Evolutionary History and Taxonomic Reclassification of the Critically Endangered Daggernose Shark, a Species Endemic to the Western Atlantic

relation to the con ﬁ guration of the Amazon River. It is closely associated with the area of the Amazon plume, and its distinctive morphological features represent autapomorphic ecological adaptations to this unique habitat and do not re ﬂ ect systematic distinction from Carcharhinus


Introduction
Most phylogenetic studies of elasmobranchs have focused on higher taxonomic levels, such as order and family, rather than lower levels, such as genera and species. This results in major drawbacks for the understanding of the phylogenetic relationships among the members of given elasmobranch families, such as the Carcharhinidae Jordan and Evermann, 1896 [1][2][3][4][5].
In recent years, the use of molecular methods has greatly improved the potential for the resolution of evolutionary questions, providing more accurate interpretations of phylogenetic arrangements, as well as important input for the establishment of conservation priorities [6][7][8][9][10]. Molecular data can also provide important insights into the relationship between diversification patterns and biogeographic events [8,11,12]. However, the management of elasmobranch stocks, and in particular the conservation of species, is still widely hindered by gaps in both basic and applied researches. These shortfalls include key data on species biology, population status, the scale and intensity of threats, and conflicts with human activities, including public safety [13], especially in the case of endemic species with limited geographic ranges [14][15][16].
The dispersal capacity of species is one of the principal factors that determine animal biodiversity and distribution patterns [17]. Hard barriers-land masses that prevent gene flow, such as the Isthmus of Panama or the closure of the Tethys Sea [18,19]-are important components of the speciation process in aquatic environments, whereas soft barriers are permeable, formed by environmental discontinuities or geographical features that restrict connectivity and gene flow between areas connected physically by bodies of water [18,[20][21][22][23]. Although they do not have a planktonic life stage, many elasmobranch species undertake major migrations [24][25][26]. This means that dispersal is the responsibility of the adults, whereas juveniles are dependent on nursery areas [27]. However, the relationships between biogeographical events and the speciation process are still mostly unclear in this group and have been elucidated only in some cases [8,9,25,28,29].
In fact, the classification of many species within the Carcharhinidae is still unresolved, and there is extensive debate on the potential for paraphyly of genera inside the family, as also found in other families such as the Triakidae Gray, 1851 and the Scyliorhinidae Gill, 1862 [10,[30][31][32]. López et al. [33] detected a paraphyletic arrangement within the carcharinids, based on DNA sequences, and noted a close relationship with the Sphyrnidae Rafinesque, 1810 and Hemigaleidae Compagno [56]. These authors recovered Galeocerdo cuvier (Péron & Lesueur, 1822) in the Hemigaleidae, in a branch external to the clade formed by the carcharhinid and sphyrnid species, which is consistent with the findings of Naylor et al. [30] and Amaral et al. [31], who identified Galeocerdo Müller and Henle, 1837 as a basal taxon and sister to the Carcharhinidae + Sphyrnidae clade.
Using nuclear markers to assess the phylogenetic relationships between Carcharhinus Blainville, 1816 and other carcharhinid genera, Dosay-Akbulut [34] recovered poten-tial paraphyly of Carcharhinus, with Prionace glauca (Linnaeus, 1758) recovered inside Carcharhinus as sister to the group (Carcharhinus falciformis (Bibron, 1839)+Carcharhinus altimus ( [35])), which lead the authors to suggest that Prionace Cantor, 1849 may not be distinct from Carcharhinus. However, they cited divergence in results of different analyses as a motive for not making any changes to the taxonomic classification.
The daggernose shark, Isogomphodon oxyrhynchus (Valenciennes, 1839), the type species of the genus Isogomphodon Gill, 1862, is a critically endangered carcharhinid, which has one of the most restricted geographic ranges of any elasmobranch, inhabiting waters off the coast of Venezuela, Trinidad and Tobago, Guyana, Suriname, French Guyana, and northern Brazil [3,4,16]. The most conspicuous characteristic of this species is its elongated and flattened snout, which resembles a dagger. The genus Isogomphodon was synonymized with Carcharhinus [36] but separated again due to its distinct morphological characteristics resurrecting the genus Isogomphodon [35]. Furthermore, the only extant congener Isogomphodon maculipinnis Poey, 1865 was subsequently synonymized with Carcharhinus brevipinna (Valenciennes, 1839) [37]. Compagno [37] proposed the designation of four tribes (Carcharhinini, Rhizoprionodontini, Triaenodontini, and Isogomphodontini) within the subfamily Carcharhinidae, based on the morphological characteristics, which implies distancing of I. oxyrhynchus from the other genera of this subfamily. These morphological differences led Garrick [38] to exclude I. oxyrhynchus from their analysis, which aimed to define the limits of the genus Carcharhinus. However, in a study of the molecular identification of the shark species harvested by fisheries using only mitochondrial 12S-16S sequences (1380 bp), Rodrigues-Filho et al. [4] found evidence that I. oxyrhynchus belongs to the genus Carcharhinus Blainville, 1816, challenging the distinction of the genus Isogomphodon, but did not make any changes to the taxonomic classification. Similarly, although other studies since then have shown paraphyly of Carcharhinus, the analyses of their studies also showed weak support [10,39] or had poor taxonomic coverage [16] and also did not make any changes to the classifications. Other recent studies using mitogenomes have also shown paraphyly of P. glauca and probably Triaenodon obesus but did not include I. oxyrhynchus and focusing on other results did not reclassify P. glauca or T. obesus [40,41].
The daggernose shark has been overfished since the 1980s and catches collapsed in the mid-1990s [15]. Since this time, the total biomass of the species harvested decreased by 90%, due to the intense targeting of juveniles, which has hampered recruitment [15]. In the 1980s and 1990s, Stride et al. [42] found that the juveniles, which are vulnerable due to the small mesh sizes of the gillnets [43], were the age class harvested most intensively in the state of Maranhão, northern Brazil. This pressure, combined with the restricted distribution of the species and its K-type life history strategy (slow growth, late sexual maturation, and reduced fecundity [2]), has led to the classification of I. oxyrhynchus as critically endangered (CR) since 2006 by the IUCN [44]. More recent demographic studies have indicated that I. oxyrhynchus populations have been decreasing by approximately 18% per annum in recent years [15,45,46].
The present study is aimed at updating the taxonomic classifications in the Carcharhinidae by using data from mitochondrial and nuclear markers to confirm the phylogenetic position of I. oxyrhynchus and evaluate the evolutionary history of this critically endangered species in relation to other carcharhinids.

Material and Methods
2.1. Study Area and Sampling. All samples utilized in the present study were collected along the northern coast of Brazil. This is one of the five distinct regions of the Brazilian coast, classified according to their climatological and oceanographic characteristics [47]. It ranges from the mouth of the Oiapoque River, in the northernmost extreme of the Brazilian state of Amapá, to São Marcos Bay, in the state of Maranhão [44,48] (Figure 1).
Samples of muscle tissue were obtained from I. oxyrhynchus and other sharks taken as bycatch and landed at Raposa Beach (n = 1 sample), in the Brazilian state of Maranhão, Bragança (n = 10) and Vigia (n = 5), both in the state of Pará, and Macapá (n = 2), in the state of Amapá (Table 1 and Figure 1). Some of the samples are whole individuals which were identified following the diagnostic key of Compagno [49]. Additional sequences represent muscle tissue samples that were sequenced in previous studies [4,16]. All samples were placed in microtubes containing 70% ethanol and stored at -20°C at the Genetics and Biotechnology Laboratory of the Capanema campus of the Federal Rural University of Amazonia.   Journal of Zoological Systematics and Evolutionary Research mouse-tail protocol. The final product of each extraction was run at 60 V in 1% agarose gel for 20 minutes and photographed under ultraviolet light. For the phylogenetic analysis, fragments of three mitochondrial loci and one nuclear locus were amplified by polymerase chain reaction (PCR). These PCRs were run using the following primers for the mitochondrial loci: fish F1 and fish R2 [50] for the cytochrome C oxidase subunit 1 gene (COI), ND4 F and Leu-Scyliorhinus [3] for the NADH dehydrogenase subunit 4 gene (ND4), and 12SA and 16SA [51] for a region encompassing the 12S rRNA, valine tRNA, and 16S rRNA genes (12S-16S). The nuclear locus was Recombination Activating Gene 1 (RAG1), for which the primers were Chon-Rag1-S026 and Chon-Rag1-R025 [3].
The following PCR protocol was used to amplify all the samples: 5 minutes for initial denaturation at 94°C, followed by 30 cycles of 30 seconds at 94°C, 40 seconds at 50-60°C, and 1 minute at 72°C, with a final extension of 5 minutes at 72°C. The resulting amplicons were purified using the polyethylene glycol 20% protocol (20% PEG), adapted from Dun and Blattner [52]. The purified material was then precipitated and sequenced using the BigDye kit (ABI Prism™ Dye Terminator Cycle Sequencing Reading Reaction-Applied Biosystems, Foster City, CA, USA) and sequenced using an ABI 3500 Genetic Analyzer at the Phylogenomics Laboratory on the UFPA Bragança campus. The additional internal primers for RAG1 (Chon-Rag1-S030 and Chon-Rag1-S031 [3]) and 12S-16S [51] were used to sequence RAG1 and 12S-16S, respectively.

Phylogenetic and Genetic Distance Analyses.
The sequences generated during this initial processing were edited, aligned, and corrected using the automatic ClustalW alignment tool [53], run in BioEdit 7.0.5 [54]. The Gblocks 0.91b program [55] was used to process the 12S-16S ribosomal regions to avoid errors of homology in the portions of this fragment that are subject to indels and other variable positions, which were removed following preestablished criteria [56]. After the alignment and sequencing corrections, all available sequences were analyzed separately for each of the four loci (four single marker datasets) and a further dataset of concatenated mitochondrial sequences was produced for phylogenetic analyses (hereafter referred to as the "mtDNA dataset"). The mtDNA dataset included one sequence representing each of the 37 species for which all three mitochondrial loci were available for concatenation.
The nucleotide composition and the numbers of conserved, variable, and parsimonious sites, as well as the most suitable nucleotide substitution models, were calculated in MEGA X [57] for each dataset as partitioned (COI: HKY+ G(1.0428)+I(0.6049), 12S-16S: GTR+G(0.4122)+I(0.3388), ND4: TN93+G(1.1641)+I(0.5130), mtDNA: GTR+G (0.5093)+I(0.4665), and RAG1: TN93+G(0.0963)). The number of transitions and transversions was plotted against the divergence of each region in DAMBE 7 to verify the saturation levels of the markers [58]. Nucleotide distance matrices, based on both the corrected and uncorrected p distances, were generated in MEGA X for all datasets [57]. Four different methods of phylogenetic inference were used to construct the phylogenetic trees, according to the evolutionary models identified for each dataset. The maximum likelihood (ML) approach was run in PhyML 3.0 [59], while Bayesian inference (BI) was applied in MrBayes 3.2 [60]. The maximum parsimony (MP) and neighbor joining (NJ) trees were generated in MEGA X [57]. The robustness of the clades generated by the ML, NJ, and MP analyses was estimated using 1000 bootstrap replicates [61].
For the concatenated dataset, BI analyses used all the concatenated fragments as separate unlinked partitions with their respective models as previously calculated. The Bayesian inference was based on MCMC (Markov chain Monte Carlo) sampling, with four simultaneous runs of four chains (one cold and three warm) and 50 million generations, with two runs. The a posteriori probabilities were defined using the 80% consensus rule, with samples being taken every 1000 generations and 10% of the initial trees being discarded as burn-in. The log-likelihood files generated by each run were visualized in Tracer v.1.7.1 [62], considering only the runs with a minimum effective sample size (ESS) of at least 200 (i.e., ESS ≥ 200).

Divergence Time Estimates.
Divergence times were estimated in BEAST 2.5 [63]. The Yule calibration model was selected as the tree prior parameter, using the relaxed molecular clock with uncorrelated rates, which assumes heterogeneous rates on the different branches [64]. This analysis was conducted using the mtDNA dataset using clock models and the unlinked evolutionary models for the concatenated fragments as previously calculated. This dataset was used given that most of the sequences used as calibration points are available for the taxa investigated here.
Three calibration points based on the fossil record were used to estimate the time of the most recent common ancestor (MRCA) of the principal clades, deployed a priori as the age of the nodes in the tree. The ancestral node is associated with the origin of the Carcharhiniformes, estimated at approximately 148.5 million years ago (mya), and the divergence times of Carcharhinus and Sphyrna were dated to~38-42 mya and~16-23 mya, respectively [65,66].
The pretrees were modeled based on the Yule speciation process calibration, with all the other parameters being based on the BEAST 2.5 default values. The MCMC analyses were based on a run of 60 million generations, with four simultaneous runs of four chains (one cold and three heated) sampled every 6000 generations, with 10% of the samples being discarded as burn-in. The results were inspected using Table 1: Species used for phylogenetic analyses with the GenBank code for each employed marker.  [62], and only runs with an ESS of at least 200 for all the marginal parameters were used. After discarding the burn-in, the results of the MCMC runs were submitted to TreeAnnotator v1.7 [67] for the inclusion of calibration times in the branches of the final tree.
The RAG1 sequence had 1,110 bp, with 122 variable sites, 72 informative sites, and 50 unique mutations. The nucleotide composition was 24.9% thymine, 17.7% cytosine, 33.0% adenine, and 24.4% guanine. None of the segments were saturated, given that the observed transition rates were higher than the transversion rates in all cases (Supplementary Figure S1).
The p (distP) and corrected (distC) distance matrices were generated for the mtDNA and RAG1 datasets ( Table 2) to calculate the pairwise divergence intervals between the study species (Supplementary Tables S1 and  S2). In the case of the corrected distances, the best nucleotide substitution model for the mtDNA dataset was the GTR+G(0.5093)+I(0.4665) model, and the best fit for the RAG1 data was the TN93+G(0.0963) model. In the case of the distP matrix, the genetic distances between I. oxyrhynchus and the other carcharhinid species ranged from 4.8% to 11.2% for the mtDNA and from 0.5% to 3.4% for RAG1. In the comparisons based on the distC matrix, the distances ranged from 5.8% to 19.3% for the mtDNA dataset and from 0.6% to 5% for RAG1.
The lowest mean intergeneric genetic distances were recorded between I. oxyrhynchus and the Carcharhinus species, for both the mtDNA (distP = 5:65% and distC = 7:06%) and RAG1 (distP = 0:77% and distC = 0:87%) datasets (Table 2 and Supplementary Tables S1 and S2). The mean values recorded between I. oxyrhynchus and the Carcharhinus species are consistent with those recorded within the genus Carcharhinus, that is, 3.6-6.8% for the mtDNA and 1.4-7.9% for RAG1 datasets (Supplementary  Table S1 and S2). Much greater genetic distance values were recorded between I. oxyrhynchus and the species of the other genera analyzed in the present study, that is, Rhizoprionodon, Loxodon, and Scoliodon (Carcharhinidae) and Galeocerdo (Galeocerdonidae) ( Table 2 and  Supplementary Tables S1 and S2).   ). The RAG1 nuclear tree showed I. oxyrhynchus within Carcharhinus but placed P. glauca as more closely related to Rhizoprionodon and Galeocerdo (Figure 2(b)), but taxon coverage for the nuclear data was lower (only seven Carcharhinus and no T. obesus), and the phylogenetic relationships were not significantly supported.
In addition to the previous phylogenetic analyses (Figure 2), the posterior probabilities for all the nodes of the time to most recent common ancestor (TMRCA, Figure 3) analysis are strong (PP > 0:9), placing Isogomphodon and Prionace conclusively within the genus Carcharhinus. P. glauca diversified from its sister species in the tree (C. falciformis) during the Miocene, approximately 16  The subdivisions of Carcharhinus observed in the TMRCA analysis are also apparent in the phylogenetic inferences based on the mtDNA dataset (Figure 2(a)), albeit with no support, except for the Carcharhinus+Prionace    Journal of Zoological Systematics and Evolutionary Research +Triaenodon+Isogomphodon clade (bootstrap > 70%, PP > 0:9). The genetic distances between these subclades were the same as those observed in the comparisons between I. oxyrhynchus and the Carcharhinus species (mtDNA: distP: 5.5-5.8%, distC: 5.9-6.2%; RAG1: distP and distC: 0.8-0.9%).
Overall, phylogenetic methods (21 trees representing NJ, ML, MP, BI, and TMRCA divergence time analyses) generated in the present study define a paraphyletic arrangement for the genus Carcharhinus in relation to what was previously classified as I. oxyrhynchus and P. glauca and possibly also Triaenodon. Given this molecular evidence, we officially recognize the previous combination Carcharhinus oxyrhynchus (vs. I. oxyrhynchus) and propose the new combination Carcharhinus glaucus (vs. P. glauca).

Carcharhinus, Isogomphodon, and the Carcharhinidae.
These results of this study reflect the long, controversial history of the taxonomic classification of C. oxyrhynchus, originally described as Carcharias oxyrhynchus, reassigned to Isogomphodon and then assigned as Carcharhinus, before Springer [35] and later studies resurrected the genus Isogomphodon (see [37]). Gill (1862) created the genus Isogomphodon to accommodate I. oxyrhynchus. Isogomphodon maculipinnis Poey, 1865, added subsequently, is now considered a synonym of Carcharhinus brevipinna [37]. Rodrigues-Filho et al. [4] raised the hypothesis that the daggernose shark was in fact a member of the genus Carcharhinus, based on a weakly supported phylogeny with low taxonomic coverage, a result of the fact that the study focused primarily on species identification rather than phylogenetic inference and did not make classification changes.
Compagno [37] proposed the allocation of tribes within the Carcharhinidae, in which case, C. oxyrhynchus would be assigned to the tribe Isogomphodontini, based on its unique morphological features. These differences led Garrick [38] to exclude C. oxyrhynchus, together with five other species, from a study that aimed to define the limits of Carcharhinus [38], based on the evidence of variables such as external morphology, morphometric and meristic parameters, and coloration. Garrick concluded that the use of these characteristics for the assessment of the systematics of this genus should be reconsidered, given that, despite their importance at the species level, they are inconclusive for the definition of subgeneric arrangements or inferences on the relationships between Carcharhinus and other, similar genera.
Carcharhinus oxyrhynchus is a demersal shark that inhabits shallow, muddy waters in coastal and estuarine areas, although few data are available on its basic biological characteristics [68]. This species nevertheless has a number of unique morphological features that distinguish it from all other carcharhinids, such as its pectoral fin to body size ratio, as well as its flattened and elongated snout, and the significantly reduced size of its eyes [37], which is probably associated with the muddy waters this shark inhabits, and its predation of benthic and nektonic organisms [37,42]. The head and fins are also sexually dimorphic [69], which may be associated with mating behavior patterns [70]. Silva [71] presented an anatomical description of C. oxyrhynchus based on an analysis of skeletal traits, including the dental series and morphology of the neurocranium, mandibular and brachial arches, claspers, fins, and the dermal denticles, which was used for systematic comparisons with other members of the family Carcharhinidae. Silva [71] initially compared C. oxyrhynchus with the fossil species attributed to the genus Isogomphodon, including Isogomphodon acuarius (Probst, 1879), Isogomphodon lerichei (Darterville and Casier, 1943), Isogomphodon gracilis (Jonet, 1966), Isogomphodon caunellensis ( [65]), and Isogomphodon aikenensis (Cicimurri and Knight, 2019), affirming that only the fossil taxa C. aikenensis and C. acuarius should be maintained in Isogomphondon while other taxa were assigned to Carcharhinus. I. aikenensis and I. acuarius should therefore now also be recognized as Carcharhinus aikenensis n. comb. and Carcharhinus acuarius n. comb. Sorenson et al. [39] also found that C. porosus and T. obesus are the carcharhinid species that are most similar to C. oxyrhynchus. Although no exclusive similarities were found between C. oxyrhynchus and C. porosus, 26 (58%) of the 45 traits analyzed were highly similar, including the morphology of the lower teeth, which have an elongated crown, with narrow roots in the anterior rows [71]. The distinctive morphology of C. oxyrhynchus does not exclude this species from the genus Carcharhinus, however, given that the tree topology and genetic distances found in the present study emphatically support the inclusion of this species in Carcharhinus. The results are also consistent with previous findings of the paraphyletic arrangement of Carcharhinus considering C. glaucus [4,10,30,31,34,37,40,41,72], but this is the first time that it is recognized as a new combination and all fossil taxa in Prionace (Prionace antiquus (Agassiz, 1856), Prionace egertoni (Agassiz, 1843), and Prionace tenuis (Agassiz, 1843)) should also be reallocated to Carcharhinus or recognized as new combinations. The relatively large dataset employed in the present study, together with the robust phylogenetic approaches applied in the analyses, supports this conclusion. However, further research using more nuclear data and more complete taxon coverage will be needed to reevaluate the classification of Triaenodon, as well as the phylogenetic structure of the genus Carcharhinus.
The well-supported subdivisions of the genus Carcharhinus established in the TMRCA analysis are consistent with the findings of Springer [35,73], Garrick [38], and Compagno [74]. Springer [35,73] suggested the division of Carcharhinus into two groups of sharks, those with smooth backs and those with ridged backs. In the present study, the TMRCA estimates also support a division between the smooth-backed sharks (clade 1: C. acronotus, C. brevipinna, C. porosus, and C. oxyrhynchus and clade 3: C. leucas, C. limbatus, and C. amblyrhynchoides) and the ridge-backed forms (clade 2: C. glaucus, C. falciformis, C. obscurus, C. plumbeus, C. perezi, and C. sorrah). Naylor [75] referred to a monophyletic clade of large sharks with interdorsal dermal crests, which included C. sorrah, C. longimanus, C. falciformis, C. plumbeus, C. altimus, C. perezi, and C. glaucus (clade 2 in our study). The results of the present study further refine 9 Journal of Zoological Systematics and Evolutionary Research this arrangement by recognizing two distinct clades (1 and 3) of smooth-backed sharks (Figure 3), of which, clade 3 is the more recent sister group of the ridge-backed shark clade (2). Garrick [38] and Compagno [74] suggested other groups within Carcharhinus, based on their morphological similarities, such as (i) C. porosus, Carcharhinus dussumieri, Carcharhinus macloti, C. sorrah, and Carcharhinus sealei; (ii) Carcharhinus albimarginatus, Carcharhinus amblyrhynchos, C. perezi, and Carcharhinus wheeleri; and (iii) C. limbatus, C. amblyrhynchos, and C. brevipinna. However, none of these arrangements were recognized by Naylor [75] or Lavery [49], and no evidence was found to support them in the present study. Naylor [75] argued that the morphological similarities found between the species of each group must be the result of convergent evolution and that there is a need for further studies of the phylogenetic structure of Carcharhinus. Future reclassifications based on these subdivisions are possible considering, but even with the full mitogenome datasets of Kousteni et al. [40] and Wang et al. [41], these arrangements are currently uncertain.
Based on an electrophoretic study of allozymes, Naylor [75] concluded that Carcharhinidae as described at the time was a paraphyletic group which included the hammerhead sharks of the genus Sphyrna, with Galeocerdo and Rhizoprionodon as the most basal genera of this family. Using DNA sequences, López et al. [33] reported paraphyly in the Carcharhinidae as classified at that time, showing marginal but consistent support for a clade corresponding to carcharhi-nids+sphyrnids to the exclusion of the Hemigaleidae+Galeocerdo cuvier, although only three carcharhinid species were included. However, these findings were corroborated by Naylor et al. [30] and Amaral et al. [31], who defined Galeocerdo as a basal group and a sister taxon to the Carcharhini-dae+Sphyrnidae clade. Rodrigues-Filho et al. [4] also observed the phylogenetic proximity of Galeocerdo and Sphyrna, with Rhizoprionodon as the sister group of this clade. These authors highlighted the genetic divergence of G. cuvier and Rhizoprionodon from the other carcharhinid species included in the analyses and suggested that these two genera may represent distinct carcharhiniform families. In the present study, Rhizoprionodon was recovered as a more basal sister group of the clade formed by Scoliodon and Loxodon. Naylor et al. [30] obtained similar findings, recovering the same arrangement for these genera, but placing them all within the carcharhinid clade. Vélez-Zuazo and Agnarsson [10] recovered Scoliodon as a sister genus of Rhizoprionodon, although their study included no representatives of the genus Loxodon. Poey [76] allocated Galeocerdo in its own family Galeocerdonidae, based on morphology, and some of these (large eyes, very long lip folds, and large spiracles) and its ovoviviparous reproductive mode are not observed in more derived taxa [37,[75][76][77][78][79], demonstrating the considerable divergence from the Carcharhinidae. Despite Poey's allocation, Galeocerdo was until recently attributed to the Carcharhinidae, only recently being widely recognized as the only member of the Galeocerdonidae [76,79]. This recent adoption of the use of the family Galeocerdonidae is strongly supported by the phylogenetic results of the present study.

Adaptations of the Daggernose Shark and Insights into
the Speciation Process. The TMRCA analysis indicates that C. oxyrhynchus diverged from its closest sister lineage in the analysis (C. porosus) during the Miocene, with a confidence interval ranging from approximately 9.38 to 23.74 mya. The geographic distribution of this species is closely associated with the area of the Amazon plume, which corresponds to more than 80% of its area of occurrence [16]. During the transition from the Oligocene to the Miocene, the proto-Amazon drained northward along the paleo-Orinoco basin to discharge into the Caribbean Sea near the presentday Lake Maracaibo, with an increase in the turbidity of the local coastal waters due to the intense deposition of sediments [80,81]. Between the early and late Miocene, river drainage patterns in northern South America changed significantly [82], with increasingly large amounts of Andean sediments being deposited into the Atlantic Ocean from the mouth of the Amazon River [83].
The uplifting of the Andes during the Miocene (~10 mya) also contributed to the formation of the Amazon River, as well as increasing the outflow of the paleo-Orinoco River system towards the Caribbean, decreasing the transparency of the water [84]. The most recent evidence indicates the occurrence of three distinct phases of sediment deposition related to the formation of the Amazon plume: (i) relatively low deposition rates (~11.8-6.8 mya), (ii) increasing volume (~6.8-2.4 mya), and (iii) high deposition rates, from~2.4 mya to the present day [82]. The Amazon River discharges approximately 6300 km 3 of water per year [85], creating a sediment-laden freshwater plume that extends approximately 200 km offshore from the mouth of the river, reaching the 30 m isobath in the middle sector of the continental shelf [86,87]. This plume represents a soft barrier (the Amazon River Barrier) to the dispersal of many marine species [18,[88][89][90][91], which has been recently shown to have an indirect influence on elasmobranch speciation processes [92].
Compagno and Compagno et al. [37,74,93] and Ebert et al. [77] evaluated a total of 24 traits to support the status of Isogomphodon as a genus distinct from Carcharhinus. However, most of these characteristics were shared with C. oxyrhynchus with only six distinct traits: (1) snout acutely triangular or subtriangular in the dorsoventral view and very elongated (preoral length much greater than internarial space or the width of the mouth); (2) extremely small eyes that lack posterior notches; (3) the teeth are not strongly differentiated in either the upper or lower jaws, with anteroposterior narrow acute cusps, and no cusplets or proximal or distal blades; (4) teeth in the upper jaw with broad, flat cusps, and those in the lower jaw with slenderer cusps and smooth edges; (5) a total of 49-60/49-56 rows of teeth; and (6) broad and triangular pectoral fins, whose length from the origin to the free posterior tip is approximately 3/ 4 of that of the anterior pectoral margins.
These distinctive features of C. oxyrhynchus almost certainly represent adaptations to the conditions found in the Amazon plume, and Compagno [37] already proposed that the distinct characteristics of this species are the result of a process of adaptation to the turbid waters of the coastal and estuarine environments associated with the discharge of the Amazon River. These adaptations include changes in the structure of the snout and an associated increase in the area of the Lorenzini ampullae that enhances its function, facilitating movement and feeding in turbid water [37]. The need to adapt to murky environments with poor visibility in northern South America would have favored the enhancement of the snout of C. oxyrhynchus to the detriment of its visual system, with this sensory, dental, and mandibular morphology favoring the predation of small prey [16,42,93].

Conservation of Carcharhinus oxyrhynchus.
Given the importance of the role played by most elasmobranchs in the ecosystems they inhabit and increasing fishery pressures around the world, conservation efforts are urgently needed to ensure the protection of this prominent group of fishes. The principal threat to their survival is the progressive expansion of fishery operations. Sharks and rays started to attract attention from conservationists in the 1990s, when research institutions began to assess this fishery resource more systematically. In Brazil, there were two prominent initiatives during this period-the REVIZEE Program (Program for the Assessment of the Potential Sustainability of Resources in the Exclusive Economic Zone), created by the Ministry of the Environment in 1994 [94], and the establishment of the Brazilian Society for Elasmobranch Studies (SBEEL), founded in 1997, that is active in the promotion of the conservation of sharks and rays in Brazil [95].
The increasing evidence of the vulnerability of elasmobranchs led to the establishment, in 1998, of the International Action Plan for the Conservation and Management of Sharks (IPOA-SHARK), with the support of the United Nations Food and Agriculture Organization (FAO). During this same period, Brazil implemented its first federal antifinning legislation, which prohibited the removal of the fins of sharks and rays (the most expensive byproducts) without the subsequent utilization of the carcass. However, landing shark and ray catches with the fins still attached to the body only became mandatory in 2012 [96,97]. Furthermore, the ability of C. oxyrhynchus to recover health after stressful interactions may result in greater survival risks compared to other Carcharhiniformes even if released alive after accidental capture [98].
The decline observed in Brazilian elasmobranch populations was considered to be an alert and provided an incentive for the implementation of the National Action Plan for the Conservation of Endangered Sharks and Marine Rays in Brazil (PAN-Tubarões), which was created by federal ordinance no. 575 of December 5th, 2014 [99][100][101]. This plan focuses on 12 species, including C. oxyrhynchus, with the support of technical advisory groups.
An associated initiative is the National Action Plan for the Conservation of Endangered and Socioeconomically Important Species in the Mangrove Ecosystem (PAN-Manguezal), which aims to protect the Brazilian mangroves, including the species contemplated by PAN-Tubarões (including C. oxyrhynchus), while respecting local traditions, which are reconciled with scientific knowledge. The PAN-Manguezal action plan encompasses the conservation of 74 elasmobranch species, 20 of which are threatened with extinction in Brazil, including C. oxyrhynchus [102].
Recently, the coastal and marine protected area (GEF-Mar) project was implemented to protect the biodiversity of marine and coastal areas from the impacts of bycatch. This project is aimed at identifying priority coastal areas in northern Brazil for both endangered species and artisanal fisheries and, in particular, the establishment of marine protected areas that can contribute to the recovery of the most threatened species, in particular, through the protection of nursery areas. This has led to the imposition of restrictions on fisheries in the estuaries of the states that make up the northern Brazil region [103].
Clearly, the strategies employed to ensure the conservation of Brazilian elasmobranchs, in particular, the most endangered species, such as the daggernose shark, will not be altered significantly in response to the present study, which refers specifically to a question of scientific nomenclature. It is nevertheless important to note that this species inhabits the shallow waters of the western Atlantic, in regions in which hot and rainy conditions predominate, where it prefers the turbid, brackish waters of estuaries dominated by mangroves and has a viviparous reproduction, late sexual maturity, and a biannual reproductive cycle [37,104]. These characteristics, together with habitat loss and fishing pressure, combine to threaten the survival of the species intensely, reinforcing its classification as critically endangered by the IUCN [44].

Data Availability
The set of sequences used and analyzed in this study is available on the GenBank website, identified by the access codes listed in Table 1.

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Authors' Contributions
Luis Fernando da Silva Rodrigues-Filho was responsible for the laboratory procedures, data analysis and interpretation, intellectual and methodological development, manuscript writing, and approval of the final version. Davidson Sodré was responsible for the sample acquisition, laboratory procedures, data analysis and interpretation, intellectual development, and manuscript revision. Paula da Costa Nogueira was responsible for the laboratory procedures and manuscript writing. José Rafael da Silva Leal was responsible for the laboratory procedures and manuscript writing. Jorge Luiz Silva Nunes was responsible for the data interpretation and manuscript revision. Getulio Rincon was responsible for the analysis and interpretation of data, intellectual and methodological development, and writing and revision of the manuscript. Rosangela Paula T. Lessa was responsible for the data interpretation and manuscript writing and revision. Iracilda Sampaio was responsible for the data interpretation and writing and revision of the manuscript. Marcelo Journal of Zoological Systematics and Evolutionary Research Vallinoto was responsible for the analysis and interpretation of the data and writing and revision of the manuscript. Jonathan S. Ready was responsible for the laboratory procedures, data analysis and interpretation, and manuscript writing and revision. João Braullio Luna Sales was responsible for the laboratory procedures, analysis and interpretation of data, intellectual and methodological development, and writing and revision of the manuscript. All the authors contributed to the article and approved the final version.