| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Bone Metabolism Group (J.A.C., R.E., A.B.) and Diabetes and Endocrinology Group (R.T.R., S.R.H.), Division of Clinical Sciences (North), University of Sheffield, S5 7 AU Sheffield, United Kingdom
Address all correspondence and requests for reprints to: Dr. Aubrey Blumsohn, M.D., Ph.D., MRCPath, University of Sheffield, Division of Clinical Sciences (North), Northern General Hospital, Herries Road, Sheffield, S5 7 AU, United Kingdom. E-mail: . ablumsohn{at}sheffield.ac.uk
Abstract
Bone turnover is acutely suppressed after feeding or oral glucose. Insulin infusion suppresses bone turnover and might mediate this effect, but this is confounded by a possible direct effect of hypoglycemia. We examined the effect of euglycemic hyperinsulinemia and hypoglycemic hyperinsulinemia on bone turnover using an insulin clamp. Sixteen men participated in this double-blind crossover study. Clamp induction involved infusion of insulin (80 mU/m2·min) while maintaining euglycemia (5 mmol/liter) for 40 min with a variable rate dextrose infusion. Glucose was lowered to 2.5 mmol/liter (hypoglycemic clamp) or maintained at 5 mmol/liter (euglycemic clamp) for a further 105 min. Nine controls received a matched saline infusion. Measurements included serum C-terminal telopeptide of type I collagen, procollagen type I N-terminal propeptide, osteocalcin, and PTH.
Induction of hyperinsulinemia resulted in a reduction in PTH (27% ± 5; P < 0.01), but no significant change in bone turnover from baseline. Hypoglycemic clamp resulted in suppression of serum C-terminal telopeptide of type I collagen by 34% ± 3, procollagen type I N-terminal propeptide by 15% ± 1, osteocalcin by 5% ± 1, and PTH by a further 12% ± 5 (all P < 0.05). By contrast, there was no significant change in any marker of bone turnover during euglycemic clamp.
Postprandial hyperinsulinemia is unlikely to explain the acute suppression of bone turnover with feeding. The reduction in bone turnover during hypoglycemia may be related to hypoglycemia itself, acute changes in PTH, or other hormones released in response to hypoglycemia.
RECENT OBSERVATIONS SUGGEST that bone turnover is suppressed after feeding (1) or after an oral glucose load (2). This change in bone turnover is rapid and may be observed over minutes to hours in human studies (1, 2). The suppression in bone resorption after a standard 75-g glucose tolerance test is approximately 50% using serum C-terminal telopeptide of type I collagen (sßCTX) as a marker of bone collagen degradation (2). The postprandial suppression of bone turnover is observed using several different markers of bone formation and resorption (1). In addition, the pronounced circadian rhythm of bone turnover is attenuated by fasting (3, 4) suggesting that this rhythm is mediated, at least in part, by food intake.
The factors that mediate the acute effect of food or glucose ingestion on bone turnover are unknown. Possible mechanisms include the postprandial rise in insulin, a direct effect of glucose on bone metabolism, or other associated endocrine responses. One recent study found that intravenous administration of insulin results in a decrease in bone resorption (2), suggesting insulin as a possible mediator. However, this design does not differentiate the direct effect of insulin from the effect of insulin-induced hypoglycemia on bone.
We examined the effect of euglycemic hyperinsulinemia on bone turnover using an insulin clamp technique. This technique enables examination of the independent effect of glucose and insulin by simultaneous infusion of glucose and insulin. The blood glucose is maintained at a predetermined value while maintaining a constant insulin concentration. The primary aim of this study was to determine the acute effect of hyperinsulinemia alone on bone turnover, while maintaining steady-state euglycemia. We also examined the acute effect of hypoglycemic hyperinsulinemia on bone turnover using a hypoglycemic clamp.
Subjects and Methods
Subjects
Twenty-five healthy men (mean age, 23 yr; range, 1835 yr) were recruited. Subjects were excluded if they had any disease known to affect bone metabolism or were taking any medications. Three subjects were current smokers. Mean body mass index was 23.0 kg/m2 (range, 19.129.0 kg/m2). The study was performed in accordance with current guidelines on Good Clinical Practice and the Declaration of Helsinki. All subjects gave written informed consent, and the study was approved by the North Sheffield Research Ethics Committee.
Protocol
On each study day, volunteers attended the investigation unit after an overnight fast from 2400 h. All subjects remained fasting throughout the study. Subjects were blinded to the study protocol, and all measurements were performed double blind. Sixteen subjects were studied in a crossover design with a euglycemic hyperinsulinemic clamp protocol or a hypoglycemic hyperinsulinemic clamp on two separate days. The order of study (hypoglycemic/euglycemic or euglycemic/hypoglycemic) was randomized at the first visit. Nine control subjects received a matched saline infusion but no glucose or intravenous insulin.
Subjects were placed in a semirecumbent position, and a cannula was inserted into the nondominant antecubital fossa for infusion of saline, glucose, or insulin. Blood glucose was monitored using arterialized venous samples from a second cannula at 5 min intervals throughout the clamp protocol. Cannulae were kept patent using 0.9% saline.
During the clamp protocols, subjects received a continuous infusion of soluble insulin (Human Actrapid, Novo Nordisk Pharma Ltd., Crawley, UK). Insulin was infused at a fixed rate (80 mU/m2·min). Glucose (20%) was infused at a variable rate to maintain blood glucose at the desired level. The infusion rate was adjusted every 5 min according to a predefined algorithm.
The induction phase of the clamp protocols began at 0900 h. The blood glucose was adjusted to 5.0 mmol/liter during the 40-min induction period for both hypoglycemic and euglycemic clamps. Steady-state was achieved within 40 min in all subjects. For purposes of statistical analysis, all results were expressed relative to this time point (time 0). For the euglycemic clamp, blood glucose was maintained at 5 mmol/liter for the duration of the clamp. For the hypoglycemic clamp, glucose was lowered in a stepwise manner from time 0 to reach 2.5 mmol/liter within 60 min. Steady-state hypoglycemia was maintained for a further 60 min. Activation of counter-regulatory responses to hypoglycemia including the suppression of insulin secretion and stimulation of glucagon release typically occurs at plasma glucose below 3.5 mmol/liter in normal subjects (5). During the control protocol, subjects received a matched infusion of saline (200 ml/h).
Plasma samples were collected at -40, 0, 15, 30, 45, 60, 75, 90, 105, and 120 min. All samples were frozen within 1 h and stored under identical conditions at -70 C.
Biochemical methods
Glucose was measured using a glucose oxidase method (YSI 2300, YSI, Inc., Yellow Springs, OH). Total insulin was measured by RIA using a double-antibody technique. The intra-assay coefficient of variation was 6%. Total calcium was measured using dry-slide technology (Ortho Clinical Diagnostics, Rochester, NY). Intact PTH was measured using an immunochemiluminometric assay (Roche Elecsys, Hoffman-LaRoche Inc., Penzberg, Germany).
Bone resorption was assessed using serum ß C-terminal crosslinked telopeptide of type I collagen (sßCTX; ß-Crosslaps/serum; Roche Elecsys). Bone formation was assessed using serum procollagen type I N-terminal propeptide (PINP; Roche Elecsys) and osteocalcin (OC; Roche Elecsys). The analytical coefficients of variation for sßCTX, PINP, and OC were less than 5%, less than 1%, and less than 2%, respectively, over the relevant analytical range. All assays were performed blind to the randomization.
Statistical analysis
Results for markers of bone turnover were expressed as a percentage of baseline (time 0 min). Responses were tested using repeated measures ANOVA with time as the within-subjects factor. The Huynh-Feldt correction was applied when the sphericity assumption was not met. Post hoc comparisons were performed using paired contrasts to compare the measurement at each time with baseline using Bonferroni adjustment for multiple comparisons. Area under the curve (AUC) was calculated for each protocol and analyte and compared using t tests. Statistical analyses were performed using Statgraphics Plus Version 4 (Manugistics, Inc., Rockville, MD) and SPSS 10.0 (SPSS, Inc., Chicago, IL).
Results
Baseline characteristics for the 25 subjects are shown in Table 1
. The absolute values for insulin and glucose during the induction of hyperinsulinemia and subsequent clamp protocols are shown in Fig. 1
. In the control protocol, there was no significant change in insulin or glucose with time (P > 0.05). Insulin increased 13-fold from a fasting (-40 min) value of 72 pmol/liter to 945 pmol/liter (P < 0.001) during the induction phase (-40 to 0 min) of the hypoglycemic and euglycemic clamps. Plasma glucose remained stable (P > 0.05) during induction of hyperinsulinemia. Plasma insulin remained stable during the clamp (0 to 120 min), and the change in plasma insulin with time was identical in hypoglycemic and euglycemic protocols (AUC; paired t test; P > 0.05). All subjects exhibited a marked adrenergic response, and some subjects exhibited neuroglycopenic symptoms. There was a 3-fold increase in cortisol during the hypoglycemic clamp (repeated measures ANOVA effect of time P < 0.01; Fig. 2
and Table 2
).
|
|
|
|
|
In contrast, the hypoglycemic clamp was associated with a significant reduction in sßCTX, PINP, and OC (repeated measures ANOVA effect of time P < 0.01; Fig. 3
). Percentage change at time 105 min (Table 2
) and AUC differed significantly between the euglycemic and hypoglycemic clamp protocols for sßCTX and PINP (P < 0.001) but not for OC (P > 0.05).
Change in PTH and serum calcium is shown in Fig. 4
and Table 2
. Induction of hyperinsulinemia (-40 to 0 min) resulted in a transient 27% reduction in PTH (P < 0.05), which partially recovered during the euglycemic hyperinsulinemic clamp (Fig. 4
). In contrast, during hyperinsulinemic hypoglycemia the reduction in PTH was progressive.
|
There is increasing evidence that intake of a variety of nutrients, including a mixed meal (1), glucose (2), or calcium (6, 7, 8), results in acute suppression of bone turnover. The pronounced circadian rhythm of bone turnover is attenuated by fasting (3, 4). The mechanism for the effect of feeding on bone turnover is unknown, and it is not clear whether different nutrients operate via a common mechanism.
The acute effect of feeding on bone resorption could be attributed to postprandial changes in glucose, insulin, or other postprandial endocrine changes. The insulin receptor is expressed by osteoblasts and osteoclast-like cells (9, 10), and insulin may modulate bone turnover directly (10). Hyperglycemia with hyperinsulinemia could also have indirect effects on bone turnover by stimulating intestinal calcium absorption (11) and renal calcium excretion (12). Postprandial changes in renal calcium handling are probably mediated by insulin (13, 14). There is, however, some evidence that insulin might contribute to the acute effect of feeding on bone resorption. Intravenous administration of insulin does decrease bone resorption (2). However, such studies are confounded by a possible indirect effect of insulin-induced hypoglycemia on bone, and the decrease observed is less than would be expected if insulin were the sole mediator (2).
In this study, we used a hyperinsulinemic clamp protocol to investigate the relative contribution of insulin and glucose on bone turnover. There was no significant change in bone turnover in the control group during the induction phase. Infusion of insulin had no effect on bone turnover when euglycemia was maintained. This suggests that insulin does not play an important role in mediating the acute effect of feeding on bone resorption.
Plasma insulin concentrations attained in this study are higher than those observed physiologically in response to a meal and approximately twice that observed after a standard glucose tolerance test. Mediators other than insulin are therefore likely to explain the acute decrease in bone turnover after a mixed meal (1) or glucose intake (2). Several other hormones such as amylin, glucose-dependent insulinotropic peptide, and calcitonin may be secreted in response to nutrient intake, and several of these have been shown to regulate bone turnover in vitro (15, 16, 17) and in vivo (18, 19, 20, 21, 22). Catecholamines and glucagon are unlikely mediators of the suppression of bone turnover after feeding because they are released in the late postprandial phase (23, 24).
In contrast to the effect of euglycemic hyperinsulinemia, hypoglycemia results in a decrease in bone resorption and formation despite equivalent insulin concentrations. This suggests that the reduction in bone turnover previously observed after an insulin infusion (2) is likely to be related to hypoglycemia itself.
There are a number of possible mechanisms for the reduction in bone turnover during hypoglycemia including: 1) a direct effect of hypoglycemia on osteoclast and osteoblast function, 2) other counter-regulatory hormones released in response to hypoglycemia, 3) changes in the metabolism of bone turnover markers, and 4) the hypoglycemia-induced reduction in plasma PTH. The most likely mechanism is hypoglycemia-induced suppression of osteoclast and osteoblast function because these cells are sensitive to glucose supply (25, 26, 27). The glucocorticoid and adrenergic response to hypoglycemia is unlikely to mediate the decrease in bone resorption with hypoglycemia. If anything, catecholamines appear to increase bone resorption (28, 29). Acute changes in serum cortisol do not alter bone resorption (30, 31). It is possible that glucagon might inhibit bone resorption indirectly by stimulating the secretion of calcitonin (32).
