Selección de los recursos y actividades para el curso de Redes III

DSS-3 is a homotaurine-containing lipid

During LC-MS analysis of lipid extracts from Phaeobacter sp. MED193 and R.

pomeroyi DSS-3, two prominent peaks in the base peak chromatogram eluted

between 4 and 5 minutes. The most prominent ions in the earlier eluting peak had

m/z values of 630.7, 656.7 and 670.7; in the second peak there was a similar pattern in the relative intensities of the masses but these were 16 m/z units higher. To elucidate the structure of these lipids, the most intense species, at 656.7 m/z (AAL1-656) was selected for high resolution MS/MS analysis on a quadrupole – time of flight (QTOF) mass spectrometer (Figure 5.1a). I scanned a range of collision energies in order to obtain maximum structural information. At low collision energy (40 eV) the major species formed corresponded to a neutral loss of 282 mass units (Figure 5.1b). This is consistent with the neutral loss of an 18:1 fatty acid. A second peak at m/z 281 is likely the carboxylate anion of an 18:1 fatty acid. Further fragmentation, at higher collision energies (up to 90 eV), yielded a major ion at m/z 237 (Figure 5.1c). This ion likely corresponds to a 16:0 fatty acid present as a ketene, which would be consistent with the fragmentation scheme proposed for ornithine lipids (Zhang et al., 2009). These results, therefore suggest a lipid class with a similar fatty acyl backbone structure to the aminolipids, such as ornithine and glutamine lipid.

125

Figure 5.1 A previously unidentified class of homotaurine lipids is present in Ruegeria pomeroyi DSS-3. a) Intact masses of R. pomeroyi lipids, measured using a high resolution, accurate mass quadrupole-time of flight mass spectrometer in negative ion mode. The identity of the most abundant ion, highlighted, was unknown, so this ion was fragmented in order to elucidate its structure. b) Fragmentation spectrum at 40 eV collision energy. The most abundant species corresponds to the loss of a 18:1 fatty acid. c) Fragmentation spectrum at 90 eV collision energy. The lower spectrum shows a peak at 237.2159, consistent with the presence of a 16:0 fatty acid. The upper spectrum shows an expanded view of the spectrum in the m/z range below 140. Ions are annotated with their predicted elemental composition. Inset is the proposed structure of the 136 m/z fragment (3-aminopropene sulfonic acid).

126

Prominent peaks at 80 and 81 m/z, respectively, were apparent in the fragmentation spectrum obtained at 90 eV collision energy (Figure 5.1c). The accurate masses for these ions were 79.9568 and 80.9643. Of the candidate formulae within 100 ppm of the measured mass, SO_;< _{and HSO}

;

< _{appear most plausible, with mass errors of 0.182}

ppm and 4.194 ppm, respectively. A smaller peak doublet at m/z 63.9611 and 64.9692 was also present in the 90 eV spectrum. These masses are unambiguously assigned to

SO_F< _{(12.506 ppm) and HSO} F

< _{(8.08 ppm). Taken together, these results demonstrate}

the presence of a sulfonate group in the lipid.

An ion at 136 m/z corresponded to the deprotonated head group. The mass determined here is larger than that of deprotonated taurine (m/z 124). Since the head group includes a sulfonate (SO_;<_{) group, the plausible formula most closely}

corresponding to the accurate mass is C3H6NSO3 (Table 5.1). This is consistent with

the structure being 3-amino propene sulfonate (Figure 5.1).

Table 5.1 Nominal and accurate masses of proposed head group fragments.

Nominal m/z Accurate m/z Formula Mass Error / ppm

136 136.0045 C3H6NSO3 12.784

120 119.9922 C3H4SO3 34.048

107 106.9743 C2H3SO3 51.315

95 94.9783 CH3SO3 16.737

Charge remote fragmentation is a gas-phase process, which can occur at both low and high collision energies, whereby fragmentation takes place without involving charge transfer reactions (Cheng and Gross, 2000). It is favoured when the charge is fixed in one location by a strongly ionised group and can result in very informative fragmentation patterns, for example along the alkyl chain of a fatty acid. Sulfonate is a strongly acidic group and consequently, when analysed in negative ionisation mode, negative charge tends to be fixed to the deprotonated sulfonic acid. This property has been used to characterise the structure of the sulfo-glycolipid SQDG (Kim et al., 1997) and also of taurine- and amino propane sulfonate-conjugated bile acids (Stroobant et

127

al., 1995). Charge-remote fragmentation would appear to explain the series of fragment peaks observed for AAL1-656 at 90eV in the m/z range between the intact head group at 136 and the sulfonate peaks at 80 and 81 (Table 5.1). Of formulae containing at least one sulfur and three oxygen atoms, only one lay within 100 ppm of 119.9922 and 94.9783. Two matches were obtained for m/z 106.9743 (C2H3SO3 and

CHNO3S). Of these, C2H3SO3 had the lower mass error (55.988 ppm compared to

61.574 ppm) and appeared more consistent with the formulae assigned to the adjacent peaks. The formulae assigned to these masses support the structure proposed in Figure 5.1. However, the mass error is somewhat high, possibly due to the low number of ions at high collision energy. To account for these results, I outline a proposed fragmentation scheme in Figure 5.2. The unsaturation in the head group arises during fragmentation, implying that the intact head group is 3-aminopropane sulfonate (homotaurine).

128

Figure 5.2 Proposed fragmentation scheme for the homotaurine lipid with m/z 656. Mass errors indicate the difference between the measured and theoretical masses of the proposed species. R1 O N H S O O O – O R2 O - C₁₇H₃₃COOH R1 O N H S O O O – O R1 H H₂N S O O O – C₁₇H₃₃ C₁₃H₂₇ C₁₃H₂₇ C₁₃H₂₇ C₃₇H₇₀O₆NS m/z = 656.4885 error = 5.917 ppm C₁₉H₃₆O₄NS m/z = 374.2364 error = 0.279 ppm C₃H₆O₃NS m/z = 136.0045 error = 17.196 ppm

129