Functional profiling of germline SDH variants associated with cancer susceptibility

Objectives: Loss-of-function (LOF) mutations in SDHA or SDHB result in SDH-deficiency and increase the lifetime risk for developing a number of cancers, including GI stromal tumors, paraganglioma, pheochromocytoma, renal cell carcinoma, and breast cancer. In most cases, tumors with these mutations arise in the setting of a heterozygous germline mutation, which is

heritable. In patients with a known pathogenic SDHA or B mutation, genetic counseling (to screen other family members) and enhanced cancer screening procedures are indicated. When SDH- deficient tumors are detected at early stages, they are often curable. Nonetheless, we currently do not have the ability to identify patients at risk and screen them appropriately. The main

limitation is our lack of knowledge of the functional consequences of various SDHA or B mutations; only SDHA or B variants that cause LOF lead to cancer. Germline variants of SDHA and SDHB are relatively common (1.5% of population), however most are currently classified as variants of unknown significance (VUS).

Plan: This proposal will generate novel models for functionally testing SDHA and SDHB VUS seen in the population, including generating human model cell lines and utilizing a deep mutational scanning approach. Human models, which can provide strong evidence to be used for clinical classification of variants will be generated and validated for testing individual variants. In

addition, we have established a yeast model that allows functional screening of thousands of SDHA/B mutations in the yeast homolog genes (ySdh1 and ySdh2, respectively). Using this yeast model coupled with saturation mutagenesis, functional screening and deep sequencing, the functional consequences of all SDHA and SDHB missense mutations will be profiled.

Methods: To fill the current void of SDH-deficient cell lines, we will use CRISPR-Cas9 to generate SDHA- and SDHB-knockout cell lines and validate them to ensure their power for determining clinical classifications for SDHA and SDHB VUS reported in human tumors (Aim 1). To address the large number of SDHA and SDHB variants seen in the population, we will create libraries of

all possible ySdh1, (Aim 2) and ySdh2 (Aim 3) amino-acid variants by saturation mutagenesis and transform these libraries into yeast models deficient in the relevant endogenous yeast protein (e.g. mutant ySdh1 expressed in ySdh1-deleted yeast). We will select against SDHA/B LOF variants by growing the libraries under conditions requiring fully functional SDH (glycerol media).

Following deep sequencing, depletion analysis will be performed. Functional interpretations (functionally normal or LOF) will be made for each variant by comparing calculated effect scores with those of nonsense/ synonymous controls. Clinical Relevance: The care of veterans with cancer represents a significant portion of the

overall Veterans Affairs Health Care budget. It is estimated that more than 300,000 US veterans harbor an SDHA or SDHB germline mutation. Our best chance of a cancer cure lies in early detection in patients with SDH-deficient tumors, thus the critical medical need lies in classifying SDHA and SDHB genetic variants to enable identification and subsequent screening of at-risk

patients. Identifying which US Veterans are at risk of developing an SDH-deficient cancer will allow focused and potentially life-saving genetic counseling and enhanced cancer screening procedures. The results from our study will also be relevant to non-veteran populations, including family members of Veterans with pathogenic germline SDHA or SDHB mutations.