Introduction

Top Summary Introduction Materials & Methods Results Discussion References

PAHs are environmental contaminants formed during incomplete combustion. Human exposure to these compounds occurs primarily through cigarette smoking, the inhalation of polluted air, and ingestion of charbroiled foods (Davila et al., 1995). These compounds are known to be carcinogens (Pelkonen and Nebert, 1982) and cause immunosuppression in laboratory animals (White et al., 1994; Davila et al., 1995). DMBA is one of the most potent carcinogenic and immunosuppressive PAHs. Mice treated with DMBA exhibit signs of immunotoxicity that include reduced spleen, bone marrow (Ward et al., 1984), and thymus cellularity (Thurmond et al., 1987) and decreased resistance to Listeria monocytogenes infection (Ward et al., 1984) and tumor growth (Dean et al., 1986). Although these in vivo studies demonstrate that DMBA cause a generalized reduction of immune cells that results in decreased immune surveillance, they do not provide a mechanistic explanation of how DMBA produces these effects.

Recent reports suggest that the bone marrow toxicity of DMBA may be dependent on the AhR in BMS cells (Yamaguchi et al., 1997a, 1997b). DMBA treatment of pre-B cells cultured with BMS cells caused apoptosis of pre-B cells, whereas DMBA treatment of pre-B cells in the absence of BMS cells did not. The cytochrome P450 and AhR antagonist -naphthoflavone blocked BMS cell-dependent pre-B cell apoptosis. Because neither BMS cells nor pre-B cells express CYP1A1 and only BMS cells express the AhR, the authors postulated that BMS cell AhR activation was necessary for DMBA-induced pre-B cell apoptosis. However, DMBA is a relatively weak AhR ligand (Bigelow and Nebert, 1982), and the potent AhR ligand TCDD did not cause pre-B cell apoptosis. These findings suggest that an event other than AhR activation is involved in DMBA-induced pre-B cell apoptosis.

Metabolism of DMBA may play a role in BMS cell-dependent pre-B cell apoptosis. It is well established that metabolism of PAHs is required for them to be carcinogenic (Pelkonen and Nebert, 1982) and that a correlation exists between the carcinogenic and immunotoxic potential of PAHs (White and Holsapple, 1984; White et al., 1985). From this relationship, it seems likely that PAH metabolism is required for immunosuppression and that the most potent toxic effects would be produced by metabolism within the target tissue. In accordance with this paradigm, bone marrow cells (Heidel et al., 1997; O'Dowd, 1987) and splenic microsomes (Kawabata and White, 1989) metabolized DMBA. In addition, the 3,4-dihydrodiol metabolite of DMBA was 65-fold more potent than DMBA in causing a reduction in the number of antibody-producing splenic B cells (Ladics et al., 1991). Furthermore, the cytochrome p450 antagonist -naphthoflavone prevented the reduction of antibody-producing B cells generated by DMBA (Ladics et al., 1991). These data provide convincing evidence that cytochrome P450-dependent metabolism of DMBA was required for mature B cell toxicity. However, a requirement for metabolism of DMBA in bone marrow progenitor B cell toxicity remains to be demonstrated.

In the current study, we sought to better understand the potential contributions of cytochrome P450-dependent metabolism and AhR activation in DMBA-induced bone marrow toxicity. To accomplish this, we characterized BMS cell cytochrome P450 expression, DMBA metabolism, and AhR activation.

	Materials and Methods

Top Summary Introduction Materials & Methods Results Discussion References

Animals. Dr. Albee Messing (University of Wisconsin, Madison, WI) generously donated C57Bl/6 mice for this study. Mice heterozygous for a disrupted AhR gene were generated (Schmidt et al., 1996) and provided by Dr. Chris Bradfield (University of Wisconsin, Madison, WI). Breeding of AhR heterozygous adult mice produced pups that were null, heterozygous, or homozygous for the wild-type AhR allele. Pups were genotyped by PCR amplification of tail DNA for the presence of a 669-bp AhR gene product (AhR^+/+, 453-1122), a 459-bp neomycin gene product (AhR^/, 782-1240), or both (AhR^+/). The 3-4-week-old pups were used to establish primary BMS cell cultures.

Antibody and cDNA probes. Rabbit antibodies to mouse AhR and mouse ARNT were generously provided by Dr. Rick Pollenz (University of South Carolina, Charleston, SC). Monoclonal anti-GAPDH was purchased from Biodesign International (Kennebunk, ME). Rabbit antibodies to CYP1B1 and CYP1A1 were prepared in our laboratory as described previously (Pottenger and Jefcoate, 1990). All primary antibodies were used at a concentration of 1 µg/ml in antibody dilution buffer [10 mM Tris·HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween 20 (v/v)]. A 1028-bp mouse CYP1B1 probe was obtained by SmaI restriction endonuclease digestion of a cDNA-containing plasmid (Savas et al., 1994). Mouse GAPDH cDNA probe was prepared by PCR amplification of control DNA using primers purchased from Stratagene (La Jolla, CA).

Cell culture and treatments. BMS2 cells were provided generously by Dr. Paul Kincade (Oklahoma Medical Research Foundation, Oklahoma City, OK) (Pietrangeli et al., 1988). C3H/10T1/2 cells were purchased from the American Type Culture Collection (Rockville, MD). BMS2 cells were grown in RPMI1640 with 5% FBS (v/v) (Intergen, Purchase, NY) and C3H10T1/2 cells were grown in Dulbecco's modified Eagle's medium with 7% FBS. All media were supplemented with 5 × 10⁵ M 2-mercaptoethanol, 2 mM L-glutamine, 50 IU/ml penicillin, and 50 mg streptomycin/ml. When monolayers of BMS2 cells and C3H10T1/2 cells were 70-80% confluent, the conditioned media were removed, and the cells were incubated with fresh media containing 10 µM DMBA [or 0.1% DMSO (v/v) or 10 nM TCDD as controls] for the times indicated in the figure legends.

Preparation of microsomes and cytoplasmic and nuclear extracts. CYP1B1 and CYP1A1 Western immunoblots were performed on microsomes isolated as described previously (Pottenger and Jefcoate, 1990). AhR and ARNT Western immunoblots were performed on cytosolic and nuclear fractions isolated from BMS2 cells. To prepare the fractions, BMS2 cells were treated for 1 hr with 10 µM DMBA, 10 nM TCDD, or 0.1% DMSO vehicle control; washed with ice-cold phosphate-buffered saline; and removed from the flask with a cell scraper. After centrifugation at 500 × g, BMS2 cell pellets were resuspended in lysis buffer [25 mM 3-(N-morpholino)propanesulfonic acid, pH 7.4, 0.02% NaN₃ (w/v), 10% glycerol (v/v), 1 mM Na₂EDTA, 5 mM EGTA, 0.5% Tween-20 (v/v), 2 mM NaVO₄, 1 mM NaF, 20 mM Na₂MoO₄, 11.7 µM leupeptin, 100 units/ml aprotinin, 5 µg/ml soybean trypsin inhibitor, and 27 µM 1-chloro-3-tosylamido-7-amino-L-2-heptanone] and incubated on ice for 30 min. The supernatant from a 325 × g centrifugation was collected and designated the cytosolic fraction. The remaining pellet (nuclear fraction) was washed three times with cold lysis buffer and sonicated on ice (four 15-sec bursts at 40% power, Sonicator Cell Disrupter; Heat Systems-Ultrasonics, Plainview, NY). Protein concentrations were determined according to the bicinchoninic acid method (Pierce Chemical, Rockford, IL).

Western immunoblots. Total cell, microsomal, cytoplasmic, or nuclear proteins were resolved in a 0.75-mm 7.5% SDS-polyacylamide gel according to standard methods (Sambrook et al., 1989), and transferred to nitrocellulose membrane (Amersham, Arlington Heights, IL) using a Hoeffer TE51 transfer apparatus at 500 mA for 1 hr. The membranes were incubated for 2 hr with primary antibody (1 µg/ml), and the immunoreactive proteins were visualized with the enhanced chemiluminescence detection method (Amersham). Immunoblots that were probed with more than one primary antibody were incubated in stripping solution (100 mM 2-mercaptoethanol, 62.5 mM Tris·HCl, pH 6.7, 2% SDS) for 30 min at 50° and washed twice (10 mM Tris·HCl, pH 8.0, 150 mM NaCl, 0.1% Tween 20) before incubation with a different primary antibody. Immunoblot signals were quantified using a Molecular Dynamics (Sunnyvale, CA) Personal Densitometer SI and ImageQuant software.

