Halobacterium Plasmid pnrc100

(1)

JOURNALOFBACTERIOLOGY,Aug. 1993,p. 4584-4596 0021-9193/93/154584-13$02.00/0

CopyrightC) 1993,AmericanSocietyforMicrobiology

Vol.

175,

No. 15

Minimal

Replication

Origin

of the

200-Kilobase

Halobacterium

Plasmid

pNRC100

WAI-LAP NGANDSHILADITYA DAsSARMA*

Department ofMicrobiology, UniversityofMassachusetts,Amherst, Massachusetts01003 Received 11February1993/Accepted24 May1993

We have identified the replication origin ofpNRC100, a 200-kb plasmid ofHalobacteriumhalobium, by assaying for replication

ability

ofminiplasmids containingclonedfragmentsofpNRC100 and the mevinolin resistance selectable marker of Haloferax vokanii. First, we showed the replication ability of plasmid pNGHCMEVl,whichcontains the 19-kb HindlI-CfragmentofpNRC100,byrecoveryofplasmidDNAfrom mevinolin-resistant transformantsof H. halobium. The minimalreplication originofapproximately3.9 kbwas definedbysubcloning

successively

smallerregions ofpNGHCMEV1 andassaying forplasmid replication in either H. halobium orH. volcanii.Thesamereplication originwasalso recovered after transformation ofH.

volcanii with a library ofpartial Sau3AI fragments ofpNRC100. The nucleotide sequence of the minimal replicationoriginwasdetermined and foundtocontainalongopenreading frame,namedrepH, transcribed

awayfroma

highly

A+T-rich region.Thetranscriptionstartsitewasidentifiedby primerextension

analysis

tobe 17to18nucleotides5'toaputative repHstartcodon. Thepredicted productof therepHgene,anacidic

proteinwitha molecularweight of113,442, showed 24to 27%o

identity

with predictedgene products ofH. volcaniiplasmidpHV2 and H. halobium plasmid

p+HL,

suggesting thateachis involvedinplasmid replication. One pNRC100 minireplicon, pNG11A12, wasanalyzed by linkerscanning mutagenesis, which showed the requirement ofrepH for replication. Restoration of the repH reading frame of one replication-defective

pNG11A12 derivative by introduction of a second small insertion resulted in reversion to replication proficiency.Thereplication

ability

ofpNG11A12waslost when theentireA+T-richregion, about550bplong,

wasdeleted butnotwhen small insertionsordeletionswereintroduced into this region.Thepresenceofonly 52 bp of the A+T-richsegmentwassufficienttopermitreplication. ThepNG11A12minirepliconwaslostat

highfrequencyfromcellsgrownwithout mevinolinselection, suggestingthat theplasmidpartitioning locus of pNRC100 is absentintheminimalreplication origin region.We discuss thepossibleroles oftherepHgeneand the A+T-richregioninreplication of pNRC100.

The genome of the extreme

halophile

Halobacterium halobium isextremely unstable

(4, 8, 12, 36, 39, 40, 42).

It consists of two physically separable components, a G+C-richmajor fraction andanA+T-rich satellitefraction(14, 24, 34). Recent mapping analysis has indicated the presence of a 2,000-kb circular

chromosome,

two megaplasmids (pNRC200 and pNRC100, 350 and 200 kb in

size,

respective-ly), and several smallerandvariable minor circularDNAs, mostof whicharedeletion derivativesofpNRC100(3, 9, 30, 32, 35). Plasmid pNRC100 sequences partition largely into theA+T-rich satellite

fraction,

while thechromosome and

pNRC200

partition mostly but not exclusively into the G+C-rich major fraction. A large number of repeated ele-ments,

including

several well-characterized insertion se-quence

(IS) elements,

arefoundinthe H. halobium genome, manyofwhich areclustered inpNRC100 and the A+T-rich satellite fraction (4, 8, 12, 15, 21, 26, 42, 47). Recombina-tional activity promotedby these repeated elements is re-sponsiblefor thegenomicinstabilityof H. halobium.

Thephysicalmap ofpNRC100was recentlyestablishedby using rarely cuttingrestriction enzymes andpulsed-fieldgel electrophoresis (Fig. 1) (30, 32). Interestingly, avery large invertedrepeatsequence(ca.35kb)wasfound inpNRC100, a structure which is reminiscent of plant chloroplast ge-nomes.Two different inversion isomersofpNRC100, named ab and

P-y,

related by the relative orientation of the

inter-*Corresponding author. Electronic mail address:

sds@rna.

micro.umass.edu.

vening small andlarge single-copy regions were identified. The finding ofinversion isomers of pNRC100 but not ofa pNRC100 deletion derivative lacking one copy of the in-verted repeats suggested that recombination between the inverted repeats is the mechanism for inversion isomeriza-tion. In addition to the large inverted repeats, three IS elements, ISH2, ISH3, and ISH8, were found in multiple copiesinpNRC100,and one element, ISH50, was found ina single copy. Two ISelements,ISH2 andISH3,werelocated at the termini of the large inverted repeats. Atotal of 17 copies of IS elements havebeen mappedto pNRC100.

Rearrangements of pNRC100 wereobserved ingas vesi-cle-deficient (Vac-) mutants of H. halobium, which occur spontaneouslyat afrequencyofabout 1% (8,10, 11, 22, 36, 44). Bothinsertion (class II) and deletion (class III) muta-tionswerefoundmappinginthesmallsingle-copyregion of pNRC100 near the major gas vesicle protein gene, gvpA. This region was sequenced and revealed a gene cluster containing13 openreading frames (Fig. 1A) (10, 19, 22, 23). The genes are organized into two divergent transcriptional units,withgvpA, -C, and -N orientedrightwardand gvpD, -E,

-F,

-G, -H,

-I,

-J,

-K, -L, and -M oriented leftward. Divergentpromoters were mapped to the gvpA-Dintergenic region (10, 22). Although genetic evidence indicated the involvement of many genes in gas vesicle synthesis, only twoproteins, GvpAandGvpC, have thus far beendetected ingasvesicles (19). Additional biochemical analysis ofgas vesicleproteins has beenhamperedbythe extremestability ofthe structure tosolubilization.

Inorder tostudytherequirement for specificgypgenes in 4584

on April 4, 2021 by guest

http://jb.asm.org/

(2)

REPLICATION ORIGIN OF PLASMID pNRC100 4585

A

repH

ISH3

ISH8.

IRW I

P0o

"

PREP

SD108 SD106

SD104

|

RI

SD120

I

4f

A

C

N

H11111 11 1 1

I

a

MLKJI HGF E

D

I gvpgenes A

#'

FIG. 1. Physicalandgeneticmapof thea8isomer ofpNRC100. (A)Gas vesicle generegionshownenlarged,with the gvpgenes,repH, andtwoIS elements indicatedbyboxes. Thepositionsof IS element insertions inclassIIVac- mutantsandmutantstraindesignationsare indicated above. Promotersareindicatedby horizontalarrows.(B)Circularplasmid map, with the locations oflargeinverted repeats marked by heavyarrows ontheoutside,thelocations of IS elementsindicatedby oxesontheoutercircle,and thepositionsofDraI and HindIII

fragmentsindicatedonthe inner circles. Theapproximatedeletion endpointsin classIII Vac-mutants(labelled)areindicatedbyinnerarcs.

ThepositionoftheISH3 element indicated inHindIII-Chas been modified from thatpreviouslyreported(32)and is basedonhybridization

analysis (datanotshown)andsequencingdata.

-VOL. 175,1993

on April 4, 2021 by guest

http://jb.asm.org/

(3)

4586 NG AND DAsSARMA

gas vesicle synthesis in H. halobium, wereconstructed the entiregvp gene cluster onrecombinant plasmids containing the Haloferax volcandi mevinolin resistance selectable marker(mev) (19,20,27). Duringthe courseof thiswork,we found that recombinant plasmids containing the 19-kb Hin-dIII-CfragmentofpNRC100nearthegyp geneclusterwere able to replicate in H. halobium. This, together with the observation that thelefthalfof the HindIII-C

fragment

was never deleted in class III Vac- mutants,

suggested

that HindIII-C contains a

region

of

pNRC100

necessary for plasmid replication or maintenance. We further observed that transformation of class III H. halobium Vac- mutants

deleted for the entire gvp gene cluster with a

plasmid

containing thereconstructedgyp gene cluster and the capa-bilitytoreplicatein H. halobium resulted in

genetic

comple-mentation of the mutants andrestoration oftheir

ability

to float(19, 20).

Inthis report, wedefine the minimal

replication origin

in the HindIII-C fragment of

pNRC100.

We show

by

DNA sequencingandmutational analysisthat this

region

contains anA+T-rich region and agene, repH,necessaryfor

repli-cation.

MATERIALS AND METHODS

Materials. Restriction

endonucleases,

DNA

polymerase

I, Klenowfragment,T4 DNA

ligase,

T4 DNA

polymerase,

and T4 polynucleotide kinase were obtained from Gibco

BRL,

New England Biolabs,

Stratagene,

Toyobo,

and United States Biochemical. Calf intestinal alkaline

phosphatase,

DNaseI, andnucleotideswerefrom

Boehringer

Mannheim. RNaseA was purchased from

Sigma.

