[Normal childbirth: physiologic labor support and medical procedures. Guidelines of the French National Authority for Health (HAS) with the collaboration of the French College of Gynaecologists and Obstetricians (CNGOF) and the French College of Midwives (CNSF) -- Text of the Guidelines (short text)]. / Accouchement normal : accompagnement de la physiologie et interventions médicales. Recommandations de la Haute Autorité de Santé (HAS) avec la collaboration du Collège National des Gynécologues Obstétriciens Français (CNGOF) et du Collège National des Sages-Femmes de France (CNSF) ­ Texte des recommandations (texte court).

OBJECTIVE: The objective of these guidelines is to define for women at low obstetric risk modalities that respect the physiology of delivery and guarantee the quality and safety of maternal and newborn care. METHODS: These guidelines were made by a consensus of experts based on an analysis of the scientific literature and the French and international recommendations available on the subject. RESULTS: It is recommended to conduct a complete initial examination of the woman in labor at admission (consensus agreement). The labor will be monitored using a partogram that is a useful traceability tool (consensus agreement). A transvaginal examination may be offered every two to four hours during the first stage of labor and every hour during the second stage of labor or before if the patient requests it, or in case of a warning sign. It is recommended that if anesthesia is required, epidural or spinal anesthesia should be used to prevent bronchial inhalation (grade A). The consumption of clear fluids is permitted throughout labor in patients with a low risk of general anesthesia (grade B). It is recommended to carry out a "low dose" epidural analgesia that respects the experience of delivery (grade A). It is recommended to maintain the epidural analgesia through a woman's self-administration pump (grade A). It is recommended to give the woman the choice of continuous (by cardiotocography) or discontinuous (by cardiotocography or intermittent auscultation) monitoring if the conditions of maternity organization and the permanent availability of staff allow it and, after having informed the woman of the benefits and risks of each technique (consensus agreement). In the active phase of the first stage of labor, the dilation rate is considered abnormal if it is less than 1cm/4h between 5 and 7cm or less than 1cm/2h above 7cm (level of Evidence 2). It is then recommended to propose an amniotomy if the membranes are intact or an oxytocin administration if the membranes are already ruptured, and the uterine contractions considered insufficient (consensus agreement). It is recommended not to start expulsive efforts as soon as complete dilation is identified, but to let the presentation of the fetus drop (grade A). It is recommended to inform the gynecologist-obstetrician in case of nonprogression of the fetus after two hours of complete dilation with sufficient uterine dynamics (consensus agreement). It is recommended not to use abdominal expression (grade B). It is recommended to carry out preventive administration of oxytocin at 5 or 10 IU to prevent PPH after vaginal delivery (grade A). In the case of placental retention, it is recommended to perform a manual removal of the placenta (grade A). In the absence of bleeding, it should be performed 30minutes but not more than 60minutes after delivery (consensus agreement). It is recommended to assess at birth the breathing or screaming, and tone of the newborn to quickly determine if resuscitation is required (consensus agreement). If the parameters are satisfactory (breathing present, screaming frankly, and normal tonicity), it is recommended to propose to the mother that she immediately place the newborn skin-to-skin with her mother if she wishes, with a monitoring protocol (grade B). Delayed cord clamping is recommended beyond the first 30seconds in neonates, not requiring resuscitation (grade C). It is recommended that the first oral dose (2mg) of vitamin K (consensus agreement) be given systematically within two hours of birth. CONCLUSION: These guidelines allow women at low obstetric risk to benefit from a better quality of care and optimal safety conditions while respecting the physiology of delivery.

Topological distribution of choline phospholipid fatty acids in trout intestinal brush-border membrane.

The transbilayer distribution of choline phospholipids in trout intestinal brush-border membrane has been investigated using phospholipase C (from Clostridium welchii). In the middle intestine, 84% of phosphatidylcholine (PC) and 60% of sphingomyelin (SP) are located in the outer membrane leaflet. In the posterior intestine, 89% of PC and 52% of SP are located in the outer membrane leaflet. The externally located PC molecular species are (n - 3) fatty acid-rich in both parts of the intestine. While the sphingomyelin molecular species containing 24:1(n - 9) are exclusively located in the outer leaflet in the middle intestine, those containing 14:0 are more abundant in the same leaflet but in the posterior intestine. This strongly asymmetric distribution of both choline phospholipids may have numerous consequences on the brush-border membrane characteristics.

Isolation and characterization of brush-border membrane from trout intestine. Regional differences.

