Species delimitation and integrative taxonomy of the Reithrodontomys mexicanus (Rodentia: Cricetidae) cryptic complex
Daily Martínez-Borrego1, Elizabeth Arellano1*, Francisco X. González-Cózatl1, Sandra M. Ospina-Garcés2, Duke S. Rogers3
1Centro de Investigación en Biodiversidad y Conservación, Universidad Autónoma del Estado de Morelos, México.
2Centro de Investigaciones Tropicales, Universidad Veracruzana, México.
3Department of Biology and Monte L Bean Life Science Museum, Brigham Young University, USA.
*Corresponding author: elisabet@uaem.mx
Running title: Integrative taxonomy of the R. mexicanus complex
ABSTRACT
Species boundaries are difficult to establish in groups with very similar morphology. As an alternative, it has been suggested to integrate multiple sources of data to clarify taxonomic problems in taxa where cryptic speciation processes have been reported. This is the case of the harvest mouse Reithrodontomys mexicanus , which has a problematic taxonomy history as it is considered a complex species. Here, we evaluate the cryptic diversity of R. mexicanus using an integrative taxonomy approach in order to detect candidate lineages at the species-level. The molecular analysis used one mitochondrial (cytb) and two nuclear (Fgb-I7 and IRBP) genes. Species hypotheses were suggested based on three molecular delimitation methods (mPTP, bGMYC, and STACEY), and cytb genetic distances values. Skull and environmental space differences between the delimited species were also tested to complement the discrimination of candidate species. Based on the consensus across the delimitation methods and genetic distance values, four species were proposed, which were mostly supported by morphometric and ecological data: R. mexicanus clade I, R. mexicanusclade IIA, R. mexicanus clade IIIA, and R. mexicanus clade IIIB. In addition, the evolutionary relationships between the species that comprise the R. mexicanus group were discussed from a phylogenetic approach. Our findings present important taxonomic implications for Reithrodontomys , as the number of known species for this genus increases. Furthermore, we highlight the importance of the use of multiple sources of data in systematic studies to establish robust delimitations between species considered taxonomically complex.
Keywords: Cricetidae, rodents, cryptic speciation, molecular delimitation, multiples lines of evidence
INTRODUCTION
A continuing challenge in systematics is the determination of which species concept is most appropriate for proposing and delimiting species. Most species have been described using morphological traits (Mayden 1997), but an increasing number of studies are reporting new species using molecular data (see Jörger and Schrödl 2013). Controversy around species concepts is largely due to the different nature of the information upon which different concepts are based (de Queiroz 2007). For example, the Phylogenetic Species Concept (Cracraft 1989) tends to be used when genetic data are analyzed, while studies focusing on reproductive isolation tend to prefer the Biological Species Concept (Mayr 1942). Bradley and Baker (2001) proposed delimiting mammal species using the Genetic Species Concept, specifically using genetic distances estimated from the mitochondrial gene Cytochrome b (cytb; Baker and Bradley 2006). In an attempt to eliminate the species problem, de Queiroz (1998; 2005; 2007) proposed the General Lineage Species Concept (GLC). This unified concept uses different species properties, considering elements of the most accepted concepts, and integrates multiple lines of evidence to establish boundaries between species. Therefore, any criteria identifying a separately evolving metapopulation lineage are considered relevant to identify species (de Queiroz 2005). This multisource approach is known as integrative taxonomy (Dayrat 2005) and has been widely recommended in systematic studies (see Padial et al. 2010; Sangster 2018).
Molecular data have been crucial for studying cryptic species in rodents; they allow both the determination of the number of entities comprising a species complex and the delimitation of those entities (D’Elía et al. 2019; and references therein). Using an integrative taxonomy approach, species phylogenetic hypotheses in rodents have been corroborated by other data sources such as morphology, ethology, biogeography, and ecology (e.g., Almendra et al. 2018; Rivera et al. 2018; Onditi et al. 2021). Thus, the congruence among several lines of evidence indicates robust species hypotheses (Dayrat 2005; Padial et al. 2010).
Reithrodontomys mexicanus (Saussure, 1860) is a cricetid rodent with a discontinuous geographic distribution from Mexico to northwestern South America (Hooper 1952; Hall 1981). Their populations occupy a variety of habitats including humid pine-oak forests, cloud forests, and lowland deciduous forests, and they can generally be found in an altitude range from around 1,000 to 3,800 m. Recent changes in its taxonomy have been proposed due to divergent lineages detected using different genetic loci and craniodental characteristics (summarized in Martínez-Borrego et al. 2020). Arellano et al. (2005), employing information from the cytb gene, identified three well-differentiated clades for this species. Based on those findings, they recommended elevating R. cherrii to the species level and proposed the existence of an undescribed species distributed in the Sierra Madre Oriental and northern Oaxaca, Mexico (Arellano et al. 2005). Using mitochondrial and nuclear genes, Miller and Engstrom (2008) also suggested that R. mexicanus constitutes a cryptic species complex, while Gardner and Carleton (2009) proposed the former R. m. garichensis as a species of the R. mexicanus group based on craniodental differences compared to the remaining Central American species of the subgenus Aporodon .
Thus, as a species complex, R. mexicanus sensu lato needs to be reevaluated from a taxonomic point of view. We hypothesize that each of the divergent lineages within R. mexicanus constitutes valid species given their significant genetic, morphological, and ecological differences. Here, our goal is to assess the cryptic diversity of this species based on Arellano et al. (2005) suggestion that it is composed of at least two species. Putative species will be delimited in accordance with the GLC under an integrative taxonomy approach.
MATERIALS AND METHODS