Specialized transduction with lambda plac5: dependence on recA and on configuration of lac and att lambda.

0022-538X/81/050497-07$02.00/0 Vol.38,No.2

Specialized Transduction

with Xplac5: Dependence

on

recA

and

on

Configuration

of

lac

and

attX

RONALD D. PORTER,l MICHAEL W. LARK,'tAND K.BROOKS LOW2

Departmentof Microbiology, Cell Biology,Biochemistry and Biophysics,The PennsylvaniaState

University, University Park, Pennsylvania16802,1andRadiobiologyLaboratories, Yale University School

ofMedicine, New Haven, Connecticut 065102

Received7November 1980/Accepted13January1981

The constructionof Xplac5 transducingphages carrying various lacZ allelesis

described. Genetically disabled(N N-P) Xplactransducing phageswereused

tostudythe dependenceof specialized transductiononhost RecA functionand

onthe location ofthe lacZgene in the recipientstrain. In the absence of

site-specific recombination at attX, transductionwascompletely dependentonhost RecAfunction. Regardlessofthe configuration ofattX,Xplac transducing phages recombinedata20-to50-foldhigher frequencywithF421acthan withalacgene

located in the cellular chromosome. Deletionmutants of lacZ in the recipient strainwereusedtoshowthattheprobability of lac recombinationresulting from Aplac infection is apparently proportionaltothe amount ofhomology between theparental lacZgenes.

Specialized transduction is a mode of gene transfer(2, 9)whichhas beenusedinfrequently inrecombination studies. We recently described asystemforthestudyofrecombination kinetics thatemploys Aplac5 transduction (13), and here we report a more detailed characterization of the Aplac5 transduction system. The isolation and characterization ofAplac derivatives con-taining specificlacZ alleles isdescribed. Wealso report the construction of

genetically

disabled

Aplac

transducing phages

and their use to ex-amine the

dependence

oftransductionon:

(i)

the hostrecAgene, (ii)thelocation of the lacZallele in the

recipient

strain,

and

(iii)

the

configuration

ofthe attA site in the genome of the recipient cell. The randomness with which recombination can occur along the lacZ gene in specialized transduction is evaluated.

MATERIALS AND METHODS Bacterial strains. TheEscherichiacoli K-12 de-rivatives used in thisstudyarelisted in Table1.The complete pedigreeofall strains isavailableonrequest. Bacteriophage. The bacteriophage lambda and Aplac5derivatives used in thisstudywere prepared from thelysogenicstrains shown in Table 1.

Media. LB medium(11)wasusedforliquidculture medium andagarplateswheneverarich mediumwas required. Modified minimal medium56and its supple-mentationtoproduceselective media have been de-scribed (8). Half-strengthmodified minimal medium 56 (56/2) was used for the dilution and washing of cells. Dilutions ofphagestocksweremadeinAbuffer tPresent address: Department of Microbiology, Case Westem ReserveUniversity,Cleveland,OH44106.

(0.006 M Tris, pH 7.2, 0.01 M MgSO4, and 0.005% Difcogelatin). Top agar was either0.6%or 0.8%Difco agar indeionized water.

Lysate production. Lysates of bacteriophage carrying theXcI857allelewereprepared byheat in-duction ofanappropriate lysogenicbacterial strain. Lysates of AcI+orAi21phagewerepreparedby induc-tion withUVirradiation ofappropriate lysogenic bac-terialstrains.

Preparationoflysogens. Lysogensforparticular bacteriophage derivatives werepreparedbyspotting alysateontoorstabbingaplaqueintoalawn of the desired host strain and then screeningcolonies ob-tainedfrom theareaofclearingfor the presence of the bacteriophageas aprophage.

Plating ofbacteriophage for plaques. Phage stockstobeplatedforplaqueswerediluted in A buffer andplated on anappropriateindicator strain which had been grown to 109 cells per ml in LB medium supplementedto0.2%maltose and 0.01 MMgSO4.

Testing plaques for the Lac character of the phage.PlatingsofAplaconA(lac-pro)strainsonLB plateswereusedto assessthelacZ+orlacZ character ofindividualplaquesorpopulationsofplaques.Phage platings that had been incubated for10 to 16hat37°C wereflooded with 0.013 M

o-nitrophenol-fi-D-galacto-pyranoside (ONPG) in 0.25 M potassium phosphate buffer(pH7.25).PlaquesfromAplacZ+phagegavea strongyellowcolorinlessthan1minat room temper-ature. Plaques fromAvir or AplacZ gave no yellow color aftermorethan2hof incubationat room tem-perature. Platings containing both

AplacZ'

and AplacZhadto be evaluated within approximately 5 min because theyellowcolor from AplacZ' plaques diffused rapidly across the plate. Stabs of AplacZ plaques from mixedpopulationsintoafresh lawn of cells demonstrated no yellow color after prolonged room-temperature incubation with ONPG. XplacZ

497

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.500.57.456.77.527.2]

498 PORTER, LARK, AND LOW

TABLE 1. E. coli K-12strains

Relevantcharacteristics F-A(lac-pro) supFtrppyrFhisrpsLthi A-RDP100(AcI857Sam7plac5I ZV Y-) RDP100(AcI857Sam7plac5 I Z118Y-) F-A(lac-pro)XlllsupG rpsLthi A-RDP100(Xi21 Pam8O)

C600(AcI857ind Nam7Nam53)

RDP100(XcI857Sam7Nam7Nam53 plac5 I- Z+ Y-) RDP 100(AcI857Sam7Nam7Nam53plac5 I Z118Y-) RDP 100(XcI857Nam7Nam53Pam8Oplac5 I Z+ Y-) RDP 100(XcI857 Nam7Nam53Pam8Oplac5 I Z118Y-)

F-lacZ813 lacI3promettiprpsLthi

A-KL761butrecAl

F-lacZ813 lacI3metA(gal-attA-bio)rpsLthi A-KL763but recAl

F-lacZ813 lacI3promethis trprpsLthi(Xind) KL765 but recAl

F421acZ813lacI3/A(lac-pro)mettrprpsLthi A-KL767 but recAl

F421acZ813lacI3/KL773 KL769 but recAl

F42lacZ813lacI3/A(lac-pro)methistrprpsLthi(Aind) KL771but recAl

KL773 F-A(lac-pro)A(gal-attX-bio) metrpsLthi A-RDP 100 F-A(lac-pro) X1llleuthiacrA(?) supE44 RDP101 F-A(lac-pro)XlllleuthisupE44 RDP 103 F42iacZ118lacI3(KL773)

RDP104 F-lacZ118 lacI3A(gal-attA-bio)metrpsL thi RDP105 F1281acZAH111proA',B I/KL773

RDP 106 F1281acZAH119proA+,B+/KL773 RDP 107 F1281acZAH114proA+,B+/KL773 RDP 108 F1281acZAH145proA+,B+/KL773 RDP 109 F1281acZAH125proA+,B /KL773 RDP110 F1281acZAH138proA+,B+/KL773 RDP111 F1281acZAH220proA+,B /KL773

a Strr,Streptomycin resistance.

Source,reference,comments CSH28(11) cured of F' Phage fromCSH66 (11) Seetext

Derived from BW113 (8)xX7151 (from J. Miller) and

sponta-neousStrr'

Phage from L.Herskowitz

Phage from C. M. Radding Seetext

Seetext Seetext Seetext

All of these strains are lacZ813-ochre (21) derivatives of KL318 (1). In eachcasethe finalRec+/1 recAl pairwasmade bya mat-ing with MA1079, which is a

serAderivative (W. K. Maas) of KL16-99 (7).

From KL320 (1)

C600derivatives with A(lac-pro) Xlllderived from PK191 (8).

FromKL334(1)xKL773 FromCHS14xKL773 FromCSH15xKL773

FromCSH16xKL773 FromCSH17xKL773 FromCSH18xKL773 FromCSH19xKL773 FromCSH20xKL773

phage containingnonsensemutations in the lacZgene could be readily detected by using an appropriate

suppressorstrain for plating and looking for yellow

colorwith ONPG.

IsolationofAplac derivatives containing par-ticular lacZalleles. Strains containingadeletionof thegal-attX-bio region of the chromosome and pos-sessing the desired lacZ alleleonthe chromosomeor

on anFlacwerelysogenized with 'placZ'asdescribed

above.These lysogenswereusedto preparelysates,

andtheresultingphagewereplatedon anappropriate strainfortestingtheLaccharacter oftheplaques with ONPGflooding. Plaques thatwereLac- werestabbed

intofresh lawns forretesting, andlysogenswere pre-paredforthosethattested Lac- twice.

Lysatesfromthenewly obtained

~placZ

lysogens weretested with ONPGafter platingonappropriate

suppressorandnonsuppressorstrainswhen lacZ

non-sense mutationswere involved. Presumptive XplacZ

phagewerealso tested by transducingstrains contain-ing thesamelacZ alleleaswell asstrainscontaining other lacZalleles(see below).

Phagecrosses.Phagecrosses weredoneby infect-ingasuitablehost strainat2x 10i cellsperml in LB mediumsupplemented to0.2% maltose and 0.01 M MgSO4withbothparental phagesatamultiplicity of Strain

KL528 KL550 KL551 KL754

KL755 KL756 KL757 KL758 KL759 KL760

KL761

KL762 KL763 KL764 KL765 KL766 KL767 KL768 KL769 KL770 KL771 KL772

on November 10, 2019 by guest

http://jvi.asm.org/

infection (MOI) of5.The plaques from these lysates weretestedasdescribed above, and the desired recom-binantphageswererecovered fromstabsaslysogens. Production of disabled AplaZ transducing

agents. The original \placZisolatesarecapable of lysing cellsthatthey infect,andthiscomplicates trans-ductionstudieswhenpotential transductantsarelost. To avoid thelossof potentialtransductants by phage-mediatedcelldeath,disabledversionsof Aplac5were made. TheNam7Nam53 markerswereintroduced by crossing

AplacZ+

with AcI857 indNam7Nam53 and scoring for lacZ -containing phage that required an

ambersuppressorin the host strainforgrowth.The resulting AplacZ+ Nam7Nam53 phage was then crossed with XplacZ118 to give a lacZ phage that requiredan ambersuppressorin the host strainfor growth.XplacZ+Nam7Nam53wasthen crossed with Ai2M Pam8OtoyieldalacZ+-containingphage which specificallyrequired supEforgrowth (Nam7Nam53is suppressed by either supEorsupF, whereas Pam8Ois

suppressed by supE butnotsupF)and whichfailedto complementeitheraA NaNamam53 phageortheAi21

Pam8O phage when the two types were spotted together on a sup+ strain. The resulting AplacZ+ Nam7Nam53Pam80 phage was then crossed with AplacZ118 Nam7Nam53 toobtainAplacZ118 Nam7-Nam53Pam80. Thesetransducing phages containing both Nam7Nam53 and Pam8O alleles produced no reductioninviable cell number whentheywereused to infect sup+ strains, andthey were used for the transductionexperimentstobedescribed below.

Transductionprocedure.Therecipientcells were

grownto2x10i cells perml in LB medium supple-mentedto0.2%maltoseand0.01MMgSO4at370C. Thetransducingphage preparationwasaddedatthe desired MOIasindicatedforindividual experiments,

and incubation was continued at 37°C for 30 min. Dilutionsweremade with56/2medium,andplatings were doneonminimal lactoseselectionplates which

were incubated at the desired temperature for the developmentofcolonies.

Chemicals and media. Tryptone, yeast extract, andagarwereobtainedfromDifco. OPNG andsugars were obtained from Sigma Chemical Co. All other

chemicalswerereagentgrade. RESULTS

Isolation and characterization ofAplc5 derivativescarryinglacZalleles. Our efforts toconstruct Xplac5derivativescontaininglacZ

alleleshave been concentrated on lacZ118and

lacZ813-ochre because of their suitability for recombination studies (13). The methodology, which is similartothetechniquesusedbyothers (5, 10, 14),is also suitable for other lacZalleles, however,andwehave alsoemployedit to obtain

aderivative of Xplac5carryingthe lacZ36allele. Strains carrying a deletion of thegal-attA-bio

regionof the chromosomeandoneof these lacZ

alleleseitheronthe chromosome oron F421ac werelysogenizedwithAplac5 containing lacZ+.

Such lysogens were characterized as

Lac'

and

Aplac5 TRANSDUCTION 499

temperature sensitive at420C since the phage carries the AcI857 marker. The lysogens ob-tained segregated Lac-, temperature-resistant colonies at a variable rate (1 to 50%). Phage lysateswereprepared and platedonKL528, and

the plaques were scored for lacZ+ or lacZ by

flooding with ONPG as described in Materials

and Methods. Theplaquescontainingthe

origi-nal lacZ+ version of

?plac5

developed yellow

color very rapidly, whereas those that had

pickedupthe lacZallele showednoyellow color.

In thecasesoflacZ118 andlacZ813-ochre,we

examined 1,000 plaques from each ofatleast 10 independent isolates containing the lacZallele

inthe chromosome and 10 independent isolates

containingthelacZalleleonF421ac.The results areshown in Table2.

Characterization of

ApkacZ

isolates. Rep-resentative plaques that appeared Lac- when platedonKL528and flooded with ONPGwere

transferredtofresh lawns of KL754 andKL528 withsteriletoothpicks and retested with ONPG after 12 h of incubationat370C. Both lacZ118 and lacZ813-ochrearesuppressible by the supG

allele of KL754, and the stabs of phage carrying lacZ118orlacZ813-ochreproducedyellowcolor within 5min of flooding with ONPG. This result

serves to demonstrate the sensitivity of the ONPGflooding method for detecting fl-galacto-sidase since the efficiency of suppression of lacZ813-ochre bysupGis only about 1% (R. D. Porter,unpublished data).

Lysogens of selected XplacZ118 and AplacZ813-ochre isolates were prepared as de-scribed in Materialsand Methods.Lysateswere thenprepared and platedonboth KL528(supF) and KL754(supG) for testing with ONPG. With both of thesephage isolates, the plaquesgave no

TABLE 2. Lac plaquesobtainedfromlysatesof AcI857Sam7placZ+lysogensofvariousstrains'

Lysogen Lac/totalplaques on strain:

isolate KL769 KL763 RDP 103 RDP 104 1 28/1,019 40/1,186 174/1,896 86/1,728 2 108/1,118 18/1,091 24/1,226 13/1,502 3 30/1,192 22/1,553 5/1,716 4/1,073

4 20/1,758 11/1,336 241/1,236 0/1,116 5 51/1,893 10/1,010 125/1,013 16/1,072 6 49/1,055 10/1,788 45/1,067 9/1,054 7 29/1,772 13/1,102 175/1,096 23/1,178

8 32/1,510 0/1,674 27/1,095 56/1,631

9 44/1,044 18/1,046 628/1,320 15/2,243 10 131/1,166 20/1,171 59/1,956 2/1,050

11 34/1,152 31/1,653

aAcI857Sam7placZ lysogensof the indicatedstrainswere

prepared using phage preparedfrom KL550. Lysates from

each lysogenwereplated onKL528, andtheplatingswere flooded with ONPG afterapproximately12h of incubationat 37°C. The numbers given represent Lac- plaques/total plaques.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.500.258.454.486.611.2]

yellow color when KL528platingswereflooded

withONPG,and theplaques yieldedauniform

level of yellow color when KL754 platingswere

tested.Theselysateswerealso usedin transduc-tionexperiments with recipientstrainscarrying the same lacZ allele as the transducing phage

and recipient strains carrying the other lacZ allele. In theselfingexperiments where both the transducingphage and the recipientstrain

car-ried the same lacZ allele,

Lac'

colonies were

observedatonly thesamelevelascontrol

plat-ings for spontaneous revertants (see Table 3). When the transducing phage and therecipient strain carried thetwodifferent lacZalleles,

Lac'

transductants were observed at levels several orders of magnitude greater than spontaneous reversion. The details of transduction experi-mentsaregiven below.

Dependence of Aplac5 transduction on

recA,attA,andthelocation of therecipient lacZ gene. We have previously reported that Aplac5transducing phage recombine more effi-cientlywithF421acorF1281acthan withalacZ

genelocatedonthechromosome,andthat lacZ

x lacZ recombination is totally dependent on

J. VIROL.

RecAfunctionin thetransduction system(13). Inthosestudies, nondisabled Aplac transducing

phages

wereused andaresidentAind

prophage

in the recipient strain was used to control the

lytic

functions of the transducing phage. We have now used thedisabled transducing phage

tomakeamoredetailedanalysisofthe

param-etersaffectingrecombinationfrequencyin

Xplac

transduction.

Aseries of strainswasconstructed to testthe importance of: (i) FlacZ813-ochre/Alac versus

F-lacZ813-ochre as the recipient DNA struc-ture, (ii) recA+ orrecA, and (iii) anormal attA

siteasopposedtoaXindprophageor a

gal-attX-biodeletion.Strainscontainingallcombinations

of these parameters were transduced with

XplacZ118

Nam7Nam53Pam80 and

XplacZ'

Nam7Nam53Pam80 atan MOI of2,and plat-ings on miniimal lactose selection plates were

incubated at 37°C. The results of these trans-duction experiments are shown in Table 3. It

canbe seenthat lacZx lacZ recombination is totally dependent on recA+, since only

back-ground reversion levels of

Lac'

colonies were

presentwhen recAstrainswereusedas

recipi-TABLE 3. Dependence ofXplactransductiononseveralparametersa

Lac'colonies perml' Strain RecipientLacgene RecA Aattachment site Expt.no.

AplacZ+

XplacZ1l8 Nophagec

KL761 Chromosome + attA 1 2.4x 105 6.5x 104 8

2 6.7x 104 3.8x104 12

KL762 Chromosome recAl attA 1 8.3x 104 7C 5

2 3.0 x 104 3C 3

KL763 Chromosome + A(gal-attA-bio) 1 4.2 x104 3.7x104 6

2 3.5 x104 4.1x 104 5

KL764 Chromosome recAl A(gal-attA-bio) 1 od 1c 3

2 4c 2c 3

KL765 Chromosome + Aind 1 4.2x105 5.5 x 104 2

2 6.6 x105 7.8 x104 3

KL766 Chromosome recAl Aind 1 7.1x 104 3c 2

2 9.4 x104 3c 2

KL767 Flac + attA 1 2.0x106 5.4x 106 9

2 1.9x 106 6.0x 106 16

KL768 Flac recAl attA 1 7.1 x104 2c 4

2 4.0x104 1c 7

KL769 Flac + A(gal-attA-bio) 1 9.8 x 105 2.5x

105

13

2 6.5x 105 2.2x106 3

KL770 Flac recAl A(gal-attA-bio) 1 od 3c 9

2 o0 2c 4

KL771 Flac + Aind 1 2.0x 106 3.8x106 13

2 2.3x 106 1.2x106 6

KL772 Flac recAl Aind 1 2.1x 104 3C 4

2 1.1x 104

5'

4

a Theindicated strainsweretransduced with

XcI857

Nam7Nam53Pam8O

placZ+ from KL759 andAcI857 Nam7Nam53Pam80 placZ118 from KL760 as described in the text at an MOI of 2. The minimal lactose selectionplateswereincubatedat 37°C for 48 to 72 h beforecolonieswerecounted.Allrecipientstrainscarry lacZ813-ochre(21).

bExceptasindicated.

cPer0.2ml.

dPer

0.1ml.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.500.64.455.351.598.2]

Aplac5 TRANSDUCTION 501

entsintransduction. When

AplacZ'

transducing phagewere used withrecA recipients that

con-tainanormalattXoraXindprophage, significant

levels of

Lac'

transductants were seen. These

Lac'

transductants didnotappearinarecAcell

containing the gal-attX-bio deletion, indicating thatthistypeoftransduction is totally depend-entonsite-specific recombination of the

trans-ducingphageat attX(3, 4).

Therewas astriking effect of location of the

recipient lac region on Lac+ transduction

fre-quency.A 20-to50-fold-higher transduction

fre-quency for Flac-carrying strains was

ob-served,comparedtoF-lac, in thecaseswithattX

orA(gal-attX-bio). In the Xind strains this

dif-ferential holds for transduction withAplacZ118, but the effect appears smaller (only fourfold)

with XplacZ+. However, this difference with

XplacZ'

andAplacZ118 hinges on therate ob-served in recA+ cells containing the resident

prophage where

Lac'

colonies can arise from

general recombination between

\placZ'

and the residentprophage but notfromrecombination between

XplacZ118

and theresidentprophage. Thisenhanced recombination of\placwith Flac isnotdue toanyfunctionssuppliedintransby the F factor since the presence of a

non-lac-containing F factor (either in the autonomous stateasF15thyorintegrated into thegenomeas

a Cavalli Hfr) had no effect on the level of

recombination observed between Aplac and a

chromosomal lac allele (13).

Transduction of kacZ deletions with

AXpkcZ+.

Aplac5 carries the entire lacZ gene

and part of the lacIgeneaswellaspartof the

lacYgene.ThelacZgenemakesupthe bulk of the lac material carried by Xplac5 since it is about 3,700 base pairsout of about 4,200 base pairs of total Lac DNA (15). To rule out the possibility of recombinational hotspots in our

transduction assay, we wanted to determine whether ornot recombinationcanoccurfairly

randomly across the Lacregion carried by the

transducing phage.To test thispossibility, trans-duction experiments were done withrecipients

thatcarrieddeletionsstartingatapoint beyond

the lacIgene and extendingvarious distances intothelacZgene(11).ThelacZ deletionswere

carried on F1281ac-pro in a A(lac-pro) strain

containingadeletion of thegal-attA-bioregion.

These strains were transduced with

XplacZ'

Nam7Nam53Pam8O at an MOI of 2, and the lactoseselectionplateswereincubatedat370C. The results of these experiments are shown in

Table 4. The level of

Lac'

transductants ob-tainedisroughlyproportionaltothe amount of lacZ material remaining in the recipient F128 episomes and indicates that recombination is

reasonably randomacrossthe Lacgenematerial carried by

Xplac5.

DISCUSSION

We have described indetailour version ofa

technique for transferring and lacZ allele onto

Xplac5

by usingattXdeletionstrains containing the desired lacZ allele. Although the frequency of XplacZ phage obtained varies considerably from isolate to isolate (Table 2), the method works well for thetwolacZalleles tested

exten-sively and also works for other lacZalleles (un-published data). For reasons that are not

en-tirelyclear,thechancesofobtaining thedesired lacZ alleleonAplac is three-tofivefold better if the lacZ allele islocated onF421acrather than onthecellulargenome (seeTable 2). Thismay

berelated, however, tothe finding that plac5 initially recombines more readily with F421ac

than withachromosomal lacgene (see Table 3

and reference 13). The mechanism for generat-ing \placZ phage presumably involves

recom-bination between the two tandem lac regions surrounding the prophage inanaddition trans-ductant. TheuseofattXdeletionsgreatly facili-tatesfinding thenecessary

Lac',

prophage-con-taining,addition transductantssinceanyphage

containing transductantsmustnecessarily have theprophage locatedinthe lacregion. It should be noted that when theXplac5 phagewas

inte-gratedat attX,wedidnotfindphage containing

the desired lacZ allele after the examination of

over25,000 plaques fromanappropriate lysate.

Theconstructionofgenetically disabled Aplac transducing phages has facilitated analysis of those factors important in determining trans-duction frequency. XcI857 Nam7Nam53 phage establishlysogenyasplasmidsinnonsuppressing

strains(6, 18) and resultinlittlecelldeath. The

TABLE 4. TransductionoflacZdeletionstrainsa

Lac+transductants

Recipientstrain lacZdeletion

peracn

_{per ml}

RDP105 Hlll 5.8 x 105

RDP106 H119 4.0x 105

RDP107 H114 3.3x 105

RDP108 H145 3.0x105

RDP109 H125 1.2x105

RDP110 H138 1.0 x105

RDP111 H220 5.1 x104

aThe indicated strainsweretransducedwith

AcI857

Nam7Nam53Pam80placZ'preparedfromKL759at an MOI of2 as described in the text. The lactose selectionplateswereincubatedat37°C.Thedeletions have beendescribed(11) andinvolve the deletion of increasingamountsof the lacZgene carriedbyF128 startingfrom Hlll up to H220, whichhas the least amountoflacZDNAremaining.

VOL. 38,1981

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.500.257.449.496.583.2]

502

PORTER, LARK, LOW

additional inclusion of the X Pam8Oallele pre-ventsthe establishment of the

plasmid

stateand providesa

nonreplicating

DNAmolecule which cannotcontributeto

Lac' colony

formation un-less the transducing phage DNA recombines withthe cellularDNAbefore it is lost

by

segre-gation (R. D. Porter,

unpublished data).

The data in Table3confirm our

previous

result that Xplac

transducing phages

recombinemore read-ily with F421ac than with a chromosomal lac gene (13), and thesenewdata showthat this is the case

regardless

ofthe

configuration

of the attXregion of the chromosome. The transduc-tants seenwith

AplacZ'

in a cell that is recA when the cell contains anormal attA site or a Xind prophage wouldappear tobe the result of site-specific recombinationatattAoratthe

hy-brid attachment sites of the

prophage,

since strains thatarebothrecA and

A(gal-attX-bio)

(KL764 andKL770) failto yieldany transduc-tantswith

XplacZV.

Thisclass of transductants probably arises because the

transducing

phage

can

lysogenize

at attA or atthe

hybrid

attach-mentsitesat alow

efficiency

as aresult of Xint gene expressionfrom thep intpromoter(16). It would be

theoretically

possible to obtain

Lac'

transductants by

site-specific

recombination at secondary X attachment sitesevenwhen

attX

is deleted (17), but such

integration

ofthe trans-ducingphageatsecondary attachment sites

ap-parently

does notoccur at significant levels in the present experiments. Similar results dem-onstrating a RecA dependence for plus-times-minus allele recombination have also been re-ported for

\pro

and Agal (19, 20). One of these reports indicated low but significant levels of transduction mediated by

Xred functions

in a recA strain with

Xgal

(19). We conclude from this that

Xred

functions are not significantly expressed inourdisabled transducingphages.

Withonenotable exception (12), minus-times-minustransductional crosses have notbeen uti-lized for recombination studies in previous re-ports. The data in Table3indicatethatlacZx lacZtransductional crosses aretotally depend-enton host RecAgenefunctioneven when site-specific recombination at

attA

ispermitted. The data in Table 3 also point up the unexpected resultthatlacZ x lacZ crosses yield more Lac' recombinants than lacZ+ x lacZ crosses when the recipient lacZ allele is located on F421ac, but not when therecipientlacZ

allele

islocated onthechromosome.We have noexplanationfor this observation, but this phenomenon may pointup someaspectof apossible mechanistic differencein the recombination betweenAplac and F421acversus F-lac.

We have also used Aplac transduction as a measureof theabilityofrecombinationtotake

J. VIROL.

place fairly randomly along the length of the recombining lacZ genes, or at least in rough proportiontothelengthofhomologypresent. A disabled

AplacZ'

transducing phage lysate pre-paredfrom KL759 was used to transduce a series of strainscontainingdeletions of variouslengths in the resident lacZgene. The datain Table 4 indicate that the level oftransductantsobtained is roughly proportional to the amount of the lacZ gene

remaining

in the

recipient

strain. Hence,therecombinationseenbetween Aplac5 derivatives and lacZgenesin recipientstrains appearsnot to involvespecific sites in thelacZ gene.

The Aplac transduction system has several advantages over previously utilized specialized transductionsystems.Numerous well-character-ized lacZallelesareavailable andcanbeplaced on the transducing phages. The separation of the lacregion from attXonthecellulargenome allows theuseof strains deleted for

attX,

which ismoredifficult with

Agal

transduction. Theuse ofattA deletions facilitates theanalysis of gen-eralversussite-specific recombination in lacZ+ x lacZtransductions. Theuseof thegenetically disabled

plaque-forming

transducing phages

de-scribed herein also allows transduction experi-ments to be done without the

killing

of cells in the recipient culture and without thepresence ofhelperphage. Thissystemshouldproveuseful for further studies of recombination mecha-nisms.

ACKNOWLEDGMENTS

We thank TomMcLaughlin,TomWitkowski,and Susan Lambert for their expert technicalassistance, and J. Beckwith forhelpfuldiscussion.

This work wassupportedby Public Health Service grants CA-06519 (to K.B.L.) and GM-26422 (to R.D.P.) from the National Institutes of Health andbyNationalScience Foun-dation grant PCM-73-01609(to K.B.L.).

LITERATURE CITED

1. Birge,E.A., and K. B.Low. 1974.Detection of

tran-scribablerecombinationproductsfollowing conjugation inRec+,RecB- and RecC- strains ofEscherichia coli K12. J.Mol. Biol. 83:447-457.

2. Campbell,A. 1964.Transduction, p. 49-85. In I. C. Gun-salasand R. T. Stanier (ed.), The bacteria, vol. 5. AcademicPress, New York.

3. Echols,H.,R.Gingery,and L.Moore.1968.Integrative recombination functions ofbacteriophageA;evidence forasite-specific recombination enzyme. J. Mol. Biol. 34:251-260.

4.Gottesman,M.E., and M. B. Yarmolinsky. 1968.

In-tegrationnegative mutants ofbacteriophage lambda. J. Mol. Biol. 31:487-505.

5. Jordan,E.,H.Saedler,and P.Starlinger.1968.0° and strong-polar mutations in the gal operon are insertions. Mol. Gen. Genet. 102:353-363.

6. Lieb,M. 1970.A mutants whichpersist asplasmids.J. Virol. 6:218-225.

7. Low,B.1968.Formation ofmerodiploidsinmatingswith

aclass of Rec- recipient strains of Escherichia coli

K-12.Proc. Natl. Acad. Sci. U.S.A.60:160-167.

on November 10, 2019 by guest

http://jvi.asm.org/

8. Low, B. 1973. Rapid mappingofconditionaland

auxo-trophic mutations in Escherichia coli K-12. J.Bacteriol. 113:798-812.

9. Low, K. B., and R. D. Porter. 1978. Modes ofgene

transfer and recombination in bacteria. Annu. Rev. Genet. 12:249-287.

10. Malamy,M.,H.Fiandt,and W.Szybalski.1972.

Elec-tronmicroscopy of polarinsertions in the lacoperonof Escherichia coli.Mol.Gen.Genet. 119:207-222.

11. Miller, J. H. 1972.Experiments inmolecular genetics.

Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

12.Morse, M. L., E. M.Lederberg, and J. Lederberg.

1956.TransductionalheterogenotesinEscherichia coli. Genetics41:758-779.

13.Porter, R. D.,T.McLaughlin,and B.Low.1978.

Trans-duction versus"conjuduction": evidence for multiple roles forexonuclease V in genetic recombination in Escherichiacoli. Cold SpringHarbor Symp. Quant. Biol. 43:1043-1047.

14. Shapiro, J. A.1969.Mutations causedbythe insertion of

genetic material into thegalactoseoperon of Esche-richia coli.J. Mol.Biol.40:93-105.

TRANSDUCTION 503

15. Shapiro,J., L MacHattie, L Eron, G. Ihler, K. Ippen, and J.Beckwith. 1969. Isolation of pure lac operon DNA.Nature(London) 224:768-774.

16. Shimada, K., and A.Campbell. 1974.Int-constitutive mutants ofbacteriophage lambda. Proc. Natl. Acad. Sci.U.S.A. 71:237-241.

17. Shimada, K.,R. A. Weisberg, and M. E. Gottesman. 1972.Prophage lambda at unusual chromosomal loca-tions. I.Location of thesecondary attachment sites and theproperties of thelysogens.J.Mol.Biol. 63:483-503. 18. Signer, E. R. 1969. Plasmidformation: a new mode of

lysogeny by phage A.Nature(London) 223:158-160. 19. Wackernagel, W., and C. M. Radding. 1974.

Transfor-mationandtransduction of Escherichia coli: the nature of recombinants formed byRec, RecF,andARed, p. 111-122.In R. F. Grell(ed.), Mechanisms in recombi-nation. PlenumPress,NewYork.

20. Weisberg, R. A., and N.Sternberg. 1974. Transduction ofrecBW hosts ispromotedbyAred function,p. 107-109.InR. F. Grell(ed.),Mechanims in recombination. PlenumPress, New York.

21.Zipser, D. 1967. Orientation of nonsense codons on the genetic map of the Lac operon.Science157:1176-1177.

on November 10, 2019 by guest

http://jvi.asm.org/