Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2010 Sep 27;588(Pt 22):4489–4505. doi: 10.1113/jphysiol.2010.197012

Monosynaptic excitatory inputs to spinal lamina I anterolateral-tract-projecting neurons from neighbouring lamina I neurons

Liliana L Luz 1,2, Peter Szucs 1, Raquel Pinho 1, Boris V Safronov 1,2
PMCID: PMC3008853  PMID: 20876196

Abstract

Spinal lamina I receives nociceptive primary afferent input to project through diverse ascending pathways, including the anterolateral tract (ALT). Large projection neurons (PNs) form only a few per cent of the cell population in this layer, and little is known about their local input from other lamina I neurons. We combined single-cell imaging in the isolated spinal cord, paired recordings, 3-D reconstructions of biocytin-labelled neurons and computer simulations to study the monosynaptic input to large ALT-PNs from neighbouring (somata separated by less than 80 μm) large lamina I neurons. All 11 connections identified were excitatory. We have found that an axon of a presynaptic neuron forms multiple synapses on an ALT-PN, and both Ca2+-permeable and Ca2+-impermeable AMPA receptors are involved in transmission. The monosynaptic EPSC latencies (1–12 ms) are determined by both post- and presynaptic factors. The postsynaptic delay, resulting from the electrotonic EPSC propagation in the dendrites of an ALT-PN, could be 4 ms at most. The presynaptic delay, caused by the spike propagation in a narrow highly branched axon of a local-circuit neuron, can be about 10 ms for neighbouring ALT-PNs and longer for more distant neurons. In many cases, the EPSPs evoked by release from a lamina I neuron were sufficient to elicit a spike in an ALT-PN. Our data show that ALT-PNs can receive input from both lamina I local-circuit neurons and other ALT-PNs. We suggest that lamina I is a functionally interconnected layer. The intralaminar network described here can amplify the overall output from the principal spinal nociceptive projection area.

Introduction

Spinal lamina I processes diverse modalities of nociceptive input and projects through ascending tracts to specific areas of the brainstem and thalamus (Willis & Coggeshall, 1991; Todd et al. 2000; Lima & Almeida, 2002). Individual lamina I neurons receive and integrate monosynaptic Aδ- and C-fibre afferent inputs (Grudt & Perl, 2002; Ikeda et al. 2003) originating from several segmental dorsal roots (Pinto et al. 2010). Lamina I neurons were shown to receive local monosynaptic inputs from excitatory and inhibitory lamina II (substantia gelatinosa, SG) neurons (Lu & Perl, 2005; Graham et al. 2007; Santos et al. 2007; Kato et al. 2009; Santos et al. 2009). The input from an SG excitatory interneuron can be sufficiently strong to excite a postsynaptic neuron (Santos et al. 2009). The functional connectivity between lamina I neurons, to our knowledge, has not been described so far.

Lamina I contains both projection neurons (PNs) and non-projection, or local-circuit, neurons. PNs form only less than 5% of the neuronal population (Spike et al. 2003; Todd & Koerber, 2006). The main axon of some PNs crosses the spinal grey matter and ascends in the contralateral anterolateral tract (ALT). It has recently been shown that the main axon of large mediolaterally oriented ALT-PNs from the lateral lamina I gives rise to ipsilateral collaterals which can spread to the dorsal, lateral or ventral regions of the spinal cord (Szucs et al. 2010). According to their size and morphology, these ALT-PNs most likely correspond to Waldeyer cells of the classical literature (Waldeyer, 1889; Cajal, 1909). The majority of lamina I neurons, however, are local-circuit neurons (Bice & Beal, 1997). Some of them can also have large cell bodies and extensive axons forming a dense network interconnecting the dorsal grey matter of several spinal cord segments (Szucs et al. 2010), suggesting a possibility of numerous contacts with other lamina I neurons. In spite of this, the functional role of lamina I local-circuit neurons as well as their interaction with the ALT-PNs is poorly understood. One reason for this is that a low percentage (<5%) of PNs, together with a low probability (∼10%) of finding pairs of monosynaptically connected superficial dorsal horn (SDH) neurons in spinal cord slices (Santos et al. 2007), makes a systematic study of the ALT-PN connections extremely difficult. Furthermore, large lamina I neurons are most sensitive to damage introduced by a slicing procedure, and therefore, their properties and functional connectivity could hardly be studied using slices.

For these reasons, we used the isolated spinal cord preparation (Szucs et al. 2009) preserving large ALT-PNs and local-circuit neurons, along with their connections via long axodendritic pathways (Szucs et al. 2010), to study whether (1) an ALT-PN can receive inputs from another lamina I neuron, (2) the synaptic transmission involves Ca2+-permeable AMPA receptors (CP-AMPARs) and Ca2+-impermeable AMPA receptors (CI-AMPARs), (3) release from a lamina I neuron can be sufficient to excite an ALT-PN, and whether (4) spinal lamina I can be considered as an interconnected cell layer.

Methods

Ethical approval

Laboratory Wistar rats (P14–P20) were killed by decapitation in accordance with Portuguese national guidelines (Direcção Geral de Veterinária, Ministério da Agricultura) after anaesthesia with intraperitoneal injection of sodium pentobarbital (30 mg kg−1) and subsequent check for lack of pedal withdrawal reflexes. The experiments were carried out according to the guidelines laid down by the institution's animal welfare committee (Comissão de Ética do Instituto de Biologia Molecular e Celular) and comply with the policies and regulations of The Journal of Physiology (Drummond, 2009). The number of animals used in this study was 84.

Preparation of the spinal cord

The vertebral column was quickly cut out and immersed in oxygenated artificial cerebrospinal fluid (ACSF) at room temperature. The lumbar spinal cord was dissected and the pia mater was locally removed in the region of interest with forceps and scissors, to provide access for the recording and stimulating pipettes. The spinal cord was glued with cyanoacrylate adhesive to a golden plate (the dorsolateral spinal cord surface was up) and transferred to the recording chamber (Fig. 1A). All measurements were done at 22–24°C.

Figure 1. Recording from large lamina I neurons in the isolated spinal cord.

Figure 1

Open in a new tab

A, preparation of the lumbar spinal cord for recordings from pairs of synaptically connected lamina I neurons. An infrared LED was used as a source of oblique illumination for the cell imaging. Both pre- and postsynaptic neurons were located in the region between the dorsolateral funiculus (lateral border) and the dorsal root entry zone (medial border). A post hoc analysis of the major axon course was done to prove that the postsynaptic neuron was an ALT-PN. B, types of large somata of lamina I neurons seen under our experimental conditions. The 3-D reconstructions (bottom) were done from 5–10 serial images (top) taken in different focal planes (∼2 μm step). Contour lines of a soma and initial dendrites were traced in individual images into a 3-D modelling software (Cinema-4D, Maxon). a, a polygonal flattened soma (dorsoventral extent, ∼10 μm) with three or four major dendrites. This cell body resembled those of flattened neurons (Lima & Coimbra, 1986). b, a pyramidal soma with four dendrites leaving from the apices similar to that of a pyramidal neuron (Lima & Coimbra, 1986). c, a spindle-shaped soma with dendrites in a bipolar organization resembling those of fusiform cells (Lima & Coimbra, 1986). d, a soma appearing as a sphere (diameter, ∼20 μm) with numerous, sometimes poorly visible, dendrites; similar to the body of multipolar cells (Lima & Coimbra, 1986). All ALT-PNs (n = 40) had somata of the first three types, while the presynaptic neurons had frequently somata of the fourth type. M, medial; C, caudal.

Visualization of lamina I neurons

Lamina I neurons were visualized (Fig. 1A) through the intact dorsolateral white matter in the lumbar spinal cord using the oblique infrared light-emitting-diode (LED) illumination technique (Safronov et al. 2007; Szucs et al. 2009). Lamina I was identified on the basis of orientation of myelinated fibres in the dorsal white matter (Pinto et al. 2010). The neurons included in this study were located in the region between the dorsolateral funiculus (lateral border) and the dorsal root entry zone (medial border) (Fig. 1A). The white matter covering this part of lamina I is thin in young rats (Pinto et al. 2008b), allowing visually controlled tight-seal recordings from the superficial neurons. Large lamina I neurons could be clearly distinguished from deeper located SG neurons whose somata were smaller and appeared as a densely packed cell layer (Szucs et al. 2009).

Selection of the ALT-PNs

The percentage of PNs in the entire population of lamina I neurons is low (Spike et al. 2003; Todd & Koerber, 2006). However, we have recently found that, among the neurons with large cell bodies (>25 μm) in the lateral SDH, the ALT-PNs represent a majority (Szucs et al. 2010). For this reason, we selected as postsynaptic neurons those which had large mediolaterally oriented or round somata with the longest diameter above 25 μm. As putative presynaptic neurons, those with soma diameter above 20 μm were tested. The major types of the somata of large lamina I neurons are shown in Fig. 1B. For all postsynaptic neurons described here, the post hoc analysis of their axon course (reaching the contralateral ALT) confirmed that they were ALT-PNs.

Recording from postsynaptic neurons

Recordings from the postsynaptic neurons were done in whole-cell mode (Melnick et al. 2004a,b;). ACSF contained (in mm): NaCl 115, KCl 3, CaCl2 2, MgCl2 1, NaH2PO4 1, NaHCO3 25, and glucose 11 (pH 7.4 when bubbled with 95%–5% mixture of O2–CO2). The pipettes were pulled from thick-walled glass (BioMedical Instruments, Germany) and fire-polished (resistance, 4–5 MΩ). The pipette solution contained (in mm): KCl 3, potassium gluconate 150, MgCl2 1, BAPTA 1, Hepes 10 (pH 7.3 adjusted with KOH, final [K+] was 160 mm) and 1% biocytin. The amplifier was an EPC10-Double (HEKA, Lambrecht, Germany). The signal was low-pass filtered at 2.9 kHz and sampled at 10 kHz. The traces in Fig. 3C were additionally off-line-filtered at 1 kHz. Offset potentials were compensated before seal formation. Liquid junction potentials were calculated (15 mV) and corrected for in all experiments using the compensation circuitry of the amplifier. The blockers CNQX and IEM 1460 were from Sigma and Tocris, respectively.

Figure 3. A monosynaptic connection with transmitter release in multiple synapses and activation of CP- and CI-AMPARs.

Figure 3

Open in a new tab

Recordings from a connection with a postsynaptic ALT-PN. The axon of the presynaptic neuron (red) released transmitter in multiple synapses. A, the shorter- and longer-latency components of a composite monosynaptic EPSC. Four non-consecutive traces show the EPSC amplitude variation caused by fluctuation of the quantal transmitter release. The longer-latency component corresponded to activation of more remote synapse 3. B, fast transitions on the rising phase of the shorter-latency component. Two subcomponents corresponding to activation of synapses 1 and 2 could be activated either individually (no transitions) or together (traces with transitions) (5 non-consecutive traces). C, 10 consecutive traces showing a polysynaptic EPSC (synapse 4). Note the lack of failures. A large latency variation (indicated by a blue box) was caused by the variation in the spike initiation time in the intercalated neuron (blue) as described by Santos et al. (2009). Laminar location of the intercalated neuron could not be determined. D, both CP- and CI-AMPARs are involved in transmission in this connection. Effects of 20 μm IEM 1460 (CP-AMPAR blocker) and 10 μm CNQX (AMPA/kainate receptor blocker) are shown for averaged traces (each obtained from 10 consecutive episodes recoded at 1 Hz). For IEM 1460, development of the block is shown. Note that monosynaptic EPSCs (corresponding to synapses 1–2 and 3) are progressively reduced, while the polysynaptic one (synapse 4, indicated by an asterisk) disappears abruptly. Because of the variable latency, the polysynaptic EPSC appears smaller in the averaged trace. The polysynaptic component was recovered only after a long wash-out at the end of the experiment.

Stimulation and labelling of presynaptic neurons

Intact presynaptic neurons were selectively stimulated and labelled through the cell-attached pipette containing 500 mm NaCl and 1.5% biocytin. The pairs of monosynaptically connected neurons were identified as described by Santos et al. (2007). After the connected neurons have been found, the strength of the current pulse (duration, 1 ms) used for the stimulation of the presynaptic neuron was reduced from 100 nA to the ‘threshold + 20 nA’ level (Santos et al. 2009), to allow a non-damaging stimulation for longer periods of time (up to 1 h 45 min). For most presynaptic neurons in this study, the adjusted stimulation strength was 50–70 nA.

To test whether our stimulation protocol, originally developed for small SG neurons, can be also used for the fast spike initiation in large lamina I neurons, we did control measurements in 15 cells with soma diameter >20 μm (not shown). Direct whole-cell recordings of the voltage changes introduced by the cell-attached stimulation (both pipettes were positioned on the same soma) showed that, as in the case of SG neurons (see Fig. 3 from Santos et al. 2009), a spike in the axon initial segment of a lamina I neuron was reliably evoked within the 1 ms of the stimulation (n = 15). For this reason, the EPSC latencies in connections were calculated from the end of the 1 ms stimulation pulse (to the time moment when the evoked EPSC reached 10% of its peak amplitude).

When recordings from a connection have been completed, the presynaptic neuron was filled by biocytin through the cell-attached pipette (Szucs et al. 2009). Biocytin diffusion into the cell was facilitated by applying depolarizing current pulses (duration, 500 ms) of increasing amplitude (1–10 nA, 1 nA increment) at 1 Hz for 10 min.

Criteria for monosynaptic connections

In most cases, the EPSCs evoked by stimulation of a presynaptic neuron consisted of several components corresponding to transmitter release in multiple synapses of the same axon (see Fig. 3; and Santos et al. 2009 for SG neurons). According to our recent estimation, the release probability for a synapse (or a group of synapses with similar spatial location) responsible for the generation of an EPSC component is about 0.5 (Santos et al. 2009). Therefore, an EPSC component was identified as monosynaptic if it was observed in at least 7 of 10 consecutive stimulations (1 Hz) and its latency variation was within 1 ms. Neurons in a pair were considered to be connected monosynaptically if at least one EPSC component was identified as monosynaptic.

