Our Drug Discovery Platform
Cell-free protein synthesis (CFPS) was the biochemical tool that allowed deciphering of both the genetic code by Marshall Nirenberg and of protein targeting by Guenter Blobel. In a CFPS system, messenger RNA sequences for proteins of interest are translated in a cell extract.
Prosetta’s founders, academics at the University of California, San Francisco and the University of Washington, Seattle, adapted CFPS to a cell-free protein synthesis and assembly (CFPSA) system. The “assembly” portion of CFPSA is concerned with the process by which newly synthesized proteins form multi-protein complexes. For technical reasons, this process is extremely difficult to detect and has led conventional molecular biology to focus on individual proteins. We believe this overlooks the essential aspect of homeostasis, the balance that characterizes life (and health).
At Prosetta, we converted the original academic test tube CFPSA assay into a moderate-throughput drug screen. We screened a drug library for “hits” which would modulate protein assembly without blocking protein synthesis itself, and we found them. We term the enzymes targeted by our drugs, responsible for catalyzing formation of protein complexes, “assembly machines.”
Initially, we identified druggable steps along the viral capsid assembly pathway. The conventional understanding has been that the viral capsid (the protein shell that houses and protects the viral genome) forms all by itself, a process termed spontaneous self-assembly. However, Prosetta’s academic founders hypothesized something very different: that host enzymes catalyze capsid formation. A “hit” molecule identified by the screen prevented formation of the completed capsid through modulating of these previously undetected host enzymes of capsid assembly.
Over the past decade we have used CFPSA to explore essentially every family of virus causing major human disease. Some of the active compounds found are viral family specific, while others are shared between one or more viral families. The antiviral activity of hit compounds have been validated against live virus in cell culture. Our most advanced lead compounds have been successfully validated in animals, providing ultimate proof of concept.
Subsets from a library of 150,000 compounds were screened through the CFPSA system to identify hit drugs that modulate assembly machine activity. The assembly requirements from one virus to another are diverse. The hits were used to curate a collection of 300 structurally diverse (<70% similarity) small molecules with the shared property of assembly modulation. This “Hit-finder” Collection of compounds is an “assembly tool kit” with which to probe the role of host assembly pathways in non- viral diseases.
An extensive literature connects specific viral infections with specific subsets of neurodegeneration in which specific protein aggregates are associated. For example, prior serious influenza infections are associated with a Parkinson’s Disease pathology that includes presence of a-SYN aggregates, herpes infections with Alzheimer’s Disease characterized by ab42 and hyperphosphorylated tau aggregates, and endogenous retroviral activation has been associated with Amyotrophic Lateral Sclerosis (ALS) in which TDP-43 aggregates are pathognomic.
We hypothesized the protein aggregates linked to neurodegeneration are indicative of dysfunctions in protein assembly and, if there is overlap in the aberrant assembly pathways exploited by viruses and responsible for degeneration, we would see activity of Hit-finder compounds against non-viral disease.
When we screened in cellular models for TDP-43, a-SYN, and ab/Tau, each screen generated different hits from the Hit-finder collection. Remarkably, in light of the association between influenza and Parkinson’s Disease, an influenza-active chemotype is able to direct a-SYN away from the pathway leading to aggregates, and to rescue rat primary dopaminergic neurons from rotenone or dopamine mediated neurotoxicity. Similarly, HIV-active chemotypes were shown effective at moving TDP-43 back to the nucleus and preventing stress-induced aggregation in stress granules in cell lines generated from patients with both familiar and sporadic ALS. The advanced compound from HIV/TDP series have shown efficacy in transgenic C. elegans, D. melonagaster, and the SODG93A mouse models of ALS.
We have used energy-dependent Drug-Resin Affinity Chromatography (DRAC) as an important tool with which to elucidate the components of assembly-complexes. In DRAC, the drug is coupled to a resin and used as a ligand to identify components of the drug’s target from a given cell or tissue extract. For several viral families studied, a free drug eluate was found to contain a number of proteins that run together upon density gradient ultracentrifugation, which suggested the that assembly enzymes are themselves multiprotein complexes. When CFPSA starting extracts were depleted by DRAC, the system was able to synthesize capsid proteins, but unable to assemble them. When the depleted extracts were complemented with the free drug eluate (that had been extensively dialyzed to remove all drug) assembly proceeded as normal.
Mass spectrometry (MS-MS) of the free drug eluate revealed a subset of these proteins are part of the disease-associated protein interactome. A comparison of DRAC eluates from uninfected vs infected cell lysates by MS-MS revealed striking changes in the MS-MS profiles upon infection, with some proteins lost, some enhanced, and most unchanged. This was consistent with the hypothesis that viral infection results in modification of the host protein complexes from normal to aberrant, with the latter serving the needs of the virus rather than of homeostasis.
The insights gained from DRAC have a practical application in assessing improvements in the compound as we progress from initial hits to advanced drugs. An advanced drug is more selective for specific protein components that correspond to an aberrant assembly machine instead of a normal assembly machine. Improving assembly-machine selectivity improves both the compound’s activity and toxicity, since the activity of the normal assembly machine will be unimpaired by drug treatment.
Normalizing aberrant multi-protein complexes and modulating protein assembly is the future of medicine and what the next generation of pharmaceuticals must confront in order to meet the demands of tomorrow’s disease (or of today’s- click here to learn about our pan-respiratory antiviral program whose lead compound is potent at eliminating COVID-19 in cell culture).