Synaptic mechanisms in sympathetic preganglionic neurons' S. 'k: WU, E. SHEN, T. MIYAZAKI,~ S. &. BUN, AND C. RBN Dep~rt~nent of Anatomy, Medical College of Ohio, Toledo, OH 43699, U.S. A.
N. J.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
Received September 23, 1991 BUN, N. J., Wu, S. Y., SHEN,E., MIYAZAKI, T., DUN, S. L., and REN, C. 1992. Synaptic mechanisms in sympathetic preganglionic neurons. Can. J. Physiol. Pharmacol. 70: S86 - S9 1. Intracellular recordings from sympathetic preganglionic neurons (SPNs) in adult cat and neonatal rat spinal cord slices reveal four types of synaptic potentials, namely, excitatory postsynaptic potentials (EPSPs), inhibitory postsynaptic potentials (IPSPs), and slow EPSPs in both preparations, and a slow IPSP in cat SPNs. Pharmacological studies show that glutamate or a related excitatory amino acid and glycine are the probable mediators of EPSPs and IPSPs. There may be heterogenous mediators of slow EPSPs; substance P, serotonin, norepinephrine, and epinephrine are all probable mediators of slow EPSPs in subpopulations of SPNs. In the case of slow IPSPs, norepinephrine appears to be the likely transmitter. Finally, stimulation of ventral roots elicits a synaptic potential that appears to be caused by glutamate released from afferent fibers in the ventral roots. Our results indicate that a multitude of synaptic mechanisms exist in the rat SPNs by means of which inputs arising from sensory and supraspinal neurons are processed in a timely and orderly manner, thus ensuring highly organized but differentiated outputs to multiple peripheral target cells. Key words: sympathetic preganglionic neurons, excitatory postsynaptic potentials, inhibitory postsynaptic potentials, slow excitatory postsynaptic potentials, glutamate, glycine. DUN, N. J., WU, S. Y., SHEN,E., MIYAZAKI, T., DUN, S. L., et REN, C. 1992. Synaptic mechanisms in sympathetic preganglionic neurons. Can. J. Physiol. Pharmacol. 190 : S86 - S9 1. Des enregistrements intracellulaires dans les neurones pr6ganglionnaires sympathiques (NPS) de tranches de moelle Cpinikre de rats nConatals et de chats adultes rCvklent quatre types de potentiels synaptiques, soit les potentiels postsynaptiques excitateurs (PPSE), les potentiels post-synaptiques inhibiteurs (PPSI), les PPSE lents, de meme que des PPSI lents observCs dans les NPS des chats. Des Ctudes pharmacologiques montrent que le glutamate ou un aeide amink exeitateur voisin et la glycine sont vraisemblablement les mCdiateurs des PPSE et des PPSH. Des mCdiateurs hCtCrogknes de PPSE lents pourraient exister; la substance P, la sCrotonine, la norCpinCphrine et 19CpinCphrinesont les mCdiatebars probables des PPSE lents dans les sous-populations de NPS. La norCpintphrine semble Ctre le transmetteur des PPSH lents. Enfin, la stimulation des racines ventrales produit un potentiel synaptique qui semble Ctre lit5 2 la liberation de glutamate par les fibres affkrentes des racines ventrales. Nos rCsultats, obtenus dans les NPS de rats, indiquent qu'il existe une multitude de mCcanismes synaptiques traitant de manikre coordonnke les influx provenant des neurones sbapraspinaux et sensoriels, Ctablissant ainsi une transmission diff6renciCe et hautement stmcturCe, dirigCe vers de multiples cellules cibles pCriphCriques. Mots c l b : neurones prbganglionnaires sympathiques, potentiels post-synaptiques excitateurs, potentiels post-synaptiques inhibiteurs, potentiels post-synaptiques excitateurs lents, glutamate, glycine. [Traduit par la rCdaction]
Introduction The sympathetic nenrous system in response to external and internal stimuli executes complex and highly differentiated commands to multiple target organs. For example, when presented with a threatening stimulus, humans as well as animals display a series of physiological responses including increased cardiac activity, pupillary dilation, constriction of the cutaneous and splanchnic vasculatures, dilation of the skeletal muscle vasculatures, and increased glucose metabolism (Cannon is 1939). The completion of this complex and organized critically dependent on the temporal and spatial coordination of hundreds or thousands of synaptic events at different levels of the neuraxis. Anatomically, the efferent portion of the sympathetic nervous system consists of two sets of serially connected neurons: the preganglionic neurons whose cell bodies reside in the thoracolumbar spinal cord and the postganglionic neurons that 'This paper was presented at the satellite symposium of the International Brian Research Organization meeting held August 12- 14, 1991, at the University of Alberta, Edmonton, Alta., Canada, entitled m e Physiology, Pharmacology, and Biophysics of Ganglionic Transmission, and has undergone the Journal's usual peer review. 2Author for correspondence. 'Present address: ' ~ e ~ a r t m e nof t Physiology, Tokyo Medical College, Tokyo, Japan. Printed in Canada / IrnprirnC au Canada
aggregate in either the para- or pre-vertebral ganglia. While much information has been gathered with respect to the synaptic mechanism involved in transmitting preganglionic and Sensory information to the postganglionic neurons (see, for
Karczmar et 1986), Our concerning transmission from sensory afferents and supraspinal to preganglionic neurons (SPNs) is at best rudimentary. In the last few years the development a cat and rat cord slice preparation has enabled several laboratories to undertake a detailed and systematic investigation of the electro~h~siological and ~ ~ characteristics~Of synaptic transmission in the SPNs. Here, the synaptic pathways, the putative transmitter systems, and the mechanism of various synaptic responses in the SPNs will be reviewed. While much of the data presented here are collected from the in vitro neonate rat SPNs, features unique to the in vitro and in vivo cat SPNs will be highlighted and, when appropriate, contrasted with those of the rat SPNs. The procedures involved in current-clamp and single-electrode voltage-clamp recordings from SPNs in thin (5M) Fm) spinal rats have been described cord 'lice' of immature (12-20 previously (Bun and Mo 1989; Miyazaki et al. 1989; Shen et al. 1990). Some of the more recent data are gathered from neonatal rat SPNs in spinal slices using the whole-cell patch configuration.
~
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
DUN BT AL.
FIG. I. A sympathetic preganglionic neuron filled with lucifer yellow, situated in the intermediolateral cell column of a transverse slice of a neonatal rat spinal cord. The axon (arrowheads) coursed along the border of the ventral horn. Two long dendrites projected medially toward the central canal and several short dendrites with cut ends oriented to the lateral white matter. Calibration bar: 100 pm. (From Shen and Dun 1990.)
A better appreciation of the synaptic mechanism of a given pathway requires, in addition to knowledge of the mechanism of action of putative transmitters on target cells, information about the spatial orientation and morphology of the cells in question as well as the content of putative transmitters within a specific pathway. A brief review of relevant information will precede the discussion of synaptic responses in the SPNs.
Localiaaticsm and morphology of SpNs SPNs in the mammalian spinal cord are organized into four nuclei, i.e., the nucleus intermediolateral pars principalis (ILp), nucleus intermediolateral pars funicularis, intercalated cell group, and central autonomic nucleus; the majority of SPNs are concentrated in the ILp (Petras and Cummings 1972). While the size, morphology ,-dendritic orientation, and axsnal projection of SPNs at these four nuclei differ (Forehand 1990), SPNs in the ILp of neonatal rats appear to be fairly uniform insofar as these parameters are concerned (Shen and Dun 1990). A typical SPN labelled with lucifer yellow in the l[Lpis shown in Fig. 1. In the present context, our recordings are made from SPNs mainly located in the HLp. Generally, the cell bodies of SPNs in the ILp appear fusiform, oval, or round in shape with a mean mediolaterd and dorsoventral diameter of 23 and 18 pm (Shen and Dun 1996). The axon arises either directly from the cell body (Fig. 1) or from the
base of a dendrite and courses ventrally along the border of gray matter before exiting the ventrolateral corner of the ventral horn. In addition to a rostro-caudal projection, several primary dendrites of SPNs in the ILp project medially and terminate dorsal to the central canal (Fig. 1; Shen and Bun 1990; Forehand 1998). This is in contrast with cat SPNs, in which the dendrites orient in a mostly rostro-caudal direction (Dembswsky et al. 198%). As the dendritic domain of neonatal rat SPNs appears to be much more outreaching compared with that of cat SPNs, the dendrites of neonatal rat SPNs may interact with inputs arising from several different directions, whereas synaptic interaction of cat SPNs may be restricted to inputs arising mainly from a rostra-caudal direction (Bembowsky et al. 198%).
Synaptic conneetioms and transmitter immunoreactivity The development of sensitive tract-tracing techniques and the availability of antibodies to an array s f putative transmitters have furnished a wealth of information concerning the afferent and central inputs to the lateral horn and the transmitter phenotypes of these inputs. Several comprehensive reviews on these topics have appeared (Gibson and Polak 1986; Coote 1988; Cabot 1990; Loewy 1990). In the present context. two broadly categorized inputs are distinguished: visceral and somatic spinal afferents, and inputs from supraspinal neurons
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
S88
CAN. J . PHYSIOL. PHARMACOL. VOL. 70, 1992
are similar insofar as the resting potential, input resistance, and time constant are concerned. However, the afterhyperpolarization (AHP) observed in in vifm and in vkvo adult cat SPNs (Dembowsky et al. 1986; Yoshimura et al. 1986) is substantially longer (1 -2 s vs. 50 -300 ms) than that present in neonatal rat SPNs (Dun and Mo 1989). It remains to be established whether the difference is species or age related. Insofar as ira vifro versus in vivo preparations are concerned, the input resistance (60- 110 Mil) measured from in v i m cat and rat SPNs is higher than that (23 Ma) obtained from in vivo cat preparations. The reason for such a difference is not clear. An abundance of ongoing synaptic activities in in vivs cat SPNs may short circuit the membrane, thus resulting in a lower input resistance (McLachlan and Hirst 1980). Under the whole-cell patch configuration a higher value of input resistances (100800 Mil) was obtained in neonatal rat SPNs. An interesting membrane property intrinsic to a small population of SPWs involves the generation of rhythmic sub- m d supra-threshold activities (Spanswick and Logan 19906; E. Shen and N. J. Dun, unpublished observations). The rhythmic activities are not abolished in a low CaB solution or by FIG. 2. Effects of membrane polarization and pharmacological C ~ Z +raising , the possibility that a small population of SPNs agents on eke amplitbade of fast excitatory postsynaptic potentials possesses the intrinsic ability to initiate rhythmic oscillations. (EPSPs) evoked by dorsal-root stimulation in two preganglionic neua population of SPNs may function as As a rons. (A) Each trace consists of an EPSP followed by a hyperpolarizpacemaker cells as has been in many other areas of ing electrotonic potential induced by a current pulse (not shown). the brain including the thalamus (Jahnsen and Llinas 1984) and Membrane hypewolarization by steady dc current at 1 0 - m ~increcerebellum (Llinas and Sugimori 1980). Interestingly, our ments reducedthd EPSP without causing a significant change of electrotonic potentials. Numbers to the left of each trace denote study in which neonatal rat SPNs are filled with Iucifer yellow membrane potential at which responses were elicited. (B) Superreveals a small number of dye-coupled cells, raising the possifusing the spinal slice with a Mg2'-free solution increased the bility that a population of SPNs may be electrically coupled amplitude as well as the duration of the dorsal root evoked EPSP. (%henand Dun 1990). Whether or not the rhythmicity is a con2-Amino-5-phosphsnovaleric acid (APV; 10 pM) completely and sequence of the activity of the electrically coupled neurons reversibly blocked the EPSP, whereas 6-cyano-7-nitroquinoxalineremains to be resolved. 2,3-dione (CNQX; 1 pM) was without significant effects. The recordings in A and B are taken from two different preganglionic neurons. Con, control. (Modified from Shen et al. 1990.)
(Loewy 1990). As will be shown later, stimulation of dorsal rootlets and the lateral hniculus, presumably activating visceral-somatic afferents and descending inputs, respectively, elicits qualitatively different synaptic potentials. Immunoreactivities to a plethora of putative transmitters have been detected in the nerve fibers and cell bodies of the spinal autonomic nuclei (Kmkoff et al. 1985; Gibson and Polak 1986; Cabot 1990). In addition to immunoreactivities to the putative excitatory and inhibitory amino acid transmitters glutamate, 7-aminobutyric acid, and glycine, immunoreactivities to a long list of mines and neuropeptides including serotonin (5-HT), norepinephrine, epinephrine, substance P, thyrstropinreleasing hormone (TWH),vasopressin, galanin, and metenkephalin have been observed in the lateral horn. Ira addition, two or more putative transmitters have been localized to the same nerve terminals; for example, some of the raphe fibers projecting to the spinal cord appear to contain 5-HT, TRH, and substance P (Johansson et al. 1981).
EIectrophysioIogid properties of SPNs The electrical properties of in vitro and in sdtu cat and in vitro neonatal rat SPNs have been reported by a number of investigators (Coote and Westbury 1979; McLachlan and Hirst 1980; Dembowsky et al. 1985a, 1986; Yoshimura et al. 1986; Dun and Mo 1988; Spanswick and Logm 1 9 9 0 ~ ~Gen). erally, the membrane properties of in vitro cat and rat SPNs
Evoked synaptic potentials Altogether, four types of synaptic potentials have been identified in rat and cat SPNs in vitro: a fast and a slow excitatory postsynaptic potential (EPSP) and a fast and a slow inhibitory postsynaptic potential (HPSP) (Dun and Mo 1988; Shen et al. 1990; Yoshimura et al. 1987a, 1987b). Except for the fast EPSP, the other three types of response may not necessarily occur in every SPN. Fast EPSPs Electrical stimulation of dorsal roots elicits in neonatal rat SPNs a long-latency (2 - 10 ms) EPSP that can be graded as a function of stimulus intensity and which, when reaching the threshold, produces one or more action potentials. Because of the long and varying latency, the dorsal root evoked EPSPs probably represent di- or poly-synaptic events. The response is decreased by either depolarizing or hyperpolarizing the membrane and is enhanced in a Mg2+-free solution (Fig. 2). The characteristics of the dorsal root evoked EPSP are then similar to that associated with activation of N-methylaaspartate (NMDA) receptors reported in other central neurons (Mayer and Westbrook 1987). Indeed, pharmacological studies show that the relatively selective NMDA receptor antagonists 2-amino-5-phosphonovaleric acid (APV) and ketamine, but not the non-NMDA receptor antagonists Q,7-dinitroquinoxaline-%,3-dione (DNQX) and 6-cyano-'7-nitroquinoxaline-2,3dione (CNQX), block the dorsal root evoked EPSPs (Shen et al. 1990). Lastly, pressure application of NMDA evokes a
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
DUN ET AL.
FIG, 3. Effects of membrane potential and pharmacological agents on the EPSPs evoked by stimulation of the lateral hniculus in two preganglionic neurons. (A) The EPSP amplitude was increased by membrane hyperpolarization and decreased by depolarization Doran the resting membrane potential sf -65 mV. (B) The EBSP was not significantly changed in a Mg2+-freesolution nor by ABV, whereas CNQX (I pM) reversibly blocked the EPSP. Recordings in A and B are taken from two different preganglionic neurons.
Frc. 4. Effects of membrane potential and low extracellular Cl- and K + concentrations on the inhibitory postsynaptic potentials (IPSPs) in two preganglionic neurons. (A) Dorsal-root stimulation elicited an IPSP at the resting membrane potential sf - 54 mV. The EPSP was increased and decreased by depolarizing and hyperpolarizing the membrane, respectively: at -85 mV, the IPSP was reversed. The graph shows the relationship between the amplitude of evoked response and membrane potential; the line intercepts at -66 mV, which is taken as the reversal potential of the IPSP. (B) Superfusing the slice with a low-Cl- (4.3 mM) solution depressed and eventually reversed the ventral root evoked IBSB, whereas superfusing the slice with a Bow-KC (1 -9 mM) solution had no significant effect on the EPSP. Recordings in A and B are taken from two different preganglionic neurons. (From Dun and Mo 1989.)
depolarization or inward current with characteristics similar to that sf dorsal root evoked EPSPs (Shen et d. 1990). As the NMDA-induced current is reduced in a low-Ca2+ solution or by Cd2+ as well as by Na-deficient solutions, activation of NMDA receptors may increase membrane permeability to CaZ+, Na+, and K+ (Dun and Miymaki 1988). It follows that a similar ionic mechanism may underlie the dorsal root evoked EPSP. On the other hand, stimulation of the lateral funiculus, presumably by descending fibers, elicits a short-latency EPSP
with a faster rise time and a shorter duration compared with that of dorsal root evoked EPSPs. More importantly, the response is increased by hyperpolarization and decreased by deplarization, and the extrapolated reversal potential is close to 0 mV. The response is not significantly enhanced by Mg2+-free solution; BNQX and CNQX block the EPSPs, whereas APV and kehmine are not effective (Fig. 3). The results collectively suggest that the EPSP evoked by lateral hniculus stimulation is mediated by non-NMDA receptors (%henet al. 1998). In the cat, focal stimulation elicits in the
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
,590
CAW. J. PWYSIOL. PHARMACOL. VOL. 70, 1992
FIG.5. Frequency dependence of slaw EPSPs and input-resistance changes in two preganglionic neurons. (A) A single stimulus (arrowhead) applied to the lateral fianiculus elicited a fast EPSP (&st upward deflection) followed by a slow EPSP. Increasing the stimulus frequency to 30 Hz and the duration to 1, 2, and 3 s (bars below the trace) caused a progressively larger and longer duration slow EPSP. (B) A train of stimuli (arrowhead; 30 Hz, 1 s) applied to the lateral funiculus elicited a slow EPSP not preceded by a fast EPSP. The input resistance as measured from the amplitude of the hyperpolarizing electrotonis potential was not significantly changed during the course of the slow EPSP. When the membrane potential was manually clamped, a clear increase of membrane resistance was noted.
SPNs an EPSP that can be characterized as an event mediated by a non-NMDA receptor (Nishi et al. 1987).
zation, and is nullified around -90 mV; and (iii) the membrane excitability is increased during the course of the slow EPSP. A representative recording of slow EPSPs is shown in Fig. 5. Studies using antagonists to putative transmitters indicate that the transmitter mediating the slow EPSP is heterogenous. For example, the slow EPSP evoked in cat SPNs is blocked by the al-antagonist prazosin (Yoshimura et al. 1987a), whereas the slow EPSPs evoked in rat SPNs can be suppressed by several different antagonists including the al-antagonist prazosin and the 5-HT antagonist ketansarin (S. Y. Wu and N. J. Dun, unpublished results). The peptide substance P can also eliminate the slow EPSPs evoked in some SPNs, presumably by desensitizing the receptors (Bun and Mo 1988). Moreover, the putative transmitters noradrenaline, 5-HT, vasopressin, and substance P when applied to SPNs produce a slow depolarization with characteristics similar to those of slow EPSPs. While the slow EPSP is a common feature in rats and cats, the slow IPSP has been observed only in cat SPNs (Yoshimura et al. 1987b). The response is associated with decreased input resistance, reverses around -90 mV, and is increased and decreased in low- and high-Kg solution, respectively. Finally, the a2-antagonist yohimbine blocks the response, whereas noradrenaline mimics the slow hyperpolarizing response. Collectively, these findings suggest that noradrenaline acting on a2-receptors is responsible for generating the slow IPSP in cat SPNs (Yoshimura et al. 1987b).
Fast bPSBs Stimulation of dorsal roots elicits in about 30% of SPNs an IPSP with a mean half-decay time of 17 ms and an amplitude of 8 mV; in a portion of SPNs the IPSP is preceded by an EPSP (Dun and Mo 1989). The HPSP is made smaller by conditioning hyperpolarization and larger by conditioning depolarization, with a reversal potential of -70 mV; the response is reversed in a low-C1- solution but not changed significantly in a low-Kg solution (Fig. 4). Finally, the glycine antagonist strychnine, but not the GABA antagonist bicuculline, blocks the IPSP. Thus, it is concluded that the IPSP is caused by an increased membrane permeability to Cl- ions owing to glycine released from inhibitory interneurons (Dun and Mo 1989). In cat SPNs the occurrence of IPSPs appears to be infrequent and the response has yet to be characterized (Nishi et al. 1987). Interestingly, activation of ventral roots elicits in a small population of SPNs an IPSP with electrophysiologica1 characteristics similar to that evoked by dorsal root stimulation. The response is blocked by nicotinic antagonists d-tubocurarine and hexamethoniurn, in addition to strychnine. These observations raise the possibility that the IPSP evoked by stimulation of ventral roots is caused by the activation of (a) glycinecontaining interneuron(s) by axon collaterals via nicotinic action (Dun and Mo 1989). Intraspinal axon branchings have been noted in a small number of neonatal rat SPNs labelled with horseradish peroxidase, thus providing a potential morphological substrate for recurrent IPSPs (Forehand 1990).
Ventral root evoked EPSBs An interesting and novel response involves the transmission of ventral root afferents to neonatal SPNs. Stimulation of ventral roots evokes in about 48% of SPNs an EPSP. The response is graded with stimulus intensity and, when reaching the threshold, fires an action potential. The ventral root evoked EBSP occurs more often in SPNs when the resting membrane potential is high ( 1 - 60 mV). Under this condition. stimulation of ventral roots with low intensities evokes an EPSP rather than an antidromic spike; the latter can be effectively induced by depolarization of the membrane. The EPSP is reversibly abolished by low-Ca2+ solution and the falling phase in some SPNs is prolonged in Mg2+-freesolutions. The response is not affected by the nicotinic antagonists mecamylamine and dihydro-P-erythroidine, indicating that it is not due to the activation of axon collaterals. Lastly. the non-NMDA receptor antagonist CNQX eliminates the EPSP, while APV shortens the falling phase in some of the neurons. These observations collectively suggest that the EPSP evoked by stimulation of ventral roots is caused by activation of afferent fibers, resulting in a release of glutamate or a related amino acid. A similar observation has been reported with respect to the motoneurons (Jiang et al. 1991).
Slow synaptic potentials In addition to the fast EPSP and IPSP, stimulation s f the lateral fianiculus and, in a few instances, dorsal rootlets elicits in rat and cat SPNs a slow EPSP and in cat SPNs a slow IPSP with a time course of seconds to minutes (Yoshimura et al. 1987a, 1987b; Dun and Mo 1988). The slow EPSPs have the following general features: (i) the amplitude and duration are graded with stimulus intensity and summated with increasing number of pulses; (ii) the response is accompanied by an increase of membrane resistance, is increased with depolarization and decreased with hyperpolari-
The autonomic nervous system controls and coordinates virtually all visceral, vascular, and glandular activities, both tonically and in response to sudden environmental demands. Accordingly, autonomic outflow to various tissues is highly differentiated even under basal conditions and must be reset rapidly to meet the moment-to-moment demands of the srganism. Recent results from in vitro rat and cat spinal cord slices show that a range of synaptic mechanisms is available to SPNs, implicating a complex and variable control of signalling hnction. Altogether four types of synaptic responses have been identified in cat and rat SPNs. The first two types, the
Summary
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
fast EPSP and fast IPSP, involve rapid excitation and inhibition, resulting from an increase of cation and anion permeability. In rat SPNs, glutamate or a closely related excitatory amino acid and glycine appear to be the excitatory and inhibitory transmitters for the fast EPSP and IPSP, respectively. The second type of synaptic mechanism involves excitation and inhibition of a relatively long time course as compared with that of fast synaptic events; the slow EPSP and slow IPSP are likely to be caused by a decrease and increase of membrane K permeability, respectively. The mediator for the slow synaptic events appears to be heterogenous. In the case of slow EPSPs, the pharmacological finding is consistent with the immnohistochemical observation that a variety of putative transmitters including amines and neuropeptides present in various afferents and descending fibers may serve as transmitters. The role of the slow IPSPs is not well established, as SPNs expressing this type of responses are small in number and may not even be present in the neonatal rat. In cat SPNs, noradrenaline acting on ar2-receptors is proposed to be the mediator. In summary, there exists in cat and rat SPNs a multitude of synaptic mechanisms via which inputs arising from sensory and supraspinal neurons can be temporally and correctly processed into orderly and purposeful outputs, thus ensuring the moment-by-rnoment regulation of individual organs. +
Acknowledgement This work was supported by National Institutes s f Health grant NS18710 from the Department of Health and Human Services. Cabot, J. B. 1990. Sympathetic preganglionic neurons: cytoarchitecture, ultrastmcture, and biophysical properties. In Central regulation of autonomic functions. Edited by A. D. Loewy and K. M. Spyer. Oxford University Press, New York. pp. 44 -67. Cannon, W. B. 1939. The wisdom of the body. 2nd ed. Norton, New York. Coote, J. H. 1988. The organisation of cardiovascular neurons in the spinal cord. Rev. Physiol. Biochem. Pharmacol. 110: 148-285. Coote, J. H., and Westbury, D. R. 1979. IntracelBuBar recordings from sympathetic preganglionic neurones. Neurosci . Lett. 15: 171- 175. Dembowsky, K., Czachurski, J., and Seller, H. 1985~.An intracellular study of the synaptic input to sympathetic preganglionic neurones of the third thoracic segment of the cat. J. Auton. Nerv. Syst. 13: 201-244. Dembowsky , K., Czachurski, J., and Seller, H. 19851s. Morphology of sympathetic preganglionic neurons in the thoracic spinal cord of the cat: an intracellular horseradish peroxidase study. J. Comp. Neurol. 238: 453 -465. Dembowsky, K., Czachurski, J., and Seller, H. 1986. Three types of sympathetic preganglionic neurones with different electrophysiological properties are identified by intracellular recordings in the cat. Pfluegers Arch. 406: 112- 126). Dun, N. J., and Miyazaki, T. 1988. NMDA-induced current in neonatal rat lateral horn cells. Soc. Neurosci. Abstr. 14: 318.16). Dun, N. J . , and Mo, N. 1988. In vitro effects of substance P on neonatal rat sympathetic preganglionic neurones. J. Physiol. (London), 399: 321-333. Dun, N. J., and Mo, N. 1989. Inhibitory postsynaptic potentials in neonatal rat sympathetic preganglionic neurones in vitro. J. Physisl. (London), 410: 267 -28 1. Forehand, C. J . 1990. Morphology of sympathetic preganglionic neurons in the neonatal rat spinal cord: an intracellular horseradish peroxidase study. J. Comp. Neurol. 298: 334 - 342. Gibson, S. J., and Polak, J. M. 1986. Neurochemistry of the spinal
cord. In Immunocytochemistry. Modern methods and application. 2nd ed. Edited by J. M. Polak and S. Van Norrdan. John Wright and Sons Ltd., Bristol. pp. 360-389. Jahnsen, H., and Llinas, R. 1984. Ionic basis for the electroresponsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro. J. Physiol. (London), 349: 227 -247. Jiang, Z. G., Shen, E., Wang, M. Y., and Dun, N. J. 1991. Excitatory postsynaptic potentials evoked by ventral root stimulation in neonate rat motoneurons in vitro. J. Neurophysisl. 65: 57-66. Johansssn, O., Hokfelt, T., Pernow, B., Jeffcoate, S. L., White, N., Steinbusch, H. W. M., Verhofstad, A. A. J., Emson, P. C., and Spindel, E. 1981. Immunohistochemical support for three putative transmitters in one neuron: coexistence of 5-hydroxytryptamine, substance P- and thyrotropin releasing hormone-like immunoreactivity in medullary neurons projecting to the spinal cord. Neuroscience, 6: 1857- 1881. Karczmar, A. G., Koketsu, K., and Nishi, S. 1986. Autonomic and enteric ganglia. Transmission and its pharmacology. Plenum Press, New York. Kmkoff, T. L., Ciriello, J., and Calaresu, F. R. 1985. Segmental distribution of peptide-like immunoreactivity in cell bodies of the thoracolumbar sympathetic nuclei of the cat. J. Comp. Neurol. 240: 90- 102. Llinas, R., and Sugimori, M. 1980. Electrophysiological properties of in vitro Wlrkinje cell somata in mammalian cerebellar slices. J . Physiol. (London), 305: 171- 195. Loewy , A. D. 1990. Central autonomic pathways. In Central regulation of autonomic functions. Edited by A. B . Loewy and K. M . Spyer. Oxford University Press, New York. pp. 88- 103. Mayer, M. L., and Westbrook, G. L. 1987. The physiology of excitatory amino acids in the vertebrate central nervous system. Prog . Neurobiol. (Oxford), 28: 197 - 276. McLachBan, E. M., and Hirst, D. G. S. 1986). Some properties of preganglionic neurones in the upper thoracic spinal cord of the cat. J. Neurophysiol. 43: 1251 - 1265. Miyazaki, T., Coote, J. H., and Dun, N. J. 1989. Excitatory and inhibitory effects of epinephrine on neonatal rat sympathetic preganglionic neurons in vitro. Brain Res. 407: 108- 116. Nishi, S., Yoshimura, M., and Polosa, C. 1987. Synaptic potentials and putative transmitter actions in sympathetic preganglionic neurons. In Organizationof the autonomic nervous system: central and peripheral mechanisms. Edited by J. Ciriello, F. R. Calaresu, L. P. Renaud, and C. Polosa. Alan R. Liss, Inc., New York. pp. 15-26. Petras, J. M., and Cummings, J. F. 1972. Autonomic neurons in the spinal cord of the rhesus monkey. A correlation of the findings of cytoarchitectonicsand sympathectomy with fiber degeneration following dorsal rhizotomy. J. Comp Neurol. 146: 189-218. Shen, E., and Dun, N. J. 1990. Neonate rat sympathetic preganglionic neurons intracellularly labelled with lucifer yellow in thin spinal cord slices. J. Auton. New. Syst. 29: 247-254. Shen, E., Mo, N., and Dun, N. J. 1990. APV-sensitive dorsal root afferent transmission to neonate rat sympathetic preganglionic neurons in vitro. J. Neurophysiol. 64: 991 -999. Spanswick, D., and Logan, S. D. 1 9 9 0 ~Sympathetic . preganglionic neurones in neonatal rat spinal cord in vitro: eBectrophysiologica1 characteristics and the effects of selective excitatory amino acid receptor agonists. Brain Res. 525: 181 - 188. Spanswick, D., and Logan, S. B. 1990b. Spontaneous rhythmic activity in the intermediolateral cell nucleus of the neonate rat thoracolumbar spinal cord in vitro. Neuroscience, 39: 395 -403. Yoshimura, M., Polosa, C., and Nishi, S. 1986. After-hyperpolarization mechanisms in cat sympathetic preganglionic neuron in vitro. J. Neurophysiol. 55: 1234- 1246. Yoshimura, M., Polosa, C., and Nishi, S. 1987a. Slow EPSP and the depolarizing action of noradrenaline on sympathetic preganglionic neurons. Brain Res, 414: 138- 142. Yoshimura, M., Polosa, C., and Nishi, S. 1987b. Slow IPSP and the noradrenaline-induced inhibition of the cat sympathetic preganglionic neuron in vitro. Brain Res. 419: 383-386.
N. J.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
Received September 23, 1991 BUN, N. J., Wu, S. Y., SHEN,E., MIYAZAKI, T., DUN, S. L., and REN, C. 1992. Synaptic mechanisms in sympathetic preganglionic neurons. Can. J. Physiol. Pharmacol. 70: S86 - S9 1. Intracellular recordings from sympathetic preganglionic neurons (SPNs) in adult cat and neonatal rat spinal cord slices reveal four types of synaptic potentials, namely, excitatory postsynaptic potentials (EPSPs), inhibitory postsynaptic potentials (IPSPs), and slow EPSPs in both preparations, and a slow IPSP in cat SPNs. Pharmacological studies show that glutamate or a related excitatory amino acid and glycine are the probable mediators of EPSPs and IPSPs. There may be heterogenous mediators of slow EPSPs; substance P, serotonin, norepinephrine, and epinephrine are all probable mediators of slow EPSPs in subpopulations of SPNs. In the case of slow IPSPs, norepinephrine appears to be the likely transmitter. Finally, stimulation of ventral roots elicits a synaptic potential that appears to be caused by glutamate released from afferent fibers in the ventral roots. Our results indicate that a multitude of synaptic mechanisms exist in the rat SPNs by means of which inputs arising from sensory and supraspinal neurons are processed in a timely and orderly manner, thus ensuring highly organized but differentiated outputs to multiple peripheral target cells. Key words: sympathetic preganglionic neurons, excitatory postsynaptic potentials, inhibitory postsynaptic potentials, slow excitatory postsynaptic potentials, glutamate, glycine. DUN, N. J., WU, S. Y., SHEN,E., MIYAZAKI, T., DUN, S. L., et REN, C. 1992. Synaptic mechanisms in sympathetic preganglionic neurons. Can. J. Physiol. Pharmacol. 190 : S86 - S9 1. Des enregistrements intracellulaires dans les neurones pr6ganglionnaires sympathiques (NPS) de tranches de moelle Cpinikre de rats nConatals et de chats adultes rCvklent quatre types de potentiels synaptiques, soit les potentiels postsynaptiques excitateurs (PPSE), les potentiels post-synaptiques inhibiteurs (PPSI), les PPSE lents, de meme que des PPSI lents observCs dans les NPS des chats. Des Ctudes pharmacologiques montrent que le glutamate ou un aeide amink exeitateur voisin et la glycine sont vraisemblablement les mCdiateurs des PPSE et des PPSH. Des mCdiateurs hCtCrogknes de PPSE lents pourraient exister; la substance P, la sCrotonine, la norCpinCphrine et 19CpinCphrinesont les mCdiatebars probables des PPSE lents dans les sous-populations de NPS. La norCpintphrine semble Ctre le transmetteur des PPSH lents. Enfin, la stimulation des racines ventrales produit un potentiel synaptique qui semble Ctre lit5 2 la liberation de glutamate par les fibres affkrentes des racines ventrales. Nos rCsultats, obtenus dans les NPS de rats, indiquent qu'il existe une multitude de mCcanismes synaptiques traitant de manikre coordonnke les influx provenant des neurones sbapraspinaux et sensoriels, Ctablissant ainsi une transmission diff6renciCe et hautement stmcturCe, dirigCe vers de multiples cellules cibles pCriphCriques. Mots c l b : neurones prbganglionnaires sympathiques, potentiels post-synaptiques excitateurs, potentiels post-synaptiques inhibiteurs, potentiels post-synaptiques excitateurs lents, glutamate, glycine. [Traduit par la rCdaction]
Introduction The sympathetic nenrous system in response to external and internal stimuli executes complex and highly differentiated commands to multiple target organs. For example, when presented with a threatening stimulus, humans as well as animals display a series of physiological responses including increased cardiac activity, pupillary dilation, constriction of the cutaneous and splanchnic vasculatures, dilation of the skeletal muscle vasculatures, and increased glucose metabolism (Cannon is 1939). The completion of this complex and organized critically dependent on the temporal and spatial coordination of hundreds or thousands of synaptic events at different levels of the neuraxis. Anatomically, the efferent portion of the sympathetic nervous system consists of two sets of serially connected neurons: the preganglionic neurons whose cell bodies reside in the thoracolumbar spinal cord and the postganglionic neurons that 'This paper was presented at the satellite symposium of the International Brian Research Organization meeting held August 12- 14, 1991, at the University of Alberta, Edmonton, Alta., Canada, entitled m e Physiology, Pharmacology, and Biophysics of Ganglionic Transmission, and has undergone the Journal's usual peer review. 2Author for correspondence. 'Present address: ' ~ e ~ a r t m e nof t Physiology, Tokyo Medical College, Tokyo, Japan. Printed in Canada / IrnprirnC au Canada
aggregate in either the para- or pre-vertebral ganglia. While much information has been gathered with respect to the synaptic mechanism involved in transmitting preganglionic and Sensory information to the postganglionic neurons (see, for
Karczmar et 1986), Our concerning transmission from sensory afferents and supraspinal to preganglionic neurons (SPNs) is at best rudimentary. In the last few years the development a cat and rat cord slice preparation has enabled several laboratories to undertake a detailed and systematic investigation of the electro~h~siological and ~ ~ characteristics~Of synaptic transmission in the SPNs. Here, the synaptic pathways, the putative transmitter systems, and the mechanism of various synaptic responses in the SPNs will be reviewed. While much of the data presented here are collected from the in vitro neonate rat SPNs, features unique to the in vitro and in vivo cat SPNs will be highlighted and, when appropriate, contrasted with those of the rat SPNs. The procedures involved in current-clamp and single-electrode voltage-clamp recordings from SPNs in thin (5M) Fm) spinal rats have been described cord 'lice' of immature (12-20 previously (Bun and Mo 1989; Miyazaki et al. 1989; Shen et al. 1990). Some of the more recent data are gathered from neonatal rat SPNs in spinal slices using the whole-cell patch configuration.
~
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
DUN BT AL.
FIG. I. A sympathetic preganglionic neuron filled with lucifer yellow, situated in the intermediolateral cell column of a transverse slice of a neonatal rat spinal cord. The axon (arrowheads) coursed along the border of the ventral horn. Two long dendrites projected medially toward the central canal and several short dendrites with cut ends oriented to the lateral white matter. Calibration bar: 100 pm. (From Shen and Dun 1990.)
A better appreciation of the synaptic mechanism of a given pathway requires, in addition to knowledge of the mechanism of action of putative transmitters on target cells, information about the spatial orientation and morphology of the cells in question as well as the content of putative transmitters within a specific pathway. A brief review of relevant information will precede the discussion of synaptic responses in the SPNs.
Localiaaticsm and morphology of SpNs SPNs in the mammalian spinal cord are organized into four nuclei, i.e., the nucleus intermediolateral pars principalis (ILp), nucleus intermediolateral pars funicularis, intercalated cell group, and central autonomic nucleus; the majority of SPNs are concentrated in the ILp (Petras and Cummings 1972). While the size, morphology ,-dendritic orientation, and axsnal projection of SPNs at these four nuclei differ (Forehand 1990), SPNs in the ILp of neonatal rats appear to be fairly uniform insofar as these parameters are concerned (Shen and Dun 1990). A typical SPN labelled with lucifer yellow in the l[Lpis shown in Fig. 1. In the present context, our recordings are made from SPNs mainly located in the HLp. Generally, the cell bodies of SPNs in the ILp appear fusiform, oval, or round in shape with a mean mediolaterd and dorsoventral diameter of 23 and 18 pm (Shen and Dun 1996). The axon arises either directly from the cell body (Fig. 1) or from the
base of a dendrite and courses ventrally along the border of gray matter before exiting the ventrolateral corner of the ventral horn. In addition to a rostro-caudal projection, several primary dendrites of SPNs in the ILp project medially and terminate dorsal to the central canal (Fig. 1; Shen and Bun 1990; Forehand 1998). This is in contrast with cat SPNs, in which the dendrites orient in a mostly rostro-caudal direction (Dembswsky et al. 198%). As the dendritic domain of neonatal rat SPNs appears to be much more outreaching compared with that of cat SPNs, the dendrites of neonatal rat SPNs may interact with inputs arising from several different directions, whereas synaptic interaction of cat SPNs may be restricted to inputs arising mainly from a rostra-caudal direction (Bembowsky et al. 198%).
Synaptic conneetioms and transmitter immunoreactivity The development of sensitive tract-tracing techniques and the availability of antibodies to an array s f putative transmitters have furnished a wealth of information concerning the afferent and central inputs to the lateral horn and the transmitter phenotypes of these inputs. Several comprehensive reviews on these topics have appeared (Gibson and Polak 1986; Coote 1988; Cabot 1990; Loewy 1990). In the present context. two broadly categorized inputs are distinguished: visceral and somatic spinal afferents, and inputs from supraspinal neurons
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
S88
CAN. J . PHYSIOL. PHARMACOL. VOL. 70, 1992
are similar insofar as the resting potential, input resistance, and time constant are concerned. However, the afterhyperpolarization (AHP) observed in in vifm and in vkvo adult cat SPNs (Dembowsky et al. 1986; Yoshimura et al. 1986) is substantially longer (1 -2 s vs. 50 -300 ms) than that present in neonatal rat SPNs (Dun and Mo 1989). It remains to be established whether the difference is species or age related. Insofar as ira vifro versus in vivo preparations are concerned, the input resistance (60- 110 Mil) measured from in v i m cat and rat SPNs is higher than that (23 Ma) obtained from in vivo cat preparations. The reason for such a difference is not clear. An abundance of ongoing synaptic activities in in vivs cat SPNs may short circuit the membrane, thus resulting in a lower input resistance (McLachlan and Hirst 1980). Under the whole-cell patch configuration a higher value of input resistances (100800 Mil) was obtained in neonatal rat SPNs. An interesting membrane property intrinsic to a small population of SPWs involves the generation of rhythmic sub- m d supra-threshold activities (Spanswick and Logan 19906; E. Shen and N. J. Dun, unpublished observations). The rhythmic activities are not abolished in a low CaB solution or by FIG. 2. Effects of membrane polarization and pharmacological C ~ Z +raising , the possibility that a small population of SPNs agents on eke amplitbade of fast excitatory postsynaptic potentials possesses the intrinsic ability to initiate rhythmic oscillations. (EPSPs) evoked by dorsal-root stimulation in two preganglionic neua population of SPNs may function as As a rons. (A) Each trace consists of an EPSP followed by a hyperpolarizpacemaker cells as has been in many other areas of ing electrotonic potential induced by a current pulse (not shown). the brain including the thalamus (Jahnsen and Llinas 1984) and Membrane hypewolarization by steady dc current at 1 0 - m ~increcerebellum (Llinas and Sugimori 1980). Interestingly, our ments reducedthd EPSP without causing a significant change of electrotonic potentials. Numbers to the left of each trace denote study in which neonatal rat SPNs are filled with Iucifer yellow membrane potential at which responses were elicited. (B) Superreveals a small number of dye-coupled cells, raising the possifusing the spinal slice with a Mg2'-free solution increased the bility that a population of SPNs may be electrically coupled amplitude as well as the duration of the dorsal root evoked EPSP. (%henand Dun 1990). Whether or not the rhythmicity is a con2-Amino-5-phosphsnovaleric acid (APV; 10 pM) completely and sequence of the activity of the electrically coupled neurons reversibly blocked the EPSP, whereas 6-cyano-7-nitroquinoxalineremains to be resolved. 2,3-dione (CNQX; 1 pM) was without significant effects. The recordings in A and B are taken from two different preganglionic neurons. Con, control. (Modified from Shen et al. 1990.)
(Loewy 1990). As will be shown later, stimulation of dorsal rootlets and the lateral hniculus, presumably activating visceral-somatic afferents and descending inputs, respectively, elicits qualitatively different synaptic potentials. Immunoreactivities to a plethora of putative transmitters have been detected in the nerve fibers and cell bodies of the spinal autonomic nuclei (Kmkoff et al. 1985; Gibson and Polak 1986; Cabot 1990). In addition to immunoreactivities to the putative excitatory and inhibitory amino acid transmitters glutamate, 7-aminobutyric acid, and glycine, immunoreactivities to a long list of mines and neuropeptides including serotonin (5-HT), norepinephrine, epinephrine, substance P, thyrstropinreleasing hormone (TWH),vasopressin, galanin, and metenkephalin have been observed in the lateral horn. Ira addition, two or more putative transmitters have been localized to the same nerve terminals; for example, some of the raphe fibers projecting to the spinal cord appear to contain 5-HT, TRH, and substance P (Johansson et al. 1981).
EIectrophysioIogid properties of SPNs The electrical properties of in vitro and in sdtu cat and in vitro neonatal rat SPNs have been reported by a number of investigators (Coote and Westbury 1979; McLachlan and Hirst 1980; Dembowsky et al. 1985a, 1986; Yoshimura et al. 1986; Dun and Mo 1988; Spanswick and Logm 1 9 9 0 ~ ~Gen). erally, the membrane properties of in vitro cat and rat SPNs
Evoked synaptic potentials Altogether, four types of synaptic potentials have been identified in rat and cat SPNs in vitro: a fast and a slow excitatory postsynaptic potential (EPSP) and a fast and a slow inhibitory postsynaptic potential (HPSP) (Dun and Mo 1988; Shen et al. 1990; Yoshimura et al. 1987a, 1987b). Except for the fast EPSP, the other three types of response may not necessarily occur in every SPN. Fast EPSPs Electrical stimulation of dorsal roots elicits in neonatal rat SPNs a long-latency (2 - 10 ms) EPSP that can be graded as a function of stimulus intensity and which, when reaching the threshold, produces one or more action potentials. Because of the long and varying latency, the dorsal root evoked EPSPs probably represent di- or poly-synaptic events. The response is decreased by either depolarizing or hyperpolarizing the membrane and is enhanced in a Mg2+-free solution (Fig. 2). The characteristics of the dorsal root evoked EPSP are then similar to that associated with activation of N-methylaaspartate (NMDA) receptors reported in other central neurons (Mayer and Westbrook 1987). Indeed, pharmacological studies show that the relatively selective NMDA receptor antagonists 2-amino-5-phosphonovaleric acid (APV) and ketamine, but not the non-NMDA receptor antagonists Q,7-dinitroquinoxaline-%,3-dione (DNQX) and 6-cyano-'7-nitroquinoxaline-2,3dione (CNQX), block the dorsal root evoked EPSPs (Shen et al. 1990). Lastly, pressure application of NMDA evokes a
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
DUN ET AL.
FIG, 3. Effects of membrane potential and pharmacological agents on the EPSPs evoked by stimulation of the lateral hniculus in two preganglionic neurons. (A) The EPSP amplitude was increased by membrane hyperpolarization and decreased by depolarization Doran the resting membrane potential sf -65 mV. (B) The EBSP was not significantly changed in a Mg2+-freesolution nor by ABV, whereas CNQX (I pM) reversibly blocked the EPSP. Recordings in A and B are taken from two different preganglionic neurons.
Frc. 4. Effects of membrane potential and low extracellular Cl- and K + concentrations on the inhibitory postsynaptic potentials (IPSPs) in two preganglionic neurons. (A) Dorsal-root stimulation elicited an IPSP at the resting membrane potential sf - 54 mV. The EPSP was increased and decreased by depolarizing and hyperpolarizing the membrane, respectively: at -85 mV, the IPSP was reversed. The graph shows the relationship between the amplitude of evoked response and membrane potential; the line intercepts at -66 mV, which is taken as the reversal potential of the IPSP. (B) Superfusing the slice with a low-Cl- (4.3 mM) solution depressed and eventually reversed the ventral root evoked IBSB, whereas superfusing the slice with a Bow-KC (1 -9 mM) solution had no significant effect on the EPSP. Recordings in A and B are taken from two different preganglionic neurons. (From Dun and Mo 1989.)
depolarization or inward current with characteristics similar to that sf dorsal root evoked EPSPs (Shen et d. 1990). As the NMDA-induced current is reduced in a low-Ca2+ solution or by Cd2+ as well as by Na-deficient solutions, activation of NMDA receptors may increase membrane permeability to CaZ+, Na+, and K+ (Dun and Miymaki 1988). It follows that a similar ionic mechanism may underlie the dorsal root evoked EPSP. On the other hand, stimulation of the lateral funiculus, presumably by descending fibers, elicits a short-latency EPSP
with a faster rise time and a shorter duration compared with that of dorsal root evoked EPSPs. More importantly, the response is increased by hyperpolarization and decreased by deplarization, and the extrapolated reversal potential is close to 0 mV. The response is not significantly enhanced by Mg2+-free solution; BNQX and CNQX block the EPSPs, whereas APV and kehmine are not effective (Fig. 3). The results collectively suggest that the EPSP evoked by lateral hniculus stimulation is mediated by non-NMDA receptors (%henet al. 1998). In the cat, focal stimulation elicits in the
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
,590
CAW. J. PWYSIOL. PHARMACOL. VOL. 70, 1992
FIG.5. Frequency dependence of slaw EPSPs and input-resistance changes in two preganglionic neurons. (A) A single stimulus (arrowhead) applied to the lateral fianiculus elicited a fast EPSP (&st upward deflection) followed by a slow EPSP. Increasing the stimulus frequency to 30 Hz and the duration to 1, 2, and 3 s (bars below the trace) caused a progressively larger and longer duration slow EPSP. (B) A train of stimuli (arrowhead; 30 Hz, 1 s) applied to the lateral funiculus elicited a slow EPSP not preceded by a fast EPSP. The input resistance as measured from the amplitude of the hyperpolarizing electrotonis potential was not significantly changed during the course of the slow EPSP. When the membrane potential was manually clamped, a clear increase of membrane resistance was noted.
SPNs an EPSP that can be characterized as an event mediated by a non-NMDA receptor (Nishi et al. 1987).
zation, and is nullified around -90 mV; and (iii) the membrane excitability is increased during the course of the slow EPSP. A representative recording of slow EPSPs is shown in Fig. 5. Studies using antagonists to putative transmitters indicate that the transmitter mediating the slow EPSP is heterogenous. For example, the slow EPSP evoked in cat SPNs is blocked by the al-antagonist prazosin (Yoshimura et al. 1987a), whereas the slow EPSPs evoked in rat SPNs can be suppressed by several different antagonists including the al-antagonist prazosin and the 5-HT antagonist ketansarin (S. Y. Wu and N. J. Dun, unpublished results). The peptide substance P can also eliminate the slow EPSPs evoked in some SPNs, presumably by desensitizing the receptors (Bun and Mo 1988). Moreover, the putative transmitters noradrenaline, 5-HT, vasopressin, and substance P when applied to SPNs produce a slow depolarization with characteristics similar to those of slow EPSPs. While the slow EPSP is a common feature in rats and cats, the slow IPSP has been observed only in cat SPNs (Yoshimura et al. 1987b). The response is associated with decreased input resistance, reverses around -90 mV, and is increased and decreased in low- and high-Kg solution, respectively. Finally, the a2-antagonist yohimbine blocks the response, whereas noradrenaline mimics the slow hyperpolarizing response. Collectively, these findings suggest that noradrenaline acting on a2-receptors is responsible for generating the slow IPSP in cat SPNs (Yoshimura et al. 1987b).
Fast bPSBs Stimulation of dorsal roots elicits in about 30% of SPNs an IPSP with a mean half-decay time of 17 ms and an amplitude of 8 mV; in a portion of SPNs the IPSP is preceded by an EPSP (Dun and Mo 1989). The HPSP is made smaller by conditioning hyperpolarization and larger by conditioning depolarization, with a reversal potential of -70 mV; the response is reversed in a low-C1- solution but not changed significantly in a low-Kg solution (Fig. 4). Finally, the glycine antagonist strychnine, but not the GABA antagonist bicuculline, blocks the IPSP. Thus, it is concluded that the IPSP is caused by an increased membrane permeability to Cl- ions owing to glycine released from inhibitory interneurons (Dun and Mo 1989). In cat SPNs the occurrence of IPSPs appears to be infrequent and the response has yet to be characterized (Nishi et al. 1987). Interestingly, activation of ventral roots elicits in a small population of SPNs an IPSP with electrophysiologica1 characteristics similar to that evoked by dorsal root stimulation. The response is blocked by nicotinic antagonists d-tubocurarine and hexamethoniurn, in addition to strychnine. These observations raise the possibility that the IPSP evoked by stimulation of ventral roots is caused by the activation of (a) glycinecontaining interneuron(s) by axon collaterals via nicotinic action (Dun and Mo 1989). Intraspinal axon branchings have been noted in a small number of neonatal rat SPNs labelled with horseradish peroxidase, thus providing a potential morphological substrate for recurrent IPSPs (Forehand 1990).
Ventral root evoked EPSBs An interesting and novel response involves the transmission of ventral root afferents to neonatal SPNs. Stimulation of ventral roots evokes in about 48% of SPNs an EPSP. The response is graded with stimulus intensity and, when reaching the threshold, fires an action potential. The ventral root evoked EBSP occurs more often in SPNs when the resting membrane potential is high ( 1 - 60 mV). Under this condition. stimulation of ventral roots with low intensities evokes an EPSP rather than an antidromic spike; the latter can be effectively induced by depolarization of the membrane. The EPSP is reversibly abolished by low-Ca2+ solution and the falling phase in some SPNs is prolonged in Mg2+-freesolutions. The response is not affected by the nicotinic antagonists mecamylamine and dihydro-P-erythroidine, indicating that it is not due to the activation of axon collaterals. Lastly. the non-NMDA receptor antagonist CNQX eliminates the EPSP, while APV shortens the falling phase in some of the neurons. These observations collectively suggest that the EPSP evoked by stimulation of ventral roots is caused by activation of afferent fibers, resulting in a release of glutamate or a related amino acid. A similar observation has been reported with respect to the motoneurons (Jiang et al. 1991).
Slow synaptic potentials In addition to the fast EPSP and IPSP, stimulation s f the lateral fianiculus and, in a few instances, dorsal rootlets elicits in rat and cat SPNs a slow EPSP and in cat SPNs a slow IPSP with a time course of seconds to minutes (Yoshimura et al. 1987a, 1987b; Dun and Mo 1988). The slow EPSPs have the following general features: (i) the amplitude and duration are graded with stimulus intensity and summated with increasing number of pulses; (ii) the response is accompanied by an increase of membrane resistance, is increased with depolarization and decreased with hyperpolari-
The autonomic nervous system controls and coordinates virtually all visceral, vascular, and glandular activities, both tonically and in response to sudden environmental demands. Accordingly, autonomic outflow to various tissues is highly differentiated even under basal conditions and must be reset rapidly to meet the moment-to-moment demands of the srganism. Recent results from in vitro rat and cat spinal cord slices show that a range of synaptic mechanisms is available to SPNs, implicating a complex and variable control of signalling hnction. Altogether four types of synaptic responses have been identified in cat and rat SPNs. The first two types, the
Summary
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by NEW YORK UNIVERSITY on 11/12/14 For personal use only.
fast EPSP and fast IPSP, involve rapid excitation and inhibition, resulting from an increase of cation and anion permeability. In rat SPNs, glutamate or a closely related excitatory amino acid and glycine appear to be the excitatory and inhibitory transmitters for the fast EPSP and IPSP, respectively. The second type of synaptic mechanism involves excitation and inhibition of a relatively long time course as compared with that of fast synaptic events; the slow EPSP and slow IPSP are likely to be caused by a decrease and increase of membrane K permeability, respectively. The mediator for the slow synaptic events appears to be heterogenous. In the case of slow EPSPs, the pharmacological finding is consistent with the immnohistochemical observation that a variety of putative transmitters including amines and neuropeptides present in various afferents and descending fibers may serve as transmitters. The role of the slow IPSPs is not well established, as SPNs expressing this type of responses are small in number and may not even be present in the neonatal rat. In cat SPNs, noradrenaline acting on ar2-receptors is proposed to be the mediator. In summary, there exists in cat and rat SPNs a multitude of synaptic mechanisms via which inputs arising from sensory and supraspinal neurons can be temporally and correctly processed into orderly and purposeful outputs, thus ensuring the moment-by-rnoment regulation of individual organs. +
Acknowledgement This work was supported by National Institutes s f Health grant NS18710 from the Department of Health and Human Services. Cabot, J. B. 1990. Sympathetic preganglionic neurons: cytoarchitecture, ultrastmcture, and biophysical properties. In Central regulation of autonomic functions. Edited by A. D. Loewy and K. M. Spyer. Oxford University Press, New York. pp. 44 -67. Cannon, W. B. 1939. The wisdom of the body. 2nd ed. Norton, New York. Coote, J. H. 1988. The organisation of cardiovascular neurons in the spinal cord. Rev. Physiol. Biochem. Pharmacol. 110: 148-285. Coote, J. H., and Westbury, D. R. 1979. IntracelBuBar recordings from sympathetic preganglionic neurones. Neurosci . Lett. 15: 171- 175. Dembowsky, K., Czachurski, J., and Seller, H. 1985~.An intracellular study of the synaptic input to sympathetic preganglionic neurones of the third thoracic segment of the cat. J. Auton. Nerv. Syst. 13: 201-244. Dembowsky , K., Czachurski, J., and Seller, H. 19851s. Morphology of sympathetic preganglionic neurons in the thoracic spinal cord of the cat: an intracellular horseradish peroxidase study. J. Comp. Neurol. 238: 453 -465. Dembowsky, K., Czachurski, J., and Seller, H. 1986. Three types of sympathetic preganglionic neurones with different electrophysiological properties are identified by intracellular recordings in the cat. Pfluegers Arch. 406: 112- 126). Dun, N. J., and Miyazaki, T. 1988. NMDA-induced current in neonatal rat lateral horn cells. Soc. Neurosci. Abstr. 14: 318.16). Dun, N. J . , and Mo, N. 1988. In vitro effects of substance P on neonatal rat sympathetic preganglionic neurones. J. Physiol. (London), 399: 321-333. Dun, N. J., and Mo, N. 1989. Inhibitory postsynaptic potentials in neonatal rat sympathetic preganglionic neurones in vitro. J. Physisl. (London), 410: 267 -28 1. Forehand, C. J . 1990. Morphology of sympathetic preganglionic neurons in the neonatal rat spinal cord: an intracellular horseradish peroxidase study. J. Comp. Neurol. 298: 334 - 342. Gibson, S. J., and Polak, J. M. 1986. Neurochemistry of the spinal
cord. In Immunocytochemistry. Modern methods and application. 2nd ed. Edited by J. M. Polak and S. Van Norrdan. John Wright and Sons Ltd., Bristol. pp. 360-389. Jahnsen, H., and Llinas, R. 1984. Ionic basis for the electroresponsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro. J. Physiol. (London), 349: 227 -247. Jiang, Z. G., Shen, E., Wang, M. Y., and Dun, N. J. 1991. Excitatory postsynaptic potentials evoked by ventral root stimulation in neonate rat motoneurons in vitro. J. Neurophysisl. 65: 57-66. Johansssn, O., Hokfelt, T., Pernow, B., Jeffcoate, S. L., White, N., Steinbusch, H. W. M., Verhofstad, A. A. J., Emson, P. C., and Spindel, E. 1981. Immunohistochemical support for three putative transmitters in one neuron: coexistence of 5-hydroxytryptamine, substance P- and thyrotropin releasing hormone-like immunoreactivity in medullary neurons projecting to the spinal cord. Neuroscience, 6: 1857- 1881. Karczmar, A. G., Koketsu, K., and Nishi, S. 1986. Autonomic and enteric ganglia. Transmission and its pharmacology. Plenum Press, New York. Kmkoff, T. L., Ciriello, J., and Calaresu, F. R. 1985. Segmental distribution of peptide-like immunoreactivity in cell bodies of the thoracolumbar sympathetic nuclei of the cat. J. Comp. Neurol. 240: 90- 102. Llinas, R., and Sugimori, M. 1980. Electrophysiological properties of in vitro Wlrkinje cell somata in mammalian cerebellar slices. J . Physiol. (London), 305: 171- 195. Loewy , A. D. 1990. Central autonomic pathways. In Central regulation of autonomic functions. Edited by A. B . Loewy and K. M . Spyer. Oxford University Press, New York. pp. 88- 103. Mayer, M. L., and Westbrook, G. L. 1987. The physiology of excitatory amino acids in the vertebrate central nervous system. Prog . Neurobiol. (Oxford), 28: 197 - 276. McLachBan, E. M., and Hirst, D. G. S. 1986). Some properties of preganglionic neurones in the upper thoracic spinal cord of the cat. J. Neurophysiol. 43: 1251 - 1265. Miyazaki, T., Coote, J. H., and Dun, N. J. 1989. Excitatory and inhibitory effects of epinephrine on neonatal rat sympathetic preganglionic neurons in vitro. Brain Res. 407: 108- 116. Nishi, S., Yoshimura, M., and Polosa, C. 1987. Synaptic potentials and putative transmitter actions in sympathetic preganglionic neurons. In Organizationof the autonomic nervous system: central and peripheral mechanisms. Edited by J. Ciriello, F. R. Calaresu, L. P. Renaud, and C. Polosa. Alan R. Liss, Inc., New York. pp. 15-26. Petras, J. M., and Cummings, J. F. 1972. Autonomic neurons in the spinal cord of the rhesus monkey. A correlation of the findings of cytoarchitectonicsand sympathectomy with fiber degeneration following dorsal rhizotomy. J. Comp Neurol. 146: 189-218. Shen, E., and Dun, N. J. 1990. Neonate rat sympathetic preganglionic neurons intracellularly labelled with lucifer yellow in thin spinal cord slices. J. Auton. New. Syst. 29: 247-254. Shen, E., Mo, N., and Dun, N. J. 1990. APV-sensitive dorsal root afferent transmission to neonate rat sympathetic preganglionic neurons in vitro. J. Neurophysiol. 64: 991 -999. Spanswick, D., and Logan, S. D. 1 9 9 0 ~Sympathetic . preganglionic neurones in neonatal rat spinal cord in vitro: eBectrophysiologica1 characteristics and the effects of selective excitatory amino acid receptor agonists. Brain Res. 525: 181 - 188. Spanswick, D., and Logan, S. B. 1990b. Spontaneous rhythmic activity in the intermediolateral cell nucleus of the neonate rat thoracolumbar spinal cord in vitro. Neuroscience, 39: 395 -403. Yoshimura, M., Polosa, C., and Nishi, S. 1986. After-hyperpolarization mechanisms in cat sympathetic preganglionic neuron in vitro. J. Neurophysiol. 55: 1234- 1246. Yoshimura, M., Polosa, C., and Nishi, S. 1987a. Slow EPSP and the depolarizing action of noradrenaline on sympathetic preganglionic neurons. Brain Res, 414: 138- 142. Yoshimura, M., Polosa, C., and Nishi, S. 1987b. Slow IPSP and the noradrenaline-induced inhibition of the cat sympathetic preganglionic neuron in vitro. Brain Res. 419: 383-386.
