• No results found

Replication and segregational stability of Bacillus plasmid pBAA1

N/A
N/A
Protected

Academic year: 2020

Share "Replication and segregational stability of Bacillus plasmid pBAA1"

Copied!
7
0
0

Loading.... (view fulltext now)

Full text

(1)

Copyright© 1989, American Society for Microbiology

Replication

and

Segregational

Stability of Bacillus Plasmid pBAA1

KEVIN M. DEVINE,* STEPHANE T. HOGAN, DESMOND G. HIGGINS,AND DAVIDJ. McCONNELL DepartmentofGenetics, Trinity College,Dublin 2, Ireland

Received31May 1988/Accepted 13 October 1988

Acrypticplasmid, pBAA1,wasidentified inanindustrial Bacillus strain. The plasmid is 6.8 kilobasesinsize and ispresentincellsatacopynumberof approximately 5perchromosomeequivalent. The

p!asmid

has been maintained under industrialfermentation conditions withoutapparentselectivepressureandsoisassumedto bepartitionproficient. The minimal repliconwaslocalizedtoa1.4-kilobasefragment which also contains the functions required for copynumber control. The verylow level ofsegregational instability of the minimal replicon suggests that it also contains functions involved in plasmid maintenance. Comparison with other plasmids indicates that pBAA1 belongs to the group of small

gram-positive

plasmids which replicate by a

rolling

cycle-type

mechanism. Asequence was identified which isrequired for the efficient conversion of the

single plus strandtothedouble-stranded form during plasmidreplication. Deletionof thissequenceresulted in alowlevelof segregationalplasmid instability.

The development ofcloning vectors for use in Bacillus

subtilis has reliedonplasmidsisolated fromStaphylococcus

aureus(6, 7, 14, 15). Both structural (10, 15, 19, 20, 27)and

segregational (1, 2, 10, 19, 28) instabilities were frequently

observed whenrecombinantvectors weretransformed into

B. subtilis. These instabilities ledtointense investigation of the modeof replication and ofthestability functions ofthese

plasmids. It has emerged that many of these plasmids

replicate byarolling cycle-type mechanism (11, 12, 31).The

essential features of this modeofreplicationare(i) anorigin of plus-strand synthesis, (ii) a replication protein which interacts with the plusorigintogenerate anickwhich allows displacement synthesis of the plus strandtooccur, and(iii)a

signal for efficient conversion of the single strand to the

double-stranded form. These features have been identified

fora variety of plasmids,

including

pT181 (12, 16, 25) and pC194 (11, 12).

Our aim is to analyze the

segregational

instability of

plasmids inB. subtilis. Itwas

hypothesized

that the

segre-gational instability of manyS. aureusplasmidsin B.subtilis is due,atleast inpart, to asuboptimalhost-plasmid

relation-ship.

Thus, wechosetoanalyze thestability functions ofa

plasmid, pBAA1, which wasresident inan industrial strain ofB. subtilis. pBAA1 has beenmaintainedunderindustrial

fermentation conditions without apparent selection

pres-sure. Preliminary experiments demonstrated that the copy

numberwaslow (approximately 5 perchromosome

equiva-lent), soitwas assumed that pBAA1mustencode an active

partitioning function. Inthis paper, we report that allof the

functions required for replication,copy number control, and

partitioningarelocatedon a2.2-kilobase(kb)

fragment.

The DNA sequence and other evidence show that pBAA1

repli-catesbyarolling-circlemechanism and that somefeaturesof itsreplicationfunctions are related both to

4X174

and tothe

gram-positive plasmids pC194,

pUB110,

and

pFTB14.

MATERIALS ANDMETHODS

Bacterial strains and plasmids. The bacterial strains and

plasmidsusedinthis study are described in Table 1. Bacteria

*Corresponding author.

were grown in LB (22). Medium was supplemented with antibiotics to thefollowingfinal concentrations: ampicillin, 50pug/ml; chloramphenicol, 5

jig/ml.

Manipulation of DNA. All manipulations of DNA were

performedasdescribed by Maniatisetal.(21). All enzymes

werepurchasedfrom commercial suppliers.Sequencingwas

carried out by the

dideoxy-chain'termination

method of

Sanger et al. (29). A sequencing kit was used (Amersham

Corp., U.K.); reactions were carried out according to the

instructions of the manufacturer. In some cases, 17-mer

syntheticoligonucleotides

(synthesized

atthe Universityof

Manchester Institute of Science andTechnology)wereused

as

primers.

Analysis ofsingle-stranded DNA production.Lysateswere

prepared by the method ofteRieleetal. (31). Lysateswere

analyzedby electrophoresisin0.8%agarosegelscontaining

0.5 ,ugof ethidium bromidepermlfor

approximately

16 hat

4 V/cm. Todetect single-stranded plasmid DNA only, gels

weretransferredtoBiodyne (PallUltrafine FiltrationCorp., EastHills, N.Y.) withoutprior denaturation bythemethod ofte Riele et al. (31). All other gels were denatured with

NaOHpriortotransfer.DNA wastransferredto

Biodyne

by

diffusion withlox SSC(lx SSC is0.15 MNaCI plus 0.015

M sodium

citrate) according

to the instructions of the

manufacturer.Prehybridization, hybridization, andwashing

wereperformed accordingtomethodAofthemanufacturer.

Autoradiograms were made with RX medical X-ray film

(Fuji Photo

Film,

Japan) with intensifying screensat room temperature.

Copynumberdeterminations.Plasmidcopy numbers were

determinedasfollows. TotalDNAwasprepared separately

fromcellscontainingtheplasmids of interestand was double

digested with BamHI-HindIII, which cuts the plasmids to

giveapBAA1 fragment of approximately2.2kb. Thisdouble

digestion

cuts chromosomal DNA, releasinga fragment of

approximately 4 kb which contains the ,-glucanase gene.

