`
`Vol. 173: 149-162, 1998
`
`MARINE ECOLOGY PROGRESS SERIES
`Mag Reo! Prog Ser
`
`d
`Published November1
`
`2
`
`Changesin lipid composition of the Antarctic krill
`Euphausia superbain the Indian sector of the
`Antarctic Ocean: influence of geographical
`location, sexual maturity stage and
`distribution among organs
`
`P. Mayzaud’, E. Albessard, J. Cuzin-Roudy
`
`Equipe d’Océanographie Biochimique et d'Ecologie, LOBEPM, URA-CNRS 2077, Observatoire Océanologique, BP 28,
`F-06230 Villefranche sur mer, France
`
`ABSTRACT: Lipid content and lipid class composition of Euphausia superba were studied at different
`levels for populations and individuals sampled in the Indian sector of the Antarctic Ocean. Strong site-
`to-site variability was recorded which could only partially be related to sex or development stage dif-
`ferences. Three groups of stations could be differentiated. Northern stations were characterized by
`‘high lipid-high triglyceride’ content, western and eastern locations by ‘high lipid - high phosphatidyl
`choline’ content and southern areas by ‘low lipid- high phosphatidyl ethanolamine/g}lycolipid’ content.
`Such variability was likely related to advected populations having spent variable lengths of time in the
`area studied. Lipid content and class among organs were studied in 5 body fractions: abdomen, stom-
`ach, digestive gland, gonad andfat body. In absolute terms, the highest concentrations were observed
`in the ovaries of mature females and the abdomens ofthe other stages. In relative terms [‘%dry weight),
`the digestive gland displayed the highest level, except in mature females. Distribution varied with
`stages, with low tnglyceride levels in abdomen tissues of most stages and in the fat body and stomach
`fractions of subadults. High triglyceride levels were recorded in the other fractions for post spawning
`females and males, as well as in the fat body fraction for mature females and in subadult gonads. A
`reverse pattern was cbserved for
`the relative content of phosphatidyl choline. Phosphatidyl
`ethanolamine showed maximum values in the abdomen and the gonad. Glycolipid percentages were
`maximum in the abdomen, suggesting a structural] role. The roles of the different lipid classes are dis-
`cussed with respect to the function of the organ.
`
`KEY WORDS: Knill- Lipids - Spatial heterogeneity - Maturity stage organs
`
`INTRODUCTION
`
`The role of lipids in Antarctic krill has been the con-
`cern of several papers in relation to reproduction
`(Clark 1980, 1984, Kolokowska 1991, Pond et al. 1995,
`Virtue et al. 1996), energy storage for overwintering
`(Quetin & Ross 1991, Hagen et al. 1996) and trophic
`interactions (Bottino 1974, Reinhardt & Van Vieet
`1986, Virtue et al. 1993a, b). Krill accumulate lipids
`mainly as triacylglycerols during the spring and sum-
`
`“E-mail: mayzaud@ccrv.obs-vlfr fr
`
`© Inter-Research 1998
`
`Resale of full article not permitted
`
`mer when phytoplankton are abundant (Clarke 1984,
`Hagen et al. 1996).
`Neutral lipids are utilized whenever energy levels
`exceed food intake. In knll the 2 major energy utilizing
`events are summer reproduction and winter survival
`under low phytoplankton conditions. Krill store signifi-
`cant amounts of lipids (Clarke 1984, Hagen et al. 1996),
`although most studies have concluded that the concen-
`trations are not sufficient to meet the energy require-
`ments during the winter, when food supply is low
`(Quetin & Ross 1991, Quetin et al. 1994}, However, the
`contribution of lipids to the overall survival strategy of
`
`RIMFROST EXHIBIT 1084
`
`page 0001
`
`

