Incorporation and Metabolism of Fatty Acids by Desaturation and Elongation In

3 the nematode, Panagrellus redivivus. Summary 26 The free-living nematode Panagrellus redivivus can be mass produced in monoxenic solid 27 culture on Saccharomyces cerevisiae and therefore could be useful as a live food for marine 28 fish or crustacean larvae in the rapidly expanding aquaculture industry. However, this will 29 depend on their lipid and fatty acid composition and so this was investigated in mass 30 produced P. redivivus grown on S. cerevisiae in three different media. Live nematodes were 31 also incubated with [1-14 C]-labelled fatty acids and their desaturation and elongation 32 determined. The combined results from the growth trials on different media and the 33 metabolic studies with labelled fatty acids indicated the presence of Δ9, Δ12, Δ6 and Δ5 fatty 34 acid desaturase activities, and elongase activities active towards C 18 , C 16 and shorter chain 35 fatty acids. The presence of Δ15, and therefore the ability to produce n-3 polyunsaturated 36 fatty acids, was suggested by the compositional data, but could not be conclusively 37 established from metabolic studies. 38 39 Keywords: live food, mass produced nematodes, fatty acid metabolism, 40 41 During their early stages of development, many fish and crustacean species important for 41 marine aquaculture rely on live food organisms (Sargent et al., 1995). The most commonly 42 used live food is the brine shrimp Artemia salina, a small branchipod crustacean representing 43 approximately 40% of the total aquaculture demand for live feeds for early stages (Lavens 44 and Sorgeloos, 2000). Following one of the worst harvests of recent times from the premier 45 source of Artemia cysts, the Great Salt Lakes in the United States, cyst production was barely 46 sufficient to satisfy the increasing demand of the rapidly growing aquaculture industry 47 (Sorgeloos et al., 2001). The lack of potential alternatives to Artemia may become an 48 obstacle to a further increase of aquaculture production especially in developing countries. It 49 has been shown that the free-living nematode Panagrellus redivivus is a suitable food for fish 50 have been proven to be an excellent food source, their use has not become widespread due to 53 problems involved in mass production. However, a low-cost technology for the mass 54 production of P. redivivus on solid medium was recently described by Ricci et al. (2003). 55 The nutritional value of nematodes can be influenced by the culture medium. For instance, 56 …

During their early stages of development, many fish and crustacean species important for marine aquaculture rely on live food organisms (Sargent et al., 1995).The most commonly used live food is the brine shrimp Artemia salina, a small branchipod crustacean representing approximately 40% of the total aquaculture demand for live feeds for early stages (Lavens and Sorgeloos, 2000).Following one of the worst harvests of recent times from the premier source of Artemia cysts, the Great Salt Lakes in the United States, cyst production was barely sufficient to satisfy the increasing demand of the rapidly growing aquaculture industry (Sorgeloos et al., 2001).The lack of potential alternatives to Artemia may become an obstacle to a further increase of aquaculture production especially in developing countries.It has been shown that the free-living nematode Panagrellus redivivus is a suitable food for fish (Kahan and Appel, 1975;Kahan et al., 1980) and crustacean larvae (Biedenbach et al., 1989;Kumlu and Fletcher, 1997;Kumlu et al., 1998;Wilkenfeld et al., 1984).Although nematodes have been proven to be an excellent food source, their use has not become widespread due to problems involved in mass production.However, a low-cost technology for the mass production of P. redivivus on solid medium was recently described by Ricci et al. (2003).
The nutritional value of nematodes can be influenced by the culture medium.For instance, lipid content and fatty acid composition of nematodes can be modified by adding lipid components to the culture medium (Rouse et al., 1992;Kumlu et al., 1998;Schlechtriem et al., 2004a,b).P. redivivus cultured on simple oat-based medium were found to feed mainly upon the yeast growing on the medium and possibly on the breakdown products of oats.
Lipids extracted from such nematodes contained highly unsaturated fatty acids (HUFA, fatty acids having carbon chain lengths of ≥C 20 and with ≥3 double bonds; n-x signifying the position of the double bond from the methyl end of the molecule) like arachidonic (20:4 n-6) and eicosapentaenoic acid (20:5 n-3) although no C 20 HUFA were found either in oats or in the yeast (Sivapalan and Jenkins, 1966;Schlechtriem et al., 2004b).Similar results were obtained by Lower et al. (1970) who cultured P. redivivus axenically (bacteria-free) on an aqueous medium composed of heated liver extract and soy-peptone yeast.Therefore, P. redivivus appear to possess fatty acid desaturase and elongase activities necessary to synthesize several HUFAs from shorter chain fatty acid precursors, as previously described for the free-living nematodes Caenorhabditis elegans (Hutzell and Krusberg, 1982;Watts and Browse, 2002) and Turbatrix aceti (Rothstein and Götz, 1968;Fletcher and Krusberg, 1973).Thus, the fatty acid pattern of the nematodes is influenced by the fatty acid composition of the culture medium and the nematodes' capability to synthesize HUFAs.In this way, both factors also influence the nutritive value of P. redivivus as live food for first feeding fish larvae.For instance, arachidonic acid and eicosapentaenoic acid are important as structural components of membrane glycerolipids and as precursors of families of signalling molecules including prostaglandins, thromboxanes, and leukotrienes in fish (Sargent et al., 2002;Tocher, 2003).To assess the effect of endogenous HUFA biosynthesis on the fatty acid composition of P. redivivus, the pathway of HUFA synthesis and the complement of fatty acid desaturase/elongase enzymes must be elucidated.In addition, it was not clear from the studies above whether P. redivivus was able to directly synthesise polyunsaturated fatty acids (PUFA, fatty acids with two or more double bonds) de novo, that is, produce 18:2 n-6 from 18:1 n-9.
In the present study, the effect of different culture media on the lipid content, lipid class composition and fatty acid composition of mass produced P. redivivus was examined.The nematodes were grown on yeast, Saccharomyces cerevisiae, in three different media.One medium represented a low lipid medium with little added lipid.The other two media contained, respectively, very high 18:2 n-6 (66.8% of total fatty acids in sunflower oil), to investigate PUFA metabolism, and high medium chain saturates (10:0, 42.1% of total fatty acids in MCT oil) to investigate effects on C 16 and C 18 metabolism.Differences in the fatty acid composition of total polar and neutral lipids were analysed.To further investigate the pathway of PUFA and HUFA biosynthesis in P. redivivus, live nematodes were incubated with different [1-14 C]-labelled fatty acids and their further metabolism by desaturation and elongation determined.

Lipid extraction and lipid class composition
Total lipid contents of nematodes and growth medium samples were determined gravimetrically after extraction by homogenization in chloroform/methanol (2:1, v/v) containing 0.01% butylated hydroxytoluene as antioxidant, basically according to Folch et al. (1957) (Henderson and Tocher 1992).

Fatty acid analysis
Samples of total lipid (2 mg) were applied as 2 cm streaks to thin-layer chromatography plates, and polar lipids separated from neutral lipids using hexane/diethyl ether/acetic acid (90:10:1, by vol.) as developing solvent.The origin area corresponding to total polar lipids and the lane above the origin corresponding to total neutral lipids were scraped into stoppered glass test tubes for transmethylation directly on the silica (Christie, 1982).Fatty acid methyl esters of total lipid, total polar lipids, and total neutral lipids, purified as above, were prepared by acid-catalyzed transesterification using 2 ml of 1% H 2 SO 4 in methanol plus 1 ml toluene as described by Christie (1982) and methyl esters extracted and purified as described previously (Tocher and Harvie, 1988).Fatty acid methyl esters were separated and quantified by gas-liquid chromatography (Fisons GC8600, Fisons Ltd., Crawley, U.K.) using a 30m x 0.32 mm capillary column (CP wax 52CB; Chrompak Ltd., London, U.K).
Hydrogen was used as carrier gas and temperature programming was from 50 o C to 180 o C at 40 o C/min and then to 225 o C at 2 o C/min.Individual methyl esters were identified by comparison to known standards and by reference to published data (Ackman, 1980).

