We have reported previously that apoptosis of intestinal epithelial Caco-2 cells is induced by co-culturing with human macrophage-like THP-1 cells, mainly via the action of TNFα (tumour necrosis factor α) secreted from THP-1 cells [Satsu, Ishimoto, Nakano, Mochizuki, Iwanaga and Shimizu (2006) Exp. Cell Res. 312, 3909–3919]. Our recent DNA microarray analysis of co-cultured Caco-2 cells showed that IEX-1 (immediate early-response gene X-1) is the most significantly increased gene during co-culture [Ishimoto, Nakai, Satsu, Totsuka and Shimizu (2010) Biosci. Biotechnol. Biochem. 74, 437–439]. Hence, we investigated the role of IEX-1 in the co-culture-induced damage of Caco-2 cells. We showed that IEX-1 expression induced in Caco-2 cells was suppressed by anti-TNFα antibody treatment. Experiments using IEX-1-overexpressing and -knockdown Caco-2 cells suggested that IEX-1 was involved in the suppression of Caco-2 cell damage. Increases in caspase 3 activity and TNFR1 (TNF receptor 1) mRNA expression were shown in IEX-1-knockdown Caco-2 cells, suggesting that IEX-1 plays a role in the suppression of apoptosis and protects cells by controlling sensitivity to TNFα under both normal and inflammatory conditions.
IBD (inflammatory bowel disease), of which Crohn's disease and ulcerative colitis are the most well-known forms, is a debilitating illness of the bowel characterized by chronic inflammation of unknown aetiology. The number of patients with IBD has recently increased, but a therapeutic method has not yet been established since the disease is induced by many complex factors .
Several models of experimental colitis, such as those induced by DSS (dextran sodium sulfate) and 2,4,6-trinitrobenzenesulfonic acid and transgenic mice in which certain cytokines are knocked out, have been used to study intestinal inflammation [2–6]. Using these models, we can dissect the pathogenic components during different stages of colitis and identify some pivotal immunological processes as well as novel genes that are intimately involved in disease susceptibility. However, detailed studies on molecular or cellular mechanisms for suppressing inflammation have seldom been reported.
Activation of immune cells in the intestines of patients with IBD and its influences on intestinal immune responses of dendritic cells, T-cells and mast cells have been reported [7–9]. Macrophages are also activated in inflamed regions of IBD, and their role in IBD has been revealed using animal models. Previous reports, using the DSS-induced colitis model  and IL-10 (interleukin-10)-deficient mice , indicated the importance of the relationship between intestinal inflammation and macrophages, but little has been elucidated in terms of the role of macrophages in inflammation and their molecular interactions with epithelial cells.
We previously constructed an in vitro inflammation model using a co-culture system of macrophages and intestinal epithelial cells to analyse the effect of the former on the latter . We showed that co-culturing with macrophage-like THP-1 cells induced apoptotic and necrotic cell death in intestinal epithelial Caco-2 cells, and this was primarily due to TNFα (tumour necrosis factor α) secreted by THP-1 cells. Recently, using DNA microarray analysis, we identified IEX-1 (immediate early-response gene X-1) as a differentially expressed gene in Caco-2 cells at an early stage of co-culture .
IEX-1 is an immediate early-response gene and is induced by stimuli such as irradiation, growth factors, LPS (lipopolysaccharide) and inflammatory cytokines, including TNFα and IL-1β . Several reports have indicated that IEX-1 is involved in the regulation of apoptosis, although contradictory results on the pro-apoptotic and anti-apoptotic properties of IEX-1 have also been reported [15–18]. Feldmann et al.  reported that IEX-1 is expressed in human tissues such as breast, liver and lymph nodes, but the expression and function of IEX-1 in intestinal epithelial cells have not been reported.
In the present study, we focused on the role of IEX-1 in inflammatory cell death occurring in the intestinal epithelium using human intestinal Caco-2 cells co-cultured with THP-1 cells.
MATERIALS AND METHODS
The Caco-2 cell line was obtained from the American Type Culture Collection and THP-1 cells were purchased from the Health Science Research Resources Bank (Osaka, Japan). DMEM (Dulbecco's modified Eagle's medium) was purchased from Nissui Pharmaceuticals and FBS (fetal bovine serum) was from Sigma. Non-essential amino acids were purchased from Cosmo Bio. Recombinant human TNFα was purchased from Peprotech. Monoclonal anti-human TNFα and biotinylated polyclonal antibodies against human TNFα were purchased from Genzyme Techne. Forskolin and hexadimethrine bromide were purchased from Sigma. Anti-IEX-1 antibody (N17, sc-8453) and anti-goat IgG horseradish peroxidase (sc-2033) were purchased from Santa Cruz Biotechnology. All other chemicals used were of reagent grade.
Cell culture (co-culture)
Human intestinal epithelial Caco-2 cell monolayers and macrophage-like THP-1 cells were used for the co-culture system, as described previously . Briefly, Caco-2 cells were monolayer-cultured for 14 days on a semi-permeable membrane support, and THP-1 cells were differentiated with PMA for 4 days in 12-well plates. Caco-2 cell monolayers were then placed in the 12-well plates in which THP-1 cells had been cultured.
To generate an expression lentiviral plasmid for human IEX-1 [NCBI (National Center for Biotechnology Information) accession number: NM_003897], full-length human IEX-1 was amplified by RT–PCR (reverse transcription–PCR) using total RNA from Caco-2 cells. The primers used for human IEX-1 were as follows: hIEX-1-forward, 5′-GGGGAATTCGCGGCCGCACCATGTGTGTCACTCTCGCAGC-3′ and hIEX-1-reverse, 5′-GGGAAGCTTGGATCCTTAGAAGGCGGCCGGGTGTTG-3′. PCR products were subcloned in the pGEM®-T Easy Vector (Promega) before cloning into the NotI site of CS II-EF-MCS-IRES-Venus (RIKEN, Tsukuba, Japan).
An expression lentiviral plasmid for shRNA (short-hairpin RNA) of human IEX-1 was constructed by recombining CS-RfA-EG (RIKEN, Tsukuba, Japan) with pENTE4-H1 (RIKEN), inserted by oligonucleotide DNA for shRNA expression. The target sequence was as follows: IEX-1, 5′-GCCTAGTATGTTCTGTGAACA-3′. Expression lentiviral plasmid for control shRNA (inserted as follows: 5′-GCGCGCTTTGTAGGATTCG-3′, Scramble II Duplex from Dharmacon) was kindly provided by Dr R. Sato (Department of Applied Biological Chemistry, University of Tokyo, Tokyo, Japan).
HEK-293T cells [HEK (human embryonic kidney)-293 cells cells expressing the large T-antigen of SV40 (simian virus 40)] were transfected with an expression lentiviral plasmid together with a packaging (pCAG-HIVgp) and a VSV-G-/Rev-expressing (pCMV-VSV-G-RSV-Rev) plasmid using the Chen–Okayama method . At 12 h post-transfection, the medium was replaced with a medium supplemented with 10 mM forskolin. After another 24 h, the medium containing lentivirus was collected and filtered. Caco-2 cells were infected with this medium supplemented with 100 μg/ml hexadimethrine bromide by centrifugation (1100 g, 90 min). The cells were then incubated in a fresh culture medium [DMEM with 10% (v/v) FBS, 1% non-essential amino acids, 2% glutamine, 100 units/ml penicillin, 100 μg/ml streptomycin and an appropriate amount of sodium bicarbonate]. Overexpression and knockdown of IEX-1 were confirmed by Western blotting using anti-IEX-1 antibody and quantitative real-time RT–PCR respectively.
