Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Biliverdin is the open tetrapyrrole oxidation product of heme (Fe-protoporphyrin IX) degradation by the heme oxygenase (HO) system (Maines, 1992, 1997). In mammals and in certain fish, biliverdin is reduced to the bile pigment, bilirubin, by the soluble enzyme, biliverdin reductase (Singleton and Laster, 1965; Tenhunen et al., 1970; Kutty and Maines, 1981). In plants, biliverdin analogs function in photoregulatory capacity (Terry and Lagarias, 1991; Cornejo et al., 1992).

The reductase is encoded by a single copy gene. Analysis of the promoter region of the rat gene (McCoubrey et al., 1995) identifies the presence of recognition sites for several regulating proteins, including INF-1, an enhancer of cytokine and virus-induced transcriptional activation (Fujita et al., 1987) and AP-1, the oxidative stress responsive proto-oncogene binding site (Karin et al., 1997). Also, two elements known to be involved in embryonic gene expression, P3A (Thiebaud et al., 1990) and engrailed (Ohkuma et al., 1990), are present in the promoter region of the gene. These criteria suggest the potential regulation of the gene by oxidizing agents and developmental factors. Biliverdin reductase is highly conserved in evolution as indicated by the predicted primary structure of the rat and human proteins (Fakhrai and Maines, 1992; Maines et al., 1996). A feature of the protein is a sequence with basic residues (Gly-Leu-Lys-Arg-Asn-Arg) in the carboxy terminal third of the reductase. Such clusters are a characteristic of the nuclear localization signal (Garcia-Bustos et al., 1991). Another feature is the presence of the nucleotide phosphate binding site (GlyXGlyXXGly).

The product of biliverdin reductase activity, bilirubin, is biologically active and has been shown to be a potent antioxidant (Stocker et al., 1987; McDonagh, 1990). In addition, bilirubin interferes with protein phosphorylation and activation of superoxide producing NADPH oxidase (Kwak et al., 1991; Hansen et al., 1996). It is noteworthy that in biological systems there are no other mechanisms for the formation of bilirubin but the catalytic activity of biliverdin reductase.

Among all organs, the kidney has the highest levels of biliverdin reductase (McCoubrey et al., 1995). As is well known, the kidney is a target organ for various chemicals that modulate gene expression. This study was undertaken to investigate the response of the kidney to nephrotoxins, bacterial lipopolysaccharide (LPS), and bromobenzene. The findings demonstrate, for the first time, that the reductase responds to environmental agents and localizes into the cell nucleus, and as such, suggest a mechanism by which bilirubin mediates its biological activities.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Male Sprague-Dawley rats (200-250 g) were purchased from Harlan Industries (Madison, WI). LPS, bromobenzene, oligo(dT)-cellulose, DNase I, salmon testis DNA, Triton X-100, dextran sulfate, 4-chloronaphthol, and paraformaldehyde were purchased from Sigma Chemicals (St. Louis, MO). Goat anti-rabbit IgG conjugated with horseradish peroxides was purchased from Cappel Organotechnica (Durham, NC). Nitrocellulose and Nytran membranes were obtained from Schleicher and Scheull (Keene, NH). [-³²P]dCTP was supplied by Amersham Pharmacia Biotech (Indianapolis, IN). Oligonucleotides for mutagenesis and sequencing were purchased from Life Technologies, Inc. (Gaithersburg, MD). Sequenase version 2.0 as well as all restriction enzymes were obtained from U.S. Biochemical Corp. (Cleveland, OH). Reagents for protein determination were obtained from Bio-Rad (Richmond, CA). Reagents for cell culture were purchased from Life Technologies, Inc. or Difco (Detroit, MI). Antibody to rat biliverdin reductase was prepared as before (Huang et al., 1989). Rats were treated with 5 mg/kg LPS and killed 6 or 16 h later, or with 2 mmol/kg bromobenzene in corn oil (s.c.) and killed after 4 or 24 h. These time points were selected based on preliminary experiments. All animal use procedures were in strict accordance with National Institutes of Health guidelines for the care and use of laboratory animals and were approved by local Animal Care Committee.