The apparent decrease in bone turnover observed during hypoglycemia could also be due to decreased clearance or metabolism of these markers. The factors influencing metabolism of these analytes are poorly understood. Concordance of response between different analytes would suggest a genuine effect because these are removed from the extravascular compartment by different mechanisms. In a previous study, a variety of bone turnover markers changed in parallel after feeding, with the exception of bone alkaline phosphatase, which has a long half-life (1).
We observed an acute decrease in PTH during the induction phase of the clamp, with partial recovery during sustained euglycemic hyperinsulinemia. The changes we observed are similar to those observed previously (33, 34, 35), including the recovery with sustained hyperinsulinemia (34). The insulin-induced change in PTH is independent of serum calcium or ionized calcium (34). There are several mediators apart from calcium, which are likely to regulate PTH secretion in this context. Glucose (36) and insulin (33, 37) may modulate PTH secretion directly. There are insulin receptors in parathyroid tissue, and insulin modulates PTH secretion in vitro (36, 37, 38). Insulin has also been shown to modulate PTH secretion acutely in vivo (33).
Hypoglycemic hyperinsulinemia was associated with a marked reduction in PTH secretion. This has also been observed in insulin-induced hypoglycemia (37), and is unlikely to be due to changes in serum calcium (37). Glucose is known to have a direct effect on PTH secretion in vitro (36). There is conflicting evidence relating to the effect of catecholamines on PTH secretion (39).
The role of PTH as a mediator of acute food-induced changes in bone turnover is uncertain. Fasting results in attenuation of the circadian rhythm of PTH secretion (3, 40). However, the suppression of PTH by calcium infusion does not alter the circadian rhythm in bone turnover (41), and reversal of the nocturnal increase in PTH by timed evening calcium administration only partially suppresses the nocturnal peak in bone resorption (6).
In conclusion, we have shown firstly that insulin has a relatively small acute effect on bone turnover in vivo. Postprandial hyperinsulinemia is therefore unlikely to explain the acute suppression of bone turnover observed with feeding. Secondly, hypoglycemia results in an acute suppression of bone turnover. The reduction in bone turnover during hypoglycemia may be related to hypoglycemia itself, acute changes in PTH, or other hormones released in response to hypoglycemia.
Acknowledgments
We thank the volunteers for generously participating in this study, Alison Eagleton for assisting with measurement of markers of bone turnover, the University of Newcastle for assisting with measurement of insulin, Roche Diagnostics (Penzberg, Germany) for providing reagents for the measurement of bone turnover markers and PTH, and Diabetes UK for supporting the clinical protocol.
Footnotes
This work was supported by Diabetes UK, which provided funding toward the cost of the clinical study. J.A.C. was supported by a fellowship from the National Health Service Executive.
Abbreviations: AUC, Area under the curve; OC, osteocalcin; PINP, serum procollagen type I N-terminal propeptide; sßCTX, serum Cterminal telopeptide of type I collagen.
Received October 18, 2001.
Accepted March 28, 2002.
References
This article has been cited by other articles:
![]() |
Q. Zhong, T. Itokawa, S. Sridhar, K.-H. Ding, D. Xie, B. Kang, W. B. Bollag, R. J. Bollag, M. Hamrick, K. Insogna, et al. Effects of glucose-dependent insulinotropic peptide on osteoclast function Am J Physiol Endocrinol Metab, February 1, 2007; 292(2): E543 - E548. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Galluzzi, S. Stagi, R. Salti, S. Toni, E. Piscitelli, G. Simonini, F. Falcini, and F. Chiarelli Osteoprotegerin serum levels in children with type 1 diabetes: a potential modulating role in bone status Eur. J. Endocrinol., December 1, 2005; 153(6): 879 - 885. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. B. Henriksen The Gut Feeling of Bone Remodeling IBMS BoneKEy, November 1, 2005; 2(11): 16 - 23. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |