hGH Variants Research Articles

Antigenic Epitope Mapping of the Human Growth Hormone Molecule: A Strategy to Standardize Growth Hormone Immunoassays

The antigenic surface of the human growth hormone (hGH) molecule has been characterized by means of monoclonal antibodies (MAbs) and their specific binding sites. From a panel of MAbs raised against the 22 kDa variant of hGH, ten distinct antigenic entities could be identified as binding sites for specific MAbs in an epitope mapping experiment. The experiment was based on combinations of the MAbs in two-site assays. By investigating the cross-reactivity of the MAbs with growth hormone (GH) variants and fragments of hGH, the underlying protein structure of some of the partially overlapping epitopes could be identified. By using antibodies of known epitope specificity as competitors in receptor binding assays of hGH, it was demonstrated that the lactogenic and somatogenic biological effects of hGH are mediated through different portions of the hormone. These observations may explain some of the differences between the various immunoassays for hGH. A strategy for selecting antibodies of known epitope specificity and biological relevance for the design of plasma GH immunoassays is outlined. The results of such an immunoassay should closely reflect the biological activity of circulating plasma hGH.

Preparation of 125I-labeled human growth hormone of high quality binding properties endowed with long-term stability.

125I-labeled human growth hormone (125I-labeled.hGH) was prepared by using two variants of the chloramine T labelling procedure and purified by polyacrylamide gel electrophoresis (PAGE) of the reaction mixture. Variant A produced a tracer with high specific activity (100 +/- 10 microCi/microgram), high maximal binding capacity to antibodies (93%) and long-term stability (at least 150 days at -20 degrees C). No diiodinated tyrosil residues could be detected in this tracer. Variant B was devised to obtain higher yields of labeled hormone. The electrophoresis of the iodination mixture revealed two radioactive components with Rm values of 0.49 and 0.55 which result from the iodination of hGH variants preexisting in the starting material. Both tracers had similar specific activities (70 +/- 10 microCi/microgram), high maximal binding capacity to antibodies or receptors (80-100%, after 80 days of their obtention) and high stability (at least 100 days at -20 degrees C). It is concluded that the iododerivatives of hGH obtained by either method are adequate to perform radioimmunoassay and receptor studies and have long-term stability.

Monoclonal Antibodies which Preferentially Bind to 22K Human Growth Hormone Rather than Its 20k Variant.

We have established 13 hybridoma cell lines which secrete mouse IgG1 monoclonal antibodies (McAbs) to human growth hormone (hGH). Binding affinity and binding specificity of McAbs were analyzed by competitive radioimmunoassay. Among these McAbs, CL. B1 showed a high affinity of 9.8 x 10(8) l/mol, and all McAbs so far tested showed very weak cross-reactivity or none at all with human prolactin (hPRL) and human chorionic somatomammotropin (hCS; human placental lactogen). Analysis of binding sites of McAbs using hGH variant and fragments in both ELISA and RIA demonstrated that McAbs could be classified into two groups. All the McAbs obtained in this study bound to plasmin-digested fragment S2 (hGH 1-134 and 141-191) and fragment alpha 3 (hGH 1-134 and 147-191). However, five (such as 1D2) out of 13 McAbs bound to fragment F1 (hGH 1-134) and others (such as CL. B1) did not. The McAb CL. B1 in the latter group showed low affinity with 20 K hGH (residue 32-46 deleted in native 22 K hGH) in contrast to high affinity with hGH (22 K). This suggests that the former McAbs recognize an epitope located at the N-terminal two-third part of hGH. In contrast, the McAbs of the latter group are likely to recognize three-dimensional structure of native 22 K hGH.

Use of calcium dependence as a means to study the interaction between growth hormones and their binding proteins in rabbit liver.

