Swimming by young Xenopus tadpoles

The most complex patterns of behaviour occur in vertebrates, and understanding the neuronal basis of these behaviours must involve unravelling the circuits of the spinal cord. One approach, which has been particularly fruitful, is to examine very simple behaviours in lower vertebrates, particularly swimming in lampreys (Grillner et al., 1995) and in tadpoles. The pattern of swimming in newly hatched tadpoles of the clawed frog Xenopus, turns out to be especially simple (Roberts, 1990). For the first few days of life, these tadpoles spend much of their time attached to leaves and stems by a cement gland on the head. If the skin is touched, a tadpole will swim by side-to-side undulating movements of its trunk and tail until it finds a new attachment site (Fig. 7.10a). At the start of a swim, the body flexes to the left and then to the right up to 20 times per second. The rate of flexion declines during a swim. Waves of movement travel towards the tail end and propel the tadpole forwards at up to 5 cm/s.

The spinal cord of a young tadpole contains about 1000 neurons of eight types, including five types of interneuron. Alan Roberts and his colleagues have pioneered a method of making intracellular recordings from these neurons during fictive swimming. A tadpole is immobilised with the drug curare, which blocks neuromuscular transmission. When the skin is touched, the pattern of spikes in motor neuron axons is the same as in free-

Touch

Touch

80ms

20mV

80ms

20mV

Rebound (c) spike

I 25mV

Current

I 25mV

Current

100ms

Control

Control

NMDA antagonist

10ms

2m V

Left

Right

Figure 7.10 The generation of the swim pattern in Xenopus tadpoles, (a) Tracings from a video recording of a short swim by a young tadpole, initiated by touching its skin. (b) Intracellular recordings from motor neurons of left and right trunk muscles during fictive swimming. During the swim, both neurons were depolarised from resting potential (dotted lines). The neurons spiked out of phase with each other, and received IPSPs in midcycle. (c) Rebound spike in a motor neuron. A small amount of steady, depolarising current was injected into the neuron, and a rebound spike occurred at the end of a brief pulse of hyperpolarising current. (d) Dual component EPSP in a motor neuron, caused by stimulating the axon of an interneuron.When the spinal cord was bathed in a solution containing an antagonist of the NMDA glutamate receptor, the slow and prolonged component of the EPSP disappeared. (e) The network that is thought to generate the rhythm for swimming. The descending interneurons (d) are involved in circuits of mutual excitation, and the commissural interneurons (c) convey inhibitory signals across the nerve cord so that excitation of motor neurons (mn) alternates in the two sides. (After Roberts, 1990; reprinted with permission from Science and Technology Letters.)

swimming tadpoles, demonstrating that the central nervous system contains a central pattern generator for swimming. Touching the skin excites sensory neurons called Rohon Beard cells and a single spike in one of these cells can be sufficient to elicit swimming. During a swim, a motor neuron remains depolarised from resting potential. It produces just one spike per swim cycle, and receives an IPSP when neurons on the opposite side fire (Fig. 7.10b). When tadpoles mature sufficiently to feed, the pattern of activity during swimming becomes considerably more complex, and motor neurons generate bursts of spikes (Sillar, Wedderburn & Simmers, 1991).

The pattern generator for swimming provides an excellent lesson on how both synaptic circuitry and intrinsic membrane properties shape the output of a central pattern generator. Like the other rhythmical behaviours examined here, interneurons generate the pattern and, as in the case of locust flight, different mechanisms for generating the rhythm operate in parallel. An essential element of the swim pattern is that motor neurons on the left and right sides are excited alternately. This is achieved by commis-sural interneurons that are excited in time with motor neurons on the same side as their cell bodies and dendrites. The axons of the commissural interneurons cross the spinal cord and inhibit motor neurons and interneurons of the opposite side.

The midcycle IPSP that these interneurons mediate has a dual role. Besides ensuring that motor neurons are inhibited when their contralateral partners are excited, an IPSP triggers the next spike in the motor neuron. The spike is called a rebound spike because it is triggered by the end of an event that hyperpolarised the motor neuron. A pulse of hyperpolarising current injected from a microelectrode can also be a trigger for a rebound spike. However, rebound spikes are not initiated if the membrane potential simply repolarises to resting at the end of a hyperpolarising pulse. A small, continual depolarising current must also be injected, to deliver a small amount of excitation at the end of the hyperpolarising pulse (Fig. 7.10c).

During a swim, steady excitation is provided through excitatory synapses from interneurons. The steady depolarising potential sets up a critical balance between currents carried through two types of voltage-sensitive channels. The first type is sodium channels, and entry of sodium through these channels tends to excite the neuron. The second type is potassium channels, through which potassium will leave the neuron. The current carried by potassium ions will increase as the neuron depolarises, opposing the excitatory action of the sodium channels. The effect of a brief hyperpolarisation is to close both types of channel. When the hyperpolarisation ends, the sodium channels open more rapidly than the potassium channels, allowing the generation of a single spike.

The steady depolarisation is necessary for the generation of a rebound spike because it enables the sodium channels to open following an IPSP. During swimming, it is provided by a population of interneurons that excite motor neurons on their own side of the nerve cord. There are probably circuits of mutual excitation among these interneurons, which helps to sustain swimming. Another factor that sustains excitation throughout a swim is that the EPSPs which these interneurons cause in motor neurons have relatively long durations, up to 200 ms, which is much longer than one swim cycle. This is because the EPSPs have two different phases, each mediated by different types of receptor for the neurotransmitter glutamate. The two components of an EPSP can be dissected apart by pharmacological agents (Fig. 7.10d). The first, fast component can be mimicked by applying the drugs kainate or quisqualate to the surface of a motor neuron, while the second can be mimicked by applying another drug, N-methyl-D-aspar-tate (NMDA). Both of these drugs are agonists of glutamate - they bind to receptors and activate their ionic channels in a similar way to glutamate. The motor neurons, therefore, have two different types of receptor for glutamate at their input synapses: one type causes the initial, fast excitation, whereas the second type, called the NMDA receptor, mediates the longer-lasting excitation. The fast EPSPs help excite the motor neuron at the appropriate time in a swim cycle, reinforcing the rebound from an IPSP.

A characteristic of the NMDA receptor is that its channel can only open if the postsynaptic neuron is already strongly depolarised, through the action of other synapses. This feature makes the NMDA receptor suitable for responding to combinations of signals, and it is heavily implicated in mechanisms of learning. In the tadpole spinal cord, strong excitation of motor neurons by the faster-acting, glutaminergic synapses is necessary to ensure that the NMDA-mediated EPSPs can switch on.

The mechanism for generating the swim pattern, therefore, depends on circuitry that ensures that left and right sides alternate in their activity and that there is mutual excitation between interneurons (Fig. 7.10e). Cellular properties underlie the sustained excitation of motor neurons by NMDA receptors, and the generation of rebound spikes. Simulation by computer has shown that circuits that have these characteristics can generate an output pattern like the tadpole swim pattern (Roberts & Tunstall, 1990). However, the interneurons that convey inhibition across the cord are not essential for the generation of a rhythm because when the spinal cord is split longitudinally, each half can generate rhythmically repeating activity on its own.

Some of the features of the swim generator of young tadpoles have also been discovered in neurons of the mammalian spinal cord, including rebound excitation and long-lasting EPSPs mediated by NMDA receptors. The relative complexity of movements that the mammalian spinal cord is capable of producing might arise partly from a greater diversity of types of neuron compared with the young tadpole. However, it is equally possible that the two spinal cords contain similar circuits of neurons but, in the mammal, there are many variations in the basic wiring of these circuits. Recent work on a relatively simple ganglion, the stomatogastric ganglion of decapod crustacea, has demonstrated that circuitry can, indeed, be reconfigured extensively, allowing groups of neurons to participate in quite different activity patterns.

Essentials of Human Physiology

Essentials of Human Physiology

This ebook provides an introductory explanation of the workings of the human body, with an effort to draw connections between the body systems and explain their interdependencies. A framework for the book is homeostasis and how the body maintains balance within each system. This is intended as a first introduction to physiology for a college-level course.

Get My Free Ebook


Post a comment