Faculty Summaries
Erica A. Golemis, PhD
Erica A. Golemis, PhD
  • Deputy Chief Scientific Officer & Vice President
  • Co-Leader, Developmental Therapeutics
  • Adjunct Associate Professor, University of Pennsylvania
  • Adjunct Professor, Drexel University College of Medicine
Office Phone: 215-728-2860
Fax: 215-728-3616
Office: W406
  • 1. Studies of NEDD9 in cancer and PKD.

    The major source of cancer morbidity and mortality is uncontrolled metastasis. Elevated expression of the protein NEDD9 (also known as HEF1 and Cas-L) is a major determinant of metastasis in multiple types of cancer. This protein has been a major focus of research in our laboratory since we first described it in 1996 (Law et al., MCB 16:3327, 1996). NEDD9 has no catalytic activity, but has a complex modular structure that allows it to dock multi-protein complexes. This modularity allows NEDD9 to act as an important signaling hub, with action in integrin signaling cascades (influencing cell migration and invasion, at focal adhesions), with Aurora-A signaling cascades (controlling cell cycle and ciliary dynamics, at the centrosome), and with ErbB cytoplasmic effectors (affecting core cell proliferation). Our work on NEDD9 addresses the complex biology of this protein in cancer and other pathologic conditions, and in therapeutic response to treatment.

  • 1(a). Incidence of HER2-induced breast cancer is greatly reduced in mice lacking NEDD9.

    In collaboration with Andres Klein-Szanto, Denise Connolly & Sachiko Seo
    Figure 1.
    Mammary tumor formation analysis shows absence of NEDD9 decreases tumor incidence.

    Inappropriately high levels of ERBB2/HER2, a growth factor receptor, drive 30% of human breast cancers.  To further investigate the role of NEDD9 in breast cancer, we have mated mice lacking NEDD9 (Nedd9-/-) with mice that transgenically express high levels of ERBB2/HER2/neu; the most physiological model for the human disease.  In this long-latency model of breast cancer, mice lacking NEDD9 are dramatically resistant to formation of tumors.  By 24 months, 88% of MMTV-neu;Nedd9+/+ mice have developed mammary tumors, but only 29% of MMTV-neu;Nedd9-/-mice (see Figure 1).  Analysis of cells derived from primary MMTV-neu;Nedd9+/+ versus MMTV-neu;Nedd9-/- tumors indicated few consistent differences in formation of new tumors following xenograft analysis, metastasis, growth, and signaling. These data combined suggest that NEDD9 contributes substantially to the initiation of neu-driven mammary tumors, but that in cases where tumors can arise, secondary changes have circumvented the need for NEDD9 function. We therefore analyzed mammary progenitor populations of mammary epithelial cells in MMTV-neu;Nedd9+/+ and MMTV-neu;Nedd9-/- animals, prior to the onset of overt cancer.  This work revealed striking defects in cell growth and polarity induced by absence of the Nedd9 gene, and decreased or altered activity of NEDD9-relevant signaling pathways that influence cell proliferation.  We hope investigation of this novel role for NEDD9 in mammary tumor initiation may reveal points of breast cancer cell vulnerability that can be therapeutically exploited. Furthermore, we are evaluating whether NEDD9 expression status conditions the ability of breast cancers to be treated by various chemotherapeutics so that, in the future, NEDD9 may be used as a biomarker for use in selecting patients for individualized treatment.

  • 1(b). Polycystic kidney disease: Role of NEDD9 as a modulator of cystogenesis

    Anna Nikovona, in collaboration with Sachiko Seo & Gregory Germina
    Figure 2.
    Figure 2

    Autosomal dominant polycystic kidney disease (ADPKD) affects about 1:700 individuals and typically manifests in middle age.  50% of affected individuals develop progressive cyst formation and ultimate loss of renal function, culminating in end stage renal disease that requires dialysis or kidney transplant as treatment. Genetically, ADPKD is caused by mutational inactivation of polycystins 1 and 2 (PC1 and PC2, encoded by PKD1 and PKD2). Importantly, on a molecular level, ADPKD has many features in common with cancer, with elevated activity of proteins such as HER2, SRC, ERK, and mTOR important in disease pathogenesis.  Our work has investigated whether NEDD9 plays an important role in PKD.

