Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`Interannual and between species comparison of the lipids, fatty acids
`and sterols of Antarctic krill from the US AMLR Elephant Island
`survey area
`
`a,b
`c
`c,d
`a,
`Charles F. Phleger *, Matthew M. Nelson , Ben D. Mooney , Peter D. Nichols
`
`a D
`
`Department of Biology, San Diego State University, San Diego, CA 92182, USA
`epartment of Zoology, University of Tasmania, Hobart, Tasmania 7001, Australia
`c
`CSIRO Marine Research, Hobart, Tasmania 7001, Australia
`d
`Antarctic CRC, Hobart, Tasmania 7001, Australia
`
`b
`
`Received 15 July 2001; received in revised form 26 November 2001; accepted 15 December 2001
`
`Abstract
`
`Antarctic euphausiids, Euphausia superba, E. tricantha, E. frigida and Thysanoessa macrura were collected near
`Elephant N Island c during 1997 and 1998. Total lipid was highest in E. superba small juveniles (16 mg g wet mass),
`y1
`ranging from 12 to 15 mg in other euphausiids. Polar lipid (56–81% of total lipid) and triacylglycerol (12–38%) were
`the major lipids with wax esters (6%) only present in E. tricantha. Cholesterol was the major sterol (80–100% of total
`sterols) with desmosterol second in abundance (1–18%). 1997 T. macrura and E. superba contained a more diverse
`sterol profile, including 24-nordehydrocholesterol (0.1–1.7%), trans-dehydrocholesterol (1.1–1.5%), brassicasterol (0.5–
`1.7%), 24-methylenecholesterol (0.1–0.4%) and two stanols (0.1–0.2%). Monounsaturated fatty acids included primarily
`18:1(ny9)c (7–21%), 18:1(ny7)c (3–13%) and 16:1(ny7)c (2–7%). The main saturated fatty acids in krill were
`16:0 (18–29%), 14:0 (2–15%) and 18:0 (1–13%). Highest eicosapentaenoic acid wEPA, 20:5(ny3)x and docosahexaenoic
`acid wDHA, 22:6(ny3)x occurred in E. superba (EPA, 15–21%; DHA, 9–14%), and were less abundant in other krill.
`E. superba is a good source of EPA and DHA for consideration of direct or indirect use as a food item for human
`consumption. Lower levels of 18:4(ny3) in E. tricantha, E. frigida and T. macrura (0.4–0.7% of total fatty acids) are
`more consistent with a carnivorous or omnivorous diet as compared with herbivorous E. superba (3.7–9.4%). The
`polyunsaturated fatty acid (PUFA) 18:5(ny3) and the very-long chain (VLC-PUFA), C and C PUFA, were not
`26
`28
`present in 1997 samples, but were detected at low levels in most 1998 euphausiids. Interannual differences in these
`biomarkers suggest greater importance of dinoflagellates or some other phytoplankton group in the Elephant Island area
`during 1998. The data have enabled between year comparisons of trophodynamic interactions of krill collected in the
`Elephant Island region, and will be of use to groups using signature lipid methodology. 䊚 2002 Elsevier Science Inc.
`All rights reserved.
`
`Keywords: Antarctica; Euphausia superba; Fatty acid; Krill; Sterol; Triacylglycerol
`
`1. Introduction
`
`Antarctic krill (Euphausia superba Dana) pro-
`vide 30–90% of the diet for marine carnivores in
`
`*Corresponding author. Tel.: q1-760-632-9447; fax: q1-
`619-594-5676.
`E-mail address:
`phleger@sunstroke.sdsu.edu (C.F. Phleger).
`
`the Southern Ocean and have an estimated standing
`stock biomass of about 500 million metric tons
`(Ross and Quetin, 1988). The global fishery for
`krill peaked prior to 1990 at about 500 thousand
`tons per year. The current fishery is about 100
`thousand tons per year. This present low level is
`due primarily to lack of demand (Nicol and Endo,
`1999). The success of Antarctic krill reflects their
`
`1096-4959/02/$ - see front matter 䊚 2002 Elsevier Science Inc. All rights reserved.
`PII: S 1 0 9 6 - 4 9 5 9 Ž 0 2 . 0 0 0 2 1 - 0
`
`RIMFROST EXHIBIT 1173 Page 0001
`
`

