Studies on Toxic Oil Syndrome: Stereoselective Hydrolysis of 3

oil syndrome (TOS), an intoxication episode that took place in Spain in 1981, ... In Vitro Bioactivation of 3-(N-Phenylamino)propane-1,2-diol by H...
0 downloads 0 Views 136KB Size
Download PDF
Chem. Res. Toxicol. 2004, 17, 889-895

889

Studies on Toxic Oil Syndrome: Stereoselective Hydrolysis of 3-(Phenylamino)propane-1,2-diol Esters by Human Pancreatic Lipase Anna Morato´, Anna Martı´nez-Cabot, Jordi Escabro´s, Jordi Bujons, and Angel Messeguer* Department of Biological Organic Chemistry, Institut d’Investigacions Quı´miques i Ambientals de Barcelona (CSIC), 08034 Barcelona, Spain Received February 2, 2004

The ingestion of rapeseed oil batches denatured with aniline and illegally refined and distributed by street vendors was responsible for toxic oil syndrome (TOS), an intoxication episode that took place in Spain in 1981, causing over 400 deaths and affecting more than 20 000 people. Despite the intense research efforts carried out to date, the compounds responsible for that intoxication have not been elucidated. Nevertheless, epidemiological studies have pointed to fatty acid mono- and diesters of 3-phenylamino-1,2-propanediol (PAP) as the biomarkers of those toxic oil batches. The structure of these esters bears common features with that of triglycerides, which suggested that PAP esters could follow the route of lipids metabolism up to a certain extent. The incubation of racemic PAP dioleyl ester with human pancreatic lipase (hPL) led to the formation of the corresponding stereoisomeric monoesters bearing the oleyl residue at C-2, although a kinetic resolution in favor of the (S)-enantiomer was observed. These monoesters are unstable and in equilibrium with their corresponding regioisomers with the acyl residue at C-1, apparently without the intervention of the lipase. Finally, incubations of these latter monoesters with hPL led to the formation of the respective PAP enantiomers. Again, the kinetic resolution of this hydrolytic process favored the formation of the enantiomer with the (S)-configuration. Taken together, these results showed that PAP esters are substrates of hPL and that the two hydrolytic steps exhibit kinetic resolution in favor of the (S)-enantiomers.

Introduction TOS1 was an outbreak of vast proportions caused by a food intoxication episode (1). The epidemic took place in 1981 in central and northern areas of Spain, caused over 400 deaths, and affected more than 20 000 people (2). It is estimated that 30-40% of this population continues to suffer mild symptoms or severe sequelae. TOS was characterized as a multisystemic disease, which resembled an allergic-toxic syndrome in the acute phase and an autoimmune condition in the chronic phase. A summary of the investigation status on this epidemic has been collected in recent reviews (3, 4). Epidemiological studies showed that the intoxication emerged from the ingestion of adulterated rapeseed oil purchased from street vendors. Rapeseed oil denatured with 2% aniline was imported for industrial purposes and illegally refined and distributed for human consumption. A strong association between TOS and ingestion of this oil was proven (5-7). Fatty acid anilides (2, 8) and fatty acid * To whom correspondence should be addressed. Tel: +34-93-400 61 21. Fax: +34-93-204 59 04. E-mail: [email protected]. 1 Abbreviations: TOS, toxic oil syndrome; OLA, oleanilide; PAP, 3-(phenylamino)propane-1,2-diol; mPAP, dPAP, and tPAP, mono-, di-, and triacyl derivatives of PAP; OO(1,2)PAP, dioleyl ester of 3-phenylaminopropane-1,2-diol; O(1)PAP, oleyl ester at C-1 of 3-(phenylamino)propane-1,2-diol; O(2)PAP, oleyl ester at C-2 of 3-(phenylamino)propane-1,2-diol; OO(N,1)PAP, bisoleyl derivative at C-1 and N of 3-(phenylamino)propane-1,2-diol; OOO(N,1,2)PAP, N,O,O-trisoleyl derivative of 3-(phenylamino)propane-1,2-diol; TEAP, triethylammonium phosphate; hPL, human pancreatic lipase.

esters of PAP (9, 10) have been identified as major contaminants of oil batches. However, the differences among the concentration of these major contaminants in fraudulent and toxic oil batches led the TOS Management Committee to consider fatty acid anilides as biomarkers of fraudulent oils, whereas PAP esters have been qualified as biomarkers of toxic oils. Moreover, the content of PAP esters in the oil has been associated with the morbidity that these oil batches caused in the corresponding households (9, 11). In particular, the dPAPs are considered to be the putative toxic substances generated during the refining process at the ITH company in Seville, the refinery where the toxic oil emerged (12). However, the toxicity mechanism of dPAP compounds in biological systems has not yet been elucidated. From the structural point of view, the aniline moiety of PAP esters shares common features with secondary arylamines. However, the fatty acid moiety of these derivatives confers a lipophilic characteristic for their distribution, and in this respect, PAP esters could be envisaged as diglyceride-like compounds (Chart 1). Moreover, glycerides and phospholipids are chiral compounds and chirality plays a crucial role in specific biological functions. Concerning PAP esters, the C-2 atom of the propane-1,2-diol moiety is also stereogenic. Therefore, the formation of two PAP enantiomers resulting from dPAP hydrolysis might be expected; nevertheless, the potential biological formation of these enantiomers has not yet been studied. In this sense, ingested OOPAP (the major

10.1021/tx049952z CCC: $27.50 © 2004 American Chemical Society Published on Web 05/28/2004

890

Chem. Res. Toxicol., Vol. 17, No. 7, 2004

Morato´ et al.

