0306-4522/92 %SOO+O.OO Pergamon Pressplc ii 1991IRRO

Neurasciettcr Vol. 46.No. 3.pp.657-667,1992 Printedin GreatBritain

FAST EXCITATORY POSTSYNAPTIC POTENTIALS AND THE RESPONSES TO EXCITANT AMINO ACIDS OF SYMPATHETIC PREGANGLIONIC NEURONS IN THE SLICE OF THE CAT SPINAL CORD H. INOKUCHI,* M. Y~SHIMURA,* S. YAMADA,? C. POLOSA~§and S. NIsHI*~ *Department

of Physiology IDepartment

and tInstitute of Physiology,

of Brain Diseases, Kurume Kurume 830, Japan McGill University, Montreal,

University Canada

School

of Medicine.

H3G lY6

Abstract-The properties

of the excitatory postsynaptic potential evoked by focal stimulation and of the responses to excitatory amino acids were examined by int~cellular recording from sympathetic preganglionic neurons in upper thoracic spinal cord slices of the adult cat. Single stimuli to the region dorsal to the intermedio-lateral nucleus evoked short-latency. presumably monosynaptic, excitatory postsynaptic potentials. The reversal potential of this response was -2.2 mV and became more negative when external Na+ or K+ concentration was decreased. The excitatory postsynaptic potential was depressed by the non-selective excitatory amino acid receptor antagonist cis-2,3-piperidine dicarboxylic acid and enhanced by a glutamate uptake inhibitor. The non-N-methyl-D-aspartate receptor antagonist 6-cyano-7nitroquinoxaline-2,3-dione abolished the excitatory postsynaptic potential in 72% of neurons. In the remaining neurons, this antagonist only depressed the potential and unmasked a slower component which was abolished by the N-methyl-t)-aspartate receptor antagonist u,r.-2-amino-5phosphonovaleric acid. In the presence of tetrodotoxin all neurons tested were depolarized by glutamate or aspartate, as well as by the selective agonists quisqualate, alpha-amino-3-hydroxy-5-methylisoxazole propionic acid, kainate and N-methyl-D-aspartate. The glutamate-evoked depolarization reversed at a membrane potential of -2.0 mV and at a more negative value when external Na+ or K + concentration was decreased. The response to alpha-amino-3-hydroxy-5methylisoxazole propionic acid was abolished by 6-cyano-7-nitroquinoxalin~-2,3-dione in all neurons tested and that to kainate in only one-third of the cells. In the remainder the response to kainate was only slightly depressed by this antagonist. The responses to glutamate and aspartate were only slightly depressed by the combined action of the various amino acid receptor antagonists used. The responses to N-methyl-D-aspartate were abolished by o,L-2-amino-5-phosphonovaleric acid. The punched-out region of the intermedio-lateral nucleus, maintained in rilro, released glutamate and aspartate in the absence of stimulation. Field stimulation (20Hz) enhanced release by between 40 and 100%. The increase was prevented by superfusion with calcium-free Krebs. It is concluded that excitatory amino acids, acting on both N-methyl-u-aspartate and non-N-methyl-Daspartate receptors, but mainly on the latter, are likely mediators of the monosynaptic excitatory postsynaptic potential evoked in sympathetic preganglionic neurons by the stimulated region. The efflux data suggest that glutamate and aspartate are among the mediators.

Intracellular recording in the sympathetic preganglionic neuron (SPN) of the cat in Z&Wshows that the background and reflex firing of this neuron is due to summating fast excitatory postsynaptic potentials (EPSPs).’ These synaptic responses are produced by the activity of axon systems originating in the spinal cord and supraspinally (for review see Refs 3 and IO). Fast EPSPs (fEPSPs) are also recorded in the spinal cord slice of the cat’9.2’.‘6 and rat.15 A study of the properties of these synaptic responses is of interest .

.- ._. .^._.. _~_~__.

jiTo whom correspondence should be addressed. AbbreCarions: AABH, r_-aspartic acid-beta-hydroxamate; AMPA, alpha-amino-3-hydroxy-S-methylisox~ole propionic acid; APV. D,L-2-amino-5-phosphonovaleric acid; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; EPSP, excitatory postsynaptic potential; fEPSP, fast excitatory postsynpatic potential; IML, intermedio-lateral nucleus; NMDA, N-methyl-o-aspartic acid; PDA, + -cis-2,3piperidine dicarboxylic acid; SPN, sympathetic preganglionic neuron; TTX, tetrodotoxin. 657

as a step towards identifying the chemical mediators and the postsynaptic receptors involved in their generation. Glutamate and aspartate can be considered as leading candidates as mediators of the fEPSP in the SPN because (i) the intermedio-lateral nucleus (IML) contains both in high concentration7.‘7.2’ as well as kainate binding sites;” (ii) glutamate-containing axon terminals make synapses on IML neurons presumed to be SPNs;” (iii) both amino acids excite SPNs when applied by superfusion to the spinal cord slice;21 (iv) spinal application of a non-selective glutamate receptor antagonist blocks the sympatho-excitation produced by stimulation of the rostra1 ventrolateral medulla;t8 and (v) in the rat an fEPSP evoked by dorsal root stimulation is blocked by the N-methyl-Daspartic acid (NMDA) receptor antagonist ILL-~amino-S-phosphonovaleric acid (APV).” The present report describes observations made with intracellular recording from SPNs in the in Gtro

658

H.

INOKIJCHI

slice of the spinal cord of the adult cat. The properties of the fast EPSP obtained by focal stimulation of a region straddling the dorsolateral funiculus and the base of the dorsal horn were studied and compared with the properties of the responses produced by application of excitant amino acids. In addition, release of glutamate and aspartate by the punchedout IML was measured at rest and during field stimulation to determine whether stimulation increased the release of these excitatory amino acids. EXPERIMENTAL

PROCEDURES

The experiments were performed on transverse slices of thoracic spinal cord segments T,-T, using methods described in detail elsewhere.z5,26 Briefly, in cats anesthetized with chloralose and pentobarbital (60 and 10 mg/kg, respectively, i.p.) the second and third thoracic spinal cord segments were excised and cut with a Vibratome into SOO-pm-thick slices. The slices were transferred to a recording chamber and superfused with Krebs solution equilibrated with 95% 0, and 5% CO,, containing (in mM): NaCl 117, KC1 3.6, NaH,PO, 1.2, CaCl, 2.5, MgCl, 1.2, glucose 11.0, NaHCO, 25.0. Temperature was maintained at 37°C. Stimulation and recording Glass micropipettes filled with 3 M KCI or a 1: 1 mixture of 1 M potassium citrate and 1 M potassium acetate or with 1M CsCl were used for intracellular recording. Tip resistances were 40-120 MQ (KCl), 50-160 MQ (potassium citrate, potassium acetate) and 3560 MQ (CsCl), respectively. Synaptic potentials were elicited by stimulation through a monopolar silver electrode (50 pm diameter, insulated except at the tip) or a concentric bipolar tungsten-stainless steel electrode (50 pm in diameter) positioned on a 200 pm wide band of gray and white matter, straddling the lateral edge of the dorsal horn and extending 200-600 pm dorsal to the IML. EPSPs were at times contaminated with inhibitory postsynaptic potentials in normal Krebs solution. To avoid any contribution of glycine or GABA-mediated inhibitory

et a/

postsynaptic potentials to the synaptic potential, bicuculhnc (30 pM) and strychnine (2 fi M) were routinely added to the Krebs solution. Intracellular recordings were obtained with a high input-impedance amplifier with an active bridge circuit, enabling simultaneous measurement of membrane potential and intracellular current injection. Membrane potential and intracellularly injected current were monitored on a VClO or VCll Nihon Kohden memory oscilloscope and recorded on a thermal pen-writer (Nihon Kohden. DC-l kHz). Neurons selected for this study were identified as described in a previous paper.‘” Resting membrane potential was -65.0 k 6.9 mV (mean + S.D., n = 116). Neuron input resistance was measured from the amplitude of the hypcrpolarization (of 0.25) in Ca2+-free medium. Efflux during field stimulation in Ca* + -free medium was 1.61 5 1.17 and 0.97 + 0.52, respectively (n = 8, P > 0.5). DISCUSSION

The observations presented in this study suggest that the presumably monosynaptic, fEPSP evoked in SPNs by focal stimulation of the region of gray and white matter just dorsal to the IML, is mediated by an excitatory amino acid transmitter. This conclusion is based on several findings. First, exogenous glutamate, in the presence of TTX, produced depolarization with increased conductance in SPNs. The fEPSP and the glutamate response reversed at a similar membrane potential, which was similarly modified by changes in extracellular ion composition, suggesting that similar membrane conductances underlie the synaptic and the agonist-evoked response. Second, a non-selective excitatory amino acid receptor antagonist produced a marked, dose-dependent depression of the EPSP, suggesting that the

mediator of the EPSP and the antagonist compete for the same receptor. Moreover, the amplitude of the EPSP was enhanced by a glutamate uptake blocker, suggesting that block of uptake results in increased concentration of the EPSP mediator at the subsynaptic receptor and that the mediator of the fEPSP is subject to uptake, as are excitatory amino acids. Third, stimulus-evoked, Ca2 + -dependent release of glutamate and aspartate was demonstrated from micropunches of the IML. The excised area comprised the IML itself and, possibly, a thin crescent of surrounding white matter. In previous studies in which the same region of the gray matter was explored with ~cr~l~trodes, the ratio of SPNs (identified by the criteria described in Ref. 26) to other neurons was very high, about 90% (Yoshimura and Inokuchi, unpublished observations). Therefore it is likely that the observed release was from axon terminals presynaptic to the SPN. Both NMDA and non-NMDA receptors are present on the SPN membrane. Evidence of the presence of non-NMDA receptors is obtained from the observation that the specific agonists q&qua&e, AMPA and kainate produced, in the presence of TTX, a dose-dependent depolarization of the SPN, with decreased RN, which was antagonized by CNQX but not by APV. Evidence of the presence of NMDA

Excitatory amino acids and synaptic excitation of SPN

665

A2 Aspartate

Glutamate

2.5mM

Ca*’

OmM Ca2+/20mM

loo-

Glu

ASP

Mg2+

J=-j

Gill

ASP

Fig. 8. (A) Typical results of efflux measurement. Content (ordinate) of glutamate (A,) and aspartate (A,) in five consecutive IO-min collection samples. Sample 3 is obtained during the stimulus period (20 Hz, 6 min). Inset shows the approximate location of the cylinder of tissue excised for et&x measurements. (8) Comparison of the effect of stimulation (20 Hz, 6 min) on efllux in 2.5 mM Ca’+, 1.2 mM Mg’+ and 0 mM Ca2+, 20 mM Mg’+ Efflux during stimulation period expressed as per cent of pre-stimulus efflux. One asterisk indicates difference from pre-stimulus (P < 0.005). Two asterisks indicates difference from pre-stimulus (P < 0.001). Number of experiments given in parenthesis.

