Molecular Carcinogenesis

Nrf2 is the key transcription factor regulating the antioxidant response. Nrf2 signaling is repressed by Keap1 at basal condition and induced by oxidative stress. Keap1 is recently identified as a Cullin 3‐dependent substrate adaptor protein. A two‐sites binding “hinge & latch” model vividly depicts how Keap1 can efficiently present Nrf2 as substrate for ubiquitination. Oxidative perturbation can impede Keap1‐mediated Nrf2 ubiquitination but fail to disrupt Nrf2/Keap1 binding. Nrf2 per se is a redox‐sensitive transcription factor. A new Nrf2‐mediated redox signaling model is proposed based on these new discoveries. Free floating Nrf2 protein functions as a redox‐sensitive probe. Keap1 instead functions as a gate keeper to control the availability of Nrf2 probes and thus regulates the overall sensitivity of the redox signaling. © 2008 Wiley‐Liss, Inc.

Several solid tumors have now been shown to contain stem cell‐like cells called cancer stem cells (CSC). These cells, although generally rare, appear to be highly tumorigenic and may be the cells that drive tumor formation, maintain tumor homeostasis, and mediate tumor metastasis. In this Perspective, we first provide our insight on how a CSC should be defined. We then summarize our current knowledge of stem/progenitor cells in the normal human prostate (NHP), an organ highly susceptible to hyperproliferative diseases such as benign prostate hyperplasia (BPH) and prostate cancer (PCa). We further review the evidence that cultured PCa cells, xenograft prostate tumors, and patient tumors may contain stem/progenitor cells. Along with our discussion, we present several methodologies that can be potentially used to identify putative tumor‐reinitiating CSC. Finally, we present a hypothetical model for the hierarchical organization of human PCa cells and discuss the implications of this model in helping understand prostate carcinogenesis and design novel diagnostic, prognostic, and therapeutic approaches. © 2006 Wiley‐Liss, Inc.

Recent studies have shown that cytosine‐5 methylation at CpG islands in the regulatory sequence of a gene is one of the key mechanisms of inactivation. The enzymes responsible for CpG methylation are DNA methyltransferase (DNMT) 1, DNMT3a, and DNMT3b, and the enzyme responsible for demethylation is DNA demethylase (MBD2). Studies on methylation‐demethylation enzymes are lacking in human prostate cancer. We hypothesize that MBD2 enzyme activity is repressed and that DNMT1 enzyme activity is elevated in human prostate cancer. To test this hypothesis, we analyzed enzyme activities, mRNA, and protein levels of MBD2 and DNMT1, DNMT3a, and DNMT3b in human prostate cancer cell lines and tissues. The enzyme activities of DNMTs and MBD2 were analyzed by biochemical assay. The mRNA expression was analyzed by reverse transcriptase–polymerase chain reaction and by Northern blotting. The protein expression was measured by immunohistochemistry with specific antibodies. The results of these experiments demonstrated that (1) the activity of DNMTs was twofold to threefold higher in cancer cell lines and cancer tissues, as compared with a benign prostate epithelium cell line (BPH‐1) and benign prostatic hyperplasia (BPH) tissues; (2) MBD2 activity was lacking in prostate cancer cell lines but present in BPH‐1 cells; (3) immunohistochemical analyses exhibited higher expression of DNMT1 in all prostate cancer cell lines and cancer tissues, as compared with BPH‐1 cell lines and BPH tissues; (4) MBD2 protein expression was significantly higher in BPH‐1 cells and lacking in prostate cancer cell lines and, in BPH tissues, MBD2 protein expression was poorly observed, as compared with no expression in prostate cancer tissues; and (5) mRNA expression for DNMT1 was upregulated in prostate cancer, as compared with BPH‐1, and mRNA expression for MBD2 was found to be significantly expressed in all cases. The results of these studies clearly demonstrate that DNMT1 activity is upregulated, whereas MBD2 is repressed at the level of translation in human prostate cancer. These results may demonstrate molecular mechanisms of CpG hypermethylation of various genes in prostate cancer. © 2002 Wiley‐Liss, Inc.

Manganese superoxide dismutase (MnSOD) scavenges toxic superoxide radicals produced in the mitochondria. Transfection of the human MnSOD gene into mouse C3H 10T1/2 cells resulted in production of active MnSOD, which was properly transported into mitochondria. Overexpression of MnSOD protected cells from radiation‐, but not chemically‐induced neoplastic transformation. This finding demonstrates that oxidative stress that occurs in the mitochondria plays an important role in the development of neoplastic transformation. © 1992 Wiley‐Liss, Inc.

Peroxisome proliferator–activated receptor (PPAR) α is a ligand‐activated transcription factor that has been linked with rodent hepatocarcinogenesis. It has been suggested that PPARα mRNA expression levels are an important determinant of rodent hepatic tumorigenicity. Previous work in rat mammary gland epithelial cells showed significantly increased PPARα mRNA expression in carcinomas, suggesting the possible role of this isoform in rodent mammary gland carcinogenesis. In this study we sought to determine whether PPARα is expressed and dynamically regulated in human breast cancer MCF‐7 and MDA‐MB‐231 cells. Having established the presence of PPARα in both cell types, we then examined the consequence of PPARα activation, by its ligands Wy‐14,643 and clofibrate, on proliferation. With real‐time reverse transcriptase–polymerase chain reaction, we showed that PPARα mRNA was dynamically regulated in MDA‐MB‐231 cells and that PPARα activation significantly increased proliferation of the cell line. In contrast, PPARα expression in MCF‐7 cells did not change with proliferation during culture and was present at significantly lower levels than in MDA‐MB‐231 cells. However, PPARα ligand activation still significantly increased the proliferation of MCF‐7 cells. The promotion of proliferation in breast cancer cell lines following PPARα activation was in stark contrast to the effects of PPARγ‐activating ligands that decrease proliferation in human breast cancer cells. Our results established the presence of PPARα in human breast cancer cell lines and showed for the first time that activation of PPARα in human breast cancer cells promoted proliferation. Hence, this pathway may be significant in mammary gland tumorigenesis. © 2002 Wiley‐Liss, Inc.

Epidemiologic, pharmacologic, clinical, and experimental studies document the importance of prostaglandin (PG) signaling in cancer development, including non‐melanoma skin cancer lesions in humans and mice. First of all, enzymes involved in PG biosynthesis, such as cyclooxygenase (COX)‐2 and/or membrane prostaglandin E synthase (mPGES)‐1, were found to be overexpressed in a wide range of premalignant and malignant epithelial tumors, including those of the skin, breast, esophagus, stomach, colorectum, pancreas, and bladder. On the other hand, 15‐hydroxy‐prostaglandin dehydrogenase (15‐PGDH), which is involved in the degradation pathway of PG including PGE_2, thus counteracting the activities of COX‐2 and PGES, was found to be downregulated in human epithelial tumors, indicating a tumor suppressor activity of this enzyme. Most remarkably, genetic studies showed that mice, which are deficient in COX‐2 and/or PGES are resistant to the development of cancer of skin, colon, and stomach. In contrast, the forced overexpression of COX‐2 in proliferative compartments of simple or stratified epithelia such as skin epidermis, urinary bladder, mammary gland, and pancreas results in spontaneous hyperplasia and dysplasia in transgenic mice. In skin, the pathological changes are found to be due to an abnormal process of terminal differentiation, while in other tissues, hyperproliferation seems to be the main contributer to the pre‐invasive neoplasms. Moreover, the COX‐2 transgenic mouse lines are sensitized for cancer development. © 2007 Wiley‐Liss, Inc.

We designed our experiments to evaluate whether fatty acid synthase (FAS), a lipogenic enzyme linked to tumor virulence in population studies of human cancer, is necessary for the malignant transformation induced by Her‐2/neu (erbB‐2) oncogene, which is overexpressed not only in invasive breast cancer but also in premalignant atypical duct proliferations and in ductal carcinoma in situ of the breast. To avoid the genetic complexities associated with established breast cancer cell lines, we employed NIH‐3T3 mouse fibroblasts engineered to overexpress human Her‐2/neu coding sequence. NIH‐3T3/Her‐2 cells demonstrated a significant upregulation of FAS protein expression, which was dependent on the upstream activation of mitogen‐activated protein kinase and phosphatidylinositol 3′‐kinase/AKT pathways. Remarkably, pharmacological FAS blockade using the mycotoxin cerulenin or the novel small compound C75 completely suppressed the state of Her‐2/neu‐induced malignant transformation by inhibiting the ability of NIH‐3T3/Her‐2 cells to grow under either anchorage‐independent (i.e., to form colonies in soft agar) or low‐serum monolayer conditions. Moreover, NIH‐3T3/Her‐2 fibroblasts were up to three times more sensitive to chemical FAS inhibitors relative to untransformed controls as determined by MTT‐based cell viability assays. In addition, pharmacological FAS blockade preferentially induced apoptotic cell death of NIH‐3T3/Her‐2 fibroblasts, as determined by an ELISA for histone‐associated DNA fragments and by the terminal deoxynucleotidyltransferase (TdT)‐mediated nick end labeling assay (TUNEL). Interestingly, the degree of Her‐2/neu oncogene expression in a panel of breast cancer cell lines was predictive of sensitivity to chemical FAS inhibitors‐induced cytotoxicity, while low‐FAS expressing and chemical FAS inhibitors‐resistant MDA‐MB‐231 breast cancer cells became hypersensitive to FAS blockade when they were engineered to overexpress Her‐2/neu. Our observations strongly suggest that inhibition of FAS activity may provide a new molecular avenue for chemotherapeutic prevention and/or treatment of Her‐2/neu‐related breast carcinomas. © 2004 Wiley‐Liss, Inc.

