ARCHIVES
Vol.
OF BIOCHEMISTRY
294, No. 2, May
AND
BIOPHYSICS
1, pp. 475-481,
1992
Characterization of a Cytochrome P450 from Di(2ethylhexyl) Phthalate-Treated Rats Which Hydroxylates Fatty Acids Richard T. Okital and Janice Rice Okit-a Department Received
of
Pharmaceutical Sciences, College of Pharmacy, Washington State University,
September
30, 1991, and in revised
form
January
6,1992
A cytochrome P450 was purified from liver microsomes of rats treated with di(2-ethylhexyl) phthalate (DEHP). DEHP is a member of a group of structurally diverse compounds which have been classified as peroxisome proliferators and are inducers of cytochromes P450 which hydroxylate lauric acid and other fatty acids. The P450 isolated from DEHP-treated rats (P4500zHp) was observed to have a &f, value of 51 kDa by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and a maximum absorbance of 452 nm in its reduced carbon monoxide bound state. The amino terminal residue for P4500sar was alanine and an l&amino acid segment at the N-terminal region was identified. The N-terminal amino acid for the P450 4A1 from clofibrate-treated rats is methionine and alignment of the N-terminal segment of the P4500sar with P450 4Al indicated that the first four amino acids were absent. There were two amino acid differences between the two P45Os in this l&amino acid segment; in P45Ousar an alanine and a phenylalanine were substituted for serines in P450 4A1. The was found to catalyze the hydroxylation of P450DEHP several saturated fatty acids, having the highest turnover activity with laurate (82.1 nmol12-OH-laurate formed/ min/nmol P450). My&state, palmitate, and stearate were also metabolized but at decreasing rates. Cytochrome bs stimulated laurate 12-hydroxylation lo-fold in a reconstituted system. Laurate was not metabolized at its llcarbon atom; however, the longer chain length fatty acids were metabolized at the (w-l)-carbon atom in addition to the o-carbon atom. A polyclonal antibody to the P4500sHp recognized three protein bands in liver microsomes from control and DEHP-treated rats on Western blot analysis, but only two protein bands from phenobarbital-treated rats. 0 1992 Academic Pm. Inc.
i To whom correspondence should macy, Washington State University, 99164-6510. 0003-9661/92 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
be addressed 105 Wegner
Pullman, Washington 99164
at College of PharHall, Pullman, WA
Fatty acid hydroxylation was one of the first reactions used to characterize the cytochrome P450 system in liver and kidney microsomes and was the first reaction to be examined in a reconstituted P450 system (l-4). The 12carbon fatty acid, dodecanoic or lauric acid, has been used extensively to characterize the fatty acid hydroxylation system (1). There has been renewed interest in this P450mediated reaction because of its induction by various peroxisome proliferators, including clofibrate and diethylhexyl phthalate (DEHP)2 (5, 6). Orton and Parker (7) were the first to report the inductive effects of hypolipidemic agents on this P450-mediated reaction and several groups have characterized the microsomal and purified P450 from clofibrate-treated rats (8-17). P45Os which catalyze the hydroxylation of fatty acids at the w-terminus are designated as belonging to P450 subfamily 4A3 and at least three genes coding for P45Os in liver and kidney exhibit increased expression when rats are administered clofibrate (18). Gonzalez’s laboratory has characterized the genomic DNA for CYP4A13 and 4A2 (18) and has expressed P45Os 4Al and 4A3 in a vaccinia virus system (19). In control rats, the hydroxylation of lauric acid occurs at both the ll- and the 12-carbon atoms in an approximately 1:l ratio (14, 20). Treatment with phenobarbital induces the 11-hydroxylase 2-fold with only minor effect on the 12-hydroxylase (21), but treatment of rats with clofibrate induces the microsomal laurate 12-hydroxylase approximately lo-fold with only a 1.9-fold induction in the 11-hydroxylase (22). P45Os induced by peroxisome proliferators also hydroxylate longer, unsaturated fatty ’ Abbreviations used: Chaps, 3-[ (3-cholamidopropyl)dimethylammonio] propanesulfonate; DEHP, di(2-ethylhexyl) phthalate; ICDH, isocitrate dehydrogenase; P450, cytochrome P450; 11-OH-Laurate, ll- or (w-l)hydroxylaurate; 12-OH-Laurate, 12- or w-hydroxylaurate; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis. 3 The nomenclature of the cytochrome P450 system which catalyzes fatty acid hydroxylation is taken from “The P450 superfamily: Update on new sequences, gene mapping and recommended nomenclature” published in DNA lO, l-14,199l. This cytochrome P450 has also been referred to as P450,,, or cytochrome P-452. 415
Inc. reserved.
