Brain Research, 103 (1976) 71-79
71
© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
H I P P O C A M P A L T H E T A R H Y T H M . 1I. D E P T H PROFILES IN T H E FREELY MOVING RABBIT
JONATHAN WINSON
The Rockefeller University, New York, N.Y. 10021 (U.S.A.) (Accepted July 14th, 1975)
SUMMARY
Depth profiles of hippocampal theta rhythm were investigated in the freely moving rabbit during three behavioral conditions: REM sleep, voluntary movement, and during sensory stimulation applied to the motionless animal. Profiles were found to be the same in all three conditions. Dorsoventral microelectrode penetration of the dorsal hippocampus revealed an approximately uniform amplitude of theta rhythm in strata oriens and pyramidale of CA1. Further microelectrode advancement revealed a sharp reversal of phase and a coincident null in amplitude in the proximal stratum radiatum. There was also a peak of theta rhythm amplitude which occurred in the molecular layer of the dorsal blade of the dentate gyrus. These data imply that in the rabbit, as in the freely moving rat, there are two generators of theta rhythm in the dorsal hippocampus, one in the dentate gyrus and the other in the overlying CA1 layer. The data also indicate the existence of a species difference in generating systems between the rabbit and the rat.
1NTRODUCTION
Species differences exist in the conditions in which theta rhythm occurs naturally in the freely moving rat and rabbit. In the rat, theta rhythm is present only during rapid eye movement (REM) sleep and voluntary motion 19,2°, while in the rabbit, theta rhythm appears in these two conditions but can also be elicited by sensory stimulation applied to the motionless animal6, s,l°. Early investigations of theta rhythm in the rabbit~, v used curarized preparations. In these experiments, the only natural condition in which theta rhythm was evoked was the application of sensory stimulation. The rhythm was also elicited by various electrical and pharmacological means. In these studies in the curarized rabbit, dorsoventral microelectrode penetrations of the hippocampus revealed the presence
72 of a type 12~ depth profile characterized by a sharp reversal of phase of the tl~eta rhythm with a coincident null point ill amplitude in the stratum radiaium of (:A~ and by a point of maximum amplitude ventral to the point of phase rcvcrsal. Recelltiv, the present author investigated the patterns of theta rhythm in the freely moving r~i~. in this study, depth profiles were determined during conditions of REM sleep and voluntary movement, and lhe profiles were found to be type I121,e~, different i~r(~in the type I profile found in the curarized rabbit. The type II profile in the fi'eely mowng rat revealed a point of maximum amplitude at approximately the level of stratum pyramidale of CA~, and a second, more ventrally located, maximum in the dorsal blade of the dentate gyrus. There was a slowly developing phase reversal between the lwo peaks. Experiments in a third preparation, the curarized rat ~', revealed a type [ profile. These data raised several possibilities. Since the type il profile in the flccly moving rat was found during REM sleep and voluntary movement and the typc 1 profile in the curarized rabbit was found during applications of sensory stimulation, it was possible that the theta rhythm profile was dependent upon the behavioral condition in which the theta rhythm was generated: type 11 for REM sleep and voluntary motion and type I for sensory stimulation. It was also possible that the profiles were the same type 1 for all behavioral conditions in the freely moving rabbit, but that the different type II profile found in the rat was due to a species difference in thctagenerating mechanisms between the rat and the rabbit. A basis for this difference might possibly lie in neuroanatomical variations among species. Such a difference may indeed exist in the medial septal hippocampal projection. The medial septal nucleus has been identified as a pacemaker of theta rhythm TM. Degeneration studies in the rat have either reported a minimal projection from the medial septal nucleus to CA~ li,v' or have found no projection at alp :~. The minimal nature of this path~ay in the rat has recently been confirmed by an autoradiographic investigation ~:. On the other hand, degeneration studies in the cat have indicated a substantial medial septum-CA~ projection n,~. A medial septal-CA~ pathway has also been demonstrated e[ectrophysiologically in the rabbit I. Further, comparative studies of acetylcholinesterase in the rat and rabbit hippocampus have reported differences in distribution is. A third possible explanation for the occurrence of the type 11 profile in the freely moving rat and the type I profile in the curarized rabbit was that the normal profiles in both the freely moving rat and rabbit were type 11 but that curare changed the profile to type 1 in the curarized rabbit. The objective of the present experiment was to attempt to resolve these questions by determining the depth profiles of theta rhythm in the freely moving rabbit during three behavioral conditions: REM sleep, voluntary movement, and the application of sensory stimulation. It will be reported that the theta rhythm profiles during the three conditions are the same. In all three conditions the profiles were not at all similar to the type II found in the freely moving rat; they were similar, though not absolutely identical, to the type 1 profile occurring in the curarized rat and rabbit.
73 METHODS
The subjects were 20 male New Zealand rabbits weighing 4-6 pounds. The methods used for electrode implantation, recording, signal filtering, data reduction, and histology were the same as described in the preceding paper 92, with the exception that Pryonin red was used instead of cresyl violet as the histological stain to increase the contrast with the Prussian blue markings. Recordings were made during REM sleep, voluntary movement, and episodes in which a sensory stimulus was presented while the animal was motionless except for the slight movements accompanying breathing. The onset of REM sleep was judged by the characteristic drooping of the ears and rapid twitching of the eyelids and nose after a period of behaviorally quiet sleep. REM sleep so identified was always accompanied by a shift from a slow-wave sleep hippocampal record observed at the macroelectrodes (Fig. le, top and bottom traces) to theta rhythm (Fig. lh). The state of motion of the animal was ascertained from the output of a motion indicator deriving its signal from the electrical motion artifact of an unshielded wire attached to the rabbit's head, as well as by direct observation. The animal was observed from a position 1 m in front of the open latticework testing enclosure. Instantaneous judgements of behavior within the categories described were expressed by depressing one of several keys on a small console. The coded output of the console was recorded simultaneously with recordings from the motion indicator and from the hippocampal electrodes. Minimal sensory stimuli, such as movement of an object within the animal's view, were generally sufficient to elicit theta rhythm. Testing was carried out by lowering the microelectrode, starting from the neocortex, and recording theta rhythm at each test point during voluntary movement, REM sleep, and during the presentation of sensory stimulation when the animal was awake but not moving. In certain penetrations, one or two behavioral test conditions were omitted. RESULTS
Phase reversal and null Fig. 1 shows recordings from a representative experiment. The location of the microelectrode path and the positions of the fixed reference electrodes are shown in the inset in Fig. 3. Upper and lower records in Fig. 1 are from the reference electrodes and the middle records are from the microelectrode. Signals from dorsal and ventral reference electrodes were approximately phase-reversed with respect to one another. Microelectrode penetration dorsal to the stratum radiatum showed that, in all behavioral conditions, signals from the microelectrode were isomorphic to signals from the dorsal reference electrode (the term isomorphic signifies that the signals were approximately in phase and had similar wave shapes and amplitude fluctuations). This isomorphism between dorsal reference electrode and microelectrode signals is shown in panel (a). The recordings were taken with the microelectrode in the pyramidal layer of CA1. On penetration of the proximal region of stratum radiatum, the
74 (£3) 1.40veN3er~ . . . . . . .
.
.
.
:
.
m;: r:;( h (;If :,::;
Movement
:,,.
%r'lD:r~
t
,~1; J,
! •
,,,j!,,,,:,:, ;
i
refit r O I elec~rr3 Je
(c)
::
!
d:ort~QI t *L':"rHI
. . . . . . . . . . . . . . .
:
[
i
i sec
( d)
I
REM
sleep
. . . . .
'
Jl,
I
(e)
Slow
wove sleep
f)
r.,1ovemen!
12oo~v
(CJ)
Slill,
sensory
slimulus
(h)
REM
sleep
Fig. I. Records from a representative penetration. Electrode positions are shown in inset of Fig. 3. Upper record is from dorsal reference electrode (No. 2), middle record is from microelectrode, lower record is from ventral reference electrode (No. I). Voltage scales for the reference electrodes are the same throughout. Amplifier gain is halved for microelectrode signal in (g) and (h). Signals from electrodes 2 and 1 are phase-reversed throughout. In recording (a) the microelectrode is dorsal to stratum radiatum. The behavioral condition is movement. In (b)-(f), the microelectrode is at the null position in stratum radiatum. Records (b)-(d) show the null at same microelectrode position in three behavioral conditions. Animal is in slow-wave sleep in (e), and moves in (f). In (g) and (h). the microelectrode is at a position ventral to stratum radiatum.
theta r h y t h m disappeared and a null signal was recorded. The null appeared at the same microelectrode depth in all behavioral conditions. Panels (b)-(d) show recordings at a single position of the microelectrode in s t r a t u m r a d i a t u m during the three behavioral conditions, in panel (b) theta r h y t h m was evoked by a visual stimulus while the a n i m a l was motionless, panel (c) was recorded while the rabbit was m o v i n g s p o n t a n e o u s l y across the test enclosure, a n d panel (d) was recorded d u r i n g R E M sleep. The transition of an episode of slow-wave sleep to m o v e m e n t with the microelectrode in the null position is shown in panels (e) a n d (f). D u r i n g slow-wave sleep all three electrodes show the high-amplitude, irregular signals characteristic of this
75
[,oo.
Fig. 2. Penetration to determine location of phase reversal and null. The penetration was terminated 40 pm ventral to the position at which the null was first detected. Three such penetrations located the null 50-150/~m below the pyramidal layer. stage of sleep s. The animal then awoke and moved, and in (f) it is seen that the signals at the reference electrodes show theta activity while the microelectrode records a null signal. The null extended through a dorsoventral span of approximately 90 #m, ventral to which the microelectrode recorded high-amplitude theta rhythm, isomorphic with the signal from the ventral reference electrode and phase-reversed from the microelectrode signals recorded dorsal to the null. Panels (g) and (h) show signals recorded during sensory stimulation and R E M sleep. The similarity of amplitude modulations between the microelectrode and the ventral reference electrode was lost during REM sleep but the in-phase relationship was maintained. On the average, signals from the two reference electrodes were phase-reversed within 5 ° and microelectrode signals were in phase with their reference electrode signals within 7 ° . In order to determine the precise location of the phase reversal and null, three penetrations were terminated within the null region. Fig. 2 shows the result of one such penetration. The experiment was terminated 40/~m after a nnll signal was first detected. The three penetrations showed the location of the null to be in the stratum radiatum, approximately 50-150 #m below the pyramidal layer.
Amplitude profile Fig. 3 is the amplitude profile of the experiment illustrated in Fig. I. Amplitude ratio profiles were plotted from the computer print-outs z2 from the three behavioral conditions. The three profiles showed a peak at the same microelectrode depth. Before plotting, the three curves were normalized to a common value at the peak. This value was 1130 #V, which was the average amplitude of theta rhythm recorded during REM sleep at this point in the brain. Fig. 3 shows that the amplitude profiles for the three behaviors were virtually
76
1200 [1
09
o I000 ::t. d o 800 (1) (19 ©
,
71
© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
H I P P O C A M P A L T H E T A R H Y T H M . 1I. D E P T H PROFILES IN T H E FREELY MOVING RABBIT
JONATHAN WINSON
The Rockefeller University, New York, N.Y. 10021 (U.S.A.) (Accepted July 14th, 1975)
SUMMARY
Depth profiles of hippocampal theta rhythm were investigated in the freely moving rabbit during three behavioral conditions: REM sleep, voluntary movement, and during sensory stimulation applied to the motionless animal. Profiles were found to be the same in all three conditions. Dorsoventral microelectrode penetration of the dorsal hippocampus revealed an approximately uniform amplitude of theta rhythm in strata oriens and pyramidale of CA1. Further microelectrode advancement revealed a sharp reversal of phase and a coincident null in amplitude in the proximal stratum radiatum. There was also a peak of theta rhythm amplitude which occurred in the molecular layer of the dorsal blade of the dentate gyrus. These data imply that in the rabbit, as in the freely moving rat, there are two generators of theta rhythm in the dorsal hippocampus, one in the dentate gyrus and the other in the overlying CA1 layer. The data also indicate the existence of a species difference in generating systems between the rabbit and the rat.
1NTRODUCTION
Species differences exist in the conditions in which theta rhythm occurs naturally in the freely moving rat and rabbit. In the rat, theta rhythm is present only during rapid eye movement (REM) sleep and voluntary motion 19,2°, while in the rabbit, theta rhythm appears in these two conditions but can also be elicited by sensory stimulation applied to the motionless animal6, s,l°. Early investigations of theta rhythm in the rabbit~, v used curarized preparations. In these experiments, the only natural condition in which theta rhythm was evoked was the application of sensory stimulation. The rhythm was also elicited by various electrical and pharmacological means. In these studies in the curarized rabbit, dorsoventral microelectrode penetrations of the hippocampus revealed the presence
72 of a type 12~ depth profile characterized by a sharp reversal of phase of the tl~eta rhythm with a coincident null point ill amplitude in the stratum radiaium of (:A~ and by a point of maximum amplitude ventral to the point of phase rcvcrsal. Recelltiv, the present author investigated the patterns of theta rhythm in the freely moving r~i~. in this study, depth profiles were determined during conditions of REM sleep and voluntary movement, and lhe profiles were found to be type I121,e~, different i~r(~in the type I profile found in the curarized rabbit. The type II profile in the fi'eely mowng rat revealed a point of maximum amplitude at approximately the level of stratum pyramidale of CA~, and a second, more ventrally located, maximum in the dorsal blade of the dentate gyrus. There was a slowly developing phase reversal between the lwo peaks. Experiments in a third preparation, the curarized rat ~', revealed a type [ profile. These data raised several possibilities. Since the type il profile in the flccly moving rat was found during REM sleep and voluntary movement and the typc 1 profile in the curarized rabbit was found during applications of sensory stimulation, it was possible that the theta rhythm profile was dependent upon the behavioral condition in which the theta rhythm was generated: type 11 for REM sleep and voluntary motion and type I for sensory stimulation. It was also possible that the profiles were the same type 1 for all behavioral conditions in the freely moving rabbit, but that the different type II profile found in the rat was due to a species difference in thctagenerating mechanisms between the rat and the rabbit. A basis for this difference might possibly lie in neuroanatomical variations among species. Such a difference may indeed exist in the medial septal hippocampal projection. The medial septal nucleus has been identified as a pacemaker of theta rhythm TM. Degeneration studies in the rat have either reported a minimal projection from the medial septal nucleus to CA~ li,v' or have found no projection at alp :~. The minimal nature of this path~ay in the rat has recently been confirmed by an autoradiographic investigation ~:. On the other hand, degeneration studies in the cat have indicated a substantial medial septum-CA~ projection n,~. A medial septal-CA~ pathway has also been demonstrated e[ectrophysiologically in the rabbit I. Further, comparative studies of acetylcholinesterase in the rat and rabbit hippocampus have reported differences in distribution is. A third possible explanation for the occurrence of the type 11 profile in the freely moving rat and the type I profile in the curarized rabbit was that the normal profiles in both the freely moving rat and rabbit were type 11 but that curare changed the profile to type 1 in the curarized rabbit. The objective of the present experiment was to attempt to resolve these questions by determining the depth profiles of theta rhythm in the freely moving rabbit during three behavioral conditions: REM sleep, voluntary movement, and the application of sensory stimulation. It will be reported that the theta rhythm profiles during the three conditions are the same. In all three conditions the profiles were not at all similar to the type II found in the freely moving rat; they were similar, though not absolutely identical, to the type 1 profile occurring in the curarized rat and rabbit.
73 METHODS
The subjects were 20 male New Zealand rabbits weighing 4-6 pounds. The methods used for electrode implantation, recording, signal filtering, data reduction, and histology were the same as described in the preceding paper 92, with the exception that Pryonin red was used instead of cresyl violet as the histological stain to increase the contrast with the Prussian blue markings. Recordings were made during REM sleep, voluntary movement, and episodes in which a sensory stimulus was presented while the animal was motionless except for the slight movements accompanying breathing. The onset of REM sleep was judged by the characteristic drooping of the ears and rapid twitching of the eyelids and nose after a period of behaviorally quiet sleep. REM sleep so identified was always accompanied by a shift from a slow-wave sleep hippocampal record observed at the macroelectrodes (Fig. le, top and bottom traces) to theta rhythm (Fig. lh). The state of motion of the animal was ascertained from the output of a motion indicator deriving its signal from the electrical motion artifact of an unshielded wire attached to the rabbit's head, as well as by direct observation. The animal was observed from a position 1 m in front of the open latticework testing enclosure. Instantaneous judgements of behavior within the categories described were expressed by depressing one of several keys on a small console. The coded output of the console was recorded simultaneously with recordings from the motion indicator and from the hippocampal electrodes. Minimal sensory stimuli, such as movement of an object within the animal's view, were generally sufficient to elicit theta rhythm. Testing was carried out by lowering the microelectrode, starting from the neocortex, and recording theta rhythm at each test point during voluntary movement, REM sleep, and during the presentation of sensory stimulation when the animal was awake but not moving. In certain penetrations, one or two behavioral test conditions were omitted. RESULTS
Phase reversal and null Fig. 1 shows recordings from a representative experiment. The location of the microelectrode path and the positions of the fixed reference electrodes are shown in the inset in Fig. 3. Upper and lower records in Fig. 1 are from the reference electrodes and the middle records are from the microelectrode. Signals from dorsal and ventral reference electrodes were approximately phase-reversed with respect to one another. Microelectrode penetration dorsal to the stratum radiatum showed that, in all behavioral conditions, signals from the microelectrode were isomorphic to signals from the dorsal reference electrode (the term isomorphic signifies that the signals were approximately in phase and had similar wave shapes and amplitude fluctuations). This isomorphism between dorsal reference electrode and microelectrode signals is shown in panel (a). The recordings were taken with the microelectrode in the pyramidal layer of CA1. On penetration of the proximal region of stratum radiatum, the
74 (£3) 1.40veN3er~ . . . . . . .
.
.
.
:
.
m;: r:;( h (;If :,::;
Movement
:,,.
%r'lD:r~
t
,~1; J,
! •
,,,j!,,,,:,:, ;
i
refit r O I elec~rr3 Je
(c)
::
!
d:ort~QI t *L':"rHI
. . . . . . . . . . . . . . .
:
[
i
i sec
( d)
I
REM
sleep
. . . . .
'
Jl,
I
(e)
Slow
wove sleep
f)
r.,1ovemen!
12oo~v
(CJ)
Slill,
sensory
slimulus
(h)
REM
sleep
Fig. I. Records from a representative penetration. Electrode positions are shown in inset of Fig. 3. Upper record is from dorsal reference electrode (No. 2), middle record is from microelectrode, lower record is from ventral reference electrode (No. I). Voltage scales for the reference electrodes are the same throughout. Amplifier gain is halved for microelectrode signal in (g) and (h). Signals from electrodes 2 and 1 are phase-reversed throughout. In recording (a) the microelectrode is dorsal to stratum radiatum. The behavioral condition is movement. In (b)-(f), the microelectrode is at the null position in stratum radiatum. Records (b)-(d) show the null at same microelectrode position in three behavioral conditions. Animal is in slow-wave sleep in (e), and moves in (f). In (g) and (h). the microelectrode is at a position ventral to stratum radiatum.
theta r h y t h m disappeared and a null signal was recorded. The null appeared at the same microelectrode depth in all behavioral conditions. Panels (b)-(d) show recordings at a single position of the microelectrode in s t r a t u m r a d i a t u m during the three behavioral conditions, in panel (b) theta r h y t h m was evoked by a visual stimulus while the a n i m a l was motionless, panel (c) was recorded while the rabbit was m o v i n g s p o n t a n e o u s l y across the test enclosure, a n d panel (d) was recorded d u r i n g R E M sleep. The transition of an episode of slow-wave sleep to m o v e m e n t with the microelectrode in the null position is shown in panels (e) a n d (f). D u r i n g slow-wave sleep all three electrodes show the high-amplitude, irregular signals characteristic of this
75
[,oo.
Fig. 2. Penetration to determine location of phase reversal and null. The penetration was terminated 40 pm ventral to the position at which the null was first detected. Three such penetrations located the null 50-150/~m below the pyramidal layer. stage of sleep s. The animal then awoke and moved, and in (f) it is seen that the signals at the reference electrodes show theta activity while the microelectrode records a null signal. The null extended through a dorsoventral span of approximately 90 #m, ventral to which the microelectrode recorded high-amplitude theta rhythm, isomorphic with the signal from the ventral reference electrode and phase-reversed from the microelectrode signals recorded dorsal to the null. Panels (g) and (h) show signals recorded during sensory stimulation and R E M sleep. The similarity of amplitude modulations between the microelectrode and the ventral reference electrode was lost during REM sleep but the in-phase relationship was maintained. On the average, signals from the two reference electrodes were phase-reversed within 5 ° and microelectrode signals were in phase with their reference electrode signals within 7 ° . In order to determine the precise location of the phase reversal and null, three penetrations were terminated within the null region. Fig. 2 shows the result of one such penetration. The experiment was terminated 40/~m after a nnll signal was first detected. The three penetrations showed the location of the null to be in the stratum radiatum, approximately 50-150 #m below the pyramidal layer.
Amplitude profile Fig. 3 is the amplitude profile of the experiment illustrated in Fig. I. Amplitude ratio profiles were plotted from the computer print-outs z2 from the three behavioral conditions. The three profiles showed a peak at the same microelectrode depth. Before plotting, the three curves were normalized to a common value at the peak. This value was 1130 #V, which was the average amplitude of theta rhythm recorded during REM sleep at this point in the brain. Fig. 3 shows that the amplitude profiles for the three behaviors were virtually
76
1200 [1
09
o I000 ::t. d o 800 (1) (19 ©
,
