Jeffrey L. Sklar MD, PhD
Professor of Pathology and of Laboratory Medicine; Director, Molecular Diagnostics Program; Director, Molecular Genetics Pathology Fellowship; Director, Molecular Tumor Profiling Laboratory; Director of Molecular and Genomic Pathology
Molecular biology of human disease, especially cancer; Molecular biologic methods of disease diagnosis; Cancer-related genes and their functions in neoplasia and normal cellular physiology; Chromosomal structure and chromosomal abnormalities in genetic and neoplastic disorders; Immunogenetics
My lab is interested in the molecular biology of human disease, especially cancer. At present, the lab is particularly interested in two genes, JAZF1 andJJAZ1/SUZ12, which we discovered and showed to be fused head to tail in the cells of certain uterine tumors. Little is known about the function of JAZF1, although single nucleotide polymorphisms in this gene are associated with altered risk for type 2 diabetes and prostate cancer. JJAZ1is a Polycomb group gene (PcG), the product of which is essential for histone methylations that regulate chromatin remodeling and gene transcription. We have made significant progress in understanding how the JAZF1-JJAZ1fusion functions in oncogenesis, and the mechanism by which it acts has features not previously described in cancer. Recently, we found that JAZF1-JJAZ1 RNA is produced by trans-splicing between the pre-mRNAs for the two genes in normal endometrium and that this production is regulated during the menstrual cycle. This discovery suggests that perhaps other examples among the many gene fusions known in cancer exert their oncogenic effects through the constitutive expression of chimeric RNAs normally generated in a regulated fashion by trans-splicing. The lab is currently vigorously testing this hypothesis and investigating its many biological implications.
Extensive Research Description
Research in this laboratory generally concerns the molecular biology of human disease, particularly in the areas of the molecular genetics of cancer, lymphocyte biology, endometrial function, and the development of molecular methods for disease diagnosis. The research is of both a basic and translational nature.
A longstanding interest of the laboratory has been in chromosomal abnormalities associated with various forms of human cancer. Over the years, DNA fragments spanning the breakpoints in a number of chromosomal translocations have been molecularly cloned and used to identify oncogenes that lie at or near the sites of interchromosomal recombination. Among the genes identified in the laboratory through such studies are BCL2, a gene up-regulated in many cancers, particularly follicular B cell lymphomas, and critical in the control of apoptosis in normal tissues; NOTCH1, a gene for a cell surface receptor structurally altered in certain T cell leukemias and involved in the normal differentiation of many types of tissues; JAGGED2, a gene encoding a ligand for the NOTCH1 receptor; and MALT1, a gene mutated in one form of B cell lymphoma and a modulator of apoptosis in normal tissues. NOTCH1 and its orthologs continue to be studied both in terms of their modes of action and the tumors in which they act as an oncogene.
Most recently, two previously unknown genes—JAZF1 and JJAZ1 (now also referred to as SUZ12)—were discovered by us to be fused head to tail in most cases of endometrial stromal tumors. These two genes are currently the focus of intense investigation within the laboratory. JJAZ1 is a Polycomb group gene, the product of which is an essential member of the protein complex (Polycomb repressive complex 2, or PRC2) that in most or perhaps all cells catalyzes specific methylations of histone 3, leading to chromatin compaction and transcriptional silencing of DNA. The function of JAZF1 is less well understood; however, separate single nucleotide polymorphisms (SNPs) within two introns of this gene have recently been shown to be associated with a strongly increased risk of type 2 diabetes and a decreased risk of prostate cancer. Both associations are now being investigated in the laboratory.
As alluded to above, we found several years ago that about 50% of endometrial stromal sarcomas (ESSs—cancers arising in the non-epithelial portion of the endometrium) contain the JAZF1-JJAZ1 gene fusion, most often as a consequence of the (7;17)(p15;q21) chromosomal translocation present in the tumor cells (Fig. 1). Additionally, the great majority of endometrial stromal nodules (ESNs—tumors of the endometrial stroma that histologically resemble ESSs but are considered benign) also contain the JAZF1-JJAZ1 gene fusion. We have also found that a consistent molecular feature distinguishing ESSs from ESNs is that the unrearranged JJAZ1 allele is silenced in ESSs but is active in ESNs. Using tissue culture cells infected with transducing retroviral vectors, we have shown that expression of the JAZF1-JJAZ1 gene fusion confers resistance to apoptosis (Fig. 2) and increased proliferative capacity, although the latter only when the normal JJAZF1 alleles are suppressed by siRNA. These data are consistent with ESNs arising from normal endometrial stroma through acquisition of the JAZF1-JJAZ1 gene fusion, and ESSs arising from ESNs after epigenetic silencing of the unrearranged JJAZF1 allele (Fig. 3). Assuming that this model is correct, it is the first known example of a sarcoma developing from a benign precursor.
A recent and related finding by the laboratory has been that chimeric JAZF1-JJAZ1 mRNA and protein can be found in normal endometrial stromal cell lines, as well as in normal endometrial tissues. Numerous studies using a variety of techniques have shown that this chimeric RNA does not derive from gene fusions but from trans-splicing of RNAs. Specifically, we have shown that extracts of normal endometrial stromal cell nuclei can catalyze trans-splicing between exogenous JAZF1 and JJAZ1 pre-mRNAs in vitro (Fig. 4). The implications of this finding are broad and include the possibility that other gene fusions, of which there are hundreds known among different types of cancers, may work similarly, by constitutively expressing chimeric RNA that is produced normally by physiologically regulated RNA trans-splicing within restricted windows during cell and tissue development. The laboratory is currently pursuing this hypothesis by attempting to identify further examples of trans-splicing in normal tissues between the pre-mRNAs of other genes involved in gene fusions within various cancers.
A continuing subsidiary interest within the laboratory has been the molecular analysis of lymphomas and the investigation of normal lymphocyte biology. In the past, this interest has led to the discovery of the frequent occurrence in normal lymphocytes of so called antigen receptor trans-rearrangements that create potentially functional chimeric genes containing V, D, and J segments from different antigen receptor genes. Another discovery relates to somatic hypermutation within rearranged immunoglobulin genes of certain subtypes of B cell cancers, and the greatly increased expression of certain DNA repair genes within germinal centers of lymph nodes, specifically at the anatomic site where this process is thought to occur. Finally, the laboratory was the first to develop evidence of the B lineage of Reed-Sternberg cells in Hodgkin’s lymphoma and to demonstrate the presence of Epstein Barr virus in these cells. New projects related to these findings are still initiated from time to time.
Lastly, methods developed by the laboratory over the years have become widely used in diagnostic medicine, especially the application of antigen receptor gene rearrangements and molecular analysis of chromosomal translocations to diagnose lymphomas and other human cancers. Within this general area, the laboratory is continuing to work on new techniques for the improved use of nucleic acid markers in the diagnosis of human disease, and predominantly of cancer.