The interaction of GH, interleukin (IL)-6 and glucocorticoids is likely to be important in regulating the GH-insulin-like growth factor (IGF)-I axis. The signalling cascades activated by GH and IL-6 appear to be very similar, as demonstrated by studies using overexpression of the receptor and other components of the Jak-Stat and mitogen-activated protein (MAP) kinase pathways. Here we show that the human embryonic kidney cell line 293 (HEK293) expresses GH and IL-6 receptors endogenously. To determine which specific pathways might be activated by the two cytokines, at physiological levels of all components, we studied GH and IL-6 mediated signal transduction both under basal conditions and in the presence of overexpressed receptors and Stat proteins. Our results suggest a receptor specificity of Jak2 for GH receptors, and Jak1 for IL-6 receptors. Stat activation in response to GH and IL-6 was determined by reporter gene induction. Both GH and IL-6 were able to induce the reporter gene containing the Stat5 responsive element (LHRE) but the IL-6 response appeared to be mediated mainly through Stat3 activation. In contrast, the reporter gene containing the Stat3 responsive element (SIE) was IL-6 specific. The levels of gene induction by GH and IL-6 were not altered by the co-stimulation with GH and IL-6, suggesting that there is little cross-talk at the Jak-Stat activation level between the two cytokines. Neither GH nor IL-6 activated the MAP-kinase responsive serum response element (SRE), unless GH receptors or gp130 were overexpressed. Transfection of Stat3 or Stat5 expression vectors enhanced the response to GH and IL-6. Stimulation with dexamethasone synergistically enhanced GH activation of the LHRE reporter gene but had no effect on the IL-6 activation of the same reporter or on the SIE reporter gene. Thus, our studies suggest that while each cytokine, GH and IL-6, may activate various members of the Jak-Stat pathway in overexpression studies, specific activation of Stat3 by IL-6 and of Jak2 and Stat5 by GH can be observed in HEK293 cells and that in this system the synergistic effect of dexamethasone appears specific for Stat5.
S von Laue, J Finidori, M Maamra, XY Shen, S Justice, PR Dobson and RJ Ross
M Maamra, A Milward, H Zarkesh Esfahani, L P Abbott, L A Metherell, M O Savage, A J L Clark and R J M Ross
Growth hormone insensitivity syndrome (GHIS) has been reported in a family homozygous for a point mutation in the GH receptor (GHR) that activates an intronic pseudoexon. The resultant GHR (GHR1–656) includes a 36 amino-acids insertion after residue 207, in the region known to be important for homodimerization of GHR. We have examined the functional consequences of such an insertion in mammalian cells transfected with the wild type (GHRwt) and mutated GHR (GHR1–656). Radio-ligand binding and flow cytometry analysis showed that GHR1–656 is poorly expressed at the cell surface compared with GHRwt. Total membrane binding and Western blot analysis showed no such difference in the level of total cellular GHR expressed for GHR1–656 vs GHRwt. Immunofluorescence showed GHR1–656 to have different cellular distribution to the wild type receptor (GHRwt), with the mutated GHR being mainly perinuclear and less vesicular than GHRwt. Western blot analysis showed GH-induced phosphorylation of Jak2 and Stat5 for both GHR1–656 and GHRwt, although reduced Stat5 activity was detected with GHR1–656, consistent with lower levels of expression of GHR1–656 than GHRwt at the cell surface. In conclusion, we report that GHIS, due to a 36 amino-acids insertion in the extracellular domain of GHR, is likely to be explained by a trafficking defect rather than by a signalling defect of GHR.
TS Johnson, M O'Leary, SK Justice, M Maamra, SH Zarkesh-Esfahani, R Furlanetto, VR Preedy, CJ Hinds, AM El Nahas and RJ Ross
GH treatment during critical illness and sepsis may increase mortality. A family of negative regulators of cytokine signalling, the suppressors of cytokine signalling (SOCS), have been characterised. SOCS provide a mechanism for cross-talk between the cytokine receptors, including GH. Here, we have investigated the impact of nutrition and GH treatment on GH receptor, SOCS1, SOCS-2, SOCS-3 and cytokine-inducible SH2-containing protein (CIS) hepatic mRNA expression in a rat model of sepsis, caecal ligation and puncture (CLP). Four groups of rats were studied: control (food given ad libitum, n=7), CLP only (n=8), CLP and total parenteral nutrition (TPN) (n=9), and CLP, TPN and GH (n=10). CLP rats underwent surgery and 18 h later received saline or TPN or TPN+GH for 6 h before they were killed. Serum IGF-I levels were lower in all CLP groups (P<0.001). The combination of TPN and GH treatment increased IGF-I levels compared with the saline-treated CLP rats (P<0.01), but IGF-I levels remained lower than control animals (P<0.001). GH receptor and GH-binding protein expression in liver was reduced in animals subjected to CLP and was unaffected by nutrition or GH treatment. Hepatic SOCS-1 was detectable in normal rats, induced in all CLP animals but was unaffected by nutrition and GH. Hepatic SOCS-2 expression was difficult to detect in normal and CLP rats but was greatly induced in CLP rats treated with GH. Hepatic SOCS-3 expression was only just detectable in the control group but was elevated in all CLP groups and unaffected by nutrition and GH. Hepatic CIS expression was difficult to detect in normal rats, was not induced by CLP but was induced by both nutrition and GH. In conclusion, CLP induced low IGF-I levels associated with increased expression of SOCS-1 and SOCS-3, both of which are known to inhibit GH receptor signalling. GH induced SOCS-2 and CIS in the CLP rat despite resistance with respect to IGF-I generation, and parenteral feeding induced CIS in the CLP rat. Thus, there is potential for a complex interaction between GH and cytokine signalling at the level of SOCS expression whereby the inflammatory response may alter GH signalling and GH may influence the inflammatory response.