Prevention of growth failure caused by glucocorticoid treatment

Sammanfattning: Children with chronic inflammation often suffer from growth failure and many of these children are treated with high doses of glucocorticoids (GCs). However, the underlying mechanisms of growth failure under conditions of chronic inflammation and the relative contributions of inflammation per se and commonly used GC treatment are still unknown. Existing treatments of chronic inflammatory conditions are often associated with various side including bone growth impairment. Therefore, new treatment strategies preventing growth failure in these children are ultimately needed. Humanin is a protective protein locally produced in various tissues with the capacity to defend cells from undesired cell death and many other cytotoxic agents. It has been reported that humanin exerts its anti-apoptotic action by interacting with Bax and interestingly, ablation of Bax in mice can protect against GC-induced bone growth impairment. This thesis aimed to investigate the molecular mechanisms of GC-induced growth failure under conditions of chronic inflammation and if humanin and its analogue HNG have the potential to prevent growth failure caused by GC treatment. In study I, we investigated the relative contributions of GC treatment and the inflammation per se to the growth failure commonly seen in conditions of chronic inflammation. In a transgenic mouse line (huTNFTg) over-expressing human TNF, we first observed that bone growth and growth plate chondrogenesis were suppressed by TNF overexpression. To investigate the effects of GCs on bone growth and growth plate chondrogenesis in a condition of chronic inflammation, huTNFTg mice were treated with saline or dexamethasone. Interestingly, dexamethasone then further suppressed bone growth, predominantly by accelerating apoptosis in the growth plate causing chondrocyte columns to become more disorganized. In study II, we assessed the potential for humanin and its analogues HNG to prevent growth failure caused by GCs. We discovered that humanin is expressed in growth plate cartilage and humanin over-expressing mice (HNtg) were resistant to GC-induced bone growth impairment. Treatment with the humanin analogue HNG prevented GC-induced bone growth retardation in cultured fetal rat metatarsal bones and also in vivo in mice. Mechanistic studies showed that humanin overexpression or HNG treatment prevented GC-induced apoptosis in vitro and in vivo, and humanin/HNG were found to be a novel regulator of Hh signalling. Importantly, HNG did not interfere with the anti-inflammatory effects of GCs. In study III, a link between humanin regulation and chronic inflammation per se was identified. We found that serum humanin levels were decreased in IBD children with known poor bone health. Mechanistic studies in ex vivo cultured human growth plate cartilage showed suppressed endogenous humanin expression in tissues exposed to IBD serum or the pro-inflammatory cytokine TNF. Interestingly, we also observed that the expression of PCNA, SOX-9 and TNF Receptor Associated Factor 2 (TRAF2) were decreased in human growth plates exposed to TNF. In conclusion, our results show that GCs and inflammation per se can suppress chondrogenesis and bone growth separately, and that humanin and its analogue HNG could could offer a novel strategy to rescue bone growth impairment caused by GC treatment in children. In addition, we propose a direct link between chronic inflammation and humanin regulation, a novel finding of potential clinical significance.