Abstract
Tunicamycin (TUN), an inhibitor of protein glycosylation and therefore a potent stimulator of endoplasmic reticulum (ER) stress, has been used to improve anticancer drug efficacy, but the underlying mechanism remains obscure. In this study, we show that acute administration of TUN in mice induces the unfolded protein response and suppresses the levels of P21, a cell cycle regulator with anti-apoptotic activity. The inhibition of P21 after ER stress appears to be C/EBP homologous protein (CHOP)-dependent because in CHOP-deficient mice, TUN not only failed to suppress, but rather induced the expression of P21. Results of promoter-activity reporter assays using human cancer cells and mouse fibroblasts indicated that the regulation of P21 by CHOP operates at the level of transcription and involves direct binding of CHOP transcription factor to the P21 promoter. The results of cell viability and clonogenic assays indicate that ER-stress-related suppression of P21 expression potentiates caspase activation and sensitizes cells to doxorubicin treatment, while administration of TUN to mice increases the therapeutic efficacy of anticancer therapy for HepG2 liver and A549 lung cancers.
Introduction
The unfolded protein response (UPR) is a conserved biochemical response that follows endoplasmic reticulum (ER) stress and aims to attain cellular homeostasis (Ron & Walter 2011). However, prolonged or severe ER stress induces ER-stress-associated apoptosis with the transcription factor C/EBP homologous protein (CHOP) playing an instrumental role (Zinszner et al. 1998, Oyadomari & Mori 2004, Puthalakath et al. 2007). CHOP belongs to the family of c/EBP transcription factors and can stimulate gene transcription directly or inhibit it by displacing other transcription factors of the c/EBP family with positive effects on gene expression (Zinszner et al. 1998, Oyadomari & Mori 2004, Puthalakath et al. 2007).
ER stress is involved in various pathologic conditions, while its modulation increasingly appears to have therapeutic value for several conditions including cancer (Farmaki et al. 2011, Firczuk et al. 2013, Ahmad et al. 2014, Liu et al. 2014). An important aspect of ER stress and subsequent UPR is the regulation of the balance between the prosurvival (adaptive) and pro-apoptotic mode of action. Consistent with this notion, during its earlier stages the UPR is pro-survival while at subsequent stages it becomes pro-apoptotic (Ron & Walter 2011). Results of previous studies at our laboratory indicated that an inverse correlation between the activation of the UPR and the expression of P21 exists (Mihailidou et al. 2010). P21 is a cell-cycle regulator with strong anti-apoptotic activity produced by both cell-autonomous and non-cell-autonomous mechanisms (Blagosklonny 2002, Roninson 2002, Weiss 2003, Trimis et al. 2008). Recently, the suppression of P21 expression by the P53 isoform P53/47 during ER stress has been demonstrated and has been linked to the reduction of the apoptotic threshold during anticancer therapy (Mlynarczyk & Fåhraeus 2014). In this study, we explored the regulation of P21 through the transcription factor CHOP at the level of transcription and evaluated how P21 activity affects the sensitivity of cells to the ER stress inducer tunicamycin (TUN), the conventional anticancer agent doxorubicin (DOX), or a combination of the two. Finally, we assessed whether TUN increases the anticancer efficacy of DOX in tumor-bearing mice.
Materials and methods
Cell culture and transfections
CHOP-deficient and P21-deficient mouse embryonic fibroblasts (MEFs) were generated from 12-day-old embryos by using standard procedures and maintained in DMEM containing 10% FBS and antibiotics/antimycotics. A549 human lung cancer cells and HepG2 (a human hepatocellular liver carcinoma cell line) cells were originally obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and subsequently maintained as described previously. These cell lines were grown at 37 °C in a humidified atmosphere of 5% CO2 95% air.
The cells were transfected using the Thermo Scientific TurboFect Transfection Reagent according to the manufacturer's instructions.
Survival clonogenic assay
For the colony formation assay, cells treated with DOX (Sigma) (0.2 μM) and TUN (Sigma) (5 μg/ml) or DMSO for 24 h. The cells were then collected and seeded into 100 mm culture plates for 2 weeks. The colonies obtained were fixed with methanol:acetic acid (3:1) and stained with hematoxylin (10% Mayer's hematoxylin). Each data point was generated from measurements performed in triplicate.
Caspase assay for apoptosis detection
Activity of caspases was determined using the Caspase Colorimetric Protease Assay Kit (Invitrogen). Briefly, 3×106 cells were used for each treatment and were lysed by the addition of 50 μl of lysis buffer and incubation on ice for 10 min. The supernant was removed by centrifugation at 14 000 g for 1 min: 200 μg of the supernatant was mixed with 50 μl of the 2× reaction buffer provided. Where indicated, 5 μl of a 4 mM solution of the appropriate substrate for caspase 2, 3, and 9 was added to each sample and preincubated with the indicated specimens at 37 °C for 1–2 h. The absorbance was measured at 405 nm microplate reader. Each data point was performed in triplicate.
