Department of Pharmacology and Therapeutics
Roswell Park Cancer Institute
Elm and Carlton Streets
Buffalo NY USA 14263
Tel: 716 - 845 - 3090
Fax: 716 - 845 - 8857
E-mail: adrian.black@roswellpark.org

Program: Cell cycle regulatory molecules as targets for chemotherapy; Alterations in transcription factors in cancer
Progression from a normal cell to cancer involves the acquisition of multiple changes. These changes, which can be genetic or epigenetic, ultimately give rise to alterations in gene expression that affect the levels and/or activities of factors involved in regulation of cell growth and survival. Thus, a major contribution to the development of rational strategies for treatment and prevention of cancer will come from characterization of proteins whose expression or activity is changed during tumorigenesis and identification of drugs that can exploit these cancer specific changes. This approach has the advantage that that it not only sheds light on the mechanisms of action of such drugs, but it also identifies molecular targets that can be used to track drug actions in animal models and eventual clinical trials. Recent advances have led to the recognition of the pathway in which cyclin/cyclin-dependent kinase complexes phosphorylate and inactivate the retinoblastoma protein as a critical regulator of cell growth. This pathway is a promising target for cancer therapy since it is disrupted in most, if not all, cancers. Of the members of this pathway, cyclin D1 is of particular interest since a) it regulates cell growth and progression through G1 phase of the cell cycle by multiple mechanisms, b) it is a critical sensor of the cellular growth environment, and c) overexpression of cyclin D1 is one of the most common abnormalities in cancer. Studies in this laboratory have determined that treatment of colon and prostate cancer cells with vanilloids (capsaicin and resiniferatoxin) leads to a dramatic downregulation of cyclin D1 which is accompanied by growth arrest, indicating that these naturally occurring substances may be useful in cancer treatment and/or prevention.

Due to their central role in regulation of gene expression, growth-related transcription factors are also of particular interest as targets for cancer therapy. The krüpel-like factor family of transcription factors has been implicated in both positive and negative growth regulation in multiple systems, and work from this laboratory has identified changes in expression of one member of this family (intestinal krüpel-like factor) as an early event in colon cancer progression (see below).

Progress
Effect of vanilloids on cell cycle regulatory molecules, cell growth and cell survival. Capsaicin (the major pungent ingredient of chili pepper, Capsicum annuum) and resiniferatoxin (derived from Euphorbia resinifera) are vanilloids that stimulate and subsequently desensitize primary afferent neurons through their binding to the vanilloid receptors. Due to these properties, vanilloids are undergoing clinical trails for pain management and treatment of diseases with neurological components such as psoriasis and overactive bladder. Recent studies have also determined that vanilloids can affect cell growth and survival in a vanilloid receptor-independent manner and it has been suggested that these effects may be specific to cancer cells. In keeping with an anticancer role, animal studies have indicated that oral capsaicin can protect against carcinogen-induced cancers, including those of the lung, liver, and intestine, and epidemiological data pointing to a possible protective effect in urological cancers have been reported. That these apparent anticancer effects of vanilloids may be exploited clinically is supported by the current characterization and acceptance of vanilloids in conventional medicine, together with the long-standing and widespread use of chili as a food additive and the use of extracts of chili and Euphorbia in alternative medicine.

Although several mechanisms for vanilloid receptor-independent action of capsaicin and resiniferatoxin have been proposed (e.g. inhibition of a tumor-specific, ubiquinol (NADH) oxidase, inhibition of protein synthesis through competition with tyrosine, disruption of mitochondrial membrane potential, oxidative stress and effects on calcium flux), the relative importance of these mechanisms in various cell types remains to be determined. Thus, rational design of vanilloid-based cancer treatments will require detailed analysis of their modes of action in vitro and in vivo. Based on data from animal and epidemiological studies, we are examining the effects of vanilloids on intestinal epithelial and prostate cancer cells. It is hoped that characterization of their actions in these models will allow optimization of vanilloids for cancer therapy/prevention and identification of molecular markers of drug action for use in in vivo/clinical studies.

Intestinal Epithelial Cells. Studies in IEC-18 intestinal crypt cells and colon cancer cell lines indicate that resiniferatoxin and capsaicin can inhibit the growth of both normal and transformed intestinal epithelial cells and that this growth inhibition is accompanied by a rapid downregulation of cyclin D1. Resiniferatoxin is at least 20-fold more potent than capsaicin and analysis of the actions of these two compounds, together with differences in their relative potencies in different cell types, indicates that resiniferatoxin may act through a novel mechanism. Thus, in addition to contributing to the potential development of vanilloids for colon cancer therapy and prevention, elucidation of the mechanisms of action of these compounds may identify novel target(s) for cancer treatment.

