الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 170 |  |  |
| | | from | 170 | | |
Abstract
In the present study we assessed the methylation of KLK10 exon 3 using methylation-specific PCR (MSP). The included samples were archival FFPE-ovarian tissue samples that are available at the Early Cancer Detection Unit, Ain Shams University Maternity Hospital. The studied ovarian tissue samples were 16 benign and 16 malignant tumors. In addition, 16 normal samples were included as controls.
The experimental protocol was: 1) DNA extraction using commercially available DNA extraction kit for FFPE-tissue, 2) DNA Bisulfite conversion using commercially available DNA Bisulfite conversion kit and 3) MSP using HotStarTaq master Mix kit. The PCR products were assessed by conventional 2% agarose gel electrophoresis and capillary gel electrophoresis.
Qualitative MSP results in agarose electrophoresis showed that only one malignant sample (serous adenocarcinoma) was positive for the methylated-specific PCR, whereas the rest of samples (47 samples) was negative for methylated-specific PCR. On the other hand, only one sample (Sertoli Leydig cell tumor) was negative for unmethylated-specific PCR, whereas the rest of samples (47 samples) was positive for unmethylated-specific PCR. Because these results obviously showed no statistically significant differences in patient group samples, we reassessed the PCR products by capillary gel electrophoresis with the hope to find more informative qualitative data and/ or compare the concentration of unmethylated PCR product concentration (ng/µl) in the samples.
The qualitative MSP results by capillary gel electrophoresis were the same as those in agarose gel electrophoresis except for only one sample whose results were included only in the qualitative data but not in the comparisons of the unmethylated PCR product concentration. Comparison of the unmethylated PCR product concentration, however, showed statistically significant differences not only in benign and malignant samples but also in normal control samples of the benign and malignant patients.
The diagnostic value of the unmethylated PCR product concentration (ng/µl) in ovarian tissue samples was further evident by means of ROC curve analysis (AUC = 0.778; p-value = 0.002). These results may denote possible diagnostic potential for KLK10 exon 3 unmethylated PCR product concentration as an epigenetic change in ovarian tissue tumors with morphologically-evident events and in otherwise histopathologically normal samples as well.
Correlating serum KLK10 level to KLK10 exon 3 unmethylated PCR product concentration (ng/µl) revealed statistically significant negative correlation by both Spearman’s correlation and by Student t-test. Because the KLK10 exon 3 unmethylated PCR product concentration as shown in the present results is higher in normal and benign samples than in malignant samples, the negative correlation between KLK10 exon 3 unmethylated PCR product and serum KLK10 level agrees with the known expression data of KLK10 gene in ovarian tumor patients. Nonetheless, Correlating KLK10 exon 3 unmethylated PCR product concentration to patient 5-year survival did not show statistical significance.
We hypothesized a theoretical rationale for MSP results based on previous knowledge about DNA methylation, MSP internal working and previous DNA methylation results by MSP and direct DNA sequencing in one and the same sample sets. The proposal was meant to rationalize for the unmethylated PCR product concentration as an indirect and reciprocal measure of the extent of methylated cytosines in the target DNA segments. It consisted of two main assumptions: 1) some partially methylated DNA segments may be MSP-inexpressible, i.e. qualitatively negative, when each target DNA segment is met with its methylation-wise opposite methylation-specific primer pair and 2) the likelihood that a methylation-specific primer would anneal to its target DNA is directly proportional to the number of methylation-wise similar cytosines in the DNA target.
We threw light on the reported significance of the exact location of the methylated cytosines, on the one hand, and methylation pattern distribution in the sample, on the other hand, for possible clues of gene functions such as gene expression. Given the accuracy of these notions, the DNA methylation data provided by MSP would be largely uninformative in the clinical and translational contexts. Accordingly, DNA methylation data should not be interpreted irrespective of their consistent and well validated type of information.
To the best of our knowledge, the present results are the first report on the possible role of KLK10 exon 3 unmethylated PCR product concentration as potential early diagnostic epigenetic marker in ovarian tumors. Taking into account the limitations of the present study, e.g. small sample size and semi-quantitative assessment of PCR product concentration, further studies with larger sample size and quantitative methods are warranted.