Histological processing and cell reconstruction

After fixation in 4% paraformaldehyde, the spinal cord was embedded in agar and parasagittal serial sections (thickness, 100 μm) were prepared with a tissue slicer (Leica, VT 1000S). To reveal biocytin, the sections were permeabilized with 50% ethanol and treated according to the avidin-biotinylated horseradish peroxidase method (ExtrAvidin-Peroxidase, diluted 1:1000) followed by a diaminobenzidine chromogen reaction. For 2-D reconstructions, sections were counterstained with 1% toluidine blue, to determine borders of the grey matter and laminae, and mounted in DPX (Fluka). For 3-D reconstructions, sections were treated with 1% OsO4 and embedded in Durcupan resin. Photomicrographs were taken with a Primo Star (Zeiss) microscope equipped with a Guppy (Allied Vision Technologies, Stadtroda, Germany) digital camera. Contrast and brightness of the images used for the figures were adjusted using Adobe Image Ready software.

The 2-D reconstructions in Figs 4 and 5 were done as follows. A window of the Neurolucida software (MBF Bioscience, Williston, VT, USA, workstation version) was dimmed and superimposed on the live digital image of the section (objective, 40×) by means of the Transparent Windows 2.2 application. All dendrites, somata and axons as well as contours of the grey and white matter were completely traced into a serial section data file in 2-D. During the reconstruction, the microscope stage was aligned manually. The location of the neuronal soma and processes were determined on reconstructions of individual sections where borders of the grey and white matter were clearly identifiable. The 2-D serial sections were overlaid, aligned using Neuroexplorer software (MBF Bioscience) and exported as bitmap images.

Figure 4. Inputs to an ALT-PN from a lamina I local-circuit neuron.

Figure 4

Open in a new tab

A, monosynaptic EPSCs with at least two components. Non-consecutive traces were chosen to show individual components and the composite EPSC. B, 2-D reconstruction from serial sections showing the major axon (blue) of the postsynaptic ALT-PN (soma and dendrites, black) reaching the rostral end of the spinal cord preparation. The presynaptic neuron: orange, soma and dendrites; red, axon. Upper part, a lateral view; lower part, a horizontal view (green line indicates the central canal). R, rostral; C, caudal; V, ventral; D, dorsal. Grey lines in B and C indicate the contours of the bottom of the serial sections. For clarity, some contour lines were omitted. C, the postsynaptic ALT-PN was a pyramidal neuron of the ventral-collateral-type. The presynaptic cell was a lamina I local-circuit neuron (multipolar type-IIa) with axon arborizations occupying the SDH of one entire segment. The region in the inset is shown amplified in D. D, three close appositions between the presynaptic axon and the postsynaptic dendrites were revealed (arrowheads). Two of them were formed by the same axon branch (right group).

Figure 5. Input to an ALT-PN from another ALT-PN.

Figure 5

Open in a new tab

Microphotographs (from different sections) and the 2-D reconstruction of two ALT-PNs (pyramidal neurons of the lateral-collateral-type). Postsynaptic neuron; soma and dendrites, black; axon, blue. Presynaptic neuron: soma and dendrites, orange; axon, red. The evoked EPSC showed only one component (5 consecutive traces). R, rostral; D, dorsal.

The 3-D reconstruction of an ALT-PN (Fig. 7) and a lamina I local-circuit neuron (Fig. 8) for computer modelling was done with the Neurolucida system (MBF Bioscience). Shrinkage was not corrected. The dendrogram in Fig. 7 and the axogram in Fig. 8 were constructed with Neuroexplorer (MBF Bioscience).

Figure 7. EPSC propagation in the postsynaptic ALT-PN.

Figure 7

Open in a new tab

A, the 3-D reconstruction of the dendrites and axon of an ALT-PN of the lateral-collateral-type. Perspective overview of the neuron depicting the orientation and course of the projecting axon and collaterals. Higher magnification image demonstrates major dendrites restricted to lamina I following the curvature of the dorsal horn surface. The projecting axon is indicated by an asterisk. Four major dendrites are shown by different colours. B, sagittal projection of the reconstructed neuron. Simulations were done for synapses (1–8) inserted on places indicated by arrowheads or a grey circle. C, dendrogram (the same colour code and synapse identification numbers as in B). D, simulations of the somatic voltage-clamp recording of the EPSCs evoked by activation of the corresponding synapses. The delays were measured as a time interval between the transmitter release (0 ms) and the moment when the corresponding EPSC reached 10% of its peak amplitude (shown by a grey bar for the slowest EPSC). E, EPSCs arising from two spatially close synapses (7 and 8) located in different major dendrites.

Figure 8. Spike propagation in the axon of a local-circuit neuron.

Figure 8

Open in a new tab

A, the 3-D reconstruction of the axon (black) and dendrites (red) of a large lamina I local-circuit neuron filled with biocytin in whole-cell mode. The highest order of the axon branches was 26. The axon had 647 endings and 4588 varicosities. Note that some of the longest branches were truncated. Axonal points for which the spike propagation simulation was done are labelled by coloured circles. B, higher magnification image demonstrating the axon (black) in the vicinity of the soma (red). The axon originated from a primary dendrite (red). In simulation, the spike was initiated in the primary axon at point 0 (stimulation was applied to the soma). C, axogram with indication of simulation points and their axon paths (indicated by the corresponding colours). The common parts of the axon paths are shown by the colour of the longer path. D, simulation of the spike propagation to points 1–8. The propagation time was counted from the moment of the spike initiation in the primary axon (point 0). Dashed lines correspond to potentials of 0 mV and −70 mV.

Computer simulations

The 3-D reconstructed neurons were imported to NEURON (version 7.1) software (Hines, 1993; Hines & Carnevale, 1997). The dendritic spines forming less than 3% of the dendritic membrane area were ignored. The resolution limit of a light microscope (0.2 μm) was used to set the diameters of all apparently thinner regions of axons and dendrites. The integration step was 50 μs. The spatial grid was set to 0.1 of the characteristic length (λ) for the axon and dendrites. The soma was divided in 10 segments. The specific membrane capacitance (Cm) and specific axial resistance (Ra) were assumed to be 1 μF cm−2 and 200 Ω cm, respectively. The specific membrane resistivity (Rm) was uniformly set to 100 kΩ cm2 to give, in agreement with our measurements, the slowest membrane time constant (τ0 = RmCm) of 100 ms. These settings of passive properties in the model of the ALT-PN gave the input resistance (RIN) of 0.33 GΩ, which was close to the experimentally determined value (0.38 GΩ) for the neuron reconstructed. Simulations were done for the resting/holding potential of −70 mV.

The ALT-PN model was used to analyse passive EPSC propagation in the dendritic tree. The synaptic input was modelled by a two-state kinetic scheme. The synaptic current (IS) was described by the following equation:

graphic file with name tjp0588-4489-m1.jpg

where τR is the conductance rise time constant (1 ms), τD is the conductance decay time constant (10 ms), gM is the maximum conductance (1.86 × 10−5μS), V is the membrane potential and ERev is the reversal potential (0 mV). The synapse locations are specified in Fig. 7.

