Marine Biology

, Volume 149, Issue 6, pp 1417–1429

Trophodynamics and seasonal cycle of the copepod Pseudocalanus acuspes in the Central Baltic Sea (Bornholm Basin): evidence from lipid composition

Authors

    • Marine ZoologyUniversity of Bremen
  • Jasmin Renz
    • Alfred Wegener Institute for Polar and Marine Research
  • Justus van Beusekom
    • Alfred Wegener Institute for Polar and Marine Research
  • Maarten Boersma
    • Alfred Wegener Institute for Polar and Marine Research
  • Wilhelm Hagen
    • Marine ZoologyUniversity of Bremen
Research Article

DOI: 10.1007/s00227-006-0290-8

Cite this article as:
Peters, J., Renz, J., van Beusekom, J. et al. Mar Biol (2006) 149: 1417. doi:10.1007/s00227-006-0290-8

Abstract

Seasonal lipid dynamics of the copepod Pseudocalanus acuspes were studied in the Bornholm Basin (Central Baltic Sea) on a monthly basis from March 2002 until March 2003 and were interpreted in light of life cycle strategies and diet selection. The individual total lipid content of females ranged from 0.9 to 1.8 μg, with relative wax ester content reaching a significant maximum in May (44% of total lipids) and minimum (17% of total lipids) in April and November. Significant changes in size, lipid content, lipid classes and fatty acid composition of structural as well as storage lipids suggested five characteristic seasonal phases that were induced by different feeding histories and environmental conditions. Storage lipids were characterized by 18:1(n−9) as major component, which ranged between 44% of total fatty acids in June and 23% in February. The strong coherence between 18:1(n−9) in the seston lipids and the occurrence of ciliates emphasized the importance of ciliates in the diet of P. acuspes. As indicated by changes in the amounts of fatty acid markers, other food sources varied over the year, suggesting an opportunistic feeding behavior. The spring period was characterized by an increase in typical diatom and dinoflagellate markers, whereas other sources, potentially cyanobacteria, became more important during summer. The life cycle strategy is discussed with respect to extant adaptations to high latitudinal habitats.

Introduction

Pseudocalanus acuspes is a key species in the Central Baltic Sea, as it serves as a major food organism for larval as well as for adult planktivorous fish (Hinrichsen et al. 2002, 2003; Möllmann and Köster 1999, 2002; Möllmann et al. 2003). Knowledge about the processes regulating population dynamics of P. acuspes in the Baltic Sea is essential to understand the principal mechanisms accounting for the high variability of copepod production and reproductive success of fish, which is a main focus of the German GLOBEC project.

Pseudocalanus acuspes mainly inhabits high latitudes (Frost 1989; Runge and Ingram 1991; Siferd and Conover 1992; Norrbin 1996) and due to its absence in the adjacent North Sea (Bucklin et al. 2003) and wide distribution in the Arctic, it is most likely a member of the Baltic glacial relict fauna. Different life cycles were described for Pseudocalanus spp. in high Arctic regions: from biennial (Cairns 1967) and annual cycles (Davis 1976; Conover and Siferd 1993; Lischka and Hagen 2005) up to cycles with two or more generations per year (Pertsova 1981; McLaren et al. 1989; Norrbin 1992). In temperate regions several generations per year are commonly observed (Marshall 1949; Digby 1950).

As a characteristic of the Baltic Sea, adult females of P. acuspes are more abundant in water layers below the thermocline and often concentrate near the halocline, presumably induced by the strong vertical stratification of the water column (Hernroth and Ackefors 1979; Henroth 1985; Renz and Hirche 2006). Hence, sinking algae, detritus or microzooplankton are most likely the only available food sources. Feeding and growth conditions might therefore be suboptimal for this originally marine species (Renz and Hirche 2006) in the temperate brackish environment, with seasonal cycle and diet differing from those of other habitats. Valuable information on the life cycle and overwintering strategy of P. acuspes in the Baltic Sea can be derived from seasonal dynamics in storage lipid content and fatty acid composition of polar, i.e. structural lipids as well as from size variations, as these attributes reflect environmental conditions and food supply during growth of different cohorts.

Studies on in situ grazing rates and food selection of Pseudocalanus spp. are scarce. It has been described that Pseudocalanus spp. exhibits a primarily herbivorous feeding behavior (e.g. Schnack 1975; Corkett and McLaren 1978; Fraser et al. 1989; Cotonnec et al. 2001), whereas other studies suggested a more omnivorous feeding mode (Båmstedt et al. 1990; Norrbin et al. 1990; Peters et al. 2004). To elucidate seasonal dynamics in diet we applied signature fatty acids to identify trophic relationships. We specifically focused on the fatty acid composition of storage, i.e. neutral lipids, in order to obtain unambiguous signals.

The use of specific fatty acids to characterize feeding on different taxonomic groups is well established, e.g. the assignment of 16:1(n−7) and 20:5(n−3) to diatoms (Nichols et al. 1993; Dunstan et al. 1994; Skerrat et al. 1997) and 18:4(n−3) and 22:6(n−3) to dinoflagellates (Sargent et al. 1987; Graeve et al. 1994). However, it is essential to validate those results in the studied ecosystem by comparing fatty acid profiles of the seston with its taxonomic composition. Beside the fatty acid dynamics in the neutral lipids of the copepods, we therefore provide data on the seasonal variation of the seston composition to reveal seasonal changes in the diet of P. acuspes.