`

`734
`
`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`ability to adapt to major differences in seasonal
`food supply. Krill are mainly herbivorous and feed
`on phytoplankton in the summer. In winter, krill
`feed on ice algae and probably bacteria and marine
`detritus, as well as depleting body protein for
`energy (Virtue et al., 1996). High recruitment and
`early spawning occur during years where there is
`a high pack ice concentration of long duration
`(Siegel and Loeb, 1995).
`The Southern Ocean is a complex ecosystem
`including planktonic herbivores (krill, salps, cope-
`pods) fed upon by squid, seals, baleen whales,
`birds and fish (Quetin and Ross 1991; Loeb et al.,
`1997). Lipid biomarkers,
`including lipid class,
`sterol and fatty acid spectra, have been used
`increasingly to help understand marine trophodyn-
`amics (e.g. Nichols et al., 1984; Sargent et al.,
`1987). Recent studies using the lipid signature
`approach have helped to clarify aspects of Antarc-
`tic ecology not visible by conventional techniques
`(Falk-Petersen et al., 1999; Phleger et al., 1999,
`2000; Nelson et al., 2000, 2001).
`There is a growing literature on Antarctic krill
`(E. superba) lipids (Saether et al., 1985; Quetin
`and Ross, 1991; Pond et al., 1995; Virtue, 1995;
`Hagen et al., 1996; Virtue et al., 1996; Mayzaud
`et al., 1998; Cripps et al., 1999). Fatty acids have
`been used extensively as bioindicators in E. super-
`ba (e.g. Virtue et al., 1993b), with sterols used to
`a lesser extent. More limited detail is available for
`E. tricantha and E. frigida (Phleger et al., 1998)
`and Thysanoessa macrura (Reinhardt and Van
`Vleet, 1986; Kattner et al., 1996; Falk-Petersen et
`al., 1999). However, few studies have examined
`interannual changes in krill lipid composition for
`animals collected from specific locations.
`The purpose of this study is therefore to examine
`comparatively lipid classes, specific sterols and
`fatty acid biomarkers of Antarctic krill E. superba
`and other less-studied euphausiids, including E.
`tricantha, E. frigida and T. macrura. The availa-
`bility of krill collected in both 1997 and 1998 in
`the oceanographic region near Elephant Island was
`possible as the area has been intensively surveyed
`for zooplankton by the United States Antarctic
`Marine Living Resources (US AMLR) Program.
`It is noted for high biological productivity and rich
`krill populations that experienced major fluctua-
`tions in density depending on sea ice extent and
`temperature (Loeb et al., 1997; Brierly et al.,
`1999). According to Loeb et al. (1998), 1993 and
`1998 were ‘salp years’ with Salpa thompsoni
`
`numerically dominant (56–89% of total zooplank-
`ton), post-larval Thysanoessa macrura second in
`abundance (8–14%) followed by post-larval krill
`and copepods. The years 1995 and 1996 were
`‘copepod years’ with copepods (presumably Metri-
`dia gerlachei) dominant taxa. Larval T. macrura
`ranked fourth and second in abundance during
`February–March 1995 and 1996, respectively. Post
`larval T. macrura and chaetognaths switched in
`order of abundance during these years. In contrast,
`S. thompsoni ranked sixth and eighth and contrib-
`uted -1.5% of
`total zooplankton. February–
`March 1994 and March 1997 appear
`to be
`‘transition periods (years)’ between ‘copepod
`years’ and ‘salp years’. During ‘transition years’,
`copepods were numerically dominant, followed in
`order by S. thompsoni, post-larval T. macrura and
`Euphausia frigida (Loeb et al., 1998). ‘Salp years’
`appear to correlate with the 4–5-year period of the
`Antarctic Circumpolar wave, which propagates
`changes in sea surface temperatures and wind
`stress direction (White and Peterson, 1996). An
`objective of our study is to utilize the lipid com-
`position data to help clarify trophodynamics,
`including examining possible interannual changes
`in feeding.
`
`2. Materials and methods
`
`2.1. Sample description
`
`Krill were obtained as part of the AMLR Field
`Study conducted annually in the Elephant Island
`region of the Antarctic Peninsula located between
`60–62.58S and 53–598W. (Loeb et al., 1997;
`Martin, 1997y8). Specimens were collected by
`Isaacs–Kidd midwater trawl fitted with a 505 mm
`mesh plankton net from the RyV Yuzhmorgeolo-
`giya during January and February, 1997 and 1998.
`The net was obliquely towed to 170 m depth for
`approximately 30 min at a speed of 2 knots, or to
`10 m above the bottom in shallower waters.
`Samples were frozen in liquid nitrogen immediate-
`ly after sorting on board ship. They were then
`transported frozen (dry ice) by air to CSIRO
`Marine Research, in Hobart, Tasmania, where they
`were maintained at y70 8C prior to analysis. 1997
`krill were 1.00–1.34 g fresh mass for E. superba
`(five and six pooled individuals for each sample,
`respectively), 0.01 g for E. tricantha (one only)
`and 0.20–0.70 g for T. macrura (20 and 70 pooled
`for each sample, respectively). Krill for 1998 were
`
`RIMFROST EXHIBIT 1173 Page 0002
`
`

