Ghrelin stimulation of growth hormone isoforms: parallel secretion of total and 20-kDa growth hormone and relation to insulin sensitivity in healthy humans

Abstract

Context: The 20-kDa human GH (hGH) is produced in the pituitary by alternative splicing of the hGH-N gene. The 20-kDa hGH promotes growth similarly to 22-kDa or total hGH, the predominant form in circulation, but the relative effects of these isoforms on glucose metabolism have been debated.

Objective: To investigate the effect of ghrelin on 20-kDa and total hGH secretion in healthy, nonobese subjects. We also studied associations between basal GH concentration and fasting glucose and insulin as well as between dynamic GH secretion and insulin sensitivity.

Design and setting: Synthetic human acyl ghrelin (0.2 or 0.6 nmol/kg · h) or saline was infused in random order in 14 healthy subjects (six males, eight females; age 27.7 ± 6.3 yr; body mass index 22.0 ± 2.7 kg/m(2), mean ± SEM) on 3 separate days. Ghrelin was infused for 45 min to achieve steady-state levels and continued through a 3-h frequently sampled i.v. glucose tolerance test. Insulin sensitivity index was quantified using the minimal model of glucose kinetics.

Results: Basal 20-kDa and total GH concentrations were 0.4 ± 0.1 and 2.2 ± 0.4 ng/ml, respectively, with a 20-kDa to total GH ratio of 0.13 ± 0.02. Females had significantly higher baseline GH levels. Ghrelin administration increased 20-kDa and total GH levels in a parallel and dose-dependent fashion, with no significant change in the ratio of the isoforms. Basal 20-kDa and total GH levels were negatively correlated with fasting glucose, insulin, and homeostasis model assessment of insulin resistance. During the frequently sampled iv glucose tolerance test, GH secretion was positively correlated with insulin sensitivity index with saline infusion.

Conclusion: Ghrelin dose-dependently increases endogenous 20-kDa and total GH secretion in a parallel fashion in healthy subjects. Both basal and stimulated levels of the different GH isoforms were positively associated with insulin sensitivity in this cohort of healthy men and women.

Fig. 1.

Serum 20-kDa hGH (A), total hGH (B) levels, and 20-kDa to total hGH ratio (C) during continuous iv infusions of saline and 0.2 and 0.6 nmol/kg · h acyl-ghrelin in healthy men and women (n = 14). A 180-min FSIVGT was performed between 45 and 225 min during the ghrelin infusion.

Fig. 2.

Serum 20-kDa GH (A) and total hGH (B) levels during continuous iv infusions of saline and 0.2 and 0.6 nmol/kg · h acyl-ghrelin in healthy men and women. A 180-min FSIVGT was performed between 45 and 225 min during the ghrelin infusion.

Fig. 3.

Baseline levels of 20-kDa and total hGH (nanograms per milliliter) were negatively associated with fasting plasma glucose (milligrams per deciliter) (A), insulin (picmoles per liter) (B), and HOMA-IR (C).

Fig. 4.

AUC_20-kDaGH and AUC_totalGH during FSIVGT (45–225 min of saline infusion) were positively associated with tissue insulin sensitivity as measured by S_I.