Choice of tracks, microtubules and/or actin filaments for chloroplast photo-movement is differentially controlled by phytochrome and a blue light receptor

Intracellular movement of organelles is an indispensable event for many cellular functions. Unlike animals, plants cannot change their location from where they start growing. However, at the subcellular level, the movement of intracellular organelles is very active and plant cells adapt to various environmental stimuli by changing their distribution. For instance, chloroplasts and nuclei relocate in the cell in response to environmental signals such as light and mechanical stimulation (Kagawa and Wada, 1993; Kennard and Cleary, 1997; Nagai, 1993; Sato et al., 1999; Williamson, 1993). In particular, chloroplast photo-movement is one of the responses observed in cells of many species and occurs throughout the plant kingdom (Haupt and Scheuerlein, 1990; Wada et al., 1993; Zurzycki, 1980). Chloroplasts move toward the illuminated area under weak light conditions, while they move away from the area when the light is too strong. The accepted interpretation of the ecological role of these responses is to optimize light harvesting for photosynthesis. Studies on the response are divided into two main areas: the light perception mechanism and the force generation system for movement.

The direction of chloroplast movement, whether toward or away from the light, depends on light quality as well as light intensity. There are two types of photosensory pigment involved in chloroplast relocation. In most plants, the effective wavelength of light for chloroplast translocation is restricted to the blue band region of the spectrum, and the response is controlled by a blue light (BL) receptor (Zurzycki, 1980). In some plant species, such as the fern Adiantum capillus-veneris and the green alga Mougeotia scalaris, red light (RL) as well as BL also induces chloroplast movement. The response to red light is mediated by another photoreceptor, phytochrome (Wada et al., 1993). Recently, we showed for the first time in bryophytes that phytochrome, as well as a BL receptor, also regulates chloroplast relocation in the moss Physcomitrella patens (Kadota et al., 2000).

The involvement of two types of motility system is well known for organelle movement in eukaryotic cells. One is a microtubule (MT)-based system and the other an actin microfilament (MF)-based system. Force-generation in chloroplast movement has been investigated in various plant cells. The mechanism of movement in all cells so far investigated, where chloroplasts show directional movement toward or away from the light, was considered to be an exclusively MF-based and not MT-based system (Izutani et al., 1990; Kadota and Wada, 1992; Tlalka and Gabrys, 1993; Wagner et al., 1972). In contrast, for many years organelle movement and vesicle transport in animal cells was thought to be an exclusively MT-based system. Since Kuznetsov et al. (Kuznetsov et al., 1992) showed that single axoplasmic organelles move along both MTs and MFs tracks in vitro, however, earlier views of the roles of the two cytoskeletal filaments have changed, and it now appears that they functionally cooperate during organelle transport (Goode et al., 2000; Langford, 1995; Rogers and Gelfand, 2000). It seems clear that melanosome movement and axonal transport of mitochondria involves both MTs and MFs in vivo, although they had previously been thought to be a model system of MT-based movement (Bridgman, 1999; Morris and Hollenbeck, 1995; Rodionov et al., 1998; Rogers and Gelfand, 1998). It is possible that the coordinated action of these filaments might be a general principle of organelle transport (Rodionov et al., 1998).

Investigating whether these two cytoskeletal components functionally cooperate in plant cells is therefore important for understanding the roles of these filaments in general. Although chloroplasts are closely associated with both MTs (Wacker et al., 1988) and MFs (Quader and Schnepf, 1989) in the moss Funaria hygrometrica, the involvement of MTs in chloroplast movement has remained obscure. The present study focuses on the contribution of these two cytoskeletal elements to chloroplast movement as a part of force generating system in the protonemal cell of P. patens. To achieve this, chloroplast photo-movement was induced by microbeam irradiation and the effect of cytoskeletal drugs on the behavior of individual chloroplasts was observed directly and examined in detail by quantitative analysis using a computer-recording system. Here we report that chloroplasts potentially move along both MTs and MFs, but that the choice of the motile system for photorelocation is differentially controlled by phytochrome and a BL receptor.

