Nutrition 29 (2013) 313–317
Contents lists available at ScienceDirect
Nutrition journal homepage: www.nutritionjrnl.com
Basic nutritional investigation
Brain lipid composition in rabbits after total parenteral nutrition with two different lipid emulsions -Botejara M.D., Ph.D. a, Jose Miguel Mora n-Penco M.D., Ph.D. b, *, Enrique Macia Marıa Teresa Espın-Jaime M.D., Ph.D. c, Francisco Botello-Martınez M.D., Ph.D. c, s Salas-Martınez M.D., Ph.D. c, Marıa Jesu s Caballero-Loscos Ph.D. d, Jesu e ndez B.Math., Ph.D. Manuel Molina-Ferna a
Internal Medicine Service, Hospital Perpetuo Socorro, Complejo Hospitalario Universitario de Badajoz, Badajoz, Spain Surgery Service, Faculty of Medicine, University of Extremadura, Badajoz, Spain c General Surgery Service, Hospital Infanta Cristina, Complejo Hospitalario Universitario de Badajoz, Badajoz, Spain d Department of Pharmacology, Faculty of Medicine, University of Extremadura, Badajoz, Spain e Department of Mathematics, Faculty of Sciences, University of Extremadura, Badajoz, Spain b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 April 2012 Accepted 24 July 2012
Objective: To study the changes occurring in brain lipid composition after the administration of total parenteral nutrition (TPN) by comparing two lipid emulsions, one with long-chain triacylglycerols (LCT) and the other with long-chain and medium-chain triacylglycerols (MCT/LCT 50%/50%). Methods: We used 21 young New Zealand rabbits divided into three groups of seven animals each. Two groups were subjected to TPN for 7 d, with each group receiving using one of two different lipid emulsions: Intralipid 20% (group LCT) and Lipofundin MCT/LCT 20% (group MCT/LCT). The third control group received an oral diet and underwent the same surgical procedure with the administration of intravenous saline solution. The energy administered in the TPN formulas was non-protein 100 kcal ∙ kg1 ∙ d1, with 40% corresponding to fats. Results: There were modest increases in plasma cholesterol and triacylglycerols. In the brain tissue, there was a decrease of phosphatidylcholine in animals with TPN, which was greater in group LCT. There were no signiﬁcant differences in the overall percentage distribution of brain fatty acids among the groups. Conclusion: The lipid emulsions administered in TPN, especially those prepared exclusively with LCT, cause changes in the brain lipid polar fractions of young rabbits. Ó 2013 Elsevier Inc. All rights reserved.
Keywords: Brain lipids Lipid emulsions Total parenteral nutrition
Introduction The brain has a high lipid content, as much as 60% of its dry weight. The fatty acids (FAs) of the brain phospholipids, especially long-chain monounsaturated FAs, are responsible for maintaining the structure of membranes, whereas polyunsaturated FAs control their ﬂuidity . These FAs are subject to continual renewal. In fact, the changes in maternal dietary lipids in pregnancy or lactation cause changes in the FA proﬁle of the brain of the offspring at birth and in early postnatal development [2,3]. Also, the adult brain can be inﬂuenced by speciﬁc exogenous FA contributions . These changes involve not only the * Corresponding author. Tel./fax: þ34-924-289-471. n-Penco). E-mail address: [email protected]
(J. M. Mora 0899-9007/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nut.2012.07.020
overall amount of polyunsaturated FAs administered but also the
u-6/u-3 ratio . Thus, dietary fat intake determines the FA composition of the brain and, hence, may inﬂuence neurologic function and learning ability, as has been demonstrated in some animals [6–8] and, less often, in humans . Also, brain lipids may be altered in different pathologies [10,11]. The lipid emulsions used in artiﬁcial nutrition, as commonly applied in everyday clinical practice, are also a source of exogenous fat. However, there have been few studies on the changes in brain lipid composition after the use of artiﬁcial nutrition, except those by Martınez and Ballabriga  in newborns and infants et al.  on fat overload in adult rabbits using and Macia Intralipid. Because parenteral nutrition including lipids is common therapeutic practice, its possible effects on brain