The model of the lamina I local-circuit neuron was used to analyse the spike propagation time in the axon of the presynaptic neuron. Voltage-gated Na+ and K+ channels were inserted at uniform densities in all branches of the axon, whereas soma and dendrites were passive. The channel models and densities were from Melnick et al. (2004b) with the only modification being that to prevent the spike self-generation in the axon at high densities of Na+ channels, their steady-state activation variable (m∞) was set to 0 at potentials negative to −65 mV. Based on our recordings from tonic-firing neurons (Melnick et al. 2004b), the Na+ conductance (gNa) was 1800 mS cm−2. Assuming the Na+ channel conductance in dorsal horn neurons of 11.6 pS (Safronov et al. 1997; Safronov, 1999), each μm2 of the axonal membrane in the model contained 433 channels, 158 of which were available at a resting potential of −70 mV (calculated for voltage steps to +20 mV). The voltage-gated K+ conductance was 76 mS cm−2. In a uniform axon with a constant diameter of 1 μm, these channel densities allowed generation of a spike with a maximum depolarization rate of 576 V s−1 propagating at a conduction velocity of 0.38 m s−1. In the axon with a diameter of 0.2 μm (thinnest axons in our reconstructed neurons), the conduction velocity was 0.16 m s−1. These values were in a good agreement with the highest and lowest conduction velocities measured at 22–24°C for unmyelinated C-fibres in isolated dorsal roots (see Table 2 from Pinto et al. 2008a).

All data are given as means ± s.e.m.

Results

A total of 199 large lamina I neurons were filled with biocytin in the whole-cell mode. Full recovery of axonal and dendritic trees was achieved for 79 neurons, 44 of which (56%) were identified as ALT-PNs. Detailed anatomical description of a subset of these ALT-PNs is given in Szucs et al. (2010). Stable voltage- and current-clamp recordings were obtained from 40 ALT-PNs. For 28 of them, we searched for presynaptic neurons by testing 234 large cells. Monosynaptic connections were found in 11 cases and were all excitatory. We did not observe monosynaptic inhibitory inputs from large neurons; all recorded IPSCs were either polysynaptic or spontaneous. These 40 ALT-PNs, including 11 with paired neurons, form the basis of the present study. According to the morphology of their axon collaterals (Szucs et al. 2010), the 11 postsynaptic ALT-PNs were of the dorsal-collateral-type (n = 1), ventral-collateral-type (n = 4), lateral-collateral-type (n = 2), mixed-collateral-type (n = 2) and no-collateral-type (n = 2). Based on their somatodendritic organization (Lima & Coimbra, 1986), these neurons were classified as fusiform (n = 3), flattened (n = 5) and pyramidal (n = 3).

Passive membrane properties of large ALT-PNs

The resting membrane potential (measured with zeroed amplifier input, Santos et al. 2004) was −70.6 ± 1.0 mV (n = 40; range, −55 mV to −86 mV). The mean input resistance (RIN) and the slowest membrane time constant (τ0) measured in the current-clamp mode from a uniformly preset potential of −70 mV were 0.67 ± 0.06 GΩ and 91.1 ± 5.2 ms, respectively (n = 40).

The mean RIN in large ALT-PNs was considerably lower than those measured under our experimental conditions (1 mm BAPTA-buffered pipette solution or perforated-patch recording) in non-identified, mostly medium-sized, lamina I neurons (1.8 GΩ, Pinto et al. 2010) and small SG neurons (1.59 to 2.26 GΩ, Santos et al. 2007; Santos et al. 2009). Lower RIN in ALT-PNs could be explained by a larger neuronal membrane area or a lower specific membrane resistivity (Rm). To distinguish between these two possibilities, we analysed the relationship between RIN and τ0 in large and small neurons, assuming that, at a uniform specific membrane capacitance (Cm), τ0 could be used as a measure of Rm (τ0 = RmCm).

In Fig. 2, we compared our measurements for large ALT-PNs (red) and small (soma diameter, ∼10 μm) SG neurons (blue, data from Santos et al. 2009). The two groups appeared well-separated in the τ0versus RIN plot. Within each group, RIN increased with τ0 (oblique dashed lines) indicating that its reduction was caused, at least in part, by the drop in Rm. However, comparison of the mean values for the two groups has revealed that ALT-PNs had significantly lower RIN (P < 10−5, independent Student's t test; ratio of mean values was 2.58) but a slightly slower τ0 (Fig. 2, continuous lines). Therefore, lower mean RIN in ALT-PNs, in comparison with SG neurons, could be explained by a larger area of the neuronal membrane with similar Rm.

Figure 2. Comparison of passive membrane properties of ALT-PNs and SG neurons.

Figure 2

Open in a new tab

RIN and τ0 were measured from membrane responses evoked in current-clamp by a 500 ms hyperpolarizing current pulse of −10 pA. The neurons were kept at −70 mV. τ0versus RIN was plotted for 40 anatomically confirmed ALT-PNs (red circles) and 122 SG neurons (blue circles) from Santos et al. (2009). Linear fit of the data points in each group is shown by oblique dashed lines. The mean RIN and τ0 are indicated by vertical and horizontal continuous lines, respectively. For SG neurons, RIN = 1.73 ± 0.06 GΩ and τ0 = 76.2 ± 3.4 ms (n = 122). The mean values for ALT-PNs are given in the text.

Monosynaptic connections with transmitter release in multiple synapses

In 9 of 11 connections with a postsynaptic ALT-PN, stimulation of the presynaptic neuron evoked composite EPSCs corresponding to transmitter release in multiple synapses of the same axon. Detailed analysis of the monosynaptic nature of individual EPSC components was done as described recently (Santos et al. 2009). Recordings from an ALT-PN receiving at least three monosynaptic inputs and one polysynaptic input from the same lamina I neuron are shown in Fig. 3.

The shorter- and longer-latency components separated by a time interval of 4 ms could be clearly distinguished (Fig. 3A). Stability of the latencies and of the interval between the components was consistent with our criteria for the monosynaptic inputs located at different electrotonic distances from the soma (Santos et al. 2009). The rising phase of the longer-latency EPSC (the synapse labelled 3 on the figure) could not be analysed due to the overlap with the preceding component. However, fast transitions on the rising phase of the shorter-latency component (Fig. 3B) revealed the presence of two subcomponents corresponding to activation of synapses (labelled 1 and 2) with only a slight difference in their electrotonic distances from the soma (Santos et al. 2009).

In addition, the ALT-PN in this connection received a strong polysynaptic input (synapse 4, Fig. 3C). Its latency variation of ∼10 ms (indicated by a blue box) was consistent with our estimation for the variation of the spike initiation time in an intercalated neuron (Santos et al. 2009). Although activity of the intercalated neuron varied during the experiment, the lack of failures of the polysynaptic EPSC in 10 consecutive stimulations shown in Fig. 3C suggested that the presynaptic lamina I neuron could efficiently excite the intercalated neuron. Similar polysynaptic inputs were observed in six other connections. Laminar location of the intercalated neuron could not be determined in our study.

Thus, an ALT-PN can receive multiple synapses from the axon of another lamina I neuron. Distinct groups of synapses can generate EPSC components with different latencies. The presynaptic lamina I neuron can additionally evoke a polysynaptic response by efficiently exciting an intercalated (intact) neuron.

Both CP- and CI-AMPARs are involved in synaptic transmission

Effects of the CP-AMPAR blocker IEM 1460 (20 μm) and the AMPA/kainate receptor blocker CNQX (10 μm) were studied in connections with postsynaptic ALT-PNs (Fig. 3D). IEM 1460 reduced the EPSC amplitude by 85.4 ± 4.2% (n = 4; range 73 to 90%), while CNQX showed a complete block (n = 5).

During the application of the blockers (shown for IEM 1460; Fig. 3D, left), the EPSC components were suppressed in a manner consistent with their classification as mono- and polysynaptic. The monosynaptic EPSCs were progressively reduced with the receptor block. In contrast, the polysynaptic component (indicated by an asterisk) disappeared abruptly at the stage of a partial receptor block (judged from the reduction of the monosynaptic EPSCs). Thus, the partial block was already sufficient to reduce depolarization and prevent spike firing in the intercalated neuron (n = 4).

Input to an ALT-PN from a local-circuit neuron or another ALT-PN

In 2 of 11 pairs, we also obtained a detailed labelling of the presynaptic neuron. In the pair shown in Fig. 4, the monosynaptic EPSC consisted of at least two components which could be evoked either separately or together (Fig. 4A). In the 2-D reconstruction, the major axon of the postsynaptic ALT-PN could be followed until the rostral end of the isolated spinal cord (Fig. 4B). The presynaptic cell was a lamina I local-circuit neuron (multipolar type-IIa) with its axon occupying the SDH of one entire segment (Fig. 4B–C). Three close appositions between the presynaptic axon and the postsynaptic dendrites (putative synaptic contacts) were revealed (Fig. 4D). Two of them were formed by the same presynaptic axon branch (right group), while the third one originated from another branch (left). In another connection (Fig. 5), a simple EPSC corresponding to activation of one synapse was recorded; both cells were ALT-PNs.

Latency and efficacy of inputs

The mean latencies of individual monosynaptic components of composite EPSCs were distributed between 1.5 and 12.4 ms (mean, 4.31 ± 0.55 ms; 21 components analysed). The latencies of the initial monosynaptic components were distributed between 1.5 and 5.0 ms (mean, 2.75 ± 0.31 ms; n = 11).

The connection with three monosynaptic contacts (analysed in Fig. 3) showed EPSC components with latencies of 2, 3 and 6 ms (Fig. 6A). In the current-clamp mode, release in all three synapses increased the membrane depolarization; however, the composite EPSP did not reach the firing threshold in the postsynaptic ALT-PN.

Figure 6. Latency and efficacy of monosynaptic inputs.

Figure 6

Open in a new tab

A, a composite EPSC with three components showing latencies of 2, 3 and 6 ms (indicated by arrowheads). The analysis of the components was done in Fig. 3. In the current-clamp mode, simultaneous activation of all components (uppermost of 3 consecutive traces) increased the membrane depolarization which, however, remained subthreshold. B, EPSC evoked in the connection with the longest latency (5 consecutive traces). The histogram shows the distribution of the EPSC latencies measured in 47 episodes. Bin width, 0.25 ms. The mean latency was 12.4 ± 0.1 ms. For 42 measurements, the latency variation was within 1 ms (a grey bar), and for all 47 measurements within 2 ms. This long-latency input evoked spike firing in each of 6 consecutive stimulations. This connection also showed a shorter-latency monosynaptic component which, however, was not activated in the episodes analysed. In current-clamp, dotted lines indicate a preset potential of −70 mV (note different amplifications in A and B).

The input with the longest mean EPSC latency of 12.4 ms was observed in a connection shown in Fig. 6B. We did a detailed analysis of the latency variation for this EPSC. The histogram constructed for 47 episodes, revealed a low latency variation typical for a monosynaptic connection. In the current-clamp mode, this input was suprathreshold and EPSPs activated one-to-three spikes in each of six consecutive stimulations.

The EPSP efficacy was tested in nine connections. The postsynaptic ALT-PN was current-clamped at −70 mV. In four pairs, the evoked EPSPs were subthreshold and had maximum amplitudes of 4–11 mV (mean 8.0 ± 1.5 mV; n = 4).

In the remaining five pairs, 59.6 ± 13.4% of the evoked EPSPs elicited spikes (range 23% to 100%; 38 EPSPs analysed). In 2 of these 5 pairs, EPSPs evoked multiple spikes (Fig. 6B). Those EPSPs which did not elicit spikes in these five connections showed maximum amplitudes of 6–13 mV.

Thus, release from one lamina I neuron can substantially depolarize and, in many cases, excite an ALT-PN. The monosynaptic responses can have a long latency.

In the following experiments we studied whether such latencies could be explained by the conduction times in pre- and postsynaptic neurons. We assumed that synaptic release is fast (<1 ms) and, therefore, major transmission delay can be caused by the electrotonic EPSC propagation in the dendrites of a large ALT-PN and/or by the active spike propagation in the axon of the presynaptic neuron. To estimate maximum delays, we did computer simulations using the models based on detailed 3-D reconstructions of the dendritic and axonal branches of a representative ALT-PN and a large local-circuit neuron.

EPSC propagation time in the postsynaptic ALT-PN

The reconstructed ALT-PN had four major dendrites (Fig. 7A and B). The highest order of the dendritic branches was 14 and the maximum dendritic path length was 804 μm (Fig. 7C). Eight uniform synapses were inserted in different locations along the somatodendritic domain of the model neuron (1–8) to simulate somatic recordings of the corresponding EPSCs (Fig. 7D and E). The somatic synapse (labelled 1 on the figure) evoked an EPSC with a short delay of 0.1 ms. The longest delay of 3.9 ms was obtained for the most distal synapse (2) in the large dorsomedial major dendrite (green). In the dendritic branch of the highest order, the most distal synapse (3) showed a delay of 1.9 ms. In the large ventrolateral major dendrite (blue), the maximum delay was 2.7 ms (synapse 4). In shorter dendrites, dorsomedial (purple) and ventrolateral (orange), the longest EPSC delays were 1.1 ms (synapse 5) and 0.8 ms (synapse 6), respectively.

We also tested whether two spatially close synapses, located, however, in different major dendrites may evoke EPSCs with substantially different propagation times. For synapses 7 and 8 (grey circle), in large dorsomedial (green) and smaller ventrolateral (orange) major dendrites, the EPSC delays were 3.4 and 0.8 ms, respectively (Fig. 7E). Thus, simultaneous release in two spatially close synapses can evoke EPSCs reaching the soma with substantially different delays depending on the electrotonic lengths of the dendritic pathways.

These simulations have shown that the electrotonic propagation of an EPSC in the dendrites of the postsynaptic ALT-PN can take up to 4 ms but, alone, cannot explain the longest latencies observed in our experiments. Therefore, the spike propagation time in the axon of a presynaptic neuron was also estimated.

