UCSF

cGMP is a ubiquitous second messenger implicated in a multitude of neurobiological processes, including sensory transduction, learning and memory. FRET-based and GFP-based cGMP reporters have been developed to complement the genetic and biochemical tools used to probe the role of cGMP in these processes. While FRET-based cGMP sensors have been expressed in C. elegans to explore the spatiotemporal regulation of cGMP in sensory neurons in response to stimuli, their use requires a dual emission system, which limits their ability to be coexpressed with other fluorophores, such as red calcium sensors that can indirectly read out neural activity. This work demonstrates that WincG2, a GFP-based cGMP sensor codon-optimized for use in C. elegans, can report changes in cGMP levels in living, behaving C. elegans. We demonstrate that coexpression of WincG2 and light-activatable guanylyl cyclases in body wall muscle cells results in an increase in WincG2 fluorescence upon light exposure that corresponds with the rate of cGMP production. Furthermore, WincG2 fluorescence changes in the cell bodies of the gustatory neuron ASER and the phasmid neuron PHB in response to [NaCl] step changes and sodium dodecyl sulfate, respectively. This provides the first visual evidence that suggests GMP levels change in these neurons in response to stimuli. Intriguingly, preliminary data suggest that in ASER, cGMP levels decrease linearly in the cell body while increasing transiently in the cilia in response to a [NaCl] downstep, which could have implications for cGMP’s potential role in both sensation and memory in ASER. Finally, we demonstrate that cGMP could act as a neuromodulator in a nociceptive neural circuit. WincG2 fluorescence increases in the nociceptive neuron ASH - which is not known to express guanylyl cyclases - in the absence of food while remaining relatively constant in the presence of LB. These results suggest that cGMP could be flowing from other neurons into ASH to signal food status, resulting in the modulation of ASH activity. Taken together, this work demonstrates that WincG2 could be used the uncover cGMP’s role in diverse neurobiological processes in living, behaving C. elegans.

Giardia lamblia is a eukaryotic protozoan parasite and the causative agent of giardiasis, a debilitating enteric disease resulting in much morbidity and mortality worldwide. It is of interest not only as a target for the development of improved giardiasis therapies, but also as a model eukaryotic system. Giardia represents the earliest branching clade of eukaryotic cells. It is thus an ideal system for investigating the evolution of cell processes, organelle compartmentalization, and critical protein families. Analysis of the structure and function of the Giardia endomembrane system, cysteine proteases, and clathrin orthologues are the focus of this dissertation project.

The ER has been studied recently as a putative endocytic organelle. Examination of Giardia endocytosis using fluorophore-labeled proteins revealed that proteins were rapidly trafficked to a tubulovesicular network with ER-like properties. Using reporter constructs, cysteine proteases that are orthologues of lysosomal hydrolases were localized to the same tubulovesicular network. Functional protease assays helped define the role that cysteine proteases play in the degradation of exogenous proteins. Organelle-specific markers were used to describe the tubulovesicular compartment in which endocytosis and subsequent proteolysis takes place.

Cysteine proteases have been implicated in life cycle transitions (encystation and excystation) of Giardia. The completion of the Giardia genome indicated that there are twenty-seven cysteine protease genes in Giardia. Cysteine protease 2 (GlCP2) was identified as the major expressed cysteine protease gene in Giardia. Biochemical analysis of heterologously expressed GlCP2 suggested that this gene indeed plays an important role in both encystation and excystation.

Giardia clathrin is also key to the processes of endocytosis and encystation. Clathrin is associated with the peripheral vacuoles of vegetative Giardia and may facilitate the inital endocytic uptake of proteins. During encystation, clathrin localizes to encystation specific vesicles and may function in cyst formation. A dominant negative clathrin heavy chain disrupted cyst formation but did not affect endocytosis.

Computational approaches for binding affinity prediction are most frequently demonstrated through cross-validation within a series of molecules or through performance shown on a blinded test set. Here, we show how such a system performs in two realistic applications: 1. An iterative, temporal lead optimization exercise, and 2. A hybrid strategy that leverages diversified information as input. In the first evaluation, a series of gyrase inhibitors with known synthetic order formed the set of molecules that could be selected for "synthesis." Beginning with a small number of molecules, based only on structures and activities, a model was constructed using the newly developed Surflex-Quantitative Modeling (QMOD) approach. Compound selection was done computationally, each time making five selections based on confident predictions of high activity and five selections based on a quantitative measure of three-dimensional structural novelty. Compound selection was followed by model refinement using the new data. Iterative computational candidate selection produced rapid improvements in selected compound activity, and incorporation of explicitly novel compounds uncovered much more diverse active inhibitors than strategies lacking active novelty selection.