Protonemata of P. patens were grown under white light (4.5 W m⁻²) for 2-4 weeks at 25°C on BCD solid medium containing 1 mM CaCl₂, TES elements, 5 mM ammonium tartrate and 0.5%(w/v) agar overlaid with cellophane (Ashton and Cove, 1977). Protonemal cells were inoculated between two layers of agar-gelatin film on a coverslip. The film was made from 0.5% (w/v) Bacto-agar (Difco Laboratories, Detroit, MI, USA) and 0.05% (w/v) gelatin (Koso Chemical Co. Ltd., Tokyo, Japan). They were cultured under continuous RL of about 0.5 W m⁻² for 1-2 weeks in the liquid BCD medium. The basal cells of RL-grown protonemata, which were subsequently incubated for 1 day in the dark, were used for the present study. All the culture and experimental procedures were conducted at 25±1°C unless otherwise stated. All the experiments shown here were repeated at least 3 times on different occasions and the same results were obtained.

Fluorescence lamps (FL40SD or FL10D, Toshiba Lighting and Technology Corp., Tokyo, Japan) were used as the source of white light. RL was obtained from the lamp through a red plastic filter (Shinkolite A, #102, Mitsubishi Rayon Co., Ltd., Tokyo, Japan).

Microbeam irradiation was performed on a custom-made microbeam irradiator as previously described (Kadota et al., 2000). Monochromatic RL and BL were provided through interference filters (Vacuum Optics Co. of Japan, Tokyo), which have their peak at 663.2 and 452.5 nm and half-band widths of 32 and 7.5 nm, respectively. Neutral density filters (either Inconel-coated quartz glass from Fujitoku Corp., Tokyo, or an ND filter from Hoya Corp., Akishima, Japan) were used to attenuate the fluence rate.

Cytochalasin B (Sigma Chemical Co., St Louis, MO, USA) was employed as an MF-depolymerizing agent. Cremart (Sumitomo Chemical Co., Osaka, Japan) was used to disrupt MTs. It was previously reported that Cremart caused complete disassembly of microtubules in the same moss protonemata, leaving microfilaments undisturbed (Doonan et al., 1986; Doonan et al., 1988). Cytochalasin B and Cremart were dissolved in DMSO as stock solutions of 20 mM and 2 mM, respectively. Final concentrations of 0.1 mM cytochalasin B and 10 or 20 μM Cremart were used in the culture medium. To help drugs gain access into the cells, a 0.2% (w/v) solution of the detergent Pluronic F-127 (Sigma) was added to the culture medium. Cells were incubated with each drug solution for at least 1 hour before continuous microbeam irradiation.

The velocities and efficiencies of movements were obtained from chloroplasts that had moved more than 10 μm toward or away from the irradiated site after microbeam irradiation.

For staining MTs, cells were fixed for 1 hour with 8% paraformaldehyde (Wako, Pure Chemical Industries, Ltd., Osaka, Japan) in piperazine-N,N′-bis (2-ethane-sulfonic acid) (Pipes) buffer (20 mM Pipes, 5 mM EGTA, 5 mM MgCl², 1% DMSO and 0.5 mM PMSF, pH 7.0). After rinsing with Pipes buffer three times, cells were treated with cell wall digestion enzyme solution containing 2% Driselase (Kyowa Hakko Kogyo, Co., Tokyo, Japan), 1% Pectryase (Seishin Pharmaceutical Co., Tokyo) and 50 μg/ml leupeptin (Sigma) in Pipes buffer for 30 minutes. Cells were subsequently immersed with cold absolute methanol for 10 minutes to stop the enzyme reaction. After rinsing with Pipes buffer again, they were treated with 0.05% Triton X-100 in Pipes buffer for 30 minutes. They were rinsed with PBS (137 mM NaCl, 2.7 mM KCl, 1.5 mM KH²PO⁴, 8.0 mM NaHPO⁴, pH 7.2) and then incubated overnight with primary antibodies, a mixture (1:1, diluted 1/500) of mouse monoclonal anti-chicken α- and β-tubulin antibodies (Amasham Japan Co., Tokyo) in PBS. Cells were rinsed three times with PBS containing 0.1% (w/v) BSA and subsequently incubated overnight with secondary antibody, FITC-conjugated anti-mouse Ig antibody from sheep (Sigma) diluted 1/10 in PBS. Specimens were then mounted on a glass slide in PBS. For labeling MFs, cells were pretreated for 30 minutes with 300 μM m-maleimidobenzoic acid N-hydroxysuccinimide ester (Sigma) and 1.5% DMSO in Pipes buffer to stabilize MFs and subsequently fixed with 2% paraformaldehyde in the same buffer for 10 minutes. We used mouse C4 anti-actin monoclonal antibody (diluted 1/500) (ICN Biomedicals, Inc., OH, USA), and other procedures were the same as for MTs.

We used a Zeiss Axiovert 135 microscope coupled to a confocal laser scanning system (LSM 410, Zeiss) to observe fluorescence images of cytoskeletons and chlorophyll. The beam splitter used was FT 488/543. The excitation wavelength was 488 nm and a barrier, BP 515-525, and an additional beam splitter, FT560, were used for FITC observation. Chlorophyll fluorescence was obtained by using OG 665 as a barrier filter after excitation with a 543 nm laser line. The pinhole size was 20. Serial optical sections were obtained every 0.75 μm in the Z-axis. Images were acquired through a ×63 Zeiss Plan-Apochromat objective (NA 1.4) at a maximum resolution of 512×512 pixels. Every optical section was gained after averaging four scans.

Moss protonemata have a simple structure composed of linearly arranged cells. We can readily observe individual chloroplasts inside the cell without any mechanical injuries. The responses of chloroplasts to a local gradient of light signal in a cell were investigated by partial irradiation of a cell with monochromatic RL or BL microbeam. RL-grown protonemata were kept in darkness for 1 day before light irradiation. This pretreatment improved the sensitivity of chloroplast photo-movement compared with that obtained previously without pretreatment (Kadota et al., 2000). Clear photo-accumulation and photo-avoidance responses of chloroplasts were induced within 2 hours by both light qualities, as shown in Fig. 1. At low fluence rates, chloroplasts accumulated mostly to the irradiated site (Fig. 1A,B). On the other hand, at high fluence rates, chloroplasts inside the beam moved away from the illuminated area, and the chloroplasts outside the beam moved toward, but did not enter, the area, resulting in ‘double bands of chloroplasts’ separated with the beam (Fig. 1C,D).

To determine the transition point in fluence rate from accumulation response to avoidance response, we used microbeam irradiation at various fluence rates of RL or BL. Fluence rate dependency of chloroplast relocation induced by RL or BL microbeam irradiation is shown in Table 1. RL-induced chloroplast relocation changed from accumulation to avoidance between the fluence rates of 3 and 10 W m⁻². On the other hand, in the case of BL, accumulation movement was induced at a fluence rate of between 0.01 and 30 W m⁻², and very strong intensity, 100 W m⁻² or more, was required to induce avoidance movement.

To study the involvement of the two filament systems on chloroplast movement, we used Cremart and cytochalasin B to disrupt MTs and MFs, respectively. MTs and MFs were visualized by indirect immunofluorescence staining and observed by fluorescence microscopy (Fig. 2). In untreated cells, MTs were aligned parallel to the cell axis, being associated with chloroplasts, and MFs also showed pronounced association with chloroplasts. Filamentous structures of MTs and MFs were completely disrupted by treatment for 1 hour with 10 μM Cremart or with 0.1 mM cytochalasin B, respectively, although some fluorescent dots were seen around the chloroplasts in both cases. We confirmed that MTs in the cytochalasin-treated cells and MFs in the Cremart-treated cells were not affected (data not shown). All the light treatments used for stimulation described below were done after drug treatment for 1 hour or more.

In the dark, chloroplasts were randomly distributed over the cell periphery, and their long axes were aligned parallel to the longitudinal axis of the cell (Figs 1, 3A-C). This longitudinal orientation of chloroplasts was distinctly disrupted by 2 hours treatment with 10 μM Cremart (Fig. 3A). Using video microscopy, chloroplast motility along the long axis of the cell was analyzed by tracking each chloroplast at 1 minute intervals (Fig. 3D-F). Chloroplasts showed movement in control cells, Cremart-treated cells and cytochalasin B-treated cells, but the motility behavior was strikingly different (Fig. 3). We found that individual chloroplasts in control cells moved in a particular fashion such that movement for a short distance in one direction along the cell axis was temporally followed by movement in the opposite direction (Fig. 3G). Chloroplasts repeated this ‘back and forth’ movement along the longitudinal axis with various amplitudes and frequencies. Typical cytoplasmic streaming, the continuous flow of small vesicles that is seen in many plant cells, could not be observed. Cytochalasin B failed to disturb the longitudinal orientation and characteristic movement of chloroplasts (Fig. 3B,E,I). On the other hand, when the cells were treated with Cremart, the ‘back and forth’ movement ceased immediately (Fig. 3D). Movement of chloroplasts was, however, still apparent at a reduced rate in any direction, causing the orientation of chloroplasts parallel to the cell axis to be disturbed (Fig. 3A,H). These results suggest that the longitudinal orientation and the ‘back and forth’ movement of chloroplasts were dependent upon an MT-based system, but the residual movement of chloroplasts in Cremart-treated cells indicates that a motor system other than an MT-based system must also be involved. Because chloroplast motility was completely blocked when the cell was treated simultaneously with both Cremart and cytochalasin B (Fig. 3C,F,J), it is suggested that the MF-based system is responsible for the irregular movement of chloroplasts under Cremart treatment. No apparent change in the intracellular distribution of chloroplasts was observed by any drug treatments during the entire observation period (6 hours). The inhibitory effects of Cremart and/or cytochalasin B were reversible, and chloroplasts resumed normal motility after washing out the drugs. We also tested other cytoskeletal drugs, Oryzalin (10 μM) for MTs, and cytochalasin D (0.1 mM) and Latrunculin B (10 μM) for MFs, with the same results as described above (data not shown). Chloroplasts can therefore move along both MTs and MFs in protonemata of P. patens.

The subsequent experiments were performed to verify the reversibility of the effects of cytoskeletal drugs on chloroplast photorelocation before motile analysis. In cells treated simultaneously with both Cremart and cytochalasin B, the chloroplasts remained completely stationary despite continuous irradiation with the BL microbeam for 2 hours. When the same samples were transferred into inhibitor-free medium, however, the chloroplasts began to move to the irradiated site within 30 minutes and normal photorelocation was visible within 2 hours after transfer (Fig. 4A). This result suggests that the drug effects were completely reversible with respect to chloroplast photo-movement. Furthermore, we ascertained that photoperception by BL receptor and by phytochrome was not affected by the drugs (Fig. 4B,C). Cells in which chloroplasts were prevented from moving toward the irradiation site by simultaneous treatments of Cremart and cytochalasin B were washed out with culture medium in the dark. Under both light qualities, chloroplasts started to move toward the pre-irradiated locus in the subsequent darkness in the absence of microbeam irradiation, indicating that the spatial information of the light signal established by the microbeam was retained at the pre-irradiated site of the cells, even though the cytoskeletons were disrupted enough to stop organelle motility. These results strongly suggest that the drugs block the motile systems, but not the photoperception processes, in the signal transduction of chloroplast photorelocation. The response reached the maximum level within 1 hour of the removal of drugs and gradually dispersed with ‘back and forth’ movement thereafter.

To determine the contribution of the two cytoskeletal tracks to chloroplast accumulation movement mediated by phytochrome and the BL receptor, we examined the effects of cytoskeletal inhibitors on the response. The fluence rate used was set to one tenth of the fluence rate required for avoidance response, that is, 1 W m⁻² for RL and 10 W m⁻² for BL (Table 1). When part of a cell was irradiated in the culture medium containing 1% DMSO, no effect of DMSO could be observed and the chloroplasts moved to the beam spot within 2 hours of the onset of microbeam irradiation (Fig. 5A,E). In the case of RL, chloroplasts moved directionally to the irradiated site, still showing MT-based ‘back and forth’ movement (Fig. 5A,B), while BL induced movement was relatively straight and chloroplasts moved toward the light spot with little backward movement (Fig. 5E,F). RL induced-relocation of chloroplasts was completely inhibited by 10 μM Cremart, but not at all by 0.1 mM cytochalasin B (Fig. 5B,C). These results suggest that phytochrome-mediated chloroplast relocation depends on the MT-based system. In the case of BL, on the other hand, neither Cremart nor cytochalasin B inhibited the accumulation of chloroplasts (Fig. 5F,G). Only simultaneous treatment with both Cremart and cytochalasin B inhibited the BL-induced accumulation response (Fig. 5H). It is notable that chloroplasts in Cremart-treated cells moved smoothly towards the BL spot with less ‘back and forth’ movement. Velocities of chloroplast movement in Cremart-treated cells were lower than those in control and cytochalasin B-treated cells (Fig. 5E-G). Exactly the same drug effects were obtained when the accumulation response was examined at 0.1 W m⁻² of BL with simultaneous background irradiation of 10 W m⁻² far-red light, the condition in which BL-activation of phytochrome is suppressed (data not shown), which suggests that chloroplasts can use both tracks, MTs and MFs, in the accumulation movement mediated by the BL receptor.

