Introduction

Top Abstract Introduction Materials and methods Results Discussion References

Cytochrome P-450s (CYPs)¹ are a multigene superfamily of enzymes known to be involved in the oxidative metabolism of a diverse range of xenobiotics, therapeutic drugs, and endogenous steroid hormones (Gonzalez, 1988; Guengerich, 1995). A more recently identified member, P-450IB1 (CYP1B1), has been classified in the CYPI family (Savas et al., 1994; Sutter et al., 1994; Tang et al., 1996), which consists of two of the most intensively studied and well characterized CYPs, CYP1A1 and CYP1A2. CYP1A1 is an inducible CYP expressed in many human tissues except in liver, and CYP1A2 is known only as a hepatic CYP (Gonzalez, 1988; Guengerich, 1995). The newest member, CYP1B1, contains at least three orthologous forms. After the cloning of human CYP1B1, the orthologs of mouse and rat have been cloned and sequences have been reported (Savas et al., 1994; Walker et al., 1995). CYP1B1 has been shown to be a 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)-inducible gene in a human squamous cell carcinoma cell line SCC12 (Sutter et al., 1994; Tang et al., 1996), a human mammary carcinoma cell line MCF-7 (Christou et al., 1994), a human renal adenocarcinoma cell line ACHN (Stefan and Greenlee, 1997), a mouse fetal fibroblast cell line C3H 10T1/2 (Christou et al., 1993), and primary cultured rat adrenocortical cells (Brake and Jefocoate, 1995). In rats, CYP1B1 was induced in liver, kidney, and lung by TCDD (Walker et al., 1995). CYP1B1 also has been suggested to be constitutively expressed, because its mRNA has been detected in many human tissues (Sutter et al., 1994; Shimada et al., 1996), especially human kidney, prostate, mammary, ovary, and uterus (Sutter et al., 1994; Shimada et al., 1996). Recent observations that CYP1B1 mRNA is extremely high in human fetal kidney and fetal heart open the possibility that CYP1B1 may play a role during embryo differentiation and development (Shimada et al., 1996).

The distribution of CYP1B1 mRNA has led to the speculation of its possible physiological function: the metabolism of steroid hormones (Sutter et al., 1994). The accumulated evidence shows its involvement in the metabolism of 17-estradiol (E₂) (Spink et al., 1992; Hayes et al., 1996) and testosterone (Crespi et al., 1997). The preferential E₂ 4-hydroxylase activity of CYP1B1 has been shown and distinguishes it from the preferential E₂ 2-hydroxylase activity of CYP1A2 and the E₂ 6-, 15-, and 2-hydroxylase activities of CYP1A1 (Spink et al., 1992; Hayes et al., 1996). This positional specificity for E₂ hydroxylation can be used to distinguish CYP1B1-specific activity from that of CYP1A1 and CYP1A2.

CYP1B1 also has been shown to be involved in the mutagenic activation of many carcinogens (Sutter et al., 1994; Shimada et al., 1996). Moreover, it has been detected in many tumors such as human breast carcinoma (McKay et al., 1995), lung squamous carcinoma (Murray et al., 1997), astrocytoma (Murray et al., 1997), and mouse sarcoma (Christou et al., 1993), and its enhanced E₂ 4-hydroxylase activity has been reported to be associated with human uterine myoma and mammary tumors (Liehr et al., 1995; Liehr and Ricci, 1996). Although the distribution of CYP1B1 mRNA infers the expression of CYP1B1 protein, there are many examples of tissues expressing CYP mRNA, but not protein (Schweikl et al., 1993). Thus, to assess its role in carcinogen and steroid hormone metabolism, it will be necessary to detect and quantitatively measure CYP1B1 protein in various human tissues and to assay enzyme activity.

Efforts have been made to develop antibodies to detect CYP1B1. An antibody to mouse CYP1B1 also recognized human CYP1B1 but cross-reacted with purified rat CYP2A1 proteins (Pottenger et al., 1991; Christou et al., 1994). In addition, a polyclonal antibody to a fusion protein containing human CYP1B1 amino acid residues 166 to 349 has been reported to react with human CYP1B1 in immunoblots (Crespi et al., 1997). In both of these studies, cross-reactivity with human CYP1A1 and CYP1A2 was not examined. Also, a peptide-specific antibody targeted to the amino acids 332 to 345 of human CYP1B1, a hinge region between two -helices in the middle of human CYP1B1 protein, has been reported (Murray et al., 1997). This antibody reacts with human CYP1B1 but does not cross-react with human CYP1A1 nor any protein from human liver microsomes, which may be expected to contain CYP1A2 (Murray et al., 1997). However, this antibody did not detect CYP1B1 in any normal human tissue, which is expected according to Northern blot analysis results (Sutter et al., 1994; Shimada et al., 1996). Thus, a highly sensitive and specific antibody is essential to study the localization and quantitation of CYP1B1 protein in different human tissues. Because the CYPs are a superfamily with many homologous isozymes and CYP1B1 reveals a 41% homology with human CYP1A1 and 40% homology with human CYP1A2 (Jaiswal et al., 1985, 1987; Sutter et al., 1994), the peptide-specific antibody strategy was chosen to raise the antibody and minimize or eliminate the potent cross-reactivity with other closely related P-450s.

	Materials and Methods

Top Abstract Introduction Materials and methods Results Discussion References

Materials. Cell microsomes containing recombinant human CYP1B1 were purchased from GENTEST Corp. (Woburn, MA), 17-estradiol [6,7-³H(N)] (50.0 Ci/mmol) was obtained from DuPont NEN (Boston, MA), protein A-Sepharose beads were purchased from Pharmacia Biotech (Piscataway, NJ), and Affi-Gel 10 and 15 were purchased from Bio-Rad (Hercules, CA). Purified human CYP1A1 protein and human CYP1A2 protein were kind gifts from Dr. F. P. Guengerich (Vanderbilt University, Nashville, TN). 4-Hydroxyestradiol (4-OH E₂) and other E₂ derivatives, 2-OH E₂, 6-OH E₂, and 16-OH E₂, used in high-performance liquid chromatography (HPLC) as standards were obtained from Steraloids Inc. (Wilton, NH). 5-OH E₂ was a kind gift from Dr. Vhagu Bhavnani through Steraloids Inc.

Peptides Selection and Antiserum Production. Unique sequence regions of human CYP1B1 distinguishable from some major human CYPs were selected as potential antibody targets by sequence comparison (Jaiswal et al., 1985, 1987; Sutter et al., 1994). The antigenicity and the surface possibility of these regions were analyzed and predicted by computer software package MacVector (International Biotechnologies, New Haven, CT). The formula used to calculate the hydrophilicity, antigenicity, and surface probability of CYP1B1 peptide was Kyte-Doolittle method over a window of seven residues. The possibility of selected regions targeting the enzyme active site has been referenced with previous studies of other P-450 members (Edwards et al., 1989, 1990, 1993; Wang and Lu, 1997). The peptides corresponding to the selected regions were synthesized on a peptide synthesizer, Advanced Chemtech, using the solid-phase chemistry by Alpha Diagnostic Inc. (San Antonio, TX). The synthesized peptides were coupled to the carrier protein, keyhole limpet hemocyanin (KLH), through the cysteine residue at the N terminus of each peptide using N-succinnidyl bromoacetate as a cross-linking reagent (Bermatowicz and Matsuedon, 1986; Wang and Lu, 1997).

Microsomal Protein Preparation and Immunoblotting. Human tissues (bladder, breast, colon, kidney, liver, lung, ovary, prostate, testis) were minced and suspended in a buffer containing 0.1 M sodium phosphate buffer (pH 7.25), 0.1 mM dithiothreitol, and 10 mM EDTA, containing the proteinase inhibitors aprotinin (0.1 mg/ml), leupeptin (0.1 mg/ml), and phenylmethylsulfonyl fluoride (0.1 mg/ml). After tissue homogenization with a Polytron (Brinkmann, Westbury, NY) and centrifugation at 15,000g for 20 min at 4°C, the supernatant was transferred to a new centrifugation tube and centrifuged again at 105,000g for 60 min at 4°C. The pellet (microsomal protein) was resuspended in the storage buffer containing 0.1 M potassium phosphate buffer, pH 7.25, containing 10 mM EDTA, 20% glycerol, 0.1 mM dithiothreitol, and 0.25 mM phenylmethylsulfonyl fluoride. The protein concentration was determined by Bio-Rad Protein Assay and referenced to the standard curve of bovine serum albumin (BSA). Microsomal protein, in the amounts indicated, was subjected to a 10% SDS-polyacrylamide gel electrophoresis (PAGE), and the banded protein was transferred to a nitrocellulose membrane using a standard procedure (Anderson and Blobel, 1983). After blocking in 5% milk in TBST (Tris-buffered saline Tween 20: 100 mM Tris-HCl, pH 7.8, 0.9% NaCl, and 0.025% Tween 20), the nitrocellulose membrane was incubated with the first antiserum at 400- to 800-fold dilutions or with the purified antibody, followed with the incubation of the secondary antibody conjugated with horseradish peroxidase (1:20,000 dilution; Pierce, Rockford, IL). A SuperSignal enhanced chemiluminescence (ECL) kit (Pierce) was used to visualize the immunodetected bands.

