Mechanism of D-cycloserine action: transport systems for D-alanine, D-cycloserine, L-alanine, and glycine

The accumulation of d-alanine, l-alanine, glycine, and d-cycloserine in Escherichia coli was found to be mediated by at least two transport systems. The systems for d-alanine and glycine are related, and are separate from that involved in the accumulation of l-alanine. d-Cycloserine appears to be primarily transported by the d-alanine-glycine system. The accumulation of d-alanine, glycine, and d-cycloserine was characterized by two line segments in the Lineweaver-Burk analysis, whereas the accumulation of l-alanine was characterized by a single line segment. d-Cycloserine was an effective inhibitor of glycine and d-alanine accumulation, and l-cycloserine was an effective inhibitor of l-alanine transport. The systems were further differentiated by effects of azide, enhancement under various growth conditions, and additional inhibitor studies. Since the primary access of d-cycloserine in E. coli is via the d-alanine-glycine system, glycine might be expected to be a better antagonist of d-cycloserine inhibition than l-alanine. Glycine and d-alanine at 10(-5)m antagonized the effect of d-cycloserine in E. coli, whereas this concentration of l-alanine had no effect.     

Mechanism of D-cycloserine action: transport mutants for D-alanine, D-cycloserine, and glycine

The accumulation of d-alanine and the accumulation of glycine in Escherichia coli are related and appear to be separate from the transport of l-alanine. The analysis of four d-cycloserine-resistant mutants provides additional support for this conclusion. The first-step mutant from E. coli K-12 that is resistant to d-cycloserine was characterized by the loss of the high-affinity line segment of the d-alanine-glycine transport system in the Lineweaver-Burk plot. This mutation, which is linked to the met(1) locus, also resulted in the loss of the ability to transport d-cycloserine. The second-step mutation that is located 0.5 min from the first-step mutation resulted in the loss of the low-affinity line segment for the d-alanine-glycine transport system. The transport of l-alanine was decreased only 20 to 30% in each of these mutants. A multistep mutant from E. coli W that is 80-fold resistant to d-cycloserine lost >90% of the transport activity for d-alanine and glycine, whereas 75% of the transport activity for l-alanine was retained. E. coli W could utilize either d- or l-alanine as a carbon source, whereas the multistep mutant could only utilize l-alanine. Thus, a functioning transport system for d-alanine and glycine is required for both d-cycloserine action and growth on d-alanine.     

Active transport of D-alanine and related amino acids by whole cells of Bacillus subtilis

Whole cells of Bacillus subtilis transported d-alanine and l-alanine by two different systems. The high-affinity system (K(m) of 1 muM and V(max) of 0.6 to 0.8 nmol/min per mg of protein) was specific for the two stereoisomers of alanine. The low-affinity system (K(m) of 10 muM for l-alanine and 20 muM for d-alanine and glycine) had a V(max) of 5 to 12 nmol/min per mg of protein. This system transported glycine, d-cycloserine, and d-serine, in addition to d- and l-alanine. Azide inhibited the uptake of these amino acids and caused the efflux of d-alanine from preloaded cells. These data suggest that transport of these amino acids is energized by the electron transport chain.     

Studies on Escherichia coli enzymes involved in the synthesis of uridine diphosphate-N-acetyl-muramyl-pentapeptide

The specific activities of l-alanine:d-alanine racemase, d-alanine:d-alanine ligase, and the l-alanine, d-glutamic acid, meso-diaminopimelic acid, and d-alanyl-d-alanine adding enzymes were followed during growth of Escherichia coli. The specific activities were nearly independent of the growth phase. d-Alanine:d-alanine ligase was inhibited by d-alanyl-d-alanine, d-cycloserine, glycine, and glycyl-glycine. l-Alanine:d-alanine racemase was found to be sensitive to d-cycloserine, glycine, and glycyl-glycine. The l-alanine adding enzyme was inhibited by glycine and glycyl-glycine.     

