Paramecium bursaria contain several hundred cells of the green algae Chlorella as endosymbionts and are designated green. Chlorella-free white cells can be obtained from natural green cells by rapid growth in constant darkness (DD). Chlorella were isolated easily from their host cells and re-infected. The infection of Chlorella was restrained by a photosynthesis inhibitor (DCMU). This result can be related with the fact that symbiotic Chlorella release their photosynthetic products. Furthermore, when green cells were cultured in DD, the number of endosymbiotic Chlorella decreased and the density of host cells increased. On the other hand, the mating reactivity rhythm of green cells disappeared in DD. The photosynthetic products of symbiotic Chlorella, maltose and oxygen, induced the rhythms of their host cells in DD, but they could not shift the phase of the rhythms. Moreover, arrhythmic mutant white cells reverted to rhythmicity after infection with Chlorella isolated from wild type green cells. Thus the photosynthetic products released by endosymbiotic Chlorella have important roles in the establishing of the endosymbiosis and the expression of circadian rhythms in P. bursaria cells.
INTRODUCTION
Cells of the unicellular ciliate, Paramecium bursaria, contain several hundred green algae (Zoochlorella) established in the cytoplasm as endosymbionts (Loefer, 1936). Chlorella-free white cells can be obtained easily from natural green cells by rapid growth in constant darkness (DD). White cells also can be restored quickly to green ones by the infection of Chlorella isolated from green cells. The re-infection process starts with the induction of Chlorella to digestive vacuoles (DV) in the cytostom. Then Chlorella individually enter the perialgal vacuoles (PV). Since PV can not combine with the lysosome, Chlorella in the PV are not digested. When Chlorella stayed in DV, they are digested in due course (Meier and Wiessner, 1989). P. bursaria is a useful material for the study of the interaction between symbionts and the host cell. Cells of symbiotic Chlorella release their photosynthetic products, 90% of which is maltose and oxygen, and the host cells use them as living energy. Free-living Chlorella did not release their photosynthetic product of sugar, therefore the maltose release could be one of the essential factors in establishing symbionts (Weis, 1979, 1980).
On the other hand, the sexual interaction of Paramecium called mating reaction occurs upon mixing cells of complementary mating types when they are in the stationary phase and are sexually mature (Sonneborn, 1957; Hiwatashi, 1981). Furthermore, cells of P. bursaria exhibit high mating reactivity in the light phase and low in the dark phase of the light and dark cycle (LD 12:12 hr). After they are transferred to constant light (LL, 1,000 lux), they continue to show a clear circadian rhythm of mating reactivity for several days (Jennings, 1939; Miwa et al., 1987). Circadian rhythms are shown in all levels of organisms including both eukaryotes and prokaryotes as basic behavior to adapt to global environments (Kondo et al., 1993). Circadian rhythms are controlled by one or more internal oscillators often called a “Circadian clock”. In spite of recent molecular genetic studies concerning circadian clocks (Aronson et al., 1994; Kondo et al., 1994; Sehgal, 1995), their mechanisms have not yet been made clear.
In this paper, we will report first the significance of photosynthetic products when Chlorella infect into Paramecium cells. We also show the disappearance of mating reactivity of green cells in DD is due to the digestion of the Chlorella. Furthermore, arrhythmic mutant cells are restored the mating reactivity rhythm by the infection of Chlorella isolated from wild type green cells.
MATERIALS AND METHODS
Strains and culture
Three strains of Paramecium bursaria, syngen 1 were used in this experiment: T316 (mating type IV, collected at Tsukuba in Ibaraki), Sj2 (mating type I, collected in Shimane), MC1w (mating type I, arrhythmic mutant strain). Strains T316 and Sj2 contain Chlorella symbionts and are thus designated green. Strains T316w and Sj2w are Chlorella-free white cells that have been induced from T316 and Sj2, respectively, by rapid growth in DD. Strain MC1w (white cells) was induced from Sj2w by treatment with MNNG (N-methyl-N′-nitroN-nitrosoguanidine). Strain MCwT (green cells) was derived from MClw by infection of Chlorella isolated from T316. All strains were cultured in a fresh lettuce juice medium, which had been inoculated with Klebsiella pneumoniae one day before use (Hiwatashi, 1968). Cultures were kept at 25°C under the light/dark cycle (LD 12:12 hr, 1,000 lux of cool-white fluorescent light) and transferred to constant light (LL) or dark (DD) condition in each experiment.
Test of mating reactivity
Mating reactivity of green cells was tested by mixing them with white “tester” cells of a complementary mating type. White cells were tested with green “tester” cells. Mating reactivity in cell populations was measured every 3 hr as follows: 10 cells were placed in each of 6 different wells of a depression glass plate, and about 100 highly reactive tester cells were added to each well. After 5 min, the percentage of mating reactive cells clumping with tester cells was counted.
To prepare tester cells with high reactivity, four groups of green and white cells were entrained to four light/dark cycles (LD 12:12 hr, staggered by 6 hr). Each group of tester cells was used twice in the mating reactivity test of consecutive 3-hr testing intervals. Since each group of tester cells showed high reactivity at least during 6 hr in a day, tester cells were always highly reactive in every test.
Treatment with maltose and oxygen
Mating reactivity rhythms of cells treated with maltose or oxygen were measured in DD. Cells entrained to a LD cycle (12:12 hr) were transferred to DD and added 10−2 M maltose or ventilated 10 ml oxygen. In this experiment, maltose or oxygen were added to the green cell cultures every 3 hr during 12 hr in three different schedules staggered by 6 hr as follows: A (12–24, 36–48, 60–72 hr); B (18–30, 42–54, 66–78 hr); C (24–36, 48–60 hr).
Re-infection of Chlorella
Chlorella were prepared for re-infection into the white cells as follows: Green cells in a 100 ml culture were concentrated by a hand centrifuge and then broken with an ultra-sonicator for 10 sec. Chlorella were collected through a 15 μm nylon mesh and suspended in K-DS solution (0.6 mM KH2P04, 1.4 mM Na2HP04, 2 mM Na3C6H507, 1.5 mM CaCl2, pH 7.0). About 4 × 107 Chlorella cells/ml were mixed with about 2 × 103 white Paramecium cells. The number of Chlorella infected to the white cells was counted on slide glass under microscope. The green cells in this experiment were treated with 10−6 M photosynthesis inhibitor DCMU (Dichlorophenyl-Dimethylurea) for 6 hr.
RESULTS
Re-infection of Chlorella
Chlorella isolated from green cells can be infected into white cells and established as endosymbionts easily. To know the effects of photosynthetic products on the process of re-infection, Chlorella were isolated from green cells treated with DCMU. DCMU was added to green cell culture at the beginning of light phase in a LD cycle (12:12 hr). Chlorella were isolated from those green cells 6 hr after treatment of DCMU. Half of them were added to white cell suspension. The other half of isolated Chlorella were added to white cell suspension with maltose. Both of these cell suspensions were kept in K-DS solution containing DCMU in LL. As a control, Chlorella were isolated from untreated green cells and they were added to white cell suspension without DCMU. Five white cells were isolated randomly every time and the number of Chlorella in the cells was counted. As seen in Fig.1, the control Chlorella increased rapidly until 6 hr after infection and then they became stable. On the other hand, Chlorella treated with DCMU only were hard to be infected to white cells. They were less than half of the control even at the maximum values, and then they decreased gradually. However, Chlorella increased to control level in presence of maltose 12 hr after addition of Chlorella. After that, they began to decrease and they could not be stable.
Fig. 1.
The effects of DCMU and maltose on the re-infection of Chlorella. Chlorella were isolated from green cells (T316) that were treated with 10−6 M DCMU for 6 hr, and they were added to white cell (T316w) suspension containing DCMU in absence (•) and presence (▪) of maltose in LL. Open square (□) shows untreated Chlorella as a control. Abscissa indicates the time after adding of Chlorella to white cell suspension. The bar at each point indicates SD of results from 5 times measurements.
Chlorella were digested by the host cells in DD
To know the necessity of photosynthesis of Chlorella for maintain the symbiont in the host cell, the transition of the number of Chlorella in green cell was examined following the time in DD. Chlorella began to decrease after 6 hr in DD and 70% of Chlorella were lost after 60 hr (Fig. 2). Moreover, the cell densities of green and white Paramecium cultures were measured in DD. As seen in Fig. 2, green cells increased in inverse proportion to the decrease in symbiotic Chlorella. White cells showed almost the same density for 3 days. All cells were not given culture medium during this experiment. Since the decrease ratio of number of Chlorella was larger than the increase ratio of number of green cells, it was considered that some symbiotic Chlorella were digested by host cells, and green cells divided in DD.