`

`150
`
`Mar Ecol Prog Ser 173: 149-162, 1998
`
`krill appears quite significant (Hagen et al. 1996) and,
`during the summer period,
`is channeled mainly into
`reproductive output.
`Krill reproduction takes place over several spawning
`phases
`(Cuzin-Roudy 1987, Quetin & Ross 1991);
`hence, their reproductive output seems directly related
`to the richness of the food supply (Ross & Quetin 1986).
`Yolk accumulation is characterized by high concentra-
`tions of neutral lipids and highlevels of triacylglycerols
`in ovaries of mature females (Clarke 1980, Hagen
`1988, Pond et al. 1995). Males do not seem to accumu-
`late neutral lipids in relation to spermatophore produc-
`tion, but various reports have suggested that
`the
`energy cost of frequent remating may be high, though
`distributed differently over time (Virtue et al. 1996).
`The resulting time integral may representa significant
`depletion in lipid content, as reported by Pond et al.
`(1995).
`Contrary to the high levels found in large calanoid
`copepods, wax esters are not present in krill lipids, and
`reliance on triacylglycerols as a storage moiety is now
`weil established (Clarke 1980, Hagen 1988, Mayzaud
`1997). Reports by Elligsen (1982), Saether et al. (1986),
`Hagen (1988}, and Hagen et al. (1996) suggested that
`polar
`lipids,
`and more
`specifically phosphatidyl
`choline (PC), may also serve as storage lipids. The
`involvement
`of
`cell
`structural components as
`an
`energy source has been reported for PC as a source of
`essential polyunsaturated fatty acids in fish for egg and
`larval development (see Fraser et al. 1988). This differ-
`entiation in the actual rele of the hpid classes could
`
`fue Aeuhion
`t
`i
`
`
`Maepn ’
`
`
`
`Fig. 1. Cruise tracks and stations surveyed as part of FIBEX (February 1981)
`
`provide additional resources during times of increased
`energy demands.
`Our knowledge on the sites of lipid synthesis and
`catabolism in krill is still limited. Lipid composition has
`been given for 3 main body fractions,
`i.e. abdomen,
`digestive gland and mature ovaries (Clarke 1980,
`Saether et al. 1985, Virtue et al. 1993a), with respect to
`neutral lipid accumulation. Little is known on the vari-
`ability of such composition with growth or sexual
`maturity stage. The potential role of the glyco-lipo-
`protein complex present in the fat body described by
`Cuzin-Roudy (1993) remains to be evaluated.
`The objective of the present study was to evaluate,
`for an open-ocean krill population,
`the influence of
`exogenous and endogenous factors in the control of
`lipids during summer. Changes in lipid concentration
`and composition were investigated at 2 different lev-
`els: population and specific organs or body fractions of
`different growth and maturity stages.
`
`MATERIAL AND METHODS
`
`Sampling. Euphausia superba were obtained from
`RMT8 oblique tows made to a depth of 100 m during 2
`cruises of the RV ‘Marion Dufresne’ in February 1981
`(FIBEX) and February 1994 (ANTARES 2}. Positions of
`sampling stations and cruise track for the first cruise
`are given in Fig. 1. Samples from the second cruise
`were obtained at 2 stations (66°41'S, 61°50'E and
`63°00'S, 70° 20'E). Krill were sorted immediately after
`
`RIMFROST EXHIBIT 1084
`
`page 0002
`
`

`

`Mayzaud et al.: Lipids in Antarctic krill
`
`151
`
` Fig. 2. Euphausia superba. Dissection of krill into 5 body fractions for
`
`the separation of the principal organs. 1 to S: Excision lines. dg: Diges-
`live gland; fb: fat body; ov: ovary; st: stomach
`
`capture, rinsed with distilled water and deep frozen
`(-80°C). They were stored at -70°C under nitrogen
`and transported to France within 4 mo.
`Additional
`samples were collected at different
`depths for particulate chlorophyll and proteins as
`descriptors of the trophic environment. Protocols and
`detailed results can be found in Mayzaud et al. (1985).
`Stage determination. For population study (FIBEX),
`at each station a subset of 20 individual krill were iden-
`tified to development or sexual stage, measured (BL:
`standard length 1; Mauchline 1980), weighed (wet
`weight: WW) and extracted. The rest of the sample was
`weighed and that part was treated as representative of
`the whole population. Female were separated into
`‘maturing’ and ‘post spawn’ categories. Maturing fe-
`males were recorded as IIIC (Makarov & Denys 1980).
`For the study of specific stages and organs, frozen
`krill were scored individually for sex, sexual develop-
`ment and maturity (Makarov & Denys 1980, Cuzin-
`Roudy & Amsler 1991) and measured for body length
`(BL) while thawing on an ice cold plate under a micro-
`scope. Mature female krill (BL = 44.52 to 58.79 mm)
`with a swollen thorax were staged IIID and SDS 7?
`(Fig. 2). Spent female krill (BL = 43.52 to 55.70 mm)
`had a small ovary and contracted and irregular lobes.
`The thoracic cavity was mainly filled with hemolymph.
`They were scored as SDS 9 rather than IIA, in order to
`take into account the ovarian regression which occurs
`normally at the end of the reproductive season (Cuzin-
`Roudy 1987, Cuzin-Roudy & Amsler 1991). Among the
`10 male krill dissected (BL = 36.75 to 60.52 mm), 4 were
`mature and 6 had empty ampullae and were scored
`‘post mature’. Immature young adults (BL = 38.35 to
`47.04 mm) will be referred to as subadults.
`Separation of main organs. The krill were dissected
`while thawing to separate either organs or body
`fractions containing a main organ. During the dissec-
`tion the specimens were placed on pre-weighted/
`pre-extracted filter paper (Whatman 42 extracted in
`chloroform:methanol 2:1}, which collected the fluids
`onginating from each step. Five fractions were sepa-
`
`rated (Fig. 2): (1} The abdomen fraction was obtained
`from excision lines 1 and 2 and contained mostly mus-
`cle and cuticle from the abdomen and the various
`
`appendages. (2) The anterior fraction containing prin-
`cipally the stomach was next obtained by excision
`line 3. Minor components were, in decreasing order:
`eyes, brain ganglia, and cuticle.
`(3) The digestive
`gland and the digestive tract were next excised as a
`whole from the anterior section of the thorax (excision
`line 4). Whenever the digestive gland was not suffi-
`ciently cohesive, recovery was accompanied by leak-
`age of greenish fluid, which was collected on the filter
`paper, extracted and combined.
`(4) The fraction con-
`taining the gonad was obtained from the dorsal part of
`the thorax. In mature females the thoracic cavity was
`overfilled with the swollen ovary and excision along
`line 5 could not be made without damaging it. The
`resulting leakage of fluid was collected on the filter
`paper, extracted and combined.
`(5) The last fraction
`contained the fat body, the conjunctive tissue which
`fills the ventral part of the thoracic cavity and comes in
`to contact with the ovary (Cuzin-Roudy 1993). A very
`minor component was the nervous tissue and cuticle.
`The wet weight of the 5 organs or body fractions was
`recorded, and additional specimens weredissected the
`same way to obtain wet weight/dry weight ratios. In
`this case dry weight was obtained after oven drying at
`60°C to constant weight.
`Lipid extraction and determination. Entire krill
`were placed frozen on crushed ice and brought to 0°C.
`Size (BL) and fresh weight
`(WW) were measured
`before lipid extraction, according to the method of
`Bligh & Dyer (1959). Either the extracted lipids were
`weighed in tarred vials or their concentration was esti-
`mated according to Barnes & Blackstock (1973) but
`with Euphausia superba lipids as standards instead of
`cholesterol. Both determinations yielded similar val-
`ues. The lipid extracts were then placed under nitro-
`gen at
`-70°C until analysis. Body fractions were
`extracted immediately after dissection using the same
`protocol,
`
`RIMFROST EXHIBIT 1084
`
`page 0003
`
`

`

`152
`
`Mar Ecol Prog Ser 173: 149-162, 1998
`
`Lipid classes were quantified after chromatographic
`separation coupled with FID detection on a Iatroscan
`Mark II] TH 10 (Ackman 1981}. Total lipid extracts
`were applied to chromarods SJ] using microcapillaries
`(1 pl} and analyzed in duplicate. Neutral lipids were
`separated using a double development procedure with
`the following solvent systems; n-hexane:benzene:for-
`mic acid 80:20:0.5 {by volume} followed by n-hexane:
`diethylether:formic acid 97:3:0.5 (v/v). Glycolipids
`were separated according to Hirayama & Morita (1980)
`with chloroform:ethyl acetate:acetone:methanol:acetic
`acid:H,O 60:12:15:16:3:3 (v/v). Phospholipids were
`separated with chloroform:methanol:H,0 65:35:4 (v/v).
`Individual calibration of rods was achieved with com-
`mercial standards according to Ackman (1981). Using
`the solvent system indicated, all listed lipid classes,
`and in particular free fatty acids, were separated and
`accounted for.
`
`To validate the Jatroscan separation and identifica-
`tion, neutral and polar lipids were further isolated ona
`preparative scale by column chromatography on silica
`gel (Bio-Sil HA, minus 325 mesh}. The neutral lipid
`fraction was eluted with 6 column volumes of chloro-
`form, the acetone mobile compounds were eluted with
`4 volumes of acetone and the phospholipids were
`eluted with 6 volumes of methanol. All operations took
`place under nitrogen. Each fraction collected was fur-
`ther separated by thin-layer chromatography (TLC) on
`pre-coated silica gel plates (Analtech, Uniplate) and
`developed with hexane:diethylether:acetic acid 80:20:
`1.5 (v/v)
`for neutral hpids, or chloroform:methanol.
`aqueous ammonia 85:30:1 (v/v) for glycolipids, and
`chloroform:methanol:acetic acid:H,0 25:15:4:2 (v/v}
`for polar lipids. Lipid classes were visualized using
`dichlorofluorescein and identification was achieved by
`comparison with standard mixtures. Specific detection
`of phospholipids and glycolipids by TLC was made
`with molybdenum blue and diphenylamine reagents
`(Stahl 1969). Each of
`these lipid classes was then
`applied to chromarods SHI] and developed as previ-
`ously to confirm both identification and retention times
`of the peaks recorded with total extract.
`No attempt was made to hydrogenate the samples
`(Shantha & Ackman 1990) because no evidenceof sub-
`fractionation effects due to the vanetyof fatty acids could
`be detected. As recently shown by Miller et al. (1998),
`the use of commercial] standards resulted in an underes-
`
`timate of the actual concentration of triglycerides but
`without changes in the shape of the FID response.
`Data analysis. The allometric relations for the differ-
`ent maturity stages (WW = aBL°)} were computed after
`log-log transformation and model ] regression (Sokal &
`Rohlf 1981}.
`One- and two-factor vanance analyses were per-
`formed for total lipids as well as lipid classes. Multiple
`
`comparisons of means were achieved using the Tukey
`test procedure. Comparison of linear regression equa-
`tions was achieved by covariance analysis. All
`the
`above procedures were carried out according to Sokal
`& Rohlf (1981). Systat 7.0 statistical package was used
`for all bivariate tests (Wilkinson 1996).
`Principal component analysis (PCA) was performed
`after arcsine transformation to normalize percentage
`data. Details on the method and means of interpre-
`tation are given in Mayzaud et al. (1989). To facilitate
`the representation of the factor scores structure, an
`ascending hierarchical clustering method was used
`(Lebart et al. 1995) to group those observations which
`displayed maximum similitude and to produce a num-
`ber of classes best represented in a dendrogram.
`Correspondence analysis
`(Benzecri 1969, Gower
`1987) was performed on a data matrix transformed to
`relative frequencies and scaled so that each row {or
`column} can be viewed as a row (or column) of condi-
`tional probability distribution. Distances between pro-
`files were computed with y? metrics. This distance
`gives symmetry to the 2 sets of data so that each facto-
`rial axis of the cloud of variables corresponds to a fac-
`torial axis of the cloud of observations. Thus,
`it was
`possible to represent simultaneously descriptors and
`observations on the plane defined by the factorial axes.
`Interpretation and representation followed that de-
`scribed above for PCA.
`
`Computation of multivariate tests was made using
`the SPAD 3.0 software (Lebart et al, 1995).
`
`RESULTS
`
`Size, weight andlipid relationships
`
`The size, wet weight and lipid content of the 4 cate-
`gories of krill collected during the FIBEX cruise are
`summarized in Table 1. Subadults displayed the small-
`est size and weight of the 4 groups, while males and
`females showed simular size ranges (31 to 44 mm) but
`slightly different wet weight distribution. Mean values
`were not statistically different (f-test, p < 0.05) but
`ranges suggested a trend towards maturing females
`and males being heavier (0.57 to 1.28 mg and 0.4 to
`1.21 mg, respectively) compared to post spawn fe-
`males. Lipid content relative to wet weight suggested
`maximum accumulation in maturing females (3.3%).
`Males tended to show minimum lipid content although
`the high variability in this case prevented statistical
`significance.
`The wet weight (WW)relationships with size or lipid
`content were established for each stage present in the
`samples collected,
`i.e. subadults, males and females.
`The log-log regressions between size and weight
`
`RIMFROST EXHIBIT 1084
`
`page 0004
`
`

`

`Mayzaud et al.; Lipids in Antarctic krill
`
`Table 1. Euphausia superba, Range and mean values of knll size, wet weight and total lipid content (% wet weight) of specific
`maturity stages of individuals collected during FIBEX. n = number of individuals analyzed
`
` n
`
`Subadults
`Males
`Maturing females
`Post spawn females
`
`
`Size
`Mean
`Wet weight Mean wet
`Total lipid
`Mean total
`range
`size
`range
`weight
`range
`lipid
`(mm)
`+S$D
`(rng)
`+SD
`(% wet wi)
`+ SD
`
`23-35
`31-43
`35-44
`35-42
`
`29.7 + 3.1
`37.1426
`39.3 + 2.6
`38.1 41.8
`
`0.20-0.59
`0.40-1.21
`O.57-1.28
`0.53-0.96
`
`0.37 +0.12
`0.74 +0.18
`0.93 + 0.22
`0.744011
`
`1.1-4.7
`0.7-5.0
`1.6-4.8
`1.1-4.8
`
`2940.9
`2.2409
`3.34 O18
`2Bi+t 1.2
`
`83)
`54
`14
`20
`
`(Fig. 3) appeared to be similar for all 3 stages (slope:
`Fy 4g = 2.07; intercept: Fo 1:2, = 0.26) corresponding to
`an overall regression equation of:
`
`logWW= —0.08 + 3.12logBL
`(t = 0.967, F, 424, = 1779, p < 0.0001}
`
`Relationships between WW and total lipids or phospho-
`lipids (Fig. 3] were all significant (p < 0.008). The re-
`gressions for male individuals showed slopes or inter-
`
`cepts different from the other 2 stages. Males and fe-
`males appeared to accumulate lipids and phospholipids
`at a similar rate (slope: F,, gg = 0.272) but with a lower in-
`tensity (intercept: F, g5 = 14.25) for a given size in males.
`Triglycerides illustrated a different pattern of changes
`with only 2 significant regressions, those for subadults (n
`= 33, r = 0.673) and females (n = 33, r = 0.556), which
`showeda significant difference in intercept (F;, g3= 5.54)
`but not in slope (Fy 65 = 0.04). In these 2 stages, triglyc-
`
`
`
`
`
`
`
`°
`
`a
`
`ig
`
`i
`
`5
`
`°
`
`+-O-+ Subudults
`—O— Males
`5
`—a~— Females
`
`
`0.00 —
`“1.00
`
`19F7
`
`0,75
`
`4,25
`-0.50
`Log Wet weight (g/ind)
`
`0,00
`
`‘
`
`a 5
`
`0.005
`
`=
`3
`cz
`& 9.25}
`ime
`»
`o
`a
`2
`a
`=,
`op
`§
`
`|
`
`0.75 7
`
`
`
`-1.00
`131
`
`
`i
`:
`" i
`141
`1.5)
`1.61
`Leg Length (mm)
`
`E 1.857
`B
`&
`e 1247
`<
`=
`=
`2 0.937
`S
`S
`7m
`S
`eB 062
`0 Subudults
`=
`a Mules
`=
`_& Fernates
`&
`S 031 L
`= *
`
`
`
`
`
`
`
`a
`
`e
`
`8
`
`.
`
`°
`
`o
`
`e
`
`ved?
`
`0.10 -——"——+
`-1.00
`0.75
`
`;
`4.25
`+),.50
`Log Wet weight (g/ind)
`
`——
`4.06
`
`£707
`=
`=
`g 1
`b
`3
`oh
`S 13er
`E as
`>
`J
`=
`&
`rs
`SF
`= 1.067
`y
`é
`3
`a
`5
`Hy
`4B 4.1
`‘S 9.74
`2
`=
`-
`—w— Females
`o
`2
`s
`
`
`s f=O6T—o~ Mates ;. 7
`
`
`
`gp 0.42
`=—orsacel we
`:
`3 Shits
`a
`oi
`= ae Females
`LAP
`—
`-1,00
`
`.
`
`14
`
`vac
`
`L
`
`7
`
`o
`
`Ca
`5 tet
`° eo *
`
`°
`
`
`
`
`
`4
`
`8.75
`
`4.25
`450
`Log Wet weight (g/ind)
`
`0.00
`
`Fig. 3. Euphausia superba. Log-log regressions, for the 3 major developmental stages, between wet weight and body length and
`between wet weight and lotal lipid, phospholipids and triglycerides content
`
`RIMFROST EXHIBIT 1084
`
`page 0005
`
`