Effects of some putative amino acid neurotransmitters on the stimulated TSH secretion in male rats

The importance of several amino acids (glycine, L-glutamic acid, L-serine, taurine and beta-alanine) in the regulation of the stimulated secretion of TSH was studied in male rats using both peripheral and central administration of the amino acids. Glycine (10-200 mg/kg i.p.), L-glutamic acid (10-500 mg/kg i.p.) and L-serine (500 mg/kg i.p.) decreased significantly the cold-induced TSH secretion whereas beta-alanine (1-500 mg/kg i.p.) and taurine (10-100 mg/kg i.p.) were not effective. The effect of L-glutamic acid (100 mg i.p.) was partially antagonized by bicuculline (1 mg/kg i.p.) but not by picrotoxin (1 or 2 mg/kg i.p.). Only glycine (50 and 100 mg/kg i.p.) inhibited the TRH-stimulated TSH secretion. When the intracerebroventricular route was used, L-serine (50 micrograms/rat) decreased the TSH could response whereas glycine and L-glutamic acid (1-50 micrograms/rat) had no clear effect. We conclude that glycine, glutamate and serine inhibit the cold-induced TSH secretion in the male rat. The action of serine and glycine is possibly mediated through the periventricular hypothalamus and the anterior pituitary, respectively. The inhibition caused by glutamate seems to be partially mediated through the bicuculline-sensitive GABA receptors in the hypothalamus. Taurine and beta-alanine play no role in the control of rat TSH secretion.     

Streptomyces dd-carboxypeptidase as transpeptidases. The specificity for amino compounds acting as carboxyl acceptors

The ability of the water-soluble dd-carboxypeptidases of Streptomyces strains albus G, R61, K11 and R39 to perform transpeptidation was studied. The donor was diacetyl-l-lysyl-d-alanyl-d-alanine, and a whole range of amino acids, peptides and structurally related amino compounds were tested for acceptor function. No compound tested was an acceptor for the enzyme from strain albus G whereas the enzymes from strains R61 and K11 could utilize with varying efficiency a wide range of substances including peptides with N-terminal glycine or d-alanine, omega-amino acids, aminohexuronic acids, 6-aminopenicillanic acid and d-cycloserine. Certain peptides, when present in higher concentration, inhibited the transpeptidation observed at lower concentration. The enzyme from strain R39 would not use any dipeptide as an acceptor, but a few compounds that were not glycine or alpha-amino acids of the d-configuration did function thus. These were d-cycloserine and the lactams of meso- or racemic-diaminoadipic acid.     

Induction of amnesia induced by disturbance of memory consolidation by antagonists of glutamate or serotonin receptors will be suppressed by the protein synthesis inhibitor

We have previously found two stages of amnesia evoked by disruption of memory reconsolidation with MK-801 (NMDA glutamate receptors antagonists) application in food aversion conditioned snails. Repeated conditioning restored the food aversion at early stage of amnesia development (<10 days), whereas repeated conditioning 10 days after MK-801 application did not restore the food aversion. In present work, amnesia was induced with MK-801/reminding 24 hours after food aversion conditioning, and antiamnestic effects of NMDA receptor glycine site agonist d-cycloserine were studied at early (3rd day) or late (12th day) stages of amnesia development. D-cycloserine injection and reminding restored memory only 3 days after amnesia induction whereas d-cycloserine injection without reminding was ineffective. D-cycloserine injection and reminding as well as repeated learning 12 days after amnesia induction were also ineffective in memory restoration. Thus, for the first time, it is revealed that NMDA receptor agonist d-cycloserine influences the memory restoration processes only at early but not the later stages of amnesia development.     

