Download Letter to Neuroscience

PII: S 0 3 0 6 - 4 5 2 2 ( 0 1 ) 0 0 1 3 1 - 2
Neuroscience Vol. 104, No. 3, pp. 619^625, 2001
ß 2001 IBRO. Published by Elsevier Science Ltd
Printed in Great Britain. All rights reserved
0306-4522 / 01 $20.00+0.00
www.elsevier.com/locate/neuroscience
Letter to Neuroscience
THETA-RHYTHMICALLY FIRING NEURONS IN THE ANTERIOR
THALAMUS: IMPLICATIONS FOR MNEMONIC FUNCTIONS OF PAPEZ'S
CIRCUIT
R. P. VERTES,* Z. ALBO and G. VIANA DI PRISCO
Center for Complex Systems and Brain Sciences, Florida Atlantic University, Boca Raton, FL 33431, USA
Key words: hippocampus, memory, mammillary body, cingulate cortex, head direction cells, diencephalic amnesia.
established that the mammillary bodies project massively
to the anterior thalamus (Shibata, 1992), which taken
together with the demonstration that mammillary body
cells ¢re synchronously with theta, suggests that the mammillary bodies may act on the anterior thalamus, possibly
in the manner that the hippocampus acts on the mammillary bodies, to rhythmically activate cells of the anterior
thalamus at theta frequency. We demonstrated that
approximately 75%
% of cells of the anterior ventral nucleus
of the thalamus ¢re rhythmically synchronous with the
hippocampal theta rhythm and the activity of 46%
% of
these anterior ventral neurons was highly correlated with
theta.
These ¢ndings, together with demonstration of thetarhythmically ¢ring cells in other structures of Papez's circuit, indicate that a theta-rhythmic signal may resonate
throughout Papez's circuit, possibly involved in the control
of mnemonic functions of the circuit. ß 2001 IBRO.
Published by Elsevier Science Ltd. All rights reserved.
In 1937 Papez described an anatomical circuit (or loop)
beginning and ending in the hippocampal formation that he
proposed subserved emotional experience (Papez, 1937).
Speci¢cally, the projections of the circuit were as follows:
hippocampal formationC mammillary bodiesC anterior
thalamusC cingulate cortexC parahippocampal gyrusC
hippocampal formation. Although the circuit has been
re¢ned based on subsequent anatomical ¢ndings (Amaral
and Witter, 1995; Shibata, 1992; Van Groen and Wyss,
1995), the major links of the circuit unquestionably represent a prominent system of connections in the mammalian
brain. Hence, the enduring nature of `Papez's circuit'.
Unlike, however, its persistence as anatomical entity, the
proposed functional role for the circuit has been less resilient. The early notion that Papez's circuit subserves emotional experience/expression has been abandoned (LeDoux,
1993) and replaced by the proposal that it is primarily
involved in mnemonic functions (Aggleton and Brown,
1999). Lesions of each of the major components of the
circuit have been shown to disrupt memory (Aggleton
and Brown, 1999; Sutherland et al., 1988; Sziklas and
Petrides, 1993). The mammillary bodies represent a
major output from the hippocampus in Papez's circuit
(Amaral and Witter, 1995). It has recently been shown
that cells of mammillary body ¢re rhythmically in bursts
synchronous with the theta rhythm of the hippocampus
(Bland et al., 1995; Kirk et al., 1996; Kocsis and Vertes,
1994, 1997) and that this rhythmical activity is dependent
upon the action of the hippocampus on the mammillary
bodies (Bland et al., 1995; Kirk et al., 1996). It is well
The present report examined the activity of cells of the
anterior thalamus (ATh) (anterior dorsal (AD), anterior
ventral (AV) and anterior medial (AM)) in urethaneanesthetized rats with respect to the hippocampal electroencephalogram (EEG)/theta rhythm. We found that
large percentages of cells (50^90%) in all three subdivisions of the ATh ¢red at signi¢cantly higher rates in the
presence than in the absence (control) of theta; that is,
`theta-on' cells. Theta-on cells were found within the
AV (49 of 52 neurons), AD (10 of 13 neurons) and
AM (5 of 7 neurons) nuclei of ATh. As depicted in
Fig. 1A, theta-on cells show a two- to four-fold increase
in mean rates of discharge from control to theta conditions with the largest increases seen with AD (3.11 þ 1.27
to 16.04 þ 3.06; P 6 0.003) and AV cells (7.86 þ 0.97 to
15.58 þ 1.16; P 6 0.005).
Interestingly, however, only cells of AV ¢red rhythmically in bursts synchronous with theta. No theta-rhythmical neurons were found in AD or AM [i.e., no strongly
rhythmical (SR) neurons, see below]. Approximately 75%
*Corresponding author. Tel.: +1-561-297-2362; fax: +1-561-2972363.
E-mail address: [email protected] (R. P. Vertes).
Abbreviations : AD, AM, and AV, anterior dorsal, anterior medial,
and anterior ventral nuclei of the anterior thalamus; ATh, anterior thalamus; EEG, electroencephalogram ; HD, head direction
(cells) ; HF, hippocampal formation; LTP, long-term potentiation; MB, mammillary bodies; SR, MR, NR, strong, moderate
and non-rhythmically ¢ring cells of ATh.
619
NSC 4976 26-6-01
620
R. P. Vertes et al.
of AV neurons ¢red rhythmically with theta (Fig. 1B)
and these AV cells were further subdivided into strongly
and moderately rhythmical (MR) neurons based on their
relative coherence with theta (Fig. 1B). The SR neurons
(46% of AV cells) showed mean coherence values of 0.58
(range: 0.284^0.782); while MR neurons (29% of AV
cells) showed mean coherence values of 0.20 (range:
0.13^0.28) (Fig. 1B). Non-rhythmical AV neurons
(NR; 25% of AV cells) exhibited mean coherences of
0.01 (range: 0.0^0.11) (Fig. 1B).
A SR cell of AV is depicted in Fig. 2. As shown in
Fig. 2A, a brief tail pinch produced a change from a
desynchronized to a theta pattern of hippocampal EEG
activity which outlasted the period of stimulation. This
was correlated with a change in AV cell discharge from
an irregular to a highly rhythmical pattern of discharge,
synchronized to theta (Fig. 2A). The change from a non
rhythmical (control) to a rhythmical pattern of activity
(theta condition) for the cell is exempli¢ed by the rhythmical peaks in the autocorrelogram (Fig. 2B), unit-theta
locked EEG oscillations (spike-triggered averaging) in
the crosscorrelogram (Fig. 2C) and by the pronounced
coherence between unit discharge and the hippocampal
EEG at theta frequency (about 3.3 Hz) (Fig. 2D). This
cell was located in the dorsomedial part of caudal AV
(Fig. 4).
Figure 3 depicts a MR AV cell (Fig. 3A) and a NR
AV cell (Fig. 3B). As shown, the MR cell ¢res rhythmically in bursts with theta (top traces of Fig. 3A). The
rhythmical activity is further exempli¢ed by the peaks in
the autocorrelogram, the unit-EEG locked oscillations in
the crosscorrelogram and the signi¢cant coherence
between unit and EEG signals at theta frequency (left
to right lower traces of Fig. 3A). Figure 3B depicts a
NR theta-on AV neuron. As shown, the cell shows no
change in patterns of activity from control to theta conditions (top traces of Fig. 3B). This is illustrated by the
absence of peaks in the auto and crosscorrelograms as
well as by the lack of coherence between unit-EEG signals at theta frequency (left to right lower traces of
Fig. 3B). Figure 4 schematically depicts the locations of
SR, MR and NR units in AV. As shown, rhythmical
cells were located throughout the rostrocaudal extent
of AV.
The present ¢ndings demonstrate, then, that approximately 75% of cells of the AV nucleus of thalamus ¢re
rhythmically with theta and the activity of about half of
them (46%) was highly correlated with the theta rhythm
(SR neurons). SR cells of AV show virtually the same
degree of pronounced rhythmicity with theta as do the
pacemaking cells of the medial septal nucleus that
directly drive theta in the hippocampus (Bland, 1986;
Vertes and Kocsis, 1997). The present ¢ndings together
with previous demonstrations of theta-rhythmic cells in
the mammillary bodies (MB) indicate that a `theta' signal
courses through the ¢rst two legs of Papez's circuit from
the hippocampus (MB and ATh) and suggest that a theta
signal may resonate throughout entire circuit. Some
theta-rhythmic cells have also been identi¢ed in the cingulate/retrosplenial cortex: the next leg of Papez's circuit
from ATh (Borst et al., 1987; Colom et al., 1988). The
essential restriction of theta-rhythmical cells to AV of
ATh is consistent with the demonstration that the intermediate MB (or lateral part of the medial MB) preferentially projects to AV, whereas the medial MB projects
to AM, and lateral MB to AD (Shibata, 1992) and the
intermediate MB contains the densest population of
theta-rhythmically ¢ring neurons (Bland et al., 1995;
Kirk et al., 1996; Kocsis and Vertes, 1994).