`

`154
`
`Mar Ecol Prog Ser 173: 149-162, 1998
`
`erides were accumulated at a similar rate but with a
`
`lower intensity in females than in subadults. The lack of
`correlation for the male data set was to a large extent as-
`sociated with an increased variance, with low concen-
`trations in the mediumto large size range.
`
`Lipid content and spatial heterogeneity
`
`To test the relative influence of geographical loca-
`tion and biological descriptors we conducted an analy-
`sis of variance. The relative lipid content recorded at
`the population level was significantly related to stage
`distribution and location of sampling but not to size or
`weight (Table 2). The overall mean lipid percentage
`(2.6% wet weight) can be divided into 2 groups of sta-
`tions, irrespective of the sex or stage composition (t-
`test = 7.65, df = 48, p < 0.0001): (1) those north and west
`of the grid (Stns 2, 3,17, 18, 19, and 23) and those to the
`extreme southeast (Stns 9A and 10} with high concen-
`trations (mean value: 3.1% wet weight), and (2) those
`from the central and southern part of the grid (Stns 5,
`5A, 6, 9, and 11} with low concentrations (mean value:
`1.9% wet weight).
`Over the spatial grid surveyed during FIBEX, a rela-
`tively complex pattern of lipid classes was abserved for
`
`the lipid distribution of
`‘Table 2. Analysis of variance for
`Euphausia superba as "« wet weight over the FIBEX grid of
`stations
`
`
`Source of
`variation
`
`Main effect
`Station
`Stage
`Covariates
`Size
`Weight
`Residuals
`
`Sumof
`squares
`
`df Mean F-ratio Signif-
`squares
`icance
`level
`
`36.33
`16.81
`
`0.28
`0.06
`80.42
`
`3.03
`5.61
`
`0.28
`0.06
`0.76
`
`3.99
`739
`
`0.37
`0.07
`
`0.0000
`0.0002
`
`0.553
`0.789
`
`12
`a
`
`1
`1
`106
`
`eas
`
`Total (corrected)
`
`144.64
`
`
`
`the different growth and maturity stages, Both struc-
`tural polar lipids and reserve tnglycerides displayed
`positive linear relationships with total lipids, suggest-
`ing that part of the polar lipids acted as reserve lipids
`(Fig. 4). Interestingly, the relationship significantly dif-
`fered with the stage considered. A covariance analysis
`of the various regressions indicated that regressions
`for polar lipids showed a trend for increasing slopes
`(p < 0.05), with subadults (r = 0.498) < male (r = 0.579}
`
`Subadults
`
`Mature Females
`
`Polar lipids
`
`Polar lipids
`
`
`
`Triglycerides
`
`
`
`Polar lipids
`
`a P
`
`Males
`
`olarlipids
`
`
`
`
`
`MainClassesofLipids(mg/100gwetweight)
`
`Fig.
`
`4. Euphausia superba. Linear regressions between total lipid content and lipid classes for sexual and developmental stage
`
`Total Lipids (mg/100g wet weight)
`
`RIMFROST EXHIBIT 1084
`
`page 0006
`
`

