C/EBP homologous protein (CHOP)/growth arrest and DNA damage-inducible gene 153 is a C/EBP family transcription factor which is involved in endoplasmic reticulum (ER) stress-mediated apoptosis. To determine whether the ER stress-CHOP pathway is involved in the pathogenesis of the lung inflammation, mice were given LPS intratracheally. Treatment with LPS induced mRNAs for CHOP and BiP. The LPS-induced inflammation in lung, including the IL-1β activity in bronchoalveolar lavage fluid, was attenuated in the Chop knockout mice. Caspase-11, which is needed for the activation of procaspase-1 and pro-IL-1β, was induced by LPS treatment in the lung and primary cultured macrophages. The induction of caspase-11 by LPS was suppressed in Chop knockout mice. Caspase-11 was also induced by such ER stress inducers as thapsigargin or tunicamycin. These results show that CHOP plays a crucial role in the pathogenesis of inflammation through the induction of caspase-11.
The endoplasmic reticulum (ER)3 performs several important functions, including posttranslational modification, folding, and the assembly of newly synthesized secretory and cell membrane proteins, and its proper function is essential to cell survival (1). When the cells are subjected to ER stress due to the accumulation of unfolded proteins in the ER, the ER stress pathways are activated to protect the cells (2). These responses involve induction of ER chaperones including Ig H chain binding protein (BiP), translational attenuation, and the degradation of unfolded proteins by a system called ER-associated degradation. Recently, the ER stress pathways have been reported to also be activated by various stresses such as oxidative stress, hypoxia, ischemia, and poly Q proteins (3, 4, 5). Therefore, these pathways function as an intracellular stress sensor and protect cells from various stresses. However, when the ER functions are severely impaired, then apoptosis occurs (6). This apoptosis is mediated by C/EBP homologous protein (CHOP)/growth arrest and DNA damage-inducible gene 153 or other factors (7). CHOP is a member of the C/EBP transcription factor family, is induced by ER stress and thus causes apoptosis. However, the target gene(s) of CHOP are still not clear at present. We found that a proapoptotic Bcl-2 family molecule Bax is translocated from cytosol to mitochondria thereby transducing the apoptosis signal in ER stress- and CHOP-mediated apoptosis (8). We also found that ER stress- and CHOP-mediated apoptosis plays a key role in mouse hereditary diabetes mellitus caused by insulin gene mutation (9), NO-induced apoptosis in pancreatic β cells (10), microglias (11), and macrophages (12), neuronal cell death induced by ischemia-reperfusion (4), and nonsteroidal anti-inflammatory drug-induced apoptosis in gastric mucosal cells (13). We speculate that CHOP-mediated apoptosis may thus be involved in many other diseases.
Severe pneumonia remains one of the major clinical problems, and the mortality rate is high. Various cytokines, secreted from macrophages and other inflammatory cells, are involved in the pathogenesis of pneumonia and many other inflammatory diseases (14). IL-1β is secreted from activated macrophages at an early stage of the inflammatory response and it also activates other inflammatory cells including the macrophage themselves (15). In activated macrophages, IL-1β is synthesized as pro-IL-1β and it is then processed into mature IL-1β by caspase-1 (16). Actually, the activation of IL-1β was suppressed in LPS-treated caspase-1 knockout mice (17), and these mice were resistant to the lethal effects of LPS. Caspase-1 is also known as IL-1β-converting enzyme (ICE). However, caspase-1 (mRNA) is also expressed in non-IL-1β-secreting cells. Therefore, it has been speculated that the function of caspase-1 is not limited to ICE. Miura et al. (18) showed that an overexpression of caspase-1 gene in a cultured cell line caused the cells to undergo apoptosis. Caspase-1 is also synthesized as procaspase-1, and it exists in cytosol as a proform in nonactivated cells. In activated macrophages, procaspase-11 is induced and activated by autoproteolysis. The active form of caspase-11 activates caspase-1 by proteolysis. Wang et al. (19) showed that caspase-11 knockout mice are resistant to endotoxin shock induced by LPS. They also showed that the expression of caspase-11 is essential for the activation of caspase-1. Both caspase-1 and caspase-11 belong to the inflammatory caspase subfamily. Caspase-11 is induced through the pathway involving NF-κB and STAT1 in LPS-treated macrophages (20).
