N-methyl-d-aspartate (NMDA) stimulates release of α-MSH from the rat hypothalamus through release of nitric oxide

Superfusion of rat hypothalamic slices with 10⁻⁴ MN-methyl-d-aspartic acid (NMDA) resulted in increased release of α-melanocyte-stimulating hormone (α-MSH). Peptide release was blocked by 10⁻⁶ MN^G-nitro-l-arginine methyl ester (l-NAME) a specific competitive inhibitor of nitric oxide synthase but not by the inactive enantiomerd-NAME at 10⁻⁶ M. The inhibition byl-NAME was reversed by the addition of 10⁻⁵ Ml-arginine, an excess of enzyme substrate. Release of nitric oxide products into tissue superfusates was stimulated by a 50 mM concentration of potassium ions and by 10⁻⁴ M NMDA. Potassium-stimulated release was blocked byl-NAME. Basal, potassium-stimulated and NMDA-stimulated release of nitric oxide products were significantly inhibited by the NMDA-receptor antagonistd-(−)-2-amino-5-phosphopentanoic acid (AP5) at 10⁻⁴ M and by the NMDA-channel blocker ketamine at 10⁻⁴ M. We conclude that nitric oxide mediates the stimulatory action of glutamic acid ont he release of α-MSH from the rat hypothalamus.     

Voltage clamp analysis of lamprey neurons — role of N-methyl-d-aspartate receptors in fictive locomotion

Spinal neurons in the lamprey have been subjected to a voltage clamp analysis of the excitatory currents generated during fictive locomotion with particular reference to the phasic activation of voltage dependent N-methyl-D-aspartate (NMDA) receptors. Voltage-clamped neurons observed during NMDA-induced fictive swimming show excitatory and inhibitory synaptic currents in phase with the ipsilateral and contralateral ventral root discharges, respectively. The excitatory synaptic currents showed a marked voltage dependence suggesting that potential sensitive conductances such as the NMDA ionophore are involved in the synaptic events underlying rhythmic locomotor activity. The effect of NMDA receptor activation during application of tetrodotoxin has also been analyzed during NMDA-induced pacemaker-like oscillations. Such NMDA-induced oscillations are essentially abolished during the voltage clamp. In the presence of NMDA current voltage plots reveal a negative slope conductance in the potential range of the inherent oscillations. The addition of tetraethyl ammonium (TEA) to NMDA solution enhanced a net steady state inward current by more than 10-fold due to a partial block of the outward currents. A kinetic analysis was done with a frequency domain technique using a white noise stimulus to linearly perturb the membrane potential over a wide range of frequencies. The analysis revealed that the induced negative conductance leads to a response which is nearly 180 degrees out of phase with the stimulus at low frequencies. This is an unstable condition which leads to the depolarizing phase of the induced oscillations.     

Role of nitric oxide, adenosine,N-methyl-d-aspartate receptors, and neuronal activation in hypoxia-induced pial arteriolar dilation in rats

In this study, we tested the hypothesis that nitric oxide (NO) and adenosine (ADO) are the principal mediators of severe hypoxia-induced vasodilation. In addition, we examined whether activation ofN-methyl-d-aspartate (NMDA) receptors and/or perivascular nerves plays a role. A closed cranial window and intravital microscopy system was used to monitor diameter changes in pial arterioles ( 40 μm) in anesthetized rats. The relative contributions of ADO, NMDA, NO, and neuronal activation to hypoxic cerebrovasodilation were assessed using the blockers 8-sulfophenyltheophylline (8-SPT), MK-801, nitro-l-arginine methylester (LNAME), and tetrodotoxin (TTX). Two experimental series were studied. In the first, we tested the effects of NOS inhibition, via topical L-NAME (1 mM), on moderate (PaO₂ ≈ 46 mmHg)then severe (PaO₂ ≈ 34 mmHg) hypoxia-induced dilation. To confirm that L-NAME was affecting specifically NO-dependent responses, we also examined, in each experiment, the vasodilatory responses to topical applications of NOS-dependent (adenosine diphosphate (ADP); acetylcholine (ACh)(and -independent (sodium nitroprusside (SNP)) agents, in the presence of L-NAME or, in controls, the presence of D-NAME or no added analogue. In the second series, topical suffusions of ADP, ADO, and NMDA were sequentially applied, followed by 5 min exposure to severe hypoxia (PaO, ≈ 32 mmHg). Following return to normoxia, a suffusion of either 8-SPT (10 μM), MK-801 (10 μM), TTX (1 μM), or 8-SPT + MK-801 was initiated (or, in controls, application of a drug-free suffusate was maintained), and the above sequence repeated. In control, TTX, and 8-SPT + MK-801 experiments, baseline conditions were then restored and hypercapnia (PaCO₂ = 70–85 mmHg) was imposed. In the series 1 control groups, moderate and severe hypoxia elicited ≈ 20% and 35–40% increases in diameter, respectively. L-NAME attenuated ADP- and ACh-induced dilations, did not alter the arteriolar responses to SNP or moderate hypoxia, but prevented further dilation upon imposition of severe hypoxia. This suggested that 45–50% of the severe hypoxia response was NO-dependent. In series 2, 8-SPT blocked the adenosine response and reduced severe hypoxia-induced dilation by 46%. MK-801 predictably blocked NMDA-induced relaxation and reduced the hypoxic response by 42%. When combined, 8-SPT and MK-801 affected hypoxic vasodilation additively. After TTX, the ADP and ADO responses were normal, but NMDA and hypoxia responses were completely blocked. Hypercapnia-induced dilation was unaffected by TTX or 8-SPT + MK-801. The results imply that severe hypoxia-induced release of NO and ADO, and the accompanying pial arteriolar dilation, are wholly dependent on the capacity to generate action potentials in perivascular nerves. The similarity of the L-NAME and MK-801 effects on hypoxic cerebrovasodilation suggests that the NO-dependency, to a large degree, derives from NMDA receptor activation.     

Interactions of dextromethorphan with theN-methyl-d-aspartate receptor-channel complex: single channel recordings

The actions of dextromethorphan (DXM) on the 50 pS conductance state of theN-methyl-d-aspartate (NMDA) receptor-operated channel were studied using outside-out patches obtained from cultured rat hippocampal pyramidal neurons. DXM (5–50 μM) had no effect on the amplitudes of unitary currents but caused concentration-dependent reductions in channel mean open times and the frequency of channel openings. Channel open probability was reduced in a concentration-dependent manner by DXM and was one-half of the control value at a DXM concentration of 6 μM, with the patch potential held at −60 mV. An IC₅₀ value of 4 μM was obtained for the reduction by DXM of NMDA-evoked rises in [Ca²⁺]_i in cultured rat hippocampal pyramidal neurons loaded with Fura-2. The results were consistent with drug block of the open NMDA channel with an onward (blocking) rate constant of 7.7 × 10⁶ M⁻¹ · s⁻¹ (at −60 mV). The estimated unblocking rate constant was about 10 s⁻¹, a value considerably higher compared to the off-rate constant found for dizocilpine block of the NMDA channel.     

Contraversive circling induced by ventral tegmental microinjections of moderate doses of morphine and [d-Pen,d-Pen]enkephalin

Unilateral ventral tegmental area (VTA) injections of morphine and [d-Pen²,d-Pen⁵]enkephalin (DPDPE), caused contraversive circling at doses of 1.2, 12, and 24 nmol. Similar doses of the selective κ-agonist U-50,488H were ineffective. These data suggest a common mechanism for the circling, locomotion and facilitation of brain stimulation reward caused by VTA morphine, and distinguish this mechanism for that of feeding which is caused by both morphine and κ-actions in this region.     

A novel interaction between dynorphin(1–13) and an N-methyl-d-aspartate site

Dynorphin injected intrathecally in the rat results in a neurotoxicity behaviorally expressed as an irreversible loss of the thermally evoked tail-flick reflex. The excitatory amino acid antagonists DL-2-amino-5-phosphonovalerate (APV) and gamma-D-glutamylglycine (DGG) blocked the loss of the tail-flick reflex. The order of potency (APV greater than DGG) suggests that the N-methyl-D-aspartate (NMDA) subclass of excitatory amino acid receptors participate in the neurotoxicity. Additionally, intrathecal injection of APV results in a reversible loss of the tail-flick reflex, whereas with DGG doses which block the tail-flick reflex also result in hindlimb paralysis. These data suggest that neurotransmission in the tail-flick reflex pathway is, in part, mediated by NMDA receptors. From these and previous findings it was concluded that dynorphin neurotoxicity results from enhanced, excitotoxic, transmission across these synapses utilizing NMDA receptors.     

