Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Dec 13;34(22):e17615. doi: 10.1111/mec.17615

Ancient Hybridisation Fuelled Diversification in Acropora Corals

Tianzhen Wu 1,2, Alan Ningyuan Xu 3, Yanli Lei 1,2,4,✉, Haijun Song 5,✉
PMCID: PMC12617331  PMID: 39670962

ABSTRACT

Introgression is the infiltration or flow of genes from one species to another through hybridisation followed by backcrossing. This may lead to incorrect phylogenetic reconstruction or divergence‐time estimation. Acropora is a dominant genus of reef‐building corals; however, whether this group has an introgression history before their diversification remains unclear, and previous divergence‐time estimates of Acropora have not considered the impact of introgression. In this study, we broke through the limitation of a few genes and a few species and proved the existence of ancient introgression in the evolution of Acropora from whole‐genome protein‐coding sequences. We inferred 21.9% of all triplet loci (homologous loci from three different species) with a history of introgression and a series of introgression events with a genetic material contribution of up to 30.9% before diversification. Furthermore, 7756 nuclear loci were clustered into three groups using a multidimensional scaling algorithm, the heterogeneity of which resulted in different phylogenetic relationships. The diversification time of Acropora was estimated to be middle to late Miocene when we retained only the gene group with the lowest degree of introgression. The collision of Australia with the Pacific arcs and the Southeast Asian margin in the early Miocene, and a series of cooling events in the middle to late Miocene, may provide geographical and climatic conditions for the diversification of Acropora, respectively. Therefore, our results indicate that at the genome‐wide level, ancient introgressive hybridisation may have promoted the radiation evolution of Acropora. Based on our results, the influence of introgression should be taken into account when reconstructing phylogenetic relationships and evaluating divergence time.

Keywords: Acropora, divergence time, hybridisation, introgression, molecular evolution, phylogenetic reconstruction

1. Introduction

Hybridisation, the crossbreeding between individuals of different species, and introgression, the transfer of genes between species‐mediated primarily by backcrossing, has always been an active area of evolutionary research (Anderson 1949; Arnold 1992; Twyford and Ennos 2012). As an important evolutionary phenomenon, introgressive hybridisation has been documented in many lineages, such as humans, butterflies, insects, corals and some plants (Bolstad et al. 2021; Edelman et al. 2019; Mao, Economo, and Satoh 2018; Stone and Wolfe 2021; Storz and Signore 2021; Suarez‐Gonzalez, Lexer, and Cronk 2018; Suvorov et al. 2022). F1 hybrids are initially produced between the two species. Although most F1 hybrids may not be fully adaptive or may be sterile, they can mate with their parental species to produce first‐generation backcross (BC1) hybrids. Then, these BC1 hybrids may backcross with pure‐species individuals, leading to gene flow from one species to another (Anderson and Hubricht 1938; Arnold and Martin 2009).

The transfer of alleles between species not only affects the inference of their phylogenetic relationship but also causes the divergence‐time estimates with large deviations (Hibbins and Hahn 2022). Introgression affects the coalescence times of the non‐sister pair of species not involved in introgression (Hibbins and Hahn 2022). In other words, in the species tree (((P1, P2), P3), O), introgression between P2 and P3 enables them to quickly coalesce at the introgressive loci, which reduces their whole‐genome divergence relative to P1 and P3. When introgression occurs from P2 into P3, a smaller divergence time is obtained. Because P3 traces its ancestry through the P2 lineage in this case, it enables P3 to coalesce with P1 earlier than normal (Hibbins and Hahn 2022). In addition, there is considerable evidence in both plants and animals that hybridisation promotes chromosomal and base mutation rates in backcross offspring. For example, the frequency of mutation was faster in the hybrid between Drosophila simulans and D. melanogaster (Belgovsky and Beljgovskiï 1937). Increased chromosomal mutation rate was found in the hybrid between two subspecies of grasshoppers (Shaw, Wilkinson, and Coates 1983). Furthermore, the hybrid progeny of Arabidopsis exhibited decreased T → G and T → A transversion rates but an increased C → T transition rate (Bashir et al. 2014). Therefore, hybridisation‐induced changes in mutation rates may also influence the inferences made about the evolutionary history of species. However, when reconstructing phylogenies or estimating species divergence times for lineages where hybridisation is common, the effect of introgression on the reliability of the results is often overlooked.

Acropora is the dominant genus of reef‐building corals across the Indo‐Pacific region and is currently the most extensively studied coral genus (Ball et al. 2021). More than 100 Acropora species have been identified in the Indo‐Pacific region, up to 76 of which are sympatrically distributed and at least 35 of which participate in synchronised mass spawning events (Willis et al. 2006). Under the influence of ocean currents, gametes from different species have the opportunity for direct contact, which increases the possibility of hybridisation and introgression. The coral reef ecosystem formed by tropical Acropora species supports a high level of marine biodiversity. However, Acropora species are very sensitive to global warming. Frequent bleaching and potential death of these corals could lead to a reduction in the biodiversity of the entire shallow marine environment (Adam et al. 2021; Fuller et al. 2020). Conservation work for Acropora species is imminent; however, an understanding of its evolutionary history is unclear because of the influence of hybridisation and introgression.

Phylogenetic reconstruction of Acropora corals inferred by a few nuclear and mitochondrial genes, such as Pax‐C intron, ATP6, CYTB and the mtDNA putative control region, has always been inconsistent (Fukami et al. 2021; Rosser et al. 2017; van Oppen et al. 2001). This was thought to be the result of factors such as introgression and incomplete lineage sorting (ILS) (van Oppen et al. 2001; Willis et al. 2006). Using high‐throughput sequencing technology, Quek et al. (2023) obtained the largest phylogenomic reconstruction of Scleractinia using 449 nuclear loci. Their tree topology of Acropora is different from that of Rosser et al. (2017), which was inferred from 10,034 single nucleotide polymorphisms. For example, A. selago was located in the basal branch of the genus in the latter but in the deeper branch in the former. The number of major clades of Acropora identified by Quek et al. (2023) was also unequal compared with the results of Cowman et al. (2020) using a mean of 1850 loci per taxon. In addition, the analysis of the combination of CYTB and ATP6 suggested that Acropora corals diversified approximately 2 million years ago (Ma) (Fukami, Omori, and Hatta 2000), yet diversification of Acropora was extended to approximately 25–50 Ma when using 818 single‐copy orthologous genes (Shinzato et al. 2021). This large discrepancy in the estimated diversification times of Acropora corals may be explained by the fact that none of these previous analyses took into account the potential effects of introgression. Therefore, detecting heterogeneity in the phylogenetic signal across loci and excluding inconsistent gene trees involving introgression as much as possible are necessary to improve the accuracy of phylogenetic reconstruction or divergence‐time estimates (Duchêne et al. 2018), particularly for Acropora corals, which are highly susceptible to bleaching events resulting from global warming. Thus, an accurate inference of their evolutionary history can facilitate scientific conservation efforts.

Even though the existence of introgression was revealed by a few genes or species (Mao, Economo, and Satoh 2018; van Oppen et al. 2001; Willis et al. 2006), a recent study indicated that hybridisation is not as frequent as expected (Ramírez‐Portilla et al. 2022). Furukawa et al. (2024) also showed that the tabular Acropora species does not involve hybridisation. These studies have introduced a fundamental ambiguity regarding the role of hybridisation in the evolutionary history of Acropora corals. In the present study, more than 7000 loci, including nuclear and mitochondrial encoding genes, across 15 Acropora species were obtained using the method of whole‐genome comparison to address the following two questions: (1) Is there introgressive hybridisation in the evolutionary history of Acropora, and what is its proportion in the genome? (2) What is the diversification time of Acropora after excluding introgression sites? Our introgression results overcame the limitations of previous studies that relied on a small number of genetic markers and species and confirmed the existence of ancient introgressive hybridisation in Acropora corals. Furthermore, the influence of introgressive genomic regions should be excluded during phylogenetic analysis and molecular dating of highly hybridised taxa.

2. Materials and Methods

2.1. Species Selection

Three databases were searched for nuclear and mitochondrial genome data, including the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/), ReefGenomics (http://reefgenomics.org/), OIST marine genomics (https://marinegenomics.oist.jp/gallery). Fifteen Acropora species were selected (Table S1), including A. awi , A. cervicornis , A. cytherea , A. digitifera , A. florida , A. hyacinthus , A. intermedia , A. microphthalma , A. millepora , A. muricata , A. nasuta , A. palmata , A. selago , A. tenuis and A. yongei . For the unselected Acropora species, they either had no nuclear genome data, no mitochondrial genome data or neither, so far. For example, although A. gemmifera and A. echinata both have genome data, we did not obtain their mitochondrial genomes using their genome sequencing reads with three different assembly methods. In addition, according to Quek et al. (2023), the lack of these two species does not affect the coverage of major clades within the genus. Besides, Montipora capitata was selected to be the outgroup for evolutionary analysis.

The mitochondrial genomes of A. awi and A. selago were assembled using genome sequencing reads (accession: DRR089900 and DRR089931 respectively, Table S1). Fastp 0.23.4 (Chen et al. 2018) was used to perform quality control on the raw reads downloaded from the Sequence Read Archive (SRA) database. Then, three mitochondria assembly software programs, including GetOrganelle1.7.7.1 (Jin et al. 2020), NOVOPlasty4.3.1 (Dierckxsens, Mardulyn, and Smits 2017) and meangs1.2.1 (Song, Yan, and Li 2022), were used to obtain the complete mitochondrial genome. Bandage 0.8.1 (Wick et al. 2015) was used for interactive visualisation of the assemblies. Finally, the mitochondrial genome annotation of 15 Acropora species was performed using the GeSeq website (https://chlorobox.mpimp‐golm.mpg.de/geseq.html).

2.2. Acquisition of Orthologous Genes

Because the vast majority of Acropora species have no nuclear genome annotation, we used the method of whole‐genome alignment to extract orthologous gene sets. A. millepora , with the best assembly quality and gene structure annotation, was selected as the reference genome. To speed up calculations and save computing resources, the genome sequence file was split into 14 separate files according to the number of chromosomes. Based on ‘local’ approaches, Lastz 1.04.22 (Harris 2007) was used to produce a large set of nucleotide‐level alignments after aligning 14 sequences files with other 15 genomes, respectively. Compared to blast, Lastz can identify orthology and paralogy, and detect homologous sequences that have low similarity but are strongly supported in their relatedness by flanking homologous sequences. Then, Multiz 11.2 (Blanchette et al. 2004) was used to merge the maf files generated in the previous step to obtain collinear blocks for all species. During this process, only the collinear blocks that included all species were retained. We extracted the protein‐coding sequences within these collinear blocks based on the coordinate positions in the gene annotation file of A. millepora . To ensure high‐quality alignments, we applied a threshold for the completeness of the protein‐coding sequences. If a species' protein‐coding sequence was incomplete by more than 1/3 (with a threshold of 0.33), the alignment was discarded. Finally, 7756 well‐aligned nuclear orthologous genes were obtained for subsequent phylogenetic analysis. The whole genome alignment method considers flanking homologous sequences, which helps filter out genes duplicated in individual species, ensuring that the final result consists of single‐copy genes. Additionally, we used Orthofinder 3.0.1b1 (Emms and Kelly 2019) to cluster these genes. These 7756 genes were assigned to a total of 7584 orthogroups. At least 90% of the genes are single‐copy orthologous (Figure S1). In addition, gffread 0.12.7 (Pertea and Pertea 2020) was used to extract all mitochondrial protein‐coding sequences according to our mitochondrial genome annotations.

2.3. Phylogenetic Analyses

Phylogenomic inferences were conducted by concatenation and coalescence‐based methods. First, 7756 nuclear loci were concatenated into one super sequence using Perl script. The fourfold degenerate sites (4dTVs) were subsequently extracted from the supersequence for phylogenetic tree construction. The maximum likelihood (ML) tree was inferred by Iqtree 2.2.0.3 (Minh et al. 2020) with parameters ‘−b 100’ and ‘−m MFP’, which instructs the ModelFinder program to select the best model. For 13 mitochondrial gene sequences, we also used the concatenation‐based method to obtain a phylogenetic tree. In addition, a coalescent‐based phylogenetic method was used to infer the species tree from a set of 7756 ML trees implemented in ASTRAL 5.7.8 (Mirarab et al. 2014) with the ‘−b’ parameter. A quartet score ranging from 0 to 1 was obtained for the inferred species tree. The higher the value, the smaller the difference between the species tree and the gene tree.

The rooted 7756 gene trees were then compared with the species tree using PhyParts (Smith et al. 2015), and the output showed the amount of conflict at each node along with the dominant alternative topology. The consensus of those gene trees was visualised using DensiTree.v3.0.2 (Bouckaert 2010). Besides, we randomly extracted three subsets of gene trees, in which each group included 100 gene trees. Then, splitstree4 (Huson 1998) was used to visualise their network evolution.

2.4. Detection of Introgression

QuIBL (Edelman et al. 2019) was used to quantify introgression through the branch lengths. We extracted all three‐taxon subtrees of Acropora species from the 7756 gene trees. We formally distinguished between two models (absence and presence of introgression) using a Bayesian information criterion (BIC) test with a strict cutoff of ∆BIC > 10 (Edelman et al. 2019). The InferNetwork_MPL method in PhyloNet 3.8.2 (Than, Ruths, and Nakhleh 2008) was also used to infer introgression based on maximum pseudo‐likelihood, which allows application to large datasets containing more than ten species and thousands of loci (Hibbins and Hahn 2022). Each operation inferred five independent evolutionary networks with introgression and their timing, extent, and potential direction. The respective −Log probability values of each inferred evolutionary network were calculated, and the smaller the value, the more reliable the inference. We ran the program three times to ensure the robustness of the results. To compare the rate of molecular evolution between ancestral and extant lineages, the nonsynonymous (d N) to synonymous substitution (d S) rate ratio (ω = d N/d S) was estimated for 7756 loci using freeratio model. We divided the obtained d N/d S values into two parts: one with 0.0002 < ω < 1 and the other with 1 < ω < 10 (removing the extreme values on both ends), because they represent purifying selection and positive selection respectively. The evolutionary rates of the ancestral and extant lineages were compared using T‐TEST, respectively.

2.5. Gene Tree Heterogeneity Analysis

There are many variables that contribute to inconsistent gene tree topologies. Dimensionality reduction condenses these variables, simplifying the representation of gene trees from high‐dimensional space to low‐dimensional space, which aids in clustering and visualising the heterogeneity between genes. The Robinson–Foulds distance between each pair of gene trees was calculated by the PH85 method of the APE package (Paradis, Claude, and Strimmer 2004). The smaller the value, the more similar the topological structure between the two trees. The multidimensional scaling (MDS) algorithm, a statistical method mapping high‐dimensional data onto a low‐dimensional space while preserving the distance relationships between data points as much as possible, describes the relative distances between gene trees by approximating these distances in Euclidean space (Borg and Groenen 2005; Hillis, Heath, and John 2005). It locates the Euclidean positions of the gene trees with the principle of minimising the sum of the distances between them. This Euclidean spatial distribution can help us understand the similarities and differences among the various genetic loci. The cmdscale function of the TreeDist package (Smith 2022) was used to represent their distances in a varying number of dimensions to reflect the tree‐topology space. Then, the kmeansPP function of the Cluster package (Maechler et al. 2013) was used to classify loci with similar evolutionary histories, setting the maximum number of clusters to eight. The optimal number of clusters was tested by calculating the Gap statistic. Finally, when using four or five MDS dimensions, 7756 loci were clustered into three groups, named group 1, group 2 and group 3, respectively.

We further estimated the d N/d S for 7756 loci. The one‐ratio model of the PAML package (Yang 2007) was used with the parameters model = 0, NSsites = 0, and fix_omega = 0. A Welch two sample t‐test was conducted to determine whether there were evolutionary differences between the three groups of loci detected as heterogeneous with each other as described above. Based on the coalescent method, we reconstructed species trees for the three groups of gene clusters using Iqtree 2.2.0.3 (Minh et al. 2020) and ASTRAL 5.7.8 (Mirarab et al. 2014). We focused on the normalised quartet score to determine the extent of introgression among species.

2.6. Divergence‐Time Estimation

Based on molecular clock theory, divergence‐time estimates were conducted using the MCMCtree program of the PAML package (Yang 2007) for the gene cluster with the highest normalised quartet score. The dates of the two nodes were constrained by fossil records, including (1) the upper and lower limit of divergence of Acropora and Montipora as 136.4 and 70.6 Ma; and (2) the upper and lower limit of the divergence of A. florida and A. tenuis as 55.8 and 20.43 Ma (Shinzato et al. 2021). Using a relaxed molecular clock (clock = 2), the posterior distribution of parameters was estimated by Markov chain Monte Carlo sampling with 1,000,000 iterations, a burn‐in of 2,000,000, and a sampling frequency of 100. The results of the MCMCtree analyses were examined using Tracer v1.7.2 (Rambaut et al. 2018) to determine the convergence between parameters, node ages and likelihood values. The estimated sample size for each parameter was > 200.

3. Results

3.1. Inconsistency of Gene Trees

Mitochondrial protein‐coding sequences of 15 Acropora corals were annotated and extracted for concatenation‐based phylogenetic reconstruction (Figures 1A and S2). The frequency of polymorphisms in all mitochondrial genes was very low (Figure 1B). In particular, ATP8 and ND4L were monomorphic. ND5, which was the most polymorphic, only has 8 haplotypes for all Acropora corals examined in this study. In contrast, nuclear loci have higher levels of polymorphisms (Figure 1C). For 7756 nuclear protein‐coding sequences, the inferred topology using the concatenation method was consistent with that evaluated by the coalescence method (Figures 2A and S3), with the bootstrap value of all nodes being 100%. However, the topologies of the mitochondrial and nuclear tree were inconsistent. For example, A. cervicornis and A. palmata were placed at the base position except ( A. tenuis , A. yongei ) in the mitochondrial tree, whereas they were placed at a deeper position in the nuclear tree (Figures 1A and 2A). Because mitochondrial loci exhibit less genetic polymorphism compared with nuclear loci (Figure 1B,C), we focused on the nuclear topology in subsequent analyses, which was overwhelmingly consistent with the phylogenetic tree of Quek et al. (2023).

FIGURE 1.

FIGURE 1

Open in a new tab

(A) Phylogenetic reconstruction based on concatenated mitochondrial protein‐coding sequences of 15 Acropora species using Iqtree 2.2.0.3. The five different colours represent the five major clades of Acropora species that we recovered, a result that differs greatly from the topological relationships reported in Quek et al. (2023), both within and between clades. (B) Polymorphism level of the mitochondrial genes, in which the same colour represents the same haplotype. The order of the species here corresponds to the arrangement of the phylogenetic tree depicted on the left in A. (C) Polymorphism level of 7756 nuclear loci is depicted. Each nuclear locus is represented by a blue dot. Each unit position on the horizontal axis corresponds to the position of a nuclear locus, with a total of 7756 loci. The vertical axis represents the number of nucleotide polymorphism in 15 species. Compared with mitochondrial genes, most nuclear genes have rich nucleotide polymorphisms.

FIGURE 2.

FIGURE 2

Open in a new tab

Visualisation of network evolution. (A) Combined species tree topology for Acropora using PhyParts software. For each branch, the two values in the upper row represent the bootstrap values from the coalescence and concatenation‐based methods, respectively. The two values on the bottom line indicate the number of loci concordant with the species tree at that node and the number of loci in conflict with that clade in the species tree. The pie charts at each node represent the proportion of loci that support the indicated topology (blue), the proportion that supports the main alternative for that clade (green), and the proportion that support the remaining alternatives (red). The five different colours represent the five major clades of Acropora species, which is consistent with the topological relationships reported in Quek et al. (2023). (B) Inference of network evolution using 100 gene trees randomly extracted from the datasets visualised by splitstree4 software. The blue lines form a complex network block, representing the evolutionary history of non‐binary divergence. The dense network blocks are primarily concentrated at the ancestral nodes of five major clades, particularly the ancestor nodes of all Acropora except A. tenuis and A. yongei . (C) Consensus of 7756 gene trees visualised by DensiTree.v3.0.2. The 7756 gene trees are shown in light purple. Blue represents the most likely tree topology, red represents the second most popular topology, and green represents the third most likely. The clade comprising A. cytherea , A. hyacinthus , A. muricata , A. millepora and A. selago exhibited the highest level of variation in gene tree topology.

The pervasive nuclear gene tree‐species tree conflict was detected in the coalescence tree (Figure 2A). In the deep branch, the ancestor node of Acropora muricata, Acropora millepora and Acropora selago was only supported by 1467 of the 7756 gene trees. Network analysis exhibited obvious reticulate evolutionary relationships at multiple ancestor nodes (Figures 2B and S4), particularly the ancestor nodes of all Acropora except A. tenuis and A. yongei . Moreover, topological conflicts among gene trees were widely observed when superimposing the 7756 gene trees in a consensus DensiTree plot (Figure 2C). There were clear differences between the most likely tree topology, the second most popular topology, and the third most likely tree topology, particularly in the clade composed of A. cytherea , A. hyacinthus , A. muricata , A. millepora and A. selago . These intricate network blocks and huge differences between different gene tree topologies clearly illustrate the non‐binary branching patterns within their evolutionary history.

3.2. Ancient Introgressive Hybridisation

The likelihoods that the distribution of internal branch lengths is consistent with introgression or with ILS only for all triplets were calculated from QUIbL. Introgression signals were identified between all nonsister taxa (Figure 3A and Table S2). The extent of introgression ranged from 0.22% between A. digitifera and A. tenuis to 1.03% between A. nasuta and A. digitifera . As the species diverged, higher levels of introgression were observed within major branches. The average extent of introgression of A. tenuis and A. yongei with other species was the lowest, whereas that of A. millepora was the highest (Figure 3A). Averaging over all of the triplets, 98% (with BIC filtering) of the loci with discordant gene trees were inferred to have a history of introgression, or 22% (with BIC filtering) of all triplet loci, indicating widespread introgression signals across the Acropora clade.

FIGURE 3.

FIGURE 3

Open in a new tab

Introgression analysis. (A) Numbers in the heat map inferred from QuIBL software represent the average degree of introgression between two species, the darker the colour, the greater the degree of introgression. The size of the blue circles represents the average level of introgression detected between that species and others. (B) Ancient hybridisation events were analysed using PhyloNet 3.8.2, which identified five hybridisation events, denoted as H1 to H5. The two blue lines in front of each hybridisation event represent the two different parental sources. The numbers indicate the genetic contributions from both parent lineages.

PhyloNet analysis identified five hybrid branches as the best fit with the statistical measures applied (log probability, Figures 3B, S4 and S5). Nearly all identified hybridisation events occurred before the diversification of Acropora, except the basal branch ( A. tenuis , A. yongei ). In the evolution network with the lowest value of −log probability, the extent of introgression ranged from 0.1% for H2 to 30.9% for H3 (Figure 3B), which was higher compared with that of the QuIBL analysis.

In Acropora, we counted the differences in the d N/d S ratio (ω < 1) between ancestral and extant lineages, which represent purifying selection. The d N/d S ratio for all ancestral nodes was significantly lower than that for all extant lineages (ω ancestral = 0.318, ω extant = 0.324, p = 3.97E‐05, Figure S9A,B). In the lineages of A. cervicornis and A. palmata , as well as the A. tenuis and A. yongei , higher ω values were identified in extant species (Figures S9C,E). In contrast, in other major clades, ancestral lineages displayed higher ω values than extant lineages (Figures S9D,F,G). In addition, we also counted the differences in d N/d S ratio (ω > 1) between ancestral and extant lineages, which represents positive selection. The d N/d S ratio for all ancestral nodes was significantly higher than that for all extant lineages (ω ancestral = 2.23, ω extant = 1.90, p = 7.24E‐27, Figure S9H). Except for A. cervicornis and A. palmata , whose d N/d S ratio was significantly higher than that of their ancestral lineages (ω ancestral = 1.67, ω extant = 2.27, p = 2.32E‐09, Figure S9K), the d N/d S ratios of the ancestral lineages in the other four major clades were higher than those of the extant lineages (Figures S9I,J,L,M).

3.3. Heterogeneity of Nuclear Protein Coding Sequences

Based on the calculated Robinson‐Foulds distances between different gene trees, which measure the differences between two phylogenetic trees, and using the MDS algorithm, these gene trees were clustered. Three topology clusters were found when using four or five MDS dimensions (Figures 4A, S6 and S7). Only a single cluster was identified when using one, two or three MDS dimensions (Figure S6), which did not demonstrate heterogeneity in the nuclear loci. Based on four MDS dimensions, the 7756 genetic loci used to construct these gene trees were correspondingly divided into three groups, consisting of 2802, 2008 and 2946 loci, respectively. Subsequent analyses, including d N/d S ratio analysis, phylogenetic tree construction and final normalised quartet score determination, revealed significant differences among these groups.

FIGURE 4.

FIGURE 4

Open in a new tab

Heterogeneity of nuclear loci. (A) Clustering of 7756 genes using the MDS algorithm with four dimensions, including 2802 loci in group 1, 2008 loci in group 2, and 2946 loci in group 3. From the first dimension, as also depicted in Figures S7A–C, the trees in Group 2 (green) were separated from the others (red and blue). From the second dimension, as also depicted in Figures S7A,D,E, Group 1 (red) and Group 3 (blue) were separated. (B) Loci from group 2 have significantly higher d N/d S values compared with that in the loci from the other two groups using a t‐test, with **** representing p ≤ 0.0001. (C) Topology A, topology B and topology C are possible topologies of A. awi , A. florida and A. intermedia . The numbers indicate the number of gene trees that support this topology. The species trees evaluated by group 2 and group 3 support topology A, shown in green and blue respectively. The species trees evaluated by group 1 support topology B, shown in red. None of the trees constructed from the three groups supported topology C.

For each group, the topologies constructed using the concatenation and coalescence methods were consistent (Figure S8). The topologies of group 2 and group 3 were consistent with that of all nuclear loci; however, the topology of group 1 was inconsistent, which was reflected in the relationship between A. awi , A. florida and A. intermedia . Topology A (( A. awi , A. florida ), A. intermedia ) was inferred from loci of group 2 and group 3, whereas topology B (( A. florida , A. intermedia ), A. awi ) was inferred from loci of group 1, which was consistent with the inference by PhyloNet analysis (Figures 3B and 4C). The d N/d S analysis revealed that the evolution rate of group 2 was significantly higher compared with the other two groups (ω 1 = 0.22, ω 2 = 0.28 and ω 3 = 0.22; P(ω 1–ω 2) < 2.2e‐16, P(ω 1–ω 3) = 0.7659 and P(ω 2–ω 3) < 2.2e‐16, Figure 4B). The QuIBL test revealed that the inference of topology B was resulted from introgression (Figure 4C and Table S2). In addition, the group with the smallest final normalised quartet score of 0.60 was group 2, whereas the largest score of 0.84 was observed in group 3. This indicates that the level of introgression for the loci in group 2 was above the average of 0.77, whereas the level of introgression for the loci in group 3 was below the average. Therefore, the factors contributing to the heterogeneity among these three groups of nuclear loci include at least different d N/d S ratios and levels of introgression.

3.4. Divergence‐Time Estimates

Because of the differences in topology and potential introgression degree observed among the three groups, we selected the third group to evaluate the species divergence time for Acropora. The ML tree from group 3 was used as the input tree for the MCMCTree analysis. The common ancestor of the Acropora genus was estimated to have existed between 131.26 and 19.91 Ma, with diversification inferred to have occurred during the middle to late Miocene epoch, extending up until the late Pliocene period (approximately 13.28–3.42 Ma, Figure 5A). Our estimated timing of Acropora diversification was later compared with the results that did not account for introgression (Shinzato et al. 2021), and earlier than the findings based solely on two mitochondrial loci (Fukami, Omori, and Hatta 2000). It was comparable to the results obtained through a combination of the nuclear Pax‐C 46/47 intron and the mitochondrial RNS‐CNOX3 control region (Richards, Miller, and Wallace 2013). Throughout the diversification of the Acropora corals, global mean temperature exhibited a downward trend, until a significant disturbance occurred at approximately 3 Ma (Figure 5A).

FIGURE 5.

FIGURE 5

Open in a new tab

(A) Divergence‐time estimates of Acropora corals inferred from 2946 nuclear genes using MCMCtree program. The grey shaded area represents the rapid radiative evolution interval. The blue bars through the nodes indicate 95% credibility intervals. (B) Fossil evidence for Acropora from Indonesia. From approximately 18 Ma to the end of the Miocene epoch, a morphological comparison of Acropora fossils found in Indonesia revealed the presence of 19 existing species, 6 extinct species and 6 open nomenclature species. (C) The bottom curve represents temperature dynamics, from Tierney et al. (2020). EM, early Miocene; MM, middle Miocene; LM, late Miocene; P, Pliocene; and Q, Quaternary.

4. Discussion

Acropora corals have important ecological roles and have garnered substantial attention from evolutionary researchers (Ball et al. 2021). A. prolifera are entirely F1 hybrids of A. cervicornis and A. palmata . They fail to meet the definition of a species and are not qualified to be protected (Vollmer and Palumbi 2002; Willis et al. 2006). The scientific conservation of this genus requires a better understanding of its evolutionary history. Identification of ancient introgressive hybridisation emphasises its important role in the evolution of Acropora corals and provides theoretical support for scientific conservation.

4.1. Extensive Ancient Hybridisation

There appears to be a consensus that Acropora has a history of hybridisation (Fogarty 2012; Isomura et al. 2016; Mao, Economo, and Satoh 2018; Van Oppen et al. 2002; Willis et al. 2006). Compatible gametes are demonstrated when eggs are provided only with the opportunity to mate with a heterospecific sperm (Van Oppen et al. 2002; Willis et al. 2006). This is even more important at the edge of a species range, where ecologically successful clonal lineages may persist even in the absence of sexual reproduction (Bullini 1994; Kearney 2005; Seehausen 2004; Willis et al. 2006). With the development of sequencing technology, studies have shown that some Acropora species seldom participates in hybridisation events (Furukawa et al. 2024; Ramírez‐Portilla et al. 2022), which complicates the evolutionary history of Acropora corals. By overcoming the limitations of the number of species and loci in previous studies, we provide genome‐wide evidence for frequent occurrences of ancient hybridisation during the evolution of Acropora.

The discovery of mitochondrial genes was unexpected, as none of the 13 protein‐coding genes were fully polymorphic across the 15 Acropora species studied (Figure 1B). Fukami, Omori, and Hatta (2000) used CYTB and ATP6 from 6 Acropora species and obtained a diversification time that was later than other results, which was likely due to the extremely low polymorphism. Expanding the species dataset revealed up to 10 mitochondrial genes, each with no more than five haplotypes across the 15 Acropora species examined. Therefore, mitochondrial genes lack sufficient variation to provide informative data for phylogenetic analysis. There are two possibilities for the highly shared haplotypes between different species. First, mitochondrial genes have been subjected to strong purifying selection during the process of radiation evolution, leaving them with few retainable mutations. However, this may be untenable because the neutral theory of molecular evolution states that the vast majority of mutations are neutral or nearly neutral (Kimura 1979). Second, their frequent hybridisation and strong capacity to reproduce asexually should be considered (Dias et al. 2019; Twyford and Ennos 2012). However, more population sampling for mitochondrial genes is needed to distinguish the effects of introgression from ILS. The classification of Acropora species is challenging, not only because of their morphological similarities (Bongaerts et al. 2021) but also because of the lack of variation in their mitochondrial genes.

Using 7756 nuclear loci across 15 species, we identified broad introgression signals. The positive results between A. cervicornis and A. palmata , which are only distributed in the Caribbean and other species distributed in the Indo‐Pacific waters, have led us to speculate that ancient introgression hybridisation occurred. The results of the PhyloNet analysis confirm this and the high level of introgression may involve hybrid speciation (Baack and Rieseberg 2007). In the present study, we provide genome‐wide evidence that hybridisation may have facilitated the radiation evolution of Acropora, as rapid species diversification occurred following complicated hybridisation events (Figures 3B and S4). This phenomenon is common in taxa with radiation evolution, such as Lake Malawi cichlid fish, salamanders and Spialia butterflies (Hinojosa et al. 2022; Patton et al. 2020; Svardal et al. 2020). Therefore, we suspect that some of the introgression signals observed between extant Acropora species may be the result of ancient hybrid introgression that was diluted over time. However, our results cannot rule out the possibility of hybridisation occurring between extant Acropora species, as gamete compatibility studies indicate that coral hybrids can be generated (Isomura, Iwao, and Fukami 2013; Kitanobo et al. 2016; Vollmer and Palumbi 2002). The degree of hybrid introgression in ancient times was likely much higher compared with today, which provides a key impetus for the evolution of Acropora (Mao 2020). Hobbs et al. (2022) provided evidence of hybridisation in 81 stony corals. Recently, the first triploid genome of Pocillopora acuta was assembled, which probably resulted from the hybridisation of two different species (Stephens et al. 2022). Therefore, as more samples are collected and sequencing costs decrease, the hybridisation of species within Scleractinia will likely be detected at the molecular level. Our d N/d S ratio analysis suggests that introgression may have accelerated the evolutionary rate in these ancestral lineages. The significantly higher levels of positive selection observed in the ancestral lineage of four major clades indicate that a higher proportion of mutations may be environmentally adaptive during the radiation evolution of Acropora. This provides potential evidence that introgression is a driving force of adaptive evolution. Previous studies have established that genes related to DNA repair and immunity play a crucial role in the evolutionary history of stony corals (Wu, Lei, and Jian 2023, 2024). However, additional research is essential to pinpoint the specific genes that contribute to the adaptive evolution of the Acropora genus.

4.2. Divergence‐Time Estimates After Excluding Loci With High Introgression Levels

Hybridisation sites may produce biased phylogenetic trees and estimates of the evolutionary timescale, reflecting the effects of heterogeneity in phylogenetic signals across loci (Duchêne et al. 2018; Mitchell et al. 2014). Nevertheless, the heterogeneity of evolutionary signals has not yet been widely detected across most taxa, particularly those that have undergone rapid radiation evolution accompanied by extensive hybridisation. Three groups of nuclear coding gene clusters were obtained using the MDS algorithm (Hillis, Heath, and John 2005). The tree from group 1 shows the different relationship between A. awi , A. florida and A. intermedia from the other groups, indicating that introgression among these three species is the potential basis for clustering. Loci from group 2 exhibited the lowest final normalised quartet score, representing the highest level of introgression or ILS. Hybridisation may lead to increased mutation rates in the offspring (Bashir et al. 2014; Belgovsky and Beljgovskiï 1937; Kimura 1979). Therefore, introgression is likely to be dominant, as a significantly higher evolutionary rate in this group was detected compared with the other two groups. In contrast, the lowest levels of introgression are detected for loci in group 3, which is considered an accurate representation of the true evolutionary relationships and timescales.

We obtained the divergence time of Acropora by excluding the influence of introgression where possible (Figure 5A). They began to diverge in the early Miocene period (approximately 19.91 Ma) and experienced an accelerated divergence during the middle to late Miocene, extending up until the late Pliocene.

4.3. External Driving Force for the Rapid Radiation of Acropora

Our estimation of Acropora divergence time is roughly consistent with the fossil record of Acropora diversity in the Neogene, during which 23 extant Acropora species appeared (Santodomingo, Wallace, and Johnson 2015). Especially from approximately 18 Ma to the end of the Miocene epoch, 19 existing species, 6 extinct species and 6 open nomenclature of Acropora were discovered in Indonesia through morphological comparison (Figure 5B). The collision of Australia with Pacific arcs and the Southeast Asian margin of approximately 23 Ma may provide fundamental geographical conditions for the diversification of species in the middle to late Miocene (Renema et al. 2008). The extensive formation of new islands and shallow seas was particularly important for a wide range of taxa in shallow marine ecosystems, including Acropora corals (Mathew et al. 2020; Renema et al. 2008). In addition, climate warming results in the breakdown of the symbiotic relationship between corals and their zooxanthellae and is considered to be the primary driver of coral bleaching and mortality (Lesser 2011; Sully et al. 2019). The middle to late Miocene period, which was a pivotal interval in the transition of the Earth's climate toward the modern state (Nairn et al. 2021), may have provided a favourable temperature environment for the diversification of Acropora corals (Figure 5C). Diversification of Acropora occurred closely after the middle Miocene climatic transition approximately 14.7–12.5 Ma, during which the global mean temperature cooled by approximately 6°C–7°C (Henderiks et al. 2020; Tierney et al. 2020). The diversification rate of Acropora accelerated from the late Miocene to the late Pliocene, corresponding to the continued decrease during this extended period (Herbert et al. 2016). Changes in ocean currents caused by the global cooling trend may have further promoted hybridisation and diversification within the Acropora corals (Willis et al. 2006). Therefore, divergence‐time estimates of Acropora, after excluding the influence of introgression loci as much as possible, are supported by fossil, geological and palaeoclimate evidence.

5. Conclusion

The results of this study provide genome‐wide evidence for ancient hybridisation and introgression in Acropora corals. During the middle to late Miocene, hybridisation as an internal driving force and the reduction of global temperatures as an external one may have jointly promoted the rapid radiation of Acropora. Furthermore, we emphasise the need to account for heterogeneity in evolutionary signals across loci when performing phylogenetic reconstructions and divergence‐time estimates for taxa with extensive hybridisation histories. For Acropora, a group with a complex evolutionary history that is extremely sensitive to temperature increases, scientific conservation efforts must be developed based on its unique evolutionary characteristics. Urgent and comprehensive results, including reticular evolution, adaptive evolution, gene editing and artificial propagation, should be conducted to ensure that Acropora corals do not disappear from the Earth by the end of this century.

Author Contributions

Tianzhen Wu, Yanli Lei and Haijun Song conceived and designed the study. Tianzhen Wu and Alan Ningyuan Xu performed the bioinformatics analysis. Tianzhen Wu, Yanli Lei and Haijun Song interpreted the data and wrote the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Benefit‐Sharing

Benefits Generated: Benefits from this research accrue from the sharing of our data and results on public databases as described above.

Supporting information

Figure S1. Statistics of single‐copy genes in 7756 genes.

MEC-34-e17615-s003.pdf (256.1KB, pdf)

Figure S2. Visualisation of the mitochondrial gene structure and functional annotation of 15 Acropora corals.

MEC-34-e17615-s002.pdf (3MB, pdf)

Figure S3. Inferred topology based on the concatenation of 7756 nuclear protein‐coding sequences.

MEC-34-e17615-s009.pdf (264.4KB, pdf)

Figure S4. Number of hybridisation events and their statistical log probability values. The smaller the absolute value, the more robust the results.

MEC-34-e17615-s001.pdf (315.6KB, pdf)

Figure S5. In addition to Figure 3B, other evolution networks with five times hybridisation events were identified from different run batches.

MEC-34-e17615-s010.pdf (306.2KB, pdf)

Figure S6. Optimal number of clusters in different dimensions from 1 to 5, determined by the gap statistic.

MEC-34-e17615-s006.pdf (284.6KB, pdf)

Figure S7. Visualisation of the clustering of 7756 trees under different dimension combinations.

MEC-34-e17615-s005.pdf (5.5MB, pdf)

Figure S8. Phylogenetic trees were reconstructed using three gene clusters, employing both the coalescence (left) and concatenation (right) methods.

MEC-34-e17615-s007.pdf (339.9KB, pdf)

Figure S9. Comparison of d N/d S between ancestral and extant lineages.

MEC-34-e17615-s004.pdf (274.3KB, pdf)

Table S1. Datasets used in this study, including 15 Acropora corals and Montipora capitata as the outgroup.

Table S2. Results of QuIBL analysis.

MEC-34-e17615-s008.xlsx (240.6KB, xlsx)

Acknowledgements

Special thanks are due to Prof. Dr. Zhimin Jian (Tongji University, China) for his valuable discussion, supports and guidance to the first corresponding author in coral research. Special thanks are due to Dr. Xin Huang for technical support. Many thanks to Dr. Haotian Li and Dr. Wenlong Fa for the technical discussion. We appreciate the anonymous reviewers and the editor for their constructive advises on this manuscript.

Handling Editor: Shotaro Hirase

Funding: This work was supported by grants from the National Key R&D Program of China (Grant No. 2022YFC2803800), the Laoshan Laboratory (No. LSKJ202204200), the Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDB42000000). In addition, we thank the supports from the international scientific program Ocean Negative Carbon Emissions (Global‐ONCE).

Contributor Information

Yanli Lei, Email: leiyanli@qdio.ac.cn.

Haijun Song, Email: haijunsong@cug.edu.cn.

Data Availability Statement

This study did not generate any new sequencing data. The data supporting the results are available from the GeneBank and SRA database (https://www.ncbi.nlm.nih.gov/) through the accession numbers listed in Table S1. The scripts and the alignment of 7756 orthologous genes are available from the GitHub repository (https://github.com/Wu‐tz/Scripts‐in‐Acropora‐analysis).

References

  1. Adam, A. A. , Garcia R. A., Galaiduk R., et al. 2021. “Diminishing Potential for Tropical Reefs to Function as Coral Diversity Strongholds Under Climate Change Conditions.” Diversity and Distributions 27, no. 11: 2245–2261. [Google Scholar]
  2. Anderson, E. 1949. Introgressive Hybridization. New York: John Wiley and Sons. [Google Scholar]
  3. Anderson, E. , and Hubricht L.. 1938. “Hybridization in Tradescantia. III. The Evidence for Introgressive Hybridization.” American Journal of Botany 25: 396–402. [Google Scholar]
  4. Arnold, M. L. 1992. “Natural Hybridization as an Evolutionary Process.” Annual Review of Ecology and Systematics 23, no. 1: 237–261. [Google Scholar]
  5. Arnold, M. L. , and Martin N. H.. 2009. “Adaptation by Introgression.” Journal of Biology 8: 1–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baack, E. J. , and Rieseberg L. H.. 2007. “A Genomic View of Introgression and Hybrid Speciation.” Current Opinion in Genetics & Development 17, no. 6: 513–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Ball, E. E. , Hayward D. C., Bridge T. C., and Miller D. J.. 2021. “Acropora—The Most‐Studied Coral Genus.” In Handbook of Marine Model Organisms in Experimental Biology, 173–193. Boca Raton, FL: CRC Press. [Google Scholar]
  8. Bashir, T. , Sailer C., Gerber F., et al. 2014. “Hybridization Alters Spontaneous Mutation Rates in a Parent‐Of‐Origin‐Dependent Fashion in Arabidopsis.” Plant Physiology 165, no. 1: 424–437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Belgovsky, B. , and Beljgovskiï M.. 1937. “A Comparison of the Frequency of Induced Mutations in Drosophila simulans and Its Hybrid With D. melanogaster .” Genetica 19: 370–386. [Google Scholar]
  10. Blanchette, M. , Kent W. J., Riemer C., et al. 2004. “Aligning Multiple Genomic Sequences With the Threaded Blockset Aligner.” Genome Research 14, no. 4: 708–715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bolstad, G. H. , Karlsson S., Hagen I. J., et al. 2021. “Introgression From Farmed Escapees Affects the Full Life Cycle of Wild Atlantic Salmon.” Science Advances 7, no. 52: eabj3397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Bongaerts, P. , Cooke I. R., Ying H., et al. 2021. “Morphological Stasis Masks Ecologically Divergent Coral Species on Tropical Reefs.” Current Biology 31, no. 11: 2286–2298. [DOI] [PubMed] [Google Scholar]
  13. Borg, I. , and Groenen P. J.. 2005. Modern Multidimensional Scaling: Theory and Applications. New York: Springer Science & Business Media. [Google Scholar]
  14. Bouckaert, R. R. 2010. “DensiTree: Making Sense of Sets of Phylogenetic Trees.” Bioinformatics 26, no. 10: 1372–1373. [DOI] [PubMed] [Google Scholar]
  15. Bullini, L. 1994. “Origin and Evolution of Animal Hybrid Species.” Trends in Ecology & Evolution 9, no. 11: 422–426. [DOI] [PubMed] [Google Scholar]
  16. Chen, S. , Zhou Y., Chen Y., and Gu J.. 2018. “Fastp: An Ultra‐Fast All‐In‐One FASTQ Preprocessor.” Bioinformatics 34, no. 17: i884–i890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Cowman, P. F. , Quattrini A. M., Bridge T. C., et al. 2020. “An Enhanced Target‐Enrichment Bait Set for Hexacorallia Provides Phylogenomic Resolution of the Staghorn Corals (Acroporidae) and Close Relatives.” Molecular Phylogenetics and Evolution 153: 106944. [DOI] [PubMed] [Google Scholar]
  18. Dias, M. , Ferreira A., Gouveia R., and Vinagre C.. 2019. “Synergistic Effects of Warming and Lower Salinity on the Asexual Reproduction of Reef‐Forming Corals.” Ecological Indicators 98: 334–348. [Google Scholar]
  19. Dierckxsens, N. , Mardulyn P., and Smits G.. 2017. “NOVOPlasty: de Novo Assembly of Organelle Genomes From Whole Genome Data.” Nucleic Acids Research 45, no. 4: e18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Duchêne, D. A. , Bragg J. G., Duchêne S., et al. 2018. “Analysis of Phylogenomic Tree Space Resolves Relationships Among Marsupial Families.” Systematic Biology 67, no. 3: 400–412. [DOI] [PubMed] [Google Scholar]
  21. Edelman, N. B. , Frandsen P. B., Miyagi M., et al. 2019. “Genomic Architecture and Introgression Shape a Butterfly Radiation.” Science 366, no. 6465: 594–599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Emms, D. M. , and Kelly S.. 2019. “OrthoFinder: Phylogenetic Orthology Inference for Comparative Genomics.” Genome Biology 20: 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Fogarty, N. D. 2012. “Caribbean Acroporid Coral Hybrids Are Viable Across Life History Stages.” Marine Ecology Progress Series 446: 145–159. [Google Scholar]
  24. Fukami, H. , Niimura A., Nakamori T., and Iryu Y.. 2021. “Species Composition and Mitochondrial Molecular Phylogeny of Acropora Corals in Funakoshi, Amami‐Oshima Island, Japan: A Proposal for Its New Taxonomic Grouping.” Galaxea, Journal of Coral Reef Studies 23, no. 1: 17–35. [Google Scholar]
  25. Fukami, H. , Omori M., and Hatta M.. 2000. “Phylogenetic Relationships in the Coral Family Acroporidae, Reassessed by Inference From Mitochondrial Genes.” Zoological Science 17, no. 5: 689–696. [DOI] [PubMed] [Google Scholar]
  26. Fuller, Z. L. , Mocellin V. J., Morris L. A., et al. 2020. “Population Genetics of the Coral Acropora millepora : Toward Genomic Prediction of Bleaching.” Science 369, no. 6501: eaba4674. [DOI] [PubMed] [Google Scholar]
  27. Furukawa, M. , Kitanobo S., Ohki S., Teramoto M. M., Hanahara N., and Morita M.. 2024. “Integrative Taxonomic Analyses Reveal That Rapid Genetic Divergence Drives Acropora Speciation.” Molecular Phylogenetics and Evolution 195: 108063. [DOI] [PubMed] [Google Scholar]
  28. Harris, R. S. 2007. Improved Pairwise Alignment of Genomic DNA. State College: Pennsylvania State University. [Google Scholar]
  29. Henderiks, J. , Bartol M., Pige N., Karatsolis B. T., and Lougheed B. C.. 2020. “Shifts in Phytoplankton Composition and Stepwise Climate Change During the Middle Miocene.” Paleoceanography and Paleoclimatology 35, no. 8: e2020PA003915. [Google Scholar]
  30. Herbert, T. D. , Lawrence K. T., Tzanova A., Peterson L. C., Caballero‐Gill R., and Kelly C. S.. 2016. “Late Miocene Global Cooling and the Rise of Modern Ecosystems.” Nature Geoscience 9, no. 11: 843–847. [Google Scholar]
  31. Hibbins, M. S. , and Hahn M. W.. 2022. “Phylogenomic Approaches to Detecting and Characterizing Introgression.” Genetics 220, no. 2: iyab173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Hillis, D. M. , Heath T. A., and John K. S.. 2005. “Analysis and Visualization of Tree Space.” Systematic Biology 54, no. 3: 471–482. [DOI] [PubMed] [Google Scholar]
  33. Hinojosa, J. C. , Dapporto L., Pitteloud C., et al. 2022. “Hybridization Fuelled Diversification in Spialia Butterflies.” Molecular Ecology 31, no. 10: 2951–2967. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hobbs, J.‐P. A. , Richards Z. T., Popovic I., et al. 2022. “Hybridisation and the Evolution of Coral Reef Biodiversity.” Coral Reefs 41, no. 3: 535–549. [Google Scholar]
  35. Huson, D. H. 1998. “SplitsTree: Analyzing and Visualizing Evolutionary Data.” Bioinformatics (Oxford, England) 14, no. 1: 68–73. [DOI] [PubMed] [Google Scholar]
  36. Isomura, N. , Iwao K., and Fukami H.. 2013. “Possible Natural Hybridization of Two Morphologically Distinct Species of Acropora (Cnidaria, Scleractinia) in the Pacific: Fertilization and Larval Survival Rates.” PLoS One 8, no. 2: e56701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Isomura, N. , Iwao K., Morita M., and Fukami H.. 2016. “Spawning and Fertility of F 1 Hybrids of the Coral Genus Acropora in the Indo‐Pacific.” Coral Reefs 35: 851–855. [Google Scholar]
  38. Jin, J.‐J. , Yu W.‐B., Yang J.‐B., et al. 2020. “GetOrganelle: A Fast and Versatile Toolkit for Accurate de Novo Assembly of Organelle Genomes.” Genome Biology 21: 1–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kearney, M. 2005. “Hybridization, Glaciation and Geographical Parthenogenesis.” Trends in Ecology & Evolution 20, no. 9: 495–502. [DOI] [PubMed] [Google Scholar]
  40. Kimura, M. 1979. “The neutral theory of molecular evolution.” Scientific American 241, no. 5: 98–129. [DOI] [PubMed] [Google Scholar]
  41. Kitanobo, S. , Isomura N., Fukami H., Iwao K., and Morita M.. 2016. “The Reef‐Building Coral Acropora Conditionally Hybridize Under Sperm Limitation.” Biology Letters 12, no. 8: 20160511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Lesser, M. P. 2011. “Coral Bleaching: Causes and Mechanisms.” Coral Reefs: An Ecosystem in Transition, 405–419. Dordrecht: Springer. [Google Scholar]
  43. Maechler, M. , Rousseeuw P., Struyf A., et al. 2013. “Package ‘Cluster’.” Dosegljivo Na, 980.
  44. Mao, Y. 2020. “Genomic Insights Into Hybridization of Reef Corals.” Coral Reefs 39, no. 1: 61–67. [Google Scholar]
  45. Mao, Y. , Economo E. P., and Satoh N.. 2018. “The Roles of Introgression and Climate Change in the Rise to Dominance of Acropora Corals.” Current Biology 28, no. 21: 3373–3382. [DOI] [PubMed] [Google Scholar]
  46. Mathew, M. , Makhankova A., Menier D., Sautter B., Betzler C., and Pierson B.. 2020. “The Emergence of Miocene Reefs in South China Sea and Its Resilient Adaptability Under Varying Eustatic, Climatic and Oceanographic Conditions.” Scientific Reports 10, no. 1: 7141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Minh, B. Q. , Schmidt H. A., Chernomor O., et al. 2020. “IQ‐TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era.” Molecular Biology and Evolution 37, no. 5: 1530–1534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Mirarab, S. , Reaz R., Bayzid M. S., Zimmermann T., Swenson M. S., and Warnow T.. 2014. “ASTRAL: Genome‐Scale Coalescent‐Based Species Tree Estimation.” Bioinformatics 30, no. 17: i541–i548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Mitchell, K. J. , Pratt R. C., Watson L. N., et al. 2014. “Molecular Phylogeny, Biogeography, and Habitat Preference Evolution of Marsupials.” Molecular Biology and Evolution 31, no. 9: 2322–2330. [DOI] [PubMed] [Google Scholar]
  50. Nairn, M. G. , Lear C. H., Sosdian S. M., Bailey T. R., and Beavington‐Penney S.. 2021. “Tropical Sea Surface Temperatures Following the Middle Miocene Climate Transition From Laser‐Ablation ICP‐MS Analysis of Glassy Foraminifera.” Paleoceanography and Paleoclimatology 36, no. 3: e2020PA004165. [Google Scholar]
  51. Paradis, E. , Claude J., and Strimmer K.. 2004. “APE: Analyses of Phylogenetics and Evolution in R Language.” Bioinformatics 20, no. 2: 289–290. [DOI] [PubMed] [Google Scholar]
  52. Patton, A. H. , Margres M. J., Epstein B., Eastman J., Harmon L. J., and Storfer A.. 2020. “Hybridizing Salamanders Experience Accelerated Diversification.” Scientific Reports 10, no. 1: 6566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Pertea, G. , and Pertea M.. 2020. “GFF Utilities: GffRead and GffCompare.” F1000Research 9: 304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Quek, Z. R. , Jain S. S., Richards Z. T., et al. 2023. “A Hybrid‐Capture Approach to Reconstruct the Phylogeny of Scleractinia (Cnidaria: Hexacorallia).” Molecular Phylogenetics and Evolution 186: 107867. [DOI] [PubMed] [Google Scholar]
  55. Rambaut, A. , Drummond A. J., Xie D., Baele G., and Suchard M. A.. 2018. “Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7.” Systematic Biology 67, no. 5: 901–904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Ramírez‐Portilla, C. , Baird A. H., Cowman P. F., et al. 2022. “Solving the Coral Species Delimitation Conundrum.” Systematic Biology 71, no. 2: 461–475. [DOI] [PubMed] [Google Scholar]
  57. Renema, W. , Bellwood D., Braga J., et al. 2008. “Hopping Hotspots: Global Shifts in Marine Biodiversity.” Science 321, no. 5889: 654–657. [DOI] [PubMed] [Google Scholar]
  58. Richards, Z. , Miller D., and Wallace C.. 2013. “Molecular Phylogenetics of Geographically Restricted Acropora Species: Implications for Threatened Species Conservation.” Molecular Phylogenetics and Evolution 69, no. 3: 837–851. [DOI] [PubMed] [Google Scholar]
  59. Rosser, N. L. , Thomas L., Stankowski S., Richards Z. T., Kennington W. J., and Johnson M. S.. 2017. “Phylogenomics Provides New Insight Into Evolutionary Relationships and Genealogical Discordance in the Reef‐Building Coral Genus Acropora.” Proceedings of the Royal Society B: Biological Sciences 284, no. 1846: 20162182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Santodomingo, N. , Wallace C. C., and Johnson K. G.. 2015. “Fossils Reveal a High Diversity of the Staghorn Coral Genera Acropora and Isopora (Scleractinia: Acroporidae) in the Neogene of Indonesia.” Zoological Journal of the Linnean Society 175, no. 4: 677–763. [Google Scholar]
  61. Seehausen, O. 2004. “Hybridization and Adaptive Radiation.” Trends in Ecology & Evolution 19, no. 4: 198–207. [DOI] [PubMed] [Google Scholar]
  62. Shaw, D. , Wilkinson P., and Coates D.. 1983. “Increased Chromosomal Mutation Rate After Hybridization Between Two Subspecies of Grasshoppers.” Science 220, no. 4602: 1165–1167. [DOI] [PubMed] [Google Scholar]
  63. Shinzato, C. , Khalturin K., Inoue J., et al. 2021. “Eighteen Coral Genomes Reveal the Evolutionary Origin of Acropora Strategies to Accommodate Environmental Changes.” Molecular Biology and Evolution 38, no. 1: 16–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Smith, M. R. 2022. “Robust Analysis of Phylogenetic Tree Space.” Systematic Biology 71, no. 5: 1255–1270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Smith, S. A. , Moore M. J., Brown J. W., and Yang Y.. 2015. “Analysis of Phylogenomic Datasets Reveals Conflict, Concordance, and Gene Duplications With Examples From Animals and Plants.” BMC Evolutionary Biology 15: 1–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Song, M.‐H. , Yan C., and Li J.‐T.. 2022. “MEANGS: An Efficient Seed‐Free Tool for de Novo Assembling Animal Mitochondrial Genome Using Whole Genome NGS Data.” Briefings in Bioinformatics 23, no. 1: bbab538. [DOI] [PubMed] [Google Scholar]
  67. Stephens, T. G. , Lee J., Jeong Y., et al. 2022. “High‐Quality Genome Assemblies From Key Hawaiian Coral Species.” GigaScience 11: giac098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Stone, B. W. , and Wolfe A. D.. 2021. “Phylogeographic Analysis of Shrubby Beardtongues Reveals Range Expansions During the Last Glacial Maximum and Implicates the Klamath Mountains as a Hotspot for Hybridization.” Molecular Ecology 30, no. 15: 3826–3839. [DOI] [PubMed] [Google Scholar]
  69. Storz, J. F. , and Signore A. V.. 2021. “Introgressive Hybridization and Hypoxia Adaptation in High‐Altitude Vertebrates.” Frontiers in Genetics 12: 696484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Suarez‐Gonzalez, A. , Lexer C., and Cronk Q. C.. 2018. “Adaptive Introgression: A Plant Perspective.” Biology Letters 14, no. 3: 20170688. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Sully, S. , Burkepile D., Donovan M., Hodgson G., and Van Woesik R.. 2019. “A Global Analysis of Coral Bleaching Over the Past Two Decades.” Nature Communications 10, no. 1: 1264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Suvorov, A. , Scornavacca C., Fujimoto M. S., et al. 2022. “Deep Ancestral Introgression Shapes Evolutionary History of Dragonflies and Damselflies.” Systematic Biology 71, no. 3: 526–546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Svardal, H. , Quah F. X., Malinsky M., et al. 2020. “Ancestral Hybridization Facilitated Species Diversification in the Lake Malawi Cichlid Fish Adaptive Radiation.” Molecular Biology and Evolution 37, no. 4: 1100–1113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Than, C. , Ruths D., and Nakhleh L.. 2008. “PhyloNet: A Software Package for Analyzing and Reconstructing Reticulate Evolutionary Relationships.” BMC Bioinformatics 9: 1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Tierney, J. E. , Poulsen C. J., Montañez I. P., et al. 2020. “Past Climates Inform Our Future.” Science 370, no. 6517: eaay3701. [DOI] [PubMed] [Google Scholar]
  76. Twyford, A. D. , and Ennos R.. 2012. “Next‐Generation Hybridization and Introgression.” Heredity 108, no. 3: 179–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. van Oppen, M. J. , McDonald B. J., Willis B., and Miller D. J.. 2001. “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?” Molecular Biology and Evolution 18, no. 7: 1315–1329. [DOI] [PubMed] [Google Scholar]
  78. Van Oppen, M. J. , Willis B. L., Van Rheede T., and Miller D. J.. 2002. “Spawning Times, Reproductive Compatibilities and Genetic Structuring in the Acropora aspera Group: Evidence for Natural Hybridization and Semi‐Permeable Species Boundaries in Corals.” Molecular Ecology 11, no. 8: 1363–1376. [DOI] [PubMed] [Google Scholar]
  79. Vollmer, S. V. , and Palumbi S. R.. 2002. “Hybridization and the Evolution of Reef Coral Diversity.” Science 296, no. 5575: 2023–2025. [DOI] [PubMed] [Google Scholar]
  80. Wick, R. R. , Schultz M. B., Zobel J., and Holt K. E.. 2015. “Bandage: Interactive Visualization of de Novo Genome Assemblies.” Bioinformatics 31, no. 20: 3350–3352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Willis, B. L. , van Oppen M. J., Miller D. J., Vollmer S. V., and Ayre D. J.. 2006. “The Role of Hybridization in the Evolution of Reef Corals.” Annual Review of Ecology, Evolution, and Systematics 37: 489–517. [Google Scholar]
  82. Wu, T. , Lei Y., and Jian Z.. 2023. “DNA Repair‐Related Genes Are the Key for Stony Coral Ancestors to Survive Under Elevated Levels of UVR.” Frontiers in Marine Science 10: 1170565. [Google Scholar]
  83. Wu, T. , Lei Y., and Jian Z.. 2024. “Amplification of Immune Genes in Ancient Stony Corals for Adapting to Unstable Marine Environments.” Ocean‐Land‐Atmosphere Research 3: 41. [Google Scholar]
  84. Yang, Z. 2007. “PAML 4: Phylogenetic Analysis by Maximum Likelihood.” Molecular Biology and Evolution 24, no. 8: 1586–1591. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Statistics of single‐copy genes in 7756 genes.

MEC-34-e17615-s003.pdf (256.1KB, pdf)

Figure S2. Visualisation of the mitochondrial gene structure and functional annotation of 15 Acropora corals.

MEC-34-e17615-s002.pdf (3MB, pdf)

Figure S3. Inferred topology based on the concatenation of 7756 nuclear protein‐coding sequences.

MEC-34-e17615-s009.pdf (264.4KB, pdf)

Figure S4. Number of hybridisation events and their statistical log probability values. The smaller the absolute value, the more robust the results.

MEC-34-e17615-s001.pdf (315.6KB, pdf)

Figure S5. In addition to Figure 3B, other evolution networks with five times hybridisation events were identified from different run batches.

MEC-34-e17615-s010.pdf (306.2KB, pdf)

Figure S6. Optimal number of clusters in different dimensions from 1 to 5, determined by the gap statistic.

MEC-34-e17615-s006.pdf (284.6KB, pdf)

Figure S7. Visualisation of the clustering of 7756 trees under different dimension combinations.

MEC-34-e17615-s005.pdf (5.5MB, pdf)

Figure S8. Phylogenetic trees were reconstructed using three gene clusters, employing both the coalescence (left) and concatenation (right) methods.

MEC-34-e17615-s007.pdf (339.9KB, pdf)

Figure S9. Comparison of d N/d S between ancestral and extant lineages.

MEC-34-e17615-s004.pdf (274.3KB, pdf)

Table S1. Datasets used in this study, including 15 Acropora corals and Montipora capitata as the outgroup.

Table S2. Results of QuIBL analysis.

MEC-34-e17615-s008.xlsx (240.6KB, xlsx)

Data Availability Statement

This study did not generate any new sequencing data. The data supporting the results are available from the GeneBank and SRA database (https://www.ncbi.nlm.nih.gov/) through the accession numbers listed in Table S1. The scripts and the alignment of 7756 orthologous genes are available from the GitHub repository (https://github.com/Wu‐tz/Scripts‐in‐Acropora‐analysis).

Articles from Molecular Ecology are provided here courtesy of Wiley

RESOURCES