Effects of alterations of primer-binding site sequences on human immunodeficiency virus type 1 replication.

(1)

0022-538X/94/$04.00+0

Effects of

Alterations of

Primer-Binding

Site

Sequences

on

Human

Immunodeficiency

Virus

Type

1 Replication

XUGUANG LI, JOHNSONMAK, ERICJ.ARTS,ZHENGXIANGU, LAWRENCEKLEIMAN, MARK A. WAINBERG,*ANDMICHAEL A. PARNIAK *

Departments of Medicine and

Microbiology,

LadyDavisInstitute-Jewish General

Hospital,

and McGillUniversity

AIDS

Centre,

Montreal,

Quebec,

Canada H3T 1E2

Received 31 March 1994/Accepted 29 June1994

Thehumanimmunodeficiency virus type1genomicRNAprimer-bindingsite(PBS)sequencecomprises18 nucleotideswhicharecomplementarytothose at the 3' end of thereplication initiationprimer

tRNAALYS.

To investigate theroleofthePBS in viralreplication,weeither deleted theoriginalwild-typePBS(complementary

to

tRNA3YS)

orreplaced it withDNA sequencescomplementarytoeither

tRNAlys

ortRNAPhe.Transfection of

COS cellswith such molecularconstructsyielded similar levels of viralprogeny thatwereindistinguishable with regard toviral proteins and tRNA content.Virus particles derived from PBS-deleted molecular clones were noninfectious for MT-4, Jurkat, and CEM-T4 cells. However, infectious viruses were derived from constructs in which the PBS had beenaltered tosequences complementary to either

tRNA'LYs

ortRNAPhe, although mutated forms showed significant lags in replication efficiency in comparison with wild types. Molecular

analysis

ofreverse-transcribed DNA in cells infectedby the mutated viruses indicated that both tRNA1LY andtRNAPhe could functionasprimersforreversetranscription duringtheearlystagesof infection. Sequencing offull-length proviralDNA,obtained 6daysafterinfection,revealedthe mutatedPBS,indicating that acomplete cycle ofreversetranscriptionhadoccurred.During subsequentrounds ofinfection,reversion ofthemutatedPBStowild-typesequences wasobserved,accompanied by increasedproductionofviralgene products. Reversion towild-type PBS sequences wasconfirmed bothbyspecificPCR

analysis,

usingdistinct primer pairs, and by direct sequencing of amplifiedsegments.We alsoperformed endogenousin vitroreverse transcription experiments in which synthesis of minus-strand strong-stop viral DNA was primed from a syntheticRNAtemplatecontainingaPBS complementarytovarioustRNAisoacceptors. These resultsshowed thattRNA3ySwas amuch moreefficientprimer ofsuchreactions than eithertRNAlLY ortRNAPh.

Anearly, criticalstepin thehumanimmunodeficiencyvirus type (HIV-1) life cycle is reverse transcription of viral RNA into proviral DNA, which can then be integrated into the infected host cellgenome. Thisprocess is carried outby the

multifunctional viral enzyme reverse transcriptase (RT) and

requires a primer annealed to a single-stranded template to initiate DNAsynthesis (9, 14, 18, 45, 47,51).All retroviruses use acell-derivedspecifictRNAasprimer,which is packaged into maturevirions (10, 11, 17, 22, 25, 34). Eighteen

nucleo-tides(nt) atthe 3' end of

tRNALYS

arecomplementary to an 18-nt sequence of HIV-1 genomic RNA, termed the

primer-binding site (PBS). The PBS is found approximately 180 nt from the 5' end of the viralgenome(39). Preferential

packag-ingof

tRNALYS

and its tight association with viralRNAinthe

virion suggests that it may function as a primer of reverse transcription from the PBS. Inaddition, HIV-1 RThasbeen shown to bind specifically to

tRNA3YS

through interactions with the anticodon loop,

TPC

loop, andDloop of

tRNALYS

(1, 2, 4, 27). However, the question of specific binding between HIV-1 RT and complexes of

tRNA3YS

annealed to a PBS template hasnotbeen studied.

Theviral genomic PBS is believed bothtoprovide a site for bindingof primer tRNA, thereby allowinginitiation of reverse transcription, andtofacilitate thesecond template switch (36, 46, 48, 50). Neither of these functions is well understood. In addition,the PBSmayplaya role inthe specific selection and

*Corresponding authors. Mailing address: Lady Davis

Institute-Jewish General Hospital, 3755 Cote Ste-Catherine Rd., Montreal, Quebec, Canada H3T 1E2. Phone: (514) 8260. Fax: (514) 340-7502.

packaging of primer

tRNA'YS.

Primer

tRNA'YS

mayalso play a dual role in HIV reverse transcription, by initiating RNA-dependent DNApolymerization from the PBS and by actingas the PBStemplate during synthesis of plus-strand DNA (9, 13). To study these multiple functions, the wild-type PBS se-quence (complementary to the 3' end of

tRNA3YS)

in an infectious molecular clone of HIV-1 was either deleted or replacedwith sequencescomplementarytoeithertRNAY2'or tRNAPhe. The rationale for choosing these particular tRNA specieswasthat (i) tRNAY2' is utilized asprimer for reverse transcription in other retroviruses,e.g., Mason-Pfizer monkey virus (27,50), whereastRNAPhehasneverbeenidentifiedas a reversetranscription initiation primer, and (ii) whiletRNALys

ispackaged intowild-type HIV-1 inamounts even greaterthan those of thewild-type primer

tRNALYS,

tRNAPhe ispresent at much lowerlevels, i.e., around 1% (22, 49).

Wenowdescribethe effects ofalterations in the HIV-1 PBS on viral replication and on virion tRNA content. We found that thequantities and patterns of tRNA species incorporated

intovirionswereunaffected either by the absenceof a PBS or bythepresenceofaltered PBSsequences,indicatingthat the PBS does not play a significant role in the selection and incorporationofprimer tRNA duringHIV-1 assembly. How-ever,deletion of the 18-ntwild-type PBS completely abolished viral infectivity, whereas its replacementwith sequences com-plementary to either tRNALys or tRNAPhe impaired but did not abort viral infectivity. Interestingly, the mutant PBS se-quences revertedtowildtype during infection. The ability of these various viruses to replicate was closely related to the statusof the PBS.Duringearlystagesofinfection, thetwoPBS mutants, in conjunction with tRNA1Ls and tRNAPhe, appar-6198

on November 9, 2019 by guest

http://jvi.asm.org/

(2)

ently functioned as primers for reverse transcription, although less efficiently than wild-type primer

tRNALYS.

(This research was largely performed by X.L., under the joint supervision of M.A.W. and M.A.P., in partial fulfillment of therequirementsfor the Ph.D. degree, Faculty of Graduate Studies and Research, McGill University, Montreal, Quebec,

Canada.)