`

`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`735
`
`1.29–1.39 g fresh mass for E. superba adults (two
`pooled for each sample), 0.28–0.38 g for E.
`superba juveniles (large), 0.22–0.31 g for E.
`superba juveniles (small) (two pooled for each
`sample), 0.70–0.86 g for E. tricantha (three to
`four pooled for each sample) and 0.24–0.28 g for
`E. frigida (six to eight pooled for each sample).
`
`2.2. Lipid extraction and analysis
`
`Oil (lipid) analyses were conducted using meth-
`ods described in Nichols et al. (1998a,b,c) and
`Phleger et al. (1998, 1999). Briefly, samples were
`extracted using a single phase Bligh and Dyer
`(1959) procedure. Oil yield was determined grav-
`imetrically. An aliquot of the oil was analyzed by
`TLC-FID to determine lipid class composition
`(Volkman and Nichols, 1991). Fatty acid and sterol
`profiles were obtained by capillary GC and GC-
`MS analysis
`following trans-methylation and
`saponification of aliquots of the oil.
`
`2.3. Statistical analyses
`
`y1
`Fatty acid profiles (mg g wet mass) of indi-
`vidual samples were compared by cluster analysis
`using Pearson’s correlation coefficient and average
`linkage (Fig. 1). Pearson’s correlation coefficient
`and non-metric multidimensional scaling (MDS)
`were also used to compare FA profiles (full pro-
`files) in two dimensions, using the Kruskal Loss
`Function. All multivariate analyses were conducted
`using SYSTAT 9 (SYSTAT, Inc., Evanston, IL,
`USA).
`
`3. Results
`
`3.1. Lipid classes
`
`Polar lipids were the major lipid class in all
`euphausiids in 1998 (56–81% of total lipid; Table
`1). Triacylglycerols (TAG) were the second most
`abundant lipid class (22–38% in Euphausia super-
`ba, 16% in E. tricantha and 12% in E. frigida).
`E. tricantha was the only euphausiid with wax
`esters (WE) (6% of total lipid). WE were below
`detection (-0.5%) in the other species. In all
`animals, sterols (ST) accounted for 4–7% of total
`lipids, with low free fatty acids (FFA) (1–3%).
`Total lipid was highest in E. superba small juve-
`y1
`niles (15.9 mg g
`wet mass) and ranged from
`
`y1
`
`in all other euphausiids (Table
`
`12.4–14.6 mg g
`1).
`
`3.2. Sterols
`
`Cholesterol was the major ST in all krill (80–
`100% of total ST) with highest values in E.
`tricantha (94–100%) and E. frigida (97%) (Table
`2). Desmosterol was the second most abundant ST
`(0–18% of total ST), and was highest in 1997 E.
`superba (18%), but markedly less in 1998 E.
`superba (2–4%). Although there was 6% desmos-
`terol in 1997 E. tricantha, none was detected in
`1998 E. tricantha (Table 2). Desmosterol was also
`not detected in E. frigida. The 1997 Thysanoessa
`macrura and E. superba samples were the only
`samples with ST other than cholesterol and des-
`mosterol. These included primarily 24-nordehydro-
`cholesterol (0.1–1.7%),
`trans-dehydrocholesterol
`(1.1–1.5%), brassicasterol (0.5–1.7%) and 24-
`methylenecholesterol (0.1–0.4%). Low levels of
`stanols (0.1–0.2% cholestanol and brassicastanol)
`were only detected in 1997 T. macrura and E.
`superba, but not in the 1998 krill (Table 2).
`
`3.3. Fatty acids
`
`Polyunsaturated fatty acids (PUFA) were 39–
`45% of the total FA in all E. superba samples, but
`were somewhat lower in 1998 E. tricantha (30%)
`and 1998 E. frigida (31%), with 11% total PUFA
`frigida (Table 3). Eicosapentaenoic
`in 1997 E.
`acid wEPA, 20:5(ny3)x and docosahexaenoic acid
`wDHA, 22:6(ny3)x were the two major PUFA in
`all samples (Table 3). Highest EPA and DHA
`values were detected in E. superba (15–21% and
`9–14%, respectively). These two PUFA were gen-
`erally lower in abundance in E. frigida (4–18%
`EPA; 5–9% DHA) than in E. tricantha (12–18%
`EPA; 14–16% DHA). Ratios of EPAyDHA for
`1997 and 1998 E. superba were similar (1.4–1.6;
`Table 3, Fig. 2) whereas this ratio was somewhat
`lower for 1997 E. tricantha (1.2 vs. 0.9 for 1998).
`In contrast, the EPAyDHA ratio for E. frigida was
`lowest (0.7) in 1997 and highest (2.0) in 1998 of
`all samples analyzed.
`Arachidonic acid wAA, 20:4(ny6)x was 1% of
`total FA in 1997 and 1998 E. tricantha, and only
`0.4–0.6% in 1997 and 1998 E. frigida. In contrast,
`AA was not detected in E. superba and T. macrura.
`Although levels of the PUFA 18:4(ny3) were 4–
`9% in E. superba from both years, it was only
`
`RIMFROST EXHIBIT 1173 Page 0003
`
`

`

`736
`
`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`Table 1
`Percentage lipid class composition of 1998 Antarctic euphausiidsa
`
`E. superba (adult)
`E. superba (juv-large)
`E. superba (juv-small)
`E. tricantha
`E. frigida
`
`n
`
`4
`2
`2
`3
`3
`
`Wax ester
`
`Triacylglycerol
`
`Free fatty acid
`
`Sterol
`
`Polar lipid
`
`y1
`Lipid as mg g wet mass
`
`Lipidyindividual (mg g
`
`)y1
`
`–
`–
`–
`5.8"1.8
`–
`
`26.0"7.4
`38.4"3.4
`22.1"0.3
`16.2"7.4
`11.6"2.6
`
`1.3"0.3
`1.1"0.5
`1.8"0.7
`3.2"1.0
`1.8"0.3
`
`6.1"0.9
`4.1"0.8
`4.0"0.2
`7.4"1.1
`5.8"0.6
`
`66.6"6.3
`56.4"2.1
`72.0"0.6
`67.5"6.4
`80.8"3.3
`
`14.5"4.3
`14.6"4.3
`15.9"2.7
`12.4"3.5
`13.9"0.9
`
`7.2"2.1
`14.6"4.3
`8.0"1.3
`3.9"1.6
`2.1"0.2
`
`a
`
`Presented as mean"S.D.; –, below detection.
`
`RIMFROST EXHIBIT 1173 Page 0004
`
`

