My primary research interest has focused on understanding the mechanistic underpinnings of the cell signaling networks that contribute to cancer progression and resistance to therapy. My laboratory is currently focused on two research areas:

  • We are testing the hypothesis that progression of prostate cancer to castration-resistance is frequently driven by changes in CHK2 signaling that regulate androgen receptor (AR) activity, facilitate cell proliferation, and minimize hormone dependence. These studies will uncover new information on how kinases regulate the AR and will reveal mechanistic information critical for the rational application of existing and novel therapies for (castration-resistant prostate cancer (CRPC).
  • We are developing an in vitro tumor microenvironment system for solid tumors that utilizes multiple cell types and recapitulates tumor capillary hemodynamics and biological transport. The applicability of this system is multifaceted and includes facilitating drug discovery and development, mechanistic analysis of drug responses, and development of biomarkers of response.
  • In addition to the two above focus areas, my collaborative interactions are focused on the molecular mechanisms of combinations of targeted molecular agents in order to overcome inherent and acquired resistance associated with traditional therapeutic approaches.


AR action and CHK2 signaling in prostate cancer.

The Androgen Receptor (AR) is essential for the growth and survival of castration-resistant prostate cancer (CRPC). The FDA approval of the CYP17 inhibitor, abiraterone, and the novel anti-androgen, MDV3100, emphasize the clinical importance of targeting AR in CRPC. Despite the justifiable excitement over these new therapies, the response to anti-androgens does not endure; the AR becomes reactivated with lethal consequences. It is essential to understand the mechanisms leading to AR reactivation, as they present targets for developing the combination therapies that will be required for effective deployment of next-gen anti-androgens. My research studies a mechanism of resistance to anti-androgens that represents a prime opportunity for therapeutic co-targeting: AR regulation by kinase signaling.

We propose a CHK2-CDC25-CDK1-AR signaling pathway which links CHK2, AR, and prostate cancer proliferation. This has particular relevance since several CHK inhibitors and second-generation CDK1 inhibitors are now in patient trials. Moreover, the CHK2 signaling pathway is activated in response to DNA damage that arises following treatments such as radiation therapy, a frontline treatment for local advanced prostate cancer. Thus, delineating how CHK2 impinges upon AR activity will provide important insights into how to more effectively combine radiation therapy with androgen blockade.

Two separate lines of investigation in my laboratory have remarkably converged to generate the hypothesis above. First, we discovered through a kinome wide RNAi screen that CHK2 knockdown significantly increases prostate cancer cell proliferation. This observation is clinically relevant since CHK2 inactivating mutations arise in ~4% of sporadic prostate cancer patients and over 60% of familial prostate cancer. Furthermore, CHK2 expression decreases as prostate cancer progresses to a castration-resistant disease. These data strongly suggest that CHK2 functions as a negative regulator or tumor suppressor in prostate cancer. We have determined that CHK2 knockdown increases AR transcriptional activity, providing evidence that CHK2 affects prostate cancer cell proliferation, at least in part, through the AR. Second, we found that AR S308 phosphorylation is catalyzed by CDK1, a downstream effector of CHK2. This phosphorylation regulates AR localization and occurs in G2/M, where a unique subset of androgen-dependent genes is expressed. These findings have significant clinical implications since CDK1 activity is elevated in CRPC. My NCI R01-funded research is currently focused on determining the mechanism of CHK2 regulation of AR activity and CRPC cell proliferation.

My research will determine the role of CDC25 and CDK1 in CHK2 regulation of AR activity and prostate cancer cell proliferation. We are studying the functional interconnectedness of each of the steps in the CHK2-CDC25-CDK1-AR signaling pathway to determine the mechanism of CHK2 regulation of AR activity and prostate cancer cell growth as well as the point or points of greatest therapeutic susceptibility. We are using prostate tumor clinical specimens to establish links between clinicopathologic features and CHK2, CDC25, CDK1, and AR expression and activation using human tissue microarrays. Using knockdown of CHK2, we are determining if CHK2 drives castration resistance in androgen dependent LNCaP cells, if overexpression of CHK2 in castration resistant C4-2 cells restores androgen dependence, and if CHK2 biological effects are dependent on CDC25 or CDK1. Moreover, we will determine the individual and cooperative roles of CHK2, CDC25, and CDK1 in promoting in vivo prostate cancer tumorigenesis using xenografts in immunodeficient mice.

We are exploring the molecular mechanisms of how the CHK2 signaling effector CDK1 regulates AR function through phosphorylation of AR S308 and the functional impact of AR S308 phosphorylation in facilitating cell proliferation. Our data suggest that CDK1 phosphorylation of the AR on S308 regulates AR localization and correlates with changes in AR transcriptional activity during G2/M. Our studies are determining if the CHK2 effects on cell proliferation and in vivo xenograft growth are mediated by AR S308 phosphorylation by CDK1. We are evaluating the effect of AR S308 phosphorylation in regulating AR localization and AR levels in mitosis. Finally, we are determining the role of AR S308 phosphorylation in regulating the G2/M specific subset of androgen-regulated genes.

