THESIS

d"
. .o'
E. ~34
I t .
g ’ g
‘ ~ ‘ ,_ I
- ~ u~_ x": f;,".,‘_ t’)\~ v‘f k
{$0.}ch ‘ .‘xxo' ...., - - t...
. .. - y
. .~ \‘. ’7'
m - -
‘1 5 .-.; wag-m

This is to certify that the

dissertation entitled

BRANCHED-CHAIN AMINO ACID METABOLISM
IN CLOSTRIDIUM SPOROGENES

presented by

Daniel J. Monticello

has been accepted towards fulfillment
of the requirements for

degree in Microbiology

Ph.D.

 

 

Datefl/Ai/fz

0-12771

M5 U is an Affirmative Action/Equal Opportunity Institution

 

 

 
  

,1”

V1"
'\

 

  
  

'1“le s

t r
~ -n“.”’
‘\ III’

 

OVERDUE FINES:
25¢ per du per 1ten

RETURNING LIBRARY MATERIALS:

Place in book return to remove
charge from circulation records

 

BRANCHED-CHAIN AMINO ACID METABOLISM
IN CLOSTRIDIUM SPOROGENES

 

BY

Daniel Joseph Monticello

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Microbiology and Public Health

1982

C..j /// [I —..

ABSTRACT

BRANCHED-CHAIN AMINO ACID METABOLISM
IN CLOSTRIDIUM SPOROGENES.

 

BY

Daniel Joseph Monticello

Preliminary studies indicated that Q. sporogenes
(ATCC 7955, NCA PA 3679) made isoleucine in a novel pathway
not involving threonine or threonine dehydratase. Data herein
show that radioactivity from [3-14C1pyruvate, [14C]propionate

14C] butyrate was incorporated into isoleu-

and a-amino-[B-
cine. Increasing concentrations of a-methylbutyrate re-
duced the incorporation of [14C]pyruvate. Crude extracts of
the bacteria grown in a minimal medium were found to contain
levels of a-acetohydroxyacid synthase activities comparable
to those found in cells of E. ggli K12 grown in minimal medium.
Stepwise degradation of isoleucine labeled with specifically-
labeled precursors supports the conclusion that g, sporogenes
makes isoleucine via the reductive carboxylation of pro—
pionate to yield a-ketobutyrate which is metabolized to
isoleucine in the classical fashion, or isoleucine can be
made via the reductive carboxylation of a-methylbutyrate to
a-keto-B-methylvalerate.

Clostridium sporogenes requires L-leucine and L-valine

for growth in a minimal medium, although valine can be re-

placed by isobutyrate, and leucine by isovalerate. Cells

l4

grown in minimal media incorporated significant C-label

14

from [14C]va1ine into leucine and C-label from [14C]leucine

into valine. These results indicate that these bacteria can

interconvert leucine and valine.

14 l

Considerable label from [ C]valine, [ 4C]leucine and

[14C]isoleucine was incorporated into cellular lipid after

growth of Q. sporogenes in a minimal medium, with six times

 

more radioactivity from labeled valine than leucine or
isoleucine being incorporated. It appears that branched-
chain amino acids were an important source of the branched-
chain fatty acids in the cellular lipid. The branched-chain
amino acids were shown to be subject to non-specific oxida-
tion to volatile fatty acids in an assay with whole cells,
and isoleucine oxidative activity appeared to be inducible

to some extent.

DEDICATION

This dissertation is dedicated to my parents, William
and Kathleen, my siblings Tom, John, Anne Marie, Mike, Jim,
George and Bob, and especially my partner, Elizabeth.

Their support, encouragement and diversionary tactics made

the difficult times bearable, and the good times wonderful.

ii

ACKNOWLEDGMENTS

I wish to express my sincere gratitude to my major
professor, Dr. R. N. Costilow, for his guidance throughout
the course of this investigation. I would also like to
express my thanks to Drs. H.L. Sadoff and R.R. Brubaker, for
the use of their laboratories and equipment, and for their

essential support and encouragement.

iii

TABLE OF

LIST OF TABLES . . . . . .
LIST OF FIGURES. . . . . .

ORGANIZATION . . . . ...

SECTION I

INTRODUCTION . . . .

LITERATURE REVIEW. .

The Clostridia

Costridium sporogenes.

 

CONTENTS

Branched-chain Amino Acid Synthesis.

Reductive Carboxylations .

Literature Cited .

SECTION II

Isoleucine synthesis from a-methylbutyrate
propionate by Clostridium sporogenes .

 

SECTION III

Interconversion of valine and leucine

by Clostridium sporogenes.

 

iv

Page

vi

. .viii

or

17

20

26

60

Page
SECTION IV
APPENDICES:

Incorporation of branched chain amino
acids into cellular lipid. . . . . . . 74

Oxidation of branched-chain amino
acids in vitro . . . . . . . . . . . . 77

Effect of a-methylbutryate on total
growth . . . . . . . . . . . . . . . . 81

SECTION V

SUMMARY. . . . . . . . . . . . . . . . . . . . 84

LIST OF TABLES

TABLE PAGE

LITERATURE REVIEW

 

1. Branch-point enzymes and their regulation
in Escherichia coli . . . . . . . . . . . . . n 12
SECTION II
1. Incorporation of 14C-propionate into

a-aminobutyrate . . . . . . . . . . . . . . . 41

2. Effect of propionate on the incorporation of

_14

[U C] serine into isoleucine. . . . . . . . 42

3. Degradation of isoleucine from g, sporogenes

 

grown in [3-14C1pyruvate, [1-14C1propionate

and [2-14

C] propionate. . . . . . . . . . . . 45
SECTION III
1. Maximum growth in basal medium supplemented
with amino acids and volatile fatty acids . . 64
2. Maximum growth in basal medium supplemented
with volatile fatty acids . . . . . . . . ... 65

14

3. Incorporation of C into aspartate, glutamate,

leucine and valine. . . . . . . . . . . . . . 67

vi

LIST OF TABLES (cont'd.)

Page
SECTION IV
1. The incorporation of 14C-label into cellular
lipid and protein . . . . . . . . . . . . . . 76

vii

LIST OF FIGURES

FIGURE PAGE

LITERATURE REVIEW

1. Biosynthesis of branched-chain amino acids. . . 11
2. Known routes to isoleucine, via a-ketobutyrate. 15
SECTION II

1. Effect of a-methylbutyrate on the incorporation

of [U-14C1pyruvate and 14CO2 into isoleucine. 36
2. Incorporation of a-amino-[3-14C] butyrate

into amino acids. . . . . . . . . . . . . . . 37
3. Incorporation of [2-14C1propionate into

amino acids . . . . . . . . . . . . . . . . . 39
4. A scheme for the degradation of isoleucine. . . 46

5. Proposed pathway for isoleucine synthesis

in Clostridium sporogenes . . . . . . . . . . 51

 

SECTION III
1. Proposed pathway for the conversion of

leucine to valine . . . . . . . . . . . . . . 7O

viii

1.

2.

3.

LIST OF FIGURES (cont'd.)

SECTION IV
The oxidation of L-isoleucine, L-valine and
L-leucine to a-methylbutyrate, isobutyrate
and isovalerate . . . . . . . . . . . . . .
The effect of a-methylbutyrate on total
growth. . . . . . . . . . . . . . . . . . .
Separation of a-methylbutyrate and isovalerate

by gas chromatography . . . . . . . . . . .

ix

Page

80

82

83

ORGAN I ZAT ION

This dissertation is organized into five sections.
Section I consists of an introduction and review of the re-
cent literature pertinent to the research discussed.
Sections II and III consist of manuscripts for a paper and
a note for publication. Section IV consists of results
obtained which are related to the research but not included
in manuscripts. A summary of all the research is presented

in Section V.

SECTION I

INTRODUCTION

The proteolytic Clostridia are a diverse group of
saprophytic organisms and commensal inhabitants of the in-
testinal tract of animals. Clostridium sporogenes is often
found in the soil, but is also encountered in contaminated
food, and occasionally in necrotic tissue where g. tetani

or C. perfringens have established an infections. 9. sporo-

 

genes is physiologically indistinguishable from Q. botulinum
type A, a causative agent of botulism poisoning. These

attributes have made 9. sporogenes an important model for

 

investigations of anaerobic physiology and energy metabolism.

It was in _C_. sporogenes that Stickland first demon—

 

strated the coupled oxidation and reduction of paired amino
acids which is the principle source of ATP in many Clostridia
and other anaerobes (24, 45). Much of our understanding of
the unique strategies for amino acid catabolism employed by
anaerobes comes from investigations with this species.

Studies of biosynthetic pathways in the anaerobes are
infrequent, largely due to the difficulties which arise when
using the complex media required by most anaerobes for

growth. The development in our laboratory of a simple de-

 

fined medium for g. sporogenes (R.S. Hadioetomo, 1980,
Ph.D. dissertation, Michigan State University) has allowed

us to investigate some biosynthetic activities of this

1

organism.

Hadioetomo found that while 9, sporogenes requires
leucine and valine for growth, it does not require isoleu-
cine. This was a surprising observation since in most mi-
crobes these amino acids are produced via a common pathway
(45). The threonine dehydratase (the first enzyme of iso-
1eucine biosynthesis in Escherichia coli) activity in Q.

sporogenes was found to be catabolic in nature, and was not

 

regulated by isoleucine in vitro. Cells grown in a minimal

14 14

medium incorporated label from C]serine and

14

coz, [
[3- C]pyruvate into isoleucine, but not from [14C1threonine.
In addition, radioactive carbon from a-[14C1methylbutyrate
was incorporated into isoleucine, presumably by reductive
carboxylation to a-ketoB-methylvalerate and subsequent trans-
amination. These observations led Hadioetomo to propose that
isoleucine was produced via a novel pathway in g. sporogenes.
The object of the experiments described in this dis-

sertation was to elucidate branched-chain amino acid meta-

bolism in Q. sporogenes. The principal goal was to determine

 

the nature of the unusual pathway for isoleucine synthesis.
The results indicate that one pathway to isoleucine origi-
nates with propionate, which is reductively carboxylated to
a-ketobutyrate. The condensation of a-ketobutyrate and
pyruvate follows (catalyzed by acetohydroxyacid synthase),
and isoleucine is then produced in the classical fashion.
a-Methylbutyrate appears to be an alternate precursor of

isoleucine. In addition, 9. sporogenes can interconvert

 

leucine and valine when grown in appropriate media.

The picture of branched-chain amino acid metabolism
which emerges is considerably more complex than that seen
in E. 921;. In addition to the drain on leucine, valine
and isoleucine resulting from protein synthesis, 9, sporo-
SEEEE must also contend with continuous non-specific oxi-
dation of these amino acids for ATP generation. a-Ketoiso-
valerate, the keto-acid analog of valine, is also the source
of pantoate in the cell. All three branched-chain amino
acids are liberally incorporated into cellular lipid as

well. To some extent 9. sporogenes takes advantage of its

 

reduced environment to meet these special demands. The
various reductive carboxylation reactions utilized are novel
and important Options for the supply and regulation of iso—

leucine, leucine and valine in these bacteria.

LITERATURE REVIEW

The Clostridia

The genus Clostridium encompasses a diverse group of

 

anaerobic, spore-forming, rod-shaped bacteria (16, 21, 38).
The Clostridia are widely distributed in nature, and their
natural habitats include soil, fresh and salt water, and

the intestinal tracts of animals. Although usually consider-
ed gram positive, their reaction to this staining procedure
is frequently variable. The bacteria are motile, relatively
large rods which frequently are swollen when endospores

form (hence, from the Greek Clostridium; small spindle)

 

although filamentous forms are sometimes observed. Spores
are located centrally, terminally or subterminally while
deve10ping in the cell. A few species (including 9. pg;-
fringens) have been observed to produce capsules. The bac-
teria possess flav0protein enzymes which will readily re-
duce molecular oxygen (02) to hydrogen peroxide (H202) and
to superoxide (02-), but they lack catalase, peroxidase and
superoxide dismutase activities which would protect them
from these toxic products. Despite this inadequacy, differ-
ent species show markedly different tolerances to oxygen;

while most are strict anaerobes, C. perfringens can tolerate

 

 

low 02 levels, and g. histolyticum grows slowly on blood

4

agar plates exposed to the air. Clostridial spores are
notoriously resistant to heat, disinfectants, and lengthy
exposure to air. Metabolically, the Clostridia can be di-
vided into two general physiological groups; the predomi-
nantly proteolytic or, the predominantly saccharolytic
Clostridia. Proteolytic Clostridial fermentations usually
result in the formation of volatile acids, ammonia, CO2 and

H The ptomaines, putrescine and cadaverine, are produced

2.
by decarboxylation of ornithine and lysine, respectively.
The Clostridia are generally thought of as saprophytic
soil organisms and commensal inhabitants of the intestinal
tracts of animals. However, several species are important

pathogens, and are responsible for three major diseases in

humans; botulism lg. botulinum types A, B, and E), tetanus

 

(Q. tetani), and gas gangrene (predominantly g. prefringens).

 

None of these diseases are caused by the invasion of bac-
teria into healthy tissues. Botulism is a classic intoxi-
cation resulting from the ingestion of food containing pre—
formed toxin. Tetanus and gas gangrene are usually the re-
sult of local Clostridial growth in traumatized tissue, from
which the soluble toxins are disseminated and a general

toxemia results. Clostridium pasteurianum is a soil organ—

 

ism capable of fixing molecular nitrogen (N2). Many other
members of this genus manifest unusual and sometimes useful
metabolic activities, and have been used in industrial fer-
mentations for many years. The investigation of Clostri-
dial intermediary metabolism has revealed several novel path-

ways and activities, unique to these anaerobic bacteria.

Clostridium sporogenes

 

The most striking and significant feature of Clostri—

dium sporogenes is its marked resemblence to Q. botulinum

 

type A. The sole reliable distinguishing property between
these species is the elaboration of botulinum toxin by g,

botulinum type A. Both species produce very resistant,

 

oval, subterminal spores. They are proteolytic, and pro-
duce large amounts of butyric acid, and smaller amounts of
acetic, isobutyric, isovaleric and isocaproic acids and
propyl, isobutyl and isoamyl alcohols during fermentations.

Q. sporogenes and Q. botulinum type A were found to ferment

 

 

only fructose and glucose among 28 carbohydrates tested (38).
The G + C content of both species is 26-28 moles %, and sev-
eral strains show 100% DNA hom010gy in DNAzDNA hybridization
experiments (28). A pyrolysis gas-liquid chromatography

study of g. botulinum and related organisms readily dis-

 

tinguished botulinum types A and B from types C and D, type
E, and nonproteolytic types B and F (23), but failed to

differentiate type A cultures from Q. sporogenes. Recent

 

investigations of the endproducts of aromatic amino acid
fermentations (20), and branched-chain amino acids (19) show
that the two species ferment these compounds in the same
fashion. In a study of the quantitative utilization of
amino acids by representative strains of the Clostridia,

g. sporogenes and Q. botulinum type A were shown to utilize

 

 

the amino acids to a similar extent (29).

6

The neurotoxin elaborated. by Q. botulinum type A is

 

the most potent biological toxin known; 1 ug is sufficient
to kill the average human. Not surprisingly, most investi-
gators pursue their studies in the metabolically indistin—

guishable Q. sporogenes, and subsequently confirm results in

 

g. botulinum type A.

 

The precise nature of the relationship of Q. sporogenes

to Q. botulinum is not clear. Ekland and Poysky (17) have

 

demonstrated that when 9. botulinum types C and D were cured

 

of their prOphages with mitomycin C, they concommitantly
ceased to produce their dominant toxins. These cured, non-
toxigenic strains of types C and D could then be converted
to either toxigenic type C or D depending on the specific
phage used. A similar interconversion of type C to g.
ngyyi type A has been reported (18). Inducible bacteri-

ophages have only recently been isolated from Q. botulinum

 

type A (43), but the relationship between toxigenicity and
lysogeny, or phage conversion to toxigenicity has yet to be

demonstrated. It is quite possible that g. sporogenes is

 

an atoxigenic strain of Q. botulinum types A by virtue of

 

the loss of a temperate bacteriOphage.

The intimate relationship of Q, sporogenes to Q,

 

botulinum type A has led to considerable interest in the

 

physiological characteristics of this species. Much of our
current understanding of intermediary metabolism in anaerobes

comes from studies with Clostridium sporogenes. The degra-

 

dation of amino acids by anaerobic bacteria has recently

been reviewed (4). These investigations have revealed

several unique and unusual metabolic activities and path-
ways, found only in the anaerobes.

The important observation that many Clostridia, while
unable to grow with a single amino acid as a fermentable
energy source would readily grow if provided with appro-
priate pairs of amino acids, was first made with Q. sporogenes
by Stickland in 1934 (4). The coupled oxidation-reduction
of amino acid pairs (the Stickland reaction) has since been

observed in many Clostridia. g. sporogenes has an absolute

 

requirement for selenium when using glycine as the oxidant

in a Stickland reaction system (14). The bacteria oxidize
the branched-chain amino acids valine, leucine, and isoleu-
cine to isobutyrate, isovalerate and methylbutyrate in this
fashion, although at least one strain converts leucine to
isocaproate and isobutyrate utilizing a novel enzyme, leucine
2,3-aminomutase, which converts a-leucine to B-leucine (34).
An investigation of ornithine metabolism in g. sporogenes

has led to the elucidation of several novel enzymes, notably
ornithine cyclase (deaminating) (31), ornithine 5,4—amino-
mutase (40), 2,4-diaminopentanoic acid C4 dehydrogenase (39),
and a reversible proline dehydrogenase which is regulated by
glutamate (15, 30). The fermentation of the aromatic amino

acids by g. sporogenes also proceeds in a novel fashion,

 

and several unique enzyme activities have been identified
including a 2-enoate reductase enzyme, which catalyzes the
interconversion of E*-cinnamate to phenylpropionate in an

NADH-dependent reaction (4, 11).

*The latest nomenclature for cis-trans isomers. E=entogen =
trans; Z = zusammen = cis (4).

 

Investigations of amino acid metabolism in the clos-
tridia have been hampered by the fact that these bacteria
require a complex and ill-defined medium for good growth.
Mead (29) was able to develop a reasonably well-defined
medium for good growth. Mead (29) was able to develop a
reasonably well-defined medium to study amino acid utiliza-
tion, but a casein- hydrolysate was always included. Such
a medium is not suitable for investigations of amino acid
synthesis. Perkins and Tsuji (32) deve10ped a synthetic

medium in which E. sporogenes could grow, although it could

 

not sporulate well. The historical development of minimal
media for the Clostridia has recently been reviewed (R.S.
Hadioetomo, Ph.D. dissertation, Michigan State University,
1980); in this investigation, a minimal medium for Q. sporo-
EEEEE was deve10ped which includes only eight amino acids,

salts, and vitamins.

Branched-chain Amino Acid Synthesis

The reactions involved in the biosynthesis of valine,
leucine, and isoleucine were first elucidated over a quarter
of a century ago by Strassman, Weinhouse, and co-workers (41,
42). In subsequent years, these reactions have been ob-
served in every biological system in which branched-chain
amino acids are synthesized from smaller (3-4 carbon) pre-
cursors (45). The most significant feature of these bio-
synthetic reactions is that they have been shown to be

catalyzed by a set of "shared" enzymes which operate both in

10

the synthesis of leucine and valine from pyruvate, and the
synthesis of isoleucine from pyruvate and a-ketobutyrate
(Fig. l). The continued extensive investigation of this sys-
tem in Escherichia coli, Salmonella typhimurium, Neurospora

crassa, Saccharomyces cerevisiae and other species has been

 

aimed at elucidating the unusual and complex regulatory pheno-
mena associated with this multi-product system. Table 1
illustrates these regulatory relationships in wild-type E.
SQli' These dynamic interactions of several overlapping re-
gulatory systems are viewed as a general model for the con-
trol of pathways with shared precursors, intermediates and
enzymes.

Control of carbon flow over the isoleucine biosynthe-
tic pathway is usually achieved at the first step in the
pathway (Fig. 1). Threonine dehydratase catalyzes this con-
version of L-threonine to a-ketobutyrate, ammonia and water.
The enzyme and its regulation have been recently reviewed
(R.S. Hadioetomo, 1980.Ph.D. dissertation, Michigan State
University). Two distinctly different forms of the enzyme
have been reported. A "biosynthetic" activity has been ob-
served and purified from several sources (44, 45). Iso-
leucine is a negative effector and valine a positive effector
that antagonizes isoleucine inhibition of this activity.
Synthesis of threonine dehydratase is subject to multiva-
1ent repression in the presence of excess isoleucine, valine,

and leucine (25, 45) in E. coli and E. typhimurium. A

 

"catabolic" threonine dehydratase, which is not allosteri—

cally modified by branched-chain amino acids and is

11

.mcflom ocHEm camcolwmnocmum mo memonucmeHm .H .mHm

$312.6 mpfisfitoxaom— new
mm<z§<mz<£ "mm
35.2.3 Eofixozezs «J

NE mi use “25335533 «mm
$.83 N5 TI... EMS 33:: 35.3323 98¢ >xo~a>zo$u< "Nu
www.ma mafia $55913 “.5535-.. "J
a
a
a? 4 ca:
m1: M12 {Diva‘s
22> 38.3 Tl... 53 TW. :5
=Wb$ " :YND "
ma . ma 0
.
m. J
. .
. I
23302 88. .IA. 38.9 VI :59
Pasha m2 Pic

ma ma

 

12

 

AIV mnH

An. qu .Aus 9mg

A+V q<> .Au. 2mg

As. one

An. n¢>

(‘0

Au. q¢>

Au. qu

HHH

HH

Dmn oDMHoEH>m0Hm0mH

mmmnucmm
A¢> .qu .qu meow axoupmsouoofi
qu ommumuomnoc

msacomunuun

 

mHOUOmM Mm mflogmm

muouoommm anofiwum

nonpoum mthncm
nocmum

 

 

.flHoo ownofluozomm ca coauoasmmn

“Hosu can dehusm ucflomlnosmum .H wanna

13

stimulated by ADP, has been observed in anaerobically-grown

E. coli and other species including Clostridium tetanomorphum

 

(46) and E. sporogenes (R. S. Hadioetomo, 1980 Ph.D. disser-

 

tation, Michigan State University).

The first "shared" enzyme activity in isoleucine syn-
thesis is a a-acetohydroxy acid synthase (AHAS), which cata-
lyzes the condensation of two pyruvate molecules to form
a-acetolactate or pyruvate and a-ketobutyrate to form a-aceto

a-hydroxybutyrate. E. coli and E. typhimurium possess three

 

isoenzymes to catalyze this reaction, but their properties,
and roles in the regulation of carbon flow through the path-
way, are not well understood. Considerable controversy has
arisen surrounding the nature of several valine-resistant
mutants of E. 99;; K-12. This strain has the genetic po-
tential to express all three AHAS isoenzymes (45) [ilXE
(AHAS I), 2125 (AHAS II), and llXHI (AHAS 111)], but is un-
able to grow in minimal medium supplemented with valine.
Since AHAS I and III are inhibited by valine, while AHAS II
is not (Table 1), and mutants of E. 22;; K-12 without AHAS I
and III require isoleucine, it has been proposed that the
EEXC gene is cryptic in E. 22;; K-12 (22). Two markedly
different explanations for the reversion of this strain to
valine resistance have been prOposed. Lawther 22.2l- (27)
have demonstrated that K-12 contains a frameshift mutation
within ilXG and that valine resistant mutants have displaced
this frameshift resulting, they say, in the expression of

ilvG and valine resistance. However, Berg EE.El-r contend

14

that a truncated product of the ilXG gene is in fact made
and that this protein may function in valine resistant
mutants, perhaps as part of a slightly altered AHAS iso-
enzyme complex (6).

Carbon flow to leucine is regulated at the first en-
zyme in this branch of the pathway, a—isopropylmatate syn-
thase. The enzyme catalyzes the initial step in the chain-
elongation mechanism, the reaction of a-keto-isovalerate
and acetyl-CoA to form a-isopropylmalate. Subsequent steps
lead to the l-carbon-longer homolog of a-keto-isovalerate,
a-keto-isocaproate. This is transaminated to yield leucine.
Usually this enzyme is quite specific for a-keto-isovalerate,
but mutants have been obtained in Serratia marcescens
in which other keto-acids are elongated (26). This leads
to the production in these mutants of the abnormal amino
acids norvaline, norleucine, homoisoleucine, homoleucine,
and homovaline. A similar activity appears to be utilized

in the spirochete Leptospira to produce a-ketobutyrate (via

 

a citramalate intermediate) from the one-carbon-shorter
homolog, pyruvate (13). It is not clear if this activity
for isoleucine synthesis is "borrowed" from the leucine
pathway or should be considered a "shared" enzyme, but the
activity appears to be regulated by isoleucine suggesting
the latter is the case. In this regard, this system is one
of several ways to bypass the usual pathway of a a-ketobu-
tyrate synthesis from threonine.

A variety of alternate pathways to form a-ketobutyrate

have been observed (Figure 2). In addition to the pathway

15

 

ASPARTATE
HOMOSERINE e~: P“°5P“°'
HOMOSERINE"““‘-* THREONINE
V
METHIONINE 11:; GPKETOBUTYRATE

 

/

GLUTAMATE —-—+ “‘METHYLASPARTATE

 

PYRUVATE 4: CITRAMALATE

 

ISOLEUCINE

Figure 2. Known pathways to isoleucine

via a-ketobutyrate

16

from threonine and the aforementioned elongation of pyruvate,
a—ketobutyrate can reportedly be synthesized from phos-
phoserine, methionine, and glutamate. In Bacillus subtilis,
phosphohomoserine is dephosphorylated and deaminated directly
to a-ketobutyrate, without the usual threonine intermediate
(37). Methionine is converted to a-ketobutyrate by some or-
ganisms (45). B-Methyl-aspartate (which can be synthesized
from glutamic acid) is converted to isoleucine in E. 22;;

W (33) and in Acetobacter suboxydans (5), probably via
B-methyloxaloacetate and a-ketobutyrate.

The reactions following the condensation of pyruvate
and a-ketobutyrate outlined in Figure l have been observed
in all organisms studied which produce isoleucine from
a-ketobutyrate. An alternate pathway to the penultimate
product of isoleucine synthesis, a-keto B-methylvalerate, is
via the reductive carbozylation of a-methylbutyric acid.
Isoleucine biosynthesis from a-methylbutyrate has been de—
monstrated in anaerobic bacteria from the rumen (35), and
in E. sporogenes (R.S. Hadioetomo, 1980 Ph.D. dissertation,
Michigan State University). Many rumen bacteria require
branched-chain volatile acids for growth (1) while for others
they stimulate growth but are not requisite. There is no
evidence to suggeSt that a-methylbutyrate is synthesized

E3 novo in these bacteria.

