Reversion In Vivo after Inoculation of a Molecular Proviral DNA Clone of Simian Immunodeficiency Virus with a Cytotoxic-T-Lymphocyte Escape Mutation

Abstract

Vaccine-based control of the replication of a simian immunodeficiency virus (SIV), SIVmac239, in macaques has recently been shown. In the process of the control, a mutant virus escaping from epitope-specific cytotoxic-T-lymphocyte (CTL) responses was rapidly selected and contained. In this study, we show that the wild-type virus appeared and became predominant in the absence of the epitope-specific CTL after inoculation of naive macaques with a molecular clone DNA of the CTL escape mutant SIV. This is the first report describing reversion in vivo from an inoculated, molecular proviral DNA clone of immunodeficiency virus with a CTL escape mutation.

Virus-specific CD8⁺ cytotoxic-T-lymphocyte (CTL) responses are crucial for the control of immunodeficiency virus infection. The importance of CTL has been indicated by temporal association of CTL appearance with the resolution of primary viremia in human immunodeficiency virus type 1-infected humans (4, 13) and by monoclonal anti-CD8 antibody-mediated CD8 depletion experiments with macaque AIDS models (10, 15, 23). Therefore, AIDS vaccine studies have been making efforts to develop methods efficiently eliciting virus-specific CTL responses (18). However, viral escape from CTL recognition can lead to viral evasion from immune control and has frequently been observed in immunodeficiency virus infection (1, 5, 8, 19, 21, 22). Under strong immune pressure exerted by CTL, viruses are often forced to mutate, with viral fitness costs, to escape from the CTL responses (7, 9, 11, 17, 20, 24). Some CTL escape mutant viruses with lower viral fitness require additional compensatory mutations to restore their replicative competence (6, 11, 20). It is important to evaluate replicative ability of CTL escape mutants in vivo.

Recently, CTL-based control of replication of a pathogenic simian immunodeficiency virus (SIV), SIVmac239 (12), has been shown in a preclinical vaccine trial using non-Indian rhesus macaques (17). In that study, macaques vaccinated with a DNA priming followed by a Gag-expressing Sendai virus vector-booster were challenged intravenously with SIVmac239. Five of eight vaccinees controlled viral replication and had undetectable levels of plasma viremia after 5 weeks of infection. All of the five macaques showed rapid selection of CTL escape mutations in gag, indicating that vaccine-induced CTL contained replication of the wild-type challenge virus. Among the five, three vaccinees that share a major histocompatibility complex class I (MHC-I) haplotype, 90-120-Ia, showed high levels of Gag_206-216 (IINEEAADWDL) epitope-specific CTL responses and rapid selection of a mutant escaping from the CTL. The replicative ability of the virus with the CTL escape mutation, Gag216S, leading to a substitution from leucine (L) to serine (S) at the 216th amino acid (aa) in Gag was diminished compared to the wild type. In the present study, we have observed replication of this mutant SIV, SIVmac239Gag216S, in the absence of Gag_206-216-specific CTL responses after its inoculation into naive macaques. All the animal experiments in this study were performed in accordance with the guidelines for laboratory animals of the National Institute of Infectious Diseases.

First, two cynomolgus macaques (Macaca fascicularis), C99049 and C99058, were coinoculated intramuscularly with 5 mg of the wild-type SIVmac239 molecular clone DNA (pBRmac239) and 5 mg of the mutant SIVmac239Gag216S molecular clone DNA (pBRmac239Gag216S) (17). We extracted RNA from plasma and quantitated plasma SIV RNA levels as described previously (17); both that of the wild type and that of the mutant are detectable in this assay. In both of the animals, plasma viremia was observed after the inoculation (Fig. 1A). Both the wild type and mutant viral genomes were detected at comparable levels by sequencing of a gag gene fragment amplified by reverse transcription and nested PCR from plasma RNA at week 1, but the mutant was poorly detected and the wild type was dominant at weeks 2 and 3. We then subcloned the amplified fragments into plasmids for sequencing and counted the numbers of clones carrying the wild-type (Gag216L) or the mutant (Gag216S) sequence at the region encoding the 216th aa in Gag. It revealed that the wild-type SIV became dominant 2 or 3 weeks after the inoculation (Fig. 1B). This result indicates that the replicative ability of this CTL escape mutant virus was diminished compared to that of the wild type, confirming the previous results obtained with rhesus macaques (17). After euthanasia of macaques at week 3, we extracted genomic DNA, by using the DNeasy kit (QIAGEN K.K., Tokyo, Japan), from the submandibular lymph node (LN), the mesenchymal LN, and the inguinal LN and subcloned proviral gag gene fragments amplified by nested PCR from the DNA into plasmids for sequencing. The mutant Gag216S was detected in most of the LNs, but the wild-type sequence was dominant in the proviral genomes in all the LNs (Fig. 1C).

FIG. 1.

Comparison of replication efficiencies in vivo of the wild-type SIVmac239 and the CTL escape mutant SIVmac239Gag216S. Macaques C99049 (left panels) and C99058 (right panels) were coinoculated with both wild-type and mutant SIV molecular clone DNAs. (A) Plasma viral loads after the inoculation. These indicate the sums of the wild-type and the mutant SIV RNA levels. The lower limit of detection in this assay is about 4 × 10² copies/ml. (B) Frequencies of the wild-type and mutant viral genomes in plasma. At each time point, a gag gene fragment was amplified by nested reverse transcription-PCR from plasma RNA and subcloned into plasmids for sequencing. The numbers of clones carrying the wild-type sequence (wt clones) and the mutant sequence (mt clones) at the region encoding the 216th aa in Gag are shown. (C) Frequencies of the wild-type and the mutant proviral genomes in the submandibular (SM) LN, the mesenchymal (Mes) LN, and the inguinal (Ing) LN. Genomic DNAs were extracted from LNs at euthanasia, and gag gene fragments amplified by nested PCR from the DNAs were subcloned into plasmids for sequencing.

Second, two cynomolgus macaques, C87072 and C87134, were inoculated intramuscularly with 5 mg of the mutant SIVmac239Gag216S molecular clone DNA alone. Plasma viremia was maintained until euthanasia of macaques at week 12 (C87072) or week 10 (C87134) after the inoculation (Fig. 2A). We subcloned viral gag gene fragments amplified from plasma RNA and found the wild-type Gag216L sequence at week 5, although it was undetectable at week 4 in both of the animals (Fig. 2B). In macaque C87072, 9 of 10 viral gag clones showed the wild-type Gag216L sequence at week 7 (Fig. 2B), and most of the proviral gag clones were the wild type in the submandibular LN, the mesenchymal LN, and the inguinal LN at week 12 (Fig. 2C). In macaque C87134, the ratio of wild-type to total viral gag clones was 3/11 at week 8 but 12/21 at week 10 (Fig. 2B). The wild-type Gag216L sequence was detected but was not predominant in the proviral gag clones from the LNs at week 10 (Fig. 2C). These results indicate that the mutant SIVmac239Gag216S proliferated in all the LNs but was outgrown by the wild-type virus.

FIG. 2.

Reversion in vivo from the CTL escape mutant molecular clone. Macaques C87072 (left panels) and C87134 (right panels) were inoculated with the mutant SIV molecular clone DNA. (A) Plasma viral loads after the inoculation. (B) Frequencies of the wild-type and mutant viral genomes in plasma. See the legend for Fig. 1B. (C) Frequencies of the wild-type and the mutant proviral genomes in LNs. See the legend for Fig. 1C.

We further examined virus-specific CD8⁺ T-cell responses in macaques by flow cytometric analysis of antigen-specific interferon-γ (IFN-γ) induction as described previously (16, 17). In brief, peripheral blood mononuclear cells (PBMCs) were cocultured with autologous herpesvirus papio-immortalized B lymphoblastoid cells infected with a vesicular stomatitis virus G-pseudotyped SIV for SIV-specific stimulation. Alternatively, PBMCs were cocultured with B lymphoblastoid cells pulsed with the Gag_206-216-epitope peptide for Gag_206-216-specific stimulation or a mixture of the peptides with a 216S mutation corresponding to the 206th through 220th aa and the 210th through 225th aa in Gag (Gag_206-220.216S and Gag_210-225.216S peptides, respectively) for Gag_206-225.216S-specific stimulation. PBMCs derived from macaques C99049 at week 3, C99058 at week 3, C87072 at week 12, and C87134 at week 10 were subjected to this assay (Fig. 3). SIV-specific CD8⁺ T-cell responses were detected but Gag_206-216-specific IFN-γ induction was undetectable, confirming no Gag_206-216-specific CTL responses in any of four animals. Gag_206-225.216S-specific CD8⁺ T-cell responses were also undetectable, indicating that the predominance of the wild-type virus in these four macaques was not due to immune pressure exerted by the mutant-specific CTL recognizing an epitope with the mutant Gag216S sequence.

FIG. 3.

SIV-specific CD8⁺ T-cell responses in macaques. We examined IFN-γ induction after SIV-specific stimulation using vesicular stomatitis virus G-pseudotyped SIV (A), Gag_206-216-specific stimulation using Gag_206-216 peptide (B), or Gag_206-225.216S-specific stimulation using a mixture of Gag_206-220.216S and Gag_210-225.216S peptides (C). The aa sequences of Gag_206-216, Gag_206-220.216S, and Gag_210-225.216S peptides are indicated. Dot plots gated on CD3⁺CD8⁺ lymphocytes are shown. PE, phycoerythrin; PerCP, peridinin chlorophyll protein.

In the previous study (17), the Gag216S mutant virus escaping from Gag_206-216-specific CTL was rapidly selected in the vaccinees possessing the MHC-I haplotype 90-120-Ia after SIVmac239 challenge. However, the CTL escape mutant with lower viral fitness was rapidly contained and became undetectable in plasma after week 5 postchallenge. The present study shows that this mutant SIV, which was rapidly contained in the vaccinees in the previous study, can replicate and is unable to be rapidly contained in naive macaques, leading to the appearance of the wild-type virus in the absence of Gag_206-216-specific CTL responses. This suggests the requirement of additional adaptive immune responses as well as Gag_206-216-specific CTLs for containment of this CTL escape mutant virus with lower viral fitness.

Viral adaptation by escape mutations under CTL pressure and reversion after transmission to MHC-I-mismatched hosts have been indicated in immunodeficiency virus infection (2, 3, 7, 14). It has recently been shown that reversion by de novo mutation can really occur after challenge of macaques with a cloned SIV with CTL escape mutations (7). In that study, preparation of the challenge virus stock from a molecular clone DNA of the mutant SIV required viral replication in vitro for more than a week. In the present study, to see the reversion by de novo mutation only in vivo by deleting the in vitro replication process for virus stock preparation, we directly inoculated macaques with a molecular clone DNA of the mutant SIV. Our results show that the reversion by de novo mutation can really occur and be detected in 5 weeks after inoculation of the mutant molecular clone DNA. Thus, this is the first report describing the reversion in vivo from an inoculated, molecular proviral DNA clone of immunodeficiency virus with a CTL escape mutation.