The isolation of brush-border membranes from trout enterocytes is described for both middle and posterior intestine. Both procedures are based on differential centrifugations combined with calcium precipitation. Classical marker enzymes are quantified and indicate a valuable purification of the membranes (13-18-fold). No difference appears when comparing the relative amounts of phospholipids, cholesterol and proteins in microvillus membranes isolated from either middle or posterior intestine. In contrast, the membranes isolated from middle intestine are more unsaturated than those from the posterior one, and their sphingomyelin/phosphatidylcholine ratio is lower. These differences are reflected by fluorescence anisotropy studies with diphenylhexatriene as lipid fluorophore which indicate a higher fluidity of the microvillus membranes from the middle intestine as compared with those from the posterior intestine. These results point out the importance of the fatty acyl chains and that of the relative amounts of phosphatidylcholine and sphingomyelin in controlling the fluidity of biological membranes in relation with their transport properties.

Topological distribution of aminophospholipid fatty acids in trout intestinal brush-border membrane.

The transbilayer distribution of aminophospholipids in trout intestinal brush-border membrane has been investigated using trinitrobenzene sulfonic acid (TNBS). In the middle intestine, phosphatidylethanolamine (PE) is symmetrically distributed between the two leaflets while 68% of the phosphatidylserine (PS) are located in the inner membrane leaflet. In the posterior intestine, 64% of the PE and 69% of the PS are located in the inner membrane leaflet. When asymmetrically distributed, the inner species of PE and PS have a higher content of 22:6(n-3) than the outer ones. This asymmetric distribution of docosahexaenoic acid in trout intestinal brush-border membrane might be related to the rod-like shape of the microvillus membrane and to its metabolism to hydroxylated derivatives.

Effect of exogenous gangliosides on the lipid composition of chick neurons in culture.

When exogenous gangliosides are added to the growth medium of neuronal cell cultures they are inserted into their plasma membranes and are afterwards metabolized in the cytoplasmic interior. The action of exogenous gangliosides brings important morphological and biochemical changes to neurons in culture. The present report shows that the treatment with exogenous gangliosides of a primary culture of chick neurons modified the distribution of fatty acids in phosphatidylinositol (PI), mainly that of arachidonic acid and the fatty acids of the (n - 3) series without affecting the other phospholipids. The composition of neutral lipids did not change but their content was increased up to 2-3-fold depending upon the concentration of gangliosides. The change of the growth medium from one containing fetal calf serum to a chemically defined one reduced dramatically the content of free fatty acids while the addition of gangliosides raised this content to normal levels. The increase in the amount of diacylglycerol (DG) confirmed the finding that gangliosides stimulate phosphoinositide degradation. Finally the fatty acid composition of DG suggests indirectly that this compound might be produced also by degradation of phosphatidylcholine and not only of PI.

Incorporation of exogenous lipids modulates insulin signaling in the hepatoma cell line, HepG2.

The lipid content of cultured cells can be experimentally modified by supplementing the culture medium with specific lipids or by the use of phospholipases. In the case of the insulin receptor, these methods have contributed to a better understanding of lipid disorder-related diseases. Previously, our laboratory demonstrated that experimental modification of the cellular lipid composition of an insulin-sensitive rat hepatoma cell line (ZHC) resulted in an alteration in insulin receptor binding and biological action (Bruneau et al., Biochim. Biophys. Acta 928 (1987) 287-296/297-304). In this paper, we have examined the effects of lipid modification in another hepatoma cell line, HepG2. Exogenous linoleic acid (LA, n-6), eicosapentaenoic acid (EPA, n-3) or hemisuccinate of cholesterol (CHS) was added to HepG2 cells, to create a cellular model in which membrane composition was modified. In this model, we have shown that: (1) lipids were incorporated in treated HepG2 cells, but redistributed differently when compared to treated ZHC cells; (2) that insulin signaling events, such as insulin receptor autophosphorylation and the phosphorylation of the major insulin receptor substrate (IRS-1) were altered in response to the addition of membrane lipids or cholesterol derived components; and (3) different lipids affected insulin receptor signaling differently. We have also shown that the loss of insulin receptor autophosphorylation in CHS-treated cells can be correlated with a decreased sensitivity to insulin. Overall, the results suggest that the lipid environment of the insulin receptor may play an important role in insulin signal transduction.

Effect of 7 beta-hydroxycholesterol on astrocyte primary cultures and derived spontaneously transformed cell lines: cytotoxicity and metabolism.

