John J. Krolewski - US grants

The receptor-type protein tyrosine kinases (PTKs) are cell surface molecules which bind growth factor ligands, and initiate pleotropic intracellular signaling cascades. In addition, most receptor PTKs are protooncogenes, and some have been implicated in human malignancies. Thus, the study of receptor tyrosine kinases has increased our understanding of neoplasia, cellular growth factors and the molecular dynamics of intracellular signaling. I have recently cloned, sequenced and characterized the full length cDNA of the human ltk gene, a new receptor PTK gene which encodes a 3.1 kB mRNA and a 100 kD protein with demonstrated tyrosine kinase activity. Preliminary evidence indicates that ltk expression is restricted to hematopoietic and neural crest derived cells, suggesting its activity is relatively tissue specific. However, its physiological function, putative ligand and oncogenic potential are all unknown. The broad aim of this proposal is to define the normal and pathologic function of the ltk gene product, with an emphasis on investigating its probable role as a receptor for a cellular growth factor and its possible role as a protooncogene. Specifically, four sets of experiments are proposed to achieve this objective. First, the characterization of the ltk gene and protein will be completed, focusing on the in vivo biosynthesis of the ltk protein and the cloning of the genomic sequences and upstream promoter region. In the second set of experiments, two approaches are proposed to identify the tissue(s) where the ltk gene acts physiologically: an expression survey and gene "knock out" experiments. The temporal and spatial expression of ltk in mouse embryos and its expression in adult tissues will be determined by in situ hybridization, RNAase protection analysis and immunofluorescence. To identify tissue(s) where ltk functions, targeted gene disruption of the ltk gene in mouse embryonic stem cells will be carried out, to produce mice which are homozygously disrupted at the ltk locus. Such mutant mice are expected to provide a powerful genetic model of ltk function. In the third aim, the oncogenic potential of the ltk gene will be investigated by determining whether various ltk derivatives, patterned after the transforming alleles of other PTKs, are able to transform fibroblasts or lymphoid cells, and by screening human tumors for the presence of activating alterations. The ultimate goal is to determine the role of ltk in human malignancy. Finally, two strategies are proposed for isolating the ltk ligand. One exploits the ability of some receptor-ligand pairs to transform cells via an autocrine loop mechanism, and the other involves screening known growth factors for their ability to induce ltk receptor autophosphorylation as an indirect assay of receptor binding. Identification of the ltk ligand would provide additional insights into the mechanism of growth control and may identify a novel growth factor with therapeutic value.

DESCRIPTION: (adapted from the investigator's abstract) Intracellular signaling pathways transduce information from the cell surface to the nucleus, controlling the growth, division and differentiation of both normal and neoplastic cells. Protein tyrosine kinases (PTKs) are responsible for triggering the cascade of events in many of these pathways. In the case of receptor-type PTKs (rPTKs), ligand binding activates the kinase, which autophosphorylates and subsequently binds and phosphorylates substrate proteins, initiating the signaling cascade. Some non-receptor kinases form a "broken" receptor complex with a cell surface molecule and function in a manner completely analogous to the rPTKs. The tyk2PTK is the prototype of a structurally distinct subfamily of non-receptor PTKs. It is activated in response to the cytokine interferon- (IFN), and it binds to and tyrosine phosphorylates one or more subunits of the IFN receptor. In addition, it is a candidate kinase for one or more members of a family of latent transcription factors, known as STATs (signal transducers and activators of transcription). Once these proteins are tyrosine phosphorylated, they translocate to the nucleus, bind an enhancer element and stimulate IFN-specific gene transcription. The resulting gene products are believed to promote an anti-viral state and/or inhibit cell growth. This latter property may account for the clinical utility of IFN in the treatment of a variety of indolent human neoplasms. Experiments are proposed to characterize, in detail, the molecular interactions between the tyk2 kinase, the IFN receptor -subunit and p113STST2 to investigate the possibility that tyk2 may be a tumor suppressor. Specifically, the regions of the tyk2 protein (p135tyk2) and the IFN receptor involved in the mutual association of these two proteins will be mapped. Next, the phosphorylation sites on both the tyk2 kinase and IFN receptor will be identified, and the role of these sites in the signaling pathway will be determined. Third, a model for STAT protein phosphorylation by p135tyk2, in which the receptor acts as a "docking" protein, will be tested. Lastly, they will test the hypothesis that constitutively active derivatives of tyk2 can act as a tumor suppressor in an IFN-sensitive cell line. In summary, the research plan outlined in this proposal should increase our understanding of the molecular interactions occurring in the IFN signal transduction pathway, as well as related cytokine-mediated pathways and provide insight into the mechanisms governing the growth of human tumors, perhaps facilitating the design of improved therapies.

