Heme Oxygenase 1, Nuclear Factor E2–related Factor 2, and Nuclear Factor κB Are Involved in Hemin Inhibition of Type 2 Cationic Amino Acid Transporter Expression and l-Arginine Transport in Stimulated Macrophages

Immortalized murine macrophages (RAW264.7 cells) were plated in cell culture dishes (60 × 15 mm) and grown in Dulbecco’s modified Eagle’s medium (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin (Life Technologies), as we previously reported.15RAW264.7 cells were incubated in a humidified chamber at 37°C in a mixture of 95% air and 5% CO².

According to our previous report,15we used bacterial lipopolysaccharide (100 ng/ml; Escherichia coli  serotype 0127:B8; Sigma-Aldrich, St. Louis, MO) to induce CAT-2 expression and l-arginine transport. Hemin (an HO-1 inducer; Sigma-Aldrich),16tin protoporphyrin (SnPP, an HO-1 inhibitor; Sigma-Aldrich),17and hemoglobin (a carbon monoxide scavenger; Sigma-Aldrich)18were used to elucidate the effects of HO-1 induction and carbon monoxide.

To investigate the effects of HO-1 induction on lipopolysaccharide-induced CAT-2 expression and l-arginine transport, confluent cells were randomized to one of the 14 groups. Each group contained six culture dishes (n = 6). Among them, two groups of cell cultures that received either 1× phosphate-buffered saline (Life Technologies; denoted as the PBS group) or lipopolysaccharide (denoted as the LPS group) served as the negative or positive control, respectively. To elucidate the effects of HO-1, another six groups of cell cultures were treated with one of the three different doses of hemin (i.e. , low dose: 5 μm; moderate dose: 50 μm; or high dose: 500 μm) or hemin (5, 50, or 500 μm) plus SnPP (50 μm) immediately after lipopolysaccharide administration [denoted as the LPS + Hemin (5), LPS + Hemin (50), LPS + Hemin (500), LPS + Hemin (5) + SnPP, LPS + Hemin (50) + SnPP, and LPS + Hemin (500) + SnPP groups, respectively]. To elucidate the role of carbon monoxide, another three groups of cell cultures were treated with hemin (5, 50, or 500 μm) and hemoglobin (10 μm) immediately after lipopolysaccharide [denoted as the LPS + Hemin (5) + Hb, LPS + Hemin (50) + Hb, and LPS + Hemin (500) + Hb groups, respectively]. Another three groups of cell cultures that received only hemin (500 μm), SnPP (50 μm), or hemoglobin (10 μm) served as the control of the effects of hemin, SnPP, or hemoglobin (denoted as the Hemin, SnPP, and Hb groups, respectively). The dose and administration timing of lipopolysaccharide, hemin, SnPP, and hemoglobin were chosen according to previously published articles15–18and further validated by a series of preliminary studies performed in our laboratory. After exposure to lipopolysaccharide for 18 h or comparable duration in groups without lipopolysaccharide, the cell cultures were harvested.

To elucidate the effects of hemin on Nrf2 and NF-κB pathways, confluent cells were randomized to one of the eight groups: the PBS, Hemin (50 μm), SnPP, Hb, LPS, LPS + Hemin, LPS + Hemin + SnPP (50 μm), or LPS + Hemin + Hb (10 μm) groups. Hemin, SnPP, and hemoglobin were administered immediately after lipopolysaccharide. Each group contained 18 culture dishes (n = 18). Three culture dishes from each group were then harvested after they were exposed to lipopolysaccharide for 0, 15, 30, 45, 60, and 120 min or comparable duration in groups without lipopolysaccharide, respectively.

To elucidate the effects of each additive on the induction of HO-1, whole cell lysates were prepared according to our previous report.19To elucidate the effects of HO-1 induction on Nrf2 and NF-κB pathway, the nuclear and cytosolic extracts of the harvested cell cultures were prepared according to protocols that were modified from a previously published article.20In brief, cells were washed, scraped, and centrifuged at 1,500g  for 5 min. The cell pellet was resuspended in 5 ml cell lysis buffer (10 mm HEPES [pH 7.9], 1.5 mm MgCl², 10 mm KCl, 0.5 mm dithiothreitol, and 0.2 mm phenylmethylsulfonyl fluoride) and centrifuged again at 1,500g  for 5 min. Cells were resuspended again in cell lysis buffer and allowed to swell on ice for 10 min followed by homogenization. Homogenates were centrifuged at 3,300g  for 15 min at 4°C. The supernatants were saved for cytosolic extracts and the pellets for nuclear extracts. The pellets were resuspended and homogenized in three volumes of nuclear extraction buffer (20 mm HEPES [pH 7.9], 1.5 mm MgCl², 400 mm KCl, 0.5 mm dithiothreitol, 0.2 mm phenylmethylsulfonyl fluoride, and 25% glycerol). After being stirred on ice for 30 min and centrifuged at 89,000g  for 30 min, the supernatant from the nuclear suspensions were collected and concentrated in a Microcon 10 concentrator (Millipore Corporation, Burlington, MA) by centrifugation at 14,000g  for 3 h at 4°C. For preparation of cytosolic extracts, the supernatant obtained after removal of nuclei was mixed with cytoplasmic extraction buffer (30 mm HEPES [pH 7.9] at 4°C, 140 mm KCl, 3 mm MgCl²) and then centrifuged at 89,000g  for 1 h. The supernatants were collected and also concentrated in a Microcon 10 concentrator by centrifugation at 14,000g  for 1 h at 4°C. The protein concentration of each sample (including whole cell lysates, nuclear extracts, and cytosolic extracts) was measured using a BCA protein assay kit (Pierce Biotechnology, Inc., Rockford, IL).

Equal amounts of protein (65 μg) were loaded into each well of a 7.5% Tris-glycine precast polyacrylamide gel (Bio-Rad Laboratories, Hercules, CA) and separated by gel electrophoresis. The proteins were then transferred from the gel to nitrocellulose membranes (Bio-Rad). For whole cell lysates, nitrocellulose membranes were incubated overnight at 4°C in primary antibody solution of HO-1 (1:1,000 dilution, polyclonal anti-HO-1 antibody; Stressgen Bioreagents, Ann Arbor, MI), HO-2 (1:1,000 dilution, polyclonal anti-HO-2 antibody; Stressgen), CAT-2 (1:500 dilution, polyclonal anti-CAT-2 antibody, provided by Lin-Cheng Yang, M.D., Associate Professor, Department of Anesthesiology, E-DA Hospital/I-Shou University, Kaohsiung, Taiwan, Republic of China), or β-actin (as an internal standard, 1:5,000 dilution, monoclonal antiactin antibody; Chemicon International, Inc., Temecula, CA). For cytosolic extracts, the nitrocellulose membranes were incubated overnight at 4°C in primary antibody solution of phosphorylated inhibitor κBα (I-κBα), the product of I-κB degradation (1:1,000 dilution, monoclonal anti–phosphorylated I-κBα antibody; Cell Signaling Technology, Inc., Danvers, MA) or β-actin (Chemicon). For nuclear extracts, the nitrocellulose membranes were incubated overnight at 4°C in primary antibody solution of NF-κB (1:500 dilution, polyclonal anti-NF-κB p65 antibody; Cell Signaling Technology, Inc.), Nrf2 (1:250 dilution, polyclonal anti-Nrf2 antibody; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), or β-actin (Chemicon). β-Actin was used as an internal standard. Horseradish peroxidase–conjugated sheep anti-mouse immunoglobulin G antibody (Amersham Pharmacia Biotec, Inc., Piscataway, NJ) was used as a secondary antibody. Bound antibody was detected by chemiluminescence (ECL plus kit; Amersham) and high-performance chemiluminescence film (Hyperfilm; Amersham). Densitometry was used to quantify the protein band densities using National Institutes of Health software (Scion Corp., Frederic, MD).

We chose to measure the bilirubin concentrations of the culture media to determine the activity of HO, as reported by Turcanu et al.  21In brief, culture supernatant (500 μl) was collected. After being mixed with barium chloride (250 mg) and benzene (750 μl), the sample was vortexed vigorously and then centrifuged at 13,000g  for 30 min. The upper benzene layer containing bilirubin was collected. Bilirubin was determined spectrophotometrically as a difference in absorbance between 450 and 600 nm using an excitation coefficient of 27.3 mm ⁻¹cm⁻¹. HO activity was expressed as picomoles of bilirubin formed per milligram of cell protein. Protein content level was determined by a BCA protein assay kit (Pierce).

We chose to measure l-[³H]arginine uptake to determine l-arginine transport in macrophages, as we previously reported.15Macrophages were cultured for 18 h in the absence or presence of test substances. Then the cellular uptake of l-arginine was determined using previously published protocols.15,22In brief, macrophages were incubated at 37°C for 2 min in uptake solutions (137 mm NaCl, 5.4 mm KCl, 2.8 mm CaCl², 1.2 mm MgSO⁴, 10.0 mm HEPES-Tris [pH 7.4]) supplemented with 0.1 mm l-arginine containing 1.0 μCi/ml l-[³H]arginine. Uptake was stopped with ice-cold stop solution (137 mm NaCl, 14 mm Tris-HCl [pH 7.4]). Cells were then lysed in 0.5 ml Tris-Triton (0.1%) solution followed by determination of the cellular radioactivity. l-Arginine transport was expressed as picomoles of l-arginine uptake per milligram of cell protein. Protein content level was also determined by a BCA protein assay kit (Pierce).

To determine the intergroup differences, one-way analysis of variance was used. The Tukey test was used for multiple comparisons. All data were presented as mean ± SD. The significance level was set at 0.05. A commercial software package (SigmaStat for Windows; SPSS Science, Chicago, IL) was used for data analysis.

Our data revealed that HO-1 protein concentration in the PBS group was low (fig. 1). Exposure to both hemin (an HO-1 inducer) and hemoglobin (a carbon monoxide scavenger) significantly increased HO-1 expression because the HO-1 protein concentrations in the Hemin and Hb groups were significantly higher than that in the PBS group (approximately 1.8-fold and 1.4-fold higher; data not shown). In contrast, SnPP (an HO-1 inhibitor) posted no significant effect on HO-1 expression (data not shown). Exposure to lipopolysaccharide significantly increased HO-1 expression because the HO-1 protein concentrations in the LPS group were significantly higher than that in the PBS group (approximately fourfold higher; fig. 1). In a dose-dependent manner, we found that hemin significantly enhanced the lipopolysaccharide-induced increases in HO-1 expression (fig. 1). Data of the HO activity assay revealed that lipopolysaccharide also significantly increased HO activity (fig. 1). Hemin significantly enhanced the effects of lipopolysaccharide on HO activity in a dose-dependent manner (fig. 1). In addition, our data revealed that the HO-1 expression induced by lipopolysaccharide plus a low or moderate dose of hemin (i.e. , 5 or 50 μm) was further enhanced by SnPP or hemoglobin (fig. 2). In contrast, the HO-1 expression induced by lipopolysaccharide plus a high dose of hemin (i.e. , 500 μm) was not affected by either SnPP or hemoglobin (fig. 2). However, in contrast to the HO-1 data, we found that the increases in HO activity induced by lipopolysaccharide plus a low or moderate dose of hemin (i.e. , 5 μm or 50 μm) were significantly inhibited by both SnPP and hemoglobin (fig. 2). In addition, the HO activity increase induced by lipopolysaccharide plus a high dose of hemin (i.e. , 500 μm) was not affected by SnPP or hemoglobin (fig. 2).

Our data also revealed that the expression of HO-2, the constitutively expressed HO isozyme, was not affected by lipopolysaccharide, hemin, SnPP, or hemoglobin because there were no between-group differences regarding the HO-2 protein concentrations measured by immunoblotting assay (figs. 1 and 2).

Our data revealed that the CAT-2 protein concentration in the PBS group was low (fig. 3). Hemin, SnPP, or hemoglobin alone posted no significant effect on CAT-2 expression because the CAT-2 protein concentrations in the Hemin, SnPP, and Hb groups were similar to those in the PBS group (data not shown). As expected, lipopolysaccharide significantly increased the protein concentrations of CAT-2 (fig. 3). Our data also revealed that hemin, in a dose-dependent manner, significantly inhibited lipopolysaccharide-induced CAT-2 expression (fig. 3). Furthermore, the inhibitory effect of a low or moderate dose of hemin (i.e. , 5 or 50 μm) on lipopolysaccharide-induced CAT-2 expression was significantly reversed by SnPP and hemoglobin (fig. 4). In contrast, the inhibitory effect of a high dose of hemin (i.e. , 500 μm) on lipopolysaccharide-induced CAT-2 expression was not affected by SnPP and hemoglobin (fig. 4). In addition, the changes in CAT-2 activity paralleled the changes of CAT-2 expression (figs. 3 and 4).

Our data revealed that the baseline protein concentrations of Nrf2 and NF-κB p65 in nuclear extracts and phosphorylated I-κBα in cytosolic extracts in all of the eight groups were almost undetectable because the protein concentrations of Nrf2, NF-κB p65, and phosphorylated I-κBα in culture dishes that were harvested at 0 min after lipopolysaccharide exposure (fig. 5) or comparable duration in the groups without lipopolysaccharide were low (data not shown). As expected, PBS did not activate Nrf2 and NF-κB because the protein concentrations of Nrf2 and NF-κB p65 in nuclear extracts and phosphorylated I-κBα in cytosolic extracts in the PBS group were low throughout the experiment (data not shown). Exposure to hemin or hemoglobin alone but not SnPP resulted in an early increase in Nrf2 protein concentrations in the nuclear extracts at 15, 30, and 45 min after the experiment began and then gradually returned to baseline level at 60 and 120 min after the experiment began (data not shown). In contrast, exposure to hemin, SnPP, or hemoglobin alone did not affect the protein concentrations of NF-κB p65 in the nuclear extracts or phosphorylated I-κBα in the cytosolic extracts (data not shown).

Exposure to lipopolysaccharide, however, resulted in prominent increases in the protein concentrations of Nrf2, NF-κB p65, and phosphorylated I-κBα (fig. 5). Hemin significantly enhanced this lipopolysaccharide-induced increase in Nrf2 protein concentrations and, in turn, inhibited the increases in NF-κB p65 and phosphorylated I-κBα protein concentrations (fig. 5). These effects of hemin were not affected by either SnPP or hemoglobin at 15–60 min after the experiment began. However, the effects of hemin on enhancing lipopolysaccharide-induced increases in Nrf2 protein concentrations in the delayed phase of the experiment (i.e. , 120 min after lipopolysaccharide) were further accentuated by SnPP and hemoglobin (fig. 6). In addition, the effects of hemin on attenuating lipopolysaccharide-induced increases in NF-κB p65 and phosphorylated I-κBα protein concentrations in the delayed phase of the experiment (i.e. , 120 min after lipopolysaccharide) were significantly reversed by SnPP and hemoglobin. (fig. 6).

Data from this study confirmed our hypothesis that superinduction of HO-1 significantly attenuated lipopolysaccharide-induced CAT-2 expression and l-arginine transport in stimulated macrophages. Our data also illustrated that inhibiting the activity of HO-1 reversed the effects of HO-1. These data clearly demonstrated the regulatory effects of HO-1 on CAT-2 expression and l-arginine transport during sepsis. The crucial pathophysiologic role of iNOS/nitric oxide pathway during sepsis is well established.1,2As mentioned before, CAT-2-mediated l-arginine transport constitutes part of the downstream regulatory pathways of iNOS activity.5,6Data from this study thus provide clear evidence to support the concept that HO-1 superinduction has significantly antiinflammatory effects against sepsis and warrants further investigation.

Our data confirmed that both hemin and lipopolysaccharide induced HO-1 expression.10,12Hemin has been reported to induce nuclear translocation of Nrf2, a crucial step for activation of this key regulator of the antioxidant responsive element that mediates the induction of antioxidant enzymes.13Nrf2 has also been reported to mediate lipopolysaccharide-induced HO-1 expression.23These concepts were confirmed by data from this study. In addition, our data demonstrated that hemin plus lipopolysaccharide superinduced HO-1 expression. Our data also revealed that lipopolysaccharide-induced Nrf2 activation was enhanced by hemin. Judging from these data, we believe that Nrf2 activation plays a crucial role in mediating the hemin-induced HO-1 superinduction in lipopolysaccharide-stimulated macrophages.

It is well established that regulation of CAT-2 expression involves NF-κB,14the essential pathway that regulates transcription of a wide array of proinflammatory molecules.24Our data demonstrated that hemin significantly enhanced Nrf2 activation and HO-1 expression and, in turn, inhibited NF-κB activation and CAT-2 expression in lipopolysaccharide-stimulated macrophages. These data, in accord with a recent report,25highlighted the interplay between Nrf2, HO-1, NF-κB, and CAT-2 during sepsis. These data also provide clear evidence to outline the mechanisms that accounted for the antiinflammatory capacity of HO-1.

Interestingly, our data revealed that SnPP, the HO-1 inhibitor, further increased the HO-1 protein concentrations but reversed the inhibitory effects of HO-1 on CAT-2 expression and l-arginine transport that were induced by lipopolysaccharide and hemin. These data seemed to contradict the concept that superinduction of HO-1 possesses antiinflammatory capacity against sepsis. However, our HO activity data clearly demonstrated that SnPP significantly inhibited the HO activity in macrophages stimulated with lipopolysaccharide plus hemin. Judging from these data, we speculate that this SnPP-induced increase in HO-1 protein concentrations in macrophages stimulated with lipopolysaccharide plus hemin may be the result of a compensatory mechanism secondary to a significant decrease in HO-1 activity.

Heme oxygenase 1 catalyzes the degradation of heme to carbon monoxide, biliverdin, bilirubin, and iron.8These end products were once considered as toxic metabolic waste products. However, recent data seem to suggest that these HO-1 end products possess certain therapeutic potentials. For example, administration of biliverdin has been demonstrated to mitigate the dextran sodium sulfate-induced experimental colitis.26Administration of bilirubin has been reported to lead to the long-term survival of allogeneic islets.27In addition, HO-1–related endogenous carbon monoxide production was reported to protect organs against ischemia–reperfusion injury.28,29Administration of exogenous carbon monoxide has also been reported to attenuate the endotoxin-induced inflammatory response in macrophages.30Data from this study clearly demonstrated that the effect of HO-1 on inhibiting LPS-induced CAT-2 expression and l-arginine transport was significantly reversed by hemoglobin, a carbon monoxide scavenger. We also found that hemoglobin, similar to SnPP, further increased the HO-1 protein concentrations but inhibited the HO activity in macrophages stimulated with lipopolysaccharide plus hemin. These data seem to support the concept that carbon monoxide plays a crucial role in mediating the therapeutic effects of HO-1. However, because this study did not assay the concentration of carbon monoxide, no definitive conclusion can be drawn.

Our data clearly demonstrated that the regulatory effects of hemin on the expression of HO-1, CAT-2, Nrf2, and NF-κB in lipopolysaccharide-stimulated macrophages were dose dependent. In addition, these effects of hemin could be significantly reversed by SnPP (i.e. , the HO-1 inhibitor) and hemoglobin (i.e. , the carbon monoxide scavenger). However, our data also revealed that SnPP and hemoglobin did not inhibit the effects of hemin at the highest dose (i.e. , 500 μm). One possible explanation is that the dose of SnPP (i.e. , 50 μm) and the dose of hemoglobin (i.e. , 10 μm) used in this study were not potent enough to reverse the effects of hemin at the dose of 500 μm. The other possibility is that the timing of SnPP and hemoglobin administration does not allow SnPP and hemoglobin to fully exhibit their inhibiting effects on 500 μm of hemin. More studies are needed before further conclusions can be drawn.

In summary, HO-1 induction significantly inhibited CAT-2 expression and l-arginine transport in lipopolysaccharide-stimulated macrophages. The mechanisms involved activation of Nrf2 and inhibition of NF-κB. In addition, carbon monoxide mediated, at least in part, the effects of HO-1 induction on inhibiting lipopolysaccharide-induced CAT-2 expression and l-arginine transport in stimulated macrophages. The proposed mechanisms are summarized in figure 7.