INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The human T-cell leukemia/lymphoma virus type 1 (HTLV-1) is the causative agent of adult T-cell leukemia (38) and of tropical spastic paraparesis/HTLV-1-associated myelopathy (11). It has also been associated with a number of inflammatory diseases, such as pediatric infectious dermatitis (23), uveitis (26), and some cases of arthropathy (18) and polymyositis (27). The overall prevalence of severe HTLV-1-associated disease is 2 to 8% among HTLV-1-infected persons, estimated to represent 15 to 25 million persons worldwide, mostly in Central and South America, equatorial Africa, and Asia (7). In regions where it is endemic, HTLV-1 is transmitted primarily from mother to child during breast-feeding; later, it is transmitted sexually between adults. In the Western world, the principal routes of infection are parenteral (transfusion and needle sharing among intravenous drug users) and sexual. Mother-to-child transmission should be prevented easily by discouraging breast-feeding, but this has proved to be impossible in areas of endemicity. Campaigns to encourage condom use in some areas of endemicity have also had disappointing results. Thus, the development of an HTLV-1 vaccine appears to be crucial.

Experimental vaccines against HTLV-1 in which the envelope protein was used for immunization have been tested in rabbits, rats, and monkeys. A series of vaccine candidates based on recombinant vaccinia virus vectors containing the HTLV-1 env gene have been tested in rabbits. Two vaccina virus-based recombinants, WR-env17 and WR-proenv1, induced an Env-specific antibody response and were protective (32). WR-SFB5env induced antibodies against gp46 that were not neutralizing and conferred only partial protection (13). We evaluated WR-SFB5env in rats: animals primed and boosted with this recombinant vaccinia virus developed antibodies against the HTLV-1 Env protein and showed partial protection against challenge from HTLV-1-infected MT2 cells (20). Cynomolgus macaques immunized with WR-SFB5env were also protected against challenge (17). Immunization by synthetic peptides with overlapping neutralizing domains in the central region of gp46 protected against challenge in rabbits (33).

The highly attenuated vaccinia virus derivative NYVAC (37) was engineered to express antigens from both animal and human pathogens. NYVAC-based recombinants expressing the rabies virus glycoprotein, a polyprotein from Japanese encephalitis virus, and seven antigens from Plasmodium falciparum were demonstrated to be safe and immunogenic in an initial study with humans (35). NYVAC-based recombinants have also been shown to protect against infection with other retroviruses, such as human immunodeficiency virus (HIV) type 2 and simian immunodeficiency virus (3, 9, 28). When NYVAC containing the HTLV-1 env gene was assessed in rabbits, immunization with this recombinant and boosting with recombinant Env protein protected against challenge from HTLV-1-infected cells. However, 5 months after the initial challenge, the immunized rabbits were not protected against exposure to a large inoculum of blood from an HTLV-1-infected animal (10).

Genetic or DNA-based immunization involves delivery of an immunogen-encoding expression plasmid to a given tissue in vivo to induce an immune response to the encoded immunogen. This novel form of immunization results in the production of correctly folded, glycosylated protein antigens de novo. Indeed, in most studies to date, DNA-based immunization has been found to induce the full range of immune responses, including neutralizing antibodies, a cytotoxic T-cell response (cytotoxic T lymphocytes), and protection against challenge (8, 15). Several DNA vaccines have been shown to be effective in nonhuman primates (4, 24), but in most cases multiple administrations were necessary to induce adequate immunity. Naked DNA has also been used to induce neutralizing antibodies against HTLV-1 Env glycoproteins in mice (1, 12). Although immunization with various naked plasmid constructs containing the env gene under the control of various promoters was not sufficient to elicit specific detectable humoral responses, booster administration with recombinant baculovirus gp62, after priming with the DNA HTLV-1 env gene, resulted in detectable humoral and cell-mediated immune responses.

We showed recently that the squirrel monkey, Saimiri sciureus, a South American primate devoid of endemic infection with simian T-cell leukemia virus, was susceptible to experimental infection with HTLV-1. Experimental inoculation led to chronic infection, similar to that in humans, with high titers of antibodies against HTLV-1 and HTLV-1 provirus being detectable by PCR in peripheral blood mononuclear cells (PBMCs) for up to 4 years after inoculation (21, 22). The squirrel monkey thus appears to be a suitable model for evaluating candidate HTLV-1 vaccines. The aim of the study reported here was to evaluate the immunogenicity and protective efficacy of a vaccination regimen involving priming with a candidate NYVAC HTLV-1 vaccine and boosting with naked HTLV-1 env DNA in squirrel monkeys.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Vaccine preparations. Recombinant NYVAC HTLV-1 env candidate vaccines were constructed from a plasmid containing DNA from the HTLV-1 1711 clone, obtained by culture of PBMCs from a West African patient (5). The procedure for generating and analyzing the recombinant NYVAC vaccine expressing the entire HTLV-1 env (gp46 and gp21) gene has been described previously (10). The DNA-based immunogen, consisting of the HTLV-1 env gene and the cytomegalovirus (CMV) promoter and long terminal repeat (CMV-env-LTR), was kindly provided by M.-C. Dokhelar, Institut de Génétique Moléculaire, Paris, France. This plasmid was constructed by Delamarre et al. (6) in order to study the expression of HTLV-1 envelope glycoprotein and its role in cell-to-cell viral transmission. The expression of this plasmid was studied by immunoprecipitation of the envelope glycoprotein from transfected COS-1 cells with sera from HTLV-1-infected individuals. This plasmid has also been studied for its capacity to induce humoral and cell-mediated immune responses against HTLV-1 in mice after boosting with recombinant Env protein (1, 12). Expression of the Env protein with this plasmid was also evaluated in transiently transfected HeLa cells by Western blotting and immunofluorescence. The transfected cells showed a high level of Env protein expression with sera from HTLV-1-infected rabbits. The functional expression of HTLV-1 Env protein was also confirmed by the observation of syncytia in HeLa cells transfected with the same plasmid (1, 12).

Animals, vaccination regimens, and challenge. For the vaccination experiments, which complied with French legislation on animal experiments, we used 8-year-old male squirrel monkeys from the primate center of the Pasteur Institute of French Guiana. In the initial protocol (Fig. 1, protocol 1), three monkeys (monkeys 1711, 1811, and 89057) were injected intramuscularly with 10⁸ PFU of NYVAC containing the HTLV-1 env gene at 0, 1, and 3 months. A control monkey (monkey 1495) was similarly injected intramuscularly with 10⁸ PFU of NYVAC-rabies G protein (NYVAC-RG) at the same time. Six months after the last administration of NYVAC-env, two of the three vaccinated monkeys (monkeys 1811 and 89057) were boosted with 500 µg of the naked DNA immunogen CMV-env-LTR intramuscularly into the tibialis anterior muscle. The third monkey (monkey 1711) and the control (monkey 1495) were injected with a naked DNA vector containing the -galactosidase gene (CMV-gal) (Table 1).

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Immunization protocols and challenge. In the first protocol (protocol 1), three monkeys were primed by three intramuscular injections of 108 PFU of NYVAC containing the HTLV-1 env gene. A control monkey was similarly injected with NYVAC-RG. Six months after the last administration of NYVAC-env, two of the three vaccinated monkeys were boosted with 500 µg of naked DNA (CMV-env-LTR) while the third monkey and the control were injected with CMV-gal. The monkeys were killed 1 year after challenge and examined for the presence of HTLV-1 provirus in various organs. In the second immunization protocol (protocol 2), three monkeys received a single intramuscular injection of 500 µg of CMV-env-LTR DNA and the control monkey received the CMV-gal vector. Six months later, the three vaccinated monkeys received a series of three booster injections, separated by 1-month intervals, of 108 PFU of the NYVAC-based candidate vaccine containing the HTLV-1 env and gag genes. The control monkey received 108 PFU of NYVAC-RG at the same times. Two months after the last boost, the monkeys were challenged with an intravenous injection of squirrel monkey HTLV-1-producing cells (EVO/798). Before and after challenge, the monkeys were bled each month and their PBMCs were separated on Ficoll plaques. The level of antibodies to HTLV-1 in the sera of immunized animals was evaluated by ELISA and Western blotting, and the presence of an HTLV-1-specific lymphocyte proliferative response (LPA) was evaluated 1 month after the last boost and 2 and 6 months after challenge.

