Endosymbiotic Chlorella variabilis reduces mitochondrial number in the ciliate Paramecium bursaria

Extant symbioses illustrate endosymbiosis is a driving force for evolution and diversification. In the ciliate Paramecium bursaria, the endosymbiotic alga Chlorella variabilis in perialgal vacuole localizes beneath the host cell cortex by adhesion between the perialgal vacuole membrane and host mitochondria. We investigated whether host mitochondria are also affected by algal endosymbiosis. Transmission electron microscopy of host cells showed fewer mitochondria beneath the algae-bearing host cell cortex than that of alga-free cells. To compare the density and distribution of host mitochondria with or without symbiotic algae, we developed a monoclonal antibody against Paramecium mitochondria. Immunofluorescence microscopy with the monoclonal antibody showed that the mitochondrial density of the algae-bearing P. bursaria was significantly lower than that of the alga-free cells. The total cell protein concentration of alga-free P. bursaria cells was approximately 1.8-fold higher than that of algae-bearing cells, and the protein content of mitochondria was significantly higher in alga-free cells than that in the algae-bearing cells. These results corresponded with those obtained by transmission electron and immunofluorescence microscopies. This paper shows that endosymbiotic algae affect reduced mitochondrial number in the host P. bursaria significantly.

. Uniquely, irrespective of the mutual relationship between P. bursaria and Chlorella spp., their relationship is not obligatory. Each can grow independently under suitable conditions. In addition, alga-free P. bursaria can be reinfected with isolated symbiotic algae by ingestion of the algae into host DVs. Based on these characteristics, P. bursaria and Chlorella spp. are considered textbook examples of endosymbiosis in protists 3 .
In previous studies, by pulse-labeling alga-free P. bursaria for 1.5 min with symbiotic algae isolated from algae-bearing paramecia and then chasing at specific times, we identified four important cytological events necessary for the establishment of endosymbiosis, as well as the timings of each during the algal infection process 7,[15][16][17][18][19][20][21][22][23] . These four cytological events are described below. (i) Three minutes after algal mixing, part of the algae is resistant to the host's lysosomal digestive enzymes in the DVs 16,18 . (ii) Within 30 min of mixing, algae in the DV begin budding from the DV membrane and enter the cytoplasm 16,20 . (iii) Fifteen minutes after budding, the DV membrane enclosing a single green Chlorella differentiates into a PV membrane, protecting the alga from host lysosomal fusion 16,18 . (iv) The alga surrounded by a PV membrane translocates beneath the host cell cortex 16,22,24 . Furthermore, the PV appears to localize near the host mitochondria and trichocysts 25 . Trichocysts are defensive organelles against predators embedded in the Paramecium cell cortex 26 . Indirect immunofluorescence microscopy using a monoclonal antibody (mAb) against the trichocysts demonstrates that the trichocysts change their localization to form algal attachment sites and decrease their number beneath the host cell cortex through algal reinfection 22 . Transmission electron microscopy (TEM) showed that some trichocysts www.nature.com/scientificreports/ near the host cell cortex were digested by host lysosomal fusion during algal reinfection 22 . These results indicate that symbiotic algae compete for their attachment sites with the pre-existing trichocysts and ensure attachment beneath the host cell cortex 22 . Many mitochondria are observed around the symbiotic Chlorella sp., similar to trichocysts in algae-bearing P. bursaria, and the symbiotic algae adhere to host mitochondria 25 . About host mitochondria in symbiotic state, Reisser reported that symbiotic algae in P. bursaria show a higher rate of photosynthetic oxygen production than those in the isolated state and thus guarantee an oxygen supply for the host 14 . Furthermore, recently He et al. also showed that the endosymbiotic algae may produce sufficient oxygen through their photosynthesis for the host cells to maintain cellular respiration in mitochondria 27 . Therefore, we had considered that the presence of symbiotic algae wrapped in the PV membrane may affect the number and function of host mitochondria. Song et al. reported that the PV membrane directly contacts the P. bursaria's mitochondrial membrane 28 . Similar results were observed in Mayorella visiris 29,30 , Toxoplasma gondii 31 , Plasmodium sporozoites 32 , and Encephalitozoon microsporidia 33 . Therefore, the association of host mitochondria with symbionts may be a universal component in symbiotic relationships.
Here, we observed cell cortex of alga-free and -bearing P. bursaria cells using TEM to examine the relationships between host mitochondria and symbiotic algae wrapped with the PV membrane. To examine whether the host mitochondria were affected by the endosymbiotic algae, we developed a mAb against Paramecium mitochondria. The relationships between the distribution of the symbiotic algae and the density of mitochondria were analyzed by an indirect immunofluorescence microscopy using mAb. The immunofluorescence localization was compared using the fluorescence of a mitochondria-specific dye, MitoTracker Green FM. Furthermore, the total cell protein concentration and that of the four kinds of cell extracts, including the mitochondrial membrane, was examined in the presence or absence of symbiotic algae. From the previous transcriptome data of host P. bursaria 34 , the expression levels of mitochondria-related genes were compared in the presence or absence of symbiotic Chlorella variabilis.

Methods
Strains and cultures. The symbiotic Chlorella sp.-free (alga-free) P. bursaria strain Yad1w was produced from the Chlorella sp.-bearing (algae-bearing) P. bursaria strain Yad1g as described previously 18 . The algaebearing Yad1g1N strain was produced by infecting Yad1w cells with cloned symbiotic Chlorella sp. 1 N cells 22 . This strain was identified as C. variabilis by rbcL gene analysis (accession number PRJDB12213, https:// ddbj. nig. ac. jp/ resou rce/ biopr oject/ PRJDB 12213 ). Both alga-free and algae-bearing P. bursaria cells were cultured in red pea (Pisum sativum) extract culture medium 35 with a modified Dryl's solution 36 (KH 2 PO 4 was used instead of NaH 2 PO 4 ·2H 2 O) and inoculated with Klebsiella aerogenes (strain ATCC 35028) 1 day before use. In all cultures, several hundred P. bursaria cells were inoculated into 2 mL aliquots of culture medium in test tubes. Subsequently, 2 mL aliquots of fresh culture medium were added every day and were cultivated for 12 days. The next day, the cultures reached to the early stationary phase of growth. All the cells used in this study were at this phase. Cultivation of the algae-bearing P. bursaria strain and all experiments were performed at 23 °C ± 1 °C under fluorescent lighting at 20-30 μmol photons m -2 s -1 using an incandescent lamp. Alga-free P. bursaria were cultivated under the same conditions, without lighting. All Paramecium strains used in this study were provided by the NBRP Paramecium Laboratory, Yamaguchi University, with support, in part, by the NBRP of the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) (http:// nbrpc ms. nig. ac. jp/ param ecium/? lang= en).
TEM. Both algae-bearing and alga-free P. bursaria were pre-fixed with 2% glutaraldehyde and prepared for TEM as described previously 37 . The paramecia embedded in Spurr's resin 38 were sectioned (70 nm thickness) using an ultramicrotome (Reichert Ultracut S; Leica Microsystems, Vienna, Austria) with a diamond knife, mounted on nickel mesh grids, and stained with lead citrate 39 . The sections were observed using TEM (CM120; Philips) at 80 kV.
Production of a mAb specific for Paramecium mitochondria. We produced a mitochondria-specific mAb as described in our previous research article 22 . The homogenate of algae-bearing P. bursaria was injected into the peritoneal cavity of an 8-week-old BALB/c mouse. A hybridoma clone, mAb2B8A8H1, was used in this study. Hybridoma cell production was performed following the guidelines for the use of animals in research at Yamaguchi University.
Indirect immunofluorescence microscopy. Indirect immunofluorescence microscopy was performed as described previously 22 . Aliquots of P. bursaria cells were air-dried on cover glasses (4.5 × 24 mm). The cells were then fixed with 4% (w/v) paraformaldehyde in phosphate-buffered saline (PBS) (137 mM NaCl, 2.68 mM KCl, 8.1 mM NaHPO 4 ·12H 2 O, 1.47 mM KH 2 PO 4 , pH 7.2) for 10 min at 4 °C. Then, the fixed cells were washed with PBST (PBS containing 0.05% Tween 20) and PBS for 10 min at 4 °C. The cells were then treated with mAb overnight at 4 °C. The cells were washed twice with PBS. Next, the cells were treated with Alexa Fluor 488 (AF488) goat anti-mouse IgG (Molecular Probes) diluted 1,000-fold with PBS for 2 h at 23 °C ± 1 °C and washed twice with PBS for 10 min. The samples were observed under differential interference contrast (DIC) and fluorescence microscopy (BX53; Olympus, Tokyo, Japan) equipped with Olympus fluorescence mirror units U-FBNA (excitation 470-495 nm, emission 510-550 nm) for AF488 and U-FGW (excitation 530-550 nm, emission 575 nm) for algal autofluorescence. Images were acquired using an Olympus DP74 system and analyzed using the Olympus cellSens Dimension software. www.nature.com/scientificreports/ Visualization of P. bursaria mitochondria with MitoTracker Green. MitoTracker Green was obtained from Molecular Probes (Eugene, OR, USA) and stored at -20 °C until further use. To stain the mitochondria with this reagent, approximately 5,000 algae-bearing P. bursaria cells/ml in MDS were mixed with MitoTracker Green at a concentration of 500 nM under constant dark conditions at 23 °C ± 1 °C, and the cells were observed using DIC and fluorescence microscopy within 10 min after mixing.
Protein quantification of the host P. bursaria with or without symbiotic algae. Alga-free and algae-bearing P. bursaria cells in the early stationary phase of growth were strained through two layers of Kimwipes to remove gross debris. The cells were transferred to a plastic beaker equipped with nylon mesh with a pore size of 15 μm. The paramecia were harvested, and the cell pellet was washed by pouring 100 mL PBS into the plastic beaker. The total cell extract was obtained as follows. Alga-free or algae-bearing P. bursaria at a density of 5 × 10 4 cells were lysed using the EzRIPA Lysis kit (ATTO, Tokyo, Japan) according to the manufacturer's instructions. After centrifugation, the supernatant was stored at − 80 °C until analysis. Extracts of the cytoplasm, including mitochondrial membrane, nucleus, and insoluble part (i.e., cytoskeleton) of P. bursaria cells were obtained as follows. Washed and harvested alga-free or algae-bearing P. bursaria at a density of 5 × 10 4 cells were treated with EzSubcell Extract (ATTO), and each extract was prepared according to the manufacturer's instructions. The extracts were stored at − 80 °C until analysis. Protein concentrations of total cell lysates and four types of cell extracts (cytoplasm, mitochondrial, nucleus, and insoluble part) were determined using the TaKaRa BCA Protein Assay Kit (Takara Bio, Shiga, Japan). Calibration curves were generated using PiCOEXPLORER (Yamato Scientific Co., Ltd., Japan, PAS-110-YU) with a color sensor R (wavelength range, 575-660 nm).

Quantification and statistical analysis. Statistical analyses were performed using Microsoft® Excel for
Mac (Ver. 16.51). Quantitative data were analyzed using the two-sided Fisher's exact test. For immunofluorescence intensity (a.u.), the statistical significance was determined using two-sided Fisher's exact test and the data of 10-12 cells. This experiment was repeated more than 10 times, and similar statistical results were confirmed. For the quantification of protein concentration, statistical significance was determined using two-sided Fisher's exact test and the data from four to five independent experiments.

Results
TEM of alga-free and -bearing P. bursaria. The cell cortex of the algae-bearing and alga-free P. bursaria were observed using TEM (Fig. 1). The host mitochondria surrounded the PV membrane wrapping the symbiotic alga (Fig. 1a). The symbiotic alga appeared to push the trichocysts aside to become fixed near the host cell cortex, as shown in our study 7 (Fig. 1b). In some algae-bearing cells, both host trichocysts and mitochondria were very few around the symbiotic alga beneath the host cell cortex (Fig. 1c). As shown by the gray arrowheads in Fig. 1b, we could observe mitochondria attached to the PV membrane, as observed in previous studies 25,28 . Meanwhile, in alga-free cells, many mitochondria and trichocysts were observed beneath the cell cortex (Fig. 1d). Mitochondria filled the gap just below the host cell cortex (Fig. 1e).
mAb specific for P. bursaria mitochondria. We successfully obtained a mAb specific for Paramecium mitochondria. Immunofluorescence was observed in the entire cell, particularly in the paramecium cell cortex (Fig. 2b,d). In the case of algae-bearing cells, fluorescence was not observed in areas where symbiotic algae were located (Fig. 2d). Immunofluorescence showed a higher number of mitochondria in the alga-free P. bursaria cells (Fig. 2a,b') than in the algae-bearing cells (Fig. 2c,e'). A merged photomicrograph of red autofluorescence of chlorophyll within the chloroplasts of the symbiotic algae and immunofluorescence clearly showed the localization of host mitochondria around symbiotic algae (Fig. 2e,e'). These immunofluorescence observations are in good agreement with the observations using TEM, as shown in Fig. 1. Images were obtained from a representative of more than ten independent experiments. The immunofluorescence intensity of the alga-free P. bursaria cells (Fig. 3, gray bar graph), was greater than that of the algae-bearing cells (Fig. 3, green bar graph). These quantitative data correspond well with the results of indirect immunofluorescence microscopy, as shown in Fig. 2.

Visualization of mitochondria in P. bursaria with MitoTracker Green. MitoTracker Green FM pas-
sively diffuses across the cell membrane, accumulates in active mitochondria, and shows green fluorescence. Figure 4a, b show a DIC image and a red autofluorescence of chlorophyll within chloroplasts of symbiotic algae image, respectively. The host mitochondria were localized around symbiotic algae (Fig. 4c). The fluorescence pattern of mitochondria of MitoTracker Green was consistent with that obtained using the mAb against mitochondria. After staining with MitoTracker Green, some DVs were formed, and nonspecific fluorescence was observed in the DVs because Paramecium takes in anything other than food from its cytopharynx (Fig. 4c, white arrowheads). This nonspecific fluorescence affects the fluorescence quantification, but this non-specific fluorescence was not observed using the mAbs, indicating the utility of mAbs for mitochondrial visualization.
Protein quantification of the host P. bursaria with or without symbiotic algae. First, the total cell protein concentrations of alga-free and algae-bearing P. bursaria were measured. Since the EzRIPA Lysis kit is used for animal cells, it cannot dissolve Chlorella cells. In fact, we confirmed that the isolated symbiotic Chlorella cells could not be dissolved using the EzRIPA Lysis buffer, and no protein was extracted (data not shown). The insoluble precipitates of alga-free and algae-bearing P. bursaria cells contained many intracellular crystals and symbiotic algae, respectively. We found that the protein concentration of 1,000 cells of alga-free P. bursa- www.nature.com/scientificreports/ ria cells was approximately 1.8-fold higher than that of algae-bearing cells (Fig. 5a). To understand the reason behind this significant difference in the concentrations, the protein concentrations of four extracts (cytosolic, including mitochondrial membrane, nuclear, and insoluble) from the 50,000 cells were examined. Of the four extracts, only the concentration of proteins of the mitochondrial extract was significantly different between algafree and algae-bearing P. bursaria cells. The protein concentration including mitochondrial membrane of algafree P. bursaria cells was significantly higher than that of algae-bearing cells (Fig. 5b). In the other three extracts, the protein concentration of alga-free P. bursaria cells was higher than that of algae-bearing cells; however, the difference was not significant. The protein concentration of the mitochondrial extract was in good agreement with the TEM and indirect immunofluorescence microscopy observations, as shown in Figs. 1, 2 and 3.  www.nature.com/scientificreports/ Differential expression of mitochondria-related genes between algae-bearing and alga-free P. bursaria. Table 1 shows the transcripts encoding mitochondria-related proteins in P. bursaria from Kodama et al 34 .
The negative values of log 2 fold change (log 2 FC) showed that mitochondrial genes were downregulated in algae-bearing cells compared to alga-free cells. The data sets are available in the DDBJ Sequence Read Archive (DRA) (accession number DRA000907, https:// ddbj. nig. ac. jp/ resou rce/ sra-submi ssion/ DRA00 0907). These results were consistent with the results obtained by TEM, indirect immunofluorescence microscopy, and quantification of the mitochondrial protein.

Discussion
In our study, TEM and indirect immunofluorescence microscopy using a mAb against Paramecium mitochondria revealed that host mitochondria are located in the proximity of symbiotic algae (Figs. 1 and 2). As shown in "Introduction" Section, because the symbiosome membrane seems to contacts the host's mitochondrial  Immunofluorescence intensity of mitochondria of alga-free and algae-bearing P. bursaria cells. In alga-free P. bursaria cells (gray bar graph), the immunofluorescence intensity is greater than that of algaebearing P. bursaria cells (green bar graph). Ten to twelve Paramecium cells were observed. Error bars show standard deviation (SD). Asterisks indicate significant differences (two-sided Fisher's exact test, ***P < 0.001).     10 . In general, most symbiotic algae do not flow in the cytoplasmic stream of P. bursaria but instead attach beneath the cell cortex. It is known that symbiotic algae fail to localize beneath the host cell cortex of mutant P. bursaria and form clusters in the host cytoplasm. This mutant cell passes an unequal distribution of the algal clusters to the daughter cell 40 . This suggests that the algal localization beneath the host cell cortex guarantees an equal distribution of symbiotic algae to daughter cells. Algal localization is an important phenomenon because the synchronization of host-symbiont cell cycles and co-segregation is critical in the permanent fusion of the two partners 41 . Hatena arenicola (Katablepharidophycota) is a single-celled eukaryote that harbors Nephroselmis. Because H. arenicola is regarded as an intermediate step organism in plastid acquisition, only one daughter cell inherits the symbiont during cell division, resulting in a symbiont-bearing green cell and a symbiont-lacking colorless cell. Interestingly, the symbiont-lacking cells have a feeding apparatus corresponding to the location of the eyespot in symbiont-bearing cells, and they can feed on prey cells [41][42][43] . In P. bursaria-Chlorella endosymbiosis, the symbiotic algae fine-tune their cell cycle pace with host Paramecium 44 . The symbiotic algal attachment in P. bursaria may be related to avoiding host lysosomal digestion or fusion because when the symbiotic algae are attached beneath the host cell cortex, the digestion never occurs (Kodama and Fujishima, unpublished. data). In some cases, the presence of symbiotic algae diminishes the mitochondrial activity in the host. Among symbiotic algae-bearing Hydra viridissima downregulated genes, mitochondria-related genes were found to be enriched using the Gene Ontology (GO) terms during differential gene expression analysis. Furthermore, most of the genes involved in the respiratory chain were significantly downregulated in symbiotic algae-bearing H. viridissima 45 . Transient suppression of mitochondrial metabolism and protein synthesis occurred when Acropora digitifera planulae were exposed to a competent strain of symbiotic algae, Symbiodinium 46 . The degradation of host mitochondria was verified when the symbiotic sea anemone Aiptasia pulchella was cultivated under hyperthermic stress conditions independent of symbiont cellular deterioration 47 . Reactive oxygen species (ROS) generated by symbionts and mitochondria have been shown to play a central role in damage to the host, such as in coral bleaching 47 . Therefore, hosts with symbiotic algae are believed to have reduced mitochondrial activity to suppress ROS production.
In a previous study, we showed that glutathione S-transferase (GST)-encoding genes were downregulated in symbiont-bearing cells compared to symbiont-free cells using transcriptome analysis 34 . This enzyme protects cells from oxidative stress 48,49 . Although it is conceivable that photo-oxidative stress in algae-bearing P. bursaria cells is greater than that in alga-free cells 50 , our previous data showed opposite results. Although a similar result of us was obtained by Hörtnagl and Sommaruga 51 , the exact mechanism remains unknown. This is the first study to report that algae-bearing P. bursaria cells have fewer mitochondria than alga-free cells do. Since algae-bearing P. bursaria receives photosynthetic products from symbiotic algae, a large number of mitochondria is not required. Because algae-bearing P. bursaria has fewer mitochondria than those of alga-free cells, the amount of ROS in algae-bearing cells may also be low. This might explain the suppression of GST in algae-bearing P. bursaria. Figure 6 shows the schematic representation of the algal attachment mechanism beneath the host cell cortex during the early algal reinfection process. This representation can be defined by our immunofluorescence microscopies using mAbs against both mitochondria and trichocysts. This is the first report that shows the change in the distribution of both mitochondria and trichocysts associated with the endosymbiotic algal localization beneath the host cell cortex. This can be explained in six points. (i) Before mixing with the symbiotic algae, P. bursaria has many mitochondria and trichocysts under the cell cortex. (ii) After mixing with the algae, (iii) the algae are Figure 6. A schematic diagram of the algal attachment mechanism beneath the host cell cortex during the early alga reinfection. Before mixing with the symbiotic algae, many mitochondria and trichocysts were observed under the cell cortex of alga-free P. bursaria (i). After mixing with algae (ii), the algae are wrapped with the DV membrane (iii). (iv) Then, the trichocysts under the cell cortex are digested, as shown in our previous study 22 . Mitochondrial number may be decreased at the same timing. (v) The alga enclosed by the PV membrane moves toward the host cell cortex for attachment to the trichocyst and mitochondria-less area after the DV-budding related to dynamin 20 . (vi) Consequently, algal reinfection process was completed. www.nature.com/scientificreports/ ingested and are wrapped with the DV membrane. Both mitochondria and trichocysts still show no remarkable changes. (iv) Three hours after mixing with the algae, trichocysts under the Paramecium cell cortex are digested as a result of the host lysosomal fusion as showed in our previous study 22 . Mitochondrial number may be decreased at the same timing, although the decreasing mechanism is still unknown. (v) After the DV-budding related to dynamin 20 , the alga enclosed by the PV membrane moves toward the host cell cortex for attachment to the trichocyst-less area 22 . (vi) Consequently, algal reinfection was completed and an algae-bearing cell is established. All oxygenic photosynthetic organisms, including algae, have an evolved P700 oxidation system. In response to high light and low CO 2 conditions, the primary electron donor in photosystem I (PSI), P700, is oxidized to suppress the production of ROS, which could oxidatively inactivate the function of PSI 52,53 . The symbiotic Chlorella spp. in P. bursaria might also retain the P700 oxidation system, although this has not yet been established. Thus, it is possible that symbiotic Chlorella spp. supply the inactivation of ROS because the algae also have an evolved P700 oxidation system; however, our study provides the first evidence that by the decrease of number of host mitochondria in the presence of a symbiont may cause the reduction of the amount of ROS in algae-bearing cells.
Algae-bearing P. bursaria are significantly larger than alga-free cells 54 . Interestingly, the quantification of total protein extract of host P. bursaria with or without symbiotic algae showed that the protein concentration of alga-free P. bursaria cells was significantly higher than that of algae-bearing cells (Fig. 5a). The reduced total protein contents of the algae-bearing P. bursaria cells may be related to the reduction of the numbers of membraned organelle of algae-bearing cells such as mitochondria as shown in Fig. 5b and trichocysts as shown in our previous study 22 . Furthermore, we have found that digestive activity of algae-bearing P. bursaria cells is highly suppressed than that of the alga-free cells, and the algae-bearing cells cannot digest as many bacteria as the algafree cells (Kodama and Fujishima, in preparation). Furthermore, previous transcriptome analysis between host P. bursaria with and without symbiotic C. variabilis showed the downregulation of ribosomal protein expression in symbiont-bearing P. bursaria cells, suggesting that algal proteins with functions equivalent to those of the host Paramecium cells are transferred to the host through the PV membrane, which consequently reduces the transcriptional activity of the host 34 . Our result of the quantification of total protein extract supports the transcriptome analysis.
A phenomenon that might explain the reduction in the number of mitochondria in the host has been observed. In Paramecium, a lot of mitochondria are found around organelles that require considerable energy utilization, such as basal bodies of the cilia, food granules, DVs, and contractile vacuoles, as well as near the cell cortex 55 . Alga-free P. bursaria swim longer distances compared to algae-bearing cells. The swimming distance and velocity of alga-free and algae-bearing P. bursaria need to be analyzed in future studies.
The following question arose from the present study. How does the number of mitochondria in Paramecium decrease during algal endosymbiosis? Bisharyan and Clark reported that the ciliates Tetrahymena thermophila and Ichthyophthirius multifillis can discard intact mitochondria to the extracellular space by triggering clusters of GPI-linked surface proteins or heat shock 56 . Their results suggest that the number of mitochondria in these ciliates can easily change.
Overall, our study provides credible evidence that the symbiosis of C. variabilis in P. bursaria induces a reduction in the number of host mitochondria. In the process which engulfed alga is acquired as plastids, there are four stages that Inouye and Okamoto have summarized in detail 57 . In stage I, the alga engulfed by the host cell is repeatedly taken up by the host cell as a temporal symbiont. In stage II, the symbiont and host cell divide synchronously, and the nuclei and mitochondria of the symbionts are retained. In stage III, mitochondria of the symbiont are erased one by one, but a vestigial symbiont nucleus is retained as a nucleomorph. In stage IV, the symbiont nucleus is erased, and only algal plastids are retained. Here, since mitochondrial erase as have shown in stage III also is occurring in algae-bearing P. bursaria cells, it is possible that the organellization of symbiotic algae may be progressing in P. bursaria. Although the endosymbiosis between P. bursaria and Chlorella spp. is not obligatory, the presence of symbiotic algae affects the number of host organelles. Since the number of mitochondria is higher in alga-free P. bursaria cells, reducing the host mitochondria is reversible. The mechanism of reduction in the number of host mitochondria caused by algal endosymbiosis needs to be investigated in future studies, which may help elucidate the mechanism by which symbionts become organelles in the process.

Data availability
The datasets used during the current study are available from the corresponding author on reasonable request.