defective transamination, mechanism resistance to ketomycin in

ANTIMICROBiAL AGENTS AND CHEMOTHERAPY, Apr. 1973, p. 510-516Copyright i 1973 American Society for Microbiology

Vol. 3, No. 4Printed in U.SA.

Defective Transamination, a Mechanism forResistance to Ketomycin in Escherichia coli

JULIUS H. JACKSON1 AND H. E. UMBARGER

Department of Biological Sciences, Purdue University, Lafayette, Indiana 47907

Received for publication 10 November 1972

A spontaneous mutant of a derivative of Escherichia coli strain K-12 resistantto 50 ug of ketomycin per ml was selected. The mutant displayed a two- tothreefold derepression of the isoleucine-valine biosynthetic enzymes and areduced growth rate in minimal medium. The lesion was found to lie in the gene(ilvE) specifying transaminase B and resulted in an isoleucine limitation. Thepresence of exogenous isoleucine during growth in minimal medium restorednormal phenotypic properties. The reduced transaminase B activity is responsi-ble for the resistance to ketomycin. An unusual derepression of the acetohydroxyacid synthetase in response to an isoleucine limitation was noted.

The antibiotic ketomycin, produced byStreptomyces antibioticus, was first described(W. D. Celmer, Belgian Patent 644682, 1964) asa 3-cyclo-hexeneglyoxylic acid. Keller-Schier-lein et al. reported that ketomycin inhibits thegrowth of Bacillus subtilis in a chemicallydefined medium and that the inhibition can bereversed competitively by some biosyntheticintermediates, leading to formation of thebranched-chain amino acids (6). Isoleucine wasreported to reverse the inhibition noncompeti-tively. Specifically, ketomycin was demon-strated to inhibit, competitively, the conversionof a-keto-,B-methylvalerate to isoleucine bymeans of an amino acid dehydrogenase activity.In the same studies, Keller-Schierlein et al.found that B. subtilis can convert ketomycin to3-cyclohexeneglycine, which inhibits growthand can be counteracted by the same com-pounds that reverse ketomycin inhibition. Thesynthesis of 3-cyclohexeneglycine and its prop-erties as an isoleucine antagonist in Escherichiacoli were reported earlier in studies of growthinhibition (5). 3-Cyclohexeneglycine competi-tively inhibits threonine deaminase in a crudeextract, but it is only slightly inhibitory toisoleucyl-transfer ribonucleic acid (tRNA) syn-thetase even when present in a 10-fold higherconcentration than isoleucine (6).

Ketomycin-resistant mutants of B. subtilishave been isolated and characterized. Noneexcreted metabolites (amino acids or a-ketoacids), other than isoleucine, that antagonize

IPresent address: Department of Microbiology, MeharryMedical College, Nashville, Tenn 37208.

the action of the analogue (K. Poralla, personalcommunication). All of the isoleucine excretersshowed resistance to 3-cyclohexeneglycine aswell. Ketomycin, although its antagonism isreversed by all of the keto acids and amino acidsin branched-chain amino acid biosynthesis, ap-pears to exert its toxic effect only after itsconversion to 3-cyclohexeneglycine, an ana-logue of isoleucine.

In E. coli K-12, transaminase B is responsiblefor the conversion of a-keto-,3-methylvalerate toisoleucine. We have selected a spontaneous,ketomycin-resistant mutant of a strain of E. coliderived from strain K-12. The biochemicallesion is in transaminase B and leads to excre-tion of a-keto acids. The mutant remains sensi-tive to 3-cyclohexeneglycine. The observationssupport a mechanism of resistance character-ized by a limited capacity to convert ketomycinto 3-cyclohexeneglycine, which is the toxicproduct of ketomycin metabolism.

MATERIALS AND METHODSOrganisms and cultivation conditions. The

strains used in this study are listed in Table 1. E. colistrain CU2001, a rbs- derivative of strain K-12(prepared by D. McGilvray), was employed as theparent strain. A mineral salts-glucose medium (mini-mal medium) previously described (12) was used,with supplements as described below, for all studiesrequiring cell growth.Mutant selection technique. Fifty separate tubes

containing 1 ml of minimal medium were inoculatedfrom single colonies of strain CU2001 growing onminimal agar plates. These tubes were incubated for24 h at 37 C in a tissue culture roller drum (NewBrunswick Scientific Co., New Brunswick, N.J.) to

510

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.

KETOMYCIN-RESISTANT MUTANT 511

TABLE 1. Strains used in this study

Strain Relevant genotype Use Source or reference

E. coliK-12 Wild-type prototroph Excretion indicatorCU2 ilvE12 Genetic mapping; excretion Formerly 11A16 (16)

indicatorCU8 iluA451 Genetic mapping; excretion Formerly JHM544 (16)

indicatorCU15 ilvD452 Genetic mapping Formerly 20A19 (16)CU2001 rbs-215 Parent of ketomycin-resistant D. McGilvray

strainsS. typhimurium

leu-477 leu V ABCD Excretion indicator P. Margolin

yield a cell density of approximately 109/ml. A 0.1-mlamount of each culture was transferred to a minimalagar plate containing 50 jug of ketomycin per ml.Spontaneous mutants resistant to ketomycin ap-peared after incubation at 37 C for 2 to 4 days. Onecolony from each plate on which growth was observedwas purified by single-colony isolation and tested forexcretion properties by its ability to feed strainsauxotrophic for branched-chain amino acid biosyn-thesis. Auxanographic tests were performed by in-cluding auxotrophs in minimal agar plates and inocu-lating the resistant mutants on the plate surface.Growth of an auxotroph in the agar in the immediatevicinity of the mutant constituted a positive test forexcretion.

Preparation of crude extracts and enzymaticassays. Crude extracts were prepared as previouslydescribed from exponential-phase cultures of thestrains being examined (13). Protein determinationswere made by the biuret method (8).

Threonine deaminase activity was measured by thecolorimetric method described by Szentirmai andUmbarger (13) and acetohydroxy acid synthetaseactivity was measured by the method of St0rmer andUmbarger (13). The colorimetric procedure of Parsonsand Burns (9) was used to assay ,B-isopropylmalatedehydrogenase activity. A colorimetric assay of trans-aminase B activity was accomplished by use of amodification of the procedure of Duggan et al. (3). Inthis procedure, the 2, 4-dinitrophenylhydrazone ofa-keto-fl-methylvalerate was extracted as it wasformed. A 1-ml enzyme assay mixture contained 100Almol of tris(hydroxymethyl)aminomethane-hydro-chloride buffer (pH 8.0), 0.1 Amol of pyridoxal phos-phate, 15 gmol of a-keto-glutarate, 50 ;1mol of L-valineor 25 gmol of L-isoleucine, 1 or 2 mg of crude extractprotein, and distilled water. A control mixture con-tained everything except the substrate valine orisoleucine. Assay mixtures were incubated at 37 C for15 min and the reactions were terminated by additionof 0.1 ml of 50% trichloroacetic acid. A suitably sizedsample of this mixture was diluted to 1.0 ml andmixed with the 2,4-dinitrophenylhydrazine reagent(3) to initiate the formation of the keto acid deriva-tive.The 5-min extraction of the 2,4-dinitrophenylhy-

drazones during their formation recommended byDuggan et al. (3) was accomplished by placing tubes

containing the extraction components in a test tuberack fastened to the center vibrator shaft of a RotaiyEvapo-Mix tube flash evaporator (Buchler Instru-ments, New York, N.Y.). This apparatus created avortex action in each tube. The speed of rotation wasadjusted so that an emulsion was clearly evident forthe duration of the extraction period. Routinely, 12tubes were conveniently handled by this method.

Portions of the toluene layers were removed andextracted with Na2CO3 in the same manner describedabove and centrifuged at low speed to separate thelayers. Samples of the Na2CO3 layers were removedand transferred to another test tube, and NaOH wasadded to develop the color. After 10 min, the colorintensity was measured spectrophotometrically at 540nm. Standards prepared from a-keto-,B-methylvaler-ate and a-keto-isovalerate were treated by the sameextraction procedure. The amino-acyl-tRNA synthe-tase activities for the branched-chain amino acidswere assayed by procedures described earlier (2).

Transductions. All transduction experiments wereconducted with the transducing phage P1 accordingto the method described by Lennox (8). Strain con-struction was performed with ultraviolet-irradiatedP1 phage.

Chemicals. All commercially available chemicalsused in the study were of reagent grade or of thehighest purity otherwise obtainable. K. Poralla andW. Shieve kindly supplied ketomycin (3-cyclohex-eneglyoxylic acid) and 3-cyclohexeneglycine, respec-tively.

RESULTSMutant selection and growth response. On

the basis of their excretion properties, fivedistinct classes of spontaneous mutants ofstrain CU2001 that are resistant to 50 jg ofketomycin per ml have been obtained. Theclasses are listed in Table 2. Isolates of class 1which did not excrete material feeding any ofthe test organisms were most common (33 of47). Strain CU2588, the mutant described inthis study, was the sole isolate from class 5. Itexcreted products satisfying the growth-factorrequirements of isoleucine, valine, and leucineauxotrophs. The growth rate of strain CU2588 in

VOL. 3, 1973

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.

512 JACKSON AND UMBARGER

TABLE 2. Classes of mutants

Class Feed Inhibit Frequency

1 None 33/472 ilvA 10/473 leuV 2/474 leuV K-12 1/475 leuV K-12 1/47

ilvA

unsupplemented minimal medium was lessthan that of the parent strain CU2001 (Fig. 1).Addition of isoleucine alone restored the normalgrowth rate. However, addition of a-aminobuty-rate, which is readily converted to the isoleucineprecursor a-ketobutyrate, failed to restore anormal growth rate to the mutant. The ketomy-cin-resistant mutant retained a sensitivity tovaline but appeared to be slightly less sensitivethan the parent strain.Phenotypic response of the isoleucine-, va-

line-, and leucine-forming enzymes to therepression signals. The relationship of thestructural genes for the biosynthetic enzymes,their regulatory loci, and the enzymes involvedin the biosynthesis of the branched-chain aminoacids is shown in Fig. 2. One enzyme from eachof the three operons involved in isoleucine,valine (11), and leucine biosynthesis (with theexception of the inducible isomeroreductase [1 ])was examined to assess what effects the muta-tion to ketomycin resistance had upon regula-tion of synthesis of the biosynthetic enzymes(Table 3). Consistently, threonine deaminaseand acetohydroxy acid synthetase in the ke-tomycin-resistant strain CU2588 were observedto have two- to threefold greater specific activ-ity than the parent strain, CU2001, when grownin unsupplemented minimal medium. 6l-Iso-propylmalate dehydrogenase appeared to beslightly elevated (50%) in the mutant. Whengrown in the presence of L-isoleucine (4 x 10-4M), L-valine (1.2 x 10-3 M), and L-leucine (4 x10-4 M), these enzymes were repressed to pa-rental, wild-type repressed levels. Threoninedeaminase and acetohydroxy acid synthetase instrain CU2588 displayed normal response tofeedback inhibition by isoleucine and valine,respectively. Examination of the branched-chain amino-acyl-tRNA synthetase activitiesyielded no evidence of enzyme alterations.Thus, the derepression of the isoleucine-valinebiosynthetic enzymes was not due to alteredamino acid-activating enzyme activities nor tobiosynthetic enzymes with altered response tofeedback inhibition. The low-grade derepres-sion was the only difference noted between themutant and wild type. It seemed unlikely that

the analogue resistance could be due solely tothe two- to threefold derepression observed.Linkage of ketomycin resistance to ilv to

P1 transduction. Phage lysates were preparedon strain CU2588 and used to determine co-transduction frequency of the excretion proper-ties (Exc) accompanying transfer of the resist-ance marker, which was designated ilv-503(T'able 4). Linkages by co-transduction fre-quency to ilvA451, ilvD452, and ilvE453 were96, 98, and 100%, respectively. The rbs markerin strain CU2588 was also used as an unselectedmarker. The co-transduction frequencies of therbs marker given in Table 4 were in goodagreement with results normally obtained inthis laboratory. The mutation for ketomycinresistance lies, therefore, very closely linked tothe ilv region of the E. coli chromosome. Fur-thermore, among 200 ilvE+ recombinants, allexhibited the Exc+ character of the donor(Table 4).

0

czaw

I-

w-J

TIME (HOURS)

FIG. 1. Effects of exogenous L-isoleucine and a-aminobutyrate on growth of the ketomycin-resistantstrain CU2588 compared to the growth of strainCU2001 in minimal medium. Symbols (strainCU2588): 0, minimal medium; A, minimal mediumcontaining 5 x 10-' M L-isoleucine; 0, minimalmedium containing 5 x 10-4 M a-aminobutyrate.Symbol (strain CU2001): 0, minimal medium. Klettreadings were made with a no. 42 filter. The lag notedin the growth of strain CU2001 was due to the factthat the inoculum was taken from a culture that hadreached the stationary phase, whereas the inoculumof the slower growing strain CU2588 was still inexponential phase when removed.

ANTIMICROB. AG. CHEMOTHER.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.


TRAa-Ketoisocoproate T Leu.cine

IPMDR- lsopropylmolate

ISOPyruvate a- Isopropylmolote

IPMS----....... -

AHSII_ IR DHTRCa-Acetolactate R a-/3-Dihydroxy - a-Ketoisovalerote Valine

AHSI isovolerate

Pyruvate I;.

AHS I IR D RAHSl a-Acetohydroxy- - * a-/3-Dihydroxy -H a-Keto-3- nmethyl- -b Isoleucine

-/ b butyrate 83-methylvalerate volerate

a-KetobutyrateTD

Threonine

ilv

P B Q C Y O A

SI .AHI IR

D EI , .

TD DH TRB

leu

O A B C D-7 / I ' I' I I I

II , I

IPMS IPMD ISO

FIG. 2. Biosynthetic pathways for isoleucine, valine, and leucine. The enzymes catalyzing the indicatedsteps are abbreviated as follows: TD, threonine deaminase (EC 4.2.1.16, L-threonine hydrolase[deaminating]); AHS I, end product-inhibited acetohydroxy acid synthetase; AHS II, end product-noninhib-ited acetohydroxy acid synthetase; IR, acetohydroxy acid isomeroreductase; DH, dihydroxy acid dehydrase;TRA, transaminase A; TRB, transaminase B; TRC, transaminase C; IPMS, a-isopropylmalate synthetase;ISO, isopropylmalate isomerase; IPMD, 8-isopropylmalate dehydrogenase. The ilv and leu genes correspondingto these enzymes are indicated below the scheme. Genes ilvP, ilvO, and leuO are repression recognition sites,ilvQ is the induction recognition site, and ilvY is the gene for the positive control element needed for inductionof isomeroreductase activity. The gene order in each cluster is the reverse (i.e., left to right is counterclockwise)of the way it is usually represented on the E. coli chromosome map. Adapted from Taylor (14) and Pledger andUmbarger (10).

Several Ilv+, Exc+ transductants were se-lected from the cross with strain CU8 (ilvA451)as the recipient, and two (strains CU8506 andCU8507) were examined for response of theisoleucine-valine biosynthetic enzymes underconditions of growth in minimal medium and inminimal medium supplemented with isoleu-cine, valine, and leucine (Table 5). Threoninedeaminase and acetohydroxy acid synthetaseactivities in both strains CU8506 and CU8507were about twice those in the parent (strainCU2001) when grown in minimal medium. Re-pressed levels of the isoleucine and valine bio-synthetic enzymes were normal in all threestrains. Thus, the phenotypic properties at-tributed to the mutation causing ketomycinresistance appear to be tightly linked to the ilvregion by P1 transduction. Furthermore, excre-

tion properties are a good indication of thepresence of the ilv-503 lesion.Examination of culture filtrates for evi-

dence of excreted isoleucine, valine, and leu-cine. Cell-free culture filtrates of strain CU2001and strain CU2588 were passed over a Dowex-50column (H+-form) to adsorb amino acids. Thepurpose was to distinguish whether excretionproducts from strain CU2588 were the keto acidprecursors (a-keto-fl-methylvalerate, a-ketoiso-valerate, and a-ketoisocaproate) or the respec-tive amino acids (isoleucine, valine, and leu-cine). The fraction binding to Dowex-50 waseluted with NH,OH, evaporated to dryness, andredissolved. This fraction, potentially contain-ing amino acids but not keto acids, was testedfor the presence of isoleucine, valine, and leu-cine by ability to feed the respective auxo-

VOL. 3, 1973

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.


TABLE 3. Phenotypic response of isoleucine, valine,and leucine biosynthetic enzymes in the mutant and

parent

Specific activity(pmol per min per mg)

Aceto- a-Iso-Strain Growth medium" Threo- hydroxy propyl-nine. ai malate

deami- syactde- dehy-nase tase drogen-

ase

CU2001 Minimal 0.076 0.056 0.108Supplemented 0.020 0.011 0.028

CU2588 Minimal 0.216 0.145 0.156Supplemented 0.036 0.014 0.034

a Supplemented medium contained L-isoleucine (5x 10-4 M), L-valine (1.2 x 10-3 M), and L-leucine (5x 10-4 M).

TABLE 4. Linkage data for P1 transductionsa

liv+Selected recon- Percent PercentCross no. marker binants Rbs+ Exc+ b

examined

1 ilvA + 500 90 962 ilvD+ 285 93 983 ilvE+ 200 93 100

a P1 lysates were prepared on strain CU2588 (ilv-503, ketomycin resistant). Recipient strains in thecrosses were (1) CU8 (ilvA451), (2) CU15 (ilvD452),and (3) CU2 (ilvE12).b Colonies found to excrete products inhibiting strainK-12 or feeding an ilvD auxotroph.

trophic mutants (Table 6). The results showthat a concentrated culture filtrate of strainCU2588 applied to agar plates seeded with theindicated auxotrophs supported the growth ofilvA and ilvD mutants but not that of an ilvEmutant, and it inhibited the growth of strainK- 12 (sensitive to valine). However, the fractionof culture filtrate that was bound to Dowex-50did not stimulate growth of any of the auxo-trophs nor did it inhibit the growth of strainK-12. This observation, together with the factthat strain CU2588 did not feed an ilvE mutantbut did feed an ilvD mutant, suggests that theketo acid precursor (a-keto-fl-methylvalerate)of isoleucine is excreted rather than isoleucine.The data suggesting excretion of a-keto acid

analogues of isoleucine and valine, the inabilityof isoleucine precursors to restore the normalgrowth rate to strain CU2588 whereas isoleucinedoes, and the genetic linkage data all indicatedthat the site of the mutation might be the ilvEgene specifying transaminase B.

Examination of transaminase B activity.The parent strain CU2001 and the ketomycin-resistant mutant CU2588 were both assayed fortransaminase B activity. Transaminase B activ-ity in crude extracts of strain CU2588 wasbarely measurable (Table 7). Higher proteinconcentrations in the extracts resulted in higherspecific activities of the mutant enzyme. Prepa-ration of the mutant extract at room tempera-ture instead of 0 C, or growth of the cells at 30 Cinstead of 37 C, did not show any stabilizingeffect on the activity.Response of the branched-chain amino

acid biosynthetic enzymes in the ketomycin-resistant strain to exogenous isoleucine.Growth studies showed that strain CU2588growing in minimal medium underwent a slightisoleucine limitation which was evidenced by alonger generation time than that of the parent,and that the normal growth rate was restored bythe addition of isoleucine alone. The presence ofa reduced transaminase B activity is consistentwith an isoleucine limitation. It was of obviousinterest to determine whether the other pheno-typic properties of the mutant were also re-stored to normal by the presence of isoleucine

TABLE 5.Examination of ilv phenotype in ilv+transductants which feed an ilvA, ilvD, and a leu

deletion auxotroph when a P1 lysate of CU2588 wasthe donor and CU8 the recipient

Specific activity

Strain Growth mediuMa Threonine Acethy-deaminase droxy acid

synthetase

CU2001 Minimal 0.076 0.056Supplemented 0.031 0.011



aSee Table 3.

TABLE 6. Feeding of indicator strains by culturefiltrates

Fed by culture fluid

Indicator strain CU2588CU2001 CU2588 after

Dowex-50

CU8 (iluA). - +CU15 (ilvD) ....... +CU2 (ilvE) ........K-12.. Inhibited

ANTIMICROB. AG. CHEMOTHER.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.


TABLE 7. Transaminase B activity in strains CU2001and Cu2588

Protein Transami-Strain Growth mediuma (mg) nase B(m) activity

CU2001 Minimal 0.94 0.043Minimal 1.88 0.042Supplemented 1.10 0.019Supplemented 2.19 0.018

CU2588 Minimal 1.00 0.0003Minimal 2.00 0.001Supplemented 1.27 0.0001Supplemented 2.54 0.001

aSee Table 3.

during growth. The results of such an experi-ment showed the derepression noted earlier forenzymes under control of ilvO, ilvP, and leuO instrain CU2588 growing in minimal medium(Table 8). The presence of isoleucine essentiallyrestored the enzyme activities of the ketomycin-resistant mutant to levels of activity observedfor the parent strain under the same growthconditions. In the presence of all three of thebranched-chain amino acids, the operons inboth the parent and the mutant were repressedto the same levels.

Effects of ketomycin and 3-cyclohexene-glycine on growth. Since the mutation causingresistance to ketomycin in strain CU2588 lies inthe ilvE region, one might predict that thisstrain should remain sensitive to cyclohexene-glycine. Minimal agar plates seeded with theparent strain, CU2001, and with the mutantstrain, CU2587, were used to test these strainsfor sensitivity to ketomycin and 3-cyclohex-eneglycine. The parent strain displayed largezones of growth inhibition by crystals of ke-tomycin and by crystals of cyclohexeneglycineplaced on the surface of the seeded plate. Themutant strain CU2588 showed no zone of inhibi-tion in the presence of ketomycin but displayedthe same inhibition by cyclohexeneglycine thatwas observed in the parent strain.

DISCUSSIONThe lesion (ilvE503) causing resistance to

ketomycin in strain CU2588 has resulted in atransaminase B activity that is very low in vitro.Because of the instability of the altered trans-aminase B, it has not been possible to deter-mine the precise manner in which the enzymehas changed. It is assumed that a similar, lowlevel of activity occurs in vivo. The extent towhich an isoleucine limitation occurs is gov-erned by the in vivo rate of conversion of

a-keto-B3-methylvalerate to isoleucine. Sincethe cells grow slowly in minimal medium andisoleucine addition restores the wild-typegrowth response, an isoleucine limitation obvi-ously occurs in vivo. Growth studies indicatethat only isoleucine is limiting, presumablybecause it is entirely dependent upon transami-nase B activity for the final step in its synthesis.Appreciable valine and leucine limitations maynot occur, since the transaminase C convertsa-keto-isovalerate to valine and transaminase Aconverts a-keto-isocaproate to leucine.

Derepression of the enzymes controlled by therepression recognition site, ilv 0, is expectedbecause of an isoleucine limitation. Conse-quently, the malfunction of transaminase Bresults in an accumulation of a-keto-fl-methyl-valerate which may enhance resistance to ke-tomycin by competing for the active site on thealtered transaminase B. Such a model for resist-ance predicts that ketomycin is not converted tocyclohexeneglycine in vivo; therefore, strainCU2588 should remain sensitive to cyclohex-eneglycine. Retention of cyclohexeneglycinesensitivity was indeed observed. It is thereforeconcluded that a low level of transaminase Bactivity is one mechanism by which E. coli maybecome resistant to the action of ketomycin.

Less easily explained in view of earlier obser-vations (4) is the derepression of the ilvP-con-trolled acetohydroxy acid synthetase. This en-zyme has been shown to respond multivalentlyto limitations of valine and leucine but notisoleucine (4). However, the derepression ofacetohydroxy acid synthetase observed in strain

TABLE 8. Effect of exogenous L-isoleucine onexpression of the isoleucine, valine, and leucine

biosynthetic operons

Specific activity(Amol per min per mg)

Aceto- a-Iso-Strain Growth mediuma Threo- hydy propyl-nine achid malate

deami- synthe- dehy-nase tase drogen-

ase

CU2001 Minimal 0.088 0.084 0.127Isoleucine 0.069 0.086 0.100Supplemented 0.028 0.014 0.032

CU2588 Minimal 0.219 0.252 0.170Isoleucine 0.065 0.097 0.118Supplemented 0.024 0.014 0.030

aIsoleucine medium was the minimal mediumcontaining 5 x 10-' M L-isoleucine. The supple-mented medium was like that described in Table 3.

VOL. 3, 1973

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.


CU2588 is restored to normal levels by thepresence of isoleucine. An obvious extrapolationis that, as a result of the mutation, the ilVB generesponds to an isoleucine-limiting signal instrain CU2588 when cultured in minimal me-dium. Since this effect has not been observed inwild-type K-12 strain, one possible explanationis that the ilvE503 lesion involves a regulatorysite that lies within or near ilvE, and that such aregulatory site functions in control of synthesisof a valine-sensitive isozyme of acetohydroxyacid synthetase.

ACKNOWLEDGMENTSWe express our appreciation to K. Poralla for the gift of

ketomycin and to W. Shive for his donation of 3-cyclohex-eneglycine. We are particularly indebted to Dennis Dugganfor describing the assay method for keto mixtures prior topublication. The technical assistance of J. Abron is gratefullyacknowledged.

This work was supported by Public Health Service grantGM12522 from the National Institute of General MedicalSciences. J.H.J. received Public Health Service PostdoctoralFellowship 5 F02 GM35763-02. This is paper XXIII in theseries "Isoleucine and Valine Metabolism in Escherichiacoli."

LITERATURE CITED1. Arfin, S. M., B. Ratzkin, and H. E. Umbarger. 1969. The

metabolism of valine and isoleucine in Escherichia coli.XVII. The role of induction in the derepression ofacetohydroxy acid isomeroreductase. Biochem. Bio-phys. Res. Commun. 37:902-908.

2. Blatt, J. M., and H. E. Umbarger. 1972. On the role ofisoleucyl-tRNA synthetase in multivalent repression.Biochem. Genet. 6:99-118.

3. Duggan, D. E., and J. A. Wechsler. 1973. An assay fortransaminase B enzyme activity in Escherichia coliK-12. Anal. Biochem. 51:67-79.

4. Dwyer, S. B., and H. E. Umbarger. 1968. Isoleucine andvaline metabolism of Escherichia coli. XVI. Pattern ofmultivalent repression in strain K-12. J. Bacteriol.95:1680-1684.

5. Edelson, J., J. D. Fisaekis, C. G. Skinner, and W. Shive.1958. 3-Cyclohexene-1-glycine, an isoleucine antago-nist. J. Amer. Chem. Soc. 80:2698-2700.

6. Keller-Schierlein, W., K. Poralla, and H. Zahner. 1969.Stoffwechselprodukte von Mikroorganismen. Isolie-rung, Identifizierung und Wirkungsweise von Ketomy-cin [(R)-3-CyclohexenylglyoxylsalureJ und dessen Um-wandlungsprodukt 3-Cyclohexenylglycin. Arch. Mik-robiol. 67:339-356.

7. Layne, E. 1957. Spectrophotometric and turbidimetricmethods for measuring proteins, p. 447-454. In S. P.Colowick and N. 0. Kaplan (ed.), Methods in en-zymology, vol. 3. Academic Presa Inc., New York.

8. Lennox, E. S. 1955. Transduction of linked geAeticcharacters of the host by bacteriophage P1. Virology1:190-206.

9. Parsons, S. J., and R. 0. Burns. 1969. Purification andproperties of fi-isopropylmalate dehydrogenase. J. Biol.Chem. 244:996-1003.

10. Pledger, W. J., and H. E. Umbarger. 1973. Isoleucine andvaline metabolism in Escherichia coli. XXII. A pleio-tropic mutation affecting induction of isomeroreduc-tase activity. J. Bacteriol. 114:195-208.

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

12. Stormer, F. C., and H. E. Umbarger. 1964. The require-ment for flavin adenine dinucleotide in the formation ofacetolactate by Salmonella typhimurium extracts. Bio-chem. Biophys. Res. Commun. 17:587-592.

13. Szentirmai, A., and H. E. Umbarger. 1968. Isoleucine andvaline metabolism of Escherichia coli. XIV. Effect ofthiaisoleucine. J. Bacteriol. 95:1666-1671.

14. Taylor, A. L. 1970. Current linkage map of Escherichiacoli. Bacteriol. Rev. 34:155-175.

15. Umbarger, H. E., and E. A. Adelberg. 1951. The role ofa-keto-fl-ethyl butyric acid in biosynthesis of isoleu-cine. J. Biol. Chem. 192:883-889.

ANTiMICROB. AG. CHEMOTHER.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 23

Oct

ober

202

1 by

42.

98.1

46.2

45.

defective transamination, mechanism resistance to ketomycin in

Documents