Multi-Strain Virus-Host Dynamics from HIV to Phage
- Author(s): Weinberger, Ariel
- Advisor(s): Lecar, Harold
- et al.
This dissertation uses mathematical modeling to probe the causes and consequences of multi-strain virus-host coexistence from the applied public health realm in which a second strain of HIV accelerates human mortality to the basic science realm in which persisting, previously dominant viruses drive the evolution of immunological memory in single-celled Bacteria and Archaea. In both applications, population-scale models are built from the ground up, utilizing experimentally measured parameters of virus and host molecules interacting within a cell to predict how virus and host populations coevolve across time. Model predictions are shown to match time-series data of virus-host dynamics in human hosts in the case of HIV and in prokaryotic hosts in the case of phage. Further, both sets of models generate experimentally testable hypotheses, suggesting therapeutic interventions against two major drivers of human mortality: HIV and pathogenic, antibiotic-resistant bacteria.
The first area of application, and the focus of Chapters 2 and 3, is HIV. No one understands why, in about 50% of HIV infections in the West, a more deadly HIV strain emerges late in infection. The new strain, known as X4, differs from its predecessor, known as R5, because X4 only infects CD4+ T cells displaying the receptor CXCR4, while R5 only infects CD4+ T cells displaying the receptor CCR5. Due to the apparent health and anti-HIV immunity of the approximately 10% of Europeans lacking a functional CCR5 receptor, some researchers have touted anti-R5 therapy as an alternative to current anti-HIV drug cocktails. Chapter 2 uses simulations of a novel mathematical model to show how anti-R5 treatment alone may accelerate X4 emergence and resultant immunodeficiency. As an alternative, I show that CCR5 blockers may be more successful in combination with effective HAART therapy or, should they become available, CXCR4 blockers. Chapter 3 probes why X4 only emerges during late-stage HIV infection, showing how X4 persists for many years at low levels, avoiding competitive exclusion to the initially fitter R5 Virus, through X4's unique, low-level infection of naïve CD4+ T cells. In this chapter, I derive a minimal target-cell based model for dual R5, X4 HIV infection in which late-stage switches (bifurcations) to X4 autonomously occur. In this simplified model, an analytic switch condition is probed, allowing us to theoretically predict how different interventions modulate the time to X4 emergence, and providing a compelling explanation for why 50% of Western HIV patients never actually switch to X4 Virus.
In Chapter 4, the focus turns to understanding the evolution of adaptive antiviral immunity--such as the T cell immunity targeted by HIV--in its most elemental setting: single-celled prokaryotes possessing the CRISPR immune system. A novel mathematical model of virus-microbe coevolution is derived to understand why prokaryotes with CRISPR-encoded specific immunity conserve old immune sequences for thousands of microbial generations despite compact prokaryotic genomes with high DNA deletion rates and rapid viral mutation, which makes old CRISPR sequences far less likely to provide immunity against current viruses. Matching metagenomic reconstructions of CRISPR sequences across time in both bacterial and archaeal populations, the model shows how CRISPRs' immunological memory protects against measured blooms of persisting, low-abundance viral sequences. Thus, CRISPR may be the first immune system tuned viruses persisting through lysogeny or remigration.