**Comparison of Fatty Acid Contents in Major Lipid Classes of Seven Salmonid Species from Siberian Arctic Lakes**

**Nadezhda N. Sushchik 1,2,\*, Olesia N. Makhutova 1,2, Anastasia E. Rudchenko 1,2, Larisa A. Glushchenko 2, Svetlana P. Shulepina 2, Anzhelika A. Kolmakova <sup>1</sup> and Michail I. Gladyshev 1,2**


Received: 24 January 2020; Accepted: 5 March 2020; Published: 8 March 2020

**Abstract:** Long-chain omega-3 polyunsaturated fatty acids (LC-PUFA) essential for human nutrition are mostly obtained from wild-caught fish. To sustain the LC-PUFA supply from natural populations, one needs to know how environmental and intrinsic factors affect fish fatty acid (FA) profiles and contents. We studied seven Salmoniformes species from two arctic lakes. We aimed to estimate differences in the FA composition of total lipids and two major lipid classes, polar lipids (PL) and triacylglycerols (TAG), among the species and to evaluate LC-PUFA contents corresponding to PL and TAG in muscles. Fatty acid profiles of PL and TAG in all species were characterized by the prevalence of omega-3 LC-PUFA and C16-C18 monoenoic FA, respectively. Fish with similar feeding spectra were identified similarly in multivariate analyses of total lipids, TAG and PL, due to differences in levels of mostly the same FA. Thus, the suitability of both TAG and total lipids for the identification of the feeding spectra of fish was confirmed. All species had similar content of LC-PUFA esterified as PL, 1.9–3.5 mg g<sup>−</sup>1, while the content of the TAG form strongly varied, from 0.9 to 9.8 mg g<sup>−</sup>1. The LC-PUFA-rich fish species accumulated these valuable compounds predominately in the TAG form.

**Keywords:** arctic; Salmoniformes; long-chain polyunsaturated fatty acids; polar lipids; triacylglycerols; eicosapentaenoic acid; docosahexaenoic acid

### **1. Introduction**

Long-chain omega-3 polyunsaturated fatty acids (LC-PUFA), such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are known to be essential compounds for human nutrition, since they can modulate the functioning of cardiovascular and neural systems and general metabolism, being the precursors for the synthesis of diverse lipid mediators and directly affecting membrane properties [1–5]. Most international and national health agencies and foundations recommended personal consumption of 0.5–1.0 g of EPA+DHA per day for reducing the risk of cardiovascular diseases and other metabolic disorders [6–8]. Although a lot of potential sources of LC-PUFA are now being considered, natural fish populations are still the major source of these compounds for human nutrition [9,10]. Recent reviews showed the deficiency of the LC-PUFA supply with fish caught from natural populations and emphasized the potential negative influence of some global threats, like climate change, pollution,

eutrophication, etc. [10–13]. To challenge above threats and to sustain the LC-PUFA supply from natural populations, one needs to know how environmental and intrinsic factors affect fish fatty acid profiles and content, including those of EPA and DHA. Causes of variations of fatty acid composition and content in wild fish are still incompletely understood [10,14].

The ability of fish to deposit fat (lipids) in muscles varies from species to species and may be a crucial intrinsic factor [15]. According to their functions, lipids in fish, like in other animals, could roughly be divided into energy-reserve and membrane-structural groups [16,17]. Fish reserve lipids are primarily represented by triacylglycerols (TAG) and include mostly fatty acids that come from food sources. Fatty acid profiles of TAG are generally considered as valuable trophic markers due to their resemblance with fatty acid profiles of particular food sources [18]. In addition, the TAG fraction in fish can also contain high levels of monoenoic C16-C18 fatty acids that are intensively synthesized in so called "fatty" fish species to provide energy reserves. TAG molecules are accumulated either directly in muscle cells as droplets or in specific adipocytes which may be integrated in muscle tissues or form separate layers of adipose tissue.

The structural polar lipids (PL) that form fish cellular and intracellular membranes mostly comprise phospholipids [19,20]. As known, fatty acid composition of PL affects physico-chemical properties of cellular membranes. Hence, PL are considered to have conservative fatty acid profiles which slightly reflect that of diet. The essential omega-3 LC-PUFA are preferentially accumulated in PL fraction of muscle tissues due to their strong membrane-modulating properties. Thus, fatty acid profiles of the major lipid classes, TAG and PL, in fish muscles are different in general [19].

TAG content per mass unit of fish muscles is highly variable due to influence of many factors [15]. In contrast, PL content per mass unit of muscles is fairly constant [21]. Thereby, we hypothesize that PL specific content has a putative upper threshold, because amounts of PL molecules in tissue are likely determined by a volume of membranes.

Contents of omega-3 LC-PUFA in muscle tissue of diverse fish species greatly vary, approximately ~400-fold [22]. The question arises what part of this variation in total EPA and DHA contents is provided by TAG or PL variability? There is a basic assumption in the current literature that a major part of omega-3 LC-PUFA presents as acyl groups of membrane phospholipid molecules [11].

To evaluate contribution of the two major lipid fractions in total content of LC-PUFA in edible muscle tissue (filets) we studied seven commercial species of the order Salmoniformes that inhabit oligotrophic non-polluted lakes in Arctic Siberia. The fish species vary in their feeding habits and habitats and have different fat content in filets. Using data on these fish we aimed to compare distribution of fatty acids, including omega-3 LC-PUFA in total lipids and two major lipid classes: TAG and PL. Specifically, we aimed (i) to check if the fish species with various feeding spectra can be differentiated basing on FA profiles of total lipids, TAG or PL, (ii) to evaluate LC-PUFA content corresponded to TAG and PL classes in muscles, (iii) to range species according to their nutritive value for humans in respect of LC-PUFA content.

### **2. Materials and Methods**

### *2.1. Sampling*

Fish specimen of commercial sizes were collected during July 2017 from catches of local authorized fishers. Following sampling was done in accordance with the BioEthics Protocol on Animal Care approved by Siberian Federal University. The catches were from two oligotrophic arctic lakes, Sobachye and Pyasino. Sobachye Lake was previously characterized elsewhere [23]. Briefly, it is located at 69◦01 N 91◦05 E and has maximum depth of 162 m and area equal to 99 km2. Pyasino Lake situates at 69◦40 N 87◦51 E, has average depth of 4 m and an area equal to 735 km2 [24].

Whitefish *Coregonus lavaretus* (Linnaeus, 1758), non-identified form of whitefish *C. lavaretus*, round whitefish *Prosopium cylindraceum* (Pennant, 1784) and charr *Salvelinus drjagini* Logashev, 1940 were caught in Sobachye Lake; broad whitefish *Coregonus nasus* (Pallas, 1776), muksun *Coregonus muksun* (Pallas, 1814) and inconnu *Stenodus leucichthys nelma* (Guldenstadt, 1772) were caught in Pyasino Lake. All studied fish species are of the Salmoniformes order. Numbers of samples, average individual sizes and weights, and main food sources for the studied fish species are given in Table 1. Stomach contents of all specimen were studied under a light microscope, and main food items were identified to a possible taxon level.

**Table 1.** The basic biological and sampling information on fish species (Salmoniformes order) from Siberian arctic lakes, 2017: *n*—number of sampled individuals; L—total length, cm (mean ± SE); W—total weight, g (mean ± SE); Food—items found in stomachs.


For biochemical analyses, we cut slices of fish white muscles of approximately 2–3 g, 2–3 cm below the dorsal fin. The samples were subdivided into two parts: for FA and moisture analyses. For FA analyses, ca. 1 g of muscle tissues was immediately placed into a volume of 3 mL of chloroform/methanol (2:1, by vol.) and kept until further analysis at −20 ◦C. Another subsample of ca. 1–2 g of wet weight was weighed, dried at 105 ◦C until constant weight, and weighed dry for moisture calculation.

### *2.2. Lipid and Fatty Acid Analyses*

In laboratory, lipids were extracted with chloroform/methanol (2:1, by vol.) in triplicate, simultaneously with homogenizing tissues with glass beads in a mortar. Prior to the extraction, an aliquot of 19:0-fatty acid methyl ester (FAME) chloroform solution, as the internal standard, was added to samples for quantification of chromatographic peaks. The extracts were combined and dried with anhydrous Na2SO4 and the solvents were roto-evaporated under vacuum at 35 ◦C. The extracted lipids were redissolved in a 1 mL portion of chloroform and separated in two equal parts. To analyze the fatty acid composition of total lipids, one part of the lipid extract was methylated in the following way. The lipids were hydrolysed under reflux at 90◦ C for 10 min in 0.8 mL of methanolic sodium hydroxide solution (8 g/L). Then the mixture was cooled for 5 min at room temperature. Next, 1 mL of methanol/sulphuric acid (97:3, by vol.) was added and the mixture was heated under reflux at 90 ◦C for 10 min to methylate free fatty acids. At the end, 5 mL of saturated solution of NaCl and 3 mL of hexane were added. The FAMEs were extracted for 1 min, the mixture was transferred to a separatory funnel, and the lower aquatic layer was discarded. The hexane layer was additionally washed once with an aliquot of the NaCl solution and twice with 5 mL of distilled water. Then the hexane solution of FAMEs was dried with anhydrous Na2SO4, and hexane was removed by roto-evaporating at 30 ◦C.

We fractionated another part of the lipid extracts by thin layer chromatography (TLC) on silica gel G with hexane-diethyl ether-acetic acid mixture (85:15:1, by vol.). We prepared a reference mixture containing triolein, oleic acid, cholesterol, and phosphatidylcholine (Sigma, St. Louis, MO, USA), which was applied on the side lanes of the silica gel plates. To identify the lipid composition of fish species, we applied aliquots of samples to the plates, and then developed them as described above. After developing, the plates were sprayed with mixture of ethanol/sulphuric acid (90:10, by vol.) and gently heated until grey spots of lipid classes appeared. To evaluate fatty acid profiles and quantify

dominant lipid classes, we separated a main portion of lipid extracts on the silica gel plates, and then visualized only the side lanes corresponding to the reference–compound mixture by a reaction with phosphomolybdic acid ethanolic solution. Lipid spots of the fish samples on the plates were blind detected according to positions of the reference compounds. The lipid spots containing TAG and PL fractions were scraped off from the silica gel plates and placed in flasks. Aliquots of 19:0-FAME hexane solution (t internal standard) were added into the flasks containing silica gel powder with the lipid fractions. To prepare FAMEs, 1ml of hexane and 0.2 mL of fresh 3 M sodium methoxide methanolic solution was added, and the mixture was shaken vigorously for 1 min. Subsequently, the mixture was kept calm at room temperature for 5 min, and finally 3 mL of hexane and 5 mL of a saturated solution of NaCl were added. The next procedure of FAME extraction and washing was the same as for those prepared from total lipids.

Analyses of all FAMEs were done with a gas chromatograph equipped with a mass spectrometer detector (model 6890/5975C; Agilent Technologies, Santa Clara, CA, USA) and with a 30-m long, 0.25-mm internal diameter capillary HP-FFAP column. Detailed descriptions of the chromatographic and mass-spectrometric conditions are given in [23].

### *2.3. Statistical Analysis*

The Kolmogorov–Smirnov one-sample test for normality *D*K-S, standard errors (SE), Student's *t*-tests, one-way ANOVA with post hoc Tukey HSD test, Kruskal–Wallis test (in the absence of normal distribution) and canonical correspondence analysis (CCA) were calculated conventionally, using STATISTICA software, version 9.0 (StatSoft Inc., Tulsa, OK, USA).

### **3. Results**

According to gut content analysis, *C. lavaretus* in Sobachye Lake was benthivorous (Table 1). Whitefish of a non-identified form in this lake consumed mostly pupa and adult insects, i.e., foraged near the water surface. Round whitefish fed on benthic invertebrates and algae (Table 1). Both broad whitefish and muksun in Pyasino Lake consumed benthic food items, including detritus. Charr in Sobachye Lake and inconnu in Pyasino Lake were piscivorous (Table 1).

Average values of moisture content, lipid content and sum of fatty acid content for total lipids in the studied fish are given in Table 2. Lower moisture values were characteristic of the species with higher values of lipid and sum FA content, charr and whitefish (Table 2). In contrast, round whitefish and whitefish of the non-identified form had the maximum moisture content and the minimal contents of lipids and sum FA. Sum FA content of total lipids significantly varied ~7-fold among the studied species (Table 2). Based on the averages of lipid and sum FA contents, charr and whitefish are further considered as "fatty" fish, muksun, inconnu and broad whitefish as "medium fat" fish, and round whitefish and the non-identified form of whitefish as "lean" fish (Table 2).

**Table 2.** Average (<sup>±</sup> SE—standard errors) moisture content (% wet weight), lipid content (mg g−<sup>1</sup> wet weight) and sum fatty acid content for total lipids (mg g−<sup>1</sup> wet weight) in muscle tissues of fish species caught in Siberian arctic lakes, 2017. Means of total fatty acids labeled with the same letter are not significantly different at *P* < 0.05 after ANOVA post hoc Tukey HSD test.


n.d.—no data.

Levels of 25 prominent individual FA and their structural groups in total lipids are showed in Table 3. Charr had the highest levels of 20:2n-6, 20:3n-3, 20:4n-3, 22:4n-3 and C24 PUFA among the studied species (Table 3). Whitefish had the significantly highest levels of 16:1n-7 and C16 PUFA, and tended to be higher in levels of 18:1n-9 and 20:5n-3. Muksun tended to have higher levels of 14:0, 20:1 and 22:5n-6 (Table 3). Broad whitefish had higher levels of 16:1n-9, C15-17 BFA (branched-chain fatty acids), 18:0, 18:1n-7, 18:2n-6, 18:3n-3 (Table 3). Whitefish of the non-identified form had higher levels of 16:0 and 22:6n-3 compared to those of the other fish. Inconnu and round whitefish had intermediate levels of all FA in total lipids (Table 3).

**Table 3.** Mean levels of fatty acids (% of the total) in total lipids of species of Salmoniformes order: charr—*S. drjagini* from Sobachye Lake; whitefish—*C. lavaretus* from Sobachye Lake; muksun—*C. muksun* from Pyasino Lake; inconnu—*S. leucichthys nelma* from Pyasino Lake; broad whitefish—*C. nasus* from Pyasino Lake; round whitefish—*P. cylindraceum* from Sobachye Lake; whitefish nd—the non-identified form of *C. lavaretus* from Sobachye Lake. Cases (fatty acids) with normal distribution are given in bold. Means labeled with the same letter are not significantly different at *P* < 0.05 after ANOVA post hoc Tukey HSD test (cases with normal distribution) or Kruskal–Wallis test. If ANOVA is insignificant, letters are absent.


\* sum of 20:1n-11, 20:1n-9 and 20:1n-7, here and in other Tables.

We performed CCA of FA profiles of total lipids of the fish species to find out their overall differences (Figure 1). Along Dimension 1, most difference in FA composition was observed between charr, on the one hand, and whitefish, on the other hand. The differences were primarily caused by higher percentages of 22:4n-3, 20:3n-3 and C24 PUFA in charr, and higher percentages of C16 PUFA and 16:1n-7 in whitefish. In Dimension 2, whitefish of a non-identified form and round whitefish located at the one end and broad whitefish was at the other end (Figure 1). Whitefish of the non-identified form and round whitefish were separated due to higher levels of 22:6n-3 and 20:5n-3, and partial separation of broad whitefish was due to higher levels of C15-17 BFA (Table 3). Samples of the non-identified whitefish were markedly scattered (Figure 1) due to the high variability in percentages of 22:6n-3 (Table 3).

**Figure 1.** Canonical correspondence analysis of fatty acid percentages (% of FA sum) in total lipids of muscles of seven fish species from arctic lakes (Siberia, Russia). **A**—individual variables, **B**—factor structure coefficients for fatty acids. Dimension 1 and Dimension 2 represented 33.9% and 21.2% of inertia, respectively.

In PL of all studied species, 22:6n-3, 16:0 and 20:5n-3 were dominant fatty acids (Table 4). Charr had the highest levels of 22:6n-3, 22:4n-3, 22:5n-6, 20:4n-3 among the studied species (Table 4). Whitefish had a significantly higher level of 20:5n-3 compared to that of the other fish. Inconnu tended to be higher in 18:1n-9 level. Broad whitefish had the highest levels of 20:4n-6 and 18:2n-6 (Table 4). Round whitefish had significantly higher levels of 16:1n-7, 18:1n-7 and C16 PUFA compared to those of the other fish. Muksun and whitefish of the non-identified form had intermediate levels of most FA in PL (Table 4).

**Table 4.** Mean levels of fatty acids (% of the total) and total content of fatty acids (Sum FA, mg g−<sup>1</sup> wet weight) in polar lipids of Salmoniformes species: charr—*S. drjagini* from Sobachye Lake; whitefish—*C. lavaretus* from Sobachye Lake; muksun—*C. muksun* from Pyasino Lake; inconnu—*S. leucichthys nelma* from Pyasino Lake; broad whitefish—*C. nasus* from Pyasino Lake; round whitefish—*P. cylindraceum* from Sobachye Lake; whitefish nd—the non-identified form of *C. lavaretus* from Sobachye Lake. Cases (fatty acids) with normal distribution are given in bold. Means labeled with the same letter are not significantly different at *P* < 0.05 after ANOVA post hoc Tukey HSD test (cases with normal distribution) or Kruskal–Wallis test. If ANOVA is insignificant, letters are absent.


To reveal overall differences in PL FA, CCA was performed (Figure 2). In the first dimension, a conspicuous difference of round whitefish versus charr was found. This difference was provided mostly by the greater levels of C16 PUFA, 16:1n-7 and 18:2n-6 in PL of round whitefish and by greater levels of C24 PUFA and 22:4n-3 in that of charr (Figure 2). The variation in the second dimension of CCA was related to differences between whitefish and broad whitefish due to levels of 15:0 and 14:0 versus levels of C16 PUFA, 18:2n-6 and 18:3n-3.

In FA composition of triacylglycerols of all the studied arctic fish, 18:1n-9, 16:1n-7, 16:0 and 20:5n-3 dominated (Table 5). Charr had the highest levels of 22:6n-3, C24 PUFA, 22:5n-3, 20:4n-3 and 20:3n-3 compared to that of the other studied species. Whitefish had the maximum levels of 18:1n-9 and 20:5n-3 (Table 5). Muksun had the significantly higher level of 14:0 than the other fish. Inconnu and whitefish of the non-identified form had intermediate FA levels in TAG (Table 5). Broad whitefish had the maximum levels of 16:0, C15-17 BFA, 17:0, 18:2n-6, 18:3n-3, and 20:4n-6 among the studied fish. Round whitefish had the significantly higher percentages of 16:1n-7 and C16 PUFA in TAG (Table 5).

**Figure 2.** Canonical correspondence analysis of fatty acid percentages (% of FA sum) in polar lipids of muscles of seven fish species from arctic lakes (Siberia, Russia). **A**—individual variables, **B**—factor structure coefficients for fatty acids. Dimension 1 and Dimension 2 represented 30.9% and 22.5% of inertia, respectively.

**Table 5.** Mean levels of fatty acids (% of the total) and total content of fatty acids (Sum FA, mg g−<sup>1</sup> wet weight) in triacylglycerols of Salmoniformes species: charr—*S. drjagini* from Sobachye Lake; whitefish—*C. lavaretus* from Sobachye Lake; muksun—*C. muksun* from Pyasino Lake; inconnu—*S. leucichthys nelma* from Pyasino Lake; broad whitefish—*C. nasus* from Pyasino Lake; round whitefish—*P. cylindraceum* from Sobachye Lake; whitefish nd—the non-identified form of *C. lavaretus* from Sobachye Lake. Cases (fatty acids) with normal distribution are given in bold. Means labeled with the same letter are not significantly different at *P* < 0.05 after ANOVA post hoc Tukey HSD test (cases with normal distribution) or Kruskal–Wallis test. If ANOVA is insignificant, letters are absent.


To study differences in fish reserve lipids, we performed CCA of FA in TAG (Figure 3). Like the multidimensional analysis for PL, Dimension 1 showed a marked difference of round whitefish versus charr. This difference was provided mostly by the greater levels of C16 PUFA and 16:1n-7 in TAG of round whitefish and by greater levels of 22:4n-3 and 20:3n-3 in that of charr (Figure 3). The second dimension of CCA for TAG also showed a similar trend to that observed in CCA of PL (Figures 2 and 3). In this dimension, most prominent difference was found between whitefish and broad whitefish due to variability in levels of C15-17 BFA, 17:0 and C16 PUFA (Figure 3).

In general, positioning of fish species in the biplot for reserve TAG well corresponded to that in biplot for structural PL (Figures 2 and 3). It should be also noted that physiologically significant EPA and DHA were not found among the FA markers responsible for separation of fish species in CCA for TAG and PL (Figures 2 and 3). Positioning of fish species in the biplot for total lipids generally corresponded to that in biplots for the lipid classes, with exception of whitefish of the non-identified form (Figures 1–3). The fatty acid markers responsible for separation of the fish samples in CCA were generally similar for total lipids, TAG and PL with exception of DHA and EPA in CCA of total lipids (Figures 1–3).

A visual analysis of all thin-layer chromatograms showed a marked dominance of TAG, PL and sterols as major lipid fractions. Spots that corresponded to other lipid classes were negligible. Therefore, we considered TAG and PL as major acyl-containing fractions, summarized their FA contents per mass unit (Tables 4 and 5) and calculated their parts in the sum of FA in the fish muscles (Figure 4A). Polar lipids constituted from 10.3 to 57.0% of the acyl-containing lipid sum, being the highest in whitefish of

the non-identified form (Figure 4A). Triacylglycerols were the dominant acyl-containing lipid fraction for majority of the studied fish and exceeded 85% in charr, whitefish and muksun (Figure 4A). Note that increase in lipid content and total fatty acids for the studied species well corresponded with the increase in the TAG proportion of the acyl-containing lipids (Table 2, Figure 4A).

**Figure 3.** Canonical correspondence analysis of fatty acid percentages (% of FA sum) in triacylglycerols of muscles of seven fish species from arctic lakes (Siberia, Russia). **A**—individual variables, **B**—factor structure coefficients for fatty acids. Dimension 1 and Dimension 2 represented 51.1% and 15.6% of inertia, respectively.

