Frequency-modulated oscillations of cytosolic Ca2+ ([Ca2+]c) are believed to be important in signal transduction, but it has been difficult to correlate [Ca2+]c oscillations directly with the activity of Ca(2+)-regulated targets. We have studied the control of Ca(2+)-sensitive mitochondrial dehydrogenases (CSMDHs) by monitoring mitochondrial Ca2+ ([Ca2+]m) and the redox state of flavoproteins and pyridine nucleotides simultaneously with [Ca2+]c in single hepatocytes. Oscillations of [Ca2+]c induced by IP3-dependent hormones were efficiently transmitted to the mitochondria as [Ca2+]m oscillations. Each [Ca2+]m spike was sufficient to cause a maximal transient activation of the CSMDHs and [Ca2+]m oscillations at frequencies above 0.5 per minute caused a sustained activation of mitochondrial metabolism. By contrast, sustained [Ca2+]c increases yielded only transient CSMDH activation, and slow or partial [Ca2+]c elevations were ineffective in increasing [Ca2+]m or stimulating CSMDHs. We conclude that the mitochondria are tuned to oscillating [Ca2+]c signals, the frequency of which can control the CSMDHs over the full range of potential activities.

Specialized neurons utilize glucose as a signaling molecule to alter their firing rate. Glucose-excited (GE) neurons increase and glucose-inhibited (GI) neurons reduce activity as ambient glucose levels rise. Glucoseinduced changes in the ATP-to-ADP ratio in GE neurons modulate the activity of the ATP-sensitive K ؉ channel, which determines the rate of cell firing. The GI glucosensing mechanism is unknown. We postulated that glucokinase (GK), a high-Michaelis constant (K m ) hexokinase expressed in brain areas containing populations of GE and GI neurons, is the controlling step in glucosensing. Double-label in situ hybridization demonstrated neuron-specific GK mRNA expression in locus ceruleus norepinephrine and in hypothalamic neuropeptide Y, pro-opiomelanocortin, and ␥-aminobutyric acid neurons, but it did not demonstrate this expression in orexin neurons. GK mRNA was also found in the area postrema/nucleus tractus solitarius region by RT-PCR. Intracarotid glucose infusions stimulated c-fos expression in the same areas that expressed GK. At 2.5 mmol/l glucose, fura-2 Ca 2؉ imaging of dissociated ventromedial hypothalamic nucleus neurons demonstrated GE neurons whose intracellular Ca 2؉ oscillations were inhibited and GI neurons whose Ca 2؉ oscillations were stimulated by four selective GK inhibitors. Finally, GK expression was increased in rats with impaired central glucosensing (posthypoglycemia and diet-induced obesity) but was unaffected by a 48-h fast. These data suggest a critical role for GK as a regulator of glucosensing in both GE and GI neurons in the brain. S pecialized neurons utilize glucose as a signaling molecule rather than as an energy substrate. Glucose-excited (GE) neurons increase, whereas glucose-inhibited (GI) neurons decrease, their firing rate as ambient glucose levels rise (1-3). It is unclear how GI neurons sense glucose, but GE neurons function much like the pancreatic ␤-cell, which utilizes the ATPsensitive K ϩ (K ATP ) channel to sense glucose and regulate insulin secretion (2-7). Although many neurons contain K ATP channels, few exhibit glucosensing properties (8,9). This makes control of glycolysis a major candidate as a regulator of ATP production and K ATP channel activity (6). The pancreatic form of glucokinase (GK; hexokinase IV, ATP:D-glucose 6-phosphotransferase) is selectively expressed in brain areas where glucosensing neurons reside (5,10 -12). GK has the physiological properties that would make it an ideal regulator of glucosensing in neurons (6,13). Previous studies examined a role for GK in hypothalamic glucosensing neurons in brain slice preparations, where presynaptic inputs could not be excluded, and under conditions outside the physiological range of brain glucose (4,5). The current studies provide new data on the neurotransmitter and peptide phenotype of neurons expressing GK, the regulatory role of GK in isolated hypothalamic GE and GI neuron activity, and the in vivo conditions associated with altered brain GK expression. RESEARCH DESIGN AND METHODSAnimals. All ...

To evaluate potential mechanisms for neuronal glucosensing, fura-2 Ca 2؉ imaging and single-cell RT-PCR were carried out in dissociated ventromedial hypothalamic nucleus (VMN) neurons. Glucose-excited (GE) neurons increased and glucose-inhibited (GI) neurons decreased intracellular Ca 2؉ ([Ca 2؉ ] i ) oscillations as glucose increased from 0.5 to 2.5 mmol/l. The Kir6.2 subunit mRNA of the ATP-sensitive K ؉ channel was expressed in 42% of GE and GI neurons, but only 15% of nonglucosensing (NG) neurons. Glucokinase (GK), the putative glucosensing gatekeeper, was expressed in 64% of GE, 43% of GI, but only 8% of NG neurons and the GK inhibitor alloxan altered [Ca 2؉ ] i oscillations in ϳ75% of GK-expressing GE and GI neurons. Insulin receptor and GLUT4 mRNAs were coexpressed in 75% of GE, 60% of GI, and 40% of NG neurons, although there were no statistically significant intergroup differences. Hexokinase-I, GLUT3, and lactate dehydrogenase-A and -B were ubiquitous, whereas GLUT2, monocarboxylate transporters-1 and -2, and leptin receptor and GAD mRNAs were expressed less frequently and without apparent relationship to glucosensing capacity. Thus, although GK may mediate glucosensing in up to 60% of VMN neurons, other regulatory mechanisms are likely to control glucosensing in the remaining ones. Diabetes 53:549 -559, 2004

Cytosolic Ca2+ signals are often organized in complex temporal and spatial patterns, even under conditions of sustained stimulation. In this review we discuss the mechanisms and physiological significance of this behavior in nonexcitable cells, in which the primary mechanism of Ca2+ mobilization is through (1,4,5)IP3-dependent Ca2+ release from intracellular stores. Oscillations of cytosolic free Ca2+ ([Ca2+]i) are a common form of temporal organization; in the spatial domain, these [Ca2+]i oscillations may take the form of [Ca2+]i waves that propagate throughout the cell or they may be restricted to specific subcellular regions. These patterns of Ca2+ signaling result from the limited range of cytoplasmic Ca2+ diffusion and the feedback regulation of the pathways responsible for Ca2+ mobilization. In addition, the spatial organization of [Ca2+]i changes appears to depend on the strategic distribution of Ca2+ stores within the cell. One type of [Ca2+]i oscillation is baseline spiking, in which discrete [Ca2+]i spikes occur with a frequency, but not amplitude, that is determined by agonist dose. Most current evidence favors a model in which baseline [Ca2+]i spiking results from the complex interplay between [Ca2+]i and (1,4,5)IP3 in regulating the gating of (1,4,5)IP3-sensitive intracellular Ca2+ channels. Sinusoidal [Ca2+]i oscillations represent a mechanistically distinct type of temporal organization, in which agonist dose regulates the amplitude but has no effect on oscillation frequency. Sinusoidal [Ca2+]i oscillations can be explained by a negative feedback effect of protein kinase C on the generation of (1,4,5)IP3 at the level of phospholipase C or its activating G-protein. The physiological significance of [Ca2+]i oscillations and waves is becoming more established with the observation of this behavior in intact tissues and by the recognition of Ca2+-dependent processes that are adapted to respond to frequency-modulated oscillatory [Ca2+]i signals. In some cells, these [Ca2+]i signals are targeted to control processes in limited cytoplasmic domains, and in other systems [Ca2+]i waves can be propagated through gap junctions to coordinate the function of multicellular systems.

Stimulation of hepatocytes with vasopressin evokes increases in cytosolic free Ca2+ ([Ca2+]c) that are relayed into the mitochondria, where the resulting mitochondrial Ca2+ ([Ca2+]m) increase regulates intramitochondrial Ca2+-sensitive targets. To understand how mitochondria integrate the [Ca2+]c signals into a final metabolic response, we stimulated hepatocytes with high vasopressin doses that generate a sustained increase in [Ca2+]c. This elicited a synchronous, single spike of [Ca2+]m and consequent NAD(P)H formation, which could be related to changes in the activity state of pyruvate dehydrogenase (PDH) measured in parallel. The vasopressin-induced [Ca2+]m spike evoked a transient increase in NAD(P)H that persisted longer than the [Ca2+]m increase. In contrast, PDH activity increased biphasically, with an initial rapid phase accompanying the rise in [Ca2+]m, followed by a sustained secondary activation phase associated with a decline in cellular ATP. The decline of NAD(P)H in the face of elevated PDH activity occurred as a result of respiratory chain activation, which was also manifest in a calcium-dependent increase in the membrane potential and pH gradient components of the proton motive force (PMF). This is the first direct demonstration that Ca2+-mobilizing hormones increase the PMF in intact cells. Thus, Ca2+ plays an important role in signal transduction from cytosol to mitochondria, with a single [Ca2+]m spike evoking a complex series of changes to activate mitochondrial oxidative metabolism.