Blood-brain glucose transfer: repression in chronic hyperglycemia

Effects of changes in peripheral and cerebral glucose metabolism on locomotor activity, learning and memory in adult male rats

Interactions of glucose and cognitive function have been reported both in the presence of elevated arterial blood glucose levels and with decreased cerebral glucose metabolism. In order to test the peripheral vs. central effects of this phenomenon, we induced irreversible hyperglycemia and depression of cerebral glucose metabolism in separate designs by means of either intraperitoneal (i.p.) or intracerebroventricular (i.c.v.) administration of streptozotocin (STZ), which is known to damage insulin-producing cells. Behavioral functions, such as locomotor activity, learning, and memory, were investigated under these different conditions. IP treatment with STZ decreased locomotor activity and increased initial step-through latencies on the passive avoidance test. No effects of elevated arterial blood glucose levels on retention of passive avoidance learning checked at 24 h and 144 h after training were observed. I.c.v. treatment of STZ increased the rate of locomotor activity and impaired retention in the passive avoidance test at 24 h, without further forgetfulness at 144 h. This finding may indicate disturbed acquisition and/or consolidation of memory, which may remain impaired but at a constant level, without further deterioration. Enhanced motor activity and impaired acquisition of passive avoidance learning without further impairment have also been reported as a characteristical behavioral pattern after disruption of the cholinergic system. It is therefore postulated that the observed behavioral abnormalities consequent on an impairment of cerebral glucose metabolism may be suggestive of cholinergic dysfunction.

Regional blood-brain glucose transfer and glucose utilization in chronically hyperglycemic, diabetic rats following acute glycemic normalization

Regional rates of brain glucose utilization (rCMRglc) and glucose influx (rJin), along with regional brain tissue glucose concentrations, were measured in chronically hyperglycemic diabetic (CHD) rats following acute glycemic normalization. These results were compared to those obtained in nondiabetic normoglycemic controls. The diabetic rats were evaluated at 6-8 weeks following i.p. streptozotocin injection. All rats were N2O (70%) sedated, paralyzed, and artificially ventilated for study. Acutely normoglycemic (plasma glucose = 8.5 mumol/ml), demonstrated significantly higher (p less than 0.05) rCMRglc and rJin values in 8 of the 11 regions analyzed. Tissue/plasma glucose concentration ratios were significantly greater than control in 9 of 11 regions. Prior to acute glycemic normalization, rCMRglc values in CHD rats were either unchanged or moderately lower than control. These findings indicate that no blood-brain barrier glucose transport repression is present in CHD rats. In fact, the results suggest an increased transport capacity. The increased rCMRglc observed in the acutely normalized CHD rats may be a manifestation of the "hypoglycemic symptoms" observed in chronically hyperglycemic patients following acute glycemic reductions to the normal range. The present results imply that these symptoms are not related to the presence of a relative cerebral glucopenia, as others have suggested.

Acute and chronic effects of hyperglycaemia on glucose metabolism

In normal man, several hormonal and metabolic adjustments allow the maintenance of the blood glucose concentration within narrow limits. Hyperglycaemia participates in this regulation via stimulation of glucose disposal and inhibition of glucose production. The effects are mediated, in addition to changes in insulin and glucagon secretion, by the mass-action effect of glucose. In both Type 1 (insulin-dependent) and Type 2 (non-insulin-dependent) diabetic patients, hyperglycaemia, by mass-action abnormally elevates the basal glucose utilization rate but compensates for reduced postprandial insulin-stimulated glucose disposal. When exposed to chronic hyperglycaemia, the body tissues seem to protect themselves, at least partly, against excessive glucose utilization. These protective mechanisms include both a reduction in insulin stimulated glucose disposal and insulin secretion. Chronic hyperglycaemia may also reduce non-insulin-dependent glucose utilization, at least in rats. In Type 1 diabetic patients with normal peripheral insulin concentrations, chronic hyperglycaemia per se could be a major cause of insulin resistance. In Type 2 diabetic patients, insulin resistance is often already present before the development of overt fasting hyperglycaemia. At the diabetic stage, hyperglycaemia could, however, maintain a self-perpetuating cycle, where the deleterious effects of high glucose concentrations on insulin action and secretion cause further deterioration of glycaemic control. The biochemical basis for hyperglycaemia-induced insulin resistance is still far from clear, but could involve changes in the glucose transporter number and gene expression.

Cerebral blood flow and plasma volume during hyperglycemia in the conscious rat

Cerebral blood flow (CBF) and cerebral plasma volume (CPV) were measured under steady-state hyperglycemic conditions in the hemispheres and brainstem-cerebellum of conscious rats. There groups of hyperglycemic animals each having a different level of plasma glucose concentration, 25, 33.3, 44.4 mmol/l, and a normoglycemic control group were studied. CBF was not affected at the hyperglycemic levels of 25 and 33.3 mmol/l. Mean hemispheric and brainstem-cerebellum CBF values appeared lower than in controls at the highest glycemic level although the differences were not statistically significant. CPV was found to be unchanged at the hyperglycemic level of 25 mmol/l, while it was found to be increased in the hemispheres of the animals whose plasma glucose concentration had been elevated to 33.3 and 44.4 mmol/l. The results of the study do not support the claim that hyperglycemia may enhance ischemic brain injury by reducing CBF.

A comparison of brain glucose metabolism in diabetes as measured by positron emission tomography or by arteriovenous techniques

[U-11-C]-glucose and positron emission tomography was used to evaluate transport and oxidative metabolism of glucose in the brain of non-diabetic and insulin dependent diabetic (IDDM) subjects. These results were compared with results obtained by the Fick principle. Blood glucose was regulated by a Biostator controlled glucose infusion during a constant insulin infusion. [U-11-C]-glucose was injected during normoglycemia as well as during moderate hypoglycemia. The tracer data were analysed using a three compartment model with a fixed correction for [11-C]-CO2 egress. During normoglycemia the influx rate constant (k1) and blood brain glucose flux of the two groups were similar. During hypoglycemia k1 increased significantly and to the same extent in both groups. During normoglycemia the tracer calculated metabolism of glucose was higher in the brain of non-diabetic than of diabetic subjects. When measured by the Fick principle the net uptake of glucose was broadly the same for the groups. During normoglycemia the molar ratio of O2 to glucose uptake was, however, lower in IDDM than in non-diabetic subjects (4.67 vs 5.50, P less than 0.05, Wilcoxon test). A significant release of lactate and pyruvate was seen in IDDM but not in non-diabetic subjects. The results imply that a larger fraction of glucose is non-oxidatively metabolized in IDDM than in non-diabetic subjects.

Cerebral blood flow and substrate utilization in insulin-treated diabetic subjects

We compared regional cerebral blood flow (rCBF) and arteriojugular vein differences of glucose, ketones, glycerol, lactate, pyruvate, and O2 in eight subjects with well-controlled insulin-dependent diabetes mellitus (IDDM) and in six healthy volunteers. Duration of diabetes was 19.4 +/- 2.1 yr. Measurements were performed before and after 120 min of insulin infusion and concomitant Biostator-controlled normoglycemia. Net uptake of ketones was seen in IDDM subjects before but not after insulin. Net uptake of glucose did not differ significantly between groups. During normoglycemia the molar ratio of O2 to glucose uptake was lower in IDDM than in nondiabetic subjects (4.68 vs. 5.50; P less than 0.05; Wilcoxon test). Small but significant release of lactate and pyruvate was seen in IDDM but not in nondiabetic subjects. The rCBF was measured by 11CH3F and position emission tomography. Global mean CBF was higher in IDDM subjects (64.9 +/- 5.9 vs. 49.3 +/- 2.7 ml.100 g-1.min-1, means +/- SE in nondiabetic subjects, P less than 0.05). rCBF was enhanced in many cortical and subcortical areas, whereas it was decreased in the head of the caudate nucleus. Neuropsychological testing did not reveal obvious cognitive dysfunction. The results imply that a larger fraction of glucose is nonoxidatively metabolized in the IDDM subjects and furthermore indicate an abnormal rCBF pattern in these subjects.

D-[U-11C]glucose uptake and metabolism in the brain of insulin-dependent diabetic subjects

We used D-[U-11C]glucose to evaluate transport and metabolism of glucose in the brain in eight nondiabetic and six insulin-dependent diabetes mellitus (IDDM) subjects. IDDM subjects were treated by continuous subcutaneous insulin infusion. Blood glucose was regulated by a Biostator-controlled glucose infusion during a constant insulin infusion. D-[U-11C]-glucose was injected for positron emission tomography studies during normoglycemia as well as during moderate hypoglycemia [arterial plasma glucose 2.74 +/- 0.14 in nondiabetic and 2.80 +/- 0.26 mmol/l (means +/- SE) in IDDM subjects]. Levels of free insulin were constant and similar in both groups. The tracer data were analyzed using a three-compartment model with a fixed correction for 11CO2 egression. During normoglycemia the influx rate constant (k1) and blood-brain glucose flux did not differ between the two groups. During hypoglycemia k1 increased significantly and similarly in both groups (from 0.061 +/- 0.007 to 0.090 +/- 0.006 in nondiabetic and from 0.061 +/- 0.006 to 0.093 +/- 0.013 ml.g-1.min-1 in IDDM subjects). During normoglycemia the tracer-calculated metabolism of glucose was higher in the whole brain in the nondiabetic than in the diabetic subjects (22.0 +/- 1.9 vs. 15.6 +/- 1.1 mumol.100 g-1.min-1, P less than 0.01). During hypoglycemia tracer-calculated metabolism was decreased by 40% in nondiabetic subjects and by 28% in diabetic subjects. The results indicate that uptake of glucose is normal, but some aspect of glucose metabolism is abnormal in a group of well-controlled IDDM subjects.

Permeability of ocular vessels and transport across the blood-retinal-barrier

This paper reviews quantitative studies on the permeability of retinal and choroidal vessels and the exchange of nutrients over the blood retinal barrier (BRB). The fenestrated capillaries in the choroid are very permeable to low molecular weight substances; sodium permeability in the choroid is probably 50 times that in skeletal muscle. This results in high concentrations and rapid turnover of nutrients in the extra-vascular compartment of the choroid. Free diffusion is restricted by the pigment epithelium barrier. Also the retinal capillaries, with tight junctions between the endothelial cells, have very low permeability even to sodium. The uptake index technique has provided evidence for several carrier systems in the BRB; hexoses, neutral and basic amino acids, and monocarboxylic acids, very similar to those found in the brain. At least for glucose and lactate these carriers operate at both levels of the BRB; the RPE and the endothelium of the retinal capillaries, and in both directions; i.e. inwards and outwards.

Glucose phosphorylation rate in rat parietal cortex during normoglycemia, hypoglycemia, acute hyperglycemia, and in diabetes-prone rats

Cerebral metabolic rate for glucose (CMRglc) was studied in rats using [6-14C]glucose. After intravenous injection, the radioactivity of the parietal cortex was corrected for loss of labeled CO2 and divided by the integral of the arterial plasma glucose concentration, determined during tracer circulation. Treatment with insulin, resulting in plasma glucose concentrations less than 2.6 mmol/l, reduced CMRglc to 64% of the values found in control animals. CMRglc did not change in animals with acute hyperglycemia produced by intraperitoneal injection of a glucose solution or in diabetes-prone rats with or without insulin treatment.

The blood-brain interface

The properties of the blood-brain barrier are those of the capillary endothelium in brain. This endothelium contrasts with that elsewhere in being sealed with tight junctions, having a high electrical resistance and low permeability to polar solutes. It is exceptional in having a paucity of pits and vesicles, a specialised enzyme content and a high density of mitochondria. Functionally, a range of transport mechanisms allow rapid movement of certain specific metabolic substrates. Ion pumps are concerned with secretion of brain interstitial fluid and regulation of its ionic concentration. The retinal capillaries are largely identical to those of the brain, but entry of solutes into retina is also determined by the properties of the pigment epithelium, functionally separating the retina from the highly vascular choroid. A clear difference lies in the greater resistance of cerebral microvessels to diabetic damage. The mechanism of this difference is unclear, but may relate to a better control of the brain interstitial fluid at a lower glucose concentration than is possible in the retinal interstitial fluid.

Age- and diabetes-associated alterations in regional brain norepinephrine concentrations and adrenergic receptor populations in C57BL/KsJ mice

The diabetes-associated changes in regional brain norepinephrine (NE) concentrations and related adrenergic receptor types were correlated with changes in blood glucose levels and body weight (obesity) in 4-16-week-old C57BL/KsJ (db/db) mice relative to corresponding age-matched control (+/?) parameters. Regional brain (i.e. frontal cortex, septal area, amygdala, hypothalamus and medulla) NE levels were determined by high performance liquid chromatography and compared to the associated changes in tissue alpha-1,2 and beta-adrenergic membrane receptor populations. All db/db mice exhibited overt hyperglycemia and obesity relative to controls between 4 and 16 weeks of age. Regional brain NE levels in diabetics were chronically elevated as compared to those of age-matched controls. All of the alpha 1 and alpha 2 adrenergic receptor populations were elevated in the regional brain samples of diabetics relative to controls. In contrast, beta-adrenergic receptor populations were depressed in diabetics as compared with age-matched controls. These data demonstrate that a marked modification in regional brain adrenergic parameters occurs in association with the overt expression of the diabetes mutation in this species. The observed changes in adrenergic influences in specific CNS loci may be causally related to the recognized diabetes-associated alterations in regional brain structure, function and metabolism in C57BL/KsJ (db/db) mice.

Morphometric analysis of obesity (ob/ob)- and diabetes (db/db)-associated hypothalamic neuronal degeneration in C57BL/KsJ mice

The influence of the obese (ob/ob) and diabetes (db/db) genetic mutations on hypothalamic structure was investigated in C57BL/KsJ and C57BL/6J mice strains by morphometric analysis of medial basal nuclei which are recognized to possess glucoregulatory neurons. Brains were collected and prepared for histomorphometric analysis at selected times following the development of expressed obesity and diabetes (Type II, non-insulin dependent) syndromes in order to compare both the strain and genomic influences on neuronal viability in the hypothalamic ventromedial (VMH) and arcuate (ARC) nuclei of mutant and age-matched control mice. The severity of each syndrome was determined by monitoring the concomitant changes in body weight and blood glucose levels in all groups. Both (db/db) and (ob/ob) mutant C57BL/KsJ mice exhibited an increase in the number and distribution of degenerated neurons in the VMH and ARC nuclei relative to corresponding controls. The mutation-associated exacerbation of the normal age-related neuronal loss, as observed in control MBH nuclei, was temporally associated with the overt expression of the hyperglycemic component of the obese and diabetes syndromes in aging C57BL/KsJ mice. No temporal or causal relationships were noted between the enhanced rate of premature neuronal degeneration, and either body weight or blood glucose levels, in either (db/db) or (ob/ob) C57BL/6J mice relative to controls. These data suggest that the hyperglycemic condition which characterizes the (ob/ob) and (db/db) mutant C57BL/KsJ mice is causally associated with the pronounced, premature MBH neuronal degeneration in these mouse strains. Neuronal changes were not pronounced when the genetic mutations were expressed in C57BL/6J mice. The accompanying alterations in brain glucose metabolism, hormone sensitivity, bioamine content and function which are recognized to occur in these mutant C57BL/KsJ mice may be causally associated consequences of the observed changes in MBH structural integrity and neuronal competence, with the severity of the mutation-associated changes being related to genetic background of the murine strain.

Blood-brain barrier glucose transporter mRNA is increased in experimental diabetes mellitus

The blood-brain barrier (BBB) glucose transporter activity in vivo is known to be down-regulated in experimental diabetes mellitus. To determine whether parallel changes in BBB glucose transporter mRNA levels occur in experimental diabetes we isolated brain microvessels, which make up the BBB in vivo, from both control and experimental diabetic rats. Microvessel RNA fractions were isolated by cesium chloride density gradient centrifugation and were applied to 1.1% agarose gels for Northern blotting. The blots were probed with [32P]-labeled cDNAs corresponding to the rat brain glucose transporter and a cDNA to alpha-actin was used to monitor the transcript level of a typical housekeeping gene. The study was repeated three times and, in all cases, the BBB glucose transporter mRNA level was increased in experimental diabetes relative to control rats. These studies suggest that factors associated with experimental diabetes mellitus in rats lead to either an increased transcription or a decreased degradation of brain capillary glucose transporter mRNA.

