- Email: [email protected]
Effects of atropine upon the hippocampal electrical activity in rats with special reference to paradoxical sleep
510
Electroencephalography and Clinical Neurophysiology, 1977, 42:510--517 © Elsevier/North-Holland Scientific Publishers, Ltd.
EFFECTS OF ATROPINE UPON THE HIPPOCAMPAL ELECTRICAL ACTIVITY IN RATS WITH SPECIAL REFERENCE TO PARADOXICAL SLEEP SETSUO USUI and SHINKURO IWAHARA
Department of Psychology, Tokyo University of Education, 3-29-10tsuka Bunkyoku, Tokyo and the University of Tsukulsa, Ibaraki Pref. 300-31 (Japan) (Accepted for publication: July 9, 1976)
Vanderwolf (1969, 1971) had argued that the hippocampal theta activity is always associated with gross body movement, such as walking, rearing and turning in rats. Although he admitted the existence of theta rhythms in the absence of body movement, it was claimed that such theta activity is generally of lower frequency, which is not high enough to initiate a movement already programmed (Vanderwolf 1971, p. 100). More recently, however, Vanderwolf changed this unitary position in favor of a dualistic theory of the hippocampal theta activity in which one type of theta rhythm with a lower frequency (4--7 c/sec) occurs during behavioral immobility in undrugged, waking animals or during ether or urethane anesthesia, and can be abolished by injections of atropine sulphate (25--50 mg/kg, i.p.), while a second type of theta rhythm with a higher frequency (7--12 c/sec) appears only during body movement or voluntary movement, and is relatively unaffected by atropine (Kramis et al. 1975; Vanderwolf 1975). Unfortunately, in this revised theory, Vanderwolf failed to refer to the hippocampal theta activity observed during the tonic phase of paradoxical sleep (PS). However, he suggested earlier that the hippocampal theta activity appearing during PS is functionally identical to that observed during body movement in the waking rat, because a great deal of activity is observed in the central motor system during PS but this is largely prevented by descending inhibitory influences from reaching the mus-
cles (Vanderwolf 1971). In addition, a similar positive relationship between the theta frequency and the rate of rapid eye movement (REM) was reported in cats not only during waking states but also during PS (Sakai et al. 1973) and the same relationship was observed during PS in rats (Sano et al. 1973). Thus if Vanderwolf's position is correct, the hippocampal theta activity during PS is expected to be relatively unaffected by atropine. However, this expectation is not quite in accordance with a study by Weiss et al. (1964), who found, contrary to control data, a spontaneous waxing and waning alternation of the theta amplitude, interrupted by periods of irregular activity during PS in atropinized rats. Although not clearly mentioned in their text, their table indicates a significant increase in theta frequency during PS (Weiss et al. 1964, Table 5). The main purpose of the present study is to examine the effects of atropine sulphate upon hippocampal electrical activity, especially during PS, in rats. As a phasic index of PS, the REM rate was used, as in our previous studies (Sakai et al. 1973; Sano et al. 1973). A similar study in cats is now in progress in our laboratory.
Method and material Five male adult Wistar--Imamichi rats, weighing 300--420 g were used as subjects. Each rat was anesthetized by pentobarbital
ATROPINE EFFECTS ON PARADOXICALSLEEP IN RATS injections (35 mg/kg, i.p.) and placed in a stereotaxic instrument. After exposure of the skull, two fixed stainless wire electrodes (0.19 mm in diameter) insulated up to the tip were implanted one above the other (0.5 mm tip separation) in both sides of the dorsal hippocampus (A4.2, L1.5, D2.7) based on de Groot's brain map (1967). For cortical and indifferent electrodes, stainless watch screws (1 mm in diameter) were placed on the dura of the frontal and occipital cortices, and in the nasal bone, respectively. The electrooculogram (EOG) was recorded with two stainless wires (0.19 mm in diameter), implanted in both sides of the left orbit and the electromyogram (EMG) was similarly obtained with two stainless wires of the same size, inserted into neck muscles. The electrode assembly was held firmly to the skull with dental cement. After a recovery period of at least a week from operation, each rat was put in an observation box {30 X 30 X 30 cm) with a front glass wall and a wood floor with sawdust, where the animal was allowed to eat food pellets and to drink tap water from a water bottle attached to the box. About 4 h later, the rat was injected i.p. either with saline or atropine sulphate, and the polygraphic recordings and visual observation started and continued for 6 h or at least until a PS episode of 100 sec or longer was obtained. All recordings were made on a 13-channel polygraph and the hippocampal EEG was simultaneously processed with a Walter-type frequency analyzer, which produced an integrated value per 5 sec for each of the following 10 frequency bands: 1--3, 3--5, 5--7, 7---9, 9--11, 11--13, 13--15, 15--20, 20--30 and 30--60 c/sec. The animal's behavior was monitored by a TV set, placed in the recording room, next to an electrically shielded, sound-attenuated room in which the observation box was located. Some rats were observed under the saline condition, and several days later, observation was repeated under the atropine condition and the procedure was reversed for the other
511
rats. Based on a pilot study, 5 mg/kg of atropine was given for one rat, 10 mg/kg for 3 rats, and 10, 20 and 30 mg/kg for the last rat. Usually no PS episodes were observed within our observation period if atropine was administered at doses higher than 10 mg/kg. Atropine sulphate was dissolved in distilled water with a concentration of 10 mg/ml and the same a m o u n t of saline was injected under the control condition. Following completion of the experiment, the animal was perfused with 10% formalin solution under deep ether anesthesia and the brain was removed. Using a frozen technique, the brain sections were cut 40 p thick and stained with hematoxylin and all hippocampal electrodes were found to be adequately located within the dorsal hippocampus proper.
Results
Paradoxical sleep PS in rats was most easily identified by the tonic aspect, including continuously maintained hippocampal theta r h y t h m and little EMG activity, as well as by the phasic aspect as represented by the REM. The PS latency (in min) or the time interval between saline or atropine injections and the first appearance of a PS episode of any length, was measured in addition to its duration (in sec). The PS latency obtained was considerably increased by atropinization for all rats except one (Rat UP-10), which, however, showed the same drug effect if the dose was increased up to 30 mg/kg; then the median PS latency across the total 5 rats was 312 min after atropine injections in contrast to 71 min after saline. Similarly, the duration of the first PS episode was reduced by atropine for all rats except the same rat UP-10; again this animal produced a similar drug effect if doses were increased up to 30 mg/kg, and the median PS duration was then 161 sec after saline vs. only 15 sec after atropine. Fig. 1 illustrates a typical time sequence of sleep--wakefulness cycles
512
S. USUI, S. IWAHARA RAT
SA
I 0 min
INF"
r!!!11-
I I
5 -
UP-6(RA-12)
,/ II 3
4
r,, il 5
6hr
under saline and atropine of one rat (UP-6). In order to examine, on a statistical basis, the detailed characteristics of a PS episode, only the first stable PS episode with a duration of 100 sec or longer was sampled under both saline and drug conditions. The sampled PS episode was divided into successive 1 sec intervals, and each interval was categorized as (++) if it included at least one REM burst, (+) if it contained isolated REMs, (-+) if it in-
A T R O P I N E IOmg/kg i.p.
i 0 min
I I
5-
i I
, I
I
t
5
4
5
r!