-Botejara et al. / Nutrition 29 (2013) 313–317 E. Macia
structure and function need to be better understood. We therefore studied the effects that parenteral nutrition with two types of fat and for a relatively short period might have on brain lipid composition in young rabbits. Materials and methods The animals used were 21 New Zealand rabbits (1700 200 g) housed in individual cages at an ambient temperature of 21 2 C, with an air-renewal rate of 14 changes per hour and light/dark cycles of 14/10 h. The animal handling followed the recommendations of the International Guiding Principles for Biomedical Research Involving Animals. Experimental groups One group (n ¼ 7) received long-chain triacylglycerols (LCTs) and total parenteral nutrition (TPN) for 7 d through a silicone catheter (outer diameter 0.65 cm, inner diameter 0.30 cm; Dow Corning Corporation, Midland, MI, USA) implanted in the internal jugular vein to end in the interscapular zone and connected to the infusion line by a swivel. The lipid source given was Intralipid 20% (Kabi-Pfrimmer, Stockholm, Sweden). Another group (n ¼ 7) received 50% medium-chain triacylglycerols and 50% LCTs and TPN under the same conditions as group LCT, except that the lipid source used was Lipofundin MCT/LCT 20% (Braun Medical SA, Rubi, Barcelona, Spain; group MCT/LCT). The control (n ¼ 7) group underwent the same surgery as groups LCT and MCT/LCT, but a 0.9% NaCl solution instead of TPN was administered through the catheter. The animals were provided food and water ad libitum, with the feed composition being 15% protein, 2.5% fats, 17% cellulose, 12% ash, 14% starch, and 2.45% minerals (BK Universal G.J. S.L., Tarragona, Spain).
Table 1 Composition of Intralipid 20% and Lipofundin MCT/LCT 20%
Soy seed oil (g/100 mL) MCTs (g/100 mL) Egg yolk phospholipids (g/100 mL) Lecithin (g/100 mL) Glycerol (g/100 mL) Phosphorus (mmol/L) Water (mL) Osmolarity (mOsm/L) Energy (kcal/L) Fatty acids (%) Caproic acid (C-6) Caprylic acid (C-8) Capric acid (C-10) Lauric acid (C-12) Myristic acid (C-14) Palmitic acid (C-16) Palmitoleic acid (C-16:1) Stearic acid (C-18) Oleic acid (C-18:1) Linoleic acid (C-18:2 u-6) Linolenic acid (C18:3 u-3) Arachidic acid (C-20) Arachidonic acid (C-20:4 u-6)
20 – 1.2 2 2.25 15 100 330 2000
10 10 1.2 1.2 2.5 15 100 380 1908
– – – – 0.07 21.31 0.22 9.64 29.42 32.97 4.21 0.1 0.1
0.076 25.71 20 1.14 – 7.14 – 2.38 13.33 25.71 3.81 – –
LCT, long-chain triacylglycerol; MCT, medium-chain triacylglycerol
Polar lipid analysis Anesthesia The animals were anesthetized with ketamine (Ketolar, Parke-Davis Laboratory, Morris Plains, NJ, USA) 25 mg/kg and phenothiazine (Combelen, Laboratories Bayer, Germany) 1 mg/kg, which were administered intramuscularly. Artiﬁcial nutrition formulas The two nutritional formulas administered to groups LCT and MCT/LCT contributed non-protein energy 100 kcal ∙ kg1 ∙ d1, with 60% corresponding to glucose (Glucoibys 40%, Ibys, Madrid, Spain) and 40% to lipids. The lipid source used in group LCT was an emulsion of LCTs (Intralipid 20%, Kabi-Pfrimmer), and that in group MCT/LCT was a 50%/50% emulsion of LCTs and MCTs (Lipofundin MCT/LCT 20%, B. Braun Medical S.A.). The composition of the two lipid emulsions is presented in Table 1. The TPN administered to these two groups also contained amino acids 3 g ∙ kg1 ∙ d1 (Trophamine 6%, Farmiberia S.A., Madrid, Spain) plus standard amounts of electrolytes, trace elements, and vitamins. The ﬁnal osmolality was 650 mOsm/kg. The infusion rate was 5 mL/kg of weight during the ﬁrst 24 h and 10 mL/kg of weight during the next 7 d. Biological controls All animals were weighed at the beginning and at the end of the experiment. Daily evaluations were made of the animals’ clinical status, and glycemia was quantiﬁed using test strips (Reﬂolux, Boehringer, Mannheim, Germany). Processing of animals At the end of the seventh day of the artiﬁcial nutrition, the animals were anesthetized, and a midline laparotomy was performed. Samples of blood 10 mL were collected by direct puncture of the vena cava. The brain tissue and cerebellum were extracted through a medial craniotomy and stored at 80 C until processing. Laboratory analyses A Hitachi 705 autoanalyzer (Hitachi Ltd, Osaka, Japan) was used to determine the plasma levels of glucose, urea, creatinine, total protein, albumin, cholesterol, triacylglycerols, total and direct bilirubin, alkaline phosphatase, glutamic oxaloacetic transaminase, glutamic pyruvic transaminase, gamma glutamil transpeptidase, sodium, potassium, chloride, and lactate dehydrogenase. Brains were homogenized in an OmniMixer apparatus (Omni, Waterbury, CT, USA), and lipids were extracted using the method of Folch et al. . The extract was separated into polar and neutral lipids as described by Bitman et al.  using Sep-Pak Silica cartridges (Waters Chromat, Dic Millipire Co., Billerica, MA, USA).
The dry polar lipid extract was re-dissolved in chloroform and analyzed by thin-layer chromatography using as the mobile phase a chloroform–methanol– water–acetic acid (60:30:6:1) mixture for 60 min. The plates were developed using a 10% molybdic acid solution, with oven heating at 120 C for 30 min. The spots were semiquantiﬁed by laser densitometry using a Shimadzu DR-13 CS9000 plate reader and integrator (Shimadzu, Kyoto, Japan). Assays were made of phosphatidylcholine, phosphatidylserine, phosphatidylethanolamine, phosphatidylinositol, and palmitoyl cerebroside.
FA brain analysis The brain FAs were extracted and methylated according to the technique of Lapage and Roy , a method of extraction, methylation, and esteriﬁcation in one step. The methyl esters extracted were assayed using a Carlo Erba GC 6000 Vega Series gas chromatograph (Carlo Erba Instruments, Milan, Italy) equipped with a 50-m capillary column of polyamide-coated fused silica of 0.22 mm internal diameter (50QC2/BP70, SGE International Pty Ltd, Victoria, Australia) and using heptadecanoic acid (C-17) as the internal standard. The data were autoprocessed in an HP-3394A integrator (Hewlett Packard, Avondale, PA, USA). The chromatographic conditions used were an injector temperature of 250 C, a detector temperature of 260 C, and an oven temperature programmed with an initial isothermal of 195 C for 5 min, followed by a gradient of 3 C/min to 250 C that remained at this temperature for 10 min. Assays were made of myristic, palmitic, palmitoleic, stearic, oleic, vaccenic, linoleic, linolenic, arachidic, icosatrienoic, dihomo-g-linolenic, arachidonic, behenic, docosatetraenoic, docosahexaenoic, nervonic, and cerotic acids.
Statistical analyses The means and standard deviations of all parameters were determined. The non-parametric two-way analysis of variance using the Friedman rank sums was used to compare within-group ﬁnal and initial values of weight and blood biochemistry parameters. For intergroup comparisons of brain fat among the LCT, MCT/LCT, and control groups, the non-parametric Kruskal–Wallis was used and differences among the groups were calculated with the Dunn approximation [17, 18]. P < 0.05 was set as the criterion for statistically signiﬁcant differences.
Results The mean weight loss of the groups receiving TPN was not signiﬁcant compared with that of the control group (Table 2).
-Botejara et al. / Nutrition 29 (2013) 313–317 E. Macia Table 2 Variations in animal weights during the 7-d study
LCT MCT/LCT Control
1.991 22 1.585 152 1.765 757
1.806 130 1.521 210 1.815 773
NS NS NS
LCT, long-chain triacylglycerol; MCT, medium-chain triacylglycerol Values are expressed as arithmetic mean SD
General plasma biochemistry The glycemia of the animals in all experimental groups showed a high variability (Table 3), with a tendency toward hyperglycemia at the end, especially in group MCT/LCT. There was a decrease in total proteins and albumin in all the experimental groups, but the intergroup differences were not statistically signiﬁcant. There were no changes in plasma levels of urea, creatinine, sodium, potassium, chloride, lactate dehydrogenase, transaminases, alkaline phosphatase, g-glutamyl-transferase, or total and direct bilirubin. Plasma lipids In the two TPN groups, the moderate increases in cholesterol and triacylglycerols were not statistically signiﬁcant compared with the control group (Table 3). Brain lipids The greater percentage of fat in relation to brain tissue weight in group MCT/LCT was not statistically signiﬁcant compared with the LCT and control groups (Table 4). The lipid fraction assays showed a signiﬁcant decrease in phosphatidylcholine and phosphatidylinositol in the TPN experimental groups (groups LCT and MCT/LCT). The decrease was greater in group LCT than in group MCT/LCT. There were also increases in palmitoyl cerebroside in group LCT and in phosphatidylethanolamine in group MCT/LCT (Table 5). As presented in Table 6, the proﬁle of the overall percentage distribution of the total brain FAs was very similar for the three groups. There were no statistically signiﬁcant differences in the u-3/u-6, C-18:2 u-6/C-20:4 u-6, or C-18:3 u-3/C-22:6 u-3 ratios. Discussion Oral nutrition has been abundantly studied, including longterm studies, in humans and other animals with respect to the inﬂuence of the types of dietary fat on lipid metabolism and brain and tissue composition, especially in the early stages of life. TPN, however, has been the subject of very few experimental or clinical studies examining how it affects lipids in the brain. There is a clear need to determine the inﬂuence of the types of fat used,
the route of administration, the duration, and the underlying pathology, etc., especially in medium-term TPN and at any age. For this reason, we selected for the present study an experiment of intermediate duration and a model of young animals. The nutritional results for the experimental groups showed a slight ﬁnal weight loss in the TPN groups (probably owing to fecal mass loss). The calorie/protein ratio and the composition of the mixture seemed adequate. Blood glucose during the course of the experimental period was not well controlled, especially in the TPN groups. The increases in total cholesterol and plasma triacylglycerols were modest and in line with those commonly described after the intravenous administration of these emulsions. The hypertriglyceridemia accompanying TPN appears to be caused by plasma increases secondary to the administration of the lipid particles and by the transfer of triacylglycerols from these particles to endogenous lipoproteins in exchange for cholesteryl ester. The increase in exogenous phospholipids could induce the mobilization of tissue cholesterol deposits to form micelles , leading to higher blood cholesterol levels. Nevertheless, although group LCT was administered larger amounts of lecithin than group MCT/LCT, there were no signiﬁcant differences in plasma levels between the two groups. Therefore, other factors seem to be involved in the phenomenon of hypercholesterolemia associated with TPN . There seemed to be no extra accumulation of lipids in the brain, because we found no intergroup differences in the total amounts of fat retrieved per gram of brain tissue. Studies of different parenteral routes of administration of LCT or MCT have found the plasma and liver FA proﬁles to reﬂect the proportions in the mixture infused . It has been shown that an increase or deﬁcit of the oral intake of linoleic or linolenic acid or a change in their proportion (u-6/u-3) may give rise to functional changes in the early stages of brain development [22,23]. One may assume that these variations in FA composition could induce the overproduction of metabolites and cytokines (prostaglandins, thromboxanes, and leukotrienes) or lipid peroxidation in brain tissue . The quantities of essential FAs administered and their relative proportions can cause changes in the synthesis of brain lipid fractions, in particular of membrane phospholipids [3,25]. We found no differences among any of the groups in the percentage distribution of these FAs in the total brain, and there were no differences in the most important FA ratios (u-6/u-3, C18:3 u-3/C22:6 u-3, C18:2 u-6/C20:4 u-6, etc.), although the total LCT amounts, the u-6/u-3 ratios, and the LCT proﬁles administered to the two TPN groups were clearly different. Indeed, previous work in our department  with an intravenous overload of LCT (Intralipid 20%) also found no signiﬁcant variations in the brain tissue LCT proﬁle, as has been conﬁrmed by other investigators . The greater relative availability of u-6 in group LCT may have induced a higher rate of peroxidation and/or production of cytokines and prostaglandins.
Table 3 Results for glucose and plasma lipids Control (n ¼ 7) Glucose (mg/dL)y Total cholesterol (mg/dL) Triacylglycerols (mg/dL)
LCT (n ¼ 7)
MCT/LCT (n ¼ 7)
137 18 70 25 74 28
146 32 109 49 159 142
162 14 56 22 74 13
200 60 119 35 154 108
133 26* 66 2* 62 15*
221 41 133 66 193 74
LCT, long-chain triacylglycerol; MCT, medium-chain triacylglycerol Values are expressed as arithmetic mean SD * P < 0.05, initial versus ﬁnal values. y P < 0.05, comparison between groups was signiﬁcant for glucose in MCT/LCT versus control group.
-Botejara et al. / Nutrition 29 (2013) 313–317 E. Macia
316 Table 4 Results for brain fat percentage
Table 6 Percentage distribution of brain fatty acids
Control (n ¼ 7)
LCT (n ¼ 7)
MCT/LCT (n ¼ 7)
LCT, long-chain triacylglycerol; MCT, medium-chain triacylglycerol; %wt fat/ tissue, percentage of lipid weight in relation to total tissue weight Values are expressed as arithmetic mean SD
Nonetheless, if this had been so, we should have observed increased FAs deriving from C18:3 u-6, which was not the case. Another possibility is that, with the amounts of LCT administered to the two TPN groups, the adult brains of our animals became saturated and, hence, ﬁnished by containing similar amounts of each LCT, but that these led to changes in the esteriﬁcation and synthesis of brain phospholipids and cholesterol, with the consequent functional alterations in cell permeability and neuronal function. Indeed, what we did observe were major variations between groups in the distribution of brain phospholipids, with important decreases in phosphatidylcholine in group LCT. This seems paradoxical because it is this group (the Intralipid group) that received the larger amount of lecithin (almost double that of group MCT/LCT). Other work by our group has shown large changes in brain lipids after an overload with Intralipid by different routes of administration, with inﬂuences from the route and the rate of infusion . Based on our present data, we must presume that the different brain phospholipid proﬁles of groups LCT and MCT/LCT is due mainly to the different LCT composition administered. In premature infants fed with Intralipid, other investigators found changes in the proﬁle of phospholipids and the composition of the LCTs that saturate them . Although we did not analyze the particular FA composition in each phospholipid fraction, the similarity of the results of the FAs in the total brain suggests that differences in lipid composition administered can be compensated in the brain or before reaching the brain, possibly the liver, at least at this age, in these animals, and in these doses, administrated in this short period. This may also occur in the adult brain, although it will be cushioned by a lower metabolic turnover of the LCTs. Although there were no structural repercussions on the brain, we found functional alterations that were more or less readily clinically detectable in our patients nourished with TPN. Indeed, changes in the lipid composition of the adult brain, mediated by the fat ingested or administered, may be related to certain functional or degenerative disorders [11,27]. Phospholipids are a fundamental part of cell membranes and
Table 5 Percentage distribution of brain lipid fractions Control (n ¼ 7) LCT (n ¼ 7) Phosphatidylinositol Phosphatidylserine Phosphatidylcholine Phosphatidylethanolamine Palmitoyl cerebroside
19.19 ND 21.50 52.65 6.64
3.66 6.11 11.11 2.05
8.72 9.28 3.55 51.28 27.15
MCT/LCT (n ¼ 7)
2.05* 13.05 3.89 1.39* 5.58 3.15 0.78* 13.64 4.35 4.18 61.10 4.72* 3.86* 6.61 3.67*
LCT, long-chain triacylglycerol; MCT, medium-chain triacylglycerol; ND, not detected Values are expressed as arithmetic mean SD * Signiﬁcant differences were found for phosphatidylinositol (LCT versus control, P < 0.01), phosphatidylserine (LCT versus control, P < 0.01; MCT/LCT versus control, P < 0.05), phosphatidylcholine (LCT versus control, P < 0.01), phosphatidylethanolamine (LCT versus MCT/LCT and control, P < 0.05), and palmitoyl cerebroside (LCT versus MCT/LCT, P < 0.01; LCT versus control, P < 0.05).
Control (n ¼ 7) C-14 C-16 C-16:1 u-7 C-18 C-18:1 u-9 C-18:1 u-6 C-18:1 u-7 C-18:2 u-6 C-18:3 u-3 C-20:1 C-20:3 C-20:3 u-6 C-20:4 u-6 C-22:0 C-22:4 u-6 C-22:6 u-6 C-24:1 C-26 S/P u-3/u-6 C-18:2 u-6/C-20:4 u-6 C-18:3 u-3/C-22:6 u-3
1.88 22.42 0.38 23.75 19.18 2.36 4.64 2.36 0.45 1.22 9.67 0.50 0.58 0.97 4.97 4.65 1.71 0.67 2.15 0.39 3.53 0.14
0.86 3.05 0.04 1.25 1.43 0.22 0.76 0.22 0.14 0.22 1.80 0.14 0.17 0.57 1.75 1.87 0.72 0.47 0.63 0.12 1.62 0.10
LCT (n ¼ 7) 1.08 20.14 0.57 23.61 20.42 2.51 4.81 2.51 0.6 1.64 9.84 0.56 0.56 1.07 5.45 3.79 1.96 0.86 2.01 0.34 4.50 0.17
0.33 2.45 0.37 0.42 1.80 0.19 0.76 0.19 0.12 0.22 0.36 0.08 0.08 0.58 0.67 1.40 0.86 0.22 0.26 0.13 0.67 0.04
MCT/LCT (n ¼ 7) 1.6 21.36 0.41 24.29 19.45 2.18 4.69 2.18 0.65 1.30 10.4 0.66 0.57 1.13 5.35 3.29 2.01 0.64 1.99 0.28 3.85 0.31
0.21 2.57 0.08 0.63 1.18 0.51 0.82 0.51 0.17 0.19 0.79 0.13 0.05 0.31 0.95 0.60 0.91 0.20 0.52 0.06 0.96 0.30
LCT, long-chain triacylglycerol; MCT, medium-chain triacylglycerol; S/P, saturated/polyunsaturated Values are expressed as arithmetic mean SD
functionally essential for permeability and diffusion across the membrane. In the central nervous system, this is even more crucial, so we analyzed the possible changes in the phospholipid proportions. Changes in brain lipid composition in the fetus and neonate after TPN administration have been known for years , although the precise mechanism is unknown. We do not know if the changes in composition and function are due to phospholipid variations or their saturation with FAs, which in turn would be related to their presence in brain tissue. We studied total FAs and, as other investigators, found no differences in their composition. However, there could be differences in the phospholipids and/or saturation with different types of FAs. Thus, the function of these FAs in the brain would be through the phospholipids and therefore neuronal membrane phenomena. Our results showed a greater variation of phospholipids than of total brain FA content, probably due to a decrease or redistribution of FAs, which is greater in older individuals. Although we do not have much data on FAs and brain phospholipid compositions in different population groups, there are studies showing an effect of dietary fat on superior function. Thus, docosahexaenoic acid brain levels depend on its presence in the diet . These levels could protect against inﬂammatory phenomena originated by a diet rich in saturated FAs, mainly at young age . Moreover, the protective role of u-3 FAs is related to the inhibition of inﬂammatory phenomena, especially when the u-3/arachidonic acid ratio is greater; therefore, human subjects receiving those diets could have better cognitive level, regardless of age and education . Monounsaturated FAs have also shown cortical activity improvement in adult mice . Therefore, further studies of each phospholipid and its FA saturation are necessary. Perhaps there is a need to evaluate the impact that such changes could have on cognition and mood in adults fed for shorter or longer times by TPN. It would also be convenient to study the speciﬁc esteriﬁcation of each phospholipid by the different Fas, depending on the type of fat used, to conﬁrm the change induced by the different emulsions used on the brain’s polar lipids.
