ABSTRACT
In the marine ciliate Euplotes raikovi, pheromone released by a complementary mating type (nonself pheromone) induces typical sexual behaviour, whereas self pheromone released by the same mating type generally has no effect. Nonself pheromone evokes a reduction of the mean walking speed by 66 %, a threefold increase in the frequency and duration of long-lasting rest phases and a doubling in the number of side-stepping reactions. Consequently, translocation is strongly reduced and the cells remain in a small area. This could increase the probability of finding a sexual partner for pair formation (conjugation). The usual pattern of rhythmic, spontaneous depolarizations controlling the walking rhythm is absent in nonself-pheromone-stimulated cells. The remaining depolarizations arise from a 4 mV hyperpolarized membrane potential and do not reach the usual amplitudes of 15–20 mV but only of 6–10 mV. In addition, the amplitudes of K+ currents are increased at depolarizations of more than 20 mV by at least 30 %. Hyperpolarization- and depolarization-activated Na+ current amplitudes are increased, whereas the Ca2+ current amplitude remains nearly unaffected.
Introduction
Some species of ciliates can be divided into a variety of mating types distinguished by chemical substances termed gamones or pheromones (Dini and Nyberg, 1993; Mulisch, 1996; Raffioni et al., 1992). These species-specific molecules are either anchored to the cell membrane, as in Euplotes crassus and Paramecium spp. (Mulisch, 1996; Kitamura and Hiwatashi, 1980; Kitamura, 1988), or released into the surrounding medium, as in Euplotes raikovi (Ortenzi and Luporini, 1995; Luporini et al., 1983, 1995). In Euplotes raikovi, they have been identified as polypeptide pheromones and are closely related to each other by their molecular architecture (Mronga et al., 1994; Luporini et al., 1995). Each mating type produces a specific protein that is either incorporated into the plasma membrane as a receptor protein or released as a pheromone protein. Both proteins are produced as isoforms by alternative mRNA splicing (Luporini et al., 1995). The pheromone protein binds with similar, but not identical, affinities to the receptor protein of its own mating type and to the receptor protein of other mating types. Pheromone of its own mating type (self pheromone) and pheromone of complementary mating types (nonself pheromone) bind in a competitive fashion to the same receptor protein, and the quantitative relationship between bound pheromone types determines whether the vegetative or the mating stage should follow (Ortenzi and Luporini, 1995). A preponderance of bound nonself pheromone induces mating and generates specific mating behaviour that leads to the formation of cell pairs (conjugation).
In ciliates, the membrane potential and the intraciliary Ca2+ concentration control the powerstroke direction, the frequency of ciliary beating (Eckert, 1972; Naitoh and Kaneko, 1972) and, thereby, the translocation of cells (Machemer and Sugino, 1989; Lueken et al., 1996). In Paramecium spp., depolarizing receptor potentials activate Ca2+ conductances within the ciliary membrane, and the resultant rise in the intraciliary Ca2+ concentration evokes ciliary reversal and thereby backward swimming. Hyperpolarizing receptor potentials, induced by augmentation of a K+ conductance, cause faster forward swimming by increasing the frequency and amplitude of ciliary beating (Machemer, 1989; Machemer and Teunis, 1996).
In Euplotes raikovi, as described for Euplotes vannus (Lueken et al., 1996), a pattern of rhythmic, spontaneous depolarizations controls a rhythm of speed decrease and increase in forward-walking cells. Prolonged depolarizations induce decelerations to zero velocity followed by fast backward jerks as in Bursaridium difficile (Berg and Sand, 1994). The present study describes nonself-pheromone-induced modifications of this depolarization pattern as well as of walking behaviour.
Materials and methods
Cells and culture
Complementary mating types II and XI of Euplotes raikovi (termed Euplotopsis raikovi by Borror and Hill, 1995) were used. They belong to the clones 1bF113 and 1aF13N, respectively (Luporini et al., 1983), and were obtained from Dr P. Luporini (Department of Cell Biology, University of Camerino, Italy). Cells were cultivated in artificial sea water and fed with bacteria as described previously for Euplotes vannus (Lueken et al., 1987). The voltage-clamp-induced individual current components from the total current pattern of Euplotes raikovi were compared with the ion currents of Euplotes vannus (clone D35; termed Moneuplotes vannus by Borror and Hill, 1995) formerly identified by our group (see Fig. 4). All experiments were performed at room temperature (20–22 °C) on cells that had been starved for 13 days.
Solutions
Artificial sea water (ASW) was prepared from analysis-grade salts (in mmol l−1): NaCl, 465; NaHCO3, 2.5; KCl, 10; CaCl2, 10.4; MgCl2, 24.8; MgSO4, 28.1. The pH was adjusted with HCl to 8.0–8.1. Behavioural as well as electrophysiological experiments were performed in ASW.
Pheromone solutions were obtained by sterile filtration of the culture medium from either mating type. They were applied to the complementary mating type or, as a control, to the pheromone-producing mating type. The effectiveness of nonself-pheromone solution was assessed by the induction of intraclonal conjugation.
Behavioural experiments
The observation chamber was a plastic Petri dish (Nunc, diameter 35 mm), designed for cell culture purposes, filled with 200 μl of medium containing cells and 1800 μl of experimental solution. Cells moving on the bottom of the chamber were observed using an inverted microscope (Olympus CK2) and video-recorded at 50 frames s−1 under darkfield illumination. Randomly selected walking tracks were digitized in real time using a frame grabber board (PIP Video digitizer board 1024B, Matrox, Canada) and BIOTRACK software (Fa. Vogel Bildverarbeitungssysteme, D-91054 Erlangen, Germany). These tracks were resolved into time segments of 80 ms using WINTRACK software, version 1.6 (Fa. Vogel). Distances were calibrated using an object micrometer, and direct measurements were made on the video monitor. Recordings were made using Panasonic cameras (WV-BL 600 and WV-CL 502) and Panasonic recorders (AG-6720-E and NV-FS200-EG). The latter has a slow-motion and single-image mode that produces half-image steps of 20 ms.
Electrophysiological recordings
Only cells of mating type II were investigated using electrophysiological methods. By analogy to behavioural experiments, 900 μl of either pheromone solution was added to 100 μl of cell suspension. No further change of solution was made because such changes induce a loss of mating competence in microelectrode-penetrated cells. Therefore, cells were investigated either in self or in nonself pheromone. The numbers of cells tested in each solution are given in the figure legends.
Shortly (5 min) after cell transfer, most cells had settled at the bottom of the chamber. One of these cells was captured using a suction micropipette, penetrated with a single microelectrode and finally released from the pipette. Cells could rotate around the tip of the inserted microelectrode. A single-electrode voltage-clamp system (npi SEC 1L, H.-R. Polder, D-71732 Tamm, Germany), an analog-to-digital interface board (Tecmar, Scientific Solutions Division, Solon, OH, USA) connected to a personal computer and pCLAMP software version 6.0.2. (Axon, Foster City, CA, USA) were used. Switching frequency was 10 kHz with a duty cycle of 50 %. Recordings were filtered on-line at 1 kHz with an eight-pole Bessel filter (48 dB per octave). Microelectrodes, filled either with 0.5 mol l−1 KCl or 0.5 mol l−1 CsCl to block K+ channels, had resistances of 30–40 MΩ. Microelectrodes filled with 0.33 mol l−1 KCl were used to reduce the electromotive force for outwardly directed K+ currents. Direct changes of the pheromone-containing extracellular K+ concentrations were not possible because the pheromone solutions were obtained from the culture medium with a standard K+ concentration and because of problems associated with mechanical agitation during solution changes. The free-running membrane potential was recorded in the current-clamp mode at zero-current injection at a sampling rate of 1000 Hz for 1 min periods. During voltage-clamp experiments, the holding potential was adjusted with respect to the cell-specific resting potential at which the transverse cirri did not move. This resting potential is not stable in Euplotes raikovi owing to spontaneously generated depolarizations. The lowest level of these membrane potential fluctuations, i.e. the resting potential, is usually preferred to the depolarized levels and will be termed the preference potential. The current–voltage diagrams show currents activated by voltage steps (ΔVm) in relation to the holding potential, which was set to equal the preference potential for each cell. Differences in the preference potentials of different cells of up to 6 mV are evened out by this method of data collection and presentation. This method also allows the calculation of a mean voltage threshold for the activation of ciliary reversal that was determined with incremental, depolarizing voltage steps of 2 mV. Passive membrane properties were determined by applying a set of ten 90 ms current injections of −0.1 nA. The input resistance was derived from the mean ohmic voltage drop at 80 ms and was used to correct the current recorded by subtracting the linear leakage current.
The significance of all electrophysiological data was tested using the Mann–Whitney U-test (P<0.05). Values are presented as means ± S.E.M.
Results
Behaviour of pheromone-stimulated cells
In self-pheromone solution, cells covered long distances by walking forward (Fig. 1) at a mean speed of 162.3±13.2 μm s−1 (N=12). In nonself-pheromone solution, the mean speed was reduced to 55.6±7.9 μm s−1 (N=12; P<0.05). Translocation of nonself-pheromone-stimulated cells was low, because cells in self pheromone performed 11±4.2 stops min−1 (N=12) and 9±2.2 ciliary reversals min−1 (N=12), whereas cells in nonself pheromone performed 34±5 stops min−1 (N=12) and 18±2.6 ciliary reversals min−1 (N=12); the number of stops does not include the stop during a ciliary reversal. Consequently, the speed distribution of nonself-pheromone-stimulated cells clearly shifted to smaller values (Fig. 1C). In solution without any pheromone, cells showed no statistically significant differences from cells in self pheromone. They walked at a mean speed of 186.6±23.1 μm s−1 (N=9) and performed 9.3±3.1 ciliary reversals min−1 (N=9) and 13.6±8.3 stops min−1 (N=9).
Effects of pheromone on membrane potential
In Euplotes raikovi, as described previously for Euplotes vannus (Lueken et al., 1996), a pattern of rhythmic, spontaneous depolarizations controls a rhythmical reduction in speed of forward-walking cells. In nonself-pheromone-stimulated cells and with 0.5 mol l−1 KCl in the microelectrode, the frequency and duration of depolarizations were reduced (Fig. 2). The amplitude of the depolarizations was 15–20 mV in controls and 6–10 mV in nonself-pheromone-stimulated cells. This range was reduced in nonself-pheromone-stimulated cells because of the reduced amplitude of depolarizations. In addition, the preference potential shifted by approximately —4.3 mV in nonself-pheromone-stimulated cells (Table 1, asterisks). Rare hyperpolarizations in the latter cells occurred spontaneously, but repeatedly. Fluctuations of membrane potential in cells measured in self-pheromone solution were similar to those in cells that were exposed to a pheromone-free solution (Fig. 2A,B). This demonstrates that the exposure to nonself pheromone itself, rather than the absence of self pheromone, inhibits the generation of spontaneous potential fluctuations.
The effects of nonself pheromone failed to occur when the microelectrode was filled with 0.33 mol l−1 KCl instead of 0.5 mol l−1 KCl (Fig. 3; Table 1). A reduction in the K+ concentration in the microelectrode might result in a decrease in the intracellular K+ concentration, i.e. a decreased K+ equilibrium potential. A lower K+ equilibrium potential reduces the electromotive force for outwardly directed K+ currents. Consequently, the preference potential would shift to more depolarized values, and the probability of generating potential fluctuations would be increased. Indeed, in nonself-pheromone-stimulated cells with 0.33 mol l−1 KCl in the microelectrode, the same fluctuation pattern occurred as was found in cells exposed to self pheromone and using 0.5 mol l−1 KCl microelectrodes (compare Fig. 3B with Fig. 2B).
Effects of pheromone on input resistances
The input resistances of cells in nonself pheromone (147±19 MΩ, N=15) were higher than those in cells in self pheromone (95±8 MΩ, N=8) when electrodes had been filled with 0.5 mol l−1 KCl. When electrodes had been filled with 0.33 mol l−1 KCl, no significant difference was found between cells in self-pheromone solution (71±4 MΩ, N=4) and in nonself-pheromone solution (80±5 MΩ, N=4).
Effects of pheromone on ion currents
The pattern of membrane currents found in E. raikovi is similar to that in E. vannus (Krüppel and Lueken, 1988). Because of the smaller surface area of E. raikovi, the current amplitudes in this species were considerably lower than in E. vannus (Fig. 4). In E. vannus, the hyperpolarization-activated inward current is carried by Na+ (Krüppel, 1993), and the depolarization-activated inward current is composed of an early, voltage-dependent, transient Ca2+ current (Krüppel and Wissing, 1996) and a late Ca2+-dependent Na+ current (Krüppel and Lueken, 1990). The fast and slow outward K+ currents (IK,fast and IK,slow; Krüppel et al., 1991; Westermann et al., 1990) (Figs 4, 5) and the inward Ca2+ and Na+ currents (see Fig. 6) could also be distinguished in E. raikovi, but the plateau of IK,slow was influenced by a small inward Na+ current (INa).
At depolarizations of more than 20 mV, IK,fast and IK,slow were increased by at least 30 % in nonself-pheromone-stimulated cells (Fig. 5A,B). In addition, the reversal potential of total currents (predominantly carried by IK,slow and to a small amount by INa, see above), measured after 140 ms of 160 ms voltage pulses, was shifted by approximately —4.2 mV (Fig. 5C). This suggests either that the Na+ current (INa) was decreased or that IK,slow was increased (or a combination of these two effects) in the presence of nonself pheromone. At low levels of depolarization, however, the late, depolarization-activating inward current component was slightly increased in nonself-pheromone-stimulated cells.
The voltage-dependent transient inward Ca2+ current remained nearly unaffected by nonself pheromone (Fig. 6). Therefore, the increase in IK,fast is not associated with any reduction in the Ca2+ current. Summation of the Ca2+ current with the late K+ current can be excluded owing to the fast inactivation of the Ca2+ current (Krüppel and Wissing, 1996).
The late hyperpolarization-dependent inward Na+ current is activated by nonself pheromone at hyperpolarizations of —10 mV, whereas in the presence of self pheromone this current occurred at hyperpolarizations of approximately —40 mV (Fig. 7).
Effects of pheromone on the threshold for ciliary reversal
Ciliates reverse the beat direction of their cilia (e. g. Paramecium spp.) or their cirri (e.g. Stylonychia/Euplotes spp.) when they are depolarized to a voltage threshold that activates inward Ca2+ currents (Machemer and Deitmer, 1987; Lueken et al., 1996). In E. raikovi, this threshold was lowered by nonself pheromone from 8.6±0.7 mV (N=8) to 6.3±0.5 mV (N=8; P<0.05) when 0.33 mol l−1 KCl electrodes were used.
Discussion
In ciliates, successful sexual stimulation by pheromones is indicated by a change of locomotory behaviour resulting in the formation of pairs (Ricci, 1996). In both E. raikovi and in E. vannus (Lueken et al., 1996), rhythmical, spontaneous fluctuations in membrane potential correspond with a rhythmical reduction in the speed of forward-walking cells. Inhibition of these rhythms allows pair formation. In the present study, we have analysed pheromone-induced changes in behaviour and in the electrophysiological properties of E. raikovi.
Behaviour of pheromone-stimulated cells
After stimulation with nonself pheromone, walking cells remain in a small area (small radius of action) because they make frequent long-lasting stops, more frequent side-stepping reactions and their walking speed is reduced. A slower speed reduces the intensity of cell/cell collisions and thereby the intensity of the typical mechano-responses (the ‘avoiding reaction’ and ‘escape response’) that would normally separate cells. The general reduction of translocation could increase the probability of finding a conjugation partner.
Effects of pheromone on membrane potential
In the ciliates Stylonychia mytilus, Euplotes vannus and Bursaridium difficile, spontaneous membrane potential depolarizations have been recorded in standard solutions (Machemer and Deitmer, 1987; Krüppel and Lueken, 1988; Berg and Sand, 1994). In Paramecium aurelia, however, such depolarizations are induced only in Na+-containing solutions (Satow and Kung, 1974). Presumably, the ratio of Ca2+, K+ and Na+ conductances at the resting potential determines whether spontaneous activity is generated (Hinrichsen and Schultz, 1988). In Euplotes raikovi, the spontaneously generated depolarizations are inhibited in nonself-pheromone-stimulated cells as a result of a slight hyperpolarization of the preference potential (see Fig. 2). Such a hyperpolarizing effect might be caused by an increased resting conductance for K+ or a reduction in the resting conductance for Ca2+ and/or Na+ (gK+, gCa2+ and gNa+, respectively). The input resistance, which is inversely proportional to the total membrane conductance, does not change in nonself-pheromone-stimulated cells when 0.33 mol l−1 KCl electrodes are used and is increased when 0.5 mol l−1 KCl electrodes are used, indicating that a reduction in gCa2+and/or gNa+ is more likely than an increase in gK+.
How the different intracellular K+ concentrations that presumably result from the use of the two types of K+ electrode produce these different effects on the input resistance remains obscure. The spontaneous membrane potential depolarizations are not inhibited by nonself pheromone when 0.33 mol l−1 KCl electrodes are used. If the intracellular K+ concentration changes from 0.5 to 0.3 mol l−1, then the K+ equilibrium potential should become 10 mV more positive (extracellular K+ concentration was 10 mmol l−1). Consequently, the total membrane potential should be depolarized to some extent, and any pheromone-induced hyperpolarizing effect might be compensated. The same result should be obtained, without the uncertainty of the real intracellular [K+], by increasing the extracellular K+ concentration. This method, however, was rejected in the present study to avoid mechanical disturbance of cells during the solution change (see Materials and methods). In fact, for this change in the electrode [K+], the preference potential shifts by approximately 15–18 mV (Table 1). This unexpectedly large shift could result from secondary effects such as an amplification of depolarization through an increase in potential-dependent Na+ and/or Ca2+ conductances.
Effects of pheromone on ion currents
In nonself-pheromone-stimulated E. raikovi, the depolarization-activated total outward current is increased. The early outward currents are predominantly carried by K+, and the late outward current is carried by K+ and to lesser extent by Na+ (see Fig. 4). A late-activating inward current is increased at low depolarizations, and a large outward current is increased at depolarizations above 20 mV (see Fig. 5). The latter data are best explained by an augmentation of a Na+/K+ conductance which passes ion currents dependent on the driving force for either ion. At low levels of depolarization, i.e. at a high driving force for Na+, an inward Na+ current would result, and at high levels of depolarization, i.e. at high driving force for K+, an outward K+ current would result. In E. vannus, two depolarization-activated, late ion currents have been identified which have been named IK,slow and INa,Ca (Westermann et al., 1990; Krüppel and Lueken, 1990). It is not yet known whether these ionic currents use the same or two different channel populations. The slight increase of inward current at small depolarizations and the general increase in the outward currents at larger depolarizations would also help to explain the occurrence of the small number of low-amplitude depolarizations observed in nonself pheromone (see Fig. 2C). Any spontaneous depolarization would be repolarized quickly before the usual amplitude is reached by an increase of these conductances.
In addition to the pheromone-induced increase in outward currents, a hyperpolarization-activated inward current is also induced in these cells. By analogy to the inward rectification in E. vannus (Krüppel, 1993), this inward current in E. raikovi may be carried by Na+. In Paramecium tetraurelia, one inward Na+ current activates upon hyperpolarization (Saimi, 1986) and another upon depolarization (Saimi and Kung, 1980), giving rise to speculations that both ionic currents flow through the same population of channels. If the hyperpolarization-induced inward current and the depolarization-induced currents in E. raikovi use the same population of channels, the nonself pheromone would act by increasing one common conductance.
Ciliary reversal is based on the free Ca2+ concentration inside the cilia (Mogami et al., 1990, 1991). The necessary increase in Ca2+ concentration is achieved by a voltage-gated inward Ca2+ current (Pernberg and Machemer, 1989, 1995). In nonself-pheromone-stimulated E. raikovi, the number of stops and ciliary reversals is augmented and the threshold potential for ciliary reversals is lowered, although the depolarization-activated inward Ca2+ current is unaltered (see Fig. 6). Consequently, the augmentation of the intraciliary Ca2+ concentration is probably not caused by an increased influx of extracellular Ca2+ through the membrane of the cilia.
A hyperpolarization-activated inward Ca2+ current has been described in Paramecium tetraurelia (Preston et al., 1992). We can speculate on the existence of such an inward current in the present study, because nonself pheromone induces hyperpolarization in E. raikovi. Hennessey et al. (1995) and Evans and Keller (1997) describe different kinds of receptor-mediated, voltage-independent inward Ca2+ currents in Paramecium tetraurelia and Chlamydomonas reinhardtii. A Ca2+ conductance could exist in E. raikovi that is affected by nonself pheromone and either is superimposed upon other conductances or is unmeasurable for reasons not yet known. The release of Ca2+ from intracellular stores (Erxleben and Plattner, 1994; Länge et al., 1995; Plattner et al., 1991) could also lead to an increase in the intraciliary Ca2+ concentration that is not measurable using standard electrophysiological methods. In Paramecium caudatum, the injection of Ca2+ results in a reversal of ciliary beating and in an increased beat frequency without a change in resting potential level (Nakaoka and Machemer, 1990). We suggest that, in E. raikovi, nonself pheromone induces the augmentation of intraciliary [Ca2+] which additionally could be involved as a second messenger in preparing the cells for conjugation. In E. vannus, both the inward Na+ current and the outward K+ currents are Ca2+-dependent (Krüppel and Lueken, 1990; Krüppel et al., 1991). Accordingly, a nonself-pheromone-mediated increase in the intracellular Ca2+ concentration could have enhanced the Na+ and K+ conductances in E. raikovi. Inositol 4,5-trisphosphate (InsP3) could act as a second messenger releasing Ca2+ from alveolar sacs or from the endoplasmatic reticulum. Miwa and Wada (1995) unravelled the role of InsP3 in conjugation in Paramecium bursaria by demonstrating that cells began conjugation 30 min after the injection of 10 pl of a 120 μmol l−1 InsP3 solution. An alternative explanation for the effects of nonself pheromone is an enhancement of the sensitivity of the motility apparatus to the intracellular Ca2+ concentration, thus inducing responses that are normally associated with elevated Ca2+ concentrations.
In conclusion, nonself pheromone changes ion currents and locomotory behaviour, resulting in conjugation. The intraciliary and/or intracellular Ca2+ concentrations seem to play an important role in this process. Since the effects of nonself pheromone are multi-layered, there is a need for further investigations to elucidate the signal transduction pathway involved in the sexual behaviour of ciliates.
ACKNOWLEDGEMENTS
The authors would like to thank Dr Pierangelo Luporini for providing E. raikovi. This work was supported by DFG (Graduiertenkolleg ‘Molekulare Zellbiologie mit Schwerpunkt Membranbiologie’ and SFB 171, C7).