Photolysis of caged calcium in cilia induces ciliary reversal in Paramecium caudatum

The beat pattern and beat frequency of cilia and flagella in many kinds of eukaryote are controlled by changes in intracellular Ca²⁺concentration. Ciliary movement in Paramecium comprises two kinds of ciliary strokes, the effective stroke and the recovery stroke, and shows metachronal coordination (Machemer,1972). When Paramecia meet an obstacle, their cilia reverse the direction of their beat, allowing the organism to swim backwards(Jennings, 1906). Collision of a Paramecium cell with an obstacle is believed to mechanically stimulate the anterior portion of the cell. This stimulation is then thought to induce activation of mechano-sensitive Ca²⁺ channels in the cell membrane, resulting in depolarization of the cell(Naitoh and Eckert, 1969; Naitoh et al., 1972). This depolarization causes an increase in Ca²⁺ influx into cilia through the activated, voltage-sensitive Ca²⁺ channels in the ciliary membrane (Ogura and Takahashi,1976; Dunlap, 1977; Machemer and Ogura, 1979).

Using permeabilized P. caudatum cells, Nakaoka and Ooi(1985) found that cAMP inhibits Ca²⁺-induced ciliary reversal. Several investigators identified a 29 kDa axonemal protein that copurified with 22S outer-arm dynein, which is phosphorylated in a Ca²⁺- and cAMP-dependent manner (Hamasaki et al., 1989, 1991; Bonini and Nelson, 1990). Brief digestion of demembranated cilia with trypsin suppressed not only the inhibitory effect of cAMP on Ca²⁺-induced ciliary reversal but also the phosphorylation of the 29 kDa axonemal protein(Noguchi et al., 2000). These results suggest that the 29 kDa protein may be a mediator of the effect of Ca²⁺ on ciliary reversal. If this were the case, Ca²⁺would control the direction of ciliary beat not by acting on the ciliary base,but throughout the cilium, because the 29 kDa dynein-associated protein is present throughout the cilium.

To clarify the region of Ca²⁺ sensitivity with respect to ciliary reversal, Hamasaki and Naitoh(1985) iontophoretically applied Ca²⁺ to detergent-permeabilized cilia of P. caudatum. Their study revealed that the basal region of the permeabilized cilium is the most sensitive to Ca²⁺ in inducing ciliary reversal. In contrast, Tamm and Tamm(1989) showed that the Ca²⁺ sensitivity extended the length of cilia, using detergent-permeabilized macrocilia of Beroë mitrata. In an effort to more faithfully represent the natural conditions of cilia, however,we used living Paramecia, rather than a detergent-permeabilized system in the present study, to prevent any possible loss of Ca²⁺sensitivity through the permeabilization procedure. We adopted a photolysis method, instead, in which caged calcium (o-nitrophenyl EGTA[NP-EGTA]) was first injected into intact P. caudatum cells, followed by application of a UV-light pulse to a restricted area to release Ca²⁺ from the NP-EGTA. The results of this study suggest strongly that Ca²⁺ causes ciliary reversal by acting upon the entire cilium above the basal body.

Wild-type Paramecium caudatum Ehrenberg cells (kyk402) were cultured in an infusion of rice straw at room temperature (20–25°C). Prior to each experiment, Paramecium cells were transferred into standard saline medium containing 1 mmol l^–1 KCl, 1 mmol l^–1 CaCl₂ and 20 mmol l^–1Pipes-Tris, pH 7.0.

The method of micromanipulation was described previously (Iwadate et al., 1997, 1999b). Briefly, standard saline medium dispersed with Paramecium cells was mixed with the same volume of a Ca²⁺-containing, highly viscous medium (1% methyl cellulose, 1 mmol l^–1 KCl, 1 mmol l^–1CaCl₂ and 20 mmol l^–1 Pipes-Tris, pH 7.0) or a Ca²⁺-chelating, highly viscous medium (1% methyl cellulose, 1 mmol l^–1 KCl, 20 mmol l^–1 EGTA and 20 mmol l^–1 Pipes-Tris, pH 7.0) before microinjection of NP-EGTA. When Ca²⁺-chelating medium was used, the pH of the mixed medium was 6.95. The free Ca²⁺ concentration in the mixed medium was estimated to be below 10^–7.5 mol l^–1(Iwadate et al., 1999b). A Paramecium cell swimming slowly in the medium was caught at the tip of a suction pipette (about 35 μm in inner diameter).

