Short-term hypoxic pretreatment is an effective approach to protect the lung from subsequent prolonged hypoxic injury under conditions such as lung transplantation, shock, and trauma. However, the signaling pathways are not well understood. By use of high-throughput, two-dimensional electrophoresis combined with mass spectrometry, we found that short-term hypoxic treatment upregulated calreticulin (CRT), an endoplasmic-reticulum stress protein, in A549 human type II alveolar epithelial cells. Genetic manipulation of CRT expression in A549 cells through small interferring RNA inhibition or overexpression demonstrated a positive correlation between CRT expression level and cell viability in subsequent prolonged hypoxia, which indicates that CRT is a key mediator of short-term hypoxia-induced cell protection. Importantly, CRT overexpression prevented reactive oxygen species (ROS) accumulation during prolonged hypoxia by inducing the expression of thioredoxin (TRX), an antioxidant, in A549 cells. Furthermore, CRT promoted the nuclear translocation of nuclear factor-E2-related factor 2, the transcription factor of TRX. Finally, overexpressing an inactive TRX mutant reversed the effects of CRT on ROS accumulation and cell protection. Our results demonstrate that CRT stimulates the anti-oxidant pathway and contributes to short-term hypoxia-induced protection in A549 type II alveolar epithelial cells, which may have potential therapeutic ramifications for hypoxic pulmonary diseases.

  • cellular protection
  • hypoxic injury
  • calreticulin
  • thioredoxin
  • reactive oxygen species

episodes of short-term hypoxic treatment increase tolerance to subsequent prolonged hypoxia (23) and represent a genetically conserved, intrinsic, and self-protective process existing in various organs, including the heart, brain, kidney, liver, and lung (27). In the lung, short-term hypoxic treatment reduces alveolar protein leakage, improves gas exchange and dynamic compliance, and enhances donor lung preservation during lung transplantation (7, 8). Type II alveolar epithelial cells, of which their functions include secreting surfactant, resorbing sodium and water, and repairing the damaged alveolar epithelium (37), play an important role in maintaining the functional and morphological integrity of alveoli. Preserving the viability and normal functions of these cells represents an important approach to ameliorating lung injury. So far, how to protect type II alveolar epithelial cells during hypoxia and how these cells respond to short-term and prolonged hypoxic treatments have not been well studied.

Reactive oxygen species (ROS) are key signaling molecules mediating many types of cell injury, especially that induced by hypoxia (34). During prolonged hypoxia, excessive ROS production and/or its insufficient removal lead to oxidative stress, an imbalance of redox status; cell dysfunction through directly interacting with DNA, proteins, and lipids. Meanwhile, a small amount of ROS elicited by short-term hypoxia may have beneficial effects through activating multiple signaling pathways such as c-Jun NH2-terminal kinase (JNK) and p38 mitogen-activated protein kinase (MAPK) (29, 32) and stimulating intracellular antioxidant defense systems such as glutathione and thioredoxin (TRX) (36). Therefore, manipulating ROS signaling has been recognized as an important approach to treat hypoxic injury.

Recently, two-dimensional electrophoresis (2-DE) has been used to study protein profile changes under pathological conditions. In the present study, we adopted this high-throughput technology to identify novel signaling molecules induced by short-term hypoxia. Among several candidates, an endoplasmic reticulum (ER) protein calreticulin (CRT) was found to be significantly upregulated by short-term rather than prolonged hypoxic treatment. Moreover, the short-term hypoxia-induced CRT expression prevented A549 cell injury during subsequent prolonged hypoxic treatment. CRT is a stress-inducible protein well known for its Ca2+ regulation and glycoprotein-folding functions (14, 22). Our data revealed a novel signaling pathway of CRT: the upregulation of antioxidant protein TRX and subsequent inhibition of ROS accumulation. We also validated that CRT protected A549 type II alveolar epithelial cells against hypoxic injury through this antioxidant pathway.



