Introduction

Parasitism, a specific form of symbiosis, is one of the most successful modes of life (Palm and Klimpel 2007). More than half of all plant and animal species on earth are parasites, and there is probably no organism without any parasitic infection during its lifetime (Palm and Klimpel 2007). One of the most widespread groups of parasites are the nematodes, using plants as well as animals and humans as their hosts on global scale (e.g. Blaxter et al. 1998). With approximately 256 families and more than 40,000 species, the phylum Nematoda represents one of the most species-rich and abundant invertebrate groups. They include not only parasitic but also free living forms in various limnic, marine as well as terrestrial habitats (McClelland 2005).

Traditionally, the Nematoda are divided into two classes, the predominantly terrestrial Secernentea and the predominantly marine Adenophorea, but whether this classification reflects the evolutionary history of the group remains highly questionable. Recent phylogenetic studies based on molecular markers divide nematodes into the two classes Enoplea (Adenophorea, Aphasmidea) and Rhabditea (Phasmidea, Secernentea; McClelland 2005). Marine fish parasitic species are found in the enoplean subclass Dorylaimia (e.g. the genera Capillaria, Pseudocapillaria) and in the rhabditean subclass Rhabditia (e.g. the genera Anisakis, Hysterothylacium, Pseudoterranova; McClelland 2005).

According to Anderson’s (1984, 1996) hypothesis, parasitic nematodes first evolved in terrestrial hosts and were only able to invade aquatic environments after the development of heteroxeny (use of intermediate hosts) and paratenesis (use of transport hosts). Heteroxenic life cycles have been frequently described in marine ascaridoid nematodes such as species of the genera Anisakis, Pseudoterranova (both subfamily Anisakinae) and Hysterothylacium (subfamily Raphidascaridinae; e.g. Klimpel 2005, Klimpel et al. 2004, Klimpel and Rückert 2005). Ascaridoid nematodes are generally associated with aquatic vertebrates as typically definitive hosts. Transmission pathways are habitat-dependent and usually involve a broad spectrum of invertebrates and intermediate or paratenic fish hosts.

The genus Hysterothylacium comprises more than 59 species and is one of the most ubiquitous parasitic nematode species in fishes of the North Atlantic (e.g. Køie 1993; Balbuena et al. 1998, 2000, Navone et al. 1998; Klimpel et al. 2006a). It can be assumed that life cycles within the genus Hysterothylacium are principally similar for all species. While sexually mature adults are mainly found in the digestive tract of various marine fish species, their third and fourth larval stages preferably parasitise in organs of the body cavity of their hosts. Over 100 different benthic and planktonic invertebrate species act as intermediate and/or paratenic hosts (Køie 1993; Navone et al. 1998).

In the North Atlantic, the North Sea, the Baltic Sea, the Mediterranean Sea and adjacent temperate and cold waters, the species Hysterothylacium aduncum is a very common fish parasite (Margolis and Arthur 1979; Palm et al. 1999; Klimpel et al. 2001), but its taxonomic status is highly ambiguous. Several authors assume H. aduncum to be a single species parasitising opportunistically in various marine fish species. In contrast, Hartwich (1975) recognised three distinctive species: H. aduncum mainly found in clupeid hosts like sprat (Sprattus sprattus) and herring (Clupea harengus), H. gadi from gadoid fishes and H. auctum frequently identified in the eelpout (Zoarces viviparus), but also in various flatfishes (Pleuronectiformes). Petter and Cabaret (1995) found variabilities between specimens collected from different fish species, but proposed only two subspecies on the basis of biometrical data. Thus, it would not be surprising if future molecular studies revealed H. aduncum as a complex of a yet unknown number of sibling species (Balbuena et al. 1998).

In the present study, we investigated the genetic variability of H. aduncum within a single host species (S. sprattus) from a phylogeographic perspective to test overall genetic diversity and the influence of geographic distance as well as known biogeographic boundaries like the Gibraltar gateway on reproductive exchange and gene flow of a abundant parasitic fish nematode.

Materials and methods

Sample collection

A total of 160 European sprat (S. sprattus), 40 samples each, were collected in 2005 at four sites within the northeastern Atlantic, the North Sea (N), the English Channel (EC), the Bay of Biscay (B) and the northern Adriatic Sea (M; Fig. 1). Immediately after catch with pelagic nets at trawling speeds of roughly four knots, fishes were deep frozen at −20°C. Before examination, fish specimens were defrosted at 0–1°C, and the standard length (SL) and total weight (TW) were recorded to the nearest 0.1 cm and 0.1 g, respectively.

Fig. 1
figure 1

Area of investigation with the four sampling stations including the North Sea (N), English Channel (EC), Bay of Biscay (B) and Adriatic Sea (M)

Full size image

Raphidascarid nematode isolation and molecular analyses

Nematodes were isolated and identified to a species level based on morphological and biometric characters by the use of existing keys and descriptions before conservation in 96% ethanol. Genomic DNA was isolated and purified from single individuals (Table 1) using genomic DNA extraction kit (Peqlab Biotechnology GmbH, Erlangen, Germany) following manufacturer’s instructions. The rDNA region comprising the ITS-1, 5.8S, ITS-2 and flanking sequences (=ITS+) was amplified using the previously described primers NC5 (5′-GTA GGT GAA CCT GCG GAA GGA TCA TT-3′) and NC2 (5′-TTA GTT TCT TTT CCT CCG CT-3′; Shih 2004; Zhu et al. 1998, 2000). The polymerase chain reaction (PCR) mix (26 μl) included 13 μl Master-Mix (QIAGEN, Hilden, Germany) containing dNTP, MgCl2, buffer and Taq-polymerase, 3 μl of each primer, 2 μl dest. water and 5 μl genomic DNA. PCR reactions were performed in a thermocycler (Biometra, Germany) under the following conditions: after an initial denaturation at 95°C for 15 min, 30 cycles of 94°C for 1 min (denaturation), 55°C for 1 min (annealing), 72°C for 1 min (extension) were performed, followed by a final extension at 72°C for 5 min. Each PCR run included a sample without DNA (negative control). PCR products were quality-checked on 1% agarose gels. The amplification products were purified with commercial PCR purification kits (E.Z.N.A. cycle-pure kit, Peqlab Biotecnology, Erlangen, Germany) before direct sequencing performed by Seqlab (Goettingen, Germany). Both strands of the two spacer regions and the 5.8S rDNA gene were sequenced for all specimens using primers NC5, NC13 (forward; 5′-ATC GAT GAA GAA CGC AGC-3′), NC13R (reverse; 5′-GCT GCG TTC TTC ATC GAT-3′), XZ1R (reverse; 5′-GGA ATG AAC CCG ATG GCG CAA T-3′) and NC2.

Table 1 Samples of H. aduncum isolated from the sprat (S. sprattus) used in this study
Full size table

Resulting sequence data of raphidascarids were aligned using CLUSTAL W (1.83) multiple sequence alignments (Thompson et al. 1994) and compared with previously published sequence data in GenBank to analyse intraspecific differences.

Results

Morphometric measurements (standard length, total weight) did not reveal any significant differences between sprat (S. sprattus) populations from the four sampling sites (Table 2). The predominant parasite species in all analysed S. sprattus was the raphidascarid nematode H. aduncum, with high values of prevalence (P%), abundance (A), intensity (I), as well as mean intensity (mI) of infestation (Table 2).

Table 2 Morphometric fish data and prevalence, abundance, intensity and mean intensity of infestation of the sprat (S. sprattus) from the four sampling stations
Full size table

Multiple sequence analyses of H. aduncum indicated low genetic diversity only for the ITS-2 region, while ITS-1 and 5.8S rDNA were identical for all investigated 40 specimens. The total lengths of the PCR products varied between 886 and 890 bp, consisting of ITS-1 with a constant length of 401 bp, ITS-2 ranging from 328 to 332 bp and the 5.8S sequence with an again constant length of 157 bp. G+C contents of the three regions varied between 50.3 to 51.3% (Table 1). Sequence divergence within ITS-2 follows a population-specific pattern with no intra- but low interpopulation differentiation based on indels separating an English Channel/Bay of Biscay group from a North Sea/Mediterranean group (Fig. 2, Table 1). The only accessible reference sequence published by Zhu et al. (1998) further differs by a single polymorphic site in ITS-1 (alignment position 61) and an additional indel not shared by any of our the here investigated specimens in ITS-2. The alignment of the ITS-1, 5.7S rDNA and ITS-2 consensus sequence representing H. aduncum from one specimen each of the different geographical areas is given in Fig. 2.

Fig. 2
figure 2

Alignment of the ITS-2 consensus sequence for third/fourth stage larva of H. aduncum (sample code H2, H11, H31, H35) isolated from sprat (S. sprattus) from the North Sea (N), English Channel (EC), Bay of Biscay (B), Adriatic Sea (M) and reference (Ref) of Zhu et al. 1998. The numbers refer to the alignment position, and asterisks indicate positions with same nucleotides

Full size image

Discussion

The present study is the first molecular characterization of raphidascarid nematodes isolated from European sprat (S. sprattus) of different biogeographic regions. The analysed nematodes could be identified morphologically as well as genetically as H. aduncum, and no cryptic sibling species could be detected.

Species of the genus Hysterothylacium are found in the adult stage predominantly in the digestive tract of teleost fishes and rarely in elasmobranches (Køie 1993; Klimpel et al. 2003). Larvae are also widely distributed and have been reported in the organs and body cavities of a variety of marine fishes and pelagic and benthic invertebrates (e.g. Palm et al. 1999; Klimpel and Rückert 2005). H. aduncum is a parasite that is not very host-specific in either its adult or its larval stages. The adults are found in the digestive tract of marine fishes; several marine invertebrates act as intermediate hosts. In general, H. aduncum is one of the most ubiquitous helminth species in the North Atlantic that was also detected in the Pacific Ocean (Shih 2004). In the North Atlantic and the Baltic Sea, H. aduncum is the only recognised species, but its taxonomy is still unresolved. The two other important ascaridoid nematodes in the North Atlantic, A. simplex and P. decipiens, use mammals as definitive hosts, while H. aduncum uses only fishes as final hosts (Balbuena et al. 1998; Klimpel and Rückert 2005). Eggs of the nematode H. aduncum pass out in the faeces of the fish final host (see Fig. 3) and develop to the second stage larvae (L2). Larvae are ingested by various benthic and/or pelagic invertebrates, hatch in their intestine and attain into their haemocoel (Køie 1993; Klimpel et al. 2006b). Development to the third stage larvae (L3) takes place in invertebrates, and the nematode may then transfer to the definitive fish host. Larval size seems important in determining their fate. Only, third stage larvae (L3) longer than 0.3 cm moult into the fourth stage larvae (L4) and subsequently into adult stages in the digestive tract lumen of the fish host (Køie 1993). Different fish species heavily acquire and accumulate different stages of H. aduncum by ingesting crustaceans, chaetognaths and small fishes that are infested and serve as carriers (Klimpel 2005; Klimpel and Rückert 2005).

Fig. 3
figure 3

Schematic illustration of the main life cycle of H. aduncum isolated from the sprat (S. sprattus) in waters of the investigated areas: the North Sea, English Channel and Adriatic Sea (I–III), and the Bay of Biscay (IV). The nematode utilizes similar pathways to reach the final benthic or pelagic fish hosts. In the areas I to III, the nematode use generally benthic and pelagic intermediate hosts, while in the area IV, pelagic intermediate host plays a more important role in the life cycle pattern

Full size image

Morphological differentiation of nematode species belonging to the genus Hysterothylacium is extremely difficult. One of the most important criteria for nematode species identification is the position and anatomy of the foregut, which often is hardly visible and therefore not useful for an unambiguous identification. Molecular genetic techniques are sometimes more reliable for a proper species identification of larval and adult ascaridoid nematodes, such as of the genera Anisakis, Contracaecum and Pseudoterranova. Different authors could demonstrate that morphospecies such as Anisakis simplex and Pseudoterranova decipiens are actually complexes of sibling species (e.g. Abollo et al. 2003) that are morphologically extremely similar but genetically distinct. At least three sibling species could be identified within the A. simplex complex in the Atlantic Ocean and three siblings of the P. decipiens complex in the North Atlantic, Norwegian and the Barents Sea (e.g. Nascetti et al. 1986; Zhu et al. 2002). Each of these cryptic sibling species was found to have much more specific host preferences and a geographically much more confided distribution than known from the original morphospecies (e.g. Marques et al. 2006). These studies show that the region covering ITS-1, 5.8S rDNA and ITS-2 is subject to moderate to fast evolutionary changes and can be successfully used for an unambiguous distinction of closely related nematode species (e.g. D’Amelio et al. 2000).

The fact that H. aduncum is also a nematode parasite with remarkably low host specificity and that siblings have been detected in congeners like Hysterothylacium fabri (Martin-Sánchez et al. 2005) led to the assumption that H. aduncum might represent a sibling species complex as well (Balbuena et al. 1998). Nevertheless, intraspecific genetic diversity in our investigation was unexpectedly low with three indels only in ITS-2, while ITS-1 and 5.8S rDNA from specimens of regions as widely separated as the North Sea and the northern part of the Mediterranean Adriatic Sea did not differ at all. Despite that, the low levels of sequence divergence within ITS-2 follow a population-specific pattern. With no intra- but low interpopulation differentiation, an English Channel/Bay of Biscay group can be separated from a North Sea/Mediterranean Sea group. This pattern can hardly be explained from a phylogeographic point of view and has to be confirmed by higher sample sizes. But it gives a hint towards a possible loss of reproductive exchange and therefore gene flow, which has to be tested by the use of more polymorphic markers.