Porcine Microsomal Vitamin D3 25-Hydroxylase (CYP2D25)

The metabolic activation of the prohormone vitamin D3 requires a 25-hydroxylation that has been reported to be catalyzed by both mitochondrial CYP27A and a microsomal vitamin D3 25-hydroxylase in the liver. CYP27A has been extensively studied, but its role as a physiologically important vitamin D3 25-hydroxylase has been questioned. The present paper reports that the microsomal vitamin D325-hydroxylase, purified from pig liver, converted vitamin D3 into 25-hydroxyvitamin D3 in substrate concentrations which are within the physiological range (apparentK m = 0.1 μm). The enzyme 25-hydroxylated vitamin D3, 1α-hydroxyvitamin D3 and vitamin D2 and also converted tolterodine, a substrate for human CYP2D6, into its 5-hydroxymethyl metabolite. Tolterodine inhibited the microsomal 25-hydroxylation, whereas quinidine, an inhibitor of CYP2D6, did not markedly inhibit the reaction. The primary structure of the microsomal vitamin D3 25-hydroxylase, designated CYP2D25, shows 77% identity with that of human CYP2D6. Northern blot and reverse transcription-polymerase chain reaction experiments revealed that CYP2D25 mRNA is expressed in higher levels in liver than in kidney and in small amounts in adrenals, brain, heart, intestine, lung, muscle, spleen, and thymus. Experiments with human liver microsomes and recombinantly expressed CYP2D6 strongly indicate that the microsomal 25-hydroxylation of vitamin D3 in human liver is catalyzed by an enzyme different from CYP2D6.

Vitamin D 3 (cholecalciferol) is essential for the absorption of calcium and phosphate in the intestine and has effects on the regulation of growth and differentiation of certain specialized cell types. The prohormone vitamin D 3 is formed from 7-dehydrocholesterol in the skin under the influence of ultraviolet irradiation. Vitamin D 3 can also be derived from dietary sources. Vitamin D 2 (ergocalciferol), which differs structurally from vitamin D 3 in the side chain, has been frequently used to treat and prevent vitamin D deficiency and is used in parenteral vitamin formulations. A cytochrome P450-dependent 25hydroxylation in the liver is the first step in the metabolic activation of both vitamin D 3 and vitamin D 2 into their active hormonal forms. It has been reported that extracts and enzyme preparations from other tissues, e.g. kidney, also contain this activity. The 25-hydroxylation is followed by the tightly regulated 1␣-hydroxylation in the kidney, to form the biologically active compounds 1␣,25-dihydroxyvitamin D 3 and 1␣,25-dihydroxyvitamin D 2 , respectively (for review, see Ref. 1).
A mitochondrial vitamin D 3 25-hydroxylase, CYP27A, 1 has been purified from several species and characterized (2)(3)(4)(5)(6)(7). It is well established that this enzyme catalyzes the obligatory 27hydroxylation of cholesterol and C 27 -sterols in bile acid biosynthesis. The tissue distribution of this mitochondrial enzyme is wide spread, its mRNA is found not only in liver but in most other tissues (3). Leading authorities in the vitamin D field have expressed skepticism that a single enzyme, i.e. CYP27A, is responsible for the hepatic 25-hydroxylation as a whole (1). There are several unexplained observations suggesting that a cytochrome P450 other than CYP27A performs 25-hydroxylation of vitamin D. Thus, studies on patients having a mutated CYP27A gene demonstrated a lack of 27-hydroxylation of C 27sterols in bile acid biosynthesis but not decreased serum levels of 25-hydroxyvitamin D 3 (8). Mice with the disrupted sterol 27-hydroxylase gene showed markedly reduced bile acid biosynthesis, but the circulating 25-hydroxyvitamin D 3 levels were increased more than 2-fold (9). Furthermore, the mitochondrial CYP27A does not 25-hydroxylate vitamin D 2 (2), evoking the question: which cytochrome P450 enzyme synthesizes 25-hydroxyvitamin D 2 (1)?
