© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Stock structure of Sebastes mentella in the North Atlantic revealed by chemometry of the fatty acid profile in heart tissue
Department of Chemistry, University of Bergen N-5007 Bergen, Norway
*Correspondence to H. Joensen; Fisheries Laboratory of the Faroe Islands, Nóatún 1, 100 Tórshavn, Faroe Islands; tel: +298 315092; fax: +298 318264. e-mail: horaldj{at}frs.fo.
Chemometric analysis of fatty acid profile in the heart tissue of redfish (Sebastes mentella, Travin 1951) from 11 locations in the waters off Norway, Faroe Islands, Iceland, and from two strata at different depths in the Irminger Sea revealed the presence of four separate stocks in the North Atlantic Ocean. The investigation has, statistically, shown: (i) a clear distinction between S. mentella on the Faroe Plateau and S. mentella in the deeps south-west of the Faroe Bank and on the Wyville Thomson Ridge; (ii) a relationship between S. mentella inhabiting the Faroe Plateau and the Norwegian Sea; and (iii) two significantly distinct stocks of S. mentella dwelling on the Icelandic Plateau. The chemometric method, consisting of methanolysis, gas chromatography of the resulting fatty acid methyl esters, and multivariate statistical treatment of the analytical data, has a discriminating power high enough to differentiate at stock level and individual level.
Keywords: discrimination, fatty acid profile, heart tissue, North Atlantic, principal component analysis, Sebastes mentella, stocks
Received 7 January 2002; accepted 1 October 2003.
 |
Introduction |
|---|
 Top Introduction Materials and methodsResultsDiscussionConclusionReferences |
|---|
The live bearing Sebastes mentella Travin 1951 is very common and widely distributed from the coast of Britain along Norway to the Barents Sea, north to Svalbard, off the Faroes, Iceland, Irminger Sea, east and west Greenland, and along the east of North America from Baffin Island to Cape Cod (Magnusson and Magnusson, 1995; Roques et al., 2001).
S. mentella is mature at an age of 15–17 years in the Northeast Arctic (ICES, 2000) and 18–20 years in waters off the Faroe Islands (Reinert, 1998). The slow growth of this redfish species, the late maturation, the inadequate knowledge of stocks and their respective sizes, and the large variation in catches have caused concern among the participating fishery nations about how to manage the economically important biological resources properly. This has resulted in national and international research activities using many different scientific methods (morphology, otolith-analysis, various DNA-analysis, electrophoresis, and isoelectric focusing of enzymes, etc.) to reveal the stock structure of S. mentella (Anon., 1998).
Different methods lead to various views of the stock structure of S. mentella. The study of chromatoblastomas and related pigmented lesions of 34 960 specimens in the North Atlantic, 21 460 of them from the Irminger Sea, resulted in the judgement that Irminger Sea S. mentella probably represents one closed population (Bogovski and Bakai, 1989). Based on the regularity of seasonal variations of lengths, sex ratio, and maturation curves in different areas, Alekseev (1999) concludes that S. mentella in the Northwest Atlantic constitutes a single population.
Other investigations based on size, weight, maturity, and infestation rate of the parasitic copepod Sphyrium lumpi pointed out two types of S. mentella in the Irminger Sea: an oceanic type with the main distribution above 400 m and a deep-sea type below this depth (Magnusson and Magnusson, 1995). Electrophoretic studies of genetic relationships were able to differentiate between deep-sea and oceanic S. mentella in the Irminger Sea as well as among other types of this redfish species in the North Atlantic (Johansen et al., 2000b).
Recognition of stocks of S. mentella is not easy in cases where even species-identification is rather difficult, and where more than one analytical method has to be used to achieve an unambiguous conclusion. There exists a variant of Sebastes in the Northeast Atlantic, which is genetically similar to S. mentella, but is morphologically more like Sebastes marinus (Altukhov and Nefyodov, 1968; Altukhov et al., 1968). This redfish variant, giant Sebastes, is recently characterized genetically by electrophoresis of haemoglobin and allozymes, but the taxonomic status and the relation to the S. marinus and S. mentella is not conclusive (Johansen et al., 2000a).
A further complicating factor, in stock identification, is indications, and even evidences, of a hybridization between S. mentella and Sebastes fasciatus (Desrosiers et al., 1999; Sévigny et al., 2000; Roques et al., 2001). Moreover, even if S. mentella, S. marinus, and Sebastes viviparus are well established species (Nedreaas and Nævdal, 1989, 1991a, b), the analysis of mitochondrial DNA indicates either a balance between mutation and selection or mitochondrial gene flow between these three species of Sebastes (Bentzen et al., 1998). The close relationship among the species is also manifested by isoelectric focusing of proteins in the filet (Rehbein, 1990) and analyses of mitochondrial RNA (Sundt and Johansen, 1998). These two methods were unable to discriminate between S. mentella and S. marinus and are therefore, in their present developmental stages, unable to identify stocks of S. mentella.
Morphology has recently been applied to distinguish three geographic variations in an approved single-management-unit of S. mentella in the Northeast Arctic (Saborido-Rey and Nedreaas, 2000). However, morphological classifications have been generally questioned (Grant, 1987).
Multivariate analysis of fatty acid profile is a new method in Sebastes research. This method was implemented after convincing results were achieved in a pilot test on two reared stocks of cod in the Faroe Islands; significantly different fatty acid profile appeared in the phospholipid-rich heart tissue in spite of identical biotic and abiotic factors during the rearing period of three years and eight months before analysis (Joensen et al., 2000). The next step was to investigate the method's discrimination power among the three species of redfish: Sebastes viviparus, S. marinus, and S. mentella in waters off the Faroe Islands and Norway (Joensen and Grahl-Nielsen, 2000, 2001). The positive results at the species level prompted an investigation into the stock structure of S. mentella in the North Atlantic. The purpose of this investigation is thus to reveal the stock structure of S. mentella in the North Atlantic by the fatty acid profile method.
|
Materials and methods |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionConclusionReferences |
|---|
Sebastes mentella were caught in 11 areas in the North Atlantic (Figure 1) in the summer and fall of 1999 (Table 1). The fish were frozen, subsequently glazed, and kept at –20°C until excision of the hearts one to two months later and four months for the samples from the Irminger Sea 1 (Ir1). The maturity stages (Table 1) were determined according to common standards (found in the Sampling Manuals of the Institutes of Marine Research) in the Nordic Countries (Jákup Reinert, personal communication).
|
|
Heart tissue from 20 specimens from each location was analyzed by the fatty acid profile method, comprising methanolysis, gas chromatography, and multivariate statistics as described earlier (Joensen and Grahl-Nielsen, 2000, 2001; Joensen et al., 2000).
The subsamples, weighing approximately 20 mg, were subjected to methanolysis in thick-walled glass tubes with Teflon-lined screw caps by 0.5 ml dry methanol containing 2 M HCl for 2 h at 90°C. After methanolysis, approximately half the methanolic HCl solution was evaporated by nitrogen gas. To this, 0.5 ml of distilled water was added, and the fatty acid methyl esters were extracted twice with 1 ml hexane.
We chromatographed 1.0 µl of the mixed hexane extracts on a 25 x .025 mm fused silica column with polyethyleneglycol as stationary phase, with a thickness of 0.2 µm (CP-WAX 52CB Chrompack) and helium at 20 psi as mobile phase. The column was mounted on Hewlett–Packard 5890A gas chromatograph equipped with a Hewlett–Packard 7673A autosampler (Manufacturer: Hewlett–Packard Company; address: Hewlett–Packard Company, 3000 Hanover St., Palo Alto, California 94304-1181, USA) and a flame-ionization detector. The injector temperature was 260°C. The temperature of the column was kept at 90°C for 4 min after injection and thereafter increased to 165°C at rate of 30°C/min, followed by an increase of 3°C/min to 225°C. This temperature was maintained for 10.5 min. The flame-ionization detector was set at 330°C. Samples were analyzed in random order with a standard solution, GLC-68D from Nu-Chek-Prep, Elysian, Minnesota, USA, containing 20 fatty acid methyl esters, and procedural blanks between every eighth sample. Altogether, 49 components were registered in the chromatograms, 40 of those were identified as fatty acids by way of standard mixture and mass spectrometry. Two of the peaks represented cholesteryl derivatives (Joensen et al., 2000) and seven of the peaks were not identified. The peaks were integrated by VG-Multicrome software and corrected by response factors determined from the standard mixture.
The relative amounts of the fatty acids, in per cent of sum, in each sample were determined. The average composition of the 20 redfish from each area was found, and these 11 average samples were subjected to multivariate treatment based on principal component analysis. The relative values of the fatty acids were logarithmically transformed, hereby levelling out differences among fatty acids present in large and small amounts. With each sample positioned in the multi-dimensional space described by the log-transformed variables (fatty acids), the three coordinates (principal components, PCs) that described the largest, second largest and third largest variances among the samples were computed using the program package SIRIUS (Kvalheim and Karstang, 1987). In this manner, the relationship between the samples could be described in three dimensions, instead of the original 49, without considerable loss of the total original variance. The dominating, systematic variance among the samples will be manifested in the principal components. The samples were displayed in the coordinate system of PC1 vs. PC2 vs. PC3.
Principal component analysis is of only a qualitative nature. The significance of the observed difference between groups of samples can be determined by SIMCA (Soft Independent Modeling of Class Analogies) (Wold, 1976, 1978; Wold and Sjøstrøm, 1977; Ugland and Massart, 1996), also available in the SIRIUS software package. Here space-filling, principal component models for the samples in each group are computed. Outliers are detected, and the models are recomputed after exclusion of the outliers. Finally, it is determined if the samples are inside a model, or how far outside it is by computing the residual standard deviation (RSD) of the samples to the models.
Material from the same fish plus additional specimens (see Table 1) was sampled for electrophoretic, morphometric, microsatellite, and otolith analysis.
|
Results |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionConclusionReferences |
|---|
It is not possible to observe any stock structure of S. mentella by univariate inspection of the data (Table 2), obtained from the analysis of the 20 specimens caught on each of the 11 locations in the North Atlantic. The 55 possible pairwise combinations of samples from the 11 locations were subjected to separate principal component analyses (PCA). This disclosed clear relationship among some groups and evident differences among others.
|
None of the 55 PC-plots are shown. However, when all the data from the 220 samples were subjected to PCA simultaneously, it was not easy to see any stock structure on the resulting PC-plot, due to overlap. Therefore, the centroide samples for each of the 11 groups were subjected to PCA. The resulting 3D-plot of PC1 vs. PC2 vs. PC3 shows four distinct groups (Figure 2).
|
The 3D-plot in Figure 2 accounts for only 72% of the total variance among the 11 centroide samples. To evaluate the true relationship among the samples, space-filling SIMCA-models were computed for the two groups containing four and five centroide samples. The distances of the 11 samples to the two models, computed as residual standard deviation, are displayed in the class distance plot in Figure 4. The centroide samples forming the two models are all inside their own model and well separated from the other model. The centroide samples for the samples from location Ic2 and for the location Ir2 are well separated from the two models, but they bear a closer resemblance to group 1 than to group 2.
