in vitro transcription escherichia coli k-12 arga, arge...

JOURNAL OF BACTE5RIOLOGY, May 1977, p. 642-655Copyright © 1977 American Society for Microbiology

Vol. 130, No. 2Printed in U.S.A.

In Vitro Transcription of the Escherichia coli K-12 argA,argE, and argCBH Operons

DON SENS, WILLIAM NATTER, AND ERIC JAMES*Department of Chemistry, University of South Carolina, Columbia, South Carolina 29298

Received for publication 6 January 1977

Deoxyribonucleic acid isolated from argA and argECBH transducing phageswas utilized to study the in vitro synthesis ofargA, argE, and argCBH messen-ger ribonucleic acid. The specific regulation of these operons by the arginineholorepressor was demonstrated, providing evidence that the majority, if not all,of the control of these operons is exercised at the transcriptional level. Data arepresented which indicate that the arginine holorepressor functions by binding tothe operator region and concomitantly prevents the binding of ribonucleic acidpolymerase to the corresponding promoter region.

The genes involved in arginine biosynthesisare scattered around the Escherichia coli chro-mosome, constituting a regulon with six orseven distinct operators (1, 16, 17) controllingnine genes (Fig. 1). The arginine genes ECBHare continuous, and studies have been pre-sented (11, 20, 35, 38) that indicate that theargECBH gene cluster is a bipolar operon di-vergently transcribed from an internal controlregion situated between argE and argCBH.Further analysis (1, 5, 15) of this gene clustersuggested that one operator controls argE withtranscription oriented counterclockwise andthat another operator controls the argCBHgenes with transcription oriented in the oppo-site direction. Extensive genetic studies byBretscher and Baumberg (3), employing four-factor crosses and deletion mapping with mu-tants involving the argE and argCBH controlregions, have not provided a definite answerregarding the structure of this important re-gion controlling the divergent transcription ofthese two operons. However, the data of theseworkers seem to exclude the possibility of oneoperator controlling both operons.The results of in vivo studies have demon-

strated that the synthesis of enzymes in thearginine pathway is repressed when wild-typecells are grown in the presence of arginine andis derepressed in the absence ofexogenous argi-nine (16, 17, 29). It has been demonstrated (21,29) that the argR gene product is a protein,which, in concert with a corepressor, regulatesthe synthesis of all the arginine genes in acoordinate but nonparallel manner. The exactnature of the corepressor for the arginine path-way is unknown, but data have been presented(4) that suggest that arginyl-transfer ribonucleicacid (RNA) is not the corepressor, and the pre-

liminary results of Cunin et al. (7) suggest thatarginine may be the corepressor for theargCBH operon.The question of whether the argR gene prod-

uct functions at the transcriptional or transla-tional stage has not been answered definitivelyfor all the genes in this regulon; however, Sensand James reported (42) that a minimum of95%of the regulation of the expression of the argFgene is mediated at the level of transcription.The studies of Sens, Natter, and James (sub-mitted for publication) have confirmed thenotion that the majority, if not all, of the regu-lation of the argF gene is effected at the tran-scriptional stage, and these workers haveshown that the argI gene is also controlled atthis level. Hybridization studies have been per-formed (6, 8, 23, 24, 25) in which the level ofargECBH messenger RNA (mRNA), producedin vivo by the argECBH cluster, has been mea-sured under conditions of repression and de-repression by hybridization to deoxyribonucleicacid (DNA) isolated from a number of special-ized transducing phages carrying the argECBHgene cluster, and it was demonstrated that thelevels ofargECBH mRNA and the correspond-ing enzymes do not correlate. Lavell6 and De-hauwer (26) reported similar results for thetryptophan operon of E. coli and proposed tworegulatory systems: one controlling transcrip-tion ofDNA, and another controlling the trans-lation of the resulting mRNA. McLellan andVogel (31) proposed a similar mechanism forthe regulation of the arginine regulon based onstudies of the argECBH gene cluster. However,the recent results of Bertrand et al. (2), involv-ing studies of the tryptophan operon, provideevidence for the existence of a second regula-tory site, the "attenuator," functioning at the

642

on June 10, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

IN VITRO TRANSCRIPTION OF argA AND argECBH 643

arg ECBH

pyrCarg D att8O

str tonBargR trparg G

arg hs

rec A

FIG. 1. Simplified genetic map of E. coli K-12.

transcriptional level, which accounts for thediscrepancy in mRNA and enzyme level. Thus,the development of a thorough understandingof the precise mechanism for the regulation ofthe arginine regulon will depend upon the fur-ther development of both in vivo and in vitrosystems for the analysis of the products of thevarious genes of the arginine regulon.The gene coding for the first enzyme of the

arginine biosynthetic pathway, argA, has re-ceived little attention in studies regarding thecontrol of gene expression in the arginine regu-lon. One reason for this situation is the insta-bility of this enzyme, N-acetylglutamate syn-thetase (EC 2.3.1.1), even in crude extracts ofE. coli; however, the synthesis of this enzyme isrepressible, and also the enzyme is subject tofeedback inhibition (27).

In this work the in vitro transcription of theargA, argE, and argCBH operons was studiedby using cytoplasmic extracts of E. coli andDNA isolated from specialized transducingphages carrying these arginine genes.

MATERIALS AND METHODSMaterials. Trizma base and dithiothreitol (DTT)

were purchased from Sigma Chemical Co., St.Louis, Mo. All radioisotopes were obtained fromNew England Nuclear Corp., Boston, Mass. Uri-dine, adenosine, guanosine, and cytidine 5'-triphos-phates (UTP, ATP, GTP, and CTP, respectively)were supplied by P-L Biochemicals, Inc., Milwau-kee, Wis. Polyuridylic acid-polyguanylic acid [poly-(U-G)] was purchased from Biopolymers, Inc.,Cleveland, Ohio, and had a U-G ratio of 1.9:1.Electrophoretically purified deoxyribonuclease(DNase) and ribonuclease (RNase) were obtainedfrom Worthington Biochemicals Corp., Freehold,N.J. Spectinomycin and streptolydigin were thegenerous gifts of The Upjohn Co., Kalamazoo,Mich. Selectron B-6 filters were supplied by ArthurThomas Co., Philadelphia, Pa. Media were pur-chased from Difco Laboratories, Detroit, Mich. Ce-sium chloride was obtained from Columbia OrganicChemicals, Columbia, S.C.

Media. Bacteria used for the preparation of phagestocks and cell extracts were grown in L-medium

(28). Bacteriophages were titered on TYE plates inH-top agar (18). F-top agar was used for plating cellson minimal medium (33). Selection plates containedmedium A (9) and supplemental growth factors asrequired, 2% agar, and 0.5% glucose as the carbonsource. Supplements were used at the concentra-tions previously described (13).

Bacterial strains and bacteriophages. The geno-type and origin of bacterial strains and bacterio-phages used in this work are listed in Table 1.

The specialized transducing bacteriophage Xh80-dargECBHl, which is the parent of the hybridphage used as DNA template in this work, wasisolated by Press et al. (39) by the technique ofepisome fusion. From this 480dargECBH phage, ahybrid phage, Oh80dargECBH1, was constructed (P.James, unpublished data). This was found to have adensity similar to that of its helper phage whichseverely hindered separation of the transducingphage from the helper phage. When DNA isolatedfrom Xh8OdargECBHl was used as hybridizationprobe for the determination of the derepressed/re-pressed ratio ofargE and argCBH mRNA formed invivo, a ratio of about 2 was determined. It was feltthat this low ratio was a result of hybridization ofmRNA from other bacterial genes, neighboring theargECBH operons, to the complementary DNA se-quence carried on the specialized transducing phage(due to the relatively large amount of bacterial in-formation carried on this phage). A light mutant(lower density) of the Xh8OdargECBHl transducingphage was isolated by the ethylenediaminetetraace-tate (EDTA) procedure of Parkinson and Huskey(37). This selection resulted in the isolation of aphage, Xh8OdargECBH2, that possessed a much de-creased density and greatly facilitated the purifica-tion of the bacteriophage (M. Cleary, unpublisheddata).The specialized transducing phage XdargECBH26

was constructed from an att lambda deletion strain(KY3304) into which XcI857sus xis6Ab515Ab519 wasinserted into the bfe gene by the technique of Shi-mada et al. (43). The bfe gene is located less than 1min from the argECBH gene cluster, and the subse-quent induction of this strain gave rise to theXdargECBH26 transducing phage (W. Natter, D.Sens, and E. James, unpublished data). Subsequentcharacterization of this phage revealed that it waseasily purified from helper phage, that the lightDNA strand of XdargECBH26 carried sense infor-mation for argCBH, and that the heavy strand car-ried sense information for argE (W. Natter, D. Sens,and E. James, submitted for publication).The specialized transducing phage XcI857dargA2

was isolated by Sens and James (unpublished data)as a light mutant ofXcI857dargA by the procedure ofParkinson and Huskey (37).

