Antibody repertoire profiling during pregnancy and proteome-wide internal degron mapping

Abstract

Recognition of specific sequence motifs is a fundamental aspect of detecting and triggering response to environmental stimuli in cellular organisms. This phenomenon is utilized for a diverse range of biological applications. Among others, two critical examples of motif-recognizing systems in vertebrate biology are the ubiquitin proteasome system and adaptive immunity. These systems serve distinct roles: the ubiquitin proteasome system degrades proteins intracellularly, while adaptive immunity works both intracellularly and extracellularly to respond to pathogens. Despite differences in biological purpose, these two systems both require precise detection of surrounding proteins in a highly tuned, sequence-specific manner. In adaptive immunity, the receptors of B and T cells develop tremendous diversity at their antigen binding sites through V(D)J recombination. In B cells, this is later refined to precisely target a specific antigen through affinity maturation. The number of distinct antibodies specificities achievable from V(D)J recombination alone is estimated to be between 1012 and 1014. In the ubiquitin proteasome system, proteins are marked for degradation by poly-ubiquitination, a process mediated by an elegantly orchestrated enzymatic cascade. One class of enzymes involved in this process are E3 ubiquitin ligases, which recognize specific sequences called degrons on their substrate and bring additional ubiquitination machinery into proximity to effect transfer of ubiquitin to the target. There are over 600 E3 ubiquitin ligases encoded in the human genome, and the cognate degron recognized by a particular E3 is only known for a handful of ligases. In the first part of this dissertation, we performed longitudinal, high-throughput serological profiling of the antibody repertoire 878 samples from 119 individuals over the course of their pregnancies. We employed orthogonal technologies to obtain data on antibodies targeting both pathogens and self-antigens, linear and conformational epitopes, and of multiple isotypes. We observed that the diversity of the anti-pathogen antibody repertoire broadens over the course of pregnancy, that the titer of anti-pathogen IgA antibodies increases over the course of pregnancy when compared to IgG or IgM, and that sharp spikes in certain antiviral IgA titers tend to increase just prior to miscarriage. As further characterization of the spikes in antiviral IgA titers that we observed just prior to miscarriage, we then identified an association between this phenomenon and an IgA-specific broadening of the antibody repertoire against self-antigens, and an increase in IgA targeting class-I HLA. In the second part of this dissertation, we perform systematic identification and mapping of critical residues of internal degrons across the human proteome. We identified 15,800 peptide sequences likely to contain sequence-dependent degrons and used scanning mutagenesis to map critical residues for over 5000 predicted degrons. We then developed fine motif fingerprints for 219 degrons using site saturation mutagenesis and implemented a custom algorithm called DegronID for clustering of degrons with similar motifs. We then used CRISPR to discover E3-degron pairs, modeled interactions using AlphaFold, and performed experimental validation of the residues critical for binding, identifying diverse sequence and structural determinants mediating 16 novel E3-degron interactions.