Assay of fatty acyl desaturation/elongation activities
Lipids were extracted from labelled nematode pellets using ice-cold chloroform/methanol (2:1, v/v) containing 0.01% (w/v) butylated hydroxytoluene essentially as described by Folch et al. (1957) and as described in detail previously (Tocher et al., 1988).Total lipid was transmethylated and fatty acid methyl esters prepared as described above.The methyl esters were redissolved in 100 µl isohexane containing 0.01% butylated hydroxytoluene and applied as 2.5 cm streaks to thin-layer chromatography plates impregnated by spraying with 2 g silver nitrate in 20 ml acetonitrile and pre-activated at 110 o C for 30 min.Plates were fully developed in toluene/acetonitrile (95:5, v/v) (Wilson and Sargent, 1992).
Autoradiography was performed with Kodak MR2 film for 6 days at room temperature.
Areas of silica containing individual PUFA were scraped into scintillation mini-vials containing 2.5 ml of scintillation fluid (Ecoscint A, National Diagnostics, Atlanta, Georgia) and radioactivity determined in a TRI-CARB 2000CA scintillation counter (United Technologies Packard, U.K.).Results were corrected for counting efficiency and quenching of 14 C under exactly these conditions.

Protein determination
Protein concentration in nematode suspensions was determined according to the method of Lowry et al. (1951) after incubation with 0.4 ml of 0.25% (w/v) SDS/1M NaOH for 45 min at 60 o C.
Thin-layer chromatography (20 cm x 20 cm x 0.25 mm) and high-performance thin-layer chromatography plates (10 cm x 10 cm x 0.15 mm), precoated with silica gel 60 (without fluorescent indicator) were obtained from Merck (Darmstadt, Germany).All solvents were HPLC grade and were obtained from Fisher Scientific U.K., Loughborough, England.

Statistical analysis
Data recorded as percentages were arcsine-transformed to ensure a normal distribution and subjected to analysis of variance (ANOVA).Duncan's Multiple Range Test (DMRT) was used to identify differences among treatment means (P<0.05)(STATISTICA 5.1 software).

Fatty acid compositions of growth media components
All nematodes were produced by culture on the yeast, Saccharomyces cerevisiae, but in three different media.The fatty acid composition of the basic yeast contained over 50% monounsaturated fatty acids, predominantly 18:1 n-9 and 16:1 n-7, nearly 18% saturated fatty acids, predominantly 16:0, and 32% PUFA composed of over 30% 18:2 n-6, and only 1% 18:3 n-3 and virtually no HUFA (Table 1).The LLM medium contained a small amount of yeast extract which contained a similar level of monounsaturated fatty acids including 13% 24:1 n-9, but a higher level of saturates and lower PUFA, than the yeast.The main PUFA in the yeast extract was 18:2 n-6 but there was 3.6% 20:3 n-6 and a small amount of 20:5 n-3.The LOM medium contained oat flour (OAT) and sunflower oil (SFO), both of which were rich in 18:2 n-6, 40% and 67%, respectively, with the remaining fatty acids being 18:1 n-9 and saturates, with virtually no n-3 PUFA (Table 1).The MTM medium was enriched with medium-chain triglycerides (MCT) which had over 42% 10:0 in a total of 54% saturates along with 30% 18:2 n-6 and 14% 18:1 n-9 (Table 1).

Effects of growth media on lipid content and lipid class composition of P. redivivus
The lipid content of P. redivivus grown in media supplemented with lipid was significantly increased in comparison to nematodes grown in the low lipid medium with the order being MTM > LOM > LLM (Table 2).The increased lipid content was accompanied by increased proportions of triacylglycerol and total neutral lipid although the percentage of triacylglycerol in nematodes grown in MTM was lower than that of nematodes grown in LOM despite having a higher lipid content suggesting an increase in the absolute amount of polar lipids in P. redivivus grown in MTM (Table 2).
The proportion of saturated fatty acids in total fatty acids was slightly lower in nematodes grown on LOM, whereas it was significantly increased in nematodes grown in MTM, compared to growth in LLM (Table 3).Growth in both lipid-supplemented media resulted in the level of 16:0 exceeding that of 18:0 compared to nematodes grown in LLM.Saturated fatty acids after growth in MTM also resulted in higher proportions of shorter chain fatty acids in total fatty acids, including 14:0, 12:0 and 10:0, but also 18:1 n-7, 18:1 n-9 and 16:1 n-7 with decreased proportions of 18:0, 20:3 n-6, 20:4 n-6 and DMA.The proportion of 18:2 n-6 and n-3PUFA in total fatty acids were not greatly affected by growth in MTM compared to growth in LLM (Table 3).

Effects of growth medium on fatty acid compositions of polar and neutral lipids
The polar lipid of P. redivivus grown in LLM contained higher proportions of n-6PUFA and n-3PUFA and lower proportions of saturated and monounsaturated fatty acids compared to neutral lipids (Table 4).Growth in LOM increased the proportion of 18:2 n-6 in both polar and neutral lipids whereas the increased 18:1 n-9 was only observed in neutral lipids.The proportions of 20:3 n-6, 20:4 n-6 and 20:5 n-3 and total n-3PUFA were reduced in both polar and neutral lipids in nematodes grown in LOM compared to growth in LLM (Table 4).In contrast, the fatty acid composition of polar lipids of nematodes grown in MTM were relatively unaffected by growth in MTM compared to growth in LLM (Table 4).However, the proportions of saturated and monounsaturated fatty acids were increased, and those of n-6 and n-3PUFA decreased, in P. redivivus grown in MTM compared to growth in LLM (Table 4).
Consistent features observed, irrespective of growth medium, were that 18:0 exceeded 16:0 in polar lipids, whereas the opposite was the case in neutral lipids, and similarly 18:1 n-7 always exceeded 18:1 n-9 in polar lipids whereas this ratio was affected by medium composition (diet) in neutral lipids (Table 4).DMA were only found in polar lipids, reflecting their origin from ether-linked phospholipid classes and their levels were relatively unaffected by growth medium.

Metabolism of 14 C-labelled fatty acids in P. redivivus
Total lipid was extracted from P. redivivus after incubation with radiolabelled fatty acids.

Discussion
The fatty acid composition of total lipid of P.redivivus grown on S. cerevisiae in LLM (Table 3) shows some interesting features such as 18:0 > 16:0 and 18:1 n-7 > 18:1 n-9, the opposite to the situation in higher animals, and in the yeast on which it was grown in which 16:0 and 18:1 n-9 both greatly exceeded 18:0 and 18:1 n-7, respectively.The composition of all the other media components were the same as the yeast in this respect (Table 1) which was most interesting, as 18:0 > 16:0 and 18:1 n-7 > 18:1 n-9 was consistently observed in the polar lipid fraction of P. redivivus, representative of the membrane lipids, irrespective of medium or diet.In contrast, the fatty acid composition of the neutral lipids did not show the same characteristic pattern in the saturated fatty acids (as 16:0 > 18:0) but 18:1 n-7 did exceed 18:1 n-9, although this could be changed by diet as shown with the LOM medium (oatmeal and sunflower oil) which was rich in 18:1 n-9.Therefore, the pattern of 18:0 > 16:0 and 18:1 n-7 > 18:1 n-9 are inherent features of P. redivivus.This appears to extend to related nematodes such as Caenorhabditis elegans, which also shows this pattern (Hutzell and Krusberg, 1982;Tanaka et al., 1996).In the former study, C. elegans were grown on a liver extract/yeast extract/soy peptone medium that was devoid of 18:1 n-7, although 18:0 exceeded 16:0, whereas in the latter study, C. elegans were grown on E.coli in which 18:1 n-7 was the predominant monoene, but 16:0 and not 18:0 was the main saturated fatty acid, but in both cases C. elegans maintained the 18:0 > 16:0 and 18:1 n-7 > 18:1 n-9 pattern despite the "dietary" influence.
The data obtained from the experiments with radioactively labelled fatty acids clearly show that a major fate of each of the fatty acids was β-oxidation.This is the explanation for the radioactivity that was recovered in fatty acids with shorter chain lengths, or more saturated, than the labelled substrate fatty acid as one round of β-oxidation would remove the labelled carbon from the fatty acid in [1-14 C]-labelled fatty acids.Thus, with [1-14 C]18:0, over 4% of radioactivity was recovered in 16:1 n-7 indicating that 18:0 was also metabolised by β-oxidation to produce 14 C-labelled acetyl-CoA which was recycled by fatty acid synthase (FAS), or an elongase, producing labelled 16:0 (although this could not be resolved from 18:0 by the chromatographic procedure) which was subsequently desaturated to 16:1 n-7.It was noteworthy that within a chain length (C 18 or C 20 ) the amount of radioactivity recovered as recycled fatty acid products increased with increasing unsaturation of the labelled substrate fatty acid suggesting that P. redivivus may prefer to β-oxidise PUFA compared to saturated and monounsaturated fatty acids.However, it could be that lower amounts of labelled acetyl-CoA were recycled with 18:0 or 18:1 n-9 than with PUFA as the data are expressed as percentages of recovered fatty acids.Although differences in recycling cannot be discounted, recoveries were generally comparable between the different fatty acids suggesting that it cannot be a full explanation.
The data clearly suggest that P. redivivus express a wide range of desaturation and elongation activities.In interpreting the data in the present study, it is necessary to be familiar with the pathways for synthesis of HUFA and the range of desaturases that are commonly found in either plants or animals (see Fig. 1; note that this figure is not representative of any one species).Production of labelled 18:1 n-9 from [1-14 C]18:0 and of 16:1 n-7, presumably arising by desaturation of 16:0, labelled as a result of recycling (see above) indicates Δ9 desaturase activity.Production of 18:2 n-6 from [1-14 C]18:1 n-9 indicates significant Δ12 desaturase activity.The presence of a Δ15 (or n-3) desaturase activity (responsible for the production of 18:3 n-3 from 18:2 n-6 in plants) is more difficult to establish as it is not possible to distinguish between 18:3 n-3 (Δ15 product) and 18:3 n-6 (a Δ6 desaturase product) and the same applies to all similar pairs such as 20:3 n-6/20:3 n-3 or 20:4 n-6/20:4 n-3.The fatty acid composition data obtained from the growth media studies suggested that P. redivivus may express Δ15 desaturase activity as significant amounts of n-3fatty acids, especially 20:5 n-3, were present despite these being very low in the growth media.
However, growth in LOM, containing very high 18:2 n-6, did not result in increased levels of 18:3 n-3 or n-3PUFA, with the proportion of 20:5 n-3 significantly decreasing in all lipid classes.In contrast, the presence of Δ6 desaturase activity is strongly suggested by the recovery of radioactivity as 18:4 n-3 when P. redivivus was incubated with [1-14 C]18:3 n-3 suggesting that at least some of the radioactivity recovered in 18:3 when P. redivivus was incubated with [1-14 C]18:2 n-6 will be 18:3 n-6, the Δ6 product.
The data obtained from P.redivivus incubated with [1-14 C]18:3 n-3 were also consistent with the presence of a Δ5 desaturase activity.This is because there is no known pathway for conversion of 18:3 n-3 into n-6 fatty acids such as 20:4 n-6 or 22:5 n-6 and it is unlikely for the amounts of radioactivity recovered in these fractions to have arisen solely from recycling.
Thus, the radioactivity recovered from [1-14 C]18:3 n-3 is almost certain to be as 20:4 n-3 and 20:5 n-3 indicating the presence of Δ5 desaturase (20:4 n-3 to 20:5 n-3).However, the best evidence for significant Δ5 activity is the high level of 20:4 n-6 in P. redivivus grown on S. cerevisiae in LLM as neither the yeast nor the yeast extract contained any significant amount of 20:4 n-6 suggesting its presence in the nematode was the result of Δ5 action on 20:3 n-6 either supplied directly or as a result of Δ6 desaturation and elongation of 18:2 n-6.
In comparison to Δ5 activity, where supporting evidence for significant activity can be demonstrated, support for Δ6* activity is lacking.The presence of 22:6 n-3 was not observed in P. redivivus under any conditions suggesting that the recovery of traces of radioactivity in the position corresponding to 22:6 n-3 was due to an unknown component or artifactual.The predominant n-3 fatty acid and, indeed, the most abundant fatty acid in C. elegans phospholipids grown on E. coli was 20:5 n-3, and no 22:6 n-3 was reported (Tanaka et al., 1996).Similarly, 20:5 n-3 was the most abundant C 20 HUFA in Steinernema carpocapsae when grown on artificial diet supplemented with lard or linseed oil, and no 22:6 n-3 was found (Fodor et al., 1994).
The presence of a Δ12 activity in an animal like P. redivivus is not without precedent.In an earlier study, a Caenorhabditis elegans cDNA encoding a Δ12 fatty acid desaturase was identified and characterized (Peyou-Ndi et al., 2000).
Although we found no unequivocal evidence for a Δ15 (n-3) desaturase activity in P.
Interestingly, the C. elegans Δ15 desaturase was actually revealed to be an ω3 desaturase as it desaturated both C 18 and C 20 n-6 substrates to the corresponding n-3 fatty acids (Spychalla et al., 1997).
Increased levels of 14:0 and 16:0 in P. redivivus cultured in MTM is evidence for elongation of 10:0 through to 16:0.The presence of C 18-20 elongase activity was also clearly demonstrated by the recovery of significant amounts of radioactivity as 20:1 when P.
redivivus was incubated with [1-14 C]18:0 or [1-14 C]18:1 n-9.However, the elongase activity towards C 20 fatty acids would not appear to be as high as towards shorter chain substrates.This is confirmed by the fatty acid composition data that showed very little evidence for C 22 fatty acids in P. redivivus under any conditions.An enzyme catalyzing the elongation of fatty acids, ELO-1, has been cloned and functional characterised from the nematode C. elegans, and heterologous expression in yeast showed it was predominantly active on C 18 PUFA with virtually no activity towards C 20 PUFA (Beaudoin et al., 2000, Watts andBrowse, 2002).
The data from the metabolic studies using radiolabelled fatty acids can perhaps help to explain the characteristic "normal" fatty acid composition of P. redivivus with 18:0 > 16:0 and 18:1n7 > 18:1 n-9 as discussed above.A possible explanation is that the desaturase and elongase enzymes are more efficient with C 16 compared to C 18 fatty acids, and therefore, 16:0, produced by fatty acid synthetase, is readily desaturated to 16:1 n-7 and elongated to 18:0 and that subsequent elongation of 16:1 n-7 to 18:1 n-7 is more efficient than desaturation of 18:0 to 18:1 n-9.The result of these differential activities being that 18:0 and 18:1 n-7 tend to accumulate more than 16:0 and 18:1 n-9.Growth in medium such as LOM, with very high 18:1 n-9 from oatmeal and sunflower oil, increases triacylglycerol rich in 18:1 n-9, thus reversing the ratio of 18:1 n-9 to 18:1 n-7 in neutral lipid.In contrast, MTM is characterized by high saturated fatty acids, particularly 10:0 which is presumably efficiently elongated to 16:0 in P. redivivus, greatly increasing 16:0 and reversing the 18:0 to 16:0 ratio, but also increasing production of 18:1 n-7 through conversion of 16:0 to 16:1 n-7 to 18:1 n-7 and so 18:1 n-7 still exceeds in both polar and neutral lipid in P. redivivus grown in MTM.
The Δ6* enzyme acting on C 24 fatty acids may or may not be the same enzyme (Δ6) that acts on C 18 fatty acids.

Figure legends Figure 1 .
Figure legends . Separation of lipid classes was performed by high-performance thin-layer chromatography.Approximately 10 µg of lipid extract was loaded as a 2 mm streak and the

Table 1 .
Fatty acid composition (percentage of total fatty acid by weight) of lipid-containing ingredients of media for culture of the nematode Panagrellus redivivus.

Table 2 .
Lipid content (mg lipid/ g dry mass) and lipid class compositions (percentage of total lipid) of Panagrellus redivivus grown on different culture media

Table 3 .
Fatty acid composition (percentage of total fatty acids) of Panagrellus redivivus grown on different culture media.

Table 4 .
). Significance of differences between means were determined by one-way ANOVA followed, where appropriate, by Tukey's multiple comparison test as described in the Materials and Methods.Values within a row with a different superscript letter are significantly different (P < 0.05).nd, not detected; LLM, lipid-free medium; LOM, lipid enriched oat based medium; MTM, medium-chain triglyceride medium; nd, not detected; PUFA, polyunsaturated fatty acids; DMA, dimethylacetals.Fatty acid compositions (percentage of total fatty acids) of total polar and total neutral lipids from Panagrellus redivivus grown on different culture media.

Table 5 .
). Significance of differences between meanswere determined by one-way ANOVA followed, where appropriate, by Tukey´s multiple comparison test as described in the Materials and Methods.Values within a row with a different superscript letter are significantly different (P<0.05).LLM, low lipid medium; LOM, lipid enriched oat based medium; MTM, medium chain triglyceride medium; DMA, dimethyl acetals; nd, not detected; PUFA, polyunsaturated fatty acids.Metabolism of 14 C-labelled fatty acids by Panagrellus redivivus