Molecular Diagnosis of Neoplasia in Barrett's Esophagus
Barrett's esophagus is a well-described risk factor and precursor to esophageal adenocarcinoma. The current standard of care is based on endoscopic surveillance. Observation through the endoscope is imprecise, however, because dysplasia, early-stage neoplasia, is difficult to see and can be missed on biopsy. Because molecular abnormalities may be more diffuse than histologic evidence of dysplasia, Alvi and colleagues tested whether aberrant DNA methylation, especially on imprinted and X-chromosome genes, could be used to improve detection of neoplastic progression on a molecular level. They used 27K methylation arrays on samples from patients with Barrett's esophagus–related lesions that included 22 cases of Barrett's esophagus (no dysplasia or neoplastic progression) and 24 esophageal adenocarcinoma samples. Interestingly, up to 70% of known imprinted genes were aberrantly methylated in association with neoplastic change. Alvi and colleagues found a set of X-linked genes with significantly different methylation patterns between the 2 patient subsets. Iterative validation led the identification of a set of 4 genes [SLC22A18, PIGR, GJA12 (GJC2), and RIN2]. These 4 genes together showed an impressive 0.988 receiver operating characteristic AUC to distinguish nonneoplastic from dysplastic/adenocarcinoma samples. Further analysis of the 4 genes showed the ability to separate patient cohorts into 3 risk groups based on the number of methylated genes (low risk, <2 genes; intermediate, 2 genes; and high, >2 genes). Of 17 patients with fewer than 2 methylated genes, none had adenocarcinoma and 3 had dysplasia. Conversely, among 55 highly methylated cases, 17 had adenocarcinoma and 7 had dysplasia. The tumors with 2 methylated markers showed an intermediate proportion of cases with adenocarcinoma and dysplasia. Overall, these results show some interesting biology of Barrett-associated neoplasia, including aberrant methylation in approximately 70% of known imprinted genes. In addition, these findings provide a potentially important diagnostic test to more precisely define neoplastic progression in this clinical setting. Given the difficulty of identifying lesional tissue by visual inspection and the resulting interobserver variability, the use of a molecular test that is insensitive to such variability could better personalize care for this disease spectrum.
Tumor-Suppressive Role of miR-31 in Regulating Androgen Receptor Expression and Prostate Cancer Progression
A significant regulatory role of microRNAs (miRNA) as important contributors to cancer development and progression is emerging. Moreover, specific miRNAs have the potential to serve as biomarkers and may provide targets for therapeutic development. Lin and colleagues examined the role of miR-31(MIR31) in regulating prostate tumorigenesis. The expression level of miR-31 appeared to be significantly reduced in approximately 93% (13/14) of paired prostate carcinoma cases. Further validation in an additional 40 cases of prostate carcinoma confirmed significant downregulation of miR-31 when these cases were compared with 15 benign prostate tissues. The investigators also uncovered cancer-specific DNA hypermethylation as one potential mechanism controlling miR-31 silencing in multiple primary prostate carcinomas. Treatment of VCaP cells with the DNA methylation inhibitor 5-aza-dC rescued miR-31 expression levels, indicating promoter methylation as a mediator of miR-31 suppression. A significant association between miR-31 promoter hypermethylation and aggressive disease development was also observed in primary prostate carcinoma samples. Other than promoter hypermethylation, further functional analysis showed that androgen receptor and EZH2/H3K27 trimethylation (H3K27me3) mediated downregulation of miR-31 expression in prostate carcinoma cells. In concurrent experiments, miR-31 was shown to suppress androgen receptor activity in prostate carcinoma cells, and this effect was abrogated by overexpression of the androgen receptor. The investigators also identified the specific miRNA recognition element (MRE) in the androgen receptor genes: MRE1 and MRE4 were identified at the 5′UTR region, whereas MRE2 and MRE3 were located at the coding sequence; MRE2 was in the ligand-binding domain, and MRE3 was near the DNA binding domain. Moreover, MRE2, located in the ligand-binding domain of the androgen receptor gene, was shown to be in the specific region for miR-31 binding and thereby regulates androgen receptor expression. Subsequent whole-genome expression analysis on miR-31–overexpressing LNCaP cells identified cell-cycle regulatory genes as the major targets of miR-31, including E2F1, a well-known transcription factor that regulates androgen receptor expression. In addition, several putative miR-31 MREs at 3′UTR of key cell-cycle regulator genes CDK1, E2F2, EXO1, FOXM1, and MCM2 were identified. The results of this study support the use of miR-31 as a potential molecule for biomarker and/or therapeutic development in prostate cancer.
Tumor Suppressor Functions of SPARC in Bladder Cancer
Secreted protein, acidic and rich in cysteine (SPARC, osteonectin), is a matricellular protein implicated in diverse processes, including the regulation of cell–cell and cell–matrix interactions and regulation of the inflammatory response. In cancer, protumorigenic and antitumorigenic roles for SPARC have both been identified. Said and colleagues tested the role for SPARC in a chemical carcinogenesis model for bladder cancer, and their findings suggest that loss of SPARC may be important in cancer progression. In addition, the authors address the contribution of tumor- and stroma-derived SPARC to tumorigenesis. Using human bladder cancer samples, the authors show an association between decreased expression of SPARC in tumor cells and prolonged disease-specific survival. To test this association, the authors used a carcinogen-induced model of bladder cancer in which SPARC expression normally decreases with urothelial transformation in combination with Sparc−/− mice. In the absence of SPARC, bladder carcinogenesis was accelerated, survival decreased, and the incidence of metastatic lesions increased. Tumors showed an increase in markers of oxidative stress, proliferation of tumor cells, and activation of the NF-κB and p38 (MAPK14)–JNK (MAPK8)-AP-1 signaling pathways. In Sparc−/− mice, tumor-associated macrophages were increased in both preneoplastic and neoplastic lesions. Using conditioned media (CM) from cultured monocyte-derived macrophages and from Sparc−/− urothelial carcinoma cells, the authors showed that a factor from the CM of either cell type could promote migration of the other cell type. Interestingly, this effect was greatest if both the cells that produced the CM and the target cells were Sparc−/−. They observed a similar stimulation of tumor cell invasion with CM from Sparc−/− carcinoma-associated fibroblasts. These data suggest that SPARC-regulated factors in stromal cells could contribute to tumorigenesis. Indeed, implantation of a urothelial carcinoma cell line subcutaneously into Sparc−/− or Sparc+/+ mice led to increased primary tumor growth and decreased survival in the Sparc−/− mice. Furthermore, intravenous injection of the cell line to model blood-borne metastasis resulted in a greater number of arrested or colonized tumor cells in the lung of Sparc−/− mice. Their finding of a similar SPARC-dependent change in tumor cell proliferation and tumor volume with gene knockdown and overexpression show the potential relevance of these results to human bladder cancer. This study suggests important roles for loss of SPARC expression in bladder cancer and highlights potential functions for SPARC in both urothelial cells and stromal cells.
Guarding the Genome against Inflammation
TP53 is one of the most important human tumor suppressor genes and is commonly mutated in human colorectal cancer. This tumor suppressor function is generally lost during transition from advanced adenomas to invasive cancers. Schwitalla and colleagues provide novel insights into the ways in which loss of p53 may encourage tumor progression. Using mice with an intestinal epithelial cell–specific p53 (Trp53) deletion, they showed that although loss of p53 was insufficient to initiate intestinal tumorigenesis, loss of p53 led to markedly enhanced carcinogen-induced tumor incidence, invasive cancers, and lymph node metastasis. Loss of p53 during tumor progression was associated with increased intestinal permeability, generating an NF-κB–dependent inflammatory tumor microenvironment and inducing epithelial–mesenchymal transition in the epithelial cells. NF-κB activation was indirectly triggered by elevated levels of circulating plasma lipopolysaccharide, a consequence of impaired intestinal epithelial barrier function in response to carcinogen exposure. Although attenuated tight-junction integrity was also observed in p53 wild-type tumors, this attenuation was aggravated by loss of p53 function. In human cancers, p53 function is presumably lost initially in only a few cells. In contrast, the entire colonic epithelium within this mouse model lacks p53. However, previous work showed that the transition from adenoma to invasive cancer in patients is also associated with an increased barrier defect. These data also underscore the relevance of intestinal microorganisms for colon cancer progression. With these findings, Schwitalla and colleagues provide further in vivo evidence that activation of certain oncogenic pathways initiates an intrinsic inflammatory response with production of cytokines and chemokines by the malignant cells and recruitment of tumor-promoting immune cells, indicating that cancer causes inflammation and that inflammation also causes cancer. This mouse model should be a useful tool for studying the actions of anti-inflammatory agents, including aspirin, on cancer progression.
TERT Mutations in Melanoma
Horn and colleagues performed linkage analysis and sequencing in a family with inherited predisposition to cutaneous melanoma and identified a promoter mutation in the telomerase reverse transcriptase (TERT) gene encoding the catalytic subunit of telomerase. The mutation created a new binding motif for Ets/TCF transcription factors, leading to increased transcription of TERT. Screening of sporadic melanomas led to identification of analogous TERT mutations in approximately 75% of melanoma cell lines and in 33% of primary tumors. In a different study, through whole-genome sequencing followed by analysis of a validation set, Huang and colleagues identified analogous TERT mutations in a majority of melanoma tumors and cell lines.
Note: Breaking Advances are written by Cancer Research Editors. Readers are encouraged to consult the articles referred to in each item for full details on the findings described.
RSS Feeds