Materials and methods

Sampling and experiments

Zooplankton and seston samples were collected in approximately monthly intervals from March 2002 until March 2003 (except for October and December) on 11 cruises in the Bornholm Basin (Fig. 1). To provide representative data for the whole basin, both stations in central and in marginal areas were sampled on each cruise and combined in average values for each month, except for January and February 2003, where only samples from the central basin were available.
https://static-content.springer.com/image/art%3A10.1007%2Fs00227-006-0290-8/MediaObjects/227_2006_290_Fig1_HTML.gif
Fig. 1

Map of investigation area, generated with ODV software (Schlitzer 2005)

Zooplankton was sampled using a WP-2 net with a 10-l bucket end (vertically towed with 0.2 m/s, mesh size 200 μm, 0.26 m2 opening). Sampling depths were adjusted to hydrography covering the water column from the lower halocline up to the surface. Copepods were sorted on board under ambient temperature conditions into −80°C precooled glass vials. Depending on availability each sample consisted of 20–150 adult females of P. acuspes or copepodite stages V (CV), respectively. On three stations of each cruise prosoma lengths of 30 females were measured using formalin preserved samples (4% in seawater).

Seston samples from five depths were taken with 10 l water sampler bottles. Vertical resolution was adapted to the hydrographic structure of the water column, with samples taken from the upper water layer (5 m), from above the thermocline (10 m), from the midwater layer, from above the halocline and in the halocline. Depending on seston concentrations 2–6 l of water were filtered with low pressure (−200 mbar) on precombusted (12 h at 400°C) GF/C filters. All zooplankton organisms were carefully removed under the stereomicroscope immediately after filtration and prior to freezing, so that they did not bias the seston data. Zooplankton samples and filters were permanently stored at −80°C until further analysis.

For analyses of taxonomic seston composition aliquots of 100 ml were taken from water sampler bottles, preserved with 2% acid Lugol’s solution and stored cool and dark until further investigation. Samples were analyzed using Uthermöhl microscopy and phytoplankton as well as protozooplankton cell size was converted to biomass according to Edler (1979) and Putt and Stoecker (1989), respectively.

Analytics

After lyophilization dry mass of copepods was determined using a Sartorius micro-balance (±2 μg). During weighing procedure, samples were temporarily stored in a vacuum desiccator to prevent unequal condensation on the tissue. Lipid extraction was performed with minor modifications as described in Folch et al. (1957) using ultrasonic disruption in dichloromethane:methanol (2:1/v:v) and a washing procedure with aqueous KCl solution (0.88%). For quantification of fatty acids, tricosanoic acid was added as an internal standard prior to extraction.

Lipid classes were separated by solid phase extraction, using 1 ml SiOH glass columns (CHROMABOND®, Macherey-Nagel) on a vacuum manifold. To remove residues the columns were washed with a solvent sequence of acetone, diethylether, and hexane:diethylether-mixtures, prior to sample load. After column conditioning with 4 ml of hexane, 4 μl of lipid extract (lipid concentration approx. 5 μg/μl) were added. The neutral lipid fraction was washed out with 2.5 ml hexane:diethylether (95:5/v:v) and 2.5 ml hexane:diethylether (1:1/v:v). Polar lipids were eluted with 2.5 ml methanol and subsequently 5 ml of dichloromethane were added. The polar fraction was then washed with 2 ml aqueous KCl solution (0.88%).

For fatty acid analyses a subsample of total lipids as well as the total neutral and polar lipid fraction were hydrolyzed and fatty acids were converted to their methyl ester derivatives (FAME) in methanol containing 3% concentrated sulfuric acid at 80°C for 4 h (Kattner and Fricke 1986). After cooling, 2 ml of Aqua bidest. were added, and FAMEs were extracted three times with 1 ml hexane. Samples were analyzed using a gas chromatograph (HP 6890A) equipped with a DB-FFAP column (30 m length, 0.25 mm inner diameter, 0.25 μm film thickness) operated with a temperature program and helium as carrier gas. Samples were injected using a hot split/splitless inlet (250°C, split mode 1:20) or a programmable temperature vaporizer injector (solvent vent mode). The FAMEs and fatty alcohols were detected by flame ionization and identified by comparing retention times with those derived from standards of known composition. The accurate identification of the substances was checked for selected peaks using GC-MS.

Calculations and statistical analyses

The proportions of wax esters (WE), triacylglycerols (TAG) and polar lipids (PL) were estimated based on comparisons of the relative fatty acid and alcohol composition of the neutral lipid fraction and the total lipid extract, whereas the composition of the polar lipid fraction was used to verify the results. Taking the non-fatty acid components into account, the usually dominating phosphatidylcholine was assumed to be the only polar lipid component and the corresponding mass ratio was used. However, this method does not account for sterol esters and cholesterol. Furthermore, the WE content was calculated based on the alcohol content in the total lipid extract.

All statistical analyses were performed using the software SPSS. For all statistical operations that require normal distribution, percentage data (e.g. relative fatty acid composition) were transformed using an arc sine square root transformation. Normal distribution and homogeneity of variances were checked using the Shapiro-Wilk- and the Leverne-test, respectively, according to sample size. For identification of coherences between fatty acid markers and seston taxa, as well as within the fatty acids of seston and storage lipids of P. acuspes females, principal component analyses (PCAs) were performed on the correlation matrix, extracting non-rotated components with eigenvalues >1. Relevant variables (i.e. length, biomass, total and storage lipid content) were analyzed using one-way ANOVA followed by a Tukey’s HSD test for post hoc comparisons with time as independent variable.

To detect seasonal changes between fatty acid compositions in the neutral and the polar lipid fraction of females, the relative amount of each fatty acid was tested between two adjacent months using a Student’s t-test. If the difference between 2 months was only due to one fatty acid on a significance level of P<0.01 or two fatty acids on a significance level of P<0.05, the months were fused to one group. Afterwards, these groups were tested against each other. For months with less than three replicates, i.e. January and February, the months were assigned to the group with the highest similarity in a cluster analysis using the PRIMER software (based on Bray-Curtis similarity and complete linkage cluster mode, data not presented). Selectivity, here understood as ratio between availability of individual fatty acids in the seston and the incorporation into the storage lipids of P. acuspes, was calculated as ratio between their relative content in the seston and in the neutral lipids of the copepods using a logarithmic scale.

Results

Pseudocalanus acuspes

Females of P. acuspes differed substantially in size between succeeding months (Fig. 2a), with a highly significant increase of prosoma length as well as biomass in May up to an average of 966 μm (significant difference to April and June P<0.001) and 12.5 μg/individual, respectively. Over the summer their size decreased, reaching a minimum of 870–880 μm in length in November (significant difference to September P<0.01) and January (significant difference to February P<0.001). From February on, females increased in size again. Dry mass-length ratios, based on monthly averages of variables, basically followed the relationship established by Hay et al. (1988) (Fig. 2b).
https://static-content.springer.com/image/art%3A10.1007%2Fs00227-006-0290-8/MediaObjects/227_2006_290_Fig2_HTML.gif
Fig. 2

a Prosoma length and dry mass of adult females of Pseudocalanus acuspes: triangles monthly average of length, circles monthly average of dry mass, error bars standard deviation. b Length dry mass relationship of adult females of P. acuspes: circles monthly average of dry mass versus monthly average of prosoma length, line length dry mass relationship determined by Hay et al. (1988). c Total lipid content of P. acuspes: circles monthly average of total fatty acids and alcohols (TL) per female, error bars standard deviation, open triangles TL per copepodite stage V (data points stations, please note axis break), dashed line monthly average of TL in percent of dry mass in adult females. d Wax ester (WE) content and egg production rates (EPR) of P. acuspes: filled circles monthly average of WE [% TL] in adult females, open circles WE [% TL] in copepodite stage V (data points stations), line without symbols monthly average of daily individual egg production, EPR modified from J. Renz et al. (submitted), error bars standard deviation. e Triacylglycerol (TAG) content of P. acuspes: filled triangles monthly average of TAG [% TL] in adult females, open triangles TAG [% TL] in copepodite stage V (data points stations), error bars standard deviation. f Monthly average of lipid concentration in the seston: filled circles upper water layer (5 m), open circles above thermocline (10 m), filled triangles midwater layer (20–30 m), open triangles above halocline (30–40 m), filled squares in halocline (40–60 m), n.d. no data

The lipid content in terms of total fatty acids and alcohols of the females ranged from 0.9 to 1.8 μg/individual and from 9 to 14% of dry mass, respectively (Fig. 2c), with a maximum in May (significant difference to April P<0.05). There was no significant difference in May between the total individual lipid content of the females and the copepodite stage V (mean 2.6 μg/individual). In all other months examined, the lipid amount of CV was clearly higher than that of the females, with an average individual lipid content between 4.7 μg in September and 1.6 μg in January.

In both stages, females and CV, wax esters (WE) as well as triacylglycerols (TAG) served as storage lipids throughout the year. Neutral lipids of females (WE and TAG) comprised about two-thirds of total lipids in May and January, respectively, and one-third of total lipids in November (Fig. 2d, e).

The relative amount of TAG ranged between 15 and 35% of total lipids, but due to a high variability no seasonal trends could be identified, neither for females nor CV. In contrast, the relative WE content of the females changed over the year, reaching a maximum of 44% of total lipids in May (significant difference to April P<0.001) and lowest values of 17% in April and November. From November the wax ester content increased until January, afterwards it declined again until March. The WE content of CV was significantly higher during all months, except for May, when values were in the same range as for the females.

Egg production data were adopted from J. Renz et al. (submitted). Daily egg production rates (EPR) showed an overall high variability in the Bornholm Basin (Fig. 2d), reaching a minimum of 0.1 eggs per female per day in January and increasing again in February. The highest EPR was measured in April, but was based only on data from one station, i.e. on the average EPR of 30 females. Whereas in summer mean EPR and WE content paralleled, EPR increased with decreasing WE content in spring 2003.

The total fatty acid composition of females was characterized by high amounts of the typical membrane components 16:0, 20:5(n−3), 22:6(n−3), as well as by elevated levels of 18:1(n−9). Alcohols were dominated by 14:0 and 16:0, while 18:0 and 18:1 were found in much lower quantities (Table ). 1
Table 1

Relative composition of fatty acids [percentage of total fatty acids] and fatty alcohols [percentage of total fatty alcohols] in total lipids of adult females and copepodite stage V (CV) of Pseudocalanus acuspes, values are calculated on basis of monthly averages, values below 1% not shown

 

Females

CV

Min

Max

Mean

SD

Min

Max

Mean

SD

Fatty acids

 14:0

<1

1.6

1.0

0.3

<1

<1

<1

 15:0

<1

1.2

<1

<1

<1

<1

 16:0

10.3

20.7

15.0

2.9

6.0

8.3

7.1

1.0

 17:0

<1

1.0

<1

<1

<1

<1

 18:0

<1

3.7

2.5

0.6

<1

1.3

<1

 16:1(n−7)

1.1

6.5

2.4

1.5

3.3

7.6

4.9

1.6

 16:1(n −9)

<1

2.6

<1

<1

<1

<1

 18:1(n−5)

1.6

4.1

2.8

0.9

<1

2.1

1.3

0.6

 18:1(n−7)

1.4

2.9

2.0

0.4

1.3

1.6

1.5

0.1

 18:1(n−9)

10.1

26.1

18.9

5.2

28.9

39.6

34.8

4.4

 24:1

1.3

3.3

2.2

0.7

<1

<1

<1

 16:2(n−4)

<1

1.2

<1

<1

<1

<1

 18:2(n−6)

2.9

8.5

5.6

1.9

4.2

10.7

7.1

2.8

 16:3(n−4)

<1

1.0

<1

<1

<1

<1

 18:3(n−3)

1.7

4.8

3.3

1.1

2.2

6.9

5.1

1.9

 18:3(n−6)

<1

1.0

<1

<1

<1

<1

 18:4(n−3)

1.3

5.8

3.6

1.4

4.1

7.2

6.1

1.2

 20:4(n−3)

<1

1.6

1.1

0.4

1.0

2.9

2.1

0.7

 20:5(n−3)

12.4

19.5

15.2

2.4

9.8

14.3

12.0

2.1

 22:6(n−3)

16.6

26.2

22.0

3.1

10.6

20.5

15.2

4.0

Fatty alcohols

 14:0

21.4

33.2

26.7

3.7

22.9

30.0

27.2

2.7

 16:0

55.2

70.2

62.7

5.1

59.5

69.6

65.6

3.9

 18:0

<1

11.5

5.1

3.7

1.5

2.6

2.0

0.4

 18:1

2.8

8.2

5.2

1.6

3.5

9.6

5.2

2.5

 20:1

<1

1.6

<1

<1

<1

<1

Min Minimum, Max maximum, SD standard deviation, – SD not calculated

Within the neutral lipid fraction ten important fatty acids (i.e. maximum values ≥2% of total fatty acids) were identified (Table 2). The fatty acid 18:1(n−9) dominated during all seasons, ranging between 44% of total fatty acids in June and 23% in February. Based on their relative fatty acid composition, females were merged into five seasonal groups, which exhibited highly significant differences (Table 2). The first group included females from March and April 2002 as well as from February and March 2003. This spring season was characterized by elevated amounts of the dinoflagellate marker 18:4(n−3), whereas the diatom marker 16:1(n−7) peaked in May. Both groups showed high percentages of 20:5(n−3) and 22:6(n−3), also indicating diatom- and dinoflagellate-based diets, respectively. In May, 18:1(n−9) strongly increased, reaching maximum values in June. From June to September fatty acids were characterized by rising levels of 18:2(n−6) and 18:3(n −3), reaching up to 12 and 8%, respectively. In winter higher amounts of the unspecific fatty acids 16:0, 16:1(n−9) and 18:0 prevailed.
Table 2

Relative fatty acid composition of neutral lipids of females of Pseudocalanus acuspes

 

I

Feb–Apr (n=11)

II

May (n=3)

III

June

(n=4)

IV

July–Sept (n=14)

V

Nov–Jan (n=4)

Level of significance (Student’s t-test)

Adjacent groups

Distant groups

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Mean

SD

I:II

II:III

III:IV

IV:V

I:V

I:III

I:IV

II:IV

II:V

III:V

16:0

13.3

2.5

8.2

1.3

9.4

3.1

14.8

2.8

17.7

4.4

**

 

**

 

*

*

 

***

**

*

18:0

3.0

1.1

1.8

0.6

2.6

1.1

3.0

0.8

6.5

1.9

   

***

***

  

*

**

**

16:1(n−7)

5.0

1.3

9.2

1.0

5.0

0.4

2.5

0.5

3.8

1.1

***

***

***

   

***

***

**

 

16:1(n− 9)

1.6

1.1

1.3

1.4

1.3

0.5

1.6

1.2

7.3

2.8

   

***

***

   

**

**

18:1(n−9)

26.7

5.0

39.1

1.4

43.8

0.5

35.9

4.5

35.7

5.0

***

**

**

 

**

***

***

  

*

18:2(n−6)

8.2

1.3

8.2

0.6

11.6

2.0

12.4

2.2

5.7

1.6

 

*

 

***

**

**

***

**

 

**

18:3(n−3)

5.7

0.9

3.0

0.3

5.2

1.5

7.9

2.2

2.7

1.0

***

*

*

***

***

 

**

***

 

*

18:4(n−3)

8.3

1.9

3.8

0.3

3.7

1.8

4.3

1.7

1.6

0.6

***

  

**

***

***

***

 

**

*

20:5(n−3)

9.4

2.2

8.5

0.5

6.6

0.4

5.6

0.8

4.7

1.4

 

**

*

 

**

**

***

***

*

 

22:6(n−3)

6.3

3.3

8.6

0.9

5.5

0.6

3.6

1.3

3.7

0.9

 

**

*

   

**

***

**

*

I–V Seasonal groups, SD standard deviation

*P<0.05, **P<0.01, ***P<0.001

The neutral lipids of CV showed a very similar fatty acid composition to those of the females (Fig. 3). Especially in May and June there was no significant difference between the fatty acids of both stages, whereas in autumn and winter the fatty acids 18:1(n−9), 18:2(n−6), 18:4(n−3) and 20:5(n−3) of CV showed higher percentages.
https://static-content.springer.com/image/art%3A10.1007%2Fs00227-006-0290-8/MediaObjects/227_2006_290_Fig3_HTML.gif
Fig. 3

Seasonal development of mean fatty acid composition of seston (total lipids) and P. acuspes (neutral lipids): n.d. no data, filled triangles seston in upper water layer (5 m), open triangles seston in halocline (40–60 m), filled circles copepodite stage V, open circles with cross adult females

Principal component analysis on the storage lipid composition of females extracted three components with eigenvalues >1. Only the major two, together explaining 68% of the variance, are presented (Fig. 4). The PCA revealed a strong coupling between the fatty acids 18:2(n−6) and 18:3(n−3), as well as between 22:6(n−3), 20:5(n−3) and 16:1(n−7). The fatty acids 16:1(n−9), 18:0 and 18:4(n−3) were important moieties to distinguish samples along component one, whereas 18:1(n−9), 22:6(n−3) and 16:1(n−7) mostly affected samples influenced by component two. Other fatty acids, like 18:3(n−3) and 18:2(n−6) had a high impact on both components and could not be assigned clearly. Samples from different months were separated, demonstrating seasonal changes in the fatty acid compositions of storage lipids.
https://static-content.springer.com/image/art%3A10.1007%2Fs00227-006-0290-8/MediaObjects/227_2006_290_Fig4_HTML.gif
Fig. 4

Principal component analysis on the relative fatty acid composition of neutral lipids of adult females of P. acuspes, scales were adjusted to combine plots: scales of principal components (PC) refer to sample plot, scale of variables reaches from −1 to +1 for both PCs

Although polar lipids remained rather uniform throughout the year, with 16:0, 20:5(n−3) and 22:6(n−3) contributing between 50 and 73% of total fatty acids (Table 3), their fatty acid profiles divided into the same seasonal groups as the storage lipids. Small but significant differences were mainly due to changes of the 18:1 isomers, as well as of 18:3(n−3) and 18:4(n−3), with largest changes between early (March–April) and late spring (May).
Table 3

Relative fatty acid composition of polar lipids of females of Pseudocalanus acuspes

 

I

II

III

IV

V

  

Level of significance (Student’s t-test)

March-Apr (n=14)

May (n=3)

June (n=4)

July–Sept (n=14)

Nov (n=3)

Jan (n=1)

Feb (n=2)

Adjacent groups

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Mean

Mean

I:II

II:III

III:IV

IV:V

I:V

16:0

16.1

1.1

17.6

0.2

17.8

1.5

16.4

1.3

18.5

0.7

22.3

19.7

*

  

*

**

18:0

3.3

1.9

2.8

0.1

5.0

3.3

3.2

0.4

4.4

1.0

6.5

4.9

   

**

 

18:1(n−5)

5.1

0.8

3.3

0.1

3.2

0.5

3.5

0.3

3.7

0.1

3.5

4.0

**

   

**

18:1(n−7)

1.9

0.2

1.1

0.0

1.3

0.3

1.6

0.3

1.6

0.1

2.2

2.1

***

   

*

18:1(n−9)

3.7

0.9

5.4

1.1

7.5

1.0

5.4

1.6

4.9

0.6

7.4

7.8

*

*

*

 

*

18:2(n−6)

3.9

1.6

2.5

0.1

6.3

3.4

5.0

1.5

0.5

1.4

1.7

6.4

 

*

 

***

***

18:3(n−3)

1.4

0.3

0.9

0.0

1.8

0.2

2.6

0.8

0.7

0.1

0.6

1.6

**

***

*

***

***

18:4(n−3)

3.3

0.9

1.6

0.1

2.1

0.6

2.9

1.1

1.0

0.2

1.0

3.4

**

  

**

***

20:5(n−3)

21.0

3.1

20.0

1.1

16.8

3.0

20.2

2.0

17.7

0.7

12.9

12.6

  

**

*

 

22:6(n−3)

33.4

4.9

40.9

0.4

33.0

5.2

34.0

2.9

37.4

2.6

29.0

29.3

*

*

   

I–V Seasonal groups, SD standard deviation

*P<0.05, **P<0.01, ***P<0.001

Seston

Maximum lipid concentrations in the seston were always found above the thermocline during spring and summer, whereas from autumn until spring, mixing caused more equally distributed lipid contents over the whole water column (Fig. 2f). In terms of total fatty acids and alcohols, maximal lipid contents with up to 52 μg/l were found in upper water layers in April. This lipid-rich seston reached lower water layers with a time delay of 1 month, resulting in a lipid peak near the halocline of 20–24 μg/l in May.

The PCA revealed a strong coherence within the relative seston composition in terms of biomass of different taxonomic groups and typical signature fatty acids (Fig. 5). There was a distinct correlation between 18:1(n− 9) and ciliates and to a lesser degree flagellates, between 16:1(n−7), 20:5(n−3) and diatoms as well as a coherence between 22:6(n−3), 18:4(n−3) and dinoflagellates. The strong connection between 18:2(n−6) and 18:3(n−3) could not be assigned to a specific algal group, but they both had a very similar impact on component one as cyanobacteria and chlorophytes, whereas component two differentiated them. Due to their relative position on component three 18:2(n−6) grouped with chlorophytes, whilst 18:3(n−3) correlated with cyanobacteria.
https://static-content.springer.com/image/art%3A10.1007%2Fs00227-006-0290-8/MediaObjects/227_2006_290_Fig5_HTML.gif
Fig. 5

Principal component analysis on the relative taxonomic (in terms of biomass) and fatty acid composition of the seston: loading plots for the extracted principal components (PC) 1–3 (open circles seston taxa, filled circles fatty acids) and sample plot for the extracted PCs 1 and 2 (numbers months of the year)

Ciliates contributed significantly to seston biomass at all seasons (Table 4), maximum proportions of diatoms were found in spring, of ciliates and dinoflagellates in May. In contrast, other flagellates and cyanobacteria increased during the summer. No data were available for the winter season.
Table 4

Relative seston composition [percentage of total biomass]

 

Depth (m)

Din

Dia

Chl

Cya

Div fl

Cil

Feb–Apr

 Upper water layer

5

24

18

0

3

9

45

 Above thermocline

10

20

23

1

1

14

41

 Midwater layer

20–30

7

38

1

2

11

42

 Above halocline

30–40

15

45

1

1

5

33

 In halocline

40–60

20

45

1

1

3

30

May

 Upper water layer

5

24

1

0

2

7

66

 Above thermocline

10

20

0

0

1

8

71

 Midwater layer

20–30

50

4

2

11

6

27

 Above halocline

30–40

23

1

1

0

2

73

 In halocline

40–60

29

0

8

0

7

55

June

 Upper water layer

5

43

1

1

19

17

19

 Above thermocline

10

23

1

17

17

15

27

 Midwater layer

20–30

25

3

1

11

29

30

 Above halocline

30–40

17

2

0

1

10

70

 In halocline

40–60

13

2

0

7

9

69

July–Sept

 Upper water layer

5

5

4

0

19

17

56

 Above thermocline

10

3

11

2

9

17

57

 Midwater layer

20–30

3

5

1

55

14

22

 Above halocline

30–40

6

0

2

18

30

42

 In halocline

40–60

14

11

1

0

15

59

Din Dinoflagellates, Dia diatoms, Chl chlorophytes, Cya cyanobacteria, Div fl all flagellates except dinoflagellates, Cil ciliates

Trophic interactions

When compared with seston lipids, some fatty acids of P. acuspes females developed with a time lag of 1–2 months at the beginning of 2002, whereas in autumn and winter seston and copepods showed relatively parallel progressions (Fig. 3). Specifically the increase of 18:3(n−3) and 18:2(n−6) in May and June was reflected with some delay in the storage lipids of females. In May, the increase of 18:1(n−9) in females co-occurred with a rise in the seston from lower water layers, whereas the peak in the upper water column in July was not found in the copepods.

As indicated by a selection index (Fig. 6), 16:0 and 18:0 were usually negatively selected during all seasons, whereas 18:1(n−9), 18:2(n−6) and 20:5(n−3) were elevated in the neutral lipids most of the time. Selectivity for all other fatty acids changed with time or depth. In early spring 18:3(n−3), 18:4(n−3) and 22:6(n−3) were accumulated in storage lipids as compared to the seston, as well as 16:1(n−7) in May and June.
https://static-content.springer.com/image/art%3A10.1007%2Fs00227-006-0290-8/MediaObjects/227_2006_290_Fig6_HTML.gif
Fig. 6

Selection index for fatty acids with vertical resolution [sorted top-down: upper water layer (5 m), above thermocline (10 m), midwater layer (20–30 m), above halocline (30–40 m), in halocline (40–60 m)]

Discussion

Seasonal cycle

The seasonal cycle and condition of P. acuspes in the Bornholm Basin were described on the basis of lipid content and composition as well as prosoma length, to relate lipid dynamics and size variations to the life cycle of this originally Arctic copepod in the Baltic Sea. Pronounced changes in body size and fatty acid composition of structural lipids of females revealed five “environmental cohorts”, which obviously experienced similar biotic and abiotic conditions during development, thus leading to constant attributes of females: early spring (February–April), late spring (May), early summer (June), late summer (July–September) and winter (November–January). Recent studies on stage composition and growth measurements (Renz and Hirche 2006; J. Renz et al., submitted) indicate that P. acuspes basically follows an annual cycle in the Bornholm Basin, although the development of a second cohort in summer was also considered possible.

Aligning the “environmental cohorts” in this context, the early spring cohort consisted of females, which successively matured from older overwintering copepodite stages. Their growth was at least partly fueled by storage lipids, as indicated by the decrease in wax ester content. This cohort was followed by females in May, which were probably larger due to better feeding conditions and lower temperatures (Vidal 1980; Klein Breteler and Gonzalez 1988). Strong changes in size co-occurred with variations in the composition of structural lipids and in storage lipid content, indicating different feeding histories during growth. The May cohort probably derived from younger overwintering copepodids of the previous year, which encountered a high food supply in the upper water column in April. In May lipid-rich seston reached lower water layers, thus providing better feeding conditions for older copepodite stages and females. It remains, however, a matter of conjecture, whether the drastic changes in May were due to successively maturing cohorts or rather to the appearance of a new generation.

To better understand the further progression of the seasonal cycle, valuable information can be derived from comparisons of storage lipid content of CV and females. In May the amounts of storage lipids of females and CV hardly differed, whereas in summer the copepodids were always richer in wax esters. A similar decrease in storage lipids, measured as oil sac volume, was observed by McLaren et al. (1989) in summer females of P. acuspes in the Bedford Basin, Nova Scotia. Two, not mutually exclusive mechanisms, causing the pronounced differences between females and copepodids, can be assumed:
  1. 1.

    Food supply: The accumulated storage lipids were used up very quickly by the females for metabolic costs of last molt, gonad maturation and egg production. Due to reduced food availability, the depletion of reserves proceeded more quickly during summer than in May, explaining a high wax ester retention of females in late spring. Lipid retention is a direct expression for surplus of food. Apparently, the food supply alone was not sufficient to sustain egg production at ambient temperatures in summer. The pronounced utilization of storage lipids signifies that in summer food limitation might have been an important factor, whereas in May egg production was primarily determined by abiotic factors. Hence, sub-optimum growth conditions might reduce the number of generations per year.

     
  2. 2.

    Onset of overwintering: Only the lipid-poor copepodids accomplished maturation during summer and autumn to produce potentially more successful offspring, whereas the lipid-rich copepodids passed into an “active diapause”, with ongoing feeding, suspended development and resting gonads (McLaren et al. 1989). According to this hypothesis, the females found in the Bornholm Basin from summer to winter would represent a still maturing but minor part of the population. This is consistent with the drastic decline of female and nauplii abundance in the water column in July and August (Renz and Hirche 2006) and the slow developmental rates in late spring and summer (J. Renz et al., submitted). Such a continued development of only a minor part of the generation of Pseudocalanus sp. was also observed in the White Sea (Pertsova 1981). Norrbin (Norrbin et al. 1990; Norrbin 1996) suggested that it is less a continuous process, triggered by the physiological state as proposed by McLaren et al. (1989), but rather a specific switching date, at which P. acuspes copepodids stop maturation but proceed to accumulate lipids. Klein Breteler and Gonzalez (1988) suspected that changes to poor food conditions are necessary to induce hormonal cessation of development in favor of lipid production. Still, it remains unclear, how an external trigger, which would be effective early in the year, should function in the Bornholm Basin, with higher temperatures and longer periods of high food abundance as compared to high latitudinal habitats.

     
In conclusion, we found evidence that the life cycle of P. acuspes in the Baltic Sea resembles that of Pseudocalanus spp. in Arctic regions (e.g. Pertsova 1981; McLaren et al. 1989; Norrbin et al. 1990; Norrbin 1991; Conover and Siferd 1993), with highest reproductive activities in spring, a successive accumulation of resting copepodite stages starting in early summer and a potential interposition of minor summer generations. This is supported by the corresponding lipid-storing strategies of P. acuspes. Wax ester levels in the Baltic were similar to those found for CIV and CV of P. acuspes in Arctic regions (Båmstedt et al. 1990; Norrbin et al. 1990), with values reaching 72% of total lipids in autumn and around 55% in summer. To our knowledge, there are no data available on the wax ester content of P. acuspes females.

Based on these fundamental analogies, we hypothesize that life cycle and lipid-storing strategies of P. acuspes in the central Baltic Sea originate from extant adaptations to high latitudinal habitats.

Trophodynamics

The five different phases of the seasonal cycle were also reflected in the fatty acid dynamics of neutral lipids, although they are less conservative than structural lipid composition and body size. We applied signature fatty acids (Lee et al. 1971; Sargent and Whittle 1981; Sargent et al. 1987; Graeve et al. 1994; Daalsgard et al. 2003) to identify feeding preferences and food selection of P. acuspes. Due to parallel analyses of the seston, we were able to assign the fatty acid markers to specific food sources.

Similar to all other studies dealing with the fatty acid composition of Pseudocalanus spp. (e.g. Kattner et al. 1981; Kattner and Krause 1989; Fraser et al. 1989; Norrbin et al. 1990; Cotonnec et al. 2001), we found 18:1(n−9) to be one of the most abundant fatty acids throughout the year. Apparently, this does not inevitably indicate similar feeding habits in different habitats, but rather a species-specific attribute, probably affected by metabolic processes. This fatty acid is not only known to be characteristic for carnivorous or detritivorous feeding (Sargent and Falk-Petersen 1981; Falk-Petersen et al. 1990), it is also synthesized de novo by copepods (Pascal and Ackman 1976; Sargent and Henderson 1986; Kattner et al. 1994; Kattner and Hagen 1998). Thus, a trophic assignment of 18:1(n−9) remains problematic. Nevertheless, as revealed by principal component analysis we found a strong coherence between 18:1(n−9) levels in the seston lipids and the occurrence of ciliates. Lipid profiles of ciliates have been reported to reflect, at least within species-specific ranges, the fatty acid composition of their diet (Ederington et al. 1995; Harvey et al. 1997; Broglio et al. 2003). Therefore, a comparison of field data with fatty acid profiles derived in laboratory studies is rather difficult. However, our data emphasize a high relevance of ciliates in the food spectrum of P. acuspes. The apparently intense use of heterotrophic organisms and/or detritus might be explained by the vertically stratified environment in the Baltic Sea. Due to the concentration of older copepodite stages of P. acuspes in deeper water layers (Hernroth and Ackefors 1979; Renz and Hirche 2006), their potential food mainly consisted of sinking material from the surface and organisms inhabiting the lower stratum of the water column. At least in May and June, those were mainly ciliates, representing approximately 75% of living biomass, co-occurring with very high 18:1(n−9) levels in the females. Feeding of Pseudocalanus spp. on ciliates (Klein Breteler et al. 2004) and heterogeneous particulate matter was documented in laboratory studies (Poulet 1974, 1976; Pavlovskaya and Pechen´-Finenko 1975 as cited by Corkett and McLaren 1978). We did not quantify detritus, although an accumulation of degraded material on the halocline is to be expected.

A comparison between the fatty acid and taxonomic composition of seston revealed a relationship between 18:4(n−3), 22:6(n−3) and the biomass of dinoflagellates, as well as coherence between 18:4(n−3) and other flagellates. Those fatty acids are known to reach high levels in dinoflagellates and cryptophytes (Sargent et al. 1987; Graeve et al. 1994, 2001 and references therein; Daalsgard et al. 2003). In our study, the biomarker 18:4(n−3) was found in significantly higher amounts in early spring, and 22:6(n−3) was also more abundant from February until May. This indicates a preferential ingestion of flagellates or dinoflagellates in spring time, although dinoflagellates showed a rather constant portion of total biomass during all seasons examined, whereas other flagellates increased later in the year. Pseudocalanus spp. selectively feeds on flagellates such as cryptophytes and dinoflagellates (e.g. Geen and Hargrave 1966; Zagorodnyaya 1974). This high quality food (Brown et al. 1997) enhances growth, egg production and lipid accumulation and also decreases mortality (Klein Breteler et al. 1990; Koski et al. 1998; Koski and Klein Breteler 2003).

Diatom blooms, which have reappeared in the Bornholm Basin since 1999, were mainly restricted to early spring (February–April) (Wasmund et al. 2003, present study). However, the diatom marker 16:1(n−7) reached its maximum in P. acuspes not until May, when diatoms were of only marginal importance in the water column and their fatty acid markers in the seston had already decreased significantly. This suggests that lipids observed in the new females in May probably derived from lipid reserves built up during earlier copepodite stages. This time shift between fatty acid levels in seston and copepods related to the period of higher lipid accumulation or retention by females. Low lipid levels reflect changes much quicker, probably causing the more synchronous progression of fatty acid composition of seston and storage lipids later in the season. Alternatively, in spite of low standing stocks of diatoms, their production rates may have been high, as the production potential of diatoms was evident from a small diatom bloom during July.

Diatom marker levels were rather low in the Baltic. Especially in Polar Regions with more pronounced diatom and ice algal blooms 16:1(n−7) may reach values of up to 20% of total fatty acids in CIV and CV of P. acuspes in the Arctic summer (Norrbin et al. 1990), thus exceeding twice the maximum value found for CV in the present study. Very similar results, indicating diatom-based feeding, were found for other Pseudocalanus species in Polar Regions with 16:1(n−7) levels reaching up to 40% of total fatty acids (Peters et al. 2004, S. Lischka and W. Hagen, submitted).

The rather limited ingestion of diatoms seems to be characteristic for temperate regions, as all studies show similarly low marker amounts (Kattner et al. 1981; Kattner and Krause 1989; Fraser et al. 1989; Cotonnec et al. 2001). Still, the levels of 16:1(n−7) found in our study belong to the lowest ever measured for Pseudocalanus spp., indicating a more intense use of other food sources. Especially cyanobacteria have to be considered as potential diet in the Baltic Sea, as they usually bloom intensively during summer, except for 2002, when only a minor bloom was registered (Wasmund et al. 2003). However, in our study cyanobacteria values reached up to 55% of seston biomass.

Cyanobacteria are very variable in their fatty acid compositions, with marked differences occurring even in the same genus (Gugger et al. 2002). Due to the coexistence of different cyanobacteria species in the Baltic Sea, it is not easy to identify a clear trophic signal. Many studies reported that a characteristic fatty acid pattern of cyanobacteria consists of 16:0, 16:1, 18:2(n−6) and 18:3(n−3) (e.g. Murata and Nishida 1987 and references therein, Vargas et al. 1998; Gugger et al. 2002). The simultaneous occurrence of 18:3(n−3) and 18:2(n−6) with cyanobacteria and chlorophytes in the seston, indicated that elevated amounts of these fatty acids in P. acuspes were due to an augmented use of cyanobacteria in summer, since chlorophytes were only of minor importance. Hoppe (1981) showed that cyanobacteria and microzooplankton often build up agglomerates especially in the late bloom phase, which might improve food quality and attractiveness for copepods, leading to a more intensive use in the later phases (Meyer-Harms et al. 1999). This might explain why cyanobacteria were reflected in the storage lipids with a delay of some months.

We conclude that P. acuspes displays a basically opportunistic feeding behavior in the Baltic Sea. Five different seasonal fatty acid profiles were determined in the neutral lipids with high levels of 18:1(n−9) at all times, indicating a species-specific storage pattern as well as a ciliate-dominated diet. Other food sources varied over the year. In early spring dinoflagellates were increasingly utilized, whereas in late spring diatom markers were most strongly reflected in the fatty acid composition. During summer, cyanobacteria, and probably to a lesser degree chlorophytes, seemed to contribute substantially to the diet of P. acuspes.

Acknowledgements

The study was funded by the German Federal Ministry for Education and Research within the GLOBEC GERMANY project (03F0320C). We wish to thank the crews and scientific parties of the RV Alkor, Heincke and A.v.Humboldt for the excellent support during the field phase. We also thank M. Graeve for GC-MS measurements. The critical revision and improvement of the manuscript by H. Auel, D. Stübing and three anonymous reviewers are gratefully acknowledged.

Copyright information

© Springer-Verlag 2006