Lipid, fatty acid and protein utilization during lecithotrophic larval development of Lithodes santolla (Molina) and Paralomis granulosa (Jacquinot)

Abstract

During the larval development of the subantarctic king crab, Lithodes santolla, and stone crab, Paralomis granulosa, we compared changes in the carbon, fatty acid and protein contents of larvae reared under constant conditions from hatching to metamorphosis, either in presence or absence of food (Artemia spp. nauplii). In both species the feeding condition had no influence on any of the chemical parameters studied, indicating a fully lecithotrophic (i.e. non-feeding) mode of development from hatching of the first zoea to metamorphosis of the late megalopa. Dry mass and carbon contents at hatching were similar in the larvae of both species, but L. santolla contained initially higher total amounts of fatty acids and protein than P. granulosa. Both species utilized considerable portions of their total fatty acid pool which decreased logarithmically throughout the time of development. At metamorphosis, it was almost exhausted in P. granulosa, while L. santolla had consumed only about 60%. Protein utilization, in contrast, was higher in L. santolla (40%) than in P. granulosa (20%). Triacylglycerol was the principal storage lipid in both species, accounting initially for about 75% of the lipid fraction; it was strongly utilized during larval development. Phospholipid constituted the second largest lipid class; it also decreased in P. granulosa, but to a lesser extent in L. santolla. The major fatty acids of both species were 18:1(n−9), 20:5(n−3) and 16:0 as well as, in lower proportions, 18:1(n−7), 22:6(n−3), 16:1(n−7) and 18:0. Monounsaturated fatty acids represented the dominant group in L. santolla, whereas P. granulosa contained similar amounts of mono- and polyunsaturated fatty acids. In L. santolla, monounsaturated fatty acids, especially 16:1(n−7), were preferentially utilized as compared to polyunsaturates. Due to a particularly strong lipid utilization in P. granulosa, all individual fatty acids were largely depleted at metamorphosis, showing similar extents of consumption. L. santolla had higher initial lipid and protein stores that seem to be used more economically as compared to P. granulosa.

Full text

Lipid, fatty acid and protein utilization during
lecithotrophic larval development of
Lithodes santolla (Molina) and
Paralomis granulosa (Jacquinot)
Gerhard Kattner
a,
*, Martin Graeve
a
, Javier A. Calcagno
b
,
Gustavo A. Lovrich
c
, Sven Thatje
a
, Klaus Anger
d
a
Alfred-Wegener-Institut fu
¨r Polar-und Meeresforschung, Am Handelshafen 12, 27570 Bremerhaven, Germany
b
Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina
c
Centro Austral de Investigaciones Cientı
´ficas (CADIC), Ushuaia, Argentina
d
Biologische Anstalt Helgoland, Stiftung Alfred-Wegener-Institut fu
¨r Polar-und Meeresforschung,
27498 Helgoland, Germany
Received 7 November 2002; received in revised form 28 January 2003; accepted 13 March 2003
Abstract
During the larval development of the subantarctic king crab, Lithodes santolla, and stone crab,
Paralomis granulosa, we compared changes in the carbon, fatty acid and protein contents of larvae
reared under constant conditions from hatching to metamorphosis, either in presence or absence of
food (Artemia spp. nauplii). In both species the feeding condition had no influence on any of the
chemical parameters studied, indicating a fully lecithotrophic (i.e. non-feeding) mode of
development from hatching of the first zoea to metamorphosis of the late megalopa. Dry mass
and carbon contents at hatching were similar in the larvae of both species, but L. santolla contained
initially higher total amounts of fatty acids and protein than P. granulosa. Both species utilized
considerable portions of their total fatty acid pool which decreased logarithmically throughout the
time of development. At metamorphosis, it was almost exhausted in P. granulosa, while L. santolla
had consumed only about 60%. Protein utilization, in contrast, was higher in L. santolla (40%) than
in P. granulosa (20%). Triacylglycerol was the principal storage lipid in both species, accounting
initially for about 75% of the lipid fraction; it was strongly utilized during larval development.
Phospholipid constituted the second largest lipid class; it also decreased in P. granulosa, but to a
lesser extent in L. santolla. The major fatty acids of both species were 18:1(n9), 20:5(n3) and
0022-0981/03/$ - see front matter D2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0022-0981(03)00143-6
* Corresponding author. Tel.: +49-471-4831-1490; fax: +49-471-4831-1425.
E-mail address: gkattner@awi-bremerhaven.de (G. Kattner).
www.elsevier.com/locate/jembe
Journal of Experimental Marine Biology and Ecology
292 (2003) 61 – 74

16:0 as well as, in lower proportions, 18:1(n7), 22:6(n3), 16:1(n7) and 18:0.
Monounsaturated fatty acids represented the dominant group in L. santolla, whereas P. granulosa
contained similar amounts of mono- and polyunsaturated fatty acids. In L. santolla, mono-
unsaturated fatty acids, especially 16:1(n7), were preferentially utilized as compared to
polyunsaturates. Due to a particularly strong lipid utilization in P. granulosa, all individual fatty
acids were largely depleted at metamorphosis, showing similar extents of consumption. L. santolla
had higher initial lipid and protein stores that seem to be used more economically as compared to
P. granulosa.
D2003 Elsevier Science B.V. All rights reserved.
Keywords: Lipids; Fatty acids; Protein; Lecithotrophy; Larval development; Lithodidae
1. Introduction
Marine organisms living in areas with scarce or highly variable food supply, for
instance in high latitudes, are especially dependent on the storage of large lipid reserves
(e.g. Clarke, 1983, 1993; Kattner and Hagen, 1995). Lipids are equally important during
critical non-feeding developmental periods such as embryogenesis and metamorphosis
(e.g. Anger et al., 1989; Kattner et al., 1994; Petersen and Anger, 1997). In subpolar
regions, where productivity occurs in short seasonal pulses (see Knox, 1994), some
lithodid crab species have evolved a non-feeding mode of larval development as a special
adaptation to strong seasonality of planktonic food availability (Anger, 1996; Shirley and
Zhou, 1997; Lovrich et al., 2003; Calcagno et al., in press).
Although extensive investigations have been carried out on the biology of commer-
cially exploited lithodids (for reviews, see Dawson, 1989; Lovrich, 1997), little is
generally known about the biochemical composition of their early life-history stages,
particularly about the energy source of food-independent larvae. Detailed information on
developmental changes in the elemental composition (carbon, nitrogen) of lecithotrophic
larvae and feeding juveniles are available only for the northern stone crab, Lithodes maja
(Anger, 1996). Similar studies were performed recently for the subantarctic congener
Lithodes santolla (Lovrich et al., 2003) and another southern lithodid crab, Paralomis
granulosa (Calcagno et al., in press). In all these species, the larvae are able to develop
successfully from hatching to metamorphosis in complete absence of food; the only major
difference is that P. granulosa passes through only two zoeal stages, while Lithodes spp.
have three, in all cases followed by a benthic megalopal stage (see McLaughlin et al.,
2001, 2003).
Although lipid deposits are generally the major energy source during non-feeding
developmental periods in decapods (e.g. Herring, 1974; Holland, 1978; Amsler and
George, 1984; Kattner et al., 1994; Wehrtmann and Kattner, 1998), practically nothing is
known about the lipid content and composition of lecithotrophic lithodid crab larvae.
Inferring from elemental and protein data, Anger (1996) estimated that lipids comprised
about 30% of the dry mass of freshly hatched larvae of L. maja, and that about one half
of these lipid reserves were catabolized during non-feeding development to metamor-
phosis.
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7462

In the present study, we measured changes in the biochemical composition of the larvae
of L. santolla and P. granulosa, both of which are commercially exploited and
economically important lithodid crab species in the Magellan region of southern Argentina
and Chile, and along the Chilean coast (Lovrich, 1997). Our major aims were (1) to
compare biochemical changes in larvae reared in presence or absence of Artemia spp.
nauplii in order to detect evidence for a possible ingestion and utilization of available food
(facultative lecithotrophy), alternatively, lack of response to different feeding conditions
should indicate a fully non-feeding mode of development (obligatory lecithotrophy); (2) to
evaluate the utilization of internally stored lipids and proteins during lecithotrophic
development. Special attention was paid to changes in the composition of lipids to
elucidate possible preferences in the utilization of different lipid classes and individual
fatty acids.
2. Materials and methods
The capture and maintenance of L. santolla and P. granulosa as well as the rearing of
their larvae were described in detail by Lovrich et al. (2003) and Calcagno et al. (in press).
Briefly, ovigerous females were collected in April 2000 in the Beagle Channel (southern
Argentina), kept in flow-through seawater aquaria, and transported with the German
Research Icebreaker ‘‘Polarstern’’ to the marine biological laboratory at Helgoland
(Germany). Actively swimming larvae (freshly hatched, all from the same female) were
randomly selected, and subsequently one individual was reared in 100-ml bowls under
constant conditions of temperature (6 jC), salinity (32), and a 12:12 h light/dark cycle. In
two different experimental treatments, the larvae were reared either with or without
addition of food (Artemia spp. nauplii; Argent Chemical Laboratories, USA). Water and
(where applicable) food were changed daily, and the larvae were checked for moults or
mortality.
Samples of larvae were collected for elemental and biochemical analyses immediately
after hatching and later after different intervals of development (see Fig. 1). Dry mass and
carbon determinations were performed with an autobalance (Mettler, UMT 2) and a Fisons
Carlo Erba 1108 Elemental Analyser, respectively (for details, see Anger and Harms,
1990). Five measurements of one larva each were performed for each interval.
Protein measurements were made in triplicate samples with one individual each,
using the Lowry method (Lowry et al., 1951) with bovine serum albumin as a standard.
The protein data were converted to values of protein-bound carbon applying a factor of
0.5, which is a realistic estimate based on the molecular structure of proteins and amino
acids.
Due to the various chemical determinations carried out during the larval development,
only one to three larvae per sample were available for lipid analyses. The larvae were
homogenized and extracted in dichloromethane/methanol (2:1; v/v) according to Folch et
al. (1957). The lipid class compositions were determined by HPTLC and scanning
densitometry (CAMAG TLC scanner III) after Olsen and Henderson (1989). Different
commercial standard mixtures and lipid extracts from marine organisms were used for
identification (Hagen, 1988).
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–74 63

The fatty acid compositions were determined by gas chromatography (Kattner and
Fricke, 1986). Briefly, fatty acids were converted to methyl esters by transesterification in
methanol containing 3% concentrated sulphuric acid at 80 jC for 4 h. Fatty acid methyl
esters were then analyzed with a gas liquid chromatograph (Chrompack 9000) on a 30
m0.25 mm i.d. wall-coated open tubular column (film thickness: 0.25 Am; liquid phase:
DB-FFAP) using temperature programming. Fatty acids were identified with standard
mixtures and quantified by internal standard. If necessary, additional confirmation was
carried out by GC-MS (Kattner et al., 1998).
Due to the small number of specimens available for lipid analysis, the determination of
total lipids was not possible. To estimate the total lipid fraction, the mass of individual
fatty acids was summed, which is a slight underestimation. Fatty acid carbon was
calculated according to the molecular structure of fatty acids which consist on average
of 75% carbon.
Fig. 1. L. santolla and P. granulosa. Changes in biochemical composition during lecithotrophic larval
development; all data are converted to Ag carbon (C) per larva. Carbon content with exponential fit for L. santolla
(y= 595.43e 0.0134x) and logarithmic fit for P. granulosa (y=68.038ln(x) + 572.05); total lipid with
logarithmic fit for L. santolla (y=39.967ln(x) + 271.55) and P. granulosa (y=43.535ln(x) + 205.85); protein
with exponential fit for L. santolla (y= 168.13e 0.0091x) and P. granulosa (y= 124.96e 0.0045x);
triacylglycerols with logarithmic fit for L. santolla (y=34.083ln(x) + 204.92) and P. granulosa
(y=32.709ln(x) + 156.36); phospholipids with exponential fit for L. santolla (y= 39.348e 0.0081x) and
logarithmic fit for P. granulosa (y=9.3307ln(x) + 41.019).
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7464