**Figure 4.** Mean percentages of polar lipids and triacylglycerols of their sum—**A**, and mean contents (mg g−<sup>1</sup> wet weight) of sum of eicosapentaenoic and docosahexaenoic acids that corresponded to polar lipids and triacylglycerols—**B**, in muscles of seven fish species from arctic lakes (Siberia, Russia): nd whf—the non-identified form of *C. lavaretus* from Sobachye Lake, round whf—*P. cylindraceum* from Sobachye Lake, broad whf—*C. nasus* from Pyasino Lake, inconnu—*S. leucichthys nelma* from Pyasino Lake, muksun—*C. muksun* from Pyasino Lake, whf—*C. lavaretus* from Sobachye Lake, charr—*S. drjagini* from Sobachye Lake. Bars represent standard errors. Means labelled with the same letter are not significantly different at *P* < 0.05 after Tukey HSD post hoc test.

Using the PL and TAG percentages of the acyl-containing lipid sum and content of EPA and DHA of total lipids per mass unit, we calculated parts of EPA + DHA that provided by polar lipids versus triacylglycerols and expressed them as mg g−<sup>1</sup> wet weight of muscle tissue (Figure 4B). Contents of EPA + DHA provided by PL fraction varied from 1.9 to 3.5 mg g−<sup>1</sup> (Figure 4B). The average value for the seven fish species was 2.4 <sup>±</sup> 0.2 mg g<sup>−</sup>1, and coefficient of variation, CV, was 8.7%. Contents of EPA + DHA in TAG were of a greater range, from 0.9 to 9.8 mg g<sup>−</sup>1; the average value was to 4.4 <sup>±</sup> 0.2 mg g<sup>−</sup>1, and CV was 28.9% (Figure 4B).

### **4. Discussion**

### *4.1. Main Finding*

All the taxonomically related species of order Salmoniformes had nearly similar content of EPA+DHA in PL, 2.4 <sup>±</sup> 0.2 mg g−1, in average. In contrast, content of EPA + DHA esterified as TAG varied ~10-fold among the studied salmonids. Thus, all variations of nutritive value, i.e., EPA + DHA content per mass unit of filet, were caused by TAG fraction, while PL had a constant species (taxon—specific physiologically optimum content. As a result, fatty fish, charr, whitefish and muksun, contained most amounts of nutritionally valuable EPA and DHA in TAG fraction of muscles. Conversely, the lean species, round whitefish and whitefish of the non-identified form, had the omega-3 LC-PUFA contained mostly in polar lipids. Regarding nutritive value, the fatty species with higher proportion of TAG in muscles, charr, whitefish and muksun, appeared to be most valuable and had 13.3 <sup>±</sup> 0.8, 10.4 <sup>±</sup> 1.1 and 8.4 <sup>±</sup> 1.4 mg g<sup>−</sup>1, respectively.

### *4.2. Fatty Acid Markers in Fish Total Lipids, TAG and PL*

In our snap-shot field study of salmonids from arctic lakes we found significant differences between fatty acid profiles of the main lipid classes, with prevalence of n-3 LC-PUFA and monoenoic 16-18 FA in structural PL and storage TAG, respectively. In addition, both FA profiles of TAG and PL, as well as profiles of total lipids, had distinct peculiarities among the studied fish that allowed separating the most species in the CCA biplots (Figures 1–3). It should be emphasized that the separations of the fish in the multivariate analyses of all three lipid fractions were provided mainly by the same marker FA: C24 PUFA, 22:4n-3, 20:3n-3, C16 PUFA, 16:1n-7 and C15-17 BFA. The conspicuous exception was 22:6n-3 and 20:5n-3 in total lipid CCA that separated whitefish of the non-identified form and round whitefish from the other species (Figure 1). PL had a predominant contribution to total lipids of these lean fish (Figure 4A). Therefore, FA profiles of total lipids of whitefish of the non-identified form and round whitefish mostly reflected FA composition of PL which were considerably rich in DHA and EPA (Table 4). As a result, DHA and EPA, dominant FA of PL, played as markers for the lean fish in CCA of total lipids.

Similar patterns of biomarker FA, characteristic of zoobenthic, algal, terrestrial and other food sources, within TAG and PL fractions of the studied species allowed to use both these fractions, as well as their sum, total lipids, for identification of food sources of wild fish (Figures 1–3). In ecological studies, analysis of FA trophic markers of various consumers is often performed for TAG assuming that they generally deposit fatty acid molecules coming from food assimilation [18,25,26]. Meanwhile, many studies used FA composition of total lipids to elucidate trophic relations of various fish species, e.g., [27–31]. In overall, both approaches base on a premise that biochemical composition of food sources is reflected by FA profiles of TAG or total lipids.

Another lipid class of high concern is polar lipids comprised mostly phospholipids and glycolipids that are main constituents of cell membranes. As known, the specific FA composition of PL provides proper membrane structure and functions. As a result, FA composition of PL are considered to be highly conserved relative to diet and tended to reflect FA biosynthetic capacities of an organism [18].

In contrast to the common opinion on conserved PL composition, in some studies FA profiles of PL were successfully used as trophic markers. For instance, fatty acid profiles of both polar and neutral lipids conspicuously differed among three species, benthivorous whitefish *Coregonus clupeaformis*, and piscivorous walleye *Sander vitreus* and northern pike *Esox lucius*, due to different feeding habits of the fish [32]. We also confirmed that PL FA profiles of muscles allow to identify feeding spectra of fish similarly that TAG profiles do (Figures 2 and 3). For instance, round whitefish was one of the most separated species in the both multivariate analyses of TAG and PL due to the greatest levels of C16 PUFA and 16:1n-7 which originated from diatoms and green algae [33]. Indeed, the algae were one of the dominant items in stomach content of this species (Table 1).

In CCA biplots of total lipids, PL and TAG, charr had a particular position due to higher levels of minor n-3 PUFA, like 22:4-3, C24 PUFA and 20:3n-3. These FA were not assigned as trophic markers, whereas some of them, C24 PUFA, were considered as intermediate compounds indicative for conversion of C20 to C22 PUFA [34,35]. In TAG of charr the percentage of 22:4n-3 accounted for 1.3% of FA sum, being absent or found in traces in other studied fish. The presence of this PUFA was previously reported for least cisco *Coregonus sardinella*, small-sized pelagic fish inhabited Sobachye Lake [23]. The studied charr from Sobachye Lake was piscivorous (Table 1), thus, it could obtain this PUFA from the consumed least cisco. Like in our study, species of the same genus and its prey, lake trout *Salvelinus namaycush* and cisco from Great Bear Lake, were together separated from other hydrobionts in a multivariate analysis due to higher levels of 22:4-3 and 20:3n-3 [31]. Alternative explanation based on coincidence between 22:4n-3 and C24 PUFA levels is that the fatty acid 22:4n-3 may be a marker of LC-PUFA conversion in fish. Anyway, we suppose that considerable levels of 22:4n-3, 20:3n-3 and C24 PUFA might be a characteristic feature of FA profiles of *Salvelinus* genus.

In both CCA analyses of TAG and PL, broad whitefish well separated from the other species due to higher levels of C15-17 BFA, 17:0, 18:2n-6 and 18:3n-3. The two former FA are known to be markers of bacterial organic matter, while the two latter are considered as markers of terrestrial organic matter [31,33]. Broad whitefish is a typical benthivorous species that likely got these marker fatty acids from detritus enriched with bacterial and terrestrial organic matter. The species had the highest levels of 18:2n-6 in TAG, and 20:4n-6 in PL, relatively. This finding likely indicates for the initial storage of dietary 18:2n-6 in TAG and its consequent conversion to 20:4n-6 with further transfer to PL.

The similarity of FA sets that are markers for food sources between TAG and PL classes likely indicates that the studied wild fish are able to directly incorporate dietary biochemical components, i.e., fatty acyl groups, into membrane PL. Besides, FA originated from food assimilation, fish are able to include in lipid molecules fatty acyl groups obtained due to biosynthesis or conversion from precursors. Freshwater fish are known to have capacity to synthesize LC-PUFA from the shorter chain precursors [35,36]. Indeed, some studied fish, e.g., charr, contained in TAG and PL certain amounts of C24 PUFA and 20:4n-3 that likely were intermediates of DHA and EPA synthesis.

### *4.3. Content of Essential LC-PUFA in Fish PL and TAG*

The studied seven salmonid species varied ~ 7-fold in total lipid and fatty acid contents per a mass unit of muscle tissues. Most of this variation was related with different TAG content in muscles (Table 5, Figure 4), whereas PL content evaluated as their FA sum varied slightly (Table 4). The observed variation in lipid class contents in the studied fish is in agreement with well-known notion that polar lipids comprise cellular membranes and, as a result, have a relatively constant content in muscle cells, in contrast to that of triacylglycerols [19,21,37]. For instance, an absence of relation between total lipid and phospholipid contents and a strong relation between total lipid and triacylglycerol contents expressed as percentages of muscle mass were previously shown for a number of marine myctophid species [38].

Fish polar lipids are commonly considered as a physiologically crucial lipid class that are rich in LC-PUFA, mostly in DHA and, to a lesser extent, in EPA [19,37,39]. Indeed, percentages of DHA and EPA in PL of various wild marine and freshwater species ranged as 11.5–55.7% and 2.6–14.6%, with average values of 31.4% and 7.4%, respectively [40–46]. The average levels of EPA and DHA in

the fish species from our study coincided with the above ranges, except the EPA value of whitefish, 16.6%, which was a bit higher than the known values.

Triacylglycerols are considered to be relatively poor in LC-PUFA and preferably accumulate monoenoic C16-22 FA [18,19,37]. According to the available data, levels of DHA and EPA in fish TAG varied in intervals of 2.3%-23.3% and 1.1%–14.1%, with average values of 8.8% and 5.3%, respectively [40–46]. The percentages of both EPA and DHA of TAG in the fish species from our study well coincided with the above ranges (Table 5).

Triacylglycerols commonly comprise a large part of total lipids in muscles of medium-fat and fatty fish species. For instance, TAG achieved 80%, 90% and 51.5% in marine species arrow-tooth flounder (*Atheresthes stomias)* and golden pompano (*Trachinotus blochii*) and freshwater whitefish (*Coregonus lavaretus*), respectively [43,47,48]. In the studied freshwater salmonids, TAG percentages varied from 43.4% to 89.7% of the sum of two acyl-containing lipid classes (Figure 4A). Such high TAG levels may be explained by adaptation of the fish species to low-temperature conditions in the studied arctic lakes [49,50].

Regarding the relatively high contents of TAG per mass unit and percentages of EPA and DHA in TAG, we hypothesized that content of EPA and DHA in TAG would appreciably contribute to total content of EPA and DHA and would increase along with lipid content in muscles of the studied fish. Hence, we compared content of EPA and DHA esterified as TAG versus that esterified as PL. Content of EPA+DHA in PL per mass unit of muscle tissues were similar among the studied salmonids, moderately varying in the interval of 1.9–3.5 mg g−<sup>1</sup> (Figure 4B). In contrast, values of EPA+DHA in TAG of the fish species greatly varied, ~10-fold. Lean fish, i.e., whitefish of the non-identified form, round whitefish and broad whitefish, contained only 25%–47% of EPA + DHA of total content of these PUFA in the muscles esterified as TAG molecules. In contrast, the medium-fat and fatty fish, inconnu, muksun, whitefish and charr, had more than half of their muscle EPA and DHA content as TAG molecules, up to 72%. Thus, the wild salmonids that had relatively high content of n-3 LC-PUFA in muscles (~ > 5 mg g−1) contained the major portion of these nutritionally valuable compounds in the storage lipids. Our finding evidently contradicts a common notion that lean and medium-fat fish that have PL as a main lipid class in the muscles are the best dietary sources of n-3 LC-PUFA for humans [19,51]. Wild fatty fish which are able to deposit large amounts of storage lipids in their muscles appear to be the most valuable sources of n-3 LC-PUFA in human diet.

Further, our results are in a good accordance with many studies showed strong relationship between total lipid and EPA, DHA or their sum contents in fish muscles. Such relation was found for marine species, sprat *Sprattus sprattus* and herring *Clupea harengus* from Baltic Sea [52], for five marine species from the northeast Pacific [38] and for several freshwater species from a subalpine lake [53]. The significant relation was also found across farmed families of Atlantic salmon *Salmo salar* [54].

It is interesting to note that percentages of the n-3 LC-PUFA and lipid (or total FA as its proxy) content were negatively correlated in aforementioned and other studies [10]. The reported negative correlation was explained by the fact that total lipids increase preferably at the expense of TAG, whereas content of the membrane phospholipids, which are rich in n-3 PUFA remains fairly constant [21,37]. As a result, the proportions of EPA and DHA in muscle total lipids become diluted due to the accumulation of neutral lipids, which have high levels of monounsaturated FA. Although increase of total lipid content at the expense of TAG in fish muscles, as a rule, leads to decrease of n-3 LC-PUFA percentage, this does not mean that a concomitant decrease of nutritional quality of a fish occurs. Nutritional quality of fish products must be estimated on quantitative base expressed as mg FA per gram of tissue rather than percentage base [10].

Quantitative (mg per gram of tissue) measurements of TAG versus PL contribution in n-3 LC-PUFA of fish muscles are very scarce. Some studies gave indirect evidence of significant TAG contribution. For instance, among four fish species commercially harvested in Alaskan waters, arrow-tooth flounder *A. stomias* had maximum contents of EPA and DHA, 7.0 mg g<sup>−</sup>1, as well as maximum levels of TAG,

80% of total lipids in edible muscles [47]. Myctophid fish species with higher total lipid content (proxy for TAG content) also had higher contents of EPA and DHA esterified as TAG [38].

Some direct measurements showed that lean fish contained less than half of EPA and DHA in their muscles esterified as TAG, e.g., wild white seabream *Diplodus sargus* [40], whitefish *Coregonus lavaretus* [43], six commercial Chilean marine species [46]. In contrast, the only studied medium-fat fish (2–4% lipid content of wet mass), Pacific sandperch *Prolatilus jugularis*, had approximately 60% of EPA+DHA esterified as TAG [46]. Similar to latter finding, farmed *C. lavaretus* which had one of the highest known values of EPA and DHA in muscles, 18.6 mg g−<sup>1</sup> wet weight, had 61% of that value in TAG [43]. In our study, the fish species were strongly variable in lipid and total FA content and, as a result, in n-3 LC-PUFA content esterified as TAG. Like in the abovementioned studies, the fatty fish, muksun, whitefish and charr, had relatively higher content of EPA + DHA per mass unit that were mostly esterified as TAG (Figure 4B).

If we take the threshold of the recommended personal daily dose of EPA + DHA as 1 g and the average per serve portion of fish as 200 g [55,56], a fish of proper nutritional value should contain EPA + DHA nearly or more than 5 mg g−<sup>1</sup> of filet [57]. The obtained data on lipid class composition and content mean that when such fish is consumed, nearly or more than half of the essential n-3 LC-PUFA comes as TAG form. Recent studies showed that bioavailability of FA, including LC-PUFA esterified as TAG may be lower than that esterified as PL [58,59], but see [60]. Thus, distribution of LC-PUFA in major lipid classes should be further addressed in studies of nutritional quality of various fish products.

### **5. Conclusions**

The studied fish with similar feeding spectra were identified similarly by a multivariate analysis of FA profiles of total lipids, TAG and PL. Marker FA characteristic of diverse food sources (benthic, terrestrial, etc.), accumulated in nearly similar proportions within TAG and PL, and thereby allow to use both these fractions, as well as total lipids, for identification of food sources of wild fish. The found incorporation of the fatty acid trophic markers in structural polar lipids similarly to that in reserve TAG deserves further studies. Regarding contribution of TAG and PL into content of essential LC-PUFA of the taxonomically closely related fish species of order Salmoniformes, we found that content of EPA+DHA esterified as PL was nearly invariable, presenting presumably a species/taxon-specific optimal level. In contrast, content of EPA+DHA esterified as TAG greatly varied among the studied fish and provided most contribution to total EPA+DHA content in the fatty fish species, charr, whitefish and muksun. We can conclude that EPA+DHA-rich fish species likely accumulate these nutritionally valuable compounds predominately in the TAG form.

**Author Contributions:** N.N.S.: conceptualization, writing—original draft; O.N.M.: conceptualization, investigation, methodology; A.E.R.: investigation, methodology; L.A.G.: investigation, methodology; S.P.S.: investigation; A.A.K.: investigation; M.I.G.: formal analysis, writing—review & editing. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Russian Science Foundation, grant number 16-14-10001.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Tiny but Fatty: Lipids and Fatty Acids in the Daubed Shanny (***Leptoclinus maculatus***), a Small Fish in Svalbard Waters**

**Svetlana A. Murzina 1,\*, Svetlana N. Pekkoeva 1, Ekaterina A. Kondakova 2,3, Zinaida A. Nefedova 1, Kseniia A. Filippova 1, Nina N. Nemova 1, Alexei M. Orlov 4,5,6, Jorgen Berge <sup>7</sup> and Stig Falk-Petersen <sup>8</sup>**


Received: 20 January 2020; Accepted: 17 February 2020; Published: 28 February 2020

**Abstract:** The seasonal dynamic of lipids and their fatty acid constituents in the lipid sac and muscles of pelagic postlarval *Leptoclinus maculatus*, an ecologically important fish species in the Arctic food nets, in Kongsfjord, Svalbard waters was studied. The determination of the qualitative and quantitative content of the total lipids (TLs), total phospholipids (PLs), triacylglycerols (TAGs), cholesterol (Chol), cholesterol esters (Chol esters) and wax esters was analyzed by TLC, the phosphatidylserine (PS), phosphatidylethanolamine (PE), phosphatidylinositol (PI), phosphatidylcholine (PC), lysophosphatidylcholine (LPC) and sphingomyelin (SM) were determined by HPLC, and fatty acids of total lipids using GC. The lipid sac is a system of cavities filled with lipids, and it is not directly connected to organs of the digestive system. The wall's inner layer is a multinuclear symplast that has a trophic function. The results provide additional knowledge on the role of lipids in the biochemical and physiological adaptation of fish to specific environments and clarify the relationship between fatty acids and the food specialization of postlarvae. Analysis of the fatty acid (FA) profile of TLs in the muscles and lipid sac of daubed shanny pelagic postlarvae showed it to be tissue- and organ-specific, and tightly associated with seasonal variations of environmental factors (temperature conditions and trophic resources).

**Keywords:** lipids; fatty acids; *Leptoclinus maculatus*; ontogenesis; adaptations; trophic nets; Arctic

### **1. Introduction**

Lipids are one of the diverse, multifunctional, and basic groups of biochemical molecules. Among the numerous functions of lipids in the body, we can distinguish the following primary functions: in any cell, lipids are the basis of biological membranes, where proteins also act (lipids as structural molecules); lipids provide energy for metabolic reactions and processes (acting as energy producers or providing a storage function); and lipids provide a regulatory function that is carried out by biologically active molecules (eicosanoids). The diverse functions of lipids are carried out both intraand intercellularly [1–6]. Individual lipid classes in the body, particularly those of fish, perform several functions, each of which is crucial in specific ecological and physiological conditions [7]. An organism's stability and sensitivity to various external influences or fluctuations of environmental factors is largely determined by peculiarities of lipid metabolism.

Fatty acids (FAs), which are mainly in a bound state in the body, are the most mobile components of lipid molecules and are characterized by multifunctionality [8]. Fatty acid constituents of lipids are incorporated into adaptive reactions relatively quickly, and under both normal and stress-induced conditions, various FAs provide the body with the choice of alternative ways (mechanisms) of response: regulation of the physical–chemical state of biomembranes, changes in enzyme activity without changing protein concentration, and the synthesis of biologically active mediators [9]. A major proportion of the fish lipid FA complexes are long-chain and highly unsaturated FAs, which emphasizes their important functional role in the body [8]. Moreover, Arctic marine organisms, including fish, accumulate high amounts of monounsaturated long-chain FAs (such as C20 and C22) that, together with polyunsaturated FAs (PUFAs), have significant biological effects on organisms, including humans.

Research on the composition and role of lipids and their FA constituents in fish inhabiting northern seas, as well as the processes of lipid transformation and transfer along the food chains of aquatic ecosystems, is essential for understanding the optimal functioning of all metabolic systems of the body in a changing environment (for example, daily, seasonal, and interannual). With the accumulation of lipids, and their dynamics and expenditure peculiarities for northern latitudes, aquatic organisms ensure the maintenance of vital activity and determine an individual survival rate under changing environmental factors and their combination, taking into account the life cycle of an organism.

Despite the severe environment in the Arctic, ichthyofauna is characterized by high plasticity and variability in life strategies and, in general, Arctic boreal marine ecosystems are very productive. The Arctic ichthyofauna is poorly studied in comparison to that of the Antarctic. There is a lack of up-to-date data on the study of lipids in ecological and biochemical adaptations of marine organisms, especially fish in the Arctic, taking into account that Arctic marine ecosystems are lipid-dependent, and lipids are considered vital molecules in northern latitudes.

The Stichaeidae family is an evolutionarily relatively young, extensive, and extremely diverse taxon [10–12] characterized by high adaptive capacities as well as plasticity to environmental factors and their fluctuations. The main directions of the research on representatives of this family around the world are aimed at obtaining new data in the fields of taxonomy, general biology, physiology, and biochemistry, studying the processes by which high-latitude aquatic organisms adapt to their environment [13,14]. Fish of the Stichaeidae family dwell in the bottom of the continental shelf in the coastal waters of marine ecosystems. The depth of their habitat varies greatly, and some species live in tidal currents. Thus, fish belong to the eurigaline group. Specific habitat conditions and a wide range of variations in environmental factors for fish from this family determine adaptability and plasticity.

Prolonged spatial continental isolation has led to the formation of suborder endemic species, including the *Lumpenus* and *Leptoclinus* genera that, among other things, define the northwestern Atlantic region of this suborder endemism [12,15,16]. The daubed shanny *Leptoclinus maculatus* (Fries, 1838) is a marine demersal species belonging to the Arctic boreal [17–19] zoogeographical group. Information about the biology of these species from the Stichaeidae family is not abundant [20–26]. Earlier [27–29], it was found that pelagic daubed shanny juveniles have a unique formation, namely, a lipid sac in the ventral part of their body. The lipid sac is a depot of lipids that ensures buoyancy and,

to some extent, phenotypic adaptation to pelagic conditions. It consists of large, densely packed lipid drops [27,28,30]. Later [28,29], it was found that postlarvae inhabit the pelagic zone until the age of 3–5 years, after which juveniles switch to a demersal mode of life. Several postlarval stages (L1–L5) have been distinguished for *L. maculatus* on the basis of age, size and weight, body color, and pigmentation, as well as the presence and parameters of the so-called lipid sac [28,29]. The lipid sac is resorbed after the fish become demersal, whereas in the pelagic young, it is distinct and formed by accumulating lipids from a diet of zooplankton (*Calanus*), which is their principal source of food [27,29,31–33].

The ecological role of *L. maculatus* becomes clearer upon a detailed examination of the trophic structure of Arctic and boreal waters, in which these species are among the main food sources for marine fish, mammals, and birds [34–36]. Studied fish species occupy a double niche in Arctic trophic chains, acting as both predator and prey [37]. The only detailed study of the contribution of the Stichaeidae family, including *L. maculatus* and *Lumpenus fabricii*, to the nutrition of northern birds [38] are results of the study of behavioral reactions of *Cepphus columba*, or pigeon guillemot, when feeding chicks in the Alaskan region, and includes an analysis of the composition and selection of objects as food. Despite the fact that some birds specialize in catching lipid-rich fish (for example, the northern Pacific sand lance *Ammodytes hexapterus* and Pacific herring *Clupea pallasii*), most prefer to produce food for chicks with fish species that are available and constant in the coastal zone, despite their lower calorie content, such as adults of Stichaeidae family.

The aim of the present study was to study the seasonal dynamics of lipids and fatty acids in the lipid sac and muscles of pelagic postlarval *L. maculatus* in Kongsfjord, Svalbard waters. The results can clarify the role of lipids in the adaptation of young fish to specific environments, and the relationship between fatty acids and postlarval food specialization.

### **2. Material and Methods**

### *2.1. Sample Collection and Description*

*Leptoclinus maculatus* pelagic postlarvae were collected in summer, winter, and spring in Kongsfjord (79◦01', 11◦21'), Svalbard, using pelagic and bottom trawls on board the R/V *Helmer Hanssen* (UiT, Norway). The hydrological characteristics of the sampling sites are presented in Table 1.


**Table 1.** Hydrological characteristics of the sampling sites of *L. maculatus* postlarvae in Kongsfjord.

Daubed shanny juveniles are characterized by long-term development in the pelagic zone and relatively recently established multistage; stages are indicated by a Latin letter and number in the proposed system, and range from L1 to L5 [28,39] (Figure 1). Stages L1–L3 were mainly collected by pelagic trawl, and L4–L4 \* by bottom trawl.

Postlarval stages of the daubed shanny differ in their morphological and physiological characteristics (length, weight, body color, pigmentation, and the presence and state of the lipid sac), and are also divided according to belonging to ecological groups (pelagic, transitional, demersal) [29,32,39] (Figure 1). The data on postembryonic growth and the developmental features of the juvenile *L. maculatus* and, in particular, changes in the length and state of the lipid sac during larval development, were presented for the first time in the work of S. N. Pekkoeva [29]. These data are unique for this species as they largely clarify some features of its ecology.

**Figure 1.** Stages of development of *L. maculatus* by Pekkoeva S.N., 2018 [29].

In the present research, lipid profiling was performed on postlarvae in the L3 developmental stage (Figure 2).

**Figure 2.** Pelagic postlarvae of *L. maculatus* (L3 stage of development) in summer. (**A**) A side view, (**B**) A view down the lipid sac.

Histomorphological studies of the lipid sac were presented on postlarvae in L2–L5 developmental stages.

All work was performed according to and within the regulations enforced by the Norwegian Animal welfare authorities and no specific permissions were required. The R/V Helmer Hanssen is owned by the University of Tromsø and has all necessary authorization from the Norwegian Fisheries Directorate to use a bottom trawl to collect fish for scientific purposes. Furthermore, the organisms are neither protected nor endangered in the coastal waters of the Svalbard Archipelago. Upon trawling, the fish were sacrificed by a sharp blow to the head and immediately dissected as specified below.

### *2.2. Lipid Extraction*

The lipid sac and muscle of postlarval *L. maculatus* (L3 developmental stage) were immediately dissected from fresh fish aboard the vessel. Samples were homogenized in glass vials in a chloroform/methanol (2:1, v/v) solution (10 mL per 1–3 g wet weight). Total lipids (TLs) were extracted using the Folch method [40]. The homogenate was filtered, and the residue, retained on the paper filter, was rinsed with 30 mL of extractive mixture. The extract was then mixed after the addition chloroform and deionized water, and left to settle in the separatory glass funnel (Schott

Duran, Hamburg, Germany) until complete separation of organic phases. Lipids remained in the lower chloroform layer, whereas non-lipid substances moved to the upper aqueous methanol phase. The chloroform layer was then withdrawn to be evaporated under vacuum on rotary evaporator Hei-VAP Advantage HL/G3 (Heidolph, Schwabach, Germany), and dried in a vacuum exicator over phosphoric anhydride to constant weight. Total lipids were dissolved in chloroform/methanol and stored at −80 ◦C.

### *2.3. Lipid Class Analysis*

Thin-layer chromatography (TLC) was used to identify the lipid classes as total phospholipids (PLs), triacylglycerols (TAGs), cholesterol (Chol), cholesterol esters (Chol esters), and wax esters. Fractionation of total lipids was performed on ultrapure glass HPTLC Silica gel 60 F254 Premium Purity plates (Merck, Darmstadt, Germany). Certain lipids, namely, PLs, TAGs, Chol esters, and wax esters, were quantified using the hydroxamate method that was modified by [41], which involves the formation of dark brown complexes of trivalent iron ions with hydroxamic acid through ester bonding between lipids and hydroxylamine [42]. Stain intensity was measured using a spectrophotometer (SF-2000, OKB "Spectr", St. Petersburg, Russia) at a wavelength of 540 nm. The quantitative determination of Chol was made on the basis of the method described in [43] by using trichloroacetic iron dissolved in perchloric acid. Stain intensity was measured using a spectrophotometer at a wavelength of 550 nm. Standard references and analytical standards for TLC (Sigma Aldrich, St. Louis, MO, USA; Avanti Polar Lipids Co., Alabaster, AL, USA) were used to distinguish the lipid classes.

The spectra of individual phospholipid fractions were determined by high-performance liquid chromatography (HPLC) using Aquilon Stayer HPLC (Aquilon LLC, Moscow, Russia) according to the method of Arduini [44], using a Nucleosil 100-7 C18 HPLC column with a acetonitrile/hexane/methanol/phosphorus acid (918:30:30:17.5, by volume) mobile phase, the rate of movement was 540 mkl/h, the volume of the injected sample was 5 μL, and the injector was Rheodyne 7725i. Detection was performed using a spectrophotometer (UV light, 206 nm), and the method was isocratic. Samples were manually injected using Rheodyne Valco Beckman and SSI Valves syringes (Hamilton, Reno, NV, USA). Phospholipid standards (Sigma Aldrich, St. Louis, MO, USA) were used for the identification and quantification of the phospholipid compounds in the sample. We identified six phospholipids: phosphatidylserine (PS), phosphatidylethanolamine (PE), phosphatidylinositol (PI), phosphatidylcholine (PC), lysophosphatidylcholine (LPC), and sphingomyelin (SM).

### *2.4. Fatty Acid Analysis*

The fatty acid profile of the total lipid extracts was analyzed by gas chromatography (GC). The methylation of fatty acids from the lipid extract was made in a glass retort in which 0.1 mL of a solution containing 20 mg/10 mL (behenic FA, C22:0) (Sigma Aldrich, St. Louis, MO, USA) in methanol was added as internal standard; then, we carried out transesterification in methanol (2 mL) containing chlorate acetyl (0.2 mL) at 70 ◦C for 90 min (using a Schott Duran glass serpentine condenser). After extraction, cooling with hexane was carried out in glass serpentine condensers that were rinsed with 5 mL of hexane for each sample. Then, 2 mL of deionized water was added to each glass retort for phase separation in the separatory glass funnels for 15 min. Fatty acid methyl esters (FAMEs) remained in the upper hexane layer, whereas other substances moved to the lower aqueous phase. The hexane layer was then withdrawn to be evaporated under vacuum on rotary evaporator Hei-VAP Advantage HL/G3 (Heidolph, Schwabach, Germany). Then, 0.9 mL of hexane for GC (Sigma Aldrich, St. Louis, MO, USA) was added to each glass retort, and the content was moved to the glass GC vials for the following GC analysis.

FAMEs were identified using a Chromatek-Crystall-5000.2 (Chromatek, Yoshkar-Ola, Russia) gas chromatograph with a flame-ionization detector (FID) and a Zebron ZB-FFAP capillary gas chromatographic column (Phenomenex, Torrance, CA, USA). An isothermal column configuration was used (200 ◦C); the temperatures of the detector and evaporator were 250 and 240 ◦C, respectively. Chromatek-Analytik-5000.2 software (Chromatek, Yoshkar-Ola, Russia) was used for recording and integrating the data. FAMEs were identified with standards of Supelco 37 Component FAME mix, bacterial acid methyl ester (BAME), and PUFA No. 1 (Sigma Aldrich, St. Louis, MO, USA), and by comparing the equivalent lengths of carbon chains and table constants according to Jamieson [45].

### *2.5. Statistical Analysis (Lipid Study)*

Data were analyzed to determine whether they exhibited normal distribution. Significant differences (*p* ≤ 0.05) in the mean of the studied lipids and fatty acids between fish collected in different seasons were tested by one-way ANOVA. To perform statistical analysis, StatGraphics (Statpoints technologies, Inc., The Plains, VA, USA) and Microsoft Excel 10 (Windows 7, USA) were used.

### *2.6. Histological Analysis*

The lipid sac of postlarval *L. maculatus* (L2–L5 stages) was dissected from fresh material and fixed in 4% formaldehyde (buffered) onboard the vessel. The tails were cut off from postlarval *L. maculatus* (L2 and L3 stages). In the postlarvae at the L4, L4 \*, and L5 stages, the heads, pectoral fins, the dorsal part of the body, and tails were removed. In one case, the lipid sac was dissected from the body cavity, but it was found to be dissatisfactory. The material was washed in PBS, dehydrated, and embedded in paraplast (Leica, Wetzlar, Germany) according to the standard procedure. Serial parasagittal sections of 6–7 μm were cut using the sleigh microtome Leica SM 2010R (Leica Microsystems, Wetzlar, Germany). The sections were stained with Carazzi's hematoxylin and eosin (Biovitrum, St. Petersburg, Russia). Histological sections were studied using the light microscope Leica DMI6000 (Leica, Wetzlar, Germany). Measurements were made using Fiji software [46]. To minimize the risk of remeasuring the same nucleus, at least four sections were left between the measured ones.

### *2.7. Statistical Analysis (Histological Analysis)*

Differences in the LSN lengths between L2 and L5 were evaluated using the Mann–Whitney U-test, with *p* < 0.05 indicating significant difference. All data are presented as mean ± SE.

