ABSTRACT OF THE THESIS

Identifying the function of cGAS-Cap2b protein complex in Type-II Short CBASS Systems

by

David Garcia

Master of Science in BiochemistryUniversity of California San Diego, 2022 Professor Elizabeth A. Komives, Chair

Two sides, in an infinite battle for survival, as one side gets stronger, so does the other. As described by the Red Queen hypothesis, this is the dilemma between bacteria and bacteriophages (or phages), where both sides are constantly evolving to try and eliminate their opponent. Phages are simple organisms and outnumber all other organisms on earth, and their means of survival requires the destruction of bacteria. During infection, phages inject their DNA into bacterium and as a response bacterium will destroy itself but not before the phage has successfully replicated. This allows for further bacterial infection upon the destruction of bacterium from initial phage infection and can be recurrent until all bacteria has been destroyed. As a means for survival, bacteria have evolved a myriad of defense systems that allow it to fight phage infection. In this thesis I will cover some of the most well-known defense systems: the restriction-modification (R-M) systems that target phage sequences, the CRISPR-Cas systems where cells obtain immunity from previous phage infection, and the abortive infection systems (Abi) that upon phage detection will program cell death. Although each of these systems is unique and important for phage defense, I will focus on a type of abortive infection systems that were recently discovered in bacteria known as cyclic-oligonucleotide based antiphage signalling systems or ‘CBASS’. In this thesis, I also discuss the defense mechanisms of the different types of CBASS currently identified and discuss my study of a unique type of CBASS systems described as Type II (short) CBASS systems. Previously identified Type II (long) systems were discovered to have operons containing ancillary genes that share similarities with the non-canonical E1 and E2 signaling machinery described in the noncanonical autophagic ubiquitination pathway. Type II (short) CBASS systems have similar operons but with only one ancilliary gene related to ubiquitin E2 proteins. I was able to demonstrate how the bacterial E2-like protein, Cap2b, and the nucleotidyltransferase, cGAS, in Type II (short) systems using the artificial intelligence (AI) structural prediction tool, AlphaFold. Using the structural prediction, I focused on three strains to study and clone these proteins in order to isolate and purify them for structural determination using X-ray crystallography.

Ubiquitin (Ub) is an 8,500 Da, 76 amino acid protein that is among a family of similar Ubiquitin-like proteins. E3 Ligases facilitate the attachment of Ub onto lysine residues of substrate proteins. Ub can also be attached to another Ub molecule, forming polyubiquitin chains which signal for proteasomal degradation and the immune response. Small ubiquitin-like modifier (SUMO) is involved in cell cycle control and the DNA damage response. Another Ubiquitin-like protein, Neural precursor cell expressed developmentally down-regulated protein 8 (NEDD8), can modify some E3 Ligases to regulate their activation.

Chapter II presents a novel method for the side-by-side identification and quantification of Ubiquitin and SUMO modifications. Trypsin digestion of a Ubiquitin-modified protein leaves a diglycyl-lysine at the site of modification, which can be detected by mass spectrometry. Unlike Ubiquitin, detection of SUMOylated proteins is limited by the lack of naturally occurring protease sites in the C-terminal tail of SUMO proteins. This chapter describes detection of endogenous SUMOylation using α-lytic protease, WaLP.

Hydrogen-deuterium exchange mass spectrometry (HDX-MS) probes the solvent accessibility and conformational dynamics of proteins by measuring deuterium uptake on backbone amides over time. Chapter III describes an open-source software package designed to automate the correction, analysis and visualization of HDX-MS data.

Chapter IV investigates the structure, dynamics and regulation of an ASB9-CUL-RING E3 Ubiquitin Ligase. The structure of the substrate-receptor complex was determined for the first time by Cryo-EM, identifying changes in both the receptor and the substrate that were not predicted by previous models.

Chapter V further investigates the dynamics, regulation and activity of the ASB9-CUL-RING E3 Ubiquitin Ligase. In vitro assays, western blots and mass spectrometry data describe the locations and intensity of ubiquitin modifications. These data show that ARIH2, another E3 ligase, is necessary for initial substrate modification, and that the addition of other E2 enzymes can modulate the poly-ubiquitin chains formed. HDX-MS studies show that modification of the ASB9-CUL-RING ligase causes conformational restructuring that facilitates the binding and activation of ARIH2.

