46937-AC2
A Paleosalinity Proxy from Lipid D/H Ratios
Over the past year we made significant progress understanding the influences on hydrogen isotope fractionation in algal lipids. Our study sites in Palau, Christmas Island and the Chesapeake Bay have proven to be well-suited to addressing the central question of how water isotope composition and salinity are preserved in lipid D/H ratios, offering large environmental gradients with few extraneous influences.
We published papers in Spring 2008 in Geochimica et Cosmochimica Acta with post-doctoral investigator Dirk Sachse on Christmas Island (Sachse & Sachs, 2008), Autumn 2007 with post-doctoral investigator Rienk Smittenberg on the purification of dinosterol by HPLC-MS in Journal of Chromatography A (Smittenberg & Sachs, 2007), and in press with post-doctoral investigator Valerie Schwab on alkenone purification from Chesapeake Bay in Organic Geochemistry (Schwab & Sachs, in press). Major findings are below.
Christmas Island
• Data from Christmas Island demonstrates that differences in αlipid–water for lipids derived from three biosynthetic pathways remained almost constant over the entire salinity range (13.6-149.2 psu) as indicated by the similar slopes of the regressions (0.00080 to 0.00107) (Figure 1). We thus postulated that biological control over the αlipid–water–salinity relationship occurred before the last common branching point, pyruvate.
• The observation that salinity is correlated with D/H fractionation in all lipids from Christmas Island, but each with a slightly different slope and intercept led us to hypothesize that changes in the isotopic composition of intracellular water were primarily responsible. The two hypotheses proposed by Sachse and Sachs (2008) invoke the role of water transport in/out of the cell as a primary control:
(1) There is reduced fractionation against HDO during water transport into the cell at higher salinities. But even if such a process occurs it is unlikely that HDO or D+ are preferentially transported into the cell relative to H2O or H+.
(2) Under high salinities less water is exchanged through the membrane and the metabolic water inside the cell is continuously ‘‘recycled’’. Over time this causes the intracellular water to become enriched in deuterium as biosynthesis preferentially consumes protium.
Palau
• Our work in Palau required a method for purifying dinosterol by HPLC–MS. 4-Methylsterols, including the dinoflagellate-specific marker dinosterol (4,23,24-trimethylcholestan-22-en-3β-ol), were successfully separated from co-eluting higher-plant-derived pentacyclic triterpenoid alcohols in sufficient purity and quantity for isotopic analysis (Smittenberg & Sachs, 2007) (Figure 2).
• These purification methods allowed the investigation of D/H fractionation during HPLC purification. D/H ratios varied widely across a chromatographic peak, demonstrating the need to collect 95% of the peak (Figure 3).
Chesapeake Bay
• In the Chesapeake Bay we sought to measure the isotopic composition of individual alkenones, prymnesiophyte biomarkers employed in paleoceanographic temperature reconstructions using the Uk37’ index (e.g., Marlowe et al., 1984; Brassell et al., 1986; Prahl and Wakeham, 1987; Prahl et al., 1988; Müller et al., 1998). Adequate purification required substantial method development by a combination of column chromatography and HPLC-MS due to co-eluting compounds (Figure 4) press (Schwab & Sachs, in press).
• Individual alkenones from Chesapeake Bay water and sediments indicated a large variation in D/H composition among different alkenones (Table 1), highlighting potential problems associated with using the total alkenone D/H for environmental reconstructions.
• The D/H composition of individual alkenones elucidates isotopic fractionation during alkenone synthesis. Similar D/H values of individual alkenones suggest a common straight-chain precursor. Constant fractionation between alkenones with different chain lengths but the same degree of unsaturation (αC37:2-C38:2 and αC37:3-C38:3 = 1.01), and those with the same chain length but different degrees of unsaturation (αC37:2-C37:3 and αC38:2-C38:3 =0.97) in all samples suggest these values represent D/H fractionation associated with elongation and desaturation.
• Substantial D/H fractionation occurred during HPLC separation of alkenones (Schwab & Sachs, in press), comparable to that observed for dinosterol, such that 92% of the peak must be collected for accurate D/H ratios (Figure 5).
