Dissection of HIV-1 Env-specific B cell responses in non-human primates

Sammanfattning: It has been almost 30 years since the identification of HIV as the causative agent of AIDS. Despite considerable efforts to halt the spread of the virus, the epidemic continues in many parts of the world. The most efficient way to stop new infections is a prophylactic vaccine that blocks transmission at the viral portal of entry. To generate an efficacious vaccine, neutralizing antibodies directed against the envelope glycoproteins (Env), the only virally expressed proteins on the surface of the virion, likely need to be elicited. Early attempts to develop an HIV vaccine used soluble recombinant monomeric Env. This vaccine candidate generated high titers of Env binding antibodies in a large clinical trial; however, no protection against infection was observed. Since then attempts have been made to design Env immunogens that more closely mimic the native Env spike with the aim to improve the quality of vaccine-elicited humoral immune responses. Due to the lack of high resolution three dimensional structures of Env in its native conformation, these design efforts are still empirical. To guide future immunogen design efforts, it is important to understand what current Env vaccine candidates elicit in relevant pre-clinical models, both at the serological and cellular level. The focus of this thesis has been to investigate B cell responses in non-human primates following immunization with soluble HIV-1 Env trimers. We show that Env immunization in adjuvant elicits high levels of binding antibodies as well as robust peripheral memory B cell and plasma cell populations. The kinetics and magnitude of the responses at early time points after immunization are similar to those elicited by the successful human influenza and rabies vaccines. Despite these potent responses, we only observed a modest protective effect upon mucosal SHIV challenge of vaccinated animals. When quantifying the levels of Env-specific antibodies in the mucosa we found them to be 1000-10 000 fold lower than in serum, which may explain the lack of protection in this heterologous challenge model. From in vitro studies it has been proposed that interactions between Env and CD4-expressing T cells may detrimentally affect T cell function. To investigate if in vivo interactions between Env immunogens and CD4-expressing cells would negatively affect the elicitation of Envspecific immune responses, CD4 binding-competent and CD4 binding-defective trimers were constructed and inoculated into non-human primates. No difference was observed between the groups in regards to Env-specific antibody titers, memory B cell or T cell levels and functionality, indicating that Env binding to CD4 in vivo was not detrimental to vaccine elicited responses. Furthermore, using the CD4 binding-competent and CD4 binding-defective Env immunogens, we assessed if in vivo CD4 binding affected the outcome of an intravenous SHIV challenge. Antibodies directed against the conserved co-receptor binding site of Env were only observed following immunization with the CD4-binding competent trimers. However, the presence or absence of this class of antibodies did not influence the level of protection against SHIV infection. These results indicate that Env-CD4 in vivo interactions, and specifically co-receptor binding site-directed antibodies, did not contribute to the control of viremia observed in this challenge model.