| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Zoology, University of Oklahoma, Norman, Oklahoma
Submitted 2 March 2005; accepted in final form 5 September 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Pearson 1993). In vertebrates, the situation is less clear, and the evidence is largely circumstantial, especially for the neuronal networks controlling limb movements (Bekoff et al. 1987; Berkinblit et al. 1978; Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b; Carter and Smith 1986; Currie and Stein 1989; Earhart and Stein 2000a,b; Juranek and Currie 2000; Mortin and Stein 1989; Perreault et al. 1999; Smith et al. 1986).
The turtle spinal cord is a convenient model system with which to address this question. It can generate multiple forms of hindlimb scratching, hindlimb swimming, and hindlimb withdrawal (flexion reflex) motor patterns, even in the absence of input from the brain and movement-related sensory feedback (Currie and Stein 1989; Juranek and Currie 2000; Lennard and Stein 1977; Mortin et al. 1985; Robertson et al. 1985; Stein et al. 1982). Previous research has shown that many individual spinal cord interneurons are rhythmically active during all three forms of ipsilateral fictive scratching (Berkowitz 2001b, 2002a,; Berkowitz and Stein 1994b). In addition, many spinal interneurons are rhythmically active during both ipsilateral and contralateral fictive scratching (Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b). Moreover, many scratch-activated spinal interneurons are also activated during fictive forward swimming and fictive hindlimb withdrawal (Berkowitz 2002a). These studies suggest that at least some spinal interneurons contribute to the generation of motor patterns for multiple, diverse hindlimb behaviors.
One concern, however, has been that correlations between single-neuron activity and motor output do not demonstrate that any of these neurons actually affect motor output for any behavior. For ascending spinal interneurons, a reasonable alternative hypothesis is that these neurons provide feedback signals to higher motor control centers in the brain as has been shown for cat ventral spinocerebellar tract neurons (Arshavsky et al. 1972, 1978, 1984) but have no direct effect on spinal motor output. Some rhythmically active spinal interneurons may mediate primary afferent depolarization (Rudomin and Schmidt 1999) and have no direct effects on motor output. If, however, some spinal interneurons do directly affect motor output for multiple hindlimb behaviors, they should have axon terminals within the ventral horn of the spinal cord hindlimb enlargement. Several physiological types of turtle spinal cord interneurons have previously been studied using a combination of intracellular recording and staining (Fernandez et al. 1996; Hounsgaard and Kjaerulff 1992; McDonagh et al. 1998, 2002; Russo and Hounsgaard 1996a,b), but those experiments were conducted using slice preparations that do not produce scratch rhythms. To test the prediction that interneurons active during multiple forms of scratching and withdrawal have axon terminals within the ventral horn of the spinal cord hindlimb enlargement, I recorded intracellularly from interneurons that were active during multiple, diverse hindlimb motor patterns in vivo and stained each with Neurobiotin to reveal its morphology. Preliminary results have been reported previously in an abstract (Berkowitz 2003).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Adult red-eared turtles (Trachemys scripta elegans, 500–1,200 g, both sexes, n = 29) were prepared for recording as previously described (Berkowitz 2001a; Robertson et al. 1985). Briefly, animals were anesthetized and surgically dissected to transect the spinal cord between the dorsal 2 (D2) and D3 postcervical segments, expose the spinal cord between the D6 and sacral 2 (S2) segments, and prepare several hindlimb motor nerves for electroneurographic recordings (ENGs), including the right and left hip flexor nerves, ventral puboischiofemoralis internus, pars anteroventralis (HF) and, in some experiments, the right knee extensor nerve, triceps femoralis, pars femorotibialis (Fig. 1). After surgery, the animal was immobilized with gallamine triethiodide (8 mg/kg im, Sigma-Aldrich, St. Louis, MO) and artificially respirated at room temperature for the duration of recording. All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Oklahoma.