RNA isolation and Northern blot analysis. Total RNA was isolated, quantified, and electrophoresed through a 1% agarose gel containing formaldehyde according to standard methods (Sambrook et al., 1989). RNA was transferred from the gel to Hybond-N⁺ membrane (Amersham) and fixed by baking for 2 hr at 80°. cDNA probes were labeled with -³²P-dCTP (3000 Ci/mmol) using Prime-a-Gene (Promega, Madison, WI) and purified with G50 Nick Columns (Pharmacia, Piscataway, NJ). Prehybridization of the membranes was performed for 2 hr at 42° in a buffer of 750 mM NaCl, 75 mM Na₃ citrate, pH 7.0, 50% deionized formamide (v/v), 0.1% SDS (w/v), 100 µg of denatured salmon sperm DNA/ml, 0.5% Ficoll 400 (w/v), 0.5% polyvinylpyrrolidone (w/v), and 0.5% bovine serum albumin (w/v). Prehybridization and hybridization were carried out in the same buffer, with the exception that a ³²P-labeled probe was added for hybridization. Blots were washed using standard methods (Sambrook et al., 1989), and the amount of specific ³²P-labeled cDNA hybridization was quantified using a Molecular Dynamics PhosphorImager and ImageQuant software.

Metabolism assays. To assess the effect of prior cytochrome P450 induction on DMBA metabolism, cells were treated with 10 nM TCDD for 24 hr at 37°. Control cells were treated with 0.1% DMSO (vehicle control). Conditioned medium was removed and replaced with fresh medium containing 1 or 10 µM DMBA. After a 1-hr incubation at 37°, the medium containing DMBA and metabolites was removed from the cells and incubated for 2 hr at 37° with -glucuronidase solution (2000 IU of -glucuronidase/ml, 0.5 M sodium acetate, pH 5.2, and 5.7 mM ascorbic acid). Cortisol was added to each sample as an internal standard, and the DMBA metabolites were extracted with ethyl acetate/acetone-containing dithiothreitol (at ratios of 2:1:0.003). The solvent phase was dried down under nitrogen gas and resuspended in 100 µl of methanol. Separation and quantification of DMBA metabolites were performed by high performance liquid chromatography analysis as described previously (Savas et al., 1993). Metabolic activities were normalized to the number of cells present in each well after completion of the DMBA metabolism assay. Cells were detached by treating them for 1-5 min with 0.05% trypsin (w/v) in calcium and magnesium-free Hanks' balanced salt solution, and the trypsin was neutralized by adding RPMI media containing 5% FBS. Cells were enumerated using a hemacytometer.

	Results

Top Summary Introduction Materials & Methods Results Discussion References

BMS cells express CYP1B1. Western immunoblot analysis demonstrated that the BMS2 mouse BMS cell line constitutively expresses CYP1B1 protein (Fig. 1). BMS2 cells had ~4-fold greater constitutive levels of 55-kDa CYP1B1 protein than the C3H10T1/2 mouse embryo fibroblast cell line, which are known to constitutively express CYP1B1 (Pottenger and Jefcoate, 1990). Treatment of BMS2 cells with DMBA or TCDD increased CYP1B1 protein by only 1.3-fold (1.3 ± 0.1 as an average of three separate experiments) and 1.4-fold (1.4 ± 0.3), respectively, relative to the vehicle control (0.1% DMSO). Incubation with DMSO had no effect compared with BMS2 cells incubated in medium alone (data not shown). In agreement with previous reports (Pottenger and Jefcoate, 1990; Savas et al., 1993), CYP1B1 protein increased by 5- and 6-fold, respectively, in C3H10T1/2 cells incubated with DMBA or TCDD. These induced levels of CYP1B1 in C3H10T1/2 cells were comparable to those in BMS2 cells after induction with DMBA or TCDD. CYP1A1 protein was not detected in either cell line.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Immunoblot analysis of CYP1B1 in BMS2 cells and C3H10T1/2 cells. Microsomal proteins were isolated after 24-hr treatments with 0.1% DMSO vehicle (C), 10 µM DMBA (D), or 10 nM TCDD (T). All lanes were loaded with 10 µg of protein, except for the far right lane, which received 2 ng of CYP1A1 (1A1 std) as a control. Proteins were separated on SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and incubated with anti-CYP1B1 or anti-CYP1A1. The enhanced chemiluminescence method was used to visualize proteins. The immunoblot was first probed with anti-CYP1B1 antibody (A) and then stripped and reprobed with anti-CYP1A1 antibody (B).