17

Reductive Carboxylations

The reductive carboxylation of organic acids is an
important mechanism for CO2 assimilation in some anaerobic
bacteria (8). While most microbes synthesize a-ketoglu-
tarate utilizing the "forward" tricarboxylic acid cycle via
isocitric dehydrogenase, some anaerobes produce this pre-
cursor for glutamate through the reductive carboxylation of
succinate. The physiological importance of the a-ketoglu-
tarate synthase activities was first demonstrated in the
green photosynthetic bacteria Chlorobium thiosulfatophilum

and Rhodospirillum rubrum (7), and has been shown to be the

 

predominant pathway in several Bacteriodes species (2).

 

In the photosynthetic bacteria, a-ketoglutarate synthase and
a—ketopropionate (pyruvate) synthase have been shown to
play central roles in the reductive carboxylation cycle
utilized by these bacteria as the principal mechanism to
produce the precursors for amino acids and other cellular
constituents (9). One complete turn of this cycle results
in the incorporation of four melecules of CO2 and the net
synthesis of one molecule of oxaloacetate.

a-Keto acid synthases catalyze reactions which rarely
occur in aerobic cells. Pyruvate synthase and a-ketoglu-
tarate synthase each reverse reactions associated with the
tricarboxylic acid cycle which are irreversible in aerobic
cells; the decarboxylation of pyruvate to yield acetyl-
CoA and C02, and the decarboxylation of a—ketoglutarate to

yield succinyl-CoA and C02. These reversals are made

18

possible by the strongly electronegative potential of fer-
redoxin found in many anaerobic bacteria which overcomes the
energy barrier for carboxylation (8). The potential of re-
duced ferredoxin is approximately equal to that of hydro-
gen gas (Hz). a-Ketoglutarate synthase activity has been
demonstrated in several genera of rumen bacteria including

Veillonella, elenomonas, and Bacteriodes (2) although the

 

 

exact nature of the low potential electron carrier is not
known.

In recent years it has become clear that the reduc-
tive carboxylation reaction is not limited to acetate and
succinate substitutes. The reactions below illustrate the
presumed steps in the reductive carboxylation, although the
exact mechanism is not well understood.

8
i) R - COOH + ATP-» R - C - Pi + ADP
ii) R - CO - Pi + HSCoA-OR - CO - SCOA + Pi

TPP
111) R - CO - SCoA + CO2 + Ferredox1nred -—a

R - CO - COOH + HSCoA + Ferredoxinbx.

The reductive carboxylation of a-methylbutyrate to
yield a-keto-B-methylvalerate for isoleucine synthesis has

been demonstrated in rumen bacteria (35) and E. sporogenes

 

(R.S. Hadioetomo, 1980 Ph.D. dissertation, Michigan State
University). Isobutyrate and isovalerate have been shown
to be precursors of valine and leucine in some rumen bac-
teria preparations (1, 10, 36). The synthesis of a-keto-

butyrate via the reductive carboxylation of pr0pionate has

19

been observed in Chromatium, Clostridium pasteurianum,

 

 

and Desulfovibrio (7), and in mixed rumen bacterial prepara-

 

tions (10, 2).

LITERATURE CITED
Allison, M. J., and M. P. Bryant. 1963. Biosynthesis
of branched-chain amino acids from branched-chain fatty
acids by rumen bacteria. Arch. Biochem. Biophys.
101:269-277.
Allison, M. J., and I. M. Robinson. 1970. Biosynthesis
of a-ketoglutarate by the reductive carboxylation of

succinate in Bacteriodes ruminicola. J. Bacteriol.

 

104:50-56.
Allison, M. J., I. M. Robinson, and A. L. Baerz. 1979.
Synthesis of a-ketoglutarate by reductive carboxylation

of succinate in Veillonella, Selenomonas and Bacteroides

 

 

 

species. J. Bacteriol. 140:980-986.

Barker, H. A. 1981. Amino acid degradation by anaerobic
bacteria. Ann. Res. Biochem. 50:23-40.

Belly, R. T., S. Greenfield, and G. W. Claus. 1970.
B-methylaspartate metabolism and isoleucine synthesis

in Acetobacter suboxydans. Bacteriol. Proc. p. 141.

 

Berg, C. M., K. Shaw, and A. Berg. 1980. The ilvG

gene is expressed in Escherichia coli K-12. Gene

 

12:165-170.
Buchanan, B. B. 1972. Ferredoxin-linked carboxylation
reactions, p. 193-216. IE,P° Boyer (ed.), The Enzymes,

V01. VI, 3rd ed., Academic Press Inc., New York.

20

10.

11.

12.

13.

14.

21

Buchanan, B. B., and D. I. Arnon. 1969. The reductive
carboxylic acid cycle, p. 170-181. £2.J- M. Lowenstein
(ed.), Methods in Enzymology, Vol. XIII. Academic
Press Inc., New York.

Buchanan, B. B., and M.C.W. Evans. 1965. The synthesis
of a-ketoglutarate from succinate and carbon dioxide by
a subcellular preparation of a phogosynthetic bacterium.
Proc. Natl. Acad. Sci. U.S.A. 54:1212-1218.

Buch, R. S., and F. D. Sauer. 1976. Enzymes of 2-oxo
acid degradation and biosynthesis in cell-free extracts
of mixed rumen microorganisms. Biochem. J. 157:325-
331.

Buhler, M., H. Giesel, W. Tischer, and H. Simon. 1980.
Amino acid metabolism. FEBS Lett. 109:244-246.

Chan, T.T.K., and E. B. Newman. 1981. Threonine as a

carbon source for Escherichia coli. J. Bacteriol.

 

145:1150-1153.

Charon, N. W., R. C. Johnson, and D. Peterson. 1974.
Amino acid biosyntheses in the spirochete Leptospira:
evidence for a novel pathway of isoleucine synthesis.
J. Bacteriol. 117:203-211.

Costilow, R. N. 1977. Selenium requirement for growth

of Clostridium sporogenes with glycine as the oxidase

 

in Stickland reaction systems. J. Bacteriol. 131:366-

368.

15.

16.

17.

18.

19.

20.

21.

22.

22

Costilow, R. N., and D. Cooper. 1978. Identity of
proline dehydrogenase and A'-pyrroline S-carboxylic

acid reductase in Clostridium sporogenes. J. Bacteriol.

 

134:139-146.

Davis, B. D., R. Dulbecco, H. Eisen, and H. Ginsberg.
1980. Microbiology, 3rd ed., p. 712-722. Harper and
Row, Publishers, Inc., Hagerstown.

Eklund, M. W., and F. T. Poysky. 1974. Interconversion

of type C and D strains of Clostridium botulinum by

 

specific bacteriophages. Appl. Microbiol. 27:251-258.
Eklund, M. W., F. T. Poyski, J. A. Meyers, and G. A.

Pelroy. 1974. Interspecies conversion of Clostridium

 

botulinum type C to Clostridium novyi type A by bacterio-

 

 

phage. Science 186:456-458.

Elsden, S. R., and M. G. Hilton. 1978. Volatile acid
production from threonine, valine, leucine, and iso-
leucine by Clostridia. Arch. Microbiol. 117:165-172.
Elsden, S. R., M. G. Hilton, and J. M. Waller. 1976.
The end-products of the metabolism of aromatic amino
acids by Clostridia. Arch. Microbiol. 107:283-288.
Freeman, B. A. 1979. Burrows Textbook of Microbiology,
let ed., p. 641-666. W. B. Saunders Company,
PHiladelphia.

Guardiola, J., M. DeFlice, and M. Iaccarino. 1974.
Mutant of Escherichia coli K-12 missing autolactate

synthase activity. J. Bacteriol. 120:536-538.

23.

24.

25.

26.

27.

28.

29.

30.

23
Gutteridge, C. S., B. M. Mackey, and J. R. Norris.
1980. A pyrolysis gas-liquid chromatography study of
Clostridium botulinum and related organisms. J. Appl.
Bacteriol. 49:165-174.
Iaccarino, M., J. Guardiola, M. DeFelice, and R. Fauro.
1978. Regulation of isoleucine and valine biosynthesis.
13 Current Topics in Cellular Regulation, Vol. 14, p.
29-73. Academic Press Inc., New York.
Kisumi, M., M. Sugiura, and I. Chibata. 1976. Bio-
synthesis of norvaline, norleucine, and homoisoleucine
in Serraria marcescens. J. Biochem. 80:333-339.
Lawther, R. P., D. H. Calhoun, C. W. Adams, C. A.
Hauser, J. Gray, and G. W. Hatfield. 1981. Molecular

basis of valine resistance in Escherichia coli K-12.

 

Proc. Natl. Acad. Sci. U.S.A. 78:922-925.
Lee, W. H., and H. Riemann. 1970. The genetic

relatedness of proteolytic Clostridium botulinum

 

strains. J. Gen. Microbiol. 64:85-90.

Mead, G. C. 1971. The amino acid-fermenting Clostridia.
J. Gen. Microbiol. 67:47-56.

Monticello, D. J., and R. N. Costilow. 1981. Purification
and partial characterization of proline dehydrogenase

from Clostridium sporogenes. Can. J. Microbiol.

 

27:942-948.
Muth, W. L., and R. N. Costilow. 1974. Ornithine
cyclase (deaminating). III. Mechanism of the conversion

of ornithine to proline. J. Biol. Chem. 254:7463-7467.

31.

32.

33.

34.

35.

36.

37.

24
Perkins, W. E., and K. Tsuji. 1962. Sporulation of

Clostridium botulinum. II. Effect of arginine and its

 

degradation products on sporulation in a synthetic
medium. J. Bacteriol. 84:86-94.

Phillips, A. T., J. I. Nuss, J. Mossic, and C. Foshay.
1972. Alternate pathway for isoleucine biosynthesis in

Escherichia coli. J. Bacteriol. 109:714-719.

 

Poston, M. J. 1976. Leucine 2,3-aminomutase, an
enzyme of leucine catabolism. J. Biol. Chem. 251:
1859-1863.

Robinson, I. M., and M. J. Allison. 1969. Isoleucine
biosynthesis from 2—methylbutyric acid by anaerobic
bacteria from the rumen. J. Bacteriol. 97:1220-1226.
Sauer, F. D., J. D. Erfle, and S. Mahadevan. 1975.
Amino acid biosynthesis in mixed rumen cultures.
Biochem. J. 150:357-372.

Schildkraut, I., and S. Greer. 1973. Threonine synthase-
catalyzed conversion of phosphohomoserine to d-keto-

butyrate in Bacillus subtilis. J. Bacteriol. 115:777-

 

785.
Smith, L.D.S., and G. Hobbs. 1974. Genus III.

Clostridium. lg R. E. Buchanan and N. E. Gibbons

 

(eds.), Bergey's manual of determinative bacteriology,
8th ed., p. 551-572. Williams and Wilkins Company,

Baltimore.

38.

39.

40.

41.

42.

43.

25
Strassman, M., A. J. Thomas, L. A. Locke, and S.
Weinhouse. 1956. The biosynthesis of isoleucine. J.
Am. Chem. Soc. 78:228-232.
Strassman, M., A. J. Thomas, and S. Weinhouse. 1955.
The biosynthesis of valine. J. Am. Chem. Soc. 77:1261-
1264.
Takami, K., T. Kinouchi, and T. Kawata. 1980. Isolation
of two inducible bacteriophages from Clostridium botulinum
type A 1906. FEMS Letts. 9:23-27.
Umbarger, H. E. 1973. Threonine deaminases. Adv.
Enzymol. 37:349-395.
Umbarger, H. E. 1978. Amino acid biosynthesis and its
regulation. Ann. Rev. Biochem. 47:533-606.
Whiteley, H. R., and M. Tahar. 1966. Threonine deaminase

of Clostridium tetanomorphum. I. Purification and

 

prOperties. J. Biol. Chem. 241:4881-8887.

SECT ION I I

Isoleucine synthesis from a-methylbutyrate or

propionate by Clostridium sporogenes
Daniel J. Monticello and Ralph N. Costilow
Department of Microbiology and Public Health

Michigan State University

East Lansing, MI 48824

Running head: Isoleucine synthesis in E. §porogenes

26

ABSTRACT
Preliminary studies showed that E. §porogenes (ATCC
7955, NCA PA 3679) made isoleucine in a novel pathway not
involving threonine or threonine dehydratase. Radioactivity

l4C]propionate and a-amino-[3-14C]

from [3-14C]pyruvate, [
butyrate was incorporated into isoleucine, and increasing
concentrations of a-methylbutyrate were found to reduce the
incorporation of [14C1pyruvate. Crude extracts of the
bacteria grown in a minimal medium were found to contain
levels of a-acetohydroxyacid synthase activities comparable
to those found in cells of E. 29;; K12 grown in minimal
medium. Stepwise degradation of isoleucine labeled with

specifically-labeled precursors supports the conclusion that

E. §porogenes can make isoleucine via the reductive carboxy-

 

lation of propionate to yield a-ketobutyrate which is
metabolized to isoleucine in the classical fashion, or
isoleucine can be made via the reductive carboxylation of a-

methylbutyrate to a-keto-B-methylvalerate.