`

`Mayzaud et al.: Lipids in Antarctic knill
`
`
`155
`
`AXIS2(22.0%)
`
`< mature females (r = 0.659) < spent
`females (r = 0.802} and the reverse for
`triglycendes. Within a given stage,
`polar hpid and triglyceride relation-
`ships were significantly different for
`males and spent females (F;,,95 = 5.76;
`F, 45 = 89.95, respectively) but not for
`subadults and mature females (Fy «7 =
`0.07 and F, 2 = 2.74, respectively),
`The spatial structure was analyzed
`using multivariate
`correspondence
`analysis. The factorial plane defined
`by the first 2 axes accounted for 97.5%
`of the total variance. The first and sec-
`
`ond axes were respectively related to
`the richness in triglycerides and phos-
`phatidyl choline (Fig. 5). Grouping of
`stations, based on proximity analysis,
`showed 3 major clusters, which com-
`prised (1) those stations with high pro-
`portions of
`triglycerides in the total
`lipids, located in the northern part of
`the grid (Stn 23 at 62°S; Stns 1? and 19
`at 63°S), (2) those stations with a high
`proportion of phosphatidyl choline in
`total Lipids, located at both the western
`and eastern edges of the grid (Stns 2, 3
`and 11), and (3)
`those stations with
`high percentages of structural compo-
`nents (phosphatidyl! ethanolamine, gly-
`colipids and to a minor extent phos-
`phatidy] choline), in the southern part
`of the grid (Stns 5A, 6, 9 and 9A at
`64° S) (Fig. 5).
`
`Distribution of lipids in body
`fractions and organs
`
`stg CHL
`.. SISA... LTOT
`
`-0.4
`
`-0.2
`
`0.0
`
`0.2
`
`0.4
`
`AXIS 1 (75.5%)
`
`b
`
`
`45%
`
`|
`
`s719
`STI7
`
`$23
`$710
`
`
`:
`
`Ss
`
`[ sté
`[I© srs
`30%
`L_[|
`sT9a
`/ STSA
`
`|| STi
`it
`ST2
`
`25%
`
`PE,
`
`
`
`
`the
`The respective contribution of
`different fractions and organs to the
`lipid composition of Euphausia superba
`was studied for different development
`and maturity stages: subadults, males,
`mature females and spent females.
`The mean sizes of the specimens dis-
`sected and the dry weight of the differ-
`ent fractions are presented in Table 3.
`Size varied over a
`relatively small
`range (42 to 51 mm) and was chosen so
`that vanability around mean size remained more or
`less constant. In terms of dry weight,
`if the mature
`females, for which the heaviest fraction was the ovary,
`are not included, the abdomen represented the largest
`body component. The other
`fractions and organs
`
`Fig. 5. Correspondence analysis of the spatial changes of total hpids and lipid
`classes of Euphausia superba collected during the FIBEX cruise.
`(a) Projection of
`biochemical descriptors and stalions on the first 2 axes. Descriptors of
`the
`trophic environment {proteins and chlorophyll) projected on the factorial piane
`as supplementary variates. LTOT:
`total
`lipids; PC: phosphatidy! choline;
`PE. phosphatidyl ethanolamine; GLYC: glycolipids; TG:
`triglycerides; CHL:
`chlorophyll; PRO: particulate proteins. For station (ST) locations refer to Fig.
`1
`(b) Dendrogram of hierarchical clustering of stations based on the scores on the
`hist 5 factorial axes. %: percent total vanance accounted for by each cluster
`
`showed dry weights ranging from 9.6 to 43.4 mg, with
`no significant differences (p < 0.05). Lipid contents in
`both absolute and relative terms are presented in
`Table 4, with the highest concentrations observed for
`the ovary in mature females and for the abdomen in
`
`RIMFROST EXHIBIT 1084
`
`page 0007
`
`

`

`156
`
`Mar Ecol Prog Ser 173: 149-162, 1998
`
`Table 3. Euphausia superba. Mean (+SD) size of individuals (mm) and dry weight
`(mg) of the different bodyfractions and organs. n = numberof individuals
`
`
`
`Females
`Mature
`508243
`n=10
`
`Spent
`471444
`n=12
`
`Subadults
`
`Males
`
`Meansize
`
`419#35
`Was
`Sod, leactions and-ongens
`Abdomen
`83.24243
`Digestive gland
`9644.3
`Fraction including:
`33.6 + 37.2
`23.8 + 10.5
`Gonads
`30.4 + 18.9
`11.2+64
`Fat body
`22.14 14.4
`10.1+4.8
`Stomach
`
`
`513448
`ASS
`
`142458
`20.6 + 7.8
`
`98.8430.2
`20.2 # 11.2
`
`23.92 139
`25.1 + 11.2
`19.9491
`
`85.4 +264
`43.4% 25.2
`
`LAS 3 82
`32.9 + 8.3
`22.6% 7.8
`
`triglyceride levels were recorded in
`abdomen tissues (mean: 28,7 + 2.1%)
`in most stages, and in the fat body and
`stomach fractions of
`the subadults
`Maximum percentages were observed
`inte Gthes traetiantor apetl semaine
`and males (respective means; 42,2 +
`0
`1.7% and 38.6= 1.8 %), as well as the
`fat body fraction in mature females
`and the gonad fraction in subadults.
`As anticipated, phosphatidyl choline
`(PC) displayed a reverse pattern with
`hidh levels in the abdomen cf all:the
`g
`;
`stages (mean: 47.6 + 1.8%) and in the
`gonad fraction in mature females.
`Minimum percentages occurred in the
`fat body and stomach of males and mature females as
`well as the digestive gland and gonad fraction of
`subadults. Phosphatidy! ethanolamine (PE) showed
`maximum values in the abdomen (mean: 6.4 + 0.4%}
`and minimum percentages in the gonad fraction
`(mean: 2.4 + 0.4%). Glycolipids (GLY) showed low per-
`centages in the abdomen (mean: 2.4 + 0.4%), mainly in
`males and post spawn females, intermediate values in
`the digestive gland and gonad fraction (respective
`means: 3.7 + 0.4% and 3.0 + 0.4%) and maximum val-
`ues in the fat body and stomach fractions (respective
`means: 4.7 + 0.5% and 4.1 + 0.4%). Monoglycerides
`(MAG) showed maximum values in the abdomen of
`most stages, except mature females. The variability in
`monoglycerides was significantly related to both stage
`and organ (ANOVA, p > 0.0003). Lysophosphatidyl
`
`most stages. On a percent dry weight basis, the diges-
`tive gland displayed the largest lipid content in all
`stages except in mature females. The significance of
`these differences was tested by covariance analyses,
`which confirmed that changes in lipid content were
`related to the above stages (p > 0.002) and organs (p >
`0.0001), and in absolute terms to size (p > 0.0001). The
`ovary in mature females is very rich and contains about
`60% of the total lipids, while in the other stages the
`abdomen contributes 30 to 40%to the total lipids, and
`the digestive gland accounts for 20 to 30%. Fat body
`and stomachfractions were in all categories minor con-
`tributors, with respective ranges of 8 to 14% and 4 to
`9% of the total lipids (Table 4).
`Distribution of lipid classes among organs showed
`different patterns with stages (Tables 5 & 6}. Low
`
`Table 4. Euphausia superba. Absolute (per mg) and relative (% dry weight of each organ] lipid content (+ SD) of the different or-
`gansor tissue fractions and lipid distribution within body components (%total lipid)
`
`
`
`Body tssue and organs
`
`Subadults
`
`Males
`
`Abdomen
`
`% total lipid located in tissue
`
`Digestive gland
`
`% total lipid located in tissue
`
`Fraction including:
`Gonads
`
`So total lipid located in fraction
`Fat body
`
`“a total lipid located in fraction
`Stomach
`
`"os total lipid located in fraction
`
`mg
`“DW
`
`mg
`% DW
`
`mg
`“DW
`
`mg
`~~DW
`
`mg
`% DW
`
`13.143.8
`[5:9 Sel
`40.8 41.6
`
`68233
`52:7 £ #8
`2A Ll
`
`69434
`28.3476
`21.4413
`2:9) 2.7
`24 Fe Oe
`2 = OF
`2441.6
`21 Get Ae
`7.4240.5
`
`15.2 +64
`11.2+2.5
`35.5 + 1.8
`
`14.1456
`63h 49
`S29 = 13
`
`4641.7
`26.1467
`10.7 #23
`5843.4
`22:5 4.4
`13.5 + 1.9
`3:1 # 1.3
`20.2 + 6.6
`7.32 0.6
`
`Spent
`
`13.5+4.6
`137 £278
`30.92 £3
`
`13.6 + 6.0
`46.2 4 11.0
`a2 1a
`
`CHa sil
`30.2 FL
`TS.206 19
`5 Gite 2.6
`24.3463
`13.6 + 1.8
`3.) ae Lies
`21.1 + G1
`9040.9
`
`Females
`
`Mature
`
`11.3 44.2
`13.5 + 3.9
`13.7 & 2.3
`
`10.4+#5.1
`25.4 + 10.0
`IZ7 £28
`
`49.8 + 16.0
`STS £76.
`605+1.8
`Falk 8
`21.9446
`8620.7
`Sh & 1G
`16.14 4.0
`45204
`
`RIMFROST EXHIBIT 1084
`
`page 0008
`
`