MATERIALS AND METHODS

Cells, viruses, and plasmids. The following reagents were obtained through the NIH AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases: the pBH10 noninfectious

plasmidused as a nick translation probe in Northern (RNA) blots (donatedby R. C. Gallo); the HXB2D infectious molec-ular clone of HIV-IIIB, containing the full-length HIV-1 proviral genome(providedby G. Shaw and B.Hahn); and the

CD4+ MT-4 cell line (contributed by D. Richman). Other cells, including the CD4- RD (rhabdosarcoma), simian-de-rived COS-7, CD4+ CEM, and Jurkat cell lines, were pur-chased from the AmericanTypeCulture Collection, Rockville, Md. Each of the CD4+ cell lines was susceptible to acute

cytopathic HIV-1 infection, while the CD4- lines were not. Cells were routinely maintained in RPMI 1640 medium

(Gibco-BRL Laboratories,Toronto,Ontario, Canada)

supple-mented with 10% fetal calf serum, 100 Uof penicillin per ml, and 100 ,ug of streptomycin per ml. pSCV21, a eukaryotic

expression vector containing the full-length HIV-1 genome withasimian virus 40origin of replication(15),wasagift of E.

Cohen, Universite de Montreal, Montreal, Quebec, Canada.

pSVK3andpSP72werepurchasedfrom Pharmacia Inc.

(Mon-treal, Quebec,Canada)and Promega(Nepean, Ontario, Can-ada), respectively. Recombinant HIV-1 RT

(p66/p5i

het-erodimer)waskindly provided by Casey Morrow, Universityof AlabamaatBirmingham. Restriction enzymes and other

mod-ifiedenzymeswereobtained from Pharmacia.

Construction of HIV-1

proviral

plasmidswithaltered PBS sequences.The PBSsofHIV-1 molecular cloneswerealtered

by usingacombinationof linkerreplacementand site-directed

mutagenesis (44). Briefly,HXB2Dwas cutwithSmaIandApaI to generatea3.7-kbfragmentcomprising the 5' region of the HIV-1 proviralgenome andcellular

flanking

sequences. This

fragment was subcloned into Smal-Apal-digested pSVK3 to give pSVPBS,whichwasused for the subsequent construction

of PBS mutants. pSVPBS was digested with NarI and then treated withamungbean nucleasetogeneratebluntends. The

plasmidwas then cutwithBssHII to remove 70 HIV nucleo-tides including thePBS and downstream sequences. We then ligated various 70-nt oligonucleotide sequences into the gap created by the NarI and BssHII

digestions

described above. These inserts contained PBS-like sequences

complementary

to

tRNA'12'

and

tRNAPhe

to

yield pSVPBS-Lysl,2

and

pSVPBS-Phe,respectively. Standard site-directed

mutagenesis

wasused todelete the entire18-nt

region

of

pSVPBS

to

yield pSVPBS(-).

The sequences ofourvarious 18-nt PBS constructs are as follows: 5'-TGG CGC CCG AACAGGGAC-3'

(pPBS-WT),

5'-TGG CGC CCAACGTGG GGC-3'

(pPBS-Lysl,2),

and 5'-TGG TGC CGA AAC CCG GGA-3'

(pPBS-Phe).

The SmaI-ApaI fragments from

pSVPBS,

pSVPBS(-),

pSVPBS-Lysl2, and

pSVPBS-Phe

were then cloned into the

appropriately

digested

pSCV21

molecular clone of HIV-1 to yield expressionplasmids

containing

full-length

HIV-1

proviral

DNA with thewild-type PBS

(pPBS-WT)

and mutated PBSs

[pPBS(-), pPBS-Lysl,2 and

pPBS-Phe].

All constructs were

sequenced to verify that the correct modifications in PBS sequencehad been achieved.

Each of the plasmids described above, containing altered PBS sequences, was cut with BglII and PstI to generate a

fragment of947 bp (473 to 1420), comprising the PBS/U5/R region of HIV-1 proviral sequences (3). Such fragments were thenligated intothe RNA expression vector pSP72, at the gap created byPstI and BglII, to generate various HIV-1 RNA

expression plasmids containing alteredPBS sequences.

Synthesis of minus-strand strong-stop DNAin an endoge-nous invitro RT assay. The HIV-1 RNA expression vectors

describedabovewerelinearized byAccI (nt 956) (39)and used inthe Promega Riboprobe Gemini Core System to generate runofftranscripts of483ribonucleotideswithalterations in the PBS region. In vitro RT assays were carried out in a volume of 20 pLl containing 10mMdithiothreitol,50 mMTris-HCl (pH 7.8),100mMKCl,10 mMMgCl2,and 0.2 mM each of the four

deoxynucleoside triphosphates as described previously (3).

tRNAlL'2s,

tRNALYS,

and

tRNALYSPhe

werepurifiedfrom human placenta(22, 42). Reaction mixturesweresupplemented with 0.3 ,ug of recombinant HIV-1 RT together with 200 U of RNasin(RNase inhibitor; Gibco-BRL,Toronto,Ontario, Can-ada)and incubated at37°Cfor upto 15 min. Reactionswere terminatedatvarioustimesby addition ofEDTAto 100 mM.

The terminated mixtures were extracted with phenol-chloro-form and chloroform and passed through a Sephadex G-25

(Pharmacia)

column to removeunincorporated free

radioac-tive nucleotides.Theproducts of these reactionswereboiled

for 4 min in formamidegel loadingbuffer

(44)

and chilledon ice for 5 min before being

loaded

onto a 5% denaturing polyacrylamide gel. The full-length minus-strand strong-stop DNAsynthesizedinthese reactionsis 249ntlong.

Infection of target cells.COS-7cellsweretransfected with thepPBSconstructsby electroporation.After 60 h of

incuba-tion,cell-free virus stockswere

prepared by centrifugation

of culture supernatants at 3,000 rpm at 4°C for 30 min in a Beckman bench-top centrifuge and filtration through a

0.2-,um-pore-size sterile membrane

(Becton Dickinson, Oxnard,

Calif.).

To remove possible contaminating

plasmid

DNA,

which could interfere with our PCR assays

(see below),

the virus stockwastreatedat37°C for30 min withexcessDNase Iat afinal concentration of 100U/mlin the presence of 10mM MgCl2 (35). Thevirus-containing clarifiedculturefluidswere storedat

-70°C

untiluse.

Infectiousnessof virus

particles produced by

transfection of

COS-7cellswasdetermined

by using

MT-4,

Jurkat,

or CEM cells as targets.

Briefly,

5 x

105

cellswere harvested

during

exponential growth,

washedonce

by

centrifugation,

and incu-bated in medium

containing

5 ngof viral

p24,

supplemented

with 10 ,ugofPolybreneper

ml,

at

37°C

for 3 hwithoccasional gentle shaking.Unbound viruseswereremoved

by

four sepa-rate centrifugation washings in serum-free

medium,

and the cellswere

resuspended

infresh medium. Toensure

complete

removal of

contaminating plasmid DNA,

medium from the

fourthwashwas checked

by

PCR

using primer

pairs specific

forthe HIV-1RTgene

(16).

Cellculturemediumwas

changed

at 3-to 4-dayintervals.

Samples

of cells and cell-free culture

supernatants were collected at

regular

intervals and

assayed

for virus content

by

HIV-1

p24

antigen

expression

and RT assay

(5). Samples

from MT-4 cells infected

by

heat-inacti-vated pPBS-WT

(60°C,

30

min)

served as

negative

controls

(52).

PCR

analysis

andDNA

sequencing.