Chart 1. PAP Ester Structures as Compared with Those of Functional Glycerides

component of the PAP diesters mixture) could be a potential substrate for lipases at the intestinal lumen. This lipase activity would yield ultimately racemic PAP or a stereoisomeric excess of one enantiomer prone to be absorbed by the intestinal cell. In addition, nonhydrolyzed OOPAP or de novo-synthesized dPAP in the intestinal cell would potentially reach the general circulation where the lipoprotein lipase activity will potentially offer the racemate or an excess of one mPAP or PAP enantiomer to the endothelial cell. We recently investigated the in vivo biotransformation of PAP administered intraperitoneally to two mouse strains. PAP, a highly polar substance, undergoes aromatic and alkyl chain oxidations, and most metabolites are eliminated in urine. Some of these metabolites have potential toxicity as reported for other similar compounds (13, 14). Thus, the putative formation of these metabolites raised the question of the potential offer of PAP at the intestinal lumen as a consequence of dPAP and mPAP lipase-mediated hydrolyses. In this context, Ruiz-Gutie´rrez and Maestro-Duran reported the detection (TLC) of PAP esters in the lipid fraction obtained after oral administration of these compounds to rats; moreover, these authors were able to detect also the presence of mPAPs and PAP in samples obtained after the administration of dPAPs and mPAPs, respectively, which represented an indirect evidence of endogen hydrolytic processes on those substrates (15). More recently, Closa and co-workers observed, after oral administration to rats of olive oil solutions of radiolabeled PAP mono- and diesters from linoleic acid, that these substrates were absorbed in the gastrointestinal tract and stored in different organs, particularly in the liver and brown adipose tissue. Noticeably, PAP esters in these organs showed different patterns of fatty acids, suggesting the

ability of the gastrointestinal tract to modify the fatty acid composition of the parent PAP ester (16). With these antecedents, the present contribution reports our results on the hydrolysis of PAP esters mediated by hPL. The aims of the study were to ascertain whether these compounds are substrates for the enzyme hydrolytic activity and if the process takes place with any regioand stereoselectivity. The study was also extended to the mono- and diesterified N-acyl derivatives of PAP, compounds also detected in toxic oil batches although as minor contaminants. The answers to these questions may help to clarify the offer of ingested PAP-derived compounds into the body and their potential implication in the toxicity mechanisms involved in TOS.

Materials and Methods Caution: Because PAP derivatives have been implicated in TOS, precautions should be taken when handling these compounds (use of gloves and ventilated hood cabinets when handling solutions in organic solvents). Chemicals and Reagents. Aniline (Aldrich, 99%) and glycidol (Merck, 98%) were previously distilled and stored at -20 °C. (R)- and (S)-glycidol (96% ee) were obtained from Aldrich (U.S.A.). Oleic acid, hPL (EC 3.1.1.3), porcine pancreas colipase, and taurodeoxycholic acid were obtained from Sigma (St. Louis, MO). n-Hexane, methanol, isopropyl alcohol, dichloromethane, and acetonitrile were HPLC or analytical grade. Synthesis of Racemic and Stereoisomeric PAP Derivatives. Unless stated otherwise, organic solutions obtained from the treatment of crude reaction mixtures were dried over MgSO4. TLC analyses were performed on Merck Kieselgel 60 F254 plates (aluminum sheets, 0.2 mm thickness) using mixtures of hexanes-EtOAc as the eluents and were developed by UV irradiation at 254 nm. The purification of metabolites by flash chromatography was performed using a Biotage system (Charlottesville, U.S.A.) provided with 32-63 µm silicagel cartridges.

Lipase-Mediated Hydrolysis of PAP Esters Routine HPLC analyses were performed on a Hewlett-Packard 1100 system provided with a DAD detector. The purifications at the semipreparative scale were performed with a system consisting of two Waters Millipore 510 pumps and an Applied Biosystems 783 Programmable Absorbance Detector. The enantiomeric excess of the mixtures of PAP and OPAP was determined by HPLC employing an OD column (250 mm × 4.6 mm) (Daicel Chem. Ind., Japan). Chromatographic profiles were normally obtained by UV monitoring at 245 nm. The IR spectra were registered with a Nicolet model Avatar 360 FT-IR apparatus, and absorptions are given in cm-1. The 1H (300 MHz) and 13C (75 MHz) NMR spectra were registered with a Varian Unity 300 spectrometer. The spectra were taken in neutralized CDCl3 unless otherwise stated. Chemical shifts are given in ppm related to tetramethylsilane for 1H and deuteriochloroform for 13C as internal standards. ESI spectra were obtained with an HP1100 LC-MS system working in the positive ionization mode. Elemental microanalyses were carried out at the “Servei de Microana`lisi” of IIQAB using a 1108 Carlo Erba analyzer. Optical rotations ([R]25) were measured in a 1.0 dm tube with a Perkin-Elmer 241 MC polarimeter. PAP, OPAP, and OO(1,2)PAP derivatives were synthesized as described previously (17). Synthesis and Purification of Racemic O(2)PAP. The monoesterification of PAP to give O(1)PAP leads to the concomitant formation of O(2)PAP (17). Purification of this crude reaction mixture by TLC (4:1 hexane: EtOAc) gave a fraction containing a 73:27 O(2)PAP:O(1)PAP mixture (HPLC monitoring). The final purification of this mixture was carried out by HPLC using a C-18 Kromasil 100 column (25 cm × 1 cm, 5 µm) and an isocratic 90:10 MeOH:H2O eluent mixture at 4 mL/min. Collected eluates were extracted with hexane, concentrated, redissolved with benzene, and evaporated to render O(2)PAP (>95% purity). Alternatively, highly pure samples of this compound (98%) could be also obtained by using the Biotage flash chromatography equipment eluting with 6:1 n-hexane: EtOAc mixtures. In any case, pure samples of this compound, i.e., without any detectable amount of its O(1)PAP isomer, could only be conserved for few days in the absence of solvent and at -20 °C. 1H NMR: δ 7.19 (t, 2 H, J ) 7.8, H-3′, H-5′), 6.74 (t, 1 H, J ) 7.4, H-4′), 6.66 (d, 2 H, J ) 7.8, H-2′, H-6′), 5.34 (m, 2 H, CHd), 5.1 (m, 1 H, CH-O), 3.82 (d, 2 H, J ) 3.9, CH2-OH), 3.40 (d, 2 H, J ) 6, CH2-N), 2.36 (t, 2 H, J ) 7.5, CH2CO), 2.01 (m, 4 H, CH2-Cd), 1.64 (m, 2 H, CH2CH2CO), 1.40-1.20 (22 H, CH2), 0.88 (t, 3H, J ) 6.6, CH3). 13C NMR: δ 173.9 (CO), 147.6 (C-1 Ar), 130.0 (CHd), 129.6 (CHd), 129.3 (C-3, C-5 Ar), 117.9 (C-4 Ar), 112.9 (C-2, C-6 Ar), 73.2 (-CHOCOR, 62.9 (-CH2OH), 44.1 (CH2N), 34.3-22.6 (12 CH2), 14.1 (CH3). ESIMS: 432 (M + H+). Synthesis of PAP Enantiomers. 1. (R)-PAP. Following the procedure previously described for racemic PAP, (S)-glycidol (0.5 g, 6.7 mmol, Aldrich, 96% ee) was added dropwise to a solution of freshly distilled aniline (1.26 g, 13.4 mmol) in methanol (5 mL) maintained at 50 °C in an argon atmosphere. When the addition was completed, the mixture was stirred under reflux for 3 h. The solvent and excess aniline were then removed in a vacuum. Final purification by preparative HPLC (as described below) afforded 0.50 g of the pure compound as a white solid in 50% yield, 99% chemical purity, and 94% enantiomeric purity. [R]25 ) +1.6 (c ) 1.03 g/100 mL, MeOH). 2. (S)-PAP. This enantiomer was obtained in 62% yield, 98% chemical purity, and 95% enantiomeric purity as described above using (R)-glycidol as the starting material. [R]25 ) -1.56 (c ) 1.03 g/100 mL, MeOH). Purification of (R)- and (S)-PAP enantiomers was performed on a Waters Prep LC 4000 system using a Waters Prepack 500 column packed with derivatized silica VYDAC C18 (300 Å, 15-20 µm, 47 mm × 300 mm). The mobile phases used were TEAP (pH 6.6) (solvent A) and CH3CN-TEAP 3:2 (solvent B). The gradient elution consisted of an initial isocratic step (from 0 to 10 min) at a rate of 100 mL/min with 99% A, followed by a linear gradient to reach 21% B [from 10 to 100 min for the (R)-PAP and from 10 to 70 min for the (S)-PAP].

Chem. Res. Toxicol., Vol. 17, No. 7, 2004 891 Enantiomers of OPAP were independently synthesized following the procedure described previously for the racemate. In this case, (R)- and (S)-PAP were used as starting materials and oleic was employed in a stoichiometric ratio. The synthesis of OO(1,2)PAP enantiomers was performed as for the monoacyl derivatives, but using two molecular equivalents of oleic acid. Purification of OPAP and OO(1,2)PAP enantiomers was carried out by preparative TLC on silicagel. (R)-O(1)PAP was obtained in 48% yield and 93% enantiomeric purity using (R)-PAP as the starting material [R]25 ) -0.57 (c ) 1.03 g/100 mL, methanol). (S)-O(1)PAP was obtained in 52% yield and 95% enantiomeric purity using (S)-PAP as the starting material. [R]25 ) +0.51 (c ) 0.99 g/100 mL, methanol). (R)-OO(1,2)PAP was obtained in 75% yield and 94% enantiomeric purity using (R)-PAP as the starting material. [R]25 ) -0.33 (c ) 1.16 g/100 mL, hexane). (S)-OO(1,2)PAP was obtained in 77% yield and 94.5% enantiomeric purity using (S)-PAP as the starting material [R]25 ) +0.48 (c ) 1.22 g/100 mL, hexane). Synthesis and Purification of Racemic OO(N,1)PAP. This N-acyl derivative was prepared from rac-O(1)PAP. Briefly, a solution of O(1)PAP (20 mg, 0.046 mmol), oleyl chloride (16.7 mg, 0.055 mmol), and triethylamine (7.7 µL, 0.056 mmol) in 1 mL of benzene was stirred for 3 h at 25 °C (HPLC monitoring). The solvent was evaporated, and the crude reaction mixture was redissolved with n-hexane and purified by flash chromatography using 5:1 n-hexane:EtOAc eluent mixture to give 22.8 mg (71% yield) of OO(N,1)PAP. IR (neat): ν 3100-3500, 1738, 1641. 1H NMR: δ 7.40 (m, 3 H, HAr), 7.20 (d, 2 H, J ) 7.6, HAr), 5.34 (m, 4 H, CH)), 4.22-4.00 (5 H, CH2N, CHOH, CH2OCO), 3.54 (d, 1 H, J ) 12.0, CHOH), 2.40-1.90 (12 H, CH2CO, CH2CHd), 1.53 (m, 4 H, CH2CH2CO), 1.40-1.10 (44 H, CH2), 0.87 (t, 6 H, J ) 6.8, CH3). 13C NMR: δ 176.2 (CO), 173.7 (CO), 143.1 (C-1 Ar), 130.0 (CHd), 129.9 (CHd), 129.7 (C-3, C-5 Ar), 128.2 (C-4 Ar), 127.8 (C-2, C-6 Ar), 69.4 (CHOH), 65.9 (CH2OCO), 54.1 (CH2N), 34.3 (CH2CO), 34.0 (CH2CO), 31.9 (CH2), 29.7 (CH2), 29.7 (CH2), 29.5 (CH2), 29.3 (CH2), 29.1 (CH2), 29.0 (CH2), 27.2 (CH2-Cd), 27.1 (CH2-Cd), 25.3 (CH2), 24.7 (CH2), 22.7 (CH2CH2CO), 14.1 (CH3). ESI-MS: 696.5 (M + H+). Elemental analysis for C45H76NO4: C, 77.75%; H, 11.02%; N, 2.01%. Found: C, 77.54%; H, 11.39%; N, 1.99%. Synthesis of Racemic OOO(N,1,2)PAP. This triacyl derivative was prepared following the procedure previously described (17) with slight modifications. Thus, the final product was obtained from the oleyl acylation of OO(1,2)PAP, purified by flash chromatography eluting with 10:1 n-hexane:EtOAc, and isolated in nearly quantitative yield. Lipase Incubations. Incubations with hPL were performed according to the procedure reported in the literature (18, 19). Incubates contained 100 mM tris-HCl, pH 8.0, 150 mM NaCl, 4 mM taurodeoxycholic acid, and colipase. The substrate concentration used was 1 mM; enzyme units used are indicated in each experiment, and they were made from a stock solution of 651 units/mL of the lipase. Colipase was added together with lipase at a ratio of 1.3 µg/lipase unit (corresponding to a 2.14 molar excess). OO(1,2)PAP and OPAP emulsion solutions were prepared from standard stocks (20 mM) dissolved in 1:1 methanol:chloroform [except for O(2)PAP, which was dissolved in n-hexane for preserving at maximum its purity]. All stock solutions were stored at -20 °C, except that of O(2)PAP, which was stored at -80 °C. The appropriate standard concentrations were dispensed into glass tubes together and dried at 45 °C under an N2 stream followed by drying under vacuum for 10 min. The substrate film was rehydrated by adding buffer solution to the tubes followed by vigorous vortexing and bath sonication for 1 min. Finally, colipase and lipase enzyme were added and the tubes were placed in a Carrousel Reaction Station (Radleys Discovery Technologies, Great Britain) at 37 °C. The hydrolysis reaction was stopped with a mixture of dichloromethane, isopropyl alcohol, and ammonia (9:1:0.1). The organic phase was collected, and the aqueous phase was extracted again. The solvents were evaporated, and then, the sample was redissolved with a 9:1 mixture of n-hexane and

892

Chem. Res. Toxicol., Vol. 17, No. 7, 2004

isopropyl alcohol and stored at 4 °C, except that of O(2)PAP, which was stored at -20 °C. All of the incubations were carried out by triplicate. Each substrate was checked in a blank assay. HPLC Analyses of Lipase Incubations. Hydrolysis products from the enzyme incubations were analyzed by HPLC using the chiral column to determine the enantiomeric excess of substrates and products. The mobile phase consisted of mixtures of n-hexane (eluent A) and isopropyl alcohol (eluent B) at 1 mL/ min. To analyze the different incubation products, two elution conditions were used. Normally, all reaction mixtures were first analyzed using an isocratic mixture of 20% B and 80% A for 18 min (elution program I); under these conditions, OO(1,2)PAP, O(2)PAP, O(1)PAP, (S)-PAP, and (R)-PAP eluted at 5.1, 6.6, 7.7, 11.0, and 14.2 min, respectively. The samples were subjected to a second analysis with the same mobile phases and gradient elution conditions (elution program II) consisting of an initial step at 7% B for 25 min, then a linear ramp from 7 to 20% B from minutes 25 to 32, and a return to the initial conditions from minutes 32 to 42. Under these conditions, (S)-O(2)PAP, (R)-O(2)PAP + (S)-O(1)PAP, and (R)-O(1)PAP eluted at 16.4, 17.9, and 19.7 min, respectively. By combining the profiles obtained from these elution conditions, relative areas for O(1)PAP and O(2)PAP enantiomers could be estimated. Unfortunately, all efforts to resolve the mixture of OO(1,2)PAP enantiomers during lipase incubations were unsuccessful.

Results and Discussion Synthesis of Standards. The synthesis of (R)- and (S)-PAP was carried out by the addition of aniline to commercially available (S)- or (R)-glycidol, respectively, in methanol solution, followed by purification by preparative HPLC. The mono- or diesterification of each PAP enantiomer with oleic acid led to the formation of the corresponding O(1)PAP and OO(1,2)PAP stereoisomers in good yields and chemical and optical purities, after the respective purification by preparative TLC. The monoesterification of PAP led to the formation of O(2)PAP and OO(1,2)PAP as minor components of the crude reaction mixture. All efforts carried out to purify O(2)PAP by TLC were unsuccessful due to its partial isomerization to give O(1)PAP during the separation process. Alternatively, highly pure samples of O(2)PAP could be obtained by rapid flash chromatography on silicagel cartridges or by reverse phase HPLC at the semipreparative scale. Despite the coincidence of most NMR absorptions for both O(1)PAP and O(2)PAP, the shift of the multiplet assigned to the proton at C-2 in the 1H NMR spectrum from 4.05 ppm in O(1)PAP to 5.1 ppm and the doublets of doublets at 3.82 and 3.40 ppm corresponding to the CH2N and CH2O hydrogen atoms confirmed the structure assignation for O(2)PAP. The corresponding shifts were also observed for the absorptions attributed to the -CH2OH and the -CHOCOR in the 13C NMR spectrum. The stability of O(2)PAP in 1 mM solution in two solvents of different polarity, i.e., n-hexane and methanol, was checked after 20 days at 20 °C to define the manipulation conditions in the lipase assays. The dry product stored at -20 °C was used as a reference. The HPLC analysis of the hexane solution afforded a 84:14:2 O(2)PAP:O(1)PAP:PAP ratio, whereas the composition in the methanol solution was 17:52:30, respectively. The reference O(2)PAP sample afforded an 11% of O(1)PAP. As expected, the acyl migration rate correlated with the increasing polarity of the solvent. Under the lipase assay buffer conditions (pH 8, 37 °C), this acyl migration took place in a few minutes; conversely, both O(1)PAP and

Morato´ et al. Chart 2. Structures of OO(1,N)PAP and OOO(1,2,N)PAP

OO(1,2)PAP were stable under the same conditions. These results evidenced the difficulty of handling O(2)PAP and the convenience of performing the incubations with freshly purified samples. The fact that similar transacylation reactions have been commonly observed for diglycerides, particularly when acid catalysis is present (20), supports the structural analogy proposed for PAP derivatives with respect to glycerolipids (cf. Chart 1). The availability of the N-oleyl derivatives of PAP esters, namely, OO(N,1)PAP and OOO(N,1,2)PAP (Chart 2), made possible the study of their stability under the lipase assay conditions. N-Acyl derivatives such as OOO(N,1,2)PAP have been identified in the toxic oil batches (21, 22). As occurred with O(2)PAP, these N-acylated derivatives were unstable in buffer solution and the loss of the N-acyl residue was mainly observed. Thus, OO(N,1)PAP afforded O(1)PAP as a major product (10%, HPLC monitoring) after 30 min at 37 °C. When the same experiment was conducted in the presence of hPL, no N-acylated products other than the starting substrate were observed. On the other hand, the triacyl derivative OOO(N,1,2)PAP gave 20% of O(1)PAP in the absence of hPL, although the reproducibility of these percentages was poor (for representative HPLC profiles, see the Supporting Information). The reactivity shown by the N-acyl substituent in the above compounds could also be envisaged as an acyl migration process. The explanation for the preference of N-acyl to O-acyl migration in these derivatives as well as the presence of O(1)PAP in the assays carried out with the triacyl derivative is currently under investigation. In any case, these results led us to discard the above N-acyl derivatives as substrates for the lipase assays and presuppose that they would in fact increase the offer of PAP esters in the intestinal tract. Assays with hPL. Despite the availability of the enantiomers of PAP and with the exception of the first experiment with OO(1,2)PAP (vide infra), we decided to perform the hPL-mediated hydrolysis assays with the corresponding racemates. In this way, both enantiomers will be present at the same time in the incubation medium and the potential kinetic resolution promoted by the lipase could be studied more accurately. Two HPLC analytical methods using a chiral column were set up to quantify the formation of OPAP and PAP enantiomers. Incubations with hPL were double-checked with two different elution conditions, and the enantiomeric excess of S/R PAP and OPAPs was estimated by combining the profiles from both analyses. Unfortunately, OO(1,2)PAP enantiomers could not be satisfactorily resolved. Initial experiments were performed seeking potential hydrolysis of OO(1,2)PAP (racemate and isolated stereoisomers) with hPL. To that aim, factors such as concentrations of taurodeoxycholic acid and porcine pancre-

Lipase-Mediated Hydrolysis of PAP Esters

Chem. Res. Toxicol., Vol. 17, No. 7, 2004 893

Figure 1. Hydrolysis by hPL: HPLC traces using the chiral OD column corresponding to the incubation of (S)-OO(1,2)PAP (1 mM) (I) or OO(1,2)PAP (1 mM) (II), in the presence of hPL (0.65 units/mL) at pH 8.0, for 3 h at 37 °C. Elution conditions: program I (see Materials and Methods).

atic colipase and the incubation conditions were optimized (see the Materials and Methods).2 Thus, the HPLC analysis of the crude reaction mixtures resulting from the incubations of OO(1,2)PAP and of its respective stereoisomers indicated, as expected, that the hydrolytic pathways were stereoselective. As an illustrative example, Figure 1 shows the chromatographic profiles obtained for (S)-OO(1,2)PAP and the racemic mixture. The time course of the reaction of the separate stereoisomers with hPL is shown in Figure 2. In both cases, the initial formation of O(2)PAP was observed, followed by the appearance of the O(1)PAP regioisomers and finally of PAP. The enzyme showed a faster hydrolysis rate for the (S)-OO(1,2)PAP isomer (vide infra). The (S)O(2)PAP concentration reached a maximum after 3 h and then decreased until the end of the experiment (24 h). Likewise, maximum (S)-O(1)PAP contents were reached at 3 h and then underwent a faster decrease than that observed for its regioisomer. This result suggested that the hydrolysis of the monoester is faster than that of the initial diester. This assumption was confirmed by comparative time course hydrolysis experiments using OO(1,2)PAP and O(1)PAP (Supporting Information). Finally, the concentration of (S)-PAP, the ultimate hydrolysis product, increased during the whole experiment. In contrast, the relative amounts of (R)-O(2)PAP reached the maximum after 6 h of incubation. Nevertheless, similar processes to those described above took place in this case to give, after 24 h of incubation, comparable amounts of (R)-PAP to those obtained for the (S)enantiomer. These preliminary results showed that PAP mono- and diesters were substrates of the hPL and that the enzyme exhibited a degree of stereoselectivity favoring the hydrolysis of compounds bearing the (S)-configuration. To study the different steps of the process more carefully and confirm the above results, we decided to run incuba2 Commercial hPL batches showed an important loss of activity on storage, which advised us to perform all assays with a given substrate in the same experiment. Indeed, the day-to-day reproducibility of the enzymatic activity was low, and it was necessary to adjust the amount of enzyme used in each experiment to achieve the desired level of activity [Lykidis, A., Mougios, V., and Arzoglou, P. (1995) Eur. J. Biochem. 230, 892-898]. For this reason, the reported amounts of hPL used reflect only the theoretical units of enzyme based on the activity stated in the label by the seller. We did not routinely perform any assays to check the actual specific enzyme activity.

Figure 2. Time course hydrolysis of (S)-OO(1,2)PAP (1 mM) (I) and (R)-OO(1,2)PAP (1 mM) (II) by hPL (0.65 units/mL) at 37 °C.

Figure 3. Time course hydrolysis of OO(1,2)PAP (I) and time course evolution of the enantiomeric excess for (S)-O(2)PAP and (S)-PAP (II) in the presence of hPL (1.3 units/mL) at 37 °C. For the estimation of the ee values of PAP derivatives by HPLC using a chiral OD column, see the Materials and Methods.

tions at shorter reaction times, using different hPL concentrations2 and working with the respective racemic mixtures of the different PAP ester substrates. The results obtained from the incubation of OO(1,2)PAP with hPL are shown in Figure 3. A representative HPLC profile of an aliquot taken after 2.5 h of incubation is enclosed as Supporting Information. As shown in Figure 3 (lower panel), after 0.5 h of incubation, the (S)-

894

Chem. Res. Toxicol., Vol. 17, No. 7, 2004

Morato´ et al.

Figure 5. Time course biotransformation of O(2)PAP in the presence or absence of hPL (4 units/mL) at 37 °C. Table 1. Apparent Rate Constants Determined by Fitting the Experimental Data Obtained in the Incubations of O(2)PAPa with hPL (4 and 8 u mL-1) to the Kinetic Model Depicted

Figure 4. Time course hydrolysis of O(1)PAP (I) and time course evolution of the enantiomeric excess for (S)-PAP (II) in the presence of hPL (0.05 units/mL) at 37 °C. For the estimation of the ee values of (S)-PAP by HPLC using a chiral OD column, see the Materials and Methods.

k1 k2 k3 a

O(2)PAP was formed in a 54% ee, thus confirming that the hPL-promoted hydrolysis of PAP diester is stereoselective in favor of the (S)-enantiomer. Moreover, although no PAP was present after 0.5 h of incubation, at 1 h, a 100% ee of (S)-PAP was detected. This result suggested also that the second hydrolytic process, i.e., from O(1)PAP to PAP, was also stereoselective for the (S)-enantiomer. This maximum ee value for (S)-PAP was maintained after 1.5 h of incubation; then, a constant decrease of this ee value was observed indicating that the overall hydrolytic process was also operating for the family of PAP esters with an (R)-configuration. An analogous study was undertaken for the case of O(1)PAP, and the results obtained are shown in Figure 4. In this case, a 23% ee value was the maximum reached by the (S)-PAP, and it was obtained after 25 min of incubation (lower panel). These results are in agreement with the higher rates of hPL-promoted hydrolysis observed for PAP monoesters in comparison with PAP diesters. In any case, in this last hydrolytic step, again a lipase preference for the member of the family with (S)configuration was maintained. A final question was to ascertain whether the intermediate step of the whole process, that is the acyl migration equilibrium between O(2)PAP and O(1)PAP, could be influenced by the lipase activity. To that aim, comparative incubations were carried out with O(2)PAP (98% purity) in the presence and absence of hPL and the results obtained are shown in Figure 5. As expected, the reactions carried out in the presence of hPL showed the disappearance of O(2)PAP with the concomitant formation of O(1)PAP and PAP. In the absence of hPL, the relative proportion of O(2)PAP also decreased with time, although at a lower rate than in the presence of the enzyme. In this case, only the formation of O(1)PAP could be detected, which suggested that the acyl migration is a process not catalyzed by the lipase. Assays were carried out at two different concentrations of hPL (4 and 8 units/

4 u mL-1

8 u mL-1

0.31 ( 0.17 2.2 ( 1.3 0.033 ( 0.002

0.22 ( 0.10 1.7 ( 0.8 0.061 ( 0.004

For additional information, see the Supporting Information.

mL) to check the dependence of the reaction rates on the amount of enzyme present. Assuming a simple reaction model as that depicted in Table 1, it was possible to fit the experimental data [relative concentrations (HPLC) of O(2)PAP, O(1)PAP and PAP vs time] to extract the apparent rate constants k1, k2, and k3. The value of k3 approximately doubles when the amount of enzyme present in the reaction mixture is doubled, while the values of k1 and k2 only suffer small variations within the experimental error, suggesting that the acyl migration equilibrium between O(2)PAP and O(1)PAP is mainly independent of the lipase under the reaction conditions used. This interpretation should be taken with care as the difference between the rates for the (R)- and (S)-isomers of each compound is not considered. In this respect, all attempts to measure the relative concentrations of those stereoisomers were unsuccessful due the instability of OPAP monoesters. In conclusion, the major findings observed in this study were that PAP esters are substrates of the hPL and that the hydrolysis of OO(1,2)PAP would generate preferentially (S)-PAP. Independently of the low content of PAP monoesters and particularly of PAP in toxic oil batches (9), the body would face an enantiomeric offer of these compounds in the intestinal lumen as a consequence of the lipase activity. Under those conditions, (S)-PAP may be absorbed and further metabolized in the liver as a highly polar substance, as we reported recently in mice (13). On the other hand, more lipophilic species such as (R)-O(1,2)PAP and (S)-O(2)PAP could be offered to the endothelial cell and may undergo an unknown metabolism such as oxidative biotransformations at the aniline moiety or at C-1 or the lipase-mediated hydrolytic cascade. Moreover, the resynthesis to form de novo PAP diesters cannot be discarded. In this last case, the resulting lipophilic compounds would reach the general circulation through lipidic absorption, possibly incorpo-

Lipase-Mediated Hydrolysis of PAP Esters

rated into lipid particles through the thoracic duct. Therefore, it would be interesting to find out whether these PAP esters could also be substrates of the lipoprotein lipase and/or endothelial lipase located in the vessel wall. It should be reiterated that the most prominent pathological feature of patients affected by TOS was a nonnecrotising vasculitis involving vessels of every type and size (3). On the other hand, the stereochemical scenario discussed above could be influenced by the initial composition of the fraudulent rapeseed oil. Preliminary assays conducted to find out whether an enantiomeric excess of one OO(1,2)PAP stereoisomer was present in a model toxic oil developed in our laboratory did not show any stereochemical difference (results not shown). However, this result cannot be taken as conclusive for all of the PAP diester components of the toxic oil batches. Actually, in parallel studies currently in progress in our laboratory, conclusive evidences on the generation of PAP diesters from the glyceride precursors present in the oils have been obtained. Thus, if OO(1,2)PAP was mainly generated from the triolein present in the rapeseed oil, the occurrence of this diester in the racemic form could be expected. However, the formation of a PAP diester from a nonsymmetric triglyceride could lead to specific diesters with slight deviations in their stereochemical composition. In any case, PAP diester intestinal biotransformation may be a crucial factor to be taken into consideration regarding the generation and distribution of further toxic metabolites after intestinal lipase hydrolysis. Further studies on the absorption and transport of all of these species and on their putative interaction with other lipases present in the body are needed to evaluate the extent of the proposed glyceride resemblance of PAP ester and the ultimate toxicological relevance of these findings.

Acknowledgment. Financial support from the WHO/ TOS Committee is gratefully acknowledged. Supporting Information Available: Representation of comparative hydrolysis rates for OO(1,2)PAP and O(1)PAP and HPLC profiles using a chiral column of crude incubation mixtures from OO(1,2)PAP and OO(1,N)PAP under two eluent conditions. Two figures showing the results of the fitting of the experimental data obtained in the incubations of O(2)PAP with hPL. This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Tabuenca, J. M. (1981) Toxic-allergic syndrome caused by ingestion of rapeseed oil denatured with aniline. Lancet 2, 567-568. (2) Grandjean, E. C. (1992) Toxic oil syndrome. Current knowledge and future perspectives. Introduction. WHO Reg. Publ. Eur. Ser. 42, 1-4. (3) Abaitua-Borda, I., and Posada de la Paz, M. (1992) Clinical findings. Toxic oil syndrome: Current knowledge and future perspectives. WHO Reg. Eur. Ser. 42 (World Health Organisation Regional Office for Europe, Copenhagen, Denmark), 26-36. (4) Yoshida, S. H., German, J. B., Fletcher, M. P., and Gershwin, M. E. (1994) The toxic oil syndrome: a perspective on immunotoxicological mechanisms. Regul. Toxicol. Pharmacol. 19, 60-79. (5) Can˜as, R., and Kilbourne, E. M. (1987) Oil ingestion and the toxic oil syndrome. Results of a survey of residents of the Orcasur neighbourhood in Madrid Spain. Int. J. Epidemiol. 16, 3-6.

Chem. Res. Toxicol., Vol. 17, No. 7, 2004 895 (6) Diaz de Rojas, F., Castro-Garcı´a, M., Abaitua-Borda, I., AlonsoGordo, J. M., Posada de la Paz, M., Kilbourne, E. M., and Tabuenca-Oliver, J. M. (1987) The association of oil ingestion with toxic oil syndrome in two convents. Am. J. Epidemiol. 125, 907911. (7) Rigau-Perez, J. G., Pe´rez-Alvarez, L., Duen˜as-Castro, S., Choi, K., Thacker, S. B., Germain, J. L., Gonza´lez-de-Andre´s, G., Canada-Royo, L., Pe´rez-Gallardo, F. (1984) Epidemiologic investigation of an oil-associated pneumonic paralytic eosinophilic syndrome in Spain. Am. J. Epidemiol. 119, 250-260. (8) Bernert, J. T., Kilbourne, J. E. M., Akins, J. R., Paz, M. P. d. l., Meredith, N. K., Borda, I. A., and Wages, S. (1987) Compositional analysis of oil samples implicated in the Spanish toxic oil syndrome. J. Food Sci. 52, 1562-1569. (9) Hill, R. H., Schurz, H. H., Posada de la Paz, M., Abaitua-Borda, I., Philen, R. M., Kilbourne, E. M., Head, S. L., Bailey, S. L., Driskell, W. J., Barr, J. R., Needham, L. L. (1995) Possible etiologic agents for toxic oil syndrome: fatty acid esters of 3-(Nphenylamino)-1,2-propanediol. Arch. Environ. Contam. Toxicol. 28, 259-264. (10) Va´zquez-Roncero, A., Janer del Valle, C., Maestro-Duran, R., and Graciane-Constante, E. (1983) New aniline derivatives in cooking oils associated with the toxic oil syndrome. Lancet ii, 10241025. (11) Posada de la Paz, M., Philen, R. M., Schurz, H. H., Hill, R. H., Gimenez-Ribota, O., Go´mez de la Ca´mara, A., Kilbourne, E. M., and Abaitua-Borda, I. (1999) Epidemiologic evidence for a new class of compounds associated with toxic oil syndrome. Epidemiology 10, 130-134. (12) Posada de la Paz, M., Philen, R. M., Borda, I. A., Sicilia Socias, J. M., Gomez de la Camara, A., and Kilbourne, E. M. (1996) Toxic oil syndrome: traceback of the toxic oil and evidence for a point source epidemic. Food Chem. Toxicol. 34, 251-257. (13) Bujons, J., Ladona, M. G., Messeguer, A., Morato´, A., and Ampurdane´s, C. (2001) Metabolism of (R)- and (S)-3-(phenylamino)propane-1,2-diol in C57BL/&- and A/J-strain mice. Identification of new metabolites with toxicological significance to the toxic oil syndrome. Chem. Res. Toxicol. 14, 1097-1106. (14) Ladona, M. G., Bujons, J., Messeguer, A., Ampurdane´s, C., Morato´, A., and Corbella, J. (1999) Biotransformation and clearance of 3-(phenylamino)propane-1,2-diol, a compound present in samples related in samples related to toxic oil syndrome, in C57BL/6 and A/J mice. Chem. Res. Toxicol. 12, 1127-1137. (15) Ruiz-Gutie´rrez, V., and Maestro-Dura´n, R. (1992) Lymphatic absorption of 3-phenylamino-1,2-propanediol and its esters. Exp. Toxicol. Pathol. 44, 29-33. (16) Closa, D. F., Calaf, E., Abia´n, R., Rosello´-Catafau, J., and Gelpı´, E. m. J. (2001) Absortion and effects of 3-(N-phenylamino)-1,2propanediol esters in relation to toxic oil syndrome. Lipids 36, 1125-1133. (17) Ferrer, M., Galceran, M., Sanchez-Baeza, F., Casas, J., and Messeguer, A. (1993) Synthesis of aniline derivatives with potential toxicological implications to the Spanish toxic oil syndrome. Liebigs Ann. Chem. 507-511. (18) Bezzine, S., Ferrato, F., Ivanova, M. G., Lopez, V., Verger, R., and Carrie`re, F. (1999) Human pancreatic lipase: colipase dependence and interfacial binding of lid domain mutants. Biochemistry 38, 5499-5510. (19) Carrie`re, F., Rogalska, E., Cudrey, C., Ferrato, F., Laugier, R., and Verger, R. (1997) In vivo an in vitro studies on the stereoselective hydrolysis of tri- and diglycerides by gastric and pancreatic lipases. Bioorg. Med. Chem. 5, 429-435. (20) Millqvist, A., Adlercreutz, P., and Mattiasson, B. (1994) Lipasecatalyzed alcoholysis of triglycerides for the preparation of 2-monoglycerides. Enzyme Microb. Technol. 16, 1042-1047. (21) Calaf, R. E., Pen˜a, J., Paytubi, S., Blount, B. C., Posada de la Paz, M., Gelpı´, E., and Abian, J. (2001) Determination of aniline derivatives in oils related to the toxic oil syndrome by atmospheric pressure ionization-tandem mass spectrometry. Anal. Chem. 16, 3828-3837. (22) Schurz, H. H., Hill, R. H., Paz, M. P. d. l., Philen, R. M., Borda, I. A., Bailey, S. L., and Needham, L. L. (1996) Products of aniline and triglycerides in oil samples associated with the toxic oil syndrome. Chem. Res. Toxicol. 9, 1001-1006.

TX049952Z