Immunoprecipitation. Immunoprecipitation was performed following a standard protocol (Kessler, 1975; Anderson and Blobel, 1983). Microsomal protein (100 µg of total protein in a 10-µl volume) containing human CYP1B1 (6.2 pmol) was diluted with 190 µl of dilution buffer (10 mM Tris-Cl, pH 8.0, 140 mM NaCl, 0.1% Triton X-100, and 0.1% bovine serum) and incubated with 2 µl of antiserum (or 2 µl of preimmune serum as control), rotated slowly on a rotator (ATR, Laurel, MD) for 2 h at room temperature. A 40-µl sample of protein A-Sepharose beads (Pharmacia Biotech), diluted 1:1 with dilution buffer, was then added, and the mixture was rotated further for 20 h at 4°C. The beads then were washed twice with 0.1% Triton X-100 in TSA solution (0.01 M Tris-HCl buffer, pH 8.0, 0.14 M NaCl), once with TSA solution, and once with 50 mM Tris-HCl buffer, pH 6.8. The bound protein was eluted from beads by adding 30 µl of SDS/sampling buffer, containing 62.5 mM Tris-HCl buffer, pH 6.8, 10% glycerol, 2% SDS, and 0.0005% bromophenol blue, incubated at 100°C for 5 min, and subjected to SDS-PAGE analysis. Protein bands were visualized by immunodetection with alkaline phosphatase staining.

CYP1B1 Enzyme Activity Assays by HPLC and Antibody Immunoinhibition Experiments. Recombinant human CYP1B1 protein (12 pmol; 200 µg of protein) was added to a reaction mixture (final volume, 0.5 ml) containing 50 µM [³H]E₂ (50 mCi/mmol), 0.1 M sodium phosphate buffer (pH 7.4), 1.0 mM NADPH, 1 mM MgCl₂, and 1 mM ascorbic acid and was incubated at 37°C for 20 min in air with vigorous shaking. The reaction was stopped and then extracted with dichloromethane (2.0 ml × 2). The organic extracts were pooled and dried in a vacuum centrifuge. The dried residues were dissolved in 40 µl of methanol containing 1.2 µg of 4-OH E₂ as a UV marker, and 20 µl of sample was injected onto an HPLC equipped with a Beckman Ultrasphere column (5 µm, 4.6 mm × 25 cm). The mobile phase consisted of 0.05% acetic acid and methanol. During the gradient elution, the methanol composition was changed from 60 to 90% over 7 min, the flow rate was 1.1 ml/min, and the elutes were detected by UV at 280 nm. The peak of 4-OH E₂ was identified by its retention time and the spectral characteristics in comparison with standard compounds. The radioactivity was detected by a flow scintillation analyzer from Packard (Downers Grove, IL).

