PATTERNS OF URINARY METHYLATED CYTOSINE DERIVATIVES IN A SAMPLE OF HEALTHY AND UNHEALTHY PEOPLE

摘要

Aims: Methylation of Cytosine residues in DNA is an epigenetic mechanisms modulating gene expression, that shows perturbations in several diseases, particularly in cancer. The mechanisms and roles of hyper- or hypo-methylation of DNA are, however, still largely unclear and both conditions may be associated with oxidative stress. The methylation status of specific genes in target tissues can be assessed relying on molecular biology, but such methods can provide only an indirect and approximate estimation of the methylation status of whole genome. We developed and validated a new method based on liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) to investigate the urinary patterns of methylated cytosine derivatives released from both DNA and RNA. Methods: We recruited 46 patients affected by Parkinson's Disease (PD) and 62 patients with non-small cell lung cancer (NSCLC), and 154 healthy subjects comparable for age. Urinary 5-Methyl-Cytosine (5-MeCyto), and 5-Methyl-Cytidine (5-MeCyt)] and their non-methylated forms [Cytosine (Cyto), and Cytidine (Cyt)] as well as cotinine (as biomarker of exposure to tobacco smoke) were measured by LC-MS/MS. Parameters of method validation, matrix effect, and stability of samples were also studied. Urinary biomarkers were expressed as a function of creatinine (creat). Due to the skewed distribution of data, parametric tests were applied after log-transformation, central tendency and dispersion being reported as geometric mean and GSD, respectively. Results: Urinary levels of both methylated forms were lower in NSCLC subjects [5-MeCyto 14.5 (2.02) μmol/mol creat and 5-MeCyt 1.18 (2.31) μmol/mol creat] as compared to both healthy subjects [5-MeCyto 19.6 (1.86) μmol/mol creat and 5-MeCyt 3.41 (2.22) μmol/mol creat] and PD subjects [5-MeCyto 24.3 (1.85) μmol/mol creat and 5-MeCyt 4.61 (2.51) μmol/mol creat]. One way ANOVA was highly significant (p<0.001), such a difference among groups being accounted for by lower values among lung cancer patients. Methylation ratios [5-MeCyto/(5-MeCyto+Cyto) and 5-MeCyt/(5-MeCyt+Cyt)] showed the same trend. Results were unaffected by either gender or smoking habits, whereas age was negatively correlated with 5-MeCyt and 5-MeCyt/(5-MeCyt+Cyt)] ratio (p<0.0001, r Pearson = -0.357 and r = - 349, respectively), but not with the other markers. Discussion: In agreement with studies suggesting DNA hypo-methylation in cancer, both 5-MeCyto and 5-MeCyt concentrations were lower in NSCLC as compared to both healthy and PD subjects. Aging was negatively correlated with urinary concentrations of 5-MeCyt (specific of RNA methylation) and 5-MeCyt/(5-MeCyt+Cyt)] ratio, whereas neither smoking habits nor gender did influence urinary levels of 5-MeCyto and 5-MeCyt. Further investigations are necessary to better clarify the mechanisms and roles of hyper- or hypo-methylation of DNA and RNA.