6 hr
SLOW WAVE SLEEP PARADOXICAL SLEEP
Fig. 1. Typical temporal shifts between wakefulness (not entered), slow wave sleep {thin lines) and paradoxical sleep (PS) (thick lines) during 6 h observation after saline and atropine injections. In this figure, 1 h observation was each divided into six 10 rain periods. Evidently, atropine retarded the first appearance of a PS episode and shortened its duration. In addition, as a whole, both slow wave sleep and PS was suppressed by the drug in this rat.
TABLE I Effects of atropine upon the REM rate and two measures of the hippocampal theta activity as a function of REM rates during paradoxical sleep, and effects of atropine upon theta frequency with and without REM during body movement in waking states. Measure
Drug
REM 'rate
(-)
(±)
(÷)
(++)
59.4(8.2) 57.3 (15.0)
6.9(3.4) 6.8 (2.8)
I. Paradoxical sleep a. Occurrence (%)
saline atropine
16.6(5.8) 19.4(7.6)
17.0 (4.8) 16.5 (7.5)
b. Theta frequency (c/sec)
saline atropine
7.2(0.2) 7.2(0.2)
7.3 (0.3) 7.5 (0.3)
7.6(0.4) 8.0(0.2)
8.4 (0.3) 9.2 (0.4)
c. Theta distortion (%) per REM rate
saline atropine
4.4(2.6) 28.2 (14.6)
0.9 (1.8) 14.4 (6.1)
1.4(0.9) 5.1(3.3)
0.0 (0.0) o.o (o.o)
saline atropine
(no REM) 6.5(0.2) 6.8(0.2)
II. Body movement Theta frequency (c/sec)
(REM) 7.4(0.2) 7.3 ( 0 . 3 )
Theta frequency during body movement in waking states was divided into only two categories of no REM and REM since more exact classification was impossible due to artifacts based on body movement itself. Standard deviations indicating individual differences are indicated in parentheses. For explanation of the four REM rates, see text.
ATROPINE EFFECTS ON PARADOXICAL SLEEP IN RATS
cluded only part of the REM, or (--) if it had no REM. Thus Table I was made with respect to PS. Table I shows the percent distribution of occurrence of each of the 4 REM categories under the two conditions. Obviously, the atropine effect upon the REM distribution was negligible and as a matter of fact, analysis of variance showed that the interaction between the REM categories and the drug states (group) was far from a statistical significance ( F = 0.05, d f = 1 / 4 , P ~ " 0.10). The REM score was also c o m p u t e d for each subject, giving weights 3, 2, 1 and 0 for (++), (+), (+) and (--) respectively, and the obtained group mean was 1.6 (SD = 1.8) and 1.5 (SD
PARADOXICAL
513
-- 2.2) under saline and atropine respectively, and again the drug effect was t o o small to be significant by the matched t test (P > 0.05). Table I also indicates the mean theta frequency for each REM category under saline and drug, and analysis of variance was conducted based on the unweighted means (Winer 1962) for significance of the differences. As was expected, theta frequency increased with the REM rate and the overall frequency differences among the 4 categories was highly significant (F = 179.03, d f = 3/12, P < 0.01). Both the drug effect (F = 49.61, d f = 1/4, p < 0.01) and the interaction between the REM and drug effects (F = 6.84, d f = 3/12, P < 0.01) were significant and
SLE
EP
RkT UP-B (RA-II)
£M DHP F EOG SALINE Fig. 2. Typical polygraphical recordings, including EMG, DHPC (dorsal hippocampus), FCX (frontal cortex), EEG and EOG, during paradoxical sleep (PS) and locomotion (gross body m o v e m e n t ) after saline and atropine injections. During PS, hippocampal theta activity was very homogeneous and rhythmic under control saline, but under atropine, theta activity increased both frequency and amplitude with REM bursts, while its regularity was somewhat distorted especially with no or little REM, and thus somewhat waxing and waning patterns appeared. The same trends were not confirmed during locomotion or gross body movement except that theta amplitude was increased after atropine. The EOG during locomotion was contaminated with artifacts due to body movement.
514
this was due to the significant atropine effect only at higher REM rates and no effect if no REM was observed (as shown in Table I). The overall theta frequency during PS was 7.53 c/sec (SD = 0.24 c/sec) and 7.86 c/sec (SD = 0.16 c/sec) under saline and atropine respectively, and the difference attained a high statistical significance (F = 40.59, d f = 1/4, P < 0.01). The representative polygraphic recordings are given in Fig. 2. Another characteristic effect of atropine was shown by the fact that the highly rhythmic theta activity during PS in control saline rats, was often changed to somewhat irregular patterns after atropinization and this disturbing effect was stronger with decreasing REM rates and was not shown with the highest REM rate (++) as indicated in Table I. As the data were highly skewed, the square-root transformation was used before they were tested for significance, and the two main effects (F for drug = 65.65, d f = 1 / 4 , P < 0.01; F for REM = 12.29, d [ = 2/8, P < 0.01) and the interaction (F = 6.02, d f = 2 [ 8 , P < 0.01) were all significant. As there was no theta irregularity with the highest REM rate both under saline and drug, this REM category was eliminated from the analysis. As irregular theta activity was produced by atropine with no or small REM rates and theta frequency was increased wth higher REM rates, the overall hippocampal EEG patterns were changed being somewhat less homogeneous and showed waxing and waning trends after atropinization, which was in contrast with the very homogeneous theta trains after controi saline administration. When cortical spindles appeared during PS, the corresponding hippocampal activity was somehow irregular; however, spindling occurred only seldom during PS, and was not significantly affected by atropine (t = --0.57, d f = 4, P < 0.10). Body movement
Gross body movement was almost always associated with the hippocampal theta activ-
S. USUI, S. IWAHARA
ity whether the animal was drugged or not. Five samples of a 1 sec hippocampal EEG recording each were taken as randomly as possible when it clearly included REM as well as when it was relatively free from REM under both saline and drug. The mean theta frequency for each of the 4 possible conditions is entered in the lowest two rows in Table I. Analysis of variance produced a significant increase in theta frequency with REM (F = 24.73, d f = 1/4, P < 0.01) but neither the drug effect (F = 0.61, d f = 1/4, P > 0.10) nor the interaction (F = 7.20, d f = 1/4, P > 0.05) was significant. As the EOG was often contaminated by artifacts due to gross body movement in waking states, it was not possible to examine more exactly the atropine effect upon theta frequency as a function of REM rates as was done during PS. Overall h i p p o c a m p a l E E G p a t t e r n s
An overall hippocampal EEG pattern was obtained in terms of the frequency spectrum based on the integrated values per 5 sec through 10 band-pass filters as describea earlier in the m e t h o d section. Fig. 3 illustrates such frequency spectra, each being based on 5 samples and averaged. Although there were considerable individual differences among the rats, there were still some c o m m o n features. When the waking rat was immobile or quiet, or was engaged in face-washing and grooming, the frequency spectrum was similar with no distinct peak components, and atropine slightly increased slower components (1--3, 3--5 c/sec). When gross body movement was clearly observed, the spectrum showed a sharp peak at the component 7--9 c/sec and a second peak at a faster c o m p o n e n t (20--30 c/sec}; this pattern remained unaltered following atropinization, excepting for a slight overall amplitude increase. A well-known pharmacological dissociation between behavior and EEG slow wave sleep patterns, first pointed out by Wikler (1952) with dogs, was not obvious in the present research.
ATROPINE EFFECTS ON PARADOXICAL SLEEP IN RATS RAT
U tel
UP-5 (RA-II)
LOCOMOTION
FACE-WASHING
80 mm
SALINE A
6
er i.i 40 0-
~
ATROPINE
~
p.
Z O
2
(J '"
SLOW WAVE SLEEP
QUIET
b. bJ Q--I 6O 811ram 40 O. 2O Z 03 W ._f <~
PARADOXICAL S L E E P 80
mm
A T R O P I N E f O m g / k g IO /~ ~ REM(+-P} REM(+)
SALINE
6O
~- 40 (.9 20 FZ
0
l 3 5 7 9 li13l~2030
(+) and (--). This was because the drug effect was dependent upon the REM rates. Although the frequency spectrum was very similar to that during locomotion or body movement in saline rats, the peak component (7--9 c/sec) was higher with higher REM rates and in addition the two faster components (9--11, 11-13 c/sec) were slightly higher with the highest REM rate (++) and this observation corresponds with higher theta frequencies with higher REM rates, as described earlier. Atropine administration showed marked effects upon frequency spectra such that the drug increased faster components (9--11, 11--13 c/sec) with REM rates (++) and (+) but it enhanced slower components (1--3, 3--5, 5--7 c/sec) with REM (--) and as a whole amplitude was augmented after atropine injections. Again this finding corresponds with the observation in which atropine increased theta frequency with higher REM rates and produced irregularity with the slow theta band with no or little REM.
r 3 5 7 S If 13 f5 2 0 3 0 3
c/s
515
s
"r
~ i'1131~
20 ' 3'0 6 '0 c/s
Fig. 3. Typical frequency spectra of the dorsal-hippocampal EEG during different states after saline and atropine injections. On the ordinate is shown the integrated amplitude (5 sec) in analyzer-pen deflection (mm) against the frequency band on the obscissa. During face-washing, and quiet-and-waking, the overall EEG spectrum was changed after atropine such that slow components were somewhat increased. During slow wave sleep, however, few changes were observed in spectrum. During locomotion or gross body movement, the spectrum remained unaltered except a slightly increased overall amplitude. During paradoxical sleep (PS), atropine augmented components faster than the dominant theta peak when REM bursts (++), (+) were observed, but with little or no REM, the same agent distorted the typical PS spectrum, especially by increasing slower components.
The frequency spectrum during slow wave sleep was dominated by slower components below 7 c/sec, and this pattern was relatively unaffected after atropinization. Lastly, the frequency spectrum during PS was obtained for each of the 3 main REM categories: (++),
Discussion In accordance with previous studies in cats (Jouvet 1961), rabbits (Khazan and Sawyer 1964) and man (Toyoda et al. 1966), atropine suppressed PS in delaying its first appearance and decreasing the duration of the first PS episode in the present experiment. Again, in agreement with our previous work with cats (Sakai et al. 1973) and rats (Sano et al. 1973), the hippocampal theta frequency was augmented with an increase in REM rate in this study and this relationship was significantly altered by atropine such that theta frequency was further increased with higher REM rates but it remained about identical with no or little REM, and in addition an increase in theta amplitude was observed especially with higher REM rates. Another important effect of atropine was that the regularity of the theta r h y t h m s was more seriously distorted with no or little REM and thus the theta activity during PS showed waxing and waning
516
patterns, as noticed by Weiss et al. (1964). Although these authors also observed a significant overall increase in theta frequency, the result was entered in the table but not stated in the text. In spite of the marked atropine effects upon the frequency, amplitude and regularity of theta rhythms, the REM rate per PS episode was not significantly affected by atropine and this finding is similar to a recent study by Henriksen et al. (1972} in which REM bursts during PS observed after 5 day PS deprivation in cats were still seen following large doses of atropine which blocked the occurrence of bursts of PGO waves. As was suggested by Vanderwolf {1975), theta activity associated with gross body movement in waking rats was not changed after atropinization and, in addition, the overall frequency spectrum as measured by a Walter-type frequency analyzer remained unaltered except that it was slightly elevated in amplitude. This observation is in contrast with the distinct effects of atropine upon theta activity during PS as described above. If this discrepancy is real, then the hippocampal theta activity observed during PS must be different in its underlying mechanisms from that appearing during body movement in waking rats. It is unfortunate that the EOG could not be clearly obtained in rats during body movement because of artifacts due to the movement itself. But as it is easier to record the EOG in cats during movement, the experiment is now in progress in our laboratory to examine the more exact atropine effect upon the relationship between theta frequency and the concomitant REM rate, as was done during PS. The present and especially Vanderwolf's data disagree with a recent report with rats by Teitelbaum et al. (1975) who claimed that the hippocampal theta activity associated with forced running behavior was blocked by scopolamine, another anticholinergic agent, at 10 mg/kg, while Vanderwolf (1975) found that scopolamine at 0.5--10 mg/kg had effects on hippocampal activity resembling those of
S. USUI, S. IWAHARA
atropine. Further research is necessary to clarify this crucial point. Finally, atropine failed to affect the overall hippocampal EEG pattern in terms of frequency spectrum, either during grooming and face-washing, during quiet and waking, or during slow wave sleep, except that the overall amplitude was increased and slower components were slightly enhanced in some but not in all cases, and thus the so-called behavior--EEG dissociation as was discovered by Wikler (1952) in atropinized dogs, was not clear in the present research. Incidentally, Weiss et al. (1964) showed that in rats the sleep-like EEG activity after atropine is present in about 85% of experiments only and thus this atropine effect is no absolute in rats.
Summary
In freely moving male rats with implanted electrodes, the influence of atropine sulphate (5--30 mg/kg, i.p.) on the hippocampal theta activity was studied with special emphasis on paradoxical sleep (PS). In accordance with previous work with cats and rabbits, atropine was found to inhibit PS, in delaying its first appearance as well as in decreasing the duration'of the first PS episode. The hippocampal theta activity during PS was also changed after atropinization. In particular, when REM occurred often, theta activity was increased not only in frequency but also in amplitude; however, with no or little REM, theta frequency was not changed but the regularity of theta rhythms was markedly disturbed. These findings are in contrast with the theta activity associated with gross body movement, which was not affected by atropine as was previous reported by Vanderwolf (1975). Thus, in contrast to commonly held views, the hippocampal theta activity during gross body movement may be different in its underlying mechanisms from that during PS.
ATROPINE EFFECTS ON PARADOXICAL SLEEP IN RATS Rdsum6
Effets de l'atropine sur l'activitd dlectrique hippocampique du rat avec intdrdt particulier pour le sornmeil paradoxal Chez des rats mfiles libres de leurs m o u v e m e n t s et p o r t e u r s d ' d l e c t r o d e s implant6es, l ' i n f l u e n c e du sulfate d ' a t r o p i n e (5 fi 30 mg/ kg, i.p.) sur l'activitd t h d t a h i p p o c a m p i q u e a dtd dtudide en a c c o r d a n t une a t t e n t i o n sp6ciale au s o m m e i l p a r a d o x a l (PS). En a c c o r d avec un travail ant6rieur p o r t a n t sur le chat et le lapin, l ' a t r o p i n e s'est av6r6e inhiber le Ps, en r e t a r d a n t sa premi6re a p p a r i t i o n ainsi q u ' e n d i m i n u a n t la durde du p r e m i e r dpisode de PS. L'activit6 thdta h i p p o c a m p i q u e au cours du PS est 6galement changde apr~s atropinisation. En particulier, lorsque le s o m m e i l avec m o u v e m e n t s oculaires rapides survient souvent, l'activit6 th6ta est a u g m e n t d e n o n s e u l e m e n t en f r 6 q u e n c e mais aussi en a m p l i t u d e ; cependant, lorsqu'il n ' y a que peu ou pas de sommeil avec m o u v e m e n t s oculaires rapides, la f r d q u e n c e t h 6 t a n ' e s t pas modifide mais la r6gularitd des r y t h m e s thdta est n e t t e m e n t perturb6e. Ces donn6es c o n t r a s t e n t avec le fait que l'activit6 t h 6 t a est associ6e aux gros m o u v e m e n t s corporels, qui ne s o n t pas affect6s par l ' a t r o p i n e c o m m e il a 6t6 p r 6 c d d e m m e n t r a p p o r t d par V a n d e r w o l f (1975). Ainsi, cont r a i r e m e n t aux donn6es g6ndralement admises, l'activit6 th6ta h i p p o c a m p i q u e au cours des gros m o u v e m e n t s c o r p o r e l s et au c o u r s du PS peut diffdrer en ce qui c o n c e r n e ses mdcanismes sous-jacents. References Groot, J. de The rat forebrain in stereotaxic coordinates. N.V. Noord-Hollandsche Uitgevers Maatschappij, Amsterdam, The Netherlands, 1963.
517
Henriksen, S.J., Jacobs, B.L. and Dement, W.C. Dependence of REM sleep PGO waves on cholinergic mechanisms. Brain Res., 1972, 48: 412-416. Jouvet, M. Telencephalic and rhombencephalic sleep in the cat. In G.E.W. Wolstenholme and M. O'Conner (Eds.), The nature of sleep. J. and A. Churchill, London, 1961: 188--206. Khazan, N. and Sawyer, C.H. Mechanisms of paradoxical sleep as revealed by neurophysiologic and pharmacologic approaches in the rabbit. Psychopharmacologia (Berl.), 1964, 5: 457--466. Kramis, R., Vanderwolf, C.H. and Brand, B.H. Two types of hippocampal rhythmic slow activity in both the rabbit and the rat: relation to behaviour and effects of atropine, diethyl ether, urethane, and pentobarbital. Exp. Neurol., 1975, 49: 58--85. Sakai, K., Sano, K. and Iwahara, S. Eye movements and hippocampal theta activity in cats. Electroenceph, clin. Neurophysiol., 1973, 34: 547--549. Sano, K., Iwahara, S., Senba, K., Sano, A. and Yamazaki, S. Eye movements and hippocampal theta activity in rats. Electroenceph. clin. Neurophysiol., 1973, 35: 621--625. Teitelbaum, H., Lee, J.F. and Johannessen, J.N. Behaviorally evoked hippocampal theta waves: a cholinergic response. Science, 1975, 188: 1114-1116. Toyoda, J., Sasaki, K. and Kurihara, M. A polygraphic study on the effect of atropine on human nocturnal sleep. Folia psychiat, neurol. Jap., 1966, 20: 275--289. Vanderwoif, C.H. Hippocampal electrical activity and voluntary movement in the rat. Electroenceph. clin. Neurophysiol., 1969, 26: 407--418. Vanderwolf, C.H. Limbic-diencephalic mechanisms of voluntary movement. Psychol. Rev., 1971, 78: 407--418. Vanderwolf, C.H. Neocortical and hippocampal activation in relation to behavior: effects of atropine, eserine, phenothiazines, and amphetamine. J. comp. physiol. Psychol., 1975, 88: 300--323. Weiss, T., Bohdaneck~, Z., Fifkov~, E. and Rold~n, E. Influence of atropine on sleep cycle in rats. Psychopharmacologia (Berl.), 1964, 5: 126--135. Wikler, A. Pharmacologic dissociation of behavior and EEG "sleep patterns" in dogs: morphine, N-allynormorphine, and atropine. Proc. Soc. exp. Biol. (N.Y.), 1952, 79: 261--265. Winer, B.J. Statistical principles in experimental design. McGraw-Hill, New York, 1962.
Electroencephalography and Clinical Neurophysiology, 1977, 42:510--517 © Elsevier/North-Holland Scientific Publishers, Ltd.
EFFECTS OF ATROPINE UPON THE HIPPOCAMPAL ELECTRICAL ACTIVITY IN RATS WITH SPECIAL REFERENCE TO PARADOXICAL SLEEP SETSUO USUI and SHINKURO IWAHARA
Department of Psychology, Tokyo University of Education, 3-29-10tsuka Bunkyoku, Tokyo and the University of Tsukulsa, Ibaraki Pref. 300-31 (Japan) (Accepted for publication: July 9, 1976)
Vanderwolf (1969, 1971) had argued that the hippocampal theta activity is always associated with gross body movement, such as walking, rearing and turning in rats. Although he admitted the existence of theta rhythms in the absence of body movement, it was claimed that such theta activity is generally of lower frequency, which is not high enough to initiate a movement already programmed (Vanderwolf 1971, p. 100). More recently, however, Vanderwolf changed this unitary position in favor of a dualistic theory of the hippocampal theta activity in which one type of theta rhythm with a lower frequency (4--7 c/sec) occurs during behavioral immobility in undrugged, waking animals or during ether or urethane anesthesia, and can be abolished by injections of atropine sulphate (25--50 mg/kg, i.p.), while a second type of theta rhythm with a higher frequency (7--12 c/sec) appears only during body movement or voluntary movement, and is relatively unaffected by atropine (Kramis et al. 1975; Vanderwolf 1975). Unfortunately, in this revised theory, Vanderwolf failed to refer to the hippocampal theta activity observed during the tonic phase of paradoxical sleep (PS). However, he suggested earlier that the hippocampal theta activity appearing during PS is functionally identical to that observed during body movement in the waking rat, because a great deal of activity is observed in the central motor system during PS but this is largely prevented by descending inhibitory influences from reaching the mus-
cles (Vanderwolf 1971). In addition, a similar positive relationship between the theta frequency and the rate of rapid eye movement (REM) was reported in cats not only during waking states but also during PS (Sakai et al. 1973) and the same relationship was observed during PS in rats (Sano et al. 1973). Thus if Vanderwolf's position is correct, the hippocampal theta activity during PS is expected to be relatively unaffected by atropine. However, this expectation is not quite in accordance with a study by Weiss et al. (1964), who found, contrary to control data, a spontaneous waxing and waning alternation of the theta amplitude, interrupted by periods of irregular activity during PS in atropinized rats. Although not clearly mentioned in their text, their table indicates a significant increase in theta frequency during PS (Weiss et al. 1964, Table 5). The main purpose of the present study is to examine the effects of atropine sulphate upon hippocampal electrical activity, especially during PS, in rats. As a phasic index of PS, the REM rate was used, as in our previous studies (Sakai et al. 1973; Sano et al. 1973). A similar study in cats is now in progress in our laboratory.
Method and material Five male adult Wistar--Imamichi rats, weighing 300--420 g were used as subjects. Each rat was anesthetized by pentobarbital
ATROPINE EFFECTS ON PARADOXICALSLEEP IN RATS injections (35 mg/kg, i.p.) and placed in a stereotaxic instrument. After exposure of the skull, two fixed stainless wire electrodes (0.19 mm in diameter) insulated up to the tip were implanted one above the other (0.5 mm tip separation) in both sides of the dorsal hippocampus (A4.2, L1.5, D2.7) based on de Groot's brain map (1967). For cortical and indifferent electrodes, stainless watch screws (1 mm in diameter) were placed on the dura of the frontal and occipital cortices, and in the nasal bone, respectively. The electrooculogram (EOG) was recorded with two stainless wires (0.19 mm in diameter), implanted in both sides of the left orbit and the electromyogram (EMG) was similarly obtained with two stainless wires of the same size, inserted into neck muscles. The electrode assembly was held firmly to the skull with dental cement. After a recovery period of at least a week from operation, each rat was put in an observation box {30 X 30 X 30 cm) with a front glass wall and a wood floor with sawdust, where the animal was allowed to eat food pellets and to drink tap water from a water bottle attached to the box. About 4 h later, the rat was injected i.p. either with saline or atropine sulphate, and the polygraphic recordings and visual observation started and continued for 6 h or at least until a PS episode of 100 sec or longer was obtained. All recordings were made on a 13-channel polygraph and the hippocampal EEG was simultaneously processed with a Walter-type frequency analyzer, which produced an integrated value per 5 sec for each of the following 10 frequency bands: 1--3, 3--5, 5--7, 7---9, 9--11, 11--13, 13--15, 15--20, 20--30 and 30--60 c/sec. The animal's behavior was monitored by a TV set, placed in the recording room, next to an electrically shielded, sound-attenuated room in which the observation box was located. Some rats were observed under the saline condition, and several days later, observation was repeated under the atropine condition and the procedure was reversed for the other
511
rats. Based on a pilot study, 5 mg/kg of atropine was given for one rat, 10 mg/kg for 3 rats, and 10, 20 and 30 mg/kg for the last rat. Usually no PS episodes were observed within our observation period if atropine was administered at doses higher than 10 mg/kg. Atropine sulphate was dissolved in distilled water with a concentration of 10 mg/ml and the same a m o u n t of saline was injected under the control condition. Following completion of the experiment, the animal was perfused with 10% formalin solution under deep ether anesthesia and the brain was removed. Using a frozen technique, the brain sections were cut 40 p thick and stained with hematoxylin and all hippocampal electrodes were found to be adequately located within the dorsal hippocampus proper.
Results
Paradoxical sleep PS in rats was most easily identified by the tonic aspect, including continuously maintained hippocampal theta r h y t h m and little EMG activity, as well as by the phasic aspect as represented by the REM. The PS latency (in min) or the time interval between saline or atropine injections and the first appearance of a PS episode of any length, was measured in addition to its duration (in sec). The PS latency obtained was considerably increased by atropinization for all rats except one (Rat UP-10), which, however, showed the same drug effect if the dose was increased up to 30 mg/kg; then the median PS latency across the total 5 rats was 312 min after atropine injections in contrast to 71 min after saline. Similarly, the duration of the first PS episode was reduced by atropine for all rats except the same rat UP-10; again this animal produced a similar drug effect if doses were increased up to 30 mg/kg, and the median PS duration was then 161 sec after saline vs. only 15 sec after atropine. Fig. 1 illustrates a typical time sequence of sleep--wakefulness cycles
512
S. USUI, S. IWAHARA RAT
SA
I 0 min
INF"
r!!!11-
I I
5 -
UP-6(RA-12)
,/ II 3
4
r,, il 5
6hr
under saline and atropine of one rat (UP-6). In order to examine, on a statistical basis, the detailed characteristics of a PS episode, only the first stable PS episode with a duration of 100 sec or longer was sampled under both saline and drug conditions. The sampled PS episode was divided into successive 1 sec intervals, and each interval was categorized as (++) if it included at least one REM burst, (+) if it contained isolated REMs, (-+) if it in-
A T R O P I N E IOmg/kg i.p.
i 0 min
I I
5-
i I
, I
I
t
5
4
5
r!
6 hr
SLOW WAVE SLEEP PARADOXICAL SLEEP
Fig. 1. Typical temporal shifts between wakefulness (not entered), slow wave sleep {thin lines) and paradoxical sleep (PS) (thick lines) during 6 h observation after saline and atropine injections. In this figure, 1 h observation was each divided into six 10 rain periods. Evidently, atropine retarded the first appearance of a PS episode and shortened its duration. In addition, as a whole, both slow wave sleep and PS was suppressed by the drug in this rat.
TABLE I Effects of atropine upon the REM rate and two measures of the hippocampal theta activity as a function of REM rates during paradoxical sleep, and effects of atropine upon theta frequency with and without REM during body movement in waking states. Measure
Drug
REM 'rate
(-)
(±)
(÷)
(++)
59.4(8.2) 57.3 (15.0)
6.9(3.4) 6.8 (2.8)
I. Paradoxical sleep a. Occurrence (%)
saline atropine
16.6(5.8) 19.4(7.6)
17.0 (4.8) 16.5 (7.5)
b. Theta frequency (c/sec)
saline atropine
7.2(0.2) 7.2(0.2)
7.3 (0.3) 7.5 (0.3)
7.6(0.4) 8.0(0.2)
8.4 (0.3) 9.2 (0.4)
c. Theta distortion (%) per REM rate
saline atropine
4.4(2.6) 28.2 (14.6)
0.9 (1.8) 14.4 (6.1)
1.4(0.9) 5.1(3.3)
0.0 (0.0) o.o (o.o)
saline atropine
(no REM) 6.5(0.2) 6.8(0.2)
II. Body movement Theta frequency (c/sec)
(REM) 7.4(0.2) 7.3 ( 0 . 3 )
Theta frequency during body movement in waking states was divided into only two categories of no REM and REM since more exact classification was impossible due to artifacts based on body movement itself. Standard deviations indicating individual differences are indicated in parentheses. For explanation of the four REM rates, see text.
ATROPINE EFFECTS ON PARADOXICAL SLEEP IN RATS
cluded only part of the REM, or (--) if it had no REM. Thus Table I was made with respect to PS. Table I shows the percent distribution of occurrence of each of the 4 REM categories under the two conditions. Obviously, the atropine effect upon the REM distribution was negligible and as a matter of fact, analysis of variance showed that the interaction between the REM categories and the drug states (group) was far from a statistical significance ( F = 0.05, d f = 1 / 4 , P ~ " 0.10). The REM score was also c o m p u t e d for each subject, giving weights 3, 2, 1 and 0 for (++), (+), (+) and (--) respectively, and the obtained group mean was 1.6 (SD = 1.8) and 1.5 (SD
PARADOXICAL
513
-- 2.2) under saline and atropine respectively, and again the drug effect was t o o small to be significant by the matched t test (P > 0.05). Table I also indicates the mean theta frequency for each REM category under saline and drug, and analysis of variance was conducted based on the unweighted means (Winer 1962) for significance of the differences. As was expected, theta frequency increased with the REM rate and the overall frequency differences among the 4 categories was highly significant (F = 179.03, d f = 3/12, P < 0.01). Both the drug effect (F = 49.61, d f = 1/4, p < 0.01) and the interaction between the REM and drug effects (F = 6.84, d f = 3/12, P < 0.01) were significant and
SLE
EP
RkT UP-B (RA-II)
£M DHP F EOG SALINE Fig. 2. Typical polygraphical recordings, including EMG, DHPC (dorsal hippocampus), FCX (frontal cortex), EEG and EOG, during paradoxical sleep (PS) and locomotion (gross body m o v e m e n t ) after saline and atropine injections. During PS, hippocampal theta activity was very homogeneous and rhythmic under control saline, but under atropine, theta activity increased both frequency and amplitude with REM bursts, while its regularity was somewhat distorted especially with no or little REM, and thus somewhat waxing and waning patterns appeared. The same trends were not confirmed during locomotion or gross body movement except that theta amplitude was increased after atropine. The EOG during locomotion was contaminated with artifacts due to body movement.
514
this was due to the significant atropine effect only at higher REM rates and no effect if no REM was observed (as shown in Table I). The overall theta frequency during PS was 7.53 c/sec (SD = 0.24 c/sec) and 7.86 c/sec (SD = 0.16 c/sec) under saline and atropine respectively, and the difference attained a high statistical significance (F = 40.59, d f = 1/4, P < 0.01). The representative polygraphic recordings are given in Fig. 2. Another characteristic effect of atropine was shown by the fact that the highly rhythmic theta activity during PS in control saline rats, was often changed to somewhat irregular patterns after atropinization and this disturbing effect was stronger with decreasing REM rates and was not shown with the highest REM rate (++) as indicated in Table I. As the data were highly skewed, the square-root transformation was used before they were tested for significance, and the two main effects (F for drug = 65.65, d f = 1 / 4 , P < 0.01; F for REM = 12.29, d [ = 2/8, P < 0.01) and the interaction (F = 6.02, d f = 2 [ 8 , P < 0.01) were all significant. As there was no theta irregularity with the highest REM rate both under saline and drug, this REM category was eliminated from the analysis. As irregular theta activity was produced by atropine with no or small REM rates and theta frequency was increased wth higher REM rates, the overall hippocampal EEG patterns were changed being somewhat less homogeneous and showed waxing and waning trends after atropinization, which was in contrast with the very homogeneous theta trains after controi saline administration. When cortical spindles appeared during PS, the corresponding hippocampal activity was somehow irregular; however, spindling occurred only seldom during PS, and was not significantly affected by atropine (t = --0.57, d f = 4, P < 0.10). Body movement
Gross body movement was almost always associated with the hippocampal theta activ-
S. USUI, S. IWAHARA
ity whether the animal was drugged or not. Five samples of a 1 sec hippocampal EEG recording each were taken as randomly as possible when it clearly included REM as well as when it was relatively free from REM under both saline and drug. The mean theta frequency for each of the 4 possible conditions is entered in the lowest two rows in Table I. Analysis of variance produced a significant increase in theta frequency with REM (F = 24.73, d f = 1/4, P < 0.01) but neither the drug effect (F = 0.61, d f = 1/4, P > 0.10) nor the interaction (F = 7.20, d f = 1/4, P > 0.05) was significant. As the EOG was often contaminated by artifacts due to gross body movement in waking states, it was not possible to examine more exactly the atropine effect upon theta frequency as a function of REM rates as was done during PS. Overall h i p p o c a m p a l E E G p a t t e r n s
An overall hippocampal EEG pattern was obtained in terms of the frequency spectrum based on the integrated values per 5 sec through 10 band-pass filters as describea earlier in the m e t h o d section. Fig. 3 illustrates such frequency spectra, each being based on 5 samples and averaged. Although there were considerable individual differences among the rats, there were still some c o m m o n features. When the waking rat was immobile or quiet, or was engaged in face-washing and grooming, the frequency spectrum was similar with no distinct peak components, and atropine slightly increased slower components (1--3, 3--5 c/sec). When gross body movement was clearly observed, the spectrum showed a sharp peak at the component 7--9 c/sec and a second peak at a faster c o m p o n e n t (20--30 c/sec}; this pattern remained unaltered following atropinization, excepting for a slight overall amplitude increase. A well-known pharmacological dissociation between behavior and EEG slow wave sleep patterns, first pointed out by Wikler (1952) with dogs, was not obvious in the present research.
ATROPINE EFFECTS ON PARADOXICAL SLEEP IN RATS RAT
U tel
UP-5 (RA-II)
LOCOMOTION
FACE-WASHING
80 mm
SALINE A
6
er i.i 40 0-
~
ATROPINE
~
p.
Z O
2
(J '"
SLOW WAVE SLEEP
QUIET
b. bJ Q--I 6O 811ram 40 O. 2O Z 03 W ._f <~
PARADOXICAL S L E E P 80
mm
A T R O P I N E f O m g / k g IO /~ ~ REM(+-P} REM(+)
SALINE
6O
~- 40 (.9 20 FZ
0
l 3 5 7 9 li13l~2030
(+) and (--). This was because the drug effect was dependent upon the REM rates. Although the frequency spectrum was very similar to that during locomotion or body movement in saline rats, the peak component (7--9 c/sec) was higher with higher REM rates and in addition the two faster components (9--11, 11-13 c/sec) were slightly higher with the highest REM rate (++) and this observation corresponds with higher theta frequencies with higher REM rates, as described earlier. Atropine administration showed marked effects upon frequency spectra such that the drug increased faster components (9--11, 11--13 c/sec) with REM rates (++) and (+) but it enhanced slower components (1--3, 3--5, 5--7 c/sec) with REM (--) and as a whole amplitude was augmented after atropine injections. Again this finding corresponds with the observation in which atropine increased theta frequency with higher REM rates and produced irregularity with the slow theta band with no or little REM.
r 3 5 7 S If 13 f5 2 0 3 0 3
c/s
515
s
"r
~ i'1131~
20 ' 3'0 6 '0 c/s
Fig. 3. Typical frequency spectra of the dorsal-hippocampal EEG during different states after saline and atropine injections. On the ordinate is shown the integrated amplitude (5 sec) in analyzer-pen deflection (mm) against the frequency band on the obscissa. During face-washing, and quiet-and-waking, the overall EEG spectrum was changed after atropine such that slow components were somewhat increased. During slow wave sleep, however, few changes were observed in spectrum. During locomotion or gross body movement, the spectrum remained unaltered except a slightly increased overall amplitude. During paradoxical sleep (PS), atropine augmented components faster than the dominant theta peak when REM bursts (++), (+) were observed, but with little or no REM, the same agent distorted the typical PS spectrum, especially by increasing slower components.
The frequency spectrum during slow wave sleep was dominated by slower components below 7 c/sec, and this pattern was relatively unaffected after atropinization. Lastly, the frequency spectrum during PS was obtained for each of the 3 main REM categories: (++),
Discussion In accordance with previous studies in cats (Jouvet 1961), rabbits (Khazan and Sawyer 1964) and man (Toyoda et al. 1966), atropine suppressed PS in delaying its first appearance and decreasing the duration of the first PS episode in the present experiment. Again, in agreement with our previous work with cats (Sakai et al. 1973) and rats (Sano et al. 1973), the hippocampal theta frequency was augmented with an increase in REM rate in this study and this relationship was significantly altered by atropine such that theta frequency was further increased with higher REM rates but it remained about identical with no or little REM, and in addition an increase in theta amplitude was observed especially with higher REM rates. Another important effect of atropine was that the regularity of the theta r h y t h m s was more seriously distorted with no or little REM and thus the theta activity during PS showed waxing and waning
516
patterns, as noticed by Weiss et al. (1964). Although these authors also observed a significant overall increase in theta frequency, the result was entered in the table but not stated in the text. In spite of the marked atropine effects upon the frequency, amplitude and regularity of theta rhythms, the REM rate per PS episode was not significantly affected by atropine and this finding is similar to a recent study by Henriksen et al. (1972} in which REM bursts during PS observed after 5 day PS deprivation in cats were still seen following large doses of atropine which blocked the occurrence of bursts of PGO waves. As was suggested by Vanderwolf {1975), theta activity associated with gross body movement in waking rats was not changed after atropinization and, in addition, the overall frequency spectrum as measured by a Walter-type frequency analyzer remained unaltered except that it was slightly elevated in amplitude. This observation is in contrast with the distinct effects of atropine upon theta activity during PS as described above. If this discrepancy is real, then the hippocampal theta activity observed during PS must be different in its underlying mechanisms from that appearing during body movement in waking rats. It is unfortunate that the EOG could not be clearly obtained in rats during body movement because of artifacts due to the movement itself. But as it is easier to record the EOG in cats during movement, the experiment is now in progress in our laboratory to examine the more exact atropine effect upon the relationship between theta frequency and the concomitant REM rate, as was done during PS. The present and especially Vanderwolf's data disagree with a recent report with rats by Teitelbaum et al. (1975) who claimed that the hippocampal theta activity associated with forced running behavior was blocked by scopolamine, another anticholinergic agent, at 10 mg/kg, while Vanderwolf (1975) found that scopolamine at 0.5--10 mg/kg had effects on hippocampal activity resembling those of
S. USUI, S. IWAHARA
atropine. Further research is necessary to clarify this crucial point. Finally, atropine failed to affect the overall hippocampal EEG pattern in terms of frequency spectrum, either during grooming and face-washing, during quiet and waking, or during slow wave sleep, except that the overall amplitude was increased and slower components were slightly enhanced in some but not in all cases, and thus the so-called behavior--EEG dissociation as was discovered by Wikler (1952) in atropinized dogs, was not clear in the present research. Incidentally, Weiss et al. (1964) showed that in rats the sleep-like EEG activity after atropine is present in about 85% of experiments only and thus this atropine effect is no absolute in rats.
Summary
In freely moving male rats with implanted electrodes, the influence of atropine sulphate (5--30 mg/kg, i.p.) on the hippocampal theta activity was studied with special emphasis on paradoxical sleep (PS). In accordance with previous work with cats and rabbits, atropine was found to inhibit PS, in delaying its first appearance as well as in decreasing the duration'of the first PS episode. The hippocampal theta activity during PS was also changed after atropinization. In particular, when REM occurred often, theta activity was increased not only in frequency but also in amplitude; however, with no or little REM, theta frequency was not changed but the regularity of theta rhythms was markedly disturbed. These findings are in contrast with the theta activity associated with gross body movement, which was not affected by atropine as was previous reported by Vanderwolf (1975). Thus, in contrast to commonly held views, the hippocampal theta activity during gross body movement may be different in its underlying mechanisms from that during PS.
ATROPINE EFFECTS ON PARADOXICAL SLEEP IN RATS Rdsum6
Effets de l'atropine sur l'activitd dlectrique hippocampique du rat avec intdrdt particulier pour le sornmeil paradoxal Chez des rats mfiles libres de leurs m o u v e m e n t s et p o r t e u r s d ' d l e c t r o d e s implant6es, l ' i n f l u e n c e du sulfate d ' a t r o p i n e (5 fi 30 mg/ kg, i.p.) sur l'activitd t h d t a h i p p o c a m p i q u e a dtd dtudide en a c c o r d a n t une a t t e n t i o n sp6ciale au s o m m e i l p a r a d o x a l (PS). En a c c o r d avec un travail ant6rieur p o r t a n t sur le chat et le lapin, l ' a t r o p i n e s'est av6r6e inhiber le Ps, en r e t a r d a n t sa premi6re a p p a r i t i o n ainsi q u ' e n d i m i n u a n t la durde du p r e m i e r dpisode de PS. L'activit6 thdta h i p p o c a m p i q u e au cours du PS est 6galement changde apr~s atropinisation. En particulier, lorsque le s o m m e i l avec m o u v e m e n t s oculaires rapides survient souvent, l'activit6 th6ta est a u g m e n t d e n o n s e u l e m e n t en f r 6 q u e n c e mais aussi en a m p l i t u d e ; cependant, lorsqu'il n ' y a que peu ou pas de sommeil avec m o u v e m e n t s oculaires rapides, la f r d q u e n c e t h 6 t a n ' e s t pas modifide mais la r6gularitd des r y t h m e s thdta est n e t t e m e n t perturb6e. Ces donn6es c o n t r a s t e n t avec le fait que l'activit6 t h 6 t a est associ6e aux gros m o u v e m e n t s corporels, qui ne s o n t pas affect6s par l ' a t r o p i n e c o m m e il a 6t6 p r 6 c d d e m m e n t r a p p o r t d par V a n d e r w o l f (1975). Ainsi, cont r a i r e m e n t aux donn6es g6ndralement admises, l'activit6 th6ta h i p p o c a m p i q u e au cours des gros m o u v e m e n t s c o r p o r e l s et au c o u r s du PS peut diffdrer en ce qui c o n c e r n e ses mdcanismes sous-jacents. References Groot, J. de The rat forebrain in stereotaxic coordinates. N.V. Noord-Hollandsche Uitgevers Maatschappij, Amsterdam, The Netherlands, 1963.
517
Henriksen, S.J., Jacobs, B.L. and Dement, W.C. Dependence of REM sleep PGO waves on cholinergic mechanisms. Brain Res., 1972, 48: 412-416. Jouvet, M. Telencephalic and rhombencephalic sleep in the cat. In G.E.W. Wolstenholme and M. O'Conner (Eds.), The nature of sleep. J. and A. Churchill, London, 1961: 188--206. Khazan, N. and Sawyer, C.H. Mechanisms of paradoxical sleep as revealed by neurophysiologic and pharmacologic approaches in the rabbit. Psychopharmacologia (Berl.), 1964, 5: 457--466. Kramis, R., Vanderwolf, C.H. and Brand, B.H. Two types of hippocampal rhythmic slow activity in both the rabbit and the rat: relation to behaviour and effects of atropine, diethyl ether, urethane, and pentobarbital. Exp. Neurol., 1975, 49: 58--85. Sakai, K., Sano, K. and Iwahara, S. Eye movements and hippocampal theta activity in cats. Electroenceph, clin. Neurophysiol., 1973, 34: 547--549. Sano, K., Iwahara, S., Senba, K., Sano, A. and Yamazaki, S. Eye movements and hippocampal theta activity in rats. Electroenceph. clin. Neurophysiol., 1973, 35: 621--625. Teitelbaum, H., Lee, J.F. and Johannessen, J.N. Behaviorally evoked hippocampal theta waves: a cholinergic response. Science, 1975, 188: 1114-1116. Toyoda, J., Sasaki, K. and Kurihara, M. A polygraphic study on the effect of atropine on human nocturnal sleep. Folia psychiat, neurol. Jap., 1966, 20: 275--289. Vanderwoif, C.H. Hippocampal electrical activity and voluntary movement in the rat. Electroenceph. clin. Neurophysiol., 1969, 26: 407--418. Vanderwolf, C.H. Limbic-diencephalic mechanisms of voluntary movement. Psychol. Rev., 1971, 78: 407--418. Vanderwolf, C.H. Neocortical and hippocampal activation in relation to behavior: effects of atropine, eserine, phenothiazines, and amphetamine. J. comp. physiol. Psychol., 1975, 88: 300--323. Weiss, T., Bohdaneck~, Z., Fifkov~, E. and Rold~n, E. Influence of atropine on sleep cycle in rats. Psychopharmacologia (Berl.), 1964, 5: 126--135. Wikler, A. Pharmacologic dissociation of behavior and EEG "sleep patterns" in dogs: morphine, N-allynormorphine, and atropine. Proc. Soc. exp. Biol. (N.Y.), 1952, 79: 261--265. Winer, B.J. Statistical principles in experimental design. McGraw-Hill, New York, 1962.