3. Results
The larval development of L. santolla was monitored for a period of 62 days after
hatching. The first juvenile stage was collected on day 63. Samples of P. granulosa were
available only until day 52, although larval development averages about 61 days
(Calcagno et al., in press). Hence, biomass losses occurring during the final ca. 9 days
of megalopal development could not be measured, implying that our data underestimate
the actual losses from hatching to metamorphosis. When values for fed and unfed larvae
were compared in each species, no significant differences were found in dry mass,
elemental composition, protein content, or in quantity or composition of the lipid fraction
(two-way ANOVA with time of development and treatment as factors). While our results
showed in both species a highly significant influence of development time, there was no
significant effect of presence or absence of food, nor was there an interaction between the
factors time and feeding condition. Thus, there was no indication of ingestion and
utilization of food, i.e. the larvae of both species did not eat, regardless of food availability.
Therefore, the data from both treatments were subsequently pooled for each species to
analyze the effect of development time on the biochemical composition of the lithodid
larvae.
3.1. Lithodes santolla
During the larval development of L. santolla from the freshly hatched zoea I to the
late megalopa (62 days), dry mass decreased by 28% from about 1040 to 750 Ag per
individual and then dropped sharply to 580 Ag in the first juvenile stage, which is a
reduction of 16%. This decline in dry mass towards the juvenile was due to the loss of
the exuviae, which is highest at metamorphosis (Lovrich et al., 2003). The utilization of
total carbon, protein and fatty acids during the time of larval development is shown in
Fig. 1; for more immediate comparison, all biochemical data are converted to carbon
(C) values. Total C amounted initially to about one half of larval dry mass. It decreased
from about 550 Ag in the zoea I to 270 Ag in the late megalopa, corresponding to a loss
of 53%. The loss in total C towards the juvenile from 270 to 220 Ag was less
pronounced than in dry mass resulting in a total loss of C from zoea I to juvenile of
about 61%. Overall, the total C content was more reduced than dry mass. This decrease
was best described as an exponential function of the time of development (r
2
= 0.9561;
Fig. 1).
The protein content decreased from about 350 Ag per larva in the zoea I to 200 Ag in the
late megalopal and first juvenile stage, representing a loss of about 40%. The initial
proportion of protein-C made up on average 15% of dry mass. The proportion of protein-C
made up on average 33% of total C during development to the late megalopal stage, but in
the juvenile stage the proportion was higher comprising 45% of total C. The decrease in
protein-C with developmental time was similar to that of total C decreasing exponentially
(r
2
= 0.7998), but this pattern was here closer to a linear (r
2
= 0.7799) than to a logarithmic
relationship (r
2
= 0.7222) (Fig. 1).
Total fatty acid C comprised about 24% of dry mass and 44% of total C immediately
after hatching. The total fatty acid content per individual decreased from about 250 to 100
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–74 65

Ag C (loss of about 60%) during development from hatching to the late megalopa without
further decrease to the juvenile. The loss of lipid-C was more pronounced during the zoeal
phase than in the megalopa. This decrease with time was best reflected by a logarithmic
(r
2
= 0.8926) or exponentially fit (r
2
= 0.8519) (Fig. 1).
The major lipid classes of L. santolla were triacylglycerols (neutral lipids) and
phospholipids (polar lipids), which together comprised about 85% of total lipid. Triacyl-
glycerols alone contributed initially ca. 70 – 75% to the total lipid fraction, but this
proportion decreased during development to the late megalopa and early juvenile stage to
about 55–60%. The amounts of triacylglycerol-C per individual decreased logarithmically
from 180 Ag at hatching to about 60 Ag C in the late megalopa and freshly metamorphosed
juvenile (r
2
= 0.9163), which is a loss of 67%. The phospholipid fraction decreased to a
lesser extent, from about 45 to 26 Ag C per larva (i.e. by 42%) with only low statistical
significance (Fig. 1).
The major fatty acids of L. santolla were 18:1(n9) and 20:5(n3), comprising up to
30% and 28% of total fatty acids, respectively, followed by 16:0, 18:1(n7), 22:6(n3),
16:1(n7) and 18:0. Other fatty acids occurred only in small amounts (Table 1).
Table 1
Lithodes santolla
Fatty acids Zoea
(days 0 – 20)
Megalopa (woA)
(days 22, 32, 42)
Megalopa (wA)
(days 32, 42)
Megalopa (woA)
(days 56, 62)
Megalopa (wA)
(days 52, 62)
Juvenile
(days 63, 65)
Mean FS.D.
(7)
Mean FS.D.
(3)
Mean FS.D.
(2)
Mean FS.D.
(2)
Mean FS.D.
(2)
Mean FS.D.
(2)
14:0 0.9 F0.3 0.5 F0.4 0.8 F0.0 0.3 F0.0 0.1 F0.1 0.3F0.1
16:0 15.7 F0.6 16.6 F0.6 16.0 F0.2 12.2 F0.4 12.5 F0.2 14.1F0.1
16:1(n7) 5.4 F0.5 5.0 F0.7 5.1 F0.1 2.4 F0.5 2.4 F0.4 2.2F0.4
16:2 0.4 F0.2 0.2 F0.3 0.2 F0.3 0.1 F0.1 0.1 F0.1 –
16:3 0.5 F0.1 0.2 F0.4 0.3 F0.5 0.2 F0.0 0.2 F0.0 0.2F0.0
16:4 0.1 F0.2 – – – – –
18:0 5.1 F1.6 5.0 F0.7 4.8 F0.3 4.6 F0.2 5.2 F0.5 6.4 F0.7
18:1(n9) 28.4 F1.5 29.7 F0.9 30.5 F0.6 24.7 F1.4 26.4F0.7 26.0 F1.3
18:1(n7) 11.0 F1.0 12.5 F0.3 12.2 F0.5 10.1 F0.3 10.9 F0.3 10.7 F0.7
18:2(n6) 1.3 F0.9 1.1 F0.1 1.1 F0.1 1.1 F0.2 0.8 F0.0 1.4 F0.7
18:3(n3) 0.5 F0.1 0.9 F0.3 0.3 F0.5 0.3 F0.0 0.3 F0.0 0.3 F0.1
18:4(n3) 0.6 F0.2 0.3 F0.3 0.3 F0.4 0.2 F0.1 0.2 F0.1 0.2 F0.0
20:1(n9) 1.7 F0.1 1.8 F0.1 2.0 F0.4 1.6 F0.1 1.4 F0.1 1.4 F0.3
20:1(n7) 1.1 F0.2 1.2 F0.0 1.6 F0.4 0.9 F0.0 1.1 F0.1 1.1 F0.1
20:4(n6) 0.8 F1.1 0.1 F0.2 0.2 F0.3 2.3 F0.1 2.5 F0.1 2.3 F0.1
20:4(n3) 0.8 F0.1 0.5 F0.4 0.3 F0.5 0.6 F0.1 0.6 F0.1 0.5 F0.1
20:5(n3) 17.9 F1.9 16.7 F0.7 17.4 F0.9 28.2 F2.6 23.9 F1.0 21.9 F1.8
22:5(n3) 0.3 F0.7 – – 1.0 F0.1 1.1 F0.0 1.1 F0.0
22:6(n3) 6.5 F0.7 6.3 F0.2 5.8 F1.2 8.4 F0.2 9.4 F0.5 8.7 F0.5
SAT 22.0 F2.3 22.1 F0.9 21.6 F0.1 17.2 F0.2 18.1 F0.2 21.0 F0.4
MUFA 48.3 F2.2 51.5 F0.4 52.5 F1.9 39.7 F2.2 42.1 F0.7 40.6 F0.9
PUFA 29.8 F1.9 26.4 F0.7 26.0 F2.1 43.1 F2.5 39.8 F0.9 37.3 F2.0
Fatty acid compositions (mass% of total fatty acids) of larval stages and first juvenile, fed without (woA) and with
(wA) Artemia nauplii presented as mean and standard deviation (S.D.); saturated (SAT), monounsaturated
(MUFA) and polyunsaturated (PUFA) fatty acids; (n) number of samples.
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7466

Monounsaturated fatty acids were the major group, mostly due to the dominance of
18:1(n9). Zoea and early megalopal larvae of L. santolla had a similar composition. A
clear change, namely an increasing dominance of polyunsaturated fatty acids, became
visible in the late megalopa and juvenile. The relative increase in 20:5(n3) and
22:6(n3) was probably due to the proportional increase of remaining phospholipids
in relation to triacylglycerols in the older stages. While total fatty acids decreased by about
60%, the losses varied among individual fatty acids. Monounsaturated fatty acids were
preferably utilized, especially 16:1(n7), decreasing by >80%, followed by the two 18:1
isomers. Only about 22% of the polyunsaturated fatty acid 20:5(n3) and 30% of
22:6(n3) were utilized from hatching to the late megalopa; including the development to
juvenile the loss was 40% for 20:5(n3) and 30% for 22:6(n3). Since no differences
were found between fed and unfed larvae, the data from the two treatments could be
pooled (Fig. 2).
3.2. Paralamis granulosa
Total dry mass and carbon content of the first larval stage of P. granulosa were slightly
lower than in L. santolla. Also the decrease in dry mass observed during larval develop-
ment from hatching to the late megalopa was weaker in P. granulosa (from 985 to 764 Ag
per individual, corresponding to a loss of 22%). The ratio of total C to dry mass was
roughly the same in both species.
The losses of total C, protein and total fatty acids are presented in Fig. 1. Nearly half of
the C initially present was lost during larval development of P. granulosa, comparable to
L. santolla, decreasing from about 540 to 280 Ag C per individual. This decrease can be fit
Fig. 2. L. santolla and P. granulosa. Fatty acid utilization during lecithotrophic larval development, expressed as
percentage decrease of major individual fatty acids (mean data from experiments with fed and unfed larvae).
Decrease was calculated from fatty acid amount of zoea after hatching (day 0) and latest megalopa analyzed (day
62 for L. santolla and day 52 for P. granulosa) (open bars). In addition, the same calculation is presented until the
juvenile stage of L. santolla (gray bars).
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–74 67

almost equally well using either a logarithmic (r
2
= 0.9089) or an exponential function
(r
2
= 0.8997).
The protein content per individual decreased moderately from hatching to the late
megalopal stage, from 260 to 200 Ag (or from 128 to 100 Ag protein-C), representing a
loss of about 22%. This decrease follows an exponential (r
2
= 0.7920) or a linear trend
(r
2
= 0.7673), comparable with the pattern observed in L. santolla (Fig. 1). The proportion
of protein-C within total dry mass remained almost constant at 13.4%, whereas the
percentage of protein-C to total C increased from 24% in the zoea I to 35% in the
megalopa. This differs conspicuously from L. santolla, where the percentage of protein-C
remained almost constant.
The fraction of total fatty acid C within dry mass or total C in hatching P. granulosa
was lower than in L. santolla, accounting for about 20% and 30%, respectively. These
proportions decreased to about 5% and 12% in the late megalopa. The utilization of lipids
was higher in P. granulosa than in L. santolla. From the first zoeal to the late megalopal
stage, the content of total fatty acid C declined from 270 to about 50 Ag C per larva, or by
about 81%. A major part of the decrease in total fatty acids occurred during the first 10 –15
days of development. This pattern could be described with a logarithmic (r
2
= 0.9200) or
an exponential fit (r
2
= 0.8947). Very similar patterns were found also in total C, showing
that these two parameters are closely linked.
Table 2
Paralomis granulosa
Fatty acids Zoea Megalopa (woA) Megalopa (wA)
Mean FS.D. (5) Mean FS.D. (3) Mean FS.D. (2)
14:0 1.3 F0.1 1.3 F0.1 1.1 F0.1
16:0 15.7 F0.6 15.3 F0.9 15.5 F2.4
16:1(n7) 5.5 F0.4 4.1 F1.7 4.0 F0.2
16:2 0.9 F0.3 1.5 F0.6 1.1 F0.0
16:3 0.8 F0.1 0.7 F0.1 0.9 F0.3
16:4 0.5 F0.0 0.9 F0.2 0.4 F0.5
18:0 5.1 F1.5 5.7 F1.0 6.2 F0.2
18:1(n9) 19.6 F0.8 19.8 F0.2 20.5 F0.9
18:1(n7) 9.5 F0.2 9.3 F0.3 10.4 F0.2
18:2(n6) 1.3 F0.1 1.5 F0.2 2.0 F0.8
18:3(n3) 0.6 F0.0 0.5 F0.0 0.6 F0.2
18:4(n3) 0.5 F0.1 0.4 F0.1 0.5 F0.4
20:1(n9) 1.1 F0.1 1.5 F0.3 1.8 F0.4
20:1(n7) 0.8 F0.0 1.0 F0.2 1.0 F0.1
20:4(n6) 5.9 F0.2 6.7 F0.6 3.7 F4.2
20:4(n3) 0.6 F0.2 0.6 F0.2 0.6 F0.2
20:5(n3) 15.1 F0.5 14.8 F0.4 16.4 F3.1
22:5(n3) 2.0 F0.1 2.2 F0.3 1.1 F1.6
22:6(n3) 11.9 F0.7 11.3 F0.3 11.0 F1.7
SAT 22.9 F2.0 23.0 F1.1 23.5 F2.8
MUFA 36.9 F1.2 35.8 F1.9 38.2 F1.9
PUFA 40.2 F1.3 41.2 F1.8 38.3 F4.7
Fatty acid composition (mass% of total fatty acids) of larval stages, fed without (woA) and with (wA) Artemia
nauplii presented as mean and standard deviation (S.D.); for abbreviations refer to Table 1.
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7468

As in L. santolla, triacylglycerols also were the major lipid class in P. granulosa,
contributing on average 75% to total lipids, while the proportion of phospholipids was
only about 17%. Triacylglycerols were utilized by more than 80%, decreasing during
larval development from about 200 to 32 Ag per individual. The same trend was observed
in phospholipids. This is different from L. santolla where proportionally less triacylgly-
cerols and only small amounts of phospholipids were utilized. The losses in triacylglycer-
ols and phospholipids were best fitted by exponential (r
2
= 0.9509 and r
2
= 0.8737,
respectively) or logarithmic functions (r
2
= 0.9462 and r
2
= 0.8307) (Fig. 1).
The same principal fatty acids as in L. santolla occurred also in P. granulosa, but with
different proportions. Mono- and polyunsaturated fatty acids contributed 36% and 41%,
respectively, while only 23% of the total fatty acid pool were saturates. The major fatty
acids were 18:1(n9), 16:0 and 20:5(n3), occurring in similar quantities each
comprising up to 20% of total fatty acids. 22:6(n3), 18:1(n7), 16:1(n7) and, in
contrast to L. santolla, also 20:4(n6) occurred in considerable amounts (up to 7%). No
significant changes in the fatty acid composition were observed during development or
between fed and unfed larvae (Table 2). Due to the strong utilization of lipids during larval
development, all fatty acids were in this species considerably depleted. When metamor-
phosis was approached, 16:1(n7) was almost exhausted, while polyunsaturates had been
utilized to a slightly lesser extent (Fig. 2).
4. Discussion
Experimental evidence of food-independent larval development (lecithotrophy) has
been shown for several lithodid crab species, namely L. maja (Anger, 1996),L.
aequispinus (Shirley and Zhou, 1997),L. santolla (McLaughlin et al., 2001; Lovrich et
al., 2003), and P. granulosa (McLaughlin et al., 2003; Calcagno et al., in press). This
reproductive trait, which is quite unusual among the marine decapod crustaceans (for
review, see Gore, 1985; Rabalais and Gore, 1985; Anger, 2001), is based on an enhanced
female energy allocation per offspring, i.e. the production of large yolky eggs. Since most
examples of lecithotrophy have been observed in subpolar regions, this trait has been
interpreted as an adaptation to strong seasonality in planktonic food availability with only
short pulses of productivity in summer (for recent discussion, see Anger et al., 2003;
Thatje et al., 2003). Hence, lecithotrophic larvae may be expected to occur also in other
lithodid species as well as in other decapods living in high latitudes (Anger, 1996, 2001).
While this generalization is supported by large egg size frequently observed in subpolar
species, comparative experimental evidence of lecithotrophic life histories is largely
missing.
Our results on lipid and fatty acid composition of the early life-history stages of L.
santolla and P. granulosa provide evidence for lecithotrophy throughout larval develop-
ment from hatching to metamorphosis. This food-independent mode of development is
based on unusually high initial organic reserves, mainly on enhanced lipid stores
remaining from the egg yolk (for recent review and comparison with planktotrophic
decapod larvae, see Anger, 2001). Since food availability did not result in an increase (or
in a lower rate of decrease) in dry mass, total carbon, or in the fatty acid content, and no
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–74 69