ThedigestedDNA wasseparatedon a

1.0%

agarosegeland

wastransferred to Biodyne. This filterwas thenhybridized

with a nick-translated pJG14 plasmid (which contains both 1.5kbof

P-glucanase

DNAand 1.4 kbofpBAA1 DNA).The

filter was then washed and exposed to X-ray film. The

developed film was scanned densitometrically with a

(2)

BACILLUS PLASMID pBAA1 1167

TABLE 1. Strains and plasmids used

Strainorplasmid Description Source or reference

E.colitgl K-12A(lac-pro) supE thi hsdRF' traD36 proABlacIlacZAM15 Amersham B. subtilis

S0113 trpC2 amy-3 26

8G-5 trpC2 tyr ade his nic ura rib met highly transformablederivative ofB. subtilis168 3

pBAA1 6.8-kb cryptic plasmid isolated from B. subtilis BAA1 BioconLtd., Ireland;

this work pDE68 pBAA1 with theCmrgene frompBD64 cloned into anEcoRIsite(7.8 kb) Thiswork pDE37 The 3.7-kbEcoRl fragment of pBAA1 plus the Cmrgene of pBD64 (4.7 kb) Thiswork pDE22 The 2.2-kbHindl.l fragment ofpBAA1 plus theCmrgeneof pBD64 (3.2 kb) This work pDE14 The 1.4-kb pBAA1 HindIII-BamHI fragment inpRP14 plus the Cmrgene of Thiswork

pBD64.This plasmid underwenta100-bpdeletion attheBamHIsite which does notdetectablyaffectthereplicationproficiency copy number orchloramphenicol resistance oftheplasmid(2.3kb)

pBD64 Plasmid derived frompUB110encoding Kmrand Cmr 13

pRP1 Replicon probe plasmid derived by insertion of theHpaIIfragment of pBD64 Thiswork containing theCmrgeneinto theEcoRIsite of pUC18

pRP2 Replicon probe plasmid derived by insertion of theHpaIIfragment of pBD64 containing theCmrgeneinto theAccIsite of pUC18

pRP3 Replicon probe plasmid derived by cutting pRP2 withSmaI andinserting aHindIII Thiswork linker

pJG14 pRP14containing a 1.5-kb fragment on which is located the P-glucanasegene Thiswork

pUC18 32

M13mpl8 24

M13mpl9 M13-derived vectors used in the dideoxy sequencing method

M13tgl3O 17

M13tgl31

matogram densitometer CD50 (DESAGA, Heidelberg, Fed-eral Republic of Germany) with an integrator to estimate peak area to determine the intensities of signals from the

3-glucanaseandpBAA1 fragments. Three quantities of total DNA (0.1, 1.0, and 4 ,ug) were loaded onto the gel, and the

filterwasexposed for various lengths of time to ensure that the signals were within the linear range of response of the

film. The similar sizes of the

P-glucanase

andpBAA1 frag-ments in the gel ensure that the efficiencies of transfer of

both fragments to Biodyne should be very similar, if not

identical. Likewise, the similar sizes of the ,B-glucanase and pBAA1 fragments ontheprobe plasmid ensure comparable

efficiencies of hybridization. Since the probe contains both the

P-glucanase

andpBAA1homologous DNAs, thespecific

activity of the probe is identical for both species of DNA. Thus the ratio of signals of pBAA1 to 3-glucanase is an

estimate ofthe number of copies ofpBAA1 per chromosome

equivalent.

Analysis of the segregational stability of plasmids. Cells

containing plasmids were grown in LB in the presence of

antibiotic. These cells wereused toinoculate antibiotic-free medium. Ten successive transfers (approximately 100

gen-erations) were performed in antibiotic-free medium over a

period of80 h. Approximately 105 cellswere transferredat

eachstage. Thecellswerethen platedonantibiotic-free LB agar, and individual colonies were tested for resistance to

antibiotic. At every second transfer, 100 colonies were

tested for eachplasmid. Plasmids whose stabilities were to

becomparedwerealwaysanalyzedin thesameexperiment.

RESULTS

Delimitation of the minimal replicon of pBAA1. Plasmid

pBAA1 was isolated from an

ot-amylase-hyperproducing

Bacillus strain. It is 6.8 kb in size,andits restriction map is shown in Fig. la. This plasmid has no known selectable

marker but has been maintained stably under industrial fermentation conditions without apparent selection pres-sure.When marked with achloramphenicol resistance gene

(derived from pC194) and transformed intoB. subtilis, the

resultant plasmid pDE68 was found to be segregationally stable. The copy number ofpBAA1 is approximately 5 per chromosomeequivalent. It is thus assumed that theplasmid isstably maintained byan active partitioning system.

To delimit the minimal replicon, fragments of pBAA1

weresubcloned ontothe replicon probe plasmidpRP1 (Fig.

lb)andweretestedfortheirabilitytosupportreplication by transformation into B. subtilis and selection for

antibiotic-resistanttransformants. It wasfound that plasmids pRP37,

pRP22, and pRP17 containing 3.7, 2.2, and 1.7 kb,

respec-tively, of pBAA1 replicatedin B. subtilis. To further delin-eatetheminimal replicon, progressive deletion of each end of the 1.7-kb HaeIIIfragmentofpBAA1 wascarriedoutby

using Bal 31 (Fig. 1c). The precise endpoints of the Bal

31-deleted derivatives were determined by sequencing.

Re-sults showed that deletion of 350 base pairs (bp) from one end of thefragment allowedreplication,whereasdeletion of afurther 230bp(regionA) abolished theabilitytoreplicate. Similarly,deletionof356

bp (region

B)from theother endof

the fragment abolished the

ability

to

replicate.

To test

whetherpUC18 containedonthe pRPplasmids contributed

to their replication proficiency,

plasmids pDE68,

pDE37,

pDE22,andpDE14wereconstructed(Table 1).Allfour

pDE

plasmidswerealsoreplication

proficient.

Thus,

pUC18

does notaffect thereplication

proficiency

ofthepRP

plasmids.

It was concluded that the minimal replicon of

pBAA1

is

contained on a 1.4-kb

fragment

and that

regions

A and B

contain functions essential for

replication

of

pBAA1

in B. subtilis.

Sequence of the 2.2-kb HindIIIfragmentofpBAA1 which contains theminimalreplicon. The 2.2-kb

HindIII fragment

ofpBAA1 was

sequenced by

the

dideoxy-chain

termination

(3)

(a) L r

1kb

H v RIH

iH

B,

IpBAA1

H H RI

(b)

1kb ,PIHP EI EH ,,

pRP22 rep+

pRP17 rep+

A AA E/

M_ ,, /

(c) E A A A ORF A

E

>j pRP17 rep+

pRP13

rep-E

E pRP14 rep+

200ba pRP11

rep-RegionA RegionB

FIG. 1. (a)Linearizedrestriction mapofplasmid pBAA1. (b)The 3.7-kbEcoRI,2.2-kbHindlIl,and 1.7-kb HaeIIIfragmentsofpBAA1 wereinserted into compatible restriction sites in the multiple cloning site ofthe replicon probeplasmid pRP1 andwere tested for their replication proficiency inB. subtilis. (c) pRP17 wasseparately digestedwith HindIII andBamHI anddigestedwith Bal31todelimit the minimal replicon. After endfilling and ligation, the plasmids weretransformed into Escherichia coli and the sizes of the deletions were

characterized.Aseries ofplasmids withincreasingamountsofpBAA1 deletedweretested for theirabilitytoreplicateinB.subtilis. Deletion ofregionA or Babolished theabilitytoreplicatein B. subtilis. Thesequenceof theregionsessential forreplicationisgiveninFig.3.rep'

andrep-, Replication proficient anddeficient,respectively. A,RsaI;B,BamHI;E, HaeIII;H,HindIll;P, PstI; RI,EcoRI; V,PvuII.

methodofSanger et al. (29). Shotgun-cloned AluI andRsaI

fragments and Bal 31-generated deletion derivatives were

sequenced. Syntheticoligonucleotides wereusedasprimers to sequence regions of nonoverlap. Both strands were

se-quenced, and each base pair was sequenced at least four

times. The sequence is shown in Fig. 2. Analysis of the

sequence reveals an open reading frame (ORF A) of307

amino acids. ORFAispreceded byaputative&43promoter

(P1). Two lines of evidence indicate that this protein is

involved in replication ofpBAA1.First, deletion of region B,

which extends 334 bp from the 3' end ofthe open reading

frame(Fig.2,regionB),abolishesthereplicationproficiency ofthe plasmid. The second line of evidence derives from a

comparison of this protein with the replication proteins of

pUB110, pC194, and pFTB14 (23) (Fig. 3). The

pBAA1-encodedproteinis

92%

homologous attheamino acidlevel

with the replication protein of pFTB14 (homology not

shown) and homologous to alesser though very significant extent with the replication proteins of pUB110 and pC194

(Fig. 3). Thus, theprotein coded by the open reading frame ofpBAA1 is designated RepA.

A 230-bp fragment (Fig. 1c and 2, region A) was also

shown tobeessential for replication of pBAA1. Analysis of the sequence in this region shows four sets of inverted repeats ranging in size from 4 to 14 bp. To discover the sequencesessential for replication, a homology search was

performedto determinewhether there are sequences within

this region conserved among other Bacillus and S. aureus

plasmids known to replicate in B. subtilis. Results (Fig. 4) showed that a 15-bp sequence was conserved in pBAA1,

pUB110,pC194, and pFTB14. No significant homology was

found with pT181, pC221, or pE194. This conserved se-quence is found in the55-bp region of pC194 shown by Gros etal.(11) to have originactivity. Within this 15-bp conserved sequence are two additional features. (i) The sequence CTTGATA is the sequence at which nicking of the plus

strandoccursin the initiation ofreplicationof the coliphage XX174. (ii) An 18-bp sequence (containingthese conserved 15bp) has been shownby Grosetal. (11)tocontainasignal sufficient to terminate replication ofplus-strand synthesis. Thesedatastrongly suggestthat this15-bp conservedregion is part of the origin of pBAA1 and is one of the features within region A which is important for replication. In addi-tion, by analogy with 4X174 and pC194, it suggests that pBAA1 replicates byarolling circle-typemechanism.

Copynumberdetermination ofparentaland deletion deriv-atives of pBAA1. The copy number of pBAA1 and the deletion derivatives were determinedas describedin Mate-rials and Methods.Densitometricscanning of the autoradio-gram indicates that the copy number of pBAA1 in the industrial strain BAA1 is 2 to 3 chromosome equivalents. The copy numbers of pDE68, pDE37, pDE22, and pDE14 in

the laboratory strain S0113 are the same as that of the

parental plasmid in the industrial strain BAA1. Under these

conditions,the copynumberofpC194 is11perchromosome

equivalent,approximatelyhalf the reported value. If normal-izedtothereportedvalue of pC194, then the copy number of pBAA1 and its derivatives is approximately 4 to 6 per chromosomeequivalent. Itcanbe concluded that the

mini-mal replicon of pBAA1 contains thefunctions required for

copy number control.

Some deletion derivatives of pBAA1 produce

(4)

BACILLUS PLASMID pBAA1 1169

Z;T!1TAGAAGLX7ITCTAATCACAAGAACAATAC 44

ATAC Al~~lllpJlCKX]OOCSTATAATATl=rAT~A rI-l-AC-l-l-AAA IZH

L-a-GotEli

Aax~lA~r~r.A~rGAC.~ri~r~ru7A~r~sA.AcT. TrCAACACi&'t~jWlMIyiM.A 284A _ X A l A l < M l . 3b44

AygAGI#AATAGOAGATl".AAXCAAITAYACA1A(TIATITXTIGACI.IrT1XCTArAAT;AAxt(~TrAG0A-AI---.AJJ-AJOT,-MA.iIIGA-1-1-1m GAT GAT'AAATx r-1-TAlTAVJIA 52444448

AAATrAlrmrAAZAArA6ATiVl-t'n rttFAT01CrITAcAmrI s-*AArmo-XIrrcAAAEroA(,ITAAA 542

W.Anm;ArA-nrArlwI"nI-II-XJI-AlrAAAArlrGX-rAAAnMAArrM-.STATk".A 6iH

__RegonA 2

_t~l 1422Co4ye4I5-mor

A>(rlTI7'Asfi ~ i---lli-mT~ r ~ YXI.Al;-IMI Ail*(Ci;]T(TATAT-ViAA J4

ATrA~ll7'rrAAAAIY'A(.l-lAA1.nrT.r71TIAA(XXTrAr.Gy.rr~.%Al5YrlAAAAi n'RAlATT 92H

Al 'AIlAATAhATAATrVk- l 1A A-OMTAAr',rrrTr.AM~rAlrAA OOMH

Rogrotna

A(1GATTfATAGCAT¶TJ.AATAAAAM'AArXTAAAGT2.A1OOAA IrATaAAJYC I(M1m

AIMGX'GGAG CATTMA GM(AMA TTAGAGAM MAGATTOG(GOCA(ITTKTAT OCAM AMG 1144

MetAla Glu HisTyr Gl AlaIue Giti HSer Lys Ile Gly Ala Pro Tyr Tyr Gly Lys lys 211

OCTGM AA CTAATTACTTT(rAGAGTAT TI= TTTMAGAGA(G aM GAGMO;GOC 12044 AlaCluG ysly u lieSeriyeAlaGluTyrlpu Ser PheLyeArtAspProGlu ThrGly 40

AAG T!A MAAi TATCM 1rrCAT mTOT MA GTGAGO TTA1MC(LAM1m7TOrr,10 1268 Lys1-ulyeleu Tyr GinAla HisPheCys LysVat ArgLouCysProMpt Cys Ala Trp 60

(tIC AGO TL TTA MA ATT OCT TATCMrMTMGTTG ATCCTAGAGGM OCC MT(TC CAG 1324 6rgArg Ser Leou lys I e AlaTyrHis AsnLyslpu I eVal Clu Glu AlaAsn ArtGin 80

TAC00C 10C OGAT1GAT!TTiTCAM MAM1M,CTGMT GTAMCGOA GM(TI CLI01 1384 Tyr Gly Cys GlyTrp lie ThLeuLe rlouThrVal Art Asn Val Lys Gly Glu Arg lou 100

AG CACMATT TIT GAGAMGAMGGM00 Tm AML AM11 Tit CAITA AAAMACIA 1448 LysProGin leSerGluMetMet Glu GlyTheAr1 Lys iJu The GinTyr Lys Lys Val 120

AAMT1MT MT `TT 0GGATTT07 AGOTTTAGAGATT AM MT CATGM GM GAT 1I50M Lys Thr Ser Val Ieu Gly PhePheArgAla LUGClulieThrLysAn His Glu Glu Ap 140 ArATAC CATCOT CAT mTT CAT 1G7 TTG ATACrA OTA AOG A MT TATTTT00MAAMC 1568

ThrTyrHis ProHis Te His VaIt u lePro Vat Arg LysAsnTyr PheGly Lysmn 160 TAT ATT MG CAG0OM2 GAG TOG ACGOCCTTTMG AA MG MAG AAATTG GATTMr AT 1628 Tyr IleLye GinAlaGluTrpT'r Ser leuTrpLyelysAiaMet Lys Leu AspTyrTIr 180 .= ATT CT! GATATT17CGA G1TGMA MAOATAMMGATT GAC(X1rGM C1G ATTlAA 1688

Pro TIe ValAsp Ile ArgArgVat Lys GlyLyseAlaLysIle AmpAlaGluLeuIleGlu 200

Mr CATG17; (00 (A WA AM AlT WA0CMMAA OCTTO1TT (T GAAAT!OGAA TAT 1748

AMn AspValArg GluAlaMet Met GluGin1y Aia Val Leu Glu lie|erLys 220

G'"AAGGATUMGAT]GTT 1r. oCAoC MAT AAG G1 mT GMGMCMAAT CMAMrA01a 1808

Asl1.3mApilir AspVal ValArgGlyAsn lys Val Thr Glu Asp Asn Leu mn ThrVal 240

CTT TM TT CATGAT xr.T2 OrA OCT(rr2 AOGTTA ATTOGA TAC017 aOt ATTTT AAGI164

IouTyr IojAsp Asp AlaLeuAlaAlaArgArgtuI le Gly Tyr Gly Gly Ile Leu1 260 GAGATA CATMAAGAG IOMTCT 017 GAT HOMGAGGfGOC GAT(C 1710 MGATT AG 1928

CliIlie HisleGblu lpuAnn luC.ly AspAlaGlu AspGlyAspltuValLyslIeGlu 280

AAA;A4A1; ATOMO; MA(i17 CAAAMr G-Ml CATTTG AGOTTA151 ("l ATT4 (rAlT 1lGCAT GliiGhIsAspUp Clii ValAia AnnGlyAlaCe GluVal Met AlaTyr TrpHlisProGly A1TAAAAAT TAr ATA ATr AAATM44.AAWAACAGa(1TMiJ C A.AMAAl-CCArTC 7AT

I lI.)AsnTyr IIe l1e yeStop 8aie8

14948 o00

'Z(4;

:107

lTiATTArYO-rATS-lr.l-lti-l-l-rAlA-l-TlrM-r-T-prATATATM-AlrATAA11G44XAC-.r

"OWN

FIG. 2. Sequence of the 2.2-kb HindIII fragment ofpBAA1.Regions A and B, which areessential for replication,arebracketed.Inverted repeats aremarked witharrowsdirectlyover thesequence, while direct repeats are boxed. Theputative active site oftheprotein isboxed, and the conserved tyrosine residue is underlined.P1isa consensus(,41promoter; aputativeribosome-binding site is underlined. Thesequence conservedinplasmidspC194,pUB110,andpFTB14(Conserved15-mer)isshown; the arrowwithin thissequenceindicates the siteatwhich nicking oftheplusstrand occursinphage fX174.

strandedplasmid DNA. Intherolling-circle model of plasmid replication, the plus strand is synthesized as a covalently closed circular single strand of DNA. Synthesis of the second strand then initiates at a minus-strand origin se-quence, generating a double-stranded plasmid DNA mole-cule (for a review, see reference 18). Such minus origins,

which are characterized by their potential for secondary

structure formation, havebeen identified forthecoliphages

M13 (9), G4 (8) and (X174 (30). In addition, minus-origin

sequences have been identified for a variety of plasmids,

e.g., pT181,pC221, andpC194 (12). Deletionof the

minus-originsequenceleadstotheaccumulation ofsingle-stranded

plasmid DNA. IfpBAA1 replicates by a rolling cycle-type

mechanism, asissuggested by thehomologyto 4X174and

pC194,then(i)itshouldhaveaminusoriginand(ii)deletion of this minus origin should lead to the accumulation of

single-strandedplasmidDNA. Todetermine the location of

the minus origin ofpBAA1, cells containing the pRP and

pDE series of plasmids were tested for the presence of

single-stranded plasmid DNA, asdescribedin Materials and

Methods.

Separate

gelswere transferredto

Biodyne

withor

withoutprior denaturation of the DNA with NaOH(31). In

the absence of denaturation, only single-stranded nucleic

acid binds to the filter. Figure 5 shows the results for the pDE series ofplasmids. Plasmids pBAA1, pDE68, pDE37,

pDE22, and pDE14 show two major

hybridizing bands,

probably corresponding to CCC monomeric and dimeric

plasmid species.These bands didnotbindto

Biodyne

in the absence of denaturation. Plasmid pDE14, in addition to thesebands,hasaband(arrow)which

migrated

aheadof the

double-stranded DNAspeciesand which boundto

Biodyne

in the absence of denaturation. This fast-migrating band is

preferentially degraded by S1 nuclease under conditions

which cause no detectable degradation of the double-stranded forms (data not shown). These data indicate that this band is single-stranded plasmid DNA. For the pRP series of plasmids, the single-stranded plasmid DNA was observed in cellscontaining pRP17 butnot in cells contain-ing pRP22 or pRP37 (data not shown). Thus, sequences presenton the 2.2-kbHindIII fragmentbut absent from the 1.7-kb HaeIIIfragmentarerequired for normal replication of theplasmid.

Two regions of DNAaredeletedfrom the 2.2-kbHindIII

fragment to give the 1.7-kb HaeIII fragment: a 355-bp

fragment from the left-hand (5') terminus and a 197-bp

fragment from the right-hand (3') terminus (Fig. 2). To

determine the location of the minus origin on the 2.2-kb

HindIII fragment, the sequences of these regions were

analyzed.The355-bpfragmentdeleted in the 1.7-kb HaeIII

subfragmenthas thepotentialfor complexsecondary

struc-tureformation. Sincenounusual featurewasevident in the

197-bp right-hand sequence, we decided to concentrate on

the355-bp sequence. Three

unique

restriction sites,BstEII

(200 bp), EagI (353 bp), and NaeI (770 bp) were used to

generatemutations in the left-hand end of the 2.2-kb HindIII

fragment. Plasmid pRP22 was restricted with

BstEII,

the

endswere filled

in,

and the

plasmid

was

religated,

yielding

pRP22F, which had a 5-bp insertion into the BstEII site. Plasmid pRP22 was also

separately

double

digested

with

BstEII-EagIand withEagI-NaeI,the endsweremade blunt,

and the vector was

religated.

This yielded plasmids

pRP22BE and pRP22EN, with 152- and

414-bp

deletions,

respectively. Whole-cell

lysates

weremade of cells

(5)

1170 DEVINE ET AL.

PW"IISSKWUPIMCE I. T

BilHpSK

--- Is i

P11 S6ZU LLHIL --' ! l

r% SL WL6tKQ XE-FI HEE

HL C6~ 1 tFLTLMR1KRLa

RK*L ___++Mt

Em!LKDYTPI

t .UZ.I+LItE" I OTl

i

1XU|ni4

I---

- --lT aTol~

t--S- --- YLS-t

ULH @+ EK1

ODE R RL .-I T e

FIG. 3. Comparison of the amino acidsequencesof the

replica-tion proteins of pBAA1, pC194, and pUB110. The amino acidsin

boxesareconserved between thepBAA1 replication proteinandthe replication proteins of pC194 andpUB110. The amino acidsequence

of thereplication protein of pFTB14 is 92% homologous tothat of

pBAA1 (homology notshown). Dashes representgaps inserted to optimize the protein alignment.

single-stranded plasmid DNA, whereas plasmids pRP17, pRP22F, and pRP22BE produce detectable quantities of single-stranded plasmid DNA (data not shown). Similarly, deletion of theBstEII-EagI fragment from plasmid pDE22 resulted in the production of single-stranded plasmid DNA (data not shown). Interestingly, the amount of single-strandedplasmid DNA produced by pRP22F is less than that produced by pRP22BE, suggesting that the 5-bp insertion attenuatesbut doesnottotally abolish this function. Itcanbe concluded thatsequences which when deleted result in the

production of plasmid single-stranded DNA, overlap the BstEII site (atbp 200; Fig. 3) but donotextendasfarasthe

EagI site (at bp 352; Fig. 2). Thus, asignal required for the

efficient conversion of the single-stranded plasmid DNAto thedouble-stranded form during replication is locatedtothis region.

dllAA GTCTTTTCTTATLCTITJkTACTATATAGAAAC

MF3TI 6TCTTTT CTTATCTT ACTATATAGCAACT

I I

Ij10GTTCTTTCATCATTIIJATAC'ATATAGAAAT

IIII I Il I I

RLM94 TTTCTTTCTTATLCTTG.LATAATAAGGGTAACT

FIG. 4. Homology at the origin for plus-strand synthesis be-tween plasmids pBAA1, pFTB14,pUB110, and pC194. The 15-bp sequenceconserved in all four plasmids is in boldface. Thesequence

CTTGATA, which is the site atwhich nicking of the plus strand

occursin4.X174, is underlined. Nickingoccursbetween the G and

A residues of this sequence. Vertical lines indicate bases not

homologous with pBAA1.

FIG. 5. Detection ofsingle-stranded plasmidDNAin cells

con-tainingpBAA1 andits deletion derivatives. Whole-celllysates ofB. subtiliscontainingnoplasmid (lanesAandA'),pBAA1 (lanes B and B'), pDE68 (lanes C andC'), pDE37 (lanesDandD'),pDE22(lanes EandE'),orpDE14(lanesFandF')wereseparated byagarosegel electrophoresis, denatured with NaOH (panel 1) or notdenatured (panel 2), and transferredtoBiodyneasdescribed in Materials and Methods. The filterwasthenhybridizedto32P-labeled pRP14. The bandofsingle-stranded plasmid DNA(4)isshown.

Segregational stability of pBAA1 and the deletion deriva-tives. Since pBAA1 wasstably maintained under industrial

fermentation conditionswithout apparentselective pressure, it is assumed to be segregationallystable. Thesegregational stabilities of the deletion derivatives were analyzed as de-scribed in Materials and Methods to detectplasmid-encoded

functions involved in this process. Results(Table2) demon-strate that pRP22 and pDE22 aretotally stable under these conditions. Plasmids pRP17 and pDE14, both of which

produce single-stranded plasmid DNA, displayeda lowbut

reproducible level of instability, with 5 to 20% of the cells

losing plasmid duringthegrowth period. The deletion ofthe

BstEII-EagI fragment from pRP22, which resulted in the

productionofsingle-stranded plasmidDNA, also resulted in

instability.Noinstabilitywasobservedforthe5-bp insertion into the BstEII site (pRP22F) despite the production of

single-stranded plasmid DNA. It can be concluded that

removal ofthe entire 355bpHindIII-HaeIIIfragment results in a low level of instability. This instability does not,

however, correlate with the production of single-stranded

plasmid DNA.

TABLE 2. Segregational stabilityof pBAA1 derivatives

Plasmid Production ofssDNA' Stability(%)

pDE22 100

pDE14 + 79

pRP22 100

pRP17 + 95

pRP22F + 100

pRP22BE + 88

pRP22EN 98

assDNA,Single-strandedDNA.Symbols:+,production of ssDNA;-, no

production.

1~~~~~~

""

r,

C' F

14.

l.

I0

41

-S

a

(6)

B L PLA1MID

pBAaa

1171 DISCUSSION

A newplasmid, pBAA1, has been identified in an

indus-trial Bacillus strain. Results presented in this study demon-strate that pBAA1 belongs to the class of small, naturally occurring plasmids of gram-positive bacteria. Although pBAA1 is different in size and restriction map from pLS11

(4) and pFTB14 (23), a high degree of homology exists

between these plasmids in regions involved in plasmid replication. Homology also exists between the plus-strand origin and Rep protein regions of pBAA1 and those regions ofpUB110and pC194, the greaterlevelof homology being found withpUB110. No homologywasdetected between the

replication regionof pBAA1 and other gram-positive

plas-mids pT181, pC221, and pE194.

The minimal replicon of pBAA1 was found to reside withina 1.4-kb fragment of DNA. The copy number ofthis

plasmid is similarto that of the parental plasmidat

approx-imately 5 per chromosome equivalent. In addition, this

plasmid is very stable, with over 80% of cells retaining

plasmid after 100 generations in the absence of selection. It

canbeconcluded that all functions required forcopynumber

determinationarecontained within the minimal replicon. In

addition, the lowcopy numberper chromosome equivalent

suggeststhat an active partitioning system may also reside onthisfragment.

Abody of evidence supports thehypothesis that pBAA1 replicates by a rolling circle-type mechanism. Key features

of thismodeof replication include (i)anorigin of plus-strand

synthesis containingasiteatwhich the DNA isnicked;(ii)a

replication protein which nicks the plus strand, forms a

covalent linkage with the DNA, and renicks and ligates the newly synthesized plus strand afteroneroundofreplication;

and (iii) an origin sequence for initiation of minus-strand

synthesis. The replication functions of pBAA1 have several features homologous to the replication functions of coli-phage 4X174 and S. aureus plasmid pC194, both of which

replicate by a rolling cycle-type mechanism (11, 18). The

similarities between(X174, pC194, and pBAA1 include all thefeatures essentialtothis mode of replication. For

exam-ple, thereareconservation ofsequenceattheplus origin and conservation of amino acidsat the active site of the replica-tion protein in all threecases. Inaddition, all three containa

genetic function which although notsimilaratthe sequence

levelis involved in efficientconversionof the single strandto thedouble-stranded form. In summary, these data strongly suggestthatpBAA1 replicates byarolling cycle-type mech-anism.

Itwas observed that deletion or mutation ofa sequence

overlapping the BstEII site results in the production of large quantities of single-stranded plasmid DNA. In view of the strong likelihood that pBAA1 replicates bya rolling

circle-typemechanism, it is probable thatthissequenceisinvolved in the efficient conversion ofthe single plus strand to the double-stranded plasmid form. Palindromic plasmid

se-quenceswith thisactivity have previously beenreportedfor gram-positive plasmids, e.g., pT181, pC221, pUB110, and pC194 (12), and forpLS1 (5). This region ofpBAA1has no

homology with the previously reportedminusorigins. Thus, thissequenceconstitutesanovelminusoriginfor thisgroup

ofplasmids.

The HindIII-EagI 350-bp fragment ofpBAA1 is almost identicaltothesequenceofthefragmentofpLS11, reported to contain a partition function(4). There are differences in thesequencesatonlyfourpositions:therearesix C residues between bases 306 and 311 in pBAA1 but only four C

residues in

pLS11;

there is a T residue at

position

316 in

pBAA1,

which isanAresidueatthesame

position

in

pLS11,

andtheCresidueatposition288 inpBAA1isnotpresentin

pLS11.

The

phenotypes

of cells

containing

similar mutant

plasmid

constructions suggest that the par function (of

pLS11)

and

minus-origin activity

(of pBAA1) may be the samefunction.

(i)

Minus

origins

are

predicted

tofunction in oneorientation only (12); Changet al. (4) report that their par sequence functions only in one orientation. (ii) The minus

origin

andparactivities both

overlap

the BstEII site but do not extend as far as the

EagI

site.

(iii)

Deletion of neitherparnorthe minus

origin

affects

plasmid

copy num-ber. A crucial difference betweenthese

activities, however,

is the effect of their removal on

segregational plasmid

stability.

Removal ofpar resultsin adramatic reduction in the

stability

of

pLS11 constructions,

whereas inactivationor deletion of the minus

origin

of

pBAA1

results in

only

alow level of

instability.

These differences may be caused

by

strain, medium,

or

plasmid

construction differences. Itcan be

concluded, however,

that increases in the

steady-state

levels of

single-stranded plasmid

DNApersedonotresultin

segregational instability

of

plasmids.

As shown in this

study,

pRP22F

produces single-stranded plasmid

DNA but is not

detectably

segregationally

unstable.

Similarly,

both

pC194

and

pE194 produce

single-stranded

plasmid

DNA in B.

subtilis

(31),

but whereas

pE194

isvery unstable in this

host,

pC194

isnot

detectably

unstable after 100

generations

in the absence of selection

(K.

M.

Devine,

unpublished

observa-tions).

A moredetailed

analysis

of minus

origins

and their

relationship

to

segregational

plasmid instability

is

required

to

resolvethese

points.

ACKNOWLEDGMENTS

This work was supported by European Economic Community

BiotechnologyActionProgramcontractBAP-0038-IRL.D.Higgins

wassupported bycontract BAP-0137-IRL.

We are grateful to Dusko Ehrlich and co-workers, especially Sandy Gruss, HeinteRiele, andMarie-FrancoiseGros,for

provi-sion ofdata prior topublication andfor manystimulating discus-sions. WearealsogratefultoBiocon Ltd. forprovisionofpBAA1.

LITERATURE CITED

1. Bron, S., P. Bosma, M. Van Belkum, and E. Luxen. 1987.

Stability function in the Bacillus subtilis

plasmid

pTA1060.

Plasmid18:8-15.

2. Bron, S., and E. Luxen. 1985. Segregational instability of

pUB110-derivedrecombinant

plasmids

inBacillus subtilis. Plas-mid 14:235-244.

3. Bron, S., and G. Venema. 1972. Ultraviolet inactivation and excisionrepairin Bacillus subtilis. I.Construction and charac-terization ofatransformable eightfold

auxotrophic

strain and twoultraviolet-sensitive derivatives. Mutat. Res. 15:1-10. 4. Chang, S., S.-Y. Chang, and 0. Gray. 1987. Structural and

genetic analysisofapar locus that

regulates

plasmid

partition

in Bacillus subtilis. J. Bacteriol. 169:3952-3962.

5. del Solar, G. H., A. Puyet, and M. Espinosa. 1987. Initiation

signalsfor theconversion of

single-stranded

todouble-stranded DNAforms inthe

streptococcal

plasmid pLS1.

Nucleic Acids Res. 15:5561-5580.

6. Ehrlich,S. D. 1977.

Replication

and

expression

of

plasmids

from

Staphlococcus aureus in Bacillus subtilis. Proc. Nati. Acad. Sci. USA 74:1680-1682.

7. Ehrlich, S. D. 1978. DNA

cloning

in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 75:1433-1436.

8. Fiddes,J.C.,B. G.Barrell,andG. N. Godson. 1978.Nucleotide sequencesoftheseparate

origins

of

synthesis

of

bacteriophage

G4 viral and

complementary

DNA strands. Proc. Natl. Acad. Sci.USA75:1081-1085.

9. Geider, K.,E.Beck,and H. Schaller.1978.AnRNAtranscribed

(7)

fromDNA atthe origin of phage fd single strand to replicative form conversion. Proc. Natl. Acad. Sci. USA 75:645-649. 10. Grandi, G., M. Mottes, and V. Sgaramella. 1981. Specific

pattern ofinstability of the E. coli his gene cloned in B.subhilis via the Saureus plasmid pCS194. Plasmid 6:99-111.

11. Gros, M. F.,H.teRiele, and S. D.Ehrlich. 1987. Rollingcircle replication ofsingle-stranded DNAplasmid pC194. EMBO J. 6:3863-3869.