Construction of RNase-negative strains. Anovernight culture of cells was subcultured andgrown to the late log phase in L-medium with glu-cose. A 1-ml amount of this culture was centrifuged,washed with TM buffer, and resuspended in 1.0 mlof TM buffer. [TM buffer is (per liter): tris-(hydroxymethyl)aminomethane (Tris), 24.2 g;maleic acid, 23.2 g; (NH4)2SO4, 1.0 g; MgSO4-7H2O,

VOL. 130, 1977


.org/D

ownloaded from

http://jb.asm.org/

TABLE 1. Bacterial strains and bacteriophages

Strain Genotype SourceE. coliCA8000 HfrH thi J. BeckwithDF634 thi leu his pyrB strA D. FraenkelGJ1 thi (proAIB argF lac)A argI RNase NTG mutagenesis of GL5

with low-RNase pheno-type

GL5 thi (proAlB argF lac)A argI N. GlansdorffJC12R15 thi met purC argR15 SpDr lac xyl mel G. JacobyKY3304 thi bfe (XcI857xis6A515SAb519) H. YamagishiMA4A4 thi argA (XcI857) (XcI857dargA) N. KelkerMG427 thi (ppc argECBH)A strr G. JacobyEJ113 thi (proAIB argF lac)A argI argR15 spcr GL5 x JC12R15EJ113* thi (proAlB argF lac)A argI argR15 RNase SpCr NTG mutagenesis of EJ113

with low-RNase pheno-type

EJ123 thi argA (AcI857) (XcI857dargA2) Our collectionEJ142 thi (ppc argECBH)A strr (Xh80cI857) (Xh8OdargECBH2) M. ClearyEJ217 thi (ppc argECBH)A strr (XcI857S7) Our collectionEJ218 thi (ppc argECBH)A strr (XcI857S7) (XdargECBH26) Our collection

Bacteriophage4)80 L. Gorini,080dargECBH R. PressXh80cI857 L. GoriniXh80dargECBH2 M. ClearyXcI857S7 N. KelkerXdargECBH26 Our collectionXcI857dargA N. KelkerXcI857dargA2 Our collection

0.1 g; Ca(NO3)2, 5.0 mg; and FeSO4 7H2O, 0.25 mg.]To the cell suspension was added 0.1 ml of a solutioncontaining 1 mg of N-methyl-N'-nitro-N-nitroso-guanidine (NTG) per ml, and the mixture was incu-bated at 37°C for 30 min without aeration. The mu-tagenized culture was then washed three times withTM buffer, resuspended in 1.0 ml of TM buffer, anddiluted to give a concentration of 5,000 viable cellsper ml; a 0. 1-ml quantity was then spread on each of100 TYE plates. The plates were incubated over-night at 30°C and then replica-plated onto fresh TYEplates. The master plates were stored at 4°C, and thereplica plates were incubated at 30°C for 3.5 h. (GSAis 0.7% agar, 0.4% EDTA, and 3.0% yeast RNA[Sigma] adjusted to pH 7.0.) After 3.5 h of incuba-tion, 1.0 ml of 1.0 N HCl was added to each plate,and any colony not surrounded by a clear halo wasscored as RNase positive, located on the masterplate, purified by streaking to single colonies, andretested for the RNase phenotype.

Propagation and purification of bacteriophages.The bacteriophages Xh8OdargECBH2, Adarg-ECBH26, and AcI857dargA2 were propogated fromthe appropriate lysogenic strains as described byMiller (32). Bacteriophages were purified as de-scribed previously (22, 41).

Resolution of phage DNA strands by poly(U-G).Strand resolution of bacteriophage DNA by com-plexing with poly(U-G) was performed as describedby Hradecna and Szybalski (19) with the modifica-tion of Sens et al. (41).

Isolation of DNA. Bacteriophages were purifiedimmediately prior to isolation of DNA by centrifu-gation to equilibrium in a cesium chloride gradient(density, 1.5 g/cm3). Extraction ofDNA for use as atemplate for in vitro transcription was performed asdescribed by Miller (32) and for use in binding tonitrocellulose filters as described previously (41).

Preparation of cell extracts. Cell extracts wereprepared from strains EJ113* and GJ1 by themethod of Miller (32). The S-30 cell extracts weredialyzed twice against 50 volumes of buffer contain-ing 20 mM Tris-hydrochloride (pH 7.9), 15 mMMgCl2, 150 mM KCl, 1 mM DDT, and 2 mM L-arginine and were then stored at -70°C. Care wastaken to select S-30 cell extracts prepared fromstrains carrying the argR and argR+ alleles, whichwere equally efficacious in directing the in vitrosynthesis of (8-galactosidase as described below.

In vitro trancription. Standard reaction mixturesfor in vitro transcription contained (per milliliter):Tris-hydrochloride (pH 7.9), 23 mM; MgCl2, 15 mM;KCl, 150 mM; DDT, 1 mM; L-arginine, 1 mM; ATP,UTP, and GTP, each at 0.15 mM; [3H]CTP (specificactivity, 23.2 Ci/mmol), 0.075 mM; DNA, 50 ,tg; andS-30 extract, 300 asl. The reaction mixture was firstincubated without nucleoside 5'-triphosphates for 5min at 37°C; the reaction was initiated by the addi-tion of nucleoside 5'-triphosphates and terminatedby the addition of 2.0 ml ofa cold solution containing100 mM Tris-hydrochloride (pH 7.0), 3 mM MgCl2,0.2 mg of carrier RNA per ml, and 25 ,ug of RNase-

.644 SENS, NATTER, AND JAMES J. BACTERIOL.


.org/D

ownloaded from

http://jb.asm.org/


free DNase per ml. After 15 min at 4°C, 20-Al frac-tions were removed in triplicate; 50 ,ug of carrierRNA, 460 plA of 10 mM EDTA, and 500 pl of 20%trichloroacetic acid were added to each of the tripli-cate samples, and the mixture was allowed to standat 40C for 25 min. The precipitate was collected on aglass-fiber filter (Whatman GF/C) and washed with100 ml of 2% trichloroacetic acid. The filter wasdried, and 3H-labeled RNA was determined by usingRedi Solv IV in a Beckman LS 230 scintillationspectrometer. The remaining [3H]RNA was ex-tracted with an equal volume of phenol (saturatedwith 50 mM sodium acetate [pH 5.2] and 30 mMMgCl2), RNA was precipitated by the addition of 2volumes of 95% ethanol (-20°C), and the mixturewas allowed to stand at -20°C until a flocculentprecipitate had formed. The precipitate was col-lected by centrifugation at 10,000 rpm for 20 min in aBeckman J-21B centrifuge. The precipitate was driedand then dissolved in 1 ml of 4 x SSC (SSC is 0.15 MNaCl plus 0.015 M trisodium citrate, pH 6.8), and 20-,up fractions were removed in triplicate for the deter-mination of [3H]RNA as described.

Hybridization procedures. The quantity of RNAsynthesized in vitro was determined by the hybridi-zation procedure of Gillespie and Spiegelman (14).In experiments for the determination of argA-spe-cific mRNA (where only a lambda phage is at pres-ent available), lambda mRNA was removed prior tothe determination ofargA-specific mRNA by prehy-bridization of a quantity ofRNA (4 x 104 cpm) for 24h at 670C to 10 ,ug of lambda DNA immobilized on anitrocellulose filter in 400 ,ul of4x SSC. Under theseconditions, lambda mRNA was removed, and spe-cific argA mRNA remained in solution. The amountof mRNA was determined by permitting 150 1LI ofthe supernatant solution to hybridize for 16 h at 67°Cto 1 ,ug of the individual, separated DNA strands ofxcI857dargA2 immobilized on a nitrocellulose filter.This general procedure also was used to determineargE- and argCBH-specific mRNA, directed byXh8OdargECBH2 template DNA, by hybridization tothe separated DNA strands of XdargECBH26.

In vitro protein synthesis. Cell-free synthesis of/3-galactosidase was performed essentially as de-scribed by Miller (32). All components (as listed inTable 2), except DNA and S-30 extract, were assem-bled at room temperature, DNA was added, andpreincubation was allowed to proceed at 37°C for 5

min. Protein synthesis was initiated by the additionof the appropriate quantity of S-30. All reactionswere performed in a total volume of 0.1 ml and wereterminated by the addition of 100 ug of chloram-phenicol per ml and by transfer to an ice-water bath.

RESULTS

DNA dependence. The in vitro synthesis ofmRNA described in this study was completelydependent upon the addition of DNA carryingthe argECBH gene cluster and the argA operon

for the production ofargE-, argCBH-, and argA-specific mRNA. The production of argE- andargCBH-specific mRNA proceeded linearlyover a range of Xh8OdargECBH2 DNA concen-trations from 0 to 10 ,ug of added DNA per 100-Aul reaction volume (as judged by hybridizationto the separated DNA strands of Xdarg-ECBH26, after removal of lambda mRNA tran-scripts by prehybridization to lambda DNA)(Fig. 2). The synthesis of argE- and argCBH-specific mRNA accounted for approximately 1.3and 3.2%, respectively, of the mRNA producedby the in vitro system when 5 ,ug of DNA tem-plate was used. Hybridization background tolambda DNA, after prehybridization, accountedfor approximately 0.3% of the total RNA syn-thesis, and the production of mRNA comple-mentary to Xh8OdargECBH2 accounted for 67%of the radioactivity incorporated into trichloroa-cetic acid-precipitable material.The in vitro synthesis ofargA -specific mRNA

was also entirely dependent upon the additionof XcI857dargA2 DNA (Fig. 3) and was a linearfunction of the XcI857dargA2 DNA concentra-tion from 0 to 10 pug of added DNA per 100-dulreaction volume (as judged by hybridization tothe light DNA strand of XcI857dargA2 afterremoval of lambda mRNA transcripts by prehy-bridization to lambda DNA). Synthesis ofargA-specific mRNA accounted for approximately 6%of the total RNA produced in vitro, and thebackground hybridization to lambda DNA(after prehybridization) was 0.4% of the total

TABLE 2. Composition of incubation mixture per milliliter

Component Quantity Component Quantity2 M Tris acetate (pH 8.2) ....... ......... 22 Al. 1 M ammonium acetate.27.3 ,ul1 M DTT ............................... 91 ,lj 1McAMP.10 t,l2 M potassium acetate ........ .......... 28.2 ul 1 M IPTG.5.5 lI5 mM amino acids ........... ........... 45.5 lI. 0.27% folinic acid.10,ul0.1 M CTP, GTP, UTP ........ .......... 5.5 1.d 1 M magnesium acetate.10 ,ul0.2 M ATP ............................. 11 ,IA 1 M calcium chloride.7.3 gl0.1 M Na3 PEP ............... .......... 210 ,lI 30% PEG6000..20 ILIDNA ................................ 50 ,g S-30 ........................ 6,500 ,ug of protein

. Water ...................... 137.1 ,ula PEP, Phosphoenolpyruvate; cAMP, cyclic adenosine 3',5'-monophosphate; IPTG, isopropyl-,8-D-thioga-

lactopyranoside; PEG 6000, polyethylene glycol 6000.

VOL. 130, 1977


.org/D

ownloaded from

http://jb.asm.org/

646 SENS, NATTER, AND JAMES

6-~~~~~~~~~~~~~~~~~-I'

O10

0 ~~~~~~~~~~0

E E

DNA CONCENTRATION (po/100 pl)

FIG. 2. Synthesis of total RNA, template-specificmRNA, and argE- and argCBH-specific mRNA as afunction of Xh8OdargECBH2 DNA concentration.The abscissa indicates DNA concentration expressedas micrograms per 100-pl reaction mixture. The left-hand ordinate indicates synthesis of argE- andargCBH-specific mRNA expressed as counts perminute per 100-pJ reaction mixture after the removalof lambda mRNA transcripts by prehybridization tolambda DNA. The right-hand ordinate indicates to-tal RNA synthesis (determined by trichloroaceticacid precipitation) and template-specific mRNA (de-termined by hybridization to Xh80dargECBH2DNA). An argR- S-30 cell extract was utilized, andsynthesis was allowed to proceed for 2 min. Thedetermination ofargE-specific mRNA (A) was moni-tored by hybridization to the heavy DNA strand ofAdargECBH26, and argCBH-specific mRNA (U)was monitored by hybridization to the light DNAstrand of XdargECBH26. In both cases, lambdatranscripts were removed by prehybridization tolambda DNA; the background hybridization tolambda DNA after removal of lambda transcriptswas less than 0.3% of the total RNA synthesized.Total RNA synthesis (0) was determined by trichlo-roacetic acid precipitation, and the quantity of tem-plate-specific mRNA (0) was determined by hybridi-zation to 2 pg ofXh80dargECBH2 DNA immobilizedon a nitrocellulose filter. Each data point is the aver-age of three determinations.

RNA produced by the in vitro system whendirected by 5 ,ug of template DNA. Synthesis ofmRNA complementary to XcI857dargA tem-plate accounted for 95% of the radioactivity in-corporated into acid-precipitable material.

In vitro transcription of Ah8OdargECBH2DNA. RNA transcripts produced by Xh80darg-ECBH2 template DNA, using an argR S-30cell extract (strain EJ113*), were analyzed by

hybridization to the heavy and light DNAstrands of XcI857S7 (Fig. 4), 080 (Fig. 4), andXdargECBH26 (Fig. 5). The quantity ofmRNAthat complexed specifically to the light DNAstrand oflambda was negligible throughout the10-min time course and comprised less than 1%of the mRNA complementary to Xh80darg-ECBH2 at times later than 1 min of synthesis(Fig. 4). The quantity ofmRNA that complexedspecifically to the heavy DNA strand oflambdarose rapidly during the first 3 min of synthesis,rose only slightly during the remainder of thetime course, and accounted for approximately29% of the total RNA synthesis (at 2 min) thatwas complementary to Xh8OdargECBH2 tem-plate DNA. The quantity of synthesized mRNAcomplementary to the heavy DNA strand of480 (Fig. 4) rose slowly during the first 0.5 minof synthesis, increased more rapidly for a shorttime, and finally slowed with only a small netaccumulation of mRNA occurring during the3- to 10-min time period. Synthesis of mRNA

0

x 3E

2

x

DNA CONCENTRATtON (pg/100p1)FIG. 3. Synthesis of total RNA, template-specific

mRNA, and argA-specific mRNA as a function ofAcI857dargA2 DNA concentration. The abscissa in-dicates DNA concentration expressed as microgramsper 100-pl reaction mixture. The left-hand ordinateindicates synthesis of argA-specific mRNA (A) ex-pressed as counts per minute per 100-,i reaction(after removal of lambda mRNA transcripts by pre-hybridization to larmbda DNA) and determined bymonitoring hybridization to the light DNA strand ofXcI857dargA2. Background hybridization to lambdaDNA was approximately 0.4%. The right-hand ordi-nate represents total RNA synthesis (0) determinedby trichloroacetic acid precipitation and synthesis oftotal template-specific RNA (0) determined by hy-bridization to XcI857dargA2 DNA. Each data pointis the average of triplicate determinations.

J. BACTERIOL.


.org/D

ownloaded from

http://jb.asm.org/


orQ

...