In this study, we report that the CHOP-involved pathway plays a crucial role in both the induction of caspase-11 and in the pathogenesis of LPS-induced lung injury.
Materials and Methods
Polyclonal Abs against CHOP, BiP, caspase-1, and mAb against heat shock cognate (Hsc) 70 were obtained from Santa Cruz Biotechnology. mAb against caspase-11 was obtained from Sigma-Aldrich.
Generation of animal models
All procedures were approved by the Animal Care and Use Committee of Kumamoto University. Mice lacking the Chop gene (C57BL/6 background) were generated as previously described (10). The tracheae of specific pathogen-free male C57BL/6 and Chop knockout mice weighing 19–25 g (6 wk of age) were cut-down and given Escherichia coli LPS (serotype 0127:B8; Sigma-Aldrich) intratracheally at 10 mg/kg body weight (total volume was set to 30 μl/body), or recombinant mouse IL-1β (Genzyme-Techne) intratracheally at 5 mg/kg body weight (total volume was set to 30 μl/body) and then killed at the indicated times after being anesthetized with ether.
Bronchoalveolar lavage (BAL), the measurement of cytokines in BAL fluid (BALF), and the staining of alveolar cells
Mice lungs were lavaged intratracheally with PBS (pH 7.4, 0.7 ml) twice, and then BALF was obtained (∼1.0 ml). To obtain samples for the measurement of cytokines, BALF samples were centrifuged at 820 × g for 5 min at 4°C, and then the supernatants were used for ELISA to measure the concentration of TNF-α and IL-1β. ELISA for TNF-α and IL-1β were performed using mice TNF-α and IL-1β ELISA kit (BioSource International) according to the protocol provided by the manufacturer. To obtain alveolar cells, BALF were applied to cytospin slides and then underwent centrifugation at 820 × g for 5 min at 4°C. The staining of alveolar cells was performed using Diff-Quik (Sysmex) according to the protocol provided by the manufacturer and then were observed by microscopy. The percentages of each kind of alveolar cell were calculated morphologically. More than 500 cells in each condition were analyzed.
Isolation and treatment of peritoneal macrophages
Mice peritoneal macrophages were prepared and grown in RPMI 1640 medium supplemented with 10% FCS as described (21). After being cultured for 3 days, nonadherent cells were removed, and then adherent cells were treated with LPS (150 μg/ml) or thapsigargin (4 μM) or tunicamycin (1 μg/ml) for the indicated periods.
The lungs were fixed by perfusing with 4% paraformaldehyde in PBS (pH 7.4), embedded in paraffin, and sectioned 5-μm thick. The procedure for H&E staining has been previously described (22).
The mice were killed with deep anesthetization at the indicated times after treatment. The lungs were removed and frozen immediately on dry ice then stored at −80°C. Total RNA from lung and cultured cells were isolated using the acid guanidium thiocyanate phenol-chloroform extraction procedure as described (23M15131); sense primer, 5′-TCCAGCCACACTTCAGCCTAG-3′ and antisense primer, 5′-ACTGTGTGGGTGGGATGTAGCT-3′ for MIP-2 (GenBank accession number X53798); sense primer, 5′-ACACGTCTTGCCCTCATTATCTGC-3′ and antisense primer, 5′-CCACTCCTTGTTTCTCTCCACG-3′ for caspase-1 (GenBank accession number NM_009807); sense primer, 5′-GCGTTGGGTTTTTGTAGATGCC-3′ and antisense primer, 5′-ATGTGCTGTCTGATGTCTGGTG-3′ for caspase-11 (GenBank accession number Y13089); sense primer, 5′-GAAAGGATGGTTAATGATGCTGAG-3′ and antisense primer, 5′-GTCTTCAATGTCCGCATCCTG-3′ for BiP (GenBank accession number AJ002387); sense primer, 5′-CATACACCACCACACCTGAAAG-3′ and antisense primer, 5′-CCGTTTCCTAGTTCTTCCTTGC-3′ for CHOP (GenBank accession number X67083); sense primer, 5′-TCTGATCGGGCGTAGGTTTG-3′ and antisense primer, 5′-CATCTCGAAGGGCCTTCTCTAC-3′ for Hsc70 (GenBank accession number M19141); sense primer, 5′-TGATGGTGGGAATGGGTCAGAA-3′ and antisense primer, 5′-CCAAGAAGGAAGGCTGGAAAAG-3′ for β-actin (GenBank accession number M12481). The primer sets for IL-1β, MIP-2, caspase-1, caspase-11, BiP, CHOP, Hsc70, and β-actin are expected to give PCR products with a size of 384, 723, 372, 439, 231, 357, 754, and 674 bp, respectively. PCR consisted of an initial denaturation cycle at 94°C for 2 min, followed by the 27 cycles at 94°C for 15 s, annealing at 55°C for 30 s, and elongation at 68°C for 1 min. An additional cycle at 72°C for 7 min completed the amplification process. Amplified PCR products were separated by 1% agarose gel electrophoresis and visualized with the use of ethidium bromide staining. The primers used for detection of X-box binding protein (XBP) 1 mRNA are as follows: sense primer, 5′-AAACAGAGTAGCAGCGCAGACTGC-3′ and antisense primer, 5′-GGATCTCTAAAACTAGAGGCTTGGTG-3′. The primer sets are expected to give PCR products with a size of 600 bp for unspliced form and 574 bp for spliced (active) form XBP1, respectively. PCR consisted of an initial denaturation cycle at 94°C for 2 min, followed by 25 cycles consisting of denaturation at 94°C for 15 s, annealing at 55°C for 30 s, and elongation at 68°C for 1 min. An additional cycle at 72°C for 7 min completed the amplification process. The amplified PCR products were separated by 3% agarose gel electrophoresis and then were visualized with ethidium bromide staining.
The mouse lung was homogenized in lysis buffer (300 mM NaCl, 50 mM Tris-HCl, 1% Triton X-100 (pH 7.5)). After centrifugation, the supernatants served as tissue extracts for an immunoblot analysis. Mouse peritoneal macrophages were homogenized in 20 mM HEPES-KOH (pH 7.5) containing 1% Triton X-100, 20% glycerol, and 1 mM DTT. After centrifugation, the supernatants were used as cell extracts for an immunoblot analysis. In the analysis of CHOP, either lung or peritoneal macrophages were homogenized in 25 mM Tris-HCl (pH 7.4) containing 300 mM NaCl, 1% Triton X-100. After centrifugation, the supernatants served as tissue or cell extracts for an immunoblot analysis. The extracts were then subjected to SDS-PAGE, and then the proteins were electrotransferred to nitrocellulose membranes. Immunodetection was performed using an ECL kit (Amersham Biosciences) according to the protocol provided by the manufacturer.
Measurement of caspase-1 activity
Peritoneal macrophages from wild-type or Chop knockout mice were treated with LPS (150 μg/ml) for 24 h. Caspase-1 activity was measured using the Caspase-1/ICE Colorimetric Assay kit (BioVision) according to the protocol provided by the manufacturer.
The lungs were fixed by perfusing them with 4% paraformaldehyde in PBS (pH 7.4), and the excised tissues were immersed in a fixative solution consisting of 4% paraformaldehyde and PBS (pH 7.4) for 3 h at 4°C, washed in PBS for 15 min, dehydrated through a graded series of ethanol and xylene, and embedded in the same paraffin block. Sections (5-μm thick) were cut, air-dried, and deparaffinized. Next, the sections were pretreated with 5 mM periodic acid for 10 min at room temperature to inhibit any endogenous peroxidase activity. The specimens were incubated for 1 h with 100-fold diluted polyclonal Ab against CHOP and/or 100-fold diluted mAb against mouse caspase-11, and then were washed three times with PBS for 5 min. For the single immunostaining of caspase-11, the sections were incubated for 1 h with 500-fold-diluted sheep-anti-mouse Ig (F(ab′)2) conjugated with peroxidase (Amersham Biosciences) as a second Ab, and peroxidase activity was visualized by incubation with a 3,3′-diaminobenzidine solution. For the double staining of CHOP and caspase-11, sections were incubated for 1 h with Alexa Fluor 488-labeled anti-rat IgG (Molecular Probes) and Cy3-labeled anti-rabbit IgG as second Abs.
Plasmids and transfection
pcDNA3.1-mCHOP, a mammalian expression plasmid for mouse full-length CHOP, was described (12
Quantitative results were expressed as the mean ± SD and analyzed using the Student t test. The significance in difference was assigned at a level of less than a 5% probability (p < 0.05).
LPS-induced lung inflammation was attenuated in Chop knockout mice
LPS was given to wild-type and Chop knockout mice intratracheally. To assess the LPS-induced lung inflammation, a histological analysis with H&E staining was performed (Fig. 1⇓A). In the lung of wild-type mice treated with LPS, severe inflammatory changes, such as thickening of the alveolar septum and the infiltration of many inflammatory cells, were diffusely observed. Such inflammatory changes were obviously attenuated in the lungs of the Chop knockout mice. The infiltration of inflammatory cells in the alveolar space is another specific marker for inflammation in the lung. Therefore, we analyzed the cell types in BALF after LPS treatment (Fig. 1⇓, B–D). Most cells (∼90%) in the BALF of the PBS-treated wild-type and Chop knockout mice were macrophages (Fig. 1⇓B). After LPS treatment, the number of cells increased markedly in the wild-type mice (∼8 × 105/ml). However, the increase was repressed in the Chop knockout mice (∼4 × 105/ml). Next, we counted neutrophils (segmented and stab) and monocytes (including macrophage) after the treatment (Fig. 1⇓D). Segmented neutrophils increased dramatically in the wild-type mice, but only slightly in the Chop knockout mice.
TNF-α and IL-1β are major inflammatory cytokines, which stimulate the synthesis of other inflammatory cytokines, and trigger the start of the inflammatory cascade. Therefore, we measured the concentrations of TNF-α and IL-1β in BALF as inflammation markers (Fig. 1⇑E). TNF-α was barely detected in either the wild-type or Chop knockout mice before LPS treatment. The treatment markedly induced TNF-α in both the wild-type and Chop knockout mice. There was no difference in the induction level of TNF-α (∼1200 pg/ml) between the wild-type and Chop knockout mice. In contrast, the induction of IL-1β (active form), which was present at a low level both in nontreated wild-type and Chop knockout mice, was markedly suppressed in the Chop knockout mice. The induced level in LPS-treated wild-type mice was ∼150 pg/ml, whereas that in Chop knockout mice was ∼50 pg/ml. These results show that LPS-induced lung inflammation is strongly attenuated in the Chop knockout mice, probably due to the suppression of the IL-1β pathway.
Inflammation of the lung induced by IL-1β was not attenuated in Chop knockout mice
To investigate whether inflammatory response pathway downstream of IL-1β is affected in Chop knockout mice, rIL-1β was given to wild-type and Chop knockout mice intratracheally. To assess the IL-1β-induced lung inflammation, a histological analysis with H&E staining was performed (Fig. 2⇓A). Severe inflammatory changes, such as thickening of the alveolar septum and the infiltration of many inflammatory cells, were diffusely observed both in the lung of wild-type and Chop knockout mice treated with IL-1β. Next, we analyzed the cell types in BALF after IL-1β treatment (Fig. 2⇓, B and C). Most cells (∼90%) in the BALF of PBS-treated (0 h) wild-type and Chop knockout mice were monocytes (including macrophages). At 2 h after IL-1β treatment, the percentage of monocytes was only a little changed both in wild-type and Chop knockout mice BALF. At 6 h after IL-1β treatment, the percentage of neutrophil was markedly increased (∼50%), and the percentage of monocytes was decreased (∼45%) both in wild-type and Chop knockout mice BALF. The percentage of lymphocytes was also suppressed at 6 h both in wild-type and Chop knockout mice BALF. These results show that the inflammation cascade downstream of IL-1β secretion is not affected in Chop knockout mice.
Induction of caspase-11 was suppressed in the LPS-treated lung in Chop knockout mice
IL-1β is known to act as a proinflammatory cytokine and it also plays a key role in regulating the inflammation cascade (24). To examine whether the activation of the IL-1β pathway by LPS is suppressed in Chop knockout mice, we analyzed mRNAs for IL-1β-related proteins in LPS-treated mouse lung by RT-PCR (Fig. 3⇓A). IL-1β mRNA was not detected before LPS treatment, it was induced almost maximally at 6 h, and thereafter changed only slightly until 24 h in the wild-type mice. In the Chop knockout mouse lung, IL-1β mRNA was induced by the treatment, but the rate of induction decreased more than in the wild-type mice. mRNA for MIP-2, a chemokine which plays a major role in mediating the chemotaxis of neutrophils in inflammation (25), was not detected before treatment, but it was markedly induced at 6 h, and then it decreased at 24 h, both in wild-type and Chop knockout mice. However, the rate of induction of MIP-2 mRNA decreased in the Chop knockout mice more than in the wild-type mice. The results of IL-1β and MIP-2 show that LPS treatment triggers the inflammation cascade in the lung of wild-type and Chop knockout mice.