The lethal effect of 7 beta-hydroxycholesterol (7 beta-OHC) on neonatal rat astrocyte primary cultures and spontaneously transformed cell lines derived from them was investigated. Confluent astrocyte primary cultures were not affected by 30 microM 7 beta-OHC over a period of 72 h. In contrast, spontaneously transformed cells were killed by 20 microM 7 beta-OHC within the first 48 h. Further studies indicated that the cell lines metabolized 7 beta-OHC to a product the polarity of which was less than that of 7 beta-OHC. The metabolite was identified as 7 beta-OHC esterified on C-3 by naturally occurring fatty acids. Incubation of the cell lines with 0.5 microM metabolite markedly affected the cells within 24 h. These observations suggest that the 7 beta-OHC metabolite is implicated in the mechanism of action of 7 beta-OHC cytotoxicity on spontaneously transformed cells.

In vitro synthesis and transbilayer movement of phosphatidylethanolamine molecules labelled with different fatty acids in chick brain microsomes.

The transbilayer fatty acid distribution of diacylglycerophosphoethanolamine and the translocation of newly synthesized phosphatidylethanolamine molecules labelled with different fatty acids has been investigated in chick brain microsomes using trinitrobenzensulfonic acid. The determination of the fatty acid composition of diacylglycerophosphoethanolamine in both the outer and the inner leaflet of the microsomal vesicles revealed a similar distribution indicating that both leaflets share the same molecular species. The in vitro incorporation of radioactive fatty acids (16:0, 18:1 and 20:4(n-6] into ethanolamine phospholipids, known to be catalyzed by the lyosphosphatidylethanolamine acyl transferase, showed that the radioactive diacylglycerophosphoethanolamine molecules appeared first in the outer leaflet and were thereafter transferred to the inner leaflet. The apparent rate of translocation of the newly synthesized ethanolamine phospholipid molecules was the highest for those labelled with 16:0 and the lowest for those labelled with 20:4(n-6). The results indicate that the active site of the acyl-CoA:lysophosphatidylethanolamine acyltransferases is located on the outer leaflet of the microsomal vesicles and that the different newly synthesized molecular species of diacylglycerophosphoethanolamine may be translocated from the outer to the inner leaflet at different rates.

Influence of lipid environment on insulin binding in cultured hepatoma cells.

The influence of alterations of plasma membrane physico-chemical properties on insulin binding have been characterized in an insulin-sensitive rat hepatoma cell line adapted to grow for several generations in culture medium enriched with linoleic acid (18:2) or with 25-hydroxycholesterol. The cells took up 18:2 and 25-hydroxycholesterol added to the culture medium, without exhibiting any sign of intolerance or intoxication. These compounds respectively increased and decreased membrane fluidity at 37 degrees C. The cells demonstrated extensive changes in insulin binding parameters in response to experimental modifications of their membrane lipid composition. When determined at 4 degrees C, insulin receptors were present in the control cells at 136,000 sites/cell but this fell to 111,000 (P less than 0.05) in cells enriched in 18:2, and rose to 176,000 (P less than 0.001) in hydroxysterol-grown cells. According to a two-site model, the main effect of 18:2 was a significant increase of the number of high-affinity sites with a concomitant decrease of low-affinity sites. The hydroxysterol had the opposite effects on these parameters. The high-affinity insulin binding capacity of the hepatoma cells was affected by lipid supplementation in a similar way, whether it was determined at 4 degrees C or at 37 degrees C. Assuming a negative cooperativity model, 18:2 enhanced the degree of negative cooperativity among the sites, while 25-hydroxycholesterol reduced it. The time-course of insulin-induced receptor down-regulation was accelerated in the cells enriched in polyunsaturated fatty acids, but reduced in cells exposed to 25-hydroxycholesterol. These insulin-binding alterations cannot be directly related to modifications of cellular growth rate, receptor internalization or membrane fluidity per se, and are discussed as being more likely due to membrane lipid composition than to overall cell metabolism modifications.

Lipid composition and structure of commercial parenteral emulsions.

