Lessepsian fish migration: genetic bottlenecks and parasitological evidence

Genetic and parasitological aspects of species invasions

Genetic data can be useful for elucidating the history and phenomena involved in species invasions. In recent years, it has increasingly been emphasized that many natural populations that derived from a founder event and/or a bottleneck represent a subset of the genetic diversity of the source population, leading to the conclusion that the genetic diversity of the newly founded populations should be significantly lower than that of the source (Tsutsui et al., 2000; Moum & Arnason, 2001; Sax et al., 2005). Conversely, other types of evidence could also be used to retrace the route taken by invasive species. For instance, parasitological data might provide useful information about the area of origin of invaders, but might also shed light on some further aspects such as the mechanism of invasion (e.g. whether they arrived as adults or in the larval stages) or the prospects of success in the new habitat (e.g. potential infection with native and/or natural parasites) (Diamant, 1989; Poulin & Mouillot, 2003; Galli et al., 2007; Merella et al., 2007; Pais et al., 2007, 2008).

When both genetic and parasitological data are available, and suggest the same invasion scenario, the history of a particular invasion can be inferred with a certain degree of reliability. However, when these data on the same invasive species cannot be reconciled because they suggest a different route or history – assuming that both genetic and parasitological evidence are equally valid – an alternative model should be considered.

In this study a comparison of genetic and parasitological data for a Lessepsian migrant fish species presents the opportunity to suggest a multiple-approach method for a better understanding of the events that led to the invasion of the new range. In fact, the complementarity of such information greatly strengthens likely invasion scenarios suggested by both types of criteria.

Lessepsian fish migrants and genetic diversity

Since the opening of the Suez Canal in 1869, the Mediterranean Sea has been subject to an invasion of Indo-Pacific species from the Red Sea (Por, 1971). As far as fishes are concerned, it has recently been estimated that the so-called Lessepsian migrants include more than 60 species (Ben Rais Lasram et al., 2008). Contrary to the evidence of a decrease in genetic diversity of invader species, analyses carried out to date on Lessepsian fish species reveal no significant reduction in genetic diversity, that is, conspecific populations from the Red Sea–Indo-Pacific region and the Mediterranean basin are genetically similar (Bucciarelli et al., 2002; Hassan et al., 2003; Hassan & Bonhomme, 2005). A likely explanation is that Lessepsian fish species invaded the Mediterranean with a conspicuous number of individuals, either as adults or larval stages, which reduced the loss of genetic diversity.

The case of Fistularia Commersonii

The bluespotted cornetfish, Fistularia commersonii Rüppell (Osteichthyes: Fistulariidae), is an Indo-Pacific species whose presence was recorded for the first time on the Mediterranean coast of the Middle East (Golani, 2000). Later, its geographical distribution rapidly extended to the eastern basin (Corsini et al., 2002), and more recently to the central and western Mediterranean (Pais et al., 2007; Dulčićet al., 2008). Knowledge of the specific dynamics of the Mediterranean invasion by F. commersonii is meagre. Golani et al. (2007), studying the partial sequence of the mitochondrial (mt) DNA control region (D-loop 1, primers in Lee et al., 1995), suggest the occurrence of a genetic bottleneck in the Lessepsian migration of this species. These authors believe that the very low richness and diversity of the haplotypes found in the Mediterranean Sea (52 individuals, 2 haplotypes, haplotype diversity 0.009), compared with those from the natural range (49 individuals, 46 haplotypes, haplotype diversity 0.997), indicate that ‘…the Mediterranean populations of bluespotted cornetfish represent a single invasion event by as few as two females’ (Golani et al., 2007, p. 544). The same authors attribute the rapid expansion of this fish in the Mediterranean Sea to larval dispersal favoured by water circulation.

The literature on the parasites of F. commersonii in its natural range is quite scarce (Pérez-Ponce de León et al., 1998; Rigby et al., 1999; Nahhas et al., 2004), and only recently Merella et al. (2007) and Pais et al. (2007) investigated the parasites of this species collected from different sites in the Mediterranean Sea (off the coasts of Sardinia, Tunisia and Libya). These authors found that the parasite assemblages of F. commersonii in the new habitat were mostly characterized by native generalist species, but also included two specific Indo-Pacific digeneans: the lepocreadiids Allolepidapedon fistulariae (Yamaguti) and Neoallepidapedon awaiiense (Yamaguti). Both of these parasites have been described exclusively in Fistularia species from the Indo-Pacific (Bray, 2005), and both genera are largely restricted to Fistularia species; in fact, the genus Neaollolepidapedon is monospecific, while the only other species of Allolepidapedon (Allolepidapedon pristipomoides Shen) is described in the lutjanid Pristipomoides filamentosus (Valenciennes) from the China Sea/Yellow Sea. Considering the short life span of adult digeneans (Cribb, 2005) and the non-migratory habits of F. commersonii, Merella et al. (2007) and Pais et al. (2007) suggested that the fish were infected in the Mediterranean. However, for parasites with complex life cycles that necessitate two or more host species (i.e. digeneans), if suitable hosts for all stages are not present establishment will not take place. As for the paucity of data on the intermediate hosts of these digeneans, Cribb et al. (2001) reported Muricoidea and Rissooidea (Mollusca) as first intermediate hosts of Lepocreadiidae, and Yamaguti (1970) described the metacercariae of A. fistulariae encysted in the muscle of Labridae and Bleniidae (Teleostei). Considering that both Muricoidea and Rissooidea are well represented in the Mediterranean Sea, as well as Labridae and Bleniidae, Merella et al. (2007) and Pais et al. (2007) argued that it is likely that the major factor for these parasites in establishing a self-sustaining population in this area could be the low population density of the definitive host (i.e. F. commersonii). Finally, considering that adult/juvenile fish acquire these digeneans by predating on second intermediate fish hosts, these authors suggested that F. commersonii colonized the Mediterranean with actively swimming individuals and not passive planktonic larvae, as previously reported by Diamant (1989) for the two Lessepsian rabbitfish species Siganus luridus (Rüppell) and Siganus rivulatus Forsskål.

Genetic versus parasitological data

The inference by Golani et al. (2007) that a few individuals of F. commersonii invaded the Mediterranean Sea, and an expansion followed through larval dispersal, does not fit well with the above-mentioned parasitological findings. In fact, according to this hypothesis it is difficult to explain how specimens caught in the western and central Mediterranean have been infected with Indo-Pacific digeneans that could only be acquired through predation on second intermediate fish hosts. Assuming that the two pioneer females cited by Golani et al. (2007) were already infected, the extremely low density of adult definitive hosts would not allow the parasites to complete their life cycle in the new habitat (Torchin et al., 2002). But even supposing this last unlikely event is possible, it is very difficult to explain how the digenean parasites arrived in the western and central basins only by means of larval dispersal of the definitive host. In fact, the larval stages of F. commersonii cannot harbour these parasites. Analogously, Galli et al. (2007) attributed the absence of monogenean parasites on another Lessepsian migrant (S. luridus) to the dispersal of uninfected larval stages of this fish host in the Mediterranean Sea.

From a genetic point of view, the hypothesis advanced by Golani et al. (2007), that a bottleneck produced ‘by as few as two females’ led to the disappearance of a high proportion of haplotype diversity, is reasonable. However, the same authors did not find the two ‘Mediterranean’ haplotypes in the Red Sea and Indian Ocean populations of F. commersonii. This point could be questioned in a scenario involving only bottlenecks (and/or founder events), since these phenomena lead to a reduction of genetic variability that stochastically affects the rarest haplogroups of a population. Therefore, considering the high genetic diversity of the specimens examined by Golani et al. (2007) from the Red Sea and the Indian and Pacific Oceans, the use of an adequate larger number of mitochondrial molecular markers, with different levels of variability, would be advisable. In this context, the occurrence of five matrilineal lineages in the Mediterranean Sea has already been detected by analysing only 12 individuals of F. commersonii (D. Sanna, Università di Sassari, pers. comm.; see GenBank accession numbers: 12S, GQ901988–GQ901999; 16S, GQ902000–GQ902011; cyt b, GQ902012–GQ9023; D-loop II, GQ902024–GQ902035).

In conclusion, using both molecular and parasitological approaches it is possible to depict a likely invasion scenario that might better explain fish migration dynamics, the consequent genetic variability and the likelihood of a non-indigenous species succeeding in the context of a biological invasion.

Editor: Alistair Crame