The solution structure of the transmembrane‐4 superfamily protein KAI1, a recently identified prostate cancer metastasis suppressor gene that encodes a 267–amino acid protein, was modeled. The structure of this four‐helical transmembrane protein was developed by defining and modeling sections individually. A complete three‐dimensional structure for the solvated protein was developed by combining the individually modeled sections. The four‐helix transmembrane bundle structure was predicted combining information from several methods including Fourier transform analysis of residue variability for helix orientation. The structure of the KAI1 large extracellular domain was modeled based on the solved crystal structure of the extracellular domain of another tetraspanin superfamily protein member, CD81 (hepatitis C virus envelope E2 glycoprotein receptor). This is a novel protein fold consisting of five alpha helices held together by two disulfide bonds for which the CD81 protein is the first solved representative. Molecular dynamics studies were performed to test stability and to relax the total model KAI1 structure's solution. The resulting KAI1 structural model should be a useful tool for predicting modes of self‐association and associations with other TM4SF proteins, integrins, cadherins, and other KAI1 binding partners. Mutations for probing the interactions of KAI1 with antibodies and with other binding partners are suggested. Published 2001 Wiley‐Liss, Inc.

Strain A/J mice received intraperitoneal injections of benz[j]aceanthrylene (B[j]A) or benzo[a]pyrene (B[a]P). At 24, 48, and 72 h, lung tissues were removed for analysis of B[a]P‐ or B[j]A‐derived DNA adduct formation during the first 3 d of exposure. One group of mice exposed to these hydrocarbons was kept for 8 mo to determine lung tumor multiplicity, the occurrence of mutations in codons 12 and 61 of the Ki‐ras gene in the tumors that arose, the relationship between Ki‐ras oncogene mutations in tumors, and the presence and quantity of genomic DNA adducts. The major DNA adduct in the lungs of mice exposed to B[a]P was N²‐(10β‐[ + B,7α, 9α‐trihydroxy‐7,8,9,10‐tetrahydrobenzo[a]pyrene]yl)‐deoxyguanosine (BPDE‐l‐dGuo) arising from bay‐region diolepoxide activation of B[a]P and was consistent with the occurrence of tumors with mutations GGT → TGT (56%), GGT → GTT (25%), and GGT → GAT (19%) in codon 12, all involving mutations of a guanine. B[j]A, a demethylated analogue of 3‐methylcholanthrene (3‐MCA) with an unsaturated cyclopenta ring, produced 16‐ to 60‐fold more tumors at equivalent doses than did B[a]P; the mutations in tumors were GGT → TGT (4%), GGT → GTT (30%), and GGT → CGT (65%). Analysis of adduction patterns in DNA suggested that B[j]A was activated to form DNA‐binding derivatives in A/J mouse lungs primarily at the cyclopenta ring even though B[j]A contains a bay region. As reported in the published literature, the mutation spectrum induced by 3‐MCA in Ki‐ras codon 12 of mouse cells is similar to that of B[a]P but not to that of its close relative B[j]A. In contrast to B[j]A, 3‐MCA is activated mostly via a bay‐region diol‐epoxide since its cyclopenta ring is saturated and not easily epoxidated. Therefore, we propose that the GGT → CGT mutations produced by B[j]A in Ki‐ras codon 12 were mostly the result of cyclopentaring‐derived adducts.