In order to study the influence of the phospholipid/triacylglycerol (PL/TG) ratio of parenteral emulsions on the distribution and the physico-chemical properties of their fat particles, commercial 10, 20 or 30% fat formulas were fractionated by centrifugation into an upper lipid cake (resuspended in aqueous glycerol) and a subnatant or mesophase, from which a PL-rich subfraction (d = 1.010-1.030 g/l) was purified by density gradient ultracentrifugation. Chemical and 31P-NMR analyses of these fractions indicated that at least two types of fat particles coexist in parenteral emulsions: (i) TG-rich particles (mean diameter: 330, 400, 470 nm in the 10, 20, 30% emulsion) which contain practically all the TG and esterified phytosterols of native emulsions, but only a fraction of their PL, unesterified cholesterol and phytosterols, and other minor lipids; (ii) PL-bilayer particles or liposomes (mean diameter: 80-100 nm) which are constituted with the remaining PL and relatively very small amounts of TG and other lipids. The higher the oil content of the emulsion, the lower the amount of these PL-rich particles, which represent the major particle population of the mesophase. Indeed, minute amounts of TG-rich particles (probably the smallest ones) are also present in the mesophase, even in the PL-rich subfraction which contains the bulk of liposomal PL. Since the PL-rich particles of the infused emulsion generate lipoprotein X-like particles, only the large TG-rich particles can be considered as true chylomicron counterparts.

Long-chain n-3 fatty acids specifically affect rat coagulation factors dependent on vitamin K: relation to peroxidative stress.

Fatty acids of marine origin have been shown to affect blood coagulation in the rat. In an attempt to gain insight into the mechanisms of this phenomenon, we studied the effects of dietary linseed and fish oils on the liver antioxidant status and plasma coagulation parameters in rats on a time-course basis. Dietary enrichment in eicosapentaenoic and docosahexaenoic acids resulted in strong hypocoagulation after only 1 week and a concomitant increase in liver lipid peroxidation and tocopherolquinone content. Enrichment in linolenic acid induced similar increases in lipid peroxidation and tocopherol catabolism but negligible alteration of coagulation. A significant correlation between plasma factor II coagulant activity and liver tocopherolquinone was found in fish oil- but not in linseed oil-fed rats. Although ingestion of tocopherolquinone led to high levels of this compound in the liver, it had only marginal effects on coagulation factors. Thus, it seems unlikely that this vitamin E metabolite could be involved in the lowering of vitamin K-dependent clotting factors through inhibition of gamma-glutamylcarboxylase. Rather, our results indicate that the effects of the n-3 fatty acids of fish oil on vitamin K-dependent coagulation factors are specific and independent of liver tocopherolquinone levels.

Effect of 7 beta-hydroxycholesterol on astrocyte primary cultures and derived spontaneously transformed cell lines. Cytotoxicity and cholesterogenesis.

The correlation between the lethal effect of 7 beta-hydroxycholesterol (7 beta-OH-CH) on spontaneously transformed cell lines derived from rat astrocyte primary cultures (normal cells) and de novo cholesterogenesis was investigated. Both 7 beta-OH-CH and 7-keto-CH were not cytotoxic on normal cells but 7 beta-OH-CH affected markedly the viability of the transformed cells. The use of [14C]acetate or [14C]mevalonate indicated that 7-keto-CH inhibits de novo cholesterogenesis upstream of 3-hydroxy-3-methylglutaryl CoA reductase (HMGR) in both cell types whereas 7 beta-OH-CH also inhibits downstream of HMGR. The accumulation of two radiolabelled products X1 and X2 between mevalonate and CH was found in unsaponifiable neutral lipids extracted from 7 beta-OH-CH treated transformed cells. HPLC and GC-MS revealed that X1 and X2 are not lanosterol and 24,25-epoxylanosterol, respectively. Incubation of the transformed cells with X1 and X2 did not affect their viability. Our data demonstrate that, under our experimental conditions, 7 beta-OH-CH cytotoxicity is not linked to the inhibition of de novo cholesterogenesis in cultured glial transformed cells.

Nuclear sphingomyelin protects RNA from RNase action.

Chromatin phospholipidic fraction, as previously demonstrated, shows the same localization as RNA inside the nuclei. DNase and RNase treatment of nuclei removed almost totally the DNA, 63% of RNA and caused a 50% loss of phospholipids. The aim of the present investigation is to study the fraction of RNase undigested nuclear RNA and its relationship with the phospholipids still present in the nuclei. Isolated hepatocyte nuclei were treated with Triton X-100 and digested with RNase and DNase. The undigested nuclear material contained proteins (98%) and a small amount of RNA (1.7%), DNA (0.4%) and phospholipids (0.18%). The analysis of phospholipids showed the presence of two components only, namely phosphatidylcholine and sphingomyelin. In the same complex, the activity of sphingomyelin synthase, phosphatidylcholine-dependent phospholipase C and neutral sphingomyelinase has been detected. Treatment of isolated RNA with neutral sphingomyelinase modified the RNA in RNase sensitive RNA, thus suggesting that the SM may represent a bridge between two RNA strands possibly regulating transcription.

