Molec. gem Genet. 156, 17-25 (1977) © by Springer-Verlag 1977

The Acetolactate Synthase Isoenzymes of Escherichia coli K-12 John Guardiola, Maurilio De Felice, Alessandro Lamberti, and Maurizio Iaccarino International Institute of Genetics and Biophysics, C.N.R., Via Marconi 10, 80125 Naples, Italy

Summary. Strains of Escherichia coli K-12 possessing only one of the three genes coding for acetolactate synthetase activity present either in the wild type or in its ilv0603 derivative were prepared and analyzed. Extracts prepared from these strains show different values of acetolactate synthase specific activity and different sensitivity to valine inhibition. These strains show a unique pattern of growth inhibition by different substances. Temperature sensitive (ts) mutations in the ilvB and ilvG genes, have been isolated a n d characterized. Extracts of these strains were found to have an acetolactate synthase activity more heat labile than that of a strain containing the corresponding wild type allele. We conclude that ilvB and ilvG are the structural genes for two different forms of acetolactate synthase activity, most likely two isoenzymes. Moreover, since the strains containing a ts mutation show a temperature sensitive auxotrophy for isoleucine and valine, these two acetolactate synthases participate in isoleucine and valine biosynthesis. Similar evidence for a third acetolactate synthase, the product of the ilvHI genes, has been reported previously. We propose the following names for the acetolactare synthase isoenzymes: acetolactate synthase I (AHAS I), the product of the ilvB gene; acetolactate synthase II (AHAS II), the product of ilvG gene; and acetolactate synthase III (AHAS III), the product of the ilvHI genes.

Introduction In Escherichia coli K-12 and in other related Enterobacteriaceae the biosynthesis of isoleucine and valine requires a carbon-to-carbon condensation step For offprints contact." Maurizio Iaccarino

performed by an acetolactate synthase activity (Umbarger, 1969; Umbarger and Davis, 1962). This condensation is the first step in valine biosynthesis and the second in isoleucine biosynthesis (Fig. 1). We have reported (Guardiola, De Felice and Iaccarino, 1974) the isolation of a strain of E. coli K-12 requiring isoleucine and valine for growth because of the absence of acetolactate synthase activity. The analysis of this strain shows that the ilvG, ilvB, and ilvI genes are required for expression of this activity and suggests the presence of three isoenzymes. The ilvI gene is the structural gene for one of these isoenzymes (De Felice et al., 1974a, b). In this paper, we present evidence that ilvG and ilvB are structural genes for the other isoenzymes. We also report some properties of strains with one isoenzyme.

Materials and Methods Reagents and Media. Reagents were of the highest purity commercially available. Phosphomycin, disodium salt, was obtained from Merck Sharp and Dohme, N.J. D, L-c~,e-diaminopimelicacid (mixture of LL, DD and meso isomers) and cycloleucine were bought from Sigma Chemical Co., St. Louis Mo. The minimal medium used was minimal citrate (Vogel and Bonner, 1956). Usual supplements, when required, were glucoseor other carbon sources (0.4%), L-tryptophan (25 gg/ml), L-arginine or L-valine (100 gg/ml), other L-amino acids (50 gg/ml) and thiamine (10 gg/ml). TY broth is 0.8% NaC1, 1% Bacto tryptone (Difco) and 0.5% yeast extract (Difco). Bacterial Strains. The strains of E. coli K-12 used are listed in Table 1. Figure 2 shows the genealogy of some of the strains used. The wild type E. coli K-12 does not grow on ]~-glucosides (arbutin or salicin) as a sole carbon source (genotype bgl+). Mutants that utilize arbutin can be selected; some of them (bglR) are located near the ilv genes and cotransduce with them between 30 and 70% by P1 transduction. Mutants altered in the cya gene were selected, following a suggestion of M.D. Alper and B.N. Ames, among those resistant to the growth inhibitory action of phosphomycin. Strains carrying

18

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

t "'', '//+--t, ara

ilv

leu

azi

CH3 CH,-?H

CH 3

CH,-C-CH,OH

"-,,../17 CH,

C,H

CH,

+ pyruvmt. Cm"O CH3"C-OH CH3 -C-COOH~ CH3-C-COOH = HCOH O OH NA~PH N~AP COOH pyruvate t f f " - ace(to~ (x, ,~ - dihydr0xy-

Ill

C'-'O CH3.CH2_C.CO0H +pyruvmt~ CH_CH2_~.COOH~

OK - k e t o b u t y r a t e

- c02

OH

NADPH

CH3-?H ~ ~'~~- HCNH2 gtut. o(wg COOH q valine VALINE

DEHYDRASE

TRANSAMINASE

I

I I

%

CH,

CH3-?H >C=O C:OOH o(- ke,o-

I

15OMEROREDUCTASE I

L- threonln°

0

q

I

ACETOLACTATE CH~-CH-CH-COOH SYNTHASE , , OH NH 2 ISOENZYMES

Ill L-THREONINE Ill

_Hz0

I I

,SOLEUC,NE

TRANSAMINASE

I

CH3 CH3 ?H, CH3-CHz'C'OH -H~O ~ CH3°CHz"1CH ~-'-'-""~" CH3-CH2",CH NADP HCOH C=O glut= a( kg H CNH2

o/. - a c e t o - oC - h y d r o x y -

butyrate

o(~.~,8-

I

I

COOH

COOH

dihydroxy -

methytvmterat°

o( - keto~

COOH isoleucme

- rnethytvaterate

Fig. 1. The biosynthetic pathway for isoleucine, leucine and valine in E. eoli K-t2 and the genes required for it. The rbs-ilv-cya cluster is located at 83 rain on the standard map (Bachmann, Low and Taylor, 1976) while the ara-leu-ilv-azi cluster is located at 1 rain. The ilvJ gene and other modifications of the standard map are reported elsewhere (Guardiola, accompanying manuscript). We show in this paper that the ilvB and ilvG genes code for a VaP and a Val r acetolactate synthase activity, respectively. We favour the idea that these are two isoenzymes (AHAS I and AHAS II) while there is no evidence, in the literature, for the hypothesis expressed by Bachmann, Low and Taylor (1976) that they are different subunits of the same isoenzyme

cya mutations grow poorly in minimal medium supplemented with glucose, and do not grow on several carbon sources (e. g. maltose, lactose, ribose, arbutin) unless 0.1 ml of 0.15 M cyclic 3',5'adenosine monophosphate is added to the agar plate. Strain MI262h was prepared by treating strain MI262c with PI phage grown on strain AB3590 and selecting cya + transductants. 99% of these (95/96) were Ilv-; 28% (27/96) grew on c(,fl-dihydroxy-isovalerate and isoleucine while 71% (68/96) did not grow and were therefore classified as containing the ilvDAC115 mutation. One of these was purified and named strain MI262h. The ilvI620 mutation of strain MI265 was induced by ultraviolet light in strain MI263. Following penicillin counterselection 10 Ilv- mutants were obtained and, after purification, cultures grown in minimal medium were analyzed qualitatively for acetolactate synthase as previously described (Guardiola, De Felice and Iaccarino, 1974). One of these cultures was found negative; P1 phage was grown on it and used to transduce MI263e with selection of Leu + transductants, of which 88% (84/96) were Ilv-, thus showing that the mutation is in ilvL Strain MI265 contains almost no acetolactate synthase activity (0.4 nmoles/min/mg) (see Table 2 for the value of strain MI263).

Strain MI269a is a bgl derivative of strain MI269 prepared by transduction with P1 phage grown on a bgl derivative of strain MI262b. The phenotype of the bgl mutation was found to cotransduce 60% (28/48) with ilvD. Strain MI269 is an Ilv- derivative of strain MI263e (see Table 1) obtained by UV mutagenesis and penicillin counterselection. It grows both on isoleucine and valine and isoleucine and c~-ketoisovalerate, but not on isoleucine and c(, ]~-dihydroxy-isovalerate (the substrate of dehydrase, the ilvD gene product). On this basis we labeled the mutation causing the Ilv- phenotype ilvD642. Extracts of this strain were found to have normal threonine deaminase specific activity (60 nmoles/min/mg protein) and normal acetolactate synthase (15 nmoles/min/mg protein). Transduetions. Transductions were performed with P1CMclrlO0 as described previously (Guardiola, De Felice, Klopotowski and Iaccarino, 1974). Most of the strains with acetolactate synthase defects show a cell envelope impairment, do not efficiently propagate P1 phage and show a low transduction efficiency (De Felice et al., accompanying manuscript). We find that addition of 50 ~tg/ml of diaminopimelic acid to rich medium is essential for the

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

19

Table 1. Bacterial strain

Strain

Genotype and Origin ilvO

ilvG

ilvB

ilvH

ilvl

AB3590

+

+

+

+

+

F - , thi-1, malA1, mtl-1, str-8 or 9, his-4, trpC, lacZ13, tsx-3, ilvDACll5. From the Coli Genetic Stock Center; see Favre et aL (1976), for further details

AT739

+

+

+

+

+

HfrH, su-, (2-); thr-lO, car-53, thi-1. From A. Taylor

AW206

603

+

+

+

+

HfrH, su-, 0 7 ) ; thr-lO, car-53, thi-1. (Favre et al., 1976)

MI2b

+

+

+

+

+

F - , argH, trpA36, ilvA601, cya. This laboratory: eya of MI2 (Favre et al., 1976)

MI149h

+

+

+

+

+

HfrH, su-, (2-); thi-1. This laboratory: Thr ~, Pyr + transductants of strain AT739

MI159

+

+

+

+

+

HfrH, su-, (2-); thr-lO, car-53, thi-1, ilvC608. (Favre et al., 1976)

MI167

+

+

+

612

+

HfrH, su-, (2-); thi-1, ara. (De Felice, Guardiola, Malorni, Klopotowski and Iaccarino, 1974)

MI226

603

605

+

+

+

HfrH, su , (2-); thr-lO, car-53, thi-1. (Favre et al., 1976)

MI244a

603

605

+

613

+

HfrH, su , ( 2 ) ; thi-1, ara. Thr +, Ara- transductants of MI244 (De Felice, Guardiola, Esposito and Iaccarino, 1974)

MI253

+

+

+

612

614

HfrH, su-, ( 2 ) ; thi-1, thr-lO, ara. (De Felice, Guardiola, Esposito and Iaccarino, 1974)

MI253c

+

+

+

612

614

HfrH, s u - , ( 2 ) ; thi-1, aro. (De Felice, Guardiola, Esposito and Iaccarino, I974)

MI261

603

605

+

612

614

HfrH, su , (2-); leu-& thi-1. This laboratory (Guardiola, De Felice and Iaccarino, 1974; De Felice et al., accompanying manuscript)

MI261c

603

605

+

612

614

HfrH, su-, ( 2 ) ; thi-1, glyA, ara, bgl. This paper

MI262

603

605

619

612

614

HfrH, su-, ( 2 ) ; thi-1, leu-8. (Guardiola, De Felice and Iaccarino, 1974)

MI262b

603

605

619

612

614

HfrH, su-, (2-); thi-1, glyA, ara. Leu ÷ transductants of a glyA derivative of MI262. (Guardiola, De Felice and Iaccarino, 1974)

MI262c

603

605

619

612

614

HfrH, su , ( 2 ) ; thi-1, glyA, ara, cya. eya derivative of MI262b

MI262e

603

+

619

612

614

HfrH, su , (2-); thi-1, glyA, ara. This paper

MI262h

+

+

+

612

614

HfrH, su-, (2-); ilvDACll5, ara. This laboratory; see Materials and Methods

MI263

603

605

619

+

+

HfrH, su , ( 2 ) ; thi-1. (Guardiola, De Felice and Iaccariuo, 1974; De Felice, Guardiola, Esposito and Iaccarino, 1974)

MI263e

603

605

619

+

+

HfrH, su , ( 2 ) ; thi-1, glyA, leu-& This laboratory; Ara +, leu-8, ilvH + , ilvI +, transductant of strain MI262b

MI265

603

605

619

+

620

HfrH, su-, ( 2 ) ; thi-1. From MI263; see Materials and Methods

MI266a

603

622

619

+

620

HfrH, su , ( 2 ) ; thi-1. This paper

619

612

614

HfrH, s u - , (;~-); thi-1, glyA, ara. This paper

623

612

614

HfrU, su , ( 2 ) ; thi-t, ara, glyA. This paper

612

614

HfrH, su-, ()o-); thi-1, glyA, ara. This paper

+

+

HfrH, su , ( 2 ) ; ilvD642, thi-1, glyA, leu-8, bgL This paper; see Materials and Methods

derivative

(ts) MI266b

603

622

(ts) MI267

603

605

(ts) MI269

603

605

623

(ts) MI269a

603

605

619

20

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

AT739 eOLI$

~transduclion AW206(ilv0603)

M I 167(ilvH612)

UV mutagenesis

Imutagenesis UV

/,ivo6o~

/i1~.612~ MI253 WIv16141

M I 2 2 6 lilvGGO5j

I transduction /.v o6o~\

/

MIz~4~ [il~Geosj

/

/

as nanomoles of product formed per minute per milligram of protein. Bacteria grown under the desired conditions were harvested by centrifugation at 4 ° C and then frozen as a pellet at -20 ° C. To 100-150 mg of frozen cell pellet 1 ml of extraction solution was added and the suspension was mixed in a Vortex mixer. The suspension was treated twice for 45 sec, with 1 min intervals for cooling, with an MSE sonic oscillator and then it was centrifuged for 20 rains at 15,000 rev/min in a Sorvall SS-34 rotor. The extracts contained between 5 and 15 mg of protein per ml. Proteins were determined by the method of Groves, Davis and Sells (1968) with crystalline bovine plasma albumin as a standard. Different extraction solutions were used. For routine assay of acetolactate synthase specific activity (as the experiments of Table 2) 0.1 M potassium phosphate, pH 8.0 was used. For the experiment of Figure 5 and of Figure 6, different extraction solutions, reported in the figure legend, were used.

Results

MI261 (il~W~ I I

\itvB

/

Strains with One Acetolactate Synthase

IUuVtagenesis "

~'~transduetion MI262e {i lv G÷)

ilvG605

~ltransductiOn~'~tran sd uc t i on MI 261c MI263 ( ilv B ÷)

( itv l ÷)

Fig. 2. Genealogy of some of the strains used in this paper. For further details see Table 1. For the sake of simplicity in some cases isogenic strains differing only because of a P1 transduction have not been considered

reproducible preparation of P1 lysates. Addition of 50 gg/ml of diaminopimelic acid to the transduction mixture and to the transduction plates is also essential to obtain a normal transduction efficiency.

Growth Stimulation or Inhibition Tests. A 0.1 ml sample of a cell suspension grown overnight in minimal citrate medium was diluted in 3 ml of 0.7% minimal citrate agar and layered on supplemented minimal citrate plates and 10 laliters of the solution to be tested for stimulation or inhibition were pipetted onto a small circle (6 m m in diameter) of W h a t m a n 3 M M paper applied on the agar surface. The result was observed after overnight incubation at 30 ° C. Values reported in the text (Table 3) represent the radius (in ram) of the inhibition or stimulation zone minus 3 m m (the radius of the paper circle). Isolation of Ilv-ts Revertants. A bacterial suspension was diluted to about 100 cells per ml in a minimal medium containing all the supplements needed for growth. The suspension was divided into 1-ml portions which were grown overnight. F r o m each tube 0.1 ml was spread on minimal plates containing no isoleucine and valine. F r o m each plate only one colony was purified and retained for further characterization. Enzyme Assays and Preparation of Cell Extracts. Acetolactate synthase was assayed by determining the rate of acetolactate formation (Stormer and Umbarger, 1964). Specific activities are expressed

We prepared strains possessing one acetolactate synthase isoenzyme in order to investigate the role of the three isoenzymes in cellular physiology. We have previously shown that strain MI262 is Ilv- and devoid of acetolactate synthase activity (Guardiola et al., 1974). This phenotype requires the combined presence of the ilvB619, ilvI614 and ilvG605 mutations. Strain MI262 also contains the ilvH612 and ilv0603 mutations which were essential for the isolation of the ilvI and ilvG mutations respectively. Thus, the genetic evidence for our hypothesis that E. coli K-12 contains the structural genes for three acetolactate synthase isoenzymes is derived from the sequence of mutagenesis and recombination experiments outlined in Figure 2. The ilv0603 mutation is necessary for the expression of the ilvG gene whose gene product is an acetolactate synthase which is not inhibited by valine (see below, Favre et al., 1976; De Felice Squires et al., accompanying manuscript). The ilvG605 mutation appeared among Val s mutant strains derived from the ilv0603 strain. A strain possessing an ilvG + allele was prepared by treating strain MI262b with P1 phage grown on strain MI2b and selecting [lv + Val r Cya + transductants. One of these transductants was purified by single colony isolation and given the designation MI262e. We verified the genotype ilv0603 ilvG ÷ (rather than ilvH612 ilvI +) by a backcross. We treated MI2b with P1 grown on strain MI262e selecting Ile + transductants. We found that 100% (96/96) of these transductants were Val r and 75% (72/96) were Cya +. We know that strain MI262e contains the ilvB619 allele of its progenitor strain, because this ilvB allele was recovered from this strain (De Felice Squires et al., accompanying manuscript).

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

We prepared an ilvB ÷ strain by crossing a bgl derivative of MI262 with P1 grown on MI261 (see Fig. 2). We selected Ilv ~ transductants and found 100% of these to be Val s and 50% bgl + (donor allele). One of the Ilv ~ transductants was retained as an ilvB ÷ strain (MI261c). Since the donor used to produce strain MI261c was strain MI26t which was also the progenitor of strain MI262 (see Fig. 2), this backcross suggests that strain MI262 (ilvB619, Ilv-) and strain MI261 (ilvB +, Ilv +) differ only in the ilvB allele (see Table 1 for complete genotypes). Thus, strains MI261 and the newly constructed MI261c should have identical phenotypes. We have assayed the acetolactate synthase activities of these two strains on several occasions, found no difference and used them interchangeably. We have already described the preparation of strains containing different ilvH and/or ilvI alleles (De Felice et al., 1974a, b). The ilvH and ilvI genes code for an acetolactate synthase activity which is sensitive to valine inhibition, but becomes resistant as a consequence of a mutation in the ilvH gene. The ilvI614 mutation was isolated among the VaP mutant strains derived from an ilvH612 containing strain. These strains (Table 1 and Fig. 2) show an Ilv + VaP phenotype when they contain ilvH612 and ilvI + and an Ilv-ts, Val r (at the permissive temperature) phenotype when they contain ilvH612 and ilvI615 (ts). In Table 2 we show the doubling time and acetolactate synthase specific activity of isogenic strains possessing one isoenzyme as compared to the wild type and the triple mutant. The strains have been chosen so that the only nutritional requirements are either thiamine or thiamine and glycine; both of these substances were present in the cultures of all strains. Two strains carrying an ilvB-- allele were used; one of them, strain MI253c, is also ilvG +, but we assume that it is equivalent to a strain containing ilvB + only, as we have evidence that ilvG is not expressed in an ilvO + strain (De Felice Squires et al., accompanying manuscript); the other strain carries also the ilv0603 and ilvG605 mutations and therefore the ilvG gene product is inactive. The growth rate of these strains was measured at 31 ° C as we have observed that in liquid culture above 35°C some of these strains lyse because of a cell envelope defect (De Felice et al., accompanying manuscript). As shown in Table 2, these strains have a normal doubling time in rich medium. In minimal medium containing an excess of isoleucine, leucine and valine all strains grow as well as the parental except strain MI261c. This strain shows a normal growth rate if the inoculum was grown in minimal medium with the same supplements. When the culture was inocu-

21 Table 2. Strains possessing a single acetolactate synthase isoenzyme a. Doubling times in different media and acetolactate synthase specific activity Strain

Doubling times b rich

minimal corepres-

minimal

Acetolactate synthase specific activity d

sors

MI149h (wild type)

52

85

90

112 (15%)

MI253c (ilvB + , ilvO +)

50

85

90

85 (8%)

MI261c (ilvB ÷ , ilvO603)

54

85 160 c

~240

MI262e (ilvG +)

55

90

~240

MI263 (ilvI +)

50

87

93

MI262 (triple mutant)

50

90

> 600

100 (10%) 6 (104%) 32 (22%) ~ 0.3 -

a See Table 1 for the complete genotype of each strain b Cultures were grown at 31 o C. Values are in min. Rich medium is T Y medium-containing 0.4% glucose. Minimal m e d i u m contains thiamine and glycine. Minimal-corepressors contains also isoleucine, leucine and valine c This value was obtained when the starting inoculum was pregrown in minimal corepressors. A higher value, 160, was obtained when the starting inoculum was in minimal (see text) d Values in parenthesis are for the activity in the presence of 1.5 m M valine

lated with 1 x 10 s bacteria pregrown in minimal medium, the doubling time is initially 160 rain, and later approaches 85 rain. This result shows that the metabolism of this strain is unbalanced when grown in minimal medium. Strains MI261c and MI262e are slow growers in minimal medium and this appears to be correlated to the low acetolactate synthase specific activity of these strains. In Table 2, we also show the acetolactate synthase specific activity and its sensitivity to valine in extracts of the strains possessing only one of the ilvB, ilvG or ilvI genes as compared to that of the wild type. The data demonstrate that the activity present in an ilvB + strain or in an ilvHI + strain is about as VaP as that of the wild type while that of the ilvG +strain is Val r. The activity present in an ilvG + strain is very low (6 compared to 112 of the wild type) and is dependent on the protein concentration used in the assay tubes so that a 2 to 4-fold difference in specific activity may be obtained by varying the protein concentration. As of now we have not succeeded in obtaining linearity and therefore the specific activity value is only indicative. A low value of specific activity of acetolactate synthase in an iIvG + strain is to

