0022-538X/85/030807-07$02.00/0
CopyrightC 1985, AmericanSocietyforMicrobiology
Expression of
the Cloned Bacteriophage PX174 A* Gene
in
Escherichia
coli
Inhibits
DNA
Replication
and
Cell Division
JOSEPH COLASANTIANDDAVID T. DENHARDT*
Cancer Research Laboratory and the Department ofMicrobiology and Immunology, University of Western Ontario,
London,
OntarioN6ASB7
CanadaReceived 27September1984/Accepted 26 November 1984
The A* geneofbacteriophage 4X174 has been cloned into the inducible expression vectorpCQV2 under conditions allowing its lethal action tobe controlledby the XcI857repressor. Upon inductionof expression, DNA synthesis inEscherichia coli carrying the recombinant plasmid is severely inhibited;however, thesesame
cellspermit 1-galactosidase inductionataratesimilartothat observed in control cellsattheinducing (for A*) temperature.Cells in which A* is expressed formifiamentsand producemoreRecA protein, indicatingatleast
apartial inductionof the SOSresponse;however, there isnoevidence of damagetothe bacterial chromosome. Itappearsthat the A* protein hasas onefunction theinhibition of cell division and DNA replicationbut not
transcriptionorprotein synthesisduring phage infection.
The functions ofmost of the genes encoded by bacte-riophage 4X174 have been elucidated by genetic
experi-ments, particularly by the isolation of mutations in those genes (see reference 31 for review). For the A* gene, however, this strategy has notbeen fruitful because of the
nature of the relation between A* and the larger A gene. Gene A codes for a polypeptide with anMr ofca. 58,000 whose essential roleas anickaseorligase in the replication
of thephagegenomehas been well documented(4), whereas the function of the A* protein (Mr, ca. 37,000) in vivo is unknown. Unlike the overlappinggenes of(X174, genesA and A*areread in thesamne reading frame; A* results from
a translational restart from a ribosome-binding site (RBS) approximately one-third of the distance from the beginning of the gene A RBS (20). Because of the nested nature of thesetwogenes, anymutation in A*also affectsA, making it difficult to differentiate their respective functions during phage infection.
An early study by Lindqvist and Sinsheimer (19) impli-catedaphage-encoded protein in the phage-induced shutoff of host DNA synthesis during infection. This protein was later found tobe aproduct of the A (or A*) gene (13, 23). Thesestudies, however, didnotreachaclear and unambig-uous position regarding the role of A* protein (7). Amore recent study has shown that A* has the ability to inhibit replication of both replicative form (RF) and DNA single-stranded(ssDNA) XX174 in vitro (11). The purified protein bindsstronglytossDNA and double-stranded DNA(34) and hasendonucleolytic activity with somesequence specificity towards ssDNA (16). In the presence ofMn2+ or the 4X gene A protein, the A* protein may be capable of nicking and binding covalently to XX RF (10, 35). The A* protein boundtothe 5' end ofaDNA strandcanbedisplaced under appropriate conditions by the 3' OH end of the same or another DNA strand, resulting in the formation ofa 3'-5' phosphodiester bond (12, 35, 36).
To elucidate further the function of A*, we used
recom-binant DNA technology to clone the A* gene to study its effects in vivo on Escherichia coli. Initial attempts at this cloning revealed that expression of the A* protein was, as
expected, lethaltocells in which itwasexpressed.
Consist-*Correspondingauthor.
ent with this, van der Avoort et al. (33) were unable to
isolateaclone with the intact AorA*geneincircumstances in which cloned segments of
4X174
RF producing other phage productswere isolated. Forour cloning strategy, we therefore undertook to use plasmid vehicles which allow control ofexpression of clonedgenes. After initial failures with vectorspPLc28 (28), pKC30 (30), and pJL6-8 (18), we succeeded with pCQV2, which contains both the APR
promoter and the c1857 repressor (27). This vector was foundtomaintain the A*genein a stable and silent format
30°C andtoallowits expression at 42°C. Upon induction of expression of the A* gene, therateof DNAsynthesis in E. coliwas drastically reduced, and the cellswerekilled.
MATERIALS AND METHODS
E. coli, phage, and plasmid strains. The amber mutants
fX174am16 and
4X174am3
(lysis defective), deficient in geneB and Efunctions, respectively,werepropagated in E. coliCRthy-Su'
(6). E. coli C Su- and CRweregrownin mT3XD(6). Other strainsweregrownin Lbrothor supple-mentedM9 minimal mediumcontaining 0.2% glucose (22)asindicated. Theplasmid pCQV2 (27)wasprovided by G. A.
Chaconas; pPLc28 (28) was obtained from W. Fiers along with itshostsK12AHlAtrp and M5219; pKC30wasfrom M. Rosenberg; pJL6-8wasobtainedfrom D. Court.
