Abstract
Cardiovascular diseases account for 80% of the premature mortality amongst individuals with Type 2 Diabetes Mellitus (DM). Diabetes is associated with detrimentally higher sympathetic drive to the heart and with disrupted circadian rhythms in autonomic function. Together, these phenomena impair the autonomic control of the heart, leading to a complication known as Cardiac Autonomic Neuropathy (CAN). Because pharmacotherapies for diabetic CAN are based on extrapolated findings from studies in other cardiovascular diseases, they tend to be less effective in diabetic than healthy individuals. This necessitates an alternative approach to managing diabetic CAN. Pharmacological agents used in treatment interventions for diabetic CAN almost exclusively target the peripheral manifestations of the higher cardiac sympathetic nerve activity rather than its unknown central origins. Therefore, the current project was designed to determine central neuronal activation patterns in brain regions involved in cardiac sympathoregulation in DM.
The rostral ventrolateral medulla (RVLM) is the final integrative site for differential sympathetic outflow from the brain and it receives inputs from sympathoregulatory regions such as the hypothalamic paraventricular nucleus (PVN), the nucleus tractus solitarius (NTS) and the area postrema (AP). However, it is not known whether the RVLM drives the cardiac sympathoexcitation in DM and what inputs to the RVLM might increase neuronal activation in the RVLM in DM. Double-label immunohistochemistry for ΔFosB (a marker of chronic neuronal activation) and markers of neuronal phenotypes in 20-week-old male DM Zucker Diabetic Fatty (ZDF) rats was used to test the hypothesis that higher activation of PVN, NTS, and AP neuronal projections to the RVLM is responsible for the higher cardiac sympathetic drive in DM. The results showed that more noradrenergic RVLM neurones co-expressed ΔFosB in DM ZDF rats (9 ± 1 neurones) than non-diabetic (nDM) ZDF rats (3 ± 0 neurones, P < 0.001, unpaired t - test). DM rats exhibited higher ΔFosB expression in the PVN (P < 0.05), NTS (P < 0.05) and AP (P < 0.01). ΔFosB expression was higher in PVN corticotropin-releasing hormone (CRH) neurones in colchicine-treated DM compared to nDM rats (P < 0.05). The axonal transport blocker colchicine was administered to enable easier visualisation of CRH. Retrograde labelling studies showed that the numbers of RVLM-projecting CRH-, oxytocin- or vasopressin-expressing PVN neurones that also expressed ΔFosB were not different between nDM and DM rats. Likewise, the numbers of RVLM-projecting noradrenergic NTS and AP neurones that also expressed ΔFosB were not different between nDM and DM rats. These data indicate that afferent inputs from the PVN, NTS or AP do not underlie the higher activation of noradrenergic RVLM neurones in DM.
Obesity is often associated with both DM and higher sympathetic nerve activity. The DM ZDF rats used in the current study also exhibit hyperphagia-induced obesity. Therefore, immunohistochemistry for ΔFosB in sympathoexcitatory energy homeostasis centres was carried out to test the prediction that higher neuronal activation in hypothalamic energy balance regions is observed in DM ZDF rats. The data revealed lower neuronal activation in the arcuate nucleus, ventromedial hypothalamus and dorsomedial hypothalamus, suggesting that obesity is unlikely to underlie the cardiac sympathoexcitation in the ZDF rat model of DM.
Physiological and behavioural functions exhibit 24-hour circadian rhythms that are largely disrupted in DM. Specifically, patients with DM fail to exhibit normal nocturnal dips in sympathetic nerve activity and heart rate. These circadian misalignments are promising therapeutic targets for diabetic CAN. Circadian variations in neuronal activation in the RVLM and NTS are likely to reflect circadian fluctuations in cardiac efferent and afferent sympathetic nerve activities respectively. Double-label immunohistochemistry for c-fos (a marker of acute neuronal activation) was used to test the hypothesis that nocturnal dipping in the activation of noradrenergic RVLM and NTS neurones is absent in DM ZDF rats. A 2-way analysis of variance (ANOVA) for the number of noradrenergic RVLM neurones that also expressed c-fos showed a significant effect of time (active phase > inactive phase; P < 0.01) and diabetic status (DM > nDM; P < 0.001). The numbers of noradrenergic NTS neurones that also expressed c-fos showed a significant effect of diabetic status (DM > nDM; P < 0.001, 2-way ANOVA). However, the effect of time of day on c-fos expression in noradrenergic RVLM and NTS neurones was not dependent on diabetic status (no interaction, 2-way ANOVA). These data support the suitability of 12- and/or 24-hour drug release systems in the treatment of diabetic CAN. The primary goal of the chronopharmacotherapy should focus on stabilising cardiac sympathetic nerve activity during the active phase.
Previous research on diabetic CAN has narrowly focused on pathologies at the heart without considering the interplay with central neural input. The current data show for the first time that DM is associated with altered neuronal activation patterns in sympathoexcitatory brain regions during the inactive and active phases. However, circadian fluctuations in neuronal activation in the regions studied are not dysregulated in DM. There is a palpable need for further research into the neuronal circuit(s) underlying the higher cardiac sympathetic drive in DM. This warrants retrograde tracer injections into the intermediolateral cell column of the upper thoracic spinal cord (from where cardiac sympathetic nerves originate) combined with immunohistochemistry in central pre-sympathetic regions such as the RVLM, PVN, and dorsal raphe nucleus.