| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INNOVATIVE METHODOLOGY
1Division of Biophysics, Center for Disease Biology and Integrative Medicine, Graduate School of Medicine, and Center for NanoBio Integration, The University of Tokyo, Tokyo; 2Precursory Research Organization for Embryonic Science and Technology, Japan Science and Technology Agency, Saitama, Japan; and 3Department of Pharmacology and Physiology, Drexel University College of Medicine, Philadelphia, Pennsylvania
Submitted 12 October 2007; accepted in final form 16 January 2008
 |
ABSTRACT |
|---|
|
600 x 600 x 100 µm) including neurons in layers 2/3 and 4 using this new technique. We mapped the density and amplitude of unitary excitatory postsynaptic currents and found that the basic microarchitecture of excitatory synaptic connections consists of two regions: a columnar, dense core region with a radius of 150 µm and an outer, sparse region. The dense core region includes the majority of strong synaptic connections in layer 2/3. Our results reveal the columnar organization of synaptic connectivity in the rat visual cortex, where functional columns have not been clearly demonstrated. Thus this technique will be a uniquely powerful tool for quantifying synaptic connectivity and manipulating neural activity at the single-cell level. |
|
INTRODUCTION |
|---|
|
; Mountcastle 1957). However, each neuron is not a homogeneous constituent of the cluster of neighboring neurons. In the cat visual cortex, each excitatory neuron connects to a small fraction of the neighboring neurons (Thomson et al. 2002) and neighboring neurons with the same orientation preference seldom have similar selectivities in other features (DeAngelis et al. 1999). In addition, since the border between columns with opposite direction preferences is extraordinarily sharp, neighboring neurons on each side of the border have opposite direction selectivity (Ohki et al. 2005). In the rat visual cortex, orientation preference often differs between adjacent neurons and the distance between cells has no correlation with their relative orientation or direction preferences (Girman et al. 1999; Ohki et al. 2005). Thus to reveal the synaptic connectivity patterns that generate the functional specificity of a neuron, it is necessary to quantify a wide distribution of synaptic connections and strengths within and across the columnar organization at the resolution of a single cell, rather than a group of cells.
Paired recording from synaptically coupled neurons has been used to identify the location and unitary postsynaptic response of the pair (Feldmeyer et al. 2002; Kalisman et al. 2005; Thomson et al. 2002). However, this method does not allow for the stimulation of many presynaptic neurons scattered widely in space. In contrast, anatomical approaches with trans-synaptic tracers are able to label many widely spaced presynaptic neurons (Wickersham et al. 2007). However, they are not able to reveal the synaptic strengths of the connections.
To spatially map both the synaptic connectivity and strength, laser-scanning photostimulation with caged glutamate has been used (Callaway and Katz 1993; Dantzker and Callaway 2000; Katz and Dalva 1994; Schubert et al. 2003; Shepherd et al. 2003; Shoham et al. 2005). These studies have used one-photon excitation of caged glutamate by UV light, which is heavily scattered within the tissue and has little axial resolution. Thus in using this method, it is difficult to map synaptic connections within the three-dimensional (3-D) tissue at the single-cell level. Similar limitations apply to one-photon excitation of channelrhodopsin-2 (ChR2), a light-gated cation channel, in a mapping study using transgenic mice expressing ChR2 (Arenkiel et al. 2007; Wang et al. 2007).
Here, we present a new photostimulation technique, which we call "two-photon macro photolysis of caged glutamate" (2pMAPG), that can be used to induce an action potential (AP) in only a small number of pyramidal neurons. 2pMAPG allows us to stimulate excitatory presynaptic neurons in a large 3-D space and measure unitary excitatory postsynaptic currents (EPSCs) in the patch-clamped postsynaptic neuron. We have used this technique to map the synaptic organization of neurons over a large area at the level of single cells in the rat visual cortex.
|
|
METHODS |
|---|
|
Coronal slices (350 µm thick) of the primary visual cortex were prepared from 15- to 19-day-old Sprague–Dawley rats. Slices were incubated at 32°C for 30 min and then stored in an incubation chamber at room temperature for 
1 h. Each slice was transferred to a recording chamber at room temperature (21–25°C). The extracellular solution contained (in mM): 125 NaCl, 2.5 KCl, 2 CaCl2, 1 MgCl2, 1.25 NaH2PO4, 26 NaHCO3, 20 glucose, and 1.5 4-carboxymethoxy-5,7-dinitroindolinyl-glutamate (CDNI-Glu; Ellis-Davies et al. 2007). The efficiency of two-photon excitation of CDNI-Glu is about fourfold higher than that of 4-methoxy-7-nitroindolinyl-glutamate (MNI-Glu), which has been used in almost all experiments with two-photon uncaging of glutamate (Ellis-Davies et al. 2007). The bathing solution also contained 50–100 µM D-2-amino-5-phosphonovaleric acid (D-APV; Sigma, St. Louis, MO) and 200 µM Trolox (Aldrich, Milwaukee, WI). To block activation of non-NMDA (N-methyl-D-aspartate) receptors and sodium channels, 10 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; Tocris, Bristol, UK) or 1 µM tetrodotoxin (TTX; Nacalai Tesque, Kyoto, Japan), respectively, were included. A total of 2–3 ml of solution was continuously oxygenated and recirculated. All experiments were approved by the animal experimental committee of the Faculty of Medicine, the University of Tokyo.
Electrophysiology
The patch-clamp electrodes (open-tip resistance: 4–8 M
) were filled with a solution containing 138 mM potassium gluconate, 4 mM MgCl2, 10 mM disodium phosphocreatine, 25–50 µM Alexa 594, 4 mM Na-ATP, 0.3 mM Na-GTP, and 10 mM HEPES-KOH (pH 7.2, 297 mOsm). In experiments to detect APs, the cells were recorded under whole cell current-clamp or extracellular loose-patch modes. The mean resting potential of the cells under the current-clamp recording was –69 ± 5 (SD) mV (n = 9). In experiments mapping synaptic inputs, recordings were obtained from layer 2/3 pyramidal cells located 50–90 µm deep in the slice (mean ± SD = 72 ± 14 µm, n = 6), 266 ± 24 (SD) µm from the pia, and 202 ± 38 (SD) µm from the layer 4 border. Series resistance was 25 ± 7 (SD) M. To detect EPSCs, the membrane potential was held at –65 mV. Since the Cl– reversal potential was measured as approximately –61 mV by the application of 
-aminobutyric acid (GABA), the driving force for GABAergic postsynaptic currents was negligible at the holding potential of –65 mV. The liquid junction potential was not corrected. Data were low-pass filtered at 2 kHz, sampled at 5–10 kHz, and recorded using FV1000-MPE software (Olympus, Tokyo, Japan).
Two-photon excitation imaging and uncaging of glutamate
Experiments were performed using an upright microscope (BX61WI; Olympus) and FV1000-MPE laser-scanning microscope system. To map a broad region, we used a water-immersion objective with a low-magnification configuration [XLUMPlanFI/IR x20, numerical aperture (NA) of 0.95]. For micro excitation, a high-magnification objective (LUMPlanFI/IR x60, NA of 0.9) was used, as in our previous experiments, to stimulate single synapses (Matsuzaki et al. 2001, 2004). Two mode-locked femtosecond-pulse Ti:sapphire lasers (MaiTai HP; Spectra Physics, Mountain View, CA) set at wavelengths of 720 and 830 nm were connected to the laser-scanning microscope via two independent scanheads. Both lasers were chirp-compensated before entering the scanheads. The diameter [(1/e2) – intensity width] of the 720-nm laser beam for photostimulation was adjusted to 4.1 mm at the back aperture of the objective by changing the distance between two convex lenses in the optical pathway before entering the scanhead using a motor-driven stage (SGSP20-85; Sigma-Koki, Tokyo, Japan). The point spread function of the focal volume at the 720-nm wavelength was estimated as full-widths of half-maximum (FWHMs) of 0.78 ± 0.06 (SE) µm laterally and 10 ± 0.2 (SE) µm axially (n = 12 and 8, respectively) using 0.1-µm fluorescent beads. The FWHM of the focal volume of the laser beam at 830 nm for imaging was estimated at 0.46 ± 0.03 (SE) µm laterally and 2.5 ± 0.04 (SE) µm axially (n = 8).
Images of neuronal morphology were acquired by X–Y scanning with the 830-nm laser at different depths, and these images were stacked along the Z-axis. The pixel length for imaging was 1.2 µm. When fluorescent imaging showed a cut in the main apical dendrite of the recorded cells, the data were discarded. In all figures with fluorescent images of neurons, the top of the image is closest to the pial surface. The locations of recorded cells and laminar borders were identified under transillumination of the laser scanning at the 830-nm wavelength.
For mapping EPSCs, a pseudorandom sequence of scanning of 32 x 32 pixels (spacing 19 µm) was constructed to maintain the distance between two successive pixels >135–270 µm. X–Y scanning was performed at three to five different depths, each of which was separated by 25–30 µm. Within each pixel, photolysis was performed consecutively at 3 x 3 points at a lateral interval of 3.6 µm with a pulse-train duration of 18 ms (2 ms at each point). It took about 2 min to complete mapping at a single plane. IPlab (BD Biosciences, Rockville, MD) and our own software programs, based on LabView (National Instruments, Austin, TX), were used for image processing.
Data analysis
The onset latency and rise time of the currents were defined, respectively, as the time from the onset of stimulation to 10% of the peak current and the time from 10% of the peak current to the current peak. For analysis of the onset latency and rise time, we used EPSCs with mean peak amplitudes ranging from 13.5 to 22.9 pA and direct currents with peak amplitudes ranging from 15.2 to 39.0 pA.
To construct a map of excitatory connected neurons, 
-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor-mediated currents (>8 pA) with peaks occurring during 15–45 ms after the onset of photostimulation were selected and their amplitudes were measured at each pixel. Pixels with large direct currents (>20 pA of the mean current amplitude at 10–18 ms after the onset of stimulation) were excluded from the analysis. Peak amplitudes of the subthreshold direct currents (<20 pA) were detected about 10–18 ms after the onset of stimulation and the decay was slow during the time window for detecting EPSCs (<1 pA/ms). As a result, at the pixels with subthreshold direct currents, it was possible to distinguish EPSCs from direct currents, although the amplitudes of the EPSCs were slightly underestimated. The sum of the amplitudes of the EPSCs at each pixel was displayed using pseudocolor coding. The EPSC density and EPSC amplitude for each distance were calculated from each of the EPSCs before summation. A Z-stacked map for each cell was reconstructed by averaging three to five maps at different depths along the Z-axis. White pixels at each depth were excluded from the averaging procedure. The positions of white pixels at all depths were represented as white pixels in the Z-stacked map. The average of the Z-stacked maps of six cells was constructed by centering the Z-stacked maps with respect to the location of the recorded cell somata and averaging the values at each pixel. White pixels in each Z-stacked map were excluded from the averaging procedure.
Data are presented as means ± SE, unless stated otherwise. Error bars on graphs correspond to the SE.
|
|
RESULTS |
|---|
|
Using a conventional two-photon microscope, we overfilled the back aperture of the objective (x60, 0.90 NA) with the laser beam to confine the focal volume of two-photon excitation to the diffraction limit of the objective (Matsuzaki et al. 2001, 2004). We now refer to this mode of illumination as two-photon micro excitation. Such diffraction-limited two-photon photolysis of caged glutamate activates glutamate receptors in a only minute focal volume and thus two-photon micro excitation does not support the generation of AP, since AP generation requires the activation of a large number of glutamate receptors distributed over the 3-D structure of a neuron. Thus to generate APs, we made the volume of two-photon excitation larger than that of the two-photon micro excitation. To increase the focal volume axially and laterally, we reduced the effective NA of the objective by underfilling its back aperture. We adjusted the diameter [(1/e2) – intensity width] of the 720-nm laser beam to 24% of the diameter of the back aperture of the objective (x20, 0.95 NA). We refer to this mode of illumination as macro excitation (for review see Helmchen and Denk 2005).
The image of a 3-µm fluorescent bead clearly shows that the dimension of two-photon macro excitation extends much further along the Z-axis than that of two-photon micro excitation (Fig. 1, A–C). Nonlinear excitation is an advantage of the two-photon method over one-photon excitation because this greatly restricts axial resolution. The linear nature of one-photon excitation results in the same number of chromophores being excited in each Z section throughout the entire light path (Fig. 1, A and B). Thus when UV light is used for uncaging, the one-photon approach will inevitably stimulate many presynaptic neurons and many dendrites of the postsynaptic neuron that are within the light cone.
|
To apply 2pMAPG to neuron stimulation, we prepared an acute slice preparation of the juvenile (P15–P19) rat visual cortex and made whole cell current-clamp recordings from layer 2/3 pyramidal neurons. We continuously recirculated the extracellular solution with CDNI-Glu (Ellis-Davies et al. 2007) in the recording chamber. Furthermore, to activate more glutamate receptors, we increased the total volume of the excitation by moving the uncaging spot in a small 3 x 3 pattern (Fig. 2, A and B).
|
50 mW, illumination time of 18 ms) induced an AP, the resting membrane potential of all recorded cells became more positive than –50 mV after some stimulation (n = 5). We assume that this response was due to the high energy of the laser beam causing photodamage when concentrated in a small area. In contrast, we found that single pulses of 2pMAPG at/near the soma of the recorded cell frequently induced three to four APs without clear photodamage (Supplemental Fig. S1).1 However, when we recorded only the postsynaptic neuron and stimulated its presynaptic neurons by 2pMAPG, it was difficult to determine whether a single or several adjoining presynaptic neurons triggered multiple EPSCs in the recorded cell. To ensure that neurons fire only once per 2pMAPG, we included an NMDA-receptor antagonist, D-APV, in the extracellular solution in all of the following experiments. D-APV perfusion reduced the long depolarization time of the stimulated neuron and considerably inhibited induction of more than one AP.
2pMAPG reproducibly induced APs 20 times at 0.5 Hz without obvious photodamage (four of five cells fired every time and one cell failed 3/20 times). The latency of the AP from the onset of stimulation was also stable (Fig. 2, C and D). Increasing the laser power induced a larger depolarization of the recorded cell supralinearly until inducing an AP (power exponent, 2.1 ± 0.2, n = 4; Fig. 2E). After inducing an AP, further power increases shortened the latency of the AP (Fig. 2E). The depolarization by 2pMAPG was abolished by perfusion with D-APV and the non-NMDA–receptor blocker CNQX (data not shown). No AP was induced by laser illumination when caged glutamate was absent from the perfusate. Thus the induction of depolarization and the AP depend on the two-photon excitation of caged glutamate and the activation of iontophoretic glutamate receptors, rather than on the excitation of endogenous chromophores (Hirase et al. 2002).
3-D resolution of mapping by 2pMAPG
To determine where 2pMAPG induced an AP from the recorded cell, we mapped the occurrence of APs over a large area (>300 x 300 µm) at several depths within the tissue slice. The axial interval was 20–25 µm. We divided the mapped area at each X–Y plane into arrays of pixels (spacing, 19 µm) and performed 2pMAPG at each pixel. The brain tissue scatters light and the strength of scattering is described by the average length of the distance between scattering events (ls) (Helmchen and Denk 2005). Thus to keep the laser power constant (P = 30–35 mW) in the mapped plane at depth z from the slice surface, we adjusted the laser power before entering the tissue slice to P/e–z/ls. We set ls at 125–130 µm, which compensated for the decrease in two-photon excitation of autofluorescence with increasing depth of the imaged plane (data not shown). We set the time window for detecting APs to be 0–40 ms after stimulation onset.
During 3-D mapping by 2pMAPG, APs occurred at pixels corresponding to some parts of the perisomatic and proximal dendritic areas (Fig. 3, A–C). Each recorded cell possessed AP-evoking pixels at one to four planes across the plane of the soma (Fig. 3B). The mean distance between each AP-evoking pixel and the Z-axial line passing through the center of the soma was 37 ± 4 µm, whereas the mean distance between each AP-evoking pixel and the X–Y plane of the center of the soma was 16 ± 2 µm (n = 13 cells, including 9 and 4 cells under the whole cell and loose-seal cell-attached recordings, respectively). Thus the mean 3-D distance between each AP-evoking pixel and the center of the soma was 44 ± 4 µm and this distance was almost independent of the depth of the cell (r = –0.21, Fig. 3D). The total number of AP-evoking pixels per 3-D mapping was 7.6 ± 1.4 (n = 13 cells) and this number was independent of the depth of the cell (r = –0.12, Fig. 3E). Figure 3F shows the distribution of the latency of APs at AP-evoking pixels in the whole cell and cell-attached modes. We found that almost all AP-evoking pixels (97/99) exhibited a single AP.
|
cell [µm–3], the total number of APs during mapping will be 109 x cell x NAPpixel, where cell and NAPpixel represent, respectively, the density of excitatory pyramidal neurons and the cumulative number of spikes per cell during mapping by 2pMAPG. The total number of APs can also be expressed as 109/Vpixel x NAPcell, where Vpixel is the volume of a single pixel and 109 [µm3]/Vpixel [µm3] is the total number of 2pMAPG for the unit volume. Thus 109 x cell x NAPpixel = 109/Vpixel x NAPcell and NAPcell = NAPpixel x Vpixel x cell. The density of neurons in the rat visual cortex ranges from about 60,000 to about 90,000 mm–3 (Gabbott and Stewart 1987; Miki et al. 1997; Peters et al. 1985; Skoglund et al. 1996) and, according to Peters et al. (1985), 85% of the neurons are pyramidal neurons. Thus assuming that NAPpixel, Vpixel, and cell in layer 2/3 were 7.6, 19 x 19 x 25 µm, and 0.85 x 80,000 mm–3 (=6.8 x 10–5 µm–3), respectively, the NAPcell in layer 2/3 was only about 5 (4.7). Mapping of excitatory neurons innervating layer 2/3 pyramidal neurons
We next determined whether we could reliably detect unitary EPSCs at the recorded cell when 2pMAPG stimulated presynaptic neurons of that cell. We made voltage-clamp recordings from the layer 2/3 pyramidal neurons at –65 mV, which is near the reversal potential for chloride ions, to detect EPSCs. Considering the distribution of AP latency (11–37 ms, Fig. 3F), we measured evoked currents whose peaks occurred within 15–45 ms after the onset of stimulation.
Evoked currents were found by performing 2pMAPG at many sites surrounding the recorded cell. After a current was detected, we fixed the site of 2pMAPG and recorded the currents evoked by repetitive 2pMAPG (10 times at 0.5 Hz, n = 16 sites for 6 cells). 2pMAPG induced two kinds of inward currents: one had a long onset latency (26.9 ± 1.7 ms, n = 5 sites for 4 cells) and a fast rise time (2.7 ± 0.4 ms) (Fig. 4, A and B), whereas the other showed a short onset latency (5.2 ± 0.8 ms, n = 11 sites for 2 cells) and a slow rise time (13.6 ± 1.2 ms) (Fig. 4, C and D). The earlier currents reflected the direct activation of glutamate receptors on the recorded cell because they were insensitive to the sodium-channel blocker TTX (Fig. 4E). These early currents were induced by 2pMAPG only at/near the soma and dendrites of the recorded cell and could be repetitively induced without failure. The mean coefficient of variation (CV) of the peak amplitudes was 0.11 ± 0.01 (n = 11). In contrast, the later currents reflected unitary EPSCs that were evoked by a neuron presynaptic to the recorded cell because they were sensitive to TTX. Moreover, these later currents showed a failure in induction of 0.08 ± 0.06 (n = 5) and the mean CV of the peak amplitudes, excluding the failures, was 0.29 ± 0.08.
|
|
). The failure of AP induction may be due to the smaller number of activated glutamate receptors. Thus in this study, almost all EPSCs that were evoked by 2pMAPG originated from medium-sized pyramidal neurons in layers 2/3 and 4 and we estimated the number of presynaptic pyramidal neurons only in layer 2/3. We performed several procedures to construct maps of synaptic connections. We first excluded pixels that showed large direct currents because these currents masked EPSCs. The pixels that showed these large currents occupied 66 ± 4 and 31 ± 2% of the total pixels in layer 2/3 whose lateral distances from the soma of the recorded cell were 0–50 and 50–100 µm, respectively (n = 6 recorded cells). We next measured the number and amplitude of the evoked currents whose rise time was <5 ms; the small direct currents with slow rise time were not included. Since the baseline current traces showed SD of 2–2.7 pA, we selected currents whose peak amplitudes were >8 pA (>3SD) as EPSCs. 2pMAPG induced one, two, and three EPSCs at 89 ± 1, 10 ± 1, and 1 ± 0.3%, respectively, of all EPSC-evoking pixels. The maximal amplitude of EPSCs was 65.9 pA.
A representative reconstruction of the map of synaptic connections is shown in Figs. 5B and S2, B and E. White pixels represent pixels that showed large, slow, direct currents (asterisk in Fig. S2C). The value at each pixel, except for the white pixels, represents the sum of the amplitudes of the EPSCs at each EPSC-evoking pixel. We detected unitary EPSCs at both neighboring and distant EPSC-evoking pixels (Figs. 5C and S2, C and F), whereas clusters of EPSC-evoking pixels were frequently found near the recorded cell, but not in distant regions. Figure 5D shows the Z-stacked map of synaptic connections in the experiment in Fig. 5B. The value at each pixel represents the average along the depth. Furthermore, we averaged the Z-stacked maps centered with respect to the location of the recorded cell somata (n = 6, Fig. 5E). This map indicates the columnar organization of synaptic connectivity. The lateral width of the cluster of synaptic connections was about 300 µm, whereas the vertical width was about 500 µm including the layer 4 area. We refer to the regions at lateral distances of <150 µm and between 150 and 300 µm as the columnar core region and the outer sparse region, respectively.
Functional distribution of connected neurons in the rat visual cortex
We next quantified the distribution of unitary EPSCs along the lateral distance (Fig. 6A). For this analysis, we examined each EPSC before the summation used for the displays in Figs. 5B and S2, B and E. To quantify the distribution of presynaptic neurons in layers 2/3 and layer 4 for each recorded cell, we defined an EPSC density (EPSC) at distances between ri–1 and ri (ri = 50 x i µm, for i = 0–6) as
 |
EPSC gives the average number of EPSCs per 2pMAPG. In both layers 2/3 and 4, EPSC fell off sharply as the distance increased to 150 µm, but small numbers of EPSCs remained at distances of 150–300 µm (Fig. 6B). To determine whether the distribution of strong connections was similar, we selected EPSCs whose amplitudes were in the top 10% in all mapped areas for each recorded cell and estimated the EPSC density in each layer again (EPSC10%) (Fig. 6C). The minimal EPSC amplitude within the top 10% was 24.8 ± 1.7 pA (n = 6). In layer 4, EPSC10% was still slightly greater than zero at distances of 150–300 µm (see also Fig. S2F), whereas EPSC10% in layer 2/3 was almost zero at >200 µm.
|
EPSC10%.
Since single 2pMAPG caused the neurons to fire once, EPSC also reflected the number of AP-evoking presynaptic neurons per 2pMAPG. Since the number of AP-evoking neurons per 2pMAPG is NAPcell, the connection probability of pairs of pyramidal neurons (pcon) is given by EPSC/NAPcell. Assuming that NAPcell in layer 2/3 is 4.7, we found that pcon in layer 2/3 within a cylinder with a radius of 50 µm was 0.10 ± 0.02 (Fig. 6B). Using the value of pcon for each distance and a vertical length of the layer 2/3 area of 339 ± 18 µm (mean ± SD, n = 6 slices), we estimated the number of presynaptic pyramidal neurons for each distance (Ncon) in layer 2/3 as pcon x cell x 
(ri2 – ri–12) x 339 (Fig. 6E). Ncon in layer 2/3 increased with increasing lateral distance 
150 µm and then decreased. The total number of Ncon in layer 2/3 within a cylinder of radius of 300 µm was about 130, approximately one third of which were scattered in the outer sparse region (Fig. 6F). In contrast, the percentage of Ncon calculated from EPSC10% in layer 2/3 was about 15% in the outer sparse region (green symbols in Fig. 6, E and F).
|
|
DISCUSSION |
|---|
|
Two-photon micro photolysis of caged glutamate has been used to stimulate single synapses in many studies (Araya et al. 2006; Beique et al. 2006; Carter and Sabatini 2004; Matsuzaki et al. 2001, 2004; Smith et al. 2003; Sobczyk et al. 2005; Tanaka et al. 2005). With 2pMAPG, we have increased the volume of two-photon excitation laterally and axially by reducing the effective NA of the objective (x20, 0.95 NA). As a result, 2pMAPG activates a very large number of glutamate receptors, which are enough to generate an AP without photodamage.
Very recently, Nikolenko et al. (2007) also performed two-photon photolysis of MNI-glutamate to induce APs in individual neurons using a diffractive optical element (DOE). In these experiments, the excitation volume was increased laterally by splitting the laser beam into five beams by DOE. It is extremely difficult to compare the actual excitation volumes obtained under our conditions with those obtained under their conditions because the effective NAs of the four objectives that they used (x20, 0.95 NA; x40, 0.8 NA; x20 0.5 NA; and x10, 0.3 NA) were not described. Nikolenko et al. (2007) performed two-photon uncaging of glutamate in the perisomatic areas of neurons that were visually identified by loading the acetoxymethyl ester of a Ca2+ indicator. They assumed that each visually identified neuron was presynaptic to the recorded cell when two-photon uncaging of glutamate at the perisomatic area of the neuron induced postsynaptic responses in the recorded cell. Using 3-D mapping by 2pMAPG, however, we have found that it is difficult to identify presynaptic neurons from the locations of the EPSC-evoking pixels because 2pMAPG presumably induced APs not only in a particular cell, but also in about 4 neighboring cells. Actually, APs were frequently induced by 2pMAPG at the proximal dendrites of neurons in addition to their perisomatic areas. The number of AP-evoking cells per 2pMAPG (5) is less than one sixth of the estimated numbers (34 and 54) of AP-evoking excitatory neurons (in layers 4 and 5, respectively) per one-photon UV uncaging of glutamate (Shepherd et al. 2005). Moreover, in our 2pMAPG, the number of AP-evoking cells was similar and independent of the depth of the mapped plane because the laser power before entering the tissue slice increased with the increase in the depth of the mapped plane, to compensate for the effect of the light scattering in the tissue. In one-photon uncaging of glutamate, however, this type of manipulation would increase the number of AP-evoking cells throughout the light cone.
Thus 2pMAPG has, for the first time, permitted 3-D mapping of unitary EPSCs. In addition, we could sometimes induce EPSCs without direct currents by illuminating pixels whose distances from the soma were less than the 3-D resolution of 44 µm for AP induction. Additionally, we could quantify the synaptic connectivity near the recorded cells. This would be difficult to accomplish by UV uncaging of glutamate, which induces large direct currents in large areas near the recorded cell (Douglas and Martin 2004).
Functional microarchitecture of synaptic connections in the rat visual cortex
For the first time, we have estimated the columnar core region with a lateral width of 150 µm in the rat visual cortex, which includes strong connections and a majority of the presynaptic neurons. Since the rodent visual cortex does not possess orientation columns (Girman et al. 1999; Hubener 2003; Ohki et al. 2005), the columnar core region of synaptic connectivity is not sufficient to form them. In contrast, the retinotopic map is preserved in the rodent visual cortex (Gias et al. 2005; Hubener 2003) and receptive fields have been mapped even at P17 in rats (Fagiolini et al. 1994). Thus the columnar core organization may play a role in sensory processing within the same and/or nearby receptive fields. Our estimation of pcon within the core region in layer 2/3 is similar to that estimated from the paired recordings of Holmgren et al. (2003). These researchers examined layer 2/3 pyramidal neurons in P14–P16 rat visual and somatosensory areas and reported that the connection probability in layer 2/3 decreased from 0.09 to 0.01 with increasing lateral distance from the target cell over the range of 25–200 µm. However, we set a threshold of 8 pA to detect EPSCs, so that weaker connections with small current amplitudes or a low release probability were discarded from our analysis. Thus the real connection probability may be higher than our estimate.
We also found that pcon in layer 2/3 in the outer region was about 1%, which has not been previously estimated. We did not find periodic clusters of presynaptic neurons in the outer region. From a functional point of view, this verifies previous anatomical reports that terminal clusters are not contained in long-range horizontal arborization of layer 2/3 axon collaterals in the rat visual cortex (Burkhalter 1989; Lund et al. 1993). Although EPSC in the outer region is lower than that in the core region, the volume of the outer region is larger than that of the core region. This geometry indicates that there are significant numbers of sparse projections from the outer region in both layers 2/3 and 4. This observation is consistent with results of Ca2+ imaging of spontaneous activity of a large population of neocortical neurons (Ikegaya et al. 2004), in which temporal sequences of cellular activity within dispersed space as well as columnar and clustered regions were detected. Long synaptic connections may prevent the columnar core region from forming a clear boundary and may play a critical role in the feedforward convergence/divergence of information from layer 4 to layer 2/3 and its horizontal spreading within layer 2/3.
Future applications of 2pMAPG
2pMPAG enables us to reveal the distribution of synaptic connectivity and the strength of unitary EPSCs in microcircuits. In the rat visual cortex, pairs of synaptically connected neurons frequently receive common inputs and it has been suggested that fine-scale connections play a role in the formation of the functional specificity of neurons (Yoshimura et al. 2005). Thus the combination of paired recording and mapping by 2pMAPG would reveal the 3-D distribution of such fine-scale networks. In the cat visual cortex, a linear boundary completely segregates functional columns with opposite direction preferences (Ohki et al. 2005). Mapping by 2pMAPG near the border between the direction columns in the cat visual cortex would clarify the functional connectivity patterns that form the sharp columnar border with single-cell resolution.
Since near-infrared light for two-photon excitation can penetrate deeply into tissues, 2pMAPG may be used for stimulating individual neurons in vivo. In fact, in vivo two-photon imaging has allowed observation of the somata of layer 2/3 pyramidal neurons, as well as those of layer 5 neurons (Helmchen and Denk 2005). If a method for applying a high concentration of caged glutamate in cortical tissues becomes available, in vivo 2pMAPG will open up a new realm of research into the perturbation of sensory processing and manipulation of neural activity with single-cell resolution.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
1 The online version of this article contains supplemental data. 
Address for reprint requests and other correspondence: M. Matsuzaki, Division of Biophysics, Center for Disease Biology and Integrative Medicine, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo, 113-0033, Japan (E-mail: mzakim{at}m.u-tokyo.ac.jp)
|
|
REFERENCES |
|---|
|
Arenkiel BR, Peca J, Davison IG, Feliciano C, Deisseroth K, Augustine GJ, Ehlers MD, Feng G. In vivo light-induced activation of neural circuitry in transgenic mice expressing channelrhodopsin-2. Neuron 54: 205–218, 2007.[CrossRef][Web of Science][Medline]
Beique JC, Lin DT, Kang MG, Aizawa H, Takamiya K, Huganir RL. Synapse-specific regulation of AMPA receptor function by PSD-95. Proc Natl Acad Sci USA 103: 19535–19540, 2006.
Burkhalter A. Intrinsic connections of rat primary visual cortex: laminar organization of axonal projections. J Comp Neurol 279: 171–186, 1989.[CrossRef][Web of Science][Medline]
Callaway EM, Katz LC. Photostimulation using caged glutamate reveals functional circuitry in living brain slices. Proc Natl Acad Sci USA 90: 7661–7665, 1993.
Carter AG, Sabatini BL. State-dependent calcium signaling in dendritic spines of striatal medium spiny neurons. Neuron 44: 483–493, 2004.[CrossRef][Web of Science][Medline]
Dantzker JL, Callaway EM. Laminar sources of synaptic input to cortical inhibitory interneurons and pyramidal neurons. Nat Neurosci 3: 701–707, 2000.[CrossRef][Web of Science][Medline]
DeAngelis GC, Ghose GM, Ohzawa I, Freeman RD. Functional micro-organization of primary visual cortex: receptive field analysis of nearby neurons. J Neurosci 19: 4046–4064, 1999.
Douglas RJ, Martin KA. Neuronal circuits of the neocortex. Annu Rev Neurosci 27: 419–451, 2004.[CrossRef][Web of Science][Medline]
Ellis-Davies GCR, Matsuzaki M, Paukert M, Kasai H, Bergles DE. 4-Carboxymethoxy-5,7-dinitroindolinyl-glu: an improved caged glutamate for expeditious ultraviolet and 2-photon photolysis in brain slices. J Neurosci 27: 6601–6604, 2007.
Fagiolini M, Pizzorusso T, Berardi N, Domenici L, Maffei L. Functional postnatal development of the rat primary visual cortex and the role of visual experience: dark rearing and monocular deprivation. Vision Res 34: 709–720, 1994.[CrossRef][Web of Science][Medline]
Feldmeyer D, Lubke J, Silver RA, Sakmann B. Synaptic connections between layer 4 spiny neurone-layer 2/3 pyramidal cell pairs in juvenile rat barrel cortex: physiology and anatomy of interlaminar signalling within a cortical column. J Physiol 538: 803–822, 2002.
Gabbott PLA, Stewart MG. Distribution of neurons and glia in the visual cortex (area 17) of the adult albino rat: a quantitative description. Neuroscience 21: 833–845, 1987.[CrossRef][Web of Science][Medline]
Gias C, Hewson-Stoate N, Jones M, Johnston D, Mayhew JE, Coffey PJ. Retinotopy within rat primary visual cortex using optical imaging. Neuroimage 24: 200–206, 2005.[CrossRef][Web of Science][Medline]
Girman SV, Sauve Y, Lund RD. Receptive field properties of single neurons in rat primary visual cortex. J Neurophysiol 82: 301–311, 1999.
Helmchen F, Denk W. Deep tissue two-photon microscopy. Nat Methods 2: 932–940, 2005.[CrossRef][Web of Science][Medline]
Hirase H, Nikolenko V, Goldberg JH, Yuste R. Multiphoton stimulation of neurons. J Neurobiol 51: 237–247, 2002.[CrossRef][Web of Science][Medline]
Holmgren C, Harkany T, Svennenfors B, Zilberter Y. Pyramidal cell communication within local networks in layer 2/3 of rat neocortex. J Physiol 551: 139–153, 2003.
Hubel DH, Wiesel TN. Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. J Physiol 160: 106–154, 1962.
Hubener M. Mouse visual cortex. Curr Opin Neurobiol 13: 413–420, 2003.[CrossRef][Web of Science][Medline]
Ikegaya Y, Aaron G, Cossart R, Aronov D, Lampl I, Ferster D, Yuste R. Synfire chains and cortical songs: temporal modules of cortical activity. Science 304: 559–564, 2004.
Kalisman N, Silberberg G, Markram H. The neocortical microcircuit as a tabula rasa. Proc Natl Acad Sci USA 102: 880–885, 2005.
Katz LC, Dalva MB. Scanning laser photostimulation: a new approach for analyzing brain circuits. J Neurosci Methods 54: 205–218, 1994.[CrossRef][Web of Science][Medline]
Lund JS, Yoshioka T, Levitt JB. Comparison of intrinsic connectivity in different areas of macaque monkey cerebral cortex. Cereb Cortex 3: 148–162, 1993.
Matsuzaki M, Ellis-Davies GCR, Nemoto T, Miyashita Y, Iino M, Kasai H. Dendritic spine geometry is critical for AMPA receptor expression in hippocampal CA1 pyramidal neurons. Nat Neurosci 4: 1086–1092, 2001.[CrossRef][Web of Science][Medline]
Matsuzaki M, Honkura N, Ellis-Davies GCR, Kasai H. Structural basis of long-term potentiation in single dendritic spines. Nature 429: 761–766, 2004.[CrossRef][Medline]
Miki T, Fukui Y, Itoh M, Hisano S, Xie Q, Takeuchi Y. Estimation of the numerical densities of neurons and synapses in cerebral cortex. Brain Res Brain Res Protocols 2: 9–16, 1997.[CrossRef][Medline]
Mountcastle VB. Modality and topographic properties of single neurons of cat's somatic sensory cortex. J Neurophysiol 20: 408–434, 1957.
Nikolenko V, Poskanzer KE, Yuste R. Two-photon photostimulation and imaging of neural circuits. Nat Methods 4: 943–950, 2007.[CrossRef][Web of Science][Medline]
Ohki K, Chung S, Ch'ng YH, Kara P, Reid RC. Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature 433: 597–603, 2005.[CrossRef][Medline]
Peters A, Kara DA. The neuronal composition of area 17 of rat visual cortex. I. The pyramidal cells. J Comp Neurol 234: 218–241, 1985.[CrossRef][Web of Science][Medline]
Peters A, Kara DA, Harriman KM. The neuronal composition of area 17 of rat visual cortex. III. Numerical considerations. J Comp Neurol 238: 263–274, 1985.[CrossRef][Web of Science][Medline]
Schubert D, Kotter R, Zilles K, Luhmann HJ, Staiger JF. Cell type-specific circuits of cortical layer IV spiny neurons. J Neurosci 23: 2961–2970, 2003.
Shepherd GMG, Pologruto TA, Svoboda K. Circuit analysis of experience-dependent plasticity in the developing rat barrel cortex. Neuron 38: 277–289, 2003.[CrossRef][Web of Science][Medline]
Shepherd GMG, Stepanyants A, Bureau I, Chklovskii D, Svoboda K. Geometric and functional organization of cortical circuits. Nat Neurosci 8: 782–790, 2005.[CrossRef][Web of Science][Medline]
Shoham S, O'Connor DH, Sarkisov DV, Wang SS. Rapid neurotransmitter uncaging in spatially defined patterns. Nat Methods 2: 837–843, 2005.[CrossRef][Web of Science][Medline]
Skoglund TS, Pascher R, Berthold CH. Heterogeneity in the columnar number of neurons in different neocortical areas in the rat. Neurosci Lett 208: 97–100, 1996.[CrossRef][Web of Science][Medline]
Smith MA, Ellis-Davies GCR, Magee JC. Mechanism of the distance-dependent scaling of Schaffer collateral synapses in rat CA1 pyramidal neurons. J Physiol 548: 245–258, 2003.
Sobczyk A, Scheuss V, Svoboda K. NMDA receptor subunit-dependent [Ca2+] signaling in individual hippocampal dendritic spines. J Neurosci 25: 6037–6046, 2005.
Tanaka J, Matsuzaki M, Tarusawa E, Momiyama A, Molnar E, Kasai H, Shigemoto R. Number and density of AMPA receptors in single synapses in immature cerebellum. J Neurosci 25: 799–807, 2005.
Thomson AM, West DC, Wang Y, Bannister AP. Synaptic connections and small circuits involving excitatory and inhibitory neurons in layers 2–5 of adult rat and cat neocortex: triple intracellular recordings and biocytin labelling in vitro. Cereb Cortex 12: 936–953, 2002.
Wang H, Peca J, Matsuzaki M, Matsuzaki K, Noguchi J, Qiu L, Wang D, Zhang F, Boyden E, Deisseroth K, Kasai H, Hall WC, Feng G, Augustine GJ. High-speed mapping of synaptic connectivity using photostimulation in channelrhodopsin-2 transgenic mice. Proc Natl Acad Sci USA 104: 8143–8148, 2007.
Wickersham IR, Lyon DC, Barnard RJ, Mori T, Finke S, Conzelmann KK, Young JA, Callaway EM. Monosynaptic restriction of transsynaptic tracing from single, genetically targeted neurons. Neuron 53: 639–647, 2007.[CrossRef][Web of Science][Medline]
Yoshimura Y, Dantzker JL, Callaway EM. Excitatory cortical neurons form fine-scale functional networks. Nature 433: 868–873, 2005.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  T. Sasaki, G. Minamisawa, N. Takahashi, N. Matsuki, and Y. Ikegaya Reverse Optical Trawling for Synaptic Connections In Situ J Neurophysiol, July 1, 2009; 102(1): 636 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Kwan What Can Population Calcium Imaging Tell Us About Neural Circuits? J Neurophysiol, December 1, 2008; 100(6): 2977 - 2980. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
