125I-Labelled human GH (hGH) was injected i.v. to male rats and its subcellular distribution in the hepatocyte was examined using fractionation techniques. Uptake into liver homogenates was maximal by 15 min after injection and represented 24% of the injected radioactivity; it was markedly inhibited by coinjection of native hGH. 125I-Labelled hGH taken up by the liver underwent a time-dependent translocation process. The peak of specific labelling of plasma membranes occurred at 3 min whereas later on the radioactivity was concentrated in low-density structures present in Golgi-endosome fractions. To characterize the ligand-associated structures better, endosome-enriched fractions were prepared from a microsomal fraction by isopycnic centrifugation in a sucrose gradient and a Nycodenz gradient. The radioactivity was in one peak with a median density of 1.096 g/cm3 in the Nycodenz gradient fractions. The peak of radioactivity was distinct from that of galactosyltransferase activity which appeared at a median density of 1.114 g/cm3. The labelled material eluted from the various subcellular fractions appeared as intact hGH. Upon in-vivo interaction with male rat hepatocytes, 125I-Labelled hGH was internalized with a sequential association with plasma membranes and endocytic structures distinct from Golgi elements.
The GH receptor (GHR) is a member of the cytokine receptor family. Short isoforms resulting from alternative splicing have been reported for a number of proteins in this family. RT-PCR experiments, in human liver and cultured IM-9 cells, using primers in exon 7 and 10 of the GHR, revealed three bands reflecting alternative splicing of GHR mRNA: the predicted product at 453 bp and two other products at 427 and 383 bp. The 427-bp product (GHR1-279) utilized an alternative 3′-acceptor splice site 26 bp downstream in exon 9; the predicted C-terminal residues are six frameshifted exon 9 codons ending in an inframe stop codon. The 383-bp product (GHR1-277) resulted from skipping of exon 9; the predicted C-terminal residues are three frameshifted exon 10 codons ending in an in-frame stop codon. RNase protection experiments confirmed the presence of the GHR1-279 variant in IM-9 cells and human liver. The proportion of alternative splice to full length was 1–10% for GHR1-279 and less than 1% for GHR1-277. The function of GHR1-279 was examined after subcloning in an expression vector and transient transfection in 293 cells. Scatchard analysis of competition curves for [125I]-hGH bound to cells transfected either with GHR full length (GHRfl) or GHR1-279 revealed a 2-fold reduced affinity and 6-fold increased number of binding sites for GHR1-279. The increased expression of GHR1-279 was confirmed by cross-linking studies. The media of cells transfected with GHR1-279 contained 20-fold more GH-binding protein (GHBP) than that found in the media of cells transfected with the full-length receptor. Immunoprecipitation and Western blotting experiments, using a combination of antibodies directed against extracellular and intracellular GHR epitopes, demonstrated that GHRfl and GHR1-279 can form heterodimers and that the two forms also generate a 60-kDa GHBP similar in size to the GHBP in human serum. Functional tests using a reporter gene, containing Stat5-binding elements, confirmed that while the variant form was inactive by itself, it could inhibit the function of the full-length receptor. We have demonstrated the presence of a splice variant of the GHR in human liver encoding a short form of the receptor similar in size to a protein previously identified in human liver and choroid plexus. Expression studies in 293 cells support the hypothesis that while the expression of the splice variant accounts for only a small proportion of the total GHR transcript, it produces a short isoform that modulates the function of the full-length receptor, inhibits signaling, and generates large amounts of GHBP. The differential expression of GHR receptor short forms may regulate the production of GHBP, and truncated receptors may act as trans-port proteins or negative regulators of GHR signaling.
The GH receptor (GHR) is a member of the cytokine/hematopoietic growth factor family, and protein tyrosine phosphorylation has been implicated in the signaling cascade of these receptors. It was recently shown that the tyrosine kinase JAK2 is associated with the GHR. GH induces the activation of JAK2, which phosphorylates itself and the receptor. Mitogen-activated protein (MAP) kinase activation and transcriptional stimulation of specific genes, such as Spi 2.1, have also been reported to be induced by GH. To identify functionally important regions in the cytoplasmic domain of the GHR, we compared the actions of the wild-type receptor, two truncated mutants, and one internal deletion mutant (similar to the intermediate Nb2 form of the PRL receptor) in transfectants of the Chinese hamster ovary cell line. A region of 46 amino acids adjacent to the membrane was found to be sufficient for activation of both JAK2 and MAP kinases. This region contains a proline-rich sequence (box 1) conserved in the cytokine receptor family that is important for signal transduction. For transcriptional activity, the C-terminal region of the GHR is required, and we found that the last 80 terminal residues contain sequences allowing activation of the Spi 2.1 promoter. Tyrosine phosphorylation of the receptor also requires the C-terminal portion of the GHR cytoplasmic domain, and we found that GHR tyrosine phosphorylation appears to be linked to activation of the Spi 2.1 transcription pathway. Thus, the GHR could be composed of at least 2 functional regions: the 46 proximal amino acids required for activation of JAK2 and sufficient to stimulate the MAP kinase pathway, and an additional carboxy-terminal region necessary for transcriptional activation.
