Alternative mRNA splicing produces a novel biologically active short isoform of PGC-1alpha

Abstract

The transcriptional co-activator PGC-1alpha regulates functional plasticity in adipose tissue by linking sympathetic input to the transcriptional program of adaptive thermogenesis. We report here a novel truncated form of PGC-1alpha (NT-PGC-1alpha) produced by alternative 3' splicing that introduces an in-frame stop codon into PGC-1alpha mRNA. The expressed protein includes the first 267 amino acids of PGC-1alpha and 3 additional amino acids from the splicing insert. NT-PGC-1alpha contains the transactivation and nuclear receptor interaction domains but is missing key domains involved in nuclear localization, interaction with other transcription factors, and protein degradation. Expression and subcellular localization of NT-PGC-1alpha are dynamically regulated in the context of physiological signals that regulate full-length PGC-1alpha, but the truncated domain structure conveys unique properties with respect to protein-protein interactions, protein stability, and recruitment to target gene promoters. Therefore, NT-PGC-1alpha is a co-expressed, previously unrecognized form of PGC-1alpha with functions that are both unique from and complementary to PGC-1alpha.

FIGURE 1.

Schematic diagrams of alternative splicing, sequence alignments, and domain structure of PGC-1α. A, schematic diagram showing alternative splicing of the Pgc-1α gene. B, alignment of genomic sequence of Pgc-1α from end of exon 6 to beginning of exon 7 for mouse, rat, human, chimp, cow, dog, horse, and cat. C, diagram of domain structure of PGC-1α and NT-PGC-1α. NLS, nuclear localization sequence; RS, arginine-serine-rich domain; ERR, estrogen-related receptor; RRM, RNA recognition motif.

FIGURE 2.

Tissue distribution and regulation of expression of NT-PGC-1α and PGC-1α. A, induction of Pgc-1α and NT-Pgc-1α mRNA and protein in brown adipose tissue by cold exposure (4 °C) for 5 h in C57BL/6J mice. β-Actin was blotted as a loading control, and data (mean ± S.E.) are representative of three experiments. Room Temp, room temperature. B, induction of PGC-1α and NT-PGC-1α protein expression in immortalized brown adipocytes from mice 12 h after treatment with vehicle or 8-CPT-cAMP (100 μm). aP2 was blotted as a loading control, and data are representative of three experiments. C, induction of hepatic Pgc-1α and NT-Pgc-1α mRNA and protein by fasting for 12 h in C57BL/6J mice. β-Actin was blotted as a loading control. Data presented as mean ± S.E. are representative of three experiments. D, relative expression of Pgc-1α and NT-Pgc-1α mRNA expression in BAT, brain, kidney, and liver of C57BL/6J mice. Data presented as mean ± S.E. are representative of three experiments. E, relative expression levels of NT-PGC-1α protein in BAT, brain, kidney, and liver of control (wild type (WT)) and PGC-1α null mice. KO, knock-out. F, induction of PGC-1α and NT-PGC-1α protein in brown adipose tissue by cold exposure (4 °C) for 5 h in 3-month-old male Zucker rats. BAT whole cell extracts for room temperature (100 μg) and cold-exposed (30 μg) rats were blotted for β-actin as a loading control, and data (mean ± S.E.) are representative of three experiments. G, relative expression of NT-PGC-1α and PGC-1α protein in whole cell extracts (30 μg) of brains from 3-month-old male Zucker rats. NT-PGC-1α and PGC-1α overexpressed in COS cells were loaded as positive controls. H, NT-PGC-1α expression in whole cell extracts (100 μg) from human heart tissue provided by the National Disease Research Interchange. NT-PGC-1α and PGC-1α overexpressed in COS cells were loaded as positive controls. I, relative targeting of NT-PGC-1α and PGC-1α to the proteosome. NT-PGC-1α and PGC-1α protein expression in CHO-K1 cells transduced with PGC-1α or NT-PGC-1α vectors for 24 h was followed by treatment with vehicle or proteosome inhibitor (1 μm MG132) for 5 h. Data are representative of three experiments.

FIGURE 3.

Regulation of subcellular distribution of NT-PGC-1α in CHO-K1 cells. A, representative confocal images of subcellular distribution of NT-PGC-1α-HA in CHO-K1 cells transduced with expression construct for NT-PGC-1α-HA for 24 h and treated with vehicle for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. B, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in CHO-K1 cells transduced with expression construct for NT-PGC-1α-HA for 24 h and treated with 100 μm 8-CPT-cAMP for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. C, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in CHO-K1 cells transduced with expression construct for NT-PGC-1α-HA for 24 h and treated with PKA inhibitor (H89) and 100 μm 8-CPT-cAMP for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. D, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in CHO-K1 cells transduced with expression construct for NT-PGC-1α-HA for 24 h and treated with p38 MAPK inhibitor (SB203580) and 100 μm 8-CPT-cAMP for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. E, image stacks from confocal imaging of cells represented in A–D were analyzed using ImageJ software, and the signal intensity from NT-PGC-1α co-localized with DAPI was expressed relative to signal intensity in the whole cell. Means ± S.E. are representative of 80–100 cells per treatment from three to four separate experiments.

FIGURE 4.

Regulation of subcellular distribution of NT-PGC-1α in undifferentiated and differentiated brown adipocytes. A, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in undifferentiated brown adipocytes stably transformed with NT-PGC-1α-HA and treated with vehicle for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. B, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in undifferentiated brown adipocytes stably transformed with NT-PGC-1α-HA and treated with 100 μm 8-CPT-cAMP for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. C, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in fully differentiated brown adipocytes stably transformed with NT-PGC-1α-HA and treated with vehicle for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. D, representative confocal images showing subcellular distribution of NT-PGC-1α-HA in fully differentiated brown adipocytes stably transformed with NT-PGC-1α-HA and treated with 100 μm 8-CPT-cAMP for 1 h prior to fixation and imaging. The images on the left and right are presented with and without DAPI staining. E, image stacks from confocal imaging of all cells represented in A–D were analyzed using ImageJ software, and the signal intensity from NT-PGC-1α co-localized with DAPI was expressed relative to signal intensity in the whole cell. Means ± S.E. are representative of 50–60 cells per treatment from three to four separate experiments.

FIGURE 5.

Physical and functional interaction of NT-PGC-1α-HA with PPARα and PPARγ. A, ligand-dependent transactivation of Gal4-PPARα and Gal4-PPARγ by PGC-1α and NT-PGC-1α in Chinese hamster ovary cells transiently transfected with GAL4-responsive luciferase reporter gene (pGK) and a pRL-SV40 plasmid expressing Renilla luciferase for normalization. Luciferase activity was measured 24 h after treatment with vehicle (DMSO), 10 μm WY14693, or 10 μm BRL49653. Fold changes in luciferase activity were expressed relative to cells transfected with pGK alone. Data represent means ± S.E. from three experiments. B, ligand-dependent interaction of ³⁵S-PPARγ or ³⁵S-PPARα by GST-tagged PGC-1α and NT-PGC-1α. The ³⁵S-labeled nuclear receptors were produced by TnT T7 in vitro translation system and incubated with GST-tagged PGC-1α or NT-PGC-1α for 1 h in the absence and presence of ligands for the respective receptors. Autoradiograms are representative of three experiments. C, co-immunoprecipitation analysis of protein-protein interaction of HA-tagged NT-PGC-1α with FLAG-tagged PPARα in COS cells. 24 h after treatment with vehicle (DMSO) or 10 μm WY14693, whole cell extracts were immunoprecipitated with anti-FLAG (top panel), anti-HA (bottom panel), mouse preimmune IgG (top panel), or rabbit IgG (bottom panel). Using Western blots (WB), the respective immunoprecipitates were sequentially probed with anti-HA and anti-FLAG (top panel) or anti-FLAG and anti-HA (bottom panel) to test for co-IP of NT-PGC-1α and PPARα and IP efficiency. Blots are representative of four experiments.

FIGURE 6.

Induction of Ucp1 mRNA and protein expression in immortalized brown adipocytes stably transformed with empty vector, HA-tagged NT-PGC-1α, or V5-tagged PGC-1α. A, Western blot analysis of NT-PGC-1α-HA and PGC-1α-V5 protein expression in brown adipocytes expressing HA-tagged NT-PGC-1α, V5-tagged PGC-1α, or empty vector. IB, immunoblot. Upper panel shows detection of respective proteins with anti-HA or anti-V5, and lower panel shows detection of both proteins in representative clones using an antibody raised against the N-terminal region of PGC-1α. B, real time PCR and Western blot analysis of ligand-dependent induction of Ucp1 mRNA and protein in differentiated brown adipocytes expressing HA-tagged NT-PGC-1α, V5-tagged PGC-1α, or empty vector. Replicate wells of each line were treated with ligand mixtures containing 1 μm 9-cis-RA and either 10 μm WY14693 or BRL49653 for 12 h, followed by 6 h with 100 μm 8-CPT-cAMP. Cell extracts were blotted for UCP1. aP2 was blotted as a differentiation and loading control. Data represent mean ± S.E. of three experiments. C, real time PCR analysis of basal expression of Cpt-1β mRNA expression in differentiated brown adipocytes expressing HA-tagged NT-PGC-1α, V5-tagged PGC-1α, or empty vector. Cell extracts from replicate cultures were blotted for aP2 as a differentiation and loading control. Data represent mean ± S.E. of three experiments. D, recruitment of NT-PGC-1α-HA to Cpt-1β promoter in differentiated brown adipocytes. Differentiated brown adipocytes expressing empty vector or NT-PGC-1α-HA were treated for 12 h with vehicle or 100 μm 8-CPT-cAMP, and cross-linked chromatin was immunoprecipitated with either anti-HA or mouse preimmune IgG. The amount of the Cpt-1β enhancer (top panel) or an intragenic region 3′ of the enhancer (bottom panel) co-immunoprecipitated with NT-PGC-1α was measured by real time PCR using 4 ODs of immunoprecipitated chromatin. Agarose gel showing specificity of PCR and data are representative of three similar experiments. E, recruitment of NT-PGC-1α-HA to UCP-1 promoter in differentiated brown adipocytes. Differentiated brown adipocytes expressing NT-PGC-1α-HA were treated for 12 h with vehicle or 100 μm 8-CPT-cAMP, and cross-linked chromatin was immunoprecipitated with either anti-HA or mouse preimmune IgG. The amount of the UCP-1 enhancer co-immunoprecipitated with NT-PGC-1α (top panel) or an intragenic region 3′ of the enhancer (bottom panel) was measured by real time PCR using 4 ODs of immunoprecipitated chromatin. Agarose gel showing specificity of PCR and data are representative of three similar experiments.

FIGURE 7.

NT-PGC-1α enhances mitochondrial biogenesis in differentiated brown adipocytes. Real time PCR analysis of the ratio of mitochondria to nuclear DNA and Mitofluor Red fluorescence intensity in differentiated brown adipocytes expressing HA-tagged NT-PGC-1α, V5-tagged PGC-1α, or empty vector is shown. Data represent mean ± S.E. of three experiments with 3–10 replicates per experiment.