In stimulated inflammatory cells, IL-1β is synthesized as pro-IL-1β, and then it is activated by proteolytic processing before secretion (24), and both caspase-1 and caspase-11 are needed for this processing (19). Caspase-1 mRNA was detected before LPS treatment in the lung of both wild-type and Chop knockout mice, it was slightly induced by LPS treatment in wild-type mice, and it showed almost no change by LPS treatment in Chop knockout mice. In wild-type mice, caspase-11 mRNA was barely detected before treatment and then it was induced markedly at 6 h, and then it decreased at 24 h. In contrast, this induction was strongly attenuated in Chop knockout mice. These results suggest that the attenuation of the induction of IL-1β (activity) in BALF after LPS treatment in Chop knockout mice (Fig. 1⇑E) is due to the lack of caspase-11 induction. To analyze the induction of caspase-1 and caspase-11 by LPS treatment at the protein level, an immunoblot analysis was performed (Fig. 3⇑, B and C). In the lungs of wild-type mice, caspase-11 was not detected before LPS treatment, while it was induced markedly at 6 h after treatment, and then decreased at 24 h. In sharp contrast, the expression of caspase-11 was not detected in the Chop knockout mice after LPS treatment. In accordance with the results in mRNAs, caspase-1 was detected before treatment, then it was slightly induced at 6 h, and then it remained little changed thereafter in wild-type mice. The induction of caspase-1 was delayed in the Chop knockout mice, and it then gradually increased up to 24 h. The reason for the difference in the time course of caspase-1 induction is unknown. These results show that the induction of caspase-11 by LPS is suppressed in Chop knockout mice. Caspase-11 is needed for the activation of caspase-1 and IL-1β. Therefore, we conclude that suppression of LPS-induced lung inflammation and the secretion of IL-1β in BALF in Chop knockout mice is due to the lack of induction of caspase-11.
We also investigated whether the ER stress-CHOP pathway is activated in the LPS-treated mouse lung (Fig. 3⇑A). In wild-type and Chop knockout mice, mRNA for BiP, a major ER chaperone, was detected before treatment, and it was slightly induced by LPS treatment at 6 h, and then it decreased at 24 h. In wild-type mice, CHOP mRNA was barely detected before LPS treatment, then it was induced markedly at 6 h, and thereafter it increased gradually until 24 h. After ER stress, XBP1 mRNA is induced and subjected to unique and specific splicing by activated IRE1, one of the ER stress sensors. Next, the active form of XBP1 is produced, and activates the expression of the ER stress-response genes (26, 27, 28). Therefore, the induction of XBP1 mRNA and the formation of the spliced form of XBP1 mRNA are good markers of the activation of the ER stress pathway. The unspliced form of XBP1 mRNA was expressed at a low level before LPS treatment in wild-type mice (Fig. 3⇑D). It was markedly induced by LPS treatment and its spliced form appeared. Similar results were observed in the Chop knockout mice. This result shows that LPS activates the ER stress pathway in the lung of both wild-type and Chop knockout mice. Therefore, we conclude that LPS treatment induces the ER stress-CHOP pathway in the mouse lung.
Induction of caspase-11 by LPS was suppressed in the peritoneal macrophages from the Chop knockout mice
Regarding inflammation, macrophages play a key role in triggering the inflammation cascade and they are also a major source of IL-1β secretion. To analyze the direct effect of LPS on macrophages, peritoneal macrophages were prepared from wild-type and Chop knockout mice, and then were treated with LPS (Fig. 4⇓). As shown in Fig. 4⇓A, BiP and caspase-1 mRNAs were moderately induced by LPS in wild-type macrophages, and CHOP, caspase-11, IL-1β, and MIP-2 mRNAs were markedly induced. In Chop knockout macrophages, the induction of mRNAs for BiP, caspase-1, IL-1β, and MIP-2 was similar to that of wild-type cells. In contrast, the induction of caspase-11 mRNA was obviously suppressed in the Chop knockout cells. To analyze the induction of caspase-11 and its related molecules at the protein level, immunoblot analysis was performed (Fig. 4⇓B). In wild-type macrophages, neither CHOP nor caspase-11 were detected before LPS treatment, and then they were induced at 24 h after treatment. In contrast, caspase-11 was not induced in the Chop knockout cells. Both in wild-type and Chop knockout cells, caspase-1 was detected before LPS treatment, but it was only slightly induced at 24 h by LPS treatment. BiP was detected at a low level before LPS treatment in both cells, and thereafter it was gradually induced until 12 h in the wild-type cells. In the Chop knockout cells, BiP was only slightly induced at 12 h, and then it decreased at 24 h. CHOP is known to be induced by ER stress. However, CHOP itself was recently reported to activate the formation of reactive oxygen species and abnormal high m.w. protein complexes in ER, and thereafter it enhances ER stress (29). This may be the reason why the induction of BiP mRNA in Chop knockout mice lung by LPS is not as obvious as that seen in wild-type mice. These results show that LPS treatment activates the ER stress-CHOP pathway in macrophages. They also show that the induction of caspase-11 by LPS depends on the CHOP expression in macrophages. As already mentioned, caspase-1 is activated by caspase-11. To examine whether the activation of caspase-1 by LPS treatment is suppressed in Chop knockout macrophages, caspase-1 activity from wild-type or Chop knockout mice macrophages was measured (Fig. 4⇓D). The activity in LPS-treated wild-type macrophages was markedly increased (by about 4-fold) compared with that in nontreated cells. In contrast, the activity in the Chop knockout macrophages was little changed by LPS treatment. These results support the finding that the induction of caspase-11 by LPS is suppressed in peritoneal macrophages from Chop knockout mice.
Thapsigargin- and tunicamycin-induced caspase-11 in a CHOP-dependent manner
We next asked whether other ER stress-inducing stimuli induce caspase-11 in macrophages (Fig. 5⇓). Peritoneal macrophages from wild-type and Chop knockout mice were treated with ER stress inducers thapsigargin (Fig. 5⇓A, left panels) and tunicamycin (Fig. 5⇓A, right panels), and mRNAs for ER stress- and IL-1β-related genes were analyzed by RT-PCR. BiP mRNA was detected before treatment, and it was slightly induced by either thapsigargin or tunicamycin both in wild-type and Chop knockout cells. CHOP mRNA was present at a low level before treatment, and it was markedly induced by either thapsigargin or tunicamycin in wild-type macrophages. These results show the ER stress pathway to thus be activated by thapsigargin and tunicamycin.
Caspase-1 mRNA was detected before treatment, and it was slightly changed by these reagents both in wild-type and Chop knockout cells. Caspase-11 was barely detected before treatment, and it was strongly induced by thapsigargin and tunicamycin in wild-type macrophages. In contrast, it was not detected either before and after treatment in Chop knockout cells. IL-1β and MIP-2 mRNAs were not detected before and after treatment in wild-type and Chop knockout macrophages (data not shown). These results show that the inflammation pathways are not activated by treatment with thapsigargin or tunicamycin. To analyze the induction of caspase-11 and related proteins, an immunoblot analysis was performed (Fig. 5⇑, B and C). When wild type macrophages were treated with thapsigargin or tunicamycin, CHOP which was not detected before treatment, while it was induced at 6 h, reached a maximum at 12 h, and thereafter decreased or remained only slightly changed at 24 h after treatment. In wild type macrophages, caspase-11 was not detected before treatment and then it gradually increased up to 24 h after treatment with thapsigargin or tunicamycin. In Chop knockout cells, caspase-11 was not induced by the treatment with thapsigargin or tunicamycin. Caspase-1 was only slightly changed by the treatment with thapsigargin or tunicamycin both in wild-type and Chop knockout cells. BiP was detected at a low level before treatment, and then it gradually was induced up to 24 h by the treatment with thapsigargin or tunicamycin in both cells. These results show that ER stress induces caspase-11 in macrophages in a CHOP-dependent manner.
CHOP and caspase-11 were coinduced in LPS-treated macrophages
To determine or see whether CHOP and caspase-11 are expressed in the same cells in the lung of LPS-treated mice, an immunohistochemical analysis was performed (Fig. 6⇓). LPS was given to wild-type and Chop knockout mice intratracheally. After 12 h, an immunostaining analysis of the lung with an Ab against caspase-11 was performed (Fig. 6⇓A). Caspase-11-positive cells were not detected in the lung of PBS-treated (control) wild-type mice. We could detect caspase-11-positive cells in the lung of LPS-treated wild-type mice. These cells were morphologically considered to be macrophages. In contrast, caspase-11-positive cells were not detected in the lungs of LPS-treated Chop knockout mice. These results correlate with those of an immunoblot analysis as shown in Fig. 3⇑. The findings of double immunostaining with Abs against CHOP and caspase-11 are shown in Fig. 6⇓B. CHOP and caspase-11 were induced by LPS in the same cells. CHOP was localized in the nuclei, whereas caspase-11 was localized in the cytoplasm. These results show that CHOP and caspase-11 are coinduced in the same cells in the lung of the LPS-treated mice.
To examine whether CHOP induces caspase-11, transfection experiments using peritoneal macrophages were performed (Fig. 7⇓A). Peritoneal macrophages from wild-type mice were transfected with an expression plasmid for CHOP, and double immunostaining with Abs against CHOP and caspase-11 was performed. We could thus detect the expression of caspase-11 in CHOP-positive cells. Fig. 7⇓B shows that CHOP and caspase-11 were coinduced in thapsigargin-treated macrophages. In PBS-treated macrophages, CHOP and caspase-11 were barely detected. CHOP and caspase-11 were also coinduced in LPS-treated macrophages (Fig. 7⇓C). These results indicate that LPS induces caspase-11 in macrophages via CHOP induction as ER stress inducers.
Inflammatory stimuli induce apoptosis through various pathways (30). However, little is known about the possible involvement of ER stress- and CHOP-mediated apoptosis in inflammatory diseases and the pathogenesis of lung diseases. Cystic fibrosis is a hereditary disorder of the exocrine glands, and its major symptom is chronic obstructive pulmonary disorder (31). However, the involvement of CHOP in this condition remains unknown. In this report, we found that LPS induces caspase-11 and the activation of IL-1β through the ER stress-CHOP pathway (Fig. 8⇓). In the early stage of inflammation, the secretion of IL-1β by macrophages is induced and the following inflammation processes involving the activation of Ag-specific lymphocytes are triggered. Therefore, the induction of caspase-11 and the following activation of caspase-1 are crucial steps in the start of the inflammation cascade. In fact, caspase-1 knockout mice have been reported to be resistant to LPS treatment (17), while caspase-11-deficient oligodendrocytes are resistant to IFN-γ- or TNF-α-induced cell death (32). Our present work suggests that the ER stress- and the CHOP-mediated pathway plays a key role in inflammation, not only in the lung but also in other tissues, because LPS activates the ER stress pathway in primary cultured peritoneal macrophages. The present study clearly shows that CHOP plays a crucial role in the pathogenesis of LPS-induced inflammation at least in the lung.
Schauvliege et al. (20) reported that caspase-11 induction by LPS and IFN-γ requires NF-κB and STAT1 in a macrophage cell line. LPS generally induces the activation of NF-κB and STAT1 in macrophages (33, 34). In the present study, we found that caspase-11 induction by LPS is attenuated in CHOP-deficient macrophages, thus indicating that CHOP is needed for the induction of caspase-11. The activation of p38 MAPK has also been reported to be involved in the induction pathway of caspase-11 (35). CHOP may be involved in p38 MAPK-mediated induction of caspase-11, because the transcriptional enhancer activity of CHOP is enhanced through the phosphorylation by p38 MAPK (36). It remains to be elucidated as to whether CHOP directly activates the transcription of the caspase-11 gene.