Effects of nitric oxide synthase inhibitors onN-methyl-d-aspartate-induced phase delay of circadian rhythm of neuronal activity in the rat suprachiasmatic nucleus in vitro

Excitatory amino acid (EAA) receptors such asN-methyl-d-aspartate (NMDA) and non-NMDA receptors have been suggested to play an important role in the regulation of photic information from the retina to the suprachiasmatic nucleus (SCN). Therefore, we investigated the role of glutamate as a retinohypothalamic transmitter by analyzing the phase-resetting effects of NMDA and a non-NMDA agonist, (R, S)-α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA), on the circadian rhythm of SCN firing activity. Nitric oxide (NO) production is believed to be an essential intermediate in NMDA-induced cGMP production in the CNS. Thus, we examined the effects of blockers of NO production on NMDA- or AMPA-induced phase delay of SCN activity rhythm.N-nitro-l-arginine methylester (l-NAME) blocked NMDA- but not AMPA-induced phase shift, indicating the involvement of NO synthesis in NMDA-induced phase changes.l-arginine but notd-arginine caused a phase delay, andl-NAME blockedl-arginine-induced phase delay. In addition, cotreatment with NMDA andl-arginine did not have an additive effect. These results suggest that NO production itself is involved in the phase change of SCN neuron activity, and NMDA-induced phase changes are also mediated via activation of NO synthesis in this nucleus.     

Polyamine effects uponN-methyl-D-aspartate receptor functioning: differential alteration by glutamate and glycine site antagonists

Polyamines such as spermidine potentiate activation of theN-methyl-D-aspartate (NMDA)-type excitatory amino acid receptor. The goal of the present study was to investigate interactions between the putative polyamine binding site and previously described sites for glutamate and glycine. Binding of the high-potency PCP receptor ligand [³H]MK-801 to well-washed rat brain membranes was used as an in vitro probe of NMDA receptor activation. Spermidine concentration-response studies were performed in the absence and presence of both glutamate and glycine, with and withoutD-(−)-2-amino-5-phosphonovaleric acid (D(−)AP-5) or 7-chlorokynurenic acid (7Cl-KYN). Incubation in the presence of spermidine alone induced a 20.4-fold increase in [³H]MK-801 binding with an EC₅₀ value of 13.3 μM. The mean concentration of spermidine which induced maximal stimulation of binding was 130 μM (n = 10,S.E.M.= 24.66,range= 25–250 μM). Glutamate (10 μM) decreased the EC₅₀ value for spermidine-induced stimulation of [³H]MK-801 binding to 3.4 μM. Glycine (10 μM) did not significantly alter either maximum spermidine-induced [³H]MK-801 binding or the EC₅₀ value for spermidine-induced stimulation of [³H]MK-801 binding. Incubation in the presence of the specific glutamate antagonistD(−)AP-5 attenuated [³H]MK-801 binding in a glutamate-reversible fashion. The competitive glycine antagonist 7Cl-KYN decreased maximum spermidine-induced [³H]MK-801 binding in a glycine-reversible fashion. In addition, 7Cl-KYN increased the EC₅₀ value for spermidine-induced stimulation of [³H]MK-801 binding whileD(−)AP-5 was without effect. These findings suggest that glutamate and glycine regulate the polyamine binding site differentially. PCP-like agents induce a psychotomimetic state closely resembling schizophrenia by inhibiting NMDA receptor-mediated neurotransmission. The ability of polyamines to modulate NMDA receptor functioning suggests a potential site for pharmacological intervention.     

Molecular and functional characterization of Xenopus laevis N-methyl-d-aspartate receptors

N-methyl-d-aspartate (NMDA) receptors of many different vertebrates have been characterized in the past. However, little information is available about amphibian NMDA receptors. Here, we investigated the South African clawed frog Xenopus laevis NR1 subunit at the molecular and functional level. In this subunit, which is obligatory for functional NMDA receptor complexes, we found three exons, the N1, C1, and C3 cassettes, being alternatively spliced. Combinations of these cassettes generated six different splice variants, which were functionally characterized in oocytes. The Xenopus NR1 isoforms generally showed the same functional properties as their mammalian homologs when coexpressed with rat NR2B. In coexpression with Xenopus NR2B, however, some properties changed significantly. This included a Zn²⁺-mediated potentiation of current amplitudes for some subunit combinations which lasted for several minutes. This mechanism presents a novel form of Xenopus NMDA receptor modulation, possibly mediating a form of short-term potentiation in the Xenopus central nervous system.     

