Whole-cell patch-clamp recordings from respiratory neurons in neonatal rat brainstem in vitro

Neuroscience Letters, 134 (1992) 153 156

153

1992 Elsevier Scientific Publishers Ireland Ltd. NSL 08288

Whole-cell patch-clamp recordings from respiratory neurons in neonatal rat brainstem in vitro Jeffrey C. Smith, Klaus Ballanyi and Diethelm W. Richter II. Physiologisches lnstitut, University of G6ttingen, G6ttingen (F.R.G.) (Received 19 July 1991; Revised version received 23 September 1991; Accepted 25 September 1991)

Key words: Brainstem spinal cord in vitro; Medullary respiratory neuron; Synaptic current; Voltage clamp; Current clamp; Neonatal rat Whole-cell recordings were obtained from respiratory neurons by applying patch-clamp techniques in the en bloc medulla of in vitro neonatal rat brainstem-spinal cord preparations. Stable voltage-clamp recordings of excitatory or inhibitory synaptic drive currents and current-clamp recordings of spike discharge of inspiratory and expiratory neurons could be maintained for periods of 1-2 h. Parameters of whole-cell recording, including membrane seal resistances and series resistances, obtained in the en bloc medulla were similar to those obtained in corresponding regions of thin slices where neurons were directly visualized to optimize conditions for whole-cell patch clamp.

In vitro preparations of neonatal rat brainstem-spinal cord that preserve respiratory network function provide the opportunity for novel investigations of neuronal mechanisms of respiratory rhythm and pattern generation in mammals [12, 14]. For analysis of cellular and synaptic mechanisms, current- and voltage-clamp recordings must be routinely obtained from neurons in the medullary respiratory network in these preparations [12]. Although conventional microelectrode techniques have been used [6, 12], there are limitations of these techniques, particularly for voltage-clamp analysis of membrane currents in the small neurons of the neonatal medullary reticular formation. Methods for whole-cell patch-clamp recording from small central nervous system (CNS) neurons in vitro that provide optimal recording of membrane currents have recently been developed [2, 3]. One method [2] does not require visualization of neurons to be patch clamped, allowing whole-cell recording from neurons in intact neural circuits in en bloc CNS tissue. Here we demonstrate that this technique can be applied to obtain current- and voltageclamp recordings from medullary respiratory neurons in the en bloc brainstem-spinal cord preparations. Preliminary reports have been published [1, 11]. Experiments were performed in the in vitro brain-

Correspondence: J.C. Smith. Current address: Systems Neurobiology Laboratory, Department of Kinesiology, University of California, Los Angeles, 405 Hilgard Ave., Los Angeles, CA 90024-1568, U.S.A.

stem-spinal cord isolated from 1- to 4-day-old SpragueDawley rats. Details of the brainstem-spinal cord preparation and techniques for nerve recording have been described elsewhere [12]. The rats were anesthetized with ether and the neuraxis was isolated by dissection in physiological solution at 27°C. The brainstem, transected at the pontomedullary junction, and spinal cord were pinned down with ventral surface upwards in a recording chamber and bathed in a solution that contained 128 mM NaC1, 3.0 mM KC1, 1.5 mM CaC12, 1.0 mM MgSO4, 24 mM NaHCO3, 0.5 mM NaH2PO4, and 30 mM D-glucose equilibrated with 95% 02-5% CO2 at 27°C (pH = 7.4). Suction electrodes used for recording respiratory motor discharge and nerve stimulation were applied to C4 spinal ventral roots, glossopharyngeal (IX) nerve roots, and vagus (X) nerve roots (Fig. 1). The ipsilateral cranial nerve roots were stimulated for antidromic activation of motoneurons; axon tracts were stimulated in the C2-C3 spinal segment with a bipolar electrode for identification of bulbospinal neurons. Stimulus parameters (10-100 pA, 1-10 V, 0.5 ms pulses, 1-5 Hz) were used that caused descending synaptic activation of C4 phrenic motoneurons. Whole-cell recordings were obtained by procedures similar to those described [2]. Patch electrodes with small tips (l-2/tm diameter, 4-7 M£2 resistances) pulled (twostage pull) from borosilicate glass, and with Sylgardcoated shanks were used. Electrodes were filled with a solution containing: 120 mM D-gluconic acid (potassium salt), 1 mM CaC12, 1 mM NaC1, 10 mM HEPES, 11 mM

154

A

1~

(-70 mV)

,,,

1~II :mvl i I C4

Fig. 1. Schematic of brainstem-spinal cord preparation illustrating arrangement of recording and stimulating electrodes for whole-cell patch-clamp recording. A D : respiratory (inspiratory) neuron currents and potentials obtained in different recording configurations. A: neuronal currents generated by action potential discharge recorded in voltage-clamp mode of amplifier with patch electrode apposed to neuronal membrane prior to seal formation. Currents are synchronous with inspiratory phase motoneuronal discharge recorded on Ca spinal ventral roots and cranial nerves. B: single-channel/channel population currents recorded in cell-attached configuration following membrane seal formation; patch electrode potential = 0 mV. C: whole-cell voltage-clamp recording of rhythmic, inspiratory synaptic drive currents following membrane rupture. D: whole-cellcurrent-clamp recording of rhythmic depolarizing potentials and neuronal spike discharge synchronous with inspiratory motor discharge.

duced reversible perturbations o f the respiratory m o t o r o u t p u t pattern when the electrode was advanced into reticular formation regions containing respiratory neurons. F o r gigaohm (GO) seal formation, the pipette pressure was restored to atmospheric in the target region; G O seals were formed in the voltage-clamp m o d e o f the amplifier by application of slight negative pressure after pushing against the neuronal membrane. Respiratory neurons could be identified prior to seal formation by recording currents ( ~ 2 0 - 5 0 pA) (or field potentials under current clamp) during the rhythmic discharge o f action potentials with the pipette tip apposed to the somal m e m b r a n e or in the loose cell attached configuration (Fig. 1A). Whole-cell recordings were obtained by m e m b r a n e rupture by applying slight negative pressure to the pipette at - 70 mV holding potential. Recordings o f synaptic drive currents and spike discharge patterns of inspiratory and expiratory neurons are shown in Figs. 2 and 3. N e u r o n a l m e m b r a n e and whole-cell recording parameters are given in Table I. G i g a o h m seals could be formed on reticular formation neurons in ~ 75 % o f attempts; seal resistances were sufficiently high ( ~ 5 GO) in some cases for g o o d resolution of single channel and channel population currents (Fig. I B). Satisfactory whole-cell recordings with relatively low m e m b r a n e leakage currents (/Leak) and series resis-

BIIIII

A h

El?vl

Em (mY)

:tJL---

-60

2 sec

C VH

-70

....

D

mV

VH (my).... t ~ , l , +20 ~

I

L,

-"

-30 ~ I100 pA

B A P T A (tetrapotassium salt), 1 m M MgCI2, 0.5 m M N a A T P or 2 m M M g A T P (pH = 7.3-7.4). M e m b r a n e currents and potentials were measured with either a List EPC 7 or A x o p a t c h 1D patch clamp amplifier (signals low-pass filtered at 3-5 kHz). The electrode was advanced with a microdrive into the ventrolateral medulla and recordings were obtained from neurons (300-500 # m deep) in the respiratory cell populations [12]. Positive pressure (140-200 m m H g ) was applied to the back o f the pipette to maintain tip patency and prevent membrane attachment, as indicated by continuous measurement o f pipette resistance. The elevation o f extracellular potassium with ejection o f the electrode solution pro-

- -

2 sec

-70

- ' h~

,.~ _,~11¢*-'--""--"

200 msec

Fig. 2. A: whole-cellcurrent-clamp recording from an inspiratory neuron. Neuron was located 390 pm below ventral surface and was classified as an interneuron from antidromic stimulation tests. B: higher gain, slower time-base current-clamp recording; membrane potential trajectories during expiratory phase are characterized by membrane hyperpolarization after synaptic drive potential and gradual repolarization. Action potentials have been truncated. C: voltage-clamp recordings of synaptic currents showing a rapidly peaking, slowly declining pattern of drive current. D: voltage-dependence of inspiratory phase synaptic current with current reversal between - 10 and 0 mV holding potential (Vn). Current measurements made with ~ 75 % series resistance compensation; Rs = 26 Mg2, CM = 19 pF.

155

A

TABLE I

04

.

.

.

.

.

al

.

NEURONAL MEMBRANE AND WHOLE-CELL RECORDING PARAMETERS Rsea~, resistance of membrane seal in patch-clamp configuration; RM, neuronal input resistance in whole-cell configuration measured from current response to 10 mV voltage pulses from - 6 0 mV holding potential; Rs, series resistance; CM, membrane capacitance; /Leak, whole-cell leakage current at - 6 0 mV holding potential; I'M, zero current membrane potential. Measurements were made with identical electrodes and solutions and with/i> 75 % series resistance compensation. Values: mean + S.D.

B VC (mY)

-95

-7s , 15 mV

-105. . . . . ~ , ~ ~ ' ~ . ~ . . , ~ w

-05

Respiratoryneurons (en bloc medulla)

~

150~

Non-respiratory Medullaryneuneurons (en bloc rons (thin slices) medulla)

Rseal (GI2)

4.9 ___ 1.8

4.6 ___ 1.9

4.8 + 2.6

RM (M~2)

310 _+ 109

301 __+ 150

480 ___ 220

Rs(Mt2)

28 -I- 11

29 + 14

27 ___ 16

CM(pF)

17 _+ 6

22 + 13

25 ___ 15

500 msae

C4 Fig. 3. A: whole-cellcurrent-clamprecordingsfrom an expiratoryneuron receivinghyperpolarizinginhibitorysynapticinput during inspiratory phase. B,C: voltage-dependenceof inhibitory synaptic potential under current-clampand synapticcurrent under voltage-clampconditions with reversal potentialnear -70 mV (Vc: current-clampholding potential; VH:voltage-clampholdingpotential). Rs = 28 M~2,CM = 13 pF. Neuron was located 340/tm below ventral surface.

tances (Rs) (Table I) were obtained in approximately 50% of these cells. In the other neurons, unsatisfactory membrane rupture resulted in high, unstable Rs ( > 40 MI2), /Leak (>100 pA), and low resting membrane potentials; these cells were excluded. Stable recordings of synaptic currents and spike discharge of respiratory neurons could be maintained for periods of 1-2 h without significant changes in/Leak and input resistance (Table I), or reductions in peak Na currents (0.5-1.5 nA) or synaptic current amplitudes, with the baseline potential maintained at or near the zero current (resting) potential. The stability of the spike discharge patterns suggests that the currents involved were not subject to rundown with dilution or washout of intracellular constituents [5, 10]. Inspiratory neurons (n = 20) exhibited large amplitude rhythmic synaptic drive potentials (10-20 mV, 600900 ms duration) and spike discharge under current clamp (Fig. 2A,B), and large inward, excitatory synaptic drive currents (0.2-0.6 nA peak currents) during the inspiratory phase under voltage-clamp conditions (mean peak synaptic current amplitudes = 410 __+ 180 pA (S.D.) at - 70 mV holding potential). Resting membrane potentials of these cells estimated from the zero current potential under current clamp ranged from - 44 to - 60 mV (Table I). Values of whole-cell capacitances (10-20 pF) are indicative of small neurons. Membrane resis-

VM(mV)

- 5 4 _+ 6

- 5 5 __+ 9

- 5 1 __+ 8

/Leak (pA)

--20 + 7

--29 ___ 17

--47 __+ 30

tances (150-500 MI2) were approximately an order of magnitude higher than values obtained for respiratory neurons in vitro by conventional single-microelectrode techniques [6, 12], as found with comparisons of resistance values obtained by whole-cell and conventional intracellular techniques for other mammalian CNS neurons [2, 3]. The inspiratory neurons typically exhibited spontaneous unitary excitatory postsynaptic currents (,-~ 10-50 pA) during the expiratory phase (Fig. 2C). Approximately 50 % of the neurons also exhibited membrane potential hyperpolarizations (3-5 mV) after the synaptic drive potential with gradual repolarization (Fig. 2B) during the expiratory phase. Membrane currents producing these hyperpolarizations typically could not be detected under voltage-clamp ( - 6 0 to - 7 0 mV holding potential, Fig. 2C), suggesting that voltage-dependent currents may be involved - - possibly voltagedependent ionic fluxes during action potential discharge (e.g. Ca 2+ fluxes) and subsequent hyperpolarizing current activation. None of the group of inspiratory neurons tested (n = 10) for axonal projections with cranial nerve stimulation generated antidromically activated action potentials and were classified as interneurons; all of the neurons received polysynaptic excitation with cranial nerve afferent stimulation. One neuron generated antidromic spikes with bulbospinal axon tract stimulation and was classified as a bulbospinal neuron. All of these inspiratory interneurons exhibited the rapidly peaking

156 -slowly declining pattern o f synaptic drive and spike discharge (Fig. 2A) characteristic o f medullary inspiratory neurons in the neonatal rat [12]. Voltage-dependence o f the synaptic currents was examined in ten o f the inspiratory cells. In four neurons it was possible to reverse synaptic drive currents (Fig. 2D) under voltageclamp, even t h o u g h pipette solutions without K + channel blockers were used. Estimated reversal potentials were between - 10 and 0 mV, consistent with equlibrium potentials o f cationic conductances associated with excitatory amino acid mediated neurotransmission [9] in the respiratory network [4]. Synaptic currents could not be reversed in the remaining neurons due to p o o r space clamp and/or inadequate series resistance compensation, particularly in cells with large capacitances, leakage currents, and high series resistances ( > 35 MI2). Expiratory neurons (n = 4), exhibited continuous spike discharge (5-10 Hz spike frequency) during the exspiratory phase, and rapidly peaking inhibitory synaptic currents ( ~ 100-250 pA) producing 10-15 mV membrane potential hyperpolarizations during the inspiratory phase (Fig. 3). Cells with this spike discharge pattern represent one o f several classes o f medullary expiratory neurons active in vitro [13]. The mean zero current potentials of these cells was - 41 mV. Estimated reversal potentials of inhibitory currents or potentials (Fig. 3B,C) were between - 65 and - 75 mV. Parameters of whole-cell recording (Rseal, Rs, and /Leak) obtained in the en bloc medulla were c o m p a r e d to those obtained for neurons in the same region o f the ventrolateral reticular formation in thin medullary slices (200 /tm) [13], where superficial neurons were directly visualized [3] to optimize conditions for patch clamp (i.e. direct placement o f electrode tip on cell soma with minimal somal deformation). Cell surface cleaning [3] was not used. Recording parameters and m e m b r a n e properties for the group o f inspiratory neurons (n = 20), a group o f non-respiratory neurons (n -- 35) in the en bloc medulla, and for neurons (n = 32) in slices are shown in Table I. Values o f Rseal, Rs, and /Leak were similar, indicating that comparable conditions for m e m b r a n e patch-clamp and whole-cell recording were achieved. The low zero current potentials in all cases m a y reflect measurement errors introduced by the existence o f liquid junction potentials which have been estimated to be as high as - 10 mV [8]. The results indicate that long-term whole-cell recordings can be obtained from different classes o f medullary respiratory neurons in addition to spinal m o t o n e u r o n s [7] in the en bloc brainstem-spinal cord preparations. The whole-cell patch-clamp techniques can be exploited to obtain information on other cellular properties, including morphological properties and axonal projec-

tions, in addition to m e m b r a n e and synaptic properties [3]. Accordingly, these techniques will enable detailed investigations o f cellular and synaptic processes underlying the generation o f respiratory r h y t h m and pattern in the m a m m a l i a n brainstem-spinal cord in vitro. The authors thank Mrs. A.A. Seebode for expert technical assistance and Drs. E. Neher and R. Penner for the disposal o f an EPC 7 amplifier. This work was supported by N I H HL40959, D F G Ba 1095/1-1, Research Career Development A w a r d N I H HL02204 (J.C.S.), and an Alexander yon H u m b o l d t F o u n d a t i o n Fellowship (J.C.S.). I Ballanyi, K., Smith, J.C. and Richter, D.W., Patch-clamp analysis of respiratory neurones in the isolated brainstem of neonatal rats, Pflfigers Arch., (Suppl. 1) 418 (1991) RI6. 2 Blanton, M.G., Lo Turco, J.J. and Kriegstein, A., Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex, J. Neurosci. Methods, 30 (1989) 203 210. 3 Edwards, F.A., Konnerth, A., Sakmann, B. and Takahashi, T., A thin slice preparation for patch clamp recordings from neurons of the mammalian central nervous system, Pfliigers Arch., 414 (1989) 600--612. 4 Greer, J.J., Smith, J.C. and Feldman, J.L., Role of excitatory amino acids in the generation and transmission of respiratory drive in neonatal rat, J. Physiol., 437 (1991) 727-749. 5 Korn, S.J. and Horn, R., Influence of sodium-calcium exchange on calcium current rundown and the duration of calcium-dependent chloride currents in pituitary cells, studied with whole cell and perforated patch recording, J. Gen. Physiol., 94 (1989) 789-812. 6 Liu, G., Feldman, J.L. and Smith, J.C., Excitatory amino acidmediated transmission of inspiratory drive to phrenic motoneurons, J. Neurophysiol., 64 (1990) 423~437. 7 Liu, G. and Feldman, J.L., Whole cell patch-clamp recording of endogenous synaptic currents in mammalian motoneurons in intact brainstem-spinal cord, Soc. Neurosci. Abstr., 16 (1990) 1184. 8 Marry, A. and Neher, E., Tight-seal whole-cell recording. In B. Sakmann and E. Neher (Eds.), Single Channel Recording, Plenum, NewYork, 1983, pp. 107 122. 9 Mayer, M.L. and Westbrook, G.L., The physiology of excitatory amino acids in the vertebrate central nervous system, Prog. Neurobiol., 28 (1987) 197 276. 10 Pusch, M. and Neher, E., Rates of diffusional exchange between small cells and a measuring patch pipette, Pflfigers Arch., 4li (1988) 204-211. 11 Smith, J.C., Greer, J.J., Ballanyi, K., Feldman, J.L. and Richter, D.W., Recent advances in studies of neural mechanisms generating respiratory rhythm in mammalian brainstem-spinal cord in vitro. In Modulation of Respiratory Pattern: Central and Peripheral Mechanisms (Proc. Soc. Neurosci. Sat. Symp.), 1990, p. 13. 12 Smith, J.C., Greer, J.J., Liu, G. and Feldman, J.L., Neural mechanisms generating respiratory pattern in mammalian brainstem spinal cord in vitro. I. Spatiotemporal patterns of motor and medullary neuron activity, J. Neurophysiol., 64 (1990) 1149-1169. 13 Smith, J.C., Ellenberger, H., Ballanyi, K., Feldman, J.L. and Richter, D.W., Pre-B6tzinger Complex: a brainstem region that may generate respiratory rhythm in mammals, Science, in press. 14 Suzue, T., Respiratory rhythm generation in the in vitro brainstem spinal cord preparation of the neonatal rat, J. Physiol., 354 (1984) 173--183.