We also observed reduced insulinotropic
We also observed reduced insulinotropic effects after intravenous administration of both GLP-1 and GIP with atropine, whereas no inhibition was observed by atropine of the direct effects of GLP-1 and GIP action on beta-cells in the islet experiments. This suggests that the reduction in circulating insulin after pretreatment by atropine is not a direct effect on the islet cells. Instead, the mechanism is more likely related to vagus nerve activity prevented by atropine in other locations. For GLP-1, this is in line with previous studies on vagus nerve activation in the portal vein (Nishizawa et al., 1996) and the fact that vagotomy of the hepatic branch of the vagus nerve in rats reduces the effects of portal GLP-1 on insulin secretion (Fujiwara et al., 2012). For GIP, the involvement of the vagus nerve in the effects of GIP has not previously been shown. Although there are studies showing lack of effect of portal GIP to activate vagal nerve fibers (Nishizawa et al., 1996), receptors for GIP have been localized in nerves in the Heparin sodium (Nyberg et al., 2005, Nyberg et al., 2007), which suggests a possibility for alternative routes of vagal nerve activation. We cannot, however, rule out that GIP stimulates the release of another factor that in turn is able to activate vagal nerve fibers, however, this needs further investigation. Another possibility of GLP-1 and GIP to activate the vagus nerve terminals is within the islets, which could tentatively affect intra-islet vessel dilatation (Dai et al., 2013), although the presence of incretin hormone receptors in these nerve fibers remains to be studied. It has previously been shown that the parasympathetic nerves stimulate the production of the liver substance HISS, which promotes insulin action in skeletal muscle (Lautt, 1999, Lautt, 2005). This, in combination with the role of the incretin hormones to stimulate vagal activity, may potentially be of relevance for an effect of incretin hormones also on insulin sensitivity, which needs to be studied in further work. Circulating glucose was not significantly affected after DPP-4 inhibition in the DGTT even though insulin levels were augmented, suggesting that the raise of insulin was short lived or insufficient in magnitude to reduce glucose in this model. When GLP-1 or GIP was given together with intravenous glucose insulin levels were raised by a higher degree and then glucose levels were lower than after glucose alone and even was reduced below baseline at 20 and 30min after administration, as is evident by comparing Fig. 4, Fig. 5. Nevertheless, atropine, although reducing insulin when given with the incretin hormones did not alter glucose levels, again suggesting that the magnitude of the insulin change was insufficient to be translated to an effect on glycemia. A strength of this study is that we used the newly developed DGTT, since an effect through changes in gastric emptying is thus bypassed. A limitation of the study, however, is that muscarinic receptors are expressed in many metabolically important tissues (Eglen et al., 1996, Thiele, 2013). A surprising effect that circulating glucose was slightly, although significantly, increased at 20 and 30min after intravenous glucose administration following atropine administration even though insulin levels were not altered might support such islet-independent mechanisms. This would suggest that atropine by some mechanism prevents the glucose uptake after intravenous glucose which would support a recent report that muscarinic stimulation stimulates glucose uptake in skeletal muscle cells (Merlin et al., 2010). However, whether such a mechanism would explain the effect of atropine on circulating after intravenous glucose remains to be studied. Nevertheless, since we had a clear reduction in the insulin response to intravenous carbachol but no effects in the insulin response to intravenous glucose, non-specific effects of atropine do probably play only a minor role, if any, in the islet effects in the current experiments.