Synthesis of Dimethyl Selenone and Studies on the Biomethylation Pt 3

Chapter IV Discussions and Results Synthesis of dimethyl selenone At the very beginning of this research, a literature search was done to find various oxidants able to oxidize dimethyl selenide to dimethyl selenone. Ozone was an effective oxidant for oxidizing dimethyl selenide to dimethyl selenone (Paetzold and Bochmann, 1968), but since it required an ozone generator, it was too expensive. Thirty percent hydrogen peroxide was also employed as oxi- dant to oxidize dimethyl selenide (Bird and Challenger, 1942). We found that it was unable to produce the corresponding selenone in our work. Then we attempted to approach this synthesis by employing the use of 3-chloroperoxybenzoic acid (m-CPBA) as oxidant with dimethyl selenide as the starting material (Krief et al, 1985). CH 3-Se-CH 3 O O CH 3 -Se-CH 3 CH 3-Se-CH 3 O Conditions: 3.0 mol. equiv. m-CPBA/CH 2Cl 2 % Yield 50 0 20 , 2 hours Scheme IV It was reported that m-CPBA was efficient for the oxidation of dimethyl selenide to dimethyl selenone (Krief et al., 1985). But under the conditions that these workers used, the formation of selenoxide was instantaneous whereas further oxidation to selenone occurred more slowly; therefore, our final product had both the selenone and selenoxide present. To minimize the problem caused by selenoxide, we modified the procedure by increasing the equivalents of m-CPBA and prolonging the reaction time. We achieved a 50% yield of pure dimethyl selenone by this means. For purifying the crude product, we selected ethyl ether to sepa- rate dimethyl selenone from the by-product, m-chlorobenzoic acid. As m-chlorobenzoic acid was freely soluble in ether but dimethyl selenone was not, dimethyl selenone was removed as a precipitate by vacuum filter and collected. Melting point consideration The melting point of dimethyl selenone published in the litera- ture (Paetzold and Bochmann, 1968) ranged from 1470C-1480C, which was higher than that of the crude product that we extracted from ethyl ether. Recrystallization was then performed using metha- nol as the solvent. Since methanol is a relatively polar solvent, it was found to work well for the recrystallization of dimethyl selenone. The melting point of our final product was 1470C-1480C, as re- ported by Paetzold and Bochmann. This was the first straight forward evidence for successful synthesis dimethyl selenone in our work. Elemental analysis The elemental analysis of dimethyl selenone showed some differ- ence between the theoretical data and the determined data. The largest difference was for selenium. The reason for this is possibly due to the degradation of dimethyl selenone to dimethyl selenoxide or other reduced selenium compounds, since this elemental analysis was performed ten days after it was synthesized. Fourier transfer infrared spectrum of dimethyl selenone FTIR instruments have three major advantages. They can give greater signal to noise ratio; they have extremely high wavelength accuracy and precision and can detect all elements of the source reaching the detector simultaneously. FTIR spectroscopy was carried out to further confirm the synthe- sis of dimethyl selenone in our experiment. Both published (Paetzold and Bochmann, 1968) and experimental infrared spectrums of dimethyl selenone are shown in Figure 1. The two spec- trum agree well. The most characteristic absorptions of dimethyl selenone (Figure 2) were two strong bands in the 889 cm-1 and 917 cm-1 regions because of asymmetric and symmetric SeO2 stretching, respectively (Paetzold and Bochmann, 1968). The stretching vibrations of C-S linkage occurred in the region of 700-600 cm-1. Two bending vibrations can occur within a methyl group. The symmetric bending vibration occurred near 1375 cm-1 and the asymmetric bending vibration near 1450 cm-1. Absorption arising from C-H stretching of methyl group generally occurred in a high frequency region which was about 3000-2840 cm-1 (data not shown). Our experimental spectrum did confirm these peaks clearly. It was re- ported that dimethyl selenoxide had a characteristic absorption at 820 cm-1 region (Paetzold and Bochmann, 1968) which was not observed in our experimental spectrum. This evidence suggested that there was little dimethyl selenoxide in our sample soon after recrystallization. Nuclear magnetic resonance spectrum of dimethyl selenone The experimental nuclear magnetic resonance spectrum for dim- ethyl selenone is shown in Figure 3. In a proton NMR spectrometry, protons in different chemical environments have different chemical shifts. If dimethyl selenone were present, only one proton NMR peak would be observed because the six protons in the two methyl groups are in the same chemical environment. H O H H C Se C H O H H Structure of dimethyl selenone We used deuterium oxide as the solvent because our sample did not dissolve well in deuterated chloroform. The TMS (reference) peak was set up at 0 Hz on the chart before the sample was run. Since the deuterium oxide was not completely deuterated or was contaminated by atmospheric moisture, a proton peak at 4.55 ppm was observed in our experimental spectrum. A water blank was run to identify this peak and we observed a proton peak appearing at the same place (4.55 ppm). In our product plus deuterium oxide sample, another singlet peak was observed at 3.45 ppm which was tentatively identified as dimethyl selenone. No literature proton NMR spectrum of dimethyl selenone was found. NMR spectrums of dimethyl selenide and dimethyl diselenide were run to identify the 3.45 peak. Both of them had nothing to do with this peak. So, given the other characterization data, we feel confident that dimethyl selenone exhibits the 3.45 ppm singlet. One month after dimethyl selenone was synthesized, we per- formed a proton NMR spectrum again (Figure 3). The spectrum showed two peaks appearing at 3.25 ppm and 2.75 ppm respectively and little or no peak at 3.45 ppm. This was probably due to the degradation or conversion of dimethyl selenone to another selenium compound. Methyl methaneseleninate CH3Se(O)OCH3 is a good pos- sibility because two methyl groups in this compound are in different environment. There should be two singlet peaks appearing in its NMR spectrum. Dimethyl selenone as an intermediate of the biomethylation in selenium resistant bacterial cultures Preparation work Initially, we used only autoclaving of solutions and media as a means of sterilization for this work; however, headspace analysis of media blanks showed that some volatile sulfur compounds such as dimethyl disulfide and dimethyl trisulfide were produced after autoclaving TSN media (Figure 4). The reason for this may be due to high temperature and high pressure degrading the medium. The sulfur containing amino acids in the complex TSN medium are pre- sumably the source of the volatile sulfur compounds observed. This problem caused confusion because we could not differentiate which gases released from our samples were from methylation process by the microorganisms and which were from sterilization process. In order to overcome this problem, we sterilized the medium by em- ploying the use of sterilized filters. With this technique, only trace amount of dimethyl diselenide were detected in medium blank in our chromatograms (Figure 4). Chromatographic analysis Table 1 summarizes the results of headspace analysis above so- dium selenate dosed culture, dimethyl selenone dosed culture, solu- tion of dimethyl selenone with DMS and DMDS (reduced sulfur compounds) in medium blank, and solution of sodium selenate or sodium selenite with DMS and DMDS in media blanks. We observed many volatile sulfur and selenium compounds such as meth- anethiol, dimethyl sulfide, dimethyl disulfide, dimethyl selenide, dimethyl diselenide and some unknown compounds in these chromatograms (Figure 5, 6, 7, 8, 9). We identified these volatile compounds by GC-SCD, using known compounds for comparison (Table 2). No selenium containing species were detected above selenium free cultures. Since 1 mL gas of headspace above selenone dosed cultures ex- tremely overloaded the column, we had to take small amounts of gas and used split injection technique for GC analysis. So, we also prepared 2 mL cultures dosed with sodium selenate and dimethyl selenone in order to achieve more headspace, smaller cell populations, and lower concentration of volatile species. When we compared these results with 10 mL samples poisoned with the same concentration, the peak areas of 2 mL samples were multiplied by five to normalize to 10 mL cultures volume. Almost all chromatograms of selenium amended cultures indi- cated that there was a peak eluted between dimethyl disulfide and dimethyl diselenide. Since the column that we used in our experi- ment was a nonpolar column, and the elution order of compounds should generally reflect