A peptide-loaded dendritic cell based cytotoxic T-lymphocyte (CTL) vaccination strategy using peptides that span SIV Tat, Rev, and Env overlapping reading frames
- Zachary Klase†1,
- Michael J Donio†1,
- Andrew Blauvelt3, 4, 5,
- Preston A Marx6,
- Kuan-Teh Jeang7 and
- Stephen M Smith1, 2Email author
© Klase et al; licensee BioMed Central Ltd. 2006
Received: 16 August 2005
Accepted: 06 January 2006
Published: 06 January 2006
CTL based vaccine strategies in the macaque model of AIDS have shown promise in slowing the progression to disease. However, rapid CTL escape viruses can emerge rendering such vaccination useless. We hypothesized that such escape is made more difficult if the immunizing CTL epitope falls within a region of the virus that has a high density of overlapping reading frames which encode several viral proteins. To test this hypothesis, we immunized macaques using a peptide-loaded dendritic cell approach employing epitopes in the second coding exon of SIV Tat which spans reading frames for both Env and Rev. We report here that autologous dendritic cells, loaded with SIV peptides from Tat, Rev, and Env, induced a distinct cellular immune response measurable ex vivo. However, conclusive in vivo control of a challenge inoculation of SIVmac239 was not observed suggesting that CTL epitopes within densely overlapping reading frames are also subject to escape mutations.
Several recent HIV vaccine strategies have focused on the induction of potent cellular immune responses . Experiments in the macaque model of HIV infection have shown that a strong cytotoxic T-cell lymphocyte (CTL) response against viral proteins can prevent disease, although such a response cannot prevent infection. Unfortunately, viruses which escape CTL-surveillance frequently occur in animals, and such escaped viruses can then engender disease .
Most vaccines have used whole viral proteins, delivered in a variety of ways, as immunogens. While some of these proteins in the context of particular major histocompatibility (MHC) antigen alleles show immunodominant epitopes in macaques [3, 4], a general strategy is to induce broad CTL responses against many different epitopes. Several CTL-eliciting epitopes can be present in a given protein. To date, HIV/SIV has been able to generate escape mutations within most, if not all, epitopes used to elicit CTL-responses. Many such mutant viruses can replicate to high levels and cause disease in vivo suggesting that these mutated viruses do not have significantly reduced viral fitness . However, the true functional content of the many CTL-eliciting epitopes used for vaccination has not been clearly defined.
Since CTL based vaccines reduce, but do not eliminate replication, it is expected that they will select for the emergence of escape viral mutants. For a CTL based vaccine to be durably effective, ideally, the target epitope(s) must be critical for function and be constrained such that any change in epitope sequence results in a significant deficit in the replicative fitness of the virus. Thus, in an operational definition, an "immutable" CTL epitope is one which may mutate in response to immune selection, but such mutations are transient and may never be observed because of their significantly deleterious effect on viral fitness.
In a recent study, we explored the concept of such an immutable epitope . We infected macaques with an engineered version of SIVmac239 (i.e. SIVtat1ex) which can only express the first coding exon of SIV Tat due to artificially inserted premature stop codons that prevented expression of the second coding exon of Tat. SIVtat1ex virus replicated well in the early phase, but much less well than wild type (i.e. SIVtat2ex) in the chronic phase of infection. In three macaques, SIVtat1ex "reverted" and opened up the stop codons that obstructed expression of the second coding exon of Tat (i.e. SIVtat1ex became SIVtat2ex). In two of these three animals, this change in Tat expression (i.e. expression of full length two-exon Tat instead of the original one-exon Tat) correlated with increased viral load and more rapid CD4+ T-cell depletion. In the third animal, the viral load initially increased, but then returned to low levels. Further investigation revealed that this third animal, although originally infected with SIVtat1ex, transiently had the emergence of a SIVtat2ex virus which surprisingly reverted quickly back to the less fit SIVtat1ex form. This third macaque has maintained low viral load and high CD4+ T-cell count. Immunologic studies demonstrated that this animal had a strong cellular response directed to the second coding exon of SIV Tat. Provocatively, after 4 years of infection, this animal continued to maintain the low-fitness SIVtat1ex virus with no evidence for the ability of the more fit SIVtat2ex to emerge by correcting the stop codons which prevent the expression of the second coding exon of Tat. Our interpretation of this scenario in the context of our operational definition of an "immutable CTL epitope" is that SIVtat2ex is a transitional "escape" virus of SIVtat1ex; and that in certain settings SIVtat1ex virus cannot durably transit to its more fit SIVtat2ex form because the host maintains a potent CTL selection targeted against an epitope within the second coding exon of Tat.
The above hypothesis posits that mutations in a CTL-epitope(s) embedded within a portion of SIV that codes three overlapping proteins, Tat, Rev and Env, might be difficult. The notion is that such CTL-epitopes might be "immutable" because "escape" changes in their sequences could alter Tat, Rev, or Env function (singularly or multiply) in ways that produce less-fit progeny viruses in vivo. Peptide loaded dendritic cells have been used in cancer immunotherapy and in viral vaccine efforts to induce a cellular response against specific epitopes [7, 8]. To test our hypothesis that triply over-lapping reading frames potentially restrict CTL-escapes, we immunized macaques with autologous dendritic cells, loaded with peptides from an SIV region with overlapping coding capacity for Tat, Rev, and Env. Here, we report findings when we challenged immunized animals with a pathogenic SIVmac239 virus.
Peptide-loaded dendritic cells elicited strong IFN-γ T-cell responses
To assess the effectiveness of the dendritic cell culture protocol, we performed flow cytometry for the MDDC phenotype. After 8 days in culture, cells were stained for HLA-DR and CD83. Immature dendritic cells express relatively low levels of HLA-DR and are CD83 negative, whereas mature dendritic cells express higher levels of HLA-DR and are CD83 positive. Flow cytometry revealed that greater than 80% of the cultured MDDC possessed the mature phenotype (data not shown).
Amino acid sequence of peptides from Tat, Rev, and Env used in the vaccine. (Peptide sequences are identical to those of challenge virus.)
Lack of control with viral amino acid changes when SIVmac239 challenge virus was used to infect peptide immunized macaques
In this study, we show that autologous dendritic cells, loaded with exogenous SIV peptides, can successfully induce cellular immune responses. These responses were moderate to strong, and, in general, increased with repeated immunization (data not shown). However, the vaccinated macaques seem not to effectively control the replication of a challenge virus, and inoculated animals developed viral loads similar to those of the control animals (Fig. 3). Curiously, rather than increasing after infection with SIV, the IFN-γ T-cell responses against the vaccine peptides decreased in three of the four vaccinated animals (Fig. 5). These findings are perplexing; and the unexpected early, study-unrelated demise of two experimental animals also contributed difficulties to a conclusive interpretation.
How could one explain the above findings? We note with interest the sequencing results on virus samples isolated from infected animals on day 28 after challenge (Figs. 6 &7). A close examination of the Tat sequences in Table 4 instructively suggests that the challenge virus appears to have commenced sequence changes, possibly evolving as a result of the host's CTL. Thus, if a dominant CTL epitope in SIV Tat were to span the trhcqpeka sequence, then in three of the four (75%) experimental animals (AT56, AT57, and BA20) viruses have initiated evasive amino acid mutations. Correspondingly, in the Rev sequence, if a major CTL epitope resided in gpgtanqrr, then viruses in AT57 and BA20 (50% of the experimental animals) would have started to change. A similar case could be made for Env. If the dominant epitope here is hypothesized as thiqqdpal, then three of the four viruses in vaccinated animals (AT56, AT57, and AV89; i.e. 75% of the experimental population) have changed by day 28. It remains to be established whether our hypothesized epitopes are truly the dominant in vivo SIV moieties. However the observation that the originally detected CTL responses faded quickly after virus challenge is compatible with these being relevant epitopes. Viral escape changes in these epitopes are expected to result in failure to re-stimulate the original CTL and would be consistent with the waning CTL profiles in Figure 5.
We note a sobering take home lesson from our study. Our data appear to tell us that one of our a priori facile assumptions is probably incorrect. We had assumed that just because a region of the virus is ORF dense that such region would be functionally constrained and difficult to mutate. The empirical results do not support that assumption. For example, the "k" in the middle of the Rev sequence seems to be easily changeable; as is the "r" in the middle of Env (Fig. 7). Neither is a result of immune selection, since viruses in the control animals also had these changes. Add to these mutations the additional changes seen in the viruses in vaccinated macaques, then the reality emerges that three densely over-lapping reading frames in a small region does not seem to greatly constrain virus mutability. Currently, we cannot formally conclude whether the viral changes in the vaccinated animals resulted in reduced fitness (however slight). Nonetheless, the in vivo viral replication profiles (Fig. 3) would seem to argue against this possibility.
We do want to point out several technical shortcomings to our study. First, our study group size was small and was unexpectedly confounded by the need to euthanize two vaccinated animals shortly after SIV challenge. One macaque became ill from an unrelated neoplasm, and the second developed severe enterocolitis, also believed to be unrelated to SIV, since the disease preceded SIV infection. This unanticipated happenstance reduced our vaccinated group from 4 to 2 animals and prevented a meaningful longer chronological follow up of viral sequence changes. Second, our CTL epitope interpretations are complicated by the current poor understanding of the MHC-context for rhesus macaques . Since CTL-responses are MHC dependent, a fuller understanding of macaque MHC would be helpful to design and study better CTL-vaccination in monkeys. Finally, our dose of challenge virus may be too high to see obvious protection. There could be a lower dose at which a CTL response would rapidly control the virus preventing the virus from replicating enough rounds to generate an escape variant. The above caveats aside, our current results suggest that a CTL vaccine based on the Tat, Rev, Env ORF-dense region of SIV is largely insufficient (under the currently utilized challenge condition) to control virus replication. Whether protocols of immunization with Tat, Rev and Env different from those currently employed here can exert control over virusreplication remain to be investigated. Currently, we also cannot distinguish between whether the immune responses observed in our animals were qualitatively ineffective at controlling infection or if higher quantitative immune responses were induced such could, in fact, control viral infection.
Six colony-bred rhesus macaques (Macca mulatta) were obtained from the Tulane National Primate Research Center (TNPRC) (Covington, LA). The six adult animals weighed between 6.15 to 10.25 kg, and were all seronegative for SIV. All aspects of this study were approved by the Tulane National Primate Research Center Institutional Animal Care and Use Committee.
The SIV peptides were obtained from the NIH AIDS Reagent Program (Rockville, MD). Each was fifteen amino acids in length, and overlapped adjacent peptides by eleven residues. Nine peptides were selected which completely overlapped the second exon of SIVMac239 tat 2nd exon (amino acids 98–130). Eleven Rev and twelve Env peptides were selected because their coding sequences completely or partially overlapped Tat's second exon (Table 1). The peptides from each protein were arbitrarily divided into two pools, A & B. Each pool contained 4–6 peptides. For instance, Tat pool A contained the first five peptides listed in Table 2 and Tat pool B contained the remaining four. The peptides for a given pool were dissolved together in water or DMSO at 5 mg/ml of each peptide. Each peptide exactly matched the encoded, cognate peptide of the challenge virus, SIVmac239.
Cell culture/vaccine generation
Primary blood mononuclear cells (PBMC) were separated from heparin treated rhesus macaque blood by centrifugation over Ficoll (Greiner Inc, Longwood, FL), washed, and cryo-preserved until needed for generation of dendritic cells. For each vaccination, 2.5 × 107 PBMC per animal were thawed, washed in PBS, plated across a 6-well costar plate in DMEM with 10% FBS, and placed in a 37°C/5% CO2 to allow monocyte adherence. After three hours, the media and non-adherent cells were aspirated, and the plates washed twice with PBS. Media was replaced with DC media (RPMI with 10% FBS, 50 ng/ml GMCSF (R&D Systems, Minneapolis, MN) and 10 ng/ml IL-4 (R&D Systems). Cells were allowed to differentiate for 4 days. On day 4 immature dendritic cells were aspirated from the plate and washed. Cells were resuspended in 5 ml DC maturation media (RPMI 10% FBS, 50 ng/ml GM-CSF, 10 ng/ml IL-4, 20 ng/ml TNF-α, 20 ng/ml IL-6, and 20 ng/ml IL-1β (R&D Systems)) in a T25 flask. Dendritic cells from experimental animals (AT56, AT57, AV89 and BA20) received 5 μg/ml each of the Tat, Rev and Env peptides. After four additional days in culture, mature monocyte-derived dendritic cells (MDDC) were removed from culture flasks, brought to 10 ml with DC maturation media, counted, and transferred to a 15 ml conical tube for shipment to TNPRC. MDDC cultures were analyzed by flow cytometry on a FACS Calibur (BD Biosciences, Franklin Lakes, NJ).
Six vaccinations were scheduled, at two-week intervals. Vaccination number five was delayed for two weeks, thus pushing back vaccinations number five and six. For each time point MDDC were generated as above and shipped overnight at room temperature to TNPRC. After centrifugation, 1 – 2 × 106 mature autologous dendritic cells were resuspended in 0.2 ml PBS and injected into a femoral lymph node in each animal. Experimental animals (AT56, AT57, AV89, BA20) received MDDC generated in the presence of Tat, Rev and Env peptides. Control animals (H405, T687) were cultured in the absence of peptides.
The challenge virus, a generous gift of David Watkins, University of Wisconsin, was SIVMac239(open) produced from transfected DNA and expanded in CEMx174 cells. Viral stock was diluted to 50 TCID50/ml in DMEM and 1 ml was administered intravenously to all animals six days after the final vaccination.
IFN-γ ELISpot assay (adapted from Amara et al) was performed on fresh PBMCs isolated from heparin treated blood. In brief, Multiscreen HA plates (Millipore, Billerica, MA) were coated with mouse anti-human IFN-γ (Pharmingen) and incubated overnight at 4°C, washed with PBS 0.1% Tween, loaded with 2 × 105 PBMC per well, and 5 μg/ml of the peptide pool, in duplicate. Plates were incubated at 37°C in a CO2 incubator for 48 hours, washed, treated with a biotinylated anti-human IFN-γ (MabTech), and then developed using streptavidin-HRP (Pierce) and Stable DAB (Research Genetics). Spot forming cells (SFC) per million PBMC were determined by subtracting the average background value for each animal from the average of the duplicate wells and multiplying by five.
Viral load and CD4+T-cell counts
Plasma samples were separated and stored at -80°C until assayed. Plasma viral loads were quantified by the Bayer SIV bDNA assay (Bayer Reference Testing Laboratory, Emeryville, CA). Peripheral blood CD4+ T-cell concentrations were quantified using standard techniques, as previously described.
At two-week intervals following challenge plasma was obtained from animals for viral sequence analysis. RNA was extracted from plasma samples by Qiagen RNA Isolation kit. The Tat 2nd exon was amplified by reverse transcription followed by two rounds of nested PCR. Primers used were; 1st round forward – TGAGACTTGGCAAGAGTGG, 1st round reverse – GGACTTCTCGAATCCTCTGTAG, 2nd round forward – GGTATAGGCCAGTGTTCTCT, 2nd round reverse – TATCAGTTGGCGGATCAGGA. Second round PCR fragment was 173 bp in length and corresponded to SIVMac239 base pairs 8762 to 8934 (GenBank accession # M33262) Fragments amplified by PCR were TA-cloned by topoisomerase into pCR2.1Topo (Invitrogen). Sequencing was performed using M13-reverse primer.
- Smith SM, Singh M, Jeang KT: A vaccine for HIV / AIDS : challenges and progress. Encyclopedia of Molecular Cell Biology and Molecular Medicine. Edited by: Meyers RA. 2005, New York, WILEY-VCH VerlagGoogle Scholar
- Barouch DH, Kunstman J, Glowczwskie J, Kunstman KJ, Egan MA, Peyerl FW, Santra S, Kuroda MJ, Schmitz JE, Beaudry K, Krivulka GR, Lifton MA, Gorgone DA, Wolinsky SM, Letvin NL: Viral escape from dominant simian immunodeficiency virus epitope-specific cytotoxic T lymphocytes in DNA-vaccinated rhesus monkeys. J Virol. 2003, 77: 7367-7375. 10.1128/JVI.77.13.7367-7375.2003.PubMed CentralView ArticlePubMedGoogle Scholar
- Allen TM, Jing P, Calore B, Horton H, O'Connor DH, Hanke T, Piekarczyk M, Ruddersdorf R, Mothe BR, Emerson C, Wilson N, Lifson JD, Belyakov IM, Berzofsky JA, Wang C, Allison DB, Montefiori DC, Desrosiers RC, Wolinsky S, Kunstman KJ, Altman JD, Sette A, McMichael AJ, Watkins DI: Effects of cytotoxic T lymphocytes (CTL) directed against a single simian immunodeficiency virus (SIV) Gag CTL epitope on the course of SIVmac239 infection. J Virol. 2002, 76: 10507-10511. 10.1128/JVI.76.20.10507-10511.2002.PubMed CentralView ArticlePubMedGoogle Scholar
- Mothe BR, Horton H, Carter DK, Allen TM, Liebl ME, Skinner P, Vogel TU, Fuenger S, Vielhuber K, Rehrauer W, Wilson N, Franchini G, Altman JD, Haase A, Picker LJ, Allison DB, Watkins DI: Dominance of CD8 responses specific for epitopes bound by a single major histocompatibility complex class I molecule during the acute phase of viral infection. J Virol. 2002, 76: 875-884. 10.1128/JVI.76.2.875-884.2002.PubMed CentralView ArticlePubMedGoogle Scholar
- Smith SM: HIV CTL escape: at what cost?. Retrovirology. 2004, 1: 8-10.1186/1742-4690-1-8.PubMed CentralView ArticlePubMedGoogle Scholar
- Smith SM, Pentlicky S, Klase Z, Singh M, Neuveut C, Lu CY, Reitz MSJ, Yarchoan R, Marx PA, Jeang KT: An in Vivo Replication-important Function in the Second Coding Exon of Tat Is Constrained against Mutation despite Cytotoxic T Lymphocyte Selection. J Biol Chem. 2003, 278: 44816-44825. 10.1074/jbc.M307546200.View ArticlePubMedGoogle Scholar
- Nestle FO, Farkas A, Conrad C: Dendritic-cell-based therapeutic vaccination against cancer. Curr Opin Immunol. 2005, 17: 163-169. 10.1016/j.coi.2005.02.003.View ArticlePubMedGoogle Scholar
- Pope M: Dendritic cells as a conduit to improve HIV vaccines. Curr Mol Med. 2003, 3: 229-242. 10.2174/1566524033479870.View ArticlePubMedGoogle Scholar
- Reinhard G, Marten A, Kiske SM, Feil F, Bieber T, Schmidt-Wolf IG: Generation of dendritic cell-based vaccines for cancer therapy. Br J Cancer. 2002, 86: 1529-1533. 10.1038/sj.bjc.6600316.PubMed CentralView ArticlePubMedGoogle Scholar
- Alexander L, Denekamp L, Czajak S, Desrosiers RC: Suboptimal nucleotides in the infectious, pathogenic simian immunodeficiency virus clone SIVmac239. J Virol. 2001, 75: 4019-4022. 10.1128/JVI.75.8.4019-4022.2001.PubMed CentralView ArticlePubMedGoogle Scholar
- Otting N, Heijmans CM, Noort RC, de Groot NG, Doxiadis GG, van Rood JJ, Watkins DI, Bontrop RE: Unparalleled complexity of the MHC class I region in rhesus macaques. Proc Natl Acad Sci U S A. 2005, 102: 1626-1631. 10.1073/pnas.0409084102.PubMed CentralView ArticlePubMedGoogle Scholar
- Amara RR, Villinger F, Altman JD, Lydy SL, O'Neil SP, Staprans SI, Montefiori DC, Xu Y, Herndon JG, Wyatt L, Candido MA, Kozyr NL, Earl PL, Smith JM, Ma HL, Grimm BD, Hulsey ML, Miller J, McClure HM, McNicholl JM, Moss B, Robinson HL: Control of a mucosal challenge and prevention of AIDS by a multiprotein DNA/MVA vaccine. Science. 2001, 292 (5514): 69-74. 10.1126/science.1058915.View ArticlePubMedGoogle Scholar
- Dailey PJ, Zamround M, Kelso R, Kolberg J, Urea M: Quantitation of simian immunodeficiency virus (SIV) RNA in plasma of acute and chronically infected macaques using branched DNA (bDNA) signal amplification assay. Journal of Medical Primatology. 1995, 24: 209-Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.