Fig. 2.
The transition of densities of symbiotic Chlorella and host cells in DD. The number of Chlorella (•) in a green cell (T316) was counted each time in DD. The densities of green cells (▪) and white cells (□) in experimental cultures were measured every 6 hr in DD as follows: 1 ml of cell suspension was putted in a depression glass plate from each cultures. They were added 2 mM NiCl2 to stop the swimming, and the number of cells was counted. The bar at each point indicates SD.
Green cells did not show a mating reactivity rhythm in DD
Green cells of P. bursaria show mating reactivity in the light period, but not in the dark period, when exposed to a light/dark cycle (LD 12:12 hr). After they were transferred to constant-light (LL) conditions, they continued to show a circadian rhythm of mating reactivity. What is the mating reactivity rhythm of green cells in DD, in which green cells divided as shown in Fig. 2? Mating reactivity rhythms of green cells and white cells in two natural strains (T316 and Sj2) were measured in LD, LL and DD. Results are shown in Fig. 3 and Fig. 4. Mating reactivity of green cells and white cells in both strains were entrained to a LD 12:12 hr cycle and their rhythms persisted in LL. But mating reactivity of green cells appeared different to the ones of white cells in DD. In strain T316, although white cells (T316w) showed the same rhythm in DD as in LL, green cells hardly exhibited any mating reactivity (Fig. 3). In strain Sj2, white cells (Sj2w) kept a high mating reactivity of more than 80% in DD, whereas green cells showed low mating reactivity in DD, about 20% (Fig. 4). These results indicate that symbiotic Chlorella inhibit the expression of the mating reactivity of Paramecium cells in DD.
The effects of maltose and oxygen on the expression of mating reactivity rhythm
Symbiotic Chlorella are doing photosynthesis and their products, maltose and oxygen, are released inside the Paramecium cell (Weis, 1979). It is expected that the lack of these products in DD cause to the disappearance or decline of the mating reactivity of green ceils as shown in Fig. 3 and Fig. 4. Then the effects of maltose and oxygen on the expression of mating reactivity were examined in DD. In this experiment, maltose or oxygen were added to green cell cultures in DD every 3 hr in three different schedules as shown in Methods. In each case, green cells showed mating reactivity in the subjective day phase in spite of the treatments of maltose and oxygen at different phases (Fig. 5). Thus the maltose and oxygen forced these cells to express the mating reactivity rhythms in DD, but not to shift the phases of these rhythms.
Fig. 5.
Mating reactivity rhythms of green ceils (T316) treated with maltose or oxygen in DD. Cells entrained to a LD cycle (12:12 hr) were transferred to DD and added 10−2 M maltose or ventilated 10 ml oxygen every 3 hr during 12 hr in three different schedules (A-C) staggered by 6 hr as follows: A (12-24, 36-48, 60-72 hr) (•); B (18-30, 42-54, 66-78 hr) (▴); C (24-36, 48-60 hr) (▪). Open circle (○) shows a control of untreated cells. Abscissa indicates the time after transference to DD. Lower bar indicates subjective day and night phases. Mating reactivity appeared in subjective day phase in spite of the treatment of maltose and oxygen at different phases.
Chlorella rescued the mating reactivity rhythm in arrhythmic mutant
One arrhythmic mutant strain MC1w was induced from Sj2w by treatment with nitrosoguanidine (Miwa et al., 1996a). As shown in Fig. 4, Sj2w appeared arrhythmicity in DD only. MC1w expressed high mating reactivity continuously not only in DD but also in LD and LL. These white cells (MC1w) were infected with Chlorella isolated from T316 and the mating reactivity of the green cells (MCwT) was measured in LD, LL and DD. As seen in Fig. 6, MC1w had kept high mating reactivity continually in each condition. On the other hand, green cells (MCwT) induced from MC1w showed rhythmic mating reactivity in LD and LL. But they did not show the rhythmicity in DD and their reactivity was less than 50 %. It seems that the photosynthetic products of symbiotic Chlorella had reverted the arrhythmic mutant to mating reactivity rhythms in LD and LL.
DISCUSSION
Unicellular ciliate, Paramecium bursaria is an interesting material as a model of a coexisting plant cell in a single animal cell. Endosymbiotic Chlorella release their photosynthetic products, maltose and oxygen into their host cells. Especially the maltose release is a feature of the Chlorella symbiont in the cells of P. bursaria. It has been reported that Chlorella released glucose could not be re-infected into the cells of P. bursaria (Weis, 1980). It is considered that maltose is a signal for infecting to Paramecium cells. In this study, Chlorella treated with photosynthesis inhibitor DCMU were tested for their infectiousness. As seen in Fig. 1, they were hardly infected to Paramecium cells. Even if many Chlorella were induced into the host cells at first, they could not establish themselves as a symbiont. Their final infectiousness was less than 50 algal cells/one Paramecium cell. On the other hand, when Chlorella added to white cell suspension with maltose, they were infected to white cells as the same amount as ones untreated with DCMU. But they began to decrease 12 hr after infection and they could not be stable. These results suggest that maltose release from symbiotic Chlorella has important roles on the process of re-infection of Chlorella. First it would function as one of triggers to induce the Chlorella into digestive vacuoles (DV). And then, maltose may be required when the Chlorella enters the perialgal vacuoles (PV) to establish a symbiont. It seems that when Chlorella were inhibited a release of their own photosynthetic products in Paramecium cells by DCMU, they could not enter the PV and were digested in the DV.
On the other hand, photosynthetic products are also necessary to keep Chlorella as the symbionts in the Paramecium cell. In DD, the number of Chlorella in the green cell decreased, and reversely, the density of host cells in the culture increased in the progress (Fig. 2). It was expected that these relationships were caused by the lack of photosynthetic products released from Chlorella. Cells of P. bursaria would digest the endosymbiotic Chlorella that could not photosynthesize in DD. Paramecium cells can not express generally their mating reactivity in growth phase. Since the cell cycle of green cells was proceeding in DD, the mating ability of green cells disappeared or declined as seen in Figs. 3 and 4. When green cells were given maltose or oxygen in DD, they could show the mating reactivity rhythms (Fig. 5). Therefore the photosynthetic products of Chlorella were necessary to express the mating reactivity of green Paramecium cells. To make sure of these assumptions, we tried to assay the mating reactivity of cells treated with the photosynthesis inhibitor DCMU in LL. However, not only the green cells, but also the white cells never expressed any mating reactivity under DCMU (data not shown). Probably, DCMU has the effect to inactivate the mating reaction itself. Thus we should try to use other photosynthesis inhibitors that do not affect to mating reactivity of Paramecium cells.
Next, the effects of photosynthetic products of Chlorella on the expression of mating reactivity rhythms of P. bursaria were considered. Since P. bursaria show many kinds of circadian behaviors including mating reactivity and photo-accumulation, they are suitable materials to analyze the rhythms at cellular level (Miwa et al., 1987; Johnson et al., 1989). in a previous paper (Miwa et al., 1996b), we reported that symbiotic Chlorella forced the host cells to shift the phase and lengthen the period of photoaccumulation rhythm in LL. In this study, several effects of symbiotic Chlorella were also observed on the mating reactivity rhythms. As described in Results, green cells containing Chlorella did not exhibit mating reactivity in DD. Whereas, green cells given maltose or oxygen showed a mating reactivity rhythm in DD similar to LD and LL (Fig. 5). Moreover, arrhythmic mutant cells (MC1w) had been restored circadian rhythmicity in LD and LL by the infection of Chlorella, but not in DD (Fig. 6). These results also indicate that the photosynthesis of symbiotic Chlorella induce the mating reactivity rhythms. Moreover, when green cells were given maltose or oxygen at different three phases staggered by 6 h in DD, they showed mating reactivity rhythms of the same pattern. Photosynthetic products seem to act to express the mating reactivity according to the clock of host cell, not to express the mating reactivity directory. It will be of interest to examine the mating reactivity rhythms of MC1w after infection of Chlorella entrained with different LD cycles.
Then, how act Chlorella on host cells? Although cells of MC1w are mutants that express arrhythmic mating reactivity continuously, they show a circadian rhythm of photo-accumulation (Miwa et al., 1996a). Therefore their circadian system is oscillating normally if these two rhythms are operated by the same oscillator. In addition, it is expected that the arrhythmicity of mating reactivity is attributed to mitochondrial mutation because this property of MC1w is inherited cytoplasmically (Miwa et al., 1996a). If this is true, it is suggested that the mitochondria play an important part in the expression process of circadian mating reactivity rhythm. Furthermore, symbiotic Chlorella might interact with the host cells through the mitochondria. Further experiments and consideration are needed on this point of view.