Infected cells

(5

x

105)

were

suspended

in 0.5 ml ofTEbuffer

(50

mM

Tris-HCl,

1mM EDTA

[pH

8.0])

containing

0.5% sodium

dodecyl

sulfate

(SDS)

and 0.5 mg of pronase per ml and incubatedat

37°C

for

on November 9, 2019 by guest

http://jvi.asm.org/

(3)

A Gag Pol Env

U3

I

R

I

US

PS Lysl,2

Phe

Lys3

PA0

I3V U-'

PA

Sequences

PS

Lys3 Lysl,2 Phe

PA

5'-AGA CCA GAT CTGAGCCTG GGA-3' 5'-GTC CCT GTT CGG GCG CCA-3

5'-GCCCCACGT TGGGCGCCA-3' 5'-TCC CGGGTT TCG GCACCA-3'

5'-CCCATC GATCTAATT CTCCCC-3'

B Pnmerpair PS/Lys3 PS/Lysl,2

PS/Phe PS/PA

SZeofprodus 185bp 185bp 185bp 367bp

meaning of products PBScomplementarytotRNkYS3 PBScomplementarytotRNA2ys PBScomplementarytotRNAhe FulllengthproviralDNA

FIG. 1. Primerlocation andstrategyfordetection of viralDNAby PCR.(A) SequencesofprimersusedinPCR.(B)PCRstrategy.Primerpairs werechosentodistinguish the threetypesofPBSstudied(PS/Lysl,2, PS/Phe,andPS/Lys3)or todetectfull-lengthproviralDNA(PS/PA).The latterprimerpairwasused toamplifya regionflankingthe PBS.Amplifiedfragmentswere subsequently analyzedby usingthe PSprimerto sequenceminus-strandDNAand the PAprimertosequenceplus-strand DNA,usingadsDNAcyclingsequencingsystem(Gibco-BRL).

5to8 h withgentle shaking.Thesampleswerethen extracted with TE-saturated phenol and chloroform-isoamyl alcohol.

Both total DNA and high-molecular-weight (HMW) DNA were isolatedbystandard methods

(44);

theformerwasused for PBS sequenceanalysis,while the latter served for

determi-nationsofintegratedHIV-1 proviralDNA.Sampleswerethen

analyzedbyPCR.

Selectedprimerpairswereused inPCR

analysis

ofthe PBS sequences of various viral DNA species (unintegrated inter-mediatesorintegrated forms).The sequences, locations, and orientations of the primer pairs, designed to detect DNA

specieswith contiguous R and PBS sequences, as well as a description of the products formed, are illustrated

schemati-callyinFig.1.Primerpair PS/PAamplifies full-lengthproviral DNA and therefore detects the completion of reverse tran-scription (13, 52). Three sets of primer pairs, PS/Lysl,2,

PS/Phe,andPS/Lys3,wereused todistinguishthe three types of PBS studied. To distinguish PBS forms, highly stringent

PCRs were performed with 50 ,ug of sample DNA, 50 mM

Tris-Cl (pH 8.0), 50 mM KCl, 2.5 mM MgCl2, 5 pmol of

32P-end-labeledsense primer, and 20 pmol of unlabeled anti-senseprimer. Reactions were run for 25 cycles of 94°C (2 min) and65°C (2 min). Other PCRs were performed with 50 pM of unlabeled primers (sense and antisense) for 30 cycles of

94°C (2

min),

60°C (2 min), and 72°C (2 min). Reactions

were standardized by simultaneous amplification of ,-globin DNA (52) (primer pair, 5'-ACACAACTGTGTTCACTAG

C-3' [sense] and5'-CAACTTCATCCACGTTCACC-3'

[anti-sense]).Fordirectsequencing of theR/U5/PBSregion(LTR/ Gag), PS/PA amplified fragments (367 bp) were resolved by

electrophoresis, purifiedby electroelution, and sequenced by

using a PCR-based double-stranded DNA (dsDNA) cycling

sequencingsystem(Gibco-BRL).

Identification of tRNAspecies invirusparticles. A dot blot

assay using DNA oligonucleotides complementary to the 3' end of

tRNA3YS

(probesequence,5'-TGGCGCCCGAACAG

GGAC-3'), tRNALys (probe sequence, 5'-TGGCGCCCAAC

GTGGGGC),

ortRNAPhe (probesequence,5'-TGGTGCCG

AAACCCGGGA-3')wasused toidentifyspecific tRNA spe-cies. Positive controls, including

tRNA1,2,

tRNA'he, and

tRNALYS,

were purified from human placenta (22, 42). The

specificitiesof theseprobesandhybridization conditions have

been described elsewhere (22, 30). Total RNAwas purified

from viruses as describedpreviously (6), and the amount of viral RNA was normalized according to copy numbers of HIV-1 genomic RNA. Total RNAcorrespondingto4 x

108

copiesof viralgenomicRNAwasused ineachanalysis.RNA

sampleswereblotted ontoHybondNfilters(Amersham) and hybridized separately with each of the three probes. Following

high-stringencywashing(22, 30),thefilterswereairdried and

exposedtoX-ray filmat -70°C.

RESULTS

Alteration of PBS sequences does not affect expression of theproviral genome. The infectious HIV-1 clone pPBS-WT,

which possessesawild-type PBS complementary to

tRNALYS,

wasaltered (i)bydeleting the 18-nt PBS to give pPBS(-) or

(ii)by replacing the

wild-tpe

PBSwith sequences complemen-taryto

tRNA1LY

ortRNA

he

togivepPBS-Lysl,2orpPBS-Phe, respectively. These mutant constructs were transfected into COS-7 cells, and virus particles were harvested after 60 h. Northern blotting and Western blotting (immunoblotting) were performed to study expression of the proviral genome. All ofourwild-type(pPBS-WT)and mutatedmolecular clones

produced the usual three major HIV RNA transcripts; no differences were noted among the various clones, nor were differences found in the patterns of proteins in the viral

j ImIn I Il= I

I

on November 9, 2019 by guest

http://jvi.asm.org/

(4)

E

cL

-%.w

0 3 6 9 12

Days post-infectio

15 18 21 24 27

B

4k

4.. . . . I . .1

4 6 8 10

Days post-infection

FIG. 2. (A) Infection ofMT-4 cells with viruses harvested from COS-7 transfections. Viral inoculawereequalizedonthe basis of eitherp24

content(5 ng)orRTactivity (800,000 cpm)toinfect5 x 105cells. Cultureswereregularly monitored for HIV-1 production byRTassayfollowing

infectionby PBS(-) (O), PBS-Lysl,2 (0), PBS-Phe (A),orPBS-WT(0).A,infection by pPBS-WT (heat-inactivated virus). No fresh cellswere

addedduring the 24-day study period, in ordertoobserveviral particleaccumulation in culture supernatants.(B)Second round of infection of MT-4cells by viruses obtained from initially infected cultures (9 daysfollowing infection by pPBS-WT and 18 days following infection by pPBS-Lysl,2orpPBS-Phe). Forsymbols, seeabove.

progeny produced after transfection with each of the

con-structs.