Motile systems for the avoidance response of chloroplasts mediated by phytochrome were also investigated. The fluence rate of RL used was set at 30 W m⁻², which increased the chloroplast-free area compared with that at 10 W m⁻² (Figs 1D, 6A). RL-induced avoidance responses were completely blocked by Cremart but not by cytochalasin B (Fig. 6B,C). These results suggest that the motile system of the phytochrome-mediated avoidance response depends on MTs, as is the case for the accumulation response. On the contrary, the avoidance response induced by BL at 100 W m⁻² was inhibited only by simultaneous treatment with both Cremart and cytochalasin B, but not with Cremart or cytochalasin B alone (Fig. 6E-H). These results indicate that the motile system of the avoidance response mediated by BL receptor is the same as that of the accumulation response, i.e. either MTs or MFs are sufficient for chloroplasts to move to the appropriate site, and this is different from the phytochrome response.

To further clarify the difference in behavior of chloroplasts between RL and BL induced-responses, the movement of each chloroplast was analyzed quantitatively. Efficiencies of chloroplast movement toward or away from the light spot were calculated as the ratio of net/total distance in movement in each of the above experiments, including those using drugs where the photo-movement was not suppressed (Table 2). The ratios were high in BL responses and low in RL responses, irrespective of whether the movements were accumulation or avoidance and of the presence or absence of drugs. This means that chloroplasts took the shortest route in the BL responses, while in phytochrome responses they moved toward or away from the spot with ‘back and forth’ movement. Furthermore, quantitative analysis of chloroplast velocities during photo-movement indicated that the kinetics for RL and BL responses were different (Fig. 7). The velocities of chloroplasts in RL accumulation responses were not changed in the presence of cytochalasin B (Fig. 7A,B), indicating that there was no contribution of MFs to the phytochrome response. In the BL accumulation response, in contrast, the velocities became higher in the cytochalasin B-treated cells and lower in Cremart-treated cells compared with those in control cells (Fig. 7C-E). It is noteworthy that the behavior of chloroplasts in cytochalasin B-treated cells, such as velocities and the ratio of net/total distance in movement, substantially differed between RL and BL responses (Table 2 and Fig. 7). This suggests that MT-based systems would be differentially controlled by phytochrome and the BL receptor. Similar results were obtained in the avoidance responses (Fig. 7F-J). No apparent difference in chloroplast behavior was evident between accumulation and avoidance in phytochrome responses or BL responses, except the movement direction with respect to light. This indicates that the choice of the motile system depends on light quality, but not on the light intensity required to determine the direction of movement.

To determine the motile system responsible for chloroplast movement, we firstly analyzed the dependency on fluence rate of the two types of chloroplast movement (accumulation and avoidance movement) induced by RL and BL, because they cause chloroplasts to move in opposite directions. From the photobiological perspective, it should be noted that the avoidance response occurred at 10 W m⁻² of RL, while 100 W m⁻² of BL was required for this response. Even RL as high as 100 W m⁻² failed to induce the avoidance response in Mougeotia (Weisenseel, 1968). So far, a phytochrome mediated-avoidance response has only been demonstrated in the fern A. capillus-veneris (Yatsuhashi et al., 1985); however, the fluence rate required for the response in fern cells (470 W m⁻²) is much higher than that in the moss cells used in the present study (10 W m⁻²). The BL-induced avoidance response is widespread throughout the plant kingdom, being found in such plants as the moss F. hygrometrica (Zurzycki, 1967), the fern A. capillus-veneris (Yatsuhashi and Wada, 1990), the monocotyledon Lemna trisulca (Zurzycki et al., 1983), and the dicotyledon Arabidopsis thaliana (Kagawa and Wada, 2000; Trojan and Gabrys, 1996). The fluence rate for the avoidance response induced by BL is within the range 10-30 W m⁻², which is much lower than that necessary in our studies. The high sensitivity to RL could be interpreted as an adaptation of cells to the preculture conditions for continuous RL because it has been found that the RL-grown cells lose the RL-response when transferred to white light conditions (Kadota et al., 2000). It is unlikely that the low sensitivity to BL in the avoidance response is the effect of preculture light conditions, because the fluence rate required for the avoidance response in white light-grown cells is the same as that in RL-grown cells (data not shown). Rather, the difference in sensitivity could be ascribed to the temperature conditions. Weisenseel (Weisenseel, 1968) reported in Mougeotia that the sensitivity of BL induced-avoidance response is high at low temperatures and low at high temperatures.

There has been a recent report that chloroplasts associate with MFs, but not with MTs, in leaf mesophyll cells of A. thaliana (Kandasamy and Meagher, 1999). In the moss used in the present study, however, we have demonstrated that chloroplasts have a structural relationship with both MTs and MFs. Our present results confirm the previous observation in F. hygrometrica (Quader and Schnepf, 1989; Wacker et al., 1988). The involvement of MTs in maintaining the longitudinal orientation of chloroplasts was clearly demonstrated. Continuous observation of chloroplast movement revealed for the first time that chloroplasts show characteristic ‘back and forth’ movement over short distances, with a range of several μm, which is reminiscent of pigment movement in melanophores in the dispersed state. The ‘back and forth’ movement may be an alternative to cytoplasmic streaming, which is not seen in moss cells. This characteristic movement of chloroplasts was found to be dependent on an MT-based system. Furthermore, treatment with Cremart alone to effectively eliminate microtubules still left small but significant chloroplast motility in a random direction. This residual motility was completely lost when the cell was treated simultaneously with both Cremart and cytochalasin B, suggesting that an actin-based system is responsible for the residual movement of chloroplasts in Cremart-treated cells. We therefore conclude that chloroplasts move along MFs as well as MTs, but with strikingly different patterns. Our results represent the first direct demonstration of organelle movement along both filaments in plants. The effects of cytoskeletal inhibitors on the behavior of chloroplasts were very similar to those on organelles in the axon (Morris and Hollenbeck, 1995). The general rule in animal cells that MTs are used as the long-range tracks, and MFs as the short-range tracks (Allan and Schroer, 1999; Langford, 1995), seems to be applicable to the chloroplast motility of moss cells in the dark.

By time-lapse observation of chloroplasts, we have clearly shown that chloroplast photo-movement in P. patens is a directional movement toward or away from the illuminated area. This behavior is different from the trap movement seen in several algae, in which chloroplasts move together with cytoplasmic streaming and slow down if they occasionally come by the irradiated site (Haupt and Scheuerlein, 1990; Wagner and Grolig, 1992). In the latter case, no directional movement of chloroplasts with respect to the light spot is found. We have also demonstrated clear effects of cytoskeletal drugs on photo-movement, leading to interesting clues about the motile systems used. RL-induced relocation was inhibited by anti-MT drugs, regardless of the type of movement, namely the accumulation or avoidance responses. This suggests that the phytochrome signal transduction pathway finally regulates only the MT-based system, but not the actin-based system. Involvement of MTs in the directional photo-movement of chloroplasts has never before been reported in literature. In contrast, treatment with neither an anti-MT drug alone nor an anti-MF drug alone could suppress the BL-induced movement, whereas simultaneous treatment with both drugs effectively inhibited it. Thus, MFs in addition to MTs are available for chloroplasts to move along as both accumulation and avoidance responses mediated by the BL receptor. Again, this type of motile system for photo-movement has never been reported in literature.

A quantitative analysis of chloroplast behavior also clearly revealed further differences in the motile system between the RL and BL responses. The effects of cytoskeletal inhibitors on the efficiency and the velocity of chloroplast movement were considerably different for RL and BL responses. Their differences in cytochalasin B-treated cells, especially, indicate that different types of MT-based systems, or at least a different use of the same system, is likely between RL and BL responses. Thus, the motile systems of chloroplasts for phytochrome- and BL receptor-mediated responses are presumably quite different, although the final cell images showing chloroplast relocation look similar. It is likely that the differences are caused by the light quality rather than by its intensity, because similar results were obtained in both accumulation and avoidance responses. Furthermore, there is no effect of light intensity on the velocity of chloroplasts in accumulation photo-movement in A. capillus-veneris (Kagawa and Wada, 1996).

Chloroplasts moved at high velocities in cytochalasin B-treated cells and low velocities in Cremart-treated cells in BL responses, indicating that the ‘dual transport’ system of chloroplasts was stimulated by BL. Notably, the velocities of chloroplasts in Cremart-treated cells in P. patens are very close to those seen in non-treated cells of A. capillus-veneris, which are known to be dependent on an MF-based system (Kadota and Wada, 1992; Kagawa and Wada, 1996). Therefore, it can be assumed that MTs are required for the rapid movement of chloroplasts, but that chloroplasts are also able to move along MFs at lower rate using myosin motors. Unlike the case of chloroplast motility in the dark, MFs as well as MTs seem to provide the tracks for long-range movement in the BL-induced movement of chloroplasts.

Chloroplast photorelocation movement of P. patens analyzed here has several novel features, namely (1) chloroplasts can utilize both MTs and MFs as traveling-tracks and (2) the motile systems are differently regulated by two different types of photoreceptor. There are three kinds of regulatory pathway involved in photorelocation of chloroplasts, as schematically shown in Fig. 8. The pathway of ‘BL receptor-MFs’ for chloroplast photo-movement seems ‘conventional’ because it is widely seen in higher plants. In addition to this, ‘phytochrome-MTs’ and ‘BL receptor-MTs’ pathways are suggested to operate in P. patens. MT-based chloroplast movement is known to occur in algal cells, although chloroplasts do not show any directional movement with respect to the light-irradiated site (Maekawa et al., 1986; Mizukami and Wada, 1981). Chloroplast movement in the moss may represent an evolutionary intermediate between an MT-dominated motile system in algal cells and an entirely MF-based system in higher plant cells. It is very interesting to see a parallel evolutionary change in the use of motile systems involved in pigment movement of melanophores in fish cells and mammalian cells (Rogers and Gelfand, 1998).

The experimental system reported here will provide an excellent model for studying ‘multi-transport’ systems of organelle movement, because the motile system used can be readily manipulated by light quality and by cytoskeletal drugs, and the direction can be also easily regulated by light intensity. At the moment, details of the molecular mechanism of chloroplast movement, such as the kinds of motor proteins involved, are unknown. In P. patens, however, gene targeting by homologous recombination is as efficient as in Saccharomyces cerevisiae (Puchta, 1998; Reski, 1998; Reski, 1999) and tagged mutagenesis has recently been developed by shuttle mutagenesis (Nishiyama et al., 2000). Incorporation of these molecular genetics techniques into the present system could reveal the molecular components of chloroplast movement and the roles of MTs and MFs in photo-movement. There are many analogous features between P. patens and higher plants in their physiological aspects (Reski, 1998). Nucleotide sequences of genes in P. patens are highly homologous to those in higher plants and high similarity of codon usage has been reported (Machuka et al., 1999; Reski et al., 1998). Thus, further analysis of chloroplast photo-movement in P. patens could lead to an understanding of the same phenomenon in higher plants as well as in lower plants.

We thank Dr M. Hasebe (NIBB, Okazaki, Japan) for the gift of Physcomitrella patens. We are grateful to Dr Jane Silverthorn (University of California, Santa Cruz, USA) for critical reading of the manuscript. This work was carried out under the NIBB Cooperative Research Program (00-133) and was partly supported by Grant-in-Aid for Scientific Research (C) (Grant No. 11640651) to A.K. from the Japan Society for the Promotion of Science, Grant-in-Aid for Scientific Research (B) (Grant No. 09440270) and PROBRAIN (Program for Promotion of Basic Research Activities for Innovative Biosciences) to M.W. and also by a grant from Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists (Grant No.12740202) and the Sasagawa Scientific Research Grant from the Japan Science Society to Y.S.