INTRODUCTION

It has been reported that in several animals generation of receptor potential in the chemoreceptor cells was accompanied by changes in membrane conductance (Ozeki, 1971; Akaike et al., 1976; Suzuki, 1977; Sato and Beidler, 1982; Trotier and MacLeod, 1983; Tonosaki and Funakoshi, 1984; Anderson and Ache, 1985). Some of these reports described effects of extracellular cations: Na⁺ and Ca²⁺ ions might play important roles in the frog taste cells in response to salt stimulation (Sato et al., 1982), whereas Ca²⁺ ions were reported to be inward current carriers in the lamprey olfactory receptors in response to amino acid stimulation (Suzuki,1978). Effects of extracellular cations were also reported in insect sugar receptor cells: Na⁺ and/or K⁺ ions were thought to be involved in the generation of taste responses (labellar sugar receptors, Morita et al., 1966; tarsal sugar receptors, Broyles and Hanson,1976).

Receptor cells of the temporal organ of the Japanese house centipede Thereuonema hilgendorfi responded sensitively to CO₂, though they showed some additional responses to other air-borne chemicals. Based on physiological and morphological studies, the temporal organ is thought to be a CO₂ receptor (Yamana et al., 1986; Yamana and Toh, 1987, 1990). The receptor cell responded to CO₂ with hyperpolarizing receptor potentials, which caused a decrease of spontaneous impulse discharges (Yamana and Toh, 1987). The hyperpolarizing receptor potential was accompanied by an increase in the membrane resistance, but it is still unclear which kinds of ions contribute to the generation of the hyperpolarizing receptor potential. In order to answer this basic question, the effects of external ionic environments on CO₂ responses have been examined in the temporal organ of T. hilgendorfi.

MATERIALS AND METHODS

Adult Japanese house centipedes T. hilgendorfi of both sexes were used throughout this study. Animals were collected in the field around Kyushu University. Because the methods for recordings and perfusion experiments have been described in detail in the previous report (Yamana et al., 1986), they are dealt with here only briefly. After the animal was immobilized by cooling it with ice, its head was fixed in the chamber (1 ml in volume), and perfused with control and test solutions at a rate of 4 ml/min (details in RESULTS). An indifferent electrode (tungsten wire of 0.1 mm diameter) was inserted into the antenna and a recording electrode (glass-coated sharpened tungsten wire) was inserted into the temporal organ in order to record impulse discharges.

Carbon dioxide-free distilled water was prepared by boiling the distilled water for decarboxylation. The boiled water was cooled down while supplying carbon dioxide free air passed through NaOH solution as a CO₂ trap. The solution containing CO₂ was prepared by bubbling air containing CO₂ into distilled water. The solution containing CO₂ in equilibrium with air containing 5% CO₂ was used throughout the present experiment, and it is referred to in convenience as “CO₂ containing solution”. Details for solution containing CO₂ were described previously (Yoshii et al.,1980; Yamana et al., 1986). Salts used in the present study were dissolved in CO₂-free distilled water or in distilled water that contained CO₂.

RESULTS

Effect of perfusion

Receptor cells of the temporal organ discharged at a rate of about 20 impulses/sec in the CO₂-free air. When the temporal organ was perfused with CO₂-free distilled water (control solution), the impulse frequency decreased to 10 impulses/sec (Fig. 1). Impulse discharges in the air and in the control solution are referred to as spontaneous discharges in the present study. Replacement of the CO₂-free distilled water with the solution containing CO₂ resulted in a decrease in the discharge frequency of the receptor cells. Even in a perfused specimen, the receptor cell of the temporal organ could respond to a CO₂ stimulation.

Fig. 1

Time courses of the changes in impulse frequencies of the receptor cells after perfusates were changed from distilled water to CO₂-free salt solutions (100 mM NaCl, open circles; 50 mM KCl, closed circles; 2 mM CaCl₂, open squares) and distilled water in equilibrium with 5% CO₂(open triangles). All recordings were taken from the same receptor cell.

Effects of salt solutions on the spontaneous activities of the receptor cells

The activities of receptor cells were changed by replacement of the CO₂-free perfusate with test solutions, which were perfusates containing CO₂, NaCl, KCl or CaCl₂ (Fig. 1). Receptor cells were first adapted to the control perfusate for more than 5 min, and then the perfusate was replaced by test solutions. The effect of changing solutions appeared about 10 sec after the replacement of the solution. This latency is thought to be due to a slow exchange of the solutions around the receptor region.

The receptor cell responded to the solution containing CO₂ in a phasic-tonic manner: the impulse frequency rapidly decreased and it recovered to a steady level within a few minutes. On the other hand, changes of the impulse frequencies in response to perfusion of salt solutions appeared to be rather tonic.

The change in the impulse frequency of the receptor cell by perfusion of solution containing Ca²⁺ appeared with a delay that was longer by about 10 sec than those caused by other salt solutions and solution containing CO₂ (n = 7). Because the change of the impulse frequency reached a steady state about 120 sec after the beginning of the perfusion, the impulse frequency around 180 sec after initiation of the perfusion was analyzed. The activities of receptor cell adapted to six kinds of salt solutions under the CO₂ free condition were recorded and compared with each other. The spontaneous activities of the receptor cell increased when the temporal organ was perfused with NaCl or KCl solution, but decreased when perfused with CaCl₂ solution. On the other hand, MgCl₂, LiCl and choline chloride solutions had little effect upon the activity of the receptor cell (Fig. 2, n = 5).

Fig. 2

Relation between the concentration of salt solutions and receptor cell activities: NaCl (open circles), KCl (closed circles), LiCl (open diamonds), MgCl₂ (×), CaCl₂ (open squares) and choline chlo-ride (+). Relative impulse frequencies were measured 180 sec after replacement of perfusates from the distilled water to each salt solution in the CO₂-free condition. The impulse frequency adapted to CO₂-free distilled water is taken as 1.0. Recordings of NaCl, KCl, CaCl₂and choline chloride were from the same receptor cell and those of MgCl₂ and LiCl were from another cell.

The activity of the receptor cell increased as NaCl concentration increased, and did not reach a saturated level even at 100 mM NaCl in the concentration-activity curve (Fig. 2). The activity also increased as the concentration of KCl solution increased, and the impulse frequency was 1.9 times higher in 10 mM KCl than in KCl-free distilled water. The activities became irregular at concentrations greater than 100 mM KCl. On the other hand, CaCl₂ solution depressed the activities of the receptor cell (Fig. 3), and the depressing effect appeared at a lower concentration than the concentration in which the elevating effect of NaCl appeared (e.g., 0.5 mM, n = 4).

Fig. 3

Effects of low concentration of CaCl₂ on impulse frequencies after adaptation to distilled water. Dose-dependent reduction of impulse frequencies was the largest in the receptor cell shown in this figure (among four cells analyzed). The average activity of the receptor cells in 1 mM CaCl₂ was 7.9 ± 0.11 impulses/sec (mean ± S.E., n = 4).

Effects of salts on the response to carbon dioxide

The amplitude of the receptor cell responses to CO₂ application in several salt solutions was measured. Time courses of the change in the impulse frequency caused by CO₂ application in salt solution appeared similar to those in distilled water (the response to distilled water in Fig. 4).

Fig. 4

Time courses of changes in the impulse frequencies after perfusion with CO₂ solution. After adaptation to a given CO₂-free salt solution, the same solution but in equilibrium with air containing 5% CO₂ was perfused. (A) Effects of CO₂ in 20 mM, 50 mM and 100 mM NaCl solutions. (B) Effects of CO₂ in 10 mM, 20 mM and 50 mM KCl solutions. (C) Effects of CO₂ in 2 mM CaCl₂ solution. Recordings shown in A, B and C were from different cells.

The effects of NaCl concentration on receptor responses were examined. The temporal organ was first perfused with CO₂-free NaCl solutions (20, 50 and 100 mM). After adaptation to each concentration of NaCl solutions, the perfusion was changed to NaCl solutions of the same concentration with CO₂ (Fig. 4A). The impulse frequency depended on the concentration of NaCl under CO₂-free condition: it is higher in the higher concentration of NaCl, as shown at time 0 in Fig. 4A. After the application of CO₂ the impulse frequencies were reduced, but they were almost independent of the Na⁺ concentration, falling to the same level in the three different NaCl concentrations (n = 4).

The effects of KCl on the responses were also examined in the same manner as those of NaCl (Fig. 4B). The receptor cells showed higher activities in the higher concentrations of KCl in CO₂-free solution as shown at time 0 in Fig. 4B. The impulse frequency was reduced by the application of CO₂, but the effects of KCl still remained at the examined concentrations (10, 20 and 50 mM), the impulse frequency being higher in the higher concentration of KCl (Fig. 4, n = 3).

Perfusion of the CaCl₂ solution led to effects on the receptor activities that were different from those of NaCl and KCl. The receptor activities adapted to CaCl₂ solution were lower than those in Ca²⁺-free solution, as shown at time 0 in Fig. 4C. Application of CO₂ reduced impulse frequencies to zero within 20 sec in both solution containing Ca²⁺ and Ca²⁺-free solution. The impulse frequencies recovered to steady state levels 180 sec after the onset of CO₂ perfusion. The impulse frequency at the steady state was almost the same between Ca²⁺-free solution and solution containing Ca²⁺ (30–50% of the initial frequencies), but the recovery appeared faster in solution containing Ca²⁺ than Ca²⁺-free solution (observed in 5 cells out of 8 experiments).

The effects of different concentrations of the salt solution on the CO₂ responses were examined (Fig. 5). In both Ca^2+-free solution and solution containing Ca²⁺, impulse frequencies increased when the KCl concentration was increased (Fig. 5A). Impulse frequencies also increased when Na⁺ concentration in CO₂-free solutions was increased, but they slightly increased (Fig. 5A) or did not change in solutions containing CO₂ (Fig. 5B). In summary the different effects of Na⁺ and K⁺mean that an increase in the concentration of KCl resulted in an increase in the impulse frequency of the receptor cell, regardless of presence or absence of CO₂ in the perfusate, whereas an increase in the concentration of NaCl affected the impulse frequency only during the absence of CO₂.

Fig. 5

Relationship between receptor cell activities and the concentration of salt solutions in the presence and absence of CO₂. Impulse frequencies adapted to salt solutions with (broken lines) or without (solid lines) CO₂ are plotted. (A) NaCl (open circles) and KCl (closed circles). (B) NaCl (open circles) and CaCl₂ (open squares). Recordings of A and B were from different cells.

The impulse frequencies of receptor cells declined when the Ca²⁺ concentration was increased in CO₂-free perfusate, whereas the cells kept their low frequency levels regardless of an increase in Ca²⁺ concentration (Fig. 5B).

The activities of receptor cells differed among different salt solutions as shown in Fig. 1. The way in which the impulse frequencies were reduced by CO₂ application is presented for each of 50 mM NaCl, 20 mM KCl and 2 mM CaCl₂solution (Fig. 6). The amplitude of the responses, and the reduction of the impulse frequencies induced by CO₂ application were strengthened in NaCl perfusion, weakened by CaCl₂perfusion, and little changed by KCl perfusion, as compared with responses to CO₂ measured for distilled water without any kind of salt.

Fig. 6

Time courses of reduction of impulse frequencies by CO₂. Receptor cells are first adapted to CO₂-free solutions of 50 mM NaCl (open circles), 20 mM KCl (closed circles), 2 mM CaCl₂ (open squares) and distilled water (open triangles). These perfusates are switched to the same solutions, but are in equilibrium with air containing 5% CO₂. The reduction of impulse frequency is the largest in the NaCl solution. The time course and degree of reduction are similar between distilled water and KCl solution. All recordings were from the same receptor cell.

DISCUSSION

It has been reported that in the temporal organ of the Japanese house centipede T. hilgendorfi, the hyperpolarizing receptor potential was accompanied with a decrease in the membrane conductance of the receptor cell, and the decrease in impulse frequency was proportional to the amplitude of the hyperpolarizing receptor potential (Yamana and Toh, 1987). Although only impulse frequencies were recorded in the present study, they can be used for the interpretation of conductance changes of the receptor membrane, owing to the above-mentioned relationship among impulse frequencies, amplitude of receptor potentials and membrane conductance.

In the present study, Na⁺, K⁺ and Ca²⁺ ions in the perfu-sate influenced the activity of the receptor cells to varying degrees depending upon their concentration. Other ions such as Mg²⁺, Li⁺ and choline⁺ had no effects on the activity of the receptor cell (Figs. 1 and 2). This lack of effects suggests that these cations would hardly permeate the receptor cell membrane.

The impulse frequency of the receptor cell increased with an increase in the concentration of K⁺ ions in both CO₂-free and solution containing CO₂ (Figs. 4B and 5A). These results suggest that the permeability of the receptor cell membrane to K⁺ ions is not changed by carbon dioxide stimulation and that K⁺ ions may contribute little to the generation of receptor potential to CO₂ stimulation.

An increase in the concentration of Na⁺ ions resulted in an increase of the receptor cell activity in CO₂-free solution, but this effect of Na⁺ ions disappeared in solution containing CO₂ (Fig. 5). Moreover, the response amplitude of the receptor cell resulting from CO₂ application, which is expressed as a reduction of impulse frequency by CO₂ perfusion, was reinforced in 50 mM NaCl (Fig. 6). These results suggest that the decrease of Na⁺ conductance of the receptor cell may be involved in the generation of the hyperpolarizing receptor potentials. If the hyperpolarizing receptor potential to CO₂ stimulus observed in the previous work (Yamana and Toh, 1987) was caused by a decrease in permeability of receptor membrane to Na⁺ ions, then the ionic mechanism of CO₂ reception in the temporal organ would appear to be quite similar to that of vertebrate photoreceptor cells (Korenbrot and Cone, 1972).

Some of the data obtained in the present work appear contradictory. Perfusion of distilled water should largely reduce or abolish impulse discharges of the receptor cell, because Na⁺ ions, supposedly inward current carriers, do not exist even if sodium-specific channels remain open under unstimulated conditions. The perfusion of distilled water may also result in a lack of differences in impulse frequency between the presence and absence of CO₂ in perfusates. In the present study the impulse frequency declined by half in the distilled water as compared with the frequency in CO₂-free air, but it was not totally suppressed (Fig. 1). Moreover, the receptor cell could further reduce impulse frequency in response to CO₂ even in the distilled water. This conflicting phenomenon may be due to structurtal complexity of the temporal organ. During perfusion with distilled water some Na⁺ ions in the external environment remained in the mucous or constantly released from supporting cells. The ultrastructural study supports this view, because there are many closely packed supporting cell processes around receptor cell dendrites (Yamana and Toh, 1990). The decreased activity of the receptor cell in distilled water may be explained by the fact that the ionic environment of the mucous is diluted with distilled water during the perfusion with distilled water.

The effects of Ca²⁺ ions on the receptor cell response to CO₂ were complex. The impulse frequency decreased as Ca²⁺ion concentration was increased in CO₂-free solution. This effect disappeared in the perfusate containing CO₂ (Fig. 5B). Moreover, the response amplitude to CO₂ application of the receptor cell, reduction of impulse frequency by CO₂ perfusion, was small in CaCl₂ solution (Fig. 6). These results suggest that Ca²⁺ ions may participate in the primary process of CO₂ reception.

Changes in Ca²⁺ conductance in the chemoreceptor cell by chemical stimuli have been reported in the channel catfish (Restrepo et al., 1990), where amino acid stimuli triggered an influx of Ca²⁺ into the receptor cells via inositol-1,4,5-trisphosphate activated channels. This Ca²⁺ influx is proposed to be one of the mechanisms of the olfactory transduction. The influx of Ca²⁺ ions through non-selective cation channels has been reported to be regulated by stimulants (Nakamura and Gold, 1987).

It is unlikely that a change in Ca²⁺ permeability in the receptor cell of the temporal organ will result in the generation of a hyperpolarizing receptor potential, because to increase the concentration of extracellular Ca²⁺ would lead to an increase in Ca²⁺ influx, and influx of cation must result in the depolarization of the receptor cell. However, in the present study, Ca²⁺ ions decreased the impulse frequency of the receptor cell. It is more likely that Ca²⁺ may be involved in or regulate intracellular signal transmission in the chemo-electric transduction as is proposed to occur in other olfactory systems (Pace et al., 1985; Sklar et al., 1986; Nakamura and Gold, 1987; Matthews, 1991).

Many questions remain to be answered by more elaborate experiments, but the involvement of the sodium ion in the generation of hyperpolarizing receptor potential, and some indirect effects of calcium ions are proposed to be involved in the primary process of the carbon dioxide reception in the temporal organ of T. hilgendorfi.