R. E. Burke

Differential control of fast and slow twitch motor units in the decerebrate cat

SummaryIn approximately 3/4 of precollicular decerebrate (unanesthetized) cats, tetanic electrical stimulation of the sural nerve, or pinching the ankle skin innervated by the sural nerve, produced predominant excitation (overall increased force output and EMG activity) in the mixed medical gastrocnemius (MG) muscle and simultaneous inhibition in its slow twitch Synergist, soleus (Sol). The present experiments were designed to test whether, as this and other evidence suggests, certain sets of cutaneous afferents can produce activation of particular groups of motor units and simultaneous inhibition of other groups within the same motor unit pool (i.e., units belonging to a single muscle).We recorded, in decerebrate cats, the activity of restricted sets of MG motor units using either fine bipolar EMG wire electrodes or bipolar hook electrodes on small natural filaments of the MG muscle nerve. In preparations exhibiting the differential effect of sural input noted above, we usually found that some low threshold MG motor units (i.e., those responding to stretch or vibration of the MG muscle) exhibited slowing of discharge or complete inhibition at the same time that higher threshold MG units, not responsive to stretch or vibration, were powerfully recruited by either electrical or natural stimulation of sural nerve afferents. The net balance of synaptic effects within the MG motoneuron population may thus be excitatory in some cells and simultaneously inhibitory in others. This finding, together with earlier evidence, suggests the existence of at least two patterns of organization of synaptic input to the MG motoneuron pool.

The use of state-dependent modulation of spinal reflexes as a tool to investigate the organization of spinal interneurons.

Abstract This review examines the proposition that state-dependent modulation of transmission through spinal reflex pathways can be used as an investigative tool to reveal details about the organization of spinal interneurons into functional circuits. The first set of examples includes the use of spinal and supraspinal lesions, as well as the administration of the drug l-dihydroxyphenylalanine (l-DOPA), to produce different, relatively stable ”states” of the central nervous system (CNS), revealing previously unsuspected spinal pathways activated by the flexor reflex afferents (FRA). The second set of examples deals with the use of fictive locomotion and scratching to investigate the organization of oligosynaptic excitatory and inhibitory reflex pathways from cutaneous and muscle afferents. As in the first set of examples, several hitherto unknown reflex pathways have been found only during the flexion or extension phases of rhythmic locomotion, which are regarded as different CNS states. Differences in the patterns of control can be used to infer the existence of distinct sets of reflex pathway interneurons that have remarkably precise input/output relations.

HRP anatomy of group la afferent contacts on alpha motoneurones

Some of the currently unresolved questions regarding the physiology of monosynaptic action of group Ia afferent fibers on alpha motoneurons can be framed in terms of the anatomical organization of Ia-motoneuron contacts. Such issues include the spatial distribution of Ia synaptic terminations on the motoneuron dendritic tree2~s~5~6,12 and the nature of the all-or-none ‘unit’ EPSPs components that make up single-fiber Ia EPSPs (see refs.

Horseradish peroxidase study of the spatial and electrotonic distribution of group Ia synapses on type-identified ankle extensor motoneurons in the cat.

9 and 11). These and some related questions that arise from electrophysiological observations require, for further clarification, direct study of the anatomy of the intraspinal arborizations of functionally identified group Ia afferents and of their contacts on similarly identified alpha motoneurons, as is now possible using intracellular iontophoresis of horseradish peroxidase (HRP)sp14. In the present work, adult (2-3 kg) cats were prepared surgically under halothane anesthesia which was later replaced by pentobarbital. The tendons of the left medial (MG) and lateral (LG) gastrocnemius and the soleus (SOL) muscles were separated, cut, and arranged for independent stretch. Their respective muscle nerves were freed but left intact and placed on stimulating electrodes. After lumbosacral laminectomy, all tissues were covered with mineral oil pools warmed by radiant heat. Micropipette electrodes (beveled tips, 1.4-2.0 pm diameter) were filled with freshly prepared 4 % HRP (type VI, Sigma) in 0.2 M KCl-0.05 M Tris buffer at pH 8.614. The left dorsal column was explored for afferents responding to slow sinusoidal stretch of the 3 muscles. Upon intra-axonal penetration of such an afferent (> 20 mV negative DC shift and positive spike amplitude), it was identified by antidromic invasion after muscle nerve stimulation (estimated conduction velocity 2 100 m/set), by silence during the resulting isometric muscle contraction, and by large dynamic responses to ramp-hold stretch and ability to follow longitudinal vibration (100 pm amplitude at 100 Hz or greater) of the appropriate muscle. HRP was then injected using trains of depolarizing current pulses (25 msec duration, 20 Hz, 25 nA) which were interrupted briefly every 1 set to check the amplitude of the antidromic action potential. Current

Supraspinal control of a short-latency cutaneous pathway to hindlimb motoneurons.

Eight functionally identified group Ia muscle afferents from triceps surae or plantaris muscles were labeled intraaxonally with horseradish peroxidase (HRP) in seven adult cats. Subsequently, HRP was injected into two to six homonymous or heteronymous α‐motoneurons per animal (total = 22), each identified by motor unit type and located near the site of afferent injection. The complete trajectories of labeled afferents were reconstructed, and putative synaptic contacts on HRP‐labeled motoneurons were identified at high magnification. Dendritic paths from each contact were also mapped and measured. A total of 24 contact systems (the combination of a group Ia afferent and a postsynaptic motoneuron) were reconstructed, of which 17 were homonymous, and seven were heteronymous.

Disynaptic excitation from the medial longitudinal fasciculus to lumbosacral motoneurons: modulation by repetitive activation, descending pathways, and locomotion

SummaryThe effects of two supraspinal systems on transmission through a short latency hindlimb cutaneous reflex pathway were studied in cats anesthetized with pentobarbital or α-chloralose. Fleshman et al. (1984) described a mixed excitatoryinhibitory input from low threshold superficial peroneal (SP) afferents to flexor digitorum longus (FDL) motoneurons with central latencies so short as to suggest a disynaptic component in the initial excitatory phase of the PSP. In the present study, conditioning stimulation of either the red nucleus (RN) or the pyramidal tract (PT) caused a marked decrease in latency and increase in amplitude of both the excitatory and inhibitory components of the SP PSP in FDL motoneurons and several other motoneuron species. The minimal central latencies of the conditioned initial excitatory phase of the PSPs were on the order of 1.5 ms, consistent with the possibility of a disynaptic linkage. The facilitatory effects of RN and PT conditioning were observed in both anesthetic conditions, although preparation-specific differences in latency were observed. Lesion experiments suggested that the interneurons involved in this pathway are located caudal to the L5 segment, most likely in segments L6 and L7.

Phasic modulation of short latency cutaneous excitation in flexor digitorum longus motoneurons during fictive locomotion.

SummaryShort-latency excitatory postsynaptic potentials (EPSPs) evoked by stimulation in the medial longitudinal fasciculus (MLF) were recorded intracellularly from motoneurons in the cat lumbosacral spinal cord. Monosynaptic and disynaptic EPSPs occurred in most flexor and extensor motoneurons studied. These EPSPs resulted from the activation of fast (> 100 m/s) descending axons from the MLF to the spinal cord. Several features distinguished monosynaptic and disynaptic MLF EPSPs. Disynaptic EPSPs exhibited temporal facilitation during short trains of stimulation, whereas monosynaptic EPSPs did not. Disynaptic EPSPs, but not monosynaptic EPSPs, were also facilitated by stimulation of the pyramidal tract and the mesencephalic locomotor region. However, disynaptic MLF EPSPs exhibited little or no facilitation when conditioned by short-latency cutaneous pathways. During fictive locomotion, the amplitude of disynaptic MLF EPSPs was modulated, with maximal amplitudes during the step cycle phase when the recorded motoneuron was active, resulting in reciprocal patterns of modulation of flexors and extensors. No comparable change was seen in the amplitude of monosynaptic MLF EPSPs during fictive stepping. These data suggest that the central pattern generator for locomotion modulates disynaptic MLF excitation at a premotoneuronal level in a phase-dependent manner. The effects of lesions made in the MLF and thoracic cord suggest that the interneurons in the disynaptic pathway from the MLF to motoneurons reside in the lumbosacral cord.

Differential control of short latency cutaneous excitation in cat FDL motoneurons during fictive locomotion

SummaryWe examined modulation of transmission of short-latency excitation produced by distal hindlimb cutaneous input, as well as fluctuations in motoneuron membrane potential and input resistance, in flexor digitorum longus (FDL) motoneurons during fictive locomotion. Fictive stepping was induced in unaesthetized, decerebrate cats either by repetitive stimulation of the mesencephalic locomotor region (MLR) or by administration of Nialamide and 1 DOPA after low spinal section. In the MLR preparations, brief depolarizing waves occurred in FDL cells during the early flexion phase of fictive stepping, immediately after cessation of activity in extensor muscles. In some FDL cells, plateau-like depolarizations also occurred during the extensor phase. Fictive stepping induced in acutely spinalized cats by administration of l-DOPA was slower and more variable; peak polarization in FDL motoneurons always occurred during the early flexion phase but there was usually no distinct depolarization during extension. In both types of preparation, the initial EPSP components in synaptic potentials (SP-EPSPs) produced by electrical stimulation of the cutaneous division of the superficial peroneal nerve (SP) were maximally facilitated during early flexion, coincident with the peak of background depolarization. This enhancement was manifested by an increase in the amplitude of initial SP-EPSP components or by decreased central latency of the initial EPSP components, or both. In most FDL motoneurons, input resistance decreased systematically during late flexion, coincident with relative membrane hyperpolarization. Correction of SP-EPSP amplitudes for changes in input resistance suggested that SP-EPSP facilitation persisted throughout the flexion phase These findings are discussed with reference to modulation of cutaneous reflexes during locomotion and the possibility that excitatory last-order interneurons in particular cutaneous reflex pathways may distribute excitatory drive from the central pattern generator for locomotion to FDL α-motoneurons

Supraspinal facilitation of cutaneous polysynaptic EPSPs in cat medial gastrocnemius motoneurons

SummaryLow threshold (≤2×T) cutaneous afferents in the superficial peroneal (SP) and medial plantar (PLNT) nerves both produce short-latency excitatory postsynaptic potentials (EPSPs) in flexor digitorum longus (FDL) α-motoneurons, with minimum central latencies (≤1.8 ms) that indicate a disynaptic connection. However, SP and PLNT EPSPs in FDL motoneurons are differentially modulated during fictive stepping in decerebrate cats. The early components in SP EPSPs are systematically enhanced during the early flexion phase of fictive stepping (Schmidt et al. 1988) while those in PLNT EPSPs are markedly depressed during flexion. In addition, transmission in the PLNT→FDL pathway is enhanced during occasional step cycles in which the FDL displays firing during the extension phase. This enhancement affects only the trisynaptic components of PLNT EPSPs, is simultaneous with the extension FDL burst, and is not found in SP EPSPs. These results indicate that the SP→FDL and PLNT→FDL pathways are composed of entirely different sets of segmental last-order interneurons, each of which receives sensory information from contiguous, relatively limited regions of skin on the most distal parts of the hindpaw. Possible functional consequences of these interneuronal organizations are discussed.

The control of movement and posture

SummaryWe examined the characteristics of postsynaptic potentials (PSPs) produced in antidromically-identified medial gastrocnemius (MG) α-motoneurons by electrical stimulation of low threshold (< 3×T) distal limb cutaneous afferents in the sural (SUR) nerve in adult cats anesthetized with α-chloralose, together with the effects on SUR PSPs of supraspinal conditioning stimulation of the contralateral red nucleus (RN) and pyramidal tract (PT). In the majority of MG motoneurons, SUR afferents with electrical thresholds < 1.5×T produced early excitatory synaptic potentials (EPSPs) with minimum central latency of about 2.0 ms, suggesting activation of a trisynaptic segmental pathway with two interposed interneurons. Such early EPSPs were often detectable with stimuli < 1.2×T, as determined by recording the compound action potential in the sciatic nerve and from the first appearance of the N1 wave of the cord dorsum potential. Inhibitory synaptic potentials (IPSPs) were regularly produced by SUR volleys of only slightly greater strength (often as low as 1.3×T) and these had minimum central latencies of about 3.0 ms (about 1.0 ms longer than the earliest EPSPs), suggesting a three interneuron central pathway.Repetitive stimulation of RN and PT regularly produced facilitation of both EPSP and IPSP components in the SUR response, suggesting that these supraspinal systems directly or indirectly excite some of the same interneurons that convey the SUR effects to MG motoneurons. When using very low strength SUR stimuli, PT conditioning produced relatively pure facilitation of the SUR EPSPs but with larger SUR volleys, PT clearly facilitated both EPSPs and IPSPs. RN conditioning produced more parallel facilitation of SUR EPSPs and IPSPs. Supraspinal control of the polysynaptic pathway producing SUR EPSPs is of particular interest because of earlier evidence that this pathway is differentially distributed to motoneurons of fast twitch versus slow twitch MG motor units.