The precise mechanism of CHOP-mediated apoptosis is still unknown. CHOP is a transcription factor. Therefore, there must be a target gene(s) which is activated by CHOP and is involved in the apoptosis signal cascade. Wang et al. (37) reported candidate target genes of the CHOP protein by using a representational difference analysis. However, these genes are distinct from the known factors involved in the ER stress response and apoptosis. McCullough et al. (38) reported that CHOP expression results in a down-regulation of the antiapoptotic molecule Bcl-2 expression, the depletion of cellular glutathione, and an exaggerated production of reactive oxygen species. Marciniak et al. (29) reported that CHOP inhibits ER stress-induced attenuation of protein synthesis by the dephosphorylation of the α-subunit of translation initiation factor 2 (eIF2α) through the induction of growth arrest and DNA damage-inducible gene 34. They thus showed that the new protein synthesis in ER stress conditions leads to the accumulation of the high m.w. protein complex in ER, and it thus impairs ER function. We showed that the apoptosis signal induced by CHOP is transmitted to the mitochondria through the translocation of a proapoptotic molecule Bax from the cytosol to the mitochondria. However, the mechanism regarding how CHOP induces the translocation of Bax is still unknown. We herein showed that CHOP induces the activation of IL-1β through the induction of caspase-11 in inflammation. Therefore, we speculate that cytokines, which are induced by IL-1β, are involved in the induction of apoptosis in inflammatory lesions. Wang et al. (19) reported that the activation of caspase-11 precedes that of caspase-1, caspase-3, and caspase-7. The activation of caspase-3 and caspase-7 was significantly suppressed in the absence of caspase-11. They also showed that the early activation of caspase-3 and caspase-7 by caspase-11 is not affected by blocking caspase-1 activity and IL-1β release, implying that caspase-11 activates caspase-3 and caspase-7 independently of caspase-1 activation. Caspase-3 and caspase-7 are classified into the apoptosis-effector subgroup of caspases. Therefore, if caspase-3 and caspase-7 are directly activated by caspase-11, then the mitochondria can be skipped in caspase-11-mediated apoptosis. Miura et al. (18) showed that an overexpression of caspase-1 induces apoptosis in cultured cells. They also showed that caspase-1-induced apoptosis is suppressed by Bcl-2. Bcl-2 suppresses apoptosis through the inhibition of the mitochondrial pathway. Therefore, this result suggests that caspase-1-induced apoptosis is mediated by the mitochondrial pathway. We previously reported that Bax is involved in ER stress- and CHOP-mediated apoptosis (18). It remains to be elucidated as to whether Bax is involved in the apoptosis pathway induced by caspase-1 and caspase-11.
The expression of caspase-1 and caspase-11 is not restricted to inflammatory cells (39, 40). Therefore, we speculated that caspase-1 and caspase-11 are also involved in ER stress-mediated apoptosis induced by noninflammatory stimuli. In fact, we found that the ER stress inducers thapsigargin and tunicamycin induce caspase-11 in primary cultured macrophages. The induction of caspase-11 by these ER stress-inducing reagents was CHOP dependent. Caspase-1 is constitutively expressed in macrophages. Therefore, we speculate that caspase-1 and caspase-11 are generally involved in the ER stress- and CHOP-mediated apoptosis pathway, at least in macrophages. The question regarding whether caspase-11 is induced by ER stress in nonmacrophage cells thus remains to be elucidated.
As far as we know, this is the first report showing that the CHOP-caspase-11 pathway plays an important role in the pathogenesis of inflammation in vivo. Severe inflammatory diseases of the lung can be fatal, and the CHOP-caspase-11 pathway may thus become a new target for the therapy of these diseases.
We thank our colleagues for their valuable suggestions and discussion, and Brian Quinn for comments on the manuscript. We also thank Rieko Shindo and Yasuko Indo for technical assistance.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This work was supported in part by Grants-in-Aid (No. 14037257 to M.M. and No. 16590233 to T.G.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and grants (to T.G.) from the Mitsubishi Pharma Research Foundation and the Inamori Foundation.
↵2 Address correspondence and reprint requests to Dr. Tomomi Gotoh, Department of Molecular Genetics, Graduate School of Medical Sciences, Kumamoto University, Honjo 1-1-1, Kumamoto 860-8556, Japan. E-mail address:
↵3 Abbreviations used in this paper: ER, endoplasmic reticulum; BiP, Ig H chain binding protein; CHOP, C/EBP homologous protein; Hsc, heat shock cognate protein; BAL, bronchoalveolar lavage; BALF, BAL fluid; XBP, X-box binding protein; ICE, IL-1β-converting enzyme.
- Received November 22, 2005.
- Accepted February 22, 2006.
- Copyright © 2006 by The American Association of Immunologists