Cytotaxonomy of the subgenus Artibeus (Phyllostomidae, Chiroptera) by characterization of species-specific markers

Abstract The genus Artibeus represents a highly diverse group of bats from the Neotropical region, with four large species occurring in Brazil. In this paper, a comparative cytogenetic study was carried out on the species Artibeus obscurus Schinz, 1821, Artibeus fimbriatus Gray, 1838, Artibeus lituratus Olfers, 1818 and Artibeus planirostris Spix, 1823 that live sympatrically in the northeast of Brazil, through C-banding, silver staining and DNA-specific fluorochromes (CMA3 and DAPI). All the species had karyotypes with 2n=30,XX and 2n=31,XY1Y2, and FN=56. C-banding showed constitutive heterochromatin (CH) blocks in the pericentromeric regions of all the chromosomes and small CH blocks at the terminal region of pairs 5, 6, and 7 for all species. Notably, our C-banding data revealed species-specific autosomic CH blocks for each taxon, as well as different heterochromatic constitution of Y2 chromosomes of Artibeus planirostris. Ag-NORs were observed in the short arms of chromosomes 5, 6 and 7 in all species. The sequential staining AgNO3/CMA3/DA/DAPI indicated a positive association of CH with Ag-NORs and positive CMA3 signals, thus reflecting GC-richness in these regions in Artibeus obscurus and Artibeus fimbriatus. In this work it was possible to identify interespecific divergences in the Brazilian large Artibeus species using C-banding it was possible provided a suitable tool in the cytotaxonomic differentiation of this genus.


Keywords
C-banding, Ag-NOR, CMA 3 /DA/DAPI, Cytotaxonomy introduction The genus Artibeus Leach, 1821 has been divided into two main groups based on body size. The species with larger body size have been classified as subgenus Artibeus and the species with smaller body size as the subgenus Dermanura Gervais, 1856. In addition, a new subgenus Koopmania Owen, 1991 was proposed by Owen (1987Owen ( , 1991 to set apart one of the species, A. concolor Peters, 1865. However, this subgenus was later disregarded by Van De Bussche et al. (1998) based on morphological, enzymatic, molecular and karyotypic analysis. The taxonomic classification proposed by Hoofer et al. (2008) recognized Artibeus as a distinct subgenus from Dermanura. Recently, Marchán-Rivadeneira et al. (2010) described the genus Artibeus as constituted by 11 large-body size species including A. concolor as the basal taxon. Its distributional range is restricted to the Neotropical region. The genus is widely distributed from Mexico to northern Argentina, including the Antillean islands in the Caribbean (Simmons 2005).
The extensive similarity of morphometric characters, high degree of shape diversity and overlapping of natural habitats have hindered accurate identification of the large Artibeus along their distribution, particularly in the Neotropical region (Haynes andLee 2004, Hollis 2005). A typical example is the northeastern region of Brazil where four species of the large Artibeus (A. obscurus Schinz, 1821, A. fimbriatus Gray, 1838, A. lituratus Olfers, 1818and A. planirostris Spix, 1823 were formally recorded living in sympatry (Taddei et al. 1998, Araújo andLangguth 2010). In this region, similarity in morphometric measurements (e.g. cranial distances and external dimensions) and geographical variation of A. planirostris are main reasons for confusing taxonomy (Guerrero et al. 2003, Simmons 2005. Since the systematic classification of subgenus Artibeus remains subject of several discussions concerning phylogenetic relationships and actual taxonomic status of species, the use of complementary information may help to define species more precisely (Larsen et al. 2010). For other Mammalian groups, such as primates, felines and rodents, classical and molecular cytogenetic analysis have been successfully allied to taxonomic studies to identify species since chromosomes are not affected by adaptation process to different feeding niches as cranial and general gross anatomies (Granjon andDobigny 2003, Garcia andPessoa 2010).
In this work a karyotypic characterization of A. obscurus Schinz, 1821, A. fimbriatus Gray, 1838, A. lituratus Olfers, 1818 and A. planirostris Spix, 1823 from northeastern of Brazil was performed by the cytogenetic techniques -conventional analysis, C-banding, Ag-NOR and triple staining CMA 3 /DA/DAPI. The data were helpful to carry a comparative analysis of those species, in terms of interspecific differences, and also to provide a better identification of them.

Material and methods
Based on literature (Handley 1989, Taddei et al. 1998, the following characters were used to diagnose the species: presence of fur on the forearms, structure of legs and interfemoral membrane; color of body, dorsal and ventral fur; facial stripes; form of nose leaf and its relationship with the upper lip; shape of the pre-and postorbital process and postorbital constriction and the presence or absence of the 3rd molar. The identification process also included the following 10 measurements: length of forearm, condylobasal length; length of maxillary tooth-row; length of lower tooth-row; length of mandible; breadth across upper canines; mastoidal breadth; zygomatic breadth; postorbital constriction; breadth across upper molars. After identification, cytogenetic studies were carried out on 53 Artibeus specimens from the state of Pernambuco, northeastern Brazil. Voucher specimens are deposited in the Mammalian collection at the Department of Systematic and Ecology, Federal University of Paraíba, João Pessoa, Paraiba, Brazil. The specimens studied were six males and eight females of Artibeus obscurus; two males and four females of A. fimbriatus; eight males and five females of A. planirostris; ten males and ten females of A. lituratus captured at different sites across the Pernambuco State: Igarassu (07°50'02"S, 34°54'21"W), Água Preta (08°42'27"S, 35°31'50"W), Rio Formoso (08°39'50"S, 35°09'32"W), Ipojuca (08°24'00"S, 35°03'45"W) and Recife (08°03'14"S, 34°52'51"W) (see also Appendix).
Metaphase spreads were obtained from bone marrow cells according to conventional procedures and staining with Giemsa. C-banding and silver staining were performed according to Sumner (1972) and Howell and Black (1980), respectively. Triple staining CMA 3 /DA/DAPI was carried out according to Santos and Souza (1998a).
For sequential staining (AgNO 3 /CMA 3 /DA/DAPI), the slides stained by silver nitrate were distained after photographing (Dos Santos Guerra 1991) and re-stained by CMA 3 /DA/DAPI. Photomicrographs were taken using Leica DMLB photomicroscope for C-banding and silver staining. Sequential staining images were captured by IM50 capture system.
C-banding revealed constitutive heterochromatin (CH) in the pericentromeric region of all the autosomes and small heterochromatic blocks were observed in the terminal region of chromosome pairs 5, 6 and 7 (Fig. 1a-d). The karyotype of A. obscurus (Fig. 1a) exhibited interstitial blocks in the short and long arms of pair 1, as well as in the long arms of pairs 2, 5, 6 and in the terminal region of the short arm of pair 9. The A. planirostris karyotype ( Fig. 1d) has the same CH pattern but lacks interstitial blocks in the short arm of chromosome 1. Absence of an interstitial block on the chromosome 6 distinguished karyotype of A. fimbriatus from the other investigated karyotypes (Fig. 1). In all the material examined the long arms of the X chromosomes were more darkly stained when compared with the euchromatin of the autosomes. The Y 2 appeared almost entirely heterochromatic in all species, except for A. planirostris which showed pericentromeric and distal blocks (Fig. 1d). The pattern of the Y 1 could not be determined with precision due to its punctiform size. Table 1 shows exhibits the C-banding pattern in chromosomal complement in all species analyzed.
Silver staining (Ag-NORs) showed three pairs of NORs in the terminal region of the short arms in chromosomes 5, 6 and 7 in all species. As a result of remarkable variation in expression and activity, Ag-NORs were counted up to 100 nuclei, which were randomly selected, and the mean number of Ag-NORs per nucleus was determined for each case ( Table 2).
The sequential staining AgNO 3 /CMA 3 /DA/DAPI showed a correlation between CMA 3 positive regions and Ag-NORs in the karyotypes of A. obscurus and A. fimbriatus (Fig. 2a-f ). Karyotypes of both species had presented CH blocks associated with Ag-NORs sites, reflecting GC-richness in these heterochromatics clusters. In addition, positive CMA 3 signals were observed in the pericentromerics regions of certain autosomes, particularly in pairs 1, 2 and 6 of A. obscurus and in pair 6 of A. fimbriatus (Fig.  2b,e). On the other hand, a uniform pattern was observed in all the chromosomes after DA/DAPI staining (Fig. 2c-d).

Discussion
Our data regarding diploid number, chromosome morphology and sex determination system obtained for Artibeus obscurus, A. fimbriatus, A. lituratus and A. planirostris karyotypes are in agreement with those previously described in the literature (Baker and Hsu 1970, Gardner 1977, Baker et al. 2003. The autosomal complements presented are morphologically similar to other, except by length of the sex chromosomes Y 1 and Y 2 , which varies from punctiform elements to well-defined acrocentric chromosomes, as firstly described by Hsu et al. (1968) for A. lituratus, A. jamaicensis Leach, 1821 and A. toltecus Saussure, 1860.
The multiple sex chromosome system XY 1 Y 2 has been widely reported within the genus Artibeus, e.g. A. aztecus Andersen, 1906, A. glaucus Thomas 1893, A. toltecus, A. concolor, A. cinereus Gervais, 1856, A. hirsutus Andersen, 1906, A. inopinatus Davis et Carter, 1964, and for other 23 species of family Phyllostomidae, as predominant type of sex-determining mechanism in this group (Baker and Hsu 1970, Varella-Garcia et al. 1989, Wetterer et al. 2000, Baker et al. 2003, Noronha et al. 2009). In mammals, this sexual system has been reported in marsupials, insectivores (shrews), carnivores (mongoose), rodents (hamsters) and in artiodactyls (gazelles) (reviewed in Gruetzner et al. 2006). Its origin involves a single sex chromosome-autosome translocation, which in meiosis leads to one sexual trivalent structure formed by XY 1 Y 2 (Rodrigues et al. 2003, Noronha et al. 2004. The CH distribution was evaluated and intercompared in the large Artibeus and with others phyllostomatids, pointing out an extensive similarity of CH pattern local- ized in the pericentromeric region (Rodrigues et al. 2000, Barros et al. 2009, Sbragia et al. 2010. Additionally, CH blocks were also found in the terminal region of chromosome pairs 5, 6 and 7 that has been considered a characteristic shared by subfamily Stenodermatinae (Souza and Araújo 1990, Santos and Souza 1998b, Silva et al. 2005).
On the other hand, a particular set of CH blocks was observed in A. obscurus, A. fimbriatus, A. lituratus and A. planirostris. This finding allowed the individualization and differentiation of each species for karyotype comparison (Fig. 1). A. fimbriatus and A. lituratus karyotypes showed a closer CH distribution differing only by one heterochromatin block. Furthermore A. obscurus and A. planirostris karyotypes presented more interstitial heterochromatin.
The occurrence of intrageneric variation on CH distribution had been described only in sporadic cases among phyllostomatids whose extensive karyotypic conservation is widely known. In turn, the genus Artibeus is widely cited as a chiropteran group that exhibits low rate of karyotype evolution whereas: (1) most of species had same diploid number (30/31) and (2) G-banding patterns are essentially identical Bickham 1980, Baker et al. 2003).
The other parameter evaluated intercomparison was the NORs localization by silver staining. The Ag-NORs were situated on the subtelocentric autosomes 5, 6 and 7 of all species. The data obtained for A. lituratus, A. planirostris and A. fimbriatus, together with the new data of A. obscurus, were similar those described by Santos et al. (2002). These authors employed FISH with 18S ribosomal probe allied to silver staining to investigate the precise localization of rDNA sites, and discovered a noncorrelation between the number and distribution of the NORs in A. cinereus Gervais, 1856, being the first report on silent NORs in bats. They also had distinguished two rDNA sites patterns for Artibeus genus: 1) in the distal regions of the short arms of pairs 5, 6 and 7 (A. lituratus, A. jamaicensis Leach, 1821 and A. fimbriatus) and 2) in the interstitial region of the long arms of pairs 9, 10 and 13 (A. cinereus). In addition, A. fimbriatus had one NOR in the interstitial region on the long arm of pair 5, that it was not observed in this work, which may indicate a chromosomal polymorphism for this species.
As only active NORs could be visualized in our data, the variation in Ag-NORs activity for cell was also investigated (Table 2). In the most of cells analyzed (> 500), the frequency of active NORs was 3 or 4 black spots (26.31 to 36.94 %). Such variability is in accordance with other studies on a NOR sites activity in Phyllostomidae bats that presents multiple NORs (Morielle and Varella-Garcia 1988, Souza and Araújo 1990, Santos et al. 2002. The association between NORs and CH by GC-specific fluorochromes staining presented in this work for A. obscurus and A. fimbriatus, has also been reported to A. NORs and CH were CMA 3 neutral. The reason for that is probably in heterogeneity of base composition of the intergenic regions related to NORs. In some cases, the triple staining with CMA 3 /DA/DAPI has also enhanced the patterns of R-bands with CMA 3, an uniform staining with DA/DAPI or a weak G-banding pattern, as it has been observed in some bat's families (Santos and Souza 1998a, 1998b, Santos et al. 2001, Leite-Silva et al. 2003, Barros et al. 2009).

Conclusion
Classical and molecular cytogenetic markers, associated to taxonomic studies, have provided a better understanding of phylogenetic relationships and the mechanisms responsible for chromosomal divergence in the different taxa in the order Chiroptera. Cytogenetic analysis of all Brazilian species of the subgenus Artibeus allowed us to reveal the conservative and specific chromosomal features among their karyotypes. Furthermore, it was possible to identify intrageneric and interespecific divergences in a group that up to today has been characterized by showing extensive karyotypic conservation. The cytogenetic techniques herein employed, demonstrated the usefulness of C-banding in the identification and correct individualization of the large Artibeus that live sympatrically in the northeastern of Brazil, thus providing an important tool in the cytotaxonomic differentiation of this genus.