The Mechanisms and Therapeutic Utility of cGAS-STING-Mediated Immune Processes

Cytosolic dsDNA is a molecular pattern indicative of microbial infection or diseased cellular states requiring an innate and eventually adaptive immune responses. Cellular detection of cytosolic DNA by the pattern recognition receptor (PRR) cGAS leads synthesis from ATP and GTP of the cyclic dinucleotide (CDN) 2’3’ cyclic GMP-AMP (cGAMP). cGAMP is a molecular pattern detected by the ER resident signaling molecule STING that, upon engagement, elicits transcription of type I interferon (IFN-I) and proinflammatory cytokines. This response is crucial for a diverse array of physiological processes that include homeostatic, inflammatory, and immune activities. A large and growing portion of our work involves molecular and immune characterization of cGAS and STING biology. This includes the manner in which these proteins are activated, the immediate intracellular reactions they produce, the systemic and immune effects of their induction, proteins and factors essential to their function, and ways in which cGAS-STING can be exploited for therapeutic outcomes. This focus of inquiry was initiated after a high throughput screen of small molecules was performed for the purpose of identifying novel compounds that induce type I IFN responses in human cells. This led to retrieval of a collection of new and unique molecules that activate STING-mediated phenotypes relevant for enhancement of antigen-directed adaptive immunity. Characterization of the mechanisms of action of these inspired us to develop and undertake reverse genetic loss of function methods to identify cellular factors required for their activities. For this CRISPR/Cas9 technology was used to generate a large and growing library of human and mouse cell lines from which key innate signaling proteins are deleted including PRRs, kinases, surface receptors, kinases, adaptor molecules, and transcription factors. This approach allowed us to definitively describe cellular factors that are essential to IFN-I induction by the most promising and potent lead molecules. Importantly, many of these appear to activate STING via molecular mechanisms that are thus far unknown. A large effort is currently underway to characterize the molecular bases by which these compounds activate STING-dependent processes, which we are confident will reveal new and fundamental molecular phenomena that will enhance our understanding of this exceptionally important innate signaling process. 

Discovery and Characterization of Novel STING Agonists as Vaccine Adjuvants

The paucity of FDA-approved vaccine adjuvants necessitates the development of new approaches to enhance the efficacy of viral vaccines and understand the processes important for pathogen-specific protective immunity. Co-administration of STING-inducing compounds with viral antigens leads to augmentation of protective antibody and T cell-mediated immune responses. Based on this we have screened a large library of small molecules and identified a substantial set of novel compounds that induce STING-dependent phenotypes. We are using structure-activity relationship and medicinal chemistry approaches with a diverse set of lead molecules to identify derivatives that display potent yet tolerable in vivo STING induction. Since these compounds are predominantly active in human and not murine cells, we have constructed a genetically humanized mouse strain in which the endogenous coding region for STING has been replaced with the human ortholog. A thorough molecular and immunological characterization of these mice indicates they are conventionally responsive to human-selective STING agonists and represent a novel and appropriate in vivo model for examining the safety and efficacy of our compounds in vaccine and tumor experiments. Based on this we are developing a set of these agonists as adjuvants for pairing with protein antigens derived from Chikungunya, Zika, SARS-CoV-2 and influenza viruses. This involves exploring the suitability of various formulations for their in vivo use, the innate immune responses they trigger, the T helper polarization they elicit, the protective capacity of the immune responses they induce, and the single cell transcriptomes they generate. Our overarching goal is to advance two candidates to a stage that attracts funding for a full adjuvant development contract from the NIH. 

