A phenotypic screening platform for chronic pain therapeutics using all-optical electrophysiology

Abstract

Chronic pain associated with osteoarthritis (OA) remains an intractable problem with few effective treatment options. New approaches are needed to model the disease biology and to drive discovery of therapeutics. Here, we present an in vitro model of OA pain, where dorsal root ganglion (DRG) sensory neurons were sensitized by a defined mixture of disease-relevant inflammatory mediators, here called Sensitizing PAin Reagent Composition or SPARC. OA-SPARC components showed synergistic or additive effects when applied in combination and induced pain phenotypes in vivo. To measure the effect of OA-SPARC on neural firing in a scalable format for drug discovery, we used a custom system for high throughput all-optical electrophysiology. This system enabled light-based membrane voltage recordings from hundreds of neurons in parallel with single cell resolution and a throughput of up to 500,000 neurons per day, with patch clamp-like single action potential resolution. A computational framework was developed to construct a multiparameter OA-SPARC neuronal phenotype and to quantitatively assess phenotype reversal by candidate pharmacology with different mechanisms of action. We screened ~3000 approved drugs and mechanistically focused compounds, yielding data from over 1.2 million individual neurons with detailed assessment of both functional OA-SPARC phenotype rescue and orthogonal "off-target" effects. Analysis of confirmed hits revealed diverse potential analgesic mechanisms including well-known ion channel modulators as well as less characterized mechanisms including MEK inhibitors and tyrosine kinase modulators, providing validation of the platform for pain drug discovery.