### **3. Results**

### *3.1. Histomorphology of the Lipid Sac of Postlarval L. Maculatus (L2–L5 Developmental Stages)*

The histomorphological study of the lipid sac (LS) of postlarval *L. maculatus* confirmed that the LS is a complex of lipid-filled compartments that is not directly connected to the digestive system (Figures 3–5). The inner layer of these compartment walls is a multinucleate symplast, and the outer layer is a connective tissue [27]. Several layers of connective tissue and blood vessels are found in the walls. We named the LS symplast the lipid syncytial layer (LSL) by analogy with the yolk syncytial layer (YSL). Unstained lipid inclusions were seen in the LSL cytoplasm. The LSL nuclei (LSN) were generally round, elliptical, or elongated in sections. The LSN could be stained weakly or intensively with hematoxylin as a likely indication of differing eu- and heterochromatin content. The linear sizes of the LSN increase significantly by stage L5 (average length of LSN rose from 8.12 ± 0.17 μm at L2 (*n* = 185 where n is the number of nuclei) to 15.94 ± 0.3 at L5 (*n* = 250)).

**Figure 3.** The general view of the lipid sac of postlarval daubed shanny. Parasagittal sections. (**A**)—L2 stage, (**B**)—L4 \* stage. Figure legend: g—gut, ls—lipid sac, m—myomere, k—kidney. The walls of the lipid sac cavity are indicated with arrowheads. Staining: Carazzi's hematoxylin and eosin. Bar: 500 μm.

**Figure 4.** The walls of the lipid sac compartments. Parasagittal sections. (**A**–**C**)—L2 stage, (**D**)—L4 \* stage. Figure legend: bv—blood vessel, lsl—lipid syncytial layer, ct—connective tissue, n—nucleus of LSL. The homogenous matrix between the compartments of LS. Staining: Carazzi's hematoxylin and eosin. Bar: 100 μm.

**Figure 5.** Nuclei of the LSL of LS at the L2 stage (**A**,**B**) and L5 stage (**C**). Nuclei generally have regular shapes, their linear sizes increase by the stage L5. Figure legend: bv—blood vessel, lsl– lipid syncytial layer, ct—connective tissue, n—nucleus of LSL. The homogenous matrix between the compartments of LS. Staining: Carazzi's hematoxylin and eosin. Bar: 100 μm.

### *3.2. Seasonal Dynamics of Total Lipids in the Lipid Sac and Muscles of Pelagic Postlarval (L3 Developmental Stage) L. Maculatus*

The seasonal variation of total lipid (TL) content in the muscles and lipid sac of pelagic postlarval daubed shanny from Kongsfjord was determined in spring, summer and winter.

TL level was the lowest in muscles in the winter and spring (13.9% and 13.4% dry weight, respectively; no significant differences), whereas the content in muscles in the summer was significantly higher—27.9% dry weight. A different pattern of dynamic in TLs was observed in the lipid sac: the content increased significantly from summer to winter—64.4% and 75.9%, and decreased to 46.96% dry weight by spring (Table 2).


**Table 2.** Seasonal dynamics of total lipids and their individual classes (% of dry weight; % of total phospholipids for individual phospholipid classes) in the muscle and lipid sac of postlarval daubed shanny from Kongsfjord.

TL—total lipid, TAG—triacylglycerol, Chol esters—cholesterol esters, PL—total phospholipids, PI—phosphatidylinositol, PS—phosphatidylserine, PE—phosphatidylethanolamine, PC—phosphatidylcholine, LPC—lysophosphatidylcholine, SM—sphingomyeline, Chol—cholesterol. Different uppercase letters indicate significant (*p* ≤ 0.05) differences when comparing the values in the same type of tissue; n—number of tissues; \*—values are given according to [29].

### *3.3. Seasonal Dynamics of Certain Lipid Classes in Lipid Sac and Muscles of Pelagic Postlarval L. Maculatus*

The content of storage lipids in the form of triacylglycerols (TAGs), cholesterol esters (Chol esters) and wax esters, and membrane lipids in the form of phospholipids (PLs) in the muscles of the daubed shanny decreased significantly by winter compared to its levels in summer (2.4%, 0.9%, and 7.2% dry weight versus 12.7%, 1.7%, and 9.6%, respectively) (Table 2). The level of TAGs in the fish increased significantly in spring compared to winter (6.1% vs. 2.4% dry weight) but remained lower than in the summer (12.7% dry weight). The level of Chol esters and wax esters in the fish in spring rose to levels observed in the muscles in the summer (1.7% and 1.7% dry weight). The PL content in postlarval muscles was the lowest in spring (2.5% dry weight).

The level of total PLs in *L. maculatus* muscles was associated with variations in both the dominant phosphatidylcholine (PC) and phosphatidylethanolamine (PE) (the former tended to decline while the latter increased from summer towards winter), as well as minor phosphatidylinositol (PI), phosphatidylserine (PS), and sphingomyelin (SM), which arose in the fish from winter towards spring (Table 2). In the muscles of the postlarval daubed shanny, the seasonal variation of total PLs consisted of slight but statistically significant variations of PC and PE, as well as minor PI, PS, and SM, especially in the winter season (Table 2).

The lipid sac of daubed shanny postlarvae had the following changes from summer towards winter: storage TAGs increased from 58.3% to 68.8% dry weight, and trace amounts of Chol esters and wax esters increased from 0.1% to 3.0%, remaining at this level towards spring (2.6% dry weight). Total PL levels declined towards winter (from 6.1% in summer to 0.8% in winter), and by spring, PL content reached a level observed in the young in summer (6.3% vs. 6.1% dry weight). The Chol content in the lipid sac of postlarvae tended to rise from summer towards spring from trace amounts (0.1%) in summer to 3.4% in winter and 10.0% dry weight in spring. The Chol/PL ratio was the highest in the lipid sac of postlarvae in the winter season (4.1), the lowest in summer (0.01), and intermediate in spring (1.6) (Table 2). In spring, when the trend was a general decrease in TLs and TAGs in the lipid sac, the Chol/PL ratio (1.6) could be considered as an indicator of lipid components being utilized for maintaining the viability of postlarval fish (and, hence, the extraction of lipids from the sac).

### *3.4. Seasonal Dynamics of Certain Fatty Acids in the Lipid Sac and Muscles of Pelagic Postlarval L. Maculatus*

Among FAs in the muscles of the studied fish in summer, monounsaturated FAs (MUFAs) were dominant (51.3% of the FA sum) due to 20:1(n-9) and 22:1(n-11) FAs (18.3% and 13.9% of the FA sum, respectively), which pointed to the prevalence of *Calanus* copepods in the diet (Figures 6 and 7). The muscles were also supplied with 16:1(n-7) and 18:1(n-9) FAs (6.2% and 6.2% of the FA sum, respectively), which are of phytoplanktonic origin (Figure 7). The second position in the FA profile in the muscles belonged to PUFAs (27.1% of the sum of FA), due to the (n-3)PUFA (23.3% of the sum of FAs) (Figure 6), including the essential 20:5 (n-3) and 22:6 (n-3), as well as another FA 18:4(n-3) (6.6%, 8.3%, and 4.6% of the FA sum, respectively) (Figure 8). The contribution of the (n-6) PUFA was not high, with 3.3% of the sum of FAs (Figure 6), where 18:2(n-6) and 20:4(n-6) accounted for 1.5% and 0.4% of the sum of the FAs, respectively (Figure 9). The saturated fatty acid (SFA) level was 21.6% of the sum of FAs due to 16:0, 14:0, and 18:0 FAs (12.5%, 6.5%, and 2.1% of the sum of FAs, respectively) (Figures 6 and 10).

**Figure 6.** Seasonal dynamic of the saturated FA (SFA), monounsaturated FA (MUFA), polyunsaturated FA (PUFA) levels, including (n-3)PUFA and (n-6)PUFA (% of the sum of FAs) in muscle and the lipid sac in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

**Figure 7.** Seasonal dynamic of certain MUFA and PUFA levels (% of the sum of FAs) in muscle in postlarval (L3 developmental stage) daubed shanny from Kongsfjord.

**Figure 8.** Seasonal dynamic of certain (n-3)PUFA levels (% of the sum of FAs) in muscle in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

**Figure 9.** Seasonal dynamic of certain (n-6)PUFA levels (% of the sum of FAs) in muscle in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

**Figure 10.** Seasonal dynamic of certain SFA levels (% of the sum of FAs) in muscle in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

The lipid profile of muscles in winter had a different quantitative FA composition with the dominance of PUFAs, in which the level increased, compared to summer, to 45% of the sum of FAs due to (n-3)PUFA (37.9% of the sum of FAs), among which the essential 20:5(n-3), 22:6(n-3), and 22:5(n-3) FAs increased in proportion (11.5%, 16.2%, and 5.6% of the sum of FAs, respectively) (Figure 8). The level of (n-6)PUFA had also increased since summer, to 4.2% of the sum of FAs due to increased contents of 18:2(n-6) and 20:4(n-6) FAs (1.8% and 0.6% of the sum of FAs, respectively). MUFA amounts in the muscles of *L. maculatus* in winter decreased to 35.4% of the sum of FAs, mainly at the expense of 20:1(n-9), 22:1(n-11) FAs (12.2% and 8.1% of the sum of FAs, respectively). The decline in 16:1(n-7) and 18:1(n-9) FAs in fish muscles in winter was the least pronounced, but reliable compared to summer levels (3.8% and 7.5% vs. 6.2% and 2.1% of the sum of FAs, respectively).

MUFAs dominated in muscles in daubed shanny in spring, but their amounts were reliably lower than in summer (31.6% vs. 51.3% of the sum of FAs). The springtime FA profile of muscles featured a PUFA level that was significantly lower than that in summer or winter at the expense of the essential 20:5(n-3) and 22:6(n-3) FAs, with 11.2% and 19.3% of the sum of FAs, respectively. Curiously, it was in spring that the quantities of another essential FA of the (n-6)PUFA (arachidonic FA) reached a maximum in *L. maculatus*, at 2.5% of the sum of FAs. Variations in PUFAs and their individual FAs were correlated with changes in the content of some minor PL classes, namely, PI and PS.

The dominant group in the lipid sac (all seasons) was MUFAs (Figure 11), mainly due to 20:1(n-9) and 22:1(n-11) FAs (respectively 63.9%, 23.7%, and 19.8% of the sum of FAs (Figure 8)), derived from a copepod diet of *Calanus* spp. *Calanus* copepods synthesize these FAs de novo and are the main food for postlarvae at the L3 stage. Another MUFA present was 16:1(n-7), amounting to 8.4% of the sum of FAs. It is produced by diatoms and consumed by postlarvae through food. The second largest FA class in the lipid sac of the daubed shanny was SFAs, due to 16:0, 14:0, and 20:0 (19.2%, 7.9%, 5.5%, and 4.1% of the sum of FAs, respectively), which are the main constituents of the biomembranes of the lipid sac (Figure 12). Another component of membrane structures, alongside SFAs, are PUFAs (16.9%), due to (n-3)PUFA 20:5(n-3) and 22:6(n-3) FAs, which contributed 13.8%, 4.6%, and 3.9% of the sum of FAs, respectively (Figure 13). The FA profile of the lipid sac of postlarvae in winter was noted for its elevated MUFA content (68.6% of the sum of FAs), due to the dietary 20:1(n-9) and 22:1(n-11) FAs.

**Figure 11.** Seasonal dynamic of certain MUFA levels (% of the sum of FAs) in the lipid sac in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

**Figure 12.** Seasonal dynamic of certain SFA levels (% of the sum of FAs) in the lipid sac in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

**Figure 13.** Seasonal dynamic of certain (n-3)PUFA levels (% of the sum of FAs) in the lipid sac in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M ± SE.

