JOuRNAL OF BACRIOLOGY, Jan. 1976, p. 94-101Copyright @ 1976 American Society for Microbiology

Vol. 125, No. 1Printed in U.S.A.

Regulatory Mutants and Control of Cysteine BiosyntheticEnzymes in Salmonella typhimurium

PEGGY R. BORUMl* AND KENNETH J. MONTYDepartment of Biochemistry, University of Tennessee, Knoxville, Tennessee 37916

Received for publication 12 September 1975

The cysB rgion in Salmonella typhimurium regulates in a positive manner thenoncontiguous structural genes for the enzymes responsible for sulfate reductionin cysteine biosynthesis. We treated three cysB mutants with chemical mutagensand selected 81 secondary mutants in which the inability to utilize sulfate wassuppressed. Growth experiments on the suppressed mutants showed that theoriginal loss of sulfate utilization had been corrected to varying degrees and thatportions of the pathway had been established in abnormal relationship to oneanother. Sixty of the suppressed mutations were mapped via transductionalanalysis, and each was very closely linked to the original cysB mutation. Wedemonstrated that the cysB product functions in the regulation of the cysteinebiosynthetic enzymes during both logarithmic growth and stationary phase.Mutation can alter the regulatory response of one enzyme in either an upward ordownward direction while the regulation of other enzymes in the pathwayremains unchanged. These data are consistent with the idea of a multivalent ormultisite regulator molecule.

Mizobuchi et al. (13) demonstrated that the14 cistrons involved in cysteine biosynthesis aredivided into five discrete groups widely scat-tered over the linkage map as illustrated in Fig.1. A scheme of the biosynthetic pathway ispresented in Fig. 2. The reductive enzymes arerepressed by growth on cysteine, and expressionof the structural genes for all the reductiveenzymes is dependent upon the integrity of thecysB region (the positive regulatory region) andthe presence of O-acetylserine (OAS) (3, 4; H.Spencer, J. Collins, and K. J. Monty, Fed. Proc.26:677, 1967). The results reported here demon-strate that for 60 suppressed mutants, thecorrecting mutation is very closely linked to theoriginal mutation and probably is locatedwithin the cysB region. Regulatory control of thecysteine biosynthetic pathway is examined atthe biochemical level by growth of bacteria onvarious sulfur sources and assay of some of theenzymes of the pathway. Several patterns ofaberrant control are documented.

MATERIALS AND METHODSBacterial strains. Salmonella typhimurium wild-

type strain LT2 and the cysteine auxotrophs cysBa25,cysBbl2, and cysBc482 were first described byDemerec and Hartman (2) and were later mapped byMizobuchi et al. (13). The pyrl l 7 marker is the result

'Present address: Department of Biochemistry, Vander-bilt University School of Medicine, Nashville, Tenn. 37232.

94

of a mutation induced by fast neutrons in the pyrFgene (7). The trp- marker was produced earlier in thislaboratory, and phenotype studies of the mutant indi-cate that the mutation is either trpE or trpB (J.Arthur and K. J. Monty, unpublished data). Thehisl35 marker, obtained from Philip Hartman, is adeletion in the E and F genes of the histidine operonthat arose spontaneously from pro2l+ (6, 7).

Bacteriophage strain. The temperate phagePLT-22 (H-1) was used for all transductional experi-ments (5, 17).Media and solutions. T-2 buffer was prepared by a

modification of the procedure of Hershey and Chase(8). Minimal medium has been described previously(2). Nutrient broth and nutrient agar were from Difco.Production and isolation of suppressor mutants.

Overnight nutrient broth cultures of cysBa25,cysBbl2, and cysBc482 were harvested by centrifuga-tion, washed, and resuspended in 2 ml of saline. Formutation by 2-aminopurine (2-AP), 0.2 ml of a4-mg/ml solution of 2-AP and then 0.1 ml of one of thesuspensions of a cysB mutant were spread onto sulfateminimal medium. For mutation by ethyl methanesulfonate (EMS) 0.1 ml of the above suspension of acysB mutant was spread onto sulfate minimal me-dium, and a sterile filter disk was placed on the centerof the plate with two drops of EMS deposited on thedisk. The mutant strains were derived and designatedin the following manner: strains Ba25-PR1 throughBa25-PR20 from EMS treatment of cysBa25; strainsBa25-PR21 through Ba25-PR30 from 2-AP treatmentof cysBa25; strains Bbl2-PR101 through Bbl2-PR120from EMS treatment of cysBbl2; strain Bbl2-PR121from 2-AP treatment of cysBbl2; strains Bc482-

