- Email: [email protected]
Regular firing patterns of suprachiasmatic neurons maintained in vitro
Neuroscience Letters, 52 (1984) 329-334
329
Elsevier Scientific Publishers Ireland Ltd.
NSL 03070
R E G U L A R FIRING P A T T E R N S OF SUPRACHIASMATIC N E U R O N S M A I N T A I N E D IN VITRO
A.M. THOMSON 1'*, D.C. WEST 2 and I.G. VLACHONIKOLIS 3
IUniversity Laboratory of Physiology, Parks Road, Oxford OX1 3PT; 2Department of Physiology, Royal Veterinary College, London; and 3Department of Biomathematics, University of Oxford, Oxford (U.K.) (Received March 8th, 1984; "Revised version received July 28th, 1984; Accepted October 9th, 1984)
Key words: suprachiasmatic nucleus - firing patterns - interspike interval distributions - regression analysis - least squares estimates - log normal distribution - anesthetics E
Cells of the suprachiasmatic nucleus (SCN) recorded in vitro display a characteristic firing pattern. Unlike many other central neurons, they have the ability to fire at a constant low rate with fixed interspike interval. This regularity is most pronounced at firing rates above 3-5 spikes/s. Spontaneous firing below 3 spikes/s was less regular but became increasingly regular as th~ firing rate was increased. Similarly, regular discharges became irregular when the firing rate was reduced below 3-5 spikes/s. The mean spontaneous firing rate was 5.6_+ 1.6, range < 1 to 12 spikes/s and cells were resistant to attempts to increase their rate of firing beyond 15-20 spikes/s. Statistical analysis showed that the firing patterns of all the cells studied formed a single continuous population in terms of their irlterspike interval distributions, and that these distributions were a function of the firing rate. Addition of either of two commonly used anesthetics, urethane or sodium pentabarbitone, disrupted previously stable, regular activity. T h e s u p r a c h i a s m a t i c n u c l e u s ( S C N ) o f t h e h y p o t h a l a m u s a r e n o w e s t a b l i s h e d as e s s e n t i a l f o r m a i n t e n a n c e o f c i r c a d i a n r h y t h m s in m a m m a l s ( f o r r e v i e w s see refs. 5 a n d 8). H o w e v e r , o n l y l i m i t e d i n f o r m a t i o n is a v a i l a b l e a b o u t t h e i r elect r o p h y s i o l o g y a n d i n t r i n s i c f i r i n g p a t t e r n s [3, 6, 9]. P r e v i o u s s t u d i e s h a v e s h o w n t h a t S C N n e u r o n s fire s p o n t a n e o u s l y in i s o l a t e d b r a i n slices a n d t h a t in s o m e cells this a c t i v i t y is r e g u l a r [2, 4, 11]. T h i s p a p e r a n a l y z e s t h e f i r i n g p a t t e r n s o f n e u r o n s distributed throughout
the SCN and the effects on these patterns of changing
neuronal firing rates. Female
Sprague-Dawley
rats,
125-200 g b.wt.,
were housed
o n a 12:12 h
l i g h t - d a r k c y c l e (lights o n 0 7 . 3 0 - 1 9 . 3 0 h) a n d s u p p l i e d w i t h f o o d a n d w a t e r a d libitum. The animals were anesthetized with ether before decapitation. The brain was r e m o v e d hypothalamus.
to an ice-cold medium
and
trimmed
to a b l o c k c o n t a i n i n g
the
O n e o r t w o 400 ttm t h i c k slices c o n t a i n i n g t h e S C N w e r e c u t a n d
t r a n s f e r r e d t o a c h a m b e r w h e r e t h e y w e r e m a i n t a i n e d at t h e i n t e r f a c e b e t w e e n w a r m *Author for correspondence. 0304-3940/84/$ 03.00 © 1984 Elsevier Scientific Publishers Ireland Ltd.
330 humidified 95% 0 2 - 5 % C02 and a medium consisting of (in mM): NaC1 124, NaHCO3 25.5, KC1 3.3, KH2PO4 1.2, CaCI2 2.5, MgSO4 1.0, D-glucose 10. The activity of single neurons was recorded extracellularly with 4 M NaCl-filled glass micropipettes (resistance 8-12 M~). Excitatory (200 mM glutamate, pH 7.5) or inhibitory (200 m M GABA, p H 3) agents were applied electrophoretically via a separate multibarrel pipette whose position was adjusted to achieve inhibition with minimum effective ejection of GABA. Rate meter records of cell firing were produced and interspike interval histograms constructed after accumulation of a fixed number of intervals. A total of 105 SCN neurons, from 15 rats, were included in this study. Each cell was recorded for not less than 15 rain and some for up to 3 h. Firing rates and patterns remained stable throughout the recording periods. The mean spontaneous firing rate was 5.6 +_+_1.6 spikes/s. Fig. 1A-C illustrates three typical interspike interval histograms obtained from cells in the SCN. Note the narrower distribution of intervals at more rapid firing rates, indicative of more regular firing. Cells were recorded throughout the nucleus and no systematic differences were found in the firing patterns of cells in different parts. In the total population, regularity appeared to be related to firing rate. Cells that fired rapidly had narrower interval distributions than more slowly fii-ing cells. Interspike interval distributions, composed of _>300 consecutive intervals (and for which we had complete confidence that no artifacts were included), were obtained for 76 cells. The remaining cells were not significantly different in terms of firing rate or interval distribution but were not included in the statistical analysis. The firing patterns of the 76 cells were analyzed by testing for interspike interval dependence on firing frequency. Statistical analysis. For the i-th cell (i = 1, 2 . . . . 76) the interspike interval distribution was assumed to be log normal, i.e. 1
Pt - - exp ai 1/2 rc
(log t - #i) e ]
where pt denotes the proportion of spikes at interspike interval t, and t~i and ai 2 are the mean and variance of log t, respectively. F r o m these distributions, we were able to derive experimental values mi and si for the unknown parameters tz~ and ai, respectively. Experimental evidence suggests that the interval distribution depends on the firing frequency in that both t~ and o~ are decreasing functions of y~, where yi is the firing rate of cell i (i= 1, 2 . . . . 76), transformed on the logarithmic scale. This hypothesis can be tested by fitting regression-type models [1]. Postulating a linear relationship m i = a + b y i , an adequate least squares fit was obtained with estimates a = 6 . 9 2 5 (S.E. (a)=0.1062), b = - 1.050 (S.E. (b) = 0.0610). The regression result is significant (FI,74 =296.291, P < 0 . 0 0 1 ) . The negative slope indicates that, as expected, an increase in firing rate is associated with a decrease in mean interspike interval. The small value of the residual sum of squares (1.095 on 74 degrees
331
A
B 6.6spikes/s
8.3sDikes/s
C
2spikes/s
50c ~5 Z
4030-
¢
=~ 20. 10, 0
,I 100
0
D
k,
. . 200ms
. . . . . . . . . .
0
100
200ms
3.2mM K
4.5raM K
0.4
2.2mM K
5spikes/s
6spikes/s
0.2
0
0.6
0.8
1.0s
16.2mM K
4spikes/s
8spikes/s
40! .E '~
30
g
_L
10"
0 0
400 0
200
E
200
400 0
400 0
F
Glutamate 9hA 4apikes/s
200
9sloikes/s
200
400ms
GABA 3nA
4.8spikes/s
3.7spikes/s
30
,°t 20
100 0
100
200
300 0
100
200
300 0
200
400
0
200
400 ms
Fig. 1. A - C : interspike interval histograms for 3 SCN neurons firing spontaneously at different frequencies. C o m p a r e with D-F, 3 SCN neurons caused to fire at different rates under the influence of: D, different external K ÷ concentrations; E, iontophoretically applied glutamate; F, ionotophoretically applied G A B A . 500 intervals per histogram.
o f freedom) indicates that all points form a single population. Similarly, it was established that the relationship si = exp (c + dyi) provided an adequate fit with least squares estimates c = - 1.336 (S.E. (c)=0.2047), d = - 0 . 2 4 3 0 (S.E. (d)=0.1176) (F1,74 =4.270, P < 0 . 0 5 ) . The negative slope indicates that o~ is also a decreasing function of yi and that more rapidly firing cells fire more regularly. In Fig. 2 the 5, 50 and 90°7o intervals are plotted against firing frequency. In an
332
interval distribution composed of 100 intervals, the 5°7o interval would be the 5th shortest, etc. Notice that the plot in Fig. 2B (50°7o interval) is a perfect straight line, while plots in Fig. 2A (5°70) and C (90°7o) are most scattered and that their slopes are slightly different from that in Fig. 2B. This is due to the fact that the variance is also a function of firing rate, and affects the 5°7o and 90o70, but not the 50°70 interval. The foregoing analysis indicates that regularity is dependent on firing rate for the population of spontaneously firing cells. To test experimentally whether regularity is also a function of firing rate for a single cell, the firing rates of recorded cells were changed. When firing rate was reduced, by reducing [K ÷ ]o, or by applying GABA electrophoretically, interval distributions became broader. Increasing firing rate, by increasing [K ÷ ]o, or by applying glutamate, resulted in narrower interval distributions (see Fig. 1D-F). The open symbols in Fig. 2 are values obtained for 4 cells caused to fire at different rates. These points clearly lie within the scatter of the population of spontaneously firing cells. Identical results were obtained for 20 SCN neurons. Increasing [K ÷ ]o, or glutamate currents beyond those required to induce firing at 10-20 spikes/s, produced no further change in rate or in interval distribution, and succeeded only in reducing spike amplitude, i.e. like supraoptic neurons 20 A
C
15 ee
o
10 8
%.
6
°o#
i
•
21 90%Interval
5%Interval
20] B 15
"% 10~ 8
.
6
~t 2O
30
40
60
8 0 100 50%Interval
200
300 400
600 ms
Fig. 2. 5, 50 and 90% intervals obtained from interspike interval histograms plotted against firing rate. For a histogram composed of 100 intervals, the 5°/0 interval would be the 5th shortest, etc. Closed circles represent cells that fired spontaneously at a given rate. Open symbols represent cells induced to fire at different rates, each symbol represents a single cell.
333
[10], but unlike m a n y other central neurons, these cells resist attempts to increase their firing rate beyond 10-20/s. Regular firing patterns are rarely reported in in vivo recordings f r o m SCN and the possibility that this difference is due to the presence of anesthetics required investigation. In two experiments urethane (1 m g / m l ) was added to the bathing medium and sodium pentobarbitone (40 #g/ml) in a further two. These concentrations would be just sub-anesthetic in vivo, assuming equal distribution throughout body water. In each experiment, one cell was followed through the change from normal to anesthetic-containing medium. In all cells there was a disruption of the previously regular firing pattern after equilibration with the anesthetic. Sampling of a further 20 neurons after equilibration with anesthetic revealed no regular firing, and all interval distributions were broader at a given firing rate than those under normal conditions. Within 2 h of a return to normal medium all effects o f anesthetic were reversed and once again regular activity was a c o m m o n finding. Other factors m a y contribute to the differences between in vivo and in vitro recordings, but these experiments indicate that the influence of anesthetics should not be overlooked if comparisons are to be made. From these results it would appear that SCN neurons in all parts o f the nucleus form a single population in terms of their firing patterns. Despite reports [2, 4] of changes in mean firing rate with time of day in isolated slices, we could find no significant difference in the mean firing rates of groups of cells recorded at all times between 10.00 and 02.00 (firing rates obtained for > 300 ceils, > 20 animals). We did however confirm the results of in vivo unit recordings [7] that at all times of day cells could be recorded that fired between < 1 and > 10 spikes/s. That is, at any
TABLE I EFFECT ON INTERSPIKE INTERVAL DISTRIBUTION OF E X P O S I N G SCN NEURONS TO C O M M O N L Y USED ANESTHETICS 5, 50 and 90070 intervals and 9007o ranges for 4 cells recorded first in standard medium, then in one that contained sodium pentabarbitone or urethane. Note the decrease in 5°7o and increase in 9007o intervals and 90% ranges on exposure to either anesthetic. The 5% interval would be the 5th shortest in a distribution composed of I00 intervals, the 50% the 50th shortest, etc. The 90°7o range is the difference between the 507o and 95% intervals. Intervals in milliseconds. Firing rates in spikes/s.
5% interval 50070 interval 90% interval 90% range Firing rate
Cell 1
Cell 2
Cell 3
Cell 4
Normal Barbiturate
Normal Barbiturate
Normal Urethane
Normal Urethane
103 148 208 125 5.6
242 275 330 105 4
139 180 217 80 5
141 168 210 90 6
41 93 252 195 5.2
120 253 366 286 4
73 153 260 257 7
96 206 506 500 4.5
334
time, cells maintain spontaneous firing rates throughout the limited range available to these cells. We would like to thank the E.P. Abraham Research Fund, Cambridge Quest Foundation and MRC for their generous financial support, A.M.T. is a Beit Memorial Fellow. 1 Armitage, P., Statistical Methods in Medical Research, Blackwell, Oxford, 1977. 2 Green, R. and Gillette, D.J., Circadian rhythm of firing rate recorded from single cells in the rat supprachiasmatic brain slice, Brain Res., 245 (1982) 198-200. 3 Groos, G. and Mason, R., Maintained discharge of rat suprachiasmatic neurones at different adaptation levels, Neurosci. Lett., 8 (1980) 59-64. 4 Kita, H., Shibata, S. and Oomura, Y., Circadian rhythmic changes of SCN neuronal activity in the rat hypothalamus slice, Soc. Neurosci. Abstr., 7 (1981) 858. 5 Moore, R.Y., The suprachiasmatic nucleus and the organisation of a circadian system, TINS, 5 (1982) 404-407. 6 Nishino, H. and Koizumi, K., Responses of neurones in the suprachiasmatic nuclei of the hypothalamus to putative transmitters, Brain Res., 120 (1977) 167-172. 7 Nishino, H., Koizumi, K. and Brooks, C. McC., The role of the suprachiasmatic nuclei of the hypothalamus in the production of circadian rhythms, Brain Res., 112 (1976) 45-59. 8 Rusak, B. and Zucker, 1., Neural regulation of circadian rhythms, Physiol. Rev., 59 (1979) 449-526. 9 Sawaki, Y., Retinohypothalamic projection: electrophysiological evidence for the existence in female rats, Brain Res., 120 (1977) 336-341. 10 Thomson, A.M., Supraoptic neurones sustain high frequency firing when extracellular Ca 2 + is replaced with other divalent cations in rat brain slices, Neuroscience, 12 (1984) 495-502. 11 Thomson, A.M. and West, D.C., The effect of picrotoxin and bicuculline on the interspike interval in the supraoptic and suprachiasmatic nuclei in slices of rat brain, J. Physiol. (Lond.), 338 (1983) 42P.
329
Elsevier Scientific Publishers Ireland Ltd.
NSL 03070
R E G U L A R FIRING P A T T E R N S OF SUPRACHIASMATIC N E U R O N S M A I N T A I N E D IN VITRO
A.M. THOMSON 1'*, D.C. WEST 2 and I.G. VLACHONIKOLIS 3
IUniversity Laboratory of Physiology, Parks Road, Oxford OX1 3PT; 2Department of Physiology, Royal Veterinary College, London; and 3Department of Biomathematics, University of Oxford, Oxford (U.K.) (Received March 8th, 1984; "Revised version received July 28th, 1984; Accepted October 9th, 1984)
Key words: suprachiasmatic nucleus - firing patterns - interspike interval distributions - regression analysis - least squares estimates - log normal distribution - anesthetics E
Cells of the suprachiasmatic nucleus (SCN) recorded in vitro display a characteristic firing pattern. Unlike many other central neurons, they have the ability to fire at a constant low rate with fixed interspike interval. This regularity is most pronounced at firing rates above 3-5 spikes/s. Spontaneous firing below 3 spikes/s was less regular but became increasingly regular as th~ firing rate was increased. Similarly, regular discharges became irregular when the firing rate was reduced below 3-5 spikes/s. The mean spontaneous firing rate was 5.6_+ 1.6, range < 1 to 12 spikes/s and cells were resistant to attempts to increase their rate of firing beyond 15-20 spikes/s. Statistical analysis showed that the firing patterns of all the cells studied formed a single continuous population in terms of their irlterspike interval distributions, and that these distributions were a function of the firing rate. Addition of either of two commonly used anesthetics, urethane or sodium pentabarbitone, disrupted previously stable, regular activity. T h e s u p r a c h i a s m a t i c n u c l e u s ( S C N ) o f t h e h y p o t h a l a m u s a r e n o w e s t a b l i s h e d as e s s e n t i a l f o r m a i n t e n a n c e o f c i r c a d i a n r h y t h m s in m a m m a l s ( f o r r e v i e w s see refs. 5 a n d 8). H o w e v e r , o n l y l i m i t e d i n f o r m a t i o n is a v a i l a b l e a b o u t t h e i r elect r o p h y s i o l o g y a n d i n t r i n s i c f i r i n g p a t t e r n s [3, 6, 9]. P r e v i o u s s t u d i e s h a v e s h o w n t h a t S C N n e u r o n s fire s p o n t a n e o u s l y in i s o l a t e d b r a i n slices a n d t h a t in s o m e cells this a c t i v i t y is r e g u l a r [2, 4, 11]. T h i s p a p e r a n a l y z e s t h e f i r i n g p a t t e r n s o f n e u r o n s distributed throughout
the SCN and the effects on these patterns of changing
neuronal firing rates. Female
Sprague-Dawley
rats,
125-200 g b.wt.,
were housed
o n a 12:12 h
l i g h t - d a r k c y c l e (lights o n 0 7 . 3 0 - 1 9 . 3 0 h) a n d s u p p l i e d w i t h f o o d a n d w a t e r a d libitum. The animals were anesthetized with ether before decapitation. The brain was r e m o v e d hypothalamus.
to an ice-cold medium
and
trimmed
to a b l o c k c o n t a i n i n g
the
O n e o r t w o 400 ttm t h i c k slices c o n t a i n i n g t h e S C N w e r e c u t a n d
t r a n s f e r r e d t o a c h a m b e r w h e r e t h e y w e r e m a i n t a i n e d at t h e i n t e r f a c e b e t w e e n w a r m *Author for correspondence. 0304-3940/84/$ 03.00 © 1984 Elsevier Scientific Publishers Ireland Ltd.
330 humidified 95% 0 2 - 5 % C02 and a medium consisting of (in mM): NaC1 124, NaHCO3 25.5, KC1 3.3, KH2PO4 1.2, CaCI2 2.5, MgSO4 1.0, D-glucose 10. The activity of single neurons was recorded extracellularly with 4 M NaCl-filled glass micropipettes (resistance 8-12 M~). Excitatory (200 mM glutamate, pH 7.5) or inhibitory (200 m M GABA, p H 3) agents were applied electrophoretically via a separate multibarrel pipette whose position was adjusted to achieve inhibition with minimum effective ejection of GABA. Rate meter records of cell firing were produced and interspike interval histograms constructed after accumulation of a fixed number of intervals. A total of 105 SCN neurons, from 15 rats, were included in this study. Each cell was recorded for not less than 15 rain and some for up to 3 h. Firing rates and patterns remained stable throughout the recording periods. The mean spontaneous firing rate was 5.6 +_+_1.6 spikes/s. Fig. 1A-C illustrates three typical interspike interval histograms obtained from cells in the SCN. Note the narrower distribution of intervals at more rapid firing rates, indicative of more regular firing. Cells were recorded throughout the nucleus and no systematic differences were found in the firing patterns of cells in different parts. In the total population, regularity appeared to be related to firing rate. Cells that fired rapidly had narrower interval distributions than more slowly fii-ing cells. Interspike interval distributions, composed of _>300 consecutive intervals (and for which we had complete confidence that no artifacts were included), were obtained for 76 cells. The remaining cells were not significantly different in terms of firing rate or interval distribution but were not included in the statistical analysis. The firing patterns of the 76 cells were analyzed by testing for interspike interval dependence on firing frequency. Statistical analysis. For the i-th cell (i = 1, 2 . . . . 76) the interspike interval distribution was assumed to be log normal, i.e. 1
Pt - - exp ai 1/2 rc
(log t - #i) e ]
where pt denotes the proportion of spikes at interspike interval t, and t~i and ai 2 are the mean and variance of log t, respectively. F r o m these distributions, we were able to derive experimental values mi and si for the unknown parameters tz~ and ai, respectively. Experimental evidence suggests that the interval distribution depends on the firing frequency in that both t~ and o~ are decreasing functions of y~, where yi is the firing rate of cell i (i= 1, 2 . . . . 76), transformed on the logarithmic scale. This hypothesis can be tested by fitting regression-type models [1]. Postulating a linear relationship m i = a + b y i , an adequate least squares fit was obtained with estimates a = 6 . 9 2 5 (S.E. (a)=0.1062), b = - 1.050 (S.E. (b) = 0.0610). The regression result is significant (FI,74 =296.291, P < 0 . 0 0 1 ) . The negative slope indicates that, as expected, an increase in firing rate is associated with a decrease in mean interspike interval. The small value of the residual sum of squares (1.095 on 74 degrees
331
A
B 6.6spikes/s
8.3sDikes/s
C
2spikes/s
50c ~5 Z
4030-
¢
=~ 20. 10, 0
,I 100
0
D
k,
. . 200ms
. . . . . . . . . .
0
100
200ms
3.2mM K
4.5raM K
0.4
2.2mM K
5spikes/s
6spikes/s
0.2
0
0.6
0.8
1.0s
16.2mM K
4spikes/s
8spikes/s
40! .E '~
30
g
_L
10"
0 0
400 0
200
E
200
400 0
400 0
F
Glutamate 9hA 4apikes/s
200
9sloikes/s
200
400ms
GABA 3nA
4.8spikes/s
3.7spikes/s
30
,°t 20
100 0
100
200
300 0
100
200
300 0
200
400
0
200
400 ms
Fig. 1. A - C : interspike interval histograms for 3 SCN neurons firing spontaneously at different frequencies. C o m p a r e with D-F, 3 SCN neurons caused to fire at different rates under the influence of: D, different external K ÷ concentrations; E, iontophoretically applied glutamate; F, ionotophoretically applied G A B A . 500 intervals per histogram.
o f freedom) indicates that all points form a single population. Similarly, it was established that the relationship si = exp (c + dyi) provided an adequate fit with least squares estimates c = - 1.336 (S.E. (c)=0.2047), d = - 0 . 2 4 3 0 (S.E. (d)=0.1176) (F1,74 =4.270, P < 0 . 0 5 ) . The negative slope indicates that o~ is also a decreasing function of yi and that more rapidly firing cells fire more regularly. In Fig. 2 the 5, 50 and 90°7o intervals are plotted against firing frequency. In an
332
interval distribution composed of 100 intervals, the 5°7o interval would be the 5th shortest, etc. Notice that the plot in Fig. 2B (50°7o interval) is a perfect straight line, while plots in Fig. 2A (5°70) and C (90°7o) are most scattered and that their slopes are slightly different from that in Fig. 2B. This is due to the fact that the variance is also a function of firing rate, and affects the 5°7o and 90o70, but not the 50°70 interval. The foregoing analysis indicates that regularity is dependent on firing rate for the population of spontaneously firing cells. To test experimentally whether regularity is also a function of firing rate for a single cell, the firing rates of recorded cells were changed. When firing rate was reduced, by reducing [K ÷ ]o, or by applying GABA electrophoretically, interval distributions became broader. Increasing firing rate, by increasing [K ÷ ]o, or by applying glutamate, resulted in narrower interval distributions (see Fig. 1D-F). The open symbols in Fig. 2 are values obtained for 4 cells caused to fire at different rates. These points clearly lie within the scatter of the population of spontaneously firing cells. Identical results were obtained for 20 SCN neurons. Increasing [K ÷ ]o, or glutamate currents beyond those required to induce firing at 10-20 spikes/s, produced no further change in rate or in interval distribution, and succeeded only in reducing spike amplitude, i.e. like supraoptic neurons 20 A
C
15 ee
o
10 8
%.
6
°o#
i
•
21 90%Interval
5%Interval
20] B 15
"% 10~ 8
.
6
~t 2O
30
40
60
8 0 100 50%Interval
200
300 400
600 ms
Fig. 2. 5, 50 and 90% intervals obtained from interspike interval histograms plotted against firing rate. For a histogram composed of 100 intervals, the 5°/0 interval would be the 5th shortest, etc. Closed circles represent cells that fired spontaneously at a given rate. Open symbols represent cells induced to fire at different rates, each symbol represents a single cell.
333
[10], but unlike m a n y other central neurons, these cells resist attempts to increase their firing rate beyond 10-20/s. Regular firing patterns are rarely reported in in vivo recordings f r o m SCN and the possibility that this difference is due to the presence of anesthetics required investigation. In two experiments urethane (1 m g / m l ) was added to the bathing medium and sodium pentobarbitone (40 #g/ml) in a further two. These concentrations would be just sub-anesthetic in vivo, assuming equal distribution throughout body water. In each experiment, one cell was followed through the change from normal to anesthetic-containing medium. In all cells there was a disruption of the previously regular firing pattern after equilibration with the anesthetic. Sampling of a further 20 neurons after equilibration with anesthetic revealed no regular firing, and all interval distributions were broader at a given firing rate than those under normal conditions. Within 2 h of a return to normal medium all effects o f anesthetic were reversed and once again regular activity was a c o m m o n finding. Other factors m a y contribute to the differences between in vivo and in vitro recordings, but these experiments indicate that the influence of anesthetics should not be overlooked if comparisons are to be made. From these results it would appear that SCN neurons in all parts o f the nucleus form a single population in terms of their firing patterns. Despite reports [2, 4] of changes in mean firing rate with time of day in isolated slices, we could find no significant difference in the mean firing rates of groups of cells recorded at all times between 10.00 and 02.00 (firing rates obtained for > 300 ceils, > 20 animals). We did however confirm the results of in vivo unit recordings [7] that at all times of day cells could be recorded that fired between < 1 and > 10 spikes/s. That is, at any
TABLE I EFFECT ON INTERSPIKE INTERVAL DISTRIBUTION OF E X P O S I N G SCN NEURONS TO C O M M O N L Y USED ANESTHETICS 5, 50 and 90070 intervals and 9007o ranges for 4 cells recorded first in standard medium, then in one that contained sodium pentabarbitone or urethane. Note the decrease in 5°7o and increase in 9007o intervals and 90% ranges on exposure to either anesthetic. The 5% interval would be the 5th shortest in a distribution composed of I00 intervals, the 50% the 50th shortest, etc. The 90°7o range is the difference between the 507o and 95% intervals. Intervals in milliseconds. Firing rates in spikes/s.
5% interval 50070 interval 90% interval 90% range Firing rate
Cell 1
Cell 2
Cell 3
Cell 4
Normal Barbiturate
Normal Barbiturate
Normal Urethane
Normal Urethane
103 148 208 125 5.6
242 275 330 105 4
139 180 217 80 5
141 168 210 90 6
41 93 252 195 5.2
120 253 366 286 4
73 153 260 257 7
96 206 506 500 4.5
334
time, cells maintain spontaneous firing rates throughout the limited range available to these cells. We would like to thank the E.P. Abraham Research Fund, Cambridge Quest Foundation and MRC for their generous financial support, A.M.T. is a Beit Memorial Fellow. 1 Armitage, P., Statistical Methods in Medical Research, Blackwell, Oxford, 1977. 2 Green, R. and Gillette, D.J., Circadian rhythm of firing rate recorded from single cells in the rat supprachiasmatic brain slice, Brain Res., 245 (1982) 198-200. 3 Groos, G. and Mason, R., Maintained discharge of rat suprachiasmatic neurones at different adaptation levels, Neurosci. Lett., 8 (1980) 59-64. 4 Kita, H., Shibata, S. and Oomura, Y., Circadian rhythmic changes of SCN neuronal activity in the rat hypothalamus slice, Soc. Neurosci. Abstr., 7 (1981) 858. 5 Moore, R.Y., The suprachiasmatic nucleus and the organisation of a circadian system, TINS, 5 (1982) 404-407. 6 Nishino, H. and Koizumi, K., Responses of neurones in the suprachiasmatic nuclei of the hypothalamus to putative transmitters, Brain Res., 120 (1977) 167-172. 7 Nishino, H., Koizumi, K. and Brooks, C. McC., The role of the suprachiasmatic nuclei of the hypothalamus in the production of circadian rhythms, Brain Res., 112 (1976) 45-59. 8 Rusak, B. and Zucker, 1., Neural regulation of circadian rhythms, Physiol. Rev., 59 (1979) 449-526. 9 Sawaki, Y., Retinohypothalamic projection: electrophysiological evidence for the existence in female rats, Brain Res., 120 (1977) 336-341. 10 Thomson, A.M., Supraoptic neurones sustain high frequency firing when extracellular Ca 2 + is replaced with other divalent cations in rat brain slices, Neuroscience, 12 (1984) 495-502. 11 Thomson, A.M. and West, D.C., The effect of picrotoxin and bicuculline on the interspike interval in the supraoptic and suprachiasmatic nuclei in slices of rat brain, J. Physiol. (Lond.), 338 (1983) 42P.