Chemically Induced Expression of Human CYP1B1 in ACHN Cells. Human renal adenocarcinoma cell line (ACHN) was cultured in minimal essential medium containing 10% fetal bovine serum (HyClone, Logan, UT) to 90% confluence and then treated with various chemicals (0.1% v/v) for 24 h. After the cells were collected, microsomes were prepared as described earlier in the text. Protein (20 µg) was applied to each lane of a 10% SDS-PAGE gel and followed the immunodetection as described previously. The density of each band was measured by an image analyzer, Alpha Imager 2000 (Alpha Innotech, San Leandro, CA), and the density of control band was normalized to 1.0.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

Selection of CYP1B1-Specific Peptide. Using the amino acid sequence data (Fig. 1A) deduced from the human CYP1B1 cDNA sequence (Sutter et al., 1994), two peptides were selected for antigenic epitopes with amino acid sequences significantly different from human CYP1A1 and CYP1A2 (Fig. 1B). Both peptides were predicted to have high hydrophilicity, antigenicity, and surface probability by computer modeling (Fig. 2). The first peptide, ME001 (CDFRANPNEPAKMN), corresponds to the amino acid sequence 491 to 504 and is located at the C terminus of human cytochrome CYP1B1. The second peptide, ME002 (CFGCRYSHDDPEF), corresponds to the amino acid sequence 209 to 221 of human cytochrome CYP1B1 and is located in a hinge region between two -helices, as predicted by a model study and the sequence alignment of several CYPs (Edwards et al., 1989).

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Peptide designed from the human CYP1B1 amino acid sequence. A, deduced amino acid sequence of human CYP1B1 and the localization of the regions chosen for peptide synthesis in the development of antipeptide antibodies. B, sequence comparison of the peptide regions among related P-450s. - represents the amino acid residue comparable to that of human CYP1B1.

View larger version (69K): [in this window] [in a new window]   Fig. 2.   Hydrophilicity, surface probability, and antigenicity of human CYP1B1 and the regions selected for antibody production. The amino acid sequence of human CYP1B1 was analyzed by the computer software MacVector, and the hydrophilicity, surface probability, and antigenicity of the whole molecule were predicted. The selected regions are as indicated.

Characterization of Peptide and Antisera. The purity of peptides ME001 and ME002 was more than 85% and 80%, respectively, as determined by HPLC analysis. These two peptides were coupled to the carrier protein, KLH, via the cysteine residue at the N terminus of each peptide and used to immunize New Zealand White rabbits. The raised antiserum presented titers as high as 10⁴ to 10⁵, as assayed by an enzyme-linked immunosorbent assay (ELISA) against antigen peptides provided by Alpha Diagnostic (data not shown). One of the raised antisera, against peptide ME001, demonstrated a high affinity for CYP1B1 in immunoblots (Fig. 3). This antibody detected amounts of recombinant human CYP1B1 protein as low as 0.34 ng (6.4 × 10³ pmol) in an immunoblot assay (data not shown). In an immunoblotting analysis of microsomal protein from human tissues, the antibody detected a single protein band with the mobility the same as the recombinant human CYP1B1 (Fig. 3) in the molecular mass region corresponding to other CYPs (43-60 kDa) (Guengerich, 1994). The apparent molecular mass of this protein thus was estimated as 56 kDa. Using the crude antiserum, another unknown protein band with a apparent molecular mass of 85 kDa was visualized on the immunoblots. In contrast, using the antibody purified by an affinity column described in Materials and Methods, only the 56-kDa band was observed. This antibody did not cross-react with either human CYP1A1 or human CYP1A2 at a relative high level (60 ng or 1 pmol) of protein (Fig. 3). The ME001 antibody also cross-reacted with mouse CYP1B1 and rat CYP1B1 as shown on the immunoblot (Fig. 3). Thus, the sensitivity and selectivity of this antibody are high and suitable for immunoblotting applications.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Immunodetection of CYP1B1 protein with the peptide-specific antibody against peptide ME-001. Lanes 1 to 4, recombinant human CYP1B1 protein, 180 ng, 60 ng, 20 ng, and 7 ng, respectively; lanes 5 and 6, human ovary microsomal protein, 20 µg; lane 7, human prostate microsomal protein, 20 µg; lane 8, human breast microsomal protein, 20 µg; lane 9, rat testis microsomal protein, 60 µg; lane 10, rat ovary microsomal protein, 60 µg; lane 11, mouse testis microsomal protein, 60 µg; lane 12, human CYP1A1 protein, 60 ng; lane 13, human CYP1A2 protein, 60 ng.

Immunodetection of Native CYP1B1 Protein and Immunoinhibition. In immunoprecipitation experiments, the observation that the recombinant human CYP1B1 protein was precipitated by the antiserum but not by the preimmune serum reveals clearly that this antibody reacts with undenatured human CYP1B1 protein (Fig. 4). This fact confirms that this selected target region is on the surface of human CYP1B1 molecule, as predicted by computer analysis.

