| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Research Laboratory and Endocrinology Service, Garrahan Pediatric Hospital, Buenos Aires, Argentina 1245
aaddress correspondence to this author at: Coordinación de Investigación, Hospital de Pediatria Garrahan, C. de los Pozos 1881-P2, Buenos Aires, Argentina 1245; fax 5411-4308-5325, e-mail abelgo{at}elsitio.net
The clinical appearance and genetic basis of familial isolated growth hormone deficiency (IGHD) are heterogeneous and are associated with at least four types of Mendelian disorders: two forms with autosomal recessive inheritance (IGHD type 1A and 1B), one with autosomal dominant inheritance (IGHD type 2), and one with X-linked inheritance (IGHD type 3) (1)(2).
Recently, Deladoëy et al. (3) reported a new form of IGHD type 2 in four unrelated families. These patients carried an ARG183HIS (R183H) GH-1 mutation. On the basis of in vivo and in vitro experiments, the authors reported that the R183H mutant GH peptide severely impaired GH-regulated secretion. Furthermore, this GH mutant was found to have an effect on GH receptor/GH-binding protein (GHR/GHBP) transcription identical to that of the 22-kDa GH when tested in a human hepatoma cell line (4).
Human GH represents a family of proteins rather than a single hormone. Indeed, GH is among the more heterogeneous polypeptide hormones, and the proportions of immunoreactive forms in plasma vary with time after a GH secretory pulse (5). The simultaneous use of an immunoassay that detects most serum GH isoforms (total GH) and one specific for the 22-kDa GH form makes it possible to uncover changes in the pattern of secretion of GH isoforms. Moreover, discrepancies between GH immunologic and biological activities have been reported (6). It has therefore been proposed that the ratio of non-22-kDa GH to 22-kDa GH isoforms may have important implications for normal and abnormal growth (7).
We studied two nonrelated patients from Argentina with severe short stature and IGHD type 2 attributable to the R183H GH-1 gene mutation and found that the evaluation of GH response to provocative tests might be misleading in these patients, depending on the type of assay used to assess GH secretion.
Serum GH was measured by two commercial immunoassays: a polyclonal IRMA, which uses WHO International Reference Preparation (IRP) human GH for RIA 66/217 (SER 66/217); and DELFIA, a time-resolved immunofluorometric assay, which uses WHO IRP for human GH for bioassay 80/505 (DELFIA 80/505). The intra- and interassay CVs were 5% (n = 90) and 7% (n = 90), respectively, for the SER 66/217 assay (range, 0.69–17.0 µg/L) and 3.3% (n = 90) and 4.7% (n = 90), respectively, for the DELFIA assay (range, 0.25–10.9 µg/L). As reported previously, the cutoff values for provocative tests in children differed markedly depending on the assay: 10 µg/L for the SER 66/217 and 4.81 µg/L for the DELFIA 80/505 (8). One reason for this discrepancy is that the SER 66/217 uses an antibody recognizing multiple GH epitopes present in several molecular forms in serum, whereas the DELFIA 80/505 is specific for 22-kDa GH (5). Indeed, cross-reactivity of 20-kDa GH in the 22-kDa GH assay was <0.001% (9). Conversion factors have been proposed to calculate equivalent values between polyclonal and monoclonal assays for serum concentrations in adults (10) and children (8). We propose designating SER 66/217 serum values as serum "total" GH and DELFIA 80/505 values as serum 22-kDa GH. We studied GH secretion, using pharmacologic stimulation with arginine and clonidine, as described previously (11)
Serum GH bioactivity was analyzed in the Nb2 bioassay according to a modification of the method published by Radetti et al. (12), using the CellTiter 96® AQueous Non-Radioactive Cell Proliferation Assay (Promega Corporation).
Intra- and interassay CVs were 11% (n = 10) and 21% (n = 16), respectively, for the GH bioassay (range, 0.5–39.1 µg/L). Normal response of bioactive GH to arginine and clonidine tests was defined in 21 control children (14 males and 7 females) 0.48–13.8 years of age with moderate short stature [-0.75 ± 1.32 standard deviation score (SDS)] and normal 22-kDa GH responses to the tests. Mean (SD) GH response was 10 (3.1) µg/L (95% confidence interval, 4.02–16.0 µg/L). The cutoff for bioactive GH response was set at the lower limit of the 95% confidence interval (mean ± 1 SD x 1.96) of the control group. This value was 4 µg/L.
Serum insulin-like growth factor-I (IGF-I) was measured with the IGF-I generation test as described previously (13). Response was considered normal when the increase in serum IGF-I was 
15 µg/L (14)(15). It was also compared with that of nine prepubertal patients (median age, 6.3 years; range, 3.7–11.2 years) with idiopathic GH deficiency.
We sequenced GH-1 (16). The gene was specifically PCR-amplified with use of sense and antisense primers corresponding to nucleotides 5101–5136 (gh1) and the complement of nucleotides 7255–7226 (gh2). The resulting GH-1 PCR product (2155 bp) was used as template for a nested PCR. For a description of the primer composition, see the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol49/issue6/.
We studied two prepubertal girls (P1 and P2) who presented severe short stature (below -3 SDS) and poor growth rate. Their main anthropometric characteristics are listed in the Data Supplement. They were monitored for 1 and 6 years, respectively, without any improvement in height. Characteristic clinical features in the absence of organic disease, skeletal dysplasia, or eating disorders and in the presence of normal intestinal, renal, and hepatic function, suggested GHD. Tests of corticotropin and thyrotropin function were normal. Serum IGF-I and IGF-binding protein-3 (IGFBP-3) values were low: serum IGF-I was 2.49 and 4.08 µg/L and serum IGFBP-3 was 0.65 and 0.6 mg/L in P1 and in P2, respectively. Several family members had severe short stature, including one parent of each index case. The pedigrees of the two nonconsanguineous families are shown in the Data Supplement. The study was approved by the Internal Review Board of the Garrahan Pediatric Hospital.
In the two patients, GH secretion was evaluated after two pharmacologic tests of provocative secretion by determining serum GH in the same samples with use of the assays for total and 22-kDa GH. The maximum response of serum total GH in P1 and P2 was 10.9 and 10.4 µg/L, respectively, above the cutoff for normal response. These results did not support the diagnosis of GHD. However, the test was reassessed by determining the serum concentration of 22-kDa GH in the same samples. Maximum responses were 2.5 and 1.5 µg/L, respectively, below the cutoff for a normal response. In contrast to the first assay, results from the second assay supported the diagnosis of GHD.
P1 and P2 were treated with 0.17 mg of recombinant human GH (rhGH) per kg of body weight per week for 1.7 and 3.7 years, respectively. For both patients, the height SDS improved markedly after treatment.
The time course of serum GH response in P1 and P2 to clonidine as determined by two immunoassays and the bioassay is shown in Fig. 1
. The relative pattern of the response was similar for the three assays, but the response was above the cutoff limit when serum total GH was assayed, whereas it was below the cutoff when 22-kDa GH or GH bioactivity was determined.
|
Serum IGF-I concentrations during the IGF-I generation test are shown in Table 1
. The increase in IGF-1 was <15 µg/L at every time point measured, suggesting that the two patients had GH insensitivity. By contrast, all nine patients with idiopathic GHD responded to the test. However, after 1 year of rhGH treatment, serum IGF-I and serum IGFBP-3 concentrations increased in both P1 and P2 (serum IGF-I to 47.4 and 60.4 µg/L and serum IGFBP-3 to 1.5 and 2.3 mg/L in P1 and P2, respectively).
|
DNA sequence analysis of PCR amplification products of the GH-1 gene is shown in the Data Supplement. A heterozygous point mutation in exon 5, which causes an Arg-to-His mutation at position 183 of the GH protein, was present in both P1 and P2. The father of P1 and the mother of P2 carried the same GH-1 mutation. Furthermore, we found two additional base changes in intron 1, at positions +52 (A-G) and +56 (A-T), in all affected members.
We report here on two prepubertal girls, belonging to two different families, with severe short stature and autosomal dominant IGHD. The initial work up was misleading because they had normal responses to two provocative tests of GH secretion, as measured by the SER 66/217 GH assay. Although we used stimulation tests that are weak for adults (17), it is accepted that they are informative tests in children (18). We have used these tests in children for more than 20 years with good results (11). We have found that our patients responded normally to stimulation tests when total GH was determined but below normal when 22-kDa GH or bioactive GH was determined. Our data indicate that secretion of the active 22-kDa GH isoform is diminished and that, consequently, a high proportion of non-22-kDa GH molecular forms circulates in the blood. Apparently these molecular forms have poor biological activity, as indicated by the Nb2 cell proliferation bioassay
In our study, the two patients failed to respond to an acute IGF-I generation test. The reliability of the IGF-I generation test has been questioned because of lack of reproducibility (19) and overlapping responses in GH-insensitive and GHD patients (7). However, short-term increments of IGF-I predict growth response to GH therapy (20). Moreover, because positive responses are found in most GHD patients (7), some degree of insensitivity to GH was present in our two patients. As in a patient reported by others (21), our patients responded to treatment with increased growth velocity and serum IGF-I after prolonged rhGH treatment. This suggests that there was a reversible inhibition of GH biological action, i.e., prolonged exogenous rhGH might have overcome GH insensitivity by decreasing secretion of all endogenous GH isoforms. The molecular mechanism of the dominant-negative effect is unknown. As suggested by Wada et al. (22), the altered 22-kDa GH/non-22-kDa GH isoform ratio in blood could impair receptor dimerization. The 20-kDa GH isoform has low affinity for GHR site 1, forming no detectable 1:1 complex but forming a 1:2 complex efficiently. Different GH analogs and fragments may interact as weak antagonists or agonists of the GHR, depending on the relative affinities of sites 1 and 2 to the receptor (23).
Our study showed that affected members of the two families carry the R183H point mutation in exon 5. We also detected two additional point mutations at positions +52 and +56 of intron 1. This novel finding suggests that genotype and phenotype variations exist among different families. These intronic mutations could represent normal polymorphism variants, or they might determine alternative splicing, generating GH isoforms that could contribute to changes in the ratio between 22-kDa GH and non-22-kDa GH.
We conclude that the evaluation of serum GH response to provocative stimuli depends on the assay selected to measure serum GH and suggests that altered circulating GH isoforms may be important in defining the GHD phenotype in these patients. GH isoforms might have a reversible dominant-negative effect at the GHR level. Impaired GH secretion may present also, as suggested by the absence of GH hypersecretion. Interestingly, the dominant-negative effects were overcome by administration of rhGH at a dose similar to that used to treat nonfamilial GHD.
Acknowledgments
We thank the National Pituitary Agency for the generous supply of IGF-I RIA reagents. This work was supported by grants from CONICET, FONCYT, and Ministerio de Salud (Beca Carrillo-Oñativia) of Argentina and from Pharmacia Endocrine Care International Fund.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  O. Hess, Y. Hujeirat, M. P. Wajnrajch, S. Allon-Shalev, Z. Zadik, I. Lavi, and Y. Tenenbaum-Rakover Variable Phenotypes in Familial Isolated Growth Hormone Deficiency Caused by a G6664A Mutation in the GH-1 Gene J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4387 - 4393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Salemi, S. Yousefi, K. Baltensperger, I. C A F Robinson, A. Eble, D. Simon, P. Czernichow, G. Binder, E. Sonnet, and P. E Mullis Variability of isolated autosomal dominant GH deficiency (IGHD II): impact of the P89L GH mutation on clinical follow-up and GH secretion Eur. J. Endocrinol., December 1, 2005; 153(6): 791 - 802. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