476
OKITA
AND
acids (23, 24). Capdevila et al. (24) reported an B-fold increase in arachidonic acid 20-hydroxylation in microsomes of ciprofibrate-treated rats which changed the o/ (w-1) hydroxy product ratio from 0.5 in control rats to 7.5 in the ciprofibrate-treated rats. Okita and Chance (20), Lake et al. (25,26), Mitchell et al. (27), and Elcombe and Mitchell (28) have also described the induction of laurate w-hydroxylase activity by another peroxisome proliferator, DEHP, a common plasticizer used in the manufacture of polyvinyl chloride. The P450 induced by DEHP was demonstrated by Hardwick et al. (12) to be immunochemically similar to the P450 induced by clofibrate. This P450 form represents a minor amount of the total P45Os in control livers but increases to 16-30% of the total P450 content following DEHP or clofibrate administration (10, 15). In the present study, we report the characterization of a P450 isolated from liver microsomes of DEHP-treated rats. MATERIALS
AND
METHODS
Materials. Male Sprague-Dawley rats were purchased from Sasco (Omaha, NE). Purina rat chow supplemented with 2% (w/w) DEHP was purchased from Ralston Purina (Richmond, IN). DEHP was obtained from Aldrich Chemicals (Milwaukee, WI). Sodium laurate, sodium myristate, pahnitic acid, stearic acid, isocitrate, ICDH, DEAE-Sephacel, deoxycholate, dilauroyl phosphatidylcholine, sodium phenobarbital, and 16-hydroxypalmitic acid were purchased from Sigma (St. Louis, MO). Chaps was purchased from Pierce Chemicals (Rockford, IL) and Emulgen 913 was donated by Kao Atlas (Japan). The l-‘V-labeled lauric, myristic, palmitic, and stearic acids were purchased from Amersham (Arlington Heights, IL). Methanol, ethyl acetate, and acetonitrile were obtained from Burdick and Jackson (Muskegon, MI). The 2’,5’-ADP-Sepharose 4B and AH-Sepharose 4B resins were purchased from Pharmacia (Piscataway, NJ). Bio-Gel HTP hydroxylapatite, SDS-PAGE chemicals, and the Biosil ODS-5 (15 X 0.4 cm) HPLC and microguard columns were purchased from Bio-Rad (Richmond, CA). All chemicals used in the purification procedures and assays were the highest grade available. Preparation of P450DEHp Microsomes were initially prepared from 170- to 200-g Sprague-Dawley rats which were fed a normal Purina rat chow diet for 1 week prior to starting them on a 2-week diet containing 2% (w/w) DEHP. However, when this diet was no longer commercially available, rats were given DEHP by intubation at a dose of 1200 mg/ kg per day for 3 days in peroxide-free corn oil. Induction of the laurate hydroxylation reactions was very similar by the two dosing procedures. Phenobarbital was dissolved in 0.9% saline and administered to rats intraperitoneally at a dose of 80 mg/kg for 4 days. Food was withdrawn 15 to 18 h prior to sacrificing the rats. Liver microsomes were prepared by the procedure of Remmer et al. (29). Microsomes (10 mg/ml) were solubilized with 1.8% (w/v) sodium cholate for 30 min, centrifuged to remove nonsolubilized material, and diluted to 0.7% (w/v) sodium cholate. The cytochrome P450 from DEHP-treated rats that hydroxylated lauric acid was isolated by a modification of the procedure of Tamburini et al. (9) using the octylamino-Sepharose 4B procedure of Guengerich and Martin (30) to initially separate the P450 fraction from cytochrome br,, NADPH-cytochrome P450 reductase, and other proteins and lipids. In the present purification scheme, the order of the hydroxylapatite and DEAE-Sephacel column procedures was reversed and a lauroyl-Sepharose column was added as the final step in the Tamburini et al. (9) procedure to aid in the purification of the P450 and in removal of the nonionic detergent. The procedure of Gibson and Schenkman (31) was used to prepare the lauroyl-Sepharose column from AH-Sepharose 4B. The lauroyl-Sepharose column (15 X 1.5 cm) was equilibrated in 50 mM potassium phosphate buffer (pH 7.25), 20% glycerol, and 0.2 M NaCl.
OKITA The P450 fraction obtained from the hydroxylapatite column was applied to the lauroyl-Sepharose column and was extensively washed to remove the nonionic detergent, Emulgen 913, which was determined by the decrease in Am readings until constant values were obtained. A Chaps detergent gradient from 0 to 20 mM in 50 mM potassium phosphate buffer (pH 7.25), 20% glycerol, and 0.2 M NaCl (50 ml of each gradient buffer) was applied to elute the bound P450. Approximately 0.5-ml fractions were collected from the lauroyl-Sepharose column and AdI readings were measured. In two of three purification attempts, the P450 fraction applied to the lauroyl-Sepharose column separated into two protein peaks; designated as fractions I and II. The two fractions were dialyzed extensively against 50 mM potassium phosphate buffer (pH 7.25), 20% glycerol, and 0.1 mM EDTA and frozen at -70°C. Both fractions I and II were found to produce P450 spectra and had identical M, values of 51 kDa by SDS-PAGE, but only fraction I supported lauric acid hydroxylation in the reconstituted system. A minor protein contaminant of lower M, was observed in both P450 fractions obtained from the lauroyl-Sepharose column on SDS-PAGE (Fig. 1). The specific contents of the purified cytochrome P450 in fractions I and II were 18 and 15 nmol/mg protein, respectively. The P450 designated as fraction II was not characterized further. In one other purification attempt, only a single protein was eluted but the sample contained two N-terminal amino acids (data not shown). Cytochrome b6 and NADPH-cytochrome P450 reductase were eluted from the octylamino-Sepharose 4B column as described by Guengerich and Martin (30) and were further purified for use in the reconstitution experiments. The NADPH-cytochrome P450 reductase was purified by applying the enzyme to a 2’,5’-ADP-Sepharose 4B column as described by Yasukochi and Masters (32). Cytochrome b6 was purified by the procedure of Chiang (33). Laurate w- and (w-1)-hydroxylation reactions were performed by the procedure of Okita (34). In reconstitution experiments, 0.05 nmol each of purified P45Os, NADPH-cytochrome P450 reductase, and cytochrome bg was added to 20 pg dilauroylphosphatidylcholine and 10 pg deoxycholate in a final volume of 1.0 ml. After 5 min, the potassium phosphate buffer (pH 7.4), EDTA, and MgClz were added at final concentrations Fifty microliters of 2 mM of 50 mM, 0.1 mM, and 10 mM, respectively. sodium laurate or 2 mM sodium myristate (in water at pH 10) containing 0.1 pCi of the respective i4C-labeled fatty acid was added. Palmitic and stearic acids (10 mM solutions) were added in lo-g1 aliquots in methanol containing 0.3-0.4 PCi of 1-‘“C-labeled fatty acid. Addition of methanol in excess of 30 ~1 was found to decrease the specific activities of fatty acid hydroxylation reactions. The reactions were started by adding a
FIG. 1. SDS-PAGE of the two P450 fractions eluted from the lauroylSepharose 4B column. Lanes 1 and 2 contained 7.4 and 14.8 pg, respectively, of the P450 fraction I which supported lauric acid hydroxylation. Lanes 3 and 4 contained 14.4 and 7.2 pg, respectively, of the P450 in the second protein peak (fraction II).
P450 TABLE
FROM
Di(Z-ETHYLHEXYL)
PHTHALATE-TREATED
TABLE
I
Specific Activities of Laurate (w-l) and w-Hydroxylation Specific Contents of P450 in Liver Microsomes from Control or DEHP-Treated Rats Hydroxylauric formed
and
ll-OH
Hydroxylauric
acid
12.OH
II
Hydroxylation of Laurate by the Two P450 Fractions Which Eluted from the Lauroyl-Sepharose Column in Reconstitution Systems in the Presence and Absence of Cytochrome b5
P450 fraction Microsomes
477
RATS
ll-OH
12-OH
P450 (nmol
(nmol Control (n = 6) DEHP treated (n = 6) Fold induction
product/min/mg)
2.0 +- 0.1” 2.6 f 0.2 1.3
The increase in 12-hydroxylase ’ Standard deviation.
activity
1.4 f 0.2 14.0 * 2.2 10.0 was significant
acid formed
~nmol/mg) 0.79 + 0.16 0.86 f 0.22
at P < 0.001.
