- Email: [email protected]
Decreased excitability of respiratory motoneurons during hypercapnia in the acute spinal cat
Brain Research, 386 (19,',;~)290-304
296
Elst~vicl BRE 12122
Decreased Excitability of Respiratory Motoneurons During Hypercapnia in the Acute Spinal Cat JOZEF S. JODKOWSKI and JANUSZ LIPSKI*
Department of Physiology, Institute of Physiological Sciences, Medical Academy, Warsaw (Poland) (Accepted 1 April 1986)
Key words: Carbon dioxide - - Phrenic motoneuron - - Intercostal motoneuron - - Cat
This study was undertaken to examine the effects of hypercapnia on the excitability of respiratory motoneurons. The action of CO 2 on phrenic (inspiratory) and internal intercostal (expiratory) motoneurons was compared with that exerted on non-respiratory motoneurons of the musculocutaneous nerve. The experiments were performed on spinalized (C1 segment), partially deafferented cats that were exposed to different CO2/O 2 mixtures (end-tidal CO: 3 + 0.3, 6 + 0.5 and 9 + 0.5%). Changes in neuronal excitability were assessed by: (1) measuring the amplitudes of antidromic field potentials recorded from a population of motoneurons; (2) analysis of the amplitude and latency of the orthodromic response recorded from a given nerve and evoked by microstimulation within the corresponding motor nucleus; (3) monitoring the membrane potentials during intracellular recordings from phrenic motoneurons; and (4) recording ongoing activity of the phrenic and internal intercostal nerves. Hypercapnia (end-tidal CO z 6 _+ 0.5 or 9 + 0.5%) decreased the excitability of phrenic and musculocutaneous motoneurons, the effect being larger at the higher CO 2 level. Internal intercostal motoneurons were generally more resistant to the effects of CO 2. A depression of their excitability was observed only at end-tidal CO: 9 _+0.5 %. The decreased excitability of phrenic motoneurons was associated with membrane hyperpolarization. It is concluded that the depressant action of CO 2 is present in both respiratory and non-respiratory spinal motoneurons. The action of hypercapnia on respiratory motoneurons may oppose the excitatory effects exerted through specific chemoreflexes.
INTRODUCTION
In most of these e x p e r i m e n t s h y p e r c a p n i a was used t o g e t h e r with h y p e r o x i a , t h e r e b y significantly
H y p e r c a p n i a is k n o w n to i n f l u e n c e the activity of various n e u r o n s in the c e n t r a l n e r v o u s system t h r o u g h its central action, apart f r o m the k n o w n excitatory reflex effects o n respiratory a n d a u t o n o m i c n e u r o n s e v o k e d t h r o u g h excitation of p e r i p h e r a l chem o r e c e p t o r s (for ref., see ref. 45). H i g h levels of car-
r e d u c i n g the reflex action of CO2 t h r o u g h excitation of p e r i p h e r a l c h e m o r e c e p t o r s ~7. It is, h o w e v e r , diffi-
b o n dioxide decrease the excitability of cortical n e u rons 2°~27, spinal a m o t o n e u r o n s (refs. 3, 6, 39; b u t see ref. 27), a n d inhibit afferent t r a n s m i s s i o n in the dorsal c o l u m n - l e m n i s c a l system 35'36. T h e activity of spinal i n t e r n e u r o n s is i n c r e a s e d or d e c r e a s e d 6'7'43"47, a n d similar variability in r e s p o n s e has b e e n o b s e r v e d in recordings from n e u r o n s in the v e n t r o l a t e r a l medulla (ref 40, cf. also ref. 18).
cult to establish w h e t h e r the C O 2, a p p l i e d in a n i m a l s with intact central n e r v o u s system, acts t h r o u g h specific m e d u l l a r y c h e m o r e c e p t o r s 42 or directly, i.e. without b e i n g m e d i a t e d by n e u r o n s i n v o l v e d in central c h e m o r e c e p t o r pathways. In fact, ' m i x e d ' direct a n d reflex effects have b e e n o b s e r v e d in s o m e n e u rons. F o r e x a m p l e , in respiratory m e d u l l a r y n e u r o n s , h y p e r c a p n i a causes a n increase in s y n a p t i c excitation due to activation of c h e m o r e c e p t o r s , as well as m e m b r a n e h y p e r p o l a r i z a t i o n ascribed to a direct action of C O 2 (ref. 33). A direct effect of h y p e r c a p n i a o n n e u r o n a l firing,
* Present address: Experimental Neurology Unit, John Curtin School of Medical Research, Australian National University, Canberra, A.C.T. 2601, Australia. Correspondence: J.S. Jodkowski. Present address: Department of Physiology and Piophysics, SJ-40, University of Washington School of Medicine, Seattle, WA 98195, U.S.A. 0006-8993/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
297 intercellular coupling or synaptic processes has been demonstrated in brain slices in vitro (e.g. refs. 10, 18), in cerebral cortex isolated from underlying structures 2°'27 and in spinal animals 5'7'15'16'25'26'5°. In acute spinal cats, a depressive action of systemic hypercapnia is usually observed 3,7'15,16,25,39, although excitation 5° has also been reported. The present study was undertaken to assess whether CO 2 (range 3-9.5%, end-tidal concentration) modifies directly the activity of motoneurons controlling respiratory muscles. The reflex effects of CO 2 through arterial chemoreceptors and central CO 2 action through chemosensitive neurons were excluded by high cervical spinalization. We have analyzed the excitability of phrenic (inspiratory) and internal intercostal (expiratory) motoneurons, and some results have been compared with those from motoneurons not directly involved in the control of respiratory movements. Preliminary results have been briefly reported 24. MATERIALS AND METHODS
General procedures Experiments were performed on 27 cats of either sex (2.1-4.0 kg) anesthetized with a-chloralose and urethane (40 and 260 mg/kg i.v., respectively). A tracheostomy was made. Blood pressure was continuously recorded from the femoral artery (P23Db, Statham). Rectal temperature was monitored and held at 37-38 °C. The C5 and C6 branches of the left phrenic nerve, the left musculocutaneous nerve (innervating biceps brachii, brachialis and coracobrachialis muscles) and the left internal intercostal nerve (T8) were exposed, cut distally and prepared for standard recording and/or stimulation. The animals were placed in a stereotaxic frame with pelvic pins and vertebral clamps at T1, T6 and T10, paralyzed with gallamine (Tricuran, Germed, 5 mg/kg/h, i.v.) and artificially ventilated with oxygenenriched air. A bilateral pneumothorax was performed and an expiratory load of 2 cm H20 imposed. Tracheal pressure was recorded continuously and end-tidal CO 2 concentration (ET CO2) monitored (LB-2, Beckman). A laminectomy was performed that extended from C3 to C7 segments (to allow recording from phrenic and musculocutaneous nerve motoneurons) and/or from T7 to T9 (for recordings
from internal intercostal motoneurons). Left dorsal roots were cut in the exposed segments. The surface of the spinal cord was covered with warm paraffin oil.
Spinalization From a midline incision the muscles of the neck were retracted to either side. The atlanto-occipital membrane was cut and vessels on the caudal portion of the medullary surface were coagulated. The spinal cord was transected just above the dorsal roots of the first cervical segment using fine scissors. Incisions were made in several steps at intervals of 3-5 min; completeness of the transection was carefully checked, and bleeding was controlled with Gelfoam. If necessary, Dextran or norepinephrine (or both) were infused intravenously to maintain a mean arterial blood pressure above 80 mm Hg. Recordings were started at a minimum of 2 h after spinalization.
Stimulation and recording The dissected nerves were stimulated with bipolar silver wire electrodes using constant-current pulses of 0.2 ms (NL 510 and NL 800, Digitimer). The stimulus amplitude was 4 times threshold value required to evoke motoneuronal antidromic field potentials. Stimulus frequency was 2 Hz, using pairs of pulses presented at an interval of 2-30 ms. Antidromic field potentials were recorded with glass microelectrodes filled with 0.5 M NaCI (resistance approximately 1 Mfl), amplified (NL 104, Digitimer), averaged over 16-64 sweeps (Anops 105, Warsaw Politechnic) and photographed. Depending on the position of the recording electrode within the ventral horn, either antidromic field potentials generated by axons or cell bodies of motoneurons were recorded. Because the axonal refractory period is shorter than that of the soma (e.g. ref. 12), field potentials unchanged by stimuli presented 2-3 ms apart were presumed to originate from axons. Mainly somatic antidromic field potentials (i.e. potentials with a clear amplitude decrease following the second stimulus of a pair, presented with a delay of 5-30 ms) were analyzed in this study (see Fig. 1). Glass microelectrodes filled with 3 M potassium citrate (resistance 10-15 Mr2) were used for intracellular recording from phrenic motoneurons (NL 102, Digitimer). The motoneurons were identified by recording antidromic action potentials following stimuli applied to the correspond-
298 ing (C5 or C6) Phrenic branch (cf. ref. 30). Glass-insulated tungsten electrodes (NL 05, Digitimer) were used for monopolar microstimulation within the ventral horn. The electrodes were placed at a point where maximal somatic antidromic field potentials could be evoked by stimulation (see above) of the selected nerves. At such loci, constantcurrent pulses (0.1-0.2 ms, 17-70 ~ A ) were applied, and orthodromic potentials evoked in the corresponding nerves were amplified and averaged. Ongoing activity present in some of the dissected nerves was amplified and processed by an RC integrator (time constant, 100-500 ms). The output signals were recorded on a chart recorder.
Experimental protocol At the beginning of each experiment, frequency and volume of the respirator were adjusted to obtain E T CO 2 of 3 _+ 0.3%. Changes in expired CO 2 level were accomplished by adding a 15% CO2/85% O 2 gas mixture to the inlet of the respirator. Desired levels of CO 2 were obtained within approximately 5 min. Recordings were made at E T CO 2 levels of 3 +_ 0.3% (control), 6 + 0.5% and 9 + 0.5%. C O 2 administration was then terminated and the last recording was made 20 min later. A series of 4 such recordings was accepted for further analysis only when there was no difference (within + 5 % ) between the first (control) and the last (recovery) records. The following parameters were analyzed: amplitudes and latencies of the evoked potentials, m e m b r a n e potential during intracellular recordings, and ongoing activity recorded from dissected nerves. Amplitudes of the evoked responses were measured from baseline level to peak of the main negative deflection. Student's paired t-test was used for statistical analysis. RESULTS
Phrenic motoneurons The influence of CO2 on somatic antidromic field potentials was analyzed in 15 series of experiments performed on 12 cats. Control amplitudes ranged from 0.3 to 1.3 mV (mean 0.6 mV) and their latencies from 0.42 to 1.37 ms (mean 0.84 ms). The field potential was depressed from control by 15.5 _ 4.2% (mean + S.E.M.; P < 0.001) at E T CO 2 6 + 0.5%
and by 30.6 _+ 6.1% (P < 0.001) at ET CO 2 9 + 0.5% (Fig. 1). The difference between depression at the two hypercapnic levels was statistically significant (P < 0.001). No changes in antidromic latency were observed. In 4 recording positions, effect of CO~ was tested on antidromic field potentials recorded from axons (see Methods). No significant effect was observed with E T CO 2 up to 9.5%. Orthodromic responses recorded from C5 or C6 phrenic branches following microstimulation within the corresponding section of the phrenic nucleus were studied in 3 cats. In 6 of 7 series of measurements, a depression of the evoked potentials was observed and this was sometimes (4 of 6 cases) associated with an increased onset latency (Fig. 2). These data are presented in Table I. Two hours after spinalization, ongoing tonic activity in the C5 phrenic branch was observed in 9 of 27 cats (33%). The number of active fibers varied. In 5 of the experiments in which there was tonic activity no clear change in firing occurred during CO2 administration (at either 6 _+ 0.5% or 9 + 0.5% level). In the remaining 4 experiments, a decrease in activity
A
B
t
I "---7 "~ , F
2
4
3.5%
I " 'k~-
5.9%
P0-
,-',,__~ A
9.0%
ET COz (%) 3±03
6±0.5
9~05
,' i" : f -
5
A" ~
0
P
20-
3.0% Ims
30-
-]lmV 40%
Fig. 1. Influence of CO 2on antidromic field potentials recorded in phrenic nucleus. A: examples of records in a representative experiment. C5 phrenic branch was stimulated with a pair of stimuli (64/~A, 0.2 ms) presented at interval of 28 ms, Stimulus artifacts are marked by arrowheads. Potentials averaged (16 sweeps). Negativity upwards. 1, control record; 2 and 3, records during CO2 administration; 4, 20 rain after CO2 withdrawal. B: influence of CO 2on peak amplitudes of antidromic field potentials in phrenic nucleus (15 experiments): Each column represents mean difference (_+ S.E.M.) between control amplitude and amplitude at a given CO 2 level expressed in percent of control value. Data for ET CO 2 of 3 __-0.3% obtained by comparing control amplitudes with amplitudes recorded after CO 2 withdrawal. *, different from control (P < 0.001). **, different from control level (P < 0.001) and from ET CO2 6 _+0.5% level (P < o.ool).
299 ETCO • •
-•
2
(%)
2'7
r n s
"j
--
5-7
"
8-8
L "
,,,,d'~
slight hyperpolarization (about 1 mV) and decreased excitability, as reflected by blockade of the IS-SD spike (Fig. 4). In the 18 other recordings, CO 2 administration was accompanied by a progressive depolarization, changes in spike shape (broadening, decreased amplitude) and failure of antidromic activation. These changes were never reversible after CO 2 withdrawal and, in many cases, recordings were lost during CO2 administration (sudden drop in measured intracellular membrane potential to a level close to 0 mV).
2.7
7 Fig. 2. Influence of CO 2 on orthodromic potentials recorded and averaged (n = 16) from C5 phrenic branch. Stimulus (17 /tA, 0.2 ms; threshold intensity, 9/~A) onset indicated by arrow. Negativity downwards in all records.
was seen, followed by recovery after CO 2 withdrawal (Fig. 3). Intracellular recordings were attempted from phrenic motoneurons in 9 spinalized animals. A recording was considered acceptable if the impaled motoneuron had a stable membrane potential more negative than -50 mV for at least 5 min and a spike potential of 60 mV or more. Twenty-one motoneurons initially met these criteria. For analysis of their responses to CO 2, however, we accepted recordings only from those motoneurons in which the membrane potential returned to control level after CO2 withdrawal (see Methods, experimental protocol). Only 3 motoneurons (3 different experiments) met these criteria. In these 3 cells, increased level of CO 2 led to
Internal intercostal motoneurons A series of 11 recordings was made in 7 cats. Control amplitudes of antidromic field potentials ranged from 0.18 to 0.83 mV (mean 0.43 mV) and their onset latencies from 0.34 to 1.1 ms (mean 0.53 ms). Moderate hypercapnia (ET CO 2 6 + 0.5%) did not significantly change the amplitudes and latencies of these potentials• At an ET CO2 of 9 + 0.5%, however, the amplitudes of the potentials decreased from control values by 11.3 _+ 3.1% (P < 0.001) with no change in the antidromic latency (Fig. 5A). Orthodromic potentials were evoked in the internal intercostal nerves by microstimulation within the internal intercostal nucleus in 5 animals. Amplitudes of the evoked potentials ranged from 28 to 198/~V (mean 117/~V, n = 12) and their latencies from 0.25 to 0.57 ms (mean 0.43 ms). During CO 2 administration, a depression of the amplitude was observed in only 3 of 12 series of measurements (cf. Fig. 5B). Onset latencies of the responses were not changed in any of these experiments• Following spinalization, ongoing tonic activity of
TABLE I Amplitudes and latencies o f orthodromic potentials evoked in phrenic nerve following microstimulation within the phrenic nucleus Measure-
Stimulus
Control
ment
(/zA)
E T C O 2 3+_0.3%
E T C O 2 6+_0.5%
ETCO 2 9+0.5%
ETCO 2 3+0.3%
Amplitude
Latency
Amplitude
Latency
Amplitude
Latency
Amplitude
(~V)
(ms)
(l~V)
(ms)
(~V)
(ms)
(~V)
(ms)
160 146 260 250 630 212 328
0.71 0.79 0.88 0.80 0.56 0.42 0.46
160 150 90 30 740 212 320
0.79 0.79 1.12 1.64 0.64 0.46 0.46
96 106 10 30 630 168 312
0.83 0.83 1.24 1.64 0.60 0.46 0.46
160 138 260 250 640 232 320
0.71 0.75 0.80 0.76 0.60 0.46 0.46
1 2 3 4 5 6 7
51 51 16.6 21.5 25 30 30
CO 2 withdrawal
Latency
300
A
IPh
03-8
~
03.8
e5.3
%.3
e6.9
e6.?
IT 8
@ 1 min
B
j•,•.•tLat_..L,_, ..... i . . . . e9"2
Li
a,..~
.... ~ _
,IIL~.
e9-4 A
e3-8
Fig. 3. Influence of CO 2 on integrated ongoing activity of C5 phrenic branch (IPh) and T8 internal intercostal nerve (ITs). Actual ET CO 2 values (non-steady-state) as indicated. Arrowhead in A indicates beginning of CO 2 administration. B: 2 rain after A. Arrowhead indicates end of CO 2 administration. C: 7 min after B. In A and C, middle line is zero reference for phrenic integrated activity.
A
C
B
m¥ -I10 •
l
I
A
J
A I
•
II
!:
,!i
~:
i!
5ms 0
E
D
F
mY -80
~ m A
L A
A
•
Fig. 4. Intracellular recordings from a phrenic motoneuron. Antidromic activation with a pair of stimuli (20 ~A, 0.2 ms) presented at interval of 8.5 ms. Stimulus artifact indicated by the arrowheads. Upper traces: intracellular records. Lower traces: differentiated records (r = 2.2/~s) of the intracellular spikes. Five superimposed sweeps in each trace. A: control (membrane potential, M P - 8 0 mV, ET CO 2 3.3%). B: 3 min after beginning of CO~ administration (MP -80 mV, ET CO 2 4.3%). C: 4 min after beginning of CO 2 administration (MP - 8 0 mV, ET CO 2 4.5%). D: 10 min after beginning of CO 2 administration (MP -81 mV, ET CO 2 5.3%). E: 5 min after CO~ withdrawal (MP -80 mV, ET CO 2 3.5%). F: 20 min after CO 2 withdrawal (MP -78 mV, ET CO 2 3.2%).
301
A
ETC02 (%) 3 + 0-3
Musculocutaneous nerve motoneurons
6 +0'5
9 + 0"5
%
B
3.0%
6.5%
i]
~'::
8.6%
2.9%
~'
ii/ 0.1 mV I
Fig. 5. Influence of CO2 on excitability of internal intercostal motoneurons. A: effect on amplitude of antidromic field potentials recorded in intercostal nucleus (11 experiments). *, different from control level (P < 0.001). For further explanation, see Fig. lB. B: effects on orthodromic potentials recorded from internal intercostal nerve. Arrowheads indicate stimulus (60 ~A, 0.1 ms, threshold current 23 ~A). Last record (ET CO 2 2.9%) taken 20 min after CO2 withdrawal. All traces are averages of 16 sweeps.
the internal intercostal nerve was present in 11 cats. Carbon dioxide caused a depression in only one experiment (Fig. 3), whereas in no change was observed. Nevertheless, in 3 experiments activity recorded from the C5 branch was depressed.
7 out of of firing 6 others of these phrenic
ET CO2 (%) 3+_0.3
o I_Tj
6+_0.5
The influence of CO 2 on antidromic field potentials evoked by stimulation of the musculocutaneous nerve was analyzed in 5 experiments (3 cats). Control amplitudes ranged from 0.51 to 2.1 m V (mean 1.07 mV) and onset latencies from 0.38 to 0.7 ms (mean 0.51 ms). At E T CO 2 6 +_ 0.5%, amplitudes of the potentials were decreased by 12.9 + 4.4% from control values (P < 0.01) and at E T C O 2 9 + 0.5% by 22.5 + 5% (P < 0.001) (Fig. 6). The difference in depression between the two hypercapnic levels was statistically significant (P < 0.015). No change in antidromic latency of any of these potentials was observed.
9+_0.5
10
Z0
30 %
Fig. 6. Influence of CO 2 on amplitude of antidromic field potential recorded from motoneurons of musculocutaneous nerve. *, different from control level (P < 0.01); **, different from control level (P < 0.01)1) and from ET CO 2 6 + 0.5% level (P < 0.015). For further explanation, see Fig. lB.
DISCUSSION Several authors have reported that profound hypercapnia ( > 10% C O 2 in inhaled air) depresses the excitability of lumbar motoneurons 6J6'39'43. Effects of lower C O 2 concentrations were less clear or not examined 3. Results of the present study show that CO2 depresses the excitability of inspiratory (phrenic) motoneurons to a degree similar to that observed in motoneurons having no apparent respiratory function. This effect was often seen at E T CO2 6 + 0.5%, and became more prominent at a higher C O 2 level (ET CO 2 9 + 0.5%). Although the effects were analyzed at only 3 different CO2 concentrations, our data indicate that the action of carbon dioxide is dose-dependent, at least in the range of 2 . 7 - 9 . 5 % . The conclusions of this report are based mainly on results obtained with two qualitative methods: one involving antidromic activation and the other orthodromic activation of motoneurons. With respect to the antidromic method, it was assumed that following supramaximal stimulation of the motor nerve, the amplitude of the antidromic field potential recorded within a m o t o r nucleus depended on the number of antidromically invaded cell bodies in the vicinity of the tip of the recording microelectrode 38. A n y factor changing the neuronal excitability (membrane potential and/or m e m b r a n e conductance) changes the proportion of antidromically excited neurons. Thus, a decrease or an increase in the amplitude of the antidromic field potential reflects, respectively, a decrease or increase in excitability of the studied motoneuron (cf., e.g. refs. 2, 34, 49). The 'orthodromic'
302 method involved microstimulation in the spinal cord, in the vicinity of phrenic or intercostal motor nuclei. Amplitude of the orthodromically evoked potentials in the corresponding nerves depends on the number of excited cell bodies. Any factor modifying excitability of the motoneurons changes the proportion of activated neurons, as fewer of more cells are excited at a certain distance from the centre of the stimulated region (where the stimulating current is close to threshold; e.g. refs. 13, 36, 37). Along with measurements of the amplitudes of antidromic field potentials, the depressive action of CO 2 was observed in 13 of 15 recording sites, at ET CO 2 6 + 0.5% and 9 + 0.5%. This result was generally confirmed in recordings of the amplitude of orthodromically evoked potentials in the phrenic nerve. The 'orthodromic' method, however, proved to be somewhat less sensitive in detecting a weak influence, as with 6 _+ 0.5% CO 2 the depression was observed in only two loci. With ET CO2 9 __ 0.5%, the decreased amplitude was observed in 6 of 7 experiments. The control latency of phrenic orthodromic responses did not exceed 0.9 ms, indicating direct stimulation of the motoneuronal cell bodies. Latencies were increased in 3 of 7 experiments with ET C O 2 6 + 0.5% and in 4 of 7 experiments with ET CO 2 9 _ 0.5%. In two of the latter experiments, a marked (> 0.35 ms) increase in latencies was observed. A likely explanation is that CO 2 first depressed the excitability of neurons whose axons had the highest conduction velocities. Another explanation (for the two most pronounced changes in latency) is based on the assumption that microstimulation within the spinal cord activates not only motoneurons, but also some excitatory interneurons and/or neuronal pathways u. When the excitability of motoneurons is decreased by CO 2, they may no longer be activated directly by the current. However, some of them could still be activated through a synaptic action, with an increase in the observed latency due to the synaptic delay (cf. ref. 22). Intracellular recordings from apparently undamaged phrenic motoneurons showed that decreased neuronal excitability accompanying hypercapnia was associated with membrane hyperpolarization. Such an influence was postulated by Gill and Kuno 21, but no direct evidence was presented. Both hyperpolarizing 6,16'39'43 and depolarizing 16'39 effects have been
reported in other studies with intracellular recordings from spinal non-respiratory motoneurons. Because in our experiments frequent depolarizations were observed that were associated with irreversible cell damage, it is probable that the depolarization is an artifact rather than a true C O 2 effect. This conclusion is supported by results of a recent study 2s in which inhibition of phrenic nerve discharges was observed following CO 2 administration in animals whose central excitatory chemoreceptor pathway was depressed by baclofen. Unfortunately, the small sample size in our experiments (only 3 motoneurons) does not allow for a definitive conclusion on this point. Carbon dioxide in concentrations of 10-20% was reported to depress transmission in nerve fibers s'16'36, The axonal effect did not occur in our experimental conditions: the M-spike 12 could still be observed in phrenic motoneurons with blocked IS-SD spike, there was no depression of 'axonal' field potentials, and the latencies of antidromic 'somatic' field potentials were not affected. Sensitivity to CO2, however, was not a uniform feature of all subpopulations of spinal motoneurons under study. Internal intercostal motoneurons were less affected than phrenic or musculocutaneous motoneurons. Occasionally, we observed that one field potential as compared to another recorded within the same motor nucleus was more resistant to the action of CO 2. Such a differential sensitivity was also observed with recordings of ongoing activity from phrenic and internal intercostal nerves. Following CO 2 administration, decreased firing frequency was observed in 4 of 9 C5 phrenic branches and in only one of 7 recordings from the internal intercostal nerves. It remains to be established whether these differences are due to a differential, direct CO: action on motoneurons or on premotor interneurons. In our experiments, only tonic, 'spontaneous' activity was observed in the C5 phrenic branch after spinalization (cf. refs. ll, 14). Several authors 1'9'a6 have observed phasic (respiratory-like) phrenic discharge under similar conditions. Since drugs (e.g. Doxapram, Nialamide, L-DOPA) were often used in these studies, the origin of this type of activity is not clear. It could be partly the result of spinal reflexes (cf. refs. 11, 14). Viala and Freton 46 reported an excitatory CO 2 effect on 'spinal' phasic activity of
303 the phrenic nerve. Although this reponse can be mediated by some interneurons that are excited by CO2 (ref. 43), our results do not support the hypothesis of a chemosensitive spinal respiratory generator. The mode of the so-called 'direct' C O 2 action is not clear. Studying neuronal effects in Aplysia, Brown et al. 4,48 observed that elevated levels of C O 2 increase the permeability of cell membranes to chloride ions. The response was shown to result mainly from a decrease in extracellular pH. In some insects, the C O 2 action can be explained by lowering the intracellular pH, which can lead to inhibition of m e m b r a n e active transport processes (ref. 8, cf. also ref. 41). These theories, based on experiments performed on invertebrates, do not necessarily explain results obtained with mammalian neurons 16, and hence other ionic mechanisms have to be taken in account 23'29'44. Changes in neuronal activity produced by high levels of CO 2 are not always similar to those produced by low p H 18,31'32. In addition, not all effects can be explained by the action of CO2 or low p H exerted upon the neurons under investigation. The effects on synaptic transmission cannot be excluded 29. Furthermore, the observed effects of CO2/pH can be evoked by action on other neurons, which change the excitability of the examined neuronal pool 19. At present it
is difficult to differentiate between the CO2 and/or low p H action per se on cell membranes and an indirect action through other mechanisms. Thus the term 'direct action' should be used with caution. Considered in conjunction with results of previous studies 5'6'16'33'43, we conclude that elevated levels of
REFERENCES
horn convergent neurons, Exp. Neurol., 70 (1980) 109-121. 8 Clark, M.A. and Eaton, D.C., Effect of COz on neurons of the house cricket, Acheta domestica, J. Neurobiol., 14 (1983) 237-250. 9 Coglianese, C.J., Peiss, C.N. and Wurster, R.D., Rhythmic phrenic nerve activity and respiratory activity in spinal dogs, Resp. Physiol., 29 (1977) 247-254. 10 Connors, B.W., Bernardo, L.S. and Prince, D.A., Carbon dioxide sensitivity of dye coupling among glia and neurons of the neocortex, J. Neurosci., 4 (1984) 1324-1330. 11 Decima, E.E. and von Euler, C., Excitability of phrenic motoneurons to afferent input from lower intercostal nerves in spinal cat, Acta Physiol. Scand., 75 (1969) 580-591. 12 Coombs, J.S., Curtis, D.R. and Eccles, J.C., The interpretation of spike potentials of motoneurones, J. Physiol. (London), 139 (1957) 198-231. 13 Eccles, J.C., Magni, F. and Willis, W.D., Depolarization of central terminals of Group I afferent fibres from muscle, J. Physiol. (London), 160 (1962) 62-93. 14 Eldrige, F.L., Gill-Kumar, P., Miilhorn, D.E. and Waldrop, T.G., Spinal inhibition of phrenic motoneurones by stimulation of afferents from peripheral muscles, J. Physiol. (London), 311 (1981) 67-79. 15 Esplin, D.W. and Rosenstein, R., Analysis of spinal depressant actions of carbon dioxide and acetalozamide,
1 Aoki, M., Moil, S., Kawahara, K., Watanabe, H. and Ebata, N., Generation of spontaneous respiratory rhythm in high spinal cats, Brain Research, 202 (1980) 51-63. 2 Bloedel, J.R. and Roberts, W.J., Functional relationship among neurons of the cerebellar cortex in the absence of anesthesia, J. Neurophysiol., 32 (1969) 75-84. 3 Brooks, C.McC. and Eccles, J.C., A study of the effects of anaesthesia and asphyxia on the mono-synaptic pathway through the spinal cord, J. Neurophysiol., 10 (1947) 349-360. 4 Brown, A.M., Walker, J.L. and Sutton, R.B., Increased chloride conductance as the proximate cause of hydrogen ion concentration effects in Aplysia neurons, J. Gen. Physiol., 56 (1970) 559-582. 5 Carpenter, D.O., Hubbard, J.H., Humphrey, D.R., Thompson, H.K. and Marshall, W.H., Carbon dioxide effects on nerve cell function. In I.G. Nahas and K.E. Schaefer (Eds.), Carbon Dioxide and Metabolic Regulations, Spilnger-Verlag, New York, 1974, pp. 49-62. 6 Caspers, H. and Speckmann, E.J., Cerebral pO2, pCO2 and pH: changes during convulsive activity and their significance for spontaneous seizure termination, Epilepsia, 13 (1972) 699-725. 7 Catchlove, R.F.H., Le Bars, D., Blanchet, F. and Besson, J.M., Effects of respiratory acidosis on the activity of dorsal
CO2 exert a dual action on spinal respiratory motoneurons. This condition can activate these motoneurons through excitation of peripheral chemoreceptors and specific medullary chemoreceptors. At the same time, C O 2 depresses excitability of respiratory motoneurons by an action at the spinal cord. The spinal action normally is outweighed by the powerful input from brainstem respiratory centers. Nevertheless, local CO2-dependent depression of respiratory motoneurons may impair these excitatory reflex mechanisms during severe hypercapnia.
ACKNOWLEDGEMENTS The authors thank Dr. A. Berger and Dr. R. Martin-Body for their critical remarks on the manuscript and H a n n a Atkins for editing of the manuscript. This study was supported by 10.4.03.5 Grant from the Polish A c a d e m y of Sciences.
304
Arch. Int. Pharmacodyn. Ther., 143 (1963) 498-513. 16 Esplin, D.W., Capek, R. and Esplin, B.A., An intracellular study of the actions of carbon dioxide on the spinal monosynaptic pathway, Can. J. Physiol. Pharmacol., 51 (1973) 498-513. 17 Fitzgerald, R.S. and Parks, D.C., Effect of hypoxia on carotid chemoreceptor responses to carbon dioxide in cats, Resp. Physiol., 12 (1971) 218-229. 18 Fukuda, Y., Difference between actions of high pCO 2 and low [HCO3- ] on neurons in the rat medullary chemosensitive areas in vitro, Pflagers Arch., 398 (1983) 324-330. 19 Fukuda, Y. and Loeschcke, H.H., A cholinergic mechanism involved in the neuronal excitation by H ÷ in the respiratory chemosensitive structures of the ventral medulla oblongata of rats in vitro, Pfliigers Arch., 379 (1979) 125-135. 20 Gellhorn, E. and French, L.A., Carbon dioxide and cortical spike frequency, Arch. Int. Pharmacodyn. Ther., 93 (1953) 427-433. 21 Gill, P.K. and Kuno, M., Properties of phrenic motoneurons, J. Physiol. (London), 168 (1963) 258-273. 22 Gustaffson, B. and Jankowska, E., Direct and indirect activation of nerve cells by electrical pulses applied extracetlularly, J. Physiol. (London), 258 (1976) 33-61. 23 Hille, B., Charges and potentials at the nerve surface. Divalent ions and pH, J. Gen. Physiol., 51 (1968) 221-236. 24 Jodkowski, J.S. and Lipski, J., Effects of carbon dioxide on spinal respiratory motoneurons, somatic motoneurons and sympathetic preganglionic neurons in cats, Neurosci. Lett., Suppl. 14 (1983) S183. 25 Kirstein, L., Early effects of oxygen lack and carbon dioxide excess on spinal reflexes, Acta Physiol. Scand., Suppl. 80, 23 (1951). 26 Kitahata, L.M., Taub, A. and Sato, I., Hyperventilation and spinal reflexes, Anesthesiology, 31 (1969) 321-326. 27 Krnjevi6, K., Randi6, M. and Siesjo, B.K., Cortical CO 2 tension and neuronal excitability, J. Physiol. (London), 176 (1965) 105-122. 28 Lalley, P.M., Biphasic effects of baclofen on phrenic motoneurons: possible involvement of two types of gammaaminobutyric acid (GABA) receptors, J. Pharmacol. Exp. Ther., 226 (1983) 616-624. 29 Landau, E.M. and Nachshen, D.A., The interaction of pH and divalent cations at the neuromuscular junction, J. Physiol. (London), 251 (1975) 775-790. 30 Lipski, J., Kubin, L. and Jodkowski, J., Synaptic action of R~ neurons on phrenic motoneurons studied with spiketriggered averaging, Brain Research, 288 (1983) 105-118. 31 Marino, P.L. and Lamb, T.W., Effects of CO 2 and extracellular H + iontophoresis on single cell activity in the brainstem, J. Appl. Physiol., 38 (1975) 688-695. 32 Miles, R., Does low pH stimulate central chemoreceptors located near the ventral medullary surface?, Brain Research, 271 (1983) 349-353. 33 Mitchell, R.A. and Herbert, D.A., The effect of carbon dioxide on the membrane potential of medullary respiratory neurons, Brain Research, 75 (1974) 345-349. 34 Morales, F. and Chase, M.H., Postsynaptic control of lure-
35
36
37
38
39
40
41 42
43
44
45
46
47
48
49 50
bar motoneuron excitability during active sleep in the chronic cat, Brain Research, 225 (1981) 279-295. Morris, M.E., The action of carbon dioxide on afferent transmission in the dorsal column-lemniscal system, J. Physiol. (London), 218 (1971) 651-669. Morris, M.E., The action of carbon dioxide on synaptic transmission in the cuneate nucleus, J. Physiol. (London), 218 (1971) 671-689. Morrison, A.R. and Pompeiano, O. An analysis of the supraspinal influences acting on motoneurons during sleep in the unrestrained cat. Responses of the alpha motoneurons to direct electrical stimulation during sleep, Arch. Ital. Biol., 103 (1965) 497-516. Nicholson, C., Generation and analysis of extracellular field potentials. In Electrophysiological Techniques, Society for Neuroscience, Atlanta, GA, 1979, pp. 93-147. Papajewski, W., Klee, M.R. and Wagener, A., The action of raised CO 2 pressure on the excitability of spinal motoneurones, Electroencephalogr. Clin. Neurophysiot., 27 (1969) 618. Pokorski, M., Neurophysiological studies on central chemosensor in medullary ventrolateral areas, Am. J. Physiol., 230 (1976) 1288-1295. Roos, A., Boron, W.F., Intracellulat pH, Physiol. Rev., 61 (1981) 296-434. Schlaefke, M.E., Central chemosensitivity: a respiratory drive, Rev. Physiol. Biochem. Pharmacol., 90 (1981) 171-244. Speckmann, E.-J. and Caspers, H., Responses of spinal and cortical neurons to changes of pO 2 and pCOz in blood and tissue. In M.J. Purves (Ed.), The Peripheral Arterial Chemoreceptors, Cambridge University Press, Oxford, 1975, pp. 163-174. Spitzer, N.C., Low pH selectivity blocks calcium action potentials in amphibian neurons developing in culture, Brain Research, 161 (1979) 555-559. Trzebski, A., Central pathways of the arterial chemoreceptor reflex. In H. Acker and R.G. O'Regan (Eds.), Physiology of the Peripheral Arterial Chemoreceptors, Elsevier, Amsterdam, 1983, pp. 431-461. Viala, D. and Freton, E., Evidence for respiratory and locomotor pattern generators in the rabbit cervico-thoracic cord and for their interactions, Exp. Brain Res., 49 (1983) 247-256. Wagman, I.H. and Price, D.D., Responses of dorsal horn cells of M. mulatta to cutaneous and sural nerve A and C fiber stimuli, J. Neurophysiol., 32 (1969) 803-817. Walker, J.L., Jr. and Brown, A.M., Unified account of the variable effects of carbon dioxide on nerve cells, Science, 167 (1970) 1502-1504. White, S.R. and Neuman, R.S., Facilitation of spinal motoneurone excitability by 5-hydroxytryptamine and noradrenaline, Brain Research, 188 (1980) 119-127. Zhang, T.-X., Rohlicek, Ch.V. and Polosa, C., Responses of sympathetic preganglionic neurons to systemic hypercapnia in acute spinal cat, J. Aut. Nerv. Syst., 6 (1982) 381-389.
296
Elst~vicl BRE 12122
Decreased Excitability of Respiratory Motoneurons During Hypercapnia in the Acute Spinal Cat JOZEF S. JODKOWSKI and JANUSZ LIPSKI*
Department of Physiology, Institute of Physiological Sciences, Medical Academy, Warsaw (Poland) (Accepted 1 April 1986)
Key words: Carbon dioxide - - Phrenic motoneuron - - Intercostal motoneuron - - Cat
This study was undertaken to examine the effects of hypercapnia on the excitability of respiratory motoneurons. The action of CO 2 on phrenic (inspiratory) and internal intercostal (expiratory) motoneurons was compared with that exerted on non-respiratory motoneurons of the musculocutaneous nerve. The experiments were performed on spinalized (C1 segment), partially deafferented cats that were exposed to different CO2/O 2 mixtures (end-tidal CO: 3 + 0.3, 6 + 0.5 and 9 + 0.5%). Changes in neuronal excitability were assessed by: (1) measuring the amplitudes of antidromic field potentials recorded from a population of motoneurons; (2) analysis of the amplitude and latency of the orthodromic response recorded from a given nerve and evoked by microstimulation within the corresponding motor nucleus; (3) monitoring the membrane potentials during intracellular recordings from phrenic motoneurons; and (4) recording ongoing activity of the phrenic and internal intercostal nerves. Hypercapnia (end-tidal CO z 6 _+ 0.5 or 9 + 0.5%) decreased the excitability of phrenic and musculocutaneous motoneurons, the effect being larger at the higher CO 2 level. Internal intercostal motoneurons were generally more resistant to the effects of CO 2. A depression of their excitability was observed only at end-tidal CO: 9 _+0.5 %. The decreased excitability of phrenic motoneurons was associated with membrane hyperpolarization. It is concluded that the depressant action of CO 2 is present in both respiratory and non-respiratory spinal motoneurons. The action of hypercapnia on respiratory motoneurons may oppose the excitatory effects exerted through specific chemoreflexes.
INTRODUCTION
In most of these e x p e r i m e n t s h y p e r c a p n i a was used t o g e t h e r with h y p e r o x i a , t h e r e b y significantly
H y p e r c a p n i a is k n o w n to i n f l u e n c e the activity of various n e u r o n s in the c e n t r a l n e r v o u s system t h r o u g h its central action, apart f r o m the k n o w n excitatory reflex effects o n respiratory a n d a u t o n o m i c n e u r o n s e v o k e d t h r o u g h excitation of p e r i p h e r a l chem o r e c e p t o r s (for ref., see ref. 45). H i g h levels of car-
r e d u c i n g the reflex action of CO2 t h r o u g h excitation of p e r i p h e r a l c h e m o r e c e p t o r s ~7. It is, h o w e v e r , diffi-
b o n dioxide decrease the excitability of cortical n e u rons 2°~27, spinal a m o t o n e u r o n s (refs. 3, 6, 39; b u t see ref. 27), a n d inhibit afferent t r a n s m i s s i o n in the dorsal c o l u m n - l e m n i s c a l system 35'36. T h e activity of spinal i n t e r n e u r o n s is i n c r e a s e d or d e c r e a s e d 6'7'43"47, a n d similar variability in r e s p o n s e has b e e n o b s e r v e d in recordings from n e u r o n s in the v e n t r o l a t e r a l medulla (ref 40, cf. also ref. 18).
cult to establish w h e t h e r the C O 2, a p p l i e d in a n i m a l s with intact central n e r v o u s system, acts t h r o u g h specific m e d u l l a r y c h e m o r e c e p t o r s 42 or directly, i.e. without b e i n g m e d i a t e d by n e u r o n s i n v o l v e d in central c h e m o r e c e p t o r pathways. In fact, ' m i x e d ' direct a n d reflex effects have b e e n o b s e r v e d in s o m e n e u rons. F o r e x a m p l e , in respiratory m e d u l l a r y n e u r o n s , h y p e r c a p n i a causes a n increase in s y n a p t i c excitation due to activation of c h e m o r e c e p t o r s , as well as m e m b r a n e h y p e r p o l a r i z a t i o n ascribed to a direct action of C O 2 (ref. 33). A direct effect of h y p e r c a p n i a o n n e u r o n a l firing,
* Present address: Experimental Neurology Unit, John Curtin School of Medical Research, Australian National University, Canberra, A.C.T. 2601, Australia. Correspondence: J.S. Jodkowski. Present address: Department of Physiology and Piophysics, SJ-40, University of Washington School of Medicine, Seattle, WA 98195, U.S.A. 0006-8993/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
297 intercellular coupling or synaptic processes has been demonstrated in brain slices in vitro (e.g. refs. 10, 18), in cerebral cortex isolated from underlying structures 2°'27 and in spinal animals 5'7'15'16'25'26'5°. In acute spinal cats, a depressive action of systemic hypercapnia is usually observed 3,7'15,16,25,39, although excitation 5° has also been reported. The present study was undertaken to assess whether CO 2 (range 3-9.5%, end-tidal concentration) modifies directly the activity of motoneurons controlling respiratory muscles. The reflex effects of CO 2 through arterial chemoreceptors and central CO 2 action through chemosensitive neurons were excluded by high cervical spinalization. We have analyzed the excitability of phrenic (inspiratory) and internal intercostal (expiratory) motoneurons, and some results have been compared with those from motoneurons not directly involved in the control of respiratory movements. Preliminary results have been briefly reported 24. MATERIALS AND METHODS
General procedures Experiments were performed on 27 cats of either sex (2.1-4.0 kg) anesthetized with a-chloralose and urethane (40 and 260 mg/kg i.v., respectively). A tracheostomy was made. Blood pressure was continuously recorded from the femoral artery (P23Db, Statham). Rectal temperature was monitored and held at 37-38 °C. The C5 and C6 branches of the left phrenic nerve, the left musculocutaneous nerve (innervating biceps brachii, brachialis and coracobrachialis muscles) and the left internal intercostal nerve (T8) were exposed, cut distally and prepared for standard recording and/or stimulation. The animals were placed in a stereotaxic frame with pelvic pins and vertebral clamps at T1, T6 and T10, paralyzed with gallamine (Tricuran, Germed, 5 mg/kg/h, i.v.) and artificially ventilated with oxygenenriched air. A bilateral pneumothorax was performed and an expiratory load of 2 cm H20 imposed. Tracheal pressure was recorded continuously and end-tidal CO 2 concentration (ET CO2) monitored (LB-2, Beckman). A laminectomy was performed that extended from C3 to C7 segments (to allow recording from phrenic and musculocutaneous nerve motoneurons) and/or from T7 to T9 (for recordings
from internal intercostal motoneurons). Left dorsal roots were cut in the exposed segments. The surface of the spinal cord was covered with warm paraffin oil.
Spinalization From a midline incision the muscles of the neck were retracted to either side. The atlanto-occipital membrane was cut and vessels on the caudal portion of the medullary surface were coagulated. The spinal cord was transected just above the dorsal roots of the first cervical segment using fine scissors. Incisions were made in several steps at intervals of 3-5 min; completeness of the transection was carefully checked, and bleeding was controlled with Gelfoam. If necessary, Dextran or norepinephrine (or both) were infused intravenously to maintain a mean arterial blood pressure above 80 mm Hg. Recordings were started at a minimum of 2 h after spinalization.
Stimulation and recording The dissected nerves were stimulated with bipolar silver wire electrodes using constant-current pulses of 0.2 ms (NL 510 and NL 800, Digitimer). The stimulus amplitude was 4 times threshold value required to evoke motoneuronal antidromic field potentials. Stimulus frequency was 2 Hz, using pairs of pulses presented at an interval of 2-30 ms. Antidromic field potentials were recorded with glass microelectrodes filled with 0.5 M NaCI (resistance approximately 1 Mfl), amplified (NL 104, Digitimer), averaged over 16-64 sweeps (Anops 105, Warsaw Politechnic) and photographed. Depending on the position of the recording electrode within the ventral horn, either antidromic field potentials generated by axons or cell bodies of motoneurons were recorded. Because the axonal refractory period is shorter than that of the soma (e.g. ref. 12), field potentials unchanged by stimuli presented 2-3 ms apart were presumed to originate from axons. Mainly somatic antidromic field potentials (i.e. potentials with a clear amplitude decrease following the second stimulus of a pair, presented with a delay of 5-30 ms) were analyzed in this study (see Fig. 1). Glass microelectrodes filled with 3 M potassium citrate (resistance 10-15 Mr2) were used for intracellular recording from phrenic motoneurons (NL 102, Digitimer). The motoneurons were identified by recording antidromic action potentials following stimuli applied to the correspond-
298 ing (C5 or C6) Phrenic branch (cf. ref. 30). Glass-insulated tungsten electrodes (NL 05, Digitimer) were used for monopolar microstimulation within the ventral horn. The electrodes were placed at a point where maximal somatic antidromic field potentials could be evoked by stimulation (see above) of the selected nerves. At such loci, constantcurrent pulses (0.1-0.2 ms, 17-70 ~ A ) were applied, and orthodromic potentials evoked in the corresponding nerves were amplified and averaged. Ongoing activity present in some of the dissected nerves was amplified and processed by an RC integrator (time constant, 100-500 ms). The output signals were recorded on a chart recorder.
Experimental protocol At the beginning of each experiment, frequency and volume of the respirator were adjusted to obtain E T CO 2 of 3 _+ 0.3%. Changes in expired CO 2 level were accomplished by adding a 15% CO2/85% O 2 gas mixture to the inlet of the respirator. Desired levels of CO 2 were obtained within approximately 5 min. Recordings were made at E T CO 2 levels of 3 +_ 0.3% (control), 6 + 0.5% and 9 + 0.5%. C O 2 administration was then terminated and the last recording was made 20 min later. A series of 4 such recordings was accepted for further analysis only when there was no difference (within + 5 % ) between the first (control) and the last (recovery) records. The following parameters were analyzed: amplitudes and latencies of the evoked potentials, m e m b r a n e potential during intracellular recordings, and ongoing activity recorded from dissected nerves. Amplitudes of the evoked responses were measured from baseline level to peak of the main negative deflection. Student's paired t-test was used for statistical analysis. RESULTS
Phrenic motoneurons The influence of CO2 on somatic antidromic field potentials was analyzed in 15 series of experiments performed on 12 cats. Control amplitudes ranged from 0.3 to 1.3 mV (mean 0.6 mV) and their latencies from 0.42 to 1.37 ms (mean 0.84 ms). The field potential was depressed from control by 15.5 _ 4.2% (mean + S.E.M.; P < 0.001) at E T CO 2 6 + 0.5%
and by 30.6 _+ 6.1% (P < 0.001) at ET CO 2 9 + 0.5% (Fig. 1). The difference between depression at the two hypercapnic levels was statistically significant (P < 0.001). No changes in antidromic latency were observed. In 4 recording positions, effect of CO~ was tested on antidromic field potentials recorded from axons (see Methods). No significant effect was observed with E T CO 2 up to 9.5%. Orthodromic responses recorded from C5 or C6 phrenic branches following microstimulation within the corresponding section of the phrenic nucleus were studied in 3 cats. In 6 of 7 series of measurements, a depression of the evoked potentials was observed and this was sometimes (4 of 6 cases) associated with an increased onset latency (Fig. 2). These data are presented in Table I. Two hours after spinalization, ongoing tonic activity in the C5 phrenic branch was observed in 9 of 27 cats (33%). The number of active fibers varied. In 5 of the experiments in which there was tonic activity no clear change in firing occurred during CO2 administration (at either 6 _+ 0.5% or 9 + 0.5% level). In the remaining 4 experiments, a decrease in activity
A
B
t
I "---7 "~ , F
2
4
3.5%
I " 'k~-
5.9%
P0-
,-',,__~ A
9.0%
ET COz (%) 3±03
6±0.5
9~05
,' i" : f -
5
A" ~
0
P
20-
3.0% Ims
30-
-]lmV 40%
Fig. 1. Influence of CO 2on antidromic field potentials recorded in phrenic nucleus. A: examples of records in a representative experiment. C5 phrenic branch was stimulated with a pair of stimuli (64/~A, 0.2 ms) presented at interval of 28 ms, Stimulus artifacts are marked by arrowheads. Potentials averaged (16 sweeps). Negativity upwards. 1, control record; 2 and 3, records during CO2 administration; 4, 20 rain after CO2 withdrawal. B: influence of CO 2on peak amplitudes of antidromic field potentials in phrenic nucleus (15 experiments): Each column represents mean difference (_+ S.E.M.) between control amplitude and amplitude at a given CO 2 level expressed in percent of control value. Data for ET CO 2 of 3 __-0.3% obtained by comparing control amplitudes with amplitudes recorded after CO 2 withdrawal. *, different from control (P < 0.001). **, different from control level (P < 0.001) and from ET CO2 6 _+0.5% level (P < o.ool).
299 ETCO • •
-•
2
(%)
2'7
r n s
"j
--
5-7
"
8-8
L "
,,,,d'~
slight hyperpolarization (about 1 mV) and decreased excitability, as reflected by blockade of the IS-SD spike (Fig. 4). In the 18 other recordings, CO 2 administration was accompanied by a progressive depolarization, changes in spike shape (broadening, decreased amplitude) and failure of antidromic activation. These changes were never reversible after CO 2 withdrawal and, in many cases, recordings were lost during CO2 administration (sudden drop in measured intracellular membrane potential to a level close to 0 mV).
2.7
7 Fig. 2. Influence of CO 2 on orthodromic potentials recorded and averaged (n = 16) from C5 phrenic branch. Stimulus (17 /tA, 0.2 ms; threshold intensity, 9/~A) onset indicated by arrow. Negativity downwards in all records.
was seen, followed by recovery after CO 2 withdrawal (Fig. 3). Intracellular recordings were attempted from phrenic motoneurons in 9 spinalized animals. A recording was considered acceptable if the impaled motoneuron had a stable membrane potential more negative than -50 mV for at least 5 min and a spike potential of 60 mV or more. Twenty-one motoneurons initially met these criteria. For analysis of their responses to CO 2, however, we accepted recordings only from those motoneurons in which the membrane potential returned to control level after CO2 withdrawal (see Methods, experimental protocol). Only 3 motoneurons (3 different experiments) met these criteria. In these 3 cells, increased level of CO 2 led to
Internal intercostal motoneurons A series of 11 recordings was made in 7 cats. Control amplitudes of antidromic field potentials ranged from 0.18 to 0.83 mV (mean 0.43 mV) and their onset latencies from 0.34 to 1.1 ms (mean 0.53 ms). Moderate hypercapnia (ET CO 2 6 + 0.5%) did not significantly change the amplitudes and latencies of these potentials• At an ET CO2 of 9 + 0.5%, however, the amplitudes of the potentials decreased from control values by 11.3 _+ 3.1% (P < 0.001) with no change in the antidromic latency (Fig. 5A). Orthodromic potentials were evoked in the internal intercostal nerves by microstimulation within the internal intercostal nucleus in 5 animals. Amplitudes of the evoked potentials ranged from 28 to 198/~V (mean 117/~V, n = 12) and their latencies from 0.25 to 0.57 ms (mean 0.43 ms). During CO 2 administration, a depression of the amplitude was observed in only 3 of 12 series of measurements (cf. Fig. 5B). Onset latencies of the responses were not changed in any of these experiments• Following spinalization, ongoing tonic activity of
TABLE I Amplitudes and latencies o f orthodromic potentials evoked in phrenic nerve following microstimulation within the phrenic nucleus Measure-
Stimulus
Control
ment
(/zA)
E T C O 2 3+_0.3%
E T C O 2 6+_0.5%
ETCO 2 9+0.5%
ETCO 2 3+0.3%
Amplitude
Latency
Amplitude
Latency
Amplitude
Latency
Amplitude
(~V)
(ms)
(l~V)
(ms)
(~V)
(ms)
(~V)
(ms)
160 146 260 250 630 212 328
0.71 0.79 0.88 0.80 0.56 0.42 0.46
160 150 90 30 740 212 320
0.79 0.79 1.12 1.64 0.64 0.46 0.46
96 106 10 30 630 168 312
0.83 0.83 1.24 1.64 0.60 0.46 0.46
160 138 260 250 640 232 320
0.71 0.75 0.80 0.76 0.60 0.46 0.46
1 2 3 4 5 6 7
51 51 16.6 21.5 25 30 30
CO 2 withdrawal
Latency
300
A
IPh
03-8
~
03.8
e5.3
%.3
e6.9
e6.?
IT 8
@ 1 min
B
j•,•.•tLat_..L,_, ..... i . . . . e9"2
Li
a,..~
.... ~ _
,IIL~.
e9-4 A
e3-8
Fig. 3. Influence of CO 2 on integrated ongoing activity of C5 phrenic branch (IPh) and T8 internal intercostal nerve (ITs). Actual ET CO 2 values (non-steady-state) as indicated. Arrowhead in A indicates beginning of CO 2 administration. B: 2 rain after A. Arrowhead indicates end of CO 2 administration. C: 7 min after B. In A and C, middle line is zero reference for phrenic integrated activity.
A
C
B
m¥ -I10 •
l
I
A
J
A I
•
II
!:
,!i
~:
i!
5ms 0
E
D
F
mY -80
~ m A
L A
A
•
Fig. 4. Intracellular recordings from a phrenic motoneuron. Antidromic activation with a pair of stimuli (20 ~A, 0.2 ms) presented at interval of 8.5 ms. Stimulus artifact indicated by the arrowheads. Upper traces: intracellular records. Lower traces: differentiated records (r = 2.2/~s) of the intracellular spikes. Five superimposed sweeps in each trace. A: control (membrane potential, M P - 8 0 mV, ET CO 2 3.3%). B: 3 min after beginning of CO~ administration (MP -80 mV, ET CO 2 4.3%). C: 4 min after beginning of CO 2 administration (MP - 8 0 mV, ET CO 2 4.5%). D: 10 min after beginning of CO 2 administration (MP -81 mV, ET CO 2 5.3%). E: 5 min after CO~ withdrawal (MP -80 mV, ET CO 2 3.5%). F: 20 min after CO 2 withdrawal (MP -78 mV, ET CO 2 3.2%).
301
A
ETC02 (%) 3 + 0-3
Musculocutaneous nerve motoneurons
6 +0'5
9 + 0"5
%
B
3.0%
6.5%
i]
~'::
8.6%
2.9%
~'
ii/ 0.1 mV I
Fig. 5. Influence of CO2 on excitability of internal intercostal motoneurons. A: effect on amplitude of antidromic field potentials recorded in intercostal nucleus (11 experiments). *, different from control level (P < 0.001). For further explanation, see Fig. lB. B: effects on orthodromic potentials recorded from internal intercostal nerve. Arrowheads indicate stimulus (60 ~A, 0.1 ms, threshold current 23 ~A). Last record (ET CO 2 2.9%) taken 20 min after CO2 withdrawal. All traces are averages of 16 sweeps.
the internal intercostal nerve was present in 11 cats. Carbon dioxide caused a depression in only one experiment (Fig. 3), whereas in no change was observed. Nevertheless, in 3 experiments activity recorded from the C5 branch was depressed.
7 out of of firing 6 others of these phrenic
ET CO2 (%) 3+_0.3
o I_Tj
6+_0.5
The influence of CO 2 on antidromic field potentials evoked by stimulation of the musculocutaneous nerve was analyzed in 5 experiments (3 cats). Control amplitudes ranged from 0.51 to 2.1 m V (mean 1.07 mV) and onset latencies from 0.38 to 0.7 ms (mean 0.51 ms). At E T CO 2 6 +_ 0.5%, amplitudes of the potentials were decreased by 12.9 + 4.4% from control values (P < 0.01) and at E T C O 2 9 + 0.5% by 22.5 + 5% (P < 0.001) (Fig. 6). The difference in depression between the two hypercapnic levels was statistically significant (P < 0.015). No change in antidromic latency of any of these potentials was observed.
9+_0.5
10
Z0
30 %
Fig. 6. Influence of CO 2 on amplitude of antidromic field potential recorded from motoneurons of musculocutaneous nerve. *, different from control level (P < 0.01); **, different from control level (P < 0.01)1) and from ET CO 2 6 + 0.5% level (P < 0.015). For further explanation, see Fig. lB.
DISCUSSION Several authors have reported that profound hypercapnia ( > 10% C O 2 in inhaled air) depresses the excitability of lumbar motoneurons 6J6'39'43. Effects of lower C O 2 concentrations were less clear or not examined 3. Results of the present study show that CO2 depresses the excitability of inspiratory (phrenic) motoneurons to a degree similar to that observed in motoneurons having no apparent respiratory function. This effect was often seen at E T CO2 6 + 0.5%, and became more prominent at a higher C O 2 level (ET CO 2 9 + 0.5%). Although the effects were analyzed at only 3 different CO2 concentrations, our data indicate that the action of carbon dioxide is dose-dependent, at least in the range of 2 . 7 - 9 . 5 % . The conclusions of this report are based mainly on results obtained with two qualitative methods: one involving antidromic activation and the other orthodromic activation of motoneurons. With respect to the antidromic method, it was assumed that following supramaximal stimulation of the motor nerve, the amplitude of the antidromic field potential recorded within a m o t o r nucleus depended on the number of antidromically invaded cell bodies in the vicinity of the tip of the recording microelectrode 38. A n y factor changing the neuronal excitability (membrane potential and/or m e m b r a n e conductance) changes the proportion of antidromically excited neurons. Thus, a decrease or an increase in the amplitude of the antidromic field potential reflects, respectively, a decrease or increase in excitability of the studied motoneuron (cf., e.g. refs. 2, 34, 49). The 'orthodromic'
302 method involved microstimulation in the spinal cord, in the vicinity of phrenic or intercostal motor nuclei. Amplitude of the orthodromically evoked potentials in the corresponding nerves depends on the number of excited cell bodies. Any factor modifying excitability of the motoneurons changes the proportion of activated neurons, as fewer of more cells are excited at a certain distance from the centre of the stimulated region (where the stimulating current is close to threshold; e.g. refs. 13, 36, 37). Along with measurements of the amplitudes of antidromic field potentials, the depressive action of CO 2 was observed in 13 of 15 recording sites, at ET CO 2 6 + 0.5% and 9 + 0.5%. This result was generally confirmed in recordings of the amplitude of orthodromically evoked potentials in the phrenic nerve. The 'orthodromic' method, however, proved to be somewhat less sensitive in detecting a weak influence, as with 6 _+ 0.5% CO 2 the depression was observed in only two loci. With ET CO2 9 __ 0.5%, the decreased amplitude was observed in 6 of 7 experiments. The control latency of phrenic orthodromic responses did not exceed 0.9 ms, indicating direct stimulation of the motoneuronal cell bodies. Latencies were increased in 3 of 7 experiments with ET C O 2 6 + 0.5% and in 4 of 7 experiments with ET CO 2 9 _ 0.5%. In two of the latter experiments, a marked (> 0.35 ms) increase in latencies was observed. A likely explanation is that CO 2 first depressed the excitability of neurons whose axons had the highest conduction velocities. Another explanation (for the two most pronounced changes in latency) is based on the assumption that microstimulation within the spinal cord activates not only motoneurons, but also some excitatory interneurons and/or neuronal pathways u. When the excitability of motoneurons is decreased by CO 2, they may no longer be activated directly by the current. However, some of them could still be activated through a synaptic action, with an increase in the observed latency due to the synaptic delay (cf. ref. 22). Intracellular recordings from apparently undamaged phrenic motoneurons showed that decreased neuronal excitability accompanying hypercapnia was associated with membrane hyperpolarization. Such an influence was postulated by Gill and Kuno 21, but no direct evidence was presented. Both hyperpolarizing 6,16'39'43 and depolarizing 16'39 effects have been
reported in other studies with intracellular recordings from spinal non-respiratory motoneurons. Because in our experiments frequent depolarizations were observed that were associated with irreversible cell damage, it is probable that the depolarization is an artifact rather than a true C O 2 effect. This conclusion is supported by results of a recent study 2s in which inhibition of phrenic nerve discharges was observed following CO 2 administration in animals whose central excitatory chemoreceptor pathway was depressed by baclofen. Unfortunately, the small sample size in our experiments (only 3 motoneurons) does not allow for a definitive conclusion on this point. Carbon dioxide in concentrations of 10-20% was reported to depress transmission in nerve fibers s'16'36, The axonal effect did not occur in our experimental conditions: the M-spike 12 could still be observed in phrenic motoneurons with blocked IS-SD spike, there was no depression of 'axonal' field potentials, and the latencies of antidromic 'somatic' field potentials were not affected. Sensitivity to CO2, however, was not a uniform feature of all subpopulations of spinal motoneurons under study. Internal intercostal motoneurons were less affected than phrenic or musculocutaneous motoneurons. Occasionally, we observed that one field potential as compared to another recorded within the same motor nucleus was more resistant to the action of CO 2. Such a differential sensitivity was also observed with recordings of ongoing activity from phrenic and internal intercostal nerves. Following CO 2 administration, decreased firing frequency was observed in 4 of 9 C5 phrenic branches and in only one of 7 recordings from the internal intercostal nerves. It remains to be established whether these differences are due to a differential, direct CO: action on motoneurons or on premotor interneurons. In our experiments, only tonic, 'spontaneous' activity was observed in the C5 phrenic branch after spinalization (cf. refs. ll, 14). Several authors 1'9'a6 have observed phasic (respiratory-like) phrenic discharge under similar conditions. Since drugs (e.g. Doxapram, Nialamide, L-DOPA) were often used in these studies, the origin of this type of activity is not clear. It could be partly the result of spinal reflexes (cf. refs. 11, 14). Viala and Freton 46 reported an excitatory CO 2 effect on 'spinal' phasic activity of
303 the phrenic nerve. Although this reponse can be mediated by some interneurons that are excited by CO2 (ref. 43), our results do not support the hypothesis of a chemosensitive spinal respiratory generator. The mode of the so-called 'direct' C O 2 action is not clear. Studying neuronal effects in Aplysia, Brown et al. 4,48 observed that elevated levels of C O 2 increase the permeability of cell membranes to chloride ions. The response was shown to result mainly from a decrease in extracellular pH. In some insects, the C O 2 action can be explained by lowering the intracellular pH, which can lead to inhibition of m e m b r a n e active transport processes (ref. 8, cf. also ref. 41). These theories, based on experiments performed on invertebrates, do not necessarily explain results obtained with mammalian neurons 16, and hence other ionic mechanisms have to be taken in account 23'29'44. Changes in neuronal activity produced by high levels of CO 2 are not always similar to those produced by low p H 18,31'32. In addition, not all effects can be explained by the action of CO2 or low p H exerted upon the neurons under investigation. The effects on synaptic transmission cannot be excluded 29. Furthermore, the observed effects of CO2/pH can be evoked by action on other neurons, which change the excitability of the examined neuronal pool 19. At present it
is difficult to differentiate between the CO2 and/or low p H action per se on cell membranes and an indirect action through other mechanisms. Thus the term 'direct action' should be used with caution. Considered in conjunction with results of previous studies 5'6'16'33'43, we conclude that elevated levels of
REFERENCES
horn convergent neurons, Exp. Neurol., 70 (1980) 109-121. 8 Clark, M.A. and Eaton, D.C., Effect of COz on neurons of the house cricket, Acheta domestica, J. Neurobiol., 14 (1983) 237-250. 9 Coglianese, C.J., Peiss, C.N. and Wurster, R.D., Rhythmic phrenic nerve activity and respiratory activity in spinal dogs, Resp. Physiol., 29 (1977) 247-254. 10 Connors, B.W., Bernardo, L.S. and Prince, D.A., Carbon dioxide sensitivity of dye coupling among glia and neurons of the neocortex, J. Neurosci., 4 (1984) 1324-1330. 11 Decima, E.E. and von Euler, C., Excitability of phrenic motoneurons to afferent input from lower intercostal nerves in spinal cat, Acta Physiol. Scand., 75 (1969) 580-591. 12 Coombs, J.S., Curtis, D.R. and Eccles, J.C., The interpretation of spike potentials of motoneurones, J. Physiol. (London), 139 (1957) 198-231. 13 Eccles, J.C., Magni, F. and Willis, W.D., Depolarization of central terminals of Group I afferent fibres from muscle, J. Physiol. (London), 160 (1962) 62-93. 14 Eldrige, F.L., Gill-Kumar, P., Miilhorn, D.E. and Waldrop, T.G., Spinal inhibition of phrenic motoneurones by stimulation of afferents from peripheral muscles, J. Physiol. (London), 311 (1981) 67-79. 15 Esplin, D.W. and Rosenstein, R., Analysis of spinal depressant actions of carbon dioxide and acetalozamide,
1 Aoki, M., Moil, S., Kawahara, K., Watanabe, H. and Ebata, N., Generation of spontaneous respiratory rhythm in high spinal cats, Brain Research, 202 (1980) 51-63. 2 Bloedel, J.R. and Roberts, W.J., Functional relationship among neurons of the cerebellar cortex in the absence of anesthesia, J. Neurophysiol., 32 (1969) 75-84. 3 Brooks, C.McC. and Eccles, J.C., A study of the effects of anaesthesia and asphyxia on the mono-synaptic pathway through the spinal cord, J. Neurophysiol., 10 (1947) 349-360. 4 Brown, A.M., Walker, J.L. and Sutton, R.B., Increased chloride conductance as the proximate cause of hydrogen ion concentration effects in Aplysia neurons, J. Gen. Physiol., 56 (1970) 559-582. 5 Carpenter, D.O., Hubbard, J.H., Humphrey, D.R., Thompson, H.K. and Marshall, W.H., Carbon dioxide effects on nerve cell function. In I.G. Nahas and K.E. Schaefer (Eds.), Carbon Dioxide and Metabolic Regulations, Spilnger-Verlag, New York, 1974, pp. 49-62. 6 Caspers, H. and Speckmann, E.J., Cerebral pO2, pCO2 and pH: changes during convulsive activity and their significance for spontaneous seizure termination, Epilepsia, 13 (1972) 699-725. 7 Catchlove, R.F.H., Le Bars, D., Blanchet, F. and Besson, J.M., Effects of respiratory acidosis on the activity of dorsal
CO2 exert a dual action on spinal respiratory motoneurons. This condition can activate these motoneurons through excitation of peripheral chemoreceptors and specific medullary chemoreceptors. At the same time, C O 2 depresses excitability of respiratory motoneurons by an action at the spinal cord. The spinal action normally is outweighed by the powerful input from brainstem respiratory centers. Nevertheless, local CO2-dependent depression of respiratory motoneurons may impair these excitatory reflex mechanisms during severe hypercapnia.
ACKNOWLEDGEMENTS The authors thank Dr. A. Berger and Dr. R. Martin-Body for their critical remarks on the manuscript and H a n n a Atkins for editing of the manuscript. This study was supported by 10.4.03.5 Grant from the Polish A c a d e m y of Sciences.
304
Arch. Int. Pharmacodyn. Ther., 143 (1963) 498-513. 16 Esplin, D.W., Capek, R. and Esplin, B.A., An intracellular study of the actions of carbon dioxide on the spinal monosynaptic pathway, Can. J. Physiol. Pharmacol., 51 (1973) 498-513. 17 Fitzgerald, R.S. and Parks, D.C., Effect of hypoxia on carotid chemoreceptor responses to carbon dioxide in cats, Resp. Physiol., 12 (1971) 218-229. 18 Fukuda, Y., Difference between actions of high pCO 2 and low [HCO3- ] on neurons in the rat medullary chemosensitive areas in vitro, Pflagers Arch., 398 (1983) 324-330. 19 Fukuda, Y. and Loeschcke, H.H., A cholinergic mechanism involved in the neuronal excitation by H ÷ in the respiratory chemosensitive structures of the ventral medulla oblongata of rats in vitro, Pfliigers Arch., 379 (1979) 125-135. 20 Gellhorn, E. and French, L.A., Carbon dioxide and cortical spike frequency, Arch. Int. Pharmacodyn. Ther., 93 (1953) 427-433. 21 Gill, P.K. and Kuno, M., Properties of phrenic motoneurons, J. Physiol. (London), 168 (1963) 258-273. 22 Gustaffson, B. and Jankowska, E., Direct and indirect activation of nerve cells by electrical pulses applied extracetlularly, J. Physiol. (London), 258 (1976) 33-61. 23 Hille, B., Charges and potentials at the nerve surface. Divalent ions and pH, J. Gen. Physiol., 51 (1968) 221-236. 24 Jodkowski, J.S. and Lipski, J., Effects of carbon dioxide on spinal respiratory motoneurons, somatic motoneurons and sympathetic preganglionic neurons in cats, Neurosci. Lett., Suppl. 14 (1983) S183. 25 Kirstein, L., Early effects of oxygen lack and carbon dioxide excess on spinal reflexes, Acta Physiol. Scand., Suppl. 80, 23 (1951). 26 Kitahata, L.M., Taub, A. and Sato, I., Hyperventilation and spinal reflexes, Anesthesiology, 31 (1969) 321-326. 27 Krnjevi6, K., Randi6, M. and Siesjo, B.K., Cortical CO 2 tension and neuronal excitability, J. Physiol. (London), 176 (1965) 105-122. 28 Lalley, P.M., Biphasic effects of baclofen on phrenic motoneurons: possible involvement of two types of gammaaminobutyric acid (GABA) receptors, J. Pharmacol. Exp. Ther., 226 (1983) 616-624. 29 Landau, E.M. and Nachshen, D.A., The interaction of pH and divalent cations at the neuromuscular junction, J. Physiol. (London), 251 (1975) 775-790. 30 Lipski, J., Kubin, L. and Jodkowski, J., Synaptic action of R~ neurons on phrenic motoneurons studied with spiketriggered averaging, Brain Research, 288 (1983) 105-118. 31 Marino, P.L. and Lamb, T.W., Effects of CO 2 and extracellular H + iontophoresis on single cell activity in the brainstem, J. Appl. Physiol., 38 (1975) 688-695. 32 Miles, R., Does low pH stimulate central chemoreceptors located near the ventral medullary surface?, Brain Research, 271 (1983) 349-353. 33 Mitchell, R.A. and Herbert, D.A., The effect of carbon dioxide on the membrane potential of medullary respiratory neurons, Brain Research, 75 (1974) 345-349. 34 Morales, F. and Chase, M.H., Postsynaptic control of lure-
35
36
37
38
39
40
41 42
43
44
45
46
47
48
49 50
bar motoneuron excitability during active sleep in the chronic cat, Brain Research, 225 (1981) 279-295. Morris, M.E., The action of carbon dioxide on afferent transmission in the dorsal column-lemniscal system, J. Physiol. (London), 218 (1971) 651-669. Morris, M.E., The action of carbon dioxide on synaptic transmission in the cuneate nucleus, J. Physiol. (London), 218 (1971) 671-689. Morrison, A.R. and Pompeiano, O. An analysis of the supraspinal influences acting on motoneurons during sleep in the unrestrained cat. Responses of the alpha motoneurons to direct electrical stimulation during sleep, Arch. Ital. Biol., 103 (1965) 497-516. Nicholson, C., Generation and analysis of extracellular field potentials. In Electrophysiological Techniques, Society for Neuroscience, Atlanta, GA, 1979, pp. 93-147. Papajewski, W., Klee, M.R. and Wagener, A., The action of raised CO 2 pressure on the excitability of spinal motoneurones, Electroencephalogr. Clin. Neurophysiot., 27 (1969) 618. Pokorski, M., Neurophysiological studies on central chemosensor in medullary ventrolateral areas, Am. J. Physiol., 230 (1976) 1288-1295. Roos, A., Boron, W.F., Intracellulat pH, Physiol. Rev., 61 (1981) 296-434. Schlaefke, M.E., Central chemosensitivity: a respiratory drive, Rev. Physiol. Biochem. Pharmacol., 90 (1981) 171-244. Speckmann, E.-J. and Caspers, H., Responses of spinal and cortical neurons to changes of pO 2 and pCOz in blood and tissue. In M.J. Purves (Ed.), The Peripheral Arterial Chemoreceptors, Cambridge University Press, Oxford, 1975, pp. 163-174. Spitzer, N.C., Low pH selectivity blocks calcium action potentials in amphibian neurons developing in culture, Brain Research, 161 (1979) 555-559. Trzebski, A., Central pathways of the arterial chemoreceptor reflex. In H. Acker and R.G. O'Regan (Eds.), Physiology of the Peripheral Arterial Chemoreceptors, Elsevier, Amsterdam, 1983, pp. 431-461. Viala, D. and Freton, E., Evidence for respiratory and locomotor pattern generators in the rabbit cervico-thoracic cord and for their interactions, Exp. Brain Res., 49 (1983) 247-256. Wagman, I.H. and Price, D.D., Responses of dorsal horn cells of M. mulatta to cutaneous and sural nerve A and C fiber stimuli, J. Neurophysiol., 32 (1969) 803-817. Walker, J.L., Jr. and Brown, A.M., Unified account of the variable effects of carbon dioxide on nerve cells, Science, 167 (1970) 1502-1504. White, S.R. and Neuman, R.S., Facilitation of spinal motoneurone excitability by 5-hydroxytryptamine and noradrenaline, Brain Research, 188 (1980) 119-127. Zhang, T.-X., Rohlicek, Ch.V. and Polosa, C., Responses of sympathetic preganglionic neurons to systemic hypercapnia in acute spinal cat, J. Aut. Nerv. Syst., 6 (1982) 381-389.