I graduated from Washington College in Chestertown, MD with a B.S. degree in Biology. This was followed by postgraduate studies on the biochemistry and endocrinology of vitamin D at the University of Maryland at College Park, where I obtained the Ph.D. degree. My postdoctoral training was conducted at Case Western Reserve University in Cleveland, Ohio, where I worked on the molecular biology of glycoprotein hormones under the mentorship of Dr. John Nilson.
My first independent faculty position as an assistant professor was in the Department of Medicine at Case Western Reserve University, where I investigated the transcriptional control of the platelet-derived growth factor genes in glioblastoma and other human cancers. I moved my laboratory in 1993 to the Department of Pharmacology at the University of Kentucky in Lexington, where I continued my work on molecular mechanisms underlying cancer progression and metastasis. I have recently moved to the University of Maryland (June 2012) to continue this work, which has evolved to a focus on genes regulating the metastatic process in melanoma.
My work has been funded almost continously since 1993 by the NIH through NIDDK, NHLBI and the NCI. The laboratory is currently funded by a recent R01 grant focused on identification of metastasis-driving mutations in our new transgenic mouse model for metastatic melanoma. We are also developing other promising avenues in drug development that we hope will lead to new treatment prospects for advanced forms of melanoma.
I am a member of the Molecular and Structural Biology Program within the University of Maryland Marlene and Stewart Greenebaum Cancer Center Program in Oncology. As such, I collaborate with both clinical and basic research investigators to elucidate molecular mechanisms underlying metastasis in melanoma and other cancers, with the ultimate goal of identifying novel therapeutic targets and molecular markers for management of cancer patients with advanced disease.
Progression of cancer to lethal forms requires the acquisition of driver mutations and aberrant programs of gene expression that endow cells with the ability to grow at distant sites, a process known as metastasis. Current cancer therapies often fail as they are almost exclusively directed to the hyperproliferative phenotype of primary tumor cells, and do not effectively target the metastatic process. Clearly, there is a dire need for new treatment options to target cancer in its advanced forms. The primary thrust of my National Cancer Institute-funded research program has been the elucidation of molecular mechanisms that mediate cancer progression and metastasis, and has centered to a large extent on the NM23 family of metastasis suppressor genes. Our studies of the NM23-H1 isoform has yielded considerable insights into molecular mechanisms that suppress metastatic potential, with one of the most significant being our identification of a novel mechanism underlying metastasis-driving mutations in melanoma.
The NM23 Family of Metastasis Suppressor Genes
Metastasis suppressors inhibit the metastatic activity of cancer cells with little effect on primary tumor growth (Smith and Theodorescu, 2009). The mouse gene nm23-m1 was the first metastasis suppressor gene to be described, with confirmation of suppressor activity of its human homolog nm23-h1 in cell lines and in melanoma, breast carcinoma and other cancers (Hartsough and Steeg, 2000). Although mechanisms underlying metastasis suppressor activity of the NM23-H1/M1 protein are not well understood, it harbors three distinct enzymatic activities that might mediate its antimetastatic functions. First to be described was its nucleoside diphosphate kinase (NDPK) activity, which maintains balance in nucleotide pools by catalyzing transfer of -phosphate between NDPs and NTPs (Agarwal et al., 1978). NM23-H1 also possesses a histidine kinase activity that may mediate an anti-motility function of the molecule (Wagner et al., 1997).
Our laboratory was the first to describe an association of 3’-5’ exonuclease activity with the NM23-H1 protein (Ma et al., 2004), and we subsequently showed this activity to be essential for its metastasis suppressor function (Zhang et al., 2011). 3’-5’ exonucleases provide proofreading during DNA replication and repair (Shevelev and Hübscher, 2002), and consistent with this function, we showed the NM23 homolog in S. cerevisiae possesses antimutator activity (Yang et al., 2009). More recently, we have also demonstrated antimutator activity of NM23-H1 in UVR-treated melanoma cell lines, and that NM23-H1 specifically promotes both the nucleotide excision/NER (Jarrett et al., 2011) and double-strand break/DSBR (unpublished observations) pathways for DNA repair. Importantly, NM23 deficiency is associated with increased rates of spontaneous mutations in a variety of melanoma and nontransformed cell lines, strongly suggesting its expression may be essential for suppression of progression- and metastasis-driving mutations. Importantly, we have recently demonstrated a direct physical association of NM23-H1 with sites of DNA damage, strongly suggesting its direct participation in the DNA repair process. Using a transgenic mouse strain harboring a concurrent deficiency in both the nm23-m1 or nm23-m2 genes, we observe vulnerability to ultraviolet light-induced melanoma in situ, consistent with a DNA repair function in vivo (Jarrett et al., 2011). Moreover, this nm23-deficient genotype confers aggressive metastasis when bred into a mouse model of melanoma with otherwise low metastatic potential (Jarrett et al., submitted for publication). Taken together, our work has revealed an unexpected and significant mechanistic role for NM23 proteins in both initiation and progression of melanoma.
