Genetic diversity analysis in the Brazilian Amazon reveals a new evolutionary lineage and new karyotype for the genus Mesomys (Rodentia, Echimyidae, Eumysopinae)

Morphological, molecular and chromosomal studies in the genera Lonchothrix and Mesomys have contributed to a better understanding of taxonomic design, phylogenetic relationships and karyotypic patterns. Recent molecular investigations have shown a yet undescribed diversity, suggesting that these taxa are even more diverse than previously assumed. Furthermore, some authors have questioned the limits of geographic distribution in the Amazon region for the species M. hispidus and M. stimulax. In this sense, the current study sought to understand the karyotypic evolution and geographic limits of the genus Mesomys, based on classical (G- and C-banding) and molecular cytogenetic analysis (FISH using rDNA 18S and telomeric probes) and through the sequencing of mitochondrial genes Cytochrome b (Cytb) and Cytochrome Oxidase—Subunit I (CO using phylogeny, species delimitation and time of divergence, from samples of different locations in the Brazilian Amazon. The species M. stimulax and Mesomys sp. presented 2n = 60/FN = 110, while M. hispidus presented 2n = 60/FN = 112, hitherto unpublished. Molecular dating showed that Mesomys diversification occurred during the Plio-Pleistocene period, with M. occultus diverging at around 5.1 Ma, followed by Mesomys sp. (4.1 Ma) and, more recently, the separation between M. hispidus and M. stimulax (3.5 Ma). The ABGD and ASAP species delimiters support the formation of 7 and 8 potential species of the genus Mesomys, respectively. Furthermore, in both analyzes Mesomys sp. was recovered as a valid species. Our multidisciplinary approach involving karyotypic, molecular and biogeographic analysis is the first performed in Mesomys, with the description of a new karyotype for M. hispidus, a new independent lineage for the genus and new distribution data for M. hispidus and M. stimulax.


Introduction
Considered the largest tropical rainforest, the Amazon is home to a diversity of ecosystems that allow a high abundance and richness of species.In this sense, several studies have sought to understand the distribution and processes that gave rise to the diversity of species and the biogeographic patterns observed in the region.From this, biogeographic units were recognized for the Amazon, whose main characteristic is an evolutionary history of the species that occur in these areas [1][2][3].Eight areas of endemism for terrestrial vertebrates are currently recognized in the Amazon biome, namely: Napo, Imeri, Guiana, Inambari, Rondo ˆnia, Tapajo ´s, Xingu and Bele ´m [4].These regions are delimited by the great Amazonian rivers and characterized as units that reflect the evolutionary history of the species found there [4,5].
Among the groups of terrestrial vertebrates, the rodents of the family Echimyidae are considered the most diverse South American hystricognaths, comprising 95 species grouped in 21 genera.This family is widely distributed in the Amazon rainforest, occurring in all recognized areas of endemism [6].Investigations with mitochondrial genes Cytochrome b (Cytb) and Cytochrome C Oxidase-Subunidade I (CO1) and nuclear genes Apolipoprotein B (apoB), Interphotoreceptor Retinoid-Binding Protein (IRBP), Recombination Activating 1 (RAG1) and Von Willebrand Factor (vWF) fragments carried out in Echimyidae have resulted in the recognition of three or four subfamilies, depending on the author and analyzes performed [6][7][8][9][10].
The subfamily Eumysopinae is composed of nine genera, four of which are terrestrial (Hoplomys, Proechimys, Thrichomys and Trinomys), three are semifossorial (Carterodon, Clyomys and Euryzygomatomys), and two are arboreal (Mesomys and Lonchothrix) [6].The genus Mesomys comprises four species (M.hispidus, M. stimulax, M. occultus and M. leniceps), which are widely distributed in the Brazilian Amazon and regions from the Guianas to the eastern part of the Andes.Mesomys hispidus occurs widely in the Amazon region, except on the right bank of the lower Tapajo ´s river south of the Amazonas river [6].Mesomys stimulax occurs from the eastern Amazon south of the river Amazonas, particularly on the right bank of the lower Tapajo ´s river up to the right bank of the Tocantins river and may be sympatric with M. hispidus on the right bank of the upper Tapajo ´s river [6].However, there are inconsistencies regarding the limits of distribution and sympatry between M. hispidus and M. stimulax, given that sympatry zones were described for these species on the left bank of the lower Tapajo ´s river south of the Amazon river [11].
In view of the difficulties in taxonomic identification, mainly associated with the overlapping of ecological and morphological characteristics and regions of sympatry, studies have sought to employ different methods for a better delineation of taxa, associating morphological [9,10], molecular [6][7][8][9][10], and karyotypic data [11,12].Analysis of Cytb sequences and morphometric data from M. hispidus samples demonstrated a high intraspecific divergence with the structuring of different clades, associated with variation in cranio-dental size, suggesting that it is a species complex [10].Despite this variation, the authors allocated all forms provisionally detected within M. hispidus, but suggested that studies using morphological, molecular and chromosomal analyzes should be carried out.
Considering the underestimated genetic diversity and inconsistencies in geographic limits in the genus Mesomys, the present study analyzed samples of Mesomys from different locations in the Brazilian Amazon by classical and molecular cytogenetics, through G-and C-banding and FISH with telomeric and 18S rDNA probes and the phylogenetic relationship of the genus using Cytb and COI genes fragments, in addition to species delimiters and timetree.We describe possible chromosomal rearrangements that contributed to evolution and diversification, in addition to investigating the distribution of the species M. hispidus and M. stimulax from our samples distributed in the Brazilian Amazon, correlating with areas of endemism.

Ethics statement
The specimens were kept stress-free with full access to food and water until euthanasia was performed in accordance with animal welfare guidelines established by Brazilian
The molecular analysis of the present study used 16 tissue samples from Mesomys for DNA extraction to obtain sequences of the mitochondrial genes (Cytb and COI).The data were complemented with sequences available in GenBank (Fig 1 ; S2 Table ).

Molecular analysis
Total DNA extraction was performed using Promega's Wizard Genomic Kit, according to the manufacturer's protocol.PCRs were standardized with 15 μl of final volume, containing ~30ng of genomic DNA; 2.4 μl of dNTPs (1.25mM); 1.5 μl 10X buffer (200 mM Tris-HCl, 500 mM KCl); 1 μl MgCl2 (25 mM); 1 μl of each primer (0.2 μM) and 1 U of Taq DNA polymerase.With the exception of the annealing temperature of the primers, the other steps of the amplification protocol were identical for the two markers, with the following conditions: initial denaturation at 95˚C for 5 min, with 35 cycles of denaturation at 94˚C for 30 s, hybridization temperature for 40 s, extension of 72˚C for 40 s, and final extension at 72˚C for 5 min.Amplifications of 792 bp of the mitochondrial Cytochrome b gene were performed with primers 5'-CGAAGCTTGATATGAAAAACCATCGTTG-3' (MVZ 05) and 5'-AAATAGGAARTATCAYTCTGGTTTRAT-3' (MVZ 16) [14] and of 631 bp of the mitochondrial Cytochrome Oxidase-Subunit I gene was performed with primers 5'-TCAACCAACCACAAAGACATTGGCAC-3' (FishF1) and 5'-TAGACTTCTGGGTGGCCAAAGAATCA-3' (FishR1) [15].Both PCR products were purified with the aid of polyethylene glycol (PEG) and alcohol.Subsequently, sequence reactions were conducted with the BigDye Kit (Applied Biosystems), and the samples were sequenced on the ABI 3500 XL automatic sequencer (Applied Biosystems).
DNA sequences were aligned in ClustalW [16] and manually edited in BioEdit v. 7.2.5 [16].Phylogenetic analyzes were performed by maximum likelihood (ML).The search for the best nucleotide substitution model was performed using the jModeltest 2.0.2 software [17].Base saturation tests were performed on DAMBE5 [18] and there was no saturation in the two markers.The ML analysis was performed in RAxML v.8, the most plausible tree was found by 1000 searches and later 1000 bootstrap pseudo-replicas were performed [19].Percentages of genetic divergence between taxa were verified in the (MEGA) program [20].

Divergence time analysis and species delimitation
The Automatic Barcode Gap Discovery (ABGD) analysis from the concatenated dataset (Cytb and COI) was performed in the online version (https://bioinfo.mnhn.fr/abi/public/abgd/abgdweb.html),using the following settings: Pmin = 0.001, Pmax = 0.1, Steps = 10, NBins = 20, Relative gap = 0.5, simple distance, the other parameters were not adjusted [21,22].The assemble species by automatic partitioning (ASAP) analysis was performed in the online version (https://bioinfo.mnhn.fr/abi/public/asap/)using the distance matrix with the simple distance model (p-distance) and settings standardized [21,22].Divergence time analyses were run in BEAST v. 1.10 [23].The uncorrelated relaxed clock [24] was the prior clock type, and the Yule process was the prior model for the tree [25].We use the fossil of the Paralonchothrix dated from 7.0 Ma for the separation between the genera Mesomys and Lonchothrix [26] as our calibration point.The run was processed with 100,000,000 generations, sampling every 5,000 generations.Trees were summarized in TreeAnnotator v.1.8.4 [23].A burn-in of 20% was used and the run was considered satisfactory when all Effective Sample Size (ESS) values checked in Tracer v.1.6[27] were equal to or larger than 200.

Cytogenetic analysis
Chromosomal preparations were obtained through bone marrow samples [28].Subsequently, the slides were prepared and submitted to G- [29] and C-banding [30], in addition to FISH (fluorescence in situ hybridization) with telomeric and 18S rDNA probes [31].All techniques followed the established protocols with adaptations.
The slides containing the chromosome preparations used in classical cytogenetic techniques were analyzed under an Olympus BX41 microscope and photographed with a Canon Powershot A95 digital camera.The karyotypes were organized using the Applied Spectral Imaging Karyotyping software, considering the morphology and size of the chromosomes.Previously hybridized slides were captured using Nis-Elements software under the Nikon H550S microscope.Chromosomes are separated into two groups: one-armed (subtelocentric/ telocentric/acrocentric) and bi-armed (metacentric/submetacentric).
The chromosomes classification was performed according to Levan et al. [32], which takes into account the quotient between the length of the long chromosomal arm (L) and the short arm (C), as the value of the L/C ratio, with the designations of chromosomes was metacentric, submetacentric, subtelocentric and telocentric due to the position of the centromere, if it occupies the median, submedian, subterminal and terminal positions, respectively.

Molecular phylogeny, species delimitation and divergence time tree
Maximum likelihood phylogenetic analysis recovered a topology with high bootstrap values for most nodes.The sister relationship between the genera Lonchothrix and Mesomys was recovered with 99% bootstrap.The sample from the municipality of Itaituba (Locality 13, Fig 1 ) did not correspond to any recognized clade of the genus Mesomys, and formed a branch with 94% support, suggesting that this is a new species for the genus (Fig 2).
The genetic distance analysis performed with the mitochondrial gene of Cytochrome b (Cytb) showed divergence among Mesomys and the other genera ranged from 15% between M. stimulax and L. emiliae to 24% between M. occultus and I. sinnamariensis (Table 1).Within Mesomys, the highest interspecific distance was 14% between M. hispidus and M. occultus, and the lowest was 7% for Mesomys sp. and M. hispidus.The latter species showed the highest value of average intraspecific nucleotide divergence (5%).The putative new species of Mesomys, herein referred as Mesomys sp., differed from the other congeneric species by 7% from M. stimulax up to 12% from M. occultus (Table 1).
The genetic distance analysis performed using the mitochondrial gene Cytochrome C Oxidase-Subunit I (COI) sequences showed divergence among Mesomys, and the other genera ranged from between 12% Mesomys sp. and L. emiliae to 22% between Mesomys sp. and C. sulcidens and also between E. spinosus and C. sulcidens (Table 2).Within Mesomys, the greatest interspecific distance was 11% between Mesomys sp. and Mesomys hispidus and also between M. stimulax and M. hispidus, and the smallest was 9% for Mesomys sp. and M. stimulax.Mesomys hispidus showed the highest value of average intraspecific nucleotide divergence (9%) (Table 2).
The timetree analysis suggests that the diversification of M. occultus occurred about 5.1 Ma with error bar (95% HPD: 6.5 Ma-3.5 Ma) in relation to the other species of the genus.Mesomys sp.diverged from M. hispidus and M. stimulax about 4.1 Ma (5.6 Ma-2.6 Ma), and the latter diverged from each other around 3.5 Ma (4.9 Ma-2.2 Ma) (Fig 3).
The species delimitation performed through the ABGD analysis from the concatenated dataset (Cytb and COI) resulted in the recognition of seven potential species, while the ASAP analysis recovered eight potential species within the genus.Mesomys occultus and Mesomys sp. were retrieved as valid species in both analyses.Mesomys stimulax was recovered as a species complex, composed of two groups in both analyses.Finally, M. hispidus was also recovered as a species complex, composed of three groups according to ABGD and four groups according to ASAP (Fig 3).

Classical and molecular cytogenetics
The sample of Mesomys hispidus from the municipality of Jacareacanga (Locality 1,  The

Phylogenetic analysis and species delimitation
Maximum likelihood (ML) analysis using the concatenated Cytb + COI markers showed the monophyly of the genus Mesomys well supported, corroborating the literature data [7,8].Our samples of M. hispidus and M. stimulax were correctly associated with the sequences available in GenBank for these species, confirming the taxonomic identification from DNA barcoding [33][34][35].
A lineage unknown to the genus Mesomys for the sample from the Tapajo ´s endemic area, in the municipality of Itaituba-PA, proves to be a sister lineage of the M. hispidus + M. stimulax grouping with 97% and it was recovered diverging in a branch with 94% support.The interspecific divergence based on the mitochondrial Cytb gene between the Mesomys species included in this study and the new lineage detected ranged between 7% and 12%, while the divergence values recorded for the mitochondrial COI gene ranged from 9 to 11% (Tables 1 and 2, respectively).Reinforcing the findings of the ML analysis, the species delimiters ABGD and ASAP (Fig 3) support the hypothesis of Mesomys sp.(JL42) as a distinct lineage.
According to the average intraspecific variation for the analyzed species, M. stimulax was the one that presented the lowest rate, with 4% in both genes (Cytb and COI) for seven samples from five localities.The species M. hispidus showed the highest intraspecific nucleotide divergence with 5% for Cytb and 9% for COI for twenty samples from sixteen localities.The high intraspecific divergence found in M. hispidus corroborates other studies, where clades with 7.2% nucleotide divergence were described, three of which are widely distributed and three are geographically isolated in the Amazon region [9][10][11].Values above 5% of nucleotide divergence for mitochondrial genes in rodents are indicative of possible cryptic species [36,37].these two possible species appear to be older groups related to the north of the Amazon River.Meanwhile, another large subclade associated with samples south of the Amazon River appears to have its most recent diversification splitting from samples to the north at approximately 2.5 Ma (Fig 3).Despite the low support of nodes (80% bootstrap), the existence of these groups evidenced by species delimiters reinforces the idea of species complexes in M. hispidus, as already postulated [9][10][11]38].

Geographical patterns, divergence time and speciation hypotheses
The genus Mesomys is widely distributed in the Amazon and in the eight areas of endemism.Mesomys leniceps occurs in the Napo area of endemism; M. occultus occurs in the areas of endemism Napo and Inambari; M. hispidus is distributed in six areas of endemism, with the exception of Bele ´m and Xingu areas; and M. stimulax is distributed in the Xingu, Tapajo ´s, Rondo ˆnia, Bele ´m and Guianas areas of endemism [6,11,12].
Regarding the distribution and sympatry between the species M. hispidus and M. stimulax, studies describe M. hispidus with the eastern limit of occurrence on the left bank of the Cristalino river, affluent of the right bank of the Teles Pires river, and M. stimulax occurring in part of central and eastern Amazonia, from the west bank of the lower Tapajo ´s River to the east bank of the Tocantins River [6].However, our analysis extends the distribution of M. hispidus to the right bank of the middle Tapajo ´s river, Para ´state, on both banks of the middle and upper Teles Pires river, Mato Grosso state, and Marajo ´island, Para ´state, in the areas of endemism Tapajo ´s and Marajo ´Island, respectively.Regarding M. stimulax, our analysis extends the occurrence of this species to the west of the Amazon, on the west bank of the lower Madeira river, state of Amazonas, and the left bank of the river Juruena, state of Mato Grosso, in the areas of endemism Inambari and Rondo ˆnia, respectively.Thus, we propose to expand the distribution of these species to these locations not recognized so far (Fig 7).
Recent studies have described new species for the Tapajo ´s region, corroborating its characterization as an endemic area, increasing the number of endemic species of different taxonomic groups for the region, including small non-flying mammals [39][40][41][42].Some of these studies proposed the hypothesis that the Amazon rivers act as a geographic barrier, being the main factor that isolated the populations, influencing the differentiation of the species that occur in this area of endemism [41][42][43][44].On the other hand, the hypothesis of rivers as a barrier is questioned, suggesting that the modifications depend on the characteristics of each river and the studied species [45,46].
Analysis of mitochondrial sequences of Mesomys hispidus collected along the Jurua ´river tested the hypothesis of the river as a barrier [38].In this study, clades grouping samples from both banks of the river were described and the gene flow data were uncertain to distinguish the intensity with which they occurred on each bank.Thus, they concluded that the Jurua Ŕiver was not acting as a barrier for M. hispidus.With regard to the molecular dating of arboreal Echimyidae, the time of diversification is estimated for the Miocene, between 17.1 and 15.3 Ma, being related to vicariant events, such as the emergence of river drainage and the formation of flooded forests, which acted on the ecomorphological modifications of these taxa, modifying the way of life for the tree extract [6][7][8][47][48][49].Our divergence time data show that Mesomys diversification started around 5.1-3.5 Ma during the transition from the Pliocene to the Pleistocene period, before the formation of the Tapajo ´s River, which occurred around 1.3-0.8Ma, and this physical barrier is not the determining factor for radiation in the genus [44].
A phylogenetic and biogeographical analysis of rodent radiation in the transition from Amazonian and Andean centers of endemism suggested that diversification between species of the genus occurred between 7 Ma-2.5 Ma, while more recent populations are observed ranging from 4 Ma-1 Ma, possibly influenced by Plio-Pleistocene climate changes, causing them to become isolated [50].In this sense, our data are congruent with those described [50], where the analyzed species diversified between 6.5 Ma-2.2 Ma and the populations varied between 3.2 Ma-1 Ma, which allows us to indicate the influence of effects that occurred in the Plio-Pleistocene period.
In view of the difficulties of taxonomic identification and the molecular issues associated with the genus Mesomys, multidisciplinary approaches play a role in elucidating the diversity and evolution of this understudied genus [51,52].Additionally, it is worth noting that different rodent groups exhibit cryptic species, and the use of these tools becomes crucial in identifying potential new species [53].

Cytogenetic analysis
The data of the present work show a new cytotype for M. hispidus, from the municipality of Jacareacanga-PA (Locality 1, Fig 1 ), having 2n = 60/FN = 112.This sample has two acrocentric pairs (28 and 29), which differs from the karyotypes described in the literature for M. hispidus from Juruti-PA (Locality 14, Fig 1) [11] and M. hispidus from the upper Jurua ´river [9] that present all chromosomal pairs with bi-armed morphology (2n = 60/FNa = 116).The conservation of 2n = 60, with variation in NF = 112 and 116 may be due to chromosomal rearrangements like pericentric inversion, being one of the chromosomal variations common in rodents [54][55][56].
The data here described for FISH with telomeric probe and 18S rDNA corroborated what was described in the literature for populations of M. hispidus from Juruti-PA (Locality 14, Fig  [11,12]. Thus, the phylogenetic and karyotypic differences found in Mesomys sp.(2n = 60/ NF = 110) may be related to the Amazon refuge hypothesis due to multiple ecological and biogeographical events that occurred during the Plio-Pleistocene periods (11.7 Ma-2.5 Ma) acting on the dispersion and isolation of populations.A study carried out [61] pointed to hypotheses of sympatric and allopatric speciation in rodents of the genus Proechimys in the endemic areas of Rondo ˆnia and Tapajo ´s, suggesting that chromosomal rearrangements that occurred within subpopulations established in allopatry could serve to reinforce the blockage of gene flow in conditions of secondary contact.In this case, considering that Mesomys sp. also occurs in the Tapajo ´s area of endemism in sympatry with Mesomys hispidus, the chromosomal differences observed between these species may be associated with vicariant effects that allowed the occurrence of independent chromosomal rearrangements in these sister species [8,[57][58][59][60][61].

Conclusions
We described a new cytotype for M. hispidus from Jacareacanga with karyotypic form 2n = 60/ NF = 112, unpublished for the genus.Our molecular and phylogenetic data revealed a new lineage for the genus Mesomys based on a sample from the municipality of Itaituba-PA, within the Tapajo ´s area of endemism.This new lineage is probably in sympatry with the species M. hispidus and M. stimulax, suggesting that it is a species not yet described for the genus, here called Mesomys sp.nov.Furthermore, M. hispidus showed a high degree of intraspecific divergence, suggesting that it may represent a species complex.Based on our data, we propose expanding the geographic distribution of M. hispidus to the right bank of the middle Tapajo ´s River, state of Para ´, to both margins of the middle and upper Teles Pires River, state of Mato Grosso, and to the Marajo ´island, state of Para ´.Additionally, we propose expanding M. stimulax geographic distribution to the western Amazonia, particularly on the lower Madeira River, state of Amazonas, and on the banks of the Juruena River, state of Mato Grosso, in the Inambari and Rondo ˆnia areas of endemism, respectively.

Fig 2 .
Fig 2. Maximum likelihood tree of the samples from the present study and those available in GenBank for the genus Mesomys, based on sequences of concatenated mitochondrial genes (Cytb and COI).The numbers near the nodes are bootstrap values for 1,000 replicates.Bootstrap values less than 80% are not shown.The numbers in square brackets precede the locations are referred to in Fig 1.https://doi.org/10.1371/journal.pone.0291797.g002 Fig 1) presented a karyotype with 2n = 60/FN = 112, with 29 autosomal pairs, ranging from large to small, with pairs 1 to 27 meta/submetacentric, the 28 and 29 acrocentric pairs and the large X chromosome as a submetacentric (Fig 4a).Constitutive heterochromatin (CH) is distributed in the centromeric region of all autosomal and sex chromosomes (Fig 4b).FISH with telomeric probes (Fig 4c) showed only distal signals and those with 18S rDNA probes (Fig 4d) marked in the interstitial region of an autosomal pair.The sample of Mesomys stimulax from the municipality of Paragominas (Locality 11, Fig 1) presented the karyotype with 2n = 60/FN = 110, with 29 autosomal pairs, ranging from large to small, with pairs 1 to 26 meta/submetacentric, pairs 27 to 29 acrocentric, a large X chromosome as a submetacentric and the Y as a small submetacentric (Fig 5a).CH is distributed in the centromeric region of all autosomal and sex chromosomes (Fig 5b).FISH with telomeric probes (Fig 5c) showed only distal signals and those with 18S rDNA probes (Fig 5d) marked in the interstitial region of an autosomal pair.
sample of Mesomys sp. from the municipality of Itaituba (Locality 13, Fig 1), presented the karyotype with 2n = 60/FN = 110, with 29 autosomal pairs, ranging from large to small, with pairs 1 to 26 meta/submetacentric, pair 27 subtelocentric, pairs 28 and 29 acrocentric, a large X chromosome as a submetacentric and the Y as a small metacentric (Fig 6a).CH is distributed in the proximal region of all autosomal and sex chromosomes (Fig 6b).FISH with telomeric probes (Fig 6c) showed only distal signals and those with 18S rDNA probes (Fig 6d) marked in the interstitial region of an autosomal pair.

Fig 4 .
Fig 4. Mesomys hispidus (2n = 60/FN = 112).a) G-banding, b) C-banding, c) Telomeric FISH and d) 18S rDNA FISH.https://doi.org/10.1371/journal.pone.0291797.g004 Furthermore, we observed the formation of geographically structured clades within Mesomys hispidus congruent with species delimiters.The first subclade to diverge about 3 Ma from the other branches in M. hispidus points to two potential species according to the ABGD and ASAP delimiters; the first corresponding to the sample from the Guyana shield (Locality 17, Fig 1), and the second corresponding to samples from the trans-Andean region (Localities 16 and 18, Fig 1), diverging from each other at approximately 2.3 Ma.Together with the samples from locations 10 and 19 (Fig 1)

1 )
and M. stimulax from Maraba ´-PA (Locality 20, Fig 1), showing distal telomeric signals and 18S rDNA signal in a chromosomal pair resolution CFMV n.1000/2012 and CFBio n. 301/2012.The necessary euthanasia was performed by intraperitoneal injection of buffered and diluted barbiturates after local anesthesia, in accordance with animal welfare guidelines established by the Animal Ethics Committee (Comitê de E ´tica Animal) from Universidade Federal do Para ´(UFPA), which authorized the present study (Permit 68-2015).JCP has a permanent scientific collecting permit (number 13248) from "Instituto Chico Mendes de Conservac ¸ão da Biodiversidade".The Cytogenetics Laboratory from UFPA has a special permit to sample transport (number 19/2003) and to use the samples for research purposes (number 52/2003) from the Ministry of Environment.

Table 1 . P-distance estimates for the Cytochrome b (Cytb) gene among the monophyletic clades recovered in the phylogenetic analysis.
VI = intraspecific variation.Scores in bold represent the lowest and highest values of interspecific and intraspecific divergence for the different species of Mesomys analyzed.https://doi.org/10.1371/journal.pone.0291797.t001

Table 2 . P-distance estimates for the Cytochrome C Oxidase-Subunit I (CO1) gene among the monophyletic clades recovered in the phylogenetic analysis.
Scores in bold represent the lowest and highest values of interspecific and intraspecific divergence for the different species of Mesomys analyzed. https://doi.org/10.1371/journal.pone.0291797.t002