Mol Gen Genet (1983) 191:353-357 © Springer-Verlag 1983

Regulation of Acetohydroxy Acid Synthase Activities in Adenyl Cyclase-Deficient Strains of Escherichia coil K-12 Arthur L. Williams School of Biological Sciences, University of Kentucky, Lexington, KY 40506, USA Summary. Previous findings suggested that cyclic AMP was involved in the regulation of ilvB(AHASI) only and that ilvG (AHASII) and ilvHI (AHASIII) were not controlled by this nucleotide. In this study, derepression patterns of total AHAS activities (ilvB and ilvHI) in adenyl cyclasenegative strains (i.e. cya-) were substantially reduced as contrasted with AHAS activity observed for cya + strains. Further, the parental strains (cya +) consistently exhibited higher levels of AHAS activity than mutant strains (cya-) during carbon and energy downshifts. Other data suggested that the valine derepression signal could not override the necessity for cya gene product to yield maximal derepression of AHAS gene activities. Cyclic AMP stimulated AHAS gene activities under both in vivo and in vitro assay conditions. Thus, these data provide evidence for an absolute requirement of cAMP for maximal expression of the genes encoding for AHAS activities of E. coli K-12.

Introduction

Regulation of the formation of the isoleucine and valine biosynthetic enzymes in Escherichia coli K-12 involves both end-product repression (Umbarger 1978) via an attenuation control process (Lawther and Hatfield 1980; Nargang et al. 1980) and guanosine tetraphosphate (ppGpp) (Freundlich 1977). In addition, other findings suggest that the ppGpp requirement for normal expression of one of the isoleucine and valine enzyme activities, acetohydroxy acid synthase, can be replaced by cyclic adenosine Y, 5'-monophosphate (cyclic AMP) (Freundlich 1977). Cyclic AMP is thought to be a signal molecule that indicates a fuel or carbon deficiency of the cell (Alper and Ames 1978) and is usually involved in regulation of degradative operons (Pastan and Adhya 1976). Thus, the participation of cAMP in control of acetohydroxy acid synthase (AHAS) is highly unusual as this is a biosynthetic enzyme of E. coli K-12. More recently, it has been reported that cyclic AMP is involved in the regulation of expression of the ilvB gene only (AHASI), while that of ilvG (AHASII) and ilvHI (AHASIII) are not affected by cyclic AMP (Sutton and Freundlich 1980). The structure genes for the isoleucine and valine biosynthetic enzymes comprise several independent operons. The ilvGEDA operon is multivalently controlled by the supply for all three branched-chain amino acids, isoleucine, valine and leucine (Umbarger 1978). The ilvG gene specifies a va-

line-resistant form of acetohydroxy acid synthase (AHASII) which is not expressed in wild type (ilvG-, Val s) E. eoli K-12 (DeFelice et al. 1978; Umbarger 1978). An explanation for the lack of expression of this gene was recently provided by sequencing the DNA of the ilvG region and by electrophoretic analysis of the proteins, in which it was shown that the ilvG gene contains a site of polarity that causes chain termination (Lawther et al. 1981). The ilvB gene, multivalently controlled by valine and leucine, encodes for a valine-sensitive AHASI (Newman and Levinthal 1980). This gene lies at 82 min on the E. coli chromosome and is thereby unlinked to the ilvGEDA cluster at 84 min (Newman and Levinthal 1980; Bachmann and Low 1980). The ilvHI operon encodes a second valinesensitive AHAS (III) whose formation is controlled by leucine alone (DeFelice and Levinthal 1977). The aims of this study were as follows (i) to examine the levels of the AHAS gene products in adenyl cyclasenegative strains (i.e. unable to synthesize cAMP); (ii) to determine the effect of exogenous cAMP on ilv gene expression in adenyl cyclase-negative strains and (iii) to monitor the effect of cAMP upon an in vitro transcription-translation system using ilv DNA. In this report, evidence is provided for an absolute requirement of cAMP for the maximal expression of the genes encoding for AHAS activities of E. coli: Materials and Methods Bacterial Strains

All of the strains used in this study are listed in Table 1. The strains are from E. coli K-12 stocks. Reagents

Amino acids, pyridoxal-5-phosphate, flavine adenine dinucleotide, sodium pyruvate, thiamine, cAMP, RNA polymerase, and ppGpp were from the Sigma Chemical Company, St. Louis, MO. Creatine hydrate was from Calibiochem and bovine serum albumin was from Miles Laboratories. [5,6-3H] Uridine-5'-triphosphate and [2,8-3H] cAMP were from New England Nuclear Corporation. All other chemicals were of reagent grade. Media

The minimal medium used was that of Davis and Mingioli (1950) modified by omitting citrate and increasing the glu-

354 R N A transcripts was performed as described by Moore et al. (1981).

Table l. E. coli strains used Strain

Genotype relevant ilv phenotype

Reference or source

CU4 CU424 CU437 CA8300 CA8306 Hfr3000 PP48

Stock Collection Stock Collection Stock Collection W. Reznikoff W. Reznikoff M. Iwaya M. Iwaya

UKt43

F-, gaiT12,2F - , galT12,ilvC462, 2F-, cya-283 Hfr, Sms, thi-1 Hfr, Sms, thi-1, cyaA Hfr, thi-1, relA1, spoT1, 2HfrC, thi-1, relA1, spoT1, eya-1, 2F-, cya-283, ilvC

UK167

Hfr, Sms, thi-1, cyaA, ilvC

UK175

HfrC, thi-1, relA1, spoT1, cya-1, ilvC

DES Mutagenesis of CU437 DES Mutagenesis of CA8306 DES Mutagenesis of PP48

2CI857,2C1857dilvB

Stock Collection

Phage UK99

cose concentration to 0.5%. Amino acid supplements were added at 0.5 raM, unless otherwise indicated; thiamine, 12 mg per liter.

Growth of Cells Batch cultures were prepared by inoculating from nutrient agar slants into 5 ml of Luria broth or tryptone broth. The tubes were shaken on a Brunswick Gyrotory Water Bath Shaker-Model G76 at 37 ° C and incubated until stationary phase was attained. The cells were centrifuged and the pellets were resuspended in minimal medium with 0.5% glucose or minimal medium supplemented with 4 × 10 4 M L-leucine and L-isoleucine, and 1 × 10 -3 M L-valine along with 0.5% glucose for repressing growth conditions. Growth of the cultures was monitored by measuring the absorbancy of the cultures at 660 nm in a Spectronic 20 spectrophotometer.

Enzyme Assays The following enzyme assays were done as previously described: acetohydroxy acid synthase (Stormer and Umbarger 1964) and Transaminase B (Duggan and Wechsler 1973). Protein was measured by the method of Lowry et al. (1951), with bovine serum albumin (100 ~tg/ml) as the protein standard. Specific activities are expressed as gmoles of product formed per minute per mg of protein.

Phage DNA Extraction The procedure of James et al. (1976) employing phenolchloroform-isoamyl alcohol was used to extract D N A from the specialized transducing phage carrying the ilvB region. The OD26 o and ODz8 o nm of the D N A suspension were measured with a Beckman-Model DBG spectrophotometer.

In Vitro Transcription Assay The transcription reactions were carried out as described by Majors (1975). The electrophoretic fractionation of

Results and Discussion

Monitoring A H A S Gene Products in AdenyI Cyelase-Negatire Strains. The effect of the eya gene product on expression of the isoleucine and valine biosynthetic enzyme was examined in isogenic cya + and eya- derivaties of E. coli K-12. The bacterial strains were grown overnight in minimal medium containing glucose and the 20 amino acids. Subsequently, the cells were centrifuged, washed with minimal medium and resuspended in the same medium (excess, Table 2) and in medium with growth limiting amounts of isoleucine and valine (minimal, Table 2). These strains were harvested at approximately A420 nm of 0.5. The cells were washed and cellular extracts prepared. The ilv enzymes were assayed and the specific activity for each enzyme in the cya- derivatives was compared to that of the cya + strains. As shown in Table 2, AHAS activity increased in the cya + strains when grown in minimal-glucose medium alone as contrasted with the activity observed for cultures grown in excess isoleucine, valine and leucine. Conversely, the AHAS activity of the eya mutants remained virtually the same (i.e. unable to respond to the derepression signal), owing to the absence of cAMP. Derepression Patterns of the ilv Operons in cya Mutants After Carbon and Energy Downshift. To further assess the role of the cya gen product on ilv operon expression, the ilv derepression response was examined during carbon and energy source transitions. These experimental conditions were similar to those described in the previous experiment, except that succinate or glycerol was substituted for glucose. The cells grown on a waterbath shaker at 37° C to A42 o nm of 0.5, collected by centrifugation and extracts were preTable 2. Acetohydroxy acid synthase activities of adenyl cyclasenegative strains of E. coli Strain

Medium

Specific activity Total AHAS

CU4 (wild type) CU437 (cya-) CA8300 (cya ÷) CA8306 (cyaA) Hfr3000 (cya +) PP48 (cya-)

Excess" Minimal b Excess Minimal Excess Minimal Excess Minimal Excess Minimal Excess Minimal

0.016 0.063 0.017 0.018 0.014 0.026 0.012 0.014 0.011 0.025 0.017 0.019

Cells were grown, cell extracts were prepared, and enzymatic assays were performed as described in the text. Specific activity is expressed as gmoles/min/mg protein. Protein concentration was determined by the method of Lowry et al. (1951) a Excess, means growth in 0.5% glucose and all 20 amino acids b Minimal, means growth in 0.5% glucose and growth limiting quantities of isoleucine and valine

355 Table 3. Expression of the AHAS conferring genes of carbon/energy source

as a

function

Strain

Medium

AHAS activity

CU4 (wild type)

Minimal-glucose Minimal-succinate Minimal glycerol Minimal-glucose Minimal-succinate Minimal-glycerol Minimal-glucose Minimal-succinate Minimal-glycerol Minimal-glucose Minimal-succinate Minimal-glycerol

0.10 0.57 0.20 0.02 0.16 0.03 0.04 0.43 0.08 0.02 0.20 0.03

CU437 (cya-283)

CA8300 (wild type)

CA8306 (cyaA)

Cells were grown and assayed as described in Table 2, except succinate and glycerol were used as carbon sources in addition to glucose. For the glycerol-grown cultures, 0.0025% glucose was also added

Table 5. Effect of limiting valine on AHAS gene expression Strain

Medium condition

AHAS activity

CU424

Excess Limiting valine Excess Limiting valine Excess Limiting valine Excess Limiting valine

0.019 0.161 0.022 0.020 0.018 0.014 0.015 0.025

(cya +, ilvC) UK143

(cya-283, ilvC) UK167

(eyaA, iloC) UK175

(eya-1, ilvC)

Cells were grown in minimal-glucose medium supplemented with excess amounts of isoleucine, leucine and valine (excess) and minimal glucose plus 4 x 1 0 - 4 M isoleucine and leucine and 0.5 x 10 -4 M valine (limiting). AHAS activity is expressed as described for Table 2 Table 6. Effect of exogenous cAMP on ilv gene expression Strain

Table 4. Effect of exogenous cAMP on AHAS gene expression Strain

AHAS activity

% Valine resistance

CU424

AHAS (ilvB/HI)

TrB (ilvE)

-cAMP

+cAMP

-cAMP

+cAMP

0.011

0.057

0.027

0.025

0.022

0.031

0.036

0.032

0.019

0.046

0.018

0.012

0.012

0.066

0.029

0.025

(cya +, ilvC) - Valine CU4

0.348

+ Valine 0.193

AHAS activity 55

(cya +) CU437

0.123

0.076

61

(cya-) CA8300

0.271

0.117

43

0.123

0.026

2t

(cya +) CA8006

(cyaA) Cells were grown in minimal-glucose medium supplemented with cAMP at a final concentration of 2 mM. AHAS activity was determined in extracts in the presence and absence of L-valine (1 raM) as described in Table 2

pared and used in the ilv enzyme assay systems. As shown in Table 3, the parental strains (cya +) consistently exhibited higher levels of A H A S activity than m u t a n t strains ( c y a ) . However, the m u t a n t strains did show some increase in A H A S activity under these conditions p r o b a b l y owing to p p G p p accumulation. Still, the level o f A H A S o f the cya- derivatives was usually severalfold less than that o f the n o r m a l strains when grown with either succinate or glycerol as the carbon/energy source (Table 3).

The Effect of Exogenous c A M P on A H A S Gene Expression. In order to examine the effect o f c A M P directly on derepression o f synthesis o f the isoleucine and valine enzymes, these activities were examined in isogenic cya- and cya + derivatives o f E. coli K-12 in the presence and absence o f exogenous c A M P (Table 4). As shown in this Table, A H A S activity was stimulated by exogenous c A M P in both m u t a n t and wild type strains. Again, the wild type strains showed higher levels of A H A S as c o m p a r e d to m u t a n t strains (cya-). Thus, the results o f these experiments indicate that

UK143

(cya-283, ilvC) UK167

(cyaA, ilvC) UK175

(cya-1, ilvC) Cells were grown under repressed conditions supplemented with 2 mM cAMP. All other procedures are as described in Table 2 for measurement of the ilvB/Hl (total AHAS, acetohydroxy acid synthase) and the ilvE (TrB, transaminase B) gene products A H A S gene products can be derepressed provided that c A M P is presence in the growth medium. The a p p a r e n t lack o f maximal derepression in the cya m u t a n t strains strongly suggests an absolute requirement of the cya gene product (i.e. cAMP).

Effect of Valine-Limitation on the Expression o f A H A S Gene. Enzyme levels in several adenyl cyclase-negative carrying an ilvC m u t a t i o n were determined under limiting-valine conditions. Since limiting valine causes maximal derepression of eya + strains, this experiment was performed to characterize the derepression patterns of cya- strains. U p o n valine limitation, there was an i m p a i r m e n t o f A H A S activity in all three cya m u t a n t strains (Table 5). Also, it a p p e a r e d that m u t a n t A H A S activities were very unstable (i.e. rapid turnover) during valine-limitation (his observation was consistently observed). Thus, these d a t a suggest that the valine derepression signal could not override the necessity of cya gene product to yield maximal derepression of the A H A S gene.

Effect of Exogenous c A M P on ilv Gene Expression. The effect o f c A M P upon the expression of two ilv gene products was examined in these cya-, itvC- derivatives mentioned above. As shown in Table 6, the a d d i t i o n of c A M P to these

356 Table 7. Effect of exogenous ppGpp on ilv gene expression Strain

CU4

AHAS (ilvB/Hl)

TrB (ilvE)

- ppGpp

+ ppGpp

- ppGpp

+ ppGpp

0.015

0.184

0.013

0.037

0.010

0.045

0.015

0.033

0.027

0.043

0.005

0.032

0.006

0.025

0.005

0.027

0.014

0.037

0.008

0.020

0.010

0.030

0.007

0.023

(cya +) CU437

Table 8. Effect of cAMP on synthesis of AHAS in cells grown under repressed and derepressed conditons Strain

Medium

AHAS activity

Relative AHAS activity

CU4

Excess Minimal Excess + cAMP Minimal + cAMP Excess Minimal Excess +cAMP Minimal + cAMP Excess Minimal Excess + cAMP Minimal+cAMP Excess Minimal Excess + cAMP Minimal + cAMP

0.013 0.057 0.065 0.080 0.016 0.018 0.067 0.069 0.014 0.039 0.049 0.051 0.012 0.012 0.037 0.049

1.00 4.38 5.00 6.15 1.00 1.13 4.18 4.41 1.00 2.79 3.50 3.64 1.00 1.00 3.08 4.08

(cya +)

(cya-283) CA8300

(cya +) CA8306

CU437

(cya-283)

(cyaA) Hfr3000

(cya +) pp48

CA8300

(cya +)

(cya=l) Cells were grown under repressed conditions supplemented with 2 mM ppGpp. All other experimental details are as described in Table 2, except prior to harvesting, these cultures were subjected to toluene treatment (1.3%) for 30 min cultures caused a moderate elevation of the A H A S gene products, but appeared not to have any effect on the ilvE gene product (transaminase B). This, these data suggest that c A M P effect is specific for expression of the A H A S gene. This is compatible with the findings of Sutton and Freundlich (1980).

