https://orcid.org/0000-0002-2567-1789
Figures
Abstract
Mesophotic habitats could be sheltered from natural and anthropogenic disturbances and act as reproductive refuges, providing propagules to replenish shallower populations. Molecular markers can be used as proxies evaluating the connectivity and inferring population structure and larval dispersal. This study characterizes population structure as well as horizontal and vertical genetic connectivity of the broadcasting coral Pocillopora verrucosa from Ludao, a small oceanic island off the eastern coast of Taiwan. We genotyped 75 P. verrucosa specimens from three sites (Gongguan, Dabaisha, and Guiwan) at three depth ranges (Shallow: 7–15 m, Mid-depth: 23–30 m, and Deep: 38–45 m), spanning shallow to upper mesophotic coral reefs, with eight microsatellite markers. F-statistics showed a moderate differentiation (FST = 0.106, p<0.05) between two adjacent locations (Dabaisha 23–30 and Dabaisha 38–45 m), but no differentiation elsewhere, suggesting high levels of connectivity among sites and depths. STRUCTURE analysis showed no genetic clustering among sites or depths, indicating that all Pocillopora individuals could be drawn from a single panmictic population. Simulations of recent migration assigned 30 individuals (40%) to a different location from where they were collected. Among them, 1/3 were assigned to deeper locations, 1/3 to shallower populations and 1/3 were assigned to the right depth but a different site. These results suggest high levels of vertical and horizontal connectivity, which could enhance the recovery of P. verrucosa following disturbances around Ludao, a feature that agrees with demographic studies portraying this species as an opportunistic scleractinian.
Citation: de Palmas S, Soto D, Ho M-J, Denis V, Chen CA (2021) Strong horizontal and vertical connectivity in the coral Pocillopora verrucosa from Ludao, Taiwan, a small oceanic island. PLoS ONE 16(10): e0258181. https://doi.org/10.1371/journal.pone.0258181
Editor: Tzen-Yuh Chiang, National Cheng Kung University, TAIWAN
Received: June 12, 2021; Accepted: September 22, 2021; Published: October 11, 2021
Copyright: © 2021 de Palmas et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study was funded by the Taiwanese Ministry of Science and Technology to CAC (MOST 103-2621-B-001-004-MY3) and VD (MOST 104-2611-M002-020-MY2). SDP is currently the recipient of a Ministry of Science and Technology postdoctoral grant (MOST 110-2811-M-002-514). SDP and DS were both supported by the Taiwan International Graduate Program (BIODIV 23-14 to SDP and DS) and the Coral Reef Ecology and Evolutionary Genetics laboratory. The funders had no role in the study design.
Competing interests: The authors declare no conflicts of interest.
Introduction
Mesophotic Coral Ecosystems (MCEs) are deeper sections of shallow tropical and subtropical coral reefs [1]. Generally located from 30 m to as deep as 150 m [2], MCEs can be sheltered from warm-water bleaching, tropical storms, and anthropogenic disturbances [3–5]. In parallel, MCEs may act as a reproductive refuge for some reefal organisms, and supply recruits to shallow water reefs following major disturbances [6–8], thus maintaining species richness and community structure, and facilitating recovery after disturbances [9–11]. However, the role of MCEs as ecological refuges, formalized as the “Deep Reef Refuge Hypothesis” (DRRH [7, 8]), depends on the frequency and intensity of depth-dependent disturbances, the local ocean dynamics, and the taxon considered—specifically that taxon’s depth distribution range and reproductive traits [5, 12–15]. Particular emphasis is now given to identifying locations and taxa that, within the DRRH framework, may persist facing major disturbances [16–27] or not [28–32].
In most benthic marine species, recruitment and settlement of new individuals are achieved through pelagic larval movement [9]. While larval dispersal and the effective recruitment at shallow and mesophotic depths are critical to the reef dynamic [33], quantifying them remains a challenge [34] considering that benthic larvae have limited mobility and small recruits. Although biophysical models predict that such movements across depth significantly contribute to demography and resilience in reef coral populations [15], genetic connectivity approaches offer a valuable proxy for assessing population structure and estimating larval flux across depths [6, 12, 35–45]. To date, genetic approaches have revealed contrasting patterns of connectivity between shallow and deep populations of reef corals, suggesting that connectivity is taxon- and location-specific [1, 46]. However, despite a growing body of literature on this topic, research evaluating connectivity between shallow and mesophotic coral populations is still considered to be preliminary [1, 47].
In Taiwan, well-developed fringing reefs have been observed around Ludao, a small volcanic island 33 km off the coast of Taitung. Along the bathymetric gradient, Pocillopora verrucosa is one of the most dominant scleractinian hard corals, distributed from the near-surface to as deep as 55 m [48]. Despite variation in corallum micro and macro morphologies with depth [49], sequencing of the mitochondrial Open Reading Frame marker (mtORF, from a gene with unknown function) has confirmed this species’ distribution across a large bathymetric gradient in Ludao [50]. Several microsatellite markers (also called Simple Sequence Repeats, SSRs) have been developed for Pocillopora species [51–54] and were used to investigate species diversity and population connectivity across their biogeographic distribution [55–58]. However, to date little has been done to assess connectivity across depths, population structures, or larval fluxes between shallow and deep populations of Pocillopora verrucosa. The present study aims to make these assessments.
Materials and methods
Sampling and DNA purification
In 2016 and 2017, Pocillopora verrucosa specimens were collected at three sites around Ludao—Gongguan, (22.683530°N, 121.496631°E), Guiwan, (22.639212°N, 121.483529°E), and Dabaisha (22.637706°N, 121.490844°E)—using compressed air or Trimix diving, with a minimum distance of 3 m between samples to avoid collecting colonies that might be clones (Fig 1). A collection permit was obtained from the Taitung County Government (1040000285) and all specimens were registered at the Biodiversity Research Museum (Academia Sinica, Taiwan). Distances between sites are < 6 km (Fig 1).
a. Ludao; b. Details of the island with position of the sampling sites Gongguan, Guiwan, and Dabaisha. Source: Ocean Data Bank, National Taiwan University (http://www.odb.ntu.edu.tw/).
