For this study, we used morphologically similar coral colonies belonging to the P. verrucosa–P. meandrina group (hereafter referred to as P. verrucosa) without attempting to precisely identify the species used. P. verrucosa was chosen as the object of study because it is a fast-growing species commonly used in aquaculture (Combillet et al., 2022). Furthermore, Pocillopora spp. host the richest known fauna of obligate and facultative ectosymbionts, which have been well studied in several regions of the tropical Indo-Pacific (Patton, 1966; Patton, 1974; Coles, 1980; Edwards and Emberton, 1980; Tkachenko et al., 2022), including the area where our research was conducted (Marin et al., 2005; Marin and Spiridonov, 2007; Britayev and Mikheev, 2013). Moreover, there is a natural population of P. verrucosa where the coral nurseries are located, meaning that both transplanted and natural coral colonies experience similar hydrological conditions, enabling an adequate comparison of their ectosymbiotic faunas.

The experiment was conducted at ​​the Dam Bay research station of the Marine Branch of the Joint Russian-Vietnamese Tropical Research and Technological Center (Tropical Center), located on Tre Island, Nha Trang Bay ( Figure 1 ). Coral fragments were transplanted and reared, and then the reared colonies were collected and processed in accordance with the methodology developed earlier (Britayev et al., 2023). Distal branch fragments measuring 5–8 cm in length and 3–6 cm in diameter were used in the experiment ( Figure 2 ). These fragments were taken from large donor colonies cultivated in a local nursery. The fragments were relieved of any symbionts before planting, as detailed in the methodology below. In April 2012, the fragments were attached by divers to fixed triangular iron frames that had been installed on coral bioherms at a depth of 2–3 m ( Figure 3A ). The frames were elevated above the ground to a height of 40 cm. The bioherms were separated from each other by a sandy bottom. In order to eliminate the influence of other possible factors, natural colonies of P. verrucosa were sampled in the same localities (bioherms) during the same season and at the same depths as the nurseries with transplanted fragments ( Figure 3B ). This allowed us to compare the structure of symbiotic communities solely based on the location of the colonies (raised above or located on the substrate) and the age of the colonies.

In recent studies on coral symbiotic communities, divers carried out the identification and abundance assessment of symbiotic animals in situ without removing coral colonies (e.g., Counsell et al., 2018). Although this approach prevents the destruction of target coral colonies, it does not enable a complete inventory of symbionts and their precise identification. However, the traditional approach of removing coral heads from the sea (Abele and Patton, 1976; Austin et al., 1980) allows obtaining more accurate information on the species composition and abundance of symbionts but is harmful to the coral population and the reef in the broader sense. In our study, we used reared colonies, which allowed us to combine the advantages of both approaches: precise counting of symbionts and avoiding the destruction of colonies in the natural population.

For comparison of symbiotic communities, we used 41 reared colonies exposed for 12 months on the frames in the natural environment and 10 natural coral colonies of unknown age, both collected in spring 2013 (April and March respectively). To minimize the impact on the local coral population, the number of natural colonies was limited to 10 heads. To investigate whether the structure of infracommunities depends on the age (exposure) of colonies, an additional 49 colonies with a 6-month exposure, transplanted at the same time as the colonies of the first group in April 2012, were also utilized.

The sampling of P. verrucosa colonies was conducted through scuba diving. Each colony was covered with a fine mesh to prevent the escape of ectosymbionts. Subsequently, the colonies were carefully detached from the nursery or natural substrate, placed in individual zip-lock plastic bags, and promptly transported to a coastal laboratory. The volume between coral branches is a major factor influencing the number of animals inhabiting a coral colony. This volume exhibits a strong correlation with the total coral colony volume, which is calculated by multiplying the maximum diameter, perpendicular diameter, and height of the colony (Austin et al., 1980). Therefore, in this study, we measured these three parameters to determine the volume of the colonies, as they were considered the most suitable for processing numerous colonies.

