| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Physiology and Center for Neuroscience, University of Wisconsin Medical School, Madison, Wisconsin; and 2Institute of Biomedical and Life Sciences, University of Glasgow, Glasgow, Scotland
Submitted 25 June 2004; accepted in final form 12 October 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Grillner and Wallen 1985; Keihn and Butt 2003). The locomotor CPG consists of spinal interneurons constituting the rhythm-generating and -coordinating networks that control hindlimb movements. In the isolated spinal cord, fictive locomotion is defined as alternating rhythms between left-right lumbar ventral roots and flexor- and extensor-related motoneuron pools (Cazalets et al. 1990; Kiehn and Kjaerulff 1996; Kudo and Yamada, 1987). Much of our knowledge about the neuronal organization of the CPG comes from studies of the swimming movements in the lamprey and Xenopus (Buchanan 1982; Buchanan and Grillner 1987; Cohen and Harris-Warrick 1984; Roberts et al. 1998). Considerably less is known about the identity of interneurons forming the locomotor networks in walking mammals. It is generally thought that networks responsible for rhythm generation are distributed along several thoracic and lumbar segments with a rostrocaudal rhythmogenic gradient (Cazalets et al. 1995; Kjaerulff and Kiehn 1996; Kremer and Lev-Tov 1997). In the transverse plane, interneurons of the hindlimb CPG seem to be located in the ventromedial regions of laminae VIII, X, and medial lamina VII (reviewed by Kiehn and Kjaerulff 1998). Using blind whole cell patch-clamp recordings, recent studies have identified heterogeneous populations of commissural inhibitory interneurons that might constitute part of the left-right coordinating networks in the rat spinal cord (Butt et al. 2002a; Butt and Kiehn 2003; reviewed by Butt et al. 2002b).
Progress in identifying interneurons that are functionally integrated components of the CPG has been slow, partly because determining synaptic circuitry requires the classification of particular neuronal populations (reviewed by Jankowska 2001). To overcome some of the technical limitations of interneuron identification in the isolated spinal cord, transgenic mice have been used with the intention of characterizing interneurons with genetic markers. A novel approach has recently become available with the discovery that ventral neurons can be distinguished by combinatorial expression of transcription factors (Briscoe et al. 2000; Goulding and Lamar 2000; Pierani et al. 1999, 2001), and distinct genetic markers for specific neuronal populations have been used to functionally identify their role in motor behavior (Goulding et al. 2002, reviewed by Jessell 2000; Kiehn and Butt 2003; Lanuza et al. 2004, Shirasaki and Pfaff 2002, Wenner et al. 2000).
In our study, a fragment of HB9, a homeobox gene crucial for motoneuron differentiation (Arber et al. 1999; Thaler et al. 1999), was used to drive the expression of the reporter gene enhanced green fluorescent protein (eGFP), generating GFP positive (GFP+) motoneurons and ventral interneurons. The objective of this study was to examine the properties of a distinct cluster of HB9+/GFP+ interneurons and to characterize their activity pattern during neurochemically induced locomotor rhythms in the isolated spinal cord. Our electrophysiological, morphological, and immunohistochemical findings suggest that the visually identified HB9/GFP interneurons are premotor glutamatergic interneurons that might constitute part of the locomotor rhythm-generating networks.
A preliminary report of this study was published in an abstract form (Hinckley et al. 2003).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). A 9-kb fragment including the 5' upstream region of the murine HB9 gene was used to control the expression of the reporter gene eGFP, generating GFP+ somatic and autonomic motoneurons and ventral interneurons. The procedures for generating the HB9/eGFP transgenic mice and the genetic elements used for the construct have been described previously (Wichterle et al. 2002). The transgenic HB9/eGFP mouse was generated by Ivo Lieberam and Thomas Jessell (Columbia University, NY). Ventral root and whole cell patch-clamp recordings
Postnatal (P1–P4) mice were anesthetized by hypothermia, then decapitated, and spinal cords were extracted in ice-cold oxygenated extracellular solution. The cord with ventral roots T12–S2 attached was isolated and equilibrated at room temperature for 30 min. The cord was then transferred to a silicone elastomer (Sylgard)-coated recording chamber, where it was continuously superfused with oxygenated extracellular solution at a rate of 2–5 ml/min at room temperature. The procedures for ventral root recordings were similar to those described previously (Hinckley et al. 2005). To verify that rhythms were related to locomotion, motor outputs were recorded in left-right L2 and the right L5 ventral roots. Ventral root potentials were amplified with AC-coupled DAM-50 amplifiers (World Precision Instruments) and were recorded using Axotape software (Axon Instruments). Potentials were filtered at 1–3 kHz, digitized at 2–5 kHz and stored on a disk for off-line analysis.