Nitric oxide synthase inhibitorN-nitro-l-arginine methyl ester decreases ischemic damage in reversible focal cerebral ischemia in hyperglycemic rats

We tested the hypothesis that the exacerbation of post-ischemic brain tissue injury associated with hyperglycemia in rats is due to toxic metabolism of nitric oxide. We used magnetic resonance imaging (MRI) techniques to measure neuronal and cerebrovascular injury in a 2-h transient focal cerebral ischemia model in normoglycemic and hyperglycemic rats at 3 and 24 h post-ischemia onset. We determined the effect of low dose (3 mg/kg i.p.) treatment with the nitric oxide synthase inhibitorN^G-nitro-l-arginine methyl ester (l-NAME). Compared to normoglycemia, preexisting hyperglycemia increased the volume of brain tissue exhibiting hyperintensity in diffusion weighted MRI (DWI) by factors of 5.6 and 6.2 at 3 h and 24 h post-ischemia, respectively. A similar increase in tissue volumes exhibiting hyperintense signal in T2-weighted MRI (T2WI) (3.3-fold and 5.6-fold) was observed. Cerebral blood volume MRI indicated a large focal no-reflow zone in hyperglycemic rats. Treatment withl-NAME eliminated the no-reflow zone in the hyperglycemic rats, and reduced tissue volumes of DWI hyperintensity by 86% and 93% at 3 h and 24 h, respectively. Similarly, tissue volumes of T2WI hyperintensity were reduced by 80% and 94% at 3 h and 24 h, respectively. Thus, nitric oxide is an important mediator in the exacerbation of post-ischemic brain injury in hyperglycemic rats. Inhibition of nitric oxide synthase limits edema formation, improves perfusion and reduces infarct volume.     

Results of N-methyl-d-aspartate antagonists in perinatal cerebral asphyxia therapy

Perinatal cerebral asphyxia, which results in significant neurologic and cognitive disabilities in infants and children, remains a major health problem. Potential neurologic sequelae include cerebral palsy, mental retardation, and epilepsy. Over the next few years, neuroprotective agents that prevent asphyxial neuronal injury and death are likely to be developed. These agents may also be effective in prophylaxis and treatment of chronic neurologic disorders, including epilepsy and neurodegenerative disorders, such as Huntington disease.     

N-methyl-D-aspartate receptor-mediated signaling in the supraoptic nucleus involves activation of a nitric oxide-dependent pathway

N-Methyl-D-aspartate (NMDA) microinjection (1 mM, 0.2 μl) into the hypothalamic supraoptic nucleus (SON) stimulated heart rate in urethane-anaesthetized rats. This effect was inhibited by coinjection of a competitive blocker of NMDA receptors, CPP (20 nmol) or by pretreatment with a sympathetic ganglionic blocker, chlorisondamine chloride (5 mg/kg i.p.), but not by prior hypophysectomy. Furthermore, the cardioexcitatory effect of intra-SON NMDA was inhibited by prior intra-SON injection of a competitive blocker of nitric oxide (NO) synthesis, N^G-nitro-L-arginine methyl ester (40 nmol) or a blocker of the soluble guanylate cyclase, Methylene blue (20 nmol), and was mimicked by intra-SON injection of a calcium ionophore, A23187 (10 nmol), which stimulates NO production by raising intracellular free calcium levels. Finally, intra-SON microinjection of a membrane-permeating cGMP analog, 8-bromo-cGMP (20 nmol) stimulated heart rate in urethane-anaesthetized rats. The results point to a functional link between a sympathetically mediated cardiophysiological effect of NMDA receptor stimulation in the SON and activation of the NO/cGMP signal transduction pathway.     

Contribution of spinal N-Methyl-d-aspartic acid receptors to control of sympathetic outflow by the paraventricular nucleus

Spinal NMDA receptors contribute to control of the cardiovascular system by the ventrolateral medulla. However, little is known about the contribution of these receptors to suprabulbar regulation of hemodynamics and sympathetic outflow. Hence, the involvement of spinal NMDA receptors in regulation of the cardiovascular system by the paraventricular nucleus (PVN) of the hypothalamus was investigated. In urethane-anesthetized rats, the change in mean arterial pressure (MAP), heart rate (HR), and renal nerve activity (RNA) produced by electrical or chemical activation of the PVN was determined before and after intrathecal administration of the NMDA receptor antagonist, 2-amino-5-phosphonovaleric acid (AP5). Intrathecal AP5 decreased resting MAP, HR, and RNA, but had no effect on the increase in RNA produced by electrical or chemical stimulation of the PVN. The presser and renal vasoconstrictor effects resulting from electrical, but not chemical, stimulation of the PVN were significantly reduced by intrathecal AP5. These data show that much of the cardiovascular control exerted by the PVN does not depend on a spinal NMDA receptor mechanism.     

Inhibition of nitric oxide formation does not protect murine cortical cell cultures from N-methyl-d-aspartate neurotoxicity

We examined the role of nitric oxide in N-methyl-d-aspartate (NMDA) receptor-mediated neurotoxicity in rat and mouse primary cortical cell cultures. In rat and mouse cultures, the NO synthase inhibitor, N^G-Nitro-l-arginine, blocked cGMP formation but not neuronal cell death following a 5–10 min exposure to 300–500 μM NMDA. N^G-Monomethyl-l-arginine was also unable to prevent neuronal death. In contrast, the non-competitive NMDA receptor antagonist, dextrophan, prevented both cGMP formation and cell death. While other data suggest that the synthesis of nitric oxide can mediate NMDA receptor-mediated neurotoxicity, present results suggest that such synthesis is not necessarily required.     

l-[H]Glutamate binds to kainate-, NMDA- and AMPA-sensitive binding sites: an autoradiographic analysis

The anatomical distribution ofl-[³H]glutamate binding sites was determined in the presence of various glutamate analogues using quantitative autoradiography. The binding ofl-[³H]glutamate is accounted for the presence of 3 distinct binding sites when measured in the absence of Ca²⁺, Cl⁻ and Na⁺ ions. The anatomical distribution and pharmacological specificity of these binding sites correspond to that reported for the 3 excitatory amino acid binding sites selectively labeled byd-[³H]2-amino-5-phosphonopentanoate (d-[³H]AP5), [³H]kainate ([³H]KA) and [³H]α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ([³H]AMPA) which are thought to be selective ligands for the N-methyl-d-aspartate (NMDA), KA and quisqualate (QA) receptors, respectively.     

Decrease in δ-opioid receptor density in rat brain after chronic [d-Ala,d-Leu]enkephalin treatment

Chronic treatment of Sprague-Dawley rats with [d-Ala²,d-Leu⁵]enkephalin (DADLE) resulted in the development of tolerance to the antinociceptive effect of this opioid peptide. When opioid receptor binding was measured, time-dependent decreases in [³H]diprenorphine binding to the P₂ membranes prepared from the cortex, midbrain and striatum were observed. Scatchard analysis of the saturation binding data revealed a decrease in B_max values and no change in the K_d values of [³H]diprenorphine binding to these brain regions, indicative of down-regulation of the receptor. This reduction in the opioid receptor binding activities could be demonstrated to be due to the DADLE effect on the δ-opioid receptors in these brain regions. When [³H]DADLE binding was carried out in the presence of morphiceptin, a significant reduction in the δ-opioid receptor binding was observed in all brain areas tested. μ-Opioid receptor binding decrease was observed only in the striatum after 5 days of DADLE treatment. Additionally, the onset of δ-opioid receptor decrease in the midbrain area was rapid, within 6 h of the initiation of the chronic DADLE treatment. Thus, analogous to previous observations in which chronic etorphine treatment preferentially reduced μ-opioid receptor binding, chronic DADLE treatment preferentially reduced δ-opioid receptor binding activity.     

Relationship between resting cytosolic Ca and responses induced byN-methyl-d-aspartate in hippocampal neurons

Cytosolic calcium concentrations ([Ca²⁺]_i) in cultured hippocampal neurons from rat embryos were measured using fura-2. Neurons with higher resting [Ca²⁺]_i showed greater [Ca²⁺]_i responses toN-methyl-d-aspartate (NMDA) and K⁺ depolarization. There was a strong relationship between resting [Ca²⁺]_i and the maximal changes in [Ca²⁺]_i (Δ[Ca²⁺]_i), which fit the our proposed equation to describe this relationship.     

Absence of side-effects in the anticonvulsant action of cortically applied antagonists of N-methyl-d-aspartate

Three compounds reportedly blocking the N-methyl-d-aspartate (NMDA) receptor, namely 2-amino-5-phosphonovalerate, γ-d-glutamylglycine and 3-hydroxy-2-quinoxalinecar☐ylic acid, were injected subdurally onto the cortex of freely moving rats. All 3 compounds effectively suppressed behavioral and electrographic seizure activity induced by strychnine, morphine and picrotoxin that were administered via the same route. The cortical application of the NMDA-receptor antagonists did not induce behavioral or electrographic changes, and behavioral side-effects commonly observed following intracerebroventricular administration of these compounds were absent. The anatomical separation of anticonvulsant action and side-effects induced by these compounds suggests that this class of compounds may eventually be useful as antiepileptic drugs.     