According to the 18:1(n-9)/18:1(n-7) ratio, juveniles belong to the group of zooplanktonophages. This is also indicated by the dominance of 20:1(n-9) and 22:1(n-11) MUFAs, which are copepod biomarkers. Seasonal variations in the ratio of these FAs allow for tracking of seasonal variations in the species composition of copepods, the main food objects of juveniles, and their availability in different seasons of the year (Figure 14).

**Figure 14.** Seasonal dynamic of 22:1/20:1 and 18:1(n-9)/18:1(n-7) ratios in the lipid sac in postlarval (L3 developmental stage) daubed shanny from Kongsfjord. The data presented as M±SE.

### **4. Discussion**

### *4.1. Histomorphology of the Lipid Sac of Postlarval L. maculatus*

The lipid sac (LS) is found only in the circumpolar fish *L. maculatus* of the Stichaeidae family. The LS is a unique temporary structure and acts as one of the mechanisms for physiological and biochemical adaptation, helping young fish to survive and develop in northern latitudes [27–29,31–33].

Interestingly, the LSL and the YSL of bony fish have several important similarities, although these structures are not analogous or homologous: during development, their nuclei grow and their shape becomes more complex with different heterochromatin content throughout, and the functioning of symplasts intensifies during transitory developmental periods. A symplast with polyploid nuclei is one of the most widespread organizational variants of the extraembryonic system.

### *4.2. Dynamics of Total Lipid and Certain Lipid Classes in the Lipid Sac and Muscles of Postlarval L. maculatus*

The lipid sac serves as a lipid depot, securing the viability and resilience of pelagic postlarvae in the severe environment of Svalbard waters, including prolonged winter and conditions of polar night. In addition, we found a significant decrease in TL content in the lipid sac by spring.

Pelagic postlarvae accumulate lipids during summer, which is the most productive but the shortest period in the Arctic. Typically, pelagic juveniles of *L. maculatus* at the L3 developmental stage are zooplanktivourous, fattening in the summer by consuming high energy food such as copepods, most abundantly represented by *Calanus finmarchicus* and *Canalus glacialis*. Copepods are the principal food components for the daubed shanny at this stage of development [29,32].

It has been presented [47,48] that copepods account for over 60% of the total number of species in the zooplankton community in Svalbard waters, the dominant ones being boreal oceanic *C. finmarchicus*; Arctic neritic *C. glacialis*; arctoboreal *Pseudocalanus* sp.; cosmopolitan *Oithona similis*; Arctic bathypelagic *Metridia longa*; oceanic cold-water *Microcalanus* sp.; and arctoboreal neritic *Acartia longiremis*. *C. finmarchicus* was shown to be abundantly present in Kongsfjord [47] due to the specific hydrobiological characteristics of this fjord. It appears that, like some other Arctic aquatic

organisms (e.g., zooplankton) that accumulate lipids in the body or in its special formations, deposition of a certain amount of lipids is a prerequisite for overwinter survival, as well as for providing energy for the growth and development of larvae and postlarvae of *L. maculatus*, as well as processes of metamorphoses. Furthermore, lipid accumulation in the lipid sac improves the buoyancy of postlarvae, and enables them to stay in the high-productive pelagic zone.

TLs in the muscles of young *L. maculatus* are accumulated and utilized exclusively for their direct function. The highest range of storage-lipid variation in the muscles of postlarval *L. maculatus* proves that these lipids are utilized to maintain energy-intensive metabolic processes that provide relevant postlarval locomotion. The decrease of storage lipids in TAGs that form from summer to winter points to their preferred utilization by the organism, as well as to a reduction of their amount due to limited nutrition in winter. The increase of TAGs and minor classes of storage lipids—represented by Chol esters and wax esters—in the muscles of the postlarval daubed shanny from winter towards spring can be regarded as an indicator of active feeding resumption, mainly on zooplankton. In muscles, the content of Chol, which is a major structural component of biomembranes, did not vary among seasons.

The resistance of aquatic organisms' biomembranes to various environmental factors largely depends on their lipid components. Changes in PL content and the ratios of their individual classes (mainly in muscles and the lipid sac) are key compensatory mechanisms in organisms that secure optimal performance of a large number of membrane-bound enzymes and their complexes [49] under a range of temperature conditions in the studied seasons. Furthermore, variations in PL quantities (mainly in PE, PS, and PI) and in FA components maintain and ensure required membrane fluidity, which is a property associated with adaptation to variable temperature conditions in the muscles of the daubed shanny [50–52]. Seasonal variations of water temperature in Kongsfjord (4.2 ◦C in summer, 1 ◦C in winter, and 0.5 ◦C in spring) activate the adaptive modifications of lipid components and influence the physical and chemical properties of cellular and subcellular membranes, as indicated by the PC to PE ratio in the muscles of postlarvae, as previously demonstrated for other aquatic organisms [53–55]. The mechanism of biomembrane adaptation to temperature reduction demonstrated for the muscles and gills of dark flounder *Pseudopleuronectes obscurus* [54] is based on oppositely directed variations of FAs within PC and PE: a rise in saturated FAs to polyunsaturated FA (SFA/PUFA) molecular form and a reduction in the ratio MUFA/PUFA, and PUFA/SFA in PC with the opposite trend observed for PE. Such changes in a certain set of the molecular forms of major membrane PLs are meant to maintain functional activity of the inner lipid monolayer and do not affect transformations in the outer monolayer of biomembranes, i.e., FA components of PLs are redistributed without a change in PL composition.

Functional biomembrane activity in the lipid sac, similar to the muscles of *L. maculatus* postlarvae exposed to variable temperature and foraging conditions in the studied seasons, is maintained both by changes in the quantities and ratios of individual PL classes, as well as through other biochemical mechanisms. One is a change in the Chol/PL ratio, which indicates variations in the microviscosity of biomembranes and their permeability mediated by key membrane lipids. A lower Chol/PL ratio in summer points to a higher permeability of biomembranes in the lipid sac as a result of utilization of lipid components for maintaining metabolic activity in this period. Thus, the lipid sac is a unique, metabolically active organ exemplifying successful physiological and biochemical adaptation of the young daubed shanny to living in high-latitude Arctic waters. Lipid sac formation, which begins with exogenous feeding in the larvae, is considered as an example of physiological and biochemical adaptation to fish habitats in the pelagic zone, contributing to their successful growth [29,56].

### *4.3. Seasonal Dynamic of Certain Fatty Acids of Total Lipids in Postlarval L. maculatus*

The qualitative and quantitative composition of FA components of TLs in the muscles and lipid sac of *L. maculatus* postlarvae inhabiting Kongsfjord in summer reflects the dietary regime and specialization in this period, as well as tissue- and organ-specificity, which is defined both by metabolic features as well as by physiological and biochemical functions performed by these organs under given environmental conditions. A manifestation of the latter is the selective integration of FAs derived from food into the lipid structure of muscles and the lipid sac, thus providing the fish with energy and structural components that the organism needs. This adaptive modification satisfies the organism's physiological demand for the accumulation of lipids with a certain structure in order to maintain a relevant metabolic level during long winters. Thus, the inclusion of 20:1(n-9) and 22:1(n-11) into the structure of energetic lipids (mainly TAGs) renders them metabolically active owing to the physical and chemical properties of these FAs, and helps satisfy the energy demand of daubed shanny postlarvae, even when water temperatures are low. Stage CV copepodites and adult copepods of the *Calanus* species have quite a high share of 20:1 and 22:1 FAs within the dominant energetic lipids class, namely, wax esters.

It was previously demonstrated that phytoplanktonic PUFAs can be partially transformed by zooplankton into SFAs and MUFAs, and can be deposited in their body as waxes, a long-term energy depot [57]. Thus, postlarval *L. maculatus* obtains essential phytoplankton-derived 20:5(n-3) and 22:6(n-3) FAs while feeding on zooplankton, which is a link for the transfer of essential FAs from primary producers to zooplankton feeders and higher-order consumers in high-latitude marine ecosystems, especially during the most productive summer period, when fish actively forage to accumulate lipid reserves.

Previously [29,58], it was shown that the dominance of young individuals (L2 and L3 developmental stages) in winter in Kongsfjord and older-age individuals (L5 stage) in the near-bottom layers of the water column corresponds with the division of the young daubed shanny into pelagic and demersal ecological groups. Similar changes in lifestyle and living conditions are also found in Antarctic fish [59,60]. In addition, studies [29,57] of the growth and early development patterns of the daubed shanny showed an increase in the length and weight of juveniles with age, increased growth at the age of 2+ (L2 stage), which is connected with the transition from phytoplankton feeding to high-energy zooplankton, while energy in the form of lipids is used on active growth. Thus, nutrition basis at the L1 development stage was phytoplankton-dominated by dinophyte algae, while a high level of biomarker 20:1(n-9) and 22:1(n-11) FAs from stage L2 reflects the nutrition of high-energy zooplankton of the *Calanus* genus, which is able to synthesize these FAs de novo [29,61]. Thus, the unique role of the postlarval daubed shanny in the transfer of matter and energy in the Arctic pelagic food chain with the participation of key chain links (phyto- and zooplankton), even in winter conditions, was also determined in this study. Interestingly, these results support the hypothesis about peculiarities of marine ecosystems functioning as a whole during polar nights in terms of the fact that the interaction of ecosystem components is determined by the season, e.g., despite the specific photoperiod, they actively interact with each other in winter, and zooplankton continue to make up a certain share in pelagic community structure, providing food to pelagic fish. The change of food objects and, accordingly, the type of food from phyto- to zooplankton in the juvenile daubed shanny, and then to carnivorous in adult individuals, was also shown for Antarctic silverfish *Pleurogramma antarcticum* [62], which is most numerous in East Antarctic waters [63].

By accumulating PUFAs in the TLs of its muscles, the postlarval daubed shanny ensures that membrane-bound enzymes and their complexes can function optimally when the functional load on the organ (e.g., swimming function and rheotaxis) or environmental conditions (e.g., temperature and salinity) changes. Polyunsaturated FAs largely define the inner structure of biological membranes and conditions for the activity of integral membrane proteins [64]. Some previously reported experiment data [64,65] suggested that if lipids with polyunsaturated chains are involved in the formation of the specific microenvironment of integral proteins then, by virtue of their physical and chemical properties, they can contribute to maintaining the relevant conformational mobility of these proteins and mitigate the negative effects of temperature changes on their activity, thus helping them to function normally. Furthermore, a sufficient supply of PUFAs during the development of *L. maculatus* postlarvae is necessary for their normal growth and development, since these compounds play important roles in regulating the activity of nerve cells in the formation of the vision system in fish, which is particularly crucial for the pelagic young. Young fish suffering a deficit of PUFAs may exhibit abnormal behavioral reactions.

The elevated PUFA level in the muscles of postlarval *L. maculatus* in winter is most probably due to their accumulation during the fattening season, which is a necessary precondition for the optimal functioning of integral membrane enzymes under low temperatures (the temperature in Kongsfjord in winter is 1 ◦C) and for these PUFAs to co-maintain their general membrane structure together with the basic MUFAs in the PL structure, 18:1(n-9) FAs, whose level remains constant in the winter season. With regard to temperature adaptations, one should always keep in mind the important role of biomembranes and their sensitivity to changes in ambient temperature. Many processes going on in cellular and subcellular membranes are essential for the organism, including the biosynthesis of the membrane itself. It is membranes that set the temperature limits within which all systems can function normally and an organism remains viable.

