Reply from Erica A. Wehrwein, Rita Basu, Ananda Basu, Timothy B. Curry, Robert A. Rizza and Michael J. Joyner

Abstract

We thank Dr Verberne for furthering the discussion of our recent paper. To address his points, we will focus on three issues: (1) the differential control of sympathetic outflow, (2) the role of a subset of RVLM neurons in hypoglycaemic counterregulation recently described by Drs Verberne and Sartor, and (3) a further comment on these RVLM neurons. (1) In our recent paper (Wehrwein et al. 2010) we do, for the sake of simplicity and generation of a ‘straw man’ argument, refer to the sympathetic nervous system as a unit as we discuss sympathoinhibition as a mechanism for suppression of counterregulation during hyperoxia. Dr Verberne is indeed correct that the modern view of the sympathetic nervous system is that of differential outflow to various tissues. Human microneurography studies from our collaborator Dr Wallin from the 1970s and 1980s made similar points repeatedly. For example, it is well established that there is differential activation of skin sympathetic versus muscle sympathetic nerves (Delius et al. 1972a,b,c; Wallin et al. 1975), especially during common manoeuvres such as orthostasis and exercise (Vissing et al. 1994; Vissing, 1997). A review on the differential control of muscle sympathetic nerve activity by the muscle chemoreflex (Joyner, 1992) summarizes at least some of these ideas from the human studies. These findings are also consistent with the idea of a ‘sympathetic signature’ as put forth by Osborn and Fink in their recent review of this topic as it relates to Angiotensin II hypertension (Osborn & Fink, 2010). Finally, May and colleagues support this notion by showing a disconnect between the cardiac and renal sympathetic outflows in a sheep model of heart failure (May et al. 2010). Thus the concept of differential sympathetic control is clearly important and is relevant to our findings as we study the role of the carotid bodies in the sympathetic and adrenal pathways involved in hypoglycaemic counterregulation. (2) The recent work of Verberne & Sartor (2010) describing a subset of cells in the RVLM is important and it was not cited and discussed in our paper due to timing. Verberne and Sartor demonstrate that a subpopulation of RVLM neurons: (1) serve as baro-insensitive, central glucose sensors, (2) play a role in mediating the counterregulatory response to hypoglycaemia, and (3) drive the adrenomedullary release of adrenaline. As such these cells provide an important mechanistic explanation for our findings and suggest generalized sympathoinhibition does not occur during carotid body inhibition with hyperoxia. Importantly, these RVLM cells do not have cardiac rhythmicity or baroreflex-driven inhibition and are directly targeted to the adrenal gland allowing adrenal stimulation independent of baroreflex driven changes in sympathetic outflow from presympathetic vasomotor neurons in the same brain region. However, it is unclear what the link is between the carotid body chemoreceptors and these specific RVLM cells, or how they might interact to influence the counterregulatory response to hypoglycaemia. Since both type 1 glomus cells in the carotid body and the RVLM subpopulation of neurons described by Verberne and Sartor are glucose sensitive and can mediate counterregulation, we do not yet understand how these cells interact to generate an integrated counterregulatory response. (3) It is well established that the RVLM controls sympathetic vasomotor output and is integral to baroreflex control of MSNA and blood pressure. Activation of adrenaline secretion was known to occur with RVLM disinhinition but until Verberne and Sartor's work described above (Verberne & Sartor, 2010), it was presumed that the neurons controlling sympathetic vasomotor tone and adrenaline secretion were fundamentally similar. By contrast, Verberne and Sartor show that there is a subpopulation of RVLM neurons that are baro-insensitive and whose activity is affected by changes in blood glucose levels. These glucose-sensitive RVLM neurons are targeted specifically to the adrenal medulla and control adrenaline secretion in response to a central hypoglycaemic stimulus (i.e. adrenaline release mediates liver glucose mobilization and an increase in plasma glucose to correct for hypoglycaemia). Their paper is important because it: (1) identifies a subpopulation of RVLM neurons that sense, and mediate changes in, blood glucose and are independent of the classic view of a more uniform role of RVLM neurons, (2) brings to light an important new understanding of integrative physiology and homeostasis, and (3) offers explanation for some of the differential control within the sympathoadrenal system. In conclusion, we are pleased to have this added discussion on mechanisms that might explain some of our findings. We agree with the view that a sympathoadrenal system that functions en masse is not applicable to most physiological states and are pleased that our paper has stimulated discussion on these and related topics.