RT-PCR analysis
Total RNA was isolated from A549 cells by using the TRIzol reagent (Invitrogen), according to the manufacturer's instructions. Semi-quantitative reverse transcriptase PCR (RT-PCR) was carried out using an RT-PCR kit (TaKaRa, Otsu, Japan) according to the manufacturer's protocol. The oligonucleotide primers used for PCR were: P21: 5′-ATGAAATTCACCCCCTTTCC-3′ and 5′-CCCTAGGCTGTGCTCACTTC-3′; actin: 5′-AGGCACCAGGGCGTGAT-3′ and 5′-GCCCACATAGGAATCCTTCTGAC-3′.
The cycling conditions were as follows: for human actin at 95 °C for 30 s, 55 °C at 30 s, and 72 °C at 30 s for 27 cycles and for human P21 at 95 °C for 30 s, 57 °C at 30 s, and 72 °C for 70 s for 28 cycles. The PCR products were electrophoresed into 2% agarose gel and visualized by ethidium bromide. Densitometric analysis was performed by using the ImageJ software (National Institutes of Health (NIH), Bethesda, MD, USA).
Western blot assays
The PBS-washed cells and tissues were lysated in RIPA buffer (Sigma) and proteins were separated following electrophoretic separation by SDS–PAGE, and transferred onto nitrocellulose membranes. The membranes were blocked with 5% milk in PBS-T (1× PBS, 0.1% Tween-20), and probed with primary antibodies (GADD153 (F-168): sc-575 Santa Cruz (1:50), p53 antibody (C-19): sc-1311 Santa Cruz (1:500), p21 (F-5) sc-6246 Santa Cruz (1:200), anti-actin, clone C4 MAB1501 Millipore (Billerica, MA, USA) (1:2000), BiP antibody (# 3183) Cell Signaling (Danvers, MA, USA) (1:1000), GRP 94 antibody (H-212): sc-11402 Santa Cruz (1:500)) overnight to detect specific proteins in whole-tissue lysates by western blotting analysis.
Construction of P21Cip1 promoter gene plasmids
The PCR primers pL1 and pL2 (deletion of the CHOP-binding site), containing KpnI restriction sites, and pR1, containing XhoI restriction sites, were used to PCR-amplify and clone the P21 promoter (positions −1298/−683) into the KpnI/XhoI polylinker restriction sites of pGL3 Basic (Promega) to produce pL1R1–pGL3 and PL2R1–pGL3 plasmids. The sequences of primers used for cloning of P21Cip1 promoter constructs were: p21 L1 (KpnI) 3′: 5′-GGGGTACCCCTTTCTTTGCTGCATGATCTGA-3′; p21 L2 (KpnI) 3′: 5′-GGGGTACCCCGCATGTGACAATCAACTT-3′; p21 R1 (XhoI) 5′: 3′-ACTTCGTGGGGAAATGTGTCCCGCTCGAGCGG-5′.
Luciferase activity assays
Promoter reporter assays were performed using the dual-luciferase reporter assay system Firefly and Renilla Luciferase Assay Kit (BiotiumCatalog Number: 30003-1). The promoter activity values were normalized by performing co-transfection with a plasmid expressing constitutively Renilla luciferase. The luciferase activity was measured by using a luminometer. The data shown are expressed in relative luciferase units (RLU) and are representative from at least three independent experiments done in triplicates, with s.d. between experiments shown in the form of error bars.
Chromatin immunoprecipitation assay
A549 cells (5×107) were seeded into culture dishes and cultured overnight. The cells were either left untreated or treated with TUN (5 μg/ml) for the indicated 0-, 6-, and 24-h time points. Briefly, the cells were crosslinked with 1% formaldehyde in culture media for 10 min at room temperature, followed by neutralization with 125 mM glycine for 5 min at room temperature. The cross-linked chromatin was extracted from the whole-cell lysates. The chromatin was then subjected to fragmentation using a Branson digital sonifier S450D. The sonicated chromatin was pre-cleared and DNA was analyzed by gel electrophoresis to ensure that 200–500-bp sized fragments were obtained before proceeding with chromatin immunoprecipitation (ChIP), with 5 μg of GADD153 (F-168): sc-575 Santa Cruz and p53 antibodies (C-19): sc-1311 Santa Cruz, which were incubated with Protein A/G agarose beads (Santa Cruz sc-2003) overnight to allow antibody coating to the beads. ChIP-enriched DNA was analyzed by PCR using primers corresponding to the human P21Cip1 promoter. The PCR products were analysed by agarose gel electrophoresis. The PCR primer sets used for the ChIP assay were as follows: chop forward 1: 5′-TACGGGCTATGTGGGGAGTA-3′; chop reverse 1: 3′-GGAAGGAGGGAATTGGAGAG-5′ (206 bp); chop forward 2: 5′-CTGCAACCACAGGGATTTCT-3′; chop reverse 2: 3′-AGAAGCACCTGGAGCACCTA-5′ (271 bp); P53 forward 1: 5′-CCCACAGCAGAGGAGAAAGA-3′; P53 reverse 1: 3′-TCTGTGCCTGAAACATTTGC-5′ (227 bp).
Animal studies
CHOP-KO and P53-KO mice have been described previously (Jacks et al. 1994, Zinszner et al. 1998). SCID mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). One group of 5- to 6-week old female mice received injections of 2.5×106 A549 cells in 100 μl serum-free DMEM and a second group of 5- to 6-week-old female mice received injections of 2.5×106 HepG2 cells in 100 μl serum-free DMEM. The animals were inspected for development of tumors externally, using vernier calipers. The tumor volumes were calculated with the formula tumor volume=(d)2×D/2, where d is width and D is the length of a tumor. Mice were ranked in closely matched groups according to their tumor volume. Treatment with drugs started when the average tumor volume per mouse was between 80 and 100 mm3. Then each group of treated mice was divided into four sub-groups. Mice from each sub-group received one i.p. injection of DOX (Sigma) (4 mg/kg) or one i.p. injection of TUN (Sigma) (1.5 mg/kg), carrier alone or a combination of DOX and TUN and tumor volumes was scored. For survival analysis, death events were scored when mice died, when they became moribund and had to be killed, or when they had to be killed because their tumor exceeded 10% of body weight.
Statistical analysis
All data were analyzed by Student's t-test and the results were considered to be significant if P<0.05. Statistical analysis for survival studies was performed using the log-rank test to determine significance and differences were considered to be significant when P<0.05.
Results
p21 expression is regulated during ER stress in vivo and in vitro
Initially, we investigated whether ER stress regulates P21 levels in vivo. To that end, WT mice received i.p. injections of the inhibitor of protein glycosylation TUN, which induces ER stress, and the levels of p21 were assessed. As shown in Fig. 1, TUN stimulated CHOP and reduced the levels of P21 protein in various organs such as the lungs, heart, liver, spleen, and pancreas, indicating that the induction of ER stress is accompanied by suppression of P21 expression. Under these conditions, the levels of BiP and GRP94 chaperones were induced, indicating the stimulation of the UPR (Fig. 1). We further investigated whether the suppression of P21 in vivo was CHOP-dependent, considering its role in ER-stress-associated death, in association with the anti-apoptotic activity of P21. To that end, we treated CHOP-deficient animals with TUN and P21 protein levels were evaluated. As shown in Fig. 1, not only was the reduction in P21 was alleviated, but also stimulation of P21 protein levels in the CHOP-deficient mice following administration of TUN was recorded. In tissues obtained from unstressed mice, CHOP ablation caused suppression in the baseline p21 levels (Fig. 1). We also examined the levels of P21 expression in tissues from P53-deficient mice as P53 is a major activator of P21 expression under various stress-inducing stimuli (el-Deiry et al. 1993). As shown in Fig. 1, P53 ablation caused an overall baseline reduction in the expression of p21, but the TUN-induced suppression of p21 levels was retained.
Regulation of P21 expression during ER stress by CHOP in vivo. P21 protein expression in organs that had been harvested from WT, CHOP-deficient, and P53-deficient mice following tunicamycin administration (i.p. 1.5 mg/kg) at the indicated time points. In wt mice tunicamycin suppresses while in CHOP-deficient mice stimulates P21 expression. P53 deficiency does not compromise the inhibitory effects of tunicamycin on P21 in vivo.
Citation: Endocrine-Related Cancer 22, 2; 10.1530/ERC-15-0019
Next we investigated whether CHOP expression in cells in culture regulates the expression of P21. Indeed, plasmid-mediated expression of CHOP in A549 lung cancer cells reduced P21 RNA and protein levels (Fig. 2A and Supplementary Fig. S1, see section on supplementary data given at the end of this article). In this experiment, cells had been exposed to DOX in order to elevate baseline P21 expression by a P53-dependent mechanism (el-Deiry et al. 1993), and thus permit recording of P21 suppression. Notably, some moderate induction of P21 protein levels but not RNA levels was recorded when smaller amounts of CHOP cDNA were transfected, which may reflect the post-transcriptional regulation of P21 expression (Fig. 2A).
