The marine mollusk Clione limacina, when swimming, can stabilize different body orientations in the gravitational field. The stabilization is based on the reflexes initiated by activation of the statocyst receptor cells and mediated by the cerebro-pedal interneurons that produce excitation of the motoneurons of the effector organs; tail and wings. Here we describe changes in the reflex pathways underlying different modes of postural activity; the maintenance of the head-up orientation at low temperature, the maintenance of the head-down orientation at higher temperature, and a complete inactivation of the postural mechanisms during defense reaction. Experiments were performed on the CNS-statocyst preparation. Spike discharges in the axons of different types of neurons were recorded extracellularly while the preparation was rotated in space through 360° in different planes. We characterized the spatial zones of activity of the tail and wing motoneurons and the CPB3 interneurons mediating the effects of statocyst receptor cells on the tail motoneurons. This was done at different temperatures (10 and 20°C). The “fictive” defense reaction was evoked by electrical stimulation of the head nerve. At 10°C, a tilt of the preparation evoked activation in the tail motoneurons and wing retractor motoneurons contralateral to the tilt and in the wing locomotor motoneurons ipsilateral to the tilt. At 20°C, the responses in the tail motoneurons and in the wing retractor motoneurons occurred reversed; these neurons were now activated with the ipsilateral tilt. In the wing locomotor motoneurons the responses at 20°C were suppressed. During the defense reaction, gravitational responses in all neuron types were suppressed. Changes in the chains of tail reflexes most likely occurred at the level of connections from the statocyst receptor cells to the CPB3 interneurons. The changes in gravitational reflexes revealed in the present study are sufficient to explain the corresponding modifications of the postural behavior inClione.
Modifications of an animal's behavior are caused by corresponding changes in the activity of the nervous mechanisms controlling the behavior. These changes can be attributed to specific nervous structures only if the control system has been analyzed in detail at the network and cellular levels. At present only a few systems, mainly those controlling rhythmic motor activity in invertebrates, meet such demands and allow exploration of the cellular substrate for modifications of their activity (for review seeAbbott and Marder 1998; Harris-Warrick 1988; Harris-Warrick and Marder 1991;Harris-Warrick et al. 1992; Katz 1995;Marder and Calabrese 1996; Selverston 1993,1995). Nervous mechanisms for postural control in the pteropod mollusk Clione limacina have been analyzed in detail in our previous studies (Deliagina et al. 1999a; Panchin et al. 1995a,b), which allowed us to address the question of what modifications of neural activity underlay different forms of postural behavior of this animal.
The marine molusc Clione limacina is a planktonic animal. It exhibits a distinct spatial orientation behavior. In the sea and in the aquarium at lower water temperature, Clione can usually be found oriented vertically, with its head up (Fig.1 A1 ).Clione swims upward or maintains itself at a particular depth by continuous beating of two wings (Arshavsky et al. 1985a,b; Satterlie and Spencer 1985;Satterlie et al. 1985). A deviation from the vertical orientation evokes a set of corrective motor responses including the upward flexion of the tail (Fig. 1 A2 ) and a change of the pattern of wing beating; these responses are aimed at restoration of the initial orientation. The postural reflexes are driven by input from two statocysts. After removal of both statocysts,Clione is not able to maintain any definite orientation and continuously loops in different planes (Panchin et al. 1995a).
The activity of the gravitational postural control system ofClione can be subjected to modifications related to different forms of behavior (Arshavsky et al. 1991a,b,1993a,b). In addition to the main mode of activity, that is stabilization of the head-up orientation, the system can switch to stabilization of the head-down orientation (Fig.1 B1 ) (Panchin 1997;Panchin et al. 1995a). This mode is observed under two conditions; when Clione leaves a water layer with higher temperature and at a certain stage of hunting behavior. WhenClione swims downwards, deviation from the vertical evokes the downward tail flexion (Fig. 1 B2 ).