Serological and molecular biology assays. At various times after challenge, the serum levels of specific HTLV-1 antibodies were determined by enzyme-linked immunosorbent assay (ELISA) (Cobas Core Anti-HTLV-I/II enzyme immunoassay; Roche, Basel, Switzerland) and confirmed by Western blot analysis (HTLV-1; 2.3 Diagnostic Biotechnology, Singapore). We also tested genomic DNA extracted from PBMCs for the presence of HTLV-1 sequences, as described by Ibrahim et al. (16). PCR was performed as previously described with the gag-specific primers gag1 and gag2 (19) or the pol-specific primers SK54 and PolAG2 (16). The amplified products were subjected to electrophoresis in a 1.4% agarose gel, transferred to nylon membranes, and hybridized with ³²P-labeled internal oligonucleotide probes specific for the gag (5' GCAAAGGTACTGCAGGAGGT 3' ) and pol (5' TTCCAGCCCTACTTTGCTTTCACTGTCCC 3' ) sequences. The membranes were washed and placed against Hyperfilm MP (Amersham, Little Chalfont, Buckinghamshire, United Kingdom) at 80°C for 24 h and then against a second film for 1 week.

Lymphocyte proliferation assays. PBMCs were collected from the eight immunized monkeys 1 month after the last boost and 2 and 6 months after challenge (Fig. 1). The PBMCs were separated on Ficoll plaques and suspended at a concentration of 5 × 10⁶ per ml in RPMI 1640 medium containing 10% fetal calf serum supplemented with glutamine (2 mmol/liter), penicillin (50 IU/ml), and streptomycin (50 µg/ml). The remaining cells were added to triplicate wells (2 × 10⁵ cells per well) in 96-well round-bottom microtiter plates (Costar; Corning, Inc., Corning, N.Y.) and incubated for 5 days in a final volume of 200 µl at 37°C in 5% CO₂ in the presence or absence of 1 µg of recombinant HTLV-1 Env gp46 protein per well corresponding to amino acids 16 to 312 (Intracel, London, United Kingdom) or two HTLV-1 Gag peptides (amino acids 110 to 130 and 175 to 191) corresponding, respectively, to portions of the HTLV-1 p19 and p24 Gag proteins (ABTeurope, London, United Kingdom). Phytohemagglutinin was used as a positive control, and cells were incubated with 4 µg of this nonspecific mitogen per ml for 24 h. Recombinant Schistosoma japonicum glutathione transferase and HIV peptide corresponding to a portion of the Env protein (V3 loop region) were used as negative controls. After stimulation, the wells were pulsed for 12 h with 0.5 µCi of [³H]thymidine (Amersham, Les Ulis, France); the cells were then lysed, and ³H incorporation was measured in a liquid scintillation counter (Rackbeta 1209-b; LKB/Wallac, Turku, Finland).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Protocol 1. With the first immunization protocol, no antibodies against HTLV-1 Env protein were detected in any of the three vaccinated animals after three administrations of the NYVAC-HTLV-1 env vaccine preparation (Fig. 2). However, one of the two animals boosted with CMV-env-LTR (monkey 1811) developed a serum antibody response 2 weeks after being boosted (Fig. 2B and 3).

View larger version (19K): [in this window] [in a new window]   FIG. 2.   Antibody responses to HTLV-1, as detected by ELISA, before and after challenge in monkeys immunized with NYVAC-env and boosted with naked DNA containing the HTLV-1 env gene. (A, B, and C) Results with monkeys 1711, 1811, and 89057, respectively; (D) control monkey 1495, which was first injected with NYVAC-RG and then boosted with CMV-gal. Arrows denote times of vaccination, and the filled arrows indicate times of challenge. O.D., optical density.

View larger version (38K): [in this window] [in a new window]   FIG. 3.   Western blot analysis of sera from immunized monkeys before and after challenge with HTLV-1-producing cells. (Protocol 1) Western blot pattern for monkeys immunized with NYVAC-env and boosted with naked DNA containing the HTLV-1 env gene. Lanes a, Western blot pattern 1 month after the last boost (before challenge); lanes b, Western blot pattern 2 months after challenge. Monkey 1711 was immunized with NYVAC-HTLV-1 env and boosted with naked CMV-gal DNA, monkeys 1811 and 89057 were immunized with NYVAC-HTLV-1 env and boosted with naked CMV-env-LTR DNA, and control monkey 1495 was injected with NYVAC-RG and boosted with naked CMV-gal DNA. (Protocol 2) Western blot pattern for monkeys primed with naked DNA (CMV-env-LTR) and boosted with NYVAC containing the HTLV-1 env and gag genes. Lanes a, Western blot pattern 1 month after the last boost (before challenge); lanes b, Western blot pattern 2 months after challenge. Monkeys 93081, 93089, and 93096 were first immunized with naked CMV-env-LTR DNA and then boosted three times with the NYVAC-HTLV-1 env and gag vaccine preparation; control monkey 93116 was injected with naked CMV-gal DNA and boosted with NYVAC-RG. Lane C, Western blot pattern for HTLV-1-negative control monkey; lane C+, Western blot pattern for HTLV-1-positive control monkey.

View larger version (31K): [in this window] [in a new window]   FIG. 4.   Lymphocyte proliferation responses before and after challenge in monkeys primed with NYVAC-env and boosted with naked DNA containing the HTLV-1 env gene. PBMCs were assayed for the proliferative response to HTLV-1 Env (gp46) protein, as described in the text. CPM, counts per minute of [3H]thymidine incorporated in the presence or absence of the antigens. Results are shown for triplicate wells. Phytohemagglutinin (PHA) was used as a positive control, and glutathione S-transferase (GST) and medium alone were used as negative controls. Monkey 1495 is a control animal.

Protocol 2. With the second immunization protocol, injection of CMV-env-LTR did not induce detectable levels of antibodies against HTLV-1 Env protein in any of the three vaccinated animals. In contrast, after being boosted with three injections of NYVAC containing the HTLV-1 env and gag genes, two of three boosted animals (monkeys 93081 and 93096) developed an antibody response to HTLV-1, as detected by ELISA (Fig. 5). In these two animals, anti-Env antibodies were also detected by Western blotting (Fig. 3). Two months after challenge, seroconversion against HTLV-1 was detected only in the control animal (monkey 93116) (Fig. 3 and 5D). One of the three vaccinated animals (animal 93089) died 2 months after challenge from clinically and pathologically confirmed acute renal failure. At necropsy, HTLV-1 provirus was not detected by PCR in the spleen, lymph nodes, liver, lung, kidney, or brain or in any other part of the central nervous system. When PBMCs from this animal were cultured, no markers of HTLV-1 infection could be detected by PCR or RT-PCR and Gag p19 was not found in the culture. No antibodies to Tax or Gag were detected in the other vaccinated animals (animals 93081 and 93096) 2 or 6 months after challenge or thereafter. pol-specific sequences were detected by PCR 2 or 6 months after challenge in PBMCs from the control animal but were not found in the vaccinated animals, and only PBMCs from the control monkey tested positive for tax/rex mRNA and HTLV-1 p19 (Table 1). In the lymphocyte proliferation test performed 1 month after boosting, high-level, specific responses were detected in the three immunized animals against both recombinant Env gp46 protein and Gag peptides but not in the control monkey (Fig. 6). This positive signal appeared to be directed more strongly against Env than against Gag. In contrast, the positive responses were similar in all immunized monkeys 2 months after challenge and could be maintained for 6 months (Fig. 6). In the control monkey, no lymphocyte proliferation response was observed before challenge but 2 and 6 months after challenge a cell-mediated immune response was detected (Fig. 6D).

View larger version (20K): [in this window] [in a new window]   FIG. 5.   Antibody responses to HTLV-1, as detected by ELISA, before and after challenge in monkeys primed with naked DNA (CMV-env-LTR) containing the HTLV-1 env gene and boosted three time with NYVAC containing the HTLV-1 env and gag genes. (A, B, and C) Results with monkeys 93081, 93096, and 93089, respectively; (D) results with the control monkey 93116, which was was first injected with CMV-gal and then boosted with NYVAC-RG. Arrows denote times of vaccination, and filled arrows indicate the time of challenge. O.D., optical density.

View larger version (34K): [in this window] [in a new window]   FIG. 6.   Lymphocyte proliferation responses before and after challenge in monkeys primed with naked DNA (CMV-env-LTR) containing the HTLV-1 env gene and boosted with NYVAC containing the HTLV-1 env and gag genes. PBMCs were assayed for the proliferative response to HTLV-1 Env (gp46) protein and to two Gag peptides (p19 and p24), as described in the text. Phytohemagglutinin (PHA) was used as a positive control, and glutathione S-transferase (GST), an irrelevant peptide (IRpep), and medium alone were used as negative controls. Monkey 93116 is a control animal. CPM, counts per minute.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we found that a vaccination regimen involving priming with naked DNA and boosting with NYVAC containing the HTLV-1 env and gag genes was more efficient in eliciting HTLV-1-specific antibody and cell-mediated responses and protective immunity against HTLV-1 than a regimen involving priming with NYVAC-env and boosting with naked DNA. With the latter protocol, only one of the three immunized animals was protected, and humoral and cell-mediated immune responses to HTLV-1 Env were detected. The curves for HTLV-1 seroconversion during the 6 months after HTLV-1 inoculation (challenge) of control monkeys and unprotected animals were similar to those observed for four nonimmunized control monkeys infected with HTLV-1 (Fig. 7). This comparison showed that protection was achieved in the immunized monkeys.

View larger version (17K): [in this window] [in a new window]   FIG. 7.   Antibody responses to HTLV-1, as detected by ELISA, in four nonimmunized control monkeys infected with HTLV-1. The arrow indicates the time of HTLV-1 inoculation. O.D., optical density.

The Env protein has been reported to confer partial protection against HTLV-1 infection in various animal models, but the mechanisms of immunity associated with the protection remain unclear. In particular, protection has been observed in the absence of an antibody response (10, 20), suggesting that cell-mediated immunity plays a key role. We used the recombinant NYVAC vaccine containing a Gag component in the second immunization protocol for the following reasons. First, rats inoculated with HTLV-1-transformed rat cells have been shown to develop cytotoxic T lymphocytes directed against Gag-specific epitopes rather than against Env-specific epitopes (29, 34). Second, vaccines containing Gag or Gag and Env have been shown to protect against Moloney and Friend murine leukemia viruses in mice (25, 30) and against feline leukemia virus in cats (36). In our squirrel monkey model, we found that monkeys primed with naked DNA and boosted with NYVAC containing the env and gag genes developed a humoral response to HTLV-1 after each boost and a cell-mediated immune response to Env and Gag after being boosted and challenged. Furthermore, these monkeys were protected against HTLV-1 challenge. Similar results were obtained by Hanke et al. (14), who showed that a combined immunization regimen involving priming with a DNA immunogen and boosting with the modified Ankara vaccina virus resulted in a more potent immune response, as determined by epitope-specific gamma interferon production and a cytotoxic T-lymphocyte response.

The combination of a recombinant virus and naked DNA containing the HTLV-1 env gene has also been tested in rats. HTLV-1-specific cytotoxic T lymphocytes were recovered from rats immunized with recombinant adenovirus 5 and boosted with naked DNA containing the HTLV-1 env gene (20). More recently, Robinson et al. (31) compared eight immunization protocols in rhesus macaques and found that the most promising protocol for protection against HIV was priming with naked DNA and boosting with recombinant vaccinia virus. This protection did not require neutralizing antibody production but was effective for a series of challenges.

Indeed, the ideal vaccine should induce long-lasting neutralizing antibodies to HTLV-1 in serum and a strong cell-mediated immune response. These conditions might be difficult to achieve with a single vaccine preparation, because the optimal immunization regimens for inducing humoral and cell-mediated immunity are often different. A cytotoxic T-lymphocyte response directed mainly against p40^tax protein is detected in asymptomatic human HTLV-1 carriers and in patients with tropical spastic paraparesis/HTLV-1-associated myelopathy. Bangham et al. (2) suggested that this response plays a major role in controlling HTLV-1 replication. A mutated tax might therefore also be included in the vaccine preparation to increase the breadth of the immune response against HTLV-1. Alternatively, separate immunization with naked DNA containing env, gag, and tax may be necessary, followed by boosting with live recombinant vector-based candidates or with recombinant subunit vaccine preparations in an appropriate adjuvant. As we have shown in this study, this approach is promising for the induction of sustained levels of both humoral and cell-mediated immunity.

In conclusion, our results show that the second vaccination protocol used protected squirrel monkeys against HTLV-1 infection. Future efforts should be directed to elucidating the qualitative aspects and the duration of this protection.