A microsomal vitamin D 3 25-hydroxylase was purified from pig liver by Axén et al. (10). The microsomal 25-hydroxylase catalyzed 25-hydroxylation of both vitamin D 3 and vitamin D 2 (11). An antibody raised against the pig enzyme recognized a protein of similar apparent molecular weight in human liver microsomes (10). A cDNA encoding the pig liver microsomal enzyme was recently isolated (12). The structure of microsomal vitamin D 3 25-hydroxylase, as deduced by both DNA sequence analysis and protein sequence analysis, showed 70 -80% identity with members of the CYP2D subfamily. The enzyme has been assigned the name CYP2D25. 2 The only CYP2D enzyme known to be expressed in man is CYP2D6. This enzyme is polymorphically expressed and is lacking in 5-10% of the caucasian population. It is not known whether vitamin D 3 25hydroxylation in man could be catalyzed by CYP2D6.
The objective of the current study was to characterize the microsomal vitamin D 3 25-hydroxylase (CYP2D25) by studying its catalytic properties, effect of inhibitors, and tissue distribu-* This work was supported by the Swedish Medical Research Council Project 03X-218. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM  RNA Isolation-Tissues were obtained from 6-months-old, castrated male pigs. Total RNA was extracted from liver, kidney, lung, heart, muscle, thymus, spleen, intestine, adrenal, and brain using RN easy Midi Kit (Qiagen) following the manufacturer's instructions.
Northern Blot Analysis-Total RNA, 30 g, from each tissue was denatured for analysis of CYP2D25 and cyclophilin transcripts, separated on 1% denaturing agarose gels, and subjected to Northern blot and hybridization with a 32 P-labeled 1624-bp KpnI-XbaI fragment of pig liver microsomal CYP2D25 cDNA and a 400-bp fragment of rat cyclophilin cDNA as control (12).
RT-PCR-Total RNA was extracted from different tissues of castrated male pigs as described above. First-strand cDNA synthesis was performed using 1 g of RNA from liver and kidney and 10 g of RNA from brain, lung, thymus, intestine, muscle, heart, adrenal, and spleen and 30 pmol of a CYP2D25-specific antisense primer. The negative controls containing RNA without reverse transcriptase enzyme were included in each experiment. The resulting cDNA was subjected to PCR amplification using 30 pmol of each of sense and antisense primers for CYP2D25. The following primers (purchased from Life Technologies, Inc.) were used to obtain the 245-bp cDNA sequence of CYP2D25: sense, 5-TCTCCTCCTCATGATCCTGC-3 (nucleotides 1025-1045) and antisense, 5-GGTTGGTGATAAGTGGTCC-3 (nucleotides 1269 -1248). The PCR profile was: 94°C denaturation for 1 min, 59°C annealing for 1 min, 72°C extension for 1 min, followed by 28 cycles with 94°C denaturation for 15 s, 59°C annealing for 1 min, 72°C extension for 1 min, and 72°C for 10 min. The 245-bp product was analyzed by electrophoresis on 2% agarose gels containing 0.5 g/ml ethidium bromide. The identity of the DNA product isolated from the electrophoretic gel was verified by restriction enzyme analysis (BsiHKA I from BioLabs Inc . ).
Source of Enzymes-Microsomal vitamin D 3 25-hydroxylase was purified from pig liver as described by Axén et al. (10). The apparently homogeneous enzyme showed a single protein band upon SDS-PAGE. Human CYP27A, recombinantly expressed in Escherichia coli, was purified as described previously (13). Human liver microsome samples from different donors were obtained from Gentest Corp. (Woburn, MA) and used in immunoblotting and incubation experiments. Recombinantly expressed human CYP2D6 (SUPERSOMES TM ) and control microsomes from nontransfected insect cells were obtained from Gentest Corp. (Woburn, MA).
SDS-PAGE and Immunoblottning-SDS-PAGE was carried out as described previously (10,16). Electrophoresis and immunoblotting were performed with CYP2D25 and different human liver microsome samples from various donors, some of them with polymorphism for CYP2D6. The monoclonal and polyclonal antibodies raised against purified pig liver microsomal vitamin D 3 25-hydroxylase (CYP2D25), used in the Western blot experiments, were prepared in this laboratory (10). The monoclonal antibody raised against human CYP2D6 was obtained from Gentest Corp.
