Abstract
Over the past several decades, the development of a cost-effective HIV-1 vaccine has faced various significant challenges. All currently approved antiviral vaccines work by eliciting protective antibody responses. These antibodies, to be an effective force against HIV-1 infection, must impede the entry process by binding to the viral envelope (Env) glycoprotein. This glycoprotein plays a critical role in facilitating the attachment of the virus to target cells via the cellular CD4 receptor and enabling membrane fusion. HIV-1 has developed varied strategies to hinder the potential for neutralizing antibody (nAb) mediated inhibition, such as masking of critical epitopes and rapid incorporation of escape mutations. This interplay between antibody and Env was the main focus of my thesis. Understanding escape patterns of viruses in response to antibody selection pressures is crucial to successfully develop protective vaccines that are able to reliably induce broadly neutralizing antibodies (bnAbs) in recipients. In the first project of my thesis, my colleagues and I investigated the effect of changes in structural domains of the Env protein and their impact on the neutralization capacities of nAbs. We conducted a comprehensive analysis of 126-point mutations on the tier 2 strain JR-CSF using weak nAbs and plasma from chronically infected people with HIV (PWH). Consequently, we were able to develop 19 mutants into a Sensitivity Env mutant panel (SENSE-19), with which we categorized conformation dependency of antibodies and DARPins based on their SENSE-19 score in combination with their breadth and potency. The results from SENSE-19 screenings showed varying degrees of tolerance to Env protein flexibility among different types of bnAbs, offering deeper understanding of their mechanisms and enhancing the scope of existing assessments of neutralization breadth. In addition, we verified these results in different PBMC-based neutralization formats, widening the avenues with which SENSE-19 could be employed in. In the second section of my dissertation, I developed a Primary HIV Isolate Library (PHILib) designed for infecting PBMC target cells in presence and absence of bnAbs in a neutralization format. Based on their known envelope glycoprotein sequence, isolates included in the library are distinguished via a naturally occurring ID region between the LTR and gag. Using MiSeq technology, sequencing of the ID region from extracted viral DNA 12h post infection directly provides information on sensitive and resistant virus isolates in the library against a tested bnAb. This allows for subsequent variant calling of their Env sequence data to determine resistance patterns. By first improving and subsequently assessing consensus sequences of isolates used in PHILib, I was able to construct a library consisting of 187 distinct isolates. My assay could be used to supplement results stemming from pseudovirus assays and offer insights into bnAb neutralization in more in vivo-like settings. In my third project, I developed a mutant virus library building on a highly unique Env isolated from the recipient in a transmission pair where both the transmitter and recipient developed highly similar CD4 binding site (CD4bs) directed immune responses. Previous work had highlighted the potential for this identified Env protein (LimE.282.G1) to harbor imprinting capacity and emerged as an ideal immunogen candidate. I provide a comprehensive report on challenges encountered in creating the mutant library, detailing the complexities involved, the obstacles I faced, and strategies to overcome them. I demonstrate that, while the initial version of the mutant library is in principle functional, the LimE library I generated will require protocol measures to enhance infectivity before it can be fully employed in resistance fingerprinting experiments. In summary, in this dissertation I established novel tools with which the Env-antibody interplay can be assessed and evaluated. These methods offer to supplement established neutralization and mapping assays and could provide valuable insights into bnAb classifications.
Foulkes, Caio