Differential Cellular and Subcellular Localization of Heme-Binding Protein 23/Peroxiredoxin I and Heme Oxygenase-1 in Rat Liver

Abstract

Heme-binding protein 23 (HBP23), also termed peroxiredoxin (Prx) I, and heme oxygenase-1 (HO-1) are distinct antioxidant stress proteins that are co-ordinately induced by oxidative stress. HBP23/Prx I has thioredoxin-dependent peroxidase activity with high binding affinity for the pro-oxidant heme, while HO-1 is the inducible isoform of the rate-limiting enzyme of heme degradation. We investigated the cellular and subcellular localization of both proteins in rat liver. Whereas by immunohistochemistry (IHC) a uniformly high level of HBP23/Prx I expression was observed in liver parenchymal and different sinusoidal cells, HO-1 expression was restricted to Kupffer cells. By immunoelectron microscopy using the protein A–gold technique, HBP23/Prx I immunoreactivity was detected in cytoplasm, nuclear matrix, mitochondria, and peroxisomes of parenchymal and non-parenchymal liver cell populations. In contrast, the secretory pathway, i.e., the endoplasmic reticulum and Golgi complex, was free of label. As determined by immunocytochemical (ICC) studies in liver cell cultures and by Western and Northern blotting analysis, HBP23/Prx I was highly expressed in cultures of isolated hepatocytes and Kupffer cells. In contrast, HO-1 was constitutively expressed only in Kupffer cell cultures but was also inducible in hepatocytes. These data suggest that HBP23/Prx I and HO-1 may have complementary antioxidant functions in different cell populations in rat liver.

Keywords

heme-binding protein 23/peroxiredoxin I heme oxygenase-1 liver oxidative stress hepatocytes Kupffer cells immunocytochemistry protein A–gold

Heme-binding protein 23 (HBP23), also called peroxiredoxin (Prx) I, is an intracellular antioxidant protein that is preferentially expressed in liver. HBP23/Prx I has thioredoxin-dependent peroxidase activity and exhibits high binding affinity for the pro-oxidant heme (Chae et al. 1994a, b; Iwahara et al. 1995). The Prxs constitute a growing protein family that has been identified in a large number of organisms (Chae et al. 1993; Ishii et al. 1993; Tsuji et al. 1995; Jin et al. 1997; Hirotsu et al. 1999; Rhee et al. 1999; Seo et al. 2000; Fujii and Ikeda 2002; Hofmann et al. 2002). The major functions of Prxs appear to be cellular protection against oxidative stress and regulation of cell signaling (Wen and Van Etten 1997; Kang et al. 1998; Rhee et al. 1999; Chang et al. 2002; Hofmann et al. 2002). We noted earlier that, in liver cell culture models, oxidative stress induces expression of HBP23/Prx I in coordination with heme oxygenase-1 (HO-1) (Immenschuh et al. 1995), and this has been confirmed in other cell systems (Siow et al. 1995; Immenschuh et al. 1999b; Nakaso et al. 2000). HO-1 is the inducible isoform of the rate-limiting enzyme of heme degradation, and induction of HO-1 constitutes an adaptive response against cellular stress (Maines 1997; Otterbein and Choi 2000).

In normal rat liver, HO-1 is mainly expressed in Kupffer cells (KC) (Goda et al. 1998; Immenschuh et al. 1999b). By contrast, although HBP23/Prx I represents a substantial portion of total hepatic protein (ca. 0.1%), its exact cellular distribution in liver is not known. Moreover, although by cell fractionation HBP23/Prx I has been detected in nucleus and cytoplasm (Kang et al. 1998; Butterfield et al. 1999), its precise subcellular distribution in liver has not been reported. Distribution of antioxidant stress proteins in various cells of an organ and in distinct subcellular compartments is important for the modulation of oxidative damage in diseases such as cancer (Oberley 2002). The major goal of the present study was to investigate the cellular localization and expression of HBP23/Prx I in comparison to that of HO-1 in rat liver. In addition, the subcellular localization of HBP23/Prx I in normal rat liver was analyzed by the protein A–gold technique.

Materials and Methods

Materials

The embedding media for light and electron microscopic immunohistochemical studies, Paraplast plus and LR White, respectively, were obtained from Sherwood Medical (St Louis, MO) and Polysciences (Heidelberg, Germany). Glutaraldehyde and paraformaldehyde were from Sigma (Munich, Germany). Media M199 and RPMI 1640 were obtained from Gibco BRL (Karlsruhe, Germany), and the chemiluminescent detection system for Western blotting, radioisotopes and nitrocellulose filters were from Amersham–Buchler (Braunschweig, Germany). Lab-Tek chamber slides were from Nunc (Naperville, IL), the nucleotide removal kit from Qiagen (Düsseldorf, Germany), and Falcon tissue culture dishes from Becton–Dickinson (Heidelberg, Germany). Protein A was from Pharmacia (Uppsala, Sweden). All other chemicals were purchased from Sigma or Roche Diagnostics.

Antibodies

The monospecificity of the polyclonal rabbit antibody against purified HBP23/Prx I was described previously (Iwahara et al. 1995) and is also shown in Figure 1. The commercial polyclonal rabbit anti-rat HO-1 antibody was obtained from Stress Gene (Victoria, BC, Canada) (Immenschuh et al. 1999b). The monoclonal antibody (MAb) ED2 was purchased from Serotec (Wiesbaden, Germany) and was used as a marker for monocyte–macrophages (Dijkstra et al. 1985). The antibodies for catalase (Beier et al. 1988) and lactate dehydrogenase (Baumgart et al. 1996) have previously been described and were used for positive controls in this study.

Animals

Male Wistar rats (2 months old, body weight 200 g) were used throughout the study. All experiments were approved by the local animal experiment review committee.

Figure 1

Expression of HBP23/Prx I protein in rat liver. Cytosolic fractions of normal rat liver (50 μg) were subjected to Western blotting analysis and probed with the antibody to HBP23/Prx I used in this study. Molecular mass markers are shown at left. Note the single band at approximately 23 kD.

Cell Isolation and Culture

Primary rat hepatocytes were isolated and cultivated as described (Immenschuh et al. 1995). They were cultured under 95% air/5% CO₂ in medium M199 with Earle's salts containing 2 g/liter bovine serum albumin (BSA), 20 mM NaHCO₃, 10 mM Hepes, 100 U penicillin/ml, 100 μg streptomycin/ml, and 1 nM insulin. In addition, 5% fetal calf serum (FCS) and 10 nM dexamethasone were present during the plating phase up to 4 hr, and cell cultures were incubated in serum-free medium for another 18 hr before treatment. KCs and sinusoidal endothelial cells (SECs) were isolated as previously described (Immenschuh et al. 1999a; Neubauer et al. 2000). Cells were resuspended in culture medium M199 containing 15% FCS, 100 U penicillin/ml, and 100 μg streptomycin/ml. Cell viability was assessed by trypan blue staining. Purity of cell isolations was determined by phase-contrast microscopy, DIL-AC-LDL incorporation of SEC, and ED2 staining of KCs as previously described (Neubauer et al. 1995). KC cultures were more than 99% pure and SECs were contaminated with 2% hepatic stellate cells and with about 10% KCs. Cells were plated on six-well plates (3 × 10⁶ cells/well). After 2 hr, cells were washed for elimination of non-adherent cells. KC cultures were maintained in medium M199 (15% FCS, 100 U penicillin/ml, 100 μg streptomycin/ml) in an atmosphere of 95% air/5% CO₂ at 100% humidity. Freshly islolated cells were used for RNA isolation within 1 hr after cell preparation.

Immunohistochemistry and Immunoelectron Microscopy

Rat livers were fixed by perfusion through the portal vein as described (Fahimi 1967). The fixative contained 4% depolymerized paraformaldehyde in PBS, pH 7.4, for light microscopic studies including immunohistochemistry (IHC). For electron microscopy (EM), it consisted of a mixture of 4% depolymerized paraformaldehyde and 0.05% glutaraldehyde in PBS, pH 7.4. After perfusion, livers were cut into small blocks that were kept in the same fixatives for several hours before further processing. Whereas for IHC the tissue blocks were embedded directly in paraffin (Paraplast Plus), for EM studies microslicer sections (c. 100 μm) were embedded in LR White. For IHC studies, 1–3-μm paraffin sections were subjected first to antigen retrieval by digestion with trypsin combined with microwave irradiation as described recently (Grabenbauer et al. 2001), followed by incubation with the specific antibodies to HBP23/Prx I, HO-1, or ED2. The antigen binding sites were detected by a peroxidase-conjugated biotin-straptividin system (Extravidin; Sigma) and visualized by the peroxidase substrate Nova Red (Vector Laboratories). The nuclei were counterstained with hematoxylin. For immunoelectron microscopic studies, ultrathin sections were incubated with the antibody to HBP23/Prx I and the antigen-antibody complexes were visualized with the protein A-gold procedure (Roth 1982), as detailed elsewhere (Baumgart 1994; Fahimi et al. 1996).

Immunocytochemistry on Isolated and Cultured Cells

For ICC studies on isolated cells, cultured cells on Lab-Tek chamber slides were fixed with methanol and acetone. Slides were stored at −20C before processing for immunolocalization. Indirect immunostaining was performed with the immunoperoxidase method with diaminobenzidine (DAB) as substrate (Immenschuh et al. 1999b). In brief, cells were washed with PBS containing 0.1% BSA and were covered with FCS for 30 min. Cells were incubated with the first antibody directed either against HBP23/Prx I, HO-1, or ED2 for 45 min at 37C. After washing, cells were incubated with peroxidase-labeled IgG against murine IgG or rabbit IgG. Cells were washed and incubated with PBS containing DAB (0.5 mg/ml) and H₂O₂ (0.01%) for 10 min, washed, and counterstained with the nuclear stain hemalaum.

RNA Isolation, Northern Blotting Analysis, and Hybridization

Total RNA for Northern blotting was isolated as previously described (Immenschuh et al. 1999a). Equal quantities of RNA were separated on 1.2% agarose/2.2 M formaldehyde gels. After electrophoresis, RNA was blotted onto nitrocellulose membranes and baked at 80C for 4 hr. After prehybridization for 3–4 hr at 42C, blots were hybridized overnight at 42C with a radiolabeled 883-bp EcoR I-Hind III fragment of pRHO1, which is a plasmid with the full-length rat HO-1 cDNA probe (Shibahara et al. 1985). The cDNA fragment was radioactively labeled with α[³²P]-dCTP by random priming using the multiprime DNA labeling kit according to the manufacturer's instructions. To correct for differences in RNA loading of Northern blots, the nitrocellulose filters were stripped and were rehybridized with a 28S rRNA oligonucleotide. The 28S rRNA oligonucleotide was labeled with γ[³²P]-dATP at the 3′-end with terminal deoxynucleotide transferase. The hybridization solution contained 6 × SSC, 5 × Denhardt's solution (0.2% Ficoll 400, 0.2% polyvinylpyrrolidone, and 0.2% BSA), 0.5% SDS, 50% formamide, and 100 μg/ml denatured salmon sperm DNA. Blots were washed with 2 × SSC/0.1% SDS (once) and 0.1 × SSC/0.1% SDS (twice) at 65C. Filters were exposed to X-ray films and quantification of the autoradiograms was performed by densitometry.

Western Blotting Analysis

After washing of cell cultures twice with 0.9% NaCl, cytosol was prepared essentially as described (Immenschuh et al. 1999a). After addition of 1 ml lysis buffer (0.1% SDS, 10 mM Tris, pH 7.4), cells were boiled for 5 min and homogenized by passing through a 25G needle. The homogenate was centrifuged for 5 min at 4C and the protein content was determined in the supernatant using the Bradford method. A total of 50 μg of total protein was loaded onto a 12% sodium dodecyl sulfate (SDS)-polyacrylamide gel and was blotted onto nitrocellulose membranes by electrophoresis. Membranes were blocked with Tris-buffered saline containing 1% BSA, 10 mM Tris-HCl (pH 7.5), and 0.1% Tween-20 for 1 hr at RT. The primary antibodies for HBP23/Prx I or HO-1 were added at 1:1000 dilution and the blot was incubated for 12 hr at 4C. The secondary anti-rabbit IgG was diluted 1:8000 and a chemiluminescent detection system for Western blotting was used according to the manufacturer's instructions (Amersham-Buchler; Poole, UK). Filters were exposed to X-ray films and quantification of the autoradiograms was performed by densitometry.

Results

IHC Localization of HBP23/Prx I and HO-1 in Rat Liver

High levels of HBP23/Prx I gene expression have previously been demonstrated in liver as compared to kidney, brain or testis (Iwahara et al. 1995). In agreement with those findings, livers probed with a monospecific polyclonal antibody directed against rat HBP23/Prx I (Figure 1) exhibited uniformly strong staining in liver parenchymal cells and in sinusoidal lining cells (SLCs) as well as in bile duct epithelial cells (Figures 2A and 2B). In contrast, staining for HO-1 protein expression was restricted to KCs (Figure 2C). The identity of KCs was confirmed by positive staining with MAb ED2, which is a marker for mononuclear phagocytes (Dijkstra et al. 1985) (data not shown). The control sections processed in the complete incubation medium but without the specific antibodies were consistently negative (Figure 2D).

Immunoelectron Microscopy

The intracellular localization of HBP23/Prx I was also investigated by immunoelectron microscopy using the protein A-gold technique. Gold particles representing antigenic sites for HBP23/Prx I were found in both the nucleus and the cytoplasm of liver parenchymal cells with prominent labeling also of the mitochondrial and peroxisomal matrix (Figures 3A and 3B). In the nucleus, the gold particles were confined to the euchromatin regions and the nucleolus, while the condensed chromatin remained unlabeled (Figure 3A). Moreover, in the cytoplasm the endomembrane system with endoplasmic reticulum and Golgi complex was negative (Figure 3C). In addition to parenchymal cells, gold labeling was also observed in all sinusoidal cells, including endothelial cells (Figure 4A), KCs (Figure 4B), hepatic stellate cells (Figure 4C), and bile duct epithelial cells (Figure 4D). The subcellular distribution of HBP23/Prx I in non-parenchymal cells was essentially similar to that in hepatocytes, with prominent labeling of nucleus and cytoplasm. The identification of these cells was based on their ultrastructural characteristics and their location in the sinusoids (Fahimi 1982) with KCs present mostly in the lumen of sinusoids (Figure 4B), the EC forming the fenestrated lining (Figure 4A), and the HSC with the typical lipid vacuoles in the perisinusoidal space (Figure 4C). Control sections incubated with nonspecific IgG (negative controls) were essentially unlabeled except for very rare gold particles distributed randomly (Figure 3D). The positive controls incubated with antibodies to catalase or lactate dehydrogenase showed specific labeling of peroxisomes (Figure 3E) or the cytoplasm (Figure 3F) as described previously (Beier et al. 1988; Baumgart et al. 1996).

Figure 2

IHC localization of HBP23/Prx I and HO-1 in rat liver. Rat liver sections were incubated with polyclonal rabbit antibodies against rat HBP23/Prx I (A,B), rat HO-1 (C), or with control serum (D) as primary antibodies. The antigen binding sites were detected by a peroxidase-conjugated biotin-straptividin system (Extravidin; Sigma) and visualized by the peroxidase substrate Nova Red (Vector Laboratories). The nuclei were counterstained with hematoxylin. The liver parenchymal cells, sinusoidal cells (SLCs), and bile duct epithelial cells (BDs) are prominently stained for HBP23/Prx I, while the HO-1 immunostaining is confined to KC. PV, portal vein; THV, terminal hepatic venule. Original magnifications: A,C,D ×250; B ×650.

