Introduction
The genus Rhagoletis (Diptera: Tephritidae) consists of at least 65 known species, most of Nearctic and Neotropic and few of Palearctic origin (Bush, Reference Bush1966; Berlocher & Bush, Reference Berlocher and Bush1982; Smith & Bush, Reference Smith and Bush1997, Reference Smith, Bush, Aluja and Norrbom2000). It currently exhibits a wide geographic distribution from New World to Eurasia (White & Elson-Harris, Reference White and Elson-Harris1992). Most Rhagoletis species are considered as major agricultural pests, despite the fact that they have a narrower host plant range than tropical tephritids, such as those of the Anastepha, Bactrocera, Ceratitis, Zeugodacus and Dacus genera (Boller & Prokopy, Reference Boller and Prokopy1976). Three species of phytophagous Rhagoletis are of utmost importance in Europe: the native European cherry fruit fly, Rhagoletis cerasi (L.), and two invasive species, the American eastern cherry fruit fly, R. cingulata (Loew), and the walnut husk fly, R. completa (Cresson), which have been recently introduced from North America and established, mainly, in central Europe. A fourth species, the north American walnut fly, R. suavis (Loew), was found for the first time in Germany (2013) in a private garden in Kleinmachnow (Brandenburg). To date, its presence has been confirmed in only two areas (Brandenburg and Berlin), and thus it is considered of minor importance for Germany (EPPO, 2016). Earlier reports of the western American cherry fruit fly, R. indifferens (Rhagin), in Europe are now attributed to misidentification of R. cingulata samples. Misidentification was driven mainly by the similarities in larval stages between the two species, even though their adults are morphological distinct (EFSA PHL Panel, 2014).
Larvae of Rhagoletis species feed on fruit mesocarp and cause extensive damage on cherries and nuts. Mature larvae leave the fruit or nut to pupate in soil, where they overwinter. Even though, Rhagoletis fruit flies are primarily univoltine (having one generation per year), some species may have a partial second generation or even exhibit long-life cycles through prolonged pupal dormancy (Moraiti et al., Reference Moraiti, Nakas and Papadopoulos2014; Moraiti & Papadopoulos, Reference Moraiti and Papadopoulos2017). Despite the huge economic losses regarding fruit and nut production (>80%), in case of unmanaged Rhagoletis populations, development of effective and environmentally friendly control methods remains a challenge (Kovanci & Kovanci, Reference Kovanci and Kovanci2006; Daniel & Grunder, Reference Daniel and Grunder2012; Daniel & Baker, Reference Daniel and Baker2013; Daniel et al., Reference Daniel, Mathis and Feichtinger2014; Nježić et al., Reference Nježić, Pećanin and Đurić2017; Verheggen et al., Reference Verheggen, Verhaeghe, Giordanengo, Tassus and Escobar-Gutiérrez2017; Florian et al., Reference Florian, Macavei, Hulujan, Vasian, Totos, Gorgan, Oltean and Florian2018). In addition, the potential hybridization events among native and invasive Rhagoletis species in European habitats could result in unpredictable pest dynamics in the area (Johannesen et al., Reference Johannesen, Keyghobadi, Schuler, Stauffer and Vogt2013). In this context, there is an urgent need of consideration of all the available population and genetic data related to the three Rhagoletis species in Europe and further explore their endosymbiotic association with Wolbachia pipientis in an effort to promote environmentally friendly, species-specific control methods, such as the sterile insect technique (SIT) and the incompatible insect technique (IIT) (Robinson et al., Reference Robinson, Franz and Fisher1999; Zabalou et al., Reference Zabalou, Riegler, Theodorakopoulou, Stauffer, Savakis and Bourtzis2004; Reference Zabalou, Apostolaki, Livadaras, Franz, Robinson, Savakis and Bourtzis2009; Dyck et al., Reference Dyck, Hendrichs and Robinson2005; Apostolaki et al., Reference Apostolaki, Livadaras, Saridaki, Chrysargyris, Savakis and Bourtzis2011; Lanzavecchia et al., Reference Lanzavecchia, Juri, Bonomi, Gomulski, Scannapieco, Segura, Malacrida, Cladera and Gasperi2014; Zacharopoulou et al., Reference Zacharopoulou, Augustinos, Drosopoulou, Tsoumani, Gariou-Papalexiou, Franz, Mathiopoulos, Bourtzis and Mavragani-Tsipidou2017; Nikolouli et al., Reference Nikolouli, Colinet, Renault, Enriquez, Mouton, Gibert, Sassu, Cáceres, Stauffer, Pereira and Bourtzis2018).