We are evaluating the therapeutic potential of targeting CHK2 signaling in combination with androgen blockade and radiation therapy using in vivo models. Our goal is to understand the regulatory pathway linking CHK2 to AR function, and identify the best approach for applying these insights to the clinical setting. Our preliminary findings on CHK2-CDC25-CDK1-AR signaling suggest that inhibiting CHK2 in conjunction with radiation and androgen deprivation provides an additive if not synergistic therapeutic response. In these studies, we are evaluating how DNA damaging agents used to treat prostate cancer alter CHK2 regulation of AR activity and prostate cancer cell proliferation. Ultimately, we will determine if androgen blockade cooperates with CHK2 inhibition and radiation therapy to inhibit in vivo prostate cancer xenograft growth.

Summary. In spite of much research, our understanding of the mechanisms by which AR functions as a driver of CRPC progression remains poorly understood, and the tools available to therapeutically target this driver are one-dimensional. My work will expand the current view of AR biology from a static snap shot of AR as an androgen regulated transcription factor, to a dynamic one that integrates the complexity of cycling cells with regulation by signal transduction pathways. We are performing an in-depth, quantitative, experimental and computational interrogation of CHK2-CDC25-CDK1-AR signaling to evaluate potential targets for novel and more effective therapies against this challenging and often fatal disease. The robust precedent for kinases as therapeutic targets has hastened the development of therapies targeting this pathway for prostate cancer, as several CHK2 inhibitors and second-generation CDK1 inhibitors are now in clinical trials for other cancers. We believe that the thorough mechanistic evaluation of these potential kinase targets will pave the way for developing novel and more effective treatments for castration-resistant metastatic prostate cancer.


Developing an in vitro translational human tumor microenvironment system.

Current in vitro systems do not accurately predict efficacy or safety of anticancer therapies in humans. In vivo mouse models have significant limitations and historically have a poor correlation with human clinical outcomes. Even for the currently favored model of patient-derived xenografts (PDXs), the predictive value is still largely unknown. My laboratory is using a system that properly recreates and mimics the human tumor microenvironment to advance the discovery and development of effective anticancer agents. In my role as the Senior Director of Cancer Biology at HemoShear Therapeutics, a discovery biotechnology company, I led the development of a tumor microenvironment system (TMeS) based on the company’s intellectual property. I am continuing this work in my laboratory, in part through a continuing partnership with HemoShear Therapeutics funded through a Phase II SBIR contract. We have developed an in vitro system combining human microvascular endothelial cells experiencing tumor capillary hemodynamics, stromal cells, and tumor cells. We have developed proof-of-concept models in two cancers: non-small cell lung carcinoma (NSCLC) and pancreatic ductal adenocarcinoma (PDAC). Our tumor microenvironment system recapitulates the in vivo xenograft transcriptional program and responds to both established chemotherapeutics and experimental small molecule inhibitors at human patient physiologic doses (Cmax or AUC). We are currently evaluating the efficacy of cytotoxic chemotherapeutics and epidermal growth factor receptor (EGFR) family kinase inhibitors on NSCLC lines and PDX derived cells using our TMeS. These experiments will establish the applicability of the TMeS for predicting clinical drug responses. We are also comparing the transcriptome, proteome, phospho-proteome, and DNA methylome of NSCLC lines and PDX derived cells in our TMeS to xenografts and 2D cultures. This will enable us to determine in an unbiased manner if the tumor cells in the TMeS more closely resemble those in the in vivo or in vitro condition. Moreover, biological themes will emerge from this analysis that illustrate common biological processes between tumor cells in the TMeS and xenografts; we predict that these themes will include drivers of oncogenesis. Finally, we expect to align changes in protein expression with changes in gene expression and with DNA methylation that identify biomarkers associated with the biological state of tumor cells in the TMeS.

Summary. To effectively develop and evaluate effective cancer treatments, we need more predictive models of patient responses. The development and validation of our innovative in vitro 3-dimensional multi-cell tumor microenvironment system that properly recreates and mimics the human in vivo-state will advance the discovery and development of effective anticancer agents.


Molecular mechanisms of combinatorial therapeutics.

Combinatorial therapies hold much promise since inhibitors targeting a single signaling molecule that is overexpressed and activated in cancer have shown only modest clinical benefit when used as single agents. Therapeutic strategies targeting multiple pathways simultaneously can hypothetically overcome the inherent compensatory, feedback, and redundant signaling mechanisms that limit the effectiveness of single agent therapy. Our work uses two general paradigms to rationally develop effective therapeutic combinations. The first is to use global analysis to identify compensatory and redundant signaling pathways induced by specific targeted therapeutics. The second is to use synthetic lethal screening with small molecule inhibitors to search for combinatorial effects and functionally identify compensatory and redundant relationships. We have been able to identify combinations of small molecule inhibitors that demonstrate synergistic inhibition of tumor growth. These combinatorial therapies are being evaluated both mechanistically and in preclinical models. This research has evolved into a team-based project drawing on the capabilities of multiple research groups. Through collaborations (with Drs Weber, Bekerinov, Conaway, Jameson and Purow), the signaling nodes predictive of drug sensitivity are being determined by mathematical modeling using proteomic, transcriptomic, and genomic analysis. The combination of both the functional and mathematical approaches will facilitate our understanding of the cell signaling network and how that network responds to therapeutic intervention. This information will then be used to both refine the selection of effective drug combinations and identify tumors that are sensitive to such combinations.



Together, each of the initiatives described above contribute to the central goal of my research program, which is to understand how crosstalk among signal transduction pathways contributes to cancer progression in complex models of disease, and how that information can be used to develop more effective cancer treatments.