Growth hormone and growth in diabetic rats: effects of insulin and insulin-like growth factor-I infusions.

The effects of streptozotocin-induced diabetes on weight gain, bone growth and GH secretion have been studied in conscious chronically cannulated male rats. In addition to the classic diabetic symptoms (hyperphagia, polydipsia, polyuria, glycosuria and hyperglycaemia), the slow body weight gain (0.95 +/- 0.5 compared with 2.63 +/- 0.5 g/day in non-diabetic controls) was associated with a reduction in bone growth (from 162 +/- 9 to 48 +/- 4 microns/day) and a reduced pituitary GH content (from 1.5 +/- 0.2 to 0.6 +/- 0.06 mg/gland). Serial blood sampling during the day or overnight showed that the normal male episodic GH secretory pattern was obliterated in the diabetic animals. The constant osmotic stimulation of hyperglycaemia and high fluid turnover was reflected in a significant reduction in pituitary oxytocin and arginine vasopressin (AVP) stores. Intravenous insulin infusions (67-1340 pmol/h for 4 or 7 days) caused a large initial weight gain (greater than 20 g in 2 days) followed by a slower increase, and stimulated tibial bone growth (to 100 +/- 16 and 126 +/- 8 microns/day after 4 or 7 days respectively). Insulin infusion for 7 days also increased pituitary GH content (to 1 +/- 0.15 mg/gland), and the normal episodic GH secretory pattern returned. Intravenous infusions of insulin which reduced, but did not completely normalize, blood glucose levels, allowed the resumption of growth and pulsatile GH secretion. Continuous infusion of recombinant human insulin-like growth factor-I (hIGF-I) at 1110 pmol/h for 54 h also caused a large initial rise in body weight in diabetic rats (17.1 +/- 1.6 compared with 7.5 +/- 2.8 g in saline-infused controls) due primarily to increased fluid retention. This effect of hIGF-I occurred without any significant changes in pituitary GH, AVP, oxytocin, blood glucose or bone growth over this short-term infusion, nor was there any obvious effect on spontaneous GH secretion, monitored over the entire infusion period. We conclude that the diabetic rat is not a good model to study growth stimulation by short-term insulin or IGF-I treatments because the insulin-like effects of these peptides obscure their specific growth-promoting activities in this model.