JAK/STAT signaling cascades provide a relative simple and direct connection between ligand binding on the cell surface and gene transcription in the nucleus. These pathways mediate the response to a diverse group of polypeptide ligands which regulate cell growth, differentiation and effector function. One of the best characterized JAK/STAT signaling cascades is the pathway stimulated by interferon- alpha (IFNalpha). Extensive studies of this signal transduction cascade by a number of laboratories has provided a basic understanding of the initial molecular events regulating IFNalpha-induced gene transcription and has established a paradigm for studying other JAK-STAT pathways, as well. During the previous term of this grant, we have explored the molecular interactions between the various components in this cascade: the Tyk2 tyrosine kinase, the IFNaR1 receptor subunit, and the Stat1 and Stat2 transcription factors. In particular, we have: i) identified and characterized the Tyk2 kinase-IFNaR1 receptor complex; ii) identified and characterized the ligand-inducible Phosphotyrosine docking site on IFNaR1 which recruits the SH2 domain of Stat2; and iii) partially reconstituted the IFNalpha signaling pathway using chimeric receptors. Many details of the molecular interactions controlling this pathway, however, remain unknown, including which parts of the Tyk2 kinase are required for the interaction with IFNaR1, how the Stat2-Stat1 heterodimer is released from the receptor, the role of the distal portion of the IFNaR1 subunit and which specific molecules are involved in the negative regulation of this pathway. Four specific aims are proposed to investigate some of the these details: 1. Identify the residues in Tyk2 required for interaction with the IFNaR1 receptor subunit. 2. Define the mechanism of Stat1-Stat2 heterodimer formation. 3. Identify additional IFNaR1-interacting proteins which regulate IFNalpha signaling. 4. Investigate the role of CIS proteins in the regulation of the Tyk2 kinase and/or the IFNaR1 subunit.

DESCRIPTION (provided by applicant): The type I interferons (IFNs) regulate the innate immune response to viral infection and have anti-proliferative and pro-apoptotic action which has been exploited in the treatment of cancer. Furthermore, type I IFN signaling is a model for the large family of helical cytokines, which regulate many aspects of proliferation, differentiation and the immune response. Thus, studying type I IFN signaling is expected to impact our understanding of cancer and other diseases and facilitate the development of therapeutic modalities. Type I IFNs can signal through the JAK-STAT pathway. In brief, receptors (IFNaRI and IFNaR2) rely on JAK tyrosine kinases (Tyk2 and Jak1) to initiate signaling. IFN binding triggers JAK activation and phosphorylation of two STATs (Statl and Stat2). The STATs then heterodimerize, complex with the interferon regulatory factor 9 (Irf9), translocate to the nucleus and bind a response element upstream of IFN-regulated genes. Previously, this grant has supported our investigation of multiple aspects of this canonical signaling cascade. Recently, in the course of investigating the interaction between Stat2 and IFNaR2, we found that type I IFNs induce a two-step proteolysis of the IFNaR2 subunit in a manner that resembles the mechanism employed by Notch and the Alzheimer's precursor protein. Cleavage also occurs spontaneously and in response to various stimuli that induce PKC activation. An initial cleavage, mediated by the metalloprotease TACE, releases most of the ectodomain and a second cleavage by the intramembrane presenilin proteases releases the intracellular domain (ICD) of IFNaR2. Preliminary data indicates the ICD is capable of nuclear translocation and that this fragment of IFNaR2 can modulate gene transcription and inhibit cell proliferation, suggesting that type I IFNs might signal via a regulated intramembranous proteolysis (RIP) mechanism. Thus, the overall goal of this renewal application is to determine if type I IFNs can signal, in a physiologically relevant context, via RIP. Moreover, does RIP act in lieu of, or in addition to, the canonical JAK-STAT signaling pathway? Four specific aims are proposed to test this hypothesis. Aim 1 determines if type I IFNs induce cleavage and nuclear translocation of the endogenous IFNaR2 ICD and characterizes the mechanisms(s) initiating production of the ICD. Next, the key experiments in this proposal will test the hypothesis that cleavage is required for physiological effects of IFN by identifying the protease cleavage sites on IFNaR2 (aim 2) and determining if mutations which prevent cleavage perturb the physiological effects of the type I IFNs (aim 3). Finally, aim 4 addresses the mechanism of ICD mediated gene regulation by testing the hypothesis that the ICD functions in a complex with Stat2 and Irf9 to regulate gene transcription.

The primary objective of the Experimental Tissue Shared Resource (ETR) is to provide basic, translational, and clinical researchers within the Cancer Center access to, and analysis of, human and animal tissues. A key advantage is that this Core leverages the technical and professional expertise of the Department of Pathology and Laboratory Medicine. Three services are currently offered: i) tissue histology and immunohistochemistry; ii) tissue banking; and iii) tissue analysis, including laser capture microscopy, morphometry and image quantitation and genotyping services. Since the Core was initiated in the prior funding period, 63 CFCCC Members have used it. There has been growth in use, especially in the histology/immunohistochemistry component, which accounts for the majority of usage within the ETR. In the prior funding period, the histology component has provided high-quality service at below-market recharge rates, with short turn-around times to 52 different CFCCC Members who are mainly pursuing translational research on human cancer specimens or studying rodent models of human cancer. When required, we customized services to meet the special needs of CFCCC Members. These features have made our histology services superior to those available from commercial vendors, contributing to usage growth. ETR services contributed to 43 publications in the prior funding period, including 36 using histology services. Tissue banking services were used by 21 CFCCC investigators. Analytical services, used by 15 investigators, have been particularly valuable to investigators pursuing translational and clinical research projects. In this renewal, we propose to expand to meet an increased demand for histology services, maintain our specialty analytical services, strengthen our tissue procurement activities by focusing on comprehensively banking prostate cancers and provide a new mouse pathology service which will assist CFCCC investigators in the initial characterization of genetically engineered mouse models of human cancer.

DESCRIPTION (provided by applicant): The diagnosis and treatment of prostate cancer (PrCa) is a complex clinical problem that affects millions of men each year in the US, producing >200,000 new diagnoses and >27,000 deaths. Most PrCa is localized and slow growing;a smaller fraction metastasizes and causes death. Since prostate cancer derives from androgen dependent epithelium, some cases of localized PrCa and most cases of metastatic PrCa are treated via androgen deprivation therapy (ADT). ADT is a more specific therapy than surgery or radiation, inducing apoptosis in both localized and dispersed prostate epithelial tumor cells, rather than ablating the entire gland and causing damage to surrounding structures. However, androgens have systemic effects beyond regulating the growth and differentiation of prostate epithelium and ADT therefore produces considerable side-effects. Moreover, ADT frequently fails. The development of novel, targeted therapies that trigger PrCa apoptosis in a selective manner offers the possibility of reduced side-effects and improved control of disease progression. Our overall goal is to elucidate the mechanism of prostate apoptosis in response to androgen withdrawal, to eventually design therapies that selectively mimic the apoptotic effects of ADT. To achieve this goal, we have investigated androgen withdrawal induced apoptosis (AWIA) in normal rodents and model prostate cell lines. Recently, we developed a novel magnetic resonance imaging (CHESS-MRI) protocol to quantitate prostate regression and found that tumor necrosis factor (TNF) signaling is required, but not sufficient, for AWIA. Previously, we showed that FLIP, an inactive homologue of caspase-8 that inhibits TNF signaling, decreases following androgen withdrawal. The expression of both FLIP and TNF are androgen regulated. Two other cytokine signaling pathways (triggered by transforming growth factor-b (TGFb) and insulin-like growth factor 1 (IGF1)) have also been implicated in AWIA, and the corresponding genes are also androgen regulated. We will test the following central hypothesis: androgen withdrawal regulates a network of paracrine-acting cytokines (TNF, TGFb, IGF1, IGFBP3) and thereby reduces the expression of the intracellular apoptosis inhibitor FLIP. The net effect is to trigger the apoptotic arm of the TNF signaling pathway. The experimental plan utilizes rodent AWIA models and a prostate-specific PTEN-deficient mouse model resembling human prostate cancer, in conjunction with CHESS-MRI to image both normal and tumorous prostates. Aim 1 determines the role of TNF, TGFb and IGFBP3 in ADT-induced prostate apoptosis and whether these same cytokines cooperate to induce apoptosis in normal and tumorigenic prostates. Aim 2 investigates the role of TGFb and NF-kB cross-talk in mediating androgen deprivation induced down-regulation of FLIP. Finally, Aim 3 tests the functional role of FLIP, determining if FLIP down-regulation can cooperate with pro- apoptotic cytokines to induce apoptosis and regression of normal prostate and PrCa in a murine model. PUBLIC HEALTH RELEVANCE: The detection and treatment of prostate cancer, the most prevalent visceral cancer among men in the US, is a complex clinical problem. Androgen-deprivation is used to treat both localized and metastatic prostate cancers, but this treatment is prone to side-effects and, in metastatic cancers, often fails. To identify novel apoptotic medical therapies for prostate cancer, we propose to investigate the cooperative role of multiple cytokines (IGF1, TNF and TGFb) in the mechanism of androgen withdrawal induced apoptosis in normal and tumorigenic prostates.

DESCRIPTION (provided by applicant): The diagnosis and treatment of prostate cancer (PrCa) is a complex clinical problem that affects millions of men each year in the US, producing >200,000 new diagnoses and >27,000 deaths. Most PrCa is localized and slow growing; a smaller fraction metastasizes and causes death. Since prostate cancer derives from androgen dependent epithelium, some cases of localized PrCa and most cases of metastatic PrCa are treated via androgen deprivation therapy (ADT). ADT is a more specific therapy than surgery or radiation, inducing apoptosis in both localized and dispersed prostate epithelial tumor cells, rather than ablating the entire gland and causing damage to surrounding structures. However, androgens have systemic effects beyond regulating the growth and differentiation of prostate epithelium and ADT therefore produces considerable side-effects. Moreover, ADT frequently fails. The development of novel, targeted therapies that trigger PrCa apoptosis in a selective manner offers the possibility of reduced side-effects and improved control of disease progression. Our overall goal is to elucidate the mechanism of prostate apoptosis in response to androgen withdrawal, to eventually design therapies that selectively mimic the apoptotic effects of ADT. To achieve this goal, we have investigated androgen withdrawal induced apoptosis (AWIA) in normal rodents and model prostate cell lines. Recently, we developed a novel magnetic resonance imaging (CHESS-MRI) protocol to quantitate prostate regression and found that tumor necrosis factor (TNF) signaling is required, but not sufficient, for AWIA. Previously, we showed that FLIP, an inactive homologue of caspase-8 that inhibits TNF signaling, decreases following androgen withdrawal. The expression of both FLIP and TNF are androgen regulated. Two other cytokine signaling pathways (triggered by transforming growth factor-b (TGFb) and insulin-like growth factor 1 (IGF1)) have also been implicated in AWIA, and the corresponding genes are also androgen regulated. We will test the following central hypothesis: androgen withdrawal regulates a network of paracrine-acting cytokines (TNF, TGFb, IGF1, IGFBP3) and thereby reduces the expression of the intracellular apoptosis inhibitor FLIP. The net effect is to trigger the apoptotic arm of the TNF signaling pathway. The experimental plan utilizes rodent AWIA models and a prostate-specific PTEN-deficient mouse model resembling human prostate cancer, in conjunction with CHESS-MRI to image both normal and tumorous prostates. Aim 1 determines the role of TNF, TGFb and IGFBP3 in ADT-induced prostate apoptosis and whether these same cytokines cooperate to induce apoptosis in normal and tumorigenic prostates. Aim 2 investigates the role of TGFb and NF-kB cross-talk in mediating androgen deprivation induced down-regulation of FLIP. Finally, Aim 3 tests the functional role of FLIP, determining if FLIP down-regulation can cooperate with pro- apoptotic cytokines to induce apoptosis and regression of normal prostate and PrCa in a murine model.

We propose to develop a prognostic mRNA immune signature for resected stage II-III melanoma, in order to stratify these patients, who have a recurrence risk of ~50%. There is an urgent need to define accurate prognostic markers for stage II-III melanoma patients because adjuvant immunotherapy to prevent recurrence is both toxic and expensive. Unfortunately, conventional staging does not allow for accurate assessment of risk and many ?low risk? patients in fact progress. Further, while the immune tumor micro- environment is a key determinant of outcomes, standard pathologic assessment of tumor infiltrating lymphocytes (TILs) is subjective and not applicable to general clinical practice. In order to better stratify patients for adjuvant immunotherapy, we seek to validate a previously defined 53-gene signature. This signature employs NanoString, a probe based technology well suited to the analysis of the partially degraded RNA typically recovered from clinical grade FFPE tissue sections. We initially defined this signature in a training set and then validated these findings in an independent test set [Sivendran, et al. (2014) J. Invest. Dermatol. 134:2202-11]. As both training and test sets populations are retrospective, the next step to develop a clinically applicable assay is to test the signature on prospectively gathered samples. Given the fact that melanomas can recur years after resection in these early stage patients, prospective validation would be lengthy, and thus we opted to use the prospective retrospective analysis [PRA] approach whereby samples are collected and annotated prospectively but analyzed retrospectively. For this purpose we use samples from the Eastern Cooperative Group (ECOG) E1697 study of adjuvant interferon randomized vs placebo in stage II-III resected melanoma. Patients in this study have been maintained on study follow-up since they were randomized between 1998 and 2010. This project is collaboration between the Herbert Irving Comprehensive Cancer Center (HICCC) at Columbia University, the Roswell Park Comprehensive Cancer Center (RPCCC), the University of British Columbia and Omniseq, a commercial biotech spin-off that is majority owned by RPCCC. In Aim 1 (UH2 phase), we perform the analytical validation of the assay including validation of gene reference controls, RNA extraction quality, and reporter binding density. The milestone for moving to the UH3 phase (Aim 2) will be submission to the NYS Clinical Laboratory Evaluation Program. In the UH3 clinical validation phase, we first evaluate two additional retrospective sample populations (from HICCC and RPCCC). As part of this study we will also correlate the 53-gene signature with state of the art immune indices including quantitative multiplex immunofluorescence (qmIF) to assess CD8 to CD68 ratio, a metric recently defined (by our group) to correlate with survival. We then perform the definitive PRA analysis using the E1697 samples.

The overall goal of the Genetics and Genomics (GG) Program is to discover novel cancer genes, to elucidate the function of diverse genetic and epigenetic states, to translate mechanistic insight into therapeutic targets and associated biomarkers, and to accelerate the delivery of precision cancer medicine to our patients. This will be accomplished in three aims: Aim 1 ? To interrogate genetic drivers of tumor hallmarks; Aim 2 ? To delineate epigenetic states as biomarkers and targets in cancer, and Aim 3 ? To translate of genetic and epigenetic discovery toward clinical application. A number of important scientific discoveries emanated from the GG Program that illustrate the overall research depth, cancer focus, and added value. Additionally, the GG Program developed unique tumor sequencing data and analysis approaches to support precision medicine concepts. For example, panels developed within the GG Program were leveraged toward the development of a molecular diagnostic company (OmniSeq) that contributes to the recruitment of Roswell Park patients into molecular basket trials, such as NCI-MATCH. The GG Program research portfolio encompasses studies addressing genetic basis of aggressiveness of hormone refractory breast and prostate cancer in African- American populations ? cancers that are of particular importance for our WNY catchment area. John Krolewski, MD, PhD and Erik Knudsen, PhD are new co-Leaders for the Program who have strong interests in basic, clinical, and population aspects of tumor genetics and epigenetics. Dr. Krolewski's leadership efforts focus on signaling/cytokine pathways and genomics of genitourinary malignancies, which complements Dr. Knudsen's focus in the areas of cell cycle control and genomic stability, and targeting such pathways in breast and gastrointestinal cancers. The Program is comprised of 27 members from nine Roswell Park departments/centers (Cancer Genetics and Genomics, Center for Personalized Medicine, Molecular and Cellular Biology, Pharmacology and Therapeutics, Urology, Thoracic Surgery, Biostatistics and Bioinformatics, Medicine, and Pathology), whose total peer-reviewed funding is $3.1M direct costs (NCI funding $1.9M direct) and a total funding of $5.4M direct costs. Since the last renewal, 12 new members (nine recruited from outside Roswell Park) have joined the Program. Of the 385 publications generated over the last funding cycle, 9% are intra-programmatic and 44% are inter-programmatic; 50 (13%) publications are in journals with an impact factor >10. Future goals of the Program are focused on multi-investigator projects that make impactful discoveries and translate basic science into the clinical arena, harnessing GG science to target cancers at the genetic and epigenetic levels.