For the second evaluation we present a hybrid structure-guided strategy that combines molecular similarity, docking, and multiple-instance learning such that information from protein structures can be used to inform models of structure-activity relationships. The Surflex-QMOD approach has been shown to produce accurate predictions of binding affinity by constructing an interpretable physical model of a binding site with no experimental binding site structural information. Here we introduce a methodological enhancement to integrate protein structure information into the model induction process in order to construct more robust physical models. The structure-guided models accurately predict binding affinities over a broad range of compounds while producing more accurate representations of the protein pockets and ligand binding modes. Structure-guidance for the QMOD method yielded significant performance improvements, especially in cases where predictions were made on ligands very different from those used for model induction.

Potassium (K+) channels are membrane-embedded proteins that selectively pass K+ ions in or out of cells in response to a variety of signals, such as membrane potential changes or binding of ligands. In an excitable cell, such as a neuron or cardiac muscle cell, delayed rectifier voltage-gated K+ channels respond to changes in membrane potential to restore the cell membrane to its resting state after an action potential.

In vertebrates, voltage-gated K+ (Kv) channels are tetramers of similar or identical subunits arranged around a central conducting pore. While these channels are primarily gated by membrane potential, their biophysical properties are set by the type of subunits in each tetramer and by interactions with other effector molecules, such as membrane phospholipids, calcium-binding proteins, kinases, and scaffolding proteins. In some cases, discrete intracellular domains control the specific assembly of pore-forming and accessory proteins. However, the molecular mechanisms that direct specific assembly of this wide range of components into a functional K channel complex are incompletely understood.

Chapter 2 of this thesis establishes the atomic-resolution structure of one such assembly domain from a Kv7 family channel (Kv7.4). This study suggests the structural basis for specific assembly properties and binding of scaffolding proteins by other members of the Kv7 channel family. Additional studies in Chapter 3 explore implications of the Kv7.4 assembly domain structure for oligomerization in other subtypes. The biochemical and functional effects of Kv7 mutations designed to disrupt or enhance assembly domain oligomerization further support the critical role for this domain in specific assembly of these channels.

The ongoing coronavirus disease 2019 (COVID-19) pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), continues to cause significant morbidity and mortality worldwide. Although most infections are mild and a majority of patients recover, some experience severe and often fatal systemic inflammation, cytokine storm, and acute respiratory distress syndrome. The innate immune system of the human body is the first line of defense against SARS-CoV-2, sensing the virus through pattern recognition receptors and activating inflammatory cascades that promote viral clearance. Simultaneously, the virus has evolved numerous strategies to escape detection and surveillance of the immune system for successful replication. An improved understanding of innate immunity and viral evasion strategies will help identify targeted therapies to mitigate disease and improve patient outcome. Here, we report two cellular epigenetic proteins, BRD4 and SIRT5, as anti- and proviral binding partners of SARS-CoV-2 envelope (E) and non-structural protein Nsp14, respectively. We identify bromodomain and extra-terminal (BET) proteins as critical antiviral factors as genetic or chemical inactivation of BRD4 exacerbates viral infection in cells and enhanced mortality in mice. BET inactivation suppresses interferon production induced by SARS-CoV-2, a process phenocopied by the “histone mimetic” E protein, supporting a model where the E protein evolved to antagonize the innate immune system via BET protein inhibition. Conversely, genetic or chemical inactivation of SIRT5 reduces SARS-CoV-2 replication in cells. While SIRT5 interacts with Nsp14 through its catalytic domain, Nsp14 does not appear to be directly deacylated by SIRT5. Depletion of SIRT5 results in higher basal levels of innate immunity and a stronger antiviral response during infection, indicating SIRT5 is a proviral factor necessary for efficient viral replication. Lastly, we compared the humoral immune responses elicited by SARS-CoV-2 variants, WA1, Delta, and Omicron. We show that without vaccination, infection with Omicron induces a limited humoral immune response in mice and humans. In contrast to WA1 and Delta, Omicron replicates at low levels in the lungs and brains of infected mice, leading to mild disease with reduced expression of proinflammatory cytokines, diminished activation of lung-resident T cells, and limited cross-variant immunity against non-Omicron variants in unvaccinated individuals. Collectively, these findings advance our understanding on the various host-pathogen interactions that need to be considered in designing novel SARS-CoV-2 therapeutics.