Eukaryotic organisms have evolved to recycle proteins in process called Ubiquitylation. Attachment of ubiquitin molecules to a substrate protein act as a signal for the substrate protein to be degraded. The last step of ubiquitylation typically requires an enzyme called the E3 ubiquitin ligase. This work investigates the interactions between the CUL5-ASB9 E3 ligase and a novel substrate, histone octamers, through Hydrogen-Deuterium exchange and MALDI-TOF mass spectrometry.

Chapter II explains the in vitro expression and purification of the CUL5-ASB9 E3 ligase (AECR). There are many components to AECR which are not very soluble if purified separately. The purification strategy of co-expression in E. coli and co-purification of AECR showed to be successful. Additionally, we show AECR is active a capable of ubiquitylating histone octamers through in vitro activity assays and SDS-PAGE analysis.

Chapter III quantitatively confirms and characterizes the activity of AECR through liquid chromatography-tandem mass spectrometry (LC-MS/MS). We further confirm through mutagenesis of AECR and in vitro assays that AECR’s histone ubiquitylation activity is not dependent on the modification called neddylation.

Chapter IV investigates the interactions between histones and ASB9, which is known to be the substrate receptor for AECR. Here we use Hydrogen-Deuterium exchange to identify changes in ASB9 upon histone binding. We observe that ASB9 increases in exchange when histones are present, but there is a small region of protection that may indicate the position where histones bind.

In this study we will present a full biophysical characterization of the NFκB-IκBα tertiary complex with a focus on intrinsically disordered regions. The identification of the intrinsically disordered N-terminus domain of IκBα exhibits protection occurring at the side chain interface, but not at the backbone interface. This protection occurs at regions near IκBα’s phosphorylation sites S32 and S36. Here we will present hydrogen deuterium mass spectrometry, fast photochemical oxidation of proteins coupled with mass spectrometry and in silico frustration calculations to support our claims.

The NF-κB family of transcription factors is a central mediator of immune and inflammatory responses. Dysregulation of NF-κB results in numerous diseases including cancers. Immediate activation and deactivation of NF-κB through macromolecular interactions is crucial. This dissertation investigated the role of conformational dynamics in these interactions by combining single-molecule FRET (smFRET) and molecular dynamics (MD) simulations. All-atom MD simulations showed that three different NF-ĸB dimers, RelA–p50, the RelA homodimer, and the p50 homodimer, display distinct relative domain motions leading to different degrees of binding cavity exposure that correlate with experimental DNA-binding rates. These motions were directly observed in real-time by smFRET, revealing a heterogeneous conformational ensemble. Interconversion between conformational states and DNA-binding events occur on the same time scale, suggesting they are coupled. Remarkably, the motions were also observed when NF-κB was bound to DNA, inferring a dynamic picture unexpected from crystal structures. These motions were allosterically altered by the inhibitor protein IκBα, which is known to accelerate DNA-dissociation and prevent DNA-binding. Coarse-grained MD simulations were also performed to predict the conformational ensemble of the disordered transactivation domain of NF-ĸB and captured the helix propensity of the motifs that become structured when bound to the transcription coactivator CBP domains. Subunit exchange between NF-ĸB dimers was also investigated with simple chemical kinetic equations. Together, this dissertation provides a molecular understanding of the conformational dynamics in NF-ĸB transcriptional regulation and may contribute to the ultimate deciphering of the dynamics-function relationship of proteins.

The NFκB signaling pathway is a central regulator of inflammatory and immune responses in virtually all human cell and tissue types, and aberrant signaling by NFκB has been implicated in a wide array of disease states including cancer, autoimmune diseases, and Alzheimer’s disease among others. Inhibition of NFκB by binding of IκB proteins represents the most robust regulatory mechanism for controlling the initiation and duration of the NFκB signaling event. NFκB and IκB both denote families of dimeric transcription factors and their inhibitor proteins, respectively. NFκB transcription factors self associate to form functionally active dimers which then bind in a specific manner to different IκB family members.

