The Regulation of Longitudinal Bone Growth
Wang, Zhou, Bondy
While insulin-like growth factor 1 (IGF1) has long been regarded as the mediator of growth hormone’s systemic effects on growth (the somatomedin hypothesis), evidence has accumulated in recent years suggesting that growth hormone (GH) may have some IGF1-independent effects on somatic growth. For example, treating GH-resistant children with IGF1 has not produced normal growth in stature, as might have been expected if IGF1 could substitute for GH in promoting long bone growth. We hypothesized that if GH does indeed have IGF1-independent effects on long bone growth, the linear growth deficit should be more severe in mice with deletion of the GH receptor (GHR) compared with mice carrying an IGF1 deletion. Supporting this view, we have recently shown that tibial linear growth rate is reduced by approximately 35 percent in IGF1 null and by about 65 percent in GHR null mice. The rate of long bone growth is determined by the rate of epiphysial growth plate chondrocyte generation and the final size of hypertrophic chondrocytes that form the scaffolding for bone elongation. Thus, to determine how IGF1 and GHR deletion impairs long bone growth, we examined these parameters in epiphysial growth plates from IGF1 and GHR null mice.

The growth plate germinal zone, which gives rise to new chondrocytes, was significantly enlarged in size and cell number in the IGF1 null but was severely attenuated in the GHR null growth plate. Given that GH levels are greatly increased in the IGF1 null (due to lack of negative feedback from IGF1) while GH effect is abolished in the GHR null mice due to lack of its receptor, the data support the view that GH stimulates the germinal zone to generate chondrocyte precursors. In the IGF1 null growth plate proliferative zone, chondrocyte proliferation and numbers were preserved, but the rate of proliferation and chondrocyte number were significantly reduced in the GHR null proliferative zone. Both dwarves had a significant reduction in chondrocyte hypertrophy. Interestingly, the dimunition in hypertrophic chondrocyte size was about 35 percent, accounting entirely for the reduction in long bone growth in IGF1 null mice and suggesting that IGF1’s major contribution to long bone growth is augmentation of chondrocyte hypertrophy. Looking into the molecular mechanisms underlying the defect in chondrocyte somatic growth, we found that the expression of the insulin-sensitive glucose transporter, GLUT4, is significantly decreased and that the insulin-regulated enzyme, glycogen synthase kinase 3b (GSK3), is hypo-phosphorylated in IGF1 null chondrocytes. Glycogen levels were significantly decreased, and ribosomal RNA levels were reduced by almost 75 percent in IGF1 null chondrocytes. The data suggest that IGF1 promotes chondrocyte hypertrophy through “insulin-like” anabolic actions.

Investigation of potential mechanisms whereby the growth plate germinal zone is expanded and chondrocyte proliferation maintained in the IGF1 null mouse revealed that IGF2 mRNA levels were increased in the germinal and proliferative zones of the knockout. Given that GH is elevated in these mice, we hypothesized that GH might stimulate increased growth plate IGF2 production and thereby germinal zone expansion and chondrocyte proliferation. Supporting this view, IGF2 mRNA levels were significantly reduced in the GHR null growth plate. In summary, we have shown for the first time that loss of GH effect produces a much more severe deficit in long bone growth than loss of IGF1, indicating that GH truly does have IGF1-independent effects in promoting growth. In addition, we have provided a mechanistic explanation for the difference in long bone growth by demonstrating that both chondrocyte proliferation and hypertrophy are impaired in the GHR null while only hypertrophy is attenuated in the IGF1 null growth plate. These observations support dual roles for GH in promoting longitudinal bone growth: an IGF1-independent role in growth plate chondrocyte generation (possibly mediated by IGF2) and an IGF1-dependent role in promoting chondrocyte hypertrophy. The accompanying figure diagrams these effects. The large and robust differences in long bone growth rate are of obvious biological and clinical importance and may explain why IGF1 treatment of GH-resistant children may be significantly less effective than GH in terms of stature.

Figure 1

Diagram of an epiphysial growth plate outlining our current views on the respective roles of GH, IGF1, and IGF2 in promotion of longitudinal bone growth. GH stimulates expansion of the germinal zone, possibly by increasing local IGF2 expression, which enhances chondrocyte proliferation. GH also stimulates local and systemic IGF1 production, which promotes chondrocyte hypertrophy.

IGF1 and Brain Development
Cheng, Kelly, Strauss, Tseng, Bondy
The growing brain consumes about 50 percent of the total fuel available to the organism during early postnatal development. How the brain competes so successfully with peripheral tissues for energy and substrates has been unclear. Insulin preferentially enhances the use of glucose by peripheral tissues, but not by the brain. Based on our observations of cellular patterns of IGF1 and IGF1 receptor expression in the developing brain, we hypothesized that endogenous brain IGF1 serves an insulin-like role in promoting neuronal glucose utilization and hence growth during postnatal development. Supporting this view, we have shown that brain growth in IGF1-/- mice falls behind that of wild-type littermates by almost 40 percent during the postnatal period when brain IGF1 expression is normally most abundant. We have also demonstrated that brain 2-deoxy-D-[1-14C]glucose uptake (2DGU) parallels IGF1 expression in wild-type mice and is profoundly reduced in IGF1-/- mice during early postnatal development, as shown by film autoradiography in Figure 2..

Figure 2

We have elucidated the molecular mechanisms whereby IGF1 regulates neuronal metabolism by demonstrating that the active, phosphorylated form of Akt/PKB is selectively colocalized with the “insulin-sensitive” glucose transporter, GLUT4, in nerve processes of wild-type, IGF1-expressing neurons. In the IGF1-/- brain, however, Akt phosphorylation is barely detectable and GLUT4 is concentrated in neuronal perikarya, implicating IGF1-induced Akt phosphorylation in translocation of GLUT4 to the neuronal membrane. Moreover, we find that the phosphorylated form of GSK3beta and glycogen stores are abundant in wild-type, IGF1-expressing neurons. In IGF1-/- neurons, however, phospho-GSK3beta and glycogen are barely detectable, suggesting that IGF1 normally augments glucose uptake and storage by familiar, insulin-like pathways. We have proposed that IGF1’s primary physiological role in normal brain development is to promote the growth of projection neurons. IGF1 expression is most abundant in neurons (e.g., Purkinje cells) destined to be the largest and most complex in the brain, implicating IGF1’s insulin-like, anabolic effects in this extraordinary growth. IGF1-expressing neurons grow large perikarya, long axons, and extraordinarily prolific dendritic arbors, which are severely hypoplastic in the IGF1 null brain (see figure below). Homozygous IGF1 deletion results in mental retardation, showing that this anabolic peptide’s effects on neuronal metabolism and growth have important effects on cognitive function. Current work in our laboratory is aimed at discovering factors regulating neuronal IGF1 expression with a view to therapeutic manipulation to increase the potential for cognitive development and successful responses to brain injury.

Estrogen Effects in the Primate Neocortex
Cheng, Bondy
Estrogen stimulates neurite growth and neurotransmitter synthesis and protects against diverse types of neural injury in vitro. In vivo, estrogen treatment is reputed to protect against Alzheimer’s disease in menopausal women. To investigate the molecular basis for estrogen’s neuroprotective effects in a model system relevant to humans, we evaluated the expression of glucose transporters and IGF1 in the prefrontal cortex of estradiol- versus placebo-treated ovariectomized rhesus monkeys. We investigated the expression of facilitative glucose transporters (Gluts) 1, 3, and 4 by using in situ hybridization, immunohistochemistry, and immunoblot analysis. Gluts 3 and 4 were concentrated in cortical neurons while Glut1 was localized in capillaries and glial cells. Estradiol (E2) treatment induced two- to four-fold increases in Glut 3 and Glut 4 mRNA and protein levels. E2 treatment also induced an increase in glial Glut1 mRNA levels but did not appreciably affect vascular Glut1 gene expression. IGF1 mRNA was concentrated in cortical neurons together with Gluts 3 and 4 and was significantly increased in E2-treated animals. These novel data suggest that the up-regulation of Gluts 3 and 4 and IGF1 expression contribute to estrogen’s salutary effects on neural tissue. In support of this view is the finding by other researchers that glucose transport and transporter expression are reduced in subjects with Alzheimer’s disease, suggesting one way in which estrogen may address the Alzheimer’s neuropathology. Current studies are aimed at profiling the entire spectrum of genes regulated by estrogen effect in the nonhuman primate neocortex.