The short-chain fatty acid receptor GPR43 is transcriptionally regulated by XBP1 in human monocytes

Abstract

G-protein coupled receptor 43 (GPR43) recognizes short chain fatty acids and is implicated in obesity, colitis, asthma and arthritis. Here, we present the first full characterization of the GPR43 promoter and 5'-UTR. 5'-RACE of the GPR43 transcript identified the transcription start site (TSS) and a 124 bp 5'-UTR followed by a 1335 bp intron upstream of the ATG start codon. The sequence spanning -4560 to +68 bp relative to the GPR43 TSS was found to contain strong promoter activity, increasing luciferase reporter expression by >100-fold in U937 monocytes. Stepwise deletions further narrowed the putative GPR43 promoter (-451 to +68). Site-directed mutagenesis identified XBP1 as a core cis element, the mutation of which abrogated transcriptional activity. Mutations of predicted CREB, CHOP, NFAT and STAT5 binding sites, partially reduced promoter activity. ChIP assays confirmed the binding of XBP1 to the endogenous GPR43 promoter. Consistently, GPR43 expression is reduced in monocytes upon siRNA-knockdown of XBP1, while A549 cells overexpressing XBP1 displayed elevated GPR43 levels. Based on its ability to activate XBP1, we predicted and confirmed that TNFα induces GPR43 expression in human monocytes. Altogether, our findings form the basis for strategic modulation of GPR43 expression, with a view to regulate GPR43-associated diseases.

Figure 1. GPR43 mRNA is up-regulated in primary human monocytes, neutrophils and PMA differentiated U937 cells as determined by quantitative PCR.

Results shown are standardized to undifferentiated U937 cells and error bars represent the mean ± s.d. of three independent cell cultures.

Figure 2. The GPR43 transcription start site is mapped to 1459 bp upstream of the ATG start codon.

(a) 5′-RACE of GPR43 transcripts from U937 cells. Total RNA from U937 cells was extracted and full-length GPR43 mRNA was reverse transcribed using a gene specific primer complementary to its open reading frame. The resulting cDNA was ligated to a known oligonucleotide sequence at the 5′ end. Nested PCR using primers flanking the 5′ end was then performed to procure an approximately 450 bp product encompassing the 5′ untranslated region (5′-UTR) and part of the GPR43 coding sequence (lane 1). As positive controls, amplification of 246 bp of the GPR43 coding region (lane 2), as well as a 900 bp-long 5′ end of the β-actin gene (lane 3) were performed. (b) Schematic mapping the GPR43 5′-UTR region and protein coding region on the human chromosome 19. The PCR product from (a, lane 1) was sequenced to reveal a 124 bp long 5′-UTR exon upstream of the GPR43 ATG start codon.

Figure 3. Deletions define a 519-bp putative GPR43 promoter containing the core and proximal promoter.

(a) ECR Browser plot spanning −4560 bp upstream of the transcription start site (+1), the entire GPR43 gene and 2000 bp after the gene. The plot represents sequence homology of the mouse chromosome 7 (Ref Seq ID: NC_000073.6 ) region flanking the murine Gpr43 (Entrez Gene ID: 233079) in comparison to the corresponding region on the human chromosome 19 (RefSeq ID: NC_000019.9; positions: 35934599–35942669), which is acting as the baseline. Sequence lengths of >100 bp with 70% sequence conservation are shown as peaks. (b) Luciferase reporter activities of 5′ deletion constructs from a 4628 bp putative GPR43 promoter in differentiated U937 cells 22 h after transfection. Analysis of the homology with mouse sequence was used to approximate the deletion sites. Results represent the average Firefly luciferase read-outs of three independent transfections (n = 3) normalized to Renilla luciferase activity and relative to the basic (empty) luciferase vector, arbitrarily set as 1. Error bars represent the mean ± s.d.. The data shown are representative of three independent experiments. Two tailed Students' T-test was used to determine the statistical significance of the difference between promoter constructs and is annotated as: * <0.05, ** <0.01, and *** <0.001.

Figure 4. In silico predictions, null mutations of the 519 bp promoter region and signalling pathway modulation identify transcription factors (TFs) and pathways regulating GPR43 expression.