A single form of GH receptor (GHR) messenger RNA (mRNA) of 4.5 kilobases, encoding the full-length GHR, has been found in man. To measure the absolute number of mRNA molecules encoding the GHR in human tissues, we developed a quantitative polymerase chain reaction assay. An internal control RNA was constructed by inserting a 50-basepair fragment of the rat PRL receptor complementary DNA into a portion of the human GHR complementary DNA. The internal control RNA and the target mRNA were amplified together with the same set of primers. Twenty-four cycles of amplification were used to satisfy an exponential phase of amplification. It was possible to detect as few as 500 molecules of target mRNA/micrograms total RNA. In 3 liver samples obtained from normal donors at the time of transplant, the amount of GHR mRNA ranged from 0.5 +/- 0.1 to 1.4 +/- 0.4 x 10(6) molecules/micrograms total RNA. These results were confirmed by slot blot analysis of the same samples. The number of receptor transcripts did not appear to be correlated with the receptor-binding capacity found in the 3 liver samples. In 7 muscle biopsies, GH receptor mRNA varied between 4.0 +/- 0.4 and 34.6 +/- 1.4 x 10(4) molecules/micrograms total RNA. This technique allows direct measurement of GHR gene expression in human tissues and represents a valuable tool, particularly for tissues such as muscle, in which the receptor protein cannot be measured using conventional binding assays.
In cirrhosis, as in other conditions of protein catabolism, there is a state of acquired GH resistance, as defined by high circulating GH levels with low insulin-like growth factor I levels. However, patients with end-stage liver failure respond to supraphysiological doses of GH with an increase in circulating insulin-like growth factor I levels. The present study represents a detailed analysis of GH receptor (GHR) expression in cirrhotic liver from 17 patients with end-stage liver disease. Specific binding of labeled GH was identified in all cirrhotic livers studied. The binding affinity for the GHR was similar in cirrhotic and normal livers, but the number of binding sites per mg protein of liver membrane was variable in both normal and cirrhotic liver, although it were generally lower in cirrhotic liver. GHR expression was identified in cirrhotic liver by Northern blotting, RT-PCR, and ribonuclease protection assay. On Northern blotting, a single transcript of 4.8 kb was identified in normal and cirrhotic tissues. RT-PCR identified expression of both full-length GHR and a truncated form of the GHR; this was confirmed by ribonuclease protection assay. In situ hybridization and immunohistochemistry confirmed the expression of GHR in regenerating hepatocytes and isolated cells in fibrous tissue. In conclusion, 1) the low level of GHR in cirrhotic liver may contribute to the acquired GH resistance found in cirrhotic patients; 2) the reduced expression of both full-length and truncated GHR is compatible with the low level of GH-binding protein found in cirrhosis, as this truncated receptor has previously been reported to generate large amounts of GH-binding protein; and 3) the demonstration of GH binding to cirrhotic liver explains why these patients with GH resistance may still respond to supraphysiological doses of GH.
The plasma GH response to a single iv bolus dose of 2 μg/kg BW synthetic GHRH-(1–44)NH2 was evaluated in 13 prepubertal children with thalassemia major (mean age, 7.6 ± 0.8 yr) with growth retardation and in 15 prepubertal children with nonendocrine short stature. All of the patients showed a significant increase in plasma GH concentration, with a mean peak of 31.4 ± 4.5 μg/L at 15 min (P < 0.001 vs. basal values; range, 18.4–65 μg/L) after GHRH, which was not different from that of the control group of idiopathic short stature children (40.1 ± 3.4 μg/L; range, 21–65.4 μg/L). All but 1 of the thalassemic patients had a normal GH response to the arginine-insulin stimulation test. The mean plasma insulin-like growth factor-I level was low (0.12 ± 0.05 U × 103/L; range, <0.02–0.61 U × 103/L). Analysis of these results as well as previously reported data indicating that older thalassemic patients have an impaired GH response indicates that there may be an age-related pituitary and/or hypothalamic dysfunction in thalassemic children. This study also confirms that the insulin-like growth factor-I decrease occurs before any alteration in GH secretion. These changes might play a role in the early growth retardation that occurs in these patients.
The growth hormone-binding protein (GHBP) which circulates in plasma is a soluble short form of the membrane growth hormone receptor (GHR). In rats and mice, GHR and GHBP originate from two alternatively spliced mRNAs (4.5 and 1.2 kb). In human and rabbit tissues, a single predominant mRNA of 4.5 kb was detected and it was hypothesized that GHBP could be produced by proteolytic cleavage of the GHR. Using gel filtration and HPLC, we have detected a high level of GH binding activity in media of cells transfected with rabbit GHR cDNA. The [125I]hGH-GHBP complex eluted at the same time as the plasma complex and both the binding affinity and specificity of the BP were comparable to that of rabbit plasma. Immunoprecipitation experiments and Western blots confirmed that GHBP in the media of transfected cells was a 55 kDa protein related to the extracellular domain of the GHR. In contrast, no BP was detected in the media of cells transfected with the cDNA encoding the rat GHR. These results strongly suggest that, in rabbit and probably in man, the GHBP could, at least in part, be produced by proteolytic cleavage of the GHR.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)