Main

HETE are monohydroxylated derivatives of arachidonic acid, mainly produced via the action of stereospecific lipoxygenases(1). The initial products are unstable hydroperoxides which are either spontaneously or peroxidatively reduced to the corresponding monohydroxylated products(1). 12-HETE and 15-HETE, two of the major forms of HETE, are synthesized mainly by activated macrophages, platelets, or neutrophils(13). Besides chemotactic effects, a number of important actions have recently been attributed to the HETE. These include effects on intracellular calcium concentration, cell proliferation, regulation of phospholipase activity, prostaglandin as well as leukotriene formation, polypeptide hormone secretion, receptor-mediated cyclic nucleotide formation, and arachidonic acid incorporation into inositol phospholipids(410).

In view of these biologic activities it is important to establish how inactivation and degradation of HETE is controlled. The intracellular location of HETE degradation is still uncertain. Previous in vitro studies have shown that peroxisome-deficient cells do not convert 12- and 15-HETE to chain-shortened, polar metabolites, suggesting that peroxisomes are the intracellular location for β-oxidation of these compounds(11, 12). However, further data demonstratedβ-oxidation of 12- and 15-HETE in purified rat liver mitochondria, suggesting that besides peroxisomes mitochondria are also competent to carry out β-oxidation of both compounds(13).

The aim of our study was therefore to investigate the importance of peroxisomes and mitochondria in HETE oxidation in humans in vivo. Inasmuch as a recent study showed that 20-HETE is present as a glucuronide conjugate in human urine(14), we studied whether and in which form 12- and 15-HETE are excreted in the urine of patients with a peroxisome deficiency disorder, patients with a defective mitochondrial long-chain fatty acid oxidation, and healthy subjects.

METHODS

Patients. Excretion of 12- and 15-HETE was studied in eight patients with a peroxisome deficiency disorder (ZS), three patients with a defect of mitochondrial long-chain fatty acid oxidation (LCAD deficiency), and eight healthy control subjects. All patients with ZS exhibited the characteristic clinical and biochemical abnormalities described for ZS(15). Specific biochemical analyses in these patients included very long-chain fatty acids (>C22) in plasma and fibroblasts as well as plasma bile acid intermediates and de novo plasmalogen biosynthesis in cultured fibroblasts. Mitochondrialβ-oxidation activity was assayed in cultured fibroblasts using[1-14C]palmitic acid and found to be in the range of normal subjects. The biochemical characteristics of these patients have been published already in detail(16). All patients with a defect of mitochondrial long-chain fatty acid oxidation(17) had LCAD deficiency as measured in cultured skin fibroblasts(18).

Urine was obtained from spontaneous micturition and stored at -80°C until analysis. The presence of pathologic constituents was excluded by a screening test (Combur9 test, Boehringer, Mannheim, Germany).

Chemicals. 5,6,8,9,11,12,14,15-[3H]-12-HETE (119 Ci/mmol) and 5,6,8,9,11,12,14,15-[3H]-15-HETE (213 Ci/mmol) were obtained from Amersham Corp. (Arlington Heights, IL). 5-, 8-, 12-, and 15-HETE were from Cascade Biochem. Ltd. (University of Reading, UK).β-Glucuronidase (Escherichia coli), 4-hydroxy-2,2,6,6-tetramethylpoperidine-N(1)-oxyl,2,3,4,5,6-pentafluorobenzylbromide as well as N,N-diisopropylethylamine were purchased from Sigma Chemical Co. (St. Louis, MO).N,O-Bis-(trimethylsilyl)-trifluoroacetamide + 1% trimethylchlorosilane was obtained from Pierce (Rockford, IL), and Sep-Pak C18 cartridges were from Waters Associates (Milford, MA). Other chemicals and solvents were of the best grades commercially available.

Urine extraction and purification. Five-milliliter aliquots of urine were labeled with 3 500 dpm each of [3H]-12-HETE and[3H]-15-HETE. The samples were acidified to pH 4.5 by addition of 0.1 M HCl, mixed, and pumped slowly through activated Sep-Pak C18 cartridges. The cartridges were washed with 50 mL of distilled H2O, and HETE were eluted with 5 mL of 90% aqueous methanol containing 1 mM 4-hydroxy-2,2,6,6-tetramethylpiperidine-N(1)-oxyl and 0.5 mM EDTA. The eluates were evaporated to dryness under reduced pressure and resuspended in 30% acetonitrile and 70% water acidified to pH 3.4 with phosphoric acid.

β- Glucuronidase treatment. Forβ-glucuronidase treatment urine samples were incubated for 2 h at 37°C with 0.5 mL of a solution of β-glucuronidase (1 mg/mL) in 0.075 M potassium phosphate buffer (pH 6.8) containing 0.1% BSA(14).

RP-HPLC. Fractions containing the different HETE were obtained by RP-HPLC on a C18 Hypersil column (4.6 × 250 mm, 5-μm particle size; Shandon, Runcorn, UK) using as mobile phase a solvent gradient consisting of phosphoric acid-acidified water and acetonitrile with the latter component increasing linearly from 30 to 100% over the 45-min run time with a constant flow rate of 1 mL/min(7, 11).

RIA. The RP-HPLC eluate was collected in 1-mL fractions, of which 300 μL were counted for calculation of the 3H recovery, and 700 μL were dried for subsequent RIA. These were performed by specific RIA kits (Paesel & Lorey, Frankfurt, Germany). Separation of antibody-bound analyte from unbound analyte was done by magnetic separation. The molar cross-reactivities at 50% binding for the 12-HETE antibody were: 12-HETE 100%; 5-HETE 0.2%; 15-HETE 0.3%; leukotriene B4 0.1%; leukotriene E4 0.1%; prostaglandin E2 0.1%; prostaglandin F 0.1%; thromboxane B2 0.1%. The corresponding values for the 15-HETE-antibody were: 15-HETE 100%; 5-HETE 0.1%; 12-HETE 0.5%; leukotriene B4 0.1%; leukotriene E4 0.1%; prostaglandin E2 0.1%; prostaglandin F 0.1%; thromboxane B2 0.1%. Data were corrected for the recovery of the 3H-labeled HETE added as internal standards.

GC/NICI-MS. For GC/NICI-MS analysis, the RP-HPLC fractions containing the different HETE were converted into their hydrogenated PFB ester/TMS ether derivatives as described elsewhere(19). Briefly, hydrogenation was performed by addition of 1 mg of a PtO2 catalyst and bubbling of hydrogen gas through a methanolic solution at 0°C for 2 min. After removal of the catalyst by filtration using a 0.2-μm microfilter unit the solution was dried under nitrogen and redissolved in 50μL of methylene chloride. Five microliters of pentafluorobenzylbromide and 10 μL of N,N-diisopropylethylamine were added, the sample was maintained for 15 min at room temperature, dried under nitrogen, and incubated with 70 mL of N,O-bis-(trimethylsilyl)-trifluoroacetamide + 1% trimethylchlorosilane at 60°C for 60 min to generate the hydrogenated PFB ester/TMS ether derivatives. The solution was cooled to room temperature, dried under nitrogen, and redissolved in 100 μL of isooctane. Two microliters of this solution were injected for a single GC/MS analysis.

GC/NICI-MS analyses were performed on a double-focusing instrument type Finnigan MAT 95 (Bremen, Germany) coupled to a GC type Hewlett-Packard 5890. Isobutane was used as reagent gas. Separation of HETE positional isomers was achieved as previously described(17) using 30-m DB-5 ms column (0.25 mm inside diameter, 0.25 μm film thickness, J&W Scientific). The GC temperature program started with rapid heating from 50°C to 230°C in 5 min followed by a linear increase from 230 to 270°C at a rate of 0.5°C/min.

RESULTS

Identification of 12- and 15-HETE. In the investigation of HETE as their hydrogenated TMS/PFB derivatives all regioisomeric HETE give rise to an intense [M - PFB] fragment ion at m/z 399. Using capillary GC the HETE positional isomers are separated in the order of the position of the oxidized carbon atom, i.e. the derivative of 3-hydroxyeicosanoic acid elutes before the corresponding derivative of 4-hydroxyeicosanoic acid, and so forth. Although separation is not complete for positional isomers near the central part of the carbon chain, it is sufficient to allow a specific detection of the four lipoxygenase-derived species oxidized at positions 5, 8, 12, and 15(19). This is demonstrated in the upper trace of Figure 1 showing the single ion monitoring analysis atm/z 399 of a standard mixture containing similar amounts of 5-, 8-, 12-, and 15-HETE (Fig. 1a). 12- and 15-HETE were isolated by HPLC from the urine of healthy subjects and patients with ZS syndrome or LCAD deficiency, as described above.

Figure 1
figure 1

GC/NICI-MS analyses of HETE with single ion monitoring of m/z 399 under identical conditions; (a) standard mixture containing 5-, 8-, 12-, and 15-HETE; (b) representative spectrum of 12- and 15-HETE HPLC fraction from a healthy control patient documenting the lack of urinary 12- and 15-HETE excretion; (c) 12-HETE HPLC fraction isolated from a ZS patient's urine; and (d) 15-HETE HPLC fraction isolated from the same ZS patient's urine.

In urine samples from healthy control subjects and patients with LCAD 12- and 15-HETE could not be detected. Figure 1b shows a representative spectrum from a healthy control patient, documenting the lack of urinary 12- and 15-HETE excretion. In contrast, the presence of 12- and 15-HETE in urine samples was verified in all eight ZS patients by GC/NICI-MS as exemplary shown in Figure 1, c and d, respectively. Because reproducibility of the GC retention times is very high (within day reproducibility ±1 s) this method provides a specific identification of the lipoxygenase-derived HETE and thus confirms the presence of 12- and 15-HETE in urine of ZS patients.

Quantification of 12- and 15-HETE. The urinary concentrations of 12- and 15-HETE for the patients with a peroxisome deficiency disorder, patients with LCAD deficiency, and normal subjects were determined by RIA and are given in Table 1. Both HETE were present in the urine from ZS patients at concentrations in the picogram/mL range. In contrast, analyses of urine samples from LCAD-deficient patients and normal subjects showed that both 12- as well as 15-HETE were below the detection limit in these groups.

Table 1 Concentrations of 12- and 15-HETE in urine of patients with peroxisomal deficiency (ZS), LCAD deficiency, and normal subjects

β-Glucuronidase treatment for 2 h had no influence on the urinary concentrations of 12- and 15-HETE in normal subjects and patients with LCAD deficiency nor in ZS patients. Additionally, no influence in 12- and 15-HETE concentrations was conserved when the urine samples were incubated withβ-glucuronidase for periods up to 20 h.

DISCUSSION

Our RIA and GC/MS data demonstrate for the first time that 12- and 15-HETE are excreted in the urine of peroxisome-deficient subjects, whereas both compounds are below the detection limit in the urine of patients with a defect of mitochondrial long-chain fatty acid oxidation and in urine of normal subjects. Up to now, the presence of 12- or 15-HETE in human urine has not been reported, either under physiologic or under pathophysiologic conditions. In contrast to 20-HETE, which is excreted as a glucuronide conjugate in human urine(14), treatment with β-glucuronidase was without any significance on 12- and 15-HETE concentrations in all urine samples analyzed.

It seems likely that 12- and 15-HETE are not detectable in the urine of normal subjects and LCAD-deficient patients due to their oxidation to more polar metabolites. The conversion of 12- and 15-HETE via β-oxidation to 16- or less carbon-reduced metabolites has already been shown in renal tubular epithelial cells(7), human vascular smooth muscle cells(20), and normal human skin fibroblasts(11). The metabolite produced from 12-HETE is 8-hydroxyhexadecatrienoic acid and that produced from 15-HETE is 11-hydroxyhexadecatrienoic acid(20). The absence of 12- and 15-HETE in normal human urine as well as in urine samples from patients with LCAD deficiency and the presence of these HETE in the urine of ZS patients implicates that oxidation of these compounds is deficient or absent in ZS, the prototypical peroxisomal deficiency disease. This is consistent with previous in vitro findings showing that human skin fibroblasts from ZS patients(11) and peroxisome-deficient wild-type mutant Chinese hamster ovary cells(12) do not convert 12- and 15-HETE to oxidative metabolites.

Recently, it was reported that isolated rat liver and kidney peroxisomes convert 12-HETE to 8-hydroxyhexadecatrienoic acid, confirming a peroxisomalβ-oxidation(21). However, mitochondrial function might also be abnormal in peroxisomedeficient cells of patients(22, 23). Additionally, in vitro data also point to mitochondria as potential sites of β-oxidation of 12- and 15-HETE(13). This could induce doubt as to whether HETE oxidation is possible only on the peroxisomal pathway in men. We therefore studied ZS patients showing a normal mitochondrial β-oxidation capacity in fibroblasts. Because these patients are peroxisome-deficient but have normal mitochondrial enzymatic activity, their urinary excretion of 12- and 15-HETE implicates a failure to convert both HETE to oxidative metabolites. Further evidence that the β-oxidation process of these HETE occurs in the peroxisomes rather than in the mitochondria was obtained from urine analyses of LCAD-deficient patients. As opposed to the 12- and 15-HETE excretion by ZS patients, the mitochondrial long-chain fatty acid β-oxidation-deficient patients showed, in analogy to normal subjects, no detectable amounts of these HETE. Thus, patients with LCAD deficiency are able to oxidize both HETE, indicating that in vivo this oxidative process does not take place in the mitochondria. All together, these results make it likely that in vivo oxidation of both HETE occurs mainly in the peroxisomes and not in the mitochondria.

Lack of HETE oxidation in peroxisome-deficient patients resulting in urinary excretion of 12- and 15-HETE represents an additional new specific diagnostic tool in patients with ZS. Because these HETE possess a diversity of important biologic properties, impaired degradation and inactivation of these potent mediators may be of pathophysiologic relevance in the course of the disease.