View larger version (41K): [in this window] [in a new window]   Fig. 4.   Immunoprecipitation of human CYP1B1 protein with the peptide-specific antiserum. Lane 1, recombinant human CYP1B1 protein, 50 ng; lane 2, immunoprecipitation with preimmune serum; lane 3, immunoprecipitation with the peptide-specific antiserum.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   HPLC analysis of hydroxylation of [3H]E2 catalyzed by recombinant human CYP1B1. Experiments were performed as described in Materials and Methods. [3H]E2 (50 µM; 50 mCi/mmol) was incubated with 12 pmol of recombinant human CYP1B1 at 37°C for 20 min. The metabolites were extracted and analyzed by HPLC. Unlabeled 4-OH E2 was added to the extracted metabolites as a marker. A, HPLC profile for control sample without recombinant human CYP1B1. B, HPLC profile for sample with 12 pmol of recombinant human CYP1B1. C, UV absorbance at 280 nm.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Lack of immunoinhibition of E2 hydroxylation catalyzed by recombinant human CYP1B1 with CYP1B1 peptide-specific antiserum. Recombinant human CYP1B1 (12 pmol) was incubated with the indicated amounts of peptide-specific antiserum for 20 min at 37°C, followed by the CYP1B1 activity assay described in Materials and Methods and in Fig. 4. The metabolite 4-OH E2 has been normalized based on the total cpm. Values represent the mean of two experiments. Interexperimental variability was less than 25%.

Detection of CYP1B1 Protein in Various Human Tissues. Using the purified ME001 antibody, a single CYP1B1 protein band was detected in all nine human tissue microsomes tested (Fig. 7). They were kidney, testis, breast, prostate, bladder, colon, liver, lung, and ovary, respectively. Four individual samples of lung, colon, bladder, and liver and two individual samples of ovary were examined by immunoblotting, and the results shown were representative of the other determinations. From the tissue microsomes surveyed, the highest level of CYP1B1 protein appeared to be in the kidney and bladder, followed by breast and lung. The levels of CYP1B1 observed in liver microsomes were very low and approached the limit of detection (Fig. 7). As a control, preimmune serum, which does not recognize CYP1B1, was used similarly and failed to detect any bands.

View larger version (31K): [in this window] [in a new window]   Fig. 7.   Detection of CYP1B1 protein from several human tissues. Twenty micrograms of microsomal protein prepared from nine different human tissues (see Materials and Methods) has been applied to a 10% SDS-PAGE gel and analyzed on an immunoblot. The affinity column-purified first antibody was at 1:40 dilution, and the second antibody was anti-rabbit IgG conjugated with horseradish peroxidase (1:20,000 dilution; Pierce). The lanes are: 1, recombinant human CYP1B1 (5 ng); 2, kidney; 3, testis; 4, breast; 5, prostate; 6, bladder; 7, colon; 8, liver; 9, lung; and 10, ovary. Three to four different individual samples of liver, lung, and colon and two different individual samples of ovary and breast have been examined and comparable results have been obtained.

Detection of CYP1B1 Protein in Chemically Induced Cells. TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin) and benz[a]anthracene have been shown to be strong inducers of CYP1B1 protein in several cell lines, including the ACHN human renal adenocarcinoma cell line (Christou et al., 1993, 1994; Stefan and Greenlee, 1997). In our experiments, we incubated ACHN cells with these chemicals as well as with several other known inducers of CYP1A1 and CYP1A2. Both TCDD and benz[a]anthracene appeared to be the strongest inducers (9- to 10-fold) of CYP1B1 protein among these chemicals (Fig. 8). In addition to TCDD and benz[a]anthracene, 3-methylcholanthrene and -naphthoflavone were moderate inducers, whereas anthramine and phenobarbitone, which have been shown as strong inducers for other CYPs (Gonzalez, 1988), showed little or no induction at CYP1B1 protein in this system.

View larger version (48K): [in this window] [in a new window]   Fig. 8.   Chemically induced expression of human CYP1B1 in ACHN cells. ACHN was cultured to 90% confluence and then treated with various chemicals (0.1% v/v) for 24 h. Samples of microsomal protein (20 µg) were resolved by a 10% SDS-PAGE gel and transferred to nitrocellulose for immunodetection as described in Materials and Methods. Lane 1, recombinant human CYP1B1, 3.3 ng. Pretreatments with: lane 2, dimethyl sulfoxide control (vehicle solvent); lane 3, TCDD, 10 nM; lane 4, benz[a]anthracene, 25 µM; lane 5, 3-methylcholanthrene, 25 µM; lane 6, -naphthoflavone, 25 µM; lane 7, anthramine, 25 µM; and lane 8, phenobarbitone, 25 µM. The density of each band was measured by an image analyzer, and the density of the dimethyl sulfoxide control band was normalized to 1.0.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

A peptide-specific polyclonal antibody was targeted successfully to CYP1B1 by immunizing rabbits with a conjugate of the peptide CDFRANPNEPAKMN, corresponding to residues 491 to 504 at the C terminus of human CYP1B1. The developed antibody revealed a relatively high sensitivity and selectivity for immunodetection of CYP1B1. This antibody bound to undenatured CYP1B1, confirming that the selected antigen epitope is surface-located. Nevertheless, the lack of inhibition of this antibody toward CYP1B1 activity suggested it was not near enough to the enzyme active site.

The CYP enzyme active site has been predicted to be located near the C terminus by analyzing the consensus sequences of major CYP members (Gonzalez, 1988). Amino acid sequence alignment of most CYPs revealed that this enzyme active site consists of a number of conserved amino acids surrounding a conserved cysteine residue. This cysteine residue has been demonstrated unequivocally as the heme iron-binding thiolate ligand by X-ray crystallography (Poulos et al., 1986; Gonzalez, 1988). According to this model, the corresponding residues for the enzyme active site of human CYP1B1 would be located at positions 463 to 488 and surround the conserved heme iron-binding thiolate ligand Cys⁴⁷⁰. Our antigen epitope sequence (ME001) was designed to cover residues 491 to 504, near this enzyme active site but downstream by a few amino acid residues, to distinguish the CYP1B1 sequence from that of CYP1A1 and CYP1A2 (Fig. 1). Nevertheless, this antibody turned out to be noninhibitory toward CYP1B1-specific activity.

The immunoinhibitory potency of this antibody was evaluated by using recombinant human CYP1B1 and increasing levels, in a manner similar to previous reports (Cooper and Peterson, 1994). Based on the estimated amount of CYP1B1 (12 pmol) and antisera (40-80 µg IgG protein or 270-530 pmol CYP1B1-specific antibody), the molecular ratio of antibody to antigen was about 20 to 40:1. This molecular ratio should have been sufficient to inhibit CYP1B1 enzyme activity if it had been bound near enough to the enzyme active site of CYP1B1. No decline of enzyme activity was observed corresponding to the increasing amount of antiserum (Fig. 6). Thus, this antibody was judged to exhibit a lack of anti-CYP1B1-inhibitory activity.

Although this antibody was not inhibitory, its ability to bind to CYP1B1 protein, with no cross-reaction toward human CYP1A1 or CYP1A2 in Western blot analysis, should make this antibody highly useful for studying the expression level of CYP1B1 protein. Many other noninhibitory antibodies have been used in this manner (Edwards et al., 1989, 1998; Shen and Strobel, 1995). Moreover, the antibody revealed cross-reactivity to mouse CYP1B1 and rat CYP1B1 (Fig. 3), which also should make this antibody highly useful for detecting rodent CYP1B1 in animal experiments. Because there were 6 to 7 amino acid differences (Fig. 1) between the human peptide antigen and that of the rat and mouse, this result was somewhat unexpected. However, the N-terminal amino acid differences between human and rodents (CDFRAN versus CNFKAN) in these peptides are structurally similar, suggesting that this area may be the primary epitope recognized by the antibody preparation.