In some experiments with vitamin D 3 and the microsomal vitamin D 3 25-hydroxylase (CYP2D25), the substrate concentrations were in the range 62-500 nM. Incubations were carried out for 10 -30 min at 37°C with 0.025-0.1 nmol of CYP2D25, 1 unit of NADPH-cytochrome P450 reductase (14), and 1.2 mol of NADPH in a final volume of 1 ml of 50 mM Tris acetate buffer, pH 7.4, containing 20% glycerol and 0.1 mM EDTA. In incubations with mitochondrial vitamin D 3 25-hydroxylase (CYP27A), 4 nmol of adrenodoxin and 0.4 nmol of adrenodoxin reductase (15) were added instead of microsomal NADPH-cytochrome P450 reductase. Incubations with human liver microsome samples (1 mg of protein) and microsomes from insect cells transfected with CYP2D6 (SUPERSOMES TM ) were carried out as described above but without addition of NADPH-cytochrome P450 reductase.
The incubations were terminated, extracted, and analyzed by straight-phase HPLC and reverse-phase HPLC as described (2,10,11). Under the conditions employed, the lower limits for detection of 25hydroxylation rates toward vitamin D 3 , vitamin D 2 , and 1␣-hydroxyvitamin D 3 , respectively, were 0.5, 2.5, and 0.5 pmol/nmol cytochrome P450/min. Incubation experiments to study the effect of different inhibitors on the 25-hydroxylation of vitamin D 3 by CYP2D25 were performed as described above but with the addition of various concentrations of respective inhibitor dissolved in 10 l of methanol. Cyclosporin A was dissolved in 10 l of acetone. Control experiments were performed with the addition of 10 l of methanol or acetone.
Incubations with Debrisoquine-Radiolabled debrisoquine (1 Ci), 3 g, was incubated for 60 min at 37°C with 0.25 nmol of CYP2D25, 1 unit of NADPH-cytochrome P450 reductase, 5 g of dilauroylphosphocholine, and 0.2 mM NADPH in a total volume of 100 l of 50 mM Tris acetate, pH 7.4, containing 20% glycerol and 0.1 mM EDTA. Before addition of NADPH, a preincubation was carried out for 5 min at 37°C. The reaction was terminated, and the mixtures were extracted and subjected to HPLC (LiChrosphere RP, 18.4 ϫ 250 mm; 5 m; Merck) as described by the manufacturer (Amersham Pharmacia Biotech).
Incubations with Tolterodine-[ 14 C]Tolterodine, 10 M, was incubated for 30 min at 37°C according to the procedures described by Postlind et al. (17). The incubation mixtures with the reconstituted vitamin D 3 25-hydroxylase system contained 0.1-0.25 nmol of CYP2D25, 2 units of NADPH-cytochrome P450 reductase, and 0.5 mg of NADPH in a total volume of 250 l of 100 mM PO 4 buffer, pH 7.4. The incubations were terminated by addition of 125 l of acetonitrile, and the mixtures were analyzed for product formation by HPLC as described previously (17). The enzymatic product formed had the same retention time as the authentic 5-hydroxymethyl metabolite of tolterodine (17).

RESULTS AND DISCUSSION
Catalytic Properties of Pig Liver Microsomal Vitamin D 3 25-Hydroxylase-The catalytic properties of the microsomal vitamin D 3 25-hydroxylase was studied with the native enzyme purified to apparent homogeneity from pig liver. The enzyme catalyzed 25-hydroxylation of various vitamin D compounds, at saturating concentrations (25 M), in the following order of efficiency: 1␣-hydroxyvitamin D 3 (760 pmol/nmol cytochrome P450/min) Ͼ vitamin D 3 (200 pmol/nmol cytochrome P450/ min) Ͼ vitamin D 2 (110 pmol/nmol cytochrome P450/min). In previous studies, the rate of hydroxylation by microsomal and mitochondrial vitamin D 3 25-hydroxylases, purified from rat and pig liver, has been analyzed with micromolar concentrations of vitamin D 3 , and apparent K m values of 5-10 M have been reported (2,10,18,19). Since the concentrations of vitamin D 3 Fig. 1 in a graph representing nonlinear regression analysis of average simulated rates of 25-hydroxylation as a function of substrate concentration. From the results of the present study, it can be concluded that the microsomal vitamin D 3 25-hydroxylase is able to work with physiological concentrations of vitamin D 3 .
Since the primary structure of the microsomal vitamin D 3 25-hydroxylase (CYP2D25) is 70 -80% identical with other members of the CYP2D subfamily (12), the activity of CYP2D25 toward some known CYP2D substrates was also studied. Debrisoquine is a commonly used marker substrate for CYP2D6, which catalyzes the 4-hydroxylation of this compound (21,22). Tolterodine, a muscarinic receptor antagonist, is converted into the 5-hydroxymethyl metabolite by CYP2D6 (17). Debrisoquine 4-hydroxylation was very low with the microsomal vitamin D 3 25-hydroxylase, less than 12 pmol/nmol P450/ min. Tolterodine, on the other hand, was efficiently converted into its 5-hydroxymethyl metabolite at a rate of 360 pmol/nmol/ min, which is comparable with that of 25-hydroxyvitamin D 3 The identity of the DNA products from liver, kidney, and spleen were verified by restriction enzyme analysis. Negative controls containing RNA without reverse transcriptase enzyme and without cDNA template, respectively, were included in each experiment and did not show any product formation. formation (Fig. 2). The results show that CYP2D25 is not specific for vitamin D 3 compounds. The very low activity toward debrisoquine is not unique to CYP2D25 among members of the CYP2D subfamily. Other CYP2D enzymes, such as those of the mouse and some of the rat CYP2D enzymes, have been reported to be inactive toward debrisoquine but catalyze hydroxylations of endogenous substrates, i.e. testosterone (21)(22)(23).
Effect of Various Cytochrome P450 Inhibitors on the 25-Hydroxylation of Vitamin D 3 -To obtain further information on the catalytic properties of the microsomal vitamin D 3 25hydroxylase, the effects of various P450 inhibitors were studied. The CYP2D6 substrate tolterodine inhibited the microsomal 25-hydroxylation of vitamin D 3 (25 M) by more than 80% at a 50 M concentration (Fig. 3). This marked inhibition is to be expected since tolterodine was shown to be a substrate also for the microsomal vitamin D 3 25-hydroxylase (cf. above). Tolterodine had no effect on the 25-hydroxylation catalyzed by the mitochondrial vitamin D 3 25-hydroxylase (CYP27A). Quinidine, an inhibitor of CYP2D6, inhibited the microsomal 25hydroxylation at most by 50% in concentrations up to 125 M (Fig. 3). Quinidine is a potent inhibitor of CYP2D6-catalyzed reactions in much lower concentrations, about 1 M (17,21). In this concentration, the inhibition of the 25-hydroxylation was less than 20% (Fig. 3). A lack of inhibition by quinidine has been reported for the N-demethylation of imipramine by CYP2D18 (24). Cyclosporin A (30 M), which has been reported to be an inhibitor of mitochondrial CYP27A-mediated 25-hydroxylation of vitamin D 3 (25), inhibited the microsomal 25hydroxylation by about 45%. For comparison, the 25-hydroxylation catalyzed by mitochondrial CYP27A was inhibited by 60% (data not shown).
Inhibitors of other cytochrome P450 subfamilies had no or very little effect on microsomal vitamin D 3  Northern Blot and RT-PCR Experiments on the Tissue Distribution of mRNA for Pig Microsomal CYP2D25-RNA from various pig tissues was isolated and examined in Northern blot experiments using the 32 P-labeled pig liver CYP2D25 cDNA as a probe (12). With this method, mRNA for CYP2D25 was detected in liver and kidney but not in adrenals, brain, heart, intestine, lung, muscle, spleen, or thymus (Fig. 4). The presence of mRNA expressing CYP2D25 in kidney is supported by previous studies reporting the purification from pig kidney of a microsomal enzyme active in 25-hydroxylation of vitamin D 3 (29,30).
The results of the Northern blot experiments showed that CYP2D25 is expressed predominantly in liver and in lower amounts in kidney. However, a low expression in other tissues could not be excluded. Indeed, experiments with RT-PCR revealed that mRNA for CYP2D25 is present not only in liver and kidney, but also in small amounts in adrenals, brain, heart, intestine, lung, muscle, spleen, and thymus (Fig. 5). CYP2D6 is predominantly expressed in liver but has been reported to be expressed also in other tissues (31,32). The CYP2D enzymes in rats are expressed both in liver and extrahepatic tissues (24,33).
Comparison of the Primary Structures of CYP2D25 and CYP2D6 - Fig. 6 shows the deduced amino acid sequence of porcine CYP2D25 and alignment with the sequence of human CYP2D6. The overall amino acid identity between the two proteins was 77%. The putative substrate recognition sites for CYP2 proteins (SRS-1 through SRS-6) described by Gotoh (34) were identified in CYP2D25 and compared with those in CYP2D6. These regions in CYP2D25 and CYP2D6 were not Comparison of the whole deduced sequences of pig CYP2D25 and human CYP2D6. Identical amino acid residues are indicated by shaded letters, while hyphens in the sequences indicate gaps introduced during the alignments. The sequences corresponding to the putative substrate recognition sites for CYP2 enzymes (SRS-1 through SRS-6) described by Gotoh (34) are underlined.
identical. The degree of identity between the substrate recognition sites of the two enzymes was as follows: 78% (SRS-1), 78% (SRS-2), 55% (SRS-3), 89% (SRS-4), 82% (SRS-5), and 79% (SRS-6). The highly conserved Asp-301 and Ser-304 in CYP2D6, which have been reported to play important roles in binding of substrate and inhibitor to this enzyme (22,23), are apparently corresponded by Asp-304 3 and Ser-307 in CYP2D25. The Asp-301 residue in CYP2D6 is considered to bind the basic nitrogen atom present in most CYP2D6 substrates. Since vitamin D 3 does not contain a basic nitrogen atom, it is likely that other residues in CYP2D25 are important for the binding of vitamin D 3 .
Since CYP2D25 and CYP2D6 have similar tissue distribution and ability to use tolterodine as substrate and similar although not identical structure, we considered it of interest to make further comparisons to examine whether CYP2D6 could be a vitamin D 3 25-hydroxylase in human liver.
Western Blot Experiments- Fig. 7 shows SDS-PAGE and immunoblotting analyses using antibodies raised against pig liver microsomal vitamin D 3 25-hydroxylase and human CYP2D6. The antibody against CYP2D6 did not react with purified pig microsomal vitamin D 3 25-hydroxylase but only recognized protein in the human liver microsomal samples containing CYP2D6. The antibodies against pig microsomal vitamin D 3 25-hydroxylase recognized the purified porcine enzyme but also protein of the same apparent M r as CYP2D6 in human liver microsomes.

25-Hydroxylation of Vitamin D 3 by Human Liver
Microsomes-Conflicting results have been reported concerning the ability of human liver microsomes to 25-hydroxylate vitamin D 3 . In an early report, data were presented that indicated that human liver microsomes do not possess vitamin D 3 25-hydroxylase activity (35). A subsequent paper reported that partially purified cytochrome P450 from human liver microsomes catalyzed 25-hydroxylation of both vitamin D 3 and 1␣-hydroxyvitamin D 3 (36). Since 1␣-hydroxyvitamin D 3 is more effectively 25-hydroxylated by purified 25-hydroxylase (10), this substrate was used in the experiments carried out to examine a possible correlation between 25-hydroxylase activity and CYP2D6 content in human liver microsomes. Liver microsome samples from nine different donors were separately incubated with 1␣-hydroxyvitamin D 3 . The 25-hydroxylase activity in the various microsome samples varied about 6-fold between 0.8 and 4.7 pmol/min/mg protein (Table I). There was no correlation between 25-hydroxylase activity and concentration of CYP2D6 in the microsome samples (r 2 ϭ 0.023). Thus, 25-hydroxylase  activity was detected in liver microsomes lacking or having very low concentrations of CYP2D6, and the microsome sample with the highest concentration of CYP2D6 showed the lowest 25-hydroxylase activity (Table I).
Experiments with Recombinantly Expressed Human CYP2D6 -CYP2D6, recombinantly overexpressed in insect microsomes, was incubated with 1␣-hydroxyvitamin D 3 for various periods of time. No 25-hydroxylated product could be detected in these incubation mixtures. In contrast, incubations with microsomal 25-hydroxylase and 1␣-hydroxyvitamin D 3 resulted in a time-dependent increase in the formation of 25hydroxylated product (Fig. 8). It may be concluded from these results that CYP2D6 is not a vitamin D 3 25-hydroxylase and that microsomal 25-hydroxylation in human liver is catalyzed by a cytochrome P450 enzyme different from CYP2D6.
The results of the present study describing structural and catalytical properties and tissue distribution of the porcine microsomal vitamin D 3 25-hydroxylase should be of value in studies on the identification and characterization of the human liver microsomal vitamin D 3 25-hydroxylase.