Colorectal cancer is one of the most common causes of cancer death throughout the world. Colorectal carcinogenesis is a multi-step process and in addition to genetic changes significant changes in DNA methylation and epigenetic events such as histone modifications can also lead to deregulation of gene expression. The histone proteins play a dynamic role in the chromatin structure and transcription. Aberrant protein acetylation, particularly on histones, has been related to cancer while abnormal expression of histone deacetlytransferases (HDACs) are observed in a broad range of cancer types. Therefore, HDACs have emerged as promising targets in cancer therapeutics.The MRG-binding protein (MRGBP / C20orf20) is a component of the NuA4 histone acetyltransferase (HAT) complex and is involved in the transcriptional activation of specific genes. The MRGBP gene is frequently upregulated in tumors and is thought to play a role in cancer development. On the other hand removal of introns from pre-mRNA is carried out by a large macromolecular spliceosome complex. PPIL1 (Peptidyl-prolylisomerase-like 1) is a component of the human spliceosome and plays a primary role in pre-mRNA splicing for catalyzing isomerization of the peptide bonds. Overexpression of the PPIL1 protein in cancer suggest a potential role of PPIL1 in the in the development and promotion of colon cancer. In the present study we analyzed MRGBP and PPIL1 gene expression levels in matched tumor and normal tissue samples from patients with colorectal cancer.Tissue samples were collected from 51 patients with colorectal cancer who underwent surgery for tumor resection. The normal tissue specimens used in the study were histologically confirmed to be free of cancer. The HPRT (hypoxanthine-Guanine phosphoribosyltransferase) gene was used as reference. The target and reference genes were amplified in the same multiplex PCR using the LightCycler 480 System. 8-9 nucleotide long UPL probes were labeled with fluorescein (FAM) and TAMRA at the 5′ and 3′ ends, respectively. Expression levels were determined using the Basic Relative Quantification analysis software and the results were evaluated by the Mann-Whitney U test.Although the median MRGBP expression levels in the tumor tissue were higher than normal tissue (1,72 vs.1,41) the difference was not significant. The median expression levels of PPIL1 in the tumors and normal tissue samples (1,97 vs 1,86) were also similar. Our results indicate that neither MRGBP nor PPIL1 gene expression levels are upregulated in colon cancer cells indicating that these are not directly associated with colorectal cancer.Citation Format: Ebru Akisik, Nejat Dalay, Sumer Yamaner, Dursun Bugra. Analysis of MRGBP and PPIL1 genes in colon cancer. [abstract]. In: Proceedings of the Third AACR International Conference on Frontiers in Basic Cancer Research; Sep 18-22, 2013; National Harbor, MD. Philadelphia (PA): AACR; Cancer Res 2013;73(19 Suppl):Abstract nr A08.

Introduction: The primary aim of this study is to investigate the effect of change in the expression levels of survivin, glutathione-S-transferase P1 (GSTP1), and topoisomerase 2 alpha (TOP2A) on the response to antracyclin-based and taxane-based neoadjuvant chemotherapy.Methods: This study included 32 locally advanced breast cancer patients. Tumoral expressions of survivin, TOP2A, and GSTP1 in serial biopsy specimens obtained before treatment, after sequential 4 cycles of doxorubicin + cyclophosphomide, and 4 cycles of docetaxel were analyzed by real-time polymerase chain reaction. Survivin expressions were additionally analyzed in serial blood samples.Results: The pathologic complete response (pCR) rate and the overall response rate (clinical complete and partial) were 28% (n = 9) and 91% (n = 29), respectively. There were no statistically significant correlations between serial TOP2A expression levels and response. There was a nonsignificant trend toward an improved response rate with decreased survivin expression. A significant decrease in the GSTP1 expression level throughout treatment (P = 0.014), which was also shown to be significantly correlated with a pCR (P = 0.0001), was seen. Downregulation of GSTP1 after 4 cycles of anthracycline-based combination was independently associated with improved progression-free survival (P = 0.01).Conclusions: Downregulation of GSTP1 is a significant predictor of pCR and improved progression-free survival during anthracyclinebased and taxane-based neoadjuvant chemotherapy in patients with locally advanced breast cancer.

Detection of cancer-related histone modifications of nucleosomes in the circulating blood may be useful in early cancer diagnosis. We have analyzed the trimethylation of histone H3 lysine 9 (H3K9me3) and histone H4 lysine 9 (H4K2Ome3), two modifications involved in heterochromatin formation, at pericentric heterochromatin of circulating nucleosomes in the blood plasma of healthy individuals, patients with either colorectal cancer or multiple myeloma. H3K9me3 was found to be decreased in colorectal cancer and increased in multiple myeloma while H4K2Ome3 levels were similar in all study groups. Therefore, we used the H3K9me3/H4K2Ome3 ratio for normalizing H3K9me3 concentrations; it significantly distinguished patients with colorectal cancer (median 0.8) from the healthy group (median 3) and multiple myeloma (median 4.7) (both p < 0.001). We conclude that if validated in a larger series of cases the ratio H3K9me3/H4K2Ome3 might be a potential diagnostic biomarker for colorectal cancer.

10539 Background: The aim of this study is to investigate the effect of change in expression levels of survivin, glutathione-S-transferase P1 (GSTP1) and topoisomerase 2 alpha (TOP2A) on response to NAC consisting of 4 cycles doxorubicin (60 mg/m2) and cyclophosphamide (600 mg/m2) (AC) followed by 4 cycles docetaxel (100 mg/m2) ± trastuzumab. Methods: Tumoral expression of markers obtained before CT, after 4 cycles of AC and completion of CT, were analyzed by real-time PCR. Survivin expressions were additionally analyzed in serial blood samples. Results: The median age of pts (n=32) was 48 (28-75) yrs and median follow-up was 36.3±12.5 months (9.4-53.6). The majority of pts had clinical stage T3-T4 (n=23, 72%), clinically lymph node positive (n=29, 91%), grade III (n=23, 72%), ER positive (n=17, 53%), PgR positive (n=21, 66%), and HER-2 negative (n=27, 84%) disease. 14 pts (43.8%) developed recurrence and 6 pts (18.8%) died during follow-up. The median PFS was 29.3±14.5 months. The table shows changes in serial expression levels of markers analyzed. Serial GSTP1 expression in pts with clinical partial and complete response (CR) decreased from 1.41 to 0.33 and to 0.19 (p:0.0001). In addition, expression levels in pts with a pathological CR in breast and axilla were 1.85, 0.33 and 0.28 (p:0.0001). Anthracycline-related decrease in GSTP1 expression was determined as the only independent prognostic variable for PFS (p=0.01). Nevertheless, there was not correlation between survivin and TOP2A expression and clinicopathologic parameters, response, and survival. Conclusions: Downregulation of GSTP1 is a significant predictor of pCR and improved PFS during anthracycline and taxane-based NAC for pts with LABC. Monitoring GSTP1 expression levels may be used as an early predictor for response to NAC in LABC pts. Expression levels of markers during treatment (median ± SD). Before CT After four cycles CT After eight cycles CT p GSTP1 2.30±59.7 0.34±25.2 0.24±2.9 0.014 TOP2A 0.27±1.3 0.27±3.8 0.14±0.8 NS Survivin 0.93±5 0.16±3.8 0.19±0.2 NS Survivin in blood 0.87±2.5 0.86±4.8 0.55±2.5 NS

Background. MicroRNAs are a type of small noncoding RNA molecules that have been shown to control gene expression in eukaryotes. Aberrant expression and alteration of miRNAs may be responsible for human diseases including cancer. An miR16-1 (C > T) + 7 gene mutation has been previously found in familial chronic lymphocytic leukemia patients, one of which reported a family history of breast cancer. miR16-1 regulates the expression of bcl-2, which is important in retinoblastoma, and is located in a genomic region that is frequently lost in nasopharyngeal and hepatocellular carcinomas (HCCs). Therefore, miR16-1 may be potentially important in the etiology of several solid tumors. To understand the power of the miR16-1 (C > T) + 7 mutation as a prognostic and diagnostic risk factor, we investigated the mutation in patients with seven different types of cancer including 188 with breast, 102 with ovarian, and 22 nasopharyngeal carcinomas, 96 HCC, 872 chronic myeloid leukemia (CML), 39 chronic lymphocytic leukemia (CLL), and 46 retinoblastoma cases from three different ethnic groups and of hereditary and sporadic etiology. Methods. 5′Nuclease TaqMan SNP genotyping assay was used to detect the miR16-1 gene C > T substitution. Results. The miR16-1 (C > T) + 7 substitution was not detected in any of the groups studied. Conclusions. Considering the large scale of our study, the representation of different ethnicities and levels of hereditary risk, we conclude that the miR-16-1 (C > T) + 7 mutation is not a good diagnostic or prognostic indicator of risk for the cancers tested.

BACKGROUNDalterations in DNA methylation and histone modifications have been implicated in carcinogenesis. Although tumor-specific alterations in DNA methylation can be detected in the serum and plasma of cancer patients, no data are available on the presence of histone modifications in circulating blood. We investigated whether histone methylation, as a model of histone modifications, is detectable in plasma. Because methylation at histone 3 lysine 9 (H3K9) has been demonstrated to be enriched at sites of repetitive ALU elements, we addressed the specificity of histone-methylation detection and hypothesized that if monomethylated H3K9 (H3K9me1) is detectable in plasma, the concentrations in mononucleosomes and oligonucleosomes would be different. We also analyzed a single-copy gene, CDKN2A.METHODSwe enrolled 21 multiple myeloma patients in the study. We used ELISA and real-time PCR analysis to evaluate nucleosomes and cell-free DNA, respectively, as evidence of the presence of histones and associated DNA in circulating blood. H3K9me1 was analyzed by chromatin immunoprecipitation.RESULTSELISA and real-time PCR assays indicated the presence of free nucleosomes and DNA in plasma, and the results were quantitatively correlated (P < 0.001). The detection of histone methylation on free nucleosomes was sequence dependent. Fragments representing mono- and oligonucleosomes differed with respect to H3K9me1 concentrations (P = 0.004), in accordance with our hypothesis. In addition, the detection rate and concentrations of H3K9me1 were significantly higher on the fragment covering both mononucleosomes and oligonucleosomes than on the CDKN2A promoter (P < 0.001).CONCLUSIONSif validated in further studies, our findings may be a basis for investigations of cancer-specific alterations in histone modifications in the circulation.

Background: The integrity of circulating cell-free DNA (cf-DNA) in serum or plasma appears to be of diagnostic and prognostic value in cancer. Here, we investigated the dynamics of serum DNA levels and the size distribution of cf-DNA during adjuvant chemotherapy of patients with breast cancer (n =73).Methods: By evaluating sera taken at the beginning and the end of the adjuvant chemotherapy, variations of serum DNA levels and the size distribution were analyzed, based on quantification of shorter apoptotic and longer non-apoptotic fragments from abundant genomic ALU fragments amplified by quantitative real-time PCR.Results: The mean DNA level did not change significantly during chemotherapy. However, individual cases revealed considerable variation in the amount of serum DNA. It increased in 43.8% of the patients, whereas it decreased in the remaining majority (56.2%). By calculating a "coefficient of variation" (both decrease and increase) in the level of total DNA and non-apoptotic DNA fragments, we compared the values at the beginning and the end of the therapy. For total DNA, the range was between 1.02- and 26-fold (mean 3.76-fold), whereas for non-apoptotic fragments it ranged from 1.01- to 73-fold (mean 6.9-fold) (p = 0.033). In accordance with these findings,the integrity of serum DNA was higher in patients with increasing DNA levels and vice versa.Conclusions: Our findings suggest that non-apoptotic fragments contribute to a higher degree to the change of the DNA level during adjuvant chemotherapy.

It is not known how chemotherapy-induced cell death influences the size distribution of circulating free DNA (cf-DNA) in serum or plasma of cancer patients. In the present study, we investigated the integrity of cf-DNA during adjuvant systemic therapy in patients (n= 41) with invasive breast cancer. Sera taken at the beginning and the end of the adjuvant chemotherapy were comparatively analyzed for the integrity of cf-DNA. The assay was based on quantification of shorter and longer fragments representing apoptotic or non-apoptotic DNA from abundant genomic ALU repeats by quantitative real-time PCR. The ratio of longer to shorter fragments showed the integrity of free serum DNA. During chemotherapy, in half of the patients (51.2%), total DNA levels increased, but decreased in the other half. The distribution of the DNA integrity in the whole patient group after the systemic therapy (median 0.31) did not significantly differ from that at the beginning (median 0.29, P= 0.39). However, in the subgroups, the variation of the DNA integrity was related to the course of the total DNA level. In the subgroup with an increasing DNA level, the median DNA integrity was elevated from 0.27 to 0.39 (P= 0.005), whereas in the group with a decrease it declined from 0.34 to 0.28 (P= 0.044). Our results show that longer fragments released from non-apoptotic cells are the main contributors to increasing DNA levels during adjuvant systemic therapy. This information might be helpful in evaluating the response of patients to adjuvant systemic therapy.

Background: Recent studies suggest that single nucleotide polymorphisms in different genes may modulate the susceptibility to chronic inyelogenous leukaemia (CML). Here, the association of the common XRCC1 gene polymorphism Arg399Gln at codon 399 in CML was investigated. Patients and Methods: Genotyping was performed by melting curve analysis in samples from peripheral blood or bone marrow. Results: The frequency of the variant allele 399Gln was similar between the control group and the patients (35.2% and 34.9%, respectively; p=0.21). Similarly, the heterozygote and homozygote variant genotypes displayed a homogenous distribution in both groups (p>0.05 for all comparisons). Moreover, distribution of the variant allele and subgenotypes did not significantly differ between the patient subgroups with a diagnosis age below or above 50 years. Conclusion: To our knowledge, this is the first study to investigate the role of any XRCC1 polymorphism in CML and our findings do not support a role of codon 399Gln polymorphism in CML.

In the detection of DNA hypermethylation as a tumor-specific epigenetic change in blood mononuclear cell fraction in patients with lymphoid and hematopoetic disorders, circulating tumor cells originating from the lymph nodes or bone marrow can be identified. However, it is still not clear whether methylation in mononuclear cells is disease specific. In the present study, we investigated whether methylation of the inhibitor of cyclin-dependent kinase (INK) 4A/alternative reading frame (ARF) locus is present in a disease-specific manner in the blood mononuclear cell fraction of patients with lymphoma, multiple myeloma, or leukemia. To increase the sensitivity of detection, a two-step methylation-specific PCR approach was used to analyze the methylation status of the promoter/exon 1 regions of both p14ARF and p16INK4A genes. Our findings indicate that although INK4A/ARF locus methylation is present in mononuclear cells, this event is not disease-specific since normal subjects also display methylated DNA in their mononuclear cells. In 85.1% of the patients and in 89% of the controls, p16INK4A gene was methylated, while the methylation rates for the p14ARF gene was 32.6 and 36.5%, respectively. The presence of methylated CpG sites in DNA in samples from normal subjects was confirmed by bisulfite genomic sequencing. The difference in the methylation rate between p16INK4A and p14ARF genes among the patients was highly significant (p<0.001). Our results demonstrate that methylation of the INK4A/ARF locus is not a disease-specific molecular change in mononuclear cell fraction and that the p14ARF and p16INK4A genes are differentially methylated.

Methylenetetrahydrofolate reductase (MTHFR), a key enzyme in folate metabolism, has been implicated in cancer risk. In the present study we used a melting curve analysis to investigate the association of the common MTHFR C677 T polymorphism with lymphoproliferative diseases. Patients (n=117) were compared with age- and sex-matched control subjects (n=154). Our results indicate that the 677 T variant occurred less frequently in patients (26%) than in the control group (33.7%; P=0.05). Investigation of the variant allele (677 T) frequency in the subgroups with Hodgkin's lymphoma (HL) and B-cell neoplasms (BCNs) revealed that this difference was a result of the significantly lower distribution of the variant allele in patients with HL (20.5%; P=0.01). This was accompanied by a significantly higher frequency of the homozygote normal genotype (677CC) among the patients with HL. In patients with BCNs the distribution of the variant allele (30.3%) was comparable to that in the control group (P=0.47). However, the difference between HL (20.5%) and BCNs (30.3%) did not reach statistical significance (P=0.09). Our results suggest that the distribution of the C677 T polymorphism may vary among lymphoproliferative diseases.

Background: Folate deficiency is implicated in cancer development. Single nucleotide polymorphisms in the methylenetetrahydrofolate reductase (MTHFR) gene can modulate the effect of folate. In this case-controlled study, a possible effect of the common MTHFR C677T (ala -> val) polymorphism on breast cancer susceptibility in Turkish patients was investigated. Materials and Methods: Polymorphism analysis was performed by melting curve analysis. Results: The variant allele valine (677T) was more frequent among the patients (30.1%) than in controls (23.9%). This difference was weakly significant (p=0.046; OR=1.37) and due to a significantly higher frequency of the valine homozygotes (677TT) among the patients (12.1% vs. 5.4%; p=0.013, OR=2.5). Among the patients diagnosed at more than 40 years of age, a more pronounced association of the valine homozygotes with breast cancer risk was observed (p=0.009; OR=3.3). Conclusion: Homozygosity for the low-activity C677T genotype (TT) may represent a genetic determinant increasing breast cancer risk.

Elevated amounts of cell-free nucleic acids are detected in the circulation of cancer patients. The type and pattern of these may vary depending on the origin. Recently, we described the presence of circulating fragmented nucleosomal DNA. In the present study, our aim was to investigate the association between nucleosomal DNA, caspase-3 expression and circulating caspase-3 mRNA, the primary activator of nucleosomal DNA fragmentation. DNA fragmentation was analyzed by gel electrophoresis, and caspase-3 was analyzed by RT-PCR. Plasma samples from 65% of patients were positive for nucleosomal DNA fragmentation, whereas none of the healthy controls showed DNA fragmentation (P = 0.0004). Expression of caspase-3 differed significantly between the cases and controls (P < 0.0001). However, we observed no direct correlation between nucleosomal DNA fragmentation and caspase-3 expression in lymphocytes (P = 0.145). Circulating plasma caspase-3 mRNA was detected by nested RT-PCR, and no significant difference was observed between the patients and the control group (P = 0.5). Our results indicate that caspase-3 expression is increased in lymphocytes from patients. When compared to healthy individuals, no differences were observed in the amount of circulating mRNA. These findings suggest that nucleosomal DNA fragmentation is not correlated with elevated levels of apoptosis and circulating caspase-3 mRNA in circulating tumor cells.

The estrogen receptor (ER) and the human epithelial growth factor receptor 2 (HER2) genes have been implicated in the development and prognosis of breast cancer. Several genetic polymorphic sites in these genes have been identified and associated with the risk of breast cancer. We have investigated the association between the estrogen receptor codon 594 (ACA to ACG) and HER2 codon 655 (ATC to GTC) polymorphisms and breast cancer risk. Genomic DNA from breast cancer patients and control subjects was analyzed by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). When allelic frequencies of the ER codon 594 and HER2 codon 655 gene were compared, no significant differences were observed between the patient and control groups. (P = 0.063, OR = 1.55, 95% CI = 0.25–9.41 and P = 0.949, OR = 1.01, 95% CI = 0.55–1.88, respectively). In conclusion, our results support the view that both the ER codon 594 and HER2 codon 655 polymorphisms are not associated with increased risk of breast cancer.

Background: Methylenetetrahydrofolate reductase (MTHFR) plays a critical role in folate metabolism and displays common genetic polymorphisms affecting the enzyme activity. The MTHFR genetic polymorphisms have been associated with a decrease in the risk of developing the lymphoid but not myeloid form of pédiatric and adult leukemias.

To date, BRCA1 and BRCA2 mutations in breast and/or ovarian patients have not been characterized in the Turkish population. We investigated the presence of BRCA mutations in 53 individuals with a personal and family history of breast and/or ovarian cancer, and 52 individuals with a personal history of breast cancer diagnosed below age 50 without additional family history. We have identified 11 mutations (nine BRCA1 and two BRCA2) using combined techniques involving protein truncation test, direct sequencing and heteroduplex analysis. We found eight out of 53 patients (15.1%) with a family history to carry BRCA gene mutations (seven BRCA1 and one BRCA2). Of these, four were found in 43 families presenting only breast cancer histories, and four were found in families presenting ovarian cancer with or without breast cancer. We also demonstrated two BRCA1 and one BRCA2 mutations in three out of 52 (5.8%) early-onset breast cancer cases without additional family history. Three of nine BRCA1 and both BRCA2 mutations detected in this study were not reported previously. These mutations may be specific to the Turkish population. The BRCA1 5382insC mutation, specific to Ashkenazi and Russian populations, was found twice in our study group, representing a possible founder mutation in the Turkish population. © 2000 Cancer Research Campaign