Paul S. G. Stein

Central program for scratch reflex in turtle

Summary1.A scratch reflex is displayed by a low spinal turtle in response to gentle mechanical stimulation delivered to specific regions of the shell anterior to a hindlimb (Valk-Fai and Crowe, 1978, 1979). The behavior consists of a rhythmic limb movement in which the foot protracts and rubs against the stimulated patch of shell once per cycle. A brief stimulus can elicit a response composed of a small number of cycles or even a fraction of a cycle (Fig. 1); maintained rubbing of the shell can elicit a response consisting of 5–20 cycles (Fig. 2A).2.The scratch motor program is divisible into an A phase in which IT-KE is active and a B phase in which IT-KE is quiescent and HR-KF is active (Fig. 2B). The A phase is further divisible into an A1 phase in which IT-KE is the only active KE muscle and into an A2 phase in which all three KE muscles are active.3.Two variations of the scratch program exist (Figs. 3 and 4). These variations indicate that there are flexibilities in the organization of the spinal generator for turtle scratch reflex.4.The scratch motor program can also be expressed in a preparation immobilized by a neuromuscular blocking agent (Figs. 5 and 6). The responses in this preparation are termed the fictive scratch reflex since they occur in the absence of a real movement. One of the variations in the scratch program which was seen in preparations with a moving limb was also seen in the fictive scratch reflex (Fig. 7).5.The scratch motor program can also be expressed in preparations with the hindlimb enlargement deafferented via bilateral dorsal rhizotomies (Figs. 8 and 9).6.These observations establish that the turtle scratch reflex is centrally programmed, i.e., it is not dependent upon phasic timing cues derived from sensory feedback. Even though the reflex is centrally programmed, it is still possible to demonstrate that sensory feedback from a moving limb can modulate the scratch rhythm (Fig. 10 of this paper and Fig. 6 of Valk-Fai and Crowe, 1979).

Mechanisms of Interlimb Phase Control

During locomotion there is a phase lag in the movement of one limb when compared to the movement of another limb. The neurons responsible for the regulation of interlimb phase have been described in the swimmeret system of the crayfish. These neurons are termed coordinating neurons. Their axons originate in one limb-bearing ganglion of the ventral nerve cord and terminate in neighboring ganglia of the nerve cord. When the axons of coordinating neurons are cut, interlimb phase regulation is destroyed. These interneurons are active during a specific fraction of the movement cycle of a limb (the “modulator” limb). This activity pattern persists in the absence of sensory discharge from the “modulator” limb. The discharge of coordinating neurons can alter the timing of locomotory movements in another limb (the “modulated” limb). Such an alteration is termed a phase shift. Both the magnitude and direction of the phase shift depend upon the arrival time of the coordinating neuron discharge in the movement cycle of the modulated limb. A systematic plot of these phase shifts is termed the “phase response curve” (PRC). The PRC can be obtained utilizing a “cut command neuron” preparation. Such a preparation leaves the coordinating neurons intact and permits separate electrical control of each limb. The magnitude of interlimb phase can be predicted if the values of the following are known: (1) the intrinsic cycle period of the modulator limb, (2) the intrinsic cycle period of the modulated limb, and (3) the PRC.

Neuronal control of turtle hindlimb motor rhythms

The turtle, Trachemys scripta elegans, uses its hindlimb during the rhythmic motor behaviors of walking, swimming, and scratching. For some tasks, one or more motor strategies or forms may be produced, e.g., forward swimming or backpaddling. This review discusses experiments that reveal characteristics of the spinal neuronal networks producing these motor behaviors. Limb-movement studies show shared properties such as rhythmic alternation between hip flexion and hip extension, as well as variable properties such as the timing of knee extension in the cycle of hip movements. Motor-pattern studies show shared properties such as rhythmic alternation between hip flexor and hip extensor motor activities, as well as variable properties such as modifiable timing of knee extensor motor activity in the cycle of hip motor activity. Motor patterns also display variations such as the hip-extensor deletion of rostral scratching. Neuronal-network studies reveal mechanisms responsible for movement and motor-pattern properties. Some interneurons in the spinal cord have shared activities, e.g., each unit is active during more than one behavior, and have distinct characteristics, e.g., each unit is most excited during a specific behavior. Interneuronal recordings during variations support the concept of modular organization of central pattern generators in the spinal cord.

Synaptic control of hindlimb motoneurones during three forms of the fictive scratch reflex in the turtle.