differences in the fatty acid composition of fed and unfed larvae were found, our data
indicate consistently that the larvae of both species are fully non-feeding (obligatory
lecithotrophy). Although the last few days prior to metamorphosis in P. granulosa were
not covered, which could allow for a slight doubt as to the degree of lecithotrophy during
the final phase of larval development in this species, observations of larval mortality,
development duration and mouthpart morphology (McLaughlin et al., 2003) as well as
more complete data of their elemental composition (total C, N, C/N ratio; Calcagno et al.,
in press) support our conclusion that all larval stages are non-feeding. First food uptake
and growth have consistently been observed after metamorphosis, i.e. in the first juvenile
crab stage (Lovrich et al., 2003; Calcagno et al., in press).
Studies of the elemental composition of lithodid larvae suggested that lipid reserves
remaining from the egg yolk should provide the principal energy source for non-feeding
development (Anger, 1996). Our study provides now the first direct biochemical evidence
for this inferred assumption. Already for embryogenesis a considerable amount of the
initially available egg lipids is invested. As known from most planktotrophic marine
decapod species utilization rates vary from about 25 to 60% of total lipid among species
from different climatic zones (Anger and Harms, 1990; Wehrtmann and Kattner, 1998;
Wehrtmann and Graeve, 1998; Anger, 2001; Graeve and Wehrtmann, 2003; Morais et al.,
2002). In the eggs of L. maja, about one third of the initial total C content is utilized during
embryogenesis (Anger, 1996). Thus, species with a lecithotrophic larval development have
to furnish their eggs with an enhanced lipid pool which may comprise at least 40% of dry
mass to cope with the extended period of non-feeding development. Comparable large
eggs contain up to 40% of lipid (Herring, 1974).
Although size, dry mass and C contents were similar in freshly hatched zoeal larvae of
L. santolla and P. granulosa, the initial lipid content was about 20% higher in L. santolla.
Its zoeal development is comprised of three stages lasting together about 22 days (at
constant 6 jC), while the megalopal stage alone requires ca. 40 days (Lovrich et al., 2003).
In P. granulosa, by contrast, there are only two zoeal stages which require in total about 16
days, followed by a megalopa which requires another ca. 45 days to reach metamorphosis
(Calcagno et al., in press). The pattern of lipid utilization during the time of development
may be described as a logarithmic decrease in total fatty acids. According to the fitted
equation, the three zoeal stages of L. santolla combined utilized within 22 days about 46%
of the total fatty acid pool that was present at hatching, whereas the megalopa used only
15% of the lipids within the following 40 days. For P. granulosa, the same calculation
results in a utilization rate of 59% for the zoeae (with only two stages and 16 days of
development), while another 28% was utilized during the 45 days of development through
the megalopal stage. These estimates are very similar to the losses that were measured
directly during the period from hatching to metamorphosis. According to these data, 60%
of the total fatty acid pool were consumed in L. santolla and 81% in P. granulosa (in the
latter species measured only until day 52, suggesting that yet higher losses may occur until
metamorphosis). The differences in the percentage utilization of total fatty acids are partly
due to the differences in the initial amounts of fatty acids in the first zoeal stages, which
were lower in P. granulosa than in L. santolla.
As in the total lipids fraction, the amount of proteins measured at hatching was lower in
P. granulosa than in L. santolla. However, L. santolla utilized during its larval develop-
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7470

ment twice as much of its initially available protein (about 40% vs. 20%), coinciding with
a lower utilization of lipids. A nearly linear pattern of protein degradation shows that this
fraction was not preferentially utilized during zoeal development. Hence, L. santolla
appears to utilize the two principal biochemical fractions of its biomass in a more balanced
fashion as energy sources, which makes it less dependent on lipids than P. granulosa.On
the other hand, freshly hatched L. santolla larvae contained more lipids than P. granulosa.
In the latter species, lipids were almost exhausted shortly before metamorphosis, while
proteins had hardly been used. This difference is confirmed also by measurements of the
C/N ratio at metamorphosis, which was lower in P. granulosa than in L. santolla.
Triacylglycerols, which are generally considered as the principal storage lipid,
constituted the major lipid class in both species, whereas phospholipids, which are typical
membrane lipids, were less abundant. The utilization of lipids was thus closely related to
that of triacylglycerols. As shown also for total fatty acids, these two lipid classes were
utilized less by L. santolla than by P. granulosa. In the latter, phospholipids also decreased
logarithmically during larval development. A high proportion of triacylglycerols, which is
the most important energy source, may thus be typical of lecithotrophic larvae, whereas
the lipids in the eggs and larvae of other decapod species are often dominated by
phospholipids (Clarke, 1977, 1979, 1993; Kattner et al., 1994; Wehrtmann and Graeve,
1998; Graeve and Wehrtmann, 2003).
The fatty acid composition of the larval stages of L. santolla and P. granulosa are not
essentially different from that in other decapod larvae (e.g. Levine and Sulkin, 1984;
Clarke et al., 1990; Kattner et al., 1994; Graeve et al., 1997). Higher proportions of
monounsaturates in L. santolla are probably due to the higher amounts of triacylglycerols.
Due to stronger utilization of lipids by P. granulosa as compared to L. santolla,
interspecific differences occurred in the overall consumption of individual fatty acids.
In both species, monounsaturated fatty acids were preferentially catabolized during larval
development, with strongest utilization of 16:1(n7) followed by degradation of the 18:1
isomers. A preferential catabolism of monounsaturates may be generalized, as other
investigations on fatty acid utilization during embryogenic and larval development of
decapods show consistently the same pattern (Rainuzzo et al., 1997; Wehrtmann and
Graeve, 1998; Wehrtmann and Kattner, 1998; Morais et al., 2002).InP. granulosa the
polyunsaturated fatty acids 20:5(n3) and 22:6(n3) were largely depleted when
metamorphosis was imminent, indicating the crucial importance of immediate feeding
in the first juvenile crab stage. In L. santolla, however, the amount of polyunsaturates
decreased only slightly during development to the late megalopa and through metamor-
phosis to the first juvenile stage, so that their proportions within the total fatty acid pool
increased. In general, polyunsaturates are essential components in marine organisms,
because they play the central role in the buildup and maintenance of membranes, requiring
that they are used less for energetic demands but conserved for membrane differentiation.
Since P. granulosa has lower initial lipid reserves than L. santolla and, moreover, it is
apparently less capable of utilizing its protein stores for additional energy production, this
species appears to approach more closely its ultimate limits of lecithotrophy.
In summary, we found clear interspecific differences in the utilization patterns of lipids
and proteins of two subantarctic lithodid species with fully lecithotrophic larvae. As a
major difference in their life cycles, these species show different degrees of abbreviation of
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–74 71

the zoeal phase (two stages in P. granulosa, three in L. santolla). As a consequence of
longer stage durations in P. granulosa, however, the total period of larval development
from hatching to the first juvenile crab stage is roughly the same in both species.
Generally, the utilization of lipids was much more pronounced during the actively
swimming zoeal phase than in the benthic megalopa. In L. maja, which shows similar
behavioral changes during its larval development, this change in larval activity was clearly
reflected by high rates of oxygen consumption in the zoeae followed by low metabolic
rates in the megalopa (Anger, 1996).
L. santolla contained higher initial amounts of storage compounds, which were used
more economically. Namely, a stronger use of protein reserves allowed for spare use of
lipids. In P. granulosa, on the other hand, lipids were nearly depleted shortly before
metamorphosis. Hence, the eggs and early larvae of L. santolla appear to be better
prepared for non-feeding development than those of P. granulosa, and the first juvenile
crab stage of L. santolla may thus depend less on benthic food availability. However, the
initial furnishing of eggs with lipids may be intraspecifically highly variable, depending on
season, population, size, age or physiological condition (feeding status) of the female and,
possibly, other internal and/or environmental factors. This is suggested by observations of
significant variability in the carbon content of eggs and larvae produced by different
females of L. maja (Anger, 1996). The extent and the consequences of intraspecific
variability in egg size and biochemistry of L. santolla and P. granulosa as well as in other
possibly lecithotrophic lithodid species living in high latitudes remains to be further
elucidated.
Acknowledgements
We would like to thank C. Harms, M. Bo¨er, A. Kaffenberger and C. Pu
¨schel for
analytical work. We greatly appreciate the help of the crew of PFS ‘‘Polarstern’’ during the
transport of live crabs. E. Heyer and R. Hottung helped in maintaining larval cultures.
Javier Calcagno is grateful to the German Academic Exchange Service (DAAD) for
funding his research visit to Helgoland. This project was partially funded by the
International Bureau of the German Ministry of Scientific Research (BMBF, project No.
ARG 99/002), and the Argentine Secretarı
´a Nacional para la Tecnologı
´a, Ciencia e
Inovacio
´n Productiva (SETCIP). [SS]
References
Amsler, M.O., George, R.Y., 1984. Seasonal variation in the biochemical composition of the embryos of
Callinectes sapidus Rathbun. J. Crustac. Biol. 4, 546 – 553.
Anger, K., 1996. Physiological and biochemical changes during lecithotrophic larval development and early
juvenile growth in the northern stone crab, Lithodes maja (Decapoda: Anomura). Mar. Biol. 126, 283 – 296.
Anger, K., 2001. The biology of decapod crustacean larvae. A.A. Balkema Publishers, Sweets and Zeitlinger BV,
Lisse, ISSN 0168-6456. Crustacean Issues 14, pp. 1– 420.
Anger, K., Harms, J., 1990. Elemental (CHN) and proximate biochemical composition of decapod crustacean
larvae. Comp. Biochem. Physiol. 97B, 69 – 80.
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7472

Anger, K., Harms, J., Pu
¨schel, C., Seeger, B., 1989. Physiological and biochemical changes during larval
development of a brachyuran crab reared under constant conditions in the laboratory. Helgol. Meeresunters.
43, 225 – 244.
Anger, K., Thatje, S., Lovrich, G.A., Calcagno, J.A., 2003. Larval and early juvenile development of Paralomis
granulosa reared at different temperatures: tolerance of cold and food limitation in a lithodid crab from high
latitudes. Mar. Ecol. Prog. Ser. (in press).
Calcagno, J.A., Thatje, S., Anger, K., Lovrich, G.A., Kaffenberger, A., 2003. Changes in biomass and chemical
composition during lecithotrophic larval development of the southern stone crab, Paralomis granulosa
(Jacquinot). Mar. Ecol. Prog. Ser. (in press).
Clarke, A., 1977. Lipid class and fatty acid composition of Chorismus antarcticus (Pfeffer) (Crustacea: Deca-
poda) at South Georgia. J. Exp. Mar. Biol. Ecol. 28, 297 – 314.
Clarke, A., 1979. Lipid content and composition of the pink shrimp Pandalus montagui (Leach) (Crustacea:
Decapoda) at South Georgia. J. Exp. Mar. Biol. Ecol. 38, 1 – 17.
Clarke, A., 1983. Life in cold water: the physiological ecology of polar marine ectotherms. Oceanogr. Mar. Biol.
Ann. Rev. 21, 341 – 453.
Clarke, A., 1993. Egg size and egg composition in polar shrimps (Caridea: Decapoda). J. Exp. Mar. Biol. Ecol.
168, 189 – 203.
Clarke, A., Brown, J.H., Holmes, L.J., 1990. The biochemical composition of eggs from Macrobrachium rosen-
bergii in relation to embryonic development. Comp. Biochem. Physiol. 96, 505 – 511.
Dawson, E.W., 1989. King crabs of the world (Crustacea: Lithodidae) and their fisheries: a comprehensive
bibliography. Misc. Publ. 101. New Zealand Oceanogr. Inst., Div. Water Sci., DSIR, Wellington.
Folch, J., Lees, M., Sloane-Stanley, G.H., 1957. A simple method for the isolation and purification of total lipides
from animal tissues. J. Biol. Chem. 226, 497 – 509.
Gore, R.H., 1985. Molting and growth in decapod larvae. In: Wenner, A.M. (Ed.), Larval Growth. Crustacean
Issues, vol. 2. Balkema, Rotterdam, pp. 1– 65.
Graeve, M., Wehrtmann, I.S., 2003. Lipid and fatty acid composition of Antarctic shrimp eggs (Decapoda:
Caridea). Polar Biol. 26, 55 – 61.
Graeve, M., Kattner, G., Piepenburg, D., 1997. Lipids in Arctic benthos: does the fatty acid and alcohol
composition reflect feeding and trophic interactions? Polar Biol. 18, 53–61.
Hagen, W., 1988. Zur Bedeutung der Lipide im antarktischen Zooplankton. Ber. Polarforsch. 49, 1 – 129 ((in
German) [English version (1989): On the significance of lipids in Antarctic zooplankton. Canadian
Translation Fish. Aquat. Sci. 5458, 1 – 149.]).
Herring, P.J., 1974. Size, density and lipid content of some decapod eggs. Deep-Sea Res. 21, 91– 94.
Holland, D.L., 1978. Lipid reserves and energy metabolism in the larvae of benthic marine invertebrates. In:
Malins, D.C., Sargent, J.R. (Eds.), Biochemical and Biophysical Perspectives in Marine Biology. Academic
Press, London, pp. 85 – 123.
Kattner, G., Fricke, H.S.G., 1986. Simple gas– liquid chromatographic method for the simultaneous determina-
tion of fatty acids and alcohols in wax esters of marine organisms. J. Chromatogr. 361, 263– 268.
Kattner, G., Hagen, W., 1995. Polar herbivorous copepods—different pathways in lipid biosynthesis. ICES J.
Mar. Sci. 52, 329 – 335.
Kattner, G., Wehrtmann, I.S., Merck, T., 1994. Interannual variations of lipids and fatty acids during larval
development of Crangon spp. in the German Bight, North Sea. Comp. Biochem. Physiol. 107B, 103–110.
Kattner, G., Hagen, W., Graeve, M., Albers, C., 1998. Exceptional lipids and fatty acids in the pteropod Clione
limacina (Gastropoda) from both polar oceans. Mar. Chem. 61, 219 – 228.
Knox, G.A., 1994. The Biology of the Southern Ocean. Cambridge Univ. Press, Cambridge, UK, pp. 1– 444.
Levine, D.M., Sulkin, S.D., 1984. Nutritional significance of long-chain polyunsaturated fatty acids to the
zoeal development of the brachyuran crab, Eurypanopeus depressus (Smith). J. Exp. Mar. Biol. Ecol. 81,
211– 223.
Lovrich, G.A., 1997. La pesquerı
´a mixta de las centollas Lithodes santolla yParalomis granulosa (Anomura:
Lithodidae) en Tierra del Fuego, Argentina. Invest. Mar. (Valparaı
´so) 25, 41 – 57.
Lovrich, G.A., Thatje, S., Calcagno, J.A., Anger, K., Kaffenberger, A., 2003. Changes in biomass and chemical
composition during lecithotrophic larval development of the southern king crab, Lithodes santolla (Molina).
J. Exp. Mar. Biol. Ecol. 288, 65 – 79.
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–74 73