Reagents for 2-DE were from Amersham Pharmacia Biotechnology (Uppsala, Sweden). Anti-CRT antibody was from Stressgen Biotech (Victoria, BC, Canada). Anti-TRX antibody and anti-nuclear factor-E2-related factor 2 (Nrf2) antibody were from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-actin antibody was from Abcam (Cambridge, MA). Cycloheximide was from Amresco (Solon, OH). 2′,7′-Dichlorofluorescin diacetate (DCFH-DA), N-acetyl-l-cysteine (NAC), and N-(2-mercaptopropionyl)-glycine (MPG) were from Sigma (St. Louis, MO). Cell culture media and supplements were from Hyclone (South Logan, UT). The BBL GasPak anaerobic system was from Becton Dickinson Biosciences (San Jose, CA). The PathDetect c-Jun transreporting system was from Stratagene (La Jolla, CA). Human full-length CRT cDNA pCMV-SPORT6 plasmid was a kind gift from Dr. Marek Michalak (University of Alberta, Edmonton, AB, Canada). The dominant-negative TRX plasmid pcDNA3-TRX (C32/35S) was kindly provided by Dr. J. Yodoi (Kyoto University, Kyoto, Japan) (11). The constitutively activated JNK plasmid pcDNA3-MKK7-JNK1 was a kind gift from Dr. Roger J. Davis (University of Massachusetts Medical School, Worcester, MA) (17).

Cell culture, short-term and prolonged hypoxia models.

The human type II alveolar epithelial A549 cell line was from ATCC (Manassas, VA). Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum. For all experiments, cells were used at 70–80% confluence. Producing the short-term and prolonged hypoxia model in A549 cells was as reported previously (38). The experimental groups were 1) controls: A549 cells cultured in serum-free DMEM for 6 h; 2) prolonged hypoxia group: A549 cells cultured with oxygen-glucose deprivation (OGD) medium (Hanks' balanced salt solution without serum, glucose and oxygen) in the anaerobic system for 6 h; 3) short-term plus prolonged-hypoxia group: A549 cells first cultured in OGD medium for 10 min, then incubated in serum-free DMEM for 10 min, and finally subjected to hypoxia for 6 h; and 4) ROS scavenger group: cells were incubated with NAC (2 mM) or MPG (1 mM) during 6-h hypoxia. Our previous observation shown in Fig. 1A indicated that prolonged hypoxia-induced cell damage was gradually alleviated with reoxygenation, therefore, we measured morphological changes, lactate dehydrogenase (LDH) release, and trypan blue staining of cells to verify cell injury at the end of 6-h hypoxia.

Fig. 1.

Setting up short-term versus prolonged hypoxia model in A549 cells and proteomic analysis. A: morphology of A549 cells after 6-h prolonged hypoxia and 24-h reoxygenation under light microscope (×200). n = 3 experiments. Note the recovery of cell viability after reoxygenation. B: short-term hypoxia model in A549 cells. Cell death rate was assessed by trypan blue exclusion test and lactate dehydrogenase (LDH) release rate after 6-h prolonged hypoxia (H) or short-term plus prolonged hypoxia (SH+H). n = 4 experiments; *P < 0.05 vs. Control; #P < 0.05 vs. H. C: representative two-dimensional gel images illustrating protein profiles of A549 cells subjected to prolonged H or SH+H. Proteins were detected by Coomassie blue R-350 staining. Arrows on the locally enlarged images (bottom panels) indicated calreticulin (CRT). n, 3 experiments.

2-DE and protein identification.

Cells were first lysed in buffer containing 7 M urea, 2.5 M thiourea, 2% CHAPS, 0.8% pharmalyte (pH 3–10), 1% DTT, and 10 mM Pefabloc. Isoelectric focusing involved the use of the IPGphor unit (Amersham Pharmacia Biotechnology, Uppsala, Sweden) at 20°C. The isoelectric focusing program accumulated up to 40,000 Vh. For the second dimension (SDS-PAGE), immobilized pH gradient strips were first equilibrated with 1% DTT, and then 2.5% iodoacetamide and 12% SDS gel were used. Gels were stained with 0.025% Coomassie Blue R-350 and then analyzed by use of PDQUEST 7.3.0 software (Bio-Rad, Hercules, CA). The differently expressed spots were digested with sequence-grade trypsin and then the peptides underwent matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS, Bruker Reflex III) for peptide mass fingerprints. Finally, proteins were identified by use of MASCOT software.

Western blot analysis and immunofluorescence.

For Western blot analysis, A549 cells were lysed with sample buffer containing 150 mM NaCl, 100 mM Tris-HCl, 1% Triton X-100, 1 mM EDTA, 10 μg/ml aprotinin, 10 μg/ml pepstatin A, and 10 μg/ml leupeptin. Nuclear proteins were extracted by use of a nuclear and cytoplasmic extraction kit from Pierce. After quantification of the protein concentration with the BCA method (Pierce, Rockford, IL), samples were loaded onto bis-acrylamide gels and separated by SDS-PAGE. Separated proteins were transferred to polyvinyldifluoride membranes and incubated with the indicated primary antibody and horseradish peroxidase-conjugated secondary antibody and revealed by use of a chemiluminescence kit (Pierce).

For immunofluorescent analysis, A549 cells were fixed in acetone at 4°C for 20 min and then permeabilized by 0.05% Triton X-100 and 0.5% BSA in PBS for 10 min. After being washed with PBS, cells were incubated with the indicated primary antibody overnight at 4°C and then with FITC-conjugated secondary antibody for 1 h at room temperature. Finally, cells were mounted and viewed on laser scanning confocal microscopy (Leica, Bannockburn, IL).

Stable transfer of CRT into A549 cells and transient transfection assays.

To construct the CRT cDNA expressed in mammalian cells, we subcloned the full-length CRT cDNA into a pcDNA3.1 vector; the resulting plasmid was named pcDNA3.1-CRT. We transferred the pcDNA3.1-CRT plasmid into A549 cells using lipofectamine (Invitrogen, Carlsbad, CA). After cultured in the presence of 800 μg/ml G418 (Calbiochem, La Jolla, CA) for 10 days, stably transferred cells were obtained and subcultured in the presence of 500 μg/ml G418 to maintain the long-term expression of the transferred gene.

The 5′-flanking region of the calreticulin gene, from position −518 to +64 relative to the putative major transcriptional start site, was amplified by PCR and then cloned into the pGL3 vector upstream of luciferase cDNA (Promega, Madison, WI). We constructed 5× ATF6-Luc, in which the firefly luciferase reporter is driven by five copies of the ATF6 binding element, as previously described (39). We transiently cotransferred plasmids from the luciferase reporter gene or PathDetect c-Jun transreporting system together with the β-gal-expressing plasmid as an internal reference using a cationic polymer transfection reagent (JetPEI, Qbiogen, Illkirch, France). The luciferase activity and β-gal activity were determined by use of kits from Promega (Madison, WI).

Real time RT-PCR.

To detect the mRNA levels of CRT and TRX, total RNA was extracted and reverse transcribed by use of kits from Promega. cDNA was amplified by real-time PCR with the following primers: CRT, 5′-ACGCAAAGAGGAGGAGGAGGCAGAGGAC-3′ and 5′-GTGGGGGAGAGTGGAGGAGGGGAACAAA-3′; and TRX, 5′-TTCATTAATGGTGGCTTCAAGC-3′ and 5′-TGAAGCAGATCGAGAGCAAGAC-3′.

Stealth small interfering RNA inhibition of gene expression.

CRT-specific small interfering RNA (siRNA) is a 25-bp duplex oligoribonucleotide with a sense strand corresponding to nucleotides 166–190 of the reported human CRT mRNA sequence (GeneBank accession number BC020493). The sense sequence of the CRT-specific siRNA was 5′-CCCGCUGGAUCGAAUCCAAACACAA-3′. A scrambled siRNA with a sense strand 5′-CCCGUAGCUAGUACCAACAACGCAA-3′ was used as the siRNA control. All of these oligoribonucleotides were synthesized by Invitrogen. A549 cells were transfected with 250 pmol siRNA by use of lipofectamine 2000 (Invitrogen). After siRNA treatment for 24 h, the efficiency of siRNA CRT was identified by RT-PCR and Western blot analysis.

Measurement of intracellular ROS generation.

Determination of intracellular ROS production was based on the oxidation of DCFH-DA by intracellular H2O2 to form the fluorescent compound: 2′,7′-dichlorofluorescin (DCF) (31). DCFH-DA (5 μM) was loaded into the cells before hypoxia. After 6 h of hypoxia, DCF fluorescence was monitored by laser scanning confocal microscopy (Leica, Bannockburn, IL).

Statistical analysis.

The results are expressed as means ± SE. Data analysis involved GraphPad Prism software. One-way ANOVA, Student-Newman-Keuls test (for comparisons between multiple groups) or unpaired Student's t-test (for comparisons between 2 groups) was used as appropriate. P < 0.05 was considered significant.


Short-term hypoxia upregulated CRT expression in human type II alveolar cells.

To study the effect of short-term hypoxia on type II alveolar epithelial cells and its signaling mechanisms, we established cellular models of short-term and prolonged hypoxia with A549 human type II alveolar epithelial cells. The basal death rate of A549 cells cultured in serum-free DMEM for 6 h was 11.93 ± 1.98%, as measured by trypan blue staining. After prolonged hypoxia for 6 h, the number of trypan blue-positive cells was greatly increased. Consistantly, the rate of LDH released into the culture medium was doubled (Fig. 1B). Short-term hypoxic treatment (10-min hypoxia and 10-min normal medium culture before 6-h hypoxia) effectively lowered the cell death rate and prevented LDH release (Fig. 1B), which suggests that the cellular model was successfully created in A549 cells.

Since hypoxia is a complex process and involves various molecules and proteins (2), we used high-throughput proteomic techniques, combining 2-DE with mass spectrometry, to identify novel candidates that may be involved in the protective effect of short-term hypoxia in A549 cells. By use of PDQUEST software, we analyzed ∼300 legible protein spots on 2-DE gel, of which 70% matched between groups (Fig. 1C). The spots with more than twofold difference in density were excised and digested by trypsin and underwent MALDI-TOF identification. At least four proteins, including CRT (Fig. 1C), thioredoxin peroxidase, annexin II, and heat shock protein 27 (data not shown), were significantly upregulated by short-term hypoxia.

Although CRT has been linked to stress-related perturbations (12, 20), no previous report has showed that short-term rather than prolonged hypoxia upregulates CRT expression. Therefore, we first confirmed the proteomic result by Western blot analysis. When compared with the control and hypoxia-alone groups, the short-term hypoxia group showed a 1.5-fold increase in CRT protein level (Fig. 2A). We next explored the underlying mechanisms of elevated CRT expression. We investigated CRT protein level in a short-term hypoxia group over a broad range of time and found that elevated CRT protein level occurred at 2 h and sustained up to 8 h after short-term hypoxic treatment (Fig. 2B). Moreover, preincubation with cycloheximide, an inhibitor of protein synthesis, abolished the increased CRT protein level, which suggests the involvement of de novo protein synthesis (Fig. 2B). Furthermore, parallel measurement of mRNA by real-time RT-PCR revealed a decrease in CRT mRNA level by prolonged hypoxic treatment and a transient increase in CRT mRNA level with short-term hypoxic treatment that peaked at 6 h and gradually decreased to the basal level at 8 h (Fig. 2C). Moreover, short-term hypoxic treatment partially reversed the decreased CRT promoter activity by prolonged hypoxia (Fig. 2D). Since CRT gene expression relies on the binding of transcription factor ATF6 to the ER stress response element in the promoter region of the CRT gene (42), we transferred the luciferase reporter gene with 5× ATF6 binding sites into A549 cells and confirmed that ATF6 binding activity decreased during prolonged hypoxia, which was partially reversed by short-term hypoxic pretreatment (Fig. 2D). Taken together, the results show that short-term hypoxia induces CRT expression mainly through translational regulation, and the transient and partial restoration of CRT transcription may play an additional role.

Fig. 2.

Short-term hypoxia pretreatment upregulated CRT in A549 cells. A: short-term rather than prolonged hypoxia upregulated CRT protein expression in A549 cells detected by Western blot analysis. B: time course of CRT protein expression induced by short-term hypoxia in A549 cells detected by Western blot analysis and its blockade by 10 μg/ml cycloheximide, an inhibitor of protein synthesis. C: short-term hypoxia upregulated CRT mRNA level detected by real-time RT-PCR. D: short-term hypoxia promoted CRT gene transcription and ATF6 transcriptional activity. A549 cells were transiently transfected with luciferase reporter gene driven by CRT promoter or 5× ATF6 binding promoter site and subjected to prolonged hypoxia alone or with short-term hypoxia pretreatment. H, prolonged hypoxia for 6 h; SH+H, short-term hypoxia pretreatment. n = 3–4 experiments. *P < 0.05 vs. Control; #P < 0.05 vs. prolonged H.

CRT protected A549 cells against prolonged hypoxia.

To study whether this upregulated CRT played a role in cell protection, we used molecular and genetic approaches to manipulate CRT expression in A549 cells. On one hand, CRT siRNA specifically attenuated both the mRNA and protein levels of CRT in A549 cells compared with the scrambled double-stranded (ds) RNA control (Fig. 3, A and B). In these endogenous CRT knockdown cells, short-term hypoxia failed to have any protective effect, as assessed by LDH release (Fig. 3C). On the other hand, transfection of full-length CRT cDNA into A549 cells resulted in CRT overexpression, as shown by Western blot analysis and immunofluorescent assay (Fig. 4, A and B). The overexpressed CRT protected A549 cells against prolonged hypoxic damage, mimicking the effect of short-term hypoxia treatment (Fig. 4C). Thus manipulations of CRT expression level lead to parallel outcomes of cell protection against hypoxic injury, indicating the elevated CRT expression induced by short-term hypoxia renders a crucial mechanism to confront hypoxic injury.

Fig. 3.

CRT mediated the protective effect of short-term hypoxia pretreatment in A549 cells. A and B: identification of the efficiency of CRT small interfering RNA (siRNA) by RT-PCR (A) and Western blot analysis (B). Scrambled double-stranded RNA (dsRNA) is used as a control of CRT siRNA. C: CRT knockdown by siRNA diminished the protective effect of short-term hypoxia in A549 cells. n = 3. *P < 0.05 vs. Control, #P < 0.05 vs. prolonged H, +P < 0.05 vs. SH+H in scrambled dsRNA group.

Fig. 4.

Overexpression of CRT mimicked the protective effect of short-term hypoxia pretreatment in A549 cells. A and B: verification of effective CRT cDNA transfection by Western blot analysis (A) and immunofluorescence (B). Vector: transferred with pcDNA3.1. Scale bar: 20 μm. C: overexpression of CRT reduced cell injury induced by prolonged hypoxia. n = 3 experiments. *P < 0.05 vs. Control, #P < 0.05 vs. Vector. D: CRT reduced reactive oxygen species (ROS) accumulation induced by prolonged H in A549 cells. Scale bar: 20 μm. n = 4 experiments. *P < 0.05 vs. Control; #P < 0.05 vs.Vector. E: morphology of A549 cells after 6 h prolonged hypoxia with or without 2 mM N-acetyl-l-cysteine (NAC) or 1 mM N-(2-mercaptopropionyl glycine) (MPG) (×200). n = 3 experiments.

CRT inhibited ROS accumulation during prolonged hypoxia.

Although ROS generated during preconditioning may have a beneficial effect (2), excessive ROS production during prolonged hypoxia is the key player mediating cell injury (34). Therefore, we examined whether CRT overexpression affected the ROS accumulation during prolonged hypoxia. Consistent with previous reports, prolonged hypoxia induced more than fourfold increase in intracellular ROS levels. However, in cells overexpressing CRT, hypoxia-induced ROS accumulation was largely abolished (Fig. 4D). Indeed, in the presence of ROS scavenger NAC (2 mM) or MPG (1 mM), prolonged hypoxia-induced cell damage was largely diminished (Fig. 4E). These results further confirm the deleterious effect of excessive ROS generated during prolonged hypoxia in A549 cells. Taken together, CRT inhibits ROS accumulation in hypoxic lung epithelial cells. The antioxidant effect of CRT may provide a novel signaling mechanism underlying its cell protective role.

CRT upregulated TRX expression in A549 cells.

Since CRT inhibited ROS accumulation during prolonged hypoxia, we hypothesized that upregulated cellular antioxidant systems may play an important role in this effect. Therefore, we examined whether CRT influenced the expression of TRX (one of the major intracellular antioxidants) in A549 cells. As shown in Fig. 5A, cells overexpressing CRT showed significantly elevated TRX protein levels under both normal and hypoxic conditions. Meanwhile, cells with endogenous CRT knocked down by siRNA showed decreased TRX protein level (Fig. 5B). Therefore, CRT is both sufficient and necessary to induce TRX expression. Consistently, measurement of TRX mRNA level by quantitative real-time PCR showed elevation of TRX mRNA level by 50% with CRT overexpression (Fig. 5C), whereas knockdown of endogenous CRT by siRNA decreased the TRX mRNA level by ∼30% (Fig. 5D). Thus CRT upregulates TRX gene expression at both the transcriptional (Fig. 5C) and translational level (Fig. 5A) in A549 cells.

Fig. 5.

CRT upregulated the expression of TRX at both mRNA and protein levels. A: overexpression of CRT increased TRX protein level under both normal and prolonged hypoxia conditions. n = 3 experiments. *P < 0.05 vs. Vector. B: CRT knockdown by siRNA decreased TRX protein level. n = 4. *P < 0.05 vs. scrambled dsRNA control. C: quantitative real-time PCR results showing that overexpression of CRT increased TRX mRNA level. n = 4 experiments. *P < 0.05 vs. Vector. D: quantitative real-time PCR results showing CRT knockdown by siRNA decreased TRX mRNA level. n = 2 experiments. *P < 0.05 vs. scrambled dsRNA.

The transcriptional regulation of TRX by CRT was further supported by studying Nrf2, the transcription factor of TRX (13). Immunofluorescence assay showed that Nrf2 was located inside nuclei in CRT-overexpressing cells but remained in the cytosol in control cells (Fig. 6A). Western blot analysis revealed that CRT overexpression significantly increased Nrf2 protein in the nuclear fraction, thus confirming its nuclear translocation (Fig. 6B). These data suggest that CRT activates Nrf2 by promoting its translocation into nuclei and, therefore, leading to TRX transcription.

Fig. 6.

CRT induced nuclear factor E2-related factor (Nrf2) nuclear translocation in A549 cells. A: immunofluorescence results showing nuclear accumulation of Nrf2 in CRT-overexpresssed cells. Scale bar: 10 μm. B: Western blot analysis results showing overexpression of CRT increased the nuclear translocation of Nrf2. Bottom band in the nuclear portion is nonspecific. n = 2 experiments.

TRX mediated the protective effect of CRT in A549 cells.

TRX is an important antioxidant inside the cell. Its redox-regulating ability relies on a conserved sequence (32–35 amino acids, Cys-Gly-Pro-Cys) for sensing redox potentials. To further explore whether CRT-induced TRX expression mediated the protective effect of CRT, we used a loss-of-function mutant of TRX (TRX-DN), in which the two cysteines in the redox center of TRX were substituted by serines (28). Transfer of TRX-DN into CRT-overexpressing A549 cells diminished the inhibitory effect of CRT on ROS accumulation (Fig. 7, A and B). As well, TRX-DN reversed the cell-protective effect of CRT (Fig. 7D). These two lines of evidence validate TRX as a crucial mediator in the protective effect of CRT in A549 cells.

