Synchronized Oscillatory Discharges of Mitral/Tufted Cells With
Different Molecular Receptive Ranges in the Rabbit Olfactory Bulb
HIDEKI KASHIWADANI,1,2 YASNORY F. SASAKI,1 NAOSHIGE UCHIDA,1 AND KENSAKU MORI1
Laboratory for Neuronal Recognition Molecules, Brain Science Institute, The Institute of Physical and Chemical Research,
2-1 Hirosawa, Wako, Saitama 351-0198; and 2Laboratory of Neuroscience, Graduate School of Engineering Science, Osaka
University, 1-3 Machikaneyama, Toyonaka, Osaka 560-0043, Japan
1
INTRODUCTION
To cope with a huge variety of odor molecules, the mammalian olfactory system expresses ;500 –1,000 types of odorant receptor in the sensory neurons of the olfactory epithelium
(Buck and Axel 1991; Lancet and Ben-Arie 1993; Mombaerts
1999; Sullivan et al. 1996). Because a particular object like a
rose, for example, emits dozens of specific odor molecules,
their nasal inhalation activates a specific combination of many
odorant receptors. Perception of the olfactory image of objects
therefore requires that the central olfactory system either combines or compares responses across the numerous types of
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1786
odorant receptor. However, the neuronal mechanisms for integrating signals from different receptors are not yet known.
In the large repertoire of odorant receptors, individual olfactory sensory neurons most probably express just one type
(Malnic et al. 1999). Thousands of olfactory sensory neurons
expressing a given odorant receptor project their axons (olfactory axons) selectively to only a few defined glomeruli in the
main olfactory bulb (MOB; Fig. 1) (Mombaerts et al. 1996;
Ressler et al. 1994; Vassar et al. 1994). Thus an individual
glomerulus is thought to be a functional unit representing a
single type (or a few types) of odorant receptor. Within the
glomerulus, olfactory axons make excitatory synaptic connections with dendritic tufts of mitral and tufted cells, which are
principal neurons of the MOB. Individual mitral/tufted cells
project a single primary dendrite to a single glomerulus. Therefore signals from different types of odorant receptor are sorted
into different glomeruli and transmitted to different mitral/
tufted cells (Fig. 1). In support of the hypothesis that individual
glomeruli represent a single odorant receptor, we have demonstrated elsewhere that individual mitral/tufted cells associated with a single glomerulus are tuned to specific features of
odor molecules (Imamura et al. 1992; Katoh et al. 1993; Mori
et al. 1992; Mori and Yoshihara 1995). The olfactory “identity”
of a given object is therefore thought to be coded by a specific
combination of activated glomeruli.
Stimulation of the olfactory epithelium with a mixture of
odorous compounds or even with a single compound causes
activation of glomeruli and mitral/tufted cells that are in many
cases distributed over several discrete regions of the MOB
(Mori and Yoshihara 1995; Shepherd 1994; Stewart et al.
1979). In addition, the inhalation of odor molecules elicits a
prominent oscillation (30 – 80 Hz) of local field potentials in
the MOB, suggesting that many mitral/tufted cells respond
with synchronized spike discharges to the odor stimulation
(see, for example, Adrian 1950; Bressler 1987; Bressler and
Freeman 1980; Mori et al. 1992; Mori and Takagi 1977).
Analysis of the oscillatory local field potential (OLFP) indicated that dendrodendritic reciprocal synaptic interactions between mitral/tufted cells and granule cells are responsible for
generating the OLFP in the MOB (Mori and Takagi 1977; Rall
et al. 1966; Shepherd and Greer 1990). These observations
raise the possibility that, during odor stimulation, synchronized
spike responses may occur in a number of mitral/tufted cells
associated with a specific subset of glomeruli representing a
selective combination of odorant receptors.