27

INTRODUCTION
The pathway for branched-chain amino acid synthesis is
well established in many aerobic organisms (32) and the
regulation is known in some detail (15). Little consider-
ation has been given to the synthesis of these molecules by
strictly anaerobic bacteria, due mainly to the lack of a
simple, defined medium for growth of these bacteria. The

recent development in our lab of such a medium for Clostridium

 

sporogenes (R. S. Hadioetomo, 1980, Ph.D. dissertation,
Michigan State University) revealed that this species
requires valine or leucine for growth, but not isoleucine.
Further investigation showed that these bacteria synthesize

isoleucine in a novel fashion from pyruvate and CO and

2,
that a-[14C1methylbutyrate is incorporated exclusively into
isoleucine. Label from [14C]threonine was not incorporated
into isoleucine, and the threonine dehydratase in these
organisms was found to be catabolic in nature.

We have therefore sought to determine the nature of
isoleucine biosynthesis in E. §porogenes. Here we report
that these bacteria incorporate [14C1propionate into iso-
leucine, probably via the reductive carboxylation of propionate
to a-ketobutyrate. Crude extracts possess significant
levels of a-acetohydroxyacid activity, and presumably metabo-
lize a-ketobutyrate to isoleucine via the classical pathway.
Isotope competition experiments and degradation of iso-

leucine isolated from cells labeled with specifically-

labeled precursors indicate the E. sporogenes produces

 

28

29

isoleucine either from the reductive carboxylation of
propionate to a-ketobutyrate, or from the reductive carboxy-
lation of a-methylbutyrate to a-keto-B-methylvalerate, which

is transaminated to form isoleucine.

MATERIALS AND METHODS

Culture and growth medium. Clostridium §porogenes
(ATCC 7955, National Canners Association PA 3679) was used
in all experiments. Procedures for maintaining the culture
and for preparation of inocula were as previously described
(12). The minimal medium used in these experiments is a
modification of the synthetic medium of Perkins and Tsuji
(23) containing salts, vitamins and eight amino acids (L-
serine, 25 mM; L-arginine, 20 mM; L-phenylalanine, 10 mM; L-
tyrosine, 2 mM; L-tryptophan, 2 mM; L-methionine, 10 mM; L-
valine, 10 mM and L-leucine, 10 mM). The bacteria were
grown in an anaerobic chamber (Coy Mfg., Ann Arbor, MI) at
37°C in an atmosphere of 5% carbon dioxide and 10% hydrogen
in nitrogen.

Labeling and fractionation of cells. [14C]-Labeled
compounds were added to exponentially growing cells (10 m1
cultures, O.D. m0.3) which were grown for 4 h (1 generation)
and then harvested by centrifugation. Typically, the cell
pellet was extracted twice with 5 ml of 10% trichloroacetic
acid (TCA) for 15 min, followed by 30 min of extraction in
10% TCA at 90°. Following this extraction, the pellet was
washed once with hot 10% TCA, suspended in 1.5 ml 6 N HCl
and hydrolysed at 110° for 24 h in an argon-flushed ampuole.
The hydrolysate was evaporated to dryness, dissolved in
distilled water and adsorbed onto a cation exchange resin
(AG 50w-x4, hydrogen form; Biorad) packed in a pasteur

pipette. The resin was washed with 15 m1 of distilled

30

31
water, and the amino acids eluted with 1 N ammonium hydroxide.
The eluate was evaporated to dryness, resuspended in 100 pl
of water and stored at -20°.

Acetohydroxyacid synthase assay. Enzyme activity was
assayed in crude extracts of E. gporogenes and E. 22;; K12
(source R. R. Brubaker). E. 22;; K12 was grown at 37° on a
rotary shaker in the minimal medium (without citrate) of
Davis and Mingoli (13). E. §porogenes was grown in the
minimal medium. Crude extracts of these organisms were made
from cells harvested from 500 ml of exponentially growing
cultures. Pelleted, washed cells were suspended in 3 m1 of
the buffer used for washing (50 mM potassium phosphate, 10%
glycerol and 0.5 mM dithiothreitol, pH 7.5), and sonicated
for 3 x 30 sec, with 1 min cooling intervals. Cellular
debris was removed by centrifugation, and the extract used
immediately.

Acetolactate synthase activity in crude extracts was
determined by the method of Stormer and Umbarger (29).

Typical reaction mixtures contained 1 M potassium phosphate,
pH 8.0, 0.1 m1; 1 mM thiamine pyrophosphate, 0.1 ml; 0.2 mg/ml
flavin adenine dinucleotide (FAD), 0.1 ml; 0.1 M MgClZ,
0.1 ml; 0.2 M sodium pyruvate, 0.2 m1; crude extract and
water to a final volume of 1 ml. The reaction was stopped
after various intervals with 0.1 ml 50% H2804. The concen-
tration of acetoin was determined with the addition of

1.0 ml 0.5% creatine hydrate and 1.0 ml 5% a-naphthol in 10%

32

NaOH. After 1 h at room temperature, absorbance at 540 nm
was measured, and the concentration of acetoin determined
from a standard curve.

A similar reaction mixture was used for the deter-
mination of acetohydroxybutyrate synthase activity, except
0.1 m1 of 0.2 M sodium pyruvate and 0.1 m1 of 0.2 M a-
ketobutyrate acid were used and the reaction stopped with

0.1 ml of 10% (w/v) ZnSO and 0.1 ml of 1 N NaOH. Enzyme

4
activity was determined using the microbiological assay
developed by Leavitt and Umbarger (17), which is based on
the fact that the inhibition of the growth of E. 22$; K12 by
L-valine is reversed by any six-carbon precursor of is
oleucine. Growth of E. 22$; K12 in the minimal medium
supplemented with 50 ug/ml L-valine and various dilutions of
the reaction mix was compared with tubes with known concen-
trations of isoleucine. The growth response of E. ggEE_K12
in this medium with added isoleucine is very similar to that
with added acetohydroxybutyrate (18).

Specific activity is expressed as umoles of acetolactate
or acetohydroxybutyrate formed per g of protein per hour.
Protein concentrations were determined by the method of
Lowry gg|§l. (19). All enzyme reactions were performed at
37° in the anaerobic chamber.

Separation of amino acids. Amino acids were separated
by either thin layer or descending paper chromatography, and
high-voltage paper electrophoresis. Whatman LK-S-D linear K

preadsorbant plates (Whatman, Inc., Clifton, NJ) developed

33

with a methy1ethylketone-pyridine-water-glacial acetic acid
(73:5:20:2) solvent system (20) were used to separate iso-
leucine from all other amino acids. L-aspartate and L-
glutamate were separated from other amino acids and each
other by paper electrophoresis on Whatman No. 1 paper in
0.25 M sodium acetate buffer at pH 4.3. Ten p1 samples of
protein hydrolysates were spotted on the paper and electro-
phoresed at 17-35 volts/cm for 90 min. Isoleucine was
readily obtained for degradation experiments utilizing
descending paper chromatography with a butanol-acetic acid-
water (120:30:50) solvent system. Hydrolysates were spotted
in 5 pl amounts across sheets of Whatman No. 1 paper (46 x
57 cm). Following development, the isoleucine/leucine
region of the chromatogram was identified and removed, and
the amino acids eluted. The coelution of isoleucine and
leucine in this solvent system was not a problem, since
leucine was never significantly labeled. Amino acids were
located by comparison with identical adjacent lanes sprayed
with ninhydrin (0.1% in acetone).

14

Isoleucine degradation. The location of C incorporated

 

into isoleucine from labeled precursors was determined with

a stepwise degradation of isoleucine from protein hydrolysates.
The procedure is essentially that of Strassman SE 3E. (30)

with some modifications. Isoleucine was decarboxylated with
ninhydrin and the a-methylbutyraldehyde formed distilled

in 25% (v/v) H SO

3 2 4
oxidized to a-methylbutyrate (14). The a-methylbutyrate was

into 20 ml of 5 M CrO and directly

34
recovered by steam distillation and degraded to gsg-butyl-
amine via the Schmidt reaction (24). This was oxidized to
methylethylketone and subsequently degraded to propionate
and C02. The CO2 was trapped in l N NaOH and precipitated

as BaCOB. The BaCO3 was washed with distilled water,

suspended in 5 ml of water, and 5 ml conc. H2504 was added

in a closed, evacuated 500 m1 jar with five vials containing

1 m1 of hydroxide of hyamine to trap the CO After 12 h,

2.
10 ml of toluene-based scintillation fluid was added to each

of the vials, and the radioactivity determined in a scintil-

lation counter.

14

Materials. C-Labeled compounds were obtained from

 

New England Nuclear, Boston, MA. Hydroxide of hyamine was
purchased from Packard Instrument Co., Inc., Downers Grove,
IL. Aqueous scintillation cocktail was obtained from New
England Nuclear and the toluene-based fluid prepared by
mixing 6.0 g of 2,5-diphenyloxasole and 0.01 g 1,4-bis-2-(5-
phenyloxazoyl)-benzene in 1 liter of toluene; both chemicals
were obtained from Research Products International Corp.,

Elk Grove Village, IL.

RESULTS

14

Effect of a-methylbutyrate on [U-14C]pyruvate and CO

2
incogporation into isoleucine. a-Methylbutyrate can be

converted to isoleucine by E. §porogenes (R. S. Hadieotomo,

 

Ph.D. dissertation, Michigan State University). Cells were
grown in the minimal medium with 0, 5, 10 or 15 mM a-
methylbutyrate, and the incorporation of label from exo-
genous [U-14C1pyruvate into isoleucine and aspartate in the
protein determined. [U-14C1Pyruvate (10 pCi) and a-methyl-
butyrate were added to 10 ml cultures with Optical densities
of about 0.3, which then were allowed to grow for 4 h. The
protein was extracted from the cells, hydrolysed, and
aspartate and isoleucine separated and the radioactivity
determined. An identical experiment was performed with
l4C-NaHCO3 (10 pCi/lO ml culture). Figure 1 shows that
increasing concentrations of a-methylbutyrate lead to a
significant decrease in the incorporation of label from [U-

14C]pyruvate into isoleucine relative to aspartate. No such

change is seen with the 14CO2 label.

14

Incorporation of a-amino-[3- Clbutyrate into isoleucine.

Although E. gpprogenes cells grown in the minimal medium do
1

 

not incorporate L-[ 4C]threonine into isoleucine, we wished
to determine if alternate sources of a-ketobutyrate might be
incorporated. A 10 ml culture was labeled for one generation

l4C]butyrate (25 pCi). The radioactive

with a-amino-[B-
amino acids present in the protein hydrolysate migrated with

the adjacent isoleucine control spot (Fig. 1). These data

35

36

 

 

DPH L-ISOLEUCINE
DPH L-ASPARTATE
5°
W
I

 

 

 

0.2L-
luC-PYRUVATE
O\
Otlh 0\
’ ‘0
o J J l

5 10 15

CONCENTRATION or Z-HETHYLBUTRYIC Acxo

Fig. 1. Effect of a-methylbutyrate on the incorporation

14

of [U-14C]pyruvate and CO into isoleucine. E.‘spongenes

 

2
was grown for one generation in a minimal medium supplemented
with 0, 5.0, 10 or 15 mM a-methylbutyrate and the 14C-
labeled compounds indicated for one generation. Since the
specific activity of aspartate did not change, the ratio of

cpm in isoleucine to cpm in aspartate is shown on the

abcissa.

37

 

 

 

 

800,
GLUTAMATE VALINE sgcxne
700 ~ . O . O
n \. //’
ASPARTATE PROLINE lSOLEUClN
600 ..
500 -
:4 3-1l'C-ALPHA-AMIHOBUTYRATE
é “00 " LABELED
E; PROTEIN HYDROLYSATE
g 300 ~
.5.
{.3 200 _
U)
\
SE
9’ 100 P
_L l l 1 J l l J 1 l J l 1 i

 

Ilia I] 1
2,4 5 81012141618202224262330323436.
SECTION NUMBER

Fig. 2. Incorporation of d-amino-[3-14C]butyrate into
amino acids. 10 ul samples of the protein hydrolysate were
applied to TLC plates. After development, 5 mm fractions
were scraped from the lanes and counted. The location of
amino acid standards chromatographed on an adjacent lane are

shown at the tOp of the figure.

38

show that the C-3 carbon of a-aminobutyrate is incorporated

14c-iabei is in

into isoleucine, since essentially all the
this spot.

a-Acetohydroxyacid synthase activity_in crude extracts.

 

Since a-ketobutyrate appeared to be a precursor of isoleucine,
we assayed crude extracts for the first enzyme in the
classical pathway for branched-chain amino acid synthesis,
a-acetohydroxyacid synthase (AHAS). The AHAS isoenzymes in

' E. EQli catalyze two reacions; the condensation of two
pyruvate molecules to form a-acetolactate and CO2 (for

valine and leucine synthesis) and the condensation of
pyruvate and a-ketobutyrate to form a-aceto-a-hydroxybutyrate
and C02. In crude extracts of E. sporogenes and E. 99;; K12
cells grown in minimal media, the specific activities of a-
acetolactate synthase were 2.2 pmoles/min/g protein and

3.0 pmoles/min/g protein respectively. The a-aceto-a-
hydroxybutyrate synthase specific activities were 2.4 pmoles/

min/g and 6.1 pmoles/min/g protein. These data show that

when E. sporogenes is grown in the minimal medium, it

 

possesses considerable AHAS activity but somewhat less than

that found in minimal medium grown E. 99;; K12.
Incorporation of 14C-propionate into isoleucine. Our

finding that a-ketobutyrate was a precursor of isoleucine in

E. gporogenes, although L-threonine and L-glutamate are not

 

(R. S. Hadioetomo, 1980, Ph.D. dissertation, Michigan State
University), led us to look for other sources of a-keto-

butyrate in the cell. The fact that E. sporogenes can

39

.800 _
GLUT/AMATE lSOLEUC INE
700 _ on I. O.\ O ..
ASPARTATE PROLINE VAL{NE LEUCINE
600 _,
500 _
.u:
S
Q.
._. uoo -
F‘ 2-1l‘C-PR0P10NATE
g LABELED
E 300 '- PROTEIN HYDROLYSATE
.7:
8
m 200 _
‘1.
a:
Q-
U
100 .—
1 ’L 1 l _i _ #1 _L J J l J J u.

 

 

 

2 Q 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

SECTION NUMBER

Fig. 3. Incorporation of [2-14

C]propionate into amino
acids. 10 ml samples of the protein hydrolysate were applied
to TLC plates. After development, 5 mm fractions were
scraped from the lanes and counted. The location of amino

acid standards chromatographed on an adjacent lane are shown

at the top of the figure.

40

reductively carboxylate d-methylbutyrate to form a-keto-B-
methylvalerate suggested that these cells might also

reductively carboxylate propionate to form a-ketobutyrate.
A 10 ml culture was labeled for one generation with 50 pCi

of [2-14

C]propionate, and the amino acids in the protein
hydrolysate separated on TLC plates (Fig. 3). Most of the
label in the protein hydrolysate migrated with the adjacent
isoleucine control spot. These data show that propionate is
also a precursor of isoleucine.

When the supernatent fluid from [1-14C], [2-14C] and
[3-14C]propionate labeled cultures were analyzed by paper

14C-label was found

chromatography, a significant amount of
in a-aminobutyrate (Table 1). The amino acids from the
supernatent media were separated in n-butanol:acetic acid:
water (120:30:50), methanol:water:pyridine (l60:40:8) and n-
butanol:pyridine:water (60:60:60). In each case, the 14C-
label amino acid had the same Rf as u-aminobutyrate. No
other amino acid had a similar Rf in all three cases.

Effect of propionate on [U-14C] serine incorporation

 

into isoleucine. Previous data showed that E. sporogenes
l4

 

incorporates a significant amount of C from [14C]-serine
into isoleucine (R. S. Hadioetomo, 1980, Ph.D. dissertation,
Michigan State University). To examine the effect of
exogenous propionate on [U-14C]serine incorporation into

isoleucine, 10 m1 cultures and E. sporogenes in the basal
l4

 

medium were supplemented with 10 pCi of [U- C]serine and

1.0, 2.0 or 3.0 mg of sodium propionate. The cultures were

TABLE 1. Incorporation of

butyrate

41
14

C-propionate into a-amino

 

Rf in solventa

 

 

Amino acid BuA BuP MeP
Radioactive spot 0.45 0.38 0.60
L-a-aminobutyrate 0.44 0.38 0.61
L-proline 0.37 0.39 -
L-methionine 0.50 0.50 0.60
L-aspartate 0.24 0.14 -
L-tyrosine - 0.61 -
L-tryptophan 0.50 0.62 -
L-leucine 0.69 0.60 0.68
L-valine 0.51 0.47 0.64
L-serine - 0.31 -
L-threonine - 0.31 0.57
L-alanine 0.34 0.35 -
L-phenylalanine 0.59 0.62 0.61

 

aAmino acids were separated by descending paper chromato-

graphy. Solvents:

BuA;

n-butanol:acetic acid:water

(120:30:50), BuP; n-butanol:pyridine:water (60:60:60) and

MeP; methanol:water:pyridine (l60:40:8).

42

 

 

 

 

TABLE 2. Effect of propionate on incorporation of [U-14C]-
serine into isoleucine.
Propionate cpm fraction of total in:
(mg/ml.) Total ASP GLU ILE ASP GLU ‘ILE

0 6430 500 966 109 0.088 0.15 0.017
0.1 5490 555 778 28 0.10 0.14 0.005
0.2 3860 276 558 25 0.07 0.15 0.006
0.3 2590 208 432 16 0.08 0.17 0.006

 

a10 p1 of the protein hydrolysate was used. Aspartate and
glutamate were separated by paper electr0phoresis, isoleucine
by thin layer chromatography.

43

allowed to grow for three generations and were then harvested
and the protein extracted. When the ratios of cpm in
aspartate/total cpm in amino acids and cpm in glutamate/
total cpm in amino acids were examined (Table 2), increasing
concentrations of propionate in the medium were found to

have no effect on the percent fraction incorporated into
these amino acids. In contrast, the cpm in isoleucine/total
cpm in amino acids ratio was found to decrease significantly
(by about 65%) with added pr0pionate (0.1 mg/ml). Incorpor-
ation of [U-14C]serine into isoleucine was reduced to a very
low level, but not zero. This isotope competition of
[14C]serine incorporation by propionate indicates that these
two substrates provide some common element in isoleucine
biosynthesis. The fact that some minimal level of [14C]serine
incorporation was attained suggests that serine may provide
some element not produced from prOpionate.

14C] threonine into isoleucine.

l

Incorporation of L-[U-

 

Although previous results indicated that L-[U- 4C]threonine

is not incorporated into isoleucine by E. sporogenes grown

 

in the minimal medium (R. S. Hadioetomo, 1980, Ph.D.
dissertation, Michigan State University), the fact that a-
ketobutyrate is incorporated into isoleucine prompted us to
reexamine this phenomenon. In the previous study, cells
growing in the minimal medium were labeled with very high
specific activity [14C]threonine and a very low total
threonine concentration (25 pCi of label with a specific

activity of 210 pc/pmole). In this experiment, the minimal

44

medium was supplemented with 10 mM L-threonine. After

several generations of growth in this medium, 25 pCi of L-

[U-14C]-threonine were added, and the cells grown for one

generation. As in the previous study, most of the label in
the protein hydrolysate was in threonine. However, while no
significant label was recovered in isoleucine in the previous

study (less than 150 dpm from a total of 124,000 dpm), there

14

was significant incorporation of C from L-[U-14C]threonine

into isoleucine (over 270 cpm of 13,200 cpm incorporated) in

this experiment.

1

Degradation of isoleucine labeled with [3- 4C]pyruvate,

[1-14C]pr0pionate and [2-14C]propionate. Specifically

 

labeled isoleucine from E. sporogenes was degraded in a

 

stepwise fashion to locate the position of the label in the
isoleucine molecule (Tables 3 and Fig. 4). The recoveries

of various carbons by this procedure were low, therefore the
results must be compared to appropriate controls. When the

l4C]pyruvate, no significant

isoleucine was labeled with [3-
incorporation occurred in C-l or C-2. All of the label was
incorporated into the last four carbons. Cleavage of Egg-
butylamine results in the formation of propionate (C-3, -

14C-label was found in

4, -5) and iodoform (C-6) and the
both fractions. In the classical pathway, the C-6 of
isoleucine arises from pyruvate, and carbons 4 and 5 arise
from a-ketobutyrate. The degradation results with [3-

14C]pyruvate labeled isoleucine show that if isoleucine is

45

TABLE 3. Degradationa of isoleucine from E. Sporogenes grown

 

in the presence of [3-14C]pyruvate, [1-14C1-
propionate and [2-14C]propionate

 

 

 

U-14C-isoleucine 14C-growth substrate

Degradation control [3-I4C]- [1-14C]- [2-14C]

product degradation pyruvate propionate pr0pionate
Isoleucine 55,000 5,600 1,300 2,200
C-l 4,900 (53)b 56 (1) None 140 (6)
a-methyl 15,000 (32) 4,400 (79) 560 (43) 1,600 (73)
butyrate
C-2 170 (2) None 290 (22) 60 (3)
Sec-butyl 3,600 (10 900 (16) None 230 (10)
amine
Propionate 750 (3) 390 (7) - 92 (4)
Iodoform 140 (2) 140 (2.5) - -

 

3See Figure 4.
Percent theoretical recovery.
CPercent recovery.

46

.mHHouoo Hmucofifluomxo Mom moonuoz pom mamwuoumz mom

.OQHU:OH0mfi mo soHumpmumop on» How ofiosom d .v .mwm

smonoooa me<zo_aoma mz~z<5>e=mummw
o m e m m e m
m_=o + =eeo-~=o-m=o ...... ......N=z-=L-N=e-m=e
m5
0
mh<m>h3m4>zhm2|o mz—o:m40w~
N N . N a
co N m e m ou sz m e m
(k 58-5915 Tl. PISBSSRERS
m=e N mzw

o o

47
made via the classical condensation of pyruvate and a-
ketobutyrate, pyruvate must also be incorporated into a-
ketobutyrate prior to this step.

All the label from [1-14C]propionate in isoleucine
appears to be in C-2. No label is released during decarboxy-
lation of isoleucine, but over 65% of the radioactivity in
a-methylbutyrate was recovered as C02. These results are as
expected if isoleucine is made from a-ketobutyrate, which is
in turn formed by the reductive carboxylation of propionate.
Although some label was released with the C-1 and C-2 carbons
of [2-14C]propionate labeled isoleucine, most of the label
remains in the Egg-butylamine and a significant amount was
recovered in the propionate fraction. These results support
the conclusion that a-ketobutyrate is made from propionate,
and is subsequently incorporated into isoleucine in the

classical manner.

DISCUSSION

These results offer a good explanation of the unusual

14C-labeling of isoleucine by E. sporogenes reported by

 

Hadioetomo (1980, Ph.D. dissertation, Michigan State Univer-

14

sity). She observed that C-label from [14C]threonine was

not incorporated into isoleucine, while radioactivity from

14

[14C]pyruvate, [14C]serine, CO or [14C]-a-methylbutyrate

2
was incorporated. She concluded that isoleucine was made in
a novel fashion. The evidence presented here indicates that
the synthesis of isoleucine in the minimal medium proceeds
via the classical pathway (30) but that the a-ketobutyrate
used is derived from a source(s) other than threonine.

The results clearly show that a-ketobutyrate is a

precursor of isoleucine. When E. sporogenes was grown with
l 14

 

a-amino-[B- 4C]butyrate, the C-label was recovered pre-
dominantly (almost exclusively) in isoleucine. It is
probably converted to isoleucine via the classical pathway,
since the first enzyme in the pathway, acetohydroxyacid
synthase (AHAS) is present in crude extracts in levels
comparable to those found in E. ggli. The data from the
degradation of specifically-labeled isoleucine supports this
conclusion, and indicates that propionate is an important
source of a-ketobutyrate in E. sporogenes. If propionate
were carboxylated to a-ketobutyrate, most of the radio-
activity from [2-14C]propionate labeled isoleucine would be

expected in the C-4 position. When isoleucine was labeled

with this substrate, the bulk of the label was recovered in

48

49

the carbon 3-5 fragment of isoleucine. Similarly, 14C-

label from [1-14C]propionate was found in C-2 of isoleucine,

which is the ultimate location of the C-2 of a-ketobutyrate

in the classical pathway. Finally, 14

14

C-label from [3-
C]pyruvate was found in two locations in isoleucine, which
suggests that pyruvate is not only condensed with a-keto-
butyrate as in the classical pathway, but also that pyruvate
is a precursor of a-ketobutyrate. There are probably other
sources of a-ketobutyrate in the medium besides propionate,
since both L-methionine and L-threonine can be converted to

a-ketobutyrate by E. sporogenes growing in complex media

 

(4). When 9. sporogenes was grown with 10 mM L-threonine in

 

the minimal medium, a small amount of radioactivity from
[14C]threonine was recovered in isoleucine.
The unusual aspect of isoleucine synthesis in E.

sporogenes is that the predominant source of a-ketobutyrate

 

appears to be propionate, not threonine. An important

source of propionate is L-serine, which is present in high
concentration (25 mM) in the medium. The fact that increasing
concentrations of propionate in the medium decreases the

4C]serine into isoleucine supports this

14

incorporation of [1
proposal, as does the observation that some C-label from
[3-14C1PYruvate appears to be incorporated into isoleucine
through a-ketobutyrate. Serine is readily converted to

alanine by E. sporogenes growing in the minimal medium (R.

 

S. Hadioetomo, 1980, Ph.D. dissertation, Michigan State

University): via pyruvate. In this fashion, serine and

50

alanine could be fermented to prOpionate via pyruvate. The
fermentation probably proceeds by the acrylate pathway

observed in E. propionicum or by the succinate-propionate

 

pathway employed by other propionate producing bacteria
(13). Hadioetomo showed that pyruvate in the cell was
subject to extensive CO

Label from 14C0

2 exchange at the C-1 position.

2 in the medium was incorporated into the C-1
of serine and alanine, and therefore would also be found in
propionate. Since the C-1 of propionate eventually becomes
the C-2 of isoleucine (Fig. 5) this may explain Hadioetomo's

observation that 14

CO2 was incorporated into at least one
carbon other than the C-1 of isoleucine. The reductive
carboxylation of propionate to a-ketobutyrate which is
proposed here has been observed in a number of other anaerobic

bacteria, including E. pasteuranum (3,6).