NADPH generating system containing 1 mM NADPH, 60 mu isocitrate, and 0.1 mg ICDH. Reactions were performed at 37’C for 5 min for laurate, 10 min for myristate, 20 min for palmitate, and 30 min for stearate. Reactions containing saturated fatty acids were stopped by the addition of 0.1 ml 1 N HCl and the fatty acids were extracted with ethyl acetate as described by Okita (34). The separation of metabolites of the different fatty acids and their detection by radiometric analysis by a Radiomatic detector (Meriden, CT) were performed by the procedure of Okita et al. (35) on a Bio-Rad ODS-5 column equipped with a precolumn. Cl2 to Cle hydroxy-fatty acids were identified by gas chromatography and mass spectroscopic analysis or cochromatography with wand (w-l)-hydroxylated standards. Separation of the hydroxy derivatives of stearate was achieved using an isocratic solvent system of 79.6% acetonitrile: 0.4% benzene: 19.8% water: 0.2% acetic acid. The unmetabolized stearate was eluted by changing to a solvent system of 99.6% acetonitrile: 0.4% benzene. Identification of the hydroxy-fatty acids of stearate by gas chromatography-mass spectroscopy was not attempted because of the minor amounts of products which were produced in either reconstituted or microsomal reactions. Two metabolites were obtained for stearate and their HPLC elution profile was similar to those for the reaction products obtained for laurate, myristate, and palmitate and it was assumed that the two peaks represent the same o- and (w-l)-hydroxy products as were determined for these fatty acids. Other procedures. SDS-PAGE was performed by the method of Laemmli (36) using an 8% gel and protein concentrations were determined by the method of Lowry et al. (37) using human serum albumin as standard Preparation of a rabbit polyclonal antibody to P450 isolated from DEHP-treated rats was performed by the procedure of Phillips et al. (38). Western blot analysis was performed by the procedure of Okita et al. (39) except the antibody to the P450 induced by DEHP was used. The specific contents of cytochromes P450 and bS were determined by the procedure of Estabrook and Werringloer (40). Amino acid sequencing was performed at the Medical College of Wisconsin Protein/Nucleic Acid Shared Facility and was performed on an Applied Biosystems Model 477A pulsed liquid phase protein sequencer equipped with an on line PTH-amino acid analyzer.
Fraction C-b& (+M
I
Fraction t-b,) (+b,)
II
Note. The reactions were performed Methods. Reactions were performed mean + standard deviation. a ND, not detected.
product/min/ nmol P450)
ND” ND
6.3 7t 2.1 63.0 k 6.0
ND ND
ND ND
as described under Materials and in triplicate and the values are the
from the lauroyl-Sepharose column with P450 present in both peaks. The two P450 fractions demonstrated identical M, values of 51 kDa on SDS-PAGE (Fig. l), but only fraction I (lanes 1 and 2) catalyzed laurate hydroxylation (Table II). The second peak or fraction II represented a minor fraction and was not characterized further. P450 fraction I exhibited a maximum reduced carbon monoxide absorbance at 452 nm (Fig. 2). This P450 supported laurate w-hydroxylation in the absence of cytochrome b5 but the reaction rate was stimulated lo-fold by the addition of an equimolar concentration of purified
1
I
I
I
,
,-
RESULTS
Rats treated with DEHP were found to increase their microsomal laurate o-hydroxylase activity lo-fold, whereas the (w-1)-hydroxylation reaction was increased by only 1.3-fold (Table I). Microsomal P450 levels were not increased following DEHP treatment (Table I). In two purification attempts, two protein peaks were eluted
410
430
450
Wavelength
470
490
(nm)
FIG. 2. Difference spectra of the dithionite reduced carbon monoxide bound P450mmp. This P450 absorbed maximally at 452 nm.
478
OKITA TABLE
AND
OKITA
III
TABLE
The Formation of the (w-l)- and w-Hydroxy-Fatty Acids of Laurate, Myristate, Palmitate, and Stearate in Reconstituted Systems Containing the Purified Cytochrome P450, NADPHCytochrome P450 Reductase, Dilauroyl Phosphatidylcholine, Deoxycholate, and Cytochrome b5
IV
Comparison of N-Terminal Amino Acid Sequence of Liver P45Os from Clofibrate (P450 4Al) and DEHP (P450nEHP) Treated Rats and from DEHP-Treated Rabbits (P450LPGAo-1) P450 4Al
P450DEHP
P450LPGAw-1
Hydroxy-fatty acid formed Fatty acid
w
(w-1)
(nmol product/min/nmol Laurie acid (n = 5) Myristate (n = 6) Palmitate (n = 9) Stearate (n = 6)
w/(w-1) P450)
N.D.b
82.1 + 8.3’
-
2.8 t 0.6 4.3 + 0.6 0.4 + 0.05
15.4 ?z 2.1 4.2 + 0.5 0.5 + 0.04
5.5 1.0 1.2
a o&l) represents the product ratio of w-hydroxy to (o-l)-hydroxy derivatives. b N.D., not detected. ’ Standard deviation.
cytochrome b5 (Table II). The metabolism of four saturated fatty acids, laurate, myristate, palmitate, and stearate, was studied and the rates are presented in Table III. Laurate was hydroxylated with the highest turnover rate (82.1 k 8.3 nmol/min/nmol P450) and only the w-hydroxylated product was found. There was no 11-hydroxy product detected (Table III and Fig. 3). The P450DEHP also catalyzed the hydroxylation of longer chain saturated fatty acids, but there was a progressive decrease in w-
12-OH
MINUTES
(ELUTION
OFF
ODS
31
COLUMN)
FIG. 3. HPLC tracing of the elution of 12-hydroxylaurate (12-OH) produced in the reconstituted system. The 12-OH product eluted at 13.2 min and represented 37% of the total cpm applied (11,537 cpm) to the HPLC column. The 11-hydroxy product elutes with a retention time of 10.2 min in this chromatographic system, but was not produced by this P450. The chromatographic procedure for separation of the hydroxy products from laurate is described in reference 35. The arrow indicates the time at which the solvent was changed to 100% methanol.