Locomotor-like rhythms were triggered in the intact spinal cords using a mixture of the neurotransmitter agonists N-methyl-DL-aspartic acid (NMA, 4–7 µM), 5-hydroxytryptamine creatinine sulfate complex, (5-HT, 7–10 µM), and dopamine (25–50 µM) (Hinckley et al. 2005; Jiang et al. 1999). For whole cell patch-clamp recordings in visually identified GFP+ interneurons, intact cords were longitudinally cut along the midline, and the hemispinal cords were placed in the recording chamber with the medial side up. GFP+ clustered interneurons, located close to the medial surface, were visualized using fluorescence optics with excitation at 490 nm (Olympus BX50WI microscope). To obtain whole cell recording configuration, the optics were switched to infrared-differential interference contrast optics (Ziskind-Conhaim et al. 2003). GFP diffusion into the patch-clamp pipette confirmed that recordings were performed in GFP+ interneurons. Whole cell recordings were carried out using electrodes pulled to tip resistances of 5–8 M
using a multi-stage puller (P-97, Sutter Instruments). Intracellular potentials were recorded with an Axoclamp 2B amplifier (Axon Instruments). Potentials were filtered at 3 kHz and digitized at 10–20 kHz. Interneurons were included in the study only if their resting membrane potentials were more negative than –50 mV and action potential peak amplitudes overshot zero. Membrane potentials were corrected for a 10-mV liquid junction potential (Gao et al. 2001).
Morphology
Axonal projections of the clustered GFP+ interneurons were examined in neurobiotin-filled neurons. Neurobiotin (0.5%) was dissolved in the pipette solution, and it diffused into the cell during intracellular recordings. Hemispinal cords were fixed overnight in 4% freshly depolymerized formaldehyde in 0.1 M phosphate buffer (pH 7.4). The cords were then washed three times (10 min each) in phosphate buffer and embedded in 2–6% agar. The hemicords were cut into transverse 70- to 100-µm sections and placed in serial order in small vials. These were rinsed and incubated overnight in avidin-rhodamine (1:1000, Jackson Immunoresearch) in phosphate-buffered saline (PBS) containing 0.3% Triton X100 (PBST). The tissue was then rinsed in PBS and mounted in serial order on glass slides with anti-fade mounting medium (Vectashield, Vector Laboratories). Sections containing labeled neurons were selected after viewing them with a fluorescence microscope, and these were then processed for immunocytochemistry.
Immunohistochemistry
Transverse slices with neurobiotin-filled interneurons were incubated for 24–72 hrs at 4°C in a mixture of rabbit anti-GFP antiserum (1:4000, Abcam) and guinea pig anti-VGLUT2 antiserum (1:5000, Chemicon) in PBST. They were then rinsed and incubated overnight in donkey anti-rabbit IgG conjugated to Alexa 488 (1:500, Molecular Probes) and donkey anti-guinea pig IgG conjugated to cyanine 5.18 (Cy5, 1:100; Jackson Immunoresearch) in PBST. Sections were rinsed in PBS and mounted in Vectashield. They were scanned with a confocal laser microscope (Bio-Rad Radiance 2100) equipped with Argon, Green HeNe and Red diode lasers, through a x60 oil-immersion lens.
To identify neurons expressing the HB9 protein, spinal cords were fixed and sectioned as described in the preceding text. Slices were incubated for 72 h at 4° C in rabbit anti-HB9 antibody (1:2500). They were then rinsed and incubated for 4–12 h in goat anti-rabbit IgG conjugated to either Alexa 546 (1:500) or Cy5 (1:200, Molecular Probes).
Data analysis of ventral root potentials
Typically, in each experiment samples of 20 successive rhythmic bursts were analyzed off-line (Hinckley et al. 2005). To facilitate the analysis of burst properties, extracellular recordings were rectified and smoothed using adjacent averaging over 100–200 points (Origin 6, MicroCal). These were used to calculate the cycle period and burst interval, duration, and amplitude. Cycle period was defined as the time between the onset of two consecutive bursts, and burst duration was measured as the time between the onset of excitation and its return to baseline. Paired t-tests were used to determine the statistical significance (P < 0.05). Rhythm stability was measured by cycle period variation coefficient (SD/mean).
Statistical analysis
Two types of statistical analysis were used to correlate the phase relations between membrane depolarizations and ventral root bursts. Circular statistics was used to analyze the correlation between subthreshold rhythms and ventral root bursts, and histograms of firing frequency tested the correlation between firing rate and the phases of the cycle period. Using the circular statistics, the onset of membrane depolarization was subtracted from the onset of ventral root bursting and divided by the cycle period of the root discharge. Twenty randomly picked rhythmic cycles were analyzed yielding values from 0 to 1. Individual values were plotted on a circle and the mean value was shown by the tip of the vector. For values from 0 to 1, 0 indicated in-phase correlation while 0.5 marked an out-of-phase correlation. The length of the mean vector is inversely correlated with the distribution of the phase values around the mean (Beato and Nistri 1999; Butt et al. 2002a; Lev-Tov et al. 2000). The significance of the mean was calculated from the length of the vector (r) using Rayleighs test (Zar 1974).
In interneurons that exhibited repetitive firing, ventral root cycle periods were divided into 10 equal bins, with bins 1–5 representing the period of ventral root burst and bins 6–10 representing the interburst period. Frequencies of instantaneous action potentials or postsynaptic currents were averaged for all events within one bin for 20 successive cycles. Interneuron-ventral root correlation was determined by comparing the averaged frequency in the burst and interburst phases. Averaged binned instantaneous frequencies were pooled for all interneurons and plotted in histograms. Correlation of the entire pool of firing interneurons was determined by averaging binned frequency values and comparing them in the burst and interburst periods using Student's t-test.
High duty cycle variability, the fraction of the cycle period during ventral root burst, adversely affects analysis of cyclic event patterns unless a dual reference procedure is used (Berkowitz and Stein 1994). The averaged duty cycle was 55 ± 0.08% (mean ± SD; n = 13), indicating that increased action potential frequency during the first half of the cycle period corresponded to the duration of ventral root burst.
Solutions and chemicals
Extracellular solution contained (in mM) 128 NaCl, 4 KCl, 1.5 CaCl2, 1 MgSO4, 0.5 NaH2PO4, 21 NaHCO3, and 30 glucose. The solution was adjusted to pH 7.3 using NaOH, and the osmolarity was 315–325 mosM. The whole cell pipette solution contained (in mM) 140 K gluconate, 9 KCl, 10 N-(2-hydeoxyethyl)piperazine-N'-(2-ethanesulfonic acid) (HEPES), 0.2 ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid (EGTA), 1 Mg-ATP, and 0.1 GTP. The solution was adjusted to pH 7.2 using KOH, and the osmolarity was 290–305 mosM. All chemicals were obtained from Sigma.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
In our study, mice were genetically altered using the HB9 promotor to drive the expression of the reporter gene eGFP. Co-expression of HB9 protein and eGFP is restricted to motoneurons in the embryonic spinal cord (Thaler et al. 1999; Wichterle et al. 2002), but widely distributed GFP+ ventral interneurons were evident in postnatal mice (Fig. 1A). The large motoneurons, clustered in the lateral and medial columns could be readily distinguished from interneurons. HB9 expression was evident in motoneurons and a small cluster of GFP+ interneurons (see following text). eGFP expression in HB9 negative (HB9–) interneurons might have resulted from the absence of silencer elements from the promotor fragment. Alternatively, it is possible that the position of the promotor at the transgene integration site directed eGFP expression in HB9– neurons.
|
15-µm-diam neurons had distinct primary dorsal and ventral dendrites (also Figs. 2, 8, and 9). Dendrites crossing the midline were also evident (Fig. 1B). On each side of the cord, the readily visualized cluster consisted of three to nine interneurons/100-µm transverse section. Examination of their topographical distribution along the lumbar rostrocaudal axis demonstrated that the clusters were located in segments L1–L3. Abundant GFP+ interneurons were distributed along the midline in segments L4–L5 (Fig. 1C), without an obvious organization of medial clusters similar to those seen in L1–L3. Preliminary observations have shown that medial clusters of GFP+ interneurons were also present in T12–T13.
|
|
|
Clustered HB9-expressing interneurons were not evident along the midline in segments L4–L5. However, a small number of GFP+, HB9-immunoreacitve interneurons were scattered in lateral lamina VII (not shown). It is possible that those HB9/GFP interneurons share similar characteristics with the interneurons clustered in L1–L3, but because of their lateral distribution they were inaccessible for visually targeted whole cell recordings in the hemisected cord. Moreover, these interneurons could not be morphologically distinguished from other lamina VII GFP+ interneurons, making it difficult to target them for whole cell recordings.
Neurochemically induced rhythms in HB9/GFP interneurons
Locomotor-like activity in the isolated cord is defined as alternating rhythms between left-right lumbar ventral roots and flexor-related and extensor-related ventral roots. Locomotor patterns can be induced in vitro using neurotransmitter agonists such as the combination of 5-HT, NMA, and dopamine (Whelan et al. 2000; Hinckley et al. 2005). Variable agonist concentrations were required to initiate rhythmic activity, but in the majority of preparations, stable alternating rhythms were induced with 10 µM 5-HT, 5 µM NMA, and 50 µM dopamine.
Whole cell patch-clamp recordings were carried out in the hemisected spinal cord in which the clustered HB9/GFP interneurons were visible when the hemispinal cord was placed with the medial side up. However, neurochemically induced ventral root rhythms could not be readily initiated in the hemisected spinal cord. Therefore rhythms were first induced in the intact spinal cord, and after 15–20 min stable rhythmic activity in the intact cord, the spinal cord was cut longitudinally along the midline (Fig. 3).
|
). In some hemispinal cords, higher concentrations of agonists were required to restore rhythmic activity (see also Kremer and Lev-Tov 1997). Clustered HB9/GFP interneurons were located 30–70 µm from the medial surface in lamina VIII and were therefore accessible for whole cell patch-clamp recording in the hemisected spinal cord. The interneurons could be visually identified based on their location, small size, spindle shape, and their primary ventral and dorsal dendrites, which were visible with a x40 objective. To confirm that interneurons recorded from expressed the HB9 protein, in a few experiments the hemicord was processed for staining with HB9 antibody after whole cell recordings were completed. HB9 staining was evident only in interneurons in which the recordings lasted for <30 min (Fig. 4, n = 3), and it was difficult to detect after longer recording periods. Native GFP fluorescence was also not as bright in recorded interneurons as in adjacent clustered interneurons. This raised the possibility that diffusion of GFP (see METHODS) and HB9 into the recording pipette over time was at least partly responsible for the weaker GFP and HB9 staining.
|
|
|
Irregular membrane depolarizations and firing, not correlated with ventral root discharges, were recorded in 12% of clustered GFP+ neurons (n = 4/32, not shown). Neurons in the proximity of the cluster did not generate rhythmic activity (GFP+, n = 5; GFP–, n = 4). Our findings suggested that exposure to neurotransmitter agonists triggered rhythmic activity in the majority of HB9/GFP interneurons but not in lamina VIII interneurons adjacent to the clusters.
To determine whether rhythmic membrane oscillations were associated with changes in synaptic activity, the frequency of excitatory and inhibitory postsynaptic currents (EPSCs and IPSCs, respectively) was analyzed during the cycle period (Fig. 7). Whole cell patch-clamp recordings were carried out at holding potentials of –40 to –45 mV (n = 9). Under our experimental conditions, glycine and GABA reversal potentials were approximately –67 mV (Gao et al. 1995), therefore EPSCs and IPSCs appeared as inward and outward currents, respectively. The average EPSC frequency increased approximately twofold during ventral root burst (Fig. 7B). The onset of increased EPSC frequency preceded the onset of ventral root burst, and it was significantly higher during the burst. Instantaneous frequency during ventral root burst was 7.0 ± 2.4 (SD) Hz (n = 9), significantly higher than the 3.6 ± 2.0 Hz recorded during the interburst period (P < 0.05). IPSC frequency was low (<1 Hz) and did not change during the cycle period. EPSC frequency was more than sevenfold higher than IPSC frequency during ventral root bursts, and more than threefold higher during the interburst phase. These findings suggested that in the hemisected spinal cord, HB9/GFP interneurons received mostly excitatory inputs rhythmically correlated with ventral root bursts. Inhibitory synapses onto these interneurons did not play an important role in regulating the amplitude and duration of membrane depolarizations.
|
Input resistance was measured by recording membrane potentials in response to steps of hyperpolarizing currents during the interburst phase. In the majority of interneurons (n = 13/15), changes in membrane potentials were linear in the range of –60 to –180 mV (not shown), and the average input resistance was 1.0 ± 0.4 (SD) G (n = 13). In only two interneurons, small hyperpolarization-dependent depolarizing sags were apparent at potentials more negative than –150 mV (not shown).
Lower input resistance of 0.7 ± 0.3 G (n = 8) was measured when neurotransmitter agonists were removed from the extracellular solution, implying that at least one of the agonists in the mixture increased interneuron input resistance. Removing the agonists also significantly reduced the frequency of excitatory postsynaptic potentials. It is conceivable that 5-HT was responsible for the increase in input resistance, higher EPSP frequency, and, possibly, the generation of oscillatory potentials as shown in motoneurons of the postnatal rat spinal cord (Ziskind-Conhaim et al. 1993).
A rheobase of 10–15 pA was measured in HB9/GFP interneurons. The threshold current for action potential generation did not change when the agonists were removed, and in most interneurons, a similar pattern of action potential firing was recorded in the presence and absence of neurotransmitter agonists.
Morphological and immunohistochemical properties of HB9/GFP interneurons
HB9/GFP projections to potential target neurons was examined in neurobiotin-filled interneurons. The 15-µm-diam, bipolar interneurons had primary dorsal and ventral dendrites extending >200 µm (Figs. 8A and 9A, n = 12). Short-distance neurobiotin-filled axon projections gave rise to bouton-like varicosities in close apposition to proximal dendrites and somata of adjacent, GFP+ clustered interneurons (Fig. 8B). This might be indicative of recurrent synaptic connections between clustered interneurons. Neurobiotin-filled axon collaterals with arborizations in the area between the lateral and medial motor columns (Fig. 8C) had bouton-like varicosities in close apposition to GFP+ motoneuron dendrites (Fig. 8D). Our morphological observations raised the intriguing possibility that HB9/GFP interneurons form synaptic contacts on motoneurons.
A recent study has shown that most GFP+ interneurons were labeled with antibody against vesicular glutamate transporter 2 (VGLUT2), suggesting that these were glutamatergic interneurons (Hartley et al. 2003). To examine whether the clustered HB9/GFP interneurons were premotor, glutamatergic interneurons, transverse slices of the hemisected cord were stained with anti-VGLUT2 antiserum. Triple-labeled GFP+, neurobiotin-filled boutons labeled with VGLUT2, were apposed to GFP+ dendrites in the area between the lateral and medial motor columns (Fig. 9, n = 4). Although it is unlikely, we cannot rule out the possibility that neurobiotin-filled, VGLUT2 boutons were in close apposition to distal dendrites of GFP+ interneurons located dorsally to the motor columns rather than to motoneuron dendrites (Fig. 9B). Only a fraction of GFP+/neurobiotin-filled boutons were labeled with VGLUT2, probably because of the limited penetration of the VGLUT2 antibody, which was restricted to axonal arborizations within 15 µm of the tissue surface. Based on our findings, it is conceivable that the clustered HB9/GFP interneurons were premotor glutamatergic interneurons that were part of the CPG network in the hemisected spinal cord.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Functional classification of HB9/GFP interneurons
Traditionally, spinal interneurons have been functionally identified by their synaptic, inputs, intrinsic properties, target neurons and the role they play in motor activity (reviewed by Jankowska, 2001). It has been suggested that some group I afferents and flexor-related afferents provide synaptic inputs to interneurons involved in rhythm generation because hindlimb reflex pathways modulate the activity pattern of the locomotor CPG circuitry in the cat spinal cord (Gossard et al. 1994; reviewed by Hultborn et al. 1998; McCrea 2001). However, the specific function of afferent projections is yet to be determined because they also control motoneuron excitability during locomotion. Identifying synaptic inputs is a challenging task in the isolated spinal cord, in which neurons with long-range descending projections are not preserved. Recent studies have provided new information about the identity of commissural interneurons that might constitute part of the CPG networks in the rodent spinal cord. Blind whole cell recordings have characterized distinct populations of functionally identified commissural inhibitory interneurons that might constitute part of the rhythm-coordinating networks in the rat spinal cord (Butt et al., 2002a; Butt and Kiehn, 2003; reviewed by Butt et al., 2002b). Using molecular probes and genetic manipulations, EphA4-positive neurons have been identified as ipsilateral excitatory interneurons that are integral components of the locomotion CPG in the mouse spinal cord (Kullander et al. 2003).
Genetic identification has become an alternative approach to functional identification based on recent findings that interneuron subtypes can be distinguished by their specific expression of transcription factors. The principle of using transcription factors as markers for identifying distinct interneuron populations relies on differences in gene expression as a basis for functional identification (Lanuza 2004; Wenner et al. 2000; reviewed by Goulding et al. 2002; Kiehn and Butt 2003, Shirasaki and Pfaff 2002). Transcription factors have been used to drive the expression of the reporter gene GFP, providing visual identification of functionally distinct group of spinal interneurons. This innovative approach has been recently used to study the properties and synaptic connections of GFP+, Engrailed-1 expressing inhibitory interneurons with ascending projections in the zebrafish spinal cord (Higashijima et al. 2004).
In our study, the HB9 promotor controlled the expression of eGFP, giving rise to HB9+/GFP+ somatic motoneurons and a small group of lamina VIII interneurons. A similar expression profile of HB9/LacZ has been reported in compound mutant in which HB9/eGFP mice were bred with HB9/nlsLacZ knockin mice (T. Jessell, personal communication). Our finding that in newborn mice the medial cluster of interneurons expresses the HB9 protein is reminiscent of its expression in a subset of motoneurons and in a distinct interneuron population in Drosophila (Odden et al. 2002). The homeobox gene HB9 has a crucial role in motoneuron differentiation. In mutant mice lacking HB9, motoneurons transiently express homeodomain proteins typically expressed only in the interneuron population adjacent to the motor nuclei (Arber et al. 1999). It remains to be determined whether the HB9 gene plays an important role in the differentiation of the medially clustered interneurons.
It is unlikely that the widely distributed HB9–/GFP+ interneurons in laminae VII-VIII and X all belong to a network of specific function. However, even if the overall GFP expression pattern throughout the ventral horn is not correlated with a distinct function, the consistency of the distribution pattern of GFP+ neurons in the ventral horn can be used to reliably target subsets of visually identified interneurons for electrophysiological and morphological studies.
The identity of neurons making synaptic contacts on HB9/GFP interneurons remains mostly unknown. Our findings that EPSC frequency is significantly higher than IPSC frequency suggest that rhythmically active excitatory synapses contact HB9/GFP interneurons in the hemisected spinal cord. Based on our morphological observations, it is conceivable that clustered interneurons make synaptic connections on each other (Fig. 8B), implying that at least a fraction of EPSCs are generated by other HB9/GFP interneurons. Reciprocal excitation between HB9/GFP interneurons might serve to effectively increase and synchronize their neuronal output. Axonal arborizations of HB9/GFP neurons in motoneuron pools and the expression of VGLUT2 in boutons apposed to motoneuron dendrites, strongly suggest that these are premotor glutamatergic interneurons. Paired electrophysiological recordings or electron micrographs could provide direct evidence for this hypothesis. Similar function of mutual excitation and motoneuron activation has been described for excitatory premotor interneurons in the lamprey spinal cord (Buchanan and Grillner 1987; Parker, 2003, Parker and Grillner 2000).
Rhythmic activity in HB9/GFP interneurons
Our findings that rhythmic firing in HB9/GFP interneurons is in-phase with ventral root rhythmic bursts indicate that they might comprise a functional component of rhythm-generating networks in the mouse spinal cord. A recent report supports the idea that these interneurons are constituents of the locomotor CPG, showing that after a locomotor task the clustered GFP interneurons are labeled with the activity-dependent immediate early gene c-fos (Wilson et al. 2003).
Activity pattern in HB9/GFP interneurons during locomotor-like rhythms was examined in the hemisected spinal cord in the absence of the transverse coupling networks that coordinate the activity between the two sides of the cord. The difficulty of evoking rhythms in the hemisected spinal cord and the reduced quality of the rhythms and their longer duration, all point to the importance of reciprocal synapses in regulating rhythm properties. The coordinated activity after perturbation of reciprocal excitation and inhibition highlights the important role of ipsilateral networks in rhythm generation. Similar to the organization of spinal networks controlling swimming movements in the lamprey and Xenopus, it is likely that hindlimb movements in the mouse are regulated by segmental rhythm generators located on each side of the cord with reciprocal inhibitory connections between them.
The low frequency of IPSCs recorded in HB9/GFP interneurons during all phases of rhythmic cycle period suggests that the amplitude and duration of rhythmic membrane depolarizations are not regulated by ipsilateral inhibitory interneurons that directly synapse onto HB9/GFP interneurons. Similarly, it has been shown that ipsilateral inhibition is not required for rhythm generation in Xenopus and lamprey hemisected cords, indicating that the interaction between excitatory interneurons is sufficient for rhythm generation (Cangiano and Grillner 2003; Soffe 1989). It remains to be determined whether HB9/GFP interneurons receive synaptic inputs from commissural inhibitory interneurons in the intact spinal cord. Preliminary data have demonstrated that significantly higher IPSC frequency is recorded in transverse spinal cord slices in which at least part of the segmental, reciprocal inhibitory coupling networks are intact. Cyclic EPSC and IPSC frequencies in unidentified interneurons in laminae VII and X in the rat spinal cord imply that the interneurons receive inputs from rhythm-generating networks (Raastad et al. 1997). Similarly, synaptic drive affects rhythmic firing in descending commissural inhibitory interneurons (Butt et al. 2002a).
The correlation between rhythmic activity in HB9/GFP interneurons and rhythmic motoneuron outputs implies that HB9/GFP interneurons receive cyclic timing inputs from the locomotor CPG. This correlation raises the intriguing possibility that this cluster of premotor glutamatergic interneurons is a functionally integrated component of rhythm-generating networks. However, the evidence that these neurons are directly involved in generating the rhythmic pattern of motor output is circumstantial. A direct experimental approach to demonstrate that HB9/GFP interneurons are essential neuronal components of the locomotor CPG is to examine the effects of their ablation or inactivation on locomotor rhythms. However, it should be noted that this strategy is based on the unsubstantiated assumption that there is no redundancy or degeneracy in rhythmic interneuron inputs to motoneurons.
In conclusion, we have identified a genetically distinct population of lamina VIII premotor interneurons that generate rhythmic membrane potentials in-phase with rhythmic motor outputs. The restricted HB9 expression in the clustered interneurons distinguishes them from other interneuron populations and might be indicative of their distinctive function in motor behavior.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other corresspondence: L. Ziskind-Conhaim, Dept. of Physiology, 129 SMI, University of Wisconsin, Medical School, Madison, WI 53706 (E-mail: lconhaim{at}physiology.wisc.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Beato M and Nistri A. Interaction between disinhibited bursting and fictive locomotor patterns in the rat isolated spinal cord. J Neurophysiol 82: 2029–2038, 1999.
Berkowitz A and Stein PS. Descending propriospinal axons in the hindlimb enlargement of the red-eared turtle: cells of origin and funicular courses. J Comp Neurol 346: 321–336, 1994.[CrossRef][ISI][Medline]
Branchereau P, Morin D, Bonnot A, Ballion B, Charpon J, and Viala D. Development of lumbar rhythmic networks: from embryonic to neonate locomotor-like patterns in the mouse. Brain Res Bull 53: 711–718, 2000.[CrossRef][ISI][Medline]
Briscoe J, Pierani A, Jessell TM, and Ericson J. A homeodomain protein code specifies progenitor cell identity and neuronal fate in the ventral neural tube. Cell 101: 435–445, 2000.[CrossRef][ISI][Medline]
Buchanan JT. Identification of interneurons with contralateral, caudal axons in the lamprey spinal cord: synaptic interactions and morphology. J Neurophysiol 47: 961–975, 1982.
Buchanan JT and Grillner S. Newly identified "glutamate interneurons" and their role in locomotion in the lamprey spinal cord. Science 236: 312–314, 1987.
Butt SJ and Kiehn O. Functional identification of interneurons responsible for left-right coordination of hindlimbs in mammals. Neuron 38: 953–963, 2003.[CrossRef][ISI][Medline]
Butt SJ, Harris-Warrick RM, and Kiehn O. Firing properties of identified interneuron populations in the mammalian hindlimb central pattern generator. J Neurosci 22: 9961–9971, 2002a.
Butt SJ, Lebret JM, and Kiehn O. Organization of left-right coordination in the mammalian locomotor network. Brain Res Brain Res Rev 40: 107–117, 2002b.[CrossRef][Medline]
Cangiano L and Grillner S. Fast and slow locomotor burst generation in the hemispinal cord of the lamprey. J Neurophysiol 89: 2931–2942, 2003.
Cazalets JR, Borde M, and Clarac F. Localization and organization of the central pattern generator for hindlimb locomotion in newborn rat. J Neurosci 15: 4943–4951, 1995.[Abstract]
Cazalets JR, Borde M, and Clarac F. The synaptic drive from the spinal locomotor network to motoneurons in the newborn rat. J Neurosci 16: 298–306, 1996.
Cazalets JR, Grillner P, Menard I, Cremieux J, and Clarac F. Two types of motor rhythm induced by NMDA and amines in an in vitro spinal cord preparation of neonatal rat. Neurosci Lett 111: 116–121, 1990.[CrossRef][ISI][Medline]
Cohen AH and Harris-Warrick RM. Strychnine eliminates alternating motor output during fictive locomotion in the lamprey. Brain Res 293: 164–167, 1984.[CrossRef][ISI][Medline]
Gao BX, Stricker C, and Ziskind-Conhaim L. Transition from GABAergic to glycinergic synaptic transmission in newly formed spinal networks. J Neurophysiol 86: 492–502, 2001.
Gao BX and Ziskind-Conhaim L. Development of glycine- and GABA-gated currents in rat spinal motoneurons. J Neurophysiol 74: 113–121, 1995.
Gossard JP, Brownstone RM, Barajon I, and Hultborn H. Transmission in a locomotor-related group Ib pathway from hindlimb extensor muscles in the cat. Exp Brain Res 98: 213–228, 1994.[ISI][Medline]
Goulding M and Lamar E. Neuronal patterning: Making stripes in the spinal cord. Curr Biol 10: R565–568, 2000.[CrossRef][ISI][Medline]
Goulding M, Lanuza G, Sapir T, and Narayan S. The formation of sensorimotor circuits. Curr Opin Neurobiol 12: 508–515, 2002.[CrossRef][ISI][Medline]
Graham-Brown T. On the fundamental activity of the nervous centres: together with an analysis of the conditioning of rhythmic activity in progression, and a theory of the evolution of function in the nervous system. J Physiol 48: 18–41, 1914.
Grillner S and Wallen P. Central pattern generators for locomotion, with special reference to vertebrates. Annu Rev Neurosci 8: 233–261, 1985.[CrossRef][ISI][Medline]
Hartley R, Todd AJ, Jessell TM, Lieberam I, and Maxwell DJ. Characterization of Hb9-expressing interneurons in the mouse spinal cord. Soc Neurosci Abstr 186.9, 2003.
Higashijima S, Masino MA, Mandel G, and Fetcho JR. Engrailed-1 expression marks a primitive class of inhibitory spinal interneuron. J Neurosci 24: 5827–5839, 2004.
Hinckley CA, Lieberam I, Jessell TM, and Ziskind-Conhaim L. Identification of interneurons involved in locomotor-like rhythms in the Hb9-GFP mouse. Soc Neurosci Abstr 277.13, 2003.
Hinckley CA, Seebach B, and Ziskind-Conhaim L. Distinct roles of glycinergic and GABAergic inhibition in coordinating locomotor-like rhythms in the neonatal mouse spinal cord. Neurosci In press.
Hochman S and Schmidt BJ. Whole cell recordings of lumbar motoneurons during locomotor-like activity in the in vitro neonatal rat spinal cord. J Neurophysiol 79: 743–752, 1998.
Hultborn H, Conway BA, Gossard JP, Brownstone R, Fedirchuk B, Schomburg ED, Enriquez-denton M, and Perreault MC. How do we approach the locomotor network in the mammalian spinal cord? Ann NY Acad Sci 860: 70–82, 1998.
Jankowska E. Spinal interneuronal systems: identification, multifunctional character and reconfigurations in mammals. J Physiol 533: 31–40, 2001.
Jessell TM. Neuronal specification in the spinal cord: inductive signals and transcriptional codes. Nat Rev Genet 1: 20–29, 2000.[CrossRef][ISI][Medline]
Jiang Z, Carlin KP, and Brownstone RM. An in vitro functionally mature mouse spinal cord preparation for the study of spinal motor networks. Brain Res 816: 493–499, 1999.[CrossRef][ISI][Medline]
Kiehn O and Butt SJ. Physiological, anatomical and genetic identification of CPG neurons in the developing mammalian spinal cord. Prog Neurobiol 70: 347–361, 2003.[CrossRef][ISI][Medline]
Kiehn O and Kjaerulff O. Spatiotemporal characteristics of 5-HT and dopamine-induced rhythmic hindlimb activity in the in vitro neonatal rat. J Neurophysiol 74: 1472–1482, 1996.
Kiehn O and Kjaerulff O. Distribution of central pattern generators for rhythmic motor outputs in the spinal cord of limbed vertebrates. Ann NY Acad Sci 860: 110–129, 1998.
Kiehn O, Kjaerulff O, Tresch MC, and Harris-Warrick RM. Contributions of intrinsic motor neuron properties to the production of rhythmic motor output in the mammalian spinal cord. Brain Res Bull 53: 649–659, 2000.[CrossRef][ISI][Medline]
Kjaerulff O and Kiehn O. Distribution of networks generating and coordinating locomotor activity in the neonatal rat spinal cord in vitro: a lesion study. J Neurosci 16: 5777–5794, 1996.
Kremer E and Lev-Tov A. Localization of the spinal network associated with generation of hindlimb locomotion in the neonatal rat and organization of its transverse coupling system. J Neurophysiol 77: 1155–1170, 1997.
Kudo N and Yamada T. N-methyl-D,L-aspartate-induced locomotor activity in a spinal cord-hindlimb muscles preparation of the newborn rat studied in vitro. Neurosci Lett 75: 43–48, 1987.[CrossRef][ISI][Medline]
Kullander K, Butt SJ, Lebret JM, Lundfald L, Restrepo CE, Rydstorm A, and Klein O. Role of EphA4 and EphrinB3 in local neuronal circuits that control walking. Science 299: 1889–1892, 2003.
Lanuza GM, Gosgnach S, Pierani A, Jessell TM, and Goulding M. Genetic identification of spinal interneurons that coordinate left-right locomotor activity necessary for walking movements. Neuron 42: 375–386, 2004.[CrossRef][ISI][Medline]
Lev-Tov A, Delvolve I, and Kremer E. Sacrocaudal afferents induce rhythmic efferent bursting in isolated spinal cords of neonatal rats. J Neurophysiol 83: 888–894, 2000.
MacLean JN, Hochman S, and Magnuson DS. Lamina VII neurons are rhythmically active during locomotor-like activity in the neonatal rat spinal cord. Neurosci Lett 197: 9–12, 1995.[CrossRef][ISI][Medline]
McCrea DA. Spinal circuitry of sensorimotor control of locomotion. J Physiol 533: 41–50, 2001.
Odden JP, Holbrook S, and Doe CQ. Drosophila HB9 is expressed in a subset of motoneurons and interneurons, where it regulates gene expression and axon pathfinding. J Neurosci 22: 9143–9149, 2002.
Parker D. Activity-dependent feedforward inhibition modulates synaptic transmission in a spinal locomotor network. J Neurosci 23: 11085–11093, 2003.
Parker D and Grillner S. The activity-dependent plasticity of segmental and intersegmental synaptic connections in the lamprey spinal cord. Eur J Neurosci 12: 2135–2146, 2000.[CrossRef][ISI][Medline]
Pierani A, Brenner-Morton S, Chiang C, and Jessell TM. A sonic hedgehog-independent, retinoid-activated pathway of neurogenesis in the ventral spinal cord. Cell 97: 903–915, 1999.[CrossRef][ISI][Medline]
Pierani A, Moran-Rivard L, Sunshine MJ, Littman DR, Goulding M, and Jessell TM. Control of interneuron fate in the developing spinal cord by the progenitor homeodomain protein Dbx 1. Neuron 29: 367–384, 2001.[CrossRef][ISI][Medline]
Raastad M, Johnson BR, and Kiehn O. Analysis of EPSCs and IPSCs carrying rhythmic, locomotor-related information in the isolated spinal cord of the neonatal rat. J Neurophysiol 78: 1851–1859, 1997.
Roberts A, Soffe SR, Wolf ES, Yoshida M, and Zhao FY. Central circuits controlling locomotion in young frog tadpoles. Ann NY Acad Sci 860: 19–34, 1998.
Shirasaki R and Pfaff SL. Transcriptional codes and the control of neuronal identity. Annu Rev Neurosci 25: 251–281, 2002.[CrossRef][ISI][Medline]
Soffe SR. Roles of glycinergic inhibition and N-methyl-D-aspartate receptor mediated excitation in the locomotor rhythmicity of one half of the Xenopus embryo central nervous system. Eur J Neurosci 1: 561–571, 1989.[CrossRef][ISI][Medline]
Thaler J, Harrison K, Sharma K, Lettieri K, Kehrl J, and Pfaff SL. Active suppression of interneuron programs within developing motor neurons revealed by analysis of homeodomain factor HB9. Neuron 23: 675–687, 1999.[CrossRef][ISI][Medline]
Wenner P, O'Donovan MJ, and Matise MP. Topographical and physiological characterization of interneurons that express engrailed-1 in the embryonic chick spinal cord. J Neurophysiol 84: 2651–2657, 2000.
Whelan P, Bonnot A, and O'Donovan MJ. Properties of rhythmic activity generated by the isolated spinal cord of the neonatal mouse. J Neurophysiol 84: 2821–2833, 2000.
). Also visible are GFP+ motoneurons in the lateral and medial columns (LMC and MMC, respectively), autonomic motor column (AMC), and widely distributed interneurons. CC, central canal. The projected image consisted of 61 sequential optical sections (0.5-µm interval Z steps). B: primary ventral and dorsal dendrites and dendrites crossing the midline are visible at higher magnification (
). The projected image consists of 71 sequential optical sections (0.5-µm interval Z steps). C: the organization of GFP+ interneurons in L5 is different than that observed in L2. Abundant GFP+ interneurons without an obvious neuronal clustering were apparent along the midline in segment L5. Note that the medial interneurons were larger than those in L2. The image is a superimposition of 59 confocal optical sections. Images are of native GFP.
, 3 possible contact points seen in a single optical section (not shown). Scale bar: 10 µm. C: neurobiotin-filled projections in motoneuron dendritic field between the lateral and medial motor columns. Projections in this slice extended to the lateral perimeter of the lateral motor column (not shown). The projected image consists of 27 confocal optical sections (0.5 µm Z steps). D: a single confocal optical section, at the site marked by 
, the cycle period as it was determined by smoothed and rectified traces (not shown). B: histogram of EPSC and IPSC frequency (mean ± SE) as a function of normalized cycle period. EPSC frequency significantly increased during the 1st half of the cycle period (P < 0.01, n = 9 interneurons), which corresponded to the duration of ventral root burst. IPSC frequency (<0.5 Hz) did not significantly change during the cycle period (P = 0.84). Bar below the x axis represents the average duration of ventral root burst ± SD.