The effect of Bordetella pertussis on lymphocyte cyclic AMP metabolism.

Bordetella pertussis culture fractions produce decreased metabolic responses to isoproterenol and epinephrine in mice and rats, suggesting the possibility of systemic beta adrenergic blockade. The present study was undertaken to elucidate the mechanism of the alteration in adrenergic responsiveness and to clarify its relationship to other biological effects of the organism. Lymphocytes were selected as a suitable tissue because of the marked alteration in lymphocyte distribution in pertussis-treated mice and rats, suggesting a change in the surface properties of these cells. Human peripheral blood lymphocytes, purified by nylon fiber chromatography, were studied. In short incubation experiments (20 min or less) B. pertussis did not alter the cyclic AMP response to isoproterenol, prostaglandin E (PGE(1)), or methacholine. However, when cells were preincubated with B. pertussis for 90 min at 37 degrees C, the responses to all three agents were markedly inhibited. Although these observations provide direct confirmation of the ability of B. pertussis to inhibit catecholamine responsiveness, the fact that PGE(1) and methacholine responses were also inhibited suggests that blockade at the level of the beta adrenergic receptor is doubtful. The inhibitory activity was localized in a nondialyzable, protein-rich fraction that is precipitated from B. pertussis culture fluid by ammonium sulfate at 90% of saturation. The bulk of the activity was obtained in the load volume after 50,000 g centrifugation in a cesium chloride gradient, density 1.2-1.5 (fraction 4). Fraction 4 produced a change in lymphocyte hormonal responsiveness at concentrations as low as 5 ng/ml. The relationship between cyclic AMP inhibitory activity in isolated human cells and leukocytosis-producing activity in intact mice was studied. The two activities seemed to parallel one another quite closely until the final Sephadex G-150 fractionation step, in which the two activities were obtained in the same column fraction, but a greater recovery of the leukocytosis-producing activity was obtained. Additional purification will be required to establish conclusively whether the same macromolecule is responsible for both activities. The availability of a bacterial product that markedly inhibits cyclic AMP accumulation in purified lymphocytes may help to clarify the role of cyclic AMP in lymphocyte activation by antigen and nonspecific mitogens.

Cholinesterase activity of nodal and internodal regions of myelinated nerve fibers of frog.

The distribution of cholinesterase (Ch-esterase) in isolated myelinated fibers of the frog has been investigated. Quantitative microgasometric measurements have confirmed the previous histochemical observations. Both approaches indicate that in frog nerve fibers acetylcholinesterase (ACh-esterase) is the only or the predominant enzyme when selective inhibitors and different substrates are used: acetylcholine (ACh), butyrylcholine, and acetyl-B-methylcholine (Mecholyl). By means of the microgasometric technique, a significant difference in ACh-esterase activity between axons isolated from ventral (37.2 +/- 1.7 micromole x 10(-5) ACh/mm(2)/hr) and dorsal roots (2.0 +/- 0.9 micromole x 10(-5) ACh/mm(2)/hr) was found. In the region of the node of Ranvier the enzyme activity (50.4 +/- 4.4 micromole x 10(-5) ACh/mm(2)/hr) appears to be considerably higher than in the internodal area (36.6 +/- 2.1 micromole x 10(-5) ACh/mm(2)/hr). The findings are discussed in relation to the theory of saltatory conduction and the ACh system.

Cholinergic sensitivity: normal variability as a function of stimulus background.

The sensitivity of the normally innervated iris sphincter to its neuro-transmitter, acetylchloline, and to relatd agents varies inversely with the preexisting physiological stimulus background, that is, the environmental light intnsity. This normal variability suggests the existence of a negative feedback mechahnism whereby sensitivity of the effector cell is modutlated by a product of neuronal activity.

Synaptic transmission: long-lasting potentiation by a postsynaptic mechanism.

Slow decreases of ionic conductance across neuronal cell membranes, which generate slow synaptic potentials, can increase the effectiveness of synaptic transmission. Slow conductance decreases sufficient magnitude increase the amplitude of monosynaptic fast excitatory postsynaptic potentials in B cells of the bullfrog sympathetic ganglion. By this postsynaptic mechanism, activation of one synaptic pathway can cause an increase in transmission, lasting several minutes, across another synapse. This may provide an important mechanism for synaptic integration and control of neuronal interaction.

Depolarization and muscarinic excitation induced in a sympathetic ganglion by vasoactive intestinal polypeptide.

The effects of vasoactive intestinal polypeptide (VIP) in the superior cervical ganglion of the cat were studied in vitro and in vivo with sucrose gap and multiunit recording, respectively. At a dose of 0.03 to 0.12 nanomole, VIP produced a dose-dependent, prolonged (3 to 15 minutes) depolarization of the ganglion and enhanced the ganglionic depolarization elicited by the muscarinic agonist acetyl-beta-methylcholine. At a dose of 1.8 to 10 nanomoles, the peptide enhanced and prolonged the postganglionic discharge elicited by acetyl-beta-methylcholine, enhanced muscarinic transmission in ganglia treated with an anticholinesterase agent, and enhanced the late muscarinic discharge elicited by acetylcholine. VIP did not affect the early nicotinic discharge elicited by acetylcholine or by electrical stimulation of the preganglionic nerve. It is concluded that VIP has a selective facilitatory action on muscarinic excitatory mechanisms in the superior cervical ganglion of the cat.

Parasympathetic ganglia: activation of an adrenergic inhibitory mechanism by cholinomimetic agents.

Electrical stimulation of the sympathetic nerves to the urinary bladder or the intraarterial administration of the cholinomimetic substances acetylcholine or methacholine produced adrenergic inhibition in parasympathetic ganglia on the surface of the bladder. The inhibition appeared to be mediated, at least in part, via adrenergic inhibitory neurons located in the pelvic plexus. Atropine blocked the inhibitory response to injected cholinomimetic agents but did not alter the response to stimulation of the sympathetic nerves. Thus, the inhibitory neurons can be activated via both muscarinic and nonmuscarinic receptors, the latter being of primary physiological importance.

Comparison of the respiratory responses to external resistive loading and bronchoconstriction.

The effects of resistive loads applied at the mouth were compared to the effects of bronchospasm on ventilation, respiratory muscle force (occlusion pressure), and respiratory sensations in 6 normal and 11 asthmatic subjects breathing 100% O2. External resistive loads ranging from 0.65 to 13.33 cm H2O/liter per s were applied during both inspiration and expiration. Bronchospasm was induced by inhalation of aerosolized methacholine. Bronchospasm increased ventilation, inspiratory airflow, respiratory rate, and lowered PACO2. External resistive loading, on the other hand, reduced respiratory rate and inspiratory flow, but left ventilation and PACO2 unaltered. FRC increased to a greater extent with bronchospasm than external flow resistive loads. With both bronchospasm and external loading, occlusion pressure increased in proportion to the rise in resistance to airflow. However, the change in occlusion pressure produced by a given change in resistance and the absolute level of occlusion pressure at comparable levels of airway resistance were greater during bronchospasm than during external loading. These differences in occlusion pressure responses to the two forms of obstruction were not explained by differences in chemical drive or respiratory muscle mechanical advantage. Although the subjects' perception of the effort involved in breathing was heightened during both forms of obstruction to airflow, at any given level of resistance the sense of effort was greater with bronchospasm than external loading. Inputs from mechanoreceptors in the lungs (e.g., irritant receptors) and/or greater stimulation of chest wall mechanoreceptors as a result of increases in lung elastance may explain the differing responses elicited by the two forms of resistive loading.

Role of gastrin supersensitivity in the pathogenesis of lower esophageal sphincter hypertension in achalasia.

Intraluminal manometric studies were carried out in 19 patients with untreated achalasia and in 20 normals. Lower esophageal sphincter (LES) pressure was 50.5 +/-4.6 mm Hg in patients with achalasia as compared with 19.4 +/-1.3 mm Hg in the normal group. In both groups, the LES pressure was lowered when exogenous 0.1 N HCl was placed into the stomach. Although the nadir of pressure attained with acid suppression was the same, the per cent inhibition was significantly greater in patients with achalasia. Serum gastrin levels were the same in the two groups studied. The patients with achalasia, pre- and postpneumatic dilatation, showed a supersensitivity to exogenous intravenous gastrin I, as compared with normals. These data suggest that high, acid-suppressible levels of LES pressure, in patients with achalasia, are due to supersensitivity to endogenous gastrin.

The pathogenesis of esophageal dysfunction in scleroderma and Raynaud's disease.

To determine the pathogenesis of esophageal dysfunction in scleroderma and Raynaud's disease, the lower esophageal sphincter (LES) was tested with: (a) methacholine acting directly at the cholinergic receptor on the muscle; (b) edrophonium, a cholinesterase inhibitor, enhancing the effect of released acetylcholine; and (c) gastrin I, acting through the release of acetylcholine. 10 patients with Raynaud's disease and 22 patients with scleroderma were compared with 20 normals and 20 patients with isolated LES incompetence. The mean basal LES pressure in normals was significantly greater than that recorded in both patients with scleroderma and Raynaud's disease. Six patients having scleroderma with normal peristalsis had an LES pressure significantly greater than that noted in 16 patients having scleroderma with abnormal peristalsis. In all groups, the per cent increase in LES pressure was similar when tested by direct muscle stimulation by methacholine. The response to agents that acted indirectly through intact cholinergic nerves differed in these groups. The LES response to gastrin I distinguished patients with normal peristalsis from those with abnormal peristalsis. The patients with normal peristalsis, either with scleroderma or with Raynaud's disease showed only a partial reduction in their response to gastrin I. The response to gastrin I was markedly reduced only in patients with abnormal peristalsis. These data indicate that in patients with scleroderma and Raynaud's disease, the LES response to direct muscle stimulation by methacholine was intact while the response to gastrin I and edrophonium was diminished.

Metabolic studies of isolated human eccrine sweat glands.

This paper describes a method for isolating and studying the metabolism of human eccrine sweat glands. (a) Electron microscopy of glands which had been isolated and then incubated for an hour revealed no apparent alteration in morphology. (b) Known variation in gland size (male > female > children) was reflected in the relative rates of lactate production. (c) Lactate production was approximately 1.5 nmoles/gland per hr in the absence of glucose and rose to 2.7 at physiological concentrations of glucose (5.6 mmoles/liter). This amount of lactate production agrees well with the amounts found in sweat. (d) Both adrenergic (epinephrine) and cholinergic (methacholine) stimuli increased lactate production. (e) Glycogen depletion was demonstrated during incubation. (f) O(2) consumption was measured and aerobic metabolism was found to account for less than 1% of the energy derived from anaerobic pathways. These studies demonstrate that the large amounts of lactate appearing in human eccrine sweat can be accounted for by glandular metabolism and that both glycogen and glucose can be used as substrates.

Mechanism of action of pentagastrin on the lower esophageal sphincter.

The effects of pentagastrin on lower esophageal sphincter (LES) pressure has been studied in trained, unanesthetized dogs. LES pressure was monitored by an infusion manometric technique. Increasing doses of pentagastrin up to 3 mug/kg given as an i.v. bolus resulted in increasing rises in LES pressure; larger doses resulted in a lesser effect of shorter duration. Increasing i.v. boluses of methacholine produced greater increases in LES pressure up to a maximum of 5 mug/kg; higher doses had similar effects. Atropine (50-100 mug/kg) slightly diminished the response of the LES to 2 or 6 mug/kg of pentagastrin. In large doses (500-2,000 mug/kg), atropine did not diminish the response to pentagastrin and prolonged the response of 6 mug/kg pentagastrin. Hexamethonium (2 mg/kg i.v.) depressed the peak response to 3 mug/kg pentagastrin slightly but the response to 6 mug/kg was increased and prolonged. Propranolol (2 mg/kg i.v.) significantly prolonged the effect of 6 mug/kg pentagastrin on the LES. We conclude that the stimulatory effect of pentagastrin is mainly due to a direct action on the LES. A lesser stimulatory effect is due to an action on preganglionic cholinergic neurons. Large doses of pentagastrin have both stimulatory and inhibitory effects. The inhibitory effect is mediated at least in part via preganglionic neurons acting through adrenergic receptors. Ganglionic transmission of the effect may be through muscarinic as well as nicotinic receptors.

Cholinergic modification of glucose-induced biphasic insulin release in vitro.

An in vitro system for perifusion of rat pancreatic islets has been utilized to define the effects of cholinergic agents on the dynamics of insulin release. In the absence of glucose the effects of either acetylcholine or acetyl-beta-methylcholine were minimal at concentrations up to 10(-5) mM. In the presence of low glucose concentration (2.4 mM), both of the muscarinic agents produced dose-dependent biphasic insulin release. Under these conditions significant insulin release was observed over both phases at concentrations of the muscarinic agents as low as 10(-8) mM. Further, the dose response curves relating muscarinic concentration to the total amount of insulin released in each of the two phases showed lack of parallelism between the curves. Nicotinic acid in concentrations up to 10(-5) mM had no effect on insulin release in the presence of 2.4 mM glucose. When the glucose concentration was increased to 16.4 mM, the effects of the muscarinic agents were significantly less than those observed in the presence of 2.4 mM glucose. This held true whether the effect was defined as absolute increment due to the muscarinic agent or as percentage of enhancement. Atropine inhibited insulin release induced by both acetylcholine and by 16.4 mM glucose. These data indicate that cholinergic stimulation can play a significant role in modifying insulin release patterns.

Vasoactive intestinal peptide in human nasal mucosa.

Vasoactive intestinal peptide (VIP), which is present with acetylcholine in parasympathetic nerve fibers, may have important regulatory functions in mucous membranes. The potential roles for VIP in human nasal mucosa were studied using an integrated approach. The VIP content of human nasal mucosa was determined to be 2.84 +/- 0.47 pmol/g wet weight (n = 8) by RIA. VIP-immunoreactive nerve fibers were found to be most concentrated in submucosal glands adjacent to serous and mucous cells. 125I-VIP binding sites were located on submucosal glands, epithelial cells, and arterioles. In short-term explant culture, VIP stimulated lactoferrin release from serous cells but did not stimulate [3H]glucosamine-labeled respiratory glycoconjugate secretion. Methacholine was more potent than VIP, and methacholine stimulated both lactoferrin and respiratory glycoconjugate release. The addition of VIP plus methacholine to explants resulted in additive increases in lactoferrin release. Based upon the autoradiographic distribution of 125I-VIP binding sites and the effects on explants, VIP derived from parasympathetic nerve fibers may function in the regulation of serous cell secretion in human nasal mucosa. VIP may also participate in the regulation of vasomotor tone.

Impaired vasodilation of forearm resistance vessels in hypercholesterolemic humans.

The effect of hypercholesterolemia on vascular function was studied in humans. To eliminate the potential confounding effects of atherosclerosis, vascular reactivity was measured in the forearm resistance vessels of 11 normal subjects (serum LDL cholesterol = 111 +/- 7 mg/dl) and 13 patients with hypercholesterolemia (serum LDL cholesterol = 211 +/- 19 mg/dl, P less than 0.05). Each subject received intrabrachial artery infusions of methacholine, which releases endothelium-derived relaxant factor, and nitroprusside which directly stimulates guanylate cyclase in vascular smooth muscle. Maximal vasodilatory potential was determined during reactive hyperemia. Vasoconstrictive responsiveness was examined during intra-arterial phenylephrine infusion. Forearm blood flow was determined by venous occlusion plethysmography. Basal forearm blood flow in normal and hypercholesterolemic subjects was comparable. Similarly, reactive hyperemic blood flow did not differ between the two groups. In contrast, the maximal forearm blood flow response to methacholine in hypercholesterolemic subjects was less than that observed in normal subjects. In addition, the forearm blood flow response to nitroprusside was less in hypercholesterolemic subjects. There was no difference in the forearm vasoconstrictive response to phenylephrine in the two groups. Thus, the vasodilator responses to methacholine and nitroprusside were blunted in patients with hypercholesterolemia. We conclude that in humans with hypercholesterolemia, there is a decreased effect of nitrovasodilators, including endothelium-derived relaxing factor, on the vascular smooth muscle of resistance vessels.

Maximum expiratory flow rates in induced bronchoconstriction in man.

We evaluated changes of maximum expiratory flow-volume (MEFV) curves and of partial expiratory flow-volume (PEFV) curves caused by bronchoconstrictor drugs and dust, and compared these to the reverse changes induced by a bronchodilator drug in previously bronchoconstricted subjects. Measurements of maximum flow at constant lung inflation (i.e. liters thoracic gas volume) showed larger changes, both after constriction and after dilation, than measurements of peak expiratory flow rate, 1 sec forced expiratory volume and the slope of the effort-independent portion of MEFV curves. Changes of flow rates on PEFV curves (made after inspiration to mid-vital capacity) were usually larger than those of flow rates on MEFV curves (made after inspiration to total lung capacity). The decreased maximum flow rates after constrictor agents are not caused by changes in lung static recoil force and are attributed to narrowing of small airways, i.e., airways which are uncompressed during forced expirations. Changes of maximum expiratory flow rates at constant lung inflation (e.g. 60% of the control total lung capacity) provide an objective and sensitive measurement of changes in airway caliber which remains valid if total lung capacity is altered during treatment.

Effect of adrenergic and muscarinic cholinergic agonists on atrial natriuretic peptide secretion by isolated rat atria. Potential role of the autonomic nervous system in modulating atrial natriuretic peptide secretion.

Stretching of the atrial wall is a known stimulant for atrial natriuretic peptide (ANP) secretion. Little is known about other factors that may influence ANP secretion. We examined the effects of the neurotransmitters of the autonomic nervous system on ANP secretion from isolated rat left atria. Superfusion with 10 muM norepinephrine produced a biphasic rise in ANP secretion with a peak response 2.5-fold above baseline secretion. To determine whether the response to norepinephrine primarily reflected alpha- or beta-adrenergic receptor stimulation, atria were superfused with 0.1 muM isoproterenol or 10 muM phenylephrine and 1 muM propranolol. ANP secretion in response to isoproterenol was biphasic, similar to the response to norepinephrine. Phenylephrine evoked a monophasic ANP secretory response, which was delayed in onset relative to that of isoproterenol or norepinephrine. Superfusion with 10 muM methacholine alone had no effect on ANP secretion, but rapidly attenuated norepinephrine-stimulated secretion by 67%. From these observations we conclude: (a) Both alpha- and beta-adrenergic agonists directly and distinctively stimulate ANP secretion; (b) Norepinephrine stimulates ANP secretion by both alpha- and beta-adrenergic mechanisms, however the secretory response pattern of norepinephrine reflects a predominence of beta-adrenergic activity; (c) Under basal conditions, methacholine does not influence ANP secretion; and (d) Methacholine inhibits norepinephrine-stimulated ANP secretion. Thus, in vivo, activation of the sympathetic nervous system may enhance ANP secretion, whereas a rise in parasympathetic tone may lower ANP secretion.

Correction of characteristic abnormalities of microtubule function and granule morphology in Chediak-Higashi syndrome with cholinergic agonists.

Chediak-Higashi (CH) syndrome is a genetic disorder of children and certain animal species including the beige mouse. We have previously described a membrane abnormality in CH mouse polymorphonuclear leukocytes (PMH). Whereas normal mouse PMN do not form surface caps with concanavalin A except after treatment with agents such as colchicine that inhibit microtubule assembly, CH mouse PMN show spontaneous cap formation. This capping is inhibited by 3',5 cyclic guanosine monophosphate and by the cholinergic agonists carbamylcholine and carbamyl beta-methylcholine that increase 3',5' cyclic guanosine monophosphate generation. These data suggested that microtubule function may be impaired in CH syndrome perhaps secondary to an abnormality in 3',5' cyclic guanosine monophosphate generation. The cholinergic agonists were also shown to prevent development of the giant granules that are pathognomonic of CH syndrome in embryonic fibroblasts isolated from CH mice and cultured in vitro. In this report it is shown that an extreme degree of spontaneous concanavalin A cap formation is also characteristic of peripheral blood PMN from two patients with CH syndrome. This indicates an abnormality of microtubule function in CH syndrome in man. 3',5' cyclic guanosine monophasphate, carbamylcholine, and carbamyl beta-methylcholine reduce spontaneous capping in CH cells. In addition, it is shown that monocytes isolated from the patients' blood and incubated in tissue culture generate a large complement of abnormal granules. When the same cells mature in vitro in the presence of carbamylcholine or carbamyl beta-methylcholine, the proportion of cells containing morphologically normal granules is significantly increased. These responses can be reproduced in vivo in the beige (CH) mouse. Animals treated for 3 wk and longer with carbamylcholine or carbamyl beta-methylcholline show normal granule morphology and a normal degree of concanavalin A cap formation in peripheral blood PMN leukocytes.

Impaired cardiac muscarinic receptor function in dogs with heart failure.

Prior physiological studies have suggested that parasympathetic control is altered in heart failure. The goal of our studies was to investigate the influence of heart failure on the muscarinic receptor, and its coupling to adenylate cyclase. Ligand binding studies using [3H]quinuclidinyl benzilate and enriched left ventricular (LV) sarcolemma, demonstrated that muscarinic receptor density in heart failure declined 36% from a control of 5.6 +/- 0.6 pmol/mg, with no change in antagonist affinity. However, agonist competition studies with both carbachol and oxotremorine showed that it was a loss of high affinity agonist binding sites in the sarcolemma from failing LV that accounted for this difference. The functional efficacy of the muscarinic receptor was also examined. When 1 microM methacholine was added to 0.1 mM GTP and 0.1 mM isoproterenol, adenylate cyclase stimulated activity was inhibited by 15% in normal LV but only 5% in LV sarcolemma from animals with heart failure even when the reduced adenylate cyclase in these heart failure animals was taken into account. Even at 100-fold greater concentrations of methacholine, significantly less inhibition of adenylate cyclase activity was observed in LV failure as compared with normal LV sarcolemma. Levels of the GTP-inhibitory protein known to couple the muscarinic receptor to adenylate cyclase, as measured with pertussis toxin labeling, were not depressed in LV failure. Thus, the inhibitory pathway regulating LV adenylate cyclase activity is defective in heart failure. The decrease in muscarinic receptor density, and in particular the specific loss of the high affinity agonist binding component of this receptor population, appears to be the major factor underlying this abnormality.

The in vitro inhibition of insulin secretion by diphenylhydantoin.

Glucose intolerance has been observed following diphenylhydantoin (DPH) intoxication. Because of this association between DPH and hyperglycemia, the effect of DPH on insulin release in vitro using preparations of isolated islets of Langerhans and pancreatic pieces was examined. In concentrations identical with those considered necessary for adequate anticonvulsant therapy in man, DPH markedly decreases the insulin secretory response of pancreatic pieces to methacholine, 1 mug/ml, tolbutamide, 250 mug/ml, and glucose, 200 mg/100 ml, without any demonstrable alteration in the oxidative conversion of glucose-1-(14)C or glucose-6-(14)C to (14)CO(2) by isolated islets. This DPH-induced inhibition of insulin secretion is not reversed by higher concentrations of glucose (600 mg/100 ml) or by increasing concentrations of extracellular calcium ion (4-6 mmoles/liter). 0.1 mM potassium and 10(-4) M ouabain, however, effectively restore the DPH-induced block of insulin secretion in pancreatic pieces. 60 mM potassium ion, on the other hand, not only restores the insulin secretory response to glucose (200 mg/100 ml) but results in an added stimulation of insulin secretion in the presence of DPH. In the presence of DPH, (22)Na accumulation by isolated islets is decreased by 26-40% as compared with controls. Such evidence is considered to indirectly support the postulate that the electrophysiological properties of DPH on the pancreas are due to a stimulation of the membrane sodium-potassium-magnesium ATPase.

Glucose metabolism of the isolated eccrine sweat gland. II. The relation between glucose metabolism and sodium transport.

This paper attempts to further clarify the characteristics of Mecholyl- or epinephrine-stimulated glucose metabolism in the isolated monkey eccrine sweat gland with special emphasis on its relationship to increased sodium transport. The Mecholyl- or epinephrine-stimulated glucose metabolism (as estimated by either lactate or (14)CO(2) production or both) is seen only in the secretory coil and not in the duct. It is markedly suppressed in the absence of glucose, Na(+), or K(+). It is inhibited by ouabain (10(-3) M) and partially suppressed in a low-sodium (40 mM), high-potassium (100 mM) medium.2,4-dinitrophenol (10(-4) M) reverses ouabain-induced inhibition of lactate and (14)CO(2) production but only partially reverses inhibition induced by Na(+) + K(+) deprivation, indicating that metabolic inhibition by ouabain is secondary to the inhibition of sodium transport. There is no synergism between Mecholyl and epinephrine. The absence of any significant inhibitory effects by acetazolamide (Diamox) or HCO(3) (-)-free media suggests that H(+) transport may not be important in sweat gland function. In contrast to a report by Wolfe et al., human eccrine sweat glands show considerable oxidative activity ((14)CO(2) production of 0.42-0.72 nmol/gland/h). These observations are discussed in terms of the linkage between sweat gland energy metabolism and sodium transport.