Kolling Institute of Medical Research Royal North Shore Hospital St Leonards, NSW 2065, Australia
Insulin-like growth factor binding protein-1 (IGFBP-1) is distinct among the members of the IGFBP family in being acutely regulated by metabolic stimuli (1). Studies with cultured human liver explants suggest that the major regulatory influences on IGFBP-1 production are insulin, which is inhibitory, and hepatic substrate deprivation, which is stimulatory, acting through a cyclic AMP-dependent mechanism (2, 3). Dexamethasone is also markedly inhibitory to IGFBP-1 production by human liver in explant culture (2).
Metabolic influences appear to account for the marked diurnal variation in serum IGFBP-1 levels seen in healthy subjects, which can be attributed to changing nutritional status throughout the day (4, 5, 6, 7, 8). The peak and the nadir of daily serum IGFBP-1 levels correspond temporally to the maximal and minimal serum cortisol levels, respectively (9), suggestive of a functional link in their regulation. However, in hypercortisolism due to Cushings disease, serum IGFBP-1 is markedly suppressed, and the diurnal variation is lost (9).
While the apparent suppression of IGFBP-1 in hypercortisolism is consistent with the inhibitory effect of dexamethasone in human liver explants, it is contrary to the well-established role of corticosteroids in the rat, where dexamethasone has been shown to increase IGFBP-1 serum levels and gene transcription (10, 11). Expression of the human IGFBP-1 gene is also stimulated by glucocorticoids (12), so that the likely explanation for the suppressive effect of hypercortisolism on serum IGFBP-1 levels is that it occurs secondarily to cortisol-induced hyperinsulinemia (13). The inhibitory effect of dexamethasone in human liver explants remains unexplained.
Insulin-induced hypoglycemia provokes a sharp rise in serum IGFBP-1, commencing an hour after the glucose nadir (1). IGFBP-1 can inhibit the effect of IGF-I on tissue glucose uptake (14). A high serum level of IGFBP-1, whether introduced by infusion or by transgenesis, is associated with hyperglycemia, and reverses the hypoglycemic action of IGF-1 (15, 16). These observations suggest that the hypoglycemia-induced rise in IGFBP-1 could serve a counterregulatory role in glucose homeostasis, as first proposed in 1988 (1). The 1-hr delay in IGFBP-1 response might render it ineffective as a counter to acute, pharmacologically-induced hypoglycemia; this delay might be caused in part by the residual suppressive effect of the administered insulin. Nevertheless, a prolonged IGFBP-1 increase in response to the gradual hypoglycemia associated with extended fasting (7) might effectively block the unwanted hypoglycemic activity of free IGFs in this more physiological situation, so that IGFBP-1 could play a true counterregulatory role.
Katz et al. (17) have recently described the increase of IGFBP-1 in 21 children in response to hypoglycemia associated with prolonged fasting. A positive association between serum cortisol and IGFBP-1 levels led the authors to conclude that the IGFBP-1 rise was "cortisol-induced." However, it is possible that the association was coincidental rather than causal, and other data suggest that the IGFBP-1 rise can occur independently of glucocorticoids. For example, in adrenalectomized rats with undetectably low serum corticosterone levels, iv insulin still induced a rise in IGFBP-1, though less than that seen in adrenal-intact animals (18). Similarly, increases in IGFBP-1 during fasting studies in a limited number of cortisol-deficient children led Cotterill et al. (7) to conclude that IGFBP-1 rose independently of cortisol levels.
In the original study on insulin-induced hypoglycemia (1), we also
reported in the text that the IGFBP-1 rise occurred independently of
the cortisol response to hypoglycemia. Figure 1illustrates the IGFBP-1 responses in 14 subjects who, after an
overnight fast, were treated at 0900 h with a single injection
containing GnRH (100 µg), TRH (200 µg), and insulin (0.1 U/Kg), as
part of an investigation of pituitary function (1). In 11 subjects, a
mean cortisol peak of 680 ± 54 nmol/L (SEM) was
achieved at 60 min, a rise of some 250 nmol/L above initial values. The
other 3 subjects, defined as failing to show a cortisol response,
peaked at a mean cortisol level of 321 ± 126 nmol/L, less than 50
nmol/L above initial values. As seen in the figure, initial IGFBP-1 in
the cortisol-responsive group (n=11) was 60.1 ± 12.6 µg/L,
compared with 142.0 ± 17.2 µg/L in the cortisol-nonresponsive
group (n=3). Despite the higher baseline IGFBP-1 levels in the latter
group, both groups showed a marked IGFBP-1 rise commencing 90 min after
insulin administration. At 180 min, IGFBP-1 in the cortisol-responsive
group was 168.0 ± 31.6 µg/L, a comparable increment above
baseline to that seen in the cortisol-nonresponsive group (277.7
± 6.7 µg/L at 180 min).
|
Footnotes
Received February 16, 1999. Address correspondence to: Robert C. Baxter, Department of Molecular Medicine, University of Sydney (E25), Kolling Institute of Medical Research, Royal North Shore Hospital, St. Leonards, NSW 2065, Australia.
References
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |