Advances in Microbial Physiology Volume 26

Advances in

MICROBIAL PHYSIOLOGY

This Page Intentionally Left Blank

Advances in MICROBIAL PHYSIOLOGY edited b y

A. H. ROSE

D. W. TEMPEST

School of Biological Sciences Bath University Bath, England

Department of Microbiology University of Shyfield Sheffield, England

Volume 26 1985

ACADEMIC PRESS (Harcourt Brace Jovanovich, P14hlishers) London Orlando San Diego New York Toronto Montreal Sydney Tokyo

COPYRIGHT 0 1985 BY ACADEMIC PRESS INC. (LONDON) LTD ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING. OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS INC. (LONDON) LTD. 24-28 Oval Road

LONDONNWI 7DX

United States Edition published by

ACADEMIC PRESS,INC. Orlando, Florida 32887

ISBN 0-1 2-027726-3 ISSN 0065-2911

PRINTED IN THE UNITED STATES OF AMERICA 85 86 87 88

9 8 7 6 5 4 3 2 I

Contents Nitrogen Catabolite Repression in Yeasts and Filamentous Fungi by JEAN-MARIE WIAME. MARCELLE GRENSON. and HERBERT N. ARST J R . 1. I1 . III . IV .

Introduction . . . . . . . . . . Nitrogen Catabolite Repression in Succ~harc~myc~es crrruisicie Nitrogen Metabolite Repression in Filamentous Fungi . Acknowledgements . . . . . . . . . . . . . . . . . References . Note Added in Proof . . . . . . . .

.

.

.

.

. . . .

. . . .

2 4 57 78 79 88

Sexual Agglutination in Chlamydomonads by H. VAN DEN ENDE 1. I1 . Ill . IV . V. VI .

Introduction . . . . . . The Mating System . . . . The Mating Process . . . . The Specificity of Sexual Agglutination Dynamics of Sexual Agglutination . Concluding Remarks . . . . . . . . . References .

.

.

.

.

.

.

. .

. .

. .

. .

. .

. .

. . . .

. . . .

. . . .

. . . .

. . . .

89 90 91

. 102 . 111 . 119 . 119

The Energy Flow in Bacteria: The Main Free Energy Intermediates and Their Regulatory Role by K. J . HELLINGWERF and W . N. KONINGS Introduction . . . . . . . . . . . The Energy Circuit in Bacteria . . . . . . . Energy Transduction in the Cytoplasmic Membrane . . . Regulation by Energy Intermediates . . . . . . Homoeostasis in the Magnitude of Free Energy Intermediates . V1 . Conclusions and Perspectives . . . . . . . VII . Acknowledgements . . . . . . . . . References . . . . . . . . . . .

I. I1 . Ill. 1V . V.

. . . . . . . .

125

126 130 138 146

148 149 150

vi

CONTENTS

Hydrogenase. Nitrogenase. and Hydrogen Metabolism in the Photosynthetic Bacteria by PAULETTE M. VIGNAIS. ANNETTE COLBEAU. JOHN C . WILLISON. and YVES JOUANNEAU 1. Introduction

.

I1. The Organisms

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

111. Hydrogen Utilization and Production by Photosynthetic Bacteria IV . Hydrogenase . . . . . . . . . . V . Nitrogenase . . . . . . . . . . . V1 . Genetics of Hydrogen Production and Utilization . . . VII . Use of Photosynthetic Bacteria as Biological Solar Energy . . . . . . . . . . Converters . VIII . Summary and Prospects . . . . . . . . IX . Acknowledgements . . . . . . . . . References . . . . . . . . . . . . . . . . . . . Note Added in Proof .

. 156 . 156

.

162

. 174 . 190

. 204 . . . . .

210 218 220 220 234

Biochemistry and Physiology of Bioluminescent Bacteria by J. WOODLAND HASTINGS. CATHERINE J . POTRI.KUS, SUBHASH C . GUPTA. MANFRED KURFURST. and JOHN C. MAKEMSON 1. Introduction . . . . . I1 . Taxonomy . . . . . I11 . Biochemistry . . . . 1V . Molecular Biology . . . V . Physiology . . . . . VI . Ecology . . . . . V11 . Analytical and Clinical Applications VIII . Acknowledgements . . . . . . . References . Note Added in Proof . . . Author Index Subject Index

.

.

.

.

.

.

.

.

. .

. .

. .

. . . . . .

. . . . . .

. . . . . .

.

.

.

. .

.

. .

.

. .

.

. .

. . . . . . . .

.

.

. .

. . . . . . . .

.

.

. .

. 236 . 237

. .

238 256 259 269 . 274 . 280 . 281 . 291

.

. 293

. . . . . .

.

. . . .

. 311

Cont ributo rs HERBERT N. ARST JR. Department of Genetics, Ridley Building, The University of Newcastle-upon-Tyne, Newcastle NEI 7 R U , England ANNETTE COLBEAU Laboratoire de Biochimie (CNRSIER 235, I N SERM U . 191, CEAIIRF), DPpartement de Recherche Fondamentale, Centre d’Etudes NuclPuires de Grenoble 85 X , 38041 Grenohle, France MARCELLE GRENSON Laboratoire de Microbiologie, FacultP des Sciences, Campus Plaine, UniversitP Lihre de Bruxelles, B-1050 Bruxelles, Belgium SUBHASH C. GUPTA Department of Cellular and Devclopmental Biology, Harvard University, Cambridge, Massachusetts 02138, USA J . WOODLAND HASTINGS Department of Cellular and Developmental Biology, Harvard University, Cambridge, Massachusetts 02138, USA K . J . HELLINGWERF Department of Microbiology, University of Groningen, 9751 N N Huren, The Netherlands Y VES JOUANNEAU Laboratoire de Biochimie (CNRSIER 235, INSERM U . 191, CEAIIRF), Dkpurtement de Recherche Fondamentale, Centre d’Etudes NiiclPaires de Grenohle 85 X , 38041 Grenoble, Frunce W. N . KONINGS Department cf Microbiology, University Qf Groningen, 9751 N N Huren, The Netherlands MANFRED KURFURST Department of Cellular and Developmentcrl Biology, Harvard University, Cambridge, Massachusetts 02138, USA JOHN C. MAKEMSON Department of Biological Sciences, Florida International University, Miami, Florida 33199, USA CATHERINE J . POTRIKUS’ Department of Cellular und Developmentul Biology, Harvard University, Cambridge, Ma.ssac~husetts02138, USA

’

Present address: Pfizer Central Research, Eastern Point Road, Groton, Connecticut 06340, USA.

viii

CONTRIBUTORS

H . VAN DEN ENDE Department of Plant Physiology, University of Amsterdam, 1098 SM Amsterdam, The Netherlands PAULETTE M . VIGNAIS Lahorutoire de Biochimie (CNRSIER 235, INSERM U . 191, CEAIIRF), Dc5partrment cie Rpchrrche Fondamentale, Centre d’Etudes NucIPaires dr Grenoble 85 X , 38041 Grenoble, Frunce JEAN-MARIE WIAME Laboratoire de Microhiologie, FacultP des Sciences, UniversitP Lihre de Bruxelles and Institlrt de Recherches du Centre d’Enseignement et de Recherches des Industries Alimentnires rt Chimiques, B-1070 Bruxelles, Belgium JOHN C . WILLISON Lahoratoire de Biochimie (CNRSIER 235, INSERM U . 191. CEAIIRF), DPpurtement de Recherche Fondumentale, Centre d’Etudes NuclPaires de Grenoble 85 X , 38041 Grenohle, Frunce

Nitrogen Catabolite Repression in Yeasts and Filamentous Fungi JEAN-MARIE WIAME,* MARCELLE GRENSON,t and HERBERT N. ARST JR* Laboratoire de Microbiologie, Faculte des Sciences, Universite Libre de Bruxelles, and lnstitut de Recherches du Centre d'fnseignement et de Recherches des Industries Alimentaires et Chimiques, Bruxelles, Belgium t Laboratoire de Microbiologie, Faculte des Sciences, Campus Plaine, Universite Libre de Bruxelles, Bruxelles, Belgium Department of Genetics, Ridley Building, The University of Newcastle upon Tyne, England

*

I. Introduction.

.

.

.

.

11. Nitrogen catabolite repression in Succhnrornyces cereuisiae

. . .

A. Methodological considerations: strains and cultures . B . Enzymes and genes-involved in the early steps in assimilation nitrogenous nutrients . . . . . . . C. Degradation of arginine and proline . . . . . D. Degradation of allantoin and urea . . . . . E. Asparaginase 11 . . . . . . . . F. Glutamate dehydrogenases . . . . . . G. Glutamine synthetase and proteinase B . . . . H . Uptake systems for nitrogen-containing compounds . . I. General amino-acid permease . . . . . . J . Transport of proline . . . . . . . K . Ureidosuccinate-allantoate permease . . . . L. Ammonia uptake systems . . . . . . M. Uptake systems for L-glutamine and L-asparagine . . N. Glutamic acid permeases. . . . . . . 0. Other ammonia-sensitive permeases . . . . . P. Comments on regulation of uptake of nitrogenous nutrients Q. General view . . . . . . . . . 111. Nitrogen metabolite repression in filamentous fungi. . . A. Background . . . . . . . . . B. Genes involved in nitrogen metabolite repression . .

. . .

. . .

. . .

2 4 4

. . . . .

. . . . .

. . .

. . . .

. .

.

8 12 24 3 1 34 3 5 37 40 48

of

. . . . . . . . .

.

.

.

. . . . . . . . . .

. . . . . . . . . . .

. .

. .

50 51 52

53 54

54 56 57 51 59

2

JEAN-MARIE WIAME E T A L .

C. L-Glutamine is the nitrogen metabolite co-repressor and Aspergillus nidulrrns. . . . . D. Interactions with other regulatory systems . E. Cis-acting regulatory mutations . . . IV. Acknowledgements . . . . . . References . . . . . . Note added in proof . . . . . .

in Nrro-osporu

. . . . .

C‘I’CI~SII

.

,

.

. . . .

. . . . .

. . . . .

. 7 0 . 72 . 7 6 . 78 . 79 . 88

I. Introduction Increasing interest in the study of gene expression and its regulation in fungi has been motivated by the desire to glimpse a critical step in the emergence of eukaryotes. Early work was conducted using concepts derived from characterization of prokaryotes and their viruses. Similarities which could be at least formally explained using terms such as repressors, activators, operators, and promoters have been described. Genetic identification of these components of regulatory circuits presents an opportunity to attempt a more direct molecular approach. At the same time, the need to select new kinds of mutants, identifying further regulatory components, remains a priority. Similarly, it is premature to sound the death knell for classical enzymology. Many enzymes and pathways resist characterization; others resist even recognition. Fungi, particularly the ascomycetes, having been brought into the main stream of genetical research in the 1930s, 1940s, and 1950s, fostered a rationale in methodology that has proved extremely fruitful in the study of prokaryotes. The ascomycetes offer a wide array of physiological diversity. The obligately aerobic growth of a number of yeasts and filamentous fungi contrasts starkly with a preference for fermentation amongst Saccharomyces spp. and some other yeasts. Between these two extremes are a number of genera, such as Kluyueromyces, Hansenula, and Debaryomyces, with less pronounced preferences. Unfortunately, little work has been done on the genetics of these organisms which are well suited to both an aerobic and an anaerobic mode of life. Amongst fungi, some well defined biochemical differences have been found in a variety of metabolic pathways. In some cases, it is precisely these pathways for which most information is available concerting regulatory processes, including pathways involved in utilization of nitrogenous compounds. In general, greater metabolic versatility accompanies a greater propensity for aerobiosis. The ability of filamentous fungi and aerobic yeasts to utilize nitrate and purines as nitrogenous nutrients is an example. For present purposes, it is important to note the frequent capability of aerobic fungi to utilize many nitrogenous compounds as sources of carbon as well as nitrogen. In contrast, Saccharomyces spp. are only able to utilize these compounds as sources of nitrogen. This imposes

NITROGEN CATABOLITE REPRESSION IN FUNGI

3

restrictions for the study of catabolism in Saccharomyces spp., where fewer nitrogen catabolic enzymes have been identified. A possible advantage is that a study of regulation of nitrogen metabolism is not complicated by an overlap in regulatory domain with regulation of carbon metabolism. With increasing propensity for aerobiosis, the greater involvement of mitochondria in energy provision is apparently accompanied by differences in the use of cornpartmentation as a regulatory device, illustrated with particular clarity in the regulation of arginine metabolism in various yeasts. Division of this review into separate sections dealing with filamentous fungi and the yeast Saccharomyces cereuisiae reflects our impressions of diversity in regulatory mechanisms involved in nutrition of nitrogenous compounds. At present, there seems little scope for broad generalizations on fungi. Glutamine is probably a key metabolite in regulation of nitrogen metabolism in fungi as well as in prokaryotes, reflecting its central position in the metabolism of both groups of organisms. However, the roles played by glutamine in the various organisms are probably very different. In filamentous fungi, glutamine apparently prevents activation of gene expression by a positive-acting regulator gene product mediating nitrogen metabolite repression.' In Saccharomyces spp., however, there is evidence only for negative regulation in the regulatory mechanism covering the domain of nitrogen nutrition, and other more localized regulatory processes are probably grafted as modulation on this more ubiquitous mechanism. Regulation of nitrogen metabolism in prokaryotes has been reviewed by Magasanik (1982) in a treatise that traces the intellectual development of the subject as well as summarizing the relevant experimental results and the conclusions that follow from them. It is now clear that the presence of additional, previously unrecognized, genes in an operon with the structural gene for glutamine synthetase was responsible for the now-discarded hypothesis for a major positive regulatory role for glutamine synthetases in enteric bacteria. However, a regulator protein that does play a major role in nitrogen regulation in Escherichia coli has been isolated (Reitzer and Magasanik, 1983). As one surveys what is presently known of the general regulation of nitrogen nutrition in fungi and prokaryotes, one is struck by the diversity and possibly led to speculate that variation in regulatory mechanisms might be responsible for a substantial portion of the diversity we see in Nature. The section of this review dealing with yeasts was the responsibility of M. Grenson and J.-M. Wiame, that dealing with filamentous fungi was the responsibility of H. N . Arst, Jr. I Throughout this review, the term nitrogen catabolite repression is used when referring to yeasts, whereas the term nitrogen metabolite repression is used when refemng to filamentous fungi, in keeping with standard usage by workers with the two groups of organisms.

4

JEAN-MARIE WIAME E T A L .

11. Nitrogen Catabolite Repression in Saccharomyces cerevisiae

A . METHODOLOGICAL CONSIDERATIONS: STRAINS AND CULTURES

1. Strains The ancestral use of yeasts in industry led to selection of many wild-type strains of Saccharomyces cereuisiae. Other fungi, primarily those of scientific interest such as Neurospora crassa and Aspergillus nidulans, are more homogeneous. Different strains of S. cereuisiae may differ in having or not having an enzyme; the finding of strains lacking asparaginase I1 is an example (Jones, 1977; Dunlop et al., 1978). In addition, among strains belonging to an accepted taxonomical species, variable regulatory processes controlling a set of identical enzymes are frequently found. It is therefore not surprising that comparing non-isogenic strains, or crossing them, will lead to confusion. Differences in the regulation of prokaryotes are well known (Escherichia coli K-12, B, and W; Jacoby and Gorini, 1969). The need to use mutations that originate from different laboratories may introduce unintentional important differences. Many wild-type strains were originally diploids and show differences in their two haploid genomes. According to the chosen strain, one may not discover a given regulation. Such a case, which concerns the subject of this section of our review, is illustrated by comparing two haploid strains which were used a long time ago in Brussels. One strain is the 1705d previously used in the study of arginase-ornithine carbamoyltransferase (OCTase) interaction (Btchet and Wiame, 1965). The other is the classical 21278b (a) strain from which the 3962c (a) mating-type mutant was derived precisely to ensure isogeny (Bechet et al., 1970). Other mutants were obtained in these strains including some with auxotrophic markers for the same purpose. Collection markers were used only for determining allelism and, if necessary, were introduced by a number of backcrosses with the original 21278b strain. Arginase production under three conditions of exponential growth illustrates the difference between the strains (Table 1). Although accidental, the use of one instead of the other strain would have led to missing unrecognized regulatory processes in yeast. For example, using strain 21278b instead of 1705d, because of poor arginase production when grown in the presence of arginine and ammonia, would not have shown arginase-ornithine carbamoyltransferase interaction in situ, whereas using strain 1705d instead of 21278b would not have shown the ammonium effect.

NITROGEN CATABOLITE REPRESSION IN FUNGI

5

TABLE 1. Arginase activity in two wild-type haploid strains of S. cereuisiae" Nitrogen nutrient in the growth medium (3% glucose as carbon source)

Strains 1705db Xl278b' a

Ammonia

Ammonia with arginine

Arginine

17 5-8

210 20-24

247 240

Arginase activity as pmol hr-' mg protein-' at 30°C. Unpublished results from C. Hennaut and J . M. Wiame. From Dubois ei al. (1974).

Along the same line, Rytka (1975) showed that strain S288c differs from Z1278b by one genetic character, which is designated amc+ in 21278b and leads to a strong "ammonia effect" on the general amino-acid permease. Differences between strains 1705d and 21278b, as well as between the two haploid genomes of M25 used by T. G. Cooper and collaborators, are multigenic. Although complementary markers are convenient for crosses, they introduce 50% of foreign genetic background at each cross. Crosses do not necessitate markers in strains; zygotes can be chosen by their morphological characters. 2. Growth Saccharomyces spp. are very different from bacteria, although in vegetative growth they can be handled in a very similar way. This is probably the origin of the choice of these organisms when prokaryotic cellular physiologists became interested in eukaryotic organisms (Watson, 1975). The similarity in behaviour is due to their common unicellular nature, submerged cultivation, and almost absence of ageing. As a result, quantitative methods developed in the past for bacteria can be applied directly to yeasts (Monod, 1949,1950; Monod et at., 1952; Novick and Szilard, 1950; Cohen and Monod, 1957). The most useful property is the prolonged exponential growth phase during which new cells resemble old ones. In other words, the growth is balanced. In a culture, if a , b, and c represent cell mass ml-', mass of protein ml-l, and amount of a given enzyme rnl-', respectively, the slopes

6

JEAN-MARIE WIAME E T A L .

of log a versus time, as well as log b and log c , are identical. A replot of a versus b or c in Monod co-ordinates ( a ml-’ versus b ml-I) gives a straight line. The necessity to recall this elementary point is that, even today, one frequently finds an expression such as “enzyme activity after 2 hours of treatment” being used without any data expressing growth and, still worse, sometimes without the values for cell mass and enzyme activity before treatment. The long balanced growth phase is a remarkable phenomenon. Cells use nutrients that are not in great excess: 1% glucose for S. cerevisiae (which grows mainly on the basis of fermentation even in aerobiosis) may be limiting at the end of the exponential phase of growth. This originates from a weak (or the absence of) Pasteur effect in species of the genus Saccharomyces (Ephrussi et al., 1956; Vissers et al., 1982). A large part of the ammonia usually present in minimal medium will be used up by the end of growth. Indeed, ammonia will be limiting when using less than the usual 20 mM concentration. So, in spite of rapid rates of nutrient consumption, growth continues without extensive modification. One of the reasons is the occurrence of efficient transport systems. With a K , value of 1 PM for ammonia, this permease will remain saturated until the ammonia concentration is above 10 pM. Ethanol production remains below its toxic concentration. Hydrogen ion production is a more difficult question. Yeasts are known to be able to grow over a large pH range. The optimal pH value is usually between 4.5 and 6.5 but usually growth is not strcngly modified between pH 3 and 8 (Rose, 1975). However, exchange of chemical compounds between growth medium and cells is under the control of pH value, especially with ionized compounds. Pyruvic acid is a good carbon source at pH 3.3, but not at pH 6.0. Oxaluric acid acts as a non-metabolizable inducer of urea degradation at pH 3.3, but it does not enter the yeast cell at neutral pH values. Thus, modifications in pH value may alter the influence of compounds used in the study of regulation, even if these pH modifications do not drastically modify growth. For example, during synthesis of a dipeptide from glucose and ammonia there is a concomitant production of one proton equivalent for each nitrogen atom assimilated: C6Ht2O6+ 2NH4++ C603N2H,2(alanylalanine) + 3HZ0+ 2H+

Unbuffered media will change in pH value over an experiment. Oxalurate could be a poor inducer or a good one depending at what stage of the growth it is applied. Commercial media are usually derived from the medium of Wickerham (1946). This medium is weakly buffered with phosphate. Phosphate with a

NITROGEN CATABOLITE REPRESSION IN FUNGI

7

pK1 value of 2.2 and a pK2 value of 7.2 does not ensure a constant pH value. Its initial pH is 4.5-5.0 and it is outside of the buffered regions which anyway are at the extremes of the useful pH scale. Saccharomyces cereuisiae does not use citric acid (Barnett and Kornberg, 1960) and so it can be used as a buffer, with the capacity to maintain a constant pH at the three useful values of 3.3, 4.7, and 6.0. Citric acid cannot be used as buffer for a number of other yeasts because they utilize it as a source of , carbon (Lodder, 1971; Barnett et a f . , 1983).

a. Starvation. Starvation for nitrogen has been often used to detect nitrogen catabolite de-repression. After being separated from growth medium, cells can be introduced in a medium lacking a nitrogenous nutrient. Usually this does not stop protein synthesis abruptly, since nitrogenous compounds present in vacuoles allow a limited synthesis. Vacuoles store basic amino acids very efficiently (Wiemken and Durr, 1974). After growth in a medium containing glucose and ammonia, starvation for ammonia led Middelhoven (1968) to observe a strong arginase synthesis and suggested nitrogen catabolite de-repression. Indeed, arginase is the key enzyme for use of arginine as a nitrogenous nutrient, and arginine is the most abundant amino acid in the vacuoles. Arginase synthesis may be under the control of induction rather than nitrogen catabolite repression when starved. We shall return to this subject later. Starvation has also been observed as a signal for modification of enzyme activity as distinct from enzyme synthesis. The best known cases concern carbon catabolism, the very first being a modification of glycogen phosphorylase in animal tissues discovered by Cori and Cori under conditions of stress or hormonal treatment. Many other examples are known today, and they include bacterial and yeast enzymes. It is established that some of these modifications may occur by covalent modifications of enzymes catalysed by converting enzymes (adenylation and phosphorlyation). Modification of enzyme activity usually is a reversible process which may help the cell to adapt to a new metabolic situation more quickly than enzyme dilution into new cells or by synthesis de nouo. Enzyme conversion can also occur in actively growing cells. This is so with glutamine synthetase in E . coli (Wulff et al., 1967; Shapiro et al., 1967; Stadtman, 1970). In the nitrogen catabolism of yeasts, glutamate dehydrogenase (NAD+) and glutamine synthetase are subject to modification of enzyme activity in addition to modification of synthesis (Hemmings, 1978). Modification of activity instead of synthesis is best analysed by the use of Monod co-ordinates during a shift from one growth condition to another (BCchet and Wiame, 1965; Legrain et al., 1982; Grenson, 1983a,b).

8

JEAN-MARIE WlAME ET AL.

To conclude, starvation should not be used to indicate occurrence of nitrogen catabolite repression without further control.

b . Nitrogen limitation. Growth of yeast with nitrogenous nutrients of different quality, but supporting exponential balanced growth, is the most usual way to modify the state of nitrogen catabolite repression. Exponential growth, however, does not exclude modifications of enzyme activity by processes distinct from synthesis. Indeed, the most immediate and frequent action in uiuo is feedback inhibition, as well as activation, which operate as soon as the composition of the medium changes and which adjust the metabolic flux before the relative concentration of enzyme is regulated as the result of gene activity.

c. Inducer exclusion. Inhibition of activity of transport may lead to inducer exclusion. This can be misleading when modifications of nitrogenous nutrition are studied in the presence of an inducer. Amino acids and even ammonia may compete with inducer transport and may inactivate permeases. This has been well illustrated for different yeast permeases. B. ENZYMES A N D GENES INVOLVED IN THE EARLY STEPS IN ASSIMILATION OF NITROGENOUS NUTRIENTS

1. Ammonia and Its Uptake Ammonia is a very good nitrogenous nutrient. Its utilization is, however, greatly decreased by the presence of other very good nitrogen sources, such as asparagine and glutamine (Dubois et al., 1974). Defects in ammonia transport have been obtained by selection of methylamine-resistant mutants. Roon and associates provided evidence for the existence of a transport system. Methylamine uptake is inhibited competitively by ammonia (Roon et al., 1975a) and non-competitively by amino acids (Roon et al., 1977). Transport of ammonia into the cell is mediated by at least two, and probably three, separate systems (Dubois and Grenson, 1979). A mutant with a double defect was selected by resistance to methylamine (100 mM) with proline as nitrogen source (C. Hennaut and J. M. Wiame, unpublished observations). This mutant bears two separate mutations: (1) a mepl mutation that leads to loss of the methylamine low-affinity transmg protein-') and (2) a mep2 port (&, 2 m M ; V,,,, 50 nmol min mutation leading to loss of a methylamine high-affinity transport (Km, 250 p ~ V,,,,,, ; 20 nmol min-I mg protein-I). Ammonia inhibits competitively

-'

NITROGEN CATABOLITE REPRESSION IN FUNGI

9

both systems with K , values of 20 p~ and 1 p ~ respectively. , So the affinity for ammonia differs by a factor of 20. The generation times of the double mutant were respectively 10,3.8, and 2 hours in the presence of 1, 4, and 20 mM ammonia. The growth rate with 20 mM ammonia indicates occurrence of an additional transport system (Dubois and Grenson, 1979). The identity of the systems described by Roon with mepl or mep2 has not been established, but mepl could be the system first described by Roon. Regulation of these transport systems will be considered separately.

2. From Ammonia to Glutamate a . Glutamate dehydrogenase and the gdhA- mutations. In Saccharomyces cerevisiae there are two glutamate dehydrogenases (GDHase), one specific for NAD+ and one for NADP+ (Holzer and Schneider, 1957). The role of these two enzymes was first deduced on the basis of their activities under different conditions of growth. There is a general agreement that the NAD+-requiring enzyme is low in activity when cells are grown with ammonia or ammonia and glutamate (Hierholzer and Holzer, 1963), and that it is higher during growth on glutamate. As a result, this enzyme has been assumed to have a catabolic function. Variations in the concentrations of the NADP+ enzyme are small or absent depending on the wild-type strain considered (Dubois et al., 1974). The biosynthetic function of the NADP+-GDHase, however, is shown by the gdhA- mutants. The gdhA- mutation that abolishes NADP+-GDHase activity causes a generation time with ammonia of 240 minutes instead of 120 minutes as in wild type, whereas a normal rate of growth is attained by addition of glutamate to ammonia-containing medium (Grenson and Hou, 1972). The gdhA- mutations are located in the structural gene for the enzyme (Grenson et al., 1974). In the mutant, NAD+-GDHaseactivity remains low when grown in the presence of ammonia or ammonia and glutamate. This raises the problem of the origin of the residual growth (generation time 240 minutes) of the gdhA- mutant. Starting from a gdhA- parent strain, a mutant with an obligate requirement for glutamate has been selected. It has an additional mutation designated as ama-. By itself, the ama- mutation does not affect growth rate with ammonia.

b. Glutamate synthase and the ama- and gdhCR- mutations. In a number of Gram-negative bacteria, glutamate can be formed from ammonia by a combination of the activity of glutamine synthetase and a new enzyme reaction discovered by Tempest and collaborators. It is usually desig-

10

JEAN-MARIE WIAME ET AL.

nated as glutamate synthase (reaction I and reaction 11) (Tempest et al., 1970): NH,' + ATP + Glutamate- + Glutamine Glutamine + 2-Oxoglutarate2- + NADPH + H+

+ ADP + P, +

(1)

2 GlutamateNADP' + H20

2-Oxoglutarate2- + NH4+ + NADPH + H+ + ATP + Glutamate- + H 2 0 + ADP

+ P, + NADP'

(11) (I

i- 11)

Enteric bacteria have only one NAD(P)+-specific glutamate dehydrogenase, and loss of this enzyme by mutation does not impair the capacity to grow on ammonia at normal rates when glutamate synthase is present (Brenchley and Magasanik, 1974). Glutamate synthase is also present in Gram-positive bacteria. Indeed, in some of them, glutamate dehydrogenase is absent (Elmerich and Aubert, 1971; Elmerich, 1972). The bacterial glutamate synthase allows the organism not only to assimilate ammonia but also to use it at a much lower concentration than is possible when the bacterium possesses only glutamate dehydrogenase. This is possible because of the expenditure of one molecule of ATP. In S. cereuisiae, there is also a glutamate synthase, but it is a NAD+-requiring enzyme (Roon et al., 1974). The measured activity, ranging from 0.8 to 2 pmol hr-' mg protein-' depending on the growth medium, is quite low compared with that of NADP+-GDHase (25-50 pmol hr-I mg protein-') and of NAD+-GDHase (1-100 pmol). Results reported in Table 2 show that the ama- mutation leads to loss of glutamate synthase activity, and explain the absolute requirement for glutamate when ama- and gdhA- mutations are present together. This shows unambiguously that glutamate synthase may participate in ammonium assimilation. However, in contrast to bacteria, when the synthase is absent from a strain with a normal NADP+-GDHase, the presence of the ama- mutation does not modify growth rate even under conditions in which ammonia is limiting. Such conditions can be obtained with cytosine as the sole nitrogen source, the compound being slowly deaminated into uracil (see Table 2). So one cannot conclude that the NAD+-dependent glutamate synthase of S. cereuisiae helps the yeast to use ammonia at low concentration and, except for the artificial use of gdhA-, the physiological function of the glutamate synthase remains obscure. Thanks to the high affinity of the mepZ ammonia transport, S. cevevisiae could dispense with a prokaryotic NADP+-specific glutamate synthase. The double mutant gdhA-ama- (strain MG1694) has a normal NAD+GDHase and this enzyme is normally regulated. Its level is low in the presence of ammonia, a situation that explains the requirement for glutamate by the MG1694 strain. From this strain, mutants were selected

11

NITROGEN CATABOLITE REPRESSION IN FUNGI

TABLE 2. Glutamate synthase activity and the rate of growth of ama- mutants of S. cereuisiae"

Strains

Genotype

Nitrogenous nutrient

Generation time (minutes)

Glutamate synthase activity (pmol hr-I mg protein-')

~~

C1278b

wild type

I 156d

ama-

4324c

gdhA-

MG1694 ED245

gdhA- amagdhCR- gdhA-

NH4+ (20 mM) Chemostat Cytosine NH,' (20 mM) Cytosine (0.1%) NH4+ (20 mM) NH,+ + glutamate NH4+ (20 mM) NH4+ (20 mM)

120 375 330 120 330 220 120

no 240

1.3-1.6 1.4 50.01

2.4 1 .o -

ama

Unpublished results of F. Ramos, M. Grenson, and J. M. Wiame.

which have recovered the capacity to grow with ammonia. This led to the isolation of the gdhCR- mutation (regulation of the catabolic glutamate dehydrogenase) and shows its regulatory property which leads to derepression of the NAD+-specific enzyme in the presence of ammonia. This explains the recovery of growth with ammonia, the NAD+-GDHase with a normal catabolic function catalysing a reversible reaction which may partly compensate for the absence of NADP+-GDHase and glutamate synthase (Grenson et al., 1974; see also Table 2). The gdhCRmutation was shown to have a pleiotropic effect on a number of nitrogen catabolic enzymes (Dubois and Grenson, 1974). Mutations ure2 (Drillien and Lacroute, 1972) and usu (Grenson, 1969) allow ureidosuccinate to enter the cell and to satisfy the pyrimidine requirement of mutants defective in aspartate carbamoyltransferase. The ure2 mutants have a de-repressed level of NAD+-GDHase in the presence of ammonia (Drillien et al., 1973); usu and ure2 mutations are allelic with gdhCR- (Grenson et al., 1974). Today, the product of the gdhCR+ gene appears to be a general element in regulating metabolism of nitrogenous compounds. The gdhCR- effect will be considered for each individual enzyme and in the general view on regulation of nitrogen metabolism. A glutaminase does not appear to have been characterized in S. cereuisiae. Mutants specifically impaired in the use of glutamine as a nitrogen source have not been obtained (C. Hennaut and J. M. Wiame, unpub-

12

JEAN-MARIE WIAME E r AL.

lished observations). Glutamine amide nitrogen is frequently used as nitrogen donor for biosynthesis and could provide enough glutamate and aspartate for normal growth. L-Asparagine is deaminated by an active asparaginase I. This enzyme is not submitted to regulation, neither by induction nor by nitrogen regulation (Jones, 1973). This absence of regulation does not lead to a futile cycle because the endogenous pool of asparagine is very low. However, such a futile cycle can be created and it leads to a waste of asparagine and an asparagine requirement. This is obtained by a mutation that lowers the efficiency (increased K , and decreased V,,, values) of the asparaginyl-tRNA synthetase. Under these conditions, asparagine requirement disappears in mutants defective in asparaginase I (Ramos and Wiame, 1979).

C. DEGRADATION OF ARGININE A N D PROLINE

Anabolism, catabolism, and regulation of arginine metabolism have been studied in parallel in Saccharomyces cereuisiae. Among other regulatory functions this led to discoveries that could reflect on the process of nitrogen catabolite repression. This was first suggested by the Middelhoven starvation effect on arginase and ornithine transaminase activities (Middelhoven, 1968). Later, an intense ammonia effect was observed, which was compared with the glucose effect observed in prokaryotes, yeasts, and other organisms. The respective parts taken by the two opposite regulations, namely induction and “repression”, was left undefined (Wiame, 1971). A number of observations could be integrated into classical concepts, mostly derived from results and mod& proposed for prokaryotes. In addition, it led to the discovery of new aspects, some of them being unique for this amino acid and this micro-organism. Studies on enzyme structure and regulation went hand in hand. Most interestingly, enzyme structures and interactions appeared to be imposed to solve regulatory problems arising from conflicting metabolic situations. Here, much help came from comparisons between Neurospora crassa, S . cereuisiae, and other yeasts (Davis, 1963; Lacroute et al., 1965; Urrestarazu el al., 1977; Brandriss and Magasanik, 1981). Nitrogen catabolite repression is a part of a global problem which needs to be viewed in an integrated form. 1. Biochemistry and Gene-Enzyme Relationships

In previous studies, arginine degradation was viewed as a three-reaction pathway leading to one urea and two glutamate molecules. The enzymes

13

NITROGEN CATABOLITE REPRESSION IN FUNGI GLUTAMATE

I \

~

TGLUT A M YL

PHOSPHATE

/

/ehydrogenase

ARGJ NINE arginase

i

!

/

mitochondria

/

A+-PYRROLINE 4 5-CARBO XYLATE

I

I

L--

pro,lne

I

$

PROLINE

i

j

I

I I

0 x 1dase

_ _ _ _ _ _ _ _ _ _ _ - - - - - - - --

I I ----J

FIG. 1. A pathway showing the possibility of a proline-futile cycle in the early view of proline and arginine metabolism in Succharomyces cerevisiue. The figure should be compared with Fig. 2. Bold arrows denote catabolism, thin arrows anabolism.

involved in the pathway were identified as arginase, ornithine transaminase, and pyrroline 5-carboxylate dehydrogenase. Ornithine and pyrroline 5-carboxylate, generated by chemical cyclization of glutamic semialdehyde which arose as a result of the activity of ornithine transaminase, are the intermediates (Middelhoven, 1964; Wiame, 1964; BCchet and Wiame, 1965). Such a mechanism could lead to the generation of two futile cycles: one because ornithine is also an intermediate in biosynthesis of arginine, the other because pyrroline 5-carboxylate is a common intermediate between proline biosynthesis and degradation of proline as well as ornithine. Proline degradation was viewed as a two-step process including its oxidation by proline oxidase, well known to be a mitochondria1 enzyme (rho- petite mutants are unable to grow with proline as a source of nitrogen; see Fig. 1). The first futile cycle could arise because ornithine, resulting from arginine degradation by arginase, could be re-utilized by ornithine carbamoyltransferase (OTCase) going back to arginine via citrulline and argininosuccinate. This futile cycle is abolished by the epiarginasic regulation. This regulation results from reversible inhibition of OTCase by arginase under conditions of arginine catabolism. Arginase, resulting from induction, binds and inhibits the enzyme when

14

JEAN-MARIE WlAME ET AL.

FIG. 2. Pathways for degradation and synthesis of arginine and proline in S. cereuisiue. Bold arrows denote catabolism, thin arrows anabolism. (A) Acetylglutamate synthase (gene urgA or urg2); (B) acetylglutamate kinase (gene argB or urg6): (C) phosphoacetylglutamate reductase (gene urgC or urg5); (D) acetylglutamic acid semialdehyde aminotransferase (gene argD or arg8); (E) acetylornthine acetyltransferase (gene argE or arg7); (F) ornithine carbamoyltransferase (gene argF or arg3); (G) argininosuccinate synthetase (gene argC or a r g l ) ; (H) argininosuccinate Iyase (gene a r g H or urg4). Step (E) transfers an acetyl group on glutamate reducing the need for a direct acetylation by step (A) (not shown). Arginine metabolism is adapted from Urrestarazu et al. (1977) and Jauniaux et al. (1978). Arginine futile cycle is avoided by epiarginasic regulation as described by Messenguy and Wiame (1969), Wiame (1971), and Vissers et al. (1982). The proline futile cycle is avoided by a different localization of A-pyrroline 5-carboxylate degradation (mitochondria) and biosynthesis (cytosol) as shown by Brandriss (1979) and Brandriss and Magasanik (1979, 1980, 1981).

arginine and ornithine are present simultaneously. As soon as arginine is needed, OTCase activity is restored (Messenguy and Wiame, 1969; Wiame, 1971; Penninckx and Wiame, 1976; see Fig. 2). In filamentous fungi as well as in many yeasts, epiarginasic regulation is absent and the futile cycle is avoided by compartmentation. Ornithine is formed and converted into citrulline inside mitochondria because OTCase and carbamoylphosphate synthetase, as well as the biosynthetic enzymes leading from glutamate to ornithine, are also mitochondria1 enzymes (Weiss and Davis, 1973; Urrestarazu et af., 1977; Jauniaux et af., 1978; Vissers et al.,

NITROGEN CATABOLITE REPRESSION IN FUNGI

15

1982). So anabolic and catabolic ornithine are in different compartments. Avoidance of a futile cycle is under the control of exchange between cytosol and mitochondria. Recently, Brandriss and Magasanik (1981) have shown that the prolinefutile cycle which could be initiated by pyrroline 5-carboxylate is also avoided by subcellular cornpartmentation (cf. Figs. 1 and 2). Degradation of pyrroline 5-carboxylate by its dehydrogenase is an intramitochondrial process; reduction of pyrroline 5-carboxylate leading to proline is most probably cytosolic. This compartmentation has an important and unexpected consequence for arginine degradation. As shown by Brandriss and Magasanik (1980), in S. cerevisiae there is only one pyrroline 5-carboxylate dehydrogenasethis is at variance with Gram-positive bacteria (De Hauwer et al., 1964)and pyrroline 5-carboxylate, arising from the activity of ornithine transaminase, must enter mitochondria to reach the dehydrogenase. This, however, does not occur unless pyrroline 5-carboxylate is first reduced to proline. Proline is an obligate intermediate in ornithine degradation and, in turn, is oxidized by the intramitochondrial tandem of proline oxidase and pyrroline 5-carboxylate dehydrogenase. These enzymes are induced by proline. The rho- petite mutants not only are unable to grow on proline, but growth on ornithine is severely restricted and they excrete proline. Growth on arginine or ornithine indirectly induces mitochondria1 degradation of pyrroline 5-carboxylate because this compound, arising from ornithine transaminase activity, is efficiently converted by the constitutive pyrroline 5-carboxylate reductase into proline. The true inducer is proline. Cytosolic pyrroline 5-carboxylate, whether it arises from arginine degradation or from action of the two first enzymes of the proline biosynthetic pathway, cannot be degraded without its conversion into proline. The central role of proline and its repartition between cytosol and mitochondria provide a basis for control. Hence, arginine degradation is really an arginine-proline degradative pathway (Fig. 2). 2. Arginase

Arginase synthesis is subject to a number of regulatory mechanisms including specific induction (Middelhoven, 1964; Wiame, 1964), de-repression by starvation (Middelhoven, 1968), a strong ammonia effect variable from strain to strain (Wiame, 1971, and this review), and de-repression by growth under nitrogen-limiting conditions (Dubois et al., 1973). These observations form the origin of the study of nitrogen catabolite repression in yeast.

16

JEAN-MARIE WIAME f T AL.

a. Nitrogen catabolite repression is distinct from induction. Inability to

induce arginase (as well as ornithine transaminase) and stabilization at high constitutive levels for both enzymes have been obtained in mutants of the 21278b wild-type strain. A model was proposed to explain the properties of these mutants (Wiame, 1971; Dubois et al., 1978; Deschamps et al., 1979). The key to this model resides in the existence of a repressor (ambivalent repressor ARGR) identified by three classes of argR- mutations. This repressor has two distinct properties. First, it represses synthesis of anabolic enzymes and its target is an element analogous to an operator. This is the case of ornithine carbamoyltransferase for which the argF+O- (Ocphenotype) mutation alters the specific operator (Messenguy, 1976). This is also the case for the cluster of the argB,argC genes. The (argB,argC)Oc mutation leads to simultaneous constitutivity of N-acetylglutamate kinase and N-acetylglutamylphosphate reductase, the second and the third enzymes in arginine biosynthesis (P. Jacobs et al., 1982). Second, it inhibits the action of the repressor specific for the catabolic pathway. This repressor is designated as CARGR, identified by three classes of the cargR- mutations which confer constitutivity for arginase and ornithine transaminse. These mutations are recessive. CARGR could act on the specific operators for the two enzymes. The existence of these operators has been suggested by the cargA+O- (carlo-) and cargB+O- (car20-) cis-dominant mutations which confer a high constitutivity to arginase and ornithine transaminase, respecti vel y . In such a system, argR- mutations (three classes argRZ-, argRZZ-, and argRZZZ-) affect the synthesis of the ambivalent repressor and lead simultaneously to non-repressible anabolic enzyme synthesis and to low noninducible catabolic enzyme synthesis (a null-type phenotype). Whatever the detailed mechanism by which ARGR carries out its ambivalent function, it results (among other devices) in a balanced exclusion mechanism between anabolism and catabolism. As experimental tools, the aforementioned regulatory mechanisms and the corresponding mutations have proved to be invaluable in the study of both anabolic and catabolic pathways (see also Fincham et a f . , 1979; Jones and Fink, 1982, and later sections of this review). As shown in Table 3 , argR- mutants have half of the arginase activity compared with wild type when grown on a very good nitrogen source (ammonia), most probably because they lack the low endogenous induction. Addition of arginine to ammonia is without any effect. Growth on glutamate, proline, or in chemostat (ammonia-limited) increases arginase activity four- to sixfold in argR- mutants as well as in the wild type.

17

NITROGEN CATABOLITE REPRESSION IN FUNGI

TABLE 3. Induction, nitrogen catabolite repression, effects of starvation, and synergism for arginase in S . cereuisiue 21278b wild type and mutants"

Strains and genotype

Nitrogen source added

Arginase activity (pmol hr-' mg protein-')

Ammon ia Ammonia and arginine Proline or serine Chemostat (ammonia limited) Arginine Two-hour starvation (ammonia omitted) Ammonia argRII--lO Ammonia and arginine (non-inducible) Arginine Ammonia and proline Proline Chemostat (ammonia limited) Two-hour starvation (ammonia omitted) Ammonia and valine Valine Ammonia and glutamate gdhAAmmonia and glutamine Proline Ammonia and arginine Arginine Chemostat (ammonia or glutamine limited) Ammonia and glutamate gdhA- argRIIcargA+O-(cargA+OC) Ammonia and glutamate Ammonia and arginine (operator Ammonia constitutive) Chemostat (ammonia limited) Glutamate Arginine Two-hour starvation (ammonia omitted) Ammonia and glutamate g d h k , cargA+OAmmonia gdhCRChemostat (glutamine limited) glnChemostat (glutamine limited, excess ammonia)

Wild type

a

From Dubois et al. (1974, 1977) and Dubois and Wiame (1976).

6-8 20 20-25 35 250 87 3-4 3.5 N o growth 3.5 16 15 3

12 63 35 50 23 350 350 47

17 129 136 I38 222 230 350 215 185 45 42 9

18

JEAN-MARIE WIAME ET AL.

In the wild type the effect of adding arginine with ammonia provokes a fourfold increase. So nitrogen limitation in cells growing normally, as well as induction, has a rather modest effect. This strongly contrasts with the 40-fold increase observed when cells are grown on arginine alone (this effect will be considered later; see Section C.2.f). The cargA+O- mutants have much greater arginase activity than the wild type grown on ammonia and arginine; they are strictly insensitive to the presence of arginine in the presence of ammonia, but are sensitive to nitrogen limitation resulting from growth with proline, or glutamate or in chemostat (ammonia limited). The fact that the relative increase then is lower is probably due to the high activity already present in cargA+O- mutants when grown on ammonia, together with an upper limit in the overall capacity of expression of this gene. Recently, Courchesne and Magasanik (1983) reported an arginine activity in proline-grown cells that is less than twice the activity in cells grown on ammonia. The strain is an authentic X1278b. This is a lower variation when compared to values reported by the Brussels group, and the authors doubt the significance of this slight variation. Although fourfold variations (Dubois et al., 1974) are not dramatic, nevertheless they may indicate operation of a mechanism. Everyone remembers that the glucose effect on P-galactosidase synthesis was a reduction in activity by a factor of three (Tyler et al., 1967; Perlman et al., 1969). This modest effect was fortunately considered an important one. b. Release of nitrogen catabolite repression of arginase in NADP+-glutamate dehydrogenase mutants (gdhA-) compensated for glutamate auxotrophy. The discovery of the ammonia effect on the activity of the general amino-acid permease and its release by the gdhA- mutation under compensatory conditions (Grenson and Hou, 1972) called attention to the possibility that NADP+-GDHase could have a regulatory function in nitrogen catabolism. It seems well to be so. The gdhA- mutants growing on ammonia with glutamate or glutamine have increased arginase activity in the range of the release of nitrogen catabolite repression by nitrogen limitation. This was shown not only with the gdhA- mutation in a wildtype background, but also in combination with a non-inducible and with an arginase constitutive mutants (gdhA-, argR-, and gdhA-, cargA+O-; see Table 3). These last controls, only availabie for arginine catabolism at that time, at least for non-inducible mutants, excluded an indirect effect of induction. Arginase is also subject to the gdhCR- effect (Dubois and Grenson, 1974). A gdhCR- strain was selected for de-repression of NAD+-GDHase and it has a very high level of this enzyme which is derepressed during chemostat-limited growth and not sensitive to the gdhA-

19

NITROGEN CATABOLITE REPRESSION IN FUNGI

mutation. This suggests another regulatory circuit for nitrogen catabolite repression (Dubois et al., 1973). It seems that the product of the gdhCR+ gene has a more general regulatory function than the product of the gdhA+ gene. These elements could be involved either in a process initiated by the same or more logically by distinct effectors. Proof would require information on the nature of the effector(s) and a study of different nitrogen catabolic enzymes. Very few such enzymes were available and, among them, not one (sensitive to nitrogen catabolite repression) was known to be impaired by mutation in the induction process, except for arginase. It was only more recently that other possible enzymes were detected, for example urea amidolyase and asparaginase 11.

c. The nature of the metabolic signal of the arginase nitrogen catabolite repression and an attempt to understand the participation of the NADP+specijc, glutamate dehydrogenase in this repression (Dubois et al., 1974, 1977). The most obvious metabolic signal for nitrogen catabolite repression is glutamine. Glutamine and glutamate are by far the most frequent nitrogen donors in biosynthesis. Addition of glutamate does not restore nitrogen catabolite repression in gdhA- mutants growing on ammonia; glutamate provokes de-repression when it is the only nitrogen source for the wild type. Glutamine does not prevent the gdhA- effect (Dubois et al., 1974) although, like ammonia, when alone it keeps arginase at a low level in the wild type. One way to distinguish if ammonia is by itself the signal or if it must be converted into glutamine is to block this conversion by a gln- mutation (a glutamine auxotroph). As already shown, the gln- mutants grown on ammonia and glutamine show normal nitrogen catabolite TABLE 4. A comparison of arginase activity in different strains of S. cerevisiae Arginase activity” Nitrogenous nutrient

M25 diploidh

81278b’

Arginine pool in strain P1278bd

~

Ammonia Proline or serine Arginine and ammonia Arginine and serine Arginine a

9 67 17 100

3 8 8 28 100

Relative values, the level on arginine alone being taken as 100. Data from Bossinger et al. (1974). Data from Dubois er al. (1973, 1974). Measured in nanomoles per milligram of cells (dry weight).

80

300 178 800

20

JEAN-MARIE WIAME E T A L .

repression as do a number of gln- mutants. Glutamine synthetase does not appear to be a regulator by itself (Dubois and Grenson, 1974). This type of mutant can be used in experiments in which glutamine is provided as a limiting source of nitrogen. A gln- mutant growing in a chemostat limited by glutamine, is de-repressed for enzymes that are sensitive to nitrogen catabolite repression. If ammonia is a signal by itself, and does not need to be converted into glutamine (or a derivative), it should exert nitrogen catabolite repression when added in excess to a gln- mutant growing with glutamine as the limiting nitrogen source. This is what occurs for arginase; arginase activity returns to the same level as in the wild type growing on ammonia or glutamine. As a control for the method, and in favour of the participation of the NADP+-GDHase, one may mention that for the NAD+-specific enzyme, for which the gdhA- mutation has no effect, this method showed that ammonia itself had no effect. This indicates that, for that enzyme, ammonia must be converted into glutamine (Table 3). The relationship between these results was apparently overlooked in other reports (Marzluf, 1981; Cooper, 1982a). The possibility that NADP+-GDHase may have a regulatory function has been disputed on a number of different points. The most interesting involve allantoin-urea degradation and transport of nitrogenous compounds. Only a few data are available for arginase. In Table 4, results from Bossinger et al. (1974) are compared with the corresponding ones from the Brussels group (Table 3). The strains, as well as the growth media, are different. Bossinger et al. (1974) concluded that serine is a better repressor than ammonia and that ammonia should be converted into amino acids to exert nitrogen catabolite repression. Ammonia when included with arginine appears to have different effects on the strains: strain M25 is almost insensitive to ammonia compared with strain 21278b (12-fold). However, this ammonia effect does not necessarily reflect repression: it could involve inducer exclusion, and this is true for addition of serine as well (in this case, the two strains behave similarly). One of the most surprising results of this analysis is that Cooper and collaborators did not conclude that the strains may show different behaviour (see also comments on the same subject concerning urea amidolyase regulation, Section 1I.D). d . Nitrogen starvation compared with limitation: The role of induction. Arginase synthesis was promoted by nitrogen starvation (Middelhoven, 1968). Ammonia, when added to cells growing with arginine alone, strongly decreased arginase synthesis (Wiame, 1971). As Middlehoven (1968) proposed, one could have interpreted the result of starvation as the expression of release from nitrogen catabolite repression. Further work

NITROGEN CATABOLITE REPRESSION IN FUNGI

21

showed that this explanation must be re-investigated. Whitney and Magasanik (1973) did show that, in a mutant auxotrophic for arginine, there is no such an effect but that one may recover arginase synthesis by addition of homoarginine, a gratuitous inducer. This is clearly an indication that the starvation effect is totally or strongly dependent on the process of induction. This was confirmed when an argR- non-inducible mutant isogenic with strain C1278b showed no effect on starvation, compared with a 10-fold increase in the wild type Z1278b (Table 3; Dubois et al., 1974). As shown in the previous section, this is in striking contrast to the effect of nitrogen limitation, which promotes arginase synthesis in non-inducible argR- mutants. This is again good proof that nitrogen limitation and starvation (at least for arginase synthesis) involve different mechanisms. The occurrence of nitrogen catabolite repression in non-inducible mutants shows that one should not assume its absence on the basis that the starvation effect is the result of an induction process. It only shows that starvation is a misleading method to study nitrogen catabolite repression. e . Non-spec@ induction. An unexpected regulation of arginase has been designated as non-specific induction. Its physiological meaning is obscure but its occurrence is very clear and its amplitude is not small. It needs to be considered in this review because it may introduce confusing results when some amino acids are used as a source of nitrogen such as one may need for studies on nitrogen catabolite repression. A number of amino acids, such as valine, leucine, and a-aminobutyrate, but not proline, glutamate, ornithine, a,y-diaminobutyrate, y-aminobutyrate, homoarginine, or arginine, provoke an increase in arginase synthesis when added to a medium containing ammonia that supports growth of non-inducible argRmutants. Therefore this effect is not linked with the process of specific induction, and is retained under conditions of nitrogen catabolite repression. This last conclusion is in agreement with the fact that this effect is also found with a gdhA- mutant. As a result, when leucine or valine is used as sole nitrogen source, one may get as much as three times more arginase than with proline, which is usually used to relieve nitrogen catabolite repression (see Table 3). The effect is also present in other wild-type strains. However, it is not observed with ornithine transaminase or NAD+-GDHase (Dubois and Wiame, 1976). f . Catabolic synergism. Part of the induction mechanism for ornithine transaminase (OTAase) and arginase is common to both enzymes. The absence of induction in argR- mutants is well established. These mutants, in contrast to wild type, do not synthesize more of both enzymes when arginine is added to ammonia-growing cells; moreover, starvation does

22

JEAN-MARIE WIAME ETAL.

not promote their synthesis. However, nitrogen limitation increases arginase activity fourfold but has no effect on OTAase activity. These were the most direct data which led to the proposal of nitrogen catabolite repression for arginase synthesis and its absence for OTAase synthesis (see Section II.C.3). This proposition has been cross-checked by the observation of a de-repressive effect of gdhCR- and gdhA- mutations on arginase activity and its absence for OTAase activity. However, the simplest idea, that full enzyme synthesis would result from conjugation of induction and release from nitrogen catabolite repression, disagrees with the fact that, with the wild type, arginine alone (induction and release from catabolite repression) leads to a strong increase in activity of both enzymes when compared with growth on arginine and ammonia. The arginine pool in cells grown on ammonia and arginine (300 nmol mg cell dry wt.-') is lower than in those grown on arginine alone (800 nmol; see Table 4). Hence, one could see in this increment in the arginine pool the origin of a supplement of induction, this process being a progressive function. Measurement of the cytoplasmic arginine pool remains a methodological difficulty, although it was tentatively shown to increase by a factor of five when the total pool increased from 300 to 800 nmol mg cell dry wt.-' (Dubois and Wiarne, 1978). Quite different total arginine pools were induced using competition between arginine and other amino acids, as well as modification of permeability by mutations. Synthesis of OTAase and arginase behave differently in these different pools. For the former, there is a continuous increase, including the last 300 to 800 pool increment, which could agree with a synthesis that is controlled by exclusion of inducer (Deschamps ef al., 1979). Full arginase synthesis is obtained with an arginine pool three to four times smaller than the one obtained with arginine alone, with simultaneous release from nitrogen catabolite repression. This comparison does not suffer from the possibility of an indirect effect. If so, it should have been expressed for OTAase as well. Indeed, one could infer that release from nitrogen catabolite repression using proline (or glutamate) modifies intracellular compartmentation. This is perfectly possible, but then OTAase should react to this modification. It is quite striking that, whatever the mutations used and the composition of the growth medium, OTAase activities always lie on the same curve depending only on the total arginine pool. In other words, this enzyme is a reference activity (C. Hennaut, J. Perez, and J.M. Wiame, unpublished observations). These results enforce the previous hypothesis for a strong molecular interaction between the mechanism of induction and nitrogen catabolite repression at the level of the regulatory region adjacent to the cargA gene coding for arginase (Dubois and Wiame, 1978).

NITROGEN CATABOLITE REPRESSION IN FUNGI

23

3 . L-Ornithine Transaminase a. Absence of nitrogen repression. Synthesis of OTAase is regulated by the same pleiotropic induction of arginase. The argR- mutants have a low minimal level of OTAase activity. Constitutivity is produced by cargRrecessive and cargB+O- or cargB+Ohcis-dominant mutations adjacent to the cargB (car2) gene coding for OTAase. The cargB+Ohmutants have the peculiarity of being sensitive to mating-type signals. They belong to what has been designated as ROAM mutations (Regulated Overproducing Alleles responding to Mating-type signals). They are the result of insertion of Tyl transposable elements (Wiame, 1971; Deschamps et al., 1979; Deschamps and Wiame, 1979; Errede et al., 1980). The regulation of OTAase synthesis, however, differs widely from that of arginase by the absence of nitrogen catabolite repression. Neither growth on a poor nitrogen source such as proline, nor the use of mutants that result in derepression for other enzymes (gdhA- and gdhCR-) or in gln- mutants auxotrophic for glutamine growing with glutamine-limited promotes OTAase synthesis. Despite absence of sensu stricto nitrogen repression, OTAase synthesis induced by arginine is much greater in the absence of ammonia. This effect is thought to result from the exclusion of the inducer arginine by ammonia (see Section 1I.C.f). Middelhoven (1968) described a large burst of OTAase synthesis after nitrogen starvation. Perhaps because of differences in strains, we have never observed such a phenomenon (Deschamps et al., 1979). This point is worth re-investigation.

b. The effect of a nitrogen-rich medium. In gdhCR- mutants asparaginase I1 and urea amidolyase avoid nitrogen catabolite repression in media containing minimal concentrations of ammonia or glutamine. Surprisingly, constitutive synthesis of these enzymes (due to gdhCR- mutation) was strongly inhibited in a complex medium containing glucose, yeast extract, and bactopeptone. This may reflect an independent mechanism distinct from induction as well as from nitrogen catabolite repression. Synthesis of OTAase is a very suitable test for such a possibility because it is not submitted to nitrogen catabolite repression and we possess cargB+O- constitutive mutants. The activity of OTAase in cargB+O--l mutant was severely lowered (16fold) in a nitrogen-rich medium compared with the activity in a medium containing minimal ammonia. The activity of cargB+O--2 mutant was lowered fourfold and the cargB+Oh mutants showed only a two- to threefold decrease in activity. The evolution of the activity as a function of growth has shown that this modification was the result of repression of synthesis and not the result of a

24

JEAN-MARIE WIAME ET AL.

modification of the activity of the enzyme molecules (C. Hennaut and J.-M. Wiame, unpublished results). This strengthens the idea that the effect of the nitrogen-rich medium involves an additional regulatory mechanism. The rationale for such a regulation is obvious. In the yeast cells the numerous permeases allow all of the nitrogen building blocks to be utilized. The need for biosynthesis being minimal, the utilization of the most frequent nitrogen donor molecules, ammonia, glutamate, or glutamine, is minimized. The synthesis of nitrogen catabolic enzymes as well as anabolic enzymes is a waste. 4. Proline Degradation

The enzymology of proline degradation and the way by which the pathway for arginine degradation intersects the proline degradative pathway are described in Fig. 2. Conversion of proline into glutamate is the result of oxidation of proline into pyrroline 5-carboxylate by proline oxidase, the electron leaving the mitochondria most probably by the respiratory chain. Pyrroline 5-carboxylate is then oxidized by a dehydrogenase, which uses NAD+ or NADP+ as cofactor. There is only one pyrroline 5-carboxylate dehydrogenase in S . cereuisiae, and proline oxidase and this dehydrogenase are both induced only by proline. The apparent induction by arginine and proline resulted from a sequential process (Brandriss and Magasanik, 1979). Nitrogen starvation induced synthesis of arginase and proline oxidase. Arginase-less strains, when starved, showed similar increases in proline oxidase indicating that starvation induced proline oxidase independently from arginine degradation. From a regulatory point of view, the arginineornithine catabolic segment is distinct from the proline one (Brandriss and Magasanik, 1980). So far, there is no indication that the proline catabolic segment is submitted to a regulation distinct from induction. It is proposed that induction and inducer exclusion are the regulatory mechanisms for the proline-specific degradation. There is no evidence for nitrogen catabolite repression of the specific proline pathway, except at the level of proline transport (see Section 1I.J).

D . DEGRADATION OF ALLANTOIN A N D UREA

The allantion-urea pathway will be considered in detail. At present, it is the most suitable tool for a study of nitrogen catabolite repression because an enzyme of this pathway, urea amidolyase (Roon and Levenberg, 1968),

NITROGEN CATABOLITE REPRESSION IN FUNGI

25

was strongly subject to nitrogen catabolite repression (Dubois et a f . , 1973), as well as to induction (Whitney et al., 1973). The effects have been separated by the discovery of non-inducible mutants which remain fully subject to nitrogen catabolite repression (Lemoine et a f . , 1978; E. Jacobs et al., 1980, 1981; Turoscy and Cooper, 1982). One of the most interesting features of this enzyme is that it raises the possibility that one enzyme could be submitted to two distinct nitrogen catabolite repression mechanisms with two different signals (Dubois et a f . , 1977). 1. Biochemistry and Genetics The nitrogen atoms included in the pyrimidine and the purine nucleus were not utilized as a source of nitrogen by Saccharomyces cereuisiae. One nitrogen was available with adenine and cytosine because of a free amino group and, with cytosine, uracil was excreted (Grenson, 1969, 1973). The four nitrogen atoms of allantoin, which result from the opening of the pyrimidine part of the purine nucleus, were used as nitrogen sources through a classical transformation into two molecules of urea and one of glyoxylic acid (Vogels and van der Drift, 1976). Urea degradation is complex. Instead of a urease, Roon and Levenberg (1968) discovered an ATP-dependent enzymic activity which they designated as urea amidolyase. First observed in Candida utilis, this system was shown to occur in S . cereuisiue. The need for ATP was correlated with a two-step enzymic process beginning with an ATP- and biotin-dependent carboxylation (urea carboxylase) leading to allophanate (urea carboxylate), followed by hydrolysis of the latter (Whitney and Cooper, 1970; Roon and Levenberg, 1970). Most budding yeasts synthesize this enzyme; some strictly aerobic yeasts as well as fission yeasts and mycelial fungi do not, and produce urease instead. In S . cereuisiae, the biochemistry and the genetics of the pathway have been investigated by Cooper and collaborators and reviewed recently (Cooper, 1982b). Figure 3 summarizes the system. Allantoin-urea degradation comprises three associations of genetic elements. These are the dull ,da14,da12 cluster (allantoinase, allantoin permease, and allantoicase), the dur3,dur4 cluster (urea active transport and facilitated diffusion), and the durZ,dur2 association first believed to be two genes (Whitney and Cooper, 1972) on the basis of complementarity between durZ- (no urea carboxylase activity) and dur2- (no allophanate hydrolase activity). Today, it is known that durZ,dur2 forms a single unit of transcription coding for a single polypeptidic protein (Cooper et al., 1980; Cooper, 1982a). A part of the (durZ,2) DNA has been cloned; it includes the dur2

26

JEAN-MARIE WlAME ET AL. dal4 ALLAN TO1 N (external)-AL

ALLA N TOATE dal5 dal2

dur4

K,

UREA (external) dur3 K , urea amidol yase bifunctional enzyme

[

1

LANTOI N (internal j I

HO

2 RE1DOGLYCOLIC

0.25mM

UREA --hntcrno// 10mM du:;To2

GLYOXYLIC ACID carboxylatton urea

AL LOPHANIC ACID allophanate hydrolysis dur2 2NH

+

2C02

FIG. 3. Pathway for degradation of allantoin and urea, showing enzymes, transport systems, and their expected structural genes. Adapted from Cooper (1982).

region, a part of the durl one, and the regulatory region including the site of insertion of a Tyl (Dubois et al., 1982). This Tyl insertion is responsible for the mating-type effect observed in durOhROAM type of mutations previously described by Lemoine et al. (1978) and by E. Jacobs et al. (1981). 2. Induction, Starvation, and the Ammonia Eflect Induction of urea amidolyase was first established by an increase in activity in response to addition of urea to minimal (ammonia) medium and with the M25 wild-type diploid. Allophanate, the product of the first catalytic activity of this bifunctional enzyme, was shown to be the true inducer. A durl- strain was not induced by urea (Whitney and Cooper, 1972; Whitney et al., 1973). Starvation for nitrogen enhanced synthesis of allophanate hydrolase, and this effect was almost abolished in a mutant lacking arginase or urea carboxylase activities (Whitney et al., 1973), an effect similar to the one observed for arginase in that it was an induction-dependent effect (Whitney and Magasanik, 1973). Oxalurate is a non-metabolizable inducer (Sumrada and Cooper, 1974).

NITROGEN CATABOLITE REPRESSION IN FUNGI

27

The first indication of nitrogen catabolite repression of urea amidolyase was obtained with strain 21278b. With this strain, there is also induction by addition of urea to cells growing in minimal (ammonia) medium (a 20fold effect). In addition, urea alone increased enzyme activity when compared with that in cells grown on urea and ammonia. This last effect could be due to two different mechanisms: either an exclusion of the inducer by ammonia or true nitrogen catabolite repression. This last possibility was favoured because growth with glutamate (which is not expected to provoke induction) also caused enhancement of urea amidolyase when compared with ammonia (a 20-fold effect). Because of these effects, urea amidolyase appeared as an appropriate enzyme to study nitrogen catabolite repression as distinct from induction (Dubois et al., 1973). These results led to a re-investigation of urea amidolyase regulation in the M25 diploid strain by Cooper and collaborators (Bossinger et al., 1974). They showed that ammonia does not lower the activity when added to ureagrown cells, a result consistent with the fact that glutamate- or aspartategrown cells have the same concentration of enzyme as ammonia-grown cells. Using strain C1278b, they observed a much smaller effect than the one reported by Dubois et al. (1973). This last observation is a more critical one because it is at the origin of the controversy concerning the mechanism of the ammonia effect. It needs further consideration (see Section II.D.4). Subsequently, Cooper (1978), for obscure reasons, discovered an extremely strong ammonia effect (40-fold) with strain C 1278b, for the same enzyme, apparently using a similar medium to the one used previously. This was reported without comment.

3. Ammonia Effect, Nitrogen Catabolite Repression, and Glutamate Dehydrogenase ( N A D P ) The finding of an “ammonia effect” on the activity of the general aminoacid permease and its release in a mutant devoid of NADP+-GDHase (gdhA- mutant; Grenson and Hou, 1972) led to the question of whether the gdhA- mutation alters the effect of ammonia on enzyme expression. This was shown to be so for some enzymes such as urea amidolyase, allantoinase, and arginase. Among different mechanisms, this opens up the possibility that the NADP+-GDHase could be the receptor of ammonia itself as a metabolic signal. Arginase was the first enzyme used to test this hypothesis in detail (Dubois et al., 1974). This study led to open controversies which need to be analysed. For urea amidolyase, the tools to study nitrogen catabolite repression have been greatly improved. Most important is the selection of mutants that have lost the ability to be induced. They belong to at least two genetic

28

JEAN-MARIE WIAME E T AL.

complementation classes, durM- (such as durM--1 in strain 13H9b; Lemoine et al., 1978) and durL- (E. Jacobs et al., 1980, 1981). The properties of these mutants do not differ substantially. Mutant du181, isolated by Turoscy and Cooper (1982), could be allelic with one or the other durMor durL- mutations. The properties of the da181-Z mutant are similar to the others as far as induction is concerned, but the nitrogen catabolite repression may be different because it has another genetic background. The most important results for this discussion are shown in Table 5 and can be summarized as follows: 1. With the wild type, growth on glutamate or proline enhanced urea amidolyase activity by some 50- to 100-fold compared with growth on ammonia or glutamine. Activity with glutamine or asparagine was lower than with ammonia (we return to this point later). 2. Non-inducible durM--Z and other similar mutants have a degree of de-repression of the same magnitude when grown with glutamate or proline, as well as with urea. In the absence of induction, and being noninducible, this is solely the result of release from nitrogen catabolite repression. 3 . The gdhA- mutation alone, or in combination with one durM- mutant, promotes enzyme synthesis under the best conditions of repression (gdhA- mutation is compensated for the lack of NADP+-GDHase activitv by addition of glutamine or glutamate). These results do not prove the validity of the hypothesis that the amm nia effect is wansmitted through NADP+-GDHase, but they confirm some earlier data in a more rigorous way than was possible before 1978. The fact that glutamine does not overcome the gdhA- mutation was already known (Dubois et al., 1977) and is confirmed with these data. These results also present an answer to an objection given before by van de Poll (1973) who observed that glutamate together with ammonia or glutamine overcome the gdhA- effect in another yeast and with a partly different methodology. One may conclude that, with the C1278b strain, the ammonia NADP+-GDHase regulatory hypothesis remains an attractive one. 4. The Possibility of More than One Regulatory Circuit for Nitrogen

Catabolite Repression The possibility of the existence of at least two distinct mechanisms of nitrogen catabolite repression was already obvious when arginase and urea amidolyase were compared with the ammonia effect on NAD+GDHase, which is completely insensitive to gdhA- mutations (Dubois et al., 1973). The simplest hypothesis was that ammonia in NAD+-GDHase

29

NITROGEN CATABOLITE REPRESSION IN FUNGI

TABLE 5. Allophanate hydrolase activity of strains of S. cerevisiae"

Strains and genotypes Wild type (21278b)

durM--l non-inducible (13H9b)b

gdhAgdhA-,durM-' gdhCR-

Nitrogen source in growth media Ammonia Ammonia and urea Urea Glutamine Asparagine Glutamate Proline Ammonia Ammonia and glutamine Ammonia and urea Proline Glutamate Urea Ammonia and glutamate Ammonia and glutamine Ammonia and glutamate Ammonia and glutamine Ammonia

Allophanate hydrolase (pmol hr-' mg protein-') 40 650 3650 3 6 590 266 13 425 630 770 740

12 3 14 810 750 705 340 454 220 335 589 273

1170

From Lemoine er al. (1978). Strains durM--2, M - - 3 ; durL--l, L--2, L -3, L--4, L--5 are not essentially different from durM--I (E. Jacobs et a / . , 1981). Unpublished data of E. Dubois, C. Hennaut, and J . M. Wiame. a

regulation is an apparent effector, the real one being a derivative. As glutamate in strain Z 1278b de-represses synthesis of nitrogen catabolic enzymes, the best remaining candidate is glutamine. The effect of ammonia on the NAD+-GDHase could be investigated only as far as synthesis of glutamine could be blocked. Glutamine auxotrophs were obtained by Dubois and Grenson (1974); they were used as described for arginase and the NAD+-GDHase. The method gave a very clear-cut answer (see Sections 1I.C and F). In contrast to arginase, NAD+-GDHase was insensitive to ammonia when it could not be converted into glutamine: a gln- mutant growing with glutamine as the limiting nitrogen source was completely insensitive to the presence of ammonia when added in excess (Dubois et al., 1977). Instead of a glutamine limitation obtained in a chemostat, one may use either a thermosensitive ginfsmutant or a combination of a glnand a gnrR- mutation (Dubois et al., 1977). The gnrR- mutation has been shown to limit transport of glutamine (M. Grenson, unpublished observations). Ammonia in excess, under conditions of glutamine limitation, resulted in an important but partial repression of urea amidolyase synthesis,

30

JEAN-MARIE WIAME € T A L .

showing that ammonia exerts its repression independently from its conversion into glutamine, which is also an effector. This lead to the conclusion that the same enzyme could be sensitive to an ammonia- as well as a glutamine (or derivative)-dependent regulation (Dubois et al., 1977). This finding was able to explain published discrepancies. Strains of S. cerevisiae such as M25 (Bossinger et al., 1974; Bossinger and Cooper, 1975), which do not show an ammonia effect (urea amidolyase production was the same with urea or urea together with ammonia), should be insensitive to gdhA- mutations when this mutation is introduced into the M25 genetic background as was the case in the work of Bossinger and Cooper (1975). The fact reported by these authors and some reviewers that this is true also for strain Z1278b contradicts more recent results (Cooper, 1978). Unfortunately, apparently being aware of the ammonia sensitivity described previously by Dubois et al. (1973), Cooper and collaborators did not reinvestigate the gdhA- effect, which they only studied in the M25 diploid genetic background. So, the obvious contradiction seems to be solved for that enzyme. The observed glutamine effect is the expression of the second mechanism.

5 . The gdhCR- Mutation and the Glutamine Circuit Dubois and Grenson (1974) showed that the gdhCR- mutation has a derepressive effect on all nitrogen catabolic enzymes sensitive to nitrogen catabolite repression. Some of these enzymes were already known to be sensitive to the gdhA- mutation but, as already reported, the NAD+GDHase was an exception that called for another regulatory circuit, distinct from the ammonia-gdhA- circuit. The product of the gdhCR gene (GDHCR) could be an element involved in the glutamine circuit as well as in the ammonia circuit. For urea amidolyase and allantoinase, for which it was shown that ammonia and glutamine are separate effectors, it was observed that the gdhCR- mutation had a stronger effect than the gdhA- mutation alone and was of the same magnitude as the one observed in chemostat limitation (Dubois et al., 1977; Lemoine et al., 1978; Table 5). 6 . Constitutivity Mutations Acting in cis and Under the Control of the Mating Type (ROAM Mutations)

Selection of mutants recovering growth on allantoin in durM- strains led to the isolation of cis-dominant mutations adjacent to the structural gene

NITROGEN CATABOLITE REPRESSION IN FUNGI

31

for urea amidolyase and designated as durOh.All of these mutations are of the ROAM type (Lemoine el al., 1978; E. Jacobs et al., 1981). These mutations result from insertion of Ty 1 transposable elements (Errede et al., 1980). These insertions may be located at different sites in the 5’ noncoding region of the gene as shown in the case of the two cargA+Ok-land cargA+Ok-2ROAM mutations inserted in the adjacent region for the arginase gene. In that case, the two insertions were separated by about 600 base pairs (Dubois et al., 1978; Jauniaux et al., 1982).The present interest in the study of these mutations for regulation of urea amidolyase is that they modify the regulation in different ways. Some have lost the interaction with the induction mechanism but retain nitrogen catabolite repression; others are inducible but do not show release from nitrogen catabolite repression by growth on glutamate (E. Jacobs et al., 1981). 7. The Effect of a Nitrogen-Rich Medium Urea amidolyase activity was very low when cells were grown in a nitrogen-rich medium such as one containing yeast extract and bactopeptone (glucose being the carbon source). This effect was not abolished in gdhCR- or gdhA- mutants. Casamino acids also lower the activity although their effect was significantly smaller. No component or group of components of this amino-acid mixture could be recognized as specifically responsible for this effect. A dur(1 ,2)Oh-I mutant was only slightly sensitive to the effect of a nitrogen-rich medium (E. Jacobs, E. Dubois, and J.-M. Wiame, unpublished results; see also sections on asparaginase I1 and ornithine transaminase).

E. ASPARAGINASE

11

Some strains of Saccharomyces cerevisiae have two asparaginases (I and 11); others have only asparaginase I. Asparaginase I is strictly intracellular, specific for L-asparagine, and is not regulated by induction or by nitrogen catabolite repression. It is absent from asp1 mutants. Asparaginase 11 hydrolyses D and L isomers, and can be excreted under conditions of nitrogen starvation. It is absent from asp3 mutants; asp2 mutations also affect its activity (Dunlop et al., 1978; Jones, 1978). Adequate crosses and backcrosses have been used to introduce asparaginase I1 in the genetic background of the 21278b wild type or isogenic gdhA- or gdhCR- mutants. Asparaginase I1 is not induced by D-asparagine; it is almost absent from cells grown with very good nutrients (ammonia, glutamine, aspara-

32

JEAN-MARIE WIAME ET AL.

gine) and is present under conditions of nitrogen-limiting nutrition and nitrogen starvation. This enzyme is a good tool to study nitrogen catabolite repression, and has been used by Roon and collaborators (Dunlop et al., 1980, 1982). Their studies do not use the classical way of expressing the level of enzyme in cells grown with nitrogen sources of different quality. This analysis (Dunlop et al., 1980; Kang et al., 1982) was essentially based on the transfer of cells into a starvation medium (without any nitrogen). The level after starvation was the highest observed and was used for comparison with other similar treatments involving individual compounds. Samples of cells were collected (after 4 or 5 hours) and activity was given per dry weight of cells introduced before the shift. Growth was not followed. Some compounds, such as lysine and methylamine, are not nitrogen sources; many amino acids are very poor substrates (e.g., histidine), whereas others (arginine, serine) are good ones. Glutamate and proline, which are typical de-repressing nutrients, were not included among the compounds tested. The evolution of activity was not followed during the 4-5 hour treatment. Their results can be summarized as follows. In all cases, activity was lower than when a nitrogenous compound was added and in most cases it was less than 15% of the activity under conditions of simple starvation. Strikingly, this was the case not only for very good or good nutrients but also for lysine and methylamine. Activity in the gdhCR- mutants was significantly higher than in the wild type (with 10 mM methylamine, it is some 80%, compared with less than 15% in the wild type). The gdhA- mutant usually behaved as wild type except with ammonia, but in that case nitrogen was limiting. The conclusion of the authors was as follows. Synthesis of enzyme produced under conditions of simple starvation was inhibited by any amino acid and even by alkylamine behaving as an analogue of ammonia. They concluded, “In summary, the data presented here raise the possibility that no single amino acid is the unique co-repressor in nitrogen catabolite repression systems and suggest that further studies on nitrogen catabolite repression should focus on the following related questions: (i) How broad is the effector specificity for nitrogen catabolite repression system? (ii) Are alkylamines other than amino acids capable of functioning as effectors? (iii) Must the effectors be metabolized to exert their repressible effect, i.e. must they donate their nitrogen atom to some acceptor?” There are obviously too few data to begin to understand these experiments. They may involve more than one mechanism. The opinion that any one compound, whatever its capacity as a nitrogen nutrient, can be a co-repressor of nitrogen catabolite repression certainly calls for a determination of asparaginase I1 inside cells being grown in permanent exponential phase with different nitrogen sources. Some

NITROGEN CATABOLITE REPRESSION IN FUNGI

33

preliminary data from our laboratory are given in Table 6 (M. Legrain, E. Dubois, and J.-M. Wiame, unpublished observations). Asparaginase I1 is under the control of the gdhCR+ gene product and probably under the control of glutamine. De-repression resulting from the gdhCR- mutation is very strong and is retained in the presence of ammonia as well as with glutamine. With the gdhA- mutant, addition of glutamine and ammonia repressed the enzyme to a low level; the NADP+-GDHase does not seem to be involved in nitrogen catabolite repression of asparaginase 11. Experiments with the gln- mutant, glutamine being limiting with excess ammonia, should be carried out as for other gdhCR-sensitive enzymes. The TABLE 6. Asparaginase 11-specific activities in strains of S. cercvisicw grown exponentially"

Strain ~~

15L4a

23L5a

Genotype

Nitrogen source added to medium

Asparaginase I1 activity (wmol hr-I mg protein-' at 30°C)

~

casnll+c

casnl-, casnIl+, gdhCR-

Ammonia Glutamine Proline Proline and ammonia Glutamate D- Asparagine D-Asparagine and ammonia D-Asparagine and proline L- Asparagine Arginine Medium 863d Ammonia Glutamine Proline D-Asparagine Medium 863d

0.4 0.34 95.0

0.43 18.0 2.0 0.7 5.0 3.0 4.0 <0.2 82.0 36.0 101.0

64.0 0.3

Unpublished data of M. Legrain, E. Dubois, and J. M. Wiame. Strain 81278b is a wild type without asparaginase II activity (casnll- strain). The casn/l+ gene (probably allelic with asp3+)has been introduced by multiple backcrosses with 21278b to ensure its usual genetic background towards nitrogen catabolite repression. This was checked for other enzymes. Asparaginase 11 activity was measured with D-asparagine as substrate in French-press extracts of cells grown exponentially. The casnl- mutation was selected from strain X1278b for absence of the capacity to use L-asparagine as the sole nitrogen source (F. Ramos and J.-M. Wiame, unpublished observations). The casnl- mutation is probably allelic with asp/ mutation. Medium 863 contained yeast extract and bactopeptone. Glucose was the carbon source.

34

JEAN-MARIE WIAME €1 AL.

effect of a nitrogen-rich medium is another interesting aspect. It overcomes the gdhCR- effect and could be the expression of another mechanism, as discussed for ornithine transaminase and urea amidolyase.

F. GLUTAMATE DEHYDROGENASES

There are probably no enzymes in fungi which have been studied more intensively than the glutamate dehydrogenases. As presented in the section dealing with the early steps of nitrogen assimilation, there is a general agreement on their functions in uiuo: NADP-GDHase is an anabolic enzyme and NAD+-GDHase is catabolic. By contrast with the importance of these enzymes in nitrogen metabolism, the genetic element governing expression of both enzymes are more limited than for some other enzymes. For Saccharomyces cereuisiae, the anabolic enzyme remains the only one for which the structural gene for the enzyme, the gdhA (gdhl) gene, has been identified (Grenson el al., 1974). The structural gene for the catabolic enzyme has not been identified. The gdhCS- mutation (Dubois and Grenson, 1974), and others obtained by Middelhoven et al. (1978), with a very low (if not null) activity may be regulatory as well as structural gene mutations. Whatever their nature, it would be important to know if they are in the structural gene, in its promoter, or unlinked. Any answer to that question is important. The gdhCR- mutation (see Section 11) selected for high NAD+-GDHase activity appears to be a true regulatory mutation and, so far, the only one with a general effect on nitrogen catabolite repression. It could represent a mutation in the structure of a negatively acting repressor of nitrogen catabolite repression (see Section 1I.Q). When the level of activity of the NAD+-GDHase was followed under many different conditions of growth, and in the same set of isogenic strains (derived from C1278b strain), it was clear that the gdhCR- mutation promoted NAD+-GDHase activity very strongly (Dubois and Grenson, 1974). It was also established that the gdhA- mutation had no effect (Dubois el al., 1973). This is in good agreement with the absence of an ammonia repression when ammonia cannot be converted into glutamine (Dubois et al. 1977). However, regulation of this enzyme remains largely unknown. A process of specific induction by glutamate was usually accepted in the early studies by Holzer and collaborators, mainly because of a higher concentration of the enzyme in cells grown on glutamate, but an ammonia effect was observed simultaneously (Hierholzer and Holzer, 1963). To date, one cannot exclude or conclusively propose a process of induction. This is especially obvious when either conditions of growth or the use of gdhCR

35

NITROGEN CATABOLITE REPRESSION IN FUNGI

mutants are compared. The NAD+-GDHase activity values vary too much when one compares cells growing in a chemostat under nitrogen limitation and the effect of gdhCR mutation (Dubois et a f . , 1977). One must suspect participation of more than one regulation process. A possibility comes from studies with Candida utilis. This yeast is very different from S. cereuisiae. It can utilize glutamate as carbon source and it is probably under the control of carbon as well as nitrogen. A reversible inactivation of the NAD+-GDHase was also observed; this inactivation was mediated by protein phosphorylation (Hemmings, 1978), which in turn was promoted by glutamate starvation. Re-activation was observed in cell-free extracts as well as with Escherichia coli phosphatase. The significance of this inactivation could be that, in the absence of glutamate, the cell saves glutamate from degradation (Hemmings, 1978). Inactivation of NAD+-GDHase in S . cereuisiae should be studied. One cannot infer from results observed with C. utifis what occurs in another physiologically different yeast. However, variations observed by Dubois et a f . (1977) may result from a process that implies a modification of enzyme activity in addition to its synthesis, in spite of the absence of starvation conditions. This was clearly the case for glutamine synthetase where repression and inactivation occurred in non-starvation conditions. The NADP+-GDHase from C. utifis was rapidly inactivated when cells were starved of carbon source; this was an irreversible process. Loss of activity parallelled loss of precipitable antigenic material when cells were transferred from a medium containing ammonia and glucose to one containing only glutamate. Re-activation was observed when cells were returned to the former culture conditions, and this was shown to be due to de nouo synthesis of enzyme (Hemmings, 1978). Mazon (1978) reported inactivation of S . cereuisiae when cells grown on ammonia and glucose were deprived of glucose. Cycloheximide prevented recovery of activity, suggesting the need for de n o w synthesis. Mazon (1978) observed also that the NAD+-GDHase activity increased four- to sixfold under conditions that provoked inactivation of the anabolic enzyme. These modifications may vary from strain to strain, but they were not observed by Roon and Even (1973).

G . GLUTAMINE SYNTHETASE A N D PROTEINASE

B

Glutamine synthesis from glutamate is catalysed by a single enzyme, glutamine synthetase. Glutamine has two functions: it is specifically needed for protein synthesis and it is needed as a nitrogen donor for synthesis of many building blocks, precursors, and macromolecules. Am-

36

JEAN-MARIE WIAME ETAL.

monia has already been reported to repress glutamine synthetase in baker's yeast (Kolhaw et a / . , 1965). Glutamine itself caused inactivation in Candida utilis (Ferguson and Sims, 1974), which was preceded by a reversible dissociation of the enzyme (Sims and Ferguson, 1974; Sims et al., 1974a,b). Reversible inactivation by glutamine was first observed by Holzer and his collaborators in enteric bacteria, a process that received much attention (Stadtman and Ginsburg, 1974). In Saccharomyces cereuisiae (strain Z1278b), Dubois and Grenson (1974) reported that, in the wild type, glutamine synthetase activity was highest in glutamate-grown cells (15 pmol hr-I mg protein-'), lower in ammonia-grown cells (5 pmol), and still lower in cells grown on ammonia and glutamate (3 pmol). Growth on glutamine led to barely detectable activity (0.1 pmol). The gdhA- mutation did not modify these values. In the gdhCR- mutant, an ammonium effect was retained but the large glutamine effect strikingly disappeared ( 5 pmol instead of 0.1 pmol), and the gdhCR- mutant did not undergo its usual release from nitrogen catabolite repression (loss of repression reported for arginase and urea amidolyase). Evolution of glutamine synthetase activity after addition of ammonium and glutamine, which has reported as a function of total protein (in the plot of Monod et al., 1952), helps elucidate the gdhCR- mutation effect on release from the glutamine effect (Legrain et al., 1982). When ammonia was added to cells growing on glutamate, synthesis of glutamine synthetase was repressed. This repression was the same in the wild type and the gdhCR- mutant. It was not likely to be an end product (glutamhe)-mediated repression, because the glutamine pool in cells growing on glutamate was three times higher than that observed with cells grown on ammonia with glutamate or on ammonia alone. Addition of glutamine resulted, in the wild type, in an inactivation process (50% loss of activity in a quarter of a generation time). In the gdhCR- mutant, this inactivation was totally absent, so revealing the repression (similar to the one due to addition of ammonia) which was masked in the wild type by the massive inactivation. After removal of glutamine, glutamine synthetase activity recovered by 50% in 40 minutes. Reversibility could not be demonstrated unambiguously. Starvation for a nitrogen source produced another type of inactivation, which proceeded independently of the product of the gdhCR+ gene. Among all reported cases of enzyme inactivation, this one appeared to be the first found to be suppressed by the Jones pep4-3 mutation (Hemmings et al., 1980). Other cases of inactivation were shown to be the result of proteolysis but by proteinases different from A, B , or C (Holzer, 1976; Wolf, 1982).

NITROGEN CATABOLITE REPRESSION IN FUNGI

37

Although glutamine synthetase in crude extracts of gdhCR- mutant was more heat labile than in the wild type, this was not true for partially purified preparations. Enhanced heat lability was shown to result from a higher proteinase B activity in the gdhCR- mutant. This opens the possibility that proteinase B is subject to nitrogen catabolite repression. Proteinases A and C did not show any change of activity in the gdhCRmutant (Legrain et al., 1982).

H . UPTAKE SYSTEMS FOR NITROGEN-CONTAINING COMPOUNDS

Uptake systems for nitrogen-containing compounds play an important role in regulating nitrogen metabolism in yeast. It is clear from the preceding sections that nitrogen sources that support fastest growth rates of Saccharomyces cereuisiae, namely ammonium ions, glutamine, and aspargine, prevent a number of other nitrogenous compounds from being utilized. The mechanisms by which this exclusion is effected involve a control at the level of uptake of nitrogenous nutrients into the cell. A number of uptake systems for nitrogenous compounds are not active in cells grown in the presence of ammonium ions. This appears to be a crucial aspect of the regulation of nitrogen catabolism in yeast, and is the subject of the present section. However, catalytic properties, namely kinetic parameters and specificity, of the uptake systems for nitrogenous substances also play a role in the physiological adaptation of yeast to nitrogen utilization. A number of uptake systems accumulate amino acids from the external medium into the cytoplasm of S . cereuisiae (Table 7). In many cases, a given substrate is transported by several permeases which are encoded by distinct structural genes, and which exhibit different properties with respect to substrate affinity, specificity, capacity, and regulation. The multiplicity and the diversity of these uptake systems allow the yeast cell to accumulate amino acids for biosynthetic as well as for catabolic activities under a wide variety of growth conditions. As a rule, there is no exclusive utilization of a given permease for a specific purpose, and no channelling has been detected in S. cereuisiae at the level of amino acid uptake systems. For instance, L-arginine can be transported equally efficiently by two permeases, namely the specific arginine permease and the general amino-acid permease (GAP), either to fulfil a specific arginine requirement, or to be used as a general nitrogen source. From a catalytic point of view, the two permeases are equivalent. However, since the GAP has a broad specificity and is not active in ammonia-grown cells, it can be used for arginine transport only in the absence of the “ammonia effect” and of

TABLE 7. Amino acid uptake systems identified in S . cereuisiae"

0

0,

Amino acids L- Arginine

L-Lysine

L-Methionine

L-Histidine

Transport systems Arginine permease General amino-acid permease Lysine permease Arginine permease General amino-acid permease Methionine permease 1 Methionine permease 2 General amino-acid permease Histidine permease 1 Histidine permease 2 General amino-acid permease

Mutations

Activity in ammoniagrown cells

arg-p = can1 gap1

Yes No

Grenson et a / . (1966) Grenson et a / . (1970)

25 200 3

Yes Yes No

Grenson (1966) Grenson (1966) Grenson er a / . (1970)

12 800

Yes Yes No

Gits and Grenson (1967) Gits and Grenson (1967) Grenson et a / . (1970)

17 4000

Yes Yes No

Crabeel and Grenson (1970) Crabeel and Grenson (1970) Grenson et a / . (1970)

Apparent K , value (WM) 10 8

References

L-Glutamic acid

L-Leucine

L-Threonine

L-Tryptophan

w a

L-Glutamine

L-Proline

Dicarboxylic amino-acid permease 1 Dicarboxylic amino-acid permease 2 General amino-acid permease Leucine permease General amino-acid permease Threonine permease General amino-acid permease Arom permease General amino-acid permease Glutamine permease 1 Glutamine permease 2 General amino-acid permease Proline-iminoacid permease General amino-acid permease

17

1000

-

Yes

Darte and Grenson (1975)

-

No

Darte and Grenson (1975)

gap1

No

Darte and Grenson (1975)

gap1

Yes No

gap1

Yes No

1000

gap1

No

Gits and Grenson (1969) M. Grenson, unpublished results Gits and Grenson (1969) M. Grenson, unpublished results C. Hennaut and M. Grenson, unpublished observations Grenson et a / . (1970)

40

gap1

Yes Yes No

Grenson and Dubois (1982) Grenson and Dubois (1982) Grenson and Dubois (1982)

25

put4 gap1

No No

Magana-Schwencke and Schwencke (1969) and others (see Section 1I.J)

Yes 10

This list of amino-acid permeases is not exhaustive; several specific uptake systems have been detected and remain to be studied.

40

JEAN-MARIE WIAME H A L .

competitors. On the contrary, the fact that the arginine permease is both specific and constitutive makes it able to transport L-arginine even in ammonia-grown cells and/or in the presence of other amino acids. Several degrees of specialization are obtained through regulatory processes that efficiently integrate amino-acid permeases into given metabolic pathways. A characteristic of this integration is that it proceeds without rigid exclusion, a condition for further adapatation to new physiological situations. Transport in yeast has been reviewed by Eddy (1982) and Cooper (1982a,b). I. GENERAL AMINO-ACID PERMEASE

1. Specificity

General amino-acid permease (GAP) has more than 20 natural substrates, about half of which are potential nitrogen sources while the others cannot give rise to ammonium ions, glutamate, or glutamine. The substrates transported by the GAP include all of the L-amino acids constitutive of proteins, i.e., the basic and neutral amino acids (Grenson et al., 1970), the dicarboxylic amino acids glutamate and aspartate (Darte and Grenson, 1975), glutamine and asparagine (Grenson and Dubois, 1982), and proline (C. Casteleyn and M. Grenson, unpublished observations; Lasko and Brandriss, 1981). The D isomers of a number of amino acids are also substrates of GAP; since they are toxic to yeast, they are used as selective agents for isolating mutant strains deprived of GAP activity (Rytka, 1975). Non-proteic L-amino acids are also transported by the GAP, as are several metabolic intermediates such as citrulline (Jairis and Grenson, 1969; Grenson et al., 1970), ornithine (M. Grenson, unpublished observations), and a-aminoadipic acid (Darte and Grenson, 1975). A number of toxic analogues of amino acids are substrates of GAP, including canavanine, thiosine, ethionine (Grenson et al., 1970), N-6-chloro-acetyl-L-ornithine (Larimore et al., 1980), and y-monohydroxamic acid (Grenson and Dubois, 1982). All of these transport activities are lost as a result of mutations in the gapl gene, the structural gene for GAP. It should be clear that the aap mutation, which was isolated by Surdin et al. (1965) and shown to affect uptake of many amino acids, does not affect the gapl gene. The aap mutation is allelic to the apfl mutation (Grenson and Hennaut, 1971), which affects a genetic locus unlinked to the gap1 gene. Mutations in the apjl-aap locus produce a strong decrease in activity of all of the aminoacid permeases tested. The activity of GAP is decreased about sevenfold,

NITROGEN CATABOLITE REPRESSION IN FUNGI

41

whereas that of the proline permease is undetectable, and glutamate uptake decreases about 50-fold. These pleiotropic effects suggest that the apfl gene product (APF1) is a common factor necessary for optimal activity of all of the amino-acid permeases, possibly an element in energycoupling mechanisms. 2. Regulation by Feedback Inhibition Like most of the amino-acid permeases studied in Saccharomyces cerevisiae, GAP is inhibited after preloading the cells with its transported substrates. Although the very specific histidine permease, for example, is transinhibited only by histidine (Crabeel and Grenson, 1970), the methionine permease by methionine, and the threonine permease by threonine and serine (Gits and Grenson, 1969), GAP is transinhibited by all amino acids. More precisely, uptake of any substrate by the GAP is inhibited by preloading cells with any amino acid that is transported by this permease (Crabeel, 1973). The simplest explanation of this behaviour is that GAP, like other permeases, is feedback-inhibited by amino acids accumulated during the preloading period. The observed inhibitions would simply reflect the low specificity of the GAP both for uptake and for feedback inhibition. It seems as difficult to prove as to disprove this view. Woodward and Kornberg (1981) think that they have ruled it out and have shown that inhibition is associated with a change in a membrane-protein component of the GAP system, and that recovery from the inhibition involves resynthesis of this protein. This interesting proposition should be supported by further rigorous experimentation. A point to be considered is that the behaviour reported might result from two (or even three) of the distinct processes which have been demonstrated in regulation of GAP (see below). On the other hand, insensitivity of GAP to feedback inhibition would be unexpected with regard to the behaviour of the other amino-acid permeases. If the feedback hypothesis is retained for a while, it becomes clear that, to appreciate the effect of inhibitors of protein synthesis or those of preloading cells with amino acids, a careful analysis (both qualitative and quantitative) of all of the potential inhibitors of the GAP activity is needed, including ammonium ions. 3. The Ammonia Effect

No GAP activity is detectable in cells of the wild-type strain Z1278b of S. cereuisiae grown in the presence of ammonium ions, whereas GAP activ-

42

JEAN-MARIE WlAME ET AL.

ity is high after growth on proline as sole nitrogen source (Grenson et al., 1970). On addition of ammonium ions to proline-grown cells, GAP is progressively and completely inactivated. This inhibition is reversible for some hours, that is, until repression of GAP synthesis has developed enough to be detected (Grenson, 1983a). That these processes are linked to two distinct regulatory mechanisms is shown by the fact that they can be lost separately as a result of mutations. The fact that GAP is active in ammonia-grown cells only when both controls are removed makes it difficult to select mutations that affect only one of them. This explains why the first mutations to be isolated affected a process that is needed for both controls and also why these mutations could not provide information about the specific mechanisms involved. The simplest mechanism for ammonia resistance is an impairment of ammonia uptake. This is observed in the mep-1 mutants in which the GAP is active in the presence of ammonium ions (Dubois and Grenson, 1979). The gdhA- mutations, located in the structural gene for the anabolic NADP+-GDHase, decrease or suppress the sensitivity of the GAP and other uptake systems to the “ammonia effect” (Grenson and Hou, 1972). With regard to glutamate formation, the catalytic deficiency in these rnutants is completely compensated when proline is included in the culture medium. It might be asked whether 2-oxoglutarate or NADPH can act as activators of the GAP system or, conversely, whether NADP+ is an inhibitor. This would be efficient from a physiological point of view, since these compounds would be good signals of the availability of a nitrogen source. However, no significant changes in the concentration of these possible effectors could be detected in gdhA- mutants (Grenson and Hou, 1972; Dubois et al., 1974). This has been the origin of the hypothesis that the enzyme might act as a regulatory protein. As already described, mutations at the gdhCR locus of S . cereuisiae suppress nitrogen catabolite repression of all of the tested catabolic enzymes for nitrogenous compounds. They also de-repress several ammonia-sensitive permeases (see Sections I-L,N). The gdhCR mutations might affect a gene coding for a general “nitrogen repressor” (Dubois and Grenson, 1974). Although these mutations also de-repress GAP, this has no detectable effect since the second control mechanism on GAP is not released (Grenson, 1983a). Roon et al. (1974) used a double (gdhA,gdhCR) mutant with the hope of demonstrating that the catabolic NAD+-GDHase,which is de-repressed as a result of the gdhCR mutation, is able to replace the defective NADP+-linked enzyme and to restore the ammonia effect on GAP. That the GAP is not active in such double mutants grown in the presence of ammonium ions is clear. However, this seems to be due not to de-repression of the catabolic NAD+-GDHase by

NITROGEN CATABOLITE REPRESSION IN FUNGI

43

itself, but rather to a secondary effect of the gdhCR mutation, namely accumulation of glutamine. It is known that one of the effects of the gdhCR mutation is to suppress a process of inactivation of glutamine synthetase which is triggered by glutamine (Legrain et al., 1982). As a consequence, glutamine accumulates in these strains and this seems to be the origin of the GAP inhibition (Grenson, 1983a). This view is supported by the fact that when a gin'" mutation (which limits glutamine formation, due to thermosensitivity of glutamine synthetase) is introduced in a double (gdhA-,gdhCR) mutant strain, it restores GAP activity in the presence of ammonium ions. The idea that de-repression of the NAD+-GDHase by itself is not responsible for the absence of activity of the GAP in ammoniagrown double (gdhA,gdhCR)mutants is further supported by the fact that the g W mutation also de-represses the NAD+-GDHase in ammoniagrown cells. The mechanism by which glutamine inhibits GAP activity is discussed later in this review. It does not exclude a direct action of NADP+-GDHase at least on one of the regulatory mechanisms involved in the ammonia effect on the GAP, namely the positive control mechanism mediated by the product of the npr gene as described later. The inactivation process is absent from mutants altered at the mut2, mut4, or p g r loci (Grenson, 1983a). On addition of ammonium ions to a mut2, a mut4, or a p g r mutant strain growing in proline-containing minimal medium, no inactivation of the GAP was observed (Fig. 4). Preexisting permease activity is maintained, whereas a further increase is prevented. On the contrary, normal inactivation of the GAP does occur in a gdhCR mutant strain, or in a g W mutant strain (Fig. 4). Repression of synthesis of a component of GAP is manifest in mutants that have lost the inactivation mechanisms, namely the mut2, mut4, and p g r strains (Fig. 4). The activity of GAP is low or negligible in ammonia-grown single mut2 or mut4 orpgr mutant strains, but it is high in double mutants that contain a gdhCR or a glnts mutation (Table 8 ) . Hence the gdhCR and the ginrs mutations provoke de-repression of the GAP. The absence of cumulative effect of other pairs of mutations has also been shown (Table 8). All of these mutations, except for one, also affect other ammoniasensitive permeases. The exceptional mutation is p g r , the effect of which is restricted to the GAP. It is strongly linked to the GAP genetic locus and located near one end of the gene (Grenson and Acheroy, 1982). The mutation is semi-dominant and its phenotype is the same as that of the mut2 and mut4 mutations, which are recessive. Thus it was concluded that the pgr region of the gap gene determines a receptor site for a negative control mediated by products of the mut2 and mut4 genes (MUT2 and MUT4). The gdhA mutation affects both controls of the GAP. Repression disap-

FIG. 4. Absence of inactivation of the general amino-acid permease upon addition of ammonium ions to proline-grown mu22 (a), mu24 (b), orpgr(c) mutants of S. cereuisiae. The inactivation process is present in gdhCR and gln" mutants as well as in the wild-type strain 81278b (d). The general amino-acid permease activity was estimated by measuring initial At the time shown by the arrow, ammonium ions [as 0.01 M (NH4)2S04final uptake rates of 0.02 mM ~-['~C]citrulline. concentration] were added to a portion of the culture. Abcissa indicates absorbance at 660 nm, as a measure of cell mass ml-' in the culture. (a) Behaviour of strain 18409c, mut2; (b) behaviour of strain 19747c, mut4; (c) behaviour of strain and 17251a, gln" (0). Behaviour of 19679a, p g r ; and (d) behaviour of strains 81278b (wild type) 12597a, gdhCR (0) proline-grown cells before ( 0 ,I),and after (0, 0) addition of ammonium ions. From Grenson (1983a).

45

NITROGEN CATABOLITE REPRESSION IN FUNGI

TABLE 8. Effect of mutations in S . cerevisiae on the activity of ammonia-sensitive permeases in cells grown in ammonia-containing minimal mediumY Initial rate of uptake (nrnol min-' mg protein-'

Strains

Genotypes

21278b 17251a 19693d 19750b 19681a 12597a 13830a 2055 1b 20570a 17314a 20607b 19213b 184O9c 19747c 19697a Proline-grown 21278b wild type (control)

Wild type gln" g W , mu12 gln'", mu14 gln", pgr gdhCR gdhCR,mur2 gdhCR,mui4 gdhCR, pgr gdhCR, glnlS mu12, mu14 mut2, pgr mu12 mu14 Pgr

a

L-citrulline (0.02 mM)

L-Proline (0.02 mM)

DL-Ureidosuccinate (0.04 mM)

0.0 0.8 19 19 15 0.2 33 9 8 2

0.0 3 15 9 3 3 16 6 3

0.0 3 8 8 3 1 4 4 1

5

-

5 5 0.8 1 18

1 1 0 10

-

0 0 0 12

From Grenson (1983a).

peared and inactivation was alleviated (Fig. 5a). In a double (gdhA,gdhCR) mutant, ammonia inactivation was completely restored (Fig. 5b). The mut2, mut4, and p g r mutations completely suppressed the ammonia effect on GAP in (gdhA,gdhCR) mutants (Grenson, 1983a). 4. Positive Control of GAP Activity

Formation of active GAP requires the integrity of two unlinked genetic loci, namely the gap1 and npr2 genes. Mutations at the gap2 locus inactivate GAP specifically; they seem to affect the structural gene for this permease (Grenson et al., 1970; Crabeel, 1973; Grenson and Acheroy, 1982). Mutations at the nprl locus have a pleiotropic effect on several ammonia-sensitive permeases (Dubois and Grenson, 1979; Grenson and

46

JEAN-MARIE WIAME E T A L .

0

0.1

0.2

0.3

0.1

0.5

0.6

Absorbance

FIG. 5. Limited inactivation of the general amino-acid permease following addition of ammonium ions to proline-grown cells of the gdhA mutant 4324c of S. cerevisiae, and full inactivation in the double (&A, gdhCR) mutant strain 12759a. The experimental procedure was as described in the legend to Fig. 4. Rate of citrulline uptake by proline-grown cells before ( 0 )and after (0) ammonium ions were added. This figure is directly comparable to Fig. 4. (a) The behaviour of strain 4324c (gdhA) and (b) the behaviour of strain 12759a (gdhA,gdhCR). From Grenson (1983a).

Dubois, 1982; Grenson and Acheroy, 1982); they completely inactivate GAP in proline-grown as well as in ammonia-grown cells. The product of the nprl gene is a protein, NPRl (Grenson and Acheroy, 1982). The behaviour of thermosensitive nprl mutants ( a p P ) shows that the continuous presence of a functional NPRl is required for maintenance of GAP activity. Indeed, GAP activity that developed at the permissive temperature of 29°C in a npr‘” mutant strain was destroyed at 36”C,whereas that of the wild-type strain was maintained for hours, even in the presence of cycloheximide (Grenson, 1983b). Although these observations do not rule out the possibility that NPRl might also act as a positive factor for synthesis of GAP, another role for this protein is strongly suggested. This role might be either to protect the permease against inactivation or to catalyse a re-activation process. This idea is further supported by the fact that

NITROGEN CATABOLITE REPRESSION IN FUNGI

47

NPRl is no longer required when the function mediated by MUT2 or MUT4 is impaired. Indeed, the mut2, mut4, and p g r mutations, which prevent inactivation of GAP on addition of ammonium ions, restored GAP activity in nprl mutants. This suggests that NPRl is necessary to counteract the action of an inactivating regulatory mechanism which is generated by the mut2 and mut4 genes, and which operates at the level of the PGR regulatory site of the GAP molecule. The nprl mutation has an additional effect which is expressed in the presence of ammonia. When associated to a mut2, a mut4, or a p g r mutation, it allows development of active GAP in the presence of ammonium ions, as do the gdhCR or the glnts mutations which de-repress synthesis of this permease in ammonia-grown cells. This effect is linked to the decreased rate of ammonia uptake in nprl mutant strains (Grenson and Dubois, 1982). The available data on regulation of GAP may be interpreted as in Fig. 6 . In proline-grown wild-type yeast cells, GAP is synthesized, but it is inactivated by the MUT2-MUT4 control mechanism acting at the level of the PGR site of the permease. This negative control process keeps GAP in an inactive form as long as the NPRl product does not intervene. In steady-state wild-type cells growing on proline as the sole nitrogen source, an equilibrium is reached between the negative effect of the MUT2-MUT4 system and the positive effect of the NPRl product. On addition of ammonium ions, the pre-existing permease is progressively inactivated by a system that involves the MUT2-MUT4 system, since gopl gene PGR

t

,active

GAPl protein , + amino acids

‘inactive

GAPl potein

FIG. 6. Tentative scheme for regulation of the general amino-acid permease in S. cerevisiue. In the scheme, the gdhCR gene product, GDHCR, with glutamine as an effector, is depicted as repressing transcription of the gap1 gene. Since the direction of the gap1 gene transcription is not known, the location of the arrow representing repression of this transcription by GDHCR is arbitrary. In addition, it is possible that GDHCR might act at any post-transcriptionallevel before formation of active GAPl protein (i.e., translation, or processing of either RNA or precursor protein). From Grenson (1983b).

48

JEAN-MARIE WIAME ET AL.

this inactivation is suppressed in mut2, mut4, and pgr mutants. The ammonia-induced inactivation of the permease seems to be linked to the incapacity of the NPRl re-activating mechanism to keep pace with the MUT2-MUT4 inactivating process, either because ammonium ions or a derivative thereof inactivate, and possibly repress synthesis of the positive-control (NPR1) protein, or because it activates the MUT2-MUT4 system. The first of these possibilities seems more likely as indicated by the effect of the gdhA mutations. Although the GAP is active in ammoniagrown gdhA single mutants, it is not in a double (gdhA,nprl)mutant strain whether grown on ammonium ions, or on proline as the sole nitrogen source. Hence, a gdhA mutation cannot replace a mut2, a mut4, or a pgr mutation in restoring GAP in nprl mutants. Moreover, a gdhA mutation acts by favouring the positive effect of the NPRl protein, i.e., by preventing either its repression or its inactivation by ammonium ions (or a metabolite derived from them). In addition to this putative regulation by inactivation-re-activation of GAP, a second regulatory mechanism seems to repress synthesis of the permease in ammonia-grown cells of wild-type yeast. This repression seems to be mediated by GDHCR which might inhibit transcription of the gap1 gene or processing or translation of the GAP1 RNA, or even a posttranslational modification necessary for giving rise to active GAP proteins. On the basis of the loss of this “repression” in glnts mutants, the thermosensitive glutamine synthetase of which provokes a moderate glutamine starvation at 29”C, glutamine might act as an effector of the repression mechanism. J . TRANSPORT OF PROLINE

A specific proline transport system was detected first by MaganaSchwencke and Schwencke (1969) in Saccharomyces chevalieri, a yeast that is now considered a strain of Saccharomyces cerevisiae. Its specificity seemed to correspond to that of an imino acid uptake system (MagaiiaSchwencke et al., 1973). This proline permease was de-repressed by nitrogen starvation (Schwencke and Magafia-Schwencke, 1969; Kuznar et af., 1973). A proline permease with similar properties was detected subsequently in the wild-type strain of S . cereuisiae C1278b. It is not active in ammoniagrown cells, and it is distinct from GAP (Grenson et af., 1970). Although GAP also transports L-proline with a low affinity (C. Casteleyn, unpublished observations), this property remained undetected in the initial work due to the limited range of proline concentrations used. Indeed, at low

NITROGEN CATABOLITE REPRESSION IN FUNGI

49

concentrations of proline (up to 0.1 mM), the contribution of the specific high-affinity proline permease to proline uptake is largely predominant. Lasko and Brandriss (1981) presented evidence for GAP transporting proline with a low affinity. In addition, they isolated mutant strains which are defective in the high-affinity proline transport system. These mutations affect a single genetic locus named put4. The ammonia-sensitive high-affinity proline permease was not induced by proline, in contrast to the two enzymes of proline catabolism (Brandriss and Magasanik, 1979). A detailed study of the high-affinity L-proline uptake system of a strain of S. cereuisiae has been reported by Horak and RihovB (1982). These authors confirmed the results of Magaiia-Schwencke et al. (1973) showing that this permease is highly specific; it is competitively inhibited only by closely related structural analogues, by sarcosine, and by L-alanine. Transport of L-proline was shown to be markedly transinhibited by a number of amino acids. With regard to the ammonia effect, the high-affinity L-proline permease of S. cereuisiae 21278b seems to be regulated by the same double control mechanism as GAP, i.e., both by an inactivation-re-activation mechanism and by repression of permease synthesis (Grenson, 1983 a,b). This conclusion is based on the following. On addition of ammonium ions to de-repressed cells, the proline permease was rapidly inactivated, and further growth in the presence of ammonia led to repression of its synthesis. Furthermore, the same genetic determinants were involved in this regulation. Thus, in double mutant strains containing a mutation (such as gdhCR or glnrs)which de-represses GAP, together with a mutation (such as mut2 or mut4) which supresses inactivation of the GAP, the proline permease was also active in the presence of ammonia. The sole exception was the gap-linked pgr mutation which suppressed ammonia inhibition of GAP only and appeared to affect a regulatory region of the gap gene. Strains bearing only one of these mutations had a low proline uptake activity under the same conditions. Finally, NPRZ, a pleiotropic factor that seems necessary for re-activation of several ammonia-sensitive permeases, was also shown to be needed for optimal activity of the proline permease (Dubois and Grenson, 1979; Grenson, 1983b). In a recent study of regulation by ammonia of the specific proline permease and the GAP of S. cereuisiae, Courchesne and Magasanik (1983) described a genetic locus (per]) which is involved in this regulation. Mutations at this locus allow development of active GAP and specific proline permease in ammonia-grown cells. They suppress inactivation of both permeases on addition of ammonia to cells with high permease activity. These per1 mutations have a phenotype that resembles that of the mut2 mutations of Grenson (1983a,b). Extending the pleiotropy of the per2

50

JEAN-MARIE WIAME H A L .

mutation to the specific histidine permease does not seem justified by the data presented by Courchesne and Magasanik since criteria are absent and histidine is also a substrate of the GAP which is indeed activated. In the same paper, Courchesne and Magasanik (1983) presented data to support the idea that the effect of gdhA mutation on regulation of permeases by ammonia is due to the existence of an undetected per1 mutation in the strain isolated by Grenson et af. (1974). Additional tests are needed to confirm this suggestion, since the genetic analysis lacks rigour. Furthermore, the conclusions of Courchesne and Magasanik (1983) are based mainly on inactivation tests, while the gdhA mutation clearly affects both repression and inactivation of GAP (see Fig. 5a). The different behaviour observed in the two gdhA mutants with regard to inactivation of GAP might be due to other elements in the genetic background. Indeed, these strains are not isogenic, as shown by the presence of mutations (like his4-41) which have not been obtained in strain 21278b. On the other hand, the effects of glutamate and glutamine should be considered with caution, since these amino acids are both substrates and strong transinhibitors of GAP.

K . UREIDOSUCCINATE-ALLANTOATE PERMEASE

A specific transport system for ureidosuccinic acid (USA) has been discovered and studied by Lacroute and colleagues (Lacroute 1966, 1968; Drillien and Lacroute, 1972; Chevallier et al., 1975; Greth et al., 1977). Mutants affected in the urepl (or u e p l ) locus have lost ureidosuccinate transporting activity. It was clear from the beginning that this permease is regulated by nitrogen metabolism, since activity was not detected in ammonia-grown cells, whereas the transport system was very active in proline-grown cells. Activity of USA permease in proline-grown cells was rapidly lost on addition of ammonium ions, and repression of the permease synthesis was indicated. A possibility of transinhibition by internal amino acids or derivatives was also suggested. Two mutations which produce de-repression of USA permease in ammonia-grown cells have been isolated by Drillien and Lacroute (1972). One of them, named urel, has been identified as affecting the gdhA gene and the second, named ure2, has been shown to be allelic to a gdhCR mutation (Grenson et al., 1974). All of these characteristics underline the similarity of behaviour of USA permease with that of GAP and proline permease. This similarity has been confirmed in two different grounds. On the one hand, USA permease has been identified as the allantoate transport system and the

NITROGEN CATABOLITE REPRESSION IN FUNGI

51

urepl (or uepl) mutations with the da1.5 mutations (Turoscy and Cooper, in preparation, cited by Cooper, 1982b). Hence, its regulation by ammonia somehow integrates this uptake system with the allantoate degradative pathway. On the other hand, the series of mutations used to demonstrate the existence of two distinct regulatory mechanisms for GAP and proline permease also affect USA-allantoate permease (Grenson, 1983a,b). The association of mutations that suppress ammonia repression of GAP (namely gdhCR and glnts) with mutations that suppress ammonia inactivation of GAP (namely mut2 and mut4) also allows the development of active USA permease in ammonia-grown cells, showing that the same genetic determinants are involved (Grenson, 1983a). In addition, NPRl also acts as a positive factor for development of active USA permease (Dubois and Grenson, 1979; Grenson, 1983b).

L. AMMONIA UPTAKE SYSTEMS

Uptake of methylamine and ammonia by Saccharomyces cerevisiae involving a specific transport system has been demonstrated by Roon et al. (1975a). The transport system exhibited maximal activity in ammoniagrown cells and was repressed 60-70% when glutamine or asparagine was added to the growth medium. There was no significant de-repression of the transport system during nitrogen starvation. L-Amino acids are noncompetitive inhibitors and this inhibition is relieved in mutant strains that have a decreased ability to transport amino acids (Roon er al., 1977). More recently, transport of methylamine and ammonia has been shown to be mediated by at least two and probably three uptake systems in the wild-type strain C1278b of S . cerevisiae (Dubois and Grenson, 1979). The two methylamine uptake systems which give rise to a biphasic Lineweaver-Burk plot of uptake rates can be lost separately as a result of the two genetically unlinked mepl and mep2 mutations. Both systems were strongly inhibited and repressed in the presence of ammonia or glutamine. Ammonia repression was lost in a gdhA, as well as in a gdhCR mutant strain. No repression occurred with glutamine in the gdhCR mutant. In the case of the gdhA mutant, repression seemed to be restored by glutamine. Experiments using a glnts mutation indicated that glutamine is a necessary effector for repression of only one of the methylamine/ammonia uptake systems, namely the MEP2 high-affinity low-capacity permease. Mutations at the nprl locus did not affect the MEP2 system, but they strongly depressed the activity of the MEPl low-affinity high-capacity

52

JEAN-MARIE WIAME ET AL.

permease, as well as the residual ammonia uptake system present in double (mepl,mep2) mutant strains. The mut4 mutation which restored activity of GAP, proline permease, and USA-allantoate permease in nprl mutants did not restore normal methylamine uptake in npr-l mutants. Moreover, neither the mut2 nor the mut4 mutation restored methylamine sensitivity in nprl mutant strains. On the contrary, a mutation named amul has been shown to specifically increase ammonia uptake in a nprl mutant strain, probably by restoring the third “MEP3” ammonia transport system (Dubois and Grenson, 1979). Hence, uptake of ammonium ions seems to be regulated very efficiently in the wild-type strain 231278b. The high-affinity MEP2 system was repressed by a mechanism that is active only in the presence of glutamine. It did not seem to depend on the positive-control factor NPRl. The highcapacity MEPl system seemed to be repressed by ammonia itself, it depended on the NPRl product, as well as the putative MEP3 seems to do. The absence of an effect of the mut2 and mut4 mutations on uptake of methylamine or ammonia in nprl mutant strains is striking and needs further investigation.

M. UPTAKE SYSTEMS FOR L-GLUTAMINE A N D L-ASPAUAGINE

L-Glutamine and L-asparagine are transported by GAP (Grenson et al., 1970; Grenson and Dubois, 1982). L-Asparagine has been shown to be accumulated in Saccharomyces cerevisiae X-2180 by a transport system that is distinct from the GAP (Gregory et al., 1982). At its optimal pH value of 4.5, this L-asparagine uptake system was inhibited competitively by L-glutamine, as well as by L-threonine and L-tryptophan. No inhibition was detected with L-glutamic acid or L-aspartic acid. Alanine was not tested. On the other hand, several observations strongly suggest that at least two distinct uptake systems are able to transport L-glutamine (Grenson and Dubois, 1982). One of them is a specific permease (GNP1) that recognizes L-glutamine, L-asparagine, and the toxic analogue of L-glutamine, L-glutamic acid y-monohydroxamate. The second system (GNP2) seems less specific since it is inhibited by L-glutamate, L-aspartate, and Lalanine. Although the specificity of these two L-glutamine transport systems has not been tested extensively, it seems possible that the first of these systems might be the L-asparagine permease reported by Gregory et al. (1982). The GNPl L-glutamine uptake system was inactivated in a nprl mutant strain, and this effect was suppressed by a mut4 mutation (Grenson and Dubois, 1982). Hence the GNPl L-glutamine permease seems to

’

NITROGEN CATABOLITE REPRESSION IN FUNGI

53

be regulated by the same mechanism as the ammonia-sensitive permeases already described.

N. GLUTAMIC ACID PERMEASES

L-Glutamic acid is taken up and accumulated by three distinct systems in Saccharomyces cerevisiae C1278b grown on L-proline as the sole nitrogen source (Darte and Grenson, 1975). One of these uptake systems is GAP, as shown by its absence from gap1 mutants, as well as by its specificity and its regulatory behaviour. The other two systems transport dicarboxylic amino acids with much higher specificity. One of them (GUP1) is constitutive in cells grown in the presence of ammonium ions; its activity is inhibited in cells grown in the presence of L-glutamic acid and ammonium ions simultaneously. The remaining system (GUP2) is ammonia sensitive. However, it reacts very differently from GAP on addition of ammonium ions to proline-grown cells. Ammonia did not produce an inactivation of the pre-existing GUP2 system, but only a clear repression of its synthesis (Darte and Grenson, 1975). In relation to this behaviour, it is interesting to note that the lack of the positive factor NPRl in nprl mutant strains had no effect on the activity of the GUP2 system in prolinegrown cells (M. Grenson, unpublished observation). This is in keeping with the hypothesis that the role of NPRl is to re-activate permeases that have been inactivated on addition of ammonia to proline-grown cells. On the contrary, mutations that affect repression of the synthesis of enzymes and permeases in ammonia-grown cells also de-repressed the GUP2 system. Among these are the gdhA mutation (Darte and Grenson, 1975) and the gdhCR mutation (Darte, 1976). Addition of ammonium ions to cultures of proline-grown gdhA mutants had no effect on GUP2 activity. Moreover, when the catalytic deficiency of the anabolic NADP+-GDHase of these mutants was compensated by adding glutamate to the ammoniacontaining culture medium, no repression of GUP2 was observed (Darte, 1976). This is one of the clearest cases in which NADP+-GDHase seems to participate in ammonia repression through a function that is distinct from its producing glutamate. The gdhCR mutation de-repressed the GUP2 system in ammonia-grown cells as it does for all the ammoniasensitive catabolic enzymes and permeases studied so far. The nprl mutation had the same effect on the GUP2 permease as on catabolic enzymes involved in metabolism of nitrogenous compounds (such as arginase, allantoinase, and urea amidolyase). Indeed, the de-repressing effect of the nprl mutation on these enzymes as well as on the GUP2 permease in

54

JEAN-MARIE WIAME ETAL.

ammonia-grown cells is a consequence of the decreased rate of ammonia uptake in npvl mutant strains (Grenson and Dubois, 1982). This interpretation is further supported by the fact that ammonia-induced repression of these systems is restored in nprZ mutant strains by the amul mutation which specifically increased ammonia uptake.

0. OTHER AMMONIA-SENSITIVE PERMEASES

Other ammonia-sensitive amino acid permeases might exist. Their presence should be tested in GAP- mutant strains. Uptake systems for other nitrogen-containing compounds also are strongly depressed in ammoniagrown cells. The active high-affinity DUR3 urea transport system of Sacchuromyces cereuisiue has been reported to be subject to nitrogen repression (Cooper and Sumrada, 1975), as has the allantoin transport system (quoted by Cooper, 1982b). The behaviour of these uptake systems in the presence of mutations that affect the regulation of other permeases by ammonia has not been studied.

P. COMMENTS ON REGULATION OF UPTAKE OF NITROGENOUS NUTRIENTS

In those cases in which the mechanism of the “ammonia effect” on uptake systems has been analysed, it has been found that several controls are involved. Repression of synthesis of the transport systems in ammonia-grown cells seems to be present in every case, as well as de-repression in the presence of a gdhCR mutation. The product of the gdhCR genetic locus thus seems to control all of the ammonia-repressible permeases as well as all of the nitrogen catabolic enzymes studied so far. It might be a general repressor which regulates uptake and catabolism of compounds used as nitrogenous nutrients, as suggested earlier in this review. Several ammonia-sensitive permeases are subject to a second regulatory mechanism which affects their activity. These permeases, the prototype of which is GAP, are inactivated in the presence of ammonium ions by a mechanism that seems to be mediated by the products, MUT2 and MUT4. Inactivation might result either from chemical modification of the permease or from binding of a regulatory protein in the presence of suitable effectors. These effectors might be either amino acids or other nitrogenous derivatives or signal molecules. In this context, it is remarkable that transinhibition by preloaded amino acids has been observed for several of these permeases. While this is expected for GAP for which these

NITROGEN CATABOLITE REPRESSION IN FUNGI

55

amino acids are natural substrates, it indicates a refined regulatory mechanism in the case of the proline permease and the ureidosuccinate-allantoate uptake system. The sole positive control detected in regulation of these uptake systems by ammonia is that mediated by the NPRZ gene product. The observations made so far on this regulation strongly indicate that the NPRl protein is needed for re-activation of the permease molecules which have been inactivated by the MUT system. As a working hypothesis compatible with the presently known facts, this regulation can be viewed as follows. Re-activation of the permease by the NPRl protein can occur only in proline-grown cells because ammonia either inhibits NPRl activity, or represses NPRl synthesis. The gdhA mutations (which affect ammonia inactivation as well as ammonia repression of GAP synthesis) seem to suppress ammonia inactivation of GAP by de-repressing NPRl formation, since nprZ mutations cancel this gdhA effect. Hence, strictly speaking, the gdhA mutations would not suppress the inactivation mechanism for the permease, but rather allow development of the re-activation mechanism in spite of the presence of ammonia. Finally, it might be interesting to observe that another possible mechanism for inactivation-re-activation of uptake systems (besides covalent modification and regulatory protein binding) could be a process analogous to modulation of cellular transport rates by hormones and other agents that operate in cells of animals (Lienhard, 1983). In these cells, rapid regulation of transport seems to be obtained by exocytotic insertion of transporter-containing vesicles into the plasma membrane and their endocytic removal. Considerable evidence indicates that this mechanism accounts for insulin stimulation of glucose transport in fat and muscle cells, for histamine stimulation of acid secrection in the stomach mucosa, and for vasopressin enhancement of water permeability across bladder and kidney epithelia. In such a process, the MUT system might catalyse excision of the sensitive permeases in response to the presence of effector amino acids, whereas the NPRl product would mediate their insertion in the plasma membrane whenever the absence of ammonia allows its formation and activity. Considering the available data on ammonia-sensitive permeases, it appears that at least three of them are regulated simultaneously by two distinct regulatory mechanisms, one acting by inactivation-re-activation, and the other by repression-de-repression.It is striking that these two regulatory mechanisms, which can be lost independently, are both found in at least three permeases. This might reflect a common origin both of these two regulatory mechanisms and of the genetic determinants of the regulatory regions of the respective permeases.

56

JEAN-MARIE WlAME ET AL.

Although details of their expression are not known, the main genetic loci involved in ammonia regulation of uptake systems seem to be identified, thus permitting study of this regulation at the molecular level.

Q . GENERAL VIEW

In the yeast Saccharonzyces cerevisiae, the activity of a number of enzymes and transport systems which make nitrogenous nutrients available for metabolism is very sensitive to the quality of the nitrogen source present in the culture medium. As a rule, the intensity of de-repression of these activities parallels the rate of cellular growth which is allowed by each nutrient. Thus, maximal growth rates are obtained with ammonium ions, glutamine, or asparagine, and growth on these compounds leads to the most intense depression of a whole range of activities linked to utilization of nitrogenous compounds. Two types of data have been obtained with regard to the mechanism of this regulation, namely identification of effectors and that of regulatory macromolecules and their receptors. Mutations in the structural gene for glutamine synthetase ( g l n l ) permitted a distinction to be made for systems for which glutamine is a necessary effector for repression or inactivation of enzymes or permeases. The product of the gdhCR gene, GDHCR, is a good candidate for the central role of general repressor exerting a negative control on synthesis of almost all of the proteins (enzymes, permeases, and most probably regulatory proteins) which are subject to nitrogen catabolite repression. Indeed, in the presence of a gdhCR mutation, synthesis of all of these proteins but one has been found to be de-repressed under nitrogen-repressing conditions. The exception is glutamine synthetase, synthesis of which is repressed by ammonia even in a gdhCR mutant. However, since this enzyme has a biosynthetic function, its exceptional behaviour seems to be concerned with its repression in ammonia-grown cells rather than the lack of gdhCR effect on this process. In several instances, de-repression of protein synthesis as a result of a gdhCR mutation can be detected only when other controls have been removed by suitable induction or mutations. This is the case for arginase which is controlled by arginine induction, and for transport systems which are subject to regulation by an inactivation-re-activation mechanism. The product of the gdhA gene, NADP+-linked glutamate dehydrogenase, has been considered as a possible regulatory molecule capable of mediating the ammonium-ion signal together with a 2-oxoglutarate signal.

NITROGEN CATABOLITE REPRESSION IN FUNGI

57

This idea is based on the fact that, for several enzymes and permeases, compensating the main consequence of the catalytic deficiency of GDHase A by making another source of glutamate available does not restore the ammonia effect in a gdhA mutant strain. Presently, this hypothesis has neither been eliminated nor proved. Available data are consistent with the view that the GDHCR molecule can be activated either by glutamine or by the GDHase A protein linked to the ammonium ion and 2-oxoglutarate simultaneously. These two types of binding might give rise to distinct repressors capable of reacting with different receptors. The “activated GDHCR repressor” might act either at the transcription level or at the level of RNA or protein-precursor processing, i.e., at any level before formation of active enzyme or permease.

111. Nitrogen Metabolite Repression in Filamentous Fungi A . BACKGROUND

A number of filamentous fungi are able to utilize a wide range of nitrogen sources. This metabolic versatility implies a strong selective pressure for a mechanism to ensure preferential utilization of the more favourable nitrogen sources. This mechanism, nitrogen metabolite repression (sometimes called “ammonium” repression or nitrogen catabolite repression) of the syntheses of many enzymes and permeases involved in nitrogen nutrition, has been extensively studied in two filamentous fungi, the ascomycetes Aspergillus nidulans and Neurospora crassa. Nevertheless, the existence of this regulatory mechanism has been documented in a number of other filamentous fungi, for example in basidiomycetes of the genus Ustilago (Lewis and Fincham, 1970a,b; Holloman and Dekker, 1971). Whether the details of nitrogen metabolite repression are common to all filamentous fungi cannot even be guessed at present, although where comparable data are available in A. nidulans and N . crassa they suggest considerable similarity. Aspergillus nidulans was probably the first organism in which a systematic genetic and biochemical study of nitrogen metabolite repression was made. The concept of a “neutral” nitrogen source for growth was a crucial advance in the study of regulation of nitrogen-metabolizing activities. A “neutral” nitrogen source neither induces nor nitrogen metabolite represses the permease(s) and/or enzyme(s) under study, enabling the

58

JEAN-MARIE WIAME ET AL.

processes of induction and nitrogen metabolite repression to be investigated separately. The earliest explicit uses of this methodology were probably those investigating the enzymes of nitrate assimilation (Cove, 1966; Pateman et al., 1967) and purine catabolism (Scazzocchio and Darlington, 1968) in A. nidulans. The first systematic attempt to study nitrogen metabolite repression was probably that of Arst and Cove (1969) who selected mutants of A. nidulans resistant to the toxic ammonium analogue methylammonium and showed that, when grown in the presence of repressing concentrations of ammonium, they were nitrogen metabolite de-repressed for the synthesis of a number of enzymes and permeases. (Throughout Section 111of this article, ammonium refers without distinction to ammoniun ions and ammonia. The experiments summarized in this section do not distinguish between the two forms but, at the pH values used, the ammonium ion would be the predominant form.) The level(s) at which nitrogen metabolite repression is exerted has not been established directly in any filamentous fungus. However, there is strong evidence for regulation of gene expression in filamentous fungi at transcription (e.g., Pate1 et al., 1981) and indirect evidence suggests that nitrogen metabolite repression prevents formation of functional mRNA (e.g., Bartnik et al., 1973a; Premakumar et al., 1978; reviewed by Marzluf, 1981). The following discussion therefore assumes that nitrogen metabolite repression acts at the transcription level, an assumption vigorously supported by both the nuclear location and the affinity for DNA (although RNA was not tested!) of the nit-2 gene product (Section III.B.l) of N. crassa (Grove and Marzluf, 1981). Before proceeding, two comments about the range of activities subject to nitrogen metabolite repression are in order. First, although it is clear that the syntheses of a very large number of enzymes and permeases involved in the provision of nitrogen are subject to nitrogen metabolite repression, it is frequently difficult to establish that synthesis of a particular enzyme is so subject. Difficulty arises where uptake of a co-inducer involves a nitrogen metabolite-repressible permease so that inducer exclusion can mimic direct nitrogen metabolite repression of enzyme synthesis. Second, it is possible for an activity not involved in nitrogen source utilization to be nitrogen metabolite-repressible. For example, the pyrimidine salvage pathway enzyme thymine 7-hydroxylase is nitrogen metabolite-repressible in A. nidulans although thymine is not a nitrogen source for this organism (Shaffer and Arst, 1984). A possible rationale is that conversion of the non-utilizable pyrimidine thymine into a utilizable pyrimidine (uracil) would economize nitrogen by making de nouo pyrimidine synthesis unnecessary (Griswold et al., 1976).

59

NITROGEN CATABOLITE REPRESSION IN FUNGI B . GENES INVOLVED IN NITROGEN METABOLITE REPRESSION

1. Nitrogen Metabolite Repression Is Mediated by a Positive-Acting Regulatory Gene All enzymes and permeases in Aspergillus nidulans whose synthesis is subject to nitrogen metabolite repression are under the control of a positive-acting regulatory gene designated areA (Arst and Cove, 1973; Hynes, 1975a). Loss of function alleles designated areA' lead to low or undetectable levels of enzymes and permeases under areA control (see Table 9 for examples) with consequent inability to utilize nitrogen sources other than ammonium (Arst and Cove, 1973; Hynes, 1975a; Polya et al., 1975; Arst et al., 1980, 1982; Tollervey and Arst, 1981; Spathas et al., 1982, 1983). They can be selected on the basis of defective utilization of nitrogen source or for conferring resistance to inhibitors whose toxicity can be reversed by nitrogen metabolite repression such as chlorate (e.g., Arst and Cove, 1973; Arst, 1982), 2-fluoroacetamide (e.g., Hynes, 1972), or m-P-aspartylhydroxamate (e.g., Kinghorn and Pateman, 1975a). Of course, the latter method poses a dilemma in the need to choose a nitrogen source able to support at least limited growth of areA' strains whilst not reversing inhibitor toxicity, a compromise frequently resulting in recovery of leaky areA' alleles. That the areA' phenotype results from loss of areA function (indicating positive regulation) was originally deduced from the observations that areA' alleles constitute the most common class of areA mutation and are recessive (Arst and Cove, 1973). It was further supported by the finding that areA'- 18 apparently results from a translocaTABLE 9. The effects of an areA' mutation on concentrations of certain nitrogenmetabolizing enzymes in A. nidulans" ~

_

_

_

_

_

_

_

Relative activity ___

~

~~

Relevant genotype

Nitrate reductase

Nitrite reductase

Urease

Formamidase

Wild type (areA+) areAr-18

100

100

100

100

100

100

0

0

20

10

2

5

~

L-GIutaminase

L-Aspa-

raginase ~

-

~

Data from Tollervey and Arst (1981) and Arst et al. (1982). Mycelia were grown under nitrogen metabolite derepressing conditions and, in the cases of nitrate and nitrite reductases, in the presence of a co-inducer.

JEAN-MARIE WIAME ET AL.

60

tion breakpoint within the areA gene (Rand and Arst, 1977; Arst, 1981; Tollervey, 1981). Still further evidence, along with an indication that the areA gene product is a protein, is provided by the selection of putative polypeptide chain termination areA' alleles (A1 Taho ef al., 1984). A far rarer class of mutation, designated areAd, leads to de-repressed expression of nitrogen metabolite-repressible activities (see Table 10 for an example). This de-repression is partially dominant in diploids (Hynes and Pateman, 1970; Cohen, 1972; Arst and Cove, 1973). Selection and genetic analysis of areAd mutations are facilitated by the existence of a number of plate tests for de-repression. For example, de-repressed synthesis of extracellular protease activity leads to production of a "halo" of milk-clearing on a repressing medium containing powdered skimmed milk (Cohen, 1972). Wild-type A. nidulans is unable to utilize certain compounds such as acrylamide, L-histidine and L-citrulline as nitrogen sources, but areAd mutations can result in their use through enhanced expression of acetamidase (Hynes and Pateman, 1970; Hynes 1975a), histidase (Polkinghorne and Hynes, 1975, 1982), and a citrulline uptake system (Arst, 1977), respectively. Pyrimidine auxotophs of A. nidulans do not respond to thymine, thymidine, or certain other pyrimidine derivatives, but an areAd allele can allow them to do so through de-repression of pyrimidine salvage pathway enzymes andfor a pyrimidine uptake system(s) (Shaffer and Arst, 1984). Still other growth tests involve the use of inhibitors of growth or conidiation whose toxicities can be reversed by exogenous ammonium or L-glutamine, including Cs+, methylammonium, chlorate, thiourea, 2-thioxanthine, 2-thiouric acid, and DL-p-aspartylhydroxamate (for examples of their use see Arst and Cove, 1969, 1973; Cohen, 1972; Pateman e f al., 1973; Kinghorn and Pateman, 1975a). Derepressed expression of an activity or activities necessary for the toxicity

TABLE 10. The effect of an areAd mutation on nitrate reductase levels in A . nidulans" Relative nitrate reductase activity Relevant genotype Wild type (areA+) xprD-1 (areAd)

Uninduced

Induced

Repressed (co-inducer present)

8

100 179

13 141

1

Data from Arst and Cove (1973).

NITROGEN CATABOLITE REPRESSION IN FUNGI

61

will enhance toxicity of the inhibitor, whereas de-repressed expression of an activity or activities involved in catabolizing the supplied nitrogen source will decrease inhibitor toxicity (Arst and Cove, 1973). As these effects can oppose each other, the phenotype of an areAd mutation in the presence of a particular combination of nitrogen source and inhibitor is not always predictable (and can be similar to wild type if the opposing effects cancel each other). Only a few pronounced areAd alleles have been reported. The areA-102 allele leads to marked de-repression of certain activities such as acetarnidase, histidase, and citrulline uptake (Hynes and Pateman, 1970; Hynes, 1975a; Polkinghorne and Hynes, 1975, 1982; Arst, 1977) whilst having a repressed phenotype for others, such as formanidase and a xanthine-uric acid permease, and having relatively little effect on the repressibility of still others such as nitrate reductase (Arst and Cove, 1973; Arst and Scazzocchio, 1975; Hynes, 1975a). This suggests that receptor sites for the product of the areA gene differ in structure, and that the mutant areA102 product can activate transcription more efficiently than the areA+ product at certain sites but less or equally efficiently at others. Alleles areA-30 and areA-31, obtained by reversion of an areA-I02 strain on xanthine and uric acid, respectively, as nitrogen sources, have a phenotype opposite at least in some respects to that of areA-102, i.e., lowered acetamidase levels and enhanced uptake of xanthine and uric acid relative to the wild type (Arst and Scazzocchio, 1975 and unpublished observations). Allele areA-200, selected by Hynes (1975a), has a similar phenotype, and areA-300 results in striking de-repression of thymine 7-hydroxylase and enhances pyrimidine uptake while decreasing expression of one or more activities involved in utilization of L-leucine and L-methionine as nitrogen sources (Shaffer and Arst, 1984). The other areAd allele was designated xprD-1 because it was selected as leading to de-repressed synthesis of exocellular protease (Cohen, 1972), and this designation was retained because uncertainty about the genetic basis for this mutation persisted until 1982. Unlike the other areAd alleles, xprD-1 leads to nitrogen metabolite de-repression of the vast majority of activities under areA control (Cohen, 1972; Arst and Cove, 1973; Pateman et a / . , 1973; Bartnik et a/., 1976). Allele xprD-1 was a genetic enigma because crosses of xprD-1 strains to wild-type strains yield a substantial proportion (up to 35%) of morphologically abnormal progeny with an areA’phenotype (Arst and Cove, 1973). Arst (1982) showed that xprD-1 is a near-terminal pericentric inversion (Fig. 7) and that these unusual progeny form a duplication-deficiency class with two copies of part of the left arm of linkage group 111 but lacking the terminal portion of the right arm including areA. Three interesting inferences arise from the genetic analy-

JEAN-MARIE WIAME H A L .

62

...

SCJadl 1.7’

-

ornC

J E A

34

FIG. 7. Partial map of linkage group I11 of A . nidulans showing positions of xprD-1 inversion breakpoints (arrows) and centromere ( 0 )(Arst, 1982). Distances are in centimorgans. A list of gene symbols for A . nidulans is given by Clutterbuck (1982).

sis of xprD-1. First, areA must be sufficiently close to the right arm telomere that no indispensable gene can lie centromere-distal to it (otherwise the duplication-deficiency class would be inviable). Second, nitrogen metabolite de-repression conferred by xprD-1 is probably a consequence of enhanced production of the areA product through fusion of the coding region to a more efficient promoter and/or ribosome-binding sequence (although the possibility of an accompanying structural alteration of the N-terminal end of the areA product must be considered). Third, if the second inference be correct, areA must be transcribed towards the right arm telomere. In the context of the xprD-1 inversion, it is interesting to note that the marginally de-repressed allele areAd-lO1 (Arst and Cove, 1973) is possibly associated with a non-reciprocal translocation (Arst, 1982). One other supposedly de-repressed allele, “areAd-520,” is mentioned in the literature (Spathas et al., 1983). However, this mutation was mistakenly classified and is correctly designated me&-520 (J. A. Pateman, personal communication). Although the areA gene product has never been isolated, there is nevertheless compelling evidence for its direct involvment in regulation of gene expression. If areA were only indirectly involved in regulating gene expression, the phenotypes of a range of mutant, but partially functional, areA‘ alleles should reveal a hierarchy of activities based on different degrees of sensitivity to nitrogen metabolite repression (Arst and Bailey, 1977). No hierarchy is observed. For example, areA‘-l is thermosensitive with respect to nitrate, nitrite, and hypoxanthine utilization but not with respect to formamide utilization, whereas the reverse is true of areA‘-2 (Arst and Cove, 1973). The existence of alleles such as areA-30, -102, and -300 having opposite effects on regulation of different activities delivers a coup de grace to any possibility of hierarchy, and indicates that areA must be directly involved. The corresponding gene in Neurosporu crassa, nit-2 (Reinert and Marzluf, 1975; Coddington, 1976; Dunn-Coleman et al., 1979), has been less well characterized genetically. No de-repressed nit-2 alleles have been reported. Nor has the phenotype of a complete loss of function allele (such as areA’-18 or the putative areA‘ chain termination mutations) been reported. This latter point is of particular interest because the character-

NITROGEN CATABOLITE REPRESSION IN FUNGI

63

ized nit-2 alleles allow considerable utilization of a number of nitrogen sources whose utilization is completely prevented by non-leaky areA' alleles such as L-arginine, L-aspartate, L-glutamate, and L-tyrosine (Facklam and Marzluf, 1978; Dunn-Coleman et al., 1979). Also, whereas no effect of a nit-2 mutation on L-glutamine utilization has been noted, nonleaky areA' alleles drastically limit, but do not prevent, glutamine utilization (Arst and Cove, 1973). It is unclear whether nitrogen metabolite repression is less stringent and less pervasive in N. crassa than in A . nidulans, or whether only leaky nit-2 mutants have been examined. Partial loss of function mutations in areA are easily obtained (Arst and Cove, 1973; Kinghorn and Pateman, 1975a; Rand and Arst, 1977). Grove and Marzluf (1981) isolated acidic proteins from nuclei of N. crassa, which were able to bind to N. crassa DNA-cellulose. A prominent protein eluted by L-glutamine, but not by L-asparagine, was identified after sodium dodecyl sulphate-polyacrylamide gel electrophoresis. Greatly decreased amounts of this protein were present in strains carrying either of two mutant nit-2 alleles, whereas a normal amount of (presumably) this protein with an altered electrophoretic mobility was present in a strain carrying a third nit-2 allele. A nit-2+ revertant selected from a mutant having only 3% as much of this protein as the wild type had regained the wild-type level. This putative nit-2 gene product bound to calf thymus DNA-cellulose but with a much lower affinity than for N. crassa DNA-cellulose. A molecular weight of 21,900 was estimated, although it is not known whether this represents the native molecular weight or whether the protein is a multimer, this being the monomer molecular weight. Grove and Marzluf (1981) estimate that the nit-2 gene product constitutes not more than 0.003% of total cell protein. Such a low concentration of nit-2 product might lead one to wonder if this protein is confined to the nucleus. If so, a nit-2+ allele in one nucleus in a heterokaryon would be unable to activate expression of a functional structural gene located in another nucleus of the heterokaryon. Chambers et al. (1983) showed rigorously that this is not the case; in heterokaryons made between nit-2+ and nit-2- strains carrying different electrophoretic variants of the nit-2controlled enzyme L-amino acid oxidase, the nit-2+ product acted in trans and both variants were synthesized (although a bias reflecting non-homogeneous distribution of nuclei in the heterokaryon was observed). This experiment confirmed much earlier demonstrations (e.g., by Sorger and Giles, 1965) that nit-2 mutations are able to complement mutations in structural genes under nit-2 control in heterokaryons. Similar complementation involving areA' alleles occurs readily in heterokaryons of A . nidulans (Arst and Cove, 1973).

64

JEAN-MARIE WlAME ET AL.

2 . Genes Having a Possible Peripheral Role in Nitrogen Metabolite Repression

a. The meaA and meaB genes. The methylammonium resistance mutations selected by Arst and Cove (1969) in A . nidulans map to two unlinked genes meaA and meaB (Arst and Page, 1973). Mutations in both genes resulted in nitrogen metabolite de-repressed expression of a wide range of activities (Arst and Cove, 1969) with extracellular protease a notable exception (Cohen, 1972). Mutations in meaA reduced net uptake of methylammonium and ammonium (Arst and Page, 1973; Pateman et al., 1973, 1974). In vivo evidence suggests that meaA mutations might lower ammonium pool sizes when nitrogen sources other than ammonium are supplied (Arst and Page, 1973). Mutations in meaB do not retard methylammonium transport (Arst and Page, 1973; Pateman et al., 1973; Cook and Anthony, 1978) although it seems likely that they lower efflux of nitrite (Rand and Arst, 1977; Tollervey and Arst, 1981). Mutations in meaB can result in resistance or hypersensitivity to toxic amino-acid analogues whose toxicity is unaffected by ammonium or ureA' mutations (H. N. Arst, Jr., unpublished observations). This suggests that nitrogen metabolite de-repression in meuB mutants might be a consequence of the intracellular concentrations or distribution of metabolites rather than some more direct effect. b. The amrA gene. The amrA-1 mutation of A . niduluns was selected as resulting in poor growth on ammonium as nitrogen source (Pateman et al., 1973). It also results in poor growth on certain nitrogen sources and nitrogen metabolite de-repression of certain activities. It leads to drastically lower rates of ammonium and methylammonium uptake under some, but not all, growth conditions (Pateman et al., 1974). In addition, amrA-1 lowers hexokinase levels about twofold and leads to poor growth on a number of carbon sources (Dunn-Coleman and Pateman, 1979). It seems futile to speculate on a possible role for amrA at this stage, but the pleiotropic effects of amrA-1 are too diverse to suppose that any involvement with nitrogen metabolite repression be other than very indirect.

c . The tamA gene. Mutations in the tamA gene of A . nidulans were first selected as conferring resistance to a number of growth inhibitors whose toxicities can be reversed by exogenous ammonium. The inhibitors included thiourea, chlorate, m-P-aspartylhydroxamate, and methylammonium (Kinghorn and Pateman, 1975a). Mutations in this class, now designated tamA-, occur frequently, are recessive, and have a highly

NITROGEN CATABOLITE REPRESSION IN FUNGI

65

pleiotropic phenotype enabling them to be selected in a variety of other ways (Rand and Arst, 1977), but they do not prevent utilization of any nitrogen source. Subsequently, Pateman and Kinghorn (1975, 1977) reported the existence of “tamAr-50” with a phenotype similar to that of areA‘ mutations and “tamAd-1” leading to nitrogen metabolite de-repression, but unfortunately these “mutations” were lost. Arst et al. (1982) attempted to reconstruct the Pateman-Kinghorn results. They showed that auxotrophies resulting from mutations in the pyroB gene respond to very high concentrations (e.g., 200 m M ) of ammonium as well as to Bg vitamers (i.e., pyridoxal 5’-phosphate and its precursors) at concentrations six or so orders of magnitude less, but that tamA- pyroB- strains will grow in the presence of much lower concentrations of ammonium (e.g., 3-10 mM), including that normally used as a nitrogen source. As pyroB- mutants respond to B6 vitamers at concentrations comparable to those required by pyroA- mutants (which do not respond to ammonium), it seems likely that the pyroB- lesion is in vitamin B6 biosynthesis, and Arst et al. (1982) speculated that the missing reaction might involve ammonium as a substrate and proceed spontaneously at an acceptable rate at very high ammonium concentrations. In the absence of a B6 vitamer, tamA- pyroB- double mutants superficially resemble areA‘ mutants in utilizing ammonium but not other nitrogen sources. In the presence of a B6 vitamer, pyroB- strains with or without a tamA- mutation utilize all nitrogen sources normally. Arst et al. (1982) found no evidence that pyroB- or ramA- mutations, alone or combination, affect regulation of a number of nitrogen metabolite-repressible enzymes. Employing the selection method used to obtain “tamAr-50”, Arst et al. were able to select a tamA- pyroB- double mutant directly from a wild-type strain, and it seems likely that “tamAr-50” was such a double mutant. Auxotrophies due to pyroB-, as might be expected, can be suppressed by mutations leading to accumulation of ammonium, such as gdhA- (loss of NADP+GDHase) and glnA- (loss of glutamine synthetase). As these mutations lead to nitrogen metabolite de-repression (Section 1II.Q it is possible that one or the other type might have been responsible for the de-repression attributed to “tamAd-l.” It is particularly pertinent that glnA- mutations, which are so leaky that no auxotrophy is evident, nevertheless result in substantial nitrogen metabolite de-repression (see Section III.C.2). The role of the tamA gene remains obscure, although involvement in membrane structure is suggested by indirect evidence (Arst et al., 1982). Elevated ammonium pool sizes (Kinghorn and Pateman, 1975a) might be the basis for the ability of tamA- mutations to lower the ammonium concentration necessary for pyroB- supplementation. It is worth

66

JEAN-MARIE WIAME E T A L .

noting that pdx-1 mutants of N. crassa respond to ammonium, especially at alkaline pH values (Perkins rt a f . , 1982, and references therein); pdn-I might therefore correspond to the pyroB gene of A. nidulans. d. The aniA and suF genes. Mutations at aniA in A. nidulans drastically lower the activities of arginase and ornithine 6-transaminase in the presence of good carbon and nitrogen sources, thereby preventing L-arginine from serving as a source of proline (through conversion into L-glutamic ysemialdehyde; Bartnik et al., 1973b, 1977). Mutations at suF physiologically suppress blocks prior to L-glutamic y-semialdehyde in the proline biosynthetic pathway by relieving carbon catabolite and/or nitrogen metabolite repression of synthesis of arginase and ornithine 6-transaminase (Bartnik et al., 1973b, 1977). As these enzymes function in utilization of arginine and ornithine as nitrogen and carbon sources as well as in this alternative route of proiine biosynthesis, and as many as yet uncharacterized genes are involved (Klimczuk and Weglenski, 1974; Bartnik and Weglenski, 1974; Bartnik e f al., 1977), it seems premature to attempt to speculate on what role, if any, aniA and suF might play in nitrogen metabolite repression. e . The nmr-1 and MS5 mutations. Mutations in the nmr-1 gene of N. crassa were selected on the basis of conferring sensitivity to chlorate in the presence of L-glutamine (which protects the wild type against chlorate toxicity by repressing nitrate reductase synthesis; Dunn-Coleman et al., 1981a; Tomsett et al., 1981). These mutations elevate levels of nitrate and nitrite reductases under both nitrogen metabolite de-repressing and repressing growth conditions. The nmr gene symbol notwithstanding, there is no convincing evidence that these mutations affect nitrogen metabolite repression rather than, for example, induction or, as Dunn-Coleman et al. (1981a) note, mRNA stability. Mutations in nit-2 are epistatic to nmr-1 mutations, leading Dunn-Coleman et al. (1981a) to propose that the nmr-1 gene product acts post-transcriptionally. It is unclear whether the effects of nmr- 1 mutations extend beyond the nitrate assimilation pathway. Premakumar e f al. (1980) selected a mutant strain of N. crassa, designated MS5, as having nitrate reductase activity on L-glutamine-containing medium (on which the wild-type nitrate reductase is nitrogen metabolite repressed). This mutation leads to elevated levels of some, but not all, nitrogen metabolite-repressible activities (Premakumar et al., 1980; Chambers et al., 1983). Unfortunately, owing to ascospore infertility, genetic data on MS5 are almost non-existent. It is not even clear that the MS5 phenotype is the result of a single mutation. Preliminary evidence that the MSS mutation(s) recombines with nif-2 was presented, but asco-

NITROGEN CATABOLITE REPRESSION IN FUNGI

67

spore viability was so low that rather tight linkage would be a possibility if wild-type recombinant ascospores germinate preferentially. Whether MS5 carries an nmr-1 mutation has not been reported.

f. The en(amf-1 tformerly i) allele. This mutation was found segregating in an N . crassa cross in which it made the L-glutamate auxotrophy of ammutants lacking NADP+-GDHase absolute (Fincham, 1981). It does not affect levels of glutamate synthase, the alternative glutamate-synthesizing enzyme (Dunn-Coleman et al., 1981b). Chambers and Wilkins (1982) showed that en (am)-I alters regulation of proline oxidase, an enzyme not under nit-2 control (Facklam and Marzluf, 1978), and that it leads to an inability to utilize several nitrogen sources including L-proline. It also has a strange effect on regulation of L-amino acid oxidase (Chambers et al., 1983). The altered regulatory patterns, although striking, are difficult to interpret and there is no reason to suppose that nitrogen metabolite repression in particular is affected. g. The gin' allele. Gonzalez et al. (1983) selected a mutation in N . crassa

designated gln' as allowing arginine to supplement a proline auxotroply in the presence of glutamine. Supplementation of proline auxotrophies by arginine requires three activities in addition to the final enzyme of de n o w proline biosynthesis: arginine uptake, arginase and ornithine &transaminase. The gln' mutant relieves glutamine repression of arginase (and also of glutamine synthetase), but regulation of the other two activites was not investigated. According to Facklam and Marzluf (1978), arginine uptake is subject to nitrogen metabolite repression but ornithine 6-transaminase (as well as arginase!) is not. Although gin' was shown to relieve repression of glutamine synthetase by ammonium, ammonium repression of arginase was not investigated and gln' actually prevents supplementation of amino acid auxotrophies in the presence of ammonium. Unfortunately, the effects of gln' on the regulation of other enzymes and permeases involved in nitrogen nutrition was not studied. Arginase is atypical in that it functions in an alternative route of proline biosynthesis as well as in utilization of arginine as nitrogen source. Genetic analysis of gln' is similarly incomplete. Gonzalez et al. (1983) report that gln' is not allelic to nit-2, en(am)-1, or structural gene mutations resulting in loss of NADP+-GDHase or glutamine synthetase, but phenotypes of double mutants are not given. The gln' mutation raised intracellular concentrations of glutamine and other amino acids and it seems possible that, rather than affecting nitrogen metabolite repression per se, it defines a gene involved in the general control of amino acid biosynthesis (see Barthelmess, 1982).

JEAN-MARIE WIAME ET AL.

68

TABLE I I . The effects of two nreB mutations on nitrate reductase levels in A . nidulans" Relative nitrate reductase activity

Relevant genotype

Uninduced

Wild type (areB') areB-401 areB-403 a

Induced

Repressed (co-inducer present)

100 78 54

5 32 11

1


Data from Tollervey and Arst (1982).

3 . A Pseudogene Able to Participate in Nitrogen Metabolite Repression

Tollervey and Arst (1982) selected in A . nidufans five mutations, unlinked to areA, which were unusual in that, unlike the vast majority of extragenic suppressors, they suppress areA' mutations for utilization of apparently all nitrogen sources (and presumably therefore for expression of all structural genes under areA control). These mutations, which are induced at an extremely low frequency, relieve nitrogen metabolite repression to a modest extent (Table 11). These mutations, designated areB and probably allelic, map to a region on the left arm of linkage group I. Although they are quite close to the regulatory gene mediating carbon catabolite repression creA, at least one unrelated gene adA has been located in between (Figs. 8-10). Mutations at areB are partially dominant and lead to various morphological abnormalities. Four of the five areB mutations are associ-

acuJ

&A I

' 3.7'

8.0 - 1 1 . 3 4

I

&A I

=A .

&D

'7.815.1' 4.3 ' 0 . 6 '

OB

E I

I

10.4

'

. . . -.

...

FIG. 8. Partial map of the left arm of linkage group I of A . nidulans (H.N. Arst Jr., unpublished observations). (a)Centromere. Distances (in centimorgans) in intervals between acuJ and adA (including the 7.8 cM distance beween medA and adA) are based on analysis of 460 progeny from a chromosomal aberration-free cross of relevant partial genotype acuJ-211 x fpaB-37 medA-15 adA-3. In intervals between medA and pyroB (including the 5.1 cM distance between medA and adA), distances are based on analysis of 510 progeny from a chromosomal aberration-free cross of relevant partial genotype medA-15 a&-3 galD5 x creAd-l pyroB-100. Using the latter cross, the relative order of creA and galD shown above was confirmed by analysis of 218 progeny selected as ndA+ creA+. Of these, 217 carry galD-5, 36 carry pyroB-100, and 13 carry medA-15.

NITROGEN CATABOLITE REPRESSION IN FUNGI

.

medA BB-403 &A ... ' 3.5 ' 2.7 '

69

. I .

FIG. 9 . Map position of the areB-403 mutationkeciprocal translocation involving linkage groups I and VII (H. N. Arst Jr., unpublished data). The 3.5 cM distance between medA and areE-403 is based on recovery of 114 medA+ areB+ recombinants out of 6501 progeny from a cross of relevant partial genotype fpas-37 medA-15 x areB-403 galD-5. Of these 114, none carry fpaE-37 and five carry galD-5. The 2.7 cM distance between areE-403 and adA is based on recovery of 38 areE+ adA+ recombinants out of 1430 adA+ progeny from a cross of relevant partial genotypefpaB-37 medA-I5 adA-3 X areE-403 galD-5. Of these 38, 36 carry medA-15, 32 carry fpaE-37, and 34 carry galD-5.

ated with reciprocal translocations (Tollervey and Arst, 1982). Remarkably, two of these, although phenotypically distinguishable, have almost identical breakpoints in linkage group VII less than 0.1 cM from nicB-8 on the side opposite to that of the palD gene. When a heteroallelic diploid is haploidized, markers in the non-translocated portions of linkage groups I and VII segregate independently with recovery of all four classes (H. N. Arst Jr., unpublished observations). This confirms that the two translocations involve exchanges which are, to the nearest indispensable gene, the same (although the orientations of the translocated segments need not be the same). The fifth allele, areB-405, which is not translocation associated, recombines with markers centromere-distal to it at drastically de-

. ..

acuJ

fmB

I

m

A EB-405 I

.

0.04 -0.1

3.9

4 A

..

4

+3.1-

FIG. 10. Map position of the Aspergillus nidulans areB-405 mutatiodchromosomal aberration (H. N. Arst Jr., unpublished observations). The 3.1 cM distance between a c d and areE-405 is based on recovery of 44 acuJ+ areE+ recombinants out of 1406 a c d + progeny from a cross of relevant partial genotype acuJ-211 x areE-405. The 0.04 cM distance between medA and areE-405 is based on recovery of 2 me&+ areE+ recombinants (both fpaB+) out of 9372 progeny from a cross of relevant partial genotype fpaE-37 medA-15 x areE-405. The 0.1 cM distance between fpaE and areE-405 is based on recovery of fivefpaE37 areB-405 recombinants (all carryingpyroE-100 and adA+, with medA-I5 not scorable in the presence of areB-405) out of 3798 fpaB-37 progeny from a cross of relevant partial genotype fpaE-37 medA-I5 adA-3 x areE-405 pyroE-100. The 3.9 cM distance between areB-405 and adA is based on analysis of 309 non-selected progeny from this last cross (with no recombination between areE-405 and medA-15 orfpaE-37). Also, using the last cross, the relative order of medA and areE-405 was confirmed by analysis of 199 areB+ adA+ recombinants. Of these 199, 198 carry medA-15, 197 carry fpaE-37 and 168 carrypyroB-100. Additionally 205 areE+ adA+ recombinants were selected from a cross of relevant partial genotype medA-15 adA-3 galD-5 X areE-405 pyroE-100. Of these 205, all carry me&-15, 12 carry gafD-5 and 171 carry pyroB-100.Using this final cross, 50 me&+ areB+ recombinants were selected. Of these 50, 49 carry adA-3, 43 carry galD-5 and 11 carry pyroB-100.

70

JEAN-MARIE WIAME

EJ AL.

creased frequencies (Fig. 10 as compared to Figs. 8 and 9 ) , indicating the presence of a chromosomal aberration (H. N. Arst Jr., unpublished observations). Recombination frequencies between areB-405 and markers centromere-proximal to it seem normal (Fig. 10; Tollervey and Arst, 1982). The extreme rarity of areB mutations, their apparently obligatory association with chromosomal rearrangements, and the nearly identical nature of two of the translocations imply that certain chromosomal rearrangements are themselves responsible for the areB genetic phenotype. An attractive hypothesis, which would also account for the ability of areB mutations to suppress any areA‘ mutation for expression of all structural genes in the areA regulatory domain, is that areB is an areA-related pseudogene which can be activated by fusion to a functional promoter and/or ribosome-binding sequence. Very recently a mutation abolishing the ability of an areB mutation to suppress areA‘ mutations has been obtained (H. N. Arst Jr., unpublished observations). It is extremely tightly linked to the original areB mutation (<0.01 cM), but is definitely separable from it and does not affect the associated translocation. This mutation has been designated areB‘-900. It is apparently silent in an areA+ background; it is recessive and it is likely to be a loss of function mutation in the areB gene. Haploidization of a diploid homozygous for the translocation shows that areB‘-900 is linked to the centromere of linkage group I. This would indicate that areB is transcribed towards the centromere, an inference supported by the fact that the reduction in recombination frequencies (and therefore the chromosomal rearrangement) associated with areB-405 occurs centromere-distal to areB. Thus areB is probably the first pseudogene to be identified by purely genetic means. C. L-GLUTAMINE IS THE NITROGEN METABOLITE CO-REPRESSOR I N Neurospora crassa AND Asperigillus nidulans

The ability of L-glutamine to elute specifically the nit-2 product from N . crassa DNA-cellulose (Grove and Marzluf, 1981) strongly suggests that it is the co-repressor and that it acts by decreasing the affinity of the nit-2 product for its receptor sites. Less direct evidence lends further support to this hypothesis in both A . nidulans and N . crassa.

1 . NADP+-Linked Glutamate Dehydrogenase Is Required for Repression by Ammonium The gdhA gene is the structural gene for NADP-linked (“anabolic”) glutamate dehydrogenase in A . nidulans (Kinghorn and Pateman, 1975b),

NITROGEN CATABOLITE REPRESSION IN FUNGI

71

whereas the extensively characterized and recently cloned am gene codes for this enzyme in N. crassa (Kinnaird et al., 1982, and references therein). Mutations at gdhA relieve repression by ammonium in A. nidulans (Arst and MacDonald, 1973; Kinghorn and Pateman, 1973; Pateman et al., 1973), but do not relieve repression by L-glutamate or L-glutamine (Hynes, 1974a). Similar observations have been made in N. crassa (Vaca and Mora, 1977; Ketchum et al., 1977; Dantzig et al., 1978; Premakumar et al., 1979; Dunn-Coleman et al., 1979). The en(am)-2 mutation in N . crassa leading to loss of glutamate synthase does not relieve repression by ammonium (Dunn-Coleman et al., 1981b). However, glutamate synthase is probably only important for glutamate synthesis under nitrogenlimited (and therefore de-repressing) growth conditions (Hummelt and Mora, 1980; Lara et al., 1982). 2. Glutamine Synthetase Is Required for Repression by Ammonium and L-Ghtamate In the presence of the glutamine synthetase inhibitor L-methionine-DLsulphoximine, ammonium is unable to repress synthesis of nitrogen metabolite-repressible enzymes in N. crassa (Premakumar et al., 1980). Mutations in gln, the structural gene for glutamine synthetase in N. crassa (Davila et al., 1980, 1983), relieve repression by ammonium and L-glutamate but not by L-glutamine (Premaknmar et al., 1979, 1980; DunnColeman et al., 1979; Dunn-Coleman and Garrett, 1980). Similar observations have been made in A . nidulans (Arst et al., 1982; MacDonald, 1982; Cornwell and MacDonald, 1984). However, Dunn-Coleman and Garrett (1981), using different growth conditions and a gln- allele different from that used in their previous work, observed some de-repression in the presence of glutamine, and interpreted this as supporting a model proposed in their 1980 paper in which the octomeric form of glutamine synthetase represses synthesis of the nit-2 gene product. There is no experimental evidence to refute this model, but it seems unnecessarily complicated. Arst et al. (1982) have pointed out that if, for any reason (such as compartmentation) endogenously synthesized and exogenously supplied glutamine are not equivalent, then comparisons of glutamine repression in gln+ and gln- strains are of dubious value. Repression of glutamine synthetase synthesis by glutamine is incomplete (Pateman, 1969; Sanchez et al., 1978; DunnColeman and Garrett, 1981; Lara et al., 1982) and in A . nidulans there is evidence that glutamine synthesis occurs in the wild type even during growth in the presence of high concentrations of glutamine (Arst et al., 1982).

72

JEAN-MARIE WlAME ET AL.

3. Repression by A m ~ o n j u mAnalogues Holloman and Dekker (1971) presented evidence that the ammonium analogues Cs+ and various alkylammonium ions, particularly methylammonium, repress nitrogen metabolite-repressible enzymes in Ustilago sphaerogena. Similar observations were made in A . nidulans (Bartnik et al., 1973a; Hynes 1974b). In view of the wealth of evidence presented in previous sections that ammonium is able to repress only through conversion into glutamine, the interpretation of these experiments is difficult. There are a variety of indirect ways in which ammonium analogues might effect repression, e.g., inhibition of glutaminase. Methylammonium might act more directly as it can be converted into the L-glutamine analogue L-yglutamylmethylamide by glutamine synthetase (Kung and Wagner, 1969; Rowe et al., 1970). Readers should note that repression by Cs+ and methylammonium is not necessarily the sole basis for the toxicities of these cations mentioned in previous sections. Sensitivity and resistance to their toxicities follows a pattern shared by a number of other inhibitors (e.g., chlorate, thiourea, and DL-P-aspartylhydroxamate) which do not resemble ammonium and whose toxicities affect different targets. This pattern is a reversal of toxicity by exogenous ammonium and L-glutamine and by mutations (of various sorts) which enhance catabolism of the nitrogen source (Arst and Cove, 1969, 1973; Bailey and Arst, 1975; Bailey et al., 1979, 1980; Arst and Bailey, 1980) and enhancement of toxicity by mutations that decrease nitrogen source utilization (Rand and Arst, 1977; Jones et al., 1981; Arst et al., 1981).

D. INTERACTIONS WITH OTHER REGULATORY SYSTEMS

1. Pathway-Specific Regulatory Genes Many structural genes whose expression is subject to nitrogen metabolite repression are also under the control of pathway-specific regulatory genes mediating induction. In these cases, structural gene expression is dependent upon both induction and absence of nitrogen metabolite repression. Some insight into the relationship between the two forms of control is available in the nitrate assimilation pathway of A . nidulans. The positiveacting regulatory gene nirA mediates induction of the syntheses of nitrate and nitrite reductases. Gain of function constitutivity mutations (nirA') and loss of function mutations (nirA-) leading to non-inducibility have been available for some time (reviewed by Cove, 1979). More recently, a second type of gain of function mutation (designated nirAd) leading to

NITROGEN CATABOLITE REPRESSION IN FUNGI

73

nitrogen metabolite de-repression and bypassing the requirement for a functional areA gene product has been described. Mutations at nirAdwere selected as suppressing an areA' mutation for utilization of nitrate and nitrite in strains also carrying an nirAc mutation (Rand and Arst, 1978; Tollervey and Arst, 1981). The resultant nirA alleles were designated nirACldto indicate that they differ from the wild-type allele by two gain-offunction mutations. An alternative selection method would be to revert an areA' meaB- or areA' ninA- strain on nitrite (but not nitrate), relying on the apparent ability of meaB- and ninA- mutations to retard nitrite efflux (Rand and Arst, 1977; Rand, 1978; Tollervey and Arst, 1981). Reversion of an areA' strain on nitrite yielded a strain carrying a ninA- mutation and an nirAd mutation (Tollervey and Arst, 1981). This allowed an nirACld allele to be constructed by recombination (as present in the nitratehitriteutilizing progeny of an areA' nirA" X areA' nirAd cross). Tables 12 and 13 illustrate the effects of nirAd and nirACldalleles on enzyme levels. The nirA gene apparently therefore contains at least two genetically separable functional domains-one in which mutations can lead to constitutivity, possibly encoding a co-inducer binding site, and another in which mutations can lead to nitrogen metabolite de-repression, possibly encoding a region interacting with the areA gene product or with initiator sites adjacent to structural genes under both areA and nirA control (Rand and Arst, 1978; Tollervey and Arst, 1981). As noted by Rand and Arst (1978) TABLE 12. The effects of nirAc and nirA'" alleles on levels of the nitrate assimilation pathway enzymes in A . nidulans" Relative activity Relevant genotype Wild type (nirA ) +

nirAc-l

nirAc'd-101

Nitrogen source(s)

Nitrate reductase

Nitrite reductase

Urea NO3NO3-+ NH4+ NH4+ Urea NO3NO3-+ NH4+ NH4+ Urea NO3NO3-+ NH4+ NH4+

6 I00 2 2 75 98 20 3 85 143 79 85

2 I00 9 2 29 117 21 6 27 111 63 40

Data from Rand and Arst (1978).

74

JEAN-MARIE WlAME ETAL. TABLE 1 3 . The effects of nirAd and nirACIdalleles on levels of the nitrate assimilation pathway enzymes in A . nidulans" Relative activity Relevant genotype Wild type (nirA+ )

nirAd-106

nirAc'd-l13

a

Nitrogen source(s)

Nitrate reductase

L-Proline NO3NO,-+ NH4+ NH4+ L-Proline NO,NO3-+ NH4+ NH4+ L-Proline NO?NO3 + NH,' NH4+

1 100 5 1

Nitrite reductase 2 100

5 1

3

5

120

134 61 6

70 10 93 155

151 126

47 81 82 60

Data from Tollervey and Arst (1981).

and Tollervey and Arst (1981), the various mutant phenotypes in this system closely parallel those in the L-arabinose regulon of Escherichia coli, in which structural gene expression also requires the positive action of two regulatory gene products, one pathway-specific mediating induction and one having a wider domain mediating carbon catabolite repression. A further point is that suppression of areA' mutations requires both a mutation relieving nitrogen metabolite repression (nirAd) and a mutation conferring constitutivity (nirAC)or elevating intracellular co-inducer concentrations (ninA- or meaB-). This indicates that the inability of areA' strains to utilize nitrate and nitrite is a consequence both of inducer exclusion and of nitrogen metabolite repression of the syntheses of nitrate and nitrite reductases. In support of this interpretation, Brownlee and Arst (1983) have shown that nitrate uptake in A . nidulans is extremely sensitive to nitrogen metabolite repression and that even the most markedly de-repressing nirAc'dalleles fail to alleviate this sensitivity.

2 . Carbon Catabolite Repression Carbon catabolite repression of the syntheses of many enzymes and permeases involved in carbon and energy provision is mediated by a negative-acting regulatory gene creA in A . nidulans (Bailey and Arst, 1975;

NITROGEN CATABOLITE REPRESSION IN FUNGI

75

Arst and MacDonald, 1975; Arst and Bailey, 1977; Arst, 1981). Maximal nutritional versatility would be achieved if carbon catabolite repression and nitrogen metabolite repression were to operate independently so that neither form of repression alone could ever provoke starvation wherever both carbon and nitrogen nutrients are present. In A . nidulans there is evidence for such independence for two enzymes: maximal concentrations of proline oxidase (Arst and MacDonald, 1975) and acetamidase (Hynes, 1972) are achieved only in the absence of both forms of repression. In support of this independence, areA' mutants are able to utilize compounds such as acetamide, y-amino-n-butyrate, L-arginine, L-asparagine, L-glutamine, and L-proline as sole carbon (or sole carbon and nitrogen) sources although they cannot utilize these compounds as sole nitrogen sources in glucose-containing medium (Arst and Cove, 1973). In the presence of de-repressing carbon sources, such as ethanol, glycerol, lactose, or L-arabinose they can utilize acetamide, y-amino-n-butyrate, Lglutamine, and L-proline (but, mysteriously, not L-arginine, L-asparagine, or a number of other carbon and nitrogen sources) as nitrogen sources. Mutations designated creAd, leading to carbon catabolite de-repression, suppress areA' mutations for utilization of these same four nitrogen sources in the presence of a repressing carbon source (Arst and Cove, 1973; Arst and Bailey, 1977).

3. Sulphur and Phosphorus Regulation Proteases of A . nidulans and N . crassa are subject to carbon, nitrogen, sulphur, and even phosphorus control (Cohen, 1973; Hanson and Marzluf, 1975; Cohen and Drucker, 1977; the subject is reviewed by North, 1982). An acid phosphatase of N. crassa is subject to both nitrogen and phosphorus control (Grove and Marzluf, 1980). Little is known about how the various regulatory systems interact but in each case starvation for any one category of nutrient provokes de-repression. Thus areA' mutants are able to utilize taurine, hypotaurine, and L-methionine as sulphur sources, although they are unable to utilize them as nitrogen sources (Arst and Cove, 1973; Shaffer and Arst, 1984; H. N. Arst Jr., unpublished observations).

4. Supplementation of Auxotrophies and Alternative Routes for Amino Acid Biosynthesis

There is no known instance in which areA' mutations prevent supplementation of an auxotrophic requirement by a compound whose use as a

76

JEAN-MARIE WIAME E T A L .

nitrogen source they prevent (Arst and Cove, 1973). The mechanisms allowing this have not been elucidated but they must reflect overriding of areA control by other regulatory mechanisms and/or the existence of alternative activities specified by structural genes whose expression is dependent on starvation for a particular metabolite but is independent of areA control.

5 . Oxygen Repression The pyrimidine salvage pathway enzyme thymine 7-hydroxylase is subject to repression by oxygen when grown under nitrogen metabolite derepressed conditions in A. nidulans (Shaffer and Arst, 1984). The areA300 allele relieves oxygen as well as nitrogen metabolite repression (Shaffer and Arst, 1984). A role for areA in oxygen repression is possibly precedented by the involvement of niJZ in both fixed nitrogen repression and oxygen repression in Klebsiella pneurnoniae (Hill et al., 1981; Buchanan-Wollaston et a f . , 1981; Merrick et a f . , 1982; Filser et al., 1983). However, recent evidence suggests that the nifZ gene product, unlike that of areA, acts indirectly (V. Buchanan-Wollaston and F. C. Cannon, as cited in Beynon et al., 1983).

E . CIS-ACTING REGULATORY MUTATIONS

1 . Identification of a Receptor Site for the areA Product The areA-102 mutation of A . nidulans results in inability to utilize xanthine and uric acid as nitrogen sources as a consequence of defective uptake of these compounds (Arst and Cove, 1973). The uap-100 mutation, selected as suppressing areA-102 for xanthine and uric acid utilization, is a cis-acting regulatory mutation, adjacent to uapA, the putative structural gene for a xanthine-uric acid permease (Arst and Scazzocchio, 1975). The uap-100 mutation alters the areA receptor site to accommodate the areA- 102 product (rather than bypassing the requirement for a functional areA gene product), exerts a 2.5-fold “up-promoter” effect on uapA expression, and results in constitutive (but not nitrogen metabolite de-repressed) expression of the uapA permease. Although uap-100 largely alleviates the requirement for co-inducers (e.g., uric acid), it does not relieve the stringent requirement for the product of ua Y , the pathway-specific positive-acting regulatory gene mediating induction (Scazzocchio and

NITROGEN CATABOLITE REPRESSION IN FUNGI

77

Arst, 1978). Thus uap-100 defines a promoter and two initiator sites (for areA and uaY), and its phenotype implies that a functional overlap exists between them. The uap-100 phenotype lends further strong support for a direct role of areA (and also uaY) in control of gene expression. The phenotypes of two other regulatory mutations closely linked to uapA suggest that they affect the areA receptor site exclusively (Gorton et al., 1978). 2 . Fusion of the Structural Gene for Nitrite Reductase in A . nidulans to an areA-independent Promoterllnitiator Region The nis-5 mutation of A . nidulans was selected as suppressing an areA' mutation for nitrite utilization and protecting against nitrite toxicity in a strain also carrying a rneaB- mutation (see Sections IIl.B.1 and III.B.2.a), a phenotype indicating enhanced nitrite reductase levels (Rand and Arst, 1977). The structural gene for nitrite reductase, niiA, is tightly linked to (probably contiguous with) niaD, the structural gene for nitrate reductase, and c r A , a gene involved in nitrate (but not nitrite) uptake, in the order crA-niiA-niaD (Tomsett and Cove, 1979; Brownlee and Arst, 1983). The nis-5 mutation raises both the uninduced and the nitrogen metaboliterepressed levels of nitrite reductase, but has no detectable effect on regulation of nitrate reductase or nitrate uptake and does not appear to affect the structure of nitrite reductase (Rand and Arst, 1977; Brownlee and Arst, 1983). In agreement with its biochemical phenotype, nis-5 partially suppresses nirA- (see Section 1II.D.I as well as areA' mutations for nitrite but not nitrate utilization (Rand and Arst, 1977). Thus nis-5 is a nonreciprocal translocation in which a large segment of another linkage group is inserted between the niiA and niaD genes (Arst et al., 1979). It is therefore likely that nis-5 fuses niiA to a new promotedinitiator region independent of areA and nirA control, implying that niiA is transcribed towards c r A . As maximal niiA expression in nis-5 strains requires nitrate (or nitrite) induction and nitrogen metabolite de-repression (Rand and Arst, 1977), the normal niiA promoter/initiator region is still present, presumably in tandem with the translocated regulatory region. The fact that the nis-5 insertion does not abolish expression of either niaD or niiA (and c r A ) shows that a single tricistronic or niaD niiA dicistronic transcript does not solely account for expression of niaD and niiA. Maximal expression of niiA in nis-5 strains is only about 40% of that in the wild type (Rand and Arst, 1977). This might indicate partial disruption of the normal niiA promoter/initiator or suggest that the nis-5 insertion abolishes synthesis of an overlapping polycistronic transcript for niiA initiated at the niiA-distal

73

JEAN-MARIE WIAME ET AL.

side of niaD (Arst et al., 1979). In either case it seems safe to conclude that one areA product receptor site in the nitrate assimilation gene cluster is located between niiA and niaD.

3 . Other cis-Acting Regulatory Mutations Selected for Suppression of areAr Mutations on Certain Nitrogen Sources

There are a few other cis-acting regulatory mutations in A . nidulans which, because of the involvement of multiple forms of regulation, do not necessarily define areA product receptor sites. Apart from their utility in studying the regulation of individual structural genes and gene clusters (outside the scope of this review), they do at least indicate that inability of areA‘ mutants to utilize a given nitrogen source is due to lack of a particular activity. Thus the ability of gab1 mutations to suppress areA‘ mutations for utilization of y-amino-n-butyrate establishes that lack of uptake is responsible for the areA‘ phenotype on this nitrogen source (Bailey et d., 1979). Similarly, suppression of areAr mutations for acetamide utilization by amdl mutations establishes that deficiency of acetamidase is responsible for the areA‘ phenotype in this case (Hynes, 1975b, 1978; Arst, 1978). The same sort of reasoning with nis-5 identifies nitrite reductase as the crucial deficiency, whereas uap- 100 confirms that uptake deficiency is responsible for the inability of areA-102 strains to utilize xanthine and uric acid. Finally, the ability of prnd mutations to suppress areA’ mutations for Lproline utilization indicates that an uptake deficiency is also crucial in this case (Arst and MacDonald, 1975; Arst et al., 1980). Forprnd mutations, a slight qualification is necessary: the prnB gene, specifying a proline permease, is part of a four-gene cluster. There is no evidence that prnd mutations affect regulation of any of the other genes of the cluster in addition to prnB, but the evidence does not eliminate this possibility either (Arst et al., 1980).

IV. Acknowledgements Work in the laboratory of H. N. Arst was supported by the Royal Society and the Science and Engineering Research Council. Alan Brownlee, Joan Kelly, Claudio Scazzocchio, and Mark Caddick kindly commented on the manuscript. Work in the laboratory of M. Grenson and J. M. Wiame was supported by Grant No. 2.2929.79 from the Fonds de la Recherche Fondamentale Collective and by Grant No. 80/85-15 from an “Action de Recherches Concertke” between the Belgian Government and the Universitk Libre de Bruxelles.

NITROGEN CATABOLITE REPRESSION IN FUNGI

79

REFERENCES

A1 Taho, N. M., Sealy-Lewis, H. M., and Scazzocchio, C. (1984). Current Genetics, 8, 245. Arst, H. N., Jr. (1977). Molecular and General Genetics 151, 105. Arst, H. N., Jr. (1978). Molecular and General Genetics 163, 23. Arst, H. N., Jr. (1981). Symposium ofthe Societyfor General Microbiology 31, 131. Arst, H. N., Jr. (1982). Molecular and General Genetics 188, 490. Arst, H. N., Jr. and Bailey, C. R. (1977). In “Genetics and Physiology of Aspergillus” (J. E. Smith and J. A. Pateman, eds.), pp. 131-146. Academic Press, London. Arst, H. N., Jr. and Bailey, C. R. (1980). Journal ofGenera1 Microbiology 121, 243. Arst, H. N., Jr. and Cove, D. J. (1969). Journal ofBacteriology 98, 1284. Arst, H. N., Jr. and Cove, D. J. (1973). Molecular and General Genetics 126, 11 1. Arst, H. N., Jr. and MacDonald, D. W. (1973). Molecular and General Genetics 122, 261. Arst, H. N., Jr. and MacDonald, D. W. (1975). Nature, London 254, 26. Arst, H. N., Jr. and Page, M. M. (1973). Molecular and General Genetics 121, 239. Arst, H. N . , Jr. and Scazzocchio, C. (1975). Nature, London 254, 31. Arst, H. N., Jr., Rand, K. N., and Bailey, C. R. (1979). Molecular and General Genetics 174, 89. Arst, H. N., Jr., MacDonald, D. W., and Jones, S. A. (1980). Journal of General Microbiology 116, 285. Arst, H. N., Jr., Jones, S. A., and Bailey, C. R. (1981). Genetical Research 38, 171. Arst, H. N., Jr., Brownlee, A. G., and Cousen, S. A. (1982). Current Genetics 6, 245. Bailey, C. and Arst, H. N., Jr. (1975). European Journal ofBiochemistry 51,573. Bailey, C. R., Penfold, H. A., and Arst, H. N., Jr. (1979). Molecular and General Genetics 169, 79. Bailey, C. R., Arst, H. N., Jr. and Penfold, H. A. (1980). Genetical Research 36, 167. Barnett, J. A., and Kornberg, H. L. (1960). Journal of General Microbiology 23, 65. Barnett, J. A., Payne, R. W., and Yarrow, D. Y. (1983). “Yeasts: Characteristics and Identification.” Cambridge University Press, Cambridge. Barthelmess, I. B. (1982). Genetical Research 39, 169. Bartnik, E., and Weglenski, P. (1974). Nature, London 250, 590. Bartnik, E., Weglenski, P., and Piotrowska, M. (1973a). Molecular and General Genetics 126, 75. Bartnik, E., Guzewska, J., and Weglenski, P. (1973b). Molecular and General Genetics 126, 85. Bartnik, E., Klimczuk, J., Kowalska, I., and Weglenski, P. (1976). Acta Microbiologica Polonica 25, 169. Bartnik, E., Guzewska, J., Klimczuk, J., Piotrowska, M., and Weglenski, P. (1977). In “Genetics and Physiology of Aspergillus” (J. E. Smith and J. A. Pateman, eds.), pp. 243-254. Academic Press, London.

80

JEAN-MARIE WIAME ET AL.

BCchet, J., and Wiame, J. M. (1965). Biochemical and Biophysical Research Communications 21, 226. BCchet, J., Grenson, M., and Wiame, J. M. (1970). European Journal ofBiochemistry 12, 3 1 . Beynon, J., Cannon, M., Buchanan-Wollaston, V., and Cannon, F. C. (1983). Cell 34, 665. Bossinger, J., and Cooper, T. G. (1975). Biochemical and Biophysical Research Communications 66, 889. Bossinger, J., Lawther, R. P., and Cooper, T. G. (1974). Journal of Bacteriology 118, 821. Brandriss, M. C. (1979). Journal of Bacteriology 138, 816. Brandriss, M. C., and Magasanik, B. (1979). Journal of Bacteriology 140, 498. Brandriss, M. C., and Magasanik, B. (1980). Journal of Bacteriology 143, 1403. Brandriss, M. C., and Magasanik, B. (1981). Journal of Bacteriology 145, 1359. Brenchley, J. E., and Magasanik, B. (1974). Journal of Bacteriology 117, 544. Brownlee, A. G., and Arst, H. N., Jr. (1983). Journal of Bacteriology 155, 1138. Buchanan-Wollaston, V., Cannon, M. C., and Cannon, F. C. (1981). Molecular and General Genetics 184, 102. Chambers, J. A. A., and Wilkins, S. A. (1982). Current Genetics 6, 87. Chambers, J. A. A., Griffon, S. M., and Marzluf, G. A. (1983). Current Genetics 7, 51. Chevallier, M. R., Jund, R., and Lacroute, F. (1975). Journal ofBacteriology 122, 629. Clutterbuck, A. J. (1982). Genetic Maps 2, 208. Coddington, A, (1976). Molecular and General Genetics 145, 195. Cohen, B. L. (1972). Journal of General Microbiology 71, 293. Cohen, B. L. (1973). Journal of General Microbiology 79, 3 1 1 . Cohen, B. L., and Drucker, H. (1977). Archives of Biochemistry and Biophysics 182, 601. Cohen, G. N., and Monod, J. (1957). Bacteriological Reviews 21, 169. Cook, R. J . , and Anthony, C. (1978). Journal of General Microbiology 109, 265. Cornwell, E. V., and MacDonald, D. W. (1983). Current Genetics 8, 33. Cooper, T. G. (1978). Biochemical and Biophysical Research Communications 82, 1258. Cooper, T. G. (1982a). In “The Molecular Biology of the Yeast Saccharomyces: Metabolism and Gene Expression” (J. N. Strathern, E. W. Jones, and J. R. Broach, eds.), p. 39. Cold Spring Harbor Laboratory, New York. Cooper, T. G. (1982b). In “The Molecular Biology of the Yeast Saccharomyces: Metabolism and Gene Expression” (J. N. Strathern, E. W. Jones, and J. R. Broach, eds.), p. 399. Cold Spring Harbor Laboratory, New York. Cooper, T. G., and Lawther, R. P. (1973). Biochemical and Biophysical Research Communications 52, 137. Cooper, T. G., and Sumrada, R. (1975). Journal of Bacteriology 121, 571. Cooper, T. G., Lam, C., and Turoscy, V. (1980). Genetics 94, 555. Cove, D. J. (1966). Biochimica et Biophysica Actu 113, 51. Cove, D. J. (1979). Biological Reviews 54, 291. Courchesne, W. E., and Magasanik, B. (1983). Molecular and Cellular Biology 3, 672. Crabeel, M. (1973). Thesis, University of Brussels.

NITROGEN CATABOLITE REPRESSION IN FUNGI

81

Crabeel, M., and Grenson, M. (1970). European Journal of Biochemistry 14, 197. Dantzig, A. H., Zurowski, W. K., Ball, T. M., and Nason, A. (1978). Journal of Bacteriology 133, 671. Darte, C. (1976). Thesis, University of Brussels. Darte, C., and Grenson, M. (1975). Biochemical and Biophysical Research Communications 67, 1028. DBvila, G., Lara, M., Guzman, J., and Mora, J. (1980). Biochemical and Biophysical Research Communications 92, 134. DBvila, G., Brom, S., Mora, Y.,Palacios, R., and Mora, J. (1983). Journal of Bacteriology 156,993. Davis, R. H. (1963). Science 142, 1652. Davis, R. H. (1975). Annual Reuiews ofGenetics 9, 39. De Hauwer, G., LavallC, R., and Wiame, J.-M. (1964). Biochimica et Biophysica Acta 81, 257. Deschamps, J., and Wiame, J.-M. (1979). Genetics 92,749. Deschamps, J., Dubois, E., and Wiame, J.-M. (1979). Molecular and General Genetics 174, 225. Drillien, R., and Lacroute, F. (1972). Journal ofBacteriology 109,203. Drillien, R., Aigle, M., and Lacroute, F. (1973). Biochimica et Biophysica Acta 53, 367. Dubois, E., and Grenson, M. (1974). Biochemical and Biophysical Research Communications 60, 150. Dubois, E., and Grenson, M. (1979). Molecular and General Genetics 175, 67. Dubois, E., and Wiame, J.-M. (1976). Biochimie 58, 207. Dubois, E.,and Wiame, J.-M. (1978). Molecular and General Genetics 164, 275. Dubois, E.,Grenson, M., and Wiame, J.-M. (1973). Biochemical and Biophysical Research Communications 50, 967. Dubois, E., Grenson, M., and Wiame, J.-M. (1974). European Journal ofBiochemistry 48, 603. Dubois, E., Vissers, S., Grenson, M., and Wiame, J.-M. (1977). Biochemical and Biophysical Research Communications 15, 223. Dubois, E., Hiernaux, D., Grenson, M., and Wiame, J.-M. (1978). Journal of Molecular Biology 122, 383. Dubois, E., Jacobs, E., and Jauniaux, J. C. (1982). The EMBO Journal 1, 1133. Dunlop, P.C., Roon, R. J., and Even, H. L. (1976). Journal of Bacteriology 125, 999. Dunlop, P. C.,Meyer, G. M., Ban, D., and Roon, R. J. (1978). Journal of Biological Chemistry 253, 1297. Dunlop, P. C., Meyer, G. M., and Roon, R. J. (1980). Journal ofBacteriology 143, 422. Dunn-Coleman, N.S.,and Pateman, J. A. (1979). Molecular and General Genetics 171, 69. Dunn-Coleman, N. S., and Garrett, R. H. (1980). Molecular and General Genetics 179, 25. Dunn-Coleman, N. S., and Garrett, R. H. (1981). Journal ofBacteriology 145, 884. Dunn-Coleman, N. S., Tomsett, A. B., and Garrett, R. H. (1979). Journal of Bacteriology 139, 697. Dunn-Coleman, N . S., Tomsett, A. B., and Garrett, R. H. (1981a). Molecular and General Genetics 182, 234.

82

JEAN-MARIE WIAME E T AL.

Dunn-Coleman, N. S., Robey, E. A., Tomsett, A. B., and Garrett, R. H. (1981b) Molecular and Cellular Biology 1, 158. Eddy, A. A. (1982). Advances in Microbial Physiology 23, 1 . Elmerich, C. (1972). European Journal of Biochemistry 27, 216. Elmerich, C., and Aubert, J. P. (1971). Biochemical and Biophysical Research Communications 42, 37 1 . Errede, B., Cardillo, T. S., Sherman, F., Dubois, E., Deschamps, J., and Wiame, J.-M. (1980). Cell 22, 247. Ephrussi, B., Slonimski, P. P., Yotsuyanagi, Y., and Tavlitzki, X. (1956). Compte Rendus des Travaux du Laboratoire de Carlsberg (Se‘riePhysiologique) 26,87. Facklam, T . J., and Marzluf, G. A. (1978). Biochemical Genetics 16, 343. Ferguson, A. R., and Sims, A. P. (1974). Journal of General Microbiology 80, 173. Filser, M., Merrick, M., and Cannon, F. (1983). Molecular and General Genefics 191, 485. Fincham, J . R. S. (1981). Neurospora Newsletter 28, 1 1 . Fincham, J. R. S., Day, P. R., and Radford, A. (1979). In “Fungal Genetics”, 4th Ed. pp. 391-396. Blackwell Sc. Publ. Oxford. Gits, J., and Grenson, M. (1967). Biochimica et Biophysica Acta 135, 507. Gits, J., and Grenson, M. (1969). Archives Internationales de Physiologie et de Biochimie 77, 153. GonzAlez, A., Tenorio, M., Vaca, G., and Mora, J. (1983). Journal of Bacteriology 155, 1. Gorton, D. J., Scazzocchio, C., and Arst, H. N., Jr. (1978). Heredity 41, 129. Gregory, M. E., Keenan, M. H. J., and Rose, A. H. (1982). Journal of General Microbiology 128, 2257. Grenson, M. (1966). Biochimica et Biophysica Acta 127, 339. Grenson, M. (1969). European Journal of Biochemistry 11, 249. Grenson, M. (1973). In “Genetics of Industrial Microorganisms, Vol. 11: Actinomycetes and Fungi” (Z. Vanek, Z. Hosfalek, and J. Cudlin, eds.), p. 179. Academia, Prague. Grenson, M. (1983a). European Journal of Biochemistry 133, 135. Grenson, M. (1983b). European Journal of Biochemistry 133, 141. Grenson, M., and Acheroy, B. (1982). Molecular and General Genetics 188,261. Grenson, M., and Dubois, E. (1982). European Journal ofBiochemistry 121,643. Grenson, M . , and Hennaut, C. (1971). Journal of Bacteriology 105, 477. Grenson, M., and Hou, C. (1972). Biochemical and Biophysical Research Communications 48, 749. Grenson, M., Mousset, M., Wiame, J.-M., and BCchet, J. (1966). Biochimica et Biophysica Acta 127, 325. Grenson, M., Hou, C., and Crabeel, M. (1970). Journal of Bacteriology 103, 770. Grenson, M., Dubois, E., Piotrowska, M., Drillien, R., and Aigle, M. (1974). Molecular and General Genetics 128, 73. Greth, M. L., Chevallier, M. R., and Lacroute, F. (1977). Biochimica et Biophysics Acta 465, 138. Griswold, W. R., Madrid, V. O., Shaf€er, P. M., Tappen, D. C., Pugh, C. S. G., and Abbot, M. T. (1976). Journal of Bacteriology 125, 1040. Grove, G., and Marzluf, G. A. (1980). Journal of Bacteriology 141, 1470. Grove, G., and Marzluf, G. A. (1981). Journal of Biological Chemistry 256, 463.

NITROGEN CATABOLITE REPRESSION IN FUNGI

83

Hanson, M. A., and Marzluf, G. A. (1975). Proceedings of the National Academy of Sciences of the United States of America 72, 1240. Hemmings, B. A. (1978). Journal of Biological Chemistry 253, 5255. Hemmings, B. A., Zubenko, G. S., and Jones, E. W. (1980). Archives of Biochemistry and Biophysics 202, 657. Hierholzer, G., and Holzer, H. (1963). Biochemische Zeitschrgt 339, 175. Hill, S., Kennedy, C., Kavanagh, E., Goldberg, R. B., and Hanan, R. (1981). Nature, London 290, 424. Holloman, W. K., and Dekker, C. A. (1971). Proceedings of the National Academy of Sciences of the United States of America 68, 2241. Holzer, H . (1976). Trends in Biochemical Sciences 1, 178. Holzer, H., andwSchneider,S. (1957). Biochemische Zeitschrgt 329, 369. HorBk, J., and RihovB, L. (1982). Biochimica et Biophysica Acta 691, 144. Hummelt, G., and Mora, J. (1980). Biochemical and Biophysical Research Communications 96, 1688. Hynes, M. J. (1972). Journal of Bacteriology 111, 717. Hynes, M. J. (1974a). Journal of Bacteriology 120, 1116. Hynes, M. J. (1974b). Molecular and General Genetics 132, 147. Hynes, M. J. (1975a). Australian Journal of Biological Sciences 28, 301. Hynes, M. J. (1975b). Nature, London 253, 210. Hynes, M. J. (1978). Molecular and General Genetics 166, 31. Hynes, M. J., and Pateman, J. A. J. (1970). Molecular and General Genetics 108, 97. Jacobs, E., Dubois, E., Hennaut, C., and Wiame, J.-M. (1981). Current Genetics 4, 13. Jacobs, E., Dubois, E., and Wiame, J.-M. (1982). Archives Internationales de Physiologie et de Biochimie 90, B198. Jacobs, P., Jauniaux, J. C., and Grenson, M. (1982).Journal of Molecular Biology 139, 691. Jacoby, G. A., and Gorini, L. (1969). Journal of Molecular Biology 39, 73. Jauniaux, J. C., Urrestarazu, L. A., and Wiame, J.-M. (1978). Journal ofBacteriology 133, 1096. Jauniaux, J. C . , Dubois, E., Vissers, S., Crabeel, M., and Wiame, J.-M. (1982). The EMBO Journal 1, 1125. Joins, C. R., and Grenson, M. (1969). Archives Internationales de Physiologie et de Biochimie 77, 154. Jones, E. W., and Fink, G. R. (1982). In “The Molecular Biology of the Yeast Saccharomyces: Metabolism and Gene Expression” (J. N . Strathern, E. W. Jones, and J. R. Broach, eds.). Cold Spring Harbor Laboratory, New York. Jones, G. E. (1973). Molecular and General Genetics 125, 9. Jones, G. E. (1977). Journal of Bacteriology 130, 128. Jones, G. E. (1978). Journal of Bacteriology 134, 200. Jones, S. A., Arst, H. N., Jr., and MacDonald, D. W. (1981). Current Genetics 3, 49. Kang, L., Keeler, M. L., Dunlop, P. C., and Roon, R. J. (1982). Journal of Bacteriology 151, 29. Ketchum, P. A., Zeeb, D. B., and Owens, M. S. (1977). Federation ofEuropean Microbiological Societies Microbiology Letters 2, 3 1.

a4

JEAN-MARIE WIAME ET AL.

Kinghorn, J. R., and Pateman, J. A. (1973). Journal of General Microbiology 78, 39. Kinghorn, J. R., and Pateman, J. A. (1975a). Molecular and General Genetics 140, 137. Kinghorn, J. R., and Pateman, J. A. (1975b). Journal of General Microbiology 86, 294. Kinnaird, J. H., Keighren, M. A., Kinsey, J. A., Eaton, M., and Fincham, J. R. S. (1982). Gene 20, 387. Klimczuk, J., and Weglenski, P. (1974). Acta Microbiologica Polonica, Series A 6, 93. Kolhaw, G., Dragert, W., and Holzer, H. (1965). Biochemische Zeitschrift 341, 224. Kung, H.-F., and Wagner, C. (1969). Journal of Biological Chemistry 244,4136. Kuznar, J., Schwencke, J., and Magafia-Schwencke, N. (1973). Biochimica et Biophysica Acta 318, 273. Lacroute, F. (1966). Thesis, University of Pans. Lacroute, F. (1968). Journal of Bacteriology 95, 824. Lacroute, F., Pierard, A., Grenson, M., and Wiame, J.-M. (1965). Journal of General Microbiology 40, 127. Lara, M., Blanco, L., Champomanes, M., Calva, E., Palacios, R., and Mora, J. (1982). Journal of Bacteriology 150, 105. Larimore, F. S., Kuisk, I., Korkowski, P. M., and Roon, R. J. (1980). Archives of Biochemistry and Biophysics 204, 234. Lasko, P. F., and Brandriss, M. C. (1981). Journal of Bacteriology 148, 241. Legrain, C., Vissers, S., Dubois, E., Legrain, M., and Wiame, J.-M. (1982). European Journal of Biochemistry 123, 61 1. Lemoine, Y., Dubois, E., and Wiame, J.-M. (1978). Molecular and General Genetics 166, 251. Lewis, C. M., and Fincham, J. R. S. (1970a). Genetical Research 16, 151. Lewis, C. M., and Fincham, J. R. S. (1970b). Journal of Bacteriology 103, 5 5 . Lienhard, G. E. (1983). Trends in Biochemical Sciences 8, 125. Lodder, J. (1971). “The Yeasts: A taxonomic study”. 2nd Ed. North-Holland Publishing Company, Amsterdam, London. MacDonald, D. W. (1982). Current Genetics 6, 203. Magafia-Schwencke, N., and Schwencke, J. (1969). Biochimica et Biophysica Acta 173, 313. Magafia-Schwencke, N., Kuznar, J., and Schwencke, J. (1973). Biochimica et Biophysica Acta 318, 281. Magasanik, B. (1961). Cold Spring Harbor Symposia on Quantitative Biology 26, 249. Magasanik, B. (1963). I n “RCgulation chez les Microorganismes”, pp. 179-183. Editions du Centre National de la Recherche Scientifique, Paris, 1965. Magasanik, B. (1970). In “The Lactose Operon” (J. Beckwith and D. Zipser, eds.), p. 189. Cold Spring Harbor Laboratory, New York. Magasanik, B. (1982). Annual Reuiews of Genetics 16, 135. Marzluf, G. A. (1981). Microbiological Reviews 45, 437. Mazon, J. J. (1978). Journal of Bacteriology 133, 780. Meers, J. D., Tempest, D. W., and Brown, C. M. (1970). Journal of General Microbiology 64, 187.

NITROGEN CATABOLITE REPRESSION IN FUNGI

85

Merrick, M., Hill, S., Hennecke, H., Hahn, M., Dixon, R., and Kennedy, C. (1982). Molecular and General Genetics 185, 75. Messenguy, F. (1976). Journal of Bacteriology l28,49. Messenguy, F., and Wiame, J.-M. (1969). FEBS Letters 3, 47. Middlehoven, W. J. (1964). Biochimica et Biophysica Acta 93, 650. Middlehoven, W. J. (1968). Biochimica et Biophysica Acta 156, 440. Middlehoven, W. J., Van Eijk, J., Van Renesse, R., and Blijham, J. M. (1978). Antonie van Leeuwenhoek 44, 3 1 1 . Monod, J. (1949). Annual Review of Microbiology 3, 371. Monod, J. (1950). Annales de l’lnstitut Pasteur 79, 390. Monod, J., Pappenheimer, A. M., Jr., and Cohen-Bazire, G. (1952). Biochimica et Biophysica Acta 9, 648. North, M. J. (1982). Microbiological Reviews 46, 308. Novick, A., and Szilard, L. (1950). Proceedings of the National Academy of Sciences of the United Stares of America 36, 708. Patel, V. B., Schweizer, M., Dykstra, C. C., Kushner, S. R., and Giles, N. H. (1981). Proceedings of the National Academy of Sciences of the United States of America 78, 5783. Pateman, J. A. (1969). Biochemical Journal 115, 769. Pateman, J. A., and Kinghorn, J. R. (1975). Heredity 35, 427. Pateman, J. A., and Kinghorn, J. R. (1977). I n “Genetics and Physiology of Aspergillus” (J. E. Smith and J. A. Pateman, eds.), pp. 203-241. Academic Press, London. Pateman, J. A., Rever, B. M., and Cove, D. J. (1967). Biochemical Journal 104, 103. Pateman, J. A., Kinghorn, J. R., Dunn, E., and Forbes, E. (1973). Journal of Bacteriology 114, 943. Pateman, J. A., Dunn, E., Kinghorn, J. R., and Forbes, E. C. (1974). Molecular and General Genetics 133, 225. Penninckx, M., and Wiame, J.-M. (1976). Journal of Molecular Biology 104,819. Perkins, D. D., Radford, A., Newmeyer, D., and Bjorkman, M. (1982). Microbiological Reviews, 46, 426. Perlman, R., de Crombrugghe, B., and Pastan, I. (1969). Nature, London 223, 810. Polkinghome, M., and Hynes, M. J. (1975). Genetical Research 25, 119. Polkinghorne, M., and Hynes, M. J. (1 982). Journal of Bacteriology 149, 93 1. Polya, G. M., Brownlee, A. G., and Hynes, M. J. (1975). Journal ofBacteriology 124, 693. Premakumar, R., Sorger, G. J., and Gooden, D. (1978). Biochimica et Biophysica Acta 519, 275. Premakumar, R., Sorger, G. J., and Gooden, D. (1979). Journal of Bacteriology 137, 1119. Premakumar, R., Sorger, G. J., and Gooden, D. (1980). Journal of Bacteriology 144, 542. Ramos, F., and Wiame, J.-M. (1979). European Journal of Biochemistry 94,409. Rand, K. N. (1978). Ph.D. Thesis, University of Cambridge. Rand, K. N., and Arst, H. N., Jr. (1977). Molecular and General Genetics 155, 67. Rand, K . N., and Arst, H. N., Jr. (1978). Nature, London 272, 732.

86

JEAN-MARIE WIAME H A L .

Reinert, W. R., and Marzluf, G. A. (1975). Archives of Biochemistry and Biophysics 166, 565. Reitzer, L. J., and Magasanik, B. (1983). Proceedings of the National Academy of Sciences of the United States of America 80, 5554. Roon, R. J., and Even, H. L. (1973). Journal of Bacteriology 116, 367. Roon, R. J., and Levenberg, B. (1968). Journal ofBiologica1 Chemistry 243,5213. Roon, R. J., and Levenberg, B. (1970). Journal ofBiological Chemistry 245,4593. Roon, R. J., Even, H. L., and Larimore, F. (1974). Journal of Bacteriology 118, 89. Roon, R. J., Even, H. L., Dunlop, P., and Larimore, F. (1975a). Journal of Bacteriology 122, 502. Roon, R. J., Larimore, F., and Levy, J. S. (1975b). Journal of Bacteriology 124, 325. Roon, R. J., Levy, J. S., and Larimore, F. (1977). Journal of Biological Chemistry 252, 3599. Rose, A. H. (1975). In “Methods in Cell Biology. Vol. XII: Yeast Cells’’ (D. M. Prescott, ed.), pp. 1-15. Academic Press, New York. Rowe, W. B., Ronzio, R. A., Wellner, V. P., and Meister, A. (1970). Methods in Enzymology 17A, 900. Rytka, J. (1975). Journal of Bacteriology 121, 562. Sanchez, F., Campomanes, M., Quinto, C., Hansberg, W., Mora, J., and Palacios, R. (1978). Journal of Bacteriology 136, 880. Scazzocchio, C., and Darlington, A. J. (1968). Biochimica et Biophysica Acta 166,557. Scazzochio, C., and Arst, H. N., Jr. (1978). Nature, London 274, 177. Schwencke, J., and Magaiia-Schwencke, N. (1969). Biochimica e f Biophysica Acta 173, 302. Shaffer, P. M., and Arst, H. N., Jr. (1984). Molecular andGeneralGenetics, 198, 139. Shapiro, B. M., Kingdom, H. S., and Stadtman, E. R. (1967). Proceedings of the National Academy of Sciences of the United States of America 58, 642. Sims, A. P., and Ferguson, A. R. (1974). Journal of General Microbiology 80, 143. Sims, A. P., Toone, J., and Box, V. (1974a). Journal of General Microbiology 80, 485. Sims, A. P., Toone, J., and Box, V. (1974b). Journal of General Microbiology 84, 149. Sorger, G. J., and Giles, N. H. (1965). Genetics 52, 777. Spathas, D. H., Pateman, J. A., and Clutterbuck, A. J. (1982). Journal of General Microbiology 128, 557. Spathas, D. H., Clutterbuck, A. J., and Pateman, J. A. (1983). Federation of European Microbiological Societies Microbiology Letters 17, 345. Stadtman, E. R. (1970). In “The Enzymes” (P. D. Boyer, ed.), 3rd Ed., Vol. I . , p. 397. Academic Press, New York. Stadtman, E. R., and Ginsburg, A. (1974). In “The Enzymes’’ (P. D. Boyer, ed.), 3rd Ed. Vol. 10, pp. 755-807. Academic Press, New York. Sumrada, R., and Cooper, T. G. (1974). Journal of Bacteriology 117, 1240. Sumrada, R., Gorski, M., and Cooper, T. G. (1976). Journal ofBacteriology 125, 1048.

NITROGEN CATABOLITE REPRESSION IN FUNGI

a7

Sumrada, R., Zacharscki, C. A., Turoscy, V., and Cooper, T. G. (1978). Journal of Bacteriology 135, 498. Surdin, Y., Sly, W., Sire, J., Bordes, A. M., and de Robichon-Sjulmajster, H. (1965). Biochimica et Biophysica Acta 107, 546. Tempest, D. W., Meers, J. D., and Brown, C. M. (1970). Journal of General Microbiology 64, 187. Tollervey, D. W. (1981). Ph.D. Thesis, University of Cambridge. Tollervey, D. W., and Arst, H. N., Jr. (1981). Current Genetics 4, 63. Tollervey, D. W., and Arst, H. N., Jr. (1982). Current Genetics 6, 79. Tomsett, A. B., and Cove, D. J. (1979). Genetical Research 34, 19. Tomsett, A. B., Dunn-Coleman, N. S., and Garrett, R. H. (1981). Molecular and General Genetics 182, 229. Turoscy, V., and Cooper, T. G. (1979). Journal of Bacteriology 140, 971. Turoscy, V., and Cooper, T. G. (1982). Journal of Bacteriology 151, 1237. Tyler, B., Loomis, W. F., and Magasanik, B. (1967). Journal of Bacteriology 94, 2001. Urrestarazu, L. A., Vissers, S., and Wiame, J.-M. (1977). European Journal of Biochemistry, 79, 473. Vaca, G., and Mora, J. (1977). Journal of Bacteriology 131, 719. van de Poll, K. W. (1973). FEBS Letters 32, 265. Vissers, S., Urrestarazu, L. A., Jauniaux, J. C., and Wiame, J.-M. (1982). Journal of General Microbiology 128, 1235. Vogels, G. D., and van der Drift, C. (1976). Bacteriological Reviews 40, 403. Watson, J. D. (1975). “The Molecular Biology of the Gene”, p. 507. W. A. Benjamin, Inc., New York. Weiss, R. L., and Davis, R. H. (1973). Journal ofBiologica1 Chemistry 248,5403. Whitney, P. A., and Cooper, T. G. (1970). Biochemical and Biophysical Research Communications 40, 814. Whitney, P. A., and Cooper, T. G. (1972). Journal of Biological Chemistry 247, 1349. Whitney, P. A., and Magasanik, B. (1973). Journal of Biological Chemistry 248, 6197. Whitney, P. A., Cooper, T. G., and Magasanik, B. (1973). Journal of Biological Chemistry 248, 6203. Wiame, J.-M. (1964). In “Comparative Biochemistry of Arginine and Derivatives” (G. E. W. Wolstenholme and M. P. Cameron, eds.), Ciba Foundation Study Group, pp. 16-19. Churchill, London. Wiame, J.-M. (1971). In “Current Topics in Cellular Regulation” (B. L. Horecker and E. R. Stadtman, eds.), Vol. 4, pp. 1-38. Academic Press, New York. Wickerham, L. J. (1946). Journal of Bacteriology 52, 293. Wiemken, A., and Durr, M. (1974). Archives of Microbiology 101, 45. Wolf, D. H. (1982). Trends in Biochemical Sciences 7 , 35. Woodward, J. R., and Kornberg, H. L. (1981). Biochemical Journal 196, 531. Wulff, K., Mecke, D., and Holzer, H. (1967). Biochemical and Biophysical Research Communications 28, 740.

88 Note Added in Proof DeBusk and Ogilvie (1984) have shown that the MS5 mutant of Neurospora crassu probably carries an allele of nmr-1 and that these mutations do not affect glutamine transport. DunnColeman er al. (1984) have described a methylammonium resistance mutation in N . crassa which is likely to reduce uptake of ammonium and methylammonium. The areB-405 mutation of Aspergillus nidulans has now been shown t o be associated with (and is probably identical to) a short paracentric inversion (M. X. Caddick and H. N. Arst, Jr., unpublished results). The ureA gene of A . niduluns has now been cloned (M. X. Caddick, A. G. Brownlee, L. H. Taylor, R. I. Johnson, and H. N. Arst, Jr., unpublished results). DeBusk, R. M., and Ogilvie, S. (1984).Journal of Bacteriology 160, 656. Dunn-Coleman, N. S., Nassiff, M. D., and Garrett, R. H. (1984). Current Genetics 8, 423.

Sexual Agglutination in Chlamydomonads H. VAN DEN ENDE Department of Plant Physiology, University of Amsterdam, Amsterdam, The Netherlands I. Introduction. . . . . . . . . 11. The mating system . . . . . 111. The mating process . . . . . . . A . Gametogenesis . . . . . . . . B. Sexual agglutination. . . . . . C. Mating structure activation . . . . . . D. Cell wall release . . . . . E. The flagellar tip . . . . . . IV. The specificity of sexual agglutination . . . . . A . The flagellar surface . . . . . . . B. The isoagglutinins . . . . . . . . C. The mt- agglutination factor in Chlrrmydomonas rugomrtos D. The mt+ agglutination factor in Chlumydoinoncis reinliardtii V. Dynamics of sexual agglutination . . . . . . A. Mechanism of sexual adhesion . . . . . . B . Receptor inactivation . . . . . . C. The signalling action of sexual agglutination . . VI. Concluding remarks . . . . . . . . References . . . . . . . . . .

.

.

.

.

. . . .

. . . .

.

.

. . .

. . .

. . . . . . . .

.

.

8 9 9 0 9 1 9 1 9 2 9 4 9 6 9 8 . 102 . 102 . 103

. .

. .

. 104 . 110

.

.

. 111

. ,

. .

,

.

.

. 116

. 111

.

.

,

.

.

.

112

119 119

I. Introduction

The way cells communicate with each other, interact, and co-ordinate their activities is a fundamental issue in developmental biology. This is particularly true in multicellular organisms, where the controlled division of labour between cells determines form and function of the whole organism as well as its parts. Also, in unicellular systems, intercellular interacADVANCES IN MICROBIAL PHYSIOLOGY VOL 26

Copwighi Q 1985 by Ac.odumk. Press. i n ( . ( l ~ i n d o f i Lld. l A / / righrs ~$rcpmudia-liunirr ail? Jiirni n w r u c d

90

H. VAN DEN ENDE

tions are important, such as in establishing cell communities, in predation, and in sexual reproduction. The last type of communication is of particular interest, because generally it surpasses others in specificity. The central event in sexual interaction, the fusion of two cells, is a dramatic event in a cell’s life, especially when there is only one nucleus per cell. The interacting cell will lose its individuality and must be sure that after fusion life still has some perspective, which of course depends to a large extent on the partner. So it is logical that elaborate mechanisms have evolved that confer a high degree of specificity on the fusion process. In addition, cells fuse only under special conditions. Often cells differentiate into a stage specialized for fusion (“gametogenesis”) and, generally, they are conditioned for fusion by a shorter or longer courtship. Essentially, there seem to be two ways by which unicellular organisms communicate with each other in anticipation of sexual fusion: (1) by means of diffusible hormones or pheromones; (2) by cellular contact, in which molecules, localized at the cell surface, play a role in generating and transmitting messages. To illustrate the latter means of communication, the genus Chlamydomonas may serve us well, as it performs sexual interaction primarily via its flagellar surface, and hormonal interactions apparently are absent. The study of sex in Chlamydomonas spp. has a long history (e.g., Kniep, 1928; Hartmann, 1956) and has been reviewed in recent times (Goodenough, 1977; van den Ende, 1981; Goodenough and Thorner, 1983). This article will describe the latest developments that have increased our understanding of the system and hopefully will stimulate further efforts in this research field. These developments largely concern two species, Chlamydomonas reinhardtii and Chlamydomonas eugametos (together with its sub-species Chlamydomonas moewusii; Gowans, 1976). The two species differ only in minor respects and, on the whole, the results obtained present a consistent picture of their sexual physiology. Presumably, they can be extrapolated to other related chlamydomonads and also to isogamous colonial relatives, such as Pandorina (Rayburn, 1974) and Gonium (Stein, 1966).

11. The Mating System

Chlamydomonas reinhardtii and Chlamydomonas eugametos are heterothallic and isogamous, which means that there is a stable inheritance of mating type ( m f +and mt-) and that there are no morphological differ-

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

91

ences between mt+ and mt- cells. Sexual phenotype is determined by a single nuclear gene, the mt locus, which acts as a “master-locus,’’ affecting the expression of selected genes at various places in the genome (Forest et al., 1978; Hwang et al., 1981; for review see Wiese, 1976). Many other Chlamydomonas species are homothallic, both mating types arising within a clone, presumably as the result of regular interconversions of the plus and minus version of the mt locus. One of these studied recently, is Chlamydomonas monoica (Vanwinkle-Swift and Burrascano, 1983; Vanwinkle-Swift and Aubert, 1983). Other species, less frequently observed, are anisogamous (e.g., Chlamydomonas capensis; Shyam and Sarma, 1976) or oogamous (Chlamydomonas zimbabwiensis; Heimke and Starr, 1979).

111. The Mating Process A. GAMETOGENESIS

Chlamydomonas reinhardtii and Chlamydomonas eugametos exist as small ovoid unicells, with two flagella arising from the anterior end, which protrude through separate pores into the surrounding cell wall. Under suitable conditions, the cells are transformed into mating-competent gametes, which are morphologically indistinguishable from sexually inactive vegetative cells. In C . reinhardtii, this sexual differentiation is triggered by nitrogen starvation (Sager and Granick, 1954). The cells leave the cycle of mitotic growth and enter a non-dividing stage in which they are competent to mate. This is most obvious in liquid cultures in which nitrogen starvation can be imposed instantaneously, but occurs probably in a more gradual way in cultures grown on a solid medium (Kates and Jones, 1964; Martin and Goodenough, 1975). Optimal gametogenesis is observed when the cells are transferred to nitrogen-free medium in the mid-GI phase. It is probably always accompanied by a mitotic division (Kates and Jones, 1964; Schmeisser et al., 1973). When the cells are returned to nitrogen-containing medium, they dedifferentiate into vegetative cells after 12-15 hours. The only two changes that becomes evident during gametogenesis (apart from apparent stress phenomena, such as an alteration in the chloroplast fine structure) are a subtle change in the properties of the flagellar surface and in the ultrastructure of the “mating structure”, as will be described below. In C. eugametos, gametogenesis is less well documented. It is more or

92

H. VAN DEN ENDE

less similar to C. reinhardtii in its susceptibility to nitrogen limitation (Bernstein and Jahn, 1955), but this sensitivity seems also to be dependent on the concentrations of other nutrients and the imposed light regime and intensity (Trainor, 1959, 1975). In recent work with this species, gametes were most conveniently obtained by flooding agar cultures with water, whereas vegetative, that is, non-mating cells, were obtained by flooding with a dilute solution of an ammonium salt (e.g., Musgrave et al., 1981). In no instance has there been any indication that sexual differentiation is induced or modulated by hormonal influences from cells of the opposite mating type, as has been found in the oogamous Volvocales, such as Volvox carteri (Starr and Jaenicke, 1974).

B . SEXUAL AGGLUTINATION

When suspensions of sexually active mt+ and mt- cells are mixed, they adhere to each other via their flagella in a species- and sex-specific fashion. In dilute suspensions, this leads to clumps of a few cells only, but in dense populations large aggregates of many cells are formed. Apparently, the initial contacts are established by random collisions, and the resulting zygote yields can be predicted from the ratio between the numbers of mt+ and mt- cells mixed and their respective mating competencies (Lewin, 1952a). There is no evidence that mt+ cells attract mt- cells, or vice versa. A characteristic feature of agglutinating cells is a violent twitching of the flagella, which gives the clumps a dancing or vibrating appearance. We will return to this behaviour in a later section. Within a clump of cells, a continuous dissociation and re-association or “sorting-out’’ activity leads to the formation of pairs of cells of opposite mating type, in which the flagella are associated by their tips only, as is seen in C . reinhardtii. In C . eugametos, the flagella become aligned along their whole length, but with the tips adhered to each other (Mesland, 1976). When both flagella of a cell pair have become connected, a specialized zone of the plasma membrane at the anterior end, between the flagellar bases, is “activated” in each cell and forms a fertilization tubule (Martin and Goodenough, 1975), also called a papilla (Mesland, 1976). By means of these organelles the cells fuse partially and form a plasma connection. At this stage, there appears a difference between C. reinhardtii and C . eugametos. In the former species, the cell wall is shed during flagellar adhesion and the naked cells rapidly merge completely. The resulting zygote takes a regularly rounded shape in which the bases of the two

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

93

flagellar pairs approach each other and the flagellar roots become connected (Friedmann et al., 1968). The four flagella lose their adhesiveness and become motile again in a co-ordinated fashion, before the gamete nuclei have fused. This situation lasts for several hours before the zygote wall is deposited (Starling and Randall, 1971). In C. eugametos, however, the cell wall remains intact, Fusion of the papillae (which penetrate the cell wall in both partners at the anterior end) results in a tandem of cells, called vis-a-vis pair, which has a life time of several hours before complete fusion of the protoplasts and karyogamy take place (Lewin, 1954; Musgrave et al., 1983). As in C. reinhardtii, the flagella lose their specific adhesiveness in a few minutes after the plasma bridge has been established. In vis-a-vis pairs, however, only the flagella of the mt+ cell resume the beating, whereas the flagella of the mt- cell remain motionless and rarely exhibit more than a slight trembling activity (Lewin, 1952a). The fusion of gametes is followed by the formation of a “primary zygote membrane” (Lewin, 1952b). Within this membrane, the thick and ornamented secondary zygospore wall is formed. The primary zygote membrane is normally discarded. In C. eugametos, zygote formation only occurs in the light. In continuous darkness, vis-a-vis pairs remain as they are (Lewin, 1954). As will be shown in the following sections, there is a strong causal relationship between sexual agglutination and the ensuing ultrastructural and physiological changes in the participating cells. Also, the establishment of a plasma connection and the loss of adhesiveness of the agglutinating flagella appear to be causally linked events. When fusion was somehow inhibited (Mesland and van den Ende, 1978a; Wiese and Wiese, 1978; Goodenough and Weiss, 1975), the flagella remained associated for many hours. The flagellar interaction between mating cells of Chlamydomonas displays a number of remarkable features: (1) it is extremely speciJc, mt+ cells only engaging with mt- cells within a species; (2) it is highly dynamic, as the mutual adhesiveness of the flagella does not prevent rearrangements of the cells into cell pairs; (3) it has a signalling action, since flagellar association not only serves to establish the correct position of the cells which is required for papillar fusion, but also induces a number of changes in the cell which make it ready for cell fusion; and (4) it is regulated, because after bridge formation flagellar adhesiveness disappears completely in a very short time. These aspects of sexual agglutination will be discussed in more detail in the following sections. But first we will consider in more detail the cellular responses to sexual agglutination.

94

H. VAN DEN ENDE C. MATING STRUCTURE ACTIVATION

Several authors have described ultrastructural aspects of the mating structure (also known as the plasma papilla; Triemer and Brown, 1975a). It consists of a differentiated, slightly protruded cortical area between the flagellar bases. In C . reinhardtii mt+, this area contains two electrondense zones of cytoplasmic material. One of these is a thin zone directly adjacent to the plasma membrane (the “membrane zone”) and the other is a more complex zone in the underlying cytoplasm (the “doublet zone”). The plasma membrane in this region is covered with fuzzy, carbohydrate-containing material (“fringe”), as shown in tannic acid/glutaraldehyde-fixed preparations (Martin and Goodenough, 1975; Goodenough et al., 1982). Figure 1 shows schematically what happens when the mating structure is activated as the result of sexual agglutination: the membrane zone separates from the cone-shaped doublet zone to form a bud which rapidly elongates into a 2 pm-long tubule. The doublet zone remains at the base. From there a large number of parallel arrays of microfilaments seem to radiate into the tubule. These microfilaments may have a supporting function, although extensive cross-linking was not observed. When cells made permeable with saponin were incubated with myosin subfragment-1 the microfilaments were decorated, indicating that they are composed of actin. The polarity of the myosin arrowheads on the actin filaments was uniform throughout the fertilization tubule, as all arrowheads point away from the tip of the tubule and toward the doublet zone at the base. Detmers et al. (1983), reporting these results, suggested that outgrowth of the fertilization tubule was accompanied by the directed polymerization of actin filaments from nucleation sites located in the doublet zone, as shown in Fig. 1. Cytochalasin D completely prevented the formation of well-organized actin filaments, suppressed fertilization tubule outgrowth, and lowered mating efficiency. The corresponding structure in C . reinhardtii mt- is less elaborate. It is also a specialized region of the cell surface between the flagella, covered with fuzzy material, but it contains only one electron-dense cytoplasmic zone and is considerably smaller. On activation, only a relatively small bud develops, without microfilaments (Goodenough et al., 1982). In C . eugametos, the activation of gametes probably takes a similar course. Brown e f al. (1968) and Triemer and Brown (1975a,b), studying the related species C . rnoewusii, found that during gametogenesis electron-dense material was deposited below the surface area adjacent to the basal bodies (the “plasma papilla”). During flagellar adhesion, a local hydrolysis of the cell wall allowed a small tubular cytoplasmic extension

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

(a)

Unactivated

(b)

Budding

95

( c ) Elongating

FIG. 1. Model for elongation of the fertilization tubule in C . reinhardtii, mt+. (a) In the unactivated mt+ mating structure an electron-dense membrane zone (mz) overlies a coneshaped doublet zone (dz). The exterior surface of the mating structure is provided with “fringe” (0 which appears to mediate recognition of and bonding to the mt- mating structure prior to fusion. (b) Following receipt of mating signals the membrane zone separates from the central portion of the doublet zone, and a bud is formed. (c) Nucleation of actin polymerization then occurs at the doublet zone, assuring the uniform polarity of the filaments, which grow by monomer addition to the barbed end (9, monomers adding in this manner). When elongation is complete, the barbed ends of the filaments are embedded in the membrane zone, which may help promote stability of the filaments. From Detmers et a / . (1983).

to become exposed in both mating types. By means of these extended papillae, the cytoplasmic bridge was established. In the resulting vis-a-vis pairs, the two nuclei were connected by endoplasmic reticulum which, after several hours, shortened and ultimately brought the nuclei into contact. Little is known about the fusion process between the two activated mating structures. The fuzzy coat covering the plasma membrane is probably essential for fusion and carries recognition determinants (see Goodenough, 1977; Mesland and van den Ende, 1978a). The i m p 1 mutant of C. reinhardtii m t + , which agglutinated normally and produced a fertil-

96

H. VAN DEN ENDE

ization tubule, but which was unable to fuse, is lacking this material (Goodenough et al., 1982). Melkonian (1980) has compared the gametic fusion of Chiamydomonas spp. with that of other green algae. Undoubtedly, there is a close connection between sexual agglutination and mating structure activation. In C. reinhardtii as well as in C . eugametos, a cell extends its fertilization tubule only when its flagella are in intimate contact with flagella of the partner cell. Activation is also achieved when cells are treated with isolated flagella or flagellar membranes derived from gametes of the opposite mating type (Mesland and van den Ende, 1978b; Weiss et al., 1977; see Table 1). Non-agglutinative mutants do not activate their mating structure (Goodenough et al., 1976). Notwithstanding the strong coupling of flagellar adhesion and activation of the cellular mating structure, both processes can be dissociated to some extent. In cells with regenerating flagella, agglutinability reappears when the flagella are short stumps of less than 3 pm, compared to the full flagellar length of 12 pm, but mating ability is only achieved after the flagella have attained their full length (Mesland, 1977; Ray et al., 1978; Pijst et al., 1983). We will return to this problem in Section V.

D . CELL W A L L RELEASE

As mentioned above, cells of C . reinhardtii rapidly shed their cell wall during sexual agglutination, and become transitory protoplasts which redevelop a cell wall after fusion. This is typical for several species (such as the ominous C . gymnogama; Miller et al., 1974). Other species such as C . eugametos and C . monoica, with a long-lived vis-8-vis pair stage following conjugation, discard their cell wall only after the secondary zygote wall has been formed (Musgrave et al., 1983). Claes (1971) showed that cell-wall shedding in C . reinhardtii is due to the secretion of an “autolysin” by the cells involved. When cell-free medium of a mixture of mt+ and mt- gametes engaged in sexual agglutination is added to single cells (vegetative cells or gametes) their cell walls are dissolved. The protoplasts slip out through a pore produced in the cell wall, round up, and swim away. The discarded cell wall becomes completely dissolved within 10 minutes (Schlosser et al., 1976). The autolytic activity is released by gametes of both mating types, since it is found equally in the medium of mt+ or mt- gametes treated with flagella of the opposite mating type (Claes, 1971; Kaska and Gibor, 1982; Snell, 1982). Its action is restricted to cell walls of C. reinhardtii and some close relatives (Schlosser et ai., 1976). It is different in specificity from the cell-wall hydrolysing activity which appears when daughter cells escape from the

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

97

sporangial wall (“hatching”), in that the action of the latter is restricted to sporangial walls only. However, there are also similarities in the nature of the enzyme involved (Schlosser, 1976). The enzyme responsible for hydrolysis of cell walls during mating has been characterized to some extent. It is precipitated by ammonium sulphate, it is non-dialysable and temperature labile, and it is inactivated by HgCI2 and EDTA. Inactivation by EDTA can be overcome by increasing the concentration of Ca2+or Mg2+(Schlosser, 1981; Snell, 1982). It is a neutral tryptic serine protease. The purified enzyme has a molecular weight of 37,000-40,000, according to Jaenicke and Waffenschmidt (1981), and consists of a single peptide chain. It hydrolyses proteins such as casein, haemoglobin, or insulin at basic amino acid residues. Various serine-specific inhibitors are effective, indicating that serine is part of the active centre. Its action is inhibited by concanavalin A (Claes, 1975) and can be restored by methyl-a-D-mannoside. Apparently concanavalin A exerts its action by binding to glucose and/or mannose residues. Whether these are part of the active site of the enzyme, or rather of the site of attack on the substrate, has not been reported. The mechanism ofautolysin reIease has hardly been investigated. It is a very rapid process. Snell (1982) observed that autolysin was secreted within 1-2 minutes after mixing mt+ and mr- gametes. Thereafter no detectable amounts of the enzyme were secreted. Claes (1977) reported that, when C . reinhardtii cells were homogenized in a French press, a sedimentable fraction was obtained which contained the autolysin in an inactive form. It could be activated by sonication. Jaenicke and Waffenschmidt (1981) observed that, together with the autolysin, a high-molecular-weight inhibitor was secreted. Thus, the enzyme might be localized in vesicles as zymogen and become released in active form in the periplasmic space by exocytosis, very similar to that which happens with chitin synthase in fungal hyphae (Cabib et al., 1973; Gooday and Trinci, 1980). Nevertheless, more experimentation is required in order to gain a better insight into these phenomena. The release of the cell-wall hydrolysing enzyme is just as closely connected to sexual agglutination as is mating structure activation. It also can be elicited by treating gametes (not vegetative cells) with flagella or flagellar membranes derived from cells of the opposite mating type (Table 1). In regenerating flagella, the capacity to induce cell-wall release is not correlated with the appearance of agglutinability (Solter and Gibor, 1977; Kaska and Gibor, 1982). Gametes that have lost mating competency by trypsin treatment regain their agglutinability within about 45 minutes of treatment, but the ability to release the cell wall reappears only after several hours. The recovery of agglutinability is inhibited by cycloheximide and

98

H. VAN DEN ENDE

not by actinomycin D, but both inhibitors block the acquisition of the capacity to hydrolyse the cell wall in response to sexual agglutination (Solter and Gibor, 1978).

E. T H E FLAGELLAR TIP

A discussion about the role of the flagellar tip in sexual agglutination is inadequate without reference to some typical dynamic properties of the flagellar membrane. These were recently reviewed by Bloodgood (1982). One of the most spectacular is the adhesiveness for, and rapid translocation of, polystyrene spheres (0.35 pm diameter) called “flagellar surface motility” (Bloodgood, 1977, 1980; Bloodgood et al., 1979; Bloodgood and May, 1982). This motility is in both longitudinal directions and is of a “saltatory” nature: particles adhered to the flagellar surface start and stop and remain at rest for various periods of time; they change directions (180”) abruptly, accelerate and decelerate very rapidly, and attain a uniform velocity of about 2 pm sec-’. Different markers on the same flagellum move independently. The motility is observed in vegetative cells and gametes alike, but only at their flagellar surface; it is not exhibited by the plasma membrane of the cell body. There is no relationship with the axoneme-driven bending of the flagellum, since a wide range of paralysed mutants of C. reinhardtii all display normal surface motility (in fact, one needs a paralysed mutant to observe flagellar surface motility accurately). Inhibition of glycoprotein synthesis by cycloheximide or tunicamycin results in a gradual loss of both adhesiveness and surface motility without any change in flagellar length. This suggests that there may be a relatively rapid turnover of one or more surface proteins involved in these phenomena. Also, proteolytic modification of the flagellar surface leads to the disappearance of adhesiveness and motility. From these studies evidence was obtained that a high-molecular-weight glycoprotein (MW 350,000) might serve as motility-coupled surface receptor. The mechanism of its propagation is unknown. Also, the physiological function of surface motility remains to be revealed. Bloodgood (1982) has suggested that it may be associated with the ability of chlamydomonads to “creep” along a solid surface (Lewin, 1952a), or with the ability to translocate the flagellar contact site during sexual agglutination (see below). Finally, this motility might have something to do with the distribution of flagellar components during assembly or normal turnover of flagellar components. A second dynamic phenomenon exhibited by flagellar membranes is called “tipping” and is specific for gametes. It bears resemblance to the energy-dependent redistribution of surface receptors called “capping,”

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

99

which has been observed in a variety of cells (e.g., Taylor et al., 1971). When gametes of C. reinhardtii were exposed to antisera raised against the flagellar surface, they responded by accumulating antibody bound to the flagellar tips. As a consequence, the gametes agglutinated but by their flagellar tips only (Goodenough and Jurivich, 1978). An explanation for this might be that flagella have a local high density of antigens at their tips. However, the authors favour the view that bound antibody is transported along the flagellar surface and is concentrated at the tip. One observation supporting this is that glutaraldehyde-fixed gametes treated with antibody agglutinated along their entire flagellar lengths and not just at their tips. Another one is that gametic flagella, detached from cells and then presented with antiserum, made agglutinating contacts at all positions, showing no preference for the tips. Clearly, the tipping response needs an intact living cell. The fact that the postulated redistribution of antigens to the flagellar tips was displayed only by gametes, suggests that it has some relevance to the sexual process. Indeed, antibody-mediated tipping is always accompanied by mating structure activation, even in non-agglutinable mutants (Goodenough and Jurivich, 1978). In this respect Mesland and van den Ende (1978b) found some dissimilar results in C. eugametos. Although mt+ gametes of this species were seen to accumulate material at the flagellar tips when exposed to flagellar membrane preparations of the opposite mating type, and consequently accumulated tip-to-tip, the mt- gametes adhered in a random fashion in a wide variety of conditions. A strong correlation between tip-to-tip agglutination and mating structure activation was only found in mt+ gametes of this species. Goodenough et al. (1980) showed in paralysed mutants of C. reinhardtii also that the contact sites between flagella of living cells of opposite mating type move gradually to the tips where they become stationary and establish the typical tip-to-tip configuration prior to fusion (“tip locking”). They proposed a relationship between the migration of the flagellar contact site and surface motility displayed by attached microspheres. If paired flagella in the contact-migration phase of the mating reaction also had bound one or more beads, the beads would migrate to the tip region and become stationary there when the flagella “locked in” at the tips (Hoffman and Goodenough, 1980). The later observation can, in turn, also be related to a third phenomenon exhibited by gamete flagella during sexual agglutination, called “flagellar tip activation” (FTA). This is a rapid change in the internal ultrastructure of the flagellar tip, first described by Mesland et al. (1980). Flagella of chlamydomonads possess a typical axoneme, consisting of a central doublet of microtubules, surrounded by nine outer doublets. The central doublet is longer than the

H. VAN DEN ENDE

100

outer doublets, and is connected to the flagellar membrane via a cap structure (Lewin and Meinhart, 1953; Dentler and Rosenbaum, 1977; Melkonian, 1982). The A tubules of the outer doublets are all longer than the B tubules, and differ considerably in length. They are connected to the flagellar membrane by so-called distal filaments (Dentler, 1980). When a flagellum is activated during sexual agglutination, the A tubules become elongated to the very tip of the flagellum and seem to become associated with the cap. A second structural change during activation is the accumulation of electron-dense material in the tip region (Mesland et al., 1980). The result is that the normally thin flagellar tip becomes swollen. From the kinetics of this change, shown in Fig. 2, it appears that FTA precedes gametic cell fusion; maximal rate of cell pairing coincides with the maximum in FTA. Moreover, FTA is reversible: it declines while cell fusion proceeds and, when cell fusion is inhibited (e.g., in the combination C . reinhardtii mt+ imp-I x wild type mtr), there is no decline of FTA. Thus, FTA and sexual agglutination are somehow linked. This also appears from the fact that FTA can be evoked by treating gametes with isolated flagella derived from cells of the opposite mating type. The response can also be obtained by other, less specific means, such as treating cells with antiserum directed against whole flagella, or with concanavalin A, both causing agglutination of single mt cells (Table 1). In C . eugametos, it was shown that Fab fragments of similarly active antiserum failed to elicit FTA, which suggests that association of multivalent ligands with flagellar membrane components is required for FTA to occur (Elzenga et al., 1982). Colchicine and vinblastine, both inhibitors of tubulin polymerization, block FTA in C . reinhardtii, without affecting sexual agglutination. Sig-

0

5

10

15 20 Time ( m i n l

25

30

FIG. 2. Time course of the mating process together with the occurrence of unactivated, activated, and intermediate flagellar tips in C. eugamefos. (0) Vis-84s pairs; (0)unactivated flagellar tips; ( x ) activated flagellar tips; (+) intermediate forms. For each point, 130300 flagella and 240-300 cells were counted. From Elzenga et al. (1982).

TABLE 1 . Effect of flagellar membrane agglutinins on parameters of the sexual response in mt+ gametes of C. reinhardtii" Mating structure activation

Flagellar tip activationb

mt+ gametes presented with No additives Polylysine-coated gridc Preimmune serum (1:6)

Isolated mt- flagella (1.3 flagella cell-') Isolated mt- flagella after polylysine adhesionc OG- extractd (I .6 mg ml-I) aGG+f(l: 160) Con A (1 mg ml-I) (200 pg ml-')c

Agglutination

Tipping

Cell wall release

(%I Non

Inter

Act

Total scored

(%)

Total scored

9

No Adhesion

No No

No No

91 76

24

0 0

98 106

0 2

51

No

No

No

90

10

0

102

0

-

Yes

Yes

Yes

23

16

61

140

40

77

Yes

Yes

18

23

59

115

44

48

Yes' Yes

Yes Yes

Yes Yes

28 15

25 44

47 41

111 74

47 45

64 84

Yes Yes

Yes Yes

No -

47 61

29 18

24 21

72 106

3 23

59 56

From Mesland et al. (1980). Flagellar tips are scored intermediate (inter) when they cannot be scored as either unactivated (Non) or activated (Act). These include tips that have a clear intermediate morphology; also included are tips that are difficult to score as, for example, when aGG' presentation has induced antibody-vesicle complexes that are resistant to OG removal. Experiment with the cw-15 mt+ mutant; occasionally a gamete of this mutant bears a spontaneously activated small mating structure. d OG- extract, dialysed octylglucoside extract of mt- gametic flagella. e Gametes adhere to the film formed by the extract. f aGG+,antiserum raised against glutaraldehyde-fixed isolated gametic mt+ flagella.

102

H. VAN DEN ENDE

nificantly, mating structure activation and the induction of cell wall release are also severely inhibited (Mesland e f al., 1980). The authors argue that FTA may be caused by the formation of patches of flagellar surface material, due to the interaction with multivalent ligands, and the subsequent movement of the patches to the flagellar tip. Snell et al. (1983b) observed that during sexual agglutination the syntheses of two particular proteins, with molecular weights of 220,000 and 165,000, respectively, were induced. A strong accumulation of these proteins was only observed in flagella or flagellar membrane preparations, derived from agglutinating gametes of C . reinhardtii. The synthesis was arrested after cell fusion. Although no role could be assigned to these proteins, they may be candidates for the “fibrous” material that accumulates at the tips during flagellar tip activation. What is the function of FTA in the sexual process? Because under all experimental conditions mating structure activation and fusion are brought about under conditions allowing FTA to occur (cf. Table l), Mesland et al. (1980) proposed that FTA is essential for generating the signal which triggers mating structure activation and accompanying responses to sexual agglutination. The elongation of the nine A tubules and their association with the cap structure in the tip might prevent them from sliding along each other. The continued action of the axonemal dynein might then create a (mechanical) impulse in the intricate microtubule system at the flagellar bases, activating the mating structure (cf. Melkonian, 1982). Alternatively, FTA might represent a morphological manifestation or tip-oriented transport of flagellar membrane material, playing no role in signal generation, but inhibiting flagellar beating. It was mentioned previously (Section 1II.B) that flagella are restricted in their motility during sexual agglutination and exhibit only a twitching activity. This is supported by the recent observation that in vis-a-vis pairs of C . eugametos, in which only the mt+ pair of flagella regains motility, the mt- flagella remain activated (Crabbendam et al., 1984; cf. Fig. 2). In C . reinhardtii, all four of the zygote flagella are de-activated and subsequently take part in propelling the cell (Starling and Randall, 1971).

IV. The Specificity of Sexual Agglutination A. T H E F LA G ELLA R SURFACE

The flagellar axoneme is surrounded by a normal unit membrane. It has a “fuzzy coat”, due to the presence of abundant carbohydrate-containing

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

103

material in or associated with the actual membrane. Projecting from the membrane are thin hair-like protrusions, called mastigonemes. On glutaraldehyde-fixed cells of C . eugametos, disk-like appendages have been observed as outgrowths of the flagellar membrane (Mesland, 1977). These may be artifacts caused by the fixation procedure, but nevertheless have been connected with mating competence, because their appearance in regenerating flagella is correlated with the ability to form vis-a-vis pairs (cf. Section 1II.C). Flagella can be detached from the cells by any irritating agent (Lewin and Meinhart, 1953). Most frequently a pH- or Ca2+-induced shock is used (Witman et al., 1972) or a treatment with local anaesthetics (Witman et al., 1978). Detached flagella are excellent material for investigating the molecular basis of sexual agglutination, because they can be rigorously purified from material such as cell bodies, by differential centrifugation. If we assume that sexual agglutination, which is extremely sex and species specific, is due to the action of surface molecules unique for species and mating type, then the most logical approach to identify them would be to search for differences in composition between the flagellar membranes of both mating types within one species. Likewise, vegetative and gamete flagella of one mating type can be compared, presuming that in vegetative cells these components are absent or present in a modified form. In this approach, attention is focused on heavily glycosylated components exposed at the outer surface of the flagellar membrane (see, e.g., Musgrave e l al., 1979). That such components may play a role in sexual agglutination is indicated by the susceptibility of gametes to glycosidases (Wiese and Wiese, 1975). Another, more direct, approach to investigating the mechanism of sexual agglutination is to isolate and purify fractions exhibiting agglutinative activity in vitro. Both routes of investigation have been pursued and will be described in the next sections.

B . THE ISOAGGLUTININS

Particularly in C . eugametos, the so-called “isoagglutinins” have been useful in research concerning sexual agglutination. They were first described by Forster and Wiese (1954). Isoagglutinins consist of membrane vesicles released by gametes into the culture medium. They can be isolated by centrifugation (McLean et al., 1974; Musgrave et af., 1979). An isoagglutinin preparation derived from a gamete suspension brings about homotypic isoagglutination when added to gametes of the opposite mating type. The same sex- and species-specificity is displayed as that exhibited in normal sexual agglutination between cells of different mating type. Presumably, the vesicles exert their action by cross-linking the flagella.

H. VAN DEN ENDE

104

This implies that they contain the surface constituents responsible for sexual recognition and binding. It is consequently assumed that these vesicles are derived from the flagellar surface (McLean et al., 1974). This is supported by the fact that electron micrographs of isoagglutinin show vesicles, some of them having mastigonemes attached. In addition, isoagglutinin has not been detected in cultures of flagella-less mutants of C. reinhardtii (Bergman et al., 1975). Their glycoprotein composition is roughly similar to that of flagella, as shown in sodium dodecyl sulphate (SDS)-gel electrophoresis, but there are some conspicuous, unexplained differences (Musgrave et al., 1981). Why gametes produce isoagglutinin material is unknown. It may be partially the consequence of the way cells are cultivated. Most isoagglutinin material is produced during flagellar assembly when agar-grown cells are transferred to an aqueous milieu. In liquid cultures very little isoagglutinin is found (W. L. Homan, unpublished observation). Isoagglutinin does not seriously influence the sexual interaction between cells, so the view expressed by McLean et al. (1974), that its production is a coincidental occurrence, is probably correct. Nevertheless, the relative ease with which isoagglutinin is obtained and purified, as well as its simple composition in relation to isolated flagella (no matrix nor axoneme constituents), made it a valuable material for identifying the mt- factor involved in sexual agglutination of C. eugametos.

c.

THE

mt-

AGGLUTINATION FACTOR I N

Chlamydomonas eugametos

When mt- isoagglutination vesicles were subjected to mild sonication or to extraction with chaotropic agents, they remained virtually intact but their isoagglutinative power was strongly diminished (Homan et al., 1980). The resulting supernatant or extract (after dialysis) had a characteristic effect on mt+ gametes. It evoked twitching of the cells, reminiscent of their movements during sexual agglutination, but they remained apart although they tended to accumulate at the surface of the liquid. The mtcells were not affected in any way. The appearance of this “twitch activity”, the strength of which could be quantified by serial dilution, was correlated with the diminished isoagglutinative power of the vesicles. The explanation of this result is that material responsible for isoagglutination was released by the treatments and, in solubilized form, was functionally univalent. A multitude of the molecules involved, located at the surface of membrane vesicles, would endow them with the ability to cross-link mt+ flagella, whereas in a soluble form, they would still bind to the mt+ flagellar surface, but then only elicit the twitch response. This was established

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

105

by cross-linking the solubilized material with glutaraldehyde. This resulted in preparations with isoagglutinative power. Another, more impressive, way to prove the point was to adsorb the material onto charcoal particles, conferring on them the ability to adsorb mt+ gametes by their flagella. The procedure, shown in Fig. 3, has become the preferred way to assay the presence of the mt- agglutination factor of C. eugametos (Musgrave et al., 1981). Extracts of mt- isoagglutinin contain carbohydrate- and protein-containing material, which with SDS-gel electrophoresis were separated into a number of high-molecular-weight glycoconjugate fractions. They were stained with the periodic acid-Schiff base reagent (PAS) and were designated as PAS-1.1, -1.2, -1.3, -1.4, -2, -3, and -4,as shown in Fig. 4. On CsCl density gradient centrifugation, activity in the “twitch” bioassay was parallelled by the presence of the PAS-1.2 fraction which sedimented at 1.473 g ~ m - ~Similar . results were obtained using other separation techniques. On gel filtration over a Sepharose 2B column, a single peak of activity was observed just after the void volume. When active fractions were subjected to gel electrophoresis, all of them contained PAS-1.2. In isoagglutinin preparations, PAS-1.2 is present in relative abundance, as shown in Fig. 4. In flagella, however, it appeared to be a minor component. By comparing the biological activities of equal packed volumes, 5 p1 charcoal suspension

2’

Sarnpledilutions/2D

22

Z3

2‘

5 rnin

20 pI mt

+

gametes

Active dilutions (e.g.,2” to 2 3 ) Titer = Z3

FIG. 3. Illustration of the “charcoal assay” to quantify the amount of mt- agglutination factor in C. eugametos.

106

H . VAN DEN ENOE

FIG. 4. (a) Separation of high-molecular-weight mtr gamete flagellar glycoproteins from C. eugametos by SDS-polyacrylamide gel electrophoresis, according to Laemmli (1970) (B) compared with that according to Weber et al. (1972) (A). The major advantage of the Laemmli system is that the PAS-1 band is separated into four banks numbered PAS-1.1 to PAS-1.4. (b) Comparison of the relative amounts of PAS-1.2 in equivalent volumes of mtr isoagglutinin from C. eugametos. Equal volumes of both particles (84pI) suspended in 1 ml were tested for isoagglutination activity (vesicles 2*, isoagglutinin 213).Equivalent samples of each (250 pl) were centrifuged at 50,000 g for 30 minutes to sediment the particles. The pellets were dissolved in the same volumes of sample buffer and equivalent volumes were subjected to electrophoresis in 4% (w/v) Laemmli gels. The PAS-stained gels are shown. From Musgrave e r a / . (1981).

isoagglutinin vesicles showed a 30-fold increase in activity as compared with equivalent flagellar material, suggesting a positive correlation with the relative abundance of PAS-1.2. This was supported by comparing flagella of vegetative cells and gametes. The first appeared to be devoid of

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

107

FIG. 4. (continued)

PAS-1.2. Since vegetative cells do not agglutinate, it confirms that PAS1.2 is responsible for this ability (Musgrave et al., 1981). An antiserum raised against the purified PAS-1.2 fraction inhibited the respective activities of PAS-1.2 and mt- isoagglutinin. Fab fragments derived from this antiserum inhibited the overall sexual reaction. Although the antiserum crossreacted considerably with other fractions derived from the mt- flagellar surface or mt- isoagglutinin, these results support the hypothesis that PAS-1.2 is the mt- agglutination factor (Lens et al., 1980, 1982). The PAS-1.2 fraction showed the same elution characteristics in gel filtration on Sepharose 2B, irrespective of detergent and salt concentrations in the eluent. Pretreatment with reducing agents or EDTA had no effect on its elution volume (Kav 0.29). The biologically active fraction produced a single sharp protein- and carbohydrate-stainable band in 320% acrylamide SDS-gel electrophoresis. Thus, it seems that PAS-1.2 is a homogeneous fraction (Homan, 1982). The fact that it only just entered the 3% region of the polyacrylamide gel, as well as the low K,, value on

108

H. VAN DEN ENDE

gel filtration on Sepharose 2B, suggest that its molecular weight is extremely large. On the other hand, a value of 8.65 was obtained on equilibrium centrifugation in a sucrose gradient, corresponding to a molecular weight of roughly 2 x lo5 for a globular protein. Compared with its behaviour in gel filtration, this implies that the agglutination factor is an asymmetric molecule. It consists for 45% of carbohydrate, with arabinose (29%) and galactose (29%) as major sugars. Notable is also a relatively high hydroxyproline content (5%). Thus, the molecule is similar to other hydroxyproline-rich arabinogalactans, which are part of the cell wall of chlamydomonads and other green algae (see, e.g., Homer and Roberts, 1979). The factor loses its biological activity rapidly when exposed to cold dilute alkali. This implies that carbohydrate chains, linked O-glycosidically to serine and/or threonine, are part of the active site of the molecule or contribute to establishing the tertiary structure of the protein. Among the oligosaccharides released by this treatment, a disaccharide, containing arabinose and galactose, binds to antiserum raised against PAS-1.2 (Lens et al., 1983). It is attractive to assume, therefore, that these sugars are constituents of the active site. This corresponds with the observation that the agglutinative activity of mt- flagella is destroyed by a-galactosidase treatment (Williams, 1981). Also, isolated PAS-1.2 is susceptible to this enzyme (H. L. Pijst and H. van den Ende, unpublished observation). N-Glycosidic glycopeptide linkages, on the other hand, are unlikely to be involved, since N-acetylglucosamine is not a constituent of PAS-1.2. Similarly, the agglutinability of mt- cells is not affected by treatment with tunicamycin or endoglycosidase H (Williams, 1981; Wiese and Mayer, 1982). The mt- agglutination factor is probably an extrinsic membrane component, anchored to the membrane by an intrinsic protein. This is suggested by the high solubility of the agglutination factor in water. Additional evidence is provided by the following results. As already mentioned isoagglutinin vesicles can be inactivated by mild sonication, which results in the release of agglutination factor from the surface of the vesicles. When these inactivated vesicles are incubated overnight with the supernatant containing the solubilized agglutination factor, they are re-activated as a result of re-binding of the factor to the membrane surface. Interestingly, both mt+ and mt- derived isoagglutinins can be re-activated by the mtagglutination factor, resulting in vesicles with isoagglutinative power for mt+ cells only. This confirms that the factor is the only mt- component carrying determinants specifying sexual agglutination. When sonicated inactive vesicles are trypsinized, their ability to become re-activated is lost. Likewise, one can destroy this ability by heating inactivated parti-

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

109

cles for 10 minutes at 80°C. Treatment with periodic acid, under conditions whereby only carbohydrate moieties were oxidized, has no effect (Homan el al., 1982). This supports the view that a proteinaceous receptor is present in the membrane, which has no mating-type specificity, and which serves as a carrier protein for the agglutination factor. This protein has so far not been characterized. The simplest model to visualize the interaction between mt+ and mtflagellar surfaces in gametes of C . eugametos assumes that the mt- agglutination factor interacts with a complementary ligand at the mt+ surface. From using the charcoal assay, described in Section IV.C, it appears that in mt+ cells a PAS-1.2-like fraction is the only glycoconjugate fraction that shows activity in agglutinating mt- gametes. The mt+ factor has not yet been sufficiently purified to make any statement with regards to its structure. There is some indirect evidence, however, indicating that it is a glycoconjugate which, in contrast to the mt- agglutination factor, is sensitive to concanavalin A and a-mannosidase treatment (Wiese and Wiese, 1975; Musgrave el al., 1979). Biosynthesis of the active mt+ factor is inhibited by tunicamycin (Wiese and Mayer, 1982) which implies that a molecule with N-glycosidic glycopeptide linkages is involved and that terminal mannose and/or glucose residues confer on the molecule its specific adhesiveness. Thus, O-glycosidically-linked carbohydrate side chains, with terminal galactose and/or arabinose at the mt- agglutination factor, may interact with mannose-containing side chains at the complementary mt+ surface component. This seems in contrast with results of Gerwig et al. (1984). They analysed the monosaccharide composition of isoagglutinins from the opposite mating types of C . eugametos, which, as mentioned in Section IV.B, contain several glycoconjugates. Besides the common occurrence of Larabinose, L-rhamnose, D-xylose, D-mannose, D-galactose, D-glucose, and N-acetyl-D-glucosamine, the mt+isoagglutinin contained 4-O-methylxylose, 2-O-methylarabinose, and 3-O-methyl-~-galactose,but those of the mt- isoagglutinin, 6-O-methyl-~-mannose,and 3-O-methyl-~-glucose. A similar analysis of gamete flagella gave an identical result. The matingtype specificity of these methyl ethers is intriguing, the more so since they are not present in the mt- agglutination factor (Homan, 1982). Recently, however, genetic analysis has proved that these differences are not sexlinked (F. Schuring and H. van den Ende, unpublished). This rules out the possibility that O-methyl ethers of hexoses and pentoses play a role in sexual recognition. On the other hand, in 65 recombinant clones isolated, the susceptibility to tunicamycin was observed exclusively in mt+ clones. This observation extends to C . moewusii, of which two syngens (sexually matching pairs of isolates; syngen I is compatible with C . eugametos, but

110

H. VAN DEN ENDE

syngen I1 is sexually incompatible with either pair of strains) are known, as well as to C . reinhardtii (Wiese and Mayer, 1982). This lends extra credence to the idea that unilateral sensitivity to tunicamycin is an inherent property of the functional structure of one partner in the sexual recognition process. The role of 0-methylated sugar moieties remains obscure. In fact, their occurrence might be coincidental. They have occasionally been described in bacterial envelopes and cell walls of algae and higher plants (Weckesser et al., 1979; Fichtinger-Schepman et al., 1981; Kennedy, 1980; Darvill er al., 1978). In Rhizobium, they allegedly play a preventive role in recognition by other, predating, bacteria (Dudman, 1977). The experimental material is not sufficient, however, to propose a similar role for sugar 0methyl ethers in Chlamydomonas.

D. THE

mt+ AGGLUTINATION FACTOR

OF

Chlamydomonas reinhardtii

There are some similarities between the mt- agglutination factor of C. eugametos and the mt+ agglutination factor in C . reinhardtii, described by Adair er al. (1982, 1983) and Cooper et al. (1983). In several respects its identification followed a similar course. Flagella or whole cells were extracted with the dialysable non-ionic detergent octylglucoside or EDTA (4 mM) and the extracts were fractionated by gel filtration. Biological activity was assayed by allowing a sample of a test sample to dry on a glass slide; a drop of gametes of opposite mating type was placed over the dried material, and adhesion of the cells to the dried material was monitored. Active fractions were found to elute in or just after the void volume of a Sepharose 6B column and were observed to have very low mobility in polyacrylamide gel electrophoresis systems. In sucrose gradient centrifugation, the activity was found to concentrate in the 12s region. After deglycosylation with anhydrous HF, a new band appeared in polyacrylamide electrophoresis gels, with an estimated molecular weight of 480,000. The identification of the mt+ agglutination factor in C . reinhardtii was greatly facilitated by the availability of non-agglutinative mutants (Goodenough et al., 1978). From none of the mutants could the active glycoprotein be extracted. However, in diploids, constructed from mutants carrying complementary mutations, which mated with an efficiency almost equal to that of mt- gametes, the active fraction was shown to be present. The purified factor was further characterized by electron microscopy, following the “deep-etch’’ procedure of Heuser (1982). Active preparations consisted almost entirely of rod-like molecules of uniform length

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

111

(220 pm). Two populations were recognized with respect to diameter, which was interpreted to mean that the population was a mixture of monomers and dimers. It is not known whether one or the other, or both subpopulations are responsible for adhesiveness in mt+ flagella. Taking the results obtained with C. eugametos and C . reinhardtii together, it now appears well established that sexual agglutination is determined by glycoconjugates of very large size, bound to the outer flagellar surface in an extrinsic fashion, These molecules exert their action by adhering to complementary molecules localized at the surface of the gamete flagella of opposite mating type. The identity of these complementary ligands has not been determined so far, but the evidence from studies with C. eugametos as well as C. reinhardtii indicates that the recognition system in these species is unipolar, which means that only one set of complementary receptors governs sexual agglutination in each species.

V. Dynamics of Sexual Agglutination A. MECHANISM OF SEXUAL ADHESION

There are two remarkable and disturbing aspects of sexual agglutination which have become evident in the studies discussed so far. The first is that the surface glycoconjugates responsible for sexual adhesion are only minor components of the flagellar surface. This seems hard to reconcile with the observed high efficiency of the agglutination process. The second is that rarely has one been able to mimic sexual adhesion in vitro, i.e., between isolated flagella of both mating types, between isoagglutinins, derived from different gamete suspensions, or even between glutaraldehyde-fixed cells of different mating type (Kohle et al., 1980). So far, no complementary binding between flagellar surface components has been detectable in vitro. Most workers agree that for a demonstrable flagellar adhesion to occur, one of the interacting “partners” must be a living cell. In the light of these problems, the following remarks can be made. It seems that the interacting forces between flagella are rather weak. A clump of interacting cells can easily be disrupted by gentle manipulation, and the vis-a-vis pair yields in C. eugametos are considerably decreased by reciprocal shaking of the mixed cell suspension. In the second place, we have seen above that flagellar surface components may be redistributed along the flagellar surface. Several authors (e.g. Mesland and Van den Ende, 1978b; Adair et al., 1982) postulated that mobile receptors might become trapped at the site of interflagellar interaction, which would

112

H. VAN DEN ENDE

expand to a patch of occupied receptors, analogous to the redistribution of cell surface receptors induced by cell-cell contact in embryo cells of Xenopus (Chow and Poo, 1982). The resulting local high density of receptors might establish a relatively stable adhesion. The mobilization of receptors to the site of interaction might be an active metabolic process, controlled by submembranous cytoskeletal elements, and executed only in living cells. A further stabilization of flagellar contact could subsequently be realized by a translocation of occupied receptor patches to the flagellar tips followed by “tip locking” (see Section 1II.C). It should be emphasized, however, that receptor patching or clustering has never been directly demonstrated. It does not explain why killing the cells should release paired receptors, whether they are patched or tipped; nor does it explain why isolated receptors do not bind to each other in uitro.

B . RECEPTOR INACTIVATION

As described in Section III.B, sexual agglutination is a highly dynamic process. Cells adhere initially in a seemingly random fashion, but gradually sort themselves out into pairs. To explain this behaviour, the suggested lateral mobility of receptors in the plane of the flagellar membrane, is certainly of relevance. Another well substantiated contribution to the explanation of this behaviour is the finding that these receptors are subject to rapid inactivation. The phenomenon was first noted by Wiese and Wiese (1978) and van den Ende (1981) and elaborated experimentally by Snell and Roseman (1979) and Snell and Moore (1980) for C. reinhardtii, and confirmed for C . eugarnetos by Pijst et al. (1984). When gametes of one mating type are mixed with isolated flagella of the other mating type, a rapid adhesion occurs, due to specific interflagellar binding. Dependent on the assay and the conditions, this binding is maximal within minutes. Its rate is only slightly dependent on the temperature and is saturable: the extent of aggregation is directly proportional to the number of flagella added to a given cell suspension, up to a ratio of one or two flagella per cell (Snell and Roseman, 1979). However, as shown in Fig. 5 , shortly after aggregation is maximal, the aggregates gradually dissociate to single cells, due to the inactivation of the isolated flagella. This appears from the following facts: (1) When fresh flagella are added to the disaggregated suspension, the aggregation and subsequent dissociation occurs once again; this does not happen when fresh gametes are added to a disaggregated suspension. This indicates that the cells maintain their capacity for adhering to fresh flagella. (2) The flagella re-isolated after the aggregationdisaggregation process have lost their capacity to agglutinate fresh cells.

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

113

Time (minutes)

FIG. 5. Binding and inactivation of isolated 35S-labelledmr- flagella mixed with mt+ cells in C. eugametos. In a volume of 165 PI, 1.75 x lo6 cells were added to 15 p1 of mt- flagella, labelled in uiuo with [35S]S042(containing 59 x lo3dpm) with an agglutination titre of z5. At various times the reaction was terminated by centrifugation through a silicone oil layer. The pellets were cut off and counted. Values are means of three separate experiments in duplicate. (0)mt+ gametes; ( x ) mt- gametes. From Pijst et a / . (1984).

Thus, the adhesion process has apparently modified the isolated flagella, not the intact cells. Similar results were obtained when mt+ cells were incubated with mt- flagella, or vice versa. The situation appears to be completely reciprocal. By various means it was demonstrated that only bound flagella were inactivated: incubation of flagella with conditioned medium (Snell and Roseman, 1979), cell-free medium of agglutinating cells, or detergent-free Triton X-100 extracts of whole cells did not affect their agglutinating properties (Pijst et al., 1984). In a mixture of mt+ and mt- flagella only those flagella that were of the opposite mating type were inactivated by incubation with living cells. Thus, the inactivation does not result from the action of some soluble factor, such as a proteolytic or glycolytic enzyme released into the medium during the agglutination process. It rather is a process occurring at the flagellar surface, with strong mating-type specificity. The most attractive supposition of course is that this process concerns those flagellar components that are responsible for binding. In C . eugametos this was tested by labelling the highly purified mt- agglutination factor

114

H. VAN DEN ENDE

with 35S and adding it to m t f cells (Pijst et a f . , 1984). It was bound specifically and at a high rate to these cells. When added in large amounts, it inhibited the binding of mt- flagella in a presumably competitive manner. Likewise, the mt- agglutination factor subsequently lost its activity, although at a lower rate than observed with mt- flagella. When the inactivated factor was recovered, it had lost not only its binding properties, but also its ability to evoke “twitching”. When it was subjected to gel electrophoresis, and the resulting autoradiogram was compared with that of an untreated sample of the factor, no difference in electrophoretic mobility was visible. This suggests that inactivation of the agglutination factor is the consequence of a relatively minor modification of the molecule that has no detectable effect on the electrophoretic mobility in the gel system used. The nature of this process and the fate of the inactivated molecules are unknown. Another interesting aspect of the interaction between living cells and isolated flagella of the opposite mating type is that the cells remain seemingly unaffected. Snell and Roseman (1979) also suggested that on live cells the adhesive sites are destroyed but are replaced by fresh ones from within the cells. Indeed, cells pretreated with cycloheximide for various lengths of time were fully able to bind and inactivate flagella but were seen to gradually lose these properties. The implication is that during sexual agglutination there is a rapid turnover of agglutination sites, and that, at least in living cells, the sites are derived from a pool in the cell body. This pool has a steady-state size under normal conditions, due to de nouo biosynthesis, but is depleted when cycloheximide-treated cells are agglutinated. There is still another argument for the existence of a pool of agglutination sites, which comes from experiments with regenerating flagella, referred to earlier in Sections 1II.C and D. When gametes are deprived of their flagella and are allowed to regenerate a new pair, these are seen to be agglutinable, even when protein synthesis is blocked by cycloheximide during regeneration. Where is the location of this pool? Solter and Gibor (1978) observed, in flagellar regeneration experiments with C. reinhardtii, that the appearance of agglutinability was much retarded if the deflagellated cells were exposed to a treatment with trypsin. Since this treatment did not seem to affect flagellar regeneration, the authors suggested that the plasma membrane of the cell might be a reservoir of components involved in agglutination, which would migrate to the flagellar membrane during flagellar regeneration. Trypsin treatment would destroy this reservoir so that the cell would become fully dependent on de nouo protein synthesis. This might cause the delay in regenerating fully agglutinable flagella.

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

115

In another approach, the amount of mt- agglutination factor in C . eugametos was determined (Pijst et al., 1983). It was shown that cells, after the removal of the flagella, contained a 25-fold quantity of the factor, as compared with that present in flagella. For the larger part, this material was located in the plasma membrane, as demonstrated by indirect immunofluorescence using an antiserum directed against the agglutination factor, and also by extraction with 2 M guanidine thiocyanate, a chaeotropic agent that leaves the cell intact and extracts only the cell surface. It is attractive to assume that during flagellar regeneration or during sexual agglutination, the factor is transported from the plasma membrane to the flagellar surface, since both compartments are in principle continuous. However, a diffusion barrier in the form of a bracelet of membrane particles at the base of each flagellum has been proposed (Weiss et al., 1977). That such a constraint exists is also borne out by the fact that some cell-surface glycoconjugates are restricted to the plasma membrane whereas others are restricted to the flagellar membrane. At any rate, a flow of agglutination receptor from the plasma membrane to the flagellar membrane has not been observed so far. If this flow really does not occur, we are left with two questions: (1) what other compartment of the cell contains a reservoir of agglutination factor that can be mobilized (maybe the Golgi-derived “gametic vesicles,” described by Martin and Goodenough (1979, cf. Section 1I.A); and (2) what is the significance of the large amount of agglutination factor in a non-functional location underlying the cell wall? One can only speculate about the functional significance of the phenomenon that adhesion sites are inactivated while performing their service. As suggested above, it may be that it lends a dynamic character to interflagellar adhesion, and allows the cells in an agglutinating clump to sort themselves out into pairs. Another possible function of this process may be to realize a “down regulation” in order to accomplish flagellar dissociation soon after establishing the plasma connection leading to vis-a-vis pairs or zygotes. As discussed in Section II.B, the flagella completely lose their adhesiveness when the fertilization tubules of two partner cells have fused. One may imagine that once fusion has taken place, the transport of agglutination factor from a cellular reservoir to the flagella is arrested, and the flagella become non-agglutinable due to the mutual inactivation of the available adhesion sites. In fact, young vis-a-vis pairs of C . eugametos contain appreciable amounts of the mt- and mt+ agglutination factors in their cell bodies (A. Musgrave, unpublished observation). The concerted activities of binding and inactivation during sexual interaction in Chlamydomonas is reminiscent of similar processes occurring during cellular interaction in sponge cells (reviewed by Muller, 1982). In

116

H. VAN DEN ENDE

the so-called secondary cellular recognition of marine sponges, a soluble, high-molecular-weight aggregation factor is involved, which binds to a “base plate,” located at the cell surface, designated as the aggregation receptor. A cell surface-bound P-glucuronidase controls the functional activity of this receptor. A deglucuronylated receptor has no binding properties, and can be reactivated by an aggregation factor-linked glucuronyl transferase. It is assumed that the enzymic activation and inactivation processes of the aggregation receptor control the cell aggregation and the “sorting out” that occurs during the formation of a functional sponge from the secondary aggregates.

C . THE SIGNALLING ACTION OF SEXUAL AGGLUTINATION

In the preceding sections it was shown that sexual agglutination serves not only to stimulate physical contact between the mating cells but also to manoeuvre them into the correct position for papillar fusion. It has also a number of morphogenetic effects on the cells involved: it induces mating structure activation, cell wall shedding, and flagellar tip activation. In various ways, these functions can be uncoupled: by low temperature (Mesland, 1977; Mesland and van den Ende, 1978a; Weiss et al., 1977), in regenerating flagella (Solter and Gibor, 1978), and by mutation (Forest and Togasaki, 1975; Forest et al., 1978). On the other hand, mating structure activation always requires the interaction of flagellar membranes of opposite mating type, but it is not necessary that this interaction is mediated by the endogenous agglutination factors. Goodenough and Jurivich (1978) showed that non-agglutinative mutants of C. reinhardtii, which reportedly are devoid of the agglutination factor (Adair et al., 1983), can be activated by non-specific antiserum directed against flagellar surface components. This indicates that the biological information of the endogenous surface components responsible for agglutination concerns only the affinity for opposite mating type ligands. An essential feature of these activating factors is that they are multivalent. Fab fragments of the above mentioned antisera do not elicit mating structure activation, nor flagellar tip activation (Elzenga et al., 1982). Instead, they act as antagonists (Lens et al., 1982). The question then is, which function of the flagellar membrane of gametes is required to create the signal for activation? At present, the simplest hypothesis seems to be that the formation of patches or clusters of occupied agglutination factors (and/or other membrane components) and the subsequent movement of these to the flagellar tips is instrumental in triggering the‘ensuing cellular changes (Goodenough and Jurivich, 1978).

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

117

This is mainly based on the observed correlation between tipping ability and the ability to activate a mating structure and to effect flagellar tip activation. In turn, flagellar tip activation might contribute to signal transmission to the cell body via the axonemal structure (Mesland et al., 1980). The proposed role of surface component clustering is highly reminiscent of the mechanism of action attributed to polypeptide hormones in animal cells. These hormones exert their action by first binding to diffusely distributed mobile membrane receptors. The hormone-receptor complexes then cluster in a temperature-sensitive process to form patches, which initially are mobile but gradually cluster to form immobile aggregates, which ultimately are internalized by an energy-dependent process to form coated vesicles. The local aggregation of hormone-receptor complexes is essential for eliciting a biological response. Monoclonal antibodies against the receptor for the epidermal growth factor in human carcinoma A 431 cells induce a number of the effects of the epidermal growth factor (e.g., Kahn et al., 1978; Schechter et al., 1979; Schreiber et al., 1981). Hormone analogues that bind to their receptor but fail to form aggregates are without effect. One of the early responses preceding internalization is the phosphorylation of internal and surface-bound proteins, including the receptor itself (Hunter and Cooper, 1981; Kasuga et al., 1982). This phosphorylation is probably CAMP andlor Ca2+-calmodulin dependent (Cohen, 1982). There are some indications that at least Ca2+ is involved as second messenger in flagellar signalling. Snell et al. (1983a) found that the local anaesthetic, lidocaine, reversibly blocked cell wall release and fusion in C . reinhardtii, without affecting flagellar adhesiveness or blocking flagellar tip activation. The external concentration of Mg2+ and Ca2+ had a modulating effect. Since lidocaine is known to interfere with Ca2+binding sites and Ca2+-dependentprocesses, this suggests that Ca2+binding and/ or movement across the membrane may play an important role in the signalling process. It seems improbable that flagellar adhesion simply depolarizes the flagellar membrane and allows passive Ca2+influx, since this is known to occur as a response to various external stimuli, such as to changes in light intensity during fobotaxis (e.g., Hyams and Borisy, 1978). The situation is undoubtedly more complicated. The Ca2+ions may also be involved in later stages of the mating process. Using C . reinhardtii gametes preloaded with 45Ca2+,Bloodgood and Levin (1983) found a rapid, transient increase in calcium efflux when mt+ and mt- cells were mated (Fig. 6), or when a single mating-type cell suspension was treated with isolated flagella of the opposite mating type. These authors also observed that unmated cells absorbed Ca2+from the medium in a linear manner for many hours to millimolar concentrations.

118

H. VAN DEN ENDE

FIG. 6. Efflux of 45Cafrom gametes of Chlamydomonas reinhardtii under mating and nonmating conditions. The mt+ gamdes were incubated in a medium containing 10 pCi of 4SCa per millilitre for 2 hours and then washed with 10 mM Tns-HCI, pH 7.4. At time zero, the labelled cells were resuspended in the same buffer at a concentration of 5 X lo6 cells ml-I. At various times after that, duplicate 0.5-ml aliquots of cells were centrifuged and 0.2 ml of each supernatant was removed and the radioactivity measured. The ordinate represents radioactivity cpm 0.2 ml-I. (A) Efflux from labelled mt+ gametes alone; the arrow indicates the time (37.5 minutes) at which an equal number of unlabelled mt- gametes were added to half of the labelled m f +gametes ( 0 ) and an equal number of unlabelled mi+ gametes were added to the other half of the labelled mt+ gametes ( 0 ) . The slope (efflux rate) for the labelled mi+ alone is 11,500 cpm ml-I min-I. During the first 6 minutes after addition of unlabelled gametes, the mt'lmt- mixture has an efflux rate (92,300 cpm ml-l min-I) 20 times that of the mt+/mtmixture (4500 cpm ml-' min-I). After the first 6 minutes, the rate of efflux of mt+lmtmixture returns to a value1(4300cpmiml-l min-I) similar to that of the mt+lmt- mixture (4500 cpm ml-I min-I). The 2.5-fold decrease in efflux rate after the labelled mt+ cells were mixed with the unlabelled mt+ cells is primarily due to dilution. At 35 minutes after mixing, the mating efficiency of the mt+/mt- mixture was 82% and the mating efficiency of the mt+/mt+ mixture was 0%. At the time of mixing (37.5 minutes), the extracellular CaZ+concentration M 6 minutes after was 0.2 p ~in ;the mt+lmt- mixture this concentration rose to 0.4 ~ L within mixing. From Bloodgood and Levin (1983).

Since the internal Ca2+ concentration is probably in the order of 1 p~ (Schmidt and Eckert, 1976), they suggested that the cells sequester Ca2+ in intracellular storage sites. Analogous to the fertilization of animal cells (cf. Shapiro et al., 1981), a signal evoked in the flagella during agglutination might temporarily release the stored Ca2+, thereby enhancing the intracellular Ca2+concentration and, as a consequence, its active extrusion from the cells. The transiently increased Ca2+ concentration might trigger one or more of the later events in the mating process, particularly mating structure activation and cell wall release.

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

119

VI. Concluding Remarks It is hoped that this article has demonstrated that sexuality in Chlamydomonas is an extremely interesting system that can be approached by a variety of methodologies to gain insight into how cells recognize and interact with each other. The increasing appreciation of this system is reflected by the publication rate, particularly in the last few years, but one must acknowledge that the basis for this interest was laid in the early 1950s by the articles of R. A. Lewin and of L. Wiese. Recently, a number of achievements have contributed to the understanding of sexual interaction in Chlamydomonas spp. : the thorough description of the morphological and structural aspects of the mating process, the identification of some of the flagellar surface components involved in sexual agglutination, and the analysis of various parameters of flagellar surface dynamics. In the near future, efforts will certainly be directed towards answering the questions which, due to this progress, have been raised: what is the molecular morphology of the flagellar surface, how are the agglutination receptors anchored to this surface, and how is the signal that prepares the cell for fusion generated and transmitted? As we have seen, the Chiamydomonas system has many features in common with other systems in cell biology, and one can predict that communication systems studied in other cell types will also be found to be operative in Chlamydomonas. Hopefully, this review will serve as an impetus to help fill these gaps in our knowledge.

REFERENCES

Adair, W. S., Monk, B. C., Cohen, R., and Goodenough, U. W. (1982). Journal of Biological Chemistry 257, 4593. Adair, W. S., Hwang, C., and Goodenough, U. W. (1983). Cell 33, 183. Bergman, K., Goodenough, U. W., Goodenough, D. A., Jawitz, J., and Martin, H. (1975). Journal of Cell Biology 67, 606. Bernstein, E., and Jahn, T. L. (1955). Journal of Prorozoology 2, 81. Bloodgood, R. A. (1977). Journal of Cell Biology 75,983. Bloodgood, R. A. (1980). Journal of Experimental Zoology 2l3,293. Bloodgood, R. A. (1982). I n “Prokaryotic and Eukaryotic Flagella”. Symposium of the Society for Experimental Biology, vol. 35, pp. 353-380. Cambridge University Press, Cambridge. Bloodgood, R. A., and Levin, E. N. (1983). Journal of Cell Biology 97, 397. Bloodgood, R. A., and May, G. S. (1982). Journal of Cell Biology 93, 88. Bloodgood, R. A., Leffler, E. M., and Bojczuk, A. T. (1979). Journal of Cell Biology 82, 664.

120

H.VAN DEN ENDE

Brown, R., Jr., Johnson, C., and Bold, H. C. (1968). Journal of Phycology 4, 100. Cabib, E., Ulane, R. E., and Bowers, B. (1973). Journal ofBiologica1 Chemistry 248, 1451. Chow, J., and Poo, M. (1982). Journal of Cell Biology 95, 510. Claes, H. (1971). Archives of Microbiology 78, 180. Claes, H. (1975). Archives of Microbiology 103, 225. Claes, H. (1977). Experimental Cell Research 108, 221. Cohen, P. (1982). Nature, London 296, 613. Cooper, J. B., Adair, W. S., Mecham, R. P., Heuser, J. E., and Goodenough, U. W. (1983). Proceedings of the National Academy of Science of the United States of America 80, 5898. Crabbendam, K. J., Nanninga, N., Musgrave, A., and Van den Ende, H. (1984). Archiues of Microbiology 138, 220. Darvill, A. G., McNeil, M., and Albersheim, P. (1978). Plant Physiology 62,418. Dentler, W. L. (1980). Journal of Cell Science 42, 207. Dentler, W. L., and Rosenbaum, J. L. (1977). Journal of Cell Biology 74, 747. Detmers, P. A., Goodenough, U. W., and Condeelis, J. (1983). Journal of Cell Biology 97, 522. Dudman, W. F. (1977). In “Surface Carbohydrates of the Prokaryotic Cell” (J. W. Sutherland, ed.), pp. 357-414. Academic Press, New York. Elzenga, J. T. M., Nanninga, N., Lens, P. F., and van den Ende, H. (1982). FEMS Microbiology Letters 15, 279. Fichtinger-Schepman, A. M. J., Kamerling, J. P., Versluis, C., and Vliegenthart, J. G. F. (1981). Carbohydrate Research 93, 105. Forest, C. L., and Togasaki, R. K. (1975). Proceedings of the National Academy of Science of the United States of America 72, 3652. Forest, C. L., Goodenough, D. A., and Goodenough, U. W. (1978). Journal of Cell Biology 79, 74. Forster, H., and Wiese, L. (1954). Zeitschrift fiir Naturforschung Teil B9, 470. Friedmann, J., Colwin, A. L., and Colwin, L. H. (1968). Journal of Cell Science 3, 115. Gerwig, G. J., Kamerling, J. P., Vliegenthart, J. F. G., Homan, W. L., van Egmond, P., and van den Ende, H. (1984). Carbohydrate Research, 127, 245. Gooday, G. W., and Trinci, A. P. J. (1980). In “The Eukaryotic Microbial Cell” (G. W. Gooday, D. Lloyd, and A. P. J. Trinci, eds.), Symposium of the Society of General Microbiologists, No. 30, pp. 207-251. Cambridge University Press, Cambridge. Goodenough, U. W. (1977). In “Receptors and Recognition” (J. L. Reissig, ed.), Series B, Vol. 3, pp. 323-351. Chapman and Hall, London. Goodenough, U. W., and Jurivich, D. (1978). Journal of Cell Biology 79, 680. Goodenough, U. W. and Thorner, J. (1983). In “Cell Interactions and Development: Molecular Mechanisms” (K. M. Yamada, ed.), pp. 29-75. Wiley, New York. Goodenough, U. W., and Weiss, R. L. (1975). Journal of Cell Biology 67 623. Goodenough, U. W., Hwang, C., and Martin, H. (1976). Genetics 82, 169. Goodenough, U . W., Hwang, C. J., and Warren, A. J. (1978). Genetics 89, 235. Goodenough, U. W., Adair, W. S., Culigor, E., Forest, C. L., Hoffman, J. L., Mesland, D. A. M., and Spath, S. (1980). In “Membrane-Membrane Interactions” (N. B. Gilula, ed.), pp. 131-152. Raven Press, New York.

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

121

Goodenough, U. W., Detmers, P. A., and Hwang, C. (1982). Journal of Cell Biology 92, 378. Gowans, C. S . (1976). In “The Genetics of Algae” (R. A. Lewin ed.), pp. 145174. Blackwell, Oxford. Hartmann, M. (1956). “Die Sexualitat.” Gustav Fischer Verlag, Stuttgart. Heimke, J. W., and Starr, R. C. (1979). Archiv fuer Protistenkunde 122, 20. Heuser, J. E. (1982). Journal of Cell Biology 95, 462 a . Hoffman, J. L., and Goodenough, U. W. (1980). Journal of Cell Biology 86, 656. Homan, W. L. (1982). Doctoral Dissertation, University of Amsterdam. Homan, W. L., Musgrave, A., Molenaar, E. M. and van den Ende, H. (1980). Archives of Microbiology 128, 120. Hornan, W. L., Hodenpijl, P. G., Musgrave, A., and van den Ende, H. (1982). Planta 155, 529. Homer, R. B., and Roberts, K. (1979). Planta 146, 217. Hunter, J., and Cooper, J. A. (1981). Cell 24, 741. Hwang, C. J . , Monk, B. C., and Goodenough, U. W. (1981). Genetics 99, 41. Hyams, J. S., and Borisy, G. G. (1978). Journal of Cell Science 33, 234. Jaenicke, L., and Waffenschmidt, S. (1981). Ber. Deutsche Botanische Gessellschaft 94, 375. Kahn, C. R., Baird, K. L., Jarrett, D. B., and Flier, J. S. (1978). Proceedings of the National Academy of Science of the United States of America 75, 4109. Kaska, D. D., and Gibor, A. (1982). Experimental Cell Research 138, 121. Kates, J. R., and Jones, R. F. (1964). Journalof CellComparative Physiology 63, 157. Kasuga, M., Zick, Y . ,Blithe, D. L., Crettaz, M., and Kahn, C. R. (1982). Nature, London 298, 667. Kennedy, L. D. (1980). Carbohydrate Research 87, 156. Kniep, H. (1928). “Die Sexualitat der niederen Pflanzen.” Jena Verlag von Gustav Fischer. Kohle, D., Lang, W., and Kauss, H. (1980). Archives of Microbiology 27, 239. Laemmli, U . K. (1970). Nature, London 227, 680. Lens, P. F., van den Briel, M. L., Musgrave, A . , and van den Ende, H. (1980). Archives of Microbiology 126, 77. Lens, P. F., Olofsen, F., Nederbragt, A., Musgrave, A., and van den Ende, H. (1982). Archives of Microbiology 131, 241. Lens, P. F., Olofsen, F., van Egmond, P., Musgrave, A., and van den Ende, H. (1983). Archives of Microbiology 135, 31 1 . Lewin, R. A. (1952a). Biological Bulletin 102, 74. Lewin, R. A. (1952b). Journal of General Microbiology 6,249. Lewin, R. A. (1954). In “Sex in Microorganisms.,” pp. 100-133. American Association of Advanced Science. Lewin, R. A., and Meinhart, J. 0. (1953). Canadian Journal of Botany 31, 711. Martin, N. C., and Goodenough, U. W. (1975). Journal of Cell Biology 67, 587. McLean, R. J., Laurendi, C. J., and Brown, R. M., Jr. (1974). Proceedings of the National Academy of Science of the United States of America 71, 2610. Melkonian, M. (1980). Journal of Cell Science 46, 149. Melkonian, M. (1982). In “Prokaryotic and Eukaryotic Flagella.” Symposium of the Society for Experimental Biology, Vol. 35, pp. 589-606. Cambridge University Press, Cambridge.

122

H. VAN DEN ENDE

Mesland, D. A. M. (1976). Archives of Microbiology 109, 31. Mesland, D. A. M. (1977). Protoplasma 93,311. Mesland, D.A. M., and van den Ende, H. (1978a). Archives of Microbiology 117, 131. Mesland, D. A. M., and van den Ende, H. (1978)b). Protoplasma 98, 115. Mesland, D. A. M., Hoffman, J. L., Caligor, E., and Goodenough, U. W. (1980). Journal of Cell Biology 84, 599. Miller, D. H., Mellman, J. S., Lamport, D. T. A., and Miller, M. (1974). Journal of Cell Biology 63,420. Miiller, W. E. G. (1982). International Reuiew of Cytology 77, 129. Musgrave, A., Homan, W. L., van den BrieI, M. L., Lelie, N., Schol, D., Ero, L., and van den Ende, H. (1979). Planta 145, 417. Musgrave, A., van Eyk, E., te Welscher, R., Broekman, R., Lens, P. F., Homan, W., and van den Ende, H. (1981). Planta 153, 362. Musgrave, A., de Wildt, P., Broekman, R., and van den Ende, H. (1983). Planta 158, 82. Pijst, H. L. A., Zilver, R. J., Musgrave, A., and van den Ende, H. (1983). Planta 158. 403. Pijst, H. L. A., Ossendorp, F. A., van Egmond, P., Kamps, A., Musgrave, A., and van den Ende, H. (1984). Planta 160, 529. Ray, D. A., Solter, K. M., and Gibor, A. (1978). Experimental Cell Research 114, 185. Rayburn, W. R. (1974). Journal of Phycology 10, 258. Sager, R., and Granick, S. (1954). Journal of General Physiology 37, 729. Schechter, Y.,Hernaez, L., Schlessinger, J., and Cuatrecasas, P. (1979). Nature, London 278, 835. Schlosser, U. G. (1976). Ber. Deutsche Botanische Gessellschaft 89, 1 . Schlosser, U. G. (1981). Ber. Deutsche Botanische Gessellschaft 94, 373. Schlosser, U. G., Sachs, H., and Robinson, D. G. (1976). Protoplasma 88, 51. Schmeisser, E. T., Baumgartel, D. M., and Howell, S. H. (1973). Developmental Biology 31, 3 1 . Schmidt, J. A. and Eckert, R. (1976). Nature, London 262, 713. Schreiber, A. B., Lax, J., Yarden, Y.,Eshar, Z., and Schlessinger, J. (1981). Proceedings of the National Academy of Science of the United States of America 78, 7535. Shapiro, B. M., Schackrnan, R. W., and Gabel, C. A. (1981). Annual Review of Biochemistry 50, 815. Shyam, R., and Sarma, Y.S. R. K. (1976). Archiv fuer Protistenkunde 118, 107. Snell, W. J. (1982). Experimental Cell Research l38, 109. Snell, W. J., and Moore, W. S. (1980). Journal of Cell Biology 84, 203. Snell, W. J., and Roseman, S. (1979). Journal of Biological Chemistry 254, 10820. Snell, W. J., Buchanan, M., and Clausell, A. (1983a). Journal of Cell Biology 94, 607. Snell, W. J., Clausell, A., and Moore, W. S. (1983b). Journal of Cell Biology 96, 589. Solter, K. M., and Gibor, A. (1977). Nature, London 265, 444. Solter, K.M., and Gibor, A. (1978). Experimental Cell Research 115, 175. Starling, D.,and Randall, J. (1971). General Research 18, 107. Starr, R. C., and Jaenicke, L. (1974). Proceedings of the National Academy of Science of the United States of America 71, 1050.

SEXUAL AGGLUTINATION IN CHLAMYDOMONADS

123

Stein, J. R. (1966). Journal of Phycology 2, 23. Taylor, R. B., Duffus, W. P. H., Raff, M. C., and de Petris, S. (1971). Nature New Biology 233, 225. Trainor, F. R. (1959). American Journal of Botany 46, 65. Trainor, F. R. (1975). Phycologia 14, 167. Triemer, R. E., and Brown Jr., R. (1975a). Protoplasma 84, 315. Triemer, R. E., and Brown, R. M., Jr. (1975b). Protoplasma 55, 99. Vanwinkle-Swift, K. P., and Aubert, B. (1983). Nature, London 303, 167. Vanwinkle-Swift, K. P., and Burrascano, C. G. (1983). Genetics 103, 429. van den Ende, H. (1981). I n “Sexual Interactions in Eukaryotic Microbes” (D. H. O’Day and P. A. Horgen, eds.), pp. 297-318. Academic Press, New York. Weber, K., Pringle, J. R., and Osborn, M. (1972). I n “Methods of Enzymology,” (C. U. W. Hirs and S. N. Timasheff, eds.), Vol. 26, pp. 3-27. Academic Press, New York. Weckesser, J., Drews, G., and Mayer, H. (1979). Annual Review of Microbiology 33, 215. Weiss, R. L., Goodenough, D. A., and Goodenough, U. W. (1977). Journal of Cell Biology 72, 144. Wiese, L. (1976). I n “The Genetics of Algae” (R. A. Lewin, ed.), pp. 174-197. Blackwell, Oxford. Wiese, L., and Mayer, R. A. (1982). Gamete Research 5, 1. Wiese, L., and Wiese, W. (1975). Developmental Biology 43, 264. Wiese, L., and Wiese, W. (1978). I n “Cell Interactions” (A. S. G. Curtis, ed.), Symposium of the Society of Experimental Biology, Vol. 32, pp. 83-104. Cambridge University Press, Cambridge. Williams, L. A. (1981). Journal of Phycology 17, Supplement, Abstract 2. Witman, G. B., Carlson, K., Berliner, J., and Rosenbaum, J. L. (1972). Journal of Cell Biology 54, 507. Witman, G. B., Plummer, J., and Sander, G. (1978). Journal of Cell Biology 76, 729.

This Page Intentionally Left Blank

The Energy Flow in Bacteria: The Main Free Energy Intermediates and Their Regulatory Role K. J. HELLINGWERF and W. N. KONINGS Department of Microbiology, University of Groningen, Haren, The Netherlands 1. Introduction . . . . . , . 11. The energy circuit in bacteria . . . . . . 111. Energy transduction in the cytoplasmic membrane. . . A . Primary transport systems . . . . . . B. Secondary transport systems . . . . , . C. Group translocation . . . . . D. ATP-Dependent solute transport systems. . . . E. The generation of a proton motive force by end product efflux IV. Regulation by energy intermediates . . . , . A. Regulation by redox potentials . . . . . B. Regulation by phosphorylation potentials. . . . C. Regulation by electrochemical potential gradients . . V. Homoeostasis in the magnitude of free energy intermediates . VI. Conclusions and perspectives . . . . . . VII. Acknowledgements . . . . . . . . References . . . . . . . . .

.

,

.

.

,

.

125 126 130 131

.

. . 132

.

.

. . . . .

.

,

.

134 136 136

. 138

.

138 140 . 142 . 146 . 148 . 149

.

. I50

,

.

.

. .

I. Introduction

Bacteria require for growth and survival chemical free energy which, in most organisms, is derived from catabolic substrates. In phototrophic bacteria light energy can be the main source of this chemical free energy. The energy supplied by these energy sources is usually not directly applied for the synthesis of cell material, or for other energy requiring ADVANCE\

VOL 26

IN MICROBIAL PHYSIOLOGY

Co,nricllr

IYN5 h\ A
$1

126

K. J. HELLINGWERF AND W. N. KONINGS

processes, but is first translated into other forms of metabolic energy. In general, three main types of “metabolic energy intermediates” can be distinguished. With these energy intermediates, energy-requiring metabolic processes can be driven and/or the energy can be transduced into other energy intermediates (Mitchell, 1968; Harold, 1972; Skulachev, 1977; Konings et al., 1981; Westerhoff et af., 1982; see Fig. 1). These three main metabolic energy intermediates are phosphorylation potentials (AG,), redox potentials (BE), and transmembrane (e1ectro)chemical potentials (ApJ, The phosphorylation potentials of energy-rich phosphorylated nucleotides such as ATP have, since the pioneering studies of Lipmann (1941), been recognized as important intermediates between free energy-releasing and energy-requiring metabolic processes. Actually, so much emphasis has been placed on the energy-currency function of ATP that the important role of other energy intermediates has largely been ignored up to the current day. Redox potentials exist in the form of reducing equivalents, functioning as cosubstrates in many enzymic reactions. Since the discovery of the process of oxidative phosphorylation, the redox potentials are known to play an important role as driving force for ATP synthesis (Slater, 1981). The important role of transmembrane electrochemical potentials as metabolic energy intermediates was realized first by Mitchell, and this led to the postulation of the chemiosmotic hypothesis (Mitchell, 1968, 1981). Many experimental data, especially since 1970, indicate the important role of these electrochemical potentials in bacterial bioenergetics (for reviews see Harold, 1972; Skulachev, 1977; Konings et al., 1981). These developments have changed considerably our picture of the energy flow in bacteria and other organisms. In this paper we discuss the current notion of energy flow in bacteria with special emphasis on the regulation exerted by the components of the three main free energy intermediates on the microbe’s metabolism.

11. The Energy Circuit in Bacteria

The flow of metabolic energy in bacteria can be presented schematically as in Fig. 1. Catabolic substrates which have been taken up by the bacterium can be metabolized via a wide variety of pathways (Stanier et al., 1978; Schlegel, 1981). During this catabolism the energy released is partially stored in redox intermediates such as NADH or succinate (process 1) and by the process of substrate level phosphorylation, partially in energy-rich phosphorylated intermediates such as ATP (process 2).

127

ENERGY FLOW IN BACTERIA

SOLUTE GMDIENTS

f

FIG. 1 . The energy flow from catabolic substrates or light to the main energy intermediates in bacteria leading to the synthesis of cell material.

The redox intermediates can subsequently be oxidized via (cytochrome-linked) electron transfer systems, located in the cytoplasmic membrane (process 4). According to the chemiosmotic hypothesis (Mitchell, 1968, 1981) this process is coupled to the translocation of protons from the cytoplasm to the external medium. Since the cytoplasmic membrane is rather impermeable to ions such as protons and hydroxyl ions (due to their low concentration at moderate pH values; see Deamer and Nichols, 1983) proton translocation leads to the generation of an electrochemical proton gradient across the cytoplasmic membrane (A&.H+).This A&H+is composed of an electrical component, the electrical potential (A+), and a chemical component, a pH gradient (ApH). This electrochemical proton gradient exerts an inwardly directed force on the

128

K. J. HELLINGWERF AND W. N. KONINGS

protons, the proton motive force Ap: expressed in the form of an equation, Ap

=

ApH+IF= A$ - ZApH (mV)

in which Z = 2.3 RTJF, where R is the gas constant, T the absolute temperature, and F the Faraday constant. The factor Z converts the pH gradient into millivolts. Z has a numerical value of about 60 mV per pH unit at 25°C. In addition to the redox energy-driven linear electron transfer systems (respiratory chain, anaerobic electron transfer systems) a proton motive force can be generated in phototrophic bacteria by light-induced cyclic electron transfer, and in halobacteria by the light-induced proton pump bacteriorhodopsin (process 7). This proton motive force is the driving force for many energy requiring processes (see Table 1 in Section 1V.C). In reversed electron flow, the A p drives electrons uphill against the redox potential, leading to the generation of redox equivalents such as, for instance, NADH (process 3). The membrane-bound transhydrogenase can catalyse the reduction of NADP by the Ap-driven transfer of electrons from NADH, which results in a more negative redox potential of the NADP(H) couple with respect to the NAD(H) couple. The electrochemical energy of the Ap also can be converted into the (e1ectro)chemical energy of solute (metabolizable substrates or ions) gradients. These Ap-driven translocation systems are termed “secondary” transport systems and ion pumps, driven by a (photo)chemical reaction, are termed primary transport systems (Konings and Michels, 1980; see below and Fig. 2). Since, in essence, all energy-transducing systems are reversible (Mitchell, 1968; Lancaster and Hinkle, 1977) the secondary solute transport systems can, under appropriate conditions, operate as Ap generating systems in which the electrochemical energy of solute gradients is converted into the electrochemical energy of a proton gradient or of other solute gradients. This aspect will be discussed in more detail below. An important role of the Ap in bacterial bioenergetics is its function as a driving force for ATP synthesis by the membrane-bound ATPase (process 5 in Fig. 1). In this process the electrochemical energy of the proton gradient is converted into the chemical energy of ATP hydrolysis. Under appropriate conditions the reversed process also can occur. The ATPase then functions as a primary proton pump. The synthesis of ATP by the membrane-bound ATPase has erroneously been considered to be obligatorily coupled to electron transfer. The terms “oxidative phosphorylation” and “photophosphorylation” originate from this misconception (Slater, 1981). The notion has gradually emerged

129

ENERGY FLOW IN BACTERIA PRIMARY TRANSPORT

SECONDARY TRANSPORT

:E-:+

syrnport

electron transfer

H+

A

hv Bacterio-

uniport

A

H+

anti p o r t

H+ ATPase A D P + Pi

FIG. 2. Schematic presentation of primary and secondary transport systems in bacteria.

that these terms are misleading and that the process of phosphorylation within the scheme of the Mitchell hypothesis is not obligatorily linked to electron transfer processes. In view of current knowledge it would be more precise to use the term “proton motive force coupled phosphorylation,” in analogy with substrate level phosphorylation. The chemical energy of ATP can be used for the synthesis of other nucleotide phosphates, phosphoenolpyruvate and acetyl phosphate via specific processes which will not be discussed further (see Thauer et al., 1977). This phosphate bond energy can be used for the accumulation of certain solutes and ions (see Section 111). It should be realized that metabolic energy also is required for the uptake of the catabolic substrates. Depending on the nature of the catabolic substrate, and on the organism, this can be electrochemical energy or phosphate bond energy. As a result of these translocation processes and subsequent metabolism, metabolites are formed intracellularly. The resulting metabolite gradients represent a form of energy which, in certain cases, can be used as metabolic energy. For instance, the efflux of end products of metabolism can occur via specific transport proteins in symport with protons, and this process can lead to the generation of a Ap which subsequently can drive energy-requiring processes. In this way energy recycling can occur (Michels et d., 1979). The amount of energy stored in the A&+ is rather small. Under certain

130

K. J. HELLINGWERF AND W. N. KONINGS

conditions more energy can be stored in ion gradients of sodium and potassium, which can occasionally function as energy buffers. For instance, Brown et al. (1983) demonstrated that, in marine organisms, Na+ ions are kept out of the cells by Ap-driven transport systems. Under conditions in which the cells are unable to maintain a high Ap, the Na+ gradient is dissipated by the influx of Na+, and this influx is coupled to the efflux of protons and leads to the generation of a AfiH+. Furthermore, efflux of potassium from Rhodopseudomonas sphaeroides, anaerobically in the dark, contributes to the maintenance of a A$ under these conditions (M. G. L. Elferink, unpublished observation). Recently evidence has been presented for the existence of primary sodium pumps. A respiration-dependent Na+ pump was found in the marine organism Vibrio alginolyticus and in the moderately halophilic Vibrio costicola (Tokuda and Unemoto, 1982). In Klebsiella aerogenes an oxaloacetate decarboxylase catalyses sodium extrusion across the cytoplasmic membrane (Dimroth, 1982a,b). The resulting inwardly directed sodium gradient can drive the uptake of solutes via Na+-solute symport systems in a number of bacteria (Lanyi, 1979; Tokuda et al., 1982). All three main energy intermediates are required for the synthesis of cell polymers. Many different anabolic pathways lead to the synthesis of the many different polymers which together form the complex structure of a bacterial cell.

III. Energy Transduction in the Cytoplasmic Membrane The energy-transducing systems in the cytoplasmic membrane are in essence transport systems for solutes (metabolizable substrates and ions). The systems can be classified mechanistically into four groups (Konings and Michels, 1980) according to the energy-transducing processes catalysed. (1) The primary transport systems convert chemical or light energy into electrochemical energy. These transport systems comprise the electrogenic proton pumps: the electron transfer systems, the Ca2+,Mg2+stimulated ATPase, and, in halobacteria, the light-driven proton pump bacteriorhodopsin. (2) The secondary transport systems are driven by electrochemical energy. These are the main solute transport systems found in bacteria. (3) The group translocation systems translocate solutes by an enzymic reaction. In this process the solute is chemically modified and the product is released in the cytoplasm. (4) The ATP-driven transport systems translocate solutes by specific membrane proteins. The energy for translocation is supplied by phosphate-bond energy directly. The

ENERGY FLOW IN BACTERIA

131

available information is not always sufficient to classify the transport systems in these mechanistic groups. In some systems a functional classification into A p generators and Ap consumers can be very useful, in view of the types of regulation to which the latter are subjected in some organisms (Elferink el al., 1983a).

A . PRIMARY TRANSPORT SYSTEMS

The main primary transport systems in bacteria are the electron transfer systems and the Ca2+,Mg2+-stimulated ATPase complex. The electrogenic extrusion of protons by these proton pumps leads to the generation of a proton motive force which is usually internally alkaline and negative (Mitchell, 1968; Konings et al., 1981). The driving force for proton extrusion in electron transfer systems is the redox potential difference across a proton translocation site (AE) and, in the ATPase complex, the change in free energy upon ATP hydrolysis, the phosphate potential A G . When thermodynamic equilibrium between the proton extruding force and the proton motive force is reached the following equations hold: nA@H+ =

AE

nA@n+ = AGp in which n is the number of protons translocated per electron transferred or per ATP hydrolysed. These equations illustrate that the electron transfer systems and the ATPase complex are coupled via the proton motive force. Both systems can also catalyse the reverse reaction: reversed electron flow leading to the generation of reducing equivalents and ATP synthesis, respectively. When a bacterial cell has a high concentration of reducing equivalents and a low phosphate potential, electron transfer can generate a proton motive force which subsequently can drive the synthesis of ATP. In the reversed situation, ATP hydrolysis can generate a proton motive force which subsequently can drive the synthesis of reducing equivalents. The proton motive force is a driving force for several energy-consuming processes in the cytoplasmic membrane. Since the activity of these energy-consuming processes will cause a dissipation of the proton motive force, a continuous proton extrusion by the proton pumps is required in order to maintain a high proton motive force. The rate at which protons can be translocated across the membrane by proton pumps, therefore, sets a limit to the rate at which work can be done. Bacteria have found

132

K. J. HELLINGWERF AND W. N. KONINGS

solutions to increase the rate of proton translocation by increasing the surface area of the cytoplasmic membrane, thus allowing the incorporation of more proton pumps. Examples are the chromatophore membranes in phototrophic organisms and the lamellar membranes of bacteria involved in the nitrogen cycle (for example, Nitrosococcus oceanus). The proton motive force has been measured in many bacteria grown under aerobic, anaerobic, and phototrophic conditions. In several organisms the Ap value could have been overestimated as a result of improper correction for probe binding (Lolkema et al., 1982). Under aerobic and phototrophic conditions, proton motive forces up to -200 mV can be generated. Anaerobically the proton motive force tends to be lower (around -100 to -150 mV; Hellingwerf el al., 1981). Since many processes in the cell’s metabolism are directly driven or influenced by the proton motive force, the activity of these processes will vary with the growth conditions. Clearly, the complex process of growth proceeds with proton motive force values ranging between -100 mV and -200 mV, indicating that a balanced regulation of the energy transducing processes must occur.

B. SECONDARY TRANSPORT SYSTEMS

The translocation of most solutes across the cytoplasmic membrane is driven by the proton motive force, or by one of its components. Some solutes, such as undissociated weak acids or bases and lipophilic ions, can cross the membrane passively without the involvement of specific carrier proteins. The transport of most solutes, however, is facilitated by a specific carrier protein. Often such proteins also are found for membranepermeable compounds, apparently to allow high rates of solute translocation. The energy-dependent carrier-mediated translocation process is usually termed “active transport.” Facilitated secondary transport can occur by three different mechanisms (Mitchell, 1968): (1) “Uniport”: only one solute is translocated by the carrier protein. (2) “Symport”: two or more different solutes are translocated by one carrier in the same direction. (3) “Antiport”: two or more different solutes are translocated by one carrier in opposite directions. The driving force for passive and facilitated solute transport is supplied by the electrochemical gradient(s) of the translocated solute(s). Figure 3 portrays the driving forces for different solute translocation processes and the steady-state levels of solute accumulation when thermodynamic equilibrium is attained by the translocation system.

133

ENERGY FLOW IN BACTERIA external med i um

membrane

cytoplasm

driving force

AuA = 0

ApA

ApA = -A$

AuA + A$

A+

steady s t a t e

.ApA

-

ZApH

ApA

+

A$

-

AuA = -ZApH

Z A ~ H ApA = -A$

+ ZApH

FIG. 3. Schematic presentation of four transport processes: (1) passive transport of a neutral solute; (2) facilitated transport via a uniport system; (3) facilitated transport via an antiport system; (4) facilitated transport via a symport system. For each process the driving force and the steady-state level of accumulation are given.

These examples demonstrate that an equilibration of the internal concentration with the external concentration will be achieved at steady state only in the passive transport of a neutral non-metabolizable solute (example 1). In the presence of a A$, facilitated (and passive) transport of a positively charged solute (example 2) can lead to an accumulation of this solute internally. On the other hand, in the absence of a A$ the uptake of A+ will generate a A$, internally positive, and net uptake will stop before equilibration has been reached. In example 3 (Fig. 3) an antiport system is shown for a positively charged solute A+ which is translocated in antiport with a proton. In this translocation process only the ApH component of the proton motive force contributes to the driving force. In example 4 a neutral solute is symported with a proton. In this translocation process the total proton motive force contributes to the driving force for solute uptake. In a similar way the driving forces and steady-state levels of transport can be derived for more complex transport processes (Konings and Michels, 1980). These examples clearly demonstrate that the driving force for solute transport is dependent on the total charge and the number of protons that are transported. Consequently when, during transport, the composition and the magnitude of the proton motive force is kept constant by primary transport systems, different levels of accumulation can be achieved for different solutes. This can be illustrated with the exam-

134

K. J. HELLINGWERF AND W. N. KONINGS

ples given in Fig. 3. When a proton motive force exists of -180 mV, which is composed of a A$ of - 120 mV and a ZApH of 60 mV, the internal concentration of solute A at steady state can maximally exceed the external concentration by a factor of 100, in example 2 [60 log(Ai,/A,,,) = + 1201, by a factor of 1/10 in example 3 [60 log(Ai,/Aout)= -601, and by a factor of 1000 in example 4 160 log(Ai,/Aout)= +180]. It should also be obvious from these examples that changes in the composition of the proton motive force will have different effects on the transport of different solutes. A dissipation of the pH gradient will strongly affect the ApHdependent transport systems. Many bacteria which grow at neutral pH maintain an internal pH value of 7.5 (see below). Consequently, transport processes in which only the ApH component contributes to the driving force can only occur in media with external pH values below 7.5. In order to make accumulation of these solutes possible above pH 7.5, bacteria have developed symport systems with variable proton-solute stoicheiometry (number of protons translocated per molecule of solute) (Rottenberg, 1976; Ramos and Kaback, 1977a; Konings and Booth, 1981). As a result, the driving force for solute translocation varies. The H+-solute stoicheiometry has been measured only for a few transport processes (Rottenberg, 1976; Ramos and Kaback, 1977a,b; Konings and Booth, 1981; West and Mitchell, 1973; Booth et al., 1979; ten Brink and Konings, 1980). Some solutes appear to be symported with a variable proton stoicheiometry : organic acids such as lactate in Escherichia coli (ten Brink and Konings, 1980) and in Streptococcus cremoris (ten Brink and Konings, 1982), and glucose 6-phosphate, amino acids, and lactose in E. coli (Ramos and Kaback, 1977a,b). The information that is currently available about the driving forces for secondary solute transport indicates that at neutral pH most solutes are taken up by the mechanisms shown in examples 2,3, and 4 of Fig. 3, with one or two protons. It seems, therefore, fair to state that, on average, the extrusion of about one or two protons are required to pay for every solute molecule taken up. This energy requirement for solute transport can be expressed in ATP equivalents by assuming that hydrolysis of ATP by the ATPase leads to the extrusion of two or three protons. The average cost for secondary solute transport is then about 0.5-1 .O ATP equivalent.

C. GROUP TRANSLOCATION

The distinctive feature of group translocation types of solute transport is that, concomitant with transport, a chemical modification of the solute occurs resulting in the appearance of the product in the cytoplasm. At

ENERGY FLOW IN BACTERIA

135

present the only thoroughly studied group translocation system in bacteria is the phosphoenolpyruvate (PEP)-dependent sugar phosphotransferase system (PTS) (Robillard, 1982). This system catalyses the transport of various sugars according to the reaction: Sugarout+ PEP

EI Eii

HPr, Mgz+

Sugar P,.

+ Pyruvate

The overall transport which is catalysed can be broken down into several independent phosphoryl-group transfer steps, as is schematically shown in Fig. 4. Several proteins are involved in this process. The soluble proteins, enzyme I and the histidine protein (HPr), are active in the transfer of phosphoryl groups from PEP to the sugar-specific membrane-bound enzymes 11. Enzyme I is phosphorylated by PEP to form phospho-enzyme I and pyruvate. Phospho-enzyme I transfers the phosphoryl group to HPr. The enzymes I1 contain sugar-specific binding sites and are active in the translocation, and subsequent phosphorylation, of the substrate. In several cases a third soluble protein, factor 111, has been shown to be involved in the transfer of the phosphoryl group from HPr to the membrane-bound enzyme 11.

i””“‘

i

FIG. 4. Schematic presentation of a PEP-dependent phosphotransferasesystem in bacteria. I, enzyme I, IIA and IIB, enzymes 11; 111, factor 111; HPr, histidine protein; S, solute transported by a secondary transport system.

K. J. HELLINGWERF AND W. N. KONINGS

136 D.

ATP-DEPENDENT SOLUTE

TRANSPORT SYSTEMS

In Gram-negative bacteria, transport systems sensitive to osmotic shock are found, in which periplasmic binding proteins play an essential role. The direct energy source for these transport systems appears to be an energy-rich phosphate intermediate, possibly acetyl phosphate (Hong et al., 1979). Such binding-protein-dependent transport systems are found for some amino acids and for phosphate in a number of bacteria (Boos, 1974). There is also evidence for the existence of phosphate-bond energydependent transport systems without the involvement of binding proteins (Heefner and Harold, 1980; Rhoads and Epstein, 1977). These systems are found in Gram-positive and Gram-negative bacteria. Examples are sodium transport in Streptococcus faecalis (Heefner and Harold, 1980) and the potassium ion TrkA transport system in E. coli (Rhoads and Epstein, 1977).

E. THE GENERATION OF PROTON MOTIVE FORCE BY E N D PRODUCT EFFLUX

Strictly fermentative bacteria have no electron transfer systems that function as proton pumps. In these organisms a proton motive force can be generated by the membrane-bound ATPase complex. Such a generation of the proton motive force by ATP hydrolysis would consume a considerable fraction of the ATP formed by substrate level phosphorylation with the consequence that less ATP is available for biosynthetic purposes. In order to avoid this drain of ATP these organisms have developed transport systems that mediate the efflux of end products of metabolism in symport with protons. In this way the reverse of the uptake process occurs: in the uptake process the energy of the electrochemical proton gradient is converted into the energy of a solute gradient; in the efflux process the energy of the end product gradient is converted into an electrochemical proton gradient (Michels et al., 1979). Efflux of end products in symport with a proton(s) will lead to the generation of an electrical potential if net charge is translocated, in the generation of a pH gradient if protons are translocated and in the generation of both an electrical potential and a ApH if net charge and protons are translocated (see Fig. 5). Since metabolic end products are continuously produced internally during fermentation, a continuous efflux of end products occurs and consequently a continuous generation of a proton motive force is achieved. As a result, carrier-mediated efflux of end products can contribute significantly to the metabolic energy during fermentation.

137

ENERGY FLOW IN BACTERIA ACID

ZH'

t FIG. 5 . Energy recycling during homolactic fermentation of glucose. Glucose is taken up by the PEP-dependent sugar transport system; lactate is excreted in symport with n protons. It is assumed that hydrolysis of ATP leads to the translocation of two protons. Secondary transport of solutes occurs in symport with i protons or charges.

It is important to realize that the maximum value of the proton motive force generated by this efflux process is limited by the chemical potential gradient of the end product, in a similar way as has been described for the uptake processes. An increased concentration of the end product in the medium will thus lead to a decrease of the end product gradient (unless the internal concentration becomes extremely high) and consequently the proton motive force generated by this efflux process will decrease. At high external concentrations of end product the organism depends on ATP hydrolysis for the generation of a proton motive force. Since, under these conditions, the inwardly directed proton motive force would prevent carrier-mediated efflux of end products, a continuous excretion can proceed only if the H+/end product stoicheiometry is lowered, for instance by translocating the end product electroneutrally (e.g., by a passive process). Experimental evidence for such a proton motive force-generating mechanism has been reported for E . coli (ten Brink and Konings, 1980) and for the homolactic fermentative bacterium Streptococcus cremoris (Otto et al., 1980). In the latter organism, glucose and lactose are taken up by a PEP-dependent phosphotransferase system and subsequently metabolized quantitatively to lactate. The concentration of lactate internally can

138

K. J. HELLINGWERF AND W. N. KONINGS

reach concentrations as high as 200 mM. The efflux of lactate is a carriermediated process which occurs in symport with protons. Under optimal conditions (external pH above 6.7 and external lactate concentrations below 10 mM) lactate leaves the cell with two protons (see Fig. 5). For each molecule of glucose, therefore, 2 molecules of ATP are formed by substrate level phosphorylation and four protons are excreted together with the lactate. Since two protons per glucose are formed during lactate production, lactate efflux leads to the net translocation of two protons per glucose molecule consumed. Assuming that two protons are taken up per ATP synthesized by the ATPase complex, the contribution to the metabolic energy of the cell by lactate efflux will be, under optimal conditions, one ATP-equivalent per glucose molecule consumed, which means an additional energy gain of 50%.

IV. Regulation by Energy Intermediates In this section we illustrate the importance of the three main energy intermediates via a description of (part of) the processes in which these energy intermediates are involved as regulators of metabolic processes. Such a regulation may be in the form of (1) a covalent modification (phosphorylation, adenylylation, etc.), (2) a redox interconversion of a particular enzyme, (3) allosteric modulation of the activity of an enzyme via binding of the regulator to a regulatory site, or (4) a change in the concentration of a (co-)substrate of an enzyme. Space and intellectual limitations force us to ignore in this report several aspects of such regulation. Both at the level of gene expression (Magasanik, 1978) and of the flux rates through metabolic pathways such as glycolysis, the citric acid cycle, and amino acid biosynthesis (Krebs and Beavo, 1979), regulations by the energy intermediates are known, but these will not be discussed. Instead, we have restricted ourselves to regulations described in the field of energy metabolism. The examples are taken mainly from metabolic processes in prokaryotes combined, where appropriate, with regulations described for eukaryotes.

A . REGULATION BY REDOX POTENTIALS

The three main intracellular redox couples of bacteria are NADH/NAD+, NADPH/NADP+, and oxidized/reduced thioredoxin. Under some conditions other important redox intermediates are functional, such as glu-

ENERGY FLOW IN BACTERIA

139

tathione or quinones, but these will not be discussed here. The redox potentials of the three couples mentioned above are closely linked: the free energy of an electrochemical potential gradient of protons is used in the energy-linked transhydrogenase reaction to lower the redox potential of the NADPH/NADP+ couple with respect to the NADH/NAD+ couple (Hanson and Rose, 1980), and the reduced form of thioredoxin is formed via an NADPH-dependent thioredoxin oxidoreductase (Holmgren, 198 1). Therefore, the redox potential of the thioredoxin couple is between those of the NADH/NAD+ and the NADPH/NADP+ couple. The latter redox couple has the most negative redox potential; it can differ with as much as 6 kT mol-I from the redox potential of the NADH/NAD+ couple through the action of the energy-linked transhydrogenase (Rydstrom et al., 1981). Regulation of the activity of enzymes by the redox potential can occur by an allosteric interaction with either the reduced or oxidized form of NAD(P). Examples are glyceraldehyde phosphate dehydrogenase, citrate synthase, and isocitrate dehydrogenase (Lehninger, 1975). Another form of regulation by the redox potential can be a control of the redox state of redox-sensitive groups such as dithiol groups, and prosthetic groups such as flavins, quinones, or Schiff bases. The rate at which the redox conversion of the various groups in enzymes takes place greatly depends on the accessibility and mid-point potential of the redox-sensitive groups and the chemical nature of the redox reagent. In general, dithioVdisulphide-controlled enzymes show the highest sensitivity towards the thioredoxin couple, whereas flavin-dependent enzymes are more sensitive towards the redox potential of the dinucleotide redox couples. In some micro-organisms multiple forms of thioredoxin may occur (Schmidt, 1980), possibly each with their specific redox (mid-point) potential. The spectrum of action of thioredoxin in bacteria was first thought to be in the area of nucleic acid synthesis only. However, in R. sphaeroides, reactions in the intermediary metabolism also were shown to be thioredoxin dependent (Clement-Metral, 1980). In chloroplasts at least six metabolic enzymes are regulated by at least one of the three thioredoxins (Buchanan, 1979). The reduced form of thioredoxin, which accumulates in the light, activates enzymes such as phosphoribulokinase, fructose 1,6bisphosphatase, and the ATPase (see also Underwood and Gould, 1980; Moroney et al., 1980). Several enzymes have been described that are liable to regulation by the redox potential, although their reactivity to the three major redox couples is not known. (1) The activity of succinate dehydrogenase in beef heart mitochondria, which contains a flavin prosthetic group, has been reported to be dependent on the redox potential (Gutman et af., 1980). The mid-point potential of the flavin group changes from -3 to -200 mV

140

K. J. HELLINGWERF AND W. N. KONINGS

during transition of the active to the inactive state, due to binding of oxaloacetate to the enzyme. (2) The activity of the membrane-bound energy-transducing ATPase complex of bacteria can, just as in chloroplasts, be modulated with redox mediators and/or SH group-specific reagents (Underwood and Gould, 1980; Moroney et al., 1980; BaccariniMelandri et al., 1979; Mills et al., 1981). In Rhodopseudomonas capsulata, optimal activity is only observed between 50 and 100 mV (Baccarini-Melandri et al., 1979). (3) The phosphorylation of one of the chloroplast-membrane proteins which is involved in the regulation of channelling of excitation energy between the two photosystems (Bennett et al., 1980; Horton et al., 1981) has been reported to depend on the redox state of a plastoquinone, which is present in the chloroplast membrane. (4) The rate of turnover of the PEP-dependent group-translocation systems (Robillard and Konings, 1982) and protodsugar and proton/amino acid symport systems (Konings and Robillard, 1982) in E. coli has recently been shown to be dependent on the redox state of the respective carriers. These observations will be discussed in Section 1V.C and are analogous to findings in eukaryotic systems: neutral amino acid transport in thymocytes (Kwock, 1981); Ca2+transport in hepatocytes (Sies et al., 1981); adenine nucleotide translocation in mitochondria (Boos and Schlimme, 1981). (5) Distinct and pronounced dependencies of phototaxis and growth of micro-organisms on the redox potential have been reported (Nultsch, 1970).

I

I

B . REGULATION BY PHOSPHORYLATION POTENTIALS

The phosphorylation potentials of three groups of intermediates in particular are important in regulation: (1) the four couples of the ribonucleic acid bases, ATP/ADP, GTP/GDP, CTP/CDP, UTP/ADP, (2) guanosine tetra- and pentaphosphate (ppGpp and pppGpp), and (3) a component of the phosphotranferase system (F-111). With respect to regulation, the phosphorylation potentials of the last two are important only with respect to their effect on the concentration of ppGpp and F-111, and these intermediates are the only reactive species in the regulation (i.e., no phosphorylation/dephosphorylation occurs during regulation). In contrast, with regards to the nucleotides (specifically the ADP/ATP couple), examples are known in which the regulation is exerted via phosphorylation-dephosphorylation. The enzyme nucleotide diphosphate kinase is rather aspecific and can convert all nucleotide diphosphates into the triphosphates with ATP as

i

ENERGY

FLOW IN BACTERIA

141

phosphate donor. As a result, the phosphorylation potentials of all four nucleotide diphosphate/triphosphate couples in bacteria are approximately the same (Lehninger, 1975). If a difference exists (most probably during anabolic growth limitation), the phosphorylation potentials of the adenine nucleotides are highest since these are the nucleotides of which the triphosphate form is generated during catabolism. Through the phosphorylation potential of adenine nucleotides, and in particular of ATP, the activities of enzymes such as glutamine synthetase can also be modulated by adenylylation-deadenylylation (Stadtman et al., 1970). The phosphorylation potential of the AMP/ADP couple also can play an important role in metabolism since many regulatory reactions of AMP (and its metabolic product, cyclic AMP) are known. We will not discuss these aspects further. Regulation of the phosphorylation potential of ppGpp (pppGpp) [or the intracellular concentration of (p)ppGpp] is quite complex. The (p)ppGpp is synthesized from ATP and GDP (GTP), via two independent pathways (Atherly, 1979). The pentaphosphate is rapidly hydrolysed by at least five different enzymes (Sommerville and Ahmed, 1979). The degradation of (p)ppGpp appears to be energy dependent. It has been postulated that both ATP and one of the ribosomal proteins are necessary for its hydrolysis (Sy, 1980), but a claim for the involvement of the proton motive force also exists (TCtu et al., 1980). The degree of phosphorylation of factor I11 (the intermediate in sugar phosphorylation during group translocation between HPr and the membrane-bound sugar-specific enzymes-11) is mainly determined by the availability of sugars to be transported by the specific transport systems (Dills et al., 1980). Usually the supply of phosphate from PEP is not the limiting factor. Examples of regulation of energy conversion by the phosphorylation potential in eukaryotes are widely available in the literature. The observations of the phosphorylation-dephosphorylation of key enzymes of glycolysis and of fat metabolism already belong to the textbook examples of biochemistry (Krebs and Fisher, 1955; Hardie, 1981; Wang and Koshland, 1978). Phosphorylation of membrane proteins has been shown to affect the distribution of photons between photosystems I and I1 in chloroplasts (Allen, 1983). Also in prokaryotes, regulations of energy-transducing processes by the phosphorylation potentials and energy-dependent phosphorylations of membrane proteins have been reported (Spudich and Stoeckenius, 1980). The activity of one of the potassium translocation systems from E. coli

142

K. J. HELLINGWERF AND W. N. KONINGS

(TrkA) is modulated by the phosphorylation potential of ATP (Rhoads and Epstein, 1977; Bakker, 1980). This transport system facilitates potassium uniport and is driven by the A$ only (Bakker, 1980). The phosphorylation potential of ATP is also involved in regulation of processes that are only indirectly related to energy conversion. For instance, in E. coli the choice between the synthesis of host or virus proteins is dependent on the magnitude of the phosphorylation potential (Wagner and Schweiger, 1980). At low potentials the synthesis of virus proteins continues, whereas translation of host mRNA halts. The mechanism of this process is still unresolved but it may have to do with the observed relation between the magnitude of the phosphorylation potential of ATP and the error frequency in protein synthesis (Jelenc and Kurland, 1979). The regulatory role of ppGpp in bacteria has been extensively discussed in the literature (e.g., Bridger and Paranchych, 1979). The ppGpp was even called (together with CAMP) an intracellular “alarmosome” (Primakoff and Artz, 1979). Here we will restrict ourselves to a summary of the various processes, known to be affected by ppGpp: gene expression (e.g., the lactose, tryptophan, and histidine operon), glycogen production (Taguchi and Katsuki, 1978), modulation of various enzyme activities (e.g., glucosephosphate isomerase, acetyl-CoA carboxylase, phosphoribosyl-ATP synthase, PEP carboxylase, and ADP-glucose pyrophosphorylase), phospholipid assembly, and intracellular proteolysis (Bridger and Paranchych, 1979). Regulation of the activity of several secondary transport systems occurs by a direct inhibition of carrier proteins by the unphosphorylated form of factor 111, (Scholte et al., 1982). In addition a stimulatory effect of the phosphorylated form of factor 111 on adenylate cyclase, the enzyme that synthesizes cAMP occurs. The activated adenylate cyclase produces increased cAMP levels. This cAMP binds to the receptor protein and activates the synthesis of catabolic enzyme systems for non-PTS substrates (Scholte et al., 1982; Nelson et al., 1983). The result of this regulation is that, in the presence of PTS sugars, the synthesis of carrier proteins for certain non-PTS substrates is repressed whereas in their absence it is initiated. C. REGULATION BY ELECTROCHEMICAL POTENTIAL GRADIENTS

The primary electrochemical potential gradient in bacteria is a gradient of protons, generated by proton translocation coupled to electron transfer chain redox reactions, light absorption, efflux of metabolic end products, and, under some conditions, ATP hydrolysis (Mitchell, 1968; Harold,

ENERGY FLOW IN BACTERIA

143

1972; Skulachev, 1977; Konings et al., 1981). This electrochemical gradient of protons is very important in bacterial physiology, and numerous processes are driven and/or regulated by the proton motive force (see Table 1). Secondary transport systems create gradients of other solutes: in most bacteria K+ and Mg2+are accumulated intracellularly whereas Na+ and TABLE 1 . Summary of processes currently assumed to be energized or regulated by the proton motive force in bacteria Process ATP synthesis Pyrophosphate synthesis Electron transfer (“secondary”) transport Phosphotransferase system Synthesis of CAMP Degradation of ppGpp Protein phosphorylation Nitrogen fixation Methane synthesis Reversed electron transfer Transhydrogenase activity Flagellar rotation Flagellar synthesis Chemotaxis Activity of autolytic enzymes Bacterial transformation Conjugation Phage infection Virus infection Intermembrane lipid transfer (from inner to outer membrane) Processing of membrane proteins Transmembrane transport of proteins Activity of colicins Phase transitions in the cytoplasmic membrane Interaction between binding proteins and the cytoplasmic membrane Cellulose synthase Conformation of membrane proteins

References Futai and Kanazawa (1983) Baltscheffsky et al. (1966); Shakhov et al. (1982) Mitchell (1968) West and Mitchell (1973) Robillard and Konings (1981) Peterkofsky and Gazdar (1979) Tetu et al. (1980) Spudich and Stoeckenius (1980) Haaker et al. (1982); Laane et al., (1979) Mevel-Ninio and Valentine (1975) Mitchell (1968) Hojeberg and Rydstrom (1977) Khan and Macnab (1980) Galperin et al. (1982) Taylor (1983) Jolliffe et al. (1981) Van Nieuwenhoven et al. (1982); Chaustova ef al. (1980) Grinius and Berzinskiene (1976) Labedan and Goldberg (1979) Wagner el al. (1980) Donohue-Rolfe and Schaechter (1980) Smith (1980); Zimmermann, R. ef at. (1982) Galperin et al. (1982); Zimmerrnan and Neupert (1980) Braun et al. (1980) Tecoma and Wu (1980) Richanme (1982) Delmer ef al. (1982) Le Grimellec et al. (1982)

144

K. J. HELLINGWERF AND W. N. KONINGS

Ca2+ are usually extruded. To some extent ion transfer, which is not facilitated by specific transport proteins, also may contribute to the establishment of secondary ion gradients. Regulation by these ion gradients can be of a very basic kind. The activities of many enzymes are regulated by the concentration of H+, K + , or Mg2+.These regulatory effects can be exerted on proteins in the cytoplasm such as proteins involved in glycolysis, protein synthesis, or even intracellular proteolysis (St. John and Goldberg, 1980), and also on intrinsic membrane proteins. Examples of the latter are found in eukaryotes: the H+/Kf exchange carrier of mitochondria (Shi et al., 1980) and several algal transport systems (U. Zimmermann et al., 1982). Another form of regulation has been observed in a potassium transport system of E . coli. The activity of the Kdp potassium transport system is dependent on the intracellular osmotic pressure (Laimins et al., 198 1). This intracellular osmotic pressure is generated mainly by secondary transport of cations and solutes such as proline or glutamate (Measures, 1975). Also other processes can depend on the intracellular osmotic pressure. In Mycoplasma gallisepticum it has been shown that the cell volume is controlled by the proton motive force (Rottem et af., 1981). A large body of indirect evidence is available in support of the suggestion that the activity of adenylate cyclase in many bacteria is directly regulated by either the proton motive force or its electrical potential component (Peterkofsky and Gazdar, 1979). The latter enzyme regulates, together with cAMP phosphodiesterase, the level of the “alarmosome” cAMP (Primakoff and Artz, 1979). Other structural or functional properties of the cytoplasmic membrane which are dependent on, or controlled by, the proton motive force are (1) phase transitions of the cytoplasmic membrane (Tecoma and Wu, 1980), (2) the activity of colicins (Braun et al., 1980), (3) the transfer of lipids from the cytoplasmic to the outer membrane (Donohue-Rolfe and Schaechter, 1980), (4) the activity of autolytic enzymes in Bacillus subtilis (Jolliffe et al., 1981), ( 5 ) the conformation of membrane proteins (Le Grimellec et al., 1982), (6) the processing of proteins (Smith, 1980; Daniels et

al., 1981; R. Zimmermann et al., 19821, and (7) even the transmembrane transfer of proteins (Galperin et al., 1982; Zimmermann and Neupert, 1980). In the latter two processes the possibility exists that the proton motive force functions as a driving force for the translocation process (Inouye et al., 1982; Weinstein et al., 1982). In that case they would fall beyond our definition of regulation. For the other processes, it is completely unknown how the proton motive force exerts its effect(s). One aspect of regulation by the proton motive force is intimately related to regulation by the redox potential (Robillard and Konings,

ENERGY FLOW IN BACTERIA

145

1982a,b). A number of membrane-bound proteins have been shown to contain redox-sensitive dithiol/disulphide groups which are located at fixed sides of the membrane. The activity of these proteins depends strongly on the redox state of these groups. This redox state will be determined by the redox state of the environment, i.e., the external medium, the cytoplasm, and/or the milieu of the cytoplasmic membrane. The redox state also will be determined by the electrical potential or the pH gradient across the membrane, if electron carriers can interact with the redox centre or if reducing equivalents can be transferred to the redox centre (Walz, 1979). Consequently, the generation of a proton motive force, internal negative and alkaline, will lead to a more oxidized state of a redox centre located at the inner surface and a more reduced state of a redox centre at the outer surface (Robillard and Konings, 1982a,b; Poolman et al., 1983). Such phenomena have recently been demonstrated to occur in E . coli for the lactose and proline proton symport systems (Konings and Robillard, 1982) and glucose transport via the PEP phosphotransferase system (Robillard and Konings, 1982a,b). These results have led to a model for solute transport, and for other energy-transducing proteins which contain redox-sensitive disulphide/dithiol groups as is presented in Fig. 6. One form of regulation, in the literature often referred to as “gating” or “a threshold phenomenon” (Lanyi and Silverman, 1979), requires some comment. These terms originate from the description of the form of flowforce diagrams, in which the rate of a reaction is plotted as a function of the thermodynamic force driving that reaction. A process with a negligible activity at low driving forces and a strongly increased activity at forces that are just above a threshold value often is referred to as a

&T-( B-SH

S BH S

SH

FIG. 6. Proposal for a redox-controlled H+/solute symport. A is the solute being transported; B is a basic group in the vicinity of a dithiolldisulphide group. ) is a high-affinity binding site and ) a low affinity binding site. After Robillard and Konings (1982a).

146

K. J. HELLINGWERF AND W.

N. KONINGS

“gated” process. This gated phenomenon is considered to be the result of “regulation” of the process (Lanyi and Silverman, 1979; Shioi et al., 1981; van der Meer et al., 1980). However, the reason for such gating can, in some processes, be the large amount of free energy that is required before a certain reaction can proceed. For instance, in intact cells of Streptococcus faecalis a large proton motive force is required before ATP synthesis starts (Maloney, 1977). This gating cannot be considered to be the result of regulation. The situation is different in those circumstances in which an inhibitor subunit functions as an ordoff switch for a membrane-bound ATP synthetase, like in the chloroplast ATPase (Bakker-Grunwald and van Dam, 1974; Harris and Crofts, 1978). In that case, the magnitude of the transmembrane electrochemical potential affects the location of the inhibitor subunit in the ATPase complex, and thereby its rate of turnover. Therefore such processes are true examples of regulation. Recently, gating phenomena have also been observed for solute transport in R. sphaeroides (Elferink e f al., 1983a). This gating appears to be due to a direct interaction between the solute carrier and electron transfer in a cyclic or linear electron transfer chain (Elferink et al., 1983a,b, 1984). A similar mechanism presumably also operates in light-driven ATP synthesis (Baccarini-Melandri et al., 1981). Although the (bio)chemical mechanism behind the direct interaction of the solute carrier with the electron transfer chain is not known, it is a form of regulation that does fit into our definition of regulation: redox changes in the electron transfer chain modulate the rate of a process in which redox equivalents are not substrates and/or products of the reaction(s). It is tempting to speculate that this direct interaction between electron transfer and solute uptake can be explained in terms of the redox-model for energy transduction (Robillard and Konings, 1982a,b).

V. Homoeostasis in the Magnitude of Free Energy Intermediates In energy transduction in bacteria, regulation is such that often a large degree of homoeostasis occurs. The best known example is the intracellular pH in bacteria (Padan et al., 1981; Schuldiner and Padan, 1981). This parameter is kept at a constant value at near neutrality in bacteria under a broad range of conditions. Primary proton pumps catalyse the extrusion of protons from the cytoplasm which is re-acidified via the action of (electrogenic) cation/proton antiporters. The sodiudproton antiporter is of primary importance for this homoeostasis in E. coli at alkaline pH

ENERGY FLOW IN BACTERIA

147

values (Zilberstein et al., 1982), but the functioning of other antiporters has been suggested too (Plack and Rosen, 1980). The precise mechanism responsible for the regulation of the intracellular pH is not known at the moment. Most probably (Zilberstein et al., 1982), the intracellular pH affects the rate at which the primary proton pumps, the antiporters, and all other components involved in the regulation (such as electrogenic uniporters) can function. A second mechanism of homoeostasis can be observed in a number of bacteria (such as E. coli and R . sphaeroides) with respect to the magnitude of the proton motive force across the cytoplasmic membrane of these bacteria. Processes consuming proton motive force in these bacteria, like ATP synthesis and solute transport in non-sulphur purple bacteria (Elferink et al., 1983a,b, 1984; Baccarini-Melandri et al., 1981) and lactose and proline uptake in E. coli (Elferink el al., 1984), are regulated by the rate of the primary proton motive force-generating proton pumps (either linear or cyclic electron transfer chains). This causes an increased rate of proton motive force consumption with increasing rates of proton motive force generation. Figure 7 shows that the uptake of alanine is measurable only at that light intensity at which the magnitude of the proton motive force decreases, indicating that indeed solute transport (andor ATP synthesis) is responsible for the lowering of the proton motive force at the higher light intensities. It is worth noting that in bacteria cyclic processes of ion translocation also can occur, for instance due to the simultaneous 70,r

1

1150 light on

I

50

100

->

I

c

-E

3 a

30 50

0

c

20-

-0

*

10O

0

L

0 Timelmin)

FIG. 7. Simultaneous measurement of alanine uptake (0) and membrane potential ( 0 )in R . sphaeroides. For experimental details see Elferink et al. (1983a). 1, 11, and I11 represent increasing light intensities. (M. G. L. Elferink, unpublished results.)

148

K. J. HELLINGWERF AND W. N. KONINGS

activity of electrogenic potassium uptake and electroneutral proton/potassium exchange. Such a cycle of potassium translocation acts as a “futile” cycle which dissipates the proton motive force and may contribute to proton motive force consumption to a variable degree, as a function of light intensity. The relative contribution to proton motive force consumption by ATP synthesis and solute transport under these conditions has not been quantified. Homoeostasis of the proton motive force strongly depends on the physiological condition of the cells. In E . coli homoeostasis can only be observed in cells that are physiologically in an optimal condition (Elferink et al., 1984) and it is rapidly lost on treatment of the cells with EDTA or dicyclohexylcarbodiimide. It is not known whether this homoeostasis of the proton motive force also occurs in bacteria such as S. cremoris which do not possess primary, electron transfer-linked, proton pumps. It was observed, however, that the magnitude of the proton motive force varied only very slightly when the growth rate of this bacterium, in a continuous culture, was varied almost 10-fold (Otto et al., 1983). A similar form of homoeostasis was observed with respect to the intracellular phosphate potential in S. cremoris (R. Otto and co-workers, unpublished results). In chloroplasts a well-resolved mechanism is functional in order to control the distribution of light quanta over the two photosystems (Bennett et al., 1980; Allen, 1983). In this process the redox state of a plastoquinone and the phosphate potential together determine the degree of phosphorylation of a light-harvesting complex via the activity of a membrane-bound protein kinase. This degree of phosphorylation regulates the distribution of light quanta over the two photosystems. Uneven distribution of light quanta over the two photosystems changes the net rate of linear electron transfer, and thus the redox state of plastoquinone. In this way an effective feedback system is obtained for optimal rates of linear electron transfer, and over-excitation of one of the two photosystems can be prevented.

VI. Conclusions and Perspectives The summarizing scheme of Fig. 1 shows clearly that the various forms of free enthalpy gradients in bacterial energy transduction are strongly interrelated. One can visualize the energy-transducing systems as communicating vessels. The energy flow from one vessel to another can be controlled by regulatory mechanisms. Several of such regulatory mechanisms have been described and discussed in this article, and other yet unknown mechanisms may exist. The molecular details of most known regulatory

ENERGY FLOW IN BACTERIA

149

mechanisms have not yet been unravelled, and further research is needed to obtain a clear picture. The many lines of communication between the three main energy intermediates make it very likely that any input of free energy into a bacterial cell will be transformed into various forms. A good picture of the energy status of a bacterial cell population therefore requires quantitative information about the three main energy intermediates. In general it is dangerous to draw conclusions on the basis of quantitative measurements of only one potential. A similar reservation applies to studies on the coupling or interaction between energy-transducing processes. Since a quantitation of the three energy intermediates of most bacterial populations is hardly possible at this moment, more simple systems often have been chosen to study the interactions between energy-transducing systems. For instance the interaction between electron transfer systems and secondary solute transport systems, or the membrane-bound ATPase system, has been studied intensively in cytoplasmic membrane vesicles and in chromatophores, respectively (Kaback, 1974; Oelze and Drews, 1972). Growth of bacteria can be described satisfactorily with mosaic nonequilibrium thermodynamics (MNET) (Westerhoff et af., 1982; Hellingwerf et af., 1982). So far in these descriptions the regulatory mechanisms such as mentioned in this report have not been taken into consideration. It will be of interest to see if the inclusion of regulatory phenomena in the mosaic non-equilibrium thermodynamic analysis will lead to an even better description of bacterial growth (cf. Westerhoff et al., 1982). Finally, it should be noted that many of the interrelations and regulatory processes described here can be helpful in understanding the functioning of bacteria in their natural environments. Several environmental parameters such as pH value, redox potential, ion concentration and composition, light intensity, and temperature will influence directly or indirectly the free enthalpy gradients. It is tempting to speculate what the effects can be of changes in the environmental conditions on the energy status of a particular group of bacteria and subsequently which response of the bacteria can be anticipated (Konings and Veldkamp, 1980, 1983).

VII. Acknowledgements We like to thank Drs. J. S. Lolkema and H. V. Westerhoff for useful suggestions concerning this manuscript and M. G. L. Elferink for the unpublished results presented in this manuscript. We are grateful to Mrs. M. Th. Broens-Erenstein and M. Pras for the help in the preparation of this manuscript. The studies performed in the laboratory of the authors were

150

K. J. HELLINGWERF AND W. N. KONINGS

supported by the Netherlands Organization for the Advancement of Pure Scientific Reseach (ZWO).

REFERENCES

Allen, J. F. (1983). Trends in Biochemical Sciences 8, 369. Atherly, A. G. (1979). Journal of Bacteriology 138, 530. Baccarini-Melandri, A., Melandri, B. A., and Hauska;G. (1979). Journal of Bioenergetics and Biomembranes 11, 1 . Baccarini-Melandri, A., Casadio, R., and Melandri, B. A. (1981). Current Topics in Bioenergetics 12, 192. Bakker, E . P. (1980). In “EBEC Short Reports,” pp. 283-284. Patron Editore, Bologna. Bakker-Grunwald, T., and van Dam, K. (1974). Biochimica et Biophysica Acta 347, 290. Baltscheffsky, H., Von Stedingk, L.-V., Heldt, H.-W., and Klingenberg, M. (1966). Science 153, 1120. Bennett, J., Steinback, K. E., and Arntzen, C. J. (1980). Proceedings of the National Academy of Sciences of the United States of America 77, 5253. Boos, K. S., and Schlimme, E. (1981). FEBS Letters 127, 40. Boos, W. (1974). Annual Review of Biochemistry 43, 123. Booth, 1. R., Mitchell, W. J., and Hamilton, W. A. (1979). Biochemical Journal 182, 687. Braun, V., Frenz, J., Handke, K., and Schaller, K. (1980). Journal of Bacteriolo g y 142, 162. Bridger, W. A., and Paranchych, W. (1979). Trends in Biochemical Sciences 4, 176. Brown, I. I., Galperin, M. Yu., Glagolev, A. N., and Skulachev, V. P. (1983). European Journal of Biochemistry W, 345. Buchanan, B. B. (1979). In “Modulation of Protein Function” (D. E. Atkinson and C. F. Fox, eds.), Ch. 7, pp. 93-112. Academic Press, New York. Chaustova, L. P., Grinius, L. L., Griniuviene, B. B., Jasaitis, A. A., Kadziauskas, J. P., and Kiausinyte, R. J. (1980). European Journal of Biochemistry 103, 349. Clement-Metral, J. D. (1980). FEBS Letters 101, 116. Daniels, C. J., Bole, D. G., Quay, S. C., and Oxender, D. L. (1981). Proceedings of the National Academy of Sciences of the United States of America 78,5396. Deamer, D. W., and Nichols, J. W. (1983). Proceedings of the National Academy of Sciences of the United States of America 80, 165. Delmer, D. P., Benziman, M., and Padan, E. (1982). Proceedings of the National Academy of Sciences of the United States of America 79, 5282. Dills, S. S., Apferson, A., Schmidt, M. R., and Saier, M. H. (1980). Microbiological Review 44, 385. Dimroth, P. (1982a). European Journal of Biochemistry 121, 435. Dimroth, P. (1982b). European Journal of Biochemistry 121, 443. Donohue-Rolfe, A. M., and Schaechter, M. (1980). Proceedings of the National Academy of Sciences of the United States of America 77, 1867.

ENERGY FLOW IN BACTERIA

151

ELferink, M. G. L., Friedberg, I., Hellingwerf, K. J., and Konings, W. N. (1983a). European Journal of Biochemistry 129,583. Elferink, M. G. L., Hellingwerf, K. J., Nano, F. E., Kaplan, S., and Konings, W. N. (1983b). FEBS Letters 164, 185. Elferink, M. G. L., Hellingwerf, K. J., van Belkum, M., Poolman, B., and Konings, W.N. (1984). FEMS Microbiology Letters, 21, 293. Futai, M., and Kanazawa, H. (1983). Microbiological Reviews 47, 285. Galperin, M. Yu.,Dibrov, P. A., and Glagolev, A. N. (1982). FEBS Letters 143, 319. Grinius, L.L., and Berzinskiene, J. (1976). FEBS Letters 72, 151. Gutman, M., Bonomi, F., Pagani, S., Cerletti, P., and Kroneck, P. (1980). Biochimica et Biophysica Acta 591,400. Haaker, H., Laane, C., Hellingwerf, K. J., Houwer, B., Konings, W. N., and Veeger, C. (1982). European Journal of Biochemistry 127, 639. Hanson, R. L., and Rose, C. (1980). Journal of Bacteriology 141, 401. Hardi, G. (1981). Trends in Biochemical Sciences 6, 75. Harold, F. M. (1972). Bacteriological Reviews 36, 172. Hams, D. A., and Crofts, A. R. (1978). Biochimica et Biophysica Acta 502, 87. Heefner, D. L., and Harold, F. M. (1980). Journal of Biological Chemistry 255, 11396. Hellingwerf, K. J., Bolscher, J. G. M., and Konings, W. N. (1981). European Journal of Biochemistry 113, 369. Hellingwerf, K. J., Lolkema, J. S., Otto, R., Neyssel, 0. M., Stouthamer, A. H., Harder, W., Van Dam, K., and Westerhoff, H. V. (1982). FEMS Microbiology Letters 15, 7. Hojeberg, B., and Rydstrom, J. (1977). Biochemical and Biophysical Research Communications 78, 11883. Holmgren, A. (1981). Trends in Biochemical Sciences 6, 26. Hong, J.-S., Hunt, A. G., Masters, P. S., and Lieberman, M. A. (1979). Proceedings of the National Academy of Sciences of the United States of America 76, 1213. Horton, P.,M e n , J. R., Black, M. T., and Bennett, J. (1981). FEES Letters 125, 193. Inouye, S., Soberon, X.,Franceschini, T., Nakamura, K., Itakura, K., and Inouye, M. (1982). Proceedings of the National Academy of Sciences of the United States of America 79, 3438. Jelenc, P. C. and Kurland, C. G. (1979). Proceedings of the National Academy of Sciences of the United States of America 76, 3174. JoWe, L. K., Doyle, R. J., and Streips, U. N. (1981). Cell 25, 753. Kaback, H. R. (1974). Science 186, 882. Khan, S.,and Macnab, R. M. (1980). Journal of Molecular Biology WS, 599. Konings, W. N., and Booth, I. R. (1981). Trends in Biological Sciences 2, 257. Konings, W. N., and Michels, P. A. M. (1980). Zn “Diversity of Bacterial Respiratory Systems” (C. J. Knowles, ed.), pp. 33-87. CRC Press, Boca Raton, Florida. Konings, W. N., and Robillard, G. T. (1982). Proceedings of the National Academy of Sciences of the United States of America 79, 5480. Konings, W. N., and Veldkamp, H. (1980). I n “Contemporary Microbial Ecology” (D. C. Ellwood, J. N. Hedger, M. J. Latham, J. M. Lynch, and J. H. Slater, eds.), pp. 161-191. Academic Press, London.

152

K. J. HELLINGWERF AND W. N. KONINGS

Konings, W. N., and Veldkamp, H. (1983). Symposium of the Society for General Microbiology 34, 154. Konings, W. N., Hellingwerf, K. J., and Robillard, G. T. (1981). In “Membrane Transport” (S. L. Bonting and J. J. H. H. M. de Pont, eds.), Ch. 10, pp. 257283. ElseviedNorth Holland, Amsterdam. Krebs, E. G. and Fisher, E. H. (1955). Journal of Biological Chemistry 216, 121. Krebs, E. G., and Beavo, J. A. (1979). Annual Review of Biochemistry 48, 923. Kwock, L. (1981). Journal of General Physiology 106, 279. Laane, C., Krone, W., Konings, W. N., Haaker, H., and Veeger, C. (1979). FEBS Letters 103, 328. Labedan, B., and Goldberg, E. B. (1979). Proceedings of the National Academy of Sciences of the United States of America 76, 4669. Laimins, L. A., Rhoads, D. B., and Epstein, W. (1981). Proceedings of the National Academy of Sciences of the United States of America 78, 464. Lancaster, J. R., and Hinkle, P. C. (1977). Journal of Biological Chemistry 252, 7662. Lanyi, J. K. (1979). Biochimica et Biophysica Acta 559, 377. Lanyi, J. K., and Silverman, M. P. (1979). Journal of Biological Chemistry 254, 4750. Le Grimellec, C., Lajeunesse, D., and Rigaud, J. L. (1982). Biochimica et Biophysics Acta 687, 28 1 . Lehninger, A. L. (1975). “Biochemistry”. Worth Publishers Inc., New York. Lipmann, F. (1941). Advances in Enzymology 1, 99. Lolkema, J. S., Hellingwerf, K. J., and Konings, W. N. (1982). Biochimica et Biophysica Acta 681, 85. Magasanik, B. (1978). In “The Operon” (J. H. Miller and W. S. Reznikoff, eds.), pp. 373-388. Cold Spring Harbor Laboratory. Maloney, P. C. (1977). Journal of Bacteriology 132, 564. Measures, J. C. (1975). Nature, London 257, 398. Mevel-Ninio, M. T., and Valentine, R. C. (1975). Biochimica et Biophysica Acta 376, 485. Michels, P. A. M., Michels, J. P. J . , Boonstra, J . , and Konings, W. N. (1979). FEMS Microbiology Letters 5, 357. Mills, J. D., Mitchell, P., and Schurmann, P. (1981). In “Photosynthesis 11. Electron Transport and Photophosphorylation” (G. Akoyunoglou, ed.), pp. 839-848. Balaban International Science Services, Philadelphia. Mitchell, P. (1968). “Chemiosmotic Coupling and Energy Transduction.” Glynn Research Ltd., Bodmin, England. Mitchell, P. (1981). In “Of Oxygen, Fuels and Living Matter, Part 1” (G. Semenza, ed.), pp. 1-160. John Wiley and Sons Ltd., London. Moroney, J. V., Andreo, C. S., Vallejos, R. H., and McCarthy, R. E. (1980). Journal of Biological Chemistry 255, 6670. Nelson, S. O., Wright, J. K., and Postma, P. W. (1983). The EMBO Journal 2, 715. Nultsch, W. (1970). In “Photobiology of Microorganisms” (P. Halldall, ed.), Ch. 8, pp. 213-251. Wiley, Interscience, London. Oelze, J., and Drews, G. (1972). Biochimica et Biophysica Acta 265, 209.

ENERGY FLOW IN BACTERIA

153

Otto, R. Sonnenberg, A. S. M., Veldkamp, H., and Konings, W. N. (1980). Proceedings of the National Academy of Sciences of the United States of America 77, 5502. Otto, R., ten Brink, B., Veldkamp, H., and Konings, W. N. (1983). FEMS Microbiology Letters 16, 69. Padan, E., Zilberstein, D., and Schuldiner, S. (1981). Biochimica et Biophysica Acta 650, 151. Peterkofsky, A., and Gazdar, C. (1979). Proceedings of the National Academy of Sciences of the United States of America 76, 1099. Plack, R. H., Jr., and Rosen, B. P. (1980). Journal ofBiologica1 Chemistry 225, 3824. Poolman, B., Konings, W. N., and Robillard, G. T. (1983). European Journal of Biochemistry 135,41. Primakoff, P., and Artz, S. W. (1979). Proceedings of the National Academy of Sciences of the United States of America 76, 1726. Ramos, S. , and Kaback, H. R. (1977a). Biochemistry 16, 854. Ramos, S., and Kaback, H. R. (1977b). Biochemistry 16, 4271. Rhoads, D. B., and Epstein, W. (1977). Journal of Biological Chemistry 252, 1394. Richarme, G. (1982). Journal of Bacteriology 149, 662. Robillard, G. T. (1982). Molecular and Cell Biochemistry 46, 3. Robillard, G. T., and Konings, W. N. (1981). Biochemistry 20, 5025. Robillard, G. T., and Konings, W. N. (1982a). European Journal of Biochemistry 127, 597. Robillard, G. T., and Konings, W. N. (1982b). In “Plasmalemma and Tonoplast: Their Functions in the Plant Cell” (D. MarnC, E. MarrC, and R. Hertel, eds.), pp. 3 13-320. Elsevier Biomedical Press, Amsterdam. Rottenberg, H. (1976). FEBS Letters 66, 159. Rottem, S . , Linker, C., and Wilson, T. H. (1981). Journal of Bacteriology 145, 1299. Rydstrom, J., Lee, C. P., and Ernster, L. (1981). I n “Chemiosmotic Proton Circuits in Biological Membranes” (V. P. Skulachev and P. C. Hinkle, eds.). pp. 483-508. Addison-Wesley Publishing Co., London. Schlegel, H. G. (1981). “Allgemeine Mikrobiologie”, 5th Ed. Georg Thieme Verlag, Stuttgart. Schmidt, A. (1980). Archives of Microbiology 127, 259. Scholte, B. J . , Schuitema, A. R. J., and Postma, P. W. (1982). Journal of Bacteriology 149, 576. Schuldiner, S., and Padan, E. (1982). In “Membranes and Transport” (A. Martonosi, ed.), pp. 65-74. Plenum Press, New York. Shakhov, Y. A,, NyrCn, P., and Baltscheffsky, M. (1982). FEBS Letters 146,177. Shi, G . -Y . , Jung, D. W., Garlid, K. D., and Brierly, G. P. (1980). Journal of Biological Chemistry 255, 10306. Shioi, J.-I., Matsuura, S., and Imae, Y. (1981). Journal ofBacteriology 144, 891. Sies, H., Graf, P., and Estrela, J. P. (1981). Proceedings ofthe NutionalAcuderny of Sciences of the United States of America 78, 3358. Skulachev, V. P. (1977). FEBS Letters 74, 1 . Slater, E. C. (1981). Trends in Biochemical Sciences 6, 226. Smith, W. P. (1980). Journal of Bacteriology 141, 1142.

154

K. J. HELLINGWERF AND W. N. KONINGS

Sornrnerville, C. R., and Ahrned, A. (1979). Molecular Genetical Genetics 169, 315. Spudich, J . L., and Stoeckenius, W. (1980). Journal of Biological Chemistry 255, 5501. Stadtrnan, E. R., Ginsburg, A., Ciardi, J. E., Yeh, J., Henning, S. B., and Shapiro, B. M. (1970). Advances in Enzyme Regulation 8, 99. Stanier, R. Y., Adelberg, E. A., and Ingraharn, J. L. (1978). “General Microbiology,” 4th Ed. Macmillan, London. St. John, A. C., and Goldberg, A. L. (1980). Journal of Bacteriology 143, 1223. Sy, J. (1980). Journal of Biological Chemistry 255, 10056. Taguchi, M., and Katsuki, H. (1978). Biochemical Biophysical Research Communications 84, 195. Taylor, B. L. (1983). Annual Review of Microbiology 37, 551. Tecorna, E. S., and Wu, D. (1980). Journal of Bacteriology 142, 931. ten Brink, B., and Konings, W. N. (1980). European Journal OfBiochemistry 111, 59. ten Brink, B., and Konings, W. N. (1982). Journal of Bacteriology 152, 682. Tetu, C., Dassa, E., and Boquet, P. L. (1980). European Journal ofBiochemistry 103, 117. Thauer, R. K . , Jungerrnann, K., and Decker, K. (1977). Bacteriological Reviews 41, 100. Tokuda, H., and Unemoto, T. (1982). Journal ofBiologica1 Chemistry 257, 10007. Tokuda, H., Sugasawa, M., and Unernoto, T. (1982). Journal of Biological Chemistry 257, 788. Underwood, C., and Gould, J. M. (1980). Biochimica et Biophysica Acta 589,297. van der Meer, R., Westerhoff, H. V., and van Darn, K. (1980). Biochimica et Biophysica Acta 591, 488. Van Nieuwenhoven, M. H., Hellingwerf, K. J., Venerna, G., and Konings, W. N. (1982). Journal of Bacteriology 151, 77 1. Wagner, E. F., and Schweiger, M. (1980). Journal of Biological Chemistry 255, 540. Wagner, E. F., Ponta, H., and Schweiger, M. (1980). Journal OfBiological Chemistry 255, 534. Walz, D. (1979). Biochimica et Biophysica Acta 505, 279. Wang, J. Y. J., and Koshland, D. E., Jr. (1978). Journal of Biological Chemistry 253, 7605. Weinstein, J. N., Blumenthal, R., van Renswoude, J., Kempf, C., and Klausner, R. D. (1982). Journal of Membrane Biology 66, 203. West, I. C., and Mitchell, P. (1973). Biochemical Journal 132, 587. Westerhoff, H. V., Lolkema, J. S., Otto, R., and HeUingwerf, K. J. (1982). Biochimica et Biophysica Acta 683, 181. Zilberstein, D., Agmon, V., Schuldiner, S., and Padan, E. (1982). Journal of Biological Chemistry 257, 3687. Zimmermann, R., and Neupert, W. (1980). European Journal of Biochemistry 109,217. Zimmermann, R., Watts, C., and Wickner, W. (1982). Journal of Biological Chemistry 257,6529. Zimmermann, U., Buchner, K.-H., and Benz, R. (1982). Journal of Membrane Biology 67, 183.

Hydrogenase. Nitrogenase. and Hydrogen Metabolism in the Photosynthetic Bacteria PAULETTE M. VIGNAIS. ANNETTE COLBEAU. JOHN C. WILLISON and YVES JOUANNEAU Laboratoire de Biochimie (CNRSIER 235. lnserm U. 191. CEAIIRF). Departement de Recherche Fondamentate. Centre d’Etudes Nucleaires de Grenoble. Grenoble. France

I . Introduction . . . . . . . I1 . The organisms . . . . . . . A . Classification . . . . . . B . Growth properties . . . . . . C . Ecological distribution . . . . . 111. Hydrogen utilization and production by photosynthetic A . Hydrogen as an electron donor . . . B . Hydrogen production . . . . . IV . Hydrogenase . . . . . . . A . General background . . . . . B . Localization . . . . . . . C . Stability and stabilization . . . . D . Structure and molecular properties . . . E . Catalytic properties . . . . . . . . . F . An inducible enzyme . G . Physiological role of hydrogen uptake . . V . Nitrogenase . . . . . . . A . Biochemistry of nitrogenase . . . . B . Regulation of nitrogenase activity . . . C . Regulation of nitrogenase synthesis . . VI . Genetics of hydrogen production and utilization . A . General background . . . . . B . Nitrogenase genetics . . . . . C . Hydrogenase genetics . . . . . ADVANCES IN MICROBIAL PHYSIOLOGY VOL . 26

. . . . .

. . . . .

bacteria

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

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

. . .

. . .

.

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

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

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

156 156 158

160 161 162 162 164 174 174 175 177 179 182

.

.

.

185

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

187 190 190 195 201 204 204 205 209

.

Copyright 0 1985 by Acudemir Press Inr . (London) Lrd . All rights r$reprodurrion in any form reserved

156

PAULETTE

M. VlGNAlS €T AL.

VII. Use of photosynthetic bacteria as biological solar energy converters

.

A. Strain screening and selection . . . . . . . B . Economical substrates . . . . . . . . . C. Cell stabilization by immobilization . . . . . . D. Advantages of using photosynthetic bacteria as hydrogen producers . VIII. Summary and prospects . . . . . . . . . IX. Acknowledgements . . . . . . . . . . References . . . . . . . . . . . . . . . . . . Note added in proof .

. . . .

210 21 1 21.5 216 . 217 . 218 .220 ,220 . 234

I. Introduction Many photosynthetic micro-organisms possess the capacity to produce molecular hydrogen by a process directly linked to their ability to capture light energy. The great potential of H2 for use as a primary or secondary source of energy, and the possibility of transforming solar radiation into stable chemical energy, in the form of hydrogen gas, have greatly stimulated the study of H2 metabolism in photosynthetic organisms. This trend of research is reflected by the tremendous output of relevant literature. Many excellent review articles on H2 metabolism in photosynthetic prokaryotes have been published in the last 5 years (Table 1). The present review is limited to the study of H2 metabolism in photosynthetic bacteria. For the sake of clarity, and because of lack of space, discussion of some aspects has been intentionally kept to a minimum, or omitted. These aspects includqthe energetics of bacterial photosynthesis, the light-harvesting pigments and organization of the reaction centres, and the membrane-bound electron transport chain; all subjects which have been excellently covered in preceding reviews. The intention of this report is to cover recent developments in this field and to discuss these developments in the context of the biochemistry and physiology of the photosynthetic bacteria, and their biotechnological applications.

11. The Organisms

The photosynthetic bacteria are aquatic Gram-negative organisms found in a wide range of environments, including marine and freshwater systems. They represent an important component of these ecosystems because of their ability to utilize solar energy for the fixation of C 0 2and N2. They are photosynthetic prokaryotes possessing bacteriochlorophylls as the characteristic pigments (Pfennig and Truper, 1974; Truper and Pfen-

H2

METABOLISM IN PHOTOSYNTHETIC BACTERIA

157

TABLE 1. Recent literature reviews of photosynthetic prokaryotes and HZmetabolismo Photosynthetic bacteria generally Clayton and Sistrom (1978) Pfennig (1977) Kondratieva (1979) Hydrogen metabolism generally Lien and San Pietro (1975) Schlegel and Schneider (1978) Zajic er a / . (1978) Bishop and Jones (1978) Weaver et al. (1980) Robson and Postgate (1980) Adams et al. (1981) Hydrogen metabolism in cyanobacteria Bothe et al. (1978) Hallenbeck and Benemann (1979) Lambert and Smith (1981) Bothe (1982) Houchins (1984) Hydrogen metabolism in photosynthetic bacteria Yoch (1978) Meyer et al. (1978a) Vignais et al. (1981b) Kumazawa and Mitsui (1982) Willison et al. (1983a) Biotechnological aspects of hydrogen production Benemann (1977) Mitsui (1978, 1979, 1981) Mitsui et al. (1980) Weaver et al. (1980) Benemann et al. (1980) Hallenbeck (1983a) Kondratieva and Gogotov (1983)

(I

Adapted from Lambert and Smith (1981).

nig, 1981). They contain only one photosystem (cf. Fig. 1) and perform anoxygenic photosynthesis (van Niel, 1931 ; Clayton and Sistrom, 1978) unlike cyanobacteria which, like green plants, contain two photosystems enabling them to split water and therefore to carry out oxygenic photosynthesis.

158

PAULETTE

M. VlGNAlS ET

AL.

=

0 2

“ 2 0

FIG. 1. Electron transport in Rhodopseudomonas capsu/ata and its relationship to Nz fixation and Hz metabolism. Light excites an electron in P870, the reaction centre bacteriochlorophyll, to a level where it can reduce the acceptor I (bacteriopheophytin). Subsequently, the electron is transferred sequentially through an iron-quinone moiety (UQ.Fe), a pool of ubiquinone (UQ), cytochrome (cyt) camers, and back to oxidized P870. In the process ATP is generated. Adapted from Weaver et al. (1980) and Melandri et a / . (1982). Abbreviations: H2ase, hydrogenase; Fd, ferredoxin.

A . CLASSIFICATION

The phototrophic green and purple bacteria, also known as photosynthetic or anoxygenic phototrophic bacteria (Triiper and Pfennig, 1981) or Anoxyphotobacteriae (Gibbons and Murray, 1978) do not produce oxygen. They are distinguishable from cyanobacteria and prochloron, the Oxyphotobacteriae, which are light-requiring bacteria that produce oxygen (Table 2). They comprise two orders, the Rhodospirillales and the Chlorobiales. This division is based on differences in the fine structure and pigment content of the photosynthetic apparatus. In the Rhodospirillales (“purple bacteria”) the entire photosynthetic apparatus, i.e., the reaction centre and antenna bacteriochlorophyll (with bchl a or b as the main pigment), is located in the cytoplasmic membrane, which invagi-

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

159

TABLE 2. Subdivision of the Kingdom “Prokaryotae” Murray (1968) as proposed by Gibbons and Murray (1978)“ Kingdom Prokaryotae Divisions:

Characteristic features

I. Gracilicutes 11. Firmacutes 111. Mollicutes

IV. Mendocutes Division I Class I

Have Gram-negative type of cell wall Prokaryotes able to carry out photosynthesis either oxygenic or anoxygenic. Photopigments may be chlorophyll a or b or bacteriochlorophyll a , b, c , d, or e Bacteria that produce oxygen during Subclass I Oxyphotobacteriae photosynthesis Cells contain chlorophyll a and Cyanobacteriales Order I ph ycobiliproteins Cells contain chlorophyll a and b but no Order I1 Prochlorales ph ycobiliproteins Subclass I1 Anoxyphotobacteriae Carry out a phototrophic metabolism under anaerobic conditions. Do not produce oxygen during photosynthesis Cell membranes contain bacteriochlorophyll a Rhodospirillales Order I or b, the pigments being located on internal membrane systems continuous with the cytoplasmic membrane Cells contain bacteriochlorophyll c, d, or e and Order I1 Chlorobiales small amounts of a. Pigments are located in the cytoplasmic membrane and in “chlorobium vesicles” Gracilicutes Photobacteria

Two proposals have been made, one on the rearrangement of the species and genera of the purple non-sulphur bacteria (Imhoff et a/., 1984) and a second on the reassignment of the genus Ectothiorhodospira to a new family, Ectothiorhodospiraceae (Imhoff, 1984). The authors suggest the use of the trivial name, purple non-sulphur bacteria, instead of the family name Rhodospirillaceae, as the heterogeneity of their group does not warrant the use of a family name. In accordance with these proposals, in the new edition of the Bergey’s Manual, the Anoxyphotobacteria will be classified under (I) Purple Bacteria, comprising (1) the family Chromatiaceae, (2) the family Ectothiorhodospiraceae, and (3) the group of purple non-sulphur bacteria and under (11) the green bacteria with ( I ) t h e group of green sulphur bacteria (including the Chlorobiaceae) and (2) the group of multicellular filamentous green bacteria (including the genus Chlorojetlexus).

160

PAULETTE M. VlGNAlS €T AL.

nates into the cytoplasm to form the intracytoplasmic membrane system. In the Chlorobiales, the bulk or antenna bacteriochlorophyll, consisting of bchl c, d, or e, is contained in distinct organelles known as chlorosomes, which underlie and are attached to the cytoplasmic membrane. The cytoplasmic membrane itself contains only the reaction-centre bacteriochlorophyll (bchl a) (Triiper and Pfennig, 1981). On the basis of physiological and ecological observations, the photosynthetic bacteria have been classified into four families. The order Rhodospirillales comprises two families: the Rhodospirillaceae, formerly Athiorhodaceae (Pfennig and Triiper, 1971 ; “purple non-sulphur bacteria”) and the Chromatiaceae, formerly Thiorhodaceae (Bavendamm, 1924; “purple sulphur bacteria”). The order Chlorobiales comprises two more families, the Chlorobiaceae, formerly Chlorobacteriaceae (Copeland, 1956; “green sulphur bacteria”) and the green filamentous gliding bacteria, Chloroflexaceae (Triiper, 1976). From their comparative analysis of the oligonucleotide sequences of the 16s RNAs, Gibson et al. (1979) and Fox et al. (1980) concluded that photosynthetic phenotypes are extremely ancient. The oldest bacteria, which were anaerobic, include the purple photosynthetic bacteria and the cyanobacteria together with the clostridia. According to these authors, the anaerobic photosynthetic bacteria would have been among the original Gram-negative bacteria, and aerobic bacteria would have appeared subsequently many times during evolution, by loss of photosynthetic properties. These evolutionary considerations are consistent with the diverse types of metabolism found in photosynthetic bacteria and, in particular, the anaerobic dark pathways of fermentation, as found also in the strict anaerobes (considered to be most closely related to the primitive anaerobic cells).

B . GROWTH PROPERTIES

The growth properties of the photosynthetic bacteria have been extensively described in recent reviews (e.g., Kondratieva, 1979; Triiper and Pfennig, 1981; Kondratieva et al., 1981). Most species of the Rhodospirillaceae family are facultative anaerobes and require B-group vitamins for growth. Many species can use COZas the main or even sole carbon source. Most of the Rhodospirillaceae are unable to use inorganic electron donors other than H2 for growth. However, a few can oxidize sulphide or thiosulphate (but not elemental sulphur). Sulphide is oxidized to sulphate, to tetrathionate, or to elemental sulphur, the latter being deposited in the medium outside the cell. Their

H2

METABOLISM IN PHOTOSYNTHETIC BACTERIA

161

preferred phototrophic electron donors and carbon sources are simple organic carbon compounds. Until recently the Chromatiaceae were considered obligate anaerobic phototrophs. However, the Russian school (cf. Kondratieva et al., 1981) has demonstrated that several of them have the capacity to grow in the dark in the presence of oxygen, although at a slow rate. Species such as Thiocapsa roseopersicina (Pfennig, 1970; Bogorov, 1974; Kondratieva et al., 1975, 1976) and Larnprobacter modestohalophilus (Gorlenko et al., 1979) have the capacity to grow in the dark under autotrophic conditions in the presence of 02,with sulphide or thiosulphate as electron donors. Ectothiorhodospira shaposhnikouii (Kondratieva et al., 1976) and Echtothiorhodospira mobilis (Krasil’nikova et al., 1980) are also capable of growing in the dark even under strong aeration. All species can utilize hydrogen sulphide and elemental sulphur as electron donors; they are therefore autotrophs. Some of them also can use thiosulphate, others reduced sulphur compounds, or Hz. Only a few purple sulphur bacteria can use organic compounds as electron donors. The Chlorobiaceae are strict anaerobic photo-autotrophs and some species require vitamin B12.All species oxidize hydrogen sulphide and sulphur and some of them can also use thiosulphate or H2 as electron donors. They can use some organic acids for photoheterotrophic growth, but only in the presence of carbon dioxide.

C. ECOLOGICAL DISTRIBUTION

The diversity of growth modes available to the photosynthetic bacteria enables them to survive under a great variety of environmental conditions, and explains the occurrence of these bacteria in many different habitats. Photosynthetic bacteria are found in great numbers in ponds, lakes, and estuarine lagoons where an active breakdown of organic matter takes place. These habitats constitute special biotopes with vertical gradients of light and oxygen (from above) and hydrogen sulphide (from below). Temperature and salinity gradients may also contribute to the stratification of these waters, in which the various species of phototrophic bacteria develop in a vertical distribution (Caumette et al., 1983; Caumette, 1984). In these aquatic ecosystems, green and brown sulphur bacteria (Chlorobiaceae), which are non-motile and sulphide dependent, are found at the greatest depths (between 5 and 7 m), closest to the bottom sediment (Caumette et a f . , 1983). Above the Chlorobiaceae, in anaerobic, sulphide-

162

PAULETTE

M. VlGNAlS ET AL.

containing water, are found the purple sulphur bacteria (mostly Chromatium species). The purple non-sulphur bacteria are found further up, in more aerated waters with a low sulphide content, which contain lowmolecular-weight carbon compounds resulting from the bacterial decomposition of organic materials. The vertical distribution of photosynthetic micro-organisms in stagnant waters is made possible by the fact that their photopigments absorb light energy in different regions of the spectrum (700-760 nm for the Chlorobiaceae, below 690 nm for the algae and cyanobacteria and above 800 nm for the purple bacteria; Pfennig, 1978) and by the high affinity of their photosynthetic apparati, which are saturated at low light intensities [lOOO-2000 lux for Chromatium (Triiper and Schlegel, 1964) and 700 lux for Chlorobium (Lippert and Pfennig, 1969)].

111. Hydrogen Utilization and Production by Photosynthetic Bacteria

A. HYDROGEN AS AN ELECTRON DONOR

1 . Anaerobic Growth in the Light a . Hydrogen consumption linked to photoreduction of carbon dioxide. It has been known for some time (Roelofsen, 1935; Gaffron, 1935) that photosynthetic bacteria, such as Chromatium, can use molecular hydrogen as sole electron donor during photo-autotrophic growth. It was later recognized that the capacity to oxidize H2 is linked to the presence of the enzyme hydrogenase (Gest, 1951). Hydrogen utilization by intact cells of photosynthetic bacteria was observed to be dependent on, or stimulated by, illumination, when the acceptor was COz (Gaffron, 1935; Roelofsen, 1935; Nakamura, 1937, 1938). However, light was not necessary for the reduction by H2 of oxidants such as 0 2 , ferricyanide, elemental sulphur, sulphite, sulphate, hyposulphite, nitrite, nitrate, or fumarate (see Gest, 1951), indicating that the hydrogenase activity per se is not light dependent. Photo-autotrophic growth with H2 as electron donor was demonstrated for Rhodospirillum rubrum (strain S-1) (Ormerod and Gest, 1962; Anderson and Fuller, 1967; Buchanan et al., 1967), Rhodopseudomonas palustris (Qadri and Hoare, 1968), Rhodopseudomonas gelatinosa (Wertlieb and Vishniac, 1967), and Rhodopseudomonas capsulata (Klemme and Schlegel, 1967). Among purple non-sulphur bacteria, R . capsulata strains

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

163

possess the fastest growth rates under photo-autotrophic conditions (Klemme, 1968). Many green bacteria can use molecular hydrogen as an electron donor. Autotrophic growth of these micro-organisms required cysteine or some other reduced sulphur compound (Lippert and Pfennig, 1969).

b. Sulphate and thiosulphate reduction. Chromatium minutissimum cells grown in the light with sodium sulphide were shown to consume HZwhen incubated in the dark in the presence of sulphate or thiosulphate (Nakamura, 1939, 1941). These results indicated that C. minutissimum could use H2 as an electron donor for the reduction of sulphur compounds. However, this type of reaction was not found in Chromatium uinosurn strain D (van Niel, 1936; Hendley, 1955). c. Nitrate reduction. Gaffron (1933, 1935) and Nakamura (1937; see also Yamagata and Nakamura, 1937) reported independently that, in two different species of Rhodospirillaceae, nitrate was reduced to ammonia by molecular hydrogen. These findings have not yet been reproduced in spite of many attempts. It is possible that, instead of using an assimilatory nitrate reductase, these bacteria first reduced nitrate to dinitrogen with a dissimilatory nitrate reductase and then reduced N2 to ammonia with nitrogenase, as shown recently with Rhodopseudomonas sphaeroides forma denitrifcans (Dunstan et al., 1982). Several strains of Rhodospirillaceae related to R . capsulata and R . sphaeroides were shown by Klemme (1979) to be capable of growing with nitrate as sole nitrogen source under anaerobic conditions in the light. These strains, and two others of the R. palustris group, produced nitrite from nitrate under photosynthetic anaerobic conditions indicating that the nitrate reductase present in those strains was of the respiratory type rather than of the assimilatory type, as is the case with the denittifying strain of R . sphaeroides described by Satoh et al. (1976). Strains possessing both a respiratory nitrate reductase and a membrane-bound hydrogenase would be expected to use H2 as an electron donor (to nitrate) and as a source of energy, as is the case for denitrifiers such as Paracoccus denitrifcans (Vignais et al., 1981a). However, this was not the case for a strain of R . rubrum S-1 “trained” to grow anaerobically in the light or in the dark with nitrate as sole nitrogen source (Katoh, 1963). The nitrate-adapted cells of R . rubrum contained a respiratory nitrate reductase; they could reduce nitrate to nitrite under anaerobic conditions, in the light or in the dark, with organic electrons donors such as succinate, malate, lactate, or 2-oxoglutarate. Molecular hydrogen was not used although hydrogenase activity was present in the adapted cells.

164

PAULETTE M. VlGNAlS ET AL.

It should be noted that, in P. denitrijicans, both nitrate reductase and hydrogenase are intrinsic membrane proteins and form part of the respiratory chain (Vignais et al., 1981a), whereas in R . sphaeroides forma denitrijicans the dissimilatory nitrate reductase is soluble (Satoh, 1981) and is located in the periplasmic space (Sawada and Satoh, 1980). Among the Chromatiaceae, Nakamura (1939) has reported that hydrogen uptake by photo-autotrophically grown C. minutissimum cells was stimulated by the addition of potassium nitrite and potassium nitrate during dark incubation. The physiology of nitrate reduction in the photosynthetic purple bacteria (both sulphur and non-sulphur) has recently been reviewed by Castillo and Cardenas (1 982). 2 . Aerobic Growth in the Dark Rhodopseudomonas capsulata is able to grow chemo-autotrophically under aerobic conditions in the dark. Under these conditions, H2 serves as the source of energy and reducing power, O2 as the terminal electron acceptor for energy transduction, and C 0 2 as the sole carbon source (Madigan and Gest, 1979; Siefert and Pfennig, 1979; Colbeau et al., 1980). Rhodopseudomonas acidophila is capable of growing in the dark with H2 and O2under micro-aerophilic conditions, and is also able to grow chemotrophically in the dark on the single carbon substrates methanol and formate (Siefert and Pfennig, 1979). The inhibitory effects of high O2tensions on R . acidophila may be due to inhibition of activity and/or synthesis of hydrogenase, and perhaps of ribulose bisphosphate carboxylase activity (Siefert and Pfennig, 1979). The relatively low sensitivity to O2 observed in R . capsulata strains is probably linked to a high respiration rate, which enables the cells to maintain an anaerobic intracytoplasmic micro-environment; for example R . acidophila versus R . capsulata (Siefert and Pfennig, 1979) and Rhodopseudomonas sulJidophila (Kelley et al., 1979) compared to R . capsulata (Colbeau et al., 1980).

8 . HYDROGEN PRODUCTION

The formation of molecular hydrogen results from the direct reduction of protons from water. As will be discussed later (Sections IV and V) two different enzymes can catalyse proton reduction: hydrogenase and the nitrogenase complex.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

165

1. Photoproduction of Molecular Hydrogen a . Discovery. Gest and Kamen (1949a,b) were the first to observe a photochemical production of molecular hydrogen by a photosynthetic bacterium, R. rubrum. They observed that photo-evolution of H2 proceeded readily under an atmosphere of 100% H2, or when noble gases were used to establish anaerobiosis, but did not occur under an atmosphere of N2. Addition of ammonium chloride or high concentrations of yeast extract to the culture medium stimulated growth but inhibited H2 production (Gest and Kamea, 1949a). Subsequent experiments with the isotope lSN indicated clearly that R. rubrum contains a N2-fixing system (Gest and Kamen, 1949b; Gest et al., 1950). Shortly afterwards, N2 fixation was demonstrated in the purple sulphur bacterium Chromatiurn, in the green sulphur bacterium Chlorobium (Lindstrom et al., 1949), and in several other species of purple non-sulphur bacteria, including R. palusiris, R. sphaeroides, R . capsulata, and R . gelatinosa (Lindstrom et al., 1951). Since H2 production did not occur when N2 fixation was repressed, there appeared to be a close relationship between these two processes. Bulen et al. (1965a), while studying nitrogen fixation by cell-free preparations of R. rubrum, demonstrated that the ATP and the low potential reductant required for N2 fixation were also necessary for the production of H2 by these preparations. This ATP-dependent hydrogenase activity was distinguishable from that of the “conventional” hydrogenases, in that the activity was independent of the partial pressure of H2, was irreversible, and was insensitive to carbon monoxide (Bulen et al., 1965a). From their studies on N2 fixation in aerobic and photosynthetic microorganisms, Bulen et al. (1965b) concluded that the same enzyme, nitrogenase, catalyses both N2 reduction and ATP-dependent H2 evolution. b . Features of hydrogen photo-evolution. Light-dependent hydrogen evolution is now recognized as a general property of photosynthetic bacteria (Gest, 1972; Yoch, 1978). The species that have been shown to evolve H2 in the light are listed in Table 3. The photo-evolution of H2 occurs in the absence of N2 and of high concentrations of ammonium ions, under conditions in which ATP from photophosphorylation and reducing equivalents from organic substrates are produced in excess (Ormerod et al., 1961; Gest et al., 1962; Bose and Gest, 1963; Hillmer and Gest, 1977a,b). Hillmer and Gest (1977a,b) studied in detail the effects of various factors on the photoproduction of H2 by R. capsulata. In growing cultures, the highest rates of H2 production (130 p1 hr-1 ml culture - I ) were ob-

TABLE 3. Maximal rate of H2 photoproduction by some photosynthetic bacteria'

Family

Species

Conditions

Electron donor

Strain

Activity (ml Hz hr-I g dry wt-')

References ~

2

g

Rhodospirillaceae

Rhodospirillum rubrum

Rhodopseudomonas capsulara

Nitrogen-limited growth

Lactate Lactate Lactate

s-I s-1

Lactate Lactate Lactate Lactate Lactate

B 10 B 10 SCJ LB2 w52 (B10 Hup-)

Malate Malate

BlOO ST410 (B100 Hup-) IR4 (B 10 Hup-) B10 N-3

Malate Malate Malate

109 146 20 115

260 168 176 144

~

~~

Ormerod et al. (1961) Weaver et al. (1980) Ziirrer and Bachofen ( 1979) Meyer e l al. (1978a) Jouanneau er al. (1982) Weaver et al. (1980) Weaver et al. (1980) Weaver et al. (1979)

100

Takakuwa et al. (1983) Takakuwa et al. (1983)

127

Willison et al. (1984)

95 119

Willison et al. (1984) Song et al. (1980)

69

Rhodopseudomonas palustris Agar entrapped cells Agar entrapped cells Agar-entrapped cells Rhodospirillurn molischianum

Chromatiaceae purple sulphur

Rhodopseudomonas sphaeroides Rhodopseudomonas acidop hila Rhodopseudomonas sulphidophila Rhodopseudomonas viridis Chrornatium sp.

Thiocapsa ruseopersicina Ectothiorhodospira shaposhnikovii Adapted from Weaver et al. (1980).

Agar entrapped cells

Lactate

EC

62

Weaver et al. (1979)

Lactate

420L

42

Vincenzini et al. (1982b) Vincenzini et al. (1982b)

Sugar refinery wastes Straw paper-mill effluent Straw paper-mill effluent Malate

S

90

Watanabe et a / . (1981)

Lactate

DSM 137

49

Lactate

BSW8

106

Siefert and Pfennig (1978) Weaver et al. (1979)

Lactate

N THC 133

3

Weaver et al. (1979)

43

Succinate Miami PBS 1071 and thiosulphate Pyruvate BBS Sulphide

E.BA 1011

50

Vincenzini et a / . (1982b)

139

Vincenzini et al. ( 1982b)

134

20 8

Ohta and Mitsui (1981)

Gogotov (1978) Matheron and Baulaigue (1983)

168

PAULETTE M. VIGNAIS E r AL.

tained with DL-lactate or pyruvate as carbon source, and either glutamate, serine, or alanine as growth-limiting nitrogen source (Hillmer and Gest, 1977a). Hydrogen production was observed only when the ratio of the concentrations of glutamate and lactate was lower than 1.0, since at higher ratios net production of NH4+from glutamate occurred, resulting in the inhibition of nitrogenase. The highest yield of H2, 72% of the theoretical maximum for the complete dissimilation of carbon substrate to H2 and C 0 2 , was obtained with m-lactate and succinate; sugars gave much lower yields than organic acids. In a standard “lactate-glutamate” system, it was found that H2 production began prior to the exhaustion of glutamate from the medium, but reached a maximum rate only when the nitrogen source had been completely consumed. Hydrogen production continued until lactate had been depleted from the medium. The rate of H2 production increased with increasing light intensity, reaching a plateau at a similar intensity (600 foot-candles or 6480 lux) to that which was saturating for growth. In R . sphaeroides, the rate of H2 production was found to be proportional to light intensity up to 12,000 lux (Macler et al., 1979). In nitrogen-limited continuous culture, the maximal rate of H2 production from m-lactate was 260 pl hr-I mg cell dry wt.-’, considerably higher than the maximal rates observed for batch culture (Jouanneau et al., 1982). In contrast to the experiments in batch culture, H2 production was saturated at a higher light intensity than that required to saturate growth (15,000 lux vs 8000 lux), indicating that the factors regulating H2 production are different in the two systems. Continuous culture is clearly preferable to batch culture for studies of the factors controlling H2 production since parameters such as cell density and growth rate can be kept constant. It has been suggested that the role of H2 photoproduction is to dissipate reducing equivalents and ATP when these are produced in excess, i.e., to regulate the intracellular redox balance (Gest, 1972). However, the observation that H2 production occurs even when the light intensity is growthlimiting suggested that H2 production may, under some conditions, constitute an energetic burden to the cells (Hillmer and Gest, 1977a). Indeed, if R . capsulata is repeatedly subcultured in a growth medium allowing HZ production (Wall and Love, 1984), or is grown in nitrogen-limited continuous culture (Allibert et al., 1984), then the appearance of Nif- mutants in the culture is observed. This could be explained if synthesis and activity of nitrogenase under these conditions was deleterious to growth. Little is known about the mechanisms involved in the regulation of H2 production. The effect of carbon and nitrogen source on the rate of H2 production probably reflects both differing rates of utilization of carbon substrates, and differing nitrogenase activities under various growth con-

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

169

ditions. The effect of increased light intensity on H2 production presumably results from an enhanced synthesis of ATP, leading to an increased activity andor synthesis of nitrogenase. In R. capsulatu strain B10, the increase in nitrogenase activity following illumination was inhibited by chloramphenicol, suggesting that nitrogenase synthesis is induced by light (Hillmer and Gest, 1977b; Meyer el al., 1978b). Experiments with specific antibodies against nitrogenase have confirmed that increased light intensity results in an increased nitrogenase synthesis (see Section V.C.2). In R. palustris, and in several other purple non-sulphur bacteria, the nitrogenase content of cells was determined by several methods, including the titration of iron-containing protein (Fe protein) with antibodies against the Fe protein of R. rubrum (Arp and Zumft, 1983a). In this way, it was established that the higher specific nitrogenase activity (four- to eightfold) of nitrogen-limited cells, as compared with that of N2-grown cells, was due to an increased synthesis (“overproduction”) of nitrogenase in the former. Most organic substrates are only partially dissimilated to H2 and C 0 2 . However, the fate of the remaining substrate carbon, and the different pathways of carbon substrate utilization, have received very little attention. A mutant of R. sphaeroides has been isolated, which, unlike the wild type, is able to produce H2 from glucose, with an efficiency appoaching 100% (Macler et al., 1979). Unlike the wild type, this mutant did not accumulate gluconate during growth on glucose, and it was suggested that the mutant had acquired an increased capacity to convert gluconate into C3 intermediates, via the enzymes of the Entner-Doudoroff pathway (Macler et ul., 1979). Mutants of R. capsulata have been isolated which are unable to grow photo-autotrophically (Aut- phenotype) and also show an increased stoicheiometry of H2 production from various organic substrates, such as m-malate (Willison et al., 1984). Analysis of the carbon balance in H2-producing cultures of these strains indicated that, in the mutants, a lower proportion of the organic carbon source provided was excreted into the medium in the form of (unidentified) end products. The excretion of acidic end products by H2-producing cultures has been confirmed by high-performance liquid chromatography (Odom and Wall, 1983; Takakuwa et al., 1983). A similar conclusion was reached by Chakrabarti and Smith (1981) who found that an unusual strain of R. cupsuluta (2.3.1 .), which does not photoproduce H2, excretes large amounts (compared to H2-producing strains) of substances absorbing at 330 nm (at pH 121, a characteristic of the a-keto acid precursors of the aromatic amino acids. These authors suggested that, in strain 2.3. l., the role of H2 production in the regulation of the intracellular redox balance is supplanted by the formation and excretion of metabolites of low redox state.

170

PAULETTE M. VIGNAIS E r AL.

The link between carbon metabolism and H2 production might be provided by changes in the redox state of nicotinamide nucleotides; this has also been suggested for the regulation of H2-dependent photoreduction of C 0 2 (Schick, 1971a,b,c; Hillmer and Gest, 1977b), although direct evidence has not yet been provided. It is clearly important to identify both the end products excreted by H2-producing cultures and the metabolic patterns involved in their formation, in order to fully understand the regulation of the H2 production process, In the early studies, where organic reductants were used and C 0 2 evolved simultaneously with H2, C 0 2 was considered necessary for the photo-evolution of H2. In 1961, Losada et al. (1961), using C . vinosum, demonstrated that an inorganic electron donor, thiosulphate, could provide the reduction equivalents for H2 production in the light. These findings were confirmed immediately by Orrnerod et al. (1961). It appeared therefore that the features of H2 photo-evolution from thiosulphate (response to light and dark and to the nitrogen source) were similar in Chromatium to those observed in R . rubrum with carbon compounds as electron donors, suggesting that the same system was involved in both cases (Losada et al., 1961). The relationship between nitrogenase-mediated H2 evolution and carbon metabolism in the Chromatiaceae is, however, not yet fully understood. Ohta and Mitsui (1981) observed that the marine Chromatium sp. Miami PBS 1071 produced hydrogen gas at two to three times higher rates when two donor substrates, succinate and thiosulphate (or succinate and sulphide), were used together than when the substrates were used separately. Matheron and Baulaigue (1983) reported recently that the photoevolution of H2 from sulphide by the purple sulphur bacteria, Chromatium, Thiocapsa, Thiocystis, and Ectothiorhodospira, was tightly linked to C 0 2 assimilation; these authors observed also that the utilization of endogenous substrates was markedly stimulated by sulphide in Ectothiorhodospira. 2 . Production of Molecular Hydrogen in the Dark a . Dark, fermentative metabolism. Purple bacteria have the capacity to ferment their endogenous substrates in the dark with production of organic acids, C02, and H2 (van Niel, 1944; see Uffen, 1978, for review). Fermentative H2 production was stimulated by the addition of pyruvate (Kohrniller and Gest, 1951). The following Rhodospirillaceae have been shown to grow under dark anaerobic conditions: R . rubrum (Schon, 1968;

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

171

Uffen and Wolfe, 1970; Uffen et al., 1971; Uffen, 1973a,b; Schon and Biedermann, 1973; Giirgiin et al., 1976), R . palustris, R. uiridis, and R . sphaeroides (Uffen and Wolfe, 1970), and R . capsulata (Yen and Marrs, 1977; Madigan and Gest, 1978; Schultz and Weaver, 1982). In R. capsulata, fermentative growth was first observed only in the presence of an additional oxidant, either trimethylammonium N-oxide (TMAO) or dimethyl sulphoxide (DMSO) (Yen and Marrs, 1977; Madigan and Gest, 1978; Madigan et al., 1980; Cox et al., 1980) which, by acting as an electron sink, allowed a more complete oxidation of the substrate. Thus, in a continuous culture, during fermentative growth with fructose as carbon substrate, lactate accumulates when TMAO is limiting; when TMAO is in excess, the pyruvate is broken down to acetate, leading to the synthesis of one molecule of ATP (Cox, 1983). The involvement of the respiratory chain during anaerobic growth is controversial. Zannoni and Marrs (1981), after comparing wild type and respiratory mutant strains, suggested that the respiratory chain was not involved in the synthesis of ATP during dark anaerobic growth, although they suggested that the membrane-bound NADH dehydrogenase might be involved in the reduction of DMSO. On the other hand, Schultz and Weaver (1982) concluded that growth on fructose in the presence of TMAO was due to anaerobic respiration; indeed, recently, the utilization of TMAO was found to generate a membrane potential in R. capsulata (McEwan e t al., 1983). Nevertheless, R . capsulata, like R. rubrum, also can grow by pure fermentation (without accessory oxidant) but at a slower rate (Schultz and Weaver, 1982). Rhodospirillum rubrum ferments fructose producing succinate, acetate, formate, propionate, C02, and H2 (Schon and Biedermann, 1973). Fermentation of fructose by R. capsulata produces H2, succinate, lactate, acetate, and C 0 2 (Schultz and Weaver, 1982). When R. rubrum was grown anaerobically in the dark, on pyruvate, cells produced formate, acetate, and equimolecular amounts of H2 and C 0 2 gases (Uffen, 1973a). Pyruvate was broken down by an inducible pyruvate-formate lyase (Gorrell and Uffen, 1978) which is the key enzyme for the fermentation of pyruvate (Jungermann and Schon, 1974; Gorrell and Uffen, 1977). The pyruvate-formate lyase reaction yields energy, ATP being formed by the classic phosphoroclastic reaction (Fig. 2). Hydrogen production results from the breakdown of formate to C02 and H2by formic hydrogenlyase (Gest, 1951), and the cleavage of formate and the synthesis of H2 are stimulated at low pH values (< 6.5) (Schon and Voelskow, 1976). Gas production from formate by the formic hydrogenlyase reaction is not influenced by light but is completely inhibited

PAULETTE M. VlGNAlS ET AL.

172 CO SENSITIVE

j

HVPOPHOSPHITE SENSITIVE

I I

H*

a

9

FORMATE

I

Acetyid P

I I

I

i I

/I

6)

pH71

( ttcetyl-5 COA

I

I

l

PVRL~E

pH 23

I

I

I

PH 13.3

I

=02

HVPOPHOSPHITE INSENSITIVE I

k A ACETATE

T

ADP. Pi P

I

I

I 0 1

I

I I

FIG. 2. Pathway for fermentative metabolism of sodium pyruvate by anaerobic, dark-grown Rhodospirillum rubrum. Enzymes: (1) formic-hydrogen lyase, (2) pyruvate-formate lyase, (3) pyruvate-ferredoxin oxidoreductase. Reproduced from Gorrell and Uffen (1977).

by carbon monoxide (Gorrell and Uffen, 1977). This is a typical fermentative type of H2 production, which is quite distinct from the CO-insensitive, ATP-dependent production of H2 mediated by nitrogenase. Formic hydrogenlyase is synthesized constitutively in R. rubrum (Gorre11 and Uffen, 1978). In cells grown and incubated anaerobically in the dark, the only hydrogenase present is that found in formic hydrogenlyase which is therefore responsible for all the H2produced from formate under these conditions (Gorrell and Uffen, 1977; Schon and Voelskow, 1976; Voelskow and Schon, 1978). However, nitrogenase and soluble hydrogenase activities can co-exist under certain conditions. When cells of R. rubrum were grown in the light to de-repress the nitrogenase, then incubated anaerobically in the dark with pyruvate, H2 was produced by both enzymes. The use of specific inhibitors (NH4+,CO) showed, however, that the activity of hydrogenase was preponderant (Voelskow and Schon, 1980). Formic hydrogenlyase was inhibited by light in the wild-type strains S1 and F1 of R. rubrum, but not in a regulatory mutant of S1 (strain Cl). Thus, when this mutant was incubated in the light with pyruvate, large amounts of H2 were produced, 50% via hydrogenase and 50% via nitrogenase (Gorrell and Uffen, 1978). Under these conditions, both enzymes served the same physiological role. Strain HA of R. rubrum is devoid of formic hydrogenlyase activity (Schon and Voelskow, 1976; Voelskow and Schon, 1978). A Nif- mutant obtained from this strain, was unable to grow by fermentation on pyruvate or under Nz-fixing conditions, but grew well under autotrophic conditions

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

173

with H2 and COz, since it is able to synthesize adaptatively an Hz-uptake hydrogenase (Voelskow and Schon, 1980). Thus, in R . rubrum, three enzymes interact in HZmetabolism according to the growth conditions: the two hydrogenases and nitrogenase. The soluble hydrogenase has not yet been purified or characterized, and it is not known whether it is distinct from the Hz-uptake hydrogenase. Formic hydrogenlyase is also present in R . palustris, and functions in the light (Qadri and Hoare, 1968). Rhodopseudomonas palustris can grow photo-autotrophicdly on formate, which is degraded to H2 and C02 by formic hydrogenlyase, the gases produced being assimilated autotrophically. It has been found that the formic hydrogenlyase of R . palustris, unlike the enzyme of R . rubrum, is inducible; it is composed of a particulate hydrogenase and a soluble formate dehydrogenase (Qadri and Hoare, 1968). It is not known whether this species contains a single reversible hydrogenase or two separate hydrogenases functioning unidirectionally . In the case of R . rubrum and R . capsulata, no H2 was evolved during anaerobic growth in the presence of DMSO or TMAO; if H2 is formed, it may be oxidized by the added electron acceptors via an anaerobic electron transport system (Schultz and Weaver, 1982; McEwan et al., 1983) involving an uptake hydrogenase. Nakamura (1939, 1941) observed that photoheterotrophically grown cells of C . minutissimum evolved HZwhen incubated in the dark with reduced organic compounds. This organism appeared to degrade pyruvate via a pyruvate formate-lyase reaction. Bennett and Fuller (1964) and Bennett et al. (1964) demonstrated that C. uinosum strain D, grown photoheterotrophicdly with L-malate, possessed a pyruvate-ferredoxin oxidoreductase pathway, and catalysed the production of acetate plus equimolecular amounts of H2 and COz both in the light and in the dark. Despite these observations, no clear evidence has yet been obtained for the presence of a soluble, reversible hydrogenase in the purple sulphur bacteria (see below, Section IV). b. Anaerobic oxidation of carbon monoxide. The anoxyphotobacteria have recently been shown to have an anaerobic CO-uptake capability (Uffen, 1981) similar to that of methanogenic bacteria, clostridia, or sulphate reducers such as Desulfouibrio. During anaerobic oxidation of CO, it is assumed that water serves as oxidant according to the reaction: CO

+ H20

-

CO,

+ 2H+ + 2e-

(1)

Strict anaerobes can only tolerate low concentrations of CO and need a metabolizable substrate to grow in the presence of CO, whereas Rhodo-

PAULETTE

174

M. VlGNAlS ET AL.

pseudomonas species are able to grow anaerobically with CO as sole reductant and carbon source, both in the light (Hirsch, 1968) and in the dark (Uffen, 1976). Strains of R . gelatinosa grew to a density of 2.2 X lo8 cells ml-’ in complete darkness under a 100% (v/v) CO gas atmosphere and in liquid medium devoid of any other carbon substrate (Uffen, 1976). Resting cell suspensions, utilizing CO, released equimolar amounts of C02 and H2 according to the following energy-yielding reaction: CO

+ H20

-

C 0 2 + Hz

AG

=

4.8 kcal (-20.1 kJ)

(2)

Consistent with this mechanism, tritiated hydrogen gas was produced in the presence of tritiated water (Uffen, 1976). This anaerobic CO oxidation is catalysed by a CO dehydrogenase which, in R . gelatinosa, is membrane bound (Wakim and Uffen, 1983), inducible, O2 labile, and insensitive to heat treatment at 70°C (Uffen, 1981). It is noteworthy that these properties are shared by the uptake hydrogenases of photosynthetic bacteria (see Section IV). Furthermore, both the uptake hydrogenases (Colbeau and Vignais, 1983) and the anaerobic CO dehydrogenases from Clostridium pasteurianum and Clostridium thermoaceticum (Drake et al., 1980; Ragsdale et al., 1983) are nickelcontaining, iron-sulphur proteins. However, the CO dehydrogenase from C . thermaceticum, which is soluble and has been purified to homogeneity, had no hydrogenase activity (Ragsdale er al., 1983). The CO oxidase of aerobic carboxydotrophic bacteria differs from that of the anaerobic bacteria in being a molybdo-iron-sulphur flavoprotein (Meyer and Schlegel, 1983). The production of H2 by the CO oxidation system of R . gelatinosa is presumably catalysed by a hydrogenase which must therefore be different from the majority of hydrogenases in being insensitive to CO.

IV. Hydrogenase A . GENERAL BACKGROUND

The hydrogenase enzymes (EC class 1.12) catalyse the “activation” or reversible oxidation of molecular hydrogen. The occurrence of an enzyme capable of using H2 as a substrate was first demonstrated in Escherichia coli by Stephenson and Stickland (1931), who named it “hydrogenase.” Since that time, hydrogenases have been found in a wide variety of bacterial and algal species (see reviews by Gray and Gest, 1965; Schlegel and Schneider, 1978; Adams et al., 1981).

H2

METABOLISM IN PHOTOSYNTHETIC BACTERIA

175

In photosynthetic bacteria, the H2-uptake hydrogenase has been studied in the most detail. More generally, this type of hydrogenase is found in all hydrogen-oxidizing and N2-fixing bacteria (Bowien and Schlegel, 1981 ; Vignais et al., 1981a). The “conventional” hydrogenase, probably involved in dark fermentative evolution of HZ,has not yet been purified and characterized.

B . LOCALIZATION

1. Localization in the Cell

Bacterial hydrogenases can have various localizations in the cell. For example, the H2-evolving hydrogenases of anaerobic bacteria, such as clostridia, are soluble and cytoplasmic (Nakos and Mortenson, 1971), whereas H2-uptake hydrogenases are found either in the periplasmic space (e.g., in Desulfovibrio vulgaris, strain Hildenborough; van der Westen et al., 1978), in the cytoplasmic space (e.g., in Alcaligenes eutrophus; Schneider and Schlegel, 1976) or in the membrane (e.g., in Paracoccus denitrijicans; Sim and Vignais, 1978). The hydrogenases of photosynthetic bacteria are generally membrane bound, although some controversy exists, as the enzyme may be more or less tightly bound to the membrane. In Rhodopseudomonas capsulata (Colbeau and Vignais, 1981) and in Chromatium (van Heerikhuizen et af., 1981) none of the treatments that removed the periplasmic proteins was able to solubilize the hydrogenase. In R . capsulata strain B10, the hydrogenase appears to be an intrinsic membrane protein and its solubilization requires treatment either with high concentrations of detergent (Colbeau and Vignais, 1981; Colbeau et al., 1983) or with acetone (Serebriakova and Gogotov, 1981); high salt concentrations (1 M NaCl or KC1) were ineffective (Colbeau and Vignais, 198 1). After sonication of Thiocapsa roseopersicina cells, hydrogenase activity was found in the soluble fraction (Gogotov et al., 1976; Zorin and Gogtov, 1975, 1982), but the hydrogenase purified from the soluble fraction, and that extracted from the chromatophores, were found to be the same enzyme (Gogotov et al., 1978b). In Rhodospirillum rubrum, the hydrogenase is membrane bound (Gest, 1952) and its total extraction from chromatophore membranes was achieved with the use of pancreatine, which contains a lipase (Adams and Hall, 1977, 1979). However, under certain conditions, hydrogenase was found in the soluble fraction (Kakuno et al., 1978) and even in the extracellular medium (Huira et al., 1979). With regards to Chromatium some discrepancies exist in the literature,

176

PAULETTE M. VlGNAlS ET AL.

since hydrogenase activity was found to be either totally membrane bound (Feigenblum and Krasna, 1970; Gitlitz and Krasna, 1975; Kakuno et al., 1977), partially soluble (Buchanan and Bachofen, 1968; van Heerikhuizen et al., 1981), or totally soluble (Weaver et al., 1965). Van Heerikhuizen et al. (1981) showed that the percentage of soluble enzyme varied greatly with the method of solubilization. Sonication, French-press treatment, or grinding with sand solubilized 20-50% of the activity. Differences in growth conditions may also have an influence on the binding of the enzyme to the membrane. Gitlitz and Krasna (1975) solubilized the enzyme by sonication in the presence of deoxycholate, but it was necessary to stir the solubilized enzyme with glass beads in the presence of a high salt concentration to obtain a completely soluble enzyme since hydrogenase binds very strongly to other proteins (Weaver et al., 1980). In Chlorobium limicola forma thiosulfatophilum nearly all of the hydrogenase activity appears to be in the cytoplasm (Kovacs and Bagyinka, 1982). 2. Orientation of Membrane-Bound Hydrogenases Few studies have dealt with the orientation of hydrogenase in the membrane, especially in photosynthetic bacteria, in spite of the fact that in these bacteria homogeneous membrane preparations of opposite polarity, namely sphaeroplasts (right-side-out) and chromatophores (inside-out), can be easily obtained. The problem has been studied by various methods. One of these is based on the use of non-permeant electron donors or acceptors. Thus, it was shown that the hydrogenase of P. denitrijicans can reduce benzylviologen on the periplasmic side of the membrane (Sim and Vignais, 1978). Since the protons resulting from the oxidation of H2 are discharged in the cytoplasmic compartment (Doussikre et al., 1980), it follows that the enzyme is a transmembrane protein. This is also the case for the hydrogenase of E. coli which, however, is oriented in the opposite direction (Jones, 1980). The difference in permeability of oxidized and reduced viologens (Jones et al., 1976) was used to demonstrate that the active site of the hydrogenase of T. roseupersicina is on the periplasmic side of the membrane (Kovacs et al., 1982; Bagyinka et al., 1981a,b, 1982). Recently, immunological methods have been employed to study the orientation of the hydrogenase in the membranes. Graham (1981) has coupled immunological and '251-labellingmethods to study the transmembrane localization of the hydrogenase of E. coli. More recently, van der Plas et al. (1983) found the enzyme to be located at the inner aspect of the

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

177

cytoplasmic membrane. In R . capsulata, hydrogenase cannot reduce the non-permeant oxidized form of benzyl viologen, either in osmotically protected sphaeroplasts or in chromatophores (Colbeau et al., 1983). Thus, the active site appears to be buried more or less deeply in the membrane. Crossed and rocket immunoelectrophoresis with specific antihydrogenase immunoglobulins showed that these could bind to hydrogenase in isolated chromatophores but not in sphaeroplats, indicating that the enzyme protrudes largely on the cytoplasmic side of the membrane. C. STABILITY AND STABILIZATION

The hydrogenases of photosynthetic bacteria are characterized by their heat stability and their resistance to 02 and denaturing agents (Zorin and Gogotov, 1982). Since they are relatively insensitive to 02 inactivation, they can be isolated under aerobic conditions (Gitlitz and Krasna, 1975; Adams et al., 1979; Serra et al., 1979), but they are inactive unless O2 is removed by sparging with an inert gas. 1 . Stability During Storage

Partially purified hydrogenase from T. roseopersicina lost 50% of its activity after 6 days; its stability was increased by immobilization in a 20% (w/v) polyacrylamide gel (?,I2 12 days) (Gogotov et al., 1978b). In contrast, the highly purified enzyme showed half-inactivation times of 60 days at room temperature and 190 days at 4°C: this increase in stability was probably due to the removal during purification of destabilizing (possibly superoxide-producing) compounds (Zorin and Gogotov, 1982). Rhodopseudomonas capsulata hydrogenase differed from that of T. roseopersicina in being cold labile and more stable at room temperature (Colbeau et al., 1978). The R . rubrum enzyme showed no detectable loss of activity after storage for 24 hours under air, at either 4 or 20°C; at -2O"C, 50% of the activity remained after 6 months (Adams and Hall, 1979). The hydrogenase from R . capsulata strain B10 was stable only when stored under H2 (Colbeau et al., 1978; Serebriakova and Gogotov, 1981); at 20°C, the loss of activity was more rapid when the enzyme was stored under N2 than under air, indicating that the enzyme is not particularly sensitive to 02. Finally, the hydrogenase of Chromatium showed no loss of activity after exposure to 100% O2 for 1 hour (Serra et al., 1979).

178

PAULETTE M. VlGNAlS ET AL.

2. Stability against Heat Inactivation Hydrogenases are resistant to temperatures up to about 80°C in T. roseopersicina (Gogotov e? al., 1978b) and in R. capsulata strain B10 (Colbeau and Vignais, 1981), and up to 70°C in R. rubrum (Adams and Hall, 1979). The thermal inactivation rate depends on the gas phase: for example, inactivation of the T. roseopersicina hydrogenase at 80°C was incomplete even after 3 days under air, but under argon it was complete within 24 hours, occurring as a two-step process with a Ki of 7 X 10 -5 sec - l and 1.8 x 10 -6 sec -*,respectively (Zorin and Gogotov, 1982). The rate of inactivation is greater when the enzyme is in the membrane, or bound to hydrophobic phenyl-Sepharose or DEAE-cellulose, than when it is solubilized, perhaps because immobilization prevents a conformational change to a more stable form. In contrast, Klibanov et al. (1980) found that the hydrogenase of Chromatiurn uinosum became heat labile (irreversible inactivation at 65°C in less than 20 minutes) after solubilization from the membrane, whereas complete stabilization was achieved only when the enzyme was bound covalently to a hydrophobic support. The R. rubrum hydrogenase is also more thermostable in the membrane-bound form than in the soluble one. This enzyme is cold labile and the cold lability was lost on solubilization (Adams and Hall, 1977), whereas the cold lability of R. capsulata B 10, hydrogenase appears only on solubilization (Colbeau and Vignais, 1981). 3 . Stability against Denaturing Agents

The hydrogenases of photosynthetic bacteria generally show a high resistance to denaturing agents. Urea, up to 5 M concentration, had no effect on the membrane-bound hydrogenase of Chromatium, or on the soluble hydrogenases of T. roseopersicina and R . capsulata strain 37V4 (Zorin and Gogotov, 1982). These hydrogenases are also resistant to dimethyl sulphoxide (up to 20% for the enzyme from Thiocapsa and 50% for that from Chromatiurn) and to 0.1% sodium dodecyl sulphate (SDS) (Zorin and Gogotov, 1982). Furthermore, it was not possible to displace the [4Fe-4S] cluster of the hydrogenase from Chromatium by treatment with thiol reagents, although the electron paramagnetic resonance spectra of the enzyme showed some changes (Strekas and Krasna, 1978). In contrast, isolated R. capsulata hydrogenase was completely inactivated by treatment with 0.1% SDS or 0.5 M urea (A. Colbeau, unpublished results), suggesting a structural difference between this and the other hydrogenases described above.

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

179

D. STRUCTURE AND MOLECULAR PROPERTIES

Hydrogenases have been isolated and purified to homogeneity from four species of photosynthetic bacteria: Chromatium (Gitlitz and Krasna, 1975; Kakuno et al., 1977; Llama et al., 1979; Strekas et al., 1980; van Heerikhuizen et al., 1981), R . rubrum (Adams and Hall, 1977, 1979), T. roseopersicina (Gogotov et al., 1978b), and R . capsulata (Colbeau and Vignais, 1981 ; Colbeau et al., 1983). These hydrogenases are intrinsic membrane proteins, and generally require detergent in order to be extracted from the membrane and solubilized; however, the hydrogenase of Chromatium has also been released from the membrane after mechanical treatment (van Heerikhuizen et al., 1981). 1 . Molecular Properties The R. rubrum hydrogenase comprises a single polypeptide chain of 65,000-67,000 Da, as determined by gel filtration and SDS-polyacrylarnide gel electrophoresis, and a p l of around 5.2 (Adams and Hall, 1977, 1979). A similar molecular weight (65,000 & 2000) and p l was recently determined for the R. capsulata hydrogenase (Colbeau et al., 1983). The hydrogenase isolated from T. roseopersicina (MW 68,000) dissociates in SDS into two subunits of 25,000 and 47,000 Da (Gogotov et al., 1976, 1978b). The soluble and the mernbrane-bound forms have an identical p l (4.2), redox potential (-270 rnV at pH 7), and amino acid composition, with alanine and glycine as N-terminal amino acids; they therefore probably represent the same enzyme. Gitlitz and Krasna (1975) isolated, after isoelectric focusing, two forms of hydrogenase from Chromatium differing only by their p l values (4.2 and 4.4). Since each form gave rise to a mixture of the two after a second isoelectric focusing, the authors concluded that the appearance of two forms was an artefact, possibly due to the complexing of hydrogenase with ampholines. The enzyme is a dimer of 100,000 Da with a single type of subunit. Llama et al. (1979, 1981), using a different purification process without heat treatment, separated two forms after ion-exchange chromatography, which differed in their activity and in their molecular weight (37,000 and 55,000 for forms I and 11, respectively) but contained the same type of subunit (3 1,000 Da); the two forms presumably represent different aggregation states of the same enzyme. On the other hand, the hydrogenase of Chromatium, solubilized without detergent by grinding the cells with sand (van Heerikhuizen et al., 1981), was composed of a single polypeptide chain of about 62,000 Da, in good agreement with Kakuno et

180

PAULETTE M. VlGNAlS ET AL.

al. (1977), who found a molecular weight of 68,000. The discrepancies between these results and those of Gitlitz and Krasna (1975) have not so far been explained, but the method used to solubilize and purify the enzyme is probably of importance. The amino acid composition has been determined for the hydrogenases of R. capsulata (Colbeau et al., 1983), Chromatiurn (Gitlitz and Krasna, 1975), and T. roseopersicina (Gogotov er al., 1978b). The compositional relatedness of the hydrogenase protein of R . capsulata with the hydrogenase proteins from other organisms was assessed by computing the compositional difference, SAQ (Marchalonis and Weltman, 1971). The SAQ values are obtained by comparing pairwise the amino acid composition of two proteins and the squares of the differences are summed. The median value for totally unrelated proteins has been found to be 300 in this type of comparison, whereas the median values within families of related proteins (haemoglobins, cytochromes, and immunoglobulins) vary from 20 to 80. Furthermore, a significant correlation was found between SAQ values and differences in amino-acid sequence (Marchalonis and Weltman, 1971). The amino acid composition of membrane-bound hydrogenases (from R. capsulata, T. roseopersicina, Chromatiurn, Proteus mirabilis, Desuljiovibrio vulgaris, D . desuljiuricans, Norway) presented a high degree of homology as indicated by SAQ values in the range of 50-80. On the other hand, R. capsulata hydrogenase showed no relationship to the soluble hydrogenase of Clostridium pasteurianum (SAQ = 266) (Colbeau et al., 1983). The visible spectra of the hydrogenases of photosynthetic bacteria show a broad absorption band at 400-440 nm which decreases in intensity by 10-15% after reduction by dithionite (Gogotov et al., 1978b; Gitlitz and Krasna, 1975; Adams and Hall, 1979; Strekas et al., 1980). These spectroscopic properties are typical of iron-sulphur proteins. Moreover, the spectrum of hydrogenases from Chromatium in 80% dimethyl sulphoxide is similar to that of other proteins having a [4Fe-4S] cluster (Strekas et al., 1980). The amount of iron and labile sulphur in these hydrogenases is consistent with the presence of a single [4Fe-4Sl cluster.

2 . Uptake Hydrogenases Are Nickel Enzymes Nickel is required for the synthesis and activity of various enzymes, including some hydrogenases (see Kaltwasser and Frings, 1981;Thauer et al., 1980 for reviews). Thus, nickel has been shown to be present in the uptake hydrogenases isolated from A. eutrophus (Friedrich et al., 1981b, 1982), from Methanobacterium thermoautotrophicum (Graf and Thauer,

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

181

1981; Jacobson et al., 1982; Albracht et al., 1982b; Kojima et al., 1983), from Desulfouibrio gigas (Cammack et al., 1982; Le Gall et al., 1982; Kriiger et al., 1982; Teixeira et al., 1983; Lalla-Maharajh et al., 1983),and from Vibrio succinogenes (Unden et al., 1982). The importance of nickel for the synthesis of hydrogenase has also been demonstrated in various N2-fixing species (Partridge and Yates, 1982; Pedrosa and Yates, 1983). Among the photosynthetic bacteria, evidence for the involvement of nickel in hydrogenase activity has been obtained for R. capsulata (Takakuwa and Wall, 1981; Colbeau and Vignais, 1983) and for C. vinosum (Albracht e f al., 1983). In R. capsulata strain B100, Ni2+at optimal concentration (8 PM),stimulated the hydrogenase activity of malate-glutamate-grown cells two- to fivefold and that of malate-NH4+-grown cells 10-fold. In a later study, with R. capsulata strain B10, nickel was found to stimulate hydrogenase activity in media which are de-repressing for hydrogenase synthesis (e.g., malate-glutamate medium) but no stimulation could be observed in a malate-NH4+ medium, in which hydrogenase activity is normally repressed (Colbeau and Vignais, 1983). The reason for this difference is not known, but it may be due to a different regulation of hydrogenase in the two strains. The presence of nickel in the hydrogenase of R. capsulata strain B10 was demonstrated by incorporation of ‘j3Niinto the hydrogenase protein. Thus, the radioactive label was found to comigrate with the protein band showing hydrogenase activity, both in nondenaturing polyacrylamide gels, and after rocket and crossed immunoelectrophoresis in agarose gels containing specific hydrogenase antibodies (Colbeau and Vignais, 1983). Furthermore, the incorporation of ‘j3Niinto the chromatophore membranes was strikingly lower when the experiments were repeated with a hydrogenase-less (Hup-) mutant (Colbeau and Vignais, 1983). The presence of nickel in M . thermoautotrophicum hydrogenase also has been demonstrated by electron paramagnetic resonance (EPR) spectroscopy (Albracht et al., 1982b). In EPR spectra, nickel produces a rhombic signal with g values around 2.3,2.2, and 2.0, detected at temperatures above 30 K. Below this temperature, the signal due to nickel is masked by the strong signal due to the Fe-S cluster(s). The signal of Ni(II1) was unambiguously identified by 61Ni substitution, which produced a nuclear hyperfine splitting mainly visible on the g, line (Lancaster, 1980; Albracht et al., 1982b). The signal disappeared after reduction of the enzyme by H2 or dithionite, and its oxidation-reduction potential was pH dependent (Cammack et al., 1982). Van Heerikhuizen et al. (1981) suggested that the Fe-S clusters may not be the primary site of interaction of H2 with hydrogenase since (1) none of the inhibitors of the hydrogenase activity had any effect on the

1UIZ

PAULETTE M. VIGNAIS E r AL.

shape or intensity of the EPR spectrum, although this spectrum is very sensitive to changes in the vicinity of the cluster; (2) the cluster is reduced by ascorbate (Eh = 50 mV), so it must have a redox potential around 0 mV that is too high to allow a direct interaction with H,; (3) the EPR spectrum remains unchanged upon replacement of H20 by 2H20,indicating that solvent protons are not in magnetic interaction with the cluster. Nickel is now thought to be the primary binding site for H2, and the fact that H2 reduces the nickel preferentially to the Fe-S cluster (Teixeira et al., 1983) agrees well with this hypothesis. With C. uinosum, Albracht et al. (1982a, 1983) have shown that the active enzyme contains a [4Fe-4S]3+(3+,2+) cluster in magnetic interaction with nickel. Due to their interaction, the two paramagnets cannot be detected in the spectrum. Upon partial reduction with ascorbate plus phenazine methosulphate, the interaction is broken down, probably by destabilization of the Fe-S cluster, and the signals due to the Fe-S cluster and nickel can be detected. Under these conditions, the Fe-S cluster is converted to a [3Fe-xS] cluster. The variable amounts of EPR-detectable nickel in different hydrogenases may therefore be due to a partial conversion of the active [4Fe-4S] centres into the active [3Fe-xS] form during purification.

E . CATALYTIC PROPERTIES

I . Hydrogenase Assays

Hydrogenase catalyses the following reversible reaction: Hz

2H+ + 2e- (Eh = -420 mV)

(3)

It is therefore possible to assess hydrogenase activity by measuring either H2 consumption or H2 production. A third method is based on the fact that hydrogen atoms from the H2 molecule are first bound to the enzyme during the activation of H2 and are then exchangeable with the solvent protons (for review see Krasna, 1979; Adams et al., 1981). a . Uptake ofhydrogen. Uptake of H2 is measured in the presence of an electron acceptor which can interact, directly or indirectly, with hydrogenase. Either the consumption of H2 is measured (manometrically, amperornetrically, or by gas chromatography) or the reduction of the electron acceptor, usually a dye whose reduction may be followed spectrophotometrically. The electron acceptors currently used are the viologens-benzylviologen (Eh = -360 mV) and methylviologen (Eh = -446 mV)-and methylene blue (Eh = + I 1 rnV at pH 7.0). The latter gives large changes in absorbance, and is often used with whole cells or

HZ METABOLISM IN PHOTOSYNTHETIC BACTERIA

183

membrane preparations, but is less specific since it can accept electrons from donors at a redox potential higher than that of the hydrogen electrode. The assays are carried out under strictly anaerobic conditions since the reduced forms of the electron acceptors used are rapidly auto-oxidized in the presence of oxygen.

b. Evolution of hydrogen. Evolution of H2 occurs when hydrogenase is incubated with a reduced acceptor whose redox potential is close to that of the H2electrode (-420 mV at pH 7), e.g., methylviologen. In this case, the hydrogenase-catalysed reaction is as follows: 2MVt

+ 2H+

-

2MV’+

+ Hz

(4)

The H2 produced is measured manometrically or by gas chromatography. The most commonly used electron donor is dithionite-reduced methylviologen, although electrolytically reduced viologen dyes also can be used (Thorneley, 1974; Vignais et al., 1982); physiological reductants such as ferredoxin or cytochrome c3are generally inactive with the hydrogenases from photosynthetic bacteria (see below). c . Exchange reaction. The exchange reaction is catalysed at the active site of hydrogenase and must be carried out in the total absence of 02. Exchange is measured either between H2 gas and deuterated or tritiated water, or between one of the isotopes of H2 (deuterium or tritium) and HzO. When 2H2or heavy water is used, labelled and unlabelled molecular hydrogen are analysed by mass spectrometry. When tritium gas is used, the radioactivity of tritiated water is counted, tritiated water being formed according to the following reaction: H3H + H 2 0

-

H2 + H03H

(5)

2. Physiological Electron Acceptor or Donor The primary electron donor andlor acceptor has been identified only for some (soluble) hydrogenases. The hydrogenase of Clostridium pasteeurianum reacts directly with ferredoxin (Nakos and Mortenson, 1971), and that of D . vulgaris reacts with cytochrome c3 (Yagi et al., 1968). In photosynthetic bacteria, the electron carrier which accepts electrons from hydrogenase is unknown. For membrane-bound hydrogenases it is presumably either a ubiquinone, as in hydrogen bacteria such as P . denitrificans (Porte and Vignais, 1980; Henry and Vignais, 1983), or a cytochrome b as reported for Rhizobium japonicum (Eisbrenner and Evans, 1982).

184

PAULETTE M. VlGNAlS €7 AL.

There is no direct interaction between ferredoxins of various origins and the hydrogenases of Chromatium (Gitlitz and Krasna, 1975; Kakuno et al., 1977), T . roseopersicina (Gogotov et al., 1978a), or R. capsulata (Colbeau and Vignais, 1981). However, the R. rubrum hydrogenase evolves H2 in the presence of ferredoxins I and I1 from the same bacterium (Adams and Hall, 1979). The purified hydrogenases of photosynthetic bacteria cannot utilize nicotinamide nucleotides directly, either as electron donors or acceptors (Adams and Hall, 1979; Colbeau and Vignais, 1981; Gogotov et al., 1978a,b), although in cell-free extracts these nucleotides are reduced in the presence of H2 (Klemme, 1969; Adams and Hall, 1979). The hydrogenase of T. roseopersicina can utilize reduced cytochrome c3 (from the same bacterium) as electron donor, and the rate of H2 production is further increased by the addition of ferredoxin from Chromatium (Gogotov, 1978). The hydrogenase of Chromatium can use both reduced cytochrome c3 and reduced flavin mononucleotide (FMN) as electron donors. It is possible that a flavoprotein component normally bound to the enzyme can couple with exogenous FMN (Kakuno et al., 1977).

3 . Kinetic Parameters In T. roseopersicina, the hydrogenase, which has a redox potential of about -280 mV, does not reduce methylene blue, ferricyanide, or azocarmine. The rate of reduction of benzylviologen is 15 times higher than the rate with methylviologen (Gogotov et al., 1978a). With the hydrogenases from R. rubrum (Adams and Hall, 1979) and R. capsulata (Colbeau and Vignais, 198l), methylene blue and benzylviologen are the best substrates for H2 uptake and have a high affinity for the . affinity for H2 is very high enzyme (apparent K , lower than 100 p ~ )The . H2 uptake reaction is as indicated by a K , for H2(uptake) of 0.25 p ~The pH dependent with a maximum in the pH range 7.5-9, which varies with the mediator used. The rate of H2 production by hydrogenase, with dithionite as electron donor, depends partially on the redox potential of the mediator. With R. rubrum, no H2 production is observed with mediators having E f ovalues greater than 50 mV (Adams and Hall, 1979). With dithionite-reduced methylviologen, the rates of H2 evolution by purified hydrogenase preparations from R. rubrum (Adams and Hall, 1979), from Chromatium (Gitlitz and Krasna, 1975), and from T. roseopersicina (Gogotov, 1978) are 26,35, and 60 pmol min-I mg protein-', respectively (see Bagyinka et al., 1984, for discussion of H2 determinations). The rate of H2 evolution

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

185

shows a sharp pH maximum in the pH range 5.5-7 (Adams and Hall, 1979; Colbeau and Vignais, 1981). It is known that the reducing power of dithionite decreases sharply at pH values below 8, thus decreasing the true concentration of methylviologen semiquinone in the assay (Mayhew, 1978). To avoid such complications, methylviologen was reduced electrochemically: under these conditions, a pH optimum of 5.7 was found for the hydrogenase of R. capsulata (Colbeau and Vignais, 1981). Despite the similarity in properties of these hydrogenases (e.g., pH dependence), the ratio of H2 uptake and H2 evolution activities, measured at their optimal pH, varies by several orders of magnitude. Thus, this ratio of H2 uptake to H2 evolution activities is 100 for R. capsulata hydrogenase (Colbeau and Vignais, 1981), 10 for Chromatium hydrogenase (Gitlitz and Krasna, 1975), and 0.2 for R. rubrum hydrogenase (Adams and Hall, 1979). Further research on the structure and catalytic mechanism of hydrogenase is required to explain this surprising variation. The two forms of hydrogenase that have been isolated from Chromatium differ in their kinetic characteristics (Llama et al., 1979, 1981) and may represent monomeric and dimeric forms of the enzyme. Form I showed hyperbolic kinetics, as confirmed by the Hill plot (Hill coefficient of l), whereas for form 11, a Hill coefficient of 0.68 indicated an apparent negative cooperativity . The affinity for reduced methylviologen of form I ([S]O.s= 19 p ~was ) higher than that of form I1 ([S]0,5= 364 p ~ ) When . form I1 disaggregated at high ionic strength, the affinity for reduced ) the methylviologen increased ([S]O.sdecreasing from 364 to 160 p ~and kinetics became more hyperbolic with an increase in the Hill coefficient from 0.68 to 0.86 (Llama et al., 1981). The hydrogen-deuterium (or hydrogen-tritium) exchange activity, like the H2 evolution activity, is maximal at acidic pH values, for example pH 6 for hydrogenase from Chromatium (Gitlitz and Krasna, 1975) and pH 4.5 for that from R. capsulata (Colbeau and Vignais, 1981). With T. roseopersicina, the exchange activity was lost during purification of hydrogenase but could be restored after addition of dithionite (Gogotov, 1978). The exchange activity was 265 pmol of 2H2 hr-' mg protein-' which compared well with the rate of H2 evolution from dithionite-reduced methylviologen (300 pmol of H2 hr-1 mg protein-') (Gogotov et al., 1978b).

F. A N INDUCIBLE ENZYME

Hydrogenases in general appear to be inducible enzymes. This is the case, for example, in aerobic bacteria such as Xanthobacter (Berndt and Wolfe,

186

PAULETTE M. VlGNAlS ET AL.

1978), Aquaspirillum autotrophicum (Aragno and Schlegel, 1978), and A . eutrophus H16 (Friedrich et al., 1981a), and also in the Enterobacteria (Krasna, 1980). With P. denitrificans, however, hydrogenase was found to be constitutive in all strains except one, the strain Stanier 381 (DSM 65) (Nokhal and Schlegel, 1980). Among the photosynthetic bacteria, the regulation of hydrogenase synthesis has been studied only in R . capsulata (Colbeau and Vignais, 1983). The increase in hydrogenase activity observed under certain growth and incubation conditions was shown to be due to an increase in hydrogenase synthesis. Thus, it was accompanied by an increase in the level of hydrogenase protein (as assayed by quantitative rocket immunoelectrophoresis) and was prevented by inhibitors of protein synthesis (chloramphenicol and tetracycline) and of mRNA synthesis. Since a low level of hydrogenase (about 10% of the maximum) was present under non-inducing conditions, the synthesis of hydrogenase in R . capsulata appears to be semi-constitutive rather than fully inducible (Colbeau and Vignais, 1983). The factors involved in the regulation of hydrogenase synthesis are poorly understood. Carbon substrates repressed completely the synthesis of hydrogenase in P. denitrificans strain Stanier 381 (DSM65), in A. eutrophus (Schlegel and Eberhardt, 1972; Friedrich et al., 1981a), and in Rhizobium japonicum (Maier et al., 1979), and the highest activity was generally found during autotrophic growth under a H2:CO2 :O2 atmosphere. However, in A . eutrophus, during growth on poor carbon substrates such as glycerol, the hydrogenase activity was higher than under nautotrophic conditions (Friedrich et al., 1981a). It appears, therefore, that H2 is not the true inducer of hydrogenase. In chemostat-grown cultures of A . eutrophus, the synthesis of hydrogenase is de-repressed when the electron donor, either organic or inorganic (i.e., H2), is limiting (Friedrich, 1982). In R . japonicum, studies with mutants have revealed that a common element is involved in the repression of hydrogenase by carbon substrates and oxygen (Merberg et al., 1983). Schlegel and Eberhardt (1972) noted that the synthesis of hydrogenase is de-repressed when the nitrogen sources are in an oxidized form (N2, N03-), but there is no synthesis of hydrogenase when the nitrogen source is in a reduced form (NH4+).In N2-fixing bacteria such as R . capsulata, Azotobacter, and Xanthobacter, hydrogenase synthesis is dependent on the nitrogen source; thus, hydrogenase activity is low in the presence of ammonium salts, which also repress nitrogenase synthesis (Klemme, 1968; Berndt and Wolfe, 1978; Colbeau et al., 1980). These data raise the question of the relationship between hydrogenase and nitrogenase. Although Siefert and Pfennig (1978) concluded from studies of mutants of R . acidophila that there is a regulatory linkage between nitrogenase and

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

187

hydrogenase, with R . capsulata it has been shown that the synthesis of the two enzymes is independent (Colbeau et al., 1980); however, Hz evolved by nitrogenase can indirectly regulate the concentration of hydrogenase (Colbeau et al., 1980; Tsygankov et al., 1982a,b). Berndt and Wolfle (1978) have suggested that the induction of the (membrane-bound) hydrogenase might be due to a higher degree of energization of the membrane; thus, under Nrfixing conditions, the increase of electron flow in the membrane would allow an increase in hydrogenase synthesis.

G. PHYSIOLOGICAL ROLE OF HYDROGEN UPTAKE

1. Oxy-Hydrogen Reaction

a . Electron transfer and coupled phosphorylation. In R . capsulata cells,

H2 is very rapidly oxidized in the presence of O2either in the light or in the dark (Meyer et al., 1978a). Oxidation of HZdoes not result from a direct interaction between hydrogenase and oxygen; instead, hydrogenase transfers electrons from H2 to the membrane-bound electron transport chain (Fig. 1). In R . capsulata, the respiratory chain is branched and comprises two terminal oxidases having markedly different sensitivities to cyanide (Marrs and Gest, 1973; La Monica and Marrs, 1976; Zannoni et al., 1974, 1976a,b) and one of them being linked to energy conservation (Baccarini-Melandri et al., 1973; Zannoni et al., 1976a). Both terminal oxidases are used during H2 oxidation, as indicated by a biphasic inhibition by cyanide (Paul et al., 1979) and by the fact that mutants of R . capsulata impaired in either one of the terminal oxidases are capable of H2 oxidation (Melandri et al., 1982). Furthermore, mutants of R . cupsulata lacking either one of the terminal oxidases can grow under chemoautotrophic conditions (Madigan and Gest, 1979). Since the two pathways branch at the level of the quinone pool, the primary electron acceptor from hydrogenase is probably a quinone (Melandri et al., 1982); although this has not yet been clearly demonstrated. Oxidation of H2 is coupled to ATP synthesis and the P/O values for H2 oxidation are as high as for NADH oxidation (Paul et al., 1979; Melandri et al., 1983). However, H2 and NADH do not share the same electron pathways to ubiquinone, since rotenone does not inhibit electron transport from H2 (Paul et al., 1979). This has been found also for other hydrogenases, for example in R . japonicum (O’Brian and Maier, 1982) and P . denitrijicans (Sim and Vignais, 1978; Porte and Vignais, 1980) and is consistent with the fact that these hydrogenases cannot transfer electrons to NAD+. The PI0 values for H2 oxidation are twice those for succinate

188

PAULETTE M. VlGNAlS ET AL.

oxidation (Paul et al., 1979; Melandri et al., 1983). Thus, the pathway from H2 must involve at least one more site of energy conservation compared to the pathway from succinate; indeed, it has been demonstrated that the electron flow from H2 to ubiquinone is linked to ATP synthesis (Melandri et al., 1983; see also Fig. 1). Various Chromatiaceae are able to grow chemo-autotrophically under semi-aerobic or aerobic conditions with H2 as electron donor (Gibson, 1967; Kampf and Pfennig, 1982). This implies that the oxidation of H2 is coupled to energy production. Indeed, it has been shown with whole cells of Chromatium sp. strain Miami PBS 1071 that the oxidation of H2 is accompanied by an outward H+ translocation; the H+le- stoicheiometries indicated that two sites of energy coupling are present (Kumazawa et al., 1983). In this bacterium the respiratory chain linked to H2 oxidation has not been characterized.

b. Oxygen scavenging. Dixon (1972, 1978) has proposed that, in N2-fixing organisms, the oxy-hydrogen reaction may help to reduce the O2 tension and therefore protect nitrogenase against O2 inactivation. Indeed, in R . capsulata, Song et al. (1980) demonstrated that nitrogenase activity was better protected against O2 inhibition under an atmosphere of H2 than under an atmosphere of argon, even in the presence of 20 mM malate. This means that H 2 was used preferentially to malate for O2 consumption. This preferential use of H2 is probably linked to a greater diffusibility of hydrogen gas compared to that of malate and/or to the very high affinity of hydrogenase for H2 ( K , 0.25 PM for the R . capsulata hydrogenase; Colbeau and Vignais, 1981). It has also been shown with whole cells of Chromatiurn that H2 oxidation represses endogenous respiration (Kumazawa et al., 1983). 2. Hydrogen Recycling Re-uptake of hydrogen (produced by nitrogenase) has been observed in R. capsulata. Resting cell suspensions producing H2 form lactate in the light were found to reverse the direction of H2 metabolism when the organic substrate was practically exhausted, and rapidly consumed the H2 that had accumulated (Kelley et al., 1977). In the absence of oxygen, H2 can serve as electron donor either to nitrogenase or for the reduction of C02 (Fig. 3). a . Hydrogen recycling to nitrogenase. Meyer et al. (1978a) and Song et al. (1980) demonstrated that addition of H2 stimulates acetylene reduction

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

189

FIG. 3. Scheme illustrating the role of hydrogenase in the recycling of H2 for photoreduction of C 0 2 .

by R . capsulata cells depleted of endogenous substrates. This corresponds to a recycling of electrons from H2 to nitrogenase, as has been observed in other bacteria such as Azotobacter chroococcum (Walker and Yates, 1978) and Rhizobium (Dixon, 1978). In uiuo, H2 recycling to nitrogenase, like nitrogenase activity with organic substrates, is inhibited by metronidazole, an inhibitor of low potential electron camers such as ferredoxin or flavodoxin (Hallenbeck and Vignais, 1981; Kelley and Nicholas, 1981). In chemostat cultures of R . capsulata, grown with N2 as sole nitrogen source, no net H2 production is observed (Jouanneau, 1982). Since H2 is an intrinsic product of the nitrogenase reaction (Mortenson, 1978; see Section V), this implies that H2 is recycled under these conditions. This was confirmed by Takakuwa et al. (1983) who showed that, in a hydrogenase-deficient (Hup-) mutant of R . capsulata, nitrogenase-mediated Hz formation occurs in N2-fixing cultures. b. Hydrogen recycling for the photoreduction of carbon dioxide. Schick (1971b,c) observed that the photoreduction of C02 by H2 in R . rubrum is strongly influenced by availability of organic substrates; H~dependent C 0 2 photoreduction is at a maximum only when organic substrates have been consumed (Schick, 1971b). On the other hand, nitrogenase-mediated H2 production occurs at maximal rates only when organic substrates are in excess (Schick, 1971b). Small amounts of ammonium chloride stimulated the photoreducing activity of resting cells, and addition of L-malate resulted in a substratedependent evolution of gas (Schick, 1971~).The amount of gas evolved depended on the ratio of L-malate to ammonium chloride, and a stoicheiometry of 2.2 mol of malate consumed for each mole of ammonium chloride added was calculated. These results indicate that photoreduction activity is controlled by the intracellular concentration of ammonium ion (or of some equivalent). In other words, nitrogen fixation and photoreduction are intimately linked: L-malate serving as the electron donor to nitrogenase and as an acceptor for NH3, the end product of nitrogen fixation. Cells maintain a high photoreducing activity as long as their end product

190

PAULETTE M. VlGNAlS ET AL.

is removed via reductive amination (Schick, 1971~).These results were confirmed by Hillmer and Gest (1977b) who observed that organic compounds inhibited the Hz-dependent photoreduction of C02 in resting cells of R . capsulata whereas ammonia stimulated H2 uptake by these cells. Jouanneau et al. (1980a) demonstrated by mass spectrometry that reuptake of H2 by R . capsulata cells depleted of organic substrates is indeed associated with COZphotoreduction. Stoppani et al. (1955) showed that the major product of COz assimilation in light is glutamic acid. More recently, Khanna et al. (1980) demonstrated that C02 stimulates nitrogenase-related activities in R . capsulata. Khanna et al. (1981) observed that 14C02fixation was markedly inhibited under aerobic conditions. By the use of respiratory inhibitors (CO, NaN3) these authors came to the conclusion that oxygen and C02 may compete for a common reductant, possibly a reduced nicotinamide nucleotide.

V. Nitrogenase

A. BIOCHEMISTRY OF NITROGENASE

I . Structure and Molecular Properties The nitrogenase complex was first isolated from an aerobic non-photosynthetic bacterium, Azotobacter uinelandii (Bulen and Lecomte, 1966). It was shown to consist of two proteins, a molybdenum and iron-containing protein (also known as MoFe protein, Component I, or dinitrogenase) and an iron-containing protein (Fe protein, Component 11, or dinitrogenase reductase). Separately, each of these two proteins has no activity; the two together form the active nitrogenase complex. Nitrogenases have now been isolated and purified from a wide range of N2-fixing micro-organisms, including photosynthetic bacteria. All these enzymes are very similar to one another in terms of their size (molecular weight), amino acid composition, and reactivity (see Winter and Burris, 1976; Mortenson and Thorneley, 1979; Burgess, 1984, for review). The MoFe protein is an a2p2tetramer of molecular weight 200,000250,000. It contains two molybdenum, about 30 iron, and 30 inorganic sulphur atoms present as sulphide (P). Sixteen out of the 30 Fe atoms are associated with S2- in four cubic [4 Fe-4SI clusters. The remaining metal atoms are arranged in two copies of a co-factor called FeMo cofactor (FeMoco) with a minimum stoicheiometry of MoFe6Sg-9(Nelson et

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

191

al., 1983). The co-factor can be isolated from the MoFe protein by extraction into N-methylformamide (Shah and Brill, 1977). The FeMoco is thought to be the site of substrate binding and reduction. The Fe protein is a dimer of about 60,000 molecular weight. The two subunits are identical and their amino acid sequences have been determined for six bacterial species. Although Fe proteins have molecular weights ranging from 29,685 for Clostridium pasteurianum to 33,000 for Anabaena, they show a high degree of homology (Burgess, 1984). Both MgATP and MgADP bind to the Fe protein at two different sites. This binding causes a shift down of the mid-point redox potential by -50 to -100 mV, as well as a conformational change of the Fe protein that enables the MgATP-Fe protein complex to reduce the MoFe protein (see Section V.A.2). Among the photosynthetic bacteria, the isolation and purification of nitrogenase in an active form has been accomplished for only two species, Rhodospirillum rubrum (Nordlund et al., 1978; Ludden and Burris, 1978) and Rhodopseudomonas capsulata (Hallenbeck et al., 1982a), although Evans et al. (1973) succeeded in purifying to homogeneity the MoFe protein from Chromatium uinosum, strain D. Molecular properties of the nitrogenase proteins from R. rubrum and R . capsulata are compared in Table 4. The subunit composition and metal content are similar in both species and fall in the range of those commonly observed for nitrogenases from non-phototrophs. Comparison of the amino acid composition of the nitrogenase polypeptides of R . capsulata with those of other sources (Hallenbeck et al., 1982a), as well as hybridization experiments with the structural nifgenes (Ruvkun and Ausubel, 1980), showed that the enzyme isolated from photosynthetic bacteria is similar to that from other species. However, in members of the Rhodospirillaceae, an unusual mechanism for the regulation of nitrogenase activity has been discovered, which involves a covalent modification of the Fe protein (Nordlund and Baltscheffsky, 1973; Ludden and Burris, 1976; Nordlund et al., 1977). Thus, one of the subunits is modified by covalent fixation of an unidentified group consisting of adenine, phosphate, and pentose which results in loss of the catalytic activity of nitrogenase. Conversely, activation of nitrogenase is catalysed by a specific membrane-bound enzyme which removes the modifying group from the Fe protein. This regulatory system (described in Section V.B.3) has been studied in R. rubrum (Gotto and Yoch, 1982a; Preston and Ludden, 1982; Ludden et al., 1982b) and in R. capsulata (Michalski et al., 1983; Jouanneau et al., 1983). None of the Fe proteins isolated so far from other nitrogen-fixing micro-organisms have exhibited such regulatory properties (see Burgess, 1984). However, nitrogenase from Azospirillum required activation in vitro (Ludden et al., 1978), sug-

PAULETTE M. VlGNAlS ET AL.

192

TABLE 4. Molecular properties of the nitrogenases from Rhodospirillum rubrum and Rhodopseudomonas capsulata"

Species Fe protein Molecular weight Subunits Active Inactive Fe content (atom molecule-')

R hodospirillum rubrum

Rhodopseudomonas capsulata

61,500

63,000

2 x 30,000 30,000 + 31,500 3.5

2 x 33,500 33,500 + 38,000 2.7 - 4.1

230.000

230.000

Mo-Fe protein Molecular weight Subunits a

2 2

P Metal content (atom molecule-I) Mo Fe

X X

58,500 58,500

2 x 55,000 2 x 59,500

1.7 20

1.3 27.7

a Data for R . rubrum are from Ludden and Bums (1978), and those for R . capsulata are from Hallenbeck et al. (1982a), except for the subunit composition of the active and inactive Fe protein which were determined by SDS-polyacrylamide gel electrophoresis (Gotto and Yoch, 1982a; Ludden et al., 1982b; Jouanneau et al., 1983).

gesting that this micro-organism also possesses a regulatory system similar to that found in Rhodospirillaceae.

2. Catalytic Mechanism: Energetics Catalytic properties of the nitrogenase complex have been studied mainly in enzymes from non-phototrophic micro-organisms, but, because of the molecular homologies that exist between nitrogenase proteins, they can be considered as general properties. Nitrogenase reduces molecular nitrogen according to the following reaction: Nz+ 6H+ + 6e-

+ nMgATP-

2NH3

+ nMgADP + nPi

(6)

and in the absence of N2 it reduces protons (Bulen et al., 1965): 2H'

+ 2e- + nMgATP-

H 1 + nMgADP + nP,

(7)

Nitrogenase also reduces triple-bonded compounds other than N2, including cyanide and acetylene, as well as nitrous oxide and protons (Fig. 4).

H2

METABOLISM IN PHOTOSYNTHETIC BACTERIA

193

240r

Time (minutes)

FIG. 4. Inhibition of nitrogenase-mediated Hz production by acetylene in resting cells of R. capsulata and restoration of Hz photoproduction by carbon monoxide. Resting cells from a phototrophic culture of R.capsufuta,grown on lactate, were suspended in medium containing lactate and incubated under argon in the light, in the reaction vessel described by Jouanneau et a/. (1980a). Deuterated acetylene (C2D2)and CO were added at the times shown, and the dissolved gases (CZD2H2and H,) produced in the incubation medium were ionized and analysed by mass spectrometry. From Jouanneau et al. (1980a).

Acetylene reduction to ethylene (Dilworth, 1966) is most often used for in vivo and in vitro nitrogenase assays. Both a low potential reductant and MgATP are required for reduction of all substrates. The electron transfer sequence in nitrogenase is as follows: Reductant

- Fe protein

MoFe protein

-

Substrate

(for review see Mortenson and Thorneley, 1979; Lowe er al., 1980). The physiological electron donors are electron carrier proteins of midpoint redox potential around -400 m V (ferredoxins and flavodoxins). Chemical reductants such as dithionite are usually preferred for in virro

experiments. Turnover studies, based on EPR and substrate binding data, suggested that the nitrogenase reaction can be broken down into four steps: (1) one

194

PAULETTE M. VlGNAlS ET AL.

electron reduction of Fe protein and binding of MgATP; (2) complexation with the MoFe protein; (3) electron transfer to the MoFe protein coupled to ATP hydrolysis; and (4) dissociation of the complex (Mortenson and Thorneley, 1979; Burgess, 1984). Kinetic studies indicated that the ratelimiting step is the dissociation of the complex (Thorneley and Lowe, 1983). ATP hydrolysis may facilitate electron transfer to the MoFe protein since binding of MgATP to the Fe protein in C. pasteurianum lowers its apparent mid-point redox potential from -290 mV to -400 mV (Zumft et at., 1974). A minimum value of two ATP molecules hydrolysed per electron transferred was calculated and, therefore, the reduction of 1 mol of N2 requires at least 12 mol of ATP. However, since during N2 fixation ATP-linked proton reduction to H2 cannot be eliminated, the minimum ATP requirement for nitrogen fixation is probably greater than 12ATP/N2. Furthermore, this stoicheiometry varies widely as a function of pH, temperature, and the ratio of the nitrogenase components (Ljones, 1979). For R. capsulata nitrogenase, Hallenbeck (1983b) found an ATPI2e- ratio of about 4 under highly reducing conditions which increased drastically above the apparent mid-point potential (-470 mV), reaching values as high as 20.

3. Electron Transport to Nitrogenase The catalytic activity of nitrogenase involves hydrolysis of ATP and the oxidation of a low potential reductant. ATP can be generated either by photophosphorylation in the light or, in facultative phototrophs, by oxidative phosphorylation or fermentation in the dark. The physiological electron donors to nitrogenase are electron carrier proteins of redox potential around -400 mV, namely ferredoxins or flavodoxins (see Buchanan and Amon, 1970; Yoch and Carithers, 1979). Ferredoxins have been isolated from different phototrophic bacteria, fimicofu including C . uinosum (Bachofen and Amon, 1966), C~forobium (Tanaka et al., 1974, 1975), R . rubrum (Shanmugam et al., 1972; Yoch et al., 1975), and R . capsulata (Hallenbeck et al., 1982c;Yakunin and Gogotov, 1983). The last two species each contain two ferredoxins, one of which is much more efficient than the other in catalysing electron transfer to nitrogenase (Yoch and Amon, 1975; Yakunin and Gogotov, 1983). Interestingly, in R. capsulata, the ferredoxin that transfers electrons at higher rates appears to be synthesized only when nitrogenase is expressed (Yakunin and Gogotov, 1983). The electron pathways leading to the reduction of ferredoxin in phototrophic bacteria have not yet been elucidated. In green bacteria, direct photoreduction of ferredoxin has been ob-

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

195

served in vitro (Buchanan and Evans, 1969). Consistent with this, titration of the primary electron acceptor of the photosystem gave a mid-point potential of about -540 mV (Olson et al., 1976), low enough to reduce ferredoxin directly. In contrast, the chromatophores of purple bacteria are unable to produce sufficient light-driven reducing power to sustain nitrogenase activity in vitro, although ferredoxins isolated from several species could couple reducing power, generated by plant chloroplasts, to nitrogenase activity in an in uitro system (Yoch and Arnon, 1970, 1975; Hallenbeck et al., 1982c; Yakunin and Gogotov, 1983). These observations suggest that the bacterial photosystem is unable to reduce ferredoxin directly; indeed, redox titration of the primary electron acceptor, an iron-ubiquinone complex, gave values more electropositive (-50 to -200 mV) than the redox potential of ferredoxins (< -400 mV) (cf. Prince and Dutton, 1978). Light-dependent reduction of NAD+ has been observed in chromatophores from several species of Rhodospirillaceae (Keister and Yike, 1967; Jones and Saunders, 1972; Klemme, 1969). It appears to be an ATP-linked process since NAD+ reduction occurred in the dark when ATP was provided, and was inhibited in the light by inhibitors of photophosphorylation. Specific blockage of ATP synthase by oligomycin did not prevent NAD+ reduction whereas uncouplers did, indicating that the membrane potential generated by light, rather than ATP itself, serves as an energy source to drive electron flow to NAD’ (for review see Knaff, 1978). In conclusion, the light reaction in purple bacteria seems incapable of gener= -340 mV); ating a stable compound more reducing than NADH this excludes the possibility of a direct light-mediated reduction of ferredoxin. The occurrence of an enzyme that could couple light-mediated NADH generation to ferredoxin reduction (e.g., an NADH-ferredoxin oxidoreductase) has yet to be demonstrated.

B . REGULATION OF NITROGENASE ACTIVITY

1. The “Switch-off’ Effect The feedback inhibition of whole cell nitrogenase activity by ammonia, first observed in R . rubrum by Kamen and Gest (1949), is a common property of the Rhodospirillaceae, since it is also found in R . capsulata (Meyer et al., 1978a), Rhodopseudomonas palustris (Zumft and Castillo, 1978), and Rhodopseudomonas sphaeroides (Jones and Monty, 1979). Glutamine and other fixed nitrogen sources are known to provoke a similar effect. After removal of excess fixed nitrogen by washing or consumption by the cells, nitrogenase activity is completely restored. Because of

196

PAULETTE M. VlGNAlS ET AL.

its rapid onset, this regulatory phenomenon has been termed “switchoff’ effect (Zumft and Castillo, 1978). Several reports (Yoch and Cantu, 1980; Alef et al., 1981; Sweet and Burris, 1981) have mentioned that nitrogen nutrition has an effect on NH4+-mediated nitrogenase “switch-off ’. Bacteria grown on N2 or on glutamate synthesized a nitrogenase highly sensitive to NH4+ inhibition, while the nitrogenase of nitrogen-starved cells is relatively insensitive to ammonia inhibition. In addition, Yoch and Gotto (1982) observed that nitrogenase activity was inhibited by ammonia to a greater extent at low light intensity than at high light intensity. A satisfactory interpretation of these results will require a better understanding of the mechanism of nitrogenase “switch-off ’. As described in detail in Section V.B.3, loss of nitrogenase activity involves inactivation of the Fe protein by covalent modification. This conversion probably requires an enzyme, still unidentified, that might be the target of various effectors. Changes in the concentration of these effectors in response to factors such as the availability of fixed nitrogen, light and reductants could act on the regulatory system in such a way that it would switch off nitrogenase activity more or less efficiently depending on the environmental conditions. 2. Role of Glutamine Synthetase and Glutamine

Several observations indicate that NH4+-mediated nitrogenase ‘‘switchoff ’ occurs through an indirect mechanism involving glutamine synthetase (GS). In glutamine auxotrophic mutants of R. capsulata (Wall and Gest, 1979), nitrogenase is active even in the presence of ammonia. Moreover, when GS was specifically inactivated by L-methionhe-DL-sulphoximine (MSX), an analogue of glutamine, nitrogenase became insensitive to ammonia inhibition (Jones and Monty, 1979; Sweet and Burris, 1981; Alef et al., 1981; Yoch and Gotto, 1982). Glutamine synthetase activity, like nitrogenase activity, is regulated in response to ammonia supply: GS is fully active only under nitrogendeficient conditions and is inactivated by adenylylation in the presence of NH4+. In R. capsulata, concomitant inactivation by covalent modification of GS and of Fe protein were observed after an ammonia shock (Michalski et al., 1983). Since concomitant adenylylation of GS and inhibition of nitrogenase occurred on addition of NH4+and, since both enzyme activities resumed after NH4+ exhaustion, Hillmer and Fahlbush (1979) suggested that GS (more particularly the adenylylated form of GS) could act directly on the regulation of nitrogenase activity in R. cupsulata. However, in a further

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

197

study on R . capsulata, there appeared to be no correlation between the adenylylation state of GS and the activity of nitrogenase (Yoch, 1980). In nitrogen-starved cells of R . palustris, nitrogenase was shown to be insensitive to NH4+ whereas GS was still adenylylated (Alef et al., 1981). These authors concluded that the regulation of nitrogenase is independent of both the concentration and the adenylylation state of GS. After studying the effect of inhibitors of NH4+ assimilation on nitrogenase “switch-off ’, different authors proposed that glutamine is the effector molecule which triggers the “switch-off’ (Jones and Monty, 1979; Yoch and Gotto, 1982; Arp and Zumft, 1983b; Michalski et al., 1983). In R . capsulata cells permeabilized by toluene and incubated with [I4C]ATPlabelled on the adenine moiety, both GS and the Fe protein of nitrogenase became labelled after addition of ammonia (from 0.1 to 15 mM) or of 30 mM glutamine. The labelling of GS and of Fe protein was prevented by MSX (Michalski et al., 1983). Methionine sulphone, an inhibitor of glutamate synthase, potentiated the inhibition of nitrogenase by glutamine in R . sphaeroides (Jones and Monty, 1979). In R. rubrum, an analogue of glutamine, 6-diazo-S-oxo-2norleucine (DON), which blocks glutamate synthase activity, was found to inhibit nitrogenase (Yoch and Gotto, 1982). In both cases, the increase of the intracellular pool of glutamine due to glutamate synthase inhibition was thought to command the nitrogenase inhibiting system. Arp and Zumft (1983b) showed that glutamine effected nitrogenase inactivation even when GS was inactivated, and proposed that the role of GS in the ammonia-mediated “switch-off’ is to convert NH4+into the active effector, glutamine (see also Falk et al., 1982). In agreement with this, methylamine, an analogue of NH4+,appeared to require processing by glutamine synthetase before it was effective in establishing “switch-off’ (Yoch et al., 1983). These results clearly indicate that glutamine is involved in triggering nitrogenase “switch-off ’ on addition of NH4+.However, direct evidence that glutamine itself is the effector has yet to be provided; glutamine (like NH4+)has no effect on nitrogenase activity in vitro (Ludden and Burris, 1979). Furthermore, factors other than NH4+ provoke nitrogenase “switch-off”, including darkness, exposure to oxygen, and phenazine methosulphate treatment (Kanemoto et al., 1984). 3. Reversible Inactivation of the Fe Protein The first attempts to isolate nitrogenase from R . rubrum were confronted by unusual properties of the enzyme in cell extracts. Crude preparations

198

PAULETTE M. VIGNAIS E r AL.

exhibited variable activities (Bulen et al., 1965a; Burns and Bulen, 1966) characterized by non-linear time courses (Munson and Burris, 1969). This was explained after it was discovered that nitrogenase was inactivated during isolation and required a membrane-bound enzyme for re-activation (Nordlund and Baltscheffsky, 1973; Ludden and Burris, 1976; Nordlund et al., 1977). Analysis of purified nitrogenase in the inactive state revealed that the Fe protein contained an adenine-like molecule, a pentose moiety, and a phosphate residue covalently attached to the molecule (Ludden and Burris, 1978). Specific labelling of the Fe protein after growth in a medium containing [3H]adenine (Nordlund and Ludden, 1983) or [32Plphosphate (Ludden and Burris, 1978) confirmed the presence of these two compounds attached to the molecule. Independently, Carithers et al. (1979) isolated nitrogenase in a fully active form from nitrogen-starved cells of R. rubrum. These results led the authors to propose that two kinds of nitrogenase existed in R. rubrum: nitrogenase A, which is always active, and a regulatory form, called nitrogenase R, which is found in NT or glutamate-grown cells and whose activity was dependent on processing by the membrane-bound activating enzyme. According to their proposal, both forms were active in uiuo and could be converted into a third species, nitrogenase R (inactive), by addition to the cells of a fixed nitrogen source such as ammonia or glutamine. A regulatory mechanism involving interconversion between the three forms of nitrogenase mentioned above was suggested (Yoch and Cantu, 1980). This hypothesis was abandoned when it was demonstrated that nitrogenase activity in R. rubrum was regulated through interconversion of the Fe protein between an active and an inactive form (Gotto and Yoch, 1982a; Preston and Ludden, 1982). These two forms were purified separately, knowing that nitrogen-starved bacteria yielded active Fe protein whereas glutamate- or N2-grownbacteria yielded inactive Fe protein. The active and inactive forms of the Fe protein showed identical properties with respect to the amino acid composition, and immunoreactivity with specific antibodies (Ludden et al., 1982b). However, they gave different patterns after electrophoresis in SDS-polyacrylamide gels. The active Fe protein consisted of two identical subunits of M, 30,000 migrating as a single band, and the inactive form was found to consist of two subunits of M , 30,000 and 31,500 (Ludden et al., 1982b; Gotto and Yoch, 1982a). On activation in uitro with the activating enzyme, release of an adenine-like molecule occurred (Ludden and Burris, 1979). Release of a phosphate-containing group from the Fe protein also occurred concomitantly with the conversion of the 3 1,500 subunit into the 30,000 unmodified subunit, suggesting that the activating enzyme catalysed the removal of the entire modifying group covalently attached to the Fe protein (Gotto

H2

METABOLISM IN PHOTOSYNTHETIC BACTERIA

199

and Yoch, 1982a). It has been shown that the in uiuo “switch-off” of nitrogenase activity coincides with the covalent modification of the Fe protein (Gotto an Yoch, 1982a; Preston and Ludden, 1982). These results clearly indicate that nitrogenase activity in R. rubrum is controlled by a mechanism involving a reversible covalent modification of the Fe protein. Attempts to isolate the inactivating system as well as to identify the donor molecule to the modifying group have hitherto been unsuccessful. However, the modifying group has recently been purified (Ludden et al., 1984) using the property that heat treatment of the inactive Fe protein results in activation and removal of the entire modifying group (Dowling et al., 1982). Analysis by mass spectrometry, high-performance liquid chromatography and by 3’P-NMR spectroscopy showed that the molecule is not AMP, although it contains one molecule each of adenine, phosphate and pentose. Comparison of the 31P-NMRspectra of the isolated modifying group and the modified protein suggested that the adenine-like molecule was attached to the protein via a phosphate linkage (Ludden et al., 1984). The presence of this modifying group on the Fe protein might alter its interaction with the MoFe protein and thus render the nitrogenase complex catalytically inactive. Indeed, spectroscopic data (Ludden et al., 1982a) and catalytic studies with different ratios of the two nitrogenase proteins (Guth and Burris, 1983b) showed that inactive Fe protein, while unaffected in MgATP binding and in its oxidoreduction properties, was impaired in its ability to transfer electrons to the MoFe protein. The occurrence of a similar regulatory mechanism in other photosynthetic bacteria is suggested by the sensitivity of nitrogenase to ammonia “switch-off’ (see Section V.B. 1). The mechanism of this regulation has also been studied in R . capsulata (Michalski et al., 1983;Jouanneau et al., 1983). A previous report by Hallenbeck et al. (1982a) mentioned that the Fe protein, isolated after an in uiuo ammonia shock, contained adenine, phosphate, and a pentose in a 1: 1 : 1 molar proportion with respect to the Fe protein. After purification of the active and inactive Fe protein, the presence of a phosphate-containing group covalently bound to one subunit was attested by incorporation of [32P]phosphateinto the slower moving subunit in SDS-polyacrylamide gels (Jouanneau et al., 1983). As in R. rubrum, attachment and removal of this modifying group correlated with inactivation and activation of nitrogenase, respectively. The inactive Fe protein of R. capsulata could be activated by the activating enzyme from R. rubrum, suggesting that the regulatory systems in the two species are homologous (Jouanneau et al., 1983). Information on the nature of the donor molecule for the modifying group came from the experiments of Michalski et al. (1983) who used toluene-permeabilized cells to permit the

200

PAULETTE M. VIGNAIS ET AL.

entry of radioactive adenine nucleotides. After submitting the cells to an ammonia shock, the authors showed that the Fe protein was specifically labelled. These results strongly suggested that either ATP or ADP is a precursor of the modifying group attached to the Fe protein. 4. The Activating Enzyme Although the existence of an activating factor for R . rubrum nitrogenase was discovered several years ago (Ludden and Burris, 1976; Nordlund et al., 1977), few data are available on the structure and properties of this component. It is a membrane-bound component which can be solubilized from chromatophores by 0.5 M NaCl (Ludden and Burris, 1976; Nordlund et al., 1977). The partially purified activating factor is a protein (now called “activating enzyme”) which is extremely sensitive to oxygen ( t l R in air of about 2 minutes) (Zumft and Nordlund, 1981) and has a molecular weight between 20,000 and 24,000 (Zumft and Nordlund, 1981; Gotto and Yoch, 1982b; Guth and Burris, 1983a). The activating enzyme catalyses the removal of either part (Ludden and Burris, 1979) or all (Gotto and Yoch, 1982a; Jouanneau et al., 1983) of the modifying group from the inactive Fe protein. The reaction requires divalent metal ions and ATP. Free Mn2+is required for maximal activation, in addition to a metal ion chelated by ATP; MgATP or MnATP are equally effective (Gotto and Yoch, 1982b; Guth and Burris, 1983a; Nordlund and Nor& 1984). A site of action for free metal ion has been postulated on the basis of kinetic activation data: the affinity of this site for Mn2+appeared much higher than for Mg2+ by a factor of 1000-fold (Guth and Bunis, 1983a). Only Fez+could substitute for Mn2+, although at much higher concentrations (7.5 m M instead of 0.5 mM for Mn2+)(Nordlund and NorCn, 1984). The activating enzyme in R . rubrum appears to be constitutively synthesized since it was present in variable amounts under all the growth conditions tested, and in Nif- mutants (Triplett et al., 1982). 5 . Oxygen Sensitivity of Nitrogenase Activity

Because of the extreme O2 lability of the nitrogenase complex, aerobic diazotrophs have developed specific mechanisms to ensure protection of nitrogenase against oxygen (Robson and Postgate, 1980). These include diffusion barrier in heterocysts of cyanobacteria, uncoupled respiratory activity, and “conformational” protection in Azotobacter species. The

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

201

latter mechanism, the occurrence of which was suggested several years ago from physiological studies with Azotobacter chroococcum (Dalton and Postgate, 1969), involves the formation of a complex between nitrogenase components and an iron-sulphur protein relatively more oxygen stable (Scherings et al., 1977; Robson, 1979). In facultative aerobes, such as Klebsiella pneumoniae, nitrogenase is repressed under aerobic conditions (Roberts and Brill, 1981). Among photosynthetic bacteria, oxygen-dependent nitrogen fixation in the dark has been demonstrated in facultative phototroph species including members of the Rhodospirillaceae (Siefert and Pfenning, 1980) and Chromatiaceae (Jouanneau et al., 1980b). This finding shows that nitrogenase can function in the presence of oxygen. However, in contrast to obligate aerobes (i.e., Azotobacteriaceae), diazotrophic growth does not occur in air but rather under much lower oxygen tension (below 2 kPa). In uiuo, nitrogenase activity is highly sensitive to oxygen since inhibition occurs at O2 tensions lower than that in air. In the purple sulphur bacterium, Thiocapsa st 581I , O2inhibited nitrogenase completely at 0.06 atm (6 kPa) (Jouanneau et al., 1980b), whereas in R. capsdata nitrogenase activity was detected at an O2 partial pressure of 0.2 atm (20.3 kPa) (Willison et al., 1983b). These differences may reflect variable efficiencies of the respiratory protection of nitrogenase against O2inactivation. In a mutant of R. capsdata lacking both respiratory terminal oxidases, nitrogenase activity was much more sensitive to O2than in the wild type (Meyer et al., 1978a), indicating that the respiratory system plays an essential role in protecting nitrogenase from O2 inactivation and denaturation.

C. REGULATION OF NITROGENASE SYNTHESIS

Knowledge of the control of nitrogenase biosynthesis has progressed considerably since the discovery of the organization of the nifgene responsible for nitrogen fixation in K . pneumoniae (Roberts and Brill, 1981). Briefly, structural genes and other nifgenes are under the control of a regulatory operon, niflA. The product of nifA is required for the activation of other nifgenes. The product of niJZ is a repressor, which is implicated in the mechanism of nitrogenase repression by 02.In addition, nifLA expression is under the general control of genes responsible for the regulation of nitrogen metabolism (ntr). The ntr gene products are thought to control the transcription of several enzymes of the nitrogen metabolism, including nitrogenase through the nifLA operon, depending on the nitrogen nutrition regime of bacteria. High concentrations of NH4+, for example, provoke complete repression of nitrogenase. Oxygen, as well as

202

PAULETTE M. VlGNAlS ET AL.

molybdenum deficiency, are also known as repressing agents of nitrogenase synthesis in several nitrogen-fixing bacteria, suggesting that regulatory mechanisms are similar from one species to another. In photosynthetic bacteria, genetic studies, performed mainly in R . capsulata, indicate that nitrogenase synthesis is also submitted to a complex regulatory system which may resemble that of K. pneumoniae (see Section V1.B). Hallenbeck et al. (1982b) studied the regulation of nitrogenase expression by oxygen and ammonia in R . capsulata, by two-dimensional gel electrophoresis. The presence (or absence) of the nitrogenase components in cell extracts was determined using the purified components as marker of position on the two-dimensional gel. Nitrogenase was absent in cells grown either photosynthetically with an excess of NH4+ or aerobically with glutamate as nitrogen source. Hence, in R . capsulata, NH4+ and O2may act as repressors of nitrogenase synthesis, as is the case in the non-photosynthetic facultative aerobe K. pneumoniae (Roberts and Brill, 1981).

1. Role of Glutamine Synthetase In R . capsulata, NH4+ is assimilated via the glutamine synthetase/glutamate synthase pathway (Johansson and Gest, 1976). The role of glutamine synthetase in the regulation of nitrogenase activity has been discussed above (Section V.B.3). Glutamine synthetase also has been implicated in the regulation of nitrogenase synthesis, for the following reasons: (1) inhibition of GS by the analogue of glutamine, L-methionine-DL-sulphoximine (MSX), resulted in a de-repression of nitrogenase synthesis in the presence of NH4+(Meyer and Vignais, 1979), and (2) mutant strains lacking GS activity are also de-repressed for nitrogenase synthesis (Wall and Gest, 1979). In strains carrying a mutation in the structural gene coding for GS, glutamine, but not NH4+,repressed nitrogenase synthesis (Scolnik et al., 1983). This implies that glutamine, rather than GS itself, is directly involved in nitrogenase repression, as indicated in Fig. 5 .

2 . Light Dependence of Nitrogenase Synthesis The synthesis of nitrogenase was originally believed to be light dependent (Hillmer and Gest, 1977a,b; Meyer et al., 1978a,b,c; Yoch, 1978). How-

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

/

rnRNA-

203

nit genes

\-

\

\

FIG. 5. Schematic representation of nitrogenase function and regulation in Rhodospirillaceae species. Abbreviations: PS, bacterial photosystem; AE, “activating enzyme”; Hlase, hydrogenase; GS, glutamine synthetase; Fd, ferredoxin; +, positive regulation; -, negative regulation; N2ase, nitrogenase. The effectors in negative regulation, and the mechanism of ferredoxin reduction, are still unknown.

ever, since several facultative phototrophs belonging to the Rhodospirillaceae are also capable of aerobic growth in the dark, the question arose as to whether nitrogenase could be expressed independently of photosynthetic activity. Siefert and Pfennig (1980) demonstrated that Rhodopseudomonas species can grow in the dark at the expense of N2 under low oxygen tensions. Furthermore, a member of the Chromatiaceae, endowed with a weak respiratory activity, was also found to sustain micro-aerobic nitrogen fixation in the dark (Jouanneau et al., 1980b). Finally, dark anaerobic dinitrogen fixation was reported to occur in a photosynthetic bacterium growing by fermentation (Madigan et al., 1979). These results demonstrate that nitrogenase synthesis can occur in the dark, at noninhibitory oxygen tensions, and therefore does not require light. Nevertheless, light was reported to strongly stimulate nitrogenase activity in whole cells (Hillmer and Gest, 1977a,b; Meyer et al., 1978a,b), suggesting that it might induce nitrogenase synthesis. The effect of light intensity on nitrogenase synthesis has been studied in continuous culture in order to maintain a constant cell density during growth (Jouanneau et at., 1982). Under these conditions, light absorption by the culture is constant and therefore the actual light intensity reaching the bacteria depends only on the incident light. At a constant light intensity, variations of the dilution rate in a nitrogen-

204

PAULETTE

M. VlGNAlS ET AL.

limited culture had little effect on the steady-state nitrogenase activity (Jouanneau, 1982). This indicates that nitrogenase synthesis is somewhat independent of growth rate. At a constant dilution rate, increase of the incident light resulted in a concomitant increase of the steady-state nitrogenase activity (Jouanneau et al., 1982). Immunoassay determinations of the enzyme components demonstrated that the increase of nitrogenase activity correlated with an increase in nitrogenase content of the cells (Y. Jouanneau, C. M. Meyer, and P. M. Vignais, unpublished observations). These results suggest that nitrogenase synthesis is related to the bacterial photophosphorylation activity. As shown by Reid1 et al. (1983), the photophosphorylation capacity of the R. capsulata photosystem is greater in cells grown under high light intensity than in cells grown under low light intensities. These authors calculated that the maximal rate of phosphorylation per reaction centre is 7.7-fold greater in cells grown under high light intensities. However, since cells synthesize fewer reaction centres at high light intensities, the rate of photophosphorylation with respect to cell protein is only 1.5-fold higher in cells exposed to a high light intensity. Further work is required to determine whether such difference in the rate of phosphorylation may account for the stimulation of nitrogenase synthesis.

VI. Genetics of Hydrogen Production and Utilization A. GENERAL BACKGROUND

Genetic studies of hydrogen metabolism in the photosynthetic bacteria have been undertaken only recently, and have been restricted to certain members of the Rhodospirillaceae, in particular Rhodopseudomonas capsulata. Mutants affected in the enzyme systems involved (nitrogenase and hydrogenase) are readily obtainable (see below) and some of these mutants have provided information on the regulation of H2 metabolism in the photosynthetic bacteria. Several methods now exist for the genetic analysis of these mutants, including a transduction-like system (“gene transfer agent”) and conjugation (Pemberton et al., 1983). The first gene transfer system to be described in a photosynthetic bacterium was the gene transfer agent (GTA) of R. capsulata (Marrs, 1974). This GTA is a bacteriophage-like particle containing about 3 x lo6 Da of double-stranded DNA that appears to be entirely derived from the bacterial chromosome (Solioz et al., 1975; Yen et al., 1979). The amount of

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

205

DNA carried by GTA is equivalent to five or six genes; it therefore can be used for fine-structure mapping of closely linked genes, such as the photopigment synthesis genes of R . capsulata (Yen and Marrs, 1976; Youvan et al., 1982). The GTA appears to be specific for R . capsulata, some strains of which act as both donors and recipients, whereas others act as donors or recipients only (Wall et al., 1975b). More recently, transfer of chromosomal genes by conjugation, using broad-host-range drug-resistance plasmids, has been demonstrated in R . capsulata and Rhodopseudomonas sphaeroides (Marrs, 1981; Pemberton and Bowen, 1981; Willison et al., 1983b) and the transformation of R . sphaeroides by isolated plasmid or bacteriophage DNA has also been described (Pemberton et al., 1983; Fornari and Kaplan, 1982). The techniques of genetic engineering and transposon mutagenesis have recently been applied to the photosynthetic bacteria. These techniques have permitted the physical mapping of the structural genes for the nitrogenase complex in R . capsulata (Avtges et al., 1983) and have revealed the presence of multiple “pseudocopies” of these genes in R . capsulata (Scolnik and Haselkorn, 1984) and R. sphaeroides (Fornari and Kaplan, 1983).

B . NITROGENASE GENETICS

The Nit- mutants (Le., mutants unable to grow with N2 as sole nitrogen source) have been isolated from R . capsulata (Wall et al., 1975a; Willison and Vignais, 1982; Avtges et al., 1983), Rhodopseudomonas acidophila (Siefert and Pfennig, 1978), Rhodospirillum rubrum (Voelskow and Schon, 1980; Triplett et al., 1982; Falk and Johansson, 1983), and R . sphaeroides (Shestakov et al., 1983). These mutants can be divided into two categories: mutants that are affected specifically in nitrogen fixation (nifgenotype), and pleiotropic regulatory mutants that are deficient in the utilization of a wide range of nitrogen sources and thus resemble the ntr mutants of Klebsiella pneumoniae (see Magasanik, 1982; Merrick, 1983), the nit mutants of Salmonella typhimurium (Broach et al., 1976), and the nac mutants of Klebsiella aerogenes (Bender et al., 1983). 1. Nitrogenase Structural Genes The Nif- mutants of R . capsulata that lack nitrogenase activity (as measured by acetylene reduction) are unable to evolve H2 in the light, confirming that H2 evolution is indeed due to nitrogenase (Wall et al., 1975a).

PAULETTE M. VlGNAlS ET AL.

206

However, in very few Nif- mutants has the biochemical lesion affecting nitrogen fixation been identified. In the non-photosynthetic diazotroph K . pneumoniae, about 17 nifgenes are known to be involved in the biosynthesis, regulation, and activity of nitrogenase (see Roberts and Brill, 1981). In R . rubrum, a mutant specifically lacking the Fe protein of nitrogenase was identified by using antibodies against the purified Fe protein (Falk and Johansson, 1983). This strain carries a mutation in the structural gene for the Fe protein ( n i m rather than in a gene for the processing of the Fe protein, since the mutation was complemented by pRDl , a plasmid carrying the nifgene cluster of K . pneurnoniae, but not by a NifF- mutant of this plasmid (Falk and Johansson, 1983). Mutants of R . capsulata carrying mutations in the structural genes for the nitrogenase complex ( n i m j D , K )were identified by complementation with the cloned nitrogenase genes (Avtges et al., 1983). An 11.8-kilobasepair (kbp) HindIII fragment of R . capsulata DNA, shown to hybridize with pSA30 containing the nijH, D , K genes of K . pneumoniae, was cloned into the plasmid vector pRK290 and introduced into different strains of R . capsulata by conjugation; the wild-type phenotypye was restored in 4 out of the 11 Nif- mutants tested. Insertion of the transposon Tn5 at different positions in the 11-kbp fragment was used to obtain a physical map of the ni$!l,D,K genes (Fig. 6). From the polar effect of Tn5

5.05

4.0

4

..-

3.0 2.6 315/275125~

.-.D

H

J56

15

H3

? K

--

p#l 5 7

J602

FIG. 6. Physical map of the nitrogenase structural genes of R. capsulara. The upper line represents a 5.05-kbp fragment of R. capsulata DNA containing restriction sites for EcoRI (Rl) and HindIII (H3). The arrows indicate the positions of Tn5 insertions, with their distance in kbp from the HindIII site. All insertions to the right of the 5.05-kbp insertion resulted in a Nif- phenotype. The line below indicates the regions of homology (determined by hybridization) between R. capsulata DNA and the K. pneumonia genes coding for the Fe protein ( n i m , the Q subunit of the MoFe protein ( n i p ) , and the P-subunit of the MoFe protein (nifK). The direction of transcription of these genes is from left to right, as determined by the polar effect of Tn5 insertions in nifH on the expression of n i p and n i f l . The third and bottom lines indicate the limits of, respectively, the extreme left- and right-hand boundaries of the nifgenes (the length of these genes was calculated on the basis of the known molecular weight of the nitrogenase components) and the positions of the mutations in four Nif- mutants. From Avtges et a/. (1983).

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

207

insertions, it was deduced that the transcriptional organization of these genes is the same as in K. pneumoniae, i.e., they form an operon that is transcribed in the direction n i m through niJX (Avtges et af., 1983). The cloned nim,D,K genes were labelled with 32Pand were found to hybridize with RNA from wild-type cells incubated in the absence of NH4+, but not with RNA from cells incubated in the presence of NH4+, showing that NH4+ represses the transcription of these genes (Scolnik et af., 1983). Cloning and transposon mutagenesis of nifgenes from R. capsulata have also been reported by Piihler et al. (1984). The niJH,D,K genes of K. pneumoniae were found to hybridize with multiple restriction fragments of chromosomal DNA from R. capsulata (Scolnik and Haselkorn, 1984) and R. sphaeroides (Fornari and Kaplan, 1983). In R. capsulata, it was suggested that one active copy of niJH,D,K genes is present, together with a number of “pseudocopies,” which are inactive under normal conditions (Scolnik and Haselkorn, 1984). Activation of the pseudocopies was obtained by selecting for Nif+ revertants of Tn5-induced insertion and deletion mutations in the active niJH,D, K genes. The nitrogenase activity developed by these “pseudorevertants” was about 10% of the activity in the wild-type strain. The genetic organization of the nifgenes in the photosynthetic bacteria, other than the nitrogenase structural genes of R. capsdata (see above), has yet to be established. However, nifmutants of R. capsulata fall into several different linkage groups in GTA-mediated crosses (J. D. Wall, personal communication, and quoted in Avtges et al., 1983), and conjugative studies using the plasmid pTH10 have also indicated that, in contrast to the nifgenes of K. pneumoniae, the nifgenes of R. capsulata are not clustered in the same region of the chromosome (J. Willison, unpublished). 2. Nitrogenase Regulatory Genes Various mutants affected in the regulation of nitrogen metabolism have been isolated, from R. capsulata (Wall et al., 1977; Wall and Gest, 1979; Willison and Vignais, 1982), R. sphaeroides (Shestakov et at., 1983), and R. rubrum (Weare, 1978). A pleiotropic Nif- mutant has been described, which is unable to use nitrogen sources other than NH4+ or urea for growth (Wall et al., 1977). This mutant contained normal concentrations of the enzymes of NH4+ assimilation (glutamine synthetase and glutamate synthase) and also developed wild-type concentrations of nitrogenase when grown with a limiting concentration of NH4+ (2.0 mM) as nitrogen source. It was suggested that the mutant contains an unspecified defect which results in an inability to utilize low intracellular concentrations of

208

PAULETTE M. VIGNAIS E r AL.

NH4+ (Wall et al., 1977). A different mutant was isolated which is unable to grow on either N2 or glutamate and is also deficient in nitrogenase activity when grown with limiting concentrations of NH4+(Willison and Vignais, 1982). The phenotype of these pleiotropic Nif- mutants indicates that regulatory genes similar to those involved in the control of nitrogen metabolism in the Enterobacteriacae are also operative in the photosynthetic bacteria. A third type of mutant has been described which shows linear nitrogenase kinetics when glutamate-grown cells are made permeable with toluene, in contrast to glutamate-grown cells of the wild type, which show non-linear kinetics. This mutant may be affected in some aspect of the inactivation of the nitrogenase Fe protein (Willison and Vignais, 1982). Further study of these mutants should provide insights into the regulation by NH4+ of both synthesis and the activity of nitrogenase. Glutamine auxotrophs of R. capsulata, lacking glutamine synthetase activity, are fully de-repressed for nitrogenase synthesis and activity, producing H2 (and reducing acetylene) in the presence of high concentrations of NH4+(Wall and Gest, 1979). The de-repressed phenotype of these glutamine auxotrophs is due to a mutation in the structural gene for glutamine synthetase (glnA), as was shown by complementation of these mutants by the cloned glnA gene of R. capsulata (Scolnik et al., 1983). This gene was cloned by screening a cosmid library of genomic DNA for strains complementing a glnA-deletion mutant of E. cofi. The Hind111 fragment carrying the glnA gene was then sub-cloned into the plasmid vector, pRK292, and transferred into R. capsulata mutant strains by conjugation. In all the glutamine auxotrophs tested, wild-type regulation of nitrogenase was restored by introduction of the wild-type glnA gene. In the same study, it was shown that nitrogenase in the glutamine auxotrophs was repressed by 20 mM glutamine, although not by a similar concentration of NH4+,indicating that the product of glutamine synthetase (i.e., glutamine) is involved in regulation (Scolnik et al., 1983; see also above). A glutamate auxotroph of R. rubrurn was also found to be derepressed for nitrogenase synthesis and H2 production in the presence of NH4+(Weare, 1978). Since this mutant is deficient in glutamate synthase activity, it might be expected to accumulate glutamine in the presence of NH4+. 3. Endogenous Plasmids Many species and strains of purple non-sulphur bacteria contain endogenous plasmids of molecular weight between 28 X lo6 and 94 X lo6 (cf.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

209

Kuhl et al., 1983). However, it is not known whether genetic information involved in nitrogen fixation or hydrogen metabolism is encoded in these plasmids (see also below). In R . capsulata strain BlO, some nifmarkers show genetic linkage to genes conferring resistance to rifampicin or streptomycin (Willison et al., 1983b), and also to genes coding for the biosynthesis of histidine, uracil, adenine, and carotenoids (J. Willison, unpublished), indicating that these nif genes are located on the chromosome. In R . rubrum, genes required for photosynthesis may be carried on an endogenous plasmid since strains “cured” of the plasmid were photosynthetically incompetent (Kuhl et al., 1983). However, in R. cupsulata (Marrs, 1981; J. Willison, unpublished) and in R . sphaeroides (Pemberton and Bowen, 1981), the photosynthesis genes appear to be located on the chromosome.

C. HYDROGENASE GENETICS

Little study has so far been made of the genetics of hydrogenase in photosynthetic bacteria. Attention has been focused instead on the aerobic hydrogen bacteria and Rhizobium sp. (Tait, 1982). In the latter case, the search has been stimulated by economical considerations, since host plants infected by Rhizobium sp. having an active hydrogenase show an improved yield compared with plants infected by the majority of wildtype strains that are devoid of hydrogenase activity (Albrecht et al., 1979). Recently, a mutant strain of Rhizobium japonicum has been obtained that is insensitive to repression by carbon substrates: in bacteroids, where carbon substrate concentration is high, the hydrogenase activity of this mutant was six to eight times higher than that of the wild type (Merberg and Maier, 1983). A method has been devised allowing the direct identification of hydrogenase-negative (Hup-) colonies of bacteria on Petri dishes (Postgate et al., 1982; Haugland et al., 1983). The involvement of plasmids in hydrogenase genetics has been suggested in the case of Alculigenes eutrophus (Lim et al., 1980; Andersen et al. 1981; Behki et al., 1983; Friedrich and Friedrich, 1983) and for other hydrogen bacteria (Pootjes, 1977). These plasmids, which have a minimum molecular weight of 200 x lo6, are either non self-transmissible but mobilizable, such as pAEl (Andersen et al., 1981) or pRMBl (Behki et al., 1983), or self-transmissible such as pHGl (Friedrich et al., 1981~). In photosynthetic bacteria, hydrogenase-deficient mutants have been isolated from R . capsulata strain BlOO (Takakuwa et al., 1983) or strain B10 (Willison et al., 1984) after treatment with either ethylmethyl sulThe specific hydrogenase acphonate or N-methyl-N’-nitrosoguanidine.

210

PAULETTE M. VlGNAlS €7 AL.

tivity of these mutants was less than 25% of that of the parent strain, but only one mutant was completely devoid of hydrogenase activity (Takakuwa et al., 1983). Rocket immunoelectrophoresis with specific hydrogenase antibodies confirmed that hydrogenase was synthesized in decreased amounts in these mutants (Colbeau and Vignais, 1983). The cloning of the hydrogenase (hup) gene has been achieved in R. japonicum (Cantrell et al., 1983) by using conjugative plasmids with a broad host range. Although the organization of the hup gene(s) is still unknown, the results obtained suggest that the Hup system involves several genes. This is in agreement with the experiments of Maier and Mutafschiev (1982) who obtained reconstitution of hydrogenase activity after prolonged incubation of the soluble cell fractions of two hydrogenasenegative mutants of R. japonicum.

VII. Use of Photosynthetic Bacteria as Biological Solar Energy Converters The photosynthetic bacteria represent a tool of great potential in various fields of biotechnology. Their rapid growth rate and their metabolic versatility enable them to survive and proliferate in a wide variety of environments. They can synthesize large amounts of biomass from organic materials; for example, in a tropical stratified lagoon the yield of phototrophic bacteria was estimated at 1.5 g C m-2 day-', corresponding to about 40% of the total photosynthetic production (Caumette et al., 1983). Consequently, the cultivation of photosynthetic bacteria for single-cell protein production for animal feed has been proposed by several authors (Shipman et al., 1975; Kobayashi and Kurata, 1978). By oxidizing low-molecular-weight carbon and sulphur compounds (including rnercaptans) to gaseous or innocuous products, the photosynthetic bacteria may contribute significantly to the purification of polluted water supplies. The treatment of industrial waste waters by processes involving photosynthetic bacteria has been advocated for several years by Kobayashi and associates (Kobayashi 1975, 1976; Kobayashi and Tchan, 1973; Kobayashi et al., 1978). More recently, the capacity of photosynthetic micro-organisms to produce molecular hydrogen (a possible alternative fuel to the fossil hydrocarbons) has attracted attention, and has stimulated research into the possibility of using these organisms as solar energy converters (for reviews see Lien and San Pietro, 1975; Mitsui, I978; Hallenbeck and Benemann, 1979; Weaver et al., 1980; Benemann at al., 1980).

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

21 1

For the purpose of solar energy conversion into hydrogen, the utilization of cyanobacteria was first considered, since these organisms are able to use water as a reductant. Photosynthetic bacteria, which cannot split water but require organic compounds as electron donors, were initially considered economically less viable. For example, Bennett and Weetall (1976) evaluated the influence of substrate cost on H2 produced from glucose by immobilized Rhodospirillum rubrum cells and concluded that H2 cannot be produced economically from a substrate having a price of more than $0.10 (U.S.$) per pound. Nevertheless, the photosynthetic bacteria present several advantages over the cyanobacteria as H2-producingorganisms. Firstly, they generally show much higher rates of H2 production. Secondly, since they conduct anoxygenic photosynthesis, the H2 produced is free of contaminating oxygen. Thirdly, they are much more amenable to genetic manipulation. The possibility of producing H2 as a by-product of processes such as biomass production or water purification also should be considered. Current approaches to the use of photosynthetic bacteria for Hz production are presented below. As discussed in the preceding sections, two different enzymes can catalyse proton reduction. The first one is hydrogenase, whose biotechnological potential has been recently evaluated by Klibanov (1983). Some authors have investigated the possibility of using hydrogenase either in whole cells (e.g., of Thiocapsa, Bagyinka et al., 1982; or of R. rubrum, Chuaqui et al., 1980) or in bacterial membranes (reaction centres of Chlorobium limicola, Bernstein and Olson, 1981). Cells containing a hydrogenase linked to formic dehydrogenase (i.e., containing formic hydrogenlyase) will also be of interest, if H2 is stored in the form of formate solution as proposed by Klibanov et al. (1982). The second enzyme is nitrogenase. The discussion below will be restricted to the use of nitrogenase in whole cells and comprises the following aspects: the screening and selection of suitable strains, the search for economical substrates, and the use of immobilized cells.

A. STRAIN SCREENING A N D SELECTION

Natural isolates show great differences in their ability to photoproduce Hz. This capacity is linked not only to the nitrogenase content and activity of the cells but also to the ability of these cells to degrade organic substrates. Additional special characteristics (e.g., capacity for growth in salt water or tolerance of high temperatures) may be of value for developing a competitive, viable process.

21 2

PAULETTE M. VlGNAlS €7 AL.

1, Marine Strains

Mitsui and colleagues at the University of Miami's Marine School have undertaken a survey of tropical marine environments for species of high hydrogen-producing capacity (Mitsui, 1981 ; Mitsui et al., 1982). The advantages of marine species is the preponderance of salt-water over freshwater environments, and the abundance in salt water of many nutrients (e.g., magnesium, sulphate, potassium) essential for growth. Among the hundreds of both purple and green photosynthetic bacteria isolated, they have selected a Chromatium sp. Miami PBS 1071, whose properties make it a good candidate for H2photoproduction. It is one of the fastest growing strains (doubling time 1.75 hours), and can sustain a high rate of H2 photoproduction capability (134 ml H2 hr-' g biomass-') (Mitsui, 1976; Ohta and Mitsui, 1981). It can use various organic and sulfur compounds, such as malate, acetate, succinate, thiosulphate, and sulphide, as electron donor (Mitsui et al., 1980). This strain has the faculty of taking up sulphide and depositing elemental sulphur inside its cells and could therefore be used for environmental treatment of sulphide-rich salt waters (Ohta et al., 1981). 2. Thermophilic or Thermostable Strains

In most strains of photosynthetic bacteria, the optimal temperature for growth and H2 production is around 30°C. This temperature may be exceeded in some climates, and it would therefore be desirable to use thermophilic, or at least thermostable, strains as solar energy bioconverters. Watanabe et al. (1981) have isolated Rhodopseudomonas sphaeroides strains from Thailand which are as active in H2 production at 40°C as at 30°C. 3. Strains with High Nitrogenase Content and Activity

As pointed out by Benemann et al. (1980), nitrogenase has a high energy requirement and a low turnover number compared to hydrogenase, suggesting that the latter enzyme might be more suitable for H2 production. However, the ATP required for the functioning of nitrogenase is provided by light, which is a primary energy source. Furthermore, it should be possible to increase the rate of H2 production by increasing the nitrogenase content of bacterial cells. This could be achieved by several means.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

213

1. Nitrogenase being an inducible enzyme, it is possible to adjust the growth conditions to have maximal expression of the enzyme. For this purpose, Ohta and Mitsui (1981) grew Chromatium sp. Miami PBS 1071 in batch culture on molecular nitrogen and observed hydrogen production rates as high as 6 pmol H2 hr-l mg protein -l in cells taken from the cultures in the middle of the logarithmic growth phase. With continuous cultures of Rhodopseudomonas capsulata growing with N2, no H2 production is observed, presumably due to inhibition of proton reduction by N2 and to recycling of the H2 evolved by nitrogenase (Jouanneau, 1982; see above). However, continuous cultures of R. rubrum exhibited increased photoproduction of H2 in the presence of molecular nitrogen, provided that N2 was in limited amounts. As pointed out by the authors (Zurrer and Bachofen, 1982), this makes possible the production of hydrogen in growth media lacking combined nitrogen. Another physiological means to maintain the catalyst (nitrogenase) fully active is to allow its regeneration during the course of the experiment. For example, Miyake et al. (1982) were able to maintain the nitrogenase activity of R. rubrum cells for more than 2 weeks in an experiment with a stimulated day-and-night rhythm of 12 hours light and 12 hours dark, by supplying the system with small amounts of ammonium sulphate or nitrogen gas at the beginning of each dark period. 2. An increase in light intensity stimulates nitrogenase synthesis. At saturating light intensities, the nitrogenase content of R . capsulata cells grown in continuous culture represented up to 25% of the total soluble proteins and the cultures maintained a steady rate of H2 production (45 ml hr-I litre-I) for at least 10 days (Y. Jouanneau, B. Wong, and P. M. Vignais, unpublished observations). In outdoor cultivation vessels, light may be a limiting factor for nitrogenase activity. This was shown with outdoor batch cultures of R. sphaeroides. Two cultures (33 litres) containing 53 mM m-lactate and 5 mM glutamate were placed on the ground, one vertically and the second inclined at 30" to receive more light. After 25 days, the total HZproduction was 155 litres in the first culture and 177 litres in the second; these productions corresponded to a conversion efficiency of lactate to H2 of 69 and 78%, respectively (Kim et al., 1982). 3 . An improved understanding of the regulation of nitrogenase activity and synthesis will make possible the construction of de-repressed cells, Weare (1978) has already obtained a mutant of R. rubrum which is not repressed by NH4+ for the photoproduction of ammonia and hydrogen. Mutants of R . capsuIata which are glutamine auxotrophs also can synthesize nitrogenase and produce hydrogen in the presence of ammonia (Wall and Gest, 1979).

214

PAULETTE M. VlGNAlS ET AL.

The glutamine analogue L-methionine-DL-sulphoximine(MSX), at concentrations lower than 0.1 mM, can relieve the photoproduction of hydrogen from ammonia inhibition in R . rubrum (Zurrer and Bachofen, 1980). 4. The techniques of genetic engineering may allow the construction of cells with an amplified copy number of nifgenes.

4. Hydrogenase-Deficient Strains (Hup-)

If the H2-uptake hydrogenase is active in the recycling of hydrogen, then strains lacking this hydrogenase should show an enhanced production of H2. This was the case for a hydrogenase-negative mutant of R . capsulata (Takakuwa et al., 1983) that produced 4.6-6.2 mol of H2 for each mole of sugar compared with 1.2-4.3 mol for the wild-type strain (Odom and Wall, 1983). On the other hand, Willison et al. (1984), using Hup- mumutagenesis, concluded tants obtained by N-methyl-N’-nitrosoguanidine that the increased H2 evolution by the Hup- mutant oxidizing DL-malate resulted from an enhanced ability of the mutant to consume D-malate. These authors had observed that some non-autotrophic mutants with apparently normal levels of hydrogenase activity also exhibited increased H2 production on DL-malate and were unable to demonstrate, with the wild-type strain in presence of an excess of organic substrate, a recycling of hydrogen as observed by Takakuwa et al. (1983). If hydrogenase is plasmid coded in photosynthetic bacteria, as it is in Pseudomonas facilis (Pootjes, 1977) or in Alcaligenes eutrophus (Friedrich and Friedrich, 1983), then the curing of plasmids may yield more appropraite Hupstrain for this type of test. Other ways to minimize the role of the uptake hydrogenase would be to place the cells under conditions where the enzyme is non-functional, namely in the absence of O2 and in the presence of an excess of organic compounds (so that C 0 2 photoreduction does not occur) or at pH values below 7.5 where the uptake activity of the enzyme is negligible.

5 . Carbon Metabolic Pathways for Complete Substrate Degradation Curiously, although the main metabolic pathways (Krebs cycle, oxidative pentose phosphate cycle, Entner-Doudoroff pathway) are known to occur in photosynthetic bacteria, the specific pathways leading to H2 formation have not been identified. If sugars or C3 and C4 organic acids are degraded only partially, there is a lower efficiency of substrate conversion

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

215

into H2. It would be of interest therefore to identify the metabolic pathways and processes of regulation capable of interfering with, and making inefficient, the production of H2. Mutants deficients in pathways leading to electron sinks other than protons could be looked for, and, indeed, may already have been isolated (Willison et al., 1984). Macler et at. (1979) have obtained a R. sphaeroides mutant which can quantitatively convert glucose into H2 and COz. In this case, 12 mol of H2 are obtained for each mole of glucose so that the combustible energy of glucose (AGI, = -686 kcal mol-I) can be recovered almost completely in the combustible energy of the evolved hydrogen (AGI, = -680 kcal mol-I). In contrast, by conversion of glucose into methane only 85% (586 kcal mol-I) of the available energy would be recovered (Mitsui et al., 1980; Zurrer, 1982).

B . ECONOMICAL SUBSTRATES

1. Production of Hydrogen

As mentioned in the preceding sections of this review, photosynthetic bacteria have a diverse metabolism that enables them to produce hydrogen from a great variety of carbon compounds. However, from an economic point of view, only very cheap substrates can be employed. Adequate electron donors may be found in organic wastes from paper mills, fruit or milk processing factories, and sugar refineries, among others. The production of H2 from cow dung has also been described (Vrati and Verma, 1983). Since lactate is a good electron donor for hydrogen production, R. rubrum and R . capsulata have been tested for their ability to produce hydrogen gas continuously from lactic acid-containing wastes. Rhodospiriffumrubrum strain S-1, in batch culture, produced hydrogen gas continuously at an average rate of 6 ml hr-' (g dry wt. cells)-I over a period of up to 80 days when supplied periodically with whey or yoghurt wastes (Zurrer and Bachofen, 1979,1981). With R. capsulata strain B10, growing in a continuous culture on whey, steady-state rates of 25 ml hr-' litre-' were attained with a conversion efficiency of lactate to H2 of 70-80% (Jouanneau et al., 1982). Stevens et al. (1983) tested nine strains of R. capsulata for their capacity to produce H2 anaerobically in the light from either DL-lactate, acetate, or butyrate. Maximum rates of 1630 ml of H2 day - I litre reactor-' were achieved with lactate, which was the best substrate of the three. The strains differed in their efficiency of conversion from the carbon substrate into H2 and C02,but produced H2 at about

216

PAULETTE M. VlGNAlS ET AL.

the same rate. These authors concluded that the choice of a strain for H2 production from waste waters can be made only on the basis of the available hydrogen donors. Mitsui et al. (1983), in the course of their search for new species, isolated Rhodopseudomonas strains able to grow on carbohydrate polymers such as cellulose, soluble starch, or peptin. Odom and Wall (1983) have described a different strategy for cellulose degradation: they grew a mixed culture of Cellulomonas, which degrades cellulose to cellobiose by fermentation, and R . capsulata which can photo-evolve H2 from cellobiose.

2. Water Depollution By oxidizing the organic substrates contained in waste waters, photosynthetic bacteria are efficient agents of water treatment. A model of a watertreatment plant utilizing photosynthetic bacteria has been proposed by Kobayashi (1976). The proposed process, which does not require dilution, can be used in areas where water is scarce, and leads to the formation of a bacterial biomass rich in protein and vitamins (Kobayashi and Kurata, 1978). This biomass is utilizable as animal feed in pisciculture and poultry industry, and as a fertilizer in horticulture.

C. CELL STABILIZATION BY IMMOBILIZATION

Immobilization of whole microbial cells by entrapment in polymeric matrices is a means to increase the stability of biological material, with a view to its use in biotechnological applications. Originally, immobilized cells were non-viable, and were used as biocatalysts for a single enzyme activity. Milder techniques have now been developed to prepare entrapped cells which are fully viable and biosynthetically active so that they can catalyse multi-step reactions (for reviews see, e.g., Chibata and Tosa, 1977, 1980; Fukui and Tanaka, 1982; Chibata and Wingard, 1983). Furthermore, the use of immobilized cells is compatible with a continuous process and eliminates the problem of product separation and harvesting of biomass. The systems using immobilized micro-organisms for H2 production have been recently reviewed and discussed by Hallenbeck (1983a). Rhodopseudomonas sp. Miami 2271 cells, immobilized with agar on an agarose-coated polyester film, produced H2 at rates as high as 445 ml hr-* g dry cells-' and were markedly protected from inhibition by oxygen and

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

217

nitrogen (Matsunaga and Mitsui, 1982).Immobilized cells exhibited wider salt tolerance than aqueous cell suspensions and photoproduced H2 at the same rate over a 10-day period. Rhodopseudomonas palustris strain 42 OL immobilized in agar gels produced H2from malate or from waste waters at a constant rate of 30-40 ml hr-i (g cell dry wt.)-' over a 30-day period. Yields of 0.78 litre of HZ from 1 litre of sugar refinery wastes (initial COD = 1200 ppm) and of 2.2 litres of H2 from 1 litre of straw paper-mill effluent (initial COD = 5600 ppm) were achieved (Vincenzini et al., 1981). Immobilized Rhodospirillum molischianum cells were even more effective, and 43% of the initial COD was converted into H2 (Vincenzini et al., 1982b). With an agar layer of 3.5 mm thickness, the rate of H2 evolution from DL-malate was limited by the diffusion of the substrate into the agar when the cell concentration was 1.7 mg cell dry ~ m - but ~ , no limitation occurred with 0.425 mg cell et al., 1982a). dry wt ~ m (Vincenzini - ~ The highest rates of hydrogen production by immobilized bacteria have been achieved with R . cupsuluta cells entrapped in a barium alginate gel, using malate as substrate (72 ml hr-' g protein-') ( Paul and Vignais, 1981). However, alginate gels did not prevent inactivation of nitrogenase with time, and the rate of H2 production and storage stability were comparable to those of free cells. Since rates as high as 260 ml hr-' g dry wt. biomass-' were obtained with nitrogen-limited continuous cultures of R. cupsuluta (Jouanneau et al., 1982), three- to five-fold higher rates are theoretically obtainable.

D. ADVANTAGES OF USING PHOTOSYNTHETIC BACTERIA AS HYDROGEN

PRODUCERS

As already pointed out by several authors (e.g., Weaver et al., 1980; Vincenzini, 1982b), the anaerobic photoproduction of HZby photosynthetic bacteria offers several advantages over other systems from the technical and economic points of view. 1. Photosynthetic bacteria produce H2 at much higher rates than other types of micro-organisms and the fundamental mechanisms by which H2 is produced are becoming well understood. 2. Molecular hydrogen is produced by an irreversible reaction catalysed by nitrogenase and proceeds even under an atmosphere of 100% hydrogen gas. 3. The reaction is anoxygenic; there is no residual OZ-evolvingactivity which would cause problems of Oz inactivation of the catalyst or of the

21 8

PAULETTE M. VlGNAlS €7 AL.

separation of O2from H2. The only gaseous contaminant produced is C02 which can be trapped easily. 4. Photosynthetic bacteria can trap light energy over a wide spectral range and can withstand high light intensities. 5. The capacity of photosynthetic bacteria to use, as reductants, a wide range of carbon compounds as metabolizable substrates (sugars, shortchain fatty acids, organic acids) make them able, in principle, to consume organic wastes. The use of these wastes may render the process economically feasible, with H2 as a useful by-product. One drawback for a nitrogenase-mediated production of H2 is that the enzyme is inhibited by ammonium ions. Suitable wastes to be used are those containing low concentrations of combined nitrogen such as sugar refinery wastes, cellulose, and lignin. Another possibility is to use mutant strains with constitutively de-repressed nitrogenase. 6. The great metabolic versatility of photosynthetic bacteria enable them to remain functional under many different environmental conditions (aerobic, anaerobic, in light, in darkness, and in salt waters). Some strains are able to degrade completely to C 0 2 and H2 organic substrates such as glucose (Macler et al., 1979). 7. Genetic techniques are rapidly being extended to photosynthetic bacteria, which now can be transformed by exogenous plasmids (Fornari and Kaplan, 1982). By genetic manipulation it should not be too difficult to introduce plasmid-borne genes coding for useful functions, such as xenobiotics degradation, and therefore develop multi-application processes based on the use of photosynthetic bacteria.

VIII. Summary and Prospects Photosynthetic bacteria possess a diverse and evolutionarily ancient metabolism, which is reflected in the different ways in which they can metabolize H2. Three enzymes have been implicated in H2 metabolism in these organisms: (1) nitrogenase, which catalyses unidirectional, ATP-dependent H2evolution, and can function either in the light, or in the dark under anaerobic or micro-aerobic conditions; (2) uptake hydrogenase, which is membrane-bound and, although capable of both H2 evolution and uptake, functions physiologically in the direction @fH2 oxidation; and (3) “classical” or reversible hydrogenase, which may be either soluble or membrane bound and functions mainly during dark anaerobic fermentation. Previous reviews have considered mainly the physiological and ecological aspects of H2 metabolism; recent developments in these fields have

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

219

been covered in the present review. However, much information has now been obtained concerning the molecular aspects of H2 metabolism, due to the purification and characterization of some of the enzymes involved. This concerns only the nitrogenase and uptake hydrogenase, since the reversible hydrogenases have yet to be isolated and characterized in detail. Nitrogenase has been purified from two photosynthetic bacteria and shown to be similar in molecular properties to nitrogenases from nonphotosynthetic bacteria. However, it differs from that of most bacteria in that it is often isolated in an inactive form, inactivation being due to covalent modification of the Fe protein. The modifying group contains adenine, a pentose moiety, and phosphate, but is not AMP. Current research involves identification of the modifying group, the mechanism of its attachment and removal, and the way in which modification of the Fe protein inhibits the catalytic activity of nitrogenase. Uptake hydrogenase has been purified from a number of photosynthetic bacteria. The solubilized enzyme is generally O2 stable and resistant to heat treatment and denaturing agents. Hydrogenases from different bacteria show similar kinetic properties, but vary widely in their relative activities of H2 evolution and H2 uptake. Some enzymes have been shown to contain nickel, which has been implicated in the active site. However, little is yet known of the catalytic mechanism of these hydrogenases, or their interaction with other electron transport components in the cytoplasmic membrane. The environmental factors affecting the rate of nitrogenase-mediated H2 production and of hydrogenase-mediated H2 uptake have been the subject of much investigation. Nitrogenase synthesis, as in non-photosynthetic bacteria, is repressed by reduced nitrogenous compounds such as NH4+, by 0 2 , and by molybdenum starvation. Light, while not essential for nitrogenase synthesis and activity, is greatly stimulatory. Continuous culture studies combined with immunoassay of the nitrogenase polypeptides have shown that the increase in H2 production with increased light intensity is due to an increased synthesis of nitrogenase. Elucidation of the molecular mechanisms involved in the regulation of nitrogenase synthesis will require further study. Uptake hydrogenase is synthesized maximally during conditions of H2 production, and is theoretically capable of recycling the H2 evolved by nitrogenase. However, H2 uptake under anaerobic conditions, due to the photoreduction of C02, appears to be inhibited by the high concentrations of organic substrates required for optimal H2 production. Experiments with hydrogenase-deficient (Hup-) mutants have shown that these mutants produce more H2 than the wild type from various substrates. This

220

PAULETTE M. VlGNAlS ET AL.

has been interpreted as indicating that H2 recycling does in fact occur in wild-type strains in the presence of excess organic substrate. However, an alternative explanation has been suggested, that these mutants are affected in pathways of carbon metabolism, resulting in an increased flow of reducing equivalents from organic substrates to nitrogenase. Identification of the pathways of carbon metabolism that occur concomitantly with H2 production is clearly of interest in resolving this discrepancy. Recent advances in the genetics of photosynthetic bacteria have permitted genetic analysis of mutants affected in nitrogen fixation (Nif-) and cloning of the structural genes for the nitrogenase complex. These techniques will permit an analysis of the organization and expression of the genes coding for enzymes of hydrogen metabolism, as well as the genetic manipulation of different strains. Finally, the large body of scientific knowledge concerning the photosynthetic bacteria has enabled an evaluation of their biotechnological potential. Their high rates of H2 production, their ability to degrade a wide variety of organic substrates to H2 and C 0 2 with a more or less high efficiency, and their amenability to genetic analysis make them interesting candidates for the bioconversion of solar energy into H2. Solar H2 production can be envisaged either as a primary process or as a secondary process accompanying water depollution or biomass production. Means of improving the economic viability of such systems were discussed.

IX. Acknowledgements This work was supported by research grants from the Centre National de la Recherche Scientifique (APP PIRSEM), Agence FranGaise pour la Maitrise de 1’Energie (AFME), and EEC Solar Energy Research and Development Program (ESD 18-F).

REFERENCES

Adams, M. W. W., and Hall, D. 0. (1977). Biochemical and Biophysical Research Communications 77, 730. Adams, M. W. W., and Hall, D. 0. (1979). Archives ofBiochemistry andBiophysics 195, 288. Adams, M. W. W., Mortenson, L. E., and Chen, J. S. (1981). Biochimica et Biophysica Acta 594, 105. Albracht, S . P. J., Albrecht-Ellmer, K. J., Schmedding, D. J. M., and Slater, E. C. (1982a). Biochimica et Biophysica Acta 681, 330. Albracht, S. P. J., Graf, E. G., and Thauer, R. K. (1982b). FEBS Letters 140,3 1 1 .

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

221

Albracht, S. P. J., Kalkman, M. L., and Slater, E. C. (1983). Biochimica et Biophysica Acta 724, 309. Albrecht, S. L., Maier, R. J., Hanus, F. J., Russell, S. A., Emerich, D. W., and Evans, H. J. (1979). Science 203, 1255. Alef, K., Arp, D. J., and Zumft, W. G. (1981). Archives of Microbiology 130, 138. Allibert, P., Odom, J. M., Wall, J. D., and Vignais, P. M. (1984). FEMS Microbiology Letters 23, 221. Andeisen, K , Tait, R. C., and King, W. R. (1981). Archives of Microbiology 129, 384. Anderson, L., and Fuller, R. C. (1967). Plant Physiology 42, 487. Aragno, M., and Schlegel, H. G. (1978). Archives of Microbiology 116, 221. Arp, D. J., and Zumft, W. G. (1983a). Journal of Bacteriology 153, 1322. Arp, D. J., and Zumft, W. G. (1983b). Archives ofMicrobiology 134, 17. Avtges, P., Scolnik, P. A., and Haselkorn, R. (1983). Journal of Bacteriology 156, 251. Baccarini-Melandri, A., Zannoni, D., and Melandri, B. A. (1973). Biochimica et Biophysica Acta 314, 298. Bachofen, R., and Amon, D. I. (1966). Biochimica et Biophysica Acta 120, 259. Bagyinka, C, Kovacs, K. L., and Rak, E. (1981a). Acta Biochimica et Biophysica Academiae Scientiarum Hungaricae 16, 235. Bagyinka, C., Dancshazy, Z, Kovacs, K. L., Ormos, P., and Keszthelyi, L. (1981b). Acta Biologica Academiae Scientiarum Hungaricae 32, 31 1. Bagyinka, C, Kovacs, K. L., and Rak, E. (1982). Biochemical Journal 202, 255. Bagyinka, C., Zorin, N. A., and Kovacs, K. L. (1984). Analytical Biochemistry 142, 7. Bavendamm, W. (1924). Die farblosen und roten Schwefelbakterien des Suss- und Salzwassers. Planzenforschung (Kolkwitz, ed.), Berlin Dahlem, pp. 1-156. Behki, R. M., Selvaraj, G., and Iyer, V. N. (1983). Canadian Journal of Microbiology 29, 767. Bender, R. A., Snyder, P. M., Bueno, R., Quinto, M., and Magasanik, B. (1983). Journal of Bacteriology 156, 444. Benemann, J. R. (1977). In “Living Systems as Energy Converters” (R. Buvet, M. J. Allen, and J. P. Massue, eds.), pp. 285-297. ElsevierNorth-Holland, Amsterdam. Benemann, J. R., Miyamoto, K., and Hallenbeck, P. C. (1980). Enzyme and Microbial Technology 2, 103. Bennett, M. A., and Weetall, H. H. (1976). Journal of Solid Phase Biochemistry 1, 137. Bennett, R., and Fuller, R. C. (1964). Biochemical Biophysical Research Communications 16, 300. Bennett, R., Rigopoulos, N., and Fuller, R. C. (1964). Biochemistry 52, 762. Berndt, H., and Wolfle, D. (1978). In “Hydrogenases: their Catalytic Activity, Structure and Function” (H. G. Schlegel and K. Schneider, eds.), pp. 327-351. Goltze Verlag, Gottingen. Bernstein, J. D., and Olson, J. M. (1981). Photochemistry and Photobiology 34, 725. Bishop, N. I., and Jones, L. W. (1978). Current Topics in Bioenergetics 8, 3. Bogorov, L. V. (1974). Mikrobiologiya 43, 326. Bose, S. K., and Gest, H. (1963). Proceedings of the National Academy of Sciences of the United States of America 49, 337.

222

PAULETTE M. VIGNAIS E r AL.

Bothe, J. (1982).Experientia 38, 59. Bothe, J., Distler, E., and Eisbrenner, G. (1978).Biochimie 60,277. Bowien, B., and Schlegel, H. G. (1981).Annual Review of Microbiology 35,405. Broach, J., Neumann, C., and Kusku, S. (1976).Journal of Bacteriology l28,86. Buchanan, B. B.,and Arnon, D. I. (1970).Advances in Enzymology 33, 119. Buchanan, B. B.,and Bachofen, R. (1968).Biochimica et Biophysica Acta 162,

607.

Buchanan, B. B.,and Evans, M. C. W. (1969).Biochimica et Biophysica Acta

180, 123.

Buchanan, B. B., Evans, M. C. W., and Arnon, D. I. (1967).Archiv fur Mikrobiotogie 59, 32. Bulen, W.A., and Le Comte, J. R. (1966).Proceedings of the National Academy of Sciences of the United States of America 56, 979. Bulen, W. A., Bums, R. C., and Le Comte, J. R. (1965a).Proceedings of the National Academy of Sciences of the United States of Amercia 53, 532. Bulen, W.A., Le Comte, J. R., Burns, R. C., and Hinkson, J. (1965b).In “Nonheme Iron Proteins: Role in Energy Conversion” (A San Pietro, ed.), pp. 261287. Antioch Press, Yellow Springs, Ohio. Burgess, B. K. (1984).In “Advances in Nitrogen Fixation Research” (C. Veeger and W. E. Newton, eds.), pp. 103-114. Martinus Nijhoff, The Hague. Burns, R. C., and Bulen, W. A. (1966).Archives of Biochemistry and Biophysics

1l3, 461.

Cammack, R.,Datil, D., Aguirre, R., and Hatchikian, E. C. (1982).FEBS Letters

142, 289.

Cantrell, M.A., Haugland, R. A., and Evans, H. J. (1983).Proceedings of the National Academy of Sciences of the United States of America 80, 181. Carithers, R. P., Yochi D. C., and-Arnon, D. I. (1979).-JournaIof Bacteriology

137. 779.

Castillo, F., and Cardenas, J. (1982).Photosynthesis Research 3, 3. Caumette, P. (1984).Canadian Journal of Microbiology 30, 273. Caumette, P., Pagano, M., and Saint-Jean, L. (1983).Hydrobiologia 106, 135. Chakrabarti, S.,and Smith, G. D. (1981).Biochemistry International 2, 59. Chibata, I., and Tosa, T. (1977).Advances in Applied Microbiology 22, 1. Chibata, I., and Tosa, T. (1980).Trends in Biochemical Sciences 5 , 88. Chibata, I., and Wingard, L. B. (eds.) (1983).“Immobilized Microbial Cells”. Applied Biochemistry and Bioengineering, Vol. 4. Academic Press, New York and London. Chuaqui, C. A., Lynn, K. R., Williams, R. E., and Martin, S. M. (1980).Makromolekulare Chemie 181,651. Clayton, R. K., and Sistrom, W. R. (eds.) (1978).The Photosynthetic Bacteria. Plenum, New York. Colbeau, A., and Vignais, P. M. (1981).Biochimica et Biophysica Acta 662,271. Colbeau, A., and Vignais, P. M. (1983).Biochimica et Biophysica Acta 748, 128. Colbeau, A., Chabert, J., and Vignais, P. M. (1978).In “Hydrogenases: their Catalytic Activity, Structure and Function” (H. G. Schlegel and K. Schneider, eds.), pp. 183-197. Goltze Verlag, Gottingen. Colbeau, A., Kelley, B. C., and Vignais, P. M. (1980).Journal of Bacteriology

144, 141.

Colbeau, A., Chabert, J., and Vignais, P. M. (1983).Biochimica et Biophysica Acta 748, 116.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

223

Copeland, H. F. (1956). “The Classification of Lower Organisms”. Pacific Books, Palo Alto. Cox, J. C. (1983). FEMS Microbiology Letters 17, 97. Cox, J. C., Madigan, M. T., Favinger, J. L., and Gest, H. (1980). Archives of Biochemistry and Biophysics 204, 10. Dalton, H., and Postgate, J. R. (1969). Journal of General Microbiology 54, 463. Dilworth, M. J. (1966). Biochimica et Biophysica Acta 127, 285. Dixon, R. 0. D. (1972). Archiu fir Mikrobiofogie 85, 193. Dixon, R. 0. D. (1978). Biochimie 60,233. Doussikre, J., Porte, F., and Vignais, P. M. (1980). FEBS Letters 114, 291. Dowling, T. E., Preston, G. G., and Ludden, P. W. (1982). Journal of Biological Chemistry 257, 13987. Drake, H. L., Shou-Ih, J. U., and Wood, H. G. (1980). Journal ofBiologica1 Chemistry 255, 7174. Dunstan, R. H., Kelley, B. C., and Nicholas, D. J. D (1982). Journal ofBacterio1ogy 150, 100. Eisbrenner, G., and Evans, H. J. (1982). Plant Physiology 70, 1667. Evans, M. C. W., Telfer, A., and Smith, R. V. (1973). Biochimica et Biophysica Acta 310, 344. Falk, G., and Johansson, B. C. (1983). FEMS Microbiology Letters 19, 145. Falk, G., Johansson, B. C., and Nordlund, S. (1982). Archives ofMicrobiology 132, 251. Feigenblum, E., and Krasna, A. I. (1970). Biochimica et Biophysica Acta 198, 157. Fornari, C. S.,and Kaplan, S. (1982). Journal of Bacteriology 152,89. Fornari, C. S., and Kaplan, S. (1983). Gene 25, 291. Fox, G. E., Stackebrandt, E., Hespell, R. B., Gibson, J., Maniloff, J., Dyer, T. A., Wolfe, R. S., Balch, W. E., Tanner, R. S . , Magrum, L. J., Zablen, L. B., Blakemore, R., Gupta, R., Bonen, L., Lewis, B. J., Stahl, D. A., Luehrsen, K. R., Chen, K. N., and Woese, C. R. (1980). Science 209, 457. Friedrich, C. G. (1982). Journaf of Bacteriology 149, 203. Friedrich, C. G., and Friedrich, B. (1983). Journal of Bacteriology 153, 176. Friedrich, C. G., Friedrich, B., and Bowien, B. (1981a). Journal of General Microbiology 122, 69. Friedrich, B., Heine, E., Finck, A., and Friedrich, C. G. (1981b). Journal of Bacteriology 145, 1144. Friedrich, B., Hogrefe, C., and Schlegel, H. G. (1981~).Journal of Bacteriology 147, 198. Friedrich, C. G., Schneider, K., and Friedrich, B. (1982). Journal of Bacteriology 152,42. Fukui, S., and Tanaka, A. (1982). Annual Review of Microbiology 36, 145. Gaffron, H. (1933). Biochemische Zeitschrift 260, 1 . Gaffron, H. (1935). Biochemische Zeitschrgt 275, 301. Gest, H. (1951). Bacteriological Reviews 15, 183. Gest, H. (1952). Journal of Bacteriology 63, 1 1 1 . Gest, H. (1972). Advances in Microbial Physiology 7 , 243. Gest, H., and Kamen, M. D. (1949a). Journal of Bacteriology 58, 239. Gest, H., and Kamen, M. D. (1949b). Science 109, 558. Gest, H., Kamen, M. D., and Bregoff, H. M. (1950). Journal ofBiologica1 Chemistry 182, 153.

224

PAULETTE M. VlGNAlS ET AL.

Gest, H., Ormerod, J. G., and Ormerod, K. S. (1962). Archives of Biochemistry and Biophysics W,21. Gibbons, N. E., and Murray, R. G. E. (1978). International Journal of Systematic Bacteriology 28, 1. Gibson, J. (1967). Archiv fiir Mikrobiologie 59, 104. Gibson, J., Stackebrandt, E., Zablen, L. B., Gupta, R., and Woese, C. R. (1979) Current Microbiology 3, 59. Gitlitz, P. H., and Krasna, A. I. (1975). Biochemistry 14, 2561. Gogotov, I. N. (1978). Biochimie 60,267. Gogotov, 1. N., Zorin, N. A., and Kondratieva, E. N. (1976). Biochemistry United Soviet Socialist Republic 41, 685. Gogotov, I. N., Zorin, N. A., and Laurinavichene, T. V. (1978a). In “Hydrogenases: Their Catalytic Activity, Structure and Function” (H. G. Schlegel and K. Schneider, eds.), pp. 171-182. Goltze Verlag, Gottingen. Gogotov, I. N., Zorin, N. A., Serebriakova, L. T., and Kondratieva, E. N. (1978b). Biochimica et Biophysica Acta 523, 335. Gorlenko, V. M., Krasilnikova, E. N., Kikina, 0. G., and Tatarinova, N. Yu (1979). Zzvestija Akademii nauk SSSR-Serija Biologiceskaja N5, 755. Gorrell, T. E., and Uffen, R. L. (1977). Journal of Bacteriology 131, 533. Gorrell, T. E., and Uffen, R. L. (1978). Photochemistry and Photobiology 27, 351. Gotto, J. W., and Yoch, D. C. (1982a). Journal ofBiologica1 Chemistry 257,2868. Gotto, J. W., and Yoch, D. C. (1982b). Journal of Bacteriology 152, 714. Graf, E. G., and Thauer, R. K. (1981). FEBS Letters 136, 165. Graham, A. (1981). Biochemical Journal 19, 283. Gray, C. T., and Gest, H. (1965). Science 148, 186. Giirgiin, V., Kirchner, G., and Pfennig, N. (1976). Zeitschrgt fur Allgemeine Mikrobiologie 16, 573. Guth, J. H., and Bums, R. H. (1983a). Biochemical Journal 2l3, 741. Guth, J. H., and Bums, R. H. (1983b). Biochimica et Biophysica Acta 749, 91. Hallenbeck, P. C. (1983a). Enzyme and Microbial Technology 5, 171. Hallenbeck, P. C. (1983b). Archives of Biochemistry and Biophysics 220, 657. Hallenbeck, P.C., and Benemann, J. R. (1979). In “Photosynthesis in Relation to Model Systems” (J. Barber, ed.), pp. 331-364. ElsevierDVorth-Holland, Amsterdam. Hallenbeck, P. C., and Vignais, P. M. (1981). FEMS Microbiology Letters 12, 15. Hellenbeck P. C., Meyer, C. M., and Vignais, P. M. (1982a).Journal ofBacterio1ogy 149, 708. Hallenbeck, P.C., Meyer, C. M., and Vignais, P. M. (1982b). Journal ofBacteriology 151, 1612. Hallenbeck, P. C., Jouanneau, Y., and Vignais, P. M. (1982~).Biochimica et Biophysica Acta 681, 168. Haugland, R. A., Hanus, F. J., Cantrell, M. A., and Evans, H. J. (1983). Applied and Environmental Microbiology 45, 892. Hendley, D.D. (1955). Journal of Bacteriology 70, 625. Henry, M. F., and Vignais, P. M. (1983). Archives of Microbiology 136, 64. Hillmer, P., and Gest, H. (1977a). Journal of Bacteriology 129, 724. Hillmer, P.,and Gest, H. (1977b). Journal of Bacteriology 129, 732. Hillmer, P.,and Fahlbusch, K. (1979). Archives of Microbiology 122,213. Hirsch, P.(1968). Nature, London 217, 555.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

225

Hiura, H., Kakuno, T., Yamashita, J., Bartsch, R. G., and Horio, T. (1979). Journal ofBiochemistry, Tokyo 86, 1151. Houchins, J. P. (1984). Biochimica et Biophysica Acta 768, 227. Imhoff, J. F. (1984). Znternational Journal of Systematic Bacteriology 34, 338. Imhoff, J. F., Truper, H. G., and Pfennig, N. (1984). Znternational Journal of Systematic Bacteriology 34, 340, Jacobson, F. S., Daniels, L., Fox, J. A., Walsh, C. T., and Orme-Johnson, W. H. (1982). Journal of Biological Chemistry 257, 3385. Johansson, C., and Gest, H. (1976). Journal of Bacteriology 128, 683. Jones, R. W. (1980). Biochemical Journal 188, 345. Jones, B. L., and Monty, K. J. (1979). Journal of Bacferioiogy 139, 1007. Jones, 0. T. G., and Saunders, V. A. (1972). Biochimica et Biophysica Acta 275, 427. Jones, R. W., Gray, T. A., and Garland, P. B. (1976). Biochemical Society Transactions 4, 671. Jouanneau, Y . (1982). Doctorat 5s Sciences. UniversitC Scientifique et Medicale de Grenoble. Jouanneau, Y., Kelley, B. C., Berlier, Y., Lespinat, P. A., and Vignais, P. M. (I980a). Journal of Bacteriology 143, 628. Jouanneau, Y., Siefert, E., and Pfennig, N. (1980b). FEMS Microbiology Letters 9, 89. Jouanneau, Y., Willison, J. C., Colbeau, A., Hallenbeck, P. C., Riolacci, C., and Vignais, P. M. (1982). In “Photochemical, Photoelectrochemical and Photobiological Processes” (D. 0. Hall and W. Palz, eds.), Solar Energy Research and Development in the European Community, Series D, Vol. 1 , pp. 174-179. Reidel, Dordrecht. Jouanneau, Y., Meyer, C. M., and Vignais, P. M. (1983). Biochimica et Biophysics Acta 749, 3 18. Jungermann, K., and Schon, G. (1974). Archives of Microbiology 99, 109. Kakuno, T., Kaplan, N. O., and Kamen, M. D. (1977). Proceedings of the National Academy of Sciences of the United States of America 74, 861. Kakuno, T., Hiura, H., Yamashita, J., Bartsch, R. G., and Horio, T. (1978). Journal of Biochemistry, Tokyo 84, 1649. Kaltwasser, H., and Frings, W. (1981). In “Nickel in the Environment” (J. 0. Nriagu, ed.), pp, 463-491. John Wiley, New York. Kamen, M. D. and Gest, H. (1949). Science 109,560. Kampf, C., and Pfennig, N. (1982). International Symposium on Photosynthetic Prokaryotes I V , Institut Pasteur, Paris. Abstract Book, p. A24. Kanemoto, H., Dowling, T. E., and Ludden, P. W. (1984). In “Advances in Nitrogen Fixation Research” (C. Veeger and W. E. Newton, eds.), p. 153. Martinus Nijhoff, The Hague. Katoh, T. (1963). Plant and Cell Physiology 4, 199. Keister, D. L., and Yike, N. J. (1967). Archives of Biochemistry and Biophysics 121, 415. Kelley, B. C., Meyer, C. M., Gandy, C., and Vignais, P. M. (1977). FEBS Letters 81. 281. Kelley, B. C., Jouanneau, Y., and Vignais, P. M. (1979). Archives of Microbiology 122, 145. KeUey, €3. C., and Nicholas, D. J. D. (1981). Archiues ofMicrobiofogy 129, 344. Khanna, S., Kelley, B. C., and Nicholas, D. 1. D. (1980). Journal ofGenera1 Microbiology 119, 28 1.

226

PAULETTE M. VlGNAlS ET AL.

Khanna, S., Kelley, B. C., and Nicholas, D. J. D. (1981). Archives of Microbiology 128,42 1 . Kim, J. S., Ito, K., and Takahashi, H. (1982). Agricultural and Biological Chemistry 46, 937. Klemme, J. H. (1968). Archiv fur Mikrobiologie 64, 29. Klemme, J. H. (1969). Zeitschrift fiir Naturforschung U b , 67. Klemme, J. H. (1979). Microbiologica 2, 415. Klemme, J. H., and Schlegel, H. G. (1967). Archiv fur Mikrobiologie 59, 185. Klibanov, A. M. (1983). Process of Biochemistry 18, 13. Klibanov, A. M., Kaplan, N. O., and Kamen, M. D. (1980). Archives ofBiochemistry and Biophysics 199, 545. Klibanov, A. M., Alberti, B. N., and Zale S. E. (1982). Biotechnology and Bioengineering 24, 25. Knaff, D. B. (1978). In “The Photosynthetic Bacteria” (R. K. Clayton and W. R. Sistrom, eds.), pp. 629-640. Plenum, New York. Kobayashi, M. (1975). Progress in Water Technology 7 , 309. Kobayashi, M. (1976). In “Microbial Energy Conversion” (H. G. Schlegel and J. Barnea, eds.), pp. 443-453. Goltze Verlag, Gottingen. Kobayashi, M., and Kurata, S. I. (1978). Process Biochemistry W, 27. Kobayashi, M., and Tchan, Y. T. (1973). Water Research 7, 1219. Kobayashi, M., Fujii, K., Shimamoto, I., and Maki, T. (1978). Progress in Water Technology 2, 279. Kohlmiller, E. F., Jr. and Gest, H. (1951). Journal of Bacteriology 61, 269. Kojima, N., Fox, J. A., Hausinger, R. P., Daniels, L., Orme-Johnson, W. H., and Walsh, C. (1983). Proceedings of the National Academy of Sciences of the United States of America 80, 378. Kondratieva, E.N. (1979). In “International Review of Biochemistry Microbial Biochemistry” (J. R. Quayle, ed.), Microbial Biochemistry, Vol. 21, pp. 117175. University Park Press, Baltimore. Kondratieva, E. N., and Gogotov, I. N. (1983). In “Advances in Biochemical Engineering/Biotechnology”(A. Fiechter, ed.), Vol. 28, pp. 139-191. Springer Verlag, Berlin and New York. Kondratieva, E. N., Petushkova, Yu, P., and Zhukov, V. G. (1975). Mikrobiologiya 44, 389. Kondratieva, N. E., Zhukov, V. G., Ivanovsky, R. N., Petushkova, Yu, P., and Monosov, E. G. (1976). Archives of Microbiology 108, 287. Kondratieva, E. N., Ivanosky, R. N., and Krasilnikova, E. N. (1981). Soviet Science Review 2, 325. Kovacs K. L., and Bagyinka, C. (1982). I V International Symposium on Photosynthetic Procaryotes. Institut Pasteur, Paris. Abstract Book p. C18. Kovacs, K. L., Bagyinka, C., and Serebriakova, L. T. (1983). Current Microbiology 9, 215. Krasil’nikova, E. N., Zaharchuk, L. M., and Linnik, L. M. (1980). Mikrobiologiya 49, 244. Krasna, A. I. (1979). Enzyme and Microbial Technology 1, 165. Krasna, A. I. (1980). Journal of Bacteriology 144, 1094. Kiihl, S. A., Nix, D. W., and Yoch, D. C. (1983). Journal of Bacteriology 156, 737. Kriiger, H. J., Huynh, B. H., Ljungdahl, P. O., Xavier, A. V., Dervartanian, D. V., Moura, I., Peck, H. D., Teixeira, M., Moura, J. J. G., and Le Gall, J. (1982). Journal of Biological Chemistry 257, 14620.

HP METABOLISM IN PHOTOSYNTHETIC BACTERIA

227

Kumazawa, S., and Mitsui, A. (1982). In “CRC Handbook of Biosolar Resources” (A. Mitsui and C. C. Black, eds.), Vol. 1 , Basic Principles (Series ed. 0. R. Zaborsky), Part 1 , pp. 299-316. CRC Press, Boca Raton, Florida. Kumazawa, S., Izawa, S., and Mitsui, A. (1983). Journal of Bacteriology 154, 185. Lalla-Maharajh, W. V., Hall, D. O., Cammack, R., Rao, K. K., and Le Gall, J. (1983). Biochemical Journal 209, 445. Lambert, G. R., and Smith, G. D. (1981). Biological Review 56, 589. La Monica, R. F., and Marrs, B. L. (1976). Biochimica et Biophysica Acta 423, 431. Lancaster, R. J. (1980). FEBS Letters 115, 285. Le Gall, J., Ljungdahl, P. O., Moura, I., Peck, H. D. Jr., Xavier, A. V., Moura, J. J. G, Teixeira, M., Huynh, B. H., and Der Vartanian, D. V. (1982). Biochemical and Biophysical Research Communications 106, 610. Lien, S., and San Pietro, A. (1975). “An Inquiry into Biophotolysis of Water to Produce Hydrogen”. NSF/RANN Report. Indiana University. Lim, S. T., Andersen, K., Tait, R., and Valentine, R. C. (1980). Trends in Biochemical Sciences 5, 167. Lindstrom, E. S., Bums, R. H., and Wilson, P. W. (1949). Journal of Bacteriology 58, 313. Lindstrom, E. S., Lewis, S. M., and Pinsky, M. J. (1951). Journal ofBacteriology 61, 481. Lippert, K. D., and Pfennig, N. (1969). Archiu f i r Mikrobiologie 65, 29. Ljones, T. (1979). FEBS Letters 98, 1. Llama, M. J., Serra, J. L., Rao, K. K., and Hall, D. 0. (1979). Biochemical Society Transactions 7, 1 1 17. Llama, M. J., Serra, J. L., Rao, K. K., and Hall, D. 0. (1981). European Journal of Biochemistry 114, 89. Losada, M., Nozaki, M., and Arnon, D. I. (1961). In “Light and Life” (W. D. McElroy and B. Glass, eds.), pp. 570-575. Johns Hopkins University Press, Baltimore. Lowe, D. J., Smith, B. E., and Eady, R. R. (1980). In “Recent Advances in Biological Nitrogen Fixation” (N. S. Subba Rao, ed.), pp. 34-87. Oxford and IBH Publishing Company, New Delhi. Ludden, P. W., and Burris, R. H. (1976). Science 194,424. Ludden, P. W., and Bums, R. H. (1978). Biochemical Journal 175, 251. Ludden, P. W., and Bums, R. H. (1979). Proceedings of National Academy of Sciences of the United States of America 76, 6201. Ludden, P. W., Okon, Y., and Bums, R. H. (1978). Biochemical Journal 173, 1001. Ludden, P. W., Hageman, R. V., Orme-Johnson, H., and Bums, R. H. (1982a). Biochimica et Biophysica Acta 700, 213. Ludden, P. W., Preston, G. G., and Dowling, T. E. (1982b). Biochemical Journal 203, 663. Ludden, P. W., Murrell, S. A., Pope, M., Kanemoto, R., Dowling, T. E., Saari, L. L., and Triplett, E. (1984). In “Advances in Nitrogen Fixation Research” (C. Veeger and W. E Newton, eds.), pp. 181-187. Martinus Nijhoff, The Hague. Macler, B. A., Pelroy, R. A., and Bassham, J. A. (1979). Journal ofBacteriology l38, 446. Madigan, M. T., and Gest, H. (1978). Archives of Microbiology 117, 119.

228

PAULETTE M. VlGNAlS ET AL.

Madigan, M. T., and Gest, H. (1979). Journal of Bacteriology 137, 524. Madigan, M. T., Wall, J. D., and Gest, H. (1979). Science 204, 1429. Madigan, M. T., Cox, J. C., and Gest, H. (1980). Journal ofBacteriology 142, 908. Magasanik, B. (1982). Annual Review of Genetics 16, 135. Maier, R. J., and Mutafschiev, S. (1982). Journal of Biological Chemistry 257, 2092. Maier, R. J., Hanus, F. J., and Evans, H. J. (1979). Journal ofBacteriology 137, 824. Marchalonis, J. J., and Weltman, J. K. (1971). Comparative Biochemistry and Physiology 38B, 609. Marrs, B . (1974). Proceedings of National Academy of Sciences of the United States of America 71, 971. Marrs, B. (1981). Journal of Bacteriology 146, 1003. Marrs, B., and Gest, H. (1973). Journal ofBacteriology 114, 1045. Matheron, R., and Baulaigue, R. (1983). Archives of Microbiology 135,21 1 . Matsunaga, T., and Mitsui, A. (1982). Biotechnology and Bioengineering Symposium N o . 12, pp. 441-450. John Wiley, New York. Mayhew, S. G. (1978). European Journal of Biochemistry 85, 535. McEwan, A. G., Ferguson, St. J., and Jackson, J. B . (1983). Archives ofMicrobiology 136, 300. Melandri, B. A . , Zannoni, D., De Santis, A., and Casadio, R. (1982). In “Photochemical, Photoelectrochemical and Photobiological Processes” (D. 0. Hall and W. Palz, eds.), Solar Energy Research and Development in the European Community, Series D, Vol. 1 , pp. 160-164. Reidel, Dordrecht. Melandri, B. A., Zannoni, D., De Santis, A., and Casadio, R. (1983). In “Photochemical, Photoelectrochemical and Photobiological Processes” (D. 0. Hall, W. Palz, and D. Pimvitz, eds.), Solar Energy Research and Development in the European Community, Series D, Vol. 2, pp. 208-213. Reidel, Dordrecht. Merberg, D., and Maier, R. J. (1983). Science 220, 1064. Merberg, D., O’Hara, E. B., and Maier, R. J. (1983). Journal ofBacteriology 156, 1236. Merrick, M.J. (1983). The EMBO Journal 2, 39. Meyer, J., and Vignais, P. M. (1979). Biochemical and Biophysical Research Communications 89, 353. Meyer, J . , Kelley, B. C., and Vignais, P. M. (1978a). Biochimie 60, 245. Meyer, J., Kelley, B. C . , and Vignais, P. M. (1978b).Journal ofBacteriology 136, 201 Meyer, J., Kelley, B. C., and Vignais, P. M. (1978~).FEBS Letters 85, 224. Meyer, O., and Schlegel, H. G. (1983). Annual Review of Microbiology 37, 277. Michalski, W.P., Nicholas, D. J. D., and Vignais, P. M. (1983). Biochirnica et Biophysica Acta 743, 136. Mitsui, A. (1976). In “Proceedings of First World Hydrogen Energy Conference: Progress in World Hydrogen Energy Projects and Planning” (T. N. Veziroglu, ed.), Vol. 2, p. 77-99. University of Miami Press, Florida. Mitsui, A. (1978). In “Bioenergy, Proceedings of International Conference on Bioenergy, Zurich”. Springer Verlag, Berlin and New York. Mitsui, A. (1979). In “Solar-Hydrogen Energy Systems” (T. Ohta, ed.), Ch. 8, pp. 171-191. Pergamon, Oxford and New York. Mitsui, A, (1981). In “Proceedings of Solar Hydrogen Production Conference”, SERUSP-624-1095, pp. 1-19. I

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

229

Mitsui, A., Ohta, Y., Frank, J., Kumazawa, S., Hill, C., Rosner, D., Barciella, S., Greenbaum, J., Haynes, L., Oliva, L., Dalton, P., Radway, J., and Griffard, P. (1980). In “Alternative Energy Sources 11” (T. N. Veziroglu, ed.), Vol. 8, pp. 3483-3510. Hemisphere Publishing Co., Washington, D. C. Mitsui, A., Matsunaga, T., and Kumazawa, S. (1982). Proceedings of Solar Hydrogen Production Conference, p. 42. Solar Energy Research Institute-DOE CT-231-1873. Miyake, J., Tomizuka, N., and Kamibayashi, A. (1982). Journal of Fermentation Technology 60, 199. Mortenson, L. E. (1978). Biochimie 60, 219. Mortenson, L. E., and Thorneley, R. N. F. (1979). Annual Review ofBiochemistry 48, 387. Munson, T. D., and Bums, R. H. (1969). Journal of Bacteriology 97, 1093. Nakamura, H. (1937). Acta Phytochimica, Japan 9, 189. Nakamura, H. (1938). Acta Phytochimica, Japan 10, 259. Nakamura, H. (1939). Acta Phytochimica, Japan 11, 109. Nakamura, H. (1941). Acta Phytochimica, Japan 12, 43. Nakos, G., and Mortenson, L. E. (1971). Biochimica et Biophysica Acta 227, 576. Nelson, M. J., Levy, M. A., and Orme-Johnson, W. H. (1983). Proceedings of the National Academy of Sciences of the United States of America 80, 147. Nokhal, T. H., and Schlegel, H. G. (1980). Antonie van Leenwenhoek 46, 143. Nordlund, S., and Baltscheffsky, H. (1973). ZX International Congress of Biochemistry Abstract No. 4113. Abstract Book, p. 240. International Union of Biochemistry. Nordlund, S., and Ludden, P. W. (1983). Biochemical Journal 209, 881. Nordlund, S., and NorCn, A. (1984). In “Advances in Nitrogen Fixation Research” (C. Veeger and W. E. Newton, eds.), p. 161. Martinus Nijhoff, The Hague. Nordlund, S., Eriksson, U., and Baltscheffsky, H. (1977). Biochimica et Biophysics Acta 462, 187. Nordlund, S., Eriksson, U., and Baltscheffsky, H. (1978). Biochimica et Biophysics Acta 504, 248. O’Brian, M. R., and Maier, R. J. (1982). Journal ofBacteriology 152, 422. Odom, J. M., and Wall, J. D. (1983). Applied and Environmental Microbiology 45, 1300. Ohta, Y., and Mitsui, A. (1981). In “Advances in Biotechnology” (M. MooYoung and C. W. Robinson, eds.), Vol. 2, pp. 303-307. Pergamon, Toronto. Ohta, Y., Frank, J., and Mitsui, A. (1981). International Journal of Hydrogen Energy 6 , 451. Olson, J. M., Prince, R. C., and Brune, D. C. (1976). Brookhaven Symposia in Biology 28, 238. Ormerod, J. G., and Gest, H. (1962). Bacteriological Review 26, 51. Ormerod, J. G., Ormerod, K. S., and Gest, H. (1961). Archives ofBiochemistry and Biophysics 94, 449. Partridge, C. D. P., and Yates, M. G. (1982). Biochemical Journal 204, 339. Paul, F., and Vignais, P. M. (1981). In “Energy from Biomass’’ (W. Palz, P. Chartier, and D. 0. Hall, eds.), pp. 665-671. Applied Science Publishers, London. Paul, F., Colbeau, A., and Vignais, P. M. (1979). FEBS Letters 106, 29. Pedrosa, F. O., and Yates, M. G. (1983). FEMS Microbiology Letters 17, 137.

230

PAULETTE M. VlGNAlS fT AL.

Pemberton, J. M., and Bowen, A. R. St. G. (1981). Journal of Bacteriology 147, 110. Pemberton, J . M., Cooke, S., and Bowen, A. R. St. G. (1983). Annales de Microbiologie, Znstitut Pasteur 134B, 195. Pfennig, N. (1970). Journal of General Microbiology 61, ii. Pfennig, N. (1977). Annual Review of Microbiology 31, 275. Pfennig, N. (1978). I n “The Photosynthetic Bacteria” (R. K. Clayton and W. R. Sistrom, eds.), pp. 3-18. Plenum, New York. Pfennig, N., and Triiper, H. G. (1971). Journal of Systematic Bacteriology 21, 17. Pfennig, N., and Triiper, H. G. (1974). Zn “Bergey’s Manual of Determinative Bacteriology” (R. E. Buchanan and N. E. Gibbons, eds), 8th Ed., pp. 24-64. Williams & Wilkins, Baltimore. Pootjes, C. F. (1977). Biochemical et Biophysical Research Communications 76, 1002. Porte, F., and Vignais, P. M. (1980). Archives of Microbiology 127, 1 . Postgate, J . R., Partridge, C. D. P., Robson, R. L., Simpson, F. B., and Yates, M. G. (1982). Journal of General Microbiology 128, 905. Preston, G. G., and Ludden, P. W. (1982). Biochemical Journal 205,489. Prince, R. C., and Dutton, P. L. (1978). In “The Photosynthetic Bacteria” (R. K. Clayton and W. R. Sistrom, eds.), pp. 439-453. Plenum Press, New York and London. F’iihler, A., Aguilar, M. O. , Hynes, M., Muller, P., Klipp, W., Priefer, U., Simon, R., and Weber, G. (1984). In “Advances in Nitrogen Fixation Research” (C. Veeger and W. E. Newton, eds.), pp. 609-619. Martinus Nijhoff, The Hague. Qadri, S. M. H., and Hoare, D. S. (1968). Journal of Bacteriology 95, 2344. Ragsdale, S. W., Clark, J. E., Ljungdahl, L. G., Lundie, L. L., and Drake, H. L. (1983). The Journal of Biological Chemistry 258, 2364. Reidl, H., Golecki, J. R., and Drews, G. (1983). Biochimica e f Biophysica Acta 725, 455. Roberts, G. P., and Brill, W. J. (1981). Annual Review of Microbiology 35, 207. Robson, R. L. (1979). Biochemical Journal 181, 569. Robson, R. L., and Postgate, J. R. (1980). Annual Review of Microbiology 34, 183. Roelofsen, P. A. (1935). Doctoral thesis, University of Utrecht. Ruvkun, G. B., and Ausubel, F. M. (1980). Proceedings of The National Academy of Sciences of the United States of America 77, 191. Satoh, T. (1981). Plant and Cell Physiology 22, 443. Satoh, T., Hoshino, Y., and Kitamura, H. (1976). Archives of Microbiology 108, 265. Sawada, E., and Satoh, T. (1980). Plant and Cell Physiology 21, 205. Scherings, G., Haaker, H., and Veeger, C. (1977). European Journal ofBiochemistry 77, 621. Schick, H. J. (1971a). Archiv fur Mikrobiologie 75, 89. Schick, H. J. (1971b). Archiv fur Mikrobiologie 75, 102. Schick, H. J. (1971~).Archiv fur Mikrobiologie 75, 110. Schon, G. (1968). Archiv fur Mikrobiologie 63, 362. Schon, G., and Biedermann, M. (1973). Biochimica et Biophysica Acta 304, 65. Schon, G., and Voelskow, H. (1976). Archives of Microbiology 107, 87. Schlegel, H. G., and Eberhardt, U. (1972). In “Advances of Microbial Physiology” (A. H. Rose and D. W. Tempest, eds.), Vol. 7, pp. 205-242. Academic Press, London and New York.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

231

Schlegel, H. G., and Schneider, K. (1978). In “Hydrogenases: Their Catalytic Activity, Structure and Function” (H. G. Schlegel and K. Schneider, eds.), pp. 15-44. Goltze Verlag, Gottingen. Schneider, K., and Schlegel, H. G. (1976). Biochimica et Biophysica Acta 452, 66. Schultz, J. E., and Weaver, P. F. (1982). Journal of Bacteriology 149, 181. Scolnik, P. A., and Haselkorn, R. (1984). Nature, London 307, 289. Scolnik, P. A., Virosco, J., and Haselkorn, R. (1983). Journal of Bacteriology 155, 180. Serebriakova, L. T., and Gogotov, I. N. (1981). Prikladnoi Biokhimiya i Mikrobiologii 17, 555. Serra, J. L., Llama, M. J., Rao, K. K., and Hall, D. 0. (1979). Biochemical Society Transactions 7, 1 119. Shah, V. K., and Brill, W. S. (1977). Proceedings of The National Academy of Sciences of the United States of America 74, 3249. Shanmugam, K. T., Buchanan, B. B., and Amon, D. I. (1972). Biochimica et Biophysica Acta 256, 477. Shestakov, S., Frolova, V., Mitronova, T., and Belavina, N. (1983). Abstracts International Symposium, Pushchino, pp. 110-1 11. Federation of the European Microbiological Society. Shipman, R. H., Kao, I. C., and Fan, L. T. (1975). Biotechnology and Bioengineering 17, 1561. Siefert, E., and Pfennig, N. (1978). Biochimie 60, 261. Siefert, E., and Pfennig, N. (1979). Archives of Microbiology 122, 177. Siefert, E., and Pfennig, N. (1980). Archives of Microbiology 125, 73. Sim, E., and Vignais, P. M. (1978). Biochimie 60, 307. Solioz, M., Yen, H. C., and Marrs, B. (1975). Journal of Bacteriology 123. 651. Song, H., Chen, H., Wu, M., Chen, B., and Yu, B. (1980). Scientia Sinica 23, 252. Stephenson, M., and Stickland, L. H. (1931). Biochemical Journal 25, 205. Stevens, P., Van Der Sypt, H., De Vos, P., and De Ley, J. (1983). Biotechnology Letters 5, 369. Stoppani, A. 0. M., Fuller, R. C., and Calvin, M. (1955). Journal ofBacteriology 69, 491. Strekas, T., and Krasna, A. I. (1978). In “Hydrogenases: Their Catalytic Activity, Structure and Function” (H. G. Schlegel and K. Schneider, eds.), pp. 141150. Goltze Verlag, Gottingen. Strekas, T., Antanaitis, B. C., and Krasna, A. I. (1980). Biochimica et Biophysica Acta 616, 1. Sweet, W. J., and Burris, R. H. (1981). Journal of Bacteriology 145, 824. Tait, R. C. (1982). Enzyme and Microbial Technology 4, 66. Takakuwa, S., and Wall, J. D. (1981). FEMS Microbiology Letters 12, 359. Takakuwa, S., Odom, J. M., and Wall, J. D. (1983). Archives of Microbiology 136, 20. Tanaka, M., Haniu, M., Yasunobu, K. T., Evans, M. C. N., and Rao, K. K. (1974). Biochemistry 13, 2953. Tanaka, M., Haniu, M., Yasunobu, K. T., Evans, M. C. N., and Rao, K. K. (1975). Biochemistry 14, 1938. Teixeira, M., Moura, I., Xavier, A. V., Der Vartanian, D. V., Le Gall, J., Peck, H. D., Huynh, B. H., and Moura, J. J. G. (1983). European Journal of Biochemistry 130, 481.

232

PAULETTE M. VlGNAlS ET AL.

Thauer, R. K., Diekert, G., and Schonheit, P. (1980). Trends in Biochemical Sciences 5, 304. Thorneley, R. N. F. (1974). Biochimica et Biophysica Acta 333, 487. Thorneley, R. N. F., and Lowe, D. J. (1983). Biochemical Journal 215, 393. Triplett, E. W., Wall, J. D., and Ludden, P. W. (1982). Journal ofBacteriology 152, 786. Triiper, H. G. (1976). International Journal of Systematic Bacteriology 26, 74. Triiper, H. G., and Schlegel, H. G. (1964). Antonie van Leeuwenhoek 30, 225. Triiper, H. G., and Pfennig, N. (1981). Zn “The Prokaryotes-A Handbook on Habitats, Isolation and Identification of Bacteria” (M. P. Starr, H. Stolp, H. G. Triiper, A. Balows, and H. G. Schlegel, eds.), Ch. 18, pp. 299-312. Springer Verlag, Berlin and New York. Tsygankov, A. A., Pavlova, E. A., and Gogotov, I. N. (1982a). Microbiology, United Soviet Socialist Republic 51, 153. Tsygankov, A. A., Yakunin, A. F., and Gogotov, I. N. (1982b). Microbiology, United Soviet Socialist Republic 51, 433. Uffen, R. L. (1973a). Journal of Bacteriology 116, 874. Uffen, R. L. (1973b). Journal of Bacteriology 116, 1086. Uffen, R. L. (1976). Proceedings of The National Academy of Sciences of the United States of America 73, 3298. Uffen, R. L. (1978). In “The Photosynthetic Bacteria” (R. K. Clayton and W. R. Sistrom, eds.), pp. 857-872. Plenum, New York. Uffen, R. L. (1981). Enzyme and Microbial Technology 3, 197. Uffen, R. L., and Wolfe, R. S. (1970). Journal of Bacteriology 104, 462. Uffen, R. L., Sybesma, C., and Wolfe, R. S. (1971). Journal ofBacteriology 108, 1348. Unden, G., Bocher, R., Knecht, J., and Kroger, A. (1982). FEBS Letters 145, 230. van Heerikhuizen, H., Albracht, S. P. J., Slater, E. C., and van Rheenen, P. S. (1981). Biochimica et Biophysica Acta 657, 26. van Niel, C. B. (1931). Archiv fiir Mikrobiologie 3, 1. van Niel, C. B. (1936). Archiv fiir Mikrobiologie 7, 323. van Niel, C. B. (1944). Bacteriological Reviews 8, 1. van der Plas, J., Hellingwerf, K. J., Seijen, H. G., Guest, J. R., Weiner, J. H., and Konings, W. N. (1983). Journal of Bacteriology 153, 1027. van der Westen, H. M., Mayhew, S. G., and Veeger, C. (1978). FEBS Letters 86, 122. Vignais, P. M., Henry, M. F., Sim, E., and Kell, D. B. (1981a). In “Current Topics in Bioenergetics”, Vol. 12, pp. 115-196. Academic Press, New York. Vignais, P. M., Meyer, J., and Hallenbeck, P. C. (1981b). Postepy Biologii Komorki8, 141. Vignais, P. M., Henry, M. F., Berlier, Y., and Lespinat, P. A. (1982). Biochimica et Biophysica Acta 681, 519. Vincenzini, M., Balloni, W., Mannelli, D., and Florenzano, G. (1981). Experientia 37, 710. Vincenzini, M., Materassi, R., Tredici, M. R., and Florenzano, G. (1982a). Znternational Journal of Hydrogen Energy 7, 23 1. Vincenzini, M., Materassi, R., Tredici, M. R., and Florenzano, G. (1982b). Znternational Journal of Hydrogen Energy 7, 725. Voelskow, H., and Schon, G. (1978). Archives of Microbiology 119, 129. Voelskow, H . , and Schon, G. (1980). Archives of Microbiology 125, 245.

H2 METABOLISM IN PHOTOSYNTHETIC BACTERIA

233

Vrati, S., and Verma, J. (1983). Journal of Fermentation Technology 61, 157. Wakim, B. T., and Uffen R. L. (1983). Journal of Bacteriology 153, 571. Walker, C. C., and Yates, M. G. (1978). Biochimie 60,225. Wall, J. D., and Gest, H. (1979). Journal of Bacteriology 137, 1459. Wall, J. D., and Love, J. (1984). In “Advances in Nitrogen Fixation Research” (C. Veeger and W. E. Newton, eds.), p. 750. Martinus Nijhoff, The Hague. Wall, J. D., Weaver, P. F., and Gest, H. (1975a). Nature, London 258, 630. Wall, J. D., Weaver, P. F., and Gest, H. (1975b). Archives of Microbiology 105, 217. Wall, J. D., Johansson, B. C., and Gest, H. (1977). Archives ofMicrobiology 115, 259. Watanabe, K. I., Kim, J. S., Ito, K . , Buranakarl, L., Karnpee, T., and Takahashi, H. (1981). Agricultural and Biological Chemistry 45, 217. Weare, N. M. (1978). Biochimica et Biophysica Acta 502, 486. Weaver, P., Tinker, K . , and Valentine, R. C. (1965). Biochemical and Biophysical Research Communications 21, 195. Weaver, P. F., Lien, S., and Seibert, M. (1979). Conference on Microbial Enzyme Reactions SERUTP-33-409. Weaver, P. F., Lien, S., and Seibert, M. (1980). Solar Energy 24, 3. Wertlieb, D., and Vishniac, W. (1967). Journal of Bacteriology 93, 1722. Willison, J. C., and Vignais, P. M. (1982). Journal of General Microbiology 128, 3001. Willison, J. C., Jouanneau, Y., Colbeau, A., and Vignais, P. M. (1983a). Annales de Microbiologie, Institut Pasteur 134B, 115. Willison, J. C., Jouanneau, Y., Michalski, W. P., Colbeau, A., and Vignais, P. M. (1983b). In “Photosynthetic Prokaryotes: Cell Differentiation and Function” (G. C. Papageorgiou and L. Packer, eds.), pp. 333-352. Elsevier, New York. Willison, J. C., Madern, D., and Vignais, P. M. (1984). Biochemical Journal, 219, 593. Winter, H. C., and Burris, R. H. (1976). Annual Review of Biochemistry 45, 409. Yagi, T., Honya, M., and Tarniya, N. (1968). Biochimica et Biophysica Acta 153, 699. Yakunin, A. F., and Gogotov, I. N. (1983). Biochimica et Biophysica Acta 725, 298. Yamagata, S . , and Nakamura, H. (1937). Actu Phytochimica 10, 297. Yen, H. C., and Marrs, B. (1976). Journal of Bacteriology 126, 619. Yen, H. C., and Marrs, B. (1977). Archives of Biochemistry and Biophysics 181, 411. Yen, H. C., Hu, N. T., and Marrs, B. L. (1979). Journal of Molecular Biology 131, 157. Yoch, D. C. (1978). In “The Photosynthetic Bacteria” (R. K. Clayton and W. R. Sistrom, eds.), Ch. 34, pp. 657-672. Plenum, New York. Yoch, D. C. (1980). Biochemical Journal 187, 273. Yoch, D. C., and Arnon, D. I. (1970). Biochimica et Biophysica Acta 197, 180. Yoch, D. C., and Arnon, D. I. (1975). Journal of Bacteriology 121, 743. Yoch, D. C., and Carithers, R. P. (1979). Microbiological Reviews 43, 384. Yoch, D. C., and Cantu, M. (1980). Journal of Bacteriology 142,899. Yoch, D. C., and Gotto, J. W. (1982). Journal ofBacteriology 151, 800. Yoch, D. C., Arnon, D. I., and Sweeneyn W. V. (1975). Journal of Biological Chemistry 250, 8330.

234

PAULETTE M. VlGNAlS €7 AL.

Yoch, D. C., Zhang, Z. M., and Claybrook, D. L. (1983). Archives of Microbiology 134, 45. Youvan, D. C., Elder, J. T., Sandlin, D. E., Zsebo, K., Alder, D. P . , Panopoulos, N. J., Marrs, B. L., and Hearst, J. E. (1982). Journal of Molecular Biology 162, 17.

Zajic, J. E., Kosaric, N., and Brosseau, J. D. (1978). In “Microbial Processes” (T. K. Chose et al, eds.), pp. 57-109. Springer Verlag, Berlin. Zannoni, D., and Marrs, B. (1981). Biochimica et Biophysica Acta 637, 96. Zannoni, D., Baccarini-Malandri, A., Melandri, B. A . , Evans, E. H., Prince, R. C., and Crofts, A. R. (1974). FEBS Letters 48, 152. Zannoni, D., Melandri, B. A., and Baccarini-Melandri, A. (1976a). Biochimica et Biophysica Acta 423, 413. Zannoni, D., Melandri, B. A., and Baccarini-Melandri, A. (1976b). Biochimica et Biophysica Acta 449, 386. Zorin, N. A., and Gogotov, I. N. (1975). Biochemistry United Soviet Socialist Republic 40, 162. Zorin, N. A., and Gogotov, I. N. (1982). Biochemistry, United Soviet Socialist Republic 47, 690. Zumft, W. G., and Castillo, F. (1978). Archives of Microbiology 117, 53. Zumft, W. G., and Nordlund, S. (1981). FEBS Letters 127, 79. Zumft, W. G., Mortenson, L. E., and Palmer, G. (1974). European Journal of Biochemistry 46, 525. Ziirrer, H. (1982). Experientia 38, 64. Ziirrer, H., and Bachofen, R. (1979). Applied and Environmental Microbiology 37, 789.

Zurrer, H., and Bachofen, R. (1980). Experientia 36, 1166. Ziirrer, H., and Bachofen, R. (1981). Studies in Environmental Science 9, 31. Zurrer, H., and Bachofen, R. (1982). Biomass 2, 165. Note Added in Proof

The following relevant articles have appeared since the literature search for this review was completed: Albracht, S. P. J., Van der Zwaan, J. W., and Fontijn, R. D. (1984). Biochimica et Biophysics Acta 766, 245. Francou, N., and Vignais, P. M. (1984). Biotechnology Letters 6, 639. Hoover, T. R., and Ludden, P. W. (1984). Journal ofBacteriology 159, 400. Jouanneau, Y.,Lebecque, S., and Vignais, P. M. (1984). Archiues ofMicrobiology 139,326. Kanemoto, R . H., and Ludden, P. W. (1984). Journal ofBacteriology 158,713. Karyakin, A. A., Yaropolov, A. I., Zorin, N. A,, and Gogotov, 1. N. (1984). Biochemistry United Soviet Socialist Republic 49, 519. Madigan, M., Cox, S. S., and Stegeman, R. A. (1984).Journal ofGeneralMicrobiology 130, 1069. Nordlund, S., and Noren, A. (1984). Biochimica et Biophysica Acta 791, 21. Segers, L., and Verstraete, W. (1983). Biotechnology and Bioengineering 25, 2843. Tsygankov, A. A., Yakunin, A. F., and Gogotov, I. N. (1984). Mikrobiologiya 52, 419. Wall,J. D., and Braddock, K. (1984). Journul ofBacreriology 158, 404. Wall, J. D., Love, J., and Quinn, S. P. (1984). Journal of Bacteriology 159, 652. Yakunin, A. F., and Gogotov, I. N. (1984). FEMS Microbiology Letters 23, 217.

Biochemistry and Physiology of Bioluminescent Bacteria J. WOOD LAND HASTlNGS,* CATH ERINE J . POTRIKUS*, SUBHASH C. GUPTA,* MANFRED KURFURST,* and JOHN C. MAKEMSONt Department of Cellular and Developmental Biology, Harvard University, Cambridge, Massachusetts, U.S.A. t Department of Biological Sciences, Florida International University, Miami, Florida, U.S.A. I. Introduction . . . . , . . . . . 11. Taxonomy. . . . . . . . . . . 111. Biochemistry . . . . . . . . . A. Luciferase: an oxygenase . , . . . . . B. Luciferase intermediates and the reaction mechanism . . C. The luciferase Havin radical . . . . . . . D. The emitter . . . . . . . . . . E. FMN reductases: NAD(P)H dehydrogenase (FMN). . . F. The luciferase system as a multi-enzyme complex . . . G . Luciferase purification: inactivation by heat, urea, and proteases H. Primary sequence and subunit structure . . . . . . I. Subunit function, substrate binding, and the active centre J. Aldehyde biosynthesis . . . . . . . IV. Molecular biology . . . . . . . . V. Physiology. . . . . . . . . . . A. Continuous or pulsed emission: the flow of electrons . B . Inhibitors of bioluminescence in uiuo and their mechanisms . C. Chemotaxis and chemosensory behaviour . . . . D. Auto-induction, luciferase synthesis, and translation in uirro . E. Arginine requirement . . . . . . . F. Catabolite repression . . . . . . . G . Iron . . . . . . . . . . . H. Oxygen . . . . . . . . . .

. 236 . 237 ,

. . . . ,

. . . . ,

.

. . ,

. . . . . .

238 238 239 241 242 244 247 248 249 251 253 256 259 259 261 262 263 264 265 266 267

I Present address: Pfizer Central Research, Eastern Point Road, Groton, Connecticut, U.S.A.

236

J. WOODLAND HASTINGS

€7 AL.

I. Luciferase inactivation in vivo . . . . . . J . Suicide reactions: light-inducible protein, b-flavin. and p-flavin. VI. Ecology . . . . . A. Planktonic bacteria . . B. Host-associated bacteria . VII. Analytical and clinical applications . A. Applications in vitro B. Applications in vivo . . VIII. Acknowledgements . . References . . . . Note added in proof . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . . . .

,267 . 268 . 269 ,269 . 271 .274 ,274 ,276 ,280 . 281 . 291

I. Introduction This review embraces studies of the luminous bacteria published between 1979 and 1983, ranging from molecular biology and biochemistry to physiology and ecology. Some 1984 publications that were available to us in preprint form are also included. References to earlier work are made in most instances without citation; these may be readily located in earlier reviews (Ziegler and Baldwin, 198la; Hastings, 1978; Hastings and Nealson, 1977, 1981; Nealson and Hastings, 1979). Bioluminescent bacteria pose at a very fundamental level the question as to the function of light emission in biological systems (Hastings, 1982a). In a cell the size of a bacterium floating in the ocean, of what possible advantage could it be to emit light? As will be discussed, one possibility is that under such conditions the light-emitting system may not be functional, or may not even be present, a result of either physiological (biochemical) or genetic controls, respectively. Alternatively, considering the variety of habitats in which these bacteria are found, the light-emitting system may indeed play an important role in their ecology or physiology. For example, light produced by bacteria growing in association with faecal debris may enhance nutrient turnover by attracting coprophagous animals. In other cases, light emission itself could be superfluous; instead, the luciferase may be functionally important as a terminal oxidase under conditions where the cytochrome pathway is not operative (e.g., in the presence of low concentrations of oxygen or iron). The luciferase in bacteria, unlike that of any other luminous group (except, perhaps, the fungi), is related to the respiratory pathway, functioning as a shunt for electrons directly to oxygen at the level of reduced flavin. In formal terms, this luciferase is an external flavin mono-oxygenase or mixed-function oxidase; electrons for reduction of flavin mononucleotide (FMN) are provided by the reducing power derived from the electron-transport pathway. The light-emitting reaction then proceeds via the reaction of molecular oxygen with reduced flavin to form an interme-

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

237

diate luciferase-flavin peroxy species, whose breakdown provides energy (about 210 kJ mol-l) sufficient to leave one of the products in an electronically excited singlet state, with subsequent light emission. In fact, energy for bioluminescent reactions generally (possibly universally) is derived bond is replaced by two from peroxide breakdown, where a weak -0 strong bonds. The bacterial (luciferase-bound) peroxide chromophore, which has been isolated and characterized, provides a model in this respect for the otherwise highly diverse and different bioluminescent reactions (Hastings, 1983). Recent work has established that, in the purified Vibrio harueyi system, the excited state is a luciferase-flavin intermediate and that, in at least some species where the colour of the emitted light is different, emission occurs from a different chromophore, associated with a different protein. Cloning of the luciferase structural (lux) genes has allowed structure (sequence) determination of the two luciferase subunits. Cloning and mapping of other genes responsible for Lux phenotype will help elucidate the molecular aspects of the control of luminescence. A rapidly developing area is concerned with the use of bioluminescent systems for detecting specific substances (DeLuca and McElroy, 1981). The bacterial luciferase system is under development for such applications in research, clinical assays, and environmental monitoring.

11. Taxonomy

The taxonomic status of the marine luminous bacteria and many allied species has undergone considerable revision over the past decade or so, most radically within the last 3 years, as described in the comprehensive review by Baumann et al. (1983). In 1971, the genus Beneckea was redefined to include a number of Gram-negative marine species, including the cosmopolitan luminous B . hurveyi and B . splendidu. The remaining marine luminous species were retained in the genus Photobacterium, namely P . Jischeri, P . phosphoreum, P . leiognathi, and, later, P . logei (Harwood et al., 1980; Baumann and Baumann, 1981). However, on the basis of subsequent studies of the sequence homologies of glutamine synthetase, superoxide dismutase, and alkaline phosphatase (Baumann and Baumann, 1980; L. Baumann et al., 1980; Bang et al., 1981), the genus Beneckea, its constituent species, as well as P . Jischeri were assigned to the redefined genus Vibrio (P. Baumann et al., 1980). The major dichotomy between Vibrio and Photobacterium is also supported by analyses of differences in their cellular fatty acid compositions (Lambert et al., 1983). Unfortunately, these authors did not include V . fischeri in their study. Some freshwater and/or estuarine strains of V . cholerae (formerly V .

238

J. WOODLAND HASTINGS ET AL.

albensis) are well known to be luminescent (Desmarchelier and Reichelt, 1981; West et al., 1983; Hada et al., 1985). A newly characterized luminous species isolated off the coast of China has been named V . orientalis (Yang et al., 1983), whereas a newly characterized non-fermentative marine luminous species was assigned to the genus Altermonas and given the species designation A . hanedai (Jensen et al., 1980). The definition of genus Photobacterium was also modified somewhat, retaining the luminous phosphoreum and leiognathi species. Newly characterized psychrophilic strains, isolated from the subarctic (Primakova et al., 1982), await definitive taxonomic status. Photobacterium belozerskii has been reassigned to Vibrio (Beneckea)harueyi (Primakova et al., 1983). To summarize, the major changes in the marine species are that Beneckea harueyi is now Vibrio harveyi, and Photobacterium jischeri is now Vibrio Jischeri. In this review these newer designations will be substituted and used throughout. In 1976, Poinar described the insect-parasitic nematode Heterorhabditis bacteriophora and noted that its bacterial symbionts were bioluminescent. Studies since then have established this group as the first welldocumented case of a terrestrial bioluminescent bacterium. They possess a luciferase system with properties and requirements similar to those of the marine species (Poinar et al., 1980) and produce antibiotic substances believed to inhibit putrefaction of the dead host (Paul et al., 1981). Although the evolutionary status of the terrestrial bacteria in relation to the marine forms has not been clarified, they were assigned to the new genus Xenorhabdus in the family Enterobacteriaceae (Thomas and Poinar, 1979, 1983; Akhurst, 1983). There are two species; the luminous one is X . luminescens. Finally, species-related kinetic differences of isolated luciferases in the in uitro reaction (Hastings and Nealson, 1977) have been further analysed and used to estimate species composition in natural populations (Vorobyeva et al., 1982).

111. Biochemistry A.

ICIFERASE: AN OXYGE lASE

The postulate that bacterial luciferase is a mono-oxygenase (Hastings, 1982b) in which one atom of molecular oxygen appears in water and the second is incorporated into the long-chain acid product has received additional experimental confirmation. In the P . phosphoreum luciferase reac-

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

239

tion, Suzuki et al. (1983) found that one atom of molecular oxygen was incorporated into the product lauric acid; the amount of acid formed was proportional to the amount of light emitted in the reaction. The balanced overall equation for the reaction is thus: FMNH2 + 0 2 + RCHO RCOOH + FMN + HzO + (c)hu (1)

-

where c is the bioluminescent quantum yield. Other evidence indicates that flavin hydroxide (FMNH-OH) is the other initial chemical product and that it promptly breaks down to give FMN and water (Kurfurst et al., 1984).

B . LUCIFERASE INTERMEDIATES AND THE REACTION MECHANISM

The major steps and several intermediates proposed for the bacterial luciferase reaction are illustrated in Fig. 1. An early step involves the rapid reaction of enzyme-bound ( 1 : 1) FMNH2 with molecular oxygen to give the key oxygenated intermediate, the structure of which has been shown to be the luciferase dihydro-4a-peroxy FMN (Lhoste et al., 1980).

““z

yE,-FMNH-OOH

* El-FMN + H2%

I

FIG. 1. Proposed steps in the catalytic cycle of bacterial luciferase. The reaction leading to light emission starts with formation of FMNHz from FMN and NAD(P)H; this is catalysed by FMN reductase (E2). Luciferase-(E,)-bound FMNH2 reacts with molecular oxygen to form the intermediate flavin 4a-peroxide, which then reacts via several steps with a longchain aldehyde to form an excited species (designated as the flavin 4a-hydroxide). The fatty acid which is formed is then released, and reconverted into aldehyde by a third enzyme (E,:myristic acid reductase). Formation of the blue radical, along with its reaction with superoxide ion, is shown at the top. Modified from Kurfurst et a / . (1982~).

240

J. WOODLAND HASTINGS ET AL.

This intermediate may also be formed by the reaction of the luciferase flavin radical with superoxide ion (Kurfurst et al., 1983) or by the reaction of FMN and hydrogen peroxide with luciferase (Watanabe and Nakamura, 1976; Hastings et al., 1979). The luciferase peroxy flavin is relatively stable; at 20°C it has a lifetime of perhaps 20 seconds, and at 0°C about 20 minutes, these values being dependent on the specific luciferase as well as on ionic strength and pH value. Its stability may be greatly enhanced by compounds such as longchain alcohols which inhibit its activity by binding, without reacting, at the aldehyde site (Tu, 1979). Certain hydrophobic compounds, such as camphor, appear to inhibit bioluminescence in vitro by binding competitively at the aldehyde site. Some authors have proposed that hydroxylation of such compounds occurs, involving the participation of cytochrome P-450 believed by them to be an essential component of the luciferase complex and to be specifically responsible for aldehyde binding (Danilov et al., 1982; Ismailov et al., 1981b). But luciferase itself has an aldehyde-binding site and, although it will bind (and be inhibited by) various other compounds, none has been found to be chemically altered in the luciferase reaction. Watanabe et al. (1980) studied the affinities and bioluminescent activities of various 8-substituted reduced flavin mononucleotides for luciferase from P. phosphoreum. The affinity of the reduced flavin for luciferase increases with a decrease of electron-donating power of the substituent at C-8 on the isoalloxazine ring, but no evidence was obtained for covalent attachment of flavin to luciferase in the reactions. Calculations suggested that the imino group at position 5 in the isoalloxazine ring contributes to binding. The inactivity of N-5-substituted flavin (see Bruice, 1982) is consistent with this. They concluded that, in general, the yield of peroxy intermediate formed with these analogues was low, possibly due to abortive side reactions, and that the affinity of a reduced flavin for luciferase is not the only determining factor for formation of the peroxy intermediate. Tu (1982) reached some similar conclusions in his study of the properties of flavin peroxides formed in the reaction of luciferase from V . harveyi with several different reduced flavin analogues as substrates. At 0"C, all peroxides were stabilized by long-chain alcohols, notably dodecanol; half lifetimes were extended by factors ranging from 3 to nearly 500. The absorption spectra of intermediates having flavins with side-chain alterations were essentially the same as the peroxy FMN whereas those with ring alterations were distinctly different, as is true also for the corresponding oxidized flavins. In all cases, decay of the intermediate in the absence of aldehyde yielded equimolar quantities of the oxidized flavin

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

241

and hydrogen peroxide. The isolated intermediates reacted with decanal to give bioluminescence with different quantum yields, lower than that observed with FMN, but higher than those recorded for the corresponding reactions starting with luciferase-bound reduced flavins as such. Thus, there is a pathway for oxidation of bound reduced flavins not involving an intermediate 4a-hydroperoxyflavin. A mechanism for the reaction of the flavin peroxy intermediate with aldehyde and formation of an excited species was proposed by Eberhard and Hastings to involve a Baeyer-Villiger type reaction (see Hastings and Nealson, 1977; Hastings and Tu, 1981). In Fig. 1, formation of a peroxy hemiacetal is followed by the formation of the flavin 4a-hydroxide in an excited state (Kurfurst et al., 1984). A Baeyer-Villiger mechanism has also been proposed for cyclohexanone mono-oxygenase (Ryerson et al., 1982; Schwab et al., 1983). In these reactions, however, no light appears to be emitted. Several alternative reaction mechanisms, including the Kosower (1980) scheme involving dissociative electron transfer, are discussed by Ziegler and Baldwin (1981a). A new proposal for the mechanism of the luminescent reaction that involves electron exchange is given in Section D (see also Fig. 3).

C . THE LUCIFERASE FLAVIN RADICAL

Presswood and Hastings (1979) reported that the reaction of luciferase with FMNH2 and molecular oxygen resulted in absorption in the 600 nm region, which had not previously been noted. Kinetic data suggested that this red-absorbing (thus blue) species was related to the light-emitting pathway, but this was later found to be incorrect (Hastings et al., 1981; Kurfurst et al., 1982b). This blue species was shown to be a luciferasebound neutral flavin semiquinone radical formed in the reaction mixture by oxidation of enzyme-bound reduced flavin, or anaerobically by reaction of FMN and FMNH2 in the presence of luciferase. The radical is kinetically stable (t1,220 hours at 0°C in air; the Arrhenius AHdecay being about 170 kJ mol-' and can be prepared in pure form by Sephadex G-25 chromatography at 0-4°C; (Kurfiirst et al., 1982~). The enzyme-bound radical is inactive for light emission either with or without aldehyde, but can nevertheless react with the superoxide radical to give light (Kurfiirst et al., 1983). The recombination step itself does not result in emission; aldehyde is required. The species formed in the recombination is deduced to be luciferase 4a-peroxy FMN, the intermediate in the normal luciferase reaction. Although this intermediate can undergo a

242

J. WOODLAND HASTINGS €T AL.

reversible homolytic dissociation to yield free superoxide and the corresponding luciferase radical (Fig. I), the slowness of these steps precludes involvement of these pathways in the normal bioluminescent reaction.

D. THE EMITTER

In a high quantum yield chemi- or bioluminescence reaction, a singlet electronically excited state is generated which, being in the ground state subsequent to its emission, should (in principle) be identifiable by its fluorescence (Hastings and Tu, 1981; Shimomura, 1982; Wilson, 1985). Oxidized FMN, a product in the reaction, has fluorescence centred at about 525 nm, but this does not correspond to the bioluminescence in uitro, which is maximal at about 490 nm. When reduced flavin analogues were used in the reaction in place of FMNH2, the colour of the bioluminescence was found to be altered, although again it did not correspond to the fluorescence emission of the analogues used (Mitchell and Hastings, 1969). These observations could be explained if the emitter is an unstable intermediate flavin species which decays to a stable product and is thus difficult to detect. Similar observations by Matheson et al. (1981) were interpreted quite differently. However, an intermediate with such properties was recently observed in the reaction of purified luciferase flavin hydroperoxide with aldehyde at 1°C (Kurfiirst et al., 1984). Simultaneous kinetic measurements of absorption and bioluminescence showed that the decay of light emission occurred more rapidly than the appearance of FMN, indicative of such a transient intermediate species subsequent to light emission. Its absorption (A,, 360 nm) and fluorescence emission (A,, 490 nm; Fig. 2) are consistent with the postulate that it is the luciferase-bound flavin 4a-hydroxide (Fig. 1). A transient fluorescence answering this description was also noted by Matheson and Lee (1983), but was also interpreted in a very different way. In some species and strains of luminous bacteria the colour of the light emission in uiuo matches that in v i m . But in other cases the spectra are different, even though the same luciferase reaction and the same flavin intermediates appear to be involved. The two documented examples are in P . phosphoreum (Gast and Lee, 1978) where blue light (A,, 475 nm) is 535 nm; emitted and a strain of V .jischeri which emits yellow light (A,, Ruby and Nealson, 1977). The blue emission is postulated to involve a chromophore identified as 6,7-dimethyl-8-ribityllumazine (Koka and Lee, 1979), bound (1 : I) to a

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

400

450

500

550

243

600

Wavelength (nm)

FIG. 2. Fluorescence emission of the putative emitter in the bacterial bioluminescence reaction, and its decay to FMN with time. Curve 0 shows the emission spectrum (excitation at 380 nm) taken at 9°C immediately after completion of bioluminescence, with a bimodal shape due to the presence of both the putative hydroxy-FMN and its breakdown product, FMN, the latter increasing with time (3, 7, and 15 minutes, as indicated). A final spectrum (dashed line) was recorded also at 9°C after warming to 25°C. The corrected fluorescence emission spectrum for the intermediate ( 0 )was calculated by subtracting the fluorescence attributable to FMN and applying fluorimeter corrections. After Kurfiirst et a / . (1984).

specific “blue fluorescent” protein having a molecular weight of about 20,000 (Small et al., 1980). Other studies have reported on the associations between the ligand and the lumazine protein and between the lumazine protein and luciferase (Irwin et al., 1980; Visser and Lee, 1980, 1982), including luciferases from different species of luminous bacteria (Lee, 1982). Evidence for involvement of the lumazine protein in emission is convincing, but the mechanism whereby the excited state is formed is less clear. Ward (1979, 1981) believed that a classical Forster-type energy transfer from a luciferase-bound flavin excited state would be unlikely, both for energetic reasons and because lumazine apparently participates in the primary reaction as judged by its stimulation of the rate of luminescence. Both Hart and Cormier (1979) and Ziegler and Baldwin (1981a) suggested that a higher energy primary excited state might be formed in the reaction which could then transfer energy (by the Forster mechanism) either to the flavin involved in the luciferase reaction or to a separate chromophore. This would resolve the first concern but not the second. The striking yellow strain of V .jischeri possesses an apparently analogous “yellow-fluorescent’’ protein (Leisman and Nealson, 1982) whose prosthetic group is a flavin, probably FMN (J.W. Hastings, G. Leisman, and K. Nealson, unpublished observations). The protein, purified to homogeneity, has a molecular weight of about 22,000 (Leisman and Nealson, 1985). With the chromophore attached, its fluorescence emis-

J . WOODLAND HASTINGS E r AL.

244

sion matches the bioluminescence emission, indicating that the yellow fluorescent protein is the in uiuo emitter. The authors discuss the results in terms of Forster-type energy transfer. Lee and his co-workers have proposed a model for the steps leading to emission (and the emitter itself) which rejects the postulate that the flavin substrate in the reaction may be formed as a product directly in the excited state, and thus that it could have a role as a primary emitter (Matheson and Lee, 1983). Instead they postulate the intermediacy of unidentified “energy rich” or “chemically energized” species as precursors to the emitting states. The latter are specified as “acceptors” which are formed in the excited state in the process of decompositon of the former. In the absence of evident secondary fluorophores, unidentified endogenous substances are assigned the role of emitters. Some confusion is caused by the earlier proposal that flavins (both reduced and oxidized) are involved as emitters with certain flavin analogues (Matheson et al., 1981), but flavin participation is not included in the 1983 scheme. Indeed, efficient bioluminescence was reported from luciferase starting with reduced lumichrome (Matheson and Lee, 1981); no further documentation or confirmation of this has appeared. We suggest a unitary mechanism in which the primary excited state could be either the flavin or, alternatively, an accessory chromophore; the luciferase hydroxy flavin would be formed as a transient product in both cases (Fig. 3). No classical Forster-type energy transfer is involved. In this mechanism, which is based on the chemically induced electron exchange mechanism proposed by Koo et al. (1978), the steps prior to excited state formation involve the transfer of an electron from a donor to an acceptor. In one of the two cases illustrated, the electron donor would be the flavin, thus the transfer would be intramolecular. In the other, which applies to systems with an accessory protein-bound chromophore, intermolecular electron transfer would occur from the chromophore as the donor. The acceptor would be the same in the two cases, namely, the peroxyhemiacetal function of the molecule; the weak oxygen-oxygen bond would then cleave to form a long-chain acid and the radical anion of the hydroxyflavin. This radical anion should be a stronger reductant, so that charge annihilation returning the electron to the donor-be it flavin or accessory chromophore-could form an excited state directly. E.

FMN

REDUCTASES:

NAD(P)H

DEHYDROGENASE

(FMN)

The luciferase reaction may be driven by coupling it to any system that produces FMNH2. In extracts of luminous bacteria and related species, as well as other bacteria (e.g., Pseudornonasputida; J.W. Hastings and I.C. Gunsalus, unpublished observations), there occur FMN reductases, or

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

245

R I

H

A

0

I RCHOH

E-FHOO-CHOH-R

E-

1 1 1-

E-FHOO-CHOH-R

f FHOTO-CHOH-R

E-FHOTO-CHOH-R

E- ? FHOH T + RCOOH E-FHOH*

E-FHOH

1

RCOOH

+/ti

E-F + H 2 0

Ib)

1+

E-FHOH

E-FHOH

E-F

1

’1

T

+H20

)\

‘Pr-R*

ill. + Pr-B

(CI

FIG. 3. Two proposed alternatives for excited-state production in the breakdown of the postulated luciferase-flavin peroxyhemiacetal (E-FHOO-CHOH-R) intermediate (a). In the system with only the peroxy intermediate (b), an intramolecular electron-transfer occurs which results in the cleavage of the Mbond, formation of acid, and a diradical; charge annihilation then results in the formation of the excited state of the hydroxyflavin. In the intermolecular electron-transfer system (c) an accessory protein-bound chromophore (Pr-B) is the donor and, following ( t o bond cleavage and acid production, it is this chromophore that is excited upon charge annihilation. In both cases ground-state luciferasebound hydroxyflavin is formed as a product.

NAD(P)H dehydrogenase (FMN) (EC 7.6.8.1) that catalyse such a reaction: NAD(P)H + H+ + FMN --z NAD(P)+ + FMNHz (2) The enzyme is also often referred to as NAD(P)H:FMN oxidoreductase. Bioluminescence may thus be stimulated in crude extracts on addition of reduced nicotinamide nucleotide in the presence of FMN and long-chain aldehyde. Such enzymes have been purified and characterized from V. harveyi and V. Jischeri, and isolated from P. phosphoreum and P . leiognathi; differences in molecular weights and kinetic mechanisms have been reported. In V. harveyi, two oxidoreductases, one specific for NADH and the other for NADPH, were reported. Recently a third type,

246

J. WOODLAND HASTINGS H A L .

able to utilize both NADH and NADPH, was found (Watanabe and Hastings, 1982). The FMN binding sites for the first two enzymes are quite similar, involving the isoalloxazine ring and a hydrophobic binding site, but differ with regard to specific participation of the two methyl groups (Nefsky and DeLuca, 1982). These authors further demonstrated that the NADH-specific enzyme exhibits an ordered single displacement mechanism with NADH binding first, whereas the NADPH enzyme exhibits ping-pong kinetics. Vibrio Jischeri appears to have only one oxidoreductase that is able to utilize either NADH or NADPH, exhibiting ping-pong kinetics (Tu et al., 1979). Attention has been directed to the question of whether or not the pathway catalysed by the oxidoreductases corresponds to the in uiuo pathway of electrons going via luciferase. As yet the question has not been resolved, but the evidence supporting the involvement of any of the isolated FMN reductases in uivo is not convincing. A key observation made some years ago is that, when substrates are saturating, the initial maximum rate (intensity) of the reaction is proportional to the product of the concentrations of reductase and luciferase, indicating that the two enzymes function independently. Free FMNH2, subject to autoxidation, would be produced under such conditions. Tu and Hastings (1980) found the coupling efficiency to be enhanced (electrons were transferred with less loss) with a system in which the luciferase was covalently linked to Sepharose. They suggested a model in which, when one of the enzymes was (membrane) bound, its affinity in uiuo for the second was greater. An even more direct approach to physical coupling has been taken by Ford and DeLuca (1981) and Wienhausen et af. (1982). These investigators co-immobilized the FMN reductase and luciferase by covalent attachment to Sepharose 4B and observed a more effkient conversion of reduced nicotinamide nucleotide into light; they postulated that the FMNH2formed in the co-immobilized system is utilized more effectively by luciferase with less loss to autoxidation. More recently, DeLuca and Kricka (1983) extended this approach with two sets of sequentially acting enzymes, one with five and a second with eleven co-immobilized enzymic species coupled to luciferase and oxidoreductase. This latter approach, although convincing as a model for the functional importance of physical localization of sequential enzymes, says nothing about the actual identity of the system feeding electrons to the bacterial luciferase. According to the authors’ interpretation, any enzyme able to produce FMNH2 would give the same result if co-immobilized with luciferase. Specific interactions between enzymes, as shown by binding induced by immobilization, effects on thermal stability, or effects on substrate binding constants, are more relevant but still inadequate to identify

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

247

the pathway whereby FMN is reduced. In fact, based on their study of NADPH- and ATP-dependent luminescence in bacterial extracts, Vysotskii et al. (1982a) concluded that the functioning of the luminescent system may not be directly connected with FMN reductase. Consistent with this view is the fact that none of these various reductases appears to be subject to co-induction with luciferase, and that most appear to function catalytically far faster than the luciferase to which they might be coupled. How electron flow via luciferase might be regulated in uiuo is unknown; Watanabe and Nakamura (1980a,b) showed that the efficiency of coupling of luciferase to electron flow can vary under different conditions.

F. THE LUCIFERASE SYSTEM AS A MULTI-ENZYME COMPLEX

A quite different and highly iconoclastic proposal for the bacterial luminescence reaction and its function was put forward by Danilov (1979) and his colleagues (Ismailov et al., 1979). The complete system, as studied by them in V. jischeri, was proposed to be a multi-enzyme complex involving, in addition to luciferase and reductase systems, cytochrome P-450 and lumiredoxin, the latter being a specific iron-sulphur protein. Purification resulted in a preparation that was either heterogeneous or lost activity if purified further (Shumikhin ef al., 1980b), results that they considered to be evidence for existence of the complex. In their model, nicotinamide nucleotide, flavin and aldehyde are believed to bind to the dehydrogenase, luciferase, and P-450 components, respectively, whereas a role for the lumiredoxin was not suggested. The multi-enzyme complex is considered to be a general hydroxylating system, functionally similar to the microsomal system and important not so much in the generation of light quanta as in protection of the cell from hydrophobic compounds, for example camphor, by their hydroxylation (Danilov e f al., 1982). Light emission is inhibited by many drugs such as dimethylaniline and hexobarbital, which are competitive with aldehyde; hydroxylation of certain compounds (aldehydes) by luciferase is distinctive in emitting light, whereas competing substrates do not produce luminescence (Ismailov et al., 1981b). The activity of luciferase genes cloned in Escherichia coli (see Section IV) suggests (but does not prove) that light emission is not necessarily dependent on lumiredoxin and cytochrome P-450.Furthermore, although cytochromes have been reported and studied in luminous bacteria (e.g., Watanabe e f af., 1979; Ismailov et al., 1980; Baranova ef al., 1980), neither they nor the two other components of the proposed complex have

248

J. WOODLAND HASTINGS ET AL.

been found by other workers to be structurally or functionally associated with the light-emitting activity of luciferase, and no active high-molecularweight (greater than 80,000) complexes have been demonstrated. An exception to this is the interesting, but still unconfirmed, report by Balakrishnan and Langerman (1977) of a membrane-bound glycoprotein luciferase from P . leiognathi; Schumikhin et al. (1980a) have also reported the isolation of a membrane-bound luciferase fraction accounting for 20-25% of the activity in the cell-free system from V . jischeri. How this might relate to the four-protein complex is unclear.

G. LUCIFERASE PURIFICATION: INACTIVATION BY HEAT, UREA,

A N D PROTEASES

Luciferase may comprise 5% or more of the soluble protein in fully induced cells. Satisfactory purification in high yield and good purity has been obtained by conventional procedures (Hastings et al., 1978). A covalently immobilized form of an inhibitor of luciferase was recently found to be an effective resin for purifying the enzyme from several different bacterial species (Holzman and Baldwin, 1982). Starting with crude ammonium sulphate precipitates, the procedure resulted in nearly homogeneous preparations with yields greater than 50% of the activity applied. The specific temperature at which denaturation of wild-type luciferases occurs may differ depending on the species from which it was obtained (Ruby and Hastings, 1980). In addition to FMN, bovine serum albumen, and reducing agents, a high concentration of phosphate (0.5 M) was found to stabilize luciferase from V . harueyi against both thermal and urea denaturation (Baldwin and Riley, 1980). Petushkov et al. (1982) found that with luciferase from P . leiognathi, tetradecanal, ethylenediamine tetraacetate (EDTA), dithiothreitol, and bovine serum albumin, but not FMN, were effective in this regard. Baldwin and co-workers have shown that on limited proteolysis, for example by trypsin or chymotrypsin, luciferase from V . harueyi is inactivated. Loss of activity is accompanied by hydrolysis of one or a small number of peptide bonds within a “protease labile region” of the a subunit, resulting in generation of a 28,000- to 30,000-Da species plus many small peptides (Baldwin et al., 1979b; Holzman and Baldwin, 1980a; Rausch et al., 1982). By contrast, the /3 subunit apparently remains unaffected. Inactivation occurs even though all or most of the resulting a fragments remain associated with the dimer, as judged by sedimentation analyses under non-denaturing conditions. From the number and sizes of the fragments generated, the protease-labile region has been estimated to

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

249

be located approximately 100 residues from one of the termini of the a subunit, and to span about 20 residues, possessing five or six trypsin- and two chymotrypsin-sensitive sites (Holzman et af., 1980; Ziegler and Baldwin, 1981b). There is substantial evidence indicating that this protease-labile region is part of (or related to) the active centre (Baldwin et al., 1981; Dougherty et af., 1982). Hydrolysis of peptide bonds in this region results in loss of activity and concomitant loss of FMN and FMNH2 bindings, whereas both FMN and orthophosphate protect the region from proteolysis (Holzman and Baldwin, 1980b). A mutant altered in the active centre has a significantly altered sensitivity to protease, and a reactive active centre (a subunit) thiol group has been shown to be within the protease-labile region. Ruby and Hastings (1979) demonstrated that luciferases isolated from three other species were similarly inactivated but were even more sensitive to the protease action ( P . leiognathi > P . phosphoreum > V .jscheri > V . harueyi). Only the a subunit was attacked and the p subunit of each retained its ability to form active luciferase when renatured with the corresponding native a subunit. Holzman and Baldwin (1980a,b) reported that the p subunits of V. jscheri and P . phosphoreum, unlike those in V. harueyi, are sensitive to proteases in low, but not high, phosphate concentrations.

H . PRIMARY SEQUENCE A N D SUBUNIT STRUCTURE

Cohn et al. (1984) have reported the primary amino acid sequence of the a subunit of V. harueyi luciferase (Fig. 4), as deduced from the nucleotide sequence of the luxA gene (see Section IV). Structurally, luciferases from the different bacterial species are all heterodimers (80 kDa) and the two luciferase subunits are clearly homologous (Baldwin et al., 1979a; Cohn et af., 1983). However, the active heterodimeric complex possesses but a single active centre, mostly, but not exclusively, associated with the heavier a chain. An earlier postulate that the reaction requires two reduced flavin molecules for each turnover remains unsubstantiated (see Hastings and Nealson, 1977). There are no confirmed reports that bacterial luciferase contains any metals, prosthetic groups, non-amino acid residues, or disulphide bonds. Some enzymes exhibit subunit exchange in buffer. Luciferase does not; formation of hybrid luciferases generally requires denaturation, separation and renaturation steps. Mutant luciferases with lesions in the two different subunits have, however, been shown to complement under nondenaturating conditions to produce wild-type luciferase (Anderson et al.,

A

s

n

m

T

h

r T y r G l n r o P r o GIU

L y c A s p T h r A s n A r g A r g -ASP

T y r Ser T y r G l u

Asn Pro1

L y 6 G l u LYS Gln ]

FIG. 4. The complete amino acid sequence of the a subunit of V. harveyi luciferase. In the sequence depicted here, deduced from the nucleotide sequence of the gene (Cohn ef al., 1985), the amino acids on a white background are those that occur most often on the surface of a protein, while those on a black background occur more often inside globular proteins (Swanson, 1984). The regions terminating with arrows indicate p sheet conformation; the helical depiction is for a helix; the remaining regions are random. All of these assignments are based on the secondary structural predictions of Chow and Fasman (1974).

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

251

1980). Thus, the subunit affinities must be lower. The specific activities of the two mutant enzymes were determined by the same authors using specific antiserum against wild-type luciferase: the a subunit mutant (AK-6) enzyme had a specific activity 500-1000 times less than that of the wild type, whereas, the p subunit mutant (flavin binding, FB-1) was about 20-fold less active.

I . SUBUNIT FUNCTION, SUBSTRATE BINDING, AND THE ACTIVE CENTRE

Specific subunit functions were investigated by examining the properties of hybrids of four different species (Ruby and Hastings, 1980). No active hybrids were formed between subunits from V . harueyi and those from any of the other species, nor between the p subunit of V.Jischeri and the LY subunit of P. phosphoreum. All other combinations of LY (heavy) and p (light) subunits yielded activity; the hybrids exhibited in uirro catalytic characteristics most like those of the parent luciferase from which the heavy subunit was derived; the light subunit of luciferase from P. leiognathi (a more thermally stable luciferase) conferred an increased thermal stability to its hybrids. These results suggest specific subunit functions but surely do not guarantee a sole or exclusive role for either subunit. Indeed, there is now good evidence that the p subunit contributes to the active centre. Among the mutant luciferases which Cline isolated by screening for defects in binding of reduced flavin, one was found to have its lesion in the p subunit (Anderson et al., 1980). Also, and in contrast to the results of Ruby and Hastings (1980), Meighen and Bartlet (1980) succeeded in preparing an active hybrid luciferase of V. harueyi ( a )and P . phosphoreum (p) which exhibited kinetics characteristic of the a subunit and flavin-binding characteristic of the p subunit. Although Holzman and Baldwin (1983) suggested an alternative interpretation of these results, further indication of involvement of the p subunit comes from the work of Welches and Baldwin (1981). They found that luciferase was completely inactivated following reaction of fluorodinitrobenzene with an amino group on either the CY or p subunit, and that modification of either one of these leads to complete inactivation of the enzyme. Following such an inactivation, the enzyme had lost all measurable FMNH2binding, again indicative of a role in binding for the p subunit. It may be noted that Kratasyuk and Fish (1980) reported that dinitrofluorobenzene inhibits bioluminescence in uif r o without modification of protein. They suggested that the mechanism involved competitive inhibition of the active site for FMNH2.

252

J. WOODLAND HASTINGS E T AL.

The importance of exposed amino groups for FMNHz binding also emerged from a study of immobilized luciferase (Watanabe et al., 1982a). Covalent attachment of luciferase to Sepharose via exposed amino groups decreased the FMNH2 binding affinity, but did not affect the kinetics of the subsequent catalytic steps leading to light emission. Luciferase linked to Sepharose via the (Y subunit had its FMNH2 binding affinity lowered more than the enzyme attached via the p subunit. It is not known whether these are the same amino groups as those reported by Welches and Baldwin (1981) where luciferase activity was completely blocked. Watanabe et al. (1982a) found that, when immobilized, neither the (Y nor the /3 subunit alone exhibited luciferase activity, but both preparations regained activity upon renaturation in the presence of the second subunit. Thus, an individual subunit does not appear to possess an active centre. A quite different approach towards the study of quaternary structure and subunit function has been reported by McCarron and Tu (1983). FMN-sensitized photo-inactivation of luciferase (possibly involving singlet oxygen) was shown to be accompanied by formation of cross-linked species, but the two processes are not coupled and the kinetics are different. Cross-linked species could be useful in probing the nature of the active site. Tu and Henkin (1983) reported photo-affinity labelling of bacterial luciferase utilizing l-diazo-2-oxoundecane. This molecule binds at the aldehyde-binding site, competitive with decanal, and irradiation results in labelling of both (Y and p subunits with concomitant enzyme inactivation. This labelling of both subunits was interpreted as an indication that the aldehyde-binding site is, like the FMNH2 binding site, at or near the interface of the two luciferase subunits. The interacting, possibly overlapping, domains of aldehyde and reduced flavin binding are further indicated by the recent demonstration that 2,2-diphenylpropylamine,an inhibitor competitive with aldehyde, when bound to luciferase from V. harveyi, causes an increase in the affinity of the enzyme for reduced flavin (Holzman and Baldwin, 1981). Likewise, the FMNH2-luciferase complex has a greater affinity for the inhibitor than does the enzyme alone. The phosphate anion, which binds at the flavin site, has a similar effect on the binding of decanal (Holzman and Baldwin, 1983). In this latter study it was shown that the previously described inhibition of the reaction by aldehyde (which occurs with luciferase from V. harueyi but not that from V. Jischeri)is chain-length dependent and reversible. In the normal reaction, the aldehyde-Iuciferase stoicheiometry was shown to be 1 : 1 , and the inhibition was attributed to an inactive complex of luciferase having two aldehyde molecules bound. The fact that excess aldehyde does not inhibit the reaction if added after

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

253

binding of FMNH2 suggests that the second aldehyde molecule may bind at a site overlapping with the flavin-binding site; the observation that high concentrations of phosphate lower the affinity for the second aldehyde is consistent with this possibility. A model proposed by these authors, in which binding of aldehyde and reduced flavin are not ordered, is consistent with all previous observations, but remains to be verified. Baumstark et al. (1979) studied the kinetics of reactions utilizing two aldehydes of different chain lengths, these being individually distinctive kinetically. The binding steps of octanal and decanal were found to be readily reversible, whereas that of dodecanal was not. This was confirmed by showing that the alcohol decanol, an inhibitor competitive with aldehyde, displaces decanal in the course of a non-turnover reaction. Stabilization of the luciferase-peroxyflavin intermediate by various compounds that inhibit the reaction and bind at the aldehyde site (e.g., long-chain aliphatic compounds, including alcohols) was investigated by Tu (1979) with luciferase from V. harueyi. He found dodecanol to be especially effective and demonstrated that the technique facilitates the study of this (and possibly other) luciferase intermediate; at 0°C the intermediate has a lifetime measured in many hours instead of many minutes, and its absorption and fluorescence were the same as, or closely similar to, those of the unstabilized intermediate. Nakamura (1982) has studied the interaction of luciferase with fatty acids of various chain lengths (Cloc 2 4 ) and found that binding affinity increased with chain length up to ($4. The existence of an active-centre thiol whose alkylation completely inactivates luciferase is well known (Ziegler and Baldwin, 1981a). Inactivation studies with long-chain N-alkylmaleimides had suggested that the active centre must have a hydrophobic character, this being expected for an enzyme with a long-chain aliphatic substrate. Studies with spin-labelled maleimides indicated that the cysteine lies in a hydrophobic cleft at least 1.7 nm (17 A) in length (Merritt and Baldwin, 1980).

J . ALDEHYDE BIOSYNTHESIS

The chain length of the natural aldehyde and its biosynthetic pathway in V. harueyi was suggested by analysis in uiuo of different types of mutants requiring exogenous aldehyde for bioluminescence (Ulitzur and Hastings, 1979b,c). Some of these mutants also emit light when myristic (tetradecanoic) acid is added; these are presumed to be blocked in a step prior to fatty acid formation [see Eq. (3)], and the stimulation by myristic acid is attributed to its conversion into aldehyde by myristic acid reductase, which also functions in recycling the fatty acid produced in the lumines-

254

J. WOODLAND HASTINGS ET AL.

cent reaction. This response is quite specific for the C14 compound, thereby conferring chain-length specificity to the recycling reaction, which is different from the chain-length specificity for luciferase alone. In such mutants, the amount of light obtained with small amounts of either tetradecanal or tetradecanoic acid (but not other chain lengths) may be increased by as much as 60-fold by cyanide or other agents that block respiration. This inhibition of respiration is postulated to increase the reducing power in the cell and thereby to stimulate recycling and the amount of light emission. In the other class of mutants, stimulation of bioluminescence occurs only with added aldehydes, indicating that the lesion affects the activity of the putative myristic acid reductase. In such mutants, chain length-specific stimulation by cyanide does not occur. Myristic acid reductase activity in V. harueyi was shown to be co-induced with luciferase. This is consistent with the recent demonstration in V. Jischeri that it maps on the same operon as the structural genes for luciferase (Engebrecht et al., 1983). Although attempts to demonstrate such an enzyme system in extracts of V. harueyi were (and until recently remained; Wall et al., 1984a) negative, Meighen (1979) isolated it from P . phosphoreum. In crude extracts, light emission following addition of FMNH2 was stimulated after incubation with ATP and NADPH, and even more so if myristic acid was also present. The enzyme was shown to be quite specific for myristic acid and to require stabilization by a reducing agent; it was partially purified and separated from luciferase as a higher molecular weight activity by gel filtration (Riendeau and Meighen, 1979). The fatty acid reductase, measured by a luciferase-independent in uitro assay, was shown to be coinduced with luciferase, and aldehyde was shown to be the only product at early times of assay (Riendeau and Meighen, 1980, 1981). A similar enzyme activity was reported in extracts of P . leiognathi (Ulitzur and Hastings, 1980; Vysotskii et al., 1982a), where its involvement in the recycling of aldehyde in uiuo was also suggested. In V . harueyi, Byers and Meighen (1984) have implicated a previously described aldehyde dehydrogenase in aldehyde biosynthesis. Proteins involved in production of aldehyde for the bioluminescent reaction were identified by their acylation (Wall el al., 1984b). The activity of such an enzyme should also account for interesting observations of Vysotskii et al. (1982a). They prepared highly concentrated (1 g of cells lysed in 5 g of water) cell-free extracts of P . leiognathi and found that luminescence occurred and continued for several hours without additions. Emission could also be stimulated further (or, in decayed extracts, initiated) simply by adding ATP and NADPH; this may

BIOCHEMISTRY

OF BIOLUMINESCENT BACTERIA

255

be attributed to recycling of endogenous long-chain acid to the active aldehyde and the reduction to FMN. In this work they extracted and chromatographed the active aldehyde factor from the brightly glowing supernatant and found that the active component was not myristic, lauric, or decyl aldehyde (Vysotskii et a/.,1981c, 1982b). This appears to conflict with the results with mutants and fatty acid reductases (see Section V), and also the earlier isolation of aldehyde by Shimomura and colleagues. Meighen and colleagues have shown that the activity responsible for reduction of long-chain (myristic) acid to aldehyde involves a fatty acid reductase enzyme complex involving at least two steps and three components with different molecular weights, these being designated according to approximate size as the 34,000,50,000, and 58,000 species (Riendeau et al., 1982; Rodriguez et al., 1983a). The reaction itself has been resolved into two steps. The first activity (50,000), designated as an acyl-protein synthetase, results in formation of an acylated protein by reaction with ATP and myristic acid (Rodriguez et al., 1983b). This then reacts with the 58,000 molecular-weight enzyme (acyl-CoA reductase) and NADPH to form myristyl aldehyde. The specific function of the 34,000 molecularweight species remains to be fully elucidated, but it has recently been shown to have a long-chain ester hydrolase activity associated with it (Carey et at., 1984). The pathway leading to formation of myristic acid [Eq. (3)] may involve several steps; a lesion in any one should give rise to a dark mutant whose luminescence could be restored by addition of either exogenous aldehyde or a compound in the pathway subsequent to the point of the lesion. If the product accumulated at the point of a lesion is stable and diffusible, it should be possible to demonstrate it by cross-feeding a mutant blocked at any earlier step. A large number of aldehyde-dependent dark mutants of P. leiognathi were tested in this way (Popova and Shenderov, 1979,1983; Popova et al., 1982). From this, they concluded that there are at least five successive steps in the synthesis of aldehyde, designated as being catalysed by enzymes 1, 2, 3, 4, and 5, the last of which is presumably myristic acid reductase: A

---El

B

EZ

C

EJ

D

E4

E

Acid L A l d e h y d e

(3)

Shenderov and Popova (I980a) make a further and interesting observation, namely, that the nearer the block is to the aldehyde, the smaller is the amount of luciferase synthesized in that mutant. This differs from certain aldehyde mutants of V. harueyi, which apparentIy do not produce less luciferase. However, the lux operons in the different species may differ (see Section IV).

256

J. WOODLAND HASTINGS E T A L .

IV. Molecular Biology There has been considerable progress in studies of the Lux phenotype, the cloning of lux genes and operons, and expression and control of bioluminescence at the DNA level. The chromosomal origin of replication in V . harueyi has been sequenced and compared with five species of enteric bacteria, with which it possesses some highly conserved sequences (Zyskind et a f . , 1983). Cohn et al. (1983) reported cloning a 1.85-kb fragment from V . harueyi B-392 which carried the coding sequence for the entire a subunit and part of the p subunit of luciferase. The clone was obtained by means of a mixed-sequence synthetic nucleotide probe. A 17-nucleotide-long sequence with minimum degeneracy was selected from the available protein sequence of the a subunit of the luciferase from V . harueyi (Baldwin et al., 1979a), and a family of eight permutations corresponding to that region of the molecule was chemically synthesized. The 1.85-kb clone was detected by screening a total gene bank of V . harueyi in bacteriophage Charon 13, using the synthetic nucleotides as probes. Restriction and sequence analysis of the clone revealed the following. First, luxA and 1uxB genes, coding for the luciferase and p subunits, respectively, are linked, separated by about 80 non-coding bases upstream from the start of the structural gene for luxB; second, both polypeptide-coding regions start with a codon for methionine on the 5' end and have consensus ribosome-binding sequences of AGGA (Shine and Dalgarno, 1974) 9- 10 base pairs upstream from the first codon, indicating similarity in the genetic arrangement in marine bacteria to other prokaryotes. Finally, comparison of the nucleotide sequence in the N-terminal coding region of luxA and luxB genes showed about 66% homology between the first 38 base pairs. This last information supports the hypothesis of Baldwin et al. (1979a) that the luciferase genes arose by duplication of an ancestral gene. The complete sequence of luxA (Fig. 4) has been recently reported (Cohn et al., 1985). Complete genes for the two polypeptides of the luciferase from V . harveyi (strain BB7) were isolated on a single BamHl fragment by Belas et al. (1982). Two mutants defective in synthesis of luciferase were isolated following transposon mutagenesis. The mutants' genomic DNA was cloned into two plasmids and Tn5-carrying segments selected by transposon-encoded tetracycline resistance. When E. coli was transformed with these two clones on compatible plasmids, an active complement of lux genes was regenerated by recombinational exchange and, thus, elimination of transposon inserts. The 5-kb BarnH1 fragment which was recov-

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

257

ered carried both luxA and luxB genes but a promoter was required for their expression. Escherichia coli carrying the BamHl clone (pBB 1 10) needed exogenous aldehyde to produce light. Furthermore, light production could be enhanced several fold by cloning the lux genes downstream from bacteriophage XPL or PR promoters. Our laboratory has independently cloned the lux genes from V. harueyi B-392 into pBR322 (Gupta et al., 1983; 1985). A gene bank of V. harueyi at the BamHl site of pBR322 was screened by exposure of the transformed E. coli colonies to decanal vapors. Several colonies emitted light, and each of these contained a 5-kb BamH1 fragment, apparently the same as the one reported by Belas et al. (1982). The 5-kb BamH1 fragment was conjugated into dark mutants of V. harueyi possessing known lesions in the lux genes, thereby complementing the defective lux genes in the V. harueyi mutants (Gupta et al., 1985). Using similar methodology, another clone (pWF101) of lux genes at the HindIII site of pBR322 was obtained. The pWFlOl clone contained an 11-kb segment of V. harueyi DNA, carried on two HindIII fragments (Gupta et al., 1983). Escherichia coli transformed with pWFlOl was similar to cells that carried the BarnH1 clone, except that the former emitted a very low level of light without exogenous aldehyde. Baldwin et al. (1984) isolated a 4-kb HindIII fragment from V. harueyi also containing luxA and luxB; expression in E. coli after insertion into pBR322 also required a promoter, and luminescence required the addition of decanal. With a plasmid carrying only luxA, no cr subunit was detected immunologically, suggesting that it is unstable unless associated with p. In these studies with V. harueyi, it appeared that only a part of the complete bioluminescence system had been cloned, as judged by the aldehyde requirement and the absence of auto-induction. By contrast Engebrecht et al. (1983) cloned a 16-kb BamHl fragment from V. Jischeri which provided E. coli with the complete Lux phenotype, not requiring exogenous aldehyde for light emission. Furthermore, the transformed bacteria showed characteristics of auto-induction comparable to wildtype V. Jischeri, thus indicating that regions for regulatory functions of luminescence were also carried on the cloned fragments. Following transposon Tn5 and mini-Mu mutagenesis with subsequent complementation analysis, two operons, L and R, were defined which were required for the complete Lux phenotype (Fig. 5). The operons mapped on a 9-kb SalI fragment, internal to the BarnHl sites. Expression of operon R carrying genes for auto-inducer, luciferase and aldehyde synthesis was auto-inducible, whereas expression of operon L was not dependent upon auto-inducer. Based on the characterization and orientation of various lux functions, Engebrecht et al. (1983) and Engebrecht and Silverman (1984)

258

J. WOODLAND HASTINGS E r AL.

+r OPERON

Lux

L

R

R

I

C

D

B

A

E

n n n n D ~ n Regulation

1

0

1

Aldehyde

1

1

1

2

1

3

Luciferase

1

1

4

~

5

1

6

Aldehyde

l

~

7

~

8

~

l

1

~

9kbp

FIG. 5. Proposed organization of the Vibriojscheri lux genes. Operon Rfrom the 5' to the 3' end cames genes for synthesis of auto-inducer (lux& for luciferase (Y and p peptides (IuxA and B ) , and for aldehyde production (luxC, D,and 8.Operon L encodes for a gene (IuxR) which encodes for a receptor molecule-binding auto-inducer, resulting in the activation of transcription of operon R. After Engebrecht et al. (1983) and Engebrecht and Silverman (1 984).

proposed a mechanism involving positive feedback regulation for operon R. According to this model, auto-inducer is produced slowly as a product of the ZuxZ gene on this operon; when the auto-inducer reaches a critical concentration, it interacts with a receptor molecule encoded by the luxR gene on operon L and activates transcription of operon R. Furthermore, since expression of operon L does not respond to auto-inducer, they proposed that the putative auto-inducer receptor (protein) might be the limiting factor in determining maximal levels of bioluminescence. Engebrecht and Silverman (1984) identified and mapped (Fig. 5) seven lux genes and specified their functions, and the apparent molecular weights of their products: ZuxR (auto-inducer receptor; 27,000), ZuxZ (auto-inducer production; 25,000); ZuxA and B (aand /3 luciferase subunits; 40,000 and 38,000); and luxC, D, and E (synthesis or recycling of aldehyde; 38,000, 53,000, and 42,000). The model for V. jischeri contrasts with one proposed by Weiser et aZ. (1981) for V. harueyi. The latter defines the lux operon as being negatively regulated, with repressor bound at the operator, whereas the auto-inducer acts by de-repression, that is, by binding the repressor and allowing transcription to occur. The model was based primarily on the effects of mutagens, DNA-intercalating agents, DNA-damaging agents, and DNA-synthesis inhibitors on the luminescence of the spontaneous dark (dim) mutants (K variants) (Ulitzur, 1982). These dark variants of V. harveyiare readily isolated and genetically stable, yet can also revert to the original luminous phenotype (Nealson and Hastings, 1979). Ulitzur et aZ. (1980) proposed that such variants occur because wild-type cells carrying an episome can be induced to form an autonomous plasmid coding for the repressor. This model was later abandoned (Ulitzur and Weiser, 1981) following the demonstration that the occurrence of plasmids is not corre-

1

~

1

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

259

lated with bioluminesence (Simon et al., 1982). A subsequent model (Weiser et al., 1981) suggests that these dark variants are super-repressed, possibly by a mutation that increases the binding affinity between the operator and the repressor. According to this model, spontaneous bright revertants would be mutants altered in the structure of either the repressor or the operator site. Mutagens that enhance reversion of dark variants to bright forms are postulated to act by increasing the rate of the latter mutations. The fact that DNA intercalating agents cause a rapid phenotypic reversion in these dark variants (Ulitzur and Weiser, 1981) is attributed to their ability to initiate transcription of luciferase genes, either directly or indirectly, by altering binding of the repressor to DNA. These agents do not induce genetic revertants. Thus, the phenotype is again dark when the agent is removed. The same is true for agents that damage or inhibit synthesis of DNA; these are suggested to stimulate “SOS” functions, inducing a protease that inactivates the repressor (Ulitzur, 1982). In P . leiognathi a still different model was postulated by Shenderov et al. (1982), involving three operons with genes for auto-inducer and luciferase being carried on different operons. This was based in part on experimental results indicating the involvement of at least five enzymes in aldehyde biosynthesis, genes for some of which are not on the operon carrying the structural luciferase gene. Greenberg et al. (1979) reported that non-luminous isolates of V. harueyi produce near-normal levels of auto-inducer; K variants of this species do not respond to exogenous auto-inducer (Nealson and Hastings, 1979). These results are consistent with the model of Ulitzur for V. harueyi, but not with the V. fischeri model, in which auto-inducer and luciferase are present on the same operon. On the other hand, V. fischeri and V . harueyi luminescence genes may, in fact, be arranged differently. This is suggested by the fact that with several isolates of V. fischeri from nature variations in luminescence intensity could be accounted for by correlated differences in auto-inducer production, and that added autoinducer increased luminescence in dim strains of this species (Nealson, 1977).

V. Physiology A. CONTINUOUS OR PULSED EMISSION: THE FLOW OF ELECTRONS

As opposed to many luminous organisms where flashes occur (lasting a fraction of a second), light emission in bacteria appears to the eye to be

260

J. WOODLAND HASTINGS ET AL.

continuous with time and quite steady. Some years ago, however, there were reports from the laboratory of Gitelson, Berzhonskaya and Fish and co-workers in Krasnoyarsk that bacterial bioluminescence exhibits an oscillation with a fundamental frequency of 8 cycles sec-l. From single isolated cells, light was reported to be emitted in discrete flashes (0.1 second) with several flashes in a packet within a 10 second period, the packets being repeated about every 50 seconds. Since the bioluminescence reaction is coupled to the electron-transport pathway, the existence of oscillations in light output would have important implications for energy metabolism and its control. However, Haas (1980) was unable to detect any oscillations, pulses, or any type of emission that differed from fluctuations of a continuous coherent source (Haas and Hastings, 1982). To account for these different results, Bartsev (1982) proposed a model in which the bacteria exhibit periodic light emission under some conditions but continuous light under others. In the 1930s, Eymers and Van Schouwenberg explored the control of the partitioning of electron flow between the cytochrome and luciferase pathways in P . phosphoreum (see Harvey, 1952). It was concluded that the luciferase reaction could be directly responsible for as much as 1020% of the total oxygen consumed in cultured cells. Similar values were reported by Dunlap (1984) for P . leiognathi. When one adds to this energy required to synthesize luciferase itself, which may constitute as much as 5% of the soluble protein in a fully induced cell, and the several other accessory proteins also required, the cellular energy which may be required for the ‘‘luxury’’ of emitting light is considerable. Karl and Nealson (1980) studied parameters of cellular energetics (pools of ATP and GTP), adenlyate energy charge, and oxygen consumption for four species of luminous bacteria and found that all levels remained constant throughout the growth cycle while bioluminescence was induced and increased many fold. They showed that an unexplained 10-fold decrease in the cellular ATP level, reported by Ulitzur and Hastings (1978) to occur at the time of induction of the luminous system, was an artefact caused by ATP hydrolysis catalysed by an inducible membrane-bound heat-stable ATPase. They estimated energetic requirements for luminescence based on both oxygen consumption and ATP pool sizes, concluding that it is significant and, interestingly, that the quantum efficiency of the luciferase reaction (quanta emitted per oxygen molecule consumed) in uiuo may be quite high, possibly approaching 1.0. Vysotskii et al. (1981d) have also suggested that the quantum yield in uiuo is higher than 0.5; further studies are needed. Watanabe and Nakamura (1980a) reported that, during growth of P . phosphoreum, bioluminescence in uiuo may rise more rapidly than cell

BiOCHEMlSTRY OF BIOLUMINESCENT BACTERIA

261

mass. Although this resembles auto-induction (see below), they showed that there was no marked change in luciferase content per cell, and attributed the increased luminescence to a change in the relative channelling of electrons between the luciferase and cytochrome systems; thus, the luciferase catalytic cycle is more rapid in cells in the late logarithmic phase compared with the early logarithmic phase. A similar “competition” for electrons has been reported in V. harueyi (Ulitzur et al., 1981); a “luciferase expression quotient” was defined as the ratio of the bioluminescence intensity in uivo to the activity of the extracted luciferase in vitro. Luciferase expression is greater in the presence of inhibitors of the electron-transport system and also at lower oxygen tensions; in both situations luciferase turnover occurs at a rate closer to its maximum. Watanabe and Nakamura (1980b) made the interesting observation that, in a dim mutant of P . phosphoreum, there is less efficient functional coupling between luciferase and the reducing system linked to it, and thus a lower expression quotient. They also reported a specific control by Na+ and K+ over electron flow via luciferase as compared with cytochromes: high concentrations of Na+ favour the flow via cytochromes relative to luciferase, and thus also a lower expression quotient. Such a control mechanism may pertain to Dunlap’s (1985) observation that P . leiognathi cells cultured from a light organ develop a higher level of luminescence in a medium of low osmolarity.

B . INHIBITORS OF BIOLUMINESCENCE

in

UiUO A N D THEIR MECHANISMS

Vysotskii et al. (1981b) studied the inhibition of bioluminescence by phenobarbital in V . harueyi and proposed that this is due to a disturbance in the synthesis of the natural aldehyde factor. There is also evidence that pargyline (known as an inhibitor of flavin) may inhibit luminescence by blocking aldehyde synthesis (Makemson and Hastings, 1979a). As already discussed above (Section 111),Vysotoskii et al. (1981b) concluded that the natural aldehyde factor may not be myristyl aldehyde as proposed by Ulitzur and Hastings (1979b) (or any of several other straight-chain aldehydes), and that stimulation of certain aldehyde-dark mutants by fatty acids is not due to the conversion of acid into aldehyde but to production of some other substance. They proposed that this (aldehyde factor) might be an aldo-acid. Their evidence supporting these proposals is interesting but certainly not definitive. If phenobarbital is added at a concentration greater than 0.3 mM, which decreases luminescence to less than 60% of its initial value, added decanal (60 p ~will ) stimulate to the 60% level; with less phenobarbital and with uninhibited cells, added decanal will

262

J. WOODLAND HASTINGS ET AL.

itself inhibit to the 60% level. This was interpreted by assuming that decanal, which is less effective in the reaction, displaces the natural aldehyde factor on luciferase in the living cell. However, neither myristyl aldehyde nor any other of the several aldehydes tested was more effective in restoring luminescence to higher levels. Thus they argued that the natural aldehyde is some other substance. A direct effect of phenobarbital on luciferase at the concentrations used was ruled out by studies of the in uitro reaction with pure luciferase. The most convincing evidence concerning the identity of the natural aldehyde includes (1) the chain-length specificity of the aldehyde synthesishecycling system and (2) the fact that tetradecanoic acid specifically restores luminescence in an aldehyde-dark mutant of V. harueyi and reverses fatty-acid inhibition (Ulitzur and Hastings, 1979a,b, 1980). Wildtype cells may be stimulated during growth to a greater or lesser extent by addition of decanal, resulting in a rapid response and a flash of luminescence (Ulitzur et al., 1981). This is similar to the response of aldehydedark mutants, but it is usually much less quantitatively. This indicates that natural aldehyde may be limiting in uiuo and thus that some luciferase-peroxy flavin intermediate may be present intracellularly poised to react with aldehyde. Independent measurements of oxygen uptake in wild type and aldehyde mutants, and inhibition of respiration by cyanide and dodecanol, have shown that this peroxy intermediate can and does decay (to give FMN and hydrogen peroxide turnover and therefore oxygen uptake) in the absence of aldehyde (J. C. Makemson, unpublished data). Together these results imply that the availability of natural aldehyde, whatever it may be, could serve as a control mechanism in bioluminescence. Bognar and Meighen (1983) have characterized a fatty aldehyde dehydrogenase which is co-induced with the luminescent system in V. harueyi and which functions to convert aldehyde into the corresponding (inhibitory) alcohol. The level of luminescence in uiuo could be regulated by this enzyme; its inducibility suggests that it is functionally important, but it is not evident exactly how and under what conditions it might act. C. CHEMOTAXIS AND CHEMOSENSORY BEHAVIOUR

Alam and Glagolev (1982a,b) advanced the novel-seemingly unlikelyhypothesis that bioluminescence in bacteria is related to negative chemotaxis. Uncoupling agents were found to act as repellants, causing clockwise flagellar rotation at very low concentrations (1 nM-10 PM),and at the same time causing an increase in bioluminescence. The way in which the proton motive force relates to bioluminescence was not specified, but the

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

263

observations are interesting. A careful survey of some 40 attractants and repellants (Alam, 1983) revealed that V . harveyi is more chemotactically sensitive to many compounds than are any other bacteria so far reported.

D . AUTO-INDUCTION, LUCIFERASE SYNTHESIS, AND TRANSLATION

in vitro

Control at the transcriptional level may involve either repression or activation or both. Auto-induction, first described in our laboratory, is unusual; it involves production of a substance by the cells themselves which, upon its accumulation in the medium (liquid or solid; Barak and Ulitzur, 1981), induces synthesis of several components of the luminescent system. V . harueyi and V . cholerae produce cross-reacting autoinducers (Hada et al., 1985), but a different one is produced by V.Jischeri; the structure and synthesis of the latter was reported by Eberhard et al. (1981) (Fig. 6). When auto-inducer is added directly to cells of V .Jischeri which are at a low density, the onset of luciferase synthesis occurs promptly. The same is true for the as yet unidentified V . harveyi autoinducer. Interestingly, a large number of non-luminous Vibrio species closely related to V . harueyi produce a substance that can induce luminescence in this latter species (Greenberg et al., 1979). Auto-induction predicts that cells kept at a low cell density should synthesize luciferase only at low uninduced levels. This was observed with V . harueyi (Ulitzur and Hastings, 1979a) by means of repetitive dilution, and for strains of five species ( V . harveyi, V.fischeri, V . splendidus, P . leiognathi, and P . phosphoreum) by a continuous dilution (chemostat) technique (Rosson and Nealson, 1981). Equally important, Rosson and Nealson (1981) demonstrated that addition of pure auto-inducer ( V . Jischeri) to a culture at a low cell density resulted in immediate de novo synthesis of luciferase. They also confirmed the existence of constitutive strains of P. leiognathi and P . phosphoreum which continue to synthesize luciferase even at low cell densities. In both of the above studies it was found that the constitutive (basal) levels of luciferase in inducible strains

FIG. 6. Structure of auto-inducer of V . fischeri. Auto-inducer was identified as N-(3-OXOhexanoyl)-3-aminodihydro-2(3H)-furanone.This material was synthesized as a racemate, and the synthetic material was biologically active. From Eberhard et al. (1981).

264

J. WOODLAND HASTINGS ET AL.

were about 1% of the fully induced level but, owing to substrate limitation (because some enzymes responsible for aldehyde synthesis are also autoinducible), the light emission in uiuo may be far lower. A continuous culture of P . leiognathi in which the flow of nutrient is controlled automatically by the intensity of luminescence was described by Zavoruev and Mazhevikin (1982). With a complex medium, undamped oscillations in luminescence occurred. They attributed this to an effect of magnesium ions; the oscillations disappeared in a medium containing low concentrations of this ion. The authors referred to the involvement of an auto-inducer in the synthesis of the luminescent system but at the same time gave data indicating that the luciferase content per cell was practically constant even though the specific luminescence varied greatly. An interesting and quite pronounced effect of phenobarbital, reported by Vysotskii et al. (1981b), is stimulation of luciferase synthesis. When phenobarbital was added to a growing culture at a concentration that inhibits luminescence by at least 99% (4 mM) the growth rate was only slightly affected but the maximum amount of luciferase attained per cell was twice as great as in control cells. Evans et al. (1983) reported in uitro translation of total V . harveyi RNA using S-30 extracts from RNAase-negative mutants of E . coli. Using RNA from highly luminescent bacteria, luciferase a and /3 subunits could be immunoprecipitated from the proteins translated in uitro, but luciferase activity was not reported. No synthesis of bacterial luciferase in uitro could be demonstrated in the translation product of RNA from uninduced luminous bacteria. This indicates that auto-inducer does indeed act at the transcriptional level.

E . ARGININE REQUIREMENT

Synthesis of luciferase in V . harueyi B-392 is repressed irrespective of cell density in cells growing in a minimal medium; under these conditions the constitutive luciferase level is also about 1%. Added arginine stimulates luciferase synthesis dramatically without stimulating growth; added autoinducer has no such effect. Mutants that have escaped this repression have been isolated and characterized (Waters and Hastings, 1977). The internal amino-acid pools of such mutants as well as wild-type cells were studied following addition of arginine (Makemson and Hastings, 1979~). Internal arginine concentration is very low and the stimulatory effect of exogenous arginine on luciferase biosynthesis is distinct from that by CAMP and auto-inducer; it was deduced to occur at the transcriptional level and the actual mediator to be either arginine or argininyl-tRNA.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

265

The mechanism by which arginine or charged tRNA-arg acts at the transcriptional level is unknown; however, the degree of stimulation of transcription is in the same range as that reported for a number of biosynthetic operons (e.g., trp, his, thr) where there is control of transcription by charged tRNA (attenuation; Yanofsky, 1981). Might lux genes be regulated by a mechanism of this sort, involving the intracellular levels of charged tRNA-arg? Sequencing of the leading regulatory region of the cloned lux genes should provide some insight into this tantalizing possibility. The arginine effect can be modulated by the salinity at which the bacteria are grown (Nealson and Hastings, 1979; Waters and Hastings, 1977; see also Dunlap, 1985). Vibrio harueyi growing in different osmolarities adjusts the intracellular concentration of certain amino acids, some so much that they appear to have an osmoregulatory function (Makemson and Hastings, 1979b). At sodium chloride concentrations greater than 2% (w/v) the arginine pool is so low that it was impossible to measure it, and at lower concentrations only a trace could be detected. At the lower salinities the bacteria grow faster and are brighter; at the same time they are less responsive to arginine. The salt effect is related to the medium; it occurs in minimal (prototrophic) media and not in complex media. In the latter case, the bacteria possess an expanded and measurable arginine pool.

F. CATABOLITE REPRESSION

Vibrio harueyi exhibits classic catabolite repression; exogenous glucose represses luciferase synthesis and addition of cyclic AMP overcomes the repression (Nealson and Hastings, 1979). In P. Zeiognathi, regulation of luciferase synthesis and catabolite repression have been studied recently by Vorobyeva et al. (1980), Shenderov and Popova (1980b), and Shenderov et al. (1982). In V. Jischeri, earlier work suggested that catabolite repression does not occur, but recent studies by Friedrich and Greenberg (1983) indicate that it does although it can be overridden by auto-induction. The concentration of iron in the medium may dictate whether glucose causes transient or permanent repression, and may also alter the response to cAMP (Makemson and Hastings, 1982). These effects of iron might be explained by a requirement of cAMP phosphodiesterase for ferrous ions for activity. Under conditions of iron limitation, the activity of this enzyme would be lowered, increasing the pool of cAMP and limiting catabolite repression. Moreover, synthesis of cAMP phosphodiesterase is re-

266

J. WOODLAND HASTINGS ET AL.

lated to growth rate; increases in growth rate increase both the specific activity of cAMP phosphodiesterase and the severity of catabolite repression (Makemson and Hastings, 1982). Consequently, in rapidly growing bacteria (with an adequate supply of iron), a catabolite-repressible operon such as the luciferase operon would not be induced as a result of a combined effect of the inhibition of adenylate cyclase (i.e., a decrease in cAMP synthesis) by substrate and the increased activity of cAMP phosphodiesterase (i.e. , an increase in cAMP removal). Catabolite repression might have functional importance in luminous bacteria growing in the gut tract, where free sugars should be present from digestion of polysaccharides, e.g. , chitin (Nealson and Hastings, 1979). Extracellular chitinase is produced by luminous bacteria, which are commonly found in the gut tract (Ruby and Morin, 1979; O’Brien and Sizemore, 1979; Baguet et al., 1983; Ohwada et al., 1980). It is fascinating to contemplate that in this case a catabolite (glucose) acts to repress synthesis of an enzyme (luciferase) which catalyses a reaction producing an ephemeral product, namely light. G . IRON

It has recently been shown that iron can affect synthesis of luciferase; when V. harveyi is grown in a minimal medium bioluminescence and luciferase synthesis are repressed by added iron (Makemson and Hastings, 1982). This effect, which is not significantly reversed by CAMP,may act through the interaction of iron with CAMP phosphodiesterase (Makemson, 1983). Iron was also proposed to be significant in relation to the function of luciferase as an electron carrier. In V .fischeri,Haygood and Nealson (1984) found that a similar repression of luminescence by iron occurs and suggested a functional significance in relation to the organism’s symbiotic habitat, where iron limitation might provide the host with a mechanism for limiting growth and maximizing bioluminescence. This is the same general consideration as was applied to the physiological effect of oxygen limitation. It is interesting that V. Jischeri is subject to both low oxygen (see below) and low iron controls. The repressive effects of iron on luciferase synthesis and luminescence suggest the intriguing possibility that auto-induction may involve iron. V. Jischeri auto-inducer (the only one whose structure is known) has structural features which suggest that it could act as a siderophore to chelate iron. However, Haygood and Nealson (1985) report that neither the binding nor the transport of iron is involved in auto-induction in this species.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

267

No information is available in the Vibrio species. V . anguillarum, which produces a potent siderophore (Crosa, 1980; Toranzo et al., 1983), is nonluminous, but can induce luminescence in V . harueyi (Greenberg et al., 1979); whether the siderophore is involved in this induction is an interesting question.

H . OXYGEN

Still another mechanism involves molecular oxygen (Nealson and Hastings, 1977). When P. phosphoreum and V.fischeri are grown at very low oxygen concentrations, growth is severely limited but luciferase synthesis continues briskly, resulting in cells with a much higher specific luciferase content than those grown in air. Since luciferase can substitute for (indeed compete with; Grogan, 1984) cytochrome as a terminal carrier of electrons to oxygen, and since its K , value for oxygen may be lower, luciferase could confer an advantage to cells growing under micro-aerophilic conditions (Makemson and Hastings, 1984). Oxygen limitation is an attractive mechanism for the maintenance of symbiotic bacteria in a light organ, limiting their growth while optimizing bioluminescence (Nealson, 1979). The biosynthesis of the luminescent system in P. leiognathi and V. harueyi appears to be independent of oxygen concentration, whereas in V. cholerae there is less luminescence per cell in cultures grown in low oxygen (Hada et al., 1985).

I . LUCIFERASE INACTIVATION

in uiuo

It has been observed and reported by many workers that the luminescence of cells growing in liquid culture declines dramatically in the early stationary phase and that, in parallel with this decline, progressively less luciferase activity can be extracted from the cells. It is well known that in order to obtain maximum yields when isolating luciferase the cells should be harvested at the time of peak luminescence in uiuo, prior to the onset of its decline (Hastings et al., 1978). Luciferase in crude extracts from cells harvested during the inactivation period is nevertheless quite stable, so this inactivation does not appear to be due to simple proteases which are known to inactivate luciferase (Holzman and Baldwin, 1980a). Reeve and Baldwin (1981, 1982) have provided a masterly analysis of this phenomenon in V . harueyi. Loss of luciferase activity in the cells, as measured in cell-free extracts, begins at about the time luciferase synthe-

268

J. WOODLAND HASTINGS E T A L .

sis terminates and is parallelled by the disappearance of luciferase crossreacting material from the 100,000 g supernatant. At lower g values, however, the inactive luciferase stays in the supernatant and retains its immunological cross-reaction with luciferase antibodies. Also, sodium dodecyl sulphate-polyacrylamide gel electrophoresis of the inactive material shows that the composition and size of the luciferase subunits remain unaltered. Thus far, the inactivation process in uiuo has been demonstrated only in the intact cell; it stops when the cells are deprived of oxygen and within 2-3 hours after addition of cyanide or chloramphenicol, and it never begins if these inhibitors are added in the logarithmic phase of growth before the onset of activation. There is circumstantial evidence that the inactivation is correlated with functioning of the bioluminescent system; in two aldehyde-dark mutants the inactivation of luciferase did not occur (Baldwin et al., 1979b). One of these mutants was temperature conditional and, at the temperature permissive for luminescence (aldehyde biosynthesis), luciferase inactivation did occur. This correlation appears to be refuted by the occurrence of mutants in which luminescence decline (and presumably also the luciferase inactivation) does not occur as the cells enter the stationary phase of growth (Reeve and Baldwin, 1981). Also, inactivation of luciferase from other types of dark mutants (e.g., AK-6; Hastings and Nealson, 1977) does occur. Elucidation of the nature of the luciferase inactivation process, and its apparent specificity, should be aided by the study of such mutants. Vysotksii et al. (1981a) independently reported similar observations in P . leiognathi (strain 54). They showed that extractable luciferase activity decreased once the cells reached the stationary phase of growth, that chloramphenicol and rifampicin could prevent the drop in its activity, and proposed that there is a special mechanism for this luciferase inactivation which is distinct from its inhibition. They also made the interesting observation that there is an inhibitor of luciferase activity present in crude extracts which is removed by centrifugation at 200,000 g for 60 minutes.

J. SUICIDE REACTIONS: LIGHT-INDUCIBLE b-FLAVIN, A N D P-FLAVIN

PROTEIN,

Some years ago Mitchell, Tu, and Hastings (see Hastings and Nealson, 1977) described a flavin (called b-flavin) from V . harueyi which was believed to be associated with catalytically inactive luciferase formed in the course of the bioluminescent reaction-in effect, a suicide reaction. This flavin did not occur in aldehyde-dark mutants, and the isolated luciferase-

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

269

bound flavin preparation had the unique property of emitting characteristic bioluminescence following irradiation, and so was termed light-inducible protein. Whether or not this protein is related to the inactivation process studied by Reeve and Baldwin (1981) is not known; it would be interesting to see if it is formed in the bright mutants of Reeve and Baldwin (1981). A possibly similar protein-bound flavin occurs in P . phosphoreum, the flavin moiety of which has recently been shown to be modified by having a long-chain hydrocarbon (aldehyde derived?) attached at either the 7 or 8 position (Kasai er al., 1982). This is interesting but difficult to account for in relation to the proposed reaction mechanism, which involves the reaction of the aldehyde with the peroxy flavin at the 4a position (see Fig. 1). However, if the peroxy adduct could migrate to other ring positions and react with aldehyde it might constitute a dead-end suicide pathway.

VI. Ecology Studies on the ecology of marine luminescent bacteria have probed beyond the question of where we find them, for indeed they are quite ubiquitous (Nealson and Hastings, 1979; Hastings and Nealson, 1981); studies now focus more precisely on why we find them in such a variety of habitats. What environmental parameters influence their distribution and survival? How are their symbiotic associations established and maintained? How might bioluminescence relate to the distribution and survival of these bacteria? Pursuit of answers to these questions continues to enlighten us as to the intricacies of the life styles of these intriguing microorganisms.

A . PLANKTONIC BACTERIA

Luminous bacteria are widely distributed in marine coastal waters and in the open ocean, including deeper water. The occurrence of plasmids in some, but not all, strains of luminous bacteria has been reported (Simon er al., 1982). The production of bacteriocin-like substances in plasmid-carrying strains of V. harueyi is the first such report for marine bacterium (McCall and Sizemore, 1979). Fluctuations in populations of luminous bacteria in coastal waters appear to correlate with seasonal changes in temperature (Ruby and Nealson, 1978; O’Brien and Sizemore, 1979) as well as with environmen-

270

J. WOODLAND HASTINGS E r AL.

tal differences in salinity, incident light intensity, and nutrient availability (Shilo and Yetinson, 1979). V. harueyi has been found in consistently high numbers during the warmer summer months, but in winter may disappear from the water column or be outnumbered by V. fischeri or P . leiognathi. Yetinson and Shilo (1979) reported “summer” and “winter” strains of V. harueyi which differed primarily in their ability to grow at 40 and lWC, respectively, suggesting selection by temperature on the bacteria present in near-shore water. Profiles of species distribution with depth from several stations in the Atlantic Ocean have revealed a vertical stratification of luminous bacteria, with V. harueyi predominating in the warmer surface waters, and decreasing in numbers with depth. P . phosphoreum, absent from or in very low numbers in surface waters, increases in numbers with depth, reaching peak populations at about 600-1000 m (Ruby et al., 1980; Orndoff and Colwell, 1980). Although temperature appears to be the most consistently observed correlate with the presence or absence of a particular species, both sunlight and pressure may also affect survival and distribution of luminous bacteria in the open ocean (Ohwada et al., 1980; Potrikus et al., 1982). The difficulty in determining which factors are most significant in creating the vertical stratification of luminous bacteria in the open ocean is made more complex by the myriad associations of these bacteria with other organisms. Fish that harbour luminous bacteria as enteric or lightorgan symbionts may serve as a reservoir for the organisms, releasing them into the environment, and thus maintaining the planktonic bacterial population (Ruby and Morin, 1979; Nealson et al., 1984; Haygood et al., 1984; Andrews et al., 1984). As one might expect, there is a correlation between the temperature range of symbiotic luminous species and the temperature of their specific host’s habitat; upon release from the light organ, the bacteria would not be at a disadvantage with respect to this physical parameter. Such released symbionts could occur in numbers high enough to produce measurable levels of light, and at least one measurement (in the Strait of Messina) suggests that this may be so (Baguet et al., 1983). Observers directly viewing the deep ocean from within submersibles have reported zones of a diffuse glow, but sea water from such zones has not yet been sampled directly to determine if luminescence is correlated with the number of luminous bacteria or with some other organism(s). Significant populations of the stomiatoid hatchet fish Argyropelecus hemigymnus possessing luminous gut bacteria do occur in the Strait of Messina (Baguet et al., 1983). Andrews et al. (1984) reported an increase in luminous faecal pellets just below the oxygen minimum in their sampling area in the Pacific Ocean; presumably, this increase is the

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

271

result of fresh organic input supporting a population of higher organisms which, in turn, provide a niche for luminous bacteria. Ruby and Morin (1979) have shown that luminous bacteria associated with faecal pellets can readily proliferate and serve as a significant source of luminous cells in the sea water. As Andrews et al. (1984) point out, the presence of luminous faecal pellets in the mid-depths of the open ocean supports the general hypothesis (Nealson and Hastings, 1979) that light production by free-living luminous bacteria may play an important role in nutrient turnover in the water column. Fish attracted to the luminous faecal debris would consume and utilize nutrients that otherwise might be ultimately lost to the ocean floor. Moreover, the bacteria associated with this debris would also be ingested, finding themselves in a more nutrient-rich environment, the gut, where they can proliferate, be excreted, and continue the cycle. It is interesting to consider whether or not this general hypothesis might also be made for the luminescence of recently identified strains of V. cholerae (Desmarchelier and Reichelt, 1981; West and Lee, 1982; West et al., 1983; Hada et al., 1985).

B. HOST-ASSOCIATED BACTERIA

Adaptations that have evolved in both teleost fish and in squid to harbour and utilize the light of luminous bacteria are numerous (Hastings and Nealson, 1981; Nealson and Hastings, 1979; Herring, 1982). In such associations there are a fascinating variety of specific mechanisms by which the light emission of bacteria is used by the host to attract prey, to defend itself, and to communicate. The light organ of the host provides a protected environment and (presumably) nutrient for the bacteria, which in turn provide light for use by the host fish. Physiological adaptations that enable the fish to make use of the bacterial light include light-conducting muscle fibres and tissues, light-absorbing melanophores, shutter mechanisms, and layers of reflective guanine crystals which play a particularly crucial role in transmitting light from the internal organ of the leiognathid fishes (McFall-Ngai, 1983). In addition, observations by McFall-Ngai and Dunlap (1983) indicate that as many as five modes of luminescent behaviour occur in a single species of leiognathid fish, emphasizing the versatility of these fishes in exploiting the light-emitting capability of their bacterial symbionts. There are extensive morphological, physiological, and behavioural adaptations of the fish that allow it to benefit from the evident expense of maintaining an active bacterial culture.

272

J. WOODLAND HASTINGS ET AL.

Symbiosis of luminous bacteria with fish light organs appears to be species specific (Nealson and Hastings, 1979; Kuwae et al., 1982a, 1983). However, immunologically distinct strains may be present in separate light organs from a population of luminous fish species (Kuwae et al., 1982a,b; Kuwae and Kurata, 1982). The mechanism by which young fish are inoculated has not been established, although bacteria in sea water may be the source. Selective pressures of the host physiology (for example, low oxygen tensions that favour growth of organisms possessing luciferase) could also play a role in preventing the “wrong” organism from invading the luminescent bacterium’s niche. Nealson (1979) described two models for the functioning of light-organ symbioses, specifying those selective pressures that could play key roles in maintaining the associations. One model considers V . jscheri and P . phosphoreum, inhabiting external or rectal (open to the exterior via the intestine) light organs (Haneda, 1980). The epithelial cells of the lining of these organs contain large mitochondria (Tebo et al., 1979), which were proposed to metabolize pyruvate excreted by the bacteria and, at the same time, to maintain the oxygen tension at a low value which would limit growth but not luciferase synthesis or activity (Nealson and Hastings, 1977). An alternate model involves nutrient limitation for P . leiognathi, which inhabits an internal, oesophogeal light organ in the leiognathid fishes. Both models predict that the fish can limit growth yet maximize luminescence of its symbionts. The possibility of unlimited growth and host digestion of symbionts was also considered. The light organs of those fish that have been examined microscopically are similar in that they consist of tubules densely packed with bacterial symbionts (Hastings and Nealson, 1977, 1981; Tebo et a / . , 1979; Dunlap, 1984). Haygood et al. (1984) reported that light organs from Monocentris japonicus and two anomalopid fishes harbour lo8 to lo9 luminous cells, and Dunlap (1983b) found that symbionts in the leiognathid fishes are packed into elongate saccules within the tubules, reaching densities up to 2 . 10” cells ml organ fluid-’. These dense concentrations of bacteria within the organs could clearly lead to nutrient and oxygen limitation, slowing growth and regulating luminescence, as the models (above) predict. The proposed slow growth rate, important from the fish’s point of view for energy conservation, has been verified by Haygood et al. (1984) who estimated an in situ doubling time as long as 135 hours for V .jscheri in M . japonicus light organs, and up to 23 hours for symbionts in two anomalopid species. Dunlap (1984) measured luminescence and respiration in P . leiognathi both within intact light organs and immediately after release from the organs into sea water. He found that, although lumines-

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

273

cence per cell was equal before and after release, oxygen uptake was twofold greater in the released cells , suggesting that oxygen limitation may be affecting cellular respiration and, consequently, growth within the organ. Differences in osmolarity between the organ and sea water may be involved (Dunlap, 1985). He found that, at osmolarities comparable to those found in fish tissue, growth and respiration of P . leiognarhi were diminished whereas luminescence was maximized. This latter point is of interest when one considers the observations of Watanabe and Nakamura (1980b) that sodium ions can affect the relative electron flow via the cytochromes and luciferase (see Section V). Clearly, these associations require highly regulated and complex metabolic adaptations on the part of both organisms. Symbionts from some light organs (e.g., flashlight fish) have not yet been cultured and may represent new species. In addition, luminescent endosymbionts (intracellular) have not been cultured. By assaying for the uniquely prokaryotic activity of bacterial luciferase, Leisman et al. (1980) verified the bacterial origin of light from some squids, tunicates, angler fish, and flashlight fish. Nealson and Hastings (1980) likewise demonstrated bacterial luciferase in extracts of pyrosome tunicates, which harbour putative bacterial endosymbionts. It is intriguing that among the as yet uncultivated host-dependent symbionts there are both extracellular and intracellular organisms; Hastings and Nealson (1980) and Nealson et al. (1981) suggest that the extracellular bacteria may be a stage in the evolutionary development of intracellular forms. The intimate associations of luminous bacteria with their hosts would be useful models for studying the theoretical origin of eukaryotic organelles from prokaryotic cells (Nealson and Hastings, 1980; Nealson et al., 1981). Hastings (1983) views the associations as a means of lateral gene transfer, in the sense that the host, in effect, acquires the desired gene, and its product, by maintaining the organism carrying that gene as a symbiont. That these intimate associations could, in fact, lead to hostsymbiont interactions at the molecular genetic level is suggested by the work of Martin and Fridovich (1981) in studies demonstrating the presence of the eukaryotic copper- and zinc-containing superoxide dismutase in the bacterium P . leiognathi, a luminous symbiont of the leiognathid fish. These authors suggest that the bacteria acquired the gene for the eukaryotic enzyme from their host and, under some unknown selective pressure (perhaps host iron sequestration?), retained that gene. If successful transfers such as this can occur from host to bacterium, perhaps these organisms are indeed a model of choice for evolutionary studies of eukaryotic cells.

274

J. WOODLAND HASTINGS €7 AL.

VII. Analytical and Clinical Applications Many readers will have heard of the use of firefly luciferase for detecting ATP. Other luciferases, including bacterial luciferase, can also be utilized to assay many other substances either directly or via appropriate coupling (DeLuca, 1978). Bioluminescent systems are also being developed as luminescence immune assays to be used in lieu of radioimmune assays (Wannlund et al., 1982). In general, bioluminescent analytical techniques offer advantages of high specificity and sensitivity (Wulff, 1983); indeed, such reactions may be capable of detecting as few as 10,000molecules in a given assay. The bacterial system is unique in that, in addition to the isolated luciferase, intact bacteria may be used. Attempts to select strains from nature with particular suitabilities were reported by Makiguchi et al. (1979). Among tests using intact bacteria are those for mutagens and carcinogens (Ulitzur, 1982). Symposium volumes (Schram and Stanley, 1979; DeLuca and McElroy, 1981; Kricka et al., 1985) provide reports on recent work in the field. Analytical applications of the bacterial system have been reviewed by Kratasyuk and Gitelson (1982), clinical applications by DeLuca (1982) and applications in monitoring energy metabolism of endocrine cells by Brolin (1982). The volume edited by Kricka and Carter (1982) has chapters on several major subject areas, including the bacterial system. Owing to space considerations, many of the relevant articles in these volumes are not cited in our references.

A. APPLICATIONS

in vitro

Perhaps the simplest assays in which the bacterial luciferase system has been used involve measurements of compounds that take part in the lightemitting reaction. Measurement of picomolar concentrations of FMN and a number of long-chain aldehydes including aldehyde pheromones of insects are important examples (DeLuca and McElroy, 1981; Thore, 1979; Meighen et af., 1981, 1982, 1983; Grant et af., 1982; Szittner et al., 1982; Morse et al., 1982; Morse and Meighen, 1984). The coupling of other enzymic reactions to luminescent systems has considerably extended the range of analytical applications. Any reaction that produces or utilizes NAD(H), NADP(H), or aldehyde, either directly or indirectly, can be coupled to a luminescent reaction.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

275

1 . Nicotinamide-Nucleotide Coupled Assays

Luminescent bacteria contain NAD(P)H:FMN oxidoreductase [NAD(P)H dehydrogenase (FMN); EC 1.6.8.1.1catalysing the formations of FMNH2 (see Section 111). Thus, reactions leading to production or disappearance of NAD(P)H can be measured by the oxidoreductaseluciferase coupled light-emitting system (Gorus and Schram, 1979; Whitehead et al., 1979; Watanabe and Hastings, 1981). Articles in the recent volumes cited above list a whole array of enzymes, substrates, drugs, vitamins, and other chemicals that have been assayed by the in uitro bacterial luminescence system. 2. Aldehyde-Coupled Assay Watanabe et al. (198213) described luminescent assays for several aminotransferases and dehydrogenases coupled to aldehyde production. For example, the keto acid produced by an aminotransferase reaction was reduced by a NADH-dependent dehydrogenase, thus oxidizing supplied NADH to NAD+. The NAD+ was in turn used by yeast alcohol dehydrogenase to oxidize a long-chain alcohol to the respective aldehyde, the limiting substrate in the luminescent system. 3. Immobilized and Co-Immobilized Coupled Luminescent Systems Bacterial luciferase and NAD(P)H:FMN oxidoreductase have been immobilized on glass beads or rods and Sepharose 4B and their properties examined (Jablonski and DeLuca, 1979; Wienhausen et al., 1982). Recently a number of coupled luminescent systems have been developed by co-immobilizing other enzymes; the results indicate enhanced sensitivity and yields. Kricka and Carter (1982) provide a table for a number of enzymes and substrates assayed by co-immobilized luminescent enzyme systems. More recent additions include D-glucose, L-lactate, 6-phosphogluconate, L-malate, L-alanine, L-glutamate, NAD+, and NADP+ (Wienhausen and DeLuca, 1982); 12 a-hydroxy bile acids (Schoelmerich et al., 1983), dinitrophenol and trinitrophenol (Wannlund and DeLuca, 1982), and primary bile acid (Roda et al., 1982). By packing Sepharose carrying co-immobolized enzymes into small flow cells, automated bioluminescent assays for certain substrates (e.g., NADH and glucose 6-phosphate) have been developed. Up to 30 samples per hour could be assayed,

276

J. WOODLAND HASTINGS ETAL.

and the flow cell was usable for 700 consecutive assays (Kricka et al., 1983). DeLuca and Kricka (1983) have likened sets of co-immobilized enzymes to membrane-bound or juxtapositioned enzymes in a living cell, modelling sequential metabolic steps. Immobilizing a set of 11 sequential enzymes on Sepharose along with bacterial luminescent enzymes yielded more efficient conversion of glucose into alcohol than did the use of soluble enzymes; luminescent enzymes acted to pull the reaction by continually regenerating NAD+. 4. Other Assays

Bacterial luciferase has been used by Schroeder to monitor specific enzyme-ligand binding, and by Baldwin to follow the activity of proteases (both cited in Kricka and Carter, 1982). Both assays are dependent upon measurement of loss of light emission as the reaction proceeds. B . APPLICATIONS in

uiuo

Intact luminous bacteria and mutants thereof have been employed as a very usable sac of enzymes, co-factors, and substrates in the development of rapid and sensitive assays for various compounds and enzymes. The ease of growing and handling non-pathogenic luminous bacteria, the stability of lyophilized cells, their ready reconstitution, and their steady light emission represent a few of the advantages of assays in uiuo. These bacteria respond to small concentrations of substrates and inhibitors or activators of metabolic processes by virtue of direct or indirect effects on luminescence. They also can provide an assessment of the influence of physiochemical or biological factors that alter the integrity of the whole cell, as well as various agents (e.g., mutagens) that affect gene stability or expression. 1 . Wild-Type Luminous Bacteria Varon and Shilo (1981) and Varon (1981) reported the effect of various environmental pollutants on the predatory activity of Bdellouibrios on P . leiognathi. Pollutants such as detergents, heavy metals and pesticides (at concentrations of 1 pg ml-I) interfered with predation and thus blocked

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

277

the decay of luminescence. They suggested that this method could be used in evaluating water quality. A. A. Bulich and colleagues at Beckman Instruments Inc. have developed and marketed an assay for toxicants and pollutants in aquatic environments using reconstituted lyophilized luminous bacteria (Bulich, 1979, 1982; Bulich and Isenberg, 1981). The method (MicrotoxT) measures the decrease in bioluminescence of luminous bacteria following exposure to the pollutant. It has been evaluated in a variety of applications and with numerous substances in water, and with environmental pollutants (Chang et al., 1981; De Zwart and Slooff, 1983). The sensitivity of this assay is also in the range of 1 pg ml-l. The site or sites of cellular action responsible for the decrease in luminescence in these tests are not known. The apparent diverse nature of active compounds would suggest that more than one site might be involved (e.g., membranes and cell permeability, electron transport, ribosomes and protein synthesis). Linked as it is to the electron-transport chain, the bacterial bioluminescent system is clearly sensitive to environmental effects and compounds of many kinds. This test for substances noxious to man, beast and plant was developed to replace a procedure that has heretofore been done by testing the ability of a fish to live 24 hours in the water in question. The possible use of the MicrotoxT system as a biological dosimeter for Y o and 100 kV X-radiation was evaluated by Mantel et al. (1983). Cells were irradiated in the freeze-dried state and reconstituted 20 hours later with the MicrotoxTprocedure. Loss of luminescence was dose dependent but independent of dose rate over the range studied. Use of the system as a toxicological assay for mycotoxins received a favourable report from Yates and Porter (1982). Luminous V. cholerae were found to be extremely sensitive to the bactericidal activity of human serum. The decrease in luminescence in uiuo was a function of serum concentration and its use in detecting abnormalities of immunoglobulins and complement factors in blood was discussed (Barak et al., 1983b). Luminous bacteria have long been used as models for analysis of the effects of temperature, pressure, drugs and anaesthetics in biological systems. Reversal of drug action by hydrostatic pressure was first demonstrated with luminous bacteria some 40 years ago. D. C. White, M. Halsey, and colleagues have shown that the potency of several general anaesthetics in mammals is closely parallelled by the chemical’s capacity to decrease bioluminescence, and that membrane phase changes may be involved (King and White, 1980). This result may help distinguish between the two common models for anaesthetic action, one involving lipid

278

J. WOODLAND HASTINGS H A L .

and the other protein as the primary site of action in the membrane. Halothane and other general anaesthetics also inhibit the in uitro bacterial luciferase system competitively with the aldehyde. Use of luminous bacteria to detect very low concentrations of oxygen goes back even further, to Beijerinck and Molisch at the turn of the century (see Harvey, 1952). The oxygen dependency of bioluminescence in V. jischeri has been used to determine the oxygen affinities of four protozoa (Lloyd et a[., 1982). The same authors (Lloyd et al., 1981) developed a membrane-covered Photobacterium probe for oxygen measurements in the nanomolar range. It responds rapidly (t1,2,less than 8 seconds) in gas or liquid phase and linearly over the range 8.4 p~ to 35 nM oxygen. 2. Mutants a. Acid- and aldehyde-requiring mutants of luminous bacteria. A dim mutant (M 17) of V. harueyi in which myristic acid or a long-chain alde-

hyde stimulates bioluminescence has been used in sensitive (picomolar concentrations) assays for lipases, lipopolysaccharides, and long-chain fatty acids and aldehydes by Utlitzur and colleagues. Ulitzur et a / . (1979) showed that an acid hydrolysate of as little as 1 ng of lipopolysaccharide containing myristic acid-induced light emission in M 17. The same principle was used to develop assays for lipases and phospholipases A and C. Enzymic activities that produced as little as 1 pmol of myristic acid min-* could be measured (Ulitzur and Heller, 1980; Ulitzur, 1979). Strain M 17 has also been used to determine antilipogenic compounds such as cerulenin as well as long-chain unsaturated fatty acids which inhibit myristic acid-stimulated induction of light emission (Ulitzur and Hastings, 1980).

b. Regulation defective mutants. Ulitzur et al. (1980) described a spontaneous dark mutant (strain 8SD18) of P . leiognafhi which was characterized as super-repressed and unable to synthesize luciferase. It was used in a number of studies for assaying various genotoxic agents and antibiotics. The in uitro microbial test developed by Bruce Ames for chemicals with mutagenic capacity is well known (Ames et al., 1975). Ulitzur and colleagues recently reported a similar bioassay method for detecting mutagens which is based on the observation that various compounds restore bioluminescence in strain 8SD18 by creating genetically stable revertants (see Section IV). As shown in Fig. 7, mutagens such as nitrosoguanidine and benzo[a]pyrene increased bioluminescence of strain 8SD18 severalfold when logarithmic-phase cells were exposed to as little as nanogram

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

1071,

0

I

I

I

10' lOe to3 CONCENTRATION(ng mi')

279

I

lo4

FIG. 7. Effect of some genotoxic agents on the reversion rate of the dark strain 8SD18 of P. leiognathi. Compounds at the concentrations indicated (abscissa) were added to cells ( lo5 ml-I) and the luminescence (ordinate) was determined after 22 hours incubation. Symbols for agents: methotrexate (0);proflavin (0);bleomycin (A); benzo-pyrene (0);nonobiocin (+); mitomycin C (B); nitrasoguanidine ((3). After Ulitzur (1982).

per millilitre concentrations of these mutagens. Table 1 lists some known mutagens and the minimum concentrations detectable by this assay. This assay has also been employed (Barak et at., 1983a) to follow the effects and mechanism of phagocytosis which is well known to proceed with genotoxic effects (Klebanoff, 1980; Weitzman and Stossel, 1981). Induction of bioluminescence in strain 8SD18 has also been observed when cells are exposed to agents that intercalate with, inhibit synthesis of, or damage DNA (Ulitzur and Weiser, 1981; Weiser et al., 1981; Levi and Ulitzur, 1983). Figure 7 and Table 1 show the sensitivity of the test to a number of these agents; however, none produced genetically stable revertants and their continued presence was required to maintain the higher level of bioluminescence. Antibiotics that block de nouo synthesis of protein inhibit induction of luminescence by DNA-intercalating agents in strain 8SD18. An assay for several such antibiotics in biological fluids (e.g., serum and milk) was recently reported; concentrations of 0.1-0.3 pg ml-I could be detected (Naveh et al., 1984).

280

J. WOODLAND HASTINGS ET AL.

TABLE I . Minimal concentrations of different mutagens that are significantly detected by the bioluminescence test"

Mutagen

Minimum concentration (pg ml-9

Direct mutagens Base substitution agents 0.002 N-Meth yl-N-nitro-N-ni trosoguanidine 0.005 Ethyl methanesulphonate 0.1 H ydrox ylamine 0.07 H ydrazine Frame-s hift agents 0.6 4-Nitroquinoline-N-oxide 0.6 2-Anthramine 0.6 Quinacrine hydrochloride 0.7 Emodine 0.7 Nitrofluorene 25 2-Amino biphenyl DNA-synthesis inhibitors and DNA damaging agents 0.005 Mitomycin C 0.02 Novobiocin 5 Nalidixic acid 2 Coumermycin 1 .O Methotrexate 1.5 6-Mercaptopunne 0.1 Bleomvcin 50 cis-Platinum 11 diaminodichloride 100 erg mm-* Ultraviolet irradiation DNA-intercalating agents Ethidium bromide 2.5 Acriflavin 0.1 9-Amino-acridine 0.2 Proflavine sulphate 0.2 Caffeine 20 Theophylline 20 3 Norharmane and Harmane From Ulitzur (1982).

VIII. Acknowledgments Work reported in this article was supported in part by a grant from t h e United States National Science Foundation PCMX3-09414. Catherine J . Potrikus was supported by a National Research Service Award No. 5F32 A1 06372 from the United States National Insti-

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

28 1

tutes of Health. Manfred Kurfiirst is a NATO Postdoctoral Fellow (program administered by DAAD).

REFERENCES

Akhurst, R. J. (1983). International Journal of Systematic Bacteriology 33, 38. Alam, M. (1983). FEMS Microbiology Letters 17, 291. Alam, M., and Glagolev, A. N . (1982a). In “Bioluminescence in the Pacific” (J. Gitelson and J. W. Hastings, eds.), pp. 362-369. USSR Academy of Sciences, Krasnoyarsk. Alam, M., and Glagolev, A. N. (1982b). Federation of European Biochemical Societies 143, 144. Ames, B. N., McCann, J., and Yamaski, E. (1975). Mutation Research 31, 347. Anderson, C., Tu, S.-C., and Hastings, J. W. (1980). Biochemical and Biophysical Research Communications 95, 1180. Andrews, C. C . , Karl, D. M., Small, L. F., and Fowler, S. W. (1984). Nature, London 307, 539. Baguet, F., Piccard, J., Christophe, B., and Marechal, G. (1983). Marine Biology 74, 221. Balakrishnan, C. V.; and Langerman, N. (1977). Archives of Biochemistry and Biophysics, 181, 680. Baldwin, T. O., and Riley, P. L. (1980). In “Flavins and Flavoproteins” (K. Yagi and T. Yamano, eds.), pp. 139-147. Elsevier, Amsterdam. Baldwin, T. O., Ziegler, M. M., and Powers, D. A. (1979a). Proceedings of the National Academy of Sciences of the United States of America 76, 4887. Baldwin, T. O., Nicoli, M. Z., and Reeve, C. A. (1979b). In “Fogarty Conference on Limited Proteolysis in Microorganisms” (H. Holzer and G. Cohen, eds.), pp, 217-223. National Institute of Health, Bethesda. Baldwin, T. O., Dougherty, J. J. Jr., Rausch, S. K., and Merritt, M. V. (1981). In “Bioluminescence and Chemiluminescence: Basic Chemistry and Analytical Applications” (M. DeLuca and W. D. McElroy, eds.), pp. 121-128. Academic Press, New York. Bang, S. S., Baumann, L., Woolkalis, M. J., and Baumann, P. (1981). Archives of Microbiology 130, 1 1 1 . Barak, M., and Ulitzur, S. (1981). Current Microbiology 5,299. Barak, M., Ulitzur, S., and Merzbach, D. (1983a). Mutation Research 121, 7. Barak, M., Ulitzur, S., and Merzbach, D. (1983b). Journal of Clinical Microbiology 18, 248. Baranova, N. A., Ismailov, A. D., Egorov, N. S., and Danilov, V. S. (1980). Mikrobiologiya 49, 477. Bartsev, S. I. (1982). In “Bioluminescence in the Pacific” (I. I. Gitelson and J. W. Hastings, eds.), pp. 349-361. USSR Academy of Sciences, Krasnoyarsk. Baumann, L., and Baumann, P. (1980). Current Microbiology 3, 191. Baumann, L., Bang, S. S., and Baumann, P. (1980). Current Microbiology 4,133. Baumann, P., and Baumann, L. (1981). In “The Prokaryotes” (M. P. Starr, H. Stolp, H. G. Truper, A. Balows, and H. G. Schlegel, eds.), pp. 1302-1331. Springer Verlag, Berlin and New York.

282

J. WOODLAND HASTINGS ET AL.

Baumann, P., Baumann, L., Bang, S. S., and Woolkalis, M. J. (1980). Current Microbiology 4, 127. Baumann, P., Baumann, L., Woolkalis, M. J., and Bang, S. S. (1983). Annual Review of Microbiology 37, 369. Baumstark, A. L., Cline, T. W., and Hastings, J. W. (1979). Archives ofBiochemistry and Biophysics 193,449. Belas, R., Mileham, A. J., Cohn, D. H., Hilmen, M., Simon, M. I., and Silverman, M. (1982). Science 218, 791. Bognar, A., and Meighen, E. (1983). Canadian Journal ofBiochemistry and Cell Biology 61,301. Brolin, S. E. (1982). In “Luminescent Assays: Perspectives in Endocrinology and Clinical Chemistry” (M. Serio and M. Pazzagli, eds.), pp. 63-77. Raven Press, New York. Bruice, T. C. (1982). In “Flavins and Flavoproteins” (V. Massey and C. Williams, eds.), pp. 265-277. Elsevier, Amsterdam. Bulich, A. A. (1979). In “Aquatic Toxicology” (L. L. Markings and R. A. Kemerle, eds.), pp. 98-106. American Society for Testing and Materials. Bulich, A. A. (1982). Process Biochemistry MarIApr, p. 45. Bulich, A. A., and Isenberg, D. L. (1981). Instrument Society of America Transactions 20, 29. Byers, D., and Meighen, E. (1984). Journal of Biological Chemistry 259, 7109. Carey, L. M., Rodriguez, A., and Meighen, E. (1984). Journal of Biological Chemistry 259, 10216. Chang, J. C., Taylor, P. B., and Leach, F. R. (1981). Bulletin of Environmental Contamination and Toxicology 26, 150. Chow, P. J., and Fasman, G. (1974). Biochemistry 13, 211, 222. Cohn, D. H., Ogden, R. C., Abelson, J. N., Baldwin, T. O., Nealson, K. H., Simon, M. I., and Mileham, A. J. (1983). Proceedings of the National Academy of Sciences of the United States of America 80, 120. Cohn, D. H., Mileham, A. J., Simon, M. I., Nealson, K. H., Rausch, S. K., and Baldwin, T. 0. (1985). Journal of Biological Chemistry, in press. Crosa, J. H. (1980). Nature, London 283, 566. Danilov, V. S. (1979). Doklady Akademii Nauk SSSR 249,477. Danilov, V. S., Baranova, N. A., Ismailov, A. D., and Egorov, N. S. (1982). European Journal of Applied Microbiology and Biotechnology 14, 125. DeLuca, M. (ed.). (1978). Methods in Enzymology 57. DeLuca, M. (1982). In “Luminescent Assays: Perspectives in Endocrinology and Clinical Chemistry” (M. Serio and M. Pazzaglia, eds.), pp. 115-123. Raven Press, New York. DeLuca, M., and Kricka, L. J. (1983). Archives of Biochemistry and Biophysics 226, 285. DeLuca, M., and McElroy, W. D. (eds.) (1981). “Bioluminescence and Chemiluminescence: Basic Chemistry and Analytical Applications.” Academic Press, New York. Desmarchelier, P. M., and Reichelt, J. L. (1981). Current Microbiology 5, 123. De Zwart, D., and Sloof, W. (1983). Aquatic Toxicology 4, 129. Dougherty, J. J., Jr., Rausch, S. K., and Baldwin, T. 0. (1982). In “Flavins and Flavoproteins” (V. Massey and C. H. Williams, eds.), pp. 379-382. Elsevier, Amsterdam.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

283

Dunlap, P. (1984). Biological Bulletin 167, 410. Dunlap, P. (1985). Archives of Microbiology, 141, 44. Eberhard, A., Burlingame, A. J., Eberhard, C., Kenyon, G. L., Nealson, K. H., and Oppenheimer, N. H. (1981). Biochemistry 20,2444. Engebrecht, J., and Silverman, M. (1984). Proceedings of the National Academy of Sciences of the United States of America 81, 4154. Engebrecht, J., Nealson, K. H., and Silverman, M. (1983). Cell 32, 773. Evans, J. F., McCracken, S., Miyamoto, C. M., Meighen, E. A.,and Graham, A. F. (1983). Journal of Bacteriology 153, 543. Ford, J., and DeLuca M. (1981). Analytical Biochemistry 110, 43. Friedrich, W. F., and Greenberg, E. P. (1983). Archives ofMicrobiology 134,87. Ghiradella, H. (1983). Journal of Morphology 177, 145. Gorus, F., and Schram, E. (1979). Clinical Chemistry 25, 512. Grant, G. G., Slessor, K. N., Szittner, R. B., Morse, D., and Meighen, E. (1982). Journal of Chemical Ecology 8, 923. Greenberg, E. P., Hastings, J. W., and Ulitzur, S. (1979). Archiues of Microbiology 120, 87. Grogan, D. W. (1984). Archives of Microbiology 137, 159. Gupta, S. C., Potrikus, C. J., Faquin, W. C., and Hastings, J. W. (1983). Federation Proceedings of the Federation of the American Society of Biologists 42, 2092. Gupta, S. C., O’Brien, D., and Hastings, J. W. (1985). Biochemical andBiophysical Research Communications, in press. Haas, E. (1980). Biophysical Journal 31, 301. Haas, E., and Hastings, J. W. (1982). In “Bioluminescence in the Pacific” (I. I. Gitelson and J. W. Hastings,eds.), pp.337-347. USSR Academy of Sciences, Krasnoyarsk. Hada, H . S., Stemmler, J., Grossbard, M. L., West, P. A., Potrikus, C. J., Hastings, J. W., and Colwell, R. R. (1985). Systematic and Applied Microbiology, in press. Haneda, Y. (1980). Science Report of the Yokosuka City Museum 27, 5. Hart, R. C., and Cormier, M. J. (1979). Photochemistry and Photobiology 29, 209. Harvey, E. N. (1952). “Bioluminescence”. Academic Press, New York. Harwood, C. S., Bang, S. S., Baumann, P., and Nealson, K. H. (1980). Znternational Journal of Systematic Bacteriology 30, 655. Hastings, 3. W., and Nealson, K. H. (1977). Annual Review of Microbiology 31, 549. Hastings, J. W., Baldwin, T. O., and Nicoli, M. Z. (1978). In “Methods of Enzymology” (N. 0. Kaplan and S. P. Colowick, eds.), Vol. 57, pp. 135-152. Academic Press, New York. Hastings, J. W. (1978). In “Critical Reviews of Biochemistry” (G.Fasman, ed.), Vol. 5, pp. 163-184. CRC Press, Boca Raton, Florida. Hastings, J. W. (1982a). Zn “Bioluminescence in the Pacific” (I. I. Gitelson and J. W. Hastings, eds.), pp. 16-30. USSR Academy of Sciences, Krasnoyarsk. Hastings, J. W. (1982b). In “Oxygenases and Oxygen Metabolism” (M. Nozaki, S. Yamamoto, Y. Ishimura, M. Coon, L. Ernster, and R. Estabrook, eds.), pp. 225-237. Academic Press, New York. Hastings, J. W. (1983). Journal of Molecular Evolution 19, 309.

284

J. WOODLAND HASTINGS E T AL.

Hastings, J. W., and Nealson, K. H. (1980). In “Endocytobiology: Endosymbiosis and Cell Biology” (W. Schwemmler and H. E. A. Schenk, eds.), pp. 467471. de Gruyter, Berlin. Hastings, J. W., and Nealson, K. H. (1981). In: “The Prokaryotes: A Handbook on Habitats, Isolation, and Identification of Bacteria” (M. P. Starr, H. Stolp, H. G. Truper, A. Balows, and H. G. Schlegel, eds.), pp. 1332-45. Springer Verlag, Berlin and New York. Hastings, J. W., and Tu, S.-C. (1981). Annals of the New York Academy of Sciences 366, 315. Hastings, J. W., Tu, S.-C., Becvar, J. E., and Presswood, R. P. (1979). Photochemistry and Photobiology 29, 383. Hastings, J. W., Ghisla, S., Kurfurst, M., and Hemmerich, P. (1981). In “Proceedings of the Second International Congress of Bioluminescence and Chemiluminescence” (M. DeLuca and W. D. McElroy, eds.), pp. 97-102. Academic Press, New York. Haygood, M. G., and Nealson, K. H. (1984). In “Current Perspectives in Microbial Ecology” (M. Klug and A. Reddy, eds.), pp. 56-61. American Society for Microbiology, Washington, D.C. Haygood, M. G., and Nealson, K. H. (1985). Journal of Bacteriology, in press. Haygood, M. G., Tebo, B. M., and Nealson, K. H. (1984). Marine Biology 75, 249. Hemng, P. J. (1982). In “Oceanography and Marine Biology Annual Review” (M. Barnes, ed.), Vol. 20, pp. 415-470. Aberdeen University Press, Aberdeen. Holzman, T. F., and Baldwin, T. 0. (1980a). Proceedings of the National Academy of Sciences of the United States of America 17, 6363. Holzman, T. F., and Baldwin, T. 0. (1980b). Biochemical and Biophysical Research Communications 94, 1199. Holzman, T. F., and Baldwin, T. 0. (1981). Biochemistry 20, 5524. Holzman, T. F., and Baldwin, T. 0. (1982). Biochemistry 24, 6194. Holzrnan, T. F., and Baldwin, T. 0. (1983). Biochemistry 22, 2838. Holzman, T. F., Riley, P. L., and Baldwin, T. 0. (1980). Archives of Biochemistry and Biophysics 205, 554. Irwin, R. M., Visser, J. W. G., Lee, J., and Carreira, L. A. (1980). Biochemistry 19, 4639. Ismailov, A. D., Baranova, N. A., Egorov, N. S., and Davilov, V. S. (1980). Mikrobiologie 49, 377. Ismailov, A. D., Danilov, V. S., Malkov, Y. A., and Egorov, N. S. (1981a). Biokhimiya 46, 40. Ismailov, A. D., Baranova, N. A., Danilov, V. S., and Egorov, N. S. (1981b). Biokhirniya 46, 234. Ismailov, A. D., Egorov, N. S., and Danilov, V. S. (1979). Doklady Akademii Nauk SSSR 249,482. Jablonski, E., and DeLuca, M. (1979). Clinical Chemistry 25, 1622. Jensen, M. J., Tebo, B. M., Baumann, P., Mandel, M., and Nealson, K. H. (1980). Current Microbiology 3, 31 1 . Karl, D. M., and Nealson, K. H. (1980). Journal of General Microbiology 117, 357. Kasai, S., Matsui, K., and Nakamura, T. (1982). In “Flavins and Flavoproteins” (V. Massey and C. H. Williams, eds.), pp. 459-463. Elsevier, Amsterdam.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

285

King, S. M., and White, D. C. (1980). In “Molecular Mechanism of Anaesthesia” (B. R. Fink, ed.), pp. 435-438. Raven Press, New York. Klebanoff, S. J. (1980). Annals of Internal Medicine 93, 480. Koka, P., and Lee, J. (1979). Proceedings of the National Academy of Sciences of the United States of America 76, 3068. Koo, J.-Y., Schmidt, S. P., and Schuster, G. B. (1978). Proceedings of the National Academy of Sciences of the United States of America 76, 30. Kosower, E. M. (1980). Biochemical and Biophysical Research Communications 92, 356. Kratasyuk, V . , and Fish, A. M. (1980). Biokhimiya 45, 1175. Kratasyuk, V . , and Gitelson, J. (1982). BioJzika 27, 937. Kricka, L. J., and Carter, T. J. N. (eds.). (1982). “Clinical and Biochemical Luminescence.” Dekker, New York and Basel. Kricka, L. J., Weinhausen, G. K., Hinkley, J. E., and DeLuca, M. (1983). Analytical Biochemistry 129, 392. Kricka, L. J., Stanley, P. E., Thorpe, G. H., and Whitehead, T. P. (eds.) (1985). “Bioluminescence and Chemiluminescence.” Academic Press, New York. Kurfiirst, M., Ghisla, S., and Hastings, J. W. (1982a). In “Flavins and Flavoproteins” (V. Massey and C. Williams, eds.), pp. 353-358. Elsevier, Amsterdam. Kurfiirst, M., Ghisla, S., Presswood, R., and Hastings, J. W. (1982b). In “Flavins and Flavoproteins” (V. Massey and C. Williams, eds.), pp. 359-364. Elsevier, Amsterdam. Kurfiirst, M., Ghisla, S., Presswood, R. P., and Hastings, J. W. (1982~).European Journal of Biochemistry 123, 355. Kurfiirst, M., Ghisla, S., and Hastings, J. W. (1983). Biochemistry 22, 1521. Kurfiirst, M., Ghisla, S., and Hastings, J. W. (1984). Proceedings of the National Academy of Sciences of the United States of America 81, 2990. Kuwae, T., and Kurata, M. (1982). Microbiology and Immunology 26, 643. Kuwae, T., Sasaki, T., and Kurata, M. (1982a). Microbiology and Immunology 26, 455. Kuwae, T., Fukasawa, S., Sasaki, T., and Kurata, M. (1982b). Microbiology and Immunology 26, 1181. Kuwae, T., Andoh, M., Fukasawa, S., and Kurata, M. (1983). Microbiology and Immunology 27, 847. Lambert, M. A., Hickman-Brennen, F. W., Farmer, J. J., and Moss, C. W. (1983) International Journal of Systematic Bacteriology 338, 777. Lee, J. (1982). Photochemistry and Photobiology 36, 689. Leisman, G., and Nealson, K. H. (1982). I n “Flavins and Flavoproteins” (V. Massey and C. Williams, eds.), pp. 383-386. Elsevier, Amsterdam. Leisman, G., Cohn, D. H., and Nealson, K. H. (1980). Science 208, 1271. Leisman, G., and Nealson, K. H. (1985). Proceedings of the National Academy of Sciences of the United States of America, in press. Levi, B.-Z., and Ulitzur, S. (1983). Archives Microbiology 134, 261. Lhoste, J.-M., Favaudon, V . , Hastings, J. W., and Ghisla, S. (1980). In “Flavins and Flavoproteins” (K. Yagi and T. Yamano, eds.), pp. 131-138. Japan Scientific Societies Press, Tokyo. Lloyd, D., James, K., Williams, J., and Williams, N. (1981). Analytical Biochemistry 116, 17.

286

J. WOODLAND HASTINGS H A L .

Lloyd, D., James, K., Williams, J., and Williams, N. (1982). Journal of General Microbiology 123, 1019. Makemson, J. C. (1983). Journal of Microbiological Methods 1, 283. Makemson, J. C., and Hastings, J. W. (1979a). Archives of Biochemistry and Biophysics 196, 396. Makemson, J. C., and Hastings, J. W. (1979b). Applied and Environmental Microbiology 38, 178. Makemson, J. C., and Hastings, J. W. (1979~).Journal of Bacteriology 140,532. Makemson, J. C., and Hastings, J. W. (1982). Current Microbiology 7, 181. Makemson, J. C., and Hastings, J. W. (1984). American Society for Microbiology, Abstracts of Annual Meeting, St. Louis, Missouri, March 1984. American Society for Microbiology, Washington, D.C. Makiguchi, N., Arita, M., and Asai, Y. (1979). Journal of General and Applied Microbiology 25, 387. Mantel, J., Freidin, M., Bulich, A. A., and P e n y , H. (1983). Physical and Medical Biology 28, 599. Martin, J. P., and Fridovich, I. (1981). Journal of Biological Chemistry 256,6080. Matheson, I . B. C., and Lee, J. (1981). Biochemical and Biophysical Research Communications 100, 532. Matheson, I. B. C., and Lee, J. (1983). Photochemistry and Photobiology 38,231. Matheson, I . B. C., Lee, J., and Muller, F. (1981). Proceedings of the National Academy of Sciences of the United States of America 78, 948. Meighen, E. A. (1979). Biochemical and Biophysical Research Communications 87, 1080. Meighen, E. A., and Bartlet, I. (1980). Journal of Biological Chemistry 255, 11181. Meighen, E. A., Slessor, K. N., and Grant, G. G. (1981). Experientia 37, 555. Meighen, E. A., Slessor, K. N., and Grant, G. G. (1982). Journal of Chemical Ecology 8, 91 1 . Meighen, E. A., Szittner, R. B., and Grant, G. G. (1983). Analytical Biochemistry 133, 179. Merritt, M. V . , and Baldwin, T. 0. (1980). Archives ofBiochemistry and Biophysics 202, 499. Morse, D., and Meighen, E. A, (1984). Journal ofBiological Chemistry 259,475. Morse, D., Szittner, R., Grant, G. G., and Meighen, E. A. (1982). Insect Physiology 28,863. McCall, J. O., and Sizemore, R. K. (1979). Applied and Environmental Microbiolo g y 38,974. McCarron, S . H., and Tu, S.-C. (1983). Photochemistry and Photobiology 38, 131. McFall-Ngai, M. J. (1983). Journal of Experimental Zoology 227, 23. McFall-Ngai, M. J., and Dunlap, P. V. (1983). Marine Biology 73, 227. Nakamura, ’I‘. (1982). In “Flavins and Flavoproteins” (V. Massey and C. H. Williams, eds.), pp. 369-374. Elsevier, Amsterdam. Naveh, A., Potasam, E., Basan, H., and Ulitzur, S. (1984). Journal of Applied Bacteriology 56, 457. Nealson, K . H. (1977). Archives of Microbiology 1l2, 73. Nealson, K. H. (1979). Trends in Biochemical Sciences 4, 105. Nealson, K. H., and Hastings, J. W. (1977). Archiues of Microbiology 112, 9. Nealson, K . H., and Hastings, J. W. (1979). Microbiological Reviews 43, 496.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

287

Nealson, K. H., and Hastings, J. W. (1980). I n “Endocytobiology: Endosymbiosis and Cell Biology” (W. Schwemmler and H. E. A. Schenk, eds.), pp. 461466. de Gruyter, Berlin. Nealson, K. H., Cohn, D. H., Leisman, G., and Tebo, B. M. (1981).Annals of the New York Academy of Sciences 361, 76. Nealson, K. H., Haygood, M. G., Tebo, B. M., Roman, M., Miller, E., and McCosker, J., (1984). Microbial Ecology 10, 69. Nefsky, B., and DeLuca, M. (1982). Archives of Biochemistry and Biophysics 216, 10. O’Brien, C. H., and Sizemore, R. K. (1979). Applied and Environmental Microbiology 38, 928. Ohwada, K., Tabor, P. S., and Colwell, R. R. (1980). Applied and Environmental Microbiology 40, 746. Orndorff, S. A., and Colwell, R. R. (1980). Applied and Environmental Microbiology 39, 983. Paul, V. J., Frautschy, S., Fenical, W., and Nealson, K. H. (1981). Journal of Chemical Ecology 7, 589. Petushkov, V. N., Tratasynk, G. A., Kratasynk, V. A., and Belobrov, P. I. (1982). Biokhimiya 47, 1773. Poinar, G. O., Thomas, G., Haygood, M. G., and Nealson, K. H. (1980). Soil Biology and Biochemistry 12, 5. Popova, L. Y., and Shenderov, A. N. (1979). Genetika 15, 1555. Popova, L. Y., and Shenderov, A. N. (1983). Biokhimiya 48, 983. Popova, L. Y., Shenderov, A. N., and Latskaya, N. I. (1982). In “Bioluminescence in the Pacific“ (I. I. Gitelson and J. W. Hastings, eds.), pp. 237-250. USSR Academy of Sciences, Krasnoyarsk. Potrikus, C. J., Greenberg, E. P., Ruby, E. G., and Hastings, J. W. (1982). Abstracts of Annual Meeting of American Society for Microbiology, p. 194. American Society for Microbiology, Washington, D.C. Presswood, R. P., and Hastings, J. W. (1979). Photochemistry and Photobiology 30, 93. Primakova, G . A., Vorobyeva, T. I., and Fish, A. M. (1982). in “Bioluminescence in the Pacific” (I. I. Gitelson and J. W. Hastings, eds.), pp. 227-236. USSR Academy of Sciences, Krasnoyarsk. Primakova, G. A., Turova, T. P., Zavoruev, V. V., and Vorobyeva, T. I. (1983). Microbiology (USSR) (English Translation) 52, 82. Rausch, S. K., Dougherty, J. J., Jr., and Baldwin, T. 0. (1982). In “Flavins and Flavoproteins” (V.Massey and C. H. Williams, eds.), pp. 375-378. Elsevier, Amsterdam. Reeve, C. A., and Baldwin, T. 0. (1981). Journal of Bacteriology 146, 1038. Reeve, C. A., and Baldwin, T. 0. (1982). Journal of Biological Chemistry 257, 1037. Riendeau, D., and Meighen, E. (1979). Journal of Biological Chemistry 254,7488. Riendeau, D., and Meighen, E. (1980). Journal of Biological Chemistry 255, 12060. Riendeau, D., and Meighen, E. (1981). Canadian Journal ofBiochemistry 59,440. Riendeau, D., Rodriguez, A., and Meighen, E. (1982). Journal of Biological Chemistry 257,6908. Roda, A., Kricka, L. J., DeLuca, M., and Hofmann, A. F. (1982). Journal of Lipid Research 23, 1354.

288

J. WOODLAND HASTINGS ET AL.

Rodriguez, A,, Riendeau, D., and Meighen, E. (1983a). Journal of Biofogical Chemistry 258,5233. Rodriguez, A., Wall, L., Riendeau, D., and Meighen, E. (1983b). Biochemistry 22, 5604. Rosson, A., and Nealson, K. H. (1981). Archives of Microbiology 129, 299. Ruby, E. G., and Hastings, J. W. (1979). Current Microbiology 3, 157. Ruby, E. G., and Hastings, J. W. (1980). Biochemistry 19, 4989. Ruby, E. G., and Morin, J. G. (1979). Applied and Environmental Microbiology 38, 406. Ruby, E. G., and Nealson, K. H. (1978). Limnology and Oceanography 23,530. Ruby, E. G., Greenberg, E. P., and Hastings, J. W. (1980). Applied and Enuironmental Microbiology 39, 302. Ryerson, C. C., Ballou, D. P., and Walsh, C. (1982). Biochemistry 21, 2644. Schoelmerich, J., Hinkley, J. E., MacDonald, I. A., Hofmann, A. F., and DeLuca, M. (1983). Analytical Biochemistry 133, 244. Schram, E., and Stanley, P. (eds.) (1979). “Proceedings of the International Symposium on Analytical Applications of Bioluminescence and Chemihminescence.” State Printing and Publishing Co., Westlake, California. Schwab, J. M., Li, W.-B., and Thomas, L. P. (1983). Journal of the American Chetnical Society 105, 4800. Shenderov, A. N., and Popova, L. Y. (1980a). Genetika USSR 16, 1109. Shenderov, A. N., and Popova, L. Y. (1980b). Proceedings of the Siberian Academy of Science, 85. Shenderov, A. N., and Popova, L. Y. (1982a). Genetika USSR 18, 1283. Shenderov, A. N., and Popova, L. Y. (1982b). In “Bioluminescence in the Pacific” (I. I. Gitelson and J. w. Hastings, eds.), pp. 251-273. USSR Academy of Sciences, Krasnoyarsk. Shenderov, A. N., Popova, L. Y., and Videletz, I. Y. (1982). In “Bioluminescence in the Pacific” (I. I. Gitelson and J. W. Hastings, eds.), pp. 251-273. USSR Academy of Sciences, Krasnoyarsk. Shilo, M., and Yetinson, T. (1979). Applied and Environmental Microbiology 38, 577. Shimomura, 0. (1982). In “Chemical and Biological Generation of Excited States” (W. Adam and G. Cilento, eds.), pp. 249-276. Academic Press, New York. Shine, J., and Dalgarno, L. (1974). Proceedings of the National Academy of Sciences of the United States of America 71, 1342. Shumikhin, V. N., Danilov, V. S., and Egorov, N. S. (1980a). Bioorganischeskaya 6, 765. Shumikhin, V. N., Danilov, V. S., and Egorov, N. S. (1980b). Biokhimiya 45, 1576. Simon, R. D., Shilo, M., and Hastings, J. W. (1982). Current Microbiology7, 175. Small, E. D., Koka, P., and Lee, J. (1980). Journal of Biological Chemistry 255, 8804. Suzuki, K., Kaidoh, T., Katagiri, M., and Tsuchiya, T. (1983). Biochimica et Biophysica Acta 722, 297. Swanson, R. (1984). Bulletin of Mathematical Biology 46, 623. Szittner, R. B., Morse, D., Grant, G. G., and Meighen, E. A. (1982). Journal of Chemical Ecology 8, 935.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

289

Tebo, B. M., Linthicum, D. S., and Nealson, K. H. (1979). Biosystems 11, 269. Thomas, G. M., and Poinar, G. O., Jr. (1979). International Journal of Systematic Bacteriology 29, 352. Thomas, G. M., and Poinar, G. O., Jr. (1983). International Journal of Systematic Bacteriology 33, 378. Thore, A. (1979). Annals of Clinical Biochemistry 16, 359. Toranzo, A. E., Barja, J. L., Potter, S. A., Colwell, R. R., Hetrick, F. M., and Crosa, J. H. (1983). Infection and Immunity 39, 1220. Tu, S.-C. (1979). Biochemistry 18,5940. Tu, S.-C. (1982). Journal of Biological Chemistry 257, 3719. Tu, S.-C., and Hastings, J. W. (1980). Proceedings of the National Academy of Sciences of the United States of America 77, 249. T u , S.-C., and Henkin, J. (1983). Biochemistry 22, 519. Tu, S.-C., Becvar, J. E., and Hastings, J. W. (1979). Archives of Biochemistry and Biophysics 193, 110. Ulitzur, S. (1979). Biochimica et Biophysica Acta 572, 211. Ulitzur, S . (1982). Trends in Analytical Chemistry 1, 329. Ulitzur, S., and Hastings, J. W. (1978). Journal of Bacteriology 133, 1307. Ulitzur, S., and Hastings, J. W. (1979a). Current Microbiology 2, 345. Ulitzur, S., and Hastings, J. W. (1979b). Proceedings of the National Academy of Sciences of the United States of America 76, 265. Ulitzur, S., and Hastings, J. W. (1979~).Journal of Bacteriology 137, 854. Ulitzur, S., and Hastings, J. W. (1980). Current Microbiology 3, 295. Ulitzur, S., and Heller, M. (1980). Methods in Enzymology 72, 338. Ulitzur, S., and Weiser, I. (1981). Proceedings of the National Academy of Sciences of the United States of America 78, 3338. Ulitzur, S., Yagen, B., and Rottem, S. (1979). Applied and Environmental Microbiology 37, 782. Ulitzur, S., Weiser, I., and Yannai, S. (1980). Mutation Research 74, 113. Ulitzur, S., Reinhertz, A., and Hastings, J. W. (1981). Archives of Microbiology 129, 67. Varon, M. (1981). Microbial Ecology 7 , 97. Varon, M., and Shilo, M. (1981). Microbial Ecology 7 , 107. Visser, A. J., and Lee, J. (1980). Biochemistry 19, 4366. Visser, A. J., and Lee, J. (1982). Biochemistry 21, 2218. Vorobyeva, T. I., Vysotskii, E. S., Zavoruev, V. V., and Mezhevikin, V. V. (1980). Mikrobiologiya 49, 517. Vorobyeva, T. I., Zavoruev, V. V., Mezhevikin, V. V., and Primakova, G. A. (1982). Mikrobiologiya 51, 420. Vysotskii, E. S., Zavoruev, V. V., and Mezhevikin, V. V. (1981a). Mikrobiologiya 50, 410. Vysotskii, E. S., Zavoruev, V. V., and Mezhevikin, V. V. (1981b). Mikrobiologiya 50, 985. Vysotskii, E. S., Zavoruev, V. V., Mezhevikin, V. V., Rodicheva, E. K., Fish, A. M., and Gitelson, I. I. (1981~).Doklady Akademii Nauk SSSR 261, 216. Vysotskii, E . S., Rodicheva, E. K., and Shcherbakova, G. Y. (1981d). Mikrobiologiya 50, 419. Vysotskii, E. S., Zavoruev, V. V., and Mezhevikin, V. V. (1982a). Biokhimiya47, 1983.

290

J. WOODLAND HASTINGS ET AL.

Vysotskii, E. S., Zavoruev, V. V., Kalacheva, G. S., and Mezhevikin, V. V. (1982b). In “Bioluminescence in the Pacific” (I. I. Gitelson and J. W. Hastings, eds.), pp. 314-323. USSR Academy of Sciences, Krasnoyarsk. Wall, L., Rodriguez, A,, and Meighen, E. A, (1984a). Journal of Biological Chemistry 259, 1409. Wall, L. A., Byers, D. M., and Meighen, E. A. (1984b). Journal of Bacteriology 159, 720. Wannlund, J., and DeLuca, M. (1982). Analytical Biochemistry 122, 385. Wannlund, J., Egghart, H., and DeLuca, M. (1982). In “Luminescent Assays: Perspectives in Endocrinology and Clinical Chemistry” (M. Serio and M. Pazzagli, eds.), pp. 125-28. Raven Press, New York. Ward, W. W. (1979). In “Photochemical and Photobiological Reviews” (K. Smith, ed.), pp. 1-57. Plenum Press, New York. Ward, W. W. (1981). Photochemistry and Photobiology 33, 965. Watanabe, H., and Hastings, J. W. (1981). In “Proceedings of the Second International Congress of Chemiluminescence and Bioluminescence” (M. DeLuca and W. D. McElroy, eds.), pp. 417-423. Academic Press, New York. Watanabe, H., and Hastings, J. W. (1982). Molecular and Cellular Biochemistry 44, 181. Watanabe, H., and Nakamura, T. (1976). Journal of Biochemistry (Japan) 79, 489. Watanabe, H., and Nakamura, T. (1980a). Journal of Biochemistry (Japan) 88, 815. Watanabe, H., and Nakamura, T. (1980b). In “Saline Environment” (H. Morishita and M. Masui, eds.), pp. 51-66. Conference on Halophilic Microbiology. Japan Science Society, Japan. Watanabe, H., Kamita, Y., Nakamura, T., Takimoto, A., and Yamanaka, T. (1979). Biochimica et Biophysica Acta 547, 70. Watanabe, H., Hastings, J. W., and Tu, S.-C. (1982a). Archives of Biochemistry and Biophysics 215, 405. Watanabe, H., Hastings, J. W., Wulff, K., Michal, G., and Staehler, F. (1982b). Journal of Applied Biochemistry 4, 508. Watanabe, T., Nakamura, T., Matsui, K., and Kasai, S. (1980). In “Flavins and Flavoproteins” (K. Yagi and T. Yarnano, eds.), pp. 125-129. Japan Scientific Societies Press, Tokyo. Waters, C. A., and Hastings, J. W. (1977). Journal of Bacteriology 131, 519. Weiser, I., Ulitzur, S., and Yannai, S. (1981). Mutation Research 91, 443. Weitzman, S. A., and Stossel, T. P. (1981). Science 212, 546. Welches, W. R., and Baldwin, T. 0. (1981). Biochemistry 20, 512. West, P. A., and Lee, J. V. (1982). Journal of Applied Bacteriology 52, 435. West, P. A., Lee, J. V., and Bryant, T. N. (1983). Journal ofAppliedBacteriology 55, 263. Whitehead, T. P., Kricka, L. J., Carter, T. J. N., and Thorpe, G. H. H. (1979). Clinical Chemistry 25, 1531. Wienhausen, G. K., and DeLuca, M. (1982). Analytical Biochemistry 127, 380. Wienhausen, G. K., Kricka, L. J., Hinkley, J. E., and DeLuca, M. (1982). Applied Biochemistry and Biotechnology 7 , 463. Wilson, T. (1985). In “Singlet Oxygen” (A. Frimer, ed.), pp. Vol. 2, pp. 37-65. CRC Press, Boca Raton, Florida. Wulff, K. (1983). In “Methods of Enzymatic Analysis” (H. V. Bergmeyer, ed.), 3rd Ed., pp. 340-368. Verlag Chemie, Weinheim.

BIOCHEMISTRY OF BIOLUMINESCENT BACTERIA

291

Yang, Y., Baumann, L., Woolkalis, M. J., and Bang, S. S. (1983). Current Microbiology 8, 95. Yanofsky, C. (1981). Nature, London 289, 751. Yates, I. E., and Porter, J. K. (1982). Applied and Environmental Microbiology 44, 1072. Yetinson, T., and Shilo, M. (1979). Applied and Environmental Microbiology 37, 1230. Zavoruev, V. V., and Mezhevikin, V. V. (1982). Mikrobiologiya 51, 224. Ziegler, M. M., and Baldwin, T. 0. (1981a). In “Current Topics in Bioenergetics” (D. Rao Sanadi, ed.), Vol. 12, pp. 65-72. Academic Press, New York. Ziegler, M. M., and Baldwin, T. 0. (1981b). In “Bioluminescence and Chemiluminescence: Basic Chemistry and Analytical Applications” (M. DeLuca and W. D. McElroy, eds.), pp. 121-128. Academic Press, New York. Zyskind, J. N., Cleary, J. M., Brusilow, W. S. A,, Harding, N. E., and Smith, D. N. (1983). Proceedings of the National Academy of Sciences of the United States of America 80, 1164. Note Added in Proof The necessity of a non-haem iron in the luciferase from Vibriu harueyi has been recently reported [Danilov, V. S.,and Malkov, Y. A. (1984). Doklady Akademii Nauk SSSR 275, 2061. The use of the luciferase system as an indicator of gene expression has been described by Engelbrecht et al. [Engelbrecht, J., Simon, M., and Silverman, M. (1985). Science 227, 13451.

This Page Intentionally Left Blank

Author Index Numhcrs in iinlics refer i o the poges on which rqferences ore lisied article.

A Abbot, M. T., 58, 82 Abelson, J . N . , 249, 256, 282 Acheroy, B., 43, 45, 46, 82 Adair, W. S., 99, 110, 1 1 1 , 116, 119, 120 Adams, M. W. W., 157, 174, 175, 177, 178, 179, 180. 182, 184, 185,220

Adelberg, E. A., 126, 154 Agmon, V., 147, 154 Aguilar, M. O., 207, 230 Aguire, R., 181, 222 Ahmed, A , , 141, 154 Ahombo, G., 207, 209, 233 Aigle, M., 9 , 11, 34, 50, 81, 82 Akhurst, R. J., 238, 281 Alam, M., 262, 263, 281 Albersheim, P., 110, 120 Alberti, B. N., 211, 226 Albracht, S. P. J., 175, 176, 179, 181, 182,

UI

the end of euch

Anthony, C., 64,80 Apferson, A , , 141, 150 Aragno, M., 186, 221 Arita, M., 274, 286 Arnon, D. I . , 162, 170, 194, 195, 198, 221, 222, 227, 231, 234

Arntzen, C. J., 140, 148, 150 Arp, D. J., 169, 196, 197, 221 Arst, H. N., Jr., 58, 59, 60,61, 62, 63, 64, 65,68,69,70,71,72,73,74.75.76,77,78, 79, 80, 82, 83, 85, 86, 87

Artz, S. W., 142, 144, 153 Asai, Y., 274, 286 Atherly, A. G., 141, 150 Aubert, B., 91. 123 Auberf, J. P., 10, 82 Ausubel, F. M., 191, 230 Avtges, P., 205, 206, 221

220, 221, 232

Albrecht, S. L., 209,221 Albrecht-Ellmer, K. J., 182, 220 Alder, D. P., 205, 234 Alef, K., 196. 197, 221 Allen, J . F., 141, 148, I50 Allen, J. R., 140, 151 Allibert, P., 168, 221 Al-Taho, N . M., 60, 79 Ames, B. N., 278, 281 Andersen, K., 209, 221, 227 Anderson, C., 249, 251, 281 Anderson, L., 162, 221 Andoh, M., 272,285 Andreo, C. S., 139, 140, 152 Andrews, C. C., 270, 271,281 Antanaitis, B. C., 179, 180, 232

Baccarini-Melandri, A,, 140, 146, 147, 150, 187, 221, 234

Bachofen, R., 166, 176, 194, 213, 214, 215, 221, 222, 233, 234

Baguet, F., 266, 270, 281 Bagyinka, C., 176, 184, 211, 221, 226 Bailey, C. R., 62, 72, 74, 75, 77, 78, 79 Baird, K. L., 117, 121 Bakker, E. P., 142, 150 Bakker-Grunwald, T . , 146, 150 Balakrishnan, C. V., 248,281 Balch, W. E., 160,223 Baldwin, T. O., 236,241,243,248,249,251, 252,253,256,257, 267,268,269,276,281, 282, 283, 284, 286, 287, 290, 291 Ball, T. M., 71, 81

293

294

AUTHOR INDEX

Balloni, W., 217, 233 Ballou, D. P., 241, 288 Baltscheffsky, H., 143, 150, 191, 198, 200, 229 Baltscheffsky, M., 143, 153 Ban, D., 4, 31, 81 Bang, S. S., 237, 238, 281. 282, 283, 291 Barak, M., 263, 277, 279, 281 Baranova, N. A., 240, 247, 281, 282, 284 Barciella, S., 157, 212, 215, 229 B a j a , J. L., 267,289 Barnett, J. A , , 7, 79 Barthelmess, I. B., 67, 79 Bartlet, I., 251, 286 Bartnik, E., 58, 61, 66, 72, 79 Bartsch, R. G., 175, 225 Bartsev, S. I., 260, 281 Basan, H., 279, 286 Bassham, J. A., 168, 169, 215, 218, 228 Baulaigue, R., 167, 170, 228 Baumann, L., 237, 238, 281, 282, 291 Baumann, P., 237, 238, 281, 282, 283, 284 Baumgartel, D. M., 91, 122 Baumstark, A. L., 253, 282 Bavendamm, W., 160, 221 Beavo, J. A., 138, 152 Bkchet, J., 4, 7, 13, 38, 80, 82 Becvar, J . E., 240, 246, 284, 289 Behki, R. M., 209,221 Belas, R., 256, 257, 282 Belavina, N., 205, 207, 231 Belobrov, P. I., 248, 287 Bender, R. A., 205, 221 Benemann, J. R., 157, 210, 212, 221, 224 Bennett, J., 140, 148, 150, 151 Bennett, M. A., 21 I , 221 Bennett, R., 173, 221 Benz, R., 144, 154 Benziman, M., 143, 150 Bergman, K., 104, I19 Berlier, Y., 183, 190, 193, 225, 233 Berliner, J., 103, 123 Berndt, H., 185, 186, 187, 221 Bernstein, E., 92, 119 Bernstein, J. D., 211, 221 Berzinskiene, J., 143, 151 Beynon, J., 76, 80 Biedermann, M., 171, 231 Bishop, N. I . , 157,221 Bjorkman, M., 66, 85

Black, M. T., 140, 151 Blakemore, R., 160, 223 Blanco, L., 71, 84 Blijham, J . M., 23, 34, 85 Blithe, D. L., 117, I21 Bloodgood, R. A., 98, 117, 118, 119 Blumenthal, R., 144, 154 Bocher, R., 181, 232 Bognar, A,, 262, 282 Bogorov, L. V., 161, 221 Bojczuk, A. T., 98, 119 Bold, H. C . , 94, 120 Bole, D. G., 144, 150 Bolscher, J. G. M., 132, 151 Bonen, L., 160,223 Bonomi, F., 139, 151 Boonstra, J., 129, 136, 152 Boos, K. S . , 140, 150 Boos, W., 136, IS0 Booth, I. R., 134, 150, IS1 Boquet, P. L., 141, 143, 154 Bordes, A. M., 40, 87 Borisy, G. G., 117, I21 Bose, S. K., 165, 221 Bossinger, J., 19, 20, 27, 30, 80 Bothe, J., 157, 222 Bothe, J., 157, 222 Bowen, A. R. S. G., 204, 205, 209, 230 Bowers, B., 97, I20 Bowien, B., 175, 186, 222, 223 Box, V., 36, 86 Brandriss, M. C., 12, 14, IS, 24, 40, 49, 80, 84 Braun, V., 143, 144, 150 Bregoff, H. M., 165, 223 Brenchley, J. E., 10, 80 Bridger, W. A., 142, I50 Brierly, G. P., 144, 153 Brill, W. J., 201, 206, 230 Brill, W. S., 191, 231 Broach, J., 205,222 Broekman, R., 92,93,96, 104, 105, 106, 107, 122 Brolin, S. E., 274, 282 Brom, S., 71, 81 Brosseau, J . D., 157, 234 Brown, C. M., 10, 84, 87 Brown, I. I., 130, IS0 Brown, R. M., Jr., 94, 103, 104, 120, 121, 123

AUTHOR INDEX

Brownlee, A. G., 59, 65. 71. 74, 77, 79, 80. 85 Bruice, T. C., 240, 282 Brune, D. C., 195, 229 Brusilow, W. S . A., 256, 291 Bryant, T. N., 238, 271, 290 Buchanan, B. B., 139, 150, 162, 176, 194, 195, 222, 231 Buchanan, M., 122 Buchanan-Wollaston, V., 76, 80 Bhchner, K.-H., 144, 124 Bueno, R., 205, 221 Bulen, W. A., 165, 190, 198, 222 Bulich, A. A., 277, 282, 286 Buranakarl, L., 167, 212, 233 Burgess, B. K., 190. 191, 194, 222 Burlingame, A. J.. 263. 283 Burns, R. C., 165, 198,222 Burrascano, C. G., 91, 123 Burris, R. H., 165. 190, 191, 192, 196, 197, 198, 199, 2 0 , 224, 227, 229. 232, 233 Byers, D. M., 254, 282, 290

C Cabib, E., 97, 120 Caligor, E., 99, 100, 101, 102, 117, 122 Calva, E., 71, 84 Calvin, M.. 190, 231 Cammack, R., 181, 222, 227 Campomanes, M., 71, 86 Cannon, F. C., 76,80, 82 Cannon, M., 76, 80 Cantrell, M. A., 209, 210,222, 224 Cantu, M., 196, 198, 234 Cardenas, J., 164, 222 Cardillo, T. S., 23, 31, 82 Carey, L. M., 255, 282 Carithers, R. P., 194, 198,222, 234 Carlson, K., 103, 123 Carreira, L. A., 243, 284 Carter, T. J . N., 274, 275, 276, 285. 290 Casadio, R., 146, 147, 150. 158, 187, 188, 228 Castillo, F., 164, 195, 196, 222, 234 Caumette, P., 161, 210,222 Cerletti, P., 139, 151 Chabert, J., 175, 177, 179, 180, 222 Chakrabarti, S., 169, 222 Chambers, J. A. A., 63, 66, 67, 80

295

Charnpomanes, M., 7 I , 84 Chang, J . C., 277, 282 Chaustova, L. P., 143, I50 Chen, B., 166, 188, 231 Chen, H., 166, 188, 231 Chen, J. S.. 157, 174, 182, 220 Chen, K. N., 160,223 Chevallier, M. R., 50, 80. 82 Chibata, I., 216, 222 Chow, J., 112, 120 Chow, P. J., 250, 282 Christophe, B., 266, 270, 281 Chuaqui, C. A., 211, 222 Ciardi, J . E., 141, 154 Claes, H., 96, 97, I20 Clark, J . E., 174, 230 Clausell, A., 102, 117. 122 Claybrook, D. L., 197, 234 Clayton, R. K., 157, 222 Cleary, J. M., 256,291 Clement-Metral, J. D., 139, I50 Cline, T. W., 253, 282 Clutterbuck, A. J., 59, 62, 80, 86 Coddington, A., 62, 80 Cohen, B. L., 60, 61, 64,75, 80 Cohen, G . N., 5,80 Cohen, P., 117, I20 Cohen, R., 110, 111, 119 Cohen-Bazire, G., 5, 36, 85 Cohn, D. H., 249, 250, 256, 257, 273, 282, 285. 287 Colbeau, A., 157, 164, 166, 168, 174, 175, 177, 178, 179, 180, 181, 184, 185, 186, 187, 188,201,203,204,205,210,215,217,222, 225, 230, 233 Colwell, R. A,, 267, 289 Colwell, R. R., 238, 263, 266, 267, 270, 271, 283, 287 Colwin, A. L., 93, I20 Colwin, L. H., 93, 120 Condeelis, J., 94, 120 Cook, R. J., 64.80 Cooke, S., 204, 205,230 Cooper, J . A , , 117, 121 Cooper, J . B., 110, 120 Cooper, T. G., 5, 19, 20, 25, 26, 27, 28, 30, 40, 51, 54, 80, 86, 87 Copeland, H. F., 160, 223 Cormier, M. J., 243, 283 Cornwell, E. V., 71, 80

AUTHOR INDEX

296

Courchesne, W. E., 18, 49, SO, 80 Cousen, S. A , , 59, 65, 71, 79 Cove, D. J., 58.60,61,62,63,64,72, 75.76, 77, 79. 80, 85, 87 Cox, J . C., 171,223, 228 Crabeel, M., 31, 38, 39, 40, 41, 42. 4s. 48, 52, 80, 81, 82, 83 Crettaz, M., 117, 121 Crofts. A. R., 146, 151, 187,234 Crosa, J . H., 267, 282, 289 Cuatrecasas, P., 117, 122 Culigor, E., 99, 120

D Dalgarno, L., 256, 288 Dalton, H., 201, 223 Dalton, P., 157, 212, 215, 229 Dancshazy, Z., 176, 221 Daniels, C. J., 144, I50 Daniels, L., 181. 225, 226 Danilov, V. S., 240, 247, 248,281, 282, 284, 288 Dantzig, A. H., 71, 81 Darlington, A. J . , 58, 86 Darte, C., 39, 40, 53. 81 Darvill, A. G., 110, 120 Dassa, E., 141, 143, 154 Datil, D., 181, 222 Davila, G., 71, 81 Davis, R. H., 12, 14, 81, 87 Day, P. R., 16.82 Deamer, D. W., 127, 150 Decker, K., 129, 154 de Crombrugghe, B., 18, 85 De Hauwer, G . , 15, 81 Dekker, C. A, , 57, 72, 83 De Ley, J., 214, 215, 231 Delmer, D. P., 143, 150 DeLuca, M., 237, 246, 274, 275, 276, 282, 283, 284, 285, 287, 288, 290 Dentler, W. L., 100, 120 de Petris, S., 99, 123 de Robichon-Sjulmajster, H., 40, 87 Der Vartanian, D. V., 181, 182, 227, 232 De Santis, A., 158, 187, 188, 228 Deschamps, J., 16, 22, 23, 31, 81, 82 Desmarchelier, P. M., 238, 271, 282 Detmers, P. A,, 94, 96, 120 De Vos, P., 214, 215, 231

de Wildt, P., 93, 96, 122 De Zwart, D.. 277,282 Dibrov, P. A., 143, 144, 151 Diekert, G., 180, 232 Dills, S. S., 141, 150 Dilworth, M . J . , 193, 223 Dimroth, P., 130, 150 Distler, E., 1.57, 222 Dixon. R., 76, 85 Dixon, R. 0. D., 188, 189, 223 Donohue-Rolfe, A. M., 143, 144, 150 Dougherty, J . J., Jr.. 248, 249,281,282,287 Doussiere, J., 176, 223 Dowling, T. E., 191, 192, 197, 198, 199,223, 225, 227, 228 Doyle, R. J., 143, 144. 151 Dragert, W., 36, 84 Drake, H. L., 174, 223, 230 Drews, G., 110, 123, 149, 152, 204, 230 Drillien, R., 9, I I , 34, SO, 81, 82 Drucker, H., 75, 80 Dubois, E., 5, 7, 8, 9, 1 1 , 15, 16, 17, 18, 19, 20.2 I , 22,23,25,26,27,28,29,30,3 I , 33, 34, 35. 36,37, 39,40,42,43,45,46,47,49, 50, 51, 52, 54, 81. 82, 83, 84 Dudman, W. F., 110, 120 Duffus, W. P. H., 99, 123 Dunlap, P., 260,261.265,271,272,273,283. 286 Dunlop, P. C., 4, 8, 31, 32, 51, 81, 83, 86 Dunn, E., 60, 61, 64,71, 85 Dunn-Coleman, N. S., 62,63,64,66,67,71, 81, 82, 87 Dunstan, R. H., 163, 223 Durr, M., 7 , 8 7 Dutton, P. L., 195, 230 Dyer, T. A,, 160, 223 Dykstra, C. C., 58, 85

E Eady, R. R.. 193, 227 Eaton, M., 71, 84 Eberhard, A., 263, 283 Eberhard, C., 263, 283 Eberhardt, U., 186,231 Eckert, R., 118, 122 Eddy, A. A., 40, 82 Egghart, H., 274, 290

297

AUTHOR INDEX

Egorov, N . S., 240, 247, 248, 281, 282, 284, 288 Eisbrenner, G., 157, 183, 222, 223 Elder, J . T., 205, 234 Elferink, M. G . L., 131, 146, 147, 148, 149, 151

Elmerich, C., 10, 82 Elzenga, J . T. M.,l00, \\6, 120 Emerich, D. W.. 209, 221 Engebrecht, J., 254. 257, 258,283 Ephrussi, B., 6, 82 Epstein, W., 136, 142, 144, 152. 153 Eriksson, U., 191, 198, 200. 229 Ernster, L., 139, 153 Ero, L., 103, 109, 122 Errede, B., 23, 31, 82 Eshar, Z., 117, 122 Estrela, J . P., 140, 153 Evans, E. H., 187, 234 Evans, H. J., 183, 186, 209, 210, 221. 222, 223, 224, 228 Evans, J. F., 264, 283 Evans, M . C. N., 194, 232 Evans, M. C. W., 162, 191, 195, 222, 223 Even, H. L., 8, 10, 35, 42, 51, 81, 86

F Facklam, T. J., 63, 67, 82 Fahlbusch, K., 196, 224 Falk, G., 197, 205, 206, 223 Fan, L. T., 210, 231 Faquin, W. C., 257, 283 Farmer, J. J., 237, 285 Fasman, G., 250,282 Favaudon, V., 239, 285 Favinger, J . L., 171. 223 Feigenblum, E., 176, 223 Fenical, W., 238, 287 Ferguson, A . R.,36,82, 86 Ferguson, S. J., 171, 173, 228 Fichtinger-Schepman, A. M. J . , 110, 120 Filser, M., 76, 82 Fincham, J. R. S., 16, 57, 67, 71, 82, 84 Finck, A., 180, 223 Fink, G. R., 16, 83 Fish, A. M.,238,251,255,260,285,287.289 Fisher, E. H., 141, 152 Flier, J. S., 117, 121 Florenzano, G., 167, 217, 233

Forbes, E., 60, 61, 64,71, 85 Ford, J., 246, 283 Forest, C. L., 91, 99, 116, 120 Fornari, C. S., 205, 207, 218, 223 Forster, H., 103, 120 Fowler, S. W., 270, 271, 281 Fox, G. E., 160,223 FQX,

J . A., \8\, 225, 226

Franceschini, T., 144, I51 Frank, J., 157, 212, 215, 229 Frautschy, S., 238,287 Freidin, M., 277, 286 Frenz, J., 143, 144, 150 Fridovich, I., 273, 286 Friedberg, I., 131, 146, 147, I51 Friedmann, J., 93, 120 Friedrich, B., 180, 186, 209, 214,223 Friedrich, C. G., 180, 186, 209, 214, 223 Friedrich, W. F., 265, 283 Frings, W., 180, 225 Frolova, V., 205, 207, 231 Fujii, K., 210, 226 Fukasawa, S., 272, 285 Fukui, S., 216, 223 Fuller, R. C., 162, 173, 190, 221, 231 Futai, M., 143, 151

G Gabel, C. A., 118, 122 Gaffron, H., 162, 163, 223 Galperin, M. Y., 130, 143, 144, 150, 151 Gandy, C., 188,225 Garland, P. B., 176, 225 Garlid, K. D., 144, 153 Garrett, R. H., 62, 63, 66, 67, 71.81, 82, 87 Gazdar, C., 143, 144, 153 Gerwig, G. J., 109, 120 Gest, H., 162, 164, 165, 166, 168, 169, 170, 171, 174, 175, 187, 190, 195, 1%. 202,203, 205,207,208,2 13,221,223,224,226,228, 230, 233 Ghiradella, H., 283 Ghisla, S., 239, 240, 241, 242. 243,284, 285 Gibbons, N. E., 158, 159,224 Gibor, A., 96, 97, 98, 114, 116, 121, 122 Gibson, J . , 160, 188, 223, 224 Giles, N . H., 58, 63, 85, 86 Ginsburg, A., 36, 86, 141, 154 Gitelson, I. I.. 255, 289

298

AUTHOR INDEX

Gitelson, J . , 274, 285 Gitlitz, P. H . , 176, 177, 179, 180, 184, 185, 224 Gits, J., 38, 39, 41, 82 Glagolev, A. N., 130, 143, 144, 150, 151, 262, 281 Gogotov, 1. N . , 157, 167, 175, 177, 178. 179, J8D, 184, 18-5. 187, 194, 195. 224,226,23J, 232. 234 Goldberg, A. L., 144, 154 Goldberg, E. B., 143, 152 Goldberg, R. B., 76, 83 Golecki, J . R., 204, 230 Gonzalez, A., 67, 82 Gooday, G. W., 97, 120 Gooden, D., 58, 66, 71, 85 Goodenough, D. A., 91, 96, 104, 115, 116, 119, 120, 123 Goodenough, U. W., 90, 91, 92, 93, 94, 95, 96, 99, 100, 101, 102, 104, 110, 1 1 1 , 115, 116, 117, 119, 120, 121, 122, 123 Gorini, L., 4, 83 Gorlenko, V. M., 161, 224 Gorrell, T. E., 171, 172, 224 Gorski, M., 86 Gorton, D. J., 77, 82 Gorus, F., 275, 283 Gotto, J . W., 191, 192, 196, 197, 198, 199, 200, 224, 234 Gould, J. M . , 139, 140, 154 Gowans, C. S., 90,121 Graf, E. G., 180, 181,220, 224 Graf, P., 140. 153 Graham, A., 176, 224 Graham, A. F., 264,283 Granick, S., 91, 122 Grant, G. G., 274, 283, 286, 288 Gray, C. T., 174,224 Gray, T. A., 176,225 Greenbaum, J., 157, 212, 215, 229 Greenberg, E. P., 259, 263, 265, 267, 270, 283, 287 Gregory, M. E., 52,82 Grenson, M., 4 , 5 , 7 , 8 , 9 , 11, 12, 15, 16, 17, 18,19,20,21,25,27,28,29,30,31,34,35, 36,38,39,40,41,42,43,44,45,46,47,48, 49, 50, 51, 52, 53, 54, 80, 81, 82, 83, 84 Greth, M. L., 50, 82 Griffard, P., 157, 212, 215, 229 Griffon, S. M., 63,66, 67,80

Grinius, L. L . , 143, 150, 151 Griniuviene, B . B . , 143, I50 Griswold, W. R., 58, 82 Grogan, D. W., 267,283 Grossbard, M. L., 238, 263, 267, 271,283 Grove, G., 58, 63, 70, 75, 82 Guest, J . R., 176, 232 Cupfa, R., 160,223, 224 GUptd, s. c.,257, 283 Giirgiin, V . , 171, 224 Guth, J. H . , 199, 200, 224 Gutman, M., 139, 151 Guzewska, J., 66, 79 Guzman, J., 71, 81

H Haaker, H., 143, 151, 152, 201, 231 Haas, E., 260, 283 Hada, H. S., 238, 263, 267, 271, 283 Hageman, R. V., 199, 227 Hahn, M., 76, 85 Hall, D. O., 175, 177, 178, 179, 180, 181, 184, 185, 220, 227, 231 Hallenbeck, P. C., 157, 166, 168, 189, 191, 192, 194, 195, 199,202,203,204,210,212, 215, 216, 217, 221, 224, 225, 233 Hamilton, W . A . , 134, I50 Hanan, R., 76, 83 Handke, K., 143, 144, 150 Haneda, Y . , 272, 283 Haniu, M., 194, 232 Hansberg, W., 71,86 Hanson, M. A., 75, 83 Hanson, R. L., 139, I51 Hanus, F. J . , 186, 209,221, 224, 228 Harder, W., 149, 151 Hardi, G., 141, 151 Harding, N. E., 256, 291 Harold, F. M., 126, 136, 142, 143, 151 Harris, D. A., 146, 151 Hart, R. C., 243, 283 Hartmann, M., 90,121 Harvey, E. N., 260, 278,283 Harwood, C. S., 237,283 Haselkorn, R., 202, 205, 206, 207, 208,221, 231 Hastings, J . W., 236,237,238,239,240,241, 242,243,245,246,248,249,251,252,253, 254,257,258,259,260,261,262,263,264,

299

AUTHOR INDEX

265,266,267,268,269,270,271,272, 273, 275,278,281,282,283,284,285.286,287, 288, 289,290 Hatchikian, E. C., 181, 222 Haugland, R. A , , 209, 210, 222, 224 Hausinger, R. P., 181, 226 Hauska, G., 140, 150 Haygood, M. G., 238, 266, 270, 272, 284, 287 Haynes, L., 157, 212, 215, 229 Hearst, J . E . , 205, 234 Heefner, D. L., 136, 151 Heimke, J. W., 91, 121 Heine, E., 180,223 Heldt, H.-W., 143, I50 Hellenbeck, P. C., 224 Heller, M . , 278, 289 Hellingwerf, K. J., 126, 131, 132, 143, 146, 147, 148, 149, 151, 152, 154, 176, 232 Hemmerich, P., 241, 284 Hemmings, B. A., 7, 35, 36, 83 Hendley, D. D., 163,224 Henkin, J., 252, 289 Hennaut, C., 5, 8, 11, 22, 24, 25, 26, 28, 29, 31, 39, 40, 81, 82, 83 Hennecke, H., 76, 85 Henning, S. B., 141, 154 Henry, M. F., 163, 164, 175, 183, 224, 233 Hernaez, L., 117, 122 Herring, P. J . , 271, 284 Hespell, R. B . , 160, 223 Hetrick, F . M . , 267, 289 Heuser, J . E., 110, 120, 121 Hickman-Brennen, F . W., 237, 285 Hierholzer, G., 9, 34, 83 Hiernaux, D., 16, 31, 81 Hill, C., 157, 212, 215, 229 Hill, S., 76, 83, 85 Hillmer, P., 165, 168, 169, 170, 190, 196, 202, 203, 224 Hilmen, M., 256, 257, 282 Hinkle, P. C., 128, 152 Hinkley, J . E . , 246, 275, 276, 285, 288, 290 Hinkson, J., 165, 222 Hirsch, P., 174, 225 Hiura, H., 175, 225 Hoare, D. S., 162, 173, 230 Hodenpijl, P. G., 109, 121 Hoffman, J . L., 99, 100, 101, 102, 117, 120, 121, 122

Hofmann, A. F., 275, 287, 288 Hogrefe, C., 209, 223 Hojeberg, B., 143, 151 Holloman, W. K., 57, 72. 83 Holmgren, A., 139, I51 Holzer, H., 7, 9, 34, 36, 83, 84, 87 Holzman, T . F., 248,249,251, 252,267,284 Homan, W., 92, 103, 104, 105, 106, 107, 109, 110, 120, I Z I , I22 Homer, R. B., 108, I21 Hong, J.-S., 136, 151 Honya, M., 183,234 Horak, J., 49, 83 Horio, T., 175, 225 Horton, P., 140, 151 Hoshino, Y., 163, 230 Hou, C., 9, 18, 27, 38, 39, 40, 42, 45,48, 52, 82 Houchins, J . P., 157, 225 Houwer, B., 143, 151 Howell, S. H., 91, 122 Hu, N. T., 204, 234 Hummelt, G., 71, 83 Hunt, A. G., 136, 151 Hunter, J . , 117, I21 Huynh, B. H., 181, 182,227. 232 Hwang, C . , 91, 94, 96, 110, 116, 119, 120, 121 Hyams, J. S., 117, 121 Hynes, M., 207,230 Hynes, M. J., 59, 60, 61, 71, 72, 75, 78, 83, 85

I Imae, Y., 146, 153 Imhoff, J . F., 159, 225 Ingraham, J. L., 126, 154 Inouye, M., 144, I51 Inouye, S., 144, 151 Irwin, R. M., 243, 284 Isenberg, D. L., 277, 282 Ismailov, A. D., 240, 247,281, 282, 284 Itakura, K., 144, 151 Ito, K., 167, 212, 213, 226, 233 Ivanovsky, R. N . , 160, 161, 226 lyer, V. N . , 209,221 Izawa, S., 188, 227

AUTHOR INDEX

300

J Jablonski, E., 275, 284 Jackson, J. B., 171, 173, 228 Jacobs, E., 25. 26, 28, 29, 31, 81, 83 Jacobs, P., 16, 83 Jacobson, F. S., 181, 225 Jacoby, G. A.. 4, 83 Jaenicke, L., 92, 97, 121. 122 Jahn, T. L., 92, 119 James, K., 278, 285. 286 Jarrett, D. B., 117, 121 Jasaitis, A. A., 143, 150 Jauniaux. J. C., 6, 14, 15, 16, 26, 31, 81, 83, 87 Jawitz, J . , 104, 119 Jelenc, P. C., 142, I51 Jensen, M. J., 238, 284 Johansson, B. C., 197, 205, 206, 207, 208, 223, 233 Johansson, C., 202, 225 Johnson, C., 94, 120 Joiris, C. R.,40, 83 Jolliffe, L. K., 143, 144, 151 Jones, B. L., 195, 196, 225 Jones, E. W., 16, 36, 83 Jones, G. E., 4, 12, 31.83 Jones, L. W., 157, 221 Jones, 0.T. G., 195, 225 Jones, R. F., 91, 121 Jones, R. W., 176, 225 Jones, S. A.. 59, 72, 78, 79, 83 Jouanneau, Y., 157, 164, 166, 168, 189, 190, 191, 192, 193, 194, 195, 199,200,201,203, 204, 205, 209, 213, 215, 217,224, 225, 233 Jund, R., 50, 80 Jung, D. W., 144, 153 Jungermann, K., 129, 154, 171, 225 Jurivich, D., 99, 116, 120

K Kaback, H. R., 134, 149, 151, 153 Kadziauskas, J . P., 143, 150 Kahn, C. R., 117, 121 Kaidoh, T., 239, 288 Kakuno, T., 175, 176, 179, 180, 184, 225 Kalacheva, G. S., 255, 290 Kalkman, M. L., 181, 182,221 Kaltwasser, H., 180, 225

Kamen, M. D., 165, 176, 178, 179, 180, 184, 195,223, 225, 226 Kamerling, J. P., 109, 110, 120 Kamibayashi, A,, 213, 229 Kamita, Y ., 247, 290 Kampee, T., 167, 212, 233 Kampf, C., 188. 225 Kamps. A., 112, 113, 114, 122 Kanazawa, H., 143, 151 Kanemoto, H., 197, 225 Kanernoto. R., 199, 228 Kang, L., 32, 83 Kao, I. C.. 210, 231 Kaplan. N . O., 176, 178, 179, 180, 184,225. 226 Kaplan, S.. 146, 147, 151, 205, 207, 218,223 Karl, D. M., 260. 270. 271, 28f. 284 Kasai, S., 240, 269, 284, 290 Kaska, D. D., 96, 97. 121 Kasuga, M., 117, 121 Katagiri, M . , 239, 288 Kates, J. R., 91, 121 Katoh, T., 163, 225 Katsuki, H., 142, 154 Kauss, H., I 1 I, 121 Kavanagh, E., 76, 83 Keeler, M. L., 32, 83 Keenan, M. H. J.. 52, 82 Keighren, M. A., 71, 84 Keister, D. L., 195, 225 Kell, D. B., 163, 164, 175, 233 Kelley, B. C., 157, 163. 164, 166, 169, 186, 187, 188, 189, 190, 193, 195,201, 202, 203, 222, 223, 225, 226, 228 Kempf, C., 144, 154 Kennedy, C., 76,83, 85 Kennedy, L. D., 110, 121 Kenyon, G. L., 263,283 Keszthelyi, L., 176, 221 Ketchum, P. A,, 71, 83 Khan, S., 143, 151 Khanna, S., 190,226 Kiasusinyte, R. J., 143, 150 Kikina, 0. G., 161, 224 Kim, J. S., 167, 212, 213, 226, 233 King, S. M., 277, 285 King, W. R., 209,221 Kingdom, H. S., 7.86 Kinghorn, J. R., 59, 60,61, 63, 64,65, 70, 71, 84, 85

301

AUTHOR INDEX

Kinnaird, J. H., 71, 84 Kinsey, J. A., 71, 84 Kirchner, G., 171, 224 Kitamura, H., 163, 230 Klausner, R. D., 144, 154 Klebanoff, S. J., 279,285 Klemme, J. H., 162, 163, 184, 186, 195, 226 Klibanov, A. M., 178, 21 I , 226 Klimcuk. J., 61, 66, 79, 84 Klingenberg, M., 143, 150 Klipp. W., 207, 230 Knaff, D. B., 195, 226 Knecht. J., 181. 232 Kniep, H., 90, 121 Kobayashi, M., 210, 216. 226 Kohle, D., 1 1 1 , 121 Kohlmiller. E. F., Jr.. 170, 226 Kojima, N., 181, 226 Koka. P., 242, 243, 285. 288 Kolhaw, G., 36, 84 Kondratieva, E. N., 157, 160. 161, 175, 177, 178, 179, 180, 184, 185, 224. 226

Konings, W. N . , 126, 128. 129, 130, 131, 132, 133, 134, 136, 137, 140, 143, 144. 145, 146, 147. 148, 149, 151, 152, 153, 154, 176, 232 KOO,J.-Y., 244, 285 Korkowski, P. M., 40, 84 Kornberg, H. L., 7, 41, 79, 87 Kosaric, N., 157, 234 Koshland, D. E., Jr.. 141, 154 Kosower, E. M., 241, 285 Kovacs, K. L., 176, 184, 21 I . 221, 226 Kowalska, I . , 61, 79 Krasilnikova, E. N., 160. 161, 224, 226 Krasna, A. I . , 176, 177, 178, 179, 180, 182, 184, 185, 186, 223, 224, 226, 231, 232 Kratasyuk, V., 248, 25 I, 274, 285, 287 Krebs, E. G., 138, 141, 152 Kricka, L. J., 246, 274, 275, 276, 282, 285, 287, 290 Kroger, A., 181, 232 Krone, W., 143, 152 Kroneck, P., 139, 151 Kriiger, H. J., 181, 227 Kiihl, S. A., 208, 209, 227 Kuisk, I., 40, 84 Kumazawa, S., 157, 188,212, 215, 216,227. 229 Kung, H.-F., 72, 84

Kurata, M., 272, 285 Kurata, S. I., 210, 216, 226 Kurfiirst, M., 239, 240, 241, 242, 243, 284, 285

Kurland, C. G., 142, 151 Kushner, S. R.,58, 85 Kusku, S., 205,222 Kuwae, T., 272, 285 Kuznar, J., 48, 49, 84 Kwock, L., 140, 152

L Laane, C., 143, 151, 152 Labedan, B., 143, 152 Lacroute, F., I I , 12, 50, 80, 81, 82, 84 Laemmli, U. K., 106, 121 Laimins, L. A., 144, 152 Lajeunesse, D., 143. 144, 152 Lalla-Maharajh, W. V., 181. 227 Lam, C., 25,80 Lambert, G. R., 157,227 Lambert, M. A., 237,285 La Monica, R. F., 187, 227 Lamport, D. T. A., 96, 122 Lancaster, J. R., 128, 152 Lancaster, R. J., 181, 227 Lang, W., 111, 121 Langerman, N., 248, 281 Lanyi, J. K., 130, 145, 146, 152 Lara, M., 71, 81, 84 Larimore, F., 8, 10, 40, 42, 51, 84, 86 Lasko, P. F., 40, 49, 84 Latskaya, N. I., 255, 287 Laurendi, C. J., 103, 104, 121 Laurinavichene, T. V., 184, 224 Lavalle, R.,15, 81 Lawther, R. P., 19, 20, 27, 30, 80 Lax, .I.. 117, 122 Leach, F. R., 277,282 Le Comte, J. R., 165, 190, 198, 222 Lee, C. P., 139, 153 Lee, J., 242,243,244,284,285,286,288,289 Lee, J . V., 238, 271, 290 Leffler, E. M., 98, 119 Le Gall, J., 181, 182, 227, 232 Legrain, C., 7, 36, 37, 43, 84 Legrain, M., 7, 33, 36, 37, 43, 84 Le Grimellec, C., 143, 144, 152 Lehninger, A. L., 138, 141, 152

302

AUTHOR INDEX

Leisman, G., 243, 273, 285, 287 Lelie, N . , 103, 109, 122 Lemoine, Y., 25, 26, 28, 29, 30, 31, 84 Lens, P. F., 92, 100, 104, 105, 106, 107, 108, 116, 120, 121, 122 Lespinat, P. A., 183, 190, 193, 225, 233 Levenberg, B . , 24, 86 Levi, B.-Z., 279,285 Levin, E. N., 117, 118, 119 Levy, J . S., 8, 51, 86 Levy, M. A., 190, 191,229 Lewin, R. A., 92, 93, 98, 100, 103, 119, 121 Lewis, B . J., 160, 223 Lewis, C. M., 57, 84 Lewis, S. M., 165, 227 Lhoste, J.-M., 239, 285 Li, W.-B., 241, 288 Lieberman, M. A., 136, 151 Lien, S., 157, 158, 166, 167, 176, 210, 217, 227, 233 Lienhard, G. E., 55, 84 Lim, S. T., 209,227 Lindstrom, E. S., 165, 227 Linker, C., 144, 153 Linnik, L. M., 161, 226 Linthicum, D. S., 272, 289 Lipmann, F., 126, 152 Lippert, K. D., 162, 163, 227 Ljones, T., 194, 227 Ljungdahl, L . G., 174, 230 Ljungdahl, P. O., 181, 227 Llama, M. J., 177, 179, 185, 227, 231 Lloyd, D., 278, 285, 286 Lodder, J., 7, 84 Lolkema, J. S., 126, 132, 149, 151, 152, 154 Loomis, W. F., 18, 87 Losada, M., 170, 227 Love, J., 168, 233 Lowe, D. J., 193, 194, 227, 232 Ludden, P. W . , 191, 192, 197, 198, 199,200, 205,223, 225, 227, 228, 229, 230, 232 Luehrsen, K. R., 160,223 Lundie, L. L., 174, 230 Lynn, K. R., 211,222

M McCdll, J. o., 269, 286 McCann, J., 278,281 McCarron, S . H., 252, 286

McCarthy, R. E., 139, 140, 152 McCosker, J., 270,287 McCracken, S., 264,283 MacDonald, D. W . , 59, 71, 72, 75, 78, 79, 80, 83, 84 MacDonald, 1. A., 275, 288 McElroy, W . D., 237, 274,282 McEwan, A. G., 171, 173,228 McFall-Ngai, M. J., 271, 286 McLean, R. J., 103, 104, 121 Macler, B. A,, 168, 169, 215, 218, 228 Macnab, R. M., 143, I51 McNeil, M., 110, 120 Madern, D., 166, 169, 209, 214, 215,233 Madigan, M. T., 164, 171, 187,203,223,228 Madrid, V. O., 58, 82 Magana-Schwencke, N., 39, 48, 49, 84, 86 Magasanik, B . , 3, 10, 12, 14, 15, 18, 21, 24, 25, 26,49, 50,80,83.84, 86, 87, 138, 152, 205, 221, 228 Magrum, L. J., 160,223 Maier, R. J., 186, 187, 209, 210, 221, 228, 229 Makemson, J. C., 261, 262, 264, 265, 266, 267, 286 Maki, T., 210,226 Makiguchi, N., 274, 286 Malkov, Y. A., 284 Maloney, P. C., 146, I52 Mandel, M., 238, 284 Maniloff, J., 160, 223 Mannelli, D., 217, 233 Mantel, J., 277, 286 Marchalonis, J. J., 180, 228 Marechal, G., 266, 270,281 Marrs, B . , 171, 187, 204, 205, 209,227,228, 231. 234 Martin, H., 96, 104, 119, 120 Martin, J. P., 273, 286 Martin, N . C., 91, 92, 94, 115, 121 Martin, S. M., 211, 222 Marzluf, G. A., 20,58,62,63,66,67,70,75, 80, 82, 84, 86 Masters, P. S., 136, 151 Materassi, R., 167, 217,233 Matheron, R., 167, 170,228 Matheson, I. B. C., 242, 244, 286 Matsui, K.,240, 269, 284, 290 Matsunaga, T., 212, 216, 217,228, 229 Matsuura, S., 146, 153

303

AUTHOR INDEX

May, G. S., 98, 119 Mayer, H., 110, I23 Mayer, R. A., 108, 109, 123 Mayhew, S. G., 175, 185,228, 233 Mazon, J . J., 35, 84 Measures, J. C., 144, 152 Mecham, R. P., 110, 120 Mecke, D., 7.87 Meers, J. D., 10, 84. 87 Meighen, E. A., 251,254,255,262,264,274, 282, 283, 286, 287, 288, 290 Meinhart, J. O., 100, 103, 121 Meister, A., 72, 86 Melandri, B. A., 140, 146, 147, 150, 158, 187, 188, 221, 228, 234 Melkonian, M., 96, 100, 102, 121 Mellman, J. S., 96, 122 Merberg, D., 186, 209,228 Menick, M., 76, 82, 85 Merrick, M. J., 205, 228 Menitt, M. V., 249, 253,281, 286 Merzbach, D., 277, 279,281 Mesland, D. A. M., 92, 93, 95, 96, 99, 100, 101, 102, 103, 111, 116, 117, 120, 121, 122 Messenguy, F., 14, 16,85 Mevel-Ninio, M. T., 143, 152 Meyer, C . M., 188, 191, 192, 199, 200, 202, 224, 225 Meyer, G. M., 4, 31, 32,81 Meyer, J., 157, 166, 169, 187, 188, 195, 201, 202, 203,228, 233 Meyer, O., 174, 228 Mezhevikin, V. V., 238, 247, 254, 255, 261, 264, 265, 268, 289. 290, 291 Michal, G., 275, 290 Michalski, W. P., 191, 196, 197, 199, 201, 205, 209, 228, 233 Michels, J. P. J., 129, 136, 152 Michels, P. A. M., 128, 129, 130, 133, 136, 151, 152 Middlehoven, W. J., 7, 12, 13, 15,20,23,34, 85 Mileham, A. J., 249, 256, 257, 282 Miller, D. H., 96, 122 Miller, E., 270, 287 Miller, M., 96, 122 Mills, J. D., 140, 152 Mitchell, P., 126, 127, 128, 131, 132, 134, 140, 142, 143, 152, 154 Mitchell, W. J., 134, 150

Mitronova, T., 205, 207, 231 Mitsui, A., 157, 167, 170, 188, 210,212,213, 215, 216, 217, 227, 228, 229 Miyake. J., 213, 229 Miyamoto, C. M., 264, 283 Miyamoto, K., 157, 210, 212, 221 Molenaar, E. M., 104, I 2 1 Monk, B. C., 91, 110, I l l , 119, 121 Monod, J., 5, 36, 80, 85 Monosov, E. G., 161,226 Monty, K. J., 195, 196, 225 Moore, W. S., 102, 112, 117, 122 Mora, J., 67, 71.81. 82, 83, 84, 86, 87 Mora, Y.,71, 81 Morin, J . G . , 266, 270, 288 Moroney, J. V., 139, 140, 152 Morse, D., 274, 283, 286, 288 Mortenson, L. E., 157, 174, 175, 182, 183, 189, 190, 193, 194, 220, 229. 234 Moss, C. W., 237, 285 Moura, I., 181, 182, 227, 232 Moura, J. J. G., 181, 182, 227, 232 Mousset, M., 38.82 Muller, F., 242, 244, 286 Miiller, P., 207, 230 Miiller, W. E. G., 115, 122 Munson, T. D., 198,229 Murray, R. G. E., 158, 159,224 Murrell, S. A., 199, 228 Musgrave, A., 92,93,96, 103, 104, 105, 106, 107, 108, 109, 112, 113, 114, 115, 121, 122 Mutafschiev, S . , 210, 228

N Nakamura, H., 162, 163, 164, 171, 173,229, 234 Nakamura, K., 144, 151 Nakamura, T., 240, 247, 253, 260, 26 I , 269. 273, 284, 286, 290 Nakos, G., 175, 183, 229 Nanninga, N., 100, 116, 120 Nano, F. E., 146, 147, 151 Nason, A., 71, 81 Naveh, A., 279, 286 Nealson, K. H., 236,237,238,241,242,243, 249,254,256,257,258,259, 260,263,265. 266,267,268,269,270,271, 272,273,282, 283, 284, 285, 286, 287, 288, 289 Nederbragt, A., 107, 116, 121

AUTHOR INDEX

304

Nefsky, B., 246, 287 Nelson, M. J., 190, 191, 229 Nelson, S. O., 142, 152 Neumann, C., 205,222 Neupert, W., 143, 144, 154 Newmeyer, D., 66, 85 Neyssel, 0. M., 149, 151 Nicholas, D. J. D., 163, 189, 190, 191, 196, 197, 199,223. 226, 228

Nichols, J. W., 127, 150 Nicoli, M. Z., 248, 267, 268, 281, 283 Nix, D. W., 208, 209, 227 Nokhal, T. H., 186, 229 Nordlund, S., t91, 197, 198, 200, 223, 229, 234

Noren, A , , 200, 229 North, M. J., 75, 85 Novick, A., 5, 85 Nozaki, M., 170, 227 Nultsch, W., 140, 152 NyrCn, P., 143, 153

0 O’ara, E. B., 186, 228 O’Brian, M. R., 187,229 O’Brien, C. H., 266, 269, 287 O’Brien, D., 257, 283 Odom, J. M., 166, 168, 169, 189, 209, 210, 214, 216, 221, 229, 232

Oelze, J., 149, 152 Ogden, R. C., 249, 256,282 Ohta, Y., 157, 167, 170, 212, 213. 215, 229 Ohwada, K., 266, 270,287 Okon, Y., 191, 227 Oliva, L., 157, 212, 215, 229 Olofsen, F., 107, 108, 116, 121 Olson, J. M., 195, 211, 221, 229 Oppenheimer, N. H., 263,283 Orme-Johnson, H., 199, 227 Orme-Johnson, W. H., 181, 190, 191. 225, 226, 229

Ormerod, J . G., 162, 165, 166, 170,224, 230 Ormerod, K. S., 165, 166, 170, 224, 230 Ormos, P., 176, 221 Orndorff, S. A., 270, 287 Osborn, M.,106, 123 Ossendorp, F. A., 112, 113, 114, 122 Otto, R., 126, 137, 148, 149, 151, 153, 154 Owens, M. S., 71, 83 Oxender, D. L., 144, 150

P Padan, E., 143, 146, 147, 150, 153. 154 Pagani, S., 139, 151 Pagano, M., 161, 210, 222 Page, M. M., 64,79 Palacios, R., 71, 81, 84, 86 Palmer, G., 194, 234 Panopoulos, N. J., 205, 234 Pappenheimer, A. M., Jr., 5, 36, 85 Paranchych, W., 142, I50 Partridge, C. D. P., 181, 209, 230 Pastan, 1.. 18, 85 Patel, V. B., 58, 85 Pateman, J . A., 58,59,60, 61,62.63,64,65, 70, 71, 81, 83, 84, 85, 86

Paul, F., 187, 188, 217, 230 Paul, v. J., 238, 287 Pavlova, E. A,, 187, 232 Payne, R. W., 7, 79 Peck, H. D.. Jr., 181, 182, 227, 232 Pedrosa, F. 0.. 181. 230 Pelroy, R. A., 168, 169, 215, 218, 228 Pemberton, J. M., 204, 205, 209, 230 Penfold, H. A., 72, 78, 79 Penninckx, M., 14,85 Perkins, D. D., 66, 85 Perlman, R., 18, 85 Perry, H., 277, 286 Peterkofsky, A,, 143, 144, 153 Petushkov, V. N., 248,287 Petushkova, Y. P., 161, 226 Pfennig, N., 156, 157, 158, 159, 160, 161, 162, 163, 164, 167, 171, 186, 188,201,203, 205,224, 225, 227, 230, 231, 232 Piccard, J., 266, 270, 281 Pierard, A,, 12, 84 Pijst, H. L. A., 96, 112, 113, 114, I 15, 122 Pinsky, M. J., 165, 227 Piotrowska, M., 9, 1 I , 34.50, 58.66, 72, 79, 82 Plack, R. H., Jr., 147, 153 Plummer, J., 103, 123 Poinar, G. O., Jr., 238, 287. 289 Polkinghorne, M., 60, 61, 85 Polya, G. M., 59, 85 Ponta, H., 143, 154 Poo, M., 112, 120 Poolman, B., 145, 146, 147, 148, 151. 153 Pootjes, C. F., 209, 214, 230 Pope, M., 199, 228

305

AUTHOR INDEX

Popova, L. Y., 255. 259, 265, 287, 288 Porte, F., 176, 183, 187, 223. 230 Porter, J. K., 277,291 Postasam, E., 279. 286 Postgate, J . R., 157, 200, 201, 209.223, 230 Postma, P. W., 142, 152, 153 Potrikus, C. J., 238, 257, 263, 267, 270, 271, 280, 283, 287 Potter, S. A., 267, 289 Powers, D. A.. 249, 256,281 Prernakumar, R., 58, 66, 71, 85 Presswood, R. P., 239, 240, 241, 284. 285, 287 Preston, G. G., 191, 192, 198, 199,223, 227, 230 Priefer. U., 207, 230 Primakova, G. A., 238,287, 289 Prince, R. C., 187, 195, 229. 230, 234 Pringle, J. R., 106, 123 Pugh, C. S. G., 58, 82 Piihler, A,, 207, 230

Q Qadri, S. M . H., 162, 173, 230 Quay, S. C., 144, 150 Quinto, C., 71, 86 Quinto, M., 205,221

R Radford, A., 16, 66, 82, 85 Radway, J., 157, 212, 215, 229 Raff, M. C., 99, 123 Ragsdale, S. W., 174, 230 Rak, E.. 176, 211, 221 Ramos, F., 11, 12, 33, 85 Ramos, S., 134, 153 Rand, K. N . , 60. 63, 64,65, 72, 73, 77, 78, 79, 85 Randall, J . , 93, 102, 122 Rao, K. K., 177, 179, 181, 185, 194, 227. 231, 232 Rausch, S. K., 248, 249, 281, 282, 287 Ray, D. A., 96, 122 Rayburn, W. R.. 90, 122 Reeve, C. A., 248, 267, 268, 269, 281, 287 Reichelt, J . L., 238, 271, 282 Reidl, H., 204, 230 Reinart, W. R., 62, 86 Reinhertz, A., 261, 262. 289

Reitzer, L. J . , 3, 86 Rever, B. M . . 58, 85 Rhoads, D. B., 136, 142, 144, 152. 153 Richarme, G., 143, 153 Riendeau, D., 254, 255, 287, 288 Rigaud, J. L., 143, 144, 152 Rigopoulos, N., 173,221 RihovB, L., 49, 83 Riley, P. L., 248, 249, 281, 284 Riolacci, C., 166, 168, 203, 204, 215, 217, 225 Roberts, G. P., 201. 202, 206, 230 Roberts. K., 108, 121 Robey. E. A.. 67, 71, 82 Robillard, G. T., 126, 131, 135, 140, 143, 144, 145, 146, 151, 152, I53 Robinson, D. G., 96, 122 Robson, R. L., 157, 200, 201, 209, 230 Roda, A., 275,287 Rodicheva. E. K., 255, 260, 289 Rodriquez, A., 254, 255. 282, 287, 288, 290 Roelofsen, P. A., 162, 230 Roman, M., 270, 287 Ronzio, R. A.. 72, 86 Roon, R. J., 4, 8, 10, 24, 31, 32, 35, 40. 42, 51, 81, 83, 84, 86 Rose, A. H., 6, 52, 82, 86 Rose, C., 139, 151 Roseman, S., 112, 114, 122 Rosen, B. P., 147, 153 Rosenbaum, J . L., 100, 103, 120, 123 Rosner, D., 157, 212, 215, 229 Rosson, A.. 263,288 Rottem, S., 144, 153, 278, 289 Rottenberg, H., 134, 153 Rowe, W. B., 72,86 Ruby, E. G., 242, 248, 249, 251, 266, 269. 270, 287. 288 Russell, S. A,, 209, 221 Ruvkun, G. B.. 191, 230 Rydstrom, J., 139, 143. 151, 153 Ryerson, C. C., 241, 288 Rytka, J.. 5 , 40, 86

5 Saari, L. L., 199, 228 Sachs, H., 96, 122 Sager, R., 91, 122 Saier, M. H., 141, 150 Saint-Jean, L., 161, 210, 222

306

AUTHOR INDEX

St. John, A. C., 144, 154 Sanchez, F., 71, 86 Sander, G., 103, 123 Sandlin, D. E., 205, 234 San Pietro, A , , 157, 210, 227 Sarma, Y. S. R. K., 91, 122 Sasaki, T., 272, 285 Satoh, T., 163, 164, 230 Saunders, V. A., 195, 225 Sawada, E., 164, 230 Scazzocchio, C., 58, 60, 61, 76, 77, 79, 82, 86 Schackman, R. W., 118, 122 Schaechter, M., 143, 144, 150 Schaller, K., 143, 144, 150 Schechter, Y., 117, 122 Scherings, G., 201,231 Schick, H. J., 170, 189, 190, 231 Schlegel, H. G., 126,153, 157, 162, 174, 175, 186, 209, 221, 222, 223, 226, 228,231, 232 Schlessinger. J., 117, 122 Schlimme, E., 140, 150 Schlosser, U. G., 96, 97, 122 Schmedding, D. J. M., 182, 220 Schmeisser, E. T., 91, 122 Schmidt, A,, 139, 153 Schmidt, J. A,, 118, 122 Schmidt, M. R., 141, 150 Schmidt, S. P., 244, 285 Schneider, K., 157, 174, 175, 180, 223, 231 Schneider, S., 9, 83 Schoelmerich, J., 275, 288 Schol, D., 103, 109, 122 Scholte, B. J., 142, 153 Schon, G., 170, 171. 172, 173,205,225,231, 233 Schonheit, P., 180, 232 Schram, E., 274, 275, 283, 288 Schreiber, A. B., 117, 122 Schuitema, A . R. J . , 142, 153 Schuldiner, S., 146, 147, 153, 154 Schultz, J. E., 171, 173,231 Schurmann, P., 140, 152 Schuster, G . B., 244, 285 Schwab, J. M., 241, 288 Schweiger, M., 142, 143, 154 Schweizer, M., 58, 85 Schwencke, J., 39, 48, 49, 84, 86 Scolnik, P. A., 202, 205, 206, 207, 208,221, 23 I

Sealy-Lewis, H. M., 60, 79 Seibert, M., 157, 158, 166, 167, 176, 210, 217, 233 Seijen, H. G., 176, 232 Selvaraj, G., 209, 221 Serebraikova, L. T., 175, 176, 177, 178, 179, 180, 184, 185, 224, 226, 231 Serra, J. L., 177, 179, 185, 227, 231 Shaffer, P. M., 58, 60, 61, 75, 76. 82, 86 Shah, V. K., 191, 231 Shakhov, Y. A , , 143, 153 Shanmugam, K. T., 195, 231 Shapiro, B. M., 7, 86, 118, 122, 141, 154 Shcherbakova, G. Y., 260,289 Shenderov, A. N., 255, 259, 265, 287. 288 Sherman, F., 23, 31, 82 Shestakov, S., 205, 207, 208, 231 Shi, G.-Y., 144, 153 Shilo, M., 259, 269, 270, 276, 288, 289, 291 Shimamoto, I., 210, 226 Shimomura, O., 242,288 Shine, J., 256, 288 Shioi, J.-I., 146, 153 Shipman, R. H., 210, 231 Shou-Ih, J. U., 174, 223 Shumikhin, V. N., 247, 248, 288 Shyam, R., 91, 122 Siefert, E., 164, 167, 186,201,203, 205,225, 23 I Sies, H., 140, 153 Silverman, M., 254, 256, 257, 258, 282, 283 Silverman, M. P., 145, 146, 152 Sim, E., 163, 164, 175, 176, 187, 231, 233 Simon, M. I . , 249, 256, 257, 282 Simon, R., 207, 230 Simon, R. D., 259, 269, 288 Simpson, F. B., 209, 230 Sims, A. P., 36, 82, 86 Sire, J., 40, 87 Sistrom, W. R., 157, 222 Sizemore, R. K., 266, 269, 286, 287 Skulachev, V. P., 126, 130, 143, 150, 153 Slater, E. C., 126, 128, 153, 175, 176, 179, 181, 182, 220, 221, 232 Slessor, K. N., 274, 283, 286 Slonimski, P. P., 6, 82 Sloof, W., 277, 282 Sly, W., 40, 87 Small, E. D., 243, 288

I

307

AUTHOR INDEX

Small, L. F., 270, 271, 281 Smith, B. E., 193, 227 Smith, D. N., 256, 291 Smith, G. D., 157, 169, 222, 227 Smith, R. V., 191, 223 Smith, W. P., 143, 144, 153 Snell, W. J . , 9 6 , 9 7 , 102, 112, 113, 114, 117, 122 Snyder, P. M.,205, 221 Soberon, X., 144, 151 Solioz, M., 204, 231 Solter, K. M., 96, 97, 98, 114, 116, 122 Sommerville, C. R., 141, 154 Song, H., 166, 188, 231 Sonnenberg, A. S. M., 137, 153 Sorger, G. J., 58, 63, 66, 71, 85, 86 Spath, S., 99, 120 Spathas, D. H., 59, 62, 86 Spudich, J. L., 141, 143, 154 Stackebrandt, E., 160, 223, 224 Stadtman, E. R., 7, 36, 86, 141, 154 Staehler, F., 275, 290 Stahl, D. A., 160, 223 Stanier, R. Y., 126, 154 Stanley, P. E., 274, 285, 288 Starling, D., 93, 102, 122 Starr, R. C., 91, 92, 121, 122 Stein, J. R., 90,122 Steinback, K. E., 140, 148, 150 Stemrnler, J., 238, 263, 267, 271, 283 Stephenson, M., 174,231 Stevens, P., 214, 215, 231 Stickland, L. H., 174, 231 Stoeckenius, W., 141, 143, 154 Stoppani, A. 0. M., 190, 231 Stossel, T. P., 279, 290 Stouthamer, A. H., 149, 151 Streips, U . N., 143, 144, I5I Strekas, T., 178, 179, 180, 231, 232 Sugasawa, M., 130, 154 Sumrada, R., 26, 54, 80, 86, 87 Surdin, Y., 40, 87 Suzuki, K., 239,288 Swanson, R., 250, 288 Sweeneyn, W. V., 194,234 Sweet, W. J., 196, 232 Sy, J., 141, 154 Sybesma, C., 171, 232 Szilard, L., 5 , 85 Szittner, R. B., 274, 283, 286, 288

T Tabor, P. S., 266, 270, 287 Taguchi, M., 142, 154 Tait, R. C., 209, 221, 227, 232 Takahashi, H., 167, 212, 213, 226, 233 Takakuwa, S., 166, 169, 181, 189, 209, 210, 214, 232 Takimoto, A., 247,290 Tamiya, N., 183, 234 Tanaka, A., 216,223 Tanaka, M., 194, 232 Tanner, R. S., 160, 223 Tappen, D. C., 58, 82 Tatannova, N. Y., 161,224 Tavlitzki, X., 6, 82 Taylor, B. L., 143, 154 Taylor, P. B., 277, 282 Taylor, R. B., 99, 123 Tchan, Y.T., 210,226 Tebo, B. M., 238, 270, 272, 273, 284, 287, 289 Tecoma, E. S., 143, 144, 154 Teixeira, M., 181, 182, 227. 232 Telfer, A., 191, 223 Tempest, D. W., 10.84, 87 ten Brink, B., 134, 137, 148, 153, 154 Tenorio, M., 67, 82 TCtu, C., 141, 143, 154 te Welscher, R., 92, 104, 105, 106, 107, 122 Thauer, R. K., 129, 154, 180, 181,220, 224, 232 Thomas, G. M., 238,287. 289 Thomas, L. P., 241, 288 Thore, A., 274, 289 Thorneley, R. N. F., 183, 190, 193, 194,229, 232 Thorner, J., 90, I20 Thorpe, G. H., 274. 275, 285, 290 Tinker, K., 176, 233 Togasaki, R. K., 116, 120 Tokuda, H., 130, 154 Tollervey, D. W., 59, 60, 64,68, 69, 70, 73, 74,87 Tomizuka, N., 213,229 Tomsett, A. B., 62,63,66,67,71,77,81,82, 87 Toone, J., 36, 86 Toranzo, A. E., 267,289 Tosa, T., 216,222

308

AUTHOR INDEX

Trainor, F. R . . 92, 123 Tratasynk. G. A , , 24X, 287 Tredici, M. R., 167, 217, 233 Trierner, R. E.. 94. 123 Trinci, A. P. J.. 97, 120 Triplett, E.. 199. 228 Triplett, E. W.. 200, 205, 232 Truper, H. G., 156, 157, 158, 159. 160, 162, 225, 230. 232 Tsuchiya, T., 239, 288 Tsygankov, A . A,, 187, 232 Tu, S.-C.. 240, 242, 246, 249, 251, 252, 253, 268,281, 284. 286, 289, 290 Turoscy, V., 25, 28, 51, 80. 87 Turova, T. P., 238, 287 Tyler, B., 18, 87

U Uffen, R. L.. 170. 171, 172, 173. 174. 224, 232, 233 Ulane, R. E., 97. I20 Ulitzur, S., 253,254,258,259,260,261,262, 263,267,274,277,278,279,280,28/ 283, 285, 286. 289, 290 Unden, G., 181, 232 Underwood. C., 139, 140, 154 Unemolo, T., 130, 154 Urrestarazu, L. A, , 6, 12, 14, 15, 83, 87 ~

v Vaca. G., 67, 71. 82, 87 Valentine. R. C., 143.152. 176,209,227,233 Vallejos, R. H.. 139, 140, 152 van Belkum, M., 146, 147, 148, 151 van Dam, K., 146, 149, 150, 151, 154 van den Briel. M. L.. 103, 107, 109.121. 122 van den Ende, H., 90,92,93,95,96,99, 100, 103, 104, 105, 106, 107, 108, 109, I 1 I , 112, 113, 114, 115, 116, 120, 121. 122. 123 van de Poll, K. W., 28. 87 van der Drift, C., 25, 87 van der Meer, R., 146, 154 van d e r Plas, J., 176, 232 van der Westen, H. M.. 175, 233 Van Det Sypt, H., 214, 215, 231 van Egmond, P., 108, 109, 112, 113, 114, 120, 121 Van Eijk, J., 23, 34, 85

van Eyk. E., 92, 104, 105, 106, 107, I22 van Heerikhuizen, H., 175. 176, 179. 181, 232 van Niel, C. B., 157, 163, 170,232 Van Nieuwenhoven, M. H., 143, 154 Van Renesse, R., 23, 34, 85 van Renswoude, J . , 144, 154 van Rheenen, P. S., 175, 176, 179. 181, 232 Vanwinkle-Swift, K. P., 91, 123 Varon, M., 276, 289 Veeger, C., 143, 151, 152. 175, 201,231, 233 Veldkamp, H., 137, 148. 149, 152, 153 Venerna, G., 143, 154 Verma, J . , 215, 233 Versluis, C., 110, 120 Videletz, 1. Y.,259, 265, 288 Vignais, P. M., 157, 163, 164. 166, 168, 169, 174, 175, 176, 177, 17X, 179, 180, 181, 183. 184, 185, 186, 187. 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 199,200,201,202. 203,204,205,207,208,209,2 10.2 13.2 14, 215,217,221,222,223,224,225,228,230, 231, 233 Vincenzini, M., 167, 217, 233 Virosco, J., 202, 207, 208. 231 Vishniac, W., 162, 233 Visser, A. J., 243, 289 Visser, J. W. G., 243, 284 Vissers, S., 6, 7, 12, 14, IS, 17, 19, 25, 28, 29,30,3 I , 34, 35, 36, 37,43.81,83,84.87 Vliegenthart, J. G. F., 109, 110, 120 Voelskow, H., 171, 172, 173. 205, 231, 233 Vogels, G. D., 25, 87 Von Stedingk, L.-V., 143, 150 Vorobyeva, T. I., 238, 265, 287, 289 Vrati, S., 215, 233 Vysotskii, E . S., 247, 254, 255, 260, 261, 264, 265, 268, 28Y. 290

W Waffenschmidt, S., 97, 121 Wagner, C., 72, 84 Wagner, E. F., 142, 143, 154 Wakirn, B. T., 174, 233 Walker, C. C., 189, 233 Wall, H. D., 205,233 Wall, J . D., 166, 168, 169, 181, 189, 196,200. 202,203,205,207,208,209,210,213,214, 216, 221, 228, 229, 232. 233

309

AUTHOR INDEX

Wall. L., 254. 255, 288, 290 Wall. L. A., 254, 290 Walsh, C., 1x1, 226, 241, 288 Walsh, C. T., 181, 225 Walz, D., 145, 154 Wang, J . Y. J . , 141, 154 Wannlund, J., 274. 275, 290 Ward, W. W., 243, 290 Warren, A. J., 110, 120 Watanabe. H., 246, 247, 252, 260, 261, 273, 275, 290 Watanabe, K . I., 167, 212, 233 Watanabe, T., 240, 290 Waters, C. A,. 264. 265. 290 Watson. J . D., 5, 87 Watts, C., 143, 144, 154 Weare, N . M., 207. 208, 213, 233 Weaver. P., 176, 233 Weaver, P. F.. 157, 158, 166, 167, 171, 173, 176, 205, 210, 217, 231, 233 Weber, G., 207, 230 Weber. K., 106. 123 Weckesser, J . , 110, 123 Weelall, H. H., 211, 221 Weglenski, P., 5 8 , 61, 66, 72, 79, 84 Weiner, J . H., 176, 232 Weinhausen, G. K., 276, 285 Weinstein, J. N., 144, 154 Weiser, I., 258, 259, 278. 279, 289. 290 Weiss, R. L., 14, 87, 93, 96, 115, 116, 120, 123 Weitzman, S. A,. 279, 290 Welches, W. R., 251, 252, 290 Wellner, V . P., 72, 86 Weltman, J . K., 180, 228 Wertlieb, D., 162, 233 West, I. C., 134, 143, 154 West. P. A.. 238, 263, 267, 271, 283, 290 Westerhoff, H. V., 126, 146, 149, 151. 154 White, D. C., 277, 285. 286 Whitehead, T. P., 274. 275, 285, 290 Whitney, P. A,, 21, 25, 26, 87 Wiame, J.-M., 4 , 5 , 6 , 7 , 8 , 1 1 , 12, 13, 14, 15, 16, 17, 18, 19,20,21,22,23,24,25,26,27, 28,29,30,3 I , 3 3 , 3 4 , 3 5 , 3 6 , 3 7 , 3 8 , 4 2 , 4 3 , 80, 81, 82, 83, 84, 85, 86, 87 Wickerham, L. J., 6, 87 Wickner, W., 143, 144, 154 Wiemken, A., 7, 87 Wienhausen, G. K., 246, 275, 290

Wiese, L., 91, 93, 103, 108, 109. 112, 119, 120. I23 Wiese, W., 93, 103, 109, 112, 123 Wilkins, S. A,, 67. 80 Williams. J . . 278, 285, 286 Williams, L. A., 108, 123 Williams. N . , 278. 285, 286 Williams, R. E., 211, 222 Willison, J. C . , 157, 166, 168, 169, 201, 203, 204,205,207, 208,209.214.215. 217.225, 233 Wilson, P. W., 165. 227 Wilson, T., 242, 290 Wilson, T. H., 144, 153 Wingard, L. B., 216, 222 Winter. H. C., 190,233 Witrnan. G. B., 103, 123 Woese, C. R.. 160, 223, 224 Wolf, D. H., 36, 87 Wolfe, D., 185, 186, 187, 221 Wolfe, R. S., 160, 171, 223, 232 Wood, H. G., 174, 223 Woodward, J. R.. 41, 87 Woolkalis, M. J., 237, 238, 281, 282, 291 Wright, J . K., 142, 152 Wu, D., 143, 144, 154 WU, M . , 166, in8,231 Wulff, K., 7, 87, 274, 275. 290

X Xavier, A. V., 181, 182, 227, 232

Y Yagen, B.. 278,289 Yagi, T., 183, 234 Yakunin, A. F., 187, 194, 195, 232, 234 Yamagata, S., 163, 234 Yamanaka, T., 247,290 Yamashita, J . , 175, 225 Yamaski, E., 278,281 Yang, Y., 238, 291 Yannai, S., 258, 259, 278, 279, 289, 290 Yanofsky, C., 265, 291 Yarden, Y., 117, 122 Yarrow, D. Y.,7, 79 Yasunobu, K. T., 194,232 Yates, I. E., 277, 291 Yates, M. G., 181, 189, 209, 230. 233

31 0

AUTHOR INDEX

Yeh, J . , 141, 154 Yen, H. C., 171, 204, 205,231, 234 Yetinson, T., 270, 288, 291 Yike, N . J., 195, 225 Yoch, D. C., 157, 165, 191, 192, 194, 195, 196, 197, 198, 199,200,202,208,209,222, 224, 227, 234 Yotsuyanagi, Y . , 6, 82 Youvan, D. C., 205, 234 Yu, B., 166, 188, 231

z Zablen, L. B . , 160,223, 224 Zacharscki, C. A , , 87 Zaharchuk, L. M., 161,226 Zajic, J. E., 157, 234 Zale, S . E., 211, 226 Zannoni, D., 158, 171, 187, 188, 221, 228, 234

Zavoruev, V. V., 238, 247, 254, 255, 261, 264, 265, 268,287, 289, 290, 291 Zeeb, D. B., 71, 83 Zhang, Z. M . , 197, 234 Zhukov, V. G., 161, 226 Zick, Y . , 117, 121 Ziegler, M. M., 236,241, 243, 249,253,256, 281, 291 Zilberstein, D., 146, 147, 153, 154 Zilver, R. J., 96, 115, 122 Zimmermann, R., 143, 144, 154 Zimmermann, U., 144, 134 Zorin, N. A., 175, 177, 178, 179, 180, 184, 185,221,224, 234 Zsebo, K., 205,234 Zubenko, G. S., 36, 83 Zumft, W. G., 169, 194, 195, 196, 197, 200, 221, 234 Zurowski, W. K . , 71, 81 Zurrer, H., 166, 213, 215,233, 234 Zyskind, J. N., 256, 291

Subject Index A Acetamidase, 75, 78 N-Acetylglutamate kinase, 16 N-Acetylglutamylphosphate reductase, 16 Acetyl phosphate, 129, 136 Acyl-CoA-reductase, 255 Acyl-protein synthetase, 255 Adenosine triphosphatase (ATPase), 128129, 139 CaZ+,Mg'+-stimulated, 130, I3 I chloroplast, 146 membrane-bound, proton motive force generated, 136 membrane-bound energy-transducing complex of bacteria, 140 Adenosine triphosphate (ATP) energy-currency function, 126 hydrolysis, for proton motive force generation, 137 phosphorylation potential, 142 synthesis, 131 Adenosine triphosphate-dependent solute transport systems, 136 Alanine, 147 Alarmosome, 142, 144 Alcohol, decanol, 253 Aldehyde biosynthesis, 253-255 Aldehyde dehydrogenase, 254 Aldo-acid, 261 Alkylamines, 32 Alkylammonium ions, 72 N-Alkylmaleimedes, 253 Allantoinase, 27 Allantoin-urea degradation, 24-3 I ammonia effect, 27-28 biochemistry, 25-26 genetics, 25-26 glutamate dehydrogenase, 27-28

31 1

induction, 26 nitrogen catabolite repression more than one circuit? 28-29 starvation effect, 26-27 Allophanate, 26 Altermonas genus, 238 Amino acids arginase synthesis provokinghon-provoking, 21 biosynthesis, alternative routes, 75-76 Ammonia, 8 conversion into glutamate, 20 effect, 27 on amino acid permease, 18, 27 glutamine synthetase repression, 35-36 nitrogen catabolite repression, 20 uptake, 8-9, 51-52 Ammonia-sensitive permeases, 54 Ammonium analogues C s + , 72 Ammonium repression, 57, 71 Anox yphotobacteriae, 158, I59 (table) anaerobic CO-uptake, 173 Antibiotics, inhibition of luminescence induction by, 279 Arginase, 13, 15-22 ammonia effect, 27 induction mechanism, 21 nitrogen catabolite repression, 16- I8 metabolic signal, 19-20 release in NADP+-glutamate dehydrogenase mutants, 18-19 non-specific induction, 21 synthesis, 22 nitrogen starvation effect, 20-2 I Arginine, 32 degradation, 12-24 pool, 22 L-Arginine, 37, 38 (table) Arginine permease, 37

SUBJECT INDEX

312

Ascomycetes, 2 Asparaginase I , 12, 31 Asparaginase 11, 31-34 L-Asparagine 12, 5 I DL-/%AspartylhydroxamaIe, 72 A.\pergillu.s nidulans, 57 allele circ,A-102, 61 allele xprD-1, 61-62 carbon catabolite repression, 74-75 cis-acting regulatory mutations, 76-78 gene crreA, 59-61 gene t w A , 74-75 gene gdhA, 70 gene nirA, 72-74 glutamate dehydrogenase, NADP-linked, 70

I.-glutamine as nitrogen metabolite corepressor, 70-71 mutation urrA-102, 76 mutation ureAd, 60-62 mutation ureB, 68-70 mulation nis-5, 77 mutation imp-100, 76-77 nitrite reductase in, 77 oxygen repression, 76 pathway-specific regulatory genes, 72-74 phosphorus regulation, 75 pyrimidine auxotrophs, 60 sulphur regulation, 75 wild type. 60 ,see ulso Nitrogen metabolite repression Autolysin. 96-97 Auxotrophies supplementation, 75-76

B Bwi1lrJ.s suhtilis. autolytic enzymes, 144 Bacteria bioluminescent see Bioluminescent bacteria brown sulphur, ecological distribution, 161 energy circuit, 126-130 energy requirements, 125-126 green, 158, 159 (table,n.) ferredoxin photoreduction, 194- 195 green sulphur, 159 (table,n.), I60 ecological distribution, I 6 I intracellular pH, 146

photosynthetic see Photosynthetic bacteria primary/secondary transport systems, 128, 129 (fig) processes energizedlregulated by proton motive force, 143 (table), 144 purple, 158, 159 (table,n.) dark, fermentative metabolism. 170 nitrate reduction. 164 nitrogenase activity, 195 purple non-sulphur, 147, 159 (table,n.), 160

ecological distribution, 162 nitrate reduction, 164 purple sulphur, maximal rate of H2 photoproduction, 167 (table) reconstituted lyophilized luminous, 277 Bacteriorhodopsine, I30 Beneckea genus, 237 b-flavin, 268-269 Binding-protein-dependent transport systems, 136 Bioluminescent bacteria. 236-291 arginine requirement, 264-265 auto-induction, 263-264 biochemistry, 238-256 catabolite repression, 265-266 chemosensory behaviour, 262-263 chemotaxis, 262-263 continuous/pulsed emission, 259-261 ecology, 269-273 host-associated bacteria, 27 1-273 planktonic bacteria, 269-27 I electron flow, 260-261 emitter, 242-244 precursors, 244 fish light organ symbiosis, 272-273 immune assays, 274 inhibitors of bioluminescence in uiuo. 26 1-262

molecular biology, 256-259 mutants, 278-279 acid-/aldehyde-requiring, 278 bioluminescence test, 279, 280 (table) regulation defective, 278-279 physiology, 259-269 taxonomy, 237-238 translation in uitro, 263-264 see ulso Luciferase

SUBJECT INDEX

C CAMP phosphodiesterase, 144 Cundidu utrilis, 35, 36 Carbon catabolism, 7 Carbon catabolite repression, 74-75 Casamino acids, 3 I Catabolic synergism, 21 Chlamydomonus spp. flagellar interaction between mating cells, 93 flagellar surface, 103 flagellar surface motility, 98 flagellar tip activation, 99-102 gametic fusion cf. that of green algae, 96 isoagglutinins, 103-104 Chlamydomonas capensis, 91 Chlamydomonus eugumetos, 90 agglutination factor m t + , 113, I15 agglutination factor mrr, 110, 113, 115 cell wall release, 96 flagellar interaction among mating cells, 92, 93 flagellar surface outgrowths, 103 flagellar surfaces in gametes, mt+-mt interaction, 109 flagellar tips, 99, 100 (fig) gametogenesis, 9 1-92 isoagglu tin ins, 103- I04 monosaccharides, 109-1 10 0-methylated sugars, 109-1 10 mating structure activation, 94, 96 primary zygote membrane, 93 sexual agglutination, 92-93, 116-1 18 receptor activation, 112-1 16 sexual adhesion mechanism, I 11-1 12 signalling action, 116-1 18 Chlamydomonas gymnogarno cell wall release, 96 Chlumydomonus moewusii, 90 Chlumydvmonas monoicu, 91 Chlamydomonas winhardtii, 90 agglutination factor m t + , 110- I I . 3 agglutination factor m f r , 113 autolysin, 96-97 CaZ+in flagellar signalling, I17 lidocaine interference, 117 calcium efflux, 118, I18 (fig) cell wall release, 96-98

31 3

fertilization tubule elongation, 94, 95 (fig) flagellar interaction among mating cells, 92, 93 flagellar membrane agglutinins effect, 100, 101 (table) flagellar regeneration experiments, 114 flagellar surface motility, 98 flagellar tip, 98-99, 99-102 FTA blocking, I 0 0 gametogenesis, 91 mating structure activation, 94, 96 non-agglutinative mutants, I16 sexual agglutination, 92-93, I 1 1-1 16 receptor inactivation, 112-1 16 sexual adhesion mechanism, 11 1-1 12 signalling action, 116-1 18 Chlamydomonas zimhuhwiensis, 91 Chlorate, 72 Chlorobiaceae, 160 ecological distribution, 161- I62 growth properties, 161 Chlorobiales, 158, 159 (table), 160 Chloroflexaceae, 160 Chloroplast(s), 148 Chloroplast ATPase, 146 Chloroplast-membrane proteins, phosphorylation of, 140 Chlorosomes, 160 Chromatiaceae, 160 growth properties, 161 hydrogen uptake stimulation, 164 maximal rate of Hz photoproduction, 167 (table) nitrogenase-mediated Hz evolutioncarbon metabolism relationship, 170 cis-acting regulatory mutations, 76-78 Citrate synthase, 139 Cow dung, hydrogen produced from, 214 Cyanobacteriales, 159 (table) hydrogen metabolism, literature reviews, 157 (table) oxygenic photosynthesis, 157 solar energy conversion into hydrogen, 21 1 Cyclohexanone mono-oxygenase, 241 Cycloheximide, 35 Cytochrome P-450, 247

SUBJECT INDEX

31 4

D Deburyornyces, 2 Decanol, 253 Dehydrogenases, anaerobic CO, 174 6-Diazo-5-oxo-2-norleucine, 197 I-Diazo-2-oxoundecane, 252 Dimethylalanine, light emission inhibited by, 247 6,7-Dimethyl-8-ribityllumazine, 242-243 Dimethyl sulphoxide, 171 Dinitrogenase, 190 Dinitrogenase reductase, 190 2,2-Diphenylpropylamine,252 Dithio-disulphide groups, redox-sensitive, I45 Dodecanol, 253

E Ectothiorh(~d(~.~phiru, I59 (table ,n.) Ectuthiorhodospiru mobilis, growth property, 161 Ectothiorhodospiru shuphoshnikouii, growth property, 161 Elect roc hemical potential gradients. 142I46 Electrogenic proton pumps, 130, 131-132 Electron transfer systems, 130, 131 Energy circuit, bacterial, 126-130 Energy intermediates regulation, 138-146 electrochemical potential gradients, 142I46 phosphorylation potentials, 140- 142 redox potentials, 138-140 Energy requirements, bacterial, 125- 126 Energy transduction in cytoplasmic membrane, 130-138 ATP-dependent solute transport systems, 136 group translocation, 134 primary transport systems, 131-132 proton motive force generation by end product efflux, 136-138 secondary transport systems, 132-133 Enzymes dithiol/disulphide-controlled,139 Ravin-dependent , 139 inducible, 185-186

Escherichiu coli amino acids, 134 glucose 6-phosphate, 134 glutamine synthetase, 7 homoeostasis, 146-147, 148 hydrogenase, 176 lactate, 134 lactose, 134, 147 lactose proton symport system, 145 nitrogen regulation, 3 phosphatase, 35 phosphorylation potential, 142 potassium ion TrkA transport system, I36 potassium translocation system, 141-142 potassium transport system, 144 proline proton symport system, 145 proline uptake, 147 proton/amino acid symport system, 140 proton motive force-generating mechanism, 137 proton/sugar symport system, 140 Eukaryotic organisms, origin from prokaryotic cells, 273

F

Factor 111, 140, 141, 142 Faecal pellets, luminous, 270-271 Fatty aldehyde dehydrogenase, 262 Ferredoxins (flavodoxins), 194-195 Firmacutes, 159 (table) Fish, luminescent bacteria-harbouring, 270-27 1, 27 1-273 Flavin, 244 b-, 268-269 protein-bound, 269 Flavin hydroxide, 239 Flavin 4a-hydroxide, 241 Flavin mononucleotide (FMN). 236, 240 reductases, 244-247 Flavin peroxyhemiacetal, 245 (fig) Flavodoxins (ferredoxins), 194- I95 Fluorodinitrobenzene, 25 1 Formic hydrogenlyase, 171-173 Fungi, metabolic pathways, 2

SUBJECT INDEX

G Gating (threshold phenomenon), 145-146 GAP uptake system, 52, 53 GDHCR, 56, 57 activated repressor, 57 General amino-acid permease, 40-48, 54 ammonia effect, 41-45 L-arginine transport, 37-40 positive control of activity, 45-48 regulation by feedback inhibition, 41 regulation in S. cereuisiae, 47 (table) specificity, 40-41 substrates, 40 transinhibition by amino acids, 41 Gene transfer agent, 204-205 Genotoxic agents, 279 (fig) Glucose, energy recycling during homolactic fermentation, 137 (fig) L-Glutamate, repression by, 71 Glutamate dehydrogenases, 7, 9, 34-35 enteric bacterial, 10 NADP+-specific, 19-20, 56 in A. nidulans, 70-71 Glutamate synthase, 9-12, 71 in Gram-positive bacteria, 10 L-Glutamic acid, 39 (table), 53 Glutamic acid permease, 53-54 Glutamine, 3, 196-197 functions, 35 reversible inactivation by, 36 L-Glutamine, 39 (table) nitrogen metabolite co-repressor, 70-71 uptake, 52-53 Glutamine amide nitrogen, 12 Glutamine synthetase, 7, 20, 35-37 ammonium and, 71 inactivation, 43 L-glutamate and, 71 modulation by adenylylation-deadenylylation, 141 mutations in structural genes, 56 nitrogenase activity regulation, 196-197 Glyceraldehyde phosphate dehydrogenase, 139 Glycoprotein luciferase, 248 GNPI, 52 GNP2, 52 Gonium, 90 Gracilicutes, I59 (table)

315

Guanosine pentaphosphate (pppGpp), 140 141 Guanosine tetraphosphate (ppGpp), 140, 141 alarmosome, 142 regulatory role in bacteria, 142 GUPl transport system, 52 GUP2 transport system, 52

H Halobacteria, 130 Hansenula, 2 Heterorhabditis bacreriophora, 238 Hexobarbital, light emission inhibited by, 247 Histidine, 32, 41 L-Histidine, 38 (table) Histidine permease, 41 Histidine protein, 135 Homoeostasis in magnitude of free energy intermediates, 146-148 Homoarginine, 21 Hydrogenases, 174-190 amino acid composition, 180 biotechnological potential, 21 1 catalytic properties, 182-185 electron acceptor/donor, 183-184 hydrogenase assays, 182-183 kinetic parameters, 184-185 “classical” (reversible), 218, 219 cytochrome cj, 184 flavoprotein component, 184 genetics, 209-2 10 hydrogen recycling, 188-190 carbon dioxide photoreduction, 189190 to nitrogenase, 188-189 inducible enzymes, 185-187 localization in cell, 175-176 molecular properties, 179-180 nickel enzymes, 180-182 nicotinamide nucleotide utilization, 184 orientation of membrane-bound hydrogenases, 176-177 oxy-hydrogen reaction, 187- 188 coupled phosphorylation, 187-188 electron transfer, 187-188 oxygen scavenging, 188

316

SUBJECT INDEX

reversible (“classical”), 218, 219 sources, 179 spectroscopic properties, 180 stability: against denaturing agents, 178 against heat inactivation, 178 during storage. 177 synthesis regulation, 186-187 uptake, 175, 187-190, 219-220 Hydrogen metabolism, literature reviews, 157 (table) Hydrogen metabolism in photosynthetic bacteria, 162-174 aerobic growth in dark, 164 anaerobic growth in light, 162-164 electron donor, 162-164 genetics, 204-205 hydrogen consumption linked to photoreduction of carbon dioxide, 162 hydrogen photoevolution, 165-170 maximal rate, 166-167 (table) hydrogen production, 164-174 literature reviews, 157 (table) molecular hydrogen photoproduction, 165- 170 molecular hydrogen production in dark, 170- I74 anaerobic oxidation of carbon monoxide, 173-174 dark, fermentative metabolism, 170I72 nitrate reduction, 163-164 sulphate reduction, 163 thiosulphate reduction, 163 Hydrogen production, biotechnical aspects, literature reviews, 157 (table)

I Inducer exclusion, 8 Inducible enzymes, 185-186 Iron, effect on luciferase synthesis, 266 Isoagglutinins, 103-104 lsocitrate dehydrogenase, 139

K Klebsiellu aerogenes, sodium extrusion, I30

K k b s i d l o p n e u m o n i w , nitrogenloxygen repression, 76 Kluyueromyce.s, 2

L Lamprobocier modesrohalophiliis, 161 Lauric acid, 239 Leucine, 21 L-Leucine, 39 (table) Luciferase, bacterial, 236, 238-280 active centre, 251-253 aldehyde-binding, 240 applications, analytical/clinical, 274-280 aldehyde-coupled assay, 275 bioluminescence test, mutagen detection, 279, 280 (table) enzyme ligand binding, 276 immobilizedko-immobilized coupled luminescent systems, 275-276 in uitro, 274-275 in uiuo, 276-279 nicotinamide-nucleotide coupled assays, 275 protease activity, 276 wild-type luminous bacteria, 276-278 b-flavin, 268-269 catalytic cycle, 239-241 FMN-sensitized photoinactivation, 252 inactivation: by heat, urea, proteases, 248-249 in uivo, 267-268 intermediates, 239-241 iron effect on synthesis, 266-267 Lux phenomenon, 256 membrane-bound fraction, 248 mukiprotein complex system, 247-248 mutant, 249 oxygen effect on synthesis, 267 photo-affinity labelling with I-diazo-2oxoundecane, 252 primary amino acid sequence, 249, 250 (fig) protein-bound flavin, 269 purification, 248-249 Sepharose-linked, 252 stabilization of luciferase-peroxyflavin intermediate, 253 substrate binding, 251

SUBJECT INDEX

subunit function, 251-253 subunit structure, 249-251 suicide reactions, 268-269 synthesis, 263-264 phenobarbital effect, 264 wild-type, 249 see also Bioluminescent bacteria Luciferase, firefly, 274 Luciferase expression quotient, 261 Luciferase tlavin radical, 241-242 Luciferase peroxy flavin, 240 Luciferase 4a-peroxy FMN, 241 -242 Lumazine protein, 243 Luminescence immune assays, 274 Lumiredoxin, 247 Lysine, 32 1.-Lysine. 38 (table)

M Mendocutes, 159 (table) Methionine, 41 L-Methionine, 38 (table) Methionine permease, 41 Methionine sulphone, 197 L-Methionine-oL-sulphoximine, 7 I , 196. 214 Methylamine, 32 uptake, 8, 51 Methylammonium, 58, 72 MicrotoxT system, 277 Mitochondria, adenine nucleotide translocation, 140 Mollicutes, 159 (table) Mosaic non-equilibrium thermodynamics (MNET), 149 M U R , 54 MUT4, 54 Mutagens, 278, 279, 280 (table) Myristic (tetradecanoic) acid, 253-255 Myristyl aldehyde, 255

N NAD+-GDHase, 21 NAD(P)H dehydrogenase (FMN), 245-247 N AD(P)H :FM N oxidoreductase, 245

31 7

NADP+-GDHase, regulatory function in nitrogen metabolism, 18, 20 Neurospora crassa, 57 acid phosphatase, 75 DNA-cellulose binding, 63 gene nif-2, 62-63 L-glutamine as nitrogen metabolite corepressor, 70-71 mutation en(am)-2, 71 phosphorus regulation, 75 sulphur regulation, 75 see also Nitrogen metabolite repression Nickel enzymes, 180-182 Nicotinamide nucleotides, redox state changes, 170 Nitrate, as nitrogenous nutrient, 2 Nitrite reductase, 77, 78 Nitrogen limitation, 8 “neutral” source, 56-57 repressor, 42 starvation, 7, 20-21 Nitrogen catabolite repression, 3n allantoin-urea degradation, 27-28 amino acid as co-repressor, 32 arginase, 18-19 synthesis, 22 more than one regulatory circuit? 28 non-inducible mutants, 21 OTAase synthesis, 22 urea amidolyase, 27 Nitrogen-containing compounds, uptake systems for, 37-40 Nitrogen metabolite repression, 3n, 57-78 mRNA formation prevention, 58 peripheral role genes, 64-67 allele en(am)-I, 67 allele gln‘. 67 gene amrA, 64 gene aniaA. 66 genes meaAlmeaB, 64 gene tamA, 64-65 mutation MS5. 66-67 mutation nmR-1. 66-67 positive-acting regulatory gene, 59-63 pseudogene, 68-70 Nitrogenase, 190-204 activating enzyme, 200 activity regulation, 195-20 1 glutamine role, 196-197

31 8

SUBJECT INDEX

glutamine synthetase role, 196-197 methionine sulphone role, 197 oxygen sensitivity, 200-201 reversible inactivation of Fe protein,

197-200 “switch-off” effect, 195-196,197,199 bacteria with high content/activity as source of energy, 212-213 biochemistry, 190-195 catalytic mechanism : energetics, 192194

construction of de-repressed cells, 213 electron transport to, 194-195 Fe protein (Component 11, dinitrogenase reductase), 190, 191,199-200 reversible inactivation, 197-198 genetics, 205-209 endogenous plasmids, 208-209 regulatory genes, 207-208 structural genes, 205-207 growth condition adjustment, 213 hydrogen recycling to, 188-189 hydrogen re-uptake, 188 light intensity increase effect, 213 MoFe protein (Component 1, dinitrogenase), 190,190-191 molecular properties, 190-192 structure, 190-192 synthesis, 186-187,201-204 glutamine synthetase role, 202 light dependence, 202-204 repression, 186 Nitrogenous nutrient assimilation enzymedgenes, 8- 12 ammonia and its uptake, 8-9 from ammonia to glutamate, 9-12 uptake regulation, 54-55 Nitrosococcus oceanus, I32 NPRl protein, 55 Nucleotide diphosphate kinase, 141 Nucleotide phosphates, 129

0 Ornithine, 14-15 Ornithine carbomyltransferase, 13, 16 Ornithine transaminase (OTAase), 13,21 induction mechanism, 21-22 synthesis, 23

L-Ornithine transaminase, 23-24 absence of nitrogen repression, 23 nitrogen-rich medium effect, 23-24 Oxaloacetate decarboxylase, 130 Oxalurate, 26 Oxidative phosphorylation, 128 Oxidoreductases, 244-247 Oxygen effect on luciferase synthesis, 267 repression, 76 Oxyphotobacteriae, 158,159 (table)

P Pandorina, 90 Pargyline, 261 Pathway-specific regulatory genes, 72-74 Permease synthesis repression in yeast, 48 Peroxy hemiacetal, 241 Phenobarbital, 261,264 Phosphate-bond energy-dependent transport systems without binding proteins, 136 Phosphoenolpyruvate, 129 Phosphoenolpyruvate (PEP)-dependent sugar phosphotransferase system,

135,140 Phosphoribulokinase, fructose 1,h-bisphosphatase, 139 Phosphorylation potentials, 126,140-142 Phosphotransferase system component FIII, 140,141,142 Photobacterium genus, 237,238 Photobacterium probe, 278 Photophosphorylation, 128- 129 Photosynthetic bacteria, 155-234 anoxygenic photosynthesis, 157 biological solar energy convertors, 210-

218 advantages, 217-218 carbon metabolic pathways for complete substrate degradation, 214-215 cell stabilization by immobilization,

216-217 economical substrates, 215-217 high nitrogenase content/activity strains, 212-214 hydrogen, production of, 215-216 hydrogenase-deficient strains, 214

31 9

SUBJECT INDEX

marine strains, 212 strain screening/selection, 21 1-215 thermophilic/thermostable strains, 212 water depollution, 216 classification, 158- 159 ecological distribution, 161-162 evolution, 160 growth properties, 160-161 hydrogen metabolism see Hydrogen metabolism literature reviews, 157 (table) Plasmids, endogenous, 208, 209 Plastoquinone, 140 Prokaryotde, 159 (table) nitrogen metabolism, 3 Proline, 15 degradation, 24 futile-cycle, 13 transport, 48-49 L-Proline, 39 (table), 78 Proline oxidase, 15, 75 Proline permease, 78 Protein “blue fluorescent”, 243 lumazine, 243 “yellow fluorescent”, 243-244 Proteinases A, B, C, 36-37 Proton motive force, bacterial processes energizedhegulated, 143 (table), 144 Protozoa, oxygen affinities, 278 Purines, as nitrogenous nutrient, 2 Pyrroline 5-carboxylate, 15, 24 mitochondrial degradation, I5 Pyrroline 5-carboxylate dehydrogenase, 13, 15, 24 Pyruvates, 171, 172 Pyruvate-ferredoxin oxidoreductase, 172 (fig). 173 Pyruvate-formate lyase, 171, 172 (fig)

R Redox potentials, 126, 138-140 phototaxis and growth of micro-organisms on, 140 Reductase systems, 247 Repressor ARGR, I6 CARGR, 16

Rhizohium, 0-methyl esters, 110 Rhodopseirdomonas capsulora, 140, I58 (fig)

Rhodopseudomonas sphaeroides alanine uptake, 147 (fig) gating, 146 homoeostasis, 147 membrane potential, 147 (fig) potassium efflux, 130 Rhodospirillaceae, 160 growth properties, 160-161 growth under dark anaerobic conditions, 170- 171 maximal rate of H2 photoproduction, 166-167 (table) nitrate reduction, 163-164 nitrogenase activity regulation, 190 nitrogenase function/regulation, 203 (fig) Rhodospirillales. 158, 159 (table), 160 Ribonucleic acid bases, 140 ROAM mutations, 23, 30-31

S Saccharomyces, 5 growth phase, 5-6 Saccharomyces cereuisiae allantoin transport system, 54 allantoin-urea degradation see Allantoin-urea degradation allophanate hydrolase activity, 28, 29 (table) amino acid uptake systems, 38-39 (table) ammonia-sensitive permeases, 54 ammonia uptake, 8-9, 51-52 arginase, 16-22 activity in different strains, 19 (table), 20 arginase-less strains, 24 arginine: degradationisynthesis, 14 metabolism, 12, 13 (fig) asparaginase I, 31 asparaginase 11, 31-34 L-asparaginase uptake, 5 1 DUR3 urea transport system, 54 general amino-acid permease see General amino-acid permease glutamate dehydrogenase, 9, 34-35

320

SUBJECT INDEX

glutamate synthase, 10, 1 I (table) glutamic acid permeases, 53-54 glutaminase, 1 I L-glutamine uptake, 51 growth, 6 methylamine uptake, 8, 51 mutation( s) constituivity, acting in cis and under control of mating type, 30-31 effect on ammonia-sensitive permease activity, 45 (table) mutation gdhA. 43 mutation gdhCR, 30-3 I , 42-43 mutation pgr, 43 nitrogen-catabolite repression, 4-5, 7 metabolism, 3 proline: metabolism, 13 (fig) transport, 48-50 proline-futile cycle, 13 (fig) pyrroline 5-carboxylate dehydrogenase, 15, 24 starvation for nitrogen, 7-8 strains, 4-5 arginase production, 4-5 uptake systems for nitrogen-containing compounds, 37-40 Serine, 20, 32 Sodium pyruvate, fermentative metabolism, 172 (fig) Solute transport, passive/facilitated, 132I33 Sponge cells, cellular interaction in, 115116

Thioredoxin oxidoreductase, 139 Thiourea, 72 Threonine, 41 L-Threonine. 39 (table) Threonine permease, 41 Threshold phenomenon (gating), 145-146 Thymine-7-hydroxylase, 58, 76 Thymocytes, neutral amino acid transport in, 140 Transhydrogenase, energy-linked, 139 Transmembrane (e1ectro)chemical potentials, 126 Trimethylammonium N-oxide, 171 L-Tryptophan, 39 (table)

U Uptake systems for nitrogen-containing compounds, 37-40 Urea amidolyase, 24-25, 26 activity in nitrogen-rich medium, 31 ammonia effect of, 27, 29-30 nitrogen catabolite repression, 27 regulation, 20 Urea degradation see Allantoin-urea degradation Ureidosuccinate-allantoate permease, 5051 Ureidosuccinic acid, 50 Uric acid, 78 Ustilago genus, 57 Ustilago sphaerogena, nitrogen metabolite-repressible enzymes in, 72

Streptococcus cremoris

homoeostasis: phosphate potential, 148 proton motive force, 148 lactate, 134 proton motive force-generating mechanism, 136-137 Streptococcus faecalis gating, 146 sodium transport, 136 Succinate dehydrogenase, 139

v Valine, 21 Vibrio alginolyricus. Na+ pump, 130 Vihrio costicolu, Na' pump, 130

W Water depollution, 216 Water quality evaluation, 277

T

X

Tetradecanoic (myristic) acid, 253-255 Thioc-upsu roseopersicinu

Thioredoxin, 139

161

Xanthine, 78 Xenopus embryo cells, 112 Xenorhahdu.~,238