Regulation of P21 expression during ER stress by CHOP in vitro. (A) RT-PCR (upper panel) and western blotting (lower panel) analysis of P21 expression in A549 lung cancer cells following plasmid-mediated expression of CHOP. Doxorubicin (DOX) was added to the media at 0.2 μM in to stimulate basal P21 expression. CHOP expression suppresses P21, notwithstanding that at lower amounts it causes stimulation of P21 protein levels. (B) Diagrammatic representation of the human P21 promoter and of the reporter plasmids L1R1 and L2R1. The relative location of a TCCAAT sequence located at position −683 bp upstream of the first exon that can be recognized by C/EBP family members is shown. A similar GCCAAT element is also present in the first intron of the mouse P21 gene. The location of the P53 binding site (−1298 bp) is also indicated. Double arrows indicate the PCR product amplified during the ChIP analysis illustrated in Fig. 2F. (C) Luciferase activity expressed in RLU of the L1R1 reporter construct in A549 lung cancer cells, WT, and CHOP-deficient (CHOPKO) MEFs cultured for 16 h with tunicamycin at the concentrations indicated. This experiment was performed at least three times in triplicates and the results of a single experiment expressed as average values±s.e.m. are shown. *P<0.05 and **P<0.01. (D) Protein extracts of the experiment described in (C) from the three triplicates pooled together were subjected to immunoblot analysis for CHOP. Actin levels were assessed as a loading control. (E) Luciferase activity expressed in RLU for the L1R1 (which includes the CHOP binding site) and the truncated L2R1 (not including the CHOP binding site) plasmids following transfection of 1 μg CHOP cDNA (left panel) or tunicamycin (5 μg/ml for 24 h) exposure (right panel) are shown for wt MEFs, CHOP-deficient MEFs, and A549 cells. Results are expressed as values normalized to Renilla luciferase activity. Experiments were performed in triplicates and error bars represent s.d. *P<0.05 and **P<0.005 (Student's t-test). (F) Results of ChIP analysis indicating that CHOP and P53 physically interact with the P21 promoter in A549 cells during ER stress. For detecting binding of CHOP, two pairs of oligonucleotides were used, while P53 was also amplified as a positive control. In the presence of tunicamycin (5 μg/ml at the indicated times), CHOP and P53 binding are increased. The relative location of the corresponding amplified fragments are indicated in (B).
Citation: Endocrine-Related Cancer 22, 2; 10.1530/ERC-15-0019
A similar trend in the expression of P21 levels had been revealed by analyzing previously published microarray data (Marciniak et al. 2004) that evaluated the consequences of 4- and 8-h treatment of primary MEFs with TUN (Supplementary Fig. 2S, see section on supplementary data given at the end of this article). In agreement with our findings, a moderate reduction in P21 levels in wt but not in CHOPKO MEFs could be seen following treatment of cells with 2 μg/ml TUN for 4 and 8 h (Marciniak et al. 2004).
P21 expression during ER stress is CHOP-dependent
In order to better understand the mechanism of CHOP-dependent regulation of P21 during ER stress, we investigated whether P21 represents a direct transcriptional target of CHOP. Thus, an approximately 700 bp fragment of the P21 promoter upstream of the gene's first exon was cloned and tested for its ability to regulate luciferase expression during ER stress (Fig. 2B). As P53 is the major transcriptional regulator of P21 and considering that P53 is differentially regulated during ER stress (Qu et al. 2004, Zhang et al. 2006), it is conceivable that P53 activity may interfere with the execution of UPR and the activation of P21 promoter transcription. Therefore, in order to eliminate the contribution of P53 to the regulation of p21, we utilized the promoter fragment that did not contain the P53 consensus binding site (Fig. 2B). As shown in Fig. 2C, TUN added at a dose of 0.1 μg/ml to the culture media of A549 lung cancer cells and MEFs moderately stimulated promoter activity. Increasing concentrations of tunicamycin, however, caused a dose-dependent reduction that was maximal at 5 μg/ml in A549 lung cancer cells and WT MEFs (Fig. 2C). In fibroblasts isolated from CHOP-deficient animals, this pattern was abolished and P21 promoter activity was consistently lower than that of controls, while TUN exposure produced only minimal effects (Fig. 2C). Under these conditions, in WT MEFs, CHOP levels were only slightly elevated by lower amounts of TUN but were maximized when TUN was added to the media at 2 and 5 μg/ml, (Fig. 2D). In A549 cells, the sensitivity to TUN, as reflected by the induction of CHOP levels, was higher as compared with that of the other cells used, reaching maximal stimulation at 0.5 μg/ml (Fig. 2D).
Subsequently, we focused on the regulation of P21 promoter by CHOP during intense ER stress induced by the relatively high dose of 5 μg/ml TUN. As shown in Fig. 2E, both TUN (right panel) and plasmid-mediated CHOP expression (left panel) significantly suppressed luciferase activity, indicating that transcriptional regulation contributes to the suppression of P21 by CHOP in MEFs and in A549 cells. Surprisingly, following TUN treatment, CHOP-deficient cells also exhibited reductions in P21 promoter activity (Fig. 2E, right panel) that were less pronounced than the ones observed in the protein levels of wt MEFs, but were still detectable. A truncated version of the P21 promoter that did not contain a CCAAT sequence considerably reduced the baseline activity of the reporter construct and abolished the suppressive effects of both CHOP transcription factor and TUN in MEFs, A549, and CHOP-deficient cells (Fig. 2E).
CHOP interacts physically with the P21 promoter during ER stress
Next, we investigated whether CHOP protein co-localizes with the P21 promoter. Thus, we performed chromatin IP (ChIP) analysis in A549 cells by using an antibody for CHOP that immunoprecipitates the chromatin–CHOP complex. As shown in Fig. 2F, in cells were exposed to TUN, CHOP was present in the protein complex that interacted with the P21 promoter. As a control for the ChIP assay, we performed immunoprecipitation analysis by using an antibody specific for P53. P21 is an established target of P53 and therefore its promoter is expected to interact physically with P53. Indeed, P53 was found to interact physically with the P21 promoter and, interestingly, this interaction was more pronounced in cells undergoing ER stress (Fig. 2F).
P21 and CHOP regulate the clonogenic survival of cells in response to TUN alone or in combination with DOX
After establishing that during ER stress P21 levels decrease by a mechanism that is CHOP-dependent, we performed a series of clonogenic survival assays to assess the sensitivity of wt, P21-deficient, and CHOP-deficient MEFs to TUN alone or in combination with DOX. Consistently with the anti-apoptotic activity of P21, the survival of p21KO MEFs following DOX exposure was compromised as compared with that of controls while CHOP ablation conferred resistance (Fig. 3A). Analogous and even more pronounced was the sensitivity of P21KO MEFs to TUN, while CHOP ablation restored their sensitivity, which is consistent with the established pro-apoptotic role of CHOP during ER stress (Fig. 3A). Combined exposure to both TUN and DOX further reduced the survival of wt MEFs and completely abolished survival in P21KO MEFs indicating that there was synergistic activity between TUN and DOX (Fig. 3A).
P21 regulates the survival of cells during ER stress. (A) Clonogenic survival of WT, P21KO, and CHOPKO MEFs following exposure to tunicamycin (5 μg/ml) and DOX (0.2 μM) singly or in combination. Representative petri dishes are shown in the left panel. Petri dishes are presented according to the diagram shown in the middle panel. Results of quantification of relative colony numbers, expressed as percentages of the value for controls, are shown in the right panel. Average values of three replicates ±s.d. are shown. *P<0.05 (Student's t-test). (B) Western blotting analysis for P21 in A549 (upper panel) and HepG2 (lower panel) cells after treatment with tunicamycin (5 μg/ml), DOX (0.2 μM), or a combination of the two. Actin was used as a loading control. (C) Clonogenic survival of A549 and HepG2 cells exposed to tunicamycin (5 μg/ml), DOX (0.2 μM), or a combination of the two. Representative petri dishes are shown in the left panel. Petri dishes are presented according to the diagram shown in (A). Results of quantification of relative colony numbers, expressed as percentages of the value for controls, are shown in the right panel. Average values for three replicates ±s.d. are shown. *P<0.05 (Student's t-test). (D) Relative activities of caspases 2, 3, and 9 after treatment of wt, P21KO MEFs, A549 lung and HepG2 liver cancer cells with tunicamycin (5 μg/ml), DOX (0.2 μM), or a combination of the two. Average values of three replicate experiments ±s.d. are shown. *P<0.05 versus controls; #P<0.05 versus DOX (Student's t-test).
Citation: Endocrine-Related Cancer 22, 2; 10.1530/ERC-15-0019
We next investigated whether TUN potentiate the cytotoxic activity of DOX in cancer cells, such as A549 lung and HepG2 liver cells. Initially, we confirmed that TUN can reduce the levels of P21 in the presence of DOX (Fig. 3B). Subsequently, we performed clonogenic survival assays that confirmed that the cytotoxicity of DOX is more potent in the presence of TUN (Fig. 3C).
Caspase activation by TUN, DOX, or their combination
In order to test whether different treatments activate distinct caspases preferentially, wt, and P21-deficient, MEFs and A549 lung and HepG2 liver cancer cells were exposed to TUN and DOX alone or in combination, and the activity of caspases 2, 3, and 9 was assessed. As shown in Fig. 3D, with the exception of caspase 2 for wt MEFs, a trend towards an increase in the activity of all three caspases was detected in all cells and for all treatment combinations. In P21KO MEFs and in A549 and HepG2 cancer cells, the activation was consistently more potent for the combined treatment for caspase 3 (Fig. 3D). Furthermore, P21KO MEFs had acquired sensitivity for caspase 2 (Fig. 3D).
TUN potentiates the anticancer activity of DOX in mice
Finally, we investigated whether administration of TUN increases the anticancer activity of DOX in tumor-bearing animals. To determine this, SCID mice received inoculations of A549 or HepG2 cancer cells and were then treated with DOX alone TUN alone, or DOX or in combination with TUN. Both DOX and TUN treatment were well tolerated by mice. As shown in Fig. 4, all treatments significantly inhibited tumor growth in both tumor models, with combination therapy showing increased efficacy compared with either DOX or TUN when administered alone. Furthermore, treatment significantly prolonged the survival of the tumor-bearing animals, with combination treatment essentially preventing the death of animals for the duration of the experiment (Fig. 4).
Tunicamycin improves the efficacy of DOX in A549 lung and HepG2 liver cancers. Tumor volume ((A) and (B)) and survival ((C) and (D)) of A549 ((A) and (C)) and HepG2 ((B) and (D)) bearing mice that received treatment with DOX (4 mg/kg), tunicamycin (1.5 mg/kg), or a combination of the two. All treatments caused significant inhibition of tumor growth (*P<0.05 Student's t-test) after about 18 days from the initiation of therapy for A549 and 14 days for HepG2. Combination therapy was significantly more potent and the difference reached significance compared with DOX treatment alone after 35 days for A549 and 32 days for HepG2 (#P<0.05 Student's t-test, (A) and (B)). For the survival analysis: *P<0.05 versus control; #P<0.05 versus TUN+DOX; log-rank test, (C) and (D)).
Citation: Endocrine-Related Cancer 22, 2; 10.1530/ERC-15-0019
Discussion
In this study, we showed that during ER stress P21 levels decrease by means of a mechanism that depends on the CHOP transcription factor that is activated during the advanced stages of UPR. The results of promoter-activity reporter assays and ChIP studies indicated that this regulation is direct and operates at the level of transcription and may involve competition with other members of the C/EBP family of transcription factors that also regulate P21 in a positive manner (Chinery et al. 1997, Cram et al. 1998; Mihailidou et al. 2015).
The regulation of P21 during ER stress appears to be a rather complex process. In the absence of CHOP, when ER stress is not induced, the baseline levels of P21 decreased, which implies a positive role rather than a negative role of CHOP in P21 transcription. Although CHOP protein remained only marginally detectable in the absence of TUN treatment, it is conceivable that low levels of CHOP, close to or below the detection limits of our assay, were sufficient to modulate P21 expression. Indeed, a positive role of CHOP in the regulation of gene transcription has been reported previously (Ubeda et al. 1996). Furthermore, during ER stress and in the absence of CHOP, the levels of P21 are increased, indicating that ER stress elicits both positive and negative signals toward P21, of which only the latter involve CHOP activity.
P53, a major regulator of P21 expression in response to various stress-inducing stimuli, appears unrelated to CHOP-induced regulation of P21, because the promoter reporter constructs utilized did not contain P53 binding sites. Furthermore, the results of our experiments carried out on tissues from P53-deficient animals indicate that the inhibitory effect of ER stress on P21 was retained. However, increased binding of P53 to the P21 promoter was revealed by the ChIP analysis, which in turn implies that positive regulation of P21 by P53 during ER stress is likely. Specifically, in mouse tissues that are deficient in CHOP, the levels of P21 increased, which may reflect the activity of P53 or of other positive regulators of P21 expression. In fibroblasts from CHOP-deficient mice, P21 promoter activity decreased during TUN exposure, indicating the operation of CHOP-independent mechanisms in the regulation of p21 expression. Why this negative effect of ER stress was not seen in tissues of CHOP-deficient mice remains unclear, but this may indicate tissue-specific effects on the regulation of P21 expression levels.
The suppression of P21 by ER stress is related to the regulation of cells' survival during ER stress because P21 possesses anti-apoptotic activity. Considering that during earlier stages ER stress is adaptive and pro-survival while during the most advanced stages it is pro-apoptotic, the suppression of P21 by CHOP may facilitate this transition to ER-stress-associated death. Indeed, P21-deficient MEFs had increased sensitivity to TUN, which is consistent with earlier results indicating that P53-deficient cells and animals are more sensitive to the induction of ER stress (Dioufa et al. 2012). Thus, when the protective activity of P21 is abolished, which may happen when P53 expression is lost, ER-stress-associated death may proceed more efficiently. Furthermore, increased sensitivity of P53-deficient cells to DNA-demethylating agents has been reported, which may also implicate ER-stress induction (Leonova et al. 2013).
The reduction in P21 expression levels as a result of ER stress sensitizes cells to the cytotoxic drug DOX because the antiapoptotic activity of P21 is eliminated. This protective action of P21 against ER-stress-associated toxicity is also reflected by the levels of caspase activation, which are increased in the P21-deficient MEFs as compared with their wt counterparts, and is also potentiated by the concomitant exposure of cells to DOX and TUN. In addition, increased sensitivity to caspase 2 activation has been seen in the P21KO MEFs that is in agreement with earlier findings indicating the inhibition of P21 caspase 2 expression (Baptiste-Okoh et al. 2008).
These results, in the context of anti-cancer therapy, indicate that TUN would sensitize cells to the therapeutic effects of DOX. Indeed, treatment of A549 lung and HepG2 liver cancer-bearing mice with TUN strongly inhibited tumor growth and prolonged the survival of the experimental mice in association with DOX, which confirms and extends earlier results revealing the chemosensitizing activity of ER-stress induction (Farmaki et al. 2011, Firczuk et al. 2013, Ahmad et al. 2014, Liu et al. 2014).
Collectively, our results indicate that P21 is suppressed during ER stress by CHOP, facilitating the transition of the UPR from its adaptive to its pro-apoptotic mode of action. Interfering with this regulation may have important implications for the management of conditions for which manipulation of ER-stress-associated death is beneficial.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/ERC-15-0019.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
C Mihailidou has been supported by a Fellowship that is co-financed by the European Union (European Social Fund – ESF) and Greek National funds through the Operational Program ‘Education and Lifelong Learning’ of the National Strategic Reference Framework (NSRF) – Research Funding Program: Heracleitus II. This study was supported by a grant from the Empirikion Foundation and EXCELLENCE-2011 (ARISTEIA) Grant (11EXC311), funded by the European Social Fund-European Union and National Resources.
Author contribution statement
C Mihailidou designed and perform experiments, analyzed results, wrote manuscript; I Chatzistamou designed experiments, analyzed results, and wrote manuscript; A G Papavassiliou designed experiments, analyzed results, wrote manuscript; H Kiaris designed and perform experiments, analyzed results, wrote manuscript.
References
Ahmad M, Hahn IF & Chatterjee S 2014 GRP78 up-regulation leads to hypersensitization to cisplatin in A549 lung cancer cells. Anticancer Research 34 3493–3500.
Baptiste-Okoh N, Barsotti AM & Prives C 2008 Caspase 2 is both required for p53-mediated apoptosis and downregulated by p53 in a p21-dependent manner. Cell Cycle 7 1133–1138. (doi:10.4161/cc.7.9.5805).
Blagosklonny MV 2002 Are p27 and p21 cytoplasmic oncoproteins? Cell Cycle 1 391–393. (doi:10.4161/cc.1.6.262).
Chinery R, Brockman JA, Peeler MO, Shyr Y, Beauchamp RD & Coffey RJ 1997 Antioxidants enhance the cytotoxicity of chemotherapeutic agents in colorectal cancer: a p53-independent induction of p21WAF1/CIP1 via C/EBPβ. Nature Medicine 3 1233–1241. (doi:10.1038/nm1197-1233).
Cram EJ, Ramos RA, Wang EC, Cha HH, Nishio Y & Firestone GL 1998 Role of the CCAAT/enhancer binding protein-α transcription factor in the glucocorticoid stimulation of p21waf1/cip1 gene promoter activity in growth-arrested rat hepatoma cells. Journal of Biological Chemistry 273 2008–2014. (doi:10.1074/jbc.273.4.2008).
el-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW & Vogelstein B 1993 WAF1, a potential mediator of p53 tumor suppression. Cell 75 817–825. (doi:10.1016/0092-8674(93)90500-P).
Dioufa N, Chatzistamou I, Farmaki E, Papavassiliou AG & Kiaris H 2012 p53 antagonizes the unfolded protein response and inhibits ground glass hepatocyte development during endoplasmic reticulum stress. Experimental Biology and Medicine 237 1173–1180. (doi:10.1258/ebm.2012.012140).
Farmaki E, Mkrtchian S, Papazian I, Papavassiliou AG & Kiaris H 2011 ERp29 regulates response to doxorubicin by a PERK-mediated mechanism. Biochimica et Biophysica Acta 1813 1165–1171. (doi:10.1016/j.bbamcr.2011.03.003).
Firczuk M, Gabrysiak M, Barankiewicz J, Domagala A, Nowis D, Kujawa M, Jankowska-Steifer E, Wachowska M, Glodkowska-Mrowka E & Korsak B et al. 2013 GRP78-targeting subtilase cytotoxin sensitizes cancer cells to photodynamic therapy. Cell Death & Disease 4 e741. (doi:10.1038/cddis.2013.265).
Jacks T, Remington L, Williams BO, Schmitt EM, Halachmi S, Bronson RT & Weinberg RA 1994 Tumor spectrum analysis in p53-mutant mice. Current Biology 4 41–47. (doi:10.1016/S0960-9822(00)00002-6).
Leonova KI, Brodsky L, Lipchick B, Pal M, Novototskaya L, Chenchik AA, Sen GC, Komarova EA & Gudkov AV 2013 p53 cooperates with DNA methylation and a suicidal interferon response to maintain epigenetic silencing of repeats and noncoding RNAs. PNAS 110 89–98. (doi:10.1073/pnas.1216922110).
Liu N, Huang H, Xu L, Hua X, Li X, Liu S, Yang C, Zhao C, Zhao C & Li S et al. 2014 The combination of proteasome inhibitors bortezomib and gambogic acid triggers synergistic cytotoxicity in vitro but not in vivo. Toxicology Letters 224 333–340. (doi:10.1016/j.toxlet.2013-11.021).
Marciniak SJ, Yun CY, Oyadomari S, Novoa I, Zhang Y, Jungreis R, Nagata K, Harding HP & Ron D 2004 CHOP induces death by promoting protein synthesis and oxidation in the stressed endoplasmic reticulum. Genes and Development 18 3066–3077.
Mihailidou C, Papazian I, Papavassiliou AG & Kiaris H 2010 CHOP-dependent regulation of p21/waf1 during ER stress. Cellular Physiology and Biochemistry 25 761–766. (doi:10.1159/000315096).
Mihailidou C, Chatzistamou I, Papavassilou AG & Kiaris H Regulation of P21 during diabetes-associated stress of the endoplasmic reticulum Endocrine-Related Cancer 2015 (In Press) doi:10.1530/ERC-15-0018).
Mlynarczyk C & Fåhraeus R 2014 Endoplasmic reticulum stress sensitizes cells to DNA damage-induced apoptosis through p53-dependent suppression of p21CDKN1A. Nature Communications 5 5067. (doi:10.1038/ncomms6067).
Oyadomari S & Mori M 2004 Roles of CHOP/GADD153 in endoplasmic reticulum stress. Cell Death and Differentiation 11 381–389. (doi:10.1038/sj.cdd.4401373).
Puthalakath H, O'Reilly LA, Gunn P, Lee L, Kelly PN, Huntington ND, Hughes PD, Michalak EM, McKimm-Breschkin J & Motoyama N et al. 2007 ER stress triggers apoptosis by activating BH3-only protein Bim. Cell 129 1337–1349. (doi:10.1016/j.cell.2007.04.027).
Qu L, Huang S, Baltzis D, Rivas-Estilla A-M, Pluquet O, Hatzoglou M, Koumenis C, Taya Y, Yoshimura A & Koromilas AE 2004 Endoplasmic reticulum stress induces p53 cytoplasmic localization and prevents p53-dependent apoptosis by a pathway involving glycogen synthase kinase-3β. Genes and Development 18 261–277. (doi:10.1101/gad.1165804).
Ron D & Walter P 2011 The unfolded protein response: from stress pathway to homeostatic regulation. Science 334 1081–1086. (doi:10.1126/science.1209038).
Roninson IB 2002 Oncogenic functions of tumour suppressor p21Waf1/Cip1/Sdi1: association with cell senescence and tumour-promoting activities of stromal fibroblasts. Cancer Letters 1 1–14. (doi:10.1016/S0304-3835(01)00847-3).
Trimis G, Chatzistamou I, Politi K, Kiaris H & Papavassiliou AG 2008 Expression of p21waf1/Cip1 in stromal fibroblasts of primary breast tumors. Human Molecular Genetics 17 3596–3600. (doi:10.1093/hmg/ddn252).
Ubeda M, Wang XZ, Zinszner H, Wu I, Habener JF & Ron D 1996 Stress-induced binding of the transcriptional factor CHOP to a novel DNA control element. Molecular and Cellular Biology 16 1479–1489.
Weiss RH 2003 p21Waf1/Cip1 as a therapeutic target in breast and other cancers. Cancer Cell 6 425–429. (doi:10.1016/S1535-6108(03)00308-8).
Zhang F, Hamanaka RB, Bobrovnikova-Marjon E, Gordan JD, Dai M-S, Lu H, Simon MC & Diehl JA 2006 Ribosomal stress couples the unfolded protein response to p53-dependent cell cycle arrest. Journal of Biological Chemistry 281 30036–30045. (doi:10.1074/jbc.M604674200).
Zinszner H, Kuroda M, Wang X, Batchvarova N, Lightfoot RT, Remotti H, Stevens JL & Ron D 1998 CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes and Development 12 982–995. (doi:10.1101/gad.12.7.982).