Fig. 7.

Thioredoxin (TRX) mediated the effects of CRT on ROS accumulation and cell injury in A549 cells. A: transfection with TRX dominant-negative plasmid (TRX-DN) reversed the effect of CRT on ROS accumulation during prolonged hypoxia. Scale bar: 20 μm. B: summarized ROS fluorescence intensity data. n = 4 experiments. *P < 0.05 vs. Control; #P < 0.05 vs. Vector; +P < 0.05 vs. CRT. C: CRT inhibited MAPKK-induced c-Jun NH2-terminal kinase (JNK) activation. n = 4 experiments. *P < 0.05 vs. Vector. D: TRX-DN and constitutively activated JNK transfection reversed the CRT effect on cell protection. n = 3. C: CRT inhibited MAPKK-induced c-Jun NH2-terminal kinase (JNK) activation. *P < 0.05 vs. Control; #P < 0.05 vs. Vector; +P < 0.05 vs. CRT.

TRX has been shown to inhibit apoptotic signaling through interacting with the apoptosis signal-regulation kinase 1 (ASK1)-JNK pathway, a pro-apoptotic pathway (30). JNK is also a downstream signaling factor of ROS (32). Therefore, we examined whether JNK is involved in the CRT-TRX-ROS signaling cascade. Using a PathDetect c-Jun transreporting system, we found that CRT significantly attenuated MAPKK-induced JNK activation (Fig. 7C). Moreover, constitutively activated JNK significantly reversed the protective effect of CRT (Fig. 7D). Taken together, these data show that CRT plays a cytoprotective role against hypoxic injury in A549 cells through upregulating TRX and subsequently inhibiting ROS signaling and JNK activity. The CRT-TRX-ROS pathway is a novel signaling pathway specifically activated by short-term hypoxia in human type II alveolar epithelial cells.


In the present study, we provide convincing evidence supporting the crucial role of a new intracellular pathway mediated by CRT in short-term, hypoxia-induced cell protection. By use of molecular biological approaches, we first demonstrate the upregulation of CRT by short-term hypoxia in these cells. Furthermore, we show that the cellular CRT level is associated with cell survival. Moreover, we identify that increased TRX expression and subsequent decreased ROS accumulation and JNK activity account for the mechanism underlying the protective effect of CRT in A549 cells.

The stress protein CRT can be induced by heat shock, heavy metals, amino acid starvation, and other ER stress signals (5, 10, 25). Here, we showed that a novel specific stimulation, brief rather than prolonged hypoxia, can induce CRT expression in A549 type II alveolar epithelial cells. In agreement with our findings, Liu and colleagues (21) recently demonstrated that ischemic preconditioning upregulated CRT expression in cultured neonatal rat cardiomyocytes. Therefore, CRT could be a common response factor induced by various stress stimulations. Furthermore, we found that short-term hypoxia upregulated the nuclear translocation of ATF6, a well-known transcription factor responsible for upregulation of many ER chaperone genes, especially CRT, following various ER stresses (42). Recently, p38 MAPK was found to be involved in the transcriptional regulation of the CRT gene following ER stress and ischemic preconditioning in HEK-293 cells, PC12 cells, and cardiomyocytes (19, 41). Also, preconditioning triggers the activation of p38 MAPK (24), which may be the upstream signaling molecule leading to ATF6 activation and CRT induction in A549 cells.

In A549 type II alveolar epithelial cells, the upregulated CRT plays a crucial role in cell survival. This finding is supported by previous studies showing the cytoprotective effect of CRT in response to such perturbations as nitric oxide, calcium ionophore, H2O2, and iodoacetamide in pancreatic β-cells, LNCaP prostate cancer cells, and renal epithelial cells, respectively (12, 20, 26, 43). Therefore, various environmental injuries may activate CRT. Of note, the protective effect of short-term hypoxic treatment disappeared when endogenous CRT was knocked down by siRNA, which suggests that endogeneous CRT is necessary for this process. As far as we know, this is the first report showing that CRT is an essential mediator of short-term, hypoxia-induced cell protection. Recently, other ER proteins, including glucose-regulated protein 78 (GRP78), GRP94, and oxygen-regulated protein 150, have been reported to protect neurons against hypoxic/ischemic-induced cell death (1, 9, 35). Also, in cardiomyocytes, GRP78 may play an important role in preconditioning (33). Our findings further indicate that ER resident proteins are central players involved in the cell protective mechanisms under stress stimulation.

The TRX system is an important self-defense system in maintaining cell viability and functions through ameliorating oxidative stress directly, via redox reaction with ROS, and indirectly, by protein-protein interactions with key signaling molecules such as ASK1 (28). TRX exerts most of its antioxidant properties through reducing thioredoxin peroxidase (40). TRX expression can be induced by many physicochemical stimuli and agents (11, 15), some of which, including calcium ionophore A23187, phorbol ester, cAMP, retinol, and androgen (6, 43), are also inducers of CRT expression. In the present study, we show that the TRX protein level was proportional to the CRT level in A549 cells, which indicates that the expression of TRX is closely related to that of CRT. Furthermore, we demonstrate that CRT regulated the expression of TRX at both protein and mRNA level. More importantly, TRX dominant-negative plasmid transfection blocked the beneficial effect of CRT. Thus we provide the first evidence of TRX as a downstream molecule of CRT in A549 type II alveolar epithelial cells. CRT is a multifunctional ER protein participating in many stress-related cellular processes through, for example, Ca2+ buffering and gene expression regulation. The present study uncovered the anti-oxidant pathway of CRT to be a novel mechanism mediating its cytoprotective role in pulmonary epithelial cells. Further studies are needed to verify whether these pathways of CRT are interrelated. For example, the buffering of Ca2+ (12, 20) may directly or indirectly contribute to the modulation of CRT with TRX expression.

A number of antioxidant genes, including TRX, have been shown to be under the control of Nrf2 through its interaction with antioxidant response element within the promoter region of these genes (13, 16). Studies have shown that Nrf2 plays a critical role protecting against oxidative stress in the liver and lung (3, 4). Recently, Nrf2 was found being activated by ischemia-reperfusion injury in epithelial cells (18), suggesting the potential role of Nrf2 against ischemia-reperfusion injury. In this study, we showed that overexpression of CRT induced Nrf2 nuclear translocation, thus indicating the involvement of Nrf2 in the cytoprotective effect of CRT. How CRT activates Nrf2 translocation and whether other Nrf2-regulating genes are also activated remain to be addressed in the future studies.

In summary, we provide convincing evidence supporting the ER protein CRT as an essential mediator responsible for the beneficial effects of short-term hypoxia in A549 type II alveolar epithelial cells. We also report, for the first time, that the antioxidant mechanism through TRX upregulation mediates the cytoprotective effect of CRT. With appropriate approaches, the present in vitro findings from A549 cell line can be judiciously extended to guide studies in normal pulmonary epithelial cells and in lung tissues in vivo. Furthermore, discovery of the short-term hypoxia-induced CRT-TRX-ROS pathway may provide new insight into the treatment of hypoxic lung diseases.


This research project was supported by the Major National Basic Research Program of People's Republic of China (No. 2006CB503802) and grants from the National Natural Science Foundation of China (No. 30330250) and the Program for Changjiang Scholars and Innovative Research Teams in Universities (PCSHT) awarded to X. Wang.


Present address of W. Wang: Dept. of Physiology, University of Michigan, Ann Arbor, MI 48109.


We thank Dr. M. Michalak, Dr. J. Yodoi, and Dr. Roger J. Davis for kind help in providing several plasmids used in this study.


  • * L. Jia and M. Xu contributed equally to this work.

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract