Long-Term Infection with Retroviral Structural Gene Vector Provides Protection Against Bovine Leukemia Virus Disease in Rabbits

Virology 329 (2004) 434–439 www.elsevier.com/locate/yviro

Long-term infection with retroviral structural gene vector provides protection against bovine leukemia virus disease in rabbits

Veronika Altanerovaa,b, Dana Holicovaa, Lucia Kucerovaa, Cestmir Altanera,b, Michael D. Lairmorec,d,e,f, Kathleen Boris-Lawriec,d,e,f,*

aCancer Research Institute, Slovak Academy of Sciences, SK-833 91 Bratislava, Slovakia bCentre of Excellence of SAS Bratislava Molecular Medicine, Slovak Academy of Sciences, SK-833 91 Bratislava, Slovakia cCenter for Retrovirus Research, The Ohio State University, Columbus, OH, 43210-1093, USA dDepartment of Veterinary Biosciences, The Ohio State University, Columbus, OH, 43210-1093, USA eMolecular Virology, Immunology and Medical Genetics, The Ohio State University, Columbus, OH, 43210-1093, USA fComprehensive Cancer Center, The Ohio State University, Columbus, OH, 43210-1093, USA

Received 7 May 2004; returned to author for revision 6 August 2004; accepted 3 September 2004 Available online 3 October 2004

Abstract

Bovine leukemia virus (BLV) infection of rabbits is a tractable model system to evaluate vaccination strategies against lymphotropic retroviruses, which represent a global human health problem. We have previously developed genetically simplified BLV structural gene vector (SGV) that replicates BLV structural and enzymatic genes independently of BLV regulatory and accessory genes. Results of a 20- month study in a rabbit model demonstrated that BLV SGV induces an antiviral immunological response and lacks pathogenicity. Here, these chronically infected-BLV SGV rabbits are assessed in a proof-of-principle study of preventative vaccination against challenge with pathogenic BLV. This study commences 24 months after BLV SGV inoculation and proceeds for an additional 20 months. The previously characterized BLV SGV rabbits and age-matched control rabbits were challenged with 1 Â 108 fetal lamb kidney/BLV producer cells. BLV SGV rabbits seroconverted upon BLV challenge, but did not progress to BLV infection nor clinical disease. By contrast, naive rabbits became infected and succumbed to lymphotropic disease. Our findings provide proof-of-principle that chronic infection with BLV SGV induces protection against BLV infection. The data indicate that SGV based on HTLV or HIV is a promising approach against lymphotropic disease by human retroviruses. D 2004 Elsevier Inc. All rights reserved.

Keywords: Replication-competent retroviral vector; Live attenuated vaccine; Rex-independent gene expression

Introduction addition to the Gag, Pol, and Env structural and enzymatic proteins that are encoded by both simple and complex Bovine leukemia virus is a complex oncogenic retrovirus retroviruses. Progression to neoplasia by BLV and HTLV is that is genetically related to human T-cell leukemia virus type associated with the virus-encoded regulatory and accessory 1 (HTLV-1) (Sagata et al., 1984). The genomes of these proteins, which induce deregulation of lymphocyte homeo- lymphotropic viruses and human immunodeficiency virus stasis. Tax regulatory protein is an essential transcriptional (HIV-1) encode essential regulatory and accessory proteins in transactivator that functions through the Tax responsive element (TRE) in the viral U3 promoter. Tax also transactivates transcription of a cadre of cellular genes involved in * Corresponding author. Center for Retrovirus Research, The Ohio State University, 1925 Coffey Road, Columbus, OH, 43210-1093. Fax: +1 cell growth control, differentiation, cell cycle control, and 614 292 6473. DNA repair processes (Akagi et al., 1996; Armstrong et al., E-mail address: [email protected] (K. Boris-Lawrie). 1993; Ressler et al., 1997; Trejo et al., 1996; Wano et al.,

1988). Tax is sufficient to induce cellular transformation in the BLV long terminal repeat (LTR) sequences, including vitro and neoplastic disease in mice (Endo et al., 2002; the TRE and RxRE. BLV LTR sequences are replaced Grossman et al., 1995; Nerenberg, 1990; Robek and Ratner, with LTRs of spleen necrosis virus (SNV). The con- 1999; Ross et al., 1996). Rex regulatory protein functions stitutive SNV U3 promoter modulates Tax-independent through the Rex responsive element (RxRE) in viral genomic vector gene expression. Rex/RxRE-independent expres- RNA to activate nuclear export of unspliced and singly sion of hybrid SNV-BLV RNA is attributed to 5V SNV spliced transcripts (Heger et al., 1999; Kusuhara et al., 1999). RU5 sequences, which in other studies have been shown While Rex is required for efficient viral replication, the to facilitate Rev/Rev responsive element (RRE)-independ- protein does not play a direct role in cellular transformation ent expression of unspliced HIV-1 gag-pol RNA (Butsch (Ye et al., 2003). The BLVand HTLVaccessory genes encode et al., 1999). BLV env is positioned downstream of an multifunctional proteins that determine efficient infection of internal ribosome entry sequence to facilitate cap-inde- lymphocytes and maintenance of high proviral loads in pendent translation of Env from polycistronic gag-pol-env animals, activation of host cells, and regulation of gene genomic RNA. BLV SGV replicates a selectable BLV transcription (Albrecht and Lairmore, 2002; Kucerova et al., vector to a titer similar to BLV in cultured D17 1999; Lefebvre et al., 2002). osteosarcoma cells (Boris-Lawrie et al., 1997). The vector Experimental infection of rabbits by inoculation of BLV- virus is replication-competent in D17 cells and in infected cells reproducibly produces persistent infection of peripheral blood mononuclear cells (PBMC) of exper- lymphocytes and wasting disease within 20 months and imentally infected rats and rabbits (Boris-Lawrie et al., provides a tractable model to evaluate BLV vaccination 1997; Kucerova et al., 1999). strategies (Altanerova et al., 1989a, 1989b; Wyatt et al., Rabbits have been infected by inoculation with BLV 1989). We have previously developed a genetically sim- SGV-producer cells or by in vivo transfection of BLV SGV plified derivative of BLV that replicates independently of proviral DNA (Boris-Lawrie et al., 1997; Kucerova et al., the BLV regulatory and accessory genes and lacks patho- 1999). The BLV SGV subsequently propagates by reverse genicity in rabbits (Boris-Lawrie and Temin, 1995; Boris- transcription and sequential cycles of error-prone retroviral Lawrie et al., 1997). Here, we test the hypothesis that long- replication generate quasi-species of BLV SGVs that have term infection of rabbits with the genetically simplified BLV potential to induce a polytypic immune response against can prevent infection by pathogenic BLV. diverse strains of BLV (Kucerova et al., 1999). Immunoblot The simplified BLV derivative is encoded by a BLV results have verified that BLV SGV infection induces structural gene vector (SGV) that is distinctly different antibodies against BLV Gag and Env structural proteins in from classic retroviral vectors. The SGV encodes BLV rats and rabbits, and antiviral seroconversion remained Gag, Pol and Env proteins from vector genomic RNA detectable for 15 months of a 20-month study in rabbits that contains BLV cis-acting replication sequences for the (Kucerova et al., 1999). Compared to BLV controls, the express purpose of producing a replication-competent BLV SGV rabbits exhibited similar levels of antiviral vector virus (Fig. 1)(Boris-Lawrie and Temin, 1995; antibodies but a proviral load that is lower by a factor of Boris-Lawrie et al., 1997). The BLV SGV lacks most of one-hundred. We speculate that the robust antiviral

