Induction of Glutathione Synthesis in Macrophages by Oxidized Low-Density Lipoproteins Is Mediated by Consensus Antioxidant Response Elements

Chemicals and Reagents

LDL (density=1.019 to 1.063 g/mL) was prepared from the plasma of fasted, healthy male donors by sequential density gradient ultracentrifugation. This protocol involving human volunteers was approved by the University of Washington Human Subjects Review Committee. Native LDL (natLDL) was dialyzed overnight in PBS, filtered, and diluted to 300 μg protein/mL in PBS containing 25 μmol/L BHT. OxLDL was prepared by diluting to 300 μg/mL natLDL in PBS without BHT and followed by incubation with 5 μmol/L CuSO₄ for 24 hours at 37°C. The oxidation reaction was stopped with addition of BHT (25 μmol/L final concentration), but the BHT in both the oxLDL and natLDL was removed by dialysis before addition to the cells and the lipoproteins were stored under nitrogen. The final preparations of oxLDL contained between 40 to 50 nmol/mg protein of thiobarbituric reactive substances (TBARS), as a measure of the extent of oxidation. Lipofectamine reagent was purchased from Gibco BRL (Gibco-BRL) and Trizol reagent was purchased from Ambion.

Tissue Culture

RAW 264.7 cells, a murine macrophage cell line (ATCC, Manassas, Va), were grown in DMEM (Gibco BRL) containing 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a 95% O₂/5% CO₂ humidified environment. Mouse peritoneal macrophages (10⁶ cells per animal on average) were obtained through peritoneal lavage with ice-cold PBS from C57BL/6J mice 4 days after intraperitoneal injection of thioglycolate (72 μg/mouse). The C57BL/6J mice were purchased from the Jackson Laboratories (Bar Harbor, Maine) and used according to a protocol approved by the University of Washington Institutional Animal Care and Use Committee. Cells were purified by adhesion to tissue culture dishes for 24 hours after harvesting prior to the experiments.

Real-Time Quantitative PCR

RAW 264.7 cells were treated with 30 μg/mL natLDL or oxLDL in DMEM plus 10% FBS, and total RNA was harvested from the cells at various time points using Trizol reagent, according to the manufacturer’s protocol. Gclc and Gclm mRNA message levels were obtained by quantitative real-time PCR of cDNA generated from 2 μg total RNA for each sample. Quantitative real-time PCR used a fluorogenic 5′ nuclease assay with primers and probes directed against the 3′-most intron/exon boundary of both the Gclc and Gclm genes. The sequences of the primers and probes are as follows (all sequences 5′ to 3′): Gclc sense primer, ATGTGGACACCCGATGCAGTATT; Gclc antisense primer, TGTCTTGCTTGTAGTCAGGATGGTTT; Gclc probe, CATTAGTTCTCCAGATGCTCTCTTCTTAA; Gclm sense primer, GCCACCAGATTTGACTGCCTTT; Gclm antisense primer, CAGGGATGCTTTCTTGAAGAGCTT; and Gclm probe, TGACTCACAATGACCCGAAAGAACTGCTCTC. Expression levels were derived from comparison with serial dilutions of a known reference standard and values are normalized to GAPDH.

Electrophoretic Mobility Shift Assays

Nuclear extracts were prepared by scraping RAW 264.7 cells in ice-cold cell lysis buffer (0.6% NP-40, 0.15 mol/L NaCl, 10 mmol/L Tris [pH 7.9], and 1 mmol/L EDTA with protease inhibitors) and incubating on ice for 5 minutes to lyse the cells. The nuclei were pelleted by centrifugation in a microfuge at 4°C at 1250g for 5 minutes before lysis on ice for 20 minutes in nuclear extraction buffer (10 mmol/L HEPES [pH 7.9], 0.1 mmol/L EGTA, 0.1 mmol/L EDTA, 1.5 mmol/L MgCl₂, 420 mmol/L NaCl, and 25% glycerol with protease inhibitors). After pelleting nuclear debris (1250g, 5 minutes, 4°C), nuclear extracts were aliquoted and stored at −80°C until needed. Protein concentrations were determined using the Bradford assay. Gel shift assays were carried out using 5 to 10 μg of nuclear proteins and ³²P-labeled oligonucleotide in the Gel Shift Assay System (Promega) according to the manufacturer’s protocol. Double-stranded oligonucleotides were prepared by boiling complementary strands in annealing buffer (20 mmol/L Tris, 1 mmol/L EDTA, and 50 mmol/L NaCl) for 10 minutes and slowly cooling to room temperature. For competition and antibody supershift assays, unlabeled oligonucleotides or rabbit polyclonal antibodies against Nrf1, Nrf2, c-jun, and c-fos (Santa Cruz Biotechnology) were added subsequent to the radiolabeled oligonucleotides. Gels were analyzed by phosphorimaging (Cyclone, Packard Instruments). The following oligonucleotide sequences (5′→3′) were used (boldface indicates the core consensus ARE sequences; underlining indicates the sites of mutation):