Gene Expression of HBP23/Prx I and HO-1 in Cultures of Isolated Parenchymal and Non-parenchymal Liver Cells

To further substantiate the cellular expression of HBP23/Prx I and HO-1 in different liver cell populations, the expression of HBP23/Prx I and HO-1 was determined by ICC in cell cultures of hepatocytes and KCs. Whereas in 24-hr cultures of parenchymal cells HBP23/Prx I exhibited a uniformly strong diffuse cytoplasmic staining (Figure 5A), HO-1 was expressed only in some hepatocytes, showing weaker and patchy staining pattern (Figure 5B). In KC cultures HBP23/Prx I and HO-1 protein were both detected by ICC, with a stronger signal for the latter (Figures 5C and 5D). HBP23/Prx I and HO-1 protein expression in cultured hepatocytes and KCs was also determined by Western blotting analysis, which confirmed their presence in both cell types (Figure 6A). By quantification of immunoblots, a higher relative level of protein expression was observed for both proteins in KCs than in hepatocytes (Figure 6B).

Figure 3

Electron microscopic immunocytochemical localization of HBP23/Prx I in parenchymal cells of rat liver. (A–C) Ultrathin sections were incubated with a polyclonal rabbit antibody to HBP23/Prx I followed by the protein A-gold procedure. (A) Gold particles representing HBP23/Prx I immunoreactivity are present in the nucleus. A higher density of gold particles was observed in the euchromatin (EC) and nucleolus (NOL) compared to the heterochromatin (HC) and nucleolus-associated heterochromatin (∗). NOL, nucleolus. (B,C) Gold particles are observed over the cytoplasm, nucleus (NUC), peroxisomes (PO), and mitochondria (MIT). No labeling was observed in the Golgi complex (Golgi) and endoplasmic reticulum (ER). (D) As a negative control, unspecific IgG antibodies were used for incubation followed by the protein A-gold procedure. Only rare gold particles were detected, indicated by white arrows. (E,F) Antibodies for either catalase or lactate dehydrogenase were used as positive controls. (E) Gold particles representing catalase were exclusively observed in peroxisomes (PO) and lactate dehydrogenase was mainly present in the cytoplasm (F). Bars = 1 μm.

Figure 4

Electron microscopic immunocytochemical localization of HBP23/Prx I in non-parenchymal cells of rat liver. Ultrathin sections were incubated with a polyclonal rabbit antibody to HBP23/Prx I, followed by the protein A-gold procedure. Gold particles representing antigenic sites for HBP23/Prx I are observed (A) in endothelial cells (END), (B) in KCs, (C) in hepatic stellate cells (HSCs), and (D) in bile duct epithelial cells (BDEPs). SD, space of Disse; F, fenestrations of endothelial cells; Lip, lipid droplets in stellate cells; BD, bile duct. Bars = 1 μm.

As determined by Northern blotting analysis, HBP23/Prx I mRNA levels were similar in freshly isolated hepatocytes and KCs, whereas a higher level of mRNA expression was observed in freshly isolated SECs (Figure 7). By contrast, HO-1 mRNA expression was high only in freshly isolated KCs, being hardly detectable in hepatocytes and SEC (Figure 7). Primary hepatocytes and KCs were cultured for up to 5 days and mRNA steady-state levels were analyzed after various times of cell culture. HBP23/Prx I mRNA was strongly expressed for up to 5 days in both hepatocytes and KCs (Figure 8). A different mRNA expression pattern was observed for HO-1, which gave a strong signal in parenchymal cells cultured for 24 hr, with decline of mRNA levels up to day 5. By contrast, HO-1 mRNA expression levels in KC remained consistently high throughout 5 days of cell culture, which is in agreement with earlier observations from our laboratory (Immenschuh et al. 1999b).

Figure 5

ICC localization of HBP23/Prx I and HO-1 in cultures of primary rat hepatocytes and KCs. Hepatocytes were prepared from livers of in vivo-perfused rats and were cultured as described in Materials and Methods. After 24 hr of cell culture, hepatocytes (A,B) or KCs (C,D) were fixed and incubated with primary polyclonal antibodies against rat HBP23/Prx I (A,C) and rat HO-1 (B,D). A peroxidase-labeled anti-rabbit antibody was applied for detection of antigenic sites, followed by the peroxidase reaction with DAB. Note the uniform strong staining of hepatocytes for HBP23/Prx I (A) in contrast to their patchy staining for HO-1 (B) and the much stronger reaction for HO-1 in KCs (D) than in hepatocytes (C). Original magnifications ×500.

Discussion

HBP23/Prx I and HO-1 are antioxidant proteins that are coordinately induced by oxidative stress in different cell culture systems (Immenschuh et al. 1995, 1999a; Siow et al. 1995; Nakaso et al. 2000). Here, the cellular expression patterns of both proteins were investigated in rat liver. Our results reveal a differential gene expression pattern in parenchymal and non-parenchymal cells of rat liver.

Localization of HBP23/Prx I and HO-1 Gene Expression in Various Liver Cell Populations

Gene expression of HBP23/Prx I was uniformly strong in both parenchymal cells and non-parenchymal cells of rat liver (Figure 2). The hepatic gene expression pattern of HBP23/Prx I in whole liver was also confirmed quantitatively by mRNA expression analysis and Western blotting in freshly isolated and cultured liver cell populations (Figures 6–8). HBP23/Prx I was consistently expressed over 5 days of cell culture in hepatocytes and exhibited an even higher expression level in cultured KCs (Figure 8). A cell type-specific expression of HBP23/Prx I mRNA and protein in different rat organs has recently been shown. Therefore, HBP23/Prx I protein was widely detectable in the peripheral and central nervous systems of rat, with the strongest expression in oligodendrocytes and Schwann cells (Mizusawa et al. 2000). In mouse intestine, HBP23/Prx I is localized in columnar epithelial cells of the middle to lower part of the intestinal villi, whereas less intense immunoreactivity for HBP23/Prx I is observed in Paneth cells, goblet cells, and myofibroblasts (Ishii et al. 2000). In rat kidney, Oberley et al. (2001) found heavy immunostaining for HBP23/Prx I in proximal tubules, with weaker reaction in glomeruli, distal tubules, and transitional epithelium of renal pelvis.

Figure 6

HBP23/Prx I and HO-1 protein levels in cultured hepatocytes and KCs of rat liver. (A) Cytosolic fractions of hepatocytes and KCs cultured for 24 hr were subjected to Western blotting analysis. Polyclonal antibodies against rat HBP23/Prx I (upper panel) or rat HO-1 (lower panel) were applied as primary antibodies and the antigen-antibody complexes were detected by the chemiluminescence technique. Molecular mass markers are shown at the left. A representative blot is shown. (B) HBP23/Prx I and HO-1 protein expression from (A) was quantitated by densitometry. The protein level in hepatocytes was set equal to 100%. Values are means ± SEM from three independent experiments. Statistics, Student's t-test for paired values: ∗, significant difference KC vs hepatocytes, p≤0.05. Hep, hepatocytes.

In contrast to HBP23/Prx I, a different pattern of hepatic gene expression was observed for HO-1. Strong HO-1 immunoreactivity was detected in KCs, but not in parenchymal cells of normal rat liver (Figure 2C), thus confirming earlier observations (Bauer et al. 1998; Goda et al. 1998; Immenschuh et al. 1999b). Different levels of HBP23/Prx I and HO-1 mRNA expression were observed in freshly isolated and cultured liver cell populations (Figure 8). HBP23/Prx I gene expression was consistently high in cell cultures of hepatocytes and KCs for up to 5 days, whereas gene expression of HO-1 decreased in hepatocytes within 5 days in culture. These findings essentially agree with reports that have consistently shown higher levels of HO-1 expression in KCs, in contrast to lower HO-1 gene expression in other liver cell populations (Bissell et al. 1972; Bauer et al. 1998; Goda et al. 1998; Immenschuh et al. 1999b). The kinetics of HO-1 mRNA expression in cultured hepatocytes, with a marked increase from day 0 to day 1 and a continous decrease thereafter, might be explained by the cell isolation procedure that can be considered a stress stimulus per se that may induce HO-1 gene expression. In addition, Schuetz et al. (1988) have previously shown that expression of various genes can be markedly affected by the hepatocyte culture conditions, such as the use of various basement membrane matrices. Nevertheless, evidence of HO-1 immunostaining was observed in liver parenchymal cells of oxidant-stressed livers (unpublished observations), confirming the inducibility of HO-1 expression also in hepatocytes (Immenschuh et al. 1995; Bauer et al. 1998).

Figure 7

HBP23/Prx I and HO-1 mRNA expression levels in freshly isolated cells of rat liver. (A) Various liver cell populations were isolated as described in Materials and Methods. Total RNA (15 μg) from freshly isolated hepatocytes (Lane 1), KCs (Lane 2) and SECs (Lane 3) was subjected to Northern blottiing analysis. Blots were sequentially probed with radiolabeled cDNAs for HBP23/Prx I, HO-1, and an oligonucleotide for the 28S rRNA. The result of a representative Northern blot from three independent experiments is demonstrated. (B) HBP23/Prx I and HO-1 mRNA expression from (A) was quantitated by densitometry and the signal of the 28S rRNA served as an internal standard. The HBP23/Prx I mRNA level in hepatocytes and the HO-1 mRNA expression in SEC was set equal to 100%. Values ± SEM represent means of relative HBP23/Prx I and HO-1 mRNA levels normalized to 28S rRNA from three independent experiments. Hep, hepatocytes.

Figure 8

HBP23/Prx I and HO-1 mRNA expression levels in cultured hepatocytes and KCs of rat liver. (A) Various liver cell populations were isolated as described in Materials and Methods. Total RNA (15 μg) from cell cultures of hepatocytes (Lanes 1–4) and KC (Lanes 5–8) at the times indicated was subjected to Northern blotting analysis. Blots were sequentially probed with radiolabeled cDNAs for HBP23/Prx I, HO-1, and an oligonucleotide for the 28S rRNA. The result of a representative Northern blot from three independent experiments is demonstrated. (B,C) HBP23/Prx I (B) or HO-1 (C) mRNA expression was quantitated by densitometry and the signal of the 28S rRNA served as an internal standard. The HBP23/Prx I or HO-1 mRNA levels in hepatocytes or KCs at day 0 were set equal to 100%, respectively. Values ± SEM represent means of relative mRNA levels normalized to 28S rRNA from three independent experiments. Hep, hepatocytes.

Physiological Functions of HBP23/Prx I and HO-1 in Liver

Prxs are multifunctional antioxidant proteins that exhibit peroxidase enzyme activity, as demonstrated by overexpression in HeLa cells (Kang et al. 1998). Therefore, Prxs similar to other antioxidant enzymes may complement the cellular defense mechanisms against oxidative stress (Scandalios 1997; Oberley 2002; Shen and Nathan 2002). The antioxidant properties of HBP23/Prx I may also be related to its high binding affinity for the pro-oxidant heme. Because non-protein bound heme produces free radicals, HBP23/Prx I might contain the pro-oxidant function of heme by non-covalent binding (Vincent 1989). Alternatively, HBP23/Prx I may serve as an intracellular transporter of heme, as was previously shown for glutathione-S-transferases (Muller–Eberhard and Nikkila 1989; Boyer and Olsen 1991). The localization of HBP23/Prx I in mitochondria and peroxisomes (Figures 3 and 4), which contain substantial amounts of cytochromes and catalase, two hemoproteins, would support either one of the above-mentioned notions. Moreover, Prxs have been shown to regulate signal transduction pathways and to modulate cell proliferation and differentiation (Jin et al. 1997; Wen and Van Etten 1997; Kang et al. 1998; Seo et al. 2000; Chang et al. 2002).

The major physiological role of hepatic HO-1 appears to be the regulation of cellular heme homeostasis (Maines 1997), and it has been shown in vivo that HO can counteract heme-induced inflammation in liver (Wagener et al. 2001). In addition, HO-1-derived carbon monoxide is considered an important signaling gas in liver (Suematsu and Ishimura 2000), with the HO-product bilirubin functioning as a radical scavenger in vitro (Stocker et al. 1987).

Subcellular Localization of HBP23/Prx I in Liver Parenchymal Cells

Cell fractionation has been used to detect subcellular distribution of Prxs and, in most cases (five of six Prxs), the proteins were recovered in cytosol except for Prx III, which was found exclusively in mitochondria (for review see Rhee et al. 1999; Fujii and Ikeda 2002). In rat kidney, Oberley et al. (2001) have shown by immunoelectron microscopy that Prxs and thioredoxins exhibit a distinct localization pattern with Prx I and II in nucleus and cytoplasm, Prx III and V in mitochondria and Prx IV in lysosomes. Independently, a similar subcellular distribution of Prxs has recently been demonstrated in human mesotheliomas (Kinnula et al. 2002). Our observations, using the immunogold technique, have clearly revealed the presence of HBP23/Prx I not only in the cytoplasm and nucleus but also in the matrix of mitochondria and peroxisomes of liver parenchymal cells (Figures 3 and 4). This further emphasizes the important anti-oxidative role of HBP23/Prx I because substantial amounts of reactive oxygen species are generated in mitochondria (Chance et al. 1979) and peroxisomes (Singh 1996). By contrast, HO-1 has previously been demonstrated to be localized in the endoplasmic reticulum (Shibahara et al. 1980) which, according to our observations, appears to lack HBP23/Prx I in hepatocytes.

In conclusion, the data show a differential localization of HBP23/Prx I and HO-1 in various liver cell populations. The expression patterns of these antioxidant stress proteins suggest a complementary physiological role in normal liver.

Footnotes

Acknowledgements

We wish to acknowledge grant supports: from the Deutsche Forschungsgemeinschaft, Bonn, Germany, SFB 402 A8 (SI), C6 (GR), and SFB 601 B1 (EB, HDF).

We wish to thank Dr A. Schad for electronic digital image processing. The technical assistance of Heike Steininger, Gabi Krämer, E. Richter, W. Stöchmann, and Inge Frommer is gratefully acknowledged.

References

Bauer

Wanner

Rensing

Alte

Miescher

Wolf

Pannen

(1998) Expression pattern of heme oxygenase isoenzymes 1 and 2 in normal and stress-exposed rat liver. Hepatology 27:829–838.

Baumgart

(1994) Morphology of peroxisomes in light- and electron microscopy. In Bugaut

Latruffe

, eds. Peroxisomes, Biochemistry, Molecular Biology, and Genetic Diseases. Heidelberg, Springer-Verlag, 37–57.

Baumgart

Fahimi

Stich

Volkl

(1996) L-lactate dehydrogenase A4- and A3B isoforms are bona fide peroxisomal enzymes in rat liver. Evidence for involvement in intraperoxisomal NADH reoxidation. J Biol Chem 271:3846–3855.

Beier

Volkl

Hashimoto

Fahimi

(1988) Selective induction of peroxisomal enzymes by the hypolipidemic drug bezafibrate. Detection of modulations by automatic image analysis in conjunction with immunoelectron microscopy and immunoblotting. Eur J Cell Biol 46:383–393.

Bissell

Hammaker

Schmid

(1972) Liver sinusoidal cells. Identification of a subpopulation for erythrocyte catabolism. J Cell Biol 54:107–119.

Boyer

Olsen

(1991) Role of glutathione S-transferases in heme transport. Biochem Pharmacol 42:188–190.

Butterfield

Merino

Golub

Shau

(1999) From cytoprotection to tumor suppression: the multifactorial role of peroxire-doxins. Antioxid Redox Signal 1:385–402.

Chae

Chung

Rhee

(1994a) Thioredoxin-dependent peroxide reductase from yeast. J Biol Chem 269:27670–27678.

Chae

Kim

Rhee

(1993) Cloning, sequencing, and mutation of thiol-specific antioxidant gene of Saccharomyces cerevisiae. J Biol Chem 268:16815–16821.

10.

Chae

Robison

Poole

Church

Storz

Rhee

(1994b) Cloning and sequencing of thiol-specific antioxidant from mammalian brain: alkyl hydroperoxide reductase and thiol-specific antioxidant define a large family of antioxidant enzymes. Proc Natl Acad Sci USA 91:7017–7021.

11.

Chance

Sies

Boveris

(1979) Hydroperoxide metabolism in mammalian organs. Physiol Rev 59:527–605.

12.

Chang

Jeong

Choi

Kang

Rhee

(2002) Regulation of peroxiredoxin I activity by Cdc2-mediated phosphorylation. J Biol Chem 277:25370–25376.

13.

Dijkstra

Döpp

Joling

Kraal

(1985) The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies. Immunology 54:589–599.

14.

Fahimi

(1967) Perfusion and immersion fixation of rat liver with glutaraldehyde. Lab Invest 16:736–750.

15.

Fahimi

(1982) Sinusoidal endothelial cells and perisinusoidal fat-storing cells: structure and function. In Arias

Popper

Schachter

Schafritz

, eds. The Liver: Biology and Pathobiolgy. New York, Raven Press, 495–506.

16.

Fahimi

Reich

Volkl

Baumgart

(1996) Contributions of the immunogold technique to investigation of the biology of peroxisomes. Histochem Cell Biol 106:105–114.

17.

Fujii

Ikeda

(2002) Advances in our understanding of peroxiredoxin, a multifunctional, mammalian redox protein. Redox Rep 7:123–130.

18.

Goda

Suzuki

Naito

Takeoka

Tsuchida

Ishimura

Tamamtani

Suematsu

(1998) Distribution of heme oxygen-ase isoforms in rat liver; Topographic basis for carbon monoxide-mediated microvascular relaxation. J Clin Invest 101:604–612.

19.

Grabenbauer

Fahimi

Baumgart

(2001) Detection of per-oxisomal proteins and their mRNAs in serial sections of fetal and newborn mouse organs. J Histochem Cytochem 49:155–164.

20.

Hirotsu

Abe

Okada

Nagahara

Hori

Nishino

Hakoshima

(1999) Crystal structure of a multifunctional 2-Cys peroxiredoxin heme-binding protein 23 kDa/proliferation-associated gene product. Proc Natl Acad Sci USA 96:12333–12338.

21.

Hofmann

Hecht

H-J

Flohe

(2002) Peroxiredoxins. Biol Chem 383:347–364.

22.

Immenschuh

Iwahara

S-I

Satoh

Nell

Katz

Muller–Eberhard

(1995) Expression of the mRNA of heme-binding protein 23 is coordinated with that of heme oxygenase-1 by heme and heavy metals in primary rat hepatocytes and hepatoma cells. Biochemistry 34:13407–13411.

23.

Immenschuh

Stritzke

Iwahara

S-I

Ramadori

(1999a) Up-regulation of heme-binding protein 23 (HBP23) gene expression by lipopolysaccharide is mediated via a nitric oxide-dependent signaling pathway in rat Kupffer cells. Hepatology 30:118–127.

24.

Immenschuh

Tan

Ramadori

(1999b) Nitric oxide mediates the lipopolysaccharide dependent upregulation of the heme oxygenase-1 gene expression in cultured rat Kupffer cells. J Hepatol 30:61–69.

25.

Ishii

Itoh

Akasaka

Yanagawa

Takahashi

Yoshida

Bannai

Yamamoto

(2000) Induction of murine intestinal and hepatic peroxiredoxin MSP23 by dietary butylated hydroxyanisole. Carcinogenesis 21:1013–1016.

26.

Ishii

Yamada

Sato

Matsue

Taketani

Nakayama

Sugita

(1993) Cloning and characterization of a 23-kDa stress induced mouse peritoneal macrophage protein. J Biol Chem 268:18633–18636.

27.

Iwahara

Satoh

Song

D-X

Webb

Burlingame

Nagae

Muller–Eberhard

(1995) Purification, characterization and cloning of a heme-binding protein (23kDa) in rat liver cytosol. Biochemistry 34:13398–13406.

28.

Jin

Chae

Rhee

Jeang

(1997) Regulatory role for a novel human thioredoxin peroxidase in NF-kappaB activation. J Biol Chem 272:30952–30961.

29.

Kang

Chae

Seo

Kim

Baines

Rhee

(1998) Mammalian peroxiredoxin isoforms can reduce hydrogen peroxide generated in response to growth factors and tumor necrosis factor-alpha. J Biol Chem 273:6297–6302.

30.

Kinnula

Lehtonen

Sormunen

Kaarteenaho–Wiik

Kang

Rhee

Soini

(2002) Overexpression of peroxiredoxins I, II, III, V, and VI in malignant mesothelioma. J Pathol 196:316–323.

31.

Maines

(1997) The heme oxygenase system: a regulator of second messenger gases. Annu Rev Pharmacol Toxicol 37:517–554.

32.

Mizusawa

Ishii

Bannai

(2000) Peroxiredoxin I (macrophage 23 kDa stress protein) is highly and widely expressed in the rat nervous system. Neurosci Lett 283:57–60.

33.

Muller–Eberhard

Nikkila

(1989) Transport of tetrapyrroles by proteins. Semin Hematol 26:86–104.

34.

Nakaso

Kitayama

Mizuta

Fukuda

Ishii

Nakashima

Yamada

(2000) Co-induction of heme oxygenase-1 and peroxiredoxin I in astrocytes and microglia around hemorrhagic region in the rat brain. Neurosci Lett 293:49–52.

35.

Neubauer

Knittel

Armbrust

Ramadori

(1995) Accumulation and cellular localization of fibrinogen/fibrin during short-term and long-term rat liver injury. Gastroenterology 108:1124–1135.

36.

Neubauer

Wilfling

Ritzel

Ramadori

(2000) Platelet endothelial cell adhesion molecule-1 gene expression in liver sinusoidal endothelial cells during liver injury and repair. J Hepatol 32:921–932.

37.

Oberley

(2002) Oxidative damage and cancer. Am J Pathol 160:403–408.

38.

Oberley

Verwiebe

Zhong

Kang

Rhee

(2001) Localization of the thioredoxin system in normal rat kidney. Free Radic Biol Med 30:412–424.

39.

Otterbein

Choi

(2000) Heme oxygenase: colors of defense against cellular stress. Am J Physiol 279:L1029–1037.

40.

Rhee

Kang

Netto

Seo

Stadtman

(1999) A family of novel peroxidases, peroxiredoxins. Biofactors 10:207–209.

41.

Roth

(1982) The protein A-gold (pAg) technique, a quantitative and qualitative approach for antigen localization on thin sections. In Bullock

Petrusz

, eds. Techniques in Immunocytochemistry. New York, Academic Press, 107–133.

42.

Scandalios

(1997) Oxidative Stress and the Molecular Biology of Antioxidant Defenses. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press.

43.

Schuetz

Omiecinski

Muller–Eberhard

Kleinman

Elswick

Guzelian

(1988) Regulation of gene expression in adult rat hepatocytes cultured on a basement membrane matrix. J Cell Physiol 134:309–323.

44.

Seo

Kang

Kim

Baines

Lee

Rhee

(2000) Identification of a new type of mammalian peroxiredoxin that forms an intramolecular disulfide as a reaction intermediate. J Biol Chem 275:20346–20354.

45.

Shen

Nathan

(2002) Nonredundant antioxidant defense by multiple two-cysteine peroxiredoxins in human prostate cancer cells. Mol Med 8:95–102.

46.

Shibahara

Müller

Taguchi

Yoshida

(1985) Cloning and expression of cDNA for rat heme oxygenase. Proc Natl Acad Sci USA 82:7865–7869.

47.

Shibahara

Yoshida

Kikuchi

(1980) Intracellular site of synthesis of microsomal heme oxygenase in pig spleen. J Biochem (Tokyo) 88:45–50.

48.

Singh

(1996) Mammalian peroxisomes: metabolism of oxygen and reactive oxygen species. Ann NY Acad Sci 804:612–627.

49.

Siow

RCM

Ishii

Sato

Taketani

Leake

Sweiry

Pearson

(1995) Induction of the antioxidant stress proteins heme oxygenase-1 and MSP23 by stress agents and oxidised LDL in cultured vascular smooth muscle cells. FEBS Lett 368:239–242.

50.

Stocker

Yamamoto

McDonagh

Glazer

Ames

(1987) Bilirubin is an antioxidant of possible physiological importance. Science 235:1043–1046.

51.

Suematsu

Ishimura

(2000) The heme oxygenase-carbon monoxide system: a regulator of hepatobiliary function. Hepatology 31:3–6.

52.

Tsuji

Copeland

Jenkins

Obinata

(1995) Mammalian antioxidant protein complements alkylhydroperoxide reductase (ahpC) mutation in Escherichia coli. Biochem J 307:377–381.

53.

Vincent

(1989) Oxidative effects of heme and porphyrins on proteins and lipids. Semin Hematol 26:105–113.

54.

Wagener

Eggert

Boerman

Oyen

Verhofstad

Abraham

Adema

(2001) Heme is a potent inducer of inflammation in mice and is counteracted by heme oxygenase. Blood 98:1802–1811.

55.

Wen

S-T

Van Etten

(1997) The pag gene product, a stress-induced protein with antioxidant properties, is an Abl SH3-binding protein and a physiological inhibitor of c-Abl tyrosine kinase activity. Genes Dev 11:2456–2467.