`

`Mayzaud et al.: Lipids in Antarctic krill
`
`La?
`
`Table 5. Euphausia superba. Lipid classes composition for the different organs and tissue fractions of subadult and male stages.
`LPC: lysophosphatidyl choline; PC: phosphatidyl choline; PE: phosphatidyl] ethanolamine; GLY: glycolipids; Chol: cholesterol,
`MAG: monoacylglycerols; TAG: triacy!glycerols. Standard deviations are given in parentheses
`
`Body tissue
`and organs
`
`Abdomen
`
`EG Pe
`_-
`
`5.7
`(1.0)
`
`45.3
`(10.0)
`
`4.9
`(1.8)
`
`3.4
`(0.7)
`
`Males
`Subadults
`
`Chol MAG TAG
`GLY
`PE
`PG
`LPC
`Chol MAG TAG
`GLY
`PE
`~~
`(%total lipids) ———
`- + (% total lipids)
`
`6.0
`(1.8)
`
`a3
`(1.8)
`
`30.9
`(7.8)
`
`1.8
`(0.7)
`
`$5.2
`(8.9)
`
`a7
`(3.6)
`
`1.9
`(0.3)
`
`32
`(1.3)
`
`11.4
`(2.4)
`
`20,1
`(4.0)
`
`37:3
`63
`0.9
`2.4
`2.9
`43.8
`2S
`39.6
`48
`1.8
`6.5
`3.8
`Send
`1.0
`Digestive gland
`
`
`
`
`
`
`
`
`
`
`
`
`(0.1) (0.7)=(2.2)(83) (1.4) (35) (1.5) (1.4) (5.3) (1.1) (12.6) (1.6) (1.6) (12.7)
`
`
` Body tissue
`
`(11.7)
`
`
`Praction including:
`Gonads
`3.0
`(2.4)
`6.1
`(1.3)
`5.5
`(2.6)
`
`Fat body
`
`Stomach
`
`30.3
`(7.0)
`50:5
`(7.2)
`38.7
`(8.1)
`
`La
`(1.4)
`a8
`(1.8)
`4.4
`(21)
`
`2:1
`(1.3)
`4.9
`(5.7)
`4.1
`(27)
`
`3.0
`(1.9)
`13
`(1.1)
`ce
`(2.2)
`
`oS
`(3.2)
`1.8
`(3.6)
`2.8
`(2.3)
`
`$1.0
`(6.0)
`29.1
`(5.4)
`85:2
`(16.3)
`
`0.9
`(0.8)
`1.6
`(1.2)
`2.6
`(1.8)
`
`46.6
`(20.5)
`34.9
`(98)
`36.8
`(9.9)
`
`17
`{1.0)
`3.4
`(41.7)
`3.9
`(23)
`
`2.4
`(2.0)
`29
`(26)
`3.2
`(1.8)
`
`2.0
`(1.8)
`28
`(1.4)
`2e'h
`(2.3)
`
`43.2
`(18.9)
`440
`(84)
`46.3
`
`3.0
`(3.0)
`94
`(1.8)
`53
`(3.2)
`
`Table 6. Euphausia superba. Lipid class composition forthe different organs and tissue fractions of females at 2 maturity stages.
`Abbreviations as in Table 5. Standard deviations are given in parentheses. tr trace
`
`
`Spent females
`Mature females
`
`Pc
`Lee
`PE
`GLY
`Chol MAG TAG
`PE
`GLY Chol MAG TAG
`(% total lipids) ----
`-
`- a (% total lipids) ——————.
`
`and organs
`
`PC
`LPC
`Sea
`
`Abdomen
`
`Digestive gland
`
`3.2
`(1.2)
`
`4.1
`(1.8)
`
`Fraction including:
`Gonads
`
`tr
`
`Fat body
`
`Stemach
`
`1.9
`(1.8)
`3.8
`(2.5)
`
`45.2
`(7.9)
`
`37.8
`(8.1)
`
`40.2
`(6.9)
`35.6
`(7.2)
`35.1
`(4.2)
`
`8.0
`(2.4)
`
`4.3
`(2.4)
`
`3.3
`(1.3)
`4.6
`(1.6)
`4.0
`(2.4)
`
`12
`(0.7)
`
`2.6
`(06)
`
`0.6
`{0.2}
`2.8
`(1.0)
`4.9
`(2.7)
`
`3.3
`(1.2)
`
`1.3
`(1.0)
`
`2:2
`(1.6)
`1.8
`(0.8)
`2.6
`{1.2}
`
`6.7
`(23)
`
`4.0
`(1.9)
`
`43
`(3.3)
`6.1
`(1.7)
`5.8
`(3.2)
`
`32.4
`(9.4)
`
`42.6
`(10.2)
`
`49.5
`(7.8)
`43.6
`(11.9)
`42.7
`(5.9)
`
`4.8
`(2.0)
`
`23
`(2.1)
`
`1.9
`(1.8)
`3:5
`{1.6)
`3.9
`(0.9)
`
`427
`(10.2)
`
`43.8
`(11.6)
`
`53.9
`(7.8)
`35.1
`(5.5)
`38.0
`(5.7)
`
`7.2
`(1.3)
`
`2.9
`(1.6)
`
`3.4
`(1.6)
`4.6
`(1.5)
`6.0
`(3.2)
`
`3.1
`(1.0)
`
`2.1
`(1.6)
`
`5.0
`(1.5)
`
`0.9
`(0.8)
`
`27
`(2.3)
`
`6.3
`(2.2)
`
`2.7
`3.0
`6.3
`(1.7)
`(1.0)
`(4.4)
`33 2 sil
`(1.7)
`(0.8)
`(0.3)
`43
`4.4
`1.4
`(2.5)
`(4.4)
`(1.2)
`
`$22)
`(8.1)
`
`37.3
`(12.8)
`
`26.7
`(8.8)
`44.3
`(5.1)
`39.2
`(9.8)
`
`no
`
`significant
`
`changes
`
`showed
`(LPC)
`choline
`(ANOVA, p < 0.05).
`lipid
`The
`relationship between distribution of
`classes, organs and stages can be more clearly
`described using a multivariate approach,
`ie. PCA.
`Three factorial axes were needed to explain 83% of
`the total variance. Correlation between axes and lipid
`descriptors showed that
`the first axis was strongly
`related to the opposition between PC and triqlycerides
`and accounted for 40.8% of the total inertia (Fig. 6).
`The second and third (not shown) axes accounted for
`22.8% and 19.6% of the total variance (Fig. 6}. They
`were respectively correlated to monoglycerides and
`PE for axis 2 and to glycolipids for axis 3. Hierarchical
`clustering of the organs and stages (Fig. 6) suggested 2
`major groupings, each associated with the dominance
`
`of 1 lipid descriptor: (1) high proportion of polar lipids
`(PC and PE) with the abdomen from all stages and the
`various fractions
`from mature females (except
`fat
`body}, and most
`fractions
`from subadults (except
`