| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Pulmonary, Allergy, and Critical Care Medicine, Department of Medicine, and
* Department of Pharmacology, School of Medicine, University of Pittsburgh, Pennsylvania, USA
1Correspondence: Division of Pulmonary, Allergy, and Critical Care Medicine, Dept. of Medicine, MUH 628NW, 3459 Fifth Ave., Pittsburgh, PA, 15213, USA. E-mail: Choiam{at}msx.upmc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: HO-1 • biliverdin reductase • HO activity
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. HO cleaves the 
-mesocarbon bridge of the heme molecule to yield equimolar quantities of biliverdin-IX, carbon monoxide (CO), and free iron in a reaction requiring NADPH:cytochrome P450 reductase (NPR) (1
2)
. Biliverdin-IX (BV) is subsequently converted to bilirubin-IX (BR) by NAD(P)H:biliverdin reductase (BVR, E.C. 1:3:1:24), whereas the iron is sequestered in ferritin or released with the aid of iron transporter (2
3)
. HO-1 plays a critical role in cellular and tissue defenses against oxidative stress (i.e., murine cells with a genomic deletion of the ho-1 gene display increased susceptibility to oxidative stress) (4)
. Recently, we and other investigators demonstrated that CO, a byproduct of heme catabolism, exhibits potent anti-inflammatory effects otherwise observed with HO-1 in animal models of endotoxin shock, acute lung injury, xenotransplantation, and aeroallergen-induced asthma (5
6
7
8)
. The BR formed by HO activity displays powerful antioxidant properties in vitro and in neuronal cell culture (9
10)
. Cellular depletion of BR by RNA interference against BVR markedly augments tissue levels of reactive oxygen species and causes apoptotic cell death (10)
.
The HO and nitric oxide synthase (NOS) systems display remarkable similarities, but also differences. NOS and HO exist as constitutive (eNOS, HO-2) and inducible forms (HO-1, iNOS). HO and NOS both require NADPH as an electron donor for oxygenase activity (11
12)
. The constitutive forms of the enzymes are differentially activated (i.e., the calcium-calmodulin complex stimulates NOS whereas the calcium-protein kinase C axis may activate HO-2) (12)
. NO and CO, when produced by NOS and HO, respectively, may stimulate soluble guanylyl cyclase (sGC) to form guanosine 3',5'-cyclic monophosphate (cGMP) (12)
. NO generated from iNOS in activated macrophages kills other cells, whereas HO-1 derived CO has a cytoprotective effect (6
, 13)
. Nevertheless, the toxicity of both gases at high concentrations is well known, with different mechanisms of action (13
14)
. Constitutive forms of NOS interact with caveolin-1 (endothelium) or caveolin-3 (muscle), which modulate NOS functions (15
16
17
18)
. Generally, protein-protein interactions with caveolin-1 exert an inhibitory role, with the exception of estrogen receptor-, protein kinase C-
-, and insulin-like growth factor-mediated signaling (19
20
21)
. Recently, Liou et al. showed that the inducible form of cyclo-oxygenase (COX-2) colocalized and interacted with caveolin-1 in human fibroblasts (22)
. In contrast, the interaction of caveolin-1 and COX-2 did not inhibit the activity of COX-2 enzyme. Caveolin-1 interacts with and regulates the activity of inducible NOS in human colon carcinoma cells (23)
.
Zabel et al. have shown that sGC dynamically translocates to the membrane fraction in human platelets and associates with the NOS-containing caveolae fraction in lung endothelial cells after cellular activation (24)
. Both HO and NOS may couple with sGC-dependent signaling pathways in brain regions (25)
. We therefore examined the subcellular localization of HO and the protein-protein interactions of HO with caveolins. To date, the HO enzymes have been characterized as endoplasmic reticulum (ER) proteins that are anchored to the ER by a single transmembrane domain at the extreme carboxyl terminus (1
, 12
, 26)
.