Prostate Cancer Cells. Analysis of the effects of vanilloids on prostate cancer cells indicates that these cells are highly sensitive to both capsaicin and resiniferatoxin. The observed growth inhibition is accompanied by rapid (<4 h) downregulation of cyclin D1 protein and mRNA and delayed (12-24 h) induction of the cyclin dependent-kinase inhibitor, p21WAF1/CIP1. Longer treatments (24-48 h) lead to a significant amount of cell killing. Comparison of androgen-dependent and -independent prostate cancer cell lines indicates that androgen-independent cells are equally, or more, sensitive to vanilloids. Thus, vanilloids offer promise for treatment of androgen-independent prostate cancer, for which no effective treatment currently exists. Levels of capsaicin that lead to growth arrest and cell killing are similar to those reported to be transferred to the blood following application of capsaicin-containing creams to the skin. This finding, together with the fact that capsaicin and resiniferatoxin are undergoing clinical trails for other urological disorders, further supports the potential of vanilloids for chemoprevention and chemotherapy of prostate cancer.

Role of krüppel-like transcription factors in intestinal tumorigenesis. The Sp1/krüppel-like factor (KLF) family of transcription factors has been implicated in growth regulation in many systems. Members of this family have a highly conserved DNA binding domain near their C-terminus which contains three krüppel-type zinc fingers, together with a non-conserved N-terminal transcriptionally active domain(s). In keeping with the lack of conservation in their N-terminal domains, these factors have diverse effects on transcription, being repressive or activating. Interestingly, individual family members can repress or activate transcription, dependent on the promoter and cell type being examined. Due to their highly conserved DNA binding domains, KLFs bind DNA with overlapping sequence specificity. These properties, together with the fact that cells express multiple family members, indicate that these factors may form a network through which transcription of individual genes can be fine-tuned. Reports have indicated that krüppel-like factors may play a role in tumor progression in the intestine. The intestinal epithelium shows reciprocal expression of two krüppel-like factors. Gut krüppel-like factor (GKLF, KLF4) is expressed in the non-growing epithelial cells of the intestinal villi and surface mucosa, but not in the growing cells of the crypts, indicating that it has a role in growth arrest, differentiation and/or mature function in this tissue. In keeping with this role, GKLF is downregulated in intestinal adenomas and tumors, relative to normal mucosa. Furthermore, overexpression of GKLF in colon cancer cells, leads to growth inhibition and cell death. In contrast, intestinal krüppel-like factor (IKLF, KLF5) is expressed in the growing cells of the crypts but is downregulated on the villus and surface mucosa, arguing that this factor may have a growth-promoting role in the intestinal epithelium. This idea is supported by studies in fibroblasts in which IKLF overexpression led to increased cell growth. In order to elucidate potential roles of IKLF in intestinal tumorigenesis, we have utilized laser-capture microdissection (LCM) and reverse-transcription-polymerase chain reaction (RT-PCR) analysis to examine the expression of IKLF in normal intestinal epithelial cells and in adenomas derived form APCmin mice and human familial adenomatous polyposis (FAP) patients. These studies indicate that IKLF mRNA is downregulated in adenomas relative to normal mucosa. Furthermore, preliminary analysis comparing gene expression between adenomas and growing cells of intestinal crypts, has revealed that, while there is a marked down-regulation of IKLF expression, there was little difference in the levels of GKLF mRNA between these populations of growing cells. This latter finding indicates that the downregulation of GKLF observed when comparing tumors with normal mucosa may reflect a failure of cancer cells to undergo normal growth arrest/differentiation in the upper crypts. Thus, while GKLF expression is clearly detrimental to colon cancer cells, it may affect cell survival rather than having a direct role in aberrant growth regulation. The finding that IKLF is downregulated in adenomas compared with both total normal mucosa and growing crypt cells indicates that loss of this factor is an early event in intestinal tumorigenesis. Thus, IKLF may have a tumor suppressive role in the intestine.

Students:
General Research Interest

• Cell cycle control in cancer
• Transcriptional alterations in cancer
• High-throughput screening for drugs affecting specific molecular targets

Research Background and Interest

The efforts of the laboratory are aimed at understanding the implications of changes in cell cycle molecules and transcription factors for cancer progression and therapy.

Cell cycle control. Current efforts are aimed at understanding mechanisms underlying altered expression of cyclin D1 and p21waf1/cip1 in prostate and colon cancer. Overexpression of cyclin D1 is one of the most common alterations in cancer and p21 has been implicated in altered growth and apoptosis in tumor cells. Thus, an understanding of the regulation of these proteins is likely to present a novel target for therapeutic intervention.

Transcriptional regulation. Particular emphasis is placed on the Sp1 and krüppel-like family of transcription factors. This family contains over 20 members many of which have been implicated in cancer progression in multiple systems. Current studies include the roles of this family in colon cancer progression and in transforming growth factor beta resistance.

High-throughput screening. Strategies are being developed to allow for high-throughput screening of chemical libraries for drugs that affect the expression of specific molecular targets.

Key Publications:

 

• Black AR, Black JD, Clifford-Azizkhan J. Sp1-related transcription factors in cell growth and cancer. J. Cell Physiol. 188: 143-160, 2001.
• Black AR, Black JD, Bateman NW. Downregulation of IKLF/KLF5 is an early event in intestinal tumor progression. Proc. AACR, Abstract 528, 2002.
• Black AR, Black JD, Clark JA, Frey MR, Roy D. Capsaicin- and resiniferatoxin-induced growth arrest in intestinal epithelial cells is associated with cyclin D1 downregulation. Digestive Disease Week, San Francisco, CA. 2002.