- Email: [email protected]
Rhythmic discharges in the perfused isolated brainstem preparation of adult guinea pig
Neuroscwnce Letters, 101 (1989) 57~1
57
Elscvler Scientific Pubhshers Ireland Ltd
NSL 06108
Rhythmic discharges in the perfused isolated brainstem preparation of adult guinea pig M.P. Morin-Surun and M. Denavit-Saubi6 Laboratolre de Physwlogte Nerveuse, C N R S , Gif-sur- Yveltc (France) (Received 14 December 1988, Revlscd version received 7 February 1989, Accepted l0 February 1989)
Key words
Rhythmic activity; Isolated brainstem preparation; Adult, Gmnea pig, Nucleus tractus sohtarn, Paraglgantocellular reticular nucleus, Ambiguus nucleus
An isolated brainstem preparation of adult guinea pig was used for an m vitro electrophyslologlcal study of rhythmic neuronal &scharge patterns More than half of the spontaneously actave neurons (40/71) exhibited a rhythmic discharge. Rhythmic discharges were recorded dorsally in the nucleus tractus sohtanus and ventrally in the amblguus and paraglgantoceUular reticular nuclei Three types of rhythmic patterns of discharge were identified in these areas repetitive single discharge, repetitive bursting discharge and spontaneous periodic discharge No rhythmic patterns were recorded m other explored parts of the brainstem Comparison of these data with those from bramstem shces shows that spontaneous periodic discharges may require extended networks within the bramstem
Neuronal networks localized m the brainstem control in vivo various rhythmic functions such as respiratory [6] and cardiovascular activities [10]. Spontaneous rhythmic neuronal discharges, consisting of trains of spikes separated by silent penods, have been recorded in different brainstem nuclei. Correlation with a rhythmic function has been established from relationships of instantaneous discharge rate to the periodic output of the function [1]. Brainstem slice preparations allowed the study of synaptic and intnnsic neuronal properties withm nuclei imphcated in these rhythmic functions [2]. However, in such preparations which include a small part of the neuronal network involved in the in vivo generation of periodic trains of spikes, the spontaneous rhythmic activity recorded is only a repetitive discharge, consisting of single spikes separated by relatively constant intervals [2]. In order to test whether peripheral or medullary inputs are required to obtain perlO&C discharges, we developed the preparation of perfused isolated brainstem of adult guinea pig in which complex neural function is preserved [7]. In the present study, we show that rhythmic patterns described both in vivo and in vitro are recorded in this preparation in the same brainstem nuclei. Fifteen guinea pigs (160-240 g) were anesthetized with sodium pentobarbltal (60 Correspondence M_P Morin-Surun, Laboratolre de Physiology Nerveuse, C.N~R.S, 91190 Gif-surYvette, France_ 0304-3940/89/$ 03 50 © 1989 Elsevier Scientific Pubhshers Ireland Ltd
58 mg/kg, l.p ) and decapitated at the level of the first cervical vertebra Brainstem and cerebellum were rapidly removed from the cranium. During this surgical procedure, cold (4°C) Ringer's solution was dripped onto exposed brain surfaces The Ringer's solution consisted o f ( i n mM) NaCI 124, KCI 5, MgSO4 2 2, KH2PO4 1 2, NaHCOa 26, CaCI2 2, glucose 10 The bralnstem was maintained by pros, the ventral surface upwards, in a recording chamber filled with room temperature Ringer's saturated with 95% 02, 5% CO2 (pH 7 2) The basilar artery was rapidly cannulated via a catheter inserted at the level of the pons, and the vascular system was perfused with saturated Ringer's at room temperature using a Gllson pump. Ends of arteries at the cerwcal first segment of the spinal cord were ligated. Perfuslon pressures [7] of 50- 70 mm Hg correspond to a flow of 1 5-2 ml/min. The chamber was also perfused with the same medmm by gravity flow. The temperature of the Ringer's perfuslng the vascular system and the tissue gradually rise to 27°C in 1 h. Single-unit recordings were initiated after temperature stabilization and intentionally interrupted after 30 mln. Preparations were recorded dunng 5-10 h Multi-barrel microp~pettes which penetrated