Temperature dependency of basal and evoked release of amino acids and calcitonin gene-related peptide from rat dorsal spinal cord

Moderate hypothermia significantly diminishes consequences of spinal and cerebral anoxia. One component of this neuroprotection has been hypothesized to be suppression of excitotoxic transmitter release. Whether this suppression is attributable to reduced hypoxic injury that induces release or an alteration of the release process itself is unclear. We sought to characterize the temperature sensitivity (Q10) of basal and evoked calcitonin gene-related peptide (CGRP) and amino acid release from dorsal horn slices of rat spinal cord over a range of temperatures from 40 to 8 degrees C. At 40 degrees C, potassium (60 mM) and capsaicin (10 microM) evoked a 21- and 32-fold increase in basal CGRP concentrations, respectively. Capsaicin had no effect on glutamate release, but potassium evoked a 2.7-fold increase. Release evoked by either potassium or capsaicin was reduced in a biphasic fashion with declining temperature. Over the range of 40 to 34 degrees C, the Q10 values for evoked release for CGRP were 11.3 (potassium) and 39.7 (capsaicin) and for glutamate, 5. 5 (potassium). Over the range of 34 to 8 degrees C, Q10 values were near unity for all evoked release (0.8 and 1.3 for CGRP and 1.2 for glutamate). Although serine, glycine, glutamine, taurine, and citrulline showed no evoked release, basal levels were reduced at temperatures below 34 degrees C. The pronounced temperature dependency of evoked transmitter release between 40 and 34 degrees C is consistent with the profound cerebral protection observed with mild hypothermia in which metabolic activity is only slightly depressed.     

The neurotoxicity of sulfur-containing amino acids in energy-deprived rat hippocampal slices

The rat hippocampal slice preparation and its electrophysiology were used to assess the toxicity of two sulfur-containing amino acids, L-cysteate (CA) and L-cysteine (CYS). Both compounds were innocuous under normal conditions but became toxic in energy-deprived (lack of oxygen or glucose) slices. CA and CYS toxicity was apparent as both reduced the number of slices that normally recover their neuronal function (evoked CA1 population spike) after a standardized period of hypoxia or glucose deprivation (GD). The competitive N-methyl-D-aspartate (NMDA) antagonist DL-2-amino-5-phosphonovalerate blocked the toxicity of both CA and CYS in hypoxic slices, but it was effective only against CYS toxicity in glucose-deprived slices. The glycine antagonist 7-chlorokynurenate blocked CA and CYS toxicity in hypoxic slices but was unable to block their toxicity in glucose-deprived tissue. Perfusing slices with medium containing a high magnesium concentration blocked the toxicity of CA in both hypoxic and glucose-deprived slices, a treatment that was ineffective against CYS toxicity under either condition. Calcium depletion from the perfusion medium completely blocked the damaging effect of both amino acids in hypoxic slices, but it only partially blocked the toxicity of CA and did not block that of CYS in glucose-deprived slices. These results suggest that CA and CYS activate different NMDA receptor subsets and other glutamate receptor subtypes. Moreover, the results indicate a possible difference between the mechanism that lead to hypoxic neuronal damage and the one that lead to hypoglycemic neuronal damage.     

Differential effects of wortmannin on the release of substance P and amino acids from the isolated spinal cord of the neonatal rat

Mice lacking hepatic lipase have been reported to express mild hyperlipidemia characterized by increased concentrations of large high density lipoproteins, but normal concentrations of lipoproteins containing apolipoprotein B. Whereas hepatic lipase has been implicated in the clearance and processing of chylomicron remnants in rats, no such defect was found in these mice. We have further characterized the abnormal lipoprotein phenotype in young hepatic lipase-deficient mice and have found more pronounced elevations of high density lipoproteins associated in particular with a 5-fold increase in plasma concentrations of apolipoprotein E. In addition, there was a reduction in the concentration of low density lipoproteins containing apolipoprotein B-100 and B-48 relative to precursor lipoproteins of lower density and a pronounced deficiency of apolipoprotein B-containing low density lipoproteins with density exceeding 1.029 g/mL. Conversion of radiolabeled rabbit intermediate density lipoproteins to low density lipoproteins was reduced by 6-fold as compared with wild-type mice. Although clearance of cholesteryl ester-labeled chylomicrons from the blood was unimpaired in the deficient mice, that of chylomicron remnants was reduced. Furthermore, endocytosis of chylomicron cholesteryl esters into liver cells occurred more rapidly than in wild-type mice. The unimpaired hepatic clearance of injected chylomicron particles in hepatic lipase-deficient mice may be the result of greater acquisition of apoE from high density lipoproteins during remnant formation. These studies thus demonstrate a critical role for mouse hepatic lipase in the formation of small, dense low density lipoproteins, as well as participation in the normal clearance and processing of chylomicron remnants.     