    The proteins PC1 and PC2 function at the cilium, a small antenna-like structure protruding from the cell surface that receives and interprets mechanical and soluble signals that influence cell growth (see Figure 2). An increasing number of mutations in genes regulating ciliary function have been found to impact cystogenesis. We have recently shown that the oncogenes NEDD9/HEF1 and its partner Aurora-A (AURKA) have multiple properties relevant to PKD. NEDD9 activation of AURKA at the ciliary basal body controls ciliary disassembly. NEDD9 and AURKA also bind and regulate the activity of the PC2 calcium channel (influencing cytoplasmic Ca2+), and are abundant and periodically active in normal kidneys, with AURKA expression and activity elevated in early PKD-associated renal cysts.
    We have now used a conditional PKD1-floxed mouse model crossed to a NEDD9 knockout mouse model to investigate the effect of eliminating NEDD9 expression on cyst formation in vivo. We used a number of different approaches to study in vivo cystogenesis, including magnetic resonance imaging (MRI), scanning electron microscopy (SEM), and immunohistochemical analysis. To more exactly analyze relevant signaling changes, we are further studying the role of NEDD9 in cystogenesis using primary kidney cells isolated from Pkd1-/- and Pkd1-/-Nedd9-/- mice.
    We have shown that deletion of NEDD9 in the Pkd1-floxed inducible mouse model significantly increases kidney size and the overall cystic burden, although loss of NEDD9 is not independently sufficient to induce cysts. Analysis of proliferation in primary kidney cells revealed accelerated rate of growth associated with a Nedd9-/- genotype. Cilia in primary kidney and kidney cells from Pkd1-/-Nedd9-/- mice were lengthened and deformed, and centrosomal defects observed. Interestingly, endogenous Nedd9 protein levels are upregulated in kidney lysates from Pkd1-/- mice compared to wild type kidney lysates, implicating NEDD9 as an intermediate in normal disease-relevant cystogenesis: analysis of related signaling is in progress. These and other findings identify NEDD9 as a physiological regulator of cystogenesis and suggest use of NEDD9 as a biomarker for prognosis.
    In ongoing work, we are testing the idea that NEDD9 function in cystic pathogenesis partially involves actions in focal adhesions. Focal adhesions serve as the mechanical linkages between the cell and the extracellular matrix (ECM), and as biochemical signaling hubs that concentrate and direct numerous signaling proteins relevant to control of cell architecture and proliferation. Abnormalities in focal adhesion formation and function are involved in fibrosis, promote tumor invasion and metastasis, and are a feature of cyst formation in ADPKD. In cancer, NEDD9 affects focal adhesion disassembly and cell attachment to the ECM, through regulation of its partner kinases SRC and FAK; tumors in NEDD9 knockout mice are characterized by depressed activation of these kinases. We are currently investigating relations between NEDD9, PC1, SRC, and FAK in the ADPKD mouse model.

  • 2. Understanding drug resistance networks.
    Figure 3.
    Figure 3.

    Drug resistance is a common problem in cancer treatment. Part of our work seeks to better understand the factors leading to drug resistance, with the goals of reversing these resistance mechanisms to improve cancer therapies and of better stratifying patients as likely or unlikely to respond to specific treatments. For this work, we are applying bioinformatics approaches coupled with siRNA screening to identify resistance genes (Figure 3).

  • 2 (a). Analysis of complex cancer signaling networks to predict effective therapeutic strategies.

    Tim Beck & Ilya Serebriiskii

    Two central elements of translational cancer research are the pursuit of optimized, personalized therapies and the elucidation of pathological processes with the aim of identifying novel drug targets.  Our research focuses on the associations between an epidermal growth factor receptor (EGFR)-centered network of genes, aspects of the transforming growth factor beta (TGF-ß) signaling pathway and heat shock protein 90 (HSP90).  EGFR and TGF-ß signaling pathway-related mutation(s) and/or overexpression are associated with chronic cell proliferation, the most fundamental attribute of cancer cell growth.  HSP90 is a molecular chaperone with a plethora of ‘clients’, many of which are also involved in cell proliferation and function within the EGFR and TGF-ß signaling cascades; many cancer cells upregulate HSP90 and are dependent on its action.  The Golemis Lab has well-established expertise regarding changes in EGFR signaling affecting cancer proliferation and survival.  One of the primary goals of this particular project is to build on and expand on knowledge gained from previous synthetic lethal screening for genes regulating response to EGFR inhibitors.  We are in the process of utilizing newly designed siRNA screens to further evaluate the sensitizing potential of inhibiting components of the EGFR signaling pathway, the functionally interacting TGF-ß signaling pathway, and the HSP90 chaperone system.  We hypothesize that these screens will allow us to construct a network of genes that sensitize cancer cells to HSP90 inhibitors.  Thus far we have identified 38 sensitizing genes, of which several have existing inhibitors, and which may be exploited to sensitize cancer cells to different therapeutics regiments.  Detailed investigation of the mechanism of sensitization is in progress.