DsdX is the second D-serine transporter in uropathogenic Escherichia coli clinical isolate CFT073

d-Serine is an amino acid present in mammalian urine that is inhibitory to Escherichia coli strains lacking a functional dsdA gene. Counterintuitively, a dsdA strain of E. coli clinical isolate CFT073 hypercolonizes the bladder and kidneys of mice relative to wild type during a coinfection in the murine model of urinary tract infection. We are interested in the mechanisms for uptake of d-serine in CFT073. d-Serine enters E. coli K-12 via CycA, the d-alanine transporter and d-cycloserine sensitivity locus. CFT073 cycA can grow on minimal medium with d-serine as a sole carbon source. The dsdX gene of the dsdCXA locus is a likely candidate for an additional d-serine transporter based on its predicted amino acid sequence similarity to gluconate transporters. In minimal medium, CFT073 dsdX can grow on d-serine as a sole carbon source; however, CFT073 dsdX cycA cannot. Additionally, CFT073 dsdXA cycA is not sensitive to inhibitory concentrations of d-serine during growth on glycerol and d-serine minimal medium. d-[(14)C]serine uptake experiments with CFT073 dsdX cycA harboring dsdX or cycA recombinant plasmids confirm that d-serine is able to enter E. coli cells via CycA or DsdX. In whole-cell d-[(14)C]serine uptake experiments, DsdX has an apparent K(m) of 58.75 microM and a V(max) of 75.96 nmol/min/mg, and CycA has an apparent K(m) of 82.40 microM and a V(max) of 58.90 nmol/min/mg. Only d-threonine marginally inhibits DsdX-mediated d-serine transport, whereas d-alanine, glycine, and d-cycloserine inhibit CycA-mediated d-serine transport. DsdX or CycA is sufficient to transport physiological quantities of d-serine, but DsdX is a d-serine-specific permease.     

Glycine modulates N-methyl-D-aspartic acid induced learning facilitation in rats

Summary Pretraining i.p. administration of N-methyl-D-aspartic acid (NMDA) at doses of 10 and 20mg/kg dose-dependently facilitated performance in a water T-maze learning task in rats. The effect of NMDA was inhibited by the competitive NMDA receptor antagonist CGP37849 [(DL)-E(E)-2-amino-4-methyl-5-phosphono-3-pentenoic acid] (CGP) at a dose of 6mg/kg, and by the NMDA receptor complex glycine site antagonist 1-hydroxy-3-amino-2-pyrrolidone (HA-966) at a dose of 10mg/kg. The NMDA site antagonist, when given alone, did not impair learning. The glycine precursor milacemide (2-N-pentylaminoacetamide HCl), at doses of 5 and 10mg/kg accelearted learning acquisition and its effect was antagonized by HA-966. The learning rate was impaired following the administration of NMDA 10mg/kg together with milacemide 5mg/kg when compared with the effect of 10mg/kg NMDA alone.The administration of 5mg/kg NMDA was associated with an elevated tissue concentration of aspartate in the hippocampus, an effect which was antagonized by 6mg/kg of CGP. NMDA at doses of 10 and 20mg/kg elevated the concentration of glycine but decreased the concentration of aspartate, glutamate and glutamine in the cortex and aspartate in the hippocampus. The cortical effects of NMDA 10mg/kg were antagonized by 6mg/kg of CGP. Milacemide at the dose of 10mg/kg elevated glycine, aspartate, glutamate and taurine concentrations. The coadministration of 5 mg/kg NMDA with 5mg/kg milacemide elevated the concentrations of glycine, glutamate and glutamine in the cortex and taurine in the hippocampus. These amino acid levels were higher than after administration of 5mg/kg either agent alone. The results demonstrate a dose-dependent facilitation effect on learning performance by NMDA and glycine receptor agonists. Antagonists at the NMDA and glycine sites counteracted the learning improvement of NMDA, and the glycine site antagonist the effect of milacemide.     

Dietary Zinc Glycine Chelate on Growth Performance,Tissue Mineral Concentrations,and Serum Enzyme Activity in Weanling Piglets

One hundred twenty crossbred piglets (Duroc × Landrace × Yorkshire) were used to determine the effects of dietary zinc glycine chelate on growth performance, tissue mineral concentrations, and serum enzyme activity. All pigs were allotted to four treatments and fed with basal diets supplemented with 0, 50, and 100 mg/kg Zn as zinc glycine chelate or 3,000 mg/kg Zn as zinc oxide (ZnO). After the 35-day feeding trial, results of the study showed that, compared to the control, average daily gain was improved (P < 0.05) for pigs fed 100 mg/kg Zn from zinc glycine chelate or 3,000 mg/kg Zn from ZnO and Zn concentrations in serum and M. longissimus dorsi were significantly enhanced by 100 mg/kg dietary zinc glycine chelate and 3,000 mg/kg ZnO. In addition, supplementation of 100 mg/kg zinc glycine chelate decreased (P < 0.05) the liver Fe level, liver Zn level, spleen Cu level, and kidney Cu level compared to that of the 3,000-mg/kg ZnO group. For feces mineral excretion, 3,000 mg/kg Zn from ZnO greatly increased the concentration of fecal Zn (P < 0.01) and Mn (P < 0.05) compared to that of the control or the 100-mg/kg zinc glycine chelate group. Moreover, alkaline phosphatase and Cu/Zn superoxide dismutase activities of pigs in 100 mg/kg zinc glycine chelate and ZnO treatments were greatly higher than that of the control. The results of present study showed that supplementation with zinc glycine chelate could improve growth and serum enzyme activities and could also decrease zinc excretion in feces in weanling pig compared to high dietary ZnO.     

Blockade of N-methyl-D-aspartate induced convulsions by 1-aminocyclopropanecarboxylates

1-Aminocyclopropanecarboxylic acid is a potent and selective ligand for the glycine modulatory site on the N-methyl-D-aspartate receptor complex. This compound blocks (ED50 234 mg/kg) the convulsions and deaths produced by N-methyl-D-aspartate (125 mg/kg) in a dose dependent fashion. In contrast, 1-aminocyclopropanecarboxylic acid does not protect mice against convulsions induced by pentylenetetrazole (80 mg/kg), strychnine (2 mg/kg), bicuculline (6 mg/kg), or maximal electroshock (50 mA, 0.2 s), and does not impair motor performance on either a rotarod or horizontal wire at doses of up to 2 g/kg. The methyl- and ethyl- esters of 1-aminocyclopropanecarboxylic acid are 5- and 2.3-fold more potent, respectively, than the parent compound in blocking the convulsant and lethal effects of N-methyl-D-aspartate. However, these esters are several orders of magnitude less potent (IC50 greater than 40 microM) than 1-aminocyclopropanecarboxylic acid as inhibitors of strychnine-insensitive [3H] glycine binding, indicating that conversion to the parent compound may be required to elicit an anticonvulsant action. These findings suggest that 1-aminocyclopropanecarboxylates may be useful in the treatment of neuropathologies associated with excessive activation of N-methyl-D-aspartate receptor coupled cation channels.     

D-cycloserine biases enrichment for auxotrophic mutants of Zymomonas mobilis

Abstract Contrary to its effect on rich medium, d-cycloserine showed no bactericidal effect on Zymomonas mobilis cells cultured on mineral medium. Addition of a mixture of glycine and glutamic acid to the mineral medium restored its bactericidal action. However, mutant enrichments run in these conditions were biased, with mostly methionine mutants isolated. A decrease of the d-cycloserine concentration only reduced the bias.     