A dendrogram (Figure 3), obtained by cluster analysis, based on Euclidian distances in the 49-dimensional space, of the same 11 centroide samples, confirmed the existence of four distinct groups of S. mentella in the Northeast Atlantic.
|
Having identified four distinct groups by way of the centroide samples, the average relative amounts and standard deviations of the heart fatty acids of the redfish within each group were calculated (Table 3). Group 1 encompasses 100 redfish from the locations N1, N2, F1, F4, and F5. Group 2 encompasses 80 redfish from the locations F2, F3, Ic1, and Ir1. Group 3 encompasses the 20 redfish from location Ic2 and group 4 the 20 redfish from location Ir2.
|
Evaluated univariately, it is obvious that the amounts of fatty acids in the four groups of S. mentella are different in most cases. When the fatty acids of the four groups were mutually treated by Student's t-test, it was found that all of them except 18:0, 22:0, 22:1n9, and 22:1n5 were discriminating significantly (p<0.01) in one or more of the six cases (Table 3). These four fatty acids were excluded from further statistical treatment, which was carried out on the remaining 45 variables (mostly fatty acids).
A proper classification and statistical evaluation including all individual redfish, based on the 45 discriminating fatty acids denoted in Table 3, where the total variation is taken into consideration, were now carried out by forming space-filling models for each group by way of SIMCA.
The number of outliers, found by plotting RSD vs. leverage and RSD vs. Hotelling's t-test, respectively (Kvalheim and Karstang, 1987), was 23, 17, 4, and 2 in the models of group 1, 2, 3, and 4, respectively. These outliers were not included in the final models. By way of cross-validation (Wold, 1978), it was found that groups 1, 2, 3, and 4 were best described by 5, 4, 4, and 2 principal components, respectively. The critical sizes of the models, based on the residual standard deviations (RSD) of the samples, were calculated at 95% level. Then the distances, expressed as residual standard deviation of all 220 samples to each of the four models, were calculated. One hundred and fifty-four redfish were correctly classified, and only one sample was misclassified, while eight fish fell into another group in addition to their own. Due to a large spread among the samples, 57 redfish fell outside all models and in this way remained unclassified (Table 4).
|
|
Discussion |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionConclusionReferences |
|---|
Of the tissues and oils investigated so far, the heart tissue is the most suitable for discrimination among Sebastes species in terms of differences in fatty acid composition (Joensen and Grahl-Nielsen, 2000, 2001). This was also the case for distinction among stocks of Striped Bass (Grahl-Nielsen and Mjaavatten, 1992), herring (Grahl-Nielsen and Ulvund, 1990), and cod (Joensen et al., 2000). Therefore the heart tissue is also the tissue of choice for discrimination among stocks of S. mentella.
A wide range of investigations have shown that the composition of fatty acids in fish tissue is influenced by the composition of the fatty acids in the diet. Still, many of these investigations have been conducted with diets of anomalous compositions, often based on vegetable oils (Bell et al., 1983; Cowley et al., 1983; Leray and Pelletier, 1985; Lie et al., 1986; Anderson and Arthington, 1989). In addition, many of the investigations were carried out on juvenile stages of fish which are more easily affected than fish in their mature stages (Muje et al., 1988; Navarro et al., 1995).
However, even in cases where natural diets have been used on mature fish, an influence on the fatty acid composition has been demonstrated, i.e. Kirsch et al. (1998) fed cod squid, Illex illecebrosus, for 6 weeks followed by mackerel for 8 weeks. The fatty acid composition of the total body lipids first changed in the direction of the pattern found in squid, and thereafter in the direction of the pattern in mackerel. The change in pattern in the body lipids occurred during the first 3–5 weeks after change in diet. However, despite the changes in the fatty acid pattern, the pattern in the cod could be readily distinguished from that in the diets.
The richer a tissue is in triacylglycerides, the closer is the resemblance of its fatty acid composition with that of the diet (Viga and Grahl-Nielsen, 1990).
Contrary to the labile profile of the fatty acids in the triacylglycerides, the profile of the fatty acid in the phospholipids may serve as natural mark for the identification of stocks. The condition is that the fatty acid profile of these lipids is typical for a stock and stable over time. Our investigation on the Faroe stocks of cod (Joensen et al., 2000) do show that the between-stock difference in fatty acid composition in heart tissue, in which the lipids are made up of between 80% and 90% phospholipids, is caused by inborn factors.
Compared with the composition of the fatty acids in the triacylglycerides, the composition of fatty acids in phospholipids is generally expected to be less sensitive to the diet. Pickova et al. (1997) demonstrated that the fatty acid composition of the total phospholipids in cod eggs was population specific and diet independent. They used the egg phospholipid fatty acids to distinguish between Skagerrak and Baltic cods. Bandarra et al. (1997) have shown that the fatty acid composition of sardine phospholipids is hardly affected by the seasonal change in dietary planktonic lipids. In rainbow trout, Oncorhynchus mykiss, dietary fatty acids are selectively incorporated into muscle phospholipids to obtain narrowly defined physiological levels (Greene and Selivonchick, 1990). In fact, this was also the case for triacylglycerols. Owen et al. (1972) detected relatively little change in the fatty acids of phospholipids of liver and extrahepatic tissue in plaice, Pleuronectes platessa, subjected to different dietary regimes.
The natural diet of redfish on the Greenlandic shelf was found to be composed of 23 common and 14 rare species (Gorelova, 1997). On the Faroe Plateau, the diet is also made up of a broad spectrum of different food items (Reinert, 1993). Redfish in the North Atlantic, except the Irminger Sea, are generally supposed to be benthic feeders, but stomach contents indicate that they leave the sea-bed occasionally (Reinert, 1993, 1998). The wide variety of ingested prey and the negligible influence of prey on the fatty acid composition of phospholipids indicate that the differences among redfish from the various areas in the North Atlantic are not diet-induced.
Environmental temperature is known to affect the fatty acid composition of the tissue of the poikilothermic fish in that the degree of unsaturation increases by decreasing temperature (Morris and Culkin, 1989). However, the temperature does not fluctuate much below the depth of 200 m in the North Atlantic (Hansen, 2000), and S. mentella has the main distribution at the depth-range of 300–750 m (Reinert and Lastein, 1992). Autumn aggregation of Oceanic S. mentella in the Irminger Sea inhabits the uppermost 250 m, in temperatures from 4.0°C to 5.5°C (Magnusson and Magnusson, 1995). Given the small temperature amplitude in the waters inhabited by S. mentella, the revealed differences in fatty acid profiles among the four stocks are expected not to be temperature induced.
Several investigations have been carried out to detect possible effects of pressure on the fatty acid composition of phospholipids in fish tissue (Patton, 1975; Macdonald and Cossins, 1985; Cossins and Macdonald, 1986; Phleger and Laub, 1989). No clear-cut conclusions about depth dependence have been drawn. In the present study, S. mentella caught at areas F5 and N2, with a depth difference of 170 m, had similar fatty acid composition of the heart tissue. Likewise, the hearts of S. mentella from areas F3 and Ir1, with a depth difference of 200 m, had similar fatty acid composition. On the other hand, S. mentella caught at the same depth in areas F1 and Ic2 had significantly different fatty acid profiles of the hearts. This means that the observed similarities and differences in fatty acid composition of the heart tissue of S. mentella from the different areas are independent of the habitat depths of the redfish.
The redfish for this investigation were caught in the autumn of 1999 in all locations except the location Ir1 (Figure 1) in the Irminger Sea, where they were caught in June 1999. In waters off the Faroes, the frequency of males and females is approximately the same in autumn, while the proportion of females is considerably lower in spring (Jákup Reinert, personal communication). This percentage-reduction of females is due to their migration to the Irminger Sea (Reinert, 1998) to extrude larvae. Autumn is the season when many reproduction-related physiological processes, like selective mobilization of lipids from liver, muscle, mesenteric fat, etc. to the gonads, are relatively passive in the Sebastes (MacFarlane et al., 1993; MacFarlane and Bowers, 1995; Norton and MacFarlane, 1995). Most of the specimens are maturing or immature (Table 1); though, the specimens caught in June in the Irminger Sea are an exception. Fifty percent of these were spent (Table 1). Anyway, it was not possible to get any specimens of the deep-sea S. mentella from the Irminger Sea 1 (Ir1) in autumn 1999, only from the summer season of July 1999 (Table 1). But, as S. mentella is a spring spawner, this gap in sampling time has probably no impact on the fatty acid profile in heart tissue. The reasons for this are firstly that lipid levels in white muscle were not significantly different among ovary maturation stages (MacFarlane et al., 1993) and secondly that heart tissue is more stable than the white muscle (Viga and Grahl-Nielsen, 1990; Mjaavatten et al., 1998).
In the autumn the stocks are believed to be well separated on the feeding and copulation areas (Anon., 1998; Reinert, 1998). Consequently, the revealed groups of S. mentella must be assumed to be equivalent to stocks, i.e. groups with unaffected gene flow (Pitcher and Hart, 1982).
The specimens were frozen (–20°C) and subsequently glazed four times on board the trawlers. This was done to minimize the deterioration. Even so, it was obvious from the appearance of the redfish that deterioration to some extent had taken place in approximately 50 of the specimens. The red, fresh colour had disappeared and several of the fins and jaws were in addition cracked and crushed, due to rough treatment, sloppiness in the pre-freezing procedure, and 2–4 months of storage. Since better samples were unavailable, these specimens were nevertheless subjected to analysis. These decayed specimens caused large variation among the samples with many outliers (Dulavik et al., 1998; Joensen et al., 2000; Joensen and Grahl-Nielsen, 2001), but did not invalidate the results of the investigation. The samples, which turned out to be outliers were not excluded from the statistical analyses, except when the SIMCA-models were created. Exclusion of outliers, when creating models, is also practised in other modern analytical methods using multivariate analysis (Saborido-Rey and Nedreaas, 2000; Roques et al., 2001).
Even if the material was deteriorated and not optimal, it was possible by using averaged data, just like many other genetic methods, to detect four significantly different stocks. When testing sample-membership in the four stock-models, calculations, based on data from each individual sample, resulted in misclassification of only one single sample (0.5%) of 220, although 57 (26%) were unclassified (Table 4). The main reason for the unclassified specimens was the deteriorated samples. Unveiled cross-breeds would probably also be unclassified. Generally, variation to some extent is unavoidable when dealing with sexually reproducing organisms. The spaciousness of the model determines the number of unclassified specimens to a certain extent as well. By decreasing the probability level from p=0.05 to p=0.01, a larger part of the variable space is embraced by the model. This decreases the risk of false negatives, but increases the risk of false positives.
The stock structure of S. mentella, which we have revealed here, corresponds in part with the results of electrophoretic analysis of the enzyme: MEP-1, from the same set of samples (Torild Johansen, personal communications). While the fatty acid method detected four distinct stocks, the enzyme method detected two. Both methods found S. mentella from the locations F2, F3, Ic1, and Ir1 to belong to the same stock (stock 2) distinct from the S. mentella from the other locations. The enzyme method assigned S. mentella from all the other locations, N1, N2, F1, F4, F5, Ic2, Ir2, to one stock only. The fatty acid method, however, divided these in three different stocks: stocks 1, 3 and 4, comprising S. mentella from the Faroe shelf (F1, F4, F5) and the Norwegian coast (N1, N2), from the northwestern shelf of Iceland (Ic2), and from the southern part off the Irminger Sea (Ir2), respectively. In addition, the fatty acid method showed that stocks 3 and 4 are more closely related to stock 1 than to stock 2 (Figures 4 and 5).
|
|
It thus appears that the discrimination power of the fatty acid method is superior to the enzyme method. This was also the case for two acknowledged stocks of cod, inhabiting the Faroe Plateau and the Faroe Bank, where electrophoretic studies displayed no significant difference in allele frequencies between the two stocks (Magnussen, 1996). On the other hand, the fatty acid profile in the heart tissue differed significantly between the two (Joensen et al., 2000).
