Mass spectrometric analysis of short-lived intermediates and ancient fossils

Abstract

The first part of this thesis focuses on mechanistic studies of two L-proline catalyzed reactions with electrospray ionization mass spectrometry. The second part describes mass spectrometric analyses of fossils, dinosaur egg shells and silicified wood.<br /> For the first L-proline catalyzed reaction of α,β-unsaturated aldehydes, a catalytic cycle with two intermediates was postulated. I studied the reaction by taking samples directly out of the reaction solution and feeding them into the mass spectrometer. Apart from the two postulated intermediates I found two additional species which fit into the mechanistic scenario. With the aid of MSn experiments, it was possible to investigate the intermediates and additional species in the gas phase and confirm their structure. The MS2 experiment of the second intermediate was riveting, as it was possible to induce it to go backwards and forwards in the catalytic cycle, thereby mimicking its behavior in the gas phase. With the aid of a charge-tagged L-proline derived catalyst, I could study the intermediates in their unprotonated form and observed the same behavior for the second intermediate. By taking samples out of the reaction solution at regular intervals, insight into the temporal progress of the reaction was gained. The rate determining step of the reaction is the liberation of the catalyst and product from the second intermediate. The reaction was studied with trans-2-hexenal and trans-2-pentenal; their results are in accordance with each other.<br /> The second L-proline catalyzed reaction of acetone with a tetrazine has a postulated catalytic cycle with three intermediates. In experiments with L-proline without a charge-tag, the first and the third intermediate could be detected. A charge-tagged tetrazine was synthesized, however only small amounts of the third intermediate could be detected with it. When the charge-tagged L-proline derived catalyst was used, I was able to detect the so far elusive second intermediate in addition to the first and third intermediate. In an MS2 experiment, the second intermediate could be induced to go backwards and forwards in the catalytic cycle, thereby mimicking its behavior in the gas phase. Through this first experimental evidence of the second intermediate, I could confirm that the reaction proceeds in a stepwise manner via the second intermediate and not in a concerted step from the first to the third intermediate.<br /> In the third project I studied dinosaur egg shells with HPLC-ESI MS. In preliminary work to the project, the color pigments biliverdin (BV) and protoporphyrin (PP) were detected in three different egg shells from Oviraptorosauria. The goal was to reproduce the results from the preliminary work, optimize the method and expand the methodology to a wider range of samples. The coloring of egg shells is of interest as it is an indicator of different ecological factors like nesting behavior. I tried to reproduce the preliminary work by extracting the same oviraptorid egg shells with the published EDTA extraction method, but was not able to detect either BV or PP. With a more effective extraction method based on hydrochloric acid (HCl) I was still not able to detect BV or PP in the oviraptor egg shells. In measurements at more sensitive instruments, PP was detected with the EDTA extraction method, but in lower concentrations than published in the preliminary work and not in all oviraptor egg shells. When testing the different extraction methods in spiking experiments the EDTA method proved to be unsuitable for the extraction of BV, while PP was acceptable. The HCl extraction method showed a better performance, however for low concentrations of PP, impossibly high extraction ratios were found. Due to this, solvent tests were performed as to see if the solvent effects the detectability of BV and PP. A solvent mimicking the sample solution after HCl extraction showed a highly improved detectability of PP over other solvents, whereas BV was undetectable in it. This does not fit to the results of the spiking experiments, where BV was well detected with the HCl extraction method. To investigate this discrepancy in the next steps of the project the effects of the solvents should be tested again. Further, the oviraptorid egg shells should be extracted with the HCl extraction method and measured at the highly sensitive Qtrap instrument, which should be promising.<br /> In the last project I analyzed with MALDI MS a specimen of 150 million years old silicified wood, which has differently colored domains. The goal for the analysis was to find an explanation for the coloration and if organic compounds can still be detected. The sample preparation needed to be adapted to a fossilized specimen. Thus, a fine powder of the specimen’s different domains was generated by drilling into it with a diamond drill. The sample powder was then suspended in a matrix solution and prepared on the MALDI target. Through this preparation, reasonable MALDI spectra could be obtained. In the dark domain I detected amorphous carbon, which is in agreement with findings of Raman measurements of the sample. As amorphous carbon is black this brings an explanation for the darker coloring of the dark domain. Fascinatingly I was further able to detect organic compounds, like lignin, cellobiose and coniferin, which are basic building blocks of wood.<br /> This work has shown that mass spectrometry is a useful tool for different scientific problems like studying reaction mechanisms to find elusive intermediates or detecting compounds in millions of year-old fossils.