Influence of mismatch of Env sequences on vaccine protection by live attenuated simian immunodeficiency virus

Abstract

Vaccine/challenge experiments that utilize live attenuated strains of simian immunodeficiency virus (SIV) in monkeys may be useful for elucidating what is needed from a vaccine in order to achieve protective immunity. Derivatives of SIVmac239 and SIVmac239Δnef were constructed in which env sequences were replaced with those of the heterologous strain E543; these were then used in vaccine/challenge experiments. When challenge occurred at 22 weeks, 10 of 12 monkeys exhibited apparent sterilizing immunity despite a mismatch of Env sequences, compared to 12 of 12 monkeys with apparent sterilizing immunity when challenge virus was matched in its Env sequence. However, when challenge occurred at 6 weeks, 6 of 6 SIV239Δnef-immunized monkeys became superinfected by challenge virus mismatched in its Env sequence (SIV239/EnvE543). These results contrast markedly not only with the results of the week 22 challenge but also with the sterilizing immunity observed in 5 of 5 SIV239Δnef-immunized rhesus monkeys challenged at 5 weeks with SIV239, i.e., with no mismatch of Env sequences. We conclude from these studies that a mismatch of Env sequences in the challenge virus can have a dramatic effect on the extent of apparent sterilizing immunity when challenge occurs relatively early, 5 to 6 weeks after the nef-deleted SIV administration. However, by 22 weeks, mismatch of Env sequences has little or no influence on the degree of protection against challenge virus. Our findings suggest that anti-Env immune responses are a key component of the protective immunity elicited by live attenuated, nef-deleted SIV.

Fig 1

Experimental design for the week 22 challenge study (A) and schematic representation of the recombinant proviral DNA constructs used for generating viral stocks (B). The numbers of animals in each group as well as the corresponding vaccine/challenge sequences are indicated in panel A. Recombinant viruses were obtained by exchanging sequences encompassing a region from the start codon of the env gene to an NheI site that is located at the junction of the region coding for the gp41 cytoplasmic domain (CD) in both SIVmac239 and nef-deleted SIVmac239 strains.

Fig 2

Multiple amino acid sequence alignment of envelope sequences from the pathogenic strains SIV239, SIVE543, and the recombinant strain of SIV239 bearing a SIVE543 sequence in the region encompassing from the beginning of the Env protein until the end of the transmembrane domain of gp41. Amino acid identities are represented with dots. The cytoplasmic domain of gp41 is indicated by a gray-shaded box.

Fig 3

Relative infectivities of the four viral stocks: SIV239, SIV239Δnef, SIV239/Env543, and SIV239/Env543Δnef. Viral stocks were generated by transfection of HEK293T cells with full-length proviral DNAs, and normalized amounts of p27 were used for subsequent assays. Normalized viral stocks were used to infect LTR-SEAP-CEMx174 indicator cells, and SEAP activity was measured on day 3 postinfection.

Fig 4

Viral loads observed along the 22 weeks of the vaccination phase of immunized animals, up to the day of challenge. (A) Animals vaccinated with SIV239Δnef (groups A and B); (B) animals vaccinated with SIV239Δnef/EnvE543 (groups C and D); (C) median viral loads of vaccinated groups: A and B (SIV239Δnef) and C and D (SIV239Δnef/EnvE543).

Fig 5

Neutralizing activity in sera taken on the day of challenge (week 22). Neutralization assays are described in Materials and Methods. Neutralization profile of animal sera taken 22 weeks after vaccination with SIV239Δnef (groups A and B), tested against p27 capsid protein-normalized stocks of SIV239 (A) and SIV239/EnvE543 (D). Neutralization profile of animal sera taken 22 weeks after vaccination with SIV239Δnef/Env543 (groups C and D), tested against p27 capsid protein-normalized stocks of SIV239/EnvE543 (C) and SIV239 (B), respectively.

Fig 6

SIV-specific IFN-γ ELISPOT responses in SIVmac239Δnef-and SIV239/EnvE543Δnef-vaccinated macaques (Mm). PBMC were obtained at the indicated time points, and SIV-specific responses to the indicated SIV peptide pool were calculated after subtraction of spots obtained in negative-control wells (R10). (A) Responses to peptide pools of SIVmac239 for 6 representative SIV239/EnvE543Δnef-vaccinated Mms at weeks 4 and 12. In all cases, the response at week −1 is indicative of background levels of reactivity. (B) Median responses to SIV peptide pools from Mms vaccinated with SIVmac239Δnef and SIV239/EnvE543Δnef at weeks 4 and 12 postinoculation. R10 is the negative control. Median week −1 represents background responses. (C) Average total responses of Mms vaccinated with SIVmac239Δnef and SIV239/EnvE543Δnef at indicated time points. (D) Median total responses from Mms vaccinated with SIVmac239Δnef and SIV239/EnvE543Δnef at indicated time points.

Fig 7

Postchallenge viral loads in animals infected with wild-type viruses and in control animals. Viral loads were determined as described in Materials and Methods. Unvaccinated control animals after challenge with SIV239 (A) and SIV239/Env543 (B). (C) Animals from group B, which became superinfected after challenge with SIV239/Env543 and the corresponding unvaccinated controls.

Fig 8

Timeline and summary of the outcome of the week 22 challenge. Twelve Indian-origin rhesus monkeys were vaccinated with SIV239Δnef (A and B), and 12 received SIV239Δnef/EnvE543 (C and D). After 22 weeks postvaccination, the animals were intravenously challenged with a virus that was either matched (A and C) or mismatched (B and D) with respect to envelope sequences. The table at the right side of the figure summarizes the results of the experiment: the pink-shaded animal numbers correspond to animals that became infected after challenge.

Fig 9

Experimental design for the week 6 challenge study. (A) The numbers of monkeys in each group as well as the corresponding vaccine/challenge sequences are indicated; (B) schematic representation of the recombinant proviral DNA constructs used for generating viral stocks; (C) trim and major histocompatibility complex (MHC) genotypes of the monkeys used in the study.

Fig 10

Outcome of week 6 challenge. (A, B) Viral load values determined at the challenge day (week 6) and at weeks 7, 8, 10, and 12 after challenge by determining either wild-type nef for groups G and I or wild-type E660/543 for groups H and J. Control viral load values for weeks 6, 7, and 8 were derived from the universal gag primers (black lines). Viral loads after week 8 in 65-09 did not decline when universal gag primers were used, suggesting the emergence of recombinant virus as a significant portion of the total.