• Gclc ARE3wt, CGGGCCGATGGTCACGTCGCC

• Gclc ARE3m1, CGGGCCGATGGTCCCGTCGCC

• Gclc ARE4wt, GCAGAGTGCTGAGTCACGGTGAG

• Gclc ARE4m1, GCAGAGTGCTGAGTCCCGGTGAG

• Gclm AREwt, CTGGAAGACAATGACTAAGCAGAAA

• Gclm AREm1, CTGGAAGACAAGGACTAAGCAGAAA

• Gclm AREm2, CTGGAAGACAATGACTAATTTGAAA

Reporter Constructs and Plasmids

The Gclc reporter constructs were prepared as follows: a 7-kb EcoRI/BamHI fragment containing exon 1 and ≈6.5-kb of Gclc promoter was initially subcloned into pBluescript (Stratagene). A 6.5-kb XhoI/NcoI fragment containing 6447 bp of Gclc promoter was then shuttled into pGL3 Basic to yield the full-length 6.5-kb Gclc luciferase reporter construct. Deletion constructs were generated by digesting the 6.5-kb Gclc reporter construct with SacI or KpnI followed by recircularization of the plasmid to yield reporter constructs containing 3869 bp (3.8-kb Gclc) and 1281 bp (1.3-kb Gclc) of Gclc promoter, respectively. Site-directed mutagenesis of the ARE sites in the Gclc promoter was carried out by Bio S&T (Bio S&T) and confirmed by DNA sequencing.

To prepare the Gclm luciferase reporter constructs, a 1292-bp NcoI fragment of the mouse Gclm promoter¹⁹ was inserted into the NcoI site of the luciferase reporter vector pGL3 (Promega). Reporter constructs containing mutations within the ARE (m1 and m2) were engineered by separately amplifying a −1685-bp F/MutR fragment and a MutF/GL2R fragment, overlapping fragments that contained the mutation of interest within the region of overlap. The full-length fragment was amplified by addition of −1685-bp F and GL2R primers to the reaction, and the product was blunt-end ligated into a shuttle vector. The 1292-bp NcoI/NcoI fragment containing the mutation was then excised and ligated into the NcoI site of the pGL3 vector. All amplifications were performed using Vent polymerase (New England Biolabs). The sequences of the mutagenesis primers were (5′→3′):

• ARE m1F, GAGGTTTCTGCTTAGTCCTTGTCTTCCAGGAAA- CAGCTCC

• ARE m1R, GGAGCTGTTTCCTGGAAGACAAGGACTAAGCAGA- AACCTC

• ARE m2F, TTTCTCGGGTGAGGTTTCAAATTAGTCATTCTC- TTCCAGG

• ARE m2R, CCTGGAAGACAATCAATGACTAATTTGAAACC- TCACCCGAGAAA

Reporter Assays

RAW 264.7 cells were seeded into 35-mm plates 24 hours before transfection at 3×10⁵ cells/well. Cells were transfected using Lipofectamine reagent according to the manufacturer’s protocol. Briefly, cells were transfected with 1.5 μg DNA/well for 5 hours. All samples were cotransfected with a CMV-Renilla luciferase construct (Promega) to compensate for variances in transfection efficiencies. Twenty-four hours after beginning transfection the medium was changed to complete medium or medium containing 30 μg/mL natLDL or oxLDL. Cells were treated for 24 hours at which time they were harvested to obtain cell extracts for luciferase assays. Cells were harvested using passive lysis buffer (Promega) according to the manufacturer’s protocol. Luciferase assays were performed using the Dual Luciferase Reporter System (Promega) according to the manufacturer’s protocol and samples were measured in duplicate on a Lumat model 9507 luminometer (EG&G Berthold).