Because the transient synchronization of spike responses
0022-3077/99 $5.00 Copyright © 1999 The American Physiological Society
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
Kashiwadani, Hideki, Yasnory F. Sasaki, Naoshige Uchida, and
Kensaku Mori. Synchronized oscillatory discharges of mitral/tufted
cells with different molecular receptive ranges in the rabbit olfactory
bulb. J. Neurophysiol. 82: 1786 –1792, 1999. Individual glomeruli in
the mammalian olfactory bulb represent a single or a few type(s) of
odorant receptors. Signals from different types of receptors are thus
sorted out into different glomeruli. How does the neuronal circuit in
the olfactory bulb contribute to the combination and integration of
signals received by different glomeruli? Here we examined electrophysiologically whether there were functional interactions between
mitral/tufted cells associated with different glomeruli in the rabbit
olfactory bulb. First, we made simultaneous recordings of extracellular single-unit spike responses of mitral/tufted cells and oscillatory
local field potentials in the dorsomedial fatty acid–responsive region
of the olfactory bulb in urethan-anesthetized rabbits. Using periodic
artificial inhalation, the olfactory epithelium was stimulated with a
homologous series of n-fatty acids or n-aliphatic aldehydes. The
odor-evoked spike discharges of mitral/tufted cells tended to phaselock to the oscillatory local field potential, suggesting that spike
discharges of many cells occur synchronously during odor stimulation. We then made simultaneous recordings of spike discharges from
pairs of mitral/tufted cells located 300 –500 mm apart and performed
a cross-correlation analysis of their spike responses to odor stimulation. In ;27% of cell pairs examined, two cells with distinct molecular receptive ranges showed synchronized oscillatory discharges
when olfactory epithelium was stimulated with one or a mixture of
odorant(s) effective in activating both. The results suggest that the
neuronal circuit in the olfactory bulb causes synchronized spike
discharges of specific pairs of mitral/tufted cells associated with
different glomeruli and the synchronization of odor-evoked spike
discharges may contribute to the temporal binding of signals derived
from different types of odorant receptor.
SYNCHRONIZED SPIKE RESPONSES OF OLFACTORY BULB NEURONS
1787
METHODS
might contribute to temporal binding of input signals from
different receptors as suggested by studies of the mammalian
visual and somatosensory systems (Gray et al. 1989; Murthy
and Fetz 1996; Singer and Gray 1995) and the insect olfactory
system (Laurent 1996; Wehr and Laurent 1996), we examined
electrophysiologically whether odor stimulation causes synchronous oscillatory discharges in mitral/tufted cells of the
MOB. We first examined the temporal relationship between the
spike discharges of individual mitral/tufted cells and the phase
of oscillation of the OLFP during odor stimulation. We then
made recordings of single-unit spike discharges simultaneously
from pairs of mitral/tufted cells that were located 300 –500 mm
apart. After establishing the molecular receptive range (MRR)
properties of each cell, we performed a cross-correlation analysis of their spike responses to odor stimulation.
RESULTS
Spike discharges of mitral/tufted cells phase-locked to
oscillatory local field potentials
Stimulation of olfactory epithelium with fatty acids of short
carbon chain elicits a prominent OLFP in the dorsomedial
region of the rabbit MOB (Mori et al. 1992). As a first step to
examine the possible synchronous firing of mitral/tufted cells,
we examined the temporal relationship between spike discharges of mitral/tufted cells and the oscillation phases of
OLFP in the dorsomedial region of the MOB in urethananesthetized rabbits. Two micropipettes were inserted into the
dorsomedial region of the MOB: one for recording OLFP in the
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
FIG. 1. Schematic diagram illustrating the basic neuronal circuit of the
main olfactory bulb and the arrangement of recording micropipettes. Olfactory
sensory neurons (S) expressing the same odorant receptor converge their axons
onto a few defined glomeruli and make synaptic terminals on dendrites of
mitral cells (M), tufted cells, and periglomerular cells (PG). Individual mitral/
tufted cells project a single primary dendrite to a single glomerulus. Mitral/
tufted cells form dendrodendritic reciprocal synapses (black and white arrows)
with granule cells (G) in the external plexiform layer (EPL), and with PG cells
in the glomeruli. GL, glomerular layer; GRL, granule cell layer; LOT, lateral
olfactory tract; MCL, mitral cell layer; MOB, main olfactory bulb; OE,
olfactory epithelium.
Experiments were performed on 17 male adult rabbits (1.8 –2.5 kg,
Japanese White, Nihon SLC, Hamamatsu) anesthetized with urethan
(1.3 g/kg). Methods for surgical preparation, artificial inhalation of
odor-containing air into the nose, stimulation with odor molecules are
described in detail in a previous paper (Imamura et al. 1992). Briefly,
double tracheal cannulation was performed, one cannula being used
for respiration and the other for artificial inhalation of odor-containing
air. The cerebrospinal fluid was drained via the atlantooccipital membrane to minimize brain pulsation. All procedures involving animal
preparation were approved by the RIKEN Brain Science Institute
Animal Committee.
We used a homologous series of normal (n)-fatty acids (carbon
chain length: 2–9) and n-aliphatic aldehydes (carbon chain length:
3–10) for odor stimulation. Each odor molecule was diluted to 2 3
1022 (vol/vol) in odorless mineral oil and stored in a glass test tube
sealed with a screw cap. For odor stimulation, the test tube was
uncapped and then placed in front of the animal’s nose. The rate of
artificial inhalation (500 ms in duration) was maintained at once per
1.5 s. Each odor stimulation lasted 5 s covering 3 cycles of the
artificial inhalation.
OLFPs and single-unit discharges were recorded from the dorsomedial region of the MOB using glass micropipettes filled with 2 M
NaCl. DC resistance of the micropipettes was 1–2 MV for OLFP and
3–5 MV for extracellular recording of single-unit spikes. A pair of
stainless steel needles was inserted into the lateral olfactory tract
(LOT) for electrical stimulation (100 ms in duration). The configuration of the LOT-evoked field potential was used for monitoring the
position of the tip of the recording micropipettes in the MOB. For
simultaneous recording of OLFP and single-unit discharges of a
mitral/tufted cell, we used two micropipettes separated at their tips
;100 mm. For simultaneous recording from two mitral/tufted cells,
the tips of two micropipettes were separated at their tips 300 –500 mm
apart. The action potentials and OLFPs were differentially amplified
using band-pass filters in the range of 150 Hz to 3 kHz and 5–300
Hz, respectively. The recorded signals were stored in the computer
(PowerMac 7300) via AD converter with Spike 2 software (Cambridge Electronic Design, Cambridge, UK), and further off-line analyses were performed.
For each pair of mitral/tufted cells, cross-correlation histograms
(binwidth of 2 ms) of spike discharges were calculated using a
standard method. We fitted a Gabor function {damped sinusoid:
F(t) 5 A exp[2(ut 2 uu/s)2] cos [2pn(t 2 u)] 1 C} to each crosscorrelogram using x2 measurement and determined five parameters
(A, the center peak amplitude; C, the offset of the correlogram
modulation; u, the phase shift; s, decay constant; n, the sinusoid
frequency). For estimation of the strength of cross-correlation, we
calculated the relative modulatory amplitude (RMA) defined as the
ratio of the center peak amplitude (A) over the offset of correlogram
modulation (C) (Engel et al. 1990; König 1994). Spike responses were
considered synchronous when the RMA exceeded 0.3.
1788
H. KASHIWADANI, Y. F. SASAKI, N. UCHIDA, AND K. MORI
external plexiform layer (EPL) and the other for recording
extracellular single-unit spikes of mitral/tufted cells in the
mitral cell layer or the EPL (Fig. 1).
Figure 2A shows an example of simultaneous recordings of
single-unit discharges (trace 2) and the OLFP (trace 3). Inhalation of enanthic acid [CH3(CH2)5COOH: C(7)-COOH] elicited a train of spike discharges of this mitral/tufted cell and a
prominent sinusoidal (;37.6 Hz) OLFP. The spike discharges
started before the beginning of OLFP and tended to phase-lock
to OLFP during the period of large oscillation. Observation
with faster sweep speeds showed that the spike discharges
occurred mostly at the falling phase of the OLFP. To examine
the temporal relationship in more detail, each cycle (360°) of
the OLFP was divided into 12 different phases at 30° intervals
starting from the peak of positivity as 0° (top panel in Fig. 2B),
and the probability of spike discharge occurrence was plotted
against the different phases of the OLFP (phase-frequency
plotting, bottom panel in Fig. 2B). In the mitral/tufted cell
shown in Fig. 2B, ;91% of spike discharges occurred during
the falling phases (between 0 and 180°) of the OLFP. Such
detailed analysis of the temporal relationship was performed in
15 mitral/tufted cells sampled in the dorsomedial region, and
11 cells (;73% of cells analyzed) clearly showed spike discharges locked to the falling phase of the OLFP. An average
histogram obtained from the 11 cells (Fig. 2C) indicated that
most of the spike discharges occurred during the period between 0 and 150° with a peak between 60 and 90°. These
results suggest that a number of fatty acid–responsive mitral/
tufted cells in the dorsomedial region elicit their spikes synchronously during odor stimulation. However, the phase-locking of spikes to OLFP does not necessarily indicate spike
synchronization because mitral/tufted cells show diverse tem-
poral patterns of odor-evoked spike discharges (Imamura et al.
1992; Mori et al. 1992).
Simultaneous recording of spike discharges from a pair of
mitral/tufted cells
To examine the synchronous discharges more directly, we
made simultaneous recordings from two mitral/tufted cells in
the dorsomedial region of the MOB. A previous study with
horseradish peroxidase labeling (Buonviso et al. 1991) has
shown in rat that cell bodies of almost all pairs of mitral cells
innervating the same glomerulus are separated by ,120 mm,
which corresponds to the average diameter of a glomerulus
(127 mm) (Royet et al. 1989). To minimize the possibility of
recording from two mitral/tufted cells innervating a same glomerulus, we separated the tips of the two recording microelectrodes by .300 mm, that is, ;1.6 times greater than the
average diameter (190 mm) of a glomerulus in rabbit (Allison
and Warwick 1949). Mitral/tufted cells extend their secondary
dendrites tangentially ;850 mm (Mori et al. 1983) and form
numerous dendrodendritic synapses with granule cells. To increase the possibility of encountering mitral/tufted cells that
interact with each other via dendrodendritic synapses, the tips
of two microelectrodes were separated by up to 500 mm. Based
on the above estimations, we aimed to record from two mitral/
tufted cells located between 300 and 500 mm apart.
Previous studies show spatially overlapping distributions of
mitral/tufted cells having distinct tuning patterns to fatty acids
and aliphatic aldehydes (Imamura et al. 1992; Mori et al.
1992). This suggests that mitral/tufted cells innervating different glomeruli interact with each other via the local neuronal
circuit and show synchronized spike discharges when they are
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
FIG. 2. Spike discharges of mitral/tufted
cells phase-locked to oscillatory local field potential (OLFP). A: simultaneous recordings of
single-unit spike responses of a mitral/tufted
cell (trace 2) and OLFP (trace 3) during odor
stimulation. Enanthic acid was applied to the
nostril. A monitor for artificial inhalation is
shown in trace 1; the downward displacement
indicates inhalation. B and C: phase-frequency
histograms of spike discharge occurrence plotted against different phases of the sinusoidal
OLFP. Each cycle of OLFP was divided into
12 different phases at 30° intervals starting
from peak of positivity as 0°. Top diagrams in
B and C indicate 1 cycle of OLFP. In B, one
cycle of OLFP lasted ;26.6 ms and thus each
time bin corresponded to ;2.22 ms. Spike
discharges of the mitral/tufted cell occurred
mostly in the falling phase of the OLFP. C:
average histogram obtained from 11 cells.
SYNCHRONIZED SPIKE RESPONSES OF OLFACTORY BULB NEURONS
1789
whether the two cells fire synchronously using cross-correlation analysis of their spike discharges. Figure 4 shows an
example of the results obtained from simultaneous recordings
from two mitral/tufted cells. MRR of one cell (S12–2) covered
C(3)- and C(4)-fatty acids (COOH; enclosed by - - -; Fig. 4A),
whereas that of the other cell (S12–1) covered C(2)- to C(5)COOH and C(3)- to C(5)-aliphatic aldehydes (CHO; enclosed
by —). Because of the overlapping MRRs of the two cells,
stimulation with C(3)-COOH elicited burst spike discharges of
both cells. A cross-correlation histogram calculated for spike
discharges evoked by C(3)-COOH showed a clear central peak
at the time lag of 3 ms (Fig. 4B) indicating synchronization.
The spikes of the cell S12–2 typically occurred between 2 ms
before and 8 ms after the spike of S12–1. Cross-correlation
analysis (Fig. 4B) together with autocorrelation analysis (data
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
FIG. 3. Simultaneous recordings of spike responses from 2 mitral/tufted
cells (trace 3 and trace 4). A: monitor for artificial inhalation is shown in trace
2. Thick bar in trace 1 indicates the period of odor stimulation (caproic acid).
During the inhalation of odor-containing air, both cells responded with burst
spike discharges. B: traces of spike discharges during the period indicated by
the bracket under trace A4 are shown with a faster sweep speed. Spike
discharges of these 2 mitral/tufted cells tended to synchronize (arrows) during
inhalation of the odor-containing air.
simultaneously activated by one or a mixture of n-fatty acids
and/or n-aliphatic aldehydes. Figure 3A shows an example of
simultaneous recordings from two mitral/tufted cells in the
dorsomedial region. When the nasal epithelium was stimulated
with caproic acid [C(6)-COOH], both cells showed burst discharges during the inhalation of odor-containing air (Fig. 3A,
trace 3 and trace 4). Observation with a faster sweep speed
(Fig. 3B) showed that spiking of the two cells tended to occur
synchronously (indicated by arrows) during the late portion of
the burst discharges.
Cross-correlation analysis of spike discharges
To further examine the synchronized spike responses, we
made simultaneous recordings from two mitral/tufted cells,
determined the MRRs of both cells using a homologous series
of n-fatty acids and n-aliphatic aldehydes, and then examined
FIG. 4. Molecular receptive range (MRR; A) and cross-correlogram (B) of
a pair of simultaneously recorded mitral/tufted cells (S12–1 and S12–2). A:
MRRs of the 2 cells were overlapped at C(3)- and C(4)-COOH. B, top panel:
cross-correlogram computed for spike discharges evoked by C(3)-COOH.
Abscissa indicates the time lag of spike occurrence of cell S12–2 with reference to S12–1. The central peak indicates synchronization. Bottom panel:
cross-correlogram computed for spike discharges recorded before odor stimulation.
1790
H. KASHIWADANI, Y. F. SASAKI, N. UCHIDA, AND K. MORI
not shown) also showed the oscillatory nature of spike discharges at a frequency of ;36 Hz. It should be noted that the
synchronized oscillatory spike discharges of the two cells
occurred only during the inhalation of odor-containing air:
without odor stimulation the spike discharges of the two cells
did not show synchronization (bottom histogram of Fig. 4B).
Cross-correlation analysis was performed on 37 pairs of
mitral/tufted cells recorded in the dorsomedial region simultaneously activated by one or a mixture of odor molecules. A
clear synchronization (RMA . 0.3; see METHODS) of spike
discharges was observed in 10 pairs (27%) of mitral/tufted
cells. In all the pairs except for one in which the determination
of MRR was not completed, the MRR of one cell differed
significantly from that of the other cell. In three pairs including
the pair shown in Fig. 4A, the MRR of one cell was much
larger and completely involved that of the other cell. In four
pairs of mitral/tufted cells, the two cells showed distinct but
partially overlapping MRRs as exemplified in Fig. 5A. In two
pairs of mitral/tufted cells, there was no overlap of MRRs (e.g.,
cell pair shown in Fig. 5B).
In the mitral/tufted cell pair shown in Fig. 5A, the overlap of
MRR occurred on C(5)- and C(6)-COOH. Thus simultaneous
activation of both cells was obtained by stimulation of the
olfactory epithelium with either C(5)-COOH or C(6)-COOH.
Cross-correlation analysis of their spike responses showed that
C(5)-COOH induced a clear synchronization of spike discharges with a mean time lag of 5 ms (Fig. 5A), whereas
C(6)-COOH induced weaker synchronization (data not
shown). In the pair shown in Fig. 5B, the MRR of two cells
showed no overlap. Therefore a mixture of odor molecules
[C(5)-CHO and C(7)-CHO] was applied to the nose to activate
both cells simultaneously. As shown in Fig. 5B, the cross-
correlation histogram showed a robust synchronization of spike
discharges with a mean time lag of 2 ms during odor stimulation. In all the pairs analyzed, the temporal nature of synchronization was evident; the synchronization was elicited only
during the inhalation of odor molecules, and no synchronization was observed during the period before the odor stimulation
(bottom histograms of Fig. 5, A and B).
DISCUSSION
A number of previous studies (e.g., Adrian 1950; Bressler
1987; Bressler and Freeman 1980; Mori et al. 1992; Mori and
Takagi 1977) demonstrated a robust OLFP in the MOB, suggesting that many mitral/tufted cells fire in synchrony during
odor stimulation. However, detailed analyses of the temporal
relationship of spike discharges of mitral/tufted cells at high
time resolutions in milliseconds have never before been made.
In this study we show that spike discharges of many mitral/
tufted cells are phase-locked to the OLFP. Furthermore, we
applied cross-correlation analysis to spike discharges of pairs
of mitral/tufted cells and show that in selective pairs, synchronous spike discharges occur within a mean time lag of ,5 ms.
Because the tips of the two recording microelectrodes were
.300 mm apart, two mitral/tufted cells that were simultaneously recorded presumably innervate different glomeruli.
This idea is supported by the observation that in all pairs, the
MRR of one cell differed significantly from that of the other
cell. The present results thus suggest that in specific pairs of
mitral/tufted cells, each associated with a distinct glomerulus,
activation of both cells by odor stimulation causes synchronized oscillatory spike discharges during odor stimulation. In
view of the evidence that different glomeruli represent differ-
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
FIG. 5. MRR properties and cross-correlograms of mitral/tufted cell pairs. A: pair of mitral/tufted cells (S07–1 and S07–2)
having distinct but partially overlapping MRRs. The cross-correlogram was computed for spike discharges elicited by C(5)-COOH.
B: pair of mitral/tufted cells (S26 –1 and S26 –2) without overlapping MRRs. The cross-correlogram was computed for spike
discharges elicited by a mixture of odor molecules [C(5)-CHO and C(7)-CHO]. Bottom panels: cross-correlograms computed for
spike discharges recorded before odor stimulation.
SYNCHRONIZED SPIKE RESPONSES OF OLFACTORY BULB NEURONS
overlap of the territories of their secondary dendrites. Furthermore spike discharges of mitral/tufted cells occurred in the
falling phase of the OLFP just before the negativity, indicating
the synaptic depolarization of granule cell dendrites (Rall and
Shepherd 1968). Mechanisms other than that connected with
the dendrodendritic synapses between mitral/tufted cells and
granule cells may also be involved in the generation of the
synchronized oscillatory spike discharges. For example, the
synchronization might be mediated by dendrodendritic synaptic connections with periglomerular cells, or by the local circuit
via axon collaterals of mitral/tufted cells.
In the present study, synchronization of spike discharges
was observed only in 27% of pairs examined in the dorsomedial region; the rest did not show clear synchronization. This
suggests that synchronization occurs only in specific pairs of
mitral/tufted cells. In addition, this raises the possibility that
strong dendrodendritic synaptic connections via granule cells
that result in synchronized spike discharges are formed among
specific subsets of mitral/tufted cells associated with selective
subsets of odorant receptors. Because previous studies have
suggested a plastic nature of the dendrodendritic synaptic connections both in the accessory and main olfactory bulbs (Brennan and Keverne 1997; Kaba and Nakanishi 1995), the present
findings might be extended to hypothesize that the degree of
synchronization among specific subsets of mitral/tufted cells
might change in response to the history of previous olfactory
inputs. In other words, a plastic change in the dendrodendritic
synaptic interactions might result in a change in the strength of
temporal binding of signals originating from different odorant
receptors. The present method of examining synchronization of
spike activities of mitral/tufted cells with defined MRR properties provides a means for examining the above hypothesis.
We thank Dr. M. W. Miller of The Institute of Physical and Chemical
Research (RIKEN) for a critical reading of the manuscript; Drs. H. Nagao, L.
Masuda-Nakagawa, M. Yamaguchi, and H. von Campenhausen of RIKEN,
and Drs. F. Murakami and N. Yamamoto of Osaka University, for a number of
helpful suggestions; and K. Aijima of RIKEN for excellent technical assistance.
This work was supported in part by grants from the Ministry of Education,
Science, Sports and Culture of Japan and the Human Frontier Science Program. H. Kashiwadani was supported by a grant from the Junior Research
Associate Program in RIKEN, and N. Uchida was supported by the Special
Postdoctoral Researchers Program, RIKEN.
Address for reprint requests: K. Mori, Laboratory for Neuronal Recognition
Molecules, Brain Science Institute RIKEN, 2-1 Hirosawa, Wako, Saitama
351-0198, Japan.
Received 26 April 1999; accepted in final form 28 June 1999.
REFERENCES
ADRIAN, E. D. The electrical activity of the mammalian olfactory bulb. Electroencephalogr. Clin. Neurophysiol. 2: 377–388, 1950.
ALLISON, A. C. AND WARWICK, R.T.T. Quantitative observations on the olfactory system of the rabbit. Brain 72: 186 –197, 1949.
BRENNAN, P. A. AND KEVERNE, E. B. Neuronal mechanisms of mammalian
olfactory learning. Prog. Neurobiol. 51: 457– 481, 1997.
BRESSLER, S. L. Relation of olfactory bulb and cortex. I. Spatial variation of
bulbocortical interdependence. Brain Res. 409: 285–293, 1987.
BRESSLER, S. L. AND FREEMAN, W. J. Frequency analysis of olfactory system
EEG in cat, rabbit, and rat. Electroencephalogr. Clin. Neurophysiol. 50:
19 –24, 1980.
BUCK, L. AND AXEL, R. A novel multigene family may encode odorant
receptors: a molecular bases for odor recognition. Cell 65: 175–187, 1991.
BUONVISO, N., CHAPUT, M. A., AND SCOTT, J. W. Mitral cell-to glomerulus
connectivity: an HRP study of the orientation of mitral cell apical dendrites.
J. Comp. Neurol. 307: 57– 64, 1991.
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
ent odorant receptors, the results described above suggest that
pairs of mitral/tufted cells each receiving different odorant
receptor inputs show synchronized spike discharges during
odor stimulation.
The function of synchronized spike discharges in the olfactory bulb is not yet known. On the basis of observations in
other sensory systems, however, it can be speculated that this
synchronization provides a basis for the integration at the level
of olfactory cortex of signals originated from different odorant
receptors. If axons of the two mitral/tufted cells were to converge on the same target neuron in the olfactory cortex, the
synchronization of spike discharges may greatly increase the
probability of driving the target neuron because of temporal
summation of synaptic inputs from the two cells. OLFPs with
similar frequencies have been reported in the olfactory cortex
(Bressler 1987; Bressler and Freeman 1980), suggesting that
synaptic inputs from the MOB occur synchronously in the
olfactory cortex. In this way, synchronization of spike discharges of mitral/tufted cells may contribute to combining
signals derived from different odorant receptors at the level of
the olfactory cortex. Extension of the present study to include
analysis of olfactory cortical neurons thus might provide us
with a clue for understanding cellular mechanisms for the
integration and decoding in the olfactory cortex of odor information that is represented by spatial and temporal patterns of
mitral/tufted cell activity.
What is the mechanism for such a precise synchronization of
spike discharges of mitral/tufted cells? Mitral/tufted cells form
dendrodendritic reciprocal synapses with local inhibitory neurons, granule cells, and perigromerular cells (Fig. 1). The local
neuronal circuit via these interneurons is thought to mediate
functional interactions among mitral/tufted cells. Previous
studies also suggest a model in which dendrodendritic synaptic
interactions with granule cells provide the basis for the generation of rhythmic oscillatory activity of mitral/tufted cells
during odor stimulation (Mori and Takagi 1977; Rall and
Shepherd 1968; Rall et al. 1966; Shepherd and Greer 1990).
According to this model, initial spike discharges of mitral/
tufted cells in response to excitatory postsynaptic potentials
(EPSPs) from olfactory axon synchronously activate mitral/
tufted-to-granule dendrodendritic excitatory synapses. The depolarization generated in granule cell dendrites by the excitatory synaptic input causes negativity of the OLFP in the EPL.
The activated granule cells then synchronously inhibit mitral/
tufted cells via granule-to-mitral/tufted dendrodendritic inhibitory synapses, resulting in the synchronous cessation of spike
discharges of mitral/tufted cells. When the inhibition subsides,
the long-lasting EPSPs from olfactory axons restimulate mitral/
tufted cell spike discharges. In this way many mitral/tufted
cells show synchronized oscillatory spike discharges. In the
insect olfactory system, blockade of GABA-mediated inhibition in the antennal lobe, which is the counterpart of the
mammalian olfactory bulb, has also been shown to impair the
generation of the OLFPs in the mushroom body (MacLeod and
Laurent 1996).
The results of the present study agree well with such a model
demonstrating that the dendrodendritic reciprocal synapses are
responsible for the synchronization of mitral/tufted cells. In our
own study the distance between the two mitral/tufted cells
recorded was ,500 mm, so they might interact with each other
via the dendrodendritic synaptic circuit because of the possible
1791
1792
H. KASHIWADANI, Y. F. SASAKI, N. UCHIDA, AND K. MORI
Chemical Senses, edited by K. Katsuki, M. Sato, S. F. Takagi, and Y.
Oomura. Tokyo: Univ. of Tokyo Press, 1977, p. 33– 43.
MORI, K. AND YOSHIHARA, Y. Molecular recognition and olfactory processing
in the mammalian olfactory system. Prog. Neurobiol. 45: 585– 619, 1995.
MURTHY, V. N. AND FETZ, E. E. Synchronization of neurons during local field
potential oscillations in sensorymotor cortex of awake monkeys. J. Neurophysiol. 76: 3968 –3982, 1996.
RALL, W. AND SHEPHERD, G. M. Theoretical reconstruction of field potentials
and dendrodendritic synaptic interactions in olfactory bulb. J. Neurophysiol.
31: 884 –915, 1968.
RALL, W., SHEPHERD, G. M., REESE, T. S., AND BRIGHTMAN, M. W. Dendrodendritic synaptic pathway for inhibition in the olfactory bulb. Exp. Neurol.
14: 44 –56, 1966.
RESSLER, K. J., SULLIVAN, S. L., AND BUCK, L. B. Information coding in the
olfactory system: evidence for a stereotyped and highly organized epitope
map in the olfactory bulb. Cell 79: 1245–1255, 1994.
ROYET, J. P., JOURDAN, F., PLOYE, H., AND SOUCHEIER, C. Morphometric
modifications associated with early sensory experience in the rat olfactory
bulb. II. Stereological study of the population of olfactory glomeruli.
J. Comp. Neurol. 289: 594 – 609, 1989.
SHEPHERD, G. M. Discrimination of molecular signals by the olfactory receptor
neuron. Neuron 13: 771–790, 1994.
SHEPHERD, G. M. AND GREER, C. A. Olfactory bulb. In: The Synaptic Organization of the Brain, edited by G. M. Shepherd. New York: Oxford, 1990,
p. 139 –169.
SINGER, W. AND GRAY, C. M. Visual feature integration and the temporal
correlation hypotheses. Annu. Rev. Neurosci. 18: 555–586, 1995.
STEWART, W. B., KAUER, J. S., AND SHEPHERD, G. M. Functional organization
of rat olfactory bulb analyzed by 2-deoxyglucose method. J. Comp. Neurol.
185: 715–734, 1979.
SULLIVAN, S. L., ADAMSON, M. C., RESSLER, K. J., KOZAK, C. A., AND BUCK,
L. B. The chromosomal distribution of mouse odorant receptor genes. Proc.
Natl. Acad. Sci. USA 93: 884 – 888, 1996.
VASSAR, R., CHAO, S. K., SITCHERAN, R., NUÑEZ, J. M., VOSSHALL, L. B., AND
AXEL, R. Topographic organization of sensory projections to the olfactory
bulb. Cell 79: 981–991, 1994.
WEHR, M. AND LAURENT, G. Odor encoding by temporal sequences of firing in
oscillating neural assemblies. Nature 384: 162–166, 1996.
Downloaded from http://jn.physiology.org/ by 10.220.32.247 on June 17, 2017
ENGEL, A. K., KÖNIG, P., GRAY, C. M., AND SINGER, W. Stimulus-dependent
neuronal oscillation in cat visual cortex: inter-columnar interaction as determined by cross-correlation analysis. Eur. J. Neurosci. 2: 588 – 606, 1990.
GRAY, C. M., KÖNIG, P., ENGEL, A. K., AND SINGER, W. Oscillatory responses
in cat visual cortex exhibit inter-columnar synchronization which reflects
global stimulus properties. Nature 338: 334 –337, 1989.
IMAMURA, K., MATAGA, N., AND MORI, K. Coding of odor molecules by
mitral/tufted cells in rabbit olfactory bulb. I. Aliphatic compounds. J. Neurophysiol. 68: 1986 –2002, 1992.
KABA, H. AND NAKANISHI, S. Synaptic mechanisms of olfactory recognition
memory. Rev. Neurosci. 6: 125–141, 1995.
KATOH, K., KOSHIMOTO, H., TANI, A., AND MORI, K. Coding of odor molecules
by mitral/tufted cells in rabbit olfactory bulb. II. aromatic compounds.
J. Neurophysiol. 70: 2161–2175, 1993.
KÖNIG, P. A method for the quantification of synchrony and oscillatory
properties of neuronal activity. J. Neurosci. Methods 54: 31–37, 1994.
LANCET, D. AND BEN-ARIE, N. Olfactory receptors. Curr. Biol. 3: 668 – 674,
1993.
LAURENT, G. Dynamical representation of odors by oscillating and evolving
neural assemblies. Trends Neurosci. 19: 489 – 496, 1996.
MACLEOD, K. AND LAURENT, G. Distinct mechanisms for synchronization and
temporal patterning of odor-encoding neural assemblies. Science 274: 976 –
979, 1996
MALNIC, B., HIRONO, J., SATO, T., AND BUCK, L. B. Combinatorial receptor
cods for odors. Cell 96: 713–723, 1999.
MOMBAERTS, P. Molecular biology of odorant receptors in vertebrates. Annu.
Rev. Neurosci. 22: 487–509, 1999.
MOMBAERTS, P., WANG, F., DULAC, C., CHAO, S. K., NEMES, A., MENDELSOHN,
M., EDMONDSON, J., AND AXEL, R. Visualizing an olfactory sensory map.
Cell 87: 675– 686, 1996.
MORI, K., KISHI, K., AND OJIMA, H. Distribution of dendrites of mitral,
displaced mitral, tufted, and granule cells in the rabbit olfactory bulb.
J. Comp. Neurol. 219: 339 –355, 1983.
MORI, K., MATAGA, N., AND IMAMURA, K. Differential specificities of single
mitral cells in rabbit olfactory bulb for a homologous series of fatty acid
odor molecules. J. Neurophysiol. 67: 786 –789, 1992.
MORI, K. AND TAKAGI, S. F. Inhibition in the olfactory bulb: dendrodendritic
interaction and their relation to the induced waves. In: Food Intake and