Mostrar el registro completo del ítem

(1)

Proc.Nati.Acad. Sci. USA

Vol. 91, pp. 12198-12202, December 1994 Biochemistry

Terminal protein-primed DNA amplification

Luis BLANCO, Jose M. LAZARO, MIGUEL

DE

VEGA,

ANA

BONNIN*,

AND MARGARITA

SALASt

Centro de BiologfaMolecular "Severo Ochoa" (Consejo Superior de Investigaciones Cientificas-Universidad Autonoma de Madrid), UniversidadAut6noma, Cantoblanco, 28049 Madrid, Spain

CommunicatedbyArthurKornberg,August9, 1994

ABSTRACT By using appropriate amounts of four bacteriophage

429

DNA replication proteins-terminal protein, DNApolymerase, protein p6 (double-stranded DNA-binding protein), and protein p5 (single-stranded DNA-binding protein)-ithas beenpossible toamplify limited amounts ofthe 19,285-bp-long 429DNA molecule by three orders of magni- tude after 1 hr of incubation at300C.Moreover,thequality of theamplifiedmaterial wasdemonstrated by transfection experiments,inwhichinfectivity ofthesynthetic(amplified)429 DNA, measured astheability toproduce phageparticles,was identical to thatof thenatural429 DNA obtained from virions.

The results presented in this paper establish some of the requisites for thedevelopmentof isothermal DNA amplification strategies based on the bacteriophage 429 DNA replication machinery thataresuitable for theamplificationofverylarge (>70 kb) segments ofDNA.

The genomeof Bacillussubtilisphage429 consists ofalinear double-strandedDNA(19,285bp),witha6-bp-longinverted terminal repeat andaterminalprotein (TP) covalentlylinked atboth 5' ends(reviewedinref. 1). Since the firstproposal by Rekosh etal. (2), bywhichafree molecule of TP would act as aprimertoinitiatesynthesisatboth endsofthe linear DNAmolecule,theavailability ofin vitroreplicationsystems for both adenovirus and bacteriophage 429 allowed the confirmation ofthishypothesis and the characterization of thefunctional role ofotherreplication proteins involved in this process(reviewedinref. 1).

Asdepicted inFig. 1, initiationof429DNAreplication istriggered bytheviralprotein p6 (double-stranded DNA- binding protein; DBP), which forms a

nucleoprotein

complex at both

429

DNA ends, producing aconformational change in the DNA thatprobablyleadstolocalopening of the DNA duplex (5). The primer TPforms a1:1 complex with

429

DNApolymerasethatrecognizes bothends of the linear

429

DNAmolecule(replication origins). Then,in the presence ofdATPand

Mg2+, 429

DNA polymerase cata- lyzestheformation ofacovalent bond between dAMP and the OHgroupof Ser-232 ofthe TPactingasprimer (6, 7).

Inthisreaction,dATPis selectedbybasecomplementarity with the second 3'-nucleotide of the template strand (8).

After this initiation step, dissociation of the TP-DNA polymerase heterodimer is likely to occur (transition) to replace the TP-DNApolymeraseinteractions requiredfor initiation by the DNA polymerase-DNA interactions requiredfor theelongationof the newly created DNA primer.

Concomitantly, anasymmetrictranslocation (sliding back) ofonly TP-dAMP, but not of the template, followed by addition ofa newdAMPresidue, allows the recovery of the information corresponding to the first template nucleotide (8). During elongation,

429

DNA polymerase catalyzes highly processive polymerization coupled to strand displacement(9),and, therefore, complete replication of both strands proceeds continuously fromeachterminal priming

event. Asthe two replication forks move, DNA synthesis is initially coupled to stranddisplacement of long stretches of single-stranded 429 DNA, producingtypeI replicative intermediates (seeFig. 1). When the tworeplication forks, movinginopposite directions, merge, a new typeof replicative intermediate (type II) is formed. Electron microscopy analysis of 429 replicative intermediates in vitro showedthat the viralprotein p5binds to thesingle-stranded portion ofboth type I and II molecules, thus acting as a single-stranded DNA-binding protein (SSB) during )29 DNAreplication (4). Oncepolymerization ofboth strands has beencompleted, the two DNApolymerase molecules dissociate fromthe DNAtoreassumeinitiation and replication ofa new 429 DNA molecule.