|
Intracellular recordings (n = 47 scratch-activated interneurons with soma-dendritic morphology recovered) were obtained from the gray matter on the right side of one or more of the five segments of the hindlimb enlargement (D8, D9, D10, S1, and S2), using sharp microelectrodes made from thin-walled borosilicate glass with filament (Sutter Instrument Company, Novato, CA), fabricated with a P-97 puller (Sutter) and filled with 4% Neurobiotin (Vector Laboratories, Burlingame, CA) dissolved in 1 M KCl, with resistances of 60–120 M
. The meninges were torn on the dorsal surface of the spinal cord at the site of each intracellular microelectrode penetration. The electrode was advanced using an Inchworm piezoelectric motorized microdrive (Burleigh Instruments, Fishers, NY); impalements were often facilitated using 5-ms negative capacitance "tickle" pulses delivered by the intracellular amplifier (World Precision Instruments, Sarasota, FL), triggered by a Master-8 stimulator (A.M.P.I., Jerusalem, Israel). Membrane potentials and spike heights were each typically within the range of 40–70 mV. In some cases, 0.1- to 0.2-nA constant hyperpolarizing current was injected to stabilize the recording. After each successful intracellular recording, sinusoidal (4 Hz) current with a peak depolarization of 0.1–3.0 nA was injected for 1–30 min to eject Neurobiotin. No more than one interneuron fill was attempted in each spinal cord segment to ensure unambiguous identification. ENGs were obtained using pairs of 100-µm silver wires in a pool of mineral oil and were amplified (1,000x) and filtered (100–1,000 Hz) using differential AC amplifiers (A-M Systems, Carlsborg, WA). Fictive scratching and fictive withdrawal were elicited by mechanical stimulation of the shell or skin with a fire-polished glass probe attached to a force transducer (Grass-Telefactor, West Warwick, RI). All recordings were stored on an RD-145 digital audio tape recorder (TEAC America, Montebello, CA), displayed on a DL 716 Scopecorder (Yokagawa Corporation of America, Newnan, GA), and redigitized and analyzed off-line using Datapac 2000 software (Run Technologies, Laguna Hills, CA).
Histology
Immediately after the recording session, the animal was killed with pentobarbital (390 mg ip); after 
1 h, the heart was exposed and perfused with 800 ml turtle saline containing 0.1% sodium nitrite, 10 units/ml heparin, and 39 mg pentobarbital, followed by 300 ml chilled 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4. The spinal cord was postfixed overnight at 4°C, cryoprotected in 20% sucrose/PB, embedded in a gelatin-albumin-glutaraldehyde-PB medium, and frozen-sectioned at 100 µm horizontally. Sections were rinsed 3 x 5 min in phosphate-buffered saline (PBS), incubated 30 min in 0.3% H2O2/PBS, rinsed 4 x 10 min in PBS, incubated overnight at 4°C in 1:100 ABC solution (Vector)/0.3% Triton X-100/PBS, rinsed 4 x 10 min in PBS, reacted 4–7 min in 0.5x Vector SG solution/PBS, and rinsed 3 x 10 min in PBS; all steps were done on a rotator. Sections were mounted, air-dried, counterstained in Fast Nuclear Red (Vector), dehydrated in graded ethanols, cleared in Histoclear (National Diagnostics, Atlanta, GA), and mounted with Permount (Fisher Chemicals, Fairlawn, NJ).
Morphological analysis
Neurons were included in this study only if they were activated or rhythmically modulated during fictive scratching, one and only one stained soma with attached stained processes was recovered at the expected site, and, for ventral horn neurons, the processes were darkly stained and yet no stained process was seen in a ventral root (to eliminate motoneurons from the study). Reconstructions were made via a camera lucida attached to an Optiphot-2 microscope (Nikon Instruments, Melville, NY); digital photomicrographs were obtained with a DP-70 camera (Olympus America, Melville NY) mounted on this microscope. Photomontages were assembled using Photoshop (Adobe Systems, San Jose, CA).
Physiological analysis
The modulation of each interneuron’s firing rate was quantitatively analyzed separately for each form of ipsilateral fictive scratching in which there were clear, alternating bursts and silent periods of the ipsilateral hip flexor motor nerve, at least two such cycles of fictive scratching occurred, and the interneuron fired 10 action potentials during these cycles. In each such case, a dual-referent phase histogram (with 5 bins for the hip flexor burst and 5 bins for the hip flexor interburst interval) was calculated with respect to the rectified and smoothed bursts and interburst intervals of the ipsilateral hip flexor nerve, using Datapac 2000 (Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b). The phase histogram data were then used to calculate the mean vector (Batschelet 1981; Berkowitz and Stein 1994b; Drew and Doucet 1991; Mardia 1972). The mean vector length is a measure of the degree of rhythmic modulation of a neuron’s activity with respect to a cycle; it can vary from 0 to 1 with 0 indicating minimal rhythmicity and 1 indicating maximal rhythmicity. The mean vector angle is a measure of the phase preference of a neuron within the hip flexor cycle; it can vary from 0 to 1 (or 0–360°) with 0 to 0.5 (or 0–180°) indicating a phase preference during hip flexor activity and 0.5 to 1 (or 180–360°) indicating a phase preference within the hip flexor interburst intervals (Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b). The null hypothesis that the interneuron’s firing occurred at random with respect to the hip flexor cycle was evaluated using the Rayleigh test (Batschelet 1981; Berkowitz and Stein 1994b; Drew and Doucet 1991; Mardia 1972). Mean vector angles were used in analyses only if the phase histogram passed the Rayleigh test with P < 0.01 (Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b). Most cells were active during multiple forms of fictive scratching, so multiple phase histograms were constructed for each cell; the phase histogram used for measurements of mean vector length, mean vector angle, and peak firing rate (Figs. 8, A–C, and 9D) was the one that passed the Rayleigh test with the lowest P value, unless multiple phase histograms had P < 0.001, in which case the one among these having the highest mean vector length was used (Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b). For each cell, the peak firing rate used (Fig. 8A) was the highest mean firing rate of the 10 values (10 bins) in this phase histogram.
