https://orcid.org/0000-0001-7004-7435
Figures
Abstract
Coral reefs are the most diverse ecosystem on the planet based on the abundance and diversity of phyla and higher taxa. However, it is still difficult to assess the diversity of lower taxa, especially at the species level. One tool for improving the identification of lower taxa are genetic markers that can distinguish cryptic species and assess species boundaries. Here, we present one such approach for an important and challenging group of reef-building corals. Porites corals are the main reef-builders of many coral reefs in the Indo-Pacific, owing to the massive growth forms of some species. The current number of valid Porites species is controversial, inflated with many synonymies, and often based on gross colony morphology although several morphospecies believed to be widespread and common can only be distinguished based on detailed microstructure analyses by taxonomic experts. Here, we test the suitability of multiple regions of mtDNA as genetic barcodes to identify suitable markers for species differentiation and unambiguous identification. Resulting sequencing data was further used for the first phylogenetic analysis of Guam’s Porites species. We tested eight different mitochondrial markers and analyzed four in detail for 135 Porites specimens: mtDNA markers were amplified for 67 Porites specimens from Guam, representing 12 nominal Porites species, and combined with 69 mitochondrial genomes, mostly from Hawaii. The combination of all 4 markers distinguished 10 common and 7 uncommon Central-West Pacific Porites species. Most clades separate species along taxonomic boundaries, which is uncommon for Porites corals and testifies to the suitability of our multi-marker approach, and a combination of the two most promising barcodes distinguished 8/10 common species. These barcodes are thus suitable to distinguish virtually cryptic species in one of the most important and challenging coral genera. They offer a cheap, fast and reliable way to identify Porites species for species-level research, monitoring and conservation.
Citation: Combosch DJ, Burdick D, Primov K, Rios D, Rios K, Fernandez J (2024) Barcoding and mitochondrial phylogenetics of Porites corals. PLoS ONE 19(2): e0290505. https://doi.org/10.1371/journal.pone.0290505
Editor: Michael Schubert, Laboratoire de Biologie du Développement de Villefranche-sur-Mer, FRANCE
Received: January 24, 2023; Accepted: August 10, 2023; Published: February 15, 2024
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All newly generated sequence data was submitted to GeneBank (https://www.ncbi.nlm.nih.gov) under the accession numbers OR509052 - OR509283 as detailed in S1 Table.
Funding: Funding for this research was provided by grants from NSF, NOAA and the National Park Service. Guam NSF EPSCoR directly supported this work through the National Science Foundation awards OIA-1457769 and OIA-1946352. NOAA supported this work by a Coral Reef Conservation Grant to DC (Grant #P16AC01681) and the National Park Service supported this work with Cooperative Agreement P21AC12027 with DC. Any opinions, findings, conclusions, or recommendations expressed in this contribution are those of the authors and do not necessarily reflect the views of any funding agency. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Corals are the main ecosystem engineers of coral reefs, the most diverse marine ecosystem in the world. Millions of people around the world rely on coral reefs for a significant portion of their diet, for coastline protection, and for the cultural significance and economic benefits reefs provide to local economies [1, 2]. Over the last several decades, coral reefs have been declining globally due to a variety of factors, including global climate change, coastal development and deteriorating water quality, and over-fishing [3–7]. This decline has led to a significant increase in coral research, addressing increasingly complex and sophisticated ecologic and evolutionary questions. Virtually all these studies, however, depend on a reliable taxonomic system to identify and classify the species in question. In addition, taxonomy is further essential for any targeted conservation and management approach [8, 9].
Reliable species identification has been a major issue for coral studies, termed “The Species Problem in Corals” [10], and presents a significant obstacle for research, e.g. [11, 12]. Traditionally, coral species identification relies on skeletal morphology [13]. On one hand, however, coral species are often distinguished using overly coarse, whole-colony characteristics, which are convenient for field studies but rarely suitable to reliably identify and distinguish actual species. Significant and widespread phenotypic plasticity has complicated species identification based on macro-morphology even further (e.g. [14]). On the other hand, sophisticated coral species descriptions are available but geographic variation and taxonomic confusion have led to an inflation of described coral species. For example, within this study subject coral genus, Porites Link 1807, over 500 species have been described but less than 10% of them are generally considered valid [11, 15] and e.g. the World Register of Marine Species currently lists only 68 valid species [16].
It was largely the application of molecular tools in population and phylogenetic studies that led to the identification and reassessment of actual species boundaries among corals [17]. Phylogenetic analyses revealed numerous discrepancies among morphology-based species descriptions, like cryptic divergences among numerous sympatric congeners (e.g. [18–24]). These discoveries have led to numerous and ongoing taxonomic revisions and/or significant readjustments of morphologic characteristics for species identification (e.g. [25–30].
For some taxa, however, morphologic differences between species are extremely subtle and micro-morphological characters may depend on sophisticated electron microscopy. A simple DNA-based barcoding test is often more convenient and reliable to assess species boundaries and identify cryptic species in many situations. Moreover, molecular phylogenies might reveal previously unrecognized species boundaries, which can then be used to subsequently identify and establish morphological characteristics, e.g. for field studies (e.g. [31]). Therefore, we set out to develop and test simple and straight-forward mitochondrial barcoding markers for the species identification in arguably the most important reef-building genera across the Indo-Pacific, Porites corals.
Mitochondrial markers have long been a favorite class for evolutionary geneticists. They tend to be rather straightforward to amplify and sequence due to their predominantly mono-allelic presence in eukaryotes [32–36]. Moreover, mitochondrial DNA (mtDNA) has been found to be highly variable among but also within species in most metazoan phyla [37], which led to numerous phylogenetic and population genetic studies based exclusively on mitochondrial sequence data (e.g. [38–46]).
However, mitochondrial markers are not without challenges. For example, they tend to be less variable in certain taxa, such as in many Anthozoan taxa, including many reef coral genera, which has been hypothesized to be due to purifying selection and/or enhanced mitochondrial DNA repair mechanisms [47–49]. Moreover, mitochondrial markers can give a clean but lopsided impression of the evolutionary history of specimens and taxa due to their matri-lineal inheritance. For example, a single introgressive hybridization event can lead to the sustained presence of distinct mitochondrial lineages within a species. Studies based exclusively on mitochondrial markers would then give the impression of an entirely different species even though generations of “inbreeding” have long purged most of the introgressed DNA in the nuclear genome. There is some evidence for mito-nuclear discordance in Porites corals [e.g. 24, 50] and this has recently been found to be widespread among coral in general [51] (see discussion as well).
Members of the Porites genus are major structural components in coral reefs [52] due to the massive skeleton structure some species build. Porites corals are widespread across most reef environments and frequently dominate brackish and murky habitats where few other corals thrive [53, 54]. Porites are also notable for their bleaching resistance, particularly when compared to other dominant coral genera like Acropora and Pocillopora [55–57]. However, Porites corals are notorious for the complexity of their species boundaries and are a prime example of “the species problem” [58]. Species can be massive, branching, encrusting, plating, columnar, and form micro atolls and corallites and several species have been found to grow in multiple colony shapes (e.g. [24, 54, 59, 60]. Moreover, differences in the arrangement and structure of corallites are traditionally used to distinguish between massive species in Porites are particularly irregular and highly varied [60, 61], making it difficult or impossible to differentiate them in the field and difficult even to taxonomic experts in the laboratory (pers. observ. and comm.).
Recently genome-wide sequencing markers have been employed to decipher the species boundaries and phylogenetic relationships among Porites corals in Hawaii [61] and the Red Sea [50]. However, the taxonomic scope and narrow geographic focus of these studies only provided significant taxonomic clarification for the Red Sea and around the Arabian Peninsula. Pacific Porites species continue to remain very challenging to distinguish, despite its enormous ecological role and resulting interest.
Due to these challenges, no unified comprehensive phylogenetic revision of this genus has yet been conducted. However, several studies have generated significant molecular data and we have combined a large amount of published data to identify suitable barcode markers. Moreover, we generated sequence data for eight mitochondrial markers for 67 Porites samples from Guam that were identified as 13 different morphospecies (S1 Table). We analyzed four of these markers in detail with a significantly expanded dataset, including 69 published mitochondrial genomes for Porites species and seven outgroup taxa, and suggest two mitochondrial markers for barcoding and Porites species identification in future studies.
2. Methods
2.1 Species, samples, sequences
In total, we analyzed 136 Porites samples, representing 18 nominal morphospecies and 4 undescribed species (S1 Table). An additional seven outgroup taxa were included for phylogenetic analyses. Only samples with high-quality sequence data for at least two out of the four analyzed markers were included. All analyzed samples are listed in S1 Table, including species name, collection and voucher identifiers, sampling locations, and GenBank accession numbers.
The final dataset consisted of 143 samples from three different sources: 1) 67 Porites samples, representing 12 nominal species and 1 undescribed species, were collected on Guam and processed and sequenced as described below. 2) 58 recently published Porites mitochondrial genomes, representing 5 nominal species and 3 undescribed species, were obtained from [61]). 3) 18 mitochondrial genomes were obtained from NCBI, including 11 Porites mtDNA genomes, representing 9 nominal species, and the following seven outgroup taxa: the poritiid Goniopora columna and the dendrophylliids Turbinaria peltata, Dendrophyllia arbuscula, D. cribrosa, Tubastrea tagusensis as well as two mitochondrial genomes of T. coccinea (downloaded in December 2019).
Samples from Guam were collected by the authors under a collection permit issued to the University of Guam Marine Lab. Small coral nubbins were collected using SCUBA or free diving and transported alive to the UOG Marine Lab. There, small pieces were preserved in 95% Ethanol and stored at -20°C until further processing. Additional sample material was bleached and preserved as skeleton voucher specimens. Tissue and voucher specimens as well as underwater photographs are available for most samples from Guam in the UOG Marine Lab Biorepository (https://specifyportal.uog.edu/) as indicated in S1 Table.
2.2 Molecular methods
Genomic DNA was extracted using GenCatch Genomic DNA Extraction Kits from Epoch Life Science (Sugar Land, TX, USA). Mitochondrial markers were PCR-amplified using primers listed in S2 Table. PCR reactions (25 μl) contained 1 ul of template DNA, 0.75 μl of each primer (10 mM), 0.75 μl of dNTP’s (10 mM; Kapa Biosystems, Wilmington, MA, USA), 0.5 μl of HIFI DNA Polymerase (Kapa Biosystems), 5 μl of 5x PCR buffer (Kapa Biosystems) and molecular grade DI water up to 25 μl. PCR conditions consisted of an initial denaturation step at 94°C for 2min, followed by 30 cycles of 30s at 94°C, 30s at 56–63°C, and 60s at 72°C, and a final 5 min extension step at 72°C. PCR products were visualized on 1% agarose gels. Successful PCR amplifications were sent off to Epoch Life Science (Missouri City, TX, USA) for sequencing in both directions. Sequence data were visualized and edited in Geneious Prime 2020.1.2 (Biomatters Limited, Auckland, New Zealand) to generate consensus sequence per sample per locus. New sequences were deposited on GenBank under the accession codes listed in S1 Table. Additional sequences were obtained from [61]) and the NCBI GeneBank, as described above and listed in S1 Table.
2.3 Phylogenetic analyses
All analyses were conducted on alignments generated with MAFFT 7.450 [62] with default parameters as implemented in Geneious. Basic sequence statistics and alignment characteristics were assessed in Geneious. Nucleotide diversity among Porites samples per marker and across markers was calculated with MEGA X [63].
For phylogenetic analyses, nucleotide positions of ambiguous homology were removed using GBlocks 0.91b [64]. Each genetic locus was processed individually using default parameters, except for gap positions that were allowed if present in less than half of all samples. All four loci were then concatenated into one alignment using Geneious. Pre- and post-GBlocks alignment lengths for each locus and the concatenated alignment are listed in Table 1.
Phylogenetic analyses were conducted using Bayesian inference and maximum likelihood. IQ-TREE 2.0.5 [65, 66] was used for maximum likelihood analyses including model selection [67] for each individual locus, for all four loci combined and for the proposed barcoding combination MT12-MT20 [68]. For each individual locus, the best model according to the Bayesian Information Criterion was GTR+F+I+G4 but in the partitioned analyses with all four loci, the best model was K3Pu+F+G4 (BIC score: 19,977; LnL: -9034, df:233). Node support was assessed with 10,000 ultrafast bootstrap replications [69], 10,000 SH approximate likelihood ratio test replicates and 10,000 approximate Bayes tests [70].
In addition, Maximum likelihood analyses of individual loci and the concatenated 4-loci alignment were conducted using RAxML 8.2.8 on XSEDE [71] as implemented on the CIPRES web portal [72]. A unique GTR model of sequence evolution was specified for each partition, i.e. each locus, with corrections for a discrete gamma distribution (G) for site-rate heterogeneity (GTRGAMMA). Nodal support was estimated via 1000 rapid bootstraps [73].
Bayesian inference analyses were carried out with MrBayes 3.2 [74]. Analyses were conducted with three different models of sequence evolution: The default model (nset = 0), the GTR model (nset = 6) selected by IQ-TREE and the HKY model (nset = 2) suggested by jModelTest [75] under the Bayesian information criterion (BIC). Gamma corrections and a proportion of invariable sites were included in all three models. MrBayes analyses started with random trees, default priors and two runs, each with 4 Markov chains. Convergence diagnostics were analyzed using the sump command as implemented in MrBayes and Tracer 1.7 [76]. The runs were allowed to proceed until the average deviation of split frequencies reached <0.01 (~1.5 M generations). Phylogenetic trees were summarized as a 50% majority-rule consensus tree with a burn-in of 25% removed using the sumt command as implemented in MrBayes.
2.4 Morphospecies determination
Species determinations for the Guam specimens were made using a combination of sources, including the original species descriptions, Veron and Pichon [77], Veron et al. [78], and unpublished species descriptions provided by the late Richard Randall. Determinations were primarily based on a qualitative assessment of colony and corallite morphology with corallite diameter measurements used to assist in assigning determinations for several of the Guam specimens.
The use of "cf." for most of the species determinations indicates uncertainty in species identification based on qualitative morphological assessment. This uncertainty is a result of slight differences between the morphology of Guam specimens and that reported in the original descriptions, the limited utility of some of the original descriptions, inconsistencies between authors, the lack of high-resolution images for type material, the lack of available genetic data for type material (or from topotypes), challenges inherent in assessing members of the genus Porites (e.g., the reliance of traditional morphology-based taxonomy on relatively subtle, qualitative differences in corallite characters; the sometimes high degree of intracolonial variability in corallite morphology; the lack of comprehensive morpho-molecular studies that identify morphological characters that reliably align with the molecular data), or a combination of all of these factors.
Even those determinations for which “cf.” was not used (e.g., P. rus, P. lutea, P. cylindrica) should be considered as at least somewhat tentative. For example, while the colony and corallite morphology of the Guam P. lutea specimens were highly similar to that described by Milne Edwards & Haime (1851) and several subsequent authors, and formed a clade with a specimen from Hainan, China, for which the whole mitogenome was published by Niu et al. [79], our study did not include P. lutea material from the type locality (Fiji), and we did not examine the voucher material associated with the published P. lutea mitogenome. Similarly, our study did not include P. cylindrica or P. rus material from the type localities. The species names provided below are meant to provide readers with the names of one or more described species to which we believe the specimen most closely compares. The use of the qualifier “-like” in the clade names is meant to convey the doubt regarding the application of species names to individual specimens within a clade, which was sometimes compounded by the inclusion of two or more morphospecies within the same clade.
3. Results & discussion
3.1 Basic results
The complete dataset for this study contained 143 samples, including 136 Porites specimens and 7 outgroup taxa. New sequence data for 67 Porites samples from Guam was generated and combined with previously published sequences for 69 specimens, mostly from Hawaii (S1 Table). For each specimen, sequencing data for 2–4 mitochondrial markers was analyzed with an average coverage of 3.8 markers/sample. For each marker, on average 128 Porites samples were analyzed on average.
Markers consisted of 800–1,100 nucleotides for a complete alignment of 3,695 nucleotides across all four markers (Table 1). This final alignment contained 104 parsimony informative sites across the 136 Porites specimens (PIS, 2.8%), which were fairly evenly distributed across all four markers (Table 1), and 3468 invariable sites (94%). The absolute nucleotide diversity was 11.6 base differences per sequence and the relative nucleotide diversity was 0.0029 base substitutions per site. Both types of nucleotide diversity varied little among markers, with slightly more diversity in MT16 and MT20 compared to MT09 and MT12 (Table 1).
Alignment curation with G-blocks removed 57 ambiguous nucleotide positions across all 4 markers (1.5%), predominantly in markers MT09 (n = 35) and MT12 (20), for a final sequence alignment of 3638 bp (Table 1). ModelTest with IQ-TREE identified GTR as the most suitable model for all individual markers but a K3Pu model across all 4 markers. In contrast, the stand-alone version of modelTest suggested HKY for individual markers and a TVM model across all 4 markers.
3.2 Phylogenetic results
All three phylogenetic algorithms generated nearly identical results with the same overall topology and similar node support. The only notable difference was the presence of Clade 9 in both Maximum Likelihood (ML) analyses (Fig 1 & S1 Fig), which was absent from the Bayesian approximation (BA) tree generated with MrBayes (S2 Fig).
This tree is based on the K3Pu+F+G4 model of sequence evolution, as identified by IQ-TREE after merging all four loci into one partition to increase model fit (BIC score: 19,958). RAxML and MrBayes analyses resulted in a similar tree with identical overall topology and identical clade composition, with the exception of clade 9 (S1 & S2 Figs; see methods and results for further details). Numbers on nodes indicate ultrafast bootstrap support percentages based on 10,000 replications, 10,000 SH approximate likelihood ratio test replicates and 10,000 approximate Bayes tests as calculated with IQ-TREE. Bold numbers in clades indicate node support statistics for each clade.
The ML analysis with IQ-TREE identified a final tree (Fig 1) with a log-likelihood of “-9023”. This tree is based on the K3Pu+F+G4 model of sequence evolution, as identified by IQ-TREE after merging all four loci into one partition to increase model fit (BIC score: 19,958). The ML analysis with RAxML based on a GTR model of sequence evolution for each locus partition generated a virtually identical phylogenetic tree with a log-likelihood of “-9006” (S1 Fig). Bayesian analyses with MrBayes using different models of sequence evolution (HKY, F81 and GTR as proposed by ModelTest and IQ-TREE) for all marker partitions resulted in identical tree topologies with very similar node support values. Trees obtained from MrBayes all had a similar overall topology, compared to the Maximum Likelihood trees, with the mentioned exception of combining clades 7 and 9 (S2 Fig)—despite a consistent nucleotide difference in marker MT20, which was available for 9 out of 11 Clade 9 samples and 30/34 Clade 7 samples.
The monophyly of the genus Porites was supported in all analyses with perfect node support, i.e. >99% bootstrap support (BS) and >0.99 Bayesian probability (BP). Many nominal morphospecies were recovered as monophyletic clades with moderate to high node support, for example Porites rus (Forskål, 1775), P. cf. lutea Porites lutea Milne Edwards & Haime, 1851, P. cf. lichen (Dana, 1846) and P. cylindrica Dana, 1846. Several other monophyletic clades contained more than one traditional species, which might be due to a lack of resolution among mitochondrial markers or due to phenotypic plasticity or hybridization as discussed in detail below.
The most basal split among Porites species separates a group of species from the Caribbean (P. porites (Pallas, 1766)), Tropical Eastern Pacific (TEP, P. panamensis Verrill, 1866, P. sverdrupi Durham, 1947), and Red Sea (P. fontanesii Benzoni & Stefani, 2012) from the remaining Central West Pacific (CWP) Porites species. This basal split has been observed before [60] and is confirmed here with strong node support. Within this basal group, P. fontanesii is sister to the remaining samples and particularly distinct compared to all other samples (Table 2), which matches its geographic isolation in this dataset. In the Caribbean and Tropical Eastern Pacific clade, the three specimens of P. panamensis (n = 2) and P. sverdrupi (1) from the Tropical Eastern Pacific clustered together to form Clade 11. This clade has been recognized by Forsman et al [60] as well, with identical species composition and is termed clade XI there (Table 3).
The last column indicates the number of species names of all morphospecies identified within each clade.
The remaining, predominantly Indo-West Pacific Porites samples fall into two separate, well-supported and clearly distinct clades. The first clade contains two subclades:
Clade 1 is well-supported (node support 93/87/100 in IQ-TREE, RAxML and MrBayes) and contains all 5 samples identified as P. annae Crossland, 1952 from Guam, the only P. cf. deformis Nemenzo, 1955 specimen, and all 6 samples identified as P. evermanni Vaughan, 1907 from Hawaii and Guam. This clade is defined by clade-specific mutations in three of the four analyzed markers (MT09, MT12 and MT20, Table 3). Clade 1 corresponds to Clade II in Forsman et al [60], where it contained 6/6 P. annae from Samoa and 1/3 P. annae from Hawaii, all P. evermanni from Hawaii and several unidentified Porites sp. from Hawaii and Panama. These species encompass significantly different and distinct morphologies, from massive P. evermanni to branching/columnar P. annae and P. cf. deformis. Using scanning electron microscopy, Forsman et al [60] noted in this clade, the corallite “walls, denticles, palli, and columella tended to have similar height, with more evenly spaced and deeper interstitial holes” than in other clades. This clade was also recovered in multi-marker phylogenetic analyses by Hellberg et al [24]. A recent phylogenomic study of massive Porites on Guam (Primov et al, in prep) recovered this clade as well and termed it clade I. The clade was absent in recent Porites studies from the Arabian Peninsula [50, 80], indicating that it may be a predominantly Pacific clade, where it is widespread, from the Tropical Eastern Pacific, across the Central, West and South Pacific.
Clade 2 is mostly well supported (85/75/91) and contains all 14 samples identified as P. rus from Guam (12 specimen), Hawaii (1) and an unknown location (from NCBI, 1). This clade is distinguished by clade-specific mutations in three out of four markers as well (MT09, MT12 and MT20, Table 3). Our Clade 2 corresponds to Clade III in Forsman et al ([60]) and Clade IV in [50, 80], where it contained P. rus from Hawaii, Tahiti and the Arabian Peninsula as well as the morphologically similar P. monticulosa Dana, 1846 from Hawaii and the Arabian Peninsula. Despite this enormous geographic range, this clade is among the most morphologically uniform and phylogenetically distinct clades of Porites in recent studies. The monophyly and species composition of this clade was further confirmed by Hellberg et al [24].
The second major Indo-West Pacific clade is more complex, starting with an unresolved polytomy (with IQ-TREE, RAxML and MrBayes) between the only P. harrisoni Veron, 2000 sample in this study from a mitochondrial reference genome (NC037435) originally collected in Saudi Arabia with seven private mutations across all 4 markers (Table 3), as well as three distinct, generally well-supported clades:
Clade 3 (95/94/100) contains a P. lutea mitochondrial reference genome (KU159432) from Hainan Island, China, as well as three P. cf. lutea-like samples from Guam. It is distinguished by clade-specific mutations in all four markers (Table 3). Our Clade 3 most likely corresponds to Clade V in Forsman et al. [60], which contains most P. cf. lutea as well as some P. lobata Dana, 1846 from Samoa and has a very similar phylogenetic position. It would then also correspond to Clade VIII in Terraneo et al [50], where it was represented by colonies identified as P. somaliensis Gravier, 1910 from Djibouti, Yemen and Saudi Arabia. Interestingly, Veron [16] states that the corallites of P. somaliensis are in fact most similar to P. lutea, which further testifies to the correspondence of these clades. Comparisons with Hellberg et al [24] are difficult here since the few P. lutea specimens were spread rather widely across their phylogenies.
Another well-supported clade (94/85/100) contains two subclades, termed Clade 4 and Clade 5 here. Clade 4 (96/100/100) consists of two morphologically similar, encrusting Porites specimens: a potentially undescribed species from Guam (P. cf. vaughani Crossland, 1952; BP13, UOG IZC #14) and a specimen identified as P. superfusa Gardiner, 1898 by Forsman et al [60] from Palmyra atoll (NCBI: SAMN06648869). It is distinguished by 6 clade-specific mutations across all four markers (Table 3). This clade was not previously reported by Forsman et al. [60] or Terraneo et al [50]. While the Palmyra specimen was identified as P. superfusa, the relatively large mammiform surface swellings and superficial calices visible in the small in situ image of the specimen in Forsman et al. [81] are more consistent with the undescribed species from the Marianas than with Gardiner’s P. superfusa description from Tuvalu. However, an examination of the skeletal material for both specimens, and ideally genetic data from the type localities would be required for proper comparison. Clade 5 (99/100/100) contains all three P. cf. lichen samples from Guam and is distinguished by ten clade-specific mutations across all four markers (Table 3). This clade corresponds to Clade VIII in Forsman et al. [60], which contains all seven P. lichen present in that study. It was not recorded by Terraneo et al [50] around the Arabian Peninsula.
The next clade (89/94/100) contains five nested subclades, which were not distinguished by Forsman et al. [60] or Terraneo et al. [50], where this undivided clade is termed Clade I and Clade V, respectively. Here, these subclades are distinguished due to shared unique mutations across enormous geographic distances, moderate to high node support and distinct, clade-specific morphologies (Fig 1, Table 2), which merit their consideration as distinct evolutionary units. Moreover, a recent ezRAD study by Forsman et al [61] in Hawaii distinguished 3 of these clades (Clades 6, 7 and 8) while the other two are likely absent in Hawaii and around the Arabian Peninsula.
The most basal split separates a clade (77/77/98) containing a well-supported subclade, termed Clade 6, as well as the only P. cf. horizontalata Hoffmeister, 1925 specimen from Guam. The single Porites cf. horizontalata colony presented in this study was composed of encrusting laminae and irregular branches, with small corallites similar to that of P. rus. The sample contains individual and/or species-specific mutations in 3 out of the 4 tested mitochondrial markers (MT09, 12 & 16; Table 3). It is interesting to note that the P. cf. horizontalata specimen here is only distantly related to P. rus, despite exhibiting a similar colony and corallite morphological characters. Clade 6 (90/84/100) is similarly distinguished by four clade-specific mutations across 3 markers (MT12, MT16 & MT20; Table 3). It contains a P. lobata specimen from Costa Rica, three P. cf. lobata from Guam, and four P. lobata and one Porites sp. from Hawaii. Samples in this clade shared four clade-specific mutations between Costa Rica and Guam, i.e. over 14,000 km of open ocean. This clade is termed Clade A in Forsman et al [61] where it contains mostly P. lobata.
The clade on the other side of the basal split (87/94/94) contained a group of “basal” specimens that were not defined by mutations in any of the four analyzed mitochondrial markers. Instead, it is defined by the absence of additional differentiation compared to Clades 8–10. This group is comprised of 8 P. cf. murrayensis Vaughan, 1918 from Guam as well as 19 P. compressa Dana, 1846 and 7 P. cf. lobata from Hawaii. Although in our phylogenetic analyses these specimens are paraphyletic and do not form a distinct monophyletic clade, this group is termed “Clade” 7 here since it was recovered as a distinct mitochondrial Clade B in Forsman et al [61] and as Clade V in Primov et al (in prep). Moreover, specimens on Guam are morphologically distinct from other clades/species (but closest to P. lobata) and were identified as P. cf. murrayensis based on their relatively small corallites (usually <1.2 mm in diameter), thick walls, and prominent palli associated with the lateral septal pairs. Interestingly, although the Hawaii P. compressa and P. cf. lobata specimens share identical mitochondrial genomes, genome-wide ezRAD analyses were able to recover them as distinct [61]. This clade thus seems to contain distinct evolutionary units, which are resolvable with mtDNA alone—at least on Hawaii.
In addition, three samples were present at this polytomy that belong to three distinct morphospecies and contain unique individual and thus potentially species-specific mutations: a single P. okinawensis Veron, 1990 specimen with one private mutation in MT09, a single P. cf. stephensoni Crossland, 1952 specimen from Guam with two private mutations in MT16, and a single “Honolulu-harbour” Porites specimen (sensu [82]) with a unique mutation in MT20. The “Honolulu-harbour” specimen was found to be significantly distinct in genome-wide ezRAD analyses [61], which supports highlighting these moderately-supported single-sample species. Since only one sample of each species was available, it is not possible to distinguish individual variation from species-level distinctiveness and other samples in this study also showed elevated levels of unique variation (e.g. MZ05_Plobata).
In addition, 3 distinct subclades were identified:
Clade 8 (85/50/63) contains 8 massive P. cf. australiensis Vaughan, 1918 from Guam as well as 21 branching P. compressa and 1 massive P. cf. lobata from Hawaii. This clade is supported by a single nucleotide change in MT12 but was also identified as a distinct clade in Forsman et al ([61] as Clade C) and Primov et al (in prep.; as Clade VII). Interestingly, while the gross colony and corallite morphology of the Guam specimens in this clade are quite similar to that of the Guam specimens in Clade 9, all of the Clade 8 specimens exhibited a brown or greenish-brown coloration, while all Clade 9 specimens were purple or purplish-brown (see e.g. Fig 3).
Clade 9 (88/16/-) contains 11 massive P. cf. australiensis, which were all sampled from two nearby populations on the southwest coast of Guam. It is characterized by a single mutation in MT20 but was not recovered in the MrBayes analyses, independent of the selected model of sequence evolution (HKY or GTR). It was recovered as Clade VI in Primov et al (in prep.) but it is unclear if it occurs outside of Guam/the Marianas. Its morphology is similar to the closely related Clade 8, but showed a different coloration, as mentioned above.
And finally, Clade 10 (84/50/67) contains all 7 P. cylindrica specimens from Guam. This species is branching and resembles P. compressa from Hawaii in overall colony morphology. It is genetically distinct by a single mutation in MT09. It was not included in Primov et al (in prep) or any other Porites phylogenetic studies to date, as far as we know.
3.3 Clades, species & barcodes
In total, 11 Porites clades were identified and are distinguished here based on a combination of node support (Table 2), morphospecies identity and comparisons with genome-wide sequence datasets [50, 61]. All 11 clades were supported by both ML analyses (Fig 1 & S1 Fig) and only one of them, Clade 9-P.australiensis-like, was not recovered in the MrBayes analyses—despite a unique mutation in marker mt20 (Fig 1, S1 & S2 Figs). Moreover, many of these clades correspond to clades identified by [24, 50, 60, 61] and Primov et al (in prep), using mitochondrial, nuclear and genomic DNA sequence markers (Table 3)–as detailed above. In addition, individual specimens of seven distinct morphospecies are highlighted here to report the potential ability of these markers to distinguish them in future studies. Additional specimens will need to be analyzed to assess if their distinctiveness is indicative for the entire species or merely represent individual variation of the sampled specimen. In general, the proposed barcode markers will all need to be tested in more species and locations to verify their general suitability.
Of the 11 multi-sample Porites clades that are distinguished here, 7 are consistently supported by all or least most of the individual mitochondrial markers (Table 2). In turn, each of the four mitochondrial barcodes was able to distinguish at least 5 well-represented clades and 10 potential species (Table 2). A combination of the two most promising markers, MT12 and MT20, distinguished 9 well-represented clades and 12 potential species overall (Fig 2). We therefore propose this combination as a suitable, low-cost barcoding approach to distinguish Porites species. For example, on Guam, all 11 well-represented clades can be identified with these two barcodes since P. cylindrica can be readily distinguished morphologically. In addition, this barcode combination identifies 14 of the 18 potential species and increases reliable clade identification compared to each individual marker by increasing the number of informative SNP for most clades and species (Table 2). RFLP approaches to distinguish between these clades without sequencing might be possible but were not designed or tested yet.
This tree is based on the HKY+F+I model of sequence evolution, as identified by IQ-TREE after merging both loci into one partition to increase model fit. Numbers on nodes indicate ultrafast bootstrap resampling percentages based on 10,000 replications, 10,000 SH approximate likelihood ratio test replicates and 10,000 approximate Bayes tests.
Although there was an overall good congruence between our phylogenetic results and the samples taxonomic classification based on morphology, there were some discrepancies as well—as has been observed previously for Porites corals (e.g. [24, 60, 61, 81]). For once, at least three clades contain samples with distinct morphologies, i.e. more than one morphospecies (Clades 1, 7 & 8). For 2 out of these 3 clades, this is due to previous taxonomic classifications [61], which could not be verified personally. In contrast, P. annae and P. evermanni were both collected and examined on Guam. Although their colony morphology is fundamentally different (Fig 3), their micromorphological corallite structures are similar and they are genetically not clearly distinct based on the 4 mitochondrial markers analyzed here. This is most likely due to incomplete lineage sorting at the species level, which is surprising for a mitochondrial marker that should sort faster than nuclear loci. Interestingly, Forsman et al [60] found both species to be indistinguishable at the nuclear ITS2 marker as well, which indicates widespread incomplete lineage sorting, i.e. including the nuclear genome, at least in Hawaii. Since both species are widespread throughout the Indo-Pacific, recent speciation seems unlikely but their presumably large effective population sizes could slow down and drag out complete lineage sorting for a very long time. The smaller populations in semi-isolated locations like Guam and Hawaii should sort faster and there is some indication for that on the trees (Fig 1, S1 & S2 Figs), which does show two subclades, one dominated by P. annae and the other dominated by P. evermanni. Lineage sorting thus seems to be happening but has not been completed yet.
On the other hand, three morphospecies appeared in more than one clade, namely P. compressa (Clades 7 & 8), P. cf. australiensis (Clades 8 & 9) and Porites cf. lobata, which appeared in three distinct clades (6, 7 & 8). In the case of P. cf. lobata, all specimens from Guam (3) and Costa Rica (1) as well as 4 of the 12 P. lobata from Hawaii were found in Clade 6. In addition, 7 P. lobata from Hawaii were found in clade 7 and one in clade 8. Since all three clades are closely related and Porites lobata has an Indo-Pacific wide distribution and is abundant in many places, incomplete lineage sorting might be responsible for the ambiguous genetic identification. However, since only Hawaiian specimens occur in more than one clade and Hawaii is more isolated and presumably less diverse than Guam, incomplete lineage sorting is less likely there and introgressive hybridization could be responsible for observed ambiguity. In fact, geographically limited introgressive hybridization has been reported involving one of these Porites species, P. lobata, in the even more isolated Tropical Eastern Pacific [24]. Ideally, the Hawaii specimen in question should therefore be re-examined carefully, focusing on characters like the corallite diameter and palli, which we found suitable to distinguish P. cf. lobata from P. cf. murrayensis (clade 7) and P. cf. australiensis (clade 8).
In contrast, P. compressa specimens were almost equally split between closely related Clades 7 (19) and Clade 8 (21). This is surprising since Porites compressa is a local endemic in Hawaii, which makes further cryptic species less likely and should facilitates lineage sorting due to presumably smaller population sizes (although it can be abundant locally, e.g. in Kaneohe Bay; [16]). Moreover, Hellberg et al [24] found significant differences across several nuclear loci among three P. compressa samples and Forsman et al [61] found genome-wide differences between those two clades using RAD-Seq data (see Table 3), which confirms their distinctiveness despite close morphological similarities. Primov et al (in prep) further detected genome-wide differences between Guam specimens of P. cf. murrayensis and P. cf. australiensis, which represent these same two clades.
Similarly, P. australiensis is also fairly evenly split between Clade 8 (8) and Clade 9 (11). All P. australiensis specimens were collected on Guam and despite repeated micromorphological examinations, no significant differences between the two groups of P. cf. australiensis could be detected. Interestingly, one of these two clades was limited to the south-west corner of Guam while the other one was more wide-spread (Primov et al, in prep). The two clades with P. cf. australiensis are closely related but both have clade-specific mutations in mt12 (Clade 8) and mt20 (Clade 9). Porites australiensis is common and widespread across the Indo-West Pacific [16] so incomplete lineage sorting is more likely for this species. It would be surprising, however, that two distinct lineages within the same species are present on a small and remote Pacific island like Guam. Moreover, a recent RAD-Seq study found genome-wide differences between P. cf. australiensis specimens from Guam, representing Clades 8 and 9 (Primov et al., in prep). Morphological stasis in two widespread and distinct species thus seems to be the most likely explanation for the congruence of morphologies in Clades 8 and 9.
Mitochondrial markers only provide a limited and somewhat particular view of genetic differences between species and specimens. It is therefore entirely possible that the proposed markers cannot fully resolve all actual Porites species boundaries. Compared to previous mitochondrial barcodes for Porites, most major clades were also recognized using the putative control regions in Forsman et al [60] and Terraneo et al [83]. In a broadly comparable dataset from the Arabian peninsula [83], the putative control region had a slightly higher number of parsimoniously informative sites (40 vs 25–28) in a longer sequencing product (1287bp). However, three of the four tested markers (MT09, MT12 and MT20) provide some resolution for Clades 6–10, which were lumped in previous studies, based e.g. on this control region marker [60, 83]. Since different species and geographic regions were analyzed in different studies, the observed increased resolution here could also be due to the absence of some of these clades in other studies. Ultimately, only a comparison with the same samples would provide a definite answer and no mitochondrial marker was or will be able to reliably distinguish P. compressa and P. cf. lobata on Hawaii, where they shared 100% identical mitochondrial genomes [61].
4. Conclusion
This study provides another perspective of the species’ diversity and phylogeny in the major reef-building coral genus Porites. Like previous studies (e.g. [24, 50, 60, 61, 81], we found some disagreements between morphology-based species designations and genetically derived clades. However, unlike other studies, we found significantly more congruence between genetic clades and micro-morphology based taxonomy, in line with more recent phylogenomic studies of Porites [50, 61]. While this study did not include a quantitative analysis of morphological characters, our results suggest that micromorphology will play an important role in resolving species boundaries in Porites, as it did in several other scleractinian genera [21, 23, 80, 84–88]. Our results also attest to the suitability of “reverse taxonomy” approaches [31] in identifying suitable characters to distinguish between genetically-identified evolutionary species. Lastly, our results support the use of carefully-tested mitochondrial sequence data to distinguish among scleractinian species, including Porites, despite conflicting results in other studies. One reason might be that we used multiple mtDNA markers since no single marker was able to resolve the complex, mostly massive Porites species (i.e. Clades 6–10). It is also important to note that our dataset was geographically and taxonomically limited and not all clades were properly vetted using genome-wide sequence data (since no comprehensive genome-wide dataset is currently available for Porites corals). Mito-nuclear discordances have been observed among Porites corals before and were recently documented to be widespread among Scleractinian corals [51]. However, Porites corals were among the least affected and only one mismatch was documented for P. cf. australiensis, which is split here between clades 8 and 9. More instances of discordance will likely be discovered over time due to incomplete lineage sorting or introgression, but this study shows that mitochondrial sequencer markers are nonetheless very useful for taxonomic assessments of Porites species.
The generation of genome-wide sequence data for phylogenetic studies is an important next step that has recently been applied to Porites corals successfully around the Arabian Peninsula [50], Hawaii [57] and Guam (Primov, in prep). Since different genome subsampling techniques were used to generate genome-wide dataset for these different studies, it will be challenging to combine them in one comprehensive analysis. However, we are working with multiple groups to address these phylogenetic questions on a larger scale with more data and better phylogenetic resolution and are active looking for additional collaborators.
Quantitative morphological analyses are further required to advance our understanding of Porites species diversity and identification. The results of such analyses would complement phylogenetic studies like this one to arrive at sets of morphological characters used to reliably delineate molecular clades, as proposed by reverse taxonomic approaches. It is also necessary to designate representative individuals for each genetic clade to facilitate comparisons among allopatric clades and across studies. Our Table 3 is another such attempt to connect with similar studies in other geographic regions and facilitate comparisons and connections. In addition, we are generating vouchers samples, permanently stored at the University of Guam Biorepository (https://specifyportal.uog.edu), including high-resolution images of micromorphological characters, DNA sequence data (as well as primers, S2 Table, and PCR protocols etc.) for the mitochondrial barcodes proposed here and links to genome-wide datasets as we generate them. Finally, we agree with Bonito et al. [89] that an attempt to collect topotypes for all nominal species should be made, and the associated morphological and molecular data be shared, in order to facilitate the proper application of Porites nomenclature to the clades that emerge from integrative morpho-molecular analyses.
Acknowledgments
We would like to thank Zac Forsman for very helpful advice and support as well as access to pre-publication data. We would also like to thank Sarah Lemer, Hector Torrado, Mikayla Reuter, Joe Proietti and Zaine Benavente for their help in the lab work and helpful discussions.
References
- 1. Cesar HSJ, Burke L, Pet-Soede L. The economics of worldwide coral reef degradation. 2003. Available: http://eprints.eriub.org/48/
- 2. Wilkinson C. Status of Coral Reefs of the World: 2008. 2014.
- 3. Hughes T, Baird A, Bellwood D, Card M, Connolly SR, Folke C, et al. Climate change, human impacts, and the resilience of coral reefs. Science. 2003;301: 929. pmid:12920289
- 4. Hughes TP, Kerry JT, Álvarez-Noriega M, Álvarez-Romero JG, Anderson KD, Baird AH, et al. Global warming and recurrent mass bleaching of corals. Nature. 2017;543: 373–377. pmid:28300113
- 5. Hoegh-Guldberg O. Climate change, coral bleaching and the future of the world’s coral reefs. Mar Freshwater Res. 1999;50: 839–866.
- 6. Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS, Greenfield P, Gomez E, et al. Coral Reefs Under Rapid Climate Change and Ocean Acidification. Science. 2007;318: 1737–1742. pmid:18079392
- 7. Carpenter KE, Abrar M, Aeby G, Aronson RB, Banks S, Bruckner A, et al. One-third of reef-building corals face elevated extinction risk from climate change and local impacts. Science. 2008;321: 560–563. pmid:18653892
- 8. Green DM. Designable Units for Status Assessment of Endangered Species. Conserv. Biol. 2005;19: 1813–1820.
- 9. Fallon SM. Genetic Data and the Listing of Species Under the U.S. Endangered Species Act. Conserv. Biol. 2007;21: 1186–1195. pmid:17883484
- 10. Bernard HM. The Species Problem in Corals. Nature. 1902;65: 560–560.
- 11. Stat M, Baker AC, Bourne DG, Correa AMS, Forsman Z, Huggett MJ, et al. Molecular Delineation of Species in the Coral Holobiont. Advances in Marine Biology. pmid:22877610
- 12. Veron J. The species concept in “Scleractinia of Eastern Australia.” Proceedings of the 4th International Coral Reef Symposium. 1981;2: 183–186.
- 13. Veron J. E. N., Pichon M. Scleractinia of Eastern Australia. Part 5: Family Poritidae. Australian Institute of Marine Science, 1982.
- 14. Todd PA. Morphological plasticity in scleractinian corals. Biological Reviews. 2008;83: 315–337. pmid:18979594
- 15. Veron JEN, Stafford-Smith M. Corals of the World. Townsville MC Qld Australia: Australian Institute of Marine Science; 2000.
- 16. WoRMS Editorial Board (2023). World Register of Marine Species. Available from https://www.marinespecies.org at VLIZ. Accessed 2023-07-14.
- 17. Fukami H, Budd AF, Levitan D, Jara J, Kersanach R, Knowlton N. Geographic differences in species boundaries among members of the Montastraea annularis complex based on molecular and morphological markers. Evolution. 2004;58: 324–337. https://doi.org/10.1111/j.0014-3820.2004.tb01648.x
- 18. Arrigoni R, Benzoni F, Terraneo TI, Caragnano A, Berumen ML. Recent origin and semi-permeable species boundaries in the scleractinian coral genus Stylophora from the Red Sea. Scientific Reports. 2016; 1–13. pmid:27713475
- 19. Arrigoni R, Berumen ML, Stolarski J, Terraneo TI, Benzoni F. Uncovering hidden coral diversity: a new cryptic lobophylliid scleractinian from the Indian Ocean. Cladistics. 2018;16: 108–28. pmid:34633682
- 20. Bongaerts P, Cooke IR, Ying H, Wels D, den SH, Hernandez-Agreda A, et al. Morphological stasis masks ecologically divergent coral species on tropical reefs. Current Biology. 2020;9: 25–24.
- 21. Keshavmurthy S, Yang S-Y, Alamaru A, Chuang Y-Y, Pichon M, Obura D, et al. DNA barcoding reveals the coral “laboratory-rat”, Stylophora pistillata encompasses multiple identities. Scientific Reports. 2013;3: 1–7. pmid:23519209
- 22. McFadden CS, Haverkort-Yeh R, Reynolds AM, Halàsz A, Quattrini AM, Forsman ZH, et al. Species boundaries in the absence of morphological, ecological or geographical differentiation in the Red Sea octocoral genus Ovabunda (Alcyonacea: Xeniidae). Mol Phylogenet Evol. 2017;112: 174–184. pmid:28467886
- 23. Stefani F, Benzoni F, Yang S-Y, Pichon M, Galli P, Chen CA. Comparison of morphological and genetic analyses reveals cryptic divergence and morphological plasticity in Stylophora (Cnidaria, Scleractinia). Coral Reefs. 2011; 1–17.
- 24. Hellberg ME, Prada C, Tan MH, Forsman ZH, Baums IB. Getting a grip at the edge: recolonization and introgression in eastern Pacific Porites corals. Journal of Biogeography. 2016;43: 2147–2160.
- 25. Fukami H, Budd AF, Paulay G, Solé-Cava A, Chen CA, Iwao K, et al. Conventional taxonomy obscures deep divergence between Pacific and Atlantic corals. Nature. 2004;427: 832–835. pmid:14985760
- 26. Fukami H. Short review: molecular phylogenetic analyses of reef corals. Galaxea. 2008;10: 47–55. https://doi.org/10.3755/galaxea.10.47.
- 27. Huang D, Licuanan WY, Baird AH, Fukami H. Cleaning up the “Bigmessidae”: Molecular phylogeny of scleractinian corals from Faviidae, Merulinidae, Pectiniidae and Trachyphylliidae. BMC Evolutionary Biology. 2011;11: 37. pmid:21299898
- 28. Huang D, Benzoni F, Fukami H, Knowlton N, SMITH ND, Budd AF. Taxonomic classification of the reef coral families Merulinidae, Montastraeidae, and Diploastraeidae (Cnidaria: Anthozoa: Scleractinia). Zoological Journal of the Linnean Society. 2014;171: 277–355.
- 29. Huang D, Benzoni F, Arrigoni R, Baird AH, Berumen ML, Bouwmeester J, et al. Towards a phylogenetic classification of reef corals: the Indo-Pacific genera Merulina, Goniastrea and Scapophyllia (Scleractinia, Merulinidae). Zoologica Scripta. 2014;
- 30. Kitahara MV, Fukami H, Benzoni F, Huang D. The New Systematics of Scleractinia: Integrating Molecular and Morphological Evidence. 2016. pp. 41–59.
- 31. Markmann M, Tautz D. Reverse taxonomy: an approach towards determining the diversity of meiobenthic organisms based on ribosomal RNA signature sequences. Philosophical Transactions Royal Soc B Biological Sci. 2005;360: 1917–1924 pmid:16214749
- 32. Hoelzer GA. Inferring Phylogenies From mtDNA Variation: Mitochondrial-Gene Trees Versus Nuclear-Gene Trees Revisited. Evolution. 1997;51: 622–626. pmid:28565366
- 33. Brown W, George M, Wilson A. Rapid evolution of animal mitochondrial DNA. Proceedings of the National Academy of Sciences of the United States of America. 1979;76: 1967–1971. pmid:109836
- 34. Harrison R. Animal mitochondrial DNA as a genetic marker in population and evolutionary biology. Trends In Ecology & Evolution. 1989;4: 6–11. pmid:21227301
- 35. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular marine biology and biotechnology. 1994;3: 294–299. http://pubmed.gov/7881515 pmid:7881515
- 36. Moore WS. Inferring Phylogenies from mtDNA Variation: Mitochondrial-Gene Trees Versus Nuclear-Gene Trees. Evolution. 1995;49: 718–726. pmid:28565131
- 37. Bucklin A, Steinke D, Blanco-Bercial L. DNA Barcoding of Marine Metazoa. Annual Rev. Mar. Sci. 2011;3: 471–508. pmid:21329214
- 38. Avise J, Arnold J, Ball R, Bermingham E, Lamb T, Neigel J, et al. Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systematics. 1987;18: 489–522. https://doi.org/10.1146/annurev.es.18.110187.002421
- 39. Martin AP, Naylor GJ, Palumbi SR. Rates of mitochondrial DNA evolution in sharks are slow compared with mammals. Nature. 1992;357: 153–155. pmid:1579163
- 40. Menotti-Raymond M O’Brien SJ. Dating the genetic bottleneck of the African cheetah. Proceedings of the National Academy of Sciences of the United States of America. 1993;90: 3172–3176. http://pubmed.gov/8475057 pmid:8475057
- 41. Romano S, Palumbi S. Molecular evolution of a portion of the mitochondrial 16S ribosomal gene region in scleractinian corals. Journal of Molecular Evolution. 1997;45: 397–411. pmid:9321419
- 42. Maynard-Smith J. Evolutionary Genetics. 1999; 1–362.
- 43. Lum JK, Cann RL. mtDNA lineage analyses: origins and migrations of Micronesians and Polynesians. American Journal of Physical Anthropology. 2000;113: 151–168. pmid:11002202
- 44. Zouros E. The exceptional mitochondrial DNA system of the mussel family Mytilidae. Genes & Genetic Systems. 2000;75: 313–318. pmid:11280005
- 45. Oppen MJH van McDonald BJ, Willis B Miller DJ. The Evolutionary History of the Coral Genus Acropora (Scleractinia, Cnidaria) Based on a Mitochondrial and a Nuclear Marker: Reticulation, Incomplete Lineage Sorting, or Morphological Convergence? Mol Biol Evol. 2001;18: 1315–1329. pmid:11420370
- 46. Chen C, Wallace C, Wolstenholme J. Analysis of the mitochondrial 12S rRNA gene supports a two-clade hypothesis of the evolutionary history of scleractinian corals. Molecular Phylogenetics and Evolution 2002;23: 137–149. pmid:12069546
- 47. Wolfe K, Li W-H, Sharp P. Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast, and nuclear DNAs. Proceedings of the National Academy of Sciences of the United States of America. 1987;84: 9054–9058. pmid:3480529
- 48. Schon I, Butlin R, Griffiths H, Martens K. Slow molecular evolution in an ancient asexual ostracod. Proceedings of the Royal Society B. 1998;265: 235–242. https://doi.org/10.1098/rspb.1998.0287
- 49. Shearer T, Oppen M van, Romano S, Wörheide G. Slow mitochondrial DNA sequence evolution in the Anthozoa (Cnidaria). Molecular Ecology. 2002;11: 2475–2487. pmid:12453233
- 50. Terraneo TI, Benzoni F, Arrigoni R, Baird AH, Mariappan KG, Forsman ZH, et al. Phylogenomics of Porites from the Arabian Peninsula. Molecular Phylogenetics and Evolution. 2021; 107173. https://doi.org/10.1016/j.ympev.2021.107173
- 51. Quattrini AM, Snyder KE, Purow-Ruderman R, Seiblitz IGL, Hoang J, Floerke N, et al. Mito-nuclear discordance within Anthozoa, with notes on unique properties of their mitochondrial genomes. Scientific Report. 2023;13: 7443. pmid:37156831
- 52. Potts D, Done T, Isdale P, Fisk D. Dominance of a coral community by the genus Porites (Scleractinia). Mar Ecol Prog Ser. 1985;23: 79–84.
- 53. Pacherres CO, Schmidt GM, Richter C. Autotrophic and heterotrophic responses of the coral Porites lutea to large amplitude internal waves. J Exp Biol. 2013;216: 4365–4374. pmid:23997201
- 54. Padilla-Gamiño JL, Hanson KM, Stat M, Gates RD. Phenotypic plasticity of the coral Porites rus: Acclimatization responses to a turbid environment. Journal of Experimental Marine Biology and Ecology. 2010;434–435: 1–10.
- 55. Levas SJ, Grottoli AG, Hughes A, Osburn CL, Matsui Y. Physiological and Biogeochemical Traits of Bleaching and Recovery in the Mounding Species of Coral Porites lobata: Implications for Resilience in Mounding Corals. Plos One. 2013;8: e63267. pmid:23658817
- 56. Loya Y, Sakai K, Yamazato K, Nakano Y, Sambali H, van Woesik R. Coral bleaching: the winners and the losers. Ecol Lett. 2001;4: 122–131.
- 57. Woesik R van, Sakai K, Ganase A, Loya Y. Revisiting the winners and the losers a decade after coral bleaching. Mar Ecol Prog Ser. 2011;434: 67–76.
- 58. Brakel WH, Taylor DL. Corallite variation in Porites and the species problem in corals. 1977. Proc. 3rd Int. Coral Reef Symp; Vol. 1; pp. 457–462.
- 59. Flora CJ, Ely PS. Surface growth rings of Porites lutea microatolls accurately track their annual growth. Northwest Science. 2003;77: 237–245.
- 60. Forsman ZH, Barshis DJ, Hunter CL, Toonen RJ. Shape-shifting corals: Molecular markers show morphology is evolutionarily plastic in Porites. BMC Evolutionary Biology. 2009;9: 45. pmid:19239678
- 61. Forsman ZH, Ritson-Williams R, Tisthammer KH, Knapp ISS, Toonen RJ. Host-symbiont coevolution, cryptic structure, and bleaching susceptibility, in a coral species complex (Scleractinia; Poritidae). Scientific Reports. 2020;10: 1–12. pmid:33046719
- 62. Katoh K, Standley DM. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution. 2013;30: 772–780. pmid:23329690
- 63. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol Biol Evol. 2018;35: 1547–1549. pmid:29722887
- 64. Castresana J. Selection of Conserved Blocks from Multiple Alignments for Their Use in Phylogenetic Analysis. Mol Biol Evol. 2000;17: 540–552. pmid:10742046
- 65. Nguyen L-T, Schmidt HA, Haeseler AV, Minh BQ. IQ-TREE: A Fast and Effective Stochastic Algorithm for Estimating Maximum-Likelihood Phylogenies. Molecular Biology and Evolution. 2014;32: 268–274. pmid:25371430
- 66. Minh BQ, Nguyen MAT, Haeseler A von. Ultrafast Approximation for Phylogenetic Bootstrap. Mol Biol Evol. 2013;30: 1188–1195. pmid:23418397
- 67. Kalyaanamoorthy S, Minh BQ, Wong TKF, Haeseler A von, Jermiin LS. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods. 2017;14: 587–589. pmid:28481363
- 68. Chernomor O, Haeseler A von, Minh BQ. Terrace Aware Data Structure for Phylogenomic Inference from Supermatrices. Systematic Biol. 2016;65: 997–1008. pmid:27121966
- 69. Hoang DT, Chernomor O, Haeseler A von, Minh BQ Vinh . UFBoot2: Improving the Ultrafast Bootstrap Approximation. Mol Biol Evol. 2018;35: 518–522. pmid:29077904
- 70. Anisimova M, Gil M, Dufayard J-F, Dessimoz C, Gascuel O. Survey of Branch Support Methods Demonstrates Accuracy, Power, and Robustness of Fast Likelihood-based Approximation Schemes. Systematic Biol. 2011;60: 685–699. pmid:21540409
- 71. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30: 1312–1313. pmid:24451623
- 72. Miller MA, Pfeiffer W, Schwartz T. Creating the CIPRES Science Gateway for Inference of Large Phylogenetic Trees. 2010. pp. 1–8.
- 73. Stamatakis A, Hoover P, Rougemont J. A Rapid Bootstrap Algorithm for the RAxML Web Servers. Systematic Biology. 2008;57: 758–771. pmid:18853362
- 74. Ronquist F, Huelsenbeck J, Teslenko M. MrBayes 3.2 Manual: Tutorials and Model Summaries. 2011; 1–103.
- 75. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more models, new heuristics and parallel computing. Nature Methods. 2012;9: 772–772. pmid:22847109
- 76. Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA. Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7. Systematic Biol. 2018;67: 901–904. pmid:29718447
- 77. Veron JEN, Pichon M. Scleractinia of Eastern Australia. Part 5: Family Poritidae. Townsville, Australia: Australian Institute of Marine Science; 1982.
- 78.
Veron JEN, Stafford-Smith M, DeVantier L. Corals of the World. In: Corals of the World [Internet]. [cited 12 Dec 2022]. Available: http://www.coralsoftheworld.org/page/home/
- 79. Niu W, Lin R, Shi X, Chen C-H, Shen K-N, Hsiao C-D. Next-generation sequencing yields the complete mitogenome of massive coral, Porites lutea (Cnidaria: Poritidae). Mitochondrial Dna Part B. 2016;1: 8–9. pmid:33490386
- 80. Terraneo TI, Benzoni F, Baird AH, Arrigoni R, Berumen ML. Morphology and molecules reveal two new species of Porites (Scleractinia, Poritidae) from the Red Sea and the Gulf of Aden. Systematics and Biodiversity. 2019;0: 1–18.
- 81. Forsman ZH, Knapp ISS, Tisthammer K, Eaton DAR, Belcaid M, Toonen RJ. Coral hybridization or phenotypic variation? Genomic data reveal gene flow between Porites lobata and P. compressa. Molecular Phylogenetics and Evolution. 2017;111: 132–148. pmid:28366817
- 82. Brown NP, Forsman ZH, Tisthammer KT, Richmond RH. A resilient brooding coral in the broadcast spawning Porites lobata species complex: a new endemic, introduced species, mutant, or new adaptive potential? Coral Reefs. 2020;324: 1–10.
- 83. Terraneo TI, Fusi M, Hume BCC, Arrigoni R, Voolstra CR, Benzoni F, et al. Environmental latitudinal gradients and host‐specificity shape Symbiodiniaceae distribution in Red Sea Porites corals. Journal of Biogeography. 2019;88: 14–13.
- 84. Stefani F, Benzoni F, Pichon M, Mitta G, Galli P. Genetic and morphometric evidence for unresolved species boundaries in the coral genus Psammocora (Cnidaria; Scleractinia). Hydrobiologia. 2008;596: 153–172.
- 85. Benzoni F, Stefani F, Pichon M, Galli P. The name game: morpho-molecular species boundaries in the genus Psammocora (Cnidaria, Scleractinia). Zoological Journal of the Linnean Society. 2010;160: 421–456.
- 86. Arrigoni R, Berumen ML, Chen CA, Terraneo TI, Baird AH, Payri C, et al. Species delimitation in the reef coral genera Echinophyllia and Oxypora (Scleractinia, Lobophylliidae) with a description of two new species. Molecular Phylogenetics and Evolution. 2016; 1–31. pmid:27593164
- 87. Benzoni F, Arrigoni R, Berumen ML, TAVIANI M, Bongaerts P, Frade PR. Morphological and genetic divergence between Mediterranean and Caribbean populations of Madracis pharensis (Heller 1868) (Scleractinia, Pocilloporidae): too much for one species? Zootaxa. 2018;4471: 473–31. pmid:30313392
- 88. Arrigoni R, Huang D, Berumen ML, Budd AF, Montano S, Richards ZT, et al. Integrative systematics of the scleractinian coral genera Caulastraea, Erythrastrea and Oulophyllia. Zool Scr. 2021;50: 509–527.
- 89. Bonito VE, Baird AH, Bridge T, Cowman PF, Fenner D. Types, topotypes and vouchers are the key to progress in coral taxonomy: Comment on Wepfer et al. 2020. Mol Phylogenet Evol. 2021;159: 107104. pmid:33609706