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

REGULATORY MUTANTS IN S. TYPHIMURIUM 95Theosulfote

Sulfate Sulfote permease Sulfate ATP sulfurylase APS kinase PAPS JSulfite \outside inside b APS / PAPS -- Sulfite Slfidecel Binding protein cel ATP A P reductose reductose

0 - ocetyserineL-Cysteine

SuIhydrylase

SerineL - Serine + Acetyl - CoA - O-Acetyl - L -seine

transocetyloe

FIG. 1. Linkage map of Salmonella typhimurium showing the cysteine loci. Abbreviations: APS, Adenosine5'-phosphosulfate; PAPS, 3'-phosphoadenosine 5'-phosphosulfate.

E Serine + acetyl - CoA-e.O-acetylserine

FIG. 2. Cysteine biosynthetic pathway. Abbrevia-tions are as in Fig. 1.

PR201 through Bc482-PR220 from EMS treatment ofcysBc482; and strains Bc482-PR221 through Bc482-PR230 from 2-AP treatment of cysBc482.

Determination of growth rates. Growth rateswere determined by using minimal medium andsulfur sources at a concentration of 10-4 M. Sulfursources used were cysteine, cysteine sulfinic acid(CSA), sodium thiosulfate, and sodium sulfate.Growth on sulfide was examined at two differentconcentrations of sodium sulfide (10-4 and 5 x 10-4M) with incubation in tubes tightly capped withpolypropylene stoppers to prevent the escape of sul-fide via the vapor phase.

Transduction experiments. Two-factor crosseswere performed with one marker selected and thetransductants were scored for the presence of an

unselected marker (9, 17).Construction of the sulfide trap. During transduc-

tional mapping, it was discovered that some recombi-nant cells containing the PR mutations were produc-ing hydrogen sulfide, which supported growth of therecipient cells. In an effort to trap the sulfide and

prevent this vapor-phase feeding, a lead acetate trapwas placed in the half of a petri dish not containingthe agar. The trap consisted of circles of filter paperdipped in a solution of 0.197 g of lead acetate per ml of50% glycerol-50% water.Growth and harvest of cultures for enzymatic

assay. Sulfate-containing minimal medium was usedwhen sulfate was provided as the sole sulfur source.Sulfate is present in this medium at a concentrationof 1.7 x 10-' M. When using other sulfur sources,sulfurless minimal medium was prepared and thesterile sulfur source was added to the sterile medium.Sulfur sources used were L-djenkolic acid (2 x 10-4M), CSA (10-4 M), and L-cysteine hydrochloridehydrate (2 x 10-4 M). The former two sulfur sourceswere sterilized by filtration, and the cysteine wassterilized by autoclaving.An overnight culture was used to inoculate 600 to

1,100 ml of medium to a cell density of 10s/ml in acotton-stoppered 2,800-ml Fernbach flask. The cul-ture was grown at 37 C with forced aeration. Unlessotherwise specified, cultures were harvested by cen-trifugation during mid-log phase of growth (5 x 108 to6 x 108 cells/ml). Storage of frozen cells over a periodof several months appeared to have no effect on thelevel of enzymatic activity exhibited by the crudeextracts prepared subsequently.

Preparation of crude extract for enzymaticassay. Frozen pellets of cells were thawed and resus-pended in 5 ml of 0.05 M sodium phosphate buffer,pH 7.0. Cells were broken by sonication with aBranson Sonifier model 5110 at a setting of 4 A. A100,000 x g clear supernatant was used as crudeextract for enzymatic assay. Crude extracts werealways prepared on the day of assay and were kept at0 to 4 C.