ICES has acknowledged four tentative stock-complexes of S. mentella in the North Atlantic. These are: one in the Northeast Arctic within the Norwegian Economic Zone and the Fishery Zone at Svalbard, another in the waters off the Faroe Islands, Iceland, and East Greenland, a third in the upper layer (0–500 m) of the open Irminger Sea, called oceanic S. mentella, and a fourth in the deeper layer (>500 m) of the open Irminger Sea, named deep-sea S. mentella. Yet the division in oceanic and deep-sea S. mentella in the Irminger Sea is disputed and a relationship between deep-sea S. mentella in the Irminger Sea and S. mentella in waters off the Faroe Islands, Iceland, and East Greenland is not ruled out (Reinert, 1993; Anon., 1998).
According to the ICES's view the redfish from the 11 sampling areas should be divided into the following four populations: (N1+N2), (F1+F2+F3+F4+F5+Ic1+Ic2), (Ir1 deep-sea S. mentella), and (Ir2 oceanic S. mentella). This is different from the stock structure which is revealed by fatty acid profile method in the present investigations, i.e. (N1+N2+F1+F4+F5), (F2+F3+Ic1+Ir1), (Ic2), and (Ir2 oceanic S. mentella).
The main discrepancy is that the fatty acid profile method divides S. mentella in the waters off the Faroe Islands into two stocks. One is inhabiting the Faroe Plateau (F1+F4+F5). The other stock is inhabiting the depth south-west off the Faroe Bank (F2) and the Wyville Thomson Ridge (F3). The second of these two stocks extends to the Icelandic Plateau (Ic1) and the deep layer of the Irminger Sea (Ir1). The Faroe Plateau stock (F1+F4+F5) extends to the Norwegian Sea (N1+N2) (Figures 2–4). This result is partly supported by a pilot study, based on Cesium-137 content in redfish, which indicated a relationship between S. mentella on the Faroe Plateau and S. mentella in the Barents Sea (Reinert et al., 1992).
Another deviation is the revelation of two significantly different stocks of S. mentella on the Icelandic Plateau. S. mentella on the shelves or plateaux and banks around the Faroe Islands, Iceland, and at East Greenland is treated as one stock unit (Anon., 1998), i.e. no sub-structuring on the Icelandic Plateau. Other fish species, however, like cod (Gadus morhua), are shown to be subdivided into genetically distinguishable groups in Icelandic waters (Jónsdóttir, 2001).
The differentiation between deep-sea and oceanic S. mentella, as established by other methods (Magnusson and Magnusson, 1995; Johansen et al., 2000b), was also substantiated by the fatty acid profile method (Figure 4).
Many families of fishes evolved in the North Pacific Ocean are presumed to have entered the temperate areas of the North Atlantic Ocean when the Bering Strait opened between three and a half and three million years ago (Grant, 1987). Great uniformity within and among the three Sebastes species in the North Atlantic is demonstrated, and it is hypothesized that the Atlantic Sebastes genus has bottlenecked from a restricted range of Pacific genetic variation (Nedreaas and Nævdal, 1991a). Moreover, close resemblance of the frequency distribution of tandem repeat numbers in redfish mtDNA indicates a close relationship among the Atlantic species of Sebastes (Bentzen et al., 1998). In addition, the broadscale introgressive hybridization between S. mentella and S. fasciatus (Roques et al., 2001) and the occurrence of several peculiar specimens from waters off the Faroe Islands could indicate reproductive barriers too low to block the gene pools entirely. The speciation process is perhaps not completed yet. It may therefore be difficult to find a single method, which is capable of giving a clear-cut discrimination at species and stock levels in all habitats. However, by developing new analytical methods (Rehbein et al., 1997; Piñeiro et al., 1999; Hold et al., 2000) and by using several different well established and new methods in coordinated collective investigations, it should be possible to get a valid picture of the stock structure of S. mentella in the North Atlantic.
The highly variable annual catches of the international fleet, ranging from 105 000 to 37 000 tons in the Irminger Sea (Johansen et al., 2000b), from 80 000 to 20 000 tons in the waters off the Faroe Islands, Iceland, East Greenland, and Rockall/Hatton Bank (Reinert, 1998), and from 269 000 to 8000 tons in the Northeast Arctic (ICES, 2000), as well as indications of overexploitation, and the fact that this redfish species is foraging on the feeding grounds extending several territorial waters are factors that have prompted concerted international efforts to unveil the stock structure of S. mentella. This clarification is crucial for a skilful management of these populations.
|
Conclusion |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionConclusionReferences |
|---|
This investigation has revealed a stock structure comprising four separate stocks of Sebastes mentella in the North Atlantic. One stock on the Faroe Plateau (F1, F4, F5) and in the Norwegian Sea (N1, N2), another extending from the Wyville Thomson Ridge (F3) and the depth (F2) south-west off the Faroe Bank to the southern part of the Icelandic Plateau (Ic1) and the deep-sea layer of the Irminger Sea (Ir1), a third on the eastern part of the Icelandic Plateau (Ic2), and a fourth in the oceanic layer of the Irminger Sea (Ir2). This stock structure corresponds partially with preliminary results of enzyme-electrophoresis carried out in another laboratory, and with earlier interpretations of results from multivariate variance analysis of morphometric measurements.
This investigation has resulted in the following statistically founded new knowledge:
- A clear division between S. mentella on the Faroe Plateau (F1, F4, F5), on one hand, and S. mentella in the depth (F2) south-west off the Faroe Bank and on the Wyville Thomson Ridge (F3), on the other.
- A relationship between S. mentella on the Faroe Plateau (F1, F4, F5) and S. mentella in the Norwegian Sea (N1, N2).
- The identification of two significantly different stocks of S. mentella on the Icelandic Plateau (Ic1, Ic2).
Chemometry of tissue fatty acid profiles may be an alternative tool to many other established and newly developed methods in discrimination at the species-level, stock-level, and individual level. However, possible impacts of biotic and abiotic factors on the stability of the fatty acid profile in selected tissues need further investigations.
|
Acknowledgements |
|---|
We want to thank Fisheries Laboratory of the Faroes and The Faroese Research Council for financial support and the managing director, Hjalti í Jákupsstovu, and the assistant director, Jákup Reinert, of Fisheries Laboratory of the Faroes for initiating this research work. Torild Johansen and Gunnar Nævdal (Department of Fisheries and Marine Biology, Bergen, Norway), Jákup Reinert, (The Fisheries Laboratories of the Faroes, Tórshavn, Faroes Islands), Anna Kristín Daniélsdóttir (Marine Research Institute, Population Genetics Laboratory, Reykjavík, Iceland), and Christoph Stransky (Institute for Sea Fisheries, Hamburg, Germany) are thanked for their benevolent participation in this project. Kjell Nedreaas (Institute of Marine Research, Bergen, Norway) and Thorsteinn Sigurdsson (Marine Research Institute, Reykjavík, Iceland) are appreciated for the provision of specimens of S. mentella. Ann-Kristin Halvorsen (Department of Chemistry, University of Bergen, Norway), Birita Jacobsen, Marit Pedersen, Rógvi Mouritsen (Fisheries Laboratory of the Faroes), and the crews on Magnus Heinason, Brettingur, Johan Hjort, G. O. Sars, Akureyrin, Vestmannaey are gratefully acknowledged for skillful technical assistance.
|
References |
|---|
|
TopIntroductionMaterials and methodsResultsDiscussionConclusionReferences |
|---|
-
Alekseev F. E. (1999) On population structure of redfish (Sebastes mentella Travin) in the Northwestern Atlantic. ICES WGAGFM meeting in Reykjavík12–15 April 1999Iceland.
Altukhov J.P. and Nefyodov G.N. (1968) A study of blood serum protein composition by agar-gel electrophoresis in types of redfish (genus Sebastes). ICNAF Research Bulletin 5:86–90.
Altukhov J.P., Nefyodov G.N., Payusova A.N. (1968) Thermostability of isolated muscle in determining the taxonomic relationship of the marinus- and mentella-types of redfish (Sebastes). Research Bulletin – International Commission for the Northwest Atlantic Fisheries 5:130–136.
Anderson A.J. and Arthington A.H. (1989) Effect of dietary lipid on the fatty acid composition on silver perch (Leiopotherapon bidyanus) lipids. Comparative Biochemistry and Physiology Part B 93:715–720.[CrossRef]
Anon. (1998) Report on the Study Group on Redfish Stocks28–30 January 1998Hamburg, Germany ICES CM 1998/G3, Ref. H.
Bandarra N.M., Batista I., Nunes M.L., Empis J.M., Christie W.W. (1997) Seasonal changes in lipid composition of sardine (Sardina pilchardus). Journal of Food Science 62:40–42.[CrossRef][Web of Science]
Bell M.V., Henderson R.J., Sargent J.R. (1983) Changes in the fatty acid composition of phospholipids from turbot (Scophthalmus maximus) in relation to dietary polyunsaturated fatty acid deficiencies. Comparative Biochemistry and Physiology Part B 81:193–198.[CrossRef]
Bentzen P., Wright J.M., Bryden L.T., Sargent M., Zwanenburg K.C.T. (1998) Tandem repeat polymorphism and heteroplasmy in the mitochondrial control region of redfishes (Sebastes: Scorpaenidae). Journal of Heredity 89:1–7.
Bogovski S.P. and Bakai Y.I. (1989) Chromastoblastomas and related pigmented lesions in deepwater redfish, Sebastes mentella (Travin), from North Atlantic areas, especially the Irminger Sea. Journal of Fish Diseases 12:1–13.[Medline]
Cossins A.R. and Macdonald A.G. (1986) Homeoviscous adaptation under pressure. III. The fatty acid composition of liver mitochondrial phospholipids of deep-sea fish. Biochimica et Biophysica Acta 860:325–335.
Cowley C.B., Wee K.L., Tacon A.G.J. (1983) Effect of fatty acid intake on growth and fatty acid composition of liver and muscle of snakehead. Bulletin of the Japanese Society of Scientific Fisheries 49:1573–1577.[Web of Science]
Desrosiers B., Sévigny J.M., Chanut J.P. (1999) Restriction fragment length polymorphism of rDNA in the redfishes Sebastes fasciatus and S. mentella (Scorpaenidae) from the Gulf of St. Lawrence. Canadian Journal of Zoology 77:267–277.
Dulavik B., Sørensen N.K., Barstad H., Horvli O., Olsen R.L. (1998) Oxidative stability of frozen light and dark muscles of saithe (Pollachius virens L.). Journal of Food Lipids 5:233–245.[Web of Science]
Gorelova T. A. (1997) Feeding of immature redfish in the Greenland shelf area. ICES CM/GG: 05.
Grahl-Nielsen O. and Mjaavatten O. (1992) Discrimination of striped bass stocks: a new method based on chemometry of the fatty acid profile in heart tissue. Transaction of the American Fisheries Society 121:307–314.[CrossRef]
Grahl-Nielsen O. and Ulvund K.A. (1990) Distinguishing populations of herring by chemometry of fatty acids. American Fisheries Society Symposium 7:566–571.
Grant W.S. (1987) Genetic divergence between congeneric Atlantic And Pacific Ocean fishes. In Ryman N. and Utter F. (Eds.). Population Genetics & Fishery Management(University of Washington Press, Seattle) pp. 225–246 420 pp.
Greene D.H.S. and Selivonchick D.P. (1990) Effects of dietary vegetable, animal and marine lipids on muscle lipid and hematology of rainbow trout (Oncorhynchus mykiss). Aquaculture 89:165–182.[CrossRef][Web of Science]
Hansen B. (2000) The ocean. School-Book Foundation of the Faroes, Thorshaven (in Faroese) 232 pp.
Hold G.L., et al. (2001) Development of a DNA-based method aimed at identifying the fish species present in food products. Journal of Agricultural and Food Chemistry 49:1175–1179.[CrossRef][Web of Science][Medline]
ICES. (2000) Report of the Arctic Fisheries Working Group23 August–1 September 1999Copenhagen ICES CM 2000/Assess: 3. 312 pp.
Joensen H. and Grahl-Nielsen O. (2000) Discrimination of Sebastes viviparus, S. marinus and S. mentella from Faroe Islands by chemometry of the fatty acid profile in heart and gill tissues and in the skull oil. Comparative Biochemistry and Physiology Part B 126:69–79.[CrossRef]
Joensen H. and Grahl-Nielsen O. (2001) The redfish species Sebastes viviparus, Sebastes marinus and Sebastes mentella have different composition of their tissue fatty acids. Comparative Biochemistry and Physiology Part B 129:73–85.[CrossRef]
Joensen J., Steingrund P., Fjallstein I., Grahl-Nielsen O. (2000) Discrimination between two reared stocks of cod (Gadus morhua) from the Faroe Islands by chemometry of the fatty acid composition in the heart tissue. Marine Biology 136:573–580.[CrossRef]
Johansen T., Nævdal G., Daniélsdóttir A.K., Hareide N.R. (2000) Genetic characterization of giant Sebastes in the deep water slopes in the Irminger Sea. Fisheries Research 45:207–216.[CrossRef][Web of Science]
Johansen T., Daniélsdóttir A.K., Meland K., Nævdal G. (2000) Studies of the genetic relationship between deep-sea and oceanic Sebastes mentella in the Irminger Sea. Fisheries Research 49:179–192.[CrossRef][Web of Science]
Jónsdóttir Ó. D. B. (2001) Atlantic cod (Gadus morhua L.) in the North Atlantic with emphasis on Icelandic waters; population genetic structure, temporal stability and genotypic dependent growth properties. PhD thesis, Department of Fisheries and Marine Biology, University of Bergen, Norway.
Kirsch P.E., Iverson S.J., Bowen W.D., Kerr S.R., Ackman R.G. (1998) Dietary effects on the fatty acid signature of whole Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences 55:1378–1386.
Kvalheim O.M. and Karstang T.V. (1987) A general-purpose program for multivariate data analysis. Chemometrics And Intelligent Laboratory Systems 2:235–237.[CrossRef][Web of Science]
Leray C. and Pelletier X. (1985) Fatty acid composition of trout phospholipids: effect of (n-3) essential fatty acid deficiency. Aquaculture 50:51–59.[CrossRef][Web of Science]
Lie Ø., Lied E., Lambertsen G. (1986) Liver retention of fat and of fatty acids in cod (Gadus morhua) fed different oils. Aquaculture 59:187–196.[CrossRef][Web of Science]
Macdonald A.G. and Cossins A.R. (1985) In Laverack M.C. (Ed.). The theory of homeoviscous adaptation of membranes applied to deep-sea animals. Physiological Adaptations of Marine Animals vol. 39:301–322 Symposia of the Society for Experimental Biology.
MacFarlane R.B. and Bowers M.J. (1995) Matrotrophic viviparity in the yellowtail rockfish Sebastes flavidus. The Journal of Experimental Biology 198:1197–1206.[Web of Science][Medline]
MacFarlane R.B., Norton E.C., Bowers M.J. (1993) Lipid dynamics in relation to the annual reproductive cycle in yellowtail rockfish (Sebastes flavidus). Canadian Journals of Fisheries and Aquatic Sciences 50:391–401.
Magnussen E. (1996) Electrophoretic studies of cod (Gadus morhua) from Faroe Bank and Faroe Plateau compared with results found in other distribution areas. ICES CM 1996/G: 10.
Magnusson J. and Magnusson J.V. (1995) Oceanic redfish (Sebastes mentella) in the Irminger Sea and adjacent waters. Scientia Marina 59:241–254.[Web of Science]
Mjaavatten O., Levings C.D., Poon P. (1998) Variation in the fatty acid composition of juvenile chinook and coho salmon from Fraser river estuary determined by multivariate analysis; role of environment and genetic origin. Comparative Biochemistry and Physiology Part B 120:291–309.[CrossRef]
Morris R.J. and Culkin F. (1989) Fish. In Ackman R.G. (Ed.). Marine Biogenic Lipids, Fats and Oils(CRC Press, Boca Raton, Florida) vol. 2: pp. 145–178 ch. 12, 515 pp.
Muje P., Ågren J.J., Lindqvist O.V., Hänninen O. (1988) Fatty acid composition of vendace (Coregonus albula L.) muscle and its plankton feed. Comparative biochemistry and Physiology Part B 92:75–79.
Navarro J.C., McEvoy L.A., Amat F., Sargent J.R. (1995) Effects of diet on fatty acid composition of body zones in larvae of the sea bass Dicentrarchus labrax: a chemometric study. Marine Biology 124:177–183.[CrossRef]
Nedreaas K. and Nævdal G. (1989) Studies of Northeast Atlantic species of redfish (genus Sebastes) by protein polymorphism. Journal du Conseil International pour l'Exploration de la Mer 46:76–93.
Nedreaas K. and Nævdal G. (1991) Genetic studies of redfish (Sebastes spp.) along the continental slopes from Norway to East Greenland. ICES Journal of Marine Science 48:173–186.
Nedreaas K. and Nævdal G. (1991) Identification of 0- and 1-group redfish (genus Sebastes) using electrophoresis. ICES Journal of Marine Science 48:91–99.
Norton E.C. and MacFarlane R.B. (1995) Nutritional dynamics of reproduction in viviparous yellowtail rockfish, Sebastes flavidus. Fishery Bulletin 93:299–307.[Web of Science]
Owen J.M., Adron J.W., Sargent J.R., Cowey C.B. (1972) Studies on the nutrition of marine flatfish. The effect of dietary fatty acids on the tissue fatty acids of the plaice Pleuronectes platessa. Marine Biology 13:160–166.[CrossRef]
Patton J.S. (1975) The effect of pressure and temperature on phospholipid and triglyceride fatty acids of fish white muscle: a comparison of deepwater and surface marine species. Comparative Biochemistry and Physiology Part B 52:105–110.[CrossRef]
Phleger C.F. and Laub R.J. (1989) Skeletal fatty acids in fish from different depths off Jamaica. Comparative Biochemistry and Physiology Part B 94:2329–334.[CrossRef]
Pickova J., Dutta P.C., Larsson P.O., Kiessling A. (1997) Early embryonic cleavage pattern, hatching success, and egg-lipid fatty acid composition: comparison between two cod (Gadus morhua) stocks. Canadian Journal of Fisheries and Aquatic Science 54:2410–2416.[CrossRef]
Piñeiro C., et al. (1999) Development of a sodium dodecyl sulfate-polyacrylamide gel electrophoresis reference method for the analysis and identification of fish species in raw and heat-processed samples: a collaborative study. Electrophoresis 20:1425–1432.[CrossRef][Web of Science][Medline]
Pitcher T.J. and Hart P.J.B. (1982) Fisheries Ecology(Croom Helm Ltd, Amsterdam).
Rehbein H. (1990) Electrophoretic techniques for species identification of fishery products. Lebensmittel Untersuchung und Forschung 191:1–10.[CrossRef]
Rehbein H., Kress G., Schmidt T. (1997) Application of PCR-SSCP to species identification of fishery products. Journal of the Science of Food and Agriculture 74:35–41.[CrossRef][Web of Science]
Reinert J. (1993) Redfish – not just redfish. Frøði 1:10–18 (in Faroese).
Reinert J. (1998) In Olsen D. (Ed.). Redfish. Fish-Stocks and Environment 48–51 School-Book Foundation of the Faroes, Thorshaven (in Faroese) 92 pp.
Reinert J. and Lastein L. (1992) Stock identification of S. marinus L. and S. mentella Travin in the Northeast-Atlantic based on meristic counts and morphometric measurements. ICES CM G: 29. 21 pp.
Reinert J., Hansen B., Joensen H. P. (1992) Stock identification of S. marinus L. and S. mentella Travin in the Norteast-Atlantic based on measurement of Cs-137 content in the fish. ICES CM/G: 28. 8 pp.
Roques S., Sévigny J.M., Bernatchez L. (2001) Evidence for broadscale introgressive hybridization between two redfish (genus Sebastes) in the North-west Atlantic: a rare marine example. Molecular Ecology 10:149–165.[CrossRef][Medline]
Saborido-Rey F. and Nedreaas K.H. (2000) Geographic variation of Sebastes mentella in the Northeast Arctic derived from a morphometric approach. ICES Journal of Marine Science 57:965–975.
Sévigny J.M., Gagné P., Lafontaine Y., Dodson J. (2000) Identification and distribution of larvae of redfish (Sebastes fasciatus and S. mentella: Scorpaenidae) in the Gulf of St. Lawrence. Fishery Bulletin 98:375–388.[Web of Science]
Sundt R.C. and Johansen T. (1998) Low levels of interspecific DNA sequence variation of the mitochondrial 16S rRNA in North Atlantic redfish Sebastes (Pisces, Scorpaenidae). Sarsia 83:449–452.[Web of Science]
Ugland K.I. and Massart B.G.J. (1996) Classification. In Nortvedt R. (Ed.). Application of Chemometry in Science and IndustryNorwegian Chemical Society & The Swedish Chemical Society pp. 153–167 615 pp. (in Norwegian).
Viga A. and Grahl-Nielsen O. (1990) Genotypic and phenotypic fatty acid composition in the tissues of salmon, Salmo salar. Comparative Biochemistry and Physiology Part B 96:4721–727.[CrossRef]
Wold S. (1976) Pattern recognition by means of disjoint principal components models. Pattern Recognition 8:127–139.[CrossRef][Web of Science]
Wold S. (1978) Cross validatory estimation of the number of components in factor and principal models. Technometrics 20:397–406.[CrossRef][Web of Science]
Wold S. and Sjøstrøm M. (1977) SIMCA: a method for analysing chemical data in terms of similarity and analogy. In Kowalski B. (Ed.). ACS Symposium Series No. 52(American Chemical Society, Washington, DC) pp. 243–282 Chemometrics: Theory and Application.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. O. Stefansson, J. Reinert, T. Sigurthsson, K. Kristinsson, K. Nedreaas, and C. Pampoulie Depth as a potential driver of genetic structure of Sebastes mentella across the North Atlantic Ocean ICES J. Mar. Sci., May 1, 2009; 66(4): 680 - 690. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||