Impact of HIV-1 Genetic Variation on Tat-RNA Binding, Anti-HIV-1 gRNA Design, and HIV-Associated NI
Friday, September 10, 2021
2:00 PM-4:00 PM
BIOMED PhD Thesis Defense
Title:
Impact of HIV-1 Genetic Variation on Tat-RNA Binding, Anti-HIV-1 gRNA Design, and HIV-Associated Neurocognitive Impairment (NI)
Speaker:
Robert William Link, PhD Candidate
School of Biomedical Engineering, Science and Health Systems
Drexel University
Advisors:
Will Dampier, PhD
Assistant Professor
Department of Microbiology and Immunology
Drexel University College of Medicine
Brian Wigdahl, PhD
Professor and Chair
Department of Microbiology and Immunology
Director
Institute for Molecular Medicine and Infectious Disease
Drexel University College of Medicine
Details:
Human immunodeficiency virus type 1 (HIV-1) infection is a major public health threat due to the formation of a latent reservoir that cannot be removed by conventional antiretroviral therapies. Infected cells secrete viral proteins which trigger the release of proinflammatory cytokines that stimulate chronic inflammation. This causes comorbidities such as an increased risk of developing various cancers and HIV-1-associated neurocognitive disorders (HAND), necessitating the continued search for a cure. Each patient harbors a variable and continuously evolving collection of viruses – referred to as viral quasispecies (vQS) – which allows HIV-1 to rapidly adapt to selective pressures and develop resistance against therapeutics and cure strategies.
To account for this variability, the composition of each patient’s vQS profile needs to be examined. Nanopore sequencing yields long reads that encompass the entire proviral genome, making it optimal for vQS characterization. However, Nanopore sequencing was error prone when first released, preventing widespread application. This was the inspiration behind the development of the HIV-Quasipore suite: a set of fast and high accuracy, HIV-specific, Nanopore basecallers designed to accurately characterize vQS. These were constructed using pretrained flip-flop Nanopore basecallers that were refined using J-Lat 10.6 cell line Nanopore trace data. This was performed by minimizing the error between the base-called reads to the true sequence found in a high-quality Sanger-sequenced J-Lat 10.6 genome. Base-calling improvement was consistent across the genome and has decreased the coverage required to confidently determine the presence of a vQS.
HIV-1 Tat variability has been implicated in HAND presentation and severity. Given this, the impact of HIV-1 Tat length on clinical parameters and neurocognitive impairment were investigated within the Drexel Medicine CNS AIDS Research and Eradication Study (CARES) cohort. Tat variants (101 and 86 residue in length) were predominantly observed within the cohort, matching observations found within Los Alamos National Laboratory’s (LANL) HIV-1 sequence database. However, unique Tat length isoforms were found within the CARES cohort that were not present within LANL. Regardless, Tat length did not correlate with CARES participant clinical parameters or neurocognitive impairment status.
While investigating the influence of Tat’s length on HAND is relatively simple, examining how specific variants influence HAND presentation and severity requires powerful computational inference. Give these observations, an optimized RNA-protein interaction (RPI) prediction model architecture and a novel negative sampling method leveraging stacked autoencoder difference magnitudes were developed to investigate how Tat variability impacts HAND severity through changes in Tat-microRNA (miRNA) interactions. When using the model to examine the Tat-specific variation’s contribution to clinical parameters, no distinct clusters could be observed encompassing any clinical parameter or neurocognitive assessment scores. When all collected Tat-miRNA combinations were passed through the model, most Tat variants were found to exclusively bind or not bind with all miRNAs, regardless of wild type Tat behavior. Tat-binder status did not significantly impact assessment-specific impairment. However more HIV-1 RPIs are required to improve Tat-miRNA prediction performance.
Variation also needs to be considered when designing anti-HIV-1 LTR-targeting guide RNAs (gRNAs) and inferring any off-target cleavage. Investigating gRNAs derived from different subtypes and groups resulted in few inferred high-quality off-target effects. Those off-target effects were predominantly located in non-coding DNA. Inferred cleavage did not frequently arise from human endogenous retrovirus (HERV)-K 5’ LTRs and were equally distributed across all chromosomes. While results are promising, all gRNA off-target events need to be validated with established off-target detection assays (e.g., GUIDE-Seq, CIRCLE-Seq).
Overall, the work in this thesis has led to advances in computational and biologic understanding of HIV-1 pathogenesis during the course of HIV-1 disease in patient samples and underlying mechanisms needed for understanding HIV pathogenesis and CNS disease and cutting-edge therapeutics.
Contact Information
Natalia Broz
njb33@drexel.edu