Design of replicating VSV vectors expressing HIV Envelope trimers and Preclinical Evaluation of Immune Responses Elicited by Candidate Vaccines.

Abstract

To prevent sexual transmission of HIV, a successful vaccine must elicit protective immune responses and durable immunologic memory that can respond rapidly upon virus exposure at mucosal sites. Thus, replication competent vesicular stomatitis virus (VSV) vectors are being investigated for HIV vaccine delivery with particular focus on the effect of delivery route on immune responses. VSV is a practical vector candidate because it naturally enters through mucosal sites, infects multiple cell types, and is highly immunogenic. Moreover, VSV typically does not cause disease in humans and pre-existing immunity is rare. Attenuated, replication-competent VSV vectors have been designed to co-express HIV Envelope (Env) trimers (from HIV-1 subtype B, strain JRFL) and the natural VSV attachment protein (G) derived from the Indiana (VSV-Env-GIN) or New Jersey serotype (VSV-Env-GNJ). To optimize Env trimer expression, several insert designs have been tested utilizing VSV or CD5 (human T-cell surface glycoprotein) signal sequences and different combinations of Env and VSV-G transmembrane (TM) and intracellular domains. Improved Env expression and vector genetic stability was observed when the CD5 signal and VSV-G TM sequences were used. Infected Vero cells were used to analyze cell surface expression and the antibody-binding profile of the Env trimer immunogen. Infected cells that expressed VSV G on their surface also expressed Env which was recognized by Env-specific broadly neutralizing antibodies b12, 2F5, 2G12, PGV04 and PGT126. Furthermore, the cells produced Env in a conformation that bound the antibody PGT145, which preferentially binds conformational determinants on trimeric Env. These results demonstrate that VSV can express membrane-bound trimeric Env in a conformation containing epitopes recognized by known broadly neutralizing antibodies. To investigate cellular and humoral responses elicited by vectors expressing HIV Env trimers, a heterologous prime-boost regimen was conducted in Balb/c mice with plasmid DNA prime and rVSV boost. DNA plasmids expressing EnvG were administered by intramuscular (IM) injection and electroporation (EP) with plasmid expressing IL-12 (pIL-12) at week 0 and 3 followed by a VSV-Env-GIN infection at week 6. The effect of administering the VSV boost by different routes also was investigated. A second study based on VSV-Env-GIN prime/VSV-Env-GNJ boost with both vectors administered by intranasal (IN) or IM route in different combinations administered at weeks 0 and 3 was used to study the immunogenicity of the two vectors without a DNA prime. Immune responses were analyzed by flow cytometry, ELISA, and competition binding assay at various times after the immunizations. Both studies elicited strong Env-specific antibody responses in the serum. pDNA+pIL-12 prime/ rVSV boost elicited Env-specific CD4+ T cells in the spleen and lung for the IN route. In the same pDNA+pIL-12 prime/ rVSV boost study, Env-specific CD8+ T cells were detected in the lungs for the DNA and IN groups. Notably, the IN route was able to elicit b6 and PGV04-like antibodies. These findings are of great interest as they seem to underline the importance of the quality of the elicited antibody response, which cannot be readily ascertained by measurement of the quantity of Env-specific antibodies. The titer of Env-specific antibodies has been used as a widespread measurement of vaccine efficacy in the HIV vaccine field. The results from the current studies point to the need for the characterization of these antibody responses and perhaps the quality of the antibody response should be used as the standard measurement of vaccine efficacy. The viral prime and viral boost study also elicited high Env-specific antibody responses in the serum but had a different cell-mediated response profile in that both CD4+ and CD8+ T cells were detected in the lungs but none in the spleen. The IM-IN route was able to elicit b6-like antibodies. The outcome of these studies highlights that VSV vectors expressing EnvG can elicit Env-specific humoral and cell-mediated immune responses in two different vaccine regimens. Even more importantly, these VSV-EnvG vectors were able to elicit antibodies similar to known broadly neutralizing antibodies. These findings give tremendous support to the use of EnvG as an immunogen in a live-replicating vector platform for future HIV vaccines.