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Northern analysis of CYP1B1 in BMS2 cells and C3H10T1/2 cells. Cells were treated for 24 hr with 0.1% DMSO as a vehicle control (C), 10 µM DMBA (D), or 10 nM TCDD (T). A, Total RNA (10 µg/lane) that was electrophoresed in a denaturing agarose gel, transferred to nitrocellulose, and hybridized with 32P-labeled cDNA probes for CYP1B1 and GAPDH. B, Level of CYP1B1 mRNA expression relative to control BMS2 cells (C). Bars, mean ± standard deviation expression of three separate experiments, as quantified using a Molecular Dynamics PhosphorImager. Loading differences were corrected using GAPDH as an internal control.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Immunoblot analysis of CYP1B1 in C57-BMS cells and C3H10T1/2 cells. Primary BMS cells from C57Bl/6 mice (C57-BMS) were established as described in Materials and Methods. After 24-hr treatments with 0.1% DMSO as a vehicle control (C) or 10 nM TCDD (T), microsomal proteins were isolated and Western immunoblotted. As a standard, 2 ng of purified CYP1A1 (1A1 std) was included in a separate lane. The immunoblot was first probed with anti-CYP1B1 (A) and then stripped and reprobed with anti-CYP1A1 (B). Proteins were visualized according to the enhanced chemiluminescence method.

BMS cells metabolize DMBA. Table 1 demonstrates that BMS2 cells metabolize DMBA with a regioselectivity that is very similar to that of C3H10T1/2 cells. DMBA metabolism in C3H10T1/2 cells has been previously characterized to be entirely due to CYP1B1 (Pottenger and Jefcoate, 1990; Savas et al., 1997). The proximate carcinogen DMBA-3,4-dihydrodiol was a similar fraction of the total dihydrodiols (13-19%) produced by both BMS2 and C3H10T1/2 cells, whereas DMBA-5,6-dihydrodiol was absent (Table 1). Although the DMBA-dihydrodiol metabolite patterns were similar, the amounts of DMBA-phenols produced by BMS2 cells were consistently higher than those produced by C3H10T1/2 cells. This suggests that BMS2 cells may have less epoxide hydrolase than C3H10T1/2 cells, similar to what has been previously reported for mouse endometrial stromal cells (Savas et al., 1993). Pretreatment with the potent AhR ligand TCDD increased the amount of each metabolite produced, although the percent of each metabolite was essentially unchanged. These findings imply that BMC cell CYP1B1 metabolizes DMBA under both basal conditions and after TCDD treatment.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   DMBA metabolism by C57-BMS. Primary BMS cells from C57Bl/6 mice (C57-BMS) were established as described in Materials and Methods. Nonadherent cells were removed by washing, and adherent cells were treated for 24 hr with 0.1% DMSO as a vehicle control (Control) or 10 nM TCDD. The monolayers were then washed and incubated for 1 hr with 1 µM DMBA, and the DMBA metabolites analyzed as described in Materials and Methods. The relative amounts of DMBA-8,9-dihydrodiol and DMBA-10,11-dihydrodiol produced by C57-BMS were compared with those produced by BMS2 cells treated for 24 hr with 0.1% DMSO (100%). Bars, mean ± standard deviation of a single experiment.

AhR is functional in BMS cells. Induction of CYP1B1 after TCDD or DMBA treatment suggests that the AhR is active in BMS cells. In addition, the high constitutive expression of CYP1B1 in BMS2 cells might be due to constitutive activation of the AhR complex. To further investigate these possibilities, cytoplasmic and nuclear extracts from BMS2 cells were immunoblotted for the AhR and ARNT. Fig. 5A shows that BMS2 cells express 97- and 104-kDa AhR proteins, which correspond with the predicted Ah^b-1 and Ah^d alleles for this cell line, which was derived from a C57Bl/6 × DBA mouse (Pietrangeli et al., 1988). The AhR proteins were clearly and reproducibly detectable (~5% of total) in the nucleus under control conditions (Fig. 5A, lane 7). Levels of DMBA and TCDD that maximally induce CYP1B1 were examined for their effect on AhR translocation. Approximately 20% (Fig. 5A, lane 8) and 70% (Fig. 5A, lane 9) of the total cellular AhR translocated to the nuclear fraction after a 1-hr treatment with DMBA or TCDD, respectively. Corresponding reductions in cytoplasmic AhR to 65% (Fig. 5A, lane 2) and 23% (Fig. 5A, lane 3) of control levels were observed after treatment with DMBA or TCDD, respectively. The substantially lower translocation of AhR in response to DMBA than TCDD, at DMBA and TCDD concentrations that fully induce CYP1B1, is noteworthy. After DMBA treatment, selective translocation of the two AhR allelotypes was observed, with more of the 97-kDa AhR than the 104-kDa AhR being depleted from the cytosol and appearing in the nucleus (Fig. 5A, lanes 8 and 9). This difference was consistently observed in four separate experiments. In the particular experiment illustrated in Fig. 5, a slight degree of selective translocation was also observed with TCDD that was not observed in the three other experiments. After 6 hr of treatment with DMBA or TCDD, most of the AhR was depleted, although this was most extensive after incubation with TCDD. These observations are in agreement with a previous report of AhR down-regulation after  2 hr of treatment with AhR ligands (Pollenz, 1996). Immunoblotting for the GAPDH cytosolic marker demonstrated that the nuclear fractions were completely free of cytosolic protein contamination.

View larger version (36K): [in this window] [in a new window]   Fig. 5.   Immunoblot analysis of cytosolic and nuclear proteins from BMS2 cells treated with DMBA or TCDD. Cytosolic (CF) and nuclear (N) protein fractions were isolated after treatments with 0.1% DMSO vehicle (C), 10 µM DMBA (D), or 10 nM TCDD (T) for 1 or 6 hr (A) and 1 hr (B). Proteins (20 µg/lane) were separated by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and incubated with anti-AhR or anti-GAPDH (A) and with anti-ARNT (B). Visualization of immunoreactive proteins was achieved by the enhanced chemiluminescence method.

AhR-deficient BMS cells express CYP1B1. These findings demonstrate that BMS cells express CYP1B1 and a functional AhR but do not establish whether the AhR is required for CYP1B1 expression. To address this question, primary BMS cell cultures were established from AhR-deficient mice (AhR^/ BMS) and their AhR heterozygous (AhR^+/ BMS) littermates. BMS2 cells were included in these experiments as a control. No obvious variation in the composition of cell types was apparent between AhR^/ BMS cells and AhR^+/ BMS cells, and many of the cells in the primary BMS cultures were morphologically similar to BMS2 cells (data not shown). Fig. 6 illustrates that AhR^/ BMS cells constitutively expressed CYP1B1 protein at levels ~60% and ~75% of those observed in BMS2 and AhR^+/ BMS cells, respectively. CYP1A1 protein was not detected in either AhR^/ BMS or AhR^+/ BMS cells. As a control, we detected no immunoblot signal for CYP1B1 in primary BMS cells obtained from CYP1B1-deficient mice (data not shown). A 24-hr treatment with TCDD had no effect on CYP1B1 in AhR^/ BMS cells but resulted in a 2-fold induction of CYP1B1 in AhR^+/ BMS cells. These findings indicate that the constitutive expression of CYP1B1 in BMS cells is largely AhR independent, whereas TCDD induction of CYP1B1 is AhR dependent. Moreover, AhR-independent constitutive CYP1B1 expression is consistent with the low levels of the AhR in the nucleus of control BMS2 cells (Fig. 5A, lane 7).

View larger version (34K): [in this window] [in a new window]   Fig. 6.   Immunoblot analysis of CYP1B1 in BMS2 cells AhR+/ BMS, and AhR/ BMS cells. Primary BMS cells from AhR heterozygote mice (AhR+/ BMS) and AhR-deficient mice (AhR/ BMS) were established as described in Materials and Methods. These cells were treated for 24 hr with 0.1% DMSO as a vehicle control (C) or 10 nM TCDD (T). Cells then were incubated for 1 hr with 1 µM DMBA, and the medium was removed for analysis of DMBA metabolites. Immediately thereafter, total protein was isolated using TRIzol reagent and Western immunoblotted for CYP1B1. The immunoblot was stripped and probed with anti-CYP1A1. Proteins were visualized with the enhanced chemiluminescence method and quantified using a Molecular Dynamics Personal Densitometer SI. Bars, level of CYP1B1 protein relative to control BMS2 cells (C). The lanes for AhR/ BMS cells represent two separate cultures within the same experiment. Amounts of protein loaded were 10 µg/lane for AhR+/ BMS cells, 20 µg/lane for BMS2 cells and AhR+/ BMS cells, and 2 ng for purified CYP1A1 (1A1 std).

+/

/

+/

/

+/

/

View larger version (11K): [in this window] [in a new window]   Fig. 7.   Metabolism of DMBA by AhR+/ BMS and AhR/ BMS cells. Chromatograms of DMBA metabolites demonstrate that primary BMS cells established from AhR-deficient mice (AhR/ BMS) and their AhR heterozygous littermates (AhR+/ BMS) metabolized DMBA producing 8,9-dihydrodiol and 10,11-dihydrodiol metabolites. Primary BMS cells were established and analyzed by high performance liquid chromatography for DMBA metabolism as described in Materials and Methods. Cells were pretreated for 24 hr with either 0.1% DMSO vehicle control (CON) or 10 nM TCDD before the DMBA metabolism assay. Peaks corresponding to the 8,9- and 10,11-dihydrodiol DMBA metabolites are indicated along with a potential 5,6-dihydrodiol metabolite. The percentages indicate the amounts of combined DMBA-8,9-dihydrodiol and DMBA-10,11-dihydrodiol produced by AhR/ BMS and AhR+/ BMS relative to those produced by BMS2 cells treated for 24 hr with 0.1% DMSO (100%).

	Discussion

Top Summary Introduction Materials & Methods Results Discussion References

The results presented here demonstrate that BMS cells constitutively express high levels of CYP1B1, which is capable of metabolizing DMBA in vitro. Although constitutive CYP1B1 expression was largely AhR independent, activation of the AhR increased CYP1B1 expression and DMBA metabolism. The former observation differs from previous investigations that indicated that constitutive CYP1B1 expression in embryo fibroblasts is dependent on AhR activation (Zhang et al., 1998). These findings have important implications for bone marrow toxicology and physiology and suggest that BMS cells may regulate CYP1B1 expression in a manner different from that previously described in embryo fibroblasts (Alexander et al., 1997; Zhang et al., 1998).

We clearly demonstrated that mouse BMS cells metabolize DMBA in vitro. Metabolites produced include a substantial proportion of 3,4 dihydrodiol (15% of total metabolites produced by BMS2 cells), the precursor to the most toxic product. The profile of DMBA metabolites formed (Table 1, Figs. 3 and 7) strongly suggests that CYP1B1 was the predominant P450 cytochrome catalyzing DMBA metabolism. This conclusion is based on similarities in the proportions of DMBA metabolites produced by BMS2 cells and C3H10T1/2 cells. In particular, production of the 3,4-dihydrodiol and 10,11-dihydrodiol DMBA metabolites has been reported for recombinant mouse CYP1B1 but not for recombinant mouse CYP1A1 (Savas et al., 1997). Moreover, neither BMS2 nor primary BMS cells produced appreciable amounts of DMBA-5,6-dihydrodiol, a major predicted product of CYP1A1 (Wilson et al., 1984). To our knowledge, this is the first report that mouse BMS cells metabolize DMBA, although human mononuclear bone marrow cells have been reported previously to metabolize DMBA to the 3,4-dihydrodiol product (O'Dowd, 1987).

Western and Northern blot analyses confirmed that mouse BMS cells constitutively express high levels of CYP1B1, which is only slightly induced by TCDD (1.4-2-fold), but no detectable CYP1A1. The basal levels of CYP1B1 in BMS2 and C57-BMS cells were consistently higher than inC3H10T1/2 cells, whereas TCDD-induced levels were comparable in all three cell types. The levels of constitutive CYP1B1 in BMS2 cells are higher than those previously reported for other mouse tissues (Savas et al., 1994, 1993) and human cells (Kress and Greenlee, 1997).

Based on our results, we hypothesize that AhR activation increases CYP1B1 expression in BMS cells. In support of this statement, we observed that BMS2 cells (1) express immunodetectable ARNT and AhR proteins (97 and 104 kDa) that are primarily cytosolic (only ~5% nuclear) under basal conditions, (2) exhibit nuclear translocation of AhR proteins after exposure to DMBA or TCDD, and (3) increase CYP1B1 expression after treatment with DMBA or TCDD and (4) TCDD did not induce CYP1B1 expression in AhR^/ BMS cells. Our observations are supported by previous findings that suggest the AhR is active in BMS2 cells (Yamaguchi et al., 1997b; Lavin et al., 1998).

Although AhR activation increased CYP1B1 expression, the constitutive regulation of CYP1B1 in BMS cells appeared to be largely AhR independent. AhR^/ BMS cells expressed levels of CYP1B1 protein that were similar (60-75%) to those in AhR^+/ BMS cells and in the BMS2 cell line. From these observations, it seems that AhR activation is responsible for at most 25-40% of constitutive CYP1B1 expression in BMS cells. The constitutive expression of CYP1B1 in AhR^/ BMS is in contrast to a recent report by our laboratory that demonstrated that CYP1B1 expression in mouse embryo fibroblasts was dependent on the presence of the AhR (Zhang et al., 1998). These separate findings suggest that the constitutive expression of CYP1B1 is regulated by a factor or factors other than the AhR in BMS cells. Because TCDD-induced levels of CYP1B1 are similar in mouse BMS cells and C3H10T1/2 embryo fibroblasts, it is possible that AhR activation overrides other cell-type-specific mechanisms to provide a maximal level of transcription.

The primary BMS cells used in this study were adherent cells prepared from Whitlock-Witte cultures (Whitlock and Witte, 1982). These types of cell cultures are reported to consist primarily of very large nonphagocytic fibroblastoid cells and macrophages (Witte et al., 1987). Stromal fibroblasts from mouse endometrium (Savas et al., 1993), mouse embryos (Pottenger and Jefcoate, 1990; Alexander et al., 1997), and rat mammary glands (Christou et al., 1995) also constitutively express CYP1B1 although at lower levels than BMS cells. Because both primary BMS cultures and BMS2 cells expressed comparable levels of constitutive and inducible CYP1B1, we conclude that most of the primary cells are CYP1B1-expressing fibroblast-like cells. However, primary BMS cell cultures, but not BMS2 cells, exhibit nonspecific DMBA oxidation in DMBA metabolism experiments (data not shown). We attribute this difference to the heterogeneity of cells in primary BMS cultures, particularly macrophages that could release peroxidative products (Johansson et al., 1995) that attack DMBA. Moreover, these nonspecific oxidation products likely reduced the amounts of specific DMBA metabolites formed by CYP1B1 in primary BMS cultures.

Bone marrow stromal cell expression of CYP1B1 has important implications for bone marrow toxicology and physiology. Metabolic activation of environmental contaminants by bone marrow CYP1B1 could increase the effective concentrations of activated toxicants in bone marrow. This is important because the human ortholog of CYP1B1 recently has been shown to activate many diverse procarcinogens to mutagenic metabolites (Shimada et al., 1996). The high constitutive expression of CYP1B1 means that activation of toxic compounds in the bone marrow is not dependent on AhR activation and therefore might precede AhR-dependent up-regulation of detoxifying enzymes in distant organs such as the liver and kidney. Furthermore, the high constitutive levels of CYP1B1 suggest that it may play a role in maintaining bone marrow physiology. This supposition is supported by the expression of catalytically active CYP1B1 in the absence of an AhR (AhR^/ BMS cells), a finding that has not been reported for any other tissue. Future experiments using CYP1B1-deficient mice will help determine whether CYP1B1 has a physiological role in bone marrow and whether the bone marrow toxicity of DMBA is dependent on the presence of CYP1B1.