Analysis of ribo- and deoxyribonucleic acids using ionpair-reversed-phase liquid-chromatography electrospray-mass spectrometry

The fast progress in natural sciences like biology, biochemistry, medicine or genetics make high demands on the analytical chemistry. The on-line coupling of ionpair-reversed phase-liquid chromatography (IP-RP-HPLC) to mass spectrometry (MS) becomes more and more the method of choice for the analysis of biomolecules. The success is based on the introduction of soft ionization methods, like electrospray ionization (ESI), which allows the transfer of intact biopolymers into the gasphase. This combination enables the on-line separation of complex biological mixtures with additional identification of the compounds by their molecular mass. The first part describes the development of a new ionsource, which combines the advantages of a micro-ESI- and a nanospray-source. In combination with additional optimization of the chromatographic conditions the new ionsource showed an improvement of the quality of the spectra by a factor of 5 and a stability of the ionspray by a factor of 2, which resulted in an overall improvement of sensitivity by a factor of 10 for the HPLC-MS system. The second part describes the quality control of synthetic RNA molecules. Using IP-RP-HPLC-ESI-MS it was possible to separate failure sequences and derivatives in raw products of short synthetic RNAs. The derivatives were formed of protecting groups, which were not removed during the deprotection step. The analysis of coupling products of the synthesis of aminoacylated transfer RNAs showed a derivative, which was formed by the addition of the used coupling reagent N-(3-dimethylaminopropyl)N'-ethylcarbodiimide (EDC). The identification of the derivatives led to the optimization of the reaction conditions which resulted in the synthesis of the wanted transfer RNA without any additional derivatives. Another experiment involved the fragmentation of RNA molecules. Tandem mass spectrometry provides the opportunity to determine the sequence of nucleic acids. Fragmentation experiments showed different fragments compared to DNA. The additional hydroxyl-group on 2'-position of the ribose caused the formation of (cn)n- and (yn)n- fragments instead of the typical (an-Bn)n- and (wn)n- fragments observed for DNA. An algorithm was adapted for the computer-aided interpretation of fragment ion spectra of multiply charged oligoribonucleotide ions generated by ESI. The method compares the experimental spectrum to the m/z values predicted by employing established fragmentation pathways. The closeness of matching is characterized by the fitness, which takes into account the difference between the measured and predicted m/z values, the intensity, the number of fragments assigned, and the number of nucleotide positions not covered by fragment ions in the experimental spectrum. This algorithm was successfully applied to verify RNA sequences in a 13-meroligoribonucleotide, although the difference of 1 Da between the nucleobases cytosine and uracil complicated their differentiation in the fragment-spectrum. Using this algorithm it was also possible to localize modifications in form of mono- and bismethylations in a RNA 13-mer. In the last part IP-RP-HPLC-ESI-MS was used to analyze single nucleotide polymorphisms (SNPs) and haplotypes in diploid and tetraploid genomes. The aim of this investigation was to determine the allelic ratio of SNPs in a tetraploid and mitochondrial genome. In contrast with diploid genomes, the ratio of SNP alleles can vary in tetraploid genomes. In tetraploids five allelic ratios, 0:4, 1:3, 2:2, 3:1 and 4:0 were distinguishable whereas in the mitochondrial genome the allelic ratio varied between 0 % and 100 %. The determination of the ratios was done upon comparison of the intensities of the mass signals for the single alleles derived from the deconvoluted mass spectra. Using IP-RP-HPLC-ESI-MS it was possible to unequivocal identify the allelic ratios in tetraploid samples. For mitochondrial samples the limit for the identification was a ratio of 5:95 respectively 95:5. Samples with less than 5 % of an allele were not distinguishable from noise. (author)