Here I present an extensive review of transcription factors and the NFκB pathway (Chapter I) to provide context for my work. I then describe a detailed analysis using hydrogen-deuterium exchange mass spectrometry (HDXMS) to probe the impact of IκBα binding on the internal dynamics of the p50/RelA NFκB heterodimer and demonstrate that long-range conformational changes induced by IκBα binding are of central importance to the inhibitory nature of the IκBα binding event (Chapter II).

Chapter III describes the first detailed biophysical analysis of the most recently discovered IκB protein, IκBε. I combined homology modeling and HDXMS to establish the presence of a seventh ankyrin repeat (AR) in IκBε and further provide a framework for utilizing HDXMS as a tool to complement homology modeling of proteins with no known structures. After identifying the seventh AR in IκBε, I performed the first comprehensive examination of IκBε’s ability to bind all NFκB dimers and the biophysical consequences on IκBε’s internal dynamics upon binding to its preferred NFκB dimers (Chapter IV). Finally, to further probe the interaction of IκBε with its preferred NFκB binding partners, I describe the impact of IκBε binding on the dynamics of the cRel heterodimers which are the NFκB proteins which bind IκBε with the highest affinity and discover the importance of entropic forces in driving these high affinity interactions (Chapter V).

The serine protease thrombin plays an essential regulatory role in blood clotting. Prothrombin is converted to the active α-thrombin in response to tissue damage, allowing the enzyme to cleave procoagulative substrates such as fibrinogen and PAR, enabling the formation of the clot. On the other hand, when thrombin binds its protein cofactor thrombomodulin (TM), the enzyme loses substrate specificity for procoagulative substrates, and engages the anticoagulative pathway by selectively targeting the substrate protein C for proteolytic activation. Despite decades of study, the mechanism by which TM enables thrombin to cleave protein C has yet to be fully characterized, indicating a need for further investigation. Crystallographic evidence shows no notable difference between the thrombin and thrombin-TM structures, suggesting the influence of TM is dynamic in nature. The work presented in this thesis uses two biophysical techniques, hydrogen-deuterium exchange mass spectrometry (HDXMS) and nuclear magnetic resonance spectrometry (NMR) to report the effects of TM on thrombin dynamics.

All organisms have mechanisms for degrading macromolecules that are critical for homeostasis and immunity. The ubiquitin proteasome system (UPS) and CRISPR-Cas System both implement targeting capabilities to degrade specific nucleic acids or proteins in response to their environment. In the UPS, an enzymatic cascade of E1 activating enzymes, E2 conjugating enzymes, and E3 ligases facilitate the post-translational modification of target proteins with ubiquitin (Ub), leading to the eventual formation of polyubiquitin chains which signal for proteasomal degradation and the immune response. CRISPR-Cas Systems utilize effector proteins that first bind CRISPR RNA (crRNA) that recognize specific stretches of nucleic acids and ultimately cleave these sequences to defend against invading viral species.In Chapter II, histones H3 and H4 are identified as novel substrates for the ASB9-CUL-RING E3 ubiquitin ligase. In vitro assays confirmed specific ubiquitylation towards histones H3 and H4 in a DNA-free form and that ASB9-CRL did not require supplementary E3 ligases to ubiquitylate histones, unlike other substrates. Additionally, specific modified lysines were detected by mass spectrometry. Chapter III investigates the structure and dynamics of the type III-E CRISPR-Cas effector, DiCas7-11 in different its apo, binary crRNA-bound state, and ternary crRNA-target bound state. Using HDX-MS, the structure and dynamics of the apo state is revealed for the first time elucidating a highly dynamic and solvent exposed structure that is heavily remodeled for target cleavage upon crRNA binding. The study also posits new hypotheses for the crRNA processing mechanism and the function of the insertion domain. Chapter IV leverages the specific and non-specific cleavage activities of the type VI CRISPR-Cas effector, RfxCas13d, to develop an CRISPR-based diagnostic to detect SARS-CoV-2. Development of a high-throughput fluorescence assay and lateral-flow strip assay are described with efficacy confirmed on SARS-CoV-2 positive patient samples.