The P2Y1 receptor is an ADP receptor antagonized by ATP and expressed in platelets and megakaryoblastic cells.

The human P2Y1 purinoceptor has been expressed in Jurkat cells and the effects of HPLC purified nucleotides on calcium movements were measured. The most potent agonist was 2-methylthio-ADP followed by ADP. ATP, Sp-ATPalphaS and beta,gamma-methylene-ATP were competitive antagonists. Suramin and PPADS inhibited the effects of ADP. This pharmacological profile is the same as that of the so-called P2T purinoceptor responsible for platelet aggregation, which has not yet been identified. Using PCR we found the P2Y1 receptor to be present in blood platelets and megakaryoblastic cell lines. These data suggest that the P2Y1 receptor may be the elusive P2T receptor.

Effect of dibutyryl cyclic AMP and isoproterenol on 7 beta-hydroxycholesterol cytotoxicity and esterification in spontaneous transformed cell lines derived from astrocyte primary cultures.

Incubation of spontaneous transformed cells derived from astrocyte primary cultures with 30 microM 7 beta-hydroxycholesterol (7 beta-OH-CH) which is lethal to the cells or with 150 microM isoproterenol reduces the intracellular level of cAMP (4- and 2-fold respectively). Treatment of the cultures with 0.5 mM dibutyryl (db)-cAMP and 7 beta-OH-CH increases 3-fold the intracellular level of cAMP and both, db-cAMP and isoproterenol, raise the lethal effect of 7 beta-OH-CH and its esterification on C-3-OH by naturally occurring fatty acids (metabolite). Kinetic studies of net steryl-3-esters hydrolysis revealed that db-cAMP and isoproterenol lower that of cholesteryl-3-esters (2-fold) whereas the opposite is found for the metabolite. These data demonstrate that (i) high cAMP intracellular levels modulate differently the net hydrolysis of cholesteryl-3-esters and metabolite, (ii) isoproterenol acts otherwise than cAMP on 7 beta-OH-CH esterification, (iii) the cytotoxicity of 7 beta-OH-CH is linked to its own esterification. The accumulation of metabolite subsequent to db-cAMP or isoproterenol treatment as a result of acyl-CoA:cholesterol acyl transferase activation is discussed.

Antithrombotic effects of (n-3) polyunsaturated fatty acids in rat models of arterial and venous thrombosis.

The antithrombotic effects of dietary lipids were investigated in rat models of arterial and venous thrombosis. In the arterial model, thrombus formation was evaluated by determination of the occlusion time and the deposition of 111In-labeled platelets and 125I-labeled fibrinogen in a collagen-coated glass capillary inserted into an arterio-arterial shunt. Venous thrombosis was evaluated by measurement of the thrombus weight after administration of thromboplastin as a source of tissue factor and establishment of stasis in the vena cava. Diets were supplemented with saturated (SAT group) or (n-3) fatty acids, the latter being added either as MaxEPA oil (MaxEPA group), or as docosahexaenoic (DHA group) or eicosapentaenoic (EPA group) ethyl ester. Only the MaxEPA group displayed a prolonged occlusion time as compared with all other groups. Platelet accumulation, similar in the MaxEPA, EPA and DHA groups (13.3, 16.7 and 17.7 x 10(6) platelets/shunt, respectively), was significantly higher in the SAT group (25.3 x 10(6) platelets/shunt), while accumulation of fibrinogen-fibrin was similar whatever the group. There was a trend towards a lower venous thrombus weight in MaxEPA fed rats relative to those fed other diets. Our data indicate that the MaxEPA diet had antithrombotic effects in arterial and to a lesser extent venous thrombosis models, best attributed to its multiple targeting of platelets and coagulation.

Morphometric and biochemical studies in trigeminal nerve of rat after trichloroethylene or dichloroacetylene oral administration.

Trigeminal nerve impairment is one of the main features of the clinical spectrum observed after trichloroethylene (Tce) exposure. A morphometric study of teased fibres, an analysis of the fatty acid composition of total lipids of the trigeminal trunk, measurements of myelin basic protein (MBP) and of 2'-3'-cyclic nucleotide 3' phosphohydrolase (CNP) activity were used to characterize this lesion. Furthermore, the effects of Tce and of dichloroacetylene (Dca), a breakdown product supposed to be the main cause of the trigeminal neuropathy, were compared under the same experimental conditions. Results showed a significant decrease of the internode length and of the mean fibre diameter in the Dca group whereas these values, although also diminished, were not significantly affected in the Tce group. In the trigeminal nerve, this was associated with a significant (46%) decrease in sigma. (n-6) fatty acid and a 22% decrease in sigma. (n-3) fatty acid contents in the Tce group whereas these decreases were respectively 32% in sigma. (n-6) fatty acid and 27% (significant) in sigma. (n-3) fatty acid contents in the Dca group. In this group, a 64% increase in sigma. Fatty aldehydes was also observed. Finally, Tce or Dca had only a slight effect on the overall profile of fatty acid content in the brain and no significant variation was observed in MBP level and CNP activity. Both compounds can thus induce trigeminal nerve alterations, but with a higher intensity for Dca.

Lipid composition of blood platelets and erythrocytes of southern elephant seal (Mirounga leonina) and antarctic fur seal (Arctocephalus gazella).

Erythrocyte and blood platelet phospholipid compositions were studied in three elephant seals and two fur seals, two species of marine mammals living in the Subantarctic region feeding on preys rich in (n-3) polyunsaturated fatty acids. Results were compared with those reported for related species and humans. In erythrocytes, the phospholipid (PL) and cholesterol (CHOL) contents were lower in pinnipeds than in humans. Phosphatidylcholine (PC) levels were higher in elephant seals than in fur seals, with a reverse trend for phosphatidylethanolamine (PE) and phosphatidylserine (PS). Both species had lower SM/PC ratios and PE plasmalogen concentrations than human. Erythrocytes were richer in (n-3) fatty acids (FA) in pinnipeds than in humans. In platelets, the PL content was lower and the CHOL content higher in elephant seals than in humans or in other phocid seal species studied to date. The SM/PC ratio was much higher than in other seal species or in man. In both species, the proportion of PE plasmalogens was higher in platelets than in erythrocytes. PL were more saturated in elephant seals than in fur seals. These results suggest that the erythrocytes and platelets of wild marine mammals may prove useful models to study the influence of dietary lipids on the structure and hemostatic function of these cells.

Effects of eicosatetraynoic acid on membrane lipids of trout liver and intestine.

After two months feeding either an (n-3) or an (n-6) fatty acid-rich diet, two groups of trout were switched to the (n-6) or the (n-3) fatty acid-rich diet, respectively. Half of each group was treated with 0.03% 5,8,11,14-eicosatetraynoic acid (ETYA) in the diet. Liver and intestinal brush border membrane lipids were analyzed. No effect was observed on their cholesterol content. ETYA induced an accumulation of 18:2(n-6), and it did not affect the 20:4(n-6) content but decreased the 22:5(n-6) content. ETYA induced an increase of 18:3(n-3) content in the brush border membrane and a decrease of the 22:6(n-3) content in the liver. Those results suggest that ETYA blocks mainly the delta 6-desaturase, which should have two different sites in the liver and one in the intestine.

Composition and structure of triacylglycerols in brown adipose tissue of rats fed fish oil.

This study examines the incorporation of highly unsaturated n-3 fatty acids (HUFA) into triacylglycerols (TAG) of brown adipose tissue (BAT), and their effect on the positional distribution of saturated (SFA) and of unsaturated (UFA) 16- or 18-carbon fatty acids. To this end, rats were fed a fish oil diet for up to four weeks. The stereospecific analysis of TAG was based on generation of sn-1,2- and sn-2,3-acylglycerols by Grignard degradation, followed by synthesis of phosphatidic acid and specific hydrolysis with phospholipase A2. From the end of the first week of fish oil feeding, a steady-state in the fatty acid composition of TAG in BAT was reached. HUFA concentration increased 30-fold, mainly at the expense of n-9 UFA and of SFA. The amount of SFA decreased selectively at position 3, where these fatty acids were progressively replaced by n-3 HUFA. By contrast, the amount of UFA decreased at all positions, and their positional distribution was not affected. About 60% of HUFA was incorporated at position 3. Nearly twice as much 22:6n-3 was incorporated into TAG than had been previously observed in white adipose tissue (WAT) [Leray, C., Raclot, T., and Groscolas, R. (1993) Lipids 28, 279-284]. At the steady-state, the distribution of HUFA was characterized by high proportions of 22:6n-3 and 20:5n-3 in position 3. Moreover, in each position of TAG, a steady level was reached rapidly (within 1 wk).(ABSTRACT TRUNCATED AT 250 WORDS)