Cryptic speciation among intestinal parasites (Trematoda: Digenea) infecting sympatric host ®shes (Sparidae) O. JOUSSON,* P. BARTOLI & J. PAWLOWSKI* *DeÂpartement de Zoologie et Biologie Animale, Universite de GeneÁve, CheÃne-Bougeries, Switzerland Centre d'OceÂanologie de Marseille, UMR CNRS 6540, Marseille, France
Keywords:
Abstract
cryptic diversity; Diplodus; hosts feeding preferences; Macvicaria; mitochondrial and ribosomal DNA; Monorchis; speciation.
In the north-western (NW) Mediterranean, the teleosts Diplodus sargus, D. vulgaris and D. annularis coexist in infralittoral habitats. These ®shes are infected by two species of the Digenea (Platyhelminthes, Trematoda): Macvicaria crassigula (Opecoelidae) and Monorchis parvus (Monorchiidae) for which we obtained Internal Transcribed Spacer rDNA sequences. Each parasite species represents a complex of two cryptic species, one restricted to D. annularis, and the other shared by D. sargus and D. vulgaris. Cytochrome b mtDNA sequences were used to infer host phylogenetic relationships which showed that the distribution of parasites in Diplodus hosts is not a consequence of coevolutionary interactions. We used diet analyses available for the ®sh hosts to assess the degree of overlap in the use of food among the three species. The feeding overlap was signi®cant only between D. sargus and D. vulgaris, but not for the other ®sh pairs. The possible mechanisms involved in the speciation of the digenean fauna of Diplodus ®shes are discussed.
Introduction Digeneans are parasitic ¯atworms (Platyhelminthes: Trematoda) characterized by complex life-cycles involving two or three hosts and several developmental stages, and by the alternation of sexual and asexual generations (Pearson, 1972). Parasites are ®rst and foremost speci®c to site; host ranges are far more subject to change than is microhabitat (Adamson & Caira, 1994; Combes, 1995). The degree of host-speci®city or host range can be de®ned by the number of host species a parasite is able to infect (Combes, 1995). Factors associated with host ecology may be the main ones generating speci®city in passively transmitted intestinal parasites (Adamson & Caira, 1994; Tinsley & Jackson, 1998). Speci®city to the hosts involved in the digenean life-cycles may differ, presumably depending on their functional role. The freeswimming miracidial stage that infects the molluscan ®rst host usually shows a strict speci®city (Gibson & Bray, 1994; Nunez & De Jong-Brink, 1997). The degree of Correspondence: Olivier Jousson, Station de Zoologie, DeÂpartement de Zoologie et Biologie Animale, Universite de GeneÁve, 154 route de Malagnou, CH-1224 CheÃne-Bougeries, Switzerland. Tel.: +41 22 349 8644; fax: +41 22 349 2647; e-mail:
[email protected]
778
speci®city of the metacercarian stage parasite of the second host is poorly known, but it is generally agreed that developmental stages that use intermediate hosts as trophic channels may exhibit reduced speci®city in these hosts to increase their chances of being channelled to the ®nal host. During de®nitive host infestation speci®city may be determined passively, by host ecological factors such as feeding habits (Adamson & Caira, 1994). In many groups, the taxonomic identi®cation of closely related species using morphological data as the unique basis may be problematical as a great time-lag must exist between the primary genetic speciation and morphological differentiation. Using molecular data, cryptic diversity has been detected in many organisms (e.g. Geiser 1 et al., 1998; De Vargas et al., 1999; Gleeson et al., 1999). Thus, given the broad de®nitive host range de®ned for most adult digenean species (Gibson & Bray, 1994), species diversity and host speci®city may have been underestimated among digenean parasites occurring in marine ®shes. Recently, several molecular techniques have been developed for taxonomic identi®cation of parasites (MacManus & Bowles, 1996). Sequences of internal transcribed spacer (ITS) of the nuclear ribosomal DNA (rDNA) region have been used for species distinction (Luton et al., 1992; Adlard et al., 1993; DespreÁs et al.,
J. EVOL. BIOL. 13 (2000) 778±785 ã 2000 BLACKWELL SCIENCE LTD
Cryptic speciation of parasites of sympatric ®shes
1995; Jousson et al., 1998a) and developmental stages identi®cation (Cribb et al., 1998; Jousson et al., 1998b, 1999) in many digenean genera. In this study, we investigated the digenean fauna of three Mediterranean sparid species common in shallow infralittoral habitats: Diplodus vulgaris (E.G. Saint-Hilaire 1817), D. sargus (L. 1758), and D. annularis (L. 1758). We focused our search on two digenean species that have been recorded from these three Diplodus species: Macvicaria crassigula (Linton 1910) (Opecoelidae) (in Bartoli et al., 1989) and Monorchis parvus Looss, 1902 (Monorchiidae) (in Looss, 1902). We obtained ITS1 ribosomal DNA sequences of several M. crassigula and M. parvus specimens isolated from the three ®sh species. The occurrence of cryptic parasite species infecting Diplodus ®shes may be explained by coevolutionary interactions if hosts and parasites show identical phylogenetic relationships, as demonstrated for other host±parasite associations (Hafner & Nadler, 1988). To test if the evolution of Diplodus parasites is driven by coevolutionary interactions, we obtained cytochrome b mitochondrial DNA sequences from the ®ve Diplodus species occurring in the Mediterranean: D. sargus, D. vulgaris, D. annularis, D. cervinus and D. puntazzo. Finally, to determine if the distribution of parasites in Diplodus ®shes could be related to host food preferences, we used the stomach content analyses available for D. sargus, D. vulgaris (Sala & Ballesteros, 1997), and for D. annularis (Rosecchi, 1985). We assessed the degree of overlap in use of food between the three species using the Schoener's index.
and D. annularis. Four M. crassigula and two M. parvus specimens were isolated from each Diplodus host species. These parasites were isolated from three individuals of D. sargus, four individuals of D. vulgaris and four individuals of D. annularis. To be used as outgroup taxa, we collected related Macvicaria and Monorchis species occurring in different ®sh species from the same geographical area: M. mormyri (from Lithognathus mormyrus), M. maillardi (from Sparus aurata), M. monorchis (from Spondyliosoma cantharus) and an undescribed Monorchis species (from Parablennius gattorugine) (Table 1).
Hosts
We obtained muscle tissues from two individuals of each of the ®ve Diplodus species occurring in the Mediterranean: D. sargus, D. vulgaris, D. annularis, D. puntazzo and D. cervinus. To be used as outgroup taxa, we obtained muscle tissues from two other Mediterranean sparids, Lithognathus mormyrus and Sparus aurata (Table 1). All parasites and ®shes were collected from the northwestern (NW) Mediterranean (Scandola Natural Reserve, Corsica, France), except for D. cervinus that was collected from the Lebanon coast. Molecular data DNA of hosts (Diplodus ®shes) and parasites (Macvicaria and Monorchis species) was extracted in guanidine lysis buffer (Maniatis, 1982), precipitated with isopropanol and dissolved in distilled water.
Parasites
Material and methods Sample collection
Parasites
Living adult specimens of M. crassigula (Opecoelidae) and M. parvus (Monorchiidae) were isolated from the digestive tract of their de®nitive hosts: D. sargus, D. vulgaris, Table 1 Host and parasite sequences analysed in this study.
779
The ITS1 region of the ribosomal DNA was ampli®ed using universal primers located about 360 bp from the 3¢ end of the 18S rDNA (s18: 5¢-TAACAGGTCTGTGATGCC3¢) and at the beginning of the 5.8S gene (5.8s1: 5¢-TGTCGATGAAGAGCGCAGC-3¢). Polymerase chain 2 reaction (PCR) ampli®cations were performed in a total volume of 50 lL with an ampli®cation pro®le consisting
Species
Host
Region sequenced
Accession number
Macvicaria crassigula M. crassigula M. maillardi M. mormyri Monorchis parvus M. parvus M. monorchis Monorchis sp. Diplodus sargus D. vulgaris D. annularis D. cervinus D. puntazzo Lithognathus mormyrus Sparus aurata
D. sargus/D. vulgaris D. annularis S. aurata L. mormyrus D. sargus/D. vulgaris D. annularis Spondyliosoma cantharus Parablennius gattorugine ± ± ± ± ± ± ±
ITS rDNA ITS rDNA ITS rDNA ITS rDNA ITS rDNA ITS rDNA ITS rDNA ITS rDNA Cytochrome Cytochrome Cytochrome Cytochrome Cytochrome Cytochrome Cytochrome
AJ 241803 AJ 277372 AJ 277373 AJ 241802 AJ 277374 Y 18935 Y 18939 AJ 277375 AJ 277369 AJ 277370 AJ 277366 AJ 277367 AJ 277368 AJ 277371 AF 165085
J. EVOL. BIOL. 13 (2000) 778±785 ã 2000 BLACKWELL SCIENCE LTD
b b b b b b b
780
O. JOUSSON ET AL.
of 40 cycles of 30 s at 94 °C, 30 s at 50 °C and 120 s at 72 °C, followed by 5 min at 72 °C for ®nal extension.
Hosts
The cytochrome b gene of mitochondrial DNA was ampli®ed using primers designed in adjacent regions coding for the transfer tRNA (tRNA) glutamate (cbtd2: 5¢AATGAYWTGAAAAACCACCGTTG-3¢) and tRNA tryptophan (cbtr2: 5¢-CGGMTTACAAGRCCGRYGCT-3¢). The complete mitochondrial genome sequences available for Gadus morhua (EMBL Accession Number NC002081), Salmo salar (NC001960) and Cyprinus carpio (NC001606) were used for designing primers. PCR ampli®cations were performed in a total volume of 50 lL with an ampli®cation pro®le consisting of 40 cycles of 30 s at 94 °C, 30 s at 54 °C and 120 s at 72 °C, followed by 5 min at 72 °C for ®nal extension. Ampli®ed PCR products were puri®ed using the high pure PCR puri®cation kit (Roche Diagnostics, Rotkreuz, 3 Switzerland), and sequenced directly with an ABI-377 automated DNA sequencer (Perkin-Elmer, Rotkreuz, 4 Switzerland). For DNA sequencing, the ampli®cation primers were used. The sequences were aligned manu5 ally using the GDE 2.2 (Larsen et al., 1993) and analysed using the following methods: the neighbour-joining (NJ) method (Saitou & Nei, 1987) using Kimura two-parameters distance matrix, the maximum parsimony (MP) method using the heuristic search option included in PAUP* (Swofford, 1998) and the maximum likelihood (ML) method using the fast DNAm1 program (Olsen et al., 1994). The reliability of internal branches was assessed using the bootstrap method (Felsenstein, 1988). The phylo-win program (Galtier & Gouy, 1996) was used for distance computations, NJ and ML tree building and bootstraping. DNA sequences analysed in this paper have been deposited in the European Molecular Biology 6 Laboratory (EMBL)/GenBank database (see Table 1). Stomach contents We used the stomach content data available for the three Diplodus species (Rosecchi, 1985; Sala & Ballesteros, 1997) (Table 2). Several coef®cients were used to determine the importance of prey in the diets. The occurrence index (F) is the percentage of nonempty stomachs that contained a particular prey item. Percentages by weight of prey (W) were calculated as the ratio of the weight of a prey category to the total weight of the stomach content. In order not to overemphasize the importance of occasional large prey, a ranking index (K) was computed by multiplying the occurrence index by the weight of a prey (K F ´ W/100). We determined the Schoener's index (T) (Schoener, 1974) for each ®sh pair, which represents the degree of overlap in use of food. The index varies from 0, when the two species use totally different resources, to 1, when they use the same prey categories in the same proportions. An overlap equal to or above 0.6 was considered signi®cant, following Keast (1978).
Results 7 Molecular data
Parasites
Given the high degree of interspeci®c variations of the ITS1 ribosomal DNA, the sequences of Macvicaria and Monorchis parasites were analysed separately. The analysed data set of Macvicaria specimens comprises 866 unambiguously aligned sites, of which 48 are variable (5.5%) and 44 are informative (5.1%). For Monorchis specimens, the data set comprises 634 sites, of which 63 are variable (10.3%) and 34 are informative (5.4%). Figure 1 shows the NJ trees of M. crassigula and M. parvus specimens isolated from D. sargus, D. vulgaris and D. annularis, inferred with 500 bootstrap replicates. Both M. crassigula and M. parvus specimens appear to be monophyletic in relation to the outgroup taxa used. The trees clearly show the presence of two well-de®ned clades (98±100% bootstrap support) within M. crassigula and M. parvus. One clade includes the specimens isolated from D. sargus and D. vulgaris (whose sequences mix together), and the other one includes the specimens isolated from D. annularis. Such tree topology was con®rmed by using ML and MP analyses (data not shown). The divergence between the two clades reaches 2.6% for M. crassigula, and 3.8% for M. parvus. The divergence within each clade does not exceed 0.7%.
Hosts
The analysis of the cytochrome b gene of mitochondrial DNA included ®rst, second and third codon positions. The analysed data set comprises 1094 sites, of which 443 are variable (40.5%) and 345 are informative (31.5%). Figure 2 shows the ML tree of Mediterranean Diplodus species, inferred with 100 bootstrap replicates. Diplodus appear to be monophyletic in relation to the outgroup taxa used (L. mormyrus and S. aurata). This monophyly was con®rmed using other Mediterranean sparid sequences as an outgroup taxa: Pagrus pagrus (EMBL Accession Number: AF143196), Boops boops (X81567) and Dentex dentex (AF143197) (data not shown). Within the Diplodus clade, D. vulgaris branches together with D. annularis, with strong bootstrap support (93%), whereas D. sargus is related to D. cervinus. D. puntazzo shows an intermediate position between D. vulgaris± D. annularis and D. sargus±D. cervinus clades. This tree topology was con®rmed using MP and NJ analyses (data not shown). Stomach contents Figure 3 shows the ranking indexes of each prey category, expressed as a percentage of the total value of the ranking indexes, for each ®sh species. The Schoener's index was signi®cant between D. sargus and D. vulgaris (T 0.69), but not between D. sargus and
J. EVOL. BIOL. 13 (2000) 778±785 ã 2000 BLACKWELL SCIENCE LTD
Cryptic speciation of parasites of sympatric ®shes
Table 2 Stomach content of Diplodus annularis, D. vulgaris and D. sargus in the NW Mediterranean.
781
Species
Amphipods Isopods Decapods Cumaceans Tanaids Copepods Pycnogonids Barnacles Cnidarians Bryozoans Holothurians Polychaetes Ophiures Echinids Asteroids Ascidians Bivalves Cephalopods Gastropods Polyplacophores Opisthobranchs Sipunculids Fishes Seagrasses Algae
D. annularis D. vulgaris Number of full stomachs examined
D. sargus
512 Length range (mm)
44
69
9±250 Source
100±240
70±400
Rosecchi, 1985
Sala & Ballesteros, 1997
F
W
K
F
W
K
F
W
K
38.4 1.2 20.3 ) ) 13.1 ) ) 18.6 ) ) 19.1 ) 5.0 ) ) 10.4 5.2 6.8 ) ) ) 6.4 ) 3.7
0.7 0.1 31.0 ) ) 0.3 ) ) 8.2 ) ) 23.2 ) 2.7 ) ) 9.8 6.6 6.2 ) ) ) 3.4 ) 0.3
0.3 + 6.3 ) ) + ) ) 1.5 ) ) 4.4 ) 0.1 ) ) 1.0 0.3 0.4 ) ) ) 0.2 ) +
59.1 36.4 13.6 ) 2.3 11.4 4.5 ) 36.4 4.5 ) 38.6 36.4 34.1 2.3 2.3 40.9 ) 6.8 9.1 2.3 2.3 ) 5.0 56.8
3.6 0.1 1.8 ) + 0.2 0.1 ) 0.9 0.1 ) 8.2 14.1 1.7 0.2 + 51.8 ) 0.2 0.7 + 0.4 ) 0.1 3.8
2.1 + 0.2 ) + + + ) 0.3 + ) 3.2 5.1 0.6 + + 21.2 ) + 0.1 + + ) + 2.2
36.2 2.9 17.4 1.4 1.4 1.4 1.4 27.5 14.5 4.3 1.4 18.8 2.9 13.0 ) 2.9 37.7 ) 4.3 1.4 ) ) ) 8.7 43.5
0.6 + 1.3 + + + + 5.0 0.1 + + 0.9 + 19.4 ) 0.5 57.5 ) 0.5 0.2 ) ) ) 0.9 10.4
0.2 + 0.2 + + + + 1.4 + + + 0.2 + 2.5 ) + 21.7 ) + + ) ) ) 0.1 4.5
F = % Occurrence frequency of prey; W = % prey weight; K = ranking index; + =
