Regulation of corepressor alternative mRNA splicing by hormonal and metabolic signaling

Abstract

Alternative mRNA splicing diversifies the products encoded by the NCoR and SMRT corepressor loci. There is a programmed alteration in NCoR mRNA splicing during adipocyte differentiation from an NCoRω isoform, which contains three nuclear receptor interaction domains, to an NCoRδ isoform that contains two nuclear receptor interaction domains. This alternative mRNA splicing of NCoR has profound effects on adiposity and on diabetes in mouse models. We report here that dexamethasone, a powerful regulator of metabolism and of adipocyte differentiation, confers this change in NCoR mRNA splicing in cultured adipocytes. We also demonstrate that changes in dietary components can consistently, if moderately, modulate the total transcript levels and the mRNA splicing of NCoR and SMRT in both cultured cells and intact mice. This ability of alternative corepressor mRNA splicing to respond to nutritional changes confirms its importance in regulating glucose and lipid metabolism, and its promise as a therapeutic candidate for metabolic disorders such as type 2 diabetes.

Keywords: Alternative mRNA splicing; Corepressors; Isoforms; NCoR; SMRT; Transcriptional regulation.

Fig. 1

Schematics of the NCoR and SMRT corepressor proteins are represented. The effects of alternative mRNA splicing at NCoR exons 28 and 37, and at SMRT exon 40 are shown. Protein coding regions deleted or added by a splicing event are indicated, and vertical ovals represent nuclear receptor interaction domains.

Fig. 2

DEX induces the change in NCoR mRNA splicing observed during 3T3-L1 cells differentiation. Cells were differentiated with various combinations of the individual factors within the tripartite differentiation cocktail. Means ± S.E.M. (n = 3 for SMRT Exon 40 and NCoR Exon 28, and n = 5 for NCoR Exon 37) are presented. p < 0.05 (*) and p < 0.01 (†) are indicated. (A) Effects of different combinations of the cocktail on alternative mRNA splicing of SMRT and NCoR. Corepressor transcript isoform abundance was determined as in Materials and Methods. (B) Effects of different combinations of the cocktail on lipid accumulation were quantified by Oil Red O staining as in Materials and Methods. Means ± S.E.M. (n = 3) are presented.

Fig. 3

Chronic ROSI treatment replaces insulin in supporting the change in NCoR mRNA splicing observed during 3T3-L1 cell differentiation. Cells were treated with the tripartite differentiation cocktail on differentiation day 0, followed by: eight days of insulin treatment (“Insulin”), four days of insulin treatment and four days of ROSI treatment (“Insulin ROSI”), or eight days of ROSI treatment (“ROSI”). Corepressor transcript isoform abundance and statistics were determined as in Fig. 2. Mean ± S.E.M. (n = 3) are presented. p < 0.01 (†) and p < 0.001 (‡) are indicated.

Fig. 4

Acute ROSI treatment also induces a change in corepressor mRNA splicing in 3T3-L1 cells. Preadipocytes (3T3-L1 cells never treated with the differentiation cocktail), or mature adipocytes (3T3-L1 cells at differentiation day 12 after exposure to the tripartite differentiation cocktail) were treated acutely for 48 h with ROSI. Corepressor transcript isoform abundance and statistics were determined as in Fig. 2. Mean ± S.E.M (n = 3) are presented. p < 0.05 (*) and p < 0.01 (†) are indicated.

Fig. 5

The availability of glucose, but not fatty acids, alters the gene expression of total (pan-splice) NCoR and total (pan-splice) SMRT. The indicated cells were exposed for 48 h to glucose or oleic acid. Total levels (all splice isoforms) of NCoR and SMRT mRNAs were determined by q RT-PCR as in Materials and Methods. Means ± S.E.M. (n = 4 for MEFs, n = 3 for Hepa1-6 and 3T3-L1 cells) are presented. p < 0.05 (*), p < 0.01 (†) and p < 0.001 (‡) are indicated.

Fig. 6

The availability of glucose, but not fatty acids, alters the alternative mRNA splicing of NCoR and SMRT. Cells were exposed for 48 h to the concentrations of glucose, oleic acid or linoleic acid indicated, using (A) differentiated 3T3-L1 adipocytes (day 12), (B) Hepa1-6 liver cells, or (C) MEFs. Relative transcript isoform abundance and statistics were determined as in Fig. 2. Means ± S.E.M. (n = 3 for all experiments, except n = 6 for NCoR Exon 28 at the 50 mM concentration) are shown. p < 0.05 (*), p < 0.01 (†), p < 0.001 (‡), and p < 0.0001 (§) are indicated.

Fig. 7

A low sucrose versus high sucrose diet alters the mRNA splicing of NCoR and SMRT in intact mice. Adult male, wild-type C57BL/6N mice were fed a low sucrose, high fat diet (LSD) or a high sucrose, low fat diet (HSD) (equivalent calories per gram diet) for 14 weeks. At the end of the study, mRNAs from the indicated tissues were analyzed as in Materials and Methods. Corepressor transcript isoform abundance is displayed in (A) visceral white adipose tissue (vWAT), (B) subcutaneous white adipose tissue (sWAT), (C) liver, or (D) skeletal muscle. Means ± S.E.M. (n = 7 for LS/HFD vWAT, n = 3 for HS/LFD vWAT, n = 4 for sWAT, n = 7 for LS/HFD liver, n = 4 for HS/LFD liver, n = 4 for skeletal muscle) are presented.