点击显示 收起
【摘要】
Objective— Thiol oxidative stress leads to macrophage dysfunction and cell injury, and has been implicated in the development of atherosclerotic lesions. We investigated if strengthening the glutathione-dependent antioxidant system in macrophages by overexpressing glutathione reductase (GR) decreases the severity of atherosclerosis.
Methods and Results— Bone marrow cells infected with retroviral vectors expressing either enhanced green fluorescent protein (EGFP) or an EGFP-fusion protein of cytosolic GR (GR cyto -EGFP) or mitochondrial GR (GR mito -EGFP) were transplanted into low-density lipoprotein receptor-deficient mice. Five weeks after bone marrow transplantation, animals were challenged with a Western diet for 10 weeks. No differences in either plasma cholesterol and triglyceride levels or peritoneal macrophage content were observed. However, mice reconstituted with either GR cyto -EGFP or GR mito -EGFP–expressing bone marrow had lesion areas ( P <0.009) that were 32% smaller than recipients of EGFP-expressing bone marrow. In cultured macrophages, adenovirus-mediated overexpression of GR cyto -EGFP or GR mito -EGFP protected cells from mitochondrial hyperpolarization induced by oxidized low-density lipoprotein.
Conclusion— This study provides direct evidence that the glutathione-dependent antioxidant system in macrophages plays a critical role in atherogenesis, and suggests that thiol oxidative stress-induced mitochondrial dysfunction contributes to macrophage injury in atherosclerotic lesions.
Thiol oxidative stress leads to macrophage dysfunction and cell injury, and has been implicated in the development of atherosclerotic lesions. We show that overexpression of mitochondrial or cytosolic glutathione reductase in macrophages decreases the severity of atherosclerosis, providing direct evidence that the glutathione-dependent antioxidant system in macrophages plays a critical role in atherogenesis.
【关键词】 atherosclerosis glutathione macrophage oxidized lowdensity lipoprotein oxidative stress
Introduction
Glutathione reductase (GR) is a homodimeric flavoprotein that maintains the cellular thiol redox state by catalyzing the reduction of glutathione disulfide (GSSG) to glutathione (GSH) with NADPH as the reducing cofactor. 1 Mammalian glutathione reductase activity is present in both the cytosol and mitochondria. These isoforms are biochemically indistinguishable, however, suggesting that both are encoded by a single nuclear gene. 2 Both the human and the mouse GR genes consist of 13 exons, and both sequences code for an N-terminal mitochondrial-targeting sequence. 3,4 Because of the fundamental role of GSH in the removal of deleterious reactive oxygen species and maintenance of the protein thiol redox state, GR is crucial to the cell?s antioxidant defense mechanism and maintenance of enzyme activities and protein functions. 5
The glutathione-dependent antioxidant system protects cells from oxidative stress-induced dysfunction and cell injury through 3 major pathways. GSH serves as an electron donor for glutathione peroxidase, which reduces hydrogen peroxide to water and organic (hydro)peroxides to the corresponding alcohols. 6,7 Glutathione peroxidase-1 also functions as a peroxynitrite reductase. 8 As a substrate for glutathione-S-transferases, GSH also contributes to the detoxification and elimination of toxic electrophilic metabolites and xenobiotics. 9 Finally, GSH selectively recycles oxidized glutaredoxin to the reduced enzyme, generating GSSG, which in turn is reduced to GSH by glutathione reductase. 10 Glutaredoxins, or thioltransferases, are small thiol-disulfide oxidoreductases that catalyze the reduction of protein disulfides and, more importantly, the reduction of mixed disulfides between GSH and protein thiols in a process referred to as deglutathionylation. 11 Together, glutaredoxins and GR form a unique protein thiol regulation and regeneration system distinct from the GSH-independent thioredoxin/thioredoxin reductase system.
Oxidative stress plays a central role in atherosclerosis. 12 However, antioxidant therapy using reactive oxygen species-scavenging antioxidants has by-and-large not proven effective in combating cardiovascular disease, 13 suggesting that any defect in the vascular antioxidant system is likely to extend beyond a simple lack in reactive oxygen species-scavenging antioxidants. Lapenna et al showed that in human atherosclerotic plaque, the GSH-dependent antioxidant system is impaired. 14 Selenium-dependent glutathione peroxidase activity was dramatically reduced, and GR activity was markedly lower in plaques than in the normal arteries. Studies in apoE-null mice showed that lipid peroxidation and atherogenesis is preceded by the depletion of GSH in the atheroma-prone aortic arch, 15 further supporting a critical role for the GSH-dependent antioxidant in protecting against atherosclerosis. Interestingly, glutathione peroxidase-1 deficiency in mice does not appear to accelerate atherosclerosis, 16 indicating that the disruption of the GSH-dependent antioxidant system in atherosclerotic lesions may occur at a more fundamental level, eg, the regeneration of GSH from GSSG by GR.
A large body of evidence now suggests that macrophage injury and macrophage foam cell death contribute to the development of atherosclerotic plaques. 17–19 The mechanisms underlying macrophage injury in vivo are still unclear, but previously we demonstrated that the GSH-dependent antioxidant system plays a critical role in protecting macrophages from oxidative stress-induced cell injury. 20–22 Oxidized low-density lipoprotein (OxLDL), a prominent component of atherosclerotic lesions, 23 is highly cytotoxic and believed to be a major factor in the development and progression of atherosclerosis. 12,18,19,24 Studies by Darley-Usmar et al 25 showed that depletion of cellular GSH enhances OxLDL cytotoxicity in macrophage-like cell lines, supporting a role for the GSH-dependent antioxidant system in protecting macrophages from OxLDL-induced cell death. We recently confirmed these findings in human monocyte-derived macrophages and demonstrated that OxLDL promotes thiol oxidative stress and cell injury by disrupting the GSH redox state. 22 Our data suggest that 3 major pathways contribute to OxLDL-induced macrophage injury: (1) depletion of GSH; (2) inhibition of GR; and (3) oxidation of protein thiols. 22 The convergence of these 3 pathways results in enhanced and sustained protein- S -glutathionylation and, subsequently, in cell death.
Inhibition of GR with pharmacological inhibitors, or with siRNA directed against GR, sensitizes macrophages to OxLDL cytotoxicity, supporting a critical role for GR in protecting macrophages from thiol oxidative stress. 22 Because OxLDL is believed to be a major factor in the development and progression of atherosclerosis, 24 we hypothesized that protecting macrophages from OxLDL-induced cell injury by preventing the collapse of the GSH redox state would decrease the severity atherosclerosis. To test this hypothesis, we overexpressed cytosolic GR in macrophages of atherosclerosis-prone LDL-receptor–deficient (LDL-R –/– ) mice. We also overexpressed GR targeted to mitochondria to examine if enhancing the regeneration of GSH from GSSG in mitochondria would protect macrophages from mitochondrial dysfunction and reduce atherosclerotic lesion formation. Our results show that enhanced expression of either cytosolic or mitochondrial GR in bone marrow-derived cells protects macrophages from OxLDL-induced mitochondrial dysfunction and reduces atherosclerotic lesion formation.
Materials and Methods
Construction of Retroviral Vectors and Virus Production
Murine stem cell virus was chosen for these studies because it achieves stable high-level gene expression in hematopoietic stem cells through a specially designed 5' long-terminal repeat from the murine stem cell virus (MSCV). 26 The retroviral vectors were generated as follows. First, we developed a Cre-loxP–compatible acceptor vector by inserting a loxp site (Creator Acceptor Vector Construction Kit; Clontech, Mountain View, Calif) into the multicloning of pMSCVneo carrying the murine stem cell virus backbone (Clontech,). Full-length cDNA for murine GR, which contained a mitochondrial targeting sequence (GR mito ), and a truncated form of the cDNA, which lacked only the mitochondrial targeting sequence (GR cyto ), 4 were cloned in frame with the expression cassette for enhanced green fluorescent protein (EGFP). The cDNAs were kindly provided by Dr Dieter Werner (German Cancer Research Center, Heidelberg, Germany). Both forms of GR are therefore expressed as C-terminal EGFP fusion proteins (GR cyto -EGFP and GR mito -EGFP). The sequences for EGFP, GR cyto -EGFP and GR mito -EGFP, were cloned into the Cre-loxP–compatible pDNR donor vectors (Clontech) and then inserted into the murine stem cell virus acceptor vector by linear recombination with Cre recombinase (Clontech). Recombinant retroviruses expressing EGFP, GR cyto -EGFP, or GR mito -EGFP were generated in the Phoenix ECO packaging cell line (ATCC, Manassas, Va). High viruses producing cells were selected for high virus production by fluorescence-activated cell sorter, and viral titers were determined in NIH 3T3 cells by flow cytometry. Viral titers were 3.6 x 10 6 cfu/mL for EGFP, 1.1 x 10 6 cfu/mL for GR cyto -EGFP, and 1.1 x 10 6 cfu/mL for GR mito -EGFP.
Transduction of Bone Marrow Cells
Three days before harvesting bone marrow cells, donor mice were injected intraperitoneally with 5-fluorouracil (150 mg/kg). Bone marrow cell suspensions were isolated from C57BL/6J mice by flushing the femurs and tibias with phosphate-buffered saline containing 2% fetal bovine serum. Cells were pooled, washed, and resuspended in DMEM with 15% fetal bovine serum. Single-cell suspensions were prepared by passing the cells through a cell strainer. Cells were plated in a 60-mm culture dishes at 2 x 10 7 /dish and incubated for 48 hours in DMEM supplemented with 15% fetal bovine serum, mr-IL-3 (10 µg/mL), mr-IL-6 (10 µg/mL), and mr-SCF (230 µg/mL). Transductions with EGFP, GR cyto -EGFP, or GR mito -EGFP retroviruses were performed by transferring bone marrow cells into Retronectin-coated dishes (Takara, Shiga, Japan) containing 1.3 mL retroviral supernatant 15% fetal bovine serum, mr-IL-3 (10 µg/mL), mr-IL-6 (10 µg/mL), mr-SCF (230 µg/mL), and polybrene (5 µg/mL). Cells were incubated for 4 hours and the transduction procedure was repeated. After incubating cells for 18 hours, a third transduction (4 hours) was performed. Transduced cells were washed with phosphate-buffered saline before tail-vein injection. An aliquot was analyzed by flow cytometry to determine the percentage of EGFP-expressing cells after transduction with retroviruses carrying EGFP, GR cyto -EGFP, or GR mito -EGFP.
