| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Physiology and Neuroscience Program, Michigan State University, E. Lansing, Michigan
Submitted 27 August 2007; accepted in final form 7 January 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Willis and Coggeshall 2004). This region also contains neurons that participate in local synaptic circuits (Mannen and Sugiura 1976; Scheibel and Scheibel 1968; Schneider 1992) and mediate signal flow from the receptors to segmental reflexes or ascending pathways to the brain and brain stem. Past electrophysiological studies indicate that deep dorsal horn neurons display diverse electrophysiological properties (Hochman et al. 1997; Jiang et al. 1995) and responsiveness to mechanical skin stimulation (Brown 1969; Heavner and de Jong 1973; Menétrey et al. 1977). More recent studies suggest that deep dorsal horn neurons with slowly and rapidly adapting firing patterns are differentially sensitive to rate of membrane depolarization (Schneider 2003) and mechanical stimulation of the skin (Schneider 2005). These findings were consistent with a long-standing hypothesis that mechanosensory inputs from hairy skin are distributed nonrandomly onto dorsal horn neurons (Tapper et al. 1973) and a later conclusion that dorsal horn networks activated by slowly and rapidly adapting mechanosensory afferents differ in their ability to integrate temporally patterned inputs (Koerber et al. 1991). However, the dorsal horn neuropil is a region of complex synaptic interactions that mediates convergence from many sensory inputs and the extent to which any connection specificity between mechanosensory afferent systems and central neurons is preserved by the intrinsic dorsal horn circuitry is unknown.
To gain insight into transformation and integration of mechanosensory information within the deep dorsal horn, we used paired whole cell electrophysiological recordings to investigate the properties and organization of synaptic connections between neurons in Rexed's laminae (L) III and IV, where the central axons of afferents innervating hair and tactile receptors terminate (Brown 1981). Our experiments asked whether firing patterns and membrane properties of the pre- and postsynaptic neurons play roles in determining organization of local synaptic connectivity. We found an unusually high incidence of connections by inhibitory interneurons and that connections involved multiple types of neurons in many combinations. Overall, reliability of synaptic transmission at these connections was relatively low and characterized by frequent failure of a presynaptic action potential to elicit a postsynaptic response. However, synaptic efficacy was found to be related to the firing properties of the presynaptic neuron. Evidence is presented for possible fundamental differences between the organization of local synaptic connectivity in LIII–IV and superficial dorsal horn (LI–II) where excitatory and inhibitory connections between a few functional types of neurons may give rise to recurring stereotyped circuitry (Lu and Perl 2003, 2005).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
All protocols involving live animals conformed to guidelines published by the National Institutes of Health and were approved by the Institutional Animal Care and Use Committee of Michigan State University. Dissection procedures and preparation of spinal slices were similar to those reported previously (Schneider 2003; Schneider et al. 1995). Nine- to 30-day-old hamsters of both sexes were anesthetized with urethan (1.5 mg/g ip) and chilled to a body temperature of 25°C. After removing the vertebral column, a 4-mm block of spinal cord (T10–L5) was dissected in an ice-cold solution containing (mM) 216 sucrose, 2.5 KCl, 0.25 CaCl2, 10 MgCl2, 1.25 NaH2PO4, 26 NaHCO3, and 25 glucose (290–300 mOsm/l) equilibrated with 95% O2-5% CO2. Slices of spinal cord 250- to 300-µm thick were prepared on a vibrating tissue slicer (Vibratome 1000) in the sagittal plane to preserve continuity of rostrocaudally oriented axon terminations characteristic of neurons in the deeper spinal laminae (Schneider 1992; Schneider et al. 1995). Slices were transferred to an artificial cerebrospinal fluid (ACSF) containing (mM) 125 NaCl, 2.5 KCl, 2.5 CaCl2, 1.5 MgCl2, 1.25 NaH2PO4, 26 NaHCO3, and 10 glucose (pH 7.35–7.45, 290–300 mOsm/l) equilibrated with 95% O2-5% CO2 at room temperature and incubated 1 h. Individual slices were transferred to a glass-bottom recording chamber (1.5 ml) mounted on a fixed stage, upright microscope and held in place with a nylon monofilament ladder attached to U-shaped platinum wire. The chamber was perfused with the same artificial cerebrospinal fluid (ACSF) solution (6–7 ml/min) maintained at 25–27°C for electrophysiological recording.
Cell identification, electrophysiology, and data analysis
Recording pipettes (6–8 M
) were fabricated from glass capillary tubing (TW150F-4, WP Instruments, Sarasota, FL) and filled with an internal solution containing (mM) 100 K gluconate, 20 KCl, 4 Mg-ATP, 10 phosphocreatine, 0.3 Li2+-GTP, and 10 HEPES (pH 7.3, 300 mosM/l). The calcium chelator ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) was omitted from the internal solution to avoid buffering intracellular calcium and subsequent reduction in synaptic transmitter release (Ohana and Sakmann 1998; Rozov et al. 2001). For most experiments, the pipette solution also contained 2% biocytin in an attempt to anatomically identify recorded neurons. Signals in voltage and current modes were amplified by AxoPatch 1D amplifiers (Axon Instruments, Foster City, CA), low-pass filtered (DC–5 kHz), digitized with a Digidata 1320A data-acquisition system (Axon Instruments), and stored on hard disk for off-line analyses (pClamp 8, Axon Instruments).