We also transfected the various PBS constructs into the CD4- RD cell line, which permits only one round of viral

replication (32). No significant differenceswereobserved with

regard to levels of RT activity in culture fluids after various times followingtransfection withWT, pPBS(-), pPBS-Lysl,2,orpPBS-Phe (datanotshown). The transientnatureof these infections isapparentby the peak in RT activity after 10 to 15 days. These results arenot surprising, consideringthat neither the PBS norreverse transcription is involved in viral replication followingtransfection with proviral DNA.

Effects ofalterations in the PBS on viral infectivity. Viral

particles were harvested from COS-7 transfection cultures after 60h, normalized accordingtop24content(approximately 5ng),and usedtoinfectavarietyofHIV-susceptiblecelllines, including MT-4, Jurkat,andCEM.Culturesupernatantswere

regularly monitored both for HIV-1 p24 antigen expression and RTactivityover4weeks.

Infection of MT-4 cells withpPBS-WTresulted in therapid

emergenceof RTactivity, syncytia, p24 Ag+cellsasmeasured by indirect immunofluorescence assay, and other cytopathic effects within 2days. Virtually, all cellswerep24 Ag+ after 1

week, at which time RT activity had peaked. RT activity graduallydeclined with death of infected cells.Incontrast,only 5 to 10% of cells transfected by pPBS-Lysl,2 or pPBS-Phe

werep24+ after thisperiod, and RTactivityinculture fluids

was low (Fig. 2A). Virus particles produced from COS cells transfected withpPBS(-)wereunabletoinfect MT-4 cells. No cytopathic effect, p24 antigen, orRTwas detectedeven after 30days.We noted that virusproduction bycells infected with thetwo PBSmutantsreached normallevelsby15 to 18days. Similar virus production kinetics were noted when either Jurkat or CEM cells were used as targets (data not shown), suggesting that cell type differences werenot responsiblefor the observed results with mutatedPBS-containingviruses.

The progeny of these MT-4 infections (9 days from pPBS-WTinfection and 18daysfromLysl,2 and pPBS-Pheinfection)werethenusedinasecond round of replication

in MT-4cells. Figure2Bshowsthathighratesof replication, equivalent to those obtained with wild-type viruses, were

observed when the progeny of pPBS-Lysl,2 or pPBS-Phe,

obtainedat18days,werestudied forthe abilitytoinfect MT-4 cells. In contrast, when the progeny of MT-4 infections, obtained after 6days, were comparedforinfectiousness in a

second-round infection,infection kinetics similar to those of Fig. 2Awereobserved (datanot shown). This resultsuggests that wild-type PBS forms had preferentially emerged during the 24-day period of study; this subject will be considered in Discussion.

PCR analysis of the PBS sequence of proviral DNA in infectedcells. ThespecificityofourPCRassay wasmonitored bymixing0.1ngof cloned HIVplasmid (wildtypeormutated)

with 50 ,ugof MT-4 DNA from uninfected cells. The R/U5/ PBSregionof eachtypeof HIVgenomicDNAwasamplified

by usingthe threesetsofprimer pairs,PS/Lysl,2,PS/Phe,and PS/Lys 3, described in Fig. 1. Each HIV clone could be amplified only byitsownspecific primer pair,e.g.,pPBS-Lysl,2 by PS/Lysl,2andnotPS/Lys3orPS/Phe (Fig. 3A,lane2). Also, primer pair PS/Lysl,2couldnotamplifyanyof theR/U5/PBS

region of PBS(-) (Fig. 3A, lane 1), pPBS-Phe (lane 3), or

pPBS-WT (lane 4).

The three sets of primer pairs were used to analyze PBS sequencesin DNA harvested from MT-4 cellsatvarioustimes after infectionbythe various viral clones.Total cellular DNA

wasusedtosimultaneously detect PBSsequencesin interme-diate viral DNA species as well as in full-length integrated proviral DNA. At all times,we detected only wild-type PBS sequences in DNA extracted from MT-4 cells exposed to pPBS-WT virions, starting3 daysafter infection (Fig. 3B).At 24 days after infection, the intensity of this band had de-creased, apparentlyas a result of virus-inducedcytopathicity. E

I-XL io6

105

104 12 14

106.

on November 9, 2019 by guest

http://jvi.asm.org/

(5)

A

Primer pai r z o.ttt4. t

TABLE 1. PBS sequencesdetected by directdsDNAsequencing'

MT-4 cells PBSsequence

infectedby: 6 days postinfection 24 dayspostinfection pPBS-WT TGGCGCCCGAACAGGGAC TGGCGCCCGAACAGGGAC

pPBS-Lysl,2 TGGCGCCCAACGTGGGGC TGGCGCCCGAACAGGGAC pPBS-Phe TGGTGCCGAAACCCGGGA TGGCGCCCGAACAGGGAC

aPrimerpair PS/PAwasusedtoamplifyaregionflanking thePBS.Amplified

fragmentswere subsequently analyzed by usingadsDNAcycling sequencing

system(Gibco-BRL).

B

PS/Lysl,2

PS/Lys3

C

PS/Lysl,2

PS/Lys3

D

PS/Lysl,2

PS/Lys3

FIG. 3. (A) S5 quences. PSserve

antisenseprimers. used to detect

tRNA3L',

respect [pPBS(-), pPBS-50,ug ofuninfecti analyses usingtht the threeprimer I range ofplasmid obtained with 0.1j

alsoshown.(Bto] TotalDNA fromi usingthe threese

Dshowresults frc PBS-Phe, respect

I-Globin serveda

W * w In contrast, only mutated PBS forms were present in cells

infected with pPBS-Lysl,2 orpPBS-Phe at 6days after infec-1 2 3 4 5 6 7 8 9 10 l1 12 tion

(Fig.

3C and

D).

By

9

days

after infection, the

wild-type

PBS had started to emerge, while mutated PBS forms had 3 days 6 days 9 days 12days 1 5 days 24 days begun to disappear, concomitant with high levels of release of infectiousprogenyby these cells(Fig.2A). To ruleoutpossible PCR artifacts, we used primer pair PS/PA(Fig. 1)toamplify a

+ + + + + + 367-bp fragment flanking the PBS region in DNA isolated

+ + + + + + from MT-4 cells infected with each of the wild-type and

+ + + + + + mutated viral forms. Sequencing confirmed that the PBS which

+

., * ,* was detected

early

(i.e.,

up to 6

days)

in cells infected

by

mutated viruses contained the

original

mutations.

However,

at later

times,

only

wild-type

PBS sequences were

found,

while

mixtures were apparently present at intermediate timepoints, assuggested by the presence of multiple bands atsuchtimes. 3 days 6 days 9 days 12days 1 5 days 24 days Asummary of DNA sequencing resultswith respect to the PBS

, mrI isshown in Table 1.

+ ++

,+

+,

+ + Determination of minus-strand

strong-stop

DNA

synthesis

+ + + + + + asmeasured by endogenous invitro reverse transcription. To

+ + + + + + determine the

efficiency

of reverse

transcription

initiated

by

different tRNA primers, we used an endogenous in vitro * _, *. * ** reverse transcription assay (3) that employed synthetic RNA templates containing mutated PBS sequences together with 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1516 17 18 their respective tRNA primers. When synthetic pPBS-WT RNA template was primed with tRNALys, full-length minus-3 days 6days 9days 12days 15days 24days strand

strong-stop

DNA

products

weredetectablewithin 1 min

days 6rdays 9days 12 days i 5 days 24 days of addition of RT to the reaction mixture

(Fig.

