Bernard W Futscher
High dimensionality has been a major problem for gene array-based cancer classification. It is critical to identify marker genes for cancer diagnoses. We developed a framework of gene selection methods based on previous studies. This paper focuses on optimal search-based subset selection methods because they evaluate the group performance of genes and help to pinpoint global optimal set of marker genes. Notably, this paper is the first to introduce tabu search (TS) to gene selection from high-dimensional gene array data. Our comparative study of gene selection methods demonstrated the effectiveness of optimal search-based gene subset selection to identify cancer marker genes. TS was shown to be a promising tool for gene subset selection. © 2007 IEEE.
PMID: 19509227;PMCID: PMC2697259;Abstract:
The timing and progression of DNA methylation changes during carcinogenesis are not completely understood. To develop a timeline of aberrant DNA methylation events during malignant transformation, we analyzed genome-wide DNA methylation patterns in an isogenic human mammary epithelial cell (HMEC) culture model of transformation. To acquire immortality and malignancy, the cultured finite lifespan HMEC must overcome two distinct proliferation barriers. The first barrier, stasis, is mediated by the retinoblastoma protein and can be overcome by loss of p16INK4A expression. HMEC that escape stasis and continue to proliferate become genomically unstable before encountering a second more stringent proliferation barrier, telomere dysfunction due to telomere attrition. Rare cells that acquire telomerase expression may escape this barrier, become immortal, and develop further malignant properties. Our analysis of HMEC transitioning from finite lifespan to malignantly transformed showed that aberrant DNA methylation changes occur in a stepwise fashion early in the transformation process. The first aberrant DNA methylation step coincides with overcoming stasis, and results in few to hundreds of changes, depending on how stasis was overcome. A second step coincides with immortalization and results in hundreds of additional DNA methylation changes regardless of the immortalization pathway. A majority of these DNA methylation changes are also found in malignant breast cancer cells. These results show that large-scale epigenetic remodeling occurs in the earliest steps of mammary carcinogenesis, temporally links DNA methylation changes and overcoming cellular proliferation barriers, and provides a bank of potential epigenetic biomarkers that may prove useful in breast cancer risk assessment. ©2009 American Association for Cancer Research.
PMID: 22064654;PMCID: PMC3262094;Abstract:
Extracellular superoxide dismutase (EcSOD) is an important superoxide scavenger in the lung in which its loss, sequence variation, or abnormal expression contributes to lung diseases; however, the role of EcSOD in lung cancer has yet to be studied. We hypothesized that EcSOD loss could affect malignant progression in lung, and could be either genetic or epigenetic in nature. To test this, we analyzed EcSOD expression, gene copy number, promoter methylation, and chromatin accessibility in normal lung and carcinoma cells.Wefound that normal airway epithelial cells expressed abundant EcSOD and had an unmethylated promoter, whereas EcSOD-negative lung cancer cells displayed aberrant promoter hypermethylation and decreased chromatin accessibility. 5-aza-dC induced EcSOD suggesting that cytosine methylation was causal, in part, to silencing. In 48/50 lung tumors, EcSOD mRNA was significantly lower as early as stage I, and the EcSODpromoter was hypermethylated in 8/10 (80%) adenocarcinomas compared with 0/5 normal lung samples. In addition, 20% of the tumors showed loss of heterozygosity (LOH) of EcSOD. Reexpression of EcSOD attenuated the malignant phenotype of lung carcinoma cells by significantly decreasing invasion and survival. Finally, EcSOD decreased heparanase and syndecan-1 mRNAs in part by reducing NF-κB. By contrast, MnSOD and CuZnSOD showed no significant changes in lung tumors and had no effect on heparanase expression. Taken together, the loss of EcSOD expression is unique among the superoxide dismutases in lung cancer and is the result of EcSOD promoter methylation and LOH, suggesting that its early loss may contribute to ECM remodeling and malignant progression. ©2011 AACR.
Both p15 and p16 are tumor suppressor genes that have 5' CpG islands; aberrant cytosine methylation of these islands has been associated with silencing of their expression. Deoxycytidine kinase (dCK) converts prodrugs to their cytotoxic form, has a 5' CpG island and is a candidate gene for inactivation by hypermethylation. In our study, we used sodium bisulfite sequencing to generate high resolution maps of 5-methylcytosine in the CpG islands associated with p15, p16 and dCK in normal human bone marrow (BM), peripheral blood lymphocytes (PBL) and cytosine arabinoside (ara-C)-resistant acute myeloid leukemia (AML) patients, and established human hematopoietic tumor cell lines. In normal cells the p15, p16 and dCK CpG islands were largely unmethylated. The p16 and dCK CpG islands were also unmethylated in the 8 AML specimens. In contrast, the p15 CpG island was aberrantly methylated in 6 of the 8 AML specimens. Furthermore, bisulfite sequencing revealed that the p15 CpG island is heterogeneously methylated in AML, with large intra-individual and inter-individual variability.
The risk of breast cancer transiently increases immediately following pregnancy; peaking between 3-7 years. The biology that underlies this risk window and the effect on the natural history of the disease is unknown. MicroRNAs (miRNAs) are small non-coding RNAs that have been shown to be dysregulated in breast cancer. We conducted miRNA profiling of 56 tumors from a case series of multiparous Hispanic women and assessed the pattern of expression by time since last full-term pregnancy. A data-driven splitting analysis on the pattern of 355 miRNAs separated the case series into two groups: a) an early group representing women diagnosed with breast cancer ≤ 5.2 years postpartum (n = 12), and b) a late group representing women diagnosed with breast cancer ≥ 5.3 years postpartum (n = 44). We identified 15 miRNAs with significant differential expression between the early and late postpartum groups; 60% of these miRNAs are encoded on the X chromosome. Ten miRNAs had a two-fold or higher difference in expression with miR-138, miR-660, miR-31, miR-135b, miR-17, miR-454, and miR-934 overexpressed in the early versus the late group; while miR-892a, miR-199a-5p, and miR-542-5p were underexpressed in the early versus the late postpartum group. The DNA methylation of three out of five tested miRNAs (miR-31, miR-135b, and miR-138) was lower in the early versus late postpartum group, and negatively correlated with miRNA expression. Here we show that miRNAs are differentially expressed and differentially methylated between tumors of the early versus late postpartum, suggesting that potential differences in epigenetic dysfunction may be operative in postpartum breast cancers.