1 2 3 4 5 6 7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Met Ser Val
-
-
-
-
Ser
-
Ala Leu Ser Ser Thr Arg Phe Thr GUY Ser Ile Ser GUY Phe Leu Gln Val Ala
Ala
Ala
LHJ
LC?U
Ser Ala Thr Arg Phe Thr GUY Phe Ile Ser GUY Phe Leu Gln Val Ala
Asn Pro Thr Arg Leu Pro Gb Ser Leu Ser GUY Leu I&U
Gln Ala Ala
Note. The amino acid sequence of P4500EHp was determined on two P450 samples except for the glycine residue at position 17, which was identified in only one of the two samples. The amino acid sequences for P450 4Al and P450 LPGAw-1 were obtained from Ref. (12) and (44), respectively.
hydroxylation rate as the fatty acid substrate increased in carbon length. Myristate and palmitate exhibited only 19 and 5%, respectively, of the activity as laurate at the o-carbon. With fatty acids greater than 12 carbons, the (w-1)-carbon atom was also hydroxylated, with palmitate exhibiting the highest turnover rate at the (w-l)-carbon atom (4.3 + 0.6 nmol/min/nmol P450). The changes in w- and (w-1)-hydroxylation rates were also reflected in the w/(w-1) hydroxy product ratios for the different fatty acids as shown in Table III. Whereas no (o-1)-product was formed for laurate, the product ratios for myristate and palmitate were 5.5 and 1, respectively. The w/(0-1) product ratio was also 1 for stearate, but the hydroxylation rates at both the w- and the (w-1)-carbon positions were decreased dramatically as the carbon length increased from 16 to 18 carbon atoms (Table III). A N-terminal amino acid sequence was performed on cytochrome P450 fraction I and an l&amino acid sequence was identified. A methionine was determined to be the N-terminal amino acid residue for P450 4Al from clofibrate-treated rats based on its cDNA sequence (12, 41), but the N-terminal amino acid for P450DEHp was determined to be an alanine as shown in Table IV. Although
P450
FROM
Di(B-ETHYLHEXYL)
the N-terminal amino acids were not identical, when the N-terminal segments of P450DEHP and P450 4Al were aligned for best correlation, the two P450 segments demonstrated close agreement if the first four amino acids of P450 4Al were not included or the P450nEHP N-terminal fragment was displaced by four amino acids, as shown in Table IV. Two amino acid differences were noted between the N-terminal sequences of P450 4Al and P450nEHP. Whereas serines were found at residues 8 and 14 in P450 4A1, alanine was found at position 8 and phenylalanine at position 14 in P450nsnr. The N-terminal segment of liver P450LPGAw-1 which was isolated from rabbits treated with DEHP had 8 amino acid substitutions from P450nEHP as shown in Table IV. As shown in Fig. 4, a polyclonal antibody to the isolated P450nsnr recognized three protein bands in liver microsomes of DEHP-treated (lanes 1 and 3) and control rats (lane 2), but only two protein bands in phenobarbitaltreated rats (lane 4). The purified P450nEHP (lanes 6 and 7) corresponded to band B of lane 1 which has a 1M, value of 51 kDa; the M, of band A was 52 kDa. Both bands were induced by DEHP treatment, but band A was not present in phenobarbital-treated rats. A band of lower M, value (band C) was also detected in control and DEHP- and phenobarbital-treated rats as shown in Fig. 4, but this band did not appear to be induced by DEHP or phenobarbital treatment. The major protein band in control rats (lane 2) and phenobarbital-treated rats (lane 4) appears to run with or at a slightly lower M, value than band B of DEHP-treated rats. DISCUSSION DEHP is a member of a diverse group of compounds which have the capacity to increase the number of liver peroxisomes and induce the peroxisomal fatty acid p-oxidation system. This group also induces a unique group of cytochromes P450 which hydroxylate fatty acids and are classified in the P450 4A subfamily. Clofibrate has been the major peroxisome proliferator studied for its effects on the P450 system and one P450, P450 4A1, has been purified from livers of clofibrate-treated rats and extensively characterized by Gibson and his colleagues (8-11, 22, 23, 41), Hardwick et al. (12), and CaJacob et al. (15). The cDNA sequences of several hepatic and renal P45Os which hydroxylate fatty acids and prostaglandins have been identified (12, 42-44) and the nucleotide sequence of two genes, CYP4Al and CYP4A2, have been identified (18). P450 4AI and 4A3 have also been expressed in a vaccinia virus system and their hydroxylation of fatty acids and prostaglandins were studied (19). DEHP also is an inducer of the fatty acid w-hydroxylase system (5) and Hardwick et al. (12) have reported that the antibody made to the clofibrate inducible form will recognize the P450 induced by DEHP. In this study, we have isolated a P450 from liver microsomes of DEHP-treated rats
PHTHALATE-TREATED
12
479
RATS
3
4
5
6
7
FIG. 4. Western blot of liver microsomes from DEHP treated (lanes 1 and 3; 15 fig), control (lane 2; 15 pg), and phenobarbital treated (lane 4; 15 ng) rats using the polyclonal antibody against P45Onznr. Lanes 6 P450 nznr (0.5 pg). Lane 5 contained preand 7 contained the purified stained molecular weight markers. The M, values of bands A and B were 52 and 51 kDa, respectively. The visualization method utilized alkaline phosphatase-conjugated secondary antibody as described in reference 39.
and found it to have the same reduced carbon monoxide absorbance spectra and similar M, value (51.0 versus 51.5 kDa) as the clofibrate inducible form. The purified catalyzed laurate hydroxylation at the o-carbon P45°DEHP atom, but not at the (w-l)-position, which differs from the laurate hydroxylase rates reported for the P450 isolated from clofibrate-treated rats which metabolized both carbon atoms at an w/(w-1) product ratio of 4.1 (8). There appear to be differences between P450 preparations for CaJacob et al. (15) reported an w/(0-1) product ratio of 17 for their purified P450 isolated from clofibrate-treated rats. Since DEHP treatment of rats produced a significant lo-fold induction in microsomal laurate w-hydroxylase activity, but only 1.3-fold induction at the (w-1)-carbon atom, it was not surprising that the isolated P45Onsnp catalyzed hydroxylation only at the w-carbon atom. Laurate demonstrated the highest turnover value of the saturated fatty acids tested. Myristate, palmitate, and stearate had only 19, 5, and 0.6% of the activity of laurate, respectively. A linear relationship was observed between the turnover rates for w-hydroxylation and the number of carbon atoms in the saturated fatty acids if the data were examined on a semilog plot (Fig. 5). A decrease in fatty acid hydroxylation rates with increasing chain length of fatty acid was also reported by Ellin and Orrenius (3) in rat kidney microsomes. In the study of Aoyama et al. (19), P450 4Al was expressed in a vaccinia virus system and was reported to hydroxylate palmitate at a greater rate than laurate with an w/(w-1) product ratio of 1.25. These results differ for our findings with the purified P45Onsnp which hydroxylated palmitate at its w-carbon atom only 5% as efficiently as laurate in a reconstituted system with dilauroyl phosphatidylcholine/deoxycholate as the lipid matrix. These results may indicate that the
480
OKITA
AND
100 OMEGA
0
: I a
HYDROXYLATION
\ 10
Vol.
