I !

77-2398 FACKLAM, Thomas John, 1950- NITROGEN REGULATION OF CATABOLIC ENZYMES OF NEUROSPORA CRASSA. The Ohio State University, Ph.D., 1976 Chemistry, biological

Xerox University MicrofilmsAnn , Arbor, Michigan 48106 NITROGEN REGULATION OF CATABOLIC ENZYMES

OF Neurospora crassa

DISSERTATION

Presented in Partial Fulfillment of the Requirement for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University

By

Thomas John Facklam, B.S.

The Ohio State University

1976

Reading Committee Approved by

G. A. Marzluf, Ph.D.

L. F. Johnson, Ph.D.

T. J. Byers, Ph.D. Advisor / / \ (J Developmental Biology Progra ACKNOWLEDGEMENTS

I wish to thank Dr. George Marzluf for his patient assistance and guidance during my graduate training.

I extend special appreciation to my wife, Nancy, for her support and encouragement during the last four years.

I gratefully acknowledge the financial support of the Developmental

Biology Program.

ii VITA

June 28, 1950 Born - Buffalo, New York

1972 B.S., Cornell University, Ithaca, New York.

1973-1974 Graduate Teaching Associate, Department of Zoology, The Ohio State University, Columbus, Ohio.

1974-1976 N.I.H. Developmental Biology Traineeship The Ohio State University, Columbus, Ohio.

r PUBLICATIONS

"Nitrogen Regulation of Amino Acid Metabolism in Neurospora crassa." Genetics 80:s29 (1975).

iii TABLE OF CONTENTS

ACKNOWLEDGEMENTS ii

VITA, iii

LIST OF TABLES v

LIST OF FIGURES vi

INTRODUCTION, 1

METHODS AND MATERIALS 26

Growth of organism Enzyme assays Allantoinase isolation Affinity column Polyacrylamide gel electrophoresis Molecular weight determination

RESULTS...... 38

Growth of Neurospora on various nitrogen sources Amino acid uptake Growth of amino acid transport mutants Smino acid transport in pm-g and pm-n Regulation of arginine, ornithine and proline catabolic enzymes Allantoinase stabilization Isolation of allantoinase Molecular weight Multi-enzyme complex Michaelis constant Effects of cations on activity Presence of metals in allantoinase Effect of sulfhydryl reagents Feedback inhibition Thermal stability Sensitivity to proteases Characteristics of type II activity In vitro turnover

DISCUSSION...... 110

BIBLIOGRAPHY 127

iv LIST OF TABLES

1. Growth of wild-type and amr on various nitrogen sources.

2. Growth of amino acid transport mutants on various amino acids.

3. Arginase specific activity.

4. Ornithine transaminase specific activity.

5. Pyrroline-5-carboxylate dehydrogenase specific activity.

6. Proline oxidase specific activity.

7. Fractionation of cell-free extracts by (NH^^SO^. precipitation.

8. Lack of association of allantoicase and uricase with allantoinase.

9. Effect of cations on allantoinase activity.

10. Effect of chelators on allantoinase activity.

11. Effect of reducing substances on allantoinase.

12. Effect of potential feedback.

13. Allantoinase sensitivity to proteases.

14. Properties of type II activity.

15. SDS treatment of allantoinase.

16. Protease activity of allantoinase preparation.

17. Treatment of allantoinase by aln extract.

18. Summary of nitrogen control.

v LIST OF FIGURES

1. Degradative pathway of purines.

2. Degradative pathway of arginine, ornithine and proline.

3. Flow diagram of allantoinase isolation.

4. Amino acid uptake of wild-type and amr.

5. Uptake of arginine, phenylalanine and aspartate by pm-n.

6. Uptake of arginine, phenylalanine and aspartate by pm-g.

7. Allantoinase stability in vivo.

8. Allantoinase stability in vivo in the presence of cyclohemimide.

9. Allantoinase stability in vitro.

10. Allantoinase stability in vitro in the presence of protease inhibitors.

11. Allantoinase stability in vitro in the presence of sulfhydryl reagents.

12. Elution of Sephadex G-150 column.

13. Elution of Sephadex G-100 column.

14. Polyacrylamide gels of G-150 and G-100 allantoinase preparations,

15. Molecular weight determination of allantoinase.

16. Molecular weight determination by SDS gel electrophoresis.

17. Lineweaver-Burk plot of allantoinase.

18. Thermal stability of allantoinase I and II.

19. Thermal stability with and without presence of substrate.

20. Stability of G-150 allantoinase.

21. Stability of G-100 allantoinase.

22. Molecular weight determination of type II.

vi 23. Gel-filtration of SDS dialyzed allantoinase.

24. Degradation of allantoinase.

25. Model of nitrogen regulation.

vii INTRODUCTION

Neurospora crassa, a member of the fungal class Ascomycetes, is a typical eukaryote and can provide a model for studying regulatory mechanisms in higher organisms. Neurospora is a typical eukaryote in that it con­ tains mitochondria, a nucleus, ribosomes, seven chromosomes, poly A-con- taining mRNA (1), histones, and other structures and metabolic functions attributable to higher eukaryotes.

As a heterotroph, Neurospora can utilize such simple compounds as acetate, glycerol, and glucose as carbon sources. Nitrate, ammonium, and various amino acids are able to provide the cells with nitrogen (2).

Neurospora crassa is the best genetically characterized eukaryote, second only to Drosophila melanogaster. The combination of a well defined genetics, plus the ease with which it grows on completely defined media make Neurospora crassa an ideal organism with which to study regulation.

Neurospora crassa is able to utilize most amides, amines, purines, and many amino acids as nitrogen sources. One aspect of my research was to examine nitrogen control of amino acid catabolism, with particular attention to arginine and proline metabolic degradation. I have also studied the closely related complex regulation of the purine catabolic enzyme, allantoinase, whose synthesis requires simultaneous induction and catabolic derepression. I will first review salient features of the regulatory mechanisms possessed by Neurospora crassa and related eukaryote organisms, and related aspects of nitrogen metabolism in these forms.

1 2 Then, my specific research objectives will be described and major conclusions will be summarized.

Davis and his coworkers have studied extensively the control of arginine and ornithine metabolism in Neurospora. Their work has lead to a description of a regulatory mechanism where metabolites and enzymes are channeled and compartmentalized. The effect of this arrangement is to maintain high local concentrations of metabolites as well as to protect them from their catabolic enzymes. A second effect is to allow the regulation of independent metabolic pathways containing identical inter­ mediates. In Neurospora crassa the arginine and pyrimidine biosynthetic pathways utilize such regulatory mechanisms.

The common intermediate shared by both the arginine and pyrimidine pathways is carbamyl phosphate (CAP). Carbamyl phosphate is synthesized by two carbamyl phosphate synthetases (CPSase), one for pyrimidines

(CPSase P) and one for the arginine pathway (CPSase A). CPSase A is located in the mitochondria and is fully repressible by arginine (4,5).

CPSase P has been localized in the nucleus complexed to arginine trans-

carbamylase (ATCase)(6). The bifunctional CPSase P within the nucleus

is feedback inhibited by UTP and is derepressed under the condition of

low uridine (7,8). The enzyme, by being compartmentalized within the nucleus, is situated where the nucleoside triphosphates are pooled and

quickly used, thus providing for a most rapid and sensitive regulation.

Both CPSases provide for a separate pool of CAP, but an accumulation of

CAP in one pathway can result in an overflow into the other pathway (4).

This overflow effect is particularly evident with mutations lacking a

particular CPSase; this deprives one pathway of CAP, but the defficiency

is relieved if the second pool overflows (9, 10). 3 The channeling of the pyrimidine CAP pool could be accomplished by binding CAP to the CPSase-ATCase aggregate (4,5). This form of molecu­ lar compartmentation (11) would commit the CAP to the pyrimidine path­ way via ATCase tp synthesize ureidosuccinate.

In the arginine pathway, CAP is believed to be confined in the mito­ chondria along with CPSase A (12,13). The intramitochondrial concentra­ tion of CAP is sufficiently high for efficient use by the next enzyme, ornithine carbamyltransferase, also located within the mitochondria

(4,7). For arginine biosynthesis CAP is compartmentalized to increase its local concentration as well as keep it segregated from being metabo­ lized by ATCase.

CPSase A, which is very sensitive to repression by arginine, is active in the presence of large concentrations of intracellular arginine.

CPSase A is insensitive to feedback inhibition by arginine (5). Along with CPSase A and ornithine carbamyltransferase, ornithine acetyltrans- ferase and two other enzymes which synthesize ornithine, are located within the mitochondria (12). The product of ornithine carbamyltrans­ ferase, citrulline, is exported from the mitochondria into the cytoplasm where the remainder of the arginine biosynthetic enzymes reside. The catabolic enzymes are also found in the cytoplasm (12). Arginine, both exogenous and biosynthetic, goes directly towards protein synthesis, bypassing a large sequestered cellular pool. The arginine which fails to be incorporated into protein enters this pool (14). Labeling studies indicate that once exogenous arginine enters the intracellular pool it remains there and is exchanged only very slowly. This maintains a low arginine concentration within the cytoplasm insufficient to maintain catabolism. Weiss (15) has shown that most of the intracellular arginine 4 is contained within an osmotically sensitive vesicle, distinct from the mitochondria. The vesicle also contains the majority of the ornithine pool. It is by this mechanism that arginine and ornithine are segregated from their catabolic enzymes as well as from sites of induction and feed­ back inhibition.

Ornithine, which is synthesized in the mitochondria, is mostly converted to citrulline (12). Ornithine which does escape from the mitochondria is utilized for polyamine synthesis, or it is accumulated in the arginine-ornithine vesicle (15). Very little biosynthetic ornithine is catabolized by ornithine aminotransferase (OATase). This is probably due to the high of OATase (Km=2mM) and the low cyto­ plasmic concentration of free ornithine (5, 15). By means of the vesicle and the separation of catabolic enzymes from the early biosynthetic enzymes, the arginine-ornithine and pyrimidine pathways are regulated with great efficiency. In prokaryotes the operon can be considered the primary regulatory unit. The operon by definition is a group of related genes that are coordinately regulated and transcribed into a polycistronic mRNA. The components of an operon include a promoter region, an operator region and a structural gene' region. This arrangement allows for the coordinate regulation of contiguous structural genes. Gene clusters have been described in eukaryotes. Two such "operons" are found in the aromatic amino acid metabolic pathways, the quinate-shikimate cluster and the arom gene cluster of Neurospora crassa.

The catabolism of quinic acid and shikimic acid is controlled by a cluster of four linked genes— the qa-cluster. Quinic acid is metabolized by quinic acid dehydrogenase (qa-3) to dehydroquinate, which is metabolized to dehydroshikimate by dehydroquinase (qa-2). Shikimic acid is converted 5 by shikimic acid dehydrogenase (qa-3) to dehydroshikimate. Dehydro- shikimate is then metabolized to the end-product protocatechuic acid by dehydroshikimate dehydrase (qa-4)(16). The gene product of qa-3 possesses both quinate dehydrogenase and shikimate dehydrogenase acti­ vities. A single mutation at the qa-3 locus will result in the simul­ taneous loss of both activities (17).

Each qa-enzyme is inducible by quinate and to a lesser degree by dehydroquinate and dehydroshikimate (18). Induction of catabolic dehydroquinase and dehydroshikimate dehydrase is initiated simultaneously and their synthesis is coordinate. The induction of quinate-shikimate dehydrogenase and dehydroquinase is also coordinate. The rate of syn­ thesis of each enzyme is identical and depends upon the inducer concen­ tration (18). The coordinate induction of these enzymes suggests that they may be synthesized via a polycistronic mRNA.

A fourth mutation, qa-1, results in the pleiotropic loss (and non- inducibility) of all qa-enzymes (16). It has been proposed that the qa-1 gene product has a positive regulatory role. Constitutive qa-1 mutations (qa-1 ) support this hypothesis. These mutations result in the synthesis of all of these enzymes in high concentrations in the absence of inducer (19,20). In heterokaryons formed between qa-1 muta­ tions and wild-type, qa-1 is recessive whereas qa-lc is semi-dormant to wild-type (20). These data indicate a positive nature of action of the regulatory gene qa-1. g There are two groups of mutations of qa-1: qa-1 , which has a x: slow rate of complementation with other qa-mutations, and qa-1 , which has a fast rate of complementation (20,21). These mutations reside in non-overlapping regions of the qa-1 locus (22). Clear complementation 6 occurs between some pairs of qa-1 fr and qa-1 s mutations so that the qa-enzymes are inducible. Temperature sensitive mutations of both f s qa-1 and qa-1 suggest that the qa-1 gene product is a protein. The qa-1 product acts both within its nucleus (cis) and in the nuclei in

a heterokaryon (trans); this indicates that the qa-1 gene product is

a diffusible protein. The two qa-1 mutant classes show varying degrees of non-complementation. One interpretation is that the qa-1 product

is a multimer and that different efficiencies of complementation reflect s f the mixing of qa-1 and qa-1 subunits. Case and Giles (22) suggest

that the two distinct types of mutants produce altered regulatory pro­

teins that fail to interact either with a DNA initiator site (qa-ls)

or with an inducer (qa-1^).

Overall, the quinate-shikimate pathway appears to be regulated by

a multimeric protein (qa-l+ gene product) which binds to the inducer

and acts in a positive fashion to induce the coordinate synthesis of the

three enzymes. This hypothesis implies that a promoter region is

located between qa-1 and the clustered qa-structural genes, but at

the present time there is no evidence for it.

In Neurospora, the second known operon-like gene arrangement is

the arom cluster. The arom cluster is composed of five contiguous

genes, which encode the five enzymes catalyzing sequential reactions

in the aromatic amino acid biosynthetic pathway prior to chorismate.

The five proteins are associated together as a multienzyme aggregate

(23). Two major groups of mutations have been isolated: single gene

mutations lacking a single enzyme and mutations having pleiotropic

effects. There are two types of pleiotropic mutants. The first group

consists of point mutants which posess either one, two, or three of the five enzyme activities, always at markedly reduced levels (23,24).

These mutants are complementing and have been identified as missense mutations. The second category of pleiotropic mutants are those which fail to complement with two, three, or all four of the other single mutant groups. Of particular interest are those which lack all five enzyme activities and map at one end of the proximal structural gene (arom-2) of the arom gene cluster (23,25). These mutations are believed to be nonsense mutations. Case and Giles (25) have identified suppressor mutations for the non-complementing pleiotropic mutants. Evidence for the identification of pleiotropic mutants, which have one, two, or three of the five enzyme activities at greatly reduced levels, as nonsense mutations, comes from sucrose density gradient studies. This group of mutations have arom aggregates of less than normal molecular weight (26). Physical studies and comple­ mentation data indicate that arom aggregate is composed of two sub-aggregates each containing the five enzymes with a molecular weight of 150.000 (26).

With identification of both missense and nonsense mutations

(which respond to suppressor mutations) within the arom region, it appears that the arom cluster is transcribed as a single mRNA in a polar fashion. The gene arom-2 is located proximal and arom-1 distal to initiation of transcription (23). The products of the polycistronic mRNA remain associated upon translation to form a large molecular weight multi-enzyme aggregate.

Positive control can be definexLas occuring when the expression I of certain genes requires the presence of an active inducer protein.

A positive regulatory system can be identified by the behavior of regulatory gene mutants: constitutive mutants will be dominant and uninducible mutants will be recessive. The regulation of sulfur metabolism in Neurospora crassa illustrates the elements of a positive control system.

The metabolism of sulfur compounds is regulated by repression and derepression (27). Repression of the sulfur-assimilatory enzymes is mediated by a metabolite of methionine and sulfate, both favored sulfur sources. Derepression occurs with limiting amounts of methionine or inorganic sulfate or in the presence of an excess of poor sulfur sources such as glucose-6-sulfate or cysteic acid. The unlinked genes for the sulfur assimilatory enzymes are regulated by repression; these include the genes for arylsulfatase, choline sulfatase (28), choline- o-sulfate permease (29), an extracellular protease (32), and methionine permease. Aryl sulfatase is fully repressible by high concentrations of methionine or inorganic sulfate (5 mM), but upon derepression the activity of the enzyme increased greatly (500-2000 fold increase in specific activity)(28). This entire group of enzymes are also con­ trolled by two unlinked regulatory genes called cys-3 and scon.

Cys-3 mutants are pleiotropic, with the loss of synthesis of all the sulfur-assimilatory enzymes (30). Temperature sensitive cys-3 revertants, e.g., cys-3 (REV 65t), have been used to demonstrate that lack of sulfatase activity in cys-3 is due to failure of its synthesis and not to selective turnover of that activity. At the permissive temperature of 25 C, aryl sulfatase is synthesized but at

35 C, the restrictive temperature synthesis ceases (30). These data suggest that the cys product is a protein. To understand the regulatory nature of cys-3, heterokaryons were constructed using mutants which specify electrophoretic variants of aryl sulfatase (33). Heterokaryons contained two nuclear types, each carrying one gene for the distinguishable aryl sulfatases, along with either cys-3 or cys-3 genes. These heterokaryons analyses showed that the cys-3 gene product is diffusible between nuclei and is dominant in action, since both aryl sulfatase genes were expressed in a cys-3 ars + cys-3 ars heterokaryon (30,33). The cys-3 locus thus appears to encode a positive regulatory element.

The second regulatory gene in this system is called scon (sulfur controller gene)(35). Constitutive mutations of scon (sconc) even when grown in the presence of high concentrations of methionine or sulfate, contain derepressed levels of the sulfur-assimilatory enzymes. Heterokaryon analysis shows that the action of sconc is restricted to its own nucleus. Since scon regulates a number of unlinked genes, the scon locus could be coding for some regulatory product which is freely diffusible within its own nucleus. A possible explanation for these data is that the scon product is synthesized on the nuclear membrane and immediately transported into its nucleus where it is restricted. A second explanation is that the arginine- ornithine biosynthetic enzymes, the scon gene product is compartmen­ talized within the nucleus.

The relationship between scon and cys-3 is.unknown at the present time. Dietric and Merzenberg (35) have isolated suppressed sconc mutations which are phenotypically identical to cys-3~ mutants. These

P c supressed scon strains (su634) are actually double mutants of scon cys-3. In heterokaryons using the two electrophoretic aryl sulfatase variants, scon is hypostatic to cys-3. Therefore, this work has shown that enzyme synthesis is dependent upon the presence of the cys-3 gene product. Marzluf (36) suggested, that assuming that both cys-3 and scon act at the level of transcription, that these two regulatory genes might act in a sequential manner. The scon gene would respond to the level of sulfur in the cell and, in turn, inhibit the cys-3 gene product or repress its synthesis. The active cys-3 gene product would "turn-on" the unlinked sulfur- metabolism genes.

In _E. coli the rate of synthesis of B-galactosidase of the lac operon is greatly reduced when the cells are grown with glucose as their carbon source. The repression of inducible enzymes by a metabolite is termed catabolic repression. Catabolite repression is found not only in prokaryotes but in fungal systems as well. Of particular interest is nitrogen catabolite repression or ammonium- repression, which controls the synthesis of a wide variety of enzymes involved with nitrogen metabolism (37,38,39). In Aspergillus nidulans, the synthesis of nitrate reductase, uricase, xanthine dehydrogenase, and extracellular proteases is repressed by ammonium and under the control of the nitrogen regulatory gene areA (40).

The areA mutations are pleiotropic with varying effects depending upon the allele. Several alleles result in the loss of the ability

to utilize nitrogen sources. Other areA mutants result in dere­ pression of one or more ammonium-repressible activities without affecting the inducibility of those enzymes. Two mutations which

lack the ability to utilize various nitrogen sources are the areAr-l and areAr-2 strains. These strains will not grow on nitrate, nitrite, adenine, hypoxanthine, uric acid, allantoin, or a number of amino acids. The transport systems in areA T -1 and araA V -2 have low activities for several nitrogenous compounds, but are not altered for others. The lack of nitrogen utilization, then is a function of some x other factor rather than uptake. The areA -1 mutation lacks urate oxidase and nitrate reductase activities and has low xanthine dehydrogenase activity, although these enzymes still retain their normal inducible character. This suggests that ammonium-repression and induction are two separate functions.

Many alleles of the areA gene have been identified, each having its own characteristics. A mutation, xprD-1, leads to the derepression of nitrate reductase, xanthine dehydrogenase, urate oxidase, and extracellular proteases in the presence of ammonium

(41,42,43). Mutation areA^-101 also results in the derepression of extracellular protease activity, but not nitrate reductase, xanthine dehydrogenase or urate oxidase (40). These two mutations are

classified as derepression mutations of areA since they result

in a loss of ammonium repression. Mutants lacking a response to

ammonium-repression still require induction, thus further supporting

the hypothesis that areA is concerned only with ammonium-catabolite r repression. These mutants act oppositely to the areA class, which

result in an inability to utilize various nitrogen sources and a

failure to synthesize nitrogen related enzymes under derepressive

conditions. V The areA -1 mutant is located on linkage group III, and very

tightly linked to it (presumably allelic) is amdT-102 (a derepression

alleles), aroAr-2, sprD-1, and amdT-19 (40.43). The alleles of 12 areAr type, the repressed phenotype, are recessive to wild-type in diploids and heterokaryons. Strains containing the derepressed phenotypes such as xprD-1 , areA^-101, and amdT-102 are semi-dormant to wild-type in heterokaryons (40,41). The epistatic character of areA suggests a positive mode of control. Arst and Cove (40) have proposed that the areA product, probably a protein, is synthesized and is required for the expression of many nitrogen-metabolizing enzymes and furthermore, that it is inactivated by ammonium or some ammonium product.

Recent work by Arst and Scazzocchio (44) has resulted in the partial description of the control regions for a nitrogen-related gene. The permease for uric acid and xanthine is regulated by both induction (uaY gene) and ammonium (areA). If a structural gene is being regulated by two unlinked control genes, there might be two regulatory sites or regions adjacent to the structural gene.

Using this assumption, cis-acting mutations were isolated which allow the expression of the uric acid-xanthine permease gene (uapA) in the presence of a areA allele (areA-102). The mutant uap-100 was selected as a revertant of areA-102, which had regained the ability to grow on uric acid and xanthine as sole nitrogen sources, but retained all other characteristics of areA-102 (derepressed for acetamidase and reduced foramidase activity) (42).

The uap-100 mutation is constitutive for uric acid and xanthine uptake, and also has 2.5 times the wild-type transport activity. The mutation still retained its repressibility by ammonium. In diploids containing areA-102, uap-100 shows dominance and acts only when uapA (the structural gene for the permease) is located cis to it. The conclusion from these data suggest that uap-100 is an initiator

constitutive mutation with an up-promoter effect. It appears that

since uap-100 is still repressible by ammonium, there remains an

absolute requirement for the areA product. This indicates that the

uap-100 initiator site is capable of recognizing the altered areA-102

gene product. The overall significance is that a specific initiator

site has been described which further defines the mechanism of areA mediated nitrogen-catabolite repression.

Purines can be utilized as the sole nitrogen source by many

fungi and the degradation of purines is under ammonium-repressible

control. The degradation of purines in prokaryotes has been described

in a number of organisms and two pathways have been found. In

Pseudomonas aeruginosa, Ps. fluorescens, Penicillium fluorescens, and

Pen, notatum the conversion of allantoic acid is in two steps

involving the enzymes allantoicase and ureidoglycolase: allantoin

allantoic acid urea + ureidoglycolic acid glyoxylate + urea

(45,46). The second pathway is found in Streptoccus allantoicus

and Pseudomonas acidovorans, which metabolize allantoic acid by

allantoate aminohydrolase and ureidoglycine aminohydrolase: allantoin

> allantoic acid— ^.ureidoglycine— ^ureidoglycolic acid— >glyoxylate

+ urea (47).

In Saccharomyces cerevisiae the degradation of purines is via:

allantoin— >allantonate— >ureidoglyoxylate——>urea — >allophanat.e-- >

NH^ + CC>2 (48). It has been shown that the loci for urea carboxylase

and allophanate hydrolase (the last two enzymes of the pathway) are

linked. (49). There is no linkage between any of the other loci

for the enzymes. The contiguous loci for urea carboxylase and allophanate hydrolase suggests that a multienzyme complex exists, which has in fact been found. These two enzymes are not synthesized

coordinately, so they may not be transcribed in a polycistronic mRNA.

Urea and urea analogues, foramide and formyl urea, can act as

inducers of the five enzymes (48). In urea carboxylase-less strains

these compounds fail to induce, indicating that they are first

carboxylated before they can induce. The true inducer has been

shown to be the usual product of urea carboxylase, allophanate

(48,50,51). Induction by allophanate is accompanied by a 2.5 minute

to 4.0 minute time lag before allophanate hydrolase activity is

increased. Removal of the inducer does not effect the increase in

activity for 15 minutes. After this period there is a decay to a

constant enzyme level. The half-life for hydrolase was calculated

to be about 3 minutes (52). Lomofungin, an inhibitor of RNA

synthesis, totally blocks allophanate hydrolase synthesis under

conditions in which protein synthesis was 45% of the uninhibited

levels. The half-life of the hydrolase mRNA is about 3 minutes.

This is in contrast to the half-life of total yeast mRNA of 20 minutes (53). When lomofungin is added at the same time as inducer,

there is no induction of hydrolase activity. RNA synthesis is

therefore required for the induction process, presumably for the

accumulation of mRNA. Protein synthesis is not required during the

induction process, but is essential for the expression of the

accumulated synthetic capacity, i.e., mRNA translation (54).

The yeast purine degradative enzymes, like those of Neurospora

crassa, are regulated by nitrogen repression. Unlike Neurospora, however, repression in yeast appears to be mediated by amino acids and not free ammonium (55). Repression exerted by ammonium is minor whereas readily utilized amino acids exert full repression. The repression caused by ammonium could actually be the result of trans­ amination to amino acids which are acting as the true repressors..

Addition of metabolically favored amino acids such as serine and asparagine to the medium results in blocking allophanate hydrolase and allantoinase synthesis after a lag of 11 minutes. The presence of serine during the expression of the accumulated mRNA does not affect the rate of hydrolase synthesis. In contrast, when serine is added during the induction, hydrolase synthesis is blocked. This indicates that nitrogen repression through certain amino acids acts at the level of transcription.

Aspergillus nidulans and Neurospora crassa utilize the same pathway for the catabolism of purines (fig. 1) (56). Unlike yeast, the last intermediate is not the inducer. In Aspergillus, purines such as hypoxanthine and uric acid are able to induce xanthine dehydrogenase. Allopurinol, a competitive inhibitor of xanthine dehydrogenase and urate oxidase, prevents the induction by hypoxanthine but not by uric acid. The action of allopurinol is to block the metabolic conversion of hypoxanthine to uric acid, so that it appears that uric acid is the true inducer of xanthine dehydrogenase. Urate oxidase is induced by both uric acid and hypoxanthine but in mutations lacking xanthine dehydrogenase (hxA and hxB) hypoxanthine is not able to induce oxidase activity. Similar results are obtained in cnx mutations, which lack functional xanthine dehydrogenase and nitrate reductase activities. Allantoinase is induced by uric acid

and allantoin, but not by hypoxanthine, Maximum induction of Fig. 1. The Degradative Pathway of Purines. The enzymes which

catalyze these reactions are: xanthine dehydrogenase, uricase, allantoinase, allantoicase, ureidoglycolate hydrolase, and urease;

they are indicated in the figure by the numbers 1-6, respectively.

16 17

ADENINE GUANINE

0 ^ 0 ^ 0 0 HN ii N — NH2_2__^ 2 . NHp Vr •NH . UO ' 0' ^ N' N O 0 N N 0 H H H H H H HYPOXANTHINE XANTHINE URIC ACID ALLANTOIN

COO0 NH2 1 1 HO N O A nh2 coo0nh2 UREIDOGLYOXYLATE o^ na ny 0 H H © COO ALLANTOIC ACID I C =0 GLYOXYLATE

2N H 3+C02 allantoinase occurs when both uric acid and allantoin are present.

Allantoin also appears to induce allantoicase, while uric acid is a weak inducer. Ureidoglycolate hydrolase is not inducible and urease is only slightly inducible by uric acid or allantoin; they both appear to be synthesized constitutively (57,58).

High concentrations of ammonium prevent the increase in uricase and allantoinase enzyme activity even if inducer is present. A pleiotropic mutation (areA) has been identified which leads to the loss of ammonium repression without affecting inducibility (40).

This system of regulation involving the areA gene product and nitrogen catabolite repression coupled with the complex induction pattern control purine degradation in Aspergillus nidulans.

The regulation of the purine catabolic pathway of Neurospora

(fig. 1) is similar to that of Aspergillus (50); The first three enzymes involved in uric acid catabolism are inducible. Uricase and allantoicase are induced by uric acid and hypoxanthine while allan­ toinase is induced by either uric acid or allantoin. Urease and ureidoglycolate hydrolase are synthesized constitutively. Hypo­ xanthine itself is not the true inducer, but presumably, it is uric acid. This was demonstrated by using two mutations lacking functional xanthine dehydrogenase, xdh-1 and nit-1. In both strains hypo­ xanthine is unable to induce either uricase or allantoicase although induction does occur in wild-type. Allopurinol also prevents induction by hypoxanthine of uricase and allantoicase in wild-type.

Finally, a uric acid analogue, 8-azaxanthine, is able to induce uricase and allantoicase without itself being metabolized. Uricase

and allantoicase are both induced by uric acid and are synthesized coordinately with each other but not with allantoinase. The first

three enzymes are not only inducible, but are repressible by the

end product, ammonium. Like Aspergillus, Neurospora appears to have a positive control gene, amr, for nitrogen metabolizing enzymes, amr mutants result in a pleiotropic loss of enzymes and

in the ability to utilize intermediates or purine degradation as nitrogen sources (59,60).

A third potential control mechanism is that of enzyme turnover.

Uricase and allantoicase are very stable in vivo and any changes in

their activity seem to result from changes is the rate of synthesis.

In contrast, allantoinase is very unstable in vivo with a calculated half-life of 20 minutes. The induction of allantoinase appears

to depend upon de-novo protein synthesis of new enzyme. It should be noted that allantoinase undergoes extensive turnover iri vitro

as well as in vivo (60).

Regulation in fungi comprises a wide range of control mechanisms

including metabolite channeling, operon-like clusters, positive

regulatory signals, catabolite repression, and dynamic turnover.

Many of these regulatory controls are probably found in other

eukaryotes and not limited to Neurospora and other Ascomycetes.

One exception are operons which appear to be quite rare in the

lower eukaryotes and even increasingly rare in higher forms. Control

schemes involving complex regulatory signals which act in a positive manner at the level of transcription could act as models for more

complex patterns. Regulatory signals may regulate batteries of

genes or could only activate a single gene which in turn is a

regulatory element for other genes in a sequential manner. 20 Transcriptional regulation has been described and no doubt can provide a mechanism for control of gene activity. However, although not mentioned, translational control may provide an additional means for regulation. Overall, the regulatory control mechanism in fungi are complex, but may provide models for regulation of gene expression at a molecular level.

Neurospora crassa is able to utilize a variety of amino acids as sole nitrogen sources in wild-type strains. The nitrogen control mutant, amr, is incapable of using certain amino acids which support growth of the wild-type (table I). Wild-type is able to grow on proline, serine and ornithine while amr is incapable of their utilization. But there are a number of amino acids which both strains can utilize, such as arginine, glutamine, and aspartate.

My research objective was to examine nitrogen-catabolic control of amino acid catabolism. The major goal of this research was to

-j- determine whether the amr gene regulates the synthesis of certain catabolic enzymes required for amino acid metabolism. The arginine, ornithine, proline degredative pathway (fig. 2) was studied since arginine is utilized by both strains and ornithine and proline are utilized by wild-type but not by amr. This work included a description of the regulation, including induction and ammonium catabolite control, of the related enzymes, arginase, ornithine transaminase, pyrro.line-5-carboxylate dehydrogenase, proline oxidase and an amino acid permease. The remaining enzymes exhibited regulation by induction alone.

The second area of my research was a detailed study of allan­ toinase. Allantoinase and other purine degradative enzymes are Table 1

Growth of Wild-type and amr on Various Nitrogen Sources

Nitrogen Growth of Strains^ Source1 Wild-type amr

Glycine Trace3 Trace Leucine 8.2 2.k Tryptophane 1.9 0.6 Ornithine 11.0 3.8 Methinoine 7.3 2.3 Valine 8.0 3.2 Phenylalanine 1.5 .2 Alaine 3.0 2.5 Proline 5-0 Trace Glutamine ^7-9 51.0 Glutamate 27.5 lh.0 Histidine 1.1 Trace Cysteine Trace Trace Threonine 1.1 0.5 Isoleucine 2.9 1.8 Tyrosine 13-1 17.0 Asparate 50.2 kh.2 Serine 3-9 1.8 Arginine 30.0 30.0 Inositol 0.3 0.2 Citrulline 19-7 20.8 Urocanate Trace Trace Nitrate 50.7 0.1 lAll nitrogen sources were added to Vogel's medium minus nitrogen to a final conc. of lOmM. "Growth in milligrams dry weight after 3 days of growth. 3Less than 0.1 mg. Fig. 2. The Arginine, Ornithine, and Proline Catabolic Pathway.

The enzymes which catalyze these reactions are: arginase, ornithine-^

-transaminase, pyrroline-5-carboxylate dehydrogenase, and proline oxidase; they are indicated in the figure by the numbers 1-4, respectively.

22 23

ARGININE ORNITHINE GLUTAMATE - y -SEMIALDEHYDE COOH COOH COOH

HN-C-H2 , NHo-C-H2 , NH2-C-H2 ( CH. CH2 CHp I 2 . i 2 ? i ch2 . . . L * c h 2 c h 2

CH CHo CH I £ i * n NH NH2 0

?=NH (NON-ENZYMATIC) Jf NH 2 PROLINE PYRROLINE -5- CARBOXYLATE

HoC CHo a ------HoC------CHo 2| | — I I h 2 „ c h - c o o h hcl c h - c o o h N H

s/ GLUTAMATE COOH NH2-C-H c i CH2 I c h 2 I COOH controlled by Induction, nitrogen repression, and the regulatory gene,

amr. Allantoinase is of special interest because of its complex

regulation and apparent instability. To facilitate a detailed

study of allantoinase, the enzyme was purified and its biochemical

characteristics examined. Allantoinase is a large enzyme of 120,000

molecular weight, make up of subunits of 30,000-35,000 molecular weight. It does not exist in a multi-enzyme complex with any of the

other catabolic enzymes. Allantoinase (in vitro) does not seem to be

feedback inhibited by any of the purine degradative intermediates.

The instability of allantoinase raises the question of whether

or not it actually undergoes dynamic turnover. If allantoinase does

indeed turnover, is it due to an active process, as modification

for degradation by proteases, or is it due instead to some inherent

instability which gives the enzyme molecule some predetermined life

expectancy? My work has shown that allantoinase has a requirement

in vitro for EDTA and mercaptoethanol for partial stabilization.

Even with these reagents present, allantoinase undergoes loss of

activity. During degradation in vitro, a small molecular weight molecule is released which has allantoinase-like properties. It

is unknown whether a functional relationship exists between this

small molecule and the process of jm vivo turnover. The purified

allantoinase preparation, containing only one other protein, can be partially stabilized by 5 mM EDTA, but other protease inhibitors,

such as PMSF, have no effect. A very low level of protease activity was detected in this same allantoinase preparation, which might

account for the degradation oai vitro. It was also observed that

crude extracts of aln strains (lacking allantoinase activity), when added to purified allantoinase, rapidly inactivated the enzyme.

The significance of these observations and their relationship to in vivo and in vitro turnover will be discussed. METHODS AND MATERIALS

Organism and growth conditions. Neurospora crassa wild-type 74-OR23-1A was used for allantoinase isolation, enzyme assays and growth studies.

All mutants used were obtained from the Fungal Genetics Stock Center,

California State University, Areata, California. Strains were main­ tained on agar plates with Vogel's minimal medium plus sucrose.

Conidial suspensions were made by inoculating sterile water with conidia harvested from plates after seven days of growth at 30 C. The conidia in water were dispersed with a vortex mixer and passed through a filter packed with glass wool. The filtered suspension was then used for inoculations. For growth assays, Erlenmeyer flasks (125 ml) containing 20 ml of Vogels medium minus nitrogen plus 2% sucrose and supplemented with various nitrogen sources were used. Each flask received a light inoculum of conidia containing 2000-3000 spores. The cultures were grown at 30 C for three days to stationary phase. The mycelial pads were collected, dried to 80 C overnight and weighed. For allantoinase isolation, large quantities of cells were required so I used six one-liter flasks, each containing 500 ml of Vogels medium minus nitrogen plus 2% sucrose and supplemented with 1.5 mM ammonium tartarate and 1.0 mM hypoxanthine. The cells were grown for 21 hours with continuous shaking. Mycelia were harvested by vacuum filtration and washed several times with distilled water. For enzyme assays using crude homogenates, cells were col­ lected on a filter and ground with a pestle in a cold mortar using an equal weight of acid-washed sand and 3-5 fold volumes of buffer

(usually 20 mM Tris-HCl, pH 7.2). The homogenate was centrifuged at 20,000 x g for 15 minutes at 0 C. The clear superantant fluid was retained for assays.

Enzyme Assays;

Arginase (1-arginine ureohydrolase EC 3.5.3.1). Acetone dried mycelia pads were homogenized in 0.025 M glycine-NaOH buffer, pH 9.5

The 20,000 x g supernatant fluid was assayed at 37 C in a 1.0 ml reaction mixture containing 25 umoles glycine-NaOH buffer, pH 9.5,

0.25 umoles MnC^, 50 umoles of L-arginine, and 0.1 ml of extract.

The reaction was run for 5 minutes and stopped by the addition of

5.0 ml of 5% TCA. The concentration of the product of the reaction, ornithine, was determined by the methol of Chinard (61). To 1.0 ml of the arginase reaction mixture were added 1.0 ml glacial acetic acid and 1.0 ml of reagent solution. The reagent solution contains

0.4 ml 6 M H^PO^, 0.6 ml glacial acetic acid, and 25 mg ninhydrin per 1.0 ml. The solution was heated to 70 C to solubilize the ninhydrin. Each sample analyzed was compared with a sample blank containing 1.0 ml of sample,' 1.0 ml glacial acetic acid and 1.0 ml reagent solution minus ninhydrin. The prepared•assay tubes were then capped and heated at 100 C for 60 minutes, which can be conviently done by autoclaving the tubes for 60 minutes. One ml glacial acetic acid was added to each of the tubes which were then allowed to cool to room temperature. The final volume of each tube was adjusted to 5.0 ml with glacial acetic acid. The absorbance of the samples was determined at 515 mn. An extinction coefficient for ornithine was established experimentally to be 5.75 x 104*7 moles (absorbance in 1 cm light path). The specific activity of arginase was given as micromoles of ornithine formed per milligram protein per minute at 37 C, under conditions of the assay described above

(62).

Ornithine Transaminase. (L-omithine:2-oxoacid amino-transferase,

EC 2.6.1.13). Harvested mycelia were homogenized in potassium phosphate buffer, pH 7.5. The assay mixture with a total volume of 1.5 ml contained: 20 umoles L-ornithine-HCl, 20 umoles -keto- glutarate, 1 umole pyridoxal phosphate, 1.2 mg O-aminobenzaldehyde, and 100 umoles K:P0. buffer, pH 7.5. To start the reaction, 0.1 ml 4 of extract was added. The mixture was incubated for 30 minutes at

37 C. At the end of this period an equal volume of cold 10% TCA was added to stop the reaction. The absorbance of the samples was read +3 -1 at 440 nm. The extinction coefficient used was 2.7 x 10 moles

(62).

A '-Pyrroline-5-carboxylate Dehydrogenase. The substrate for this enzyme is A*-pyrroline-5-carboxylic acid which must be chemically synthesized since the compound is not available commercially. The synthesis of this substrate was accomplished in two parts; First

I synthesized P, ^-decarbethoxy- T -acetamidobutyraldehyde which was acid hydrolyzed to glutamic- $ -semialdehyde, and this in turn was cyclized to A -pyrroline-5-carboxylate. Preparation of the first intermediate was carried out in a two-necked flask (500 ml) with a CaCl^ drying tube and a drop funnel connected. Ethyl acetamidomalonate (100 gm) was suspended in 152 ml benzene along with 0.3 gm sodium methoxide and was stirred at 20 C. Redistilled acrolein (32 ml) in 31 ml benzene was added dropwise with continued stirring. After two hours the reaction mixture was filtered through a sintered-glass funnel. The product remaining behind was t- ^ dicar- bethoxy- If -acetamidobutyraldehyde. This product was resuspended in a mixture of 95% ethanol (220 ml), 30 ml benzyladehyde, and 2 ml glacial acetic acid. After refluzing this mixture for 30 minutes, the alcohol was removed by rotary evaporation. From the aqueous phase an oily residue was recovered by further evaporation. This residue was dissolved in 6 N HC1 and refluxed for 20 minutes. The acid- dissolved residue was passed through a Dowex 50-X4 column (2x20 cm) and eluted in a stepwise manner, first with water, followed by 0.01 N

HC1 and then 0.5 N HC1. The 0.5 N HC1 elutate, containing the final product, was collected and the pooled fractions were lypholized.

The dried product was stored under a vacuum (63).

The enzyme was assayed in crude homogenates in 100 mil Tris-HCl buffer, pH 7.5 containing 2.0% glycerol. The reaction mixture, total volume 3 ml, included 0.1 ml 40 umoles/ml A'-pyrroline-5-carboxylate,

0.2 ml 3.39 umoles/ml NAD, and 0.3 ml 0.5 M Tris-HCl buffer, pH 8.2.

Added last was 0.1 ml extract, and the enzyme reaction was assayed at room temperature in quartz cuvettes, the assay being run in the

Perkin-Elmer Double Beam Spectrophotometer-Coleman 124. The rate of NAD reduction was followed over a 20 minute period at 339 nm. ”13 —1 The extinction coefficient used for NADH was 6.2 x 10 moles and specific activity was expressed as umoles oxidized per mg protein per minute.

Proline Oxidase. (EC 1.4.3.2). The assay was carried out in 1 cm glass cuvettes in a Perkin-Elmer Double Beam Spectrophotometer at room temperature. The assay mixture contained in a total volume of 3.0 ml: 200umoles Tris-HGl buffer, pH 8.5, 10 umoles p-iodoni- trotetrazolium violet, 1.3 umoles N-methylphenazonium methosulphate,

42 umoles FAD, and 0.1 ml extract. The reaction was started with the addition of 150 umoles proline and the increase in absorbance at 500 nm was recorded over a 10 minute period. The blank contained the assay mixture and enzyme but not proline was not added. The extinction coefficient used was c= 1.091 moles (65).

Amino Acid Transport. Neurospora conidia were suspended in Erlen- meyer flasks (250 ml) containing either 100 ml Vogels minimal medium plus 2% sucrose or, for experiments comparing nitrogen sources,

Vogels minimal medium minus nitrogen plus 2% sucrose and supplemented with the appropriate nitrogen .compound. The conidia were inoculated / 0 c\ to a concentration of 0.1 A units and germinated for six hours with shaking. After this germination period, 1.0 ml samples of the conidial suspension were removed and added to assay mixtures containing

0.1 ml labeled amino acid (0.01 mCi/0.01 umoles or % ) and 0.1 ml

0.1 M amino acid as carrier. The samples were incubated at 30 C for varying times (0, 2, 4, 6, 8, 10, or 15 minutes) with shaking, when the conidia were collected on a millipore filter and washed with 5 volumes of water. The filters were dried and placed in scintillation vials containing 10 ml Beckman Universal Scintillation Fluid (100 gm napthalene, 5 gm PPO, 0.1 gn POPOP, and 905 ml dioxane) and counted in a Beckman LS-133 Liquid Scintillation Counter.

Allantoinase. (allantoin amidohydrolase, EC 3.5.2.5). To 1.2 ml 100 mM

Tris-HCl buffer, pH 7.2 and 5.7 umoles allantoin/ml, 0.1 ml enzyme was added. This reaction mixture was incubated for 10 minutes at 30 C. At the end of incubation, 0.1 ml 11% HC1 and 0.1 ml 1% phenylhydrazine

HC1 was added. The sample was mixed and placed in a boiling water bath for two minutes and then cooled in an ice bath. After cooling,

1.0 ml 24% cold HC1 and 0.1 ml 5% K^Fe(CN)g were added and the reaction was allowed to stand for 15 minutes at room temperature. The absorbance of the sample was read at 535 nm against a zero time blank (the enzyme was added to a tube containing buffer, allantoin, 11% HC1, and phenylhydrazine HC1). The molar extinction coefficient used was

4.93 x 10^ moles/liter (cm^) (66).

Allantoicase. (allantoate amidohydrolase EC 3.5.3.4). Allantoicase was assayed in a mixture of 0.9 ml 100 mM Tris-HCl buffer, pH 7.2 and

1.0 mg/ml allantoic acid: 0.1 ml enzyme was added and the mixture was incubated for 10 minutes at 30 C. The reaction was stopped by the addition of 0.5 ml 0.5 N NaOH. After a two minute period 1.5 ml

0.4 M phosphate buffer, pH 7.0 and 0.1 ml 1% phenylhydrazine HC1 were added. This was mixed and incubated for 10 minutes at room temperature, when the assay mixture was cooled in an ice bath. When cool, 2.0 ml of cold concentrated HC1 and 0.1 ml 5% K^Fe(CN)g was. added. The absorbance of the sample was determined at 535 nm against a zero-time blank containing enzyme added to a tube containing buffer, allantoic acid and NaOH (66).

Urease. (urea aminohydrolase EC 3.4.1.5). Urease was assayed in a mixture of 2.7 ml 100 mM Tris-HCl buffer, pH 8.6 and 211 ugm/ml uric acid in a 50 ml Erlenmeyer flask. Upon addition of 0.3 ml enzyme

0.5 ml was withdrawn and placed in a tube containing 4.5 ml 12% HC1; this sample was the zero-time tube. The remaining assay mixture was incubated for 10 minutes at 30 C with shaking, when 0.5 ml samples 32 were withdrawn and to each 4.5 ml 12% HC1 was added. The absorbance of these samples was read at 290 nm, and its decrease was recorded (67).

Protein Determination. The amount of protein in a sample was deter­ mined by the method of Lowry et al (68). A standard curve was established using bovine serum albumin.

Allantoinase Isolation. The purification of allantoinase is diagrammed in Fig. 3. Mycelia (60-90 gm wet weight) collected from six flasks were homogenized in 24 ml of 20 mM Tris-HCl buffer, pH 7.2 plus

1 mM EDTA and 1 mM 2-mercaptoethanol. The.homogenate was centri­ fuged at 20,000 x g for 15 minutes and the supernatant fluid was retained. To this, (NH^^SO^ was dissolved slowly to a concentration of 30%. This mixture was centrifuged at 40,000 x g for 30 minutes and the precipitate discarded. The supernatant fluid was brought up to 50% saturation of (NH^^SO^ and centrifuged at 40,000 x g for 30 minutes. The precipitate was saved and suspended in 2 ml

20 mM Tris-HCl buffer, pH 7.2 plus 1 mM EDTA and 1 mM 2-mercapto­ ethanol. This sample was passed through a Sephadex G-150 column

(2.5 x 40 cm) and eluted with the same Tris-HCl buffer. The fractions containing activity were collected, pooled, and concentrated to about 1.0-1.5 ml by passive dialysis against Aquacide II (Calbiochem).

The concentrated material was then passed through a Sephadex G-100 column (1.0 x 20 cm) and eluted with the Tris-HCl buffer. The fractions containing allantoinase activity were found to contain only two proteins, the enzyme and a single contaminating protein of

30,000 molecular weight. This preparation was used for all experiments which involved a purified enzyme. The contaminating protein can be removed by one additional step, affinity chromatography, as described below. Fig. 3. Flow Diagram of Allantoinase Isolation

33 34

HARVESTED MYELIA

HOMOGENIZE CENT. 20,000 Xg FOR 15 MIN. r PPT. SUPERNATANT 30% (NH4)2S04 v

' I SUPERNATANT PPT. 50% (NH4)2S04 v/ I V SUPERNATANT PPT.

v SEPHADEX G-150

v SEPHADEX G-IOO

& r ' AFFINITY COLUMN ALLANTOINASE

v ALLANTOINASE 5-Hydantolnacetic Acid Affinity Column. The allantoin analogue,

5-hydantoinacetic acid, was reacted with one equivalent of water-soluble

carbodiimide and one equivalent of alkylamine glass (provided by

Dr. G. Royer) in redistilled n,n-dimethylformamide. It is important

to use a tube with a teflon screw cap to carry out the reaction. The

reaction was first run for one hour at 0 C and then at room temperature

for four hours. A gentle shaking was used, but magnetic stirring was avoided. After incubation, the glass was washed well with

several volumes of water on a sintered glass funnel. The washed

glass was then degassed and stored in water at 3-4 C. A disposable

pasteur pipet was used as a column and when packed with glass had a

void volume of 2.0 ml. A sample was introduced onto the column, _2 which was developed stepwise, first with buffer, and then with 10 M

5-hydantoinacetic acid in 20 mM Tris-HCl buffer, pH 7.2, which eluted

the allantoinase. The removal of 5-HAA from the enzyme is very

difficult. Repeated dialysis against several liters of buffer plus

activated charcoal reduced the 5-HAA concentration. To monitor this

process, samples of the dialysate and enzyme preparation were scanned

from 200 nm to 300 nm in a Perkin-Elmer Model 202 UV-Visible Spectro­

photometer. 5-HAA has a characteristic absorbance peak at 217 nm

which can be quantitated.

Polyacrylamide Gel Electrophoresis. Gels routinely used for deter­

mination of purity of enzyme preparations were 5% polyacrylamide.

The gels were made up in a total volume of 25 ml containing: 1.25 gm

acrylamide (Cyanogum 4.1) , 50 ml TEMED and 12 mg ammonium persulfate

were dissolved in 0.37 M Tris-HCl buffer, pH 8.9. Gel tubes (75 mm

x 5 mm i.d.) sealed at one end with Parafilm, were filled to within 5 mm of the top with the acrylamide solution followed by a top layer of water. After 10 minutes the water layer and Parafilm were removed and the gels placed in the electrophoresis apparatus (Buchler Instru­ ments). The electrolyte buffer was a Tris-Glycine buffer, pH 8.3

(0.6 gm/liter Tris-HCl and 3.0 gm/liter glycine). The samples, containing bromophenol blue saturated with sucrose, were layered on top of the gel with a Hamilton Syringe (Microliter, model 170). The samples were run with 1 milliamp per gel tube until the sample had entered the gel, when the current was increased to 3 milliamps/gel using a Buchler Instruments power source.

The gels were stained with Coomassie blue R250. After the gels were removed from the tubes, the proteins were fixed by suspending the gels in a 10-40 fold volume of 12.5% TCA for 30 minutes. After the fixing period, the gels were placed in a solution prepared by diluting a 1% aqueous stock solution of Coomassie blue with 19 volumes of

12.5% TCA (69). One hour was sufficient for staining after which the gels were transfered to 10% TCA for destaining and storage.

Molecular Weight Determinations. The molecular weight of the purified allantoinase was determined with a calibrated Sephadex G-200 column

(1.5 x 25 cm). The column was calibrated with proteins of known size and their Kav calculated. The standards used were aldolase

(40.000 MW), bovine serum albumin (68,000 MW), and myoglobin (12,000 MW).

An independent molecular weight determination was made with SDS polyacrylamide gel electrophoresis. The concentration of the gels were 10% containing 0.1% SDS; the buffer used was Tris-Glycine buffer

(6gm/liter Tris-HCl and 30 gm/liter glycine) plus 0.1% SDS. The gels were run in the same manner as the 5% polyacrylamide gels. The gels were standardized with bovine serum albumin (68,000 MW), ovalbumin

(43,000 MW), trypsin (23,000 MW), and myoglobin (12,000 MW). Each

sample containing allantoinase was prepared by adding 0.1% SDS and

1% 2-mercaptoethanol and heating in a boiling water bath for 5 minutes.

The gels were stained with 0.25% Coomassie blue in 40% methanol and

7.5% acetic acid for one to two hours and then destained in 20% methanol and 7.5% acetic acid overnight.

The molecular weight determination of a small degradation product with allantoinase-like activity was accomplished by using a

Sephadex G-15 column (1.5 x 25 cm). Because of the small size of

this product, the column was calibrated with arginine (174 MW),

leucylglycine (206 MW), leucylglycylleucine (291 MW) , and glycyl-

leucylglycylleucine (412 MW). To determine the fractions in which

the standards were eluted, 10 ul of each fraction was spotted onto

filter paper, dried, and sprayed with ninhydrin reagent (Sigma). RESULTS

Growth of Neurospora crassa on various nitrogen sources. Wild-type is able to utilize many nitrogen sources as indicated by its vigorous growth on glutamine, glutamate, aspartate, arginine, and nitrate.

(Table 1). However it cannot grow using glycine, cysteine, or urocanate as a nitrogen source. The amr mutant likewise cannot use these same potential nitrogen sources, and, furthermore is incapable of utilizing proline, histidine phenylalanine, and nitrate. The remainder of the amino acids are utilized to a greater degree by wild-type, as seen in increased growth, compared to amr. Particu­ larly significant differences are observed with nitrate, ornithine, proline, leucine, valine, and methionine.

Amino Acid Uptake. To account for the different abilities of wild- type and amr in their utilization of amino acids, the transport of arginine, phenylalanine, aspartate, glutamate, and ornithine was examined (Fig. 4). The uptake of each amino acid was assayed in wild-type and amr under both repressing (25 mM ammonium) and dere­ pressing (1 mM urea) conditions. Both strains, when grown in

25 mM ammonium, had essentially the same rate of uptake. Wild-type, when grown in 1 mM urea, showed a marked increase in amino acid uptake, generally a 10-fold increase was observed. The amr strain grown under derepressing conditions did not exhibit such an increase, but maintained the same levels of uptake present when grown under repressing conditions. These data indicate that amr cannot be Fig. 4. Amino Acid Uptake of Wild-Type and amr. Conidia were germinated for six hours in either Vogel's minimal medium (25 mM

NH^ *) or in Vogel's medium minus nitrogen supplemented with 1 mM urea. Arginine, phenylalanine, aspartate, glutamate, and ornithine were assayed for uptake. Wild-type in 1 mM urea (0) and in 25 mM

NHa* fa) and amr in 1 mM urea (O) and in 25 mM NH.* O^s). 4

39 40

1 5 , 0 0 0

u> 10,000 s_ < I O ’S-

o9- 5 , 0 0 0

0 2 4 6 8 10 12 14 15 TIME (min) derepressed for amino acid transport by low nitrogen concentrations as is wild-type.

Growth assays of amino acid transport mutants. Mutations lacking functional general amino acid permease (pm-g), neutral amino acid permease (pm-n), and double mutants (pm-g,n) were grown in medium containing various amino acids as the sole nitrogen sources (Table 2).

The data show that the pm-g strain is unable to utilize all amino acids tested, except for leucine which is used poorly. The strain lacking the neutral permease (pm-n) was capable of maintaining various degrees of growth on all of the amino acids. This indicates that the general amino acid permease is the dominant uptake system for use of the amino acids tested.

Amino acid transport in pm-g and pm-n. The pm-n strain was able to increase its uptake of arginine, phenylalanine, and aspartate by

2-3 fold when grown under nitrogen-limiting conditions (1 mM urea)

(Fig. 5). In contrast, the pm-g strain had a greatly reduced uptake of the amino acids under similar conditions, (Fig. 6). It appears that pm-g is also derepressed to the same degree (2-3 fold) as pm-n. The diminished ability to transport the amino acids in pm-g agrees with the growth data (Table 2), indicating that the uptake of arginine, phenylalanine, and aspartate occurs mainly by the general permease. These data also indicate that both permeases are derepressed by low concentrations of ammonium and therefore are likely to be under nitrogen-catabolite repression.

Regulation of arginine, ornithine, and prollne catabolic enzymes.

Arginase in wild-type is induced by 10 mM arginine, resulting in a five-fold increase in specific activity (Table 3). This is in Table 2 -

Growth of Amino Acid Transport Mutants on Various Amino Acids

Amino Growth of Strains2, Acids1 pm-g pm-n pm-g,n-

Phenylalanine Trace3 2.7 Trace

Proline Trace 6.2 Trace

Leucine 5-35 11.7 7.3

Glutamate Trace 10. b Trace

Aspartate Trace 17-5 Trace

Glycine Trace 0.8 Trace

Arginine 1.1 UU.2 1.6

■’•All amino acids were added to Vogel's medium minus nitrogen to a final concentration of lOmM. 3Growth in milligrams dry weight after 3 days of growth. aLess than 0.1 mg dry weight. Fig. 5. The Uptake of Arginine, Phenylalanine, and Aspartate by pm-n. Conidia of each strain were germinated for six hours in medium containing either 25 mil or 1 mM urea as the sole nitrogen source. Arginine uptake was assayed in minimal medium (£1 Q) , phenylalanine uptake in minimal medium is) a aspartate uptake in minimal medium (©--- &) , arginine uptake in repressing medium

43

Fig. 6. The Uptake of Arginine, Phenylalanine, and Aspartate by pm-g. Conidia were germinated for six hours in either Vogel's minimal medium (25 mM NH +) or Vogel's medium minus nitrogen and 4 supplemented by 1 mM urea. Arginine uptake was assayed in' the

-f- presence of 25 mM NH^ (£3--- ^ and 1 mM urea (Q □) , phenyl­ alanine in the presence of 25 mM (& A) , and 1 mM urea

A), and aspartate in the presence of 25 mM NH,+ (©--- ©) , 4 and 1 mM urea (O C) .

45 cpm 0 0 0 , 5 0 0 0 , 3 7,000 000 0 2 DEI 4 6 ie mn ) (min Time 8 10 12 14 16 4N ON agreement with Davis and Mora (62) who also found that arginase was the inducer of arginase. The amr strain also responds to arginine as an inducer of arginase. Neither strain displayed any change in arginase activity when grown on repressing and derepressing concen­ trations of ammonium. It seems clear that arginase synthesis is not subject to ammonium-repression.

Ornithine transaminase (OTA), in both wild-type and amr, is inducible by arginine and to a lesser extent by ornithine (Table 4).

There is no derepression of activity when either wild-type or amr is grown on 1 mM ammonium. This lack of derepression is also apparant by the high specific activity in cells grown in medium which includes arginine and 25 mM ammonium. If OTA was sensitive to nitrogen catabolite control, the specific activity expected would be similar to that for uninduced conditions, which clearly was not found. Like arginase, OTA is not under nitrogen-catabolite repression, but is inducible.

Pyrroline-5-carboxylate (p-5-C) dehydrogenase is induced by proline (and to a lesser extent by ornithine) in wild-type and amr.

(Table 5). The specific activities for wild-type grown on high and low concentrations of ammonium are about equal, indicating that the activity is not derepressed. Additional evidence are the data comparing activities between cells grown on 25 mM ammonium plus inducer and 1 mM ammonium plus inducer. There are no significant differences between specific activities under the two conditions.

If high (25 mM) ammonium repressed P-5-C dehydrogenase synthesis, the enzyme level would have been much lower under high ammonium conditions. The wild-type response to ammonium concentrate and 48

Table 3

Arginase Specific Activity

Genotype K-Source lOrnM Arg 25mM NH„+ 1-rnM W H / lOmM Arg ■t 25n'M N1L,+ W.T. (7^-OR23-1A) 1.0 1.0 1J-'J

AMR 0.9 1.0 h.9 »i.5 49

Table U

Ornithine Transaminase Specific Activity (XIO"2)

Genotype N--Source lOrrM Arg. 2 5 mM Ml/ lmM Ml/ lOinM A rg. t 2 f>iM Nil.,1 lOinM Dni, W.T. (7 ^-OR2 3 -lA) .29 .28 2 . 1 1.6 1.2 CO lf\ AMR • 36 3-8 2.5 1.3 Table 5

iyrroline-5-Carboxylate Dehydrogenase Specific Activity

Genotype Ii-Source lOirli Pro ICteM Pro ICnM Orn lOinM Orn 25mM NH4+ i e m k h 4+ lOr.M Pro + 25rrM K H 4+ + lrM N H 4+ lOmM Orn + 25n24 H H 4+ + ijt.m n h 4+

W.T. (7^-0323-LA.) 6.6 5.8 3^.6 23.5 31.7 13.7 20.U 18.6

AMR 6.8 7.0 19-3 27.1 23.0 11.8 21.8 21.6

U1 o to inducers is mirrored by amr. These data indicate that the synthesis of P-5-C dehydrogenase is controlled only by induction.

Proline oxidase is induced by proline and ornithine when compared to the basal level activity found under conditions of 25 mM ammonium

(minimal medium) (Table 6). Wild-type grown on 1 mM ammonium results in a 5-6 fold increase in specific activity. High concentrations of ammonium result in repression of activity even when ornithine or proline are present. These data indicate that proline oxidase in wild-type is repressed by ammonium and therefore regulated by nitrogen control. Two nit-2 alleles were assayed for proline oxidase activity, amr and nit-2 (nr-37); both are mutations of the ammonia regulatory systems. Proline oxidase in both mutations was induced by proline and ornithine similar to wild-type. There is the additional similarity that both strains were derepressed by 1 mil ammonium with or without inducer present. This result is contrary to the expected behavior of nit-2 mutants, which cannot derepress synthesis of many nitrogen- related enzymes during growth on low-ammonium medium. Proline oxidase was also assayed in an ota mutant, which is deficient in ornithine transaminase (OTA). The data show that the activity in ota after growth in the presence of 25 mM ammonium plus ornithine was only slightly greater than basal levels in wild-type. It appears, then, that ornithine can only partially induce proline oxidase in the ota mutant. In Aspergillus nidulans, ornithine induction of proline oxidase requires the conversion of ornithine to glutamate-

(T-semialdehyde by OTA (65). The data presented here indicate that, as in Aspergillus, the induction of proline oxidase by ornithine in

Neurospora probably occurs via a metabolized form of ornithine. Table 6

Proline Oxidase Specific Activity (xio-?)

K-Source lOxM Pro 10x14 Pro lOmM Orn 10x1! Orn Genotype 25rl! NH*+ LxM 10x1' Pro +25nM riH4+ +1x14 KH4+ 10mM Orn +LnM N H 4+ +25x11 M U + CO 0

W.T. (7^-0R23-lA) • 5.8 1.5 0.9 2.1+ 3-7 5.1 1.1 anr 1.0 6.5 2.5 1.1+ 5.1+ 2.3 1+.1+ 1.7

Nit-2a 167 1.0 6.7 2.5 1.5 5-1+ 2.3 5-3 0.86 ota 2.5 1.6

Ui fO Allantoinase stabilization. It has been stated-that allantoinase is very unstable in vivo (60). Fig. 7 shows that in vivo, allantoinase activity has decayed 50% in 38 minutes. This half-life represents the decay of activity under repressing conditions but in the presence of continued protein synthesis. Repeating this experiment: with the additions of 10 ug/ml cycloheximide to the repressing medium (10 mM ammonium) yielded a half-life of 14 minutes. (Fig. 8).

These data agree well with the calculated half-life of 20 minutes by Reinert (60). Allantoinase turnover is not prevented or slowed by hypoxanthine nor is its degradation rate enhanced by ammonium

(70). Furthermore, allantoinase activity is not subject to feedback inhibition by ammonium, since even 10 mil ammonium does not inhibit the enzyme in_ vitro. Therefore an increase in the cellular level of allantoinase activity appears to result from an increase in its rate of synthesis and not from a change in its degradation rate.

Since allantoinase activity is very labile in vivo, it was of interest to determine whether the enzyme was likewise inactivated in cell-free extracts. At 25 C, allantoinase activity was found to decay rapidly with a half-life of about 38-40 minutes, which agrees favorably with its in vivo half-life in the absence of cycloheximide.

(Fig. 9). The decay of activity at 0 C was noticeably slower than at 25 C. Allantoinase undergoes decay at a rapid rate both ir^ vivo and iui vitro. The difference between decay at 0 C and 25 C iii vitro may indicate that allantoinase turnover is an active process mediated by proteolysis. If this is the case, then the rate of decay may be reduced by the addition of- protease inhibitors. A retardation of decay would obviously be advantageous for the isolation of allantoinase. Fig. 7. Allantoinase Stability In Vivo. Mycelia were grown for

20 hours in Vogels medium minus nitrogen supplemented with 1.5 mM ammonium tartrate plus 1.0 mM hypoxanthine, after which the cultures were transfered to 10 mM ammonium tartrate and sampled at various times as indicated.

54 o o MAX. ACTIVITY IOOQ- 50 75 2:5 30 0150 60 TIME (min) 90 120 180 Ln Ui Fig. 8. Allantoinase Stability In Vitro in the Presence of Cyclo­ heximide. Wild-type cells were grown in medium with 10 mM ammonium tartrate and 1.5 mM hypoxanthine for 20 hours, and then transfered to medium containing 10 mM ammonium tartrate plus 10 ug/ml cycloheximide and cells sampled at various times as indicated.

56 % % MAX. ACTIVITY 100 50 5 - 25 75 10 20 IE ( TIME min ) 30 050 40 60 -vjLn Fig. 9. Stability of Allantoinase Iii Vitro. Wild-type cells were grown in medium containing 0.5 mM ammonium tartrate and 1.0 mM hypoxanthine for 20 hours to induce allantoinase. A cell-free extract was prepared and samples were incubated for 0-4 hours at either 0 C (o) or 25 C ( ).

58 % % MAX. ACTIVITY 100 75 50 25 IE r ) (hr TIME VO Ln Stabilization of allantoinase in vitro. In order to study allan­ toinase it was necessary to stabilize the enzyme preparation. Three protease inhibitors were individually added to crude, cell-free extracts of allantoinase and activity was assayed after various times at 0 C

( Fig. 10). The chelator EDTA was able to provide a significant degree of stabilization and about 80% activity was retained over four hours. PMSF (phenylmethylsulfonyl fluoride), and the potent enzyme inhibitor PCMB (p-chloromercuribenzoate), do not reduce the rate of decay at 25 C. It was concluded that EDTA is capable of stabilizing allantoinase in vitro and it was therefore used in all buffers during subsequent enzyme isolations.

In addition to adding protease inhibitors, sulfhydryl reagents were also added to cell-free extracts to determine if they would stabilize allantoinase by reducing its rapid decay rate. Sulfhydryl reagents such as 2-mercaptoethanol and dithiothreitol (Cleland's reagent) protect many enzymes by maintaining sulfhydryl groups in the reduced state. Fig. 10 shows that 2-mercaptoethanol stabilized allantoinase since 95% of the original activity was retained over a six hour period. Dithiothreitol gave partial protection. In addition to EDTA, 2-mercaptoethanol was thus included in the isolation buffer: 20 mM Tris-HCl buffer, pH 7.2, 1 mM EDTA, 1 mM 2-mercapto­ ethanol.

Isolation of allantoinase. Allantoinase in cell-free extracts was purified by precipitation and Sephadex gel filtration

(Fig. 3). Allantoinase was precipitated by (NH^^SO^ at 50% satu­ ration, resulting in a 4-fold purification (Table 7). The 50% fraction was then passed through a Sephadex G-150 column and fractions 52-60 were Fig. 10. Allantoinase Stability In Vitro in the Presence of

Protease Inhibitors. Wild-type cells were grown in medium containing 1.5 mM ammonium tartrate and 1.0 mM hypoxanthine for 20 hours. Cell-free extracts were prepared and samples were incubated at 0 C in the presence or absence of EDTA (10 M) o, PMSF (10 \l)A, PCMB (10-^M)Q, and extract alone, ©, for various times when remaining enzyme activity was assayed.

61 loon-

75 >- b > o < 50 ✓s, <

25 a

0 0 4 TIME ( hr) O'* ro Fig. 11. Allantoinase Stability In Vitro in the Presence of

Sulfhydryl Reagents. Wild-type cells were grown in medium containing 1.5 mM ammonium tartrate and 1.0 mM hypoxanthine

for 20 hours. Cell-free extracts were prepared and samples were incubated at 0 C with either 1 mil 2-mercaptoethanol, (o) or 1 mM dithiothreitol, (A) . Control values (®) from experiment of Fig. 10 are provided for comparison.

63 ON o cji r\3 r\3 cji -si o ' ' % MAX. ACTIVITY o oi o o ro oj TIME ( hr) 65

Table 7

Fractionation of Cell-free Extracts by (NH4)2S04 Precipitation

Fraction Allantoinase (j (NH4 )pSO4. Saturation)______mg Protein/ml______(Specific Activity)

Crude extract 36.0 0.88

0-30 5.2 0.11

30-50 11.3 3.92

50-70 15.8 0 .6^

70-80 3.2 0.0 66 collected and concentrated (Fig. 12). The degree of purification at this step is 20 to 30 fold; however, if the enzyme preparation is stored overnight at -4 C, its loss of activity was of such magnitude

(70-80% loss) to give an apparent purification factor of only six.

Fractions containing activity were then passed through a Sephadex G-100 column. This step results in the separation of the enzyme into two peaks of activity, I and II (Fig. 13). It will be shown that type I is the normal enzyme while type II is a small molecular weight degra­ dation product of allantoinase. Fractionation of the concentrated

G-150 fractions by the Sephadex G-100 results in an increased purity of 13-fold (considering only type I activity). The actual purification can be evaluated with the use of gel electrophoresis. (Fig. 14).

Fractions which possess activity contain about 11 major proteins and some minor ones as well. The G-100 gel filtration step results in the purification of allantoinase to only two proteins, the enzyme and one contaminant.

Molecular weight of allantoinase. Purified allantoinase was passed through a calibrated Sephadex G-200 column, and by its elution position, its molecular weight was calculated to be about 120,000 MW (Fig. 15).

Similar samples of allantoinase were also run on 10% polyacrylamide gels containing 0.1% SDS. These data (Fig. 16) give a molecular weight of about 37,000 MW, thus indicating that allantoinase is a multi-, meric enzyme of 120,000 MW composed of subunits of molecular weight approximately 37,000.

Does allantoinase exist as a component of a multi-enzyme complex?

Allantoinase and uricase, enzymes for two steps in the purine catabolic pathway, were assayed in cell-free extracts and allantoinase preparations Fig. 12. Elution of Sephadex G-150 Column for Allantoinase.

A sample of the 50% (NII^) 2SO^-precipitation-fraction was run through a Sephadex G-150 column (2.5 x 45 cm), and eluted with

20 mM Tris-HCl buffer, pH 7.2 plus 1 mM EDTA and 1 mM 2-mercapto- ethanol. Allantoinase activity was assayed and fractions 50 to 62 OOQ (1.5 ml per fraction) were collected and pooled. A (o o) , allantoinase activity, (o o).

67 oo 1 ----- 1 to to o ----- CD 1 ABSORBANCE280nrn ALLANTOINASE SPECIFIC ACTIVITY

FRACTION NUMBER Fig. 13. Elution Profile of Sephadex G-100 Column for Allantoinase.

A sample of the combined fractions containing allantoinase activity from the G-150 column was applied to a Sephadex G-100 column and eluted with 20 mM Tris-HCl buffer, pH 7.2 plus 1 mil EDTA and 1 mM 230 5 35 2-mercaptoethanol. A (o o), and allantoinase activity (A ),

(o ©) .

69 O o bi o o o j ABSORBANCE280nm ro o ALLANTOINASE ACTIVITY A535 Ol 03 t\3 OJ

FRACTION NUMBER Fig. 14. Polyacrylamide Gels of G-150 and G-100 Allantoinase

Preparations. 1) A 50 ul sample of allantoinase-containing activity from Sephadex G-150 column run on 5% polyacrylamide gel. 2) A 50 ul sample of type I allantoinase isolated from G-100 column and run on

10% polyacrylamide gel. 3) A 50 ul sample of type I allantoinase and run on a 10% polyacrylamide gel containing 0.1% SDS.

71 1 2 3 Fig. 15. Molecular Weight Determination of Allantoinase.

A Sephadex G-200 column was calibrated with aldolase, 158,000 MW,

(Kav = .132), bovine serum albumin, 68,000 MW (Kav = .276), and myoglobin, 18,500 MW (Kav = .516). Allantoinase had a Kav = .180 which corresponds to a molecular weight of 120,000.

73 MOLECULAR WEIGHT (xIO '4) 20 0 5 10 0 2 5 0 0.2 L OI E T ) I E P TY ( SE A IN TO N LLA A 4 . 0 3 . 0 v a ^ 5 . 0 0.6 7 . 0 74 Fig. 16. Molecular Weight Determination of Allantoinase by SDS-

Polyacrylamide Gel Electrophoresis. Standard proteins and allan­

toinase were run on 10% polyacrylamide gels containing 0.1% SDS, with a Tris-HCl buffer. Proteins used for calibration were: bovine

serum albumin, 68,000 MW (Rf = .240), ovalbumin, 43,000 MW (Rf = .40),

trypsine 23,000 MW (Rf = .82), and myoglobin, 17,200 MW (Rf =1.0).

Allantoinase had a Rf = .50 and a corresponding molecular weight

of 37,000.

75 MOLECULAR WEIGHT (x|0-3 )

ro CJl o

o .

O CD

O CT>

o 00 o CD o from both G-150 and G-100 gel•filtration steps (Table 8.) In crude extracts both activities were found in addition to allantoinase. In both G-150 and G-100 allantoinase fractions, uricase was not detected.

Allantoicase activity was not found associated with pure allantoinase and only as a residual activity in. the G-150 allantoinase fraction.

These data suggest then that allantoinase does not exist in a multi­ enzyme complex with uricase or allantoinase.

Michaelis constant of allantoinase. The Km of allantoinase under the assay conditions described is 0.13 mM (Fig. 17). Vogels (71) reported Km values for allantoinase with a different set of assay conditions and degrees of purity of the enzymes. This could account for the difference between the Km reported here for Neurospora crassa and that found in various prokaryotes.

Effect of cations on allantoinase activity. The data of Table 9 indicate that additions of Co+^, Fe+^, Zn+^, Ca+^, and Cd+^ signi­ ficantly inhibit allantoinase activity. The effect of these cations on allantoinase from different organisms is varied. In mung beans, + 2 4-2 Cu inhibits while Mn only slightly inhibits activity (72).

In Saccharymyces cerevisia, allantoinase was insensitive to a variety of cations (83). Vogels (71), in his survey of allantoinase from +9 +2 +2 different organisms, found that Zn , Co , and Cd inhibited allantoinase activity from most organisms (Streptococcus allantoicus,

Arthrobacter allantoicus, Escherichia coll, Pseudomonas acidoverans,

Ps. fluorescens, livers of Rana esculenta and Carassius auratus) except for higher plants (Glycine hisplda and Phaseolus hysterlnus).

+2 Mn stimulated the activity of enzyme from higher plants, had only

a slight inhibitory effect on animal liver allantoinases and strongly 78 Table 8

Lack of Association of Allantoicase and Uricase with Allantoinase

1 Enzyme Activity Fraction Allantoinase Uricase Allantoicase

Crude .180 .088 .05*+

G-150 0.2k 0.00 .021

G-100(I) o.Qk 0.00 0.003 lAllantoinase and allantoicase activities were assayed at A535 and uricase activity was assayed at A A990. Fig. 17. Lineweaver-Burk Plot of Allantoinase Data. The initial velocity of allantoinase activity was determined at varied substrate concentrations. The incubation mixtures contained varying allantoin concentrations in 1.2 ml 100 mM Tris-HCl buffer, pH 7.2, plus 0.1 ml allantoinase in 20 mM Tris-HCl buffer, pH 7.2, plus 1 mM EDTA and

1 mM 2-mercaptoethanol. The Km of Neurospora allantoinase calculated from these data is 0.13 mM.

79 I

I o I 03 I CD

I

i l/v {jj. moles /m in) r o

• • — r\3 o j Ul v 01 o o

0 3 o Table 9

•'•Effect of Bivalent Cations on the Activity of Allantoinase

'•'Cation '■°lo Activity

No addition 100

CaCl2 70.0

Co (N03)2 k 8

Na2Mo 81

FeCl3 11.3

ZnS04 2.6

CdCl2 0.9

BaClp 72

MgS04 82.0

MnS04 83.6

CuS04 10.6

1Allantoinase activities were assayed in mixtures containing: 1.2 ml, 5.7 M-g/ml allantoin in lOOmM Tris-HCl buffer, pH 7 -^5 0.1 ml enzyme and 0.1 ml cation. ^Cation concentration was lmM, except CUSO4 which was O.lmM. 3A11 activities were reported as % of full activity of purified allantoinase. inhibited Pseudomonas allantoinase.

The diversity of responses to cations in a number of organisms does not allow conclusions to be drawn except that there is no correlation between the response of allantoinase activity to cations and an organism's phylogenetic position.

Presence of metal ions in allantoinase. Table 9 indicates that Co +2 ,

+2 +2 +2 +2 Fe , Zn , Cd , and Cu strongly inhibit allantoinase activity. The mechanism of inhibition could be the exchange of these ions for a metal associated with the enzyme. To examine more closely the question of a metal present in allantoinase a series of chelators were added to allantoinase to determine their effect on activity

(Table 10). The general chelator EDTA exerted little inhibitory effect at a concentration of 1 mM. Sodium azide is known for its inhibitory effect on Fe-containing enzymes like catalase at concen­ trations of 1 mM. The data indicate that azide did not inhibit allantoinase activity to a great extent, and therefore Fe is not likely to be present in the enzyme. This conclusion is further supported by the lack of inhibition by 1, 10-phenanthroline which +3 +2 is a chelator of Fe and Zn , of metalo-enzymes. Sodium fluoride +2 is a strong chelator of Mg , but its effect on allantoinase activity is relatively insignificant, thus Mg+2 probably is not associated with allantoinase. KCN and phosphate buffer both had a significant inhibitory effect on allantoinase activity. Cyanide acts as a general inhibitor by complexing with metals in metalo-enzyme complexes at concentrations of 10 — 3 -10 -5 M. Phosphate, depending upon the pH,

“12 +2 +2 is known to bind metals, particularly Mg , Ca , and Mn . A third

inhibitor, thiourea, resulted in significant inhibition of activity Table 10 aEffect of Chelators of Metal Ions on Allantoinase Activity

Chelator Concentration 2cjo Activity

KCN ImM 20.0

Ha"Azide ImM 73.3 lOmM 63.8

NaF ImM 89.1 lOmM 78.7

Na’Arsenate ImM 6 0 .1*

1 ;10-Phenunthroline ImM 97.7

P04 buffer, pH 7-0 20mM 20.0

Thiourea ImM 6U.7 lOmM 15.1*

EDTA JmM .91.5 lOmM 61.9

•LAllantoinase activities were assayed, in mixtures contain­ ing: 1.2 ml, 5-7 M-g/ml allantoin in lOOmM Tris-HCl buffer, pH 7.1*, 0.1 ml enzyme and 0.1 ml chelator. •^Activities reported as % of full allantoinase activity. particularly at 10 mM. Thiourea forms complexes with metal groups,

+2 particularly Cu . These data (Tables 9 and 10) suggest that a metal group might be associated with allantoinase activity, but are not sufficient to indicate its identity.

Effect of sulfhydryl reagents. Allantoinase is inhibited by 1 mM glutathione (reduced), cysteine, iodoacetate, and PCMB (p-chloro- mercuribenzoate)(Table 11). Iodoacetate and PCMB each react with sulfhydryl groups and can be used to detect essential groups by their inhibition of enzymatic activity. Cysteine and glutathione inhibition also indicate that a sulfhydryl group in the enzyme is essential for activity. PMS (phenazine methosulphate), is also reactive with sulfhydryl groups and the data of Table 11 reveal that increasing concentrations of PMS inhibit allantoinase activity, again supporting the hypothesis that sulfhydryl groups are essential for activity. Yeast allantoinase was found to bev inhibited by high concentrations of iodoacetate, cysteine, and glutathione (73).

Allantoinase from fish and frog livers and from higher plants is sensitive to reducing compounds but that from prokaryotes is not

(71).

Feedback inhibition. Many enzymes are regulated by feedback inhibition by end products of intermediates of the pathways. To determine if allantoinase is regulated in this manner, uric acid, urea, glutamate, and ammonia were incubated with allantoinase and their effect on activity was noted. The data (Table 12) indicates that allantoinase was not significantly inhibited by any of these compounds (tested at

1 mM concentration). Allantoinase, then is not sensitive to either an early intermediate or final products of purine catabolism, nor Table 11

1Effect of Reducing Substances on Allantoinase

Substance Concentration WcJ3 Activity

None (control) - 100

Glutathione ImM 0.0

Cysteine ImM 0.0

Iodoacetate ImM 13.3

PCMB ImM 10.0

PMS O.lmM 85.5 ImM 62.k 5mM 27.6

-‘Allantoinase activities were determined at 30°C in mixtures containing 1.2 ml substrate containing 5-7 ng/nil allantoin in lOQmM Tris-HCl buffer, pH 7.h, 0.1 ml enzyme and 0.1 ml reducing agent. Activities were reported as °/0 of specific activity of purified allantoinase. 86 Table 12

1Effect of Purine End Products and Glutamate as Potential Feedback Inhibitors of Allantoinase

■-Substance A Activity

Uric acid 71

Urea 95

Glutamate 92

n h 4+ 6k

■’■Allantoinase activities were assayed in mixtures contain­ ing: 1.2 ml, 5-7 |ig/ml allantoin in lOOmM Tris-HCl buffer, pH 7.U, 0.1 ml enzyme and 0.1 ml potential inhibitor. 2 The concentration of each compound tested was ImM. Activities reported as °]o of full allantoinase activity. to glutamate, which represents a key intermediate in nitrogen metabolism.

Thermal stability of allantoinase. The effect of a 10 minute pre­ incubation of allantoinase at various temperatures with and without substrate present is shown in Figs. 18 and 19. Allantoinase activities, type I and II were assayed over a 60 C range (Fig. 19). Allantoinase I is rapidly inactivated at temperatures of 40 C and higher. Type II activity is stable at most temperatures. Allantoinase I was also treated at various temperatures in the presence and absence of sub­ strate, allantoin (Fig. 19). There appears to be no significant protection of allantoinase by its substrate from heat inactivation.

Vogels (71) showed that various allantoinases in crude preparations were stable at 40 C but inactivated by heating at 55 C, which is in agreement with my data.

The stability of purified allantoinase in both G-150 and G-100 preparations was investigated by incubating the enzyme for several hours at either 0 C or 24 C and then assaying activity (Figs. 20,21').

Allantoinase from the G-150 fractions undergoes a rapid inactivation at 24 C such that by two hours only 21% activity remained (Fig. 20).

At 0 C, inactivation is slox^er; after six hours, 42% activity remained

(Fig. 20). Allantoinase I, recovered from G-100 column, is not stable at either 0 C or 24 C (Fig. 21). After three hours, an 87% loss of allantoinase activity had occured. Type I activity in a purified state, maintained under the conditions described, is very labile as seen by the rapid enzyme inactivation, even at 0 C.

Allantoinase sensitivity to proteases. Allantoinase, both type I and

II, were preincubated with proteases (2000:1 wt/wt ratio) for 30 minutes at 37 G prior to assay (Table 13). Type I enzyme was very Fig. 18. Thermal Stability of Allantoinase I and II. Allantoinase samples (0.1 ml) were incubated at each temperature for 10 minutes and then assayed under standard conditions for activity. Allantoinase type I (0 o), and allantoinase type II (o -o).

88 CO VO CJJ OJ -£> 0 1 o j OJ CD r o SPECIFIC ACTIVITY! SPECIFIC SPECIFIC ACTIVITY 31 ACTIVITY SPECIFIC OJ Ol o o o OJ TEMPERATURE (C°) Fig. 19. Thermal Stability of Allantoinase With and Without the

Presence of Substrate. Type I allantoinase was incubated at various temperatures for 10 minutes either in the presence or absence of allantoin (5.7ug/ml). Allantoinase activity was assayed at the end of each incubation period. Allantoinase incubated without substrate

(0--- 0), allantoinase incubated with substrate (o o).

90 SPECIFIC ACTIVITY I

ro o j

H m OJ “□ o m a J $ cz o rn

CD o

CD o

-n! o ro ro CD CO ro CD CD o SPECIFIC ACTIVITY H

o Fig. 20 • Stability of Allantoinase at 0 C and 24 C. Allantoinase isolated from Sephadex G-150 fractions was incubated at either 0 C or 24 C and activity at various times as indicated. 0 C, (©--- ©),

24 C (o o).

92 ALLANTOINASE ACTIVITY

vOu> Fig. 21. Stability of Allantoinase I at 0 C and 24 C. Allantoinase, type I activity, recovered from the Sephadex G-100 step, was incubated at either 0 C or 24 C and at various times as indicated, activity assayed. 0 C (o--- o) , 24 C (o o) .

94 Lf» VO o * *00 bo ALLANTOINASE ACTIVITY ro oj O TIME (hr) Table 13

Allantoinase Sensitivity to Proteases1

% Activity? Protease Type I Type II

Trypsin 6.3 73.2

Chymotrypsin 0.0 86.5

Protease V 39-1 83.7

Pronase ^•3 98.0

Control (no protease) 100 100

1Allantoinase, type I and II were incubated with protease at a ratio of 2000:1 (allantoinase to protease) for 30 min at 37°C. ^Activity remaining is reported as % of allantoinase activity present in a control not subjected to protease treatment. sensitive to trypsine. chymotrypsin, pronase, and protease V. Type II activity was relatively stable to protease treatment. Since allantoinase (type I) activity is very sensitive to protease digestion in vitro, it may be equally sensitive iin vivo and could thus be the mechanism for its turnover in vivo.

Characteristics of allantoinase type II activity. It was already described that upon passing semi-pure allantoinase through a Sephadex

G-100 gel filtration column, two peaks of allantoinase activity are resolved. Allantoinase II is very different from type I in many characteristics, such as protease sensitivity and thermal stability as already shown. A particular characteristic of type II activity is that when it is simply added to a substrate mixture as a zero time blank, the blank has a large absorbance value that is equal to that of an incubated sample. This fact allows for a very rapid identi­ fication of type II activity versus type I activity! Table 14 summarizes properties of type II activity... Type II activity is not due to a subunit of type I. The evidence for this includes its loss, without any loss of total protein, upon dialysis.,: thus indicating a small size.

The size of type II is about 320 molecular weight (Fig. 22). This very small size explains why type II could never be stained by Coomassie blue in polyacrylamide gels or cellulose acetate strips. When type II material is spotted on filter paper and treated with ninhydrin spray, there is no detectable staining, indicating no free amino acids or small peptides having free primary amino groups present. These data imply that type II activity is not associated with a protein, which thus accounts for its protease resistance and thermal stability. The exact nature of type II is unknown, but it appears to be derived from Table lb

Properties of Type II Allantoinase-like Activity

Property Results Found with Type II "Activity"

Kinetics Non-linear

Thermal stability Stable after 5 ®in at 100°C

Dialysis Loss of activity upon dialysis. Recovery in dialysate.

Inhibition by 5-HAA Not inhibited

Protease sensitivity Insensitive- to digestion

Ninhydrin reaction Is not detected by ninhydrin reagent

Molecular weight Approximately 320 Fig. 22. Molecular Weight Determination of Allantoinase Type II.

A Sephadex G-15 column was calibrated with glycylleucylglycylleucine

(412 MW), leucylglycylleucine (291 MW), leucylglycine (209 MW), and arginine (174 MW). Type II allantoinase-like activity had a size corresponding to 320 MW.

99 100 o MOLECULAR WEIGHT (XIO"2) WEIGHT MOLECULAR o w O ho o 0.4 0.5 o type I allantoinase, the naturally occuring enzyme, as a degradation product of the enzyme. There are several lines of evidence supporting

this conclusion. When type I allantoinase is dialyzed in the presence of 0.1% SDS, a type II activity is released and recovered in the dialysate (Table 15). Without SDS present dialysis does not alter activity of type I enzyme, nor is any "activity" found in the dialysate. The concentrated dialysate of SDS-treated type I enzyme was passed through a Sephadex G-15 column to determine its position

in relation to the dialysate of dialyzed type II material (Fig. 23).

A peak of allantoinase-like activity from the dialysate of type I

appears at the same position as that of type II activity indicating

that type II material was released from SDS-treated enzyme. It can

also be shown that allantoinase degradation results in the shift of 280 A material and allantoinase-like activity to a small form which

coincides with type II behavior. Fig. 24 illustrates this shift of 280 A material upon incubation of allantoinase at 30 C for 12 hours.

The data indicate that purified allantoinase (isolated from the G-100

column) undergoes degradation and, probably, a concomitant release,

of a small molecular weight product which has allantoinase-like

behavior.

In vitro turnover of allantoinase. It has already been shown that various preparations of allantoinase undergo decay of activity, the

release of small degradation products, and some degradation of protein.

If it is assumed that these observations are the result of the same

process, then in vitro degradation could be occuring in one of two ways: instability of allantoinase molecules or proteolytic digestion

or some related process. Type I allantoinase isolated from 102 Table 15

SDS Treatment of Normal Allantoinase (Type i)1

-Activity Initial After Dialysis

Sample • 035 • 035 - SDS Dialysate 0.0 0.0

Sample .03^ 0.0 + SDS Dialysate 0.0 .061

IPuxe type I allantoinase was dialyzed (dialysis tubing with 6000 m.w. pore size against buffer in the presence and absence of 0.1$, SDS. Dialysis was for 18 hours (l ml sample against 100 ml buffer), and at the end of this period the dialysate was concen­ trated by rotory evaporation at 37°C under vacuum. Allantoinase activity of the sample and concentrated dialysate was assayed. ^Activity is in A535 units. Fig. 23. Gel-Filtration of SDS-Dialyzed Type I Allantoinase. Type I allantoinase was dialyzed against buffer plus 0.1% SDS for 18 hours.

The dialysate was concentrated by rotary evaporation and then passed 280 through a Sephadex G-15 column. Fractions were assayed for both OD

(o ©) and allantoinase-like activity (o o). The arrow designates the elution position of type II activity resolved from type I by

Sephadex G-100 step of purification.

103 FRACTIONS Fig. 24. Degradation of Type I Enzyme. Purified allantoinase, type I, was incubated for 12 hours at 30 C and run through a Sephadex G-100 column. Fractions were monitored for their absorbance at 280 nm.

Zero time, (©--- ©); 12 hr. (o o). I indicates the position of type I activity and II indicates the position of type II activity.

105 2 8 0

o o o ro oj o

ro

G) ~n X) o H oo o

co o

ro 1 = 1

oo

901 the Sephadex G-100 column was,assayed for protease activity (Table 16).

The protease activity detected by two different assays is so low as to make analysis difficult. When crude preparations of proteases are assayed, values as given in Table 16 would be considered as background readings for no protease activity. The question then is whether this low level of activity is responsible for the In vitro stability of allantoinase?

If proteases are responsible for the turnover of allantoinase, then an in vitro turnover system can be constructed with extracts of aln strains. The aIn strain lacks allantoinase activity and in cell-free extracts contains an activity which induces the decay of allantoinase activity when added to pure allantoinase (Table 17).

The data show that allantoinase is sensitive to some factor in the aln extract which is non-dialyzable, insensitive to EDTA and sensitive to PMSF. This may indicate that a protease is actively degrading allantoinase in this in vitro system. Although the data imply that allantoinase is labile to protease degradation, caution is advised when extending these iri vitro experiments to the situation which occurs in vivo. Table 16

1Protease Activity of Allantoinase I Preparation

Time (min) Assay 20 40 50 100

Lowry a780 .029 .060 2Sp. act. .006 .007

Casein Yellow A428 .011 .035 3Sp. act. .0009 .0014

•’•Allantoinase preparations were assayed for proteclytic activity according to the Casein Yellow and Lowry Protease assays. 2Specific activity was reported as A'"5°/min/mg protein. 3Specific activity was reported as A42s/min/:mg protein. Table 17

Treatment of Allantoinase by aln Extract

Addition to Allantoinase A Activity

None (control) 100

aln extract 0

aln extract dialyzed 0

aln extract + ImM EDTA 28

aln extract + ImM PMSF 73

Allantoinase (isolated from G-100 step) was incubated for 5 min at 30°C with 0.05 ml of aln cell-free extract and then assayed as described previously. Activity was reported as °jo of maximum activity of allanto­ inase after 5 min incubation at 30°C. DISCUSSION

Neurospora crassa is able to utilize certain amino acids as sole nitrogen source, whereas other amino acids cannot support growth. Wild-type and the amr mutant both can use the same favored amino acids: glutamine, glutamate, aspartate, and arginine. The difference between these two strains is in their ability to utilize the remaining amino acids, wild-type being able to maintain a more vigorous growth than amr. It was a goal of this research to determine whether the amr gene regulated the catabolic enzymes for certain amino acids. The arginine, ornithine, proline degradative pathway was chosen for study for three reasons: (1) arginine is utilized very well by both wild-type and amr and possible regulation of its use can be contrasted with (2) ornithine, which wild-type used with greater efficiency than amr, and (3) proline which is poorly utilized by both strains. This pathway is involved in the metabolism of three amino acids, each of which is utilized to a different degree.

Arginase, the first enzyme of arginine metabolism is not under nitrogen regulation. Ornithine is next deaminated by ornithine transaminase (OTA), which, like arginase, is also regulated by induction in both wild-type and amr. Glutamate-semialdehyde or its cyclic form. Fyrrol:i.ne-5~carboxylate is converted to glutamate by pyrroline-5-carboxylate dehydrogenase and this enzyme is not under • ammonium repression control. This is in contrast to Aspergillus,

110 where P-5-C dehydrogenase and proline oxidase are both nitrogen- controlled. The only enzyme of this pathway in Neurospora which was shown to be regulated by ammonium was proline oxidase, the first committed step of proline degradation.

Since the catabolism of arginine and ornithine is not regulated in a way which could account for the observed differences in their use by wild-type and amr, I examined the transport of these amino acids, another point at which control might be exerted. The results show clearly that in wild-type, amino acid uptake is derepressed by low concentrations of ammonium and that this response is lacking in amr. Amino acid transport in Neurospora is commonly described as occurring via three major systems and several minor ones. The neutral system (AA I) is specific for most aromatic and neutral amino acids

(high affinity for tyrosine, leucine, and phenylalanine and a moderate affinity for alanine, histidine, serine, glycine, and methionine).

A broader range of specificity is exhibited by the general system

(AA II) which transports neutral and basic amino acids, including arginine, phenylalanine,' glycine, tyrosine, and leucine. The third major transport system is the basic (AA III) system, specific for basic amino acids and ornithine (74). Each of these systems have been defined by mutations (such as pm-n, pm-g, and pm-b). The general and neutral amino acid transport systems were examined using the pm-g and pm-n mutants to study transport of amino acids under conditions of high and low ammonium. The results indicated that both transport systems are controlled by ammonium-repression. This is in agreement with previous work in which all three amino acid transport systems were derepressed by low ammonium (75). A recent 112 study (76) has indicated that AA II decreased in activity in parallel with a decrease in the concentration of ammonium in the medium. This report is contrary to the evidence presented here. It may indicate that the general transport system is not under ammonium repressible control.

Proline is unlike the other amino acids examined in that the first committed step in its use is catalyzed by an ammonium-regulated enzyme. Proline oxidase is regulated in wild-type and in amr. Arst and Cove (40) found that the synthesis of certain nitrogen-regulated enzymes was different in various areA mutant strains of Aspergillus.

The derepression of proline oxidase in Neurospora could be similar and different amr alleles may have different responses to ammonium.

Only one other allele (nr-37) was tested for proline oxidase activity and it behaved similar to amr. Examination of additional amr alleles might show that proline oxidase is under ammonium-repressible control and that some amr mutants are unable to respond by derepression during low ammonium concentration.

The regulation of proline degradation is interesting, not only because of proline oxidase, but because the uptake of proline is not under ammonium-repressible control as is transport for the other amino acids tested (77). Similarly, proline uptake in Aspergillus is not controlled by ammonium-repression.(44). Proline transport occurs by a separate permease and, therefore, its control can be separate from that of the other amino acids.

In general terms, amino acids which lack at least one ammonium- regulated enzyme in their degradative pathway seem to transported by permeases which are subject to control. Those amino acids which have a catabolic enzyme under nitrogen-regulation need not have their specific permease controlled as well. This generalization is attractive because nitrogen-regulation could easily be exerted over a number of amino acid degradative pathways at one common point, their transport systems. This would provide an efficient means to control a number of pathways with the minimum number of regulatory sites. Amino acids such as proline which have their own specific permeases can be regulated at either a critical enzyme within the pathway or at the permease.

Table 18 presents a summary of nitrogen-control of amino acid and purine metabolism. A diversity of controls are applied to these enzymes. Arginine, ornithine-transaminase and P-5-C dehydrogenase are only inducible and not subject to ammonium-repressible control.

The purine degradative enzymes uricase, allantoicase, and allantoinase require not only that inducer be present but also need a functional amr product as well for their expression. Amino acid transport represents a repressible system where only nitrogen-regulation is operating.

The amr mutants lack or have only basal levels of ammonium- repressible enzymes related to nitrogen metabolism. This amr locus appears to act as a positive regulatory gene for synthesis of these enzymes. It is postulated that amr may encode a protein which binds at a promoter-like site adjacent to each structural gene and is necessary for transcription. Ammonium is though to interact with the amr gene product which alters it, such as by a change in confor­ mation, so that it is no longer capable of binding to the promoter.

Thus, the presence of high concentrations of ammonium would result in the repression of synthesis of the nitrogen-regulated enzymes. Table 18

Summary — Nitrogen Control of Amino Acid. Metabolism

Regulated Synthesis Altered Enzyme Inducer (Repressed) by NH4+ in AMR Mutant PRESENT RESULTS

A.rginase Arginine NO NO

Ornithine Transaminase Arginine > Ornithine NO NO

pyrroline-5-carboxylate Proline > Ornithine NO NO dehydrogenase

Proline Oxidase Proline. Ornithine YES . NO

Amino acid transport — YES YES

OTHER WORK (W.R. Reinert)

Uricase uric acid YES YES

A.llantoicase uric acid YES YES

Allantoinase uric acid YES YES 114 115 This suggested arrangement of regulatory genes and structural genes

is illustrated in Fig. 25. The structural genes which are both

inducible and repressible, such as the purine catabolic enzymes,

are predicted to have two regulatory sites, a promoter for the amr protein and another to bind a second regulatory protein. This second

regulatory protein would bind the inducer (uric acid for uricase,

allantoicase for allantoinase) and the active protein-inducer complex

could then bind to the second promoter. It is further proposed that

the simultaneous presence of both of these regulatory proteins would be required for transcription to occur (e.g. for the purine catabolic

enzymes). Genes which are regulated by ammonium-repression only would

possess only an amr-protein promoter site. Inducible enzymes like

arginase and ornithine transaminase which are not repressible would

likely require a regulatory protein which binds the particular

inducer and which is specific for that structural gene. These

specific regulatory proteins would be encoded by minor nitrogen control

genes.

This model contains several assumptions: promoter-like sites

exist adjacent to structural genes, and that ammonium is a regulatory

effector or co-repressor. To date, there is no evidence for promoter­

like sites for amr in Neurospora, but in Aspergillus, Arst (44) has

identified an initiator constitutive mutation with an up-promoter

effect for the ammonium-regulated activity, xanthine-uric acid permease.

This may represent a binding site or promoter for the areA gene

product. It is not unreasonable to expect to find such sites in

Neurospora which will act .as specific promoters for the amr gene

product. Fig. 25. Model of Nitrogen Regulation.

116 117

6

Major Nitrogen Minor Nitrogen Control Gene Control Gene , a m r. pur-c nit-c, ,arg-c,

+ Inducer (ay)

Active 1+) In active (+) Active {+) + Regulatory Regulatory Regulatory Protein Protein Protein -Mnducer

Inactive Form Active Form + / Repressible System JInducible Systems: / H.H. Permease Inducible/Rep. arginase Genes: Purine + f PmG Cat. Enz. arg , H

Permease Allantoinase Arginase 118 The second assumption is that ammonium itself is the co-repressor

for nitrogen-regulation. It is possible that the repression which occurs in the presence of high ammonium concentrations is due instead

to the metabolism of ammonium into the true co-repressor. This could be accomplished by transamination reactions or through glutamate dehydrogenase, with the aminated product then interacting with the amr gene product. Nitrogen repression of the yeast purine degradative pathway is fully repressed by certain amino acids as serine and asparagine.

NADP-Glutamate dehydrogenase (GDH) is the first enzyme in ammonium assimilation. A mutation in yeast, gdhA , is a mutation

in the structural gene of GDH. The general amino acid permease in yeast is repressed by high ammonium concentrations in wild-type but

in gdhA strains the mutation relieves ammonium repression (78).

This suggests that nitrogen-catabolite control of amino acid transport

is not mediated directly by ammonium, but by some metabolite derived

from it. This same effect is seen when gdhA~ released arginase, allantoinase, and urea amidolyase from ammonium repression (79,80).

Dubois et al (80), have concluded that GDH (NADP) is a regulatory element for ammonium repression of nitrogen-related enzymes. This hypothesis is also favored by Pateman (81) as a mechanism for nitrogen-

repression in Aspergillus. Pateman found that gdh mutants lacking normal GDH (NADP) activity had enzyme systems which were derepressed

for ammonium-regulation. These enzymes included nitrate reductase,

xanthine dehydrogenase and the permeases for thiourea and glutamine.

The role proposed for GDH .(NADP) is that of a bifunctional protein

having a catalytic activity and a role in ammonium regulation. Pateman 119 (81) suggested that GDH was located in the cell membrane where it could bind extracellular ammonium and that this complex would act to regulate by repression-derepression of nitrogen-related enzymes.

The hypothesis of the role for GDH in nitrogen regulation could be adapted to the amr system. Instead of ammonium binding directly with the amr product, a metabolized form of ammonium would interact, not

GDH itself, but a GDH-catalyzed product. Recent work with nitrate reductase in Neurospora suggests that ammonium must be first metabolized before repression can occur (82). In wild-type, ammonium will partially repress nitrate reductase activity. The extent of repression is a function of ammonium concentration. In a strain lacking GDH activity

(am-la), various concentrations of ammonium had no effect on reductase activity. There are two possible explanations: a) that ammonium must be first metabolized via GDH to repress nitrate reductase or

2) the am-la mutation is lacking some other component which interacts with ammonium to result in repression. The former hypothesis was favored because when glutamate was added to both am-la and wild-type cultures, the synthesis of nitrate reductase was repressed in both cases (in a manner similar to ammonium-repression in wild-type). The question of whether the actual co-repressor is ammonium or a metabolized form is not resolved at this time. Overall, the model for the regulation of nitrogen-related enzymes can provide for the control of a wide number of diverse genes, an essential requirement in a eukaryotic regulatory scheme.

The rational for studying allantoinase was our interest in obtaining a nitrogen-regulated enzyme in pure form which could be

J. used as a tool to further the understanding of the role of the amr 120 gene. We intended to use purified allantoinase to prepare antibodies

and then use these antibodies to obtain allantoinase mRNA. Briefly,

this could be accomplished by the binding of the antibodies to nascent

allantoinase that is still bound to polysomes, so that the whole

complex could be precipitated. Once precipitated, the ribosomes

and protein could be removed leaving the allantoinase mRNA. The

mRNA could be used to make a cDNA probe, and that used to determine

if the amr+gene acts to regulate transcription. Allantoinase, however, is not an ideal enzyme to use for such a project. The results have shown that allantoinase is too unstable in vitro to be useful,

and that a more stable nitrogen-regulated enzyme could better serve

the purpose.

Allantoinase isolated from Neurospora appears to be a tetramer

of 120,000 molecular weight made up of identical or similar sized

subunits with an appromimate molecular weight of 37,000. To date,

this is the only estimate of the size of allantoinase from any

organism. In some instances, related enzymes of a pathway have been

found to be associated together in a multi-enzyme complex. The

results presented here show that neither uricase nor allantoicase,

the enzymes which catalyze the reaction prior to and, subsequent

to allantoinase, respectively, are physically associated with

allantoinase in a stable complex. The results reported here indicate

that allantoinase is not subject to allosteric regulation by the major nitrogen metabolite glutamate as well as by purine catabolite

intermediated uric acid and ammonium.

My results have provided strong evidence for the existence

of essential sulfhydryl groups in allantoinase. Several different classes of sulfhydryl reagents were used and, collectively, they

indicate that allantoinase has reactive sulfhydryl groups with

functional significance. Iodoacetate reacts by alkylating sulf­

hydryl groups (RSH + RI — R-S-R' + HI). This reagent is highly

specific and yields essentially an irreversible reaction (83).

PCMB, an organic mercurial, reacts with sulfhydryl groups readily, but at high concentrations will also react with other protein groups

(84). Arsenate and some heavy metals react with sulfhydryl groups,

as does PCMB, by forming mercaptides. Metals which react include mercury, copper, cadmium, and palladium. The severe inhibition of

allantoinase activity by CuSO^ is largely reversed when allantoinase

is protected by thiol groups, this indicating that some of the Cu"*-*"

inhibition may be due to reaction with sulfhydryl groups. The

inactivation by cysteine and glutathione can be understood if these

reagents are considered as low molecular weight thiols. These thiols

are known to react with a disulfide bond in an exchange reaction to

form mixed disulfides (X-SH+Y-S-S-Y X-S-S-Y+Y-SH). The formation

of these mixed disulfides could alter the structural integrity of the

enzyme and therefore cause a loss of activity (84). The inhibition

of allantoinase by PMS (phenazine methosulphate) is interesting

because PMS is an electron acceptor and carrier. Inhibition can be

accounted for because sulfhydryl groups can readily donate electron

pairs, and thus are one of the most reactive protein groups. PMS

thus can attack exposed sulfhydryl groups (85) . The inhibition by

these various sulfhydryl reagents indicate that one or more sulfhydryl

groups are sufficiently near the active center to interfere with

catalysis, either directly or by structural changes, when they are 122 modified.

Iodoacetate was found to inhibit allantoinase activity in yeast

(73), but not for the bacterial enzymes examined by Vogels (71). A reason for the lack of inhibition could be that iodoacetate was not allowed sufficient time to react. Generally, it requires 30-60 minutes to fully react with a protein's sulfhydryl groups. Allantoinase from yeast, soybeans, and Phaseolus hysterinus were inhibited by cysteine and glutathione (71,73,86). Cysteine was also inhibitory for enzymes isolated from livers of frogs and goldfish (71).

Allantoinase from most sources appears to have reactive sulfhydryl groups which are essential for activity.

The effect of bivalent cations on allantoinase activity is extremely varied depending on the source of enzyme. Yeast allantoinase was only slightly reduced in activity by numerous ions (73). Allan­ toinase from many bacterial sources was found to be inhibited by many heavy metals: Hg+^, Cu+^, Zn+^ , Fe+^, Co+^, and Cd+^ (71).

The metals Zn +2 , Co'1'+9 , and Cd +2 had an inhibitory effect on animal allantoinases but not on that from higher plants (71»73). Inhibition by the above metals (the Neurospora enzyme is inhibited by each of them), could be due to a specific interactions with sulfdydryl groups or to non-specific interactions with the polypeptides. Potential metal ion-binding sites on proteins are principally the ionizable side-chains and the N-terminal and C-terminal groups (87,88). For example, zinc binds to serum albumin by interacting with the imidazole side chains (89). Pancreatic ribonuclease activity is inhibited by both Zn+2 and Cu'-'■? ‘ binding at the imidazole side chains and at a ot-amine site (8 7). A third mechanism of inhibition by metal ions is that they are capable of substituting themselves for a metal ion which is normally present in a metallo-enzyme. If a metal were present in allantoinase, then the use of chelators could give some idea of its identity.

Cyanide, a strong chelator of Ag+^, Fe+^, Zn+^, Cd+^, Hg+^, Cu*^, and Co 4*? , had a marked inhibitory effect on allantoinase activity.

Cyanide inhibition of enzymes is due to its complexing with these metals in metallo-enaymes. Azide which has a high specificity :for iron-containing metals had little significant effect on allantoinase at 1 mM concentration; therefore, iron is probably not associated with the enzyme. Allantoinase was not inhibited by o-phenanthroline +2 which primarily binds to Zn in zinc-containing metallo-enzymes. + 2 Mg does not appear to be essential for allantoinase activity since

NaF, a strong chelator of Mg+2 , did not cause inhibition of enzyme activity. Thiourea complexes with heavy metals, particularly with +2 Cu , and it was found to inhibit allantoinase activity. Phosphate

a_0 i also inhibited allantoinase activity and is known to bind Mg , Ca , and Mn+ 2 . Although it is difficult to interpret these various results obtained with metal ions and chelators, it seems clear that there does exist a metal ion involvement with allantoinase. Although

■f 2 ~\“2 that data is not conclusive, either Cu or Mn may be associated with the enzyme.

My work has shown quite convincingly that Neurospora allantoinase is unstable iri vitro and jLn vivo. Allantoinase Tn vivo decays with a half-life of 15 minutes. In vitro partially purified enzyme (G-150 isolate) upon storage at -5 C overnight will, lose 80-90% activity.

Purified allantoinase (from G-100 column) was found to degrade at 124 both 0 C and 24 C. Allantoinase's instability iri vitro is a problem that has plagued all allantoinase studies. Vogels (71) found puri­ fication of the enzyme from various organisms very difficult and as a result was limited to (NH )„S0, precipitation and DEAE-cellulose 4 2 4 chromatography steps. Allantoinase from goldfish liver was so unstable that all activity was lost during DEAE-chromatography. Yeast allantoinase was similarly affected by this step, resulting in only an apparent 6-fold purification. The enzyme from Candida utilis under the same conditions lost all activity during purification (73).

Allantoinase activity from Neurospora was completely lost when DEAE- cellulose chromatography was attempted as a purification step. This required the use of successive Sephadex gel-filtration steps. As I have already stated, there is a small amount of residual protease activity associated with allantoinase in the highly purified G-100 preparation. This level of protease activity is so very low that it is difficult to attribute in vitro turnover solely to it. The co­ purification of a protease is not unusual with enzymes from Neurospora; tryptophan synthase in a partially pure state was inactivated by a

PMSF-sensitive factor (90). This inactivation was due to a serine endopeptidase which had been co-purified with the synthase (91).

Allantoinase in vitro is very sensitive to exogenously added proteases and, therefore, an endogenous protease could account for the extreme instability of allantoinase. Purified allantoinase was rapidly inactivated upon adding a cell-free extract of aln~ by an

EDTA-sensitive inactivating factor. This agent has the same basic characteristics as the alkaline protease found in Neurospora (92). 125 Inactivation of allantoinase in vivo has also been studied in

Pseudomonas aeruginosa (93). Allantoinase activity decreased in

stationary phase when the cells were grown in medium containing allantoin or allantoic acid as the sole carbon, nitrogen, and energy source. The addition of chloramphenicol, in inhibitor of protein synthesis, to cell-free extracts prevented the decrease of allantoinase activity. When chloramphenicol was added at various times during the growth phase, there was an almost immediate cessation of the decrease of allantoinase activity. It appears that turnover of allantoinase is due to the synthesis of a protein which has no proteolytic activity toward allantoinase. This protein is thought to be a specific binding protein for allantoinase which inactivates enzyme activity.

A different type of turnover system controls nitrate reductase in Neurospora. Nitrate reductase is rapidly inactivated in the presence of ammonium or the removal of nitrate (94). Cycloheximide affords partial protection from inactivation. Inactivation requires that nitrate reductase be in its active form, a complex containing both of its subunits. This may be because the inactivating factor, a protease for example, can only recognize a complete structure (the native configuration). Nitrate is not only required for induction of nitrate reductase synthesis but it also protects the enzyme from degradation (95).

Is allantoinase in Neurospora being degraded by a protease in vivo? It is not clear at this time. Allantoinase does decay very rapidly in vivo, but unlike Pseudomonas, the addition of an inhibitor of protein synthesis does not stabilize allantoinase activity. The purification of an allantoinase-inactivating factor is one possible approach to this problem. Caution must be exhibited, however, since the decay-inducing agent would be tested iai vitro against an enzyme which my work has shown to be very labile even after purification. I have isolated a protease activity which is

EDTA-insensitive and PMSF-sensitive, but was unable to isolate a fraction which resulted in significant allantoinase degradation.

The mechanism of allantoinase turnover is an interesting and challenging problem. A protease is the likely agent for this turnover, but inherent instability of the enzyme cannot be totally discounted.

However, because of the very short half-life of allantoinase, a specific mechanism is probably involved. A protein which would physically associate with allantoinase jin vivo could modify allan­ toinase so that a general protease could recognize it and degrade it.

Thus, an allantoinase-specific protease is not required, but one which could readily recognize specific functional groups or conformations.

This hypothesis could be applied as a general mechanism to regulate the levels of intracellular enzymes. The initiation of enzyme degradation could result from conformational changes in the enzymes, causing conversion from a nonsusceptible to a susceptible protease substrate. This could be by interactions with ligands (substrate or co-enzymes) or modifications like phosphorylation or adenylation.

Once it is in a susceptible form, group-specific proteases could act to degrade an enzyme by limited proteolysis. The remaining large fragments could then be completely degraded by other general proteases

(90,96). The ability of the cell to control the level of enzymes by dynamic turnover adds another dimension of regulation to the better understood controls exerted over enzyme synthesis and activity. BIBLIOGRAPHY

1. Reed, J. and Wintersberger, E. (.1973)

2. Davis, R. H. and DeSerres, F. J. (1970) in "Methods in Enzymology, 17 A," (Tabor, H. and Tabor, G. W., eds.) pp. 79, Academic Press, N. Y.

3. Williams, L. G., Bernhardt, S. A. and Davis, R. H. (1971) J. Biol. Chem. 246:973.

4. Davis, R. H. (1972) Science 178:835.

5. Davis, R. H. (1974) Stadler Sympos. 6:61.

6. Williams, L. G., Bernhardt, S. A. and Davis, R. H. (1970) Biochem. 9:4329.

7. Davis, R. H. (1967) in "Organizational Biosynthesis", (Vogels, H. J., Lampen, J. 0. and Bryson, V., eds.) pp. 303, Academic Press, N. Y.

8. Williams, L. G. and Davis, R. H. (1970) J. Bacteriol. 103:335.

9. Davis, R. H. (1962) Genetics 47:351.

10. Davis, R. H. (1965) Biochim. Biophys. Acta. 107:54.

11. Lue, P. F. and Kaplan, J. G. (1970) Biochim. Biophys. Acta. 220:365.

12. Weiss, R. L. and Davis, R. .11. (1973) J. Biol. Chem. 248:5403.

13. Bernhardt, S. A. and Davis, R. H. (1972) Proc. Nat. Acad. Sci. U. S. 69:1868.

14. Subramanian, K. N., Weiss, R. L. and Davis, R. H. (1973) J. Bacteriol. 115:284

15. Weiss, R. L. (1973) J. Biol. Chem. 248:5409.

16. Rhines, H. W. (1968) Unpublished PhD thesis, Yale University, New Haven, Conn.

17. Chaleff, R. S. (1974) J. Gen. Microbiol. 81:337.

18. Chaleff, R. S. (1974) J. Gen. Microbiol. 81:357.

.127 128 19. Partridge, C. W. H., Case, M. F. and Giles, N. H. (1972) Genetics 72:411.

20. Valone, J. A., Case, M. E. and Giles, N. H. (1971) Proc. Nat. Acad. Sci. U. S. 68:1555.

21. Rines, H. W. (1968) Genetics 60s:215.

22. Case, M. E. and Giles, N. H. (1975) Proc. Nat. Acad. Sci. U. S. 72:553.

23. Giles, N. H., Case, M. E., Partridge, C. W. H. and Ahmed, S. I. (1967) Proc. Nat. Acad. Sci. U. S. 58:1453.

24. Rhines, H, W., Case, M. E. and Giles, N. H. (1969) Genetics 61:789.

25. Case, M. E. and Giles, N. H. (1968) Genetics 60:49.

26. Case, M. E. and Giles, N. H. (1971) Proc. Nat. Acad. Sci. U. S. 68:58.

27. Metzenberg, R. L. (1972) Ann. Rev. Genetics 6:111.

28. Metzenberg, R. L. and Parson, J. W. (1966) Proc. Nat. Acad. Sci. U. S. 55:629.

29. Marzluf, G. A. (1972) Biochem. Genet. 7:219.

30. Marzluf. G. A. and Metzenberg, R. L. (1968) J. Mol. Biol. 33:423.

31. Marzluf, G. A. (1972) Arch. Biochim. Biophys. 150:714.

32. Hanson, M. A. and Marzluf, G. A. (1973) J. Bacteriol. 116:785.

33. Metzenberg, R. L. and Ahlgran, S. K. (1971) Genetics 68:369.

34. Burton, E. G. and Metzenberg, R. L. (1972) J. Bacteriol. 109:140.

35. Dietrich, P. S. and Metzenberg, R. L. (1973) Biochem. Genet. 8:73.

36. Marzluf, G. A. (1976) Manuscript in preparation.

37. Arst, H. N. and Cove, D. J. (1969) J. Bacteriol. 98:1284.

38. Dubois, E., Gransen, M. and Wiame, J. M. (1973) Biochim. Biophys. Res. Commun. 50:967.

39. Garrett, R. H. (1972) Biochim. Biophys. Acta. 264:481.

40. Arst, H. N. and Cove, D. J. (1973) Molec. Gen. Genet. 126:111.

41. Cohen, B. L. (1972) J. Gen. Microbiol. 71:293.

42. Hynes, M. J. and Pateman, J. A. (1970) Molcc. Gen. Genet. 108:97. 129 43. Hynes, M. J. (1972) J. Bacteriol. 111:717.

44. Arst, H. N. and Scazzocchio, C. (1975) Nature 254:31.

45. Trijbels, F. and Vogels, G. D. (1966) Biochim. Piophys. Acta. 113:292.

46. Trijbels, F. and Vogels, G. D. (1966) Biochim. Biophys. Acta.. 118:387.

47. Vogels, G. D., Trijbels, F. and Uffink, A. (1966) Biochem. Biophys. Acta. 122:482.

48. Cooper, T. G. and Lawther, R. P. (1973) Proc. Nat. Acad. Sci. U. S. 70:2340.

49. Lawther, R. P., Riemer, E., Chojnacki, B. and Cooper, T. G. (1974) J. Bacteriol. 119:461.

50. Whitney, P. A., Cooper, T. G. and Magasanik, B. (1973) J. Biol. Chem. 248:6203.

51. Cooper, T. G. and Lawther, R. P. (1973) Biochim. Biophys. Res. Commun. 52:137.

52. Lawther, R. P. and Cooper, T. G. (1973) Biochim. Biophys. Res. Commun. 55:1100.

53. Lawther, R. P., Phillips, S. L. and Cooper, T. G. (1975) Molec. Gen. Genet. 137:89.

54. Lawther, R. P. and Cooper, T. G. (1975) J. Bacteriol. 121:1064.

55. Bossinger, J., Lawther, R. P. and Cooper, T. G. (1974) J. Bacteriol. 118:821.

56. Darlington, A. J., Scazzocchio, C. and Pateman, J. A. (1965) Nature 206:599.

57. Scazzocchio, D. and Darlington, A. J. (1968) Biochim. Biophys. Acta. 166:557.

58. Scazzocchio, C. (1973) Molec. Gen. Genet. 125:147.

59. Reinert, W. R. and Marzluf, G. A. (1975) Arch. Biochim. Biophys. 166:565.

60. Reinart, W. R. and Marzluf, G. A. (1975) Molec. Gen. Genet. 139:39.

61. Chinard, F. P. (.1.952) J. Biol. Chem. 199:91

62. Davis, R. II. and Mora, J. (1968) J. Bacteriol.. 96:383.

63. Strecker, II. J. (1971) in "Methods .in Enzymology, 17B, (Tabor, H. and Tabor, C. W., eds.) pp. 262, Academic Press, N. Y. 130 64. Strecker, H. J. (1971) in "Methods in Enzymology, 17B", (Tabor, H. and Tabor, C. W., eds.) pp. 254, Academic Press, N. Y.

65. Arst, H. N. and MacDonald, D. W. (1975) Nature 254:26.

66. Vogels, G. D. and Van der Drift (1970) Anal. Biochem. 33:143.

67. Reinert, W. R. and Marzluf, G. A. (1975) Arch Biochim. Biophys. 166:565.

68. Lowry, 0. H., Rosenbrough, N. J., Farr, A. and Randell, J. R. (1951) J. Biol. Chem. 193:265.

69. Chrambach, A., Reisfeld, R. A., Wycokoff, M. and Zaccari, J. (1967) Anal. Biochem. 20:150.

70. Reinert, W. R. (1975) Unpublished PhD thesis, Ohio State University, Columbus, Ohio.

71. Vogels, G. D, and Van der Drift (1966) Biochim. Biophys. Acta. 122:497.

72. Nagai, K. and Funahashi, S. (1961) Agr. Biol. Chem. Tokyo 25:265.

73. Lee, K. W. and Roush, A. H. (1964) Arch. Biochim. Biophys. 108:460.

74. Scarborough, G. A. (1973) Inter. Rev. Cytology 34:103.

75. Rao, E. Y., Rao, T. K. and Debusk, A. G. (1975) Biochim. Biophys. acta. 413:45.

76. Railey, R. M. and Kinsey, J. A. (1976) Can. J. Microbiol. 22:115.

77. Marzluf, G. A. (1976) Personal Communication.

78. Grenson, M. and Hou, C. (1972) Biochim. Biophys. Res. Commun. 48:749.

79. Dubois, E., Grenson, M. and Wiame, J. M. (1973) Biochim. Biophys. Res. Commun. 50:967.

80. Dubois, E., Gresnen, M. and Wiame, J. M. (1974) Eur. J. Biochem. 48:603.

81. Pateman, J. A., Kinghorn, J. R., Dunn, E. and Forbes, E. (1973) J. Bacteriol. 114:943.

82. Chang, H. C. P. and Sorger, C-. J. (1976) J. Bacteriol. 126:1002.

83. Webb, J. L. (1966) in "Enzyme and Metabolic Inhibitors, Vol. Ill", pp. 1025, Academic Press, N. Y. 131 84. Webb, J. L. (1966) in "Enzyme and Metabolic Inhibitors, Vol. II", pp. 1235, Academic Press; N. Y.

85. Webb, J. L. (1968) in "Enzyme and Metabolic Inhibitors, Vol. IV", pp. 1143, Academic Press, N. Y.

86. Van der Drift, C. and Vogels, G. D. (1966) Acta Botan. Neerl. 15:209.

87. Eichlan, G. L. ed. (1973) in "Metal-Protein Complexes in Inorganic Biochemistry", pp. 227. Elsevier Scientific Publ. Comp., N. Y.

88. Sigel, II. (1973) "Metal Ions in Biological Systems, Vol. 2", pp. 219, Marcel Dekker, Inc. N. Y.

89. Gurd, F. R. N. and Goodman, D. S. (1952) J. Am. Chem. Soc. 74:670.

90. Tsai, H. and Suskind, S. R. (1972.) Biochim. Biophys. Acta. 284:324.

91. Yu, P. H., Kula, M. R. and Tsaik H. (1973) Eur. J. Biochem. 32:129.

92. Siepen, D., Yu, P. H. and Kual, M. R. (1975) Eur. J. Biochem. 56:271.

93. Rujnierse, V. F. M. and Van der Drift, C. (1974) Arch. Microbiol. 96:319.

94. Subramanian, K. H. and Sorger, G. J. (1972) J. Bacteriol. 110:538.

95. Sorger, G. J., Debanne, M. T. and Davies, J. (1974) Biochem. J. 140:395.

96. Katunanma, N. (1975) Rev. Physiol. Biochem. Pharmacol. 72:83.