Transcribing the dynamic multicellular immune orchestra during acute HIV infection

Abstract

The development of novel vaccines and therapeutics requires comprehensive understanding of the immune system and its functional responses in health and disease. For infections, blueprinting the complex immune response during the initial stages of disease is essential. Human immunodeficiency virus-1 (HIV-1) infection is a model for studying host-pathogen interactions and has led to the development of several far-reaching concepts in immunology and infectious disease (e.g. antibody affinity maturation, host-derived pathogen sensors, etc.). Thus, exploring primary HIV-1 infection could not only impact the advancement of HIV-specific vaccines and treatments, but also serve as a model for early immune response in other viral infections. Utilizing a unique prospective cohort of young women at high risk of contracting HIV, we have begun to profile the immune response to HIV infection at its earliest detectable timepoints.

To maximize the utility of these rare samples from the FRESH study, we applied bulk and single-cell RNA-sequencing (RNA-seq) approaches to characterize changes in cellular phenotype as a function of time during acute infection. Specifically, we first characterized changes to peripheral innate lymphoid cells (ILCs), which are irreversibly depleted in acute infection. ILCs express gene programming associated with apoptosis and cell death near peak peripheral viremia, suggesting that they are depleted throughout the body. Second, we profiled HIV-specific CD8⁺ T cells, comparing between early treated and untreated participants. We show strong transcriptional responses near peak viremia consisting of broad cellular activation and cytotoxic activity that was mitigated by early treatment. Cells from treated participants demonstrated higher levels of anti-apoptotic markers and displayed long-lasting memory phenotypes.

Finally, we applied single-cell RNA-seq to total PBMCs from four individuals in FRESH throughout the course of acute infection. We developed a novel computational framework to discover gene modules significantly varying in expression as a function of time, enabling us to link distinct cellular activity between cell subsets. Moreover, we identify early subsets of monocytes and NK cells that associate with future disease control. These transcriptomic approaches have allowed an unprecedented view into the cellular dynamics of infection response, corroborating and contextualizing flow-cytometry and in vitro culture experiments. Together this body of work broadens our understanding of the first moments of HIV infection on a cellular and molecular level, highlighting cell-subsets and signaling pathways for perturbation in future vaccines and treatments.

Moreover, we pioneer the application of bulk and single-cell transcriptomics to longitudinal infection data on the days-to-weeks timescale, providing approaches and tools for others to apply to new datasets and studies in humans and other model organisms.