The theta rhythm of the hippocampal formation (HF)
is a large amplitude (1^2 mV) nearly sinusoidal oscillation of 5^12 Hz found in all mammalian species, including primates (Bland, 1986; Vinogradova, 1995; Vertes
and Kocsis, 1997). A growing body of evidence, including recent work in humans (Kahana et al., 1999;
Klimesch, 1999), indicates that the theta rhythm is
directly involved in mnemonic functions of the hippocampus (Huerta and Lisman, 1995; Staubli and Lynch,
1987; Vertes and Kocsis, 1997; Winson, 1978). Perhaps,
the strongest support for this position derives from the
demonstration that long-term potentiation (LTP) is optimally induced in the hippocampus with stimulation at
theta frequency (5^7 Hz) (Huerta and Lisman, 1995;
Staubli and Lynch, 1987; Vertes and Kocsis, 1997).
This has led to the proposal that theta may act as a
natural tetanizer producing the same (or similar types)
Fig. 1. (A) Mean discharge rates of neurons in the AV (n = 49),
AD (n = 10), and AM (n = 5) of the ATh during control and theta
conditions. (B) Mean discharge rates during control and theta
conditions and mean coherence values during theta conditions for
three types of AV neurons ^ SR (n = 24), MR (n = 15), and NR
(n = 13) cells.
NSC 4976 26-6-01
Theta rhythmic neurons in the anterior thalamus
621
Fig. 2. The discharge characteristics of a single cell in the AV of the ATh that ¢red rhythmically in bursts synchronous with
the theta rhythm of the hippocampus. (A) Upper traces : recordings of the hippocampal EEG and unit activity before and
during theta elicited with a tail pinch (horizontal bar). Lower traces: expanded record (from A) of a period after tail pinch
showing a strong correlation between unit bursts and theta. (B, C) Autocorrelograms and crosscorrelograms (spike-triggered
averaging) depicting the rhythmical discharge of the cell (B) locked to the theta rhythm (C) during theta but not control
conditions. (D) spectral and cross-spectral (coherence) plots showing peaks in the EEG and unit signals at theta frequency
and signi¢cant coherence between EEG and unit signals at theta frequency during theta (solid lines) but not during control
conditions (dotted lines).
long-term changes in the hippocampus (e.g., LTP) as
produced by external (arti¢cial) stimulation (Staubli
and Lynch, 1987; Vertes and Kocsis, 1997). A theta signal resonating through Papez's circuit might produce
LTP-like modi¢cations throughout the circuit possibly
to code complex behavioral sequences involving the circuit.
There is a remarkable correspondence in rats between
structures that contain `theta-rhythmic' neurons and
those containing `head direction' (HD) cells. These
include the MB, ATh, posterior cingulate (retrosplenial)
cortex and subiculum (Blair et al., 1997; Taube, 1998).
HD cells ¢re selectively when a rat is facing or oriented
in a particular direction (e.g., northeast) irrespective of
its location in its environment (Taube, 1998). HD cells
appear to be critically involved in spatial/navigational
learning; lesions of structures containing HD cells disrupt both hippocampal place cell activity and spatial
NSC 4976 26-6-01
622
R. P. Vertes et al.
Fig. 3.
NSC 4976 26-6-01
Theta rhythmic neurons in the anterior thalamus
Fig. 4. Series of schematic transverse sections through the rostral thalamus showing the locations of the three types of cells
recorded in the AV nucleus of the thalamus: SR (circle) and MR (triangle) theta-rhythmic cells and NR (square) cells. Open
symbols indicate cells marked by lesions, black symbols represent neurons identi¢ed by reference to marked cells. LD, lateral
dorsal nucleus of thalamus; MD, mediodorsal nucleus of thalamus; RT, reticular nucleus of thalamus; VAL, ventral anterior-lateral complex of thalamus; VL, lateral ventricle. Schematic sections adapted from Swanson, 1998.
6
Fig. 3. The discharge characteristics of a MR ¢ring cell (A) and a NR ¢ring cell (B) of the AV nucleus of the thalamus.
Upper traces in A: recordings of the hippocampal EEG and unit activity for an MR AV cell before and after theta elicited
with tail pinch (horizontal bar). Lower traces in (A) autocorrelograms (left), crosscorrelograms (center) and spectral/coherence
plots (right) depicting the rhythmical discharge of the cell (autocorrelogram) locked to the theta rhythm (crosscorrelogram) as
well as peaks in the EEG and unit signals (autospectra) and strong coherence between these signals at theta frequency during
theta (solid lines) but not during control conditions (dotted lines). Upper traces in (B) recordings of the hippocampal EEG
and unit activity for an NR AV cell before and after theta elicited with tail pinch (horizontal bar). Lower traces in (B)
autocorrelograms (left), crosscorrelograms (center) and spectral/coherence plots (right) showing an absence of a rhythmical
pattern of discharge of the cell during both control and theta conditions.
NSC 4976 26-6-01
623
624
R. P. Vertes et al.
learning (Aggleton and Brown, 1999; Mizumori et al.,
1994; Taube et al., 1992).
It would appear that directional information is particularly critical for a rat (and other species) when engaged
in locomotor/exploratory behaviors and less so during
non-locomotor activities such as grooming or consumatory acts. Accordingly, theta may serve as an important
signal involved in the di¡erential processing of HD activity under the two conditions (locomotion and grooming); that is, only when HD activity is coupled with
theta-rhythmic discharge (of structures of Papez's circuit)
is HD activity processed and used to guide spatial behaviors.
ATh and associated structures of Papez's circuit also
appear to serve a role in non-spatial behaviors. For
instance, Gabriel et al. (1983, 1995) demonstrated in
behaving rabbits that cells in the MB, ATh, cingulate
cortex and HF show marked changes in activity during
the development of a conditioned avoidance response,
and that lesions of each of these structures severely disrupts conditioned avoidance learning.
Finally, the involvement of subcortical structures of
Papez's circuit in mnemonic functions is underscored
by well documented cases of human amnesia resulting
from select destruction of major components of the circuit, notably to the MB or ATh. Restricted damage to
the either MB (Korsako¡'s or Wernicke^Korsako¡ syndrome) (Tanaka et al., 1997) or to the ATh (diencephalic
amnesia) (Von Cramon et al., 1985) produces severe
anterograde amnesia in humans.
In summary, we described a large subset of cells in the
AV of ATh that ¢re rhythmically synchronous with the
hippocampal theta rhythm. This ¢nding together with
previous demonstrations of theta-rhythmic neurons in
the HF, MB, and retrosplenial cortex suggests that a
theta-rhythmic signal may resonate throughout Papez's
circuit. We suggest that this theta-rhythmic signal may
serve a critical role in mnemonic functions associated
with Papez's circuit.
EXPERIMENTAL PROCEDURES
Experiments were conducted on 35 male Sprague^Dawley rats
(Charles River Laboratories, Wilmington, MA, USA) weighing
250^350 g. These experiments were approved by the Florida
Atlantic University Institutional Animal Care and Use Committee and conform to all Federal regulations and National Institutes of Health Guidelines for the Care and Use of Laboratory
Animals. All e¡orts were made to minimize the number of animals used and their su¡ering.
Experimental procedures have previously been described in
detail (Kocsis and Vertes, 1992, 1994, 1997). A catheter was
inserted in the femoral vein under methoxy£urane (metofane)
anesthesia. Anesthesia was then maintained by intravenous
administration of urethane for the duration of the experiment.
Bipolar electrodes (125-Wm Te£on-coated stainless steel wires,
separated by 1 mm at their tips) were used to record the hippocampal EEG. Reference and ground screw electrodes were
placed over frontal and occipital regions of the cortex. Commercial tungsten microelectrodes (5^10 M6) were used to record
single cell activity in the ATh. Well-isolated single units were
sampled without bias along multiple tracts through the ATh.