Individual neurons were visualized with infrared differential contrast (IR-DIC) microscopy using a water-immersion objective (Zeiss x40 Plan, 0.75 NA, 1.5 mm working distance). Under brightfield illumination, Rexed's laminae (L) III and IV were recognized as the region ventral to a translucent band corresponding to LII, the substantia gelatinosa (Fig. 1 A). (Recordings were not sought from deeper laminae because access to cells in these areas was obscured by myelinated fibers.) Neuronal somata near the slice surface (within 
10–20 µm or 1 cell body diameter) and separated by no more than 160 µm (constrained by the optical viewing field) were selected for recording. Recording pipettes were brought into proximity of two candidate somata (Fig. 1B). One cell was contacted by a recording pipette, establishing a high-resistance seal between tip and cell membrane. Whole cell recording was initiated by applying negative pressure to the pipette, rupturing the membrane patch, as signaled by abrupt appearance of membrane current responses to 2-mV pulses under voltage clamp (Vh = –60 mV). The process was repeated for the second cell. Amplifiers were later switched to current-clamp mode for most subsequent analyses. Recordings were attempted in one or two slices per experiment and only one pair of neurons was recorded per slice. No effort was made to select neurons for recording based on size of the cell body imaged in situ.
|
.
Neuronal firing properties and steady-state current-voltage relations were examined by injecting pulses of constant current through the recording pipettes (Schneider 2003). Instantaneous firing frequency was computed from measurements of interspike intervals. Small hyperpolarizing voltage changes in response to anodal current were used to measure membrane time constant (
m) and neuronal input resistance (Rin). The slowest time constant of the voltage response was used as an estimate of m. Measurements of the amplitude and time course of single-fiber synaptic responses recorded in voltage and current mode were made from individual traces or averages computed from 25 to 100 individual traces, excluding failures. Rise time was defined as the time taken to reach from 10 to 90% of the peak amplitude. Latency was defined as the time from the peak of the presynaptic action potential to 5% of the postsynaptic potential amplitude (Markram et al. 1997). Half-maximum width was defined as the time between points on the trace at which the postsynaptic response is at half of its maximum value. Time constants of synaptic response decay were estimated by fitting the response to a standard exponential of the form f(t) = 
Ai e–t/ + C. Failure rate (% failures) was calculated from trials of 25–100 stimuli and used to compare reliability of synaptic connections.
Statistical analyses
Numerical data are presented as means ± SD. Statistical differences between data groups were analyzed with t-test (2-tailed probability), 
2 (1-sample), or one-way ANOVA with Neuman-Keul posttest comparisons (GBStat, Dynamic Microsystems, Silver Spring, MD).
Pharmacology
-Amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) glutamatergic receptors were blocked by adding 10 µM 6-cyano-7 nitroquinoxaline-2,3-dione (CNQX; Tocris or Sigma) to the bathing medium. Transmission involving 
-aminobutyric acid receptors (GABAARs) and glycine receptors was blocked by 20 µM bicuculline (Sigma) and 1 µM strychnine (Sigma), respectively.
Histological procedures
After recordings, slices were fixed overnight in phosphate-buffered 4% paraformaldehyde/4% sucrose (pH 7.4) and washed 12–48 h (4°C) in 0.1 M phosphate-buffered 30% sucrose (pH 7.4). Tissue was reacted en bloc with ABC complex (Vector Labs, PK-4000) according to standard procedures (Metz et al. 1989), mounted on gelatin-coated glass slides, dehydrated, cleared, and fitted with a coverslip. Labeled neurons were examined under brightfield illumination at x20, x40, and x100 to assess laminar location of somata and classify recorded cells based on previously described morphological criteria (Schneider 1992).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
m were, respectively, 498 ± 355 M and 36 ± 25 ms. Cell capacitance (m/Rin) was 81 ± 35 pF. These values are within ranges reported previously for LIII–IV neurons in this preparation (Schneider 2003). Action potentials in one neuron generated time-locked, short-latency depolarizing (excitatory) or hyperpolarizing (inhibitory) membrane potentials in the other in 32 cell pairs (Fig. 1, C–F, and Table 1), giving an overall connection probability of 0.3 for these experiments. Cell pairs linked by inhibitory connections (n = 22) were encountered twice as often as those with excitatory connections (n = 10) and represented 69% of the sample. The average resting membrane potential for linked neurons was not significantly different from cells without connections (–64 ± 7 vs. –62 ± 8 mV, P = 0.31). Similarly there was no difference between the membrane potentials of pre- and postsynaptic neurons linked by excitatory or inhibitory connections (P = 0.51). Linkage between most cell pairs (91%, 29/32) was unidirectional. The probability of recording connected cell pairs was unrelated to age of animals from which slices were obtained (Fig. 1G) and the distance separating the cells (Fig. 1H). Moreover, there was no evidence that synaptically connected cells exhibited rostral-to-caudal or dorsal-to-ventral orientations (n = 17, P = 0.49).
|
Thirteen cell pairs were recovered after histochemical processing for biocytin labeling. Although intracellular staining was attempted in most experiments, labeling quality was highly variable—in many cases, only one neuron per pair was stained well enough to permit morphological identification. Three successfully labeled pairs evidenced synaptic connectivity and two of these are illustrated in Fig. 2.