Effect of Exogenous ppGpp on ilv Gene Expression. Since p p G p p also acculumates during shiftdown conditions (Table 3), the effect of p p G p p alone was examined in toluenetreated mutant and parental strains. The data in Table 7 indicate that p p G p p can stimulate A H A S activity, but the stimulatory effect is much less than the c A M P effect. Also, under these experimental conditions, the ilvE gene product (transaminase B) appeared to be elevated by the presence of p p G p p which was not found with cAMP. Table 8 shows a summary of c A M P effect on A H A S activity under repressed (excess) and derepressed (minimal) conditions, in which it is evident that the derepression patterns of A H A S are impaired in the cAMP-deficient cells.

Effect of c A M P on Derepression of the ilv m R N A (In vitro). To further examine the role of c A M P with respect to the ilv gene expression D N A isolated from 2transducing phages (2dp8Odilv and 2dilvB) was employed as templates for the in vitro transcription/translation system (i.e. Zubay-type). Several bacterial strains were employed as a source of S-30 preparation; these strains were isogenic, except for the super induced allele. The transcription/translation conditions were identical to those described by Zubay (1973), except various amounts of c A M P were added to this synthetic system alone as well as in combination with another nucleotide p p G p p (Guanosine tetraphosphate). As shown in the Table 9, c A M P stimulated the transcription of the ilvB m R N A . Further, it had no effect on the translation process per se (data not shown). Moreover, c A M P influenced the activity of R N A polymerase activity as judged by D N A binding studies. Our preliminary data suggested that c A M P may play a role in the maximal expression o f the ilvB gene (AHASI)

CA8306

(cyaA)

All conditions were as described in Table 2. Relative AHAS was determined by comparing all specific activities to that of each strain grown in minimal medium plus excess isoleucine, valine and leucine which is set to equal 1.0 Table 9. The effect of cAMP on the in vitro transcription of [3H] RNA from 2dilvB DNA a Gel No. b

In vitro transcription conditions

CPM in [3H] RNA sample

1 2 3 4

Minus ilv DNA Minus cAMP and CRP Plus cAMP and CRP Plus cAMP, CRP and ppGpp

90 161 751 1037

a The transcription reactions were carried out as described in the Materials and Methods Section b Electrophoretic fraetionation of RNA transcripts was performed as described in the Materials and Methods Section and the RNA was removed from the gels and resuspended in electrophoresis buffers and the radioactivity determined for 50 tll samples of E. coli. The present study was specifically undertaken to address this observation. Results obtained show an absolute requirement of c A M P for the maximal expression of the A H A S gene(s). Further, data from this study revealed that the site of activity of c A M P is at the transcriptional level, influencing the R N A polymerase activity for this operon. It was observed that c A M P probably functions best in combination with ppGpp. Evidence is provided for an absolute requirement of c A M P for maximal expression of the genes encoding for A H A S activities o f E. eoli K-12. Specifically, A H A S activity increased in the cya + strains when grown in minimal-glucose medium alone as contrasted with the activity observed for cultures grown in excess isoleucine, valine and leucine. Conversely, the A H A S activity of the eya mutants remained virtually the same (i.e. unable to respond to the derepression signal), owing the absence of cAMP. The growth conditions were performed in a manner whereby p p G p p levels should have been minimum. Further, during carbon and energy

357 transitions the parental strains (cya +) consistently exhibited higher levels of A H A S activity than mutant strains (cya-). However, the mutant strains did show elevated levels of A H A S activity under these conditions, probably owing to p p G p p accumulation. In addition, A H A S activity was stimulated by exogenous c A M P in both mutant and wild type strains, but the wild type strains (cya +) showed higher levels of A H A S activity as compared to mutant strains (cya-). Thus it appears that the A H A S gene products can be derepressed provided that c A M P is present in the growth medium of cya mutant strains. Under these experiments, the ilvE gene product (transaminase B) appeared elevated by the presence of p p G p p which was not found with cAMP. Valine-limitation failed to derepress A H A S gene products (ilvB, ilvHi) in c y a - strains. Thus, it appeared that the valine-derepression signal was impaired in c-AMP-deficient strains. Lastly, c A M P stimulated the transcription of ilvB m R N A in an in vitro transcription system. In addition, the stimulatory effect was even greater when p p G p p and c A M P were used in combination in the transcription assay. While the study has not revealed all the components and/or conditions for this regulatory system, I have shown the effect of one important effector (cAMP). This is an unusual observation whereby c A M P alters the expression of a biosynthetic operon. Presently, experiments are being performed (in this laboratory) to assess further the interconnection of c A M P and p p G p p role(s) with respect to A H A S gene expression. Acknowledgements. This study was supported in part by Biomedical Sciences Mini Grant (201-65-8E300-A5270). I would like to thank The University of Kentucky, School of Biological Sciences, for financial assistance in support of this study. I am also grateful to Dr. L.S. Williams for helpful discussions throughout this study. References

Alper MD, Ames BN (1978) Transport of antibiotics and metabolic analogs by systems under cyclic AMP control : positive selection of Salmonella typhimurium cya and crp mutants. J Bacteriol 133:14~157 Bachmann BJ, Low KB (1980) Linkage map of Escherichia coli K-12, ed 6. Microbiol Rev 44:1-56 Davis BD, Mingioli ES (1950) Mutants of Escherichia coli requiring methionine or Vitamin B12. J Bacteriol 60:17-28 DeFelice M, Newman T, Levinthal M (1978) Regulation of synthesis of the acetohydroxy acid synthase I. isoenzyme in Escherichia coli K-12. Biochim Biophys Acta 541:1-8

DeFelice M, Levinthal M (1977) The acetohydroxy acid synthase III isoenzyme of Escherichia coli K-12: regulation of synthesis by leucine. Biochim Biophys Res Comm 79:82-87 Duggan DE, Wechsler JA (1973) An assay for transaminase B enzyme in Escherichia coli K-12. Anal Biochem 51:67-79 Freundlich M (1977) Cyclic AMP can replace the relA-dependent requirement for derepression of acetohydroxy acid synthase in E. coli K-12. Cell 12:1121-1126 James PM, Sens W, Natter W, Moore S, James E (1976) Isolation and characterization of the specialized transducing bacteriophages ~b80dargF and 2h80ci857 dargF: specific cleavage of arginine transducing deoxyribonucleic acid by the endonucleases, EcoRl and SmaR. J Bacteriol 126:487-500 Lawther RP, Calhoun DH, Adams CW, Hauser CA, Gray J, Hatfield GW (1981) Molecular basis of valine resistance in Escherichia coli K-12. Proc Natl Acad Sci USA 78:922-925 Lawther RP, Hatfield GW (1980) MultiValent translational control of transcription at the attenuator of ilvGEDA operon of Escherichia coli K-I 2. Proc Natl Acad Sci USA 77:1862-1866 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the folin-phenol reagent. J Biol Chem 193 : 265-275 Majors J (1975) Initiation of in vitro mRNA synthesis from the wild type lac promoter. Proc Natl Acad Sci USA 72:4394~398 Moore SK, Garvin RT, James E (1981) Nucleotide sequence of the argF regulatory region of Escherichia coli K-12. Gene 16:119-132 Nargang FE, Subrahmanyam CS, Umbarger HE (1980) Nucleotide sequence of ilvGEDA operon attenuator region of Escherichia coli. Proc Natl Acad Sci USA 77:1823-1827 Newman TC, Levinthal M (1980) A new map location of the ilvB locus of Eseherichia coll. Genetics 96: 59-77 Pastan I, Adhya S (1976) Cyclic adenosine 5'-monophosphate in Escherichia coli. Bacteriol Rev 40:527-551 Stormer FC, Umharger HE (1964) The requirement for flavin adenine dinucleotide in the formation of acetolactate by Salmonella typhimurium extracts. Biochem Biophys Res Commun 17 : 587-592 Sutton A, Freundlich M (1980) Regulation of cyclic AMP of the ilvB-encoded biosynthetic acetohydroxy acid synthase in Escherichia coli K-12. Mol Gen Genet 178:179-183 Umbarger HE (1978) Amino acid biosynthesis and its regulation. Annu Rev Biochem 47:533-606 Zubay G (1972) In vitro synthesis Of protein in microbial systems. Annu Rev Genet 7:267-287 Communicated by G. O ' D o n o v a n Received November 30, 1982