Thesymmetrical modeof429DNAreplicationis similarin several aspects(primed initiationattheendsofalinear DNA molecule and continuous synthesis of both strands) tothe mostwidely used DNAamplification technique:PCR. How- ever, the natural 429 DNAamplification systemhas signif- icantdifferencesderived from the nature of theprimers(TP), from the fact that both DNA endsare truereplication origins and, therefore, no thermaldenaturationisrequiredtoposi- tion primers; and from the specific strand displacement coupledtoDNAsynthesis catalyzed by 429DNApolymer- ase, which allows this enzyme to replicate extremely long double-stranded DNAmolecules inisothermal conditions.

In this paper, as the initial step in the development of amplification vectors based on the 429 DNA replication machinery, wehavecharacterized the minimal protein fac- torsrequiredin vitrofortheefficientTP-primed amplification oflimitedamounts of429DNA.

MATERIALS

AND

METHODS

Nucleotides, Proteins, and DNA Templates. Unlabeled dNTPs and

[a-32P]dNTPs

(410

Ci/mmol;

1Ci=37GBq)were obtained from Amersham. 429 DNA polymerase (Mr =

66,520)and TP (Mr = 30,918)were overproduced in Esche- richiacolicellsandpurified essentiallyasdescribed(10, 11).

Protein p6 (Mr = 11,873) and protein p5 (Mr = 13,212), obtained from429-infectedB. subtiliscells,werepurifiedas described (12, 13). TP-linked DNA from

429 susl4

(1242) virions, isolated as described (14), was used as input template forinvitroamplification experimentsand as acontrol DNA intransfection experiments.

Phage and Bacterial Strains. Bacteriophage

429 susl4

(1242) hasanonsense (sus)mutation in gene 14 (15), which produces a delayed lysis phenotype when plated in the nonpermissive strain B. subtilis11ONATry-spoA (16). B.

subtilisBG295 (Sup3) (17) was usedfor transfection experiments.B.subtilisMO-101-PspoA Thr-

(Met-)+

Su+44(18) Abbreviations: TP, terminal protein; pfu, plaque-forming unit(s);

SSB, single-stranded DNA-binding protein (p5); DBP, double- strandedDNA-binding protein(p6).

*Permanent address: Facultad de Medicina, Universida Com- plutense,28040Madrid, Spain.

tTowhomreprintrequestsshould be addressed.

12198 Thepublicationcostsof this article weredefrayedin partby page charge payment.This articlemust therefore behereby marked "advertisement"

in accordance with 18 U.S.C. §1734 solelytoindicate this fact.

(2)

Proc. Nati. Acad. Sci. USA 91 (1994) 12199

,-..t

I)BI' p6

I

Unr -1)BI]) phl

(Iplxopenlling

pr.-ll

s lrailnd-disp Ice IlellIi

OSSMSSSYSMSM'"Ssll P5\-

TvpeI 0~~o

r

⁰

/.S1)

^~ ^~^__ ^~~^;~oI^S\St2 i3Ala;ildtIol}\l^{1. ()} ^,G\^lI^O^)N NSSI3p-5remnova

I)X

Tvpe~~~~~

I I

119;

Jo4 3

comiipletionof

po1liyerizationi

TE'RNIINA'l11()N

,I,;,>,qgr~r\\r\\rJ-x>XACR~gRSII! ^{1)N'A pol.}

discsiation

was usedaspermissive host (suppressor strain)for titration of 429susl4 (1242).

*29 DNA Amplcn Assay. To optimizethe initiation step of

029

DNA replication, an equimolar (1:1) complex between highly purified 429 DNA polymerase and TPwas

obtainedbyincubation of bothproteinsin thepresenceof 20 mM ammonium sulfate, as described (19). The incubation mixturecontained,in10Il,50 mM Tris*HCl(pH 7.5),10mM

MgCl2,20 mM(NH4)2SO4, 1 mMdithiothreitol,4%glycerol, bovine serum albumin (0.1 mg/ml), 80

ALM

each dCTP, dGTP, dTTP,and

[a-32P]dATP

(2

AQCi),

15ngofpreformed TP-DNA polymerase complex, and 20 ng offree TP. As indicated,differentamountsof )29 DNA(containingparen-

talTP), obtainedfromphage #29mutantsusl4(1242),were

used as input DNA template. When indicated, different amounts ofpurified proteins p6 (DBP) and p5 (SSB) were

eitherindividuallyorsimultaneouslyadded. After incubation for 1 hrat30TC,thereactionwasstopped byaddition of 10 mMEDTA,and theunreacted[a-32P]dATPwasremovedby filtrationthroughaSephadex G-50 spincolumn in thepresenceof 0.1% SDS. Quantitationof the DNAsynthesizedin vitro,measuredasthetotalamount(in nanograms) ofdNTP incorporated, wascarried outfromthe amount ofradioac- tivity (Cerenkov radiation) corresponding to the excluded volume. Whenindicated,the size of theamplifiedDNAwas

analyzed by alkaline agarose gel electrophoresis (20) fol- lowedby autoradiographyandethidium bromide staining.

FIG. 1. Differentstagesandvi- ralgeneproducts involved in

029

DNA replication. For simplicity, onlyone4)29 DNA end(replication origin) is represented. 029 ^TP is indicatedin black(when actingas

primer) or shadowed (parental).

429 DNApolymerase is depicted byatriangle. Thetwomaintypesof replicative intermediates(I and II)

produced during the elongation stageandobserved both during in vivo(3) and in vitro (4)4)29DNA replication are represented in boxes.

Infectivi Assayfor the in Vitro-Ampf ed 4¢9DNA. In this

case, the in vitroamplificationconditionswere asdescribed aboveexceptthat theinput 429DNA(0.5 ng)wasincubated with TP-DNApolymerase complex (15 ng),free TP(20 ng), DBP(10 pg), and SSB (8

jig),

in the presence of the four unlabeled dNTPs each at 80

AM.

After 2 hr at 30"C, the reactionwasstoppedwith 10 mMEDTA,and the DNAwas

precipitated with ethanol and resuspendedin10mMTris-HCl (pH

7.5)/0.1

mM EDTA. Asanin vivocontrol ofinfectivity, 500ngof the 429 DNA usedas inputforamplificationwas

incubated under exactly the same conditions but in the absenceof dNTPs. After 2 hrat300C,thecontrol 429 DNA

was processed as described for the in vitro-amplified 4)29 DNA.Aliquotsof theamplified 429 DNAwerequantitated by alkaline agarose gel electrophoresis (20) and ethidium bromidestaining.Differentamountsof 429 DNA(controlor

amplified)wereusedtotransfectB.subtilis BG295(suppres-

sorstrain)-competent cells, preparedasdescribed(21). After transfection, infectivity, expressed asplaque-forming units (pfu), wasdeterminedby plating^onB. subtilis Su+44sup- pressorstrain.

RESULTS

Requirementsfor in VitroAmplfationof 029 DNA. Itwas

reported that the complete replication ofboth

029

DNA strands in vitrorequired onlytwoproteins: the 429 TP, which actsasinitiationprimer,andthe viralDNApolymerase (22).

T'P/ DNApJul.

CompI)lex

Biochemistry:

Blanco etal.

0

(3)

Proc. Natl. Acad. Sci. USA 91 (1994)

ws 675

10 'W 450

._

t 225 z

,- 100

_lQa 0To

fj' 60

= 40

0 20 5.