Current Projects Under Study
In a project which has just been funded by the National Cancer Institute (R01 CA159871, “Suppression of Melanoma Initiation and Progression by NM23-H1”, 5th percentile), we have proposed three aims to better understand how NM23 deficiency drives progression of melanoma to metastatic forms. The first aim will employ transgenic mice harboring ablations of either the nm23-m1 or nm23-m2 loci, or the tandem ablation of both loci described above, to measure their individual and combined contributions to metastasis suppression. The second aim will probe more deeply into the molecular mechanisms through which NM23 proteins participate in the NER and DSBR pathways of DNA repair, including assessment of the individual roles played by the NDPK and 3’-5’ exonuclease functions. Developing and applying NM23 mutants that disrupt various aspects of its repair function and protein interactions with other key participants in these pathways (e.g. MRN complex, ATM, XRCC1), we plan to determine the contribution of the DNA repair function of NM23 to metastasis suppressor activity in melanoma cell lines and transgenic mice. In perhaps the most ambitious and exciting aim, we are applying the powerful technologies of whole genome sequencing and RNA-seq to identify candidate metastasis-driving genomic alterations and gene expression profiles in the melanoma lesions of our transgenic mouse model. These studies are being conducted in collaboration with Dr. John Carpten at TGen (Phoenix), and preliminary results have already identified a number of unique mutations in these metastatic melanoma tumors that represent excellent candidates for further study. For example, we have identified an indel in the Grin2a locus, which encodes an NMDA receptor isoform recently identified by exome sequencing to be a potential metastasis driver in human melanoma (Wei et al., 2011). The detection of the Grin2a mutation in metastatic melanomas of our NM23 knockout mice indicates the relevance of this model to the human disease, and its immediate promise for identifying other novel metastasis-driving events caused by NM23 deficiency.
Our laboratory is also highly focused on understanding the molecular mechanisms underlying the anti-invasive activity of NM23 proteins. Microarray analysis of our NM23-H1-transfected melanoma cell lines has revealed its potent regulation of a broad spectrum of mRNAs, with siRNA and forced expression approaches validating the functional relevance of interesting candidate mRNAs, encoding such molecules as lipocalin-2, IQ-GAP and BRAP. In addition, we have determined that NM23-H1 upregulates expression of fibronectin, which we have found to be essential for motility-suppressing interactions with a4b1 integrin (Novak et al., in preparation). These observations have opened a wealth of opportunities to better understand how NM23 acutely suppresses the metastatic process, which we plan to pursue through the submission of an NCI-directed R01 proposal in the coming 2012 calendar year.
We are also actively engaged in a project to identify small molecules that reactivate NM23 expression in advanced melanoma and other relevant human cancers. This represents a novel and potentially fruitful approach for therapy, as nm23 genes are almost invariably down-regulated in cancer and rarely damaged by mutation or deletion. In collaboration with the Drug Discovery Center at the University of Cincinnati (UC-DDC), we developed a mass spectrometry-based method for screening their extensive Procter & Gamble compound library for induction of NM23/NDPK protein expression. This effort has been quite successful, yielding > 20 distinct small molecules with NM23-inducing activity, at least four of which exhibit efficacy at low nanomolar concentrations. Moreover, these “first hits” have exhibited potent motility-suppressing activity in cultured melanoma and breast carcinoma cell lines. We have recently entered into a second phase of development to optimize activity and pharmacokinetic properties of these promising agents, using both analogs in hand at UC-DDC as well as new chemical syntheses. After preliminary testing of these second generation compounds in cell culture and our mouse model of metastatic melanoma, we will submit another R01 proposal to support their further development as potential therapeutic agents.
Lab Techniques and Equipment:
State-of-the-art molecular biological techniques (e.g. plasmid construction, measurement of steady-state levels for nucleic acids and proteins) Technologies for forced expression and siRNA-mediated silencing of gene expression Identifying protein-protein interactions (e.g. pull-downs, yeast 2-hybrid) Chromatin immunoprecipitation Static and live microscopic imaging (e.g. confocal, TIRF) for assessing cell motility NextGen sequencing (whole genome, RNAseq) High-throughput screening for metastasis-inhibiting compounds HPLC for protein purification (state-of-the-art Waters system) Mouse models for metastatic melanoma, including transgenic mouse strains for spontaneous melanoma and explantation of cell lines in immunodeficient and syngeneic strains Methodology for measuring activity of DNA repair pathways (NER, DSBR, BER, etc.)
Grants & Contracts:
9/10/12- 6/30/17 P.I.