Neuroscience Letters, 130 (1991)237-242 © 1991 ElsevierScientificPublishers Ireland Ltd. 0304-3940/91/$03.50 ADONIS 030439409100520H
237
NSL 08039
Whole-cell recordings from sympathetic preganglionic neurons in rat spinal cord slices A.E. Pickering, D. S p a n s w i c k a n d S.D. L o g a n Department of Physiology, Medical School, Universityof Birmingham, Birmingham (U.K.)
(Received26 April 1991;Revisedversion received6 May 1991;Accepted6 June 1991) Key words: Sympatheticpreganglionicneuron; Whole-cellpatch-clamp; Spinal cord slice;Anomalous rectification;Transient rectification;Rat
Whole-cellpatch-clamp recordings(WCR) were made from sympatheticpreganglionicneurons (SPN) in neonate rat spinal cord slices. SPN were identifiedhistologicallyby fillingthem with the fluorescentdye LuciferYellowcontained within the patch pipette solution. Current clamp recordings were obtained from SPN with a potassium based pipette solUtion. The cells exhibited many of the characteristicproperties of SPN seen previously with intracellular recordingsin both the rat and the eat. However,we found an order of magnitude increase in both cell input resistance (950 M~2) and time constant (I 18 ms) over those seen with conventional recordings. We believethese values approximate better the situation in intact cells, and will have a vital bearing upon how SPN integrate inputs. We conclude that WCR in spinal cord slicesprovides a powerful tool for investigating the cellularproperties of SPN.
Autonomic preganglionic neurons along with motoneurons represent the final common pathway for output from the central nervous system to the periphery. The intrinsic cellular properties o f these neurons are therefore vital in determining the way in which activity within the CNS is communicated to peripheral organs. We have studied the electrophysiology and morphology of sympathetic preganglionic neurons in vitro using whole-cell patch-clamp recording from brain slices [3, 9]. In the rat these neurons are predominantly located in the intermediolateral cell column (IML) of the thoracolumbar spinal cord [27]. Some o f this data has appeared previously in abstract form [17, 18]. All experiments were performed upon transverse thoracolumbar spinal cord slices (350-500 /tm) prepared from neonate Sprague-Dawley rats aged 7-14 days, as described previously [24]. Slices were maintained at room temperature in a recording chamber and superfused with an artificial cerebrospinal fluid (aCSF) at a rate of 2-5 ml/min. The aCSF had the following composition (raM): NaCI 127, KCI 1.9, KH2PO 4 1.2, CaCI 2 2.4, MgCI2 1.3, N a H C O 3 26, D-glucose 10; saturated with carbogen (95% O2/5~ CO2). Whole-cell recordings were obtained from neurons in the lateral horn using patch pipettes of resistance 3-7 MI2. The pipettes were filled with the following solution Correspondence: A.E. Pickering, Department of Physiology,Medical School, Universityof Birmingham, Birmingham,U,K.
(mM): potassium gluconate 130, KC1 10, MgCI 2 2, CaCI 2 1, E G T A - N a 11, HEPES 10, Na2ATP 2; pH 7.4 with NaOH. Recordings were made using a patch-clamp amplifier (EPC-7, List Electronic, Germany) and stored on videotape for later analysis (via a Sony 701es PCM modified after [14]). To estimate the time constant of the cell exponential fits to charging curves were made using a least squares error minimisation algorithm (Systat, Systat Inc., U.S.A., running on an Apple Mac Ilfx). For subsequent epifluorescent examination of cell morphology 1 mg/ml Lucifer Yellow [25] was added to the pipette solution. Slices were fixed in 0.1 M phosphate-buffered saline with 10% formaldehyde before being cleared in dimethyl sulfoxide and viewed wholemount [12]. Thirty-eight cells .were identified histologically as being SPN (see Fig. 1) using the following criteria: 1. Location o f the soma in the I M L or in the lateral funiculus. 2. An axon projecting towards the ventral horn. 3. Presence of characteristic medial dendrites. Additionally, 5 cells were further identified antidromically as SPN by stimulating with a bipolar electrode at the ventral root exit site, as has been described by us previously [24]. SPN had either ovoid, spherical or multilobed somata usually with their long axis orientated in a mediolateral direction. The somata had mean dimensions ( + S.E.M.) 26.5+1.0 /~mx18.5__+0.9 /tm. The cell nucleus was
238
Fig. 1. Photo-montage showing a Lucifer Yellow-filled sympathetic preganglionic neuron. This cell shows a morphology typical of SPN seen in this study. The inset shows the transverse spinal cord slice with the cell soma located in the lateral horn. DH, dorsal horn; LF, lateral funiculus; CA, central autonomic region. Bar-- 100/am. The cell can be seen more clearly when the slice is not transilluminated. Bar= 100 jam. The soma is ovoid with dimensions 32 × 20/am, The axon (arrowed) is seen to leave the soma from its ventral edge and travels medially briefly before heading out of the ventral horn. The cell sends two primary dendrites (over 500/am) to the central autonomic region where they ramify to form a plexus over both sides of the cord (not shown). Notice also the dendrite projecting to the lateral funiculus.
visible in several of the less intensely stained somata, it was usually ovoid and filled approximately 2/3 of the soma. Axons left from either a medial primary dendrite or from the soma and projected towards the ventral root. The axon was seen to leave the ventral horn in about 30% of cells, most axons being truncated at the cut surface of the slice. No axon collaterals were seen in this study in agreement with previous findings for SPN located in the lateral horn [7, 11, 22]. The majority of the cells showed a mediolateral dendritic organisation. This was manifest most obviously as a very large medial dendritic projection seen intact in many SPN. Typically one or two primary dendrites extended through Rexed's lamina VII to end just dorsal and dorsolateral to the central canal, in the ipsilateral or contralateral central autonomic (CA) region. Such dendrites were observed to reach the contralateral lamina VII in three cells. The dendrites branched infrequently, usually some distance from the cell soma, en route to the
CA, however, profuse branching was often observed within the CA forming a fine dendritic plexus. Most of the cells (34 of 38) also had dendrites which projected to the lateral funiculus and were often seen to loop back towards the cell soma. Major rostrocaudal dendritic projections were seen in 5 cells. In these neurons the dendrites were truncated at the edge of the slice. The morphology described here is in close agreement with two previous dye injection studies upon neonate rat SPN in vitro [11, 22]. It contrasts with the strictly rostrocaudal organisation of dendrites documented in a study in adult cats in vivo [7]. In a retrograde labelling study, Bacon and Smith clearly labelled dendrites in SPN projecting to the adrenal gland in adult rats [l]. The cells seen in our study correspond to their types B and C, which had a predominantly mediolateral organisation. The SPN had a mean resting potential of - 5 2 + 1.6 mV (n = 32, range - 40 to - 82 mV). They typically exhibited on-going synaptic activity that resulted in the dis-
239
B
A
20 mV
5
\
' '
I
25mV
ms
J
SO0 ms
100 ms
I
1
100 pA
500 ms
C
1:/t .~o
.~o
/I 820 l~b'4 240 k~4
39
e/1 1 3'' i i i i
/
i / s/ s /s
~
.
i
10
J
145, 9
I
/ eI
i
230.
.
~
.:;o ,' f
i
~ (rm)
(pA)
Current
---- -
0
,
. olt t
.~
J
-30 ~
t'
>
>
42
-50
-60
28 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
"rime (s)
Fig. 2. Membrane properties of a typical sympathetic preganglionic neuron revealed in response to current pulses. The cell had a resting potential of - 5 8 mV. A: voltage responses to a series of injected current pulses (+30 to -140 pA x 1.1 s). B: spikes evoked by current injection on two different timebases. The upper trace shows the spike to have a characteristic inflexion in the downstroke of repolarization (threshold = - 3 8 mV, amplitude = 59 mV, overshoot = 21 mV, duration = 8.1 ms). The lower trace illustrates the large afterhyperpolarization following the spike (amplitude = 16 mV, duration =250 ms). C: current-voltage relationship. The input resistance was calculated to be 820 Ms'2 over the linear range of the relationship close to resting (+ 10 to - 2 5 mV). It is clear that with large hyperpolarizations the input resistance falls, to a value of about 240 MI2. D: exponential fits to the voltage responses. The responses to large pulses were fitted well with single exponentials, however the smaller responses were significantly better approximated by two exponentials. The passive time constant (3) close to rest was estimated to be 230 ms, this is observed to fall by up to 90% with large hyperpolarizations.
charge o f action potentials, usually at rates below 1 H z . This synaptic activity could be largely blocked in the presence o f t e t r o d o t o x i n ( T T X , 0.2 p M ) , h o w e v e r spontaneous miniature post-synaptic potentials were still distinguishable. In cells with higher input resistances steplike fluctuations in m e m b r a n e potential remained in the presence o f T T X , these appeared to be due to single channel activity. Six o f the cells s h o w e d s p o n t a n e o u s rhythmic oscillations in m e m b r a n e potential which were
very similar in nature to the population o f oscillating cells seen previously in the lateral horn [23]. T o e x a m i n e the cellular m e m b r a n e properties a series o f positive and negative current pulses were injected into the cells (Fig. 2A). The input resistance o f the cells at rest was calculated f r o m the linear portion o f the currentvoltage relationship close to rest (Fig. 2C). S P N had a m e a n input resistance o f 950___98 Mr2 ( n = 3 1 ) with a range o f 3 4 0 - 2 8 0 0 MI2. These are an order o f magnitude
240
ls
120mV
600ms [20mV
75pA ls
600ms
Fig. 3. Transient rectification in a sympathetic preganglionic neuron. The cell had a resting potential of -43 mV. A: this figure shows the delayed recovery to rest, upper traces, following the end of hyperpolarizingcurrent pulses of different amplitudes ( - 10 to - 100 p A x I. 1 s, lower traces). The return to rest becomes delayed following the end of hyperpolarizationsgreater than 15 mV and is maximal from about 50 mV hyperpolarized. There is a clear inflexion marking the end of the passive decay towards rest and the activation of the rectification. This cell also demonstrates pronounced anomalous rectification. B: this figure shows the response of the same neuron, upper traces, to a seriesof variable duration, constant amplitude current pulses (I00 p A x 10 ms to 1.05 s, lower traces). This shows the return to rest taking much longer than the initial hyperpolarization, indicating the activation of the transient rectification. Note that for maximal activation the pulse must last for at least 400 ms.
larger than previously reported values; in the cat in vivo of 12-50 MI2 [16, 8] and in vitro of 68 Mr2 [33] and in the neonate rat in vitro of 106 M~2 [15] and 110 MI2 [24]. As this high input resistance was seen immediately following the establishment of W C R it seems unlikely that it was caused by the loss of some vital intracellular constituents. We saw similar resistances when using the nystatin perforated patch technique [13] which largely preserves the intracellular constituents (unpublished observations). It seems probable that the increased input resistance was largely due to the improved seal between recording electrode and the cell membrane. Similar increases in input resistances have been seen when using W C R in other cells including chromaffin cells (500 Mr2 to 5 GO) [10], salamander amacrine cells (100 MI2 to > 2 GO) [2], turtle visual cortex (80 MI2 to 314 MI2) [3], hippocampal granule cells (48 MI2 to 1.1 GO) [9], motoneurons (34 Mr2 to 290 MI2) [28] and salamander olfactory receptors (200 MI2 to 5 GI2) [19]. The time constant of the ceils was estimated by fitting exponentials to the membrane voltage charging curves from small negative current pulses (Fig. 2D). The data was often significantly better fitted by the sum of two or three exponentials. In these cases the longest exponential was used as the passive membrane time constant (as sug-
gested by Rail [20]). The SPN had an average time constant 118 +__14 ms (n = 28) with a range of 31-360 ms. The time constant of SPN has been previously estimated in adult cats as being in the range 10-25 ms in vivo [8] and 7-20 ms in vitro [33]. We believe that the increase in time constant seen here is also due to the better cell-electrode seal. The presence o f multiple exponentials indicates that these neurons are not well modelled as a single isopotential c o m p a r t m e n t [20]. This is consistent with the extensive dendritic arbor possessed by these cells (see Fig. 1). Most of these cells showed a fall in input resistance with hyperpolarizations greater than 20 mV from rest (25 of 28 SPN). The fall in input resistance was evident from the current-voltage plots from these cells (Fig. 2C). The fall was shown even more clearly by the reduction in the time constant of large hyperpolarizing responses (Fig. 2D). This data shows that this rectification could reduce the input resistance by up to 80%. This rectification was very rapid in onset and no decay was observed. A similar rectification has previously been seen in cat SPN by D e m b o w s k y et al. [8]. In other cells it has been termed an instantaneous anomalous rectification, m a n y vertebrate neurons possess this conductance including those of the olfactory cortex [5], inferior olive [30] and substantia gelatinosa [31].
241 All of the cells recorded from in this study fired action potentials in response to depolarizing currents (Fig. 2A, B), they were also seen to fire spontaneously or in response to orthodromically evoked synaptic potentials and antidromic activation. The spike characteristics were measured from the threshold for firing (mean of - 3 5 + 1.3 mV). The spikes had a mean amplitude 6 6 _ 1.8 mV (n = 32) and overshot zero by an average of 30 mV. Such large overshoots have been documented in vivo in the cat where a range of 12-50 mV was seen [8]. The spikes had durations (time spent above threshold) that averaged 8.8___0.4 ms (n=32, range 4.0-14.5 ms). Spikes characteristically exhibited an inflection in the downstroke of repolarization, suggestive of the presence o f a calcium component (Fig. 2B). Such shoulders have been documented in the cat both in vivo [8] and in vitro [33] and the rat in vitro [24]. In all cells to which T T X (0.2-1 a M ) was applied the spike was blocked reversibly suggesting that it is sodium dependent. All SPN exhibited an after-hyperpolarization (AHP) following the discharge of action potentials. The A H P had a mean amplitude (measured from threshold) of 2 4 _ 0.8 mV (n = 32, 14-33 mV) and an average duration of 400 ms (n=32, range 120-1000 ms). They are of similar duration but are larger in amplitude than those reported previously in the rat in vitro [15, 24], cat in vivo [8] and in vitro [33]. The spikes recorded in this study have longer durations and larger AHPs than those seen previously using intracellular recordings. Preliminary experiments have indicated that increasing the bath temperature to 36°C during a recording shortens the spike duration and decreases the size of the AHP. These observations are consistent with a previous report o f temperature effects upon CA1 hippocampal neurons [29], where decreased temperature lengthened spike durations and slowed calcium buffering and hence increased A H P amplitudes. A transient outward rectification was seen in 13 of 16 SPN (see Fig. 3). It took the form o f a delayed recovery to rest following hyperpolarizing pulses o f amplitude greater than -,, 15 mV (Figs. 2A and 3A). This recovery could take as long as 2 s following the end o f the pulse, considerably longer than might be expected from the passive time constant of the cells (Fig. 3B). A fast component of this rectification could be blocked reversibly by 2-4 m M 4-aminopyridine (4-AP) in the presence of 1 a M TTX, but was not significantly altered by lower concentrations of 4-AP (30-100 aM). This data suggests this rectification is due in part to the activation of an Acurrent [4, 21] rather than a D-current [26]. 4-AP also increased spike duration, shortened the A H P and increased the spontaneous synaptic activity onto the cells in the absence o f TTX. This transient rectification is
similar to that described previously in cat SPN both in vivo and in vitro [8, 32]. Using the WCR technique we have been able to obtain high resolution, stable intracellular recordings from rat SPN in vitro. These neurons have shown many o f the characteristic properties previously reported for both rat and cat SPN when using intracellular recording [6, 33], however we have seen that when using W C R the input resistances and time constants of these cells are considerably larger. These increased values indicate that SPN are more electrotonically compact [20] than previous conventional intracellular studies have suggested [8, 15, 16, 24, 33]. This may be vital in allowing SPN to integrate effectively inputs arising from such an extensive dendritic tree. We would like to express our gratitude to Mr. I.C. Gibson for invaluable discussions, to Mr. H.F. Ross for sharing his computer expertise with us and to Sue Ethel for photographic assistance. A.E.P. is in receipt of a Wellcome Trust prize studentship. This work is supported by the M R C and the British Heart Foundation.
1 Bacon, S.J. and Smith, A.D., Preganglionicsympathetic neurons innervatingthe rat adrenal-medulla:immunocytochemicalevidence of synaptic input from nerve-terminals containing substance-P, GABA or 5-hydroxytryptamine,J. Auton. Nerv. Syst., 24 (1988) 97-122. 2 Barnes,S. and Werblin, F., Gated currents generatesingle spike activity in amacrine cells of the tiger salamander retina, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 1509-1512. 3 Blanton, M.G., LoTurco, J.J. and Kriegstein, A.R., Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex, J. Neurosci.Methods, 30 (1989)203-210. 4 Connor, J.A. and Stevens, C.F., Prediction of repetitive firing behaviour from voltage clamp data on an isolated neurone soma, J. Physiol., 213 (1971) 31-53. 5 Constanti, A. and Galvan, M., Fast inward-rectifyingcurrent accounts for anomalous rectificationin olfactory cortex neurones, J. Physiol., 385 (1983) 153-178. 6 Coote, J.H., The organizationof cardiovascularneurons in the spinal cord, Rev. Physiol.Biochem.Pharmacol., 110 (1988) 147-285. 7 Dembowsky, K., Czachurski, J. and Seller, H., Morphology of sympathetic preganglionic neurons in the thoracic spinal cord of the cat: An intracellular horseradish peroxidase study, J. Comp. Neurol., 238 (1985)453--465. 8 Dembowsky,K., Czachurski, J. and Seller,H., Three types of sympathetic preganglionicneurones with differentelectrophysiological properties are identified by intracellular recordings in the cat, PflfigersArch., 406(2)(1986) 112-20. 9 Edwards, F.A., Konnerth, A., Sakmann, B. and Takahashi, T., A thin slicepreparation for patch clamp recordingsfrom neurones of the mammalian central nervous system,PfliigersArch., 414 (1989) 600-612. 10 Fenwick, E.M., Marty, A. and Neher, E., A patch-damp study of bovine chromaffinceils and of their sensitivityto acetylcholine,J. Physiol., 331 (1982)577-597.
242 11 Forehand, C.J., Morphology of sympathetic preganglionic neurons in the neonatal rat spinal-cord: an intracellular horseradish peroxidase study, J. Comp. Neurol., 298(3) (1990) 334-342. 12 Grace, A.A. and Llinas, R., Morphological artifacts induced in intracellularly stained neurons by dehydration: circumvention using rapid dimethyl sulfoxide clearing, Neuroscience, 16 (1985) 461-475. 13 Horn, R. and Marty, A., Muscarinic activation of ionic currents measured by a new whole-cell recording method, J. Gen. Physiol., 92 (1988) 145-159. 14 Lamb, T.D., An inexpensive digital tape recorder suitable for neurophysiological signals, J. Neurosci. Methods, l 5 (1985) 1-13. 15 Ma, R.C. and Dun, N.J., Vasopressin depolarizes lateral horn cells of the neonatal rat spinal cord in vitro, Brain Res., 348 (1985) 3643. 16 McLachlan, E.M. and Hirst, G.D.S., Some properties of preganglionic neurons in upper thoracic spinal cord of the cat, J. Neurophysiol., 43 (1980) 1251 1265. 17 Pickering, A.E., Spanswick, D. and Logan, S.D., Rectification in identified rat sympathetic preganglionic neurones in vitro, J. Physiol. (in press), Royal Free meeting, February, 1991. 18 Pickering, A.E., Spanswick, D. and Logan, S.D., Whole-cell recording from lateral horn cells in the neonate rat spinal cord in vitro, J. Physiol., 435 (1991) 36P. 19 Pongracz, F., Firestein, S. and Shepherd, G.M., Electrotonic structure of olfactory sensory neurons analyzed by intracellular and whole cell patch techniques, J. Neurophysiol., 65(3) (1991) 747 758. 20 Rail, W., Time constants and electrotonic length of membrane cylinders and neurons, Biophys. J., 9 (1969) 1483-1508. 21 Rogawski, M.A., The A-current: how ubiquitous a feature of excitable cells is it?, Trends Neurosci., 8 (1985) 214-219. 22 Shen, E. and Dun, N.J., Neonate rat sympathetic preganglionic neurons intracellularly labeled with Lucifer Yellow in thin spinal cord slices, J. Auton. Nerv. Syst., 29(3) (1990) 247-254.
23 Spanswick, D. and Logan, S.D., Spontaneous rhythmic activity in the intermediolateral cell nucleus of the neonate rat thoracolumbar spinal cord in vitro, Neuroscience, 39(2) (1990) 395-403. 24 Spanswick, D. and Logan, S.D., Sympathetic preganglionic neurons in neonatal rat spinal-cord in vitro: electrophysiological characteristics and the effects of selective excitatory amino-acid receptor agonists, Brain Res., 525 (1990) 181-188. 25 Stewart, W.W., Intracellular marking of neurons with a highly fluorescent napthalimide dye, Cell, 14 (1978) 741-759. 26 Storm, J.F., Temporal integration by a slowly inactivating K + current in hippocampal neurons, Nature, 336 (1988) 379-381. 27 Strack, A.M., Sawyer, W.B., Marubio, L.M. and Loewy, A.D., Spinal origin of sympathetic preganglionic neurons in the rat, Brain Res., 455 (1988) 187-191. 28 Takahashi, T., Membrane currents in visually identified motoneurones of neonatal rat spinal cord, J. Physiol., 423 (1990) 27-46. 29 Thompson, S.M., Masukawa, L.M. and Prince, D.A., Temperature dependence of intrinsic membrane properties and synaptic potentials in hippocampal CA1 neurons in vitro, J. Neurosci., 5 (1985) 817-824. 30 Yarom, Y. and Llinas, R., Long-term modifiability of anomalous and delayed rectification in guinea pig inferior olivary neurons, J. Neurosci., 7 (1987) 1166-1177. 31 Yoshimura, M. and Jessell, T.M., Membrane properties of rat substantia gelatinosa neurons in vitro, J. Neurophysiol., 62(1) (1989) 109-118. 32 Yoshimura, M., Polosa, C. and Nishi, S., A transient outward rectification in the cat sympathetic preganglionic neuron, Pfliigers Arch., 408 (1987) 207-208. 33 Yoshimura, M., Polosa, C. and Nishi, S., Electrophysiological properties of sympathetic preganglionic neurons in the cat spinal cord in vitro, Pfliigers Arch., 406(2) (1986) 91-8.
237
NSL 08039
Whole-cell recordings from sympathetic preganglionic neurons in rat spinal cord slices A.E. Pickering, D. S p a n s w i c k a n d S.D. L o g a n Department of Physiology, Medical School, Universityof Birmingham, Birmingham (U.K.)
(Received26 April 1991;Revisedversion received6 May 1991;Accepted6 June 1991) Key words: Sympatheticpreganglionicneuron; Whole-cellpatch-clamp; Spinal cord slice;Anomalous rectification;Transient rectification;Rat
Whole-cellpatch-clamp recordings(WCR) were made from sympatheticpreganglionicneurons (SPN) in neonate rat spinal cord slices. SPN were identifiedhistologicallyby fillingthem with the fluorescentdye LuciferYellowcontained within the patch pipette solution. Current clamp recordings were obtained from SPN with a potassium based pipette solUtion. The cells exhibited many of the characteristicproperties of SPN seen previously with intracellular recordingsin both the rat and the eat. However,we found an order of magnitude increase in both cell input resistance (950 M~2) and time constant (I 18 ms) over those seen with conventional recordings. We believethese values approximate better the situation in intact cells, and will have a vital bearing upon how SPN integrate inputs. We conclude that WCR in spinal cord slicesprovides a powerful tool for investigating the cellularproperties of SPN.
Autonomic preganglionic neurons along with motoneurons represent the final common pathway for output from the central nervous system to the periphery. The intrinsic cellular properties o f these neurons are therefore vital in determining the way in which activity within the CNS is communicated to peripheral organs. We have studied the electrophysiology and morphology of sympathetic preganglionic neurons in vitro using whole-cell patch-clamp recording from brain slices [3, 9]. In the rat these neurons are predominantly located in the intermediolateral cell column (IML) of the thoracolumbar spinal cord [27]. Some o f this data has appeared previously in abstract form [17, 18]. All experiments were performed upon transverse thoracolumbar spinal cord slices (350-500 /tm) prepared from neonate Sprague-Dawley rats aged 7-14 days, as described previously [24]. Slices were maintained at room temperature in a recording chamber and superfused with an artificial cerebrospinal fluid (aCSF) at a rate of 2-5 ml/min. The aCSF had the following composition (raM): NaCI 127, KCI 1.9, KH2PO 4 1.2, CaCI 2 2.4, MgCI2 1.3, N a H C O 3 26, D-glucose 10; saturated with carbogen (95% O2/5~ CO2). Whole-cell recordings were obtained from neurons in the lateral horn using patch pipettes of resistance 3-7 MI2. The pipettes were filled with the following solution Correspondence: A.E. Pickering, Department of Physiology,Medical School, Universityof Birmingham, Birmingham,U,K.
(mM): potassium gluconate 130, KC1 10, MgCI 2 2, CaCI 2 1, E G T A - N a 11, HEPES 10, Na2ATP 2; pH 7.4 with NaOH. Recordings were made using a patch-clamp amplifier (EPC-7, List Electronic, Germany) and stored on videotape for later analysis (via a Sony 701es PCM modified after [14]). To estimate the time constant of the cell exponential fits to charging curves were made using a least squares error minimisation algorithm (Systat, Systat Inc., U.S.A., running on an Apple Mac Ilfx). For subsequent epifluorescent examination of cell morphology 1 mg/ml Lucifer Yellow [25] was added to the pipette solution. Slices were fixed in 0.1 M phosphate-buffered saline with 10% formaldehyde before being cleared in dimethyl sulfoxide and viewed wholemount [12]. Thirty-eight cells .were identified histologically as being SPN (see Fig. 1) using the following criteria: 1. Location o f the soma in the I M L or in the lateral funiculus. 2. An axon projecting towards the ventral horn. 3. Presence of characteristic medial dendrites. Additionally, 5 cells were further identified antidromically as SPN by stimulating with a bipolar electrode at the ventral root exit site, as has been described by us previously [24]. SPN had either ovoid, spherical or multilobed somata usually with their long axis orientated in a mediolateral direction. The somata had mean dimensions ( + S.E.M.) 26.5+1.0 /~mx18.5__+0.9 /tm. The cell nucleus was
238
Fig. 1. Photo-montage showing a Lucifer Yellow-filled sympathetic preganglionic neuron. This cell shows a morphology typical of SPN seen in this study. The inset shows the transverse spinal cord slice with the cell soma located in the lateral horn. DH, dorsal horn; LF, lateral funiculus; CA, central autonomic region. Bar-- 100/am. The cell can be seen more clearly when the slice is not transilluminated. Bar= 100 jam. The soma is ovoid with dimensions 32 × 20/am, The axon (arrowed) is seen to leave the soma from its ventral edge and travels medially briefly before heading out of the ventral horn. The cell sends two primary dendrites (over 500/am) to the central autonomic region where they ramify to form a plexus over both sides of the cord (not shown). Notice also the dendrite projecting to the lateral funiculus.
visible in several of the less intensely stained somata, it was usually ovoid and filled approximately 2/3 of the soma. Axons left from either a medial primary dendrite or from the soma and projected towards the ventral root. The axon was seen to leave the ventral horn in about 30% of cells, most axons being truncated at the cut surface of the slice. No axon collaterals were seen in this study in agreement with previous findings for SPN located in the lateral horn [7, 11, 22]. The majority of the cells showed a mediolateral dendritic organisation. This was manifest most obviously as a very large medial dendritic projection seen intact in many SPN. Typically one or two primary dendrites extended through Rexed's lamina VII to end just dorsal and dorsolateral to the central canal, in the ipsilateral or contralateral central autonomic (CA) region. Such dendrites were observed to reach the contralateral lamina VII in three cells. The dendrites branched infrequently, usually some distance from the cell soma, en route to the
CA, however, profuse branching was often observed within the CA forming a fine dendritic plexus. Most of the cells (34 of 38) also had dendrites which projected to the lateral funiculus and were often seen to loop back towards the cell soma. Major rostrocaudal dendritic projections were seen in 5 cells. In these neurons the dendrites were truncated at the edge of the slice. The morphology described here is in close agreement with two previous dye injection studies upon neonate rat SPN in vitro [11, 22]. It contrasts with the strictly rostrocaudal organisation of dendrites documented in a study in adult cats in vivo [7]. In a retrograde labelling study, Bacon and Smith clearly labelled dendrites in SPN projecting to the adrenal gland in adult rats [l]. The cells seen in our study correspond to their types B and C, which had a predominantly mediolateral organisation. The SPN had a mean resting potential of - 5 2 + 1.6 mV (n = 32, range - 40 to - 82 mV). They typically exhibited on-going synaptic activity that resulted in the dis-
239
B
A
20 mV
5
\
' '
I
25mV
ms
J
SO0 ms
100 ms
I
1
100 pA
500 ms
C
1:/t .~o
.~o
/I 820 l~b'4 240 k~4
39
e/1 1 3'' i i i i
/
i / s/ s /s
~
.
i
10
J
145, 9
I
/ eI
i
230.
.
~
.:;o ,' f
i
~ (rm)
(pA)
Current
---- -
0
,
. olt t
.~
J
-30 ~
t'
>
>
42
-50
-60
28 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
"rime (s)
Fig. 2. Membrane properties of a typical sympathetic preganglionic neuron revealed in response to current pulses. The cell had a resting potential of - 5 8 mV. A: voltage responses to a series of injected current pulses (+30 to -140 pA x 1.1 s). B: spikes evoked by current injection on two different timebases. The upper trace shows the spike to have a characteristic inflexion in the downstroke of repolarization (threshold = - 3 8 mV, amplitude = 59 mV, overshoot = 21 mV, duration = 8.1 ms). The lower trace illustrates the large afterhyperpolarization following the spike (amplitude = 16 mV, duration =250 ms). C: current-voltage relationship. The input resistance was calculated to be 820 Ms'2 over the linear range of the relationship close to resting (+ 10 to - 2 5 mV). It is clear that with large hyperpolarizations the input resistance falls, to a value of about 240 MI2. D: exponential fits to the voltage responses. The responses to large pulses were fitted well with single exponentials, however the smaller responses were significantly better approximated by two exponentials. The passive time constant (3) close to rest was estimated to be 230 ms, this is observed to fall by up to 90% with large hyperpolarizations.
charge o f action potentials, usually at rates below 1 H z . This synaptic activity could be largely blocked in the presence o f t e t r o d o t o x i n ( T T X , 0.2 p M ) , h o w e v e r spontaneous miniature post-synaptic potentials were still distinguishable. In cells with higher input resistances steplike fluctuations in m e m b r a n e potential remained in the presence o f T T X , these appeared to be due to single channel activity. Six o f the cells s h o w e d s p o n t a n e o u s rhythmic oscillations in m e m b r a n e potential which were
very similar in nature to the population o f oscillating cells seen previously in the lateral horn [23]. T o e x a m i n e the cellular m e m b r a n e properties a series o f positive and negative current pulses were injected into the cells (Fig. 2A). The input resistance o f the cells at rest was calculated f r o m the linear portion o f the currentvoltage relationship close to rest (Fig. 2C). S P N had a m e a n input resistance o f 950___98 Mr2 ( n = 3 1 ) with a range o f 3 4 0 - 2 8 0 0 MI2. These are an order o f magnitude
240
ls
120mV
600ms [20mV
75pA ls
600ms
Fig. 3. Transient rectification in a sympathetic preganglionic neuron. The cell had a resting potential of -43 mV. A: this figure shows the delayed recovery to rest, upper traces, following the end of hyperpolarizingcurrent pulses of different amplitudes ( - 10 to - 100 p A x I. 1 s, lower traces). The return to rest becomes delayed following the end of hyperpolarizationsgreater than 15 mV and is maximal from about 50 mV hyperpolarized. There is a clear inflexion marking the end of the passive decay towards rest and the activation of the rectification. This cell also demonstrates pronounced anomalous rectification. B: this figure shows the response of the same neuron, upper traces, to a seriesof variable duration, constant amplitude current pulses (I00 p A x 10 ms to 1.05 s, lower traces). This shows the return to rest taking much longer than the initial hyperpolarization, indicating the activation of the transient rectification. Note that for maximal activation the pulse must last for at least 400 ms.
larger than previously reported values; in the cat in vivo of 12-50 MI2 [16, 8] and in vitro of 68 Mr2 [33] and in the neonate rat in vitro of 106 M~2 [15] and 110 MI2 [24]. As this high input resistance was seen immediately following the establishment of W C R it seems unlikely that it was caused by the loss of some vital intracellular constituents. We saw similar resistances when using the nystatin perforated patch technique [13] which largely preserves the intracellular constituents (unpublished observations). It seems probable that the increased input resistance was largely due to the improved seal between recording electrode and the cell membrane. Similar increases in input resistances have been seen when using W C R in other cells including chromaffin cells (500 Mr2 to 5 GO) [10], salamander amacrine cells (100 MI2 to > 2 GO) [2], turtle visual cortex (80 MI2 to 314 MI2) [3], hippocampal granule cells (48 MI2 to 1.1 GO) [9], motoneurons (34 Mr2 to 290 MI2) [28] and salamander olfactory receptors (200 MI2 to 5 GI2) [19]. The time constant of the ceils was estimated by fitting exponentials to the membrane voltage charging curves from small negative current pulses (Fig. 2D). The data was often significantly better fitted by the sum of two or three exponentials. In these cases the longest exponential was used as the passive membrane time constant (as sug-
gested by Rail [20]). The SPN had an average time constant 118 +__14 ms (n = 28) with a range of 31-360 ms. The time constant of SPN has been previously estimated in adult cats as being in the range 10-25 ms in vivo [8] and 7-20 ms in vitro [33]. We believe that the increase in time constant seen here is also due to the better cell-electrode seal. The presence o f multiple exponentials indicates that these neurons are not well modelled as a single isopotential c o m p a r t m e n t [20]. This is consistent with the extensive dendritic arbor possessed by these cells (see Fig. 1). Most of these cells showed a fall in input resistance with hyperpolarizations greater than 20 mV from rest (25 of 28 SPN). The fall in input resistance was evident from the current-voltage plots from these cells (Fig. 2C). The fall was shown even more clearly by the reduction in the time constant of large hyperpolarizing responses (Fig. 2D). This data shows that this rectification could reduce the input resistance by up to 80%. This rectification was very rapid in onset and no decay was observed. A similar rectification has previously been seen in cat SPN by D e m b o w s k y et al. [8]. In other cells it has been termed an instantaneous anomalous rectification, m a n y vertebrate neurons possess this conductance including those of the olfactory cortex [5], inferior olive [30] and substantia gelatinosa [31].
241 All of the cells recorded from in this study fired action potentials in response to depolarizing currents (Fig. 2A, B), they were also seen to fire spontaneously or in response to orthodromically evoked synaptic potentials and antidromic activation. The spike characteristics were measured from the threshold for firing (mean of - 3 5 + 1.3 mV). The spikes had a mean amplitude 6 6 _ 1.8 mV (n = 32) and overshot zero by an average of 30 mV. Such large overshoots have been documented in vivo in the cat where a range of 12-50 mV was seen [8]. The spikes had durations (time spent above threshold) that averaged 8.8___0.4 ms (n=32, range 4.0-14.5 ms). Spikes characteristically exhibited an inflection in the downstroke of repolarization, suggestive of the presence o f a calcium component (Fig. 2B). Such shoulders have been documented in the cat both in vivo [8] and in vitro [33] and the rat in vitro [24]. In all cells to which T T X (0.2-1 a M ) was applied the spike was blocked reversibly suggesting that it is sodium dependent. All SPN exhibited an after-hyperpolarization (AHP) following the discharge of action potentials. The A H P had a mean amplitude (measured from threshold) of 2 4 _ 0.8 mV (n = 32, 14-33 mV) and an average duration of 400 ms (n=32, range 120-1000 ms). They are of similar duration but are larger in amplitude than those reported previously in the rat in vitro [15, 24], cat in vivo [8] and in vitro [33]. The spikes recorded in this study have longer durations and larger AHPs than those seen previously using intracellular recordings. Preliminary experiments have indicated that increasing the bath temperature to 36°C during a recording shortens the spike duration and decreases the size of the AHP. These observations are consistent with a previous report o f temperature effects upon CA1 hippocampal neurons [29], where decreased temperature lengthened spike durations and slowed calcium buffering and hence increased A H P amplitudes. A transient outward rectification was seen in 13 of 16 SPN (see Fig. 3). It took the form o f a delayed recovery to rest following hyperpolarizing pulses o f amplitude greater than -,, 15 mV (Figs. 2A and 3A). This recovery could take as long as 2 s following the end o f the pulse, considerably longer than might be expected from the passive time constant of the cells (Fig. 3B). A fast component of this rectification could be blocked reversibly by 2-4 m M 4-aminopyridine (4-AP) in the presence of 1 a M TTX, but was not significantly altered by lower concentrations of 4-AP (30-100 aM). This data suggests this rectification is due in part to the activation of an Acurrent [4, 21] rather than a D-current [26]. 4-AP also increased spike duration, shortened the A H P and increased the spontaneous synaptic activity onto the cells in the absence o f TTX. This transient rectification is
similar to that described previously in cat SPN both in vivo and in vitro [8, 32]. Using the WCR technique we have been able to obtain high resolution, stable intracellular recordings from rat SPN in vitro. These neurons have shown many o f the characteristic properties previously reported for both rat and cat SPN when using intracellular recording [6, 33], however we have seen that when using W C R the input resistances and time constants of these cells are considerably larger. These increased values indicate that SPN are more electrotonically compact [20] than previous conventional intracellular studies have suggested [8, 15, 16, 24, 33]. This may be vital in allowing SPN to integrate effectively inputs arising from such an extensive dendritic tree. We would like to express our gratitude to Mr. I.C. Gibson for invaluable discussions, to Mr. H.F. Ross for sharing his computer expertise with us and to Sue Ethel for photographic assistance. A.E.P. is in receipt of a Wellcome Trust prize studentship. This work is supported by the M R C and the British Heart Foundation.
1 Bacon, S.J. and Smith, A.D., Preganglionicsympathetic neurons innervatingthe rat adrenal-medulla:immunocytochemicalevidence of synaptic input from nerve-terminals containing substance-P, GABA or 5-hydroxytryptamine,J. Auton. Nerv. Syst., 24 (1988) 97-122. 2 Barnes,S. and Werblin, F., Gated currents generatesingle spike activity in amacrine cells of the tiger salamander retina, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 1509-1512. 3 Blanton, M.G., LoTurco, J.J. and Kriegstein, A.R., Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex, J. Neurosci.Methods, 30 (1989)203-210. 4 Connor, J.A. and Stevens, C.F., Prediction of repetitive firing behaviour from voltage clamp data on an isolated neurone soma, J. Physiol., 213 (1971) 31-53. 5 Constanti, A. and Galvan, M., Fast inward-rectifyingcurrent accounts for anomalous rectificationin olfactory cortex neurones, J. Physiol., 385 (1983) 153-178. 6 Coote, J.H., The organizationof cardiovascularneurons in the spinal cord, Rev. Physiol.Biochem.Pharmacol., 110 (1988) 147-285. 7 Dembowsky, K., Czachurski, J. and Seller, H., Morphology of sympathetic preganglionic neurons in the thoracic spinal cord of the cat: An intracellular horseradish peroxidase study, J. Comp. Neurol., 238 (1985)453--465. 8 Dembowsky,K., Czachurski, J. and Seller,H., Three types of sympathetic preganglionicneurones with differentelectrophysiological properties are identified by intracellular recordings in the cat, PflfigersArch., 406(2)(1986) 112-20. 9 Edwards, F.A., Konnerth, A., Sakmann, B. and Takahashi, T., A thin slicepreparation for patch clamp recordingsfrom neurones of the mammalian central nervous system,PfliigersArch., 414 (1989) 600-612. 10 Fenwick, E.M., Marty, A. and Neher, E., A patch-damp study of bovine chromaffinceils and of their sensitivityto acetylcholine,J. Physiol., 331 (1982)577-597.
242 11 Forehand, C.J., Morphology of sympathetic preganglionic neurons in the neonatal rat spinal-cord: an intracellular horseradish peroxidase study, J. Comp. Neurol., 298(3) (1990) 334-342. 12 Grace, A.A. and Llinas, R., Morphological artifacts induced in intracellularly stained neurons by dehydration: circumvention using rapid dimethyl sulfoxide clearing, Neuroscience, 16 (1985) 461-475. 13 Horn, R. and Marty, A., Muscarinic activation of ionic currents measured by a new whole-cell recording method, J. Gen. Physiol., 92 (1988) 145-159. 14 Lamb, T.D., An inexpensive digital tape recorder suitable for neurophysiological signals, J. Neurosci. Methods, l 5 (1985) 1-13. 15 Ma, R.C. and Dun, N.J., Vasopressin depolarizes lateral horn cells of the neonatal rat spinal cord in vitro, Brain Res., 348 (1985) 3643. 16 McLachlan, E.M. and Hirst, G.D.S., Some properties of preganglionic neurons in upper thoracic spinal cord of the cat, J. Neurophysiol., 43 (1980) 1251 1265. 17 Pickering, A.E., Spanswick, D. and Logan, S.D., Rectification in identified rat sympathetic preganglionic neurones in vitro, J. Physiol. (in press), Royal Free meeting, February, 1991. 18 Pickering, A.E., Spanswick, D. and Logan, S.D., Whole-cell recording from lateral horn cells in the neonate rat spinal cord in vitro, J. Physiol., 435 (1991) 36P. 19 Pongracz, F., Firestein, S. and Shepherd, G.M., Electrotonic structure of olfactory sensory neurons analyzed by intracellular and whole cell patch techniques, J. Neurophysiol., 65(3) (1991) 747 758. 20 Rail, W., Time constants and electrotonic length of membrane cylinders and neurons, Biophys. J., 9 (1969) 1483-1508. 21 Rogawski, M.A., The A-current: how ubiquitous a feature of excitable cells is it?, Trends Neurosci., 8 (1985) 214-219. 22 Shen, E. and Dun, N.J., Neonate rat sympathetic preganglionic neurons intracellularly labeled with Lucifer Yellow in thin spinal cord slices, J. Auton. Nerv. Syst., 29(3) (1990) 247-254.
23 Spanswick, D. and Logan, S.D., Spontaneous rhythmic activity in the intermediolateral cell nucleus of the neonate rat thoracolumbar spinal cord in vitro, Neuroscience, 39(2) (1990) 395-403. 24 Spanswick, D. and Logan, S.D., Sympathetic preganglionic neurons in neonatal rat spinal-cord in vitro: electrophysiological characteristics and the effects of selective excitatory amino-acid receptor agonists, Brain Res., 525 (1990) 181-188. 25 Stewart, W.W., Intracellular marking of neurons with a highly fluorescent napthalimide dye, Cell, 14 (1978) 741-759. 26 Storm, J.F., Temporal integration by a slowly inactivating K + current in hippocampal neurons, Nature, 336 (1988) 379-381. 27 Strack, A.M., Sawyer, W.B., Marubio, L.M. and Loewy, A.D., Spinal origin of sympathetic preganglionic neurons in the rat, Brain Res., 455 (1988) 187-191. 28 Takahashi, T., Membrane currents in visually identified motoneurones of neonatal rat spinal cord, J. Physiol., 423 (1990) 27-46. 29 Thompson, S.M., Masukawa, L.M. and Prince, D.A., Temperature dependence of intrinsic membrane properties and synaptic potentials in hippocampal CA1 neurons in vitro, J. Neurosci., 5 (1985) 817-824. 30 Yarom, Y. and Llinas, R., Long-term modifiability of anomalous and delayed rectification in guinea pig inferior olivary neurons, J. Neurosci., 7 (1987) 1166-1177. 31 Yoshimura, M. and Jessell, T.M., Membrane properties of rat substantia gelatinosa neurons in vitro, J. Neurophysiol., 62(1) (1989) 109-118. 32 Yoshimura, M., Polosa, C. and Nishi, S., A transient outward rectification in the cat sympathetic preganglionic neuron, Pfliigers Arch., 408 (1987) 207-208. 33 Yoshimura, M., Polosa, C. and Nishi, S., Electrophysiological properties of sympathetic preganglionic neurons in the cat spinal cord in vitro, Pfliigers Arch., 406(2) (1986) 91-8.