22

J. G u a r d i o l a et al. : A c e t o l a c t a t e Synthase I s o e n z y m e s

be expected from the small difference in resistance to valine previously found in extracts of a wild type strain compared to that of an ilv0603 strain (Favre et al., 1976). Strain MI262e grows slowly in minimal medium, probably because the level of acetolactate synthase activity is too low for normal growth. This strain is often contaminated with " r e v e r t a n t s " having a higher growth rate and a higher value of acetolactate synthase specific activity. Strain MI262, the triple mutant, as previously reported, contains almost undetectable acetolactate synthase activity and does not grow in the absence of isoleucine and valine. The strains possessing a single acetolactate synthase were tested for growth inhibition by different substances with the aim of finding specific inhibitors for each strain and thus to corroborate the evidence that the three enzymes are separate entities. The plates used for this experiment contained thiamine and glycine and were incubated at 30 ° C because strains missing acetolactate synthase genes show a temperature sensitive cell envelope defect (De Felice et al., accompanying manuscript). The data in Table 3 show that valine (glycylvaline showed the same effect), as expected, inhibit the growth of all strains except MI262e (ilvG+). Leucine inhibits the growth of strain MI261c and especially of strain MI263 (ilvI+); methionine inhibits very strikingly strain MI262e (ilvG +) and lysine has a small effect on strain MI253c. Among many analogues tested only those reported in Table 3 show a specific effect: glycylleucine stimulates the growth of strain MI262e, aminobutyric acid stimulates MI261 c and inhibits MI263, cycloleucine inhibits strain MI263.

Table 3. G r o w t h i n h i b i t i o n or s t i m u l a t i o n of strains possessing a single a c e t o l a c t a t e synthase a Substance tested b

MI149h

valine inh 12 leucine methionine lysine glycylleucine inh 10 aminobutyric cycloleucine . .

MI253c

MI261c

MI262e

MI263

inh 12 inh 7 inh 12 .

inh 12 inh 6 stim 6 inh 12 stim 12

inh 22 stim 10 -

inh inh inh inh inh inh

.

12 15 6 15 20 15

" S y m b o l s are: inh for inhibition, stim for s t i m u l a t i o n , - for no effect. Figures represent m m of i n h i b i t i o n or s t i m u l a t i o n . F o r further details see M a t e r i a l s a n d M e t h o d s b The a m o u n t s used were: 0.43 g m o l e s of L-valine (glycyl-Lv a l i n e gave the same effect), 1.19 ~tmoles of L-leucine, 2.16 g m o l e s of L - m e t h i o n i n e , 7.8 g m o l e s of L-lysine, 1.73 ~tmoles of glycyl-Lleucine, 9.8 txmoles of L - e - a m i n o - n - b u t y r i c acid, 2.43 ~xmoles of cycloleucine

--

P1(MI269a)..

4-

--

619

.

bgl

"G.D A O B

MI267 4-

--

@

4--

623

(is) Fig. 3. Cross between strain M I 2 6 7

(bgl +, ilvD +, EvB623 (ts))

a n d P I p h a g e g r o w n on strain M I 2 6 9 a (bgl-, ilvD642, ilvB619) w i t h selection of bgl- iIvD + t r a n s d u c t a n t s

Isolation of a ts Mutation in the ilvB Gene Mutants with a ts Ilv + phenotype were isolated from strain MI262b, Ilv- because of the ilvB619, ilvG605 and ilvI614 mutations. Strains altered in acetolactate synthase activity grow on plates incubated at 37 ° C but lyse when the plates are incubated at higher temperatures (De Felice et al., accompanying manuscript); for this reason we looked for ts revertants that failed to grow at 37 ° C. Independent Ilv ÷ spontaneous mutants were isolated by plating different cultures of strain MI262b at 30 ° C on minimal plates. In this way 43 mutants were isolated and 10 of them were found to be ts; at the permissive temperature 5 of these ts revertants were Val r, the other 5 being Val S. Since the ilvB + allele is essential for the expression of an acetolactate synthase activity sensitive to valine, we chose a ts, VaP revertant for further study. This strain was named MI267 and the mutation conferring the Ilv ts requirement was tentatively named ilvB623. A preliminary experiment showed that this mutation is 59% (57/96) linked to bgl. Therefore, an ilvI mutation is excluded. The ilvB623 mutation was located on the chromosome by means of the cross shown in Figure 3. Strain MI267 was treated with P1 grown on strain MI269a (see Materials and Methods for the preparation of this strain) and bgl-, ilvD + transductants were selected on minimal plates containing arbutin (as a carbon source), isoleucine and e,/3-dihydroxy-isovalerate. The plates were incubated at 30 ° C and 99% (95/96) of the transductants were found to be ts. This establishes the order bgl, ilvD, ilvB623 and excludes the possibility of a ts mutation in iIvG giving a valine sensitive acetolactate synthase, since in this case the ts transductants would be expected to be much less than 99%.

Isolation of a ts Mutation in the ilvG Gene For this experiment we decided not to analyze the Val r, Ilv-ts revertants of strain MI262b because they might be altered not only in ilvG but also ilvI ~or even in ilvB. The strain we chose, MI265 (ilvG605, ilv0603, ilvB619, ilvI620) contains an ilvH + allele and

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

P1(MI262h) M1266a

,

r ...... //

bgl ~\D A O C / B ......... L. . . . . . . J -6 2 2 + - I - - H- -

HI

purified, n a m e d strain M I 2 6 6 b , a n d k e p t for further study.

//

(ts)

Acetolactate Synthase Activity in the ts Mutants

Fig. 4. Cross between strain MI266a (bgl , ilvG622 (ts), ilvB619) and P1 phage grown on strain MI262h (bgl+, ilvDAC115, ilvB+) with selection of Ilv ÷ transductants at 37° C

Table 4. The 4 classes of recombinants found in the cross of Figure 4 Val r at 30° C

Val' at 37° C

bgl b

+ + + +

+ +

+

a

23

+

Frequency %

29 (28/96) 2 (2/96) 1 (1/96) 68 (65/96)

" + and - mean growth or absence of growth respectively on plates supplemented with valine b -t- and - mean genotype

t h e r e f o r e its ilvI + r e v e r t a n t s s h o u l d be VaP. W e isolated 50 Ilv + i n d e p e n d e n t r e v e r t a n t s at 30 ° C; 10 o f t h e m were ts and, o f these, one was Val r. A bglderivative o f this strain was i s o l a t e d on m i n i m a l plates c o n t a i n i n g a r b u t i n as c a r b o n source. This strain was n a m e d M I 2 6 6 a a n d the m u t a t i o n c o n f e r r i n g t h e Ilv-ts r e q u i r e m e n t was t e n t a t i v e l y n a m e d ilvG622. T h e Ilv-ts p h e n o t y p e a n d t h a t due to the bgl- m u t a t i o n ( g r o w t h o n a r b u t i n ) were f o u n d to be l i n k e d to the ilvC608 allele o f strain M I 1 5 9 (100% (48/48) a n d 6 3 % (30/48) c o n t r a n s d u c t i o n respectively). S t r a i n M I 2 6 6 a was t r e a t e d with P1 g r o w n on strain M I 2 6 2 h (bgl +, ilvDAC115, ilvH612, ilvI614) and Ilv + t r a n s d u c t a n t s were selected at 3 7 ° C (see Fig. 4). T h e ilvDAC115 m u t a t i o n has been s h o w n to be ilvG + a n d ilvB + ( G u a r d i o l a , D e Felice a n d I a c c a r i n o , 1974) a n d therefore the Ilv ÷ t r a n s d u c t a n t s s h o u l d be either ilvG + o r ilvB + (less f r e q u e n t l y both). I f the m u t a t i o n conferring the Ilv-ts p h e n o t y p e is in ilvG the p h e n o t y p e o f this m u t a t i o n s h o u l d change, in the ilvB + t r a n s d u c tants, to Valr-ts. T h e d a t a in T a b l e 4 show that, indeed, 2 9 % (28/96) o f the t r a n s d u c t a n t s were Valr-ts. M o r e o v e r , they were bgl- while the t r a n s d u c t a n t s showing a Val r p h e n o t y p e at b o t h t e m p e r a t u r e s (30 ° a n d 37 ° C) a n d s u p p o s e d l y ilvG +, ilvB619, were, as expected, bgl ÷. W e c o n c l u d e t h a t the m u t a t i o n conferring the Ilv-ts p h e n o t y p e to strain M I 2 6 6 a is in ilvG. P1 p h a g e g r o w n on this strain was used to transduce strain M I 2 6 2 b ( I l v - b e c a u s e o f m u t a t i o n s in ilvB, ilvG a n d ilvI) a n d Ilv + t r a n s d u c t a n t s were selected a n d f o u n d to be all ts. One o f these was

T h e e x p e r i m e n t o f F i g u r e 5 shows the rate o f inactivation o f a c e t o l a c t a t e synthase activity o f strain MI261 a n d strain MI267. S t r a i n MI267 c o n t a i n s the ilvB623 (ts) m u t a t i o n a n d m u t a t i o n s in the genes c o d i n g for the r e m a i n i n g a c e t o l a c t a t e synthases. Strain MI261, otherwise isogenic to strain MI267, c o n t a i n s the wild t y p e ilvB gene. T h e rate o f i n a c t i v a t i o n is strikingly different in the extracts o f the two strains. W e conclude that ilvB is the s t r u c t u r a l gene for a n acetolacrate synthase. T h e e x p e r i m e n t o f F i g u r e 6 shows the rate o f i n a c t i v a t i o n o f a c e t o l a c t a t e synthase activity o f strains M I 2 6 2 e a n d MI266b. Strain M I 2 6 6 b c o n t a i n s the ilvG622 (ts) m u t a t i o n a n d m u t a t i o n s in the genes coding for the r e m a i n i n g a c e t o l a c t a t e synthases. Strain MI262e, otherwise isogenic to strain M I 2 6 6 b , contains the wild type ilvG gene. F o l l o w i n g a suggestion o f Dr. M a r t i n F r e u n d l i c h , we l o w e r e d the p H o f the e x t r a c t i o n s o l u t i o n in o r d e r to increase the difference in stability o f the two extracts. The rate o f i n a c t i v a t i o n is strikingly different in the extracts o f the two strains. A s a l r e a d y m e n t i o n e d above, the activity was n o t p r o p o r t i o n a l to the a m o u n t o f extract a n d t h e r e f o r e we used a fixed a m o u n t o f it in each assay t u b e ; if the a m o u n t o f extract was v a r i e d t h e specific activity

100 80-

60cn c

E

E

40

2 "C

20-

io

io min

6o

8'0

at

40 C

Fig. 5. Rate of inactivation at 40° C of extracts from strain MI261 (o) and MI267 (e). The extraction solution was 10mM potassium phosphate, pH 7.1, containing 5 mM MgC12, 20 gg/ml flavin adenine dinucleotide, 200 ~g/ml thiamine pyrophosphate and 20% glycerol. Protein concentration was 8.2 mg/ml for strain MI261 and 6.9 mg/ml for strain MI267. The extracts were put at 40° C and at different times portions were withdrawn, put at 0° C and assayed for acetolactate synthase activity

24

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

1001~ o e; OI

80-tt 60-

""

v

o

_

~

o

.E E

~

40-

~L.

E

>

20-

4o

8o

li0

16o

26o

24o

min at 40 C Fig. 6. Rate of inactivation at 42 ° C of extracts from strain MI262e (o) and MI266b (e). The extraction solution was 0.05 M potassium phosphate, pH 7.5. Protein concentration was 13.4 mg/ml for strain MI262e and 12 mg/ml for strain MI266b. The specific activities were 2.7 for strain MI262e and 23 for strain MI266b. The extracts were put at 42 ° C and at different times portions were withdrawn, put at 0 ° C and then 0.2 ml aliquots were put in each tube for the assay of acetolactate synthase activity. The buffer in the assay tubes for this experiment was 20 m M potassium phosphate, p H 6.5, final concentration (the resulting p H was 6.8)

was different but the difference between the two curves remained. Under the conditions of the experiment described in Figure 6 the specific activity of the two extracts is quite different (2.7 versus 23); this difference is reproducible and is also present if an extract of the parental strain is used (MI266a) and we believe it is a consequence of the mutation. Although it is difficult to give a simple explanation of these effects in molecular terms until more experiments with purified enzymes are performed, the difference in rate of inactivation shows that the acetolacrate synthase activity of strain MI266b is temperature sensitive. We conclude that ilvG is the structural gene for an acetolactate synthase.

Discussion

The results presented in this paper show that ilvB and ilvG are structural genes for acetolactate synthase activity in E. coli K-12. We have previously reported similar evidence for the ilvHI genes. While a strain mutated in ilvB, ilvG and ilvHI is auxotrophic for isoleucine and valine and devoid of acetolactate synthase activity, strains carrying a wild type allele of a single gene become prototrophic and show acetolactate synthase activity. On this basis we conclude that E. coli K-12 contains three different isoenzymes and propose for them the following names: acetolactate

synthase I (AHAS I) for the product o f the ilvB gene, acetolactate synthase II (AHAS lI) for the product of the iIvG gene and acetolactate synthase III (AHAS III) for the product of the ilvHI genes. The availability of strains containing a specific isoenzyme permits further genetic and biochemical analysis. We believe that purification and characterization of the three isoenzymes is necessary to confirm their existence in the cell as separate entities. The prototrophy of the strains carrying a single isoenzyme shows that each of them is capable of catalyzing the biosynthesis of both the isoleucine and valine intermediates. However, it is possible that in a wild type strain each isoenzyme has a more specific role. Indeed, when ilvG is expressed there is excretion ofisoleucine but not ofvaline (Ramakrishnan and Adelberg, 1964), while, when the Val s AHAS III becomes Val r because of a mutation, there is an increase in the pool size of valine, but not of isoleucine (De Felice et al., 1974b). Moreover, the slow growth rate of the strain containing only AHAS II (strain MI262e, ilv0603, ilvG +) is in contrast with the normal growth rate in valine of a strain expressing all acetolactate synthases: if, in the presence of valine, AHAS I and AHAS III are inactive in vivo, then all isoleucine biosynthesis should be carried by AHAS II. Therefore, this isoenzyme would be capable of catalyzing enough isoleucine synthesis for a normal growth rate when AHAS I and AHAS III are present, but not when they are absent. We think it is likely that interaction between acetolactate synthases in vivo plays an important role in the channeling of biosynthesis toward a specific end product. Indeed, this might be one of the evolutionary advantages for the presence of these isoenzymes in E. coli K-12. Evidence for the presence of acetolactate synthase isoenzymes in E. coli K-12 has been reported previously. Radhakrishnan and Snell (1960) were able to distinguish two activities on the basis of their pH optima (pH 8 and pH 6) and believed that one of them was involved in isoleucine and valine biosynthesis while the other was present for catabolic purposes. Ramakrishnan and Adelberg (1965a) isolated a mutant in a gene named by them ilvB. The mutant strain, although a slow grower, was prototrophic, thus showing that t h e " catabolic" activity, at least in these conditions, could provide intermediates for isoleucine and valine biosynthesis. We show that a VaV mutation, ilv0603, causes the appearance of AHAS II, presumably not expressed in the wild type. Mutations with a similar phenotype are ilv0264, ilv0268 and ilv0269 isolated by Ramakrishnan and Adelberg (1964) who mapped ilv0264 between ilvA and ilvC, where ilv0603 is located. The ilv0268 and ilv0269 mutations have been

J. Guardiola et al. : Acetolactate Synthase Isoenzymes

mapped by Cohen and Jones (1976) between rbs and ilvE and found to be also cis-dominant as ilv0603 and ilv0264. We think it is likely that ilv0264 is a mutation in the same site altered by ilv0603, while ilv0268 and ilv0269 should be renamed with a new gene symbol. The alteration caused by the ilv0268 and ilv0269 mutations has effects on the total acetolactate synthase activity, similar to those reported for the ilv0603 mutation, thus suggesting that ilvG is expressed when one of two cis-dominant elements on either side ofa gene cluster is altered. More experiments should be performed to clarify this point. Another mutation causing the appearance of a Val r phenotype and a VaV acetolactate synthase activity is the ilvF465 mutation described by Pledger and Umbarger (1973) and located far from the ilv cluster. At this moment, it is premature to speculate on the effect of the ilvF465 mutation. This might (a) cause the appearance of AHAS II or of a previously unrecognized synthase or, (b) prevent the sensitivity to valine of AHAS I or AHAS III (a mutation in ilvB causing a Val r phenotype has been described by Ramakrishnan and Adelberg (1965 b). Evidence for multiple forms of acetolactate synthases has been reported also in Salmonella typhimurium (Blatt, Pledger and Umbarger, 1972; O'Neill and Freundlich, 1972, 1973). Two genes have been described, ilvB and ilvG, coding for a Val s and a Val r acetolactate synthase respectively; the presence of mutants devoid of acetolactate synthase activity has been claimed (Blatt, Pledger and Umbarger, 1972) but not yet described. In Salmonella and E. coli W it is possible to separate two peaks of activity (one Val s and one VaV) by column chromatography (O'Neill and Freundlich, 1972, 1973; Favre et al., 1976). Acknowledgements. We thank N. Della Volpe, B. Esposito and C. Migliaccio for skillful technical assistance, Drs. Martin Freundlich and Mark Levinthal for useful discussions, Drs. M.D. Alper and B.N. Ames for communicating to us their method for selection of eya mutants and for generously sending a sample of phosphomycin, Dr. Barbara Bachmann for sending strains, information and genealogy of some of them.

References Bachmann, B.J., Low, K.B., Taylor, A.L.: Recalibrated linkage map of Escherichia coli K-12. Bact. Rev. 40, 116-167 (1976) Blatt, J.M., Pledger, W.J., Umbarger, H.E. : Isoleucine and valine metabolism in Escherichia coli. XX. Multiple forms of acetohydroxy acid synthetase. Biochem. biophys. Res. Commun. 48, 444 450 (1972) Cohen, B.M., Jones, E.W. : New map location ofilvO in Escherichia coli. Genetics 83, 201-225 (1976~ De Felice, M., Guardiola, J., Esposito, B., Iaccarino, M.: Structural genes for a newly recognized acetolactate synthase in Escherichia coli K-12. J. Bact. 120, 1068-1077 (1974a)

25 De Felice, M., Guardiola, J., Malorni, M.C., Klopotowski, T., Iaccarino, M. : Regulation of the pool size of valine in Escherichia coli K-12. J. Bact. 120, 1058-1067 (1974b) De Felice, M., Guardiola, J., Schreil, W., Levinthal, M., Iaccarino, M. : Metabolic interlock between the acetolactate synthase isoenzymes and lysine biosynthesis in Escherichia coli K-12. Molec. gen. Genet. 156, 9-16 (1977) De Felice, M., Squires, C., Levinthal, M., Guardiola, J., Lamberti, A., Iaccarino, M.: Growth inhibition of Escherichia coli K-12 by L-valine: a consequence of a regulatory pattern. Molec. gen. Genet. 156, 1 7 (1977) Favre, R., Wiater, A., Puppo, S., Iaccarino, M., Noelle, R., Freundlich, M.: Expression of a valine-resistant acetolactate synthase activity mediated by the ilvO and ilvG genes of Escherichia coil K-12. Molec. gen. Genet., 143, 243-252 (1976) Groves, W.E., Davis, F.C., Jr., Sells, B.: Spectrophotometric determination of microgram quantities of protein without nucleic acid interference. Anal. Biochem. 22, 195 210 (1968) Guardiola, J. : Requirement of the newly recognized ilvJ gene for the expression of a valine transaminase activity in E. coli K-12. Molec. gen. Genet. (in press) (1977) Guardiola, J., De Felice, M., Iaccarino, M. : Mutant of Escherichia coli K-12 missing acetolactate synthase activity. J. Bact. 120, 536 538 (1974) Guardiola, J., De Felice, Klopotowski, Iaccarino, M. : Mutations affecting the different transport systems for isoleucine, leucine and valine in Escherichia coli K-12. J. Bact. 117, 393405 (1974) O'Neill, J.P., Freundlich, M.: Two forms of biosynthetic acetohydroxy acid synthetase in Salmonella typhymurium. Biochem. biophys. Res. Commun. 48, 437~443 (1972) O'Neill, J.P., Freundlich, M. : Temperature-sensitive growth inhibition by valine in Sahnonella typhymurium: alteration of one form of acetohydroxy acid synthetase. J. Bact. 116, 98-106 (1973) Pledger, W.J., Umbarger, H.E. : Isoleucine and valine metabolism in Escherichia coli. XXI. Mutations affecting derepression and valine resistance. J. Bact. 114, 183 I94 (1973) Radhakrishnan, A.N., Snell, E.E.: Biosynthesis of valine and isoleucine. II. Formation of c~-acetolactate and c~-aceto-c~-hydroxybutyrate in Neurospora crassa and Escherichia coli. J. biol. Chem. 235, 2316-2321 (1960) Ramakrishnan, T., Adelberg, E.A. : Regulatory mechanisms in the biosynthesis of isoleucine and valine. I. Genetic derepression of enzyme formation. J. Bact. 87, 566-573 (1964) Ramakrishnan, T., Adelberg, E.A.: Regulatory mechanisms in the biosynthesis of isoleucine and valine. II. Identification of two operator genes. J. Bact. 89, 654 660 (1965a) Ramakrishnan, T., Adelberg, E.A. : Regulatory mechanisms in the biosynthesis of isoleucine and valine. III. Map order of the structural genes and operator genes. J. Bact. 89, 661-664 (1965 b) Stormer, F.C., Umbarger, H.E. : The requirement for flavine adenine dinucleotide in the formation of acetolactate by Salmonella typhymurium extracts. Biochem. biophys. Res. Commun. 17, 587-592 (1964) Umbarger, H.E. : Regulation of the biosynthesis of the branchedchain amino acids. In: Current topics in cellular regulation (B.L. Horecker and E.R. Stadtman, eds.), Vol. 1, pp. 57-76. London-New York: Academic Press 1969. Umbarger, H.E., Davis, D.B.: Pathways of amino acid biosynthesis. In: The bacteria (I.C. Gunsalus and R.Y. Stanier, eds.), Vol. III, pp. 167-251. London-New York: Academic Press 1962 Vogel, H.J., Bonner, D.M.: Acetylornithinase of Escherichia coli." partial purification and some properties. J. biol. Chem. 218, 97-106 (1956)

Communicated by F. Gros Received March 6~May 26, 1977