Determination of protein concentration. Proteinconcentration was determined by the microbiuretmethod (10). Solutions of bovine serum albumin wereused as reference standards.Assay for sulfite reductase (EC 1.8.1.2). Sulfite

reductase activity was determined by measuring thesulfide produced from a saturating concentration ofsulfite. The reaction mixture consisted of 50 umol ofsodium phosphate buffer, pH 7.7, 5 Mmol of glucose-6-phosphate, 100 nmol of nicotinamide adenine dinu-cleotide phosphafe, 0.5 units of glucose-6-phosphatedehydrogenase, 5 ,sg of flavine adenine dinucleotide, 1mol of sodium bisulfite, crude extract, and water to

VOL. 125, 1976

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

96 BORUM AND MONTY

yield a total volume of 1.0 ml. The reaction wasstarted with the addition of sodium bisulfite, and thetube was capped with a polyethylene stopper toprevent loss of hydrogen sulfide. After incubation for10 min at 37 C, the reaction was stopped by additionof the acidic reagents for sulfide determination usedin the method of Siegel (16). One milliunit of sulfitereductase is defined as the amount of enzyme catalyz-ing the formation of 1 nmol of sulfide per min underthese assay conditions.Assay for OAS sulfhydrylase (EC 4.2.99.8). OAS

sulfhydrylase activity was determined by a modifica-tion of the method of Kredich et al. (1, 12), whichmeasures the cysteine produced from OAS and sul-fide. The reaction mixture contained 20 grmol oftris(hydroxymethyl)aminomethane-hydrochloride, pH7.8, 1 umol of sodium sulfide, 53 Amol of OAS,crude extract, and water to yield a final volume of 1.0ml. The incubation was at room temperature for twomin, and the reaction was stopped by addition of 0.05ml of 6 N sulfuric acid. Three solutions were used fordevelopment of the azo dye. Solution I was 34.5 mg ofsodium nitrite per 10 ml of water made immediatelybefore use; solution II was 0.5 g of ammoniumsulfamate per 100 ml of water; and solution III, madeeach week, was 0.57 g of mercuric chloride, 7.88 g ofsulfanilamide, and 0.12 g of N-1-naphthylethylene-diamine dihydrochloride dissolved in that sequence

in 100 ml of 1.64 N hydrochloric acid. An aliquotof 0.05 ml of solution I was added to the re-

action mixture, which had been treated as describedabove, and the mixture was incubated for 5 min atroom temperature followed by addition of 0.1 ml ofsolution II. After another incubation period of 2 min,0.2 ml of solution III was added. Color developedmaximally within 5 min, and absorbance was deter-mined at 540 nm with a Beckman model DU spectro-photometer. One milliunit of OAS sulfhydrylase isdefined as the amount of enzyme catalyzing theformation of 1 nmol of cysteine per min under theseassay conditions.

RESULTS

Phenotypic expression of suppressedmutants. Strains cysBa25, cysBbl2, andcysBc482 are all unable to utilize sulfate as asulfur source. Thus, any colony growing on asulfate minimal medium plate spread with oneof the above cysB mutants would be the resulteither of a reversion event or of the introductionof a second mutation that suppressed the origi-nal cysB mutation. Strains isolated from suchclones will be designated as suppressed mu-tants.Eighty of 81 such suppressed mutants were

characterized according to their rates of growthwith cysteine, CSA, thiosulfate, or sulfate as thesole sulfur source in liquid media. The 81stmutant (Bc482-PR215) failed to grow in anyliquid medium. Similar growth studies wereperformed on cysBa25, cysBbl2, and cysBc482.

J. BACTERIOL.

None of the cysB mutants exhibited impairedability to utilize sulfide. Suppressed mutantsexhibiting growth rates similar to that of strainLT2 on all sulfur sources used were designatedclass 1 mutants. Suppressed mutants growingat a slower rate or to a lower final populationdensity on sulfate than does strain LT2 weredesignated class 2 mutants. Each of theseclasses could be further subdivided as explainedbelow.Class 1-A mutants. During preliminary

mapping experiments, five suppressed mutantsof the class 1 group produced from cysBc482gave apparent 90+% linkage with both the trpand pyr markers. This aberrant behavior wasthe result of the release of large quantities ofhydrogen sulfide from certain recombinants andcould be controlled by the presence of thesulfide trap. These strains were classified asclass 1-A mutants and are listed in Table 1.

Previous liquid growth studies of class 1-Amutants indicated that their growth rates wereindistinguishable from those of strain LT2.Growth curves on sulfate minimal medium foreach class 1-A mutant and strain LT2 weregenerated with removal of all possible sulfide byconstant vigorous bubbling of sterilized airthrough each tube. Growth rates of class 1-Amutants and strain LT2 were identical; there-fore, the sulfide given off by the class 1-Amutants was not necessary for their optimalgrowth on sulfate minimal medium.Class 1-B mutants. Suppressed mutants in

class 1 that exhibited no sulfide feeding weredesignated class 1-B mutants. Suppressed mu-tants produced from all three original cysB mu-tants had representatives in the class 1-B phe-notype as shown in Table 1.Class 2 mutants. Class 2 mutants included

mutants whose growth on sulfate was noticeablyimpaired compared with the growth of strainLT2. Class 2 mutants could be divided into thefollowing subgroups. Class 2-A mutants demon-strated a wild-type rate of growth on cysteine,CSA, and thiosulfate, but experienced an initiallag in growth on sulfate. Class 2-B mutants gavea wild-type rate of growth on cysteine, CSA,and thiosulfate but grew slowly on sulfate. Class2-C mutants showed a wild-type rate of growthon cysteine, CSA, and thiosulfate but grow verylittle on sulfate. Class 2-D mutants gave awild-type rate of growth on cysteine, a slow rateof growth on CSA and thiosulfate, and verylittle growth on sulfate. Suppressed mutantswithin these classes are listed in Table 1.Genetic analysis. The suppressed mutants

were mapped versus trp and pyr markers. Since

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

REGULATORY MUTANTS IN S. TYPHIMURIUM 97

TABLE 1. Phenotypic classification of suppressed strainsSuppressedmutants Class 1-A Class 1-B Class 2-A Class 2-B Class 2-C Class 2-Dfrom:

cysBa25 Ba25-PR2 Ba25-PR14 Ba25-PR1 Ba25-PR23Ba25-PR4 Ba25-PR18 Ba25-PR3 BA25-PR24

thruBa25-PR7 Ba25-PR20 Ba25-PR5Ba25-PR8 Ba25-PR6Ba25-PR11 Ba25-PR9

thruBa25-PR13 Ba25-PR10Ba25-PR15 Ba25-PR22

thruBa25-PR17Ba25-PR21Ba25-PR25

thruBa25-PR30

cysBbl2 Bbl2-PR101thru

Bbl2-PR121

cysBc482 Bc482-PR223 Bc482-PR201 Bc482-PR216 Bc482-PR221 Bc482-PR226thru

Bc482-PR224 Bc482-PR214 ic482-PR222 Bc482-PR225Bc482-PR227 Bc482-PR217 Bc482-PR230

thru thruBc482-PR229 Bc482-PR220

growth of class 2 mutants on sulfate was varia-ble, growth on thiosulfate was used for map-ping. Selection media for Cys+ were supple-mented with 10-4 M thiosulfate. A solution ofthiosulfate was made fresh daily and spreadonto each plate just before spreading the trans-duction mixture or replica plating. To avoidpossible feeding by escaping sulfide, the sulfidetrap was used in replica plates during themapping of class 2 mutants.

Consistent with the findings of Demerec andHartman (2) using crosses of Trp+ CysB+ xTrp- CysB-, we found that there were usuallymore Trp Cys+ than Trp+ Cys recombinants.Ozeki has suggested that genetic material nearthe ends of a transducing fragment would beincorporated less frequently into the genome ofrecipient bacteria than would material locatednear the center of the transducing fragment(14). When using this system, it is necessary toallow for this unequal incorporation of regionsin the transducing fragment when calculatingmap distances. Unfortunately, the mappingsystem will not separate Ba, Bb, and Bc regionsor show definitively that the suppressed mu-tants map within the cysB region. However, allof the suppressed mutations are definitely very

closely linked to the cysB region. The sup-pressed mutations are about 75 to 90% linked totrp and 45 to 60% linked to pyr.Enzyme levels of strain LT2. It has been

shown many times that growth on cysteine orcystine will lower the activity of sulfite reduc-tase to a level that is not detectable (3, 4).Growth on L-djenkolic acid, sulfate, or CSAgave induced levels of sulfite reductase activity.The levels of sulfite reductase activity found instrain LT2 grown under these different condi-tions are given in Table 2. Most of the en-zymatic activities listed in Table 2 are averagesof several different experiments. All the en-zymes of the reductive part of the pathwayresponded in parallel to each sulfur source;thus, sulfite reductase activity could be used asa representative of the entire reductive pathway(11; Spencer et al., Fed. Proc. 26:677, 1967).Enzyme levels of cysB mutants. The data in

Table 2 show that cysB mutants when grown oncysteine had no detectable levels of sulfitereductase. Although L-djenkolic acid was asulfur source yielding a high level of sulfitereductase activity in strain LT2, cysBc482 hadno detectable sulfite reductase activity whengrown on L-djenkolic acid. This response to

VOL. 125, 1976

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

98 BORUM AND MONTY

TABLE 2. Enzymatic activities during mid-log growth on various sulfur sourcc

Bacterial strain Mutant class Sulfur source OAS sulfhydrylase(mU/mg of protein)

L-Djenkolic acidSulfateCysteine sulfinic acidL-CysteineL-Djenkolic acidCysteine sulfinic acidL-CysteineL-CysteineL-CysteineL-CysteineL-CysteineL-CysteineL-CysteineL-CysteineL-CysteineL-Cysteine

1-A L-Djenkolic acidSulfateCysteine sulfinic acidL-Cysteine

1-A SulfateL-Cysteine

1-A SulfateL-Cysteine

1-A SulfateL-Cysteine

1-A SulfateL-Cysteine

1-B L-Djenkolic acidCysteine sulfinic acidL-Cysteine

2-A Cysteine sulfinic acid2-A Cysteine sulfinic acid2-B Cysteine sulfinic acid2-C Cysteine sulfinic acid2-B Cysteine sulfinic acid2-B L-Djenkolic acid

Cysteine sulfinic acidL-Cysteine

2-C L-Cysteine2-C L-Cysteine2-C L-Cysteine2-C L-Cysteine2-B L-Cysteine2-B L-Cysteine2-B L-Cysteine2-B L-Cysteine2-C L-Cysteine2-D L-Cysteine

5,3145,4413,712593-1,8842,9714,5862,065192295401291257581394

2,429226

6,1574,8105,1173,750

11,11310,10313,76111,6058,74713,2878,593

12,3017,9963,476692-2,0855,5854,3157,2277,3107,8198,3797,7326,839752

1,002890

1,120709788961704929605

Sulfite reductase(mU/mg ofprotein)

9.920.97.9003.80

<0.2<0.100000009.86.29.18.97.58.26.98.95.6

10.05.2

10.610.19.104.44.02.72.62.91.93.51.10000000000

L-djenkolic acid is common to all cysB auxo-

trophs and illustrates the requirement of anintact cysB region for expression of the reduc-tive pathway. Mutant strains cysBc482 andcysBa25 gave slow or leaky growth on CSAmedium. Growth of cysBc482 on CSA yieldedexpression of sulfite reductase but at a level

lower than that of strain LT2 grown under thesame conditions.

All cysB mutants except cysBc482 andcysBc484 when grown on cysteine exhibitedlevels of OAS sulfhydrylase activity that were

the same as or lower than the activity found instrain LT2 grown on cysteine. Strains cysBc482

LT2

cysBc482

cysBbl2cysBa25cysBbl4cysBcl5cysBa24cysBb87cysBb4O3cysBc484cysBbl6Bc482-PR224

Bc482-PR223

Bc482-PR227

Bc482-PR228

Bc482-PR229

Bc482-PR206

Bc482-PR216Bc482-PR222Bc482-PR225Bc482-PR226Bc482-PR230Bc482-PR221

Ba25-PR3Ba25-PR6Ba25-PR9Ba25-PR10Ba25-PR14Ba25-PR18Ba25-PR19Ba25-PR20Ba25-PR22Ba25-PR24

es

J. BACTERIOL.

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

REGULATORY MUTANTS IN S. TYPHIMURIUM

and cysBc484 when grown on cysteine had alevel of OAS sulfhydrylase activity that washigher than that of cysteine-grown strain LT2.The level of activity found in cysBc482 was notincreased in response to growth on L-djenkolicacid or CSA. These data for cysB auxotrophsare presented in Table 2. Since mutations in thecysB region yielded altered levels of expressionof OAS sulfhydrylase, the cysB region appar-ently functions in the control of OAS sulfhydryl-ase. These data are consistent with those ofKredich (11).Enzyme levels of class 1-A mutants. All

class 1-A mutants were the result of a secondmutation in cysBc482. The nonvarying level ofOAS sulfhydrylase that was noted withcysBc482 was retained in the suppressed cysBmutants of class 1-A but at a level of activitythat was higher than that seen in cysBc482.Actually two general levels of OAS sulfhydryl-ase activity were found in class 1-A mutants(Table 2). The level of activity of the sulfhydryl-ase found in Bc482-PR224 was comparable tothe maximum seen in strain LT2, and the levelof activity found in the other four mutants wasnoticeably higher. Although the data for sulfitereductase activity in class 1-A mutants pre-sented in Table 2 are rather scattered, there isno systematic variation. The level of sulfitereductase activity does not appear to be de-pendent on the sulfur source in the growthmedium. In contrast to what has been seen inall other mutants discussed so far, sulfite reduc-tase activity in this class never fell to undetecta-ble levels.Enzyme levels of class 1-B mutants. Sup-

pressed mutants that mimic wild-type behaviorare called class 1-B mutants. The enzymaticactivity of Bc482-PR206 is presented in Table 2as an example of results obtained from this classof mutants. Cultures grown on cysteine gavelower levels of sulfite reductase and OAS sulf-hydrylase activity than those grown on CSA orL-djenkolic acid.Enzyme levels of class 2 mutants. The

enzymatic activities for class 2 mutants areshown in Table 2. The levels of OAS sulflhydryl-ase activity in class 2 mutants that are theresult of a second mutational step in cysBc482were the same as or higher than the maximumlevel found in strain LT2, and varying the sulfursource in the growth medium had no substantialeffect on the level of activity. The level of sulfitereductase of these mutants when grown on CSAor L-djenkolic acid was similar to that ofcysBc482. However, sulfite reductase activitydid not fall to undetectable levels by growth on

cysteine as in the case with strain LT2 andcysBc482. When class 2 mutants that are theresult of a second mutational step in cysBa25were grown on cysteine, they had levels of OASsulfhydrylase activity that were similar to theactivity level found in strain LT2 grown oncysteine.Enzyme levels during stationary phase. As

seen in Table 2, we have some difficulty ingrowing strain LT2 and class 1-B mutants oncysteine and in reproducing the expected lowlevels of OAS sulfhydrylase activity (-600mU/mg of protein). In some experiments thelevel of OAS sulfhydrylase activity was threetimes too high. Kredich has reported similarvariability of enzymatic activity of culturesgrown repeatedly on the same sulfur source andsuggests that it is due to imperfect control ofculture conditions (11). The following series ofobservations will be used to argue that thesedifficulties reflect the attainment by the cells ofa high level of the sulfhydrylase enzyme duringresting phase at population limit and then thevariable dilution of this presynthesized enzymeas the cells are reestablished in logarithmicgrowth.We observed that OAS sulfhydrylase activity

in strain LT2 harvested after attainment ofpopulation limit on cysteine overnight wasmuch higher than the level of activity normallyfound during mid-log growth on cysteine. Bycontrast, sulfite reductase remained at un-detectable levels. We asked whether OAS sulf-hydrylase would reach zero by continuous loga-rithmic growth for five days without exposure tostationary phase. A culture of strain LT2 oncysteine minimal medium was grown to anoptical density at 424 nm of 1.0 and divided intothree portions. The first was used to inoculatean identical medium, the second was used forenzymatic assay, and the third was grownovernight into stationary phase and then usedfor enzymatic assay. This procedure was carriedthrough several cycles. The OAS sulfhydrylasein mid-log-phase cultures of strain LT2 was thelowest seen for the strain, but stationary cul-tures had a very high level (Table 3). The lack ofrigorous control of the physiological state of theinoculum could result in widely varying levels ofOAS sulfhydrylase activity. Sulfite reductaseremained at undetectable levels of activityduring mid-log-phase growth on cysteine anddid not increase during stationary phase. Thereason for the appearance of sulfite reductaseactivity during the 4th and 5th days of growthwas not established.

Strains cysBc482, Bc482-PR224, and Bc482-

VOL. 125, 1976 99

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

TABLE 3. Enzymatic activities during different phases of growth on cysteine minimal medium

OAS sulfhydrylase Sulfite reductase(mU/mg of protein) (mU/mg of protein)

growth Ph Strain cys Bc482 Bc482- Strain cys Bc482 Bc482-LT2 PR224 LT2 PR224

1 Mid-log 229 2,226 3,154 0 0.17 6.47Late-log 2,338 2,433 0 8.23Stationary 5,622 1,823 3,960 0 0 1.31

2 Mid-log 216 1,211 3,818 0 0 5.90Late-log 1,742 4,953 0 10.23Stationary 2,900 1,736 4,490 0 0 0.61

3 Mid-log 276 1,761 4,275 0 0 8.47Late-log 1,958 5,207 0 8.40Prestationary 2,840 6,940 0 7.54Stationary 4,796 2,766 6,395 0 0 0.82

4 Mid-log 342 0Stationary 3,650 0.36

5 Mid-log 641 0Prestationary 2,259 1.8

PR227 were treated in a manner similar to thatdescribed above. Table 3 illustrates the data forcysBc482 and Bc482-PR224. An additional pel-let was harvested at an optical density at 424nm of 1.0 and termed late log phase. The OASsulfhydrylase activity of all three mutants wasnot affected by the phase of growth or bycontinuous log growth without exposure to sta-tionary phase. The enzymatic activity levels ofstrain Bc482-PR227 followed the same patternas those of strain Bc482-PR224. Strain cysBc482maintained a level intermediate between themaximum and minimum levels seen in strainLT2, and strain Bc482-PR227 had a fixed levelapproximately equal to the maximum level ofstrain LT2, whereas the level of activity foundin strain Bc482-PR224 was noticeably higher.Sulfite reductase remained at undetectable lev-els during all phases of growth in the cysBc482strain. The two suppressed mutants had a levelof activity comparable to the maximum wild-type level until stationary phase was reached,but it never reached zero.

DISCUSSIONIt has been shown previously (11; Spencer et

al., Fed. Proc. 26:277, 1967) that the cysBregion in S. typhimurium exerts control in apositive manner over the enzymes concernedwith the biosynthesis of L-cysteine with theexception of serine transacetylase. The six phe-notypic patterns of the suppressed mutantsdemonstrate that the inability of the parent

cysB mutants to produce the catalytic compo-nents of the sulfur reduction pathway has beencorrected to varying degrees in the suppressedmutants.The structural genes for the enzymes of the

cysteine pathway are scattered throughout thechromosome. In order for the product of thecysB region to interact with scattered structuralgenes, we proposed that this product is multiva-lent with perhaps a different controlling site foreach set of structural genes. It has been demon-strated that independent mutations within thecysB region can alter drastically the quantita-tive relationships of the enzymes of the pathwayto one another. The second mutation in class1-B mutants returns the control of both sulfitereductase and OAS sulfhydrylase to the type ofcontrol exhibited by strain LT2. The secondmutation in class 1-A mutants and the class 2mutants that were derived from cysBc284 canintroduce a totally new form of control forsulfite reductase (where the level of enzymedoes not reach zero) and at the same time alterthe level of activity of sulfhydrylase withoutaltering the lack of response of sulfhydrylase tothe sulfur source in the growth medium. In theclass 2 mutants that were derived from cysBa25,the second mutation does not alter the ability ofcysteine to depress sulfite reductase to un-detectable levels, and yet the level of OASsulfhydrylase during cysteine growth has beenincreased.Although these investigations are the first

100 BORUM AND MONTY J. BACTERIOL.

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from

REGULATORY MUTANTS IN S. TYPHIMURIUM 101

concerted efforts to characterize control of thecysteine biosynthetic pathway during station-ary phase, a suggestion of change in regulationof cultures in the resting state can be found inearlier literature. Pardee et al. (15) reportedthat transport-negative cells grown on cysteinedo not bind sulfate but that the same cellsgrown on L-djenkolic acid do. It is interesting tonote that in those studies, cysteine-grown cellsacquire some ability during the late restingstate to combine with sulfate.The parallel behavior of sulfite reductase and

OAS sulfhydrylase of strain LT2 characteristicof logarithmic-phase growth is lost during sta-tionary phase. On cysteine minimal medium,sulfite reductase remains at undetectable levelsduring both logarithmic and stationary phases,but OAS sulfhydrylase activity increasesgreatly upon reaching stationary phase. Onsulfate minimal medium, sulfite reductase de-creases in activity upon reaching stationaryphase but OAS sulfhydrylase remains at a highlevel of activity. However, mutation in the cysBregion can alter these relationships during sta-tionary phase just as during logarithmic growth.On cysteine minimal medium the cysBc482mutant and two representatives of class 1-Amutants do not exhibit the expected increase inOAS sulfhydrylase activity upon reaching sta-tionary phase. As another example of diver-gence, the two class 1-A mutants have highlevels of sulfite reductase activity during thelogarithmic phase of growth and, although theactivity decreases in stationary phase, it doesnot fall to zero. In addition to proposing amultivalent cysB product that is able to controlseveral scattered structural genes in a parallelmanner during logarithmic growth, one mustconsider that each controlling site can have anindependent and possibly different response tothe metabolic circumstance of stationary phase.

ACKNOWLEDGMENTSThis work was supported in part by research grant E-337

from the American Cancer Society and contract AT (40-1)3082 from the United States Atomic Energy Commission.P.R.B. was supported by National Science Foundationtraineeship 199005-220 7R69 and grant GZ0962.

LITERATURE CITED1. Becker, M. A., N. M. Kredich, and G. M. Tomkins. 1969.

The purification and characterization of 0-acetylserinesulfhydrylase-A from Salmonella typhimurium. J.Biol. Chem. 244:2418-2427.

2. Demerec, M., and Z. Hartman. 1956. Tryptophan mu-tants in Salmonella typhimurium. In Genetic studieswith bacteria. Carnegie Inst. Washington Publ.612:5-25.

3. Dreyfuss, J., and K. J. Monty. 1963. The biochemicalcharacterization of cysteine-requiring mutants of Sal-monella typhimurium. J. Biol. Chem. 238:1019-1024.

4. Dreyfuss, J., and K. J. Monty. 1963. Coincident repres-sion of the reduction of 3'phosphoadenosine 5'phospho-sulfate, sulfite and thiosulfate in the cysteine pathwayof Salmonella typhimurium. J. Biol. Chem.238:3781-3783.

5. Hartman, P. E. 1956. Linked loci in the control ofconsecutive steps in the primary pathway of histidinesynthesis in Salmonella typhimurium. Carnegie Inst.Washington Publ. 612:25-62.

6. Hartman, P. E., Z. Hartman, and D. Serman. 1960.Complementative mapping by abortive transduction ofhistidine-requiring Salmonella mutants. J. Gen. Mi-crobiol. 22:354-368.

7. Hartman, P. E., J. C. Loper, and D. Serman. 1960. Finestructure mapping by complete transduction betweenhistidine-requiring Salmonella mutants. J. Gen. Mi-crobiol. 22:322-353.

8. Hershey, A. D., and M. Chase. 1952. Independent func-tions of viral protein and nucleic acid in the growth ofbacteriophage. J. Gen. Physiol. 36:39-56.

9. Hayes, W. 1968. The genetics of bacteria and theirviruses. John Wiley and Sons, Inc. New York.

10. Itzhaki, R. F., and D. M. Gill. 1964. A micro-biuretmethod for estimating proteins. Anal. Biochem.9:401-410.

11. Kredich, N. M. 1971. Regulation of L-cysteine biosynthe-sis in Salmonella typhimurium. J. Biol. Chem.246:3474-3484.

12. Kredich, N. M., and G. M. Tomkins. 1966. The enzymicsynthesis of L-cysteine in Escherichia coli and Salmo-nella typhimurium. J. Biol. Chem. 241:4955-4965.

13. Mizobuchi, K., M. Demerec, and D. H. Gillespie. 1962.Cysteine mutants of Salmonella typhimurium. Genet-ics 47:1617-1627.

14. Ozeki, H. 1959. Chromosome fragments participating intransduction in Salmonella typhimurium. Genetics44:457-470.

15. Pardee, A. B., L. S. Prestidge, M. B. Whipple, and J.Dreyfuss. 1966. A binding site for sulfate and itsrelation to sulfate transport into Salmonellatyphimurium. J. Biol. Chem. 241:3962-3969.

16. Siegel, L. M. 1965. A direct microdetermination ofsulfide. Anal. Biochem. 11:126-132.

17. Zinder, N., and J. Lederberg. 1952. Genetic exchange inSalmonella. J. Bacteriol. 64:679-699.

VOL. 125, 1976

at Univ of F

lorida on March 4, 2008

jb.asm.org

Dow

nloaded from