The 4-hydroxylation activity of E₂ has been used to measure CYP1B1-specific activity. An HPLC method was used to provide improved identification and quantification of CYP1B1 E₂ 4-hydroxylation activity. Under our HPLC conditions, all of the standards (E₂, 4-OH E₂, 2-OH E₂, 6-OH E₂, 16-OH E₂, and 5-OH E₂) were well separated (data not shown). We determined the E₂ 4-hydroxylase activity mediated by the recombinant human CYP1B1 expressed in human lymphoblasts to be 2.5 pmol/min/mg microsomal protein (Fig. 5). This value is lower than the activity of a recombinant human CYP1B1 expressed in yeast cell (20 pmol/min/mg microsomal protein) (Hayes et al., 1996) but is much higher than the activity assayed in TCDD-induced human renal adenocarcinoma cells (0.7 pmol/min/mg microsomal protein) (Spink et al., 1997).

Compared with the detection of CYP1B1 by immunoblot method, the sensitivity of enzymatic detection of CYP1B1 is much lower. To detect CYP1B1 enzymatically, 12 pmol of recombinant human CYP1B1 was used in each assay, and only a small peak corresponding to 4-OH E₂ could be reliably detected by HPLC, whereas, in immunoblot analyses (Fig. 7), 5 ng (0.088 pmol) could be readily detected, showing an intensely stained band, which is at least 140-fold more sensitive than the enzyme activity assay for CYP1B1. Moreover, CYP1A1 and 1A2 show some 4-hydroxylation activity on E₂, although the major activity is 2-OH, 15-OH, and 6-OH-E₂ formation for CYP1A1 and 2-OH E₂ formation for CYP1A2 (Spink et al., 1992). In addition, recent studies have revealed that other hepatic CYPs, CYP3A4 and CYP2C9, are also capable of hydroxylating estradiol at the C-4 position (Yamazaki et al., 1998). Because it is now clear that there are multiple catalytic activities for the 4-hydroxylation activity of E₂, distinguishing CYP1B1 activity becomes problematic. Although the use of CYP-selective inhibitors merit further study (Shimada et al., 1997), immunodetection of CYP1B1 clearly provides a promising tool to study CYP1B1 tissue distribution and expression levels at this time.

Another peptide epitope, ME002 (CFGCRYSHDDPEF), corresponds to the amino acid residues 209 to 221 of human CYP1B1 and is located in a hinge region between two -helices, predicted by a model study and the sequence alignment of several CYPs (Edwards et al., 1989). It failed to raise an antibody against human CYP1B1. One of the possible explanations for this failure is the presence of two cysteine residues in this 13-mer peptide. We cannot exclude the possibility of formation of an intramolecule disulfide bond that blocked the linkage of peptide to carrier protein KLH or inappropriately changed the antigen-peptide configuration.

With this antibody, we detected CYP1B1 protein in various nontumor tissues (Fig. 7). This result is significant and is inconsistent with a previous study that found CYP1B1 protein only in tumor tissues and not in surrounding tissue (Murray et al., 1997). However, CYP1B1 mRNA has been detected in many normal human tissues by Northern blot analysis (Shimada et al., 1996) and by reverse transcription-polymerase chain reaction approaches (Huang et al., 1996; Hakkola et al., 1997; Vadlamuri et al., 1998). Thus, our immunodetection of CYP1B1 protein in nontumor tissues is consistent with the observations of CYP1B1 mRNA in several human tissues (Shimada et al., 1996). The constitutive expression and wide distribution of CYP1B1 protein supports the speculation of an important hormonal role(s), although its precise mechanism is not fully clear at this time. However, the higher levels of CYP1B1 expression in kidney and bladder are striking and may suggest that this CYP plays some role in corticosteroid metabolism and should be investigated further.

The variation of the level of CYP1B1 mRNA in nontumor breast tissue has been reported using reverse transcription-polymerase chain reaction (Huang et al., 1996; Hakkola et al., 1997; Vadlamuri et al., 1998). In this study, we evaluated the relative level of CYP1B1 protein in various tissues from multiple individuals by immunoblotting. Although only four individual samples each of lung, colon, bladder, and liver and two individual samples of ovary were surveyed, the levels of CYP1B1 protein within each tissue type were comparable. With this caveat, the tissues containing the highest level of CYP1B1 protein appeared to be the kidney and bladder, followed by the breast and lung. The tissue with lowest level of CYP1B1 protein was liver (Fig. 7). CYP1B1 protein was detected in all four liver samples, although its amount was the lowest among these tissues surveyed.