1. The turtle spinal cord produces three forms of the fictive scratch reflex in response to tactile stimulation of sites on the body surface. Common to all three forms is the rhythmic alternation of activity between hip protractor and hip retractor motoneurones. Hip protractor motoneurone activity is monitored via nerves innervating the hip protractor muscle puboischiofemoralis internus pars anteroventralis (VP‐HP). Hip retractor activity is monitored via nerves innervating several monoarticular hip retractor muscles, one hip adductor muscle, and several biarticular hip retractor‐knee flexor muscles (HR‐KF). Each form is characterized by the timing of activity of motoneurones innervating femorotibialis (FT‐KE), a monoarticular knee extensor muscle, within this alternating cycle (Robertson, Mortin, Keifer & Stein, 1985). In the present study, intracellular recordings revealed a corresponding regulation of synaptic drive to knee extensor motoneurones with respect to the synaptic drive to the motoneurones innervating antagonist muscles of the hip. These patterns of synaptic activation give rise to the distinct motor pattern underlying each form of the scratch reflex. 2. VP‐HP, HR‐KF and FT‐KE motoneurones all exhibited phasic depolarizing and hyperpolarizing changes in membrane voltage during the production of the rhythmic motor patterns underlying each stratch form. Membrane depolarization is caused by synaptic excitation (EPSPs) and gives rise to motoneurone discharge. Hyperpolarization is primarily the result of postsynaptic inhibition (IPSPs) mediated by an increased conductance of chloride ions (Cl‐) and ensures motor pool quiescence during antagonist activation. 3. VP‐HP motoneurones depolarized during activation of the VP‐HP motor pool and hyperpolarized during activation of the HR‐KF motor pool. HR‐KF motoneurones depolarized during activation of the HR‐KF motor pool and hyperpolarized during activation of the VP‐HP motor pool. In many cases, the amplitude of hyperpolarization was directly related to the intensity of the antagonist motor pool burst. During the rostral scratch, HR‐KF motor pool activity was sometimes deleted, along with the depolarizing wave in HR‐KF motoneurones and the hyperpolarizing wave in VP‐HP motoneurones. The interneurones providing the synaptic drive to these antagonist motoneurones appear, therefore, to have reciprocal activation patterns. 4. FT‐KE motoneurones depolarized during FT‐KE motor pool activation and hyperpolarized during FT‐KE motor pool quiescence. This alternation of opposing synaptic drive underlies the rhythmic activation of the FT‐KE motor pool during all scratch forms.(ABSTRACT TRUNCATED AT 400 WORDS)

Motor pattern deletions and modular organization of turtle spinal cord

The turtle spinal cord contains a central pattern generator (CPG) that produces rhythmic hindlimb motor patterns during a rostral scratch. This review describes evidence in support of the hypothesis that the turtle rostral scratch CPG has a modular structure similar to that described in the Unit-Burst-Generator hypothesis for cat locomotion by Grillner. During normal rostral scratch in turtle, activity bursts rhythmically alternate with quiescence for each motor neuron pool; agonist activity rhythmically alternates with antagonist activity at each degree of freedom, e.g., hip, knee; and a transition from knee flexor to knee extensor motor neuron activity occurs midway during each hip flexor motor neuron burst. Hip extensor deletions, knee flexor deletions, and knee extensor deletions are motor pattern variations of rostral scratch. During each of these variations, agonist activity is rhythmic; antagonist activity and agonist quiescence are absent. Several classes of evidence during both normal and variation motor patterns support a modular organization of the turtle rostral scratch CPG: electroneurographic recordings from axons of motor neurons, intracellular recordings of synaptic potentials in motor neurons, and extracellular unit recordings from spinal interneurons. These data support the hypotheses that the knee extensor module is different from the hip extensor module and that the knee flexor module is different from the hip flexor module. Potential mechanisms for rhythmogenesis include reciprocal connections between agonist and antagonist modules at each degree of freedom, and agonist module rhythmogenesis. Additional tests of the modular hypothesis for turtle rostral scratch include unit recordings from knee-related interneurons during normal rostral scratch, as well as during knee-related deletions.

