We studied the mechanisms and physiological relevance of the cephalic insulin response to meal ingestion in 12 healthy women (age 63 ± 0.4 years; BMI 27.7 ± 1.7 kg/m). The ganglionic antagonist, trimethaphan, which impairs neurotransmission across parasympathetic and sympathetic autonomic ganglia, or atropine or saline was given intravenously during the first 15 min after ingestion of a standard meal (350 kcal). During saline infusion, insulin levels increased during the first 10 min after meal ingestion, whereas the first increase in glucose was evident at 15 min. The preabsorptive 10-min insulin response was reduced by 73 ± 11% by trimethaphan (P = 0.009), accompanied by impaired reduction of glucose levels from 25 to 60 min after meal ingestion (
glucose = 1.27 ± 0.5 [with saline] vs. 0.1 ± 0.4 mmol/l [with trimethaphan]; P = 0.008). This reduction at 2560 min in glucose levels correlated significantly to the 10-min insulin response (r = 0.65, P = 0.024). The 10-min insulin response to meal ingestion was also reduced by atropine, but only by 20 ± 9% (P = 0.045), which was lower than the reduction with trimethaphan (P = 0.004). The preabsorptive insulin response was not accompanied by any increase in circulating levels of gastric inhibitory polypeptide (GIP) or glucagon-like peptide 1 (GLP-1). In conclusion, 1) the early preabsorptive insulin response to meal ingestion in humans can be largely attributed to autonomic activation mediated by noncholinergic and cholinergic mechanisms, 2) this cephalic insulin response is required for a normal postprandial glucose tolerance, and 3) GIP and GLP-1 do not contribute to the preabsorptive cephalic phase insulin response to meal ingestion.
The preabsorptive or cephalic phase insulin response, which lasts for ~10 min, is initiated by meal ingestion, as has been demonstrated in humans and rats . It is abolished by vagotomy in rats and by atropine in rats and humans , suggesting mediation by cholinergic mechanisms. However, the mechanism is probably more complex than that executed solely by cholinergic actions on islet ß-cells; noncholinergic mechanisms might also contribute to the response. For example, islet parasympathetic nerves harbor several neuropeptides in addition to acetylcholine, such as vasoactive intestinal polypeptide (VIP), pituitary adenylate cyclase-activating polypeptide (PACAP), and gastrin-releasing polypeptide (GRP) . These neuropeptides are released after vagal nerve activation of the pancreas and stimulate insulin secretion . Prevention of their effects inhibits the insulin response to oral administration of glucose in mice . It is also possible that the gut hormones, gastric inhibitory polypeptide (GIP) and glucagon-like peptide 1 (GLP-1), contribute to the cephalic phase insulin response to meal intake, as both these hormones are potent insulin secretory hormones released during meal ingestion and their secretion is under neural control . Finally, there is a problem in the interpretation of atropine data in previous studies, as atropine reduces postprandial glycemia, thereby also indirectly influencing insulin secretion . Therefore, the mechanism of the cephalic phase insulin response to meal ingestion has not been established in humans.
It has recently been established that the early insulin response to meal ingestion is of great importance for subsequent glucose tolerance. This was first suggested by results demonstrating a negative correlation between the 30-min insulin response to oral glucose, as a marker for early insulin secretion, and the 120-min glucose value, as a marker of glucose tolerance . Furthermore, prevention of the early insulin response by somatostatin results in glucose intolerance , and sham feeding, which increases circulating insulin, improves glucose tolerance after intragastric glucose . The importance of the early insulin response for postprandial glucose tolerance is also illustrated by studies reporting that brief administration of a minute amount of insulin during the first 15 min after food intake markedly improves glucose tolerance in obese and type 2 diabetic subjects . Whether specifically the neurally mediated cephalic insulin response to meal ingestion is of importance for postprandial glucose homeostasis has, however, not been established.
The current study was designed to explore the mechanism and significance of the preabsorptive insulin response to meal ingestion in humans. Healthy overnight fasted subjects were given a standard meal in the presence of the ganglionic blocker, trimethaphan, or the muscarinic antagonist, atropine. Trimethaphan inhibits neurotransmission across autonomic ganglia and therefore blocks the neural influences of islet function. Its usefulness for studies of neural regulation of islet function in humans was shown in a previous study, in which we explored the neural contribution to the islet hormone responses to insulin-induced hypoglycemia . By comparing the results obtained with trimethaphan with those obtained with saline, the contribution of neural influences to the insulin response to meal ingestion could be established; furthermore, by comparing results with trimethaphan with those of atropine, the relative contribution of cholinergic versus noncholinergic neural mechanisms could be established. We also determined the GIP and GLP-1 responses to meal ingestion with or without administration of trimethaphan or atropine to examine whether any cephalic phase response of these incretins might contribute to the cephalic phase insulin response. Finally, by examining glycemia in the presence or absence of trimethaphan, the importance of the neurally mediated insulin response to meal ingestion for postprandial glycemia could be established.