Horizontal Transmission of Symbiotic Green Algae Between Hydra Strains.
The genus Hydra is classified into four species groups: viridissima, braueri, oligactis, and vulgaris, based on morphological and molecular phylogenetic analyses (Campbell, 1983, 1987, 1989; Hemmrich et al., 2007; Kawaida et al., 2010; Martinez et al., 2010; Schwentner and Bosch, 2015). Molecular phylogenetic analyses also suggest that the viridissima group (green hydras) diverged from the other three groups (brown hydras), and then the braueri group diverged from the oligactis and vulgaris groups. Green hydras form stable symbiotic relationships with green algae, typically Chlorella species, and this symbiosis has been investigated as an intriguing issue for research into the mechanisms of symbiosis (Kovacevic, 2012; Kobayakawa, 2017). Most of the brown hydras, on the other hand, do not form such symbiotic relationships.
However, some strains of vulgaris group hydras collected in Japan show symbiotic relationships with green algae. In the description of Hydra magnipapillata (a synonym of Hydra vulgaris) as a new species, Ito (1947) noted that some polyps were greenish-brown or green due to the presence of symbiotic algae. Two such greenish-colored strains of vulgaris group hydras (strains J7 and J10) were collected and have been maintained in the National Institute of Genetics (NIG, Mishima, Japan). These strains (described as H. magnipapillata) are identified as H. vulgaris based on recent phylogenetic analyses (Kawaida et al., 2010; Martinez et al., 2010; Schwentner and Bosch, 2015). Rahat and Reich (1985) described the endosymbiotic relationship between vulgaris group hydras and green algae, and at that time they suggested that the symbiotic algae belonged to the genus Chlorococcum; a few years later those authors described the symbiotic green algae as Symbiococcum hydrae gen. et sp. (order Chlorosarcinales), based on the morphological data and the reproduction mode (Rahat and Reich, 1989).
Recent molecular phylogenetic analyses suggest that the symbiotic green algae in hydras of strains J7 and J10 belong to the genus Chlorococcum (Kawaida et al., 2013; Ishikawa et al., 2016a). Rahat and Reich (1989) isolated the symbiotic green algae from the host polyps of the vulgaris group and succeeded in culturing them in vitro. They also found that the cultured symbiotic algae could form flagellated, motile zoospores under certain conditions (Rahat and Reich, 1991).
Rahat and Reich (1986) and Rahat and Sugiyama (1993) proposed that a specific relationship between the symbiont and the host may exist in the symbiosis between chlorococcum' and brown hydra, based on the results of their artificial transmission experiments using symbiotic algae. Ishikawa et al. (2016a) recently suggested that a specificity of the relationship between the symbiont and the host exists in the endosymbiosis between the vulgaris group hydras and the chlorococcum, based on their molecular phylogenetic analysis and an artificial transmission experiment using the symbiotic algae.
In artificial symbiotic algae transmission experiments, Rahat and Reich (1986) fed isolated symbiotic algae to Artemia larvae, and then the larvae were fed to non-symbiotic hydra polyps. Rahat and Sugiyama (1993) fed freshly hatched Artemia larvae to non-symbiotic hydra polyps, and then immediately gave the feeding polyps concentrated suspensions of isolated symbiotic algae in order to make an artificial transmission. The non-symbiotic polyps subsequently engulfed the algae together with the Artemia larvae. In some cases, engulfed algae were phagocytized by the endodermal epithelial cells (digestive cells) and formed symbiotic relationships with the host polyps. Other studies revealed that during feeding by the green hydras (i.e., viridissitna group hydras), the expulsion of symbiotic chlorellae into the gastric cavity was observed (Rahat and Reich, 1984; Fisman et al., 2008; Kawaida et al., 2013).
As described above, the symbiotic chlorococcum harbored in the symbiotic polyps may be transmitted to non-symbiotic polyps that do not ordinarily harbor symbiotic chlorococcum in nature. Indeed, in our maintenance of chlorococci-harboring H. vulgaris, that is, strains J7 and J10 in our laboratory, we found that the horizontal transmission of symbiotic chlorococci into some non-symbiotic strains, for example, H. vulgaris strain 105, occurred accidentally. Our preliminary observations indicated the horizontal transmission of symbiotic chlorococci from strain J7 polyps to other strains of vulgaris group hydra polyps (Kobayakawa, 2017).
In this study, we confirmed the occurrence of the horizontal transmission of symbiotic chlorococci from strain J10 to other hydra strains, including non-vulgaris group hydras. We also investigated the effects of symbiosis with chlorococci on the new host hydra polyps with respect to the morphology, behavior, and asexual proliferation rate.
Materials and Methods
We used five strains of vulgaris group hydras (J10, a symbiotic strain harboring green algae Chlorococcum sp. in the endodermal epithelial cells; 105, the standard strain of Hydra vulgaris Pallas, 1766, originally called Hydra magnipapillata; AEP, produced by mating between hydras from Pennsylvania and California; and K5 and K6, collected in Switzerland), two strains of oligactis group hydras (G7, a German Hydra oligactis strain; and L5, a Japanese H. oligactis strain, originally called Pelmatohydra robusta), two strains of braueri group hydras (HcKT1, a Japanese strain of Hydra circumcincta; and M7, a European strain of H. circumcincta), and one strain of viridissima group hydras (B5, an aposymbiotic Hydra viridissima strain K10). We have maintained strain 105, into which symbiotic chlorococci were transmitted and which has shown a symbiotic relationship with the algae for more than five years in our laboratory. We named the symbiotic strain 105G and used it in this study.
Hydras were maintained in 50-100 mL of hydra culture solution (HCS; 1 mmol [L.sup.-1] NaCl, 1 mmol [L.sup.-1] Ca[Cl.sub.2], 0.1 mmol [L.sup.-1] KC1, 0.1 mmol [L.sup.-1] MgS[O.sub.4], 1 mmol [L.sup.-1] tris-(hydroxymethyl)-aminomethane; pH 7.4, adjusted with HC1) in glassware at 20 [degrees]C under 14 h : 10 h light: dark (14L10D) illumination cycles. Polyps were fed newly hatched Artemia nauplius larvae two times per week. The day after feeding, polyps were transferred into the fresh HCS.
We could completely eliminate symbiotic chlorococci from strains J10 and 105G by treatment with [10.sup.-6] mol [L.sup.-1] 3-(3,4-dichlorophenyl)-1,1-dimethyl urea (DCMU) in darkness for two months. We named the obtained aposymbiotic strains J10_Apo and 105G_Apo, respectively.
Observation of the horizontal transmission of symbiotic Chlorococcum sp. from strain J10 symbiotic polyps to non-symbiotic polyps
To determine whether symbiotic chlorococci could be transmitted from symbiotic strain J10 polyps to other non-symbiotic hydra strains, we cultured a strain J10 polyp together with a non-symbiotic strain polyp in a small plastic container (12-well cell culture plate, 2-3 mL of HCS in each well) for at least 2 months. During the observation, newly hatched Artemia larvae were fed to polyps two times per week. Polyps newly formed by budding were removed from the container.
At one-week intervals, we used a fluorescence microscope (Optiphot, with EFDA2 and DXM1200F, Nikon, Tokyo) to examine each non-symbiotic strain polyp for a determination of whether symbiotic chlorococci were transmitted into the polyp. We observed more than six pairs for each combination between the J10 and non-symbiotic strains.
At 8 weeks after the start of co-culturing, we treated the upper-half gastric columns of the non-symbiotic strain polyps with hydra maceration medium (glycerol: acetic acid:[H.sub.2]O, 1:1:13) and dissociated the tissue into single cells to confirm the transmission of chlorococci into the endodermal epithelial cells. We measured the proportion of endodermal epithelial cells harboring chlorococci in the cytoplasm. If there were symbiotic chlorococci being harbored in an observed endodermal epithelial cell, we would also count the number of algae in the cell, under a Nomarski differential interference microscope (NISSR, Nikon).
Measurement of polyp length and count of endodermal epithelial cells in a polyp
We measured the length of the relaxed and elongated mature polyps that had buds in strains J10, 105, and 105G. It is difficult to measure the size of hydra polyps because of their elastic body; polyps usually repeatedly expand and contract, and the body column is not always standing straight. For the measurements, we thus observed each relaxed and elongated polyp that was floating under the water surface in an upside-down orientation.
To confirm the body size difference between non-symbiotic 105 polyps and symbiotic 105G polyps, we macerated each polyp; removed its bud, head, and basal disk; and counted the number of endodermal epithelial cells in each body column of the polyp with the use of a hemocytometer.
Observation of tentacles and stenoteles
As in the case of polyp length, it is difficult to measure tentacle length. We thus estimated the relative tentacle length by counting the number of stenoteles (a type of nematocyst) throughout a tentacle. One tentacle was cut from a polyp and anesthetized for 20 min in 100 [micro]L of HCS containing 1 % urethane. Then, 100 [micro]L of 8% paraformaldehyde was added to the solution and incubated for 20 min. The fixed tentacles were observed under a Nomarski differential interference microscope (Nikon), and the number of stenoteles was counted. We also measured the length of stenoteles by following the procedure of Campbell (1983).
Measurement of food ingestion
After 4 days of starvation, 1 polyp each of strains 105 and 105G was placed in the center of a plastic container (a 12-well cell culture plate). Then, 50 newly hatched Artemia larvae were released gently into each container, and the polyps were left overnight to allow them to eat the prey until satiety. The number of larvae eaten by each polyp was calculated using the number of dead larvae remaining in the container the next day.
Obsewation of endosymbiotic and floating free-living algae
Two or three symbiotic polyps of strains J10 and 105G were homogenized with a BioMasher (Nippi, Tokyo) in 100 [micro]L of HCS. and the homogenates were centrifuged at 2000 x g for a few minutes to extract endosymbiotic algae. The extracted algae were precipitated at the bottom of 1.5-mL microtubes. The precipitates were re-suspended in 500 [micro]L of HCS and were centrifuged at 2000 x g for a few minutes to wash the algae. After 3 washes, the precipitated algae were re-suspended in 50 [micro]L of HCS and mounted on a glass slide, and we observed them with a Nomarski differential interference microscope (Nikon).
To correct for the free-swimming algae from HCS that contained symbiotic hydras, strains J10 and 105G, 20-30 mL of HCS was centrifuged at 2000 x g for a few minutes. A small amount of precipitate was formed. We re-suspended it and observed the algae with the microscope, as for endosymbiotic algae. We also counted the free-swimming algae (in some cases using a hemocytometer) and estimated the concentration of algae in HCS.
Nucleotide sequences of algal rbcL and 18S rDNA genes
Because it is difficult to identify algal species on the basis of morphological characteristics, we determined the nucleotide sequences of the partial chloroplast genome-encoded gene, ribulose-bisphosphate carboxylase large subunit (rbcL), and the 18S ribosomal RNA gene (18S rDNA).
To obtain DNA from the endosymbiotic algal species, we used two or three symbiotic polyps. The polyps were homogenized with a BioMasher (Nippi) in a 50-[micro]L polymerase chain reaction (PCR) buffer (50 mmol [L.sup.-1] KCl, 10 mmol [L.sup.-1] tris-(hydroxymethyl)-aminomethane HC1; pH 8.3, 0.1% NP-40) and incubated with proteinase K (2 [micro]L of 20 mg m[L.sup.-1] proteinase K stock solution; Takara Bio, Shiga, Japan) at 55 [degrees]C for 15-60 min. After the inactivation of proteinase K (95 [degrees]C for 15 min), we used the solution as a DNA template for a PCR. We also obtained algal DNA from the precipitated free-living and swimming algae collected from the HCS of the symbiotic polyps. DNA was extracted with a DNeasy Plant Mini Kit (Qiagen, Hilden, Germany), following the manufacturer's instructions.
The PCR amplification of the 18S rDNA gene of symbiotic chlorococcum was performed with the primer pair 5'-GAGGATTGACAGATTGAGAGC-3' and 5'-GAACACTT CACCAGCACACC-3'. For the chloroplast rbcL gene, we used the primer pair 5'-AGGTCCTCCACACGGTATTCA-3' and 5'-TCAATAACAGCGTGCATAGC-3'. The PCR conditions were as follows: an initial denaturation step at 94 [degrees]C for 1 min, followed by 35 cycles of 30 s at 94 [degrees]C, 1 min annealing at 50 [degrees]C, and 1 min elongation at 72 [degrees]C; and a final elongation step of 5 min at 72 [degrees]C. The PCR products were directly sequenced using an ABI3730 DNA Analyzer and BigDye Terminator v3.1 Cycle Sequencing Kits (Thermo Fisher Scientific, Waltham, MA).
Measurement of the asexual proliferation rate of strains 105 and 105G
We put one polyp in a plastic container to measure the asexual proliferation rate under 14L10D illumination cycles, and polyps were fed two times per week. We counted the number of polyps reproducing by budding from the one polyp every day for four or eight weeks. We measured the rate under two different lighting conditions: light intensity 84 [micro]mol [m.sup.-2] [s.sup.-1 ]or 17 [micro]mol [m.sup.-2] [s.sup.-1] For each strain and each condition, more than six samples were counted. We counted a bud as 0.5.
Endosymbiotic algae Chlorococcum sp. of strain J10 polyps could be transmitted into different strains of non-symbiotic hydra polyps
We observed that endosymbiotic algae Chlorococcum sp. of strain J10 polyps were transmitted horizontally into several non-symbiotic hydra strains. As shown in Figure 1, at one or two months after the start of the co-culture with a strain J10 polyp, some of the examined polyps harbored green algae in their body. Most of the transmitted algae were located in the endoderm but not in the ectoderm of the new host polyps (Fig. 1 A*, B*, E*). To confirm the identity between the green algae in the new host polyps and the ones in the native symbiotic strain J10 polyps, we compared the partial nucleotide sequences of the rbcL gene in the algae. The examined nucleotide sequences (accession no. LC381699) were identical to those of Chlorococcum sp. in strain J7, which were the same as the sequences in strain J10 in the National Center for Biotechnology Information database (accession no. AB713414, rbcL). Our morphological observation of extracted symbiotic algae from strain J10 polyps and the new host polyps also suggested the identity between them (Fig. 2A, B).
We also observed that there were free-swimming flagellate algae in the HCS of strain J10 and 105G polyps (Fig. 2C, D). We estimated that the population of the zoospores in the HCS in which strain J10 polyps and other non-symbiotic polyps were co-cultured was 250-2000 m[L.sup.-1] (888 [+ or -] 608, mean [+ or -] SD, n= 12). Each free-living chlorococcum had a pair of flagella (Fig. 2C, D, arrows), which was not observed in the endosymbiotic algae in endodermal epithelial cells of the host polyps (Fig. 2A, B). We confirmed the identity between the endosymbiotic algae and the free-swimming flagellate algae by sequencing the rbcL and 18SrDNA genes (accession nos. LC381698 [rbcL] and LC381701 [18SrDNA]). The nucleotide sequences were identical to those of strain J10. These free-swimming zoospores were morphologically similar to the particular flagellated motile zoospores that Rahat and Reich (1991) reported had been induced when culturing endosymbiotic chlorococci.
The extent of transmission of symbiotic algae from the source of strain J10 polyps differed between strains. In the three vulgaris group strains, the extent of transmission depended on the strain (Fig. 1A-C) and was correlated with the phylogenetic distance. In particular, strain 105, which has a close relationship with strain J10, harbored as many green algae as strain J10 (Fig. 1 A1, A2, A*). However, almost no green alga was observed in the polyps of strains K5 and K6, which have distant relationships with strain J10 (Fig. 1C1, C2). The polyps of strain AEP harbored an intermediate number of algae (Fig. 1B1, B2, B*). The horizontal transmission was not restricted to the vulgaris group (Fig. 1E, F). The braueri group strain HcKT1 polyps harbored a relatively large number of green algae (Fig. 1E1, E2, E*).
We identified chlorococci in macerated endodermal epithelial cells of strains in each group (Fig. 3). In strains 105 and HcKT1, the proportion of the endoderm cells that harbored algae and the number of green algae per endoderm cell were higher than those of strain AEP and other strains (Fig. 4). After 2 months of co-culture with strain J10 polyps, about 75% and 30% of the endodermal cells harbored algae in the cytoplasm in strains 105 and HcKT1, respectively. On the other hand, < 15% of the endodermal cells harbored algae in strain AEP, and almost none harbored algae in strains K6 and L5 (Fig. 4, white bars). The numbers of endosymbiotic algae in a single endodermal cell in strains 105 and HcKTl (about 3.5 and 2.0, respectively) were less than that of strain J10 (about 6.0; Fig. 4, dark bars).
Only strain 105 polyps had maintained the symbionts without co-culturing of strain J10 polyps after the eight-week co-culture, whereas we found that some other vulgaris group strains, that is, BW and SW collected in Lake Biwa and Fukuoka, Japan, respectively, maintained horizontally transmitted chlorococci without co-cultured strain J10 polyps and maintained the symbiosis for more than several months. We confirmed the increase of transmitted symbionts in 105 strain polyps after removal of co-culturing J10 polyps (Fig. 1, bottom row, A2W and A2W-1W). In contrast, other non-vulgaris strains, that is, HcKT1 and B5, could not maintain transmitted symbionts after the removal of the co-cultured strain J10 polyps from the culture wells. Strain AEP polyps also lost the transmitted symbionts within a few months after the removal of co-cultured strain J10 polyps. The decreased process of transmitted symbionts in HcKTl polyps after the removal of co-culturing J10 polyps is shown in Figure 1 (bottom row, E2W, E2W-1W, and E2W-2W).
Endosymbiosis with Chlorococcum sp. affected the morphology and behavior of strain 105G
As described above, in our laboratory we have maintained the symbiotic strain 105G (which originated from strain 105), to which symbiotic algae from strain J10 polyps were transmitted; and it has maintained a stable symbiotic relationship for more than five years. Strain 105G polyps showed characteristics that are remarkably different from those of the original non-symbiotic strain 105 polyps. We therefore investigated the effects of newly transmitted symbiotic algae on the host polyps from several different perspectives.
The length of symbiotic strain 105G polyps was significantly shorter than that of non-symbiotic strain 105 polyps (Fig. 5A). The average length of strain 105 polyps was 9.3 [+ or -] 2.5 mm, and that of 105G polyps was 5.8 [+ or -] 2.0 mm (mean [+ or -] SD, P < 0.01 by Student's t test, n = 12). Strain J10 polyps were almost the same length (9.3 [+ or -] 2.2 mm) as strain 105 polyps.
We counted the total endodermal epithelial cells of body columns of polyps to further examine the size decrease of the host polyps by the symbiosis (Fig. 5B). The number of endodermal epithelial cells in a body column (excluding the head, basal disk, and bud) of non-symbiotic strain 105 polyps was 8908 [+ or -] 1460 (mean [+ or -] SD, n = 6). In symbiotic strain 105G polyps, the number of cells decreased to 2928 [+ or -] 1363 (mean [+ or -] SD, n = 6; the difference between strains 105 and 105G was significant, P < 0.01 by Tukey-Kramer method). The sizes of endodermal epithelial cells were not significantly different between these strains. In aposymbiotic strain 105G polyps (105G_Apo), the number of cells was 6835 [+ or -] 1329 (mean [+ or -] SD, n = 6). The difference between strains 105 and 105G_Apo was not significant. We also noticed that polyp size increased in the aposymbiotic strain J10 (J10_Apo) compared with the original symbiotic strain J10. To confirm the size change, we counted the total endodermal epithelial cells of body columns of polyps. The number of cells was 14,190 [+ or -] 3769 (mean [+ or -] SD, n = 6) in strain J10_Apo polyps and 8520 [+ or -] 3070 (mean [+ or -] SD, n = 6) in strain J10 polyps.
As can be seen from Figure 5A, tentacle length was shortened in symbiotic strain 105G polyps compared to non-symbiotic strain 105 polyps. To further examine the change of tentacle length, we counted the stenoteles in each tentacle of strains 105G and 105. The number of stenoteles in each tentacle of strain 105G polyps was about one-fourth that of strain 105 polyps (128 [+ or -] 53.6 vs. 471 [+ or -] 31.3, mean [+ or -] SD, n = 6 in each group), indicating that the tentacle length of symbiotic polyps of strain 105G is about one-fourth of that of non-symbiotic strain 105 polyps, if the stenoteles are distributed uniformly in the tentacles.
We next focused on the morphology of nematocysts in tentacles and compared the size of stenoteles between strains 105G and 105. As shown in Figure 6, the stenoteles of strain 105G polyps tended to be shorter than those of strain 105 polyps. The size distribution in strain 105 polyps had a peak at about 15-18 [micr]m, whereas the peak of size distribution in strain 105G polyps was about 9-12 jum. In the case of strain J10, however, the distribution was similar to that of the non-symbiotic strain 105 (Fig. 6).
The main role of stenoteles in hydra tentacles is to capture prey. It is intriguing to consider whether these morphological changes in tentacles and stenoteles affect their feeding ability. We quantified the amount of food ingested by polyps of strains 105G and 105. When we fed 50 larvae to each polyp, strain 105 polyps ingested 22.3 [+ or -] 6.2 larvae, whereas symbiotic strain 105G polyps ingested 6.0 [+ or -] 4.2 larvae (P < 0.01 by Student's t test, n = 8 in both groups).
Endosymbiosis with Chlorococcum sp. affected the asexual proliferation of strain 105
The asexual proliferation rate by budding was increased by endosymbiosis (Fig. 7A; 14L10D illumination cycles, light intensity 84 [micro]mol [m.sup.-2] [s.sup.-1]). We also found that light conditions influenced the growth rate of strain 105G polyps. We examined the effects of algal symbiosis on asexual proliferations under two different light conditions: high-light condition, 84 [micro]mol [m.sup.-2] [s.sup.-1]; low-light condition, 17 [micro]mol [m.sup.-2] [s.sup.-1 ](Fig. 7B). Under the high-light condition, the asexual proliferation rate became higher by symbiosis. The doubling times of the 105 and 105G polyps under the high-light condition were 21.2 [+ or -] 2.2 days and 12.4 [+ or -] 1.4 days (mean [+ or -] SD, n = 6), respectively.
The transmitted chlorococci could establish symbiotic relationships with only restricted strains of vulgaris group hydras
Based on the "phylosymbiotic" (Brooks et al., 2016) analysis between viridissima group hydras and Chlorella sp., the co-evolution of the host and the symbionts was suggested (Kawaida et al., 2013). In the brown hydras, Rahat and Reich (1986) succeeded in achieving artificial chlorococcum symbiosis with some hydra strains belonging to the vulgaris group (which were Australian and African strains regarded as Hydra attenuate at the time), but not with some other strains. However, they did not examine the phylogenetic closeness between the hydra strains that successfully established symbiosis with the chlorococci. Ishikawa et al. (2016a) suggested that only a few phylogenetically close strains in the vulgaris group hydras could establish endosymbiosis with chlorococci of strain J7. They also proposed that the evolution of the endosymbiosis of chlorococcum in the vulgaris group hydras occurred in two steps: (1) endosymbiotic potential was gained once in an ancestor of a lineage of the vulgaris group hydras, and (2) only the two strains J7 and J10 obtained the symbiotic algae independently after the divergence of the strains.
In this study, we recognized three types of hydra strains by the difference in reaction to chlorococci of strain J10 that were taken into the gastric cavity. Endodermal epithelial cells of the first type of hydra strains do not phagocytize as many chlorococci and do not allow chlorococci to stay in their cytoplasm, probably through digestion (e.g., vulgaris group: K5, K6; oligactis group: L5, G7). In the second type of hydras, endodermal epithelial cells phagocytize the chlorococci well, and the incorporated chlorococci can survive for a certain period of time in the cytoplasm but cannot form symbiotic relationships with the host (e.g., vulgaris group: AEP; braueri group: HcKT1, M7). The third type of hydras can incorporate chlorococcum in their endodermal cells, and the chlorococci survive and reproduce there. The relationship may be able to develop into symbiosis (e.g., vulgaris group: 105, BW, SE). The third type of hydras may be equivalent to the hydras that have gained the "endosymbiotic potential" (Ishikawa et al., 2016a). The process of acquiring the endosymbiotic potential may be clarified by a comparative study of reactions to the symbiotic chlorococcum by strain AEP and 105 hydras.
Endosymbiotic Chlorococcum sp. enhanced the asexual proliferation of host strain 105 polyps, but the polyps became smaller
Sugiyama and Fujisawa (1973) reported that there was a negative correlation between the polyp size of hydras and their asexual proliferation rate, and they suggested a model mechanism of budding signals in which the time required by buds to grow larger adult polyps will be longer. Conversely, small polyps can have a higher asexual proliferation rate. The data in this study on the size and asexual proliferation rate of strain 105 and 105G polyps are consistent with their report.
Morphological and asexual proliferation rate changes similar to those observed herein for strain 105G were not observed in a prior study comparing a native symbiotic hydra strain (J7) with an aposymbiotic hydra strain whose symbionts were removed artificially (Ishikawa et al., 2016b). In our study, however, elimination of symbiotic chlorococci of the J10 polyps caused polyp size to increase (Fig. 5B). It is possible that the interaction with symbiotic chlorococci in strain 105G is similar to that in strain J10 but different from that in strain J7. The two native symbiotic hydra strains reacted to the elimination of symbionts with different polyp size changes. In the well-established symbiosis between green hydras and Chlorella sp., under moderate and low feeding conditions (once per 7 days and once per 14 days, respectively), asexual growth was reduced in polyps lacking algae, without polyp size change (Habetha et al., 2003). On the other hand, the two native symbiotic vulgaris group hydra strains, J7 and J10, reacted to the elimination of symbionts with different changes of growth rate and of polyp size. These erratic reactions suggest that a symbiotic relationship between chlorococcum and vulgaris group hydras still has not been strongly established in nature.
The location of chlorococci in an endodermal cell also suggests the unstable symbiosis of vulgaris group hydras (e.g., J7, J10, 105G). Chlorococci were not restricted to the basal side (Fig. 3), whereas most of the symbiotic chlorellae in green hydras locate on the basal side. It is suggested that the location of symbiotic chlorellae in green hydras is essential for the avoidance of digestion by the host endodermal epithelial cells (Kobayakawa, 2017).
The viridissima group hydras, which have endosymbiotic chlorellae, regulate the number of symbionts so that the symbionts do not increase in size excessively (Douglas and Smith, 1984;Dunn, 1987; Fishman et al., 2008). The hydra cells must digest the symbionts in accord with the growth of the host in order to maintain constant algal density. Slobodkin et al. (1991) proposed that it was difficult for large polyps of the viridissima group to control the increase in algae due to the greaterrate of algae removal. Overgrowth of the symbiotic algae causes the death of a hydra's digestive cells (Taylor et al., 1989). It disrupts the symbiotic relationship between hydras and algae. Strain 105G seems to shrink polyp sizes to facilitate the control of the number of symbionts.
However, it was reported that there is no significant difference in body size between symbiotic hydras and aposymbiotic hydras in the viridissima group (Habetha et al., 2003). This pattern shows the stable symbiotic relationship of the green hydras. It is assumed that the origin of the symbiosis in a common ancestor of the viridissima group was before the Cretaceous period (Kawaida et al., 2013; Schwentner and Bosch, 2015). If strain 105G has not yet evolved the function of efficiently removing the algae, the strain needs to shrink its polyp size for its survival in order to avoid overgrowth of the symbiotic algae.
Another hypothesis is that the symbionts force the hosts to shrink their polyp sizes. The symbionts of strain 105G changed their host from strain J10; in other words, they moved to a new host by horizontal transmission. Mutualism and parasitism show similar evolutionary patterns, and symbionts may change between mutualism and parasitism under conflicts of interest between the host and the symbionts (Sachs et al., 2011). Symbionts that are transmitted horizontally tend to be more harmful to their hosts than those transmitted vertically (Douglas, 1998; Sachs and Wilcox, 2006). For symbiotic chlorococcum, strain 105 is a new host through horizontal transmission.
Hydras have few predators, due to the presence of nematocysts and their low mobility (Massaro et al., 2013). Accordingly, symbiotic algae can be dispersed while being protected by host polyps. The benefits of the dispersal and immigration of symbionts often conflict with host fitness (Frank, 1996). Larger polyp size is probably advantageous for hydras in their ability to prey on food (Bossert and Dunn, 1986). On the other hand, in regard to the dispersion of the symbionts, it is advantageous for the symbionts to make hydras increase their budding rates at the cost of polyp size. As the number of hosts increases, it becomes easier for the symbionts in the hosts to disperse. The symbionts in strain 105G may behave "selfishly" for their own dispersal and immigration, and they may shrink the polyp size of the host, unlike those in strain J10. The reason that symbiotic hydras of the vulgaris group are rarely seen in the wild may be because a shrinking polyp size causes selective pressure on the host hydras. If this is the case, it indicates that strain 105G and its symbionts have not yet established a stable mutualistic relationship.
Our present findings provide evidence that the endosymbiotic Chlorococcum sp. can horizontally transmit into all hydra species groups via surrounding water, at least temporarily. Especially in some vulgaris group strains, we observed that symbiosis with chlorococci in the non-symbiotic strains caused smaller polyp size and higher asexual proliferation rates, compared to those in the individuals without symbionts. These changes may occur by regulating the number of symbiotic chlorococci and/or because of conflicts of interest between the host and the symbiont. For the elucidation of the precise mechanisms underlying symbiosis through horizontal transmission, it is necessary to conduct further molecular analyses of symbiotic hydras.
We thank the National Institute of Genetics (Mishima, Japan) for providing hydra strains. We sincerely appreciate the referees' helpful suggestions. This research was partially supported by a Grant in Aid (16K07465) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan to YK, HT, and JK.
Bossert, P., and K. Dunn. 1986. Regulation of intracellular algae by various strains of the symbiotic Hydra viridissima. J. Cell Sci. 85: 187-195.
Brooks, A. W., K. D. Kohl, R. M. Brucker, E. J. van Opstal, and S. R. Bordenstein. 2016. Phylosymbiosis: relationships and functional effects of microbial communities across host evolutionary history. PLoS Biol. 14: e2000225.
Campbell, R. D. 1983. Identifying hydra species. Pp. 19-28 in Hydra: Research Methods. H. Lenhoff, ed. Springer, New York.
Campbell, R. D. 1987. A new species of Hydra (Cnidaria: Hydrozoa) from North America with comments on species clusters within the genus. Zool. J. Linn. Soc. 91: 253-263.
Campbell, R. D. 1989. Taxonomy of the European Hydra (Cnidaria: Hydrozoa): a re-examination of its history with emphasis on the species H. vulgaris Pallas. H. attenuata Pallas and H. circumcincta Schulze. Zool. J. Linn. Soc. 95:219-244.
Douglas, A. 1998. Host benefit and the evolution of specialization in symbiosis. Heredity 81: 599-603.
Douglas, A., and D. C. Smith. 1984. The green hydra symbiosis. VIII. Mechanisms in symbiont regulation. Proc. R. Soc. Biol. Sci. B 221: 291-319.
Dunn, K. 1987. Growth of endosymbiotic algae in the green hydra. Hydra viridissima. J. Cell Sci. 88: 571-578.
Fishman, Y., E. Zlotkin, and D. Sher. 2008. Expulsion of symbiotic algae during feeding by the green hydra: a mechanism for regulating symbiont density? PLoS One 3: e2603.
Frank, S. A. 1996. Host-symbiont conflict over the mixing of symbiotic lineages. Proc. R. Soc. Biol. Sci. B 263: 339-344.
Habetha, M., F. Anton-Erxleben, K. Neumann, and T. C. G. Bosch. 2003. The Hydra viridislChlorella symbiosis: growth and sexual differentiation in polyps without symbionts. Zoology 106: 101-108.
Hemmrich, G., B. Anokhin, H. Zacharias, and T. C. G. Bosch. 2007. Molecular phylogenetics in Hydra, a classical model in evolutionary developmental biology. Mol. Phylogenet. Evol. 44: 281-290.
Ishikawa, M., H. Shimizu, M. Nozawa, K. Ikeo, and T. Gojobori. 2016a. Two-step evolution of endosymbiosis between hydra and algae. Mol. Phylogenet. Evol. 103: 19-25.
Ishikawa, M., I. Yuyama, H. Shimizu, M. Nozawa, K. Ikeo, and T. Gojobori. 2016b. Different endosymbiotic interactions in two Hydra species reflect the evolutionary history of endosymbiosis. Genome Biol. Evol 8:2155-2163.
Ito, T. 1947. A new fresh-water polyp, Hydra magnipapillata, n. sp. from Japan. Sci. Rep. Tohoku Univ. 4th Ser. Biol. 18: 6-10.
Kawaida, H., H. Shimizu, T. Fujisawa, H. Tachida, and Y. Kobayakawa. 2010. Molecular phylogenetic study in genus Hydra. Gene 468: 30-40.
Kawaida, H., K. Ohba, Y. Koutake, H. Shimizu, H. Tachida, and Y. Kobayakawa. 2013. Symbiosis between hydra and chlorella: Molecular phylogenetic analysis and experimental study provide insight into its origin and evolution. Mol. Phylogenet. Evol. 66: 906-914.
Kobayakawa, Y. 2017. Symbiosis between green algae and Hydra. Pp. 347-369 in Algal and Cyanobacteria Symbiosis, M. Grube. J. Seckbach. and L. Muggia. eds. World Scientific. Singapore.
Kovacevic, G. 2012. Value of the Hydra model system for studying symbiosis. Int. J. Dev. Biol. 56: 627-635. doi:I0.1387/rjdb.l23510gk.
Martinez, D. E., A. R. Iniguez, K. M. Percell, J. B. Willner, J. Signorovitch, and R. D. Campbell. 2010. Phylogeny and biogeography of Hydra (Cnidaria: Hydridae) using mitochondrial and nuclear DNA sequences. Mol. Phylogenet. Evol. 57: 403-410.
Massaro, F. C., N. F. Negreiros, and O. Rocha. 2013. A search for predators and food selectivity of two native species of Hydra (Cnidaria: Hydrozoa) from Brazil. Biota Neotrop. 13: 35-40.
Rahat, M., and V. Reich. 1984. Intracellular infection of aposymbiotic Hydra viridis by a foreign free-living Chlorella sp.: initiation of a stable symbiosis. J. Cell Sci. 65: 265-277.
Rahat, M., and V. Reich. 1985. A new alga/hydra symbiosis: Hydra magnipapillata of the "nonsymbiotic" vulgaris group hosts a Chlorococcum-like alga. Symbiosis 1: 177-184.
Rahat, M., and V. Reich. 1986. Algal endosymbiosis in brown-hydra: host/symbiont specificity. J. Cell Sci. 86: 273-286.
Rahat, M., and V. Reich. 1989. Symbiococcum hydrae gen. et sp. nov. (Chlorosarcinales. Chlorophyta): an endosymbiotic green alga from cells of the Japanese Hydra magnipapillata (Coelenterata). Phycologia 28: 181-187.
Rahat, M., and V. Reich. 1991. Zoospore formation by Symbiococcum hydrae (Chlorosarcinales. Chlorophyta), an alga endosymbiotic in Hydra magnipapillata (Coelenterata). Phycologia 30: 226-230.
Rahat, M., and T. Sugiyama. 1993. The endodermal cells of some "brown" hydra are autonomous in their ability "to host or not to host" symbiotic algae: analysis of chimera. Endocytobiosis Cell Res. 9: 223-231.
Sachs, J. L., and T. P. Wilcox. 2006. A shift to parasitism in the jellyfish symbiont Symbiodinium microadriaticum. Proc. R. Soc. Biol. Sci. B 273: 425-429.
Sachs, J. L., C. J. Essenberg, and M. M. Turcotte. 2011. New paradigms for the evolution of beneficial infections. Trends Ecol. Evol. 26: 202-209.
Schwentner, M., and T. C. G. Bosch. 2015. Revisiting the age, evolutionary history and species level diversity of the genus Hydra (Cnidaria: Hydrozoa). Mol. Phylogenet. Evol. 91: 41-55.
Slobodkin, L. B., P. Bossert, C. Matessi, and M. Gatto. 1991. A review of some physiological and evolutionary aspects of body size and bud size of Hydra. Hydrobiologia 216/217: 377-382.
Sugiyama, T., and T. Fujisawa. 1973. Genetic analysis of developmental mechanisms in hydra. VII. Statistical analyses of developmental-morphological characters and cellular compositions. Dev. Growth Differ. 21: 361-375.
Taylor, C. E., L. Muscatine, and D. R. Jefferson. 1989. Maintenance and breakdown of the hydra-chlorella symbiosis: a computer model. Proc. R. Soc. Biol. Sci. B 238: 277-289.
RYO MIYOKAWA (1,2,*), TAKUYA TSUDA (1,*), HIROYUKI J. KANAYA (3,*), JUNKO KUSUMI (4), HIDENORI TACHIDA (5), AND YOSHITAKA KOBAYAKAWA (6,[dagger])
(1) Graduate School of Systems Life Sciences, Kyushu University, 744 Moto-oka, Nishi-ku, Fukuoka 819-0395, Japan; (2) Graduate School of Integrated Science for Global Society, Kyushu University, 744 Moto-oka, Nishiku, Fukuoka 819-0395, Japan; (3) School of Science, Kyushu University, 744 Moto-oka, Nishi-ku, Fukuoka 819-0395, Japan; (4) Graduate School of Social and Cultural Studies, Kyushu University, 744 Moto-oka, Nishi-ku, Fukuoka 819-0395, Japan; (5) Department of Biology, Faculty of Science, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan; and (f) Facultyof Arts and Science, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan
Received 26 December 2017; Accepted 10 July 2018; Published online 21 September 2018.
(*) These authors equally contributed to this study.
([dagger]) To whom correspondence should be addressed. E-mail: email@example.com.
Abbreviations: 14L10D, 14 h : 10 h light : dark; HCS, hydra culture solution; PCR, polymerase chain reaction.
(1) The use of "chlorococcum" and "chlorococci" throughout the article reflects familiar forms of Chlorococcum sp.
|Printer friendly Cite/link Email Feedback|
|Author:||Miyokawa, Ryo; Tsuda, Takuya; Kanaya, Hiroyuki J.; Kusumi, Junko; Tachida, Hidenori; Kobayakawa, Yos|
|Publication:||The Biological Bulletin|
|Date:||Oct 1, 2018|
|Previous Article:||Comparative Oxygen Consumption of Gastropod Holobionts from Deep-Sea Hydrothermal Vents in the Indian Ocean.|
|Next Article:||Morphological Variability Among Broods of First-Stage Blue Crab (Callinectes sapidus) Zoeae.|