Characterization of a radula opener neuromuscular system in Aplysia.

1. Several lines of evidence suggest that the I7-I10 muscle group contributes to the radula opening phase of behavior in Aplysia; 1) extracellular stimulation of these muscles in reduced preparations causes the halves of the radula to separate, 2) synaptic activity can be recorded from muscles I7-I10 in intact animals when the radula is opening, and 3) motor neurons innervating I7-I10 are activated out of phase with retractor/closer motor neurons during cycles of buccal activity driven by the cerebral-to-buccal interneuron 2 (CBI-2). 2. All of the opener muscles are innervated by the B48 neurons, a bilaterally symmetrical pair of cholinergic motor neurons. B48 neurons produce excitatory junction potentials (EJPs) in opener muscle fibers that summate to produce muscle contractions. Contraction size is determined by the size of depolarization in muscle fibers and/or by action potentials that are triggered by summation of B48-evoked EJPs. 3. In addition to input from B48 neurons, opener muscles also receive excitatory input from the cholinergic multiaction neurons B4/B5. EJPs evoked by stimulation of neurons B4/B5 are 1/10 the size of B48-evoked EJPs. Consequently, changes in muscle tension produced by B4/B5 activity are relatively small. In contrast to B48 neurons, neurons B4/B5 are likely to be active during the closing/retraction phase of behavior. During cycles of buccal activity driven by neuron CBI-2, neurons B4/B5 fire in phase with closer/retractor motor neurons. Thus opener muscles may develop a modest amount of tension during the closing/retraction phase of behavior as a result of synaptic input from neurons B4/B5. 4. Opener muscles may also develop tension during closing/retraction simply by virtue of the fact that they have been stretched. When isolated opener muscles are lengthened, depolarizations are recorded from individual muscle fibers, and muscle tension increases. With sufficient changes in fiber length, action potentials are elicited. These action potentials produce twitchlike muscle contractions that become rhythmic with maintained stretch. Stretch-activated depolarizations are generally first apparent when muscle length is increased by 1 mm. Length changes of 4-5 mm are generally necessary to elicit twitchlike muscle contractions. Changes of 1-2 mm in muscle length are observed when the opener muscle's antagonist, the accessory radula closer, is activated in reduced preparations. 5. Stretch may also modulate B48-induced contractions of the opener muscles. When muscle length is increased, B48-elicited contractions of the I7 muscle are larger. These increases in contraction amplitude are accompanied by decreases in contraction latency. 6. We conclude that muscles I7-I10 contract vigorously in response to strong excitatory input from neuron B48 and contribute to radula opening. Stretch may potentiate this activity. Thus, if radula closer muscles contract vigorously and pull on the opener muscles, the opener muscles will respond by contracting more vigorously themselves. This may be a mechanism for maintaining amplitude relationships between antagonistic muscles. Additionally, it is likely that the opener muscles will develop at least a modest amount of tension during closure/retraction of the radula. Part of this activation may derive from the weak excitatory input that the muscles receive from neurons B4/B5. Another part may derive from the stretch. One function of this co-contraction may be to act as a brake on closure, bringing this phase of feeding behavior to a smooth halt.