TIME (min)

FIG. 4. Time course of synthesis of mRNA com-plementary to Xh8OdargECBH2 DNA and the sepa-rated DNA strands of lambda and of 480, withXh8OdargECBH2 DNA template and an argR S-30cell extract. Synthesis of mRNA complementary toXh8OdargECBH2 DNA (O) was determined by hy-bridization to Xh8OdargECBH2 DNA. Synthesis ofmRNA complementary to the light DNA strand oflambda (A) and the heavy DNA strand of lambda(0) was determined by hybridization to the appropri-ate separated DNA strands of lambda. Synthesis ofmRNA complementary to the light DNA strand of480 (A) and the heavy DNA strand of 480 (0) wasdetermined by hybridization to the separated DNAstrands of 080. Samples (50 pl) were removed ateach time point, and a portion (2 x 104 cpm) wasused to determine the amount ofeach ofthe differentmRNA transcripts present. Each time point repre-sents that amount that would have been present in a100-pl reaction mixture and is the average of threedeterminations.

complementary to the light DNA strand of 480(Fig. 4) rose rapidly during min 1 of synthesisand then increased more slowly until at 3 minno further net accumulation of mRNA occurred.Synthesis ofRNA complementary to the heavyDNA strand of 480 accounted for approximately34% of the total synthesis of RNA complemen-tary to Xh8OdargECBH2 template DNA,whereas synthesis of RNA complementary tothe light DNA strand of 4)80 accounted for ap-proximately 27% of the total synthesis of RNAcomplementary to Xh8OdargECBH2 templatedDNA at the 2-min time point.

Analysis of mRNA produced from Xh8O-dargECBH2 template DNA by hybridization tothe separated DNA strands of XdargECBH26,after the removal of lambda and )80 mRNAtranscripts by prehybridization, is shown inFig. 5. The rate of specific argE mRNA synthe-sis was determined by hybridization to the

heavy DNA strand of XdargECBH26 and ex-hibited three distinct phases. Synthesis ofargE-specific mRNA occurred rapidly duringthe first 1 min of synthesis, much more slowlyduring the next 2 min, and then exhibited nonet accumulation of argE-specific mRNA dur-ing the remainder of the time course (Fig. 5).The rate of argCBH-specific mRNA synthesiswas monitored by hybridization to the lightDNA strand of XdargECBH26 and exhibitedtwo distinct phases ofmRNA synthesis. Duringthe first 1 min of synthesis, argCBH-specificmRNA rose rapidly, followed by a much re-duced rate of synthesis for the remainder of thetime course (Fig. 5). At a synthesis time of 2min, approximately three times as muchargCBH-specific mRNA was produced com-pared with argE-specific mRNA (argE mRNAaccounted for 1.3% of the total RNA synthe-sized, and argCBH-specific mRNA accountedfor 3.2% of the total RNA synthesized). Back-ground hybridization to lambda DNA was ap-

0

It

A0

a.

C)

2 4 6TIME (min)

8 10

FIG. 5. Time course ofargE- and argCBH-specificmRNA synthesis directed by Ah8OdargECBH2 DNAwith an argR- S-30 cell extract. The quantity ofargE- and argCBH-specific mRNA was determinedby analysis ofa portion ofeach 50-pJ portion used inthe experiment depicted in Fig. 4. The quantity ofargE-specific mRNA was determined by hybridiza-tion to the heavy DNA strand of XdargECBH26 (A),and the quantity of argCBH-specific mRNA was

measured by hybridization to the light DNA strandof XdargECBH26 (0) after removal of lambdamRNA transcripts by prehybridization to lambdaDNA. The quantity ofmRNA synthesized representsthe amount that would have been present in a 100-plreaction mixture. Each data point represents the av-

erage of three determinations.

d

I I

VOL. 130, 1977


.org/D

ownloaded from

http://jb.asm.org/


proximately 0.3% and has been subtracted fromthe values presented.

In vitro transcription of XcI857dargA2DNA. RNA produced by AcI857dargA2 tem-plate DNA with an argR- S-30 extract wasanalyzed by hybridization directly to the sepa-rated DNA strands oflambda (Fig. 6) and to theseparated DNA strands of XcI857dargA2 (Fig.7) after removal of lambda RNA transcripts byprehybridization to lambda DNA as describedin Material and Methods. The quantity ofmRNA produced by XcI857dargA2 templateDNA which specifically complexed to the lightDNA strand of lambda increased steadily dur-ing the first 5 min of the time course, afterwhich time net synthesis rapidly decreased tozero (Fig. 6). Synthesis of mRNA complemen-tary to the heavy strand of lambda DNA pro-ceeded linearly for 7 min, at which time netaccumulation of mRNA approached zero (Fig.6). Synthesis of mRNA complementary toXcI857dargA2 DNA template accumulated in alinear fashion for the first 7 min of synthesis, at

TIME (ml.)

FIG. 6. Time course of synthesis of mRNA com-

plementary to XcI857dargA2 DNA and to the sepa-rated DNA strands of lambda with XcI857dargA2DNA template and an argR- S-30 cell extract. Syn-thesis of mRNA complementary to XcI857dargA2DNA (0) was determined by hybridization directly to2 pg of XcI857dargA2 DNA immobilized on a nitro-cellulose filter. Synthesis of mRNA complementaryto the lightDNA strand oflambda (A) and the heavyDNA strand oflambda (0) was measured by hybrid-ization to the appropriate separated DNA strands oflambda (1 pg) immobilized on a nitrocellulose filter.Samples (50 p1) were removed at each time point,and a portion (2.0 x 104 cpm) was used to determinethe amount ofmRNA complementary to the templateDNA and the separated strands of lambda DNA.Each time point represents the amount that wouldhave been present in a 100-y4 reaction mixture and isthe average of three determinations.

J. BACTERIOL.

o r-SC

U

7b' 2 4 6 8 10TIME (min)

FIG. 7. Time course of argA-specific mRNA syn-thesis directed by XcI857dargA2 template DNA withan argR- S-30 cell extract. The quantity of argA-specific mRNA was determined by analysis ofa por-tion of each of the 50-pl portions used in the experi-ment described in the legend ofFig. 6. The quantityof argA-specific mRNA was determined by hybridi-zation to the light DNA strand of XcI857dargA2(A), whereas antisense argA mRNA was measuredby hybridization to the heavy DNA strand ofXcI857dargA2 (0) after (in both cases) the removal oflambda mRNA transcripts by prehybridization tolambda DNA. Background hybridization to lambdaDNA (after removal of lambda transcripts by prehy-bridization) was 0.4% of the total template-specificmRNA produced. The quantity of mRNA 'synthe-sized represents the amount that would have beenpresent in a 1O00-p reaction mixture. Each data pointrepresents the average of three determinations.

which time net accumulation of template-spe-cific mRNA ceased (Fig. 6). The synthesis ofmRNA complementary to the light DNA strandof lambda accounted for approximately 34% ofthe template-specific mRNA, whereas synthe-sis ofmRNA complementary to the heavy DNAstrand of lambda accounted for approximately50% of the template-specific mRNA (at 2 min).Synthesis of mRNA complementary toXcI857dargA template accounted for 95% of theradioactivity incorporated into trichloroaceticacid-precipitable material.

Analysis of mRNA produced from AcI857-dargA2 template DNA by hybridization to theseparated DNA strands of XcI857dargA2 (afterthe removal of lambda mRNA transcripts byprehybridization) is shown in Fig. 7. The rate ofargA-specific mRNA synthesis was determinedby hybridization to the light DNA strand ofXcI857dargA2 and was found to be a linear

CI_

oto _


.org/D

ownloaded from

http://jb.asm.org/


function of time during the first 7 min of syn-thesis, after which time the net accumulation ofargA-specific mRNA rapidly became zero (Fig.7). Synthesis complementary to the heavy DNAstrand of XcI857dargA2 (antisense strand) alsoproceeded in a linear fashion during the first 7min of synthesis, after which time the rate offormation ofmRNA steadily decreased (Fig. 7).The synthesis of mRNA complementary to theheavy DNA strand of XcI857dargA2 wasgreater than that produced complementary tothe light DNA strand of XcI857dargA2, espe-cially after 5 min of synthesis (Fig. 7). At asynthesis time of 2 min, mRNA complemen-tary to the light DNA strand (sense) ofXcI857dargA2 (after lambda transcripts hadbeen removed) accounted for approximately6.5% of the total mRNA complementary to theXcI857dargA2 template (a similar amount wasdetermined to complex to the heavy DNAstrand of XcI857dargA2). The background hy-bridization to lambda DNA (after removal oflambda transcripts by prehybridization) wasapproximately 0.4% of the total template syn-thesis.

Repression of argE-specific mRNA synthe-sis by an argR+ S-30 cell extract, usingAh8OdargECBH2 template DNA. The result ofa set of experiments demonstrating in vitroregulation ofargE mRNA synthesis directed byDNA isolated from the specialized transducingphage Xh8OdargECBH2 is shown in Fig. 8.These experiments utilized varying proportionsofS-30 cell extract, simultaneously added to thereaction mixture, isolated from strains carry-ing either argR+ or argR- alleles and the heavyDNA strand of XdargECBH26 as the hybridiza-tion probe. It was demonstrated that argE-spe-cific mRNA was repressed significantly (Fig. 8)by increasing proportions of an S-30 extractobtained from a strain possessing the argR+allele when a synthesis time of 2 min was em-ployed. When 40% of the total S-30 extract wasprepared from a strain carrying the argR+ al-lele, argE-specific mRNA synthesis was re-pressed approximately 30%, and this repressionvalue increased as the proportion ofargR + S-30was increased, until a final repression value of86% was obtained when the argR+ extract com-prised 100% of the S-30 cell extract. When syn-thesis was permitted to proceed for 15 min nosignificant difference in the extent ofrepressionwas noted (Fig. 8). When synthesis was allowedto continue for 30 min, a slight loss of repres-sion was noted (66% of argE-specific mRNAsynthesis was under the control of the arginineholorepressor, as shown in Fig. 8).

Repression of argE-specific mRNA synthesisby an argR+ S-30 cell extract was shown to be

I0

PERCENT R EXTRACT

FIG. 8. Repression of argE-specific mRNA as afunction ofS-30 composition. A series ofexperimentswas performed in which S-30 cell extracts ofdifferentcomposition were used while Xh8OdargECBH2 tem-plate DNA was used to direct argE-specific mRNAsynthesis. The cell extract used in each experimentwas derived either from argR S-30, argR+ S-30, orfrom specific mixtures of the two S-30 cell extracts;the percentage of argR+ extract is indicated on theabscissa. The ordinate represents the observed per-centage of repression of argE mRNA synthesis. Inthis experiment, 60,000 cpm of total RNA were pre-hybridized (for each data point) to 30 pg of lambdaDNA, and the resulting supernatant solution wasanalyzed for argE-specific mRNA by hybridization tothe heavy DNA strand of XdargECBH26. When cell-free transcription was performed with S-30 entirelyderived from an argR- strain, approximately 700cpm of radioactive argE-specific mRNA were deter-mined. Repression ofargE-specific mRNA was mea-sured by hybridization to the heavy DNA strand ofXdargECBH26 after removal oflambda mRNA tran-scripts at a synthesis time of2 min (5), 15 min (A),and 30 min (0). These values were corrected forbackground hybridization to lambda DNA, whichwas less than 0.25%. An identical experiment wasperformed with XcI857S7 DNA as the template (A);the amount of lambda-specific mRNA was deter-mined by hybridization to lambda DNA, and thevalue observed from synthesis directed by the S-30extract prepared from a strain carrying the argR+allele was 2,000 cpm. In this experiment, the S-30preparations were matched by selecting cell extractswith comparable activity in a coupled in vitro tran-scription-translation system as described by Cleary,Garvin, and James (submitted for publication).Synthesis of ,3galactosidase is represented on theright-hand ordinate by the value ofthe absorbance at420 nm determined per hour, per 100-pi reactionmixture, as a function of S-30 composition (a). Thedata presented are the average of duplicate determi-nations.

specific for the argE gene by employing twodifferent controls. First, synthesis of (8-galacto-sidase directed by 50 ,ug of480dlac DNA per mlwas monitored in a coupled transcriptional-translational protein-synthesizing system us-

VOL. 130, 1977

z

0Fe*

:I

z

rV


.org/D

ownloaded from

http://jb.asm.org/


ing the respective S-30's used in the repressionexperiment presented herein, and it was dem-onstrated (Fig. 8) that the argR+ S-30 was

slightly superior in directing the synthesis of 63-galactosidase. The second control involved themeasurement of transcription of XcI857S7 DNAby both the argR - and argR+ S-30 cell extractsand combinations thereof. No difference in syn-thesis of lambda transcripts was observed (Fig.8).

Repression of argCBH-specific mRNA syn-thesis by an argR+ S-30 cell extract, usingAh8OdargECBH2 template DNA. The ability ofan argR+ S-30 cell extract to repress argCBH-specific mRNA synthesis directed by DNA fromthe specialized transducing phage Xh80darg-ECBH2 was determined as a function of addedargR+ S-30 cell extract by using the lightDNA strand of XdargECBH26 as the hybridiza-tion probe. At a synthesis time of 2 min, theaddition of increasing amounts of argR+ cellextract caused a concomitant decrease in theamount ofargCBH mRNA produced; attaininga value of 90% repression when the S-30 cellextract was entirely derived from a straincarrying the argR+ allele (Fig. 9). At a synthe-sis time of 15 min, increasing amounts ofargR+S-30 cell extract decreased argCBH-specificmRNA synthesis to a slightly smaller extent,resulting in a maximum repression value of

0-3

204 \

- 0240-

i60

so

1001 -I -I -I

I0

0 20 40 60 S0 100

PERCENT R* EXTRACT

FIG. 9. Repression of argCBH-specific mRNAsynthesis as a function of S-30 composition. Theexperiment was performed as described in the legendofFig. 8, except that the light DNA strand of Xdarg-ECBH26 was used to monitor argCBH-specificmRNA synthesis at times of2 (A), 15 (0), and 30 (0)min ofsynthesis. As in Fig. 8, an input of60,000 cpmofRNA was used, and 2,100 cpm ofargCBH-specificmRNA were determined to be synthesized when theS-30 was entirely derived from a strain carrying theargR - allele. The controls were performed as de-scribed in the legend ofFig. 8. Each data point is theaverage of duplicate determinations.

81% when the S-30 cell extract was derivedentirely from an argR + strain (Fig. 9). Increas-ing the synthesis time to 30 min resulted in afurther loss of repression (66% when the S-30cell extract was entirely derived from an argR +strain [Fig. 9]). It was also shown (Fig. 9) thatsynthesis of lambda mRNA directed byXcI857S7 template DNA was unaffected by theuse of S-30 cellular extract derived from strainscarrying either the argR allele or mixturesthereof (Fig. 9). Synthesis of f8-galactosidasewas also shown to be independent of the compo-sition of the S-30 utilized, as would be expectedsince S-30 extracts with matched protein syn-thesis characteristics were used.

Repression of argA-specific mRNA synthe-sis by an argR+ S-30 cell extract, usingXcI857dargA2 template DNA. The ability of anargR + S-30 cell extract to repress argA-specificmRNA synthesis directed by DNA isolatedfrom the specialized tranducing phageXcI857dargA2 was determined as a function ofthe quantity of argR+ S-30 cell extract presentin the incubation mixture. At a synthesis timeof 2 min, the addition of increasing amounts ofargR + extract caused a concomitant decrease inthe amount of argA-specific mRNA producedcomplementary to the sense strand (lightstrand) ofXcI857dargA2 DNA, reaching a valueof 90% repression when the S-30 was entirelyderived from a strain carrying the argR + allele(Fig. 10). Increasing the time of synthesis to 15or 30 min resulted in only a slight loss of repres-sion (84% repression at 15 min and 78% repres-sion at a synthesis time of 30 min [Fig. 10]).Synthesis of argA mRNA complementary tothe heavy strand of XcI857dargA2 DNA (theantisense strand) was also monitored as a func-tion ofthe concentration ofargR+ cell extract atsynthesis times of 2, 15, 30 min, and it wasdemonstrated (Fig. 10) that the arginine holo-repressor had no effect on argA-specific mRNAproduced complementary to this strand. Syn-thesis of ,B-galactosidase was shown to be inde-pendent of the argR allele present in the S-30cell extract (Fig. 10), and synthesis of lambdamRNA directed by XcI857S7 DNA was not af-fected by the argR allele present in the S-30 cellextract (data not presented).

