- Top of page
- Materials and methods
XBP1 is a key transcription factor in the endoplasmic reticulum (ER) stress response pathway. In a previous study, we suggested a possible link between XBP1 and bipolar disorder, but its role in neuronal cells has not yet been clarified. Here we examined the target genes of XBP1, using DNA microarray analysis in SH-SY5Y cells transfected with an XBP1-expressing vector. Among the genes up-regulated by XBP1, the most significant p-value was observed for WFS1, which is an ER stress response-related gene. Examining the promoter region of WFS1, we found a conserved sequence (CGAGGCGCACCGTGATTGG) that is highly similar to the ER stress response element (ERSE). A promoter assay showed that this ERSE-like motif is critical for the regulation of WFS1 by XBP1. An electrophoretic mobility shift assay suggested that XBP1 does not directly bind to this sequence. Our results demonstrate that WFS1 is one of the target genes of XBP1 in SH-SY5Y cells.
The endoplasmic reticulum (ER) is responsible for protein folding within each cell. When unfolded proteins accumulate in the ER, the ER stress response begins. The ER stress response consists of four signaling cascades: (i) induction of ER chaperones such as HSPA5 (GRP78/BiP), which promotes the folding of unfolded proteins (unfolded protein response, UPR); (ii) inhibition of protein synthesis; (iii) induction of the ER-associated degradation pathway; and (iv) induction of apoptosis (Yoshida 2004; Schroder and Kaufman 2005). The UPR begins when HSPA5 proteins are used to fold unfolded proteins. Dissociation of HSPA5 from ATF6 protein on the ER membrane causes cleavage of ATF6, and cleaved ATF6 protein induces the expression of ER chaperones and XBP1. In parallel, dissociation of HSPA5 from IRE1 protein on the ER membrane causes dimerization of IRE1, which splices XBP1 mRNA. The spliced XBP1 mRNA encodes an active form of XBP1 that strongly induces the expression of target genes such as ER chaperones (Yoshida 2004).
We previously showed, by DNA microarray analysis, that XBP1 and HSPA5 are down-regulated in the lymphoblastoid cells of monozygotic twins with bipolar disorder, compared with healthy co-twins (Kakiuchi et al. 2003). However, the role of the ER stress response pathway in the brain has not yet been clarified, except for the possible role of the UPR in the modulation of glutamate receptor trafficking (Shim et al. 2004; Vandenberghe et al. 2005).
To clarify the role of XBP1 in neuronal cells, we investigated the target genes of XBP1 in SH-SY5Y cells. First, we performed DNA microarray analysis in SH-SY5Y cells transfected with an XBP1-expressing vector, and we identified WFS1 as the most up-regulated gene. Next, we found an ER stress response element (ERSE)-like sequence in the promoter region of WFS1. We further searched for ERSE-like sequences in the genes altered by XBP1 overexpression, and found that the gene for glycine cleavage system H protein (GCSH) is another candidate gene with an ERSE-like sequence in its upstream region. We confirmed by a promoter assay that this sequence is critical for the response of WFS1 to XBP1 overexpression. However, XBP1 did not directly bind to the ERSE-like sequence in a gel-shift assay. Our results demonstrate that WFS1 has an ERSE-like sequence in its promoter and XBP1 regulates WFS1 expression indirectly through this motif.
- Top of page
- Materials and methods
We demonstrated in this report that WFS1 expression is regulated by XBP1 through an ERSE-like sequence, not by binding directly.
The ERSE is observed in ER chaperones (GRP78, GRP94, calreticulin), as well as CHOP and XBP1 (Yoshida et al. 1998, 2001; Ubeda and Habener 2000). Although the consensus sequence of ERSE is CCAAT-N9-CCACG, its variant, CCAAT-N9-GCACG, is also reported to be functional in the CHOP gene. The ERSE-like sequence of WFS1 may also be another variant of ERSE. Recently, WFS1 was reported to be induced by ER stressors such as thapsigargin and dithiothreitol in pancreatic islets (Yamaguchi et al. 2004; Ueda et al. 2005). This previously reported induction via transcriptional activation might be through this ERSE sequence.
In addition to the ERSE of WFS1, a 11-bp sequence including CCAAT near the ERSE is also highly conserved in mammals. Although we could not observe the effect of sXBP1-dependent promoter activity, it cannot be ruled out that this element enhances the promoter activity of the ERSE-like sequence.
Several ERSE-binding proteins have been reported, such as ATF6, NF-Y, YY1 and TFII-I (Yoshida et al. 1998, 2000; Li et al. 2000; Parker et al. 2001). Of these, expression of the gene for YY1 was significantly altered, but it was down-regulated. None of these was found to be up-regulated by XBP1. XBP1 itself is also known to bind to ERSE in the presence of NF-Y (Yoshida et al. 1998; Yamamoto et al. 2004). However, anti-c-myc antibody did not cause a supershift in the EMSA, and nuclear extracts from XBP1–/– MEFs could also cause a band shift. These findings suggest that XBP1 does not directly bind to the ERSE-like sequence of WFS1. The amount of shifted band was smaller in XBP1–/– MEFs, suggesting that some transcription factor regulated by XBP1 binds to the ERSE-like sequence on the WFS1 promoter.
WFS1 was initially identified as a causative gene for Wolfram disease (Inoue et al. 1998; Strom et al. 1998; Hardy et al. 1999). Wolfram disease (OMIM in NCBI website 222300) is a rare autosomal-recessive disorder characterized by diabetes insipidus, diabetes mellitus, optic atrophy and deafness. Swift and colleagues reported that about 60% of altered individuals have some mental disturbance, such as severe depression, psychosis or organic brain syndrome, as well as impulsive verbal and physical aggression (Swift et al. 1990). They further reported that even the heterozygotes who do not have Wolfram disease are 26 times more likely than non-carriers to have a psychiatric hospitalization, mainly because of severe depression (Swift and Swift 2000). WFS1 protein predominantly localizes to ER (Takeda et al. 2001), and induces cation channel activity on the ER membrane (Osman et al. 2003). However, its molecular function is still not well characterized. Our results, as well as those in two other recent reports (Yamaguchi et al. 2004; Ueda et al. 2005), clearly indicate that WFS1 is involved in the UPR pathway.
In addition to WFS1, we suggest that the gene encoding GCSH is another candidate for XBP1-mediated regulation through an ERSE-like sequence. GCSH has an ERSE-like sequence in its upstream region near the first exon, where the known ERSEs, as well as the ERSE-like sequence of WFS1, are typically found. However, GCSH expression was down-regulated when XBP1 was overexpressed. GCSH is a component of the enzyme system for cleavage of glycine (OMIM 238330), and a defect in GCSH is implicated in glycine encephalopathy (OMIM 605899). A link between this gene and the ER stress response has not yet been reported. Further studies are required to clarify the functional role of this motif.
In summary, we screened for target genes of XBP1 in neuroblastoma cells and identified WFS1 as a strong candidate. An as-yet-unknown transcription factor is thought to mediate this induction. Further studies are needed to clarify further details of ER stress signaling in neuronal cells.