We have used mRNA differential display to compare gene expression in normal and GH receptor-deficient dwarf chickens, and report here the characterization of one differentially expressed gene, which shows significant sequence identity to the sulfotransferase gene family. Partial cDNA clones were isolated from a chicken liver cDNA library and an additional sequence was obtained using 5' rapid amplification of cDNA ends. A complete cDNA probe hybridizes to three transcripts (2.4, 2.0 and 1.45 kb) on Northern blots of chicken liver RNA, which differ in the length of the 3' untranslated region. All three transcripts are expressed at higher levels in normal vs dwarf chickens, as expected for a GH-regulated gene. The expression of this sulfotransferase mRNA was also detected in skeletal muscle, but not other tissues. The administration of GH to chickens increased the hepatic expression within 1 h, suggesting this sulfotransferase could be directly regulated by GH. Sulfotransferase activity, using estradiol or corticosterone as substrate, is detected in cells transfected with an expression vector containing the full-length cDNA. The sequence of this sulfotransferase does not show significant similarity with any subfamily of the sulfotransferases and its endogenous substrate is presently unknown. However, we speculate that GH activation of sulfotransferase activity could play a role in reducing concentrations of growth-antagonistic steroid hormones in GH target tissues. These results demonstrate the usefulness of differential display in this model system to identify genes that play a role in mediating GH action.

Abstract

The sex-linked dwarf (dwdw) chicken represents a valuable animal model for studying GH insensitivity and the consequence of mutations in the GH receptor (GHR) gene. We have recently reported undetectable hepatic GH-binding activity and an aberrantly sized transcript in a strain of dwdw chickens obtained from Arbor Acre Farms, Inc. (Glastonbury, CT, USA). Southern blot analysis of the chicken GHR (cGHR) gene revealed a restriction-fragment length polymorphism in HindIII and EcoRI digests of genomic DNA in this strain of dwdw chicken. In order to localize the molecular mutation, we analysed the gene structure and determined the complete sequence of the 3' untranslated region (3' UTR) of the normal cGHR. With the use of this information, we located a large deletion in the 3' end of the cGHR gene of the Connecticut (CT) strain of dwdw chicken. This deletion (1773 bp) contained 27 highly conserved amino acids of the 3' end of the coding region, the in-frame stop codon, a less frequently used poly(A) signal that is normally found 445 bp downstream of the stop codon, and a large portion of the 3' UTR. Because of this deletion, 27 novel amino acids were substituted and the open reading frame was extended for an additional 26 amino acids before reaching the transcriptional termination site. The predicted amino acid sequence of the novel carboxyl-terminus of the dwdw cGHR is largely hydrophobic with a polylysine tail, whereas the carboxyl-terminus of the wild-type (DwDw) cGHR is composed of hydrophilic amino acids. Western blot analysis using antisera directed against the extracellular domain of the cGHR confirmed that the mutant transcript was translated. These observations suggested that the mutation in the intracellular domain results in expression of a dysfunctional cGHR which could account for the phenotype of the CT dwdw chicken.

Journal of Endocrinology (1994) 142, 427–434

The purpose of this study was to determine the relationship between genetic selection for growth traits and tissue expression of the chicken growth hormone receptor (cGHR) gene. Two different populations of broiler chickens were studied. One population consisted of strain (S) 80, selected for 14 generations for high 9-week body weight (BW), and its progenitor, S90 (a 1950's strain). The second population consisted of S21, selected for 10 generations for high 4-week BW and low abdominal fat, and its progenitor S20 (a 1970's strain). Tissue (liver, fat, breast and leg muscle) and blood samples were collected from six birds/strain at 2-week intervals between 1 and 11 weeks of age. An RNase protection assay was developed to measure mRNA levels of full-length cGHR (3.2 and 4.3 kb) transcripts and chicken glyceraldehyde 3-phosphate dehydrogenase (for normalization) in total RNA prepared from tissue. Analysis of the area-under-curve (AUC) was used for strain comparisons of certain developmental profiles (BW, plasma hormones and tissue cGHR mRNA). The BW AUC showed that the growth rates are different (P < 0.05) among the four strains (S21 > S20 > S80 > S90). Both slow-growing strains (S90 and S80) had a higher (P < 0.05) plasma GH AUC than the two fast-growing strains (S20 and S21). The plasma T3 AUC was highest (P < 0.05) in S90 due to maintenance of higher T3 levels after 3 weeks of age. At 11 weeks of age, hepatic and plasma GH-binding activities were positively related to growth rate (S21 > S20 > S80 > S90). However, the developmental increase in cGHR mRNA in liver and fat was similar among these different populations of growth-selected broiler chickens. Steady-state levels of cGHR mRNA increased in a developmental manner in the liver (5-fold at 9 weeks of age) and abdominal fat (4.5-fold at 11 weeks of age) of all strains. In contrast, there was no developmental increase or strain difference in cGHR mRNA levels in breast and leg muscle. There is a discrepancy between GH-binding activity in liver and plasma, which is different among strains, and steady-state levels of tissue cGHR mRNA which are similar among strains. These observations suggest that the cGHR is under translational or post-translational regulation which would determine the amount of cGHR protein available for GH binding.