receptors is found from the observation that the specific agonist NMDA, in the presence of TTX, produced depolarization associated with increased R, in Mg’+ -containing Krebs but with decreased R, in Mg2 +-free Krebs. The NMDA-evoked depolarization was antagonized by APV but not by CNQX. The SPN response to NMDA was depressed at hyperpolarized levels of membrane potential. This depression, which in other neurons has been shown to be a characteristic of NMDA receptor-mediated responses,‘” was abolished in Mg’+-free Krebs. The constant latency of the EPSP, the absence of response failures at relatively high stimulation frequencies, and the continuously graded increase in EPSP amplitude with increasing stimulus intensity, without changes in shape, strongly suggest that the focal stimulus was activating axons with monosynaptic connections to the SPN. These axons may belong to spinal interneurons (segmental or intersegmental) or to fiber tracts of supraspinal origin. Performing the stimuiation in a segment below a chronic spinal transection may provide the data for a choice between these two possibilities. In this connection it is important to mention that glutamate-like immunoreactivity in axon terminals of the rat IML has been

demonstrated with light and electron microscopy. The immunoreactivity of the IML disappears below a chronic spinal transection, suggesting that the glutamate-contajning axon terminals in the IML are of supraspinal originI Since the path connecting the stimulus site to the SPN can be assumed to be monosynaptic. the observed EPSP pharmacology can be interpreted in terms of the excitatory amino acid receptors on the SPN membrane demonstrated by the observations with the various agonists. The fEPSP evoked by focal stimulation seems to result mainly from activation of non-NMDA receptors, with a contribution from NMDA receptor activation in only approximately one-third of the neurons. This conclusion is based on the observation that in all neurons tested the EPSP was depressed or abolished by CNQX. whereas in only one-third of the neurons the EPSP was depressed by APV. Therefore, of the two amino acids released by the IML, aspartate, which is considered to be a rather pure NMDA receptor agonist,” is unlikely to be involved in the generation of the fast, CNQX-sensitive component of the EPSP. Previous observations in the rat, of SPN firing in response to stimulation of the ventrolateral medulla, which was

666

H.

INOKUCHI

blocked by kynurenate but was insensitive to APV and ketamine,18 also suggests a dominant role of nonNMDA receptors in SPN excitation by supraspinal input. The discrepancy between the number of neurons showing APV-sensitive depolarizations in response to exogenous NMDA (100%) and the number of neurons showing APV-sensitive EPSPs (approximately 30%) may be due to the fact that in this neuron NMDA receptors are mainly at synapses other than those activated by focal stimulation and/or situated extrasynaptically. Still another possible explanation for the discrepancy is that while NMDA is not subject to cellular uptake6 the endogenous NMDA receptor ligand is, resulting in higher concentration at the subsynaptic receptor of NMDA compared with the endogenous ligand. At any rate, the observation, reported under Results, that field stimulation releases aspartate, a rather pure NMDA receptor agonist,” from IML micropunches suggests that NMDA receptors are likely to have a role in synaptic transmission at this neuron. From the observation that APV blocks the EPSP evoked in SPNs of neonatal rats by dorsal root afferent stimulation, the conclusion was reached that the NMDA receptor of the SPN has a dominant role in the synaptic excitation of this neuron.lS If the NMDA receptors responsible for this finding were on the SPN, the contrast between this conclusion and that reached in the present as well as in another study in the rat,r8 suggests that the NMDA receptor may only be involved in transmission of spinal inputs and/or that this receptor may have a role at some stage of development but not at maturity. An additional possibility, since the resting potential of SPNs was more hyperpolarized in the present study than in the study on neonatal ratsI and since the Mg block of NMDA receptors is voltage dependent,” is that the different results of the two studies may be due to a difference in the extent of Mg block between the two preparations. Alternatively, since the EPSPs studied by MO and Dun were polysynaptic, as suggested by their long latency as well as by previous data on dorsal root afferents-SPN connections,‘,5 the NMDA receptors responsible for their finding may have been on interneurons rather than on the SPN.

