br Materials and methods br Results br Discussion Early

Materials and methods
Results
Discussion Early studies in the mid 1980s tentatively identified cCMP in tissues using fast atom bombardment MS [3], [4], but specificity and sensitivity of the method are insufficient for unequivocal identification as compared to modern MS analytical methods. Specificity of cCMP BQ-788 sodium salt is also problematic [19]. In the present study, we took advantage of the most advanced HPLC–MS/MS and HPLC–MS/TOF methods currently available to detect cNMPs. In this study, the cyclic pyrimidine nucleotides cCMP and cUMP were unambiguously detected in substantial concentrations in representative mammalian cell lines besides cAMP and cGMP. The identification of cCMP and cUMP in intact cells raises the question which enzyme is responsible for this cNMP formation. There is no evidence for a role of pGC-A in this process [20], but highly purified GC generates cCMP and cUMP in a NO-dependent manner [13]. However, purified sGC generates cCMP and cUMP only in the presence of Mn2+, but not in the presence of Mg2+[13]. Because cellular Mg2+ concentrations are much higher than Mn2+ concentrations [21], we were skeptical about a role of sGC in the regulation of cCMP and cUMP levels in intact cells. Nonetheless, we clearly detected NO-stimulated and ODQ-inhibited cCMP and cUMP formation in cells endogenously expressing sGC and cells transiently overexpressing sGC. Hence, sGC regulates not only cellular cGMP- but also cCMP- and cUMP levels. These data also confirm our hypothesis that in intact cells, Mn2+ rather than Mg2+ is the relevant cofactor for sGC [14]. Thus, our study will also stimulate research on the physiological role of Mn2+, a relatively poorly addressed field so far [14], [21]. Based on the present sGC data with intact cells, it is also likely that the bacterial toxins edema factor and CyaA generate cCMP and cUMP in intact cells [12]. Compared to cells overexpressing sGC, cells endogenously expressing sGC revealed relatively large increases in cCMP and cUMP. The absolute basal and NO-stimulated cGMP- and cUMP levels in RFL-6 cells are very similar. These findings argues against the notion that cUMP is a trivial by-product of a leaky sGC. The cell type-specific differences in NO-dependent cCMP and cUMP formation may be due to different hydrolysis and/or export rates of cNMPs. Basal cCMP formation in cells may also be mediated by sGC considering the fact that the sGC inhibitor ODQ effectively inhibits basal cCMP formation by purified sGC [13] and almost abolishes cCMP levels in intact cells. The MS methods used in the study are very well suited for unequivocal cNMP identification (HPLC–MS/TOF) and cNMP quantitation (HPLC–MS/MS) in a cell population. However, MS is not feasible for analysis of spatiotemporal cNMP patterns in cells so far, although advances in metabolomic MS imaging have been described [22]. Currently, for spatiotemporal analyses, fluorescence sensor-based methods are the most feasible approach. For cGMP and cAMP, fluorescence-based detection methods are available [23], and studies with such sensors have revealed substantial spatiotemporal differences in cNMP levels [24]. Thus, relatively small cNMP- and cUMP changes observed over a large cell population as assessed by traditional MS methods do not exclude the possibility that in individual cells or cell compartments at a specific time point, large cCMP- and cUMP increases do occur. Accordingly, the next major challenge in the cCMP- and cUMP field is the development of highly sensitive and specific fluorescence sensors. As a first step, specific cCMP- and cUMP-binding proteins have to be identified. Appropriate biochemical analysis methods are already in place [25]. As an alternative approach, specific cCMP- and cUMP-binding proteins may be engineered. This is not an impossible task considering that even naturally occurring uracil base-specific receptors exist [26]. In this context, it will also be important to identify cCMP and cUMP not only in cultured mammalian cell lines but also in primary mammalian cells and intact organs. We predict that such analyses will demonstrate the presence of cCMP and cUMP in native vertebrate systems. In the nematode Caenorhabditis elegans, cCMP and cUMP were not identified [27], suggesting that these cyclic nucleotides are evolutionary relatively new molecules.