Designing Foods: Animal Product Options in the Marketplace (1988)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Hormonal Regulation of Growth F. C. LEUNG Animal growth is a complex physiological process regulates] by the endocrine system (Figure 1), which also mediates the effects of nutritional, environmental, and genetic factors in animals. To enhance growth and improve feed conversion efficiency in agri- cultural animals, scientists must understand the roles of hormones (pepticle and steroid) and peptide growth factors in these pro- cesses and identify the limiting factors so that these processes can be modulated. The hormones that affect growth in ani- mals are growth hormone, insulin, thyroid hormones, glucocorticoids, prolactin, and gonaclal steroids (androgens and estrogens). Their role in growth and development has traditionally been investigated by examining the effect of hormone deprivation after organ ablation; the effects of excess amounts of hormones can be observed by administering the hormones to animals in vivo. Growth hormone (GH) is generally be- lieved to be the most important hormone affecting growth and development. Clinical observations show that GH deficiency in children results in dwarfism ant] that excess GH results in acromegaly and gigantism (Underwood! and Van Wyk, 1981~. This has 135 led to the assumption that an increase in the circulating concentration of GH would result in faster growth. This hypothesis has been confirmed] by the gene insertion tech- nique. Palmiter et al. (1983) proclucec] trans- genic mice by direct injection of cloned rat GH or human GH recombinant DNA, li- gated with a mouse metallothionein pro- moter, into the pronuclei of fertilized eggs. Transgenic mice that carried the extra GH gene, and that therefore had high circulating concentrations of GH, grew to twice the size of their control littermates. Hammer et al. (1984) also user! this technique to correct dwarfism in a strain of"Little" mice, which are deficient in GH; the transgenic mice grew even larger than normal mice. Injected GH has been reported to im- prove the growth rate and feed conversion efficiency of normal pigs (Chung et al., 1985; Machlin, 1972), calves (Brumby, 1959), and lambs (Wagner and Veenhuizen, 1978~. Administration of GH to dairy cows report- edly increases the efficiency of milk pro- cluction (see the papers by Gorewit ant]  Linn in this volume), and in pigs and lambs shifts carcass composition from fat toward protein ant] moisture (Chung et al., 1985;

136 Effects ~ I Hypothalamus Growth Hormone Releasing Somatostatins Factor ~ 1 Indirect Effects , ~ , E] ~ I Other Organs I Skeletal: Chondrogenesis, Skeletal Growth Extraskeletal: DNA-, RNA-, Protein Synthesis; Cell Prollferatlon 1 ~ Direct Effects 1 A+ |Other Organs | Effects Glucose Transport Glucose Homeostasis Amino Acid Transport Upolysis RNA- Protein Synthesis (Ever) FIGURE 1 Regulation and effects of growth hor- mone. Machlin, 1972; Wagner and Veenhuizen, 1978~. The effects of exogenous GH on growth in fish (salmon and trout) and chick- ens have recently been reported by Kawau- chi et al. (1986) and Leung et al. (1986b). However, responses in these animals were much less marked than those observed in . . transgen~c mice. To investigate the impact of increased circulating GH concentration on growth and feed efficiency, Leung et al. (1986b) used various experimental approaches to manip- ulate the endocrine systems of chicks. A discussion of their methodologies and re- sults follows.  THE INFLUENCE OF GlI ON GROWTH Pituitary GH synthesis and reaction are generally believed to be regulates! by the hypothalamic releasing factor, GH releasing factor (GRF) and inhibiting factor, and so- matotropin releasing/inhibiting factor. In avian species, a thirc] hypothalamic factor, thyrotropin releasing hormone (TRH), which stimulates thyrotropin stimulating hormone APPENDIX at the pituitary level, is also a potent GH releaser (Harvey et al., 1978~. In contrast to mammalian species, where there is only one GH releasing factor, avian species ap- pear to have two. It is widely thought that the lipolytic effect of GH is direct but that somatomedin-C (SM-C) mediates the growth response of GH (Chawla et al., 1983; Un- clerwood and Van Wyk, 1981~. There is also evidence that GH may act directly in the tibia to promote bone growth (Isaksson et al., 1982; Russell and Spencer, 1985~. The various experimental methods used to ele- vate serum concentrations of GH are listed in Table 1. Effects of Chicken GH on Body Weight Gain in Chickens Large quantities of chicken pituitary GH were purified to examine its eject on growth (Leung et al., 1986b). The purified chicken GH (cGH), which was biologically active in the rat tibia bioassay, gave a dose-dependent response parallel to that of the bovine GH stanclard. The amino acid composition of cGH was similar to that of mammalian GH, and particle-sequencing analysis of cGH shower! 79 percent homology with bovine GH. Four-week-old Hubbard x Hubbard broiler cockerels were user] in all experi- ments. Thirty-six bircis were individually caged in a temperature- and light-controlled TABLE 1 Methods for Elevating Serum Concentration of Growth Hormone I. Treat with GH. 2. Treat with GRF for TRH. 3. Increase secretion of endogenous GRF or TRH by control of neuroregulators. 4. Decrease secretion or action of endogenous so- matomedin releasing/inhibiting factor. 5. Increase secretion of endogenous GRF or GH by inserting multiple copies oftheir structural genes, linked to an appropriate promoter, into the chicken genome.

HORMONAL REGULATION OF GROWTH room; they were randomly divided into four treatment groups of nine birds each, with food and water available ad libitum. The purified cGH was dissolved in phys- iological saline and given daily by intrave- nous injection via the brachial vein at concentrations of 5, 10, ant! 50 ~g/bird in 100- volumes. Body Sleight ant! feed con- sumption were recorded twice weekly for 2 weeks. At the end of the experiment, birds were killed, clefeathered, and ground in a meat grinder. Tissues were analyzer! by New Jersey Feet] Laboratory, Inc. (Plains- field, Nail.), for moisture, protein, ant! fat content, according to the procedure rec- ommenclec] by the Association of Official Analytical Chemists. Birds that received 5 log of cGH daily shower! significant weight gains (20.6 and 13.5 percent over control birds) on clays 3 and 6, respectively. Birds that received 10 fig of cGH also showed significant weight gains over control bircis after 3 and 6 days of treatment (19.6 and 11.3 percent, re- spectively). Birds that received 50 log of cGH shower! an improvement in weight gain over control bircis, but the increase was not statistically significant. Overall, the increase in body weight gain seemed to be transient, so that the stimulating effect of cGH was diminished by the end of the experiment. There was no difference in the effect of feed conversion efficiency on car- cass composition between cGH-treated and control birds. Effects of Human Pancreatic GRF and TRH on Body Weight Gain in Chickens Chicken hypothalamic GRF has not yet been isolated and purified, but a synthetic human pancreatic GRF (hpGRF) has been shown to be active in stimulating cGH release in chickens both in viva and in vitro (Leung ant! Taylor, 1983; Scanes et al., 1984~. In addition, TRH, which is a hypo- thalamic peptide, has been shown to stim 137 ulate cGH release in viva. The objective of the studies clescribed below was to deter- mine the effect of hypothalamic peptides on growth in chickens. Four-week-olc! Hubbard x Hubbard broiler cockerels were used in all experi- ments. In the hpGRF experiment, bircis were individually caged and randomly dis- tributec3 into four treatment groups of nine birds each. In the TRH experiment, bircis were individually cager! and randomly di- vided into four treatment groups of 8 to 10 birds. All birds were housed in a tempera- ture- ant! light-controllecl room (25°C; 14 hours of light, 10 hours of darkness) and provided with food and water ad libitum.  Food consumption and weight were re- corded twice weekly for 2 weeks. At the end of the experiment, birds were killed and defeathered, and carcass composition was analyzed as described in the previous section. The hpGRF44 (Bachem, Torrance, Calif. ~ and TRH (Beckman, Palo Alto, Calif. ~ were dissolved in physiological saline and injected via the brachial vein at concentra- tions of 0. 1, 1. 0, or 10.0 Catbird in a 100- ~1 volume. Control birds received 100 Al of a saline solution. Birds that received 0.1 log of hpGRF daily showed a significant increase in body weight gain early on, but that soon diminished. The similarly transient stimulating effect of cGH and hpGRF on body weight gain suggests that hpGRF is also mediated through pituitary GH. Birds that received 1.0 or 10.0 log of TRH dally showed significant increases in body weight compared to controls. In contrast to the effect of hpGRF, the growth response to TRH injections was not transient (Leung et al., 1984c). The difference between the effects of the two hormones is probably due to the additional stimulation of thyroid hor- mone by TRH. Thyroid hormones (triio- dothyronine tT3] and thyroxine itch) have been shown to influence body weight gain in chickens (Leung et al., 1985~.

138 Somatome~in-c The growth activity of GH is believed to be mediated by SM-C growth factor, gen- eratec] mainly in the liver. Somatomedin-C is GH-`lepenclent, and purified SM-C has been shown to stimulate body weight gain in both hypophysectomized and intact rats (Hizuka et al., 1986; SchoenIe et al., 1982~. Since chicken SM-C has not been isolated ant] purified, a human SM-C raclioimmu- noassay (RIA) was used to measure serum immunoreactive SM-C when purified cGH was injected into 4-week-old cockerels (Leung et al., 1986b). Purified cGH clid not affect weight or incorporation of 3H-proline or 35SO4 in 9- to 10-day-old chicken embryo cartilage cultured in vitro, but purified hu- man SM-C hac3 a significant effect (Burch et al., 1985~. Thus' it seems that the growth promotion axis of hypothalamic GRF-pitui- tary GH-hepatic SM-C in chickens is similar to that in mammals, but investigation of the biological effects of purifier! chicken GRF and chicken SM-C is neecled to validate this hypothesis. Growth Hormone Receptor Hormone-receptor interaction is the first step in hormone action, but receptor phys- iology has only recently been given atten- tion. Many human diseases are known to result from receptor clefects, but the bio- logical significance of the receptor is only beginning to be recognized. For example, analysis of the amino acid] and nucleoticle sequences of purified epidermal growth fac- tor receptor (EGF-R) has enabled scientists to link the structure-function relationships of oncogenes (v-erbB) ant! EGF-R (Down- warc! et al., 1984~. Although there is no structural analysis (amino acid response) for the GH receptor as yet, its eventual deter- mination will lead to an understanding of the molecular basis of GH action. Leung et al. (1984a) demonstrated a spe- cific hepatic GH receptor in chickens and obs erve cl paradoxically high blood co ncen APPENDIX "rations of GH, as measurer] by a homolo- gous cGH RIA (Leung et al., 1984b), in sex- linkecl dwarf chickens (Lilburn et al., 1986~. These chickens grew to less than half the size of normal chickens, leacling Leung et al. (1984a) to examine GH receptor binding in the same strain. There was a significant decrease in hepatic receptor binding at 6, 8, and 20 weeks of age compared to that of normal, fast-growing broiler chickens (Leung et al., 1987~. Huybrechts et al. (1985) re- portec] that sex-linkec3 dwarf chickens also hac! significantly lower circulating immu- noreactive SM-C concentrations compared  to those of normal birds. And Leung et al. (1984a) observed that sex-linked dwarf chickens had significantly higher hepatic (IGF-I) receptor binding. These observations may provide evidence that dwarfism is sex-linked and may be clue to a defect in the GH receptor. Based on preliminary results, we believe that GH receptors may be the limiting factor in the growth promoter axis in chickens. For ex- ample, normal Leghorn chickens, which grow at a much slower rate than broiler chickens, possess significantly fewer GH receptors than broiler chickens (Leung et al., 1987~. However, that hypothesis floes not agree with data reported for mammalian species. Growth hormone has been shown to maintain its own receptors in rat a(lipo- cytes ant] to up-regulate its hepatic recep- tors (Baxter and Zaltsman, 1984~. Recently, Chung ant] Etherton (1986) reported that the number of hepatic GH receptors is increased in pigs that have received GH injections. The method of regulating GH receptors in other agricultural animals is not known. However, if GH up-regulates its receptors at the target tissue, it is logical to assume that an increase in circulating GH would! result in an amplified biological response to GH. Gene Insertion The technology for introducing foreign genes into mammalian embryos forms the

HORMONAL REGULATION OF GROWTH basis of a powerful approach for studying gene regulation and the genetic basis of development (Palmiter and Brinster, 1985~. A dramatic growth increase in transgenic mice from eggs that were microinjectec] with a metallothionein GH foreign gene suggests that this technology could be val- uable for agricultural applications. Indeed, Hammer et al. (1985) successfully intro- duced foreign genes into the genes of rab- bits, sheep, and pigs by microinjecting eggs, using mouse metallothionein-human GH recombinant DNA. The foreign DNA was integrated ant] expressed in transgenic rab- bits and pigs. Thomas E. Wagner (Ohio University, personal communication, 1986) also successfully introduced foreign genes in pigs by microinjection. Leung and co- workers have attempted to directly inject foreign DNA into the blastoderm of freshly laid eggs with recombinant DNA technology (unpublished! data). And Souza et al. (1984) user] the retroviral approach in introducing foreign genes into chickens. Kopchick et al. (1985) constructed a re- combinant DNA (pbGH-4~-that is an avian retroviral long-terminal repeat (LTR), li- gatec! to the structural bovine GH (bGH) gene. This recombinant DNA is biologically active in a transient eukaryotic expression assay system. When this recombinant DNA was totally integrated into a mouse fibroblast cell line, mature bGH was expressed and secreted into the culture medium. Leung et al. (1986a) purified and characterized the recombinant bGH from culture medium and shower] that the recombinant bGH pos- sesses the same physiochemical and physical properties as native pituitary bGH. This recombinant bGH DNA was then intro- clucec3 into the germinal disk of the freshly lair! egg by opening a window in the egg and injecting various amounts of DNA in circular or linear form with a micropipette. Only seven of the chicks that hatched from the 3, 000 injected eggs had measurable circulating immunoreactive bGH. When serum samples were measured with both a homologous cGH RIA and a bGH RIA, the 139 cross-reactivity of purified cGH and bGH in the RIA was less than 5 percent. The expression of bGH was transient; no de- tectable immunoreactive bGH was present after 10 weeks of age. All the chickens were killed or crossed after sexual maturity. Tis- sue DNA was analyzed by dot blot and Southern gel assays. No measurable im- munoreactive bGH was detected by RIA from seven samples collected from first- generation offspring. It appears, therefore, that this method is inefficient. In addition, since the germinal disk in freshly laid eggs  consists of at least 500 to 1,000 cells, even if the foreign DNA is integrated in the host cell genome it is unlikely that the foreign DNA will enter the germ line. Use of a retroviral vector to introduce foreign genes into chicken genes provides an alternative experimental approach. In- deed, Souza et al. (1984) generated a re- combinant retrovirus by cloning chicken GH cDNA into a modified Rous sarcoma virus Schmiedt-Ruspin A genome in which the sac gene was entirely deleted. Recom- binant infectious virus that expresses cGH was generated to infect 9-day-old chick embryos. Subsequently born chicks ex- pressed circulating concentrations of cGH that were two- to threefold higher than those of normal birds. In addition, the birds were uremic. Salter et al. (1986) obtained similar results using a different retroviral vector. These results suggest that the retro- viral approach may be more elective than direct injection of foreign DNA in intro- clucing foreign genes into the germ line of chickens. CONCLUSIONS AND FUTURE DIRECTIONS Our preliminary information that the GH receptor, rather than GH itself, may be the limiting factor in the growth production axis in chickens opens up new research direc- tions. Pituitary GH has been purified from many agricultural animals, and antibodies to these preparations have also been gen

140 crated for RIA. Somatomedin-C has been purified only from humans and rodents (Spencer et al., 1983~; with recombinant DNA technology, scientists should be able to clone the SM-C gene and express syn- thetic recombinant SM-C using prokaryotic and eukaryotic cell expression systems. Only then can the biological activities of SM-C in agricultural animals be determined. The techniques for inserting foreign DNA into genes by microinjection into the pronucleus of fertilized eggs have been successful in agricultural animals (Hammer et al., 1986), and the retroviral vector approach in chick- ens is also promising. However, further research is needed to determine which genes are most desirable for use in gene insertion, define the sites of integration, and attain the fine control for expressing the exogenous genes that is necessary to make such technology useful to agriculture. ACKNOWLEDGMENTS I am grateful for the collaboration of Drs. John Kopchick, km Smith, H. Chen, and Mike Lilburn and for the expert assistance of I. Taylor, A. Van Iderstine, C. A. Ball, K. N. Ngiam-Rilling, B. Goggins, C. I. Rosenblum, R. Malavarca, E. Mills, and F. Macks. I also thank M. E. Mer~cka and H. B. Crow for typing this manuscript and D. L. FeIton for her expert editing. REFERENCES Baxter, R. C., and Z. Zaltsman. 1984. Induction of hepatic receptors for growth hormone (GH) and prolactin by GH infusion is sex dependent. Endo- crinology 115:2009. Brumby, P. J. 1959. The influence of growth hormone on growth in young cattle. N.Z. J. Agric. Res. 2:683. Burch, W. M., G. Corda, J. J. Kopchick, and F. C. Leung. 1985. Homologous and heterologous growth hormones fail to stimulate avian cartilage growth in vitro. J. Clin. Endocrinol. Metab. 60:747. Chawla, R. K., J. S. Parks, and D. Rudman. 1983. Structural variants of human growth hormone: Bio chemical, genetic and clinical respects. Annul Rev. Med. 34:519. Chung, C. S., and T. D. Etherton. 1986. Characteri APPENDIX zation of porcine growth hormone (pGH) binding to porcine liver microsomes: Chronic administration of pGH induces pGH binding. Endocrinology 119:780. Chung, C. S., T. D. Etherton, and J. P. Wiggins. 1985. Stimulation of swine growth by porcine growth hormone. J. Anim. Sci. 60:118. Downward, J., Y. Yarden, E. Mayes, G. Scarce, N. Totty, P. Stockwell, A. Ullrich, J. Schlessinger, and M. D. Waterfield. 1984. Close similarity of epider- mal growth factor receptor and v-erb-B oncogene protein sequences. Nature 307:521. Hammer, R. E., R. D. Palmiter, and R. L. Brinster.  1984. Partial correction of murine hereditary growth disorder by germ-like incorporation of a new gene. Nature 311:65. Hammer, R. E., V. G. Pursel, C. E. Rexroad, R. J. Wall, D. J. Bolt, K. M. Ebert, R. D. Palmiter, and R. L. Brinster. 1985. Production of transgenic rab- bits, sheep and pigs by microinjection. Nature 315:680. Hammer, R. E., V. G. Pursel, C. E. Rexroad, R. J. Wall, D. J. Bolt, R. D. Palmiter, and R. L. Brinster. 1986. Genetic engineering of mammalian embryos. J. Anim. Sci. 63:269. Harvey, S., C. G. Scanes, N. J. Bolton, and A. Chadwick. 1978. Effect of thyrotropin-releasing hor- mone (TRH) and somatostatin (GH-RIH) on growth hormone and prolactin secretion in vitro and in viva in the domestic fowl (Gallus domesticus). Neuroen- docrinology 26:249. Hizuka, N., K. Takano, K. Asakawa, M. Miyakawa, I. Tanaka, R. Harikawa, and K. Shizume. 1986. Insulin- like growth factor I stimulates growth in normal growing rats in viva. In Proceedings of the 68th Annual Endocrine Society Meeting, June 25-27, 1986, Anaheim, Calif. Bethesda, Md.: Endocrine Society. Huybrechts, L. M., D. B. King, T. J. Lauterio, J. Marsh, and C. G. Scanes. 1985. Plasma concentra- tions of somatomedin-C in hypophysectomized, dwarf and intact growing domestic fowl as determined bY heterologous radioimmunoassay. J. Endocrinol. 104:233. Isaksson, O. G. P., J.-O. Jansson, and I. A. M. Cause. 1982. Growth hormone stimulates longitudinal bone growth directly. Science 216:1237. Kawauchi, H., S. M. Ama, A. Yasuda, K. Yamaguchi, K. Shirahata, J. Kubota, and T. Hirano. 1986. Isolation and characterization of chum salmon growth hormone. Arch. Biochem. Biophys. 244:542. Kopchick, J. J., R. Malavarca, T. Livelli, and F. C. Leung. 1985. Use of avian retroviral-bovine growth hormone DNA recombinants to direct expression of bovine growth hormone by cultured fibroblasts. DNA 4:23. Leung, F. C., and J. E. Taylor. 1983. In viva and in vitro stimulation of growth hormone release in chick- ens by synthetic human pancreatic growth hormone releasing factor (hpGRF). Endocrinology 113:1913. , ,

HORMONAL REGULATION OF GROWTH Leung, F. C., J. Gillett, M. S. Lilburn, and J. Kopchick. 1984a. Analysis of growth hormone re- ceptors and genes in sex-linked dwarf chickens. J. Steroid Biochem. 20:1557. Leung, F. C., J. E. Taylor, S. L. Steelman, C. D. Bennett, J. A. Rodkey, R. A. Long, R. Serio, R. M. Weppelman, and G. Olson. 1984b. Purification and properties of chicken growth hormone and the de- velopmentofahomologousradioimmunoassay. Comp. Endocrinol. 56:389. Leung, F. C., J. E. Taylor, and A. Van Iderstine. 1984c. Thyrotropin-releasing hormone stimulates body weight gain and increases thyroid hormones and growth hormone in plasma of cockerels. Endo- crinology 115:736. Leung, F. C., J. E. Taylor, and A. Van Iderstine. 1985. Effects of dietary thyroid hormones on growth, plasma T3, T4 and growth hormone in normal and hypothyroid chickens. Gen. Comp. Endocrinol. 59:91. Leung, F. C., B. Jones, S. L. Steelman, C. I. Rosen- blum, and J. J. Kopchick. 1986a. Purification and physiochemical properties of a recombinant bovine growth hormone produced by cultured murine fi- broblasts. Endocrinology 119:1489. Leung, F. C., J. E. Taylor, S. Wien, and A. Van Iderstine. 1986b. Purified chicken growth hormone (cGH) and a human pancreatic growth hormone releasing factor (hpGRF) increased body weight gain in chickens. Endocrinology 118:1961. Leung, F. C., W. J. Styles, C. R. Rosenblum, M. S. Lilburn, and J. A. Marsh. 1987. Diminished hepatic growth hormone receptor bindings in sex-linked dwarf broiler and Leghorn chickens. Proc. Soc. Exp. Biol. Med. 184:234. Lilburn, M. S., K. N. Ngiam-Rilling, J. H. Smith, and F. C. Leung. 1986. The relationship between age and circulating concentrations of triiodothyronine (T3), thyroxine (T4), and growth hormone in com- mercial meat strain chickens. Proc. Soc. Exp. Biol. Med. 182:336. Machlin, L. J. 1972. Effect of porcine growth hormone on growth and carcass composition of the pig. J. Anim. Sci. 35:794. Palmiter, R. D., and R. L. Brinster. 1985. Transgenic mice. Cell 41:343. 141 Palmiter, R. D., G. Norstedt, R. E. Gelines, R. E. Hammer, and R L. Brinster. 1983. Metallothionein- human GH fusion genes stimulated growth of mice. Science 222:809. Russell, S. M., and E. M. Spencer. 1985. Local injections of human or rat growth hormone or of purified human somatomedin-C stimulate unilateral tibial epiphyseal growth in hypophysectomized rats. Endocrinology 116:2563. Salter, D. W., E. J. Smith, S. H. Hughes, S. E. Wright, A. M. Fadly, R. L. Witter, and L. B. Crittenden. 1986. Gene insertion into the chicken germ line by retroviruses. Poultry Sci. 65:1445. Scanes, C. G., R. V. Carsia, T. J. Lauterio, L. Huybrechts, J. Rivier, and W. Vale. 1984. Syn- thetic human pancreatic growth hormone releasing factor (GRF) stimulates growth hormone secretion  in the domestic fowl (Gallus domesticus). Life Sci. 34:1127. Schoenle, E., J. Zapf, R. E. Humbel, and E. R. Froesch. 1982. Insulin-like growth factor I stim- ulates growth in hypophysectomized rats. Nature 296:252. Sonza, L. M., T. C. Boone, D. Murdock, K. Langley, J. Wypych, D. Fenton, S. Johnson, P. H. Lai, R. Everette, R. Y. Hsu, and R. Bosselman. 1984. Application of recombinant DNA technologies to studies on chicken growth hormone. Exp. Zool. 232:465. Spencer, E. M., M. Ross, and B. Smith. 1983. The identity of human insulin-like growth factors I and II with somatomedins C and A and homology with rat IGF I and II. Proceedings of a Symposium on Insulin-Like Growth Factors/Somatomedins, Nai- robi, Kenya, November 13-15, 1982. Berlin: Walter de Gruyter. Underwood, L. E., and J. J. Van Wyk. 1981. Hormones in normal and aberrant growth. P. 1149 in Textbook of Endocrinology, R. H. Williams, ed. Philadelphia: W. B. Saunders. Wagner, J. F., and E. L. Veenhuizen. 1978. Growth performance, carcass deposition and plasma hor- mone levels in wether lambs when treated with growth hormone and thyroprotein. J. Anim. Sci. 45:397.

Muscle Cell Growth and Development RONALD E. ALLEN Skeletal muscle from domestic animals is a major source of high-quality protein in the human diet. Past technological advances in production of animal muscle protein have been baser] on empirical and fundamental biological research. Future technological advances, however, are less likely to occur unless research is firmly grounder! in the basic biology of muscle and animal growth. The primary function of this paper is to review information about the structure and composition of muscle, muscle clifferentia- tion and development, and key elements of protein metabolism as they relate to muscle growth. It also describes current areas of active research interest and speculates on applications of new research knowledge and future research needs. MUSCLE CELL STRUCTURE AND COMPOSITION The differentiated muscle cell in postnatal muscle is the muscle fiber, a highly spe- cializecI, long, cylindrical cell that can range in diameter from 10 to 100 Em en c! in length from millimeters up to many centimeters. The primary differences in fibers of different 142 species are fiber length and number of fibers per muscle. Each fiber is surrounded by a 7.5- to 10-nm-thick plasmalemma, caller! the sarcolemma. The sarcolemma is a lipid bilayer like the cell membranes of other cells and has a lipid composition of roughly 60 percent protein, 20 percent phospho- lipid, and 20 percent cholesterol. Surround- ing the sarcolemma is the basal lamina, or basement membrane. This somewhat amor- phous structure, 50 to 70 nm thick, is composed of mucopolysaccharicles an(1 col- lagen (types III ant] V). The cell membrane of muscle has a specialized structure the motor endplatc which accommodates in- teraction with an axon from a motoneuron. In addition, the membrane maintains an electrical potential that is propagated from the motor en(lplate, clown the membrane, ant] finally into the cell by a complex set of invaginations that form the transverse tu- bular system. Muscle fibers contain the major orga- nelles present in most cells. The most strik- ing difference between muscle cells an(1 the majority of other cells is their multinu-  cleated nature. Depending on its size, an in(lividual fiber may contain hundreds of

MUSCLE CELL GROWTH AND DEVELOPMENT nuclei. They are found just beneath the sarcolemma and seem to be randomly dis- tributed along the length of the fiber. Mi- tochondria are present between the con- tractile elements of muscle; their concentration varies with the metabolic ac- tivity of the particular fiber. Ribosomes are dispersed within the cytoplasm, but very few are associated with endoplasmic retic- ulum, primarily because muscle fibers syn- thesize few secreted proteins. The enclo- plasmic reticulum in muscle has formed a specialized set of membrane structures called the sarcoplasmic reticulum. The primary function of this structure is regulation of free calcium ion concentration. When free calcium ion concentration is maintainer] be- low approximately 0.1 ~M, contraction does not occur. But when the membrane is depolarized, the action potential reaches the interior of the cell through the trans- verse tubular system, calcium is released from the sarcoplasmic reticulum, the con- centration approaches 1 ,uM, and contrac- tion is activated. Lysosomes are not readily seen in muscle fibers, although lysosomal enzymes are present. The lysosomes are most likely sequestered in the sarcoplasmic reticulum. By far the most unique subcellular aspect of muscle fibers is the contractile machinery, the myofibril. This is an aggregation of 12 to 14 proteins into highly organized con- tractile threads that are insoluble at the ionic strength of the cytoplasm in muscle cells. It is noteworthy that this specialized set of proteins constitutes about 55 percent of the total protein in muscle. Conse- quently, many developmental studies of muscle have focuses] on myofibrillar protein gene expression and synthesis, which are discussed later in this paper. Myofibrils are composed of two main classes of filaments: thick filaments and thin filaments. Thick filaments measure approx- imately 15 rim by 1,500 nm. The major protein in thick filaments is myosin, which has the active site that hydrolyzes adenosine 143 triphosphate (ATP) and the site that binds to actin in the thin filament. The thin filament is roughly 6 nm by 1,000 nm and is composed of actin, which forms the beaded backbone of the filament, ant! tropomyosin and troponin, which perform regulatory functions. At one end, thin filaments insert into a protein lattice called the Z-line; at the other end, they overlay with thick filaments in a hexagonal array. Aciclitional small-diameter filament systems are present within myofibrils to provide an elastic com- ponent. Also, an intermecliate-diameter fil- ament system, found outside the periphery  of the myofibril, links adjacent myofibrils and maintains their contractile units in reg- ister. Specific details of the ultrastructure of myofibrils and the biochemical properties of this intercligitating array of filaments can be found in Goll et al. (1984~. These features of muscle cells are com- mon to all skeletal muscle fibers, but specific fibers have cli~erentiated somewhat de- pending on their purpose. Some populations of fibers are primarily responsible for rapid contractions on an intermittent basis, while others have slower contraction speed ant! sustain contractile activity over extended periods of time. Muscle fiber types have been described extensively in many species; and their biochemical, physiological, and morphological cli~erences are significant to problems of muscle growth and meat qual- ity. A generalized scheme for describing fiber types classifies them on the basis of their contraction speed and on the energy metabolism pathways primarily used to pro- vide energy for contraction. Peter et al. (1972) provided one of the most descriptive classification systems by grouping fibers into three general categories. Fibers that were dependent on oxiclative metabolism and had slower contraction speecis were classified as slow-twitch, oxidative fibers (SO). Fibers with faster contraction times that were de- pendent on anaerobic, or glycolytic, energy metabolism pathways were termed fast- twitch, glycolytic fibers (FG). A third broac!

144 category contained fast-twitch fibers that had glycolytic metabolic capabilities but also a significant capacity for oxidative metabo- lism; these were termed] fast-twitch, oxida- tive-glycolytic fibers (FO G). Contraction speed is correlated with myosin aclenosine triphosphatase (ATPase) activity and, therefore, with the particular myosin isozymes synthesizer! by the fiber. Other myofibriliar protein isoform varia- tions may also be associated with contractile properties. The complexity and degree of development of the sarcoplasmic reticulum, t-tubule system, and neuromuscular junc- tions have all been associated with contrac- tion speed and fiber class. As expected, mitochondrial content and glycolytic en- zyme content vary among fiber types, as (lo energy substrates such as glycogen and triglyceride. Aspects of fiber type variation that affect muscle growth include the not- able differences in fiber size that generally correlate with muscle fiber type. SO fibers are smaller in diameter than FG fibers, and FO G fibers tent] to be intermediate in size. Smaller fiber diameters may facilitate effi- cient gas exchange in oxidative fibers. In addition, SO fibers tend to have higher nuclei concentrations and, therefore, lower protein concentrations per nucleus. Satellite cell frequency, however, is reportedly higher for SO fibers (Kelly, 1978b). Because indi- vidual muscles vary in fiber type composi- tion, factors that clifferentially affect the development or growth of specific fiber types can result in alterations in muscle mass (for example, the transition from FG to FO G that can accompany aerobic con- ditioning). Reductions in fiber diameter and, consequently, muscle mass would be ex- pected. Alterations in gene expression and in quantitative aspects of protein metabo- lism that are responsible for such fiber type transitions are poorly understood. Chemical composition of muscle tissue can be quite variable, and the primary source of variation is intramuscular adipose tissue. It is clear that most of the variation APPENDIX in major constituents is minimized when expressed on a fat-free basis. Some com- positional variation can be found in associ- ation with aging, but, in general, it is attributable to changes in moisture content. Skeletal muscle from very young animals has a high moisture content that decreases with maturity. As a result, protein concen- tration increases with maturity. Subtle changes in other constituents, such as gly- cogen, can vary among muscles and species, but these differences may not have major nutritional significance when considering the composition of muscle as a food.  The primary lipid fraction contributing to muscle tissue variation is triglyceride, which is stored in adipocytes within the muscle. These depositions are commonly referred to as marbling, and within the range of marbling found in the longissimus muscle of beef, the ether-extractable lipid (primar- ily triglycericle) varies from 1.77 to 10.42 percent on a wet weight basis (Savell et al., 1986~. Cholesterol content, on the other hand, is less variable. This can best be understood in light of its role in muscle tissue. Choles- tero} is an integral part of cell membranes, mainly the plasma membrane. On a tissue basis across maturity groups and marbling contents within maturity groups, cholesterol content of beef muscle floes not vary (Stromer et al., 1966~. In addition, the amount of cholesterol per gram of whole steak was not significantly different among the five yield gracles examined by Rhee et al. (1982~. Furthermore, neither breed type nor nu- tritional background affected cholesterol content of lean muscle tissue in beef cows (Eichhorn et al., 1986~. It is possible to find variation in cholesterol content of meat, however, because adipose tissue tends to have a higher cholesterol concentration than do muscle fibers. Consequently, variations in the amount of subcutaneous or inter- muscular fat consumed with the lean portion can alter cholesterol intake. It has been calculated that 37 to 56 percent of the

MUSCLE CELL GROWTH AND DEVELOPMENT cholesterol in a cooker] rib steak of beef originates from subcutaneous and inter- muscular adipose tissue (Rhee et al., 1982~. In looking only at muscle cells, however, significant variations in cholesterol content have not been seen, even among most of the species used for muscle foods (Reiser, 1975; Watt and Merrill, 1963~. This is also true for the amino acid composition of muscle. The majority of muscle cell proteins are myofibrillar ant! are very highly con- served across species. In addressing topics such as alteration of tissue composition to enhance nutritional quality, it is important to keep in mind that the biology of the animal or tissue must come first. Our ability to manipulate cells in animals has both physiological limits en c] ramifications. MUSCLE FIBER DEVELOPMENT Prenatal Development Myogenesis originates in cells of the em- bryonic mesoderm and apparently follows a similar course in all species examined. Per- haps the most detailed] descriptions come from studies of human (Hauschka, 1974) and chick (White et al., 1975) embryo clevel- opment. In the human, no apparent orga- nization is noted in the limb mesoderm on day 28 of development, but by day 43 loose connective tissue cell regions and compact myogenic cell regions are visible. By clay 45 the first small multinucleated myotubes (the precursors of muscle fibers) have formed; by day 50 the general organization of major muscles and bones is essentially complete. Beyond this point, the rate of muscle his- togenesis occurs at different rates between ant] within individual muscles. In the gas- trocnemius on day 62, well-developed, my- ofibril-containing muscle fibers are present, but the majority of cells are still mononu- cleated. This population decreases to about 50 percent of the total by clay 72, while fibers increase two- to threefolcl. During the next 2 weeks, fiber formation proceeds 145 rapidly, with the percentage of mononu- cleated cells diminishing to 20 percent by clay 95 and further decreasing to the point that only a few single cells persist in asso- ciation with fibers by day 146. In other vertebrate species, comparable developmental patterns are discernible. One striking observation in rat and chick muscle is the development of two populations of fibers (Kelly and Zacks, 1969; McLennan, 1983~. The "primary fibers" develop early and are surrounded by closely associated mononucleatec] cells. In the chick embryo, "seconclary fiber" formation proceeds rap- idly after about 12 days of development  until most of the mononucleated cell pop- ulation is exhausted and fiber formation is complete. This occurs before hatching in the chick and before birth in most mammals. A similar biphasic developmental pattern has been documentecl in fetal lamb skeletal muscle (Ashmore et al., 1972~. In general, fiber formation is complete near the time of birth. The stucly of myogenesis focuses on the muscle development process and has cen- tered around efforts to unravel myogenic lineages and the mechanisms responsible for alterations in the synthetic programs of muscle cells that lead to the formation of fibers and the expression of muscle-specific cell characteristics. One of the most impor- tant initial observations on the mechanisms of myogenesis came from a series of exper- iments reported by Stockdale and Holtzer (1961) that directly (demonstrate that mul- tinucleated myotubes arise from the fusion of mononucleated myogenic cells (myo- blasts). Furthermore, only mononucleatec] cells have the ability to proliferate; the nuclei in myotubes cannot replicate their DNA and divide. Consequently, the tran- sition from a proliferating myoblast to a nonproliferating myotube that can synthe- size muscle-specific macromolecules rep- resents the terminal step in muscle differ- entiation. There now appear to be several different

146 types of myogenic cells that are actively proliferating and differentiating during spe- cific periods of development. Their collec- tive developmental patterns are responsible for the general pattern of muscle histoge- nesis. At least two broad types and four subtypes of myogenic cells have been iden- tified by White et al. (1975), based on the in vitro morphology and medium require- ments of cloned myogenic cells from various stages of embryo development. One general type is the early muscle-colony-forming cell, which predominates in early development; the colonies are noted for having small, thick myotubes with few nuclei. In contrast, the predominant form of myogenic cells in later periods of development form colonies in vitro that are extensively fused and con- tain large myotubes with many nuclei; these are the late muscle-colony-forming cells. Miller and Stockdale (1986) have identified four types of myogenic cells based on the presence of specific isoforms of the myosin heavy chains present in early and late mus- cle-colony-forming cells. Early and late classes of cells appear to be distinct, since they can maintain their class-specific characteristic when subcloned up to five times, until proliferative senes- cence (Rutz and Hauschka, 19821. Addi- tional experiments reported by Seed and Hauschka (1984) have shown that transplant- ing limb buds at various stages results in the absence of late myogenic cells in the transplant, even though the early class of muscle-colony-forming cells was present. The late class apparently migrates into the limb bud from the somite at a later stage than the early class and, furthermore, does not appear to descend from the early class, in agreement with the previous in vitro experiments (Rutz and Hauschka, 1982~. The appearance of early and late muscle- colony-forming cells appears to correlate well with the anatomical appearance of pri- mary and secondary fibers that are formed during development. Different myogenic classes of cells are further implicated in the APPENDIX formation of primary and secondary fibers because the in viva formation of secondary fibers is nerve-dependent (McLennan, 1983), as is the in vitro development of fibers from one of the later muscle-colony-forming types (Bonner and Adams, 1982~. An additional class of myogenic cells, or branch of the myogenic lineage, is the satellite cell, which is discussed further in the subsection on postnatal development. As mentioned previously, a striking tran- sition takes place in muscle development with the differentiation of mononucleated myoblasts into multinucleated myofibers.  This terminal step in differentiation is ac- companied by the cessation of proliferation and the expression of genes responsible for the muscle phenotype. For many years, there were two general theories to explain myogenesis. The first postulated that a ma- jor reorganization in gene expression took place in specific mitotic cycles, and the resultant daughter cells had protein synthe- sis capabilities that differed from those of the mother cell. This special cell cycle was referred to as a "quanta!" cell cycle (Holtzer and Bischoff, 1970~. This theory has now been expanded to hypothesize that a fixed number of cell divisions occur between the stem cell compartment to the terminally differentiated, fusion-competent myoblast compartment (Quinn et al., 1984~. Key to this description of myogenesis is the "com- mitment" step of myoblasts to withdraw from the cell cycle, fuse, and initiate the synthesis of muscle-specific proteins. In contrast, a second theory of myogenesis (Buckley and Konigsberg, 1974) was based on a model that predicted that myoblasts remaining in the Go phase of the cell cycle had an increasing probability of fusion that resulted in permanent withdrawal from the cell cycle and the initiation of muscle protein synthesis. The probability of remaining in the cell cycle or fusing depended on the presence or absence of environmental fac- tors that stimulate these activities. In this model, withdrawal from the cycle and ini

MUSCLE CELL GROWTH AND DEVELOPMENT tiation of muscle gene expression was thought to be the result of the fusion process itself. A current, and more likely, explanation encompasses elements of both the original theories. It appears that during the early part of the Go phase of the cell cycle, proliferating myoblasts have the option of continuing to proliferate or of differentiating and fusing into myotubes (Nadal-Ginard, 1978~. The commitment to withdraw from the cell cycle is made before fusion, not as a result of fusion. This commitment, how- ever, clepends on the presence of growth- stimulating factors in the environment (probably mitogens) that keep myoblasts in the cell cycle. For many years, it appeared that withdrawal from the cell cycle, fusion, and expression of the muscle phenotype were coupled events; recent experiments with a temperature-sensitive mutant of the muscle cell line L6E9 have cast doubts on the obligatory relationship of these events. In experiments with wild-type and mutant L6E9 myoblasts, Nguyen et al. (1983) clem- onstrated that muscle-specific isoforms of certain myofibrillar proteins could be in- duced in the mutant cells under conditions that did not permit commitment to with- drawal from the cell cycle. In fact, these cells could be stimulated to reenter the cell cycle even after induction of myofibrilIar protein synthesis. Aciclitional experimenta- tion with wild-type L6E9 myoblasts arrester] in a low-calcium medium indicated that induction of myofibrillar protein synthesis occurred in cells that could] subsequently be stimulated to synthesize DNA and di- vide. Reentry into the cell cycle, however, resulted in a rapic! cessation of myofibrillar protein synthesis ant! degradation of existing muscle-specific messenger RNAs (Nadal- Ginard et al., 1984~. Similar experiments were reporter! with primary cultures of quail embryo muscle that were arrested in a low- calcium medium (Devlin and Konigsberg, 1983~. Apparently, induction of the gene expression transitions leading to the muscle phenotype can be uncoupled from perma 147 nent withdrawal from the cell cycle. In normal muscle development, however, the commitment to withdraw from the cell cycle and the induction of the muscle phenotype are closely correlated and occur simulta- neously. The in vivo signals that affect the com- mitment decision maple by myoblasts during fetal development and myofiber formation have not been iclentified. One class of pro- tein growth factors, the fibroblast growth factor (FGF), has been shown to be mito- genic for myoblasts in culture and can re- duce the tendency to differentiate (Allen et  al., 1984; Gospodarowicz et al., 1976; Link- hart et al., 1981~. A second growth factor, transforming growth factor beta (TGF-~), is a very notent inhibitor of myoblast differ- entiation and coulc! be responsible for reg- ulating myogenic cell activities in viva (Flor- ini et al., 1986~. In contrast to the two inhibitors of differentiation, the insulin-like growth factors have been reported to stim- ulate myoblast proliferation ant! differentia- tion in culture (Ewton ant] Florini, 1980, 1981~. The means by which these two an- tagonistic processes can be stimulated by the same hormone, however, has not been completely clarified. In general, the activ- ities of the growth factors and hormones in embryonic muscle development have yet to be verified in vivo. Although the specific regulatory agents involved in stimulating differentiation have not been thoroughly documented, many of the gene transitions that occur in association with the terminal step in muscle cell differ- entiation have been reported (Young and Allen, 1979~. From the standpoint of gene regulation, some of the interesting events center around the contractile proteins. The major myofibrillar proteins are synthesized in a coordinate fashion shortly after fusion (Devlin and Emerson, 1978, 1979~. These events seemed relatively straightforward, until it became possible to examine them in greater molecular detail. It now appears that there are a series of subtle transitions

148 in expression of specific skeletal muscle isoforms of individual proteins cluring the in vivo ant! in vitro development of muscle (reviewed by Caplan et al., 1983~. The actin that is first synthesized after myoblast clif- ferentiation is of the alpha isoform, but it is alpha-cardiac actin en c] not alpha-skeletal actin. The transition from alpha-cardiac to alpha-skeletal actin occurs as the myotube matures (Bains et al., 1984; Paterson and Eldricige, 1984~. Similarly, myosin light chains and heavy chains (Bandman et al., 1982; Crow et al., 1983; Gauthier et al., 1982; Lowey et al., 1983; Lyons et al., 1983; Whalen et al., 1978) progress through iso- form transitions that inclucle fetal, neonatal, and, finally, adult isoforms of the subunits of these proteins. These transitions occur in viva and are also recapitulated in regen- erating muscle (Marechal et al.. 19841. Re~- ulators of this developmental scheme have not been elucidatecI; however, innervation and load-bearing functions may be involved in the feedback that is responsible for al- terations in gene expression (Hoffman et al., 1985; Rubinstein and Kelly, 1978~. The environmental factors that regulate the synthesis of specific isoforms and the rate at which these proteins are accumulated are not specifically known, but the mecha- nisms will be resolved in the near future because genes for these proteins are being studied in detail (reviewed by Bobbins et al., 1986; Young et al., 1986~. For example, the regulation of alpha-skeletal actin may depend on the DNA sequence in regions of the gene preceding the 5 -untranslated part of the message-coding region (Bergsma et al., 1986; Hu et al., 1986; Mellou! et al., 1984~. It has been suggested that "trans- acting factors in the cytoplasm of myogenic cells interact with nuclear genes to activate their expression (Chin and Blau, 1984), but the nature of these factors has not been clescribed. In the case of myosin, thyroid hormone may be involved in myosin heavy- chain synthesis (Butier-Brown et al., 1986; Gambke and Rubinstein, 1984; Izumo et APPENDIX al., 1986~. The chemical mediators of the eject of activity level (Brevet et al., 1976; Hoffman et al., 1985) and neurogenic influ- ences (Rubinstein ant! Kelly, 1978) remain unclefinec3. Detailed information about the structure of important muscle-specific genes, including identification of regulatory se- quences, will open the door to studies that are critical to un(lerstancling quantitative aspects of muscle protein synthesis regula- tion, one of the key problems in animal growth research. Postnatal Development  Understanding the regulation of postnatal muscle growth requires an appreciation of the cellular events underlying the process. Postnatal muscle growth is frequently con- siderec! to be (lue to muscle fiber hypertro- phy, in contrast to prenatal muscle growth. This assumption stems from the docu- mented fact that muscle fiber number does not increase dramatically after birth in most animals; consequently, increases in size must be clue to hypertrophy (reviewer] by Gold- spink, 1972; Swatland, 1976~. Although postnatal muscle growth is often thought of in terms of fiber hypertrophy, and not hyperplasia, proliferation and (lif- ferentiation of myogenic cells are central to the process of postnatal muscle growth. For example, Winick and Noble (1966) (lem- onstrated an 8.5-fold increase in rat muscle DNA from 21 to 133 days of age, corre- sponding to an 88 percent increase in muscle DNA. Moreover, the relationship between DNA accretion and muscle growth was more firmly established by the findings of Moss (1968) ant! Swatland (1977), which demon- stratec! that muscle fiber diameter in grow- ing chicken and pig muscle, respectively, is directly relater] to the total number of muscle fiber nuclei. Additional studies sup- porting these results have been reviewed by Allen et al. (1979) and continue to appear regularly in the literature. Consistent with the point of view that

MUSCLE CELL GROWTH AND DEVELOPMENT myogenic cell proliferation is critical to the attainment of maximum muscle mass in livestock are studies involving strains of swine that differ in muscle growth potential (Harbison et al., 1976; Powell and Aberle, 1981) and growth studies in cattle (Trenkle et al., 1978~. Of the biochemical parameters evaluated in these experiments, DNA ac- cretion anc] protein/DNA ratios were most intimately related to muscle growth. In addition, the most rapic] period! of DNA accretion coincided with the most rapid period] of muscle growth. The cumulative evidence presenter] by these and other studies suggests that most muscle fiber DNA fount] in mature muscle is accumulated postnatally, ant! the accretion of DNA in muscle is a key factor in regulating muscle growth. The idea that muscle fiber number is constant beyonc! the neonatal period! hac! been accepted for years, as had the notion that nuclei within muscle fibers do not replicate their DNA or divide. However, these observations were clearly inconsistent with the large increases in DNA occurring in postnatal muscle. This is explained by the role of satellite cells, the small mon- onucleated cells that reside between the sarcolemma ant] basement membrane of muscle fibers (Mauro, 1961~. These cells have the ability to proliferate, differentiate, and fuse into adjacent fibers (Moss and Leblond, 1971), which results in the addi- tion ofthe satellite cell nucleus to the muscle fiber. Satellite cells are only discernible at the electron microscope level because they look like normal myonuclei that are located ad- jacent to the sarcolemma inside the fiber. Satellite cells are evenly clistributed across the surface of muscle fibers, except for an increased density arounc] the neuromuscu- lar junction (Gibson and Schultz, 1983; Kelly, 1978a). In normal adult muscle from many species, the cells generally make up only a small fraction of the total nuclei associated with fibers, usually ranging from 149 2 percent to less than 10 percent (Allbrook et al., 1971; Cardasis and Cooper, 1975; Schultz, 1974; Snow, 1977) and varying from one type of fiber and muscle to another; slow-twitch fibers often have a higher per- centage of satellite cells than do fast-twitch fibers (Gibson and Schultz, 1983; Kelly, 1978b). Also, there seems to be a greater percentage present in muscles of very young animals ant] a smaller percentage in muscles of old animals; this is particularly evident in fast-twitch muscle fibers (Gibson and Schultz, 1983~. The myogenic potential of satellite cells  and their ability to synthesize DNA, divide, and fuse into existing fibers was established by Moss ant] Leblond (1971~. Their my- ogenic properties were further documented by isolating mononucleated cells from minced muscle digests (Bischoff, 1974) or by isolat- ing individual muscle fibers (Bischoff, 1975; Konigsberg et al., 1975) and monitoring the division of mononucleatec] cells in culture. Not only did these mononucleatec3 cells clivicle but they eventually fused to form multinucleated myotubes. Myotubes formed by satellite cells in vitro synthesize muscle- specific proteins and spontaneously contract in culture (Allen et al., 1980; Cossu et al., 1980~. Although qualitatively they resemble em- bryonic myogenic cells, satellite cells may well be a separate type of myogenic cell. Cossu et al. (1980) first noted major differ- ences in the morphology of the two, and Allen et al. (1982) found that myotubes derived from satellite cells were only able to synthesize one-third to one-half as much alpha-actin as myotubes formed from neo- natal rat muscle. Cossu et al. (1983, 1985) also demonstrated that satellite cells and embryonic myogenic cells responded dif- ferently to a tumor promoter, 12-O-tetra- decanoylphorbol-13-acetate (TPA). TPA did not stimulate division or inhibit differentia- tion of satellite cells, as it did with myogenic cells of embryonic origin. Therefore, factors that stimulate the proliferation or differen

150 tiation of embryonic myogenic cells may or may not have the same effect on satellite cells. Even though the importance of satellite cells to muscle regeneration and normal growth has been appreciated for some time, details of their regulation are only now beginning to emerge. The stimulatory effect of five different growth factors and hormones and the inhibitory effect of one growth factor on satellite cell proliferation have been documented in vitro (Allen, 1986; Allen et al., 1984; Dodson et al., 1985~. Three of these proteins are insulin-like growth factors I ant! II (IGF-I and IGF-II) and insulin (Dodson et al., 1985~. These proteins are members of the same gene family and share high (legrees of sequence homology (Klap- per et al., 1983; Marquardt and Tociaro, 1981; Rinclerknecht and Humbel, 1978~. Insulin is active only at supraphysiological concentrations, which has been explained in terms of its action as an IGF-I analog. Both IGFs (commonly referred to as so- matomedins) stimulate satellite cell prolif- eration at concentrations well within the physiological range. The significance of the IGFs particularly IGF-I lies in their re- lationship to growth hormone. IGF-I me- diates the growth hormone signal at the target cell level. Consequently, in vitro data clirectly link the action of the IGFs to an authentic target cell in postnatal skeletal muscle. Two additional growth factors active in promoting satellite cell proliferation are the basic (Allen et al., 1984) and acidic (R. E. Allen, University of Arizona, unpublished] data) forms of fibroblast growth factor. Un- like the IGFs, however, the basic form of FGF only stimulates proliferation and ac- tually inhibits clifferentiation. Unfortu- nately, the physiological role of FGFs or similar proteins has not been established. FGFs have been isolated from a variety of cells ant] tissues; brain and pituitary tissue are the two most commonly used sources for purification (Gospodarowicz et al., 1976~. APPENDIX It is particularly noteworthy that similar protein fractions have been isolated from skeletal muscle (Kardami et al., 1985) and from peritoneal macrophages (Baird et al., 19854. The observations that this growth factor is not freely circulating but can be found in a variety of cells and tissues make it a reasonable candidate for an autocrine or paracrine hormone. This concept may have particular importance in regulation of skeletal muscle regeneration and work-in- ducec] hypertrophy, where a local signaling mechanism would seem to be necessary. Insights into the molecular mechanisms of  FGF action are sparse, although receptors have been identified (Olwin and Hauschka, 1986~. The possible role of FGF or FGF- like proteins as local signals for myogenic cell proliferation is an interesting concept that should be addressed. Satellite cell culture systems have also been used to evaluate the response of sat- ellite cells to growth hormone, prolactin, luteinizing hormone, thyroid stimulating hormone, epidermal growth factor, platelet- derived growth factor, and nerve growth factor. None ofthese proteins had the ability to stimulate satellite cell growth in vitro (Allen et al., 1986~. As mentioned previously, an inhibitor of satellite cell proliferation and differentiation has been identifie~l: transforming growth factor beta (TGF-~. In vitro, very low concentrations of TGF-,B ('0.5 ng/ml) can affect both processes (Allen, 1986~. This factor is interesting because it can be found in many cell types and has a variety of effects on their functions. It can be either stimu- latory or inhibitory, depending on cell type en c] the presence of other growth factors (Moses et al., 1985~. TGF-'B apparently is identical to the differentiation inhibitor de- scribecl by Evinger-Hodges et al. (1982) and Florini et al. (1986~. In summary, it appears that satellite cell activity can be controlled by several protein hormones/growth factors, and it may be the interplay of these factors that determines ;

MUSCLE CELL GROWTH AND DEVELOPMENT the state of the cell (quiescence, prolifera- tion, or differentiation). Nutritional and en- vironmental factors that influence muscle fiber DNA accretion in postnatal muscle may be mediated through one or more of these proteins. MUSCLE FIBER PROTEIN METABOLISM Muscle protein metabolism encompasses a broad range of cellular activities, many of which are integral parts of energy metabo- lism in the whole animal. Most notable among these biochemical processes are the deamination of amino acids and the utili- zation of the carbon skeletons for energy production; supplying amino acicis to the liver for gluconeogenesis is another impor- tant function. These aspects of muscle pro- tein metabolism are obviously critical to the physiology of the animal, but they are not necessarily directly related to muscle growth. Consequently, this discussion dwells on two broad growth-related processes in muscle: protein synthesis and protein degradation. The quantitative balance between these two activities determines the net accumulation of protein in muscle. A fundamental concept that has been widely appreciated only within the past decade or so is the fact that muscle protein is in a constant state of flux. Protein is constantly being degraded. It would not be out of the ordinary, for example, to expe- rience a 5 to 10 percent rate of degradation of protein per day. To maintain muscle mass, the muscle would have to synthesize an amount of protein equivalent to 5 to 10 percent of its protein content on a daily basis. The ramifications of this are enormous when one considers the energetic costs of synthesizing one peptide bond and the total number of peptide bonds that must be degrader] and resynthesized per clay. It is easy to understand why protein turnover represents a significant factor in the "main- tenance" energy requirements of an animal. 151 It is also easy to see how the efficiency of growth or production could be enhanced if protein turnover could be altered in a fa- vorable way. A number of studies have clemonstrated the balance between protein synthesis and degradation in domestic animals, laboratory animals, and humans and have revealed a general trend: In growing animals, synthesis and degradation rates are elevates] with synthesis rate exceeding degradation rate; as maturity is approached, both synthesis ant! degradation rates decrease and ulti- mately reach a low and equal rate. With  only minor variations, these trends have been observed in cattle, chickens, and lab- oratory animals (Lewis et al., 1984; MacDonald and Swick, 1981; McCarthy et al., 1983; Millwarc] en cl Waterlow, 1978; Millwarc! et al., 1976~. Certain metabolic hormones influence protein turnover; glucocorticoicls, for ex- ample, cause muscle atrophy by depressing synthesis and degradation (McGrath and Goldspink, 19824. Synthesis rate is appar- ently depressed to a greater extent than degradation rate. Insulin, on the other hand, causes net accretion of protein, primarily by affecting synthesis rate (Tischler, 1981), and generally antagonizes the glucocorticoid effect on synthesis and degradation (Tomes et al., 1984~. Thyroid hormone, T3, can increase degradation rate, but this modu- ration tends to follow the rate of synthesis (Millward, 1985), so there is a minimal change in protein accretion. Metabolites such as branchecl-chain amino acids or the keto acids of these amino acids may also be involves] in depressing degradation (Mitch ant! Clark, 1984; Tischler et al., 19824. The integrated response of muscle to the inter- play of metabolites and metabolic hormones is not completely understood but represents an important feature of muscle protein ac- cretion regulation. In addition to the homeostatic regulation of protein turnover, relative rates of syn- thesis and degradation are altered during

152 growth. Thus far, the only growth-related hormones that have been implicated in regulating protein degradation are the in- sulin-like growth factors, the somatomedins. Most of the work in this area has been conducted in vitro, where potent inhibitory effects have been observed (Ballard et al., 1986; laneczko and Etlinger, 1984~. The involvement is somewhat perplexing, since rapid growth rates in young animals are accompanied by increased rates of degra- dation, not decreased degradation. This point of contention, however, may be related to the in vitro assay system; the key element in the observation may be the decrease in degradation rate relative to synthesis rate. Several physiological conditions have been shown to affect the rates of synthesis and ciegradation in skeletal muscle. Included among these are physical influences such as muscle stretching, which leads to hypertro- phy (Golcispink, 1978; Summers et al., 1985~. In vitro muscle stretching decreases protein degradation (Baracos and Goldberg, 1985~. Inflammation, fever, and burns also have a dramatic effect by accelerating protein turn- over (Goldberg et al., 1984~; the common denominator in these observations and in the stretch-induced alteration in turnover may be calcium metabolism. In vitro, an influx of calcium into cells increases protein degradation (Silver and Etlinger, 1985~. Furthermore, the calcium-incluced eleva- tion in degradation is of nonlysosomal origin (Furuno and Goldberg, 1986), as evidenced by the failure of lysosomal protease inhibi- tors to inhibit this calcium-inducec3 re- sponse. At present, an inadequate mechanistic understanding of the biochemical details of protein synthesis and degraclation espe- cially degradation is blocking progress in research on the regulation of these proc- esses. Nutritional/physiological experimen- tation has provided an important descriptive base, but future progress depends on cel- lular and molecular details. As mentioned previously, new information on the regu- lation of myofibrillar protein isoform tran APPENDIX sitions and the structure ant] regulation of genes encoding these proteins will have a dramatic impact on our view of muscle protein synthesis regulation. Molecular (le- tails of the interaction of key hormones or their second messengers with myofibrillar protein genes should be forthcoming within the next clecacle. It is traditionally assumed that lysosomal enzymes are responsible for intracellular protein degradation. These proteases are contained in lysosomes and are active at acidic pH. Several clifferent proteases are  grouped in this class and called cathepsins. Not all cathepsins are able to cleave peptize bonds in myofibrillar proteins; only cath- epsins Be, D, H. and L have been fount] in muscle and are active on myofibrilIar protein substrates (see Goll et al., 1983~. A problem with attributing myofibrillar protein cleg- radation in skeletal muscle to catheptic proteases is the fact that myofibrils or my- ofilaments have not been observed in ly- sosomal structures in muscle. Nor have lysosome-like organelles been observed in association with myofibrils. In addition, treatment of cells with lysosomal enzyme inhibitors failed to suppress calcium-in- cluced protein degradation (Furuno ant] Goldberg, 19864. Although it has been pos- sible in some cases to show correlations between lysosomal enzyme activity and pro- tein degradation, the cause-and-effect re- lationship has not been proved. A more likely mechanism for explaining myofibrillar protein degradation begins with the action of nonlysosomal cytoplasmic pro- teases that selectively cleave certain myofi- brillar proteins, resulting in the disassembly of filaments in the myofibril (Dayton et al., 1975~. Individual myofibrillar proteins or fragments of these proteins can then be taken up by lysosomes and degraded to individual amino acids. If such a scheme is accurate, one of the rate-limiting steps in the process would be the initial degradation steps accomplished by nonlysosomal pro- teases. Recent evidence suggests that acti- vation of calcium-induced and injury-in

MUSCLE CELL GROWTH AND DEVELOPMENT cluced protein degradation in muscle does not involve a lysosomal mechanism (Furuno and Goldberg, 1986~. A couple of strong cancliciates have been suggested for this clegraclation role, the first of which is the calcium-clepencient neutral protease described by Dayton et al. (1976~. This protease, with a molecular weight of 110,000 daltons, is located inside skeletal muscle cells, as well as many other cell types, and is active at neutral pH. In skeletal muscle cells, it is found in the sarcoplasm and not in lysosomal structures or other intracellular membrane-bounc! organelles. The specificity of this protease is somewhat limited in that it generally cleaves only one or a few peptide bonds in a protein. In the myofibril, the proteins affected are tro- ponin-T, troponin-I, tropomyosin, C-pro- tein, filamin, desmin, the Z-line structure, and possibly titin (Golf et al., 1983~. Many of these proteins have regulatory and struc- tural significance. Note, however, that the primary proteins in the myofibril actin and myosin are apparently not hydrolyzes] by this protease. Although the regulatory details of this protease have not been elucidated, it is clear that calcium ions and a free sul~ydryl group are requires! for activity. It is also accepted that two forms of the protease exist, one that requires millimolar concen- trations of calcium and another that only requires micromolar concentrations for ac- tivity. These are distinctly different proteins that share a high degree of sequence ho- mology. The active sites of these proteins are similar to those of papain, and the caTcium-bincling regions are similar to those of calmodulin (Emori et al., 1986~. To add to the complexity of the system, an inhibitor of these proteases is also found in skeletal muscle. The physiological regulation of these different forms of the enzyme and inhibitor is not clear, but it may be crucial to an understanding of protein degradation ant] turnover. Other soluble proteases may also be im- portant components of the myofibril cleg 153 radation process. Several alkaline or neutral proteases have the ability to hydrolyze actin or myos~n, but most of these are not found in muscle cells. Perhaps the degradative system understood] in greatest detail is an ATP-depenclent protease system (Hershko and Ciechanover, 1982), which is found in many cells but has been most extensively stuclied in reticulocytes. This system is com- posecl of a small, heat-stable protein caller] ubiquitin (because of its presence in a highly conserved form in most cells) that interacts with an activating enzyme in an ATP-de-  pendent process to ultimately form a cova- lent isopeptide bond between the carboxyl group of the C-terminal glycine residue of ubiquitin and an epsilon amino group of a lysine on the target protein. The covalent attachment of ubiquitin is thought to target the protein for protease attack. The pro- teases responsible are ill-defined, but the end products are pepticles ant] a released ubiquitin that can recycle. This system could be responsible for identifying proteins that were damager! structurally or otherwise in- activated. Other protease systems requiring ATP may also be present in cells, but their characterization is far from complete. The primary problems with proposed roles for these ATP-dependent proteolytic systems in muscle protein degradation are the lack of detailed information about the specificity of these systems for muscle proteins and the presence ant] location of these systems in muscle cells. During normal growth and in many met- abolic states, rates of synthesis and (legra- dation tend to move in tandem. Even during fasting, clegraclation is clepressec] and not increased, presumably to spare protein. These observations suggest that during nor- mal growth, protein synthesis may repre- sent the primary site of regulation, and degradation may follow (Millward, 1985~. It is virtually impossible to tie together en- docrine and nutritional influences on animal protein degradation en cl the subcellular events that mediate these effects because of the present gap that exists between our

154 knowledge of the cellular ant] biochemical mechanisms involved in skeletal muscle protein degradation and the whole animal and tissue level descriptions of the process This does not eliminate the possibility of targeting degradation as a site for muscle growth regulation, but it makes it difficult to devise strategies to manipulate protein clegraciation to enhance the efficiency of muscle growth in meat animals. STRATEGIES FOR REGULATING MUSCLE DEVELOPMENT AND GROWTH IN MEAT-PRODUCING ANIMALS Significant research areas that can be layered over the muscle-specific problem are the integration of metabolism during growth and the manner in which tissue growth is coordinated within the animal. These topics are more general and, at face value, more pertinent to altering efficiency of protein accretion and the composition of the product than are the studies of specific cellular and biochemical events in devel- oping ant! growing muscle. But progress in these areas can only proceed as rapidly as progress toward a mechanistic understand- ing of muscle growth. Establishing muscle cellularity, in its broadest sense, involves prenatal fiber de- velopment and nuclear accretion during postnatal growth. Fiber development is the result of myogenesis that takes place in the developing embryo or fetus. The final event in this cascade of proliferation and cliffer- entiation is the fusion of myoblasts into multinucleated myotubes that mature into fibers. Currently, hormones and growth factors that stimulate and inhibit the prolif- erative and differentiative events in my- ogenic cells are being iclentifiec3, but the factors that regulate the number of fibers that are former! from a given cohort of myoblasts have not been considered exper- imentally. It may be that innervation plays a key role in establishing fiber number ant] APPENDIX organization in muscle, since innervation is required to sustain fibers. For many years, it has been accepted that major differences in muscle mass in mature animals can be attribute`] in large part to differences in fiber number. Consequently, alterations in fiber number during late prenatal life would! likely result in differences in muscularity. At present, however, there is probably insufficient mechanistic (retail to suggest specific approaches. A critical question in this regard is whether it is advisable to increase muscularity prenatally; in cattle, for example, increased management prob- lems associated with dystocia could offset  any advantages due to increased muscle growth potential. In swine or poultry, this problem may not be so acute. Cellularity could conceivably be altered by nuclear accretion postnatally without reproductive problems. Again, we are be- ginning to understand more about the ac- tivation, proliferation, and differentiation of satellite cells, although specific physiologi- cal regulators have not yet been confirmed. Assuming that satellite cell activity could] be altered and nuclear accretion in fibers coup! be influencecl, satellite cells may be more receptive to manipulation efforts dur- ing certain periods of growth than others. Early postnatal growth is the time of greatest satellite cell activity and would correspond to the period of greatest sensitivity to hor- mones and growth factors. On the other hand, later phases of growth are marked by decreasing nuclear accretion rate; therefore, stimulating additional satellite cell prolif- eration and (differentiation could result in an extension of the rapid! muscle growth phase that is normally associated with mus- cle growth in younger animals. Affecting changes in muscle growth by altering protein metabolism has been the most commonly considered avenue, pri- marily because of the erroneous assumption that cellularity does not change after birth. During normal growth, synthesis and deg- radation tend to move in parallel, with

MUSCLE CELL GROWTH AND DEVELOPMENT synthesis rate exceeding degradation rate. Consequently, accelerated growth rate is accompanied by an accelerated degradation rate; hence, there is no increase in efficiency of protein accretion. Because these proc- esses seem to be coupled, Millward (1985) suggested that manipulating synthesis may be the most reasonable way to affect protein accretion. Specific alterations in synthesis await increased knowledge of the mecha- nisms of muscle protein gene regulation and the elucidation of hormones or other exter- nal signals, such as electrical stimulation or stretch, that modulate the expression of these genes. Likewise, strategies for manipulating deg- radation rate in muscle will not progress beyond the empirical stage without a me- chanistic unclerstancling of the proteases involved and their regulation. Protein deg- radation is, however, an attractive target for postnatal growth manipulation. If degrada- tion rate could be decreased, net rate of protein accretion would be accelerated and less energy would be expended on resyn- thesizing degracled protein. To illustrate that the present level of cellular and molecular understanding of im- portant regulatory events is grossly inade- quate and, indeed, limiting, consider some current growth-manipulating techniques. Take three growth-altering treatments: growth hormone (GH), steroid hormones and their analogs, ant! beta-adrenergic ag- onists. With all three, scientists are still dependent on information that is often one or two decades old, or on empirical obser- vations, the biology of which is still not fully unclerstoocI. In these cases current biology is not leading the way to new applications; rather, new applications are leacling basic biological investigation. Let us begin with GH. Direct adminis- tration of GH to domestic meat animals was first reporter] in pigs by Truman and An- drews (1955), Henricson and Ullberg (1960), and Machlin (1972~. Later, Chung et al. (1985) also reported direct administration of 155 GH to pigs. Dramatic increases in muscle growth in GH-treated pigs (Etherton et al., 1986) could be the result of action at several sites, such as adipose tissue, where GH could be having an antilipogenic eject. If energy is not stored in adipose tissue, it may be more available for growth. It is also possible that when growth processes are stimulated, they demand more energy than do adipose tissue triglyceride storage activ- ities. GH could also be having part of its effect by stimulating higher levels of so- matomeclins that are, in turn, stimulating satellite cells. Arguments can be maple for  increaser! muscle growth as a result of nu- clear accretion and subsequent protein ac- cumulation directed by new nuclei. Another plausible alternative may be somatomeclin- mediated depression of muscle protein deg- raclation. Or, the net effect could be clue to a combination of the above. The point is that it is application that is leading scientists to undertake basic biological research. Next consider the steroid hormones and their analogs. Studies that were mostly empirical in nature gave us cliethyistilbes- trol. Its application has come and gone from agriculture, yet we still do not know its precise mode of action. Even the action of testosterone on muscle growth is unclear. Trenbolone acetate (TBA) is another ex- amplc- it stimulates growth, but again, the mechanism is unknown. In terms of the biological events responsible for muscle growth, it must either directly or in(lirectly stimulate nuclear accretion, stimulate pro- tein synthesis, or decrease protein degra- dation. At least one report suggests that protein synthesis is depresses] but that pro- tein clegradation is depressed to a greater extent, thus leacling to a net increase in rate of protein accretion as well as in efficiency of protein gain (Vernon and Buttery, 1976). This in viva study was not able to address the direct or indirect nature of the action of TBA. In vitro studies are limited; how- ever, TBA floes not appear to have a direct effect on protein clegradation in LO muscle

156 cells in culture (Ballard and Francis, 1983~. Another example of application leacling basic investigation concerns a class of agents that has receiver] a great deal of attention in recent years, the beta-adrenergic ago- nists. One of these-clenbuterol was orig- inally designed as a respiratory drug but was subsequently shown to have a stimu- latory eject on rat growth. Since then, it has been used to stimulate growth ant! feed efficiency in poultry, sheep, and cattle (Baker et al., 1984; Dairymple et al., 1984; Ricks et al., 1984~. A great deal of effort is cur- rently being devoted to understanding how it works. An obvious site of action would be as a lipolytic agent for adipose tissue; how- ever, this alone could not explain the ex- treme muscle hypertrophy observed in sheep (Beermann et al., 1986~. Recently, Kim et al. (1986) reporter] that the major eject appeared to be on hypertrophy offast-twitch muscle fibers and that muscle DNA con- centration actually decreased in the cima- terol-treated group. Beermann, however, inclicated that a significant increase in DNA content was noted in 12-week studies with sheep but that DNA content increased after muscle hypertrophy (D. H. Beermann, per- sonal communication, 1986~. In another re- port, cimaterol was demonstrated to have an inhibitory eject on protein degradation in cultured myotubes from a rat muscle cell line (Forsberg ant] Merrill, 19864. Evi- clently, the beta-adrenergic agonists may have multiple sites of action, especially for protein degradation and adipose tissue me- tabolism, but this conclusion remains highly speculative. These examples of a few of the most interesting agents currently being investi- gatec3 for use in stimulating muscle growth not only demonstrate that application is leading investigation, they also provi(le striking demonstrations that muscle growth in meat animals can be manipulated to increase protein production and decrease triglyceride deposition beyond the normal physiological limits of a particular animal. APPENDIX They also suggest that the muscle growth processes mentioned earlier protein syn- thesis ant! clegradation ant! pre- and post- natal muscle cellularity alterations-repre- sent legitimate targets for growth-regulating strategies. In the future, several approaches may be used to enhance rate and efficiency of mus- cle growth, but for now the most promising are administration of recombinant hor- mones. As indicated, recombinant GH has been shown to have impressive stimulatory effects on growth, feed efficiency, and car- cass composition in pigs. Other hormones  will surely be investigated in a similar man- ner. Based on recent research in muscle development, somatomedin-C/IGF-I is a logical choice for such application. New growth factors or combinations of growth factors that affect muscle development, such as fibroblast growth factor and IGF-II, are also candidates. At present, GH administration entails regular infections during later stages in postnatal life. In a seconct generation ot studies, researchers may wish to eject a permanent change in the cellularity of the animal, such as increased fiber number or myonuclei content. In contrast to ap- proaches that are designed primarily to alter protein metabolism, cellular/developmental changes may only require acute treatments during early, critical stages of development. Therefore, the need for costly, labor-inten- sive administration schemes could be elim- inated, as would potential questions about the presence of drug residues in the final product. At another level of sophistication, trans- genic animals may have a place in livestock production systems. Growth has already been accelerated in transgenic mice carrying a metallothionein-human growth hormone fusion gene (Palmiter et al., 1982) and in mice expressing the metallothionein-human growth hormone releasing factor minigene (Hammer et al., 1985~. In addition, these genes have been shown to be transmittable - O ~

MUSCLE CELL GROWTH AND DEVELOPMENT to subsequent generations, although repro- duction suffered in some ofthe initial studies (Hammer et al., 1985~. These techniques will uncloubtedly be applied in large do- mestic animals to produce new germplasm. Furthermore, it may be possible to con- struct and perpetuate the genes of important hormones that can be regulated by coupling the genes to promoters that can be turned on or off at critical periods through nutri- tional, pharmacological, or environmental manipulation. These approaches will ob- viously require a more detailed description ofthe significant regulatory events in muscle growth and the important factors that me- diate these events so that appropriate mo- lecular targets can be selectecI. CONCLUSIONS Major obstacles exist. New fundamental knowledge of cellular and molecular mech- anisms of growth is desperately needed. Technical advances are also needed in the area of delivery systems for effectively ac3- ministering exogenous agents at specific times and in appropriate amounts. Of critical importance are practical means for targeting the clelivery of agents to specific tissues. It is conceivable that a factor could have a beneficial effect on one tissue or organ ant! a detrimental effect on another. This may be a major impediment to the application of certain hormones or growth factors. Tech- nical advances are still needler! in gene transfer and gene construct technology, but progress is occurring rapidly. These are only a few of the problem areas that need to be acIdressecI. Advances in the production of nutritious muscle protein foods will probably not come by altering the cellular composition of a muscle fiber. Membrane systems ant! my- ofibriliar proteins in muscle are highly con- servec] ancl may not be amenable to efforts to inflict gross alterations that would] provide a more desirable balance of amino acids or reduced cholesterol content. An approach 157 to improving the nutritional attributes of meat products that holds greater promise is one that attempts to reduce the amount of adipose tissue associated with meat products while maintaining palatability. Great ad- vances can be macle in the efficient produc- tion of muscle protein by providing a grow- ing knowledge base in biology, by rapidly adopting new scientific technologies, and by fostering innovative applied research. REFERENCES Allbrook, D. B., M. F. Han, and A. E. Hellmuth. 1971. Population of muscle satellite cells in relation  to age and mitotic activity. Pathology 3:233-243. Allen, R. E. 1986. Transforming growth factor-beta inhibits the IGF-I-induced proliferation and differ- entiation of skeletal muscle satellite cells. J. Cell Biol. 10345~:120a. Allen, R. E., R. A. Merkel, and R. B. Young. 1979. Cellular aspects of muscle growth. Myogenic cell proliferation. J. Anim. Sci. 49~1~:115-127. Allen, R. E., P. K. McAllister, and K. C. Masak. 1980. Myogenic potential of satellite cells in skeletal muscle of old rats. Mech. Age. Dev. 13:105-109. Allen, R. E., P. K. McAllister, K. C. Masak, and G. R. Anderson. 1982. Influence of age on accumulation of Martin in satellite-cell-derived myotubes in vitro. Mech. Age. Dev. 18:89-95. Allen, R. E., M. V. Dodson, and L. S. Luiten. 1984. Regulation of skeletal muscle satellite cell prolifer- ation by bovine pituitary fibroblast growth factor. Exp. Cell Res. 152:154-160. Allen, R. E., M. V. Dodson, L. K. Boxhorn, S. L. Davis, and K. L. Hossner. 1986. Satellite cell proliferation in response to pituitary hormones. J. Anim. Sci. 62:1596-1601. Ashmore, C. R., D. W. Robinson, P. Rattray, and L. Doerr. 1972. Biphasic development of muscle fibers in the fetal lamb. Exp. Neurol. 37(2~:241-255. Bains, W., P. Ponte, H. Blau, and L. Kedes. 1984. Cardiac actin is the major actin gene product in skeletal muscle cell differentiation in vitro. Mol. Cell. Biol. 4:1449-1453. Baird, A., P. Mormede, and P. Bohlen. 1985. Im- munoreactive fibroblast growth factor in cells of peritoneal exudate suggests its identity with mac- rophage-derived growth factor. Biochem. Biophys. Res. Commun. 126:358-364. Baker, P. K., R. H. Dalrymple, D. L. Ingle, and C. A. Ricks. 1984. Use of a ,B-adrenergic agonist to alter muscle and fat deposition in lambs. J. Anim. Sci. 59:1256-1261. Ballard, F. J., and G. L. Francis. 1983. Effects of C7 - - - _

158 anabolic agents on protein breakdown in LO myo- blasts. Biochem. J. 210:243-249. Ballard, F. J., L. C. Read, G. L. Francis, C. J. Bagley, and J. C. Wallace. 1986. Binding properties and biological potencies of insulin-like growth factors in L6 myoblasts. Biochem. J. 233:223-230. Bandman, E., R. Matsuda, and R. C. Strohman. 1982. Developmental appearance of myosin heavy and light chain isoforms in viva and in vitro in chicken skeletal muscle. Dev. Biol. 93:508-518. Baracos, V. E., and A. L. Goldberg. 1985. Ca2+, interleukin-1 and failure to maintain normal length stimulate protein degradation in isolated skeletal muscle. Pp. 603~05 in Intracellular Protein Catab- olism, E. A. Khaviallah, J. S. Bond, and J. W. C. Bird (eds.~. New York: Alan R. Liss. Beermann, D. H., D. E. Hogue, V. K. Fishell, R. H. Dalrymple, and C. A. Ricks. 1986. Effects of cima- terol and fishmeal on performance, carcass charac- teristics and skeletal muscle growth in lambs. J. Anim. Sci. 62:37~380. Bergsma, D. J., J. M. Grichnik, L. M. A. Gossett, and R. J. Schwartz. 1986. Delimitation and charac- terization of cis-acting DNA sequences required for the regulated expression and transcriptional control of the chicken skeletal a-actin gene. Mol. Cell. Biol. 6:246~2475. Bischoff, R. 1974. Enzymatic liberation of myogenic cells from adult rat muscle. Anat. Rec. 180:645-662. Bischoff, R. 1975. Regeneration of single skeletal muscle fibers in vitro. Anat. Rec. 182:215-236. Bonner, P. H., and T. R. Adams. 1982. Neural induction of chick myoblast differentiation in culture. Dev. Biol. 90:175-184. Brevet, A., E. Pinto, J. Peacock, and F. E. Stockdale. 1976. Myosin synthesis increased by electrical stim- ulation of skeletal muscle cell cultures. Science 193:1152-1154. Buckley, P. A., and I. R. Konigsberg. 1974. Myogenic fusion and the duration of the post-mitotic gap (G~). Dev. Biol. 37:193-212. Butler-Brown, G. S., G. S. Cambon, C. Laurent, G. Prulier, M. Toutant, S. Watkins, and R. G. Whalen. 1986. Control of myosin isoenzyme transitions by thyroid hormone in the dwarf mouse strain. J. Physiol. 377:93. Caplan, A. I., M. Y. Fiszman, and H. M. Eppenberger. 1983. Molecular and cell isoforms during develop- ment. Science 221:921-927. Cardasis, C. A., and G. W. Cooper. 1975. An analysis of nuclear numbers in individual muscle fibers during differentiation and growth. A satellite cell-muscle fiber growth unit. J. Exp. Zool. 19143):347~58. Chiu, C. P., and H. M. Blau. 1984. Reprogramming cell differentiation in the absence of DNA synthesis. Cell 37:87~887. Chung, C. S., T. D. Etherton, and J. P. Wiggins. APPENDIX 1985. Stimulation of swine growth by porcine growth hormone. J. Anim. Sci. 60(1):118-130. Cossu, G., B. Zani, M. Colletta, M. Bouche, M. Pacifici, and M. Molinaro. 1980. In vitro differen- tiation of satellite cells isolated from normal and  dystrophic mammalian muscles. A comparison with embryonic myogenic cells. Cell Differ. 9:357-368. Cossu, G., M. Molinaro, and M. Pacifici. 1983. Dif- ferential response of satellite cells and embryonic myoblasts to a tumor promoter. Dev. Biol. 98:52 524. Cossu, G., P. Cicinelli, C. Fieri, M. Coletta, and M. Monlinaro. 1985. Emergence of TPA-resistant 'Sat- ellite' cells during muscle histogenesis of human limb. Exp. Cell Res. 160:403~11. Crow, M. T., P. S. Olson, and F. E. Stockdale. 1983. Myosin light-chain expression during avian muscle development. J. Cell Biol. 96:73~744. Dalrymple, R. H., C. A. Ricks, P. K. Baker, J. M. Pensak, P. E. Gingher, and D. L. Ingle. 1984. Use of the beta-agonist clenbuterol to alter carcass com- position in poultry. Poultry Sci. 63:237~2383. Dayton, W. R., D. E. Goll, M. H. Stromer, W. J. Reville, M. G. Zeece, and R. M. Robson. 1975. Some properties of a Ca++-activated protease that may be involved in myofibrillar protein turnover. Pp. 551-577 in Proteases and Biological Control, E. Reich, D. B. RiEkin, and E. Shaw (eds.). Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. Dayton, W. R., D. E. Goll, M. G. Zeece, R. M. Robson, and W. J. Reville. 1976. A Ca2+-activated protease possibly involved in myofibrillar protein turnover. Purification from porcine muscle. Bio- chemistry 15:215~2158. Devlin, R. B., and C. P. Emerson, Jr. 1978. Coordinate regulation of contractile protein synthesis during myoblast differentiation. Cell 13:599~11. Devlin, R. B., and C. P. Emerson, Jr. 1979. Coordinate accumulation of contractile protein mRNAs during myoblast differentiation. Dev. Biol. 69:20~216. Devlin, B. H., and I. R. Konigsberg. 1983. Reentry into the cell cycle of differentiated skeletal myocytes. Dev. Biol. 95:17~192. Dodson, M. V., R. E. Allen, and K. L. Hossner. 1985. Ovine somatomedin, multiplication-stimulating ac- tivity, and insulin promote skeletal muscle satellite cell proliferation in vitro. Endocrinology 117:2357- 2363. Eichhorn, J. M., L. J. Coleman, E. J. Wakayama, G. J. Blomquist, D. M. Bailey, and T. G. Jenkins. 1986. Effects of breed type and restricted versus ad libitum feeding on fatty acid composition and cho- lesterol content of muscle and adipose tissue from mature bovine females. J. Anim. Sci. 63:781-794. Emori, Y., H. Kawasaki, H. Sugihara, S. Imajoh, S. Kawashima, and K. Suzuki. 1986. Isolation and sequence analyses of cDNA clones for the large

MUSCLE CELL GROWTH AND DEVELOPMENT subunits of two isozymes of rabbit calcium-depend- ent protease. J. Biol. Chem. 261:9465-9471. Etherton, T. D., C. M. Evock, C. S. Chung, P. E. Walton, M. N. Sillence, K. A. Magri, and R. E. Ivy. 1986. Stimulation of pig growth performance by long-term treatment with pituitary porcine growth hormone (pGH) and a recombinant pGH. J. Anim. Sci. 63:219 (Abstr.~. Evinger-Hodges, M. J., D. Z. Ewton, S. C. Seifert, and J. R. Florini. 1982. Inhibition of myoblast differentiation in vitro by a protein isolated from liver cell medium. J. Cell Biol. 93:395~01. Ewton, D. Z., and J R. Florini. 1980. Relative effects of the somatomedins, multiplication-stimulating ac- tivity, and growth hormone on myoblasts and my- otubes in culture. Endocrinology 106:577-583. Ewton, D. Z., and J. R. Florini. 1981. Effects of the somatomedins and insulin on myoblast differentia- tion in vitro. Dev. Biol. 86:31~9. Florini, J. R., A. B. Roberts, D. Z. Ewton, S. L. Falen, K. C. Flanders, and M. B. Sporn. 1986. Transforming growth factor-,13. A very potent inhib- itor of myoblast differentiation, identical to the differentiation inhibitor secreted by buffalo rat liver cells. J. Biol. Chem. 261:16509-16513. Forsberg, N. E., and G. Merrill. 1986. Effects of cimaterol on protein synthesis and degradation in monolayer culture of rat and mouse myoblasts. J. Anim. Sci. 63:222 (Abstr.~. Furuno, K., and A. L. Goldberg. 1986. The activation of protein degradation in muscle by Cat+ or muscle injury does not involve a lysosomal mechanism. Biochem. J. 237:859-864. Gambke, B., and N. A. Rubinstein. 1984. A monoclonal antibody to the embryonic myosin heavy chain of rat skeletal muscle. J. Biol. Chem. 259:12092-12100. Gauthier, G. F., S. Lowey, P. A. Benfield, and A. W. Hobbs. 1982. Distribution and properties of myosin isozymes in developing avian and mammalian skel- etal muscle fibers. J. Cell Biol. 92:471~84. Gibson, M. C., and E. Schultz. 1983. Age-related differences in absolute numbers of skeletal muscle satellite cells. Muscle Nerve 6:57~580. Goldberg, A. L., V. Baracos, P. Rodemann, L. War- man, and C. Dinarello. 1984. Control of protein degradation in muscle by prostaglandins, Ca2+, and leukocytic pyrogen (interleukin 1). Proc. Fed. Am. Soc. Exp. Biol. 43:1301-1306. Goldspink, G. 1972. Postembryonic growth and dif- ferentiation of striated muscle. P. 179 in The Struc- ture and Function of Muscle, Vol. 1, 2nd Ed., G. H. Bourne (ed.). New York: Academic Press. Goldspink, D. F. 1978. The influence of passive stretch on the growth and protein turnover of the denervated extensor digitorum longus muscle. Biochem. J. 174:595-602. Goll, D. E., Y. Otsuka, P. A. Nagainis, J. D. Shannon, ]59 S. K. Sathe, and M. Muguruma. 1983. Role of muscle proteinases in maintenance of muscle integ- rity and mass. J. Food Biochem. 7:137-177. Goll, D. E., R. M. Robson, and M. H. Stromer. 1984. Skeletal muscle, nervous system, temperature reg-  ulation, and special senses. Pp. 548-580 in Dukes' Physiology of Domestic Animals, 10th Ed., M. J. Swenson (ea. ). Ithaca, N. Y.: Cornell University Press. Gospodarowicz, D., J. Weseman, J. S. Moran, and J. Lindstrom. 1976. Effect of fibroblast growth factor on the division and fusion of bovine myoblasts. J. Cell Biol. 70:395 405. Hammer, R. E., R. L. Brinster, and R. D. Palmiter. 1985. Use of gene transfer to increase animal growth. Cold Spring Harbor Symp. Quant. Biol. 50:379- 387. Harbison, S. A., D. E. Goll, F. C. Parrish, V. Wang, and E. A. Kline. 1976. Muscle growth in two genetically different lines of swine. Growth 40:25 283. Hauschka, S. D. 1974. Clonal analysis of vertebrate myogenesis. III. Developmental changes in the muscle-colony-forming cells of the human fetal limb. Dev. Biol. 37:345 368. Henricson, B., and S. Ullberg. 1960. Effects of pig growth hormone on pigs. J. Anim. Sci. 19:1002. Hershko, A., and A. Ciechanover. 1982. Mechanisms of intracellular protein breakdown. Annul Rev. Biochem. 51:335 364. Hoffman, R. K., B. Gambke, L. W. Stephenson, and N. A. Rubinstein. 1985. Myosin transitions in chronic stimulation do not involve embryonic isozymes. Muscle Nerve 8:796-805. Holtzer, H., and R. Bischoff. 1970. Mitosis and my- ogenesis. Pp. 29-51 in Physiology and Biochemistry of Muscle as a Food, Vol. 2, E. J. Briskey, R. G. Cassens, and B. B. Marsh (eds.~. Madison: Univer- sity of Wisconsin Press. Hu, M. C. T., S. B. Sharp, and N. Davidson. 1986. The complete sequence of the mouse skeletal a- actin gene reveals several conserved and inverted repeat sequences outside of the protein-coding re- gion. Mol. Cell. Biol. 6:15-25. Izumo, S., B. Nadal-Ginard, and V. Mahdavi. 1986. All members of the MHC multigene family respond to thyroid hormone in a highly tissue-specific man- ner. Science 231:597-600. Janeczko, R. A., and J. D. Etlinger. 1984. Inhibition of intracellular proteolysis in muscle cultures by multiplication-stimulating activity. Comparison of effects of multiplication-stimulating activity and in- sulin on proteolysis, protein synthesis, amino acid uptake, and sugar transport. J. Biol. Chem. 259:6292- 6297. Kardami, E., D. Spector, and R. C. Strohman. 1985. Myogenic growth factor present in skeletal muscle

160 is purified by heparin-affinity chromatography. Proc. Natl. Acad. Sci. USA 82:804~8047. Kelly, A. M. 1978a. Perisynaptic satellite cells in the developing and mature rat soleus muscle. Anat. Rec. 190:891-904. Kelly, A. M. 1978b. Satellite cells and myofiber growth in the rat soleus and extensor digitorum longus muscles. Dev. Biol. 65:1-10. Kelly, A. M., and S. I. Zacks. 1969. The histogenesis of rat intercostal muscle. J. Cell Biol. 42:13~153. Kim, Y. S., Y. B. Lee, C. R. Ashmore, and R. H. Dalrymple. 1986. Effect of the repartitioning agent, cimaterol (CL 263,780), on growth, carcass charac- teristics and skeletal muscle cellularity of lambs. J. Anim. Sci. 63:221 (Abstr.). Klapper, D. G., M. E. Svoboda, and J. J. Van Wyk. 1983. Sequence analysis of somatomedin-C. Confir- mation of identity with insulin-like growth factor I. Endocrinology 112:2215-2217. Konigsberg, U. R., B. H. Lipton, and I. R. Konigsberg. 1975. The regenerative response of single mature muscle fibers. Dev. Biol. 45:26~275. Lewis, S. E. M., F. J. Kelly, and D. F. Goldspink. 1984. Pre- and post-natal growth and protein turn- over in smooth muscle, heart and slow- and fast- twitch skeletal muscles of the rat. Biochem. J. 217:517-526. Linkhart, T. A., C. H. Clegg, and S. D. Hauschka. 1981. Myogenic differentiation in permanent clonal mouse myoblast cell lines: Regulation by macro- molecular growth factors in the culture medium. Dev. Biol. 86:19-30. Lowey, S., P. A. Benfield, D. D. LeBlanc, and G. S. Waller. 1983. Myosin isozymes in avian skeletal muscles. I. Sequentialexpressionofmyosinisozymes in developing chicken pectoralis muscles. J. Mus. Res. Cell Motil. 4:695-716. Lyons, G. E., J. Haselgrove, A. M. Kelly, and N. A. Rubinstein. 1983. Myosin transitions in developing fast and slow muscles of the rat hindlimb. Differ- entiation 25:16~175. MacDonald, M. L., and R. W. Swick. 1981. The effect of protein depletion and repletion on muscle-protein turnover in the chick. Biochem. J. 194:811~19. Machlin, L. J. 1972. Effect of porcine growth hormone on growth and carcass composition of the pig. J. Anim. Sci. 35(4):794. Marechal, G., K. Schwartz, G. Beckers-Bleukx, and E. Ghins. 1984. Isozymes of myosin in growing and regenerating rat muscles. Eur. J. Biochem. 138:421- 428. Marquardt, H., and G. J. Todaro. 1981. Purification and primary structure of a polypeptide with multi- plication-stimulating activity from rat liver cell cul- tures. Homology with human insulin-like growth factor II. J. Biol. Chem. 256:6859-6865. Mauro, A. 1961. Satellite cell of skeletal muscle fibers. J. Biophys. Biochem. Cytol. 9:493~95. APPENDIX McCarthy, F. D., W. G. Bergen, and D. R. Hawkins. 1983. Muscle protein turnover in cattle of differing genetic backgrounds as measured by urinary N'r- methylhistidine excretion. J. Nutr. 113:2455-  2463. McGrath, J. A., and D. F. Goldspink. 1982. Gluco- corticoid action on protein synthesis and protein breakdown in isolated skeletal muscles. Biochem. J. 206:641 645. McLennan, I. S. 1983. Neural dependence and inde- pendence of myotube production in chicken hind- limb muscles. Dev. Biol. 98:287-294. Melloul, D., B. Aloni, J. Calvo, D. Yaffe, and U. Nudel. 1984. Developmentally regulated expression of chimeric genes containing muscle actin DNA sequences in transfected myogenic cells. Eur. Mol. Biol. Org. J. 3:983-990. Miller, J. B., and F. E. Stockdale. 1986. Develop- mental origins of skeletal muscle fibers. Clonal analysis of myogenic cell lineages based on expres- sion of fast and slow myosin heavy chains. Proc. Natl. Acad. Sci. USA 83:386() 3864. Millward, D. J. 1985. The physiological regulation of proteolysis in muscle. Biochem. Soc. Trans. 13:102 1026. Millward, D. J., and J. C. Waterlow. 1978. Effect of nutrition on protein turnover in skeletal muscle. Fed. Proc. 37:2283-2290. Millward, D. J., P. J. Garlick, D. O. Nnanyelugo, and J. C. Waterlow. 1976. The relative importance of muscle protein synthesis and breakdown in the regulation of muscle mass. Biochem. J. 156:185- 188. Mitch, W. E., and A. S. Clark. 1984. Specificity of the effects of leucine and its metabolites on protein degradation in skeletal muscle. Biochem. J. 222:579- 586. Moses, H. L., R. F. Tucker, E. B. Leof, R. J. Coffey, Jr., J. Halper, and G. D. Shipley. 1985. Type-,B transforming growth factor is a growth stimulator and a growth inhibitor. Pp. 65-71 in Cancer Cells 3/Growth Factors and Transformation, J. Feramisco, B. Ozanne, and C. Stiles (eds.~. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. Moss, F. P. 1968. The relationship between the dimensions of the fibres and the number of nuclei during normal growth of skeletal muscle in the domestic fowl. Am. J. Anat. 122:555-564. Moss, F. P., and C. P. Leblond. 1971. Satellite cells as the source of nuclei in muscles of growing rats. Anat. Rec. 170:421~36. Nadal-Ginard, B. 1978. Commitment, fusion and bio- chemical differentiation of a myogenic cell line in the absence of DNA synthesis. Cell 15:855-864. Nadal-Ginard, B., H. T. Nguyen, and V. Mahdavi. 1984. Induction of muscle-specific genes in the absence of commitment is reversible. Exp. Biol. Med. 9:25~259.

MUSCLE CELL GROWTH AND DEVELOPMENT Nguyen, G. T., R. M. Medford, and B. Nadal-Ginard. 1983. Reversibility of muscle differentiation in the absence of commitment. Analysis of a myogenic cell line temperature-sensitive for commitment. Cell 34:281-293. Olwin, B. B., and S. D. Hauschka. 1986. Identification of the fibroblast growth factor receptor of Swiss 3T3 cells and mouse skeletal muscle myoblasts. Bio- chemistry 25:3487-3492. Palmiter, R. D., R. L. Brinster. R. E. Hammer. M. . . E. Trumbauer, M. G. Rosenfeld, N. C. Birnberg, and R. M. Evans. 1982. Dramatic growth of mice that develop from eggs microinjected with metallo- thionein-growth hormone fusion genes. Nature 300:611-614. Paterson, B. M., and J. D. Eldridge. 1984. a-Cardiac actin is the major sarcomeric isoform expressed in embryonic avian skeletal muscle. Science 224: 143 1438. Peter, J. B., R. J. Barnard, V. R. Edgerton, C. A. Gillespie, and K. E. Stempel. 1972. Metabolic profiles of three fiber types of skeletal muscle in guinea pigs and rabbits. Biochemistry 11:2627- 2633. Powell, S. E., and E. D. Aberle. 1981. Skeletal muscle and adipose tissue cellularity in runt and normal birth weight swine. J. Anim. Sci. 52:748-756. Quinn, L. S., M. Nameroff, and H. Holtzer. 1984. Age-dependent changes in myogenic precursor cell compartment sizes. Exp. Cell Res. 154:65-82. Reiser, R. 1975. Fat has less cholesterol than lean. J. Nutr. 105:15-16. Rhee, K. S., T. R. Dutson, and G. C. Smith. 1982. Effect of changes in intermuscular and subcutaneous fat levels on cholesterol content of raw and cooked beef steaks. J. Food Sci. 47:1638-1642. Ricks, C. A., R. H. Dalrymple, P. K. Baker, and D. L. Ingle. 1984. Use of a ,B-agonist to alter fat and muscle deposition in steers. J. Anim. Sci. 59:1247- 1255. Rinderknecht, E., and R. E. Humbel. 1978. Primary structure of human insulin-like growth factor II. FEBS 89:28~286. Robbins, J., T. Horan, J. Gulick, and K. Kropp. 1986. The chicken myosin heavy chain family. J. Biol. Chem. 261:660~6612. Rubinstein, N. A., and A. M. Kelly. 1978. Myogenic and neurogenic contributions to the development of fast and slow twitch muscles in the rat. Dev. Biol. 62:473 485. Rutz, R., and S. Hauschka. 1982. Clonal analysis of , , vertebrate myogenesis. VII. Heritability of muscle colony type through sequential subclonal passages in vitro. Dev. Biol. 91:10~110. Savell, J. W., lI. R. Cross, and G. C. Smith. 1986. Percentage ether extractable fat and moisture con- tent of beef longissimus muscle as related to USDA marbling score. J. Food Sci. 51:838~40. 161 Schultz, E. 1974. A quantitative study of the satellite cell population in postnatal mouse lumbrical muscle.  Anat. Rec. 180:589~96. Seed, J., and S. D. Hauschka. 1984. Temporal sepa- ration ofthe migration of distinct myogenic precursor populations into the developing chick wing bud. Dev. Biol. 106:389-393. Silver, G., and J. D. Etlinger. 1985. Regulation of myofibrillar accumulation in chick muscle cultures. Evidence for the involvement of calcium and lyso- somes in non-uniform turnover of contractile pro- teins. J. Cell Biol. 101:238~2391. Snow, M. H. 1977. Myogenic cell formation in regen- erating rat skeletal muscle injured by mincing. II. An autoradiographic study. Anat. Rec. 188:201-218. Stockdale, F. E., and H. Holtzer. 1961. DNA synthesis and myogenesis. Exp. Cell Res. 24:508-520. Stromer, M. H., D. E. Goll, and J. H. Roberts. 1966. Cholesterol in subcutaneous and intramuscular lipid depots from bovine carcasses of different maturity and fatness. J. Anim. Sci. 25:1145-1147. Summers, P. J., C. R. Ashmore, Y. B. Lee, and S. Ellis. 1985. Stretch-induced growth in chicken wing muscles. Role of soluble growth-promoting factors. J. Cell. Physiol. 125:288-294. Swatland, H. J. 1976. Recent research on postnatal muscle development in swine. Proc. Recip. Meat Conf. 29:86-103. Swatland, H. J. 1977. Accumulation of myofiber nuclei in pigs with normal and arrested development. J. Anim. Sci. 44:759-764. Tischler, M. E. 1981. Hormonal regulation of protein degradation in skeletal and cardiac muscle. Life Sci. 28: 256~2576. Tischler, M. E., M. Desautels, and A. L. Goldberg. 1982. Does leucine, leucyl-tRNA, or some metab- olite of leucine regulate protein synthesis and deg- radation in skeletal and cardiac muscle? J. Biol. Chem. 257:161~1621. Tomas, F. M., J. Murray, and L. M. Jones. 1984. Interactive effects of insulin and corticosterone on myofibrillar protein turnover in rats as determined by N-methylhistidine excretion. Biochem. J. 220:469- 479. Trenkle, A., D. L. DeWitt, and D. G. Topel. 1978. Influence of age, nutrition and genotype on carcass traits and cellular development of the M. Longissi- mus of cattle. J. Anim. Sci. 46:1597-1603. Truman, E. J., and F. N. Andrews. 1955. Some effects of purified anterior pituitary growth hormone on swine. J. Anim. Sci. 14:7. Vernon, B. G., and P. J. Buttery. 1976. Protein turnover in rats treated with trienbolone acetate. Br. J. Nutr. 36:575-579. Watt, B. K., and A. L. Merrill. 1963. Composition of foods- raw, processed, prepared. In USDA Agri- cultural Handbook 8. Washington, D.C.: U.S. De- partment of Agriculture.

162 Whalen, R. G., G. S. Butler-Browne, and F. Gros. 1978. Identification of a novel form of myosin light chain present in embryonic muscle tissue and cul- tured muscle cells. J. Mol. Biol. 126:41~431. White, N. K., P. H. Bonner, D. R. Nelson, and S. D. Hauschka. 1975. Clonal analysis of vertebrate myogenesis. IV. Medium-dependent classification of colony forming cells. Dev. Biol. 44:346-361. Winick, M., and A. Nobel. 1966. Cellular response in APPENDIX rats during malnutrition at various ages. J. Nutr. 89:300-306. Young, R. B., and R. E. Allen. 1979. Transitions in gene activity during development of muscle fibers. J. Anim. Sci. 48:837-852. Young, R. B., D. M. Moriarity, and C. E. McGee. 1986. Structural analysis of myosin genes using recombinant DNA techniques. J. Anim. Sci. 63:259- 268.

The Role of Growth Hormone in Fat Mobilization H. M A U RICE G O O D M A N Around 1931, several papers appeared in both the English and German literature suggesting that the pituitary gland contained a fat mobilizing or fat metabolism substance (Anselmino and Hoffman, 1931; Burn and Ling, 1929, 1930~. The first indication that growth hormone might be that substance came from Lee and Shaffer (1934), who showecl, by analysis of carcass composition, that animals treated with a pituitary prep- aration rich in growth-promoting activity had less fat than untreated animals ant] that the composition of the growth that ensued largely favorer! the accumulation of protein. Rats treated with highly purified growth hormone had consiclerably less body fat than dic3 control rats; growth hormone favored the deposition of more protein and less fat (Li et al., 1949~. The decrease in the pro- portion offal seen in rats treated with growth hormone reflects a decrease in the amount of lipid stored in the adipose tissue (Good- man, 19631. This decrease in adipose mass could be the result of changes in several aspects of lipid metabolism. For example, there was a decrease in fat synthesized within the tissue, as well as a decrease in the deposition 163 of fat synthesized in the liver or consumed in the cliet. There was also an increase in mobilization of fat from the adipose tissue. These data suggest that adipose tissue might be a target tissue for growth hormone. The fat cell readily stores preformed fat that enters by way of the gut or is synthe- sizec! in the liver. In adclition, it can syn- thesize fat from glucose or amino acids. Lipic! is stored in adipose tissue in the form of triglyceride, which is a triester composed of three molecules of Tong-chain fatty acids per molecule of glycerol. Stored lipids can be mobilized from the fat cell to meet the energy needs of muscle and other tissues. Fat leaves the adipose cell in the form of free fatty acids (FFAs) after cleavage of the three ester bonds of the triglyceride. FFAs released from adipose tissue can be con- sumed directly by muscle. It appears that muscle takes up FFAs from the circulation in proportion to the amount that is there (Armstrong et al., 1961), although muscle may not immecliately burn all the FFAs extracted from the circulation. In addition,  muscle and other tissues consume the car- bons of fatty acids after conversion of FFAs to ketone bodies in the liver. Thus, regu ~. ~.. ~. ~

164 ration of lipid storage, mobilization, and oxidation is really determined by events that take place at the level of the fat cell. The glycerol released along with the FFAs travels to the liver, where it can serve as a substrate for gluconeogenesis. Growth hormone might act in several ways to decrease the amount of fat in adipose tissue. It might promote fatty acid mobili- zation and thereby oxidation, or it might decrease fatty acid synthesis. Either would be consistent with previous reports, in which growth hormone was reported to decrease the respiratory quotient (the ratio of CO2 produced to oxygen consumed) (Astwood, 1955; DeBodo and Altszuler, 1957; Ketterer et al., 1957~. To determine whether growth hormone decreased carbohydrate utilization and fatty acid synthesis in adipose tissue, Goodman (1968b) injected hypophysectom- ized rats with growth hormone. At various times thereafter, the epididymal fat was removed, divided into segments, and in- cubated in vitro along with various radio- active substrates. In tissue segments from animals that were treated with growth hor- mone 3.5 hours earlier, there was a decrease in the utilization of glucose both in terms of its oxidation to CO2 and its conversion to fatty acids. Oxidation of pyruvate and fruc- tose and the incorporation of their carbons into long-chain fatty acids were similarly reduced. Thus, growth hormone, even as early as 3.5 hours after injection, decreased the conversion of carbohydrate to fat. There- fore, one of the ways in which growth hormone decreases carcass fat is to decrease the synthesis of triglycerides in adipose tissue. The rat is one species that relies heavily on its adipose tissue for synthesis of long-chain fatty acids. In other species, the liver is the principal site of lipogenesis; thus, it is reasonable to expect a similar effect of growth hormone on the liver. Another way in which growth hormone may decrease the content of lipid in adipose tissue is by promoting fatty acid release. Goodman and Knobi} (1959) treated intact APPENDIX and hypophysectomized rhesus monkeys with growth hormone at 8:00 a.m., imme- diately after removing food from their cages. Blood samples were obtained from the fem- oral vein at various times during the day. In the control animals, plasma concentra- tions of FFAs increased about fourfold in 8 hours. When these animals were given 50 fig of simian growth hormone per kilogram of body weight, FFA concentrations in- creased even more rapidly and were si~nif- icantly higher at 4 and 8 hours. Similar results were obtained in hypophysectom- ized animals, except that the rate of mobi-  lization of FFAs in the untreated monkeys was significantly lower than normal!. Two important points are illustrated by these experiments. First, the effects of growth hormone are slow to appear and last for a Tong time. Second, animals must be fasted for this effect of growth hormone to be seen. When FFAs were measured in monkeys or rats that were allowed to eat during the experiment, the effects of growth hormone on fat mobilization were small and difficult to show. This is largely because there are many other influences, in addition to growth hormone, that affect fat and carbohydrate metabolism. Certainly insulin, and also glu- cose, have very marked effects on the re- lease of FFAs from adipose tissue. This has complicated studies of the actions of growth hormone and has contributed to the contro- versy over whether growth hormone is a lipolytic agent. It therefore appears that in order to see a growth hormone effect, some other signal that operates simultaneously is needed for fatty acid mobilization (Goodman ant] Schwartz, 1974~. Growth hormone appears to enhance the efficacy of other signals for lipolysis. Because energy metabolism is gov- erned by redundant control systems in the intact animal, compensatory adjustments that can be made when we disrupt the system may mask the actions of a hormone such as growth hormone that does not have very large effects in the short time span of

ROLE OF GROWTH HORMONE an experiment. The effects of growth hor- mone may be relatively small and slow to develop and dissipate, but even small changes can be quite meaningful over a long period of time. Fat is stored in adipose tissue in the form of triglycerides, which are synthesized con- tinuously from fatty acids, en c! alpha-glyc- erol phosphate, which is derived from glu- cose. Triglycerides, in turn, are broken down by an enzyme, the hormone-sensitive lipase, which is dependent on cyclic aclen- osine monophosphate (cyclic AMP) (Stein- berg and Huttunen, 1972) and stimulated primarily by epinephrine and to a lesser extent by a wide variety of other hormones. The activity of this enzyme is probably the major rate-determining factor in lipolysis and involves the splitting off of the first fatty acid molecule from the triglyceride. The cycle of lipolysis en c] esterification appears to be ongoing. Growth hormone can change the rate of fatty acid mobilization in two ways-either by accelerating lipolysis, which would make the cycle spin faster, or by slowing reesterification, which would in- crease the fraction of fatty acids escaping from the cell. As mentioned, growth hormone de- creases glucose utilization in fat. In orcler for fatty acids to be incorporated into tri- glyceride, alpha-glycerol phosphate must be present. The glycerol that is releaser! in lipolysis cannot be reutilized in adipose tissue, which is almost totally devoid] of the enzyme glycerol kinase (Margolis ant] Vaughan, 19621. Hence, all free glycerol produced by lipolysis escapes from the fat cell. Therefore, the rate of glycerol release can be used as an index of the rate of lipolysis. The fatty acids that are liberated in this process can be either recycled or released as FFAs. In fact, if adipose tissue were studied in vitro, it would be seen that only a very small fraction of the fatty acids that are released by lipolysis actually get out of the tissue. If there were no reester- ification, the ratio of FFAs to glycerol re 165 leased from the tissue ought to be 3:1. Actually, it is usually closer to 1:1, or perhaps less, suggesting that at least two- thirds of the fatty acids producer! by lipase activity are normally reconverted to triglyc- ericle. If reesterification were blocked, there would be potential for tripling the release of FFAs without changing the speed of the cycle. This can be accomplishes] just by limiting the rate of alpha-glycerophosphate production. Certainly this is one of the important effects that growth hormone has on adipose tissue, and it follows directly from limiting glucose metabolism.  Because of the reciprocal relationship between glucose and fatty acid metabolism, virtually anything that interferes with glu- cose metabolism is reflected] in increased fatty acid mobilization. Thus, how an ex- periment is conducted very much influences the results, and such variables as time of last feeding and amount fed may be crucial. This was made quite clear by Goodman and Knobil's (1959) studies on the effects of growth hormone on plasma FFAs in mon- keys. Growth hormone readily produced an increase in plasma concentrations of FFAs when given to animals that were accustomed to eating ad libitum until the time of hor- mone administration. When the same pro- tocol of giving growth hormone immediately upon removal of food was used with mon- keys that were accustomed to eating only one meal a day, no such effect was seen. Growth hormone increased FFAs in these animals only when given at the end of a 24- hour fast. It appeared that in these animals, which were accustomed to a nearly 24-hour interval between meals, removal of food at the time of hormone administration was not a sufficient stimulus to activate fasting re- sponses. Hormones acting at the surface of the adipocyte activate adenylate cycIase by a receptor-driven mechanism that is dis- cussed in more detail later. Adenylate cy- ciase catalyzes the conversion of adenosine triphosphate to cyclic AMP, which binds to

166 the cyclic AMP-dependent protein kinase enzyme complex and releases free catalytic units that catalyze the transfer of the ter- minal phosphate group of adenosine tri- phosphate to the lipase (Steinberg, 1976~. It appears that the hormone-sensitive lipase is an 84, OOO-dalton protein that is converted from an inactive to an active enzyme by phosphorylation of a single serine residue (Stralfors et al., 1984~. Presumably there is also a phosphatase that restores the enzyme to its inactive dephospho-form. This cycle appears to be responsible for all known hormone stimulation ofthe lipolytic process. The reaction is very rapid, and the phys- iologically important hormone that activates the lipase is epinephrine. The effects of growth hormone are presumably expressed through the same enzyme. Before cliscuss- ing growth hormone, however, the effects of epinephrine, which are typical of the other lipolytic hormones and therefore color expectations for the effects of growth hor- mone, should be examined. Birnbaum and Goodman (1977) incubated segments of adipose tissue from normal rats in bicarbonate buffer in the presence or absence of epinephrine. To obtain frequent measurements of glycerol production, the tissue segments were transferred to a fresh medium every 5 minutes. The amount of glycerol that was released into the medium during each of those 5-minute intervals was measured with a sensitive enzymatic assay. Again, glycerol production served as an indicator of how fast the lipolytic cycle was turning. Within just a few minutes, epi- nephrine increased glycerol production about fivefold. This effect persisted as long as the hormone was present, and dissipated within minutes after removing epinephrine. Growth hormone has been shown to be among the most potent hormones in causing an increase in FFAs in vivo and is the only pituitary hormone to produce such an effect (Goodman ant] Knobil, 1959~. Yet when growth hormone was added to adipose tissue in vitro, very little or no effect was seen. APPENDIX Initially, investigators looked for the same rapidly activated lipolysis resulting with epinephrine, or at least for some effect in the first hour of incubation a time when growth hormone has absolutely no lipolytic eject (Goodman and Schwartz, 1974~. Even when tissues were incubated for 3 or 4 hours, growth hormone clic] not do very much by itself. Studies by Fain ant] col- leagues expanded on some earlier findings of an apparent interaction of growth hor- mone and acirenal hormones (Fain et al., 1965) and produced the first convincing in vitro lipolytic results with growth hormone.  They showed that when adrenal glucocor- ticoid hormones were added along with growth hormone, a lipolytic effect of growth hormone was obtained but that the response hac] a built-in delay. Goodman ant] Knobil's (1959) in vivo studies fount] that the effects of growth hormone take a couple of hours to develop. In fact, if one allows enough time and examines lipolysis in the presence of some other agents, particularly glucocor- ticoids, lipolytic effects of growth hormone are obtaine(l with reasonable consistency. Goodman et al. (1986) transferred seg- ments of normal epididymal fat to a fresh medium every hour. Tissues were incubated with a small amount (0.1 ~g/ml) of clexa- methasone, a synthetic glucocorticoid, and 1 ~g/ml of bovine growth hormone. During the first hour, the rate of glycerol production in the absence of hormones and the rate in the presence of growth hormone and steroid were the same. No effect of the combination of growth hormone and dexamethasone was seen until the end of the second hour. The effect was initially small, but it gradually increased cluring the third and fourth hours, when it was relatively large. Neither growth hormone nor dexamethasone alone had any eject. The underlying mechanisms for the actions of glucocorticoid and growth hor- mone are not yet established. Their effects require the synthesis of new proteins and ribonucleic acid (Fain, 1967; Fain and Sap- erstein, 1970), but the nature of those pro

ROLE OF GROWTH HORMONE teins is not yet defined. Part of the effect of glucocorticoic3 may be mediatecl by in- cluction of an inhibitor of the activity of the enzyme phospholipase A2, which releases the arachidonic acid precursor of prosta- glanclins from membrane phospholipids (Flower and Blackwell, 1979~. It is unlikely, however, that this action can explain all the effects of glucocorticoids on adipose tissue. The effects of growth hormone may also involve protein synthesis, but the nature of the inducer] proteins is unknown. In a slightly different experimental situ- ation, Goodman (1968a) studied adipose tissue of hypophysectomized animals to cle- termine whether there was an absolute dependence on the steroid. Dexamethasone was replacer] with theophylline, which, at the time of these experiments, was thought to act solely by inhibiting cyclic nucleoticle phosphodiesterase and thereby allowing cyclic AMP to accumulate. It now appears that theophylline has at least one other effect: blocking the adenosine receptor (Londos et al., 1978), which may account for its lipolytic activity. Tissues were incu- batec! in Krebs Ringer bicarbonate buffer and transferred to a fresh medium every hour; theophylline (0.3 mg/ml) was always present. The lipolytic effect of growth hor- mone was seen only after a lag period of 1 hour. The standard errors were always about 10 percent at the mean, and the response to growth hormone was always statistically significant by the second hour after hormone addition. Curiously, the effect of growth hormone seen in the presence of theophyl- line was not blocked with inhibitors of RNA or protein synthesis (Goodman, 1968b). Using this model to study the reversibility of the lipolytic action of growth hormone, Goodman (1981) adcled neutralizing anti- bodies at various times after growth hor- mone and measured glycerol production each hour. In the control tissues, glycerol production was highest in the first hour ant! then declined very rapidly. In the presence of growth hormone, the initial rapid rate of ]67 lipolysis was sustained as long as the hor- mone was present. With the antiserum alone, or with growth hormone plus anti- serum abides! at zero time, there was a similar, rapic! decline in glycerol production after the first hour. When antiserum was added 1 or 2 hours after growth hormone, the high lipolytic rate was maintained for at least 1 hour and then decliner! to the same level as the control, whereas when growth hormone was addled without antiserum, the initial high lipolytic rate persisted through- out the experiment. These results provide a further illustration that actions of growth  hormone are slow in onset and dissipate slowly an(l, in this respect, are very different from the effects of epinephrine. Goodman et al. (1986) next investigated the concentration dependency of the lipo- lytic response by using bovine growth hor- mone prepares] by Dr. Martin Sonnenberg of the Memorial Sloan Kettering Institute in New York City. Tissues from normal rats were preincubated for 3 hours with dexa- methasone, and the various concentrations of growth hormone and lipolysis were meas- urecl in the fourth hour (Goodman and Grichting, 1983~. Significant effects were obtained with-3 ng/ml, but in many ex- periments significant effects were seen with 1 ng/ml, and sometimes a maximum effect was observed at around 10 nglml. This is an extremely sensitive response. The pro- tocol adopted, which takes into account the glycerol released only in the fourth hour, provides more sensitive conditions for show- ing the lipolytic effect than simply meas- uring glycerol released over the entire 4 hours. When glycerol release is measured over the entire 4 hours, the hormonal effect is partially obscured by the low rate of glycerol production cluring the rather long lag period. When only that narrow window of just the fourth hour is observed, when the response is largest, it is more likely that a lipolytic eject will be detected. The magnitude of the lipolytic effect of growth hormone was comparer] with that of

168 epinephrine (Good men ant] Grichting, 1983~. In this experiment, the concentration re- sponse range was narrow, and a maximum lipolytic effect of growth hormone was ob- tained with 3 ng/mI. Growth hormone in- creased glycerol production about twofold, from 1.5 to 4 Meg of tissue per hour, whereas 100 ng/ml of epinephrine, which is a submaximal concentration, increaser! glycerol production sixfoicI, to 9 ,uM/g per hour. The response could not be increased beyond 4 Meg per hour by adding more growth hormone, even though the tissues had ample capacity for a more rapid rate of lipolysis. Exposure of tissues of hypophysectom- ized rats to growth hormone in the presence of theophylline yields a similar concentra- tion/response relationship (Goodman et al., 1986~. In this case, the maximum response to growth hormone was seen at a concen- tration of about 1~30 ng/mI. The response was significant and almost maximal at 3 no/ ml. Once again, lipolysis was measured only in the fourth hour of incubation. In comparing the concentration of growth hormone needed for lipolysis to the con- centration of growth hormone circulating in rat blood, it is evident that maximum stim- ulation of lipolysis usually occurs at the low end of the range found in blood. The data of Tannenbaum et al. (1976) illustrate a peculiar ultradian secretory pattern in the rat, in which every 3.5 hours there is a burst of growth hormone secretion. The rat rarely has a growth hormone concentration lower than 50 ng/mI. Yet, a maximum li- polytic effect is often seen at around 10 ng/ ml. If the in vitro data are in any way representative of in viva events, it is difficult to see how growth hormone could be an activator or signal for increased fatty acid mobilization, because increaser] lipolysis is seen at concentrations that are as low or lower than the usually prevailing concen- trations in blood. It is likely, therefore, that growth hormone acts as a facilitator or potentiator of the effects of other agents, APPENDIX such as epinephrine, which are the primary signals for fatty acid mobilization. Growth hormone might act as a gain control, being a regulator only in the sense that it increases or decreases responsiveness to other signals. The effects of growth hormone on lipolysis are multiple. Goodman (1968a) investigated growth hormone in adipose tissue from normal rats and from hypophysectomized rats that were either untreated or given growth hormone for 2 days. In normal tissues, glycerol production was nearly dou- bled in 4 hours of incubation with growth hormone and dexamethasone. The control  tissues released less fatty acid than glycerol, instead of the theoretical threefold-greater amount of fatty acids. Most of the fatty acid that was formed was reconverted to triglyc- eride. In the presence of growth hormone and dexamethasone, the ratio of glycerol to fatty acid production decreased from about 4 to about 1.2. Thus, growth hormone and dexamethasone increased fatty acid mobi- lization in at least two ways: (1) by increasing glycerol production and (2) by decreasing the amount of fatty acids reconverted to triglyceride. In contrast to its effects in the presence of theophyIline, growth hormone had no effect on lipolysis when examined in the presence of dexamethasone in tissues of hypophysectomized rats. Treatment of the rats with growth hormone for 2 clays, but not 1 clay, enabled the tissues of hy- pophysectomized rats to respond when growth hormone and dexamethasone were later added in vitro. It appears that growth hormone has some long-term effect on the lipolytic system that takes days to develop. That effect is distinct from the shorter term stimulation of lipolysis, which requires an hour or two, and both, in turn, are (different from the lipolytic effects that growth hor- mone produces in tissues of hypophysec- tomized rats when theophylline is present. In conclusion, there are multiple effects of growth hormone on adipose tissue that are ultimately rejecter! in increaser! lipolysis. Also, it is evident that the effects of growth

ROLE OF GROWTH HORMONE hormone are expressed in increased fatty acid release as well as increased glycerol release. In adipose tissue obtained from hypophy- sectomized rats, the lipolytic response to epinephrine is severely curtailed (Good- man, 1970~. Hypophysectomy grossly de- creases the sensitivity of these tissues to virtually any lipolytic agent. The hypophy- sectomizec] rat obviously lacks more than just growth hormone. At least two other hormones that are related to pituitary se- cretions are also involves! in maintaining responsiveness of the lipolytic apparatus (Goodman, 1970~: thyroid! hormone and ad- renal glucocorticoicT. The ejects of growth hormone and dexamethasone on the re- sponse to epinephrine were examined by Goodman (1969~. Eight segments of adipose tissue were taken from each of eight hy- pophysectomized rats and preincubated for 3 hours. Two segments from each rat were incubates! without any hormone, two were incubated only with clexamethasone, two with only growth hormone, and two with a combination of growth hormone and dexa- methasone. The tissues were then trans- ferred to a fresh medium for incubation in the fourth hour in the presence or absence of a test close of 0.01 ~g/ml epinephrine. Tissues pretreater! for 3 hours with dexa- methasone produced an almost threefold increase in the response to the test dose of epinephrine. Preincubation with growth hormone alone had little or no eject, but when growth hormone was addec] along with clexamethasone in the 3-hour prein- cubation period, there was a significant increase above the response evoked by epinephrine in the presence of dexameth- asone alone. Thus, growth hormone clearly increaser! the lipolytic effects of another agent, ant] this response also required glu- cocorticoi(l. In an effort to pinpoint where within the lipolytic cycle growth hormone may be working, epinephrine was replaced in the previous protocol with clibutyryl cyclic AMP, 169 which is an analog of cyclic AMP that readily penetrates fat cells (Goodman, 1969~. Nei- ther clexamethasone nor growth hormone alone or in combination increaser] the li- polytic response to clibutyryl cyclic AMP. This suggests that the potentiating ejects of growth hormone and dexamethasone on lipolysis are more likely related to cyclic AMP formation than to cyclic AMP action. Receptor-mediated generation of cyclic AMP is complex, and there are many sites at which growth hormone might have an effect. It appears that in adipose tissue, and other tissues as well, the cyclic AMP-gen-  erating system is uncler the control of both stimulatory and inhibitory agents. Stimu- latory agents such as epinephrine act through beta-adrenergic receptors. Inhibitory agents include catecholamines (which might affect alpha-2 receptors in some species), the pros- taglandins, and adenosine. Both prostaglan- din and aclenosine seem to be formed in adipose tissue by endogenous mechanisms (Schwabe et al., 1973; Shaw and Ramwell, 1968~. Experimentally, it can be shown that the activity of adenylate cyclase under "rest- ing conditions" represents a balance cleter- mined by the combined influence of inhib- itory and stimulatory agents. What is caller! resting or basal activity actually represents the preponderance of inhibitory influences that keep the system shut clown (Kather et al., 1985). The prostaglandins seem to be important endogenous inhibitors of lipolysis and aclen- ylate cyciase (Steinberg et al., 1964~. Along with the recognition that one of the major effects of glucocorticoic] hormones in viva is to inhibit the release of arachidonic acid from phospholipids in cell membranes (Flower and Blackwell, 1979) arose the pos- sibility that clexamethasone may promote lipolysis by blocking prostaglanclin forma- tion. Arachi(lonate is the precursor for pros- taglanclins. Therefore, Goodman et al. (1986) attempted to (letermine whether indometh- acin, which is an inhibitor of the conversion of arachiclonic acid to prostaglandin (Vane,

170 1971), might mimic the effect of clexameth- asone in the lipolytic system described above. Using the same protocol, tissues were prein- cubatecl with 50 ~g/ml indomethacin, dex- amethasone, or growth hormone. Neither clexamethasone alone nor growth hormone alone had much effect on lipolysis. The combination of clexamethasone and growth hormone significantly increased lipolysis, as clid the combination of indomethacin and growth hormone. At least in this experi- ment, inclomethacin and dexamethasone seemed to have a similar effect, suggesting that at least part of the effect of dexameth- asone on lipolysis in response to epinephrine or growth hormone may be to block pros- taglanclin formation. This, in turn, somehow allows growth hormone, wherever it might be acting in the lipolytic system, to express its effects. Thus, when tissues were exposed to both agents, lipolysis was evident even though neither indomethacin nor growth hormone alone hac! much of a lipolytic effect. It is not certain that all the effects of dexamethasone can be explained in this way. The other prominent enclogenous inhib- itory agent in adipose tissue is adenosine, which is released from fat cells by the breakdown of cyclic AMP. Therefore, Good- man et al. (1986) used adenosine deaminase to eliminate the endogenous adenosine pro- cluced during the experiment. The effects of aclenosine are most clearly shown in isolates] adipocytes, rather than tissue seg- ments; but with at least some preparations of adenosine deaminase, these effects can be demonstratecl in tissue segments as well. Tissue segments rather than isolated cells were studied, largely because segments are easier to study and introduce fewer artifacts. Tissues were preincubated for 3 hours in the presence or absence of aclenosine de- aminase and hormones, and lipolysis was examined in the fourth hour. Acceleration of adenosine destruction increased glycerol production in a manner that is probably analogous to what has been seen with theo- phylline, which seems to block the adeno APPENDIX sine receptor (Londos et al., 1978~. Again, growth hormone alone had very little effect, but when added along with adenosine de- aminase, a substantial lipolytic effect was observed. Thus, it appears that there are two antilipolytic agents present in the tissue, at least in vitro, and that these agents contribute to the low basal activity of lipase. Presumably, they are also present in viva. During incubation in vitro, spontaneous production of prostaglandins and adenosine appear to inhibit lipolysis; growth hormone and glucocorticoids apparently relieve that inhibition.  To gain insight into how enclogenous inhibitors might interact with the cyclic AMP-generating system, the regulation of aclenylate cyclase should be looked at in more detail. The catalytic component re- sponsible for conversion of ATE to cyclic AMP relates to the receptors for stimulatory or inhibitory hormones by way of two other proteins, called G-proteins, because they bind guanine nucleotides (Spiegel et al., 1985~. For stimulatory input, the recogni- tion subunit of the receptor complex com- municates with the catalytic unit by way of the stimulatory guanine nucleotide binding protein (Gs), which somehow activates aden- ylate cyclase. Inhibitory effects appear to be mediated in a similar fashion through an inhibitory subunit (Gi). The inhibitory and stimulatory subunits can be examiner] by taking advantage of the fact that certain bacterial toxins specifically affect these sub- units. When plasma membranes are incu- bated in the presence of cholera toxin and 32P-labeled NAD (nicotinamide-aclenine di- nucleotide), there is a marked increase in the incorporation Of 32p into Gs, reflecting NAD ribosylation of the stimulatory sub- unit. Ribosylation of Gs in intact cells results in irreversible activation of adenylate cy- clase. Pertussis toxin catalyzes the NAD ribosylation of the inhibitory subunit, which irreversibly inactivates Gi and, in intact cells, blocks all inhibitory input to adenylate cyclase (Spiegel et al., 1985~.

ROLE OF GROWTH HORMONE The effects of pertussis toxin on lipolysis in normal adipose tissue were examined Goodman et al., 1986~. Tissue segments were preincubated with the toxin for 3 hours, and lipolysis was measured in the fourth hour. The intense lipolysis seen when the inhibitory influence was removed in the absence of an activator of adenylate cyclase substantiates the idea that aclenylate cyclase is under powerful inhibitory control under basal conditions. When that inhibitory con- trol is removed, activation of lipolysis is as profound as when a strong lipolytic agent is added. One possible site of action of growth hormone could! be on the linkage of the recognition subunits for either excitatory or inhibitory signals to the adenylate cyclase catalytic subunit. Incubating adipocyte plasma membranes with excess NAD, guan- osine triphosphate, and toxin (that is, con- clitions in which the subunit is limiting), gives some idea of whether these inhibitory or stimulatory guanine nucleoticle-bincling subunits are subject to change as a result of hormonal treatment. Membranes prepared from a(lipocytes of hypophysectomizec! or normal rats and from hypophysectomized rats treated with growth hormone 3 hours earlier were incubated with 32P-labeled NAD and cholera toxin or a mixture of cholera toxin and pertussis toxin (Goodman et al., 1986~. The membranes were then dissolved in sodium dodecyl sulfate and subjectec! to electrophoresis on slabs of polyacrylamide gel. NAD-ribosylated proteins were visu- alized by autoradiography. When cholera toxin was present alone, two bands with apparent molecular weights of about 45,000 ant] 53,000 daltons appeared] and were of about equal intensity regardless of whether the membranes were obtainer! from normal rats, hypophysectomizec] rats, or hypophy- sectomized rats treated with growth hor- mone. The bane] at 45,000 daltons is thought to be the alpha-subunit of the Gs protein. When pertussis toxin was present, another band (molecular weight, 41,000 claltons) 171 appeared that corresponds to the alpha- subunit of Gi. In tissues of hypophysectom- ized rats, the incorporation Of 32p indicative of NAD ribosylation of the inhibitory sub- unit was greatly increased. It is likely that in these tissues there is either more Gi or that it is in a form that is more susceptible to NAD ribosylation. Although these two possibilities cannot be separated at this time, there is clearly something different about the inhibitory subunit in adipocytes after hypophysectomy. Three hours after growth hormone treatment, the change was not restored to normal, but was at least  partly reversed. Similar results have been obtained in 8 or 10 experiments, and al- though the data are still preliminary, this is probably a very real phenomenon. The data suggest that growth hormone may affect the inhibitory subunit in a way that allows stimulatory inputs to produce greater changes in lipolysis or that sets the basal activity of adenylate cyclase at a higher level by re- ducing inhibitory input. REFERENCES Anselmino, K. J., and F. Hoffman. 1931. Das Fett- stoffwechselhormon des Hypophysenvorderlappens. I. Nachweis, Darstellung und Eifenschaften des Hormons. II. StufEwechselwirkungen und Regula- tionen des Hormons. Klin. Wochenschr. 10:2380. Armstrong, D. T., R. Steele, N. Altszuler, A. Dunn, J. S. Bishop, and R. C. DeBodo. 1961. Regulation of plasma free fatty acid turnover. Am. J. Physiol. 201:9. Astwood, E. B. 1955. Growth hormone and cortico- tropin. P. 235 in The Hormones, Vol. 3, G. Pincus and K. V. Thimann feds.). New York: Academic Press. Birnbaum, R. S., and H. M. Goodman. 1977. Studies of the mechanism of epinephrine stimulation of linolvsis. Biochem. Biophys. Acta 476:292. Burn, J. H., and H. W. Ling. 1929. The effect of pituitary extract and adrenalin on ketonuria and liver glycogen. J. Pharm. Pharmacol. 2:1. Burn, J. H., and H. W. Ling. 1930. Ketonuria in rats on a fat diet (a) after injections of pituitary (anterior lobe) extract, (b) during pregnancy. J. Physiol. (Lon- don) 69:x)x. DeBodo, R. C., and N. Altszuler. 1957. The metabolic effects of growth hormone and their physiological significance. Vitamins and Hormones 15:205.

172 Fain, J. N. 1967. Studies on the role of RNA and protein synthesis in the lipolytic action of growth hormone in isolated fat cells. Adv. Enzyme Reg. 5:39. Fain, J. N., and R. Saperstein. 1970. Involvement of RNA synthesis and cyclic AMP in the activation of fat cell lipolysis by growth hormone and glu- cocorticoids in adipose tissue. P. 20 in Adipose Tissue: Regulation and Metabolic Functions, B. Jeanrenaud and D. Hepp (eds.~. New York: Aca- demic Press. Fain, J. N., V. P. Kovacev, and R. O. Scow. 1965. Effect of growth hormone and dexamethasone on lipolysis and metabolism in isolated fat cells of the rat. J. Biol. Chem. 240:3522. Flower, R. J., and G. J. Blackwell. 1979. Antiinflam- matory steroids induce biosynthesis of a phospholi- pase A2 inhibitor which prevents prostaglandin gen- eration. Nature 278:456. Goodman, H. M. 1963. Effects of chronic growth hormone treatment on lipogenesis by rat adipose tissue. Endocrinology 72:95. Goodman, H. M. 1968a. Effects of growth hormone on the lipolytic response of adipose tissue to theo- phylline. Endocrinology 82:1027. Goodman, H. M. 1968b. Growth hormone and the metabolism of carbohydrate and lipid in adipose tissue. Ann. N.Y. Acad. Sci. 148:419. Goodman, H. M. 1969. Endocrine control of lipolysis. P. 115 in Progress in Endocrinology. Proceedings of the Third International Congress of Endocrinol- ogy, Mexico City, C. Gual (ed.~. Amsterdam: Ex- cerpta Medica Foundation. Goodman, H. M. 1970. Permissive effects of hormones on lipolysis. Endocrinology 86:1064. Goodman, H. M. 1981. Separation of early and late responses of adipose tissue to growth hormone. Endocrinology 109:120. Goodman, H. M., and G. Grichting. 1983. Growth hormone and lipolysis: A reevaluation. Endocrinol- ogy 113:1697. Goodman, H. M., and E. Knobil. 1959. Effects of fasting and of growth hormone on plasma fatty acid concentration in normal and hypophysectomized Rhesus monkeys. Endocrinology 65:451. Goodman, H. M., and J. Schwartz. 1974. Growth hormone and lipid metabolism. P. 211 in Handbook of Physiology: Endocrinology, Part 2, E. Knobil and W. Sawyer (eds.). Bethesda, Md.: American Phys- iological Society. Goodman, H. M., E. Gorin, and T. W. Honeyman. 1986. Biochemical basis for the lipolytic activity of growth hormone. In Perspectives in Growth Hor- mone Research, B. Sherman and L. Underwood (eds.). New York: Marcel Dekker. Kather, H., W. Bieger, G. Michel, K. Aktories, and K. H. Jakobs. 1985. Human fat cell lipolysis is primarily regulated by inhibitory modulators acting through distinct mechanisms. J. Clin. Invest. 76:1559. APPENDIX Ketterer, B., P. J. Randle, and F. G. Young. 1957. The pituitary growth hormone and metabolic proc- esses. Ergeb. Physiol. Biol. Chem. Expt. Pharmacol.  49:127. Lee, M. O., and N. K. Shaffer. 1934. Anterior pituitary growth hormone and the composition of growth. J. Nutr. 7:337. Li, C. H., M. E. Simpson, and H. M. Evans. 1949. Influence of growth and adrenocorticotropic hor- mones on the body composition of hypophysectom- ized rats. Endocrinology 44:71. Londos, C., D. M. F. Cooper, W. Schlegel, and M. Rodbell. 1978. Adenosine analogs inhibit adipocyte adenylate cyclase by a GTP-dependent process: Basis for action of adenosine and methylxanthines on cyclic AMP production and lipolysis. Proc. Natl. Acad. Sci. USA 75:5362. Margolis, S., and M. Vaughan. 1962. a-Glycenophos- phate synthesis and breakdown in adipose tissue. J. Biol. Chem. 237:44. Schwabe, U. R., R. Ebert, and H. G. Erbler. 1973. Adenosine release from isolated fat cells and its significance for the effects of hormones on cyclic 3',5'-AMP levels and lipolysis. Naunyn Schmiede- berg's Arch. Pharmacol. 276:133. Shaw, J. E., and P. W. Ramwell. 1968. Release of prostaglandin from epididymal fat pad on nervous and hormonal stimulation. J. Biol. Chem. 243:1498. Spiegel, A. M., P. Gierschik, M. A. Levine, and R. W. Downs, Jr. 1985. Clinical implications of guanine nucleotide-binding proteins as receptor-effector cou- plers. N. Engl. J. Med. 312:26. Steinberg, D. 1976. Interconvertible enzymes in adi- pose tissue regulated by cyclic AMP-dependent protein kinase. P. 157 in Advances in Cyclic Nu- cleotide Research, Vol. 7, P. Greengard, and G. A. Robinson (eds.). New York: Raven Press. Steinberg, D., and J. K. Huttunen. 1972. The role of cyclic AMP in activation of hormone-sensitive lipase in adipose tissue. P. 47 in Advances in Cyclic Nucleotide Research, Vol. I, P. Greengard, R. Paoletti, and G. A. Robinson (eds.). New York: Raven Press. Steinberg, D., M. Vaughan, P. Nestel, O. Strand, and S. Bergstrom. 1964. Effects of the prostaglandins on hormone-induced mobilization of free fatty acids. J. Clin. Invest. 43:1533. Stralfors, P., P. Bjorsell, and P. Belfrage. 1984. Hor- monal regulation of hormone-sensitive lipase in in- tact adipocytes: Identification of phosphorylated sites and effects of phosphorylation by lipolytic hormones and insulin. Proc. Natl. Acad. Sci. USA 81:3317. Tannenbaum, G. S., J. Martin, and E. Colle. 1976. Evidence for an endogenous ultradian rhythm gov- erning growth hormone secretion in the rat. Endo- crinology 99:720. Vane, J. R. 1971. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature (New Biol.) 231:232.

The Use of Bioassays To Detect and Isolate Protein or Peptide Factors Regulating Muscle Growth in Meat-Producing Animals WILLIAM R. DAYTON PEPTIDE FACTORS AFFECTING MUSCLE GROWTH Several peptide or protein factors that have the potential to regulate muscle growth in meat-producing animals have been iclen- tified. These are discussed below. Somatotropin The effect of somatotropin deficiency on muscle growth has been well established for many years. Long-term administration of somatotropin to pituitary-intact animals has been reporter] to increase muscling, decrease fat content, ant] improve feed efficiency in swine (Chung et al., 1985; Machlin, 1972~; increase nitrogen retention in steers (Moseley et al., 1982) and sheep (Davis et al., 1969~; increase growth rate in lambs (Wagner ant] Veenhuien, 1978~; and improve milk production in dairy cattle (Peel et al., 1981~. However, it appears unlikely that somatotropin directly affects proliferation and protein turnover in muscle cells. Although there is an increased incor- poration of 3H-thymicline into DNA in mus- cle from somatotropin-treated hypophysec tomizec] rats as compared to untreated controls (Breuer, 1969), this may reflect a direct effect of somatotropin on proliferation of nonmuscle cells or an indirect effect of somatotropin on proliferation of muscle cells. It has also been reported that in in vitro incubations of rat diaphragm muscle, 10-8M somatotropin stimulates amino acid uptake (Albertsson-Wikland and Isaksson, 19761. However, recent observations that many types of cells can secrete somatomedin (Ad- ams et al., 1984; Hill et al., 1986a) raise the possibility that responses seen in the intact diaphragm are the result of locally produced somatomedins. In fact, it is generally be- lieved that many if not all of the effects of somatotropin on muscle growth are me- diated through somatotropin-clependent plasma factors somatomedins produced in response to somatotropin. In culture, muscle cells do not appear to respond! to the addition of physiological levels of somatotropin. Ewton and Florini (1980) have reported that somatotropin has no detectable effect on anabolic processes in embryonic muscle cell cultures. Ad(li- tionally, Allen et al. (1983) have reported that somatotropin has no direct effect on 173

174 the rate of actin synthesis in myotube cul- tures derived from rat satellite cells. These findings support the theory that the eject of somatotropin on muscle is an indirect one mediated through the somatomedins. Insulin-Like Growth Factors (Somatomedins) Insulin-like growth factors are small poly- peptides (approximate molecular weight of 7,500 gallons) extracted and purified from human serum. They possess insulin-like properties in vitro but do not cross-react with insulin antibodies. Multiplication stim- ulating activity (MSA) is the name given to a family of polypeptides isolated from media conditioner! by a Buffalo rat liver (BRL) cell line (BRL 3A). To date, two classes of insulin-like growth factors (IGFs) have been characterized: IGF-I, also referrer] to as basic somatomedin (pH 8.~8.4), or soma- tomeclin-C (SM-C), and IGF-II, or neutral somatomedin. Multiplication stimulating activity appears to be the rat form of IGF- II, since the primary structure of M SA shows 93 percent identity with that of hu- man IGF-II (Marquarcit et al., 19814. At concentrations of 10-9 to 10-~°M, IGFs are mitogenic for a variety of cultured cell types. Biologically active receptors for both IGF- I/SM-C ant] IGF-II/MSA have been iden- tified on the surface of cultured muscle cells (Ballard et al., 1986~. IGF-I/SM-C has been shown to stimulate growth of hypophysec- tomized rats (Schoenle et al., 1982), prolif- eration of cultured myoblasts (Ballard et al., 1986), amino acid uptake in cultured myob- lasts (Hill et al., 1986a), differentiation of cultured myoblasts (Ewton and Florini, 1981), and RNA synthesis and polypepticle chain initiation in an isolated muscle (Monier and Le Marchand-Brustel, 1984~. IGF-II/MSA has been shown to stimulate proliferation of cultured myoblasts (Ewton and Florini, 1981; Florini and Ewton, 1981; Florini et al., 1984), amino acid transport into cultured muscle cells (Janeczko and Etlinger, 1984), APPENDIX and the rate of protein synthesis in cultured myotubes (.Janeczko and Etlinger, 1984~. MSA has also been shown to decrease the rate of protein degradation in cultured my- otubes (Janeczko ant] Etlinger, 19844. In addition to their well-documented presence in serum, both IGF-I/SM-C and IGF-II/ MSA have been reported to be released by rat myoblasts (Hill et al., 1986b), thus raising the possibility that these peptides may be involved in autocrine or paracrine regulation of muscle growth. On the basis of this information, it appears likely that insulin- like growth factors are potent stimulators of all aspects of muscle growth and clevelop-  ment. Insulin The role of insulin in regulating general cell metabolism has been recognized for many years, but its mechanism of action is still not well understood. Similarly, its role in controlling muscle growth is not clear. Several lines of evidence suggest that insulin may have an anabolic effect on muscle tissue. Studies of a variety of animal models have demonstrated that wasting of skeletal muscle is a prominent feature of diabetes mellitus and that it is reversed by admin- istration of insulin (Pain and Garlick, 1974~. Aclclitionally, ribosomes isolated from mus- cle of diabetic rats are less active in in vitro protein synthesis systems than in ribosomes from nondiabetic controls. In vitro stu(lies with isolates! muscles (Fulks et al., 1975) and the perfused rat hemicorpus Ue~erson et al., 1977) have shown that insulin in- creases the rate of protein synthesis and decreases the rate of protein degradation in these systems. In cultured muscle cells as well as in fibroblasts and fibroblastic cell lines, supra- physiological concentrations of insulin (21 ,u~g/ml) are required to elicit a maximum response. In muscle cell cultures, these high concentrations stimulate both prolif- eration and differentiation of myogenic cells

BIOASSAYS (Ewton ant] Florini, 1981~. Insulin at high concentrations (10-6M) is a component of synthetic media used to support growth and differentiation of myogenic cells in culture (Dollenmeier et al., 1981; Florini and Rob- erts, 1979~. It has been proposed that the stimulation of growth of fibroblasts by in- sulin is mediated by insulin's weak binding to receptors for insulin-like growth factors. Affinity cross-linking studies have shown the existence of two classes of IGF recep- tors. Type I receptors (Massague and Czech, 1982) have a higher affinity for IGF-I than for IGF-II and a low affinity for insulin. The structure and subunit composition of type I receptors are very similar to those of the insulin receptor. Type II receptors bind IGF-II with a higher affinity than they do IGF-I and do not appear to have appreciable affinity for insulin (Massague ant! Czech, 1982~. At high concentrations, insulin may bind to the type I receptor, and in so doing affect cell growth in a manner similar to that observed for much lower concentrations of IGF-I. This hypothesis is based on work by King et al. (1980), who showed that blockade of high-affinity insulin receptors with anti-receptor Fab fragments blocked high-affiinity insulin binding but did not prevent insulin-induced stimulation of DNA synthesis in cultured fibroblasts. Further- more, these investigators showed that anti- insulin-receptor immunogIobulin G (IgG), which triggers a number of acute insulin- like metabolic effects, floes not stimulate DNA synthesis. They concluded that the growth-promoting effects of insulin on hu- man fibroblast were clue to binding of insulin to the type I receptor. Although this has not been prover! in cultured muscle cells, it would seem likely that the well-clocu- mented effects of supraphysiological con- centrations of insulin on proliferation and differentiation of cultures] muscle cells are the result of this spillover action of insulin through IGF-I receptors. Insulin has a wide range of effects on cell metabolism. Consequently, it is possible 175 that physiological levels of insulin facilitate muscle cell growth by maintaining cells in a metabolic state that allows them to respond] to other hormones and growth factors that stimulate cell proliferation. Differentiation Inhibitor Coon's BRL cells secrete a protein that is a potent inhibitor of skeletal myoblast differentiation in vitro (Evinger-Hodges et al., 1982; Florini et al., 1984~. In skeletal myoblast cultures, this protein reversibly blocks fusion, elevates creatine kinase, and increases binding of alpha-bungarotoxin. It  has also been isolates] from sera of embry- onic origin, prompting the suggestion that it may play a role in embryonic growth of myoblasts and in satellite cell formation (Evinger-Hociges et al., 1982~. Transferrin Tr ransterr~n Is an iron-bin(ling glycopro- tein that is present in serum (Ozawa and Kohama, 1978) and embryo extract (Ii et al., 1981~. Additionally, transferrin-like mol- ecules have been isolated from both nerve and muscle extracts (Matsuda et al., 1984~. In muscle cell cultures, iron-saturated trans- ferrin stimulates both proliferation and dif- ferentiation and is essential for maintenance of healthy myotubes. The effect of transfer- rin on muscle growth in culture is absolutely dependent on the presence of iron and appears to be class specific (that is, mam- malian transferring do not affect avian myo- blasts, nor do avian transferring affect mam- malian myoblasts) (Shimo-Oka et al., 1986~. Fibroblast Growth Factor In cell cultures, fibroblast growth factor (FGF) stimulates proliferation of myogenic cells and delays their clifferentiation (Gos- podarowicz et al., 1976; Linkhart et al., 1981~. Allen et al. (1984) have proposed that FGF regulates satellite cell proliferation in

176 skeletal muscle. However, they do not be- lieve that serum is the source of the FGF that is affecting satellite cell proliferation. Rather, they hypothesize that FGF-like molecules are producer! locally in muscle and trigger a localizes! response of satellite cells cluring muscle regeneration. Paracrine and Autocrine Control of Muscle Growth Reports that various cell types secrete growth factors have sparked interest in au- tocrine and paracrine regulation of muscle growth. It has been reported that cultured fibroblasts secrete IGF or IGF-like mole- cules (Adams et al., 1984) ant] that fetal rat chondrocytes sequentially elaborate sepa- rate growth- ant! clifferentiation-promoting peptides cluring their development (Shen et al., 1985~. Aclditionally, cultured myo- blasts have been reported to synthesize and secrete IGF-I/SM-C (Hill et al., 1986a). Because all these cell types are fount! in muscle tissue, their ability to produce growth factors raises the possibility that muscle growth may be at least partially regulated by factors procluced locally. This hypothesis is supported by reports of the purification of an FGF-like muscle growth factor present in skeletal muscle tissue (Karkami et al., 1985~. The mechanism by which this factor is accumulated in skeletal muscle and the relationship of this accumulation to regula- tion of muscle growth and regeneration is of interest. BIOASSAYS FOR FACTORS INFLUENCING MUSCLE GROWTH To develop effective strategies for con- trolling animal growth, a better understand- ing is needed of the mechanism by which known growth factors regulate proliferation, (differentiation, and protein turnover in mus- cle cells. The potential for autocrine and paracrine regulation of muscle growth, as well as the discovery of factors such as the APPENDIX differentiation inhibitor, emphasize the im- portance of efforts to isolate currently un- known peptide factors that significantly in- fluence the development of muscle tissue. In adclition to mitogenic growth factors, factors that inhibit the growth of cells have been reported (Hare] et al., 1985; Harring- ton ant! Gociman, 1980; Salmon et al., 19831. Although these factors have not been well characterized, it seems reasonable to as- sume that they modulate the growth-pro- moting effects of mitogenic serum factors such as the IGFs. In fact, both specific and nonspecific inhibitors of IGF action have been reported (Kuffer ancl lIerington, 1984; Salmon et al., 1983~. Although these inhib-  itors have been detected in normal sera (Kuffer ant! Herington, 1984), their level and activity appear to be increased by cat- abolic conditions in both humans and ex- perimental animals (Salmon et al., 1983; Unterman and Phillips, 19851. Under the proper conditions, transforming growth fac- tor-,B (TGF-'B) has also been shown to inhibit proliferation of certain types of cultured cells (Roberts et al., 1985~. Because these inhibitory factors appear to have the poten- tial to attenuate the action of growth-pro- moting factors, it is important that more is learned about their mode of action and physiological significance in meat-pro(lucing animals. Radioimmunoassays (RIAs) cannot be used effectively to detect and characterize un- known or poorly characterized muscle growth factors. Consequently, bioassays capable of reliably detecting factors influencing muscle growth are necessary. These bioassays will augment existing RIAs by enabling us to detect ant] study currently unknown factors that may stimulate or inhibit muscle growth in meat-producing animals. The current lack of understan(ling of the mechanisms con- trolling muscle growth in meat animals is largely the result of (lifficulties encountered in devising a satisfactory bioassay system in which to study these processes. Experi- mental animals, isolated muscles, and mus

BIOASSAYS cle cell culture have been the primary systems used to study the effects of specific peptides on the growth of muscle tissue. While experimental animals provide the most biologically complete system in which to study muscle growth, the complex inter- actions of their hormonal systems and large animal-to-animal variation often make it difficult to evaluate the role of any specific factor in muscle growth. Additionally, ex- periments with animals are expensive and labor intensive and often require several weeks or months to complete. In order to evaluate the effect of a specific factor on muscle growth, it is also necessary to meas- ure the muscle mass of control and experi- mental animals. At present, this is a labo- rious and inaccurate procedure. In vitro incubation of excised muscle tissue has also been used to study the effects of various peptides on muscle growth, pri- marily the influence of different substances on the rates of protein synthesis and deg- radation in skeletal muscle tissue (Fulks et al., 1975~. This technique provides a more controlled experimental environment and easier measurement of protein synthesis and degradation rates than does the whole ani- mal. However, excised muscles are gener- ally in a catabolic state relative to protein turnover (for example, protein degradation exceeds protein synthesis) (Clark and Mitch, 1983; Fulks et al., 1975~. Muscle cell culture has been used exten- sively to study the effects of specific DeDtides on both protein turnover and muscle cell proliferation. In culture, muscle precursor cells differentiate and proliferate to form myoblasts that fuse to form multinucleated myotubes. Myotubes synthesize contractile proteins, assemble them into myofibrils, and develop the ability to contract. How- ever, for these processes to occur, the culture media must contain blood serum or serum factors. Presumably, serum contains specific factors that are necessary for the differentiation and proliferation of muscle cells in culture. Consequently, muscle cell 177 culture has been used to study the effect of specific factors on proliferation, protein turnover, and differentiation in muscle cells. Although cell culture lends itself well to these kinds of studies, there is some concern about whether the findings are valid for muscle tissue in viva. Therefore, cell culture data must ultimately be confirmed in the animal. EFFECT OF PORCINE GROWTH HORMONE ON BIOACTIVITY AND IGF-I CONCENTRATION IN SWINE SERUM  Although all the systems discussed in the preceding section may be useful as bioassays under the proper circumstances, my col- leagues and I have focused our efforts on developing and statistically standardizing a muscle cell culture bioassay that can be used to identify factors influencing muscle growth and to determine their mode of action in meat animals. This muscle cell culture bioassay and an IGF-I radioimmu- noassay have been used to measure the bioactivity and IGF-I concentration, re- spectively, in sera obtained from pigs before and after injection with porcine growth hormone (pGH). Although there have been conflicting re- ports about the effect of exogenous growth hormone (GH) on muscle growth in pitui- tary-intact swine, it now appears that long- term injection of highly purified pGH in- creases muscling, decreases fat, and im- proves feed efficiency in growing pigs (Chung et al., 1985; Machlin, 1972). However, very little is known about the mechanism through which pGH affects muscle deposition in pituitary-intact swine. Although it appears likely that the GH-induced increases in the circulating level of somatomedin-C may be responsible for increased muscle deposition, little information is available on the effect of artificially increased growth hormone lev- els on the concentration and bioactivity of somatomedins and other growth factors whose

178 levels might be affected by this increase. Comparison of the muscle cell culture bioas- say response and the raclioimmunoassayable IGF-I concentration of sera obtained from pigs before and after pGH injection should help determine whether IGF is uniquely responsible for increases in muscle growth resulting from growth hormone treatment. Methods Standardized bioassays for measuring the elect of porcine serum on proliferation in cultured LO muscle cells were done accord- ing to procedures described in detail by Kotts et al. (1987a,b). Briefly, L6 cells were plated at 600/cm2 (25-cm2 flasks) in Dulbec- co's modified Eagle's medium (DMEM) containing 10 percent fetal calf serum. After 24 hours of attachment, the medium was removed and the cells were rinser! with 37°C DMEM without serum (SF media). Test media were applied ant] cells were incubated for 72 hours. The cells were removed for counting by trypsinization for 5 minutes at 37°C, and the reaction was stopper] by adding ice-col(1 DMEM con- taining 10 percent fetal calf serum. Cells from each flask were quantitatively trans- ferred to glass tubes on ice. The contents of each tube were diluted and counted in triplicate, and the counts were averaged. Triplicate flasks were assayed for each serum sample tested, ant! the results were ex- pressed as the mean number of cells/cm2 per flask + standard error. The intraassay coefficient of variation was 2.6 percent (Kotts et al., 1987b). Test media consisted of DMEM containing 3 percent (volume/volume) test sera. Porcine growth hormone was purchased from Dr. A. F. Parlow (Torrance, Calif.~. The pGH used for injection was lot no. 7024-C (specific activity = 1.5 U/mg) and that used for radioimmunoassay standard was lot no. APE 6400. i25I-pGH and rabbit anti-bovine GH were supplier! by Monsanto Company (St. Louis, Dog. Crossbred bar APPENDIX rows (19 to 36 kg) from separate litters were individually penner! and fee! ad libitum a corn- and soybean-basec] diet containing 21 percent protein. Five pigs were injected with 143 log of pGH/kg of body weight per clay for 3 clays. Catheters were inserted into both jugular veins, and after a 2-day recovery period, 12-ml blood samples were removed from the catheters at 6-hour intervals (6 a.m., noon, 6 p. m., and midnight) throughout the duration of the study. Injections of pGH were given at 2 p.m. on days 4 through 6. On days 1 through 3 and 7 through 9, all pigs received sham injections containing  sterile saline. Injection and postinjection blooc! samples were collected on clays 4 through 9. The blood was allowed to clot, and serum was prepared for use in the muscle cell culture bioassays and radioim munoassays. Solutions of pGH for injection were pre- pared by dissolving the pGH in 44 mM NaHCO3, pH 11.5, and then immediately lowering the pH to 9.5 by addition of 1 N HC1. Solutions were prepared on the day of the first injections and filtered through a 0.22-,um filter. Protein content of the fil- tered solution was determined by the mi- crobiuret method. The basic electrophoresis system used for analytical sodium clodecy} sulfate (SDS) polyacrylamide slab gels was that of Laem- mli (1970) and consisted of a 3.5 percent acrylamide stacking gel ant! a 12 percent separating gel. Radioimmunoassays were done on the individual 6-hour serum samples obtained from each pig during the study. Radioim- munoassay kits from Micromedic Systems (Horsham, Pa.) were user! to quantify the levels of insulin and cortisol in the sera. The insulin kit was a homologous RIA for porcine insulin and used rabbit anti-porcine insulin antisera. The cortisol kit used rabbit anti-cortisol sera. A heterologous radioimmunoassay for porcine growth hormone was used to quan

BIOASSAYS tify levels of pGH in the sera. This raclioim- munoassay used pGH (pituitary; lot AFP 6400) as a standard, i25I-pGH as a trace, and rabbit anti-bovine growth hormone anti- sera. The sensitivity of the assay at 95 percent binding was 5 ng/ml. Serial dilution of porcine serum at 100, 150, 200, and 250 ~1 yielded a curve that was parallel to the pGH standard curve. Recovery of standard in the presence of 200 ~1 of serum was 98.6 percent. The intraassay variability was 2.95 percent, and the interassay variability was 9.8 percent. All samples compared to each other in this work were assayed in the same experiment to avoic] interassay variation. Somatomedin-C levels in serum were quantified with a kit from the Nicholls Institute (San Juan Capistrano, Calif. ). Sera were treated in 1 M glycine-glycine HC1 buffer (pH 3.5) for 24 hours at 37°C prior to assay. All sera were measured against a human serum SM-C standard (1 U = 36 ng of purified SM-C). The trace was i25I-human SM-C; rabbit anti-human somatomedin-C antisera were used. The intraassay variabil- ity was 5.4 percent, and the interassay variability was 9.2 percent. When acidified swine serum was assayed in the presence of ~25I-human SM-C standard, 100 percent recovery was achieved. A titration of various dilutions (1:4 to 1:20) of swine sera resulted in curves parallel to those obtained with purified SM-C. To verify that the observed increases in mitogenic activity resulted from the oGH injections ant] were not random daily vari- ations in serum activity, the data obtained from the bioassay were subjected to analysis of variance. A randomized block design was used, with blocks representing pigs. To test for differences owing to pGH injection, the bioassay results from the preinjection days (1 through 3) were compared to those during (days 4 through 6) and after (days 7 through 9) injection by using the single degree of freedom contrasts on treatments. 179 Results and Discussion SDS-polyacrylamicle gel electrophoresis of the pGH preparation used in this study shower! a major band at 21.9 kilodaltons (Al) en cl a minor band at 20 kit, along with several minor hands between 15 anal 9 kd. The molecular weights of the 21.9- and 20- k] bands correspond to those reported for human growth hormone (Chambach et al., 1973~. The peptides banding between 9 and 15 k(l may be proteolytic fragments of pGH, or they may be impurities in the prepara- tion. Whatever their origin, any single one of these peptides represents an extremely minor contaminant in the pGH preparation. To determine whether the pGH prepa-  ration contained contaminants that affected muscle cell proliferation, it was added at various concentrations to media containing 2.5 percent (volume/volume) control swine serum (CSS). Radioimmunoassay of the CSS showed that it contained 5.56 ng of pGH/ m} and 2.18 U of SM-C/mI. Consequently, the contribution of the CSS to the final pGH or SM-C level in the bioassay was 6 x 10-12 M pGH and 2.58 x 10-1° M SM- C (based on a molecular weight of 7.6 kd and 36.1 ng of human SM-C/U and a mo- lecular weight of 22 Ed for pGH). The proliferation rate of cultured muscle cells was not significantly affected by pGH con- centrations below 10-8 M, but 10-8 M pGH or higher resulted in a slight, though sig- nificant, increase in cell numbers (10 to 12 percent above control levels). The inability of pGH to stimulate proliferation of cultured muscle cells is in agreement with results obtained by others using primary myogenic cultures or L6 myogenic cells (Ewton and Florini, 1980; Gospodarowicz et al., 1976~. The slight stimulation of proliferation ob- served at higher pGH concentrations (~10-8 M) is consistent with the stimulation of alpha-aminoisobutyric acid uptake in 8-day- old cultures of L6 myotubes exposed to 10 - 7

180 M bovine GH (Ewton and Florini, 1980~. It is possible that impurities in the GH preparation or biologically active fragments of the GH molecule (Liberti and Miller, 1978) are responsible for these increases in mitogenic activity observed at supraphy- siological concentrations of GH. In contrast to the lack of response ob- served when pGH was aciclec3 directly to muscle cells, sera from four out of five pigs injected with pGH exhibiter] increased mi- togenic activity. Analysis of variance on the bioassay data from all five pigs showed that the treatment elects were highly significant (P < 0.005~. The single degree of freedom contrasts on treatment revealer! that the mitogenic activity of sera obtained during and after the pGH injections was signifi- cantly higher (P < 0.005) than preinjection levels. Aciclitionally, all pigs receiving pGH showed increases in SM-C levels in their sera during and after the injections. The pGH concentration in the 24-hour pooled serum samples from the pigs on pGH injection days (clays 4 through 6) was approximately 100 ng/ml, and these pools were diluted 29-fold for use in the prolif- eration bioassay. Thus, the maximum con- centration of pGH in the bioassay media was 10-~° M. Since 10-~° M pGH had no elect on proliferation when added directly to the muscle cell cultures, the increases in bioassayable mitogenic activity of serum pools obtained cluring ant] after pGH injec- tion were not a direct result of the increased level of pGH in the culture media. Serum pGH levels were increased ap- proximately 30-folcl by 4 hours after each pGH injection and declined to preinjection levels by approximately 16 hours after each injection. Increases in serum SM-C levels were observed 6 to 12 hours after the increase in serum pGH concentration (10 to 16 hours after each pGH injection). The magnitude of the SM-C response was dif- ferent for each pig, even though all pigs received the same dose of pGH and attained similar blood levels of pGH 4 hours after APPENDIX injection. SM-C increases ranged from 1.7 to 4 times the preinjection levels. In all the pigs, the second and third! injections re- sulted in higher concentrations of serum SM-C than the first injection. In two cases, SM-C concentrations appeared to increase in a stepwise manner with each successive injection of pGH. A similar stepwise in- crease in SM-C production upon successive injections of human growth hormone into hypopituitary patients was reported by Copeland et al. (1980~. Serum SM-C levels remained high for 2 to 6 days after the last pGH injection.  Insulin en c! cortiso} levels in the sera dicI not change during the treatment period ant! ranged from 3.8 to 10.6 ~U/ml and 2.0 to 6.9 ~g/~l, respectively. It is well established that GH stimulates the production of somatomedins (IGFs) by the liver and possibly by other tissues as well. Administration of IGF-I/SM-C to hy- pophysectomized rats has been reported to restore growth to a level equivalent to that seen with GH replacement (SchoenIe et al., 1982~. Additionally, IGF-I/SM-C and IGF- II/MSA stimulate the proliferation of my- ogenic cells in culture (Ballard et al., 1986; Ewton en cl Florini, 1981; Florini et al., 1984; Hill et al., 1986a). Consequently, it appears likely that the increased levels of IGF-I/SM-C observed in sera obtained from pigs cluring and after pGH injection play a role in the increased mitogenic activity of these sera. Nonetheless, there were several instances when changes in serum IGF-I/ SM-C levels did not appear to be directly related to changes in serum mitogenic ac- tivity in the bioassay. For example, sera from pig 90 showed a significant increase in SM-C concentration cluring and after pGH injection (2.5 U/ml preinjection to 6.5 U./ ml postinjection); however, no correspond- ing increase in serum mitogenic activity was detectable. In contrast, sera from pig 85 exhibited a similar change in serum SM-C concentration during and after pGH injec- tion (2 U/m! preinjection to 7 U/m} postin

BIOASSAYS jection), and this corresponded to a signif- icant increase in mitogenic activity. In acI- dition, sera from pig 87 exhibiter! a relatively large increase in SM-C concentration (3.5 U/m! preinjection to 10 to 13 U/ml postin- jection) but shower] only a modest increase in mitogenic activity. Conversely, sera from pig 7, which exhibited relatively little in- crease in SM-C concentration (2 U/ml prein- jection to 4 to 5.5 U/ml postinjection), showed a relatively large increase in mito- genic activity (24 percent) over the injection - period. These results suggest that factors in acI- dition to radioimmunoassayable IGF-I/SM- C may contribute to the alterations in mi- togenic activity observed in sera during and after pGH injection. There are several fac- tors that could! be involved in the mitogenic response, either by directly affecting muscle cell proliferation or by modulating the bioac- tivity of IGF-I. For example, IGF-II has been reported to increase fourfold in the sera of GH-deficient humans after GH administration (Schalch et al., 1982~. Ad- clitionally, inhibitors of IGF-stimulated syn- thesis of DNA anchor sulfate incorporation in costar cartilage have been reported in sera from starved, diabetic, or hypophysec- tomized rats (Kuffer and Herington, 1984; Salmon et al., 1983; Unterman and Phillips, 1985), and a specific inhibitor of IGF has been isolated ant! partially purified from normal sera (Kuffer and Herington, 1984~. Somatomeclin-binding proteins ranging in molecular weight from 40 to 70 kit have also been reported to bind and inactivate IGF (Hossenlopp et al., 1986; Martin and Baxter, 1985; Romanus et al., 1986~. In addition, a protein that inhibits differentiation of my- ogenic cells has been identified in fetal calf serum and in merlin obtained from BRL cells in culture (Evinger-Hodges et al., 1982; Florini et al., 1984~. It is possible that these factors or other, as yet unidentified, factors are affecting the mitogenic activity of sera in the muscle cell culture bioassay used in this study. 181 Results of this study demonstrate the importance of developing bioassays for mus- cle growth. Used in conjunction with ra- dioimmunoassays, bioassays can help elu- cidate the mode of action of known growth factors such as somatotropin. They also provide a valuable tool for use in identifying unknown growth factors that affect muscle growth in meat animals. Identification of these factors ant] clarification of their mode of action is crucial to an eventual under- stancling of the biological control of muscle growth.  REFERENCES Adams, S. O., M. Kapadia, B. Mills, and W. H. Daughaday. 1984. Release of insulin-like growth factors and binding protein activity into serum-free medium of cultured human fibroblasts. Endocrinol- ogy 115:520. Albertsson-Wikland, K., and 0. Isaksson. 1976. De- velopment of responsiveness of young normal rats to growth hormone. Metabolism 25:747. Allen, R., K. C. Masak, P. K. McAllister, and R. A. Merkel. 1983. Effects of growth hormone, testos- terone and serum concentration on actin synthesis in cultured satellite cells. J. Anim. Sci. 56:833. Allen, R. E., M. V. Dodson, and L. S. Luiten. 1984. Regulation of skeletal muscle satellite cell prolifer- ation by bovine pituitary fibroblast growth factor. Exp. Cell Res. 152:154. Ballard, F. J., L. C. Read, G. L. Francis, C. J. Bagley, and J. C. Wallace. 1986. Binding properties and biological potencies of insulin-like growth factors in L6 myoblasts. Biochem. J. 233:223. Breuer, C. B. 1969. Stimulation of DNA synthesis in cartilage of hypophysectomized rats by native and modified placental lactogen and anabolic hormones. Endocrinology 85:989. Chambach, A., R. A. Yadley, M. Ben-David, and D. Rodbard. 1973. Characterization of human growth hormone by electrophoresis and isoelectric focusing in polyacrylamide gel. Endocrinology 93:848. Chung, C. S., T. D. Etherton, and J. P. Wiggins. 1985. Stimulation of swine growth by porcine growth hormone. J. Anim. Sci. 60:118. Clark, A. S., and W. E. Mitch. 1983. Comparison of protein synthesis and degradation in incubated and perfused muscle. Biochem. J. 212:649. Copeland, K. C., L. E. Underwood, and J. J. Van Wyk. 1980. Induction of immunoreactive somato- medin-C in human serum by growth hormone: Dose response relationships and effects on chromato- graphic profiles. J. Clin. Endocrinol. Metab. 50:690.

182 Davis, S. L., U. S. Garrigus, and F. C. Hinds. 1969. Metabolic effects of growth hormone and diethyl- stilbestrol in lambs. II. Effects of daily ovine growth hormone injections in plasma metabolites and nitro- gen-retention in fed lambs. J. Anim. Sci. 30:236. Dollenmeier, P., D. C. Turner, and H. M. Eppen- berger. 1981. Proliferation of chick skeletal muscle cells cultured in a chemically defined medium. Exp. Cell Res. 135:47. Evinger-Hodges, M. J., D. Ewton, Z. S. C. Seifert, and J. R. Florini. 1982. Inhibition of myoblast differentiation in vitro by a protein isolated from liver cell medium. J. Cell Biol. 93:395. Ewton, D. Z., and J. R. Florini. 1980. Relative effects of the somatomedins, MSA and growth hormone on myoblasts and myotubes in culture. Endocrinology 106:577. Ewton, D. Z., and J. R. Florini. 1981. Effects of the somatomedins and insulin on myoblast differentia- tion in vitro. Dev. Biol. 86:31. Florini, J. R., and D. Z. Ewton. 1981. Insulin acts as a somatomedin analog in stimulating myoblast growth in serum-free medium. In Vitro 17:763. Florini, J. R., and S. B. Roberts. 1979. A serum-free medium for the growth of muscle cells in culture. In Vitro 15:983. Florini, J. R., D. Z. Ewton, M. J. Evinger-Hodges, S. L. Fallen, R. L. Lau, J. F. Ragan, and B. M. Vertel. 1984. Stimulation and inhibition of myoblast differentiation by hormones. In Vitro 20:942. Fulks, R. M., J. B. Li, and A. L. Goldberg. 1975. Effects of insulin, glucose and amino acids on protein turnover in rat diaphragm. J. Biol. Chem. 250:290. Gospodarowicz, D., J. Weseman, J. S. Moran, and J. Lindstrom. 1976. Effect of fibroblast growth factor on the division and fusion of bovine myoblasts. J. Cell Biol. 70:395. Harel, L., C. Blat, and G. Chatelain. 1985. Regulation of cell proliferation inhibitory and stimulatory factors diffused by 3T3 cultured cells. J. Cell. Physiol. 123:139. lIarrington, W. N., and G. C. Godman. 1980. A selective inhibitor of cell proliferation from normal serum. Proc. Natl. Acad. Sci. USA 77:423. Hill, D. J., C. J. Crace, S. P. Nissley, D. Morrell, A. T. Holder, and R. D. G. Milner. 1986a. Fetal rat myoblasts release both rat somatomedin-C (SM-C)/ insulin-like growth factor I (IGF I) and multiplica- tion-stimulating activity in vitro: Partial characteri- zation and biological activity of myoblast-derived SM-C/IGF I. Endocrinology 117:2061. Hill, D. J., C. J. Crace, A. J. Strain, and R. D. G. Milner. 1986b. Regulation of amino acid uptake and deoxyribonucleic acid synthesis in isolated human fetal fibroblasts and myoblasts: Effect of human placental lactogen, somatomedin-C, multiplication stimulating activity, and insulin. J. Clin. Endocrinol. Metab. 62:753. APPENDIX Hossenlopp, P., D. Seurin, B. Segoria-Quinson, S. Hardouin, and M. Binoux. 1986. Analysis of serum insulin-like growth factor binding proteins using Western blotting: Use of the method for titration of  the binding proteins and competitive binding stud- ies. Anal. Biochem. 154:138. Ii, I., I. Kimura, T. Hasegawa, and E. Ozawa. 1981. Transferrin is an essential component of chick embryo extract for avian myogenic cell growth in vitro. Proc. Jpn. Acad. 57:211. Janeczko, R. A., and J. D. Etlinger. 1984. Inhibition of intracellular proteolysis in muscle cultures by multiplication-stimulating activity. Comparison of effects of multiplication-stimulating activity and in- sulin on proteolysis, protein synthesis, amino acid uptake, and sugar transport. J. Biol. Chem. 259:6292. Jefferson, L. S., J. B. Li, and S. R. Rannels. 1977. Regulation by insulin of amino acid release and protein turnover in perfused rat hemicorpus. Proc. Natl. Acad. Sci. USA 69:816. Karkami, E., D. Spector, and R. C. Strohman. 1985. Myogenic growth factor present in skeletal muscle is purified by heparin-affinity chromatography. Proc. Natl. Acad. Sci. USA 82:8044. King, G. L., C. R. Kahn, M. M. Rechler, and S. P. Nissley. 1980. Direct demonstration of separate receptors for growth and metabolic activities of insulin and multiplication-stimulating activity (an insulin-like growth factor) using antibodies to the insulin receptor. J. Clin. Invest. 66:130. Kotts, C. E., M. E. White, C. E. Allen, and W. R. Dayton. 1987a. Stimulation of in vitro muscle cell proliferation by sera from swine injected with porcine growth hormone. J. Anim. Sci. 64:623. Kotts, C. E., M. E. White, F. Martin, C. E. Allen, and W. R. Dayton. 1987b. A statistically standard- ized bioassay for measuring the proliferation rate of myogenic cells in culture. I. Anim. Sci. 64:615. Kuffer, A. D., and A. C. Herington. 1984. Partial purification of a specific inhibitor of the insulin-like growth factors by reversed-phase high-performance liquid chromatography. J. Chromatogr. 336:87. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680. Liberti, J. P., and M. S. Miller. 1978. Somatomedin- like effects of biologically active bovine growth hormone fragments. Endocrinology 103:680. Linkhart, T. A., C. H. Clegg, and S. D. Hauschka. 1981. Myogenic differentiation in permanent clonal mouse myoblast cell lines: Regulation by macro- molecular growth factors in the culture medium. Dev. Biol. 86:19. Machlin, L. J. 1972. Effect of porcine growth hormone on growth and carcass composition of the pig. J. Anim. Sci. 35:794. Marquardt, H., G. J. Todaro, L. E. Henderson, and S. Oroszlan. 1981. Purification and primary structure

BIOASSAYS of a polypeptide with multiplication-stimulating ac- tivity from rat liver cell cultures. J. Biol. Chem. 256:6859. Martin, J. L., and R. C. Baxter. 1985. Antibody against acid-stable insulin-like growth factor binding protein detects 150,000 mol wt growth hormone-dependent complex in human plasma. J. Clin. Endocrinol. Metab. 61:799. Massague, J., and M. P. Czech. 1982. The subunit structures of two distinct receptors for insulin-like growth factors I and II and their relationship to the insulin receptor. J. Biol. Chem. 257:5038. Matsuda, R., D. Spector, and R. C. Strohman. 1984. There is selective accumulation of a growth factor in chicken skeletal muscle. I. Transferrin accumu- lation in adult anterior latissimus dorsi. Dev. Biol. 103:267. Monier, S., and Y. Le Marchand-Brustel. 1984. Effects of insulin and IGF-I on RNA synthesis in isolated soleus muscle. Mol. Cell. Endocrinol. 37:109. Moseley, W. M., L. F. Krabill, and R. F. Olsen. 1982. Effect of bovine growth hormone administered in various patterns on nitrogen metabolism in the Holstein steer. J. Anim. Sci. 55:1062. Ozawa, E., and K. Kohama. 1978. Partial purification of a factor promoting chicken myoblast multiplication in vitro. Proc. Jpn. Acad. 49:852. Pain, V. M., and P. J. Garlick. 1974. Effect of strep- tozotocin diabetes and insulin treatment on the rate of protein synthesis in tissues of the rat in vivo. J. Biol. Chem. 249:4510. Peel, C. J., D. E. Bauman, R. C. Gorewit, and C. J. Sniffen. 1981. Effect of exogenous growth hormone on lactational performance in high yielding dairy cows. J. Nutr. 111:1662. Roberts, A. M., M. A. Anzano, L. M. Wakefield, N. S. Roche, D. F. Stern, and M. B. Sporn. 1985. Type beta transforming growth factor: A bifunctional 183 regulator of cellular growth. Proc. Natl. Acad. Sci. USA 82:119. Romanus, J. A., J. E. Terrell, Y. W.-H. Yang, S. P. Nissley, and M. M. Rechler. 1986. Insulin-like growth factor carrier proteins in neonatal and adult rat serum are immunologically different: Demon- stration using a new radioimmunoassay for the carrier protein from BRL-3A rat liver cells. Endocrinology 118:1743. Salmon, W. D., L. A. Holladay, and V. J. Burkhalter. 1983. Partial characterization of somatomedin inhib- itors in starved rat serum. Endocrinology 112:360. Schalch, D. S., S. E. Tollefsen, G. J. Klingensmith, R. W. Gotlin, and M. J. Diehl. 1982. Effects of growth hormone administration on serum somato- medins, somatomedin carrier proteins and growth rates in children with growth hormone deficiency. J. Clin. Endocrinol. Metab. 55:49. Schoenle, E., J. Zapf, R. E. Humbel, and E. R. Froesch. 1982. Insulin-like growth factor I stimulates growth in hypophysectomized rats. Nature 296:252. Shen, V., L. Rifas, G. Kohler, and W. Peck. 1985. Fetal rat chondrocytes sequentially elaborate sepa- rate growth- and differentiation-promoting peptides  during their development in vitro. Endocrinology 116:920. Shimo-Oka, T., Y. Hagiwara, and E. Ozawa. 1986. Class specificity of transferrin as a muscle trophic factor. J. Cell. Physiol. 126:341. Unterman, T. G., and L. S. Phillips. 1985. Glucocor- ticoid effects on somatomedins and somatomedin inhibitors. J. Clin. Endocrinol. Metab. 61:618. Wagner, J. F., and E. L. Veenhuien. 1978. Growth performance, carcass deposition and plasma hor- mone levels in wether lambs when treated with growth hormone and thyroprotein. J. Anim. Sci. 46(Suppl. 1):397.