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
All 47 cells were active during at least two of the three forms of ipsilateral fictive scratching and most were active during all three forms of scratching (Figs. 3A, 5A, and 6A and Table 1). Each interneuron showed some degree of rhythmic modulation of its spike rate and/or its membrane potential, phase-locked to each scratch rhythm with which it was active (Figs. 3A, 4A, 5A, 6A, and 7A). The strength of rhythmic modulation during fictive scratching, as assessed for firing rate by the mean vector length (Fig. 8B) and for membrane potential by the amplitude of the averaged membrane potential oscillation (Fig. 9B), varied substantially among neurons, from extremely rhythmic (e.g., Fig. 6A) to only slightly rhythmic (e.g., Fig. 7A).
|
|
|
|
|
|
), which is the same neuron as is illustrated in Fig. 5. The conductance increases were more subtle in other neurons, such as the one shown in Fig. 10B, which is the same neuron as is illustrated in Fig. 7. While conductance increases were evident in all phases of fictive scratching, the amplitude of the conductance increases could vary with the phase of the scratch rhythm (indicated by the bursts and quiescent intervals of the ipsilateral hip flexor motor nerve, iHF). For example, for the neuron shown in Figs. 5 and 10A, the conductance increases were smallest during the transitions from hyperpolarization to depolarization and back (Fig. 10A3, ).
|
) and extracellular recording of spinal interneurons with D8–S2 somata (Berkowitz 2001b). Nonetheless, a few neurons exhibited depolarization and spiking in substantially different phases of the iHF rhythm during different forms of scratching. For example, the neuron shown in Fig. 3 depolarized and fired action potentials during iHF quiescence in rostral scratching (Fig. 3A1), during late iHF quiescence and early iHF activity in pocket scratching (Fig. 3A2), and during late iHF activity in caudal scratching (Fig. 3A3). The peak phases (and the trough phases) of the membrane potential also varied considerably among neurons. The trough phase occurred about half a cycle after the peak phase on average (Fig. 9C). However, the mean phase difference between the peak and the trough (a measure that can vary only from 0 to 0.5; see METHODS) was 0.38 ± 0.01 (SE); thus the peak and trough occurred in substantially different phases of the hip flexor cycle, but were not entirely out of phase in all cells. The preferred phase of firing was similar to the phase of peak depolarization in most cells (Fig. 9D); the difference between these phases was on average 0.12 ± 0.02 (SE).
In addition to being active during multiple forms of ipsilateral fictive scratching, 43 (91%) of these interneurons were rhythmically active during both ipsilateral and contralateral fictive scratching (Figs. 4A, 5A, 6A, and 7A and Table 1). During unilateral fictive scratching, bursts of the iHF and the contralateral hip flexor nerve (cHF) typically alternate with each other, with much larger iHF bursts during ipsilateral fictive scratching and much larger cHF bursts during contralateral fictive scratching (e.g., Fig. 5A, 1–3, and 6A, 4 and 5) (see also Berkowitz and Stein 1994b; Currie and Stein 1989; Stein et al. 1995). Some interneurons were depolarized and fired action potentials in the same phase of the iHF rhythm during ipsilateral and contralateral scratching (Figs. 5A and 6A), while other interneurons depolarized and fired in different phases during ipsilateral and contralateral scratching (Fig. 4A).
Thirty-two (68%) of these same interneurons also fired during ipsilateral fictive hindlimb withdrawal (flexion reflex; Figs. 3A4, 5A4, and 7A2 and Table 1). This was true even for several neurons that were active in fictive scratching mainly when the iHF was quiescent (e.g., Fig. 5A). An additional six cells (13%) displayed subthreshold depolarization and seven others (15%) displayed hyperpolarization alone during ipsilateral fictive withdrawal (Fig. 6A3 and Table 1). In contrast, only 14 cells (30%) fired or were depolarized during contralateral fictive withdrawal and four others (9%) displayed hyperpolarization alone during contralateral fictive withdrawal (Table 1).
Intracellular staining
The soma and processes of the scratch-activated interneurons included in this study were stained black, typically with the dendritic processes contiguously stained (Figs. 3B, 4B, 5B, 6B, and 7B). The somata of scratch-activated interneurons were found in the dorsal horn, usually deeply (Figs. 2 and 3), in the intermediate zone, usually in its lateral half (Figs. 2, 4, 5, and 7), and in the ventral horn, usually in its dorsal half (Figs. 2 and 6). The dendrites often extended laterally completely across the ipsilateral lateral funiculus to the edge of the spinal cord (Figs. 3B1, 4B1, 5B1, and 6B1), as has been seen previously for turtle spinal motoneurons (Berkowitz and Stein 1994c; McDonagh et al. 1998, 2002; Ruigrok et al. 1984; Skydsgaard and Hounsgaard 1994) and interneurons (Berkowitz and Stein 1994c; Fernandez et al. 1996; McDonagh et al. 1998, 2002); in several cases, stained dendrites even appeared to penetrate and extend beyond the pia (not shown). In contrast, axons often showed patchy staining, in which black or gray axonal segments alternated with unstained segments along the visible length of the axon, as has been observed previously with intracellular injection of biotinylated compounds (McDonagh et al. 2002). In the best-stained cases, stained funicular axons could be followed 1–2 mm from the soma. Axons were usually so thin (<<1 µm) along most of their visible length that the axon diameter could not be measured precisely.
This report focuses on the axonal projections and terminations of these interneurons; correlations between soma-dendritic parameters and physiological parameters (Berkowitz et al. 2004, 2005) will be described separately. Some spinal neurons with endogenous biotin staining (Berkowitz 2002b) were also observed in these sections, but such neurons were always stained gray instead of black and never had more than the soma and the most proximal dendrites stained. In contrast, Neurobiotin-stained cells were always much more darkly stained with numerous distant and thin processes stained and occurred only at the expected locations; thus it was never difficult to distinguish Neurobiotin-stained neurons from endogenously stained cells.
Intracellularly stained axons were distinguishable from stained dendrites using one or more of the following criteria: axons ascended or descended rostrocaudally for a long distance, usually within a funiculus; axons either did not branch or gave off individual, thin collaterals at approximately right angles (Figs. 4B1, 5B1, 6B1, and 7B, 1 and 2), in contrast to dendrites, which branched repeatedly and at a variety of oblique angles (Figs. 3B1, 5B1, 6B1, and 7B1); and axon collaterals terminated, almost always within the gray matter, by branching into extremely thin processes, near the limits of visibility, with clusters of black, ovoid varicosities along their lengths and at their tips, like grapes on a vine (Figs. 3B3, 4B3, 5B3, 6B3, and 7B3). If a process could not be unambiguously identified as an axon by these criteria, the process was not assumed to be an axon. The somata of cells with axon terminals stained, cells with funicular axons stained but no detectable terminals, and cells with neither funicular axons nor terminals stained were interspersed in the same regions of the gray matter (Fig. 2); thus it seems likely that these groups of cells simply differed in how completely their axons were stained and do not represent distinct morphological or functional populations of interneurons.
Axon trajectories and terminals of scratch-activated interneurons
For 33 (70%) of the scratch-activated interneurons studied here, either a stained axon could be seen entering and traveling rostrocaudally within one or more funiculi, or axon-terminal arborizations could be identified within the gray matter, or both (Fig. 2 and Table 1). (Some axon-terminal arborizations were also seen in the white matter, usually adjacent to the gray matter, but such terminals were not studied systematically.) Interneurons with identifiable axon terminal arborizations within the gray matter appeared to be representative of the recorded population of scratch-activated interneurons with respect to soma location (Fig. 2), peak firing rate during fictive scratching (Fig. 8A), degree of rhythmic modulation (mean vector length) during fictive scratching (Fig. 8B), firing rate phase preference within the hip flexor activity cycle during fictive scratching (Fig. 8C), and amplitude of phase-normalized membrane potential oscillations (Fig. 9B).
Figures 3–7 illustrate the range and diversity of morphological properties of typical scratch-related interneurons. Figure 3 shows an interneuron that was rhythmically activated during all three forms of ipsilateral scratching as well as ipsilateral withdrawal (Fig. 3A); it had axon branches ascending in both the ipsilateral lateral funiculus and the contralateral ventral funiculus (Fig. 3B1) as well as axon-terminal arborizations in the local ipsilateral dorsal horn and intermediate zone (Fig. 3B3). Figure 4 shows an interneuron that was rhythmically active during both ipsilateral and contralateral scratching (Fig. 4A) as well as ipsilateral withdrawal (not shown); its axon ascended and descended within the ipsilateral gray matter (and perhaps the lateral funiculus, further caudally; Fig. 4B1), giving off collaterals with abundant terminal arborizations in the ipsilateral intermediate zone and ventral horn (Fig. 4B, 1 and 3). Figure 5 shows an interneuron that was rhythmically active during multiple forms of ipsilateral scratching (Fig. 5A, 1 and 2) and contralateral scratching (Fig. 5A3), as well as being excited to spike during hindlimb withdrawal, both ipsilaterally (Fig. 5A4) and contralaterally (not shown); its axon descended in the ipsilateral lateral funiculus and gave off short collaterals with terminal arborizations in the ipsilateral ventral horn and the immediately adjacent lateral funiculus (Fig. 5B, 1 and 3). Figure 6 shows an interneuron that was activated in brief, high-frequency bursts during multiple forms of ipsilateral (Fig. 6A, 1 and 2) and contralateral (Fig. 6A, 4 and 5) scratching as well as being depolarized during withdrawal, both ipsilaterally (Fig. 6A3) and contralaterally (not shown); its axon bifurcated into ascending and descending branches within the ipsilateral ventral funiculus (Fig. 6B1) and had local axon-terminal arborizations in the ipsilateral ventral horn (Fig. 6B3). Figure 7 shows an interneuron that was strongly activated but with weak rhythmic modulation, during ipsilateral scratching (Fig. 7A1) as well as being activated during contralateral scratching (Fig. 7A, 3 and 4) and ipsilateral withdrawal (Fig. 7A2); its axon crossed into the contralateral ventral funiculus, where it bifurcated into ascending and descending branches and gave off axon collaterals with terminal arborizations in the contralateral ventral horn (Fig. 7B, 1–3).
Despite the great morphological diversity of scratch-activated interneurons, there were several repeated patterns of axonal trajectories (Table 1). Of the 30 interneurons the axon trajectories of which could be seen, 7 (23%) had an axon in the ipsilateral lateral funiculus, 7 (23%) had an axon in the ipsilateral ventral funiculus, and 14 (47%) had an axon that crossed and traveled in the contralateral ventral funiculus; in addition, 4 (13%) only had an axon that remained within the ipsilateral gray matter and 1 only had an axon within the contralateral gray matter. None of the scratch-activated interneurons stained had an axon in the dorsal funiculus, either ipsilaterally or contralaterally, and no axons traveled in the contralateral lateral funiculus. Sixteen (53%) of the 30 axons with identifiable trajectories bifurcated into ascending and descending branches; this included 3 of the 7 ipsilateral lateral funiculus axons, 4 of the 7 ipsilateral ventral funiculus axons, 3 of the 4 ipsilateral gray matter axons, 5 of the 14 contralateral ventral funiculus axons, and the 1 contralateral gray matter axon.
Of the 24 scratch-activated interneurons with identifiable axon-terminal arborizations, 21 (88%) terminated in the ventral horn, including 16 (84%) of the 19 cells with ipsilateral terminals and all 5 (100%) of the cells with contralateral terminals (Table 1). Axon-terminal arborizations in the ipsilateral ventral horn were seen for 12 (71%) of the 17 neurons with identifiable ipsilateral axon trajectories and arborizations in the contralateral ventral horn were seen for 5 (33%) of the 15 neurons with identifiable contralateral axon trajectories.
All 16 (100%) of the scratch-activated interneurons with ipsilateral ventral horn axon terminals were activated during multiple forms of ipsilateral fictive scratching (Table 1). Fourteen (88%) of these 16 cells were also active during contralateral fictive scratching. During ipsilateral fictive withdrawal, 11 (69%) of these 16 cells fired, an additional 3 cells displayed subthreshold depolarization, and 1 displayed hyperpolarization alone. In contrast, during contralateral withdrawal, only two of these cells fired, two additional cells displayed subthreshold depolarization, and one displayed hyperpolarization alone.
All five scratch-activated interneurons with contralateral ventral horn axon terminals were activated during multiple forms of ipsilateral fictive scratching and multiple forms of contralateral fictive scratching. Four of these five cells also fired during ipsilateral withdrawal. During contralateral withdrawal, one of these cells fired and an additional one displayed subthreshold depolarization.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
The primary finding of this study was that many spinal interneurons that are rhythmically active during multiple forms of fictive scratching have axon-terminal arborizations in the ventral horn of the spinal cord hindlimb enlargement and thus are not simply relaying corollary discharge signals to the brain. This finding strongly suggests that many scratch-activated spinal interneurons have direct effects on hindlimb motoneurons and/or hindlimb motor pattern-generating circuitry for multiple forms of scratching. This finding provides some of the strongest evidence to date that individual interneurons contribute to more than one type of rhythmic limb movement in vertebrates. Some of these interneurons may themselves contribute to pattern generation for multiple behaviors, while others may relay the outputs of pattern-generating circuitry to motoneurons.
This evidence strengthens the conclusions of a large set of studies that relied on more circumstantial evidence, such as extracellular single-unit recordings correlated with fictive motor patterns, to suggest that some interneurons are shared components of the interneuronal networks underlying distinct types of limb movements, in turtles (Berkowitz 2001b, 2002a; Berkowitz and Stein 1994b; Currie and Stein 1989; Earhart and Stein 2000a,b; Juranek and Currie 2000; Mortin and Stein 1989), chicks (Bekoff et al. 1987), and cats (Arshavsky et al. 1972, 1978, 1984; Berkinblit et al. 1978; Carter and Smith 1986; Deliagina and Feldman 1981; Deliagina and Orlovsky 1980; Feldman and Orlovsky 1975; Perreault et al. 1999; Pratt and Jordan 1987; Smith et al. 1986). Similar types of studies have also suggested that vertebrate interneuronal networks controlling multiple forms of breathing, vocalization, mastication, and swallowing in mammals (Grelot et al. 1993; Larson et al. 1994; Lieske et al. 2000; Oku et al. 1994; Westberg et al. 1998; Yajima and Larson 1993) and axial swimming and escape movements in fish (Svoboda and Fetcho 1996) involve some shared interneurons. Moreover, paired intracellular recordings of interneurons and motoneurons have directly demonstrated that individual spinal interneurons contribute to generating multiple forms of axial locomotion in embryonic tadpoles (Soffe 1993; Soffe et al. 1984). Taken together, these studies strongly suggest that there is some sharing among the vertebrate neuronal networks that generate diverse behaviors involving the same muscles, as has been demonstrated most clearly for invertebrates (Marder and Calabrese 1996; Pearson 1993).
Interneurons contribute to scratching bilaterally
A second finding of this study was that many spinal interneurons that are rhythmically active during both ipsi- and contralateral fictive scratching have axon-terminal arborizations in the ventral horn of the spinal cord hindlimb enlargement. This finding strongly suggests that many scratch-activated interneurons have direct effects on motor output for both right and left hindlimb scratching. Some have axon terminals in the ipsilateral ventral horn and others in the contralateral ventral horn. Previous work has shown that although scratching is essentially a unilateral behavior, important contributions to scratch motor pattern generation nonetheless come from the contralateral spinal cord (Berkowitz and Stein 1994b; Stein et al. 1995, 1998). Interneurons with the activity and morphology seen in this study may mediate contralateral contributions to ipsilateral scratching.
Interneurons contribute to scratching and withdrawal
A third finding was that spinal interneurons that are activated during both fictive scratching and fictive withdrawal have ventral horn axon terminals and thus are likely to have direct effects on hindlimb motor output. It was much more common for such cells to be activated during ipsilateral withdrawal than contralateral withdrawal. Previous research showed that spinal interneurons active during both fictive scratching and fictive swimming ("scratch/swim" neurons) have a high probability of also being activated during ipsilateral withdrawal, while interneurons active during scratching but suppressed during swimming ("scratch-specialized" neurons) are less likely to be activated during withdrawal (Berkowitz 2002a). Taken together with the current results, it seems likely that at least some scratch/swim neurons have axon terminals in the spinal cord ventral horn, but further experiments will be required to test this hypothesis.
Interneuronal synaptic inputs
A fourth finding was that scratch-activated interneurons typically exhibited alternating de- and hyperpolarizing waves and substantial conductance increases throughout fictive scratching, suggesting that they receive alternating excitatory and inhibitory scratch-related synaptic inputs. The peaks and troughs of these membrane potential oscillations were nearly half a cycle apart in most cells. Similar de- and hyperpolarizing waves and conductance increases have been described in turtle motoneurons during in vivo fictive scratching (Robertson and Stein 1988) and in turtle spinal motoneurons and interneurons during in vitro turtle rostral scratch motor patterns (Alaburda et al. 2005). In the current study, cells with very rhythmic activity often had hyperpolarizing waves that were larger than the depolarizing waves. This result adds to the previous finding from single-unit recording that a cell’s mean firing rate and its degree of rhythmic modulation were negatively correlated (Berkowitz and Stein 1994b). Together, these findings suggest that rhythmic inhibition may play a large role in shaping the rhythmic activity of turtle spinal interneurons.
Axon trajectories
Finally, there was great diversity in the axon trajectories of scratch-activated interneurons. About half could be seen to bifurcate into ascending and descending branches, and the total numbers of ascending and descending axons were approximately equal. A recent retrograde labeling study (Nissen et al. 2004) indicates that a minority of turtle D9 segment interneurons have bifurcating axons, so scratch-activated interneurons may be more likely than other interneurons to have bifurcating axons. Because many of the interneurons in the current study had a descending axon, this set of interneurons should be largely overlapping with the descending propriospinal neurons whose single-unit activities and axon projections have been studied previously (Berkowitz and Stein 1994a–c; Stein and Daniels-McQueen 2002, 2003); moreover, the soma locations of the scratch-activated interneurons studied here (Fig. 2) were quite similar to the soma locations of descending propriospinal neurons (Berkowitz and Stein 1994c). The axons of the scratch-related interneurons studied here traveled in either the lateral funiculus or the ventral funiculus ipsilaterally but exclusively in the ventral funiculus contralaterally. Within this group of scratch-activated interneurons, there were no obvious correlations between a neuron’s physiological behavior and its axon trajectory. The axon trajectories of turtle scratch-related spinal interneurons are generally consistent with the axon trajectories of spinal interneurons involved in swimming in the lamprey (Buchanan 1996; Grillner et al. 1991), zebrafish (Fetcho 1991; Lewis and Eisen 2003), and embryonic tadpole (Li et al. 2001; Roberts and Clarke 1982), but there are apparently fewer morphological types in those cases, and the morphologies seen are more clearly correlated with particular physiological roles. These differences may be due to the greater complexity of the spinal cord in adult, limbed vertebrates. Correlations between the soma-dendritic morphologies of the turtle scratch-related spinal interneurons and their physiological behavior during fictive scratching (Berkowitz et al. 2004, 2005) will be examined in more detail in the near future.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: A. Berkowitz, Dept. of Zoology, University of Oklahoma, Norman, OK 73019 (E-mail: ari{at}ou.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Arshavsky YI, Berkinblit MB, Fukson OI, Gelfand IM, and Orlovsky GN. Origin of modulation in neurons of the ventral spinocerebellar tract during locomotion. Brain Res 43: 276–279, 1972.[CrossRef][ISI][Medline]
Arshavsky YI, Gelfand IM, Orlovsky GN, and Pavlova GA. Messages conveyed by spinocerebellar pathways during scratching in the cat. II. Activity of neurons of the ventral spinocerebellar tract. Brain Res 151: 493–506, 1978.[CrossRef][ISI][Medline]
Arshavsky YI, Gelfand IM, Orlovsky GN, Pavlova GA, and Popova LB. Origin of signals conveyed by the ventral spino-cerebellar tract and spino-reticulo-cerebellar pathway. Exp Brain Res 54: 426–431, 1984.[ISI][Medline]
Batschelet E. Circular Statistics in Biology. London: Academic, 1981.
Bekoff A, Nusbaum MP, Sabichi AL, and Clifford M. Neural control of limb coordination. I. Comparison of hatching and walking motor output patterns in normal and deafferented chicks. J Neurosci 7: 2320–2330, 1987.[Abstract]
Berkinblit MB, Deliagina TG, Feldman AG, Gelfand IM, and Orlovsky GN. Generation of scratching. II. Nonregular regimes of generation. J Neurophysiol 41: 1058–1069, 1978.
Berkowitz A. Broadly tuned spinal neurons for each form of fictive scratching in spinal turtles. J Neurophysiol 86: 1017–1025, 2001a.
Berkowitz A. Rhythmicity of spinal neurons activated during each form of fictive scratching in spinal turtles. J Neurophysiol 86: 1026–1036, 2001b.
Berkowitz A. Both shared and specialized spinal circuitry for scratching and swimming in turtles. J Comp Physiol [A] 188: 225–234, 2002a.
Berkowitz A. Endogenous biotin staining in a subset of spinal neuronal cell bodies: a potential confounding factor for neuroanatomical studies. Brain Res 938: 98–102, 2002b.[Medline]
Berkowitz A. Morphology and physiology of scratch-related spinal interneurons. Soc Neurosci Abstr 188: 4, 2003.
Berkowitz A and Stein PSG. Activity of descending propriospinal axons in the turtle hindlimb enlargement during two forms of fictive scratching: Broad tuning to regions of the body surface. J Neurosci 14: 5089–5104, 1994a.[Abstract]
Berkowitz A and Stein PSG. Activity of descending propriospinal axons in the turtle hindlimb enlargement during two forms of fictive scratching: Phase analyses. J Neurosci 14: 5105–5119, 1994b.[Abstract]
Berkowitz A and Stein PSG. Descending propriospinal axons in the turtle hindlimb enlargement: cells of origin and funicular courses. J Comp Neurol 346: 321–336, 1994c.[CrossRef][ISI][Medline]
Berkowitz A, Yosten GLC, and Ballard RM. A distinct morphological class of rhythmic, scratch-related, spinal interneurons in turtles. The 7th Congress of the International Society for Neuroethology PO226, Nyborg, Denmark, 2004.
Berkowitz A, Yosten GLC, and Ballard RM. Correlations between morphology and in vivo physiology of scratch-related spinal interneurons. Soc Neurosci Abstr 630: 16, 2005.
Buchanan JT. Lamprey spinal interneurons and their roles in swimming activity. Brain Behav Evol 48: 287–296, 1996.[ISI][Medline]
Carter MC and Smith JL. Simultaneous control of two rhythmical behaviors. II. Hindlimb walking with the paw-shake response in spinal cat. J Neurophysiol 56: 184–195, 1986.
Currie SN and Stein PS. Interruptions of fictive scratch motor rhythms by activation of cutaneous flexion reflex afferents in the turtle. J Neurosci 9: 488–496, 1989.[Abstract]
Deliagina TG and Feldman AG. Activity of Renshaw cells during fictive scratch reflex in cat. Exp Brain Res 42: 108–115, 1981.[ISI][Medline]
Deliagina TG and Orlovsky GN. Activity of Ia inhibitory interneurons during fictitious scratch reflex in the cat. Brain Res 193: 439–447, 1980.[CrossRef][ISI][Medline]
Drew T and Doucet S. Application of circular statistics to the study of neuronal discharge during locomotion. J Neurosci Methods 38: 171–181, 1991.[CrossRef][ISI][Medline]
Earhart GM and Stein PS. Scratch-swim hybrids in the spinal turtle: blending of rostral scratch and forward swim. J Neurophysiol 83: 156–165, 2000a.
Earhart GM and Stein PS. Step, swim, and scratch motor patterns in the turtle. J Neurophysiol 84: 2181–2190, 2000b.
Feldman AG and Orlovsky GN. Activity of interneurons mediating reciprocal Ia inhibition during locomotion. Brain Res 84: 181–194, 1975.[CrossRef][ISI][Medline]
Fernandez A, Radmilovich M, Russo RE, Hounsgaard J, and Trujillo-Cenoz O. Monosynaptic connections between primary afferents and giant neurons in the turtle spinal dorsal horn. Exp Brain Res 108: 347–356, 1996.[ISI][Medline]
Fetcho JR. Spinal network of the Mauthner cell. Brain Behav Evol 37: 298–316, 1991.[ISI][Medline]
Grelot L, Milano S, Portillo F, and Miller AD. Respiratory interneurons of the lower cervical (C4–C5) cord: membrane potential changes during fictive coughing, vomiting, and swallowing in the decerebrate cat. Pfluegers 425: 313–320, 1993.
Grillner S, Wallen P, Brodin L, and Lansner A. Neuronal network generating locomotor behavior in lamprey: circuitry, transmitters, membrane properties, and simulation. Annu Rev Neurosci 14: 169–199, 1991.[CrossRef][ISI][Medline]
Hounsgaard J and Kjaerulff O. Ca2+-mediated plateau potentials in a subpopulation of interneurons in the ventral horn of the turtle spinal cord. Eur J Neurosci 4: 183–188, 1992.[CrossRef][ISI][Medline]
Juranek J and Currie SN. Electrically evoked fictive swimming in the low-spinal immobilized turtle. J Neurophysiol 83: 146–155, 2000.
Larson CR, Yajima Y, and Ko P. Modification in activity of medullary respiratory-related neurons for vocalization and swallowing. J Neurophysiol 71: 2294–2304, 1994.
Lennard PR and Stein PS. Swimming movements elicited by electrical stimulation of turtle spinal cord. I. Low-spinal and intact preparations. J Neurophysiol 40: 768–778, 1977.
Lewis KE and Eisen JS. From cells to circuits: development of the zebrafish spinal cord. Prog Neurobiol 69: 419–449, 2003.[CrossRef][ISI][Medline]
Li WC, Perrins R, Soffe SR, Yoshida M, Walford A, and Roberts A. Defining classes of spinal interneuron and their axonal projections in hatchling Xenopus laevis tadpoles. J Comp Neurol 441: 248–265, 2001.[CrossRef][ISI][Medline]
Lieske SP, Thoby-Brisson M, Telgkamp P, and Ramirez JM. Reconfiguration of the neural network controlling multiple breathing patterns: eupnea, sighs and gasps. Nat Neurosci 3: 600–607, 2000.[CrossRef][ISI][Medline]
Marder E and Calabrese RL. Principles of rhythmic motor pattern generation.
other cells with axon terminal arborizations recovered; 
, cells with axon trajectories but no terminal arborizations recovered; 
, cells with neither axon trajectory nor terminal arborizations recovered. cc, central canal; VF, ventral fissure.