Monoaminergic control of the release of calcitonin gene-related peptide- and substance P-like materials from rat spinal cord slices

The possible control by monoamines of the spinal release of substance P- and calcitonin gene-related peptide-like materials (SPLM and CGRPLM, respectively) was investigated in vitro, using slices of the dorsal half of the rat lumbar enlargement superfused with an artificial cerebrospinal fluid. Whereas the spontaneous outflow of SPLM and CGRPLM was changed by none of the agonists/antagonists of monoamine receptors tested, the overflow of both peptide-like materials due to 30 mM K+ was differentially affected by alpha 2-adrenoreceptor and dopamine D-1 receptor ligands. Noradrenaline (10 microM to 0.1 mM) and clonidine (0.1 mM) significantly reduced the K(+)-evoked overflow of SPLM, and both effects could be prevented by idazoxan (10 microM) and prazosin (10 microM) as expected from their mediation through the stimulation of alpha 2B-adrenoreceptors. In contrast, CGRPLM overflow remained unaffected by alpha 2-adrenoreceptor ligands. Dopamine D-1 receptor stimulation by SKF 82958 (10-100 nM) significantly increased the K(+)-evoked overflow of both SPLM and CGRPLM, and this effect could be prevented by the selective D-1 antagonist SCH 39166 (1 microM). Further studies with selective ligands of other monoamine receptors indicated that neither alpha 1- and beta-adrenergic receptors, dopamine D-2, nor serotonin 5-HT1A and 5-HT3 receptors are apparently involved in some control of the spinal release of CGRPLM and SPLM. These data are discussed in line with the postulated presynaptic control by monoamines of primary afferent fibres conveying nociceptive messages within the dorsal horn of the spinal cord.     

Na+ channel block prevents the ischemia-induced release of norepinephrine from spinal cord slices

To assess the role of beta-phenylethylamine in aspects of dopamine release, we measured the level of beta-phenylethylamine in the rat striatum after killing the rats by microwave irradiation. We then investigated the effect of beta-phenylethylamine on electrically evoked dopamine release from rat striatal slices in vitro. The striatal beta-phenylethylamine level was 46.5 +/- 3.5 ng/g wet tissue, equivalent to 0.3 micromol/l. Superfusion with low concentrations of beta-phenylethylamine up to 1 micromol/l had no effect on spontaneous or electrically evoked dopamine release from striatal slices. Quinpirole reduced the evoked dopamine release from slices in a concentration-dependent manner. The quinpirole-induced reduction of evoked dopamine release was attenuated 30% by superfusion with 0.3 micromol/l beta-phenylethylamine. Moreover, the (-)-sulpiride (0.1 micromol/l)-induced increase in evoked dopamine release was also attenuated by superfusion with 0.3 micromol/l beta-phenylethylamine. These data indicate that submicromolar levels of beta-phenylethylamine could modify the dopamine autoreceptor mediated changes in evoked dopamine release from rat striatal slices.     

The serotonin receptor agonist 5-methoxy-N,N-dimethyltryptamine facilitates noradrenaline release from rat spinal cord slices and inhibits monoamine oxidase activity

1. The influences of the purported serotonergic agonist 5-methoxy-N,N-dimethyltryptamine (MeODMT) on noradrenaline release and metabolism were investigated in a rat spinal cord release model and a monoamine oxidase (MAO) assay. 2. MeODMT inhibited the basal outflow of tritium from rat spinal cord slices preincubated with [3H]noradrenaline and enhanced the electrically-evoked overflow. 3. Effects on basal outflow were not observed, when monoamine oxidase (MAO) was inhibited by pargyline. Effects on the evoked overflow were not observed in the presence of metitepine or phentolamine. 4. Preferential inhibition by MeODMT of MAO A-type enzyme activity was found in a direct assay. 5. The results provide evidence for two different effects by which MeODMT reinforces noradrenergic neurotransmission in the rat spinal cord: facilitation of stimulation-evoked noradrenaline release and inhibition of noradrenaline metabolism by MAO inhibition.     