Lowry, D.H., Rosenberg, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin phenol
reagent. J. Biol. Chem. 193, 265 – 275.
McLaughlin, P.A., Anger, K., Kaffenberger, A., Lovrich, G.A., 2001. Megalopal and early juvenile development
in Lithodes santolla (Molina, 1782) (Decapoda: Anomura; Paguroidea: Lithodidae), with notes on zoeal
variations. Invertebr. Reprod. Dev. 40, 53 – 67.
McLaughlin, P.A., Anger, K., Kaffenberger, A., Lovrich, G.A., 2003. Larval and early juvenile development in
Paralomis granulosa (Jacquinot) (Decapoda: Anomura: Paguroidea: Lithodidae), with emphasis on abdomi-
nal changes in megalopal and crab stages. J. Nat. Hist. (in press).
Morais, S., Narciso, L., Calado, R., Nunes, M.L., Rosa, R., 2002. Lipid dynamics during the embryonic develop-
ment of Plesionika martia martia (Decapoda; Pandalidae), Palaemon serratus and Palaemon elegans (Deca-
poda; Palaemonidae): relation to metabolic consumption. Mar. Ecol. Prog. Ser. 242, 195– 204.
Olsen, R.E., Henderson, R.J., 1989. The rapid analysis of neutral and polar marine lipids using double-develop-
ment HPTLC and scanning densitometry. J. Exp. Mar. Biol. Ecol. 129, 189 – 197.
Petersen, S., Anger, K., 1997. Chemical and physiological changes during the embryonic development of the
spider crab, Hyas araneus L. (Decapoda: Majidae). Comp. Biochem. Physiol. 117B, 299 – 306.
Rabalais, N.N., Gore, R.H., 1985. Abbreviated development in decapods. In: Wenner, A.M. (Ed.), Larval
Growth. Crustacean Issues, vol. 2. Balkema, Rotterdam, pp. 67 – 126.
Rainuzzo, J.R., Reitan, K.I., Olsen, Y., 1997. The significance of lipids at early stages of marine fish: a review.
Aquaculture 155, 103 – 116.
Shirley, T.C., Zhou, S., 1997. Lecithotrophic development of the golden king crab Lithodes aequispinus (Anom-
ura: Lithodidae). J. Crustac. Biol. 17, 207 – 216.
Thatje, S., Calcagno, J.A., Lovrich, G.A., Sartoris, F.J., Anger, K., 2003. Extended hatching rhythms in the
Subantarctic lithodid crabs Lithodes santolla and Paralomis granulosa (Crustacea: Decapoda: Lithodidae).
Helgol. Mar. Res. (in press).
Wehrtmann, I.S., Graeve, M., 1998. Lipid composition and utilization in developing eggs of two tropical marine
caridean shrimps (Decapoda: Caridea: Alpheidae, Palaemonidae). Comp. Biochem. Physiol. 121B, 457 – 463.
Wehrtmann, I.S., Kattner, G., 1998. Changes in volume, biomass, and fatty acids of developing eggs in Nauti-
caris magellanica (Decapoda: Caridea): a latitudinal comparison. J. Crustac. Biol. 18, 413 – 422.
G. Kattner et al. / J. Exp. Mar. Biol. Ecol. 292 (2003) 61–7474