The affinity of 22,000-Mr human growth hormone (22 K-hGH) for GH binding proteins in rabbit liver is increased approx. 19-fold by 25 mM-Ca2+. In contrast, ovine growth hormone (oGH) binding is Ca2+-independent up to 10 mM, and decreased by greater Ca2+ concentrations. The 20,000-Mr hGH variant (20K-hGH), lacking residues 32-46, exhibits intermediate behaviour. Without Ca2+ there is a residual 40% of maximum specific binding to liver microsomes, and this increases to 65% with liver cytosolic GH binding proteins. In contrast with 22K-hGH, Scatchard analysis of 20K-hGH binding to liver microsomes produces curvilinear plots in the presence of 25 mM-Ca2+. From these results and inhibition studies with monoclonal antibodies to the GH binding proteins, it is concluded that deletion of the region 32-46 from 22K-hGH has eliminated one component of high-affinity Ca2+-potentiable binding. The Ca2+-mediated increase in Ka for the 22K-hGH-binding protein interaction is consistent with convergence of unit negative charges on the hormone and binding protein towards an intercalated Ca2+ ion. A positive charge in the critical region of nonprimate GHs would render their interactions Ca2+-independent and of lower Ka compared with 22K-hGH. A likely candidate for the negatively charged interactive residue is glutamate-33, since it is unique to human GH and is replaced by a positively charged arginine in non-primate GHs. Its absence in 20K-hGH could explain the altered calcium-dependence of 20K-hGH binding to what is probably the type 2 binding protein [Barnard & Waters (1986) Biochem. J. 237, 885-892]. The Ca2+-dependence of 20K-hGH binding to a subset of GH binding proteins provides both a verification and a mechanistic basis for the proposal [Hughes, Tokuhiro, Simpson & Friesen (1983) Endocrinology (Baltimore) 113, 1904-1906] that 20K-hGH binds with high affinity to only a subset of binding proteins in rabbit liver membranes.

Biological activities of human growth hormone and its derivatives estimated by measuring DNA synthesis in Nb2 node rat lymphoma cells.

Nb2 cell is a rat lymphoma cell line that responds to lactogens such as prolactin and human growth hormone (hGH) with an increased rate of proliferation. We explored the relationship between the biochemical events induced by hGH and its derivatives and their receptor binding activities. hGH stimulated RNA, DNA and protein synthesis of Nb2 cells as a function of time. Stimulation of RNA and protein was maximal at 2-3 h and 12 h, respectively, after the addition of hGH. DNA synthesis, measured by the rate of [3H]thymidine incorporation, reached a maximum after 18-h incubation with hGH. Stimulation of DNA synthesis was elicited by hGH in a dose-dependent manner between 0.45 and 45 pmol/l. The activity of the 20 K hGH variant in stimulating DNA synthesis was approx 30% of that of hGH. In contrast, S1-hGH, which lacks a sequence of ten amino acids (140-149) of hGH, showed a 3.2-fold greater activity than hGH. F1 (amino-terminal sequence 1-134 of hGH) was only 0.06% as active as hGH, and the activity of F2 (C-terminal 42 amino acid residue of hGH) was less than 0.01%. Both fragment 1-15 and 32-46 were without effect. The relative potencies of these hGH derivatives in stimulating DNA synthesis were similar to their relative abilities to inhibit [125I]hGH binding to lactogenic receptors on Nb2 cell. Nb2 cells provide a suitable model to study the relationship between receptor binding and the biochemical events induced by lactogens.

A new natural hGH variant--17.5 kd--produced by alternative splicing. An additional consensus sequence which might play a role in branchpoint selection.

From a human pituitary cDNA library, we have cloned 3 distinct human growth hormone (hGH) cDNAs, coding respectively for the 22 K hGH, the 20 K variant, and a yet unknown 17.5 K variant. S1 mapping analysis using human pituitary RNA confirms the existence of at least four distinct hGH mRNAs originating from alternative acceptor sites at the second intron of the primary transcript. We have analysed the hGH gene sequence to explain the high frequency of alternative splicings which occur only at this location. In this study we propose CTTGNNPyPyPy as an additional consensus sequence guiding the selection of the branched nucleotide.

Purification of pituitary and biosynthetic human growth hormoneusing monoclonal antibody immunoadsorbent

This report describes the purification of human growth hormone from crude pituitary extract and lysate of recombinant E. coli by an immunoadsorbent purification with monoclonal antibody coupled to solid phase. By a single-immunoaffinity chromatography step pure hGH can be obtained from either origin as revealed by SDS-PAGE followed by silver staining or immunoblotting. An additional ion-exchange chromatography step results in homogeneous 22 kDa hGH preparations. Furthermore, this method may be used for isolation of a pituitary hGH variant which has higher binding affinity for this monoclonal antibody than the major 22 kDa form. This study clearly illustrates the potential of monoclonal antibody immunoadsorbents for purification of different molecular forms of hGH.

Pituitary and Recombinant Deoxyribonucleic Acid-Derived Human Growth Hormones Alter Gloucose Metabolism in 3T3 Adipocytes*

3T3-F442A adipocytes were used to compare the effects on glucose metabolism of pituitary human (h) GH and methionyl-hGH produced by recombinant DNA techniques (met-hGH 22K). Pituitary hGH and met-hGH 22K were similar in their ability to inhibit glucose oxidation and lipid accumulation after 48 h. After 4 h of incubation, both forms of hGH stimulated glucose oxidation transiently in 3T3 adipocytes. A bacterially produced form of the 20,000-dalton variant of hGH (met-hGH 20K) also stimulated glucose oxidation at 4 h and inhibited glucose oxidation and lipid synthesis after 48 h. All three forms of hGH had a similar ability to inhibit [125I]iodo-met-hGH 22K binding to 3T3-adipocytes. Thus, met-hGH 22K and 20K directly produce in 3T3 adipocytes the transient stimulation and delayed inhibition of glucose metabolism attributed to pituitary hGH, indicating that these metabolic effects are intrinsic to the hGH molecule.

Release of the 22,000- and the 20,000-dalton variants of growth hormone in vivo and in vitro by human anterior pituitary cells.

Using a combination of polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and immunoblotting with antihuman GH (anti-hGH) serum, we quantitatively and independently measured the major 22,000-dalton form of hGH (hGH22K) and the 20,000-dalton form (hGH20K). This technique was equally effective in assaying cell culture media or human sera. We studied isolated human pituitary cell cultures during a 16-day incubation period with and without stimulation by GH-releasing hormone (GHRH). Under basal conditions, the cells released 2.83 +/- 0.24 (+/- SEM) microgram hGH22K/ml . day and 0.67 +/- 0.17 microgram hGH20K/ml . day. GHRH (10(-8) M) treatment resulted in stimulation of both hGH22K and hGH20K by 24 h. We also measured both hGH22K and hGH20K in the sera of normal subjects before and after an iv bolus injection of 100 micrograms GHRH. hGH20K increased as did hGH22K. Peak concentrations of both variants occurred 45 min after GHRH administration. The results of this study indicate that the combination of sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting is an accurate and effective means of separately assaying hGH22K and hGH20K. We also demonstrated that primary monolayer cultures of human pituitary cells are an excellent model system for study of the secretion of these two hGH variants.

Molecular forms of human growth hormone secreted in vivo: nonspecificity of secretory stimuli.

Several molecular forms of human GH (hGH) are present in blood, but their individual regulation is largely unknown. To examine the factors controlling the secretion of individual hGH variants, the relationship between the mixture of circulating hGH forms and the type of preceding secretory stimulus was studied in 18 normal subjects and 5 acromegalic patients. The stimuli employed were L-dopa, GH-releasing hormone-(1-40), exercise, spontaneous sleep-related and daytime secretory bursts, estrogens, and TRH in the acromegalic patients. Three monomeric hGH forms (22K, 20K, and acidic hGH) were identified in all samples; their mean relative proportions were 76.4%, 15.8%, and 7.9%, respectively. These proportions were similar in all subjects, regardless of stimulus, sex, or presence of acromegaly (P greater than 0.25). We conclude that the release of individual hGH forms is not stimulus specific, but, rather, that the secretory granule contents are released in toto upon somatotroph stimulation.

A Specific Growth Hormone-Binding Protein in Human Plasma: Initial Characterization*

Human (h) GH in plasma exists as a series of size isomers, which are in part explained by the presence of hGH oligomers. However, certain aspects of circulating large mol wt hGH, such as its high relative proportion compared to that in the pituitary, are poorly understood. To explore whether binding of hGH to plasma protein(s) could contribute to the phenomenon of large mol wt hGH, we incubated freshly prepared monomeric [125I]hGH or biosynthesized [3H]hGH with normal human plasma or serum at pH 7.4 for various time periods at 22 and 37 C. Plasma radioactive hGH patterns were then analyzed simultaneously with unincubated tracer hGH by Sephadex G-100 and G-200 chromatography. We found that part of the radioactivity was converted to a component with an apparent mol wt of 85,000, suggesting binding to a plasma protein(s). This phenomenon was inhibited in a dose-dependent fashion by unlabeled hGH. Saturation/Scatchard analysis indicated an association constant (Ka) of 2-3 X 10(8) M-1 and a maximum binding capacity of 20 ng hGH/ml plasma. Binding was rapid, reversible, and specific. A number of polypeptide hormones, including human placental lactogen, hPRL and rat GH, did not inhibit hGH binding. However, the 20K variant of hGH interacted weakly with the plasma binding component (Ka, 1.2 X 10(7) M-1; maximum binding capacity, 450 ng/ml). The binding component was heat labile and could be partially purified by gel permeation chromatography and affinity chromatography on a hGH-Sepharose column. Its estimated mol wt is 60,000-65,000, and it appears to bind one molecule of hGH to form a complex of 80,000-85,000 mol wt. The binding component is neither albumin nor an immunoglobulin. Under physiological conditions, a minimum of 15-18% of circulating hGH is presumably bound to this plasma component. We conclude that a specific, high affinity, low capacity binding protein for hGH with mol wt of 60,000-65,000 exists in normal and hypopituitary human plasma. hGH complexed with this protein forms part of big-big hGH. The biological significance of this binding protein remains to be investigated.

Biosynthetic 20-kilodalton methionyl-human growth hormone has diabetogenic and insulin-like activities.

The anterior pituitary gland produces a 20-kilodalton (kDa) variant of human growth hormone (hGH) that differs from the predominant 22-kDa form of hGH in that amino acid residues 32-46 are deleted. Previous work has suggested that the 20-kDa variant possesses the full growth-promoting and lactogenic activities of 22-kDa hGH but lacks its intrinsic diabetogenic and insulin-like activities. In the present study, recombinant DNA techniques were used to prepare biosynthetic 20-kDa hGH, and some of the biological properties of the purified hGH variant were examined. The biosynthetic 20-kDa hGH variant was found to share the propensity for aggregation exhibited by its native counterpart. Moreover, like the native variant, biosynthetic 20-kDa hGH possessed full growth-promoting activity in the weight gain test in hypophysectomized rats. However, contrary to previous work suggesting that native 20-kDa hGH lacks diabetogenic and insulin-like activities, biosynthetic 20-kDa hGH was found to have substantial diabetogenic activity when administered chronically to ob/ob mice and to possess approximately 20% the in vitro insulin-like activity of biosynthetic 22-kDa hGH on isolated epididymal adipose tissue of hypophysectomized rats. The diabetogenic and insulin-like activities of biosynthetic 20-kDa hGH cannot be ascribed to contamination of the hormone preparation with the 22-kDa form of hGH or with other diabetogenic or insulin-like pituitary peptides. Therefore, the results strongly suggest that diabetogenic and insulin-like activities are also intrinsic properties of the 20-kDa variant of hGH.

Highly purified human growth hormone fails to stimulate lipolysis in rabbit adipocytes in vitro or in rabbits in vivo

To determine whether the rapid lipolytic effect observed with human growth hormone (hGH) preparations in rabbits and rabbit adipose tissue is an intrinsic property of the hormone, we examined the lipolytic effects in vivo and in vitro of clinical grade preparations, hGH purified by DEAE cellulose chromatography, and hGH prepared by recombinant DNA techniques. Using isolated rabbit perirenal adipocytes, ACTH and clinical-grade hGH preparations both stimulated glycerol release to the same maximal rate with half-maximal hGH effects observed between 8 and 50 μg/mL. Purification of the most potent lipolytic preparation of clinical grade hGH by DEAE cellulose chromatography yielded a preparation (2 IU/mg of growth activity that retained insulinlike effects on rat fat pad [U- 14C] glucose metabolism) that, at concentrations up to 0.2 mg/mL, failed to stimulate lipolysis by adipocytes incubated for one or four hours in the presence or absence of dexamethasone or trypsin inhibitor or when preincubated for three hours prior to addition of hGH. Recombinant DNA-derived hGH did not stimulate glycerol release at 0.1 mg/mL. While antiserum to purified hGH blocked the increase in glucose oxidation in rat fat pads produced by clinical grade hGH, it did not inhibit its lipolytic effect using rabbit adipocytes. Purified hGH (0.1 mg/kg IV) was also unable to elicit a rise serum free fatty acid (FFA) levels of conscious rabbits while, at the same dose, clinical grade hGH increased FFA levels to 900 μEq/L over basal. Rapid lipolytic stimulation in rabbit adipocytes by hGH preparations could not be attributed to the 20,000 molecular weight variant of hGH (hGH20K) or the peptide corresponding to positions 32 to 46 of hGH (deletion peptide). Glycerol release was accelerated by a < 5000 molecular weight fraction obtained during the preparation of purified hGH. In conclusion, the lipolytic effect produced by clinical-grade hGH preparations in rabbits in vivo and in vitro could not be attributed to intact hGH, hGH20K, or deletion peptide, but is elicited by a contaminant separable from hGH on further purification.

Antigenic sites of human growth hormone and related molecules detected by monoclonal antibodies to human growth hormone

Four major antigenic sites for human growth hormone (hGH) were identified by 27 mouse monoclonal antibodies to hGH. Sites 1 and 2 are spatially close whereas sites 3 and 4 are located in other parts of the molecule. There also appears to be a subdivision of antigenic sites. A panel of 10 monoclonal antibodies, which included representatives from each antigenic site group, were used to determine cross-reactivities between hGH and human placental lactogen (hPL), human prolactin (hPRL), the 20,000 mol. wt variant of hGH (liGH 20k) and a disulfide-linked dimer of hGH (diS-dimer). The data suggest a high conformational dependence of antigenic sites in hGH. DiS-dimer retains all four antigenic sites of hGH, although all have been altered. hGH 20k retains sites 2–4 but site 1 has been dramatically altered. hPL retains site 3, whereas sites 1 and 4 have been dramatically altered and site 2 may be lacking. The extremely low cross-reactivity observed for hPRL is consistent with the dissimilarity between hGH and hPRL. Antigenic site 3 is the most conserved of all sites. The lack of structural similarity compared with hGH of site 1 in hGH 20k and of a portion of site 3 in diS-dimer suggests that it may be possible to develop specific radioimmunoassays for these structural variants of hGH.

Growth hormone and lipolysis: a reevaluation.

A recent report claimed that the lipolytic activity found in clinical grade human GH (hGH) was entirely attributable to an acidic peptide contaminant. Using the same preparations as studied in the earlier report, we found that delayed lipolytic activity, as demonstrated in the presence of dexamethasone, was fully evident in the highly purified hGH preparation which had been reported to be devoid of lipolytic activity. We also studied the acidic peptide contaminant in which all of the lipolytic activity of clinical grade hGH was alleged to reside and found its lipolytic activity was equivalent to that of the highly purified hGH. Both compounds increased lipolysis in the presence of 0.1 micrograms/ml dexamethasone after a lag period of 1 h. The minimum effective concentration of both fell within the range of 1 and 10 ng/ml, and both were capable of producing a 2- to 3-fold increase in glycerol production. When saturating amounts of each were added simultaneously they produced no greater effect than either alone, although the lipolytic rate was far from saturated as evidenced by the more than 2-fold further increase in glycerol production when 100 ng/ml epinephrine was added along with the combined peptides. The lipolytic response to highly purified hGH was comparable to that seen with bacterially synthesized hGH which should be free of pituitary peptide contaminants. The acidic peptide appeared to be a variant of hGH as judged by its ability to produce both an insulin-like response and refractoriness. Furthermore, the acidic peptide was indistinguishable from hGH in competing for specific binding sites on adipocytes. Although an ability to increase glycerol production in rat adipose tissue in vitro thus appears to be an intrinsic property of hGH, the physiological significance of this effect remains to be established.

Phosphorylation of human growth hormone by the epidermal growth factor-stimulated tyrosine kinase.

In the present study, we have demonstrated that human growth hormone (hGH) can be phosphorylated by the epidermal growth factor (EGF)-stimulated tyrosine kinase of A431 cell membranes. Phosphotyrosine was the predominant phosphoamino acid released from phosphorylated hGH on partial acid hydrolysis. All five tyrosine-containing tryptic peptides of hGH are also phosphorylated by the EGF-stimulated tyrosine kinase. The highest phosphate incorporation was found for peptide T4 (residues 20-38), which is distinguished by a high frequency of acidic amino acids. The phosphorylated peptides have been characterized by HPLC and two-dimensional mapping on paper. Comparison with the labeled peptides obtained on tryptic digestion of phosphorylated hGH suggests that tyrosine phosphorylation is restricted to two tryptic peptides, T4 (tyrosine-28 or -35) and T6 (tyrosine-42). It is suggested that the absence of early insulin-like activity in the naturally occurring Mr 20,000 variant of hGH, which has an internal deletion spanning residues 32-46, may be a consequence of the loss of the tyrosine phosphorylation sites at residues 35 and 42.

In vitro deletional mutagenesis for bacterial production of the 20,000-dalton form of human pituitary growth hormone.

The 20,000-dalton (20K) variant form of human growth hormone (hGH) present in extracts from pituitary glands differs from the major form of hGH (22K, 191 amino acids) by the deletion of amino acid residues 32-46. Using oligonucleotide-mediated mutagenesis, the DNA coding for these amino acids was deleted from the gene previously constructed by us (Goeddel et al., 1979) for microbial hGH production. The DNA to be deleted was looped out by the annealing of a synthetic oligodeoxyribonucleotide to the coding strand of the hGH gene contained on recombinant phage M13 mp8 DNA. Resulting heteroduplex structures were stabilized using primer-directed in vitro DNA synthesis in the presence of T4 DNA ligase. On transformation of Escherichia coli, these heteroduplex DNAs yielded phage whose genomes contained either the original or the partially deleted hGH gene, and genotypes were distinguished by in situ plaque hybridization with synthetic oligonucleotide probes. A gene with the correct deletion was used to express the short hGH variant in E. coli.

Monomeric pituitary growth hormone and prolactin variants in man characterized by immunoperoxidase electrophoresis

Immunoperoxidase electrophoresis, combining SDS-ME-PAGE and the ‘double bridge’ immunoperoxidase staining was applied to crude human pituitary homogenates. With anti-hGH and anti-hPL sera, 4 hGH-related monomers were characterized: a M r 22 000 peptide corresponding to hGH; a M r 20 000 peptide corresponding to the known hGH variant and two unknown hGH variants ( M r 65 000 and M r 75 000). With anti-ovine, rat and human PRL sera, 4 PRL-related monomers were immunostained: one comigrated with purified hPRL ( M r 25 000), and 3 were unknown ( M r 29 000; M r 45 000; M r 16 000).

Structural variants of human growth hormone: biochemical, genetic, and clinical aspects.

Human growth hormone (hGH) is a mixture of peptides in which the major physiologic component is a single chain polypeptide of 191 residues with a molecular weight of 22,000 ("22K" form). The minor components differ from the 22K form in terms of mass (e.g. the 20K form, a single-chain peptide synthesized by deletion of residues 32-46 of the 22K isomer, and the 45K variant formed by aggregation of the 22K molecule), or of charge (e.g. the more acidic two-chain forms alpha 2 and alpha 3 which are generated by proteolytic deletion of residues 135-140 and 135-146 from the 22K variant). The minor components also differ from the 22K molecule in (a) their metabolic effects; (b) their capacity to bind to antiserum raised against the 22K form in a radioimmunoassay (RIA); and (c) their ability to bind to membrane receptors for the 22K hGH in a radioreceptor assay (RRA). Two genes, N and V, involved in the biosynthesis of hGH, are located on human chromosome 17. Heritable alterations of the N gene may cause deficiency of immunoreactive hGH and growth failure. A pathologic variant of hGH has also been identified which is indistinguishable from the 22K form on RIA, but which has low reactivity on hGH-RRA and low somatomedogenic activity. Several mechanisms to account for this bioinactive variant are discussed.

Separation and Assignment of the Tryptic Peptides of Human Growth Hormone (hGH) and the 20K Dalton hGH Variant by Reversed Phase High Performance Liquid Chromatography

Abstract Reversed-phase high performance liquid chromatography (RP-HPLC) has been used to separate the peptides generated by tryptic cleavage of human growth hormone (hGH) and the 20K dalton human growth hormone variant. The total amino acid compositions of both these pituitary proteins has been accounted for on the basis of these chromatographic mapping procedures. Structural analyses of the peptides isolated from semi-preparative RP-HPLC separations has confirmed that the primary structure of the variant differs from that of the sequence of hGH by deletion of the amino acid residues 32–46.