Here, we report that HO–1 interacts with caveolin-1 and localizes in part to specialized plasma membrane caveolae in rat pulmonary artery endothelial cells (PAEC). In the absence of caveolin-1 protein, the activity of HO-1 markedly increased, suggesting inhibitory regulation of the enzyme activity by caveolin-1. We also observed the localization of BVR and NPR in caveolae of PAEC. CO and BV increased the activity of BVR, indicating that caveolae may act as a platform that offers a functional and spatial proximity to heme degradation pathways in cells. Because caveolae are rich in signaling molecules, caveolae compartmentalization of heme degradation pathways may play a role in cellular protection, possibly by facilitating caveolae-mediated signaling cascades.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell culture and treatments
Primary cultures of rat pulmonary artery endothelial cells (PAEC) were prepared as described previously (27)
, and used for experiments as completely confluent monolayers at passages 7–12. PAEC were cultured in DMEM high-glucose media containing 10% FBS and antibiotics. Human hepatocytes were isolated by collagenase digestion of human donor livers not used for transplantation (28)
. Cells were plated in Dulbecco’s modified Eagle medium with 10% fetal bovine serum for 12–24 h and the medium was changed to serum-free medium before cells were used for experiments. Normal NIH 3T3 cells harboring vector alone were grown in DMEM supplemented with L-glutamine (2 mM) and 10% donor calf serum (DCS, JRH Biosciences, Lenexa, KS, USA). NIH 3T3 cells harboring caveolin-1 antisense (Cav-AS) were grown in DMEM supplemented with L-glutamine, 10% DCS, and hygromycin B (200 µg/mL) (29)
. RAW 264.7 macrophages were cotransfected (1:10) with pcDNA3.0 containing the neomycin selection marker, the expression vector pCAGGS containing caveolin-1 cDNA in the sense orientation, or with the pCAGGS control vector. Transfected cells were grown in DMEM, 10% FBS containing G418 (400 µg/mL). PAEC cells overexpressing HO-1 were prepared and cultured as described previously (30)
. Cells were grown in humidified incubators containing an atmosphere of 5% CO2 and 95% air at 37°C. Lipopolysaccharide (LPS), hemin, and tin protoporphyrin-IX (Sn-PPIX) (Frontier Scientific, Logan, UT, USA) were applied from concentrated stock solutions directly to the culture media at the final concentrations indicated.
Hypoxia exposures
Cells were exposed to hypoxic conditions in humidified, tightly sealed modular chambers (Billups-Rothenberg, Del Mar, CA, USA) filled to 1 atm with a premixed hypoxic gas mixture (1% O2, 5% CO2, 94% N2) (Valley National Gas, Pittsburgh, PA, USA), and placed at 37°C for the indicated time (31)
. Corresponding normoxic controls were maintained for equivalent times in humidified incubators filled with an atmosphere of 95% air-5% CO2.
Cell fractionation and validation of caveolae
PAEC were grown to confluence in 150 mm dishes (or three 100 mm dishes) and used for fractionation. The homogenates in MBS (25 mM 2-(N-morpholino)-ethanesulfonic acid (MES), pH 6.5, 0.15 M NaCl) containing 1% Triton X-100 were adjusted to 40% sucrose by the addition of 2 mL of 80% sucrose prepared in MBS and placed at the bottom of an ultracentrifuge tube. A 5–30% discontinuous sucrose gradient was formed above (4 mL of 5% sucrose/4 mL of 30% sucrose, both in MBS lacking detergent) and centrifuged at 39,000 rpm for 18 h in an SW41 rotor (Beckman Instruments, Palo Alto, CA, USA). A light-scattering band at the 5–30% sucrose interface was collected or fractionated into 12 subfractions (32)
. A 100 µL aliquot of each fraction was used to measure alkaline phosphatase activity. After direct addition of substrate solution (p-nitro-phenyl phosphate, R&D systems, Minneapolis, MN, USA), the absorbance at 405 nm was measured.
Western blot analysis and immunoprecipitation
The following antibodies were used for immunoblotting: monoclonal anti-caveolin-1, anti-caveolin-2, anti-flotillin-1 (BD Transduction Lab, San Diego, CA, USA), anti-heme oxygenase-1, anti-heme oxygenase-2 (Stressgen, Victoria, BC, Canada), anti-BVR (Stressgen), anti-NPR (Santa Cruz, Santa Cruz, CA, USA). Proteins were resolved by SDS-PAGE under reducing conditions, then transferred to a nitrocellulose membrane (33)
. Protein bands were visualized by staining with Ponceau S, washed with Tris-buffered saline (TBS) containing 10 mM Tris (pH 8.0), 150 mM NaCl, and 0.05% Tween 20, then placed in a blocking solution (TBS, 0.05% Tween 20, 2% nonfat dried milk, and 1% bovine serum albumin) for 1 h. The blots were incubated an additional hour with primary antibodies and washed with TBS containing 0.05% Tween 20. Blots were then incubated with horseradish peroxidase-conjugated secondary antibodies. Bound IgGs were visualized using an enhanced chemiluminescence detection kit (Amersham, Uppsala, Sweden).
To identify the protein-protein interactions between caveolin and HO, PAEC were subjected to lysis in the immunoprecipitation buffer (50 mM Tris, pH 7.5; 150 mM NaCl, 1% w/w IGEPAL CA 630, and 0.05% w/v deoxycholate) supplemented with 0.1% w/v SDS, 0.1 mM Na3VO4, and protease inhibitors. Cells were sheared by brief sonication on ice and cellular debris was removed by centrifugation at 12,000 x g for 10 min. Lysates were cleared initially by incubation with protein A/G-Sepharose (Santa Cruz) for 1 h at 4°C. Lysates were incubated with a specific polyclonal caveolin-1 antisera (Santa Cruz), HO-1 antisera (StressGen), or a preimmune rabbit IgG at a final concentration of 4 µg/mL each for 4 h at 4°C. Protein A/G-Sepharose was then added for 4 h at 4°C. Immune complexes were collected by centrifugation, washed eight times with 1 mL of the immunoprecipitation buffer lacking Na3VO4 and protease inhibitors, then disrupted by boiling in 1% SDS.
Immunocytochemistry
Immunofluorescence labeling was performed as described (33)
. Briefly, cells were fixed in 2% paraformaldehyde and immunostained with polyclonal caveolin-1 (Santa Cruz) or HO-1 (Stressgen) antibodies. Bound primary antibodies were visualized with Cy-3 or Alexa 488-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, Inc. PA, USA). Cells were viewed with an Olympus Fluoview BX 61 confocal microscope and images were collected using a DC-330S cooled CCD camera (DAGE-MTI Inc., Michigan, USA).
Enzyme activity assays
The protein concentration of cellular fractions was determined using the commercial BioRAD assay (BioRad, Hercules, CA, USA) with bovine serum albumin as the standard. The HO and BVR activity was measured by the spectrophotometric determination of bilirubin production, as described previously (2
, 34)
. For HO activity, final reaction concentrations were: 25 µM heme, 2 mM glucose 6-phosphate, 2 U glucose 6-phosphate dehydrogenase (Sigma, Type XV from Baker’s Yeast), 1 mM ß-NADPH, 1 mg/mL endothelial cell extract, and 2 mg/mL partially purified rat liver biliverdin reductase preparation.
Reaction mixtures were incubated for 60 min in a 37°C water bath in the dark. The reactions were terminated by addition of 2 volumes of chloroform (Aldrich, Milwaukee, WI, USA). Bilirubin concentration in the chloroform extracts was determined on a Beckman DU640 scanning spectrophotometer (Beckman Instruments) by measuring O.D. (464–530 nm). HO activity were reported as pmol BR/mg protein/h assuming an extinction coefficient of 40 mM–1 cm–1 for bilirubin in chloroform. The eNOS activity assay was performed as described previously (33)
. eNOS activity upon acute treatment with bradykinin (1 µM for 5 min) was expressed as nmol NO2–/106 cells.
Preparation of biliverdin reductase
Partially purified BVR was prepared from fresh rat liver with modification of the protocol of Tenhunen et al., stopping at step III (2)
. Briefly, Sprague Dawley rats were killed by CO2 inhalation. The livers were excised, washed in ice-cold PBS, and homogenized in 0.1 M KPO4, pH 7.4 containing protease inhibitors (complete mini protease inhibitor tablets, Roche Molecular Biochemicals, Indianapolis, IN, USA). The homogenates were centrifuged at 18,000 x g and the resulting supernatant was centrifuged at 104,000 x g in a Beckman Ti70.1 rotor (4°C). The resulting supernatant was recovered and brought to 40% saturation (242 g/L) with ammonium sulfate, incubated on ice for 30 min, then centrifuged 10 min at 10,000 x g. The supernatant was brought to 60% saturation by another addition of ammonium sulfate (132 g/L). The preparation was centrifuged again at 10,000 x g for 10 min at 4°C. The resulting pellet was resuspended in 0.1 M phosphate buffer, pH 7.4, then dialyzed against the same buffer for 24 h at 4°C. The preparation was diluted to 25 mg/mL with 0.25 M sucrose, 20 mM Tris, pH 7.4 containing 20% glycerol and stored at –80°C. The resulting fraction was assayed for the rate of BV conversion to BR as described above.
Statistical analysis
All values were expressed as means ± SD obtained from at least three independent experiments. Differences in measured variables between experimental and control group were assessed using the Student’s t test (Statview II Statistical Package, Abacus Concepts, Berkeley, California). Statistically significant difference was accepted at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The plasma membrane localization of constitutive NOS supports this consensus (15
16
17
18)
. To further examine this hypothesis, we first determined the localization of HO–1 in PAEC. We chose endothelial cells because they constitutively express HO-2 and can induce HO-1 in response to a variety of stimuli including oxidative stress, cytokines, and heavy metals (35
36)
. As shown in Fig. 1
A, a portion (25–40%) of HO-1 induced by LPS, heme, or hypoxia localized to the caveolae of PAEC obtained from sucrose density gradient ultracentrifugation of cell lysates. Caveolae were concentrated in fractions ranging from number 4 to 5 (Fig. 1A
). We further evaluated the gradients from unstimulated cells for the localization of additional markers. eNOS, which is known to reside in caveolae, displayed a concentrated distribution in caveolae (fractions 4–5). The mitochondrial marker cytochrome c appeared in fractions 9–12 and did not coincide with caveolin-1-containing fractions (Fig. 1A
). NPR is a well-known ER marker (37)
. NPR was distributed in ER (fraction 8) and caveolae (fractions 4–5) (Fig. 1A
). Under basal conditions HO-1, as expected, appeared only in the ER fraction (fraction 8) (Fig. 1A
). Curiously, a portion of inducible HO-1 expression after treatment with the various inducing stimuli (heme, LPS) appears in cytochrome c-containing fractions (fractions 10–11) (Fig. 1A
). These results suggest the possible appearance of inducible HO-1 in the mitochondria; however, a detailed examination of this phenomenon is beyond the scope of this study. In the absence of cellular stimulation, HO-2 was also localized, in part, to the caveolae of PAEC (data not shown). For validation of caveolae fractionation, we measured AP activity in each fraction obtained from sucrose density gradient centrifugation. In agreement with previous results (38)
, AP activity peaked in fractions 4–5 (Fig. 1B
). The quantitative distribution of proteins in each fraction was determined by Ponceau S staining of the membrane (Fig. 1B
). In PAEC expressing HO-1 from an adenoviral construct (AdHO-1), HO-1 content in the detergent-resistant fraction increased compared with that of control infected cells (Fig. 1C
). Caveolin-1 occurred almost exclusively in the detergent-resistant fraction in control and AdHO-1 infected cells. The distribution of ß-actin between the two fractions was identical in both cell types. AP activity appeared only in detergent-resistant fractions (Fig. 1C
). Because the distribution of caveolin-1 is sensitive to cholesterol depletion, we tested the location of HO-1 after treatment with 5 mM methyl-ß-cyclodextrin (CD) for 100 min. The treatment removed 
70% of the membrane cholesterol (data not shown). Most of LPS-inducible HO-1, as well as caveolin-1, disappeared from the detergent-resistant fraction after cholesterol removal in the plasma membrane (Fig. 1D
). These data, taken together, demonstrate that a portion of HO-1 localizes to the caveolin-1-containing caveolae of the plasma membrane.
|
HO-1 interacts with caveolin-1
To investigate possible protein-protein interactions, coimmunoprecipitation was carried out in PAEC. Caveolin-1 bound physically with HO-1 that was induced by hypoxia treatment (1% O2, 8 h), LPS (100 ng/mL for 18 h), or heme (20 µM for 18 h) and vice versa (Fig. 2
A). Under basal conditions, HO-2 also interacted with caveolin-1 in endothelial cells (data not shown). Compartmentalization and interaction of the proteins was confirmed under confocal microscopy. As shown in Fig. 2B
, induced HO-1 was distributed partially in the plasma membrane, cytosol, and even the nucleus. The cytosolic staining could be attributed to ER localization. However, caveolin-1 stained exclusively at the plasma membrane (Fig. 2B
). The partial interactions of the proteins shown in immunoprecipitation experiments were also supported by merged images of both under confocal microscopic examination (Fig. 2B
, right panel).
|
HO activity localizes to caveolae
To investigate whether HO activity is retained in the plasma membrane or caveolae, we isolated the plasma membrane fraction using ultracentrifugation (160,000xg) of cell lysates. We observed that HO activity appeared in cytosolic and plasma membrane fractions, but not in the nuclear fraction (Fig. 3
A). After combining the caveolae fractions (4
5)
harvested from sucrose gradients, HO activity was determined. LPS-inducible HO-1 activity appeared in caveolae, which could be abolished by addition of Sn-PPIX (20 µM) for 18 h, indicating a functional compartmentalization of HO-1 to caveolae of PAEC (Fig. 3B
). Similar results were obtained from PAEC treated with hemin (data not shown). Basal or stimulated eNOS activity was detected in the caveolae-containing fractions (Fig. 3B
, right). Immunoreactivity to HO-1 antibody was shown in the combined caveolae from PAEC treated with LPS (Fig. 3C
).
|
Caveolin-1 regulates HO activity
It is well known that caveolin-1 inhibits eNOS activity (15
, 17)
. In contrast, COX-2 and caveolin-1 complexes retained specific COX-2 activity (22)
. To elucidate the role of caveolin-1 in HO activity, we used NIH 3T3 cells expressing caveolin-1 antisense mRNA (Cav-AS). Treatment with LPS (100 ng/mL) and sodium nitroprusside (SNP) (100 µM), an NO donor, for 8 h increased HO activity in Cav-AS and control cells containing the vector alone. Longer duration of stimuli (>12 h) or a high concentration of SNP (>500 µM) resulted in apoptotic cell detachment from substrates in both cells. Surprisingly, however, the induced enzyme activity was dramatically up-regulated in Cav-AS cells compared with that of NIH 3T3 cells transfected with vector alone (Fig. 4
A). Protein expression of HO-1 in both cells increased to similar levels in response to treatment, and therefore did not account for the discrepancy of enzyme activity (Fig. 4B
). In RAW 264.7 macrophages overexpressing caveolin-1, the induction of HO activity by LPS was down-regulated compared with that in RAW 264.7 macrophages transfected with the expression vector alone (Fig. 4D
). Figure 4C
shows the relative level of caveolin-1 expression in RAW 264.7 macrophages transfected with caveolin-1-containing expression vector or expression vector alone. In summary, the above results indicate that caveolin-1 interacts with HO-1 and intimately inhibits HO enzymatic activity in cells.
|
Biliverdin reductase localizes to caveolae
In catabolizing heme, HO generates three products: CO, iron, and BV. BV is subsequently reduced to BR by BVR (1
2)
. We examined the distribution of BVR in caveolae, where BV may be produced by HO-1. BVR localized in caveolae of rat PAEC (33 kDa) (Fig. 5
) and human hepatocytes (42 kDa) (data not shown). Upon LPS treatment, BVR translocated to caveolin-1-containing fractions (4
5)
(Fig. 5)
.
|
Functional proximity of HO-1 and BVR
Yoshinaga et al. demonstrated that purified BVR and NPR can form binary or tertiary complexes with HO-1 in vitro (39)
. Recently, Wang et al. demonstrated competitive binding of NPR and BVR with HO-1 (40)
. These studies suggest that all three heme degradation enzymes should localize to the same subcellular compartment(s) when induced. To test this, we measured the LPS-inducible HO activity in caveolae (fractions 4–5) or noncaveolae (fractions 10–12) in the absence of exogenous BVR supplementation. The detection of BR production (HO activity) under these conditions implied the presence of functional endogenous BVR activity in these fractions (Fig. 6
). The distribution of BVR appears to resemble that of HO-1 (Fig. 1A
and Fig. 5
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. These dynamic structures constantly recycle between the plasma membrane, endosomes, and the trans-Golgi network. Thus, caveolae associated HO-1 may undergo internalization and recycling, in parallel with the caveolin-1 pathway. As reported previously, HO-1 and HO-2 also reside in the ER (25
26)
. The data obtained from our microscopic examination support the relevance of intracellular HO staining (Fig. 2B
).
Caveolin-1 is ubiquitous and binds to signaling molecules such as Gi protein, Src family kinase, Ha-Ras, and eNOS (41
42)
. These signaling molecules contain a common sequence motif that recognizes the caveolin-1 scaffolding domain. Our data suggest that HO-1 may behave like PKC- and COX-2 in that the molecules do not contain the typical caveolin-1 binding motif (42)
, 
XXXXX or XXXXXX ( is an aromatic residue and X is any amino acid residue). Unlike HO-1, HO-2 contains a caveolin-1 binding motif in the peptide (226FEYNMQIF233), which is highly conserved among species from mouse to human (43)
. These observations suggest that HO-1 interacts with caveolin-1 at a site distinct from the scaffolding domain. On the other hand, the interaction of HO-2 with caveolin-1 potentially involves the scaffolding domain. HO-1 contains a di-acidic motif (AspXGlu, where X is any amino acid) preceded by a TyrXX motif (where is a hydrophobic residue), which is known to be required for selective export of certain proteins from the ER into the Golgi. The TyrXX motif can be found in many membrane-associated proteins (i.e., vimentin, cystic fibrosis transmembrane regulator, glucose transporter-4, transferrin receptor, epidermal growth factor receptor, and insulin-related receptor) (44
45)
. Some of these proteins are also harbored in caveolae (41)
. These observations may explain the wide distribution of HO proteins in cytosol and membrane (36
, 46
47)
. The relevance of this motif in HO-1 sorting requires further investigation.
In addition to the anatomical distribution, HO enzymes were functionally compartmentalized to specialized plasma membrane caveolae. Due to the lower sensitivity of assay methods available so far, we could not discern the exact contribution of HO-1 and HO-2. However, the difference of BR production between the control and the SnPP-IX-treated group can be attributed to basal HO activity. Likewise, the difference between stress-induced and control groups can be attributed to inducible HO activity. (Fig. 3A, B
). Although HO-1 immunoreactivity appears in the nucleus, the nuclear fraction, unlike caveolae-containing fractions, was devoid of enzymatic activity under basal and LPS-inducible conditions. It remains possible that HO-1 may appear in the nucleus and participate in nuclear functions unrelated to enzymatic activity.
CO, like NO, is more soluble in hydrophobic environments than in water, and activates sGC (12
, 35)
. A recent study has demonstrated that sGC associates with the plasma membrane (24)
, raising the possibility that the CO/cGMP signaling cascade might be spatially organized at cellular membranes. Byproducts of HO could easily access their target molecules in caveolae. In a variety of experimental settings, CO modulated mitogen-activated protein kinases (6
7)
whose components are connected through scaffolding proteins, suggesting that CO generation may be membrane oriented. To verify this hypothesis, we first determined whether BVR could be distributed to the membrane. Stocker et al. noted that BR possesses strong antioxidant potential against peroxyl radicals (9)
. At physiological oxygen tension, BR surpassed -tocopherol as the most potent protector against lipid peroxidation. Moreover, BR serum concentrations (5–17 µM) are high enough to provide a substantial portion of the total antioxidant capacity of serum. However, the concentration of BR is very low in tissues (20–50 nM) compared with serum levels or that of established antioxidants such as glutathione. Baranano et al. proposed that an intracellular amplification cycle governed by BVR could overcome the low levels of tissue bilirubin (10)
. Here, we report another important finding that BVR is compartmentalized to caveolae of endothelial cells. These results support the proposed hypothesis that BVR and glutathione may be the principal endogenous antioxidants associated with the membrane and cytoplasmic compartments, respectively (10)
. Next, we asked whether BVR or NPR translocated to caveolae in parallel with HO-1 induction and its subsequent localization to the plasma membrane. Treatment of cells with LPS resulted in translocation of BVR to caveolin-1-containing fractions 4–5 (Fig. 5)
. We identified the presence of endogenous BVR in caveolae, which was inducible by LPS (Fig. 6)
. Likewise, a portion of NPR translocated to caveolae (Fig. 1A)
. In a functional study using a cell-impermeable NADPH:bovine serum albumin conjugate, Osada et al. revealed that NPR could partition to plasma membrane as well as ER (37)
. The above results clearly define the caveolae compartmentalization of heme degradation enzymes.
Our data using caveolin-1 knockdown cells by delivery of antisense transcripts demonstrate that caveolin-1 negatively regulates the enzyme activity of HO-1. The data not only suggest a novel molecule that modulates HO-1 activity but evoke a question: If HO-1 is cytoprotective and induced in response to stressful conditions, could the harmful stimuli also affect the caveolin-1 expression or its protein-protein interactions? Accordingly, Lei and Morris have demonstrated that LPS down-regulated caveolin-1 expression of RAW 264.7 cells in a time- and concentration-dependent manner (48)
. It appears that LPS augmented HO-1 activity bidirectionally by an increase in HO expression and a decrease in caveolin expression. Moreover, it is known that stress conditions can trigger the phosphorylation but inhibit the acylation and trafficking of caveolin-1 (49
50)
. HO-1 is notably induced by various stresses. Therefore, the interactions of HO-1 with caveolins under various experimental conditions deserve further examination.
These results suggest that the caveolae act as a platform for HO activity and all the potential functional consequences of HO activity in signaling processes. This study did not address the spatial modulation of active HO-1/-2 by caveolin. The possible interaction of caveolin-1 with other enzymes (NPR or BVR) should be studied in detail. Our study does not address the export of CO, BR, BV, or iron. The reaction products of HO may have signaling and cytoprotective functions when directed inside the cells, but at high concentrations also have toxicological sequelae that potentially require their export or efflux from the cells. Endogenous CO production may activate endogenous sGC or MAPK activities, but may also act in a paracrine fashion as it diffuses from the cells. HO-1 prevents cell death by augmenting the efflux of iron, because iron chelators protect HO-1 null fibroblasts (51)
. Whether caveolae offer an iron efflux site is not presently known. It is not clear whether the iron released from caveolae-associated HO activity remains within the cell or is exported in a caveolae-assisted process. Likewise, it is not clear whether the caveolae facilitate the export of bilirubin from the cell, another molecule that has been implied in cytoprotective and toxic processes. Therefore further studies are required to determine the fate and functional consequences of HO reaction products that are generated specifically within the caveolae compartment. Whether byproducts of HO could affect the export pathways of other molecules warrants further investigation.
Collectively, these results clearly define the caveolae compartmentalization of heme degradation enzymes (Fig. 7
) and their regulation by caveolin-1. HO/CO-mediated signaling cascades and/or the antioxidant properties of bile pigments may play important roles in pathological conditions such as inflammation. The activation of HO activity in caveolae may represent an important prerequisite in the manifestation of the anti-apoptotic, anti-inflammatory, and cytoprotective functions for HO-1 after cellular stimulation.
|
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 27, 2004. Accepted for publication March 19, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Jin, H. P. Kim, M. Chi, E. Ifedigbo, S. W. Ryter, and A. M. K. Choi Deletion of Caveolin-1 Protects against Oxidative Lung Injury via Up-Regulation of Heme Oxygenase-1 Am. J. Respir. Cell Mol. Biol., August 1, 2008; 39(2): 171 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Ryter and A. M. K. Choi Caveolin-1: a critical regulator of pulmonary vascular architecture and nitric oxide bioavailability in pulmonary hypertension Am J Physiol Lung Cell Mol Physiol, May 1, 2008; 294(5): L862 - L864. [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Lin, S. Weis, G. Yang, Y.-H. Weng, R. Helston, K. Rish, A. Smith, J. Bordner, T. Polte, F. Gaunitz, et al. Heme Oxygenase-1 Protein Localizes to the Nucleus and Activates Transcription Factors Important in Oxidative Stress J. Biol. Chem., July 13, 2007; 282(28): 20621 - 20633. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D.-J. Slebos, S. W. Ryter, M. van der Toorn, F. Liu, F. Guo, C. J. Baty, J. M. Karlsson, S. C. Watkins, H. P. Kim, X. Wang, et al. Mitochondrial Localization and Function of Heme Oxygenase-1 in Cigarette Smoke-Induced Cell Death Am. J. Respir. Cell Mol. Biol., April 1, 2007; 36(4): 409 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Maniatis, V. Brovkovych, S. E. Allen, T. A. John, A. N. Shajahan, C. Tiruppathi, S. M. Vogel, R. A. Skidgel, A. B. Malik, and R. D. Minshall Novel Mechanism of Endothelial Nitric Oxide Synthase Activation Mediated by Caveolae Internalization in Endothelial Cells Circ. Res., October 13, 2006; 99(8): 870 - 877. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nakahira, H. P. Kim, X. H. Geng, A. Nakao, X. Wang, N. Murase, P. F. Drain, X. Wang, M. Sasidhar, E. G. Nabel, et al. Carbon monoxide differentially inhibits TLR signaling pathways by regulating ROS-induced trafficking of TLRs to lipid rafts J. Exp. Med., October 2, 2006; 203(10): 2377 - 2389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. P. Converso, C. Taille, M. C. Carreras, A. Jaitovich, J. J. Poderoso, and J. Boczkowski HO-1 is located in liver mitochondria and modulates mitochondrial heme content and metabolism FASEB J, June 1, 2006; 20(8): 1236 - 1238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Ryter, J. Alam, and A. M. K. Choi Heme Oxygenase-1/Carbon Monoxide: From Basic Science to Therapeutic Applications Physiol Rev, April 1, 2006; 86(2): 583 - 650. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. M. Wang, H. P. Kim, R. Song, and A. M. K. Choi Caveolin-1 Confers Antiinflammatory Effects in Murine Macrophages via the MKK3/p38 MAPK Pathway Am. J. Respir. Cell Mol. Biol., April 1, 2006; 34(4): 434 - 442. [Abstract] [Full Text] [PDF]
|
|
