HIF prolyl hydroxylase inhibition protects skeletal muscle from eccentric contraction-induced injury

Abstract

Background: In muscular dystrophy and old age, skeletal muscle repair is compromised leading to fibrosis and fatty tissue accumulation. Therefore, therapies that protect skeletal muscle or enhance repair would be valuable medical treatments. Hypoxia-inducible factors (HIFs) regulate gene transcription under conditions of low oxygen, and HIF target genes EPO and VEGF have been associated with muscle protection and repair. We tested the importance of HIF activation following skeletal muscle injury, in both a murine model and human volunteers, using prolyl hydroxylase inhibitors that stabilize and activate HIF.

Methods: Using a mouse eccentric limb injury model, we characterized the protective effects of prolyl hydroxylase inhibitor, GSK1120360A. We then extended these studies to examine the impact of EPO modulation and infiltrating immune cell populations on muscle protection. Finally, we extended this study with an experimental medicine approach using eccentric arm exercise in untrained volunteers to measure the muscle-protective effects of a clinical prolyl hydroxylase inhibitor, daprodustat.

Results: GSK1120360A dramatically prevented functional deficits and histological damage, while accelerating recovery after eccentric limb injury in mice. Surprisingly, this effect was independent of EPO, but required myeloid HIF1α-mediated iNOS activity. Treatment of healthy human volunteers with high-dose daprodustat reduced accumulation of circulating damage markers following eccentric arm exercise, although we did not observe any diminution of functional deficits with compound treatment.

Conclusion: The results of these experiments highlight a novel skeletal muscle protective effect of prolyl hydroxylase inhibition via HIF-mediated expression of iNOS in macrophages. Partial recapitulation of these findings in healthy volunteers suggests elements of consistent pharmacology compared to responses in mice although there are clear differences between these two systems.

Keywords: Eccentric injury; HIF activation; Prolyl hydroxylase; Protection; Skeletal muscle.

Fig. 1

Prolyl hydroxylase inhibitors protect skeletal muscle from exercise-induced injury. a Maximal limb force in mice before and after eccentric muscle exercise. Mice were treated with vehicle or GSK360 daily, starting 10 min after eccentric exercise. Limb force was measured for 28 days. Uninjured limb force included as a control. Data are normalized to pre-exercise force (n = 8). b Representative images of immunohistochemical staining for albumin (red), laminin (green), and nuclei (blue) in gastrocnemius muscle 3 days after eccentric exercise (n = 4/group). Upper panels: low power images of full cross section of gastrocnemius. Lower panels: higher magnifications of the rectangular areas above (scale bars = 250 μM). Arrows: albumin-positive myofibers. Asterisks: albumin-negative fibers. c Quantification of albumin-positive fibers measured as a proportion of total fibers (n = 4/group). d Maximal limb force in mice before and after eccentric muscle damage. Mice were treated daily with vehicle or GSK360 24 h before, at the time of injury, or 24 h after eccentric exercise. Limb force was measures for 8 days (n = 8). Data shown as the mean ± SEM. Two-way ANOVA followed by Holm-Sidak’s test corrected for multiple comparisons; *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle-dosed cohort (unless otherwise indicated)

Fig. 2

Prolyl hydroxylase inhibitors exert their protective effects independently of EPO. a Serum EPO concentration and b Serum VEGF concentration after oral administration of 10 mg/kg GSK360 in mice (n = 8). c Force deficit 24 h after eccentric injury. Mice were treated daily with rEPO, with and without an EPO neutralizing antibody (n = 8). d Force deficit 24 h after eccentric injury of mice co-treated with EPO neutralizing antibody and GSK360 (n = 8). e Force deficit 24 h after eccentric injury of mice co-treated with soluble EPO receptor and GSK360 (n = 8). f Western blot of HIF1α protein levels from muscle lysates 1 h and 24 h after limb injury. GAPDH used as a loading control

Fig. 3

Prolyl hydroxylase inhibitors exert their protective effects via macrophage HIF1α and iNOS. a Force deficit 24 h after eccentric injury in indicated strains of myeloid HIF1α KO mice administered vehicle or GSK360 (n = 8). b Force deficit 24 h after eccentric injury in myeloid HIF1α KO mice with exogenous BMDM (either HIF1α KO or WT) injected intramuscularly following eccentric injury and treated with vehicle or GSK360 (n = 8). c FACS analysis of PKH67-green labeled myeloid populations recovered from injured muscles 24 h after injection and treatment. d Force deficit 24 h after eccentric injury in WT mice treated with vehicle, iNOS inhibitor 1400 W and/or GSK360. Right side; force deficit 24 h after eccentric injury in iNOS KO mice administered vehicle or GSK360 (n = 8). e iNOS gene expression measured by qPCR. Bone marrow-derived macrophages (BMDM) were cultured and polarized to promote classical M1 or M2 activation or left inactivated (M0) and treated with vehicle or GSK360 for 24 h. f iNOS protein expression in M1-polarized macrophages treated with GSK360 for 24 h before protein analysis by Western blot. Rab5 was used as a loading control. All data in the figure are shown as the mean ± SEM. Two-way ANOVA followed by Holm-Sidak’s test corrected for multiple comparisons; *P < 0.05, **P < 0.01, ***P < 0.001 versus the vehicle dosed cohort (unless otherwise indicated)

Fig. 4

Prolyl hydroxylase inhibitor daprodustat reduces markers of muscle damage in healthy male volunteers after eccentric exercise. a Clinical study design. b Maximum voluntary force at an arm angle of 90^o before and after eccentric exercise in two separate studies. Boxed area represents a 75% range of values; central line indicates the median and the cross the mean value. Individual values are displayed to the left (top figure, placebo n = 14, daprodustat 5 mg n = 12; bottom figure, placebo n = 15, daprodustat 50 mg n = 15). c Serum creatine kinase before and after eccentric exercise in the two separate studies. d Heatmap and hierarchical clustering of the top 16 serum proteins changed between pre-exercise and 3 days post exercise and the effect of 50 mg daprodustat on their levels. The first two columns display fold changes pre- vs. postexercise for daprodustat-treated and placebo-treated participants, respectively. The third column shows daprodustat fold change treated vs. placebo-treated samples postexercise. Protein names in bold denote a significant association with muscle-related phenotypes and/or pathways (each group n = 15). All data in the figure are shown as the mean ± SEM. Repeat measure ANOVA for multiple comparisons; ***P < 0.001)