Statistical Analysis

All data are expressed as mean±SEM. Significant differences between means were determined by using the student’s two-tailed t test. All experiments were performed at least three times and representative results are shown.

OxLDL Increases Gclc and Gclm mRNA Levels

To determine whether oxLDL induces an increase in Gcl subunit message levels, mouse RAW 264.7 cells were treated with 30 μg/mL natLDL or oxLDL in the tissue culture medium for up to 24 hours (Figure 1). We did not observe any toxicity using this concentration of LDL. Cellular RNA was harvested for real-time quantitative PCR determination of Gclc and Gclm mRNA levels. Note that RNA was collected from untreated control cells at each time point as Gclc and Gclm subunit message levels fluctuate with changing cellular confluence. We observed induction of both Gclc and Gclm steady-state mRNA levels in response to treatment with oxLDL with induction occurring as early as 3 hours and, for the modifier subunit, persisting for 24 hours (Figure 1). Peak induction for both subunit mRNAs occurred after 6 hours of treatment. In contrast to oxLDL, treatment with natLDL does not appreciably alter message levels for either subunit. Induction of Gclm message after oxLDL treatment is consistently more robust than that observed for Gclc, with a peak 16-fold induction of Gclm message versus 8-fold induction for Gclc message relative to untreated controls.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Time course of Gclc and Gclm mRNA expression in RAW 264.7 cells after treatment with oxLDL or natLDL. Cells were treated for the indicated time points with 30 μg/mL of oxLDL or natLDL in tissue culture medium. Gclc (A) and Gclm (B) mRNA levels were determined by real-time RT-PCR. Values are normalized to GAPDH expression and are presented as mean±SEM.

OxLDL Induces Expression of Gclc and Gclm Reporter Constructs

An increase in the steady-state mRNA levels could be due to both increased de novo transcription and/or message stability. To determine whether the observed increases in Gclc and Gclm mRNA levels after oxLDL treatment are due to transcriptional upregulation, we transfected RAW 264.7 cells with luciferase reporter constructs containing 6.5 kb of the mouse Gclc or 1.3 kb of the mouse Gclm proximal promoter regions. Transiently transfected RAW 264.7 cells were treated with 30 μg/mL natLDL or oxLDL for 24 hours at which time cell extracts were harvested and assayed for luciferase activity. This time point demonstrated maximum luciferase activity as evaluated through time course experiments. OxLDL increased luciferase reporter expression from both the Gclc (Figure 2A) and Gclm (Figure 2B) promoters with a 1.8-fold increase in luciferase activity for the Gclc construct and a 2.5-fold increase observed for the Gclm construct. The activity of the internal standard, Renilla luciferase, was not affected by treatment with oxLDL (data not shown). These data provide evidence for increased transcription of the Gcl subunits as part of a coordinated response to upregulate GSH production upon exposure to oxLDL in macrophages.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Induction of Gclc and Gclm reporter constructs in RAW 264.7 cells after treatment with oxLDL or natLDL. RAW 264.7 cells were transiently transfected with 6.5-, 3.8-, or 1.3-kb Gclc/pGL3 constructs (A) or 1.3-kb Gclm/pGL3 construct (B) for 24 hours and cotransfected with a pRL-CMV construct to normalize for transfection efficiency. After transfection, cells were treated with 30 μg/mL natLDL or oxLDL for 24 hours. Luciferase activity in cell lysates was determined. Values are normalized to Renilla luciferase activity and are presented as mean±SEM. *P<0.05 vs untreated control.

As oxLDL triggers an oxidative stress response, we postulated that AREs present within the Gcl subunit promoters mediate the observed increase in transcription after oxLDL treatment. The GCLC promoter contains several putative AREs, of which two distal AREs have been characterized as functional for the human GCLC promoter in response to β-naphthoflavone.¹⁷ Transfection of reporter constructs containing truncated fragments of the mouse Gclc promoter into RAW 264.7 cells and subsequent treatment with oxLDL or natLDL reveal that promoter elements necessary for induction by oxLDL lie between −3.8 kb and −6.5 kb relative to the translation start codon (Figure 2A). This region of mouse Gclc promoter contains two putative AREs bearing semblance to the human ARE3 and ARE4, which mediate induction of human GCLC in response to oxidative stress; hence, we refer to these putative AREs as mouse ARE3 and ARE4. In the Gclm promoter, a single ARE located at −500 relative to the translation start codon has been described for both the mouse and human Gclm subunit and is contained within the 1.3-kb promoter fragment of the mouse Gclm reporter construct.