A different modification is a complete cessation of postural activity. This is observed during the defense reaction caused by the tactile stimulation of the top of the head (Fig.1 C1 ) (Arshavsky et al. 1991a,1992; Satterlie et al. 1985). In response to this stimulation, Clone terminates wing beating and passively sinks. Becausae of a larger specific weight of the head,Clione gradually turns the head down, but no postural corrective responses are generated (Fig.1 C2 ) until the defense reaction ceases (Fig. 1 C3 ). Similar passive sinking, not accompanied by postural reflexes, is observed in a different form of behavior; the periodic vertical migrations (Litvinova and Orlovsky 1985; Panchin et al. 1995a).
Experiments with intracellular recording and staining have revealed the main neuron groups constituting the postural networks inClione (Panchin 1990; Panchin et al. 1995a,b). Information about spatial orientation is delivered by the statocyst receptor cells (SRCs) lining the wall of the statocyst cavity (Fig. 2). The SRCs contacting the statolith get excited. They activate different groups of cerebro-pedal interneurons which, in turn, evoke excitation of motoneurons of the effector organs (tail and wings).
Functional relationships between different cell groups of the tail and wing postural networks were established in the experiments on the CNS isolated with statocysts, when the activity of inter- and motoneurons was recorded during natural gravitational stimulation (Deliagina et al. 1999a). These experiments were performed at low temperature. They revealed the functional organization of the tail and wing reflexes responsible for the maintenance of the head-up orientation, which is observed at this temperature (Fig. 2,A and D). The diagram of Fig. 2 A shows the organization of tail reflexes. It illustrates the effect of a leftward deviation of Clione from the vertical. The deviation leads to excitation of the ipsilateral (left) SRCs. These SRCs, through the CPB3 interneurons, activate the motoneurons eliciting the contralateral (rightward) tail flexion, that is a corrective motor response aimed at restoration of the normal orientation.
Figure 2 D shows the organization of wing reflexes. The SRCs excited by the leftward tilt evoke excitation of a different group of cerebro-pedal interneurons (CPB2). In their turn, the CPB2 neurons activate the locomotor motoneurons 1A and 2A of the ipsilateral (left) wing. This leads to an increase of the amplitude of beating of the left wing and to turning of Clione to the right, toward the head-up position.
In the present study, we investigated the changes in the activity of the basic networks eliciting the tail and wing postural reflexes (Fig.2, A and D) when the animal modifies its postural behavior, that is during downward swimming (Fig.1 B1 ) and passive sinking (Fig.1 C2 ). The results of this study are summarized in the diagrams of Fig. 2, B, C, andE.
Experiments (n = 21) were carried out at the White Sea Marine Biological Station Kartesh. Mollusks were collected locally and kept in aquaria at 5–12°C not longer than 2–3 days. In preliminary experiments it was found if the mollusks were kept for a longer period their reaction to an increase of temperature (turning the head down) was considerably reduced.
Preparation and experimental chamber
The preparation and design of experiments were described in detail in the previous paper (Deliagina et al. 1999a). In brief, the experiments were performed on the in vitro preparation of the CNS (without buccal ganglia) isolated together with the statocysts and positioned in the recording chamber (petri dish, diam = 35 mm). Five silver wires were inserted into the chamber through its bottom to make contact with the electrodes, which were pieces of the filter paper soaked in the sea water. The CNS was positioned dorsal side up on the larger electrode and the nerves used for recording and stimulation were positioned on the smaller electrodes. The chamber was filled with paraffin oil and closed. A tubing system with the circulating water, attached to the bottom of the chamber, allowed us to maintain and regulate the temperature in the chamber within the range of 10–20°C.
Recording and stimulation
The method of extracellular recording from inter- and motoneurons was described earlier (Deliagina et al. 1999a). Activity of the tail motoneurons was recorded from the tail nerves N2(1) and N3 responsible for the tail flexion in the dorso-ipsilateral and ventral directions, respectively. Activity of the wing motoneurons was recorded from the wing nerve. Activity of the CPB3 interneurons was recorded from a stump of the transected subpedal commissure.
Gravitational stimulation of the statocysts was performed by rotating the chamber with the preparation (Deliagina et al. 1999a). Three modes of rotation were employed. In two of them, the longitudinal axis of the preparation (which corresponds to the longitudinal axis of the animal) was positioned vertically, and then the preparation was moved either in the sagittal plane (“sagittal sway,” α) or in the frontal plane (“lateral sway,” β). In the third mode, the axis was positioned horizontally and the preparation was rotated around it (“horizontal roll,” γ). Two patterns of rotation were used: 1) the full turn rotation, which was performed in successive 45° steps, each step was ∼5 s in duration (see e.g., Fig. 3 A) and2) trapezoid angular movements between two positions with an amplitude of ±90° (Fig. 3 B).
To evoke the “fictive” defense reaction, one of the head nerves supplying the top of the head was electrically stimulated (Arshavsky et al. 1993b).
In addition to natural gravitational stimulation of the statocysts, we used electrical stimulation of a limited area of the statocyst - either its right side or its left side. This was performed with a platinum-iridium wire electrode (diam = 40 μm) introduced into the chamber with the lid opened and positioned close to the corresponding aspect of one of the statocysts.
Temperature-dependent reflex reversal
As has been found in the previous study (Deliagina et al. 1999a), the tail motoneurons projecting to the right and left nerves N2(1), when tested at 10°C, are activated if the ipsilateral side is directed upwards. This was confirmed in the present study. As shown in Fig. 3 A, motoneurons in RN2(1) were activated within a wide zone around the right-side-up (R) position. When tested by the periodical trapezoid tilts, the neurons were also activated in this position and silent in the left-side-up (L) position (Fig.3 B).
At 20°C, the motoneurons dramatically changed their pattern of gravitational response—now they were activated in a wide zone around the L position (Fig. 3 C). A test with trapezoid tilts has also shown that the neurons were firing in the L position and silent in the R position (Fig. 3 D).
In contrast to the nerves N2(1), motoneurons both in the left and in the right nerves N3, when tested at 10°C, were activated in and around the ventral-side-up (V) position (Fig.4, E and F) (see also Deliagina et al. 1999a). At 20°C, the motoneurons from N3 were activated in and around the dorsal-side-up (D) position. In some cases it was evident that the gravitational response in the “warm” zone (around the D position) appeared when the response in the “cold” zone (around the V position) was still present (Fig.3 G). With further warming, the response in the V position disappeared (Fig. 3 H). In some other cases, the response in the cold zone disappeared earlier in the process of warming than the response in the warm zone appeared, so that gravitational responses in motoneurons were absent at some temperature (Fig. 4 B). The reversal of responses in the motoneurons at 20°C was observed in 14 of 16 experiments.
We also examined the effect of warming on the cerebro-pedal interneurons of group CPB3, mediating influences of SRCs on tail motoneurons. Activity of interneurons was recorded from their axons in the transected subpedal commissure (see methods). As shown in Fig. 4 A, at 10°C the interneuron in the right stump of the commissure (RSPC) was activated in the R position along with the motoneurons in RN2(1). During warming, gravitational responses in the inter- and motoneurons disappeared (Fig. 4 B) and reappeared at 20°C. At this temperature however, the responses were reversed; the RSPC interneuron and RN2(1) motoneurons were activated in the L position (Fig. 4 C). The responses reversed back with cooling (Fig. 4 D). The reversal of interneuron response was clearly seen in five of eight experiments. In the remaining three experiments, the activity of interneurons at 20° was difficult to detect at the background of high-amplitude units that appeared in the SPC at this temperature. These findings strongly suggest that the reversal of tail reflexes occurs at the level of connections from SRCs to interneurons.
The reversal at this level may be caused by two different reasons:1) by modification of the connections between the SRCs and CPB3 interneurons and 2) by modification of the response of SRCs to gravitational stimulation. To distinguish between these two hypotheses, we stimulated electrically a lateral aspect of the statocyst (see methods). Figure5 A shows a response to stimulation of the left side of the statocyst at 10°C. The stimulation led to a long-lasting activation of both CPB3 interneurons from the right SPC and tail motoneurons from the right N2(1) nerve. In contrast, the motoneurons in the left N2(1) were inhibited.
Warming the preparation initially caused a reduction of the excitatory responses of interneurons and motoneurons on the right side (Fig. 5,B and C). These responses disappeared completely and an excitatory response of motoneurons in the left N2(1) nerve appeared instead (Fig. 5 D). Cooling the preparation caused a return of the initial pattern of responses (Fig. 5 E). These effects were observed in all three experiments with electrical stimulation of the statocyst. Thus stimulation of the same SRCs at 10 and 20°C resulted in activation of different reflex pathways. This result strongly suggests that warming causes a switch of SRC-CPB3 connections.
The chain of gravitational wing reflexes is also subjected to temperature-dependent reconfiguration. Figure6 A shows gravitational responses of the wing motoneurons recorded at 10°C from the left and right wing nerves (LNW and RNW) along with the responses of tail motoneurons recorded from RN2(1). The larger units in the wing nerves represent discharges of 1A and 2A wing motoneurons (Arshavsky et al. 1985a,b). Because of the input from the locomotor rhythm generator, these motoneurons fired periodically at ∼0.7 Hz, and activity of the 1A and 2A alternated in each locomotor cycle. Under the effect of gravitational input, the 1A and 2A motoneurons exhibited their rhythmic locomotor activity only in the contralateral side-up position, and were silent in the ipsilateral side-up position. In this position however, the nonrhythmic motoneurons responsible for wing retraction (Huang and Satterlie 1990) were activated (see Deliagina et al. 1999a).
At 20°C (Fig. 6 B), the gravitational responses of tail motoneurons from RN2(1) and of the retractor wing motoneurons were reversed. In the 1A and 2A motoneurons, however, no changes in activity related to gravitational input were observed at 20°C, suggesting that they were functionally disconnected from the statocysts. Similar results were obtained in all four experiments with recording from the wing nerve.
Electrical stimulation of the head nerve (see methods) evoked an essential component of the defense reaction, suppression of locomotor activity, in all tested preparations (n = 5). Initially the preparation exhibited “fictive” locomotion (Fig.7 A), the discharges of 1A and 2A motoneurons in the wing nerve alternated with a frequency ∼1 Hz. Stimulation of the head nerve led to cessation of this regular rhythmic activity. In response to stimulation, the wing motoneurons fired a high-frequency burst and then became silent for >1 min. The locomotory rhythm returned afterwards. During all this trial, a sensitivity to gravitational input in the tail motoneurons was tested by trapezoid tilts. These stimuli evoked alternating responses of the tail motoneurons from LN2(1) and RN2(1) before nerve stimulation. During the defense reaction, the gravitational responses were initially almost completely inhibited and then gradually recovered to reach their normal value by the end of the defense reaction. Inhibition of gravitational responses, lasting for 0.5–2 min, was observed in four of five experiments with the head nerve stimulation. Recording from the subpedal commissure (n = 3) has shown that gravitational responses in the CPB3 interneurons disappeared along with the responses in the tail motoneurons (Fig. 7 B). This finding strongly suggests that suppression of gravitational reflexes occurs at the level of connections from SRCs to interneurons.
In the present study we examined changes in the activity of postural neuronal mechanisms in Clione responsible for different modifications of its postural behavior. It has been found that the basic postural networks, generating gravitational tail and wing reflexes and responsible for the maintenance of the head-up orientation (Fig. 2, A and D) (Deliagina et al. 1999a; Panchin et al. 1995a,b), are subjected to dramatic reconfigurations when Clione switches from the head-up to the head-down orientation at higher temperatures, or when it switches from the postural stabilization to the passive sinking during defense reaction. Two principal reconfigurations of the networks have been revealed: 1) reversal of gravitational reflexes and 2) suppression of gravitational reflexes.
Reversal of gravitational reflexes occurs with warming the CNS. In the network generating tail reflexes (Fig. 2 A) the reversal occurred in the reflexes tested by tilting the preparation in different planes [frontal (Fig. 3, A–D) and sagittal (Fig.3, E–H)]. These findings strongly suggest that the whole system of the tail reflexes is subjected to the temperature-dependent reconfiguration.
The reversal of responses in the tail motoneurons was accompanied by that in the cerebro-pedal interneurons. This was demonstrated in the experiments with the natural gravitational stimulation of the statocysts (Fig. 4) and with their electrical stimulation (Fig. 5). These findings strongly suggest that the reversal was caused by a suppression of the excitatory connections from the SRCs to the contralateral CPB3 interneurons and by an expression of the excitatory connections from these SRCs to the ipsilateral CPB3 interneurons (Fig.2 B). A different explanation for the modification of network was given by Panchin (1997) who suggested that an excitatory effect of SRCs on CPB3 neurons, observed at lower temperature, is transformed into an inhibitory effect at higher temperature. This suggestion contradicts, however, to the observations (Figs. 4 and 5) that stimulation of SRCs at higher temperature also evokes excitation but in a different cell group.
Our technique allowed recording only from a part of the cerebro-pedal neurons; namely, from those having axonal branches in the subpedal commissure (subgroups CPB3b and CPB3c, Panchin et al. 1995a,b). The neurons with a different morphology, the subgroup CPB3a also projecting onto the tail motoneurons, were not recorded in the present study, and their involvement in the reflex reversal can only be suggested by analoguey with the recorded neurons.
The present study has shown that a part of the gravitational wing reflexes, the wing retraction, is also subjected to reversal at higher temperatures, as demonstrated by the reversal of responses to tilt in the retractor wing motoneurons (Fig. 6). The interneurons mediating gravitational influences on the retractor wing motoneurons were not recorded in the present study, however, and it remained unclear if the switch of connections occurred at the level of connections from SRCs to interneurons (as in the tail reflexes, Fig. 2 B) or at the lower level.
Suppression of gravitational reflexes was observed during defense reaction both in the tail and in the wing postural networks. For the tail network it was also demonstrated that the suppression takes place at the level of connections from SRCs to interneurons (Fig.7 B); this is reflected in the diagram (Fig. 2 C). One can suggest that a similar mechanism such as inhibition of input from the statocysts to the CPB2 interneurons causes suppression of gravitational responses in the wing locomotor motoneurons (1A and 2A) observed at higher temperatures (Fig. 6). This supposition is reflected in the diagram (Fig. 2 E).
It seems most likely that the reflex reversal and the reflex suppression in Clione have common mechanisms. This idea is supported by the observations that a complete suppression of gravitational responses may occur at a certain stage of warming and precedes the reversal of responses (Fig. 4 B). Also, simultaneous presence of both cold and warm angular zones of gravitational responses at a certain stage of warming (Fig.3 G) can be explained by slightly different rates of suppression and expression of different connections from the SRCs to the CPB interneurons.
The cellular bases of this mechanism remains unclear, however. One possibility is that the reflex reversal is the result of a specific temperature sensitivity of the SRC-CPB synapses. It is known that synaptic transmission in mollusk may be strongly affected by changes of temperature (for review see Kehoe and Marty 1980). Another possibility is that special thermosensitive neurons regulate the efficiency of the SRC-CBP synapses. One cannot completely exclude the possibility that modulation of gravitational sensitivity of the SRCs under the direct effect of temperature or via their efferent innervation (Arshavsky et al. 1993a,b) may also play some role in the modifications of gravitational reflexes.
This work was supported by National Institute of Neurological Disorders and Stroke Grant NS-38022, Howard Hughes Medical Institute International Research Scholars Grant 75195-544801, Swedish Medical Research Council Grant 11554, and a Royal Swedish Academy of Science research grant for Swedish-Russian scientific cooperation.
Address for reprint requests: T. G. Deliagina, The Nobel Institute for Neurophysiology, Dept. of Neuroscience, Karolinska Institutet, SE-171 77 Stockholm, Sweden.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2000 The American Physiological Society