Les Barbus
Trine Gritz
The site is secure.
The ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.
Biological invasions and introgressive hybridization are major drivers for the decline of native freshwater fish. However, the magnitude of the problem across a native species range, the mechanisms shaping introgression as well as invader's dispersal and the relative role of biological invasions in the light of multiple environmental stressors are rarely described. Here, we report extensive (N = 665) mtDNA sequence and (N = 692) microsatellite genotypic data of 32 Northern Adriatic sites aimed to unravel the invasion of the European Barbus barbus in Italy and the hybridization and decline of the endemic B. plebejus. We highlight an exceptionally fast breakthrough of B. barbus within the Po River basin, leading to widespread introgressive hybridization with the endemic B. plebejus within few generations. In contrast, adjacent drainage systems are still unaffected from B. barbus invasion. We show that barriers to migration are inefficient to halt the invasion process and that propagule pressure, and not environmental quality, is the major driver responsible for B. barbus success. Both introgressive hybridization and invader's dispersal are facilitated by ongoing fisheries management practices. Therefore, immediate changes in fisheries management (i.e. stocking and translocation measures) and a detailed conservation plan, focussed on remnant purebred B. plebejus populations, are urgently needed.
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Random amplified polymorphic DNA (RAPD) markers were used to estimate the population structure and phylogenetic relationships among the eight species of the genus Barbus that inhabit the Iberian Peninsula. Ten random oligodecamers were used to amplify DNA from 232 fish sampled from 15 populations. A total of 270 markers were detected that revealed low levels of genetic variability. The conclusions of cluster analysis indicate two main branches and three well-differentiated groups: north-eastern, Mediterranean and Atlantic. This clustering mainly reflects the evolutionary history of the genus, which is closely related to the paleogeography of the Iberian Peninsula. The contribution and application of these results to the conservation of the species, to their taxonomic status and to the process of colonization of the Iberian Peninsula by the genus Barbus are discussed.
The phylogenetic relationships and taxonomic identity of Barbus have been the subject of debate for decades. Prior to 1990, their study in Europe was based on morphological data and geographical distribution (Tsigenopoulos et al, 1999). Osteological characters have also been employed in some studies to determine the relationships among the different species (Doadrio, 1990). Nevertheless, due to the considerable morphological diversity of Barbus (mainly related to growth, biogeographical and ecological factors (Encina and Granado, 1988)), the interpretation of these data have sometimes been very difficult.
The development of the random amplified polymorphic DNA technique (RAPD) (Williams et al, 1990; Welsh and McClelland, 1990) has provided a useful tool for research into genetic variability. It consists of the PCR amplification of small, inverted repeats scattered throughout the genome, using a single, short primer of arbitrary sequence. Thus, the genome can be scanned more randomly than with conventional techniques. The ability to examine genomic variation without previous sequence information (Williams et al, 1990), the relatively low cost of the technique, and the requirement of only nanograms of template DNA, are all advantages of RAPD in population and other genetic studies.
There is now increasing evidence that the RAPD technique, which has been used in different fields, can detect nuclear variation in fish (Borowski et al, 1995; Naish et al, 1995; Sultmann et al, 1995; Bielawski and Pumo, 1997; Caccone et al, 1997; Callejas and Ochando, 1998; Mamuris et al, 1999; Allendorf and Seeb, 2000). These studies have shown that RAPD is an extremely sensitive method for detecting DNA variation and for establishing genetic relationships in closely related organisms. In the present work, RAPD markers were used to infer phylogenetic relationships among the Spanish species of genus Barbus.
Genomic DNA was extracted from 2 g of liver according to Benito et al (1993), with minor modifications. Samples were treated with SDS and genomic DNA isolated using phenol:chloroform extraction and isopropanol precipitation. The resulting pellets were washed in 70% ethanol, dried and resuspended in 10 mM Tris (pH 8.0), 1 mM EDTA.
Amplification products were resolved by electrophoresis in 2% agarose gels with TAE buffer (40 mM Tris-Acetate, 1 mM EDTA pH 8.0) containing ethidium bromide. A 100 bp ladder marker was used as a molecular size standard.
The RAPD-PCR method has some limitations, such as sensitivity to reaction conditions, occasionally non-reproducible amplification products, and possible co-migration of amplified fragments. However, these problems can be solved by following a strict protocol with standardised conditions, repeating the amplification reactions twice in order to score clearly reproducible bands, and increasing the resolution of band separation (Hadrys et al, 1992). In the present work, amplification results were routinely repeatable.
In the 232 specimens analysed, a total of 270 scorable bands were observed for the eight species of Barbus, ranging in size from 2000 to 300 bp. Only 17 (6.30%) were monomorphic for all species. Differences among the eight species were detected with all 10 primers. Figure 2 shows the RAPD profiles of the different species with the primer C08. As observed in Table 2, RAPD analysis also revealed intraspecific polymorphism in barbel populations, ranging from 0% in B. meridionalis, where all specimens analysed showed identical patterns, to 29.36% in B. comizo (GUA population). Frequencies of polymorphic bands were also included for species comparisons.
In order to visualise the gathering of populations, PCA was used to examine the variability assessed by RAPD. The first three axes covered a large portion of the total variation (89.5%), with three main groups observed (Figure 4). This PCA topology is congruent with the distribution of the groups as revealed by the dendrogram.
Bands that were monomorphic in one species (and therefore its diagnostic markers) or some species, but absent in the rest, are those mainly responsible for the greater inter- rather than intraspecific polymorphism detected (Callejas and Ochando, 2001). Levels of intraspecific polymorphism ranged between 0 and 29.36% (Table 2). The higher polymorphism of some populations, such as CIJ, GUA and MAN, are due mainly to the presence of a high number of bands in low frequencies (Table 2). In general, levels of polymorphism are low compared with those seen in other fish RAPD studies (Bielawski and Pumo, 1997; Caccone et al, 1997; Gomes et al, 1998; Mamuris et al, 1999).
Table 3 shows the similarity values within- (on the diagonal) and between-populations (above the diagonal). Within-population indices were very high, ranging from 95.36% to 100%, and greater than those found among other species of fish using RAPD analysis (Bardakci and Skibinski, 1994; Foo et al, 1995; Bielawsky and Pumo, 1997). Likewise, the genetic similarity between populations of the same species was smaller than the intrapopulation similarities. This implies that individuals within each population are genetically more similar to each other than to individuals from any other population of the same species. In general, similarity values ranged from 94.06% between populations of B. comizo, to 97.81% between populations of B. sclateri, and were higher than interpopulation similarity values reported for other species of fish (Bielaswky and Pumo, 1997). Thus, although there is much genetic similarity between these populations, some geographical species differentiation appears to exist.
These parameters allow quantification of genetic diversity, and indicate levels of genetic variability in barbel populations to be rather low. Diversity was low within populations and was mainly found between populations of different species. Previous surveys in several species of European barbels, using allozymes, also showed their genetic variability to be low (Berrebi and Cattaneo-Berrebi, 1993; Machordom et al, 1995). Nowadays, although barbel populations are larger in numbers when compared with other cyprinids species, they are in decline due to environmental changes such as drought, freshwater contamination, river dredging, excessive water supply for human, agricultural or industrial purposes and the introduction of exotic species (Garca-Berthou and Moreno-Amich, 2000; Penczak and Kruk, 2000; Vila-Gispert et al, 2000; Doadrio, 2001). Thus, the low genetic variability of Barbus spp. is probably attributable, at least in part, to genetic drift owing to small population sizes arising as a consequence of habitat alteration. However, we cannot forget another important process that must be implicated in such low variability, ie founder effect in the evolutionary history of the genus due to its recent radiation in the Iberian Peninsula (Callejas and Ochando, 2000).
An understanding of the inter- and intraspecific distribution of genetic variation within Barbus is essential for the development of appropriate conservation strategies. At present, conservation programmes are being promoted to protect Barbus species (Economidis, 1998). The results of the present RAPD study reveal that nuclear DNA variation in Barbus Spanish species is low, and moreover this kind of study can provide important information for the management of future conservation strategies.
c80f0f1006