OF BIOCHEMISTRY
294, No. 2, May
AND
BIOPHYSICS
1, pp. 475-481,
1992
Characterization of a Cytochrome P450 from Di(2ethylhexyl) Phthalate-Treated Rats Which Hydroxylates Fatty Acids Richard T. Okital and Janice Rice Okit-a Department Received
of
Pharmaceutical Sciences, College of Pharmacy, Washington State University,
September
30, 1991, and in revised
form
January
6,1992
A cytochrome P450 was purified from liver microsomes of rats treated with di(2-ethylhexyl) phthalate (DEHP). DEHP is a member of a group of structurally diverse compounds which have been classified as peroxisome proliferators and are inducers of cytochromes P450 which hydroxylate lauric acid and other fatty acids. The P450 isolated from DEHP-treated rats (P4500zHp) was observed to have a &f, value of 51 kDa by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and a maximum absorbance of 452 nm in its reduced carbon monoxide bound state. The amino terminal residue for P4500sar was alanine and an l&amino acid segment at the N-terminal region was identified. The N-terminal amino acid for the P450 4A1 from clofibrate-treated rats is methionine and alignment of the N-terminal segment of the P4500sar with P450 4Al indicated that the first four amino acids were absent. There were two amino acid differences between the two P45Os in this l&amino acid segment; in P45Ousar an alanine and a phenylalanine were substituted for serines in P450 4A1. The was found to catalyze the hydroxylation of P450DEHP several saturated fatty acids, having the highest turnover activity with laurate (82.1 nmol12-OH-laurate formed/ min/nmol P450). My&state, palmitate, and stearate were also metabolized but at decreasing rates. Cytochrome bs stimulated laurate 12-hydroxylation lo-fold in a reconstituted system. Laurate was not metabolized at its llcarbon atom; however, the longer chain length fatty acids were metabolized at the (w-l)-carbon atom in addition to the o-carbon atom. A polyclonal antibody to the P4500sHp recognized three protein bands in liver microsomes from control and DEHP-treated rats on Western blot analysis, but only two protein bands from phenobarbital-treated rats. 0 1992 Academic Pm. Inc.
i To whom correspondence should macy, Washington State University, 99164-6510. 0003-9661/92 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
be addressed 105 Wegner
Pullman, Washington 99164
at College of PharHall, Pullman, WA
Fatty acid hydroxylation was one of the first reactions used to characterize the cytochrome P450 system in liver and kidney microsomes and was the first reaction to be examined in a reconstituted P450 system (l-4). The 12carbon fatty acid, dodecanoic or lauric acid, has been used extensively to characterize the fatty acid hydroxylation system (1). There has been renewed interest in this P450mediated reaction because of its induction by various peroxisome proliferators, including clofibrate and diethylhexyl phthalate (DEHP)2 (5, 6). Orton and Parker (7) were the first to report the inductive effects of hypolipidemic agents on this P450-mediated reaction and several groups have characterized the microsomal and purified P450 from clofibrate-treated rats (8-17). P45Os which catalyze the hydroxylation of fatty acids at the w-terminus are designated as belonging to P450 subfamily 4A3 and at least three genes coding for P45Os in liver and kidney exhibit increased expression when rats are administered clofibrate (18). Gonzalez’s laboratory has characterized the genomic DNA for CYP4A13 and 4A2 (18) and has expressed P45Os 4Al and 4A3 in a vaccinia virus system (19). In control rats, the hydroxylation of lauric acid occurs at both the ll- and the 12-carbon atoms in an approximately 1:l ratio (14, 20). Treatment with phenobarbital induces the 11-hydroxylase 2-fold with only minor effect on the 12-hydroxylase (21), but treatment of rats with clofibrate induces the microsomal laurate 12-hydroxylase approximately lo-fold with only a 1.9-fold induction in the 11-hydroxylase (22). P45Os induced by peroxisome proliferators also hydroxylate longer, unsaturated fatty ’ Abbreviations used: Chaps, 3-[ (3-cholamidopropyl)dimethylammonio] propanesulfonate; DEHP, di(2-ethylhexyl) phthalate; ICDH, isocitrate dehydrogenase; P450, cytochrome P450; 11-OH-Laurate, ll- or (w-l)hydroxylaurate; 12-OH-Laurate, 12- or w-hydroxylaurate; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis. 3 The nomenclature of the cytochrome P450 system which catalyzes fatty acid hydroxylation is taken from “The P450 superfamily: Update on new sequences, gene mapping and recommended nomenclature” published in DNA lO, l-14,199l. This cytochrome P450 has also been referred to as P450,,, or cytochrome P-452. 415
Inc. reserved.
476
OKITA
AND
acids (23, 24). Capdevila et al. (24) reported an B-fold increase in arachidonic acid 20-hydroxylation in microsomes of ciprofibrate-treated rats which changed the o/ (w-1) hydroxy product ratio from 0.5 in control rats to 7.5 in the ciprofibrate-treated rats. Okita and Chance (20), Lake et al. (25,26), Mitchell et al. (27), and Elcombe and Mitchell (28) have also described the induction of laurate w-hydroxylase activity by another peroxisome proliferator, DEHP, a common plasticizer used in the manufacture of polyvinyl chloride. The P450 induced by DEHP was demonstrated by Hardwick et al. (12) to be immunochemically similar to the P450 induced by clofibrate. This P450 form represents a minor amount of the total P45Os in control livers but increases to 16-30% of the total P450 content following DEHP or clofibrate administration (10, 15). In the present study, we report the characterization of a P450 isolated from liver microsomes of DEHP-treated rats. MATERIALS
AND
METHODS
Materials. Male Sprague-Dawley rats were purchased from Sasco (Omaha, NE). Purina rat chow supplemented with 2% (w/w) DEHP was purchased from Ralston Purina (Richmond, IN). DEHP was obtained from Aldrich Chemicals (Milwaukee, WI). Sodium laurate, sodium myristate, pahnitic acid, stearic acid, isocitrate, ICDH, DEAE-Sephacel, deoxycholate, dilauroyl phosphatidylcholine, sodium phenobarbital, and 16-hydroxypalmitic acid were purchased from Sigma (St. Louis, MO). Chaps was purchased from Pierce Chemicals (Rockford, IL) and Emulgen 913 was donated by Kao Atlas (Japan). The l-‘V-labeled lauric, myristic, palmitic, and stearic acids were purchased from Amersham (Arlington Heights, IL). Methanol, ethyl acetate, and acetonitrile were obtained from Burdick and Jackson (Muskegon, MI). The 2’,5’-ADP-Sepharose 4B and AH-Sepharose 4B resins were purchased from Pharmacia (Piscataway, NJ). Bio-Gel HTP hydroxylapatite, SDS-PAGE chemicals, and the Biosil ODS-5 (15 X 0.4 cm) HPLC and microguard columns were purchased from Bio-Rad (Richmond, CA). All chemicals used in the purification procedures and assays were the highest grade available. Preparation of P450DEHp Microsomes were initially prepared from 170- to 200-g Sprague-Dawley rats which were fed a normal Purina rat chow diet for 1 week prior to starting them on a 2-week diet containing 2% (w/w) DEHP. However, when this diet was no longer commercially available, rats were given DEHP by intubation at a dose of 1200 mg/ kg per day for 3 days in peroxide-free corn oil. Induction of the laurate hydroxylation reactions was very similar by the two dosing procedures. Phenobarbital was dissolved in 0.9% saline and administered to rats intraperitoneally at a dose of 80 mg/kg for 4 days. Food was withdrawn 15 to 18 h prior to sacrificing the rats. Liver microsomes were prepared by the procedure of Remmer et al. (29). Microsomes (10 mg/ml) were solubilized with 1.8% (w/v) sodium cholate for 30 min, centrifuged to remove nonsolubilized material, and diluted to 0.7% (w/v) sodium cholate. The cytochrome P450 from DEHP-treated rats that hydroxylated lauric acid was isolated by a modification of the procedure of Tamburini et al. (9) using the octylamino-Sepharose 4B procedure of Guengerich and Martin (30) to initially separate the P450 fraction from cytochrome br,, NADPH-cytochrome P450 reductase, and other proteins and lipids. In the present purification scheme, the order of the hydroxylapatite and DEAE-Sephacel column procedures was reversed and a lauroyl-Sepharose column was added as the final step in the Tamburini et al. (9) procedure to aid in the purification of the P450 and in removal of the nonionic detergent. The procedure of Gibson and Schenkman (31) was used to prepare the lauroyl-Sepharose column from AH-Sepharose 4B. The lauroyl-Sepharose column (15 X 1.5 cm) was equilibrated in 50 mM potassium phosphate buffer (pH 7.25), 20% glycerol, and 0.2 M NaCl.
OKITA The P450 fraction obtained from the hydroxylapatite column was applied to the lauroyl-Sepharose column and was extensively washed to remove the nonionic detergent, Emulgen 913, which was determined by the decrease in Am readings until constant values were obtained. A Chaps detergent gradient from 0 to 20 mM in 50 mM potassium phosphate buffer (pH 7.25), 20% glycerol, and 0.2 M NaCl (50 ml of each gradient buffer) was applied to elute the bound P450. Approximately 0.5-ml fractions were collected from the lauroyl-Sepharose column and AdI readings were measured. In two of three purification attempts, the P450 fraction applied to the lauroyl-Sepharose column separated into two protein peaks; designated as fractions I and II. The two fractions were dialyzed extensively against 50 mM potassium phosphate buffer (pH 7.25), 20% glycerol, and 0.1 mM EDTA and frozen at -70°C. Both fractions I and II were found to produce P450 spectra and had identical M, values of 51 kDa by SDS-PAGE, but only fraction I supported lauric acid hydroxylation in the reconstituted system. A minor protein contaminant of lower M, was observed in both P450 fractions obtained from the lauroyl-Sepharose column on SDS-PAGE (Fig. 1). The specific contents of the purified cytochrome P450 in fractions I and II were 18 and 15 nmol/mg protein, respectively. The P450 designated as fraction II was not characterized further. In one other purification attempt, only a single protein was eluted but the sample contained two N-terminal amino acids (data not shown). Cytochrome b6 and NADPH-cytochrome P450 reductase were eluted from the octylamino-Sepharose 4B column as described by Guengerich and Martin (30) and were further purified for use in the reconstitution experiments. The NADPH-cytochrome P450 reductase was purified by applying the enzyme to a 2’,5’-ADP-Sepharose 4B column as described by Yasukochi and Masters (32). Cytochrome b6 was purified by the procedure of Chiang (33). Laurate w- and (w-1)-hydroxylation reactions were performed by the procedure of Okita (34). In reconstitution experiments, 0.05 nmol each of purified P45Os, NADPH-cytochrome P450 reductase, and cytochrome bg was added to 20 pg dilauroylphosphatidylcholine and 10 pg deoxycholate in a final volume of 1.0 ml. After 5 min, the potassium phosphate buffer (pH 7.4), EDTA, and MgClz were added at final concentrations Fifty microliters of 2 mM of 50 mM, 0.1 mM, and 10 mM, respectively. sodium laurate or 2 mM sodium myristate (in water at pH 10) containing 0.1 pCi of the respective i4C-labeled fatty acid was added. Palmitic and stearic acids (10 mM solutions) were added in lo-g1 aliquots in methanol containing 0.3-0.4 PCi of 1-‘“C-labeled fatty acid. Addition of methanol in excess of 30 ~1 was found to decrease the specific activities of fatty acid hydroxylation reactions. The reactions were started by adding a
FIG. 1. SDS-PAGE of the two P450 fractions eluted from the lauroylSepharose 4B column. Lanes 1 and 2 contained 7.4 and 14.8 pg, respectively, of the P450 fraction I which supported lauric acid hydroxylation. Lanes 3 and 4 contained 14.4 and 7.2 pg, respectively, of the P450 in the second protein peak (fraction II).
P450 TABLE
FROM
Di(Z-ETHYLHEXYL)
PHTHALATE-TREATED
TABLE
I
Specific Activities of Laurate (w-l) and w-Hydroxylation Specific Contents of P450 in Liver Microsomes from Control or DEHP-Treated Rats Hydroxylauric formed
and
ll-OH
Hydroxylauric
acid
12.OH
II
Hydroxylation of Laurate by the Two P450 Fractions Which Eluted from the Lauroyl-Sepharose Column in Reconstitution Systems in the Presence and Absence of Cytochrome b5
P450 fraction Microsomes
477
RATS
ll-OH
12-OH
P450 (nmol
(nmol Control (n = 6) DEHP treated (n = 6) Fold induction
product/min/mg)
2.0 +- 0.1” 2.6 f 0.2 1.3
The increase in 12-hydroxylase ’ Standard deviation.
activity
1.4 f 0.2 14.0 * 2.2 10.0 was significant
acid formed
~nmol/mg) 0.79 + 0.16 0.86 f 0.22
at P < 0.001.
NADPH generating system containing 1 mM NADPH, 60 mu isocitrate, and 0.1 mg ICDH. Reactions were performed at 37’C for 5 min for laurate, 10 min for myristate, 20 min for palmitate, and 30 min for stearate. Reactions containing saturated fatty acids were stopped by the addition of 0.1 ml 1 N HCl and the fatty acids were extracted with ethyl acetate as described by Okita (34). The separation of metabolites of the different fatty acids and their detection by radiometric analysis by a Radiomatic detector (Meriden, CT) were performed by the procedure of Okita et al. (35) on a Bio-Rad ODS-5 column equipped with a precolumn. Cl2 to Cle hydroxy-fatty acids were identified by gas chromatography and mass spectroscopic analysis or cochromatography with wand (w-l)-hydroxylated standards. Separation of the hydroxy derivatives of stearate was achieved using an isocratic solvent system of 79.6% acetonitrile: 0.4% benzene: 19.8% water: 0.2% acetic acid. The unmetabolized stearate was eluted by changing to a solvent system of 99.6% acetonitrile: 0.4% benzene. Identification of the hydroxy-fatty acids of stearate by gas chromatography-mass spectroscopy was not attempted because of the minor amounts of products which were produced in either reconstituted or microsomal reactions. Two metabolites were obtained for stearate and their HPLC elution profile was similar to those for the reaction products obtained for laurate, myristate, and palmitate and it was assumed that the two peaks represent the same o- and (w-l)-hydroxy products as were determined for these fatty acids. Other procedures. SDS-PAGE was performed by the method of Laemmli (36) using an 8% gel and protein concentrations were determined by the method of Lowry et al. (37) using human serum albumin as standard Preparation of a rabbit polyclonal antibody to P450 isolated from DEHP-treated rats was performed by the procedure of Phillips et al. (38). Western blot analysis was performed by the procedure of Okita et al. (39) except the antibody to the P450 induced by DEHP was used. The specific contents of cytochromes P450 and bS were determined by the procedure of Estabrook and Werringloer (40). Amino acid sequencing was performed at the Medical College of Wisconsin Protein/Nucleic Acid Shared Facility and was performed on an Applied Biosystems Model 477A pulsed liquid phase protein sequencer equipped with an on line PTH-amino acid analyzer.
Fraction C-b& (+M
I
Fraction t-b,) (+b,)
II
Note. The reactions were performed Methods. Reactions were performed mean + standard deviation. a ND, not detected.
product/min/ nmol P450)
ND” ND
6.3 7t 2.1 63.0 k 6.0
ND ND
ND ND
as described under Materials and in triplicate and the values are the
from the lauroyl-Sepharose column with P450 present in both peaks. The two P450 fractions demonstrated identical M, values of 51 kDa on SDS-PAGE (Fig. l), but only fraction I (lanes 1 and 2) catalyzed laurate hydroxylation (Table II). The second peak or fraction II represented a minor fraction and was not characterized further. P450 fraction I exhibited a maximum reduced carbon monoxide absorbance at 452 nm (Fig. 2). This P450 supported laurate w-hydroxylation in the absence of cytochrome b5 but the reaction rate was stimulated lo-fold by the addition of an equimolar concentration of purified
1
I
I
I
,
,-
RESULTS
Rats treated with DEHP were found to increase their microsomal laurate o-hydroxylase activity lo-fold, whereas the (w-1)-hydroxylation reaction was increased by only 1.3-fold (Table I). Microsomal P450 levels were not increased following DEHP treatment (Table I). In two purification attempts, two protein peaks were eluted
410
430
450
Wavelength
470
490
(nm)
FIG. 2. Difference spectra of the dithionite reduced carbon monoxide bound P450mmp. This P450 absorbed maximally at 452 nm.
478
OKITA TABLE
AND
OKITA
III
TABLE
The Formation of the (w-l)- and w-Hydroxy-Fatty Acids of Laurate, Myristate, Palmitate, and Stearate in Reconstituted Systems Containing the Purified Cytochrome P450, NADPHCytochrome P450 Reductase, Dilauroyl Phosphatidylcholine, Deoxycholate, and Cytochrome b5
IV
Comparison of N-Terminal Amino Acid Sequence of Liver P45Os from Clofibrate (P450 4Al) and DEHP (P450nEHP) Treated Rats and from DEHP-Treated Rabbits (P450LPGAo-1) P450 4Al
P450DEHP
P450LPGAw-1
Hydroxy-fatty acid formed Fatty acid
w
(w-1)
(nmol product/min/nmol Laurie acid (n = 5) Myristate (n = 6) Palmitate (n = 9) Stearate (n = 6)
w/(w-1) P450)
N.D.b
82.1 + 8.3’
-
2.8 t 0.6 4.3 + 0.6 0.4 + 0.05
15.4 ?z 2.1 4.2 + 0.5 0.5 + 0.04
5.5 1.0 1.2
a o&l) represents the product ratio of w-hydroxy to (o-l)-hydroxy derivatives. b N.D., not detected. ’ Standard deviation.
cytochrome b5 (Table II). The metabolism of four saturated fatty acids, laurate, myristate, palmitate, and stearate, was studied and the rates are presented in Table III. Laurate was hydroxylated with the highest turnover rate (82.1 k 8.3 nmol/min/nmol P450) and only the w-hydroxylated product was found. There was no 11-hydroxy product detected (Table III and Fig. 3). The P450DEHP also catalyzed the hydroxylation of longer chain saturated fatty acids, but there was a progressive decrease in w-
12-OH
MINUTES
(ELUTION
OFF
ODS
31
COLUMN)
FIG. 3. HPLC tracing of the elution of 12-hydroxylaurate (12-OH) produced in the reconstituted system. The 12-OH product eluted at 13.2 min and represented 37% of the total cpm applied (11,537 cpm) to the HPLC column. The 11-hydroxy product elutes with a retention time of 10.2 min in this chromatographic system, but was not produced by this P450. The chromatographic procedure for separation of the hydroxy products from laurate is described in reference 35. The arrow indicates the time at which the solvent was changed to 100% methanol.
1 2 3 4 5 6 7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Met Ser Val
-
-
-
-
Ser
-
Ala Leu Ser Ser Thr Arg Phe Thr GUY Ser Ile Ser GUY Phe Leu Gln Val Ala
Ala
Ala
LHJ
LC?U
Ser Ala Thr Arg Phe Thr GUY Phe Ile Ser GUY Phe Leu Gln Val Ala
Asn Pro Thr Arg Leu Pro Gb Ser Leu Ser GUY Leu I&U
Gln Ala Ala
Note. The amino acid sequence of P4500EHp was determined on two P450 samples except for the glycine residue at position 17, which was identified in only one of the two samples. The amino acid sequences for P450 4Al and P450 LPGAw-1 were obtained from Ref. (12) and (44), respectively.
hydroxylation rate as the fatty acid substrate increased in carbon length. Myristate and palmitate exhibited only 19 and 5%, respectively, of the activity as laurate at the o-carbon. With fatty acids greater than 12 carbons, the (w-1)-carbon atom was also hydroxylated, with palmitate exhibiting the highest turnover rate at the (w-l)-carbon atom (4.3 + 0.6 nmol/min/nmol P450). The changes in w- and (w-1)-hydroxylation rates were also reflected in the w/(w-1) hydroxy product ratios for the different fatty acids as shown in Table III. Whereas no (o-1)-product was formed for laurate, the product ratios for myristate and palmitate were 5.5 and 1, respectively. The w/(0-1) product ratio was also 1 for stearate, but the hydroxylation rates at both the w- and the (w-1)-carbon positions were decreased dramatically as the carbon length increased from 16 to 18 carbon atoms (Table III). A N-terminal amino acid sequence was performed on cytochrome P450 fraction I and an l&amino acid sequence was identified. A methionine was determined to be the N-terminal amino acid residue for P450 4Al from clofibrate-treated rats based on its cDNA sequence (12, 41), but the N-terminal amino acid for P450DEHp was determined to be an alanine as shown in Table IV. Although
P450
FROM
Di(B-ETHYLHEXYL)
the N-terminal amino acids were not identical, when the N-terminal segments of P450DEHP and P450 4Al were aligned for best correlation, the two P450 segments demonstrated close agreement if the first four amino acids of P450 4Al were not included or the P450nEHP N-terminal fragment was displaced by four amino acids, as shown in Table IV. Two amino acid differences were noted between the N-terminal sequences of P450 4Al and P450nEHP. Whereas serines were found at residues 8 and 14 in P450 4A1, alanine was found at position 8 and phenylalanine at position 14 in P450nsnr. The N-terminal segment of liver P450LPGAw-1 which was isolated from rabbits treated with DEHP had 8 amino acid substitutions from P450nEHP as shown in Table IV. As shown in Fig. 4, a polyclonal antibody to the isolated P450nsnr recognized three protein bands in liver microsomes of DEHP-treated (lanes 1 and 3) and control rats (lane 2), but only two protein bands in phenobarbitaltreated rats (lane 4). The purified P450nEHP (lanes 6 and 7) corresponded to band B of lane 1 which has a 1M, value of 51 kDa; the M, of band A was 52 kDa. Both bands were induced by DEHP treatment, but band A was not present in phenobarbital-treated rats. A band of lower M, value (band C) was also detected in control and DEHP- and phenobarbital-treated rats as shown in Fig. 4, but this band did not appear to be induced by DEHP or phenobarbital treatment. The major protein band in control rats (lane 2) and phenobarbital-treated rats (lane 4) appears to run with or at a slightly lower M, value than band B of DEHP-treated rats. DISCUSSION DEHP is a member of a diverse group of compounds which have the capacity to increase the number of liver peroxisomes and induce the peroxisomal fatty acid p-oxidation system. This group also induces a unique group of cytochromes P450 which hydroxylate fatty acids and are classified in the P450 4A subfamily. Clofibrate has been the major peroxisome proliferator studied for its effects on the P450 system and one P450, P450 4A1, has been purified from livers of clofibrate-treated rats and extensively characterized by Gibson and his colleagues (8-11, 22, 23, 41), Hardwick et al. (12), and CaJacob et al. (15). The cDNA sequences of several hepatic and renal P45Os which hydroxylate fatty acids and prostaglandins have been identified (12, 42-44) and the nucleotide sequence of two genes, CYP4Al and CYP4A2, have been identified (18). P450 4AI and 4A3 have also been expressed in a vaccinia virus system and their hydroxylation of fatty acids and prostaglandins were studied (19). DEHP also is an inducer of the fatty acid w-hydroxylase system (5) and Hardwick et al. (12) have reported that the antibody made to the clofibrate inducible form will recognize the P450 induced by DEHP. In this study, we have isolated a P450 from liver microsomes of DEHP-treated rats
PHTHALATE-TREATED
12
479
RATS
3
4
5
6
7
FIG. 4. Western blot of liver microsomes from DEHP treated (lanes 1 and 3; 15 fig), control (lane 2; 15 pg), and phenobarbital treated (lane 4; 15 ng) rats using the polyclonal antibody against P45Onznr. Lanes 6 P450 nznr (0.5 pg). Lane 5 contained preand 7 contained the purified stained molecular weight markers. The M, values of bands A and B were 52 and 51 kDa, respectively. The visualization method utilized alkaline phosphatase-conjugated secondary antibody as described in reference 39.
and found it to have the same reduced carbon monoxide absorbance spectra and similar M, value (51.0 versus 51.5 kDa) as the clofibrate inducible form. The purified catalyzed laurate hydroxylation at the o-carbon P45°DEHP atom, but not at the (w-l)-position, which differs from the laurate hydroxylase rates reported for the P450 isolated from clofibrate-treated rats which metabolized both carbon atoms at an w/(w-1) product ratio of 4.1 (8). There appear to be differences between P450 preparations for CaJacob et al. (15) reported an w/(0-1) product ratio of 17 for their purified P450 isolated from clofibrate-treated rats. Since DEHP treatment of rats produced a significant lo-fold induction in microsomal laurate w-hydroxylase activity, but only 1.3-fold induction at the (w-1)-carbon atom, it was not surprising that the isolated P45Onsnp catalyzed hydroxylation only at the w-carbon atom. Laurate demonstrated the highest turnover value of the saturated fatty acids tested. Myristate, palmitate, and stearate had only 19, 5, and 0.6% of the activity of laurate, respectively. A linear relationship was observed between the turnover rates for w-hydroxylation and the number of carbon atoms in the saturated fatty acids if the data were examined on a semilog plot (Fig. 5). A decrease in fatty acid hydroxylation rates with increasing chain length of fatty acid was also reported by Ellin and Orrenius (3) in rat kidney microsomes. In the study of Aoyama et al. (19), P450 4Al was expressed in a vaccinia virus system and was reported to hydroxylate palmitate at a greater rate than laurate with an w/(w-1) product ratio of 1.25. These results differ for our findings with the purified P45Onsnp which hydroxylated palmitate at its w-carbon atom only 5% as efficiently as laurate in a reconstituted system with dilauroyl phosphatidylcholine/deoxycholate as the lipid matrix. These results may indicate that the
480
OKITA
AND
100 OMEGA
0
: I a
HYDROXYLATION
\ 10