through the ventral surface were filled with: NaC1 (3 M) for recording and Methyl blue for histological identification of recording sites. The lntersplke interval histograms were bmlt up using a computer and mean values + S.D. of the lntersplke interval were calculated for each cell Seventy-one spontaneously active neurons were recorded m the isolated brainstems. Forty neurons exhibited a rhythmic pattern of discharge. All of them were located either in the nucleus tractus solitarius (NTS) (n = 16), in the paragigantocellular reticular nucleus (PGi) (n = 16) and in the ambiguus nucleus (AMB) (n = 8). Three types of rhythmic activities were distinguished by their pattern: the repetitive single discharge, the repetitive bursting discharge and the periodic discharge as defined below. These different types of rhythmic discharges could be recorded in the same nucleus from the same preparation. Likewise, the same type of discharge could be recorded in different nuclei of the same preparation. A The pattern of repetitive single discharge (n = 22, Fig. 1A) was composed of single spikes separated by relatively constant intervals. Fig. 2A (left) illustrates the spike interval histogram of one of these cells. For this category of cells, the interval between spikes was in a range of 60-450 ms (195+ 108 ms) (Fig 2A, right), giving a firing frequency of about 2 16 splkes/s. The S.D of the intersplke interval for all cells was within 5-20% of the mean. These neurons were recorded in the ventral part of NTS (n= 10), In AMB ( n = 4 ) and in PGI ( n = 8 ) (Fig IA). No relation was observed between the different firing frequencies of these neurons and the recording site, e.g. a cell firing with a frequency of 3 spikes/s was recorded in the vicinity o f a cell firing at l0 spikes/s B. The pattern of repetitive bursting discharge (n = I l, Fig. 1B), consisted of bursts (2-5 spikes) separated by relatively constant intervals. The intersplke intervals hrstogram (Fig. 2B, left) shows two populations of intervals: the first population (truncated) corresponds to the intersplke intervals within bursts and the second corresponds to the intervals between bursts The intervals between bursts were distributed in a range of 100-400 ms (216+88 ms) (Fig. 2B right) gavlng a burst frequency of
59
A
II LI L,,,t.L,I 1 I l I I I 1 s
1 ms
B
1 s
5 ms
1 s
20 ms
C
m
Fig 1 Different types of rhythmic activities, recorded In the Isolated bramstem preparation For each type extracellular recording at slow sweep, fast sweep and anatomical locahzatlon projected on the plane which corresponds to the most rostral recording A: repeUtwe discharge B repet~tlvebursting C peno&c Amb, nucleus (n) amblguus, Cu, n cuneatus, Io, inferior ohve, Mve, me&al vestibular n_, PGI, paragqgantocellular reticular n, PrH, preposltus hypoglossal n, Py, pyramidal tract, SP5, spmal tngemlnal n, SPV, spinal vestibular n, TS, tractus sohtanus 3-10 bursts/s. The S.D. o f interburst intervals for all these cells was between 8 and 20% o f the mean. This repet~ttve bursting acttvtty was usually recorded m the ventral part of the bramstem, in the P G t and m the A M B (n = 10), c o m p a r e d to n = 1 m the N T S (Fig. 1B). In some cases (n = 3), during a recording period, a burst was replaced by a single spike lndicatmg that the repetitive burstmg and repetitive single discharges are v a n a t t o n o f a same basic pattern. C. The pattern o f periodic discharge (n = 7, Fig. 1C) was characterized by a train o f spikes (5-20 sptkes) with a decreasing firing frequency, separated by long and relatively constant intervals. Depending on the cell, the train o f spikes lasted 100-800 ms (570+__ 110 ms). Interspike interval histograms show two populations (Fig. 2C, left) The first one (truncated) corresponds to the intervals wtthm the trams o f spikes and the second one to the intervals between the trains. These long intervals could
60
INTERSPIKE INTERVALS
MEANS OF INTERSPIKE INTERVALS DISTRIBUTION OF n CELLS)
cells
n :1ooo
A
200 1 1001
i
I
0
200
I
B
I
I
I
50
t°°1
,I,I I
100
200 300400
1+o1 ~
w
l
~'
i
I
200
5O
30-p
n
r~
10-
I
=
ms
n= 11
200 1
l
JII I
100
IIIII I
|
I I
I
200 300400
, o 0 0
,
I
200
I
'
400
i
I
600
rl/ loins 30
....