PI al

CNQX, at concentrations which blocked the EPSP, antagonized the AMPA-evoked depolartration in every cell tested and the kainatc-evoked depolarization in only one-third of the cells. This suggests that in at least two-thirds of cells the fEPSP is mainly mediated by quisqualate receptors. The apparent low sensitivity of the glutamate and aspartate responses to CNQX and APV, previously observed in other neurons,*,” may be tentatively explained in several ways, none of which, however, is entirely satisfactory. The use of high concentrations of these uptake-prone agonists may result in a concentration at the receptor higher than for the endogenous ligand and hence in responses less sensitive to a given antagonist concentration than the synaptic responses. In relation to this explanation. it is of interest that responses to AMPA, an agonist not subject to uptake,” of magnitude comparable to those to glutamate and aspartate, were blocked by CNQX. The antagonist potency may be greater in the presumably non-equilibrium conditions associated with the short-lasting life span of the endogenous ligand than in the presumably equilibrium conditions associated with the long-lasting presence of the exogenous agonists. However, the AMPA responses were blocked by CNQX at concentrations no higher than required for block of the EPSP, making this explanation unlikely. Finally, on the assumption that nerve-released and exogenous ligands interact with different excitatory amino acid receptor subsets. the endogenous mainly with subsynaptic receptors, the exogenous mainly with extrasynaptic receptors, and that the extrasynaptic subset is heterogeneous as far as sensitivity to antagonists is concerned (see, for example, the large variation in CNQX sensitivity of kainate responses), this heterogeneity may be responsible for the apparent insensitivity of glutamate and aspartate responses to CNQX and APV.

Acknowledgements-This work was supported by a Grantin-Aid for Scientific Research to S. Nishi by the Ministry of Education, Science and Culture of Japan, and by a grant to C. Polosa by the Medical Research Council of Canada. C. Polosa was Visiting Professor in the Department of Physiology, Kurume University School of Medicine.

REFERENCES 1. Beacham W. S. and Per1 E. R. (1964) Background and reflex discharge of sympathetic preganglionic neurons in the spinal cat. .I. Physiol., Land. 172, 400-416. 2. Berry M. S. and Pentreath V. N. (1976) Criteria for distinguishing between monosynaptic and polysynaptic transmission. Bruin Rex 105, l-20. 3. Coote J. H. (1988) The organization of cardiovascular neurons in the spinal cord. Reu. Physiol. Biochem. Pharmac. 110, 148-285. 4. Davies J., Evans R. H., Francis A. A., Jones A. W. and Watkins J. C. (1981) Antagonism of excitatory amino acid-induced and synaptic excitation of spinal neurons by cis-2,3-pip&dine dicarboxylate. J. Neurochem. 36, 1305-1307. 5. Dembowsky

K., Czachurski J. and Seller H. (1985) An intracellular study of the synaptic input to sympathetic preganglionic neurons of the third thoracic segment of the cat. J. auron. neru. Sysr. 13, 201-244. 6. Garthwaite J. (1985) Cellular uptake disguises action of L-glutamate on N-methyl-D-aspartate receptors. Br. J. Pharmac. 85. 297-307.

Excitatory

amino

acids and synaptic

excitation

of SPN

667

7. Graham L. T. Jr, Shank R. P., Werman R. and Aprison M. H. (1967) Distribution of some synaptic transmitter suspects ,n cat spinal cord: glutamic acid, aspartic acid, gamma-aminobutirric acid, glycine and glutamine. J. Neurochem. 14, 465412. 8. Honor6 T., Davies S. N., Drejer J., Fletcher E. J., Jacobsen P., Lodge D. and Nielsen F. E. (1988) Quino~linediones: potent competitive non-NMDA glutamate receptor antagonists. Science 241, 701-703. 9. Johnston D. and Hablitz J. J. (1980) Voltage clamp discloses slow inward current in hippocampal burst-firing neurones. Nature 286, 391-393. 10. Laskey W. and Polosa C. (1988) Characteristics of the sympathetic preganglionic neuron and its synaptic input. Prog. Neurobiol. 31, 47-84. Il. Lodge D. (1981) Uptake inhibitors, amino acids. and spinal neurons. In Amino Acid Neurotrunsmitters (eds DeFeudis F. V. and Mandel P.). pp. 327-332. Raven Press, New York. 12. Mayer M. L.. Westbrook G. L. and Guthrie P. B. (1984) Voltage-dependent block by My’+ of NMDA responses in spinal cord neurones. Nuture 309, X1-263. 13. Mayer M. M. and Westbrook G. L. (1985) The action of N-methyl aspartic acid on mouse spinal neurons in culture. J. Physiol., Lond. 361, 65-90. 14. Mayer M. L. and Westbrook G. L. (1987) The physiology of excitatory amino acids in the vertebrate central nervous system. Prog. Neurobiol. 28, 197-276. 15. MO N. and Dun N. J. (1987) Excitatory postsynaptic potentials in neonatal rat sympathetic preganglionic neurons: possible mediation by NMDA receptors. Neurosci. Lett. 77, 327-332. lb. Moroni F. and Pepeu G. (1984) The cortical cup technique. In ~eu~ur~~~ent nf ~eurotran.~mitter Release In Vivo (ed. Marsden C. A.), pp. 63-76. Wiley, New York. 17. Morrison S. F.. Callaway J.. Milner T. A. and Reis D. J. (1989) Glutamate in the spinal sympathetic intermediolateral nucleus: localization by light and electron microscopy. Brain Res. 503, 5-15. IX. Morrison S. F.. Ernsberger P., Milner T. A., Callaway J., Gong A. and Reis D. J. (1989) A glutamate mechanism in the intermediolateral nucleus mediates sympathoexcitatory responses to stimulation of the rostra1 ventrolateral medulla. In The Central Neural Organization of Cardiovascular Control (eds Ciriello J., Caverson M. M. and Polosa c’.), pp. 1599169. Elsevier, Amsterdam. 19. Nishi S., Yoshimura M. and Polosa C. (1987) Synaptic potentiafs and putative transmitter actions in sympathetic preganglionic neurons. In Organizaiion of the Autonomic Nemous System: Central and Peripheral ~e~honi~~rn.~ (eds Ciriello J., Calaresu F. R., Renaud L. P. and Polosa C.), pp. 15-26. Alan R. Liss, New York. 20. Patneau D. K. and Mayer M. L. (1990) Structure-activity relationships for amino acid transmitter candidates acting at N-methyl-D-aspartate and quisqualate receptors. J. Neurosci. IO, 23852399. 21. Patrick J. T., McBride W. J. and Felten D. L. (1983) Distribution of glycine, GABA, aspartate and glutamate in the rat spinal cord. Brain Res. Bull. 10, 415418. 22. Rethelyi M. (1972) Cell and neuropil architecture of the intermediolateral (sympathetjc) nucleus of cat spinal cord. Brriitr Rex 46. 2032 13. 23. Srinivasan V.,‘Neal M. J. and Mitchell J. F. (1969) The effect of electrical stimulation and high potassium concentrations on the efflux of [‘HP;-aminobutyric acid from brain slices. J. Neurochem. 16, 1235-1244. 24. Watkins J. C. and Evans R. H. (1981) Excitatory amino acid transmitters’rl. Rev. Pharmuc. Toxic. 21, IbS--204. 25. Yoshimura M. and Nishi S. (1982) Intracellular recordings from lateral horn cells of the spinal cord in vitro. J. auton. nerr. Spt. 6, S-l I. 26. Yoshimura M., Polosa C. and Nishi S. (1986) ElectrophysioIogical properties of sympathetic pregang~ioni~ neurons in the cat spinal cord in vitro. &hTgers Arch. ges. Physiol. 406, 91-98. 27. Yoshimura M. and Jesse11 T. (1990) Amino acid-mediated EPSPs at primary afferent synapses with substantia gelatinosa neurones in the rat spinal cord. J. Ph~y.yio(., Lond. 430, 315-335. (Accepted

1 July 1991)