The rate of fall of blood glucose determines the necessity of forebrain-projecting catecholaminergic neurons for male rat sympathoadrenal responses

Abstract

Different onset rates of insulin-induced hypoglycemia use distinct glucosensors to activate sympathoadrenal counterregulatory responses (CRRs). Glucosensory elements in the portal-mesenteric veins are dispensable with faster rates when brain elements predominate, but are essential for responses to the slower-onset hypoglycemia that is common with insulin therapy. Whether a similar rate-associated divergence exists within more expansive brain networks is unknown. Hindbrain catecholamine neurons distribute glycemia-related information throughout the forebrain. We tested in male rats whether catecholaminergic neurons that project to the medial and ventromedial hypothalamus are required for sympathoadrenal CRRs to rapid- and slow-onset hypoglycemia and whether these neurons are differentially engaged as onset rates change. Using a catecholamine-specific neurotoxin and hyperinsulinemic-hypoglycemic clamps, we found that sympathoadrenal CRRs to slow- but not rapid-onset hypoglycemia require hypothalamus-projecting catecholaminergic neurons, the majority of which originate in the ventrolateral medulla. As determined with Fos, these neurons are differentially activated by the two onset rates. We conclude that 1) catecholaminergic projections to the hypothalamus provide essential information for activating sympathoadrenal CRRs to slow- but not rapid-onset hypoglycemia, 2) hypoglycemia onset rates have a major impact on the hypothalamic mechanisms that enable sympathoadrenal CRRs, and 3) hypoglycemia-related sensory information activates hindbrain catecholaminergic neurons in a rate-dependent manner.

Figure 1

Fluorescence photomicrographs of coronal sections through the hypothalamus of rats showing catecholaminergic innervation revealed with DBH immunocytochemistry. Animals were injected in the PVH with either a MIgG-SAP (A and B) or a DSAP (C and D). Sections correspond to levels 26 (A and C) and 28 (B and D) of Swanson (24). Note the profound loss of catecholaminergic innervation in the mediobasal hypothalamus of the animal injected with DSAP (C and D) compared with that injected with MIgG-SAP (A and B). 3V, third ventricle; AHN, anterior hypothalamic nucleus; ARH, arcuate nucleus; fx, fornix; ic, internal capsule; LHA, lateral hypothalamic area; ot, optic tract; RCH, retrochiasmatic area; TU, tuberal nucleus; VMH, ventromedial nucleus of the hypothalamus. Scale bar = 500 μm.

Figure 2

Animals injected in the PVH with a DSAP show a significant loss of DBH (A and C) and PNMT (B and D) immunoreactive neurons in the A1, A1/C1, and C1 regions of the VLM compared with animals injected with a MIgG-SAP. Animals were exposed to either rapid- (open bars) or slow-onset (solid bars) hyperinsulinemic-hypoglycemic clamp. Results in A and B are expressed as mean (SEM) number of labeled cells per section. C and D show DBH and PNMT in the A1 (DBH only) and C1 regions of the VLM. Scale bar = 100 μm.

Figure 3

Animals injected in the PVH with a DSAP show a significant loss of DBH (A and Cb) and PNMT (B and Db) immunoreactive neurons in the AP and NTS compared with animals injected with a MIgG-SAP (Ca and Da). Animals were exposed to either rapid-onset (open bars) or slow-onset (solid bars) hyperinsulinemic-hypoglycemic clamp. Results in A and B are expressed as mean (SEM) number of labeled cells per section. ns, not significant; c, central canal, spinal cord/medulla; co, commisural part of the NTS; ge, gelatinous part of the NTS; m, medial part of the NTS; ts, solitary tract. Scale bar = 200 μm.

Figure 4

Sympathoadrenal hormone responses to rapid- and slow-onset IIH. Panels A and B show the mean (SEM) arterial glucose concentrations in animals injected in the PVH with either a MIgG-SAP (solid circles) or a DSAP (open circles) and exposed to a rapid- or slow-onset hyperinsulinemic-hypoglycemic clamp. C: Corresponding glucose infusion rates. Plasma epinephrine (D–F) and norepinephrine (G–I) responses to the hyperinsulinemic-hypoglycemic clamps. F and I show the plasma epinephrine and norepinephrine concentrations at 0 min (open bars) and after 60 min (solid bars) of hypoglycemia. In all panels except F and I, the light gray area denotes the onset duration of hypoglycemia and the dark gray area denotes the duration of hypoglycemia (2.5 mmol/L). *P < 0.02; **P < 0.001; ***P < 0.0002; ****P < 0.0001; ns, not significant.

Figure 5

Mean (SEM) numbers of Fos (A and C) and Fos/DBH (B and D) colocalized neurons per section in the A1, A1/C1, and C1 regions of the VLM (A and B) and dorsal medulla (C and D). Animals were injected in the PVH with either a MIgG-SAP or a DSAP and exposed to a rapid-onset (open bars) or slow-onset (solid bars) hyperinsulinemic-hypoglycemic clamp. ns, not significant.

Figure 6

Confocal photomicrographs showing DBH (green fluorophore) and Fos (red fluorophore) immunoreactive elements in the C1 region of the VLM (A–D) and dorsal medulla (E–H) of animals injected in the PVH with either a MIgG-SAP or a DSAP. Animals were exposed to either rapid- or slow-onset hyperinsulinemic-hypoglycemic clamp. The white rectangle in each main panel is the region shown at higher magnification in the accompanying panel. The arrowheads in A–D indicate representative DBH/Fos double-labeled neurons (solid) and DBH or Fos singly labeled neurons (open). The brightness and contrast of each panel was adjusted uniformly to achieve balance across the entire panel using combinations of the brightness/contrast and curves (γ-correction) tools in Adobe Photoshop CS3 (Adobe Inc.) Please note that brightness/contrast manipulations were only used for the photomicrographs and were not performed on images from which the data used to derive the quantitative results shown in Figs. 2, 3, and 5 were obtained. None of the manipulations changed the nature of the results. c, central canal, spinal cord/medulla; co, commisural part of the NTS; ge, gelatinous part of the NTS; m, medial part of the NTS; ts, solitary tract; XII, hypoglossal nucleus. Scale bars = 100 μm (A–D); scale bars = 200 μm (E–H).