Mice
Female LDLR –/– recipient mice (B6.129S7-Ldlr tmIHer /J, stock no. 002207) on a C57BL/6J background and C57BL/6J mice donor mice were obtained from The Jackson Laboratories (Bar Harbor, Me). All mice were maintained in a barrier facility and fed normal mouse laboratory diet. To induce hypercholesterolemia, mice were fed a diet supplemented with fat (21% wt/wt) and cholesterol (0.15% wt/wt; diet no. TD88137; Harlan Teklad) beginning at 5 weeks after bone marrow transplantation for a total of 10 weeks. All studies were performed with the approval of the University of Kentucky Institutional Animal Care and Use Committee.
Bone Marrow Transplantation
Two weeks before irradiation and bone marrow transplantation, recipient LDLR –/– mice were put on acidified water containing sulfamethoxazole (160 ng/mL) and Trimetoprim (32 ng/mL). Mice were maintained on antibiotics for 7 weeks. Before transplantation, recipient LDLR –/– mice received 2 equal doses of 4.5 Gy 3 hours apart (9 Gy total; Mark I-68 Irradiator). Bone marrow cells transduced with EGFP, GR cyto -EGFP, or GR mito -EGFP retroviruses were injected into the tail vein of the irradiated recipients (n=10 per group). Two mice in the GR mito -EGFP group died within 2 weeks after bone marrow transfer. Thus, for all experiments shown, animal numbers were 10, 10, and 8 for the EGFP, GR cyto -EGFP, and GR mito -EGFP groups, respectively, unless stated otherwise.
Macrophages
Resident peritoneal cells were harvested by lavage with 5 mL of ice-cold RPMI (Gibco BRL, Carlsbad, Calif) supplemented with 2 mmol/L L-alanyl- L -glutamine (GLUTAMAX-1; Gibco BRL), 1% v/v nonessential amino acids (Gibco BRL), penicillin G/streptomycin (100 U/L and 100 µg/mL, respectively; Gibco BRL), referred to from here on simply as culture medium. Cell viability was 95% in all cell preparations. Cells were plated in culture medium supplemented with 10% fetal bovine serum and cultured in a humidified atmosphere at 5% CO 2. After 3 hours, nonadherent cells were removed through washing with culture medium. Human monocyte-derived macrophages were isolated and cultured as described previously. 22
Plasma Cholesterol and Triglycerides
Mice were fasted for 4 hours before euthanasia. After macrophages were harvested, blood was drawn by cardiac puncture. Plasma cholesterol and triglyceride levels were determined using enzymatic assay kits (Wako Chemicals, Richmond, Va).
Analysis of Atherosclerosis
After mice were euthanatized, the right atrium was removed and hearts and aortas were perfused with phosphate-buffered saline through the left ventricule. Hearts were embedded in OCT and frozen on dry ice. Aortas were fixed overnight with 4% paraformaldehyde in phosphate-buffered saline, dissected from the proximal ascending aorta to the bifurcation of the iliac artery, and had adventitial fat removed. For en face analysis, aortas were stained with oil red O, opened longitudinally, pinned flat onto black paper placed over dental wax, and digitally photographed at a fixed magnification. Total aortic area and lesion areas were calculated using ImagePro Plus 5.1 (Media Cybernetics, Silver Spring, Md). As a second measure of atherosclerosis, lesions of the aortic root were analyzed. Serial sections were cut through a 400-µm segment of the aortic root. For each mouse, 8 sections (8 µm) separated by 70 µm were examined. Each sections was stained with oil red O, counterstained with hematoxylin (Vector Labs, Burlingame, Calif), and digitized. Lesion area was measured using ImagePro Plus 5.1 (Media Cybernetics) and expressed as millimeters squared. Macrophages were detected with anti-CD68 antibodies (1:300; Serotech, Westbury, NY). 27
Immunoprecipitation and Western Blotting Analysis
Adherent peritoneal macrophages (1 x 10 6 ) from each mouse were lysed at 4°C for 20 minutes in buffer A consisting of 25 mmol/L MES, pH 6.4, 150 mmol/L NaCl, 1% (v/v) Triton X-100, and 60 mmol/L octyl glucoside. Lysates from mice in each group were pooled and precleared with protein A Sepharose CL4B beads (GE Health Sciences, Piscataway, NJ) suspended in buffer A. Beads were removed by centrifugation, and immunoprecipitation was performed by adding 2 µg anti EGFP antibodies (Abcam, Cambridge, Mass) and overnight incubation at 4°C. Immune complexes were precipitated with preblocked protein A Sepharose CL4B beads, washed with buffer A, and resuspended in reduced sample buffer. Proteins were separated by SDS-PAGE, transferred onto polyvinylidene fluoride membrane (Immobilon-FL; Millipore, Bedford, Mass) and detected with antibodies directed against EGFP (Cell Signaling Technologies, Boston, Mass).
Glutathione Reductase Activity Assay
Macrophage lysates were prepared with 50 mmol/L KP i, pH=7.5, containing 1 mmol/L EDTA and 1% Triton X-100. 20 The assay was performed at 37°C in 50 mmol/L KP i, pH=7.5, containing 1 mmol/L EDTA, bovine serum albumin (1 mg/mL), and 333 µmol/L NADPH. The enzymatic reaction was started by adding 1 mmol/L GSSG, and absorbance was monitored at 340 nm for 15 minutes on a VERSAmax plate reader (Molecular Devices, Sunnyvale, Calif). GR from Baker?s yeast served as a standard.
Adenoviral Vectors
To control GR expression levels and prevent localization artifacts caused by overexpression, we constructed a doxycycline-controlled Tet-On adenoviral gene expression system. The adenoviral vector containing both the Tet-on transcriptional activator and the Tet-responsive element were kindly provided by Dr George Smith, Department of Physiology at the University of Kentucky. To facilitate the process of cloning of genes of interest into this large construct, we inserted a loxp cassette into the multicloning site of the vector. Sequences for EGFP, cytosolic GR cyto -EGFP, and mitochondrial GR mito -EGFP were first cloned into pDNR donor vectors (Clontech) and then inserted into the adenoviral vector by linear recombination using Cre recombinase (Invitrogen, Carlsbad, Calif). Expression of the GR cyto -EGFP and mitochondrial GR mito -EGFP yielded enzymatically active proteins; the latter was expressed in mitochondria, as confirmed by immunohistochemistry with antibodies directed at the mitochondrial HSP60 and by confocal microscopy ( Figure 1 B). Infection of human monocyte-derived macrophages with adenoviruses was performed in Opti MEM-I medium (Invitrogen) supplemented with 2 mmol/L L-alanyl- L -glutamine (GLUTAMAX-1; Gibco BRL), 1% v/v nonessential amino acids (Gibco BRL), penicillin G/streptomycin (100 U/L and 100 µg/mL respectively; Gibco BRL), 1% sodium pyruvate, and 2% human AB serum. After 6 hours, supernatants were replaced with culture medium containing 5% human AB serum. Transgene expression was induced with doxycycline (0.01 to 0.1 µg/mL; Calbiochem, San Diego, Calif).
Figure 1. Expression of EGFP fusion proteins of cytosolic GR (GR cyto -EGFP) and mitochondrial GR (GR mito -EGFP) in macrophages in vivo and in vitro. A, Western blot analysis of transgene expression in peritoneal macrophages isolated from lethally irradiated LDL-R –/– mice reconstituted with EGFP, GR cyto -EGFP, and GR mito -EGFP expressing bone marrow. Cell lysates from peritoneal macrophages isolated from each animal were prepared, and lysates from mice in the same group were pooled. EGFP (27 kDa), GR cyto -EGFP (79 kDa), and GR mito -EGFP (81 kDa) were immunoprecipitated from 25 µg total protein from each pool of lysates with an antibody directed against EGFP (Abcam). Proteins were separated by SDS-PAGE and immunoblots were probed with anti-EGFP antibody (Cell Signaling Technologies) as described in "Experimental Procedures." Cell lysates from HEK cells and HEK cells overexpressing GR cyto -EGFP were subjected to the same immunoprecipitation procedure and load on the same gel as controls. Results shown are representative of 2 separate SDS-PAGE/Western blot analyses. B, Subcellular localization of GR cyto -EGFP and GR mito -EGFP in adenovirus-infected macrophage. Human monocyte-derived macrophages infected with inducible adenovirus carrying the genes for EGFP, GR cyto -EGFP, and GR mito -EGFP were incubated for 30 hours with 0.01 µg/mL doxycycline to induce transgene expression (green). Mitochondria were stained with an antibody directed against the mitochondrial marker HSP60 (red). Co-localization (yellow) of GR mito -EGFP with mitochondrial HSP60 is shown in the overlay. Images were acquired by confocal microscopy and represent a single focal layer (60 x ).
JC-1 Assay
Mitochondrial function was assessed with the fluorescent marker JC-1 (Invitrogen), an indicator of mitochondrial membrane potential ( m ) and integrity. Macrophages overexpressing EGFP, GR cyto -EGFP, or GR mito -EGFP were exposed to OxLDL (75 µg/mL) for 16 hours, washed, and incubated with JC-1 (5 µg/mL) for 15 minutes at 37°C. Fluorescence was measured on a SpectraMax fluorescence plate reader set to excitation wavelengths of 485 nm (JC-1 monomers) and 535 nm (JC-1 aggregates), and emission wavelengths of 530 nm (JC-1 monomers) and 590 nm (JC-1 aggregates). JC-1 monomers, which fluoresces in the green spectrum (FL green ), accumulate in mitochondria, and with increasing m form aggregates, which fluoresce in the red spectrum (FL red ). Results are expressed as fluorescence ratios FL red /FL green and normalized to the value of untreated control macrophages.
Statistics
Data were analyzed using ANOVA (SigmaStat; SPSS Inc.). Data were tested for use of parametric or nonparametric post hoc analysis, and multiple comparisons were performed by using the Holm-Sidak method. All data are presented as mean±SE. Results were considered statistically significant at the P <0.05 level.
Results
Increased Expressions of Mitochondrial and Cytosolic GR in Bone Marrow Cells Increase GR Activity in Resident Macrophages
Expression levels of endogenous GR protein are low in macrophages. Therefore, to determine if transgene expression in macrophages was maintained throughout the experiment, lysates from equal numbers of peritoneal macrophages harvested from all mice in a group were pooled and subjected to immunoprecipitation and Western blot analysis. Expressions of EGFP, GR cyto -EGFP, and GR mito -EGFP fusion protein were confirmed in the respective macrophage lysates ( Figure 1 A), but expression levels of GR mito -EGFP appeared to be very low ( Figure 1 A). Because samples had to be pooled for Western blot analysis, these data only indicated an average expression levels for the transgene in each group. To examine whether transgene expression actually increased macrophages GR activity in individual mice, we used a highly sensitive enzymatic activity assay to measured GR activity in macrophage lysates from each mouse. Macrophages from GR cyto -EGFP mice showed a 57% ( P <0.004) increase in GR activity compared with macrophages from EGFP mice, whereas GR activity was increased by 70% ( P <0.002) in macrophages from GR mito -EGFP mice ( Figure 2 ). Our results confirm that all three transgenes were expressed throughout the duration of the experiment.
Figure 2. Glutathione reductase activity in peritoneal macrophages from LDL-R –/– mice reconstituted with EGFP, GR cyto -EGFP, or GR mito -EGFP expressing bone marrow. Peritoneal macrophages isolated from each mouse were plated for 3 hours. Cells were washed and lysed, and GR activity was measured as described in "Experimental Procedures." Results are expressed as mean±SE. * P <0.004 vs EGFP.
To confirm that GR mito -EGFP is targeted and expressed in macrophage mitochondria, we cloned the sequences for EGFP, GR cyto -EGFP, and GR mito -EGFP into inducible adenoviral vectors and infected human monocyte-derived macrophages with purified adenoviruses. Macrophages that were infected with viruses carrying the GR mito -EGFP gene and treated with 0.01 to 0.1 µg/mL doxycycline to induce transgene expression showed EGFP expression exclusively in the mitochondria ( Figure 1 B), whereas adenovirus-mediated expression of GR cyto -EGFP showed a cytosolic expression pattern ( Figure 1 B).
Increased Expression of Mitochondrial and Cytosolic GR in Bone Marrow Does Not Affect Peritoneal Macrophage Content or Plasma Lipid Parameters
Resident peritoneal macrophage yields were not different between mice that had received EGFP, GR cyto -EGFP, or GR mito -EGFP bone marrow ( Table ). This result indicates that increased expression of GR did not appear to affect monocyte extravasation or macrophage tissue content. We also observed no statistically significant differences in plasma cholesterol levels between the 3 groups ( Table ). There was a trend toward lower plasma triglyceride levels in GR cyto -EGFP and GR mito -EGFP mice compared with EGFP mice. Even though the differences plasma triglyceride levels did not reach statistical significance, we cannot rule that changes in lipoprotein composition may have contributed to the decrease in atherosclerotic lesions.
Peritoneal Macrophage Content, Plasma Total Cholesterol, and Triglyceride Levels in LDL-R –/– Mice Reconstituted With Bone Marrow Expressing EGFP, Cytosolic GR cyto -EGFP, or Mitochondrial GR mito -EGFP
Increased Expression of GR in Bone Marrow Decreases Atherosclerosis in LDL-R –/– Mice
To determine the effect of increased cytosolic or mitochondrial GR expression in hematopoietic cells on the formation of atherosclerotic lesions, we first measured the total aortic lesion area between the proximal ascending aorta and the bifurcation of the iliac artery by en face analysis of paraformaldehyde-fixed and oil red O-stained aortas. Compared with mice that received EGFP-expressing bone marrow cells, both GR cyto -EGFP and GR mito -EGFP mice showed a 32% ( P <0.009) reduction in lesion area ( Figure 3 A). To confirm the anti-atherosclerotic effects of GR cyto -EGFP and GR mito -EGFP bone marrow cells, we also quantified lesion size in the aortic root. Lesion area measurements were performed in 8 equidistant sections obtained from the aortic root of each animal ( Figure 3 B). Analysis of the mean lesion area by 2-way ANOVA showed a significant ( P =0.002) reduction in lesion size. However, the decrease in lesion burden was only statistically significant for the GR mito -EGFP group. Although we observed a similar trend for the GR cyto -EGFP group, the differences in lesion size did not reach statistical significance (unadjusted P =0.041). Immunocytochemical analysis of sections from the aortic root of each animal revealed no overt differences in the macrophage content of lesions from EGFP and those from GR cyto -EGFP and GR mito -EGFP mice ( Figure 4 ). In summary, our data support a protective effect of increased GR expression in macrophages on atherosclerotic lesion development.
Figure 3. (A) Analysis of atherosclerosis in LDL-R –/– mice reconstituted with EGFP, GR cyto -EGFP, or GR mito -EGFP expressing bone marrow. A, En face lesion analysis. Lesion size is expressed as percent of total intimal area. Sq symbols represent mean for each group±SE. EGFP (n=10), GR cyto -EGFP (n=9), and GR mito -EGFP (n=8). B Lesion area in the aortic root. Lesion area was measured in 8-µm sections separated by 70 µm after oil red O staining. Results are expressed as mean±SE. Two-way ANOVA showed that the difference in the mean values among the different transgenes was significant ( P =0.002). Unadjusted probability values obtained by pairwise multigroup comparisons (Holm-Sidak method) were GR mito -EGFP vs EGFP: P =0.00066 (significant); GR cyto -EGFP vs EGFP: P =0.041 (not significant), and GR mito -EGFP vs GR cyto -EGFP P =0.054 (not significant).
Figure 4. Representative lesions in the aortic root of LDL-R –/– mice with EGFP, GR cyto -EGFP, or GR mito -EGFP expressing bone marrow. Sections were stained with oil red O (A to C). Adjacent sections were stained with a macrophage-specific antiserum and shown at 4-fold higher magnification (D to F) as indicated by the box in the corresponding oil red O-stained section (A to C). Scale bar=500 µm.
Overexpression of Mitochondrial and Cytosolic GR in Cultured Macrophages Prevents OxLDL-Induced Mitochondrial Dysfunction
To explore potential mechanisms underlying the protective effect of cytosolic and mitochondrial GR, we examined if increased expression of GR protects macrophages from thiol oxidative stress-induced mitochondrial dysfunction. Previously we reported that atherogenic OxLDL promotes thiol oxidative stress in human macrophages by disrupting the glutathione redox buffer 22 and that OxLDL cytotoxicity in human macrophages involves mitochondrial dysfunction. 28 We found that stimulation of human macrophages with OxLDL for up to 16 hours promoted a slow but continuous increase in the membrane potential of macrophage mitochondria ( Figure 5 A). After 20 to 24 hours, coinciding with the onset of cytotoxicity, we observed a collapse of the mitochondria membrane potential, indicating the loss of mitochondrial integrity (not shown). Addition of the peroxyl radical scavenger Trolox, which prevents OxLDL cytotoxicity in macrophages, 28 prevented mitochondrial hyperpolarization induced by OxLDL, implicating OxLDL-induced peroxide formation in this process ( Figure 5 B). To examine if increased regeneration of GSH from GSSG also protects macrophages from OxLDL-induced mitochondrial dysfunction, we generated inducible adenoviral vectors to overexpress GR cyto -EGFP, GR mito -EGFP and EGFP in monocyte-derived macrophages. Doxycycline-induced expression of GR cyto -EGFP and GR mito -EGFP, but not EGFP, prevented mitochondrial hyperpolarization ( Figure 5 C), suggesting that increasing GR activity either in the cytosol or the mitochondria protects against OxLDL-induced mitochondrial dysfunction.
Figure 5. Mitochondrial dysfunction in OxLDL-stimulated macrophages. A, Time course of OxLDL-induced mitochondrial hyperpolarization. Human monocyte-derived macrophages were treated with either OxLDL (75 µg/mL, -) or the uncoupler p-trifluoromethoxy carbonyl cyanide phenyl hydrazone (25 µmol/L, ) for up to 18 hours and stained with the potentiometric fluorescent dye JC-1. Mitochondrial membrane polarization was determined as the fluorescence ratios FL red /FL green and values within each experiment were normalized to the value obtained in untreated control macrophages as described in "Experimental Procedures." * P <0.05 vs RPMI (t=0). B, Effect of Trolox on OxLDL-induced mitochondrial hyperpolarization. Macrophages were incubated for 16 hours either in the absence (open bars) or in the presence of OxLDL (75 µg/mL, solid bars), washed, and stained with the potentiometric fluorescent dye JC-1. The peroxyl radical scavenger Trolox (250 µmol/L) was present during stimulation where indicated. * P <0.05 vs control (no OxLDL, white bar). C, Effect of increased cytosolic or mitochondrial GR expression on OxLDL-induced mitochondrial hyperpolarization. Macrophages were infected with adenoviruses carrying EGFP, GR cyto -EGFP, or GR mito -EGFP, and gene expression was induced with doxycycline (0.01 mg/mL, 36 hours). Cells were then incubated for 16 hours either in the absence (white bars) or in the presence of OxLDL (75 µg/mL, black bars), washed, and stained with the potentiometric fluorescent dye JC-1. Results are expressed as mean±SE from 4 independent experiments measured in triplicate. * P <0.05 vs control (uninfected, no OxLDL, white bar); ** P <0.05 vs EGFP.
Discussion
In this study, we provide evidence that thiol oxidative stress contributes to atherogenesis and that protecting macrophages from (thiol) oxidative stress by strengthening their GSH-dependent antioxidant system decreases the severity of atherosclerotic lesion formation. Reconstitution of LDL-R –/– mice with bone morrow cells overexpressing either cytosolic or mitochondrial GR resulted in a significant reduction in the size of lesions found in the aortic arch and descending aorta as compared with mice reconstituted with EGFP-overexpressing bone marrow. While the protective effects of cytosolic or mitochondrial GR were identical in the aortic arch and descending aorta, only GR mito -EGFP significantly reduced the severity of atherosclerosis in the aortic root. The anti-atherosclerotic effect of GR cyto -EGFP was less pronounced in this vascular bed. The reason for this region-specific difference is not clear, but it suggests that the mechanisms of thiol oxidative stress are likely to differ between the different vascular beds.
Several studies demonstrated that atherosclerotic lesions are areas of increased thiol oxidative stress. Biswas et al 15 observed decreased GSH synthesis in lesion-prone areas of the aortic arch from apoE-null mice that appears to precede the onset of oxidative stress. The GSH-dependent antioxidant systems also appears to be suppressed in human atherosclerotic lesions, 14 and exogenous supplementation with GSH has shown significant benefits on the hemodynamic parameters of patients with atherosclerosis. 29 Our study demonstrates that strengthening macrophages? GSH-dependent antioxidant system by increasing their GR activity is sufficient to significantly attenuate atherosclerotic lesion formation. Plasma lipid levels were not significantly altered by the expression of the GR cyto -EGFP and GR mito -EGFP transgenes, indicating that mechanisms other than changes in plasma lipids contributed to the reduction in atherosclerotic lesions. Decreased monocyte recruitment could have accounted for reduced lesion size; however, we found no evidence that monocyte recruitment was impaired. We did not observe a reduction in the numbers of resident peritoneal macrophages in mice reconstituted with GR cyto -EGFP and GR mito -EGFP compared with EGFP-expressing mice. Furthermore, the macrophage content of lesions from these mice also did not appear to be reduced, suggesting that increasing GR activity in hematopoietic cells did not limit monocyte recruitment into the vessel wall.
We found that increasing mitochondrial GR activity in macrophages reduced lesion formation in all vascular beds examined, indicating that macrophage mitochondria might be a primary target of vascular (thiol) oxidative stress. OxLDL, which is proatherogenic and accumulates in atherosclerotic lesions, 30 promotes thiol oxidative stress in macrophages. 22 OxLDL not only depletes GSH but also inhibits GR activity and promotes protein thiol oxidation, ultimately resulting in macrophage death. Here, we demonstrated that OxLDL also promotes chronic mitochondrial dysfunction, which was prevented in macrophages overexpressing either mitochondrial or cytosolic GR. Although not shown directly, increasing macrophage GR expression in vivo may therefore protect macrophage mitochondria from thiol oxidative stress-induced injury, thereby preventing macrophage dysfunction and/or cell death and thus the progression of atherosclerotic lesions. In support of a role for mitochondrial injury in atherosclerotic lesion formation, Ballinger et al 31 reported that atherosclerosis-prone apoE-null mice deficient in mitochondrial manganese superoxide dismutase exhibit increased mitochondrial damage and accelerated atherosclerosis. Our data are also in good agreement with recent findings by Oliveira et al, 32 which suggest that a low antioxidant status of mitochondria from LDL-R –/– mice may contribute to atherogenesis. The authors provide evidence that a shift in the GSH/GSSG ratios to a more oxidized state contributes to the lower antioxidant status of mitochondria isolated from atherosclerosis-prone LDL-R –/– mice. The decreased GSH/GSSG ratio suggests that mitochondria from hypercholesterolemic LDL-R –/– mice lacked sufficient GR activity to maintain a normal GSH redox state.
In summary, our studies show for the first time that increased expression of macrophage GR activity reduces the severity of atherosclerotic lesion formation, providing evidence that the GSH redox state in macrophages plays a critical role in the development and progression of atherosclerotic lesions. Our data also suggest that strategies aimed at protecting macrophages from oxidative stress-induced mitochondrial dysfunction may suppress atherosclerosis.
Acknowledgments
The authors thank Wuqiong Ma for her technical assistance, and Kirsten Gallagher for her comments and critical reading of the manuscript.
Sources of Funding
This work was supported by grants (to R.A.) from the National Institutes of Heath (HL-70963) and the American Heart Association (455176B).
Disclosures
None.
【参考文献】
Tutic M, Lu XA, Schirmer RH, Werner D. Cloning and sequencing of mammalian glutathione reductase cDNA. Eur J Biochem. 1990; 188: 523–528.
George DL, Francke U. Gene dose effect: regional mapping of human glutathione reductase on chromosome 8. Cytogenet Cell Genet. 1976; 17: 282–286.
Kelner MJ, Montoya MA. Structural organization of the human glutathione reductase gene: determination of correct cDNA sequence and identification of a mitochondrial leader sequence. Biochem Biophys Res Commun. 2000; 269: 366–368.
Iozef R, Becker K, Boehme CC, Schirmer RH, Werner D. Assembly and functional expression of murine glutathione reductase cDNA: a sequence missing in expressed sequence tag libraries. Biochim Biophys Acta. 2000; 1500: 137–141.
Chaudiere J, Ferrari-Iliou R. Intracellular antioxidants: from chemical to biochemical mechanisms. Food Chem Toxicol. 1999; 37: 949–962.
Hayes JD, McLellan LI. Glutathione and glutathione-dependent enzymes represent a co-ordinately regulated defence against oxidative stress. Free Radic Res. 1999; 31: 273–300.
Chu FF, Esworthy RS, Doroshow JH. Role of Se-dependent glutathione peroxidases in gastrointestinal inflammation and cancer. Free Radic Biol Med. 2004; 36: 1481–1495.
Sies H, Sharov VS, Klotz LO, Briviba K. Glutathione peroxidase protects against peroxynitrite-mediated oxidations. A new function for selenoproteins as peroxynitrite reductase. J Biol Chem. 1997; 272: 27812–27817.
Commandeur JN, Stijntjes GJ, Vermeulen NP. Enzymes and transport systems involved in the formation and disposition of glutathione S-conjugates. Role in bioactivation and detoxication mechanisms of xenobiotics. Pharmacol Rev. 1995; 47: 271–330.
Srinivasan U, Mieyal PA, Mieyal JJ. pH profiles indicative of rate-limiting nucleophilic displacement in thioltransferase catalysis. Biochemistry. 1997; 36: 3199–3206.
Shelton MD, Chock PB, Mieyal JJ. Glutaredoxin: Role of reversible protein-S-glutathionylation and regulation of redox signal transduction and protein translocation. Antioxid Redox Signal. 2005; 7: 348–366.
Chisolm GM, Steinberg D. The oxidative modification hypothesis of atherogenesis: an overview. Free Radic Biol Med. 2000; 28: 1815–1826.
Asplund K. Antioxidant vitamins in the prevention of cardiovascular disease: a systematic review. J Intern Med. 2002; 251: 372–392.
Lapenna D, de Gioia S, Ciofani G, Mezzetti A, Ucchino S, Calafiore AM, Napolitano AM, Di Ilio C, Cuccurullo F. Glutathione-related antioxidant defenses in human atherosclerotic plaques. Circulation. 1998; 97: 1930–1934.
Biswas SK, Newby DE, Rahman I, Megson IL. Depressed glutathione synthesis precedes oxidative stress and atherogenesis in Apo-E(–/–) mice. Biochem Biophys Res Commun. 2005; 338: 1368–1373.
de Haan JB, Witting PK, Stefanovic N, Pete J, Daskalakis M, Kola I, Stocker R, Smolich JJ. Lack of the antioxidant glutathione peroxidase-1 does not increase atherosclerosis in C57BL/J6 mice fed a high-fat diet. J Lipid Res. 2006; 47: 1157–1167.
Kockx MM. Apoptosis in the atherosclerotic plaque: quantitative and qualitative aspects. Arterioscler Thromb Vasc Biol. 1998; 18: 1519–1522.
Mitchinson MJ, Hardwick SJ, Bennett MR. Cell death in atherosclerotic plaques. Curr Opin Lipidol. 1996; 7: 324–329.
Colles SM, Maxson JM, Carlson SG, Chisolm GM. Oxidized LDL-Induced injury and apoptosis in atherosclerosis - Potential roles for oxysterols. Trends Cardiovasc Med. 2001; 11: 131–138.
Asmis R, Wang Y, Xu l, Kisgati M, Begley JG, Mieyal JJ. A novel thiol oxidation-based mechanism for adriamycin-induced cell injury in human macrophages. FASEB J. 2005; 19: 1866–1868.
Asmis R, Qiao M, Rossi RR, Cholewa J, Xu l, Asmis LM. Adriamycin promotes macrophage dysfunction in mice. Free Radic Biol Med. 2006; 41: 165–174.
Wang Y, Qiao M, Mieyal JJ, Asmis LM, Asmis R. Molecular mechanism of glutathione-mediated protection from oxidized LDL-induced cell injury in human macrophages: Role of glutathione reductase and glutaredoxin. Free Radic Biol Med. 2006; 41: 775–785.
Palinski W, Rosenfeld ME, Yla Herttuala S, Gurtner GC, Socher SS, Butler SW, Parthasarathy S, Carew TE, Steinberg D, Witztum JL. Low density lipoprotein undergoes oxidative modification in vivo. Proc Natl Acad Sci U S A. 1989; 86: 1372–1376.
Steinberg D, Witztum JL. Lipoproteins, lipoprotein oxidation, and atherogenesis. In: Chien KR, ed. Molecular Basis of Cardiovascular Disease. Philadelphia, PA: W.B. Sanders Co.; 1998: 458–475.
Gotoh N, Graham A, Nikl E, Darley-Usmar VM. Inhibition of glutathione synthesis increases the toxicity of oxidized low-density lipoprotein to human monocytes and macrophages. Biochem J. 1993; 296 (Pt 1): 151–154.
Hawley RG, Lieu FH, Fong AZ, Hawley TS. Versatile retroviral vectors for potential use in gene therapy. Gene Ther. 1994; 1: 136–138.
Huby T, Doucet C, Dachet C, Ouzilleau B, Ueda Y, Afzal V, Rubin E, Chapman MJ, Lesnik P. Knockdown expression and hepatic deficiency reveal an atheroprotective role for SR-BI in liver and peripheral tissues. J Clin Invest. 2006; 116: 2767–2776.
Asmis R, Begley JG. Oxidized LDL promotes peroxide-mediated mitochondrial dysfunction and cell death in human macrophages: a caspase-3-independent pathway. Circ Res. 2003; 92: E20–E29.
Coppola L, Grassia A, Giunta R, Verrazzo G, Cava B, Tirelli A, D?Onofrio F. Glutathione (GSH) improved haemostatic and haemorheological parameters in atherosclerotic subjects. Drugs Exp Clin Res. 1992; 18: 493–498.
Witztum JL, Steinberg D. The oxidative modification hypothesis of atherosclerosis: does it hold for humans? Trends Cardiovasc Med. 2001; 11: 93–102.
Ballinger SW, Patterson C, Knight-Lozano CA, Burow DL, Conklin CA, Hu Z, Reuf J, Horaist C, Lebovitz R, Hunter GC, McIntyre K, Runge MS. Mitochondrial integrity and function in atherogenesis. Circulation. 2002; 106: 544–549.
Oliveira HC, Cosso RG, Alberici LC, Maciel EN, Salerno AG, Dorighello GG, Velho JA, de Faria EC, Vercesi AE. Oxidative stress in atherosclerosis-prone mouse is due to low antioxidant capacity of mitochondria. FASEB J. 2005; 19: 278–280.
作者单位：Division of Nephrology (M.Q., R.A.), University of Texas Health Science Center at San Antonio and South Texas Veterans Health Care System, San Antonio, Tex; Department of Laboratory Medicine (M.K.), Kenezy Gyula Hospital, Debrecen, Hungary; Graduate Center for Nutritional Sciences (J.M.C.) and the D