Cytogenetic studies in six species of Scinax (Anura, Hylidae) clade Scinax ruber from northern and northeastern Brazil

Abstract

Scinax species are still underrepresented in cytogenetic studies, mainly with respect to populations from northeastern and northern Brazil. In this study, we provide new chromosomal information on Scinax boesemani, S. camposseabrai, S. garbei, S. pachycrus, S. trilineatus and S. x-signatus, all belonging to clade S. ruber. They were collected at two locations in the Caatinga biome (northeastern Brazil) and at one in the Amazon (northern Brazil) biomes. Chromosomes were analyzed by conventional staining, C-banding, Ag-NOR staining, and fluorochrome staining. All species shared a modal diploid value of 2n = 24 and fundamental arm number (FN) of 48. Moreover, both chromosomal size and morphology were similar to other species in this Scinaxclade. C-banding revealed centromeric heterochromatin in all species, along with terminal species-specific C-bands in some species. Active nucleolar organizer regions (Ag-NORs) were identified at 11q in most species, except for S. boesemani and S. garbei (Ag-NORs at interstitial region of 8q). Differing from most anurans, GC-rich regions were not restricted to NORs, but also coincident with some centromeric and terminal C-bands. These data contribute to the cytotaxonomy of Scinax by providing chromosomal markers and demonstrating the occurrence of microstructural rearrangements and inversions on chromosomal evolution of Scinax.

Introduction

The genus Scinax (Anura: Hylidae) comprises 113 species widespread from southern Mexico to Argentina, Uruguay, Trinidad and Tobago and Santa Lucia islands (Frost, 2014). Based on molecular markers, morphology, osteology, myology, reproductive biology and chromosomes, this genus was divided into two clades: S. catharinae (which includes two species groups - S. catharinae and S. perpusillus) and S. ruber (species groups S. rostratus, S. uruguayus and the remaining species) (Faivovich, 2002; Faivovich et al., 2005). The clade S. ruber comprises nearly 65 species (Frost, 2014) widespread over open ares of tropical and subtropical regions (Faivovich, 2002).

So far, karyotypes are known for only 39 species of Scinax and only few studies included banding methods, such as Ag-NOR staining, C-banding, BrdU, fluorochrome staining, and in situ hybridization (Pombal et al., 1995; Kasahara et al., 2003; Nunes and Fagundes, 2008; Cardozo et al., 2011). These cytogenetic studies showed that all species in this genus share a modal diploid value of 2n = 24.

In spite of numerical conservativeness in diploid number, the two clades can be differentiated by cytogenetic analyses. In the clade S. catharinae, the first and second chromosomal pairs are submetacentric and NORs are usually located on the sixth pair, while, in the clade S. ruber, the first and second pairs are metacentric and NORs have been frequentle been detected on pair 11 (Cardozo et al., 2011).

Since Scinax species and populations are still under-represented in terms of cytogenetic data, and species-specific or population differences might be overlooked, we herein provide the first chromosomal information about samples of the following species within the clade S. ruber: S. boesemani, S. camposseabrai, S. garbei, S. pachycrus, S. trilineatusand S. x-signatus, collected in northern and northeastern Brazil.

Materials and Methods

Twenty one individuals of six species of Scinax from the clade S. ruber collected in distinct localities in northern (Bragança, PA) and northeastern (Maracás and Jequié, BA) Brazil were cytogenetically analyzed (Table 1). The specimens were deposited in the Herpetological Collection at Universidade Estadual de Santa Cruz (UESC). Metaphasic chromosomes were obtained from cells of intestinal epithelium as described by Schmid (1978). Slides were stained with 10% Giemsa solution in phosphate buffer 0.1 M (pH 6.8) for about 10 min, washed in distilled water and air dried. The best metaphases were selected and photographed for karyotyping and chromosomal measurements. Active nucleolar organizer regions (Ag-NORs) were located by silver nitrate staining (Howell and Black, 1980), and heterochromatin was detected by C-banding according to Sumner (1972), modified by Siqueira et al.(2008). Fluorochrome staining using chromomycin A₃(CMA₃) and 4′,6-diamidino-2-phenylindole (DAPI) was performed to detect GC- and AT-rich regions, respectively (Schmid, 1980). All images were captured by epifluorescence microscopy (Olympus BX-51 equipped with digital image software Image Pro-Plus version 6.2). The chromosomes were classified based on centromere position as: m (metacentric), sm (submetacentric) and st (subtelocentric), according to Green and Sessions (1991).

Table 1. Analyzed species, number and sex of specimens (N) and collection site.

Species	Voucher identification	N	Locality
S. boesemani	MZUESC12012(M)	1	Bragança - PA
S. camposseabrai	MZUESC11022(M), MZUESC11023(M), MZUESC11024(M), MZUESC11025 (F)	4	Maracás - BA
S. garbei	MZUESC12006(M), MZUESC12007(F), MZUESC12008(F), MZUESC12009(F)	4	Bragança - PA
S. pachycrus	MZUESC 11008(NI), MZUESC11035 (F), MZUESC11036(M) MZUESC11037(M)	4	Jequié - BA
S. trilineatus	MZUESC12013 (M)	1	Bragança - PA
S. x-signatus	MZUESC11001(M), MZUESC11010(F), MZUESC11018(NI)	3	Jequié - BA
	MZUESC12001 (M), MZUESC12002 (F), MZUESC12003 (F), MZUESC12004 (F)	4	Bragança - PA

M: male. F: female.

Results

All analyzed species presented a modal diploid number of 2n = 24 and FN = 48. The chromosomal morphology was similar in nearly all species, comprising eight metacentric pairs (1, 2, 7, 8, 9, 10, 11, and 12) and four submetacentric ones (3, 4, 5, and 6). However, pair 7 in S. boesemani was submetacentric and pairs 11 and 12 in S. x-signatus were submetacentric (Figure 1 and Table 2).

Figure 1. Giemsa stained karyotypes of Scinax species: (A) S. boesemani, (B) S. camposseabrai, (C) S. garbei, (D) S. pachycrus, (E) S. trilineatus and (F) S. x-signatus. The NOR-bearing chromosomes after silver nitrate staining are highlighted in boxes.

Table 2. Chromosomal measurements of analyzed species (RL - relative length; CI = centromeric index and CT = chromosomal type) according to Green and Sessions (1991).

Species		Chromosomal pairs
		1	2	3	4	5	6	7	8	9	10	11	12
S. boesemani	RL	12.34	10.71	10.05	8.90	7.51	6.21	5.82	5.63	5.62	4.65	4.21	3.84
	CI	0.49	0.41	0.29	0.31	0.25	0.32	0.30	0.42	0.47	0.45	0.47	0.46
	CT	M	M	SM	SM	SM	SM	SM	M	M	M	M	M
S. camposseabrai	RL	14.42	11.58	10.51	9.67	7.45	7.25	6.61	6.30	5.14	4.71	4.52	4.08
	CI	0.46	0.38	0.28	0.31	0.25	0.28	0.41	0.41	0.44	0.46	0.45	0.49
	CT	M	M	SM	SM	SM	SM	M	M	M	M	M	M
S. garbei	RL	16.17	13.57	11.51	10.55	8.96	8.82	8.81	7.02	6.85	5.56	5.47	3.61
	CI	0.46	0.41	0.29	0.33	0.25	0.3	0.38	0.46	0.39	0.48	0.46	0.47
	CT	M	M	SM	SM	SM	SM	M	M	M	M	M	M
S. pachycrus	RL	11.33	9.48	8.85	8.04	6.66	6.61	5.11	4.80	3.88	3.78	3.55	2.60
	CI	0.46	0.39	0.25	0.26	0.26	0.28	0.42	0.45	0.45	0.47	0.4	0.4
	CT	M	M	SM	SM	SM	SM	M	M	M	M	M	M
S. trilineatus	RL	9.76	8.36	7.18	6.76	5.68	5.29	4.24	4.02	3.63	3.43	3.41	3.02
	CI	0.47	0.43	0.31	0.3	0.32	0.28	0.38	0.45	0.44	0.47	0.46	0.49
	CT	M	M	SM	SM	SM	SM	M	M	M	M	M	M
S. x-signatus	RL	11.77	9.26	8.50	8.41	6.76	6.45	5.45	4.80	4.11	4.06	3.87	3.00
	CI	0.48	0.41	0.27	0.35	0.27	0.27	0.41	0.38	0.47	0.42	0.28	0.3
	CT	M	M	SM	SM	SM	SM	M	M	M	M	SM	SM

Silver nitrate staining showed that NORs were usually located on the long arms of pair 11 (11q), either at an interstitial position (S. campossabrai, S. trilineatus and S. x-signatus) or in the terminal region (S. pachycrus). NORs were also detected at interstitial positions on 8q in S. boesemani and S. garbei. It is worthy of note that the Ag-NORs were coincident with secondary constrictions observed in Giemsa-stained karyotypes of S. boesemani, S. campossabrai and S. x-signatus (Figure 1).

Heterochromatin was visualized at centromeric regions of most chromosomes in all species (Figure 2). In addition, some species-specific C-bands represented heterochromatin segments at telomeric regions of pairs 3 and 4 in S. boesemani (Figure 2A). S. campossabrai was characterized by the presence of exclusively centromeric C-bands in all chromosomes (Figure 2B). S. garbei and S. pachycrus presented terminal C-bands on arms of pairs 1 and 2 (Figure 2C) and pairs 1, 2, 3, 4, 5, 6, 7, and 9 (Figure 2D), respectively. S. x-signatus also presented heterochromatin at terminal regions of pair 4 (Figure 2E). Heterochromatic blocks interspersed with NORs were observed in S. garbei, S. pachycrus and S. x-signatus (Figure 2C,D,E).

Figure 2. C-banded karyotypes of Scinax species: (A) S. boesemani, (B) S. camposseabrai, (C) S. garbei, (D) S. pachycrus and (E) S. x-signatus.

Fluorochrome staining was successful in S. boesemani, S. pachycrus and S. x-signatus. In these species, GC-rich segments (CMA₃ ⁺ and DAPI⁻) were coincident with NORs (Figure 3). In addition, all centromeric regions of S. boesemani were positively stained, following the C-banding pattern (Figure 3A), while fluorescent signals were observed in several centromeres and terminal regions in 3q and 5 of S. pachycrus (Figure 3B) and in most centromeric regions of S. x-signatus (Figure 3C).

Figure 3. Mitotic metaphases after fluorochrome staining showing CMA₃ ⁺ signals in green and chromosomes counterstained with DAPI (blue): (A) S. boesemani, (B) S. pachycrusand (C) S. x-signatus.

Interestingly, no chromosomal differences could be found among samples of S. x-signatus, in spite of a high degree of isolation by distance of collection sites from distinct biomes (Amazon in Bragança-PA and caatinga in Jequié, BA).

Discussion

We herein provide the first karyotypic data about S. boesemani, S. camposseabrai, S. garbei, S. pachycrus, S. trilineatus and S. x-signatus. The present results confirm the modal diploid values of 24 chromosomes and FN = 48 proposed for Scinax and most species in Hylinae (Duellman, 2001; Faivovich, 2002; Kasahara et al., 2003; Cardozo et al., 2011).

In general, the karyotype formulae are similar among species of the S. ruber clade. Nonetheless, we noted variation in morphology has been reported for pairs 7, 11 and 12 from submetacentric to metacentric (Cardozo et al., 2011), as well as in pair 7 of S. boesemani (submetacentric) and pairs 11 and 12 (submetacentric) in S. x-signatus (Figure 1 and Table 1). These results suggest that microrearrangements such as deletion/duplication of chromosomal segments or inversions may have taken place independently in the abovementioned species.

The results also should prove to be useful for cytotaxonomy and systematics of Scinax species. Brusquetti et al. (2014) revised S. fuscomarginatus and related species, suggesting that S. trilineatus is a synonym for S. fuscomarginatus, since the genetic differences in 16S and COI sequences could be associated with geographic distance among populations. Moreover, these authors found no signification differences in vocalization and morphometric patterns between the two putative species. On the other hand, we observed chromosomal differences between S. trilineatus (8 metacentric and 4 submetacentric pairs) (Table 2), and S. fuscomarginatus (6 metacentric and 6 submetacentric pairs) (Cardozo et al., 2011). Thus, the cytogenetic data support the separation of both cryptic species based on structural rearrangements, such as inversions.

The nucleolar organizer regions are also considered efficient markers for the identification of new species (Bruschi et al., 2012) or cryptic forms (Siqueira et al., 2008), as well as to establish infragenus subdivisions (Raber and Carvalho, 2004; Cardozo et al., 2011). In this sense, the Ag-NOR located at 11q is regarded as an ancestral condition in Scinax (Cardozo et al., 2011). In the present study, two species of the Amazon region (S. boesemani and S. garbei) showed GC-rich NORs on 8q, as confirmed by Ag-NOR, C-banding and fluorochrome staining, suggesting a derived feature. An additional nucleolar region in pair 8 was also reported by Cardozo et al. (2011) in another species of S. ruber clade, Scinax hayii. The two species did not come out as closely related in the phylogenetic analysis carried out by Wiens et al. (2010) and, hence, the changes in Ag-NOR location should be regarded as independent events leading to convergent distribution pattern of NORs.

A single NOR-bearing chromosome was visualized in Scinax trilineatus, and this may be related to the small size of ribosomal cistrons in one homologous or an inactive site, since silver nitrate staining can only reveal previously active NORs (Schmid, 1978; Kasahara, 2009). This Ag-NOR pattern is similar to that reported in Scinax alter and Scinax hiemalis (Cardozo et al., 2011).

Heterochromatin at centromeric region, as detected in this study, has been commonly reported in species of the S. ruber clade, such as S. acuminatus, S. alter, S. curicica, S. duartei, S. eurydice, S. fuscovarius, S. granulatus, S. hayii, S. nasicus, S. perereca, S. similis and S. squalirostris (Kasahara et al., 2003; Cardozo et al., 2011). Another common feature observed in the analyzed species, except for S. camposseabrai, was the presence of heterochromatin interspersed with NORs, as described in S. argyreornatus, S. curicica, S. eurydice, S. hiemalis, S. similis and S. squalirostris (Cardozo et al., 2011).

A distinguishable C-banding pattern was observed in S. boesemani, S. garbei, S. pachycrus and S. x-signatus, as these species presented additional heterochromatic segments at terminal regions of some chromosomes. Even though terminal C-bands have been reported in other hylids (Busin et al., 2006; Kasahara et al., 2003; Gruber et al., 2012), this is the first description in Scinax. Therefore, C-banding patterns can be potentially used to identify species in this genus, and more studies based on this technique should be performed within Scinax.

Fluorochrome staining in amphibians usually identify GC-rich regions associated with NORs and more occasionally C-bands (Ananias et al., 2007; Campos et al., 2008). Indeed, NORs and several centromeric regions in the species anayzed herein were positively stained by CMA₃. In the case of S. pachycrus, C-bands at telomeric regions of pairs 3 and 5 were also GC-rich, indicating a homogeneous base composition of heterochromatin. However, there are only few studies on CMA₃/DAPI staining in Scinax to provide a reliable scenario of heterochromatin composition.

In conclusion, the present work increased the number of karyotyped species in Scinax by providing the first chromosomal data in S. boesemani, S. camposseabrai, S. garbei, S. pachycrus, S. trilineatusand S. x-signatus. Comparisons with previous reports suggest that chromosomal evolution in Scinax (S. ruber clade) may have been mainly driven by microstructural rearrangements and inversions associated with stable karyotype fomulae, particularly among species within a same clade (e.g. S. camposseabrai and S. pachycrus). As indicated by Cardozo et al.(2011), the chromosomal inversions in the clade S. ruberare restricted to pairs 7, 11 and 12. Therefore, banding techniques should prove essential to provide cytotaxonomic markers, as observed in relation to NOR location of the Amazon species and heterochromatin distribution.