Quantitative real-time RT–PCR
Total RNA in Caco-2 and THP-1 cells was extracted using Isogen (Nippon Gene, Tokyo, Japan) according to the manufacturer's instructions. RT of RNA was performed using the ExScript® RT reagent kit (TaKaRa, Shiga, Japan) according to the manufacturer's instructions. A quantitative real-time RT–PCR analysis was performed using a LightCycler (Roche Diagnostics). One-twentieth of the first-strand cDNA was used as a template for detecting each gene of interest by the SYBR Green PCR assay (TaKaRa, Shiga, Japan). Sequences of primers and annealing temperatures were as follows: human IEX-1, NCBI accession number: NM_003897, 5′-CGAGCCCTTTAATCTGACTTT-3′ and 5′-GTACGCCTGGTGTTTCTTT-3′, 58°C; human TNFR1 (TNF receptor 1), NCBI accession number: NM_001065, 5′-GAATGTTAAGGGCACTGAG-3′ and 5′-CACAAACAATGGAGTAGAGC-3′, 60°C; human GAPDH (glyceraldehyde-3-phosphate dehydrogenase), NCBI accession number: NM_002046, 5′-TTCAACAGCGACACCCACTG-3′ and 5′-CACCCTGTTGCTGTAGCCA-3′, 56°C. Gene expression levels were normalized to that of the housekeeping gene (GAPDH). Each relative gene expression change was calculated as a number-fold change compared with the control sample.
Determination of damage to Caco-2 cell monolayers
The viability of Caco-2 cells was mainly assessed by their ability to maintain LDH (lactate dehydrogenase) inside them, which was measured using an LDH-cytotoxic test kit (Wako).
Measurement of caspase 3 activity
Caspase 3 activity in Caco-2 monolayers was measured by the CaspACE™ colorimetric assay system (Promega) according to the manufacturer's protocol.
Time-dependent IEX-1 mRNA expression in Caco-2 cells during co-culture with THP-1 cells
Effect of anti-TNFα antibody on IEX-1 mRNA expression in Caco-2 cells during co-culture with THP-1 cells
Effect of overexpression and knockdown of IEX-1 on the characteristics of Caco-2 cells
Effect of overexpression and knockdown of IEX-1 on the characteristics of Caco-2 cells co-cultured with THP-1 cells
IEX-1 mRNA expression in Caco-2 cells was induced at an early stage of co-culturing with THP-1 cells
To elucidate the time course of IEX-1 mRNA expression during co-culture with THP-1 cells, RNA was extracted from Caco-2 cells co-cultured with THP-1 cells for 0, 0.5, 1 and 3 h and used for real-time RT–PCR. Figure 1 shows that IEX-1 mRNA expression was induced within 0.5 h after co-culturing with THP-1 cells and reached a peak at 1 h.
IEX-1 mRNA expression in Caco-2 cells co-cultured with THP-1 cells was elicited by TNFα
Next, we investigated the effect of TNFα secreted by THP-1 cells on IEX-1 mRNA expression by adding anti-TNF-α antibody to the incubation medium. The co-culture-induced up-regulation of IEX-1 mRNA expression was markedly ameliorated by treating the cells with anti-TNFα antibody (Figure 2). This indicates that TNFα is primarily involved in the increase in IEX-1 mRNA expression level in Caco-2 cells.
Overexpression or knockdown of IEX-1 in Caco-2 cells altered the damage to cells
IEX-1 mRNA expression in Caco-2 cells was up-regulated at an early stage of co-culturing with THP-1 cells as shown in Figure 1. As shown previously, the TER (transepithelial electrical resistance) value decreased and the released LDH increased in Caco-2 cells by co-culturing with THP-1 cells. Morphological characteristics of both apoptosis and necrosis were also observed in the disrupted Caco-2 cells . These results indicate disruption of barrier function and induction of cell damage. To investigate the function of IEX-1 in the co-culture-induced damage of Caco-2 cells, we overexpressed IEX-1 in Caco-2 cells by lentiviral infection. IEX-1 protein expression in Caco-2 cells infected with IEX-1 expressing vector was confirmed by Western blotting 2 weeks after culturing (Figure 3A). The amount of LDH release from Caco-2 cells was significantly reduced by IEX-1 overexpression, as shown in Figure 3(B).
We subsequently examined the functions of endogenous IEX-1 in Caco-2 cells through lentiviral infection with a lentiviral vector expressing shRNA to achieve long-term knockdown of IEX-1 expression. In 2 weeks, the mRNA level of IEX-1 in the knockdown cells was reduced by 60% compared with that in the control cells (Figure 3C). In contrast with the Caco-2 cells overexpressing IEX-1 (Figure 3B), the LDH release from IEX-1-knockdown Caco-2 cells showed a tendency to increase compared with that from the control cells (Figure 3D). These results suggest that IEX-1 is involved in the suppression of damage in Caco-2 cells.