DNA purification and manipulation. Plasmid DNA was purified fromalysozyme-Brij 58 clearedlysateby isopycnic centrifugationincesium chloride-ethidiumbromide followed by extraction with n-butanol, dialysis, phenol-chloroform extraction, and a final dialysis and alcohol precipitation. Phage RF DNAwas purified from E. coli CR infected with XX174am16atahighmultiplicity (2to3) when the cells had reachedadensity of 2 x 108/ml; chloramphenicolwasadded
12 min after infection to a concentration of 20 ,ug/ml, and
aeration of the cultures was continued at37°C for5 h. RF
DNA was isolated by the same method used to isolate plasmid DNA.
Allenzymeswereobtained from BethesdaResearch Lab-oratories, Boehringer-Mannheim, orNew England Biolabs and were used as recommended by the suppliers unless
indicated otherwise. To fill in recessed 3' ends, restricted duplexDNAwasincubated with 65 mMTris-hydrochloride
(pH 7.6)-40mM KCl-25 mM MgCl2 in the presence of the 807
on November 10, 2019 by guest
http://jvi.asm.org/
*PA
pCQV2 - 0X174 StuI A
-GGAGQTTGTATfGATCC
[5IGGAGGCCTCCACTATGA
-CCTCCAACATACCTAG(3-
bpCCTCCGGAGGTGATACT
BamHI
FIG. 1. Construction ofpJA*1 and pBCK5. The recombinant pJA*1wasconstructedby inserting the fragmentcontaining the A* gene of
4X174 into thecloning vehicle pCQV2 (see the text). This construction results in placement of the A* gene in the correct orientation for transcription from the A PR promoter. Repressor protein (cI857) prevents transcription from the PR promoter at 30°C; at 42°C, the temperature-sensitive repressor is renderednonfunctional, and transcription occurs, allowing the A* proteintobesynthesized starting from itsownRBS. TheplasmidpBCK5wasconstructed frompJA*1 bydeleting the BamHI-BstXIfragment,blunting withT4DNApolymerase, andreligating. The deletion eliminates A* expression but leaves therest of the plasmid intact, thereby allowing the effects of A* to be separated from those of theaml6geneBproduct and the gene C- and geneK-pCQV2 fusions. The predicted sequence of the
pCQV2-4X174
jointatthe A*Nterminus isshownin detail. Apeptide initiated from thecroRBS and AUG codon terminatesneartheA* startcodon.
fourdeoxynucleoside triphosphates (0.5 mM) and thelarge (Klenow)fragment ofDNA polymerase I (60 to 100 U/ml) for30to60 minat25°C. DNAfragmentswerepurified from 0.7%
low-melting-point
agarosegels by electroelutiononto adialysis membrane placed inatrough inthegel (22). Recom-binant DNA was used to transform E. coli RR1 cells to
ampicillin resistance by a modified Hanahan procedure for cells other than X1776 (22). Cells were freeze-shocked in a
-70°C
ethanol-dry
ice bath before addition of dimethyl sulfoxide, which was added only once. Heat shock wasperformed at 36°C, and the transformation mixture was incubatedat30°C.
Determination of the rate of DNA synthesis. Cells with various plasmids were growntomid-logphase in L broth in the presenceof carbenicillin (100
p.g/ml)
at 30°C. When the cell density reached 108/ml, half of each culture was trans-ferredto 42°C. Portions (1 ml) were taken at various times beforeandaftertheshift and addedto10,uCi of[3H]thymidine
andsufficientdeoxyguanosinetoproduce a50 mMsolution. The cultures wereincubated for2 min, and then incorpora-tion was terminated with 2 ml of cold 10% trichloroacetic acid. Trichloroacetic acid-precipitable material was
col-lected onglassfiber filters andwashed,and itsradioactivity was determined.
I-Galactosidase
assay.E. coli RR1 cellscarrying
plasmids
were grown at 30°C in M9 minimal medium supplemented with0.2% Casamino Acidsand 100 ,ugof carbenicillinperml to a density of 5 X
107
cells perml, at whichpoint half ofeach culture was shifted to 42°C; 6 min later, isopropyl
,-D-thiogalactopyranoside
was added to a final concentra-tion of 2 mM.Duringthecourseofgrowth, aliquots (0.25 ml) weretaken and added toan equal volume of Z-buffer (25), and the cellswerepermeabilizedwithadropof toluene. The level of 3-galactosidase production was determined as de-scribedbyMiller(25). The apparentA6Eo
of the cultureswasalsodetermined atvarious times during growth.