10 10

DII

500 50 5 0.5

029DNAinput,ng

FIG. 2. Invitroamplificationof429DNA.(A)DNAsynthesized in vitro with differentcombinationsof429DNAreplication proteins, as a function of the amount ofinput 429 DNA. The 429 DNA amplificationassay was carriedoutasdescribed in Materials and Methods,in the presenceof15 ngofpreformedTP-DNApolymerase complex, 20 ng of free TP, and different amounts of 429 DNA (containing parental TP)asinputDNAtemplate.Whenindicated,10 pgofpurified protein p6 (DBP)and 8pgofpurifiedprotein p5 (SSB) were added eitherindividually(o and A) orsimultaneously (e)tothe minimalsystem(o)previouslydescribed. After incubation for 1 hrat 30'C,thereactionwasstoppedandquantitatedasthe totalamount (innanograms) of dNTPincorporated. (B)DNAamplificationfac- tors, corresponding to the different conditions shown inA, were calculatedastheratio betweentheamountof DNAsynthesized in vitroand the amountofinput429DNA.o, DNApolymerase+TP;

a,DNApolymerase+TP+p6;A, DNApolymerase+TP+p5;*, DNApolymerase + TP+ p6 + p5.

However, this minimal systemwasdefined using aconsid- erableamountof

429

DNAtemplate (0.5

Mg;

1.6nM),and, therefore, wereconsidered the characterization ofthe pro- teinsrequiredforanefficient DNA

amplification

in vitro(i.e., startingfrom limitedamounts of429 DNA).

Fig.2A showsthat theamountofDNAsynthesizedwhen usingaminimal systemformedbyTPand DNA

polymerase

is strongly dependent on the amount ofinput 429 DNA.

Thus, in vitrosynthesisoccurredatinputdoses of 500 ng(4 nM) and 50 ng (0.4 nM), but itwas almost undetectable at

A

DNApoI+TP+p5(SSB)

input doses corresponding to 0.5 ng (4 pM) and 5 ng (0.04 nM); no reaction was observed even after overnightincuba- tion. Therefore, initiation is probably the rate-limiting step:it is very inefficient at low ratios between DNA replication origins and the minimal initiation proteins (TP-DNA polymerase complex). As expected from previous data (23), additionof DBP to the in vitro minimal systemproduced only a 1.5- to 2-fold increase in theamountof DNAsynthesized at anyof the input )29 DNA doses tested. Also in agreement with previous reports (24), addition of SSB to the minimal system produced astimulation of about 4- to 5-fold in the amountofsynthesizedDNA whenstarting with either500 ng or50 ngofinput )29 DNA; however, at the lower doses of input DNA tested, the effect of adding SSB was negligible.

On theother hand, the simultaneous addition of viral DBP andSSBto theminimal system resultedinahigh yield ofin vitro-synthesizedDNA, even whenstartingwith either 5 ng or 0.5 ng ofinput DNA template. Analysis of the titration curvesshown inFig. 3 indicates that both DBP and SSB are required inhighamounts toobtainanefficientamplification.

Thus, in the presence of the four viral proteins, it was possibletoamplify bythree ordersofmagnitudetheamount ofinput 429DNA(seeFig. 2B). Under theseconditions, an exponential increase ofin vitro-synthesized DNAcould be obtained as a function of the time of incubation (data not shown), indicating not only a high efficiency in terms of number ofinitiation events but also the occurrence offull elongationof both DNA strands.

Size Analysis ofthe in Vitro-Ampliflied

029

DNA. Fig. 4 (lanes 1-4) shows the sizeanalysisof theamplified

4)29

DNA obtained in vitroinanexperiment similartothat shownin Fig. 2,usingeachofthe fourcombinations ofviralproteins andstartingwith 0.5ngofinput 429DNA.Interestingly,the amplified DNA, obtained only in the presence of the four viral proteins (Fig. 4, lane 4), corresponded tofull-length429 DNA. Moreover, electronmicroscopy analysis ofthe DNA amplifiedunder theseconditions indicatedtheaccumulation ofmature 429 DNAmolecules (datanot shown). Theam- plificationfactor obtained(1460-fold)impliesamaximal time of about6min perduplication cycle; theamountof429DNA synthesizedunder theseconditions (730ng/58 fmol)isclose tothelimitimposed bythe amount ofdNTPs provided.