Motor neuron synaptic potentials during fictive scratch reflex in turtle

Summary1.A scratch reflex motor program, termed the ‘fictive’ scratch reflex (Stein and Grossman 1980), is displayed by an immobilized low spinal turtle in response to gentle mechanical stimulation delivered to specific regions of the shell. The fictive scratch is a cyclic program; each cycle is divisible into three phases, the A1, the A2 and the B phases (Fig. 1). Rhythmic A1 and A2 activities may be produced even when the B phase is deleted (Fig. 2).2.Intracellular recordings from hindlimb motor neurons during the fictive scratch reveal that each motor neuron is depolarized to fire action potentials during its active period and is hyperpolarized during at least part of its quiescent period (Figs. 3–5). Moreover, the activation pattern of A1 motor neurons is the inverse of the activation pattern of B motor neurons (Figs. 3 and 5). The synaptic activation patterns of motor neurons during the A1 and A2 phases are preserved even during B phase deletions (Fig. 6).3.A fictive flexion reflex motor program is produced in this preparation in response to gentle mechanical pressure applied to the dorsum of the foot (Figs. 7 and 8). The synergies observed in fictive flexion reflex differ from those observed during the fictive scratch reflex.4.These data support a model for turtle of a three phase scratch generator that is asymmetrically arranged. A similar conclusion has been reached in studies of the scratch program generator in the cat (Berkinblit et al. 1978a, b). Our data also indicate that the motor neuron activation pattern of flexion reflex is different from that of scratch reflex. Therefore data obtained from turtle flexion reflex can not be utilized to construct a model of the turtle scratch generator.

Swimming movements elicited by electrical stimulation of the turtle spinal cord: The high spinal preparation

Summary1.The movements of each limb of the turtle,Pseudemys scripta elegans, are out-of-phase with the movements of its ipsilateral neighboring limb during normal swimming. Protraction of one limb coincides with retraction of an adjacent limb. The period of the movement cycle in one limb is equal to that of its neighbor. This coordination pattern is termed 1∶:1 or absolute coordination.2.Swimming movements of the limbs are not produced spontaneously in a turtle with an acute section of the spinal cord at the first cervical segment. This “high spinal” turtle can produce swimming movements in several limbs, however, in response to unpatterned, i.e. constant frequency, electrical stimulation applied to the dorsolateral funiculus (DLF) caudad to the spinal transection. On many occasions, it was possible to place a stimulating electrode in the DLF so that the swimming movements exhibited the normal pattern of absolute coordination (Figs. 2, 4, 5, 8).3.On a few occasions, it was possible to place a stimulating electrode in the DLF so that the swimming movements elicited by electrical stimulation displayed an abnormal pattern of coordination. In these cases, protraction of the hindlimb coincided with every second occurrence of retraction of the forelimb. The period of the movement cycle of the hindlimb was twice as long as that of the ipsilateral forelimb. This coordination pattern is termed 2∶:1 coordination, and is an example of one type of relative coordination (Figs. 7, 11).4.These results are consistent with the hypotheses (i) that each limb has its neural control center resident mainly in the spinal cord and (ii) that the neural elements necessary for interlimb coordination are not dependent upon supraspinal connections.

In vitro motor program for the rostral scratch reflex generated by the turtle spinal cord

Abstract Tactile stimulation applied to the turtle shell bridge elicits a rostral scratch reflex in the spinal turtle. The triphasic rostral scratch motor program produced in vivo can also be also produced by an in vitro turtle shell and spinal cord preparation. The in vitro production of this scratch program provides support for the generalization that coordinated motor programs can be generated by the spinal cord. This in vitro preparation can be utilized to examine the pharmacological sensitivities of the spinal cord neurons generating the scratch motor program.

Spinal Cord Circuits for Motor Pattern Selection in the Turtlea

The turtle spinal cord contains neural circuitry capable of generating the motor patterns responsible for the flexion reflex and each of the three forms of the scratch reflex. This neural circuitry selects the appropriate motor pattern in response to a specific cutaneous stimulus. We understand some of the neuronal mechanisms responsible for motor pattern selection and generation; additional work is still required to understand other neuronal mechanisms functioning at the spinal cord level. These mechanisms, once understood at the spinal level, may be used to generate hypotheses concerning the roles of supraspinal structures in motor pattern selection and generation.

N-Methyl-d-aspartate antagonist applied to the spinal cord hindlimb enlargement reduces the amplitude of flexion reflex in the turtle

APV (D(-)-2-amino-5-phosphonovalerate), an NMDA (N-methyl-D-aspartate) antagonist, was applied in situ onto segments of the hindlimb enlargement of the turtle spinal cord. APV reduced the response amplitude of the flexion reflex. In contrast, APV did not alter the responsiveness of the rostral scratch reflex. Afferents for the flexion reflex enter the spinal cord via the dorsal roots of the middle segment of the hindlimb enlargement; afferents for the rostral scratch reflex enter the spinal cord via dorsal roots located anterior to the hindlimb enlargement. The results are consistent with the hypothesis that sensory interneuron NMDA receptors, synaptically activated either directly or indirectly by nearby cutaneous afferent axons, play a role in the spinal cord processing of cutaneous information.