Repression of argE-, argCBH-, and argA-specific mRNA synthesis as a function of theorder of addition ofargR+ and argR- S-30 cellextract. The effect of the order of addition ofargR + and argR - cell extract on the regulationof argE-, argCBH-, and argA-specific mRNAsynthesis was determined by performing cell-free synthesis as described previously, exceptthat, in one case, the argR+ cell extract was

J. BACTERIOL.

9co9)wIatab,

I.-zwcoa

a.W


.org/D

ownloaded from

http://jb.asm.org/


0.1

2b 4b to do b-o-o.oPERCENT R* EXTACT

FIG. 10. Repression of argA-specific mRNA syn-

thesis as a function ofS-30 composition. The experi-ment was performed as described in the legend ofFig. 8, except that in this case XcI857dargA2 DNAwas used as template. Synthesis of argA-specificmRNA representing sense information was moni-tored at 2 (A), 15 (U), and 30 (0) min of synthesisby hybridization to the light DNA strand ofXcI857dargA2. The input oftotal RNA in this experi-ment was 40,000 cpm, and, after removal of lambdatranscripts by prehybridization to lambda DNA, ap-

proximately 2,400 cpm ofmRNA were determined tobe argA specific when the S-30 cellular extract was

entirely derived from a strain carrying the argR-allele. Background hybridization to lambda DNAvaried between 87 and 123 cpm. Synthesis of argA-specific mRNA representing antisense informationwas also monitored at 2, 15, and 30 min (0) byhybridization to the heavy DNA strand ofXcI857dargA2 after the removal of lambda tran-scripts by prehybridization to lambda DNA. The in-put ofRNA was 40,000 cpm, and, after prehybridi-zation, approximately 2,700 cpm were argA anti-sense specific regardless ofthe argR allele present inthe S-30. The ,3galactosidase control (a) was per-

formed as described in the legend of Fig. 8. Eachdata point represents the average of duplicate deter-minations.

added first and then incubated for 1 min at 40C,followed by the addition of the argR S-30 cellextract. In the other case, the argR- cell ex-tract was added first and allowed to incubate at40C for 1 min before the addition ofthe argR+ S-30 cell extract. In both cases, the reaction mix-ture was allowed to preincubate at 37°C for 5min, and the reaction was initiated by the addi-tion of nucleoside 5'-triphosphates.

Experiments performed with the prior addi-tion of cellular extract derived from a straincarrying the argR allele followed by the addi-tion of the appropriate quantity of S-30 derived

from a strain carrying the argR + allele resultedin the synthesis of steadily decreasing amountsofargE-specific mRNA with increasing quanti-ties of argR+ S-30 (repression value of 15 and38%, respectively, corresponding to an S-30composition comprising 20 and 60% ofthe cellu-lar extract derived from the argR+ strain).

In sharp contrast, when the order of additionwas reversed, with the prior addition of S-30derived from a strain carrying the argR+ allelefollowed by the addition of the appropriatequantity of argR- S-30, a dramatic increase ineffectiveness of the arginine holorepressor wasapparent. The synthesis ofargE-specific mRNAwas repressed 54% by the prior presence ofarginine holorepressor present in only 20% ofthe S-30 utilized in the experiment. Increasingthe quantity of S-30 derived from the straincarrying the argR+ allele to 40% of the totalresulted in the repression of the specific argEmRNA synthesis by 81% (Fig. 11). Similar re-sults were obtained when the effect of the orderof addition of argR+ and argR- S-30 cellularextracts was determined for the argCBH op-eron (data not presented). When DNA isolatedfrom XcI857dargA2 was used in this order ofexperiments, it was found that the repression ofargA-specific mRNA followed a pattern quali-tatively similar to that described for the argEand argCBH operons (data not presented). Syn-thesis of mRNA directed by DNA template iso-lated from XcI857 and cell-free synthesis of /3-galactosidase directed by DNA isolated from480dlac was unaffected by the order of additionof S-30 cellular extracts (Fig. 11). In the case ofexperiments performed with the argA operon,synthesis of antisense argA-specific mRNA wasalso determined to be unaffected by the order ofaddition of S-30 cellular extracts.

DISCUSSION

We have described the in vitro transcriptionof the argECBH gene cluster and the argAoperon by using DNA template isolated fromthe specialized transducing phages Xh80darg-ECBH2 and XcI857dargA2 and cytoplasmic(S-30) extracts prepared from strains of E. coliK-12 carrying the argR- and argR+ alleles.The system used for in vitro transcription isidentical to that used by McGeoch et al.(30) for the study of the tryptophan operon,except that S-30 extracts, which are capable ofcoupled transcription-translation, were usedinstead of S-180 cell extracts.The system was shown to be entirely depend-

ent on the addition of DNA containing theargECBH gene cluster and the argA operon forthe synthesis of mRNA specific for the argE,

VOL. 130, 1977


.org/D

ownloaded from

http://jb.asm.org/


PERCENT R EXTRACT

FIG. 11. Determination ofthe effect of the order ofaddition of S-30 extracts (prepared from strainscarrying the argR- and argR+ alleles) on the repres-sion of argE-specific mRNA synthesis as a functionof S-30 composition. Synthesis of RNA was per-formed as described in the legend ofFig. 8, except theeffect of the order of addition of the argR+ and theargR- S-30 cell extract was determined. TemplateDNA isolated from Xh8OdargECBH2 was used todirect synthesis of argE-specific mRNA with a syn-thesis time of 1 min. The amount of argE-specificmRNA produced in the individual reactions was

determined by hybridization to the heavy DNA strandof XdargECBH26. In one case, the quantity ofargR-S-30 used in each experiment was added to the reac-tion mixture at 4°C and permitted to incubate for 1

min before the quantity ofargR+ S-30 cell extract wasadded (O). The reciprocal experiment was performedby adding the argR+ S-30 cell extract first, incubat-ing the mixture at 4°C for 1 min, and adding theappropriate quantity of argR- S-30 cell extract (A).Polymerization was initiated by the addition of nu-

cleoside 5'-triphosphates and terminated as de-scribed in Materials and Methods. The same quan-

tity ofRNA was used in the hybridization reaction as

described in the legend ofFig. 8. Synthesis oflambdamRNA was also monitored by using XcI857S7 tem-plate DNA when the argR+ was added first (0) andwhen the argR- was added first (0); the data pointsshown represent approximately 2,000 cpm oflambda-specific mRNA. Synthesis of(&galactosidasewas also monitored when the argR+ was added first(-) and when the argR- was added first (-). Thedata represent the average of duplicate determina-tions.

argCBH, and argA operons (Fig. 2 and 3). Thesystem was not completely dependent on addedDNA for total RNA synthesis; this was notunexpected since McGeoch et al. (30) andRogers et al. (40) reported similar data. Themajority (90%) of radioactivity incorporatedinto trichloroacetic acid-precipitable material

J. BACTERIOL.

was found to complex specifically to DNA probeisolated from XcI857dargA when this sameDNA was used as template; however, only 67%of the radioactivity incorporated into acid-pre-cipitable material was determined to hybridizeto the DNA probe isolated from XdargECBH2when DNA template was isolated fromXh80dargECBH2. It appears that the largeamount of endogenous synthesis exhibitedwhen Xh8OdargECBH2 DNA template wasused to direct mRNA synthesis was a functionof the particular template and not a deleteriousaspect of the transcriptional system.

Synthesis of argE- and argCBH-specificmRNA was found to exhibit uncomplicated ki-netics, with the rate of accumulation ofmRNAsteadily decreasing at extended periods of syn-thesis, indicating that the rate of formation ofRNA was becoming balanced by degradationdue to the action ofRNase present in the S-30 ex-tracts. Synthesis ofargCBH-specific mRNA ac-counted for approximately 70% of the arg-ECBH-specific mRNA produced. This agreeswell with the ratio of argE mRNA to argCBHmRNA reported to be formed in vivo (8), as wellas the ratio of the mRNA species formed invitro (36). Initiation at the argCBH promoterappears to be the principal source of argCBH-specific mRNA synthesis, since 90% of theargCBH-specific mRNA synthesis was undercontrol of the arginine holorepressor (Fig. 9)and since the time course ofargCBH synthesisexhibited no plateau followed by a subsequentincrease in synthesis (Fig. 5), as was observedfor the argF operon carried on 480dargF DNA,where readthrough from one or more upstreambacterial promoters was observed (Sens et al.,submitted for publication).

Since experiments performed on the regula-tion of the argI and argF operons (Sens et al.,submitted for publication) demonstrated poorrepression values at extended times of mRNAsynthesis due to readthrough of mRNA initi-ated at upstream bacterial promoters, these ex-periments were repeated for the argECBHgene cluster.,Increasing the time of synthesishad little effect on the extent of repression ofargE- and argCBH-specific mRNA synthesis,with substantial repression values being at-tained even at 30 min of synthesis (Fig. 8 and9). These data support the notion that the DNAsequence carried on the specialized transducingphage Xh8OdargECBH2 in the region of theargECBH gene cluster probably has effectiveterminator sequences associated with anyneighboring bacterial genes that have pro-moters functioning under the conditions uti-lized in this work, thus preventing readthrough


.org/D

ownloaded from

http://jb.asm.org/


of mRNA transcripts initiated at other bacte-rial promoters. The demonstration of regula-tion to the extent of 90% by the arginine holore-pressor indicates that relatively little transcrip-tion is occurring from other bacterial operons.Furthermore, it shows that the majority, if notall, of the regulation ofthe argECBH operons ismediated at the level of transcription. It is nec-essary to reconcile this notion that regulation ofgene expression for the argE and argCBH op-erons is mediated entirely at the level of tran-scription with the observaton of Cunin et al. (8)and Krzyzek and Rogers (24) that the dere-pressed/repressed ratios of the enzyme activi-ties for the corresponding enzymes do not corre-late. One factor contributing to this noncorre-spondence is the difficulty in measuring accu-rately the quantity of mRNA formed in vivo,particularly under conditions of physiologicalrepression. Another potential explanation is toinvoke the possibility of post-transcriptionalcontrol operating in a manner analogous tothat observed for the trp operon (2). This possi-bility has been espoused by Krzyzek and Rogers(24), who have reported preliminary resultsconcerning the measurement of levels for twospecies of mRNA formed in vivo for the arg-ECBH operons; however, these workers did notutilize separated DNA strands for monitoringthese species.To obtain meaningful data when performing

repression experiments with two sets of cellextracts-one isolated from an argR- strainand the other prepared from an argR + strain -it is necessary to insure that the S-30 cell ex-tracts are of equal competence in directingtemplate-specific mRNA synthesis. This wasaccomplished in the present work by perform-ing three different controls. First, in all cases,cellular extracts were utilized that exhibitedcomparable activity in an in vitro transcription-translation system producing f3-galactosidasedirected by 480dlac DNA (M. Clearly, R. T.Garvin, and E. James, submitted for publica-tion). Second, the S-30 cell extracts werematched in their ability to synthesize argAantisense (heavy-strand-specific) RNA directedby XcI857dargA2 DNA and mRNA specific toXcI857S7. It is therefore clearly evident thatrepression of arginine-specific mRNA was dueto the presence of functional arginine holo-repressor.Synthesis of argA-specific mRNA was moni-

tored by hybridization using the separatedDNA strands of XcI857dargA2, and the rate ofsynthesis during a 10-min time period wasfound to exhibit uncomplicated kinetics (Fig.7). Previous characterization of the bacterio-

phage (Natter et al., submitted for publication)has demonstrated that the light DNA strand ofXcI857dargA2 carries sense information for theargA gene. Since synthesis of mRNA comple-mentary to both DNA strands of XcI857dargA2exhibited similar kinetics, it was of interest toascertain what effect the presence of thepromoter on the heavy DNA strand of XcI857-dargA2 would have opposite the argA operonand its interaction with the arginine holo-repressor. The results (Fig. 10) show that sub-stantial repression of argA-specific mRNAwas obtained at all synthesis times and that norepression was noted for mRNA complemen-tary to the heavy DNA strand ofXcI857dargA2.Thus, it appears that the promoter(s) located onthe heavy DNA strand of XcI857dargA2 has noadverse effect on mRNA produced that is com-plementary to argA sense gene DNA. This maybe due to the argA operon lying downstreamfrom the promoter on the heavy strand, thusgiving rise to a situation in which RNA polym-erase initiated at one promoter does not tran-scribe the other promoter region.Since both Ah8OdargECBH2 and XcI857-

dargA2 DNA template provide an excellentsource of DNA for regulation studies involvingthe argECBH gene cluster and the argAoperon, it was of interest to ascertain if infor-mation regarding the control regions of theseoperons could be obtained by altering the condi-tions employed when performing studies on theeffect of the arginine holorepressor as describedpreviously for the argF and argI operon (Senset al., submitted for publication). From the re-sults (Fig. 11), it is clear that there is competi-tion between the RNA polymerase and holore-pressor. Experiments performed with smallquantities of holorepressor added together witha small proportion of RNA polymerase (both inthe argR + S-30 cell extract) prior to the additionof the major proportion of RNA polymerase(contained in the argR - S-30) demonstrate thatrepression was considerably greater than whenthe reverse order of addition of holorepressorand RNA polymerase was employed. It may bepostulated, therefore, that repression of theseoperons occurs by the arginine holorepressorpreventing the binding of RNA polymerase atthe respective promoters and the subsequentformation of mRNA polymerase initiation com-plex as reported by Squires et al. (44) for thetryptophan operon.

This work describes the study of two DNAtemplates, one carrying the argA operon andthe other carrying the argECBH gene cluster.It has been shown that at least 86% of argE,90% of argCBH, and 90% of argA gene expres-

VOL. 130, 1977


.org/D

ownloaded from

http://jb.asm.org/


sion is mediated by the arginine holorepressorat the level of transcription and that repressionappears to result from the holorepressor inter-fering with the binding of RNA polymerase tothe promoter site. The in vitro system describedprovides the basis for further study regardingrepressor-corepressor interactions and will per-mit the development of studies aimed at deter-mining whether or not other control sites arepresent as described by Bertrand et al. (2) forthe trp operon, whether autoregulation occursas has been postulated for other operons (45),and whether the presence or removal (by dele-tion of the gene in question) of other arginineenzymes has an effect on the particular argi-nine operon under study.

ACKNOWLEDGMENTSWe are happy to acknowledge the excellent technical

assistance of Robin Hendrix and the expertise of Katay Pat-terson, who typed the manuscript, and Susan Wise, whohelped prepare the manuscript. We also thank PhillipJames and Michael Cleary, who constructed the hybridargECBH phage used in this work. We are most happy toacknowledge the generous gift of the specialized transduc-ing phage AcI857dargA made by N. Kelker.We are grateful for financial support through American

Cancer Society grants VC131, VC131A, and VC131B to E. J.

LITERATURE CITED1. Baumberg, S., D. F. Bacon, and H. J. Vogel. 1965.

Individually repressible enzymes specified by clus-tered genes of arginine biosynthesis. Proc. Natl.Acad. Sci. U.S.A. 53:1029-1032.

2. Bertrand, K., L. Korn, F. Lee, T. Platt, C. L. Squires,C. Squires, and C. Yanofsky. 1975. New features onthe regulation of the tryptophan operon. Science189:22-26.

3. Bretscher, A. P., and S. Baumberg. 1976. Control mu-tations within the divergently transcribed argECBHcluster of E. coli K-12 J. Mol. Biol. 102:205-220.

4. Celis, T. F. R., and W. K. Maas. 1971. Studies on themechanism of repression of arginine biosynthesis inEscherichia coli. IV. Further studies on the role ofarginine transfer RNA repression of the enzymes ofarginine biosynthesis. J. Mol. Biol. 62:179-188.

5. Cunin, R., D. Elseviers, G. Sand, G. Freundlich, andN. Glansdorff. 1969. On the functional organizationof the argECBH cluster of genes in Escherichia coliK-12. Mol. Gen. Genet. 106:32-47.

6. Cunin, R., and N. Glansdorff. 1971. Messenger RNAfrom arginine and phosphenol pyruvate carboxylasegenes in argR+ and argR- strains of E. coli. FEBSLett. 18:135-137.

7. Cunin, R., N. Kelker, A. Boyen, H. Yang, G. Zubay, N.Glansdorff, and W. K. Maas. 1976. Involvement ofarginine in in vitro repression oftranscription of argi-nine genes C, B and H in Escherichia coli K-12.Biochem. Biophys. Res. Commun. 69:377-382.

8. Cunin, R., P. Pouwels, N. Glansdorff, and M. Crabeel.1975. Parameters of gene expression in the biopolarargECBH operon of E. coli K-12. The question oftranslational control. Mol. Gen. Genet. 140:51-60.

9. Davis, B. D., and E. S. Mingioli. 1950. Mutants ofEscherichia coli requiring methionine or vitamin B,2.J. Bacteriol. 60:17-28.

10. Dickson, R. C., J. Abelson, W. Barnes, and W. S.Reznikoff. 1975. Genetic regulation: the lac control

J. BACTERIOL.

region. Science 187:27-29.11. Elseviers, D., R. Cunin, N. Glasdorff, S. Baumberg,

and E. Ashcroft. 1972. Control regions within theargECBH gene cluster of Escherichia coli K-12. Mol.Gen. Genet. 117:349-366.

12. Eron, L., and R. Block. 1971. Mechanism of initiationand repression of in vitro transcription of the lacoperon in E. coli. Proc. Natl. Acad. Sci. U.S.A.68:1828-1832.

13. Eshenbaugh, D. L., D. Sens, and E. James. 1974. Asolid phase radioimmune assay for ornithine trans-carbamylase, p. 61-73. In R. B. Dunlap (ed.), Ad-vances in experimental medicine and biology, vol. 42.Plenum Publishing Corp., New York.

14. Gillespie, D., and S. Spiegelman. 1965. A quantitativeassay for DNA-RNA hybrids with DNA immobilizedon a membrane. J. Mol. Biol. 12:829-842.

15. Glanedorff, N., and G. Sand. 1965. Coordination ofenzyme synthesis in the arginine pathway of Ewche-richia coli K-12. Biochim. Biophys. Acta 108:308-311.

16. Glanadorff, N., G. Sand, and C. Verhoef. 1967. Thedual genetic control of ornithine transcarbamylasesynthesis in Escherichia coli K-12. Mutat. Res. 4:742-751.

17. Gorini, L., W. Gunderson, and M. Burger. 1961. Ge-netics of regulation of enzyme synthesis in the argi-nine biosynthetic pathway of Escherichia coli. ColdSpring Harbor Symp. Quant. Biol. 26:173-182.

18. Gottesman, S., and J. R. Beckwith. 1969. Directedtransposition of the arabinose operon: a technique forthe isolation of specialized tranducing bacteriophagefor any Ewcherichia coli gene. J. Mol. Biol. 44:117-127.

19. Hradecna, A., and W. Szybalski. 1967. Fractionation ofthe complementary strands ofcoliphage X DNA basedon the asymmetric distribution of the poly UG bind-ing sites. Virology 32:633-643.

20. Jacoby, G. 1972. Control of the argECBH cluster inEscherichia coli. Mol. Gen. Genet. 117:337-348.

21. Jacoby, G., and L. Gorini. 1969. A unitary account ofthe repression mechanism of arginine biosynthesis inEscherichia coli. J. Mol. Biol. 39:73-87.

22. James, P., D. Sens, W. Natter, S. K. Moore, and E.James. 1976. Isolation and characterization of thespecialized transducing bacteriophages +80dargF andXh8OcI867dargF: specific cleavage of arginine trans-ducing deoxyribonucleic acid by the endonucleasesEcoRI and SmaR. J. Bacteriol. 126:487-500.

23. Krzyzek, R., and P. Rogers. 1972. Arginine control oftranscription ofargECBH messenger ribonucleic acidin Escherichia coli. J. Bacteriol. 110:945-954.

24. Krzyzek, R., and P. Rogers. 1976. Dual regulation byarginine of the expression of the Escherichia coliargECBH operon. J. Bacteriol. 126:348-364.

25. Krzyzek, R., and P. Rogers. 1976. Effect of arginine onthe stability and size ofargECBH messenger ribonu-cleic acid in Escherichia coli. J. Bacteriol. 126:365-376.

26. Lavalld, R., and G. Dehauwer. 1970. Tryptophan mes-senger translation in Escherichia coli. J. Mol. Biol.51:435447.

27. Leisinger, T., and D. Haas. 1975. N-acetylglutamatesynthetase ofEscherichia coli regulation of synthesisand activity by arginine. J. Biol. Chem. 250:1690-1693.

28. Lennox, E. 1955. Transduction of linked genetic charac-ters of the host by bacteriophage P1. Virology 1:190-206.

29. Maas, W. K. 1961. Studies on repression of argininebiosynthesis in E8cherichia coli. Cold Spring HarborSymp. Quant. Biol. 26:183-191.

30. McGeoch, D., J. McGeoch, and D. Morse. 1973. Cell-free synthesis of tryptophan operon mRNA. Nature


.org/D

ownloaded from

http://jb.asm.org/


(London) New Biol. 245:137-140.31. McLellan, W. L., and H. J. Vogel. 1970. Translational

repression in the arginine system ofEscherichia coli.Proc. Natl. Acad. Sci. U.S.A. 67:1703-1709.

32. Miller, J. H. 1972. Experiments in molecular genetics,p. 423. Cold Spring Harbor Laboratory, Cold SpringHarbor, New York.

33. Miller, J. H., K. Ippen, J. G. Scaife, and J. R. Beck-with. 1968. The promoter-operator region of the lacoperon ofEscherichia coli. J. Mol. Biol. 38:413-420.

34. Nakanishi, S., S. Adhya, M. E. Gottesman, and I.Pastan. 1973. In vitro repression of the transcriptionofgal operon by purifiedgal repressor. J. Biol. Chem.248:5937-5942.

35. Panchal, C., S. Bagchee, and A. Guha. 1974. Divergentorientation of transcription from the arginine geneECBH cluster of Escherichia coli. J. Bacteriol.117:675-680.

36. Pannekock, H., R. Cunin, A. Boyen, and N. Glans-dorff. 1975. In vitro transcription of the argECBHgene cluster. FEBS Lett. 51:143-145.

37. Parkinson, J., and R. Huskey. 1968. Isolation of dele-tion mutants of bacteriophage lambda. J. Mol. Biol.56:369-384.

38. Pouwels, P., R. Cunin, and N. Glansdorff. 1974. Diver-gent transcription in the argECBH cluster ofgenes inEscherichia coli K12. J. Mol. Biol. 83:421-424.

39. Press, R., N. Glansdorff, P. Miner, J. Devries, R. Kad-ner, and W. K. Maas. 1971. Isolation of 4680 transduc-

ing particles carrying different regions of the E. coligenome. Proc. Natl. Acad. Sci. U.S.A. 68:795-798.

40. Rogers, P., T. Kaden, and M. Toth. 1975. Repression ofarg mRNA synthesis by L-arginine in cell-free ex-tracts of Escherichia coli. Biochem. Biophys. Res.Commun. 65:1284-1291.

41. Sens, D., D. L. Eshenbaugh, and E. James. 1975. Reso-lution of the DNA strands of the specialized tranduc-ing bacteriophage Xh80cI857dargF. J. Virol 16:85-93.

42. Sens, D., and E. James. 1975. Regulation of argFmRNA synthesis, performed in vitro. Biochem. Bio-phys. Res. Commun. 64:169-174.

43. Shimada, K., R. Weisberg, and M. E. Gottesman. 1972.Prophage lambda at unusual chromosomal locations.I. Location of the secondary attachment sites andproperties of the lysogens. J. Mol. Biol. 80:297-314.

44. Squires, C. L., F. Lee, and C. Yanofsky. 1975. Interac-tion of the trp repressor and RNA polymerase withthe trp operon. J. Mol. Biol. 92:93-111.

45. Umbarger, H. E. 1975. Regulation ofamino acid biosyn-thesis in microorganisms, p. 13-18. In H. R. V. Am-stein (ed.), Synthesis of amino acids and proteins.International review of science and biochemistry, se-ries I. vol. 7. University Park Press, Baltimore.

46. Williams, L. S. 1973. Control of the arginine biosyn-thesis in Escherichia coli: role of arginyl-transfer ri-bonucleic acid synthetase in repression. J. Bacteriol.113:1419-1432.

VOL. 130, 1977


.org/D

ownloaded from

http://jb.asm.org/

in vitro transcription escherichia coli k-12 arga, arge...

Documents