-Botejara et al. / Nutrition 29 (2013) 313–317 E. Macia
Conclusions The two types of fat emulsion led to changes in the brain phospholipid proﬁle, but those in group MCT/LCT were smaller and closer to those of the control group. A lighter LCT load and a lower u-6/u-3 ratio are preferable to using LCTs as the sole lipid source in prolonged nutrition for all ages. References re braux synthe se in situ et apports  Bourre JM. Origine des acides gras ce nutritionnels. Rev Fr Corps Gras 1984;6:225–31.  Innis SM. Essential fatty acids in growth and development. Prog Lipid Res 1991;30:39–103.  Lutz M, Durand G. Inﬂuence of dietary lipids on fatty acid composition of nervous membranes (myelin and synaptosomes) in rats. Nutr Res 1994;14:1365–73.  Ohkubo T, Tanaka Y. Administration of DHA-PS to aged mice was suitable for increasing hippocampal PS and DHA ratio. J Oleo Sci 2010;59:247–53.  Guesnet P, Couet C, Alessandri JM, Antoine JM. Effects of linoleic acid content (18:2u-6) and 18:2u-6/18:3u-3 ratio of breast milk lipids on docosahexaenoic acid content (22:6u-3) of nervous structures (brain, retina) in suckling pups. Ann Pediatr 1995;42:289–94.  Lim SY, Suzuki H. Effect of dietary docosahexaenoic acid and phosphatidylcholine on maze behavior and fatty acid composition of plasma and brain lipids in mice. Int J Vitam Nutr Res 2000;70:251–9.  Hiratsuka S, Koizumi K, Ooba T, Yokogoshi H. Effects of dietary docosahexaenoic acid connecting phospholipids on the learning ability and fatty acid composition of the brain. J Nutr Sci Vitaminol (Tokyo) 2009;55:374–80.  Cao D, Kevala K, Kim J, Moon HS, Jun SB, Lovinger D, Kim HY. Docosahexaenoic acid promotes hippocampal neuronal development and synaptic function. J Neurochem 2009;111:510–21.  Farquharson J, Cockburn F, Ainsliepatrick W, Jamieson EC, Logan RW. Infant cerebral cortex phospholipid fatty-acid composition and diet. Lancet 1992;340:810–3.  Igarashi M, Ma K, Gao F, Kim HW, Greenstein D, Rapoport SI, Rao JS. Brain lipid concentrations in bipolar disorder. J Psychiatr Res 2010;44:177–82.  Huss M, Völp A, Stauss-Grabo M. Supplementation of polyunsaturated fatty acids, magnesium and zinc in children seeking medical advice for attention-deﬁcit/hyperactivity problemsdan observational cohort study. Lipids Health Dis 2010;24:105–12.  Martınez M, Ballabriga A. Effects of parenteral nutrition with high doses of linoleate on the developing human liver and brain. Lipids 1987;22:133–8.  Maci a-Botejara E, Mor an-Penco JM, Salas Martınez J, Mahedero Ruiz G, n G, Climent Mata V, Molina Fern Cruz Villalo andez M. Effect of IntralipidÒ overdose on brain lipid composition in the adult rabbit and its relation with the route of administration. Nutr Res 1995;6:881–8.
 Folch J, Less M, Sloane G. A simple method for the isolation and puriﬁcation of total lipids from animal tissues. J Biol Chem 1957;226:497–505.  Bitman J, Wood L, Mehta NR, Hamosh P, Hamosh M. Comparison of the phospholipids composition of breast milk from mothers of term and preterm infants during lactation. Am J Clin Nutr 1984;40:1103–19.  Lapage G, Roy CC. Direct transesteriﬁcation of all classes of lipid in a onestep reaction. J Lipid Res 1986;27:114–20.  Hollander M, Wolfe D. Nonparametric statistical methods. In: Probability and mathematical methods. 2nd ed. New York: Wiley Series in Probability and Statistics; 1999. p. 189–362.  Dunn O. Multiple comparisons using rank sums. Technometrics 1964;6: 241–54.  Ekelund M, Roth B, Trelde H, Ekstrom U, Nilsson-Ehle P. Effects of total parenteral nutrition on lipid metabolism in rats. JPEN 1994;18:503–9.  Weinberg RB, Singh KK. Short-term parenteral nutrition with glucose and IntralipidÒ: effects on serum lipids and lipoproteins. Am J Clin Nutr 1989;49:794–8.  Bourre JM, Dumont O, Pascal G, Durand G. Dietary alpha-linolenic acid at 1.3 g/kg maintains maximal docosahexaenoic acid concentration in brain, heart and liver of adult rats. J Nutr 1993;123:1313–9.  Bowena RA, Clandinin MT. High dietary 18:3n-3 increases the 18:3n-3 but not the 22:6n-3 content in the whole body, brain, skin, epididymal fat pads and muscle of suckling rats pups. Lipids 2000;35:389–94.  Ward GR, Huang YS, Xing HC, Bobik E, Wauben I, Auestad N, et al. Effects of gamma-linoleic acid and docosahexaenoic acid in formulae on brain fatty acid composition in artiﬁcially reared rats. Lipids 1999;34:1057–63.  Hagi A, Nakayama M, Shinzaki W, Ajı S, Ohyanagi H. Effects of the omega6:omega-3 ratio on fat emulsions on the fatty acid composition in cell membranes and the antiinﬂammatory action. JPEN 2010;34:263–70.  Song JH, Fujimoto K, Miyazawa T. Polyunsaturated (n3) fatty acids susceptible to peroxidation are increased in plasma and tissue lipids of rats fed docosahexaenoic-acid containing oils. J Nutr 2000;130:3028–33. nchez M, Hyde MJ, Laws J, Clarke L, Herrera E. Inﬂuence of  Amusquivar E, Sa fatty acid proﬁle of total parenteral nutrition emulsions on the fatty acid composition of different tissues of piglets. Lipids 2008;43:713–22.  Fernstrom JD. Can nutrient supplements modify brain function? Am J Clin Nutr 2000;71:1669S–73S.  Tang X, Li ZJ, Xu J, Xue Y, Li JZ, Wang JF, et al. Short term effects of different omega-3 fatty acid formulation on lipid metabolism in mice fed high or low fat diet. Lipids Health Dis 2012;11:70. Available at: http://www. lipidworld.com/content/11/1/70. Accessed June 2012.  Boitard C, Etchamendy N, Sauvant J, Aubert A, Tronel S, Marighetto A, et al. Juvenile, but not adult exposure to high-fat diet impairs relational memory and hippocampal neurogenesis in mice. Hippocampus 2012;22: 2095–100.  Chiu C-C, Frangou S, Chang C-J, Chiu W-C, Liu H-C, Sun I-W, et al. Associations between n-3 PUFA concentrations and cognitive function after recovery from late-life depression. Am J Clin Nutr 2012;95:420–7.  Sartorius T, Ketterer C, Kullmann S, Balzer M, Rotermund C, Binder S, et al. Monounsaturated fatty acids prevent the aversive effects of obesity on locomotion, brain activity, and sleep behavior. Diabetes 2012;61:1669–79.