(a) Predicted TF binding sites by MatInspector software with corresponding WT and mutant sequences and their relative positions indicated above each TF binding site. The core recognition sequences are underlined while mutated bases are indicated in red. (b) Relative luciferase reporter activities of 519-bp promoter mutated at indicated sites were measured in differentiated U937 monocytes. The multiple binding sites of PU.1 or NFAT residing within the 519 bp promoter were simultaneously mutated in their respective constructs. Wild type (WT) promoter is arbitrarily set to 1 luciferase unit. (c) Quantitative PCR analysis of GPR43 mRNA levels upon modulation of signalling pathways after 1 h pre-treatment with activators/inhibitors followed by immune challenge for 3 h with 100 ng/mL LPS. (d) Quantitative PCR analysis of GPR43 mRNA levels upon 4 h treatment of monocytes with inhibitor/activator or 12 h treatment with inhibitor. (c) and (d) Inhibitor/Activator + (Targeted signalling proteins) are shown: SB203580, 10 μM —| p38; U73122, 5 μM —| phospholipase C (PLC); MG132, 10 μM —| Proteasome; Wortmannin, 2 μM —| PI3kinase (PI3K); BisI, 4 μM —| protein kinase C (PKC); Forskolin, 20 μM → Adenylyl cyclase (AC). Expression levels of two reference genes, beta-2-microglobulin (B2M) and cyclophilin B (CYPB) were also measured and presented as Supplementary Figure S2. All measurements were standardized to the RPL27 as the reference gene. Experiments were performed in triplicate transfections or treatments, where error bars represent the mean ± s.d. In (b), Dunnett's test (α = 0.05) was performed while in (c) and (d), the two tailed Students' T-test was used to determine statistical significance. p-value * <0.05, ** <0.01, and *** <0.001. The data shown are representative of three independent experiments.

Figure 5. XBP1 binds to the GPR43 promoter in vivo.

(a) Schematic showing the relative regions of GPR43 promoter amplified by quantitative PCR following ChIP assay with anti-XBP1 antibody. P1–P6 spans a region of ~1200 bp surrounding the XBP1 binding site on the GPR43 promoter. Only promoter regions, P3 and P4 contain the XBP1 binding site. The GPR43 enhancer region and coding sequence, located more than 3 kb and 2.5 kb up- and downstream of the XBP1 binding site respectively, were also amplified as negative controls. (b) Quantitative PCR analysis of GPR43 promoter regions following ChIP assay on U937 cells using anti-XBP1 antibody or IgG isotype control antibody. Results are expressed as fold enrichment relative to GPR43 coding sequence (GPR43_CDS) negative control region and error bars represent the mean ± s.d. of three independent pull-down. p-value * <0.05, ** <0.01, and *** <0.001. The data shown are representative of two independent experiments.

Figure 6. XBP1 is critical for the expression of GPR43.

(a) Quantitative PCR analysis of GPR43 and spliced XBP1 (XBP1s) mRNA in XBP1-stable knockdown U937 cells relative to the negative control (non-targeting) siRNA expression plasmid, arbitrarily set as 1. (b) Western blot analysis of unspliced XBP1 (XBP1u) protein levels in XBP1-stable knockdown U937 cells. XBP1s levels were undetectable. (c) Quantitative PCR analysis of GPR43 and XBP1 mRNA after 48 h overexpression of XBP1s in A549 cells, relative to the pcDNA empty vector which is set arbitrarily as 1. (d) Western blot analysis of XBP1s protein levels after 48 h overexpression of XBP1s in A549 cells. (b) and (c) Measurements were standardized to RPL27 as the reference gene. (b) to (d) Results shown are the average of three independent transductions (for stable knockdown) or transfections (for overexpression), with error bars representing the mean ± s.d.. Two tailed Students' T-test was used to calculate statistical significance. p-value * <0.05, *** <0.001.

Figure 7. Human monocyte GPR43 expression is up-regulated by LPS, TNFα and GM-CSF treatment.

(a) Quantitative PCR analysis of GPR43 mRNA and (b) spliced XBP1 (XBP1s) mRNA levels under 3 h treatment with either 100 ng/mL LPS; 10 ng/mL TNFα; 100 ng/mL GM-CSF or 10 mM Acetate (Ac). (a) and (b) The results represent average fold change of the treated samples relative to the non-treated control (NT) and error bars represent the mean ± s.d. of three independent treatment wells. Measurements were standardized to RPL27 as the reference gene. Two tailed Students' T-test was used to determine the statistical significance of the difference between treated and NT samples. p-value * <0.05, ** <0.01, and *** <0.001. The data shown are representative of at least two independent experiments.