  • 2 (b). Drug resistance networks in ovarian cancer.

    Ilya Serebriiskii, in collaboration with Denise Connolly, Lainie Martin & David Proia

    Epithelial ovarian cancer (EOC) is the second most commonly diagnosed and most frequently lethal gynecologic malignancy. At the time of presentation, most EOCs no longer depend on single genetic determinants for growth and survival, and targeted therapies used as single agents typically do not work in this disease. We hypothesized that network-based synthetic lethality screens would help identify critical pathways to target in combination therapy; further, by comparing the results of multiple screens, we might identify core regulators of tumor cell viability. We have used RNAi screening with a targeted library to identify a set of gene candidates that influence the response of different cell lines to drug treatment and that are frequently active in ovarian tumors. Mapping the pattern of hits back to the network revealed suggestive groups of very closely interacting proteins clustering with well-validated modulators of cancer cell survival and drug sensitivity, an indication of key cancer signaling nodes.
    Analysis of the mode of action of this gene group found that a subset of candidates physically or functionally clustered with HSP90. Building on these results, we also validated of a set of candidate siRNAs that influence the response of EOC cell lines to the clinically promising HSP90 inhibitor ganetespib. We performed in depth evaluation of a cluster of hits that bind HSP90, to assess their regulation of the growth, survival, and signaling in EOC cells. We found that treatment of multiple EOC cell lines with ganetespib significantly decreased expression of both known HSP90 clients, and new clients we identified through functional screening, including PKC epsilon, PKC alpha, phospho-MAPK, phospho-STAT3, and others.
    Building from these data, we collaborated with the Connolly group to assess the therapeutic value of combining ganetespib (STA9090) and paclitaxel (a standard drug used for treatment of EOC) in an orthotopic xenograft model. Ganetespib was extremely effective when combined with the standard EOC therapeutic agent paclitaxel. Together, these data are providing strong preclinical support for the design of the first trial of a promising, well-tolerated HSP90-targeting inhibitor in EOC. Ganetespib also synergized with other drugs (cisplatin and dasatinib) in vitro, suggesting other potential combinations for in vivo validation in the near future. Assessment of the antagonizing/sensitizing protein sub-network for HSP90 will be completed in additional EOC cell lines to determine which have the broadest activity profile. We will then study the functional roles of specific proteins of interest identified in our antagonizing/sensitizing sub-network for HSP90, emphasizing their role in drug response in EOC cells. Based on these studies, we are working with the company that developed ganetespib (Synta Pharmaceuticals) and with the Clinical Trials Office at our institute, Fox Chase Cancer Center, to initiate a phase I/II clinical trial using ganetespib in combination with paclitaxel or docetaxel; the intended PI for this trial effort is Dr. Lainie Martin, M.D.

  • 2 (c). Polycystic kidney disease: Testing new therapeutic targets

    In collaboration with David Proia, Gregory Germina & Stefan Somlo
    Figure 4.
    Figure 4.

    As noted above, most ADPKD arises from mutations in PKD1, encoding polycystin-1 (PC1), leading to deregulated renal cell growth, loss of epithelial cell polarity and fibrosis. Development of ADPKD is associated with the elevated activity of numerous signaling proteins that control proliferation and cell-environment interactions.  These include Aurora-A, SRC, STAT3, ERK, and others, all of which are oncogenes or oncogene effectors being targeted in the treatment of cancer.  To date, there are few treatment options for ADPKD.  We have hypothesized that some of the more well-tolerated pathway-targeted agents initially developed for use in cancer may find practical use in therapy of ADPKD.
    We are currently testing specific small molecule inhibitors for oncogenic proteins in different models for ADPKD using a treatment schedule adjusted for a chronic disease such as ADPKD. To monitor the dynamics of ADPKD progression during the treatment we are performing MRI to quantify cyst and kidney volume. In addition, to understand the mechanism of the specific inhibitors in the context of ADPKD, we are using immunohistochemical and biochemical analysis of renal mouse tissue, and in vitro analysis of primary kidney cells derived from ADPKD mouse models. As shown in Figure 4, specific inhibitors are showing excellent activity in reducing cyst formation.