`

`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`737
`
`Table 2
`Percentage sterol composition of 1997 and 1998 Antarctic euphausiidsa
`
`Sterol
`
`24-Nordehydrocholesterol
`24-Nordehydrocholestanol
`Occelasterol
`trans-Dehydrocholesterol
`Cholesterol
`Cholestanol
`Desmosterol
`Brassicasterol
`Brassicastanol
`24-Methylenecholesterol
`Other
`
`1997
`
`E. superba
`(ns2)
`0.1"0.2
`–
`–
`1.1"0.4
`80.0"2.5
`0.1"0.1
`18.2"1.8
`0.5"0.0
`–
`0.1"0.1
`–
`
`E. tricantha
`
`–
`–
`–
`–
`94.3
`–
`5.7
`–
`–
`–
`–
`
`T. macrura
`(ns2)
`1.7"2.1
`tr
`0.1"0.1
`1.5"1.0
`81.2"15.3
`0.1"0.1
`6.5"1.3
`1.7"1.8
`0.2"0.2
`0.4"0.2
`6.7
`
`1998
`
`E. superba
`(adult, ns4)
`–
`–
`–
`–
`92.8"5.0
`–
`1.7"1.3
`–
`–
`–
`5.5
`
`a
`
`Presented as mean"S.D.; –, below detection; tr, trace (below integration).
`
`E. superba
`(juv-large, ns2)
`–
`–
`–
`–
`86.9"0.9
`–
`3.8"0.5
`–
`–
`–
`9.3
`
`E. superba
`(juv-small)
`
`–
`–
`–
`–
`88.6
`–
`2.9
`–
`–
`–
`8.5
`
`E. tricantha
`(ns3)
`–
`–
`–
`–
`100.0"0.0
`–
`–
`–
`–
`–
`–
`
`E. frigida
`(ns3)
`–
`–
`–
`–
`96.8"2.8
`–
`–
`–
`–
`–
`3.2
`
`RIMFROST EXHIBIT 1173 Page 0005
`
`

