Research

The optimal manner of achieving seamless translational cancer research is on a single platform of research, clinical care and education. On such a base, the broad thrust of the Eng laboratory can be characterized as clinical cancer genetics translational research, which involves the utilization and integration of multiple –omics-based platforms to identify, characterize and understand genes which cause susceptibility to high penetrance Mendelian and complex heritable cancers, to determine their role in sporadic carcinogenesis and to perform molecular epidemiologic analyses as they might relate to near-future clinical applications. Upon this framework, we are investigating the following broad topics:

Genetic and Functional Characterization of the Harmatoma-Tumor Syndromes: PTEN, KILLIN and Beyond
(Clinical research of PTEN syndromes)
The translational objective of this group of projects is to obtain evidence to ensure accurate molecular diagnosis to sub-set individuals for genotype-specific risk assessment and medical management such as surveillance and prevention, as well as for family-specific gene predictive testing. Functional interrogation will then point to signaling pathways which will help inform future therapeutic and preventative targets, whether by naturally occurring compounds or by synthetic compounds.

Our prototype hamartoma-tumor syndrome is Cowden syndrome (CS), which is a difficult-to-recognize under-diagnosed autosomal dominant disorder characterized by multiple hamartomas and a high risk of breast and thyroid cancers. We mapped the CS predisposition to 10q23 and subsequently identified germline mutations in the PTEN tumor suppressor gene as associated with the majority of classic CS. Because of the difficulty in recognizing CS, despite the International Cowden Consortium Operational Diagnostic Criteria, many more (perhaps 10-100-fold more) individuals who have incomplete clinical features of CS present to be evaluated. Only 5-10% of such CS-like individuals have germline PTEN mutations. Relatedly, after analyzing 3,042 probands, we have recently created a PTEN risk calculator which is based on demographics and clinical features to come up with the most parsimonious features to predict prior probability of finding a PTEN mutation. As another strategy of examining mutational spectra and variable clinical expression, PTEN and a series of other relevant genes are being analyzed for individuals and families presenting with different clinical features, one of which includes unexplained hamartomatous polyps.

PTEN is ubiquitously expressed and plays broad cell signaling roles, chief of which includes G1 cell cycle arrest and apoptosis. In 2000, after we noticed PTEN nuclear expression by immunohistochemistry, we demonstrated that PTEN can traffick in and out of the nucleus. PTEN has 4 nuclear localization-like signals and only works in pairs. We subsequently also showed that PTEN has two ATP-binding motifs that are required for nuclear exit. We are particularly interested in the mechanisms of nuclear entry and exit, their partitioned function and how natural compounds such as resveratrol affects intracytoplastic localization in the context of neoplasia.

In 2008, we identified germline variants of SDHB and SDHD associated with 10% of individuals with PTEN mutation negative CS and CS-like. The succinate dehydrogenase genes were first described as the susceptibility genes for pheochromocytoma-paraganglioma syndromes, which we also characterized. We are currently examining the interaction of the SDH and PTEN pathways in modifying neoplasia risks.

A novel tumor suppressor gene KILLIN which shares the same transcription start site as PTEN was recently found to be transcribed in the opposite direction. Presumably PTEN and KILLIN share the same bidirectional promoter. Recently, we found that germline methylation of the bidirectional promoter was associated with downregulation of KILLIN but not PTEN. Germline KILLIN methylation was found in ~37% of individuals with PTEN mutation negative CS and CS-like individuals. Notably, individuals with germline KILLIN methylation have a 2-3-fold increased prevalence of breast and renal carcinomas over those with germline PTEN mutations. This is important for genetic counseling and risk management between the two genes. We are currently validating the role of KILLIN in CS/CS-like and its function.

Integrative Genomic Analysis for Systems Medicine Approach to Common Cancer Risk
We are pursuing several strategies, from single platform to multiple platform integration, to obtain in point-of-care risk calculators to facilitate risk assessment and subsequently personalized medical management. We are particularly interested in homozygosity mapping and its integration with various relevant platforms to come up with low-to-moderate penetrance and eventually interaction with environment. We focus on such solid tumors as carcinomas of the breast, lung, oropharyngeal, and prostate as well as uveal melanoma.

Integrative Genomic Analysis of the Micro and Macro Environment of Solid Tumors
A solid tumor is heterogeneous comprising the carcinomatous component and its microenvironment, in which we have shown genomic and expressional alterations relevant to clinical outcome. We are utilizing single and multiple platform strategies to examine the alterations in solid tumor microenvironment in the context of clinical outcomes. Furthermore, a solid tumor does not exist in isolation and in certain sites, are exposed to its non-human macroenvironment including microbial and other populations. We are utilizing metagenomic profiling to describe microbial populations in oropharyngeal squamous cell carcinomas as they relate to other somatic alterations and clinical outcome.

Funded by National Cancer Institute, National Institutes of Health, Breast
Cancer Research Foundation, ACS Clinical Research Professorship, Arthur
Blank Foundation, Healthnet Foundation (per Sharon and Roger Vail), Susan G.
Komen for the Cure