# Identification of Covalent Transcriptional Dependencies in Melanoma

> **NIH NIH R01** · MASSACHUSETTS GENERAL HOSPITAL · 2024 · $697,629

## Abstract

PROJECT SUMMARY/ABSTRACT
Advances in human genetics have identified 400+ genes that when amplified or mutated promote
tumorigenesis. While there has been substantial success in developing drugs for kinases and metabolic
enzymes deregulated in cancer, they represent only a tiny fraction of the cancer drivers discovered to date.
The vast majority (~80%) of these cancer drivers remain undrugged, in part, because of their druggability
remains unknown. Among these undrugged oncogenes are transcription factors which represent a large
fraction of cancer drivers but are notoriously considered undruggable. To address this challenge, our lab has
deployed state-of-the art cysteine-focused chemical proteomics to profile the ligandable cysteines in 300+
cancer models representing 23 tumor types. This pan-cancer DrugMap identifies thousands of ligandable sites
across the cancer proteome, revealing both common and uniquely ligandable sites across cancer subtypes. By
incorporating gene essentiality with cysteine ligandability, we uncovered that multiple lineage restricted
transcription factors may make promising candidates for small molecule inhibitor development. At the top of
this list, was SOX10 a transcription factor (TF) required for the proliferation of >85% of melanoma models,
which contained a highly ligandable cysteine in a conserved domain. We demonstrate that mutation of this
cysteine disrupts SOX10 transcriptional activity, forming the basis to undertake a covalent small molecule
screen, which identified a chemical probe that disrupts SOX10 activity. An exploratory medicinal chemistry
campaign yielded more potent analogs, that disrupt the SOX10 transcriptional complex, block melanoma
transcriptional networks and abrogate the proliferation of melanoma models that are addicted to SOX10. In
this grant application, we build on our core chemical proteomic platform and incorporate advances in
biochemical/biophysical methodologies, synthetic chemistry and patient derived melanoma models to: 1)
define the mechanism by which SOX10 liganding disrupts its transcriptional activity; 2) develop potent and
specific tool compounds targeting SOX10 and 3) characterize the impact of its pharmacological disruption on
transcriptional networks and proliferation of advanced melanoma models. The research proposed herein,
takes full advantage of advances in human genetics and functional genomics which nominates SOX10 as a
primary target required for melanoma growth and combines them with ultra-high throughput chemical
proteomic technologies to define the druggability of this TF cancer driver. This approach is a critical first step in
the development of targeted therapies against a class of proteins traditionally considered beyond the realm of
pharmacological intervention. If successful, the development of this chemical probe has the potential to
radically reshape the next-generation of targeted anti-cancer therapies.

## Key facts

- **NIH application ID:** 10978684
- **Project number:** 1R01CA285415-01A1
- **Recipient organization:** MASSACHUSETTS GENERAL HOSPITAL
- **Principal Investigator:** Liron Bar-Peled
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $697,629
- **Award type:** 1
- **Project period:** 2024-06-14 → 2029-05-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10978684

## Citation

> US National Institutes of Health, RePORTER application 10978684, Identification of Covalent Transcriptional Dependencies in Melanoma (1R01CA285415-01A1). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/10978684. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*