PFKFB3-mediated glycolysis rescues myopathic outcomes in the ischemic limb

Figures

• 

  Figure 1. D257A+/+ (mtDNA mutator) mice display a distinct phenotype. WT, heterozygous (D257A+/–), and homozygous (D257A+/+) mice were aged to 12 months. (A) Visual representation of aged mice demonstrates typical phenotype (graying of hair, weight loss, kyphosis). (B) Quantification of body weight at 12-months of age. (C) Quantification of skeletal muscle contractile function in cage-control transgenic mice. (D) Mitochondrial phenotyping in mtDNA mutator mice. Mitochondria were isolated from hind limb skeletal muscle of control limbs for the direct analysis of respiratory capacity. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 using ANOVA (1-way in B and C, 2-way in D) with Tukey’s post hoc for comparisons. Values are presented as mean ± SEM.

• 

  Figure 2. D257A+/+ (mtDNA mutator) mice are protected from ischemic muscle injury. Mice were aged to 12 months before performing unilateral hind limb ischemic (HLI). (A) Following HLI, homozygous mutants displayed greater limb perfusion recovery (A and B, n = 14/group) and increased perfused capillary area (C, n = 6/group; HLI d7), protection against ischemic injury to the skeletal muscle (D and E, n = 13 for WT and D257A+/–, n = 14 for D257A+/+), and no impairment of skeletal muscle contractile function in ischemic EDL muscles (F, n = 6–7/group; HLI d7). (G) Respiratory function in ischemic limb muscle mitochondria supported by pyruvate, malate, glutamate, and succinate under state 4 (S4) and S3 (ADP-stimulated) conditions. (H) Percent change in mitochondrial respiration (from control limb). *P < 0.05, ***P < 0.001, ****P < 0.0001 as indicated or versus WT (between group) using ANOVA (1-way in E, 2-way in B, C, F, G, and H) with Tukey’s post hoc for comparisons. φP < 0.05 versus nonischemic control (within group) using 2-way ANOVA with Tukey’s post hoc for comparisons. Values are presented as mean ± SEM.

• 

  Figure 3. D257A+/+ (mtDNA mutator) mice are metabolically reprogrammed to enhance limb muscle glycolysis. (A) Top enriched gene ontology (GO) terms obtained from differentially expressed genes suggest a reprogramming of glycolytic metabolism. (B) Heatmap of top glycolytic genes. (C–F) Values are reported as the log2 of the fold change from the WT control limb. D257A+/+ also had increased blood lactate (C, n = 5–6/group), elevated glycolytic flux over time in primary muscle cells (D, n = 14/group; graph to the right shows basal and maximal rates), increased mRNA expression of PFKFB3 (E, n = 6/group) and PFKM (F, n = 6/group), as well as increased PFKFB3 protein expression (G and H n = 4/group). *P < 0.05, ** P < 0.01, *** P < 0.001, and ****P < 0.0001 versus WT using ANOVA (1-way in C and H, 2-way in D–F) with Tukey’s post hoc for comparisons. φP < 0.05 versus nonischemic control using 2-way ANOVA with Tukey’s post hoc for comparisons. Values are presented as mean ± SEM. Homozygous D257A+/+ mice were given 25 mg/kg PFK15, a small molecule specific inhibitor of PFKFB3, via i.p. injection 24 hours before HLI to inhibit PFKFB3, or equal volume DMSO/saline mixture as control. (I and J) PFKFB3…

• 

  Figure 4. Chronic mitochondrial dysfunction in skeletal muscle of BALB/c mice. (A) Representative laser Doppler perfusion images (LDPI) from BL6 and BALB/c mice prior to and during the recovery from unilateral hind limb ischemia surgery (HLI). (B) Quantification of LDPI limb perfusion. (C) Ex vivo force production (expressed as a percentage of the contralateral control limb) measured in the extensor digitorum longus muscle of BL6 and BALB/c. Mitochondrial function was assessed using high-resolution respirometry in mitochondria isolated from the plantarflexor muscles. (D) Representative electron micrographs of the tibialis anterior muscle of BL6 and BALB/c mice under control and ischemic (HLI day 7) conditions. Scale bars: 1 µm of 500 nm. (E) Mitochondrial respiratory function was not different between BL6 and BALB/c mice under nonsurgical conditions. (F) Oxygen consumption rate (OCR) in the presence of 10 mM glutatmate + 0.5 mM malate. (G) Complex I supported state 3 respiration in the presence of 10 mM glutatmate + 0.5 mM malate + 4mM ADP. (H) State 3 respiration supported by 10 mM glutatmate + 0.5 mM malate + 10 mM Succinate + 4mM ADP. (I) Complex II supported state 3 respiration was…