Eighty-five colonies originating from three depth ranges—Shallow (7–15 m, hereafter “S”), Mid-depth (23–30 m, hereafter “M”), and Deep (38–45 m, hereafter “D”)—were previously confirmed to be Pocillopora verrucosa by mtORF sequencing [50]. Nine locations were defined accordingly: Gongguan S, Gongguan M, Gongguan D, Guiwan S, Guiwan M, Guiwan D, Dabaisha S, Dabaisha M, and Dabaisha D. Locations are separated by vertical distances (i.e., depth) and horizontal distances (i.e., perpendicular to the depth axis) so that the frontal distances between locations from the same site are < 100 m.
DNA extraction process followed De Palmas et al. [50]. Additionally, DNA extracts were purified and concentrated using a sodium acetate (3M, pH 5) precipitation method. For each DNA sample, 1/10 volume of sodium acetate was added, and mixtures were gently inverted for 10 min. Three volumes of pure ethanol were added, and solutions were kept at -20°C overnight. Samples were centrifuged at 14,000 g for 30 minutes and DNA pellets were rinsed three times with -20°C ethanol. The DNA pellets were re-suspended in 30 μL TE buffer (1X, USB Corporation, Cleveland, OH, USA). DNA concentration and quality were measured on a Nanodrop 2000 (Thermo Fisher Scientific) and diluted aliquots were prepared to homogenize the DNA concentration of each sample before Polymerase Chain Reaction (PCR).
Microsatellite marker selection
Microsatellite marker effectiveness was determined by amplifying eight samples haphazardly picked with 30 microsatellite markers, accounting for most microsatellites developed for Pocillopora species. Each PCR mixture contained 15 μL of Master Mix RED (Ampliqon, Odense M, Denmark), 15 μL of ddH2O, 2 μL of each pair of primers (2.5 μM), and ~50 ng of DNA template. PCR conditions were as follows: 60 s of denaturation step at 94°C; 40 cycles of 60 s at 94°C, 60 s at the annealing temperature (Ta), and 60 s at 72°C; and an extension step at 72°C for 5 min. PCR amplifications were checked on a 2% agar gel to confirm amplification success rate and product size. Successful amplifications were diluted 22X before processing on a 5400 Fragment Analyzer™ Automated CE System (Agilent, CA, USA) using a dsDNA 905 reagent kit (1 to 500 bp). Any markers displaying low PCR amplification success or non-specific PCR amplification (multiple additional PCR products) were removed from subsequent analyses.
Microsatellite screening and allele scoring
Loci that passed the selection process were amplified to genotype the 85 P. verrucosa. Each PCR product was run three times on the fragment analyzer (as described above) to account for the accuracy of the fragment analyzer (i.e., 2–3 bp, per the manufacturer’s instructions). The three repeated electropherograms were scored using Prosize 3.0 (Agilent, CA, USA) and allele repeat size was then estimated using GenAlEx 6.5 [59] based on the three scored electropherograms. The estimation of repeat size and repeated allele scoring were used to solve allele scoring ambiguity. Micro-Checker 2.2.3 [60] was used to detect and correct eventual scoring inconsistencies and scoring errors due to stuttering and large allele drop-out. Samples showing low standard scoring quality were removed from the analysis.
Computational analysis
Arlequin 3.5.2.2 [61] was used to calculate the total number of alleles (Na), the observed (Ho) and expected (He) heterozygosities, and the exact probability of departure from Hardy-Weinberg equilibrium for each locus (pHWE) with 106 steps in a Markov chain and 105 dememorization steps. Arlequin was also used to investigate linkage disequilibrium between all pairs of loci with 5 x 105 permutations. Null allele frequency was inferred using MicroChecker 2.2.3 [60] with a 95% confidence interval and 105 randomizations.
Genetic differentiation among all locations was estimated using a pairwise FST comparison at all loci and their statistical significance with 5 x 105 permutations in Arlequin. Pairwise FST were also calculated using FreeNa software [62] with the number of replicates fixed at 50,000. Unlike Arlequin, FreeNa takes into consideration the presence of null alleles and provides an accurate estimation of FST values [62]. To account for the low number of samples per location, pairwise FST were also computed at the site level (pooling 3 depth ranges at one site together) and at the depth level (pooling similar depth ranges from all 3 sites together).
Genetic structuring was assessed using STRUCTURE 2.3.4 [63] with 106 Markov Chain Monte Carlo simulations, a burn-in period of 105 and three iterations for each K from k = 1 to k = 12. The LOCPRIOR option was used (to account for the location characteristics) under the admixture model and assuming independent allele frequency. The web-based program STRUCTURE HARVESTER 0.6.94 [64] was used to estimate the K number of populations and compare them to the web-based program CLUMPAK [65]. A principal coordinate analysis (PCoA), which summarizes genetic distances between individual multi-locus genotypes to cluster individuals/locations relative to each other, was performed at the location level using GenAlEx 6.5 [59].
In order to detect individuals with a recent migration history, we use the “Detection of first-generation migrants” option from Geneclass 2 [66] which assigns potential new migrant to their most likely source population [67]. This option calculates the likelihood Lhome/Lmax (with Lhome the likelihood computed from the population where the individual was sampled and Lmax the highest likelihood value among all population including the population where the individual was sampled) using the frequencies-based method from Paetkau et al. [68] and Monte-Carlo resampling algorithm with 10,000 simulated individuals and a type I error threshold α = 0.05 [67]. To account for the low number of samples per locations, this simulation was also computed at the site (pooling 3 depth ranges from one site together) and at the depth level (pooling similar depth range from all 3 sites together).
Results
Of 85 P. verrucosa specimens genotyped in this study, 75 produced satisfactory amplification and scoring results (≤ 4% of data missing) for eight microsatellite markers, Psp_16, Psp_23, Psp_29, Psp_32, Psp_35, Psp_39, Psp_41, and Psp_48 (Table 1). Full dataset with allele scoring, mtORF haplotype and museum numbers is provided in S1 Table.
Initial PCR amplification success was evaluated for a subset of eight samples. Loci displaying PCR amplification success >75% and expected fragments size for the initial subset (in bold) were used to genotype the 85 Pocillopora verrucosa or otherwise excluded from this study.
Loci were polymorphic, but not for all combinations of location and marker (48 combinations in total): Psp_29 was monomorphic at Dabaisha D and Gongguan S; Psp_35 was monomorphic at Dabaisha M, Gongguan S, and Gongguan M; and Psp_39 was monomorphic at Gongguan M (Table 2). The number of alleles per location and locus ranged from one to 19 (Table 2). Twelve combinations of location and loci deviated from Hardy-Weinberg equilibrium (Table 2), representing 16% of the total combinations. Linkage disequilibrium was found in Psp_29/Psp_39, and Psp_32/Psp_48. Null alleles were detected in Psp_29, Psp_35, Psp_39, Psp_41, and Psp_48. No clone was detected.
N: number of samples, n: number of specimens which produced data at each locus, Na: number of alleles, Ho: observed heterozygosity, He: expected heterozygosity, pHWE: p-value for exact tests of Hardy–Weinberg equilibrium (significant values are in bold, ns for non-significant,—indicates the presence of monomorphic loci).
Except for the Dabaisha M and Dabaisha D comparison, which displayed significant differentiation (FST = 0.106, p < 0.05, Fig 2), all FST values were low (0.000 < FST < 0.058) and not significant. Pairwise FST computed at the site and depth levels were also low (0.000 < FST < 0.007) and non-significant (S1 File). In addition, FST values that were corrected for the presence of null alleles were not significantly different from non-corrected ones. The PCoA did not reveal any site-specific or depth-related pattern (Fig 3).
FST (below) with their respective p-values (above). Significant values are in bold.
Neither STRUCTURE nor CLUMPAK revealed any population structuration (k = 1, Mean LnP(K) = - 1748.400, Prob (k = 1) = 0.688, (Table 3)).
Mean LnP (K), Ln’ (K), Ln” (K), and Delta K are given using the Evanno method (see [69] for details). Prob (K = k) is the probability for each K given by STRUCTURE. *computed in CLUMPAK.
Assignment tests yielded 40% of our specimens (30/75) as potential new migrants: 10 came from deeper locations, 10 came from shallower locations and the remainder relocated from similar depth but from a different site (Table 4). This result was consistent when specimens were grouped by site (10 potential migrants) and depths (10 potential migrants, including 3 detected as coming from shallower locations and 7 detected as coming from deeper locations; S1 File).
For each specimen (Assigned sample) from the location where it was sampled (Origin) the table gives the likelihood (-log (Lhome/Lmax) of this specimen to be a new migrant with its respective probability p < α and the likelihood of originating from another location (-log(L)). Shaded lines highlight potential new migrants their possible location of origin in bold. Number of loci (Nb. of loci) are the loci used for calculations (loci with missing data were not included into the calculation).
Discussion
Pocillopora verrucosa colonies originating from three sites and three depth ranges were genotyped at eight microsatellite loci to evaluate genetic differentiation and genetic structure, and to estimate migration flux between sites and across depths. Overall, P. verrucosa exhibits a low level of genetic differentiation, with the exception of Dabaisha M and Dabaisha D, which showed a mild genetic differentiation despite being separated by a short horizontal distance. Pairwise FST values generally support high genetic connectivity between sites and across depths (Fig 2 and S1 File), and the absence of clustering suggested that P. verrucosa specimens were collected from the same, well-mixed population unit (Table 3). We further identified possible migrants originating from sites/depths different from those where they were collected (Table 4 and S1 File), highlighting that MCEs could act as a source of propagules for nearby shallow water communities (and vice versa). Our data indicate that populations of P. verrucosa around the oceanic island of Ludao may approximate panmixia, thus we hypothesized that the mesophotic zone around this island could 1) act as an ecological refuge against disturbances for several other depth generalist species inhabiting this area and 2) benefit from the Kuroshio Current for dispersion of coral larvae and increase the overall connectivity along its path.
Low genetic differentiation and absence of genetic structure across depth in P. verrucosa
In this study, P. verrucosa from Ludao exhibited low genetic differentiation between sites located less than 6 km apart, and across depth zonations occurring at shorter horizontal distances. Only one pairwise comparison showed weak but significant differentiation (FST = 0.106), which corresponded to two adjacent locations at Dabaisha (M: 23–30 m and D: 38–45 m, Fig 3). Scleractinian coral populations may exhibit strong genetic breaks between shallow and mesophotic depths, even when separated by a short horizontal distance (< 1 km) [12, 39, 43, 70]. High levels of genetic differentiation could be linked to biotic factors such as selection against migrants (or genotypes) in their non-native environment [12], temporal differences in reproductive timing [71], and/or depth-dependent parental effects [72]. Alternatively, high levels of genetic differentiation could also be linked to abiotic factors such as contrasting reef geomorphology and water currents between shallow and mesophotic habitats that create a physical barrier to dispersal [39, 70]. In Ludao, there are evidences for shallow and mesophotic reefs to produce recruits in synchrony [73]. For Pocillopora, similar molecular operational taxonomic units (MOTU) are recruiting at both shallow and mesophotic depths [73]. In addition, few scleractinian corals from the mesophotic zone have been observed to have gametes, in synchrony with shallow water corals, affirming that these scleractinians are reproductively active (Fymbriaphyllia ancora: -45 m and Acropora spp.: -40 m, De Palmas S, personal observation). Despite the short horizontal distance separating Dabaisha M from Dabaisha D (< 100 m), both locations are well-known for their strong currents. The slight difference between these two adjacent locations could be due to diverging currents generating a physical barrier to P. verrucosa larval dispersal. Overall, low genetic differentiation and the absence of any genetic structures within and between sites further suggest that P. verrucosa specimens originated from a single population, and that this population was probably panmictic.
Previous investigations have found contrasting patterns of connectivity between shallow and mesophotic populations, varying from open to closed populations, with connectivity patterns occurring over thousands of kilometers [39, 43, 70, 74–76]. For instance, populations of the coral Montastraea cavernosa showed genetic differentiation with depth in Florida [39] and Little Cayman Island [41], but not in Bermuda or the US Virgin Islands [39, 41]. Similar findings were seen in Seriatopora hystrix from the Great Barrier Reef: the species displayed genetic differentiation with depth at Yonge Reef and Day Reef [36], but not at Scott Reef [40]. Interestingly, while it was proposed that broadcast spawners (i.e., species that release gametes into the water column and undergo external fertilization) may have higher dispersal abilities than brooders (i.e., species that release sperm and larvae into the water column and undergo internal fertilization) [36], low genetic differentiation was observed in the broadcasting species Stephanocoenia intersepta [12]. In addition, divergence by depth has also been reported in both brooders and spawners [36, 39–41, 43]. Pocillopora verrucosa is a broadcasting spawner, which spawns in the early morning [77], 2–3 days after the full moon in April and/or in May in Taiwan [78]. Mulla et al. [78] found larvae of P. verrucosa to display positive photo-movements, i.e., phototaxis or photokinesis. They further suggested that P. verrucosa larvae could disperse further and benefit from stronger surface currents as they move upward into the water column. This larval behaviour could explain the absence of genetic structure observed in the present study. In general, the ability of mesophotic populations to reseed shallow water reefs should be regarded as species- and location-specific [1] with local hydrology playing a key role in disseminating coral larvae [15]. In Ludao, mesophotic and shallow zonations have large scleractinian diversity overlap (37%, [79]), so that many other reef coral species with large bathymetric distribution may contribute to the reproductive refuge.
Possible new migrants and their origins
Effective ecological connectivity between shallow and deep zonations is achieved under three prerequisite criteria: 1) suitable habitats supporting robust populations, 2) habitats that are physically connected via water or propagule movements, and 3) organisms with life-history traits that enable successful replenishment of both their own populations and those connected to them [45]. In Ludao, Denis et al. [48] and De Palmas et al. [50] reported that P. verrucosa is one of the dominant species found across the depth gradient. Our data strongly suggest that larval movement creates connectivity between shallow and deep zonations. Our simulation test detected potential migrants between sites and across depths, which reinforces the notion that mesophotic populations can support themselves as well as adjacent populations. About 40% of our specimens were assigned to different populations from where they were collected (Table 3). In this study, loci and sample numbers were relatively low but our methodology produced a similar outcome when individuals were grouped by sites and depths (S1 File). Among the 30 individuals detected as potential migrant, 1/3 were assigned coming from deeper locations, 1/3 were assigned coming from shallower locations and the rest were assigned to similar depth range from different sites. This result suggests strong exchange of individual between sites and depths, showing that both vertical and horizontal larval exchange shape the populations of P. verrucosa at Ludao. Moreover, this is congruent with the low observed genetic differentiation and absence of population structure between sites and across depths, and suggests that P. verrucosa disperses effectively around Ludao. Assignment methods based on the analysis of genetic data should still be taken with caution due to the relative limitations induced by the presence of null alleles and the sensitivity of such methods to imperfect data (missing data and/or an unbalanced number of loci) [66]; however, concordant evidence suggests that P. verrucosa shows effective ecological connectivity around the island. Our study design includes adult colonies that are probably different ages, so the connectivity pattern is likely the result of multi-generational larval exchange, rather than associated with a specific P. verrucosa cohort. Population genetics comparing adult colonies to newly recruited P. verrucosa could tackle this issue and allow researchers to evaluate the current populations’ dynamics.
Deviation from Hardy-Weinberg, linkage disequilibrium, and null alleles
Among all the combinations of sites, depths, and loci, 12 (out of 96) deviated from Hardy-Weinberg equilibrium (Table 1). In addition, linkage disequilibrium—i.e., non-random association of alleles—was found in Psp_29/Psp_39, and Psp_32/Psp_48. These markers were not reported to have linkage disequilibrium in Pocillopora populations from the Ryukyus for which they were developed [53]. Linkage disequilibrium can be a result of evolutionary forces such as natural selection or genetic drift [80]. However, null alleles could also slightly affect linkage disequilibrium estimators [81]. Microsatellite null alleles occur when flanking regions have accumulated enough mutations to prevent primers from annealing during PCR amplification [62]; they could also be the result of preferential amplification of short alleles (i.e., Short Allele Dominance, SAD) [82]. Consequently, some alleles simply do not amplify during PCR, in which case individuals that are heterozygous will appear to be homozygous following fragment analysis [83]. In this study, five loci (Psp_29, Psp_35, Psp_39, Psp_41, and Psp_48) were detected as null alleles, but no SAD effect was found at these loci, suggesting that the flanking regions of those microsatellite loci could have accumulated enough variation to prevent marker annealing during PCR. As a deviation from Hardy-Weinberg equilibrium results from a shortage in heterozygotes (or excess in homozygotes), we hypothesize that the high number of potential null alleles found in this study could influence both linkage disequilibrium and deviation from Hardy-Weinberg equilibrium. Null alleles can also influence population differentiation estimators [68], so they require correction for the potential bias they induce, but in our case, no differences were found between biased and corrected FST. Null alleles are common in a large variety of taxa and, interestingly, are often found in taxa with a large population [62]. P. verrucosa is indeed locally abundant in Ludao, as it is one of the major components of the scleractinian community, residing from very shallow to at least 55 m deep [48, 50]. Population censuses for scleractinian corals inhabiting the mesophotic zone could reveal important elements for evaluating the dynamic of their populations and in fine the role of the mesophotic zone as an ecological refuge.
Cross-specific amplification for Pocillopora microsatellite markers
The microsatellite markers used here were developed for a variety of Pocillopora species (P. damicornis [54, 84], P. verrucosa, P. meandrina and P. damicornis [51, 52], P. acuta, and P. meandrina [53]), and are known to amplify closely related taxonomic groups [85–87]. However, our analyses found that only eight markers produced both satisfactory amplification and scoring results for P. verrucosa from Ludao, and several others did not provide any exploitable results (low amplification success, low specificity, Table 1). Cross-specific amplifications of microsatellites among Pocillopora species display contrasting results when compared to previous work. Nakajima et al. [53] confirmed that 14 previously described Simple Sequence Repeat (SSR) markers can amplify Pocillopora types 1 (P. meandrina and P. eydouxi), 3 (P. verrucosa), 4 (P. damicornis), 5 (P. acuta), and 8 (an undescribed Pocillopora species), while several other SSR markers could not (PV3, PV5, PV6, Pd3-002, Pd2-003, Pd3-010, and Pd13). None of these previously developed markers produced satisfactory results in P. verrucosa from Ludao. For instance, PV3, PV5, PV6, Pd3-002, Pd2-003, and Pd13 had low amplification rates (<50%) and Pd3-010 did not amplify at all (Table 1). Markers PV2, PV5, PV6, and PV7 successfully amplified P. verrucosa specimens from Southeast Africa [88], and most of these markers generally amplify Pocillopora species from the Western Indian Ocean, the Southwestern Pacific, Southeast Polynesia [55, 58, 89], and the Red Sea [90]. Among the 13 markers developed by Nakajima et al. [53] for populations from Japan and predicted to amplify most Pocillopora species, four (Psp_1, Psp_2, Psp_10, and Psp_18) showed no or poor amplification (<25%), and an additional marker (Psp_33) was found to have major additional products (Table 1). We hypothesize that the low amplification success of most SSR markers on P. verrucosa in Ludao, and differences in cross-specific amplification, could be the result of the accumulation of important mutations in the microsatellite flanking regions [91], which may reflect strong differentiation between populations from Taiwan and those for which these markers were developed and tested. Verifying this hypothesis would require sequencing both microsatellite regions and their flanking regions or applying single nucleotide polymorphisms (SNPs), such as used in Taninaka et al. [92], to reveal population difference of P. verrucosa across large biogeographic scale in the future.
Conclusions
Our results demonstrate that P. verrucosa populations around Ludao from 7 m to as deep as 45 m are close to panmictic. This species is more likely to recover from acute disturbances occurring in shallow waters (typhoons or warm water bleaching events) through the recruitment of larvae originating from adjacent shallow and deep habitats. We think that the reproductive status and reseeding capacity of mesophotic corals in this area require further investigations. Ludao, as an oceanic island could act as an important source of recruits along the Kuroshio Current for populations located on its path [93]. However, to the best of our knowledge, no other study addresses genetic connectivity or larval flux between shallow and deep coral populations in Taiwan [48]. In fact, deeper coral ecosystems around Taiwan are generally not the focus of conservation plans [48], and this study calls for greater consideration to be paid to these ecosystems, as they may constitute important reproductive refuge on the path of the Kuroshio current.
Supporting information
S1 Table. Raw microsatellite data at each locus.
Data are reported in allele size in bp and missing data are coded 0. Mitochondrial Open Reading Frame (mtORF) haplotype and museum numbers are indicated for each specimen.
https://doi.org/10.1371/journal.pone.0258181.s001
(DOCX)
S1 File. Pairwise FST comparisons and migration simulation tests at the site (pooling all depth ranges together) and depth (pooling similar depth range from different sites together) levels.
https://doi.org/10.1371/journal.pone.0258181.s002
(DOCX)
Acknowledgments
The authors would like to thank our dear friend Su Yeng, who dived with us to ensure our safety during diving. The authors would like to thank the reviewers and editor for their time and valuable comments to improve the overall quality of this manuscript.
References
- 1.
Pyle RL, Copus JM. Mesophotic Coral Ecosystems: Introduction and overview. In: Loya Y., Puglise K. A. & Bridge TCL, editor. Mesophotic Coral Ecosystems. Springer, Cham; 2019. pp. 3–27. https://doi.org/10.1007/978-3-319-92735-0_1
- 2. Hinderstein LM, Marr JCA, Martinez FA, Dowgiallo MJ, Puglise KA, Pyle RL, et al. Theme section on “Mesophotic Coral Ecosystems: Characterization, ecology, and management.” Coral Reefs. 2010;29: 247–251.
- 3. Glynn PW. Coral reef bleaching: Facts, hypotheses and implications. Global Change Biology. 1996;2: 495–509.
- 4. Riegl B, Piller WE. Possible refugia for reefs in times of environmental stress. International Journal of Earth Sciences. 2003;92: 520–531.
- 5. Lesser MP, Slattery M, Leichter JJ. Ecology of mesophotic coral reefs. Journal of Experimental Marine Biology and Ecology. 2009;375: 1–8.
- 6. Slattery M, Lesser MP, Brazeau D, Stokes MD, Leichter JJ. Connectivity and stability of mesophotic coral reefs. Journal of Experimental Marine Biology and Ecology. 2011;408: 32–41.
- 7.
Bongaerts P, Smith TB. Beyond the “Deep Reef Refuge” Hypothesis: A conceptual framework to characterize persistence at depth. In: Loya Y., Puglise K. A. & Bridge TCL, editor. Mesophotic Coral Ecosystems. Springer, Cham; 2019. pp. 881–895. https://doi.org/10.1007/978-3-319-92735-0_45
- 8. Bongaerts P, Ridgway T, Sampayo EM, Hoegh-Guldberg O. Assessing the ‘deep reef refugia’ hypothesis: focus on Caribbean reefs. Coral Reefs. 2010;29: 309–327.
- 9. Cowen RK, Sponaugle S. Larval Dispersal and Marine Population Connectivity. Annual Review of Marine Science. 2009;1: 443–466. pmid:21141044
- 10. McCook LJ, Almany GR, Berumen ML, Day JC, Green AL, Jones GP, et al. Management under uncertainty: Guide-lines for incorporating connectivity into the protection of coral reefs. Coral Reefs. 2009;28: 353–366.
- 11. Guest JR, Edmunds PJ, Gates RD, Kuffner IB, Andersson AJ, Barnes BB, et al. A framework for identifying and characterising coral reef “oases” against a backdrop of degradation. Journal of Applied Ecology. 2018;55: 2865–2875.
- 12. Bongaerts P, Riginos C, Brunner R, Englebert N, Smith SR, Hoegh-Guldberg O. Deep reefs are not universal refuges: Reseeding potential varies among coral species. Science Advances. 2017;3: e1602373. pmid:28246645
- 13. Semmler RF, Hoot WC, Reaka ML. Are mesophotic coral ecosystems distinct communities and can they serve as refugia for shallow reefs? Coral Reefs. 2017;36: 433–444.
- 14. Laverick JH, Piango S, Andradi-Brown DA, Exton DA, Bongaerts P, Bridge TCL, et al. To what extent do mesophotic coral ecosystems and shallow reefs share species of conservation interest? A systematic review. Environmental Evidence. 2018;7: 15.
- 15. Holstein DM, Paris CB, Vaz AC, Smith TB. Modeling vertical coral connectivity and mesophotic refugia. Coral Reefs. 2016;35: 23–37.
- 16. van Woesik R, Houk P, Isechal AL, Idechong JW, Victor S, Golbuu Y. Climate-change refugia in the sheltered bays of Palau: analogs of future reefs. Ecology and Evolution. 2012;2: 2474–2484. pmid:23145333
- 17. Chollett I, Mumby PJ. Reefs of last resort: Locating and assessing thermal refugia in the wider Caribbean. Biological Conservation. 2013;167: 179–186.
- 18. Fine M, Gildor H, Genin A. A coral reef refuge in the Red Sea. Global Change Biology. 2013;19: 3640–3647. pmid:23959950
- 19. Bridge TCL, Hoey AS, Campbell SJ, Muttaqin E, Rudi E, Fadli N, et al. Depth-dependent mortality of reef corals following a severe bleaching event: implications for thermal refuges and population recovery. F1000Research. 2013;2: 187. pmid:24627789
- 20. Smith TB, Glynn PW, Maté JL, Toth LT, Gyory J. A depth refugium from catastrophic coral bleaching prevents regional extinction. Ecology. 2014;95: 1663–1673. pmid:25039230
- 21. Smith TB, Gyory J, Brandt ME, Miller WJ, Jossart J, Nemeth RS. Caribbean mesophotic coral ecosystems are unlikely climate change refugia. Global change biology. 2016;22: 2756–2765. pmid:26648385
- 22. Holstein DM, Smith TB, Gyory J, Paris CB. Fertile fathoms: Deep reproductive refugia for threatened shallow corals. Scientific Reports. 2015;5: 12407. pmid:26196243
- 23. Gress E, Arroyo-Gerez MJ, Wright G, Andradi-Brown DA. Assessing mesophotic coral ecosystems inside and outside a Caribbean marine protected area. Royal Society open science. 2018;5: 180835. pmid:30473832
- 24. Keppel G, Kavousi J. Effective climate change refugia for coral reefs. Global Change Biology. 2015;21: 2829–2830. pmid:25846057
- 25. Baird AH, Madin JS, Álvarez-Noriega M, Fontoura L, Kerry JT, Kuo CY, et al. A decline in bleaching suggests that depth can provide a refuge from global warming in most coral taxa. Marine Ecology Progress Series. 2018;603: 257–264.
- 26. Slattery M, Moore S, Boye L, Whitney S, Woolsey A, Woolsey M. The Pulley Ridge deep reef is not a stable refugia through time. Coral Reefs. 2018;37: 391–396.
- 27. Kramer N, Eyal G, Tamir R, Loya Y. Upper mesophotic depths in the coral reefs of Eilat, Red Sea, offer suitable refuge grounds for coral settlement. Scientific Reports. 2019;9: 1–12. pmid:30626917
- 28. Frade PR, Bongaerts P, Englebert N, Rogers A, Gonzalez-Rivero M, Hoegh-Guldberg O. Deep reefs of the Great Barrier Reef offer limited thermal refuge during mass coral bleaching. Nature Communications. 2018;9: 3447. pmid:30181537
- 29. Morais J, Santos BA. Limited potential of deep reefs to serve as refuges for tropical Southwestern Atlantic corals. Ecosphere. 2018;9: e02281.
- 30. Rocha LA, Pinheiro HT, Shepherd B, Papastamatiou YP, Luiz OJ, Pyle RL, et al. Mesophotic coral ecosystems are threatened and ecologically distinct from shallow water reefs. Science. 2018;361: 281–284. pmid:30026226
- 31. Soares M de O, Tavares TCL, Carneiro PB de M. Mesophotic ecosystems: Distribution, impacts and conservation in the South Atlantic. Diversity and Distributions. 2018;25: 255–268.
- 32. Soares M de O. Marginal reef paradox: A possible refuge from environmental changes? Ocean and Coastal Management. 2020;185: 105063.
- 33.
Shlesinger T, Loya Y. Sexual reproduction of scleractinian corals in Mesophotic Coral Ecosystems vs. shallow reefs. In: Loya Y., Puglise K. A. & Bridge TCL, editor. Mesophotic Coral Ecosystems. Springer, Cham; 2019. pp. 653–666. https://doi.org/10.1007/978-3-319-92735-0_35
- 34. Cowen RK, Gawarkiewicz G, Pineda J, Thorrold SR, Werner FE. Population connectivity in marine systems: An overview. Oceanography. 2007;20: 14–21.
- 35. Eytan RI, Hayes M, Arbour-Reily P, Miller M, Hellberg ME. Nuclear sequences reveal mid-range isolation of an imperilled deep-water coral population. Molecular Ecology. 2009;18: 2375–2389. pmid:19457199
- 36. Bongaerts P, Riginos C, Ridgway T, Sampayo EM, van Oppen MJH, Englebert N, et al. Genetic divergence across habitats in the widespread coral Seriatopora hystrix and its associated Symbiodinium. PLoS ONE. 2010;5: e10871. pmid:20523735
- 37. Bongaerts P, Sampayo EM, Bridge TCL, Ridgway T, Vermeulen F, Englebert N, et al. Symbiodinium diversity in mesophotic coral communities on the Great Barrier Reef: A first assessment. Marine Ecology Progress Series. 2011;439: 117–126.
- 38. Bongaerts P, Frade PR, Ogier JJ, Hay KB, Van Bleijswijk J, Englebert N, et al. Sharing the slope: Depth partitioning of agariciid corals and associated Symbiodinium across shallow and mesophotic habitats (2–60 m) on a Caribbean reef. BMC Evolutionary Biology. 2013;13: 205. pmid:24059868
- 39. Serrano XM, Baums IB, O’Reilly K, Smith TB, Jones RJ, Shearer TL, et al. Geographic differences in vertical connectivity in the Caribbean coral Montastraea cavernosa despite high levels of horizontal connectivity at shallow depths. Molecular Ecology. 2014;23: 4226–4240. pmid:25039722
- 40. van Oppen MJH, Bongaerts P, Underwood JN, Peplow LM, Cooper TF. The role of deep reefs in shallow reef recovery: An assessment of vertical connectivity in a brooding coral from west and east Australia. Molecular Ecology. 2011;20: 1647–1660. pmid:21410573
- 41. Brazeau DA, Lesser MP, Slattery M. Genetic Structure in the Coral, Montastraea cavernosa: Assessing Genetic Differentiation among and within Mesophotic Reefs. PLoS ONE. 2013;8: e65845. pmid:23734263
- 42. Prada C, Hellberg ME. Long prereproductive selection and divergence by depth in a caribbean candelabrum coral. Proceedings of the National Academy of Sciences of the United States of America. 2013;110: 3961–3966. pmid:23359716
- 43. Serrano XM, Baums IB, Smith TB, Jones RJ, Shearer TL, Baker AC. Long distance dispersal and vertical gene flow in the Caribbean brooding coral Porites astreoides. Scientific Reports. 2016;6: 21619. pmid:26899614
- 44. Hammerman NM, Rivera-Vicens RE, Galaska MP, Weil E, Appledoorn RS, Alfaro M, et al. Population connectivity of the plating coral Agaricia lamarcki from southwest Puerto Rico. Coral Reefs. 2018;37: 183–191.
- 45.
Sponaugle S, Cowen RK. Coral Ecosystem Connectivity Between Pulley Ridge and the Florida Keys. In: Loya Y., Puglise K. A. & Bridge TCL, editor. Mesophotic Coral Ecosystems. Springer, Cham; 2019. pp. 897–907. https://doi.org/10.1007/978-3-319-92735-0_46
- 46.
Goodbody-Gringley G, Noyes T, Smith SR. Bermuda. In: Loya Y., Puglise K. A. & Bridge TCL, editor. Mesophotic Coral Ecosystems. Springer, Cham; 2019. pp. 31–45. https://doi.org/10.1007/978-3-319-92735-0_2
- 47. Kahng SE, Copus JM, Wagner D. Recent advances in the ecology of mesophotic coral ecosystems (MCEs). Current Opinion in Environmental Sustainability. 2014;7:72–81.
- 48.
Denis V, Soto D, De Palmas S, Lin YTV, Benayahu Y, Huang YM, et al. Taiwan. In: Loya Y., Puglise K. A. & Bridge TCL, editor. Mesophotic Coral Ecosystems. Springer, Cham; 2019. pp. 249–264. https://doi.org/10.1007/978-3-319-92735-0_14
- 49. Soto D, De Palmas S, Ho MJ, Denis V, Chen CA. Spatial variation in the morphological traits of Pocillopora verrucosa along a depth gradient in Taiwan. PLoS ONE. 2019;13: e0202586. pmid:30118513
- 50. De Palmas S, Soto D, Denis V, Ho MJ, Chen CA. Molecular assessment of Pocillopora verrucosa (Scleractinia; Pocilloporidae) distribution along a depth gradient in Ludao, Taiwan. PeerJ. 2018;6: e5797. pmid:30386700
- 51. Magalon H, Samadi S, Richard M, Adjeroud M, Veuille M. Development of coral and zooxanthella-specific microsatellites in three species of Pocillopora (Cnidaria, Scleractinia) from French Polynesia. Molecular Ecology Notes. 2004;4: 206–208.
- 52. Pinzón JH, LaJeunesse TC. Species delimitation of common reef corals in the genus Pocillopora using nucleotide sequence phylogenies, population genetics and symbiosis ecology. Molecular ecology. 2011;20: 311–25. pmid:21091563
- 53. Nakajima Y, Nishikawa A, Iguchi A, Nagata T, Uyeno D, Sakai K, et al. Elucidating the multiple genetic lineages and population genetic structure of the brooding coral Seriatopora (Scleractinia: Pocilloporidae) in the Ryukyu Archipelago. Coral Reefs. 2017;36: 415–426.
- 54. Starger CJ, Yeoh SSR, Dai C-F, Baker AC, Desalle R. Ten polymorphic STR loci in the cosmopolitan reef coral, Pocillopora damicornis. Molecular Ecology Resources. 2008;8: 619–621. pmid:21585850
- 55. Gélin P, Postaire B, Fauvelot C, Magalon H. Reevaluating species number, distribution and endemism of the coral genus Pocillopora Lamarck, 1816 using species delimitation methods and microsatellites. Molecular Phylogenetics and Evolution. 2017;109: 430–446. pmid:28219759
- 56. Poquita-Du R, Ng CSL, Loo J Bin, Afiq-Rosli L, Tay YC, Todd P, et al. New evidence shows that Pocillopora ‘damicornis-like’ corals in Singapore are actually Pocillopora acuta (Scleractinia: Pocilloporidae). Biodiversity Data Journal. 2017;5: e11407. pmid:28325983
- 57. Oury N, Gélin P, Magalon H. High connectivity within restricted distribution range in Pocillopora corals. Journal of Biogeography. 2021;48: 1679–1692.
- 58. Gélin P, Fauvelot C, Bigot L, Baly J, Magalon H. From population connectivity to the art of striping Russian dolls: the lessons from Pocillopora corals. Ecology and evolution. 2018;8: 1411–1426. pmid:29375807
- 59. Peakall R, Smouse PE. GENALEX 6: Genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes. 2006;6: 288–295.
- 60. Van Oosterhout C, Weetman D, Hutchinson WF. Estimation and adjustment of microsatellite null alleles in nonequilibrium populations. Molecular Ecology Notes. 2006;6: 255–256.
- 61. Excoffier L, Lischer HEL. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources. 2010;10: 564–567. pmid:21565059
- 62. Chapuis MP, Estoup A. Microsatellite null alleles and estimation of population differentiation. Molecular Biology and Evolution. 2007;24: 621–631. pmid:17150975
- 63. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155: 945–959. pmid:10835412
- 64. Earl DA, vonHoldt BM. STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources. 2012;4: 359–361.
- 65. Kopelman NM, Mayzel J, Jakobsson M, Rosenberg NA, Mayrose I. Clumpak: A program for identifying clustering modes and packaging population structure inferences across K. Molecular Ecology Resources. 2015;15: 1179–1191. pmid:25684545
- 66. Piry S, Alapetite A, Cornuet J-M, Paetkau D, Baudouin L, Estoup A. GENECLASS2: A Software for Genetic Assignment and First-Generation Migrant Detection. Journal of Heredity. 2004;95: 536–539. pmid:15475402
- 67. Paetkau D, Slade R, Burden M, Estoup A. Genetic assignment methods for the direct, real-time estimation of migration rate: A simulation-based exploration of accuracy and power. Molecular Ecology. 2004;13: 55–65. pmid:14653788
- 68. Paetkau D, Strobeck C. The molecular basis and evolutionary history of a microsatellite null allele in bears. Molecular Ecology. 1995;4: 519–520. pmid:8574449
- 69. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Molecular Ecology. 2005;14: 2611–2620. pmid:15969739
- 70. Eckert RJ, Studivan MS, Voss JD. Populations of the coral species Montastraea cavernosa on the Belize Barrier Reef lack vertical connectivity. Scientific Reports. 2019;9: 7200. pmid:31076586
- 71. Levitan DR, Fukami H, Jara J, Kline D, McGovern TM, McGhee KE, et al. Mechanisms of reproductive isolation among sympatric broadcast-spawning corals of the Montastraea annularis species complex. Evolution. 2004;58: 308–323. pmid:15068348
- 72. Shlesinger T, Loya Y. Depth-dependent parental effects create invisible barriers to coral dispersal. Communications Biology. 2021;4: 202. pmid:33589736
- 73. Soto D, De Palmas S, Denis V, Ho MJ, Chen CA. A molecular census of early-life stage scleractinians corals in shallow and mesophotic zones. Ecology and Evolution. 2021.
- 74. Garavelli L, Studivan MS, Voss JD, Kuba A, Figueiredo J, Chérubin LM. Assessment of mesophotic coral ecosystem connectivity for proposed expansion of a marine sanctuary in the Northwest Gulf of Mexico: Larval dynamics. Frontiers in Marine Science. 2018;5: 174.
- 75. Studivan MS, Voss JD. Assessment of mesophotic coral ecosystem connectivity for proposed expansion of a marine sanctuary in the Northwest Gulf of Mexico: Population genetics. Frontiers in Marine Science. 2018;5: 152.
- 76. Studivan MS, Voss JD. Population connectivity among shallow and mesophotic Montastraea cavernosa corals in the Gulf of Mexico identifies potential for refugia. Coral Reefs. 2018;37: 1183–1196.
- 77. Bouwmeester J, Coker DJ, Sinclair-Taylor TH, Berumen ML. Broadcast spawning of Pocillopora verrucosa across the eastern and western coast of the central Red Sea. Ecosphere. 2021;12: e03340.
- 78. Mulla AJ, Lin C-H, Takahashi Shunichi, Nozawa Y. Photo-movement of coral larvae influences vertical positioning in the ocean. Coral Reefs. 2021; 40: 1297–1306.
- 79. De Palmas S, Denis V, Derek Soto, Lin YV, Ho M-J, Chen CA. Scleractinian diversity in the upper mesophotic zone of Ludao (Taiwan): a museum collection with new records from Taiwanese waters. Marine Biodiversity. 2021.
- 80. Slatkin M. Linkage disequilibrium—Understanding the evolutionary past and mapping the medical future. Nature Reviews Genetics. 2008;9: 477–485. pmid:18427557
- 81. Waples RS. Null Alleles and FIS × FST Correlations. Journal of Heredity. 2018;109: 457–461. pmid:29554281
- 82. Wattier R, Engel CR, Saumitou-Laprade P, Valero M. Short allele dominance as a source of heterozygote deficiency at microsatellite loci: experimental evidence at the dinucleotide locus Gv1CT in Gracilaria gracilis (Rhodophyta). Molecular Ecology. 1998;7: 1569–1573.
- 83. Crooks L, Carlborg Ö, Marklund S, Johansson AM. Identification of null alleles and deletions from SNP genotypes for an intercross between domestic and wild chickens. G3: Genes, Genomes, Genetics. 2013;3: 1253–1260. pmid:23708300
- 84. Torda G, Schmidt-Roach S, Peplow LM, Lundgren P, van Oppen MJH. A rapid genetic assay for the identification of the most common Pocillopora damicornis genetic lineages on the Great Barrier Reef. PLoS ONE. 2013;8: e58447. pmid:23505507
- 85. Rico C, Rico I, Hewitt G. 470 million years of conservation of microsatellite loci among fish species. Proceedings of the Royal Society B: Biological Sciences. 1996;263: 549–557. pmid:8677258
- 86. Guichoux E, Lagache L, Wagner S, Chaumeil P, Léger P, Lepais O, et al. Current trends in microsatellite genotyping. Molecular ecology resources. 2011;11: 591–611. pmid:21565126
- 87. Barbará T, Palma-Silva C, Paggi GM, Bered F, Fay MF, Lexer C. Cross-species transfer of nuclear microsatellite markers: potential and limitations. Molecular Ecology. 2007;16: 3759–3767. pmid:17850543
- 88. Ridgway T, Riginos C, Davis J, Hoegh-Guldberg O. Genetic connectivity patterns of Pocillopora verrucosa in southern African Marine Protected Areas. Marine Ecology Progress Series. 2008;354: 161–168.
- 89. Chiazzari B, Magalon H, Gélin P, Macdonald A. Living on the edge: Assessing the diversity of South African Pocillopora on the margins of the Southwestern Indian Ocean. PLOS ONE. 2019;14: e0220477. pmid:31374077
- 90. Robitzch V, Banguera-Hinestroza E, Sawall Y, Al-Sofyani A, Voolstra CR. Absence of genetic differentiation in the coral Pocillopora verrucosa along environmental gradients of the Saudi Arabian Red Sea. Frontiers in Marine Science. 2015;2: 1–10.
- 91. Brohede J, Ellegren H. Microsatellite evolution: Polarity of substitutions within repeats and neutrality of flanking sequences. Proceedings of the Royal Society B: Biological Sciences. 1999;266: 825–833. pmid:10343406
- 92. Taninaka H, Maggioni D, Seveso D, Huang D, Townsend A, Richards ZT et al. Phylogeography of blue corals (genus Heliopora) across the Indo-West Pacific. Frontiers in Marine Science. 2021.
- 93. Chen CA, Shashank K. Taiwan as a connective stepping-stone in the Kuroshio traiangle and the conservation of coral ecosystems under the impacts of climate change. Kuroshio Science. 2009;3: 15–22.