Molecular and Cellular Endocrinology,
11
13 (1990) 11-14
Elsevier Scientific Publishers Ireland, Ltd. MOLCEL 02358
The 20,000 Da variant of human growth hormone does not bind to growth
hormone receptors in human liver James McCarter, Melissa A. Shaw, Lori A. Winer and Gerhard Baumann Center for Endocrinology, Metabolism and Nutrition, Department of Medicine, Northwestern
University Medical School,
Chicago, IL 6061 I, U.S.A.
(Received 9 May 1990; accepted 26 June 1990)
Key words: Growth
hormone; Somatotropin; Growth hormone receptors (human liver); 20,000 Da variant of growth hormone
The 20,000 Da variant of human growth hormone (hGH) (2OK) exhibited no specific binding to hGH receptors in human liver plasma membranes. This contrasts with the 22,000 Da form of human growth hormone (22K), which bound with high affinity to the same hepatic receptor preparation. Since the liver is considered a major target organ for the somatogenic pathway of growth hormone action, this finding implies that in humans the 20K form plays little role in that pathway. The homologous hormone-receptor system examined here yielded results that differ from heterologous receptor binding experiments in animals. The differences are likely explained by the presence in non-primate mammals of more than one type of growth hormone receptor with varied specificities. In man, the 20K form of growth hormone may have a biologial role distinct from that of the main 22K form of growth hormone.
Introduction
Human growth hormone (hGH) is produced and secreted by the pituitary gland in multiple molecular forms (see Lewis et al., 1980a; Baumann, 1988 for review). The second most prevalent form is the 20,000 Da variant (2OK), which lacks amino acids 32-46 and arises by alternative mRNA splicing (Lewis et al., 1978, 1980b; DeNoto et al., 1981; Baumann and Stolar, 1986; Cooke et al., 1988). Since its discovery in 1978 (Lewis et al., 1978), the biological properties of 20K have been investigated in a variety of animal
Address for correspondence: G. Baumann, M.D., Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 60611, U.S.A.
systems. However, very little is known about the activity of 20K in man, and its biological role has remained somewhat of a mystery. In the present study, we examined the binding of 20K to receptors in human liver, a major target organ for hGH. The study was prompted by the finding that the binding of 20K to circulating growth hormone (GI-I) binding proteins, one of which is related to the hepatic GH receptor (Leung et al., 1987), differed from that of the main, 22,000 Da hGH form (22K) (Baumann and Shaw, 1989). Materials and methods
Human livers were obtained at autopsy (8-12 h post mortem) from two persons deceased from causes not thought to affect liver function. Fresh frozen human liver, obtained at the time of liver
0303-7207/90/$03.50 6 1990 Ekevier Scientific Publishers Ireland, Ltd.
transplant, was a gift from Dr. P. Whittington, University of Chicago. Liver plasma membranes were prepared immediately according to the procedure of Tsushima and Friesen (1973) and Carr and Friesen (1976). Microsomal membranes (100,000 x g pellet) were stored in aliquots at -70°C for up to 3 weeks until assay. Natural pituitary hGH (22K and 20K) were gifts from Dr. U.J. Lewis (La Jolla, CA). The 20K preparation (bioactivity 1.7 U/mg in the rat tibial line assay) contained 3% 22K, as assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) and densitometry. Biosynthetic methionyl20K was a gift from Drs. G.W. Becker and J.H. Holcombe, Eli Lilly and Co. (Indianapo~s, IN). The molecular integrity of both the 22K and the 20K preparation was verified by determining their ability to bind to GH receptors in rabbit liver membranes (Tsushima and Friesen, 1973); all were found to be receptor-active in that system. hGH was iodinated with i2? by a lactoperoxidase method, and > 98% monomeric tracers prepared by Sephadex G-100 chromatography as described (Baumann et al., 1986; Baumann and Shaw, 1990). Conditions for radioreceptor assay were as described by Tsushima and Friesen (1973) and Carr and Friesen (1976). Briefly, liver membranes (600 pg protein) were incubated in duplicate with [‘251]hGH (22K or 20K) and varying concentrations of unlabeled 22K or 20K in a final volume of 0.5 ml 25 mM T&s-Cl, pH 7.6, 20 mM MgSO,, 10 mM CaCI,, 0.1% bovine serum albumin (BSA) for 24 h at 4*C in a shaking water bath. At the end of the incubation, 2.5 ml cold 25 mM Naacetate, pH 5.4, 0.1% BSA was added, the mixture was centrifuged at 4”C, the supernatant aspirated and the pellet counted. Non-specific binding was defined as residual binding in the presence of 2000 ng/rnl 22K. The three liver preparations were tested separately; all gave comparable specific binding results. Data were therefore pooled and are expressed as means reflecting all liver preparations. Experiments were also conducted with liver membranes that had been desaturated by exposure to 4 M MgCl, for 10 min, according to the procedure of Kelly et al. (1979). Results obtained with desaturated membranes were corrected for protein loss.
Results
The relative ability of 22K and 20K to displace [‘251]22K hGH from receptors in human liver membranes is shown in Fig. 1. The 22K binding/displacement curve had an apparent Kd of 1.4 nM. In contrast, natural 20K showed very weak displacing ability, with an apparent Kd of 45 nM. This displacing ability could be entirely accounted for by the content of 22K in the 20K preparation used. Biosynthetic 2OK, which is devoid of 22K contamination, showed no displacing activity up to a concentration of 10 ,ug/ml. To further probe the binding of 20K (or lack thereof) to liver, analogous binding studies were conducted with [‘251]20K as the tracer. With that tracer no specific binding was detected in any of the liver preparations. The use of MgCl,-desaturated liver membranes yielded slightly (1%20%) higher binding values for 22K, but again 20K failed to show any intrinsic binding activity. Discussion
The present study demonstrates that the 20K variant of hGH does not bind appreciably to GH receptors in human liver. This contrasts with re100-I
80-
?
2 f
.!3
60-
40-
ii *
20-
I
I
I
IO
I
Xx)
I
ICOO
I
1omo
hGH Cnglml)
Fig. 1. Bind~g/~~pla~ment curves for 22K and 20K hGH in human liver membranes. [“‘1]22K is used as the tracer. 20K (nat.) denotes pituitary-derived 20K, which contains 3% 22K. 20K (synth.) denotes pure biosynthetic methionyL20K. Points represent mean values of five different experiments (three in the case of synthetic 20K). B/B, denotes the bound fraction relative to maximal binding.
13
sults reported with GH receptors in animals, such as rabbit and rat liver (Sigel et al., 1981; Wohnlich and Moore, 1982; Closset et al., 1983; Hughes et al., 1983), and rat adipocytes (Smal et al., 1987). However, in all these systems, 20K bound with substantially lower affinity than 22K, and part of the apparent binding may have been due to varying degrees of contamination of the 20K preparations used with 22K. The only other report on the interaction of 20K with a human receptor is that of Smal et al. (1985), employing cultured IM-9 lymphoblasts as a receptor source. In that system, 20K bound with about half the affinity of 22K, a finding clearly different from the present results. It is not clear whether the GH receptor in IM-9 cells, an established cancerous cell line, is representative of GH receptors in normal human cells, or in non-lymphatic tissues. (Normal resting or mitogen-stimulated human lymphocytes do not express detectable amounts of GH receptors (Baumann G., unpublished results).) Our present data suggest that GH receptors in human liver differ in their specificity for 20K from those in IM-9 cells. The results with human liver are congruent with our earlier work on the high affinity GH binding protein in human plasma (Baumann et al., 1986). The binding protein, a soluble and truncated hepatic GH receptor encompassing the extracellular domain (Leung et al., 1987), bound 20K weakly with an apparent affinity 20-fold lower than 22K (Baumann et al., 1986). That degree of binding can also be largely attributed to contamination of the 20K preparation with 22K. In plasma, 20K binds to another, specific BP that is distinct from the hepatic receptor-related BP (Baumann and Shaw, 1989, 1990). The possible relationship of that BP to a 20K-specific receptor remains speculative. Based on the present data, the liver does not appear as a likely site for such a receptor. This study also shows that previous results obtained in animals are difficult to extrapolate to humans. Studies in rat and rabbit liver showed that 20K is nearly as effective as 22K (70% and 94% respectively) at displacing labeled rut GH (Hughes et al., 1983). However, when hGH (22K) was used as the tracer, the effectiveness of 20K was only about 3-12% (Sigel et al., 1981; Wohnlich and Moore, 1982; Hughes et al., 1983). Hughes
et al. (1983) suggested that rat GH and hGH are identifying two distinct populations of receptors with very different affinities for 20K in rabbit liver. In rat adipocytes, 20K bound with only 3% of the affinity of 22K (Smal et al., 1987). In rabbit mammary glands, 20K exhibited 22-1008 of the binding activity of 22K or ovine prolactin (Sigel et al., 1981; Closset et al., 1983). The findings in various animal systems have been difficult to interpret or reconcile with each other; they suggest the existence of several types of GH receptors. More important, however, is the realization that ultimately relevant data about a species-specific hormone, such as hGH, can only be secured in a homologous system. In this regard, the present study may serve to clarify a difficult area. The biological role of 20K remains unknown. The fact that this GH variant is evolutionarily preserved (Sinha and Gilligan, 1984; Howland et al., 1987; Sinha, 1987) suggests that it has biological significance. Although 20K variants in animals have been identified, they have not yet been biologically characterized. Human 20K has comparable activity to 22K as a growth promoter and as a lactogen in the hypophysectomized rat (Lewis et al., 1978; Spencer et al., 1981; Kostyo et al., 1987; Cameron et al., 1988), but has diminished ‘insulin-like’ activity (Frigeri et al., 1979; Kostyo et al., 1985, 1987; Cameron et al., 1988). Nothing is known about its bioactivity in man. In view of the present receptor data, its growth promoting activity in rats may not necessarily indicate similar activity in man. Rather, this activity in rats may be due to ‘crosstalk’ of human 20K with somatogenie rat GH receptors (Hughes et al., 1983), a property which appears to be absent in man. If indeed the liver is the major target organ for somatogenic GH activity, our data may be taken to indicate that 20K has very little somatogenic potential in man. Its biological role may instead lie in an as yet unrecognized arena. To address these issues, biological testing of 20K in humans will be required. Acknowledgements This work was supported by NIH grants DK38128, DK07169, and a grant from Northwestern Memorial Foundation.
14
References Baumann, G. (1988) in Basic and Clinical Aspects of Growth Hormone (Bercu, B.B., ed.), pp. 13-31, Plenum, New York. Baumann, G. and Shaw, M.A. (1989) Program 71st Meeting Endocrine Society, Abstract no. 1653. Baumann, G. and Shaw, M.A. (1990) J. CIin. Endocrinol. Metab. 70, 680-686. Baumann, G. and Stolar, M.W. (1986) J. Chn. Endocrinol. Metab. 62, 789-790. Baumann, G., Stolar, M.W., Ambum, K., Barsano, C.P. and DeVries, B.C. (1986) J. CIin. Endocrinol. Metab. 62, 134141. Cameron, CM., Kostyo, J.L., Adamafio, N.A., Brostedt, P., Roos, P., Skottner, A., Forsman, A., FrykIund, L. and Skoog, B. (1988) Endocrinology 122, 471-474. Carr, D. and Friesen, H.G. (1976) J. CIin. Endocrinol. Metab. 42, 484-493. Closset, J., Smal, J., Gomez, F. and Hennen, G. (1983) Biothem. J. 214, 885-892. Cooke, N.E., Ray, J., Watson, M.A., Estes, P.A., Kuo, B.A. and Liebhaber, S.A. (1988) J. Clin. Invest. 82, 270-275. DeNoto, F.M., Moore, D.D. and Goodman, H.M. (1981) Nucleic Acids Res. 9, 3719-3730. Frigeri, L.G., Peterson, S.M. and Lewis, U.J. (1979) B&hem. Biophys. Res. Cornmun. 91, 778-782. Howland, D.S., Farrington, M.A., Taylor, W.D. and Hymer, W.C. (1987) B&hem. Biophys. Res. Commun. 147, 650657. Hughes, J.P., Tokuhir, E., Simpson, J.S.A. and Friesen, H.G. (1983) Endocrinology 113, 1904-1906. Kelly, P.A., Leblanc, G. and Djiane, J. (1979) Endocrinology 104, 1631-1638.
Koszyo, J.L., Cameron, C.M., Olson, K.C., Jones, A.J.S. and Pai, R.C. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 42504253. Kostyo, J.L., Skottner, A., Brostedt, P., Roos, P., Cameron, C.M., Forsman, A., Fryklund, L., Adamafio, N.A. and Skoog, B. (1987) B&him. Biophys. Acta 925, 314-324. Leung, D.W., Spencer, S.A., Cachianes, G., Hammonds, R.G., Collins, C., Henzel, W.J., Barnard, R., Waters, M.J. and Wood, W.I. (1987) Nature 330, 537-543. Lewis, U.J., Dunn, J.T., BonewaId, L.F., Seavey, B.K. and VanderLaan, W.P. (1978) J. Biol. Chem. 253, 2679-2687. Lewis, U.J., Sin& R.N.P., Tutweiler, G.F., Sigel, M.B., VanderLaan, E.F. and VanderLaan, W.P. (1980a) Recent Prog. Horm. Res. 36, 477-508. Lewis, U.J., Bonewald, L.F. and Lewis, L.J. (1980b) Biochem. Biophys. Res. Commun. 92, 511-516. Sigel, M.B., Thorpe, N.A., Kobrin, MS., Lewis, U.J. and VanderLaan, W.P. (1981) Endocrinology 108, 1600-1603. Sinha, Y.N. (1987) Clin. Res. 35, 183A (Abstract). Sinha, Y.N. and Gilligan, T.A. (1984) Proc. Sot. Exp. Biol. Med. 177, 465-474. Smal, J., Closset, J., Hennen, G. and DeMeyts, P. (1985) B&hem. J. 225, 283-289. Smal, J., Closset, J., Hennen, G. and DeMeyts, P. (1987) J. Biol. Chem. 262, 11071-11079. Spencer, E.M., Lewis, L.J. and Lewis, U.J. (1981) Endocrinology 109, 1301-1302. Tsushima, T. and Friesen, H.G. (1973) J. CIin. Endocrinol. Metab. 31, 334-331. Wohnhch, L. and Moore, W.V. (1982) Horm. Metab. Res. 14, 138-141.
11
13 (1990) 11-14
Elsevier Scientific Publishers Ireland, Ltd. MOLCEL 02358
The 20,000 Da variant of human growth hormone does not bind to growth
hormone receptors in human liver James McCarter, Melissa A. Shaw, Lori A. Winer and Gerhard Baumann Center for Endocrinology, Metabolism and Nutrition, Department of Medicine, Northwestern
University Medical School,
Chicago, IL 6061 I, U.S.A.
(Received 9 May 1990; accepted 26 June 1990)
Key words: Growth
hormone; Somatotropin; Growth hormone receptors (human liver); 20,000 Da variant of growth hormone
The 20,000 Da variant of human growth hormone (hGH) (2OK) exhibited no specific binding to hGH receptors in human liver plasma membranes. This contrasts with the 22,000 Da form of human growth hormone (22K), which bound with high affinity to the same hepatic receptor preparation. Since the liver is considered a major target organ for the somatogenic pathway of growth hormone action, this finding implies that in humans the 20K form plays little role in that pathway. The homologous hormone-receptor system examined here yielded results that differ from heterologous receptor binding experiments in animals. The differences are likely explained by the presence in non-primate mammals of more than one type of growth hormone receptor with varied specificities. In man, the 20K form of growth hormone may have a biologial role distinct from that of the main 22K form of growth hormone.
Introduction
Human growth hormone (hGH) is produced and secreted by the pituitary gland in multiple molecular forms (see Lewis et al., 1980a; Baumann, 1988 for review). The second most prevalent form is the 20,000 Da variant (2OK), which lacks amino acids 32-46 and arises by alternative mRNA splicing (Lewis et al., 1978, 1980b; DeNoto et al., 1981; Baumann and Stolar, 1986; Cooke et al., 1988). Since its discovery in 1978 (Lewis et al., 1978), the biological properties of 20K have been investigated in a variety of animal
Address for correspondence: G. Baumann, M.D., Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 60611, U.S.A.
systems. However, very little is known about the activity of 20K in man, and its biological role has remained somewhat of a mystery. In the present study, we examined the binding of 20K to receptors in human liver, a major target organ for hGH. The study was prompted by the finding that the binding of 20K to circulating growth hormone (GI-I) binding proteins, one of which is related to the hepatic GH receptor (Leung et al., 1987), differed from that of the main, 22,000 Da hGH form (22K) (Baumann and Shaw, 1989). Materials and methods
Human livers were obtained at autopsy (8-12 h post mortem) from two persons deceased from causes not thought to affect liver function. Fresh frozen human liver, obtained at the time of liver
0303-7207/90/$03.50 6 1990 Ekevier Scientific Publishers Ireland, Ltd.
transplant, was a gift from Dr. P. Whittington, University of Chicago. Liver plasma membranes were prepared immediately according to the procedure of Tsushima and Friesen (1973) and Carr and Friesen (1976). Microsomal membranes (100,000 x g pellet) were stored in aliquots at -70°C for up to 3 weeks until assay. Natural pituitary hGH (22K and 20K) were gifts from Dr. U.J. Lewis (La Jolla, CA). The 20K preparation (bioactivity 1.7 U/mg in the rat tibial line assay) contained 3% 22K, as assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) and densitometry. Biosynthetic methionyl20K was a gift from Drs. G.W. Becker and J.H. Holcombe, Eli Lilly and Co. (Indianapo~s, IN). The molecular integrity of both the 22K and the 20K preparation was verified by determining their ability to bind to GH receptors in rabbit liver membranes (Tsushima and Friesen, 1973); all were found to be receptor-active in that system. hGH was iodinated with i2? by a lactoperoxidase method, and > 98% monomeric tracers prepared by Sephadex G-100 chromatography as described (Baumann et al., 1986; Baumann and Shaw, 1990). Conditions for radioreceptor assay were as described by Tsushima and Friesen (1973) and Carr and Friesen (1976). Briefly, liver membranes (600 pg protein) were incubated in duplicate with [‘251]hGH (22K or 20K) and varying concentrations of unlabeled 22K or 20K in a final volume of 0.5 ml 25 mM T&s-Cl, pH 7.6, 20 mM MgSO,, 10 mM CaCI,, 0.1% bovine serum albumin (BSA) for 24 h at 4*C in a shaking water bath. At the end of the incubation, 2.5 ml cold 25 mM Naacetate, pH 5.4, 0.1% BSA was added, the mixture was centrifuged at 4”C, the supernatant aspirated and the pellet counted. Non-specific binding was defined as residual binding in the presence of 2000 ng/rnl 22K. The three liver preparations were tested separately; all gave comparable specific binding results. Data were therefore pooled and are expressed as means reflecting all liver preparations. Experiments were also conducted with liver membranes that had been desaturated by exposure to 4 M MgCl, for 10 min, according to the procedure of Kelly et al. (1979). Results obtained with desaturated membranes were corrected for protein loss.
Results
The relative ability of 22K and 20K to displace [‘251]22K hGH from receptors in human liver membranes is shown in Fig. 1. The 22K binding/displacement curve had an apparent Kd of 1.4 nM. In contrast, natural 20K showed very weak displacing ability, with an apparent Kd of 45 nM. This displacing ability could be entirely accounted for by the content of 22K in the 20K preparation used. Biosynthetic 2OK, which is devoid of 22K contamination, showed no displacing activity up to a concentration of 10 ,ug/ml. To further probe the binding of 20K (or lack thereof) to liver, analogous binding studies were conducted with [‘251]20K as the tracer. With that tracer no specific binding was detected in any of the liver preparations. The use of MgCl,-desaturated liver membranes yielded slightly (1%20%) higher binding values for 22K, but again 20K failed to show any intrinsic binding activity. Discussion
The present study demonstrates that the 20K variant of hGH does not bind appreciably to GH receptors in human liver. This contrasts with re100-I
80-
?
2 f
.!3
60-
40-
ii *
20-
I
I
I
IO
I
Xx)
I
ICOO
I
1omo
hGH Cnglml)
Fig. 1. Bind~g/~~pla~ment curves for 22K and 20K hGH in human liver membranes. [“‘1]22K is used as the tracer. 20K (nat.) denotes pituitary-derived 20K, which contains 3% 22K. 20K (synth.) denotes pure biosynthetic methionyL20K. Points represent mean values of five different experiments (three in the case of synthetic 20K). B/B, denotes the bound fraction relative to maximal binding.
13
sults reported with GH receptors in animals, such as rabbit and rat liver (Sigel et al., 1981; Wohnlich and Moore, 1982; Closset et al., 1983; Hughes et al., 1983), and rat adipocytes (Smal et al., 1987). However, in all these systems, 20K bound with substantially lower affinity than 22K, and part of the apparent binding may have been due to varying degrees of contamination of the 20K preparations used with 22K. The only other report on the interaction of 20K with a human receptor is that of Smal et al. (1985), employing cultured IM-9 lymphoblasts as a receptor source. In that system, 20K bound with about half the affinity of 22K, a finding clearly different from the present results. It is not clear whether the GH receptor in IM-9 cells, an established cancerous cell line, is representative of GH receptors in normal human cells, or in non-lymphatic tissues. (Normal resting or mitogen-stimulated human lymphocytes do not express detectable amounts of GH receptors (Baumann G., unpublished results).) Our present data suggest that GH receptors in human liver differ in their specificity for 20K from those in IM-9 cells. The results with human liver are congruent with our earlier work on the high affinity GH binding protein in human plasma (Baumann et al., 1986). The binding protein, a soluble and truncated hepatic GH receptor encompassing the extracellular domain (Leung et al., 1987), bound 20K weakly with an apparent affinity 20-fold lower than 22K (Baumann et al., 1986). That degree of binding can also be largely attributed to contamination of the 20K preparation with 22K. In plasma, 20K binds to another, specific BP that is distinct from the hepatic receptor-related BP (Baumann and Shaw, 1989, 1990). The possible relationship of that BP to a 20K-specific receptor remains speculative. Based on the present data, the liver does not appear as a likely site for such a receptor. This study also shows that previous results obtained in animals are difficult to extrapolate to humans. Studies in rat and rabbit liver showed that 20K is nearly as effective as 22K (70% and 94% respectively) at displacing labeled rut GH (Hughes et al., 1983). However, when hGH (22K) was used as the tracer, the effectiveness of 20K was only about 3-12% (Sigel et al., 1981; Wohnlich and Moore, 1982; Hughes et al., 1983). Hughes
et al. (1983) suggested that rat GH and hGH are identifying two distinct populations of receptors with very different affinities for 20K in rabbit liver. In rat adipocytes, 20K bound with only 3% of the affinity of 22K (Smal et al., 1987). In rabbit mammary glands, 20K exhibited 22-1008 of the binding activity of 22K or ovine prolactin (Sigel et al., 1981; Closset et al., 1983). The findings in various animal systems have been difficult to interpret or reconcile with each other; they suggest the existence of several types of GH receptors. More important, however, is the realization that ultimately relevant data about a species-specific hormone, such as hGH, can only be secured in a homologous system. In this regard, the present study may serve to clarify a difficult area. The biological role of 20K remains unknown. The fact that this GH variant is evolutionarily preserved (Sinha and Gilligan, 1984; Howland et al., 1987; Sinha, 1987) suggests that it has biological significance. Although 20K variants in animals have been identified, they have not yet been biologically characterized. Human 20K has comparable activity to 22K as a growth promoter and as a lactogen in the hypophysectomized rat (Lewis et al., 1978; Spencer et al., 1981; Kostyo et al., 1987; Cameron et al., 1988), but has diminished ‘insulin-like’ activity (Frigeri et al., 1979; Kostyo et al., 1985, 1987; Cameron et al., 1988). Nothing is known about its bioactivity in man. In view of the present receptor data, its growth promoting activity in rats may not necessarily indicate similar activity in man. Rather, this activity in rats may be due to ‘crosstalk’ of human 20K with somatogenie rat GH receptors (Hughes et al., 1983), a property which appears to be absent in man. If indeed the liver is the major target organ for somatogenic GH activity, our data may be taken to indicate that 20K has very little somatogenic potential in man. Its biological role may instead lie in an as yet unrecognized arena. To address these issues, biological testing of 20K in humans will be required. Acknowledgements This work was supported by NIH grants DK38128, DK07169, and a grant from Northwestern Memorial Foundation.
14
References Baumann, G. (1988) in Basic and Clinical Aspects of Growth Hormone (Bercu, B.B., ed.), pp. 13-31, Plenum, New York. Baumann, G. and Shaw, M.A. (1989) Program 71st Meeting Endocrine Society, Abstract no. 1653. Baumann, G. and Shaw, M.A. (1990) J. CIin. Endocrinol. Metab. 70, 680-686. Baumann, G. and Stolar, M.W. (1986) J. Chn. Endocrinol. Metab. 62, 789-790. Baumann, G., Stolar, M.W., Ambum, K., Barsano, C.P. and DeVries, B.C. (1986) J. CIin. Endocrinol. Metab. 62, 134141. Cameron, CM., Kostyo, J.L., Adamafio, N.A., Brostedt, P., Roos, P., Skottner, A., Forsman, A., FrykIund, L. and Skoog, B. (1988) Endocrinology 122, 471-474. Carr, D. and Friesen, H.G. (1976) J. CIin. Endocrinol. Metab. 42, 484-493. Closset, J., Smal, J., Gomez, F. and Hennen, G. (1983) Biothem. J. 214, 885-892. Cooke, N.E., Ray, J., Watson, M.A., Estes, P.A., Kuo, B.A. and Liebhaber, S.A. (1988) J. Clin. Invest. 82, 270-275. DeNoto, F.M., Moore, D.D. and Goodman, H.M. (1981) Nucleic Acids Res. 9, 3719-3730. Frigeri, L.G., Peterson, S.M. and Lewis, U.J. (1979) B&hem. Biophys. Res. Cornmun. 91, 778-782. Howland, D.S., Farrington, M.A., Taylor, W.D. and Hymer, W.C. (1987) B&hem. Biophys. Res. Commun. 147, 650657. Hughes, J.P., Tokuhir, E., Simpson, J.S.A. and Friesen, H.G. (1983) Endocrinology 113, 1904-1906. Kelly, P.A., Leblanc, G. and Djiane, J. (1979) Endocrinology 104, 1631-1638.
Koszyo, J.L., Cameron, C.M., Olson, K.C., Jones, A.J.S. and Pai, R.C. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 42504253. Kostyo, J.L., Skottner, A., Brostedt, P., Roos, P., Cameron, C.M., Forsman, A., Fryklund, L., Adamafio, N.A. and Skoog, B. (1987) B&him. Biophys. Acta 925, 314-324. Leung, D.W., Spencer, S.A., Cachianes, G., Hammonds, R.G., Collins, C., Henzel, W.J., Barnard, R., Waters, M.J. and Wood, W.I. (1987) Nature 330, 537-543. Lewis, U.J., Dunn, J.T., BonewaId, L.F., Seavey, B.K. and VanderLaan, W.P. (1978) J. Biol. Chem. 253, 2679-2687. Lewis, U.J., Sin& R.N.P., Tutweiler, G.F., Sigel, M.B., VanderLaan, E.F. and VanderLaan, W.P. (1980a) Recent Prog. Horm. Res. 36, 477-508. Lewis, U.J., Bonewald, L.F. and Lewis, L.J. (1980b) Biochem. Biophys. Res. Commun. 92, 511-516. Sigel, M.B., Thorpe, N.A., Kobrin, MS., Lewis, U.J. and VanderLaan, W.P. (1981) Endocrinology 108, 1600-1603. Sinha, Y.N. (1987) Clin. Res. 35, 183A (Abstract). Sinha, Y.N. and Gilligan, T.A. (1984) Proc. Sot. Exp. Biol. Med. 177, 465-474. Smal, J., Closset, J., Hennen, G. and DeMeyts, P. (1985) B&hem. J. 225, 283-289. Smal, J., Closset, J., Hennen, G. and DeMeyts, P. (1987) J. Biol. Chem. 262, 11071-11079. Spencer, E.M., Lewis, L.J. and Lewis, U.J. (1981) Endocrinology 109, 1301-1302. Tsushima, T. and Friesen, H.G. (1973) J. CIin. Endocrinol. Metab. 31, 334-331. Wohnhch, L. and Moore, W.V. (1982) Horm. Metab. Res. 14, 138-141.
