Postnatal Growth Responses to Insulin-Like Growth

Organ weight was compared in adult mice with deletion of one (IRS-1) or both (IRS-1) copies of the insulin receptor substrate-1 (IRS-1) gene and IRS-1 littermates. IRS-1 mice showed modest reductions in weight of most organs in proportion to a decrease in body weight. IRS-1 mice showed major reductions in weight of heart, liver, and spleen that were directly proportional to a decrease in body weight. In IRS-1 mice, kidney and particularly small intestine and brain exhibited proportionately smaller weight reductions, and gastrocnemius muscle showed a proportionately greater weight reduction than the decrease in body weight. Growth deficits in IRS-1 mice could reflect impaired actions of multiple hormones or cytokines that activate IRS-1. To assess the requirement for IRS-1 in insulinlike growth factor I (IGF-I)-dependent postnatal growth, IRS-1 mice were cross-bred with mice that widely overexpress a human IGF-I transgene (IGF1) to generate IGF1 and wild-type mice on an IRS-1, IRS-1, and IRS-1 background. IGF-I overexpression increased body weight and weight of brain, small intestine, kidney, spleen, heart, and gastrocnemius muscle in IRS-1 mice. IGF-I overexpression could not completely reverse the body growth retardation in IRS-1 mice. Absolute or partial IRS-1 deficiency impaired IGF-I-induced body overgrowth more in females than in males. In males and females, IGF-I stimulated similar overgrowth of brain regardless of IRS-1 status, and intestine and spleen showed dose dependence on IRS-1 for IGF-I-induced growth. IGF-I-induced growth of gastrocnemius muscle had an absolute requirement for IRS-1. IGF-I-induced growth of kidney and heart was impaired by IRS-1 deficiency only in females. In vivo, therefore, most organs do not require IRS-1 for IGF-I-induced growth and can use alternate signaling molecules to mediate IGF-I action. Other organs, such as gastrocnemius muscle, require IRS-1 for IGF-I-induced growth in vivo. (Endocrinology 140: 5478–5487, 1999) I growth factor I (IGF-I) plays a key role in regulating body and organ growth during postnatal development (1). Serum levels of IGF-I correlate with postnatal increases in body weight gain and linear growth in humans and rodents (1). Transgenic mice that overexpress a metallothionein promoter-driven human IGF-I transgene show modest increases in plasma levels of IGF-I and exhibit widespread overexpression of the IGF-I transgene in multiple tissues (2–4). The IGF-I transgenic mice show increased body weight gain from about 2 weeks of age through adulthood and have significant increases in the sizes of a number of organs (2–4). The type 1 IGF receptor (IGF1R) is the primary mediator of the growth-promoting actions of IGF-I (5). The IGF1R is a receptor tyrosine kinase that is related in structure and function to the insulin receptor (IR) (5). Ligand-dependent autophosphorylation of IGF1R or IR leads to binding and tyrosine phosphorylation of a 185-kDa signaling intermediate, insulin receptor substrate-1 (IRS-1) (6). Phosphorylated tyrosines within IRS-1 provide binding motifs for multiple proteins that contain SH2 domains, including the p85 regulatory subunit of phosphatidyl inositol 3-kinase (PI 3kinase), growth factor receptor bound protein-2, phosphotyrosine phosphatase Syp, and the oncogenic proteins Abl, Crk, Nck, and Fyn (6). These interactions mediate activation of multiple downstream pathways, including PI 3-kinase, extracellular signal-related kinase/mitogen-activated protein kinase, and p70 s6 kinase, and induction of immediate early genes such as c-fos (6). Mice that are homozygous for targeted deletion of the IRS-1 gene are growth retarded at birth and remain growth retarded to adulthood (7, 8). These IRS-1 null mice show impaired glucose tolerance and a decrease in the magnitude of insulinor IGF-mediated glucose uptake in vivo, but do not develop type 2 diabetes (7, 8). Insulin-stimulated activation of PI 3-kinase and mitogen-activated protein kinase also is impaired in IRS-1 null mice, but the level of impairment is tissue specific (6–9). Liver, for example, is affected less than skeletal muscle (9). In liver of IRS-1 null mice, the relative levels of insulin-dependent tyrosine phosphorylation of IRS-2, a molecule that is structurally and functionally related to IRS-1, are greater than those in muscle (9). IRS-2 therefore can substitute for IRS-1 as a mediator of insulin action in a tissue-specific manner (9). Whether different organs or tissues show differential dependence on IRS-1 for normal postnatal growth in vivo has not been examined in detail. To address this question, the present study compared organ Received March 11, 1999. Address all correspondence and requests for reprints to: Dr. Pauline Kay Lund, Ph.D., Department of Cell and Molecular Physiology, CB# 7545, University of North Carolina, Chapel Hill, North Carolina 275997545. E-mail [email protected]. * This work was supported by NIH Grants DK-40247 (to P.K.L.), DK-33201 (to C.R.K.), AG09973 (to D.S. and P.K.L.), and HD-08299 (to A.J.D.) and the New Zealand/USA Cooperative Science Program of the ISAT Linkages Fund (to J.M.O.). 0013-7227/99/$03.00/0 Vol. 140, No. 12 Endocrinology Printed in U.S.A. Copyright © 1999 by The Endocrine Society