`

`738
`
`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`Table 3
`Percentage fatty acid composition of 1997 and 1998 Antarctic euphausiidsa
`
`Fatty acid
`
`1997
`
`1998
`
`E. tricantha
`
`E. frigida
`
`a
`
`?
`
`E. superba
`T. macrura
`E. superba
`E. superba
`E. superba
`E. tricantha
`E. frigida
`(ns4)
`(ns3)
`(adult, ns4)
`(juv-large, ns2)
`(juv-small, ns2)
`(ns3)
`(ns3)
`–
`–
`2.8
`–
`–
`–
`–
`–
`–
`12:0
`5.9"4.0
`15.3"43.8
`8.1"1.6
`8.6"0.5
`6.1"1.4
`5.1"1.2
`7.2"0.8
`2.3
`9.5
`14:0
`21.1"5.2
`22.9"0.7
`22.0"0.5
`22.4"1.1
`23.2"1.1
`17.6"0.7
`25.6"3.1
`25.1
`29.1
`16:0
`0.8"0.1
`4.0"5.9
`0.8"0.2
`0.8"0.0
`1.1"0.4
`0.8"0.1
`0.9"0.1
`1.5
`13.3
`18:0
`27.8"9.3
`42.2"11.4
`30.9"2.3
`31.8"1.6
`30.4"2.9
`23.5"2.0
`33.7"4.0
`28.9
`54.7
`Sum SFA
`16:1(ny7)c
`4.5"1.1
`3.1"0.3
`4.3"0.7
`3.5"1.3
`2.9"0.5
`5.6"0.9
`7.4"0.3
`2.3
`1.5
`16:1(ny5)c
`0.7"0.3
`1.0"0.3
`0.8"0.1
`0.5"0.2
`0.5"0.1
`0.6"0.1
`0.3"0.0
`0.2
`0.1
`18:1(ny9)c
`6.7"2.6
`9.6"0.5
`8.0"1.0
`7.7"1.6
`7.6"0.7
`18.9"1.7
`10.5"0.1
`11.9
`20.5
`18:1(ny7)c
`8.0"2.9
`6.4"1.4
`8.9"0.4
`7.7"0.8
`7.2"0.8
`9.6"0.7
`9.6"0.1
`13.0
`2.9
`20:1(ny9)c
`0.5"0.2
`0.8"0.4
`0.3"0.1
`0.1"0.0
`0.2"0.0
`2.2"0.6
`0.7"0.1
`1.0
`0.7
`20.4"7.1
`20.9"2.9
`22.3"2.3
`19.5"3.7
`18.4"2.1
`36.9"4.0
`28.5"0.6
`28.4
`25.7
`Sum MUFA
`C PUFA
`0.9"0.7
`0.2"0.3
`0.7"0.3
`1.5"0.8
`1.0"0.3
`0.3"0.0
`0.2
`0.1
`–
`16
`18:5(ny3)
`0.6"0.2
`1.2"0.5
`1.0"0.1
`0.2"0.1
`0.5"0.6
`–
`–
`–
`–
`18:4(ny3)
`5.2"4.6
`0.7"0.3
`3.7"1.2
`9.4"1.6
`8.9"0.5
`0.5"0.1
`0.6"0.1
`0.4
`–
`18:2(ny6)
`3.2"0.1
`2.1"0.8
`3.8"0.3
`2.4"0.3
`3.2"0.5
`2.1"0.1
`1.9"0.0
`1.7
`1.6
`18:3(ny6)
`1.3"0.7
`0.3"0.2
`1.7"0.4
`3.5"0.7
`3.7"0.3
`0.5"0.0
`0.8"0.0
`0.3
`0.3
`20:4(ny6)
`1.1"0.3
`0.6"0.1
`–
`1.2
`0.4
`–
`–
`–
`–
`20:5(ny3)
`20.5"5.7
`17.2"2.0
`17.1"1.5
`15.2"1.6
`15.0"1.1
`11.9"1.8
`17.6"2.9
`18.0
`3.6
`22:6(ny3)
`14.2"7.2
`10.2"2.9
`11.8"2.2
`9.3"0.5
`10.5"0.0
`13.5"2.5
`8.9"1.8
`15.6
`5.0
`45.3"19.0
`30.7"6.6
`39.4"6.1
`42.5"6.0
`43.3"2.8
`29.8"4.9
`31.2"5.5
`37.3
`11.0
`Sum PUFA
`6.5
`5.4
`8.6
`6.2
`7.4
`6.2
`7.9
`9.8
`6.6
`Other
`)y1
`11.7"5.1
`17.7"6.0
`9.2"2.5
`10.5"3.2
`12.7"0.6
`9.3"3.0
`9.6"1.1
`Total (mg g
`9.7
`7.0
`)y1
`2.2"0.6
`2.9"0.7
`1.5"0.3
`1.6"0.3
`1.9"0.1
`1.1"0.3
`1.7"0.5
`EPA (mg g
`1.7
`0.2
`)y1
`1.5"0.5
`1.7"0.2
`1.0"0.1
`1.0"0.2
`1.3"0.1
`1.2"0.2
`0.9"0.3
`DHA (mg g
`1.5
`0.3
`Ratio EPAyDHA
`1.44
`1.15
`0.72
`1.68
`1.44
`1.63
`1.43
`0.88
`1.99
`Presented as mean"S.D.; SFA, saturated fatty acids; MUFA, monosaturated fatty acids; PUFA, polysaturated fatty acids; EPA, eicosapentaenoic acid w20:5(ny3)x; DHA,
`docosahexaenoic acid w22:6(ny3)x; other components present at -1%: 12:0, 13:0, 15:0, 17:0, 20:0, 22:0, i14:0, i15:0, a15:0, i16:0, i17:0, a17:0, i19:0, 4,8,12-TMTD, 14:1(ny
`7)c, 14:1(ny5)c, 15:1, 16:1(ny7)t, 17:1, 18:1, 18:1(ny7)t, 18:1(ny5), 19:1, 20:1(ny11y13), 20:1(ny7)c, 22:1(ny11)c, 22:1(ny9)c, 22:1(ny7)c, C PUFA, 20:3(ny6),
`20:4(ny3), 20:2, 20:2(ny6), C PUFA, 22:4(ny6), 22:5(ny3), 22:2(ny6), C PUFA, C PUFA.
`
`21
`
`26
`
`28
`
`RIMFROST EXHIBIT 1173 Page 0006
`
`

