Islets Have a Lot of Nerve! Or Do They?

Abstract

The autonomic nervous system influences insulin and glucagon secretion. In this issue, Rodriguez-Diaz et al. (2011) show that mouse and human islets differ in their innervation patterns, yet the effect of neural activation on islet hormone secretion is similar. Key questions raised by this species difference have potential relevance to diabetic therapeutics.

Neural Control of the Islet

Glucagon and insulin are secreted from islet alpha and beta cells, respectively, and have major effects on metabolism. Circulating metabolites in turn control the secretion of these hormones. The landmark review of Woods and Porte (Woods and Porte, 1974) summarized the early evidence that autonomic nerves also influence pancreatic hormone secretion. Later, animal studies established that sympathetic nerves inhibit insulin and stimulate glucagon secretion and that parasympathetic nerves markedly stimulate insulin and modestly stimulate glucagon secretion (Ahrén, 2000). These effects were assumed to be due to the release of neurotransmitters directly onto alpha and beta cells. However, the finding of Rodriguez-Diaz et al. (2011) that islet nerves in humans do not contact alpha or beta cells suggests that neural regulation of islet hormone secretion is not so straightforward.

The sophisticated and detailed studies of Rodriguez-Diaz et al. (2011) advance our understanding of the neuroanatomy of the islet in two major ways. First, they determine whether islet nerves actually contact blood vessels, alpha cells, or beta cells by coupling immunohistochemistry for neural, islet cell, and blood vessel markers with confocal microscopy. Second, they provide the first detailed description of the neuroanatomy of the human islet based on high-quality samples of human pancreas, which gives credence to their conclusions. In the mouse islet they find that sympathetic fibers preferentially innervate alpha cells and that parasympathetic fibers innervate alpha cells and beta cells equally. In marked contrast, they find in the human islet that sympathetic fibers preferentially innervate central islet blood vessels and that parasympathetic fibers are few and far between, a conclusion hedged somewhat by the extensive coexpression of their parasympathetic marker in islet endocrine cells.

The Physiologic Role of Islet Nerves

Although the animal studies mentioned above demonstrate that nerves can influence islet hormone secretion, they did not establish when they actually do so. Studies measuring norepinephrine spillover demonstrated that pancreatic sympathetic nerves were activated by hypoglycemic stress, but not by either hypoxic or hypotensive stress (Havel et al., 1988). Analogous measurements of acetylcholine spillover from parasympathetic nerves are rarely possible because of avid and ubiquitous acetylcholinesterases. Fortunately, pancreatic polypeptide, secreted from the islet F cell, was found to be under strong parasympathetic control (Schwartz et al., 1978). Use of this surrogate established that pancreatic parasympathetic nerves are activated during both hypoglycemia and feeding. Studies using autonomic blockade or ablation established autonomic mediation of the glucagon response to severe hypoglycemia (Taborsky et al., 1998). This increased glucagon secretion helps to restore euglycemia by stimulating hepatic glucose production. Studies using vagotomy or cholinergic antagonists suggested that parasympathetic nerves help mediate the very early insulin response to feeding (Ahrén, 2000), which primes the liver to store the incoming glucose load.

Human versus Animal Studies

The finding that the endocrine cells of the human islet are not directly innervated raises the question of whether the conclusions derived from animal studies apply to humans. However, three representative studies of neural control of the islet in humans produce results similar to those in animals. First, the insulin response to the cephalic phase of feeding (Teff et al., 1991) is parasympathetically mediated in humans, as it is in animals (Ahrén, 2000). Second, autonomic blockade eliminates 75% of the glucagon response to insulin-induced hypoglycemia in humans (Havel and Ahrén, 1997), just as it does in animals (Taborsky et al., 1998). Third, activation of sympathetic nerves inhibits insulin secretion in humans (Gilliam et al., 2007), as it does in animals (Ahrén, 2000). Given the differences in islet innervation between mice and humans, it is likely that animals and humans use different mechanisms to achieve similar autonomic effects on pancreatic hormone secretion.

Mechanisms of Neural Control

In mice it is likely that activation of islet sympathetic nerves inhibits insulin secretion by releasing norepinephrine directly onto the islet beta cell. In contrast, the authors suggest two indirect mechanisms for this inhibition of insulin in humans: (1) a decrease of islet blood flow or (2) a spillover of norepinephrine into the arteriole that perfuses the islet. Although a decrease of islet blood flow could explain a decrease of insulin secretion, it cannot easily explain a simultaneous increase of glucagon secretion, like that seen in animals. Regarding the norepinephrine spillover mechanism, there is precedent in another organ under sympathetic control, the liver. Low frequencies of sympathetic nerve stimulation, which are insufficient to increase hepatic norepinephrine spillover, do not increase hepatic glucose production. They do decrease hepatic arterial blood flow, suggesting that the norepinephrine is released at the arteriole. In contrast, frequencies of nerve stimulation high enough to increase norepinephrine spillover do stimulate hepatic glucose production (Mundinger et al., 1997). These data suggest that sympathetic nerves can influence downstream targets by releasing their neurotransmitters into the perfusing arteriole.

In contrast, it is unlikely that acetylcholine spillover mediates the effect of parasympathetic activation to stimulate insulin secretion in humans because of their rapid degradation by acetycholinesterases. However, vagal-mediated release of gastrointestinal hormones could stimulate insulin release indirectly.

Implications and Future Studies

The discovery and elucidation of these indirect mechanisms of neural control of the human islet will yield new insight into modulation of alpha and beta cell function, with the potential to influence future diabetic therapeutics. For example, understanding how autonomic activation mediates the glucagon response to hypoglycemia in humans may provide a way to correct the impaired response seen in type 1 diabetes. Treatment-induced hypoglycemia, which is a major barrier to patient compliance with intensive insulin therapy, may thus be avoided. Likewise, understanding how activation of the parasympathetic nervous system helps mediate early insulin responses to feeding in humans may provide a way to augment these responses and thereby avoid the glucose intolerance which precedes type 2 diabetes. Thus, the elegant studies of Rodriguez-Diaz et al. are not only a major advance for the academic field of neural control of the islet, but also raise heretofore unrecognized questions whose answers may impact the treatment of diabetes.