Proc. Natl. Acad. Sci. USAVol. 86, pp. 2448-2452, April 1989Medical Sciences

Molecular cloning of feline immunodeficiency virus(feline immunodeficiency virus proviral done/cross-hybridization/animal model/acquired immunodeficiency syndrome)

ROBERT A. OLMSTED*t, ANDREA K. BARNESt, JANET K. YAMAMOTO§, VANESSA M. HIRSCH*,ROBERT H. PURCELLt, AND PHILIP R. JOHNSON**Division of Molecular Virology and Immunology, Department of Microbiology, Georgetown University, Rockville, MD 20852; tLaboratory of InfectiousDiseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892; and §Department of Medicine,School of Veterinary Medicine, University of California, Davis, CA 95616

Contributed by Robert H. Purcell, December 30, 1988

ABSTRACT Feline immunodeficiency virus (FIV) is aT-lymphotropic retrovirus associated with immunodeficiencyand opportunistic infections in cats. The discovery of FIVprovides an opportunity for the development of a small animalmodel for AIDS. To initiate the molecular and biologicalcharacterization of FIV, cDNA clones were synthesized andused to isolate a proviral clone of FIV. Molecular cross-hybridization analysis of FIV with five lentiviruses revealedthat nucleotide-sequence similarities exist between FIV andthese lentiviruses in the gag-pol genes. However, nucleotide-sequence similarities were not seen upon comparison of the FIVlong terminal repeat sequence with known viral sequences.Common antigenic determinants appeared to be shared byFIV, caprine arthritis encephalitis virus, and visna virus asshown by serological cross-reactivity of rabbit antibodies tocaprine arthritis encephalitis virus and visna virus with theputative FIV core protein p28. These studies demonstrated thatFIV is a member of the lentivirus subfamily and is distantlyrelated to the AIDS lentiviruses of primates. Importantly,progeny virions of our molecular clone were infectious forexperimentally inoculated cats. The availability ofan infectiousmolecular clone will make possible a detailed dissection of themolecular pathogenesis of FIV, which may facilitate the devel-opment of vaccine and therapeutic strategies for AIDS.

Feline immunodeficiency virus (FIV) was isolated in 1986from a feline leukemia virus (FeLV) negative cat with chronicopportunistic infections (1). FIV was initially classified as alentivirus on the basis ofvirion morphology, Mg2"-dependentreverse transcriptase activity, and the propensity for persis-tent infection in cats (1, 2). In addition, FIV shares twoimportant features with the AIDS lentiviruses, human im-munodeficiency virus (HIV), and simian immunodeficiencyvirus (SIV): a tropism for T lymphocytes and the associationwith an immunodeficiency syndrome in the infected host (1).Seroepidemiologic studies indicate that FIV infection occursthroughout the world. Seropositivity rates of 14%, 19%, and30% in the United States, Canada, and Japan, respectively,were found in sick cats, compared with a 1.2% rate in healthycats (3, 4). Although FeLV is the best characterized agent offeline AIDS (5), these data suggest that FIV may be asignificant cause of immunodeficiency in cats.Relevant animal models of AIDS-like infection and disease

are required for the development of safe and effectivestrategies for the prevention and therapy of HIV infection inhumans. The discovery of FIV presents an excellent oppor-tunity for the development of a small animal model for AIDS.We have undertaken molecular cloning and characterizationof FIV and describe in this report (i) the derivation andcharacterization of a biologically active molecular clone of

FIV that is infectious for experimentally inoculated cats, (ii)the cross-hybridization of FIV genomic probes to proviral-clone DNA of five lentiviruses, (iii) the immunologicalcross-reactivity patterns of rabbit anti-lentiviral antibodies toFIV, and (iv) the nucleotide sequence of the long terminalrepeats (LTR) of FIV. These molecular and immunologicaldata provide conclusive evidence for the inclusion of FIV inthe lentivirus family.

MATERIALS AND METHODSVirus and Cell Lines. The FIV-Petaluma strain was prop-

agated in mitogen-stimulated peripheral blood mononuclearcells (PBMCs) from specific pathogen-free (SPF) cats, orCrandell feline kidney (CRFK) cells, as previously described(1, 2). To obtain a large amount of stock virus (designatedFIV/PBMC-4), SPF cat PBMCs were infected with a seedstock of FIV-Petaluma strain; this seed stock had beenpropagated exclusively in cat PBMCs. At the peak of reversetranscriptase activity, culture supernatants were clarified,filtered (0.45 ,um), and stored in aliquots under liquid nitro-gen. CRFK cells (American Type Culture Collection) wereinfected with FIV/PBMC-4 to establish a low passage,persistently infected cell line (designated CR/FIV-4). Thiscell line was the source of FIV for large-scale purification ofvirion genomic RNA and infected cell DNA and RNA.FIV cDNA Synthesis and Cloning. FIV was sucrose-

gradient purified (1) from culture fluids obtained from CR/FIV-4 cells. Viral RNA was purified by phenol/chloroformextraction and ethanol precipitation and used as the templatefor oligo(dT)-primed cDNA synthesis (6). Spin column-purified cDNAs were ligated into the EcoRI site of theplasmid vector pT7T318U (Pharmacia). Transformants con-taining viral-specific cDNAs were detected by differentialhybridization with radiolabeled single-stranded cDNAs ofmRNAs from uninfected and FIV-infected CRFK cells.

Molecular Cloning of FIV Proviral DNA. Genomic DNAwas isolated from CR/FIV-4 cells (passage 4), partiallydigested with Mbo I, and size-selected by sucrose-gradientcentrifugation (7). The 15- to 20-kilobase (kb) DNA fractionswere pooled and ligated into BamHI-digested bacteriophageA DASH vector (Stratagene). Recombinant phage (2 X 106)were transferred to duplicate nitrocellulose filters andscreened with radiolabeled EcoRI fragments of FIV-cDNA

Abbreviations: FIV, feline immunodeficiency virus; FeLV, felineleukemia virus; HIV-1, human immunodeficiency virus type 1;CAEV, caprine arthritis encephalitis virus; EIAV, equine infectiousanemia virus; BIV, bovine immunodeficiency virus; SIV, simianimmunodeficiency virus; LTR, long terminal repeat; SPF, specif-ic pathogen free; PBMC, peripheral blood mononuclear cell.tTo whom reprint requests should be addressed at: National Insti-tutes of Health/Twinbrook II, 12441 Parklawn Drive, Rockville,MD 20852.IThe sequence reported in this paper is being deposited in theEMBL/GenBank data base (accession no. J04541).

2448

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

June

22,

202

1

Proc. Natl. Acad. Sci. USA 86 (1989) 2449

clones pE20 and pD32. The library was screened underhigh-stringency hybridization conditions [50% formamide/1o dextran sulfate at 42°C, washed at 55°C in 0.2x SSC (1xSSC = 0.15 sodium chloride/0.015 M sodium citrate, pH 7),0.1% NaDodSO4]. Filters were exposed to x-ray film with anintensifying screen for 20-24 hr at -70°C.

Transfection. Lambda clone FIV-14 DNA (0.5 ug; seebelow) was introduced into CRFK cells by the DEAE-dextran transfection method (8). After transfection, thepresence of replicating FIV was monitored by reverse tran-scriptase activity in culture supernatant fluids (9).

Nucleotide Sequence Analysis. FIV cDNA clones, pE20 andpD32, and selected plasmid subclones of the proviral cloneFIV-14 were sequenced by the dideoxynucleotide chain-termination method (10) with T7 DNA polymerase accordingto the manufacturer's protocol (United States Biochemical).Proviral DNA sequences were confirmed on both strands ofthe plasmid subclones. DNA sequence analysis was per-formed with the programs PC/Gene, FASTA-Mail, and IFIND(IntelliGenetics).

RESULTSMolecular Cloning of FIV. We synthesized and isolated

putative FIV cDNA clones by the strategy described. Tworecombinant plasmids, pE20 and pD32, both containingcDNA inserts of =400 base pairs (bp), were chosen forauthentication of FIV specificity. A Southern blot of FIV-infected and uninfected CRFK genomic DNA (Fig. LA)demonstrated hybridization of E20 cDNA (or D32 cDNA,data not shown) exclusively to FIV-infected cell DNA.Digestion ofinfected cellDNA with Nhe I generated two viralfragments that had a combined length of9-9.5 kb. These dataindicated that Nhe I cut three times within the proviralgenome: once in each LTR and once internally. Northern blot

A

c.)

m cm I wUl I UI I I I

analysis of total and poly(A)+ RNA isolated from FIV-infected CRFK cells (Fig. 1B) revealed that E20 hybridizedto three viral-specific RNAs of =9, 4.5, and 2 kb. TheseRNAs were similar in size to those viral RNA speciesdetected in HIV-1-infected cells (11, 12). Thus, FIV post-transcriptional splicing events may resemble those of otherlentiviruses.A recombinant A phage library constructed from genomic

DNA isolated from CR/FIV-4 cells was screened with cDNAclones E20 and D32. Fifteen recombinant clones were iden-tified, and eight were analyzed by Nhe I digestion (data notshown). Recombinant proviral clone FIV-14 appeared tocontain the full-length genome and was selected for furthercharacterization. A partial restriction endonuclease site mapof FIV-14 (Fig. 2D) revealed that the overall length of theproviral genome was =9.4 kb, a figure consistent withestimates generated by Southern blot analyses (see above).Limited nucleotide sequence analysis demonstrated thatcDNA clone E20 was derived from the 3' LTR and that theD32 cDNA originated just downstream of the internal Nhe Isite.

Biological Activity of Clone FIV-14. We tested the in vitrobiological activity of FIV-14 A DNA by transfection intoCRFK cells. A productive infection was established by 3days posttransfection as demonstrated by the presence ofMg2+- dependent reverse transcriptase activity in the culturefluids. Also, intracellular FIV antigens were detected at 3days by indirect immunofluorescence (data not shown).Culture supernatants from FIV-14-transfected CRFK cellswere filtered and used to infect SPF cat PBMCs. Reversetranscriptase activity was detected 4 days after infection,confirming that progeny virus of FIV-14 was infectious forcat PBMCs (data not shown).

Preliminary experiments have demonstrated that virusproduced by our biologically active molecular clone of FIV

B

.c:_C c

z D

I UI I

Total Polv A+

Ul I I

23 -

9.4- 1 1

6.7- 40

4.4-

9.5 -7.5-

4.4-

2.4-

1.4

40

2.3-2.0-

FIG. 1. Southern and Northern (RNA) blot analyses of genomic DNA and RNA from FIV-infected CRFK cells. (A) For Southern blotanalyses, genomic DNA was isolated from CR/FIV-4 cells (I) or uninfected CRFK cells (UI). DNA (10 ,ug) was digested with the restrictionendonuclease indicated above each lane, electrophoresed, and transferred to duplicate nitrocellulose filters. The blots were probed with theradiolabeled EcoRI fragments of cDNA clones pE20 under high-stringency conditions. Lambda HindIII fragments were used as size markers(kb) and are shown at the left. (B) For Northern blot analyses, total RNA or poly(A)+ RNA from either CR/FIV-4 cells (I) or uninfected (Ul)CRFK cells were electrophoresed on a formaldehyde-agarose gel, transferred to nitrocellulose, and probed with E20 under the conditionsdescribed above. Size estimates (kb) were determined by standard RNA markers (Bethesda Research Laboratories) run in parallel and are shownat the left.

Medical Sciences: Olmsted et al.

Dow

nloa

ded

by g

uest

on

June

22,

202

1

2450 Medical Sciences: Olmsted et al.

BA

khf[

3.0-

1.5 -

D

C

RI B N B K Bg, , b ~~~~~~~~~~~~~~fIV-14-or-I- -i if XE- a -00 f; |pm

LkbO0

,, I1 2 3 4 5 6 7 8 9 10

FIG. 2. Cross-hybridization of FIV to other retroviruses. Plasmid DNA (1 ,ug) containing the complete or nearly complete proviral genomesof FIV-14, FeLV (13), HIV-1 (14), SIV isolate from sooty mangabey monkeys (SIVsin) (15), visna virus (16), equine infectious anemia virus(EIAV) (17), or caprine arthritis encephalitis virus (CAEV) (18) were digested with the restriction endonucleases Nhe I, BssHII, Sac I, Sst I,Sst I, BamHI, or HindIII, respectively, and then separated on a 0.9%o agarose gel (A). Duplicate filters of the gel shown in A were preparedand hybridized under low-stringency conditions (B) (30%o formamide, 370C, washed in 4x SSC/0.1% NaDodSO4 at 500C) or high-stringencyconditions (C) (50%o formamide, 370C, washed in 0.2x SSC/0.1% NaDodSO4 at 600C) with equimolar amounts of the radiolabeled FIV 6-kb and3-kb Nhe I fragments (D, shaded bars). The DNA fragments containing the genomes of FeLV, SIVsmm, visna virus, and CAEV migrated at -9kb or greater. The HIV-1 genome is contained in the two comigrating DNA fragments at 5.5 kb. BamHI digestion of the EIAV plasmid exciseda 5-kb fragment that contained the complete gag-pol region of the genome. The 3- and 6-kb Nhe I fragments are estimated to contain >95%of the FIV-14 proviral genome (D). (D) Schematic representation of a partial restriction endonuclease site map of the FIV-14 proviral genome.The map was generated by Southern blot analyses of FIV-14 using probes derived from the cDNA clones pE20 and pD32. Open boxes, proviralLTRs. N, Nhe I; B, BamHI; Bg, BgI II; K, Kpn I; RI, EcoRI; S, Sac I; X, Xba I; Sp, Sph I. Shaded bars, 3- and 6-kb Nhe I fragments usedas probes in B and C.

is infectious for cats. FIV-14 virus established an infection intwo experimentally inoculated SPF cats. FIV was recoveredfrom cultures ofPBMCs from both inoculated cats at 3 weekspostinoculation (data not shown).

Sequence Similarities Between FIV and Other Lentiviruses.Cross-hybridization analyses were performed to detect se-quence similarities between FIV and six retroviruses (Fig. 2).The FIV genome hybridized at low stringency to all of thelentiviral genomes examined (Fig. 2B). Importantly, cross-hybridization with the oncornavirus, FeLV, was not de-tected. We predicted that the cross-hybridizing sequenceswere contained within the gag-pol genes based on thefollowing data: (i) FIV hybridized to the single EIAV 5-kbBamHI fragment that contains the complete gag-pol genes ofEIAV (17), and (ii) nucleotide sequence alignments of the 5'1.5 kb of the FIV-14 pol gene with the GenBank viralsequence library revealed significant nucleotide sequenceidentity to the po1 genes of SIV, HIV-2, HIV-1, EIAV, visnavirus, and CAEV (55-65% identity, data not shown). Thesedata indicated significant genetic relatedness between FIVand the five other lentiviruses examined.

Detection of Cross-Reactive Antibodies to FIV. Recently,rabbit polyclonal antibodies to bovine immunodeficiencyvirus (BIV) and HIV-1 were seen to react reciprocally withthe core proteins (p26 and p24, respectively) of either virus(19). Rabbit anti-HIV-1 antibodies also recognized p24 ofCAEV. We therefore examined the reactivity of rabbit anti-sera against CAEV, visna virus, BIV, and HIV-1 with FIVpolypeptides by immunoblot analysis. As shown in Fig. 3,rabbit polyclonal antibodies against CAEV and visna virus

bound to FIV p28. Specific reactivity was noted for anotherpolypeptide (15 kDa), but this protein was not detected byfeline anti-FIV antibodies (Fig. 3, lanes b and c). Based on sizesimilarities, FIV p28 may be the counterpart to the major coreprotein of other lentiviruses (2). Rabbit antibodies againstHIV-1 or BIV did not react with any FIV-specific polypep-tides. This immunologic cross-reactivity indicated that FIV,CAEV, and visna virus core proteins share some commonantigenic determinants.

Sequence Analysis of the FIV LTR. The complete nucleo-tide sequence (355 bp) of the proviral FIV LTR is presentedin Fig. 4. The LTR length was similar to those reported forCAEV, visna virus, and EIAV, but was considerably shorterthan the LTRs of H4IV-1 or SIV (17, 20-24). The boundariesofthe LTRs were demarcated by (i) the presence ofthe highlyconserved dinucleotide inverted repeats (5' TG and CA 3') atthe ends of the LTR and (ii) a 5-bp direct repeat (CTTAC) inthe 5'- and 3'-flanking cellular DNA (25).A purine-rich region, believed to be the site for initiation of

plus-strand DNA synthesis (26), was identified immediatelyupstream of the 3' LTR (Fig. 4, polypurine tract). Theprimer-binding site for initiation of minus-strand DNA syn-thesis was identified three bases downstream from nucleotide355 in the 5' LTR (data not shown). The sequence wascomplementary to the 3'-terminal 21 nucleotides of mamma-lian tRNAYS (27) and was similar to the primer binding sitesdescribed for HIV-1 and EIAV (17, 23).

Putative transcription signals (Fig. 4) were identified byanalogy to previously defined lentiviral LTR signal se-quences. Interestingly, two potential TATA boxes, usually of

NsLp...

1:1.

Proc. Natl. Acad. Sci. USA 86 (1989)

Dow

nloa

ded

by g

uest

on

June

22,

202

1

Proc. Natl. Acad. Sci. USA 86 (1989) 2451

i k Im no p

100-I

62-

54- _

40-

30-

28- O -P

17- -

A

FIG. 3. Immunoblot analysis of cross-reactivity of lentiviruspolyclonal antisera with FIV polypeptides. Purified FIV was frac-tionated by 8% NaDodSO4/PAGE and transferred to nitrocellulose(2). Immunoblot strips were incubated with serum samples (1:200dilution) and processed for detection of bound antibodies as de-scribed (2). Lanes: a, normal SPF cat serum; b and c, FIV-infectedSPF cat sera; d and e, rabbit preimmune CAEV sera; f and g, rabbitpostimmune CAEV sera; h and i, rabbit preimmune visna virus sera;j and k, rabbit postimmune visna virus sera; and m, rabbitpreimmune BIV sera; n and o, rabbit postimmune BIV sera; p, rabbitpostimmune HIV-1 sera. Arrows indicate reactivity with FIV p28and p15. The positions of the molecular mass markers (in kDa) aregiven at the left. The smear detected near 70 kDa in lanes c, f, g, j,k, n, and o probably represents reactivity to bovine serum albumin.Reactivity to the 17-kDa polypeptide was detected with several oftherabbit pre- and postimmune serum samples and therefore wasconsidered to be nonspecific. The protein species detected near 100kDa in lane c may represent the extracellular envelope glycoproteinof FIV (R.A.O., V.M.H., and P.R.J., unpublished observations).The paired rabbit sera (pre- and postimmunization) for CAEV, visnavirus, BIV, and HIV-1 (postimmunization only) were provided byMatthew A. Gonda (National Cancer Institute-Frederick CancerResearch Facility, Frederick, MD) (19).

the form TAT(W)XG (where W = A or T and x = 3 or 4) (26),were identified at base positions 164-169 and 187-193. Thefirst TATA box was similar to the signal described for HIV-1,TATAAG (11, 23). The second TATA box was nearlyidentical to that reported for CAEV, visna virus, and EIAV(TATATAAC vs. TATATAAG) (17, 20-22).

A consensus polyadenylylation signal spanned nucleotides269-274 (28, 29). Retroviral (and many eukaryotic mRNAs)are polyadenylylated at a site located 10 to 20 bases 3' of thepolyadenylylation signal (28, 29), which delineates the R-U5junction. The dinucleotide CA has been reported to be apreferred site for transcription termination (29). However,the dinucleotide GA also can be used (29) and was designatedas the poly(A) site for CAEV (20). We identified the putativesite for poly(A) addition in the FIV LTR at the dinucleotideGA (bases 286 and 287).

Direct repeat structures identified in the U3 regions ofCAEV and EIAV were not detected in the U3 region of theFIV LTR. A consensus "core" enhancer sequence (GTG-GTTTG or GTGGAAAG) also was not detected (30). How-ever, a potential enhancer sequence located from nucleotide51 through 59 is similar to the sequence GGGACTTTCC thatis contained in the B site within the K immunoglobulin geneenhancer (31). This site has been observed in the U3 regionsof HIV-1 and SIV (24, 31), as well as in the enhancersequences of simian virus 40 and cytomegalovirus (31).The precise lengths of the U3, R, and U5 regions in the

LTR have not been determined. Primer extension of FIVgenomic RNA will be required to identify the RNA initia-tion/cap site and the R/U5 boundary.Attempts to align the FIV LTR nucleotide sequence with

other lentiviral LTR sequences were not successful. Further-more, a global search of GenBank viral sequences did notreveal any significant similarities between the FIV LTR andany reported viral sequences.

DISCUSSIONWe have described the derivation of an infectious molecularclone of FIV and provided molecular and immunologicalevidence that conclusively demonstrates genetic relatednessbetween FIV and other lentiviruses. The size of the full-length genome (-9.4 kb) suggests that FIV may have acomplex genetic structure similar to other lentiviruses. De-termination of the complete nucleotide sequence and studiesof molecular function will be required to address this issue.

Cross-hybridization of FIV genomic probes to proviralDNA of HIV-1, SIVsmm, visna virus, EIAV, and CAEVdemonstrated that FIV was significantly related to otherlentiviruses. The specificity of this assay was illustrated bythe failure to detect cross-reactivity to FeLV, a type Cretrovirus of cats. The cross-hybridization patterns sug-gested that the gag-pol genes contained the sequences ofgreatest similarity. In addition, a search of the GenBank database for sequences related to the FIV pol gene (5' half)

Polypurine tract , *U3

Enhancer elementTi ACG ACGAACAAATGATAAAAGGA

ATGCTTATGGAC 48

LAATAiCTGALGCATGACTCATAGTTAAAGCGCTAGCAG 117

CTGCCTAACCGCAAAACCACATCCTATGGAAAGCTTGCTAATGACGj ~rTGTTCCATTGTAAGAG 186* ~~~~~~~~*

t X..IiCAGTGCTTTGTGAAACTTCGAGGAGTCTCTTTGTTGAGGACTTTTGAGTTCTCCCTTGAGG 255Poly A R U5

CTCCCACAGATAC A~rATTTGAGATTGAACCCTGTCGAGTATCTGTGTAATCTTTTTTACCTGTG 324

AGGTCTCGGAATCCGGGCCGAGAACTTCGCA 355IR

FIG. 4. FIV LTR nucleotide sequence. The FIV LTRs and flanking cellular DNA were sequenced as described. U3 designates base number1 of the LTR. Two putative TATA boxes are shaded. Asterisks above bases 184 and 216 designate two potential RNA initiation/cap sitesdownstream from their respective TATA box and would represent the 5' end of R. IR, inverted repeat; Poly A, polyadenylylation signal.

a bc d e f g h

Medical Sciences: Olmsted et al.

Dow

nloa

ded

by g

uest

on

June

22,

202

1

2452 Medical Sciences: Olmsted et al.

revealed significant similarity scores only for the pol genes ofEIAV, HIV-1, CAEV, visna virus, HIV-2, and SIV.

Previous studies demonstrated that natural host-derivedserum antibodies to CAEV, visna virus, HIV-1, or SIV didnot cross-react with FIV antigens in immunoblot assays (1,2). However, serologic cross-reactivity was observed be-tween FIV and both CAEV and visna virus when specificrabbit antisera to these two viruses were assayed. Polyclonalrabbit antibodies to either of these viruses bound specificallyto the FIV p28 polypeptide, which may represent the FIVmajor core protein. The results of these molecular and sero-logic analyses are consistent with the observed conservationof gag-pol genes across the lentivirus subfamily (20, 32, 33).The inability to align the FIV LTR nucleotide sequence

with other lentiviral LTR sequences was not surprisingbecause LTR sequences are not highly conserved amonglentiviruses in general (17, 20). An unusual feature in the FIVLTR was the presence of two potential TATA boxes.Site-directed mutagenesis of putative transcription and reg-ulatory signals in the LTR of our infectious FIV molecularclone should clarify their respective functions during viralreplication.

Importantly, the proviral molecular clone, FIV-14, wasbiologically active upon transfection into CRFK cells. Prog-eny virus ofFIV-14 infected feline PBMCs in culture and wasshown to infect two experimentally inoculated SPF cats.Thus, with the availability ofan infectious molecular clone ofFIV, informative studies designed to define the moleculardeterminants of FIV pathogenesis and to develop vaccinestrategies in the feline model system are now feasible.

We acknowledge the excellent technical assistance provided by G.Dapolito, C. McGann, and S. Kitov and we thank Drs. RobertChanock and John Gerin for their continued support. We also thankDr. Neils Pederson for providing the FIV-Petaluma strain and Drs.Janice Clements and Fred Fuller for the plasmid clones ofCAEV andvisna virus and for EIAV, respectively.

1. Pederson, N. C., Ho, E. W., Brown, M. L. & Yamamoto,J. K. (1987) Science 235, 790-793.

2. Yamamoto, J. K., Spager, E., Ho, E. W., Anderson, P. R.,O'Connor, T. P., Mandell, C. P., Lowenstine, L., Munn, R. &Pederson, N. C. (1988) Am. J. Vet. Res. 49, 1246-1258.

3. Yamamoto, J. K., Pederson, N. C., Ho, E. W., Okuda, T. &Thelen, G. H. (1988) Leukemia 2 (Supplement), 2045-2155.

4. Yamamoto, J. K., Hansen, H., Ho, E. W., Morishita, T. Y.,Okuda, T., Sawa, T. R., Nakamura, R. N. & Pederson, N. C.(1989) J. Am. Vet. Med. Assoc. 194, 213-220.

5. Hoover, E. A., Mullins, J. I., Quackenbush, S. L. & Gasper,P. W. (1986) in Animal Models of Retrovirus Infection andTheir Relationships to AIDS, ed. Salzman, L. A. (Academic,Orlando, FL), pp. 59-74.

6. Gubler, U. & Hoffman, B. J. (1983) Gene 25, 263-269.7. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1981) Molecular

Cloning:A Laboratory Manual (Cold Spring Harbor Lab., ColdSpring Harbor, NY), pp. 269-307.

8. Davis, L. G., Dibner, M. D. & Battey, J. F. (1986) BasicMethods in Molecular Biology (Elsevier, New York), pp. 290-292.

9. Willey, R. L., Smith, D. H., Lasky, L. A., Theodore, T. S.,Earl, P. L., Moss, B., Capon, D. J. & Martin, M. A. (1988) J.Virol. 62, 139-147.

10. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl.Acad. Sci. USA 74, 5463-5467.

11. Muesing, M. A., Smith, D. H., Cabrodilla, C. D., Benton,C. V., Lasky, L. A. & Capon, D. J. (1985) Nature (London)313, 450-458.

12. Feinberg, M. B., Jarrett, R. F., Aldovini, A., Gallo, R. C. &Wong-Staal, F. (1986) Cell 46, 807-817.

13. Reidel, N., Hoover, E. A., Gasper, P. W., Nicolson, M. O. &Mullins, J. I. (1986) J. Virol. 60, 242-250.

14. Fisher, A. G., Collalti, E., Ratner, L. Gallo, R. C. & Wong-Staal, F. (1985) Nature (London) 316, 262-265.

15. Hirsch, V. M., Dapolito, G., McGann, C., Olmsted, R. A.,Purcell, R. H. & Johnson, P. R. (1989) J. Primatol., in press.

16. Molineaux, S. & Clements, J. E. (1983) Gene 23, 137-148.17. Kawakami, T., Sherman, L., Dahlberg, J., Gazit, A., Yaniv,

A., Tronick, S. R. & Aaronson, S. A. (1987) Virology 158, 300-312.

18. Pyper, J. M., Clements, J. E., Gonda, M. A. & Narayan, O.(1986) J. Virol. 58, 665-670.

19. Gonda, M. A., Braun, M. J., Carter, S. G., Kost, T. A., Bess,J. W., Jr., Authur, L. 0. & Van Der Maaten, M. J. (1987)Nature (London) 330, 388-391.

20. Hess, J. L. Pyper, J. M. & Clements, J. E. (1986) J. Virol. 60,385-393.

21. Sonigo, P., Alizon, M., Staskus, K., Klatzman, D., Cole, S.,Danos, O., Retzel, E., Tiollais, P., Haase, A. & Wain-Hobson,S. (1985) Cell 42, 369-382.

22. Derse, D., Dorm, P. L., Levy, L., Stephens, R. M., Rice, N. R.& Casey, J. W. (1987) J. Virol. 61, 743-747.

23. Starich, B., Ratner, L., Josephs, S. F., Okamoto, R., Gallo,R. C. & Wong-Staal, F. (1985) Science 227, 538-540.

24. Hirsch, V., Reidel, N. & Mullins, J. I. (1987) Cell 49, 307-319.25. Temin, H. M. (1981) Cell 27, 1-3.26. Coffin, J. (1984) in RNA Tumor Viruses, Molecular Biology of

Tumor Viruses, eds. Weiss, R., Teich, N., Varmus, H. &Coffin, J. (Cold Spring Harbor Lab., Cold Spring Harbor, NY),2nd Ed., pp. 262-345.

27. Raba, M., Limberg, K., Burghagen, M., Katzl, J. R., Simsek,M., Heckman, J. E., Rajbhandary, U. L. & Gross, H. J. (1979)Eur. J. Biochem. 97, 305-318.

28. Birnsteil, M. L., Busslinger, M. & Strub, K. (1985) Cell 41,349-359.

29. Fitzgerald, M. & Shenk, T. (1981) Cell 24, 251-260.30. Weiher, H., Konig, M. & Gruss, P. (1983) Science 219, 626-

631.31. Sen, R. & Baltimore, D. (1986) Cell 47, 921-928.32. Stephens, R. M., Casey, J. W. & Rice, N. R. (1986) Science

231, 589-594.33. Gonda, M. A., Braun, M. J., Clements, J. E., Pyper, J. M.,

Wong-Staal, F., Gallo, R. C. & Gilden, R. V. (1986) Proc. Natl.Acad. Sci. USA 83, 4007-4011.

Proc. Natl. Acad. Sci. USA 86 (1989)

Dow

nloa

ded

by g

uest

on

June

22,

202

1