Nitric Oxide-induced Apoptosis in RAW 264.7 Macrophages Is Mediated by Endoplasmic Reticulum Stress Pathway Involving ATF6 and CHOP*

Excess nitric oxide (NO) induces apoptosis in some cell types including macrophages; however, the cascade of NO-mediated apoptosis is not fully understood. We investigated the initial steps of NO-mediated apoptosis in mouse macrophage-like RAW 264.7 cells. When cells were treated with bacterial lipopolysaccharide (LPS) plus interferon-γ (IFN-γ), NO-mediated apoptosis occurred. Under these conditions, p53 accumulation was not observed, indicating that DNA damage is not the main trigger of NO-mediated apoptosis. On the other hand, mRNA and protein for CHOP, a transcription factor known to be induced by endoplasmic reticulum (ER) stress, were induced. The CHOP induction by LPS/IFN-γ treatment preceded cytochrome c release from mitochondria. In addition, p90ATF6, an ER membrane-bound transcription factor involved in ER stress response, was cleaved to its active soluble form p50ATF6, which was transported to nucleus and bound to the ER stress response element of the CHOP gene. In the luciferase reporter assay, both the CHOP-binding element of the Rous sarcoma virus long terminal repeat and ER stress response element of theCHOP gene were activated by LPS/IFN-γ treatment. When RAW 264.7 cells or COS-7 cells were transfected with expression plasmids for CHOP, p90ATF6, or p50ATF6, cell death was observed. In addition, apoptosis induced by p50ATF6 was prevented by a CHOP dominant negative form as well as by an ATF6 dominant negative form, and LPS/IFN-γ-induced apoptosis was prevented by the CHOP dominant negative form. Peritoneal macrophages from CHOP knockout mice showed resistance to NO-induced apoptosis. These results indicate that the ER stress pathway involving ATF6 and CHOP plays a key role in NO-mediated apoptosis in macrophages.

Nitric oxide (NO) is a multifunctional biomolecule involved in a variety of physiological and pathological processes (1). In pathological conditions, NO functions as a bactericidal or tumoricidal agent. However, excess NO production has been implicated in diseases such as septic shock, autoimmune disease, cerebral infarction, and diabetes mellitus, in which NO-mediated apoptosis is often observed (2,3). NO has several cytotoxic effects, including reactions with proteins and nucleic acids, and causes apoptosis. NO-induced apoptosis is generally considered to be mediated by DNA damage or mitochondrial damage (4); however, the cascade of the cell death has not been fully clarified.
CHOP, also known as GADD153, is a member of the C/EBP family that heterodimerizes with other members of the C/EBP transcription factor family. This factor is induced in response to cellular stresses, especially endoplasmic reticulum (ER) 1 stress (5)(6)(7)(8)(9). CHOP is involved in the process of apoptosis associated with ER stress, although the mechanism is still unclear (6,10,11). Recently, ATF6 (12) was shown to be involved in the induction of CHOP in ER stress (13). ATF6 exists constitutively as a transmembrane protein p90ATF6 in the ER under nonstressed conditions (14,15). ER stress induces proteolysis of p90ATF6 and releases a soluble transcription factor p50ATF6, which is transported into the nucleus, binds to the ER stress responsive element (ERSE) of the CHOP gene, and activates its transcription (13, 16 -19). Here we report that the ER stress pathway involving CHOP is also important in bacterial lipopolysaccharide (LPS) plus interferon-␥ (IFN-␥)-induced and NOmediated apoptosis in RAW 264.7 macrophages and that this requires ATF6 activation.
Materials-A polyclonal antibody against mouse CHOP and a monoclonal antibody against p53 were obtained from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). A polyclonal antibody against mouse Bip was obtained from StressGen Biotechnologies Corp. (Victoria, Canada). Monoclonal antibodies against human/mouse cytochrome c, human hsp60, and rabbit GAPDH were obtained from R&D Systems Inc. (Minneapolis, MN), StressGen Biotechnologies Corp., and CHEMICON International Inc. (Temecula, CA), respectively. A polyclonal antibody against human ATF6 was reported (14). CHOP knockout mice were obtained from Shizuo Akira (Osaka University).
