A Novel Specific Bioassay for Serum Human Growth Hormone

0021-972X/00/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2000 by The Endocrine Society

Vol. 85, No. 11 Printed in U.S.A.

A Novel Specific Bioassay for Serum Human Growth Hormone MAYUMI ISHIKAWA, ATSUKO NIMURA, REIKO HORIKAWA, NORIYUKI KATSUMATA, OSAMU ARISAKA, MITSUFUMI WADA, MASARU HONJO, AND TOSHIAKI TANAKA Department of Endocrinology and Metabolism, National Children’s Medical Research Center, (M.I., A.N., R.H., N.K., T.T.) Setagaya-ku, Tokyo 154-8509; The 1st Department of Internal Medicine, Toho University of School Medicine (M.I.), Ota-Ku, Tokyo 143-8541; Department of Pediatrics, Dokyo University School of Medicine (O.Y.), Shimotuka-gun, Ibaragi 321-0267; and Life Science Laboratories Mitsui Chemicals, Inc., (M.W., M.H.), Mobara, Chiba 297-0017, Japan ABSTRACT Human GH receptor (hGHR) was recently expressed on a Ba/F3 cell line, which is a mouse pro-B cell lymphoma that has been induced to become a cloned cell line (Ba/F3-hGHR). Using a Ba/F3-hGHR cell line, we have established a bioassay for serum hGH. hGH stimulated cell proliferation in a dose-dependent manner in concentrations ranging from 1 ng to 100 ng/mL. Cell proliferation was not influenced by other hormones or growth factors in the bioassay, with the exception of insulin-like growth factor I (IGF-I) and GH binding protein. Free IGF-I significantly stimulated the proliferation of Ba/F3-hGHR cells at concentrations over 25.85 ng/mL in this bioassay system, but serum IGF-I did not stimulate cell proliferation because the sensitivity of cell proliferation was insufficient for free IGF-I in serum. GH binding protein, however, did suppress cell proliferation at the highest concentration (100 ng/mL), but did not at the average concentration (20 ng/mL). Human serum stimulated cell pro-

I

T IS CLINICALLY useful to measure the bioactivity of serum human GH (hGH) to determine whether short stature is due to biologically inactive hGH, as suggested by Kowarski et al. (1). Various GH bioassays, both in vivo and in vitro have been proposed. The two most common types of in vivo bioassays of GH are body weight gain in hypophysectomized rats (2– 4) or in dwarf “little” mice (5) and the tibial plate assay on hypophysectomized rats (6). In both in vivo bioassays, however, the response is easily influenced by stress and metabolic state, and the sensitivity is insufficient for measuring serum GH bioactivities. In vitro GH bioassays suitable for clinical use include radioreceptor assays (7, 8), receptor modulation assays (9), and cell proliferation bioassays using the Nb2 cell line (10). Recently, expression of the hGH receptor (hGHR) in the mouse pro-B cell lymphoma cell line. Ba/F3 has led to a cloned cell line (Ba/F3-hGHR) the proliferative response of which is hGH dose dependent (11). Using this cell line, we established in this study a bioassay system for hGH and

Received December 7, 1999. Revision received July 27, 2000. Accepted August 1, 2000. Address correspondence and requests for reprints to: Mayumi Ishikawa, Department of Endocrinology and Metabolism, National Children’s Medical Research Center, 3-35-31, Taishido, Setagaya-ku, Tokyo 154-8509, Japan.

liferation, which was completely suppressed by anti-GH antibody. The GH bioactivity of serum samples from normal children and patients with non-GH deficient short stature correlated strongly with the serum hGH concentration determined by immunoradiometric assay (IRMA) (r ⫽ 0.967, r ⫽ 0.924, P ⬍ 0.0001, respectively). The ratio of bioactivity/IRMA was 1.01 ⫾ 0.26 in sera from normal children and 1.18 ⫾ 0.24 and 1.00 ⫾ 0.29 at basal values and peak values in GH stimulation tests, respectively, in sera from patients with non-GH deficient short stature. The bioactivity/IRMA ratio for the serum GH bioactivity of a patient who had biologically inactive GH caused by an amino acid substitution was 0.333 ⫾ 0.056 (mean ⫾ SD). In conclusion, we established a new sensitive bioassay for hGH that is specific for hGH somatogenic action and is useful for screening of patients with short stature caused by biologically inactive hGH. (J Clin Endocrinol Metab 85: 4274 – 4279, 2000)

measured hGH bioactivity in sera from normal children and patients with non-GH-deficient short stature. Materials and Methods Hormone preparations hGH and human PRL (hPRL) were provided by the Polypeptide Hormone Laboratory (University of Manitoba, Canada). Bovine GH was purchased from Biogenesis Ltd. (England, UK), and human IGF-1 was purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). Insulin (Novorin R 40 IU/mL) was purchased from Novo Nordisk (Gentafte, Denmark). FSH (Fertinorm P) and hCG (Prophacy) were provided by Serono Japan (Tokyo, Japan). Fibroblast growth factor was purchased from Collaborative Research (Waltham, MA), and epidermal growth factor was purchased from Toyobo Co., Ltd. (Osaka, Japan). TSH was purchased from Zymed Laboratories Inc. (California), and l-thyroxine sodium was purchased from Teikoku Zouki (Tokyo, Japan). Hydrocortisone (Solu-cortef) was provided by Sumitomo-Pharmacia-Upjohn (Stockholm, Sweden). GHBP was provided by Novo Nordisk. GRF (GRF Sumitomo) was purchased from Sumitomo Pharmaceuticals (Tokyo, Japan). Anti-GH antibody (clone 5801) was provided from Oy Medix Biochemica Ab (Kaunianinen, Finland).

Cell cultures Ba/F3-hGHR (11), which was established by Mitsui Chemicals, Inc. (Tokyo, Japan), was maintained as suspension cultures in 75 tissue culture flask (Falcon) Life Technologies, Inc., (Grand Island, NY) supplemented with FCS (10%; JRH Biosciences, Australia), 2-mercaptoethanol (2-ME) (50 ␮m; Nacalai Tesque, Inc., Kyoto, Japan), penicillin (50

4274

A BIOASSAY FOR hGH U/mL), streptomycin (50 ␮g/mL; Life Technologies, Inc.), and hGH (10 nm) in an atmosphere of 5% CO2, 95% air at 37 C.

Bioassay with Ba/F3-cell line Approximately 4 – 6 h before the start of the bioassays, the cells were washed twice with assay medium (RPMI 1680, supplemented with 5% FCS, 50 ␮m 2-ME, and antibiotics, without hGH) and were transferred to the assay medium and incubated for 4 – 6 h to slow down the rate of cell replication. After incubation, the cells were collected by centrifugation (3 min at 1000 rpm) and resuspended in the assay medium at a concentration of 1 ⫻ 105 cells/mL. Two-hundred microliter aliquots were distributed in each well of 96-well microplate (Nalge Nunc International, Roskilde, Denmark). Standard hGH was diluted with 0.01 m PBS supplemented with 0.1% BSA, (Sigma Chemical Co., St. Louis, MO) at each concentration (0 –1000 ␮g/L). Samples were incubated at 56 C for 40 min to inactivate the serum. To each well was added 25 ␮L of standard or sample. The cultures were incubated in a CO2 incubator (5% CO2 ⫹ 95% air) for 48 h at 37 C. At the end of the incubation, the colorimetric end point was determined by an eluted stain bioassay (ESTA) described by Marshall et al. (12) and Ealey et al. (13) with a slight modification. Briefly, 20 ␮L MTT solution (3-[4, 5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) (5 mg/mL in 0.01 m PBS; Sigma) were added to each well and incubated at 37 C for 4 h in a CO2 incubator. During this time, activated cells reduced the yellow MTT salt to purple formazan. The plate was centrifuged at 800 rpm for 10 min, and the stain was eluted into dimethyl sulfoxide (Nacalai Tesque, Inc.), of which 100 ␮L were added to each well. Bioactive responses were determined with a kinetic microplate reader (Molecular Devices, Menlo Park, CA), reading optical densities at the test wavelength of 550 nm and a reference wavelength of 650 nm to correct for differential scattering. All samples were assayed in duplicate (coefficient of variation, ⬍12%), and a control serum with known bioactivity was used at every assay for the quality control. In the blocking study, 25 ␮L anti-hGH antibody (diluted ⫻50 or ⫻100 with 0.01 m PBS containing with 0.1% BSA) were added to wells of standard or sample. The influence of GHBP was determined by diluting the GH standard with GHBP, in concentrations of 20 or 100 ng/mL in 0.01 m PBS containing 0.1% BSA. The bioactivity of other hormones and growth factors— bGH (0.1– 1000 ␮g/L), PRL (0.1–1000 ␮g/L), human IGF-I (0.1–1000ng/mL), human insulin (0.1 ␮ IU/mL to 1.0 mIU/mL), hydrocortisone (10-1000 ng/mL), (25-1000 IU/L), FSH (100-100,000 IU/L), TSH (1–100 mU/L), l-thyroxin Na (80.4 – 643.5 nmol/L), epidermal growth factor (10 –1000 ng/mL), fibroblast growth factor (10 –1000 ng/mL), and GRF (0.1–1000 ng/mL)—was assayed by adding 25 ␮l to each well.

Bioassay with Nb2 cell line Approximately 48 h before the start of the bioassays, the cells were transferred to the pre-assay medium (Fisher’s Medium supplemented with 1% horse serum, 50 ␮m 2-ME, and antibiotics) to slow down the rate cell of replication. After incubation, cells were collected by centrifugation (10 min at 800 rpm) and resuspended in assay medium (Fisher’s Medium supplemented with 10% horse serum, 50 ␮m 2-ME, and antibiotics) at a concentration of 1 ⫻ 105 cells/mL. Two-hundred microliter aliquots were distributed in each well of a 96-well microplate (Nalge Nunc International). Standard hGH was diluted with 0.01 m PBS supplemented with 0.1% BSA (Sigma) at each concentration. To each well were added 25 ␮l of standard or samples. The cultures were incubated in a CO2 incubator (5% CO2 ⫻ 95% air) for 48 h at 37 C. After incubation, the colorimetric end point was determined to be the same as the bioassay with the Ba/F3-hGHR cell line.

Subjects Serum samples were obtained from 23 normal children (14 boys and 9 girls) aged 0 yr, 10 months to 20 yr, 6 months (9 yr, 10 months ⫾ 3 yr, 11 months, mean ⫾ sd) and 10 non-GH deficient short children (6 boys and 4 girls) aged 4 yr, 10 months to 13 yr, 6 months (8 yr, 3 months ⫾ 2 yr, 11 months). Basal and peak hGH samples of GH provocative tests in non-GH-deficient short children (arginine tolerance test, GRH stim-

4275

ulation test, clonidine tolerance test) were also used in non-GH-deficient short children. Serum samples from a patient with severe short stature caused by bioinactive GH, reported previously by Takahashi et al. (14), were obtained after GRF stimulation (Table 1). The concentration of GH and IGF-I in serum sample from a patient from acromegaly were 24.9 ␮g/L and 440 ng/mL, respectively. Another serum sample was collected from a patient with precocious puberty on TSH provocative test (TRH). The concentration of PRL and GH were 47.8 ␮g/L and 6.9 ␮g/L, respectively. All samples were collected after obtaining informed consent.

Measurements Immunoactivity of hGH in serum was determined by IRMA (Daiichi Radio Isotope, Japan). GHBP was measured by ligand-mediated immunofunctional assay, described previously (15, 16). IGF-I was measured by Somatomedin C-RIA kit (Chiron, Yuka Medias Company Ltd.).

Results Growth stimulatory activity of hGH

hGH stimulated Ba/F3-hGHR cell proliferation in a dosedependent manner between 1 ␮g/L and 100 ␮g/L (Fig. 1), which was used as the standard curve for the following GH bioassay. None of the other hormones and growth factors, except IGF-I and bGH, showed any mitogenic activity. IGF-I stimulated the cell proliferation in a dose-dependent manner (Fig. 1), and the cell proliferation by IGF-I was suppressed by anti-IGF-1 antibody. The bGH increased cell proliferation at a concentration of more than 1000 ng/mL. The highest GHBP concentration (100 ng/mL) slightly suppressed the cell proliferation by hGH, but the average GHBP concentration (20 ng/mL) in the serum did not (Fig. 2). The growth stimulatory activity of hGH was completely blocked by anti-hGH antibody (Fig. 3). Bioassay of hGH in human serum

Human serum stimulated cell growth in the dilution range of 6.25–100% (the final concentration in the wells was 0.69 – 11%). The dose-dependent cell growth paralleled that produced by the standards (Fig. 4). The anti-hGH antibody also completely inhibited the stimulatory activity of serum samples. Figure 5 shows the correlation between bioactivity and immunoactivity in serum samples from normal children. There was a strong positive correlation between them, and the ratio of bioactivity to immunoactivity was 1.01 ⫾ 0.26 (mean ⫾ sd) (Fig. 5). The bioactivity of samples from non-GH-deficient short children also showed a close correlation with immunoactivity. The mean ratios of bioactivity/IRMA at basal values and TABLE 1. Immunoactivity and bioactivity in the patient with bioinactive GH Time (min)

Immunoactivity (ng/mL)

Bioactivity (ng/mL)

Immunoactivity/bioactivity

0 15 30 60 90

0.587 10.384 9.351 18.446 3.563

Not done 2.918 3.097 7.560 1.100

Not done 0.281 0.331 0.410 0.308

4276

JCE & M • 2000 Vol. 85 • No. 11

ISHIKAWA ET AL.

FIG. 1. Stimulation of the proliferation of Ba/F3-hGHR by hGH and other hormones or growth factors. Points are the mean values of triplicate (except IGF-I) or duplicate (IGF-I) wells (SD ⫾ 10% of the mean).

FIG. 2. Effect of GHBP on the proliferation of Ba/F3-hGHR cell cultures. Points are the mean values of duplicate wells (SD ⫾ 10% of the mean).

peak values in GH stimulation tests were 1.18 ⫾ 0.24 and 1.00 ⫾ 0.29 (mean ⫾ sd), respectively (Fig. 6). There was no significant difference in the mean ratios of bioactivity/IRMA between basal values and peak values. The bioactivity of samples measured by Ba/F3-hGHR showed a close correlation with the bioactivity by Nb2 assay (Fig. 7). The bioactivity of samples from a patient with severe short stature caused by bioinactive GH were lower than immunoactivity (0.33 ⫾ 0.06, mean ⫾ sd) (Table 1). The bioactivity of samples from acromegaly were almost equal to the immunoactivity and the ratio of bioactivity/IRMA was 1.04. The cell growth of diluted serum samples from acromegalic patient paralleled with the standards (Fig. 8). The ratio of bioactivity/IRMA of hyperprolactinemia with precocious puberty, which was under the TRH stimulation test, was 1.5.

FIG. 3. Effect of anti-GH antibody on the growth stimulatory activity of hGH. Ba/F3-hGHR cell cultures were incubated for 48 h with hGH in the presence or absence of anti-GH antibody. (The dilutions of the antibody were ⫻50 or ⫻100. The final concentrations of antibody were ⫻5000 or ⫻1000.)

FIG. 4. Effect of human serum on the proliferation of Ba/F3-hGHR cell cultures after 48-h incubation with and the anti-hGH antibody.

The mean GHBP concentrations in the sera from normal children and non-GH-deficient short children were lower than 20 ng/mL [5.1 ⫾ 3.8 ng/mL (2–17.4 ng/mL) and 4.9 ⫾ 3.4 ng/mL (0.8 –9.4 ng/mL), respectively] did not correlate with bioactivity. IGF-I concentrations in the serum from normal children and non-GH-deficient short children were 216.9 ⫾ 144.7 ng/mL (76 – 430 ng/mL) and 120.4 ⫾ 64.9 ng/mL (43–210 ng/mL), respectively, and did not correlate with bioactivity, either. Discussion

Among the hGH bioassays suitable for the measurement of serum GH bioactivity, radioreceptor assays and receptor modulation assays only reflect the binding of hGH to its receptor (7–9, 17), whereas the cell proliferation assay reflects

A BIOASSAY FOR hGH

4277

FIG. 5. A, Correlation between hGH bioactivity and immunoactivity in normal children. B, Ratio of bioassay/ IRMA. Mean value of bioassay to IRMA ratio was 1.013 ⫾ 0.260 (mean ⫾ SD).

FIG. 6. A, Correlation between hGH bioactivity and immunoactivity in short children. E, basal values; F, peak values. B, Ratio of bioassay/IRMA at basal and peak values. Mean value of bioassay to IRMA ratio was 1.181 ⫾ 0.243 or 0.997 ⫾ 0.293 (mean ⫾ SD), respectively.

not only receptor binding but also intracellular function activated by hGH. Although the Nb2 bioassay does not need a radioactive ligand, it becomes more convenient and useful after ESTA system modification (12, 13). However, bioactivity in the Nb2 bioassay is mediated through the PRL receptor (10, 18). The bioactivity of hGH is measurable by the Nb2 bioassay because hGH is the only animal GH that has lactogenic activity. The Nb2 bioassay, therefore, measures not somatotrophic but lactogenic activity of hGH, and the addition of anti-hPRL antibody to block hPRL lactogenic activity is necessary to measure the hGH bioactivity in the serum (10). Another method to measure hGH bioactivity has been reported: for example, suppressive activity of hGH on lipid accumulation has been measured using cell lines with hGHR (19). However, this method is not sensitive enough to establish a bioassay system (20) and does not detect soma-

totrophic activity (19). The Ba/F3-hGHR cell line is both suitable and sensitive enough to serve as an hGH bioassay. The Ba/F3-hGHR cell line proliferates dose-dependently when hGH is added, and cell proliferation is blocked by anti-hGH antibody. Cell proliferation is stimulated by no other hormone or growth factor but free IGF-I and bGH. Although free IGF-I did stimulate cell growth, the bioactivity of the serum sample did not correlate with total IGF-I concentrations but only with hGH concentration. In addition, the cell growth by diluted samples from acromegaly paralleled with the standards. Those demonstrate that the free IGF-I concentration is insufficient to affect the assay system under both normal and abnormal conditions, because most IGF-I binds to the binding proteins in serum (21–23). The cell proliferation was suppressed by anti-IGF-I antibody. The other factor influencing cell proliferation is GHBP,

4278

JCE & M • 2000 Vol. 85 • No. 11

ISHIKAWA ET AL.

FIG. 7. Correlation between hGH bioactivity measured with the Ba/ F3-hGHR cell line and with the Nb2 cell line.

drome and with non-GH-deficient short stature children, serum GH bioactivity measured by the suppression of lipid accumulation with GH on 3T3-F442A embryonic murine fibroblasts is reported to be higher than the GH concentration determined by RIA both in basal and peak values of the GH stimulation test (20). In healthy adult volunteers, serum GH bioactivity using the Nb2 bioassay is accompanied by bigger rises in bioactivity than in immunoactivity at the peak value of GH stimulation tests (25). Contrastingly, the bioactivity/ immunoactivity ratio is normal in peak GH samples by the provocative test in non-GH-deficient short stature and acromegaly (26, 27). Because the bioactivity of the Ba/F3hGHR cell line correlates with the bioactivity of the Nb2 cell line in normal and non-GH-deficient short children, this bioassay system confirms the Nb2 cell line bioassay system in use since the 1980s. Only two cases with severe growth retardation caused by bioinactive hGH confirmed by DNA analysis have been reported (14, 28). We measured bioactivity in the serum of one of these cases by the Ba/F3-hGHR cell line assay. The patient was a 3-year-old girl whose height was 79.4 cm (3.6 sd below the mean for her age and sex). A heterozygous single-based substitution (AG) in exon 4 of the GH-1 gene was found; the mutation is located in binding site 2 of the GH molecule to hGHR (14). The recombinant mutant GH is less potent than the wild-type GH in phosphorylation of tyrosine residues in hGHR (14). This demonstrates that the bioactivity measured by the Ba/F3-hGHR cell line assay reflects the somatogenic activity of mutant hGH. In summary, we have established a novel hGH bioassay that is specific for hGH, one that will be useful for screening of patients with short stature caused by biologically inactive hGH. References

FIG. 8. Effect of serum from acromegaly of the proliferation of Ba/ F3-hGHR cell line.

which also influences the Nb2 bioassay (24). Yet cell proliferation is not disturbed significantly when the GHBP concentration is less than 20 ng/mL. Therefore, to measure hGH bioactivity when the GHBP concentration in serum is higher than 20 ng/mL, GHBP should be added to the standard samples at the same concentration as in serum samples. Bovine GH increases cell proliferation at a concentration of over than 1000 ng/mL, but does not influence assay system because a very low concentration of bGH was used in assay medium 5% FCS and 0.1% BSA/PBS). In the samples determined by IRMA in our study, the bioactivity of serum samples from normal children and from non-GH-deficient short children is observed to be very close to the concentration of hGH. In patients with Turner syn-

1. Kowarski AA, Schneider J, Ben-Galim E, Welden VV, Daughady WH. 1978 Growth failure with normal serum RIA-GH and low somatomedin activity: somatomedin restoration and growth acceleration after exogenous GH. J Clin Endocrinol Metab. 47:461– 464. 2. Evans HM, Uyei N, Bartz QR, Simpson ME. 1938 The purification of the anterior pituitary growth hormone by fraction with ammonium sulfate. Endocrinology. 22:483– 492. 3. Marx W, Simpson ME, Evans HM. 1942 Bioassay of the growth hormone of the anterior pituitary. Endocrinology. 30:1–10. 4. Greesbeck MD, Palon AF. 1987 Highly improved precision of the hypophysectomized female rat body weight gain bioassay for growth hormone by increased frequency of injection, avoidance of antibody formation, and other simple modifications. Endocrinology. 120:2582–2590. 5. Bellini MH, Bartolini P. 1993 In vivo bioassay for the pituitary determination of human growth hormone dwarf “little” mice. Endocrinology. 132:2051–2055. 6. Greespan FS, Li CH, Simpson ME, Evans HM. 1949 Bioassay of hypophyseal growth hormone: the tibia test. Endocrinology. 45:455– 463. 7. Tsushima T, Friesen HG. 1973 Radioreceptor assay for growth hormone. J Clin Endocrinol Metab. 37:344 –337. 8. Lesniak MA, Roth J, Gorden JR. 1973 Human growth hormone radioreceptor assay using cultured human lymphocytes. Nature (New Biol). 241:20 –22. 9. Rosenfeld RG, Hiatz RL. 1980 Modulation of homologous receptor concentration: a sensitive radioassay for human growth hormone in acromegalic and newborn, and stimulated plasma. J Clin Endocrinol Metab. 50:62– 69. 10. Tanaka T, Shiu RPL, Gout PW, Bear CT, Noble RL, Friesen HG. 1980 A new sensitive and specific bioassay for lactogenic hormones: measurement of prolactin and growth hormone in human serum. J Clin Endocrinol Metab. 51:1058 –1063. 11. Wada M, Uchida H, Ikeda M, et al. 1998 The 20kDa human growth hormone (hGH) differs from the 22kDa hGH in the complex formation with cell surface hGH receptor and hGH-binding protein circulating in human plasma. Mol Endocrinol. 12:146 –156.

A BIOASSAY FOR hGH 12. Marshall NJ, Ealey PA, Gaines-Das R, Yateman ME, Claffey D, Holt SJ. 1990 The ESTA bioassay system. J Endocrinol. 127 (suppl): 2. 13. Ealey PA, Yateman ME, Sandhu R, et al. 1995 The development of an eluted stain bioassay (ESTA) for human growth hormone. Growth Regul. 5:36 – 44. 14. Takahashi Y, Shirono H, Arisaka O, et al. 1997 Biologically inactive growth hormone caused by an amino acid substitution. J Clin Invest. 100:1159 – 1165. 15. Calsson LMS, Rowland AM, Clark RG, Gesundheit N, Wong WLT. 1991 Ligand-mediated immunofunctional assay for quantitation of growth hormone-binding protein human blood. J Cin Endocrinol Metab. 73:1216 –1223. 16. Horikawa R, Tanaka T, Kokai Y, et al. 1994 Growth hormone-binding protein measurements in normal children and children with short stature using ligand-mediated immunofunction assay. Jpn Hoc Periatr Endocrinol. 3:35– 43. 17. Banghum DR. 1989 Assays for human growth hormones. J Phar Biomed Anal. 7:169 –172. 18. Hughes JP, Tanaka T, Gout PW, Beer CT, Noble RL, Friesen HG. 1982 Effect of ionization on human growth hormone and prolactin: characterized by bioassay, radioimmunoassay, radioreceptor assay, and electrophoresis. Endocrinology. 111:827– 832. 19. Rosewall EC, Mukku UR, Chen AB, et al. 1996 Novel assays based on human growth hormone receptor as alternative to the rat weight gain bioassay for recombinant human growth hormone. Biologicals. 24:25–39. 20. Foster CM, Borondy M, Padmanabhan V, et al. 1993 Bioactivity of human growth hormone in serum validation of an in vitro bioassay. Endocrinology. 132:2073–2081.

4279

21. Hizuka N, Takano K, Asakaw K, et al. 1991 Measurement of free form of insulin-like growth factor I inhuman plasma. Growth Regul. 1:51–55. 22. Hasegawa Y, Hasegawa T, Takeda M, Tsuchiya Y. 1996 Plasma free insulinlike growth factor I concentrations in growth hormone deficiency in growth hormone deficiency in children and adolescents. Eur J Endocrinol. 134:184 – 189. 23. Juul A, Holm K, Kastrup KW, et al. 1997 Free insulin-like growth factor I serum levels in 1430 healthy children and adults, and its diagnostic value in patients suspected of growth hormone deficiency. J Clin Endocrinol Metab. 82:2497–2502. 24. Dattani NT, Hindmarsh PC, Brook CGD, Robinson ICAF, Marshall NJ. 1994 Inhibition of growth hormone bioactivity by recombinant human growth hormone-binding protein in the eluted stain assay system. J Endocrinol. 140:445– 453. 25. Dattani MT, Hindmarsh PC, Pringle PJ, Brook CGD, Marshall NJ. 1995 The measurement of growth hormone bioactivity in patient serum using an eluted stain assay. J Clin Endocrinol Metab. 80:2675–2683. 26. Tanaka T, Shishiba Y, Gout PW, et al. 1983 Radioimmunoassay and bioassay of human growth hormone and human prolactin. J Clin Endocrinol Metab. 56:18 –24. 27. Tanaka T, Nakayama M, Hayashi K, et al. 1983 Bioactivity of serum growth hormone in children assessed by lymphoma cell bioassay. Clin Endocrinol. (Tokyo). 31:577–581. 28. Takahashi Y, Kaji H, Okimura Y, et al. 1996 Brief report. Short stature caused by a mutant growth hormone. N Eng J Med. 15:432– 436.