 

a-Methylbutyrate is an alternate precursor of isoleucine

for E. sporogenes growing in the minimal medium. Hadioetomo

 

showed that [14CJ-a-methylbutyrate was incorporated exclusively
into isoleucine, and no radioactivity was released upon
decarboxylation. We have shown here that increasing concen-
trations of a-methylbutyrate in the medium decreased the
incorporation of [14C]pyruvate but did not change the

relative incorporation of radioactivity from 14

CO2 lnto
isoleucine. If propionate is reductively carboxylated to a-
ketobutyrate and processed to isoleucine in the classical
fashion, or a-methylbutyrate carboxylated directly to a-

keto-a-methylvalerate (the keto-acid analog of isoleucine),

 

.mmcmmonomm EsaofluumoHo

Ga mflmonuswm ocwosoaomfl How mmzaumm oomomoum .m .mHm

mh<mm4<>4>zhm2| Q Eggs
”Etuamxm; Ohyfio ozfizro w5§Sbafiflmzuo ubSfichzu

N12 me a N8

1 5395.915 «kl 58. ..-PNSLE Tl gage .me law-”Lie 1k: TSUNILS
5 m6 MS a

N8

L .5 he

1895.915 T8915... L
Me. 1.1.

Egg. 8 m._.<>:m>m

 

52

the net incorporation of 14

CO2 would not change regardless
of the prOportion of isoleucine made from a-methylbutyrate.
The data presented here support the proposal that propionate
and a-methylbutyrate are alternate sources of isoleucine.
The carboxylation reactions which occur are presumably
similar to the reductive carboxylation reactions which
produce a-ketobutyrate from propionate (3,6), a-keto-
glutarate from succinate (2,7) and a-ketoisocaproate from
isovalerate and a-ketoisovalerate from isobutyrate (1,8) in
other anaerobic bacteria. The incorporation of [14C1-a-

methylbutyrate into isoleucine in this fashion has been

demonstrated in the rumen species Bacteroides ruminicola

 

(26).

When E. sporogenes was grown in the minimal medium

 

considerable a-ketobutyrate was produced, leading to the

accumulation of a-amino-[14C]butyrate in the medium when the

1

cells were grown with [1- 4C]propionate, [2-14C]propionate

or [3-14C]propionate. E. sporogenes also accumulates a-

 

aminobutyrate when grown in a complex medium (21). In E.
EEEE, a-ketobutyrate is usually produced from threonine, and
the levels of a-ketobutyrate in the cell are strictly
controlled by feedback regulation of threonine dehydratase
by isoleucine and to a lesser extent valine and leucine
(16). Threonine is not the primary source of a-ketobutyrate

in E. sporogenes, nor is threonine dehydratase activity

 

regulated by isoleucine (R. S. Hadioetomo, 1980, Ph.D.

dissertation, Michigan State University). The accumulation

53

of a-aminobutyrate in the medium is of special significance
since it has been shown to be an antimetabolite of valine in
Bacillus anthracis (12) and E. coli K-12 (25). Gladstone

(12) observed that the growth of E. anthracis in a minimal

 

medium was inhibited by a-aminobutyrate in concentrations as
low as 0.04 mM, and that the inhibition could be overcome by
the addition of L-valine to the medium. The inhibitory
effects of a-aminobutyrate may be responsible for the in-

ability of E. sporogenes to grow in the minimal medium

 

without valine and leucine. a-Aminobutyrate inhibition may
be the result of reduced acetolactate production by the
cell. In E. ggli_the condensation of two pyruvate molecules
to form acetolactate (for valine and leucine) and the
condensation of a-ketobutyrate to form a-aceto-a-hydro-
xybutyrate (for isoleucine) are catalyzed by the same AHAS
isoenzymes. Consequently, increased concentrations of a-
ketobutyrate will competitively inhibit the formation of
acetolactate. In experiments with purified AHAS I from E.
Egli! Grimminger and Umbarger (14) showed that as the
pyruvate/a-ketobutyrate ratio in their assay was changed
from 10/1 to 1/1, the ratio of acetolactate/a-aceto-a-
hydroxybutyrate formed changed from 5.3/1 to l/l.5. When a-
aminobutyrate resistant mutants were selected from E. 99;;
K-12 they were found to have derepressed levels of AHAS.
Increasing enzyme concentration is one way to overcome
competitive inhibition. Amino acid tRNA-syntheses play a

role in the regulation of branched-chain amino acid synthesis

54

(16), and although a-aminobutyrate is activated by valyl-

1 and has

tRNA synthase (5), it is not transferred to tRNAva
no effect on the formation of the isoleucine-valine bio-
synthetic enzymes in E. ggli_(1l,31). Regardless of the
mechanism of inhibition, it seems likely that the same

system which allows E. sporogenes to make isoleucine from

 

propionate via a-ketobutyrate may also be indirectly
responsible for the cell's inability to make valine and
leucine, since the accumulation of a-aminobutyrate may

preclude growth without exogenous valine and leucine.

LITERATURE CITED
Allison, M. J., and M. P. Bryant. 1963. Biosynthesis
of branched-chain amino acids from branched-chain
fatty acids by rumen bacteria. Arch. Biochem.
Biophys. 101:269-277.
Allison, M. J., and I. M. Robinson. 1970. Biosynthesis
of a-ketoglutarate by the reductive carboxylation

of succinate in Bacteriodes ruminicola. J.

 

Bacteriol. 104:50-56.
Allison, M. J., I. M. Robinson, and A. L. Baetz. 1979.
Synthesis of a-ketoglutarate by reductive carboxy-

lation of succinate in Veillonella, Selenomonas

 

 

and Bacteroides species. J. Bacteriol. 140:980-

 

986.

Barker, H. A. 1981. Amino acid degradation by anaerobic
bacteria. Ann. Res. Biochem. 50:23-40.

Belly, R. T., S. Greenfield, and G. W. Claus. 1970.
B-Methylaspartate metabolism and isoleucine

synthesis in Acetobacter suboxydans. Bacteriol.

 

Proc. p. 141.

Buchanan, B. B. 1969. Role of ferredoxin in the
synthesis of a-ketobutyrate from propionyl coenzyme
A and carbon dioxide by enzymes from photosynthetic
and nonphotosynthetic bacteria. J. Biol. Chem.

244:4218-4223.

55

10.

ll.

12.

13.

56

Buchanan, B. B. 1972. Ferredoxin-linked carboxylation
reactions, p. 193-216. 12 P. Boyer (ed.), The
enzymes, vol VI, 3rd ed. Academic Press Inc., NY.

Buchanan, B. B., and D. I. Arnon. 1969. The reductive
carboxylic acid cycle, p. 170-181. £2.3- M.
Lowenstein (ed.), Methods in enzymology, vol XIII.
Academic Press Inc., NY.

Buchanan, B. B., and M.C.W. Evans. 1965. The synthesis
of a-ketoglutarate from succinate and carbon
dioxide by a subcellular preparation of a photo-
synthetic bacterium. Proc. Natl. Acad. Sci. USA
54:1212-1218.

Bush, R. S., and F. D. Saver. 1976. Enzymes of 2-oxo
acid degradation and biosynthesis in cell-free
extracts of mixed rumen microorganisms. Biochem.
J. 157:325-331.

Charon, N. W., R. C. Johnson, and D. Peterson. 1974.

Amino acid biosynthesis in the spirochete Leptospira:

 

evidence for a novel pathway of isoleucine synthesis.
J. Bacteriol. 117:203-211.

Costilow, R. N. 1977. Selenium requirement for growth
of Clostridium sporogenes with glycine as the
oxidant in Stickland reaction systems. J.

Bacteriol. 131:366-368.
Davis, B. D., and E. S. Mingioli. 1950. Mutants of

Escherichia coli requiring methionine or vitamin

 

B J. Bacteriol. 60:17-28.

12'

14.

15.

16.

17.

18.

19.

20.

57
Hoare, D. S., and J. Gibson. 1964. Photoassimilation

and biosynthesis of amino acids by Chlorobium

 

thiosulphatophilum. Biochem J. 91:546-559.

 

Iaccarino, M., J. Guardiola, M. DeFelice, and R. Fauro.
1978. Regulation of isoleucine and valine bio-
synthesis, p. 29-73. Eg Current topics in cellular
regulation, vol 14. Academic Press, Inc., NY.

Kisumi, M., M. Sugiura, and I. Chibata. 1976. Bio-
synthesis of norvaline, norleucine, and homo-

isoleucine in Serratia marcescens. J. Biochem.

 

80:333-339.

Leavitt, R., and H. E. Umbarger. 1960. Colorimetric
method for the estimation of growth in cap assays.
J. Bacteriol. 80:18-20.

Leavitt, R. I., and H. E. Umbarger. 1961. Isoleucine
and valine metabolism in Escherichia coli. X. The
enzymatic formation of acetohydroxybutyrate. J.
Biol. Chem. 236:2486-2491.

Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J.
Randall. 1951. Protein measurement with the
folin phenol reagent. J. Biol. Chem. 193:265-
275.

McBride, R. W., D. W. Jolly, B. M. Kadis, and T. E.
Nelson Jr. 1979. Rapid thin-layer chromatographic
separation of isoleucine, leucine and phenylalanine.

Chrom. 11:424-425.

21.

22.

23.

24.

25.

26.

27.

28.

58

Mead, G. C. 1971. The amino acid-fermenting Clostridia.
J. Gen. Microbiol. 67:47-56.

Moss, C. W., and J. L. Vester. 1967. Characterization
of Clostridia by gas chromatography. I. Differ-
entiation of species by cellular fatty acids.

Appl. Microbiol. 15:390-397.
Perkins, W. E., and K. Tsuji. 1962. Sporulation of

Clostridium botulinum. II. Effect of arginine and

 

its degradation products on Sporulation in a
synthetic medium. J. Bacteriol. 84:86-94.
Phares, E. F. 1951. Degradation of labeled propionic
and acetic acids. Anal. Chem. 23:168-174.
Phillips, A. T., J. I. Nuss, J. Mosic, and C. Foshay.
1972. Alternate pathway for isoleucine biosynthesis

in Escherichia coli. J. Bacteriol. 109:714-719.

 

Robinson, I. M., and M. J. Allison. 1969. Isoleucine
biosynthesis from 2-methy1butyric acid by anaerobic
bacteria from the rumen. J. Bacteriol. 97:1220-
1226.

Sauer, F. D., J. D. Erfle, and S. Mahadevan. 1975.

Amino acid biosynthesis in mixed rumen cultures.
Biochem. J. 150:357-372.

Schildkraut, I., and S. Green. 1973. Threonine synthase-
catalyzed conversion of phosphohomoserine to a-
ketobutyrate in Bacillus subtilis. J. Bacteriol.

115:777-785.

29.

30.

31.

32.

33.

59

St¢rmer, F. C., and H. E. Umbarger. 1964. The require-
ment for flavine adenine dinucleotide in the

formation of acetolactate by Salmonella typhimurium

 

extracts. Biochem. BiOphys. Res. Comm. 17:587-
592.

Strassman, M., A. J. Thomas, L. A. Locke, and S. Weinhouse.
1956. The biosynthesis of isoleucine. J. Amer.
Chem. Soc. 78:228-232.

Umbarger, H. E. 1973. Threonine deaminases. Adv.
Enzymol. 37:349-395.

Umbarger, H. E. 1978. Amino acid biosynthesis and its
regulation. Ann. Rev. Biochem. 47:533-606.

Whitely, H. R., and M. Tahara. 1966. Threonine
deaminase of Clostridium tetanomorphum. I. Purifi-
cation and properties. J. Biol. Chem. 241:4881-
4889.

Interconversion of Valine and Leucine by

Clostridium sporogenes
Daniel J. Monticello and Ralph N. Costilow
Department of Microbiology and Public Health

Michigan State University

East Lansing, Michigan 48824

Running head: Interconversion of Valine and Leucine

60

SECT ION I I I

ABSTRACT
Clostridium sporogenes has been found to require L-
leucine and L-valine for growth in a minimal medium,
although valine can be replaced by isobutyrate, and leucine
by isovalerate. Cells grown in minimal media were found to

14

incorporate significant C-label from [14C]valine into

leucine and 14

C-label from [14C]leucine into valine. These
results indicate that these bacteria can interconvert

leucine and valine.

61

62

It is well-established that many microbes can convert
L-valine to L-leucine (14). This is accomplished by the
elongation of a-ketoisovalerate (the keto-acid analog of
valine) to a-ketoisocaproate (the keto-acid analog of leucine).
This chain-elongation is similar in function to the elongation
system in the tricarboxylic acid cycle which converts oxalo-
acetate to a-ketoglutarate. Up to this point, there has
been no evidence for the conversion of L-leucine to L-
valine.

E. sporogenes (ATCC 7955, National Canners Association
PA 3679) was used in all experiments. It was maintained by
refrigeration of a culture that had sporulated in a medium
containing 4% trypticase, 2 mg of thiamine hydrochloride per
liter, and 0.05% sodium thioglycolate. Vegetative cultures
for inoculation of experimental media were initiated by
inoculating tubes of this same medium (10 ml/tube) with
0.1 ml of the sporulated culture. These were heat-shocked
(60°C for 10 min) and incubated at 37°C in an anaerobic
chamber (Coy Manufacturing Company, Ann Arbor, MI). After
growth for 6 to 8 h, tubes with 10 ml of the basal medium
containing 10 mM valine and 10 mM leucine were inoculated
with 0.1 m1 of this culture. When this second culture had
growth to late exponential phase, 0.05 ml was used to
inoculate experimental growth media. In the labeling
experiments, cells were preadapted to the appropriate medium
before addition of label. The basal medium used in these

experiments was a modification of the synthetic medium of

63

Perkins and Tsuji (11). It contained salts, vitamins, 0.05%
sodium thioglycolate and six amino acids - L-tyrosine (2 mM),
L-tryptophan (2 mM), L-phenylalanine (10 mM), L-methionine
(10 mM), L-arginine-HCl (20 mM) and L-serine (25 mM). For
growth experiments optical densities at 600 nm were measured
with a mini-spec 20 spectrophotometer (Bausch and Lomb
Optical Co., Rochester, NY).

E. §porggenes grew well in the basal medium supple-
mented with 10 mM valine and 10 mM leucine; 10 mM valine
and 10 mM a-methylbutyrate; 10 mM leucine and 10 mM
a-methylbutyrate or 10 mM leucine and 10 mM isobutyrate
(Table 1). These results encouraged us to attempt to grow
the bacteria without either valine or leucine. Some growth
occurred when 10 mM isobutyrate and 10 mM a-methylbutyrate,
were added to the medium (Table 2). Slightly more growth
was achieved by adding 10 mM isovalerate along with the
other two volatile acids. These results suggested that the
keto-acid analogues of leucine (a-ketoisocaproate) and

valine (a-ketoisovalerate) can be produced by E. gporogenes

 

from isovalerate and isobutyrate, respectively.

Individual lots of cells were grown for one generation
(m4 h) in the basal medium supplemented as follows: (a)
50 pCi L-[U-14C11eucine; 10 mM leucine, 10 mM isoleucine,
and 10 mM a-methylbutyrate; (b) 50 pCi L-[U14C]valine,

10 mM valine and 10 mM a-methylbutyrate; and (c) 10 pCi [U-

14 14

C]pyruvate or 10 pCi CO in the medium with 10 mM valine

2
and 10 mM leucine. Cells were harvested, extracted, and

64

Table 1. Maximum growth in basal mediuma supplemented with

amino acids and volatile fatty acids.

 

Concentrations (mM)

 

a-Methyl- Final

 

L-Valine L-Leucine Isovalerate Isobutyrate butyrate O.D.
0 0 0 0 0 0.039
10 0 10 0 0 0.081

10 0 0 10 0 0.13

10 0 0 0 10 0.95

0 10 10 0 0 0.086

0 10 0 10 0 0.34

0 10 0 0 10 0.75

10 10 0 0 0 0.92

 

aThe basal medium is described in the text. Triplicate 10 m1
cultures were grown at 37° in the anaerobic chamber, and the

optical density (O.D.) monitored at 2 h intervals.

65

Table 2. Maximum growth in basal medium supplemented with

volatile fatty acids.

 

’Concentrations (mM)

 

a-Methyl- Final

Isobutyrate~ Isovalerate butyrate ‘ O.D.
0 0 0 0.039

10 0 0 0.066

10 10 0 0.066

10 10 10 0.40

0 10 0 0.056

0 10 10 0.056

0 0 10 0.080

10 0 10 0.30

 

66

protein hydrolysates prepared as described elsewhere
(Monticello and Costilow, manuscript submitted). Aspartate,
glutamate, valine, and leucine were separated from the
protein hydrolysates and the specific radioactivity of each
amino acid determined. Whatman LK-S-D linear K preadsorbant
thin layer plates (Whatman, Inc., Clifton, NJ) developed
with a methy1ethylketone-pyridine-water-glacial acetic acid
(73:5:20:2) solvent system (9) were used to separate iso-
leucine, leucine, and valine. L-Aspartate and L-glutamate
were separated from other amino acids and each other by
paper electrophoresis on Whatman No. 1 paper in 0.25 M
sodium acetate buffer at pH 4.3. Samples (10 p1) were
spotted on the paper and electrophoresed at 17-35 volts/cm
for 90 min. Amino acids were located by comparison with
identical adjacent lanes sprayed with ninhydrin (0.1% in
acetone). The relative abundance of each of these amino

acids in the protein from E. sporogenes grown in the basal

 

medium is known (R. S. Hadioetomo, 1980 Ph.D. dissertation,
Michigan State University), so the specific activity
(cpm/nmole) of each amino acid could be estimated. Table 3
shows the results of these incorporation experiments.

A significant amount of label was incorporated into
valine when the cells were labeled with [14C1-1eucine, and
into leucine when the cells were labeled with [14C]-valine.
When amino acids are oxidized by E. sporogenes the carboxyl-
group is released into the medium, and other amino acids may

become labeled in this way. To determine the influence of

67

Table 3. Incorporation of 14C into aspartate, glutamate,

leucine, and valine.

 

cpm/10 pl Protein Hydrolysate Sp. Act. Ratios

 

 

 

 

Growth

Substrate Aspartate Glutamate Leucine Valine Leu/Asp Val/Asp

[14c]- 30 a 50 3200 160 140 6.6

Leucine (0.67) (0.98) (94) (4.4)

[14C]- 140 110 370 3400 3.7 30

Valine (3.0) (2.1) (11) (91)

[3-14c1- 2700 4400 20 o 0.01 o

Pyruvate (60) (86) (0.6) (0)

14CO2 3400 2800 580 440 0.23 0.16
(75) (54) (17) (12)

aEstimated specific activities (cpm/nmole).

68

such reactions on the incorporation of label from [14C]-

14

1eucine into valine, cells were labeled with CO and the

2
amino acids examined. The ratio of radioactivity in valine/

14

aspartate should remain roughly equal if the C-label in

the valine is from exogenous 14C02, or 14CO

[1

of valine/aspartate was 41 times larger with [l

as compared to 14CO

14

2 released from

4C]-leucine. In fact, the ratio of the specific activities

4C]-leucine

2 as substrate (Table 3). This rules out

CO2 as the principle source of radioactivity from [14C]-
leucine which was found in valine. The same reasoning
applies for the incorporation of [14C]-valine into leucine;
in this case, the ratio of specific activities in leucine/
aspartate was 16 times that obtained when the cells were

14

labeled with CO As found in many other experiments

2.
(Hadioetomo, 1980 Ph.D. dissertation, Michigan State Uni-

14C from [3-14C] pyruvate

versity) no significant amount of
was found in either leucine or valine.

Branched-chain amino acid synthesis and its regulation
in Escherichia coli, Salmonella gyphimurium, Neurospora

crassa, Saccharomyces cerevisiae and several other organisms

 

have been the objects of intense investigation (15). Little
work has been done in this area with the anaerobes, however.
The results presented here indicate that although E. sporogenes
usually requires L-valine and L-leucine for growth in a
minimal medium, the bacteria can interconvert leucine and

valine.

69

The production of isobutyrate, acetate, and isocaproate
from cultures grown with leucine has been observed in
several strains of E. gporogenes [NCIB 10696 (10), EC-9 (12)
and ATCC7956 (our unpublished results), and other Clostridia
(4,10,12)]. The oxidation of leucine to isobutyrate by E.

sporogenes has been shown to proceed via an unusual series

 

of reactions; a-leucine is converted to B-leucine, which is

oxidized to isobutyrate and acetate (12). It is likely that

14C-leucine is converted to l4C-isobutyrate in this fashion,
and the l4C-isobutyrate is reductively carboxylated to a-
l

[ 4C]-ketoisovalerate, the keto-acid precursor of valine.
The proposed pathway is shown in Fig. l.

a-Keto-acid synthase catalyzed reductive carboxylations
are relatively common among the anaerobes. The reductive

carboxylation of a-methylbutyrate to a-keto B-methylvalerate

has been demonstrated in E. gporogenes (R. S. Hadioetomo,

 

1980 Ph.D. dissertation, Michigan State University) and

Bacteriodes ruminicola (13). The reductive carboxylation of

 

isobutyrate for valine synthesis, and isovalerate for leucine
synthesis has been observed in extracts prepared from mixed
rumen bacteria (1) and photosynthetic bacteria (8). The
formation of a-ketobutyrate from propionate has been
demonstrated in cell-free preparations from Chromatium,
Clostridium pasteurianum, Desulfovibrio desulfuricans (5),
mixed rumen extracts (7) and Veillonella alcalescens (3),
which along with other rumen bacteria (2) and photosynthetic

bacteria (6) produces a-ketoglutarate from succinate.

.ocflam>

on ocflosoa mo coflmuo>coo may now hm3suom pomomoum .H .mam

 

 

MII
MII
IINNII INNN-IN-IL-MIN .m.
m5 0
wheem5<>oNIoemxua Ioouuwuzuumzu
.
MIN
> I I
N wh<._.mu< m._.<m .SmowH 1000 :w mzu Now
M MIN
INNN- IN
mzuuamuuu mznuzmuno
mh<omd<uow_0hmxlu sz N12
I . .
INNN-NIN-N-IN-MIN .A I\.. INNN-NIN-IN-Iw-MIN .o.....INNN-IN-NIN-IL-MIN
MIN M MIN MIN

:2

71

Reductive carboxylation reactions have only been observed in
anaerobic heterotrophs from the rumen and other anaerobic
habitats. The requirement for a very electronegative
electron donor such as reduced ferredoxin to overcome the
energy barrier for carboxylation essentially precludes this
activity from aerobically grown organisms.

In light of the apparent ubiquity of the keto-acid
synthase activity among the anaerobes, many of these bacteria
may be capable of interconverting leucine and valine. The
other key enzyme activity, leucine 2,3-amino mutase, which
converts a-leucine to B-leucine, has been found in several
Clostridia, in rats, sheep, rhesus and African green monkey
livers, and in human leukocytes (12). The overall roles of
such reactions in the branched-chain amino acid metabolism

by Clostridium §porogenes is not clear. Further experi-

 

mentation is required to determine this, and the role of a-

methylbutyrate in this process.

Literature Cited
Allison, M. J., and M. P. Bryant. 1963. Biosynthesis
of branched-chain amino acids from branched-chain fatty
acids by rumen bacteria. Arch. Biochem. Biophys.
101:269-277.
Allison, M. J., and I. M. Robinson. 1970. Biosynthesis
of a-ketoglutarate by the reductive carboxylation of
succinate in Bacteriodes ruminicola. J. Bacteriol.
104:50-56.
Allison, M. J., I. M. Robinson, and A. L. Baerz. 1979.
Synthesis of a-ketoglutarate by reductive carboxylation

of succinate in Veillonella, Selenomonas and Bacteroides

 

 

 

species. J. Bacteriol. 140:980-986.

Britz, M. L., and R. G. Wilkinson. 1981. Conversion
of leucine to isovaleric and isocaproic acids by broken
cell preparations of Clostridium bifermentans. Proc.
Aust. Biochem. Soc. 14:59.

Buchanan, B. B. 1969. Role of ferredoxin in the
synthesis of a-ketobutyrate from propionyl coenzyme A
and carbon dioxide by enzymes from photosynthetic and
nonphotosynthetic bacteria. J. Biol. Chem. 244:
4218-4223.

Buchanan, B. B., and M.C.W. Evans. 1965. The synthesis
of a-ketoglutarate from succinate and carbon dioxide by
a subcellular preparation of a photosynthetic bacterium.

Proc. Natl. Acad. Sci. U.S.A. 54:1212-1218.

72

10.

ll.

12.

l3.

14.

73

Bush, R. S., and F. D. Sauer. 1976. Enzymes of 2-OXO
acid degradation and biosynthesis in cell-free extracts
of mixed rumen microorganisms. Biochem. J. 157:325-
331.

Hoare, D. S., and J. Gibson. 1964. Photoassimilation

and biosynthesis of amino acids by Chlorobium thio-

 

sulphatophilum. Biochem. J. 91:546-559.

McBride, R. W., D. W. Jolly, B. M. Kadis, and T. E.
Nelson, Jr. 1979. Rapid thin-layer chromatographic
separation of isoleucine, leucine and phenylalanine.
Chrom. 11:424-425.

Mead, G. C. 1971. The amino acid-fermenting Clostridia.
J. Gen. Microbiol. 67:47-56.

Perkins, W. E., and K. Tsuji. 1962. Sporulation of

Clostridium botulinum. II. Effect of arginine and its

 

degradatiOn products on Sporulation in a synthetic
medium. J. Bacteriol. 84:86-94.

Poston, M. J. 1976. Leucine 2,3-aminomutase, an
enzyme of leucine catabolism. J. Biol. Chem. 251:
1859-1863.

Robinson, I. M., and M. J. Allison. 1969. Isoleucine
biosynthesis from 2-methylbutyric acid by anaerobic
bacteria from the rumen. J. Bacteriol. 97:1220-1226.
Umbarger, H. E. 1978. Amino acid biosynthesis and its

regulation. Ann. Rev. Biochem. 47:533-606.

SECTION IV

Incorporation of Branched-chain Amino Acids
into Cellular Lipid

 

Many gram positive bacteria produce branched-chain
fatty acids which are incorporated into cellular lipid. The
initial step in this synthesis is the production of branched-
chain acyl-CoA molecules derived from branched-chain amino
acids. A characterization of the cellular fatty acids from

E. sporogenes showned that these bacteria produce signifi-

 

cant amounts of 15:0 (from leucine), 16:0 (from valine)
branched-chain fatty acids, and perhaps some anti-iso 16:0
fatty acids (from isoleucine)* These fatty acids were pre-
sent in the cellular lipid from cells of various ages grown
in various media. Since the production of these fatty acid
entails a significant drain on the pools of isoleucine,
leucine and valine in the cell, the extent to which these
amino acids were incorporated into cellular lipid during
growth in a minimal medium was examined.

To determine the relative incorporation of valine,
leucine and isoleucine into cellular lipid, the incorpora-

14

tion of C-label from these amino acids into lipid was

determined. The incorporation of [3-14C] pyruvate, which
*Moss, C.W., and J.L. Vester. 1967. Characterization of
Clostridia by gas chromatography. I. Differentiation

of species by cellular fatty acids. Appl. Microbiol.
15:390-397.

74

75

should be incorporated to a large extent, and [U-14C] lysine,
which is not readily incorporated into lipid was also exam-.
ined. Cells were grown in the basal medium (section II,

this dissertation) with 10 mM leucine and 10 mM valine, and
labeled with [U-14C]isoleucine (10 pCi/10 ml), [U-14C]1ysine
(10 pCi/10 ml) or [3-14C] pyruvate (10 pCi/10 m1). For
labeling with [U-14C]leucine, cells were grown in the basal
medium with 10 mM valine and 10 mM a-methylbutyrate, and

14C] valine label-

with 10 pCi/10 ml [U-l4c11eucine. For [U-
ing, cells were grown with 10 mM leucine and 10 mM isobu-
tyrate. The cells were grown with the label for four hours
(about 1 generation) and then harvested and fractionated

as described elsewhere (section I, this dissertation). After
extraction of the cellular material at 90°C for 30 min, the
material was centrifuged. The supernatent contained cellu-
lar lipid and nucleic acids. Lipid was extracted for this
solution with and equal volume of chloroform:methanol (1:1).
The misture was shaken for 30 min, left overnight at 4°C,

and the lipid fraction drawn off, dried, and counted in a
toluene based scintillation cocktail (section I).

14

Table 1 shows that significant amounts of C-label

-14 14C]valine

from [U-14C]isoleucine, [U C]leucine and [U-
were incorporated into cellular lipid. The ratios of the
cpm in lipid/protein indicate that approximately six times
more valine than leucine or isoleucine was incorporated into

cellular lipid.

76

14

Table 1. The incorporation of C-label into cellular

lipid and protein.

 

 

 

a LIPID
SUBSTRATE LIPID PROTEIN PROTEIN
[U-14C]isoleucine 42,000 15,000,000 0.0028
[U-14C]leucine 64,000 30,000,000 0.0021
[U-l4c1valine 13,000 840,000 0.015
[3-14C1PYruvate 180,000 4,600,000 0.039
[U-14C]1ysine 590 180,000 0.0032
a 10 pCi of 14C-labeled substrate was added to a 10 ml cul-

ture (optical density = 0.3). The cells were grown for
4h (about 1 generation), harvested, and the total radio-
activity in cellular lipid and protein determined.

Oxidation of Branched-chain Amino Acids by Whole Cells.

One mechanism utilized by E. sporogenes for ATP genera-

 

tion is the coupled oxidation and reduction of amino acid
pairs (the Stickland reaction). An attempt was made to
evaluate the contribution of these oxidations to the overall
disposition of the branched-chain amino acids in cells grown
in various minimal media. In some media, this drain on the
amino acid pools may play an important part in diminishing
the isoleucine, valine and leucine levels in the cell.

Cells were grown in the basal medium described in
Section II, with 10 mM leucine and 10 mM valine (designated
medium LV in figures 1-3); 10 mM leucine and 10 mM a-methyl-
butyrate (medium VMB); 10 mM isoleucine, 10 mM leucine and
10 mM valine (ILV) or the standard Trypticase medium (STM).
50 ml cultures of late exponential phase cells were har-
vested, washed with buffer (50 mM potassium phosphate, pH 7.5),
and suspended in 20 ml of the same buffer with 50 mM L-proline
and 25 mM L-isoleucine, L-leucine or L-valine. The suspen-
sions were incubated at 37° in an anaerobic chamber. At
15 min intervals, 0.20 ml samples were removed and 0.05
ml of 41 mM n-valerate (final concentration 1 mM) added to
them as an internal standard. Samples without n-valerate
showed that no valerate was produced in these experiments.
After acidification with 9 N H2804, the volatile fatty acids
were extracted with 1 ml of ether, and the preparation

frozen overnight.

77

78

Volatile fatty acid concentrations were determined by
comparison with standards after separation by gas-liquid
chromatography. Separations were effected on a 15% SP-1220/
1% H

PO on 100/120 Chromosorb W AW, 6ft. x 4mm ID glass

3 4
column, column temp. 145°C, flow rate 70ml/min. Volatile
fatty acids were detected by flame ionization with an F&M
Research Chromatogram.

The results of these E3 33359 oxidations of isoleu-
cine, leucine and valine are shown in Fig. 1. Regardless
of the medium the cells were grown in , they retained the
capacity to oxidize all three amino acids to some extent.
This suggests the presence of a relatively nonspecific
amino acid oxidation system in these bacteria. Fig. l-A
shows that cells grown in STM could oxidize isoleucine sig-
nificantly better than those grown in the other media.
Since the complex STM medium contains considerable isoleucine,
this may indicate the presence of an inducible isoleucine
degradation system in the bacteria. A similar result was
not observed for leucine and valine oxidation. However,
cells grown in the VMB medium oxidized both valine (Fig. l-B)
and leucine (Fig. l-C) better than cells grown in any other
medium, including STM. The role of a-methylbutyrate in stimu-

lating the valine and leucine oxidation system is not clear.

Figure 1.

79

The oxidation of L-isoleucine (A),
L-valine (b) and L-leucine (C) to
a-methylbutyrate, isobutyrate or
isovalerate. Cells were harvested
from basal media supplemented as
described in the text, and resus-
pended in buffer with 50 mm proline
and 25 mm L-isoleucine, L-valine or
L-leucine.

MOLEs / NO PROTEIN

IINOLEs / HG PROTEIN

IIHOLES / HG PROTEIN

80

 

 

9 . ISOLEucINE —o d-NETNYLwTYuTE
8 P o—o STM
7 >-
6 .-
5 .— °\o/o
‘l I
3 K /.\O—O ILV
" O /.
A LV
2 - 55.44 LIB
EXP—:0: L l L
0 15 30 ‘15 60 75 90 105 120
TINE (NIN)
‘1 )-
VALINE —0 lSOBUTYRATE D ‘DVMB
3 ..
8

1 . /°7‘

15/:/" I /A t)”

L/

 

 

75 105 1210
TINE (mu)
8 .
l
7 ..
6 U——I:I VHB

- LUEcINE —-0 lsowILEPATE /
5 ..

\
ll
.\

 

04>Afifi/‘/ :5 38

0 15 30 45 60 75 90 105 120

TINE (NIN)

Effect of a-Methylbutyrate on Total Growth

It has been shown that E. sporogenes grows well in

 

the basal medium with 10 mM valine and either 10 mM leucine
or 10 mM a-methylbutyrate (R.S. Hadioentomo, 1980, Ph.D.
dissertation, Michigan State University). To further exam-
ine the role of a-methylbutyrate, cells were grown in the
basal medium (section III) with 10 mM valine and increas-

ing concentrations of either leucine, isoleucine or a-methyl-
butyrate (Figure 2.). The cells grew significantly better
with added a-methylbutyrate than they did with added isoleu-
cine. When the a-methylbutyrate was analyzed by separation
by gas chromatography, it was found to be contaminated with
approximately 10% isovalerate (Figure 3). Two different lots

of this product (Eastman Co.) were similarly contaminated.

81

FINAL OPTICAL DENSITY (600 NM)

Figure 2.

82

1.2

1.0-
0* 'METHYLBUTYRATE A 87
_. :::::::;;;,’15:UCINE

0'8

0.6- / o

0.4 _ / /
1 o .
.A
002b// ‘/‘——-—_‘/ 7
IA-’”"
‘,II.

IsOLEUCINE

 

I J I I; l
2.0 4.0 6.0 8.0 10

CONCENTRATION (NM)

The effect of a-methylbutyrate on the final op-
tical density (total growth) of cultures in 3
different media. All of the media include the
basal medium and 10 mM L-valine, with a-methyl-
butyrate, L-leucine or L-isoleucine.

Figure 3.

83

 

 

 

 

 

3
AN .

L_
_T ‘1' l

Separation of a-methylbutyrate and isovalerate
by gas chromatography. Shown above is the re-
corder printout following the injection of a

"10 mm a-methylbutyrate" sample. Peak 1 is the
solvent front, peak 2 a-methylbutyrate, and peak
3 is isovalerate. Column and Conditions; Varian
model 2400 equipped with flame ionization de-
tector, 6ft x 2 mm glass column packed with car-
bopak c coated with 0.3% carbowax and 0.1% H3PO4.
Injector temp - 175°, Detector temp. - 175°,
oven temp. - 135°, H2 carrier at 30 cc/min,
sample size 1 pl.

SECTION V

SUMMARY

The results presented illustrate the complexity of

branched-chain amino acid metabolism in Clostridium sporo-

 

genes. This investigation was initiated by the observa-

tion that E. sporogenes required valine and leucine for

 

growth in a minimal medium, but not isoleucine.

There is considerable evidence (section II) that iso-
leucine is made by the classical pathway (4) using propion-
ate as a precursor of a-ketobutyrate, or that isoleucine can
be produced directly from a-methylbutyrate. In an isotOpe
competition experiment, exogenous a-methylbutyrate was shown

to decrease the incorporation of [14

l4

C]pyruvate into isoleu-
cine, while having no effect on CO2 incorporation. Since
L-threonine is not a major source of isoleucine, cells

were labeled with a-amino-[3-14C]butyrate to determine if
a-ketobutyrate was produced from some other source. The

14C-label, and the demonstra-

incorporation of considerable
tion of AHAS activity (the first enzyme in the classical
apthway) suggested that a-ketobutyrate was a precursor

of isoleucine. Further labeling experiments showed that

[2_14

C]propionate was also a precursor of isoleucine, and
an isotope competition experiment demonstrated that exo-

genous propionate reduced the incorporation of

84

85

[14

C]serine into isoleucine. These experiments prompted

14

the labeling of isoleucine with [l- C]propionate, [2-14C1-

propionate and [3-14C]pyruvate and subsequent sequential de-

14C-label from [1-14

gradation. It was found that the C]-
propionate was in the C-2 of isoleucine, [2-14C]propionate
was in the carbon 3-5 fragment, and [3-14C]pyruvate in the
carbon 3-5 fragment and C-6. These results support the con-
clusion that propionate is reductively carboxylated to
a-ketobutyrate in the cell, which is then converted to iso-
leucine via the classical pathway, and that a-methylbutyrate
is an alternate source of isoleucine.

a-Amino-[14

C]butyrate was found to accumulate in the
medium following labeling with [14C1propionate. Since
a-aminobutyrate is a potent inhibitor of valine synthesis

in E. anthracis (1) and E. coli (3), this may explain E.

 

sporogenes' inability to grow in the minimal medium without

 

exogenous valine and leucine.
Further investigations of the reductive carboxylation

capabilities of E. spgrogenes showed that these bacteria

 

could make valine from isobutyrate and probably leucine
from isovalerate. The results presented in section III

indicate that E. §porogenes, like some other bacteria (4),

 

is capable of the conversion of valine to leucine via the
chain elongation pathway. This organism probably converts
leucine to valine by first forming B-leucine using a leucine
2,3-aminomutase enzyme, which is then oxidized to acetate
and isobutyrate (2). The isobutyrate can then be reduc-

tively carboxylated to the keto-acid precursor of valine.

86

The branched-chain amino acid pools in E. sporogenes

 

were shown to be subject to several unusual activities not
observed in the enteric bacteria (4). Isoleucine, valine
and leucine are subject to non-specific oxidation, presum-
ably by the cellular machinery for ATP synthesis via the
coupled oxidation/reduction of amino acid paris (the

14C-Labeled branched-chain amino acids

Stickland reaction).
were also incorporated into cellular lipid, especially va-
line. Branched-chain fatty acids have been shown to re-

present a considerable portion of the fatty acids in E.

sporogenes lipid.

 

It is clear that the substitution of a-methylbutyrate
for leucine in the minimal medium is possible because of
the considerable amounts of isovalerate in the a-methyl-
butyrate preparation. Media which were believed to contain
10 mM a-methylbutyrate actually contained almost 1 mM iso-
valerate. Reductive carboxylation of this fatty acid yields
the keto-acid precursor of leucine.

The results presented here illustrate the unique
strategies and mechanisms employed by anaerobic bacteria to
survive in their highly reduced environment. The reductive
carboxylations observed here for the production of iso-
leucine from propionate or a-methylbutyrate, valine from
isobutyrate and leucine from isovalerate can only proceed
under such reduced conditions. These reactions and their
regulation represent a unique system for branched-chain

amino acid synthesis and utilization in Clostridium

 

Sporogenes.

 

LITERATURE C ITED

Gladstone, G. P. 1939. Inter-relationships between
amino-acids in the nutrition of B. anthracis. Brit.

J. Exp. Pathol. 20:189-200.

Poston, M. J. 1976. Leucine 2,3-aminomutase, an en-
zyme of leucine catabolism. J. Biol. Chem. 251:1859-
1863.

Ramakrishnan, T. and E. A. Adelberg. 1965. Regulatory
mechanisms in the biosynthesis of isoleucine and valine.
II Identification of two operator genes. J. Bacteriol.
89:654-660.

Umbarger, H. E. 1978. Amino acid biosynthesis and its

regulation. Ann. Rav. Biochem. 47:533-606.

87