Lipid classes and fatty acid composition in two parasitic copepods Peroderma cylindricum and Lernaeocera lusci and their respective fish hosts Sardina pilchardus and Merluccius merluccius from the Tunisian waters

SUMMARY: The present study investigates the detailed lipid classes and their fatty acid (FA) compositions from two parasitic copepods Lernaeocera lusci and Peroderma cylindricum and their respective fish host species Merluccius merluccius and Sardina pilchardus . The lipid classes, including phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), triacylglycerol (TAG), wax ester/cholesterol ester (WE/CE), mono-diacylglycerol (MDG), and free fatty acids (FFA) were separated by thin layer chromatography. The results revealed that TAG and PC were the major lipid classes in parasites; while WE/CE and PS were the most abundant in hosts. As for FA composition, C16:0, C18:0, C18:1n-9, C20:5n-3, and C22:6n-3 were recurrently found to be dominant in all lipid classes of the different organisms studied. However, some differences concerning the abundance and the distribution of several FAs were observed. Overall, the obtained results highlighted that despite the quite strong trophic connection between the parasites and their respective hosts, the parasites could be distinguished by specific lipid profiles.


INTRODUCTION
Peroderma cylindricum (Heller, 1865) and Lernaeocera lusci (Bassett-Smith, 1896) are two common parasitic copepods belonging to the Pennellidae family which infect several marine fish. These parasites have complex and heteroxenous cycles that comprise different larval stages (Brooker et al., 2007). While the life cycle of L. lusci is well established, that of P. cylindricum remains only partially elucidated. The typical host species of L. lusci in the Tunisian coastal area are the sole Solea solea (intermediate host) and the European hake Merluccius merluccius is the definitive host (Kabata, 1979). Once passing through copepodid and chalimus developmental stages, L. lusci males and females attain maturity on the gills of their intermediate host. After copulation, the females leave the sole and swim actively to infect the definitive host where they embed deeply into the gill arches and continue to develop (Kabata, 1979). As for P. cylindricum, only the adult metamorphosed female embedded into pilchard Sardina pilchardus (definitive host) is known to date. It was reported that P. cylindricum inserts its holdfast into the pilchard's kidney and spine while its genital segment, producing two egg strings, emerges on the exterior (Becheikh et al., 1997). As a result of their attachment and feeding, these two hematophagous parasites P. cylindricum and L. lusci can affect the survival, physiology and fitness of their hosts (van Damme et al., 1994;Hajji et al., 1998). Particularly, these parasites were found to be able to inflect substantial alteration in their host's lipids (Hajji et al., 2015;Telahigue et al., 2017;Telahigue et al., 2019). However, little is known about their own lipid profiles.
Lipids and their building block fatty acids are fundamental components for animal health and survival. They are involved in a wide range of biological functions due to their complexity and structural diversity (Tracey et al., 2018). Neutral lipids, predominantly constituted by triacylglycerols (TAG) and wax esters (WE), have the primary function of energy storage and are widely related to the physiological status of the organism (Şen Özdemir et al., 2019). However, phospholipids and sterols serve as the principal structural constituents of cell membranes (Dufourc, 2008). They may also play significant roles in cell signaling pathways and serve as precursors to bioactive compounds such as steroid hormones (Tocher, 2003). The common phospholipid classes including phosphatidylethanolamine (PE), phosphatidylinositol (PI), phosphatidylcholine (PC), and phosphatidylserine (PS), possess significant structural diversity and complexity which mainly depend on the type of fatty acid (length of the fatty acid chain and number of double bonds) which is incorporated (Şen Özdemir et al., 2019;Zhou et al., 2020).
Owing to their involvement in these diverse cellular and physiological processes, lipids are attractive targets for parasites and are recognized as key players in host-parasite interactions (Vallochi et al., 2018). It was reported that various pathogens such as obligate hematophagous and intracellular parasites are able to hijack host's lipids to their own benefit in order to modulate and escape the host response and to complete their own life cycle (O'Neal1 et al., 2020). Although some authors have reported that the fatty acid composition of parasites may largely reflect that of their hosts (Tocher et al., 2010), others have highlighted that several parasites may have their own specific fatty acid fingerprints (Taraschewski et al., 1995).
The main objective of this study was to illustrate the lipid profiles of two common parasitic copepods from the Mediterranean waters: L. lusci and P. cylindricum, and to investigate whether these parasites share common patterns with their respective hosts or whether they have their own specific profiles. For this purpose, the lipid class fatty acid (FA) compositions of the two parasitic copepods as well as those of their respective host fish were studied. For more relevance, the specific fixation sites of these parasites (i.e. M. merluccius Gills for L. lusci and S. pilchardus kidney for P. cylindricum) were considered in this work. The obtained results will help to better elucidate the trophic connections in the two studied parasite-host systems and to bring new comprehensive knowledges about some biochemical aspects of L. lusci and P. cylindricum, which remain hitherto poorly understood.

Sample collection
A total of 243 specimens of Sardina pilhcardus (16.5±1.5 cm) and 276 specimens of Merluccius merluccius (18.3±2.2 cm) were purchased from fishermen at the port of Bizerte (Northeast of Tunisia). The samples were directly transported to the laboratory in polystyrene ice-cooled boxes. All specimens were carefully examined for the presence of P. cylindricum and L. lusci. Specimens from the two parasite species (adult females with egg strings) as well as M. merluccius gills and S. pilhcardus kidney tissues were collected, weighed, and stored at -20 °C until analysis.

Total lipid extraction and lipid classes' separation
Total lipids from the whole body (including egg strings) of both P. cylindricum and L. lusci and their respective hosts' tissues were extracted according to the method of Folch et al. (1957) using a mixture of the chloroform-methanol (2:1 V/V) solvent containing 0.01% butylated hydroxytoluene (BHT) as an antioxidant. An analysis of lipid classes was performed by one-dimensional double development high performance thin layer chromatography (HPTLC) as described by Olsen and Henderson (1996). Briefly, aliquots of 500 μl of lipid extracts were spotted onto a TLC plate (20x20 cm, silica gel 60, Merck, Germany), and hexane/diethyl ether/glacial acetic acid (80: 20: 2. V/V) was used as developing solvent system for the neutral lipid classes and methyl acetate/ isopropanol/chloroform/methanol/0.25% KCl (25: 25: 25: 10: 9. V/V), for the polar lipid separation. Lipid fractions were visualized under UV light after spraying with 0.1% 2'-7' dichloro-fluorescein in absolute methanol. Eight lipid classes, including phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI), phosphatidylserine (PS), triacylglycerol (TAG), wax ester/cholesterol ester (WE/CE), mono-diacylglycerol (MDG), and free fatty acids (FFA) were identified by comparison with known standards. Each lipid class band was then scraped from the TLC plates and subjected to transmethylation to yield fatty acid methyl esters (FAME) as given by Cecchi et al. (1985).

Fatty acid analysis
FAMEs were analyzed on a HP 6890 gas chromatograph with a split/splitless injector equipped with a flame ionization detector at 275 °C, and a 30 m HP Innowax capillary column with an internal di-ameter of 250 μm and a film thickness of 0.25 μm. The injector temperature was held at 250 °C. The oven temperature was programmed from 50 to 180 °C at a rate of 40 °C/min, then from 180 to 220 °C at 1.33 °C/min and to stabilize at 220 °C for 7 min. Nitrogen was the carrier gas. Methyl nonadecanoate 19:0 (Sigma) was added as internal standard. The identification of FAMEs was based on the comparison of their retention times with those of a mixture of methyl esters (SUPELCO PUFA-3 and Supelco 37 component FAME). Fatty acid peaks were integrated and analyzed using HP chemstation software.

Statistical analysis
The R software version 4.0.2. (R Core Team, 2020) was used to carry out statistical analyses. The normality of data distribution and homogeneity of variance were evaluated using the Shapiro-Wilk test and Levene's test, respectively. One-way analysis of variance (ANOVA) followed by the Bonferroni test were performed to check the significant differences between fatty acid amounts of lipid classes from the two studied parasites. The Bonferroni conservative correction method adjusts p values because of the increased risk in a type I error (significance level). Additionally, the hierarchical clustering analysis (HCA) was made by "FactoMineR" R package (Zhao et al., 2014) using Ward's method.

Lipid classes and their fatty acid composition from Lernaeocera lusci and its host Merluccius merluccius
The percentages of each lipid class in relation to total lipid content in the whole body of L. lusci and in the gills of M. merluccius are presented in Figures 1 and 2, respectively. TAG was the major lipid class in L. lusci, constituting ~46% of total lipids followed by PC (with ~26%) and PE (with ~11%). WE/CE was found to represent around 7%, while PI and MDG were found in lesser amounts (~1.5%) ( Figure 1). As for the host, it was found that WE/CE was by far the most dominant lipid class (up to 40%) followed by TAG and PS, which constituted around 16% of the total lipids. However, PC and FFA were found to be minor constituents, each representing around 3% of the TL in M. merluccius gill tissue ( Figure 2).
The fatty acid composition of the various lipid classes isolated from L. lusci and its host M. merluccius are respectively given in Tables 1 and 2. Overall, our results revealed that all L. lusci lipid classes were dominated by saturated fatty acids (SFA), which accounted for about 50% of the total FA followed by polyunsaturated fatty acids (PUFA), ranging from 28.41 to 39.93% and monounsaturated fatty acids (MUFA) by about 15% (Figure 2). Among the 33 FA species identified, palmitic (C16:0), stearic acid (C18:0), oleic acid (C18:1n-9), eicosapentaenoic acid (C20:5n-3, EPA) and docosahexaenoic acid (C22:6n-3, DHA) were recurrently found to be major in all lipid classes. Among the phospholipid classes, PC had significantly (p < 0.05) higher proportions of C14:0 (9.97%) than the other lipid classes. Significantly high amounts of DHA were also recorded in PC and PI fractions. In addition, substantial amounts of C16:0, reaching 30.81 and 32.83% of the total FA were recorded for PI and PS (p > 0.05). As for neutral lipids, TAG and FFA appeared to be richer in PUFA, mainly in terms of EPA and DHA when compared to the other lipid classes. The WE/ CE fraction was characterized by a higher amount     % Total lipids % Total lipids % Total lipids % Total lipids of C15:1 (4.33%, p < 0.05) and C22:5n-3 (9.14%), while MDG showed significantly elevated amounts of SFA, mainly C16:0, with 27.57%.
A quite different pattern was observed for the host, where the PUFA group was found to be dominant mainly in the phospholipid classes (with around 50%), followed by SFA (varying from 28.86% in PI to 37.73% in PE) and MUFA (ranging from 14.07 to 21.20% in PE and PI, respectively). Within saturates, C16:0 was the major FA in all lipid classes, exhibiting significantly highest levels in PS (27.16%) and FFA (29.9%) fractions (p < 0.05). Along with C16:0, C18:0 was also found to be abundant in all analyzed polar and neutral lipid fractions with a percentage exceeding 5% of total FAs (p > 0.05). Within the MUFA group, we particularly noticed significantly high amounts of C16:1n-7 in PC (~10%) and C18:1n-9 in PI, reach-ing 12.09% of total FAs. The latter FA was also found to be dominant in all neutral lipid classes (p > 0.05). With respect to PUFA, DHA was by far the most dominant FA in all lipid classes, varying from 14.58% in FFA to 20.72% in PE. In addition to DHA, EPA was also found at quite important percentages (up to 5%), mainly in phospholipid and FFA fractions. Another FA, C18:2n-6, was also found at significantly high amounts (around 10% of TFA) in both TAG and WE/CE fractions (p < 0.05).

Lipid classes and their fatty acid composition from Peroderma cylindiricum and its host Sardina pilchardus
PC and PE were the major lipid classes in P. cylindricum polar lipids, representing 30.63%, and 11.21% of TL, respectively. Among neutral lipids, a high proportion of TAG (41.38%) was found while WE/CE and FFA represented 7.75 and 5.20%, respectively. The remaining lipid classes (i.e. PI, PS, and MDG) occurred in smaller proportions (ranging from ~1 to 1.6%) (Figure 3). Regarding the host, we found that the lipid fraction of the S. pilchardus kidney was characterized by the dominance of the WE/ CE, which accounted for over 30% of TL followed by PS (18.16%), TAG (15.62%), PE (14.42%), and PC (10.26%). The other lipid classes, including MDG, FFA and PI were found to be minor components with proportions ranging from 6.27 to 3.30% and to 1.65% of TL, respectively (Figure 4). The fatty acid compositions of the different P. cylindricum lipid classes are presented in Table 3. The obtained results showed that almost all lipid class-es contained higher proportions of SFA than MUFA and PUFA, except for PC and PE fractions, where PUFA constituted the major group with 47.81 and 45.86%, respectively. Among the total of 28 identified FA, five (i.e. C16:0, C18:0, C18:1n-9, EPA and DHA) appeared to be dominant in all lipid fractions. Within phospholipids, PC and PE showed quite sim- All values are averages of triplicate analyses (n=3) performed on the pool of 9 samples. Values labelled with different letters in the same line by considering polar and neutral lipids separately, are significantly different (by the Bonferroni test at the 5% probability level). SFA (saturated fatty acids that have no double bonds), MUFA (monounsaturated fatty acids that contain one double bond), PUFA (polyunsaturated fatty acids that contain two or more double bonds). ilar compositions and exhibited the highest level of DHA, which constitutes almost 20% of total FAs (p < 0.05). C16:0 and C18:0 were mostly found in PI and PS (~33%). As for neutral lipid classes, large proportions of C16:0 (33.04%) and DHA (18.24%) were recorded in the TAG fraction. Furthermore, substantial amounts of DHA were also recorded in WE/CE, FFA, and MDG. The fatty acid profiles of polar and neutral lipid classes extracted from the host kidney are shown in Table 4. All neutral lipids were dominated by SFA (up to 50%), followed by PUFA and MUFA. C14:0, C16:0 and C18:0 were the major FAs of SFA group, while C18:1n-9 was the most abundant MUFA. In the PUFA group, C18:2n-6, EPA and DHA tended to be dominant, mainly in the FFA fraction. Regarding the polar lipid classes, PUFA percentages were higher than SFA and MUFA except for PS, where SFA dominated (by representing 42.28% of total FA). C16:0 and C18:1n-9 were the most abundant SFA and MUFA, culminating to 29.13 and 14.44% (p < 0.05) in PS, respectively. EPA and DHA were the major FA within the PUFA group with significantly higher levels of DHA in PC, PE, and PI. In addition, C18:2n-6 was found in relatively important amounts (up to 5%) in all phospholipid fractions.