Northern Blot Analysis. An HO-1 cDNA corresponding to HO-1 nucleotides +71 to +833 reported by Shibahara et al. (1985) was generated using PCR and cloned into PBS(+) vector as described before (Ewing and Maines, 1991). A PCR product consisting of nucleotides +401 to +926 of biliverdin reductase cDNA (Fakhrai and Maines, 1992) and -actin (loading control) were used. All probes were labeled with [³²-P]dCTP by the random primers DNA labeling system (U.S. Biochemical Corp.), according to the manufacturer's instructions, and further purified by spin column chromatography. Total RNA and poly(A)+ RNA was isolated from pooled kidneys of three rats by oligo(dT)-cellulose chromatography, and the formaldehyde-denatured RNA was fractionated on a 1.2% (w/v) agarose gel and subsequently transferred to Nytran membrane. Prehybridization and hybridization of the membranes with the appropriate ³²P-labeled cDNA were performed essentially as described previously (Sun et al., 1990). The membranes were exposed at 70°C to Kodak X-OMAT film with intensifying screens, and autoradiographs were quantified using Bio-Rad model GS-700 imaging densitometer and Molecular Analyst v.1.5 software.

Western Blot Analysis. Analysis was carried out as before (Huang et al., 1989). Pooled kidneys of six rats were used for each preparation. Protein samples were fractionated by SDS-polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane using an LKB2005 Transphor Apparatus (Bio-Rad). Antigen-antibody complexes were immunochemically visualized using horseradish peroxidase-conjugated goat anti-rabbit antibody.

Cell Fractionation and Measurement of Biliverdin Reductase Activity. Rats were killed by decapitation, and kidneys were perfused with saline. The organs are homogenized with 5 volumes of 0.1 M sodium phosphate buffer, pH 7.4, and used for preparation of cytosol (105,000g fraction). To obtain nuclear extract, kidneys pooled from 6 rats were homogenized in 5 volumes of 0.35 M sucrose homogenization buffer (10 mM Hepes, pH 7.6, 25 mM KCl, 1 mM EDTA, 10% glycerol, and 0.5 mM dithiothreitol). The homogenate was centrifuged at 10,000g for 20 min, and the pellet was rehomogenized and subjected to gradient centrifugation (26,000g for 40 min), using 2 M sucrose homogenization buffer (pH 7.6, containing the above indicated compounds plus sucrose). The nuclear pellet was resuspended in 0.35 M sucrose homogenization buffer (pH 7.6, containing the above indicated compounds plus sucrose) containing 0.5% Triton X-100 and centrifuged at 4,000g for 8 min. The pellet was resuspended again in 0.35 M sucrose homogenization buffer and centrifuged at 4,000g for another 8 min. The nuclear pellet was resuspended in 225 µl of water and after 10 s, 0.5 µl of 0.5 M EDTA, 20 µl of 1 M sodium phosphate buffer, pH 7.4, and 20 µl of DNase (10 mg/ml) was added, and the pellet was centrifuged at 4,000g for 10 min. The supernatant fraction was used for measurement of biliverdin reductase activity. Contamination of the nuclear extract with cytosol was assessed by measuring the cytoplasmic lactate dehydrogenase activity in the nuclear extract and the cytosol (Kornberg, 1955).

Mutagenesis of Biliverdin Reductase. The expression clone hBVR-1 (Maines et al., 1996) was used to generate nuclear localization signal mutants designated "NLS mut" using the oligonucleotide primer: GGAAGCTTAAATATCCTGTGGATCCTATAACAGGTCCTTTTTC, consisting of the reverse complement of nucleotides +652 to 694 with mismatches (underlined) which result in the change of amino acids 222-227: Gly-Leu-Lys-Arg-Asn-Arg to Val-Ile-Gly-Ser-Thr-Gly. The changes made were conservative ones except that charged residues were replaced with uncharged ones. The mutations introduced a BamHI site, the presence of which was used in screening for mutants, and the nucleotide sequence of those mutants was subsequently confirmed.