Immunohistochemical evidence for flupirtine acting as an antagonist on theN-methyl-d-aspartate and homocysteic acid-induced release of GABA in the rabbit retina

When rabbit retinas are exposed in vitro to specific excitatory amino acid receptor agonists certain GABAergic amacrine cells are activated to cause a release of GABA. The GABA that is not released can be detected by immunohistochemistry. Exposure of tissues to kainate or NMDA each caused a characteristic change in the GABA immunoreactivity. CNQX antagonised the kainate effect specifically while MK-801 counteracted the influence of NMDA The effect produced by kainate was mimicked by domoic acid while the influence of homocysteic acid was identical with NMDA. Flupirtine alone did not influence the nature of the GABA immunoreactivity and so did not act as a kainate or NMDA agonist. However, flupirtine counteracted the influence produced by NMDA and homocysteic acid but had no effect on the kainate and domoic acid responses. Thus in this system flupirtine acts as an NMDA antagonist.     

Role of nitric oxide, adenosine,N-methyl-d-aspartate receptors, and neuronal activation in hypoxia-induced pial arteriolar dilation in rats

In this study, we tested the hypothesis that nitric oxide (NO) and adenosine (ADO) are the principal mediators of severe hypoxia-induced vasodilation. In addition, we examined whether activation ofN-methyl-d-aspartate (NMDA) receptors and/or perivascular nerves plays a role. A closed cranial window and intravital microscopy system was used to monitor diameter changes in pial arterioles (∼ 40 μm) in anesthetized rats. The relative contributions of ADO, NMDA, NO, and neuronal activation to hypoxic cerebrovasodilation were assessed using the blockers 8-sulfophenyltheophylline (8-SPT), MK-801, nitro-l-arginine methylester (LNAME), and tetrodotoxin (TTX). Two experimental series were studied. In the first, we tested the effects of NOS inhibition, via topical L-NAME (1 mM), on moderate (PaO₂ ≈ 46 mmHg)then severe (PaO₂ ≈ 34 mmHg) hypoxia-induced dilation. To confirm that L-NAME was affecting specifically NO-dependent responses, we also examined, in each experiment, the vasodilatory responses to topical applications of NOS-dependent (adenosine diphosphate (ADP); acetylcholine (ACh)(and -independent (sodium nitroprusside (SNP)) agents, in the presence of L-NAME or, in controls, the presence of D-NAME or no added analogue. In the second series, topical suffusions of ADP, ADO, and NMDA were sequentially applied, followed by 5 min exposure to severe hypoxia (PaO, ≈ 32 mmHg). Following return to normoxia, a suffusion of either 8-SPT (10 μM), MK-801 (10 μM), TTX (1 μM), or 8-SPT + MK-801 was initiated (or, in controls, application of a drug-free suffusate was maintained), and the above sequence repeated. In control, TTX, and 8-SPT + MK-801 experiments, baseline conditions were then restored and hypercapnia (PaCO₂ = 70–85 mmHg) was imposed. In the series 1 control groups, moderate and severe hypoxia elicited ≈ 20% and 35–40% increases in diameter, respectively. L-NAME attenuated ADP- and ACh-induced dilations, did not alter the arteriolar responses to SNP or moderate hypoxia, but prevented further dilation upon imposition of severe hypoxia. This suggested that 45–50% of the severe hypoxia response was NO-dependent. In series 2, 8-SPT blocked the adenosine response and reduced severe hypoxia-induced dilation by 46%. MK-801 predictably blocked NMDA-induced relaxation and reduced the hypoxic response by 42%. When combined, 8-SPT and MK-801 affected hypoxic vasodilation additively. After TTX, the ADP and ADO responses were normal, but NMDA and hypoxia responses were completely blocked. Hypercapnia-induced dilation was unaffected by TTX or 8-SPT + MK-801. The results imply that severe hypoxia-induced release of NO and ADO, and the accompanying pial arteriolar dilation, are wholly dependent on the capacity to generate action potentials in perivascular nerves. The similarity of the L-NAME and MK-801 effects on hypoxic cerebrovasodilation suggests that the NO-dependency, to a large degree, derives from NMDA receptor activation.