Each colony was immersed in a clove oil emulsion (15 drops per liter of seawater) for 3 min and intensively washed to immobilize and remove symbionts. The resulting solution was filtered through a 1 × 1-mm mesh to collect the washed-off animals, excluding those that were smaller than 1 mm in length. The sample processing method we used does not allow us to obtain quantitative data on endosymbionts; therefore, we did not take them into account in this study. After being washed, the colonies were carefully inspected, and all found ectosymbionts were collected with tweezers. The collected animals were then photographed to capture their coloration and preserved in 70% or 100% alcohol for further examination. Finally, the treated coral colonies were planted in a coral nursery for subsequent use in the coral reef restoration program of the Tropical Center.

Each animal was identified to the lowest possible taxonomic level using specialized literature (Banner and Banner, 1982; Bruce, 1993; Bruce, 1998; Castro et al., 2004; Marin et al., 2005; Marin and Spiridonov, 2007). Small animals like polychaetes, amphipods, and copepods were not identified and excluded from the analysis. The species associated with the colonies of P. verrucosa were grouped into two main categories, “obligatory ectosymbionts” and “facultative ectosymbionts”, according to the existing literature (Coles, 1980; Black and Prince, 1983; Stella et al., 2011; Britayev and Mikheev, 2013). The latter group includes both well-known facultative ectosymbiont species and poorly studied species with uncertain status. Juvenile Trapezia spp. crabs, which could not be identified at the species level, were used only for comparing the density of main groups of symbiotic species. All identified material has been deposited at the Institute of Ecology and Evolution Russian Academy of Sciences (IPEE RAS), Moscow, Russia.

The diversity (total number of obligate and facultative species), species richness (median number of species per 1 dm³ of colony volume of infested hosts), density (median number of individuals per 1 dm³ of colony volume of infested hosts), and prevalence (ratio between number of infested and total number of hosts) were compared between communities associated with planted and natural coral colonies. Typically, when comparing symbiotic communities, species richness and average abundance are calculated per host (Bush et al., 1997). However, to adjust for the correlation between the number of associated species and their abundance with the volume of the coral colony (Abele and Patton, 1976; Austin et al., 1980), the number of species and individuals per 1 dm³ were calculated instead of per colony (Britayev and Mikheev, 2013). Therefore, we used a more accurate term “density” instead of “average abundance”. These terms were employed with minor changes following Bush et al. (1997). The assemblage formed by all ectosymbionts from different species harbored by the same host individual is known as an infracommunity, while all infracommunities from a target host population are considered as the component community (Holmes and Price, 1986).

Each infracommunity associated with a given coral colony was characterized by the number of obligate and facultative species and the number of their individuals. These data and colony volume were tested for normality, using the Shapiro–Wilk test (Shapiro et al., 1968), with variables considered normal when both W and P were greater than 0.05.

Since none of the variables met this criterion, the Kruskal–Wallis H test was employed to compare differences between natural and transplanted colonies, as well as between transplanted colonies with 6 and 12 months of exposure. The analyzed parameters had non-normal distribution, so the median was used as the central tendency descriptor instead of the mean.

All tests were performed using Statistica v. 12.0 software (TIBCO Inc., Palo Alto, CA, USA).

A total of 37 taxa, including 11 obligate and 26 facultative species, were found in both reared and natural colonies ( Table 1 ). Thirty species belonged to the crustacean order Decapoda, while other higher-order taxa, Polycladida, Sipuncula, Gastropoda, Ophiurida, and Osteichthyes, were represented by only one or two species each ( Table 1 ).

The reared colonies with 12-month exposure exhibited a higher total diversity, as well as a higher diversity of obligate and facultative species (25, 9, and 16 species), compared to the natural colonies (19, 8, and 11 species; Table 1 ). However, the species accumulation curve was higher for symbionts from natural colonies, indicating potentially higher diversity ( Figure 4 ). Only 11 species were shared between planted and natural colonies (seven obligate and four facultative), 15 species were found exclusively on reared colonies (three obligate and 12 facultative), and nine species were found only on natural ones (two obligate and seven facultative). The ratio of obligate to facultative species was somewhat higher in natural than reared colonies with 12-month exposure (0.73 and 0.56, respectively).

All coral colonies were found to be populated by obligate ectosymbionts, resulting in 100% prevalence, while the prevalence for facultative ones was lower at 80.0% and 53.7% for natural and reared colonies with 12-month exposure, respectively ( Figure 5A ).

The most common species in all samples were obligate ectosymbionts, the crab Trapezia septata, and the shrimp Alpheus lottini ( Table 1 ; Figures 5B , 6A, C ). Several obligate species, such as the shrimps Harpiliopsis spinigera ( Figure 6B ) and Synalpheus charon, the gastropod mollusk Coralliophila monodonta, and the facultative symbiont Calcinus gaimardii, which were common in natural colonies, were rare or absent in planted colonies with 12-month exposure. Conversely, one facultative symbiont, the shrimp Palaemonella rotumana ( Figures 5A , 6D ), was quite frequent in planted colonies but absent in natural ones.

The diversity and abundance increased with colony volume in both natural and reared colonies with 12-month exposure in all ecological groups of symbionts ( Figure S1 ).

Natural colonies exhibited significantly higher species richness than reared ones for all, obligate, and facultative species ( Table 2 ; Figures 7A–C ). These differences were significant in all cases (Mann–Whitney U test, p ≤ 0.0001, p ≤ 0.001, and p ≤ 0.05; Table S1 ).

Density in natural colonies was 2.7 times higher for all and obligate species and nearly three times higher for facultative ectosymbionts than in reared ones ( Table 2 ; Figures 7D–F ). These differences were also significant in all cases (p ≤ 0.0001, p ≤ 0.001, and p ≤ 0.05; Table S1 ).

The reared colonies with 12-month exposure exhibited a higher total diversity, as well as a higher diversity of facultative species (25 versus 16, and 16 versus 8 species, respectively). The number of obligate ectosymbionts was the same in both cases, with eight species ( Table 2 ). Although the diversity of ectosymbionts was close to saturation in both samples, the asymptote was substantially higher in 12-month exposure colonies, indicating an increase in biodiversity with exposure ( Figures 4 , S1 ). We identified 10 species that were shared between two samples, with seven obligate and three facultative. Fifteen species were found only in 12-month exposure colonies (two obligate and 13 facultative) and five species only in 6-month ones (one obligate and four facultative. The ratio of obligate to facultative species was higher in 6-month exposure colonies than in 12-month exposure colonies (1.0 and 0.56, respectively).

All planted coral colonies were inhabited by ectosymbionts and almost all by obligate ectosymbionts. However, the prevalence of facultative ectosymbionts was higher in 12-month exposure colonies than in 6-month exposure colonies (53.7% and 42.8%, respectively; Table 1 ; Figure 5A ).

The most common obligate species, the crab T. septata and the shrimp A. lottini, and the most common facultative ectosymbiont, the shrimp P. rotumana, were the same in both samples. However, the prevalence of T. septata and A. lottini was higher in 12-month exposure colonies, while the prevalence of facultative symbiont P. rotumana was lower compared to that in 6-month exposure colonies ( Table 1 ; Figure 5B ).

The diversity and abundance increase with the volume of colonies, and the regression lines are almost identical for all ecological groups of symbionts in colonies with 6- and 12-month exposure (except for facultative symbionts in colonies with 6-month exposure; Figure S1 ).

Species richness was lower in 12-month exposure colonies compared to colonies with 6-month exposure for all, obligate, and facultative ectosymbionts ( Table 2 ; Figures 8A–C) . These differences were significant for all species and obligate species, but not for facultative ectosymbionts ( Table S1 ).

Density was also lower in 12-month exposure colonies than in colonies with 6-month exposure for all, obligate, and facultative symbionts ( Table 2 ; Figures 8D–F ). These differences were significant for all species and obligate species, but not for facultative ectosymbionts ( Table S1 ).

The Indo-Pacific pocilloporid-associated communities are known for their high species diversity, ranging from 36 to 127 species depending on geographical location (Britayev et al., 2017). However, the majority of species forming coral-associated communities are non-specialized facultative ectosymbionts, while the diversity of obligate species is much lower and ranges just from nine to 19 species (Britayev et al., 2017). The overall diversity of Poсillopora-associated species in Dam Bay is relatively low, with only 36 species, which is comparable to the diversity in the Red Sea (36 species, Britayev et al., 2017). However, the diversity of obligate ectosymbionts, which is 11 species, is very close to the numbers recorded in symbiotic communities associated with pocilloporids in other regions of the tropical Indo-Pacific (8–15 species; Patton, 1966; Black and Prince, 1983; Stella et al., 2010). In fact, it is even higher than values reported for Southeastern Australia and the Pacific coast of Panama (Abele and Patton, 1976; Black and Prince, 1983).

It is well established in the literature that the diversity of coral ectosymbionts increases with sample size (Austin et al., 1980; our data, Figure 3 ) and the size of coral heads (Abele and Patton, 1976; Austin et al., 1980; Edwards and Emberton, 1980). However, the number of natural colonies in our study was relatively low, and we only used relatively small coral heads. Therefore, we can assume that the diversity we identified in this study is only a small fraction of the overall diversity of Pocillopora-associated fauna of Nha Trang Bay. Increasing the sample size of natural colonies and examining larger coral heads would undoubtedly lead to a higher diversity of both facultative and, to a lesser extent, obligate coral ectosymbionts. In this study, we documented rather rich fauna of invertebrates and fish (30 species, including 10 obligate and 20 facultative ectosymbionts), associated with colonies of P. verrucosa reared in an in situ coral nursery (both 12-month and 6-month exposure). In fact, this is one of the first data on the actual diversity of ectosymbiotic fauna associated with reared colonies. Earlier studies showed the presence of several ectosymbionts on reared corals (Shaish et al., 2010). Later, Wee et al. (2019) found 63 taxa of symbiont species on alive and dead colonies of three species of reared corals, including one species of Pocillopora, based on the limited number of sampled colonies. Our study, based on large sample sizes, confirms the existence of a rich and abundant fauna of obligate and facultative ectosymbionts on reared corals and significantly expands our knowledge of their diversity. Furthermore, these findings highlight the growing importance of coral farms in not only propagating corals but also conserving the diversity of animals closely associated with them in the face of global coral reef degradation.

The comparison of the diversity of communities associated with natural and reared coral colonies revealed a low level of similarity between them. Out of the 33 ectosymbiotic species, only 10 (most of them obligate symbionts) were shared between natural and nursery-reared colonies. Although the total number of species was higher in reared colonies, the position of the species accumulation curves indicates a higher potential diversity of the symbiont fauna in natural colonies. We also registered substantial differences in the structure of symbiotic communities between natural and reared coral colonies. Species richness and density indices were higher for natural colonies. Although all natural and reared colonies were populated with ectosymbionts, the prevalence of facultative symbionts and the dominant species of obligate symbionts were higher for natural colonies. These results raise the question of why such dissimilarities exist.

The environmental conditions for these two groups of coral colonies were similar. However, there were two main differences in the habitat properties of communities on reared and natural colonies that can affect their structure. First, there was a difference in location, with reared colonies being elevated above the substrate, while natural ones are attached to a hard substrate. Second, the exposure (age) of the colonies is unknown for natural colonies and may differ from that of the reared colonies at similar sizes.

(1) The location of colonies above the substrate has two main ecological consequences for ectosymbionts. Symbionts living on natural colonies attached to the substrate are only accessible to predators (mainly coral fish) from above. The space between the substrate and the coral head provides protection against predator attacks and serves as a safe zone for symbionts ( Figure 2B ). In contrast, ectosymbionts living in reared colonies lack this refuge, making them more vulnerable to predators ( Figure 2A ).

Migration of post-settled juvenile or adult ectosymbionts from one host individual to another is a fairly common phenomenon in marine ecosystems, playing an important role in shaping symbiotic communities (e.g., Wirtz and Diesel, 1983; Bell, 1984; Yanagisawa and Hamaishi, 1986; Thiel et al., 2003; Dgebuadze et al., 2012). Such migrations were also observed in the obligate symbiont of pocilloporid corals, the crab Trapezia bidentata (as Trapezia ferruginea, Castro, 1978). Obligatory ectosymbionts are poorly adapted to survive outside their hosts, making host-to-host migrations risky due to high levels of predation pressure (Castro, 1978). It can be assumed that the movement of ectosymbionts between colonies located in the water column is riskier than migration between colonies located on the substrate (Mekhova et al., 2015), which could reduce the recruitment of symbiont populations and adversely affect the species richness and density of symbiotic communities in aquaculture.

Both of these factors may contribute to a decrease in species richness, density, and prevalence of dominant species, especially of obligate ectosymbionts, in symbiotic communities in reared colonies.

Another potential predictor of species richness and abundance is the extent of partial coral colony mortality (Stella et al., 2010; Britayev et al., 2017). Although we lack information on the proportion of dead tissue, our subsequent observations of natural and reared colonies at the same site (Britayev et al., unpublished) suggest that natural colonies exhibit a higher incidence of partial mortality. This characteristic could lead to an increase in diversity, species richness, and abundance, particularly of facultative ectosymbionts, in natural colonies (Stella et al., 2010; Counsell et al., 2018). However, further research is needed to test this hypothesis.

(2) It is important to consider that the exposure (age) of natural colonies may be significantly longer than that of nursery-reared colonies of the same size. The size of P. verrucosa fragments used for planting roughly corresponded to, or even exceed, the size of 2-year-old colonies formed from settled larvae (Zakaria, 2004). According to this, we assume that the exposure (age) of natural colonies in our samples could be at least 3 years. Consequently, the succession of communities on natural colonies likely lasted at least three times longer compared to reared ones. This extended time frame could also account for the significant differences observed in the composition and structure between communities of natural and reared colonies.

The development of symbiotic communities in scleractinian corals is a poorly understood process (Rouzé et al., 2019) and is completely unknown in Pocillopora species. In order to investigate the effect of exposure on symbiotic communities, we compared reared colonies exposed for 6 and 12 months.

As expected, communities with different exposure times differed in ectosymbiont diversity and abundance, which increased with longer exposure and in accordance with the increase of the colony size ( Figure S1 ). Although almost all colonies in both exposure groups were populated with ectosymbionts, the prevalence of facultative symbionts and the dominant species of obligate symbionts was higher in the 12-month exposure colonies ( Table 2 ; Figure 5B ). The regression lines between the number of species and individuals of symbionts and the volume of colonies were almost identical in corals with different exposure ( Figure S1 ). These findings suggest that communities in 6- and 12-month-old colonies are not different communities but rather different stages of succession in the same community.

On the contrary, the species richness and density of symbionts in 6-month exposure colonies were significantly higher than in colonies exposed for 12 months ( Table 2 ). This finding may be attributed to the territorial behavior of coral ectosymbionts. Previous studies have shown that pairs of obligate shrimp A. lottini and crabs of Trapezia spp., which settle in the early stages of colony formation, continue to control the entire coral head as it grows, excluding individuals of both their own species (Patton, 1974; Abele and Patton, 1976) and other species (Preston, 1973). Therefore, the decrease in species richness and density of symbionts in 12-month exposure colonies can be attributed to the expected territorial defense by resident symbionts. As the coral grows, resident symbionts expand their territory, preventing the invasion of new individuals. However, to distinguish the contribution of each factor to the formation of symbiotic communities, further experimental studies with planted coral colonies are needed.