Generation of Hemagglutinin-Tagged Constructs for Transfection. DNA from plasmids encoding wild type or NLS mutant BVR was used as substrate for PCR using the primers: GGGATCCATGTACCCCTACGACGTGCCCGACTACGCCAATGCAGAGCCCGAGAGGA, representing nucleotides +4 to +22 placed, in-frame, downstream of a hemagglutinin recognition sequence (bold), a methionine start codon (underlined), and a BamHI linker (italics) and (GCTCGAGCTCCTCCTCTTACTTCCTTG), which is the reverse complement of nucleotides +881 to +900 including the stop codon (underlined) and an XhoI linker (italics). The product was cloned into the vector pCR2.1, and the insert was excised using BamHI and XhoI and subcloned into pcDNA3, which had been cut with the same enzymes.

Cell Culture and Immunofluorescence Staining. HeLa cells were grown on glass coverslips in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum under an atmosphere of 5% CO₂, 95% air. Cells were transfected with different BVR constructs using LipofectAMINE (Life Technologies, Inc.). Briefly, cells were treated with a mixture of 2 µg of DNA and 20 µl of LipofectAMINE for 5 h, followed by the addition of Dulbecco's modified Eagle's medium. After 48 h, cells were treated with 500 µM 8-Br-cGMP for 1, 5, 20, or 60 min, and fixed with 4% paraformaldehyde for 10 min. Cells were then rinsed with phosphate-buffered saline, and incubated with 5% goat serum to block nonspecific binding sites. Immunohistochemistry was carried out using monoclonal anti-hemagglutinin as the primary antibody and fluorescein isothiocyanate-conjugated secondary antibody. Cells were visualized by direct fluorescence microscopy.

Immunohistochemistry of Kidney Biliverdin Reductase. Control and treated rats were killed by an overdose of pentobarbital (100 mg/kg). For single antibody staining, kidneys were perfused, fixed with 4% paraformaldehyde (v/v), and postfixed for an additional 16 h at 4°C, then an additional 8 h at 25°C, before dehydration and paraffin embedding (Ewing and Maines, 1995). Kidney sections (5-µm thick) were cut and mounted on Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA). Tissue was rehydrated by sequential equilibration with graded ethanol solutions (99-50%, v/v), and finally with 0.1 M phosphate buffer (pH 7.2). Thereafter, sections were preincubated in filtered (0.45 mm) 0.1 M phosphate buffer containing 0.3% (v/v) Triton X-100 and 10% (v/v) normal goat serum (antibody buffer). Tissue was incubated with polyclonal rabbit anti-rat biliverdin reductase antibody diluted 1:1000 in antibody buffer under paraffin coverslips. After 16 h of incubation at 4°C, slides were allowed to equilibrate to 25°C for 1 h before being rinsed with phosphate buffer. Tissue was stained for antibody-antigen complexes by means of the avidin-biotin detection system, according to the manufacturer's instructions (Vector Laboratories, Burlingame, CA). Control tissues from treated animals were processed histologically under identical conditions at the same time. For double staining, free floating samples were used. Kidney tissue was sucrose-ethylene glycol cryoprotected, and 35-µm-thick specimens were cut on dry ice using a sliding microtome (Microm 400, Carl Zeiss Inc., Thornwood, NY). Free floating specimens were first processed for BVR immunocytochemistry as described above and then counterstained with thionin (Vector Laboratories). The experiments were repeated a minimum of three times. The controls and treated samples were processed simultaneously using the same experimental conditions and solutions.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Figure 1 shows the effect of 6 and 16 h LPS treatment on the pattern of biliverdin reductase distribution in the rat kidney, as assessed by immunohistochemistry. For this experiment the kidney tissue was double-labeled: immunostained for biliverdin reductase using anti-rat biliverdin reductase antibody (red) and counterstained with a nuclear stain, thionin (blue). Panels a, b, and c show tissue staining at 0, 6, and 16 h of LPS treatment at low magnification, respectively. Panels d, e, and f depict the higher magnification of the same samples. As shown in panels a and d, under normal conditions, the reductase displays a diffuse staining of the cytosol with some scattered nuclei showing immunoreactivity with the reductase antibody. Subsequent to LPS treatment, there is a pronounced increase in population of nuclei that display staining for the reductase at 6 h (panels b and e) as well as at 16 h (panels c and f). Under higher magnification in panels e and f, nuclear staining for the reductase, which localizes with methyl green, is clearly visible in the kidney of LPS-treated rats.

View larger version (70K): [in this window] [in a new window]   Fig. 1.   Cellular distribution of biliverdin reductase in the kidney of lipopolysaccharide-treated rats. Rats were treated with LPS (5 mg/kg, i.p.) and were killed 6 or 16 h later. Kidneys were processed for immunohistochemistry as described under Experimental Procedures, using rabbit anti-rat biliverdin reductase antibody for visualization of biliverdin reductase (red) and thionin for nuclear counterstaining (blue). Panels a and d, control; b and e, 6 h LPS-treated; c and f, 16 h LPS-treated. Magnification: panels a, b, and c, 40×; panels d, e, and f, 100×.

The next set of experiments (Fig. 2) confirmed that the nuclear staining for the reductase is due to an increase in the concentration of the enzyme and that the nuclear biliverdin reductase is catalytically active. The findings with Western blot analysis of the cytosol and the nuclear extract 16 h after LPS treatment are shown in Fig. 2, a and b, respectively. As noted in the cytosol, the reductase migrated as two bands with molecular mass of ~33 kDa and ~30 kDa. Lane 1 is that of control kidney cytosol, and lane 2 is that of LPS-treated sample. As noted in panel a, the intensity of the 33 kDa is modestly increased in the treated rat kidney cytosol. In contrast, Fig. 2b shows that in the nucleus the larger band is not detected; rather when compared with the control, there is a notable increase in intensity of the signal near the 30 kDa region (lane 1 versus 2). Moreover, the nuclear signal shows microheterogeneity in size. Previous studies have shown (Huang et al., 1989) that post-translational modification of the reductase is responsible for generation of molecular weight and charge variants. It is likely that post-translational modification of the protein is the basis for the heterogeneity of the nuclear protein.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Western blot and activity analyses of nuclear biliverdin reductase. Cytosol and nuclear extracts were prepared from control and LPS-treated (16 h) rat kidney as described in the text. The cell fractions were obtained from organs pooled from six rats. Western blot and activity analyses were carried out as described under Experimental Procedures. Panel a, Western blot analysis of the cytosol obtained from control and LPS-treated rat kidney; 50 µg of protein was loaded in each lane. Lanes: 1, control; 2, LPS-treated. Panel b, Western blot analysis of kidney nuclear extract fraction obtained from control and LPS-treated rats; 200 µg of protein was loaded in each lane. Lanes: 1, control; 2, LPS-treated. Purified rat kidney cytosol biliverdin reductase was used as the standard (ST). Panels c and d, activity analysis. The reductase activity was measured in the cytosol (c) and nuclear extract (d) 16 h after LPS treatment. The rate of activity was assessed by measuring the reduction of biliverdin to bilirubin. *p  0.05 when compared with corresponding control. Panel e, absorption spectrum of bilirubin formed by the control rat kidney nuclear extract in the presence of NADPH as the cofactor and biliverdin as the substrate. Experimental details are provided in the text.

The results of activity analysis at 16 h after treatment are shown in panel c. Data show that nuclear reductase is active, and that subsequent to LPS treatment, there is an increase in nuclear extract reductase activity. The activity in the cytosol was modestly increased and measured about 160% of that of the control; the nuclear extract activity, however, was increased by nearly 2.5-fold. It is noteworthy that the specific activity of the reductase in control kidney cytosol exceeds that of nuclear fraction by about 20-fold, whereas in the LPS-treated rat kidney the ratio is decreased to about 13-fold. This observation is consistent with an increased nuclear concentration of the reductase in the treated kidney and may suggest increased nuclear traffic of the reductase in LPS-treated tissue. This, in turn, may reflect changes in signal transduction pathways. The activity of the nuclear extract was not due to contamination with the cytosolic fraction, as was indicated by finding that lactate dehydrogenase activity in the nuclear extract measured 1% of that of the cytosol. Figure 2d shows that the product of the nuclear reductase activity displays typical bilirubin absorption spectrum with peak absorption at 450 nm. For this experiment, control nuclear fraction was used.

Next, we examined whether the nuclear localization of the reductase is a phenomenon specific to LPS or extends to other nephrotoxins. For this, response of the kidney biliverdin reductase to bromobenzene was investigated. Findings of this investigation using immunohistochemical analysis are shown in Fig. 3. As shown in panel a at low magnification (10×) and panel b at high magnification (100×), under normal conditions the reductase shows diffuse staining in the cell, with isolated nuclei exhibiting staining. This is very similar to the observation made with double-staining experiments shown in Fig. 1. In response to bromobenzene treatment, as with LPS treatment, there is an increase in biliverdin reductase immunostaining (panels c and d). As noted, there is a striking increase in the population of nuclei that display immunostaining as depicted using different magnifications (panels: c = 20×; d = 100×).

View larger version (129K): [in this window] [in a new window]   Fig. 3.   Immunohistochemical detection of biliverdin reductase reactivity in bromobenzene-treated rat kidney. Male Sprague-Dawley rats were treated with 2 mmol/kg bromobenzene in corn oil (s.c.) and killed 24 h later. Control rats received the vehicle. Paraffin-embedded kidney samples (5-µm-thick) were processed for immunohistochemistry as described under Experimental Procedures. Panels a and b, control kidney at low (20×) and high (100×) magnifications, respectively; c and d, treated rat kidney samples at 20× and 100× magnifications, respectively.

To determine the molecular basis for increase in kidney biliverdin reductase levels, Northern blot analysis was carried out (Fig. 4). Poly(A) mRNA was isolated from kidneys of control, LPS- or bromobenzene-treated rats and was analyzed for biliverdin reductase transcript levels (panels a and b, respectively). Actin was used as the control for loading and HO-1 was used as the positive control. As noted, neither treatment increased the level of biliverdin reductase transcript in the kidney. The effectiveness of the treatment to modulate responsive genes is, however, indicated by the observation that HO-1 mRNA levels were prominently increased in the treated kidneys. When assessed by densitometry, a nearly 25-fold and a 5-fold increase in the HO-1 mRNA levels was detected for LPS and bromobenzene, respectively. As noted, under control conditions HO-1 mRNA levels in the kidney is minimal. It is noteworthy that this is the first demonstration of transcriptional activation of HO-1 in the kidney by these compounds.

View larger version (49K): [in this window] [in a new window]   Fig. 4.   Northern blot analysis of biliverdin reductase and HO-1 in the LPS and bromobenzene-treated rat kidney. Poly(A) mRNA was isolated from the control and LPS-treated (5 mg/kg, i.p., 6 h) or bromobenzene-treated (2 mmol/kg, s.c., 4 h) rats for Northern blot analysis. The blot was probed sequentially with cDNA probes for biliverdin-reductase, HO-1, and last, with actin, which was used as a loading control. Experimental details are provided in the text. Each lane contained 8 µg of poly(A) mRNA isolated from pooled organs of three rats. Panel a, LPS-treated; lanes: 1, control; 2, treated. Panel b, bromobenzene-treated; lanes: 1 and 2, control; 3 and 4, treated.

Next, whether nuclear localization of the reductase extends to human cell lines was examined, and the molecular basis for nuclear localization was investigated. For the latter, the primary features of the protein were considered. The cluster of positively charged residues near the carboxy terminal third of the protein (amino acids 222-227), which as mentioned, appeared to be a good candidate for being a nuclear localization signal (Garcia-Bustos et al., 1991). Such clusters are present in certain phosphotransferases, such as those that phosphorylate G-protein-coupled receptors, and those that translocate in the cell in response to phosphorylation activators, such as cGMP (Hanks and Quinn, 1991). Moreover, the gaseous monoxide cell signals, NO and CO, share affinity for the heme molecule, and they resemble one another in mediating those activities that are cGMP-related (Furchgott and Jothianandan, 1991; Shraga-Levine et al., 1994; Morita et al., 1995). It follows, elevation of renal HO-1 expression in response to LPS and bromobenzene are generally believed to be accompanied by increased CO-mediated cGMP production. Accordingly, nuclear localization of a biliverdin reductase construct in which the sequence encompassing the putative signal (Gly-Leu-Lys-Arg-Asn-Arg) was mutated by the conservative replacement of residues (Val-Ile- Gly-Ser-Thr-Gly) was tested in HeLa cells. Nuclear localization was examined under normal growth conditions and in response to the addition of nonhydrolyzable cGMP analog, 8-Br-cGMP. Comparison was made with localization of the wild-type reductase under the same conditions. To monitor cellular location, the constructs were designed with a short hemagglutinin tag at the amino terminus; immunoflourescence staining, using anti-hemagglutinin as the primary antibody, was used to trace the tagged proteins. Results are shown in Fig. 5. Under normal conditions, wild-type biliverdin reductase was found diffused and distributed throughout the cytoplasm with a somewhat greater concentration in the perinuclear region (panel a). Five minutes after treatment with the cGMP analog (panel b) there was a lessening of cytoplasmic staining and the appearance of strong punctate staining associated with the nucleus. The cellular localization of the mutated protein, under control conditions, was nearly indistinguishable from what was observed with the wild-type protein (panel c versus a). However, as shown in panel d, nuclear localization of the mutant protein was not detected in response to 8-Br-cGMP stimulation. Nuclear localization at shorter (1 min) and longer time points (up to 60 min) were also examined, and again translocation of the mutant protein was not detected. It appears certain that mutation in this region of the reductase alters intracellular trafficking of the protein in response to the cyclic nucleotide.

View larger version (32K): [in this window] [in a new window]   Fig. 5.   The presence of the putative nuclear localization signal is necessary for biliverdin reductase translocation into the nucleus. Constructs were made in pcDNA3 expressing hemagglutinin tagged: full-length wild-type biliverdin reductase or a full-length protein in which mutations were introduced in the stretch of BVR amino acids, encompassing the putative NLS motif of the reductase. The constructs were used to transfect HeLa cells. After 48 h, the cells were treated for 5 min with 8-Br-cGMP and then harvested. Cells were examined by fluorescence immunocytochemistry as described in the text. Panels: a, control wild-type; b, treated wild-type; c, control NLS mutant; d, treated NLS mutant. Magnification, 100×.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Prior to this study, biliverdin reductase was assumed to be exclusive to the cytosol and nonresponsive to chemicals (Kutty and Maines, 1981; Huang and Maines, 1990). Hence, the enzyme was only examined in the cytosol, and primarily with respect to its molecular and biochemical properties. The present study reports, for the first time, on the nuclear presence of the reductase and increases in its nuclear localization in response to renal toxins, LPS and bromobenzene. The study also suggests that nuclear localization of biliverdin reductase is not exclusive to the rat kidney and extends to human cell lines. As noted above, biliverdin reductase is the product of a single copy gene (McCoubrey et al., 1995) that undergoes post-translational modification to give rise to variants with different molecular weight (Huang et al., 1989; Huang and Maines, 1990). The present study detects two bands in the rat kidney cytosol; however, as noted, molecular variants close in size to the smaller cytosolic variant are detected in the nucleus. It is noteworthy that in the reductase preparation purified from the liver cytosol of bromobenzene-treated rats and examined by two-dimensional electrophoresis system, a major depression in the M_r 30,400 form of the reductase is detected (Huang and Maines, 1990). It is also noteworthy that the reductase displays extensive microheterogeneity when examined by two-dimensional electrophoresis.

Nuclear transport of cytosolic proteins across the nuclear envelope is an active process and requires a nuclear localization signal contained within the transported protein (Garcia-Bustos et al., 1991). Transport into the nucleus differs from traffic of proteins into other organelles, which involve movement through the hydrophobic environment of a membrane (Verner and Schatz, 1988). The nuclear localization of biliverdin reductase as shown here (Fig. 5), as that of other nuclear proteins (Garcia-Bustos et al., 1991), requires both the nuclear localization signal and is responsive to cGMP. The transport of the reductase could involve either interaction of the reductase nuclear localization signal with nuclear pore complex or with other cytoplasmic components that would deliver the reductase. Considering that nuclear localization of the reductase is blocked when the signal is mutated, it would appear that the latter mechanism is not involved in nuclear transport of the protein, and suggests that the signal present in the reductase itself is essential for its translocation; this, however, does not preclude the possibility that the reductase serves as a vehicle for transport of cytoplasmic components into the nucleus. At this time a speculation can be offered with respect to this possibility. Previous studies have described a high-affinity metalloporphyrin-binding site of the reductase, which is separate from its substrate (biliverdin) binding site (Bell and Maines, 1988). Given the fact that metalloporphyrins are effective regulators of gene expression, including that of -aminolevulinate synthase, the rate-limiting enzyme in heme biosynthesis, and that of HO-1 (Granick et al., 1975; Foresti et al., 1997; Jacobs et al., 1998; Shan et al., 1999), it would not be unreasonable to speculate that the reductase may serve as an intracellular shuttle for heme. In this context, it is noteworthy that HO-1 mRNA levels were found increased in LPS- and bromobenzene-treated rat kidneys (Fig. 3). Previous studies have demonstrated induction of HO-1 activity in the liver in response to bromobenzene treatment (Guzalian and Elshourbagy, 1979; Maines et al., 1986)

The cellular basis for LPS and bromobenzene promoting nuclear transport of the reductase may relate to these agents and/or their products activating kinases and altering signal transduction pathways; LPS has been reasoned to exert such effect (Haga et al., 1996). Moreover, given the fact that LPS activates production of NO, a potent inducer of HO-1(Motterlini et al.,1996) and NO induces HO-1 via mitogen-activated protein kinases (MAPK) pathway, ERK and p38 (Chen and Maines, 2000), it is likely that LPS treatment alters kinase pathways. Bromobenzene, perhaps the most potent inducer of liver HO-1 isozyme, is known to be metabolized to reactive metabolites that are capable of interaction with regulatory macromolecules and -SH groups (Brodie et al., 1971; Jollow et al., 1974; Aniya et al., 1988). The effect of bromobenzene on signal transduction pathways has not been examined. However, based on the reports that indicate oxidative stress-mediated HO-1 induction involves the MAPK signal transduction pathway (Elbirt et al., 1998; Chen and Maines, 2000), it would be reasonable to suspect that bromobenzene and/or its metabolite(s) effect the signal transduction pathways.

The toxicological significance of nuclear localization of biliverdin reductase may reflect a multiple of factors that stem from the activity of the reductase. The product of the enzyme activity, bilirubin, is biologically active and has diverse kinds of activities in the cell. Bilirubin could protect the nuclear components against free radical damage by virtue of being an effective chain-breaking antioxidant (Stocker et al., 1987; McDonagh, 1990) and an inhibitor of the superoxide-producing NADPH oxidase (Kwak et al., 1991). In addition, the reductase has serine/threonine kinase activity (Salim et al., 2001) and as such could modulate signal transduction pathways. Serine/threonine kinases are regulated by a host of extracellular stimuli, including growth factors, mitogens, cytokines, oxidative stress, and environmental agents. In its capacity as a kinase, biliverdin reductase could modulate signal transduction pathways and function as an important mechanism for regulation of HO-1 activity. The increased activity of which, in turn, would enhance cellular defense mechanism by degrading heme and generating bile pigments.