4,

lane

1).

However, when reactions involving synthetic pPBS-Lysl,2 and

+ ++ + + + pPBS-Phe RNA templates were primed with

tRNAL2s

and

+ + + + + +

tRNAPhe,

respectively, full-length minus-strand strong-stop

+ + + + + _ + DNA (249 nt) was not detectable until after 15 min (Fig.

4,

K

__

:*'

* * lanes 7 and

11).

None of the tRNAs were able to

prime

minus-strand strong-stop DNA synthesis from a synthetic 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 161718 pPBS(-) RNA template (Fig. 4, lanes 4, 8, and 12), indicating the specificity of the 249-nt products generated. The faster-pecificity of PCR for differentiation of PBS se- movingbands (molecular size,<249nt)inFig. 4 are probably d as asenseprimer,whileLysl,2, Phe,andLys3 were due to pausing of DNA synthesis by RTduring DNA synthesis ,.PrimerpairsPS/Lysl,2, PS/Phe, and PS/Lys3were (19, 24).

PBS complementary to

tRNALy~2,

tRNAPhe, and Alteration of

HIV-1

PBS does not abolish the reverse ively. For this purpose, linearized PBS constructs transcription cycle. The retrovirus PBS is thought to facilitate Lysl,2,pPBS-Phe,andpPBS-WT]were mixed with the second

template

switch whichoccurs

dunn

reverse tran-;edcellular DNAandsubjected to PCR in separate

gcription.

tete fllt h

occurstd

rovi

ran-eabove-describedprimer pairs. The sensitivities of scription To detect full-length integrated proviral DNA by

pairs

weresimilar, andspecificity was present over a PCR,we studied HMW DNA of infected MT-4 cells by using concentrations.Theintensitiesof reaction products primer

pair

PS/PA

(Fig. 1),

which amplifiesthe lastregion of ngplasmid and an exposure time of 2 h at -70°C are reverse-transcribed proviral DNA (13). Figure 5 shows that D)PCRanalysisofPBS inreverse-transcribed DNA. full-length proviral DNA was detected throughout the course infectedcells at five time points wasanalyzed by PCR ofinfection by wild-type and mutated viruses. This was true

tsofprimer pairsdescribedabove (A). Panels B to even at early time points of infections (6 days) involving

im

MT-4cells infectedbyPBS-WT,PBS-Lysl,2,and mutatedviruses, when

only

mutated PBS forms were

present

tively; 50 ,ug of DNA was used in each reaction.

(Fig.

5).

isanintemnalcontrol (not shown). The

identity

of the PBS sequences of integrated proviral DNA was next confirmed by sequencing 367-bp fragments amplified by primer pair PS/PA (Fig. 1). The PBS sequences in PS/Lys1

,2

PS/Phe PS/Lys3

+ + + +

+ ++ +

+ + + +

Ah

Adel

on November 9, 2019 by guest

http://jvi.asm.org/

(6)

CCc_C C

C-time _{_}E E E EE f E EE C

mf _- - _n - - L

_ 310

o 281

271

_ 234

(-)ssDNA-pPBS-Lysl,2 pPBS-Phe pPBS-WT infection infection infection

Ir

11

1

Ln> r >> Entn 00>8tn tn En ns V\r 'I -aas tt VsVn c-s c; v nV

m Vm V V N m ' N L - m

HIV-1 LTR/gag

-9-globin

-FIG. 5. Detection of full-length integrated proviral DNA from infected cells. HMWDNA was analyzed by PCR using primer pair PS/PA, which detects full-length proviral DNA (13, 52). Freshcells _ 118 were added into parallel infection cultures to

facilitate

isolation of relatively intactHMW DNA(on day12afterpPBS-WT infectionand onday 18 after infections bypPBS-Lysl,2 and pPBS-Phe).Amplified productswereelectrophoresedon1.5% agarose gels and detected by staining the gels with ethidium bromide.Lane -, HMW DNA from MT-4 cells infected by heat-inactivated pPBS-WT; lane m, 100-bp DNAladder(Gibco-BRL).

1 2 3 4 5 6 7 8 9 10 1112

FIG. 4. Detection of minus-strandstrong-stopDNAsynthesis inan

invitro RTassay. Synthetic RNA templates containing altered PBS sequences wereprimed with

tRNAL",

tRNAPhe, and

tRNAVs,

respec-tively,inminus-strandstrong-stopDNAsynthesis reactions. Lanes: 1 to3, pPBS-WT RNA template primed by

tRNA3LYs;

4,pPBS(-) RNA template primed by

tRNA3L3";

5 to 7, pPBS-Lysl,2 RNA template

primed by

tRNAjYs;

8, pPBS(-)RNAtemplate primed by

tRNA,'r2;

9 to 11, pPBS-Phe RNA template primed by tRNAPhe; 12, pPBS(-)

RNA template primed by tRNAPhe. The full-length minus-strand

strong-stop DNA [(-) ssDNA] product is 249 nt long. Bands with molecular sizes smaller than 249ntareincomplete minus-strand DNA products, possibly duetopausing by RT during DNAsynthesis (19, 24).

HMW DNA after 6dayswereall of the mutatedvariety in cells

infected with themutantvirusPBS-Lysl,2or PBS-Phe.

How-ever, only wild-type PBS sequences were detected after 24 days.The PBSsequencesatintermediate times(e.g., day12 in PBS-Lysl,2-infectedcells andday9 inPBS-Phe-infectedcells)

were amixture of bothmutantsand wildtypes.

Identification of the tRNAspeciesinmutantviruses.Three differentoligonucleotide probes, complementarytothe 3' ends

oftRNAl,tRNAPhC,and

tRNA,

,respectively,wereusedto

detect these tRNA species in purified virus particles, using hybridizationconditions and purifiedhumanplacental tRNA isoacceptor species, as described previously (22, 30). These

analyseswerecarriedoutbydot blothybridization;allpatterns of reactivity were specific, and no cross-hybridization was

observed amongtheprobesused with anytRNAisoacceptor species.Forexample,thetRNAPhe-specific probedid not show cross-hybridization toeither

tRNAIL2S

or

tRNA3L3"

(Fig. 6A to C),which sharemorethan 60%homologyatthe 18-nt3' end. Nosignificantdifferences in levels oftRNAl , tRNAPhe, or

wild-type primer

tRNA3LYS

were found in each of the three mutantcomparedwithwild-typeviruses(Fig. 6AtoC). Thus,

replacement of the

wild-type

PBS with sequences complemen-tary to

tRNAIL2S

ortRNA edid not alter the pattern of tRNA

species incorporation into viral progeny(Fig. 6). Two-dimen-sional polyacrylamide gel electrophoresis analysis of tRNA species in these various virus particles did not reveal significant

differences among the wild type and pPBS(-), pPBS-Lysl,2, andpPBS-Phe mutants(datanotshown). Thus, the PBS does not appear to be involved in the selective incorporation of tRNAspecies into mature virions.

DISCUSSION

Reversetranscription of retroviral genomic RNA into pro-viral DNA isanearlyand essential step in the HIV life cycle. The role of the PBS in this process is to provide a complemen-tary region for the binding of the specifictRNAisoacceptor

species, which serves as a primer for RNA-dependent DNA

polymerization, and to facilitate the secondtemplate switch,

allowing completion of full-length double-stranded proviral DNA(12, 26, 28, 31, 33, 38, 43).

This study of PBS sequences was based on the utility of

tRNA as a primer of retroviral RT and the abundance of tRNA species in HIV-1 virions. We were unable to detect

significant differences among tRNA isoacceptors in terms of

packaging into mature virions, consistent with results that showed deletion of either long terminal repeat or PBS se-quences did notdisrupttRNApatterns inviruses

(22).

Viral

proteins, includingthe

Pr1606'""'

precursor,probably playan important roleinthe selection oftRNAisoacceptors(30).

Virusparticlesharvested fromtransfectedCOS-7 cellswere assessed forinfectivity, using MT-4,CEM-T4,and Jurkat cells as targets. Not surprisingly, virus produced from

pPBS(-)

transfectionswerenoninfectious, consistent with

(40).

A novel

finding

of this

study

is that virus

produced after transfection

by

the two PBS replacement

mutants,

pPBS-Lysl,2

and

pPBS-Phe,

were infectious, al-though lesssothanwild-typeviruses.

However,

despite

a

delay

inproductionof viralp24

(CA [capsid])

and

RT,

therate of

on November 9, 2019 by guest

http://jvi.asm.org/

(7)

A

1 2 3 4 5 6

::*

'3..3*

B

1 2 3 4

C

1 2 3 4

tRNALys1,2 tRNAPhe tRNALys3

FIG. 6. (AtoC) Identificationof tRNAspeciesin viralparticles bydotblotting usingDNAoligonucleotides complementarytothe 3' endof tRNALys(A), tRNAPhe (B),ortRNALYS(C).Thespecificityandhybridizationconditions of thisassayhave been described elsewhere(22, 30).RNA

sampleswereobtained frompurifiedvirusesproduced byCOS-7 cells after transfection withpPBSconstructs. Dots1 to 4designateviralRNA

from COS-7 cells transfected by pPBS-WT, pPBS(-), PBS-Phe, andPBS-Lysl,2, respectively. Dots5to7designate purifiedhumantRNALY,

tRNAPhC, andtRNALYS,respectively. (D) Comparisonof tRNAisoacceptors packagedinto viruses.Eachexperimentwasrepeatedthree times.The

relativeintensity of each dot fromthehybridizationsinpanelsAtoCwasestimatedbylaserscanninginanLKBfilm laserscanner.Resultsare

expressed ±standard deviation. *, pPBS-Phe; O, pPBS-Lysl,2; El, pPBS(-); E, pPBS-WT.

virusreplication bythesemutantseventuallyreachedwild-type levels. Mutant viruses derived from early and late stages of infection were used to reinfect MT4cells; we found that the

early-stage-derivedvirus continuedtolagininfectionkinetics, whilelate-stagevirus behavedindistinguishably from the wild type.As discussedbelow,this shift inphenotype corresponded to a reversion to a wild-type PBS. Other workers have also reported differential results with use of mutated PBS

se-quencesin adifferentsystem involving onlyasingle round of viralreplication (29).

During plus-strand DNA synthesis, tRNA serves as the templatefor thegeneration of the PBS,thus enabling identi-fication of the tRNA isoacceptor used as a primer. PCR

analysisatearly time pointsof the PBSsequencesfound within theHMW DNAof cellsinfectedby PBS-Lysl,2orpPBS-Phe

showed complementarity to

tRNA1,2

and tRNAPhe,

respec-tively. Incontrast, the PBS sequences ofproviralDNA from later stagesof these infections were complementaryto wild-type primer

tRNALYS,

indicating that synthesis and apparent selection ofwild-typeforms had occurred.Thus,both

tRNAlLY

and tRNAPheappeartoserve asprimers forreverse

transcrip-tionduring earlystagesofinfection. However,atlatestages,it islikelythat

tRNALYs

alonecan servethis role. Our data show

that the replication competence of HIV-1 clones is closely related to the status of the PBS. Ouruse ofsynthetic RNA

templatesandmutatedPBSincell-freeRTassaysshowed that

tRNA1LY

and tRNAPhe were inefficient primers of

minus-strandstrong-stopDNAsynthesis. Thismayexplain the lag in

virus production kinetics when the two PBS mutants were

studied inareplication-competent system.

Despite the ability of tRNA1LY and tRNAPhO to serve as

primers, both mutated PBS forms eventually reverted to

wild-type complementary to

tRNALYS.

It is unlikely that this development was due to contamination and amplification of smallquantitiesofwild-type forms, sinceourmolecular

provi-ral clones had been repeatedly subcloned and purified and

werepureby sequencing. Also, specific PCR showed thatonly mutant PBS forms were present in proviral DNA at early stages of infection by mutated viruses. No reversion of the

PBS- mutantwas noted,as might have been expected if the

reversions withPBS-Lysl,2andPBS-Pheweredueto contam-ination.

Although the mechanisms responsible for the observed reversion are uncertain, it is known that the HIV RT binds

preferentially to

tRNALYS

(4, 21, 37, 41)and that interaction between retroviral U5 RNA and the TTCloop of the tRNATrP

mayberequiredfor efficient initiation ofreversetranscription

(1, 2, 7, 8, 27).Recent studies have shown thata4-ntsequence

in the anticodon loop of tRNALYS interacts with HIV-1 genomic RNA in a region upstream from thePBS (20). The resulting loop-loopinteraction between tRNA and RNA tem-plate, combined with normal PBS-tRNAbinding, might give rise to significant alterations in secondary structure of the primer-template complexrelativetothatoccurringwhenonly the 18ntof the viral PBS interact with tRNA(asin thecaseof the pPBS-Lysl,2 and pPBS-Phe mutants). The stability of additional tRNA-RNAtemplateinteractionsmightbe depen-dentonparticularbase modifications foundonlyin

tRNALYS.

Such interactions couldplayarole in formation of RT-tRNA-RNAtemplate transcription complexes, thereby affecting tran-scription efficiency.Results ofourendogenous,invitroreverse

transcriptionreactionshowed that initiation withtRNALYSand

tRNAPiTe

occurred less efficiently than with

tRNA3Ys

when minus-strand strong-stopDNA synthesiswas primed froman

RNAtemplate containingaPBS complementarytothe

respec-tive tRNAisoacceptors (Fig. 4).

tRNALYs

hasextensive 3'-endhomologywithbothtRNALys

(71%)and tRNAPhe(62%) andwasincorporated intoourPBS

mutantvirusesatlevels similartothose found with wildtypes. Homology among these tRNA species implies that

tRNALYS

could conceivably anneal to a mutant PBS; such annealing mightbe furtherstabilizedby thetRNA-RNAtemplate inter-actions discussedabove. Thus,

tRNALYS

might be abletoprime

reversetranscription even froma mutant PBS. Sincereverse

transcription with

tRNALys

is more efficient than that with

tRNALys or tRNAPhe, it is conceivable that mutant viruses might preferentially use a wild-type primer, leading to the

reversetranscriptionfrom

tRNA3YS

ofawild-type PBS and a

D 10

7

8

5 6

is

7

0)

cc 6

4

2

o

S 6 7

56

on November 9, 2019 by guest

http://jvi.asm.org/

(8)

consequent increase in viral production. Other factors that may affect reversion of the PBS to wild type include specific

interactions between the HIV-1 RT and

tRNA3YS

and the

preferential incorporation of

tRNALYS

isoacceptors into

viri-ons. Since the PBSs of both proviral plus-strand strong-stop

DNAand full-length plus-strand DNA reflect the identity of thetRNAprimer, the use of

tRNALYS

as a primer for viruses

containinga mutated PBS might eventually result in reversion to a wild-type sequence. It is still unclear whether cellular

factors may be involved in reverse transcription (24, 52).

Furtherinsight will be possible once the factors involved in the

selection, incorporation, and placement of primer tRNA onto theHIV-1 PBS in vivo are better understood (23, 25, 30, 32).

ACKNOWLEDGMENTS

We aregrateful toC. Morrow, University of Alabama at Birming-ham,for a gift ofrecombinantwild-typep66/p5i heterodimeric HIV-1 RT. WethankFrancine Busschaert for assistance in preparation of the manuscript.

We thank Health and Welfare Canada, the Medical Research CouncilofCanada, and theAmerican Foundation for AIDS Research forgrantsupport to M.A.W. andM.A.P. X.L. was the recipient of a predoctoral studentship fromthe Medical Research Council of Can-ada.

REFERENCES

1. Aiyar, A., D. Cobrinik, Z. Ge, H. J. Kung, and J. Leis. 1992. InteractionbetweenretroviralU5 viral RNA and the T'C loop of thetRNATrP primer isrequiredfor efficient initiation of reverse transcription.J.Virol.66:2464-2472.

2. Aiyar, S., Z. Ge, and J. Leis. 1994. A specific orientation of RNA secondary structuresisrequired for initiation of reverse transcrip-tion. J. Virol. 68:611-618.

3. Arts, E., X. Li, Z. Gu, L. Kleiman, M. A. Parniak, and M. A. Wainberg. 1994. Comparison of deoxyoligonucleotide and tRNALYS3 as primers in an endogenous HIV-1 in vitro reverse transcription/template switching reaction. J. Biol. Chem. 269: 14672-14680.

4. Barat, C., V. Lullien, 0. Schatz, G. Keith, M. T. Nugeyre, F. Gruninger-Leitch, F.Barre-Sinoussi, S. F. J. Le Grice, and J. L. Darlix. 1989. HIV-1 reverse transcriptase specifically interacts with the anticodondomain of itscognateprimer tRNA. EMBO J. 8:3279-3285.

5. Boulerice, F., S. Bour, R. Geleziunas, A. Lvovich, and M. A. Wainberg. 1990. Highfrequencyofisolation of defective human immunodeficiencyvirus type 1 and heterogeneity of viral gene expressioninclones ofinfected U-937cells. J. Virol. 64:1745-1755. 6. Chomczynski,P.,and N.Sacchi. 1987. Single step method of RNA

isolationbyacidguanidinium thiocyanate-phenol-chloroform ex-traction.Ann.Biochem.162:150-159.

7. Cobrinik, D., A.Aiyar,Z. Ge, M.Katzman, H. Huang, and J.Leis. 1991.OverlappingU5sequence elements are required for efficient integration and initiation of reverse transcription. J. Virol. 65: 3864-3872.

8. Cobrinik, D., L. Soskey, and J. Leis. 1988. A retroviral RNA secondary structure required for efficient initiation of reverse transcription.J.Virol.62:3622-3630.

9. Coffin, J. M. 1990. Retroviridae and their replication, p. 1437-1500. In B. N. Fields et al. (ed.),Virology. Raven Press, New York. 10. Cordell, B., R Swanstrom,H. Goodman, and J. M. Bishop. 1979. tRNATrPasprimer forRNA-directedDNApolymerase: structural determinantsoffunction.J.Biol. Chem.254:1866-1874. 11. Dahlberg, J. E., R C. Sawyer, J. M. Taylor, A. J. Faras, W. E.

Levinson,H. M.Goodman,and J. M. Bishop. 1974. Transcription ofDNAfrom the70S RNA ofRoussarcomavirus.I. Identifica-tion of a specific 4S RNA which serves as primer. J. Virol. 13:1126-1133.

12. DeStephano, J. J., R A. Barbara, and P. J. Fay. 1994. The mechanism of HIV reversetranscriptase-catalyzed strand transfer frominternalregions ofheteropolymeric RNA templates. J. Biol. Chem. 269:161-168.

13. Gilboa, E., S. W. Mitra, S. P. Goff, and D. Baltimore. 1979. A detailed model of reverse transcription and tests of crucial aspects. Cell 18:93-100.

14. Goff, S, P. 1990. Retroviral reverse transccriptase: synthesis, structure, and function. J. Acquired Immune Defic. Syndr. 3:817-831.

15. Gottlinger, H. E., J. G. Sodroski, and W. A. Haseltine. 1989. Role of capsid precursor processing and myristoylation in morphogen-esis and infectivity of HIV-1. Proc. Natl. Acad. Sci. USA 86:5781-5785.

16. Gu, Z., Q. Gao, X. Li, M. A. Parniak, and M. A. Wainberg. 1992. Novel mutation in the human immunodeficiency virus type 1 reverse transcriptase gene that encodes cross-resistance to 2',3'-dideoxyinosine and 2',3'-dideoxycytidine. J. Virol.66:7128-7135. 17. Hu, J. C., and J. E. Dahlberg. 1983. Structural features required

for the binding oftRNATIP to avian myeloblastosis virus reverse transcriptase. Nucleic Acids Res. 11:4823-4833.

18. Hu, W.-S., and H. M. Temin. 1990. Retroviral recombination and reverse transcription. Science 250:1227-1233.

19. Huber, H. E., and C. C. Richardson. 1990. Processing the primer for plus strand DNA synthesis by human immunodeficiency virus 1 reverse transcriptase. J. Biol. Chem. 265:10565-10573. 20. Isel, C., R Marquete, G. Keith, C. Ehresmann, and B.

Ehres-mann. 1993. Modified nucleotides oftRNALPS`3 modulateprimer/ template loop-loop interaction in the initiation complex of HIV-1 reverse transcription. J. Biol. Chem. 268:25269-25272.

21. Jacques, P.S., B. M. Wohrl,K.J. Howard, and S. F. J. Le Grice. 1994. Modulation of HIV-1 reverse transcriptase function in "selectively deleted" p66/51 heterodimers. J. Biol. Chem. 269: 1388-1393.

22. Jiang, M., J. Mak, A. Ladha, E. Cohen, M. Klein, B. Rovinski, and L. Kleiman. 1993. Identification of tRNAs incorporated into wild-type and mutant human immunodeficiency virus type 1. J. Virol.67:3246-3253.

23. Khan, R, and D. Giedroc. 1992. Recombinant human immunode-ficiency virus type 1 nucleocapsid protein unwinds tRNA. J. Biol. Chem.267:6689-6695.

24. Klarmann, G. J., C. A. Schauber, and B. D. Preston. 1993. Template-directed pausing of DNA synthesis by HIV-1 reverse transcriptase during polymerization of HIV-1 sequences in vitro. J. Biol. Chem. 268:9793-9802.

25. Kohlsteadt,L.A., and T. A. Steitz. 1992. Reverse transcriptase of HIV can use either human tRNALYS3 or E. coli tRNAglU as a primer in an in vitro utilization assay. Proc.Natl. Acad. Sci. USA 89:4652-4656.

26. Lee, Y. M. H., and J. M. Coffin. 1991. Relationship of avian retrovirus DNA synthesis to integration in vitro. Mol. Cell. Biol. 11:1419-1430.

27. Leis, J., A. Ashok, and D.Cobrinik 1993. Regulation of initiation of reverse transcription of retroviruses, p. 33-47. In A. M. Skalka and S. P. Goff(ed.), Reverse transcriptase. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

28. Lobel, L. I., and S. P. Goff. 1985. Reverse transcription of retroviral genomes: mutations on the terminal repeat sequences. J. Virol.53:447-455.

29. Lund, A. H., M. Duch, J. Lovmand, P. Jorgensen, and F. S. Pedersen. 1993. Mutated primer binding sites interacting with different tRNAs allow efficient murine leukemia virus replication. J. Virol.67:7125-7130.

30. Mak, J., M. Jiang, M. A. Wainberg, M.-L. Hammarskjold, D. Rekosh, and L. Kleiman. 1994. Role of

Pr1609a9P`

in mediating the selective incorporation oftRNALYS into human immunodefi-ciency virus type 1 particles. J. Virol.68:2065-2072.

31. Murphy, J. E., and S. Goff. 1989. Construction and analysis of deletion mutations in the U5 region of Moloney murine leukemia virus: effects on RNA packaging and reverse transcription. J. Virol. 63:319-327.

32. Nagashunmugam, T., A. Velpandi, C. S. Goldsmith, S. R. Zaki, V. S. Kalyanaraman, and A. Srinivasan. 1992. Mutation in the primer binding site of the type 1 human immunodeficiency virus genome affects virus production and infectivity. Proc.Natl.Acad. Sci.USA89:4114-4118.

33. Olsen, J. C., C. Bova-Hill, D. P. Grandgenett, T. P. Quinn, J. P.

on November 9, 2019 by guest

http://jvi.asm.org/

(9)

Manfredi, and R. Swanstrom. 1990. Rearrangements in uninte-gratedretroviral DNA arecomplexand are the resultof multiple geneticdeterminants.J.Virol.64:5475-5484.

34. Omer,C.A.,andA.J. Faras. 1982.Mechanism of release of the avian retrovirus tRNAtrP primer molecule from viral DNAby ribonucleaseHduringreversetranscription. Cell30:797-805. 35. Panacino, P. S., H. A.Liang, E. J. Firpo, and E. A.Clarlk 1993.

T-cell activationinfluences initialDNAsynthesisofsimian immu-nodeficiency virus in resting T lymphocytes from macaques. J. Virol. 67:7008-7016.

36. Panganiban, A. T., and D. Fiore. 1988. Ordered interstrandand intrastrand DNA transfer during reverse transcription. Science 241:1064-1069.

37. Prats, A. C., L. Sarih, C. Gabus, S. Litvak, G. Keith, and J. L. Darlix. 1988.Smallfingerprotein of avian and murineretroviruses has nucleic acid annealing activity and positions the replication primertRNAontogenomicRNA.EMBO J. 7:1777-1783. 38. Pullen, K. A., and J. J.Champoux. 1990. Plus-strand origin for

humanimmunodeficiency virus type 1: implications for integra-tion.J.Virol.64:6274-6277.

39. Ratner, L., W. Haseltine, R. Patarca, K. J. Livak, B. Starcich, S. F. Josephs, E. R. Doran, J. A. Rafalski, E. A. Whitehorn, K. Baumeister, L.Ivanoff, S. R Petteway, Jr., M. L. Pearson, J. A. Lautenberger,T.S. Papas,J.Ghrayeb,N. T.Chang,R. C.Gallo, and F.Wong-Staal. 1985. Complete nucleotide sequence of the AIDS virus,HTLV-III. Nature(London) 313:277-284.

40. Rhim, H., J. Park, and C. D. Morrow. 1991. Deletions in the tRNALYSprimer-binding site of human immunodeficiency virus type1identify essential regions forreversetranscription.J. Virol. 65:4555-4564.

41. Richter-Cook, N. J., K. Howard, N. M. Cirino, B. M. Wohrl, and S. F.J. Le Grice. 1992.Interaction oftRNAlYS3 with multiple forms of humanimmunodeficiencyreversetranscriptase.J.Biol. Chem. 267:15952-15957.

42. Roe, B. A.1975.StudiesonhumantRNA. 1.Therapid, largescale isolation and partial fractionation of placenta and liver RNA. Nucleic Acids Res. 2:21-42.

43. Roth, M. J., P. L. Schwartzberg, and S. P. Gof.1989. Structureof the termini of DNA intermediates in theintegration of retroviral DNA: dependenceon INfunction and terminalDNA sequence. Cell 58:47-54.

44. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning:alaboratory manual, 2nd ed. Cold Spring Harbor Labo-ratoryPress,ColdSpring Harbor,N.Y.

45. Swanstrom, R., J. M.Bishop,andH. E. Varmus.1982. Structure of areplication intermediateinthesynthesis ofRoussarcomavirus DNAin vivo.J.Virol. 42:337-341.

46. Telesnitsky, A., and S. P. Goff. 1993.Strong-stopstrandtransfer duringreversetranscription,p.49-83.InA.M. Skalka and S.P. Goff(ed.),Reversetranscriptase. Cold Spring Harbor Laboratory Press,ColdSpring Harbor,N.Y.

47. Varmus,H.E., and R Swanstrom. 1985.Replication of retrovi-ruses, p.75-134.InR.Weiss,N.Teich,H.Varmus,and T.Coffin (ed.), Molecular biology oftumorviruses, 2nd ed.RNA tumor viruses,part2.Supplements and appendixes. Cold Spring Harbor Laboratory, ColdSpring Harbor,N.Y.

48. Wakefield, J.,H.Rihm,andC. Morrow.1994. Minimalsequence requirements of a functional human immunodeficiency virus 1 primer binding site.J.Virol. 68:1605-1612.

49. Waters,L. C.,and B. C. Mullin. 1977. Transfer tRNAinRNA tumorviruses.Prog.NucleicAcidRes.Mol. Biol.20:131-160. 50. Weiss, R, N. Teich, H. Varmus, and J. Coffin (ed.). 1985.

Molecularbiologyoftumorviruses, 2nd ed. RNA tumorviruses, parts 1 and 2. Cold Spring Harbor Laboratory, Cold Spring Harbor,N.Y.

51. Whitcomb, J. M., R Lumar, and S. H. Hughes.1990.Sequence of the circle junction of human immunodeficiency virus type 1: implications for reverse transcription and integration. J. Virol. 64:4903-4906.

52. Zack, J. A., S. J.Arrigo,S. R. Weitsman, A. S.Go, A. Haislip, and L.S. Y. Chen. 1990. HIV-1 entryinto quiescent primary lympho-cytes:molecularanalysis revealsalabile,latent viralstructure.Cell

61:213-222.