Electroblotting ofproteinsand immunodetection. Cultures ofE. coli RR1 harboring plasmids were grown in L broth with carbenicillin (100 ,ug/ml) at 30°C to adensity of _108
cells per ml. Theculturesweredivided,onehalfwasshifted to42°C,and theincubationswerecontinued foranother 2 h. Aliquots were taken and centrifuged, and the pellets were
suspended in Laemmli sample buffer and boiled for 5 min. Equivalent amounts of each sample (determined by the
on November 10, 2019 by guest
http://jvi.asm.org/
[image:2.612.152.479.71.411.2]apparent
A6N
of the cultures) were then run on a sodium dodecyl sulfate-polyacrylamide gel (12.5% acrylamide, 2.25% N,N'-diallyltartardiamide) with a 4.5% acrylamide stacking gel. Immunoblotting was performed essentially as described by Towbin et al. (32) with thefollowing modifica-tions. Afterelectrophoresis, proteins were transferred from the gelontonitrocellulosepaper(Schleicher&Schuell, Inc.) inLaemmlirunningbuffercontaining20% methanol with an E-C Electroblot system. The paper was stained with 0.1% amido black, destained for 1min with 10%acetic acid-10% methanol, washed three times for 5 min each in distilled water, and incubated overnight with Tris-buffered saline (TBS;20 mM Tris-hydrochloride [pH 7.5], 150 mM NaCl). Nonspecificadsorptionwas blocked by incubating the paper with asolution of TBS containing 0.05% Tween 20 (Sigma) for2h atroomtemperature(2). Rabbit antiserum (provided by Shlomo Eisenberg)raisedagainst the 4XX gene A protein was diluted 1:750 in TBS containing 0.05% Tween 20 and incubated withthe blot in a volume of 20 ml for 2 h at room temperature. Rabbit antiserum (provided by Jeff Roberts) raised against E. coli RecA protein was diluted 1:4,000 in the samebuffer and incubatedsimilarly. The blots were washed sixorseventimesfor10min eachin TBSplus Tween 20 and then incubated with 125I-labeled protein A (New England NuclearCorp.), ca. 30,000 cpm per lane, for 2 to 3 h. After extensive washing with TBS plus Tween 20, blots were subjectedto autoradiography.RESULTS
Controlledexpression of a lethal gene. Phage X regulatory elements have been usedtocontroltheexpression of cloned genes in several systems (3, 24, 29). Shimatake and Ro-senberg (30), for example, used the X PL promoter-c1857 thermolabile repressor system of pKC30to control expres-sion of the lethal X cIIgene. In attempting toclone the A* geneof
4X174,
however,wediscoveredthatseveralvectors bearing the PL promoter apparently exerted insufficient repression of thisgene at 30°C (the restrictivetemperature forexpression)topermittheisolationof stableclones. This was possibly due to an inadequate concentration of repres-sor, since the cI857 gene was located on the host chromo-somein adefective prophage in the systems we tried.This prompteda search fortemperature-inducible expres-sion vectors that used eitherthe X
PL
or PR promoter and encoded the X cI857 repressor. The plasmid pCQV2 was found tobesatisfactory (seeFig. 1). Itcontainsthe XcI857 repressor gene transcribed from thePRM promoter(which can be stimulated by cI) and the PR rightward promoter (which is repressed by thecIorc1857 protein)just upstream from the RBS and theAUG startcodon ofthe X cro gene.(These
latterwere notessential incloning
the A*fragment
sinceitcarried its ownRBSand startcodon).
Construction, diagrammed in Fig. 1, of a recombinant with the A*gene in the properorientation relativeto the PR promoter involved cleaving pCQV2 with BamHI, creating bluntends on theduplexwith the KlenowfragmentofDNA polymerase I, and then cutting with SaI. The digestion of
4~X174
double-stranded RF DNAwithNruI, which cleaves 74base pairsupstream fromthe A* start codon, andXhoI, which cleaves 32 base pairs after the A-A* stop codon, yields a 1,142-base-pair fragment containing the A* gene. The normal promoter for the A* gene is the PA promoter, which is just before the start of the A gene and, thus, not present. Consequently, transcription of this fragment is dependentontranscription from the PRpromoter.Ligation of theprepared pCQV2 DNA with the XX174 A* fragment, followed by transformation of E. coli RR1 to ampicillin resistance, permitted the isolation of the plasmid pJA*1. The flush-ended BamHI-NruI joining results in re-generation of theBamHI site, whereas the SalI-XhoI cohe-sive end joining does not regenerate either site. Given transcription from the PR promoter, the nature of this construction allows synthesis ofapeptide fromthe cro RBS-startcodon into
4XX174
DNA but out offrame with the A* gene andresulting intermination oftranslation upon reach-ing the A* start region. (Expressionof this peptide does not affect E. coli metabolism because clones have been con-structed in which it is absent [by cloning the StuI-XhoI fragment of4X174
intopCQV2, see Fig. 1]and theybehave as pJA*1 does [data not shown]. These recombinants were not used in this study because the A* produced was altered in that it had five additional amino acids fused in its N terminus.) Since the cloned segment of XX DNA in pJA*1 carried theRBS and initiationcodonfor A*, itwaspossible to obtainexpression of A* protein. Expressionwasinitially evidenced by the inability of the cells to form colonies at 42°C. The presenceof(X174 DNA wasconfirmedbycolony hybridization with labeled (by nick translation) XX174 RF DNA orfragments containing the A*gene.PropertiesofpBCK5, a deletion mutant of A*. Because of thepresence ofoverlapping reading frames in this segment of XX174 DNA, it isimpossibletoclone the A* gene without cloning all or partsofothers, in this case the entire B gene andthe 5' ends ofgenesC and K. Tominimize interference
M
a
b
c
d e f
g
h
i
j k
--_
-_
=
=_~-
58k
_~~
--37k
A~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
FIG. 2. Electroblotting and immunodetection of gene A* pro-tein. Equivalentamounts of cell extracts (E. coli RR1) were elec-trophoresedon a sodium dodecylsulfate-polyacrylamide gel, elec-troblottedontonitrocellulose paper, and probed first with antibody to XX gene A protein and then with 1251I-labeled protein A as described in thetext. Lanes athroughccontain2,0.4, and 0.04 ,ug, respectively, ofa purified preparation of
4X174
gene A protein provided by S. Eisenberg. Lanes d and e are extracts of strain RR1(pCQV2)at 30 and 42°C (2 h),respectively; lanes f and g are extractsof strainRR1(pJA*1)at30 and42°C(2 h),respectively;and lanes hand iareextractsof strainRR1(pBCK5)at 30and42°C (2 h), respectively.Lanesjand k are extracts of E. coli Cuninfected(lane j) and infected(lane k)with 4X174am3for1.5 h at37°C. Molecular weight markers (Bio-Rad Laboratories) onthe leftare phosphor-ylase b(Mr,92,500), bovineserumalbumin (Mr,66,200),ovalbumin (Mr, 45,000), carbonic anhydrase (Mr, 31,000), soybean trypsin inhibitor (Mr, 21,500)andlysozyme(Mr, 14,400).on November 10, 2019 by guest
http://jvi.asm.org/
[image:3.612.320.555.379.565.2]A
E
0.
I
B
700-600.
0)
.-:
500-0)
U)
0 'o
O' 400.
0
0
0%
3100-\/
1
2X1
4X'I00
21
~~~~t
10' 10 20 30 40 50 0 15 30 45 60
Time (min) Time (min)
FIG. 3. Effect of A* protein on DNA synthesis and 3-galactosidase induction in E. coli. (A) DNA synthesis was determined by measurementof[3H]thymidineuptake intoacid-precipitable material. CellsweregrowninLbroth(containing100,ugofcarbenicillin perml) to108cells per ml at 30°C, and 1-ml portions were taken(0min)and added to 10,uCiof[3H]thymidinefor2min (see the text). At20 min after thefirst sampling, half of each culture wastransferredto420C, and the measurements werecontinued. (B) Cells weregrownin minimal medium(containing100,ugof carbenicillin per ml) to5x 107cells per ml at30°C, atwhich point half of each culture wastransferredto42°C (O min). Isopropyl 3-D-thiogalactopyranoside was added six min laterto a 2 mM finalconcentration, and 0.25-mlsamples were takento determine levels ofP-galactosidaseinduction. Strains used:E.coliRR1(pCQV2)grownat30°C(0)andat420C(0),E.coli RR1(pJA*1) grown at30°C
(E)
andat42°C (E), andE. coliRR1(pBCKR) grownat30°C (A)and at42°C (A).fromgeneB,anambermutant,4X174am16,wasused in the cloning, and studies were done in a nonpermissive host. However,onecouldstillarguethatitwasthetruncatedgene B product or gene C-pCQV22 or gene K-pCQV2 fusion products whichwereresponsible for the effects attributedto
the A*protein. DNA-directedtranscription-translation reac-tions performed in vitro with recombinant plasmids (not shown) indicated thepresenceofputative fusionpeptides of theapproximate molecular weight predicted for the XX174-pCQV2construct.
To prove that the lethal effects were indeed due to A* expression, we constructed a mutant, pBCK5, by deleting theBamHI-BstXI fragment from pJA*1 (Fig. 1). This dele-tionremovesthe A* startsitealong with the first half of the genebut stillallows transcription ofgeneB and thegene C and Kregions asbefore. The cro start siteisjoined to the remaining 4X174sequencesuch that thepeptide produced is in adifferent reading frame from A* andterminates almost
immediately after it starts. This is borne out by immuno-blotting results(see Fig. 2). The ability of cells bearing this plasmid to form colonies at42°C indicates that these other polypeptides did notkill the cell and thatexpression of the A* proteinwas necessaryfor the lethal effect.
Anadditional argumentthatgene B,geneC, and geneK
products are not lethal to the cell can be based on the
existenceof the
PB
promoter. Thispromoterprecedesgene B(Fig. 1) and presumably permits constitutive transcription of thisregion in the cell. Since the cellssurvivedat30°C, this transcription was notlethal. Inaddition, vander Avoort etal. (33) showed that cells thatwere expressing theB and C genes (and, hence, able to complement 4X gene B and C mutations) grew well.
Detection of the A*protein.In casesin whichgenes are put undercontrol ofAPL or PR promotersand
c1857, induction
at 42°C often results inthe
production
ofalargeamount of protein (27, 30). Unfortunately, thiswas notobserved with induction ofE.colicarryingpJA*1.