`

`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`739
`
`the exception of 1997 E. frigida (0.024) and one
`1997 E. superba sample (0.032; Fig. 1). Separation
`between species was obtained with some overlap
`occurring.
`
`4. Discussion
`
`4.1. Lipid content and class composition
`
`Lipid content varied between krill species and
`between juvenile size classes in E. superba. Large
`y1
`juveniles of E. superba (14.6 mg g wet mass)
`showed the highest TAG (38.4% of total lipid;
`Table 1). Lipid content was 12.4–15.9 mg gy1
`wet mass in all other euphausiid species, which
`also had relatively less TAG (11.6–26.0%; Table
`1). TAG is a short-term energy reserve lipid (Lee
`et al., 1971), and a better fuel than PL. Higher
`TAG levels probably reflect greater feeding activ-
`ity in E. superba small juveniles. All krill (Table
`1) were collected in February, during the Antarctic
`summer, a time when phytoplankton production is
`at its maximum. The lipid level, lipid class and
`fatty acid composition of Thysanoessa species
`from Norwegian waters have been shown to
`change dramatically during this intense growth
`phase (Falk-Petersen, 1981). E. superba adults
`from the Weddell Sea have a large seasonal accu-
`mulation of lipid (average 28.2% lipid as% dry
`mass), whereas after winter, lipid values fall to
`10.5% (Hagen et al., 1996). Phosphatidylcholine
`appears to also be an energy storage lipid in E.
`superba (Hagen et al., 1996; Mayzaud, 1997).
`Polar lipid was high in all krill species from the
`AMLR study area (56–81% of total lipid; Table
`1). A similar PL level was found for Thysanoessa
`inermis (Falk-Petersen et al., 1982). Low lipid
`levels (high PL and low neutral
`lipid) usually
`indicate a low energy status. Individual phospho-
`lipids were not determined in this study. It has
`also been suggested that proteins contributed most
`energy during long term starvation in krill (Ikeda
`and Dixon, 1982). WE, a long-term energy reserve
`molecule were only detected in E. tricantha (6%
`of total lipid; Table 1). Moderate levels of WE
`(5–8%) were also reported form E. tricantha from
`the AMLR study area in 1996 (Phleger et al.,
`1998). Thysanoessa macrura, not analyzed in this
`study, has also been reported to have high WE
`(Hagen et al., 1996; Kattner et al., 1996; Falk-
`Petersen et al., 1999; Phleger and Nichols, unpub-
`lished data).
`
`RIMFROST EXHIBIT 1173 Page 0007
`
`Fig. 1. Dendrogram of cluster analysis comparing Antarctic
`euphausiids using fatty acids (performed using all fatty acids
`analysed), (individual samples; mgyg wet mass).
`
`present at lower levels (0–0.6%) in other krill
`species (Table 3; Fig. 3). The PUFA 18:5(ny3),
`not detected in 1997 samples, comprised 0.2–1.2%
`in all 1998 samples.
`The C and C very-long chain-polyunsaturat-
`26
`28
`ed fatty acids (VLC-PUFA) were identified by GC
`retention time data, with confirmation by GC–MS.
`The mass spectra of these components are similar
`to those of the other more common marine-derived
`PUFA with prominent ions at myz 79 and 91. The
`VLC-PUFA were not present in 1997 samples, but
`were detected as minor components (trace–0.1%)
`in 1998 samples of all krill except E. frigida.
`Monounsaturated fatty acids (MUFA) were 18–
`37% of the total FA (Table 3). MUFA included
`primarily 18:1(ny9)c (7–21%), 18:1(ny7)c (3–
`13%) and 16:1(ny7)c (2–7%) (Table 3). Highest
`levels of 18:1(ny9)c were found in 1997 E.
`frigida (21%) and 1998 E. tricantha (19%; Table
`3, Fig. 2). Although low levels (-1%) of
`20:1(ny9) and 20:1(ny7) were present in most
`krill species, 2.2% of 20:1(ny9) was detected in
`1998 E. tricantha (Table 3). The principal saturat-
`ed FA in all samples were 16:0 (18–29%), 14:0
`(2–15%) and 18:0 (1–13%) (Table 3).
`y1
`The fatty acid profiles (mg g
`wet mass) of
`individual samples were compared by cluster anal-
`ysis and MDS. All euphausiids were joined at a
`distance of F0.013 in the cluster analysis, with
`
`

`

`740
`
`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`Fig. 2. Ratio 18:1(n-7) to 18:1(n-9) versus ratio eicosapentaenoic acid (EPA) to docosahexaenoic acid (DHA) in Antarctic euphausiids (mean values). High 18:1(n-17)y18:1(n-
`9) ratios are consistent with an herbivorous diet. High EPAyDHA ratios reflect a diatom diet, while low ratios reflect a flagellate diet. (d), our data; (s), data from Virtue et al.
`(1993a,b), Virtue (1995).
`
`RIMFROST EXHIBIT 1173 Page 0008
`
`

`

`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`741
`
`Fig. 3. Percent 18:4 versus ratio 16:1 to 16:0 in Antarctic euphausiids (mean values). Low levels of 18:4 are consistent with an omnivorous diet. High 16:1y16:0 reflect a diatom
`diet. (d), our data (s), data from Virtue et al. (1993a,b), Virtue (1995).
`
`RIMFROST EXHIBIT 1173 Page 0009
`
`

`

`742
`
`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`4.2. Sterols
`
`The principle ST of all euphausiids analyzed
`was cholesterol (80–100%, of total ST) with lesser
`amounts of desmosterol (0–18%) (Table 2). Crus-
`taceans are incapable of de novo ST synthesis and
`depend on diet and phytosterol dealkylation (Goad,
`1978). Desmosterol is produced as an intermediate
`from phytosterol dealkylation and is also found in
`marine microalgae Nitzschia closterium (100%
`desmosterol) and Rhizoselenia setigera (94% des-
`mosterol) (Barrett et al., 1995). The waters where
`some of these krill were collected, known as
`Bransfield Strait waters, are dominated by micro-
`planktonic diatoms, such as Chaetoceros, Nitzschia
`and Rhizoselenia (Villafane et al., 1995). Desmos-
`˜
`terol present in krill may be from dealkylation of
`dietary phytosterols or, alternatively, directly sour-
`ced from the diet. Testing of the latter hypothesis
`will require examination of the ST content of
`phytoplankton from this region.
`The presence of ST other than cholesterol and
`desmosterol in 1997 E. superba and T. macrura
`(not in 1998 samples of these species) may indi-
`cate different food sources between these two
`years. 24-Nordehydrocholesterol and trans-dehy-
`drocholesterol (0.1–1.7 and 1.1–1.5%, respective-
`ly; Table 2) may be intermediates in cholesterol
`synthesis or derived from animal sources as well
`as phytoplankton. 24-Nordehydrocholesterol was
`3–12% of the total ST, and trans-dehydrocholes-
`terol 2–24% in a number of Cnidaria and Cteno-
`phora species collected from the AMLR study area
`during 1997 and 1998 (Nelson et al., 2000).
`Similar
`levels of 24-nordehycholesterol were
`observed in 1996 Cnidaria from the same study
`area (Phleger et al., 1998). trans-Dehydrocholes-
`terol was the major ST in the Antarctic ice diatom
`Nitzschia cylindrus (Nichols et al., 1986), and the
`structurally
`similar
`27-nor-24-methylcholest-
`5,22E-3b-ol (occelasterol) has also been reported
`as the major ST in an Antarctic dinoflagellate
`(Thompson, unpublished data). Care is needed, by
`examination of other signature lipids and by other
`procedures, before the exact algal source of these
`minor ST present in krill can be ascertained. The
`low levels of brassicasterol observed in 1997 E.
`superba and T. macrura also may have originated
`from dietary prymnesiophytes, such as Phaeocys-
`tis, and diatoms, where it is a major ST (Nichols
`et al., 1991; Tsitsa-Tzardis et al., 1993). The low
`24-methylenecholesterol (0.1–0.4%; Table 2) may
`
`have originated from sources such as the diatom
`is the major ST (Tsitsa-
`Chaetoceros, where it
`Tzardis et al., 1993). High 24-methylenecholester-
`ol was detected in the solitary salp, Salpa
`thompsoni (52% of total ST) and 5–12% 24-
`methylenecholesterol was reported in aggregate
`salps of the same species (Phleger et al., 2000).
`This indicates different
`feeding preferences or
`sources for these salps as compared to the krill
`reported in this study.
`Low levels of stanols were only detected in
`1997 E. superba and T. macrura but not in 1998
`species (Table 2). A similar difference
`krill
`between these years was also observed in salps (S.
`thompsoni) and their commensal hyperiid amphi-
`pods (Vibilia antarctica, Cyllopus lucasii and C.
`magellanicus) from the AMLR study area. Four
`stanols were detected in 1997 salps, and in their
`commensal
`amphipods;
`the
`stanols
`neither
`occurred in the 1998 salps nor in their amphipods
`(Phleger et al., 2000). The presence of stanols
`reflects diet. For example, in the marine dinoflag-
`ellate Scrippsiella sp., cholestanol comprised 24%
`of total stanols (Mansour et al., 1999) and a
`species of Gymnodinium isolated from Australian
`waters contained 24% dinostanol (Nichols et al.,
`1984). Dinostanol was not detected in the krill
`analyzed in this study. The fact that stanols were
`only detected in some 1997 krill species is proba-
`bly indicative of changes in the importance of the
`dietary sources of these components or products
`of bacterial degradation.
`
`4.3. Fatty acids
`
`Separation between species was represented in
`cluster analysis (Fig. 1). A scatterplot of MDS
`y1
`using FA profiles (mg g wet mass) of individual
`samples showed similar separation, although the
`1998 E. superba grouped tighter than 1997 E.
`superba (data not shown). The separations used
`the full suite of FA analyzed; similar separation
`also occurs when the FA suite is reduced (e.g.
`selection of components )1% of total FA). All E.
`superba were closely joined in the cluster analysis,
`except one 1997 E. superba sample (Fig. 1). Since
`1998 was considered as a ‘salp year’ and 1997
`was a ‘transition year’ (Phleger et al., 2000), 1997
`E. superba may receive a more diverse diet, which
`is reflected in the more varied FA profile. Although
`the composition of zooplankton for these years is
`known (Loeb et al., 1998), the composition of the
`
`RIMFROST EXHIBIT 1173 Page 0010
`
`

`

`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`743
`
`diet of E. superba (mostly phytoplankton) was not
`determined during the AMLR field studies and is
`therefore not known.
`For 1997 T. macrura, cluster analysis joined the
`animals at a distance of 0.009 (Fig. 1). Addition-
`ally, the small and large 1998 E. superba juveniles
`group together separated from the adults. 1998 E.
`tricantha and E. frigida are distinct and separate,
`however, 1997 samples do not cluster near their
`1998 counterparts (Fig. 1). These differences may
`also reflect variations in diet, as a separation was
`also evident for plots of selected biomarkers (Figs.
`2 and 3, and see below).
`Our data for the PUFA eicosapentaenoic acid
`wEPA, 20:5(ny3)x and docosahexaenoic acid
`wDHA, 22:6(ny3)x (Table 3) in E. superba from
`the Elephant Island AMLR study area are similar
`to those reported by Virtue et al. (1993a, 1997)
`for E. superba from the Prydz Bay region of
`Antarctica. These long chain PUFA, EPA and
`DHA, are dominant in the phospholipid fraction
`of krill where they appear to function in membrane
`structure (Virtue et al., 1993b). Higher proportions
`of EPA were present in phosphatidylcholine of E.
`superba compared with phosphatidyethanolamine
`(Mayzaud, 1997). A dietary source of these PUFA
`is considered essential in krill and other crustacean
`species, and phytoplankton, especially diatoms and
`other phaeophytes, are rich in PUFA which they
`synthesize (Brown et al., 1989). Higher EPAy
`DHA ratios in krill may be due to metabolic rate
`variations of these two PUFA. However, they more
`probably reflect differences in the diet. For exam-
`ple, diatoms contain high EPA (Falk-Petersen et
`al., 1998), whilst dinoflagellates and other flagel-
`lates have elevated DHA (Falk-Petersen et al.,
`2000). E. superba had EPAyDHA ratios of 1.4–
`1.6 for 1997 and 1998 adults and juveniles (Fig.
`2; Table 3). Major diatom FA include EPA,
`16:1(ny7) and C PUFA (Volkman et al., 1989;
`16
`Dunstan et al., 1994). The ratios of EPAyDHA
`were less in 1997 E. frigida (0.7) and 1998 E.
`tricantha (0.9) than in E. superba (Fig. 2; Table
`3). These may reflect a different dietary input
`since E.
`frigida is carnivorous, E.
`tricantha is
`omnivorous, and E. superba is herbivorous in
`earlier life stages and omnivorous as an adult.
`Higher levels of oleic acid w18:1(ny9)cx were
`frigida (21%) and 1998 E.
`found in 1997 E.
`tricantha (19%) with lower levels in E. superba
`w7–8% of 18:1(ny9)cx. Higher levels of oleic
`acid are also more consistent with a carnivorous
`
`diet (Nelson et al., 2000). Supporting evidence is
`provided by the ratios of 18:1(ny7)cy18:1(ny
`9)c which were mostly lower (0.1–1.1 for 1997
`and 0.5–0.9 for 1998; Fig. 2; Table 3) for these
`species, whereas the ratios for E. superba of
`18:1(ny7)cy18:1(ny9)c were higher (0.9–1.2
`for both years). In addition, relatively low levels
`of 20:1(ny9)c (-1% of total FA) were detected
`in all krill species except 1998 E. tricantha (2.2%)
`and -1% 20:1(ny7)c (Table 3). These FA are
`found in high levels in copepods and other crus-
`taceans (Kattner et al., 1994). Levels of 4–10%
`of 20:1(ny9)c and 20:1(ny7)c were found in
`1996 E. tricantha from the same AMLR study
`area (Phleger et al., 1998). E.
`tricantha is a
`mesobathypelagic euphausiid which is predomi-
`nantly carnivorous and the carnivory is reflected
`in these higher 20:1(ny9)c values (Table 3).
`Lower 16:1(ny7)c values (2–3%) were noted
`in 1997 E. tricantha, E. frigida and T. macrura,
`which indicates less importance of diatoms in the
`diet (Cripps et al., 1999). The 16:1(ny7)c values
`for E. superba were somewhat higher (5% for
`1997, and 3–4% 1998) as were 1998 values for
`the other krill species (6–7%; Table 3). 16:1y16:0
`ratios for all krill were in the range 0.05–0.57
`(Fig. 3; Table 3). Higher values have been pro-
`posed to indicate a diatom-containing diet (Volk-
`man et al., 1989). In E. superba, conversion of
`16:1(ny7)c to 18:1(ny7) is occurring.
`The PUFA 18:4(ny3), which was 4–9% of
`total FA in E. superba from 1997 and 1998 (Table
`3), is a major FA in the prymnesiophyte Isochrysis
`sp. (T-ISO) and in the cryptomonad Chroomonas
`salina (Volkman et al., 1989). The increased levels
`of 18:4(ny3) in particulate matter samples and
`the copepod Paralabidocera antarctica from coast-
`al waters, at Davis Station in Eastern Antarctica,
`was suggested to be indicative of Cryptomonas
`spp. (Swadling et al., 2000). In composition, the
`proportions of the various PUFA and other FA in
`the krill are suggestive of a phytoplankton diet,
`specifically diatom and cryptomonads for E. super-
`ba. The lower levels of 18:4(ny3) (0–0.6%) in
`other krill species, including E. tricantha, E. fri-
`gida and T. macrura, is more consistent with a
`carnivorous or omnivorous diet (Fig. 3). Another
`source of 18:4(ny3) could be from 18:5(ny3),
`which is rapidly metabolized to 18:4(ny3) in
`cultured fish cells (Ghioni et al., 2001).
`The PUFA octadecapentaenoic acid w18:5(ny
`3)x, present at 0.2–1.2% of total FA in all 1998
`
`RIMFROST EXHIBIT 1173 Page 0011
`
`

`

`744
`
`C.F. Phleger et al. / Comparative Biochemistry and Physiology Part B 131 (2002) 733–747
`
`krill, was not detected in any 1997 samples (Table
`3). This unusual PUFA was first recognized as an
`important biomarker in marine zooplankton by
`Mayzaud et al. (1976). It was also only detected
`in 1998 AMLR Antarctic salps and their commen-
`sal amphipods, AMLR gelatinous zooplankton,
`and AMLR free amphipods, but neither in 1997
`nor 1996 zooplankton (Phleger et al., 1998, 2000;
`Nelson et al., 2000, 2001). 18:5(ny3) is a major
`FA in dinoflagellates (Nichols et al., 1984) and
`coccolithophorids (Volkman et al., 1981), and is
`synt