OxLDL Induces Binding of Nuclear Factors to Gclc and Gclm AREs

We conducted electrophoretic mobility shift assays (EMSAs) to determine whether oxLDL induces binding of nuclear factors to AREs located in the mouse Gclc and Gclm subunit promoters. RAW 264.7 cells were treated with oxLDL or natLDL for up to 12 hours during which time cells were harvested at various time points for the preparation of nuclear extracts. Nuclear proteins (5 to 10 μg) were incubated at room temperature for 20 minutes with a radiolabeled double-stranded oligonucleotide corresponding to mouse Gclc ARE3, Gclc ARE4, or Gclm ARE, and the resulting protein:DNA complexes were separated by polyacrylamide gel electrophoresis. As demonstrated in Figure 3, oxLDL induced binding of nuclear factors to both the Gclc ARE3 and ARE4 (Figure 3A) as well as the Gclm ARE (Figure 3B). Addition of excess unlabeled oligonucleotide to the binding reaction successfully competed away nuclear factors, indicative of binding specificity. The time course of induction is similar for all AREs examined, with increased nuclear factor binding occurring within 1 hour after treatment and returning to basal levels by 12 hours. However, for the two Gclc AREs, we observe different patterns of nuclear factor binding, particularly in the magnitude of the changes. The induction is much greater at ARE4 compared with ARE3, whereas the basal level of nuclear factor binding is greater at ARE3. Interestingly, natLDL also strongly induces binding of nuclear factors to these AREs, but the observed induction lags behind the induction after oxLDL treatment. This is perhaps due to oxidation of natLDL by the macrophages in the tissue culture dish. Gclm ARE also exhibits constitutive binding of nuclear factors, but induction is clear by 1 hour of treatment with oxLDL. In contrast to the catalytic subunit AREs, natLDL only slightly induced binding of nuclear factors to the modifier subunit ARE and, similar to the Gclc AREs, the binding is delayed (Figure 3B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Induction by oxLDL of nuclear protein-DNA binding to consensus oligonucleotides of the AREs in the Gclc and Gclm promoters. RAW 264.7 cells were treated with 30 μg/mL natLDL or oxLDL for the indicated time period and nuclear proteins were harvested. Protein complexes for the ARE3 and ARE4 of the Gclc promoter (A) as well as the ARE of the Gclm promoter (B) were analyzed by EMSA. Competition studies (lane 3 in all shift assays shown) were performed using 100-fold excess of unlabeled oligonucleotide. Binding activity to the Gclc promoter AREs in the no-treatment controls (data not shown) was similar to the binding activity observed at 12 hours after treatment with oxLDL (panel A, lane 8 for both ARE3 and ARE4).

We also conducted gel shift experiments using oligonucleotides for each of three AREs, which contained single base pair mutations within the ARE consensus sequence, and observed that mutations in the AREs strongly inhibited binding of nuclear factors. As a representative example, the reduced binding of nuclear proteins to the mutated ARE4 is presented in Figure 4, lane 7. Furthermore, addition of anti-Nrf1, anti-Nrf2, or anti-c-jun antibodies to the binding reaction resulted in displacement of the protein:DNA binding complex indicating involvement of the Nrf1, Nrf2, and c-jun transcription factors at this site. In contrast, addition of an anti-c-fos antibody had only minor effects. These transcription factors are known to bind to the AREs and influence transcription of certain ARE-containing genes. The displacement by Nrf1, Nrf2, and c-jun could be observed for ARE4 as demonstrated in Figure 4, lanes 3 through 6, and similar results were also observed for ARE3 and Gclm ARE (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4. Identification of transcription factors that bind to the Gclc promoter ARE4 and inhibition of nuclear protein-DNA binding using mutated oligonucleotides of the ARE4. RAW 264.7 cells were treated with 30 μg/mL oxLDL for 2 hours and nuclear proteins were harvested. Protein complexes for the ARE4 of the Gclc promoter were analyzed by EMSA. Lane 2, Cells were treated for indicated time; lanes 3 through 6, addition of indicated antibody; lane 7, radiolabeled oligonucleotides for the mutant ARE4; and lane 8, addition of 100-fold excess unlabeled ARE4.

Mutations in Gclc and Gclm Promoter AREs Attenuate Induction by OxLDL

We transfected RAW 264.7 cells with reporter constructs containing targeted mutations within the AREs of the Gclc and Gclm promoters to assess the functional significance of the AREs in induction of these genes in response to oxLDL. Cells were transiently transfected with a luciferase reporter construct containing 6.5 kb of the Gclc promoter with mutations in either ARE3 or ARE4 or both, and luciferase activities were measured in cell extracts obtained after 24 hours of treatment with oxLDL. The introduced mutation targets the core TGAC sequence of these AREs and contains a T>G single base pair transversion. This single base pair mutation in ARE3 diminished constitutive expression of the luciferase reporter but had no effect on induction by oxLDL (Figure 5A). In contrast, the same mutation in ARE4 affected both constitutive and oxLDL-induced expression of the luciferase reporter. Mutation of both AREs resulted in a luciferase expression pattern similar to that observed for the ARE4 mutation alone, suggesting that both ARE3 and ARE4 coordinate to influence constitutive expression, whereas ARE4 predominates in the induction response to oxLDL.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5. Functional analysis of the AREs in the Gclc and Gclm promoters in RAW 264.7 cells treated with oxLDL. RAW 264.7 cells were transfected with wild-type or mutated Gclc/pGL3 constructs (A) and wild-type or mutated 1.3-kb Gclm/pGL3 construct (B). Mutations of the reporter constructs were generated as described in Materials and Methods. After transfection for 24 hours, cells were treated with 30 μg/mL oxLDL for 24 hours. Luciferase activity in cell lysates was determined. Values are normalized to Renilla luciferase activity and are presented as mean±SEM. *P<0.05 vs wild-type treated with oxLDL; †P<0.05 vs untreated wild-type; and ‡P<0.05 vs control of the same construct.

For the Gclm reporter construct, we separately introduced two mutations into the ARE of the luciferase reporter construct containing 1.3 kb of the mouse Gclm promoter. The mut1 mutation is the same mutation introduced into the AREs of the Gclc promoter as described above, and the second mutation (mut2) targets the GCA box just downstream and contains a GCA>TTT triple base pair mutation. These mutated ARE constructs were transiently transfected into RAW 264.7 cells and their ability to drive oxLDL-inducible expression was compared with the wild-type ARE luciferase construct. Figure 5B reveals that both mutations in the Gclm ARE inhibit induction of luciferase expression on oxLDL treatment, with no effect on basal luciferase expression. These data, coupled with the EMSA results, support a functional role for the AREs in the observed induction of Gclc and Gclm expression by oxLDL.

Induction of Gcl mRNA and Binding of Nuclear Proteins to the AREs Is Also Observed in Mouse Peritoneal Macrophages

The experiments described above were all conducted in RAW 264.7 cells, an immortalized macrophage cell line. To test whether oxLDL also induces expression of the Gcl subunits and binding of nuclear proteins to the AREs in primary cells, we conducted real-time quantitative PCR and gel shift experiments with mouse peritoneal macrophages. As indicated in Figure 6A, oxLDL but not natLDL induced expression of Gclc and Gclm after 6 hours of treatment. Furthermore, oxLDL induced the binding of nuclear proteins to all AREs. As already shown for the RAW 264.7 cells, the transcription factors Nrf1, Nrf2, and c-jun were identified in the primary macrophages. As a representative example, ARE4 is shown in Figure 6B. However, functional analyses of the AREs could not be determined, because peritoneal macrophages could only be transfected with the constructs used previously for the RAW 264.7 cells at very low transfection efficiency (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 6. Gclm mRNA expression and protein-DNA binding to the Gclm promoter ARE4 in mouse peritoneal macrophages treated with oxLDL. A, Mouse peritoneal macrophages were treated with 30 μg/mL oxLDL or natLDL for 6 hours. Gclc and Gclm mRNA levels were determined by real-time RT-PCR. Values are normalized to GAPDH expression and presented as mean±SEM. B, Mouse peritoneal macrophages were treated with 30 μg/mL oxLDL for 2 hours and nuclear proteins were harvested. Protein complexes for the ARE4 of the Gclc promoter were analyzed by EMSA. Lane 2, Cells treated for indicated time; lanes 3 through 6, addition of indicated antibody; lane 7, radiolabeled oligonucleotides for the mutant ARE4; and lane 8, addition of 100-fold excess of unlabeled ARE4.