Multivariate analysis
To provide an overview of similarities and discrepancies within each studied host-parasite system, hierarchical clustering analysis (HCA) was applied to the whole FA data set ( Figure 5). Figure  5a presents the dendrogram of HCA, giving information about L. lusci and its host M. merluccius. Two major groups, comprising two clusters each, were separated out in the resulting dendrogram. The first group was constituted by the hosta' TAG and PL classes, while the second group involved the remaining data. Interestingly, we found that among the second group, almost all neutral lipid classes (including FFA, WE/CE, and MDG) of the host clustered with the parasite phospholipid. Similarly, the dendrogram generated from P. cylindricum and S. pilchardus data revealed two major groups. The first one constituted two clusters related to the host neutral lipid moieties. The second group was composed of two distinct clusters: Cluster I comprised the polar lipids of the host; while cluster II grouped all lipid classes of P. cylindricum. ing upon several parameters such as species, type of food, latitude, season, developmental stage, and life cycle strategy (van der Meeren et al., 2008). Although a substantial amount of literature exists on free-forms, information on the lipid composition of parasitic forms is still limited. Thus, this study was conducted to explore similarities and differences among the lipid fractions and their fatty acid compositions in L. lusci and P. cylindricum, and their respective hosts. According to the obtained results, clear differences in the relative percentages of the different lipid classes were noticed between parasites and their respective hosts. Indeed, TAG turned out to be the most predominant neutral lipid class in both L. lusci and P. cylindricum, while WE/CE prevailed in fish tissues. These results are in accordance with those of Tocher et al. (2010) and Kotani (2006), who reported that parasitic and opportunistic feeder copepods store their lipids mainly as TAGs. However, it is worth noting that by feeding regularly on their host's blood, the parasitic copepod females are able to maintain a sufficient lipid content to sustain their own survival and to fuel maturation and egg production as recently reported in L. lusci by Hajji et al. (2021). Thus, it is thought that TAG is rather stored in lipovitellin as a main source of energy for the next stages (e.g. nauplii and infective copepodids) to ensure development, basic metabolism, swimming activity and infectivity. In this context Tocher et al. (2010) demonstrated that the eggs of parasitic copepods contain significantly higher proportions of TAG than adulate females. The same authors reported that, furthermore, due to the short time period between fish host infections, long-term energy deposits such as WEs are not required for parasitic copepods (Tocher et al., 2010). In the other hand, the high proportions of WE/CE recorded in the host's tissues (i.e. gills and kidney) could be explained by the essential role played by cholesterol in the physiological regulation of the physical properties of cell membranes (Díaz et al., 2016). Differences in the proportion of polar lipid classes were herein observed between the two studied parasites and their respective hosts. Although PC, and to a lesser extent, PE were found to be the major phospholipid classes in parasites, PS appeared among the most dominant phospholipid classes in host tissues. In line with this, Tocher et al. (2010) also reported that PC and PE were the ma-jor phospholipid classes in both the females and egg strings of parasitic caligid copepods belonging to the genus Lepeophtheirus. These key components of biological membranes are known for their functional and structural roles. Moreover, it was reported that PC is the principal lipid component in crustacean lipovitellin and may serve as metabolic energy storage for reproduction and embryonic development (Lee et al., 2006).
Regarding the fatty acid composition of the different lipid classes, a general pattern characterized by the dominance of five FAs comprising C16:0, C18:0, C18:1n-9, C20:5n-3 and C22:6n-3 was herein clearly observed in both copepods and fish. Such findings corroborate previous studies which emphasized that the FA profile of the parasite can largely reflect that of its host (Tocher et al., 2010;Telahigue et al., 2017;Hajji et al., 2021). However, some dissimilarities were noticed as for the abundance and the distribution of some fatty acids. For instance, SFA was found to be the major FA group in almost all lipid classes of the two parasites L. lusci and P. cylindricum, mainly due to the substantial level of palmitic acid (C16:0). Although C16:0 was also found in a considerable amount in the host's tissues, their lipid fractions (mainly polar ones) tended to be more unsaturated and affected by the high DHA and EPA percentages. Remarkably high proportions of saturated fatty acids were also recorded in the phospholipid molecular species of other parasites such as Isoparorchis hypselobagri which infect the catfish Wallago attuand (Mondal and Dey, 2013) and Paratenuisentis ambiguous, an intestinal helminth parasite in eels (Aitzetmüller et al., 1994). Next to SFA, relatively high proportions of PUFA were also recorded in all lipid molecular species of the two studied parasitic copepods. Interestingly, it was observed that L. lusci neutral lipid classes, mainly TAG and FFA were particularly rich in PUFA (~40%), mostly in terms of long-chain FA such as C20:5n-3, C22:5n-3 and C22:6n-3. This may reflect a direct diversion of these essential FA from the host by the adult female and their probable incorporation in their egg strings. In fact, these fatty acids are known for their key roles in the reproductive success, development, and somatic growth of copepods as reported by Arendt et al. (2005). It is worth noting that the high unsaturation level in FFA could also be considered as an indicator of a good conser-vation of the acyl lipids during the storage and/or extraction process. High proportions of PUFA (up to 40% of the total FAs) with substantial amounts of DHA and EPA were also recorded in P. cylindricum phospholipid classes, chiefly in PC and PE. This pattern, reflecting the importance of PUFAs as essential structural components of the cell membrane phospholipids, was also reported by Tocher et al. (2010) in the parasitic female lice and its egg strings. According to several authors, these components are involved in various physiological functions such as the modulation of physicochemical properties of the membrane, ion exchange and transport, and cell signaling (Tocher, 2003). Furthermore, it has been reported that some C20-PUFAs such as ARA and EPA, recognized as main precursor molecules for eicosanoids, are thought to be involved in the modulation/ suppression of the host's immune responses, which is crucial for parasite survival (Fast et al., 2004;Tocher et al., 2010).
The scarce available literature on the lipid profile of parasitic copepods has pointed out that neutral lipids (principally TAG) are characterized by high levels of MUFA (Lee, 1975;Tocher et al., 2010). This pattern contrasts with that observed in our study where monoenes were found to not exceed 20% of the total FAs in all lipid classes of both L. lusci and P. cylindricum. This seems to be reflective of the host's lipid profiles, where MUFAs were found to compose approximately 20% of the total FAs across almost all the analyzed lipid classes. According to our results, the MUFA group was mainly comprised of oleic acid (C18:1n-9). This FA, known as a potential metabolic energy source (Tocher, 2003), has been found to occur in high proportions in several free and parasitic forms of marine copepods (Lee, 1975;Tocher et al., 2010;Escribano and Pérez, 2010). Furthermore, this FA is generally considered a trophic marker for omnivorous and carnivorous copepods (Dalsgaard et al., 2003). Although the main source of oleic acid is food intake, its synthesis de novo or by conversion of 18:0 was also reported in copepods (Kattner et al., 1994).
To better visualize the trophic connection within each parasite-host system, hierarchical clustering analysis was applied. It was found that L. lusci lipid classes clustered with the neutral lipids of its host M. merluccius, which suggested that they were closely related. However, the lipid classes of P. cylindricum linked more with the phospholipids of its host S. pilchardus. These observations could be attributed to several parameters such as the host/parasite species, the fixation site, and the ability of each parasite to modulate its FA composition.
Overall, our data revealed that although the two studied host-parasite systems exhibited differences concerning the abundance of lipid classes as well as the distribution of some of their fatty acids, they appeared to share general similar FA patterns due to the strong trophic connection between them. Yet, to be more conclusive, further investigations (taking apart the eggs from the adult females) are needed to go deeper into the knowledge of their molecular biology and biochemistry in order to better understand host-parasite interactions.