Download Glutamate Receptors Form Hot Spots on Apical Dendrites of

Glutamate Receptors Form Hot Spots on Apical Dendrites of
Neocortical Pyramidal Neurons
A. FRICK, W. ZIEGLGÄNSBERGER, AND H.-U. DODT
Clinical Neuropharmacology, Max-Planck-Institute of Psychiatry, 80804 Munich, Germany
Received 29 December 2000; accepted in final form 14 May 2001
The apical dendrites of layer V pyramidal neurons of the
neocortex are a major receptive zone for excitatory glutamatergic synaptic input (Peters 1987). Release of glutamate activates N-methyl-D-aspartate (NMDA)– and ␣-amino-3-hydroxy5-methyl-4-isoxazolepropionic acid (AMPA)–type ionotropic
glutamate receptors at these synapses (Cauller and Connors
1994; Jones and Baughman 1988; Thomson et al. 1993), and it
has been shown immunocytochemically that these glutamate
receptor subtypes are differentially distributed in the various
layers of the neocortex (Huntley et al. 1994a; Petralia and
Wenthold 1992). Until recently, however, the specific location
and density of these receptors could not be determined on the
surface of single neocortical neurons (Currie et al. 1994; Dodt
et al. 1998).
Although electrophysiological and immunocytochemical
studies have shown that NMDA and AMPA receptors are
clustered at spines and often colocalized (Bekkers and Stevens
1989; Craig et al. 1994; Jones and Baughman 1991; Kornau et
al. 1995; O’Brien et al. 1998; Rao and Craig 1997), glutamate
receptors have also been detected and recorded at extrasynaptic
sites (Clark et al. 1997; Häusser and Roth 1997; Rao and Craig
1997; Rosenmund et al. 1995; Spruston et al. 1995). Dendritic
recordings in hippocampal pyramidal neurons and in cerebellar
Purkinje cells have provided evidence that the functional properties of extrasynaptic and synaptic glutamate receptors are
similar (Häusser and Roth 1997; Spruston et al. 1995).
To functionally map glutamate receptor activation on single
neurons, we developed an infrared-guided laser photostimulation technique (Dodt et al. 1999), combined with whole cell
patch-clamp recording, that allowed us to resolve local responses with high temporal and spatial resolution; necessary
criteria for mimicking synaptic transmission (Denk 1994; Dodt
et al. 1999; Katz and Dalva 1994; Pettit and Augustine 2000).
The focus of an ultraviolet (UV) laser beam was visually
guided by infrared videomicroscopy to a neuronal structure in
the brain slice, and flashes of UV light caused the release of
glutamate by localized photolysis from the caged compound.
The ability to focally release glutamate (⬍10 ␮m) at multiple
sites on a single neuron in brain slices overcomes limitations
inherent to other modes of glutamate application, such as
iontophoretic application or electrical stimulation. Furthermore, and in contrast to binding studies using labeled ligands,
functional receptors can be assigned to a particular compartment of a neuron.
We report a high resolution-mapping of functional glutamate
receptors, including areas of high density (hot spots), along the
apical dendrite of neocortical pyramidal neurons in brain
slices. The results described herein cannot distinguish between
synaptic and extrasynaptic glutamate receptors, but, given the
significantly greater density of glutamate receptors at synaptic
sites compared with extrasynaptic sites (Craig et al. 1994;
Jones and Baughman 1991; Kornau et al. 1995; O’Brien et al.
1998; Rao and Craig 1997), our results are more likely to
reflect this distribution of receptor sites. Hot spots of glutamate
receptors might be strategically located on the dendritic tree
and could have an important influence on neural information
processing for these neurons. We show that the generation of
Ca2⫹ spikes is facilitated by stimulation of these hot spots,
Present address and address for reprint requests: A. Frick, Div. of Neuroscience, S 700, Baylor College of Medicine, One Baylor Plaza, Houston, TX
77030 (E-mail: [email protected]).
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.
INTRODUCTION
1412
0022-3077/01 $5.00 Copyright © 2001 The American Physiological Society
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
Frick, A., W. Zieglgänsberger, and H.-U. Dodt. Glutamate receptors form hot spots on apical dendrites of neocortical pyramidal
neurons. J Neurophysiol 86: 1412–1421, 2001. Apical dendrites of
layer V cortical pyramidal neurons are a major target for glutamatergic synaptic inputs from cortical and subcortical brain regions. Because innervation from these regions is somewhat laminar along the
dendrites, knowing the distribution of glutamate receptors on the
apical dendrites is of prime importance for understanding the function
of neural circuits in the neocortex. To examine this issue, we used
infrared-guided laser stimulation combined with whole cell recordings
to quantify the spatial distribution of glutamate receptors along the
apical dendrites of layer V pyramidal neurons. Focally applied (⬍10
␮m) flash photolysis of caged glutamate on the soma and along the
apical dendrite revealed a highly nonuniform distribution of glutamate
responsivity. Up to four membrane areas (extent 22 ␮m) of enhanced
glutamate responsivity (hot spots) were detected on the dendrites with
the amplitude and integral of glutamate-evoked responses at hot spots
being three times larger than responses evoked at neighboring sites.
We found no association of these physiological hot spots with dendritic branch points. It appeared that the larger responses evoked at hot
spots resulted from an increase in activation of both ␣-amino-3hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methylD-aspartate (NMDA) receptors and not a recruitment of voltageactivated sodium or calcium conductances. Stimulation of hot spots
did, however, facilitate the triggering of both Na⫹ spikes and Ca2⫹
spikes, suggesting that hot spots may serve as dendritic initiation
zones for regenerative spikes.
GLUTAMATE HOT SPOTS ON NEOCORTICAL DENDRITES
which are, for the most part, not associated with dendritic
branch points (sites that have been hypothesized to be Ca2⫹
spike “hot spots”) (Llinas and Hess 1976; Llinas and Sugimori
1980). These findings might contribute to the understanding of
the synaptic connectivity in neocortical circuits (Denk 1994;
Huntley et al. 1994b; Katz and Dalva 1994), as well as of the
synaptic integration in dendrites (Johnston et al. 1996; Magee
et al. 1998; Stuart et al. 1997; Yuste and Tank 1996). Part of
this study has been presented in abstract form (Frick et al.
1998).
METHODS
1413
component of the photolytic reaction is in the microsecond range
(Wieboldt et al. 1994). The beam of the laser was directed via a
multimode optical fiber into the epifluorescence port of the microscope (Zeiss Axioskop), and focused through a ⫻60/0.9 NA objective
lens (Olympus) to a spot about 1 ␮m wide (see also Fig. 1). The
intensity of the UV light at the front of the objective was 0.5–2 mW.
The microscope was mounted on a motorized, three-dimensional stage
(Luigs and Neumann, Ratingen, Germany) connected to a personal
computer. After establishing a stable whole cell recording, 1-mM
caged glutamate was added to a small volume (3 ml) of the bath
solution, reoxygenated, and recirculated. Caged glutamate was bath
applied for at least 15 min before a visually guided light spot (using
IR-GC optics) was moved in small steps (with glutamate responses
Electrophysiology and data acquisition
Infrared-guided laser photostimulation
Caged glutamate (␥-CNB-caged L-glutamic acid, Molecular
Probes, Amsterdam) was flash photolyzed by a 354-nm UV laser
beam (Liconix), using 3- to 5-ms shuttered pulses (UniBlitz shutter
driver/timer, Rochester, New York) (Callaway and Katz 1993). The
properties of the caged glutamate have been described elsewhere
(Wieboldt et al. 1994). In short, the caged glutamate used in this study
has a quantum product yield of 0.14, and the half-life of the major
J Neurophysiol • VOL
FIG. 1. Infrared-guided laser photostimulation combined with whole cell
patch-clamp recording in the somatosensory cortex of rat brain slices. A:
schematic representation of the experimental set-up. B: schematic representation of the recording and stimulation geometry [the diameter of the ultraviolet
(UV) spots not to scale]. Montage of photographs of an infrared (IR) videomicroscopy image of a layer V pyramidal neuron. The pial surface of the
neocortex is toward the top. A and B: neurons in the brain slice were visualized
by using a combination of illumination with infrared light and the gradient
contrast system. At the same time, light pulses of a UV laser were directed via
a quartz fiber into the microscope and by a dichroic mirror onto the recorded
neuron. The intensity of the UV light at the front of the objective was 0.5–2
mW. The microscope could be positioned in x-y-z by remote controls. The
laser spot of an optical diameter of 1 ␮m formed by the objective in the
specimen plane was made visible before the experiment by a fluorescent paper,
and its position was marked on the TV monitor. By positioning the site of the
neuron to be stimulated at this point, the laser stimulation could be precisely
guided by visual control. C: lateral resolution of the laser stimulation. The data
were fitted with a Gaussian function (R ⫽ 0.998, n ⫽ 6). Half-maximal peak
amplitude was evoked when the light spot was moved 4.8 ␮m away from the
dendrite. Data are normalized to values at 0 ␮m. D: focal photolysis of caged
glutamate elicits fast glutamatergic responses that are mediated by N-methyl-Daspartate (NMDA) and ␣-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
(AMPA) receptors. Bath application of the AMPA receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoylbenzo(f)quinoxaline (NBQX) (1 ␮M) attenuates glutamatergic responses. The residual response is abolished by addition of the NMDA
receptor antagonist D-2-amino-5-phosphonovaleric acid (D-APV; 100 ␮M).
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
Patch-pipette recordings were obtained from the somata of layer V
pyramidal neurons in parasagittal neocortical slices (300 ␮m thick)
from 15- to 25-day-old male Sprague-Dawley rats, as described previously (Dodt et al. 1998). Individual neurons were visualized using
the newly developed infrared “gradient contrast” (IR-GC) optics
(Dodt et al. 1999). The extracellular solution consisted of the following (in mM): 125 NaCl, 25 NaHCO3, 25 glucose, 2.5 KCl, 1.25
NaH2PO4, 2 CaCl2, and 1 MgCl2 (saturated with 95% O2-5% CO2,
pH 7.4). Patch pipettes (DC resistance 4 –7 M⍀) were filled with the
following (in mM): 130 K-gluconate, 5 KCl, 10 HEPES, 0.5 EGTA,
2 Mg-ATP, and 5 glucose, pH 7.2, and in some cases Neurobiotin (5
mg/ml). Whole cell patch-pipette recordings were made with a SEC
1 l amplifier (npi Electronics, Tamm, Germany) with appropriate
bridge and capacitance compensation. Series resistances ranged from
8 to 25 M⍀ and were monitored on-line. Recordings were made at
room temperature (22–24°C). Data were low-pass filtered at 0.3–1.5
kHz, digitized at 1–5 kHz (ITC 16 A/D board, Instrutec, New York),
and stored and analyzed using Pulse 8.11 (HEKA, Germany) and Igor
Pro 3 (WaveMetrics). Surface and image plots were constructed using
Surfer32 (Golden Software). Neurons included in this study had
resting potentials of ⫺61 ⫾ 3 mV (mean ⫾ SD). Cell input resistances
ranged from 44 to 179 M⍀, and action potential amplitudes, measured
from resting membrane potential to peak, ranged from 90 to 116 mV.
Video images were taken during the experiments to document the
position of the soma being recorded, of the stimulation sites on the
neuronal surface and of a reference cell, which was filled with Neurobiotin at the end of the experiment. The distance between the
recorded neuron and the reference cell served as a measure of a
potential tissue shrinkage after fixation. After electrophysiological
recording, the slices were fixed in 4% paraformaldehyde, subsequently stained using cy3-streptavidin, and stained neurons were
imaged with a fluorescence microscope. Slices were not dehydrated
after fixation to minimize tissue shrinkage and were mounted in an
aqueous mounting medium. If necessary, minimal corrections for the
position of the branch points were made for tissue shrinkage. Each
stained cell had a pyramidal-shaped soma in layer V and an apical
dendrite extending to the pial surface with a terminal tuft, and with the
first major branch point 484 ⫾ 129 ␮m from the soma (mean ⫾ SD,
n ⫽ 11). When 500 ␮M CdCl2 was used, NaH2PO4 was omitted in the
bath to avoid precipitation. Statistical analysis was performed using
unpaired two-tailed t-tests.
1414
A. FRICK, W. ZIEGLGÄNSBERGER, AND H.-U. DODT
evoked at each step) from the soma to the distal apical dendrite with
a frequency of the flash photolysis of ⱕ0.1 Hz. The displacement of
the laser beam was performed by moving the microscope. Application
of 1-mM caged glutamate itself had no effect on the neuronal activity
of the recorded cells. It has previously been shown that ␥-CNB-caged
glutamate, at 1 mM concentration, does not desensitize glutamate
receptors in hippocampal neurons, and that caged glutamate does not
inhibit the activation of the glutamate receptors by 50 ␮M glutamate
(Wieboldt et al. 1994). Likewise, exposure of cells to the relatively
brief periods of uncaging light used in these experiments (in the
absence of caged glutamate) had no effect on neuronal activity. The
depth of the somata and dendrites stimulated was generally within the
first 60 ␮m from the surface of the brain slice, where dendrites could
still be visualized, but in most cases even closer to the surface.
RESULTS
Profile of glutamate responsivity along the apical dendrite
Mapping glutamate receptors along the apical dendrites (n ⫽
36) revealed a nonuniform distribution of the receptors (Figs.
2, 3, and 5B). UV flashes were standardized across experiments
for laser intensity and duration by eliciting responses of 4 –9
mV when delivered to the soma. The response amplitude
J Neurophysiol • VOL
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
After establishing stable whole cell patch-clamp recordings
from the somata of layer V pyramidal neurons in neocortical
slices, the recording chamber was perfused with solution containing 1 mM caged glutamate (Wieboldt et al. 1994). With
apparatus developed for infrared-guided laser photostimulation
(Fig. 1, A and B), we were able to visually direct a 1-␮m spot
of UV light onto the surface of the neuron of interest. Because
the uncaging in the z-axis depends on the precise focus onto the
center of the dendrite, the focal points of the IR image of the
neuron and of the UV-light flashes were well aligned (Fig. 1).
The measured spatial variations in the glutamate responsivity
were therefore not due to a variability in the z-axis uncaging of
glutamate. UV-light flashes (duration: 3–5 ms) delivered to the
soma and various sites along the apical dendrite (Fig. 1B)
elicited fast depolarizing deflections of the membrane potential. We first determined the resolution of the active region by
recording responses evoked as the UV-light spot was moved
incrementally away from the dendrite. In this series of experiments, photolytically evoked responses decayed with increasing distance. A Gaussian curve was fitted to the data (R ⫽
0.998, n ⫽ 6), and the accuracy was determined to be within
4.8 ␮m laterally (half-maximal peak amplitude, Fig. 1C) and
18 ␮m in the z-axis. The value for the depth resolution has
been established previously (Dodt et al. 1999). Based on these
results, we tested the glutamate responsivity for each neuron by
applying the UV-light flashes to 30 –120 different sites (mean:
76 sites) along the apical dendrite at ⬃5-␮m steps. In one case,
a region on the apical dendrite ⬃640 ␮m from the soma was
evaluated. The evoked responses (Fig. 1D) were strongly attenuated when 1 ␮M 2,3-dihydoxy-6-nitro-7-sulfamoylbenzo(f)quinoxaline (NBQX) was added to the perfusion medium
(response amplitude 33 ⫾ 4% of control, mean ⫾ SE, n ⫽ 11),
and the residual response was almost completely blocked by
adding 100 ␮M D-2-amino-5-phosphonovaleric acid (D-APV;
7 ⫾ 1%, n ⫽ 3). These findings demonstrate that the responses
evoked by glutamate photolysis resulted from a co-activation
of AMPA and NMDA receptors, and that dendritic kainate
receptors play only a minor role in these neurons.
FIG. 2. Glutamate receptor mapping. The peak amplitude and decay values of
the responses are plotted as a function of the stimulation distance from the soma.
A: glutamate responsivity profile for 7 neurons recorded in solution containing
TTX. All points are response amplitudes that were normalized to the amplitudes
evoked by stimulation at the soma (100%). Ca2⫹ spikes were often evoked at the
hot spots. B: summary data of control (●, n ⫽ 14), and in solution containing 1 ␮M
TTX ⫹ 200 –500 ␮M Cd2⫹ (䊐, n ⫽ 16). Dendritic response amplitudes are
normalized to the amplitude evoked by UV flashes delivered to the soma (mean ⫾
SE). C: the decay kinetics of the glutamate responses became faster as the site of
dendritic stimulation was moved farther from the soma (R2 ⫽ 0.79, linear regression fit, n ⫽ 14). All points are mean ⫾ SE.
decreased monotonically as the site of stimulation was moved
along the first 50 ␮m of the apical dendrite away from the soma
(Fig. 2B), which is probably due to a small density of spines in
86 • SEPTEMBER 2001 •
www.jn.org
GLUTAMATE HOT SPOTS ON NEOCORTICAL DENDRITES
1415
this region (Kunz et al. 1972). With distances ⬎50 ␮m from
the soma, however, the membrane of the apical dendrite became progressively more sensitive with distance, eventually
exceeding the somatic sensitivity (i.e., the same intensity and
duration of the UV flashes evoked bigger responses compared
with a somatic stimulation). The increase in responsivity with
distance is most easily explained by an increase in spine
density that occurs beyond the proximal 50 ␮m of the dendrite
(Kunz et al. 1972). Up to four restricted areas of high sensitivity (hot spots) were found at dendritic regions 80 – 600 ␮m
from the soma (Figs. 2, A and B, 3, and 5B). An electrophysiological hot spot was defined as having a response to glutamate, under conditions where voltage-dependent Na⫹ and
Ca2⫹ conductances are blocked, that was twofold or more
greater than the response recorded by stimulation at an adjacent
site, 5 ␮m away. The large responses evoked at hot spots often
triggered TTX-sensitive Na⫹ spikes (not shown) and Cd2⫹sensitive Ca2⫹ spikes (see Enhancement of dendritic calcium
spikes at hot spots). In 34 of 36 neurons, at least one hot spot
was located on the apical dendrite at a distance between 80 and
250 ␮m from the soma. The amplitude and integral of glutamate-evoked responses elicited beyond 250 ␮m from the soma
decreased with increasing distance from the soma (Fig. 2B).
To verify that the photostimulation method was reliable and
that the glutamate responsivity profile of a single neuron could
be measured in a reproducible manner, we repeated the stimulation protocol along the same apical dendrite. In all cases
(n ⫽ 6), the overall profile of glutamate responsivity was the
same, and most notably, the location of the hot spots was
J Neurophysiol • VOL
identical, suggesting that the hot spots were not due to some
experimentally introduced artifact such that might occur with
slice drift or temporal changes in dendritic excitability.
The decay of the responses was slowest at the most proximal
test sites and became progressively faster as the stimulation site
was moved to the more distal test sites (R2 ⫽ 0.79, linear
regression fit, n ⫽ 14; Fig. 2C). For responses induced at 500
␮m, the decay time constant (␶) was ⬃50% of responses
evoked at the soma (n ⫽ 8; Fig. 2C). We also found a decrease
in the half-maximal width of the responses from proximal to
more distal dendritic stimulation sites, but no significant
change in the 20 – 80% rise time values with distance from the
soma (not shown).
Responsivity profile does not reflect activation of voltageactivated calcium and sodium conductances
To test whether voltage-activated Na⫹ and Ca2⫹ conductances might be responsible for the spatial variations in responsivity, we bath applied both TTX (1 ␮M) and various concentrations of CdCl2 (200 –500 ␮M, n ⫽ 16 cells). Blocking these
conductances accelerated the time course and decreased the
peak amplitude and integral of the responses evoked at the
soma and along the dendrite. For subthreshold responses in the
presence of TTX and Cd2⫹, peak amplitude, integral, 20 – 80%
rise time, half-width, and the decay time constant were decreased by 23 ⫾ 2%, 47 ⫾ 3%, 40 ⫾ 2%, 44 ⫾ 2%, and 35 ⫾
2%, respectively (relative to control; mean ⫾ SE, n ⫽ 5 cells).
These data show that TTX- and Cd2⫹-sensitive conductances
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
FIG. 3. Dendritic hot spots of glutamate sensitivity
in a neocortical pyramidal neuron of layer V. Fluorescent image (A) taken from the neuron following intracellular filling with Neurobiotin and staining with
cy-3. Scale bar, 20 ␮m. B: surface plot of responses to
focal glutamate photolysis at the soma and along the
1st 400 ␮m of the apical dendrite, showing several
dendritic hot spots in this neuron. Note that Ca2⫹
spikes were evoked at the hot spots. Recordings were
obtained in solution containing 1 ␮M TTX.
1416
A. FRICK, W. ZIEGLGÄNSBERGER, AND H.-U. DODT
soma and the adjacent sites and for the half-maximal width
values between the three stimulation sites. In addition, the
decay time constants of responses at hot spots and at adjacent
sites were not significantly different.
To explore whether the activation of both AMPA and
NMDA receptors contribute to the hot spots of glutamate
responsivity, we selectively blocked AMPA receptors by applying 1 ␮M NBQX. NBQX reduced the peak amplitude of
responses evoked at hot spots by 72%, and the integral by 48%
(n ⫽ 8). NBQX had a similar effect on responses evoked at
dendritic sites adjacent to hot spots, strongly suggesting that
the ratio of AMPA and NMDA receptors does not differ
between hot spots and neighboring dendritic regions.
Hot spots and dendritic branch points do not overlap
significantly increased the charge that reached the soma and
axon. However, regardless of the degree of blockade of these
conductances in the presence of TTX and Cd2⫹, the overall
profile of glutamate responsivity (including the hot spots)
persisted (Figs. 2B and 5B), indicating that it was not caused by
activation of either Na⫹ or Ca2⫹ conductances. These results
suggest strongly that the glutamate responsivity profile reflects
a nonuniform distribution of glutamate receptors along the
apical dendrite of neocortical pyramidal neurons.
Both AMPA and NMDA components are elevated
at hot spots
Under the conditions where Na⫹ and Ca2⫹ conductances
were blocked (Figs. 2B, 4, and 5B), the most pronounced hot
spots were found at a distance of 170 ⫾ 61 ␮m (mean ⫾ SD;
range: 88 –357 ␮m) from the soma with an extent of 22 ⫾ 10
␮m (range: 9 – 48 ␮m; n ⫽ 16; Fig. 4A). Figure 4C compares
responses evoked at these hot spots, a dendritic region adjacent
to them, and the soma (examples are given in Fig. 4B). Peak
amplitudes and integrals of responses evoked at hot spots were
on average three times larger as compared with the soma and
adjacent sites (t-test, P ⬍ 0.001). Responses elicited at the
soma had 20 – 80% rise times of 8 ⫾ 1 ms (mean ⫾ SE) and
decay time constants (␶) of 72 ⫾ 5 ms. For responses induced
at hot spots, 20 – 80% rise times were longer (11 ⫾ 2 ms) and
decay times faster (57 ⫾ 4; P ⬍ 0.01). No significant differences were observed for the 20 – 80% rise times between the
J Neurophysiol • VOL
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
FIG. 4. Properties of responses evoked at hot spots of neurons recorded in
solution containing 1 ␮M TTX ⫹ 200 –500 ␮M CdCl2 (n ⫽ 16). A: the most
prominent hot spots were measured 170 ⫾ 61 ␮m from the soma with an extent
of 22 ⫾ 10 ␮m (mean ⫾ SD). B: example of the significantly larger response
evoked at a hot spot compared with the soma or an adjacent region in one neuron.
C: comparison of the peak amplitude, integral, 20 – 80% rise time, half-width
(duration at 1⁄2 of peak amplitude), and decay values (␶) for responses evoked at hot
spots (■), at regions close to the hot spots (s), and at the somata (䊐) in 16 neurons.
P ⬍ 0.01 for hot spot/adjacent site (⫹), and for hot spot/soma (*).
Previous reports showing intradendritic recordings from cerebellar Purkinje cells suggested that multiple spike generation
zones (“hot spots”) exist in the dendritic tree, probably at or
near dendritic bifurcations (Llinas and Hess 1976; Llinas and
Sugimori 1980). To examine the question of whether hot spots
of glutamate responsivity are located at branch points, and
therefore could be the basis for spike generation zones at these
sites, we injected Neurobiotin intracellulary to label the dendritic tree and determined the position of the branch points
using fluorescence microscopy (n ⫽ 15; Fig. 5). The experiments were carried out in control solution (Fig. 5B, neurons
12–15) or in the presence of TTX (1 ␮M) and Cd2⫹ (200 –500
␮M; Fig. 5B, neurons 1–11). The glutamate responsivity profile along the somato-dendritic axis of the neurons is shown by
color coding the response amplitudes (Fig. 5B). The dendritic
branch points (small horizontal bars) are superimposed on this
activity profile to compare the position of the hot spots and
dendritic branch points. Although in several cases the branch
points occurred within the range or in close proximity of the
hot spots, the majority of the branch points did not overlap with
the hot spots. In addition, in cases where branch points could
be detected with infrared videomicroscopy, photolytic stimulation at these sites did not result in bigger responses as
compared with adjacent sites. Plotting the number of hot spots
and branch points as a function of the distance from the soma
revealed that most hot spots and branch points were found
between 50 and 250 ␮m from the soma (Fig. 5C). In contrast
to branch points, which were also located along the first 50 ␮m
of the dendrite and beyond 250 ␮m from the soma, hot spots
were not found in these regions, providing further evidence
against an association of hot spots and dendritic branch points.
To demonstrate this statistically, we divided our data into two
groups (branch points and hot spots) and performed unpaired
two-tailed t-tests to determine the relationship between hot
spots and branch points within each group. When we considered the group of hot spots separately (n ⫽ 24), we found that
there was no bias for their colocalization either with or without
branch points. However, when we considered the group of
branch points, there was a strong significant difference between the number of branch points associated with a hot spot
and the number that was not associated with a hot spot (P ⬍
0.001, n ⫽ 147). This analysis confirmed that there was no
particular bias for hot spots to occur at branch points.
GLUTAMATE HOT SPOTS ON NEOCORTICAL DENDRITES
1417
Enhancement of dendritic calcium spikes at hot spots
The dendrites of neocortical pyramidal neurons are known to
contain a rich collection of active Ca2⫹ conductances (Deisz et
al. 1991; Kim and Connors 1993; Markram and Sakmann
1994; Markram et al. 1995; Schiller et al. 1995; Yuste et al.
1994) that can support Ca2⫹ action potentials (Amitai et al.
1993; Kim and Connors 1993; Schiller et al. 1997; Schwindt
and Crill 1997; Seamans et al. 1997). In particular, the observation that Ca2⫹ spikes can arise in the dendrites of several
neural cell types has changed our conception of how a neuron
processes synaptic input (for reviews see Johnston et al. 1996;
Yuste and Tank 1996). To determine whether Ca2⫹ spikes can
be initiated anywhere along the dendrite, or whether glutamate
receptor hot spots might provide a trigger site for the initiation
of regenerative Ca2⫹ responses, we recorded from neurons in
the presence of TTX (1 ␮M). We found that Ca2⫹ spikes were
triggered most easily following stimulation of hot spots as
compared with other sites on the neuron (Figs. 2A and 3),
consistent with the hypothesis that hot spots provide a faciliJ Neurophysiol • VOL
tatory role for generation of Ca2⫹ spikes. To photolytically
evoke Ca2⫹ spikes at the soma and multiple dendritic sites, we
adjusted the duration or intensity of the light flashes (n ⫽ 17;
Fig. 6). At distances ⬎100 ␮m from the soma, the amplitude
and threshold of the Ca2⫹ spikes, measured at the soma,
decreased as the UV-light spot was moved farther from the
soma (see Fig. 6A for determination of amplitude and threshold). At ⬃450 ␮m, the amplitude of the Ca2⫹ spikes had
declined from 53 mV by stimulation at the soma to 30 mV
(P ⬍ 0.01; Fig. 6B), and the threshold from 23 to 13 mV (P ⬍
0.01; Fig. 6C). In the presence of 1 ␮M TTX and 10 mM TEA
to block both Na⫹ and K⫹ channels, the amplitudes of the
photolytically evoked Ca2⫹ spikes were large (n ⫽ 9; Fig. 6).
Although the amplitude of the dendritic Ca2⫹ spikes was large
in TEA (on average 83 mV; Fig. 6B), their apparent threshold
was still quite low (55% at 450 ␮m compared with soma, P ⬍
0.01, n ⫽ 13; Fig. 6C). The similarity between amplitude, but
not threshold, of the photolytically evoked Ca2⫹ spikes in
TTX/TEA suggests that the spikes were initiated locally in the
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
FIG. 5. Localization of physiological hot spots with respect
to branch points of the apical dendrites (n ⫽ 15). Fluorescent
image (A) taken from 2 neurons following intracellular filling
with Neurobiotin and staining with cy-3. Scale bar, 20 ␮m. The
scaling does not relate to B. B: color-coded plots of response
amplitudes to focal glutamate photolysis at the soma and along
the 1st 500 ␮m of the apical dendrite, showing several dendritic
hot spots in these neurons. Recordings were obtained in control
solution (neurons 12–15) or in the presence of TTX (1 ␮M) and
Cd2⫹ (100 –500 ␮M; neurons 1–11). Hot spots are colored in
red. The positions of the dendritic branch points was determined
using fluorescence microscopy of the Neurobiotin-cy3–stained
neurons and superimposed on the activity pattern as small horizontal bars. C: statistical distribution of hot spots and dendritic
branch points along the dendrite. The normalized numbers of
branch points and of hot spots are plotted against the distance
from the soma. As not all dendrites could be followed in the
infrared image for stimulation up to a distance of 500 ␮m from
the soma, the numbers were normalized, by dividing the absolute
values by the number of neurons for this dendritic segment
(given in parentheses).
1418
A. FRICK, W. ZIEGLGÄNSBERGER, AND H.-U. DODT
2⫹
FIG. 6. Ca
spikes evoked by somatic and dendritic depolarization. A:
examples of Ca2⫹ spikes evoked at the soma and at a dendritic site in TTX
(black line) and in TTX/TEA (gray line) in one neuron. Arrows and triangles
indicate apparent threshold and peak amplitude of the responses, respectively.
Plot of Ca2⫹ spike amplitude (B), measured from somatic membrane potential
to peak, and apparent threshold (C), evoked by somatic and dendritic glutamate
photostimulation at various distances from the soma. Ca2⫹ spikes were measured in solution containing 1 ␮M TTX (●, n ⫽ 17), and in solution containing
1 ␮M TTX ⫹ 10 mM TEA (‚, n ⫽ 9).
dendrite and, because dendritic and somatic K⫹ currents were
reduced by TEA, propagated actively to the soma.
DISCUSSION
Glutamate receptors are non-uniformly distributed along
apical dendrites of neocortical pyramidal neurons
We have used local photolysis of caged glutamate to map
regional differences in the distribution of glutamate receptors
across a wide range of the main apical dendrites in layer V
pyramidal neurons in neocortical slices. The main findings are
that the responsivity to glutamate is non-uniformly distributed
along the apical dendrite, and that the glutamate receptors form
up to four hot spots of glutamate responsivity (extent 22 ␮m)
within the region of the dendrite under investigation (up to 640
␮m). The evidence showing this non-uniform distribution pattern of glutamate receptors is twofold: first, photolytically
induced glutamate responses vary as a function of distance
along the dendrite, suggesting that the density of glutamate
receptors differs (see Figs. 2, A and B, 3B, and 5B); second,
blocking voltage-activated sodium and calcium conductances
J Neurophysiol • VOL
Densities of AMPA and NMDA receptors are elevated
at hot spots
We found that both AMPA and NMDA receptors were
concentrated in higher numbers at hot spots, and that stimulation at these sites under conditions where Na⫹ and Ca2⫹
conductances were blocked, resulted in a threefold greater
response compared with adjacent dendritic sites (see Fig. 4).
The kinetics of the responses evoked at hot spots and adjacent
sites were not statistically different and blockade of the AMPA
receptors with NBQX reduced the responses for both sites to
the same degree. Electrophysiological and anatomical studies
of cultured rat visual cortex (Jones and Baughman 1991),
hippocampal neurons (Bekkers and Stevens 1989; Benke et al.
1993; Craig et al. 1994), and spinal cord (Vogt et al. 1995), as
well as of neocortical neurons in brain slices (Aghajanian and
Marek 1997) suggest that subtypes of glutamate receptors are
colocalized (Bekkers and Stevens 1989; Jones and Baughman
1991) and clustered (Aghajanian and Marek 1997; Benke et al.
1993; Craig et al. 1994; Jones and Baughman 1991; Vogt et al.
1995) at synapses. On the basis of these and other studies,
extrasynaptic receptors may make an insignificant contribution
to the profile of glutamate responsivity reported here.
Hot spots and dendritic branch points are not spatially
correlated
Previous studies of Purkinje cells (Llinas and Hess 1976;
Llinas and Sugimori 1980) suggested that many dendritic sites
were capable of initiating Ca2⫹ spikes (hot spots), and that
these sites were located probably at or near dendritic bifurcations. Our study of pyramidal neurons in the neocortex shows
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
does not change the responsivity profile obviating the possibility that the differences in response magnitude are due to a
non-uniform distribution of voltage-activated conductances
(see Fig. 2B). These results are consistent with the results of
other studies showing that the distribution of AMPA receptors
are non-uniform with respect to distance along apical dendrites
of neocortical neurons (Dodt et al. 1998) and hippocampal
pyramidal neurons (Magee and Cook 2000; Pettit and Augustine 2000). With the exception of hot spots, the amplitude and
integral of the glutamate-evoked responses elicited beyond 250
␮m from the soma decreased with increasing distance. This
decline could be caused by a decreased membrane area exposed to glutamate, electrotonic attenuation (Rall 1977; Spruston et al. 1993), or higher densities of transient A-type K⫹
currents (Hoffman et al. 1997) or hyperpolarization-activated
currents (Ih) (Magee 1998; Stuart and Spruston 1998) in the
distal apical dendrites.
The decay of the responses was slowest at the soma and
became progressively faster as the stimulation site was moved
to the more distal dendrite (see Fig. 2C). Although not investigated here, this might result from an elevated dendritic density of A-type K⫹ channels (Hoffman et al. 1997) or of Ih
channels (Magee 1998; Stuart and Spruston 1998), a stronger
contribution of AMPA compared with NMDA receptors
(Cauller and Connors 1994; Dodt et al. 1998), a lower membrane resistivity in the distal dendrites (Stuart and Spruston
1998), or some combination thereof.
GLUTAMATE HOT SPOTS ON NEOCORTICAL DENDRITES
that there is no statistical relationship between hot spots of
glutamate responsivity and morphological identified branch
points on the apical dendrites (see also Fig. 5). One might
speculate that the hot spots of glutamate responsivity described
in our study are strategically located, probably in a layer- and
input-specific manner, but not necessarily at branch points. It
remains to be shown whether the density of Ca2⫹ channels is
elevated at bifurcations, and whether spike initiation zones are
located at these sites.
Hot spots of glutamate responsivity are preferential trigger
sites for calcium spikes
Mechanism underlying the hot spots of glutamate
responsivity
What is the mechanism of the hot spots of glutamate
responsivity? There are several possibilities: dendritic
branch points, increase in agonist-affinity or single-channel
conductance, elevated glutamate receptor number or density
per synapse, or increases in the density of spines. Branch
points seem unlikely to be responsible. If this was the case,
we should see an association of the dendritic branch points
with the physiological hot spots, but our statistical analysis
has indicated that this does not occur. Increases in affinity or
single-channel conductance also seem unlikely to account
for hot spots. Dendritic recordings in hippocampal pyramidal neurons and in cerebellar Purkinje cells have provided
evidence that the functional properties of dendritic glutamate receptors, at least along the first part of the dendrite,
are similar (Häusser and Roth 1997; Spruston et al. 1995).
Therefore it seems most likely that the hot spots (⬃22 ␮m
in extent) described in our study arise from either higher
densities of synapses or more receptors per synapse in these
regions. It has been shown that Schaffer collateral synapses
have a wide range of surface areas, and that the number of
glutamate receptors at these synapses is directly related to
the area (Nusser et al. 1998; Takumi et al. 1999). Thus it is
J Neurophysiol • VOL
possible that glutamate uncaging at larger synapses would
open more glutamate receptors, and a cluster of those synapses with a larger area could explain an elevated dendritic
glutamate sensitivity of a membrane area the size of a hot
spot (extent ⬃22 ␮m). Unitary excitatory postsynaptic potentials evoked between layer V pyramidal neurons have
been found to be variable in size (e.g., Markram et al. 1997;
Thomson et al. 1993), but both pre- and postsynaptic mechanisms could account for this observation. It is difficult to
compare these data with our own study, because the spatial
resolution of the laser photostimulation system is unlikely to
activate single but rather several postsynaptic sites, and
presysnaptic mechanisms only play a minor or no role.
Either of the two mechanisms (cluster of particularly strong
synapses or increase in density of synapses) should correlate
with an elevated dendritic glutamate sensitivity of a hot
spot. Ninety-four percent of the cells investigated had at
least one hot spot in layer IV, and hot spots may be a
correlate of functional synaptic connections; it remains to be
established which connections form these hot spots.
In conclusion, hot spots may provide a mechanism for
weighting dendritic inputs and/or triggering local Na⫹- and
Ca2⫹-dependent action potentials, either of which could
increase the effectiveness of synapses by compensating for
electrotonic attenuation. Infrared-guided laser photostimulation may also be used to clarify how changes in glutamate
receptor density and active properties of the dendrites interact to allow location-independent transmission of synaptic responses to the soma (Cook and Johnston 1999; Magee
and Cook 2000), or linear summation of synaptic responses
activated at different dendritic locations (Cash and Yuste
1999). Other relevant issues related to the distribution and
clustering of glutamate receptors concern synaptic plasticity
or synaptic connectivity; for example, what are the implications of hot spots for the events that underlie synaptic
remodeling during associative synaptic modification (Carroll et al. 1999), and how might changes in receptor density
correlate to specific inputs from the thalamus or cortical
areas?
We thank D. Johnston, N. Spruston, and M. Yeckel for discussions and
comments on the manuscript.
This work was supported by a grant from Sonderforschungsbereich (SFB)
391 to H.-U. Dodt and W. Zieglgänsberger.
REFERENCES
AGHAJANIAN GK AND MAREK GJ. Serotonin induces excitatory postsynaptic
potentials in apical dendrites of neocortical pyramidal cells. Neuropharmacology 36: 589 –599, 1997.
AMITAI Y, FRIEDMAN A, CONNORS BW, AND GUTNICK MJ. Regenerative
activity in apical dendrites of pyramidal cells in neocortex. Cereb Cortex 3:
26 –38, 1993.
BEKKERS JM AND STEVENS CF. NMDA and non-NMDA receptors are colocalized at individual excitatory synapses in cultured rat hippocampus.
Nature 341: 230 –233, 1989.
BENKE TA, JONES OT, COLLINGRIDGE GL, AND ANGELIDES KJ. N-methyl-Daspartate receptors are clustered and immobilized on dendrites of living
cortical neurons. Proc Natl Acad Sci USA 90: 7819 –7823, 1993.
CALLAWAY EM AND KATZ LC. Photostimulation using caged glutamate reveals
functional circuitry in living brain slices. Proc Natl Acad Sci USA 90:
7661–7665, 1993.
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
We demonstrated that when Na⫹ channels were blocked
(with the addition of TTX), Ca2⫹ spikes could be triggered in
the dendrite by photolytically released glutamate (see Figs. 2A,
3B, and 6). Ca2⫹ spikes could be evoked at multiple sites along
the dendrite, although most easily at hot spots, and the amplitude and apparent threshold of the spikes decreased with increasing distance of the stimulation site from the soma. These
findings support the hypothesis that in neocortical pyramidal
neurons many dendritic sites are capable of initiating Ca2⫹
spikes (Schwindt and Crill 1997). In the presence of TTX and
TEA, the amplitudes of the photolytically evoked Ca2⫹ spikes
were large, but their apparent thresholds were 45% lower
compared with the Ca2⫹ spikes evoked by photolysis at the
soma. Similar findings were obtained by Schwindt and Crill
(1997), who applied glutamate iontophoretically at sites of the
apical dendrite to evoke Ca2⫹ spikes. In agreement with several previous reports (Reuveni et al. 1993; Schwindt and Crill
1997; Stafstrom et al. 1985; Yuste et al. 1994), we conclude
that dendritic and somatic K⫹ channels normally prevent the
active propagation of large-amplitude Ca2⫹ spikes from the
site of initiation to the soma.
1419
1420
A. FRICK, W. ZIEGLGÄNSBERGER, AND H.-U. DODT
J Neurophysiol • VOL
MAGEE J, HOFFMAN D, COLBERT C, AND JOHNSTON D. Electrical and calcium
signaling in dendrites of hippocampal pyramidal neurons. Annu Rev Physiol
60: 327–346, 1998.
MAGEE JC. Dendritic hyperpolarization-activated currents modify the integrative properties of hippocampal CA1 pyramidal neurons. J Neurosci 18:
7613–7624, 1998.
MAGEE JC AND COOK EP. Somatic EPSP amplitude is independent of synapse
location in hippocampal pyramidal neurons [see comments]. Nature Neurosci 3: 895–903, 2000.
MARKRAM H, HELM PJ, AND SAKMANN B. Dendritic calcium transients evoked
by single back-propagating action potentials in rat neocortical pyramidal
neurons. J Physiol (Lond) 485: 1–20, 1995.
MARKRAM H, LUEBKE J, FROTSCHER M, ROTH A, AND SAKMANN B. Physiology and anatomy of synaptic connections between thick tufted pyramidal
neurones in the developing rat neocortex. J Physiol (Lond) 500: 409 – 440,
1997.
MARKRAM H AND SAKMANN B. Calcium transients in dendrites of neocortical
neurons evoked by single subthreshold excitatory postsynaptic potentials via
low-voltage-activated calcium channels. Proc Natl Acad Sci USA 91: 5207–
5211, 1994.
NUSSER Z, LUJAN R, LAUBE G, ROBERTS JDB, MOLNAR E, AND SOMOGYI P. Cell
type and pathway dependence of synaptic AMPA receptor number and
variability in the hippocampus. Neuron 21: 545–559, 1998.
O’BRIEN RJ, LAU LF, AND HUGANIR RL. Molecular mechanisms of glutamate
receptor clustering at excitatory synapses. Curr Opin Neurobiol 8: 364 –369,
1998.
PETERS A. Number of neurons and synapses in primary visual cortex. In:
Cerebral Cortex: Further Aspects of Cortical Function, Including Hippocampus, edited by Jones EG and Peters A. New York: Plenum, 1987, p.
267–294.
PETRALIA RS AND WENTHOLD RJ. Light and electron immunocytochemical
localization of AMPA-selective glutamate receptors in the rat brain. J Comp
Neurol 318: 329 –354, 1992.
PETTIT DL AND AUGUSTINE GJ. Distribution of functional glutamate and
GABA receptors on hippocampal pyramidal cells and interneurons. J Neurophysiol 84: 28 –38, 2000.
RALL W. Core conductor theory and cable properties of neurons. In: Handbook
of Physiology. The Nervous System. Cellular Biology of Neurons. Bethesda,
MD: Am. Physiol. Soc., 1977, sect. 1, vol. I, p. 39 –97.
RAO A AND CRAIG AM. Activity regulates the synaptic localization of the
NMDA receptor in hippocampal neurons. Neuron 19: 801– 812, 1997.
REUVENI I, FRIEDMAN A, AMITAI Y, AND GUTNICK MJ. Stepwise repolarization
from Ca2⫹ plateaus in neocortical pyramidal cells: evidence for nonhomogeneous distribution of HVA Ca2⫹ channels in dendrites. J Neurosci 13:
4609 – 4621, 1993.
ROSENMUND C, FELTZ A, AND WESTBROOK GL. Synaptic NMDA receptor
channels have a low open probability. J Neurosci 15: 2788 –2795, 1995.
SCHILLER J, HELMCHEN F, AND SAKMANN B. Spatial profile of dendritic calcium
transients evoked by action potentials in rat neocortical pyramidal neurones.
J Physiol (Lond) 487: 583– 600, 1995.
SCHILLER J, SCHILLER Y, STUART G, AND SAKMANN B. Calcium action potentials restricted to distal apical dendrites of rat neocortical pyramidal neurons.
J Physiol (Lond) 505: 605– 616, 1997.
SCHWINDT PC AND CRILL WE. Local and propagated dendritic action potentials
evoked by glutamate iontophoresis on rat neocortical pyramidal neurons.
J Neurophysiol 77: 2466 –2483, 1997.
SEAMANS JK, GORELOVA NA, AND YANG CR. Contributions of voltage-gated
Ca2⫹ channels in the proximal versus distal dendrites to synaptic integration
in prefrontal cortical neurons. J Neurosci 17: 5936 –5948, 1997.
SPRUSTON N, JAFFE DB, WILLIAMS SH, AND JOHNSTON D. Voltage- and spaceclamp errors associated with the measurement of electrotonically remote
synaptic events. J Neurophysiol 70: 781– 802, 1993.
SPRUSTON N, JONAS P, AND SAKMANN B. Dendritic glutamate receptor channels
in rat hippocampal CA3 and CA1 pyramidal neurons. J Physiol (Lond) 482:
325–352, 1995.
STAFSTROM CE, SCHWINDT PC, CHUBB MC, AND CRILL WE. Properties of
persistent sodium conductance and calcium conductance of layer V
neurons from cat sensorimotor cortex in vitro. J Neurophysiol 53:
153–170, 1985.
STUART G AND SPRUSTON N. Determinants of voltage attenuation in neocortical
pyramidal neuron dendrites. J Neurosci 18: 3501–3510, 1998.
86 • SEPTEMBER 2001 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
CARROLL RC, LISSIN DV, VON ZASTROW M, NICOLL RA, AND MALENKA RC.
Rapid redistribution of glutamate receptors contributes to long-term depression in hippocampal cultures. Nature Neurosci 2: 454 – 460, 1999.
CASH S AND YUSTE R. Linear summation of excitatory inputs by CA1 pyramidal neurons. Neuron 22: 383–394, 1999.
CAULLER LJ AND CONNORS BW. Synaptic physiology of horizontal afferents to
layer I in slices of rat SI neocortex. J Neurosci 14: 751–762, 1994.
CLARK BA, FARRANT M, AND CULL-CANDY SG. A direct comparison of the
single-channel properties of synaptic and extrasynaptic NMDA receptors.
J Neurosci 17: 107–116, 1997.
COOK EP AND JOHNSTON D. Voltage-dependent properties of dendrites that
eliminate location-dependent variability of synaptic input. J Neurophysiol
81: 535–543, 1999.
CRAIG AM, BLACKSTONE CD, HUGANIR RL, AND BANKER G. Selective clustering of glutamate and gamma-aminobutyric acid receptors opposite terminals releasing the corresponding neurotransmitters. Proc Natl Acad Sci USA
91: 12373–12377, 1994.
CURRIE SN, WANG XF, AND DAW NW. NMDA receptors in layers II and III of
rat cerebral cortex. Brain Res 662: 103–108, 1994.
DEISZ RA, FORTIN G, AND ZIEGLGÄNSBERGER W. Voltage dependence of
excitatory postsynaptic potentials of rat neocortical neurons. J Neurophysiol
65: 371–382, 1991.
DENK W. Two-photon scanning photochemical microscopy: mapping ligandgated ion channel distributions. Proc Natl Acad Sci USA 91: 6629 – 6633,
1994.
DODT HU, FRICK A, KAMPE K, AND ZIEGLGÄNSBERGER W. NMDA and AMPA
receptors on neocortical neurons are differentially distributed. Eur J Neurosci 10: 3351–3357, 1998.
DODT H, EDER M, FRICK A, AND ZIEGLGÄNSBERGER W. Precisely localized
LTD in the neocortex revealed by infrared-guided laser stimulation [published erratum appears in Science 1999 Dec 17;286(5448): 2273]. Science
286: 110 –113, 1999.
FRICK A, ZIEGLGÄNSBERGER W, AND DODT HU. Glutamate sensitivity of
neocortical pyramidal neurons investigated with infrared-guided laser stimulation. Soc Neurosci Abstr 24: 325, 1998.
HÄUSSER M AND ROTH A. Dendritic and somatic glutamate receptor channels
in rat cerebellar Purkinje cells. J Physiol (Lond) 501: 77–95, 1997.
HOFFMAN DA, MAGEE JC, COLBERT CM, AND JOHNSTON D. K⫹ channel
regulation of signal propagation in dendrites of hippocampal pyramidal
neurons [see comments]. Nature 387: 869 – 875, 1997.
HUNTLEY GW, VICKERS JC, JANSSEN W, BROSE N, HEINEMANN SF, AND
MORRISON JH. Distribution and synaptic localization of immunocytochemically identified NMDA receptor subunit proteins in sensory-motor and
visual cortices of monkey and human. J Neurosci 14: 3603–3619, 1994a.
HUNTLEY GW, VICKERS JC, AND MORRISON JH. Cellular and synaptic localization of NMDA and non-NMDA receptor subunits in neocortex: organizational features related to cortical circuitry, function and disease. Trends
Neurosci 17: 536 –543, 1994b.
JOHNSTON D, MAGEE JC, COLBERT CM, AND CRISTIE BR. Active properties of
neuronal dendrites. Annu Rev Neurosci 19: 165–186, 1996.
JONES KA AND BAUGHMAN RW. NMDA- and non-NMDA-receptor components of excitatory synaptic potentials recorded from cells in layer V of rat
visual cortex. J Neurosci 8: 3522–3534, 1988.
JONES KA AND BAUGHMAN RW. Both NMDA and non-NMDA subtypes of
glutamate receptors are concentrated at synapses on cerebral cortical neurons in culture. Neuron 7: 593– 603, 1991.
KATZ LC AND DALVA MB. Scanning laser photostimulation: a new approach
for analyzing brain circuits. J Neurosci Methods 54: 205–218, 1994.
KIM HG AND CONNORS BW. Apical dendrites of the neocortex: correlation
between sodium- and calcium-dependent spiking and pyramidal cell morphology. J Neurosci 13: 5301–5311, 1993.
KORNAU HC, SCHENKER LT, KENNEDY MB, AND SEEBURG PH. Domain interaction between NMDA receptor subunits and the postsynaptic density
protein PSD-95. Science 269: 1737–1740, 1995.
KUNZ G, KIRSCHE W, WENZEL J, WINKELMANN E, AND NEUMANN H. Quantitative studies on the dendrite spines of pyramidal neurons in the rat sensory
cortex. Z Mikrosk Anat Forsch 85: 397– 416, 1972.
LLINAS R AND HESS R. Tetrodotoxin-resistant dendritic spikes in avian Purkinje cells. Proc Natl Acad Sci USA 73: 2520 –2523, 1976.
LLINAS R AND SUGIMORI M. Electrophysiological properties of in vitro Purkinje
cell dendrites in mammalian cerebellar slices. J Physiol (Lond) 305: 197–
213, 1980.
GLUTAMATE HOT SPOTS ON NEOCORTICAL DENDRITES
STUART G, SPRUSTON N, SAKMANN B, AND HÄUSSER M. Action potential
initiation and backpropagation in neurons of the mammalian CNS. Trends
Neurosci 20: 125–131, 1997.
TAKUMI Y, RAMIREZ-LEON V, LAAKE P, RINVIK E, AND OTTERSEN OP. Different modes of expression of AMPA and NMDA receptors in hippocampal
synapses. Nature Neurosci 2: 618 – 624, 1999.
THOMSON AM, DEUCHARS J, AND WEST DC. Large, deep layer pyramid-pyramid
single axon EPSPs in slices of rat motor cortex display paired pulse and
frequency-dependent depression, mediated presynaptically and self-facilitation, mediated postsynaptically. J Neurophysiol 70: 2354 –2369, 1993.
VOGT K, LUSCHER HR, AND STREIT J. Analysis of synaptic transmission at
1421
single identified boutons on rat spinal neurons in culture. Pflügers Arch 430:
1022–1028, 1995.
WIEBOLDT R, GEE KR, NIU L, RAMESH D, CARPENTER BK, AND HESS GP.
Photolabile precursors of glutamate: synthesis, photochemical properties,
and activation of glutamate receptors on a microsecond time scale. Proc Natl
Acad Sci USA 91: 8752– 8756, 1994.
YUSTE R, GUTNICK MJ, SAAR D, DELANEY KR, AND TANK DW. Ca2⫹ accumulations in dendrites of neocortical pyramidal neurons: an apical band and
evidence for two functional compartments. Neuron 13: 23– 43, 1994.
YUSTE R AND TANK DW. Dendritic integration in mammalian neurons, a
century after Cajal. Neuron 16: 701–716, 1996.
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on April 29, 2017
J Neurophysiol • VOL
86 • SEPTEMBER 2001 •
www.jn.org