New karyotype for Mesomys stimulax (Rodentia, Echimyidae) from the Brazilian Amazon: A case for species complex?

Abstract Mesomys Wagner, 1845 (Rodentia, Echimyidae, Eumysopinae) currently has four recognized species, three of which occur in Brazil: Mesomys hispidus (probably a species complex), M. occultus, and M. stimulax. Mesomys leniceps is found in montane forests of northern Peru. Mesomys stimulax, the focus of the present study, has a distribution that is restricted to the central and eastern Amazonia south of the Amazon River, extending from the left bank of the Tapajós River to the right bank of the Tocantins River, and south to the southeast portion of Pará State. The genus presents karyotypes with diploid number 2n = 60 and Fundamental Number (FN) = 116 for M. hispidus and M. stimulax, and 2n = 42, FN = 54 for M. occultus. We studied the karyotype of a female specimen of M. stimulax collected from the Tapirapé‐Aquiri National Forest, Marabá, Pará, Brazil, in the Xingu/Tocantins interfluvium. The obtained karyotype (2n = 60 and FN = 110) differs from that described in the literature for both M. stimulax and M. hispidus by exhibiting more biarmed chromosomes, probably due to pericentric inversions and/or centromeric repositioning, and exhibiting differences in the amount and distribution of constitutive heterochromatin (CH). These results suggest that, similar to what has already been proposed for M. hispidus, M. stimulax may represent a species complex and/or cryptic species. The mechanisms of chromosomal diversification in Mesomys and the biogeographic implications are discussed reinforcing the need for broad systematic review for Mesomys.

M. ecaudatus Wagner, 1845, Echimys ferrugineus Günther, 1876, and M. ferrugineus spicatus Thomas, 1924; Mesomys stimulax Thomas, 1911;Mesomys leniceps Thomas, 1926; and Mesomys occultus Patton, da Silva andMalcolm, 2000 (Patton et al., 2000;Patton & Emmons, 2015;Woods & Kilpatrick, 2005). Of them, M. leniceps is the unique species not reported for Brazil, being restricted to northern Peru; Mesomys hispidus is the most widely distributed and is present throughout nearly all of the Amazonia; Mesomys occultus occurs on the left bank of the Juruá River, south of the Solimões River and Rio Urucú, Tefé, Amazonas, Brazil (Patton et al., 2000); and Mesomys stimulax is restricted to central and eastern Amazonia, south of the Amazon River, with a distribution that extends from the left bank of the lower/medium Tapajós River to the right bank of the Tocantins River, and south to the southeastern region of Pará State (Miranda & Silva, 2015;Patton & Emmons, 2015; Figure 1). Molecular approaches indicate that the species diversity of Mesomys is underestimated, with M. hispidus likely representing a species complex (Orlando et al., 2003;Patton et al., 2000).
The basic karyotypes from three of the four species of the genus have been described. The karyotype 2n = 42 and FN = 54 was reported for individuals of M. occultus collected from the region of the Juruá River (Patton et al., 2000). The karyotype 2n = 60 and FN = 116 was described for individuals of M. hispidus collected from sites at the Samuel Dam in Madeira River (Leal-Mesquita, 1991), the Juruá River south of the Solimões River, the upper Urucu River, Jaú River north of the Solimões River, Brazil, and Tambopata, Peru (Patton et al., 2000;Emmons, personal communication). This same karyotype (2n = 60 and FN = 116) was assigned to specimens of M. stimulax collected from both banks of the lower Xingu River (Patton et al., 2000; personal communication) and from the left bank of the lower Tapajós River (Dias de Oliveira et al., 2019). Only the M. stimulax karyotype (2n = 60; FN = 116) has been analyzed with chromosome banding and molecular cytogenetics (Dias de Oliveira et al., 2019). More detailed cytogenetic studies of these species are needed to improve our understanding of the real karyotypic diversity in this genus and shed light on the mechanisms involved in its diversification.
In the present study, we report a new karyotype for M. stimulax, from an individual collected in Tapirapé-Aquiri National Forest, Marabá, Pará, Brazil. The mechanisms of chromosomal diversification, the biogeographic implications, and the possibility of cryptic speciation are discussed.

| Sample
The sample consisted of a female specimen of Mesomys (Figure 2), which was collected using a live animal trap (Sherman) baited with a mixture of peanut butter, sardine, and cornflour, set in the understory (ca. 1.5 m above the ground) at Igarapé Mano, Tapirapé-Aquiri National Forest, Marabá, Pará (05°46′21″S, 110 50°33′21″W,   (Patton & Emmons, 2015) with highlights indicating the collection sites for karyotyped samples described in the literature and the present work. The map was made using QUANTUM-GIS (Q-GIS) v. 3.8.0 by Willam Oliveira da Silva. The database was obtained from DIVA and REDLIST. Scale bar: 5 cm

| Identification procedures
The specimen was identified by morphological analysis following Patton and Emmons (2015) and Miranda and Silva (2015). In addition, tissue sample was used to extract DNA and obtain a partial  (Table S1).
Sequences of M. leniceps and representatives of clades B, E, and F recognized by Orlando et al. (2003) are not available in the GenBank.
The sequences were aligned and edited in the program BioEdit 7.0.5.2 (Hall, 1999). The data matrix was best represented by the

| Cytogenetic analysis
Chromosomal preparations were obtained from the bone marrow in the field (Ford & Harmerton, 1956). As our sample is a female, the definition of the X chromosome was made by comparing it with the literature. The following techniques were applied, with adaptations: G-banding (Sumner et al., 1971), C-banding (Sumner, 1972), Ag-NOR staining (Howell & Black, 1980), and FISH (Fluorescence In Situ Hybridization) with telomeric probes (All Human Telomere Probe: Oncor, P5091) (Nagamachi et al., 2013) and 18S rDNA probes (Hatanaka & Galetti, 2004). Images of classic cytogenetics were obtained using an Olympus BX41 microscope (bright field/phase) with a digital CCD 1300QDS camera and analyzed using the SpectraView software (Applied Spectral Imaging). Images of FISH were obtained using a Nikon H550S microscope and analyzed using Nis-Elements software. The images were edited using the Adobe Photoshop CS4 program.

| Morphological and molecular identification
The specimen karyotyped in this study is a nonadult individual in age class 5 according to age criteria provided by Patton and Rogers (1983) and Leite (2003) for echimyid rodents, molting to the adult pelage. The adult part of its pelage agrees with M. stimulax descriptions provided by Miranda and Silva (2015) and Patton and Emmons

| Karyotype
The specimen of Mesomys stimulax studied herein has 2n = 60 and FN = 110, with 26 pairs of biarmed chromosomes, three small acrocentric pairs, and a medium-size submetacentric X chromosome

| D ISCUSS I ON
Both the morphological and phylogenetic analyses carried out in the present study allowed to characterize our sample as M. stimulax. The phylogenetic analysis (Table S1 and Figure S1) also showed that two samples from the GenBank must be reidentified. Mesomys hispidus (Dias de Oliveira et al., 2019;Orlando et al., 2003;Patton et al., 2000) shares the same 2n and FN of the previously reported specimens of M. stimulax (Patton et al., 2000) and probably shares the same chromosomal differences with the karyotype of M. stimulax herein described. In our sample, more CH This 2n = 60 and FN = 116 karyotype is also found in the Isothrix bistriata (Patton et al., 2000), whose genus is a sister group of the clade formed by Mesomys and Lonchothrix (e.g., Emmons & Fabre, 2018;Fabre et al., 2016). These data suggest that this is the ancestral karyotype of the clade formed by Isothrix and Mesomys. Therefore, the karyotype described in the present study (2n = 60, As large Amazonian rivers can act as primary or even secondary geographic barriers for rodents (e.g., Antonelli et al., 2018;Leite & Rogers, 2013;Oliveira da Silva et al., 2017;Patton et al., 2000;Patton & Emmons, 2015), the biota of the Amazon region may have a complex evolutionary history (Antonelli et al., 2018). However, one of the clades of M. hispidus is believed to have crossed the Amazon River (or the original population was divided by the Amazon River), as it was distributed from the Guiana Shield to the Bolivian Chaco (Orlando et al., 2003). Thus, it is not clear whether rivers are effective barriers for Mesomys. The existence of Mesomys with the same karyotype on both banks of the Xingu River suggests that this river may not be a strong barrier for this genus. As this is the karyotype that we supposed to be the ancestral for Mesomys, an alternative possibility is that this distribution is consequence of its ancestral condition.
The Oecomys (Malcher et al., 2017). Morphological and molecular studies in specimens of M. hispidus from different locations suggest that this taxon must comprise more than one species (Orlando et al., 2003). The situation may be similar for M. stimulax. In our phylogenetic analysis ( Figure S1), we found a genetic distance of 4.78% between the M. stimulax sequence of Upham and Patterson (2015) and the other samples, which supports the possibility that M. stimulax is a species complex.
Cytogenetic studies of more samples of M. stimulax are needed to define the existence and territorial extent of a possible population with 2n = 60, FN = 110. Moreover, molecular and morphological studies will be useful to properly evaluate and describe the nature of the diversity found in this still largely unstudied group of rodents.

| CON CLUS ION
The species M. stimulax shows variation in its karyotypic formula, suggesting that this name may refer to more than one species. This could have originated from populations in the western region whose distribution expanded from the west to the east of the Amazon. Our results emphasize the need for a systematic and biogeographic study with more samples and an integrative approach.

ACK N OWLED G M ENTS
We would like to thank Dr. Willam Oliveira da Silva for making the map in Figure 1 and Dr. Cleuton Lima Miranda for the image courtesy of Figure 2, for sending us his notes on external morphology of the specimen, and for the critical reading of the manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data used in this research are available in the article. There was no need to deposit in public databases. The authors are available for any further explanation.