|
; Schneider et al. 1995). Neurons were relatively small (average major and minor soma diameters, 17 and 13 µm) and were located within LIII/IV near the slice surface. Each had two to four tapered dendrites extending 200–400 µm and a thin, cylindrical axon that ramified in a tangle of fine, rostrocaudal fibers bearing numerous terminal and en passant varicosities (e.g., see Fig. 2, A and B). [Postsynaptic location of contacts could not be confirmed due to complexity of the intermingled processes (e.g., Fig. 2C).] Axon terminations projected 400–1,300 µm beyond the parent cells' dendritic field and were contained within the dorsal horn neuropil. Labeled processes showed considerable overlap (Fig. 2, A and B), independent of synaptic linkage. None of the cells gave rise to an axon that could be traced into the spinal white matter or ventromedially into lamina X. Thus the pattern of labeling would indicate that recordings were made from interneurons that participate in longitudinally oriented local circuits within LIII/IV. Amplitude and time course of single-axon PSPs
Examples of postsynaptic responses initiated by single presynaptic action potentials are shown in Fig. 3. Excitatory (E) postsynaptic potentials (PSPs) (Fig. 3A1) averaged 1.9 ± 2.5 mV (range, 0.5–8.5 mV). Mean rise time was 5.9 ± 6.8 ms (range, 2–23 ms). The average half-width was 31 ± 22 ms (range, 10–87 ms), and the average decay time constant was 31 ± 18 ms (range, 12–65 ms). Transmission at excitatory connections was associated with inward membrane currents (Fig. 3A2) with mean amplitude and decay time constant of 19 ± 11 pA and 3.7 ± 2.1 ms at Vh = –60 mV (n = 3) and never elicited action potentials from the postsynaptic neuron. IPSPs (Fig. 3B1) averaged –0.8 ± 0.7 mV (range, –0.1 to –3.0 mV) at resting membrane potentials around –60 mV. Mean rise time was 9.7 ± 6.9 ms (range, 2–25 ms), and average half-width and decay time constants were 31 ± 18 ms (range, 9–82 ms) and 39 ± 23 ms (range, 4–77 ms), respectively. Amplitude and half-width of single-axon PSPs were unrelated to postsynaptic cell Rin, m, or pipette access resistance. Small synaptic currents were recorded at five inhibitory connections, four of these were associated with outward currents (Fig. 3B2) and one with an inward current (not shown). Mean amplitude and decay time constants at a holding potential of –60 mV were 8 ± 3 pA and 8 ± 5 ms, respectively. Transmission at connections between LIII/IV neurons was temperature sensitive: increasing the bath temperature to 35°C (n = 2 connections) resulted in a 14–29% reduction in peak PSP amplitude and 19–41% decrease in half-width.
|
; Iansek and Redman 1973) and between pyramidal cells in the neocortex (Thomson et al. 1988). Bidirectional linkages
For three cell pairs (9%), action potentials in either cell evoked synaptic responses in the other, indicating bidirectional (reciprocal) connectivity (not shown). In all cases, connections were inhibitory. There was no evidence of electrical coupling and the distance separating reciprocally linked cells was not different from pairs evidencing unidirectional linkage (P = 0.64). The time course of IPSPs evoked at unidirectional and reciprocal linkages were similar (half-widths: 29 ± 20 vs. 32 ± 15 ms, P = 0.67). However, IPSP amplitude at bidirectional connections was –0.4 ± 0.1 mV (range, –0.2 to – 0.5 mV), less than half the amplitude of IPSPs evoked at unidirectional connections (–0.9 ± 0.7 mV; range, –0.1 to –2.4 mV) at resting potentials around –60 mV (P = 0.01). This suggests that synaptic coupling is much weaker between reciprocally connected neurons than between cells connected by unidirectional linkages.
Latency and amplitude fluctuations of single-axon PSPs
Latency and amplitude of individual PSPs varied from trial to trial (Fig. 4). The mean latency for EPSPs and IPSPs was not significantly different (3.0 ± 1.8 vs. 1.9 ± 1.7 ms, P = 0.15), and the amount of jitter was also similar at these connections. Latency histograms for EPSPs and IPSPs showed a single peak and could be fitted with a Gaussian function (Fig. 4D, 1 and 2). Histograms of PSP amplitude were also approximated by Gaussian functions with no difference between excitatory and inhibitory connections (Fig. 4E, 1 and 2). Overall there was a weak negative correlation (r = –0.08 to –0.47) between PSP latency and amplitude. The mean slope of the best-fit linear regression for eight cell pairs was –0.15 ms/mV (range, –0.05 to –0.35 ms/mV) with no significant difference between excitatory and inhibitory connections (P = 0.3).
|
The contributions of glutamate and GABA receptor activation to single-axon PSPs were examined in the presence of pharmacological blockade of either class of receptor. EPSPs were regularly and reversibly antagonized (4 of 4 connections) by the glutamate AMPA receptor antagonist CNQX (10 µM; Fig. 5 A, 1 and 2). Reduction in peak EPSP amplitude was nearly complete (93–100%) 5 min after drug application. The GABAA receptor antagonist bicuculline (20 µM) reversibly reduced peak IPSP amplitude (42 ± 16%; range 22–71%; n = 6) within 14 min of perfusion (Fig. 5B1). It should be noted that all cells tested showed at least a 30% bicuculline-resistant component. For one connection, application of strychnine (1 µM) subsequent to bicuculline caused an additional decrease in IPSP amplitude from 22 to 59% (Fig. 5B2). None of the bath-applied drugs had any effect on action potentials evoked in the presynaptic neuron (not shown).
|
Low background activity enabled detection of responses and failures at most connections (Fig. 6 A). Overall reliability of synaptic transmission between spinal LIII–IV neurons was much lower than transmission reported at connections between cortical neurons (Feldmeyer et al. 1999, 2002, 2005; Markram et al. 1997) and between thalamic relay cells (Gentet and Ulrich 2003). The apparent failure rate for PSPs at unidirectional connections averaged 44 ± 29% (range, 0–84%). Failures occurred in less than 15% of trials for one-third of the pairs (9/28) and exceeded 50% for half of the sample (14/28). The relatively high rate of transmission failure was not readily explained by low slice temperature (see also Allen and Stevens 1994) because elevating the bath temperature from 25 to 30°C did not alter transmission reliability (mean failure rate 46%, n = 3 connections). Failure rates for EPSPs and IPSPs at unidirectional connections were similar (36 ± 33% versus 48 ± 25%, P = 0.32; Fig. 6B). However, inhibitory transmission at bidirectional connections was almost twice as likely to fail (81 ± 2%, n = 3; P = 0.015). Failure rate was positively related to latency jitter (Fig. 6C) for IPSPs (P = 0.0052) but not for EPSPs (P = 0.35), suggesting that transmission along local inhibitory pathways is less secure than for excitatory pathways. Overall transmission at connections between LIII and IV neurons exhibited a decrease in failures with increasing PSP amplitude (Fig. 6D), suggesting that release probability may be the main determinant of PSP amplitude, similar to intracortical synapses (Feldmeyer et al. 1999; Markram et al. 1997). Note that one exception of 34 connections is shown in Fig. 6D (indicated by 
) where a large, long latency EPSP was associated with an unusually high failure rate.
|
Past studies showed that neurons in LIII/IV display substantial diversity in their discharge properties (Hochman et al. 1997; Jiang et al. 1995; Schneider 1992, 2003, 2005; Schneider et al. 1995). Therefore analyses were performed in an effort to determine whether synaptic connectivity is related to the firing patterns of the pre- and postsynaptic cells.
Depolarizing current injected through the recording pipettes activated rapidly adapting ("phasic") firing from 53% of cells (n = 100) (e.g., Figure 7, A2 and C2, cell 1) and slowly adapting ("tonic") discharge from 37% of neurons (n = 69; e.g., Fig. 7B2). Another 15% of cells (n = 20) responded to membrane depolarization with sustained excitation after a prolonged delay (e.g., Fig. 7C2, cell 2). Phasic cells were preferentially activated by fast current ramps (Fig. 7, A3 and C3, cell 1) similar to properties of LIII–IV neurons reported previously using this preparation (Schneider 2003, 2005). The percentage of linked cells sharing similar firing patterns (41%, 12 of 29 pairs) was not significantly different (P = 0.24) from unlinked cells (29%, 18 of 63 pairs). [Firing patterns were considered "similar" if they shared common spike frequency adaptation and timing characteristics (Fig. 7, A and B) and "dissimilar" if they did not (Fig. 7C).] Furthermore, whether connections between linked neurons were excitatory or inhibitory was also unrelated to neuronal firing pattern (P = 0.69). Three of the presynaptic neurons in cell pairs with excitatory connections had tonic firing patterns, five had phasic firing patterns, and one was delayed-firing (Table 1). Of the postsynaptic cells in these linkages, four were tonic, three were phasic, and two were delay. In 17 cell pairs with unidirectional inhibitory connections, 4 of the presynaptic cells had tonic patterns, 8 had phasic patterns, and 5 were delayed-firing. Four of the postsynaptic cells were tonic, and 13 were phasic. Thus neither linkage probability nor the type of synaptic connection could be predicted by the firing properties of the neurons.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Schneider and Walker 2007). Synaptic connections between LIII and IV neurons are predominantly unidirectional, and the majority of these are inhibitory. The finding that transmission efficacy at these linkages is correlated with intrinsic firing properties of the presynaptic cells may have important implications for integration of mechanosensory information carried by rapidly conducting myelinated afferents. Postsynaptic location of local inputs
Assuming a somatic location of the recording pipette and that PSPs originate from restricted loci on the dendritic membrane, rise time and half-width measurements can be used to approximate the postsynaptic distribution of axon terminals from which synaptic potentials originate. PSPs with shape indices near the origin in Fig. 3C are inferred to be activated from contacts closer to the pipette tip, consistent with locations on proximal dendrites. Data points up and to the right on the plot originate farther from the cell body, indicative of synapses at distal dendritic sites. The location of these inputs cannot be known with certainty without anatomical visualization of the contacts themselves. However, the shape indices of single-axon PSPs are consistent with our previous studies that found that axon terminals of LIII/IV interneurons target dendrites (Schneider et al. 1995). Moreover, activation of presynaptic impulses never elicited action potentials in the postsynaptic cell. Taken together, these results indicate that local circuit inputs to interneurons in the deep spinal dorsal horn are located some distance from the soma and spike-generating membrane of the initial segment. To trigger a suprathreshold response in the postsynaptic cell activation and summation of many excitatory inputs seem to be required.
Strength and reliability of single-axon PSPs
This study revealed that synaptic connections between lamina III and IV neurons are relatively weak, similar to linkages between superficial dorsal horn neurons (Lu and Perl 2003, 2005) and at synaptic connections in the neocortex (Deuchars et al. 1994; Markram et al. 1997; Tamás et al. 1997; Thomson et al. 1995) and hippocampus (Ali et al. 1998; Gulyás et al. 1993; Scharfman 1994). However, reliability of transmission at LIII/IV synapses is relatively low (average failure rate 44%). It has been reported that excitatory synapses in the thalamus (Gentet and Ulrich 2003) and neocortex (Feldmeyer et al. 1999, 2002, 2005, 2006; Markram et al. 1997) have failure rates in the range of 0–14%, indicative of connections with substantially greater reliability than those described here. The relatively high failure rates at LIII/IV connections are probably not due to lowered bath temperature (27°C) because elevating the temperature to a more physiological 35°C had no effect on transmission reliability and actually reduced PSP amplitude. It is also unlikely that synaptic transmission was decreased by chelation of intracellular calcium during whole cell recordings because the pipette internal solution did not contain EGTA (Ohana and Sakmann 1998; Rozov et al. 2001). Transmission failures between LIII and IV cells could reflect impairment of presynaptic action potential invasion at the branch points of complex axon terminations (Henneman et al. 1984; Lüscher and Shiner 1990) that are characteristic of interneurons in this region (Schneider 1992; Schneider et al. 1995). It is also possible that the high failure rates seen in the present study result from co-release of enkephalin at synaptic terminals containing glutamate or GABA (Schneider and Walker 2007; Todd et al. 1992, 2003) via activation of inhibitory presynaptic opioid receptors (Grudt and Henderson 1998; Hori et al. 1992).
Chemical mediators at local synaptic linkages
The present results are consistent with our previous finding that the VGLUT2 vesicular glutamate transporter, a specific and reliable marker of glutamatergic synapses, is localized within axon terminals of dorsal horn interneurons (Schneider and Walker 2007). Together these studies show that glutamate, acting on AMPA type receptors, mediates excitatory synaptic linkages between lamina III and IV interneurons. The observation that single-axon IPSPs were reduced by bicuculline agrees with immunolocalization of glutamic acid decarboxylase, the major synthetic enzyme for GABA, within axon terminals of laminae III/V interneurons (Schneider and Lopez 2002) and is consistent with studies showing expression of GABA and GABAA receptor subunits in this region of the dorsal horn (Alvarez et al. 1996; Magoul et al. 1987; Persohn et al. 1991; Todd and McKenzie 1989). The presence of a substantial bicuculline-resistant component of single-axon IPSPs that may also be strychnine-sensitive is supported by previous suggestions that GABA and glycine are co-transmitters at inhibitory connections between lamina III and IV interneurons (Basbaum 1988; Curtis et al. 1968; Todd and Sullivan 1990; Todd et al. 1995). Moreover, the present results are also consistent with evidence that glycinergic transmission is concentrated in the deep layers of the dorsal horn (Cronin et al. 2004) and support a significant role for glycine in local circuits receiving large fiber inputs from mechanoreceptors.
Neural circuitry in superficial and deep dorsal horn
It is tempting to compare the results of this study to recent studies of synaptic connectivity in the substantia gelatinosa, a region traditionally associated with nociceptive and thermoreceptive sensory inputs, to illuminate functional organization in adjacent spinal zones having different sensory functions. First, our observation that a surprisingly high percentage (69%) of LIII/IV cell pairs was connected by inhibitory linkages is similar to Lu and Perl's (2003) finding that a majority of connected cell pairs in LII evidenced inhibitory linkages. Taken together, these results suggest that inhibitory interneurons comprise a substantial proportion of intrinsic circuitry in the superficial and deep dorsal horn laminae. However, it should be noted that in their study of LII connectivity, Santos et al. (2007) reported evidence that only 10% of cell pairs were linked by inhibitory connections and concluded that excitatory interneurons dominate in this layer. Disparity between studies regarding the incidence of inhibitory linkages may be due in part to the use of different electrophysiological recording procedures. Whereas the present study and the one by Lu and Perl (2003) tested synaptic connectivity using paired whole cell recordings and recorded PSPs in current-clamp mode, Santos et al. (2007) determined synaptic linkage by an approach using cell-attached and whole cell recording configurations and identified linkages by recording synaptic currents under voltage clamp conditions. Our observations that single-axon synaptic currents could be recorded at only a few LIII/IV connections suggest that Santos et al.'s procedures may have underestimated connectivity, especially linkages mediated by weak inhibitory synapses that are situated on distal dendrites.
The present study of LIII–IV and paired recordings from LII (Lu and Perl 2003, 2005) document similar observations that all excitatory linkages, and the preponderance of inhibitory connections, are unidirectional. Bidirectional (reciprocal) synaptic connections between dorsal horn neurons are rare. The few cases of bidirectional interactions seen in the present study were inhibitory; this is in agreement with a recent description of mutual inhibitory connections between a small subset of LII neurons (Zheng et al. 2007). However, the organization of local LIII/IV circuits may differ from those in the superficial laminae where it has been reported that connections are made between preferred combinations of neurons having certain firing properties (Lu and Perl 2003, 2005). The present results suggest that intrinsic circuits of the deep dorsal horn appear to involve multiple categories of neurons in many combinations.
The present study reports a probability of finding synaptic connections between randomly selected neuron pairs in LIII–IV that is about three times higher than for cell pairs in the superficial dorsal horn (Lu and Perl 2003, 2005). The high degree of neuronal interconnectivity in LIII/IV is consistent with anatomical descriptions that show the axons of many LIII/IV interneurons give rise to numerous synaptic terminations (Schneider 1992, 2003; Schneider et al. 1995). However, differences in linkage ratios could also result from species, age of animals, or recording technique. Furthermore, studies may in fact underestimate neuronal connectivity to different degrees. Axonal branches are pruned during sectioning, removing synaptic inputs to neurons near the slice surface. Moreover, weak connections with high failure rates may be particularly difficult to recognize, leading to variability between studies. A study of connectivity among presumptive deep dorsal horn neurons in cat was previously published (Brown et al. 1979) using paired simultaneous, extracellular recordings and revealed an unexpected paucity of monosynaptic linkage between cells that shared similar receptive field locations on the skin. As the results reported here show, synaptic connections within this region are relatively weak, making it unlikely they could be detected in extracellular recordings.
Functional implications for mechanosensory processing
The present observations shed new light on how multiple sensory channels originating from mechanoreceptors innervating the skin may be organized in down-stream circuits in the spinal dorsal horn. First, the fact that synaptic connections between lamina III and IV neurons are predominantly unidirectional, with a majority of these linkages being inhibitory, suggests that reciprocal ("feed-back") connections do not contribute significantly to information transfer within lamina III/IV circuits. Although we observed similar transmission reliability for local excitatory and inhibitory connections, the increase in synaptic jitter with increasing failure rate for single-axon IPSPs suggests that transmission along local inhibitory pathways may be less secure than for excitatory pathways.
Second, as summarized in Table 1, there is no evidence that linkage probability for LIII/IV neurons is related to congruence in firing patterns of the pre- and postsynaptic cells as might be expected if signal flow from mechanosensory channels originating in the skin is simply preserved along modality-specific circuits within the deep dorsal horn. Tapper et al. (1973) suggested that a matching of signaling characteristics of primary receptors with postsynaptic membrane properties of dorsal horn neurons assists in organizing cutaneous mechanoreceptive input to the spinal cord. Our previous work (Schneider 2005) finding that tonic- and phasic-firing deep dorsal horn neurons display differential responsiveness to sustained and dynamic skin stimulation supports this concept. The present results suggest that flow of information within intrinsic spinal circuits receiving input from cutaneous mechanoreceptors is conducted along multiple pathways that may be parallel and complementary: pathways that transform sensory information based on the differential intrinsic properties of the pre- and postsynaptic cells and pathways made of elements with similar intrinsic properties that preserve the informational content of the afferent signal.
Last, afferent information from different sensory channels originating in the skin does not appear to be transmitted with uniform fidelity within dorsal horn circuits. As previously reported (Schneider 2003, 2005), deep dorsal horn neurons with phasic firing patterns are activated selectively by rapid, time-dependent membrane depolarizations and rapid skin indentation or movements parallel to the skin surface, whereas the discharge of tonic cells more effectively signal static skin displacements. The present study suggests that connections made by phasic cells are more reliable than those by tonic cells and PSPs mediated at these connections have a longer duration. Thus local circuits within spinal LIII/IV may be specialized to preserve input from cutaneous mechanoreceptive afferents signaling stimulus velocity or rate of change. Input from rapidly adapting afferents may predominate over slowly adapting mechanoreceptors that signal static stimuli and be subject to greater temporal summation, thereby contributing to feature extraction processes in the deep spinal dorsal horn.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: Dept. of Physiology, 2196 Biomedical Physical Sciences, Michigan State University, East Lansing, MI 48824-3320 (E-mail: schnei98{at}msu.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Allen C, Stevens CF. An evaluation of causes for unreliability of synaptic transmission. Proc Natl Acad Sci USA 91: 10380–10383, 1994.
Alvarez FA, Taylor-Blake B, Fyffe REWF, De Blas AL, Light AR. Distribution of immunoreactivity for the β2 and β3 subunits of the GABAA receptor in the mammalian spinal cord. J Comp Neurol 365: 392–412, 1996.[CrossRef][Web of Science][Medline]
Basbaum AI. Distribution of glycine receptor immunoreactivity in the spinal cord of the rat: cytochemical evidence for control of lamina I and V nociceptive neurons. J Comp Neurol 278: 330–336, 1988.[CrossRef][Web of Science][Medline]
Brown AG. Organization in the Spinal Cord. Berlin, Germany: Springer-Verlag, 1981.
Brown PB. Response of cat dorsal horn cells to variations of intensity, location and area of cutaneous stimuli. Exp Neurol 23: 249–265, 1969.[CrossRef][Web of Science][Medline]
Brown PB, Koerber HR, Yezierski RP. Cross-correlation analysis of connectivities among cat lumbosacral dorsal horn cells. J Neurophysiol 42: 1199–1211, 1979.
Cronin JN, Bradbury EJ, Lidierth M. Laminar distribution of GABAA- and glycine-receptor mediated tonic inhibition in the dorsal horn of the rat lumbar spinal cord: effects of picrotoxin and strychnine on expression of Fos-like immunoreactivity. Pain 112: 156–163, 2004.[CrossRef][Web of Science][Medline]
Curtis DR, Hosli L, Johnston GAR. A pharmacological study of the depression of spinal neurons by glycine and related amino acids. Exp Brain Res 6: 1–18, 1968.[Web of Science][Medline]
Deuchars J, West DC, Thomson AM. Relationships between morphology and physiology of pyramid-pyramid single axon connections in rat neocortex in vitro. J Physiol 478.3: 423–435, 1994.
Feldmeyer D, Egger V, Lübke J, Sakmann B. Reliable synaptic connections between pairs of excitatory layer 4 neurons within a single "barrel" of developing rat somatosensory cortex. J Physiol 521.1: 169–190, 1999.
Feldmeyer D, Lübke J, Silver RA, Sakmann B. Synaptic connections between layer 4 spiny neuron-layer 2/3 pyramidal cell pairs in juvenile rat barrel cortex: physiology and anatomy of interlaminar signaling within a cortical column. J Physiol 538.3: 803–822, 2002.
Feldmeyer D, Lübke J, Sakmann B. Efficacy and connectivity of intracolumnar pairs of layer 2/3 pyramidal cells in the barrel cortex of juvenile rats. J Physiol 575: 583–602, 2006.
Feldmeyer D, Roth A, Sakmann B. Monosynaptic connections between pairs of spiny stellate cells in layer 4 and pyramidal cells in layer 5A indicate that lemniscal and paralemniscal afferent pathways converge in the infragranular somatosensory cortex. J Neurosci 25: 3423–3431, 2005.
Fyffe REW. Laminar organization of primary afferent terminations in the mammalian spinal cord. In: Sensory Neurons: Diversity, Development and Plasticity, edited by Scott SA. New York, NY: Oxford, 1992.
Gentet LJ, Ulrich D. Strong, reliable and precise synaptic connections between thalamic relay cells and neurons of the nucleus reticularis in juvenile rats. J Physiol 546.3: 801–811, 2003.
Grudt TJ, Henderson G. Glycine and GABAA receptor-mediated synaptic transmission in rat substantia gelatinosa: inhibition by µ-opioid and GABAB agonists. J Physiol 507.2: 473–483, 1998.
Gulyás AI, Miles R, Sík A, Tóth K, Tamamaki N, Freund TF. Hippocampal pyramidal cells excite inhibitory neurons through a single release site. Nature 366: 683–687, 1993.[CrossRef][Medline]
Harrison PJ, Jack JJB, and Kullmann DM. Monosynaptic EPSPs in cat lumbosacral motoneurons from group Ia afferents and fibers descending in the spinal cord. J Physiol 412: 43–63, 1989.
Heavner JE, de Jong RH. Spinal cord neuron response to natural stimuli. A microelectrode study. Exp Neurol 39: 293–306, 1973.[CrossRef][Web of Science][Medline]
Henneman E, Lüscher H-R, Mathis J. Simultaneously active and inactive synapses of single Ia fibers on cat spinal motoneurons. J Physiol 352: 147–161, 1984.
Hochman S, Garraway SM, Pockett S. Membrane properties of deep dorsal horn neurons from neonatal rat spinal cord in vitro. Brain Res 767: 214–219, 1997.[CrossRef][Web of Science][Medline]
Hori Y, Endo K, Takahashi, T. Presynaptic inhibitory action of enkephalin on excitatory transmission in superficial dorsal horn of rat spinal cord. J Physiol 450: 673–685, 1992.
Iansek R, Redman SJ. The amplitude, time course and charge of unitary excitatory post-synaptic potentials evoked in spinal motoneurone dendrites. J Physiol 234:665–688, 1973.
Jiang MC, Cleland CL, Gebhart GF. Intrinsic properties of deep dorsal horn neurons in the L6 –S1 spinal cord of the intact rat. J Neurophysiol 74: 1819–1827, 1995.
Koerber HR, Seymour AW, Mendell LM. Tuning of spinal networks to frequency components of spike trains in individual afferents. J Neurosci 11: 3178–3187, 1991.[Abstract]
Lu Y, Perl ER. A specific inhibitory pathway between substantia gelatinosa neurons receiving direct C-fiber input. J Neurosci 23: 8752–8758, 2003.
Lu Y, Perl ER. Modular organization of excitatory circuits between neurons of the spinal superficial dorsal horn (laminae I and II). J Neurosci 25: 3900–3907, 2005.
Lüscher H-R, Shiner JS. Simulation of action potential propagation in complex terminal arborizations. Biophys J 58: 1389–1399, 1990.[Web of Science][Medline]
Magoul R, Onteniente B, Geffard M, Calas A. Anatomical distribution and ultrastructural organization of the GABAergic system in the rat spinal cord. An immunocytochemical study using anti-GABA antibodies. Neurosci 20: 1001–1009, 1987.[CrossRef][Web of Science][Medline]
Mannen H, Suguira Y. Reconstruction of neurons of dorsal horn proper using Golgi-stained serial sections. J Comp Neurol 168: 303–312, 1976.[CrossRef][Web of Science][Medline]
Markram H, Lübke J, Frotscher M, Roth A, Sakmann B. Physiology and anatomy of synaptic connections between thick tufted pyramidal neurones in the developing rat neocortex. J Physiol 500.2: 409–440, 1997.
Menétrey D, Geisler GJ, Besson JM. An analysis of response properties of spinal cord dorsal horn neurons to nonnoxious and noxious stimuli in the spinal rat. Exp Brain Res 27: 15–33, 1977.[Web of Science][Medline]
Metz CB, Schneider SP, Fyffe REWF. Selective suppression of endogenous peroxidase activity: application for enhancing appearance of HRP-labeled neurons in vitro. J Neurosci Methods 26: 181–188, 1989.[CrossRef][Web of Science][Medline]
Ohana O, Sakmann B. Transmitter release modulation in nerve terminals of rat neocortical pyramidal cells by intracellular calcium buffers. J Physiol 513.1: 135–148, 1998.
Persohn E, Malherbe P, Richards JG. In situ hybridization histochemistry reveals a diversity of GABAA receptor subunit mRNAs in neurons of the rat spinal cord and dorsal root ganglia. Neuroscience 42: 497–507, 1991.[CrossRef][Web of Science][Medline]
Rozov A, Burnashev N, Sakmann B, Neher E. Transmitter release modulation by intracellular Ca2+ buffers in facilitating and depressing nerve terminals of pyramidal cells in layer 2/3 of the rat neocortex indicates a target cell-specific difference in presynaptic calcium dynamics. J Physiol 531.3: 807–826, 2001.
Santos SFA, Rebello S, Derkach VA, Safronov BV. Excitatory interneurons dominate sensory processing in the spinal substantia gelatinosa of rat. J Physiol 581.1: 241–254, 2007.
Scharfman HE. Evidence from simultaneous intracellular recordings in rat hippocampal slices that area CA3 pyramidal cells innervate dentate hilar mossy cells. J Neurophysiol 72: 2167–2180, 1994.
Scheibel ME, Scheibel AB. Terminal axonal patterns in cat spinal cord. II. The dorsal horn. Brain Res 9: 32–58, 1968.[CrossRef][Medline]
Schneider SP. Functional properties and axon terminations of interneurons in laminae III–V of the mammalian spinal dorsal horn. J Neurophysiol 68: 1746–1759, 1992.
Schneider SP. Spike frequency adaptation and signaling properties of identified neurons in rodent deep spinal dorsal horn. J Neurophysiol 90: 245–258, 2003.
Schneider SP. Mechanosensory afferent input and neuronal firing properties in rodent spinal laminae III–V: re-examination of relationships with analysis of responses to static and time-varying stimuli. Brain Res 1034: 71–89, 2005.[CrossRef][Web of Science][Medline]
Schneider SP, Lopez M. Immunocytochemical localization of glutamic acid decarboxylase in physiologically identified interneurons of hamster spinal laminae III–IV. Neuroscience 115: 627–636, 2002.[CrossRef][Web of Science][Medline]
Schneider SP, Sandiford DR, Kavookjian AM, Johnson BD. Synaptic connectivity of local circuit neurons in laminae III and IV of hamster spinal cord. J Comp Neurol 355: 380–391, 1995.[CrossRef][Web of Science][Medline]
Schneider SP, Walker TM. Morphology and electrophysiological properties of hamster spinal dorsal horn neurons that express VGLUT2 and enkephalin. J Comp Neurol 501: 790–809, 2007.[CrossRef][Web of Science][Medline]
Tamás G, Buhl EH, Somogyi P. Fast IPSPs elicited via multiple synaptic release sites by different types of GABAergic neuron in the cat visual cortex. J Physiol 500.3: 715–738, 1997.
Tapper DN, Brown PB, Moraff H. Functional organization of the cat's dorsal horn: connectivity of myelinated fiber systems of hairy skin. J Neurophysiol 36: 817–826, 1973.
Thomson AM, Girdlestone D, West DC. Voltage-dependent currents prolong single-axon postsynaptic potentials in layer III pyramidal neurons in rat neocortical slices. J Neurophysiol 60: 1896–1907, 1988.
Thomson AM, West DC, Deuchars J. Properties of single axon excitatory postsynaptic potentials elicited in spiny interneurons by action potentials in pyramidal neurons in slices of rat neocortex. Neuroscience 69: 727–738, 1995.[CrossRef][Web of Science][Medline]
Todd AJ, Hughes DI, Polgár E, Nagy GG, Mackie M, Ottersen OP, Maxwell DJ. The expression of vesicular glutamate transporters VGLUT1 and VGLUT2 in neurochemically defined axonal populations in the rat spinal cord with emphasis on the dorsal horn. Eur J Neurosci 17: 13–27, 2003.[CrossRef][Web of Science][Medline]
Todd AJ, McKenzie J. GABA-immunoreactive neurons in the dorsal horn of the rat spinal cord. Neurosci 31: 799–806, 1989.[CrossRef][Web of Science][Medline]
Todd AJ, Spike RC, Chong D, Neilson M. The relationship between glycine and gephyrin in synapses of the rat spinal cord. Eur J Neurosci 7: 1–11, 1995.[CrossRef][Web of Science][Medline]
Todd AJ, Spike RC, Russell G, Johnston HM. Immunohistochemical evidence that Met-enkephalin and GABA coexist in some neurons of the rat dorsal horn. Brain Res 584: 149–156, 1992.[CrossRef][Web of Science][Medline]
Todd AJ, Sullivan AC. Light microscopic study of the coexistence of GABA-like and glycine-like immunoreactivities in the spinal cord of the rat. J Comp Neurol 296: 496–505, 1990.[CrossRef][Web of Science][Medline]
Willis WD, Coggeshall RE. Sensory Mechanisms of the Spinal Cord. Primary Afferent Neurons and the Spinal Dorsal Horn. (3rd ed.) New York: Kluwer Acedemic/Plenum, 2004, vol. 1, chapt. 2, p. 19–90.
Zheng J, Lu Y, Perl ER. Inhibitory interactions including mutual inhibition in the spinal substantia gelatinosa (lamina II). Soc Neurosci Abstr 714.18, 2007.
This article has been cited by other articles:
|
|
 |
|
  V. B. Lu, J. E. Biggs, M. J. Stebbing, S. Balasubramanyan, K. G. Todd, A. Y. Lai, W. F. Colmers, D. Dawbarn, K. Ballanyi, and P. A. Smith Brain-derived neurotrophic factor drives the changes in excitatory synaptic transmission in the rat superficial dorsal horn that follow sciatic nerve injury J. Physiol., March 1, 2009; 587(5): 1013 - 1032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Lu Synaptic Wiring in the Deep Dorsal Horn. Focus on "Local Circuit Connections Between Hamster Laminae III and IV Dorsal Horn Neurons" J Neurophysiol, March 1, 2008; 99(3): 1051 - 1052. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) produced transient de- or hyperpolarizing membrane responses in the postsynaptic cell. 
, transmission failure. E and F: single-axon excitatory postsynaptic potentials (EPSPs, E) and inhibitory postsynaptic potentials (IPSPs, F) produced in the postsynaptic cell (Post) by stimulating presynaptic APs (Pre, 
). - - -, approximate boundary between LII and LIII. Dendrites (*) of cell 1 in B penetrate dorsally into the substantia gelatinosa (LII). Electrophysiological recordings from cells 1 and 2 of A and B are shown in 
Ai e–t/
