Previous studies by LeRoith and co-workers and Ueki et al have demonstrated that selective loss of liver-derived insulin-like growth factor-1 (IGF-1) or of acid labile subunit (ALS) does not substantially impair murine growth and development despite marked decline in circulating levels of IGF-1.^1,2  This has led to the suggestion that only the IGF-1 produced locally by bone is necessary for linear growth.³ 

In order to explore this question further, LeRoith and his colleagues developed double “knock-out” animals which were deficient in both liver IGF-1 and ALS (LID-ALSKO), and compared these with animals deficient only in liver IGF-1 (LIDKO) or ALSKO. As anticipated, serum concentrations of IGF-1 were decreased markedly, -65% in ALSKO, -75% in LIDKO, and -90% in LID-ALSKO relative to control animals with normal hepatic IGF-1 and ALS production.  However, the rate of IGF binding protein-3 (IGFBP-3) degradation was also increased in these animals; thus free IGF-1 values were increased modestly in LIDKO (+150%), minimally in ALSKO (+108%), and markedly in LID-ALSKO animals (+350%).  Growth hormone and insulin concentrations were greatly increased in LID-ALSKO mice.  The clearance of IGF-1 was markedly accelerated in ALSKO (32 minutes) and LID-ALSKO (18 minutes) as compared with control (69 minutes) and LIDKO (73 minutes) mice, reflective of lack of binding of IGF-1 to IGFBP-3/ALS.

Intrauterine growth of all animals was apparently normal.  By 3 weeks and 4 weeks of post natal age (Figures), the length and weight of the LID-ALSKO mice were less than those of the intact animals.  Linear growth of the LIDKO and ALSKO animals did not differ from controls.   However, the rate of weight gain of ALSKO mice was impaired to the same extent as that of the LID-ALSKO group. Tibial length, and heights of germinal, proliferating, and hypertrophic zones of the proximal tibial growth plate, were significantly diminished in the LID-ALSKO mice but not in the two single “knock-out” groups.  On the other hand, femoral length, total and cortical bone density, periosteal circumference, and cortical and trabecular bone volume were diminished in all “knock-out” groups, but to a substantially greater degree in the LID-ALSKO animals.  Administration of exogenous IGF-I increased linear growth, femoral length, and size of the proximal tibial growth plate, as well as IGFBP-3 concentrations, in all groups. IGF-1 mRNA levels in bone were similar in all groups.

The investigators concluded that circulating IGF-1 was important for linear and appositional bone growth and bone mineralization and that its effects were mediated through actions on periosteal osteoblasts as well as upon chondrocytes within the epiphyseal growth plates.

Yakar S, et al.  J Clin Invest 2002;110:771-781.

First Editor’s Comment:  This important paper establishes the necessity of circulating IGF-I for normal growth and bone mineralization.  It demonstrates that osseous synthesis of IGF-I alone is insufficient for normal linear growth of bone and mineral deposition.  Thus, reexamination of the “somatomedin hypothesis” suggests that both liver derived and locally synthesized IGF-I are necessary for normal bone metabolism.  Interestingly, “knock-out” of any of the IGFBPs has little effect upon the phenotype of the mutant mouse, but their over expression results in inhibition of growth.⁴ One wonders what the phenotype of the mouse that lacks IGF-I, IGFBP-3, and ALS might be ... possibly lethal?

Allen W. Root, MD

References

1.  LeRoith D, et al.  Endocrine Rev 2001;22:53-74.

2.  Ueki I, et al. Proc Natl Acad Sci USA 2000;97:6868-6873.

3.  Kaplan SA.  Growth Genetics & Hormones 2002;18:38-39.

4.  Silha JV, Murphy LJ. Endocrinology 2002;143:3711-3714.

Second Editor’s Comment: In Growth, Genetics & Hormones (Vol. 18, No. 3), an important lead article entitled Somatomedin Hypothesis: Time for Reexamination was written by Dr. Solomon Kaplan.  He has been asked to write an editorial comment.

Dr. Kaplan’s Comment: The paper by Yakar et al extends and amplifies the findings in a previous publication by the authors¹ on the role of circulating IGF-1 in promoting longitudinal growth in mice.  They had already shown that despite inactivation of the IGF-1 gene in the liver, resulting in reduced concentrations of circulating IGF-1 by as much as 75%, the growth of the animals was not impaired.  Their findings were consistent with the growing body of evidence against the validity of the somatomedin hypothesis, which holds that the effects of growth hormone on longitudinal growth are mediated through hepatic production of IGF-1.²

IGF-1 circulates in the serum largely as a 150-kDa complex comprised of the IGF-1 molecule, IGF binding proteins (mostly IGFBP-3), and the acid labile subunit (ALS).  Others had previously shown that ALS knockout (ALSKO) mice experienced only mild growth retardation despite profound disruption of the circulating IGF system.

Yakar’s current paper reported the effects of double gene disruption of the IGF system: inactivation of the hepatic gene for IGF-1 (LID) combined with ALSKO, on bone growth and density.  In the mice carrying the double gene deletion, there was a reduction of circulating IGF-1 concentrations by as much as 85 to 90%; the animals also experienced significant growth impairment.  There was a diminution in the amount of circulating IGFBP-3 protein and also in the free IGF-1 fraction.  Loss of ALS led to more rapid disappearance of 125-I labeled IGF from the serum because absence of the ALS protein leads to proteolytic cleavage of IGFBP-3 and loss of its protective binding of IGF-1. The authors conclude that a minimum concentration of IGF-1 in the serum, higher than what they observed in the double gene-deletion mice, is necessary for normal bone and somatic growth.

Following administration of IGF-1 by injection, the animals with the double gene deletion experienced increased serum IGF and IGFBP-3 concentrations accompanied by restoration of normal bone growth and modeling, as well as increased somatic growth.  These findings are consistent with their observation that the restoration of normal growth can be accounted for by increased serum IGF-1 concentrations above the minimal levels necessary for normal growth to occur.

This paper provides confirmatory evidence that hepatic derived IGF-1 and acid labile subunit are not necessary for normal growth provided minimal serum levels are maintained from non-hepatic sources including autocrine/paracrine production by target tissues.

Solomon A. Kaplan, MD

References

1.  Yakar S, et al. Proc Natl Acad Sci USA 1999;96:7324-9.

2.  Daughaday WH, et al. Nature 1972;235:107.