so intimately concerned with the planning of bacteriology quarters in Kirkwood and Jordan and provided extraordinary service to all courses and research programs-to acknowledge the invaluable assistance of Helen Ruggles Arthur, who was first appointed on June 12, 1945, to the position of “MediaTechnician. She was a graduate of a local rural high school, without any training in science much less a degree in bacteriology. The position had just been made full time, and carried the added duties of animal caretaker in the summer when NYA students were not available. She has grown with the position during the years, and now has the title of “Administrative Assistant to the Chairman.” She quickly learned how to correctly prepare, dispense, and sterilize various media, varying in quality from simple to complex and in volume from small amounts to 100-liter batches. She learned how to order supplies and equipment for the best price advantage, to service equipment, to read blueprints, and a variety of other skills, including being “mother confessor” to the majority of the graduate majors and the faculty! ”

APPENDIX LISTOF PH.D. RECIPIENTS, 1949-1979 Number

Year

Name

Thesis title

1

1949

2 3

1950

1950

Bard, Raymond Camillo Roessler, William George Wood, Willis Avery

4

1951

DeMoss, Ralph Dean

5 6

1951 1951

Dolin, Morton Irwin Feldman, Louis Israel

7

1951

Rhuland, Lionel Edward

8

1951

Gary, Norman Dwight

9

1953

Pine, Martin Joseph

10

1954

Shankar, Kirpa

11 12 13

1956 1957

Sames, Richard William Josten, John J. Cook, Elizabeth Anne

14 15

1959 1959

Bain, William Murray Stone, Robert Louis

Transaminases: Bacterial enzymes for aliphatic, aromatic, and polyfunctional amino acids The mechanism of tryptophane synthesis by Lactobacillus arabinosus: The role of anthranilic acid The effect ofnutrition on the carbohydrate metabolism of Bacillus subtilis (Marburg) The formation of formic dehydrogenase and other adaptive enzymes by micro-organisms The effect of magnesium on the growth and metabolism of Cbscridium perfirngens BPGK Studies on the bacteriophages of Clostridium perji-ingens Oxidation and electron transport in Azotobacter extracts Studies on the influence of the metallic ion environment of the growth of yeast and on the antimicrobial activity of juglone and kojic acid An investigation of cross-resistance in Escherichia coli An examination for phage-mediated genetic transfer mechanisms in

16

1959

Wittenberger, Charles Louis

Hydrogen oxidation and electron transport in species of Hydro-

1958

A function of iron in the metabolism of Clostridium perfirngens Cbstridium spwogenes and related proteolytic anaerobic bacteria The mechanism of Damino acid formation: An alanine racemase from stfeptococcus faecalis The intermediary carbohydrate metabolism of Leuconostoc mesenteroides Studies on the pyruvate and alanine metabolism of Streptococcus faecalis

Advisor 1. C. Gunsalus L. S. McClung

I.C. Gunsalus

R. C. Bard I. C. Gunsalus I. C. Gunsalus R. C. Bard

R. C. Bard J. L. Stokes R. C. Bard

L. S. McClung R. Repaske E. D. Weinberg W. A. Konetzka W. D. Fraser

Staphylococcus aureus

R. Repaske

genomonas

(Continued)

APPENDIX (Continued) Number

Year

17

1960

Crane. Anatole

18 19 20 21

1960 1960 1961 1962 1963 1963 1963 1965 1965 1965

Hau, Charles Julius Woodings, Eric Torrance Kaufman, Bernard Smith, James Lee Berrah, Ghoulem Escobar, Mario Rene Lukemeyer, Jack W. Kessler, Donald P. Roeser, Joanne Treick, Ronald W.

30

1965 1966 1966

Vidaver, Anne M. Kopecky Boyle, James Vincent Davie, Joseph Myrten

31

1966

Malacinski, Gorge Michael

32 33 34

1967 1967 1968

Fuscaldo, Anthony Alfred Molholt, Bruce Andrew Craddall, Marjorie Ann

35

1968

Steenbergen, James F.

36

1969

37

1969

Hsie, Abraham Wuhsiung Kelly, Michael Thomas

22 23 24

D

2s 26 27

28 29

Name

Thesis title Changes in surface properties associated with cross-resistance.in Escherichia coli Properties of E. coli susceptible to the protoplast-infective agent Pi The microbial dissimilation of aromatic methyl ethers The physiology of carbohydrate formation by Azotobacter oinelandii Biochemical mechanisms of action of bacitracin The dissociation of macromolecular biosynthesis by phenethyl alcohol Serological studies of the genus Serratiu Studies on the antimicrobial properties of perphenazine Methylglumside transport and catabolic repression in E . coli Chromosome transfer and the DNA replication cycle in Escherichia coli Control of deoxyribonucleic acid synthesis in Escherichia coli by B-phenethyl almhol Characteristics of a Streptococcus faecium bacteriophage Studies on an antifungal antibiotic produced by Streptomyces sp. JM57h The molecular basis of resistance. to the bacteriocins of Streptomcus zymogenes Physiological and biochemical aspects of the bacterial oxidation of orthophosphite The radiological and physical properties of coliphage C-1 Hostcontrolled modification and restriction of T-even bacteriophages Biochemical and genetic studies of sexual agglutination in the yeast Hamenulo wingei The role of zinc and cadmium in the nutrition and physiology of TorurOpsisfennentam and certain other yeasts Aspects of glyconeogenesis and catabolic repression in Escherichia coli Physiological ecology and biochemical geography of Leucothrix mucor

Advisor W. A. Konetzka W. D. Fraser W. A. Konetzka R. Repaske E. D. Weinberg W. A. Konetzka L. S. McClung E. D. Weinberg H. V. Rickenberg W. A. Konetzka W. A. Konetzka T. D. Brock E. D. Weinberg T. D. Brock

W. A. Konetzka W. D. Fraser W. D. Fraser T. D. Brock E. D. Weinberg

H. V. Rickenberg T. D. Brock

E

38 39

1969 1970

Witmer, Heman John Bernstein, Kenneth

40 41

1970 1970

Choe, Byung-Kil Dancis, Barry Martin

42 43

1970 1970

Doemel, William Naylor Fernald, Nancy Jane

44

1970

Lee, Kuo-Yung

45 46

1970 1970

Nath, Kamalandu Noms, Thomas E E e d

47

1970

Ray, Paul Herman

48

1970

Sedmak, Gerald Victor

49

1970

Sedmak, Joseph James

50 51

1970 1970

Zeikus, Joseph Gregory

52

1970

Zilinsky, Joseph William

53 54

1972 1972 1972

Bland, Judith Ann Fliermans, Carl Bernard Johnson, Roosevelt Young

55

Wall, Thomas Randolph

Photocynamic action of proflavine on coliphage T3 Physiological and evolutionary studies on the “RC” locus of Escherichia coli Transfer RNA synthesis in mammalian cells The effectof amino acids on the total rate of RNA synthesisin E s c k h i a coli The physiological ecology of Cyanidium caZu’uriun Purification and properties of o-amidase from Bacillus subtilis 168 and Thennus aquaticus YT-1 Effect of magnesium, manganese, iron, calcium and zinc on the growth and sporulation of Bacillus megaterium Protein degradation in Escherichia coli Controls affecting RNA synthesis in a chemostat system of Escherichia coli involving different growth rates The effect of temperature on the spheroplasts of two mesophilic bacteria and on the lipid composition of Thennus aquaticus I. Biochemical and physical studies of coliphage C-1 11. The binding of cations to T2 DNA Purification and properties of B . subtilis 168 nucleoside diphosphokinase and its phosphorylated form The hostdependent restriction of mengovirus replication Studies of the protein synthesizing apparatusof an extreme thermophile, Thennus aquaticus Studies on the physiology and biochemistry of the photosynthetic bacterium, Rhodopseudomonas capsuhta Leucothrix mucor as an algal epiphyte in the marine environment The ecological significance ofchemoautotrophicbacteria in hot acid soils The cell wall of Myrococcus xanthus: Chemical changes related to morphogenesis

W. D, Fraser A. L. Koch

M. W. Taylor A. L. Koch T. D. Brock R. Ramaley E. D. Weinberg A. L. Koch A. L. Koch

T. D. Brock W. D. Fraser

R. Ramaley

M. W. Taylor T. D. Brock H. Gest T. D. Brock T. D. Brock D. White

(Continued)

APPENDIX (Continued) Number

Year

56

1972

Ouellette, Andrew Joseph

57 59

1972 1973 1973

Weiss, Richard Lawrence Cordell-Stewart. Barbara Lueking, Donald Robert

60

1974

Orlowski, Michael Edward

61 62

1974 1974

Poos, Jocelyn Carol Prather, Suzanne Virginia

63

1974

Yen, Huei-Che

64 65 66 67

1975 1975 1976 1976

Kottell, Randall Henry Su, Tzyh Chuan Robert Failla, Mark Lawrence Nielson, Allen Madsen

68

1977

Dubos, Michael Scott

69

1978

Bowman, LleweUyn Harper, 111

70 71 72

1978 1978 1979

Kirkconnell, Scott Wolfe Moench, Thomas Taylor Beremand, Phillip

73

1979

Blumberg, Gerald R.

58

Name

Thesis title Comparison of phenylalanyl-tRNAsynthetase and transfer RNAs of liver and Moms hepatoma 5123D Survival of bacteria in extreme environments Mechanism of bovine enterovirus-1-inducedoncolysis and cytopathology Studies on the glycerol metabolism of the non-sulfur purple photosynthetic bacterium, Rhodopseudomonas capsulata Regulation and properties of isocitrate lyase from developingmyxospores of Myxococcus xanthus Aspects of growth and cell division in the bacterium Fkxibacter I. Restricted mengovirus replication in MDBK cells. 11. Effects of cordycepin on mengovirus replication in HeLa cells Regulation of biosynthesis of aspartate family amino acids in the photosynthetic bacterium Rhodopseudomonas pa1ustt-k Isolation and characterization of the myxospore of Myxococcus xanthus Bovine enterovirus-1: Characterization, replication, and maturation Zinc transport and metabolism in Candida utilis Studies on the photosynthetic metabolism of acetate by Rhodopseudom0na.s capsulata strain “St. Louis” The function and distribution among procaryotes of host factor for coliphage QP RNA replication Studies on the distribution of glycerophosphate dehydrogenase in the membranes of the non-sulfur photosynthetic bacteria Rhodopseudomonas capsulata and Rhodopseudomonas sphaeroider Bacterial oxidation of carbon monoxide Distribution, isolation and characterization of a magnetotactic bacterium Events associated with energy perturbations and amino acid starvation in Rhodopseudomonas capsulata The nucleoid of Escherichia coli B/r and related physiological studies

Advisor M. W. Taylor

T. D. B m k M. W. Taylor G. Sojka D. White

D. White M. W. Taylor H. Gest D. White

M. W. Taylor E. D. Weinberg G. Sojka T. Blumenthal G . Sojka

G. Hegeman W. A. Konetzka G. Sojka A. L. Koch

MICROBIOLOGY AT INDIANA UNIVERSITY

205

REFERENCES Campaigne, E. E. (1968).The development of the science departments at Indiana University. Proc. Indiana Acad. Sci. 17, 340-345. Clark, P. F. (1961).“Pioneer Microbiologists of America.” Univ. of Wisconsin Press, Madison, Wisconsin. Clark, T. D. (1976a)“Indiana University: Midwestern Pioneer. Vol. 1. The Early Years.” Indiana Univ. Press, Bloomington, Indiana. Clark, T. D. (1976b).“Indiana University: Midwestern Pioneer. Vol. 2. In Mid-Passage.” Indiana Univ. Press, Bloomington, Indiana. Clark, T. D. (1977).“Indiana University: Midwestern Pioneer. Vol. 3. Years of Fulfillment.” Indiana Univ. Press, Bloomington, Indiana. McClung, L. S. (1941).History of bacteriology at Indiana University. Proc. Indiana Acad. Sci. 53, 59-61. Myers, B. D. (1951).“The Trustees and Oficers of Indiana University: 1820 to 1950.” Indiana Univ. Press, Bloomington, Indiana. Myers, B. D. (1956).“The History of Medical Education in Indiana.” Indiana Univ. Press, Bloomington, Indiana. Sarles, W. B. (1973).Development of applied microbiology at the University of Wisconsin. Ado. Appl. Miwobiol. 16, 301322. Torrey, T. W. (1940).Zoology and its makers at Indiana University.Bios (Madison, N . ] . ) 20,

67-99. Wylie, T. N. (1890). “Indiana University, Its History from 1820,When Founded, to 1890.”Wm. E. Burford, Lithographer, Printer and Binder. Indianapolis, Indiana.

This Page Intentionally Left Blank

INDEX A

Actinophages, of streptomyces, as DNA host-vector systems, 41-43 Anabolic pathways, transfer by recombinant DNA technology, 58-61 Animals, vanishing, gene banks from, 61 Antibiotics, common subunits of, 59 ATP, monitoring in cell cultures, 163 Auxotrophic complementation, gene identification by, 6-9

B BaciUus subtilis, plasmids of, as DNA hostvector, systems, 38-39 Bacteriophage A as DNA host-vector system, 35-37 genome of, map, 36

Bacteriophages, as DNA host-vector systems, 3738 Blunt end ligation, recombinant DNA synthesis by, 27-29

C

Canned foods, nisin use in, 118-119 Carbon dioxide, control in cell cultures, 156159 Catabolic pathways, transfer by recombinant DNA technology, 58 cDNA, in uitro synthesis of, 13-16 Cell culture ATP measurement in, 163 carbon dioxide control in, 156-159 ce1l-d interactions in, 141-142 cell-environment interactions in, 141 cell-growth model for, 140-142 cytofluoremetry use in, 163-164 dialyzable component control in, 152-153 207

instrumentation for process control in, 137-167 evaluation, 142-144 ionic strength control in, 149-152 mixing and viscosity in, 146-147 nicotinamide nucleotide monitoring in, 159-162 oxidation-reduction potential in, 162 oxygen control in, 153-156 pH control in, 148-149 temperature control in, 145-146 Cells animal and microbial compared, 139 mammalian, as DNA host-vector systems, 4547 nisin as growth regulator for, 91-94 Cheese, nisin-producing starters in, 113-115 Chemiluminescent methods, for coliform counting, 178-179 Chocolate milk, nisin use in, 117-118 Chromatographic methods, for coliform counting, 178 Clorobiocin, structure of, 128 Coliform-counting procedures (rapid ), 169183 automation of, 179-180 chemiluminescent type, 178-179 chromatographic type, 178 criteria for, 170-171 electrochemical type, 174-175 enzymatic type, 175-177 radiometric type, 171-172 review of, 171-179 serological type, 172-174 Cosmids, of E. coli, as DNA host-vectors, 35 Coumermycins, 125-136 biological developments on, 128-133 biosynthesis of, 127 chemical developments on, 125-128 Crops, improvement of, by recornbinant DNA technology, 61-64 Cytofluoremetry, use in cell culture process control, 163-164

208

INDEX

D Density, of genes, 9-10 Dialyzable components, control in cell cultures, 152-153 DNA of plasmids, purification of, 3-4~ recombinant, see Recombinant DNA replication of, recombinant DNA studies on, 54 Drift, of culture process instruments, 144

physical differences in, 9-13 selection for function, 5-9 sib selection of, 9 size differences in, 10-13 structure of, recombinant DNA use in,

5253 types of, 4 Genome library, 6 screening of, 6-9

H E Electrochemical methods, for coliform counting, 174-175 Enzymatic methods, for coliform counting,

Homopolymer tailing, recombinant DNA synthesis by, 23 Host-vector systems, for recombinant DNA,

3247 Human growth hormone, plasmid for, 20-21

175-177 Enzymes, for DNMRNA modification, 65-66 Escherichia coli, host-vector systems for recombinant DNA from, 34-39

F Faculty, of Microbiology Department of Indiana University, 191-197 Fecal contamination, coliform-counting procedures for, 169-183 Fermentation industry, instrumentation for,

137-167

I Immunoassays, of genes, 8 Indiana University, history of Microbiology Department at, 185-205 Industrial organisms, improvement by recombinant DNA technology, 5641 Instruments, for cell-culture process control,

137-167 Interferon, production of, by recombinant DNA technology, 54-55 Ionic strength, control in cell cultures, 149-

152 G

L Gelatin, nisin use in, 115-116 Genes amplification of, by recombinant DNA technology, 56-57 auxotrophic complementation of, 6-8 chemical synthesis of, 19-22 cRNA and, 13-16 density of, 9-10 in situ hybridization of, 8 in situ immunoassays of, 8-9 isolation of, 4-22 mapping of, recombinant DNA use in,

5152 novel, production of, 5758

Linkers, for recombinant DNA synthesis,

2931

M Meats, semipreserved, nisin use in, 116-117 Merodiploids, recombinant DNA in construction of, 48 Metabolic pathways, transfer by recombinant DNA technology, 58-61 3-Methylpyrrole-2,4-dicarboxylic acid biosynthesis of, 126

209

INDEX Microbiology Department at Indiana University, 185-205 early period of, 187 faculty and research of, 191-197 modern period of, 187-191 physical facilities of, 197-201 Microcarriers, for cell culture, 139-140 Milk products, nisin use in, 118 Mixing, control in cell cultures, 146-148 Mutagensis, localized, using recombinant DNA, 5 0 5 1

N Nicotinamide nucleotides, monitoring in cell cultures, 159-162 Nisin, 85-123 antibacterial spectrum of, 94 autibiotics cross resistant to, 110, 112-113 assay of, 98-100 biology of, 87-98 biosynthesis of, 105-108 as a cell growth regulator, 91 -94 chemistry of, 98-109 composition, structure, and molecular weight of, 101-103 culture, strains, and media for, 87449 as a dominance factor, 90-91 function to producer organism, 90-94 effect on sporeformers, 94-96 enzymatic inactivation of, 104-105 factors affecting effectiveness of, 96 genetic control of synthesis of, 89-90 growth stimulation by, 113 mode of action of, 94-98 molecular basis, 96-96 as peptide mixture, 101 preparation of, 100-101 properties of, 103-104 resistance to, 97-98 as a secondary metabolite, 90 toxicity of, 109-110 use of, 108-119 as cheese starter, 113-115 countries permitting, 111 in heat-processed foods, 115-119 Noviosylcoumarin, biosynthesis of, 126 Nucleotides, synthetic, as primers for gene isolation, 18-19

0 0perons definition of, 4 fusion of, technology for, 4 8 4 9 Oxidation-reduction potential, monitoring in cell culture, 162 Oxygen, control in cell cultures, 153-156

P pH, control in cell cultures, 148-149 Ph.D. recipients, of Microbiology Department of Indiana University, 201-204 Plants DNA host-vector systems from, 44-45 vanishing, gene banks from, 61 Plasmids of B. subtilis, as DNA host-vector systems, 38-39 of E. coli, as DNA host-vectors, 3 4 3 5 purification of, 3-4 of Streptomycetes, as DNA host-vector systems, 4 0 4 1 Proteins direct secretion of, recombinant DNA in, 50 human production by recombinant DNA technology, 5 4 5 5

R Radiometric methods, for coliform counting, 171-172 Recombinant DNA (See olso Genes) electron microscopy of, 12 gel electrophoresis of, 10-12 in genome organization studies, 5 1 5 3 host-vector systems for, 3 2 4 7 from B . subtilis, 38-39 from E. c d i , 3 4 3 8 from mammalian cells, 4 5 4 7 from plants, 44-45 from yeast, 4 3 4 4 synthesis of, 2 2 3 1 by blunt end ligation, 27-29 by DNA linkers, 29-31 by homopolymer tailing, 23 by ligation of restriction fragments, 23-27

210

INDEX

Recombinant DNA (cont.) technology, see Recombinant DNA technology transformation by, 31-32 Recombinant DNA technology, 1-84 in antiviral vaccine production, 55-56 for crop improvement, 61-64 for directed protein secretion, 50 in DNA replication studies, 53-54 for gene banks of vanishing organisms, 61 in human protein production, 9-55 for improvement of industrial organisms,

Sib selection, of genes, 9 Signal-to-noise ratio, of culture process instruments, 144 Sporeformers, nisin effects on, 94-96 Streptomycetes, DNA host-vector systems from, 3943 SV40 virus, as DNA host-vector, 46

T Temperature, control in, cell cultures, 145-

56-61 in in vftro localized mutagensis, 50-51 in merodiploid construction, 48 for operon fusions, 4849 in resource recovery and waste disposal, 61 uses of, 47-67 Regulons, definition of, 4 Resources, recovery of, recombinant DNA use in, 61 Response time and gain, of culture process instruments, 142-144 Restriction fragments, ligation of, in recombinant DNA synthesis, 23-27 R-loop hybridization, use in gene isolation, 18 RNA complementary, 13-18 as a probe, 16-17 translation of, 17-18

S

146 Transformation, by recombinant DNA, 3132

V Vaccines antiviral, production by recombinant DNA technology, 55-56 Viruses, vaccines against, production by recombinant DNA technology, 55-56 Viscosity, control in cell cultures, 146-148

W

Waste disposal, recombinant DNA use in, 61

Y

Serological methods, for coliform counting,

172-174

Yeast, DNA host-vector systems from, 4344

CONTENTS OF PREVIOUS VOLUMES Vdumo 1

Volume 2

Protected Fermentation

Newer Aspects of Waste Treatment

Nandor Porges

Miloi Herold and Jan NeEasek

Aerosol Samplers

The Mechanism of Penicillin Biosynthesis

Harold W . Batchelor

Arnold L. Demain Preservation of Foods and Drugs by Ionizing Radiations

A Commentary on Microbiological Assaying F . Kavanagh

W . Dexter Bellamy Application of Membrane Filters The State of Antibiotics in Plant Disease Control

Richard Ehrlich Microbial Control Methods in the Brewery

David Prumer

Gerhurd 1. H a s

Microbial Synthesis of Cobamides D . Perlman

Newer Development in Vinegar M a n u k tures

Factors Mecting the Antimicrobial Activity of Phenols E. 0. Bennett Germfree Animal Techniques and Their Applications Arthur w' and James E' Smith

Rudolph 1. Allgeier and Frank M. Hildebrandt The Microbiological Transformation of Steroids T,H . Stoudt ~ i ~ l ~~ @ ~~ ~ l~of Solar ~Energy f William I. Oswald and Clarence G .

Golueke

Insect Microbiology

S. R. Dutky

The Production of Amino Acids by Fermentation Processes

Shukuo Kinoshita

SYMPOSIUM ON ENGINEERING ADVANCES IN FERMENTATION PRACTICE Rheolo@cal Properties of Fermentation Broths Fred H . Deindoerfer a n d l o h n M. West

Continuous Industrial Fermentations

Philip Gerhardt and M. C . Bartlett Fluid Mixing in Fermentation Process

The Large-Scale Growth of Higher Fungi Rua!clijfe F. Robinson and R. S. D a d s o n

1.Y. Oldshue Scale-Up of Submerged Fermentations W . H . Burtholemew

AUTHOR I N D E X 4 U B J E C T INDEX

211

~

~

212

CONTENTS OF PREVIOUS VOLUMES

The Metabolism of Cardiac Lactones by Microorganisms

Air Sterilization

Arthur E. Humphrey

Elwood Titus Sterilization of Media for Biochemical Processes Uoyd L. Kempe Fermentation Kinetics and Model Processes Fred H . Deindoerfer

Intermediary Metabolism and Antibiotic Synthesis J . D. Bu’Lock Methods for the Determination of Organic Acids

A. C. H u l m

Continuous Fermentation W . D. Maron

AUTHOR INDEX-SUBJECT

Control Applications in Fermentation George J . Fuld AUTHOR INDEX-SUBJECT

INDEX

Volume 3 Preservation of Bacteria by Lyophilization Robert J . Heckly

Sphaerotilus, Its Nature and Economic Sig-

INDEX

Volume 4 Induced Mutagenesis in the Selection of Microorganisms S. I. Alikhaniun The Importance of Bacterial Viruses in Industrial Processes, Especially in the Dairy Industry F . J . Babel

nificance

Norman C. Dondero Large-Scale Use of Animal Cell Cultures Donald J. Merchant and C. Richard

Eidam

Applied Microbiology in Animal Nutrition Harlow H . Hall Biological Aspects of Continuous Cultivation of Microorganisms

T. H o l m Protection against Infection in the Microbiological Laboratory: Devices and Procedures

Mark A . Chatigny Oxidation of Aromatic Compounds by Bacteria Martin H . Rogoff

Maintenance and Loss in Tissue Culture of Specific Cell Characteristics Charles C . Morris Submerged Growth of Plant Cells L. G. Nickel1 AUTHOR INDEX-SUBJECT

Screening for the Biological Characterizations of Antitumor Agents Using Microorganisms Frank M . Schabel, Jr., and Robert F. Pit-

till0

INDEX

Volume 5 Correlations between Microbiological Morphology and the Chemistry of Biocides

Adrkrn Albert The Classification of Actinomycetes in Relation to Their Antibiotic Activity Elio Baldacci

Generations of Electricity by Microbial Action J . B. David

213

CONTENTS OF PREVIOUS VOLUMES

Microorganisms and the Molecular Biology of Cancer G . F . Cause

Microbial Formation and Degradation of Minerals Melvin P. Silverman and H e n y L. Eht-lich

Rapid Microbiological Determinations with Radioisotopes

Enzymes and Their Applications Zrwin W. Sizer

Gilbert V. Levin The Present Status of the 2,SButylene Glycol Fermentation Sterling K . Long and Roger Patrick Aeration in the Laboratory W . R . Lockhart and R. W . Squires

A Discussion of the Training of Applied Microbiologists B. W . Koft and Wayne W. Umbreit AUTHOR INDEX-SUBJECT

INDEX

Volumo 7 Microbial Carotenogenesis

Stability and Degeneration of Microbial Cultures on Repeated Transfer

Fritz Reusser

Alex Ciegler Biodegradation: Problems of Molecular Recalcitrance and Microbial Fallibility

Microbiology of Paint Films Richard T. Ross The Actinomycetes and Their Antibiotics

M. Alexander Cold Sterilization Techniques John B. Opfell and Curtis E . Miller

Selman A. Waksman Microbial Production of Metal-Organic Compounds and Complexes

Fuse1 Oil

A. Dinsmoor Webb andJohn L. Zngraham AUTHOR INDEX-SUBJECT

INDEX

Volume 0 Global Impacts of Applied Microbiology: An Appraisal Carl-Giiran He&n and Mortimer P. S t a r Microbial Processes for Preparation of Radioactive Compounds

D. Perlman Development of Coding Schemes for Microbial Taxonomy S . T. Cowan

Effects of Microbes on Germkee Animals

Thomas D. Luckey Uses and Products of Yeasts and Yeast-Like Fungi Walter J. Nickerson and Robert G . Brown

D. Perlmun, A r i s P. Bayan, and Nancy A . Microbial Amylases Giufie Walter W . Windish and Nagesh S. Mhatre Secondary Factors in Fermentation Processes P. Margalith

The Microbiology of Freeze-Dried Foods Gerald J . Silverman and Samuel A .

Goldblith Nonmedical Uses of Antibiotics

Herbert S . Goldberg Microbial Aspects of Water Pollution Control K. W u h m n n

Low-Temperature Microbiology Judith Fatrell and A. H. Rose AUTHOR I N D E X s U B J E C T INDEX

214

CONTENTS OF PREVIOUS VOLUMES

Volum 8 Industrial Fermentations and Their Relations to Regulatory Mechanisms Arnold L. Demain

Antiserum Production in Experimental Animals Richard H. Hyde Microbial Models of Tumor Metabolism G. F. Cause

Genetics in Applied Microbiology S . G. Bradley

Cellulose and Cellulolysis Brigitta Norkrans

Microbial Ecology and Applied Microbiology Thomas D. Brock

Microbiological Aspects of the Formation and Degradation of Cellulose Fibers L. Juraiek, J . Ross Colvin, and D. R. Whita ker

The Ecological Approach to the Study of Activated Sludge Wesley 0.Pipes Control of Bacteria in Nondomestic Water Supplies Cecil W. Chambers and Norman A. Clarke The Presence of Human Enteric Viruses in Sewage and Their Removal by Conventional Sewage Treatment Methods Stephen Alan Kollins

Bulking of Activated Sludge Wesley 0.Pipes Malo-Lactic Fermentation Ralph E. Kunkee AUTHOR INDEX-SUBJECT

Oral Microbiology Heiner Hoffman

INDEX

Volum 10

Media and Methods for Isolation and Enumeration of the Enterococci Paul A. Hartman, George W. Reinbold, and Devi S . Saraswat Crystal-Forming Bacteria as Insect Pathogens Martin H. Rogofl Mycotoxins in Feeds and Foods Emanuel Borker, Ntno F. Insalata, Colette P. Levi, andJohn S . Witzeman AUTHOR INDEX-SUBJECT

The Biotransformation of Lignin to HumusFacts and Postulates R. T. Oglesby, R. F. Christman, and C. H. DftVer

INDEX

Volume Q The Inclusion of Antimicrobial Agents in Pharmaceutical Products A. D. Russell,]une]enktns, and 1. H. Harrison

Detection of Life in Soil on Earth and Other Planets, Introductory Remarks Robert L. Starkey For What Shall We Search? Allan H. Brown Relevance of Soil Microbiology to Search for Life on Other Planets G. Stotzky Experiments and Instrumentation for Extraterrestrial Life Detection Gilbert V. Levin Halophilic Bacteria D. J . Kushner Applied Significance of Polyvalent Bacteriophages S . C . Bradley

215

CONTENTS OF PREVIOUS VOLUMES

Proteins and Enzymes as Taxonomic Tools Edward D. Garber and ]ohn W . Rippon

Ergot Alkaloid Fermentations William 1.Kelleher

Mycotoxins Alex Ciegler and Eivind B. Lillehoj

The Microbiology of the Hen’s Egg R. G . Board

Transformation of Organic Compounds by Fungal Spores Claude V6dna. S. N . Sehgal, and Kamar Singh

Training for the Biochemical Industries 1. L. Hepner

Microbial Interactions in Continuous Culture Henry R. Bungay, Ill and M a y Lou Bungay

Volume 12

Chemical Sterilizers (Chemosterilizers) Paul M . Borick Antibiotics in the Control of Plant Pathogens M . J . Thirumalachar AUTHOR INDEX-SUBJECT

AUTHOR INDEX-SUBJECT

INDEX

History of the Development of a School of Biochemistry in the Faculty of Technology, University of Manchester Thomas Kennedy Walker Fermentation Processes Employed in Vitamin C Synthesis Miloi Kulhanek

INDEX

Flavor and Microorganisms CUMULATIVE AUTHOR I N D E X ~ U M U L A T I V EP. Margalith and Y. Schwartz TITLE INDEX Mechanisms of Thermal Injury in Nonsporulating Bacteria Volume 11 M . C . Allwood and A. D. Russell Successes and Failures in the Search for Antibiotics Selman A. Waksman Structure-Activity Relationships of Semisynthetic Penicillins K . E. Price Resistance to Antimicrobial Agents J . S . Kiser, G . 0. Gale, and G . A. Kemp

Micromnospora Taxonomy George Luedemunn

Collection of Microbial Cells Daniel 1. C . Wang and Anthony J . Sinskey Fermentor Design R . Steel m d T. L. Miller

The Occurrence, Chemistry and Toxicology of the Microbial Peptide-Lactaws A . Taylor Microbial Metabolites as Potentially Useful Pharmacologically Active Agents D. Perlman and G . P. Peruzzotti

Dental Caries and Periodontal Disease Considered as Infectious Diseases William Gold

AUTHOR INDEX-SUBJECT

The Recovery and Purification of Biochemicals V k t o r H. Edwards

Chemotaxonomic Relationships Among the Basidiomycetes Robert G . Benedkt

INDEX

Volume 13

216

CONTENTS OF PREVIOUS VOLUMES

Proton Magnetic Resonance SpectroscopyAn Aid in Identifcation and Chemotaxonomy of Yeasts P. A. J . Gwin and]. F. T. Spencer

Mathematical Models for Fermentation Processes A. G . Frederickson, R. D. Megee, 111, and

Large-Scale Cultivation of Mammalian Cells R. C . T e h g and P. J . Radlett

AUTHOR INDEX-SUBJECT

H. M. Tsuchija INDEX

Volum 14 Large-Scale Bacteriophage Production

K. Sargent Microorganisms as Potential Sources of Food

Jnanendra K. Bhattacharjee Structure-Activity Relationships among Semisynthetic Cephalosporins M . L. Sassioer and Arthur Lewis Structure-Activity Relationships in the Tetracycline Series Robert K. Blackwood and Arthur A . En-

glish Microbial Production of Phenazines J . M. IngramandA. C . Blackwood The Gibberellin Fermentation E . G . Jeffeys Metabolism of Acylanilide Herbicides

Richard Bartha and David P r a m Therapeutic Dentrifrices J . K. Peterson Some Contributions of the U.S. Department of Agriculture to the Fermentation Industry George E . Ward Microbiological Patents in International Litigation

John V . Whittenburg Industrial Applications of Continuous Culture: Pharmaceutical Products and Other Products and Processes A . C . Righelato and R . Elsworth

Development of the Fermentation Industries in Great Britain John J . H. Hastings Chemical Composition as a Criterion in the Classification of Actinomycetes

H. A. Lecheoalier, M a y P. Lecheoalier, and Nancy N . Gerber Prevalence and Distribution of AntibioticProducing Actinomycetes

John N. Porter Biochemical Activities of Nocardia

R. L. Raymond and V . W . Jamison Microbial Transformations of Antibiotics

Oldrich K. Sebek and D. Perlman In Vioo Evaluation of Antibacterial Chemotherapeutic Substances

A. Kathrine Miller Modifcation of Lincomycin Barney J . Magerlein Fermentation Equipment G . L. Solomons The Extracellular Accumulation of Metabolic Products by Hydrocarbon-Degrading Microorganisms Bernard J . Abbott and William E. Gledhill AUTHOR INDEX-SUBJECT

INDEX

Volume 15 Medical Applications of Microbial Enzymes

Irwin W . Sizer

CONTENTS OF PREVIOUS VOLUMES

Immobilized Enzymes K. L. Smiley and C. W . Strandberg

Intestinal Microbial Flora of the Pig

Microbial Rennets

Antimycin A,, a Piscicidal Antibiotic

Joseph L. Sardinas

217

R. Kenwwthy

Robert E. Lennon and Cloude Vhzina

Volatile Aroma Components of Wines and Other Fermented Beverages

A . Dimmoor Webb and Carlos]. MuUer Correlative Microbiological Assays Ladisloo J . Harika

Ochratoxins Kenneth L. Applegate and John R.

.chipley Cultivation of Animal Cells in Chemically Defined Media, A Review

Kiyoshi Higuchi Insect Tissue Culture W . F . Hink

Genetic and Phenetic Classification of Bacteria

Metabolites from Animal and Plant Cell Culture lming S. Johnson and George B. Boder Structure-Activity Relationships in Coumermycins John C . Godfrey and Kenneth E . Price Chloramphenicol

R. R . Colwell Mutation and the Production of Secondary Metabolites Arnold L. Demain Structure-Activity Relationships in the ACtinomycins

Johannes Meienhofer and Eric Atherton

Vedpal S. Malik Microbial Utilization of Methanol Charles L. Cooney and David W . Leoine Modeling of Growth Processes with TWOLiquid Phases: A Review of Drop Phenomena, Mixing and Growth P. S . Shah, L. T.Fan, 1. C . Kao, and L. R.

Erickson Microbiology and Fermentations in the Prairie Regional Laboratory of the National Research Council of Canada 1946-1971 R. H. Haskins AUTHOR INDEX-SUBJECT

INDEX

volume 1s Public Health Significance of Feeding Low Levels of Antibiotics to Animals

Thomas H. lukes

Development of Applied Microbiology at the University of Wisconsin William B. S a r h AUTHOR I N D ~ ~ U B INDEX J E ~

Volume 17 Education and Training in Applied Microbiology Wayne w. "mbreit Antimetabolites from Microorganisms David L. Pruess andJames P. Scannell Lipid Composition as a Guide to the Classification of Bacteria

Norman Shaw Fungal Sterols and the Mode of Action of the Polyene Antibiotics

1. M . T . Hamilton-Miller

218

CONTENTS OF PREVIOUS VOLUMES

Methods of Numerical Taxonomy for Various Genera of Yeasts I . Campbell

Recent Developments of Antibiotic Research and Classification of Antibiotics According to Chemical Structure

Ja.nos B k d y Microbiology and Biochemistry of Soy Sauce Fermentation F . M. Young and B. J . 8. Wood

SUBJECT INDEX

Volum 19 Contemporary Thoughts on As&cts Applied Microbiology P. S. S. Dawson and K . L. Phillips

of Culture Collections and Patent Depositions T. G. Prtdham and C. W . Hesseltine

Some Thoughts on the Microbiological Aspects of Brewing and Other Industries Utilizing Yeast G . G . Stewart

Production of the Same Antibiotics by Members of Different Genera of Microorganisms

Linear Alkylbenzene Sulfonate: Biodegradation and Aquatic Interactions

Antibiotic-Producing Fungi: Current Status of Nomenclature C . W. Hesseltine and J . J . Ellis

William E. Gledhill The Story of the American Type Culture Collection-Its History and Development (1899-1973)

Hubert A . Lechevalier

Significance of Nucleic Acid Hybridization to Systematics of Actinomycetes S . G . Bradley

William A . Clark and Dorothy H. G e a y Microbial Penicillin Acylases E . J . Vandamme and J . P. Voets SUBJECT INDEX

Current Status of Nomenclature of AntibioticProducing Bacteria

Erwin F . Lessel Microorganisms in Patent Disclosures

Irving Marcus

Volum 18 Microbial Foundation of Environmentd Pollutants

Martin Alexander Microbial Transformation of Pesticides lean-Marc Bollag

Microbiological Control of Plant Pathogens Y. Henis and 1. Chet Microbiology of Municipal Solid Waste Composting Meluin S. Finstein and Merry L. Morris

Taxonomic Criteria for Mycobacteria and Nocardiae S. G . Bradley a n d ] . S. Bond

Nitrification and Dentrification Processes Related to Waste Water Treatment

Effect of Structural Modifkations on the Biological Properties of Aminoglycoside Antibiotics Containing 2-Deoxystreptamine

The Fermentation Pilot Plant and Its Aims D. J . D. HockenhuU

Kenneth E. Price, John C . Godfrey, and Hiroshi Kawguchi

D. D . Focht and A. C . Chang

The Microbial Production of Nucleic AcidRelated Compounds

Koichi Ogata

CONTENTS OF PREVIOUS VOLUMES

Synthesis of L-Tyrosine-Related Amino Acids by &Tyrosinase

Hideaki Yamado and Hidehiko Kumagai Effects of Toxicants on the Morphology and Fine Structure of Fungi

Donald V . Richmond SUBJECT INDEX

Volunn 20

The Current Status of Pertussis Vaccine: An Overview

Cytotoxic and Antitumor Antibiotics duced by Microorganisms J . Fuska and 8 . Proksa

219 Pro-

SUBJECT INDEX

V d u m 21 Production of Polyene Macrolide Antibiotics Juan F . Martin and Uoyd E . McDaniel Use of Antibiotics in Agriculture Tomonuasa Misato, Keido KO, and Isamu

Yanuaguchi

Charbs R. Manclark Biologically Active Components and Properties of Bordetella pertussis Stephen 1. Morse Role of the Genetics and Physiology of Bordetella pertussis in the Production of Vaccine and the Study of Host-Party Relationships in Pertussis

Charlotte Parker

Enzymes Involved in &Lactam Antibiotic Biosynthesis E. J . Vandamme Information Control in Fermentation Development D. J . D. Hockenhull Single-Cell Protein Production by Photosynthetic Bacteria R. H. Shipnuan, L. T. Fan, a n d l . C . Kao

Problems Associated with the Development and Clinical Testing of an Improved Pertussis Vaccine George R. Anderson

Environmentd Transformation of Alkylated and Inorganic Forms of Certain Metals

Problems Associated with the Control Testing of Pertussis Vaccine

Bacterial Neuraminidase and Altered Immunological Behavior of Treated Mammalian Cells

Jack Cameron

Jitendra Saxena and Philip H. Howard

Prasanta K. Ray Vinegar: Its History and Development

Hubert A. Conner and Rudolph J . AUgeier Microbial Rennets

M. Sternberg Biosynthesis of Cephalosporins

Toshihiko Kanzaki and Yukio Fujcpowa Preparation of PharmaceuticalCompounds by Immobilized Enzvmes and Cells Bernard J . Abbott

Pharmacologically Active Compounds kom Microbial Origin Hewitt W. Matthews and Barbara Fritche

Wade SUBJECr INDEX

V d u m 22 Transformation of Organic Compounds by Immobilized Microbial Cells

Ichiro Chibata and Tetsuya Tosa

220

CONTENTS OF PREVIOUS VOLUMES

Microbial Cleavage of Sterol Side Chains Christuph K. A. Martin

The Biodegration of Polyethylene Glycols Donald P. Cox

Zearalenone and Some Derivatives: Production and Biological Activities P. H. Hidy, R. S. Baldwin, R. L. Greasham, C. L. Keith, and]. R. McMul-

Introduction to Injury and Repair of Microbial Cells

Zen Mode of Action of Mycotoxins and Related Compounds F . S . Chu Some Aspects of the Microbial Production of Biotin

Yoshikazu Izumi and Koichi Ogata

F . F. Busta Injury and Recovery of Yeasts and Mold K. E. Stevenson and T. R. Graumlich Injury and Repair of Gram-Negative Bacteria, with Special Consideration of the Involvement of the Cytoplasmic Membrane L. R. Beuchat Heat Injury of Bacterial Spores

Daniel M . Adams Polyether Antibiotics: Versatile Carboxylic Acid Ionophores Produced by Streptomyces 1.W. Westley

The Involvement of Nucleic Acids in Bacterial Injury M . D. Pierson, R. F. Comet, and S . E .

Martin The Microbiology of Aquatic Oil Spills

R. Bartha and R. M . Atlas Comparative Technical and Economic Aspects of Single-Cell Protein Processes John H . Litchfkld

SUBJECT INDEX

Volume 24 Preservation of Microorganisms

Robert]. Heckly SUBJECT INDEX

Streptococcus mutans Dextransucrase: A Re-

Volume 23 Biology of Bacillus popilliae Lee A . BuUa, Jr., Ralph N . Costilow, and

Eugene S . Sharpe Production of Microbial Polysaccbarides M. E . Slodki and M . C. Cadmus Effects of Cadmium on the Biota: Influence of Environmental Factors H . Babich and G . Stotzky Microbial Utilization of Straw (A Review) Youn W . Han The Slow-Growing Pigmented Water Bacteria: Problems and Sources

Lloyd G . Herman

view Thorns J . Montville, Charles L. Cooney and Anthony J . Sinskey Microbiology of Activated Sludge Bulking Wesley 0. Pipes Mixed Cultures in Industrial Fermentation Processes David E. F . H a d s o n Utilization of Methanol by Yeasts

Yoshiki Tani, Nobuo Kato, and Hideaki Yamada Recent Chemical Studies on Peptide Antibiotics Jun’ichi Shoji

CONTENTS OF PREVIOUS VOLUMES

221

The CBS Fungus Collection I. A. Von A n and M . A. A. Schipper

Bacteriophages of the Genus Cbstddtum

Microbiology and Biochemistry of Oil-Palm Wine

SUBJECT INDEX

Seiya Ogata and Motoyoshf Hongo

Nduka Okajor

Volume 26

Bacterial-Amylases

M . B. lngle and R . J . Erickson Microbial Oxidation of Gaseous Hydrocarbons

Ching-Tsang Hou

SUBJECI INDEX

Volume 25 Introduction to Extracellular Enzymes: From the Ribosome to the Market Place

R u d y ] . Wodzinski Applications of Microbial Enzymes in Food Systems and in Biotechnology Matthew J . Taylor and Tom Richardson Molecular Biology of Extracellular Enzymes

Epoxidation and Ketone Formation by C1Utilizing Microbes Ching-Tsang Hou, Ramesh N . Patel, and

Allen 1. Laskin Oxidation of Hydrocarbons by Methane Monooxygenases from a Variety of Microbes

Howard Dalton

Robert F. Ramaley

Propane Utilization of Microorganisms

Increasing Yields of Extracellular Enzymes

Douglas E. Eoeleigh and Montenecourt

Ecology and Diversity of Methylotrophic Organisms R. S. Hanson

Bland

S.

Regulation of Chorismate-Derived Antibiotic Production

Vedpal S . Malik Structure-Activity Relationships in Fusidic Acid-Type Antibiotics W . von Daehne, W . 0. Godtfi-edsen, and P. R. Rasmussen Antibiotic Tolerance in Producer Organisms Leo C . Vining

Microbial Models for Drug Metabolism John P. Rosazza and Robert V . Smith Plant Cell Cultures, a Potential Source of Pharmaceuticals W. C.W . Kurz and F. Constabel

Jerome]. Perry Production of Intracellular and Extracellular Protein from ndutane by P s e u d m n a s butanooora sp. nov.

Joji Takahashi Effects of Microwave Irradiation on Microorganisms John R . Chipley Ethanol Production by Fermentation: An Alternative Liquid Fuel N. Kosaric, D. C . M . Ng, I . Russell, and G. C. Stewart Surface-Active Compounds from Microorganisms D. C. Cooper and J . E . Zajic INDEX

This Page Intentionally Left Blank