We have reported that apoptosis occurs in Caco-2 cells when they are co-cultured with THP-1 cells . A significant increase in caspase 3 activity in IEX-1 knockdown cells was observed, suggesting the suppressive nature of IEX-1 to apoptosis induction (Figure 3E). We have previously reported that TNFR1 mRNA expression was up-regulated in Caco-2 cells by co-culturing with THP-1 cells . An increase in the TNFR1 mRNA level was clearly observed in IEX-1-knockdown Caco-2 cells by real-time RT–PCR experiments (Figure 3F).
Co-culture-induced damage in Caco-2 cells altered by overexpression or knockdown of IEX-1 in Caco-2 cells
LDH release from IEX-1-overexpressing and -knockdown Caco-2 cells was determined after co-culturing with THP-1 cells to investigate functions of IEX-1 in inflammed intestinal epithelial cells. LDH release from Caco-2 cells increased by co-culturing with THP-1 cells, but was significantly lower in IEX-1-overexpressing Caco-2 cells (Figure 4A). In contrast, LDH release from co-cultured Caco-2 cells was much higher in IEX-1-knockdown cells (Figure 4B). A pronounced increase in the TNFR1 mRNA level was also observed in IEX-1-knockdown Caco-2 cells on co-culturing with THP-1 cells (Figure 4C).
In the present study, we showed that IEX-1 mRNA expression in Caco-2 cells was significantly increased at an early stage by co-culturing with THP-1 cells (Figure 1). Furthermore, IEX-1 mRNA expression in Caco-2 cells was blocked by adding anti-TNFα antibody (Figure 2), suggesting that IEX-1 plays a crucial role in intestinal inflammatory reactions in which TNFα is involved. The role of IEX-1 in Caco-2 cells was elucidated using IEX-1-overexpressing and -knockdown Caco-2 cells. LDH assay using these Caco-2 cells showed that IEX-1 was likely to be involved in the suppression of Caco-2 cell damage under both normal and inflammatory conditions (Figures 3B, 3D, 4A and 4B). Furthermore, increases in caspase 3 activity and TNFR1 mRNA expression were shown in IEX-1-knockdown Caco-2 cells (Figures 3E, 3F and 4C). These results suggest that IEX-1 protects Caco-2 cells by reducing sensitivity to TNFα, thereby suppressing TNFα-induced apoptosis.
Our attempts to reveal the mechanism for the protective function of IEX-1 against TNFα were, however, not successful for the following reasons. We previously showed that the cell damage in Caco-2 cells induced by co-culturing with THP-1 cells was ameliorated by the treatment with an anti-TNF-α antibody. However, the same concentration of TNFα as that secreted from co-cultured THP-1 cells did not induce cell damage in Caco-2 cells. These results strongly suggest that TNFα is a key player in this phenomenon (co-culture-induced damage), but some other factors, unknown except for TNFα, may be involved in the damage in Caco-2 cells. Furthermore, IEX-1-overexpressed Caco-2 cells were found to be more resistant to the THP-1-derived damaging factors, showing no apoptotic change during the co-culture. Analysing the mechanism of co-culture-induced damage of Caco-2 cells by a treatment with TNFα alone is therefore highly difficult. Identification of the above-mentioned unknown factors in the future may hopefully make it possible to elucidate the interaction between IEX-1- and THP-1-derived factors.
Although the regulatory mechanisms for TNFR1 expression are not yet fully understood, Bristol et al.  have recently reported that the TNFR1 promoter contains a C/EBP (CCAAT/enhancer-binding protein) binding motif located at position −88 to −80 and that binding of either C/EBPα or C/EBPβ induces an increase in TNFR1 expression. Association of IEX-1 with the promoter of anti-apoptotic genes was revealed by Arlt et al.  by ChIP (chromatin immunoprecipitation) assay. These reports indicate the possibility that IEX-1 suppresses C/EBP expression or interferes with the interaction of C/EBP and its binding site although further studies are needed to reveal the mechanism for the regulation of TNFR1 expression by IEX-1.
As mentioned above, IEX-1 suppressed induction of apoptosis in Caco-2 cells co-cultured with THP-1 cells. In contrast, previous reports using hepatocytes and HEK-293 cells have indicated that IEX-1 promotes TNFα-induced apoptosis [16,23,24]. Osawa et al.  reported that IEX-1 promoted TNFα-induced hepatocyte apoptosis by blocking the PI3K (phosphoinositide 3-kinase)/Akt (protein kinase B) survival pathway. Furthermore, Arlt et al. [23,24] showed that IEX-1 interfered with the turnover of IκBα (inhibitory κBα) by inhibiting the assembly and activity of the 26 S proteasome, and suggested that IEX-1 prevented NF-κB (nuclear factor κB)-dependent protection of cells from apoptosis.
The present study has reported for the first time that TNFα-induced IEX-1 may be involved in the suppression of apoptosis in intestinal epithelial cells. The discrepancy between previous findings that IEX-1 promotes apoptosis [16,23,24] and our results can be explained as follows. First, cells used were Caco-2 cells differentiated into small-intestinal epithelial-like cells. In general, cells proceeding in the cell cycle are prone to apoptosis, whereas cells in G1 arrest are more refractory [25–27]. Disruption of IEX-1 expression using anti-IEX-1 ribozymes was reported to decrease the number of HEK-293 cells and HeLa cells in the G2/M phase, which was accompanied by a reduction in apoptosis triggered by ligation of Fas or anti-cancer drugs stoposide and doxorubicin [28,29]. This suggests that, although continuously growing hepatocytes and HEK-293 cells easily undergo apoptosis, differentiated Caco-2 cells, which have stopped growing, are not susceptible to apoptosis. Furthermore, it is possible that the cellular response to given stimuli is different among tissues. Schilling et al.  reported that overexpression of IEX-1 in HaCaT keratinocytes increased the growth rate of cells under basal conditions. However, the rate of apoptosis increased in IEX-1-overexpressing HaCaT keratinocytes when they were subjected to stresses such as serum deprivation and UV irradiation. This indicated the possibility that the function of IEX-1 differs depending on the cell milleu.
Caco-2 cells finally die by co-culturing with THP-1 cells even if IEX-1 expression is induced. When infected by pathogenic bacteria, macrophages generally induce apoptosis in approx. 1 h [31,32], but apoptosis induction in intestinal epithelial cells requires 12–18 h . This may suggest that intestinal epithelial cells slow down the progression of apoptosis by up-regulating IEX-1 expression to maintain the epithelial barrier as long as possible and prevent the spread of bacterial invasion and infection.
Up-regulation of IEX-1 was demonstrated by DNA microarray analysis of a mucosal gene expression profile in IBD . In addition, Sina et al.  reported that ablation of gly96, which is a mouse homologue of IEX-1, aggravated DSS-induced colitis and triggered inflammation in mice. These reports seem to emphasize the importance of IEX-1 in intestinal inflammation. Our present results support the previous findings and suggest the usefulness of this co-culture system in analysing cell–cell interactions in inflammed intestinal epithelium.
In conclusion, the present results show induction of IEX-1 expression at an early stage of co-culturing and indicate that IEX-1 functions in intestinal epithelial cells. Although many questions on the role of IEX-1 in intestinal epithelial cells still remain to be elucidated, the findings from this study suggested that substances that modulate IEX-1 expression may be used as active ingredients in medicines and functional foods for controlling intestinal inflammation.
Dulbecco's modified Eagle's medium
dextran sodium sulfate
fetal bovine serum
human embryonic kidney
inflammatory bowel disease
immediate early-response gene X-1
National Center for Biotechnology Information
tumour necrosis factor α
TNF receptor 1
All experiments were done by Yoko Ishimoto under the supervision of Makoto Shimizu. Technical advice and comments were given by Hideo Satsu and Mamoru Totsuka throughout this study, which were extremely useful in designing the experiments and also preparing the paper.
We thank Dr H. Miyoshi (RIKEN) for lentiviral plasmids (pCAG-HIVgp, pCMV-VSV-G-RSV-Rev, CS II-EF-MCS-IRES-Venus, pENTE4-H1 and CS-RfA-EG). We thank Dr R. Sato for the expression lentiviral plasmid for control shRNA.
This work was supported by a Grant-in-Aid for Scientific Research [grant number A-20248013] from the Japan Society for the Promotion of Science.