Avian

myeloblastosis

virus reverse

transcriptase

wasfrom International Biotech-nologies. The Sequenase version 2.0 DNA

sequencing

kit was from United States Biochemical.

32P-labelled

nucle-otides were

purchased

from Amersham. Mevinolin

(Lova-statin) was agiftfrom

Merck, Sharp,

andDohme Research Lab. Oligonucleotides were

synthesized by

the

synthesis

facilityatthe

University

of Massachusetts.

Bacterial strains.Wild-typeH. halobium NRC-1was cul-turedas

previously

described

(11).

H. volcandiWFD11was

kindly provided by W. F. Doolittle and cultured as previ-ously described

(5).

Escherichia coli DH5a was used for cloning andsubcloning

(37).

Cloningand subcloningof the minimal

origin

region. Plas-mids pNGHC1 and

pNGHCMEV1,

containing the Hin-dIII-CfragmentofpNRC100,werepreviously described(20, 32).To constructpNG11and-12,a9-kbEcoRIsubfragment of HindIII-C was excisedand subcloned inbothorientations into the unique EcoRI site ofpNGMEV100, a previously described pTZ19R derivative containing the 3.5-kb

Kpnl-SphI fragment

of H. volcanii with the mev gene

(27).

Deletion derivatives ofpNG11 and -12 were constructed by linearizing plasmidswithpartial HaeIII digestion, purifica-tion of15.4-kbsizefragments by agarose gel electrophoresis, cleavage atthemultiple cloningsite with SmaI, circulariza-tion with T4 DNAligase,andtransformationofE.coli DH5a (37).PlasmidspNG11A4,

-A10, -A12,

-A32, -A36, -A51,

-A55,

and

-A69

and

pNG12A6, -A14, -A15,

and

-A21

were obtained bythisprocedure. Anadditional set of deletion derivatives was constructed from pNG11A12 by linearization

with

EcoRI,

second digestion with

BglII,

ClaI,

KpnI, or

CvhI,

generation of blunt ends with

Klenow

fragment of DNA polymerase I orT4 DNA polymerase, and then recircular-ization with T4 DNA ligase and transformation of E. coli

DH5a.

Plasmids

pNG11A1201,

-A1205,

-A1210,

and-A1215

were obtainedbythisprocedure. During construction, one ofthe ClaI sites (proximaltothe SphI site)was protected fromdigestionby Dammethylation inE. coli DH5a because of thepresence of twocontiguous GATCsequences, GATC-GATC, containing the ATCGAT ClaI site. The plasmid constructs aretabulated inTable 1.

Construction of apNRC100 library andcloning ofa mini-replicon. Plasmid pNRC100 was partially digested with Sau3AI, and fragments of 5to10kbwerepurified byagarose gel electrophoresis. The partial Sau3AI fragments were inserted into theBamHIsite of pNGMEV101,aderivative of pNGMEV100 lacking one of the two BamHI sites, and transformed intoE. coli DH5a.Approximately 1,200 trans-formants containing recombinant plasmids were obtained. Thelibrarywasamplifiedin E. coliand used totransformH. volcanii(5, 37). PlasmidDNA wasisolated from mevinolin-resistant (Mevr) H. volcanii transformants and used to retransformH. volcanii andE. coli. Plasmids prepared from individual transformantswere then analyzed by restriction mapping and partialDNAsequencing.

Transformation of halobacteria and replication assay. DNA-mediated transformationof H. halobium andH. vol-canii was carried out by using the EDTA-polyethylene glycol transformation procedures ofClineand Doolittle (5, 7). Mevr transformantswereselected by platingonrichsolid mediumcontainingeither 25,uM(forH. halobium)or20,uM (for H. volcandi) mevinolin. Plasmids were prepared from transformants by a previously described alkaline sodium dodecyl sulfateprocedure

(31)

andvisualizedbyagarosegel electrophoresis. Transformation with plasmids ableto repli-cate in halobacteria gave -104 transformants per ,ug of plasmid, while transformation with those unabletoreplicate gave a 4- to30-fold lowerfrequency. Plasmidswereprepared from at least three separate Mevr colonies obtained from each transformation. When plasmids abletoreplicatewere used to generate Mev' transformants, plasmid DNA was recoveredfromnearly all transformants, but when plasmids lacking theabilitytoreplicatewereused,plasmid DNAwas neverrecovered.

DNAsequence

analysis.

DNAsequenceanalysiswas car-riedoutby the chain termination procedure ofSanger et al. (38), using synthetic oligodeoxyribonucleotide primers on double-stranded templates. External primers (5'-CCCAAC

GTCGTCGAG-3',

5'-AATGCGTCCGTCGGG-3',

and the universalprimer)were used forsequencingofthe deletion series

pNG11A12, -A1201, -A1205, -A1210, -A1215,

-A4, and -A32and

pNG12A14

and -A6.Tocovergapsin thesequence, 18 additional synthetic primers were used. Theprimer se-quences were

5'-GCATCGCTGTCATCG-3',

5'-CGGCAAG

CAGTTCCC-3',

5'-TGCATTGAGAATATGAT3',

5'-GGG

ATGCGCTGTTGTAT-3',

5'-CGAGCACITGCAGCTAT-3',

5'-TGACGATCCCGCACAGT-3',

5'-AGATTGACGTG

CAGGCA-3',

5'-AGTCTGGATACCGTCGC-3', 5'-GTCT

GGTGTTCTCGA-3',

5'-CTCCGTTCGCAGTCT-3', 5'-CC

GAAGCTGCTGTAC-3', 5'-CGAGGACGCCATTGA-3',

5'-TCAATGGCGTCCTCG-3', 5'-CAGCAACGAAGCTCC-3',

5'-AGACTGCGAACGGAG-3',

5'-GCGGCCATTCCTCTG-3', 5'-TACCTCTACCGCATA-5'-GCGGCCATTCCTCTG-3',

and 5'-CTCCATCCGTG TTGA-3'.

Primer extension analysis. Primer extension analysis was carriedoutby usinga 15-nucleotide-longsynthetic oligode-oxyribonucleotide

(5'-CCAGCAGTAGAGAGG-3')

hybrid-izingnearthe5' end of therepH gene. The oligodeoxyribo-nucleotide was first labelled at the 5' end by T4 polynucleotidekinase and

[_y-32P]ATP

and then extended by

using

avian myeloblastosis virus reverse transcriptase on J.BAcTERIOL.

on April 4, 2021 by guest

http://jb.asm.org/

(4)

REPLICATION ORIGIN OF PLASMID pNRC100 4587 TABLE 1. Plasmids used in this study

Plasmid Size

DsrpinSource

or

Plasmid (kb) Descnption reference

pTZ19R 2.9 E. coli cloning vector carrying amp gene Pharmacia

pNGHC1 22 pTZ19R containingthe 19-kb HindIII-C fragment of pNRC100 32

pNGMEV100 6.4 pTZ19R containingthe 3.5-kbKpnI-SphIfragmentcarrying the mev 20 gene ofH.

volcani

pNGMEV101 6.4 pNGMEV100derivative with one of the two BamHI sites deleted This work pNGHCMEV1 25.5 pNGMEV100containing the 19-kb HindIII-C fragment of pNRC100 20 pNG11 15.4 pNGMEV100containing the 9-kbEcoRIfragment ofpNGHC1 This work pNG12 15.4 Same as pNG11,except opposite orientation of the insert This work pNG11A4 10.9 Deletion derivative of pNG11 lacking 4.5-kbsegment; see Fig. 2 This work pNG11A1O 14.6 Deletion derivative of pNG11 lacking 0.8-kbsegment; see Fig. 2 This work pNG11A12 11.5 Deletion derivative of pNG11 lacking 3.9-kbsegment;see Fig. 2 This work pNG11A32 10.8 Deletion derivative of pNG11 lacking 4.6-kbsegment;see Fig. 2 This work pNG11A36 11.8 Deletion derivative of pNG11 lacking 3.6-kbsegment;see Fig. 2 This work pNG11A51 7.6 Deletion derivative of pNG11 lacking 7.8-kbsegment; see Fig. 2 This work pNG11A55 11.5 Deletion derivative of pNG11 lacking 3.9-kbsegment; see Fig. 2 This work pNG11A69 9.0 Deletion derivative of pNG11 lacking 6.4-kb segment;see Fig. 2 This work pNG12A6 12.7 Deletion derivative of pNG12 lacking 2.7-kbsegment; see Fig. 2 This work pNG12A14 11.8 Deletion derivative of pNG12 lacking 3.6-kbsegment; see Fig. 2 Thiswork pNG12A15 10.5 Deletionderivative of pNG12 lacking 4.9-kbsegment; see Fig.2 This work pNG12A21 7.8 Deletionderivative of pNG12 lacking 7.6-kb segment; see Fig.2 Thiswork pNG11A12O1 10.8 Deletion derivative of pNG11A12 lacking0.7-kbsegment;seeFig. 2 Thiswork pNG11A1205 10.0 Deletion derivative ofpNG11A12lacking 1.5-kbsegment; see Fig. 2 This work pNG11A121O 8.7 Deletionderivative of pNG11A12 lacking 2.8-kb segment; seeFig.2 This work

pNG11A&1215 7.7 Deletion derivativeof

pNG11A&12

lacking 3.8-kb segment;seeFig. 2 This work

pNG11A12i6 11.5 pNG11A12derivative with insertion of linker This work

pNG11A12i1O 11.5 pNG11A12derivative with insertion of linker This work

pNG11A12i28 11.5 pNG11A12derivative with insertion of linker This work

pNG11A12i41 11.5 pNG11A12 derivative with insertion of linker This work

pNG11A12iS7 11.5 pNG11A12derivative with insertionof linker This work

pNG11A12dil 11.2 pNG11A12derivative with linker insertion and small deletion This work pNG11A12di8 11.3 pNG11A12 derivative with linker insertion and small deletion This work pNG1112di21 11.3 pNG11A12derivative with linker insertion and small deletion This work pNG11A12di33 11.3 pNG11A12 derivative with linker insertion and small deletion This work pNGllAl2i28r 11.5 pNG11A12i28derivative with restoredreading frame This work pNG11A12i41r 11.5 pNG11A12i41derivative with restoredreadingframe This work pNG100 11.4 pNGMEV101containinga5-kbpartial Sau3AI fragment of Thiswork

pNRC100;see Fig.2

RNA which was purified from H. halobium NRC-1 and

NRC-1(pNG11A12)

and H. volcandi WFD1l and WEDli

(pNG11A12)

atlate

logarithmic growth

phase.

Productswere

analyzed on a 6% polyacrylamide-8.3 M urea

gel,

with a

Sanger

sequencing

ladder

generated

by

the same

phospho-rylatedprimer as asize standard

(37,

38).

Linker scanning mutagenesis and restoration of

reading

frame. Forlinker

scanning

mutagenesis,

plasmid

pNG11A12

was linearized by

partial

Sau3AI

digestion,

purified

by

agarose gel

electrophoresis,

and

dephosphorylated by

calf intestinal alkaline

phosphatase.

The

overhanging

endswere filled in with Klenow

polymerase,

and the DNA

population

was circularized with T4 DNA

ligase

in the presence of a SmaI linker

(5'-CCCCCGGGGG-3')

(37).

After transforma-tion ofE.

coli

DH5a,

derivatives of

pNG11A12

containing

14-bp insertions

(5'-TCCCCCCGGGGGGA-3')

atthecenter of Sau3AI siteswere isolated

(pNG11A12i6,

-i57,

-ilO,

-i28,

and-i41). Several

plasmid

derivatives hadalso suffered small deletions between Sau3AI sites in addition to the linker insertion

(pNGl15l2dil,

-di2l, -di33,

and

-di8).

To restore the frame of these frameshift

mutations,

plasmids

were linearized with aSmaI

isoschizomer,

PspAI,

which gener-ates 4-base 5'

overhangs,

and the ends were filled in with Klenow polymerase. These linear

fragments

were

purified

by

contour-clamped homogeneous

electric field agarose

gel

electrophoresis

(6)

and circularized with T4 DNA

ligase.

Intramolecular

ligation

resulted in insertion ofanadditional 4

bp

(18 bp

total:

5'-TCCCCCCGGCCGGGGGGA-3')

atthe

center of Sau3AI sites. The constructed

plasmids

are

de-scribed in Table1.

Measurement of

plasmid

loss in the absence of mevinolin selection.Threecolonies ofH.

volcanii

WFD11(pNG11A12)

wereused to inoculate culturemedia

containing

mevinolin. Whenthecultures reached late

logarithmic phase,

appropri-ate dilutions were

plated

on agar

plates

with or without

mevinolin.The cellswerealsodiluted in fresh

liquid

medium

lacking

mevinolinand grown for 15to90

generations

before further

plating

on agar

plates

with orwithoutmevinolin.

Nucleotide sequence accession number. The nucleotide sequence data

reported here,

has been

deposited

in the

EMBL,

GenBank,

and DDBJ databases

under,

the

acces-sion numberL19296.

RESULTS

Minimal

replication

origin

of

pNRC100.

Previously,

we found that

plasmid

DNAcould be recovered from H. halo-bium transformed with

pNGHCMEV1,

which contains the HindIII-C

fragment

of

pNRC100

(Fig.

1B)

and the H.

volcanii

mev gene

(20).

This resultindicated that HindIII-C containsasequence

capable

ofautonomous

replication

in H. halobium.In

addition,

wefoundthat the

part

ofHindIII-Cto VOL.175, 1993

on April 4, 2021 by guest

http://jb.asm.org/

(5)

4588 NG AND DASSARMA

Spbl SnaBI SnaBi EooR!EcoRI BooR!

V-

~

%'

ISH3 ISH8 gvpMLKJIHGFE3'-D

I .. Ec0oRI laal,KpnI . . ClaI 1-kb + pNG11/12 I LJ

SphI

K*1-

Cviii

_dCWC*W~

C

_N/s

l Ka BgUlBcoR! 1-kb ,w| l I II I I I -Ai iI IL I I iiIi~ II I A I

--i

S _ I

KpiI

Cia ChI Cla KpilBjII EcoRi

I .I . N , 05-kb LJ. I I I I I I I LI I II I I I I I + pNG11/12 _ pNG1221 _ pNG12A15 _ pNG12A14 _ pNG12A6 + pNG11A10 + pNG11A36 + pNG11AS5 + pNG11A12 + pNG11A12 + pNG11A1201 _ pNG11A1205 _ pNG11A1210 - pNG11A1215 _ pNG11A4 - pNG11A32 - pNG11A69 - pNG11A51 _ _ __ 1 _ + pNG100

FIG. 2. Restrictionandgeneticmapsof thepNRC100replication origin region in miniplasmidderivatives of pNRC100. The replication ability(Rep)andplasmiddesignationsareindicatedinthe right-hand columns. The plasmids includepNGHCMEV1, pNG11, and pNG12 (A)

andpNG11andpNG12deletionderivatives (pNG11AnandpNG12An), pNG11A12deletionderivatives (pNG11A12n),and pNG100 (B and C). PanelsB andCare shownenlargedcomparedwith panel A, with correspondingsections indicated by dashed lines. Theopenboxes

represent therepHgene,the lightly shaded boxesrepresentsgypgenes, and black boxesrepresentISelements (orpartsof ISelements).

Restrictionenzymecleavagesitesareindicated by verticallines, and scalesareshown for each panel.

the left of theISH8 element wasnot deleted in anyof the class III Vac- mutants characterized (e.g., strains SD109, SD116, zndSD112A) (Fig. 1B), indicating that this region is important forplasmid replicationormaintenance.Inorderto further define the replication origin, subclones of

pNGHC-MEV1 were constructed by using EcoRI. Of these

con-structs, pNG11 and -12 (Fig. 2Aand B), containing the 9-kb internal EcoRI fragment of HindIII-C cloned in opposite orientations in an E. coli plasmid carrying the mev gene, were reproducibly recovered from H. halobium NRC-1

A

_Hindu! _EcoRI a a Hiundll Rep +'%,

B

pNGHCMEV1 I I

C

I I 1

4

, -.. . i I I I I 5 I I I J. 0 . i I J. BACTERIOL.

on April 4, 2021 by guest

http://jb.asm.org/

Downloaded from

(6)

FIG. 3. ReplicationassayforpNRC100 miniplasmidsin H.

vol-candi. An ethidium bromide-stained agarosegel ofplasmid DNA

isolated fromMevrH.volcaniitransformedwithpNG11A4 (lane 1),

pNG11A12 (lane 2), pNG11A32 (lane 3), pNG11A36 (lane 4), pNG11A55 (lane 5),pNG11A1201 (lane 6), pNG11A1205 (lane 7),

pNG11A1210(lane 8), pNG11A1215 (lane 9),andpNG100 (lane 10)

is shown. Lane 11containsplasmidDNAisolatedfrom the

trans-formationhost H. volcaniWFD11,and lanes Mcontain1-kb DNA

ladder(2to 12kb)sizemarkers.

(Mevr)transformants inanextrachromosomal plasmid form, indicating that the 9-kb EcoRI fragment contains the entire regionnecessaryforautonomousreplication.

Plasmids pNG11 and pNG12wereusedto generate dele-tion derivatives(Fig. 2B and C) andto testfor replication ability. The plasmids were linearized by partial digestion

with the frequently cutting restriction enzymeHaeIII

fol-lowed by cleavage within the multiple cloning region with SmaI. Fragments of the appropriate sizeswere

recircular-ized and usedtotransform E. coli. Twelve deletion deriva-tives of pNG11 and -12 were isolated in this way. The deletion derivatives were used for transformation of H. halobium NRC-1 and H. volcanii WED11. Four of the deletionderivatives,pNG11A10, -A36,-A55,and-A12,were able to replicate, as judged by recovery of plasmid DNA from Mevr transformants(Fig. 3,lanes 1to5 [datanotshown forpNG11A10]).Each ofthese had deletionsextendingfrom the EcoRI site distal tothegvpgene cluster (Fig. 2B).The smallest deletion derivative with replication ability was

pNG11A12,which contained a5.1-kb region ofHindIII-C. Since theplasmidwith thesmallest deletionattheright end, extending 2.7 kb from the EcoRI siteproximal to thegyp

genecluster,wasunabletoreplicate (pNG11A6; Fig. 2B)we used anotherstrategytogeneratesmaller deletions. Plasmid pNG11A12wasdigestedwithin themultiple cloningsite with EcoRI and at the right end of the insert with CvnI,

KpnI,

ClaI, and BglII (Fig. 2C). Only one of the four resulting plasmids, pNG11A1201,with the smallest (0.7-kb) deletion attheright endwas abletoreplicate (Fig. 3, lanes 6to9). The extent of deletion in these plasmids indicated that a regionof about 4.4 kb of HindIII-C isrequiredforreplication. We tookasecondapproach toidentifying thereplication originofpNRC100. Alibrary ofpNRC100,constructed by using partialSau3AIfragmentscloned in anE. coliplasmid containingthe H. volcani mevgene,wasusedtotransform H. volcandiWFD11. The transformed cellsweregrown upin

liquid culture under mevinolinselection, and plasmid DNA wasprepared. Useofthis DNA for transformation of both E. coli andH. volcanii ledto recoveryof only a singleplasmid, pNG100 (Fig. 3, lane 10), containing a 5-kb restriction fragmentof pNRC100 from the same regionof HindIII-Cas in pNG11A1201 (see above). Plasmid DNA was isolated from a total of 12 E. coli transformants and shown to be identical by restriction mappingusing SstI. Further restric-tionmapping analysis of pNG100showed that the insertwas 0.5 kbsmallerthanthatin pNG11A1201at the left endand 1.2kblargerattheright end (Fig. 2C).This result confirmed the presence of a replication origin in HindIII-C and, to-gether with the deletion mappingresults, showed the mini-mal origin tobe 3.9 kb. Importantly, the isolation ofonly pNG100 fromalibraryof pNRC100 fragments suggestedthat the autonomous replicating regionin HindIII-C constitutes thereplication origin ofpNRC100.

Sequenceand transcription analysis.The minimal replica-tion origin identified in the small pNRC100 minireplicons wasfurther analyzed by DNA sequencing. Figure 4 shows the sequence of a4,377-bp region containing the 3,874-bp minimalorigin, defined by the left end of pNG100 and the rightendofpNG11A1201. The entireregion wassequenced on both strands. The sequence showed several interesting features, includingahighly A+T-richregion(58.5%A+T) of about 550bpatthe leftendandarightwardlyoriented open

reading

frame, repH, 3,027 bp long. TherepH gene product ispredictedtobe anacidic (pI 4.4)protein with a molecular weight of 113,442. The codonusageissimilar to that of other genes ofH. halobium, with G or Cusually in the wobble position. The base composition is about 55% G+C, whichis typical forpNRC100. An ISH3 elementwas found 230 bp downstream oftherepHstopcodon.

The start site for transcription was mapped by primer extension analysis of RNA isolated from H. halobium NRC-1 and H. volcanii

WFDli

containing or lacking the pNRC100 minireplicon pNG11A12 (Fig. 5). For both H. halobium strains, a major transcript with one nucleotide heterogeneityatthe5'end, beginning17 to18nucleotides 5'

to the

putative

repH start codon, was observed. Aminor

transcript starting

about 75nucleotides furtherupstream was alsovisible.For an H. volcanii strain containing pNG11A12,

a

heterogeneous

setofrepHtranscripts originating90 to145 bp upstream of the start codon was observed, while the strain

lacking pNG11A12

did not contain any detectable

transcripts.

The presence ofrepH

transcripts

iscorrelatedto the presence of therepHgene, eitheron pNG11A12 oron pNRC100. The differences in the major start site for

tran-scription

ofrepHin H.halobium

compared

withH. volcanji suggest that different promoters are utilized in these two

organisms.

All but52

bp

ofthe

550-bp

A+T-rich

region

located 5' to the promotercanbedeletedfrom the left endof

pNG11A12

without

losing replication ability,

as shown

by

the isolation of

pNG100

from the

pNRC100

library (see above).

Plasmids

pNG11A4

and

pNG11A32,

which contain 52- or

146-bp

deletions, respectively, beyond

the left end of

pNG100,

were not ableto

replicate (Fig. 3,

lanes 1 and

3).

Linker scanning mutagenesis and reversion

analysis.

To furtherdetermine the necessary

requirement

for

replication,

linker

scanning

mutagenesis

was carried out on

plasmid

pNG11A12.

The

plasmid

was linearized with

Sau3AI,

and the endswerefilled in with Klenow

polymerase.

The popu-lation of blunt-ended

fragments

was recircularized in the presence ofa

10-bp

SmaIlinkerandtransformed intoE. coli. The site ofinsertionwascharacterizedinmutated

plasmids

VOL. 175, 1993

on April 4, 2021 by guest

http://jb.asm.org/

(7)

first by double digestion with SmaI plus EcoRI and with SmaI plus HindIII. Nine plasmids with linker insertions in thereplicationorigin regionwereidentified, andeightwere further characterized byDNA sequencing (Fig. 4). Four of theplasmids(pNG11A12dil, -di2l, -di33,and-di8)had small deletionsbetween Sau3AIsitesin addition toinsertion of the linker, while five plasmids

(pNG11A12i6,

-i57, -ilO,-i28,and -i41) contained simple insertions. pNG11A12i28 and -i41 were independent isolates of the same insertion. Onlytwo plasmids, pNG11A12dil and -i6, with mutationslocated in theregionupstreamofrepH, showed the

ability

to

replicate

in H. volcaniiWFD11 (Fig.6, lanes2and3). Plasmidswith theothersevenmutations, which

mapped

withinrepH,were all unable toreplicate,showingthat therepHgeneproduct is required forreplication(Fig. 6).

In order to verify that the linker insertions inrepH are indeed responsible forreplication deficiency, the four plas-mids with simple insertions

(pNG11A12iS7, -ilO, -i28,

and -i41) and without the ability toreplicatewere cleaved with PspAI, an isoschizomer ofSmaI which generates 4-bp 5' overhangs. Fillingin of thePspAI sitesfollowedby recircu-larization resulted in an increase of theoriginal 14-bp inser-tion to 18 bp. Thus, the

original

frameshift mutation was converted to an in-frame insertion. After the sizes of the in-frame mutationswereconfirmedbyDNA

sequencing,

the plasmids were transformed into H. volcanii to assay for replication ability. Twoof theplasmids,

pNG11A12i28r

and

-i41r,

with identical in-frame

insertions, regained

the

ability

toreplicate (Fig. 6, cf. lanes 6 and 9 andlanes 8 and

10),

demonstratingtherequirementofrepHfor

replication.

The other twoplasmids,

pNG11Al2i57r

and

-ilOr,

did not

regain

replicationability,suggesting that mutationsatthose sitesin repHblockreplication.

Plasmid

stability

inthe absence ofselection. Todetermineif theminimal replication origin

region

contains the functions necessary for stable plasmid

maintenance,

cultures of H. volcanii containing

pNG11A12

were grown for 15 and 90 generationsin the absence of selection andassayedfor Mev phenotype.

pNG11A12

waslostrapidly:initially99 + 5%of the colonies were resistant to

mevinolin,

74 + 2% were resistant after 15 generations, and only 23 + 4% retained mevinolin resistance after 90 generations (range of three separate experiments). Since there is no substantial differ-ence in the growth rates for plasmid-free and plasmid-containing strains, the rapidrateof

pNG11A12

loss observed indicates that the plasmid does not contain functions for partitioning during celldivision.

DISCUSSION

Wehave isolated and characterized the minimal replica-tionoriginofpNRC100,whichconferstheabilitytoreplicate on recombinant plasmids in H. halobium and H. volcanii. The minimal origin is approximately 3.9 kb in size and is

located near the gvp gene cluster in the small single-copy region of pNRC100. The origin contains a long open reading frame, named repH, and a highly A+T-rich region 5' to repH.Theisolation of thisreplicationorigin fromalibraryof pNRC100 sequences by selectingitsabilitytoreplicate inH. volcanii aswellas the lack of deletions extending into this region in class III Vac- mutants together suggest that it functionsasthereplication origin of pNRC100. The minimal replication region,however, isnotsufficient for stable plas-midmaintenance in theabsenceofpositive selection.

Interestingly, Blaseio and Pfeifer independently con-structed aminiplasmid, pUBP2, by usingadeletion deriva-tive of pHH1, a 150-kb plasmid of H. halobium NRC817 similar to pNRC100 (2). The available information on this plasmid indicates thatitcontainsaregion witharestriction map very similar to those of our replicating pNRC100 miniplasmids (cf.

KpnI,

SphI, and EcoRI sites). However, an unambiguous conclusion ofidentity is precluded bythe lack of sequence information and mutagenic studies on pUBP2at present (see addendum in proof).

The predicted product of the pNRC100 repHgene is a large acidicproteinwith amolecularweightof113,442. The results of linkerscanning mutagenesis show that insertions which interfere withthe reading frame, but not one which does not, preventreplication of theminiplasmids. Therefore, the repH gene product must be necessary for plasmid replication. When we compared the sequence of the pre-dicted RepH protein against the nucleotide sequence data basesbyusingtheTFASTAprogram(13),twohypothetical proteins withsignificant (24to 27%) homologywere identi-fied (Fig. 7). Interestingly, bothof thesewere encoded by halobacterial plasmids,oneby pHV2,a6,354-bp crypticH. volcanii plasmidcommonlyused as acloningvector(5),and the secondby

p4HL,

a 12,041-bp plasmid derivativeof an H. halobium bacteriophage, 4fH (16).This finding suggests that these homologs are also involved in the replication of thecorrespondinghalobacterialplasmids.

The region immediately 5' to the repH gene probably contains the major promoter, as judged by the finding of transcripts starting 17 to 18 nucleotides from the putative ATG start codon. The lackoftranscripts starting substan-tially upstream indicates that another possible ATG start codon,192nucleotidesfurther5' to theATG start(Fig. 4),is notused fortranslation initiation. Interestingly,in H. volca-nii,therepH geneis transcribed frompromoterslocated75 to 125 bp upstream of the major promoter used in H. halobium. The use of different promoters in these two organisms is surprising, considering that the mevinolin re-sistance gene is expressed in both species (27), that halo-phage

4H

canreplicateinbothspecies(2,7, 27), and that a consensus promoter sequence for all halophiles has been proposed(28).Onepossibleexplanation for this result is that a factor required for transcription of repH from its native promoter in H. halobiumisabsent in H. volcanii.

FIG. 4. Nucleotidesequence of the pNRC100minireplicon

pNG11A1201

and the derived amino acid sequenceof RepH.The top strand of thenucleotidesequenceis shown above, and the amino acid sequence is shown below in the single-letter amino acid code. The transcription startsitesareindicatedbyheavy arrows, oligo(A) tracts and restriction sites are underlined, inverted repeats, including one terminus of an ISH3element,areshown by half arrows, and iteron-like repeats are doubleunderlined[CAA(T/A)]orindicated by arrows

[(G/T)ATl'T(A/

T)]. The 5' termini of inserts in pNG11A4,pNG11A12, pNG11A32,pNG11A1201, and pNG100 and the 3' terminus inpNG11A1201 are indicated. TheBglII, ClaI, CvnI,KpnI, SnaBI, and SphIrecognition sites are labelled. The boldface type indicates theSau3AIsites of

pNG11A12linkerinsertion (pNG11A12i6,pNG11A12i1O, pNG11A12i28,pNG11A12i41, andpNG11A12i57) and substitution

(pNGl15l2dil,

pNG11A12di8, pNG11A12di21, andpNG11A12di33)mutants and theputative repH start codon. The six-amino-acid insertions encoded in

pNGllAl2i28rand-i41rareindicatedbelowthe normal RepH sequence (dashed lines). The positions of all mutations except forpNG1112di8 wereverifiedbyDNAsequencing.

J. BAcrERIOL.

on April 4, 2021 by guest

http://jb.asm.org/

(8)

VOL. 175, 1993 REPLICATION ORIGIN OF PLASMID pNRC100 4591

r-pUG11A12,pMG11A1201 . . .

1 GGCCTCGGwTwTTCGGTTGTATTGTCCACCATCTTCACAGTGTCCGCAAGTTCCGTCTCCGTTGCCCTTACGACATTTATCTGAGCCCGGAATCTGAATG

0 . . . dil,i6

101 TTGCTCGTGTTGTTTCTCATACCAGTTCCGATCCCTAAGTGGACAGACATGG

TG&ATAmTTOlR:~TCTACACTCG AACTAGAGTTG&GTAGILATATTTGACGAATGGGAAACACATC& TCGAAATTTGGTGGTTCTACACTGTGC

pUG100 . pN{G11,4 . .

w_TTACCTTTAA6TTGCATCCG _ _ _ _TAATCACTGGTCGGCGTTGGA-GCCTGAAG

dil

.

rgp-pNG11A32. .

AAGGCCCCGGCACGACCGTGTCCCCACAGGTTGCTACCC CTTCTTCCGCTGTGGGCTTAAGCGTTACCCCACGTAGCCGCTGCGCTGCCA

AATGAAAT6SCC CAGACACACCAAGACCGAGGTGACCGCGAGTAGCTGACGCTACTCCAGACCAATGACTC ACAGGTACAG&CTCACG&CAATCCGGGTCCGATGCTGCC&GGCTTCGCCCCGGATGTCGTTGTGAAGCATCGTCGCAGCGGACCAGCAGAGCCGGTCC

TCAACACGG&TGOGGCGGTATTATCAG TCAAATTGG CCCTCGCATTTGCCGGCTTCCACG

SphI. . . . .

TCCACGG&CTGCA!GCAC&CG^CCTA&C _CAGG~ATCCGGA GGGACCCTCTCTACTGCTGGAATTACC

M H T P N Q Q Q G I R X I V P G G T L 8 T A G I T ATTACCGAGGCAC T GGAACTACTGCGGTATTCAATCGGTTCGAAAATTCATCCGGCAAG I T E V T P R V T E W I P D L L E E L L P R S I Q S V R K F I R Q E . . . . CvnI . AAGACCCAGAAGTCCTCQCGCACGCACGATACAACACCGTCTATCGCCGTCTTCAAGAGGAAACCCTGAGGTTCGACCATCAAGAATGGTGCTCCACAAC D P E V L T H A R Y N T V Y R R L Q E TT L R F D H Q E W C S T T * . . SnaBI a . . . . di21 0 GGATATTTGGAGTGATGCAGAGGCTGAAGCGGT13A&TACGTAGA&TCACTTGTCGAGTTCGCAGTCAAATATTCTGACGTTGACGAAGTGATCTCGAC D I W 8 D A Z A E A V E Y V 9 8 L V 2 Y A V X Y 8 D V D E D D L D

GAACT UGGAGTACCAGCAG -CGTGCAAATCGCTTAAAC&GACTCTCACTACGATCAGTACCGGACGTGGCCCACTCAACGCTGGCCTTGAAGCCC

E L 8 E Y H Q Q a C K 8 L X Q T L T T I 8 T G R 6 P L N A G L E A L * . . . .di21,157 . . . . TCGCCAAGGGACCCGTACGACTCCACGATGAGCTCGATGACGCACCGCAACCGATCACGCTTGTCCTTGATGGCGAGTTGTGGTCAAAACTCGACGATAG A K G P V R L H D 2 L D D A P Q P I T L V L D G E L W 8 X L D D R *KGPVRLHDE* DD* PQP* TLVLD* EL*SKLDDR AGGAACAGGTATCCGAGCACTTGCAGCTATCGCCGTGCTCGGCTCCACCTTCGACGTCCGCCTGGTTATTTCACCAGCGTTAGACGCCGCGATTGAGCGA G T G I R A L A A I A V L G S T P D V R L V I S P A L D A A I E R CGGTATCCAGACTGGTATGACTCCC TCTCCGTCTTACTGAAACCCGTGAGACCTC-TCTGTAGAATCAGCGGGTGGCGACGGACAACCATCGGCGGAGC R Y P D W Y D 8 H L R L T Z T R 2 T 8 8 V E S A G G D G Q P 8 A E Q

AGC1GAGGAhGCGGA CA&TCAGACCTCC AG&GGAATCAGGGAGACTCCGTCTTCTCCGGAACCTCCCTATAGAGGGCTCTCGAGACTATCG

L I E A W N A I Q N L P E E 8 G R L R L L R N L P I E G S R D Y R

i10 . . .

T C_AGCAGGATGATT TGACT TTTACATCCTCGATCTAGAAaTA GCTTGTGGATATCGACCGTCGT

D L K Q D D Z I D V Q A G T V G R Y I L D L E E L G L V D I D R R DL*DDE D V .**ED E GGACAATACAACAGCGCATCCCTCACCGGCTTAGGACAAGTAGCAGTTGAGCAGTATGTCACCACGGACTACCGGGTGATCCATCCGACCCAATCGACGC G Q Y N S A S L T G L G Q V A V E Q Y V T T D Y R V I H P T Q 8 T L TGGAGACGCATCTTACGCCGACCCCTCAGCCCCAAGCAAGTACAGTGTATCCCGCGCGATCGGACACGAGGGAGGGGGATCAGCCTGGGACAGCGGAGGA T THLTPT P Q P Q A S T V Y P A R S D T R G D Q P G T A E D TTGGATAGCTGCGACAGGCAGTCCTAGTGAGGGTGCTGACTACGTTCAATGGCTCGATGGGCCGTCTGGTGTTCTCGACGCTTGGGGAATGCATCAGCGG W I A A T G S P 8 E G A D Y V Q W L D G P S G V L D A W G 1 H Q R 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 201 301 401 501 601 701 801 901 1001 1101 1201 1301 1401 1501 1601 1701 1801 1901 2001 2101 2201

on April 4, 2021 by guest

http://jb.asm.org/

Downloaded from

(9)

4592 NG AND DAsSARMA J.BACrERIOL. KpnI . di33 . 2301 TACCTTGCTGGCCGTCGAGATCGTGGTGTCACCCTAGTCVGACCGTATCGAGCGTTTCGAGGACGGTCGCGTATCGTATCTCAGTTGCTTCGACGATG 2400 Y L A G R R D R G V T L V D D R I E R F E D G R V S Y L S C P D D D * . . . di33 2401 ATCTCTTCGTAGCCACTCAGTGGGGCGGCCCCCTTCCTAC=CTGGGACGTATTGCGGGTGCCTTACTCAGCGCAAAGCACTGAGTAAGATCCTTACCCC 2500 L F V A T Q W G G P L P T L G R I A G A L L 8 D X A L 8 K I L T P

2501 TTCTCGTCTCGGAACCAGTTCGAAGAAAT TC A GAT GCGG CGTAGAGGACACCAAATCGGC 2600

S R L G N Q P E E I N D A V V Z Q L D R E A G E I I R R G H Q I G 2601 TGGTTCAGCGAGGACGAAGAAGACTACGATGGTTGGCGTGAGCGTATTGGCTCCGTTCGCAGTCTGTGTTTGCAACAAGTTGGTGAACTCACAAACAGTG 2700 W P S E D E D Y D G W R E R I G S V R S L C L Q Q V GCE L T N S D 2701 ACGACGTCGAAGCCCGGACTGAGTTACTACGTGACCTGCACGGCTTGGTTGCCTCAGCGACACAGCTGTACTACGCAGCCGGTGTAGACGTCACAATCAA 2800 D V E A R T E L L R D L H G L V A S A T Q L Y Y A A G V D V T I N 2801 CGTTCGAGTCCCACACGGGA CTCATCAGCGATGAACGCCGTCTCGACGATTTCCTCGGTTTCGCCCGGTACACCATTCCGAAACAGTCCGTGTAC 2900 V R V P D T G M L I S D E R R L D D P L G P A R S T I P K Q 8 V Y

2901 GGGATTCATTCGGGGTATAGGATGCTCC %AftfTTTCCTACGAGGTGGATGATGCTGATTCCACAATGCATC 3000

G I S8 G Y R M L L E D R P E X L X R R L P Y E V D D A D S T M H L

* . . C1aI...

3001 TGACCGCATCG TTTC TTCCCGATCAACGATGATCGATC ACCACGAAATTCGT 3100

T A S W V P S G 8 T M I D L H D D I E D A I E M E T N E I R E A I 3101 TGCTAATGGACAGGAGTCAGCACCGGTAATGGAAATCCCCGTCCAGATTGGGAACTCCTACTCAGCAATTCGTAATCACGTCGAAGACTACGCTTCAGCG 3200 A N G Q E S A P V M R I P V Q I G N S Y S A I R N H V E D Y A S A * . . . i28,±41 . . 3201 AAGAACTACCAGGTAGCTCACCAGGAGGATATTCACGAAGGAAAACAGGATCTCGAGCGACTCGTCCGCCTGTTCCTCCGTGTCCTCGGGACAGAGGATC 3300 X N Y Q V A H Q E D I H E G X QD

A.

E R L V R L F L R V L G T E D R iP a R a 3301 GACCACATCGAGCGTGTCCTCACGACGTTGCCGAGGCAATGCTTCATGTTGCTCAATCCTCCCGGAACTACGATTTCATCACCGTGCGAGACATCTCGTA 3400 P H R A C P H D V A E A M L H V A Q 8 S R N Y D P I T V R D I S Y 3401 CGGACTGTCGAATCTCCCCACGAAACGACTCTTACCCGAGCTCCCACCAACAGCAACGAAGCTCCTCAAAACCCTGCTTGATGCGGATGACCCGATGGGA 3500 G L S N L P T K R L L P E L P P TA T X L L X T L L D A D D P M G * ClaI . . . . 3501 CGGTCTGAAATCATCGATACC3C60A0TTTCGAAAGTAGCTATGATCGCTACATCAACGAAC3600 R S E I I D T A D I S E S S Y D R Y I N E L A A W D I I E P R E I E 3601 AGGGACACCGTCGGTGGGAAGCTCACTTAGAGCCGTGGTGGACACCGCAGAGTGATCGAGACGAACCCTATGCCGACCCTGACCCCGACACGGGAATACT 3700 G H R R W E A H L 2 P W W T P Q S D R D E P Y A D P D P D T G I L 3701 GTACGCTGAATTTCCCCGTGATGTCGCTAGTGCGGTGATGTGCCACCTCATCACCCACTACGACTTACCCGACCTTGAGACAGCGTATCTTGAGGGTATC 3800 Y A E P P R D V A S A V M C H L I T H Y D L P D L E T A Y L E G I * . . di8 . . . . 3801 CAACCGGGGGACGACATCAAGGCCCTCTTCGATGATCACGACCGACTCAGACGATGGCGGCCATTCCTCTGGGGTGCATTCGCCGACTCGGACAAACTTG 3900 Q P G D D I K A L P DDHD R L R R W R P P L W G A P A D S D X L E 3901 AAAGAGGCCCATCTGGTACAGCAGCTTCGGACTCGACCGTAGTTCGTCTCGGTCAATCTCCAGGACCCGACACAGCACAATCGAGCTTCCAAGACGTCTC 4000 R G P S G T A A S D S T V V R L G Q S P G P D T A Q S S P Q D V S * .di8 . . . . ClaI . 4001 AGAAACAGCAACCCAACGAGATCGTCTCAGTCAACCGTCGCCAGGGCTGGACTAATGATTGACGGCAATCGATAACTACGCAATGAATCCTCCACAACCA 4100 E T A T Q R D R L S Q P S P G L D * . . . KpnI . 4101 GCATCTGAACCGTACTCGAAAGGTGATGCAGTAACCGTATACGTGGGCGAAGACGACCCCGATGTCCGGTACCACGGCGTTAAATGCGTCGTCACTGACC 4200 4201 GCCTGCAAGATGATTTGAACACCGAGACAGGCCGTGACCGTGACCAGTACCTCTACCGCATAAAAAGACGGTCAACGGGCAAGGTTCAGTACCTCACAAA 4300 ISH3 . . . pNG11A1201--1 4301 GCATTCTCGGCTAGCTGTTTCTGAAGCCTGAGTTCCACGGCGGAGCTGTTGTGCTGGTGGTCTTGACGAGAAGATCT 4377 BglII FIG. 4-Continued.

on April 4, 2021 by guest

http://jb.asm.org/

Downloaded from

(10)

G

1

2

3

4 GATC

_{M 1 2 3}

₄

₅

₆

₇

_{8 910M}

5, C

TT

C C A C G C A G T C G C C 3,

FIG. 5.

Mapping

ofrepH

transcription

start

points by primer

extension

analysis.

A

synthetic

oligodeoxyribonucleotide

wasused to

prime synthesis

ofcDNAonH.

halobiun

NRC-i

(lane 1),

H. halobium

NRC-1(pNG11A12) (lane 2),

H. volcanu WFD11

(lane

3),

and H. volcanji

W*FD11(pNG11A12)

(lane

4)

RNAtemplates, and

products

werefractionated on a

sequencing gel.

Lanes labelledG, A, T, and C are

sequencing

ladders

generated by

using

the same

primer

and pNG11A12 as the

template.

The G lanewas loaded twice. The

complementary (coding)

sequence deduced from the

sequencing

ladders is shown beside the

autoradiogram.

The

tran-scription

start sites corresponding to the

major

cDNA

products

obtained with H. halobium are indicated beside the nucleotide sequencebyarrows.

An A+T-rich

region

was located about 470

bp

5' to the

repH

start codon

(Fig.

4).

This

region,

sequenced

in

pNG11A12,

isatleast550

bp long

andcontains58.5%A+T.

Surprisingly, nearly

all of the A+T-rich

region

(all

but52

bp)

wasdeletedin

pNG100,

whichwasselected for its

capability

to

replicate

in H. volcanii.

However,

when theentire A+T-rich

region

was deleted in

pNG11A4

and

pNG11A32,

the

plasmids

couldno

longer replicate, indicating

the

importance

of this

region

for

replication.

This

region

contains many

oligo(A)

and

oligo(T)

tracts and six

copies

of the sequence

(GIT)A1TIT(A/T)

withina130nucleotide stretch.

Interest-ingly,

the

region

maycontaina

sharp

bend

resulting

from the

phasing

of several of the

oligo(A)

tracts

(18).

Within the

470-bp

region

between the A+T-rich

region

and the

repH

structural gene isa

segment

of

relatively high

G+Ccontent

(42% A+T).

Several repeats were found in this

region,

including

an

8-bp

inverted repeat

(GGACACGA),

a

7-bp

inverted repeat

(GAGTAGC),

and seven

copies

of a very short repeat

(CAA[TIA]).

Several similar short

repeats

were also found in the

putative

replication origin

ofthe 1.8-kb pHSB1 family of

miniplasmids

of Halobacte-ium

species

(17).

One

possible

function of

repeats

couldbe in

binding

of

replication

proteins

such as

RepH,

with the formation ofa melted

replication

complex being

facilitated

by

theA+T-rich

region (25, 41).

FIG. 6. Replication assay forpNRC100miniplasmid mutants in H. volcani. Anethidiumbromide-stained agarose gel of plasmid DNAisolated fromMevrH. volcanii transformed withpNG11A12 (lane 1),pNG1l/l2dil(lane 2),pNG11A12i6(lane 3),pNG1112di8 (lane 4), pNG11A12di21 (lane 5), pNG11A12i28 (lane 6), pNG11 A12di33(lane7),pNG11A12i41(lane 8),pNGl1A12i28r(lane 9), and pNG11A12i41r (lane 10) is shown. Lanes M contain 1-kb DNA ladder(2 to 12kb) size markers.

Replication originsof several plasmids, phages, and bac-terial chromosomes havebeen studiedin detail (25, 43, 45, 46). In most cases, such as those of antibiotic resistance plasmids Rl andpSC101, conjugative plasmid F, bacterio-phagesXandP1,and theBacillus subtilisandPseudomonas putida chromosomes, genes required for replication (rep) have been found proximal to replication origins containing A+T-richregions and multiple copiesofreplication protein-binding sites(iterons).For manyof thesereplicons,rep gene transcription is oriented away from the replication origin, althoughin atleastone case

(Ri),

transcription isoriented toward the origin. ForpNRC100, the positionof the A+T-rich segment and repeated sequences 5' to the repH gene suggest that initiation ofDNAreplication occurs in this 5' region, with transcription ofrepHaway from the putative

origin.

Miniplasmid

pNG11A12

was frequently lost from cells grownwithout mevinolin selection. Both therateof lossand the low copy number of

pNG11A12

were similar to those observedforpSC101 mutantsdefective in

partitioning

(29).

The instability was also apparent in H. halobium strains harboring pNRC100

miniplasmids containing

the gyp gene cluster, with formation of sectored colonies on agar

plates

lacking mevinolin

(1).

In contrast, we have observed that although pNRC100

frequently

suffers rearrangements, it is very rarely

spontaneously

lost from H. halobium. These observations suggest that the minimal

replication

origin

region in

pNG11A12

is lacking the

partitioning

locus

nor-mallypresent onpNRC100. The absenceof the

partitioning

locus in

pNG11A12

is also consistent with ourobservation that pNRC100 or its deletion derivatives are

stably

main-tained(compatible)instrains transformedwith the

miniplas-mid (la,

33).

Thepresence of bothpNRC100anda

pNRC100

miniplas-mid in H. halobium transformants

produced

the

potential

for recombinationbetween the

homologous

regions

of the two VOL. 175,1993

on April 4, 2021 by guest

http://jb.asm.org/

(11)

A `0L~ Y D S MSH

E KR T GRI

0GD

M'IW

LVDDR I F VSYLSCF ...

T. DQV1AED-R AEL--DO.IEDG TY.VaGSr

FRE VEA FDG.. DL. LLNGN RWAL1 ...LG

,FSE D E E D Y D~~~~A DE RS

L-'-A

H1-bD Y E C E T KQ

-DD

VE L

Z..N

D DGWAERIGSV 1SLCLQQVGE L-TN....D V~RT LD L LVASATQ H PE R Y EDNA AKN DIyV101

I'SI

?H DRGR~ LSMR MTQS%

R'

EY

LYYVAAOVL T INV JTGM . LISDJRR L DD ARYT IjK Q$VS in

VVR

RLDF..

Q

.SF

D V Y

AN.

:~QP. T.3K EV IGN G GTVQG LT -.:GVSK ~~~~~~L RL P Y E.M.fViL-DA -F SM ADHVAGV I ED LRPYLDHLD. I EDAI EMETN RLSN P GEI AE I .ALK> ..HGP . G.DLQ EGFQPFTL DF .. ... . . LDRRR -P

EI.3|AJANGQ ES AP VMEI P. QIGNSYSAIR

NHVI.D..YAS-AK

V C R RD.RL G .ART

IK .A LQ IS I LEA.FC. D A.. S

NY QYAHQOEDI H LRVL.

SQ.ATI SRSS -VRLS 1SKRLEU tVR IST. ATNL :K

E G P I ,..H I Q DT PMf

QS'SRNYDFIT,

RDI S EAT 4D MR

TTQ'VNRM I:LPT GLVD

E1:.

N LJS

F-T.

T-A`

5''D,

S.-.-YD RYI PR EIEGHiR EPWW

pHV2

p+HL

RepH pHV2

p+HL

RepH

pHV2 p HL

;epH

E.RHKRILPSS

MLQES

F. TRI.:.S

LR.R..R.

N .. T VVDD.I.. S TLWD...L.

.WEAAMY,

It

D2DEPYAD..D PD¶GI LYAEF PRDWgAWAC HLITH P..

. . . .

DPSRFGDPDD

LETAYLEGIQ

* . . . * . . * * . . . ... . . .. .. .. . . .. . . .

ft

.I:F~E.A SW LM... ... D

RWLsIAAALA

tGDDIKALFD -I RtRRWRM FLWGAFADSD KLERGPSGTA ASd>TVVRtG

* . . . * * . * * * .:. . . * . . * . . . * *. * . * .

NP.RRNPDTI

TVGKIP.QSS

LQEPL:AR-'TA

NAD.W

Q...GPDTAQS

SFQDVSQTAT

QRDR.>SQ;SP GL.E

FIG. 7. Alignment of the RepHsequencewithsequencesfor predicted proteins encoded byH.volcanii plasmidpHV2(5)and H.halobium plasmid p4HL (16). Sequencesareindicated in the single-letter amino acid code. Identical aminoacidsareindicated by shading. Thepercent

identitiesof amino acid residuesare24.2% between RepHand the pHV2-encoded protein, 27.1% between RepH and the

p4HL-encoded

protein, and 30.2% between the pHV2- and

pOHL-encoded

proteins.

plasmids. Southern hybridization analysis of genomicDNA from Mev' H. halobium transformants confirmed that

re-combination was occurring (data not shown), particularly when miniplasmids unableto replicatewereused for trans-formation. Usually, these plasmids were integrated into pNRC100.Inaddition,recombination between themevgene

onthetransforming miniplasmid and themevinolin-sensitive alleleonthechromosomewaspossible in both H. halobium andH. volcandi. Suchrecombinants probably accounted for theMev'H. volcandi transformants obtained after transfor-mation withminiplasmidsunabletoreplicate autonomously (27). In contrast, integrated forms ofminiplasmids able to pHV2

p+HL

RepH pHV2 pHL pH pHV2 p+HL RepH pHV2

p+HL

RepH

pHV2

p+HL

RepH

pHV2

p+HL

RepH pHV2

p+HL

RepH pHV2

p+HL

RepH pHV2

p+HL

RepH J. BACTERIOL.

on April 4, 2021 by guest

http://jb.asm.org/

Downloaded from

(12)

replicateautonomously were negligible in comparison to the extrachromosomal form.

Our studies on pNRC100 documented in the present report and in earlier reports have provided some detail on the structure, rearrangement, and replication of a large halobacterial replicon. Furthermore, miniplasmid deriva-tives of pNRC100 have also served as useful vectors for moleculargenetic studiesof gas vesicle synthesis. However, additional questions regarding pNRC100 remain, for exam-ple, questions regarding its mechanism for copy number control, thegenes andsites involvedin partitioning and their possible involvement in sectored class I mutants, and the coding function of much of the plasmid, including all of the large inverted repeats and large single-copy region.

ACKNOWLEDGMENTS

This workwassupported bygrant DMB-8703486from the Na-tional Science Foundation and by Public Health Service grant GM41980 fromthe National Institutes of Health.

ADDENDUM INPROOF

After the acceptance of this paper for publication, the sequence of theorigin of replication of pHH1was reported

(F.

Pfeifer and P. Ghahraman, Mol. Gen. Genet. 238:193-200,

1993).

Assuggested by the similar restrictionmap, the sequence is verysimilar to the pNRC100 minimal originof

replication.

We compared thetwo sequences and found 20

single

nucleotideinsertions, deletions, and substitutionsout of4,377 bp. For pHH1, two smaller open reading frames werereportedinplaceof asingle largeopenframe(repH) for pNRC100. No sequenceswithsimilarityto thepHH1 open reading frame sequences were found in the EMBL data library.

REFERENCES

1. Arora, P.,F.Lin,andS. DasSarma.Unpublisheddata. la.Austin, S. J. 1988. Plasmidpartition.Plasmid 20:1-9.

2. Blaseio, U.,and F.Pfeifer.1990.Transformation of Halobacte-rinumhalobium:developmentofvectorsandinvestigationof gas vesiclesynthesis.Proc. Natl. Acad.Sci. USA87:6772-6776. 3. Bobovnikova, Y., S.DasSarma,K.Grey,and N. Hackett. 1991.

Restrictionmapping theHalobactenum halobium genome, p. 187.Abstr. 91st Gen. Meet.Am. Soc. Microbiol. 1991. Ameri-canSocietyforMicrobiology,Washington,D.C.

4. Charlebois, R. L., and W. F. Doolittle. 1989. Transposable elements and genomestructureinhalobacteria, p.297-307.In D. E. Berg and M. M. Howe (ed.), MobileDNA. American SocietyforMicrobiology, Washington,D.C.

5. Charlebois,R.L.,W. L.Lam,S.W.Cline,and W. F. Doolittle. 1987. Characterization of pHV2fromHalobacterium volcanii and itsuseindemonstratingtransformation ofan archaebacte-rium.Proc. Natl. Acad. Sci. USA84:8530-8534.

6. Chu,G.,D.Vollrath,and R. W. Davis.1986.Separationoflarge

DNA molecules by contour-clamped homogeneous electric fields.Science234:1582-1585.

7. Cline,S.W.,andW. F.Doolittle. 1987.Efficienttransfection of the archaebacterium Halobacterium halobium. J. Bacteriol. 169:1341-1344.

8. DasSarma,S. 1989.Mechanismsofgeneticvariabilityin Halo-bacterium halobium: the purple membrane and gas vesicle mutations. Can. J.Microbiol. 35:65-72.

9. DasSarma, S.,P.Arora, J. Perkel,W.-L.Ng,and N. Hackett. 1992. The second copy of the gas vesicle gene cluster of Halobacterium halobium NRC-1 is locatedon alargeplasmid,

p. 204. Abstr. 92nd Gen. Meet. Am. Soc. Microbiol. 1992. AmericanSocietyforMicrobiology, Washington,D.C.

10. DasSarma, S., T. Damerval, J. G. Jones, and N. Tandeau de Marsac. 1987. A plasmid-encoded gas vesicle protein gene in a halophilic archaebacterium. Mol. Microbiol. 1:365-370. 11. DasSarma, S., J. T. Halladay, J. G. Jones, J. W. Donovan, P. J.

Giannasca, and N. Tandeau de Marsac. 1988. High-frequency mutations in a plasmid-encoded gas vesicle gene in Halobacte-riumhalobium. Proc. Natl. Acad. Sci. USA 85:6861-6865. 12. DasSarma, S., U. L. RajBhandary, and H. G. Khorana. 1983.

High-frequency spontaneous mutation in the bacterio-opsin gene inHalobacterium halobium is mediated by transposable elements. Proc. Natl. Acad. Sci. USA 80:2201-2205.

13. Devereux, J., P. Haeberli, and0. Smithies. 1984. A comprehen-sivesetof sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.

14. Ebert,K., and W. Goebel. 1985. Conserved and variable regions inthe chromosomal and extrachromosomal DNA of halobacte-ria.Mol. Gen. Genet. 200:96-102.

15. Ebert, K., C. Hanke, H. Delius, W. Goebel, and F. Pfeifer. 1987. A new insertion element, ISH26, from Halobacterium halo-bium. Mol. Gen. Genet. 206:81-87.

16. Gropp, F.,B. Grampp, P. Stolt, P. Palm, and W. Zillig. 1992. The immunity-conferring plasmid p+HLfrom the Halobacte-riumsalinarium phage4H: nucleotide sequence and transcrip-tion.Virology 190:45-54.

17. Hackett,N.R., and S. DasSarma. 1989. Characterization of the smallendogenous plasmid ofHalobacterium strain SB3 and its use in transformation of H. halobium. Can. J. Microbiol. 35:86-91.

18. Hagerman,P.J. 1990. Sequence-directed curvature of DNA. Annu. Rev.Biochem. 59:755-781.

19. Halladay, J. T., J. G. Jones, F. Lin, A. B. MacDonald, and S. DasSarma. 1993. Therightward gas vesicle operon in Halobac-tenium plasmid pNRC100: identification of the gvpA and gvpC geneproducts byuseofantibody probesandgenetic analysis of theregion downstream of gvpC. J. Bacteriol. 175:684-692. 20. Halladay, J. T., W.-L. Ng, and S. DasSarma. 1992. Genetic

transformation ofahalophilic archaebacterium with a gas ves-icle gene clusterrestoresitsabilitytofloat. Gene 119:131-136. 21. Hofnan, J. D., L. C. Schalkwyk, and W. F. Doolittle. 1986. ISH51: a large, degenerate family of insertion sequence-like elements in the genome of thearchaebacterium, Halobacterium volcandi. Nucleic AcidsRes. 14:6983-7000.

22. Jones, J.G.,N. R.Hackett, J.T.Halladay,D.J.Scothorn,C.-F. Yang,W.-L.Ng,andS. DasSarma.1989.Analysis of insertion mutantsrevealstwo newgenes inthepNRC100 gas vesicle gene cluster of Halobacterium halobium. Nucleic Acids Res. 17: 7785-7793.

23. Jones, J. G.,D. C.Young,andS. DasSarma. 1991. Structure and organizationof the gasvesicle gene clusteronthe Halobacte-rium halobiumplasmidpNRC100. Gene 102:117-122.

24. Joshi, J.G.,W.R.Guild,and P.Handler.1963.Thepresenceof twospecies of DNA insomehalobacteria. J. Mol. Biol.6:34-38. 25. Kornberg, A.,and T. A. Baker. 1992.DNAreplication.W.H.

Freeman &Co.,NewYork.

26. Krebs,M. P., U. L. RajBhandary, and H. G.Khorana. 1990. Nucleotide sequence ofISHII,a newHalobacterium halobium insertion element isolated from the plasmid pGRB1. Nucleic AcidsRes. 18:6699.

27. Lam,W.L., andW. F. Doolittle.1989.Shuttlevectorsforthe archaebacterium Halobacterium volcanii. Proc. Natl. Acad. Sci. USA86:5478-5482.

28. Leffers, H., F. Gropp, F. Lottspeich, W. Zillig, and R. A. Garrett.1989. Sequence,organization,transcriptionand evolu-tion ofRNApolymerasesubunit genesfrom thearchaebacterial extreme halophilesHalobacterium halobium and Halococcus morrhuae.J. Mol.Biol. 206:1-17.

29. Miller,C.A.,S. L.Beaucage, andS. N. Cohen. 1990.Role of DNAsuperhelicityinpartitioningofthepSC101plasmid.Cell. 62:127-133.

30. Ng, W.-L.,andS. DasSarma. 1991. Physical andgenetic map-ping of the unstable gas vesicle plasmid in Halobacterium halobium NRC-1, p. 305-311. In F. Rodriguez-Valera (ed.),

General andappliedaspects ofhalophilicmicroorganisms. Ple-VOL. 175,1993

on April 4, 2021 by guest

http://jb.asm.org/

(13)

4596 NG AND DAsSARMA

numPress, NewYork.

31. Ng, W.-L., J. T. Halladay, and S.DasSarma. 1991.Isolation of large Halobacterium halobium plasmids, p. 3.5.1-3.5.3. In

E. M.Fleischmann,A.R.Place,F. T.Robb,and H. J.Schreier (ed.), Protocols for archaebacterial research. University of Maryland, Baltimore.

32. Ng,W.-L., S. Kothakota, andS. DasSarma. 1991.Structureof thegasvesicleplasmid in Halobacterium halobium: inversion

isomers, invertedrepeats,andinsertionsequences.J.Bacteriol. 173:1958-1964.

33. Novick, R. P. 1987. Plasmid incompatibility. Microbiol. Rev. 51:381-395.

34. Pfeifer, F., and M. Betlach. 1985. Genome organization in Halobacterium halobium:a70 kb islandofmore(AT) richDNA in the chromosome. Mol. Gen. Genet. 198:449-455.

35. Pfeifer,F., and U.Blaseio. 1989.Insertion elementsand deletion formation in a halophilic archaebacterium. J. Bacteriol. 171: 5135-5140.

36. Pfeifer, F., G.Weidinger, and W.Goebel. 1981. Genetic vari-abilityin Halobacteriumhalobium. J. Bacteriol. 145:375-381. 37. Sambrook, J.,E.F.Fritsch,and T. Maniatis. 1989. Molecular

cloning: a laboratory manual, 2nd ed., Cold Spring Harbor

Laboratory, ColdSpring Harbor,N.Y.

38. Sanger, F.,S.Nicklen, and A. R. Coulson. 1977. DNA

sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467.

39. Sapienza, C., and W. F. Doolittle. 1982. Unusual physical organization oftheHalobacterium genome. Nature (London) 295:384-389.

40. Sapienza, C., M. R. Rose, and W. F. Doolittle. 1982. High-frequency genomic rearrangements involving archaebacterial repeat sequence elements. Nature(London)299:182-185. 41. Schnos, M., K. Zahn, R. B.Inman,and F. R. Blattner. 1988.

Initiation protein induced helixdestabilizationatthe ? origin: a prepriming step inDNAreplication. Cell 52:385-395.

42. Simsek, M., S. DasSarma, U. L. RajBhandary, and H. G. Khorana. 1982. Atransposable element from Halobacterium halobium which inactivates thebacteriorhodopsin gene. Proc. Natl. Acad. Sci. USA79:7268-7272.

43. Vocke, C., and D. Bastia. 1983. Primary structureof the es-sentialreplicon of the plasmid pSC101.Proc.Natl. Acad. Sci. USA80:6557-6561.

44. Weidinger,G., G. Klotz,and W.Goebel. 1979.Alarge plasmid fromHalobactenum halobium carrying genetic information for gasvacuole formation. Plasmid 2:377-386.

45. Willetts, N.,and R.Skurray. 1987. Structureand function of the Ffactor andmechanism of conjugation, p. 1110-1133. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. Amer-ican Society for Microbiology, Washington, D.C.

46. Womble, D. D., and R. H. Rownd. 1988. Genetic and physical mapofplasmidNR1: comparison with other IncFII antibiotic resistance plasmids. Microbiol. Rev. 52:433-451.

47. Xu, W.-L., and W. F. Doolittle. 1983. Structure of the archae-bacterialtransposable element ISH50. Nucleic Acids Res. 11: 4195-4199.

J. BACTERIOL.