Multiple surveys employing a new sample‐processing protocol reveal the genetic diversity of placozoans in Japan

Abstract Placozoans, flat free‐living marine invertebrates, possess an extremely simple bauplan lacking neurons and muscle cells and represent one of the earliest‐branching metazoan phyla. They are widely distributed from temperate to tropical oceans. Based on mitochondrial 16S rRNA sequences, 19 haplotypes forming seven distinct clades have been reported in placozoans to date. In Japan, placozoans have been found at nine locations, but 16S genotyping has been performed at only two of these locations. Here, we propose a new processing protocol, “ethanol‐treated substrate sampling,” for collecting placozoans from natural environments. We also report the collection of placozoans from three new locations, the islands of Shikine‐jima, Chichi‐jima, and Haha‐jima, and we present the distribution of the 16S haplotypes of placozoans in Japan. Multiple surveys conducted at multiple locations yielded five haplotypes that were not reported previously, revealing high genetic diversity in Japan, especially at Shimoda and Shikine‐jima Island. The observed geographic distribution patterns were different among haplotypes; some were widely distributed, while others were sampled only from a single location. However, samplings conducted on different dates at the same sites yielded different haplotypes, suggesting that placozoans of a given haplotype do not inhabit the same site constantly throughout the year. Continued sampling efforts conducted during all seasons at multiple locations worldwide and the development of molecular markers within the haplotypes are needed to reveal the geographic distribution pattern and dispersal history of placozoans in greater detail.

Placozoans have been reported at 76 locations in temperate to tropical seas worldwide, and 19 haplotypes forming seven clades have been identified using 16S sequences thus far (see Figure 2; Eitel et al., 2013;Nakano, 2014). The number of existing haplotypes is estimated to be approximately 200 (Eitel & Schierwater, 2010;Eitel et al., 2013), and a habitat suitability model analysis suggested that placozoans are probably present in numerous nonsampled regions (Paknia & Schierwater, 2015). The current worldwide distribution map of placozoan 16S haplotypes is based on the genotyping results from 47 locations, but sampling has been performed only once at 38 of these locations (Eitel et al., 2013). Multiple surveys conducted in the  (Eitel & Schierwater, 2010;Signorovitch et al., 2006), leading the authors to believe that continued sampling during different seasons would also yield different haplotypes at a single location in temperate regions. In Japan, placozoans have been reported at nine locations in the three surrounding seas: the Sea of Japan, the Northern Pacific Ocean, and the East China Sea (Eitel & Schierwater, 2010;Nakano, 2014;Pearse, Uehara, & Miller, 1994;Pearse & Voigt, 2007;Sudzuki, 1977;Ueda, Koya, & Maruyama, 1999), and haplotypes collected at two locations have been reported: H15 for Shirahama, Wakayama (Miyazawa, Yoshida, Tsuneki, & Furuya, 2012) and H2 for Chatan, Okinawa (Eitel & Schierwater, 2010).
Placozoans have been sampled from shallow (<20 m depth) seawater environments using two distinct methods (Maruyama, 2004): (1) In slide sampling, glass or plastic slides are placed in natural seawater and are generally retrieved after more than 10 days. Placozoans on the slides are collected under a stereomicroscope. (2) In substrate sampling, natural substrate materials such as stones and molluscan shells are collected together with ambient seawater inside a container. There are several protocols for subsequent processing of the substrates after collection. One method for processing these substrates, a process we refer to as "passive substrate sampling," is to place the substrates and seawater into a tank and collect the animals from the walls of the tank (Miyazawa et al., 2012). Slides may also be placed in the tank, and placozoans on the slides can then be collected under a stereomicroscope (Miyazawa et al., 2012). Another method is what we refer to as "agitation substrate sampling," in which the container with the substrate and seawater is vigorously shaken for several seconds to detach placozoans from substrates. The seawater is then decanted into Petri dishes, and placozoans are collected from the Petri dishes under F I G U R E 1 Photograph of Placozoa sp. H2, collected at Shimoda. Scale bar = 200 μm F I G U R E 2 Cladogram of placozoan 16S haplotypes. Phylogenetic relationships, Bayesian posterior probabilities supporting nodes, clade names, and color codes are based on Eitel et al. (2013). For the haplotypes collected in Japan, the locations presented in Figure 3 are each shown on the right. Locations genotyped in previous studies: asterisks; those in this study: bold a stereomicroscope. Slide sampling requires special equipment to prevent the submerged glass slides from being carried away or broken by the current and requires a minimum of approximately 10 days to collect the animals. Passive substrate sampling generally requires a few days, as it takes time for the animals to migrate from the substrates to the walls or slides. Agitation substrate sampling can yield placozoans at multiple locations in a single day, but the collected placozoans are frequently damaged as a result of the shaking process (Maruyama, 2004).
In a previous study, we identified placozoans at all six surveyed locations in Japan, suggesting a wide distribution of placozoans around Japan and in the Northern Pacific Ocean (Nakano, 2014). Here, we introduce a new processing protocol for substrate sampling referred to as "ethanol-treated substrate sampling," which can be performed at multiple locations in a single day and does not appear to damage placozoans. We succeeded in collecting placozoans using this protocol in three new sampling locations: Shikine-jima Island, Chichi-jima Island, and Haha-jima Island. We further determined the 16S rRNA sequences of the placozoans from eight locations within Japan, and we report five haplotypes that have not been previously described in the country. On Shikine-jima Island, we performed slide and ethanoltreated substrate sampling from 2015 to 2016, yielding multiple haplotypes, including haplotypes that were not found in other areas of Japan. Our results show that the haplotype distribution at a site varies with the seasons, and we propose that sampling efforts conducted during different seasons at multiple locations worldwide, together with analyses of molecular markers to reveal the genetic structure within haplotypes, are essential for revealing the distribution and dispersal history of placozoans.

| Sampling methods
In preliminary experiments, when an approximately 5% volume of 99.5% ethanol (Wako, Osaka, Japan) was added to seawater in glass     (1) Stones, shells, and ambient seawater are collected in containers; (2) 99.5% ethanol is added to each container to achieve an ethanol concentration of approximately 5%; (3) (Figures 3 and 4). The observation and collection of placozoans were carried out using a light stereomicroscope (SZX7; Olympus, Tokyo, Japan) and a micropipette (P200; Gilson, Villiers, France).

Permission to collect the substrates and seawater in the Ogasawara
National Park was obtained from the Ogasawara Ranger Office for Nature Conservation, Ministry of the Environment, Japan.

Ethanol-treated substrate sampling performed at three sites on
Shikine-jima Island ( Figure 3J) yielded 58 placozoans, and slide sampling at two sites resulted in 10 collected specimens (Figure 4). On Chichi-jima Island ( Figure 3K) and Haha-jima Island ( Figure 3L), four and seven specimens were isolated by ethanol-treated substrate sampling, respectively (Figures 5 and 6).

The mitochondrial 16S genotyping of placozoans collected in
Japan is summarized in Figures 2 and 3 and Table 1. H2 and H11 placozoans showed a wide distribution around Japan, with both haplotypes being found along the coasts of the Pacific Ocean and the Sea of Japan ( Figure 3). H11 placozoans have previously only been collected in Monterey Bay (California, USA) (Pearse & Voigt, 2007; nevertheless, placozoans of this haplotype were found at five locations around Japan. The haplotypes of the placozoans collected on Shikine-jima Island were H2, H9, H15, H17, and H19 ( Figure 4). As H17 and H19 placozoans have only been reported in Monterey Bay (California, USA) and Adelaide (Australia), respectively, the specimens collected on Shikine-jima Island represent the second report of these haplotypes worldwide. On the Bonin Islands (Chichi-jima and Haha-jima Islands; Figure 3K,L), H4 and H9 placozoans were isolated (Figures 5 and 6).
Our sampling efforts resulted in the collection of 79 placozoans from three previously unreported locations (Shikine-jima, Chichi-jima, and Haha-jima Islands), increasing the number of locations from which these animals have been collected to 12 in Japan and 79 worldwide.
In addition to the two haplotypes previously reported from Japan (H2 and H15) (Eitel & Schierwater, 2010;Miyazawa et al., 2012), five haplotypes (H4, H9, H11, H17, and H19) were newly collected in Japan in the present study. No new haplotypes that did not belong to the 19 known haplotypes were detected in our surveys.

| DISCUSSION
Six sampling efforts conducted using ethanol-treated substrate sampling on Shikine-jima Island yielded a total of 58 placozoans (Figure 4), showing that ethanol-treated substrate sampling is an efficient protocol for collecting placozoans from natural environments. However, 13 sampling efforts performed on the Bonin Islands yielded only 11 placozoans ( Figures 5 and 6). We are of the opinion that this disparity resulted not from the sampling method but from the biological nature of placozoans. The success of placozoan collection has been reported to depend largely on the weather and the microenvironment at collection sites (Nakano, 2014;Paknia & Schierwater, 2015;Pearse & Voigt, 2007). In the present study, the number of collected placozoans from site J-a on Shikine-jima Island fluctuated with the sampling date ( Figure 4). Therefore, we consider ethanol-treated substrate sampling to be an efficient collection protocol, and we expect that performing ethanol-treated substrate sampling on different dates and at different sites on the Bonin Islands will yield more placozoans.
As more placozoans are collected from slides suspended above the sea bottom than those placed on the sea bottom, it has been suggested that a pelagic phase of placozoans exists (Pearse & Voigt, 2007). For example, swarmers, which bud off from the dorsal surface of individuals, have been reported to float in seawater (Thiemann & Ruthmann, 1991). Embryos released by adults have also been suggested to be planktonic . Additionally, it is likely that adults that detach from the substrate are easily carried away by ocean currents. In Japan, the warm Kuroshio Current originates from east of the Philippines and reaches the Pacific coast of the country via Taiwan (Figure 3). In the Sea of Japan, the Tsushima Current, a branch of the Kuroshio Current, flows along the Japanese coast from the south through the Tsushima Strait (Figure 3). It has been reported that the Kuroshio and the Tsushima Currents impact the distribution of various animals, including sponges, clams, mantis shrimp, sunfish, and halfbeaks (Cheng & Sha, 2017;Hoshino, Saito, & Fujita, 2008;Yamada, Ishibashi, Toyoda, Kawamura, & Komaru, 2014;Yoshita et al., 2009;Yu, Kai, & Kim, 2016). H2 placozoans have been reported from Okinawa and Hong Kong (Eitel & Schierwater, 2010;Pearse & Voigt, 2007) and were collected on the coasts of both the Pacific and the Sea of Japan during this study (Figure 3). Therefore, the Kuroshio  (Eitel & Schierwater, 2010), in Shirahama (Honshu Island) (Miyazawa et al., 2012, and this study) and in Shikine-jima Island (this study), suggesting that the Kuroshio Current has an effect on the dispersal of this haplotype. Performing further collections on the coast of the Sea of Japan could reveal the effects of the Tsushima Current on its distribution.
H4 placozoans were shared between Shimoda and Chichi-jima Island, but only one H4 placozoan was collected during our continuous survey conducted in Shimoda from 2010 to 2015 (Table 1).
Furthermore, despite the wide distribution of H2, H11, and H15 placozoans along the Pacific coast of Japan, no H2, H11, or H15 placozoans were found on the Bonin Islands. Ethanol-treated substrate sampling was performed in a similar environment (shallow water along a rocky shore) on both the Shikine-jima and Bonin Islands, but haplotypes collected at Shikine-jima, such as H2, H15, H17, and H19, were not found on the Bonin Islands. These results suggest that successful dispersal of placozoans between the Honshu/Shikine-jima Islands and the Bonin Islands may be rare or absent. The Izu Islands stretch between Honshu Island and the Bonin Islands, and carrying out collections on these islands will illustrate the extent of both northward and southward dispersal and may clarify the reason for the lack of dispersal between Honshu Island and the Bonin Islands.
The 16S haplotypes of placozoans reported from the Northern Pacific Ocean, including Japan, are diverse (Figure 7), and it is difficult to estimate the dispersal routes of each haplotype based on its known  (Eitel et al., 2013).
As different haplotypes were found in the samplings performed in different months on Shimoda and Shikine-jima Island (Figure 4; There were some differences between the habitat types where the specimens of certain haplotypes were collected in this study and in previous studies. For example, H4 and H9 were sampled from an outdoor tank and a rocky shore, respectively, in this study and have not been previously reported at such sites (Table 1). H19, which was previously collected once on a stony beach (Eitel et al., 2013), was collected on a rocky shore in this study (Table 1). H11 and H17 have been reported only from tanks at the Monterey Bay Aquarium (California, USA) (Pearse & Voigt, 2007. In this study, H11 was collected on a boat dock, from an outdoor tank and on a rocky shore, and H17 was also collected from an outdoor tank and on a rocky shore ( Table 1), suggesting that H11 and H17 are present in various environments.
These results suggest that haplotypes that have previously only been reported in certain environments may be found in other environments with further sampling.
An effect of weather on the number of collected placozoan individuals has been previously suggested (Nakano, 2014;Pearse & Voigt, 2007. In the present study, multiple samplings during different seasons resulted not only in a change in the number of collected specimens, but a change in the collected 16S haplotypes, especially for those from Shimoda and Shikine-jima Island (Figure 4; Table 1). This observation suggests that the 16S haplotypes found at a certain site might also easily change, depending on the weather and other environmental conditions.
In previous studies, increasing the number of sampling locations has resulted in the discovery of new haplotypes, and it has been estimated that more than 200 haplotypes exist worldwide (Eitel & Schierwater, 2010;Eitel et al., 2013;Pearse & Voigt, 2007.
However, we failed to collect new haplotypes among our samples.
Our sampling was mostly performed in outdoor tanks and along rocky shores (Table 1), all at depths shallower than 50 cm. These conditions are similar to those of previous studies sampling placozoans in Japan.
Further sampling efforts in other environments where placozoans F I G U R E 7 16S haplotype distribution in the Pacific Ocean. Multiple sampling locations on the Honshu/Shikine-jima Islands and in Panama are shown with rectangles. The colors of the haplotypes are based on the color codes presented in Figure 2. Map source: Natural Earth (http:// www.naturalearthdata.com/) have not been previously reported in Japan, such as open ponds, mangroves, or algae beds (Eitel et al., 2013), or at a greater depth, may be needed to obtain new haplotypes from Japan.
A recent study revealed that population growth rates of placozoans are negatively affected by acidity stress (Schleicherová et al., 2016). In Mikama Bay, on Shikine-jima Island, where sites J-a and J-b are located (Figure 4), shallow CO 2 seeps result in reduced pH zones (Agostini et al., 2015). No placozoans were collected from J-b, whereas 26 were collected from J-a. Collection at J-a was performed in tide pools separated from the open sea at low tide, and the shallow CO 2 seeps are expected to have little impact on the pH of the tide pools; thus, it is assumed that the placozoans that entered the tide pool were able to proliferate. Moreover, slides placed at J-b were constantly exposed to a reduced pH, probably resulting in the failure of collection. The effect of low pH has been reported only in laboratory-cultured placozoans (Schleicherová et al., 2016).
Further studies at multiple sites exhibiting various pH levels within Mikama Bay would illustrate the effects of long-term exposure to a reduced pH on wild placozoans and the ecological differences among the 16S haplotypes.
Our results revealed high genetic diversity of placozoans around