Molecular Phylogenetics and Evolution 130 (2019) 269–285
Contents lists available at ScienceDirect
Molecular Phylogenetics and Evolution
journal homepage: www.elsevier.com/locate/ympev
The evolutionary history of the goby Elacatinus puncticulatus in the tropical eastern pacific: Effects of habitat discontinuities and local environmental T variability
E.R. Sandoval-Huertaa,b, R.G. Beltrán-Lópezc,d, C.R. Pedraza-Marróna,b, M.A. Paz-Velásqueze, ⁎ A. Angulof, D.R. Robertsong, E. Espinozah, O. Domínguez-Domínguezb, a Programa Institucional de Maestría en Ciencias Biológicas, Facultad de Biología, Universidad Michoacana de San Nicolás de Hidalgo, Edificio “R” planta baja, Ciudad Universitaria, Morelia, Michoacán 58030, Mexico b Laboratorio de Biología Acuática, Facultad de Biología, Universidad Michoacana de San Nicolás de Hidalgo, Edificio “R” planta baja, Ciudad Universitaria, Morelia, Michoacán 58030, Mexico c Programa Institucional de Doctorado en Ciencias Biológicas, Facultad de Biología, Universidad Michoacana de San Nicolás de Hidalgo, Edificio “R” planta baja, Ciudad Universitaria, Morelia, Michoacán 58030, Mexico d Laboratorio de Ictiología, Centro de Investigaciones Biológicas, Universidad Autónoma del Estado de Morelos, Av. Universidad no. 1001, Cuernavaca, Morelos 62209, Mexico e Centro de Estudios del Mar y Acuicultura, Universidad de San Carlos de Guatemala, Guatemala City, Guatemala f Museo de Zoología y Centro de Investigación en Ciencias del Mar y Limnología, Universidad de Costa Rica. 11501–2060, San Pedro de Montes de Oca, San José, Costa Rica g Naos Marine Laboratory, Smithsonian Tropical Research Institute, Balboa, Panama h Dirección del Parque Nacional Galápagos, Puerto Ayora, Islas Galápagos, Ecuador
ARTICLE INFO ABSTRACT
Keywords: Habitat discontinuities, temperature gradients, upwelling systems, and ocean currents, gyres and fronts, can Biogeographic barriers affect distributions of species with narrow environmental tolerance or motility and influence the dispersal of Habitat discontinuities pelagic larvae, with effects ranging from the isolation of adjacent populations to connections between them. The Mesoscale eddies coast of the Tropical Eastern Pacific (TEP) is a highly dynamic environment, with various large gyres and Upwellings upwelling systems, alternating currents and large rocky-habitat discontinuities, which may greatly influence the Reef-fishes genetic connectivity of populations in different parts of the coast. Elacatinus puncticulatus is a cryptic, shallow- Tropical Eastern Pacific living goby that is distributed along the continental shore of virtually the entire TEP, which makes it a good model for testing the influence of these environmental characteristics in the molecular evolution of widespread species in this region. A multilocus phylogeny was used to evaluate the influence of habitat gaps, and ocea- nographic processes in the evolutionary history of E. puncticulatus throughout its geographical range in the TEP. Two well-supported allopatric clades (one with two allopatric subclades) were recovered, the geographic dis- tribution of which does not correspond to any previously proposed major biogeographic provinces. These po- pulations show strong genetic structure and substantial genetic distances between clades and sub-clades (cytb 0.8–7.3%), with divergence times between them ranging from 0.53 to 4.88 Mya, and recent population ex- pansions dated at 170–130 Kya. The ancestral area of all populations appears to be the Gulf of Panama, while several isolation events have formed the phylogeographic patterns evident in this species. Local and regional oceanographic processes as well as habitat discontinuities have shaped the distribution patterns of the genetic lineages along the continental TEP. Large genetic distances, high genetic differentiation, and the results of species-tree and phylogenetic analyses indicate that E. puncticulatus comprises a complex of three allopatric species with an unusual geographic arrangement.
⁎ Corresponding author. E-mail addresses: [email protected] (R.G. Beltrán-López), [email protected] (D.R. Robertson), [email protected] (E. Espinoza), [email protected] (O. Domínguez-Domínguez). https://doi.org/10.1016/j.ympev.2018.10.020 Received 10 May 2018; Received in revised form 5 September 2018; Accepted 15 October 2018 Available online 23 October 2018 1055-7903/ © 2018 Elsevier Inc. All rights reserved. E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
1. Introduction they occur and on the region as a whole (Fiedler, 2002; Pennington et al., 2006; Fernández-Álamo and Färber-Lorda, 2006). Genetic connectivity among populations of marine organisms is The red-head goby, Elacatinus puncticulatus (Ginsburg, 1938), the strongly determined by the dispersal capability of the species and by only member of this neotropical genus in the eastern Pacific, is wide- barriers that may prevent or limit dispesal (Pineda et al., 2007; Cowen spread along the mainland of TEP, ranging from the central Gulf of and Sponaugle, 2009). For marine organisms some barriers are im- California to southern Ecuador. Malpelo Island, a tiny rock offshore passible and completely obstruct marine corridors, e.g. land masses from Colombia, is the only oceanic island in the TEP at which E. such as the Central American isthmus (Lessios, 2008); while others are puncticulatus has been recorded (Robertson and Allen, 2015), although permeable, and even large stretches of open ocean can be crossed by that likely is as a vagrant, as that record is new and it was not seen at some species (e. g. the Eastern Pacific Barrier; Robertson et al., 2004; Malpelo by divers searching for it there in 2018 (A. Polanco, pers. com. Lessios and Robertson, 2006). to DRR). Adults of E. puncticulatus live on rocky reefs, frequently in Large habitat discontinuities are considered to be one of the main association with sea urchins in shallow water (< 21 m) (Patzner et al., barriers to dispersal and expansion of the geographic ranges of reef- 2011; Robertson and Allen, 2015). Elacatinus species produce benthic fishes (Riginos and Nachman, 2001), by eliminating available suitable eggs, from which pelagic larvae hatch (Colin, 1975). The pelagic larval habitat for establishment at the end of the larval stage. In the Tropical stage of three Caribbean species of Elacatinus ranges from means of Eastern Pacific (TEP) two large stretches of sandy and muddy shoreline 21–38 days (Luiz et al., 2013; Rickborn and Buston, 2015). The larval represent such barriers for shallow-water reef fishes that have been duration of E. puncticulatus is not known. proposed as the limits of biogeographic provinces in the mainland TEP: The adult and likely larval characteristics of E. puncticulatus, to- the ∼525 km long Central American Gap between southern Mexico and gether with its widespread distribution in a range of subtropical and El Salvador, and the 370 km wide Sinaloan Gap on the southeast shore tropical environments across 34 degrees of latitude and > 5000 km of of the Gulf of California. (Ekman, 1953; Springer, 1959; Walker, 1960; the continental shore of virtually the entire TEP, make this species a Briggs, 1974; Hastings, 2000; Robertson and Cramer, 2009; Briggs and good model subject for testing how large habitat discontinuities and Bowen, 2012). Three biogeographic provinces relating to effects of oceanographic process may have affected genetic connectivity and these two gaps have been defined by the differences in species richness variation across populations, and whether the distribution of genetic and levels of endemism (see Fig. 1A). If effects of these sandy gaps as lineages parallels that of major recognized biogeographic subunits of barriers have produced such differences then their restrictive effects on the TEP. To do so we used mitochondrial and nuclear markers from the distributions of species should also be reflected in a genetic differ- individuals collected at sites between the northern and southern limits entiation between populations of the same species in different pro- of its continental range, in all three generally recognized continental vinces (Muss et al., 2001). However, in various reef fishes that are biogeographic provinces, to examine genetic relationships between widely distributed along the continental shore of the TEP, no evidence populations at different locations and provinces. of phylogeographic patterns related to the location of the two sandy gaps or the biogeographic provinces have been reported. These include 2. Materials and methods the blenniid Ophioblennius steindachneri (Muss et al., 2001); the serranid Epinephelus labriformis (Craig et al., 2006), four species of the haemulid 2.1. Specimen collection genus Anisotremus (Bernardi et al., 2008); the pomacanthid Holacanthus passer, and the pomacentrid Abudefduf troschelii (Bernardi et al., 2014). A total of 202 individuals of E. puncticulatus were collected in 39 Lessios and Baums (2017) provide examples of various invertebrate localities throughout the continental range of the species (Fig. 1B, taxa and circumtropical reef-fishes that lack of geographic genetic Table 1). The specimens were collected while SCUBA diving, using an structure related to those two barriers. In such cases, larval dispersal anesthetic solution of 5% clove oil in ethanol. Tissue samples comprised capability and biological and ecological traits evidently are more im- a pectoral fin of each individual, preserved in 95% ethanol and, stored portant than barriers produced by oceanographic conditions or large at −76 °C. Voucher specimens were fixed in formalin and preserved in habitat discontinuities (Rocha et al., 2007; Bernardi et al., 2014). There 70% ethanol. Tissue samples and voucher specimens were deposited in are a few exceptions however: For the sygnathid Hippocampus ingens,a the Ichthyological Collection of the Laboratorio de Biología Acuática, genetic break has been found between the Cortez Province and eight Universidad Michacana de San Nicolás de Hidalgo, México (SEMAR- sites along the remainder of the TEP coast, from Mexico to Peru, thus NAT registration number MICH-PEC-227-07-09). spanning all three major mainland biogeographic provinces (Saarman et al., 2010). In addition, data presented in Taylor and Hellberg, 2006 2.2. DNA amplification and alignment (their Fig. 2) indicates there is genetic differentiation in Elacatinus puncticulatus somewhere between the Gulf of California and Panama. Total genomic DNA was extracted using the phenol-chloroform Not only habitat discontinuities but also oceanographic character- protocol of Sambrook et al. (1989). The polymerase chain reaction istics, such as temperature gradients and upwelling systems, as well as (PCR) was used to amplify the mitochondrial Cytochrome b gene (cytb: major ocean currents, gyres and fronts, can affect distributions of spe- 1140 bp) of 202 specimens, using the primers LA and HA (Dowling cies with narrow environmental tolerance or motility (Hellberg et al., et al., 2002). Taking into account variation found in the cytb gene, a 2002; Craig et al., 2006; Rocha et al., 2007), and influence the cap- subset of 65 specimens were sequenced to obtain nuclear Rhodopsin ability for pelagic larvae to disperse (Rocha et al., 2005; Taylor and (RHO: 816 bp) with the primers RH193F and RH1073R (Chen et al., Hellberg, 2006; Floeter et al., 2008; Weersing and Toonen, 2009), with 2003). In addition, the first intron of the S7 ribosomal protein gene (S7: effects ranging from the isolation of (Gilg and Hilbish, 2003) to the 735 bp) was obtained from 76 specimens using the primers S7RPEX1F connection of adjacent populations (Mitarai et al., 2009). and S7RPEX2R (Chow and Hazama, 1998), following the conditions In this context, the coast of the TEP is a highly dynamic environ- described in Table S1. The PCR products were visualized electro- ment, with several gyres and fronts that develop at predictable loca- phoretically on 1.5% agarose gels and submitted to MACROGEN Korea tions and can change their direction or strength, including mesoscale for sequencing. gyres of the Gulfs of Tehuantepec, Papagayo and Panama, where there Sequences were manually aligned in Mega v.6.06 (Tamura et al., are strong seasonal upwelling systems (Fig. 1A; Chelton et al., 2000, 2013) and examined using chromatograms. For the nuclear genes, a 2004; Rodríguez-Rubio et al., 2003; Fiedler and Lavín, 2006; Kessler, Bayesian computational inference of nuclear genes gametic phase was 2006). These oceanographic systems have a great impact on the phy- performed using PHASE v.2.0 (Stephens and Donelly, 2003) module sical, chemical and biological characteristics of parts of the TEP where implemented in the software DNAsp v.5.0 (Librado and Rozas, 2009).
270 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Fig. 1. Sampling locations and biogeographical provinces where Elacatinus puncticulatus is distributed. (A) The biogeographic provinces, the upwelling systems, (a) represent the most common anticyclonic eddies in the entrance of the Gulf of California (taken from Pantoja et al., 2012). (B) Samples locations, the colors correspond to the clade and haplogroups in Figs. 2 and 4, (b) represents the less common cyclonic eddy in the entrance of the Gulf of California is also show (taken from Pantoja et al., 2012). Light blue represent the Tropical Eastern Pacific (TEP). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
271 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Table 1 Samples localities and sequences information.
Site Locality, Municipality and State Country Biogeographic province Sequences number GPS Coordinates Longitude Cytb/RHO/S7 Latitude
1 Isla Tortuga, Mulege, Baja California Sur Mexico Cortez 4/2/2 27.4306389 −111.866194 2 San Marcos, Mulege, Baja California Sur Mexico Cortez 6/0/0 27.2096667 −112.042528 3 Punta Pulpito, Mulege, Baja California Sur Mexico Cortez 2/0/0 26.5145278 −111.443111 4 Bahía Concepción, Mulege, Baja California Sur Mexico Cortez 2/0/0 26.8942222 −111.823778 5 Santa Inés, Mulege, Baja California Sur Mexico Cortez 3/0/0 27.0330833 −111.913639 6 Bahía de los Sueños, La Paz, Baja California Sur Mexico Cortez 3/3/3 23.9936389 −109.825694 7 Barco CSB, Isla Espíritu Santo, La Paz, Baja California Sur Mexico Cortez 3/1/0 24.44 −110.377111 8 El Empachado, Isla Espíritu Santo, La Paz, Baja California Sur Mexico Cortez 10/4/4 24.4468806 −111.377031 9 Playa Sta. María, La Paz, Baja California Sur Mexico Cortez 1/0/0 22.9311667 −109.785083 10 Caleta del Zorro, Guaymas, Sonora Mexico Cortez 1/1/1 27.9370278 −111.094917 11 Isla San Pedro Nolasco, Guaymas, Sonora Mexico Cortez 1/1/1 27.9751667 −111.379083 12 Las Gringas, Guaymas, Sonora Mexico Cortez 10/2/2 27.8854167 −110.954917 13 Santa Catalina, Guaymas, Sonora Mexico Cortez 3/0/0 27.9306556 −111.060372 14 Pescadores La Manga II, Guaymas, Sonora Mexico Cortez 1/1/1 27.9594278 −111.001081 15 Las Cruces, Isla Tiburon, Hermosillo, Sonora Mexico Cortez 2/2/2 28.7587139 −112.343411 16 Punta Carrizales, Manzanillo, Colima Mexico Mexican 6/2/4 19.0952944 −104.427433 17 Barco Hundido, Faro de Bucerias, Aquila, Michoacán Mexico Mexican 4/4/4 18.35245 −103.511469 18 El Chato, Zihuatanejo, Guerrero Mexico Mexican 2/2/1 17.6499139 −101.600086 19 Zacatoso, Ixtapa, Zihuatanejo, Guerrero Mexico Mexican 14/3/3 17.6539806 −101.610964 20 Morros del Potosí, Zihuatanejo, Guerrero Mexico Mexican 11/3/3 17.5341111 −101.469333 21 Las Gatas, Zihuatanejo, Guerrero Mexico Mexican 4/1/1 17.6220611 −101.548058 22 La Virgen, Zihuatanejo, Guerrero Mexico Mexican 6/0/0 17.6516111 −101.595639 23 Morro Ahogado El Tejón, Santa María Huatulco, Oaxaca Mexico Mexican 7/3/3 15.78 −96.0402778 24 La Blanca, Santa María Huatulco, Oaxaca Mexico Mexican 5/3/3 15.7488056 −96.1145556 25 El Arrocito, Santa María Huatulco, Oaxaca Mexico Mexican 8/0/2 15.7488889 −96.0708333 26 Isla Montosa, Santa María Huatulco, Oaxaca Mexico Mexican 8/0/1 15.7629167 −96.0813889 27 Isla Cabo Blanco, Península de Nicoya, Puntarenas Costa Rica Panamic 10/5/5 9.54405556 −85.0861389 28 Los Frailes, Península de Azuero, Provincia Los Santos Panama Panamic 9/5/6 7.34766667 −80.1301389 29 Archipielago de Las Perlas, Provincia de Panamá Panama Panamic 8/3/7 8.641416667 −79.0411111 30 El Pelado, Santa Elena, Guayaquil Ecuador Panamic 5/3/3 −1.93597222 −80.7778056 31 Isla Salango, Manabí, Guayaquil Ecuador Panamic 10/1/1 −1.62044444 −80.8068056 32 Ballena, Bahía Salinas, Santa Elena Ecuador Panamic 2/0/0 −2.19652778 −80.9424444 33 Isla Salango, Puerto López, Manabí Ecuador Panamic 5/1/1 −1.59205556 −80.8366944 34 Punta Mala, Puerto López, Manabí Ecuador Panamic 3/1/1 −1.56527778 −80.8262778 35 Isla de la Plata, Puerto López, Manabí Ecuador Panamic 3/0/0 −1.28401667 −81.0499083 36 Pozas la Chocolatera, Salinas, Santa Elena Ecuador Panamic 1/1/1 −2.19014722 −80.989625 37 La Pared del Viejo, El Pelado, Santa Elena Ecuador Panamic 7/1/1 −1.93445278 −80.7777361 38 Bajo 40, El Pelado, Santa Elena Ecuador Panamic 3/0/0 −1.93541111 −80.7781 39 La Roñosa, Nuqui, El Choco Colombia Panamic 9/3/9 5.58627778 −77.492278
Because sequences of S7 showed heterozygous position defined by in- 2005). dels, the gametic phase reconstruction followed the procedure de- A Bayesian Inference analysis was carried out in MrBayes v.3.2.1 scribed by Sousa-Santos et al. (2005). The pairwise homoplasy index in (Ronquist et al., 2012), for 10 million generations, with two in- both nuclear genes was calculated with the software Splitstree v.4.13 dependent runs implementing four Markov Chain Monte Carlo (MCMC) (Huson and Bryant, 2006). We found no significant evidence of re- processes and sampling every 500 generations. The substitution model combination in either of those genes (RHO, p = 0.5; S7, p = 0.45). All was assigned for each gene according to PartitionFinder v.1.1.0 sequences were deposited in GenBank under the accession numbers (Lanfear et al., 2012). We evaluated chain convergence with the log- (cytb: MH814154-MH814355; RHO: MH814078-MH814153; and S7: likelihood (−InL) values of two independent runs on Tracer v.1.5 MH813998-MH814077) (see Table S2). (Rambaut and Drummond, 2007), discarding the initial 10% of the generations (burn-in) to construct the consensus tree. Species-tree analyses under a multispecies coalescent model were 2.3. Phylogenetic analyses and species tree performed using *BEAST v.1.8.1 (Heled and Drummond, 2010). Here we tested two different hypotheses: (i) H1 assumes that each major DNA sequences of each gene (cytb, RHO and S7) were collapsed to clade and sub-clade recovered in the phylogenetic analyses represents a haplotypes using the web-based program ALTER (González-Peña et al., different species: (a) the Gulf of Panama, (b) Ecuador, West Panama 2010). Model selection was based on Akaike information criteria (AIC) and Costa Rica, and (c) Mexico; and (ii) H2 considers groups within and optimal partition setting analysis was conducted using Parti- those sub-clades as separate species. The model parameters were un- tionFinder v.1.1.0 (Lanfear et al., 2012). Optimal partition setting was linked across cytb, RHO and S7. Substitution models were set according obtained by assigning one substitution model to each gene, using model to PartitionFinder v.1.1.0 (Lanfear et al., 2012). We applied a log- GTR + G for cytb, model TrN + I + G for RHO, and model normal relaxed clock (uncorrelated) model on branch length and cali- TVM + I + G for S7. The phylogenetic analyses were performed with brated the cytb partition using the mutation rate of cytb in teleosts of the concatenated dataset (cytb, RHO and S7). A Maximum Likelihood 0.76–2.2%/million years (Zardoya and Doadrio, 1999; Machordom and tree was calculated with RAxMLGUI v.1.3.1 (Silvestro and Michalak, Doadrio, 2001; Near and Benard, 2004). The evolutionary rate of RHO 2012; Stamatakis, 2014), using the GTR + Gamma + I substitution and S7 genes were estimated relative to the cytb gene. Due to the high model, and 10,000 bootstrap replicates. For the outgroup, we used divergences found between populations we used the tree prior: Yule Elacatinus randalli (Bohlke and Robins, 1968), one of the group of West Process model (Gernhard, 2008) and estimated a starting tree using the Atlantic Elacatinus species that is sister to E. puncticulatus and Tigrigobius random method. Markov Chain Monte Carlo analysis was run for 50 saucrus (Robins 1960), a sister genus of Elacatinus (Taylor and Hellberg,
272 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285 million generations, sampled every 1000 generations. The chains con- obtained by PartitionFinder. The molecular clock was also calibrated vergence was evaluated with the -InL values in Tracer v.1.5 (Rambaut using the mutation rate of cytb in teleosts of 0.76–2.2%/million years and Drummond, 2007), and the results were summarized using (Zardoya and Doadrio, 1999; Machordom and Doadrio, 2001; Near and TreeAnnotator v.1.8.1 (Drummond et al., 2012). Benard, 2004). An uncorrelated relaxed clock model, and 50 million generations, sampling every 500 generations. Convergence was as- 2.4. Haplotype networks sessed with Tracer v1.5 (Rambaut and Drummond, 2007).
In order to determine the geographic distribution of haplotypes for 2.8. Ancestral area reconstruction all populations of E. puncticulatus, three independent phylogenetic network estimations were reconstructed using statistical parsimony In order to reconstruct the ancestral area for all populations of E. (TCS), corresponding to sequences of each gene (cytb, RHO and S7) puncticulatus, a statistical Dispersal-Vicariance (S-DIVA) method (Yu using PopArt v.1.7 (http://popart.otago.ac.nz). et al., 2010), and the Dispersal-Extinction-Cladogenesis (DEC) model (Ree et al., 2005; Ree and Smith, 2008) were performed in RASP v.3.2 2.5. Genetic variation and genetic structure software (Yu et al., 2015). The ultrametric and dichotomous tree ob- tained in BEAST for estimated the divergence time was used as the tree Genetic diversity was estimated by calculating the numbers of topology on which ancestral areas were mapped. In both analyses, the haplotypes (Hn) and polymorphic sites (S), nucleotide diversity (π), and number of maximum areas was kept as two. The analyses were per- haplotype diversity (h), at different hierarchical levels using the soft- formed selecting six areas established in relation to the results of the ware Arlequin v.3.5.1.3 (Excoffier and Lischer, 2010). phylogenetic analyses: A. Cortez province; B. Mexican province; C. Genetic differentiation levels among groups of E. puncticulatus were Costa Rica and Western Panama; D. Ecuador; E. Gulf of Panama and F. estimated with paired test fixation indices (ΦST) for each gene. A El Choco, Colombia. Bonferroni correction (Rice, 1989) was applied to each p-value obtained in the paired test of genetic differentiation. Genetic structure was es- 3. Results timated via analysis of molecular variance (AMOVA) with 10,000 permutations implemented in the software Arlequin v.3.5.1.3. Both 3.1. Dataset information analyses were implemented for two different groupings: (1) the Cortez, Mexican and Panamic provinces recognized by Ekman (1953); Walker A total of 1140 base pairs (bp) were obtained for the cytb gene, out (1960), and Hastings (2000); and (2) the four major clades and sub- of which 788 were constant, 352 were variable and 182 were parsi- clades (a) Gulf of Panama and El Choco, Colombia, (b) Ecuador, (c) mony-informative sites. For the 816 bp of RHO, 795 were constant, 21 Costa Rica and western Panama, (d) Cortez and the Mexican Coast. were variable and 20 were parsimony-informative sites; while for 735 bp of S7, 719 were constant, 15 variable and 11 were parsimony- 2.6. Divergence-times estimation and genetic distances informative sites.
The program BEAST v.1.8.1 (Drummond et al., 2012) was used to 3.2. Phylogenetic analyses and species tree construction estimate the most recent common ancestor (MRCA) for the main ge- netic groups of E. puncticulatus. The molecular clock was calibrated The phylogenetic analyses recovered two highly supported clades using the mutation rate of cytb in teleosts of 0.76–2.2%/million years for all populations of E. puncticulatus. One comprised individuals from (Zardoya and Doadrio, 1999; Machordom and Doadrio, 2001; Near and the Gulf of Panama (the Perlas Archipelago) plus El Choco, Colombia Benard, 2004). The evolutionary rate of the RHO and S7 genes was (Clade I; HgI), and the second clade grouped individuals from the rest of estimated relative to the cytb gene. The model parameters were un- the TEP (Clade II; HgII, HgIII and HgIV) (Fig. 2). This second clade is linked across cytb, RHO and S7 genes, and substitution models were set divided in two subclades (IIa and IIb). Sub-clade IIa clustered in- according to the selected model of each gene by PartitionFinder v.1.1.0 dividuals from the remainder of the Panamic province but segregated (Lanfear et al., 2012). A lognormal relaxed clock (uncorrelated) model them among two groups, one from Ecuador (HgII) and the other from was selected (Drummond et al., 2006). Due to the high divergences Costa Rica and western Panama (HgIII). Sub-clade IIb grouped in- found between populations, we used the tree prior: Yule Process model dividuals from both the Cortez and Mexican provinces (HgIV) (Figs. 1B (Gernhard, 2008), and estimated a starting tree using the random and 2). method. An MCMC analysis was conducted with 50 million generations, For the species-tree analyses, the arrangement of four putative sampling every 1000 generations. Whether parameter values had species was only weakly supported (< 0.61), whereas that for three reached effective sample size (> 200) and convergence was assessed in putative species was much more strongly supported (> 0.90). One of Tracer v.1.5. (Rambaut and Drummond, 2007). To build the maximum those three species spanned the Cortez and Mexican provinces, another clade credibility tree, we discarded the first 10% of the trees as burn-in comprised individuals from the Gulf of Panama (the Perlas Archipelago using TreeAnotator v.1.8.1 (Drummond et al., 2012). For these analyses and Choco, Colombia), and the third clustered individuals from Costa were used several species of Elacatinus and Tigrigobius genus from the Rica, Western Panama and Ecuador (Fig. 3). tropical West Atlantic (see Table S3) The uncorrected genetic distances between the recovered clades 3.3. Haplotype networks obtained in phylogenetic analyses were calculated for cytb, RHO and S7 in Mega v.6.06 (Tamura et al., 2013), using a bootstrapping process From the 202 sequences obtained for cytb, 101 haplotypes were performed with 100 repetitions. obtained. The haplotype network showed four haplogroups, which are in agreement with the phylogenetic results. The first one is represented 2.7. Demographic history of Elacatinus puncticulatus by the populations of the Gulf of Panama and El Choco, Colombia (HgI) separated by at least 73 mutational steps from all other haplotypes. The In order to infer the population-size fluctuations through time of second haplogroup includes specimens from Ecuador (HgII) and is se- each of the recovered major clades and sub-clades of E. puncticulatus, a parated by at least seven mutational steps from the third haplogroup, Coalescent Bayesian skyline plot analysis with the cytb gene which include the samples from Costa Rica and western Panama (Drummond et al., 2005), was implemented in BEAST v.1.8.1 (HgIII). These three haplogroups belong to the Panamic province. The (Drummond et al., 2012). The substitution model used was that final haplogroup is represented by individuals from the Mexican and
273 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
26525_Las Perlas_Panama_Panamic 50706_El Choco_Colombia_Panamic 26737B_Las Perlas_Panama_Panamic Gulf of Panama 26738_Las Perlas_Panama_Panamic Clade I 50701_El Choco_Colombia_Panamic and El Choco, Colombia 100/100 50705_El Choco_Colombia_Panamic HgI 50705B_El Choco_Colombia_Panamic (Panamic province) 26537_Las Perlas_Panama_Panamic 26737_Las Perlas_Panama_Panamic 17209_Ecuador_Panamic 17210_Ecuador_Panamic 17878_Ecuador_Panamic 18464_Ecuador_Panamic 25909_Ecuador_Panamic HgII, Ecuador 25919_Ecuador_Panamic 25919B_Ecuador_Panamic (Panamic province) 25980_Ecuador_Panamic 25988_Ecuador_Panamic 99/85 21586_Costa Rica_Panamic Sub-clade IIa 100/100 21590_Costa Rica_Panamic 21590B_Costa Rica_Panamic 21606_Costa Rica_Panamic 21607_Costa Rica_Panamic Panama and HgIII 22283_Panama_Panamic Costa Rica 100/93 22283B_Panama_Panamic 22284_Panama_Panamic (Panamic province) 22284B_Panama_Panamic 22289_Panama_Panamic 22285_Panama_Panamic 8102_Mexico_Mexican 19101_Mexico_Mexican Clade II 19827_Mexico_Mexican 100/100 19955_Mexico_Mexican 19956_Mexico_Mexican 19956B_Mexico_Mexican 20015_Mexico_Mexican 20015B_Mexico_Mexican 13208_Mexico_Mexican 19099_Mexico_Mexican 19300_Mexico_Mexican 19300B_Mexico_Mexican 19828_Mexico_Mexican 19957_Mexico_Mexican Cortez and 19957B_Mexico_Mexican 20122_Mexico_Mexican Sub-clade IIb Mexican coast 19832_Mexico_Mexican 12452_Mexico_Mexican HgIV (Cortez and Mexican 100/96 14012_Mexico_Mexican 20013_Mexico_Mexican provinces) 11037_Mexico_Cortez 11196_Mexico_Cortez 15715_Mexico_Cortez 15774_Mexico_Cortez 15774B_Mexico_Cortez 37396_Mexico_Cortez 37396B_Mexico_Cortez 15716_Mexico_Cortez 15717_Mexico_Cortez 12445_Mexico_Cortez 14006_Mexico_Mexican 19100_Mexico_Mexican 11032_Mexico_Cortez 14716_Mexico_Cortez 100/100 Elacatinus randalli Tigrigobius saucrus
0.02
Fig. 2. Bayesian inference tree of Elacatinus puncticulatus from concatenated sequences of the three genes. Bayesian posterior probabilities (> 90%) and maximum likelihood bootstrap values (> 80%) (BPP/MLB) are indicated.
Cortez provinces (HgIV), and is separated by at least 23 mutation steps network and the phylogenetic analyses showed the highest haplotype from other haplotypes in groups HgII and HgIII. and nucleotide diversity in Cortez - Mexican sub-clade (HgIV), followed Within the HgIV haplogroup some asymmetrical segregation be- by the Costa Rica - western Panama sub-clade (HgIII). tween the Mexican and Cortez provinces was evident, with some hap- The highest genetic diversity of RHO was found for Gulf of Panama - lotypes found in the Mexican province being closely related to haplo- El Choco, Colombia clade (HgI), followed by the clade from Ecuador types in the Cortez province, but no haplotypes found in the Cortez sub-clade (HgII). Finally, for S7, the highest genetic diversity was found province being closely related to Mexican haplotypes (Fig. 4). As ex- for the Gulf of Panama - El Choco, Colombia clade (HgI), followed for pected (Taillebois et al., 2013), fewer mutational steps separated the Costa Rica - western Panama from the Panamic province (HgIII).The different phylogenetic groups in nuclear genes and some admixture was lowest genetic diversity found in cytb was for the Ecuador sub-clade, found. For S7 (Fig. 5a) three haplogroups were recovered (HgI, while for RHO and S7 the lowest genetic diversity was found in the HgII + HgIII and HgIV), with one mutational step between them. In Cortez - Mexican sub-clade. addition, haplogroups HgII and HgIII comprised a single group, and one sequence from Ecuador is grouped with three sequences from the ff Cortez province. For RHO (Fig. 5b) all haplogroups share haplotypes 3.4.2. Genetic di erentiation (Table 3) ff Φ between them, while some haplotypes of HgI and HgIII were unique A paired test of genetic di erentiation ( ST) among haplogroups of and peripheral (Fig. 5b). E. puncticulatus for cytb ranged between 0.327 and 0.984, for RHO from 0.048 and 0.545, and for S7 from −0.018 to 0.780, all statistically significant comparisons except the comparison of Cortez province vs. 3.4. Genetic variation and genetic structure Mexican province for RHO, and Ecuador vs. Costa Rica - western Panama of the Panamic province for S7. The highest values for cytb 3.4.1. Haplotype groups (Table 2) were observed between the Gulf of Panama-El Choco (Colombia) and The genetic diversity calculated for each haplogroup in the cytb both the Cortez province and Ecuador haplogroups. For RHO the
274 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
)a )b Sub-clade IIb = Sp. 4 Sub-clade IIb = Sp. 3
0.81 0.90 Hg III = Sp. 3
0.61 Sub-clade IIa Sub-clade IIa = Sp. 2
1 1 Hg II = Sp.
Clade I = Sp. 1 Clade I = Sp. 1 0.3 0.4
Fig. 3. Species tree, (a) three putative species, considering the two clades, and the two recovered sub-clades within clade II, (b) four putative species, considering the segregation obtained within sub-clade IIa. The posterior probabilities are show above the branches.
10 samples
1 sample
BCS-Cortez Sonora-Cortez Hg III Colima-Mexican Michoacan-Mexican 4 Hg IV Guerrero-Mexican Oaxaca-Mexican 6 CostaRica-Panamic 13 Panama-Panamic Gulf of Panama-Panamic Ecuador-Panamic El Choco-Colombia-Panamic 8 Hg II 64
Hg I
Fig. 4. Haplotypes network of cytb. Size of the circles indicate the frequency of the haplotype and colors correspond to the regions were the samples belong as show in the box. Numbers inside circles correspond to mutation steps. White circles correspond to hypothetical haplotypes. Dash lines correspond to mutation steps. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
275 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
)a )b Hg I
Hg IV
Hg II + Hg III BCS-Cortez CostaRica-Panamic Sonora-Cortez Panama-Panamic 10 samples Colima-Mexican Gulf of Panama-Panamic 1 sample Michoacan-Mexican Ecuador-Panamic Guerrero-Mexican El Choco-Colombia-Panamic Oaxaca-Mexican
Fig. 5. Haplotypes networks, (a) of S7 gene, (b) of RHO gene. Size of the circles indicate the frequency of the haplotype and colors correspond to the regions were the samples belong as show in the box. White circles correspond to hypothetical haplotypes. Dash lines correspond to mutation steps. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) highest value was between Cortez and Ecuador haplogroups, and for S7 have occurred ca. 19.25 million years ago (Mya) (95% HPD: the highest value found was between the Mexican province and 12.44–26.06 Mya), while the cladogenetic event within E. puncticulatus Ecuador sub-clades. The lowest values for each gene were found be- that separated clades I (Gulf of Panama - El Choco, Colombia) and II (all tween the Cortez and Mexican provinces. the remainder) was estimated to have occurred at the junction of the Miocene and Pliocene ca. 4.88 (Mya) (95% HPD: 2.73–7.04 Mya). The 3.4.3. Molecular variance (Table 4) cladogenetic event that formed sub-clades IIa (remainder of the Analyses of molecular variance for cytb, RHO and S7 detected sta- Panamic Province) and IIb (Cortez and Mexican provinces) was esti- tistically significant higher levels of variation when individuals were mated to have occurred at about the start of the Pleistocene ca. 1.51 – grouped according to haplogroup clades and sub-clades, than when Mya (95% HPD: 0.75 2.27 Mya). Within sub-clade IIa, the separation of – they were grouped according to biogeographic provinces. the Ecuador (HgII) and Costa Rica western Panama groups (HgIII), was estimated at ca. 0.53 Mya (95% HPD: 0.17–0.89 Mya) (Fig. 6). The uncorrected mean genetic distances (p-distances) calculated 3.5. Estimation of divergence time and genetic distances between the various clades and subclades for cytb gene ranged from 0.8 to 7.3% (Table 5). The smallest genetic distance occurred between The cladogenetic event that separated the West Atlantic species of Ecuador (HgII) and Costa Rica - western Panama (HgIII) samples, and Elacatinus from the ancestor of E. puncticulatus (TEP) was estimated to
Table 2 Genetic diversity according with the phylogenetic arrangement of Elacatinus puncticulatus for the three genes. N, sample size, S, polymorphic sites, H, number of haplotypes, π, nucleotide diversity h, haplotype diversity.
Phylogenetic arrangement N S H Π h
cytb Las Perlas-Panama and El Choco-Colombia-Panamic 17 9 9 0.001 ± 0.000 0.735 ± 0.117 Ecuador-Panamic 39 16 16 0.0008 ± 0.000 0.712 ± 0.081 Costa Rica-Panama-Panamic 19 14 12 0.001 ± 0.000 0.836 ± 0.086 Cortez-Mexican 127 66 64 0.003 ± 0.001 0.943 ± 0.013
RHO Las Perlas-Panama and El Choco-Colombia-Panamic 14 5 6 0.001 ± 0.001 0.835 ± 0.070 Ecuador-Panamic 16 4 5 0.001 ± 0.001 0.800 ± 0.065 Costa Rica-Panama-Panamic 20 2 3 0.0006 ± 0.0006 0.468 ± 0.104 Cortez-Mexican 78 8 10 0.0006 ± 0.0006 0.452 ± 0.066
S7 Las Perlas-Panama and El Choco-Colombia-Panamic 32 8 7 0.001 ± 0.001 0.629 ± 0.088 Ecuador-Panamic 16 5 4 0.001 ± 0.001 0.525 ± 0.136 Costa Rica-Panama-Panamic 22 6 5 0.001 ± 0.001 0.588 ± 0.114 Cortez-Mexican 82 6 7 0.0008 ± 0.0007 0.419 ± 0.066
276 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Table 3 3.7. Demographic history of Elacatinus puncticulatus
Genetic differentiation using pairwise ΦST for cytb, RHO and S7 among re- covered groups in phylogenetic analyses. The comparisons that were not sig- The Coalescent Bayesian skyline plot analyses for each of the four fi ni cant after Bonferroni correction (p < 0.05), highlighted in bold. main clades and subclades of E. puncticulatus indicates they have had Cortez Mexican Ecuador Costa Rica stable effective population sizes following an expansion that occurred province province and Panama about 170–130 Kya for all clades and sub-clades (Fig. 8).
cytb Cortez province 4. Discussion Mexicana province 0.327 Ecuador 0.955 0.887 The heterogeneity of habitat distribution, the history of develop- Costa Rica and Panama 0.951 0.877 0.866 ment of the Central American isthmus and the spatially- and temporally Las Perlas-Panama and 0.984 0.958 0.984 0.980 highly dynamic nature of oceanographic conditions within the TEP El Choco-Colombia have promoted the development of unusual evolutionary processes in RHO many types of organisms that form the biota of that region (Hurtado Cortez province Mexicana province 0.048 et al., 2007; Duda and Lessios, 2009; Lessios and Baums, 2017). To date Ecuador 0.545 0.438 these processes have received little research attention, and remain Costa Rica and Panama 0.517 0.296 0.291 poorly understood, particularly among reef- and other types of marine Las Perlas-Panama and 0.223 0.222 0.319 0.449 fishes (Stepien et al., 2001; Szabó et al., 2014; Lessios and Baums, El Choco-Colombia 2017). S7 The phylogenetic (Fig. 2) and biogeographic (Figs. 4, 5 and 7) Cortez province patterns that we found in E. puncticulatus show that it has had a complex Mexicana province 0.193 Ecuador 0.625 0.784 spatial and temporal evolutionary history, with several events of dis- Costa Rica and Panama 0.645 0.776 −0.018 persal and genetic isolation (Fig. 7), that promoted cladogenetic events Las Perlas-Panama and 0.630 0.780 0.657 0.679 and the formation of well-differentiated genetic groups within what has El Choco-Colombia been regarded as a single species (Figs. 2 and 6). This is the first study to report substantial genetic structure in a widely distributed shore-fish species along the continental shore of the TEP, among either the re- the greatest between clade I (Gulf of Panama - El Choco, Colombia) and gional endemics or non-endemics (Muss et al., 2001; Craig et al., 2006; sub-clade IIb (HgIV; Cortez and Mexican provinces). Distances were Bernardi et al., 2014; Lessios and Baums, 2017). Our results and ana- also high between IIb (HgIV) and both HgIII (western Panama - Costa lyses (Tables 3 and 4) indicate that local oceanographic processes are Rica samples) and HgII (Ecuador samples). For nuclear genes, the ge- more important than the largest habitat discontinuities in the region for netic distances between the various clades and sub-clades ranged be- the genetic differentiation and isolation of populations of this species. tween 0.1 and 0.5% (Table 6). For RHO the highest value was obtained Similar results have been found for both fishes (e.g. Henriques et al., between clade I/HgI (Gulf of Panama - Choco, Colombia) and HgII 2012, 2014) and invertebrates in other dynamic marine ecosystems (Ecuador). For S7 the genetic distances ranged between 0.1 and 0.5%, (White et al., 2010), in which it has been proposed that genetic struc- with the highest value between clades I/HgI (Gulf of Panama - El ture is associated with coastal oceanographic features such as upwel- Choco, Colombia) and HgIII (Costa Rica – western Panama). ling, ocean currents and fronts, as well as with local environmental gradients. 3.6. Ancestral area reconstruction 4.1. Biogeographic scenario The analyses of ancestral-area reconstruction using S-DIVA and DEC 4.1.1. Isolation of the Gulf of Panama clade (Fig. 7) showed low resolution for basal nodes (DEC = 26.79% prob- Elacatinus puncticulatus is sister to all Atlantic species of Elacatinus ability; S-DIVA = 33.33% probability). However, both analyses placed (Rüber et al., 2003). Our analysis indicates that E. puncticulatus and the the Gulf of Panama as an ancestral area for E. puncticulatus. The results 13 species of Atlantic Elacatinus included in that analysis became se- of these two methods produced some differences in events relating to parated at ca. 19.25 Mya (95% HPD: 12.44–26.06 Mya). This is very this basal node. In S-DIVA, 1 = dispersal events from the Gulf of Pa- similar to the 18 Mya separation time estimated by Rüber et al. (2003), nama, Costa Rica and western Panama towards the Mexican province and also similar to the time of separation (ca. 14 Mya) between Atlantic were followed by a vicariant event that separated the Mexican province and Pacific species of the reef fish genus Apogon estimated by Piñeros and Costa Rica - western Panama from the Gulf of Panama (marginal et al. (in press). Both of these isolation events occurred long before the probability = 0.166), while, for DEC, 2 = dispersal events from Gulf of final closure of the Isthmus of Panama, ∼3 Mya (Bacon et al., 2015). Panama and Mexican province toward Costa Rica, western Panama and For the two biogeographic analyses we performed within the E. El Choco, Colombia followed by vicariant events that separated the puncticulatus groups, there was congruence in the occurrence of Gulf of Mexican province and Costa Rica - western Panama from the Gulf of Panama as the basal node and ancestral area for Elacantinus puncticu- Panama - El Choco, Colombia (marginal probability = 0.091). For DEC latus (Fig. 7). Such a spatial pattern has been proposed for other tran- and S-DIVA the rest of the biogeographical routes followed the same sisthmian sister species (Lessios, 2008). historical pattern: 3 = Dispersal towards Ecuador from Costa Rica - The isolation of the Gulf of Panama – El Choco, Colombia clade from western Panama and the Mexican province were followed by a se- the clade comprising all other populations within E. puncticulatus was paration of the Mexican province from Ecuador and Costa Rica – wes- dated at about 4.88 Mya. The isolation of the Gulf of Panama – El Choco tern Panama (DEC = 0.501, S-DIVA = 0.500); and, subsequently, area seems surprising given that the western Panama sampling site is 4 = the Ecuador population separated from Costa Rica - western Pa- only 140 km from the Perlas Archipelago (Gulf of Panama) sampling nama (DEC = 1.0, S-DIVA = 1.0). Event 5 = dispersal from the Mex- site, and only 30 km west of the western edge of the Gulf of Panama, but ican province to the Cortez province (DEC = 0.681, S-DIVA = 1.0). belongs to a very different and divergent genetic group (Sub-clade IIa, Finally, event 6 was a dispersal event toward El Choco, Colombia from HgIII, with a cytb genetic distance from the Gulf of Panama Gulf of Panama (DEC = 0.521, S-DIVA = 1.0) (Fig. 7). clade = 7.1%). The Gulf of Panama is located on the inner part of a substantial area
277 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Table 4 Analyses of molecular variance for groups according to: (1) biogeographic provinces sensu Hastings (2000); (2) phylogenetic analyses (1 = Las Perlas, Panama, Panamic province; 2 = Ecuador, Panamic province; 3 = Costa Rica and Panama of Panamic province; 4 = Cortez and Mexican provinces); and (3) phylogenetic analyses, considering Cortez and Mexican provinces as separated groups, for the three genes. Ns = Not significant.
Testing assumptions Source of variation % of variance Fixation index P-value
Cytb
(1) Cortez, Mexican and Panamic Among groups 35.56 ΦCT: 0.35 Ns
Among populations within 57.14 ΦSC: 0.88 < 0.001 groups
Within populations 7.30 ΦST: 0.92 < 0.001 Total 100
(2) Las Perlas-El Choco-Panamic, Ecuador-Panamic, Costa Rica-Panama-Panamic, Cortez- Among groups 92.65 ΦCT: 0.92 < 0.001
Mexican Among populations within 2.10 ΦSC: 0.28 < 0.001 groups
Within populations 5.25 ΦST: 0.94 < 0.001 Total 100
(3) Las Perlas-El Choco-Panamic, Ecuador-Panamic, Costa Rica-Panama-Panamic, Cortez Among groups 92.51 ΦCT: 0.93 < 0.001
province, Mexican province Among populations within 0.63 ΦSC: 0.08 < 0.001 groups
Within populations 6.86 ΦST: 0.93 < 0.001 Total 100
RHO
(1) Cortez, Mexican and Panamic Among groups 10.65 ΦCT: 0.10 Ns
Among populations within 20.38 ΦSC: 0.22 < 0.001 groups
Within populations 68.97 ΦST: 0.31 < 0.001 Total 100
(2) Las Perlas-El Choco-Panamic, Ecuador-Panamic, Costa Rica-Panama-Panamic, Cortez- Among groups 37.96 ΦCT: 0.38 < 0.001
Mexican Among populations within 0.74 ΦSC: 0.01 Ns groups
Within populations 61.30 ΦST: 0.38 < 0.001 Total 100
(3) Las Perlas-El Choco-Panamic, Ecuador-Panamic, Costa Rica-Panama-Panamic, Cortez Among groups 32.83 ΦCT: 0.32 < 0.001
province, Mexican province Among populations within −0.17 ΦSC: −0.002 Ns groups
Within populations 67.34 ΦST: 0.32 < 0.001 Total 100
S7
(1) Cortez, Mexican and Panamic Among groups 36.42 ΦCT: 0.36 Ns
Among populations within 33.35 ΦSC: 0.52 < 0.001 groups
Within populations 30.23 ΦST: 0.69 < 0.001 Total 100
(2) Las Perlas-El Choco-Panamic, Ecuador-Panamic, Costa Rica-Panama-Panamic, Cortez- Among groups 67.04 ΦCT: 0.67 < 0.001
Mexican Among populations within 5.91 ΦSC: 0.17 < 0.001 groups
Within populations 27.04 ΦST: 0.72 < 0.001 Total 100
(3) Las Perlas-El Choco-Panamic, Ecuador-Panamic, Costa Rica-Panama-Panamic, Cortez Among groups 64.07 ΦCT: 0.64 < 0.001
province, Mexican province Among populations within 5.26 ΦSC: 0.14 < 0.001 groups
Within populations 30.67 ΦST: 0.69 < 0.001 Total 100 of continental shelf at the head of the Gulf of Panama. In this area jimenezi and Notarius lentiginosus). Further, various fishes known from seasonal winds move surface waters offshore, producing strong surface western Panama do not occur in the Gulf of Panama (Robertson and upwelling, which dramatically lowers sea temperatures (as much as Allen, 2015). These differences in distributions between western Pa- 15 °C), oxygen and pH levels and produces large inputs of nutrients nama and the Gulf of Panama – El Choco area are consistent with the (Glynn et al., 1972; D’Croz and Robertson, 1997). The effect of this hypothesis that strong local variation in ecological conditions can de- upwelling system extends to and further south than the Choco area in termine not only the nature of fish assemblages, but also connectivity Colombia (Devis-Morales et al., 2008; GA Castellano-Galindo pers. com. between closely related populations. Upwelling conditions have pro- to DRR July 9, 2018, concerning upwelling conditions he has experi- duced local adaptation and genetic divergence has occurred in organ- enced at El Choco). In contrast Western Panama and Northern Ecuador isms found throughout the coast of Panama. For example, in the coral lack such surface upwelling. Upwelling in Panama bay has strong in- Pocillopora damicornis, in which there are differences in genetic struc- fluence on the nature of local reef biota in the Perlas Archipelago ture of populations in the Gulf of Panama and western Panama, corals (Glynn and Stewart, 1973; Abele, 1979; Glynn et al., 2017). In addition from those sites differ significantly in their thermal tolerance to up- there are distinct differences in the marine faunas of Gulf of Panama welling conditions (Glynn and D’Croz, 1990; D’Croz and Mate, 2004; and western Panama. Some species are known only from the Gulf of Combosch and Vollmer, 2011). The Gulf of Panama also has a large Panama, including some soft corals (H. Guzman pers. com. to DRR) and gyre that is influenced by the wind regime, and likely transports marine various fishes: two chaenopsids (Protemblemaria perla and Tanyem- larvae well offshore (Rodríguez-Rubio et al., 2003), reducing the po- blemaria alleni), a sciaenid (Paranebris bauchotae), two microdesmids tential for connectivity between populations in the Perlas Islands and (Microdesmis hildebrandi and M. multiradiatus), and two ariids (Ariopsis western Panama.
278 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Elacatinus of the Atlantic ocean
19.25 Mya HPD 95% [12.44-26.06] HgI, Gulf of Panama and El Choco, Colombia (Panamic province)
4.88 Mya HPD 95% [2.73-7.04] HgIV, Mexican and Cortez provinces
1.51 Mya HPD 95% [0.75-2.27]
HgII, Ecuador (Panamic province)
0.53 Mya HPD 95% [0.17-0.89] HgIII, Panama and Costa Rica (Panamic province) 3.0 25,0 20,0 15,0 10,0 5,0 0,0
Fig. 6. Chronogram showing the estimated divergence times. Branch lengths are proportional to the divergence times. The value at each node is the mean estimated divergence time (in Mya), with the 95% confidence interval in parentheses.
Table 5 Uncorrected p genetic distances within (bold numbers) and between recovered clades and sub-clades based on cytb for Elacatinus punticulatus.
Cytb Clade I/HgI (Las Perlas- Sub-clade IIa/HgII Sub-clade IIa/HgIII (CR- Sub-clade IIb/HgIV (Cortez and Colombia) (Ecuador) Panama) Mexican)
Clade I/HgI (Las Perlas-Colombia) 0.1% Sub-clade IIa/HgII (Ecuador) 6.8% 0.1% Sub-clade IIa/HgIII (CR-Panama) 7.1% 0.8% 0.1% Sub-clade IIb/HgIV (Cortez and Mexican) 7.3% 2.2% 2.4% 0.3%
Thus the large gyre, the upwelling system and the current regime in explain much of the gobiid diversity in the Neotropics (Sato et al., 1999; the Gulf of Panama – El Choco area (Figs. 1A and S1) may have pro- Streelman and Danley, 2003). Allopatric distributions of many pairs of duced local adaptation and genetic divergence in Elacatinus puncticu- Elacatinus sister species in the Greater Caribbean suggests that geo- latus in response to local ecological conditions while promoting isola- graphic speciation at small spatial scales may be the most common tion through larval transport offshore (see also discussion of the mode of speciation in this genus (Taylor and Hellberg, 2005). The si- Sinaloan Gap, below). Adaptation to different ecological conditions tuation of the Gulf of Panama population of E. puncticulatus re- works as an important factor that fuels the formation and maintenance presenting a very distinctive lineage living in a highly stressful en- of species (Schluter, 2009; Prada and Hellberg, 2013; Nosil et al., vironment may be another case in point. Such a lineage would not 2005). For a number of organisms upwelling conditions in Gulf of Pa- necessarily have to be completely isolated from dispersal from neigh- nama are stressful – some reef-fishes halt reproduction (Robertson, boring lineages (subclades IIa and IIb) if individuals from those popu- 1990) during upwelling season, as do some invertebrates (Collin and lations are unable to cope with environmental conditions in the Gulf of Ochoa, 2016; Collin et al., 2018). Adaptation to such conditions would Panama. Rocha et al. (2005) found both geographical and within-site be obligatory for an organism restricted to the upwelling area, as E. habitat separation of lineages of a labrid reef fish in the tropical west puncticulatus clade I appears to be. Neotropical gobies exhibit patterns Atlantic and proposed that such can lead to ecological speciation (see of ecological differentiation relating to habitat usage (Rüber et al., also discussion relating to the Sinaloan Gap, below). 2003), and adaptive radiation at small geographic scales has been proposed to account for phylogenetic patterns of the genus Elacatinus, fl which has at least 21 species in the wider Caribbean area, and may 4.1.2. In uence of large habitat discontinuities: The two sandy-shore gaps The Central American and Sinaloan gaps have been proposed to
Table 6 Uncorrected genetic distances within (bold numbers) and between recovered clades and sub-clades based on nuclear genes RHO and S7 for Elacatinus punticulatus, the italic letters corresponded to the genetic distances with RHO.
RHO/S7 Clade I/HgI (Las Perlas- Sub-clade IIa/HgII Sub-clade IIa/HgIII (CR- Sub-clade IIb/HgIV (Cortez and Colombia) (Ecuador) Panama) Mexico)
Clade I/HgI (Las Perlas-Colombia) 0.2%/0.1% 0.4% 0.5% 0.3% Sub-clade IIa/HgII (Ecuador) 0.3% 0.2%/0.1% 0.1% 0.2% Sub-clade IIa/HgIII (CR-Panama) 0.2% 0.2% 0.1%/0.2% 0.3% Sub-clade IIb/HgIV (Cortez and Mexico) 0.2% 0.2% 0.1% 0.1%/0.1%
279 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
5 5 DEC S-DIVA EF E E E EF EF EF F E EF EF F Hg I F F
B B
B B B B
B B BE B DE B B B 2 1 Hg IV B B B B
B B
AB A AB A A A A A BD BD
3 A 3 A
C C C C Hg II C C
CD CD D D 4 D 4 D Hg III D D
D D
ANCESTRAL AREAS A CD AB D BD DE B E A BE F C EF Area without info with lower pr B Dispersal Vicariance Dispersal followed by vicariance D A = Cortez province E B = Mexican province C = Ecuador-Panamic province F D = Costa Rica-Panama-Panamic province E = Gulf of Panama-Panamic province C F = El Choco-Colombia-Panamic province
Fig. 7. Ancestral area reconstruction with DEC and S-DIVA base in BEAST tree using the biogeographical provinces proposed by Hastings (2000). Color and letters of the circles correspond to the areas shown in the box. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) represent not only the limits of the biogeographic provinces within the 2006; Robertson and Cramer, 2009; Briggs and Bowen, 2012). In ad- TEP, but also to act as barriers to the larval dispersal of various reef dition, in many widespread species of both reef-fishes and invertebrates fishes (Hastings, 2000). However, their efficiency as biogeographic and there is no evidence that those barriers promote population level ge- dispersal barriers for reef fishes has been questioned due to inconsistent netic structure (Muss et al., 2001; Craig et al., 2006; Bernardi et al., patterns of the geographic distributions of many species (Craig et al., 2008, 2014; Lessios and Baums, 2017 for a general review).
280 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Ne 1.0E3 Ne 1.0E3 HgI HgII
1.0E2 1.0E2
1.0E1 1.0E1
1.0E0 1.0E0
(HPD = 95%) (HPD = 95%) 1.0E-1 1.0E-1 0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4 Time (Mya) Time (Mya)
Ne 1.0E3 Ne 1.0E3 HgIII HgIV
1.0E2 1.0E2
1.0E1 1.0E1
1.0E0 1.0E0
(HPD = 95%) (HPD = 95%) 1.0E-1 1.0E-1 0.0 0.05 0.1 0.15 0.2 0.25 0.3 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Time (Mya) Time (Mya)
Fig. 8. Demographic history of populations of each haplogroup recovered using Bayesian Skyline Plots (BSP) from cytb sequences.
4.1.3. Central American gap (CAG) distribution of E. puncticulatus south of the CAG. At present E. puncti- In both the phylogenetic analyses and haplotype networks a clear culatus has not been recorded between the southern edge of the CAG segregation of the samples distributed South (sub-clade IIa) and North (western El Salvador) and southern Nicaragua, a distance of 500 km of (sub-clade IIb) of the CAG was evident. These two sub-clades represent coastline that mainly consists of sandy shoreline. Our collecting ex- independent evolutionary units that have been isolated for ∼1.51 Mya, peditions that sampled both intertidal pools and subtidal reef in El and, based on the species-tree analysis, likely represent distinct species. Salvador, habitats used by E. puncticulatus, failed to produce any spe- Thirteen mutation steps separate those two subclades, which is similar cimens of this species. According to J.E. Barraza (pers. com. to DRR, to the number of mutation steps (∼10) that separate six recognized and 2018), a marine biologist at the Universidad Francisco Gavidia in El morphologically distinct species of Elacatinus in the tropical west Salvador, local scuba divers have not seen this species in reef habitats in Atlantic: E. oceanops, E. evelynae, E. chancei, E. horsti, E. lori and E. El Salvador. Thus the distribution gap for E. puncticulatus seems likely to louisae (Taylor and Hellberg, 2006). ∼1000 km, twice as wide as the CAG, and oceanographic conditions in In addition to being a large habitat gap, the CAG does have two the CAG may not entirely explain the lack of gene flow between the characteristics that reinforce that effect, enabling it to act as a strong Mexican and Panamic provinces. biogeographic barrier. The northern half of the CAG on the Pacific side of the Tehuantepec Isthmus has a large, strong upwelling system with 4.1.4. Sinaloan gap ff associated o shore current jets and mesoscale gyres (Fig. 1A) (Trasviña A degree of separation of the Cortez and Mexican province popu- et al., 1999; Rodríguez-Rubio et al., 2003). As upwelling conditions can lations is indicated by statistically significant Φ values for pairwise fi ST be stressful to adult reef shes (Robertson, 1990) so may they be to genetic differentiation between those populations for all three genes, fi larval reef shes. In addition, currents of that upwelling system trans- although some Mexican-province fish have cytb haplotypes that are ff port invertebrate larvae o shore into the open ocean (Valles-Jiménez most closely related to common haplotypes from the Cortez province ff et al., 2005), e ectively reducing connectivity across the CAG. The (Fig. 4; Table 3). Further, high genetic diversity in both provinces fl in uence of smaller-scale gyres on the distribution of larvae has been suggest large and stable population sizes in each case (Table 2). This documented in the Gulf of California, where one meso-scale gyre pro- pattern reflects a present day disjunct population distribution, with ff fi duces the o shore concentration and retention of sh larvae (e.g. occasional migrants and a historically large effective population size, as Aceves-Medina et al., 2003, 2004). The existence of a genetic break at has observed for other marine fishes (e.g. Henriques et al., 2014). This the Gulf of Tehuantepec in a holothurian has been attributed to a cytb pattern indicates that the Sinaloan gap per se does not act as a ff combination of lack of rocky habitat, and the e ects of the gyre and significant barrier to dispersal in E. puncticulatus, and that there is upwelling there (Prieto-Rios et al., 2014). unidirectional dispersal across that gap, from the Cortez province This information is consistent with the CAG acting as a strong bio- southwards to the Mexican province. Such a north-to-south pattern has geographic barrier for E. puncticulatus (Fig. 1B). However, while that not been described before for organisms in the Sea of Cortez and may well be the case there is an issue with the size of the gap in the Mexican provinces. In the coral (Porites panamensis) gene flow is more
281 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285 restricted between the Cortez and Mexican provinces than within each population resident in that gulf, which spans a large area between those province, and gene flow between those provinces is northwards from occupied by those two haplogroups. (Romero-Torres et al., 2017). the Mexican to the Cortez province (Saavedra-Sotelo et al., 2013). In However, the fact that subclades HgII and HgIII are sister clades se- the seahorse Hippocampus ingens the population in the Cortez province parated geographically by a much older major clade (clade I) indicates is genetically separated from those in the Mexican and Panamic pro- that larvae of E. puncticulatus are capable of dispersal over considerable vinces, which themselves are not separated (Saarman et al., 2010). distances of open water, if only rarely. They do occasionally manage to Those authors attributed this pattern to retention of pelagic juveniles, get to Malpelo Island, which is ∼370 km from the nearest point on the which spend an unknown period of time in the plankton, resulting from mainland, in the southern Gulf of Panama. the action of gyres in the Gulf of California and a persistent front at the mouth of that gulf (Figs. 1 and S1). However, there is no information 4.2. Demographic scenario available on the directionality of gene flow between the Cortez and Mexican/Panamic provinces in that seahorse. Our data and analyses - the star-like haplotype network, high hap- This partial isolation between the Cortez and Mexican populations lotype diversity and low nucleotide diversity, and BSP analyses - reveal of E. puncticulatus could result from effects of mesoscale gyres and the that a population expansion of the four haplogroups of E. puncticulatus upwelling in the mouth of Cortez Sea brought about by the convergence started ∼170–130 kya, and continued approximately until ∼25 kya of three current systems there: the cold southward flow of the California (Fig. 8). That period extended from the end of the Illinois/Riss glacial Current (CC), the warm northward surface flow of the Mexican Coastal stage to the relatively cold Wisconsin/Würm glacial maximum (∼20 Current (MCC) from the coast of southern Mexico to the mouth of the Kya). Historical climatic changes during the Pleistocene produced Gulf of California, where it meets the high-salinity Gulf of California fluctuations in sea surface temperatures, sea levels and ocean circula- Current (GCC) flowing out of the Gulf (Figs. 1 and S1) (Zamudio et al., tion patterns, which have been linked to significant changes in demo- 2007; Pantoja et al., 2012). However, currents inside the Gulf of Cali- graphic history of various marine fishes (Fauvelot et al., 2003; Craig fornia that potentially connect the Cortez and Mexican provinces by et al., 2006; Janko et al., 2007 Gaither et al., 2011). At least three larval dispersal flow in both directions along the southeastern coast of glacial cycles have occurred during the last 800 Kya, during which the Gulf, across the Sinaloan Gap, rather than exclusively or pre- there were substantial eustatic sea level fluctuations that affected ha- dominantly southwards (Pantoja et al., 2012; Zamudio et al., 2007). bitat distributions on continental shelves (Lambeck and Chappell, 2001; This indicates that bidirectional larval dispersal between those pro- Siddall et al., 2003), and left footprints in patterns of genetic diversity vinces is quite possible in E. puncticulatus. If that is the case then genetic among marine organisms (Hewitt, 2000). Although demographic ex- relations between populations of E. puncticulatus in those two provinces pansions found in several species of fishes have been linked to the last may be related more to adaptation to local environmental conditions glacial maximum (LGM) around 20 Kya (Bernatchez and Wilson, 1998; than dispersal per se. In this scenario, the limited unidirectional gene Fauvelot et al., 2003; Hewitt, 2004; Woodroffe et al., 2010; Delrieu- flow from the Cortez to Mexican province may arise because the po- Trottin et al., 2017), our data instead indicate that population expan- pulation adapted to a highly seasonal subtropical environment in the sions of E. puncticulatus began around 170–130 Kya, long before the Gulf, which includes a tropical summer, can tolerate less seasonal, more LGM. While similar results have been found in other marine organisms uniformly tropical conditions in the Mexican province, while fish in various parts of the globe (Hoarau et al., 2007; Derycke et al., 2008; adapted to year-round tropical conditions in the Mexican province are Marko et al., 2010; Laakkonen et al., 2013; Ni et al., 2014; Kawamura not able to tolerate winter conditions in the Cortez province. Such et al., 2017), this is the first such case for a marine fish species in the variation in environmental tolerance has been implicated in producing TEP. This result highlights the importance of pre-LGM events on the local disjunctions in the distributions of lineages adapted to different demography of tropical marine shore-fish populations, as well as in- environmental conditions in species of Halichoeres wrasses in the tro- dicating that the LGM did not affect all species in a demographically pical West Atlantic, fishes that show genetic connectivity over large similar manner, and that some species even maintained stable popu- distances in areas of similar environmental conditions and dispersal lations during that period. across boundaries between lineages (Rocha et al., 2005). Thus, in the case of E. puncticulatus, both sandy gaps may have had 4.3. Taxonomic implications effects on the geography of genetic structure of its populations, al- though different effects: unidirectional gene fl ow across the smaller The results of the species-tree construction, supported by AMOVA
(370 km) Sinaloan Gap, and a lack of gene flow across the larger and paired ΦST values, indicate that Elacatinus puncticulatus actually (∼1000 km) distributional gap centered on the Central American Gap. represents a species complex (Fig. 3; Tables 3 and 4). That conclusion is However, neither has had as strong an effect as the factors that pro- corroborated by comparison of the genetic distances between three duced the strong differentiation of the Gulf of Panama haplogroup. populations: the Gulf of Panama – El Choco, Cortez – Mexican pro- Consequently, the pattern of geographic genetic structure found in E. vinces, and Costa Rica – Western Panama/Ecuador. These range from puncticulatus evidently is determined by a combination of mechanisms: 2.2 to 7.3% in cytb gene, distances that are higher than the lowest di- most importantly adaptation to strong local variability in environ- vergence (1.27%) reported for sister species of Elacatinus in the west mental conditions, enhanced to some degree by effects of ocean cur- Atlantic (Taylor and Hellberg, 2005), and higher than the 2% limit for rents on retention vs offshore (i.e. removal) transport of larvae, and cytb that is often used as the basis for the recognition of vertebrate some effects of one large habitat discontinuity. species (Avise, 1998). Accordingly, we propose that E. puncticulatus comprises a species complex with least three allopatric members: the 4.1.5. Ecuador vs Costa Rica – Western Panama split first is Clade I/HgI, Gulf of Panama - El Choco, Colombia, which is Our results showed a clear segregation of individuals from two areas separated by uncorrected genetic distances of 7.3% from the second separated by the Gulf of Panama: Costa Rica – Western Panama (hap- species, which is found in the Cortez and Mexican provinces (Sub-clade logroup HgIII) vs Ecuador (haplogroup HgII). This was evident in IIb/HgIV), and from the third species by distances of 6.7% from haplotype networks for all three genes, the AMOVA results, highly Ecuador (Sub-clade IIa/HgII), and 7.1% from Costa Rica – Western significant paired ΦST values, and a genetic distance of 0.8% in cytb and Panama (Sub-clade IIa/HgIII). The third species, incorporates the re- 0.2% in S7, that indicate isolation of those two populations of ∼530 maining populations from the Panamic province (Costa Rica, Western Kya (Fig. 4 and; Tables 3–6). This separation is linked to effects of the Panama and Ecuador), and is separated by uncorrected genetic dis- upwelling dominated environmental conditions and gyre system in the tances of 2.2–2.4% from the Mexican and Cortez populations. Gulf of Panama that produced the even stronger isolation of the Elacatinus puncticulatus (Ginsburg, 1938) was described from
282 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285 specimens collected in the Gulf of California, which is occupied by the Funding sources Mexican-Cortez sub-clade. The only other species of Elacatinus named from the TEP is E. rubrifrons (Fowler, 1944), the formal description of This work was supported by the Universidad Michoacana de San which was based on specimens collected in the Perlas Archipelago of Nicolás de Hidalgo (CIC-2013-2017) and Consejo Nacional de Ciencia y the Gulf of Panama, the same locality where we collected fish in Clade Teconología (CONACyT, Grant No. CB-2014-240875). I/HgI. Although E. rubrifrons was synonymized with E. puncticulatus by Hoese and Reader (2001), they noted geographic variation in fin-ray Declarations of interest counts, and our genetic data support the notion that it actually re- presents a valid species. In addition our data also indicate that E. The authors declare no conflict of interest. puncticulatus is restricted to the Cortez and Mexican provinces. Thus at least two of the three species indicated by our species-tree analysis have Appendix A. Supplementary material been described, named, and have allopatric distributions. Assessment of the third species indicated by our species-tree (Western Panama/Costa Supplementary data to this article can be found online at https:// Rica and Ecuador, with a cytb separation from the other two ranging doi.org/10.1016/j.ympev.2018.10.020. from 2.2 to 7.1%) must await a detailed morphological analysis of specimens from all major clades and subclades, as it has not yet been References described and named. As well as the genetic differences described above there also are Abele, L.G., 1979. The community structure of coral associated decapod crustaceans in indications of interspecific color variation among these TEP species of variable environments. In: Livingston, R.J. (Ed.), Ecological Processes in Coastal and Marine Systems. Plenum, New York, pp. 265–290. Elacatinus: Inspection of photographs of live individuals taken in the Aceves-Medina, G., Saldierna-Martínez, R.J., González, E.A., 2003. Distribution and field show that fish from Coiba and Nicaragua (Sub-clade IIa/HgIII) abundance of Syacium ovale larvae (Pleuronectiformes: Paralichthyidae) in the Gulf have dark red ground color with paler red stripes restricted to the top of of California. Rev. Biol. Trop. 51, 561–567. fi Aceves-Medina, G., Jiménez-Rosenberg, S.P.A., Hinojosa-Medina, A., Funes-Rodríguez, the head, and red bars through the eye (Fig. S2a); multiple sh from R., Saldierna-Martínez, R.J., Smith, P.E., 2004. Fish larvae assemblages in the Gulf of Baja California and Puerto Vallarta (E. puncticulatus; Sub-clade IIb/ California. J. Fish Biol. 65, 832–847. HgIV) have orange-red ground color on the top of the head that extends Avise, J.C., 1998. The history and purview of phylogeography: a personal reflection. Mol. – down the top of the snout and surrounds the mouth, with bright yellow Ecol. 7, 371 379. Bacon, C.D., Silvestro, D., Jaramillo, C., Smith, B.T., Chakrabarty, P., Antonelli, A., 2015. stripes and yellow bars through the eye (Fig. S2b); while, in a single fish Biological evidence supports an early and complex emergence of the Istmus of from Ecuador, the red-head coloration is similar to that in E. puncticu- Panama. Proc. Natl. Acad. Sci. 112, 6110–6115. Bernardi, G., Alva-Campbell, Y.R., Gasparini, J.L., Floeter, S.R., 2008. Molecular ecology, latus, but more extensive around the mouth (Fig. S2C). Unfortunately fi fi speciation, and evolution of the reef sh genus Anisotremus. Mol. Phylogenet. Evol. we have no photographs of sh from the Perlas Archipelago to compare 48, 929–935. with those from Cortez, western Panama and Ecuador. Bernardi, G., Ramon, M., Alva-Campbell, Y., Mccosker, J.E., Bucciarelli, G., Garske, L.E., Victor, B.C., Crane, N.L., 2014. Darwin’s fishes: phylogeography of Galápagos reef fishes. Bull. Mar. Sci. 90, 533–549. 5. Conclusions Bernatchez, L., Wilson, C.C., 1998. Comparative phylogeography of nearctic and pa- learctic fishes. Mol. Ecol. 7, 431–452. We found strong subdivisions between populations of E. puncticu- Briggs, J.C., 1974. Marine Zoogeography. McGraw-Hill, California, New York. Briggs, J.C., Bowen, B.W., 2012. A realignment of marine biogeographic provinces with latus scattered throughout its geographic range. This pattern included particular reference to fish distributions. J. Biogeogr. 39, 12–30. five discrete subdivisions, with an unusual geographic arrangement. Chelton, D.B., Freilich, M.H., Esbensen, S.K., 2000. Satellite observations of the wind jets This is the first case in which such a degree of genetic differentiation off the Pacific coast of Central America. Part I: Case studies and statistical char- fi acteristics. Mon. Weather Rev. 128, 1993–2018. has been found in any shore- sh species in the TEP. The limited pub- Chelton, D.B., Schlax, M.G., Freilich, M.H., Milliff, R.F., 2004. Satellite measurements lished information available on other endemic TEP shore fishes in- reveal persistent small-scale features in ocean winds. Science 303, 978–983. dicates at most either one subdivision (between the Cortez province and Chen, W.J., Bonillo, C., Lecointre, G., 2003. Repeatability of clades as a criterion of re- the remainder of the region) in the only seahorse found in that region liability: a case study for molecular phylogeny of Acanthomorpha (Teleostei) with larger number of taxa. Mol. Phylogenet. Evol. 26, 262–288. (Saarman et al., 2010) or gene flow across all three provinces in four Chow, S., Hazama, K., 1998. Universal PCR primers for S7 ribosomal protein gene introns species of the haemulid genus Anisotremus (Bernardi et al., 2008; in fish. Mol. Ecol. 7, 1247–1263. Lessios and Baums, 2017). Only future work, both on other genera and Colin, P.L., 1975. The Neon Gobies: The Comparative Biology of the Gobies of the Genus fi ff Gobiosoma, subgenus Elacatinus, (Pisces: Gobiidae) in the Tropical Western North species of gobies, and members of other sh taxa with di erent life Atlantic Ocean. TFH Publications, Neptune, NJ. histories will show how much variability there is the extent and geo- Collin, R., Ochoa, I., 2016. Influence of seasonal environmental variation on the re- production of four tropical marine gastropods. Mar. Ecol. Prog. Ser. 555, 125–139. graphic patterning of genetic subdivisions of widespread TEP shore ff fi Collin, R., Nieto, N., Peña, C., 2018. Seasonal di erences in egg size in three species of shes, and the extent to which that is related to life history char- crabs from a tropical upwelling zone. Mar. Biol. Res. 1–11. https://doi.org/10.1080/ acteristics. 17451000.2017.1406664. Combosch, D.J., Vollmer, S.V., 2011. Population genetics of an ecosystem-defining Reef Coral Pocillopora damicornis in the Tropical Eastern Pacific. PLoS ONE 6 (8), e21200. Acknowledgements Cowen, R.K., Sponaugle, S., 2009. Larval dispersal and marine population connectivity. Annu. Rev. Mar. Sci. 1, 443–466. We thank Georgina Palacios, Salvador Romero, Alfrancis Craig, M.T., Hastings, P.A., Pondella, D.J., Robertson, D.R., Rosales-Casián, J.A., 2006. Phylogeography of the flag cabrilla Epinephelus labriformis (Serranidae): implications Arredondo, Eloisa Torres, Francisco Martínez, Edgar Acevedo, Yareli for the biogeography of the Tropical Eastern Pacific and the early stages of speciation López, Omar Valencia, Victor Piñeros and Ernesto Peña for their help in in a marine shore fish. J. Biogeogr. 33, 969–979. fieldwork, and to Juan Armando Sánchez of Universidad de los Andes D’Croz, L., Robertson, D.R., 1997. Coastal oceanographic conditions affecting coral reefs for his collaboration in the collect of El Choco, Colombia. The samples on both sides of the isthmus of Panama. In: Proc. 8th Int. Coral Reef Symp., vol. 2, pp. 2053–2058. were collected under permits 013/2012 PNG (Ecuador), PPF/DGOPA- D’Croz, L.D., Mate, J.L., 2004. Experimental responses to elevated water temperature in 035/15 (México), 78-Panama (Panama), R056-2015-OT-CONAGEBIO genotypes of the reef coral Pocillopora damicornis from upwelling and non-upwelling – (Costa Rica) and MARN-AIMA-004-2013 (El Salvador). The research environments in Panama. Coral Reefs 23, 473 483. Delrieu-Trottin, E., Mona, S., Maynard, J., Neglia, V., Veuille, M., Planes, S., 2017. has been carried out within an appropriate ethical framework. All Population expansions dominate demographic histories of endemic and widespread procedures performed, including field sampling techniques and la- Pacific reef fishes. Sci. Rep. 7, 40519. fl boratory protocols, as anesthetised and euthanasia techniques used in Derycke, S., Remerie, T., Backeljau, T., Vierstraete, A., Van eteren, J., Vincx, M., Moens, T., 2008. Phylogeography of the Rhabditis (Pellioditis) marina species complex: this study were reviewed and approved by a committee of Mexican evidence for long-distance dispersal, and for range expansions and restricted gene Ministry of Environmental and Natural Resources (SEMARNAT).
283 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
flow in the northeast Atlantic. Mol. Ecol. 17, 3306–3322. demography and genetic characteristics of the Pacific Oyster Crassostrea gigas in Devis-Morales, A., Schneider, W., Montoya-Sánchez, R.A., Rodríguez-Rubio, E., 2008. Japan. Biochem. Syst. Ecol. 70, 211–221. Monsoon-like winds reverse oceanic circulation in the Panama Bight. Geophys. Res. Kessler, W.S., 2006. The circulation of the eastern tropical Pacific: a review. Prog. Lett. 35, L20607. https://doi.org/10.1029/2008GL035172. Oceanogr. 69, 181–217. Dowling, T.E., Tibbetts, A.C., Minckley, W.L., Smith, G.R., 2002. Evolutionary relation- Laakkonen, H.M., Lajus, D.L., Strelkov, P., Väinölä, R., 2013. Phylogeography of amphi- ships of the Plagopterins (Teleostei: Cyprinidae) fom Cytochrome b sequences. boreal fish: tracing the history of thePacific herring Clupea pallasii in North-East Copeia 2002, 665–678. European seas. BMC Evol. Biol. 13, 67. https://doi.org/10.1186/1471-2148-13-67. Drummond, A.J., Rambaut, A., Shapiro, B., Pybus, O.G., 2005. Bayesian coalescent in- Lambeck, K., Chappell, J., 2001. Sea level change through the last glacial cycle. Science ference of past population dynamics from molecular sequences. Mol. Biol. Evol. 22, 292, 679–686. 1185–1192. Lanfear, R., Calcott, B., Ho, S.Y., Guindon, S., 2012. PartitionFinder: combined selection Drummond, A.J., Ho, S.Y.W., Phillips, J., Rambaut, A., 2006. Relaxed phylogenetics and of partitioning schemes and substitution models for phylogenetic analyses. Mol. Biol. dating with confidence. Plos Biology 5, e88. Evol. 29, 1695–1701. Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics with Lessios, H.A., Robertson, D.R., 2006. Crossing the impassable: genetic connections in 20 BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29, 1969–1973. reef fishes across the eastern pacific barrier. Proc. R. Soc. B 273, 2201–2208. Duda, T., Lessios, H., 2009. Connectivity of populations within and between major bio- Lessios, H.A., 2008. The great American schism: divergence of marine organisms after the geographic regions of the tropical PacificinConus ebraeus, a widespread marine rise of the Central American Isthmus. Annu. Rev. Ecol. Evol. Syst. 39, 63–91. gastropod. Coral Reefs 28, 651–659. Lessios, H.A., Baums, I.B., 2017. Gene flow in coral reef organisms of the Tropical Eastern Ekman, S., 1953. Zoogeography of the Sea. Sidgwick and Jackson, London. Pacific. In: Glynn, P.W., Manzello, D., Enochs, I.C. (Eds.), Coral Reefs of the Eastern Excoffier, L., Lischer, H.E., 2010. Arlequin suite ver 3.5: a new series of programs to Tropical Pacific. Coral Reefs of the World. Springer, Dordrecht, pp. 477–499. perform population genetics analyses under Linux and Windows. Mol. Ecol. Resour. Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis for DNA 10, 564–567. polymorphism data. Bioinformatics 25, 1451–1452. Fauvelot, C., Bernardi, G., Planes, S., 2003. Reductions in the mitochondrial DNA di- Luiz Jr., O.J., Allen, A.P., Robertson, D.R., Floeter, S.R., Kulbiki, M., Vigliola, L., Becheler, versity of coral reef fish provide evidence of population bottlenecks resulting from R., Madin, J.S., 2013. Adult and larval traits as determinants of geographic range size Holocene sea-level change. Evolution 57, 1571–1583. among tropical reef fishes. Proc. Natl. Acad. Sci. USA 110, 16498–16502. Fernández-Álamo, M.A., Färber-Lorda, J., 2006. Zooplankton and the oceanography of Machordom, A., Doadrio, I., 2001. Evidence of a cenozoic Betic-Kabilian connection the eastern tropical Pacific: a review. Prog. Oceanogr. 69, 318–359. based on freshwater fish phylogeography (Luciobarbus, Cyprinidae). Mol. Phylogenet. Fiedler, P.C., 2002. Environmental change in the eastern tropical Pacific Ocean: review of Evol. 18, 252–263. ENSO and decadal variability. Mar. Ecol. Prog. Ser. 244, 265–283. Marko, P.B., Hoffman, J.M., Emme, S.A., McGovern, T.M., Keever, C.C., Cox, L.N., 2010. Fiedler, P.C., Lavín, M.F., 2006. A review of eastern tropical pacific oceanography. Prog. The ‘Expansion-Contraction’ model of Pleistocene biogeography: rocky shores suffer a Oceanogr. 69, 91–398. sea change? Mol. Ecol. 19, 146–169. Floeter, S.R., Rocha, L.A., Robertson, D.R., Joyeux, J.C., Smith-Vaniz, W.F., Wirtz, P., Mitarai, S., Siegel, D.A., Watson, J.R., Dong, C., McWilliams, J.C., 2009. Quantifying Edwards, A.J., Barreiros, J.P., Ferreira, C.E.L., Gasparini, J.L., Brito, A., Falcón, J.M., connectivity in the coastal ocean with application to the Southern California Bight. J. Bowen, B.W., Bernardi, G., 2008. Atlantic reef fish biogeography and evolution. J. Geophys. Res. 114 (C10026). https://doi.org/10.1029/2008JC005166. Biogeogr. 35, 22–47. Muss, A., Robertson, D.R., Stepien, C.A., Wirtz, P., Bowen, B.W., 2001. Phylogeography of Gaither, M.R., Bowen, B.W., Bordenave, T.R., Rocha, L.A., Newman, S.J., Gomez, J.A., Ophioblennius: the role of ocean currents and geography in reef fish evolution. Van Herwerden, L., Craig, M.T., 2011. Phylogeography of the reef fish Cephalopholis Evolution 55, 561–572. argus (Epinephelidae) indicates Pleistocene isolation across the indo-pacific barrier Near, T.J., Benard, M.F., 2004. Rapid allopatric speciation in logperch darters (Percidae: with contemporary overlap in the coral triangle. BMC Evol. Biol. 11, 189. https://doi. Percina). Evolution 58, 2798–2808. org/10.1186/1471-2148-11-189. Ni, G., Li, Q., Kong, G., Yu, H., 2014. Comparative phylogeography in marginal seas of the Gernhard, T., 2008. The conditioned reconstructed process. J. Theor. Biol. 253, 769–778. northwestern Pacific. Mol. Ecol. 23, 534–548. Gilg, M.R., Hilbish, T.J., 2003. The geography of marine larval dispersal: coupling ge- Nosil, P., Vines, T.H., Funk, D.J., 2005. Perspective: reproductive isolation caused by netics with fine-scale physical oceanography. Ecology 84, 2989–2998. natural selection against immigrants from divergent habitats. Evolution 59, 705–719. Glynn, P., Stewart, R., McCosker, J., 1972. Pacific coral reefs of Panama: structure, dis- Pantoja, D.A., Marinone, S.G., Parés-Sierra, A., Gómez-Valdivia, F., 2012. Numerical tribution and predators. Int. J. Earth Sci. 61, 483–519. modeling of seasonal and mesoscale hydrography and circulation in the Mexican Glynn, P., Stewart, R., 1973. Distribution of coral reefs in the Pearl Islands (Gulf of Central Pacific. Ciencias Marinas 38 (363–379), 363. Panama) in relation to thermal conditions. Limnol. Oceanogr. 18, 367–379. Patzner, R.A., Van Tassell, J.L., Kovacic, M., Kapoor, B.G., 2011. The Biology of Gobies. Glynn, P., D’Croz, L., 1990. Experimental evidence for high temperature stress as the Science Publishers, Enfield, NH. cause of El Niño-coincident coral mortality. Coral Reefs 8, 181–191. Pennington, J.T., Mahoney, K.L., Kuwahara, V.S., Kolber, D.D., Calienes, R., Chavez, F.P., Glynn, P.W., Manzello, D.P., Enochs, I., 2017. Coral Reefs of the Eastern Tropical Pacific: 2006. Primary production in the eastern tropical Pacific: a review. Prog. Oceanogr. Persistence and Loss in a Dynamic Environment. Springer. 69, 285–317. González-Peña, D., Gómez-Blanco, D., Reboiro-Jato, M., Fernández-Riverola, F., Posada, Pineda, J., Hare, J.A., Sponaugle, S., 2007. Larval transport and dispersal in the coastal D., 2010. ALTER: program-oriented conversion of DNA and protein alignments. Nucl. ocean and consequences for population connectivity. Oceanography 20, 22–39. Acids Res. 38, 14–18. Piñeros, V.J., Beltrán-López, R.G., Baldwin, C.C., Barraza, E., Espinoza, E., Martínez, J.E., Hastings, P.A., 2000. Biogeography of the tropical Eastern Pacific: distribution and Domínguez-Domínguez, O., 2018;al., in press. Diversification of the genus Apogon phylogeny of chaenpsid fishes. Zool. J. Linn. Soc. 128, 319–335. (Lacepède, 1801) (Apogonidae: Perciformes) in the tropical eastern Pacifi c. Mol. Hellberg, M.E., Burton, R.S., Neigel, J.E., Palumbi, S.R., 2002. Genetic assessment of Phylogenet. Evol (in press). connectivity among marine populations. Bull. Mar. Sci. 70, 273–290. Prada, C., Hellberg, M.E., 2013. Long prereproductive selection and divergence by depth Heled, J., Drummond, A.J., 2010. Bayesian inference of species trees from multilocus in a Caribbean candelabrum coral. Proc. Natl. Acad. Sci. USA 110, 3961–3966. data. Mol. Biol. Evol. 27, 570–580. Prieto-Rios, E., Solís-Marín, F.A., Borrero-Pérez, G.H., Díaz-Jaimes, P., 2014. Henriques, R., Potts, W.M., Sauer, W.H.H., Shaw, P.W., 2012. Evidence of deep genetic Filogeografía de Holothuria (Halodeima) inornata Semper, 1868 (Echinodermata: divergence between populations of an important recreational fishery species, Lichia Holothuroidea). Rev. Peru. Biol. 21, 155–162. amia L. 1758, around southern Africa. Afr. J. Mar. Sci. 34, 585–591. Rambaut, A., Drummond, A., 2007. Tracer v1.5. Available: http://tree.bio.ed.ac.uk/ Henriques, R., Potts, W.M., Santos, C.V., Sauer, W.H.H., Shaw, P.W., 2014. Population software/tracer. connectivity and phylogeography of a coastal fish, Atractoscion aequidens Ree, R.H., Moore, B.R., Webb, C.O., Donoghue, M.J., 2005. A likelihood framework for (Sciaenidae), across the Benguela Current Region: evidence of an ancient vicariant inferring the evolution of geographic range on phylogenetic trees. Evolution 59, event. PLoS ONE 9, e87907. 2299–2311. Hewitt, G.M., 2000. The genetic legacy of the Quaternary ice ages. Nature 405, 907–913. Ree, R.H., Smith, S.A., 2008. Maximum likelihood inference of geographic range evolu- Hewitt, G.M., 2004. Genetic consequences of climatic oscillations in the Quaternary. tion by dispersal, local extinction, and cladogenesis. Syst. Biol. 57, 4–14. Philos. Trans. R. Soc. B 359, 183–195. Rice, W.R., 1989. Analyzing tables of statistical test. Evolution 43, 223–225. Hoarau, G., Coyer, J., Veldsink, J., Stam, W., Olsen, J., 2007. Glacial refugia and re- Rickborn, A.J., Buston, P.M., 2015. Life-history transitions of the coral-reef fish Elacatinus colonization pathways in the brown seaweed Fucus serratus. Mol. Ecol. 16, lori. J. Fish Biol. 86, 637–650. 3606–3616. Riginos, C., Nachman, M.W., 2001. Population subdivision in marine environments: the Hoese, D.F., Reader, S., 2001. A preliminary review of the eastern Pacific species of contributions of biogeography, geographical distance and discontinuous habitat to Elacatinus (Perciformes: Gobiidae). Rev. Biol. Trop. 49 (Suppl. 1), 157–167. genetic differentiation in a blennioid fish, Axoclinus nigricaudus. Mol. Ecol. 10, Hurtado, L.A., Frey, M., Gaube, P., Pfeiler, E., Markow, T.A., 2007. Geographical sub- 1439–1453. division, demographic history and gene flow in two sympatric species of intertidal Robertson, D.R., 1990. Differences in the seasonalities of spawning and recruitment of snails, Nerita scabricosta and Nerita funiculata, from the tropical eastern Pacific. Mar. some small neotropical reef fishes. J. Exp. Mar. Bio. Ecol. 144, 49–62. Biol. 151, 1863–1873. Robertson, D.R., Grove, J.S., McCosker, J.E., 2004. Tropical transpacific shorefishes. Pac. Huson, D., Bryant, D., 2006. Application of phylogenetic networks in evolutionary stu- Sci. 58, 507–565. dies. Mol. Biol. Evol. 23, 254–267. Robertson, D.R., Cramer, K.L., 2009. Shore fishes and biogeographic subdivisions of the Janko, K., Lecointre, G., DeVries, A., Couloux, A., Cruaud, C., Marshall, C., 2007. Did Tropical Eastern Pacific. Mar. Ecol. Prog. Ser. 380, 1–17. glacial advances during the Pleistocene influence differently the demographic his- Robertson, D.R., Allen G.R., 2015. Shore fishes of the Tropical Eastern Pacific: online tories of benthic and pelagic Antarctic shelf fishes? - inferences from intraspecific information system. Version 2.0. Smithsonian Tropical Research Institute, Balboa, mitochondrial and nuclear DNA sequence diversity. BMC Evol. Biol. 7. https://doi. Panamá. org/10.1186/1471-2148-7-220. Rocha, L.A., Robertson, D.R., Roman, J., Bowen, B.W., 2005. Ecological speciation in Kawamura, K., Miyake, T., Obata, M., Aoki, H., Komaru, A., 2017. Population tropical reef fishes. Proc. Royal Soc. B 272, 573–579.
284 E.R. Sandoval-Huerta et al. Molecular Phylogenetics and Evolution 130 (2019) 269–285
Rocha, L.A., Craig, M.T., Bowen, B.W., 2007. Phylogeography and the conservation ge- Stepien, C.A., Rosenblatt, R.H., Bargmeyer, B.A., 2001. Phylogeography of the spotted netics of coral reef fishes. Coral Reefs 26, 501–512. sand bass, Paralabrax maculatofasciatus: divergence of Gulf of California and Pacific Rodríguez-Rubio, E., Schneider, W., Abarca del Río, R., 2003. On the seasonal circulation Coast populations. Evolution 55, 1852–1862. within the Panama Bight derived from satellite observations of wind, altimetry and Streelman, J.T., Danley, P.D., 2003. The stages of vertebrate evolutionary radiation. sea surface temperature. Geophys. Res. Lett. 30, 1410. https://doi.org/10.1029/ Trends Ecol. Evol. 18, 126–131. 2002GL016794. Szabó, Z., Snelgrove, B., Craig, M.T., Rocha, L.A., Bowen, B.W., 2014. Phylogeography of Romero-Torres, M., Acosta, A., Treml, E.A., 2017. The regional structure of spawning the manybar goatfish, Parupeneus multifasciatus, reveals isolation of the Hawaiian phenology and the potential consequences for connectivity of coral assemblages Archipelago and a cryptic species in the Marquesas Islands. Bull. Mar. Sci. 90, across the Eastern Tropical Pacific, ICES. J. Mar. Sci. 74, 613–624. 493–512. Ronquist, F., Teslenko, M., Van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget, Taillebois, L., Castelin, M., Ovenden, J.R., Bonillo, C., Keith, P., 2013. Contrasting Genetic B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: efficient Bayesian Structure among Populations of Two Amphidromous Fish Species (Sicydiinae) in the phylogenetic inference and model choice across a large model space. Syst. Biol. 61, Central West Pacific. PLoS ONE 8 (10), e75465. https://doi.org/10.1371/journal. 539–542. pone.0075465. Rüber, L., Van Tassell, J.L., Zardoya, R., 2003. Rapid speciation and ecological divergence Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular in the American seven-spined gobie (Gobiidae, Gobiosomatini) inferred from a mo- evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30, 2725–2729. lecular phylogeny. Evolution 57, 1584–1598. Taylor, M.S., Hellberg, M.E., 2005. Marine radiations at small geographic scales: spe- Saarman, N.P., Louie, K.D., Hamilton, H., 2010. Genetic differentiation across eastern ciation in Neotropical reef gobies (Elacatinus). Evolution 59, 374–385. Pacific oceanographic barriers in the threatened seahorse Hippocampus ingens. Taylor, M.S., Hellberg, M.E., 2006. Comparative phylogeography of a genus of coral reef Conserv. Genet. 11, 1989–2000. fishes: biogeographic and genetic concordance in the Caribbean. Mol. Ecol. 15, Sambrook, J., Fritschi, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual. 695–707. Cold Spring Harbor Laboratory Press, New York. Trasviña, A., Lluch-Cota, C.D., Filonov, A.E., Gallegos, A., 1999. Oceanografia y El Niño. Sato, A., O’hUigin, C., Figueroa, F., Grant, P.R., Grant, B.R., Tichy, H., Klein, J., 1999. In: Magaña, R.V. (Ed.), Los Impactos de El Niño en México: UNAM, pp 69–101. Phylogeny of Darwin’s finches as revealed by mtDNA sequences. Proc. Natl. Acad. Valles-Jiménez, R., Cruz, P., Pérez-Enríquez, R., 2005. Population genetic structure of Sci. USA 96, 5101–5106. Pacific White shrimp (Litopenaeus vannamei) from Mexico to Panama: microsatellite Saavedra-Sotelo, N.C., Calderon-Aguilera, L.E., Reyes-Bonilla, H., Paz-García, D.A., DNA variation. Mar. Biotechnol. 6, 475–484. López-Pérez, R.A., Cupul-Magaña, A., Cruz-Barraza, J.A., Rocha-Olivares, A., 2013. Walker, B.W., 1960. The distribution and affinities of the marine fish fauna of the Gulf of Testing the genetic predictions of a biogeographical model in a dominant endemic California. Syst. Zool. 9, 123–133. eastern Pacific coral (Porites panamensis) using a genetic seascape approach. Ecol. Weersing, K., Toonen, R.J., 2009. Populations genetics, larval dispersal, and connectivity Evol. 3, 4070–4091. in marine systems. Mar. Ecol. Prog. Ser. 393, 1–12. Schluter, D., 2009. Evidence for ecological speciation and its alternative. Science 323, White, C., Selkoe, K.A., Watson, J., Siegel, D.A., Zacherl, D.C., Toonen, R.J., 2010. Ocean 737–741. currents help explain population genetic structure. Proc. Royal Soc. B 277, Siddall, M., Rohling, E., Almogi-Labin, A., 2003. Sea-level fluctuations during the last 1685–1694. glacial cycle. Nature 423, 853–858. Woodroffe, C.D., Brooke, B.P., Linklater, M., Kennedy, D.M., Jones, B.G., Buchanan, C., Silvestro, D., Michalak, I., 2012. RAxmlGUI: a graphical front-end for RAxML. Org. Mleczko, R., Hua, Q., Zhao, J.X., 2010. Response of coral reefs to climate change: Divers. Evol. 12, 335–337. expansion and demise of the southernmost Pacific coral reef. Geophys. Res. Lett. 37 Sousa-Santos, C., Robalo, J.I., Collares-Pereira, M.J., Almada, V.C., 2005. Heterozygous (15), L15602. https://doi.org/10.1029/2010GL044067. indels as useful tools in the reconstruction of DNA sequences and in the assessment of Yu, Y., Harris, A.J., He, X.J., 2010. S-DIVA (statistical dispersal-vicariance analysis): a ploidy level and genomic constitution of hybrid organisms. DNA Sequence 16, tool for inferring biogeographic histories. Mol. Phylogenet. Evol. 56, 848–850. 462–467. Yu, Y., Harris, A.J., Blair, C., He, X., 2015. RASP (Reconstruct Ancestral State in Springer, V.G., 1959. Systematics and zoogeography of the clinid fishes of the subtribe Phylogenies): a tool for historical biogeography. Mol. Phylogenet. Evol. 87, 46–49. Labrisomini Hubbs. Inst. Mar. Sci. Publ. 5, 7–492. Zamudio, L., Hurlburt, H.E., Metzger, E.J., Tilburg, C., 2007. Tropical wave-induced Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis oceanic eddies at Cabo Corrientes and the María Islands, Mexico. J. Geophys. Res. of large phylogenies. Bioinformatics 30, 1312–1313. 112, C05048. https://doi.org/10.1029/2006JC004018. Stephens, M., Donelly, P., 2003. A comparison of Bayesian methods for haplotype re- Zardoya, R., Doadrio, I., 1999. Molecular evidence on the evolutionary and biogeo- construction from population genotype data. Am. J. Hum. Genet. 73, 1162–1169. graphical patterns of european Cyprinids. J. Mol. Evol. 49, 227–237.
285
Figure S1. Map of the currents of the tropical Eastern pacific (TEP) taken from Mariano, A.J. and E.H. Ryan, 2018. http://oceancurrents.rsmas.miami.edu/. Scale bar is in 0C; currents shown are long term average for April.
Figure S2. Photographs of E. Punticulatus, from different biogeographical regions, a) fish from Coiba and Nicaragua (Taken by Alicia Hermosillo); b) Fish from Baja California and Puerto Vallarta (Taken by Frank Krasovec); and c) Fish from Ecuador (Taken by Graham Edgar).
Table S1. Primers, PCR conditions, and References.
Cytb S7 RHO Primers LA S7RPEX1F RH193F HA S7RPEX3R RH1073R
Size (pb) 1140 735 816
Reference Dowling et al., Chow and Hazama, Chen, Bonillo & (2002). (1998) Lecointre (2003)
Denaturing 94ºC, 2 min. 94ºC, 4 min. 94ºC, 4 min. (step 1)
Cycles 5 5 5 (step 2) 30 32 30
Denaturing 94ºC, 30 s. 94ºC, 30 s. 94ºC, 30 s.
Annealing 53ºC, 45 s. 53ºC, 45 s. 56ºC, 45 s. 58ºC, 45 s. 58ºC, 45 s. 50ºC, 45 s.
Extension 72ºC, 60 s. 72ºC, 45 seg. 72ºC, 45 s.
Final Extension 72ºC, 10 min. 72ºC, 7 min. 72ºC, 7 min. (step 3)
Table S2. Tissue voucher number at Universidad Michoacana de San Nicolas de Hidalgo, and access number of GenBank.
Site Locality Biogeographic Voucher Access number of GenBank region number Cytb RHO S7 1 Isla Tortuga, Mulege, Baja California Sur Cortez 11032 MH814154 MH814078 MH813998 1 11033 MH814155 NO NO 1 11037 MH814156 MH814079 MH813999 1 11044 MH814157 NO NO 2 San Marcos, Mulege, Baja California Sur Cortez 14573 MH814161 NO NO 2 14581 MH814162 NO NO 2 14582 MH814163 NO NO 2 15605 MH814180 NO NO 2 15878 MH814200 NO NO 2 15879 MH814201 NO NO 3 Punta Pulpito, Mulege, Baja California Sur Cortez 14785 MH814176 NO NO 3 14787 MH814177 NO NO 4 Bahía Concepción, Mulege, Baja California Sur Cortez 14799 MH814178 NO NO 4 14808 MH814179 NO NO 5 Santa Inés, Mulege, Baja California Sur Cortez 15637 MH814181 NO NO 5 15638 MH814182 NO NO 5 15639 MH814183 NO NO 6 Bahía de los Sueños, La Paz, Baja California Sur Cortez 15715 MH814184 MH814084 MH814004 6 15716 MH814185 MH814085 MH814005 6 15717 MH814186 MH814086 MH814006 15715B ------MH814007 7 Barco CSB, Isla Espíritu Santo, La Paz, Baja Cortez 15754 MH814187 NO NO California Sur 7 15755 MH814188 NO NO 7 15756 MH814189 MH814087 NO 8 El Empachado, Isla Espíritu Santo, La Paz, Baja Cortez 15774 MH814190 MH814088 MH814008 California Sur 15774B --- MH814089 --- 8 15775 MH814191 NO NO 8 15776 MH814192 NO NO 8 15777 MH814193 NO NO 8 15779 MH814194 NO NO 8 15780 MH814195 NO NO 8 15781 MH814196 MH814090 MH814009 8 15782 MH814197 MH814091 MH814010 8 15783 NO MH814092 MH814011 8 15784 MH814198 NO NO 8 15785 MH814199 NO NO 9 Playa Sta. María, La Paz, Baja California Sur Cortez 28513 MH814202 NO NO 10 Caleta del Zorro, Guaymas Sonora Cortez 11196 MH814158 MH814080 MH814000 11 Isla San Pedro Nolasco, Guaymas, Sonora Cortez 11226 MH814159 MH814081 MH814001 12 Las Gringas, Guaymas, Sonora Cortez 14222 MH814160 NO NO 12 14710 MH814164 NO NO 12 14712 MH814165 NO NO 12 14715 MH814166 MH814082 MH814002 12 14716 MH814167 MH814083 MH814003 12 14717 MH814168 NO NO 12 14718 MH814169 NO NO 12 14719 MH814170 NO NO 12 14720 MH814171 NO NO 12 14726 MH814172 NO NO 13 Santa Catalina, Guaymas, Sonora Cortez 14747 MH814173 NO NO 13 14748 MH814174 NO NO 13 14749 MH814175 NO NO 14 Pescadores La Manga II, Guaymas, Sonora Cortez 37396 MH814203 MH814093 MH814012 15 Las Cruces, Isla Tiburón, Hermosillo, Sonora Cortez 37496 MH814204 MH814094 MH814013 15 37497 MH814205 MH814095 MH814014 16 Punta Carrizales, Manzanillo, Colima Mexican 13944 MH814213 NO MH814020 16 14000 MH814214 NO NO 16 14006 MH814215 MH814102 MH814021 16 14012 MH814216 MH814103 MH814022 16 14035 MH814217 NO MH814023 16 14045 MH814218 NO NO 17 Barco Hundido, Faro de Bucerías, Aquila, Michoacán Mexican 12445 MH814209 MH814098 MH814016 17 12451 MH814210 MH814099 MH814017 17 12452 MH814211 MH814100 MH814018 17 13208 MH814212 MH814101 MH814019 18 El Chato, Zihuatanejo, Guerrero Mexican 8102 MH814206 MH814096 MH814015 18 8103 MH814207 MH814097 NO 19 Zacatoso, Ixtapa, Zihuatanejo, Guerrero Mexican 8153 MH814208 NO NO 19 19171 MH814225 NO NO 19 19184 MH814226 NO NO 19 19191 MH814227 NO NO 19 19205 MH814228 NO NO 19 19823 MH814244 NO NO 19 19824 MH814245 NO NO 19 19825 MH814246 NO NO 19 19826 MH814247 NO NO 19 19827 MH814248 MH814110 MH814028 19 19828 MH814249 MH814111 MH814029 19 19829 MH814250 NO NO 19 19832 MH814251 MH814112 MH814030 19 19834 MH814252 NO NO 20 Morros del Potosí, Zihuatanejo, Guerrero Mexican 19099 MH814219 MH814104 MH814024 20 19100 MH814220 MH814105 MH814025 19100B --- MH814106 --- 20 19101 MH814221 MH814107 MH814026 20 19102 MH814222 NO NO 20 19103 MH814223 NO NO 20 19104 MH814224 NO NO 20 19497 MH814233 NO NO 20 19498 MH814234 NO NO 20 19499 MH814235 NO NO 20 19501 MH814236 NO NO 20 19502 MH814237 NO NO 21 Las Gatas, Zihuatanejo, Guerrero Mexican 19300 MH814229 MH814108 MH814027 19300B --- MH814109 --- 21 19367 MH814230 NO NO 21 19383 MH814231 NO NO 21 19386 MH814232 NO NO 22 La Virgen, Zihuatanejo, Guerrero Mexican 19629 MH814238 NO NO 22 19630 MH814239 NO NO 22 19651 MH814240 NO NO 22 19652 MH814241 NO NO 22 19653 MH814242 NO NO 22 19655 MH814243 NO NO 23 Morro Ahogado El Tejón, Santa María Huatulco, Mexican 19955 MH814253 MH814113 MH814031 Oaxaca 23 19956 MH814254 MH814114 MH814032 19956B --- MH814115 --- 23 19957 MH814255 MH814116 MH814033 19957B --- MH814117 --- 23 19971 MH814256 NO NO 23 19972 MH814257 NO NO 23 19973 MH814258 NO NO 23 19977 MH814259 NO NO 24 La Blanca, Santa María Huatulco, Oaxaca Mexican 20011 MH814260 NO NO 24 20012 MH814261 NO NO 24 20013 MH814262 MH814118 MH814036 20013B --- MH814119 --- 24 20014 MH814263 MH814120 MH814037 24 20015 MH814264 MH814121 MH814038 20015B --- MH814122 --- 25 El Arrocito, Santa María Huatulco, Oaxaca Mexican 20039 MH814265 NO NO 25 20048 MH814266 NO MH814034 25 20050 MH814267 NO MH814035 25 20051 MH814268 NO NO 25 20052 MH814269 NO NO 25 20057 MH814270 NO NO 25 20058 MH814271 NO NO 25 20059 MH814272 NO NO 26 Isla Montosa, Santa María Huatulco, Oaxaca Mexican 20122 MH814273 MH814123 MH814039 26 20123 MH814274 NO NO 26 20124 MH814275 NO NO 26 20125 MH814276 NO NO 26 20126 MH814277 NO NO 26 20178 MH814278 NO NO 26 20179 MH814279 NO NO 26 20180 MH814280 NO NO 27 Isla Cabo Blanco, Península de Nicoya, Puntarenas Panamic 21584 MH814296 NO NO 27 21585 MH814297 NO MH814044 27 21586 MH814298 MH814128 NO 27 21587 MH814299 NO NO 27 21588 MH814300 NO NO 27 21589 MH814301 MH814129 MH814045 27 21590 MH814302 MH814130 MH814046 21590B --- MH814131 --- 27 21606 MH814303 MH814132 MH814047 27 21607 MH814304 MH814133 MH814048 27 21608 MH814305 NO NO 28 Los Frailes, Península de Azuero, Provincia Los Panamic 22282 MH814306 NO MH814049 Santos 28 22283 MH814307 MH814134 MH814050 22283B --- MH814135 --- 28 22284 MH814308 MH814136 MH814051 22284B --- MH814137 --- 28 22285 MH814309 MH814138 MH814052 28 22286 MH814310 NO NO 28 22287 MH814311 NO NO 28 22288 MH814312 MH814139 MH814053 28 22289 MH814313 MH814140 MH814054 28 22290 MH814314 NO NO 29 Archipielago de Las Perlas, Provincia de Panamá Panamic 26525 MH814339 MH814145 MH814060 29 26526 MH814340 NO NO 29 26537 MH814341 MH814146 MH814061 29 26717 MH814342 NO MH814062 29 26718 MH814343 NO MH814063 29 26737 MH814344 MH814147 MH814064 29 26738 MH814345 MH814148 MH814065 29 26739 MH814346 NO MH814066 30 El Pelado, Santa Elena, Guayaquil Panamic 17209 MH814281 MH814124 MH814040 30 17210 MH814282 MH814125 MH814041 30 17211 MH814283 NO NO 30 17212 MH814284 NO NO 30 18464 MH814294 MH814127 MH814043 31 Isla Salango, Manabí, Guayaquil Panamic 17878 MH814285 MH814126 MH814042 31 17881 MH814286 NO NO 31 17882 MH814287 NO NO 31 17883 MH814288 NO NO 31 17884 MH814289 NO NO 31 17885 MH814290 NO NO 31 17886 MH814291 NO NO 31 17887 MH814292 NO NO 31 17888 MH814293 NO NO 31 18583 MH814295 NO NO 32 Ballena, Bahía Salinas, Santa Elena Panamic 25640 MH814315 NO NO 32 25641 MH814316 NO NO 33 Isla Salango, Puerto López, Manabí Panamic 25909 MH814317 MH814141 MH814055 33 25910 MH814318 NO NO 33 26189 MH814332 NO NO 33 26190 MH814333 NO NO 33 26192 MH814334 NO NO 34 Punta Mala, Puerto López, Manabí Panamic 25919 MH814319 MH814142 MH814056 25919B ------MH814057 34 26225 MH814335 NO NO 34 26226 MH814336 NO NO 35 Isla de la Plata, Puerto López, Manabí Panamic 25932 MH814320 NO NO 35 26254 MH814337 NO NO 35 26291 MH814338 NO NO 36 Pozas la Chocolatera, Salinas, Santa Elena Panamic 25980 MH814321 MH814143 MH814058 37 La Pared del Viejo, El Pelado, Santa Elena Panamic 25987 MH814322 NO NO 37 25988 MH814323 MH814144 MH814059 37 26097 MH814327 NO NO 37 26098 MH814328 NO NO 37 26099 MH814329 NO NO 37 26100 MH814330 NO NO 37 26101 MH814331 NO NO 38 Bajo 40, El Pelado, Santa Elena 25999 MH814324 NO NO 38 26000 MH814325 NO NO 38 26001 MH814326 NO NO 39 La Roñosa, Nuqui, El Choco Panamic 50700 MH814354 NO MH814067 50700B ------MH814068 39 50701 MH814347 MH814149 MH814069 39 50702 MH814348 NO MH814070 39 50703 MH814349 NO MH814071 50703B ------MH814072 39 50704 MH814350 NO MH814073 39 50705 MH814351 MH814150 MH814074 50705B --- MH814151 --- 39 50706 MH814352 MH814152 MH814075 50706B --- MH814153 --- 39 50707 MH814353 NO MH814076 39 50708 MH814355 NO MH814077 Table S3. Access number of GenBank for the Elacatinus and Tigrigobius spp., used in the present work
Species Access number of GenBank Cytb RHO Elacatinus genus 1 Elacatinus randalli AY846441.1 AY846614.1
2 Elacatinus figaro AY846438.1 AY846611.1 3 Elacatinus prochilos AY846436.1 AY846609.1 4 Elacatinus illecebrosus AY846435.1 AY846608.1 5 Elacatinus oceanops AY846430.1 AY846603.1 6 Elacatinus evelynae AY846427.1 AY846600.1 7 Elacatinus genie AY846422.1 AY846595.1 8 Elacatinus atronasus AY846421.1 AY846594.1 9 Elacatinus xanthiprora AY846419.1 AY846592.1 10 Elacatinus louisae AY846417.1 AY846590.1 11 Elacatinus horsti AY846415.1 AY846588.1 12 Elacatinus lori AY846413.1 AY846586.1 13 Elacatinus chancei AY846409.1 AY846582.1 Tigrigobius genus 1 Tigrigobius saucrus AY846449.1 AY846622.1 2 Tigrigobius digueti AY846393.1 AY846566.1 3 Tigrigobius pallens AY846453.1 AY846626.1 4 Tigrigobius gemmatus AY846450.1 AY846623.1 5 Tigrigobius janssi AY846401.1 AY846574.1 6 Tigrigobius inornatus AY846395.1 AY846568.1 7 Tigrigobius nesiotes AY846397.1 AY846570.1