An elevated content of the essential 20:4(n-6) FA in the muscles of postlarvae may point to the ongoing synthesis of lipid signal mediators, which are synthesized in a majority of tissues and are involved in the regulation of a series of physiological processes: immune response, inflammatory reactions, and functioning of vision and the nervous system [66,67].

The FA profile of the lipid sac of the postlarvae in winter was noted for its elevated MUFA content (68.6% of the sum of FAs) due to dietary 20:1(n-9) and 22:1(n-11) FAs, corroborating its key role as an energy storage organ and demonstrating a successful accumulation of lipids, both in quality and quantity, during the fattening season. Studies showed [47] that the wintering stocks of *C. glacialis* are chiefly made up of stage IV–V copepodites, concentrated at 150 m depth, and for *C. finmarchicus,* this is copepodite stage V, living at 50–150 m depths. An 18:1(n-9)/18:1(n-7) ratio allowed us to categorize the daubed shanny as a carnivorous group of animals. Juveniles belong to the group of zooplanktonophages according to the 18:1(n-9)/18:1(n-7) ratio (above 1).

The lipid sac of the postlarvae contained lower levels of PUFAs in winter, with a simultaneous rise in Chol content compared to that in summer. This combination may point to an increase in biomembrane viscosity and a corresponding reduction in the activity of lipid transport from the lipid sac to preserve them for utilization later during the winter.

There probably exists some threshold metabolic level in this age group (L3 developmental stage) of *L. maculatus*, tightly linked to the stressfulness of environmental conditions in winter, and until this threshold is reached, lipids stored in the lipid sac are moderately utilized to support basic functions. This statement is confirmed by the reduced MUFA level (to 53% of the sum of FAs), including 20:1(n-9) and 22:1(n-11), in the lipid sac of postlarvae. However, the spring sampling of the young daubed shanny in Kongsfjord took place on 21 April, when phytoplankton was blooming (*Chl a* level was within 1.6–1.9 μg/L; 51% due to diatoms), and *Calanus* spp. copepods were present in the area [68]. Phytophagous copepods with pronounced ontogenetic (seasonal) migrations display a distinct correlation with the onset of the high-productivity period, when the vernal phytoplankton outburst takes place. At this point, the overwintered stage V–VI copepodites concentrate close to the surface for the start of the breeding season [47]. Meteorological conditions in this period of spring strongly fluctuated (wind speed up to 8.2 m/s and its directions), causing the *Chl a* level to fall below 0.5 μg/L and resulting in vertical water mixing and cooling, especially down to a depth of 60 m. The vertical redistribution of the phytoplankton biomass happening in the process caused transient breaks in its blooming, and a redistribution of phytoplankton species ratios and its low biomass were therefore observed [69,70]. Following these events, in the period from 28 April to 12 May, the conditions at our study site in the fjord were shown to have stabilized [68]. Thus, the lowest MUFA level in the lipid sac was detected in springtime, possibly due to their utilization within lipids to support the vitality of postlarval fish under the specific conditions of this period.

Observed quantitative MUFA and PUFA variations testified to their particular roles and evidence that the FA composition of lipids is specific to the studied organs, in accordance with the function they perform under given and variable environmental conditions. Certain FAs are retained in the structure of lipids to enhance their metabolic value, both to facilitate a general adaptation of postlarvae to overwintering, and to provide specific compensatory reactions adjusting an organism to seasonal environment.

MUFAs dominated in muscles in the studied fish in spring, but their amounts were reliably lower than those in summer (31.56% vs. 51.30% of the sum of FAs). The springtime FA profile of the muscles of the postlarvae featured a PUFA level that was reliably lower than that in summer or winter at the expense of the essential 20:5 (n-3) and 22:6 (n-3) FAs—11.15% and 19.34% of the sum of FAs, respectively. Curiously, the quantities of another essential FA of the n-6 family (arachidonic FA) reached a maximum in *L. maculatus* (2.45% of the sum of FAs) in spring. Variations in PUFAs and their individual FAs were correlated with changes in the content of some minor PL classes, namely, PI and PS. This resulted in an alteration in the activity of biomembrane enzymes due to modification of their lipid environment. Furthermore, in a stressful environment, FA components perform the function of lipid mediators, becoming integrated in the structure of their molecules. As mentioned above, the water temperature in Kongsfjord is 0.5 ◦C in spring. The observed lipid and FA profile of postlarval muscles in springtime is indicative of their role in the adaptation to the highly variable meteorological and hydrological conditions in the habitat in this season, which most probably demand intensified locomotion, orientation in the water column, and relevant provision for corresponding metabolic functions (energy supply and signal transduction in tissue by means of lipid-type mediators). In addition to their structural function, the minor PI and its physiologically active metabolites (inositol, triphosphate, diacylglycerols) are known to modulate the activity of the phosphotransferase protein kinase C, which is particularly important for cell growth and differentiation in young fish [71].

Thus, compensatory modifications in the composition of lipids and their FA components in the muscles and lipid sac of typically pelagic L3 stage postlarvae of the daubed shanny living under highly variable environment (primarily temperature and foraging conditions, as well as photoperiod) secure the viability of this Arctic fish species. Owing to some of its specific physiological and biochemical traits (e.g., the lipid sac in its postlarvae), the daubed shanny has a marked ecological role and significance in the marine ecosystem's food chains [72].

### **5. Conclusions**

This paper presents results on the seasonal dynamics of lipids and their FA components in the muscles and the lipid sac of postlarval *L. maculatus* inhabiting Kongsfjord in the summer, winter, and spring in the Arctic, with particular attention paid to the role of lipids in the adaptation of young fish to seasonal variations in temperature and photoperiods (abiotic factors), as well as an important biotic factors associated with the number, species composition, and availability of food objects. Biochemical analysis was carried out for the typically pelagic L3 developmental stage of the juvenile *L. maculatus*, which is predominantly zooplanktovorous.

Both general and numerous specific biochemical adaptation mechanisms of the compensatory and exploitative variety, which were found in daubed shanny juveniles, involve lipid participation, and their high plasticity can be explained by the evolutionary "youth" of this species, formed due to the movement of its Pacific ancestors to the North Atlantic about 3.0–3.5 million years ago [12], in addition to prolonged spatial continental isolation, which led to formation of an endemic species of the *Leptoclinus* genus [12].

Among specific adaptations, we have to highlight:


copepods in the studied fjords (mainly Kongsfjord, Svalbard) for the pelagic juveniles of the daubed shanny. Tracking the movement of food-derived FAs determines the basis for the chemoindication of qualitative and quantitative relationships between zooplankton and zooplanktophages in Arctic marine food chains;


Among the general adaptation mechanisms, a high unsaturation of lipids was shown for daubed shanny juveniles due to their FA components, characteristic of inhabitants of marine ecosystems of northern latitudes. A high degree of lipid unsaturation is supported, first, due to the dominance of MUFAs, among which oleic acid 18:1(n-9), that is present in the structure of almost all lipids, both structural and primarily storage lipids, is one of the main biochemical component. The second position is often competitively occupied by PUFAs and/or SFAs. In this case, we should mention the relationship between the results of this work and theoretical studies of the properties of FA chains (the shape and flexibility of FA chains) in an unperturbed state, obtained by computer simulation with the Monte Carlo method [63,64,73,74]. In particular, it was found that FA chains (16:0, 18:1(n-9)*cis*, 18:2(n-6)*cis*, 18:3(n-3)*cis*, 20:3(n-6)*cis*, 20:4(n-6) *cis*, 20:5(n-3)*cis*, 22:5(n-3)*cis*, 22:6(n-3)*cis*), i.e., key FAs for the studied fish, have equal longitudinal sizes due to which molecular substitutions, manipulations, and complementarity (in the properties) of these FAs in the PL structure achieve constant thickness of the membrane hydrocarbon layer and adjust its fluidity, which ensures the functionality of the lipid bilayer under changing environmental factors.

In addition, for fish of the Stichaeidae family, both juveniles and adults [75], lipid unsaturation is achieved due to PUFAs, especially EPA and DHA, in muscle tissue, which is consistent with the increased motor activity of fish in specific habitat conditions (currents, turbulent flows, etc.) and, ultimately, is directed to ensure homeostasis of the internal environment to maintain optimal activity of membrane-bound enzymes with a general increase in the level of metabolism. At the same time, attention should be drawn to the fact that this phenomenon is apparently based on a mechanism that is associated with certain properties: increased flexibility (which is associated with the fluidity of the biomembrane) for PUFAs, i.e., EPA and DHA, compared to SFAs. For example, with an increase in the metabolic load in fish muscle tissue in a specific environment (low temperatures, increased concentration of dissolved oxygen, fluctuations in salinity, etc.), the full functioning of biomembranes is maintained due to the control of such signs as thickness and resistance to mechanical action (with the active work of proteins) and the provision of a thermally insulating function for membrane-bound enzymes in the individual domains of the lipid bilayer. For young marine fish living at low temperatures, the increased level of EPA and DHA is due to their adaptive significance of overcoming the negative impact of fluctuations of this factor on the growth and development of juveniles. Thus, the experimental results of FA status in fish of the northern seas obtained in this work are consistent with the previously formulated concept of the functions of polyunsaturated chains [63,64,73,76].

The conducted studies allowed us to identify the general and specific features of the biology of the daubed shanny, an ecologically important representative of the Arctic ichthyofauna. Further research on the daubed shanny as a component of the Arctic trophic net will be focused on the study of the abilities and/or limitations of biosynthesis of certain MUFAs and PUFAs, including food-derived ones, that can compensate environmental changes and maintain the presence of the species in the northern marine ecosystems.

**Author Contributions:** The individual contributions of authors: conceptualization, S.A.M., N.N.N., J.B. and S.F.-P.; methodology, S.A.M., Z.A.N. and E.A.K.; software, S.A.M., S.N.P. and E.A.K.; validation, S.A.M., S.N.P..; formal analysis, S.A.M., Z.A.H., S.N.P. and E.A.K; investigation, S.A.M.; resources, S.A.M., J.B., S.F.-P.; data curation, S.A.M., Z.A.N., N.N.N. and S.F.-P.; writing—original draft preparation, S.A.M.; writing—review and editing, N.N.N., A.M.O. and S.F.-P.; visualization, S.A.M., S.N.P., E.A.K. and K.A.F.; supervision, N.N.N., J.B. and S.F.-P.; project administration, X.X.; funding acquisition, J.B., S.F.-P.. Authorship must be limited to those who have contributed substantially to the work reported. All authors have read and agreed to the published version of the manuscript.

**Funding:** The research was supported by the State budgetary theme 0218-2019-0076, RFBR project No. 17-04-00466, The President of the Russian Federation grant to support young researchers MK-2188.2020.4; the Research Council of Norway projects—SpitsEco, ES504895, and Arctic ABC, No 244319.

**Acknowledgments:** The biochemical analysis was performed at the Scientific Center of collective usage platform of the Karelian Research Centre of the Russian Academy of Sciences. The histological research was performed at the Research Resource Center for Molecular and Cell Technologies of Saint-Petersburg State University. Authors are deeply thankful to crew R/V Helmer Hanssen to their help during trawling and sampling.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Review*
