Geochemical Journal, Vol. 46, pp. 235 to 247, 2012 Changes in alkenone and alkenoate distributions during acclimatization to salinity change in Isochrysis galbana: Implication for alkenone-based paleosalinity and paleothermometry MAKIKO ONO,1 KEN SAWADA,1,3* YOSHIHIRO SHIRAIWA2,3 and MASAKO KUBOTA2 1Department of Natural History Sciences, Faculty of Science, Hokkaido University, N10W8, Kita-ku, Sapporo 060-0810, Japan 2Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba 305-8572, Japan 3CREST, Japan Science and Technology Agency (JST), Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan (Received February 8, 2012; Accepted April 30, 2012) A cultured strain of Isochrysis galbana UTEX LB 2307 was grown at 20°C and 15°C under salinity of 35‰, 32‰, 27‰, 20‰, and 15‰, and analyzed for long chain (C37–C39) alkenones and (C37–C38) alkyl alkenoates. It was character- ized by abundant C37 alkenones and C38 ethyl alkenoates (fatty acid ethyl ester: FAEEs) and a lack of C38 methyl alkenones. There were no tetra-unsaturated (C37:4) alkenones, which are frequently found in natural samples from low salinity waters. K′ ° The alkenone unsaturation index (U 37) did not vary in response to change in salinity at 15 C. The alkenone unsaturation K′ ° ° index (U 37) clearly changed in response to change in salinity at 20 C, but not at 15 C, where algal growth was and was ° ° K′ not limited by temperature at 15 C and 20 C, respectively. The U 37-temperature calibration for I. galbana UTEX LB 2307 is quite different from those of E. huxleyi and another strain of I. galbana (CCMP1323), while is resemble to that reported for C. lamellosa isolated from a Chinese lake. Also, variation in the alkenone chain length ratio values (K37/K38) did not correlate with salinity. These results implied that a high abundance of tetra-unsaturated alkenone and high K37/K38 values might be attributed to a taxonomic factor rather than a physiological response to salinity change. Interestingly, our culture experiments showed that the ethyl alkenoates/alkenones ratio (EE/K37) correlates with salinity. Hence, it is sug- gested that the EE/K37 ratios are affected by the cellular and physiological factors against salinity condition in single haptophyte cells. Keywords: alkenone, Isochrysis, acclimatization, paleothermometry, saleosalinity alkenones are derived from Haptophycean algae, includ- INTRODUCTION ing the Noelaerhabdaceae and Isocrysidaceae families. In K Unsaturation indices for C37 alkenones (U 37 and particular, cosmopolitan haptophycean algae such as K′ U 37) are established as proxies for paleotemperature of Emiliania huxleyi and Gephyrocapsa oceanica in the open sea surface water (Brassell et al., 1986; Prahl and ocean are thought to be main sources of alkenones and Wakeham, 1987; Prahl et al., 1988) and have been widely related compounds. Also, coastal species such as used to reconstruct paleoceanograhic variation from ma- Isochrysis galbana and Chrysotila lamellosa are known rine sediments (e.g., Rostek et al., 1993; Bard et al., 1997; to synthesize alkenones, but the distribution and abun- Sawada and Handa, 1998; Sachs and Lehman, 1999; dance of I. galbana in the marine environment are virtu- Kienast et al., 2001; Seki et al., 2004; Haug et al., 2005) ally unknown (Versteegh et al., 2001). A strain of C. and to infer continental paleoclimate from lacustrine lamellosa was isolated from a continental lake (Lake sediments (e.g., Zink et al., 2001; Coolen et al., 2004; Xiarinur, China) and then its cultured strain was confirmed Chu et al., 2005; Pearson et al., 2008; Toney et al., 2010; to produce alkenones and alkenoates (Sun et al., 2007). D’Andrea et al., 2011). Moreover, the relative abundance Using 18S ribosomal RNA (rRNA) analysis, Coolen et of tetra-unsaturated alkenones (C37:4 alkenones) could be al. (2004) reported firstly that the alkenones in sediments potentially used as a proxy for paleosalinity (Rosell-Melé, from Ace Lake in Antarctica originated mainly from or- 1998; Schulz et al., 2000; Harada et al., 2003; Blanz et ganisms related to I. galbana. A recent 18S rRNA inves- al., 2005; Liu et al., 2006). Long-chain (C37–C39) tigation of 15 alkenone-containing sediments from lakes in Eurasia and North American continents (Theroux et al., 2010) demonstrated that multiple haptophyte species *Corresponding author (e-mail: [email protected]) of order Isochrysidales are possibly present as alkenone Copyright © 2012 by The Geochemical Society of Japan. producers in continental lakes worldwide. More recently, 235 Toney et al. (2012) reported that the enrichments of two tion of alkenone producers against change in salinity and/ haptophyte phylotypes (Hap-A and Hap-B) from George or specific hydrological conditions (Schulz et al., 2000). Lake, United States, could be individually cultured, and In addition, C37:4% variation is associated with low tem- major alkenone producer in the lake system was likely to perature (Sikes et al., 1991; Sikes and Volkman, 1993; be the Hap-A phylotype, which was related to I. galbana, Rosell-Melé et al., 2002) and is also speculated to be af- rather than the Hap-B, which was related to C. lamellosa fected by limited nutrients (Harada et al., 2003). The ra- and Pseudoisochrysis paradoxa. These results suggest that tio of C37 to C38 alkenones (K37/K38 or C37/C38) also vary the Chrysotila, Isochrysis and other Isochrysidales can in a marginal sea and a continental lake (Schulz et al., be the main producers of alkenones and related com- 2000; Chu et al., 2005). Shultz et al. (2000) reported that pounds in continental lake systems. K37/K38 (C37/C38) values varied depending on the salin- K′ U 37-temperature calibrations have been studied by ity of surface water in the Baltic Sea, decreasing culturing alkenone-synthesizing algae (Marlowe et al., exponentially below 9 PSU and constant from 9 to 30 1984; Prahl and Wakeham, 1987; Conte et al., 1994, 1995, PSU. They inferred that such differences might be attrib- 1998; Volkman et al., 1995; Sawada et al., 1996; utable to variations in species and/or strain of alkenone Versteegh et al., 2001; Rontani et al., 2004; Sun et al., producers. Ono et al. (2009) showed the variation in 2007) as well as analysis of water column particulates alkenone and alkenoate distributions in a culture of the (Conte and Eglinton, 1993; Sikes and Volkman, 1993; haptophycean algae Emiliania huxleyi and Gephyrocapsa Sawada et al., 1998; Hamanaka et al., 2000; Bentaleb et oceanica under various salinity conditions, and found no al., 2002) and core top sediments (Sikes et al., 1991; C37:4 alkenone and that K37/K38 values increased with Rosell-Melé et al., 1995; Pelejero and Grimalt, 1997; decreasing salinity (Ono et al., 2009). Sonzogni et al., 1997; Müller et al., 1998). It is known In this study, we have examined variation in the K′ that there are large variations in U 37-temperature cali- unsaturation ratios and distributions of alkenones and brations from both culture and field samples. These vari- alkenoates for a culture of Isochrysis galbana UTEX LB ations have been attributed to ecological bias in depth 2307, a different strain from that examined by Versteegh and seasonality of alkenone production, and diagenetic et al. (2001), as well as the reliability of these parameters alteration during and after deposition (Grimalt et al., 2000 as paleotemperature and paleosalinity proxy. and references therein). Also, many researchers have pointed out that UK′ -temperature relationships vary as 37 MATERIALS AND METHODS a result of physiological factors such as growth phase and genetic factors such as species and genotype (e.g., Conte Culture conditions et al., 1998; Yamamoto et al., 2000). Isochrysis aff. galbana UTEX LB 2307 was obtained Recent studies of marine environments have indicated from the Algal Collection of the University of Texas at that the relative abundance of the C37:4 alkenone (C37:4/ Austin (Texas, USA; Starr and Zeikus, 1987), and had K37 or C37:4%) as a proportion of total C37 (di-, tri- and been isolated from an aquaculture pond in Society Islands, tetra-unsaturated components) is inversely correlated with Tahiti (Wikfors and Patterson, 1994). It was cultured con- salinity, especially at salinity <ca. 30 PSU (Schulz et al., tinuously in batch systems at 21°C and 34‰ salinity un- 2000; Harada et al., 2003; Blanz et al., 2005). Further- der cool white fluorescent light and continuous bubbling more, C37:4% values have been found to be applicable to of sterilized air. It was transferred to 500 ml Mericron reconstruction of relative changes in salinity for Holocene flasks (Iwaki Co. Ltd., Tokyo, Japan) containing 300 ml sediments from a continental lake (Lake Qinghai; Liu et artificial seawater (Marine Art SF-1 (MA); produced by al., 2006). However, the cause(s) of the variability of Tomita Pharmaceutical Co. Ltd. (Naruto, Tokushima, Ja- %C37:4 values in ocean and continental lake systems is pan) and obtained from Osaka Yakken Co. Ltd. (Osaka, (are) not clear. Most researchers suggest that the C37:4% Japan; formerly Senju Pharmaceutical Co. Ltd.)) enriched values could increase by way of a contribution of with modified ESM in which soil extract was replaced haptophycean species or genetic strains that contained with 10 nM sodium selenite (Danbara and Shiraiwa, 1999) ° C37:4 alkenone, and that salinity might be an indirect fac- at 20 C under 32‰, 27‰, 20‰ and 15‰ salinity, as well ° tor affecting the relative abundance of the C37:4 alkenone as at 15 C under 35‰, 32‰, 27‰, 20‰ and 15‰ salin- (e.g., Blanz et al., 2005; Chu et al., 2005). In addition, ity. They were harvested at day 2 (48 h), day 4 (96 h) and Toney et al. (2012) reported that the C37:4 alkenone- day 8 (192 h).