Both eastern and western populations showed a karyotypic structure that corresponds to cytotype A in the Bertollo's classification (Bertollo et al., 2000). However, some chromosomal differentiation was found between the populations, particularly in respect of the Ag-NORs and GC-rich regions locations. Indeed, the Ag-NORs located on the telomeric region of the smaller submetacentric chromosomes and in the interstitial region of the 16th chromosome (Fig. 3b) were only seen in the eastern population. In fact, interstitial NOR is an uncommon feature for H. malabaricus. On the other hand bitelomeric NORs, as those observed in both eastern and western populations, have been commonly found in H. malabaricus cytotypes (Bertollo, 1996; Born and Bertollo, 2001), representing a probable sinapomorphic condition for this fish group. Concerning the distinct numbers of Ag-NORs found in the eastern (2-7) and in the western (3-8) populations, although this variation may be related to a differential expression of some rDNA cistrons, the results may also represent real karyotypic differences, as verified by Born and Bertollo (2001) and Vicari et al. (2003) for distinct H. malabaricus populations belonging to cytotype A.
Two classes of heterochromatin were observed in both populations. The first one possesses a GC-rich DNA, which is present only in few chromosomes. This heterochromatin was not found to be related to NORs in the western population, which might represent an exception feature amongst fishes (Souza et al., 1996; Artoni et al., 1999; Margarido and Galetti Jr., 2000). Indeed, GC-rich heterochromatin associated with NORs is a common character in fish species (Mayr et al., 1985; Schmid and Guttenbach, 1988; Phillips and Hartley, 1988; Sola et al., 1992). The location of this heterochromatin in distinct chromosome pairs, besides its relation with NOR only in the eastern population, are good evidences that a karyotypic diversification have already occurred at some degree between the populations. A second heterochromatin class, present on the centromeric region of most chromosomes, does not present any fluorescent signal after chomomycin A3 staining.
Lemos et al. (2002) found sympatric populations of H. malabaricus belonging to cytotype A (2n=42 chromosomes without a sex chromosome system) and cytotype B (2n=42 chromosomes with a XX/XY sex chromosome system) in the Iguaçu river, the latter being also found in coastal Brazilian basins (Bertollo et al., 2000). The occurrence of the cytotype B in the Iguaçu basin could be explained by dispersion from the coastal basins, as a consequence of a preterit contact between the basins, a headwater capture in small streams or even an occasional communication during the wet seasons (Lemos et al., 2002). Whereas cytotype A show a wide distribution along the Iguaçu basin (Bertollo et al., 2000; Lemos et al., 2002; present study), cytotype B was found thus far only in a restricted area of the first plateau of that basin (Fig. 1d). Thus, it is likely that the dispersal event of cytotype B have a recent origin, in addition to the low vagility of H. malabaricus and the possible geological barrier formed by the Ponta Grossa arc, that is, the Devonian scarp that delimits the first and the second plateaus of the Paraná State (Fig. 1b).
The Iguaçu river basin is not recent in origin. As other tributaries of the Paraná river, as well as the Paraná river itself, its origin dates from the basaltic settlement during the formation of Serra Geral, in the Cretaceous (Potter, 1997). Its fish fauna finds itself isolated from that of the Paraná river due to the formation of the Iguaçu falls, approximately 22 millions years ago (Oligo-Miocene period). This fact may have favored the speciation and the considerable degree of endemism of that basin (Severi and Cordeiro 1994). H. malabaricus has been considered a non-native species in the Iguaçu basin. However, cytogenetics data show that cytotype A finds itself broadly distributed in Brazilian southeast and southern regions (where the Iguaçu basin is inserted), reaching Uruguay and Argentina (Bertollo et al., 2000). A priori, it seems very unlikely that H. malabaricus was originally absent from the Iguaçu basin, although present in many other neighboring basins. Thus, the endemism of the Iguaçu river appears to be valid for many fish species, but does not seem to apply for others, as H. malabaricus (Lemos et al., 2002). Our hypothesis is that H. malabaricus cytotype A is a native form in the Iguaçu river, and that the karyotypic diversification of the populations here analyzed is a throughout time consequence of vicariant events.
This study was supported by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico), Fundação Araucária (Fundação de Apoio ao Desenvolvimento Científico e Tecnológico do Estado do Paraná), Universidade Estadual do Oeste do Paraná and Universidade Estadual de Ponta Grossa. The authors also thank Dra. Onildes Maria Taschetto and Dr. Horácio Ferreira Julio Jr. for their contributions, and Miguel Airton Carvalho for field and laboratory assistance.