NP-EGTA medium containing 60 mmol l^–1 Ca(OH)₂,100 mmol l^–1 NP-EGTA, 20 mmol l^–1dithiothreitol (DTT) and 100 mmol l^–1 Hepes-KOH, pH 7.2, or control medium containing 40 mmol l–1 Ca(OH)₂, 100 mmol l^–1 ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA), 20 mmol l^–1 DTT, 120 mmol l^–1Hepes-KOH, pH 7.2, was injected into the Paramecium cell according to the method of braking micropipette(Hiramoto, 1974). The free Ca²⁺ concentration in each medium was estimated to be 10^–6.9 mol l^–1(Iwadate et al., 1999a). The injected volume of each medium was approximately 8 pl, corresponding to approx. 3% of the whole cell volume.

Micromanipulation of the suction pipette and micropipette was carried out using two micromanipulators (MO-102N and WR-60, Narishige, Tokyo, Japan) and injection through the micropipette was performed using a microinjector(IM-50A, Narishige, Tokyo, Japan).

UV application to the restricted area was performed according to the methods described previously (Iwadate et al., 1999a), with slight modifications. An inverted microscope(TE300; Nikon, Tokyo, Japan) was used throughout the experiment. A mercury short arc lamp (USH102D; Ushio, Tokyo, Japan) was attached to the side light path of the microscope and used as the UV light source. A pinhole 400 μm in diameter (43-5305, Coherent, Tokyo, Japan) was arranged at the field plane conjugated to the plane of the specimen. The UV light through the pinhole was filtered through a band-pass filter of 270–400 nm (U340; Hoya, Tokyo,Japan) and a cutoff filter of 300 nm (UV30; Hoya, Tokyo, Japan) to pass 300–400 nm UV light, as described by Funatsu et al.(1993). The light was collected by a UV-transmitting objective lens (CFI S Fluor 40× H, NA 1.30; Nikon,Tokyo, Japan) to form a small image of the pinhole on the plane of the specimen (Fig. 1). Thus, the 300–400 nm UV light was applied to the restricted portion of the Paramecium cell, about 2 min after microinjection of NP-EGTA. The application time was controlled with an electromagnetic shutter (No. 0; Copal,Tokyo, Japan). UV light applied to the inner portion of the Paramecium cell was scattered by cytoplasmic particles. We determined the area where the rise in [Ca²⁺]_i(Δ[Ca²⁺]_i) occurred as described below.

As illustrated in Fig. 2,the basal body is located just under the terminal plate. The diameter of cilium is about 200 nm. Ciliary beating is always highly active, so imaging methods using a Ca²⁺ indicator such as Calcium Green are predicted to suffer from poor spatial and temporal resolution and are unlikely to allow accurate determination of Δ[Ca²⁺]_i.

A Paramecium cell contains several thousands of secretory vesicles named trichocysts, which in the resting state are located just beneath the plasma membrane (Fig. 2). The release of Ca²⁺ from cortical Ca²⁺ stores, called alveolar sacs (reviewed by Plattner and Klauke, 2001), triggers the fusion of trichocyst membranes with the plasma membrane. Just after the membrane fusion, released trichocysts expand 25 μm in length instantaneously (<1 ms) if the external medium contains Ca²⁺. Thus, the released trichocysts are easily observed by optical microscopy, as shown by an arrowhead in Figs 7A and 8A, for example. Klauke and Plattner (1997) indicated that the threshold level of [Ca²⁺]_i for fusion of trichocyst membranes with the plasma membrane is >1 μmol l^–1,which is the threshold level of [Ca²⁺]_i for ciliary reversal (Naitoh and Kaneko,1972). Iwadate et al.(1999a) directly demonstrated that the threshold level of [Ca²⁺]_i for membrane fusion is higher than that necessary for ciliary reversal. Thus, trichocyst exocytosis upon photolysis of NP-EGTA can be used as a marker of whether the[Ca²⁺]_i is greater than the threshold level for ciliary reversal. Electron microscopy studies revealed that the three structures(basal body, trichocyst and alveolar sac) are positioned very close to each other (Stelly et al., 1990; Knoll et al.,1993) (Fig. 2) and that Ca²⁺ release from the alveolar sacs reaches the neighboring basal bodies (Knoll et al.,1993). According to this logic, we judgedΔ[Ca²⁺]_i to have occurred at the basal bodies when trichocyst exocytosis was observed as shown in Figs 7A and 8A.

A photomultiplier tube (PMT) (R374; Hamamatsu Photonics, Hamamatsu, Japan),equipped with an electromagnetic shutter (No. 0; Copal, Tokyo, Japan), a high-pass filter of 500 nm (SC-50, Fuji Photo Film, Tokyo, Japan) and band-pass filter of 530 nm (35-3607, Coherent, Tokyo, Japan), was attached to the top light path of the microscope.

Calcium Green medium (2 pl) containing 1 mmol l^–1 Calcium Green-1 dextran 10 000 MW (Molecular Probes, Eugene, OR, USA) and 120 mmol l^–1 Hepes-KOH, pH 7.2, was injected simultaneously with NP-EGTA medium or control medium into the Paramecium cell. We used a blue light-emitting diode (blue LED, λ_max=473 nm; LSPB500S,Nichia, Tokushima, Japan) to excite the Calcium Green. Details of the method have been previously described (Iwadate and Kikuyama, 2001). The fluorescence intensity of Calcium Green was detected with the PMT.

To detect images of ciliary reversal, the PMT at the top light path of the microscope was removed and a CCD camera (XC-ST50, Sony, Tokyo, Japan) was attached in its place. The images were recorded on VHS videotape with a videotape recorder (HR-VX200, Victor, Tokyo, Japan). The images were then transferred to a computer (PC-9821Nr13, NEC, Tokyo, Japan) and analyzed with NIH-image version 1.62.

In highly viscous medium containing 0.5% methyl cellulose, ciliary movement composed of the effective stroke and the recovery stroke is altered. Under these conditions, the cilia remain relatively straight, and the apical end moves in a circular path as if drawing a small circle, such that the movement of each cilium traces a cone-like path. The ciliary angle (θ) was measured as shown in Fig. 3. A cilium was selected randomly, within (or outside of) the UV application area. Then, its angle (θ) was measured against the line (the line with one arrowhead in Fig. 3)perpendicular to the cell surface. If the cilium was positioned in an apical portion of the cell to the perpendicular line, the angle (θ) was shown as positive.

The threshold level of [Ca²⁺]_i for trichocyst exocytosis is higher than that necessary for ciliary reversal, as explained in Materials and methods. To use trichocyst exocytosis as a marker forΔ[Ca²⁺]_i at the neighboring basal bodies, the duration of UV application should be sufficient to raise[Ca²⁺]_i above the threshold level for trichocyst exocytosis.

We applied a UV light for 125 ms to an entire Paramecium cell that had been previously injected with NP-EGTA medium and Calcium Green medium. UV application caused a significant increase in fluorescence of Calcium Green(N=4) and caused both trichocyst exocytosis and ciliary reversal at the whole cell surface in all Paramecium cells tested (N=4). Representative results are shown in Fig. 4. By contrast, in cells injected with control medium alone, the fluorescence intensity did not change in response to UV application(N=3), nor did the cells show any trichocyst exocytosis or ciliary reversal (N=3). The duration of UV application was set as 125 ms in all subsequent experiments.

To determine whether Ca²⁺ influx through the Ca²⁺channels in the ciliary membrane was contributing to ciliary reversal, we applied UV light to cilia of Paramecium cells that had been first immersed in a Ca²⁺-chelating medium and injected with NP-EGTA medium. It is known that a low-Ca²⁺ condition is critical for Paramecia, so all experiments were carried out within 2 min after Paramecium cells were mixed with the Ca²⁺-chelating medium.

Application of UV light to cilia at various portions in the anterior of cells treated in this manner led to ciliary reversal, indicating that it was the increase in [Ca²⁺]_i caused by the photolysis of NP-EGTA, and not the Ca²⁺ influx through the Ca²⁺channels that directly induced ciliary reversal (N=16). Typical results are shown in Fig. 5A(Movie 1). Ciliary reversal occurred strongly only in the area of UV application. The angle of ciliary reversal was 2.2 rad.

The relationship between the distance from the rim of the UV-applied area to the cell surface and the angle of ciliary reversal is summarized in Fig. 5B. Smaller distances led to greater angles of ciliary reversal.

We next immersed a Paramecium cell that had been injected with NP-EGTA medium in Ca²⁺-containing medium and applied UV light to selected cilia at the anterior of the cell.

UV light was first applied outside the cell body. In these cases, only the cilia in the area of UV application reversed their direction. It is important to note that no trichocyst exocytosis took place (N=22), even though Ca²⁺ was present outside the cell. Typical results are shown in Figs 6A (Movie 2) and 6B (Movie 3). In the example shown in Fig. 6A, UV light was applied to the entire cilium above the basal body. Ciliary reversal took place strongly only in the area of UV application. The angle of ciliary reversal was 1.9 rad. When the distance from the rim of the UV-irradiated area to the cell surface was increased to 6 μm, UV application induced only weak ciliary reversal in the area (Fig. 6B,Movie 3). In this case, the angle of ciliary reversal was 0.9 rad. The relationship between the distance of UV irradiation and the angle of ciliary reversal is summarized in Fig. 6C. Smaller distances led to greater angles of ciliary reversal,although the ciliary angle before UV application was independent of the distance of the irradiated area from the cell surface.

These results are similar to those obtained when ciliary reversal was induced by the photolysis of NP-EGTA in the Ca²⁺-chelating medium. The following experiments were carried out in Ca²⁺-containing medium.

When UV light was applied to the portion of the cilium including both the basal body and the upper ciliary region in the anterior of Parameciumcell, ciliary reversal and trichocyst exocytosis took place in response to UV application in the following order. At first, ciliary reversal took place only in the area of UV application. Then, after a slight delay equivalent to about one video frame (33 ms), trichocyst exocytosis also occurred but only in the immediate area, indicating that [Ca²⁺]_i had reached the threshold level for ciliary reversal. After a small further delay (one video frame) or concomitant with the exocytosis, ciliary reversal at the whole cell surface took place (N=13). Typical results are shown in Fig. 7 and Movie 4. As shown in the image at 0.23 s in Fig. 7A,cilia in the area of UV application reversed their direction, although cilia outside that area did not. After trichocyst exocytosis in the area of UV application (arrowheads in Fig. 7A), cilia outside the application area reversed their direction(at 0.3 and 0.37 s in Fig. 7A). Fig. 7B shows that ciliary reversal in the area of UV application (solid line) took place before trichocyst exocytosis, whereas ciliary reversal outside the application area(broken line) took place after the trichocyst exocytosis (triangle).

When UV light was applied to the inside at the anterior of the Paramecium cell, trichocyst exocytosis took place just outside the area of UV application, indicating that [Ca²⁺]_i had reached the threshold level for the ciliary reversal. Then, ciliary reversal at the entire cell surface occurred after a short delay (one video frame) or concomitant with trichocyst exocytosis (N=16). Typical results are shown in Fig. 8 and Movie 5. As shown in the 0.23 s image in Fig. 8A, cilia within and outside the area of trichocyst exocytosis reversed their direction simultaneously after trichocyst exocytosis(arrowheads in Fig. 8A). Fig. 8B shows that ciliary reversal within (solid line) and outside (broken line) the area of trichocyst exocytosis occurred simultaneously just after trichocyst exocytosis. It should be noted that, even in the area where trichocyst exocytosis took place (in response to sufficient Δ[Ca²⁺]_i at the neighboring basal bodies), ciliary reversal never occurred prior to trichocyst exocytosis(triangle in Fig. 8B).

All of these results are summarized in Fig. 9, as described in the Discussion.

It is widely accepted that ciliary reversal is induced by an increase in intra-ciliary Ca²⁺ levels. Utilizing living Paramecia,Saeki and Hiramoto (1975)directly demonstrated that an injection of Ca²⁺ buffer into the cell body caused ciliary reversal. In this study, we used living Paramecia and have shown that Δ[Ca²⁺]_i,through photolysis of NP-EGTA, directly induces ciliary reversal. Ciliary reversal took place even when performed in a Ca²⁺-chelating medium(Fig. 5), indicating that ciliary reversal by photolysis of NP-EGTA is independent of Ca²⁺influx through the ciliary membrane.

When the experiment was repeated in a Ca²⁺-containing medium,ciliary reversal also took place (Fig. 6) at levels similar to that seen in Ca²⁺-chelating medium. In both cases, the larger the area of UV application in cilia, the greater the maximum angle of ciliary reversal (Figs 5B, 6C). These results strongly suggest that, even when the external medium contains Ca²⁺, theΔ[Ca²⁺]_i induced by the photolysis of NP-EGTA is directly responsible for ciliary reversal.

The relationship between the cellular localization of the photolysis of NP-EGTA (=Δ[Ca²⁺]_i) and the induction of ciliary reversal is summarized in Fig. 9. (i) When Δ[Ca²⁺]_i took place at the cilia but not at the basal bodies, ciliary reversal was restricted to the area of UV application (Figs 5 and 6, cf. Fig. 9A,B). In such cases,ciliary reversal was probably induced directly by theΔ[Ca²⁺]_i in each cilium, because the reversal took place only in cilia subjected to UV application. (ii) WhenΔ[Ca²⁺]_i took place at the basal bodies,trichocyst exocytosis, but not ciliary reversal, was evident in the area of UV application, and after a slight delay ciliary reversal occurred over the entire cell surface (Fig. 8,cf. Fig. 9D). This ciliary reversal over the whole cell surface was probably not induced directly by theΔ[Ca²⁺]_i upon UV application, but through a signal transduction cascade that begins with membrane depolarization due to the trichocyst exocytosis. Trichocyst exocytosis from the anterior of the cell is known to bring about a membrane depolarization(Erxleben and Plattner, 1994). This idea is supported by the following observations: (iii) When UV light was applied to the area containing both the basal bodies and the upper ciliary region, cilia in the area of UV application reversed their direction prior to trichocyst exocytosis, whereas cilia outside the area reversed their direction concomitantly with, or just after, trichocyst exocytosis(Fig. 7, cf. Fig. 9C). This suggests that ciliary reversal in the UV application area was induced directly byΔ[Ca²⁺]_i, while ciliary reversal outside the area was secondarily induced by membrane depolarization accompanying trichocyst exocytosis. Thus it is likely that the basal bodies do not receive the Ca²⁺ signal for ciliary reversal. We note that larger areas of photolysis were associated with greater angles of ciliary reversal (Figs 5B, 6C). There are two possible explanations for these observations. (1) The Ca²⁺ sensitive region for ciliary reversal may not be restricted to the ciliary base, but may lie over the entire cilium above the basal body. (2) Ca²⁺ released from the NP-EGTA in the upper portion of the cilium may diffuse to the base,causing Δ[Ca²⁺]_i around the ciliary base, leading subsequently to ciliary reversal. As shown in Figs 5B and 6C, ciliary reversal occurred even when the distance between the region of the cilium irradiated with UV light and the cell surface (l) was >6 μm. When l=6μm, the concentration of Ca²⁺ at the ciliary base was estimated to be 10^–6.7 mol l^–1, assuming that all NP-EGTA molecules in the target area released their Ca²⁺ upon UV application, and that the released Ca²⁺ distributed itself evenly throughout the whole cilium, including the ciliary base. This Ca²⁺concentration is much lower than that required to induce ciliary reversal(10^–6 mol l^–1; Naitoh and Kaneko, 1972, 1973). Thus, we propose that the Ca²⁺ sensitive region for ciliary reversal exists over the entire cilium (above the basal body).

In the present study, ciliary reversal occurred in response toΔ[Ca²⁺]_i even when the latter took place only at the tip of the cilia. This result coincides well with observations that local iontophoretic application of Ca²⁺ to any site along the length of demembranated macrocilia of B. mitrata elicits oscillatory bending,indicating that Ca²⁺ sensitivity extends the length of the cilia(Tamm and Tamm, 1989), that Ca²⁺ inward current decreases upon removal of cilia and recovers accompanying ciliary regrowth, indicating that voltage-sensitive Ca²⁺ channels are present throughout the cilia of P. caudatum (Machemer and Ogura,1979), and that fluorescence of Calcium Green rises at a similar rate along the ciliary length during ciliary reversal in Mnemiopsis leidyi (Tamm and Terasaki,1994). Our findings are not necessarily consistent, however, with the fact that iontophoretic application of Ca²⁺ to the ciliary base of a permeabilized P. caudatum cell induces ciliary reversal, though the application to the ciliary tip does not(Hamasaki and Naitoh, 1985). Hamasaki and Naitoh (1985) also found that the ciliary reversal is always preceded by tremulous beating,indicating that tremulous beating and ciliary reversal are probably part of a single chain of events. They found that Ca²⁺ sensitivity in producing tremulous beating is roughly equivalent between the base and the tip in the permeabilized cilium. Thus, Ca²⁺ sensitivity in producing tremulous beating, but not in ciliary reversal, in permeabilized cilia is similar to that in ciliary reversal in living Paramecium. Thus, the sensitivity should extend the length of the cilium beyond the basal body.

Blue LEDs were kind gifts of the Nichia Corporation, Tokushima, Japan. We thank Dr M. Kikuyama (Niigata University, Niigata, Japan) for his invaluable suggestions throughout this work and critical reading of the manuscript. We also thank Dr K. Katoh (AIST, Tukuba, Japan) for his invaluable comments on the manuscript and Dr T. Harumoto (Nara Women's University, Nara, Japan) for providing wild-type P. caudatum cells (kyk402). This study was supported by a Grant-in-Aid for Encouragement of Young Scientists from the Ministry of Education, Science and Culture (no. 12780539), a Grant from Saneyoshi Scholarship Foundation and a Grant from Nissan Science Foundation.