Keywords: Cestoda, cryptic species, DNA barcoding, Felis silvestris, Taenia taeniaeformis, wildcat. Received 21 ..... Taenia crassiceps (7 sequences). Taenia ...
Molecular Ecology Resources (2012)
doi: 10.1111/j.1755-0998.2011.03110.x
Integrative taxonomy at work: DNA barcoding of taeniids harboured by wild and domestic cats A. GALIMBERTI,* D. F. ROMANO,* M. GENCHI,† D. PAOLONI,‡ F. VERCILLO,‡ L. BIZZARRI,‡ D. SASSERA,† C. BANDI,† C. GENCHI,† B. RAGNI‡ and M. CASIRAGHI* *Universita` degli Studi di Milano Bicocca, ZooPlantLab, Dipartimento di Biotecnologie e Bioscienze, Piazza della Scienza 2, 20126 Milano, Italy, †Universita` degli Studi di Milano, Sezione di Patologia Generale e Parassitologia, Dipartimento di Patologia Animale, Igiene e Sanita` Pubblica Veterinaria, Via Celoria 10, 20133 Milano, Italy, ‡Dipartimento di Biologia Cellulare e Ambientale, Universita` degli Studi di Perugia, Via Elce di Sotto, 06123 Perugia, Italy
Abstract In modern taxonomy, DNA barcoding is particularly useful where biometric parameters are difficult to determine or useless owing to the poor quality of samples. These situations are frequent in parasitology. Here, we present an integrated study, based on both DNA barcoding and morphological analysis, on cestodes belonging to the genus Taenia, for which biodiversity is still largely underestimated. In particular, we characterized cestodes from Italian wildcats (Felis silvestris silvestris), free-ranging domestic cats (Felis silvestris catus) and hybrids populations. Adult taeniids were collected by post-mortem examinations of the hosts and morphologically identified as Taenia taeniaeformis. We produced cox1 barcode sequences for all the analysed specimens, and we compared them with reference sequences of individuals belonging to the genus Taenia retrieved from GenBank. In order to evaluate the performance of a DNA barcoding approach to discriminate these parasites, the strength of correlation between species identification based on classical morphology and the molecular divergence of cox1 sequences was measured. Our study provides clear evidence that DNA barcoding is highly efficient to reveal the presence of cryptic lineages within already-described taeniid species. Indeed, we detected three welldefined molecular lineages within the whole panel of specimens morphologically identified as T. taeniaeformis. Two of these molecular groups were already identified by other authors and should be ranked at species level. The third molecular group encompasses only samples collected in Italy during this study, and it represents a third candidate species, still morphologically undescribed. Keywords: Cestoda, cryptic species, DNA barcoding, Felis silvestris, Taenia taeniaeformis, wildcat Received 21 September 2011; revision received 29 November 2011; accepted 8 December 2011
Introduction Traditionally, the taxonomical identification of cestodes and other endoparasites has been based on both morphological and ecological (e.g. host specificity) criteria. However, these approaches could lead to unreliable results, especially when cryptic species complexes are considered (e.g. the case of Taenia krabbei and T. ovis reported by Priemer et al. 2002). In addition, the sampling of some endoparasites is often based on post-mortem examination of the hosts, while less-invasive approaches like the collection of eggs or proglottids in host faeces cannot permit an easier identification owing to the loss of many diagnostic tracts (Zhang et al. 2007). Besides these criticisms, one of the main aspects affecting the taxonomy of cestodes is Correspondence: Maurizio Casiraghi, Fax: +39 02 64483568; E-mail: maurizio.casiraghi@unimib.it
2012 Blackwell Publishing Ltd
the complex influence of their symbiotic relationship with hosts (Bouzid et al. 2008) and their facultative selfsufficient hermaphroditic reproduction. In these cases, reproductive isolation is not necessary to observe strong changes in the genetic structure of populations, and various patterns of sympatric speciation could ultimately occur without the opposing effects of interbreeding. As a consequence, cestode species richness could be underestimated because of the occurrence of several unreported cryptic species potentially co-occurring in the same host at the same time (Fontaneto et al. 2009; Birky et al. 2010). It is thus not surprising that in the last decades, the advent of molecular approaches was considered the future for the identification of poorly known parasite groups, among which are cestodes (Azuma et al. 1995; McManus 2006; Lavikainen et al. 2008). However, despite such highly positive claims, most of the species belonging to the genus Taenia are still genetically uncharacterized.
2 A. GALIMBERTI ET AL. Interesting patterns of intraspecific genetic variability have been observed within several taxa belonging to this genus, supporting the existence of lineages that could be ranked at the species level if further coherent diagnostic characteristics would be provided (Lavikainen et al. 2008; Michelet et al. 2010). Owing to the large amount of variability detected in the genus Taenia, it would be useful to integrate morphological and molecular approaches to investigate the microtaxonomy of these parasites. In particular, a case of veterinary interest deserving to be further investigated concerns the taeniid panel of wildcat populations (Felis silvestris Schreber, 1777). The host taxon is polytypic, being represented by a domesticated subspecies (Felis silvestris catus Linnaeus, 1758) and five recognized wild subspecies, among which Felis silvestris silvestris occurs in Italy and in the rest of Europe (Randi et al. 2001). Concerning the cestode species harboured by F. s. silvestris, only few studies are available (e.g. Krone et al. 2008), while more detailed surveys have been provided for F. s. catus that seems to play an active role as a vector of infectious diseases for indigenous wildcat (Calvete et al. 1998; Delahay et al. 1998; Changizi et al. 2007; Zibaei et al. 2007; Arbabi & Hooshyar 2009; Schuster et al. 2009). In addition, the endoparasite panel of Mediterranean F. silvestris ssp. populations has been poorly investigated when compared with the moderate amount of data currently available for the rest of Europe (Calvete et al. 1998; Krone et al. 2008). These studies identified Taenia taeniaeformis as the most abundant cestode species harboured by European wildcat, while it occurs less frequently in domestic cat populations belonging to the same study areas. Strong evidence for the presence of cryptic diversity within T. taeniaeformis was provided both morphologically and molecularly (Azuma et al. 1995; Okamoto et al. 1995; Lavikainen et al. 2008), but any attempt was performed to clearly synthesize these results towards a taxonomic reassessment of the taxon. In this context, DNA barcoding could help in clarifying such a problematic issue. With this approach, the sequence variability of a standardized region of the genome is used to support morphological characterizations, allowing us to assign biological samples to a given species (Hebert et al. 2002). In particular, the chosen marker (e.g. the mitochondrial gene for cox1 in metazoans) has to satisfy a simple condition: its genetic variability among species must be higher than the variability within species. Recently, DNA barcoding has emerged as an efficient tool for the identification of many metazoans at the species level (Hebert et al. 2003; Fre´zal & Leblois 2008), and for the diagnosis of pathogenic infective agents like endoparasites (Ferri et al. 2009). Moreover, DNA barcoding results can be used to predict new cryptic or undescribed taxa, that is, at the basis of the so-called DNA taxonomy (Lefe´bure et al. 2006; Casiraghi et al. 2010).
In this study, we compiled a mitochondrial data set of cox1 barcode sequences for several taeniid species of medical and veterinary interest. In particular, our analyses allowed us to cover the following topics: (i) to verify the efficacy of a DNA barcoding approach in the recognition of taeniid species, testing also for the coherence of a molecular approach with the morphology-based taxonomy; (ii) to characterize taeniid spectrum in wildcat Italian populations; and (iii) to investigate the morphological and molecular variability of T. taeniaeformis.
Materials and methods Collection of biological samples Forty-eight wildcats (Felis silvestris silvestris), two freeranging domestic cats (Felis silvestris catus) and 12 undetermined individuals and hybrids between these two taxa were collected from 29 different localities of Central and Southern Italy (see Table 1 for details). All specimens were found dead because of roadkills. The carcasses were collected and subjected to post-mortem examination. Taxonomical determination was performed by means of coat colour and pattern markings, cranial index, intestinal index and DNA sequencing (Ragni 1981; Ragni & Possenti 1996; Randi et al. 2001). We determined host body conditions by measuring the thickness of subcutaneous, abdominal and kidney fat tissues. Stomach and intestine contents were collected and the parasites were isolated. Sixty-two adult taeniids were recovered by post-mortem examination of the hosts listed in Table 1 and were stored in 96% ethanol. The scolex and the most part of the strobila were isolated for the morphological examination, while 10–15 proglottids were transferred into plastic tubes with 96% ethanol to be used for the DNA barcoding molecular analyses. Following the protocol specified by the Biorepositories initiative (http:// www.biorepositories.org), samples were vouchered with the id string ‘MIB:zpl:’.
Morphological identification procedure Rostellum samples were washed in physiological solution, stored overnight at 4 C and fixed in 70% ethanol. Scolices were stored in lactophenol for 3 h, with sufficient coverslip pressure to cause the hooks to lie flat but not damaged. Measurements of the total length and blade length were made on large and small hooks per rostellum using a Zeiss Axioskop 2 microscope. Measurements were taken for 15 rostellum, four large hooks and four small hooks of each considered sample. The morphological data were statistically analysed using the Wilcoxon test, and data were considered significant when probability values were