Unit and EEG signals were ampli¢ed and ¢ltered (unit:
300 Hz to 10 kHz; EEG: 1^75 Hz). Spike trains were taken
online using a voltage/time window discriminator and together
with EEG and unit signals were sampled at 14.3 kHz (70-Ws
sampling interval) with a 12-bit AD converter (RC Electronics)
during 60-s epochs and saved to computer disk. For purposes of
analysis the data were digitally re-sampled at 476 Hz.
Interspike interval histograms and perievent histograms were
calculated from the standard pulse train. Firing rates were computed during at least 10 s stationary segments and changes in
rates of discharge across conditions were assessed using a twotailed t-test. Autocorrelograms and unit/EEG crosscorrelograms
were calculated to assess unit periodicity and synchrony with
hippocampal theta. Autospectra and coherence functions relating single unit activity to the hippocampal EEG were computed
with a customized program (Kocsis and Vertes, 1992). Brie£y,
spike trains were convolved with a sinc function; a FFT was
then done on at least eight contiguous window segments 2.1 s in
duration; ¢nally, the spectra were smoothed with a three-point
moving average before coherence functions were computed. A
95% con¢dence level was determined for testing statistical signi¢cance.
Microelectrode recording sites were determined by histological analysis of locations of electrolytic lesions. At the end of the
recording session, rats were perfused intracardially (10% formalin), brains removed and stored. Histological (50-Wm) sections were taken with a freezing microtome and stained with
Cresyl Violet.
AcknowledgementsöThis work was supported by NIH Grant
NS35883 and NIMH Grant MH01476. These studies serve as
partial ful¢llment for requirements toward the degree of Doctor
of Philosophy for Zimbul Albo at Florida Atlantic University.
We thank Drs. J.A.S. Kelso, B. Kocsis, J.E. Lisman and G.
Lynch for their critical reading of this manuscript.
REFERENCES
Aggleton, J.P., Brown, M.W., 1999. Episodic memory, amnesia, and the hippocampal-anterior thalamic axis. Behav. Brain Sci. 22, 425^489.
Amaral, D.G., Witter, M.P., 1995. Hippocampal formation, in: G. Paxinos (Ed.), The Rat Nervous System, 2nd edn., Academic Press, London,
1995, pp. 443^493.
Blair, H.T., Lipscomb, B.W., Sharp, P.E., 1997. Anticipatory time intervals of head-direction cells in the anterior thalamus of the rat: implications
for path integration in the head-direction circuit. J. Neurophysiol. 78, 145^159.
Bland, B.H., 1986. The physiology and pharmacology of hippocampal formation theta rhythms. Prog. Neurobiol. 26, 1^54.
Bland, B.H., Konopacki, J., Kirk, I.J., Oddie, S.D., Dickson, C.T., 1995. Discharge patterns of hippocampal theta-related cells in the caudal
diencephalon of the urethan-anesthetized rat. J. Neurophysiol. 74, 322^333.
Borst, J.G.G., Leung, L.-W.S., MacFabe, D.F., 1987. Electrical activity of the cingulate cortex. II. Cholinergic modulation. Brain Res. 407, 81^93.
Colom, L.V., Christie, B.R., Bland, B.H., 1988. Cingulate cell discharge patterns related to hippocampal EEG and their modulation by muscarinic
and nicotinic agents. Brain Res. 460, 329^338.
Gabriel, M., Lambert, R.W., Foster, K., Orona, E., Sparenborg, S., Maiorca, R.R., 1983. Anterior thalamic lesions and neuronal activity in the
cingulate and retrosplenial cortices during discriminative avoidance behavior in rabbits. Behav. Neurosci. 97, 675^696.
Gabriel, M., Cuppernell, C., Shenker, J.I., Kubota, Y., Henzi, V., Swanson, D., 1995. Mammillothalamic tract transection blocks anterior
thalamic training-induced neuronal plasticity and impairs discriminative avoidance behavior in rabbits. J. Neurosci. 15, 1437^1445.
NSC 4976 26-6-01
Theta rhythmic neurons in the anterior thalamus
625
Huerta, P.T., Lisman, J.E., 1995. Bidirectional synaptic plasticity induced by a single burst during cholinergic theta oscillation in CA1 in vitro.
Neuron 15, 1053^1063.
Kahana, M.J., Sekular, R., Caplan, J.B., Kirschen, M., Madsen, J.R., 1999. Human theta oscillations exhibit task dependence during virtual maze
navigation. Nature 339, 781^784.
Kirk, I.J., Oddie, S.D., Konopacki, J., Bland, B.H., 1996. Evidence for di¡erential control of posterior hypothalamic, supramammillary, and
medial mammillary theta-related cellular discharge by ascending and descending pathways. J. Neurosci. 16, 5547^5554.
Klimesch, W., 1999. EEG alpha and theta oscillations re£ect cognitive and memory performance: a review and analysis. Brain Res. Rev. 29, 169^
195.
Kocsis, B., Vertes, R.P., 1992. Dorsal raphe neurons: synchronous discharge with the theta rhythm of the hippocampus in the freely behaving rat.
J. Neurophysiol. 68, 1463^1467.
Kocsis, B., Vertes, R.P., 1994. Characterization of neurons of the supramammillary nucleus and mammillary body that discharge rhythmically
with the hippocampal theta rhythm in the rat. J. Neurosci. 14, 7040^7052.
Kocsis, B., Vertes, R.P., 1997. Phase relations of rhythmic neuronal ¢ring in the supramammillary nucleus and mammillary body to the hippocampal theta activity in urethane anesthetized rats. Hippocampus 7, 204^214.
LeDoux, J.E., 1993. Emotional memory systems in the brain. Behav. Brain Res. 58, 69^79.
Mizumori, S.J.Y., Miya, D.Y., Ward, K.E., 1994. Reversible inactivation of the lateral dorsal thalamus disrupts hippocampal place representation
and impairs spatial learning. Brain Res. 644, 168^174.
Papez, J.W., 1937. A proposed mechanism of emotion. Arch. Neurol. Psychiatr. 38, 725^734.
Shibata, H., 1992. Topographic organization of subcortical projections to the anterior thalamic nuclei in the rat. J. Comp. Neurol. 323, 117^127.
Staubli, U., Lynch, G., 1987. Stable hippocampal long-term potentiation elicited by `theta' pattern stimulation. Brain Res. 435, 227^234.
Sutherland, R.J., Whishaw, I.Q., Kolb, B., 1988. Contributions of cingulate cortex to two forms of spatial learning and memory. J. Neurosci. 8,
1863^1872.
Swanson, L.W., 1998. Brain Maps: Structure of the Rat Brain. Elsevier, Amsterdam.
Sziklas, V., Petrides, M., 1993. Memory impairments following lesions to the mammillary region of the rat. Eur. J. Neurosci. 5, 525^540.
Tanaka, Y., Miyazawa, Y., Adalka, F., Yamada, T., 1997. Amnesia following damage to the mammillary bodies. Neurology 48, 160^165.
Taube, J.S., 1998. Head direction cells and the neurophysiological basis for a sense of direction. Prog. Neurobiol. 55, 225^256.
Taube, J.S., Kesslak, J.P., Cotman, C.W., 1992. Lesions of the rat postsubiculum impair performance on spatial tasks. Behav. Neural Biol. 57,
131^143.
Van Groen, T., Wyss, J.M., 1995. Projections from the anterodorsal and anteroventral nucleus of the thalamus to the limbic cortex in the rat.
J. Comp. Neurol. 358, 584^604.
Vinogradova, O.S., 1995. Expression, control, and probable functional signi¢cance of the neuronal theta-rhythm. Prog. Neurobiol. 45, 523^583.
Vertes, R.P., Kocsis, B., 1997. Brainstem-diencephalo-septohippocampal systems controlling the theta rhythm of the hippocampus. Neuroscience
81, 893^926.
Von Cramon, D.Y., Hebel, N., Schuri, U., 1985. A contribution to the anatomical basis of thalamic amnesia. Brain 108, 993^1008.
Winson, J., 1978. Loss of hippocampal theta rhythm results in spatial memory de¢cit in the rat. Science 201, 160^163.
(Accepted 21 March 2001)
NSC 4976 26-6-01