Cooperation between HMGA1, PDX-1, and MafA is Essential for Glucose-Induced Insulin Transcription in Pancreatic Beta Cells

Abstract

The high-mobility group AT-hook 1 (HMGA1) protein is a nuclear architectural factor that can organize chromatin structures. It regulates gene expression by controlling the formation of stereospecific multiprotein complexes called "enhanceosomes" on the AT-rich regions of target gene promoters. Previously, we reported that defects in HMGA1 caused decreased insulin receptor expression and increased susceptibility to type 2 diabetes mellitus in humans and mice. Interestingly, mice with disrupted HMGA1 gene had significantly smaller islets and decreased insulin content in their pancreata, suggesting that HMGA1 may have a direct role in insulin transcription and secretion. Herein, we investigate the regulatory roles of HMGA1 in insulin transcription. We provide evidence that HMGA1 physically interacts with PDX-1 and MafA, two critical transcription factors for insulin gene expression and beta-cell function, both in vitro and in vivo. We then show that the overexpression of HMGA1 significantly improves the transactivating activity of PDX-1 and MafA on human and mouse insulin promoters, while HMGA1 knockdown considerably decreased this transactivating activity. Lastly, we demonstrate that high glucose stimulus significantly increases the binding of HMGA1 to the insulin (INS) gene promoter, suggesting that HMGA1 may act as a glucose-sensitive element controlling the transcription of the INS gene. Together, our findings provide evidence that HMGA1, by regulating PDX-1- and MafA-induced transactivation of the INS gene promoter, plays a critical role in pancreatic beta-cell function and insulin production.

Keywords: HMGA1; MafA; PDX-1; beta-cells; diabetes; insulin gene; transcription.

Figure 1

Schematic representation of the human (INS), rat (InsI), and mouse (InsII) promoter regions. HMGA1, PDX-1, and MafA binding sites are underlined in each gene sequence. The numbers indicate positions in base pairs relative to the transcriptional start site (+1). The rat insulin mini-enhancer element E2A3/4 is indicated.

Figure 2

Physical association between HMGA1, PDX-1, and MafA. (A) SDS-PAGE of ³⁵S-PDX-1, ³⁵S-MafA, and ³⁵S-NeuroD bound to GST or GST-HMGA1 resin (lanes 1–9). In lanes 1, 4, and 7 labeled protein was added directly onto the gel without binding to and elution from GST protein resin. (B) Immunoprecipitation (IP) of HMGA1, PDX-1, and MafA by using the anti-HMGA1 antibody followed by immunoblotting with the anti-PDX-1 antibody (lanes 1–3), the anti-MafA antibody (lanes 4–6), or the anti-HMGA1 antibody (lanes 7–9) after reprobing the same transfer. Lanes: 1, 4, and 7, 10 ng of pure HMGA1, 2, 5, and 8, INS-1 nuclear extract (NE; 500 μg). In lanes 1, 4, and 7, protein was directly applied to the gel without binding to and elution from protein A beads. To prove specificity, pure Sp1 (15 ng) was used for immunoprecipitation by the anti-HMGA1 antibody (lanes 3, 6, and 9, control). The faint band in lane 8 at 43 kDa represents non-specific signal. A representative of three separate assays is shown for each condition.

Figure 3

Functional significance of HMGA1, PDX-1, and MafA for INS gene transcription. HEK-293 cells were cotransfected with 1 μg of human (phINS-Luc) (A) or mouse (pmINSII-Luc) (B) Luc reporter plasmids, in the absence or presence of effector vectors for HMGA1, PDX-1, and MafA (0.5 μg each), either alone or in double or triple combinations. Data represent the means ± SE for three separate experiments; values are expressed as factors by which Luc activity increased as compared with the level of Luc activity obtained in transfections with reporter vector alone (slashed bar), which is assigned an arbitrary value of 1. Open bar, mock (no DNA); black bar, pGL3-basic (vector without an insert). *p < 0.05; **p < 0.01; ***p < 0.001 versus control (slashed bar).

Figure 4

Suppression of INS promoter and mRNA by siRNA to HMGA1. INS-1 (A) and HeLa (B) cells were incubated without or with siRNA targeting HMGA1 (100 pmol) and transfected thereafter with 1 μg of mouse (pmINSII-Luc) or human (phINS-Luc) Luc reporter plasmid, respectively, either in the absence or presence of effector vectors (0.5 μg each) for HMGA1, PDX-1, and MafA. Values are expressed relative to the Luc activity obtained in transfections with the Luc reporter plasmid alone (slashed bar) that is assigned an arbitrary value of 1. Results are the means ± SE of triplicates from three independent transfections. *p < 0.05 versus control (slashed bar). (C) INS-1 cells were transfected with siRNA against HMGA1 or a non-targeting control siRNA, and mRNA levels for HMGA1, Insulin, PDX-1, and MafA were measured by quantitative RT-PCR (qRT-PCR). Results are the mean ± SEM of at least three separate transfections, each in triplicate. *p < 0.05 versus the relative control. Western blots (WB) of HMGA1, either before or after HMGA1 siRNA treatment, are shown in the autoradiograms and are from four independent experiments. Densitometric slot blot analysis, using the ImageJ software program, is shown. Numbers on the peaks are the size of the corresponding slot as a percentage of the total size of the two slots in each condition.

Figure 5

Functional significance of HMGA1 for glucose-induced INS gene expression and insulin secretion. (A) ChIP of the rat Ins promoter gene in INS-1 cells treated with either basal (3 mM) or stimulatory (15 mM) glucose concentrations, using an anti-HMGA1 specific antibody (Ab). A representative assay is shown, together with qRT-PCR of ChIP-ed samples. p < 0.05 versus control (white bar). (B) Insulin secretion from HMGA1 siRNA-treated INS-1 cells. Results are the mean ± SEM for three independent experiments conducted in triplicate. p < 0.05 versus control (white bar). A representative WB of HMGA1 is shown in the autoradiogram, together with densitometric slot blot analysis in each condition.