Conduction time in the axon of a local-circuit neuron

We reconstructed a lamina I local-circuit neuron with its axon covering the SDH area of three segments (Fig. 8A), to model possible long conduction times. The axon originated from one of the primary dendrites (Fig. 8B) and its length to the first branching point was 228 μm. The axon diameter was 2 μm in the 19 μm-long initial part, but then rapidly decreased to less than 1 μm. The maximum axon path length was 3725 μm (Fig. 8C) and the thinnest branches had a diameter of 0.2 μm.

When a 1 ms current pulse was applied to the soma in the model, the spike was initiated in the primary axon, about 20 μm away from its origin (point 0). It took 0.65 ms for the spike to reach the first branching point (1) (Fig. 8C and D).

We did simulations for several terminal boutons which were within the range of the dendritic field of a typical ALT-PN with the soma located closely to the soma of the reconstructed neuron. Terminal boutons 2 and 3 (Fig. 8B) located 64 μm and 118 μm away from the soma (3-D distances) had substantially longer axon paths of 646 μm and 922 μm, respectively (Fig. 8C). The spike propagation times to these boutons, located on the same principal branch, were 2.7 and 4.5 ms, respectively (Fig. 8D). For the terminal bouton 4 (another axon branch) located 436 μm away from the soma (3-D distance) and connected via the axon path of 952 μm, the spike propagation time was 4.15 ms. For bouton 5, in a dorsomedial position (3-D distance, 585 μm; the axon path length, 2076 μm), the conduction time was 11 ms.

We also did simulations for two terminal boutons located rostrally (6) and caudally (7) beyond the area of a possible contacts with the mediolaterally oriented dendrites of a neighbouring (somata separated by <80 μm) ALT-PN. The spike propagation times were 9.2 ms and 10.7 ms, respectively. The longest propagation time (17.4 ms) was obtained for the most remote labelled point (8) of the axon (Fig. 8C and D).

Thus, a lamina I local-circuit neuron can form synapses with dendrites of neighbouring neurons via long axonal pathways. The conduction in the narrow and highly branched axon may last for about 10 ms. For the connections with more distant neurons, the conduction time may be even longer.

Discussion

We used the isolated spinal cord preparation to record from large lamina I neurons with preserved dendritic trees and axonal arborizations. Our results show that ALT-PNs receive excitatory inputs from other lamina I neurons both directly and indirectly, via intercalated neurons. The axon of a presynaptic neuron forms multiple functional synapses on the dendrites of an ALT-PN and transmitter release activates both CP- and CI-AMPARs. An ALT-PN can receive input from a local-circuit neuron as well as from another ALT-PN. Transmission between lamina I neurons can have a long latency, mostly caused by the spike propagation in the axon of the presynaptic neuron. An input from a lamina I excitatory neuron can be sufficiently strong to evoke spikes in an ALT-PN.

Large lamina I neurons

In our sample from the lateral half of the SDH, 54% of large mediolaterally oriented neurons were identified as ALT-PNs (Szucs et al. 2010). The remaining neurons had either a highly branched local axon (like the presynaptic neuron in Fig. 4 and the neuron in Fig. 8) or projected via pathways other than the ALT. In comparison with smaller SG and lamina I neurons (Santos et al. 2007, 2009; Pinto et al. 2010), the ALT-PNs had lower RIN but a similar slowest membrane time constant, which could be explained by a larger area of the neuronal membrane with a similar density of a K+ leak conductance. It should be noted that our experiments were focused on large laterally located neurons and, therefore, smaller or medially located ALT-PNs may have different properties.

Organization of synaptic connections in lamina I

Our experiments revealed synaptic connections between large lamina I neurons. The EPSC latencies were distributed in a broad range implying that, in some cases, the signal propagated via long axodendritic pathways.

The postsynaptic delay is caused by the electrotonic EPSC propagation in the dendrites of an ALT-PN. It can be less than 1 ms for synapses in short dendrites or in the proximal parts of long dendrites. However, the propagation may take about 4 ms for distal synapses in long dendrites. This maximum delay was similar to our estimation for SG neurons (Santos et al. 2009). In an ALT-PN, different major dendrites may have some overlapping branches. In this case, two spatially close synapses located, however, in different dendritic branches may generate EPSCs with substantially different latencies.

The spike propagation time in the unmyelinated axon of the presynaptic neuron contributes substantially to the EPSC latency. Lamina I local-circuit neurons have extensive and dense axonal arbours and can synapse on ALT-PNs. The axon path can be several times longer than the 3-D distance between the terminal bouton and the soma of the presynaptic neuron. Furthermore, the small diameter of the axon (0.2–1 μm) lowers its conduction velocity, thus increasing the spike propagation time. The architecture of the axon of a local-circuit neuron predicts the spike conduction times of 1–11 ms for a postsynaptic ALT-PN with a closely located soma, and longer times for a more distant neuron. In terms of these findings, shorter EPSC latencies reported by Santos et al. (2009) for the SG excitatory interneurons can imply their short axon paths. It is also possible that in that study, using spinal cord slices, the long axodendritic paths were truncated.

In the majority of the cases, an axon of a presynaptic neuron forms multiple synapses on an ALT-PN, which may be a general principle of organization of neuronal connections in the SDH (Santos et al. 2009). The inputs from some synapses can be only distinguished on the basis of fast transitions on the EPSC rising phase, whereas others are better separated in time. Each component of a composite EPSC may represent a synapse (or a group of synapses) with its specific dendritic location and presynaptic axon path. Multiple monosynaptic contacts between lamina I neurons may provide a basis for the functional plasticity related to the activation of silent synapses (Santos et al. 2009).

In addition to direct contacts, lamina I neurons are also connected to ALT-PNs via intercalated neurons. The polysynaptic inputs show variable latencies consistent with the variation of the spike initiation time in the intercalated neuron (Santos et al. 2009). Polysynaptic inputs can be sufficiently strong to increase the excitation of an ALT-PN. Thus, the recruiting of intercalated neurons may be an additional mechanism increasing efficacy of neuronal connections in the SDH.

We have found that, in many cases, monosynaptic inputs from one neuron can be sufficiently strong to evoke spikes in an ALT-PN. Inputs from different groups of synapses can be integrated to increase the postsynaptic depolarization. Furthermore, monosynaptic connections via different axodendritic paths may broaden the temporal pattern of synaptic input from one neuron. It should be noted that the presence of numerous polysynaptic inputs implied an excitation of an intercalated (intact) neuron. This also supports the idea that release from one SDH neuron can be sufficient to evoke a spike in another neuron (Santos et al. 2009).

We show that ALT-PNs communicate with other lamina I neurons via synapses equipped with CP- and CI-AMPARs. Expression of GluR2-lacking CP-AMPARs and GluR2-containing CI-AMPARs in the SDH was shown by immunocytochemistry and EM studies (Nagy et al. 2004; Larsson & Broman, 2006; Antal et al. 2008; Polgar et al. 2008). Physiological activation of CP-AMPARs by transmitter release has recently been reported for synapses formed by the SG excitatory interneurons (Santos et al. 2009). CP-AMPARs may provide Ca2+ influx and underlie different forms of synaptic plasticity in the SDH (Santos et al. 2009) and hippocampus (Oren et al. 2009). The CP-AMPAR-dependent plasticity in synapses formed by the interneurons (Santos et al. 2009) may be an alternative to the NMDA-receptor-dependent plasticity in the synapses of primary afferents (Ikeda et al. 2003). The contribution of CP-AMPARs to transmission in the SDH may further increase under chronic pain conditions (Katano et al. 2008; Larsson & Broman, 2008; Vikman et al. 2008).

Thus, multiple mono- and polysynaptic inputs from one lamina I neuron can increase its efficacy of exciting an ALT-PN and provide the basis for different forms of functional plasticity.

Lamina I and nociceptive processing

Lamina I is traditionally considered as a relay station conveying thin primary afferent input to supraspinal centres (Molander & Grant, 1995; Lima, 1998; Todd & Koerber, 2006). However, recent studies show that lamina I is an interconnected layer involved in intralaminar, interlaminar and intersegmental spinal processing. Individual lamina I neurons can integrate monosynaptic Aδ- and C-afferent-inputs from several segmental dorsal roots (Pinto et al. 2010). Highly branched axons of lamina I local-circuit neurons can interconnect large SDH areas of several segments (Szucs et al. 2010; and this study). In addition, the projecting axons of the ALT-PNs give rise to local collaterals connecting lamina I with the dorsal, ventral and lateral spinal cord regions of several neighbouring segments (Szucs et al. 2010). As we show here, lamina I neurons are interconnected through groups of functional synapses and intercalated neurons. The intralaminar neuronal connectivity has also been shown by the laser-scanning photostimulation technique (Kato et al. 2009). An overall prevalence of local-circuit neurons over the PNs in lamina I (Bice & Beal, 1997; Spike et al. 2003; Todd & Koerber, 2006) can further emphasize the functional importance of the intralaminar processing. Therefore, future investigations of the neuronal network organization in lamina I will be necessary for the better understanding of signal processing in this spinal nociceptive projection area.

Acknowledgments

The work was supported by the grant from the Portuguese Foundation for Science and Technology funded by COMPETE and FEDER. We thank Miklós Antal for providing the Neurolucida system.

Glossary

Abbreviations

ALT

anterolateral tract

CI-AMPAR

Ca2+-impermeable AMPA receptor

CP-AMPAR

Ca2+-permeable AMPA receptor

PN

projection neuron

SDH

superficial dorsal horn

SG

substantia gelatinosa (lamina II)

Author contributions

The experiments were performed in IBMC (University of Porto). All authors have contributed to (1) conception and design of the experiments, (2) collection, analysis and interpretation of data, and (3) drafting the article or revising it critically for important intellectual content. All authors approved the final version of the manuscript.

References

  1. Antal M, Fukazawa Y, Eordogh M, Muszil D, Molnar E, Itakura M, Takahashi M, Shigemoto R. Numbers, densities, and colocalization of AMPA- and NMDA-type glutamate receptors at individual synapses in the superficial spinal dorsal horn of rats. J Neurosci. 2008;28:9692–9701. doi: 10.1523/JNEUROSCI.1551-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bice TN, Beal JA. Quantitative and neurogenic analysis of the total population and subpopulations of neurons defined by axon projection in the superficial dorsal horn of the rat lumbar spinal cord. J Comp Neurol. 1997;388:550–564. doi: 10.1002/(sici)1096-9861(19971201)388:4<550::aid-cne4>3.0.co;2-1. [DOI] [PubMed] [Google Scholar]
  3. Cajal RyS. Histologie du systeme nerveux de l'homme et des vertebres. Paris: A. Maloine; 1909. [Google Scholar]
  4. Drummond GB. Reporting ethical matters in The Journal of Physiology: standards and advice. J Physiol. 2009;587:713–719. doi: 10.1113/jphysiol.2008.167387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Graham BA, Brichta AM, Callister RJ. Moving from an averaged to specific view of spinal cord pain processing circuits. J Neurophysiol. 2007;98:1057–1063. doi: 10.1152/jn.00581.2007. [DOI] [PubMed] [Google Scholar]
  6. Grudt TJ, Perl ER. Correlations between neuronal morphology and electrophysiological features in the rodent superficial dorsal horn. J Physiol. 2002;540:189–207. doi: 10.1113/jphysiol.2001.012890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Hines M. NEURON – A program for stimulation of nerve equation. In: Eeckman F, editor. Neural Systems: Analysis and Modeling. Boston: Kluwer Academic Publishers; 1993. pp. 127–136. [Google Scholar]
  8. Hines ML, Carnevale NT. The NEURON simulation environment. Neural Comput. 1997;9:1179–1209. doi: 10.1162/neco.1997.9.6.1179. [DOI] [PubMed] [Google Scholar]
  9. Ikeda H, Heinke B, Ruscheweyh R, Sandkuhler J. Synaptic plasticity in spinal lamina I projection neurons that mediate hyperalgesia. Science. 2003;299:1237–1240. doi: 10.1126/science.1080659. [DOI] [PubMed] [Google Scholar]
  10. Katano T, Furue H, Okuda-Ashitaka E, Tagaya M, Watanabe M, Yoshimura M, Ito S. N-ethylmaleimide-sensitive fusion protein (NSF) is involved in central sensitization in the spinal cord through GluR2 subunit composition switch after inflammation. Eur J Neurosci. 2008;27:3161–3170. doi: 10.1111/j.1460-9568.2008.06293.x. [DOI] [PubMed] [Google Scholar]
  11. Kato G, Kawasaki Y, Koga K, Uta D, Kosugi M, Yasaka T, Yoshimura M, Ji RR, Strassman AM. Organization of intralaminar and translaminar neuronal connectivity in the superficial spinal dorsal horn. J Neurosci. 2009;29:5088–5099. doi: 10.1523/JNEUROSCI.6175-08.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Larsson M, Broman J. Pathway-specific bidirectional regulation of Ca2+/calmodulin-dependent protein kinase II at spinal nociceptive synapses after acute noxious stimulation. J Neurosci. 2006;26:4198–4205. doi: 10.1523/JNEUROSCI.0352-06.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Larsson M, Broman J. Translocation of GluR1-containing AMPA receptors to a spinal nociceptive synapse during acute noxious stimulation. J Neurosci. 2008;28:7084–7090. doi: 10.1523/JNEUROSCI.5749-07.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lima D. Anatomical basis for the dynamic processing of nociceptive input. Eur J Pain. 1998;2:195–202. doi: 10.1016/s1090-3801(98)90015-5. [DOI] [PubMed] [Google Scholar]
  15. Lima D, Almeida A. The medullary dorsal reticular nucleus as a pronociceptive centre of the pain control system. Prog Neurobiol. 2002;66:81–108. doi: 10.1016/s0301-0082(01)00025-9. [DOI] [PubMed] [Google Scholar]
  16. Lima D, Coimbra A. A Golgi study of the neuronal population of the marginal zone (lamina I) of the rat spinal cord. J Comp Neurol. 1986;244:53–71. doi: 10.1002/cne.902440105. [DOI] [PubMed] [Google Scholar]
  17. Lu Y, Perl ER. Modular organization of excitatory circuits between neurons of the spinal superficial dorsal horn (laminae I and II) J Neurosci. 2005;25:3900–3907. doi: 10.1523/JNEUROSCI.0102-05.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Melnick IV, Santos SF, Safronov BV. Mechanism of spike frequency adaptation in substantia gelatinosa neurones of rat. J Physiol. 2004a;559:383–395. doi: 10.1113/jphysiol.2004.066415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Melnick IV, Santos SF, Szokol K, Szucs P, Safronov BV. Ionic basis of tonic firing in spinal substantia gelatinosa neurons of rat. J Neurophysiol. 2004b;91:646–655. doi: 10.1152/jn.00883.2003. [DOI] [PubMed] [Google Scholar]
  20. Molander C, Grant G. Spinal cord cytoarchitecture. In: Paxinos G, editor. The Rat Nervous System. 2nd edn. San Diego: Academic Press; 1995. pp. 39–44. [Google Scholar]
  21. Nagy GG, Al-Ayyan M, Andrew D, Fukaya M, Watanabe M, Todd AJ. Widespread expression of the AMPA receptor GluR2 subunit at glutamatergic synapses in the rat spinal cord and phosphorylation of GluR1 in response to noxious stimulation revealed with an antigen-unmasking method. J Neurosci. 2004;24:5766–5777. doi: 10.1523/JNEUROSCI.1237-04.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Oren I, Nissen W, Kullmann DM, Somogyi P, Lamsa KP. Role of ionotropic glutamate receptors in long-term potentiation in rat hippocampal CA1 oriens-lacunosum moleculare interneurons. J Neurosci. 2009;29:939–950. doi: 10.1523/JNEUROSCI.3251-08.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Pinto V, Derkach VA, Safronov BV. Role of TTX-sensitive and TTX-resistant sodium channels in Aδ- and C-fiber conduction and synaptic transmission. J Neurophysiol. 2008a;99:617–628. doi: 10.1152/jn.00944.2007. [DOI] [PubMed] [Google Scholar]
  24. Pinto V, Szucs P, Derkach VA, Safronov BV. Monosynaptic convergence of C- and Aδ-afferent fibres from different segmental dorsal roots on to single substantia gelatinosa neurones in the rat spinal cord. J Physiol. 2008b;586:4165–4177. doi: 10.1113/jphysiol.2008.154898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Pinto V, Szucs P, Lima D, Safronov BV. Multisegmental Aδ- and C-fiber input to neurons in lamina I and the lateral spinal nucleus. J Neurosci. 2010;30:2384–2395. doi: 10.1523/JNEUROSCI.3445-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Polgar E, Watanabe M, Hartmann B, Grant SG, Todd AJ. Expression of AMPA receptor subunits at synapses in laminae I–III of the rodent spinal dorsal horn. Mol Pain. 2008;4:5. doi: 10.1186/1744-8069-4-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Safronov BV. Spatial distribution of Na+ and K+ channels in spinal dorsal horn neurones: role of the soma, axon and dendrites in spike generation. Prog Neurobiol. 1999;59:217–241. doi: 10.1016/s0301-0082(98)00051-3. [DOI] [PubMed] [Google Scholar]
  28. Safronov BV, Pinto V, Derkach VA. High-resolution single-cell imaging for functional studies in the whole brain and spinal cord and thick tissue blocks using light-emitting diode illumination. J Neurosci Methods. 2007;164:292–298. doi: 10.1016/j.jneumeth.2007.05.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Safronov BV, Wolff M, Vogel W. Functional distribution of three types of Na+ channel on soma and processes of dorsal horn neurones of rat spinal cord. J Physiol. 1997;503:371–385. doi: 10.1111/j.1469-7793.1997.371bh.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Santos SF, Luz LL, Szucs P, Lima D, Derkach VA, Safronov BV. Transmission efficacy and plasticity in glutamatergic synapses formed by excitatory interneurons of the substantia gelatinosa in the rat spinal cord. PLoS One. 2009;4:e8047. doi: 10.1371/journal.pone.0008047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Santos SF, Melnick IV, Safronov BV. Selective postsynaptic inhibition of tonic-firing neurons in substantia gelatinosa by mu-opioid agonist. Anesthesiology. 2004;101:1177–1183. doi: 10.1097/00000542-200411000-00018. [DOI] [PubMed] [Google Scholar]
  32. Santos SF, Rebelo S, Derkach VA, Safronov BV. Excitatory interneurons dominate sensory processing in the spinal substantia gelatinosa of rat. J Physiol. 2007;581:241–254. doi: 10.1113/jphysiol.2006.126912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Spike RC, Puskar Z, Andrew D, Todd AJ. A quantitative and morphological study of projection neurons in lamina I of the rat lumbar spinal cord. Eur J Neurosci. 2003;18:2433–2448. doi: 10.1046/j.1460-9568.2003.02981.x. [DOI] [PubMed] [Google Scholar]
  34. Szucs P, Luz LL, Lima D, Safronov BV. Local axon collaterals of lamina I projection neurons in the spinal cord of young rats. J Comp Neurol. 2010;518:2645–2665. doi: 10.1002/cne.22391. [DOI] [PubMed] [Google Scholar]
  35. Szucs P, Pinto V, Safronov BV. Advanced technique of infrared LED imaging of unstained cells and intracellular structures in isolated spinal cord, brainstem, ganglia and cerebellum. J Neurosci Methods. 2009;177:369–380. doi: 10.1016/j.jneumeth.2008.10.024. [DOI] [PubMed] [Google Scholar]
  36. Todd AJ, Koerber R. Neuroanatomical substrates of spinal nociception. In: McMahon SB, Koltzenburg M, editors. Wall and Melzack's Textbook of Pain. 5th edn. Edinburgh, UK: Churchill Livingstone; 2006. pp. 73–90. [Google Scholar]
  37. Todd AJ, McGill MM, Shehab SA. Neurokinin 1 receptor expression by neurons in laminae I, III and IV of the rat spinal dorsal horn that project to the brainstem. Eur J Neurosci. 2000;12:689–700. doi: 10.1046/j.1460-9568.2000.00950.x. [DOI] [PubMed] [Google Scholar]
  38. Vikman KS, Rycroft BK, Christie MJ. Switch to Ca2+-permeable AMPA and reduced NR2B NMDA receptor-mediated neurotransmission at dorsal horn nociceptive synapses during inflammatory pain in the rat. J Physiol. 2008;586:515–527. doi: 10.1113/jphysiol.2007.145581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Waldeyer W. Das Gorilla-Rückenmark. Berlin: Königliche Akademie der Wissenschaften; 1889. [Google Scholar]
  40. Willis WD, Coggeshall RE. Sensory Mechanisms of the Spinal Cord. New York: John Wiley & Sons; 1991. [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES