11β-Hydroxysteroid dehydrogenase type 1 within muscle protects against the adverse effects of local inflammation

Abstract

Muscle wasting is a common feature of inflammatory myopathies. Glucocorticoids (GCs), although effective at suppressing inflammation and inflammatory muscle loss, also cause myopathy with prolonged administration. 11β-Hydroxysteroid dehydrogenase type 1 (11β-HSD1) is a bidirectional GC-activating enzyme that is potently upregulated by inflammation within mesenchymal-derived tissues. We assessed the regulation of this enzyme with inflammation in muscle, and examined its functional impact on muscle. The expression of 11β-HSD1 in response to proinflammatory stimuli was determined in a transgenic murine model of chronic inflammation (TNF-Tg) driven by overexpression of tumour necrosis factor (TNF)-α within tissues, including muscle. The inflammatory regulation and functional consequences of 11β-HSD1 expression were examined in primary cultures of human and murine myotubes and human and murine muscle biopsies ex vivo. The contributions of 11β-HSD1 to muscle inflammation and wasting were assessed in vivo with the TNF-Tg mouse on an 11β-HSD1 null background. 11β-HSD1 was significantly upregulated within the tibialis anterior and quadriceps muscles from TNF-Tg mice. In human and murine primary myotubes, 11β-HSD1 expression and activity were significantly increased in response to the proinflammatory cytokine TNF-α (mRNA, 7.6-fold, p < 0.005; activity, 4.1-fold, p < 0.005). Physiologically relevant levels of endogenous GCs activated by 11β-HSD1 suppressed proinflammatory cytokine output (interkeukin-6, TNF-α, and interferon-γ), but had little impact on markers of muscle wasting in human myotube cultures. TNF-Tg mice on an 11β-11β-HSD1 knockout background developed greater muscle wasting than their TNF-Tg counterparts (27.4% less; p < 0.005), with smaller compacted muscle fibres and increased proinflammatory gene expression relative to TNF-Tg mice with normal 11β-HSD1 activity. This study demonstrates that inflammatory stimuli upregulate 11β-HSD1 expression and GC activation within muscle. Although concerns have been raised that excess levels of GCs may be detrimental to muscle, in this inflammatory TNF-α-driven model, local endogenous GC activation appears to be an important anti-inflammatory response that protects against inflammatory muscle wasting in vivo. © 2016 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.

Keywords: 11β-HSD1; animal models; chronic inflammation; glucocorticoids; muscle wasting.

Figure 1

(A) Clinical scoring of behaviour, mobility, weight loss, mouse grimace, evidence of joint inflammation and duration of joint swelling (as outlined in supplementary material, Table S2) in TNF‐Tg and WT littermate controls between 5 and 9 weeks of age (N = 6 per group). (B) mRNA abundance (AU) of 11β‐HSD1 determined by RT‐qPCR in tibialis anterior muscles isolated from TNF‐Tg and WT controls at 9 weeks (N = 6 per group). (C) Rate of corticosterone generation by ex vivo biopsies of whole quadriceps muscles isolated from TNF‐Tg and WT controls following incubation for 16 h with DHC, determined by scanning thin‐layer chromatograms (N = 6 per group). Values are expressed as mean ± SE. Statistical significance was determined with Student's unpaired t‐test. *p < 0.05, **p < 0.005, ***p < 0.0005.

Figure 2

(A) Representative images of undifferentiated primary myoblasts isolated from human quadriceps. Primary myoblasts syncytialize over a 5‐day differentiation period in selective medium, forming primary myotubes in vitro. (B) Fold change in mRNA expression of the muscle differentiation gene MYOG in undifferentiated human myoblasts and mature mytotubes at day 5 after differentiation, determined by RT‐qPCR (n = 3 per group). (C, D) mRNA expression of the 11β‐HSD1 gene (C) and the rate of generation of cortisol from cortisone (D) in differentiated human myotubes over a period of 16 h following 24 h of pretreatment with TNF‐α (10 ng/ml), determined by RT‐qPCR and scanning thin layer chromatograms, respectively (n = 3 per group). (E) mRNA expression of the 11β‐HSD1 gene in differentiated human myotubes following treatment with TNF‐α (10 ng/ml) for 2 h in the presence or absence of the selective NF‐κB signalling inhibitor Mln‐4924 at 2 µm (n = 3 per variable). (F) Paraffin sections of quadriceps muscles isolated from OA patients were stained for 11β‐HSD1 by immunohistochemistry, and counterstained with Gill's haematoxylin (N = 3) (scale bars: 50 µm). (G) Gene expression (AU) of 11β‐HSD1 was measured by RT‐qPCR in ex vivo OA quadriceps biopsies following incubation with TNF‐α (10 ng/ml) for either 0, 3, 6 or 16 h (N = 3 per variable). Values are expressed as mean ± SE. Statistical significance was determined with using Student's unpaired t‐test (B–D) and one‐way anova with a Dunnett post hoc analysis (E, G). **p < 0.005, ***p < 0.0005.

Figure 3

(A, B) Gene expression (AU) of Il6 and Gilz in murine primary cultures of differentiated myotubes isolated from WT quadriceps biopsies (n = 3 per variable). Cultures were incubated for 16 h with either the inactive corticosterone precursor DHC (100 nmol/l), the selective 11β‐HSD1 inhibitor LJ2 (1000 nmol/l), or a combination of the two. (C–G) Gene expression (AU) of IL6 and GILZ in human primary cultures of differentiated myotubes isolated from quadriceps muscles (n = 3 per variable). Cultures were stimulated with either vehicle (C, D) or TNF‐α (E, F) at 10 ng/ml for 48 h (n = 3 per variable). Cultures were then maintained in medium containing either vehicle control, cortisol (100 nmol/l), the inactive corticol precursor cortisone (100 nmol/l), the selective 11β‐HSD1 inhibitor LJ2 (1000 nmol/l) or a combination of both cortisone and inhibitor for 16 h. (G), The decrease in IL6 mRNA level relative to respective vehicle controls following incubation with cortisone in the vehicle and TNF‐α‐prestimulated groups was then measured (n = 3 per variable). (H–J) Concentrations of IL‐6, TNF‐α and IFN‐γ within 48‐h conditioned medium of primary human myotubes treated with either control, cortisol (100 nmol/l), cortisone (100 nmol/l), or LJ2 (1000 nmol/l) and cortisone, determined by multiplex cytokine analysis (n = 3 per variable). Values are expressed as mean ± SE. Statistical significance was determined with one‐way anova with a Dunnett post hoc analysis. *p < 0.05, **p < 0.005. LJ2 = PF‐877423.

Figure 4

Levels of mRNA in primary cultures of differentiated myotubes isolated from quadriceps muscle biopsies from WT mice (A–C) and human quadriceps biopsies (F–H) (n = 3 per variable), determined by RT‐qPCR. To induce 11β‐HSD1 expression, human primary cultures were pretreated with TNF‐α (10 ng/ml) for 48 h prior to a 12‐h washout. Cells were then incubated for 16 h with either control, active cortisol (100 nmol/l), or its inactive precursor cortisone (100 nmol/l). In murine culture, cells were treated for 24 h with either control medium, the inactive corticosterone precursor DHC (100 nmol/l), or DHC in combination with the selective 11β‐HSD1 inhibitor LJ2 (1000 nmol/l). Fibre width and cell viability were determined with Image J analysis software and MTT assay, respectively, in murine (D, E) and human (I, J) primary myotubes following incubation with normal control medium and DHC (100 nmol/l)‐containing medium over a period of 5 days (n = 3 per variable). Values are expressed as mean ± SE. Statistical significance was determined with one‐way anova with a Dunnett post hoc analysis. *p < 0.05, **p < 0.005.

Figure 5

(A) Clinical scoring of behaviour, mobility, weight loss, mouse grimace, evidence of joint inflammation and duration of joint swelling (as outlined in supplementary material, Table S2) in TNF‐Tg/11βKO mice versus TNF‐Tg controls between 5 and 9 weeks of age (N = 6 per group). (B) Rate of corticosterone generation by ex vivo biopsies of whole quadriceps muscles isolated from WT, TNF‐Tg, 11βKO and TNF‐Tg mice on an 11βKO background following incubation for 16 h with DHC, as determined by scanning thin‐layer chromatograms (N = 6 per group). (C) Total quadriceps muscle weight normalized to total body weight in WT, TNF‐Tg, 11βKO and TNF‐Tg mice on an 11βKO background (N = 6 per group). (D) Representative images of quadriceps muscle sections taken from WT and 11βKO mice on either control or TNF‐Tg backgrounds stained with Gill's haematoxylin and eosin (scale bars: 50 µm) (N = 6 per group). (E–G) Average fibre size (AU), and quantification of small (<30 µm) and large (≥20 µm) fibres (AU) determined with Image J in paraffin sections of quadriceps muscles isolated from WT mice, TNF‐Tg mice, and TNF‐Tg mice on an 11βKO background (N = 6 per group). (H–J) Levels of mRNA (ΔCt) for human TNF‐α and the murine proinflammatory genes Il6 and Tnf in WT mice, TNF‐Tg mice and TNF‐Tg mice on an 11βKO background at 9 weeks (N = 6 per group). Values are expressed as mean ± SE. Statistical significance was determined with one‐way anova with a Dunnett post hoc analysis. ND, not detected; NS, not significant. *p < 0.05, **p < 0.005, ***p < 0.0005.