12. Gruss, A. D., H. F. Ross, and R. P. Novick. 1987. Functional analysis ofapalindromic sequencerequiredfornormal replica-tion ofseveralstaphylococcal plasmids. Proc.Natl. Acad. Sci. USA84:2165-2169.

13. Gryczan,T., A. G.Shivakumar,and D.Dubnau. 1980. Charac-terization of chimeric plasmid cloning vehicles in Bacillus subtilis. J. Bacteriol. 141:246-253.

14. Gryczan, T. J., S. Contente, and fl.Dubnau. 1978. Characteri-zation ofStaphlococcus aureus plasrpids introduced by trans-formation into Bacillus subtilis. J.Bacteriol. 134:318-329. 15. Gryczan, T. J., and D. Dubnau. 1978. Constructionand

prop-erties of chimeric plasmids in Bacillus subtilis. Proc. Natd. Acad. Sci. USA 75:1428-1432.

16. Khan, S. A., G. K. Adler, and R. P. Novick. 1982. Functional origin of replication ofpT181plasmidDNAiscontained within a168-basepair segment. Proc. Natl. Acad. Sci. USA 79:4580-4584.

17. Kieny, M. P., R.Lathe, and J. P. Lecocq. 1983. Newversatile cloningand sequencing vectors basedon bacteriophage M13. Gene26:91-99.

18. Kornberg, A. 1980. DNAreplication, p. 473-570. W. H. Free-man &Co., San Francisco.

19. Kreft,J., J.Parrisius, K.Burger,and W.Goebel.1982. Expres-sion and instability of heterologous genesinBacillussubtilis, p. 145-157. In A. T. Ganesan, S. Chang, and J. A. Hoch (ed.), Molecular cloning and gene regulation in bacilli. Academic Press, Inc.,NewYork.

20. Lopez, P., M. Espinosa, B.Greenberg, andS. A. Lacks. 1984. Generation of deletions inpneumococcal malgenes clonedin Bacillus subtilis. Proc. Natl. Acad. Sci. USA81:5189-5193. 21. Maniatis, T., E. F. Fritsch, andJ.Sambrook. 1982. Molecular

cloning: alaboratory manual.ColdSpring Harbor Laboratory,

ColdSpring Harbor, N.Y.

22. Miller, J. H. 1972. Experiments in molecular genetics. Cold SpringHarborLaboratory, Cold SpringHarbor, N.Y. 23. Murai, M., H. Miyashita, H. Araki, T. Seki, and Y. Oshima.

1987.Molecular structure of thereplication origin ofaBacillus amyloliquefaciens plasmid pFTB14. Mol. Gen. Genet. 210: 92-100.

24. Norrander, J., T. Kempe, and J. Messing. 1983.Construction of improved M13 vectors using oligodeoxy-nucleotide directed mutagenesis. Gene26:101-106.

25. Novick, R. P., G. K. Adler, S. Majumder, S. A. Khan, S. Carlton, W. D. Rosenblum, and S. Iordanescu. 1982. Coding sequenceforthepT181 repC product: aplasmid-coded protein uniquely required for replication. Proc. Natl. Acad. Sci. USA 79:4108-4112.

26. Ortlepp, S. A., J. F. Ollington, and D. J. McConnell. 1983. MolecularcloninginBacillussubtilis ofaBacilluslicheniformis geneencodingathermostablealphaamylase. Gene 23:267-276. 27. Ostroff, G. R., and J. J. Pene. 1984. Molecular cloning with bifunctional plasmid vectors in Bacillus subtilis. 11. Transfer of sequencespropagated inE.coli toB.subtilis. Mol. Gen. Genet. 193:306-311.

28. Rabinowitz, P. M., M. Y.Haykinson, L. S. Arutyunova, Y. V. Yomantas,andA. I.Stepanov.1985. Thestructureandsourceof plasmid DNA determine the cloning properties ofvectors for Bacillus subtilis, p. 635-656. In D. R. HIelinski, S. N. Cohen, D. B. Clewell-, D. A. Jackson, and A. Hollaender (ed.), Plas-midsinbacteria. PlenumPublishing Corp., New York. 29. Sanger,F., S.Nicklen, and A. R.Coulson. 1977.DNA

sequenc-ing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA74:5463-5467.

30. Shlomnai, J., and A. Kornberg. 1980. An E. coli replication proteinthatrecognizesauniquesequencewithahairpin region ofOX174DNA.Proc. Natl. Acad. Sci. USA 77:799-803. 31. teRiele, H.,B.Michel,andS.D.Ehrlich. 1986. Single-stranded

plasmid DNA in Bacillus subtilis and Staphylococcusaureus. Proc. Natl. Acad. Sci. USA83:2541-2545.

32. Yanisch-Perron, C., J. Viera, andJ. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide

References

Related documents

Health care professionals? knowledge, attitudes and practices relating to umbilical cord blood banking and donation an integrative review RESEARCH ARTICLE Open Access Health

Towards this goal, we performed immuno- histochemical labeling for GATA4 protein in fetal tissues for which GATA4 has been shown to play a role in

Conservation of current does NOT depend on the properties of charge.. Conservation of current depends on the existence of J but not on

Using the Connector/J JDBC driver, all sorts of Java applications can access a database and its data.. In the next chapter, we provide a comprehensive overview of the JDBC

Table 3 2 Multinomial Logistic Regression Analysis of the effect age, sex, the distal and intermediate factors on moderate and severe stunting versus normal growth among

Associations between urinary phthalate metabolite concentrations (log transformed, μ g/g creatinine) and handgrip strength, stratified by the 90th quantile of the omega-6 to

State Service database Access Center Authorization &amp; Authentication Enumeration Access Indexing Engine Search Engine Index Server Gateway Client. Advanced Gateway