D-serine transport system in Escherichia coli K-12

The d-serine transport system in Escherichia coli K-12 was studied by use of a mutant unable to form d-serine deaminase, yet resistant to d-serine. The mutant is greatly impaired in its ability to accumulate d-serine, d-alanine, and glycine. Transport of l-alanine is partially affected but transport of l-serine is unaffected. The mutant is also resistant to d-cycloserine, indicating that d-serine is transported by the system responsible for uptake of d-cycloserine. The d-serine transport system is not inducible, but appears to be formed constitutively, as are the transport systems of most amino acids. The transport mutation appears to be multistep and maps to the right of malB on the E. coli linkage map.     

Resistance to d-Cycloserine in the Tubercle Bacilli: Mutation Rate and Transport of Alanine in Parental Cells and Drug-Resistant Mutants

A single transport system was found to accumulate l- and d-alanine, glycine and d-serine in Mycobacterium tuberculosis. The results of inhibition experiments suggested that the antibiotics d-cycloserine and O-carbamyl-d-serine were also transported by the alanine-glycine-d-serine system. A d-cycloserine-resistant permease-competent (d-CS(r)/perm(+)) mutant and a d-cycloserine-resistant permease-defective (d-CS(r)/perm(-)) mutant were isolated. The d-CS(r)/perm(-) mutant was not found to be more resistant to the drug than was the d-CS(r)/perm(+) mutant. The data were consistent with the conclusion that resistance to d-cycloserine in the tubercle bacilli is primarily due to mutations in the gene(s) controlling the enzyme d-alanyl-d-alanine synthetase. The mutation rate was calculated to be about 10(-10) mutations per bacterium per generation.     

Selective increase in the extracellular D-serine contents by D-cycloserine in the rat medial frontal cortex

A partial agonist of the N-methyl-D-aspartate (NMDA) receptor, D-cycloserine, acting at its glycine modulatory site, ameliorates the neuropsychiatric symptoms that are mimicked by NMDA antagonists and include cognitive disturbances, antipsychotic-resistant schizophrenic symptoms and cerebellar ataxia. To obtain a further insight into the mechanisms of the therapeutic efficacies of D-cycloserine, we investigated the effects of the systemic administration of D-cycloserine on the extracellular contents of an endogenous NMDA co-agonist, D-serine, in the medial frontal cortex of the rat using an in vivo dialysis technique. An acute intraperitoneal injection of D-cycloserine (50 and 100 mg/kg) caused an increase in extracellular concentrations of D-serine without significant effects on those of L-serine, glycine, L-glutamate, L-aspartate, L-glutamine, L-asparagine, L-alanine, L-threonine and taurine in the medial frontal cortex. The selective increase in the extracellular D-serine contents may, at least partially, be associated with the facilitating effects of D-cycloserine on the NMDA receptor functions in addition to its direct stimulation of the NMDA receptor glycine site.     

The effects of various cognition-enhancing drugs on in vitro rat hippocampal synaptosomal sodium dependent high affinity choline uptake

The purpose of the present study was to compare the effect of seven drugs, that have been reported to enhance cognitive functions, on rat hippocampal cholinergic neuronal activity. The latter was assessed by measuring the effects of the drugs on in vitro sodium-dependent high affinity choline uptake (HACU) into rat hippocampal synaptosomes 30 minutes after their in vivo administration. 3,4-Diaminopyridine (0.1 mg/kg IP), like pramiracetam (44 and 88 mg/kg IP), increased HACU with higher or lower doses being ineffective. Centrophenoxine (100 mg/kg IP) decreased HACU. Piracetam (100 and 500 mg/kg IP), aniracetam (10-200 mg/kg PO), lysine vasopressin (0.005-0.05 mg/kg IM) and 4-aminopyridine (0.01-3.0 mg/kg IP) were ineffective. The results indicate that 3,4-diaminopyridine and centrophenoxine, like pramiracetam may be increasing cognitive function in part by affecting hippocampal cholinergic neuronal activity. In addition, the findings indicate the usefulness of using in vitro HACU as a biochemical measurement to assess the potential effect of cognitive-enhancing drugs on cholinergic neuronal activity in vivo.     

Combined embryotoxic action of toluene, a widely used industrial chemical, and acetylsalicylic acid (aspirin)

CFY rats were exposed to inhalation of fresh air at days 10-13 of gestation; at day 12 the dams were given 0, 125, 250, 500, or 1,000 mg/kg acetylsalicylic acid (ASA) by gavage. During the same period of gestation (days 10-13) further groups of rats were exposed to toluene at 1,000, 2,000, and 3,600 mg/m3 atmospheric concentration and were given 250 mg/kg ASA by gavage; two subgroups of animals treated with 250 mg/kg ASA in combination with 3,600 mg/m3 toluene inhalation were given 0, 2.5, or 5 gm/kg glycine 2 hours before the ASA dose. At day 21 the animals were killed and examined for teratogenic effects and histological changes. After 48 hours toluene exposure other groups of rats were treated with ASA or with ASA plus glycine (administered 2 hours earlier) on day 20 of gestation. These animals were killed 2 hours later and the salicylic acid concentration in maternal and embryonic plasma and in amniotic fluid was measured by gas chromatography. With the rising ASA doses both maternal toxicity (increased mortality, decreased food consumption, and weight gain) and embryonic toxicity (postimplantation loss, increased incidence of weight-retarded fetuses, increased minor anomalies and malformations, decreased average weight of fetuses) increased. Toluene was found to potentiate the toxic effect of ASA and to increase both maternal and embryonic toxicity. The type of ASA-induced minor anomalies and malformations was also found to be altered under the effect of toluene pretreatment. By raising the toluene concentration the salicylic acid level in the maternal and embryonic plasma and in the amniotic fluid was increased above the expected concentration. The mechanism of the potentiating interaction should be looked for in the depletion of the glycine pool by toluene (and its metabolites) and in the resultant increase of salicylic acid level. Increasing ASA embryotoxicity caused by toluene can be warded off by glycine administration.     

Hyperglycinemia due to folate deficiency in rats: evidence for the lack of involvement of the hepatic glycine cleavage system

In addition to a well-recognized hyperhomocysteinemic state, folate deficiency also leads to profound hyperglycinemia. To further characterize the latter observation, two trials were conducted using a folate-deficient rat model to (1) determine the sensitivity of plasma glycine to folate repletion and (2) test the hypothesis that hyperglycinemia results from a reduced flux through the folate-dependent glycine cleavage system (GCS). Weanling male Sprague–Dawley rats were used, and they consumed an amino acid-defined diet with either 0 (FD) or 1 (FA) mg/kg of crystalline folic acid. In Trial 1, 30 rats consumed the FD diet for 28 days. Rats then consumed diets containing 0.1, 0.2, 0.3 or 0.4 mg/kg of folic acid for 14 days before termination. In Trial 2, 16 rats were allocated to receive either the FA (n=8) or FD (n=8) diet for 30 days before termination. Liver mitochondria were isolated and flux through the GCS (measured as ¹⁴CO₂ production from 1-¹⁴C-glycine) was determined. Plasma from blood collected at termination was analyzed for folate, homocysteine and glycine. In Trial 1, both homocysteine and glycine responded linearly to increased dietary folic acid (milligrams per kilogram) levels (P<.05). In Trial 2, plasma folate (FA=25.85 vs. FD=0.66; S.E.M.=1.4 μM), homocysteine (FA=11.1 vs. FD=55.3; S.E.M.=1.7 μM) and glycine (FA=564 vs. FD=1983; S.E.M.=114 μM) were significantly affected by folate deficiency (P<.0001). However, glycine flux through hepatic GCS was not affected by folate deficiency (P>.05). These results provide evidence that in a folate-deficient rat model, both homocysteine and glycine are sensitive to dietary folic acid levels; however, the observed hyperglycinemia does not appear to be related to a reduced flux through the hepatic GCS.     

Regional Heterogeneity of Nicotine Effects on Neurotransmitters in Rat Brains <Emphasis Type="Italic">in vivo</Emphasis> at Low Doses

In our recent studies on nicotine-induced changes in neurotransmitters in brain areas associated with cognitive function using a nicotine dose of 0.5 mg/kg administered subcutaneously to conscious freely moving rats, we found changes in dopamine, norepinephrine, and serotonin, and their metabolites, in the areas examined. For the present report we examined changes in these neurotransmitters following administration of lower nicotine doses, to test regional differences in nicotine response and possible threshold levels for some effects of nicotine. The doses used were 0.15 mg/kg and 0.03 mg/kg nicotine administered subcutaneously. Nicotine levels in the brain reached peak values in less than 10 min and decreased with a half-life of about 60 min (0.15 mg/kg) or 30 min (0.03 mg/kg) to values below detection limits (1 ng/g), by the later time points of the 0.03 mg/kg experiments. Nicotine-induced dopamine (DA) increase (and increase in DA metabolites) and decrease in 5-HT levels at 0.15 mg/kg were significant in the cortex, less so in the hippocampus. Norepinephrine (NE) increase at 0.15 mg/kg was much less significant than found previously at 0.5 mg/kg. At a low nicotine dose (0.03 mg/kg), the significant changes observed were a decrease in 5-HT in the hippocampus and small increases of DA and NE in the prefrontal cortex and of NE in the medial temporal cortex. In the nucleus accumbens DA, NE, and 5-HT and their metabolites in the ventral tegmental area, mostly DA and metabolites were increased. We conclude that in areas of cognitive function nicotine-induced DA changes are more concentration dependent than changes in NE or 5-HT, and that there are regional differences in neurotransmitter changes induced by nicotine, with NE changes detectable only in the cortex and 5-HT changes only in the hippocampus at a low nicotine dose, indicating significant regional variation in sensitivity to nicotine-induced neurotransmitter changes in brain areas associated with cognitive function. The decrease in 5-HT shows that nicotine also has indirect effects caused by neurotransmitters released by nicotine. The effects of low nicotine dose are more significant in areas of reward function, indicating differences in sensitivity between cognitive and reward functions.     

Dehydroevodiamine.HCl prevents impairment of learning and memory and neuronal loss in rat models of cognitive disturbance

We previously reported that dehydroevodiamine.HCl (DHED) has anticholinesterase and antiamnesic activities. To verify the effects of DHED on cognitive deficits further, we tested it on the scopolamine-induced amnesia model of the rat using the passive avoidance and eight-arm radial maze tests. A single (20 mg/kg p.o.) and repeated (10 mg/kg p.o.) administrations of DHED could significantly reverse the latency time shortened by scopolamine (1 mg/kg i.p.) to control level. The impaired spatial working memory induced by scopolamine (1 mg/kg i.p.) was also improved significantly by a single injection (6.25 mg/kg i.p.) and repeated administrations of DHED (10 mg/kg p.o.) in the eight-arm radial maze test. In addition, we examined the effects of DHED on the memory impairment and the histological changes of the brain after unilateral electrolytic lesion of the entorhinal cortex (EC) and middle cerebral artery occlusion in rats. The cognitive deficits caused by EC lesion and middle cerebral artery occlusion were improved significantly by repeated administrations of DHED (6.25 mg/kg i.p.) after EC lesion or ischemic insult once a day for 7 days in the passive avoidance test. Histological analysis showed that the neuronal loss in the DHED-treated group was notably reduced in the hippocampal area (CA1) of ischemic rats and in the dentate gyrus and hippocampal area (CA1 and CA3) of EC-lesioned rats compared with the nontreated group. The infarction area was decreased significantly by a single administration of DHED (6.25 mg/kg i.p.) 30 min before ischemic insult for 6 h. These results suggest that DHED might be an effective drug for not only the Alzheimer's disease type, but also the vascular type of dementia.