Advertisement

Regulation of the Biosynthesis of 4,7,10,13,16,19-Docosahexaenoic Acid*

  • Devanand L. Luthria
    Affiliations
    Department of Medical Biochemistry, The Ohio State University, Columbus, Ohio 43210
    Search for articles by this author
  • B. Selma Mohammed
    Affiliations
    Department of Medical Biochemistry, The Ohio State University, Columbus, Ohio 43210
    Search for articles by this author
  • Howard Sprecher
    Correspondence
    To whom correspondence should be addressed: Dept. of Medical Biochemistry, The Ohio State University, 337 Hamilton Hall, 1645 Neil Ave., Columbus, OH 43210. Tel: 614-292-4933; Fax: 614-292-4118
    Affiliations
    Department of Medical Biochemistry, The Ohio State University, Columbus, Ohio 43210
    Search for articles by this author
  • Author Footnotes
    * This work was supported by National Institutes of Health Grants DK20387 and DK48744. The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:July 05, 1996DOI:https://doi.org/10.1074/jbc.271.27.16020
      The synthesis of 4,7,10,13,16,19-docosahexaenoic acid (22:6(n-3)) requires that when 6,9,12,15,18,21-tetracosahexaenoic acid (24:6(n-3)) is produced in the endoplasmic reticulum, it preferentially moves to peroxisomes for one cycle of β-oxidation rather than serving as a substrate for membrane lipid synthesis. Both 24:6(n-3) and its precursor, 9,12,15,18,21-tetracosapentaenoic acid (24:5(n-3)), were poor substrates for acylation into 1-acyl-sn-glycero-3-phosphocholine (1-acyl-GPC) by rat liver microsomes. When peroxisomes were incubated with 1-14C- or 3-14C-labeled 7,10,13,16,19-docosapentaenoic acid (22:5(n-3)), [1-14C]22:6(n-3), [3-14C]24:5(n-3), or [3-14C]24:6(n-3), only small amounts of acid-soluble radioactivity were produced when double bond removal at positions 4 and 5 was required. When microsomes and 1-acyl-GPC were included in incubations, the preferred metabolic fate of acids, with their first double bond at either positions 4 or 5, was to move out of peroxisomes for esterification into the acceptor rather than serving as substrates for continued β-oxidation. When [1-14C]22:6(n-3) or [3-14C]24:6(n-3) was incubated with peroxisomes, 2-trans-4,7,10,13,16,19-22:7 accumulated. The first cycle of 20:5(n-3) β-oxidation proceeds through 2-trans-4,8,11,14,17-20:6 and thus requires both Δ3,52,4-dienoyl-CoA isomerase and 2,4-dienoyl-CoA reductase. The accumulation of the substrate for 2,4-dienoyl-CoA reductase, as generated from 22:6(n-3), but not from 20:5(n-3), suggests that this enzyme distinguishes between subtle structural differences. When 22:6(n-3) is produced from 24:6(n-3), its continued degradation is impaired because of low 2,4-dienoyl-CoA reductase activity. This slow reaction rate likely contributes to the transport of 22:6(n-3) out of peroxisomes for rapid acylation into 1-acyl-GPC by microsomes.

      INTRODUCTION

      It has generally been accepted that polyunsaturated fatty acid biosynthesis takes place in the endoplasmic reticulum which is also the major intracellular site for phospholipid biosynthesis (
      • Bell R.M.
      • Coleman R.A.
      ,
      • Das A.K.
      • Horie S.
      • Hajra A.K.
      ,
      • Vance J.
      ). Our recent studies have shown that 7,10,13,16,19-22:5 and 7,10,13,16-22:4 were not desaturated, respectively, to 4,7,10,13,16,19-22:6 and 4,7,10,13,16-22:5 by a microsomal acyl-CoA-dependent 4-desaturase. Instead, they were metabolized to 24-carbon (n-3) and (n-6) acids in the endoplasmic reticulum. The synthesis of 4,7,10,13,16,19-22:6 and 4,7,10,13,16-22:5 thus requires that the 24-carbon precursors move to a site for partial β-oxidation followed by the esterification of the chain-shortened products (
      • Voss A.
      • Reinhart M.
      • Sankarappa S.
      • Sprecher H.
      ,
      • Mohammed B.S.
      • Sankarappa S.
      • Geiger M.
      • Sprecher H.
      ).
      Peroxisomes partially degrade long chain fatty acids, and the chain-shortened intermediates may then move to mitochondria where β-oxidation is completed (
      • Schulz H.
      ,
      • Osmundsen H.
      • Bremer J.
      • Pedersen J.I.
      ). Long chain unsaturated acids are also partially degraded with the subsequent esterification of chain-shortened products into membrane lipids (
      • Voss A.
      • Reinhart M.
      • Sankarappa S.
      • Sprecher H.
      ,
      • Mohammed B.S.
      • Sankarappa S.
      • Geiger M.
      • Sprecher H.
      ,
      • Voss A.
      • Reinhart M.
      • Sprecher H.
      ). Several types of evidence suggest that peroxisomes are the intracellular site for partial β-oxidation. Liver phospholipids, from a patient with Zellweger's syndrome, which lack peroxisomes, had reduced levels of both esterified 4,7,10,13,16-22:5 and 4,7,10,13,16,19-22:6 (
      • Martinez M.
      ). When 3-14C-labeled 7,10,13,16-22:4 and 7,10,13,16,19-22:5 were incubated with fibroblasts, they were metabolized to yield, respectively, esterified 5,8,11,14-20:4 and 5,8,11,14,17-20:5 in control, but not in cells from patients with Zellweger's syndrome (
      • Christensen E.
      • Woldseth B.
      • Hagve T.-A.
      • Poll-The B.T.
      • Wanders R.J.A.
      • Sprecher H.
      • Stokke O.
      • Christophersen B.O.
      ). Normal fibroblasts metabolized [1-14C]9,12,15-18:3 to yield radioactive 4,7,10,13,16,19-22:6 while 24-carbon (n-3) acids were the end metabolites when identical incubations were carried out with fibroblasts from patients with Zellweger's syndrome (
      • Moore S.A.
      • Hurt E.
      • Yoder E.
      • Sprecher H.
      • Spector A.A.
      ).
      The above studies show that peroxisomal partial β-oxidation is, in essence, an anabolic reaction used to chain-shorten endoplasmic reticulum-generated products for subsequent esterification into membrane lipids. We recently demonstrated that peroxisomes do not contain acyl-CoA:1-acylglycero-3-phosphocholine acyltransferase (
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ). The implication of this finding is that partial β-oxidation products, generated by peroxisomes, must move back to the endoplasmic reticulum for esterification into membrane lipids (
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ,
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ). The synthesis of 4,7,10,13,16, 19–22:6 from dietary precursors, and its subsequent esterification into membrane lipids, are thus processes that require the extensive movement of fatty acid between the endoplasmic reticulum and a site for partial β-oxidation which is most likely peroxisomes. The study reported here was undertaken to determine what intracellular processes regulate the synthesis and subsequent esterification of 4,7,10,13,16,19-22:6 into membrane lipids.

      RESULTS

      The results in Table I compare rates of esterification of fatty acids into 1-acyl-GPC with microsomes from rats fed Purina chow with, and without, 0.5% clofibrate. Rates of acylation of 20:5(n-3) and 22:6(n-3), from rats fed Purina chow, were similar to those reported by Lands et al. (
      • Lands W.E.M.
      • Inoue M.
      • Sugiura Y.
      • Okuyama H.
      ). Microsomal esterification of 24-carbon acids into 1-acyl-GPC would curtail their rapid movement to peroxisomes for chain-shortening. Both 24-carbon acids and 22:5(n-3) were poor substrates for acylation with peroxisomes from rats fed chow. When clofibrate was added to the diet, it still was not possible to detect any activity with 24:5(n-3), and the rate of 22:6(n-3) acylation was considerably greater than for its precursor, 24:6(n-3).
      Table I.Rates of acylation of fatty acids into 1-acyl-GPC by liver microsomes from rats fed Purina chow with and without 0.5% clofibrate
      Acyl-CoAChow dietChow diet with 0.5% clofibrate
      nmol/min/mg microsomal protein
      20:5(n-3)88230
      22:5(n-3)35
      22:6(n-3)1644
      24:5(n-3)ND
      ND, no activity as detected.
      ND
      ND, no activity as detected.
      24:6(n-3)38
      a ND, no activity as detected.
      Acyl-CoAs are the true substrates and products of both microsomal desaturation (
      • Strittmatter P.
      • Spatz L.
      • Corcoran D.
      • Rogers M.J.
      • Setlow B.
      • Redline B.
      ,
      • Okayasu T.
      • Nagao M.
      • Ishibashi T.
      • Imai Y.
      ) and malonyl-CoA-dependent chain elongation reactions (
      • Bernert J.T.
      • Sprecher H.
      ). Presumably, when 24-carbon acids are produced in the endoplasmic reticulum, they also are acyl-CoAs. It is not known whether acyl-CoAs move directly to peroxisomes or whether they are hydrolyzed by cytosolic acyl-CoA hydrolases (
      • Yamada J.
      • Matsumoto I.
      • Furihata T.
      • Sakuma M.
      • Suga T.
      ,
      • Svensson L.T.
      • Wilcke M.
      • Alexson S.E.
      ). In the latter case, they must be reactivated by peroxisomes in order to serve as substrates for acyl-CoA oxidase. The results in Table II show that reaction rates for the microsomal activation of all five acids increased when clofibrate was added to the diet. With peroxisomes, the addition of clofibrate increased the rates of activation of 20:5(n-3) and 22:5(n-3) but not the other three (n-3) acids. Since all fatty acids were activated by peroxisomes, albeit at different rates, the subsequent studies were all carried out by generating acyl-CoAs in situ.
      Table IIRates of activation of fatty acids by peroxisomes and microsomes from rats fed Purina chow with and without 0.5% clofibrate
      Fatty acidPeroxisomesMicrosomes
      Without clofibrateWith clofibrateWithout clofibrateWith clofibrate
      nmol acyl-CoA synthesis/min/mg protein
      20:5(n-3)9212671138
      22:5(n-3)58724296
      22:6(n-3)33371742
      24:5(n-3)22201626
      24:6(n-3)16161527
      We reported that when [3-14C]7,10,13,16-22:4 was incubated with peroxisomes, microsomes, and 1-acyl-GPC, the preferred metabolic fate of [1-14C]5,8,11,14-20:4, when it was produced via β-oxidation, was to come out of peroxisomes for esterification by microsomal acyl-CoA:1-acyl-GPC acyltransferase, rather than to serve as a substrate for continued β-oxidation (
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ,
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ). These results suggested that selectivity for peroxisomal β-oxidation, versus microsomal acylation, was a major intracellular control for determining membrane lipid fatty acid composition. In the studies depicted in Figs. 1, 2, 3, 100 µM levels of labeled (n-3) acids were incubated alone with peroxisomes from rats fed clofibrate, as well as when incubations also contained microsomes and 1-acyl-GPC. In addition, 50 µM levels of the labeled acids were incubated with peroxisomes from rats fed Purina chow with, and without, 0.5% clofibrate. With all substrates, i.e. [3-14C]9,12,15,18,21-24:5 and [3-14C]6,9,12,15,18,21-24:6 (Fig. 1), [1-14C]- and [3-14C] 7,10,13,16,19-22:5 (Fig. 2), [1-14C]5,8,11,14,17-20:5 and [1-14C] 4,7,10,13,16,19-22:6 (Fig. 3), the amount of acid-soluble radioactivity produced after 5 min, from rats fed clofibrate, was similar when the substrate concentration was 50 and 100 µM. Rates of β-oxidation at 5 min, from rats fed chow, ranged from 6 to 14% of that found for peroxisomes from rats fed clofibrate.
      Figure thumbnail gr1
      Fig. 1The time-dependent β-oxidation of [3-14C]9,12,15,18,21-24:5 (A) and [3-14C]6,9,12,15,18,21-24:6 (B). Peroxisomes (300 µg of protein) from rats fed Purina chow containing clofibrate were incubated with 50 µM (•) or 100 µM (○) substrate as well as when incubations with 100 µM substrate also contained 100 µM 1-acyl-GPC and 300 µg of microsomal protein (▴). In addition, 300 µg of peroxisomal protein, from rats fed Purina chow, were incubated with 50 µM levels of substrate (▵).
      Figure thumbnail gr2
      Fig. 2The time-dependent β-oxidation of [1-14C]7,10,13,16,19-22:5 (A) and [3-14C]7,10,13,16,19-22:5 (B). Peroxisomes (300 µg of protein) from fats fed Purina chow containing clofibrate were incubated with 50 µM (•) or 100 µM (○) substrate as well as when incubations with 100 µM substrate also contained 100 µM 1-acyl-GPC and 300 µg of microsomal protein (▴). In addition, 300 µg of peroxisomal protein, from rats fed Purina chow, were incubated with 50 µM levels of substrate (▵).
      Figure thumbnail gr3
      Fig. 3The time-dependent β-oxidation of [1-14C]5,8,11,14,17-20:5 (A) and [1-14C]4,7,10,13,16,19-22:6 (B). Peroxisomes (300 µg of protein) from fats fed Purina chow containing clofibrate were incubated with 50 µM (•) or 100 µM (○) substrate as well as when incubations with 100 µM substrate also contained 100 µM 1-acyl-GPC and 300 µg of microsomal protein (▴). In addition, 300 µg of peroxisomal protein, from rats fed Purina chow, were incubated with 50 µM levels of substrate (▵).
      The results in Fig. 1, A and B, show after 5 min that 7.8 and 0.6 nmol of acid-soluble radioactivity were produced, respectively, from 3-14C-labeled 9,12,15,18,21-24:5 and 6,9,12,15, 18,21-24:6 with peroxisomes from rats fed clofibrate. Two cycles of β-oxidation of the acid with five double bonds requires only the enzymes of saturated fatty acid degradation while both NADPH-dependent 2,4-dienoyl-CoA reductase and the trifunctional enzyme, with Δ32-enoyl-CoA isomerase activity, are required for the second cycle of 24:6(n-3) degradation (
      • Palosaari P.M.
      • Hiltunen J.K.
      ). When incubations also contained microsomes and 1-acyl-GPC, there was a reduction in the production of acid-soluble radioactivity from [3-14C]6,9,12,15,18,21-24:6 (Fig. 1B) but not from [3-14C]9,12,15,18,21-24:5 (Fig. 1A).
      The addition of microsomes and 1-acyl-GPC to peroxisomes did not decrease the production of acid-soluble radioactivity from [1-14C]7,10,13,16,19-22:5 (Fig. 2A). With [3-14C]7,10, 13,16,19-22:5, there was about a 90% reduction in the production of acid-soluble radioactivity (Fig. 2B), which is a measure of the second cycle β-oxidation. In a similar way, the addition of microsomes and 1-acyl-GPC to peroxisomes decreased the rate of β-oxidation of both [1-14C]5,8,11,14,17-20:5 (Fig. 3A) and [1-14C]4,7,10,13,16,19-22:6 (Fig. 3B).
      The results in Table III compare the nanomoles of radioactive acids esterified into 1-acyl-GPC when peroxisomes were incubated with microsomes and 1-acyl-GPC. When 1-14C-labeled acids were incubated under these conditions, the fate of the substrate is dictated by its preference for peroxisomal β-oxidation versus esterification by microsomal acyl-CoA:1-acyl-GPC acyltransferase. With 1-14C-labeled 5,8,11,14,17-20:5, 7,10,13,16,19-22:5, and 4,7,10,13,16,19-22:6, 53, 5, and 30 nmol, respectively, were esterified into 1-acyl-GPC. The rank order relative to the nanomoles of fatty acids esterified was the same as was the rate of esterification when rates of acylation were measured directly with microsomes (Table I). At 5 min, 10.2, 31.2, and 1.2 nmol of acid-soluble radioactivity, respectively, were generated when 1-14C-labeled 5,8,11,14,17-20:5, 7,10,13,16,19-22:5, and 4,7,10,13,16,19-22:6 were incubated with peroxisomes Fig. 2, Fig. 3. With the exception of 5,8, 11,14,17-20:5, there was an inverse relationship between rates of peroxisomal β-oxidation versus acylation into 1-acyl-GPC. However, as shown in Table III, when [1-14C]5,8,11,14,17-20:5 was generated from [3-14C]7,10,13,16,19-22:5, via β-oxidation, 42 nmol of 5,8,11,14,17-20:5 were esterified versus 53 nmol when [1-14C]5,8,11,14,17-20:5 was used directly as the substrate. A major metabolic fate of 5,8,11,14,17-20:5, when it was generated from 7,10,13,16,19-22:5, was thus esterification.
      Table IIIIncorporation of radioactive fatty acids into phospholipids when peroxisomes were incubated with 1-acyl-GPC and microsomes
      Substrate20:5 (n-3)22:5 (n-3)22:6 (n-3)24:5 (n-3)24:6 (n-3)
      nmol esterified radioactive fatty acids
      [1-14C]20:5(n-3)53
      [1-14C]22:5(n-3)5
      [3-14C]22:5(n-3)427
      [1-14C]22:6(n-3)30
      [3-14C]24:5(n-3)63
      [3-14C]24:6(n-3)416
      Neither 24-carbon acid was readily esterified into 1-acyl-GPC when they were incubated alone with microsomes (Table I) or in the microsomal-peroxisomal mixing studies (Table III). When 3-14C-labeled 9,12,15,18,21-24:5 and 6,9,12,15,18, 21–24:6 were incubated alone with peroxisomes, 7.8 and only 0.6 nmol of the two substrates, respectively, were metabolized through two cycles of β-oxidation after 5 min (Fig. 1) As shown in Table I, the rate of acylation of 4,7,10,13,16,19-22:6 into 1-acyl-GPC by microsomes was 8.3-fold greater than for 7,10,13,16,19-22:5. In the mixing experiments (Table III), 41 and 6 nmol, respectively, of 4,7,10,13,16,19-22:6 and 7,10, 13,16,19-22:5 were esterified when they were generated from the appropriate 3-14C-labeled 24-carbon precursors. There is thus an inverse relationship between rates of β-oxidation of 7,10,13,16,19-22:5 and 4,7,10,13,16,19-22:6 versus their rates of acylation into the acceptor.
      Of the substrates that were used in these studies, [1-14C]4,7,10,13,16,19-22:6 and [3-14C]6,9,12,15,18,21-24:6 were unique in that. when they were incubated alone with peroxisomes, radioactive catabolites accumulated prior to CoASH-dependent thiolytic cleavage of β-ketoacyl-CoAs. The compound that eluted immediately prior to unmetabolized [1-14C]4,7,10,13,16,19-22:6, in the HPLC radiochromatogram in Fig. 4A, had an absorbance maximum at 265 nm, which is what was observed for authentic methyl 2-trans-4-cis-10:2 (
      • Luthria D.L.
      • Baykousheva S.P.
      • Sprecher H.
      ). Compound 3 is thus 2-trans-4,7,10,13,16,19-22:7. The ultraviolet spectra of compounds 1 and 2 had absorbance maxima at 256, 267, and 277 nm, which, based on our previous studies with 4,7,10-16:3 (
      • Luthria D.L.
      • Baykousheva S.P.
      • Sprecher H.
      ), are isomeric 3,5,7,10,13,16,19-22:7 trienes. These conjugated trienes could be produced from 2-trans-4,7,10,13,16,19-22:7 by peroxisomal Δ3,52,4-dienoyl-CoA isomerase operating in the reverse direction (
      • He X.-Y.
      • Shoukry K.
      • Chu C.
      • Yang J.
      • Sprecher H.
      • Schulz H.
      ). The accumulation of 2-trans-4,7,10,13,16,19-22:7 suggests that the reaction catalyzed by NADPH-dependent 2,4-dienoyl-CoA reductase, in addition to fatty acid oxidase, may play a regulatory role in peroxisomal β-oxidation.
      Figure thumbnail gr4
      Fig. 4HPLC radiochromatogram obtained when 100 µM[1-14C]4,7,10.13,16,19-22:6 was incubated with 300 µg of peroxisomal protein in the presence of both NAD+and NADPH (A). After 30 min, the incubation was terminated, methyl esters were made and analyzed by HPLC. Compounds 1 and 2 were tentatively identified as 3,5,7,10,13,16,19-22:7 isomers which could be produced from 2-trans-4,7,10,13,16,19-22:7 (compound 3), by Δ3,52,4-dienoyl-CoA isomerase. In B, 100 µM [1-14C]5,8,11,14,17-20:5 was incubated with 300 µg of peroxisomal protein in the absence of NAD+ but with NADPH. After 30 min, the incubation was terminated, methyl esters were made and analyzed by HPLC. Compound 1 is methyl-3-hydroxy-8,11,14,17-20:4; compounds 2 and 3 were tentatively identified as isomeric methyl esters of 3,5,8,11,14,17-20:6; compound 4 is methyl-2-trans-4,8,11,14,17-20:6; compound 5 is probably methyl-2-trans-8,11,14,17-20:5 while compound 6 was shown to be the methyl ester of 8,11,14,17-20:4.
      Since the above studies were carried out with 100 µM 4,7,10,13,16,19-22:6, additional experiments were carried out to determine if the substrate for NADPH-dependent 2,4-dienoyl-CoA reductase accumulated when 4,7,10,13,16,19-22:6 was generated in situ. The results in Table IV show that there was a time-dependent accumulation of [1-14C]4,7,10,13,16,19-22:6 when 100 µM [3-14C]6,9,12,15,18,21-24:6 was incubated with peroxisomes. At each time point, the amount of 2-trans-4,7,10,13,16,19-22:7 that accumulated was approximately equal to the amount of substrate that underwent two complete cycles of β-oxidation.
      Table IVTime-dependent β-oxidation of [3-14C]6,9,12,15,18,21-24:6 and the accumulation of radioactive catabolites
      MinutesAcid-soluble radioactivity3,5,7,10,13,16,19-22:7 isomers2-trans-4,7,10,13,16,19-22:74,7,10,13,16,19-22:6
      nmol radioactive catabolites
      2.50.80.52.5
      5.02.02.37.1
      10.04.312.613.0
      20.09.65.78.719.5
      30.013.36.812.624.9
      The accumulation of 2-trans-4,7,10,13,16,19-22:7, when either 22:6(n-3) or 24:6(n-3) was the substrate, implies that the enzyme NADPH-dependent 2,4-dienoyl-CoA reductase catalyzes a control step in peroxisomal β-oxidation. When [1-14C]5,8,11,14,17-20:5 was incubated with peroxisomes, in the presence of all cofactors, no radioactive catabolites were detected. We recently showed that the first cycle of arachidonate β-oxidation requires both NADPH-dependent-2,4-dienoyl-CoA reductase and Δ3,5Δ2,4-dienoyl-CoA isomerase (
      • Luthria D.L.
      • Baykousheva S.P.
      • Sprecher H.
      ). When [1-14C]5,8,11,14,17-20:5 was incubated without NAD+, but with NADPH, the mass spectrum of the methyl ester-trimethylsilyl ether of compound 1 (Fig. 4B) had a molecular ion at m/z = 406 (3%) and another at m/z = 391 (M − 15; 32%) showing that it is 3-hydroxy-8,11,14,17-20:4. Double bond removal at position 5 thus takes place prior to completion of the first cycle of β-oxidation. The two metabolites, compounds 2 and 3, had absorbance maximum at 233 nm, which is what would be expected for acids with two conjugated double bonds, i.e. 3,5,8,11,14,17-20:6 isomers. Compound 4 had an absorbance maximum identical with authentic 2-trans-4-cis-10:2 showing that it is the substrate for NADPH-dependent 2,4-dienoyl-CoA reductase, i.e. 2-trans-4,8,11,14,17-20:6. Compound 5 was not identified, but most likely it is 2-trans-8,11,14,17-20:5 which upon NADPH-dependent reduction (
      • Dommes V.
      • Luster W.
      • Cvetanovic M.
      • Kunau W.H.
      ,
      • Horie S.
      • Suzuki T.
      • Suga J.
      ) would yield compound 6 whose chromatographic properties and mass spectrum were identical with synthetic methyl 8,11,14,17-20:4. These results, which are identical with those for the peroxisomal β-oxidation of arachidonate (
      • Luthria D.L.
      • Baykousheva S.P.
      • Sprecher H.
      ), show that NADPH-dependent 2,4-dienoyl-CoA reductase is required for the first cycle of fatty acid degradation for both 4,7,10,13,16,19-22:6 and 5,8,11,14,17-20:5. However, 2-trans-4,7,10,13,16,19-22:7 and 2-trans-4,8,11,14,17-20:6 are the respective substrates for NADPH-dependent 2,4-dienoyl-CoA reductase. These two substrates differ in that the third double bonds are located, respectively, at carbons 7 and 8.

      DISCUSSION

      The diagram in Fig. 5 depicts a simplified overview of the factors regulating the biosynthesis and subsequent incorporation of 4,7,10,13,16,19-22:6 into membrane lipids. The diagram implies that dietary linolenate is metabolized to 6,9,12,15,18,21-24:6, via 5,8,11,14,17-20:5, by a series of chain elongation and desaturation reactions in the endoplasmic reticulum. Any fatty acid in this metabolic pathway may also be removed and esterified into precursors for phospholipid biosynthesis. These processes are also localized primarily in the endoplasmic reticulum (
      • Bell R.M.
      • Coleman R.A.
      ,
      • Das A.K.
      • Horie S.
      • Hajra A.K.
      ,
      • Vance J.
      ). If 24-carbon acids were rapidly esterified into acceptors in the endoplasmic reticulum, it would, in essence, curtail their movement to a site for partial β-oxidation. When 24-carbon (n-3) and (n-6) acids were incubated with hepatocytes, only small amounts were esterified into phospholipids (
      • Voss A.
      • Reinhart M.
      • Sankarappa S.
      • Sprecher H.
      ,
      • Mohammed B.S.
      • Sankarappa S.
      • Geiger M.
      • Sprecher H.
      ). By direct measurement, it was now shown that both 24:5(n-3) and 24:6(n-3) are poor substrates for microsomal acyl-CoA:1-acyl-GPC acyltransferase. The implication of these findings is that when 24-carbon acids are produced in the endoplasmic reticulum they preferentially move to another site for further metabolism.
      Figure thumbnail gr5
      Fig. 5Pathway for the metabolism of (n-3) fatty acids. The schematic shows that when 5,8,11,14,17-20:5 is metabolized to 6,9,12,15,18,21-24:6 in microsomes, any fatty acid in this pathway may also be removed by esterification into precursors for phospholipid biosynthesis. In addition, any fatty acid may move from microsomes to peroxisomes for partial β-oxidation followed by the transfer of the chain-shortened metabolite back to microsomes for phospholipid biosynthesis.
      Based on other desaturation and chain elongation studies (
      • Strittmatter P.
      • Spatz L.
      • Corcoran D.
      • Rogers M.J.
      • Setlow B.
      • Redline B.
      ,
      • Okayasu T.
      • Nagao M.
      • Ishibashi T.
      • Imai Y.
      ,
      • Bernert J.T.
      • Sprecher H.
      ), it is assumed that when 24:5(n-3) and 24:6(n-3) are produced in the endoplasmic reticulum they are acyl-CoAs and not free fatty acids. Rat liver cytosol contains acyl-CoA hydrolases which are up-regulated by feeding clofibrate (
      • Yamada J.
      • Matsumoto I.
      • Furihata T.
      • Sakuma M.
      • Suga T.
      ,
      • Svensson L.T.
      • Wilcke M.
      • Alexson S.E.
      ). It is not known whether acyl-CoAs move from the endoplasmic reticulum to other subcellular organelles or whether they are hydrolyzed in the cytosol and subsequently reactivated. In the study reported here we observed that peroxisomes have the capacity to activate 24-carbon (n-3) fatty acids, albeit at a somewhat slower rate than for 20- and 22-carbon acids. The efficient synthesis of 4,7,10,13,16,19-22:6 requires that 6,9,12,15,18,21-24:6 is targeted to an intracellular site for partial β-oxidation. Microsomes, peroxisomes, and mitochondria all contain an identical protein that is designated as a long chain acyl-CoA synthetase (
      • Philipp D.P.
      • Parsons P.
      ,
      • Tanaka T.
      • Hosaka K.
      • Hoshimura M.
      • Numa S.
      ,
      • Miyazawa S.
      • Hashimoto T.
      • Yokoto S.
      ,
      • Suzuki H.
      • Kawarabayasi Y.
      • Kondo J.
      • Abe T.
      • Nishikawa K.
      • Kimura S.
      • Hashimoto T.
      • Yamamoto T.
      ). Microsomes and peroxisomes, but not mitochondria, also contain a second enzyme, which has not been purified, but it is designated as a very long chain acyl-CoA synthetase (
      • Lagewig W.
      • Wanders R.J.A.
      • Tager J.M.
      ,
      • Mannaerts G.P.
      • Van Veldhoven P.P.
      ). If 24-carbon acyl-CoAs are hydrolyzed in the cytosol, and not reactivated by mitochondrial long chain acyl-CoA synthetase, it would be an efficient type of control to channel them to peroxisomes for activation and subsequent partial β-oxidation. This hypothesis is currently being evaluated.
      The diagram in Fig. 5 implies that when fatty acids enter peroxisomes they undergo one or more cycles of β-oxidation followed by the transfer of the chain-shortened metabolite back to the endoplasmic reticulum for use in membrane lipid synthesis. In this study, as well as in previous publications (
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ,
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ), it was observed that generally there is an inverse relationship between rates of peroxisomal β-oxidation versus rates of acylation into 1-acyl-GPC. If double bond removal is not required for peroxisomal β-oxidation, that acid is a poor substrate for microsomal acyl-CoA:1-acyl-GPC acyltransferase. For example, 9,12,15,18,21-24:5, and its chain-shortened product, 7,10,13,16,19-22:5, are relatively poor substrates for acylation into 1-acyl-GPC. The metabolism of 9,12,15,18,21-24:5, via two cycles of β-oxidation to 5,8,11,14,17-20:5, requires only the enzymes of saturated fatty acid degradation. The first cycle of 6,9,12,15,18,21-24:6 β-oxidation uses only the enzymes of saturated fatty acid degradation, but when 4,7,10,13,16,19-22:6 is produced its continued β-oxidation requires NADPH-dependent 2,4-dienoyl-CoA reductase. When 4,7,10,13,16,19-22:6 or 6,9,12,15,18,21-24:6 was incubated with peroxisomes, 2-trans-4,7,10,13,16,19-22:7 accumulated. This finding suggests that the reaction catalyzed by NADPH-dependent 2,4-dienoyl-CoA reductase is a control step in regulating unsaturated fatty acid degradation. The preferred pathway for 4,7,10,13,16,19-22:6 metabolism, when it is produced in peroxisomes, is to move out of peroxisomes, via some unknown pathway, for rapid esterification into 1-acyl-GPC rather than to serve as a substrate for continued β-oxidation.
      In contrast to 4,7,10,13,16,19-22:6, the first cycle of 5,8,11,14,17-20:5 β-oxidation is not only rapid, but, moreover, 2-trans-4,8,11,14,17-20:6, an obligatory catabolite, does not accumulate. This finding implies that when the double bonds in the substrate are in the 2-trans-4,7-positions the substrate is slowly metabolized by NADPH-dependent 2,4-dienoyl-CoA reductase. Conversely, when the third double bond is shifted to position 8, i.e. 2-trans-4,8,11,14,17-20:6, it is readily reduced. This hypothesis is further supported by the observation that 2-trans-4,7,10-16:4 accumulated when 4,7,10-16:3 was incubated with peroxisomes. No radioactive catabolite was detected when [1-14C]5,8-14:2 or [1-14C]5,8,11,14-20:4 were incubated under identical conditions whose β-oxidation was shown to proceed, respectively, through 2-trans-4,8-14:3 and 2-trans-4,8,11,14-20:5 (
      • Luthria D.L.
      • Baykousheva S.P.
      • Sprecher H.
      ).
      In summary, the results obtained here and in other studies (
      • Voss A.
      • Reinhart M.
      • Sankarappa S.
      • Sprecher H.
      ,
      • Mohammed B.S.
      • Sankarappa S.
      • Geiger M.
      • Sprecher H.
      ,
      • Christensen E.
      • Woldseth B.
      • Hagve T.-A.
      • Poll-The B.T.
      • Wanders R.J.A.
      • Sprecher H.
      • Stokke O.
      • Christophersen B.O.
      ,
      • Moore S.A.
      • Hurt E.
      • Yoder E.
      • Sprecher H.
      • Spector A.A.
      ,
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ,
      • Baykousheva S.P.
      • Luthria D.L.
      • Sprecher H.
      ) show that there is considerable recycling of fatty acids between peroxisomes and the endoplasmic reticulum. The slow rate of NADPH-dependent 2,4-dienoyl-CoA reductase, a catabolic reaction, may be viewed as a major anabolic control point in the biosynthesis of 22:6(n-3). Once this acid is produced, it preferentially moves out of peroxisomes for acylation into 1-acyl-GPC rather than serving as a substrate for continued β-oxidation. The major objective of the above studies has focused on determining how peroxisomes and the endoplasmic reticulum interact relative to the synthesis of 20- and 22-carbon unsaturated fatty acids. It has recently been shown that when 5–14:1 and 5,8-14:2 are formed, respectively, by β-oxidation of oleate and linoleate, they are used as substrates to covalently modify photoreceptor-specific proteins (
      • DeMar J.C.
      • Wensel T.G.
      • Anderson R.E.
      ). A number of different studies now clearly show that peroxisomes are required for producing acids that are subsequently used as substrates for a number of different anabolic reactions.

      REFERENCES

        • Bell R.M.
        • Coleman R.A.
        Annu. Rev. Biochem. 1980; 49: 459-487
        • Das A.K.
        • Horie S.
        • Hajra A.K.
        J. Biol. Chem. 1992; 267: 9724-9730
        • Vance J.
        J. Biol. Chem. 1990; 265: 7248-7256
        • Voss A.
        • Reinhart M.
        • Sankarappa S.
        • Sprecher H.
        J. Biol. Chem. 1991; 266: 19995-20000
        • Mohammed B.S.
        • Sankarappa S.
        • Geiger M.
        • Sprecher H.
        Arch. Biochem. Biophys. 1995; 317: 179-184
        • Schulz H.
        Biochim. Biophys. Acta. 1991; 1081: 109-120
        • Osmundsen H.
        • Bremer J.
        • Pedersen J.I.
        Biochim. Biophys. Acta. 1991; 1085: 141-158
        • Voss A.
        • Reinhart M.
        • Sprecher H.
        Biochim. Biophys. Acta. 1992; 1127: 33-40
        • Martinez M.
        Lipids. 1989; 24: 261-265
        • Christensen E.
        • Woldseth B.
        • Hagve T.-A.
        • Poll-The B.T.
        • Wanders R.J.A.
        • Sprecher H.
        • Stokke O.
        • Christophersen B.O.
        Scand. J. Clin. Lab. Invest. 1993; 536: 61-74
        • Moore S.A.
        • Hurt E.
        • Yoder E.
        • Sprecher H.
        • Spector A.A.
        J. Lipid Res. 1995; 36: 2433-2443
        • Baykousheva S.P.
        • Luthria D.L.
        • Sprecher H.
        FEBS Lett. 1995; 367: 198-200
        • Baykousheva S.P.
        • Luthria D.L.
        • Sprecher H.
        J. Biol. Chem. 1994; 269: 18390-18394
        • Bar-Tana J.
        • Rose G.
        • Shapiro B.
        Methods Enzymol. 1975; 35B: 117-120
        • Lands W.E.M.
        • Inoue M.
        • Sugiura Y.
        • Okuyama H.
        J. Biol. Chem. 1982; 257: 14968-14972
        • Bartlett K.
        • Hovik R.
        • Eaton S.
        • Watmough N.J.
        • Osmundsen H.
        Biochem. J. 1990; 270: 175-180
        • Strittmatter P.
        • Spatz L.
        • Corcoran D.
        • Rogers M.J.
        • Setlow B.
        • Redline B.
        Proc. Natl. Acad. Sci. U. S. A. 1974; 71: 4565-4569
        • Okayasu T.
        • Nagao M.
        • Ishibashi T.
        • Imai Y.
        Arch. Biochem. Biophys. 1981; 206: 21-28
        • Bernert J.T.
        • Sprecher H.
        Biochim. Biophys. Acta. 1979; 573: 436-442
        • Yamada J.
        • Matsumoto I.
        • Furihata T.
        • Sakuma M.
        • Suga T.
        Arch. Biochem. Biophys. 1994; 308: 118-125
        • Svensson L.T.
        • Wilcke M.
        • Alexson S.E.
        Eur. J. Biochem. 1995; 230: 813-820
        • Palosaari P.M.
        • Hiltunen J.K.
        J. Biol. Chem. 1990; 265: 2446-2449
        • Luthria D.L.
        • Baykousheva S.P.
        • Sprecher H.
        J. Biol. Chem. 1995; 270: 13771-13776
        • He X.-Y.
        • Shoukry K.
        • Chu C.
        • Yang J.
        • Sprecher H.
        • Schulz H.
        Biochem. Biophys. Res. Commun. 1995; 215: 15-22
        • Dommes V.
        • Luster W.
        • Cvetanovic M.
        • Kunau W.H.
        Eur. J. Biochem. 1982; 125: 335-341
        • Horie S.
        • Suzuki T.
        • Suga J.
        Arch. Biochem. Biophys. 1989; 274: 64-73
        • Philipp D.P.
        • Parsons P.
        J. Biol. Chem. 1979; 254: 10776-10784
        • Tanaka T.
        • Hosaka K.
        • Hoshimura M.
        • Numa S.
        Eur. J. Biochem. 1979; 98: 165-172
        • Miyazawa S.
        • Hashimoto T.
        • Yokoto S.
        J. Biochem. (Tokyo). 1985; 98: 723-733
        • Suzuki H.
        • Kawarabayasi Y.
        • Kondo J.
        • Abe T.
        • Nishikawa K.
        • Kimura S.
        • Hashimoto T.
        • Yamamoto T.
        J. Biol. Chem. 1990; 265: 8681-8685
        • Lagewig W.
        • Wanders R.J.A.
        • Tager J.M.
        Eur. J. Biochem. 1991; 196: 519-523
        • Mannaerts G.P.
        • Van Veldhoven P.P.
        Biochimie (Paris). 1993; 75: 147-158
        • DeMar J.C.
        • Wensel T.G.
        • Anderson R.E.
        J. Biol. Chem. 1996; 271: 5007-5016