A delicate equilibrium (stoichiometry) of initiation and elongation factorsrequired fortheefficientin vitroamplifi- cationof429DNAcanbeclearly observedinFig. 4, lanes 5-9. At a fixed concentration ofDBP and SSB, a small increase in the amount of TP-DNApolymerase complex with respect to the one allowing optimal amplification of full- length 429DNAleads to thegeneration of immature elongation products, althoughthe amount ofDNA synthesized wasroughlysimilar. This result is interpreted as the conse- quence ofanimbalance between the number of initiations (increased) and the ratio of elongation factors to growing B DNApoI+TP+p6(DBP)

2 4 6

p5(SSB), tg

8 10

FIG. 3. Requirement ofproteins p6 (DBP)andp5(SSB)for

4)29

DNAampli- fication. The

429

DNAamplificationas- saywascarried out asdescribed in Ma- terials andMethods, but using 0.5 ng of 4)29 DNA (containing parental TP) as inputDNAtemplate, 15 ng of preformed TP-DNApolymerase complex, 20 ng of free TP, and either 8 pg of SSB and different amounts of DBP (A) or 10pgof protein DBP and different amounts of SSB(B).After incubation for 1 hr at300C, thereaction wasstopped andquantitated as the total amount (in nanograms) of dNTPincorporated.

._

la

U)

z

500'

X 400'

N

.= 300

0

200 z

100

4 6

p6 (DBP),

jg

⁰

12200

Biochemistry:

Blanco etal.

(4)

A

o291)N.\

i1928 nii

I 2 3 4 5 (6 7 X 9 11) 11

_-

*9

BI

o2')D)N\A

;I)2Xs ,[II -O

f,2)1)NAilII)l[t. II1 ' - ' ( (.'--- 17 3-7.$ 75 ¹1)

P 1(SSI 8^L - + - + + + + + +

p4 ori-I)I BItP).I( Ll -- + + ++ + + + amp1if1i~icaliinfactor I I II 1400 320 II461 1 (0

611 1'141)

-....

-1(1)

51}1

FIG. 4. Sizeanalysis of in vitro-amplified 429DNA.Amplifica- tionassayswerecarriedoutessentiallyasdescribed in Materialsand Methods, using the indicatedamountsofinput 429 DNA. Different 429 DNAreplication proteins [TP-DNApolymerasecomplex, p6 (DBP), and p5 (SSB)] were added as indicated. After 1 hr of incubationat300C,the reactionwasstoppedwith 10 mMEDTA,the amountof synthesized DNAwasquantitatedasdescribed, andthe size of the amplified DNAwas analyzed by alkalineagarose gel electrophoresis followed byautoradiography (A)and ethidium bromide staining (B). The amplification factor correspondingtoeach particular condition,calculatedasdescribedinthelegendtoFig.2, is indicated. The arrows at the left indicate the electrophoretic mobility of full-length 429 DNA obtained fromvirions.

DNAchains(reduced). Moreover,the size of thesynthesized DNAwaslinearlydependentontheamountof SSB(datanot shown). Ontheotherhand,in theabsence ofSSB,andusing

alargeamount (60 ng) ofTP-DNApolymerase complex,it

waspossibletoobservealarge stimulation intheamountof shortelongation productsas aconsequenceofaddingDBP (Fig. 4,lanes 10 and11).This resultallowsustoconclude that binding ofDBPtothe 4)29 DNAreplication originsisthe main factorstimulating the initiationstageinthese in vitroamplification conditions. Therefore, it is tempting to speculate that, in addition to facilitating the opening of 429 DNA replication origins, DBP increases the affinity of the TP- DNA polymerase complex by the 429DNA ends. On the otherhand, SSB,inadditiontopreventingthenonproductive bindingofreplication proteinstothedisplaced strands (25),

appearstobecriticalfor the progression(maturation) of the multiple replication forks initiatedatboth 4)29DNAreplica-

tionorigins.

All theseresultsagreewith theimportance of the four viral proteins for 429 DNA replication in vivo (26, 27) and also emphasize the necessity ofaprecisestoichiometry of these proteins to maintain the equilibrium between the initiation andelongation stages of replication.

Infectivityof the inVitro-Amplified 429 DNA. Asaquality controlof the in vitro amplification procedure described in thispaper,theinfectivityofthesynthetic (invitro) 429DNA,

-

p-b 1;.

0.

cJ4

0.1 1 10 100 1000

029DNA, ng

FIG. 5. Infectivity of thein vitro-amplified 4)29DNA. In vitro amplification conditionswere asdescribed in Materials and Methods for theinfectivityassay.Acontrol 429 DNA(invivo)wasprocessed essentially in the same form but in the absence of dNTPs. The indicatedamountsofeitherin vivo(control)or in vitro429DNA were used to transfect B. subtilis BG295 (suppressorstrain) competent cells. Aftertransfection, infectivity, expressed aspfu,was deter- minedby platingonpermissiveB.subtilisSu+44 cells.

measuredastheabilitytoproduce phage 429 particles,was compared with that of the natural(in vivo) 429 DNA.Thus, similaramountsof 429DNA,either obtained from

029

susl4 (1242) virions orby in vitroamplification of 429 susl4 (1242) DNA, were used to transfect B. subtilis competentcells,and thenumber of infectivecenters wasdetermined by platingon permissive(suppressor)B.subtilis cells. As shown inFig. 5, infectivity ofboth natural (in vivo)andsynthetic (in vitro) 429 DNA wasvirtuallyidentical;thetransfectionefficiencywas '104 pfu/pg. Moreover, the size distribution of individual plaques was alsoidentical, indicatingtheabsence of in vitro mutationsaffecting phage growth. These results indicate that the in vitro429 DNAamplification system produces mature molecules and also suggest that DNA synthesis is being carried out withhigh fidelity. This could be expected ifone takes into account previous estimations of the insertion

fidelity

of 4)29 DNA

polymerase

(28) andits strong 3' -* 5' exonuclease activity (29, 30). A minimal estimate of the fidelity of thisamplification system is the fact that noplaques wereobtained when thetransfectionmixture corresponding tothe invitro-amplified 429 DNA wasplatedin nonpermissive B. subtilis cells. This was the expected result if the nonsensemutation present in theinput 429 DNA wasmain- tained in the in vitro-amplified 429 DNA molecules.

DISCUSSION

In thispaper, one of the most efficient in vitro DNA repli- cationsystems described sofar ispresented. Byusing four 429 DNAreplication proteins-TP(primer), DNA polymerase, protein p6 (DBP), and protein p5 (SSB)-it has been possibletoamplifythe19,285-bp-long429 DNAmolecule by three ordersofmagnitudeafter 1hrofincubation at 300C. As itoccursduring 4)29 DNAreplication in vivo, each of these proteinswas anabsoluterequirement.Thebiological activity of the invitro-amplified 429 DNA was demonstrated by its ability to producephage429 particles when transfected into B. subtilis cells.

Itwasexpected that, under the different conditions used, the amount of DNA synthesized (amplified) would largely depend on the efficiency of initiation and thefrequency of reinitiations both on the sameand/ordifferent DNA molecules. Moreover, elongation was not expected to be rate-

TP: DNAwd,nu I ;

(5)

Proc. Natl. Acad. Sci. USA 91 (1994) limiting because once a molecule isinitiated, complete elon-

gation could be achieved in about 5 min(9). However, size analysisof the synthesizedDNAindicatesthat at ahighratio among initiation factors (TP, DNA polymerase, and DBP) and input DNA (replication origins), there is a blockage of elongationthatcanbeonly preventedby addition oftheSSB protein. This blockage could be due to a "mechanical"

hindrance produced bythe multiple displacedstrandsorigi- natedfrom frequent reinitiationsatthe sameorigin.

Avariety of nucleic acid amplificationtechniques, devel- oped as tools for nucleicacid analysis andmanipulation,are also being successfully applied for clinical diagnosis of ge- netic and infectious diseases. Amplification techniques (re- viewed in ref. 31) can be grouped into (i) those requiring temperature cycling(PCRand ligase chain reaction)and(ii) isothermal systems [transcription-based amplification systems(3SRandNASBA), strand-displacementamplification, and

QP

replication systems]. Twoaspects are frequent ca- veats intheseprocedures: fidelity ofsynthesis andlengthof theamplified product. As it has beenrecentlyreported, the

useof

proofreading polymerases (containing

3'^-- 5'exonu-

clease

activity)

in combination with 3' -- 5' exo-deficient enzymessignificantlyincreasesnotonlyfidelity but also the size limit ofthe amplification reaction (32, 33).

The resultspresented inthis paperestablishbasic require- mentsforthe development of anamplificationsystemrelying on the mechanism of phage 429 DNA replication. This systemwould be adequateforfaithful amplification of DNA molecules longer than 70kb (9), largely over the size limit obtained with the amplification systems available to date.

This procedureofisothermalTP-primedamplification would exploit thepeculiarproperties of 429 DNA polymerase: (i) ability to use aprotein asprimer, (ii)intrinsic high proces- sivity(>70kb),and(iii) stranddisplacementcoupled to DNA synthesis. This procedurewould alsorequire the action of the viral DBP,toefficiently recognize andactivate the 429 DNA replicationorigins (whichshould bepresent atthe ends of the target DNAtobeamplified),and SSB, to allow maturation of replicative intermediates andreinitiation on new DNA molecules. The sequences at 429 DNA ends that are active as minimalreplication origins (34) and thelocation of the protein p6nucleation siteabsolutelyrequired for protein p6 binding andactivation of the 429 DNA replication origins (3) have been determined. Therefore, DNA molecules suitable for TP-primed amplification could be directly obtainedby liga- tion ofdouble-stranded DNA fragments containing the 429 DNAreplication origins to both ends of the DNA molecule to beamplified.Alternatively, selective (targeted)amplification would require initial stages of single-stranded DNA primingbased on theuse of dualprimers (specific sequence plus 429 DNA origin sequence) to obtain specific double- stranded DNA linearmolecules flanked by 429 DNA replication origins. Once thesemolecules areformed andthe 4)29 DNAreplication proteins are provided, initiation, progression, and resolution steps would occur continuously, producing an amplification cascade similar to that occurring during viral infection.

This work isdedicatedto thememory of Severo Ochoa. We are grateful to Dr. C.Gutidrrezforthe electron microscopy analysis of 429DNAamplificationproducts. We are also grateful to Dr. J. C.

Alonso for the gift of B.subtiliscompetent cells. This investigation wasaided byResearchGrant 5R01GM27242-15 from the National Institutes of Health, by GrantPB90-0091from Direcci6n General de Investigaci6nCientificayTdcnica,byGrantBIOTCT 91-0268 from

theEuropeanEconomic Community, and by an Institutional grant fromFundaci6n Ram6nAreces.

1. Salas, M.(1991) Annu. Rev.Biochem. 60, 39-71.

2. Rekosh, D. M. K., Russell, W. C. &Bellett, A.J. D. (1977) Cell 11, 283-295.

3. Serrano, M., Gutidrrez, J.,Prieto,I.,Hermoso, J. M. & Salas, M. (1989)EMBOJ. 8, 1879-1885.

4. Gutidrrez, C., Sogo,J. M.&Salas,M.(1991) J.Mol.Biol. 222, 983-994.

5. Serrano, M., Salas, M. & Hermoso, J. M. (1990) Science 248, 1012-1016.

6. Blanco, L.& Salas,M.(1984) Proc. Natl. Acad. Sci. USA 81, 5325-5329.

7. Hermoso,J.M., Mdndez, E.,Soriano,F. &Salas,M.(1985) Nucleic Acids Res. 13, 7715-7728.

8. M6ndez,J.,Blanco,L.,Esteban,J. A.,Bernad,A.&Salas,M.

(1992)Proc. Natl.Acad. Sci. USA89, 9579-9583.

9. Blanco, L.,Bernad, A.,LIzro,J. M.,Martin, G., Garmendia, C.&Salas,M. (1989)J.Biol. Chem. 264,8935-8940.

10. LAzaro,J. M.,Blanco, L.& Salas,M.(1994) Methods Enzym- ol., inpress.

11. Zaballos,A.&Salas, M. (1989) Nucleic Acids Res. 17, 10353- 10366.

12. Pastrana, R., LAzaro,J.M., Blanco,L.,Garcfa,J.A., Mdn- dez,E. &Salas,M.(1985)NucleicAcids Res. 13, 3083-3100.

13. Martfn, G., Lazaro, J. M., Mdndez, E. & Salas, M. (1989) NucleicAcids Res. 17, 3663-3672.

14. Pefialva, M. A.&Salas,M.(1982) Proc. Nadl. Acad. Sci. USA 79,5522-5526.

15. Jimdnez,F.,Camacho,A.,de la Torre, J.,Vifluela,E.&Salas, M.(1977)Eur.J.Biochem. 73, 57-72.

16. Moreno, F., Camacho, A., Vifuela, E. & Salas, M. (1974) Virology62, 1-16.

17. Chai, S., Szepan, U.,Luder, G., Trautner, T. A.& Alonso, J. C.(1993)Gene129, 41-49.

18. Mellado, R. P., Vifuela, E. & Salas, M. (1976) Eur.J. Bio- chem. 65, 213-223.

19. Blanco,L.,Bernad, A.,Esteban,J. A. &Salas, M. (1992)J.

Biol. Chem. 267, 1225-1230.

20. McDonell, M.W.,Simon, M. N. & Studier, F. W. (1977)J.

Mol.Biol. 110, 119-146.

21. Rottlander,E.& Trautner, T. A. (1970) Mol. Gen. Genet. 108, 47-60.

22. Blanco,L.&Salas, M.(1985)Proc.Nail. Acad.Sci. USA82, 6404-6408.

23. Blanco, L., Bernad, A. & Salas, M. (1988) J. Virol. 62, 4167-4172.8.

24. Salas,M.,Bernad, A.,Zaballos,A.,Martfn,G., Otero,M.J., Garmendia, C., Serrano, M., Blasco, M.A., Lkzaro, J.M., Pares, E., Hermoso, J. M. & Blanco,L. (1990) inMolecular Mechanisms onMolecularandCellularBiology,eds. Richard- son, C. &Lehman,R.(Liss,NewYork),Vol.127,pp. 277-288.

25. Martin,G. &Salas,M. (1988)Gene67, 193-201.

26. Talavera,A.,Salas,M. &Vinluela,E.(1972)Eur.J.Biochem.

31, 367-371.

27. Carrascosa, J. L., Camacho, A., Moreno, F., Jimdnez, F., Mellado, R. P., Viniuela, E. & Salas, M. (1976) Eur. J. Bio- chem.66, 229-241.

28. Esteban,J. A.,Salas,M. &Blanco, L.(1993)J.Biol. Chem.

268, 2719-2726.

29. Blanco, L. & Salas, M.(1985) Nucleic Acids Res. 13, 1239- 1249.

30. Garmendia, C., Bernad, A., Esteban, J.A., Blanco, L. &

Salas,M.(1992)J.Biol. Chem. 267, 2594-2599.

31. Landegren, U. (1993) Trends Genet. 9, 199-202.

32. Barnes, W. M. (1994) Proc. Natl. Acad. Sci. USA91, 2216- 2220.

33. Cheng, S.,Fockler, C., Barnes, W. M. &Higuchi, R. (1994) Proc.Natl. Acad. Sci. USA 91,5695-5699.

34. Gutierrez,J.,Garcia,J. A.,Blanco,L.&Salas,M.(1986) Gene 43, 1-11.

12202

Biochemistry:

Blancoetal.