I ....
1'"'1
n=7
' "~' 9
D
I
rl = 1000
0
C
n=22
800
1000
5o
I
I
I
I
500
J llJJ
' [''I'"'I""]
1000
2000
13= I000
lO ms i
0
,
i
200
,
i
400
600
I
800
l
1000
Fig 2 Example of mtersplke Interval histograms (10 ms bins, n = 1000) for each group of cells (left) and d~strlbutlon of the means mtersplke intervals for all rhythmic cells (right) Abscissa time m mllhseconds (ms), arithmetic scale (left) and logarithmic scale (right) A repetmve discharge (mean_+ S D Intersplke interval 100_+ 14 ms) B bursting repetlt+ve The first histogram (truncated) corresponds to the intervals within the bursts (4_+0 77 ms) and the second histogram to the intervals between bursts (97_+ 15 ms) The &strlbutlon for n cells (right) corresponds to the second histogram C peno&c The first histogram (Iruncatedl represents the intervals within the trams of spikes (21 + 2 ms) and the second histogram, the intervals between the trams of spikes (573 + 190 ms) The dlstrlbuUon (right) corresponds to the lntertraln intervals D non-rhythmic(316_+271 ms)
vary, a c c o r d i n g to the cell, f r o m 680 to 1850 ms (1320_+ 190 ms) ( F i g 2C right), giving a spikes trains f r e q u e n c y o f 20~75 t r a I n s / m l n (38 + 8.5 ms). T h e S D. o f m t e r t r a l n m t e r v a l s f o r all cells was b e t w e e n 20 a n d 30% o f the m e a n T h e s e n e u r o n s w e r e l o c a t e d in the v e n t r o l a t e r a l N T S (n = 5) a n d in the l a t e r a l p a r t o f the P G I (n = 2) ( F i g
IC) D T h e r e m a m m g 31 n e u r o n s fired w i t h f r e q u e n c i e s o f I 20 spikes/s w i t h o u t a n y rhythmtclty
S p i k e i n t e r v a l h i s t o g r a m s o f these n e u r o n s s h o w e d a n o n - n o r m a l distri-
b u t i o n ( F i g 2D)_ T h e s e n e u r o n s w e r e f o u n d in all r e c o r d e d b r a i n s t e m a r e a s the retic u l a r f o r m a t i o n ( n = 17), the P G I ( n = 3 ) ,
the N T S ( n = 3 ) , the spinal n u c l e u s o f the
61
trigeminal nerve (n = 1) and the cuneate nucleus (n = 7) where neurons presented nonrhythmic bursting discharges. The present study shows that the perfused tsolated brainstem preparation of an adult animal is viable since spontaneous neuronal activities are present m all the recorded structures. Thts preparation exhibits 3 types of rhythmic neuronal actlvittes which are localized in the same nuclei. Both single and bursting groups of repetitive discharges (patterns A and B) may be closely related since their discharge frequencies are m the same range and since it was possible to observe, dunng a recording period, transient changes from a bursting to a single repetitive discharge. These repetitive discharges have also been descrtbed in the same nuclei in the shce preparation [2, 3,
ll]. On the other hand, the patterns of periodic discharges (pattern C) clearly differed from the repettttve patterns. They may be compared to the periodic activity of resptratory neurons recorded m the same areas in m vivo studies m the cat [1] and in the guinea pig [8]. Their spikes trains frequency, 20-75/mm, is comparable to respiratory frequency observed in the guinea pig m vivo [4] and in the m sltu perfused brain preparatlon [9]. These periodic dtscharges were not recorded in slice preparattons unless excitatory neurotransmitters were added to the perfusion [2, 5]. Thus, periodic firing patterns may be present m the in vitro preparatton when connections between different parts of the extended networks are conserved. Furthermore, peripheral feedbacks such as primary sensory afferences which are cut in the brainstem preparation, are not necessary for the occurrence of periodic trains of spikes. This work was supported by INSERM 866004 and DRET 87041. 1 Blanchl, A L , Modahtrs de drcharge et propnrtrs anatomofonctlonnelles des neurones respiratolres bulbalres, J Phymol (Paris), 68 (1974) 555-587 2 Champagnat, J Denavlt-Sauble, M and Siggms, G R , Rhythmic neuronal actlvttms in the nucleus of the tractus sohtanus isolated m vitro, Brain R e s , 280 (1983) 155~159 3 Champagnat, J , Denavit-Saubm, M , Grant, K and Shen, K F , Organization of synaptlc transmission in the mammalian solitary complex, studied m vitro, J Phystol (Lond), 381 0986) 551 573 4 Crosfill, M.L and Wlddicombe, J G , Physical characteristics of the chests and lungs and the work of breathing in different mammahan species, J Physiol, 158 (1961) 1-14. 5 Dekm, M S, Rlcherson, G B and Gettmg, P A , Thyrotropm-releasmg hormone induces rhythmic bursting in neurons of the nucleus tractus sohtarms, Scmnce, 229 (1985) 67~9 6 Euler, C von, Bramstem mechamsms for generation and control of breathing pattern_ In N S Chefmack and J_C. WIddtcombe (Eds), The Respiratory System, Handbook of Physiology Control of Breathing, Vol II, 1986, pp 1~7 7 Lhnas, R and Muhlethaler, M , An electrophysiologmal study of the in vitro, perfused bralnstem-cerehelium of adult guinea-pig, J Physiol. (Lond), 404 (1988) 215-240. 8 Rlcherson, G B and Getting, P A , Characteristms of respiratory neurons in the guinea pig, Soc Neuroscl. Abstr, 13th Annual Meeting, Boston, MA, 1983, 1163 p 9 Rmherson, G B and Getting, P.A_, Maintenance of complex neural function during perfuston of the mammalian brain, Brain Res, 409 (1987) 128-132 l0 Spyer, K M , Central nervous integration of cardiovascular control, J Exp Biol, 100 (1982) 109-128 l 1 Sun, M_K, Hackett, J T and Guyenet, P G , Sympathoexcttatory neurons of rostral ventrolateral medulla exhibit pacemaker properties m the presence of a glutamate-receptor antagomst, Brain Res, 438 (1988) 23-40 ,
57
Elscvler Scientific Pubhshers Ireland Ltd
NSL 06108
Rhythmic discharges in the perfused isolated brainstem preparation of adult guinea pig M.P. Morin-Surun and M. Denavit-Saubi6 Laboratolre de Physwlogte Nerveuse, C N R S , Gif-sur- Yveltc (France) (Received 14 December 1988, Revlscd version received 7 February 1989, Accepted l0 February 1989)
Key words
Rhythmic activity; Isolated brainstem preparation; Adult, Gmnea pig, Nucleus tractus sohtarn, Paraglgantocellular reticular nucleus, Ambiguus nucleus
An isolated brainstem preparation of adult guinea pig was used for an m vitro electrophyslologlcal study of rhythmic neuronal &scharge patterns More than half of the spontaneously actave neurons (40/71) exhibited a rhythmic discharge. Rhythmic discharges were recorded dorsally in the nucleus tractus sohtanus and ventrally in the amblguus and paraglgantoceUular reticular nuclei Three types of rhythmic patterns of discharge were identified in these areas repetitive single discharge, repetitive bursting discharge and spontaneous periodic discharge No rhythmic patterns were recorded m other explored parts of the brainstem Comparison of these data with those from bramstem shces shows that spontaneous periodic discharges may require extended networks within the bramstem
Neuronal networks localized m the brainstem control in vivo various rhythmic functions such as respiratory [6] and cardiovascular activities [10]. Spontaneous rhythmic neuronal discharges, consisting of trains of spikes separated by silent penods, have been recorded in different brainstem nuclei. Correlation with a rhythmic function has been established from relationships of instantaneous discharge rate to the periodic output of the function [1]. Brainstem slice preparations allowed the study of synaptic and intnnsic neuronal properties withm nuclei imphcated in these rhythmic functions [2]. However, in such preparations which include a small part of the neuronal network involved in the in vivo generation of periodic trains of spikes, the spontaneous rhythmic activity recorded is only a repetitive discharge, consisting of single spikes separated by relatively constant intervals [2]. In order to test whether peripheral or medullary inputs are required to obtain perlO&C discharges, we developed the preparation of perfused isolated brainstem of adult guinea pig in which complex neural function is preserved [7]. In the present study, we show that rhythmic patterns described both in vivo and in vitro are recorded in this preparation in the same brainstem nuclei. Fifteen guinea pigs (160-240 g) were anesthetized with sodium pentobarbltal (60 Correspondence M_P Morin-Surun, Laboratolre de Physiology Nerveuse, C.N~R.S, 91190 Gif-surYvette, France_ 0304-3940/89/$ 03 50 © 1989 Elsevier Scientific Pubhshers Ireland Ltd
58 mg/kg, l.p ) and decapitated at the level of the first cervical vertebra Brainstem and cerebellum were rapidly removed from the cranium. During this surgical procedure, cold (4°C) Ringer's solution was dripped onto exposed brain surfaces The Ringer's solution consisted o f ( i n mM) NaCI 124, KCI 5, MgSO4 2 2, KH2PO4 1 2, NaHCOa 26, CaCI2 2, glucose 10 The bralnstem was maintained by pros, the ventral surface upwards, in a recording chamber filled with room temperature Ringer's saturated with 95% 02, 5% CO2 (pH 7 2) The basilar artery was rapidly cannulated via a catheter inserted at the level of the pons, and the vascular system was perfused with saturated Ringer's at room temperature using a Gllson pump. Ends of arteries at the cerwcal first segment of the spinal cord were ligated. Perfuslon pressures [7] of 50- 70 mm Hg correspond to a flow of 1 5-2 ml/min. The chamber was also perfused with the same medmm by gravity flow. The temperature of the Ringer's perfuslng the vascular system and the tissue gradually rise to 27°C in 1 h. Single-unit recordings were initiated after temperature stabilization and intentionally interrupted after 30 mln. Preparations were recorded dunng 5-10 h Multi-barrel microp~pettes which penetrated through the ventral surface were filled with: NaC1 (3 M) for recording and Methyl blue for histological identification of recording sites. The lntersplke interval histograms were bmlt up using a computer and mean values + S.D. of the lntersplke interval were calculated for each cell Seventy-one spontaneously active neurons were recorded m the isolated brainstems. Forty neurons exhibited a rhythmic pattern of discharge. All of them were located either in the nucleus tractus solitarius (NTS) (n = 16), in the paragigantocellular reticular nucleus (PGi) (n = 16) and in the ambiguus nucleus (AMB) (n = 8). Three types of rhythmic activities were distinguished by their pattern: the repetitive single discharge, the repetitive bursting discharge and the periodic discharge as defined below. These different types of rhythmic discharges could be recorded in the same nucleus from the same preparation. Likewise, the same type of discharge could be recorded in different nuclei of the same preparation. A The pattern of repetitive single discharge (n = 22, Fig. 1A) was composed of single spikes separated by relatively constant intervals. Fig. 2A (left) illustrates the spike interval histogram of one of these cells. For this category of cells, the interval between spikes was in a range of 60-450 ms (195+ 108 ms) (Fig 2A, right), giving a firing frequency of about 2 16 splkes/s. The S.D of the intersplke interval for all cells was within 5-20% of the mean. These neurons were recorded in the ventral part of NTS (n= 10), In AMB ( n = 4 ) and in PGI ( n = 8 ) (Fig IA). No relation was observed between the different firing frequencies of these neurons and the recording site, e.g. a cell firing with a frequency of 3 spikes/s was recorded in the vicinity o f a cell firing at l0 spikes/s B. The pattern of repetitive bursting discharge (n = I l, Fig. 1B), consisted of bursts (2-5 spikes) separated by relatively constant intervals. The intersplke intervals hrstogram (Fig. 2B, left) shows two populations of intervals: the first population (truncated) corresponds to the intersplke intervals within bursts and the second corresponds to the intervals between bursts The intervals between bursts were distributed in a range of 100-400 ms (216+88 ms) (Fig. 2B right) gavlng a burst frequency of
59
A
II LI L,,,t.L,I 1 I l I I I 1 s
1 ms
B
1 s
5 ms
1 s
20 ms
C
m
Fig 1 Different types of rhythmic activities, recorded In the Isolated bramstem preparation For each type extracellular recording at slow sweep, fast sweep and anatomical locahzatlon projected on the plane which corresponds to the most rostral recording A: repeUtwe discharge B repet~tlvebursting C peno&c Amb, nucleus (n) amblguus, Cu, n cuneatus, Io, inferior ohve, Mve, me&al vestibular n_, PGI, paragqgantocellular reticular n, PrH, preposltus hypoglossal n, Py, pyramidal tract, SP5, spmal tngemlnal n, SPV, spinal vestibular n, TS, tractus sohtanus 3-10 bursts/s. The S.D. o f interburst intervals for all these cells was between 8 and 20% o f the mean. This repet~ttve bursting acttvtty was usually recorded m the ventral part of the bramstem, in the P G t and m the A M B (n = 10), c o m p a r e d to n = 1 m the N T S (Fig. 1B). In some cases (n = 3), during a recording period, a burst was replaced by a single spike lndicatmg that the repetitive burstmg and repetitive single discharges are v a n a t t o n o f a same basic pattern. C. The pattern o f periodic discharge (n = 7, Fig. 1C) was characterized by a train o f spikes (5-20 sptkes) with a decreasing firing frequency, separated by long and relatively constant intervals. Depending on the cell, the train o f spikes lasted 100-800 ms (570+__ 110 ms). Interspike interval histograms show two populations (Fig. 2C, left) The first one (truncated) corresponds to the intervals wtthm the trams o f spikes and the second one to the intervals between the trains. These long intervals could
60
INTERSPIKE INTERVALS
MEANS OF INTERSPIKE INTERVALS DISTRIBUTION OF n CELLS)
cells
n :1ooo
A
200 1 1001
i
I
0
200
I
B
I
I
I
50
t°°1
,I,I I
100
200 300400
1+o1 ~
w
l
~'
i
I
200
5O
30-p
n
r~
10-
I
=
ms
n= 11
200 1
l
JII I
100
IIIII I
|
I I
I
200 300400
, o 0 0
,
I
200
I
'
400
i
I
600
rl/ loins 30
....
I ....
1'"'1
n=7
' "~' 9
D
I
rl = 1000
0
C
n=22
800
1000
5o
I
I
I
I
500
J llJJ
' [''I'"'I""]
1000
2000
13= I000
lO ms i
0
,
i
200
,
i
400
600
I
800
l
1000
Fig 2 Example of mtersplke Interval histograms (10 ms bins, n = 1000) for each group of cells (left) and d~strlbutlon of the means mtersplke intervals for all rhythmic cells (right) Abscissa time m mllhseconds (ms), arithmetic scale (left) and logarithmic scale (right) A repetmve discharge (mean_+ S D Intersplke interval 100_+ 14 ms) B bursting repetlt+ve The first histogram (truncated) corresponds to the intervals within the bursts (4_+0 77 ms) and the second histogram to the intervals between bursts (97_+ 15 ms) The &strlbutlon for n cells (right) corresponds to the second histogram C peno&c The first histogram (Iruncatedl represents the intervals within the trams of spikes (21 + 2 ms) and the second histogram, the intervals between the trams of spikes (573 + 190 ms) The dlstrlbuUon (right) corresponds to the lntertraln intervals D non-rhythmic(316_+271 ms)
vary, a c c o r d i n g to the cell, f r o m 680 to 1850 ms (1320_+ 190 ms) ( F i g 2C right), giving a spikes trains f r e q u e n c y o f 20~75 t r a I n s / m l n (38 + 8.5 ms). T h e S D. o f m t e r t r a l n m t e r v a l s f o r all cells was b e t w e e n 20 a n d 30% o f the m e a n T h e s e n e u r o n s w e r e l o c a t e d in the v e n t r o l a t e r a l N T S (n = 5) a n d in the l a t e r a l p a r t o f the P G I (n = 2) ( F i g
IC) D T h e r e m a m m g 31 n e u r o n s fired w i t h f r e q u e n c i e s o f I 20 spikes/s w i t h o u t a n y rhythmtclty
S p i k e i n t e r v a l h i s t o g r a m s o f these n e u r o n s s h o w e d a n o n - n o r m a l distri-
b u t i o n ( F i g 2D)_ T h e s e n e u r o n s w e r e f o u n d in all r e c o r d e d b r a i n s t e m a r e a s the retic u l a r f o r m a t i o n ( n = 17), the P G I ( n = 3 ) ,
the N T S ( n = 3 ) , the spinal n u c l e u s o f the
61
trigeminal nerve (n = 1) and the cuneate nucleus (n = 7) where neurons presented nonrhythmic bursting discharges. The present study shows that the perfused tsolated brainstem preparation of an adult animal is viable since spontaneous neuronal activities are present m all the recorded structures. Thts preparation exhibits 3 types of rhythmic neuronal actlvittes which are localized in the same nuclei. Both single and bursting groups of repetitive discharges (patterns A and B) may be closely related since their discharge frequencies are m the same range and since it was possible to observe, dunng a recording period, transient changes from a bursting to a single repetitive discharge. These repetitive discharges have also been descrtbed in the same nuclei in the shce preparation [2, 3,
ll]. On the other hand, the patterns of periodic discharges (pattern C) clearly differed from the repettttve patterns. They may be compared to the periodic activity of resptratory neurons recorded m the same areas in m vivo studies m the cat [1] and in the guinea pig [8]. Their spikes trains frequency, 20-75/mm, is comparable to respiratory frequency observed in the guinea pig m vivo [4] and in the m sltu perfused brain preparatlon [9]. These periodic dtscharges were not recorded in slice preparattons unless excitatory neurotransmitters were added to the perfusion [2, 5]. Thus, periodic firing patterns may be present m the in vitro preparatton when connections between different parts of the extended networks are conserved. Furthermore, peripheral feedbacks such as primary sensory afferences which are cut in the brainstem preparation, are not necessary for the occurrence of periodic trains of spikes. This work was supported by INSERM 866004 and DRET 87041. 1 Blanchl, A L , Modahtrs de drcharge et propnrtrs anatomofonctlonnelles des neurones respiratolres bulbalres, J Phymol (Paris), 68 (1974) 555-587 2 Champagnat, J Denavlt-Sauble, M and Siggms, G R , Rhythmic neuronal actlvttms in the nucleus of the tractus sohtanus isolated m vitro, Brain R e s , 280 (1983) 155~159 3 Champagnat, J , Denavit-Saubm, M , Grant, K and Shen, K F , Organization of synaptlc transmission in the mammalian solitary complex, studied m vitro, J Phystol (Lond), 381 0986) 551 573 4 Crosfill, M.L and Wlddicombe, J G , Physical characteristics of the chests and lungs and the work of breathing in different mammahan species, J Physiol, 158 (1961) 1-14. 5 Dekm, M S, Rlcherson, G B and Gettmg, P A , Thyrotropm-releasmg hormone induces rhythmic bursting in neurons of the nucleus tractus sohtarms, Scmnce, 229 (1985) 67~9 6 Euler, C von, Bramstem mechamsms for generation and control of breathing pattern_ In N S Chefmack and J_C. WIddtcombe (Eds), The Respiratory System, Handbook of Physiology Control of Breathing, Vol II, 1986, pp 1~7 7 Lhnas, R and Muhlethaler, M , An electrophysiologmal study of the in vitro, perfused bralnstem-cerehelium of adult guinea-pig, J Physiol. (Lond), 404 (1988) 215-240. 8 Rlcherson, G B and Getting, P A , Characteristms of respiratory neurons in the guinea pig, Soc Neuroscl. Abstr, 13th Annual Meeting, Boston, MA, 1983, 1163 p 9 Rmherson, G B and Getting, P.A_, Maintenance of complex neural function during perfuston of the mammalian brain, Brain Res, 409 (1987) 128-132 l0 Spyer, K M , Central nervous integration of cardiovascular control, J Exp Biol, 100 (1982) 109-128 l 1 Sun, M_K, Hackett, J T and Guyenet, P G , Sympathoexcttatory neurons of rostral ventrolateral medulla exhibit pacemaker properties m the presence of a glutamate-receptor antagomst, Brain Res, 438 (1988) 23-40 ,