Fig. 1. Comparison of genomic structures of BLV and BLV SGV with cis-acting replication sequences. (A) BLV genome. Terminal white boxes, BLV LTRs with U3, R, U5 regions labeled. Labeled rectangles, open reading frames; TRE, Tax responsive element; asterisk, major BLV splice donor; RxRE, Rex- responsive element; BLV cis-acting replication sequences: att, provirus integration sequence; E, viral RNA encapsidation signal, PBS, primer binding site; PPT, polypurine tract. (B) BLV SGV. Terminal black boxes, SNV LTRs with U3, R, U5 regions labeled. Shaded rectangle labeled IRES, internal ribosome entry sequence of poliovirus. 436 V. Altanerova et al. / Virology 329 (2004) 434–439 response relative to lower proviral load is attributable to ciency, reminiscent with the outcome of AIDS in HIV-1 constitutive expression from the Tax-independent SNV U3 infected patients (Altanerova et al., 1989a, 1989b). The promoter in a small population of activated lymphocytes BLV SGV rabbits exhibited diminishing anti-Gag antibody (Kucerova et al., 1999). The BLV SGV rabbits lacked any levels, which became undetectable in rabbit 44-6 at 3 clinical sign of disease progression for the duration of the months and rabbit 44-5 after 14 months. The results verified study, whereas all BLV rabbits exhibited opportunistic exposure to BLV in the BLV SGV and control rabbits. infections that culminated in a clinical outcome of death within a period of 3 to 20 months. Here, we used these BLV SGV infection protects against BLV infection previously characterized BLV SGV rabbits to determine whether or not chronic infection with BLV SGV induces PCR assays to assess BLV challenge infection of PBMC protection against challenge with wild type BLV. were performed at four sampling points between 1 and 20 months. First, PCR was performed with primers complementary to BLV pol sequences, which are present in both Results the BLV SGV and BLV genome. PCR with primers complementary to h-actin was used to assess any minor Challenge with BLV producer cells induces anti-BLV differences in sample loading. Proviral loads were assessed antibodies in parallel against a dilution series of PBMCs from a BLV- infected rabbit. BLV SGV rabbit 44-5 displayed low, but This is a long-term pilot study that commenced at 24 consistently positive signal from 1 to 20 months (Fig. 2). By months after infection with BLV SGV with a total duration comparison to the titration series, the intensity of the of 4 years. At the onset, the BLV SGV rabbits displayed no proviral signal is similar to 100 PBMC from a BLV-infected clinical symptoms and there were no differences in control. BLV SGV rabbit 44-6 also produced consistently hematocrit values between challenged and control rabbits, positive signal from 1 to 20 months and the level of which ranged from 0.3 to 0.5 and verified lack of intensity is similar to 50 PBMC. BLV rabbit 46-3 displayed lymphoproliferation. The BLV SGV rabbits (44-5 and 44- a robust signal beginning at 1 month post-challenge and 6) and two age-matched naRve control rabbits (46-3 and 46- displayed intensity similar to 50,000 BLV PBMC. Similar to 5) were challenged with 1.0 Â 108 BLV-infected fetal lamb the trend in BLVantibody level, the proviral load declined in kidney (FLK/BLV) cells, which harbor multiple BLV the sample collected just before death. The signal at 20- proviruses (Altaner et al., 1985). The challenged rabbits months declined to an intensity similar to 1000 BLV PBMC. were evaluated at monthly intervals for antiviral antibody The intensity of the h-actin signal remained similar between levels by immunoblot, BLV proviral load by PCR and both groups of samples and verified similar sample loading. clinical signs of disease progression. A second PCR was performed with primers comple- Challenge with the FLK/BLV cells induced BLV Gag mentary to BLV tax sequences, which are present exclu- antiserum in both the BLV SGV rabbits and the age- sively in the BLV genome. Southern blot analysis with BLV matched controls (Table 1). Consistent with our previous tax probe was used to verify the specificity of the PCR observations (Kucerova et al., 1999), the control BLV- product and DNA from BLV-infected rat cell line R(BLV) rabbits sustained anti-Gag antibody levels until they provided a positive control for the size of the 1009-bp tax succumbed to clinical decline. The BLV rabbits BLV amplicon (Fig. 3). Beginning at 1 month, BLV tax presented with severe weight loss, bronchopneumonia, sequences were detectable in 50,000 PBMC of BLV rabbit diarrhea, abscesses, and leg paralysis. The cause of death 46-3 (Fig. 3, panel B). The observation of signal from was opportunistic infections due to severe immunodefi- 50,000 PBMC indicates that the sensitivity of the tax and

Table 1 Reactivity to BLV Gag in BLV-challenged rabbits Status Animal no. Level of BLV Gag antisera Month post-challenge with BLVa 012369121420 BLV SGV-infected for 24 months 44-5 bMD 3+ 2+ 2+ 2+ 1+ 1+ 1+ bMD 44-6 bMD + F bMD bMD bMD bMD bMD bMD NaRve age-matched control 46-3 bMD 2+ 2+ 2+ 2+ 2+ 2+ 2+ 1+b 46-5 bMD 2+ 2+ NAc NA NA NA NA NA a Western immunoblot with alkaline phosphatase detection of BLV-specific antisera. bMD, no visible signal at 1:50 dilution; F, barely detectable signal at 1:50 dilution; 1+, detectable at 1:50 dilution; 2+, detectable at 1:500 dilution; 3+, detectable at 1:5000 dilution. b Rabbit 46-3 died during month 21 of opportunistic infections attributable to immunodeficiency. c NA, not available because death of rabbit at month 3 of opportunistic infections attributable to immunodeficiency. V. Altanerova et al. / Virology 329 (2004) 434–439 437

over a 6-month period. Each recipient rabbit became seropositive for anti-BLV Gag, positive by PCR for pol and negative for tax (data not shown). The data verify the presence of transcriptionally active BLV SGV provirus in PBMCs and that the vaccinated animals are protected against BLV infection.

Discussion

Infection of bovine with BLV is a barrier to cattle exportation and infection of humans with HIV-1 and HTLV- 1 represents a global health problem. Therapeutic elimi- nation of these viral infections is complicated by at least two hallmark features of the retrovirus lifecycle: permanent integration of provirus into the genome of infected cells and capacity to produce genetically diverse progeny. These features of the replication cycle of retroviruses present challenges to conventional vaccination strategies and provide rationale for development of vaccines based on replication-competent retroviral vector (Boris-Lawrie and Temin, 1995; Temin, 1993). A preventative vaccine against BLV infection is a step toward development of a preventative treatment against human lymphotropic infections. Our results validate that delivery by in vivo DNA transfection is

Fig. 2. Semi-quantitative PCR analysis of BLV pol detects differences in proviral load between BLV SGV and BLV-rabbits. PBMC were harvested at month indicated after challenge with 1 Â 108 FLK/BLV cells from indicated rabbits that had been infected with BLV SGV for 24 months or from age-matched naı¨ve controls. DNA from 50,000 PMBC was subjected to nested PCR and gel electrophoresis and stained with ethidium bromide. The position of the 591-bp amplicon is designated by the labeled arrow. In parallel, DNA from 20,000 PBMC was subjected to PCR with primers complementary to h-actin. The position of the 594-bp amplicon is designated by the labeled arrow and the signals were quantified by densitometry. pol PCR assays is similar in magnitude. The tax sequences were not detectable in equivalent PBMC samples from the BLV-challenged BLV SGV rabbits (44-5, panel A and 44-6, panel C) nor from a naive rabbit kept in close proximity of the challenged animals (46-4, panel D). As summarized in Table 2, both BLV rabbits consistently exhibited presence of BLV provirus, whereas tax was not detectable in either BLV SGV rabbit, despite antiviral seroconversion. The data indicate that the BLV SGV rabbits do not harbor detectable Fig. 3. BLV tax sequences are not detectable in PBMC of BLV SGV rabbits BLV provirus and that BLV SGV infection protects against after challenge with BLV. DNA from 50,000 PMBC was subjected to PCR BLV infection. The BLV SGV rabbits have remained with BLV tax primers and the products were subjected to Southern blot clinically healthy for longer than 24 months. hybridization with gel-purified radiolabeled tax DNA probe. R(BLV), A second approach was used to confirm BLV status of positive control DNA from rat cells chronically infected with BLV [100 ng] the challenged BLV SGV rabbits. Ten ml of blood from (Altanerova et al., 1989a,b). (A) BLV SGV rabbit 44-5 does not exhibit detectable tax sequences. (B) BLV rabbit 46-3 exhibits tax sequences the BLV SGV rabbits (44-5 and 44-6) was inoculated beginning one-month post-challenge. (C) BLV SGV rabbit 44-6 does not intravenously to naRve adult rabbits and PBMC were exhibit detectable tax sequences. (D) negative control naRve rabbit 46-4 harvested and subjected to immunoblot and PCR analysis does not exhibit detectable tax sequences. 438 V. Altanerova et al. / Virology 329 (2004) 434–439

Table 2 BLV tax sequences in rabbit PBMCs by PCR and Southern blot hybridization Status Animal no. Responsea Month post-challenge with BLV 012369121420 BLV SGV-infected for 24 months 44-5 À NDb ÀÀ À À ND ND À 44-6 ÀÀÀÀÀÀÀÀÀ NaRve age-matched control 46-3 À + +ND+++++c 46-5 À ++NAd NA NA NA NA NA a Presence (+) or absence (À) of positive signal after gene-specific PCR and Southern blot hybridization on DNA from 50,000 PBMC. b ND, not determined, sample not available for analysis. c Rabbit 46-3 died during month 21 of opportunistic infections attributable to immunodeficiency. d NA, not available because death of rabbit at month 3 of opportunistic infections attributable to immunodeficiency. a safe and effective approach to induce chronic BLV SGV intradermal injection at five sites between the mid-thoracic infection. Four years after infection, BLV SGV produced and inguinal region of the dorsal side. Blood transfusion protection against challenge with pathogenic BLV. Cell- from BLV SGV rabbits to naRve control rabbits was mediated immunity is induced by BLV infection (Gatei performed using whole blood (10 ml) and intravenous et al., 1993; Kabeya et al., 2001) and replication of BLV inoculation into the central ear vein of adult rabbits. The SGV is expected to induce an anti-BLV cell-mediated rabbits were maintained with unlimited access to food and immune response that is essential protection against BLV water in the approved animal care facility of the Cancer infection. Another possible explanation for protection Research Institute, Slovak Academy of Sciences, Bratislava, against BLV infection is virus interference. However, this Slovakia. mechanism would be expected to require a significant percentage of BLV SGV-positive cells, and this is not PCR analysis of PBMC DNA observed in either tissue culture experiments nor in rabbits (Boris-Lawrie et al., 1997; Kucerova et al., 1999). We To prepare peripheral blood mononuclear cell (PBMC) speculate that a small population of PBMC harbor the BLV samples for PCR, PBMCs were isolated by a Ficoll-Paque SGV and these cells are sufficient to provide immunological (Pharmacia) gradient centrifugation and washed three times memory of exposure to BLV antigens and elicit a cell- with phosphate buffered saline. Lymphocytes were lysed mediated immune response. with SDS, treated with proteinase K overnight at 37 8C, and In conclusion, this small but long-term pilot study was a extracted with phenol-chloroform. One microgram of DNA necessary step in evaluation of SGV as a protective was combined in a 50-Al PCR mixture with specific primers, vaccination approach. We observed that both of the BLV incubated for 1 min at 94 8C, 1 min at 54–62 8C, 1 min at SGV rabbits, but not the naive control animals, are protected 72 8C for 25–30 cycles, and 10 min at 72 8C. The products against BLV infection and disease and that BLV SGV were analyzed on agarose gels. BLV pol primers KB2341 infection over a 44-month period does not produce collateral (GAA CGC CTC CAG GCC CTT CAA GA) and KB561 disease. These results provide proof-of-principle that the (CAT TGG AGG TCT CCT AAG ACC) amplify a 709-bp SGV is a promising novel live-attenuated virus vaccination primary PCR product. Aliquots of primary PCR product strategy and warrants investigation in a large study were subjected to seminested PCR with BLV pol primers population that provides statistical prediction of protective KB560 (GGA GGT TTG TGC ATG ACC TAC) and efficacy. Furthermore, this study indicates that in vivo KB561 to amplify a 591-bp PCR sequence. BLV tax transfection with SGV based on HTLV or HIV-1 is a sequences were amplified by nested PCR and detected by promising vaccination approach against human lympho- Southern blot hybridization. The primers for the primary tropic disease. reaction were (sense) AACGACAAAATTATTTCTTGTC and (antisense) CGGGGCGGTGGCGGCGCCTAG and the primers for the secondary reaction were (sense) GCAAQ Material and methods GTGTTGTTGGTTGG and (antisense) TCAAAAAAGGQ CGGGAGAGCC. BLV tax-specific PCR products were Animal procedures detected by Southern blot hybridization to [a32P]-dCTP- labeled probe that was prepared by PCR using BLV provirus As described previously, BLV SGV infections were pBL913 as a template and purified from agarose gels by produced in outbred grey Chinchilla rabbits by transfection QIAEX II Gel Extraction Kit (Qiagen). The probe was with vector DNA (Kucerova et al., 1999). Six-week-old labeled by the random-primer method with Redi-Primek rabbits were treated with three 50-Ag doses of BLV SGV reagent (Amersham). Hybridization was performed in Rapid provirus pU5gag-pol-env (Boris-Lawrie et al., 1997)by Hybk solution (Amersham) under stringent conditions and V. Altanerova et al. / Virology 329 (2004) 434–439 439 signal detected by autoradiography. Control reactions were ficiency virus Rev/RRE-independent Gag production. J. Virol. 73, performed with actin primers (Kucerova et al., 1999). 4847–4855. Endo, K., Hirata, A., Iwai, K., Sakurai, M., Fukushi, M., Oie, M., Higuchi, M., Hall, W.W., Gejyo, F., Fujii, M., 2002. Human T-cell Western immunoblot analysis leukemia virus type 2 (HTLV-2) Tax protein transforms a rat fibroQ blast cell line but less efficiently than HTLV-1 Tax. J. Virol. 76, As described previously (Boris-Lawrie et al., 1997), 2648–2653. whole BLV was prepared by ultracentrifugation of cell-free Gatei, M.H., Good, M.F., Daniel, R.C., Lavin, M.F., 1993. T-cell responses to highly conserved CD4 and CD8 epitopes on the outer membrane culture supernatants of FLK/BLV cells. Virus proteins were protein of bovine leukemia virus: relevance to vaccine development. separated on polyacrylamide gels, transferred to nitro- J. Virol. 67, 1796–1802. cellulose membrane. Membrane strips were subjected to Grossman, W.J., Kimata, J.T., Wong, F.H., Zutter, M., Ley, T.J., Ratner, L., Western immunoblot analysis with the designated rabbit 1995. Development of leukemia in mice transgenic for the tax gene of sera prepared at various dilutions and positive control serum human T-cell leukemia virus type I. Proc. Natl. Acad. Sci. U.S.A. 92, 1057–1061. was polyspecific anti-BLV rabbit serum. Western blue Heger, P., Rosorius, O., Hauber, J., Stauber, R.H., 1999. Titration of cellular stabilized substrate for alkaline phosphatase (Promega) export factors, but not heteromultimerization, is the molecular was used for color development. mechanism of trans-dominant HTLV-1 rex mutants. Oncogene 18, 4080–4090. Kabeya, H., Ohashi, K., Onuma, M., 2001. Host immune responses in the course of bovine leukemia virus infection. J. Vet. Med. Sci. 63, Acknowledgments 703–708. Kucerova, L., Altanerova, V., Altaner, C., Boris-Lawrie, K., 1999. Bovine We thank Drs. Patrick Green, Stacey Hull, Lawrence leukemia virus structural gene vectors are immunogenic and lack pathogenicity in a rabbit model. J. Virol. 73, 8160–8166. Mathes, and Tiffiney Roberts for critical comments on the Kusuhara, K., Anderson, M., Pettiford, S.M., Green, P.L., 1999. Human T- manuscript. This research was supported by: the National cell leukemia virus type 2 Rex protein increases stability and promotes Institutes of Health (FIRCA TW01217-01 and NCI nuclear to cytoplasmic transport of gag/pol and env RNAs. J. Virol. 73, P30CA16058); American Cancer Society-Ohio Division; 8112–8119. VEGA Agency of the Slovak Academy of Sciences (2/3095/ Lefebvre, L., Vanderplasschen, A., Ciminale, V., Heremans, H., Dangoisse, O., Jauniaux, J.C., Toussaint, J.F., Zelnik, V., Burny, A., Kettmann, R., 24); Centre of Excellence of SAS Bratislava Molecular Willems, L., 2002. Oncoviral bovine leukemia virus G4 and human T- Medicine. cell leukemia virus type 1 p13(II) accessory proteins interact with farnesyl pyrophosphate synthetase. J. Virol. 76, 1400–1414. Nerenberg, M.I., 1990. An HTLV-I transgenic mouse model: role of the tax gene in pathogenesis in multiple organ systems. Curr. Top. Microbiol. References Immunol. 160, 121–128. Ressler, S., Morris, G.F., Marriott, S.J., 1997. Human T-cell leukemia virus Akagi, T., Ono, H., Shimotohno, K., 1996. Expression of cell-cycle type 1 Tax transactivates the human proliferating cell nuclear antigen regulatory genes in HTLV-I infected T-cell lines: possible involvement promoter. J. Virol. 71, 1181–1190. of Tax1 in the altered expression of cyclin D2, p18Ink4 and p21Waf1/ Robek, M.D., Ratner, L., 1999. Immortalization of CD4(+) and CD8(+) T Cip1/Sdi1. Oncogene 12, 1645–1652. lymphocytes by human T-cell leukemia virus type 1 Tax mutants Albrecht, B., Lairmore, M.D., 2002. Critical role of human T-lymphotropic expressed in a functional molecular clone. J. Virol. 73, 4856–4865. virus type 1 accessory proteins in viral replication and pathogenesis. Ross, T.M., Pettiford, S.M., Green, P.L., 1996. The tax gene of human T- Microbiol. Mol. Biol. Rev. 66, 396–406 (table). cell leukemia virus type 2 is essential for transformation of human T Altaner, C., Ban, J., Zajac, V., Rossler, H., Rosenthal, S., Kettmann, R., lymphocytes. J. Virol. 70, 5194–5202. Burny, A., 1985. Isolation and characterization of cell clones Sagata, N., Yasunaga, T., Ohishi, K., Tsuzuku-Kawamura, J., Onuma, M., producing various amounts of bovine leukosis virus. Folia Biol. Ikawa, Y., 1984. Comparison of the entire genomes of bovine leukemia (Praha) 31, 107–114. virus and human T-cell leukemia virus and characterization of their Altanerova, V., Ban, J., Altaner, C., 1989a. Induction of immune deficiency unidentified open reading frames. EMBO J. 3, 3231–3237. syndrome in rabbits by bovine leukaemia virus. AIDS 3, 755–758. Temin, H.M., 1993. A proposal for a new approach to a preventive vaccine Altanerova, V., Portetelle, D., Kettmann, R., Altaner, C., 1989b. Infection against human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. of rats with bovine leukaemia virus: establishment of a virus-producing U.S.A. 90, 4419–4420. rat cell line. J. Gen. Virol. 70 (Pt. 7), 1929–1932. Trejo, S.R., Fahl, W.E., Ratner, L., 1996. c-sis/PDGF-B promoter Armstrong, A.P., Franklin, A.A., Uittenbogaard, M.N., Giebler, H.A., transactivation by the Yax protein of human T-cell leukemia virus Nyborg, J.K., 1993. Pleiotropic effect of the human T-cell leukemia type 1. J. Biol. Chem. 271, 14584–14590. virus Tax protein on the DNA binding activity of eukaryotic tran- Wano, Y., Feinberg, M., Hosking, J.B., Bogerd, H., Greene, W.C., 1988. scription factors. Proc. Natl. Acad. Sci. U.S.A. 90, 7303–7307. Stable expression of the tax gene of type I human T-cell leukemia virus Boris-Lawrie, K., Temin, H.M., 1995. Genetically simpler bovine leukemia in human T cells activates specific cellular genes involved in growth. virus derivatives can replicate independently of Tax and Rex. J. Virol. Proc. Natl. Acad. Sci. U.S.A. 85, 9733–9737. 69, 1920–1924. Wyatt, C.R., Wingett, D., White, J.S., Buck, C.D., Knowles, D., Reeves, R., Boris-Lawrie, K., Altanerova, V., Altaner, C., Kucerova, L., Temin, Magnuson, N.S., 1989. Persistent infection of rabbits with bovine H.M., 1997. In vivo study of genetically simplified bovine leukemia leukemia virus associated with development of immune dysfunction. virus derivatives that lack tax and rex. J. Virol. 71, 1514–1520. J. Virol. 63, 4498–4506. Butsch, M., Hull, S., Wang, Y., Roberts, T.M., Boris-Lawrie, K., 1999. Ye, J., Silverman, L., Lairmore, M.D., Green, P.L., 2003. HTLV-1 Rex is The 5V RNA terminus of spleen necrosis virus contains a novel required for viral spread and persistence in vivo but is dispensable for posttranscriptional control element that facilitates human immunode- cellular immortalization in vitro. Blood 102, 3963–3969.