A Novel Self-Adjuvanting Virus-Like Particle Vaccine Platform

cGAMP endogenously synthesized by mammalian cells is transmitted intercellularly by being packaged in membraned exosomes and virus particles. This allows protection of the molecule from extracellular degradative phosphodiesterases and induction of innate responses in the surrounding tissue. We are exploiting this phenomenon for the purpose of generating vaccines that are safe, inert, and display enhanced immunogenicity. Virus-like particles (VLP) are replication incompetent membraned structures that lack genomes but exhibit proper surface display of viral antigens important for neutralization. We show that human cells engineered to express a constitutively active cGAS mutant generate VLP that contain encapsulated cGAMP at high levels. Moreover, similar expression of bacterial enzymes that synthesize cyclic di-AMP or di-GMP also leads to VLP encapsulation of those molecules. When administered in vivo these stimulate STING-dependent activity (e.g. IFN-I induction) and also enhanced antigen-directed immune responses. We are using this approach to create self-adjuvanting VLPs against clinically relevant and potentially emerging members of the Alphavirus and Flavivirus families. These will then be used in mouse and nonhuman primate models to assess their ability to protect against viral challenge. In addition, these vaccines will be leveraged as unique research tools for pursuing a mechanistic understanding of the molecular and innate correlates of STING-mediated adaptive immune responses. For this we will perform systemic multiplex cytokine analysis and single cell RNA sequencing of draining lymph nodes following VLP administration. Ultimately it is our hope that this approach can be applied to a wide variety of virus types and develop into a generalizable platform to fight emerging disease.    

Development of an Adjuvanted Vaccine Against Mpox Virus

Mpox is a poxvirus that has re-emerged in the United States in 2003 and 2022. Currently the only option available for conferring protective immunity to infection is the JYNNEOS live attenuated smallpox vaccine, which is contraindicated for both immunocompromised patients and those prone to severe allergic reactions. Moreover, whether JYNNEOS elicits a durable immune response to Mpox infection that is sufficiently potent is not clear. Given the increasing probability of future Mpox outbreaks as well as the virus’ potential as an agent of bioterrorism new strategies for generating protective immunity are needed that are safer, more effective, and widely scalable. As such, we are investigating new vaccines against Mpox that include viral protein antigens paired with novel TLR activating adjuvants using a nonhuman primate model of Mpox infection and immunity developed by Scott Wong (VGTI, ONPRC). This work first involves assessment of various antigen/adjuvant pairings and suitable formulations in a C57Bl/6 mouse model. The best combinations will then be used to vaccinate Rhesus macaques (using JYNNEOS as a control vaccine) and then viral challenge of vaccinated animals to assess protective capacity. We hope to show that our novel vaccine approach represents an improved strategy for generating anti-Mpox immunity that can be applied clinically.

Anti-Tumor Activity of Novel cGAS- and Human-Selective Small Molecule STING Agonists

Pharmacologic activation of STING in the tumor microenvirenment (TME) elicits cytokine and immune signaling that potentiates anti-tumor activity of cytotoxic T and NK cells, especially when paired with checkpoint inhibition factors. While this has been shown repeatedly and to great effect in numerous syngeneic mouse tumor models, multiple clinical trials using similar approaches with diverse STING agonists have failed for unknown reasons. We are thus employing two alternative approaches to explore new methods for leveraging the anti-tumor effects of the STING pathway. First, we have discovered novel molecules that activate cGAS-mediated synthesis of 2’3’ cGAMP. These successfully activate canonical STING-dependent signaling and in this way lead to tumor regression in vivo. Based on this we hypothesize that molecular targeting of cGAS (rather than STING directly) may represent an alternative strategy for therapeutic usage of this pathway to eliminate tumors. Second, we hypothesize that uncharacterized biological differences exist between human and mouse STING orthologs that engender disparate molecular, innate, and adaptive responses. These manifest as anti-tumor activity following agonist-mediated activation that differs in fundamental efficacy between species. To address this, we are using the genetically humanized STING mouse strain described above in combination with human-selective STING inducers to evaluate observable anti-tumor effects as well as the mechanisms responsible. We are also characterizing the transcription-independent phenotypes such as autophagy and inflammasome activation activated by STING in both species and across agonist types. We hypothesize that these differences may be responsible for the disparity in agonist efficacy observed between humans and mic