QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online February 1, 2008
Journal of Experimental Biology 211, 613-629 (2008)
Published by The Company of Biologists 2008
doi: 10.1242/jeb.006270

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hill, A. A. V. Articles by Cattaert, D. Search for Related Content PubMed PubMed Citation Articles by Hill, A. A. V. Articles by Cattaert, D.

Recruitment in a heterogeneous population of motor neurons that innervates the depressor muscle of the crayfish walking leg muscle

Andrew A. V. Hill and Daniel Cattaert^*

Université de Bordeaux, Centre de Neurosciences Intégratives et Cognitives (CNIC), CNRS, UMR 5228, Bâtiment B2 Biologie Animale, Avenue des Facultés, 33405 Talence Cedex, France

View larger version (24K): [in this window] [in a new window]   Fig. 1. The in vitro walking leg preparation. (A) The 5th walking leg was dissected out together with the 3rd to 5th thoracic ganglia (T3–T5) and the 1st abdominal ganglion (A1) of the ventral nerve cord. In the intact animal the coxo-basipodite chordotonal organ (CBCO) is attached to dorsal edge of the coxopodite and an apodeme at the proximal-dorsal edge of the basipodite. Thus, the tension on the CBCO, which is composed of sensory neurons embedded in an elastic strand, is released during upward movements of the leg and is increased during downward movements. The levator (LEV) and depressor (DEP) muscles are located within the coxopodite. When the depressor muscle contracts, there is a rotation of the basipodite around a pivot point causing the downward movement of the leg and deformation of soft cuticle (dotted line) above and below this point. (B) Extracellular recordings were made from the various motor nerves as well as the sensory nerve of the CBCO (a CBCO neuron is represented in red) using en passant electrodes (not shown, see Materials and methods). Intracellular recordings of the motor neurons were made from within the neuropil (a Dep MN is represented in blue). Movements were imposed on the CBCO by a mechanical puller. Stretching the elastic strand mimicked downward movements of the leg, whereas releasing the strand mimicked upward movements. The dotted line marks the midline of the thoracic ganglia.

View larger version (17K): [in this window] [in a new window]   Fig. 2. Recruitment of three motor neurons in response to sensory input. Sinusoidal movements (mvt) were imposed on the CBCO strand while a depressor neuron (Dep MN) was recorded intracellularly from a neuritic branch in the posterior-lateral quadrant of the neuropil (see Fig. 5A for an explanation of quadrants) and extracellular spikes in the depressor nerve were monitored. In the uppermost trace the CBCO strand was stretched (S) and released (R). In this figure and in all subsequent figures an upward deflection of the movement trace corresponds to the release of tension on the CBCO strand, while a downward deflection corresponds to a stretch. The dotted line in the 2nd trace from the top indicates a membrane potential of –65 mV. It was possible to identify three distinct spike shapes in the extracellular record, using a method based on template matching. These spike shapes are shown at the bottom of the figure. A raster plot of the occurrence of these three spike shapes is shown below the raw extracellular record. A short vertical line represents a single spike. Note that the largest extracellular spike (L) corresponds one-for-one with the intracellular spikes.

View larger version (23K): [in this window] [in a new window]   Fig. 3. Spiking timing represented as phase of the sinusoidal movement. (A) Data from the preparation shown in Fig. 2 presented in terms of phase (i.e. a value from 0 to 1 rather than 0° to 360°). (B,C) Data from other preparations. The colored vertical bars show the relative amplitude of the extracellular spikes (large is blue, medium is red, small is black). To the right of each bar are a number of cycles of a raster plot corresponding to that spike. Each row represents one movement cycle with the first cycle on top and subsequent cycles displaced downwards. The small square is the mean spike phase, and the error bars are the standard deviation of this phase (for the calculation of mean spike phase, see Materials and methods). Note that in all three preparations the smallest spike was active through out the entire movement cycle, whereas the medium and large spikes were generally active only during the release phase; however, the mean phases do not differ very much between the different size spikes. (D) The mean vector length, a measure of the degree of clustering of spikes, is correlated with the mean number of spikes per cycle. The mean vector length is highest for neurons that fire very few spikes per cycle. An equation with a single exponential was used to fit the data (R2=0.718).

View larger version (10K): [in this window] [in a new window]   Fig. 4. Mean phase and mean vector length of small, medium and large spikes. Mean phase was calculated as the mean of the mean phases from 5 preparations in which three different sized spikes could be clearly identified as belonging to individual neurons. For an explanation of the calculation of mean vector length see Materials and methods. (A) There was no significant difference between mean phases of small, medium and large spikes. (B) There was a significant difference between the mean vector length of the small spikes and the medium spikes, and between small spikes and large spikes. There was no significant difference between the mean vector length of the medium and large spikes.

View larger version (21K): [in this window] [in a new window]   Fig. 5. The somata of the depressor motor neurons lie posterior to the neuropil. (A) Drawing of a rhodamine backfill of the depressor nerve reveals 13 somata (filled circles). For clarity only the somata are shown. The dotted line indicates the midline of the 5th thoracic ganglion. To aid the discussion of anatomical figures that follow, we divided the neuropil into four quadrants: AL (anterior-lateral), AM (anterior-medial), PL (posterior-lateral) and PM (posterior-medial). The cell body of the common inhibitor, which lies on the midline, was not filled in this particular preparation. (B) A cross section of the depressor nerve shows 18 circular profiles, including 13 with thick walls. The five thin-walled profiles are among the smallest in diameter. (C) The diameters of the 13 thick-walled profiles vary in a continuous manner.

View larger version (29K): [in this window] [in a new window]   Fig. 6. The anatomical and physiological characterization of an individual depressor motor neuron (Dep MN) that, when viewed from above, has branches that cover the whole neuropil. (Ai) This neuron was recorded from its main neurite, the thick branch in the posterior-lateral quadrant that leads directly to the axon. This depressor motor neuron has the 4th largest extracellular spike of the 12 depressor motor neurons in this experiment. The response properties of this neuron are summarized above the histogram of extracellular spike amplitude. The neuron shows a resistance response (R) to movements imposed on the CBCO strand and receives monosynaptic (E) and polysynaptic (P) excitatory input. (Aii) This neuron is represented from three different viewpoints. Based on previous work describing the anatomy of the CBCO sensory neurons, it may receive monosynaptic input from CBCO sensory neurons in the region indicated by the gray oval (El Manira et al., 1991). In particular, the lateralmost branch, which is also the most ventral, is a likely area of contact. The direction indicated by the arrows is dorsal (d). (B) This neuron depolarizes during the release (R) phase of a sinusoidal movement (mvt) imposed on the CBCO strand. The resting membrane potential, indicated by the dotted line, is –69 mV. The data shown are averages of eight cycles triggered by a timing pulse that was phase locked to the movement trace. (C) A ramp-and-hold stimulus reveals that this neuron is phaso-tonic. It strongly depolarizes during the release phase of ramp movements (Ci; mvt) and also shows a small, slowly decaying depolarization. Note that the ramp movements may appear to be instantaneous (perfectly vertical) at the time scale in Ci but are in fact ramps (Cii). The data shown are averages of eight cycles. (D) Stimulation of the CBCO nerve reveals that this neuron receives mono- and polysynaptic excitatory inputs, which were distinguished based on the delay of the peaks of the compound EPSPs from the stimulus artifact marked with an asterisk (5 ms and 22.7 ms, respectively).

View larger version (26K): [in this window] [in a new window]   Fig. 7. Two example experiments in which the extracellular spike amplitudes, orthodromic conduction delays and response properties of many depressor motor neurons (Dep MN) were characterized. (A) 12 depressor motor neurons (Dep MN) were characterized. (Ai) The amplitude of the extracellular spike varies continuously. Spike amplitudes were normalized to the largest amplitude found in a given experiment. The neurons showed a resistance (R) reflex, and assistance (A) reflex, or no (N) response to movements imposed on the CBCO strand. Neurons received monosynaptic EPSPS (E), polysynaptic EPSPs (P), and IPSPs (I). (Aii) The neurons with the smallest spikes have the longest orthodromic delays. (Aiii) Conduction velocity was calculated for each neuron by dividing the distance between the neuropil and the site of the extracellular recording (7.5 mm) by the orthodromic delay. Conduction velocity varied with extracellular spike amplitude. (B) In a second experiment, we recorded from nine depressor motor neurons with similar trends to those found in the first experiment.

View larger version (35K): [in this window] [in a new window]   Fig. 8. Four examples of Fast resistance whole neuropil (FR/W) depressor motor neurons (Dep MN) that have neuritic processes extending throughout most of the neuropil when viewed from above. According to our classification scheme, these four neurons (A–D) belong in the same group as the neuron represented in Fig. 6. These neurons have relatively large extracellular spikes (insets), and show a resistance response (R). They receive monosynaptic excitatory (E) input, polysynaptic (P) excitatory input and polysynaptic inhibitory (I) input. Also, similar to the neuron shown in Fig. 6, these neurons presumably receive monosynaptic input on the lateral-most branches in the posterior-lateral quadrant. Note that the neurons in C and D lack medial-most branches. Although it is possible that neurons in C and D may represent another class of neurons, due to the great anatomical and physiological similarity of these neurons with those in A and B and in Fig. 6A, we consider them also to belong to the class FR/W.

View larger version (54K): [in this window] [in a new window]   Fig. 9. Two fast oblique (F/O) class depressor motor neurons. These neurons have large diameter main neurites that lie obliquely along the medial edge of the neuropil and a large extracellular spike. (Ai) Two neurons were recorded from and filled with different dyes in a single animal. While there are many similarities in the morphology of the two neurons, there are also some notable differences. For example, the rhodamine-filled neuron (red) has four branches in the lateral-most part of the lateral-posterior quadrant where CBCO sensory neurons may make monosynaptic contact with depressor motor neurons, whereas the Lucifer Yellow-filled neuron (green) has only one branch in this region. (Aii) The rhodamine-filled neuron shows a monosynaptic EPSP with a delay of 5.02 ms from the stimulus artifact and amplitude of 6.08 mV in response to CBCO stimulation (an average of 23 traces is shown). The dotted line indicates a resting membrane potential of –74 mV. (Aiii) The Lucifer Yellow filled neuron shows only an IPSP in response to CBCO stimulation. (Bi) The morphology of the rhodamine-filled neuron. (Ci) The morphology of the Lucifer Yellow-filled neuron. (Bii) The rhodamine-filled neuron showed a resistance response to movement of the CBCO strand. The resting potential was –74 mV. The data shown are averages of 10 cycles triggered off of the movement trace. (Cii) The Lucifer Yellow-filled neuron showed no response to movement of the CBCO strand. The resting potential was –72 mV. The data shown are averages of 11 cycles triggered from the movement trace.

View larger version (25K): [in this window] [in a new window]   Fig. 10. Three depressor motor neurons (Dep MN) that belong to the Medium (M) class. These neurons have relatively small extracellular spikes, thin neuritic branches and unique branching patterns. Each M neuron that we recorded from was unique in morphology from all other M neurons. (A) This neuron has a relatively sparse neuritic arbor. It shows a resistance and an assistance response to movement (RA), and receives monosynaptic excitatory input (E) and polysynaptic inhibition (I). (B) This neuron has very fine neuritic branches that cover most of the neuropil when viewed from above. It does not respond to movement of the CBCO strand (N) and only receives polysynaptic EPSPs (P). (C) This neuron has neuritic branches that cover most of the neuropil except for the anterior-most parts of the anterior-lateral and anterior-medial quadrants. This neuron shows no response to movement and receives only IPSPs.

View larger version (24K): [in this window] [in a new window]   Fig. 11. Two examples of the Fast assistance depressor motor neuron (Dep MN). (Ai) This Fast assistance (FA) neuron has sparse neurites and lacks branches in the anterior-lateral quadrant. It receives mono- (E) and polysynaptic EPSPs (P) as well as IPSPs (I). The orthodromic delay is 1.62 ms. (Aii) The neuron depicted in Ai depolarizes during the stretch phase of sinusoidal movement (mvt). In response to a ramp-and-hold stimulus the neuron strongly depolarizes phasically during stretches and weakly hyperpolarizes during releases. The resting membrane potential (dotted lines) was –69 mV. The response to sinusoidal movement is an average of 9 cycles triggered from the movement trace. The response to ramp-and-hold movement is an average of 5 cycles. (Bi) The morphology of this Fast assistance neuron is similar to that of the one shown to the left. In this particular experiment we did not record from other depressor motor neurons. Therefore, there is no histogram of spike amplitudes. The orthodromic delay is 2.3 ms. (Bii) The response of this neuron to movement is very similar to that of the neuron shown to the left. The resting membrane potential (dotted lines) was –50 mV. The response to sinusoidal movement is an average of 22 cycles triggered from the movement trace. The response to ramp-and-hold movement is an average of 13 cycles.

View larger version (12K): [in this window] [in a new window]   Fig. 12. An example of a Slow assistance (SA) depressor motor neuron (Dep MN). (A) The branches of this neuron are relatively sparse and are restricted to the posterior-medial and anterior-medial quadrants. This neuron receives only polysynaptic inhibition. The orthodromic delay is 5.6 ms. (B) In response to sinusoidal movement (mvt), this neuron hyperpolarized during the release phase and depolarized during the stretch phase, which is consistent with an assistance response. The ramp movement stimuli reveal that the input to this neuron is inhibitory. Similarly electrical stimulation of the CBCO nerve revealed only IPSPs (I). The resting membrane potential indicated by the dotted lines was –55 mV.

View larger version (14K): [in this window] [in a new window]   Fig. 13. A summary of extracellular spike amplitude and orthodromic delay for the various classes of neurons. (A) The orthodromic delay of Slow assistance (SA) neurons was significantly different from that of all other classes of neurons. In the inset a solid circle at the intersection of the initials representing each class indicates a significant difference. (B) There were no significant differences in spike amplitude among the Slow assistance (SA) neurons, the Fast assistance (FA) neurons, and the Medium (M) neurons. However, these three classes had significantly different spike amplitudes than the F/O and FR/W classes. There was no significant difference between the F/O and FR/W class neurons.

View larger version (21K): [in this window] [in a new window]   Fig. 14. Soma (A), neurite (B) and axon diameters (C) of the five classes of motor neurons. (A) FR/W class neurons had significantly larger somata than SA and M neurons. Because the somata of these motor neurons were not perfectly round, the diameter was measured by taking the mean of the smallest diameter and the largest diameter. (B) The F/O class neurons had significantly larger neuritic diameters than all other classes of neurons. We measured neurite diameter at the point just posterior to the junction where the neurite from the soma reaches the main branch. (C) FR/W class neurons had significantly larger diameter axons than FA and M class neurons, and F/O class neurons had significantly larger diameter axons than FA and M class neurons. We measured axon diameter at the point where the neurite exits the neuropil. This location is not ideal since the neurites tend to narrow at this point before becoming larger in diameter in the nerve, but unfortunately not all of our dye-fills included the true axon in the nerve.