Passive electrical properties of ventral horn neurons in rat spinal cord slices

Recordings were made from large neurons located in the ventral horn of transverse spinal cord slices from young rats (7-15 days). Whole cell recordings were made simultaneously with two electrodes from the soma of these neurons, visualized using infra-red differential interference contrast optics. Positive identification of motoneurons could not always be achieved. The response of a neuron to a brief pulse of current delivered by one electrode, and recorded by the other electrode, were matched optimally to responses of a compartmental model of the same neuron with an identical current pulse as input. The compartmental model was based on a reconstruction of the neuron, using Biocytin staining. The compartmental model had three free parameters: specific membrane capacitance (Cm), membrane resistivity (Rm), and cytoplasmatic resistivity (Ri), all assumed to be uniform throughout the neuron. The experimental and model responses could be matched unequivocally for four neurons, giving Cm = 2.4 +/- 0.5 microF/cm2, Rm = 5.3 +/- 0. 9 kOmega/cm2, and Ri = 87 +/- 22 Omega/cm. No somatic shunt was required. For the remaining six neurons, a less perfect fit (but still within 95% confidence limits) was indicative of nonhomogeneous membrane properties. The electrotonic length of uncut dendrites was 0.85 +/- 0.14 lambda. The results resolve the issue of a somatic shunt conductance for motoneurons, relegating it to a microelectrode impalement artifact. They are consistent with previous reports on the electrical compactness of motoneurons to steady state currents and voltages. However, the much higher value of Cm (than the previously assumed 1 microF/cm2) implies much greater dendritic attenuation of fast synaptic potentials, and a much enhanced integrative response of motoneurons to synaptic potentials.     

Passive electrical properties of ventral horn neurons in rat spinal cord slices

Recordings were made from large neurons located in the ventral horn of transverse spinal cord slices from young rats (7-15 days). Whole cell recordings were made simultaneously with two electrodes from the soma of these neurons, visualized using infra-red differential interference contrast optics. Positive identification of motoneurons could not always be achieved. The response of a neuron to a brief pulse of current delivered by one electrode, and recorded by the other electrode, were matched optimally to responses of a compartmental model of the same neuron with an identical current pulse as input. The compartmental model was based on a reconstruction of the neuron, using Biocytin staining. The compartmental model had three free parameters: specific membrane capacitance (Cm), membrane resistivity (Rm), and cytoplasmatic resistivity (Ri), all assumed to be uniform throughout the neuron. The experimental and model responses could be matched unequivocally for four neurons, giving Cm = 2.4 +/- 0.5 microF/cm2, Rm = 5.3 +/- 0. 9 kOmega/cm2, and Ri = 87 +/- 22 Omega/cm. No somatic shunt was required. For the remaining six neurons, a less perfect fit (but still within 95% confidence limits) was indicative of nonhomogeneous membrane properties. The electrotonic length of uncut dendrites was 0.85 +/- 0.14 lambda. The results resolve the issue of a somatic shunt conductance for motoneurons, relegating it to a microelectrode impalement artifact. They are consistent with previous reports on the electrical compactness of motoneurons to steady state currents and voltages. However, the much higher value of Cm (than the previously assumed 1 microF/cm2) implies much greater dendritic attenuation of fast synaptic potentials, and a much enhanced integrative response of motoneurons to synaptic potentials.     

NMDA receptor activation modulates evoked release of substance P from rat spinal cord

The possible modulation exerted by glutamate on substance P (SP) release from the rat spinal cord has been investigated. The N-methyl-D-aspartate (NMDA) receptor agonist, NMDA (1 microM), increased SP basal outflow by 46.5+/-10.9% (n = 3, P<0.01) without changing the evoked release of the peptide. Conversely, NMDA antagonists but not 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) inhibited both electrically-evoked and capsaicin-induced release of SP. In particular, D-2-amino-5-phosphonopentanoate (D-AP5; 50 microM) inhibited electrically-evoked and capsaicin-induced release of SP by 93+/-2.4% and 93.2+/-3.8% (n = 12, P<0.01), respectively. Functional pharmacological evidence is provided for glutamate exerting a positive feedback on SP release evoked by C fibre stimulation via NMDA receptor activation.     

Transient spinal ischemia in rat: characterization of spinal cord blood flow, extracellular amino acid release, and concurrent histopathological damage

Extracellular concentrations of amino acids in halothane-anesthetized rats were measured using a microdialysis fiber inserted transversely through the dorsal spinal cord at the level of the lumbar enlargement in conjunction with HPLC and ultraviolet detection. After a 2-h washout and a 1-h control period, 20 min of reversible spinal cord ischemia was achieved by the inflation of a Fogarty F2 catheter passed through the femoral artery to the descending thoracic aorta. After 2 h of postischemic reperfusion, animals were transcardially perfused with saline followed by 10% formalin or 4% paraformaldehyde. The glutamate concentration in the dialysate was significantly elevated after 10 min of occlusion and returned to near-baseline during the first 30 min of reperfusion. Taurine was elevated significantly 0.5 h postocclusion and continued to increase throughout the 2 h of reperfusion. Glycine concentrations showed a tendency to be slightly above baseline during the reperfusion period. Glutamine concentrations modestly increased following 2 h of reperfusion. No significant changes in aspartate, asparagine, and serine were detected. In control animals no significant changes in any amino acids were detected. To assess the role of complete spinal ischemia on spinal glutamate release, studies were carried out using cardiac arrest. Twenty minutes after induction of cardiac arrest, the glutamate concentration was increased about 350-400%. In a separate group of animals, spinal cord blood flow (SCBF) and its response to decreased CO2 were measured using a laser probe implanted into the epidural space at the level of the L2 vertebral segment. SCBF decreased to 5-6% of the control during aortic occlusion. After reversible ischemia, marked hyperemia was seen for the first 15 min, followed by hypoperfusion at 60 min. Under control-preischemic conditions a decrease in arterial CO2 content caused a decrease in SCBF of about 25%. This autoregulatory response was almost completely absent when assessed 60 min after a 20-min interval of aortic occlusion. Histopathological analysis of spinal cord tissue from these animals demonstrated heavy neuronal argyrophilia affecting small and medium-sized neurons located predominantly in laminae III-V. These changes corresponded to signs of irreversible damage at the ultrastructural level. Occasionally, small areas of focal necrosis, located in the dorsolateral part of the dorsal horn and anterolateral part of the ventral horn, were found. The results are consistent with a role for glutamate in ischemically induced spinal cord damage and suggest that taurine elevation detected during the early reperfusion period may serve as an important indicator of irreversible spinal cord neuronal damage.     

Dimethylamine formation in the rat from various related amine precursors

Inflammatory processes are a key event in various respiratory disorders. Adhesion to microvascular endothelial cells is essential for the migration of leukocytes into inflamed tissues. Intercellular adhesion molecule-1 (ICAM-1) is a member of the immunoglobulin superfamily and a ligand for lymphocyte function-associated antigen-1 alpha (LFA-1 alpha). It has been shown to be required for leukocyte migration into inflamed tissues, and to play a significant role in inflammatory diseases, including rhinovirus infections. Clinical and experimental investigations have demonstrated a close relationship between the ICAM-1/LFA-1 alpha pathway and the pathogenesis of asthmatic responses and acute respiratory distress syndrome. Further progress in research on the antagonism of ICAM-1 and LFA-1 alpha may provide new therapeutic approaches to inflammatory lung diseases.     

Chromogranin A inhibits dopamine release from rat striatal slices

Chromogranin A (CGA), a prohormone and a protein component of endocrine and neural secretory granules, neuritic plaques in Alzheimer's disease and Lewy bodies in Parkinson's disease, inhibited the release of dopamine (DA) from perfused rat striatal slices. Dopamine release was stimulated by a pulse of high potassium (40mM) medium introduced at 20 minutes (K1) and 55 minutes (K2) following equilibration. The ratio of K2/K1 was 0.80+/-0.04 in control tissues, but fell significantly to 0.26+/-0.08 when 100nM purified CGA was added prior to the second potassium pulse. This reduction in DA release was equivalent to that seen when calcium was excluded from the buffer (0.19+/-0.05). Pancreastatin, a centrally active peptide product of CGA, had no effect on stimulated DA release (0.77+/-0.06), although it, as well as the other treatments, did reduce basal DA release. It is likely that the parent molecule itself, CGA, or an as yet unidentified product is responsible for inhibition of K-stimulated striatal DA release.     

NMDA and AMPA receptors evoke transmitter release from noradrenergic axon terminals in the rat spinal cord

The organic extracts of soil collected at parks in residential areas in Osaka and neighboring cities in the Kansai area, Japan, showed mutagenicity in Salmonella typhimurium strain TA98 in the presence or absence of a mammalian metabolic activation system (S9 mix). The soil extracts from Ibaraki and two different sites in Osaka, i.e., Sumiyoshi-ku and Minato-ku, were mutagenic in strain TA100 as well as in strain TA98. Direct-acting mutagenicity of soil extracts from Sumiyoshi-ku and Minato-ku toward strain TA98 were 66 or more times higher than that of the other cities. Both extracts exerted stronger mutagenicity in strains YG1021 and YG1024 than TA98 and TA100, and the potency was especially high in strain YG1024: Sumiyoshi-ku, 153 000 revertants/g of soil; and Minato-ku, 246 000 revertants/g of soil. Two mutagenic compounds (I and II) were isolated from the Soxhlet extract of soil from the park in Sumiyoshi-ku by repetitive separation using normal-phase and reversed-phase column chromatography. By comparing the mass and UV spectra and retention times for HPLC on two individual ODS columns of compounds I and II with those of authentic chemicals, we identified these two compounds as 1,6- and 1,8-dinitropyrene (DNPy) isomers. Amounts of DNPy isomers in soil from Sumiyoshi-ku and Minato-ku were 1.7-2.2 ng/g. Forty-three percent and 40% of the mutagenicity of soil from Sumiyoshi-ku and Minato-ku could be attributed to these DNPy isomers, respectively.     

Met-enkephalin inhibits 5-hydroxytryptamine release from the rat ventral spinal cord via delta opioid receptors

The effect of opioid receptor agonists and antagonists on the electrically evoked release of endogenous serotonin (5-hydroxytryptamine, 5-HT) was studied in superfused slices of the rat ventral lumbar spinal cord. Met-ENK (1 x 10(-8)M-1 x 10(-6)M) and DPDPE (1 x 10(-8)M-1 x 10(-6)M) reduced the evoked 5-Ht release in a concentration dependent fashion. DAMGO (1 x 10(-8)-1 x 10(-6)) and (-)-trans-(1S,2S)-U-50488 (1 x 10(-6)M) had no effect on the 5-HT release. The inhibitory effect of met-ENK was completely abolished by ICI-174,864, but neither by naloxonazine nor nor-binaltorphimine. Following i.c.v. treatment with 5,7-dihydroxytryptamine (5,7-DHT), the tissue concentration of 5-HT was reduced by 97%, whereas the concentration of noradrenaline was reduced by only 5%. The tissue concentration of met-ENK, as measured by radioimmunoassay, was not significantly altered. The results suggest that met-ENK is present in the rat ventral spinal cord mainly in non-serotonergic nerve terminals and exerts an inhibitory action on 5-HT release via delta opioid receptors.     

Structure-activity relations of excitatory amino acids on frog and rat spinal neurones

1 A series of compounds structurally related to glutamic acid has been tested on frog and rat spinal neurones. The substances were added to procaine-containing medium bathing the isolated hemiscected spinal cord of the frog, and their potencies in depolarizing motoneurones were assessed by the magnitude of the potential produced in the ventral root. The electrophoretic technique was used to administer the substances around single interneurones of the rat spinal cord and the relative potencies of the compounds as excitants assessed by the magnitude of the currents required to produce similar rates of neuronal firing. 2 Parallel structure-activity relations were observed in the two series of experiments, suggesting that the receptors for excitatory amino acids on frog and rat spinal neurones are similar. 3 Quisqualate, domoate and kainate were the strongest excitants in both animals, with potencies around two orders of magnitude higher than that of L-glutamate. 4 2,4,5-Trihydroxyphenylalanine (6-OH-DOPA) was a stronger excitant and L-3,4-dihydroxyphenylalanine (L-dopa) a weaker excotamt than L-glutamate on frog spinal motoneurones. The former compounds was also a more potent convulsant than L-glutamate on intraventricular injection into mouse brain. The lack of activity of 6-OH-DOPA on electrophoretic administration was attributed to oxidation. 5 Unlike the majority of amino acid excitants, several of the compounds shown in the present work to have moderate excitatory activity are not anionic at physiological pH. This indicates either that two negatively charged groups are not essential for interaction with a common excitatory receptor, or that more than one type of receptor is involved in the actions demonstrated.     

Interactions between excitatory amino acids and tachykinins in the rat spinal dorsal horn

Whole-cell patch-clamp technique of freshly isolated rat spinal dorsal horn (DH) neurons, intracellular recording from DH neurons in a slice preparation, and high performance liquid chromatography with fluorimetric detection of release of endogenous glutamate and aspartate from spinal cord slice following activation of primary afferent fibers were employed to investigate interactions between excitatory amino acids (EAA) and tachykinins [substance P (SP) and neurokinin A (NKA)]. Potentiation of N-methyl-D-aspartate (NMDA)-, quisqualate (QA)- and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)-, but not kainate-induced currents by SP and NKA was found. Spantide II, a claimed novel nonselective tachykinin antagonist, effectively blocked the SP (2 nM)-induced potentiation of the responses of DH neurons to NMDA. In the presence of glycine (0.1 microM), the SP-evoked increase of the NMDA-induced current was prevented. However, 7-chlorokynurenic acid (2 microM), a competitive antagonist at the glycine allosteric site of the NMDA receptor, led to the reestablishment of the SP effect. Brief high frequency electrical stimulation of primary afferent fibers produced a long-lasting potentiation of presumed monosynaptic and polysynaptic excitatory postsynaptic potentials and sustained enhanced release of endogenous glutamate (218.3 +/- 66.1%) and aspartate (286.3 +/- 58.0%). Possible functional implications of the observed phenomena are discussed in relation to transmission and integration of sensory information, including pain.     

Primary afferent-evoked release of immunoreactive galanin in the spinal cord of the neuropathic rat

We have determined if peripheral nerve stimulation altered the increased spontaneous release of immunoreactive (ir)-galanin that is found in the superficial dorsal horn of the spinal cord of neuropathic rats. Using the antibody microprobe technique to study the localized sites of ir-galanin release in vivo, we found that high intensity electrical stimulation of the injured nerve resulted in a further increase in ir-galanin release in the superficial dorsal horn, with no significant persistence of ir-galanin after release. Release of ir-galanin at stimulus strengths sufficient to activate C fibres, in an area of the spinal cord thought to be concerned with nociceptive transmission, indicates a possible role for this peptide in the spinal modulation of pain after peripheral nerve injury.     

Nocistatin reverses nociceptin inhibition of glutamate release from rat brain slices

We have examined the effects of the recently described heptadecapeptide nocistatin on K+-evoked glutamate release from rat cerebrocortical slices in vitro. In vivo, nocistatin reverses the action of nociceptin. Nocistatin (100 nM, n = 7) did not inhibit K+-evoked glutamate release alone. Nociceptin (100 nM) inhibited glutamate release by 51.7 +/- 8.3% (P < 0.05, n = 6) and this was fully reversed by nocistatin (100 nM). Nocistatin also appears to be an antagonist of nociceptin action in vitro.     

Influence of adrenergic agonists on the release of amino acids from rat skeletal muscle studied by microdialysis

Epidermal growth factor (EGF) has been considered to be a candidate for neurotrophic factors on the basis of the results of several in vitro studies. However, the in vivo effect of EGF on ischemic neurons as well as its mechanism of action have not been fully understood. In the present in vivo study using a gerbil ischemia-model, we examined the effects of EGF on ischemia-induced learning disability and hippocampal CA1 neuron damage. Cerebroventricular infusion of EGF (24 or 120 ng/d) for 7 days to gerbils starting 2 hours before or immediately after transient forebrain ischemia caused a significant prolongation of response latency time in a passive avoidance task in comparison with the response latency of vehicle-treated ischemic animals. Subsequent histologic examinations showed that EGF effectively prevented delayed neuronal death of CA1 neurons in the stratum pyramidale and preserved synapses intact within the strata moleculare, radiatum, and oriens of the hippocampal CA1 region. In situ detection of DNA fragmentation (TUNEL staining) revealed that ischemic animals infused with EGF contained fewer TUNEL-positive neurons in the hippocampal CA1 field than those infused with vehicle alone at the seventh day after ischemia. In primary hippocampal cultures, EGF (0.048 to 6.0 ng/mL) extended the survival of cultured neurons, facilitated neurite outgrowth, and prevented neuronal damage caused by the hydroxyl radical-producing agent FeSO4 and by the peroxynitrite-producing agent 3-morpholinosydnonimine in a dose-dependent manner. Moreover, EGF significantly attenuated FeSO4-induced lipid peroxidation of cultured neurons. These findings suggest that EGF has a neuroprotective effect on ischemic hippocampal neurons in vivo possibly through inhibition of free radical neurotoxicity and lipid peroxidation.