Cell Culture and Transfection-Mouse macrophage-like RAW 264.7 cells were grown in Eagle's minimal essential medium supplemented with 10% fetal calf serum. COS-7 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. Mouse peritoneal macrophages were prepared and grown in RPMI 1640 medium supplemented with 10% fetal calf serum as described (21). CHOP knockout mice were described (22). Transfection of RAW 264.7 cells and COS-7 cells with plasmids was carried out using TransIT-LT1 polyamine (PanVera Corp., Madison, WI) according to the protocol provided by the manufacturer. In each experiment, the same total amounts of plasmids were transfected by adding insert-less expression plasmids.
Detection of Apoptosis-RAW 264.7 cells or COS-7 cells were treated as described above. To analyze morphological changes of nuclei, the cells were fixed, stained with Hoechst dye 33258 (8 g/ml) for 5 min, and washed with phosphate-buffered saline. The stained cells were observed under a fluorescence microscope. To analyze mitochondrial depolarization, peritoneal macrophages were stained with a mitochondrial membrane potential-dependent dye DePsipher (Trevigen, Gaithersburg, MD).
RNA Blot and Immunoblot Analyses-RNA blot analysis was performed as reported (23). Hybridization was performed using digoxigenin-labeled antisense RNA as a probe for mouse CHOP or rat GAPDH. The antisense RNA was synthesized using as a template pT7Blue-mCHOP or pcDNAII-GAPDH digested with an appropriate restriction enzyme. Chemiluminescence signals derived from the hybridized probe were detected using a chemiluminescence image analyzer Las-1000 Plus (Fuji Photo Film Co, Tokyo, Japan). Immunoblot analysis was performed as described (23). Nuclear extracts from RAW 264.7 cells were prepared as described (24). Immunodetection was performed using an ECL kit (Amersham Biosciences, Inc.) according to the protocol provided by the manufacturer. Filters were analyzed using Las-1000 Plus.
Gel Shift Assay-Nuclear extracts of RAW 264.7 cells were prepared as described above. The sequence of double-stranded oligonucleotide used for the ERSE of the CHOP gene is as follows: 5Ј-CCTACCAATC-AGAAAGTGGCACGC-3Ј, 3Ј-GGATGGTTAGTCTTTCACCGTGCG-5Ј. This probe was 3Ј end-labeled with digoxigenin-11-ddUTP and terminal transferase using a DIG gel shift kit (Roche Molecular Biochemicals). The binding reaction was carried out for 30 min on ice as described (25). Non-labeled oligonucleotides (1 pmol) of the ERSE site or the p53 consensus site were used as competitors. The sequence of doublestranded oligonucleotide used for the p53 consensus site (Santa Cruz Biotechnology, Inc.) is as follows: 5Ј-TACAGAACATGTCTAAGCATG-CTGGGGACT-3Ј, 3Ј-ATGTCTTGTACAGATTCGTACGACCCCTGA-5Ј. Electrophoresis was done as described (25). After electrophoresis, the oligonucleotide probe was transferred to nylon membranes. Chemiluminescent signals derived from the probe were detected using a DIG luminescence detection kit (Roche Molecular Biochemicals) and Las-1000 Plus.
Luciferase Assay-Transfection was carried out with a total of 15 g of DNA mixture containing 10 g of a luciferase reporter plasmid and 5 g of an internal standard plasmid pAc-lacZ (26) bearing the E. coli ␤-galactosidase gene. Cells were cultured with or without LPS plus IFN-␥ or thapsigargin for 10 h prior to harvesting. 34 h after transfection, luciferase activity of cell extracts was measured using Luciferase Assay Systems (Promega Corp.) and then normalized for ␤-galactosidase activity.

p53 Is Not Induced in NO-mediated Apoptosis in RAW 264.7
Cells-When mouse macrophage-like RAW 264.7 cells were treated with LPS/IFN-␥, a NO-donor SNAP, an ER stressinducing reagent thapsigargin, or a DNA-damaging reagent camptothecin for 12 h, morphological changes characteristic of apoptosis were observed (Fig. 1A). Round-shaped cells and apoptotic bodies were observed in phase-contrast images, and chromatin condensation and nuclear fragmentation were seen in Hoechst dye 33258 staining. We showed previously that apoptosis induced by LPS/IFN-␥ treatment is mediated by NO (23). However, the pathway of NO-mediated apoptosis is not fully understood. We asked whether p53 is involved in NOmediated apoptosis (Fig. 1B). In general, p53 increases in response to DNA damage. In fact, when cells were treated with camptothecin, p53 protein increased. In contrast, treatment with LPS/IFN-␥, SNAP, or thapsigargin did not increase p53.
CHOP Is Induced in NO-mediated Apoptosis in RAW 264.7 Cells-CHOP is a transcription factor that is involved in ER stress-induced apoptosis. We found that CHOP mRNA, which was undetectable in untreated RAW 264.7 cells, was induced when cells were treated with LPS/IFN-␥ or SNAP ( Fig. 2A). The mRNA was induced by thapsigargin but not by campto- thecin. Fig. 2B shows immunoblot analysis of nuclear extracts for CHOP. CHOP, which was barely detectable before LPS/ IFN-␥ treatment, was induced after treatment. The level of LPS/IFN-␥-induced CHOP was close to that obtained with thapsigargin. This induction was prevented by a NO scavenger carboxy-PTIO, indicating that the induction is mediated by NO. CHOP was induced slightly 2 h after LPS/IFN-␥ treatment, increased markedly at 4 h, and increased up to 10 h (Fig.  2, C and D). Bip/GRP78, an ER chaperone that is known to be induced by ER stress (7), was induced by LPS/IFN-␥ as well as by thapsigargin (Fig. 2E), indicating that LPS/IFN-␥ produces ER stress. Camptothecin did not induce Bip.
We have shown previously that cytochrome c release from mitochondria takes place in LPS/IFN-␥-induced apoptosis of RAW 264.7 cells (27). Fig. 2F shows the time course of cytochrome c release in RAW 264.7 cells treated with LPS/IFN-␥. Cells were fractionated with digitonin into a soluble fraction and a particulate fraction that includes mitochondria, and these fractions were subjected to immunoblot analysis for cytochrome c. Cytochrome c release was not seen up to 6 h after LPS/IFN-␥ treatment but was observed at 10 h. At this time, about 70% of cytochrome c was recovered in the soluble fraction. Under these conditions, a mitochondrial matrix protein, hsp60, was recovered exclusively in the particulate fraction, whereas a cytosolic protein, glyceraldehyde 3-phosphate dehydrogenase, was recovered almost exclusively in the soluble fraction. Therefore, we conclude that CHOP is induced prior to cytochrome c release in LPS/IFN-␥-induced apoptosis in RAW 264.7 cells.
ATF6 Is Activated in LPS/IFN-␥-treated RAW 264.7 Cells-Yoshida et al. (13) found that p50ATF6 binds to ERSE of the CHOP gene and transactivates the gene. Therefore, we asked whether ATF6 is activated in LPS/IFN-␥-treated RAW 264.7 cells. Fig. 3A shows the time course of ATF6 activation and CHOP induction. In control cells, p90ATF6, which is the ER transmembrane form of ATF6, was detected, but p50ATF6, which is the processed and active form, was not detected. p90ATF6 was decreased with time after LPS/IFN-␥ treatment, whereas p50ATF6 appeared and increased with time, indicating that p90ATF6 was processed to p50ATF6. CHOP was induced concomitantly with the processing of ATF6. Processed p50ATF6 appeared in the nuclei 2 h after LPS/IFN-␥ treatment and increased with time up to 10 h (Fig. 3, B and C). The level at 10 h was about half of that produced in response to thapsigargin. Binding activity to ERSE of the CHOP gene was also monitored (Fig. 3D). When cells were treated with thapsigargin for 6 h, two binding complexes (complexes I and II) were detected. Treatment with LPS/IFN-␥ also produced two complexes. An antibody specific to ATF6 diminished complex II, whereas control serum caused no change. Therefore, we conclude that p50ATF6 is involved in complex II. This agrees well with the results obtained with purified p50ATF6 and NF-Y, another transcription factor involved in the complex (13). In that report, it was shown that p50ATF6 and NF-Y were involved in complex II and only NF-Y was involved in complex I. Complex II was not detected in control nuclear extract, began to increase 2 h after LPS/IFN-␥ treatment, and increased up to 10 h. These results indicate that LPS/IFN-␥ treatment processed to activate ATF6 and that the active form of ATF6 (p50ATF6) binds to ERSE of the CHOP gene.
To confirm that LPS/IFN-␥ treatment induces CHOP through stimulation of ERSE, we made reporter constructs where ERSE of the CHOP gene or the CHOP binding site of Rous sarcoma virus long terminal repeat was inserted just upstream of the SV40 promoter that drives the luciferase gene. RAW 264.7 cells were transfected with the reporter constructs and treated with LPS/IFN-␥ or thapsigargin for 10 h, and then cell extracts were assayed for luciferase activity (Fig. 4). Luciferase activity derived from the SV40 promoter was little af- The results in panel C and in a parallel experiment were quantified and are shown by means Ϯ ranges (n ϭ 2) (D). The maximal value at 10 h is set at 100%. Cells were treated with LPS (150 g/ml) plus IFN-␥ (100 units/ml), TG (2 M), or CPT (30 nM) for 10 h, and cell extracts (20 g of protein) were subjected to immunoblot analysis for Bip (E). Immunoblots from two dishes for each condition are shown. Cells were treated with LPS (150 g/ml) plus IFN-␥ (100 units/ml) for indicated periods and fractionated into the soluble fraction (S) and the particulate fraction (P) as described under "Experimental Procedures" (F). Distribution of total protein in the soluble fraction and the particulate fraction was 65 and 35%, respectively. The fractions (10 g of protein) were subjected to immunoblot analysis for cytochrome c, hsp60, and GAPDH. fected by treatment with LPS/IFN-␥ or thapsigargin. When ERSE of the CHOP gene was inserted, luciferase activity was enhanced 2.2-fold by LPS/IFN-␥ treatment and 3.4-fold by thapsigargin treatment. When the CHOP binding site was inserted, luciferase activity was enhanced 3.2-fold by LPS/ IFN-␥ treatment and 5.9-fold by thapsigargin treatment. We therefore conclude that activated ATF6 enhances transcription of the CHOP gene through ERSE in vivo in LPS/IFN-␥-treated RAW 264.7 cells.
Overexpression of CHOP, p50ATF6, or p90ATF6 Induces Apoptosis in RAW 264.7 and COS-7 Cells-CHOP was reported to induce apoptosis in M1 myeloblastic leukemia cells (10) and mouse embryonic fibroblasts (6). We asked whether CHOP and ATF6 can induce apoptosis in RAW 264.7 cells. Cells were cotransfected with an EGFP expression plasmid and an expression plasmid for CHOP, p50ATF6, or p90ATF6. When cells were transfected with only the EGFP plasmid, many cells became fluorescent (Fig. 5A). Cotransfection with the CHOP plasmid markedly reduced EGFP-positive cells. When the p50ATF6 plasmid was cotransfected, the number of EGFP-positive cells was even more markedly reduced. In the case of the p90ATF6 plasmid, the number of EGFP-positive cells was moderately reduced. It was shown that overexpression of p90ATF6 leads to partial processing of p90ATF6, resulting in the production of small amounts of p50ATF6 (14) (see also Fig. 6E). These results indicate that ATF6 as well as CHOP induces the death of RAW 264.7 cells. Fig. 5B shows the effect of dominant negative forms of ATF6 and CHOP on ATF6-and CHOP-induced apoptosis in RAW 264.7 cells. Reduction of EGFP-positive cells by transfection with the p50ATF6 plasmid was markedly prevented by cotransfection with a plasmid for its dominant negative form, and more importantly, by cotransfection with a CHOP dominant negative plasmid. This indicates that p50ATF6-induced apoptosis in RAW 264.7 cells is mediated by CHOP. CHOPinduced apoptosis was prevented by cotransfection with the CHOP dominant negative plasmid.
We then asked whether NO-mediated apoptosis in RAW 264.7 cells can be prevented by expression of the CHOP dominant negative form (Fig. 5C). The number of EGFP-positive cells was decreased markedly by LPS/IFN-␥, and this decrease was effectively prevented by cotransfection with the CHOP dominant negative plasmid. The CHOP dominant negative plasmid alone had little effect on the number of EGFP-positive cells. These results indicate that LPS/IFN-␥-induced apoptosis in RAW 264.7 cells is mediated by CHOP. However, CHOP was barely detected in RAW 264.7 cells transfected with p50ATF6 or p90ATF6. Therefore, we used COS-7 cells in which higher levels of expression can be obtained.
When COS-7 cells were cotransfected with the EGFP plasmid and the plasmid for CHOP, p50ATF6, or p90ATF6, the FIG. 3. Activation of ATF6 in LPS/IFN-␥-treated RAW 264.7 cells. Cells were treated with LPS (150 g/ml) plus IFN-␥ (100 units/ ml) for indicated periods, and whole cell extracts (100 g of protein for ATF6 and CHOP, 10 g of protein for GAPDH) were subjected to immunoblot analysis for ATF6, CHOP, and GAPDH (A). Cells were treated with LPS (150 g/ml) plus IFN-␥ (100 units/ml) for indicated periods or TG (2 M) for 6 h, and nuclear extracts (30 g of protein) were subjected to immunoblot analysis for the activated form of ATF6, p50ATF6 (B). The results in panel B and in a parallel experiment were quantified and are shown by means Ϯ ranges (n ϭ 2) (C). The maximal value at 10 h is set at 100%. Gel shift analysis of a factor(s) binding to ERSE of the CHOP gene is shown (D). Cells were treated with LPS (150 g/ml) plus IFN-␥ (100 units/ml) for indicated periods or TG (2 M) for 6 h, and nuclear extracts were prepared. A digoxigenin-labeled probe for the ERSE site of the CHOP gene was incubated with nuclear extracts (5 g of protein), and a gel shift assay was performed as described under "Experimental Procedures." An antiserum against ATF6 (␣ATF6, 1 l) or a nonimmune serum (Control Ab, 1 l) was added to the binding mixture halfway through the reaction. As competitors, unlabeled oligonucleotides for the ERSE site (ERSE) or the p53 binding site (Control oligo) were added in the reaction mixture. The positions of complexes I and II are indicated on the right. number of EGFP-positive cells was decreased, and the results were similar to those for RAW 264.7 cells (Fig. 6A). Expression of EGFP was quantified by immunoblot analysis, and the results are shown in Fig. 6, B and C. CHOP protein, which was not detectable in nuclear extracts from control cells, was detected in nuclear extracts from cells transfected with the plasmid for p50ATF6 or p90ATF6 as well as for CHOP (Fig. 6D). In cells transfected with the p50ATF6 plasmid, p50ATF6 protein was detected in addition to endogenous p90ATF6 (Fig. 6E). When the p90ATF6 plasmid was transfected, processed p50ATF6 as well as unprocessed p90ATF6 was detected. Therefore, overexpression of p90ATF6 leads to a partial processing to active p50ATF6, which in turn induces CHOP and following apoptosis. Fig. 6F shows the time course of GFP expression cotransfected with the EGFP plasmid and the plasmid for CHOP or p50ATF6. At 12 h after transfection, GFP expression was detected in control cells, and GFP levels in CHOP-or p50ATF6-transfected cells were similar to that in control cells. In contrast, the GFP expression increased markedly in control cells at 24 h, whereas it increased much less markedly in CHOP-transfected cells and did not increase in p50ATF6-transfected cells. These results indicate that reduction of GFP-positive cells in CHOP-or ATF6-transfected cells is due to cell death.
We next asked whether cell death induced by ATF6 or CHOP is prevented by their dominant negative forms in COS-7 cells (Fig. 7, A-C). The results were similar to those for RAW 264.7  were subjected to immunoblot analysis for GFP and GAPDH (B). Immunoblots from two dishes for each condition are shown. The results in panel B were quantified and are shown by means ϩ ranges (n ϭ 2) (C). COS-7 cells were treated as in panel A (D and E). Cell extracts (50 g of protein) were subjected to immunoblot analysis for CHOP (D) and ATF6 (E). Cells were cotransfected with an EGFP expression plasmid and an expression plasmid for CHOP or p50ATF6 (F). After incubation for indicated times, cell extracts (20 g of protein) were subjected to immunoblot analysis for GFP and GAPDH. cells. The decrease in EGFP-positive cells by p50ATF6 was partially reversed by the dominant negative forms of ATF6 and CHOP, and the decrease by CHOP was almost completely reversed by its dominant negative form. Expression of p50ATF6 and its dominant negative form (Fig. 7D) and that of CHOP and its dominant negative form (Fig. 7E) were confirmed by immunoblot analysis.
We then asked whether expression of CHOP or ATF6 induces apoptosis in COS-7 cells (Fig. 8). When CHOP or p50ATF6 was coexpressed with EGFP, 68 or 79% of EGFPpositive cells showed apoptotic changes, respectively. When p90ATF6 was coexpressed, 48% of EGFP-positive cells were apoptotic, apparently due to partial processing to p50ATF6 (see Fig. 6E).
Peritoneal Macrophages from CHOP-deficient Mice Are Resistant to NO-mediated Apoptosis-We finally asked whether CHOP is involved in NO-mediated apoptosis in primary-cultured peritoneal macrophages by using CHOP knockout mice. When peritoneal macrophages from wild-type mice were treated with 1.5 mM SNAP, CHOP mRNA was induced with time (Fig. 9A). When wild-type macrophages were treated with 1.5 mM SNAP for 10 h, most cells (82%) lost mitochondrial membrane potential (Fig. 9, B and C). In contrast, more than half (55%) of peritoneal macrophages from CHOP knockout mice retained membrane potential after SNAP treatment. These results demonstrate that NO-induced apoptosis of peritoneal macrophages is mediated at least partly by CHOP. DISCUSSION Excess NO production induces apoptosis in various cell types. It is generally believed that NO-induced apoptosis is mediated by the DNA damage pathway involving accumulation of p53 (4). However, in this report we showed that p53 induction is not evident in NO-mediated apoptosis in RAW 264.7 macrophages. It was reported that there is a p53-independent pathway in addition to a p53-dependent one in NO-mediated apoptosis (28). From the results shown here, we conclude that the p53-independent pathway is the major one in NO-mediated apoptosis in RAW 264.7 cells. Recently, we showed that NO induces apoptosis in p53-deficient microglia (29).
We found that CHOP is induced in NO-mediated apoptosis in RAW 264.7 cells and peritoneal macrophages. NO is known to inhibit many enzymes and ion channels (3). We found that NO depletes ER Ca 2ϩ and causes ER stress in mouse ␤ cell-derived MIN6 cells (22). NO was reported to inhibit Ca 2ϩ -ATPase activity of SERCA2a by tyrosine nitration within the channel-like domain (30). We have therefore proposed that NO induces ER stress by disturbing ER Ca 2ϩ homeostasis in cells (22). CHOP induces apoptosis in some cell types (6). Embryonic fibroblasts derived from CHOP knockout mice exhibit significantly less apoptosis as compared with wild-type cells when challenged with ER stress-inducing reagents (6). CHOP was found to induce apoptosis in M1 myeloblastic leukemia cells in a p53-independent manner, and Bcl-2 delayed this process (10). In the present work, we showed that CHOP-deficient peritoneal macrophages are more resistant to NO-induced apoptosis than wild-type cells. Because CHOP functions as a transcription factor, there must be a target gene(s) whose transcription is activated by CHOP and whose product(s) works in the apoptosis signal cascade. Wang et al. (11) found candidate target genes of the CHOP protein using representational difference analysis. However, these genes are distinct from known factors involved in the ER stress response and apoptosis. Recently, McCullough et al. (31) reported that CHOP expression results in down-regulation of Bcl-2 expression, depletion of cellular glutathione, and exaggerated production of reactive oxygen species. The precise apoptosis cascade downstream of CHOP remains to be clarified.
ATF6 is a type 2 ER membrane protein with its NH 2 terminus in the cytosol (15). When ER stress such as unfolded protein accumulation in ER occurs, ATF6 (p90ATF6) is cleaved to release its cytosolic domain (p50ATF6), which is a transcription factor of the basic leucine zipper family (14). Recently, Ye et al. (16) reported that this proteolysis is mediated by the Site-1 protease (S1P) and Site-2 protease (S2P). However, it is still unknown how the ER stress signal triggers proteolysis of ATF6. At least four genes, GRP78, GRP94, CHOP, and calreticulin, have been shown to be activated by p50ATF6 (13,17,32). The cleavage of the transmembrane protein in response to cell signaling to liberate cytosolic fragments that enter the nucleus to control gene transcription is called regulated intramembrane proteolysis (Rip) (15). In the present work, we found that the Rip type proteolysis of ATF6 takes place in the process of NO-mediated apoptosis in RAW 264.7 cells.
An ERSE was identified in the CHOP gene. The consensus sequence of ERSE is CCAAT-N 9 -CCACG, and this cis-acting element is necessary and sufficient for the response to ER stress (32,33). The sequence CCACG provides specificity for the response to ER stress. The general transcription factor NF-Y binds to CCAAT and p50ATF6 binds to CCACG, thereby accounting for the specificity of the ER stress response (13). p50ATF6 was shown to bind directly to CCACG only when CCAAT exactly 9 bp upstream of CCACG is bound by NF-Y. We suggest that complex II in Fig. 3C consists of ATF6 and NF-Y, whereas complex I consists of NF-Y, as shown previously by Yoshida et al. (13). On the other hand, Fawcett et al. (34) reported that the CHOP gene can be activated by ATF4, another basic leucine zipper-type transcription factor, via its binding to the C/EBP-ATF site present in the CHOP promoter region and distinct from the ERSE. Recently, Harding et al. (35) revealed that translation of ATF4 is selectively increased during ER stress and that this translational induction is mediated by PERK, a type 1 transmembrane protein kinase in the ER, which senses ER stress and transmits signals by phosphorylating the ␣ subunit of eukaryotic initiation factor 2. Therefore, it is very likely that not only ATF6 but also ATF4 is activated in RAW 264.7 cells by LPS/IFN-␥ and that both factors are involved in NO-induced apoptosis. Islet cells and macrophages are highly active in protein secretion, and it is tempting to speculate that "secretory" cells active in protein secretion are more sensitive to NO-mediated apoptosis than "non-secretory" cells, which are less active in protein secretion. This remains to be tested. FIG. 9. Peritoneal macrophages from CHOP knockout mice are resistant to NO-induced apoptosis. Peritoneal macrophages from wild-type mice were stimulated with 1.5 mM SNAP for indicated periods on the top, and total RNA (6 g) were subjected to RNA blot analysis for CHOP and GAPDH (A). Control shows total RNA (2 g) from RAW 264.7 cells stimulated with 1.5 mM SNAP for 6 h. Peritoneal macrophages were prepared from wild-type (WT) and CHOP knockout mice (CHOP Ϫ/Ϫ) (B). Cells were treated with 1.5 mM SNAP for 10 h, stained with a mitochondrial membrane potential dye DePsipher, and observed by fluorescence microscopy. The red fluorescence represents intact potential, and the green fluorescence represents disrupted potential. Original magnifications: ϫ200. Bar, 50 m. Experiments were performed as in panel B, and the percentages of apoptotic cells are shown as means ϩ S.E. for three dishes (C). More than 100 cells in each dish were analyzed.