• 

  Figure 5. AAV-mediated expression of PFKFB3 decreases ischemic muscle necrosis and improves muscle function in BALB/c mice. BALB/c mice are known to respond poorly to hind limb ischemia, exhibiting significant muscle necrosis and — in some cases — limb loss. BALB/c mice were given intramuscular injections of AAV-GFP or AAV-PFKFB3 to the hind limb musculature at 1 × 1011 vg/muscle 3 weeks before HLI. (A and B) AAV expression was validated at both the mRNA (A; n = 3/group) and protein (B) level for PFKFB3. (C and D) AAV-PFKFB3 treatment significantly improved perfusion recovery (n = 10/group), measured by laser Doppler perfusion imaging. (E and F) AAV-PFKFB3–treated mice also displayed less limb necrosis (E, n = 10/group) and improved muscle histology (F). Scale bar: 200 µm. (G) Importantly, AAV-PFKFB3 also increased muscle force production, which was undetectable in all AAV-GFP mice. n = 10/group. **P < 0.01 versus control (AAV-GFP or DMSO) using 2-way ANOVA with Tukey’s post hoc for comparisons (D) and Student’s t test (A and G). Data in E were analyzed by nonparametric Mann-Whitney U test. Values are presented as mean ± SEM.

• 

  Figure 6. PFKFB3 expression increases glycolytic flux in skeletal muscle and endothelial cells in vitro, resulting in enhanced hypoxia tolerance and angiogenesis. Skeletal muscle cells (myotubes) and HUVECs (endothelial cells) were treated with control (AAV-GFP) and overexpression (AAV-PFKFB3) constructs. (A and B) Treatment was validated at both the mRNA (A, n = 3/group) and protein (B) level. (C and D) PFKFB3 overexpression increased both basal and maximal glycolytic flux in skeletal muscle cells (C, n = 7–8/group) resulting in improved cell survival/viability in hypoxia (D, n = 4/group). (E–G) PFKFB3 overexpression also increased both basal and maximal glycolytic flux in endothelial cells (E and F, n = 3–4/group), which resulted in enhanced endothelial cell tube formation (G, n = 3-4/group), an in vitro model of angiogenesis. Increased angiogenesis could be blocked by treatment with PFK15 (inhibitor of PFKFB3). **P < 0.01, ***P < 0.001, and ****P < 0.0001 versus control (AAV-GFP or DMSO) using ANOVA with Tukey’s post hoc for comparisons. φP < 0.05 for virus effect (E) using ANOVA (1-way in A and G, 2-way in C–E) with Tukey’s post hoc for comparisons. Values are presented as mean ± SEM.

• 

  Figure 7. Reactome enrichment in RNA sequencing data from muscle biopsies of peripheral arterial disease patients. Gene expression profiles were determined by whole genome sequencing of RNA isolated from muscle biopsy samples of the gastrocnemius. (A) Reactome enrichment analysis indicated that the most significant gene expression changes were related to bioenergetics (including mitochondria and glucose metabolism). (B) A heatmap of gene expression differences (log2 [fold change from healthy adult (HA)]) for the Reactome terms and genes associated with glucose metabolism. (C) mRNA fold-changes from HA in WTSS data set used for Reactome enrichment. (D) mRNA directional changes were validated by qPCR of selected genes. Statistical testing performed in C using 2-tailed Student’s t test. Values are presented as mean ± SEM.

• 

  Figure 8. PFKFB3 protein and muscle progenitor cell glycolytic flux are reduced in CLI patients. Western blotting was performed on gastrocnemius muscle lysates from age-matched healthy adults without PAD (n = 8) and severe PAD patients with critical limb ischemia (CLI, n = 16). (A) Western blot images for PFKFB3, GAPDH, and total protein on PVDF membrane. (B) Quantified densitometry of images in A. Western blotting was performed on primary muscle cell myoblast and differentiated myotube lysates from age-matched healthy adults without PAD (HA; n = 3) and severe PAD patients with critical limb ischemia (CLI, n = 3). (C) Western blot images for PFKFB3, GAPDH, and total protein on PVDF membrane. (D and E) Quantified extracellular acidification rate (Seahorse) assay under baseline and maximal stimulation conditions in myoblasts (D) and differentiated myotubes (E). ***P < 0.001 using 2-tailed Student’s t test. Values are presented as mean ± SEM.