Turkish Journal of Zoology Turk J Zool (2013) 37: 470-487 http://journals.tubitak.gov.tr/zoology/ © TÜBİTAK Research Article doi:10.3906/zoo-1208-25 Chromosomal evolution of the genus Nannospalax (Palmer 1903) (Rodentia, Muridae) from western Turkey Ferhat MATUR*, Faruk ÇOLAK, Tuğçe CEYLAN, Murat SEVİNDİK, Mustafa SÖZEN Department of Biology, Faculty of Arts and Sciences, Bülent Ecevit University, Zonguldak, Turkey Received: 29.08.2012 Accepted: 17.02.2013 Published Online: 24.06.2013 Printed: 24.07.2013 Abstract: We used 33 blind mole rats belonging to 10 different chromosomal races from 10 localities in western Turkey. We applied G- and C-banding techniques to compare chromosomal races as well as clarifying relationships between them. We discussed cytogenetic similarities and differences between chromosomal races. We concluded that 2n = 60C is the ancestor of the other chromosomal races. However, as a result of ongoing evolution processes 2n = 38 and 2n = 60K have become ancestors to chromosomal races on their peripherals. We discovered which rearrangements contribute to the evolution of such a complex chromosomal race system in a genus. With this study we provide a comprehensive comparison of the 10 chromosomal races and perform a cladistic analysis using chromosomal rearrangement character states. According to our tree, chromosomal races with a low diploid number formed a monophyletic group. Key words: Blind mole rat, comparative cytogenetic, G- and C-banding, chromosome differentiation, phylogeny, Anatolia 1. Introduction assumed that ancestral karyotype diverged into the 2n The genus Nannospalax includes blind rodents that have = 60W and R chromosomal races, and independent adapted to living underground. Currently more than 30 translocations of short arms of some chromosomes caused chromosomal races have been determined in Turkish blind this differentiation. Arslan et al. (2011) studied variation mole rats but there is still doubt about the taxonomy of this of C- and AgNOR-banding of 3 chromosomal races (2n = taxon (Nevo et al., 1994; Sözen et al., 1998a, 1998b, 1999, 40, 58, and 60) of N. xanthodon from southern Anatolia. 2000a, 2000b, 2006a, 2006b, 2011; Sözen, 2004; Matur and They found differentiation among the chromosomal Sözen, 2005; Kankiliç et al., 2007; Ivanitskaya et al., 2008; races. Matur et al. (2011) banded 4 chromosomal races Arslan et al., 2011). Kandemir et al. (2012) discussed the with 2n = 50 from different localities in Anatolia. These taxonomic name problem of blind mole rats and so here 4 chromosomal races had the same diploid numbers but we will follow Kandemir et al. (2012). their G-banding patterns were different. The complements Ivanitskaya and Nevo (1998) analyzed Jordanian blind of all these chromosomal races included 2 identical mole rats using C-, G-, and AgNOR-banding techniques metacentric autosomes and the sex chromosomes were and compared these data with previous results obtained also always the same. The studied chromosomal races in Turkish and Israeli blind mole rats. They found should have their own evolutionary pathway but they have that NF values were useful for differentiation due to a common ancestor. pericentric inversions and centromeric shifts. So far, only Dobigny et al. (2004) indicated that chromosomal a few banding studies of Turkish N. nehringi have been data have been underutilized in phylogenetic research, performed. These were conducted in populations from and chromosomal changes could be used as a character. Malatya (Ivanitskaya et al., 1997) and Kastamonu and We set out to identify chromosomal characters that Çankırı provinces (Ivanitskaya et al., 2008). Additionally, could be used to reconstruct the evolutionary history of a banding study was performed by Ivanitskaya et al. (1997) these blind mole rats. The aim of the present study was with southeastern Anatolian blind mole rats (N. ehrenbergi) to compare 9 chromosomal races of N. xanthodon and using G-, C-, and AgNOR-banding techniques. Ivanitskaya a chromosomal race from N. leucodon by determining et al. (2008) assigned the 2n = 60 populations in Turkey to which Robertsonian translocations are prevailing (fissions 2 chromosomal races as 2n = 60W and 2n = 60R, based vs. fusions). By finding the main chromosomal changing on G-bands, C-bands, AgNOR staining, fluorochrome mechanism, we may explain chromosomal evolution of staining, and FISH of telomeric and rDNA probes. They the genus Nannospalax in western Turkey. * Correspondence: [email protected] 470 MATUR et al. / Turk J Zool 2. Materials and methods and C- (Sumner, 1972) banding techniques were applied In this study, 33 animals were studied from 10 localities to each specimen. Pictures of metaphases were taken using (Table 1; Figure 1). According to their geographical a Canon DP 21 digital camera. location in Turkey, these chromosomal races were The G-banding patterns allowed us to assess all the designated as N for north (52N, 54N, 58N), W for west chromosomal homologies among chromosomal races and (50W), E for east (50E), Tr for Thrace (56Tr), C for central to identify the structural differences among karyotypes. (60C), and K for Kastamonu (60K). We recognized the 2n The 2n = 60J fromN. ehrenbergi from Jordan was used = 36, 2n = 38, 2n = 40, 2n = 50W, 2n = 52N, 2n = 54N, as an outgroup (Ivanitskaya and Nevo, 1998). In order 2n = 56W, 2n = 58N, and 2n = 60C chromosomal races to determine whether fusion or fission is the main from N. xanthodon and 2n = 56Tr from N. leucodon. rearrangement we identified the specific arm combination Karyotypes were prepared from bone marrow according of a particular metacentric (Figure 2). If an arm was to Ford and Hamerton (1956). Then G- (Seabright, 1971) included in different metacentrics, fusion was accepted Table 1. Localities, sample size (M: males, F: females), diploid chromosome numbers (2n), and chromosomal arm numbers (NF) of animals examined. 2n NF Localities Province N 36 70 Kemer Cemetery AYDIN 3 38 74 Kırkağaç, Gelenbe MANİSA 3 40 72 Beyşehir 12 km SW KONYA 4 50W 74 Alaşehir MANİSA 5 52N 72 Yalova YALOVA 3 54N 70 Eflani KASTAMONU 2 56W 72 Kula 7 km S MANİSA 5 56Tr 78 Hayrabolu KIRKLARELİ 3 58N 72 Taşköprü KASTAMONU 3 60C 78 Kızılcasöğüt 1 km S UŞAK 2 30 35 40 56 Tr 38 40 36 56W 50W 52N 60K 54N 60 58N 100 0 100 200 300 km Figure 1. Map of the study area in Turkey and the geographical distribution of the chromosomal races studied. 471 MATUR et al. / Turk J Zool Populaton A Populaton B Populaton C a c e a c a Fsson b d f b d e f b c d e f Metacentrc chromosomes Populaton B Populaton C a c d c e Populaton A Fuson b e a b c d e f f d b a f Acrocentrc chromosomes Figure 2. Schematic diagram showing how to specify the rearrangement that plays a main role. If in a hypothetical Population A we always found “a” and “b” arms together like Population B and C then we can claim that fission is responsible for differentiation, or if the reverse situation is observed—“a” and “b” arms are combined with different arms in different condition such as Population B or Population C—it can be said the fusion is the main rearrangement responsible. as the responsible mechanism (references within Sumner 3. Results (2003)). We coded all the characters as indicated by Dobigny 3.1. Karyotype results of C-banding patterns and et al. (2004). First, the chromosomes or chromosomal G-banding comparisons segments were treated as characters, and their presence/ After C- and G-banding, heterochromatin variation in absence or the changes they had undergone represented the heterochromatin distribution (Figures 3a–j) and the character states. Secondly, the chromosomal changes rearrangements among chromosomal races (Figures themselves were considered to represent the characters. 4a–i) were identified. In karyotypes of the 2n = 60, 9 of Using both approaches, we treated 30 chromosome states 30 chromosomes were bi-armed. The X chromosome and 60 chromosomal rearrangements as absent/present. was submetacentric, while the Y chromosome was The matrix of chromosomal characters (Table 2) was subtelocentric. According to the C-band pattern of the 2n analyzed by maximum parsimony using the heuristic = 60, 16 pairs of chromosomes (pairs 1, 2, 3, 5, 6, 7, 10, search option in PAUP 4.0b.10 (Swofford, 2001) with 11, 14, 17, 18, 19, 20, 22, 23, and 26) had heterochromatin bisection–reconnection (TBR) and 10,000 random taxon areas (Figure 3a). The X chromosomes had a centromeric addition replicates. Bootstrap resampling (Felsenstein, heterochromatin area. 1985) was applied to assess the support for individual The karyotype of the 2n = 36 showed 17 pairs of bi- nodes using 10,000 bootstrap replicates with 100 random armed chromosomes. The X chromosome was a middle- additions. TBR was also conducted by both a NJ search sized submetacentric and the Y chromosome was a with 10,000 heuristic bootstrap analysis and NJ bootstrap small-sized acrocentric. C-band results showed interstitial analysis in PAUP 4.0b.10 (Swofford, 2001) dibranch blocks in 7 pairs (pairs 3, 6, 10, 11, 13, 14, and 17) and swapping. The karyotype preparations and animals centromeric heterochromatin in 2 pairs (pairs 7 and 12) examined were deposited in the Department of Biology, (Figure 3b). Two chromosomal rearrangements relative to Faculty of Arts and Sciences, Bülent Ecevit University. 2n = 38 were recognized from G-banding patterns. These 472 MATUR et al. / Turk J Zool Table 2. The matrix of first 30 out of 90 chromosomal characters identified in Nannospalax and used for the phylogenetic reconstruction. The karyotype of 2n = 60 Jordanian (Ivanitskaya and Nevo, 1998) has been used as outgroup. These 30 characters are present/absent in chromosomes; others are rearrangement and every identified rearrangement is coded as 1. 2n/Chr.no.