TheA*protein (Mr, ca. 37,000) did not stand out on Coomassie blue-stainedpolyacrylamide
gels or in autoradiographs ofgels
of[35S]methionine-labeled
cell extracts induced at 42°C (not shown). However, major E. coliproteins migrated
in this region and could have obscured moderate amounts of aninducedprotein.
Theavailability ofananti-A antiserummadeitpossibleto usethe more sensitive immunoblotting procedureto detect the A* protein. Figure 2 shows the presence of the A* protein after induction of expressionat42°C,whereas none at all was detectable at 30°C. Cells
carrying
pBCK5 pro-duced neither A* protein nor a truncated product after induction.Interestingly, the anti-A antiserum seemed to react spe-cificallywith anotherE. coliprotein intheregion ofthegel indicating an Mr of ca. 56,000. The nature ofthis bacterial protein is unknown; however, since at
420C
heat shock increases itsexpression, it could be the B56.5protein (also called the A protein and the groE protein [26]). The cross-reactionwith anti-+X174gene Aproteinserummayindicate a sharedepitope.
The other(less intense) bands on thegel0
0 a
0
o
XZ/
on November 10, 2019 by guest
http://jvi.asm.org/
[image:4.612.127.501.73.342.2]A
B
FIG. 4. Induction offilamentsby A* protein.E.coli RR1 cellsharboringeitherpCQV2orpJA*1weregrowntomid-log phaseinLbroth (with 100 ,ug of carbenicillin per ml)at30°C, transferredto Lagarplates(withcarbenicillin), and allowedtogrowateither30 or42°Cfor1.5 h.Blocks of agarcontainingcellswerethen fixed with 1%Os04;the cellsweretransferredtoglassslides and then stained withHCl-Giemsa. This procedure stains the nuclear regions ofE. coli darkly. Strains shown: E. coliRR1(pCQV2) at 30°C (A) and42°C (B) and E. coli RR1(pJA*1)at30°C (C) and42°C(D).
appeared with nonimmune rabbit serum also (data not shown).
Effect ofA*protein on E. coli DNA synthesis. One ofthe effects ofA* protein in vitro is inhibition of(X174 DNA synthesis (11). Induction ofA*expressioninvivoallows one to determine its effects on E. coli replication
directly.
Measurement oftherateofDNAsynthesis via[3H]thymidine uptake revealed a marked decrease in synthesis in cells in which A* expression had beeninduced (Fig. 3A), whereas cellsat 30°C and those
carrying
theparent plasmid atboth temperatures continuedsynthesizing
DNA. Similarly, cells harboringpBCK5 showed noinhibitionof DNAsynthesisat 42°C, providing further evidence that A* protein directly inhibitsE. coliDNAreplication. The residual DNA synthe-sis in the E. coliRR1/(pJA*1)
culture at42°Cmay be due to a subpopulation of cells lacking the ability to express A* protein.Effect of A* on
I(-galactosidase
induction in E. coli. To ascertain theeffect ofthe A*proteinoncellulartranscription and translation, we monitored ,B-galactosidase production.Cells carrying either pCQV2 or pJA*1 were grown toearly log phase at 30°C, shifted to 42°C, and then induced to produce ,3-galactosidase with isopropyl ,,D-thiogalactopy-ranoside. The effect of A* protein on enzyme production was assayed colorimetrically. There was not a noticeable difference in
P-galactosidase
levels incellscarryingeither of the plasmids at either temperature (Fig. 3B). This may indicate that A* acts specifically to inhibit host DNA syn-thesis and does not significantly affect transcription or translation. Consistent with this conclusion is the fact that the cells bearing pJA*1 form filaments (Fig. 4) and continue to increase in mass at42°C several hoursafter induction of A* synthesis.Are SOS functions induced? It is usually the case that when cellular DNA is damaged orcell DNA replication is inhibited or both, the cell responds by expressing a set of genes under the control of the lexA repressor (21). The phenomenoniscalled the SOSresponse and inlargepart is directed atfacilitating repairof thedamagedDNA. Wehave investigated whether thisresponse occurs afterinductionof
on November 10, 2019 by guest
http://jvi.asm.org/
[image:5.612.147.469.71.467.2]ja
b
c
d
ellf
gh
i
la"elIf'
FIG. 5. Quantification of RecA protein by immunoblotting. EquivalentamountsofE.coliRR1carrying pCQV2(lanesathrough e)orEcoli RR1carryingpJA*1 (lanesfthrough j)were
electropho-resed on an sodium dodecyl sulfate-polyacrylamide gel and im-munoblotted withantiserum raisedagainstE.coliRecAprotein (see the text).Cells in early log phase eitherwereexposedtomitomycin C(1.5 ,ug/ml) for30min (lanes bandg) and 90min (lanescandh)
or wereshiftedto42°Cfor 30min (lanesd andi)and 90min (lanes
eandj). Lanesaand frepresentcellskeptat 30°Cfor 90min.The fourlanes onthe right representtwofold dilutions ofthesamples
seeninlanesa,e,f,andj. Arrowsonthe left indicate thepositions
of themolecular weight markers (Bio-Rad) phosphorylase b (Mr, 92,500), bovine serum albumin (Mr, 66,2000), ovalbumin (Mr, 45,000),andcarbonicanhydrase(Mr, 31,000).
A* expression. One of the diagnostic features of the SOS responseis increasedsynthesis of the RecA protein, and the experiment presented in Fig. 5 reveals that this occurred when A*wasinduced,although theaugmentation of
synthe-siswasnotasgreataswasfoundwhen the cellsweretreated
with mitomycin C,a DNA-damagingagent. This is
consist-ent with a partial turn-on of the SOS function and could
accountfor the filamentationobserved, since LexA repres-sion of the sfiA gene, whose product inhibits septation, wouldbe lifted.However, thecomplete explanation is likely
more complex, since A*-induced filamentation was also observed in recA56 and lexA3(Ind-) strains in which the SOSresponse cannotbeinduced, and in whichmitomycin C
treatmentdidnotcausefilamentation (datanot shown). DISCUSSION
Earlyattemptsbyourlaboratory and others(33) indicated that successful cloning ofgene A* required that its lethal action berepressed. To accomplish this,weused thevector
pCQV2 (27), which uses the A
PR
and cI857 controlling elementstoregulate the expression of clonedgenes. SeveralPL-containing
vectors(pPLc24,pKC30,andpJL6-8)used in earlierexperiments apparently provided insufficient repres-sionofA* synthesistoallow isolation ofstableclones;this may be because these vectors rely on c1857 repressor provided byasingle-copy defective lysogen, whereas pCQV2carries a copy of the repressor gene itself, thus increasing thegene copynumberof therepressorgene.
When expression of A* protein in cells carrying pJA*1
was induced by shifting the cellsto 42°C, large amountsof theproteinwerenotmade, raising thepossibility of
autoreg-ulation of synthesis. Detection of expression required the immunoblotting technique (Fig. 2) to visualize the A* pro-tein, which was otherwise obscured by E. coli proteins.
Upon expression ofA*, cellularDNA synthesiswas
inhib-ited,and thecellsstoppeddividing.Howeverthey increased in mass,formedfilaments, and,wheninducedto
synthesize
a newprotein (here by isopropyl
,,D-thiogalactopyranoside
induction of
P-galactosidase
synthesis), responded as did control cells(Fig.
3). Ghosh and Poddar (14) found that 4X174 infection caused inhibition ofP-galactosidase
produc-tion in E. coli, and they speculated it was caused by a phage-encoded protein. Ourresults suggestthat it isnotA* thatisreponsible for this shutdown. Apparently, the mech-anism that A* uses to shut down E. coli DNAreplication
doesnotresult ina
generalized
breakdownof the bacterial DNA, and doesnotinterfere withtranscription
and transla-tion.The greatersensitivity of ssDNAto cleavage bythe gene A*
protein
(12, 16, 17) encourages thethought thatthe A* proteinmay actspecificallyatsingle-stranded regions in the E. coli growing fork, breaking the DNA at that point and blocking furtherreplication. Otherpossibilities
include phys-icalobstruction merelyasthe resultofA*binding
toduplex DNA(11)or totheRepprotein,
ahelicasenormally
involved in separating the strands of the bacterial chromosome. Inhibition ofDNA replication causes induction ofthe SOS response, which includes the increasedproduction
ofthe SfiAprotein,
an inhibitor ofseptation
(9, 15). However, sinceseptation
is inhibited under conditions in which the LexArepression
of thesfiA
geneis notlifted(in
recA56orlexA3 mutants), it may be that inhibition of cell division occurs astheresultofan SOS-independent
coupling
of cell divisionto DNAreplication
(5).The role of the
4X174
gene A*protein
in the viral lifecycle
is unknown.That it is madeby
alltheisometricphages
examined and that it
seemingly
has a counterpart in the filamentous phages(8) implies thatit isadvantageous ifnot necessaryforsomeaspectof the viral lifecycle.
Itsactionson ssDNA suggest a role in
superinfection
exclusion(7).
Other
possible
roles includeshutting
off host cell DNAsynthesis
andfacilitating
the transition fromRFreplication
toDNA
synthesis
(10, 11, 13, 17, 23). Earlierresearch,
themore recent studies of
Eisenberg
and Ascarelli(11)
andLangeveld
etal.(16),and our owndataarguestrongly
foraneffect on host DNA
synthesis;
presumably
a cessation of host DNAsynthesis increasesthepool of nucleotide precur-sors available forphage DNAsynthesis.
Evidence
against
adirect roleforthe A*protein
inXX174 ssDNAsynthesis
comesfrom in vitrostudiesindicating
that it is notrequired
forRFreplication
orssDNAsynthesis
in vitro (1). Oneexplanation
fortheinferredessential function of the A*protein
could be the existence of a host cellprotein,
required
forreplication
ofthe bacterialDNA,
that interferes with viralssDNAsynthesis (1).
Therole of the A*protein
would be topreventthis interference.Oligonucleo-tide-directed in vitro
mutagenesis
of the A*coding
sequence inplasmid pJA*1
will makepossible
incisiveexperiments
into A* function.
ACKNOWLEDGMENTS
Thisresearchwassupported byagranttoD.T.D. anda Student-ship to J.C., both awarded by the Medical Research Council of Canada.Supportwasalsoprovided bytheNational Cancer Institute of Canada.
We thank those individuals who provided us with strains and
plasmids. WearegratefultoC.F.Robinowfor thephotographyin
Fig.4,toShlomoEisenbergforthegiftsof antiserumagainstthe4OX
AproteinandasampleofAprotein,andtoJeff Roberts for thegift
of anti-RecA antibody.
[125I]protein
Awas agiftfromSam Dales. Wealso thankDylanEdwards andNancyNgfor usefuldiscussions,on November 10, 2019 by guest
http://jvi.asm.org/
[image:6.612.70.298.73.223.2]Sharon Wilton and Sven Beushausen forhelp withimmunoblotting, andGeorge Mackie for doing the in vitro transcription-translation reaction. The figures and manuscript were skillfully prepared by Dale Marsh and Linda Bonis.
LITERATURE CITED
1. Aoyama, A., R. K. Hamatake, andM. Hayashi. 1983. Invitro synthesis of bacteriophage 4)X174 by purified components. Proc. Natl. Acad.Sci. U.S.A. 80:4195-4199.
2. Batteiger, B.,W.J.Newhall V,and R. B.Jones. 1982.Theuse of Tween 20 as ablocking agent in theimmunological detection of proteins transferredtonitrocellulose membranes. J. Immun. Methods 55:297-307.
3. Bollen, A., R. Loriau, A. Herzog, and P.Herion. 1984. Expres-sion of human a1-antitrypsin in Escherichia coli. FEBS Lett. 166:67-70.
4. Brown, D. R., J. Hurwitz, D. Reinberg, and S. L. Zipursky. 1982. Nuclease activities involved in DNA replication, p. 187-202.InS. M. Linn and R. J. Roberts (ed.), The nucleases. ColdSpring Harbor Laboratory, Cold SpringHarbor,N.Y. 5. Burton, P., and I. B. Holland. 1983. Twopathways of division
inhibition in UV-irradiated E. coli. Mol. Gen. Genet. 190: 309-314.
6. Denhardt, D. T.1969. Formation of ribosylthymine in Escheri-chiacoli.Studiesonpulselabeling with thymine and thymidine. J. Biol. Chem. 244:2710-2715.
7. Denhardt, D. T. 1977. The isometric single-stranded DNA phages, p. 1-104. In H. Fraenkel-Conrat and R. R. Wagner (ed.),Comprehensive virology, vol.7.PlenumPublishing Corp., NewYork.
8. Denhardt,D.T. 1978.Overview:acomparisonof the isometric and filamentous phages, p. 645-655. In D. T. Denhardt, D. Dresslerand D.S. Ray (ed.), The single-stranded DNA phages. ColdSpringHarborLaboratory, ColdSpringHarbor, N.Y. 9. Devoret,R. 1983. Workshop summary: inducible responsesto
DNAdamage, p. 571-576.In E.C.Friedberg andB.A.Bridges (ed.), Cellular responses toDNAdamage. Alan R. Liss, Inc., NewYork.
10. Dubeau, L., and D. T. Denhardt. 1981. The mechanism of replication of4)X174.XVIII.Gene A and A* proteins of
+X174
replicative formDNA. Biochim.Biophys. Acta 653:52-60. 11. Eisenberg, S.,and R.Ascarelli. 1981. The A*protein of4)X174
is an inhibitor of DNA replication. Nucleic Acids Res. 9:1991-2002.
12. Eisenberg, S., andM.Finer.1980. Cleavage and circularization ofsingle-stranded DNA: anovel enzymatic activity of
+X174
A*protein. Nucleic Acids Res. 8:5305-5316.
13. Funk, F. D., and D. Snover. 1976. Pleiotrophic effects of mutantsin geneAofbacteriophage4)X174.J.Virol. 18:141-150. 14. Ghosh, A., and R. K. Poddar. 1977. Reduced synthesis of ,B-galactosidase inEscherichiacoli infected withphage
+X174.
Can. J. Microbiol.23:1069-1077.
15. Huisman, O.,R. D'Ari, andS. Gottesman. 1984. Cell-division control in Escherichia coli: specific induction of the SOS function SfiAproteinis sufficienttoblockseptation. Proc. Natl. Acad. Sci. U.S.A. 81:4490-4494.
16. Langeveld,S.A.,A.D. M. vanMansfeld, A. vander Ende, J. H. van dePol, G. A. van Arkel, and P. J. Weisbeek. 1981. The nuclease specificity of the bacteriophage
+X174
A* protein. NucleicAcids Res. 9:545-562.17. Langeveld,S.A.,A. D. M. vanMansfeld, J. M. de Winter, and P.J.Weisbeek.1979.Cleavage of single-stranded DNA by the A and A* proteinsof bacteriophage 4)X174. Nucleic Acids Res. 7:2177-2189.
18. Lautenberger, J. A.,D.Court, and T. S. Papas.1983. High-level expression in Escherichia coli of the carboxy-terminal
se-quences of the avian myelocytomatosis virus (MC29) v-myc protein. Gene 23:75-84.
19. Lindqvist, B. H., and R. L. Sinsheimer. 1967. Process of infection with bacteriophage
+X174.
XIV. Studieson macromo-lecularsynthesis during infection withalysis-defectivemutant. J.Mol. Biol. 28:87-94.20. Linney, E.,and M. Hayashi. 1974. Intragenicregulationof the synthesis of
+X174
gene A proteins. Nature (London) 249: 345-348.21. Little, J. W., and D. W. Mount. 1982. The SOS regulatory systemof Escherichia coli. Cell 29:11-22.
22. Maniatis, T.,E. F.Fritsch,andJ. Sambrook. 1982. Molecular cloning,alaboratory manual. Cold Spring HarborLaboratory, ColdSpring Harbor, N.Y.
23. Martin,D.F.,and G. N. Godson.1975.Identification ofa
+X174
codedprotein involved in the shut off of host DNA replication. Biochem.Biophys. Res. Commun. 65:323-330.24. Meilado, R. P., and M. Salas. 1982. High level synthesis in Escherichiacoli of the Bacillus subtilis phage+29proteins p3 and p4 under the control of the phage lambda PL promoter. Nucleic Acids Res. 10:5773-5784.
25. Miller, J. H. 1972. Experiments in molecular genetics, p. 352-355. ColdSpringHarborLaboratory,ColdSpring Harbor, N.Y.
26. Neidhardt, F. C., T. A. Phillips, R. A. VanBogelen, M. W. Smith,Y.Georgalis,and A. R. Subramanian. 1981. Identityof theB56.5protein, theA-protein, and thegroE geneproduct of Escherichiacoli. J. Bacteriol. 145:513-520.
27. Queen, C. 1983. A vectorthatuses phagesignals for efficient synthesisofproteins in Escherichia coli. J. Mol.Appl. Genet. 2:1-10.
28. Remaut, E.,P. Stanssens,and W. Fiers. 1981. Plasmidvectors forhigh-efficiency expressioncontrolledby the PL promoter of coliphagelambda.Gene 15:81-93.
29. Remaut, E., P. Stanssens, and W. Fiers. 1983. Inducible high levelsynthesis ofmaturehumanfibroblast interferon in Esch-erichiacoli. Nucleic Acids Res. 11:4677-4688.
30. Shimatake, H.,and M.Rosenberg. 1981. PurifiedX regulatory protein cIIpositively activates promoters for lysogenic devel-opment. Nature(London) 292:128-132.
31. Tessman, E. S., and I. Tessman. 1978. The genes of the isometric phages and their functions, p. 9-25. In D. T. Den-hardt, D. Dressler, and D. S. Ray (ed.), The single-stranded DNA phages. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
32. Towbin, H.,T.Staehelin,andJ. Gordon. 1979.Electrophoretic transfer ofproteins from polyacrylamide gelsto nitrocellulose sheets:procedureandsomeapplications.Proc.Natl. Acad.Sci. U.S.A. 76:4350-4354.
33. van derAvoort,H.G.A.M.,R.Teertstra,R.Versteeg,and P.J. Weisbeek. 1983. Genes and regulatory sequences of bacterio-phage
+X174.
Biochim. Biophys.Acta741:94-102.34. vanderEnde, A., S.A.Langeveld, G. A.vanArkel,and P. J. Weisbeek. 1982. Theinteraction of the A and A* proteins of bacteriophage
+X174
withsingle-stranded and double-stranded 4)XDNAin vitro. Eur. J.Biochem. 124:245-252.35. vanderEnde, A., S.A.Langeveld,R.Teerstra, G. A.van Arkel, and P.J. Weisbeek. 1981. Enzymatic propertiesof the bacterio-phage
+X174
A*proteinonsuperhelical+X174
DNA: amodel for the terminationof the rolling circle DNA replication. Nu-cleic Acids Res. 9:2037-2053.36. vanMansfeld,A.D.M.,H.A. A. M. vanTeeffelen, J. Zandberg, P. D.Baas,H.S. Jansz, G.H. Veeneman,andJ. H. van Boom. 1982. A*protein of bacteriophage+X174carriesan oligonucle-otidewhich itcantransfertothe 3'-OH ofa DNA chain. FEBS Lett. 150:103-108.