Brain perfusion in acute and chronic hyperglycemia in rats

Recent studies show that acute and chronic hyperglycemia cause a diffuse decrease in regional cerebral blood flow and that chronic hyperglycemia decreases the brain L-glucose space. Since these changes can be caused by a decreased density of perfused brain capillaries, we used 30 adult male Wistar rats to study the effect of acute and chronic hyperglycemia on 1) the brain intravascular space using radioiodinated albumin, 2) the anatomic density of brain capillaries using alkaline phosphatase histochemistry, and 3) the fraction of brain capillaries that are perfused using the fluorescein isothiocyanate-dextran method. Our results indicate that acute and chronic hyperglycemia do not affect the brain intravascular space nor the anatomic density of brain capillaries. Also, there were no differences in capillary recruitment among normoglycemic, acutely hyperglycemic, and chronically hyperglycemic rats. These results suggest that the shrinkage of the brain L-glucose space in chronic hyperglycemia is more likely due to changes in the blood-brain barrier permeability to L-glucose.

Changes in histamine metabolism in the brains of mice with streptozotocin-induced diabetes

Histamine (HA) metabolism in the brain of mice with streptozotocin (STZ)-induced diabetes was examined. The levels of tele-methylhistamine (t-MH), a major metabolite of brain HA, significantly increased 3 and 4 weeks after STZ injection. However, the HA turnover rates in the diabetic mice, determined from the accumulation of t-MH after the administration of pargyline, were not different from the control values when the animals were allowed free access to food. When the mice were starved for 15 h 4 weeks after STZ treatment, the brain levels of L-histidine decreased significantly, whereas HA turnover increased significantly. Such changes were not observed in starved control mice. Histidine decarboxylase or HA N-methyltransferase activity did not change after starvation in either diabetic or control mice. These results show that the histaminergic (HAergic) activity in the brains of diabetic mice remains within normal range as long as the animals are allowed free access to food. However, they also indicate that a marked enhancement of HAergic activity accompanied by a decrease in the brain L-histidine level occurs in starved diabetic mice.

Glucose transporter localization in brain using light and electron immunocytochemistry

A polyclonal antibody to a synthetic 13 amino acidpeptide found at the carboxyl-terminal end of the glucose transporter protein was raised in rabbit and used in light and electron immunocytochemical studies of human and canine brain. This antibody identified a broad band of polypeptide of average Mr 55,000 on immunoblots (immunogold-silver stains) of electrophoresed membrane proteins from human red blood cells. A similar polypeptide band (Mr 45,000-60,000) was identified on immunoblots of microvessel membrane proteins isolated from canine cerebrum, suggesting that this antibody is a useful tool for studying the distribution and abundance of the glucose transporter protein in mammalian nervous tissue. Peroxidase antiperoxidase stains of cerebrum using this antibody demonstrated that transporters are abundant in the intima pia, in the endothelium of blood vessels in the subarachnoid space, and in the endothelium of arterioles, venules, and capillaries of gray and white matter. In cerebellum, reaction product was localized in the vessels of the subarachnoid space and in microvessels of the molecular layer, the granular layer, and the white matter. However, transporters were not found in the intima pia of cerebellum. In medulla oblongata, transporters were found in the intima pia, the endothelium of some subarachnoid vessels, and the microvessels of gray and white matter. In pituitary, microvessels in adenohypophysis contained no reaction product, but the antigen was detected in some microvessels in neurohypophysis. Electron microscopy of cerebral cortex using a protein A-gold technique demonstrated that glucose transporters are equally abundant on the luminal and abluminal membranes of microvessel endothelial cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of glucose starvation on glucose transport in neuronal cells in primary culture from rat brain

The regulation of glucose transport into cultured brain cells during glucose starvation was studied. On glucose deprivation for 40 h, 2-deoxy-D-glucose (2-DG) uptake was stimulated twofold in neuronal cells but was not changed significantly in astrocytes. On refeeding, the increased activity of neuronal cells rapidly returned to the basal level, an observation indicating that the effect of glucose starvation was reversible. The increase was due solely to change in the Vmax, a finding suggesting that the number of glucose transporters on the plasma membrane is increased in starved cells. Cycloheximide inhibited this increase. In the presence of cycloheximide, the activity of 2-DG uptake of starved cells remained constant for 12 h and then slowly decreased, whereas that of fed cells decreased rapidly. These findings suggest that glucose starvation regulates glucose transport by changing the rate of net synthesis of the transporter in neuronal cells in culture.

Cerebral mitochondrial respiration in diabetic and chronically hypoglycemic rats

The respiratory function of cerebral mitochondria harvested from genetically diabetic (BB/W) and streptozotocin-diabetic rats deprived of insulin for 3-4 weeks was found to be unchanged from control values. Furthermore, insulin-deprived BB/W rats subjected to 30 min of insulin-induced hypoglycemic coma demonstrated a normal mitochondrial respiration following a 60 min period of glucose restitution, a finding consistent with earlier results in non-diabetic rats. However, in rats exposed to 1 week of moderate hypoglycemia (plasma glucose = 3.0 mumol.ml-1), both state 3 respiration and the respiratory control ratio (RCR) were reduced from control. In fact, when the chronic hypoglycemia was imposed following a 3-4 week period of diabetic hyperglycemia, the state 3 rate and RCR were found to be reduced to a greater degree than in chronically hypoglycemic, non-diabetic, previously normoglycemic rats. Finally, when 1 week of moderate hypoglycemia preceded a 30 min period of insulin-induced hypoglycemic coma, a disturbed pattern of mitochondrial respiration (i.e. increased state 4, decreased RCR) was found at 60 min of recovery following coma. These results indicate that chronic increases in glucose (and insulin deprivation) have no effect on cerebral mitochondrial respiratory function, whereas prolonged, albeit moderate, reductions in cerebral glucose supply result in perturbations in mitochondrial respiration. These results demonstrate the importance of an adequate glucose supply for normal mitochondrial activity.

Glucose enhancement of performance on memory tests in young and aged humans

Recent findings indicate that glucose administration enhances memory processes in rodents. This study examined the effects of glucose on memory in humans. After drinking glucose- or saccharin-flavored beverages, college-aged and elderly humans were tested with modified versions of the Wechsler Memory Scale. Beverages and tests were administered in a counter-balanced, crossover design, enabling within subject comparisons. The major findings were: (1) glucose enhanced memory in elderly and, to a lesser extent, in young subjects; and (2) glucose tolerance in individual subjects predicted memory in elderly, but not in young subjects on both glucose and saccharin test days.

Glucose transporter of the blood-brain barrier and brain in chronic hyperglycemia

The effect of chronic hyperglycemia on the glucose transporter moiety of the blood-brain barrier and cerebral cortex was studied in rats 3 weeks after the administration of a single intravenous dose of streptozotocin (60 mg/kg), using specific [3H]cytochalasin B binding methods. Streptozotocin-treated rats developed hyperglycemia, as well as polydipsia and polyuria, and failed to gain weight. The density of D-glucose-displaceable cytochalasin B binding sites in the brain microvessels of streptozotocin-treated hyperglycemic rats was increased by about 30% compared with those of control rats, without change in the affinity of binding. Chronic hyperglycemia had no effect on the density or affinity of specific binding of cytochalasin B to cerebral cortical membranes. These findings do not support the hypothesis that glucose transporters in brain microvessels comprising the blood-brain barrier are "down-regulated" in chronic hyperglycemia.

Vascular perfusion and blood-brain glucose transport in acute and chronic hyperglycemia

We studied the effects of acute and streptozotocin-induced chronic hyperglycemia on regional brain blood flow and perfusion characteristics, and on the regional transport of glucose across the blood-brain barrier in awake rats. We found (1) a generalized decrease in regional brain blood flow in both acute and chronic hyperglycemia; (2) that chronic, but not acute, hyperglycemia is associated with a marked and diffuse decrease in brain L-glucose space; and (3) that chronic hyperglycemia does not alter blood-to-brain glucose transport. Taken together, these results suggest that in streptozotocin-induced chronic hyperglycemia, there is a reduction in the proportion of perfused brain capillaries and/or an alteration in brain endothelial membrane properties resulting in decreased noncarrier diffusion of glucose.

Glucose transfer into rat brain during acute and chronic hyperglycemia

Chronic hyperglycemia has been reported to decrease the maximum velocity of glucose transport across the blood-brain barrier by 30 to 40%. However, available measurements of brain glucose content during chronic hyperglycemia are consistent with an unaltered transport system. Because of this discrepancy the brain capillary permeability-surface area product (PA) was measured in awake-restrained rats during acute and chronic hyperglycemia. Acute hyperglycemia was produced by intraperitoneal injection of glucose, and chronic hyperglycemia was produced by treatment with streptozotocin. PA was measured using an intravenous tracer method. PA decreased during hyperglycemia, consistent with saturation kinetics for transfer. However, PA was similar in acutely and chronically hyperglycemic rats. These data suggest that down-regulation of facilitated glucose transport into the brain does not occur during chronic hyperglycemia.

Modest decrements in plasma glucose concentration cause early impairment in cognitive function and later activation of glucose counterregulation in the absence of hypoglycemic symptoms in normal man

To establish the glycemic threshold for onset of neuroglycopenia (impaired cognitive function, measured by the latency of the P300 wave), activation of hormonal counterregulation and hypoglycemic symptoms, 12 normal subjects were studied either under conditions of insulin-induced, glucose-controlled plasma glucose decrements, or during maintenance of euglycemia. A decrement in plasma glucose concentration from 88 +/- 3 to 80 +/- 1 mg/dl for 150 min did not result in changes in the latency of the P300 wave nor in an activation of counterregulatory hormonal response. In contrast, a greater decrement in plasma glucose concentration from 87 +/- 3 to 72 +/- 1 mg/dl for 120 min caused an increase in the latency of the P300 wave (from 301 +/- 12 to 348 +/- 20 ms, P less than 0.01), a subsequent increase in all counterregulatory hormones but no hypoglycemic symptoms. Finally, when plasma glucose concentration was decreased in a stepwise manner from 88 +/- 2 to 50 +/- 1 mg/dl within 75 min, the increase in the latency of the P300 wave was correlated with the corresponding plasma glucose concentration (r = -0.76, P less than 0.001). The glycemic threshold for hypoglycemic symptoms was 49 +/- 2 mg/dl. Thus, in normal man the glycemic threshold for neuroglycopenia (72 +/- 1 mg/dl) is greater than currently thought; the hormonal counterregulation follows the onset of neuroglycopenia; the hypoglycemic symptoms are a late indicator of advanced neuroglycopenia.

Counterregulatory hormonal responses to hypoglycaemia in type 1 (insulin-dependent) diabetes: evidence for diminished hypothalamic-pituitary hormonal secretion

Acute insulin-induced hypoglycaemia in humans provokes autonomic neural activation and counterregulatory hormonal secretion mediated in part via hypothalamic stimulation. Many patients with Type 1 (insulin-dependent) diabetes have acquired deficiencies of counterregulatory hormonal release following hypoglycaemia. To study the integrity of the hypothalamic-pituitary and the sympatho-adrenal systems, the responses of pituitary hormones, beta-endorphin, glucagon and adrenaline to acute insulin-induced hypoglycaemia (0.2 units/kg) were examined in 16 patients with Type 1 diabetes who did not have autonomic neuropathy. To examine the effect of duration of diabetes these patients were subdivided into two groups (Group 1: 8 patients less than 5 years duration; Group 2: 8 patients greater than 15 years duration) and were compared with 8 normal volunteers (Group 3). The severity and time of onset of hypoglycaemia were similar in all 3 groups, but mean blood glucose recovery was slower in the diabetic groups (p less than 0.01). The mean responses of glucagon, adrenaline, adrenocorticotrophic hormone, prolactin and beta-endorphin were similar in all 3 groups, but the mean responses of growth hormone were lower in both diabetic groups than in the normal group (p less than 0.05). The mean increments of glucagon and adrenaline in the diabetic groups were lower than the normal group, but these differences did not achieve significance; glucagon secretion was preserved in several diabetic patients irrespective of duration of disease. Various hormonal responses to hypoglycaemia were absent or diminished in individual diabetic patients, and multiple hormonal deficiencies could be implicated in delaying blood glucose recovery.(ABSTRACT TRUNCATED AT 250 WORDS)

Plasma glucose concentrations at the onset of hypoglycemic symptoms in patients with poorly controlled diabetes and in nondiabetics

We tested the hypothesis that during decrements in plasma glucose concentration, symptoms of hypoglycemia may occur at higher glucose concentrations in patients with poorly controlled insulin-dependent diabetes mellitus than in persons without diabetes. Symptoms of hypoglycemia and counterregulatory neuroendocrine responses were quantified during hypoglycemic and euglycemic clamp studies in eight patients with insulin-dependent diabetes mellitus selected because their hemoglobin A1 levels were above 10 percent. These data were compared with similar observations in 10 nondiabetic subjects studied previously. Glycemic thresholds--the plasma glucose concentrations during each hypoglycemic clamp study at which a given symptom or biochemical measurement first exceeded its 95 percent confidence interval determined in the euglycemic clamp studies--were calculated for each variable. The mean (+/- SE) glycemic threshold for the symptoms of hypoglycemia was 4.3 +/- 0.3 mmol per liter (78 +/- 5 mg per deciliter) in patients with poorly controlled diabetes--significantly higher (P less than 0.001) than the value of 2.9 +/- 0.1 mmol per liter (53 +/- 2 mg per deciliter) in subjects without diabetes. The mean glycemic thresholds for growth hormone, epinephrine, and cortisol secretions were not significantly different in the two groups. Thus, during decreases in the plasma glucose concentration, patients with poorly controlled insulin-dependent diabetes mellitus may experience symptoms of hypoglycemia at higher plasma glucose concentrations than persons without diabetes. The mechanism underlying this observation remains to be defined.

Interactions between the blood-brain barrier and endogenous peptides: emerging clinical implications

The effects of peptides on brain function suggest therapeutic and pathologic roles for these substances. Many peptides cross the blood-brain barrier (BBB) by transmembrane diffusion as a function of their lipid solubilities. Other peptides, such as the enkephalins, Tyr-MIF-1, vasopressin-related peptides, and peptide T-like peptides, are transported by carrier-mediated systems. Passage is influenced by aging, stress, lighting, drugs, amino acids, and neurotoxins. Disruption of the BBB results in complex changes in the blood and CSF levels of peptides. Peptides influence the passage of glucose, amino acids, and inorganic acids and may affect the integrity of the BBB. Peptide-BBB interactions have been suggested to play direct roles in dialysis dementia and maple syrup urine disease; they may be expected to be involved in other disorders of the CNS.

Antecedent chronic hyperglycaemia blocks phlorizin-induced insulin resistance in the dog

Hyperglycaemia may enhance insulin resistance typical of non-insulin dependent diabetes mellitus, as well as insulin dependent diabetes mellitus, and thus initiate a vicious pathogenetic cycle. We sought to test the hypothesis that reduction in chronic hyperglycaemia in the diabetic dog by methods that do not employ insulin may improve insulin resistance. We used the glucuretic agent phlorizin in dogs rendered chronically hyperglycaemic and diabetic by alloxan treatment. To analyse glucose disposition the euglycaemic clamp was performed. To minimize the role of counterregulatory influences that might be at play when glucose is reduced, the hyperglycaemic clamp with continuous somatostatin infusion was performed. Although phlorizin normalised plasma glucose in the diabetic dog and reduced plasma glucose in normal, non-diabetic dogs, insulin dependent glucose disposition rate did not improve. While phlorizin itself was associated with insulin resistance in the normal animals, the insulin resistance of diabetes mellitus was not further augmented. We conclude that phlorizin is associated with insulin resistance perhaps by a common pathway shared by chronic hyperglycaemia. Care must be taken when phlorizin is used as an agent to study glucose disposition.

Effects of hypoglycemia and diabetes on fuel metabolism by rat brain microvessels

Glucose and beta-hydroxybutyrate metabolism were compared in isolated cerebral microvessels from chronically diabetic and hypoglycemic rats. As noted previously, glucose oxidation and conversion to lactate are diminished in rats with streptozotocin-induced diabetes. The decrease in glucose metabolism did not result from selective damage to diabetic vessels during isolation, since the ATP level and the ATP/ADP ratio were similar to those of nondiabetic rats, and O2 consumption was increased. In addition, cerebral microvessel oxidation of beta-hydroxybutyrate was enhanced by diabetes. By contrast, microvessels from rats made chronically hypoglycemic by insulinoma engrafting 30 days earlier had a more than twofold increase in glucose oxidation and conversion to lactate, whereas their oxidation of beta-hydroxybutyrate was diminished by 50%. Unlike the insulinoma rats, no consistent increase in glucose metabolism was observed in microvessels from rats made hypoglycemic either by acute insulin administration or by a 4-day infusion of insulin. These results indicate that diabetes, and under some circumstances chronic hypoglycemia, markedly alters fuel metabolism in the cerebral microvasculature.

Effect of aging on the blood-brain barrier

Aging is commonly associated with progressive deterioration in central nervous system (CNS) function. Nutritional factors or environmental toxins have important effects on CNS degenerative changes. The blood-brain barrier (BBB) is a major modulator of nutrient delivery to the CNS. The tight junctions and the paucity of pinocytosis or fenestrations in brain capillary endothelium act as an effective barrier between the CNS and the circulating toxic agents. Senescence is associated with significant, though often subtle, changes in BBB. Conditions which are commonly associated with aging, such as hypertension and cerebrovascular ischemia, aggravate the age-related alterations in BBB function. The histologic changes in brain vasculature with aging is region selective and species specific. The common age-related histologic changes include loss of capillary endothelial cells, elongation of the remaining endothelial cells, and decreased capillary diameter in rat cortex, but not in the monkey or human cortex, and a decrease in the number of mitochondria in endothelial cells of the brain capillaries in the monkey but not in the rat. The age-related alterations in BBB transport function include a decrease in BBB choline transport with aging and decreased brain glucose influx. The BBB neutral amino acid transport appears to be unaltered in the aged mice. Most of the studies reported so far have failed to show a significant age-related alteration in BBB permeability to water-soluble substances and high molecular weight solutes in the absence of neurological disease. A more profound change in BBB permeability appears to be associated with Alzheimer's disease. Immunohistological studies have demonstrated the presence of serum proteins in the cerebrovascular amyloid in patients with Alzheimer's disease.(ABSTRACT TRUNCATED AT 250 WORDS)

Brain energy metabolism in streptozotocin-diabetes

Regional brain glucose use was measured in rats with streptozotocin-induced diabetes (65 mg/kg intravenously) of 1 or 4 weeks duration, by using [6-14C]glucose and quantitative autoradiography. The concentrations of several metabolites were measured in plasma and brain. Results were compared with those from normal untreated rats. Glucose concentrations were increased in both plasma and brain, to similar degrees in both diabetic groups. Plasma ketone-body concentrations were 0.25, 1.0, and 3.15 mumol/ml in the control, 1-week and 4-week groups respectively (sum of acetoacetate and 3-hydroxybutyrate). Glucose use was increased throughout the brain (differences were statistically significant in 55 of 59 brain areas) after 1 week of diabetes, with an increase of 25% for the brain as a whole. In contrast, normal rates were found throughout the brain after 4 weeks of diabetes. None of the brain areas was affected significantly differently from the others, in either diabetic group. There was no significant loss of 14C as lactate or pyruvate during the experimental period, nor was there any indication of net production of lactate in any of the groups. Other methodological considerations that could have affected the results obtained in the diabetic rats were likewise ruled out. Because the ketone bodies are expected to supplement glucose as a metabolic fuel for the brain, our results indicate that brain energy consumption is increased during streptozotocin-diabetes.

Regional amino acid transport into brain during diabetes: effect of plasma amino acids

Transport of phenylalanine and lysine into the brain was measured in 4-wk streptozotocin-diabetic rats to assess the effect on the neutral and basic amino acid transport systems at the blood-brain barrier. Amino acid concentrations in plasma and brain were also measured. Regional permeability-times-surface area (PS) products and influx were determined using a continuous infusion method and quantitative autoradiography. The PS of phenylalanine was decreased by an average of 40% throughout the entire brain. Influx was depressed by 35%. The PS of lysine was increased by an average of 44%, but the influx was decreased by 27%. Several plasma neutral amino acids (branched chain) were increased, whereas all basic amino acids were decreased. Brain tryptophan, phenylalanine, tyrosine, methionine, and lysine contents were markedly decreased. The transport changes were almost entirely accounted for by the alterations in the concentrations of the plasma amino acids that compete for the neutral and basic amino acid carriers. The reduced influx could be responsible for the low brain content of some essential amino acids, with possibly deleterious consequences for brain function.

Effect of chronic hyperglycemia on in vivo insulin secretion in partially pancreatectomized rats

We have examined the effect of chronic (4 wk) hyperglycemia on insulin secretion in vivo in an awake, unstressed rat model. Three groups of animals were examined: control, partial (90%) pancreatectomy, and partial pancreatectomy plus phlorizin, in order to normalize plasma glucose levels. Insulin secretion in response to arginine (2 mM), hyperglycemia (+100 mg/dl), and arginine plus hyperglycemia was evaluated. In diabetic compared with control animals three specific alterations were observed: (a) a deficient insulin response, in both first and second phases, to hyperglycemia; (b) an augmented insulin response to the potentiating effect of arginine under basal glycemic conditions; and (c) an inability of hyperglycemia to augment the potentiating effect of arginine above that observed under basal glycemic conditions. Normalization of the plasma glucose profile by phlorizin treatment in diabetic rats completely corrected all three beta cell abnormalities. These results indicate that chronic hyperglycemia can lead to a defect in in vivo insulin secretion which is reversible when normoglycemia is restored.

Altered glucose kinetics in diabetic rats during gram-negative infection

The present study examined the purported exacerbating effect of sepsis on glucose metabolism in diabetes. Diabetes was induced in rats by an intravenous injection of 70 or 45 mg/kg streptozotocin. The higher dose produced "severe" diabetes, whereas the lower dose of streptozotocin produced a milder, "latent" diabetes. After a chronic diabetic state had developed for 4 wk, rats had catheters implanted and sepsis induced by intraperitoneal injections of live Escherichia coli. After 24 h of sepsis the blood glucose concentration was unchanged in nondiabetics and latent diabetics, but glucose decreased from 15 to 8 mM in the septic severe diabetic group. This decrease in blood glucose was not accompanied by alterations in the plasma insulin concentration. Glucose turnover, assessed by the constant intravenous infusion of [6-3H]- and [U-14C]glucose, was elevated in the severe diabetic group, compared with either latent diabetics or nondiabetics. Induction of sepsis produced a slight decrease in the glucose turnover in the severe diabetic group but did not alter turnover in the latent diabetics. The rate of glucose disappearance, used to quantitate the alterations in plasma glucose after an intravenous glucose tolerance test, was decreased in both groups of diabetics and was proportional to the severity of the diabetic state. Sepsis increased the rate of glucose disappearance in nondiabetic rats but had no effect in either group of diabetic animals. Sepsis also failed to alter the insulinogenic index, used to estimate the insulin secretory capacity, in diabetic rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Regional cerebral glucose utilization during hyperglycemia

Previous data indicate that regional cerebral blood flow (rCBF) decreases during acute and chronic hyperglycemia. To test the hypothesis that the decrease in rCBF is secondary to a decrease in cerebral metabolic rate, the rate of regional cerebral glucose utilization (rCMRgl) was measured in awake-restrained rats during acute and chronic hyperglycemia. Acute hyperglycemia was produced by intraperitoneal injection of glucose, and chronic hyperglycemia was produced by treatment with streptozotocin. The rCMRgl was measured over a 10-min period using [6-14C]glucose. Glucose utilization was normal during acute hyperglycemia but decreased by 13% following 3 weeks of chronic hyperglycemia. The absence of a decrease in rCMRgl measured during acute hyperglycemia indicates that decreased rCBF cannot be explained by a change in the metabolic rate of the brain. The decrease in rCMRgl measured during chronic hyperglycemia does not necessarily indicate the presence of a drop in the metabolic rate of the brain because ketone bodies are available as an alternate fuel for oxidative metabolism. Therefore, it is unlikely that the decrease in rCMRgl measured during chronic hyperglycemia accounts for decreased rCBF.

Regional cerebral blood flow decreases during chronic and acute hyperglycemia

The presence of hyperglycemia prior to stroke or cardiac arrest can increase neuronal damage caused by brain ischemia. Acute hyperglycemia shows this effect in animal models of stroke. However, chronic hyperglycemia and chronic hyperglycemia with additional acute elevation of blood glucose are more common premorbid states for stroke patients. The effect of chronic hyperglycemia on regional cerebral blood flow (rCBF) is unclear but blood flow changes may play a role in this ischemic cell damage. We measured rCBF in awake restrained rats that had chronic hyperglycemia induced by treatment with streptozotocin. This was compared to that measured in rats made acutely hyperglycemic by injecting glucose into the peritoneal space. rCBF was measured in 17 brain regions using [14C]iodoantipyrine. During chronic hyperglycemia, when plasma glucose was 29 microns/ml, rCBF was decreased and a regional distribution of this effect was noted; 9 hindbrain regions showed a mean flow decrease of 14% while forebrain regions demonstrated less flow reduction. Acute elevation of plasma glucose during normoglycemia or superimposed on chronic hyperglycemia produced flow reductions of 7% for each 10 microns/ml increment in plasma glucose up to 60 microns/ml. Both chronic and acute hyperglycemia are associated with decreased rCBF and the mechanism for this effect does not appear to adapt to chronic hyperglycemia.

Regional studies of blood-brain barrier transport of glucose and leucine in awake and anesthetized rats

D-Glucose and L-leucine are transported across the blood-brain barrier (BBB) by two separate carrier-mediated facilitated diffusion mechanisms. In the awake rat there are regional differences in blood-to-brain glucose transport among the cerebral cortex, cerebellum, hippocampus, and striatum. To determine whether these are due to variations in the regional density or affinity of the glucose transporter moiety of brain capillaries or are secondary to regional tissue perfusion and capillary arrangement characteristics, we studied regional blood-to-brain transport of L-leucine in awake rats; regional blood-to-brain transport of both glucose and leucine under chloral hydrate anesthesia, a condition associated with altered regional brain blood flow (BF) and metabolism; and regional brain vascular volume, derived from the L-glucose and insulin spaces, in both awake and anesthetized rats. We found the same regional differences in blood-to-brain leucine transport in awake rats as we previously described for D-glucose transport. These regional differences in glucose and leucine transport disappear under chloral hydrate anesthesia, as regional differences in BF are abolished. However, we found regional differences in the brain vascular volumes, which are evident in wakefulness and persist during anesthesia. These results suggest that the regional differences in blood-to-brain transport are due mainly to local tissue perfusion and capillary arrangement characteristics rather than to intrinsic regional differences in the transport systems of the BBB.

Studies on the relationship between cerebral glucose transport and phosphorylation using 2-deoxyglucose

Regional rates of blood-brain glucose transfer and phosphorylation have been measured in anaesthetized fasted and conscious fed and fasted rats using a dual-label 2-deoxyglucose technique that exploits differences in the early-time distribution of analogue and native glucose between blood and brain. Regional cerebral blood flow was also measured in comparable groups of rats. Estimates of glucose influx in the anaesthetized group were compared with those calculated from previously published kinetic constants obtained using [14C]D-glucose as tracer. The close agreement of these two sets of results served to validate estimates of influx obtained using the glucose analogue. Comparisons between all three groups showed that regional rates of glucose influx were maintained at levels appropriate to the rate of cerebral glucose phosphorylation. This occurred despite wide variations in plasma glucose concentration. The results indicate that at least two factors are involved in the adaptation of glucose supply to meet metabolic demand. One is related to blood flow, and probably reflects changes in the surface area of the capillary endothelium perfused. The second involves changes in the blood-brain barrier permeability to glucose and could reflect changes in the density of functioning glucose transporters within capillary endothelial cell membranes.