the order of increasing boiling points, the compound eluted between DMDS and DMDSe in our chromatograms was probably dimethyl selenenyl sulfide based on previous work (Chasteen, 1993); however, GC/MS confirmation using these cul- tures has not been performed. Another unknown peak was noted at 13.2 min, which was probably due to the dimethyl trisulfide. A dimethyl trisulfide standard has recently been synthesized in our laboratory and this 13.2 min peak has the same retention time as the major peak in that standard. The most exciting aspect of this research was that we observed much larger amount of volatile selenium compounds (DMSe and DMDSe) above selenone dosed cultures than those above selenate dosed cultures. (All comparisons are based on integrated peak ar- eas). Gas phase concentrations of DMSe and DMDSe above selenone dosed cultures were respectively about 4000 times and 1000 times more than those above selenate dosed cultures poisoned with the same concentration (Table 1). This suggests that dimethyl selenone could indeed be one of the biometabolic intermediates of the biom- ethylation of selenium in the Se cultures. This lends some credence to the methylation mechanism proposed by Challenger (Challenger, 1945). If it were not true, we might not observe an increase of vola- tile selenium compounds released from the cultures if a stable intermediate such as dimethyl selenoxide, (CH3)2SeO, were created. These results demonstrate the ready and efficient bioreduction of dimethyl selenone; however, the mechanism of reduction may actually be a sulfur pathway which accommodates the structurally related and isoelectronic selenium species. Meanwhile, we also noticed that gas phase concentrations of vola- tile sulfur compounds evolved from the 10 mM selenone dosed cultures were less than those evolved from the selenate dosed cul- tures poisoned with the same 10 mM concentration. Further experi- ments were undertaken to identify the correlation between selenone and volatile sulfur compounds. One mL of headspace gas of three media blanks separately dosed with 10 mM selenone, 10 mM selenate and 10 mM selenite in the presence of 1 ng DMS and 1 ng DMDS were analyzed by GC. The chromatograms showed that some volatile selenium compounds such as dimethyl selenide and dim- ethyl diselenide were formed in the selenone dosed medium in the presence of reduced sulfur compounds (Figure 7); however, no volatile selenium compounds were observed in the selenate or selenite dosed medium with DMS and DMDS (Figure 8, 9). This evidence proved that DMS and DMDS could reduce dimethyl selenone to DMSe and DMDSe by a chemical reaction without microorganisms. But the reduced sulfur compounds could not methylate and reduce selenate or selenite salts to volatile selenium compounds directly. On the basis of the present research, it seems plausible to suggest that the K27 bacteria can methylate nonvolatile sulfur compounds in TSN media to volatile sulfur compounds such as DMS and DMDS and thereby to provide a reducing environment. In the meantime, the K27 bacteria also reduce and methylate selenate or selenite salts to some intermediate, possibly dimethyl selenone, which may then be reduced by both chemical and biological means to volatile re- duced selenium species simultaneously. Whether or not this final step is only a chemical reaction with no biological involvement can not be determined at this point. Conclusions The results of this research show that: 1) 3-Chloroperoxybenzoic acid was efficient at oxidizing dim- ethyl selenide to dimethyl selenone. This was confirmed by FTIR, NMR, elemental analysis and melting point. 2) Pure dimethyl selenone was found to be unstable in the presence of reduced sulfur species in trypticase soy broth medium with 0.1 % nitrate added. 3) TSN medium was better than two different minimal media for the growth of Pseudomonas fluorescens K27 bacteria. 4) Dimethyl selenone is possibly one of the intermediates of the biological reduction and methylation of inorganic selenium. This conclusion is based on the observation of an increase amount of volatile selenium species such as dimethyl selenide and dimethyl diselenide above selenone dosed cultures when compared with analogous selenate dosed cultures. 5) Chemical reduction and bioreduction by microorganisms may be simultaneously involved in the conversion of dimethyl selenone to volatile selenium compounds in Pseudomonas fluorescens K27 cultures.