PGC-1α pathway dysregulation disrupts myofiber specification in a mouse model of SBMA
Curtis J. Kuo, Laura B. Chopp, Zhigang Yu, Luhan Ni, Hien T. Zhao, Janghoo Lim, Andrew P. Lieberman
Curtis J. Kuo, Laura B. Chopp, Zhigang Yu, Luhan Ni, Hien T. Zhao, Janghoo Lim, Andrew P. Lieberman
View: Text | PDF
Research Article Muscle biology Neuroscience

PGC-1α pathway dysregulation disrupts myofiber specification in a mouse model of SBMA

Abstract

Skeletal muscle pathology is a critical but poorly understood contributor to neuromuscular degeneration in spinal and bulbar muscular atrophy (SBMA), a CAG/polyglutamine (polyQ) expansion disorder caused by mutation in the androgen receptor (AR). Using a gene-targeted SBMA mouse model, we applied single-nucleus RNA sequencing to identify a disease-specific population of skeletal muscle myonuclei that replaced normal myonuclear subtypes. This transition was associated with dysregulation of the pathway governed by PGC-1α, a central regulator of myofiber specification and metabolic identity. PGC-1α dysfunction in SBMA muscle was age, hormone, and polyQ length dependent and was partially rescued by subcutaneous delivery of AR-targeted antisense oligonucleotides. Integrated ChIP-seq and RNA-seq analyses revealed that aberrant PGC-1α activity promoted the expression of a distinct set of myofiber specification genes while downregulating those that define healthy Type IIb and Type IIx myonuclei. We propose a model in which this dysfunction arose downstream of polyQ-mediated sequestration of PGC-1α cofactors MEF2, CREB, and CBP, leading to transcriptional reprogramming and cellular dysfunction. These findings implicated PGC-1α dysregulation as a key event linking AR polyQ expansion to skeletal muscle degeneration and suggested a shared mechanism for polyQ-mediated muscle pathology across related neurodegenerative diseases.

Authors

Curtis J. Kuo, Laura B. Chopp, Zhigang Yu, Luhan Ni, Hien T. Zhao, Janghoo Lim, Andrew P. Lieberman

×

Figure 5

PGC-1α pathway dysfunction is hormone and glutamine tract length dependent in AR113Q mice and is conserved in 2 other polyQ mouse models.

Options: View larger image (or click on image) Download as PowerPoint
PGC-1α pathway dysfunction is hormone and glutamine tract length depende...
(A) Relative gene expression by qPCR of 6 PGC-1α target genes in the TA of male mice at 52 weeks from independent cohorts of AR21Q, AR113Q, and AR113Q castrated (“CAST”) mice (n = 5/group for AR21Q and AR113Q, n = 4 for the castrated group). (B) Heatmap of log2(FC) values of the PGC-1α target genes studied in Figure 4G. RNA from TA of AR113Q and WT males at 52 weeks (n = 4/genotype), TA of SCA1 Q154 and WT males at 24 weeks (n = 3/genotype), and quadriceps of R6/2 transgenic and WT males at 12 weeks (n = 4/genotype). The SBMA AR113Q column is repeated from Figure 4G for ease of comparison. (C) UpSet plot visualization of overlap between PGC-1α target genes with differential expression in AR113Q, SCA1 Q154, and R6/2 transgenic skeletal muscle from B. Genes were assigned as “unchanged” when their Padj values were ≥ 0.05 or when |log2(FC)| < 0.5849, “up” when Padj < 0.05 and |log2(FC)| > 0.5849, and “down” when Padj < 0.05 and |log2(FC)| < 0.5849 in the comparisons of polyQ disease to their corresponding age-matched WT controls. (D) Heatmap of |log2(FC)| values of the top 25 cluster defining genes for the 113Q-myo population as seen in Figure 3D, among disease vs. WT bulk RNA-seq analyses as in B. While indications for statistical significance are not shown here, they are included in the accompanying Supporting Data Values file; 21 out of 25 genes met thresholds for differential expression in both SCA1 Q154 and R6/2 transgenic disease vs. WT. Order of genes is as shown in Figure 3D. Data are mean ± SD. NS, not significant. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 by 2-way ANOVA with Tukey’s multiple-comparison test (A). All results of statistical analyses are in left-to-right order (Nmrk2, Clec2d, Slc2a4, Ankrd23, Smyd2, Atp1b1). (A) Main ANOVA results: F = 103.9, 38.19, 30.38, 17.98, 14.93, 32.70; df = 2; P = <0.0001, 0.0002, 0.0004, 0.0017, 0.0030, 0.0003. For Padj values by multiple comparisons testing, see the Supporting Data Values file.