CD8 T Cells Show Protection against Highly Pathogenic Simian Immunodeficiency Virus (SIV) after Vaccination with SIV Gene-Expressing BCG Prime and Vaccinia Virus/Sendai Virus Vector Boosts

Abstract

Toward development of a dual vaccine for human immunodeficiency virus type 1 (HIV-1) and tuberculosis infections, we developed a urease-deficient bacillus Calmette-Guérin (BCG) strain Tokyo172 (BCGΔurease) to enhance its immunogenicity. BCGΔurease expressing a simian immunodeficiency virus (SIV) Gag induced BCG antigen-specific CD4⁺ and CD8⁺ T cells more efficiently and more Gag-specific CD8⁺ T cells. We evaluated its protective efficacy against SIV infection in cynomolgus monkeys of Asian origin, shown to be as susceptible to infection with SIVmac251 as Indian rhesus macaques. Priming with recombinant BCG (rBCG) expressing SIV genes was followed by a boost with SIV gene-expressing LC16m8Δ vaccinia virus and a second boost with SIV Env-expressing Sendai virus. Eight weeks after the second boost, monkeys were repeatedly challenged with a low dose of SIVmac251 intrarectally. Two animals out of 6 vaccinees were protected, whereas all 7 control animals were infected without any early viral controls. In one vaccinated animal, which had the most potent CD8⁺ T cells in an in vitro suppression activity (ISA) assay of SIVmac239 replication, plasma viremia was undetectable throughout the follow-up period. Protection was confirmed by the lack of anamnestic antibody responses and detectable cell-associated provirus in various organs. Another monkey with a high ISA acquired a small amount of SIV, but it later became suppressed below the detection limit. Moreover, the ISA score correlated with SIV acquisition. On the other hand, any parameter relating anti-Env antibody was not correlated with the protection.IMPORTANCE Because both AIDS and tuberculosis are serious health threats in middle/low-income countries, development of a dual vaccine against them would be highly beneficial. To approach the goal, here we first assessed a urease-deficient bacillus Calmette-Guérin (BCG) for improvement of immunogenicity against both Mycobacterium tuberculosis and SIV. Second, we demonstrated the usefulness of Asian-origin cynomolgus monkeys for development of a preclinical AIDS vaccine by direct comparison with Indian rhesus macaques as the only validated hosts that identically mirror the outcomes of clinical trials, since the availability of Indian rhesus macaques is limited in countries other than the United States. Finally, we report the protective effect of a vaccination regimen comprising BCG, the highly attenuated vaccinia virus LC16m8Δ strain, and nontransmissible Sendai virus as safe vectors expressing SIV genes using repeated mucosal challenge with highly pathogenic SIVmac251. Identification of CD8⁺ T cells as a protective immunity suggests a future direction of AIDS vaccine development.

Keywords: CTL; HIV; SIV; human immunodeficiency virus; recombinant BCG; recombinant vaccinia; simian immunodeficiency virus.

FIG 1

Pathogenicity of various BCG strains in SCID mice. SCID mice were inoculated intravenously with 10⁷ CFU of each BCG, which was calculated by a calibration curve between the OD of 7H9-ADC liquid culture of the BCG strains and the numbers of colonies on a Middlebrook 7H10 agar plate containing oleic albumin dextrose catalase supplement (Becton, Dickinson and Co., Tokyo, Japan). (a) Survival of SCID mice uninfected (●) or infected with the Danish (□), Pasteur (△), or Tokyo172 (▾) strain of BCG (n = 10 for each). (b) Body weights of SCID mice over time after BCG infection. (c) The numbers of living BCG that persisted in the lungs and spleens of SCID mice. Photographs of spleens at 8 weeks after infection show splenomegaly of Danish and Pasteur strain-infected SCID mice. (d) Hematoxylin-and-eosin (HE)-stained paraffin sections of mouse spleens prepared at 8 weeks after infection.

FIG 2

Pathogenicity and immunogenicity of urease-deficient BCG Tokyo172 (BCGΔurease) harboring plasmid bearing the SIV gag gene. (a) Body weights of SCID mice over time after wild-type BCG (▴), BCGΔurease (●), or mock (◆) infection (n = 5 for each). SCID mice were inoculated intravenously with 10⁷ CFU of each BCG. One of the tubes containing frozen BCG was molten, and the number of CFU was measured by the colony assays. (b) Numbers of living BCG that persisted in the spleens and lungs of SCID mice at 8 weeks after infection. (c) ELISPOT assay of splenocytes prepared from C57BL/6J primed with BCG-SIV-Gag and boosted with m8Δ-SIV-Gag.

FIG 3

Vaccination of cynomolgus monkeys and challenge with SIVmac251. (a) Schematic vaccination and challenge protocols. Arrows represent time of vaccination or challenges. s.c., subcutaneous; i.n., intranasal; IR, intrarectal. (b) Outcome of repeated, limiting-dose intrarectal SIVmac251 challenge. Upper and lower panels show plasma viral loads of vaccinated and control monkeys, respectively. (c) The rates of SIVmac251 acquisition per exposure were compared using the log rank test of the discrete-time proportional hazards model, and viral loads (VL) at peak and set point (17 weeks after the first challenge, except 13 weeks after the first challenge for two control monkeys) were compared using the Mann-Whitney test. (d) Proviral DNA loads in lymphoid tissues prepared at euthanasia corresponding to the last time point of viral load measurements were detected by PCR. Information for the monkeys used in this experiment, including sex, age, and MHC typing data, is given in Table 3.

FIG 4

Quantification of anti-Env (gp140) antibodies elicited in the six vaccinated monkeys (monkeys 1, 7, 8, 9, 10, and 11) during the vaccination phase (BCG, VV, and SeV) and after challenge (cha) with SIVmac251. Numbers in the figure show weeks after vaccination or the first challenge. Anti-gp140 antibody titers in the plasma were assessed by ELISA.

FIG 5

Association between in vitro suppression activity (ISA) at first challenge and rate of infection acquisition in immunized animals. Using the two-tailed Spearman rank test, the rate of infection acquisition was found to significantly correlate with the ISA score at first challenge. The graph is plotted as if the protected monkey was infected at the sixth challenge.

FIG 6

Intracellular staining (ICS) for cytokines in CD4 and CD8 T cells among PBMC from selected monkeys (monkeys 9, 10, and 11) prepared at 4 weeks after boosting with SeV. PBMC were incubated with SIV Gag or SIV Env peptides.