In the current review, we focus on population genetics, cytogenetics and Wolbachia symbiosis of R. cerasi, R. cingulata and R. completa, with all of them currently present in the European region. Genetic and population data can provide insight into population structuring, origin of invasions, population expansion patterns and possible hybridization events that can subsequently lead to incipient speciation and unpredictable pest dynamics. On the other hand, data on symbiosis, together with the genetic and population data, can be exploited for the enhancement of SIT and other related techniques such as IIT.
Origin and dispersion
Rhagoletis cerasi is believed to have originated in the Caucasian area of western Asia, and until recently it was considered to be widespread throughout Europe and the temperate regions of the Middle and Near East, and Russia (White & Elson-Harris, Reference White and Elson-Harris1992; fig. 1). In 2016, R. cerasi was detected into Ontario, Canada, and in 2017 in Niagara County in New York State, USA, demonstrating the entry of this pest in North America (Barringer, Reference Barringer2018). Rhagoletis cerasi mainly infests fruits of the genus Prunus spp., such as those of Prunus avium, P. cerasus, P. serotina and P. mahaleb. Wild growing cherries (Prunus spp.) and Lonicera spp (Lonicera xylosteum, L. tatarica) can serve as reservoirs or secondary hosts (White & Elson-Harris, Reference White and Elson-Harris1992). To date, sweet and sour cherry producing areas of Europe are under a serious threat since fruit infestation can reach up to 100%, if no insecticide control is applied, while the tolerance level of the market for damaged fruits is less than 2% (Daniel & Grunder, Reference Daniel and Grunder2012). Moreover, R. cerasi populations express extensive geographic variability in life-history traits (i.e. life span, reproduction) and diapause traits, as a result of local adaptation to the ecological heterogeneity of their European habitats (Vallo et al., Reference Vallo, Remund and Boller1976; Papanastasiou et al., Reference Papanastasiou, Nestel, Diamantidis, Nakas and Papadopoulos2011; Moraiti et al., Reference Moraiti, Nakas, Köppler and Papadopoulos2012a, Reference Moraiti, Nakas and Papadopoulos2014, Reference Moraiti, Nakas and Papadopoulos2017). In addition, R. cerasi can plastically respond to unpredictable environmental variation of their local habitats through diapause bet-hedging strategies ensuring persistence in European orchards (Moraiti et al., Reference Moraiti, Nakas and Papadopoulos2014; Moraiti & Papadopoulos, Reference Moraiti and Papadopoulos2017). Since the biology and ecology of the European cherry fruit fly populations are extensively studied in European habitats compared to those of the other two currently introduced Rhagoletis species (Boller & Bush, Reference Boller and Bush1974; Boller & Prokopy, Reference Boller and Prokopy1976; Papanastasiou et al., Reference Papanastasiou, Nestel, Diamantidis, Nakas and Papadopoulos2011; Daniel & Grunder, Reference Daniel and Grunder2012; Moraiti et al., Reference Moraiti, Nakas, Köppler and Papadopoulos2012a, Reference Moraiti, Nakas and Papadopoulosb, Reference Moraiti, Nakas and Papadopoulos2014, Reference Moraiti, Nakas and Papadopoulos2017; Moraiti & Papadopoulos, Reference Moraiti and Papadopoulos2017), R. cerasi could serve as a model species for predicting the adaptation patterns of the two new residents in European orchards.
Geographic distribution of R. cerasi, R. cingulata and R. completa in Europe. Marks show indicative areas where the species occur.
Rhagoletis cingulata is another cherry-infesting Rhagoletis species that has been established over the last few decades in several central European countries. This species is endemic in North America, distributed from Southeastern Canada to eastern USA (Florida and Texas) (Bush, Reference Bush1966; Smith & Bush, Reference Smith and Bush1997; Rull et al., Reference Rull, Aluja and Feder2011; Yee et al., Reference Yee, Hernández-Ortiz, Rull, Sinclair and Neven2014) and Mexico (Rull et al., Reference Rull, Aluja and Feder2011; Yee et al., Reference Yee, Hernández-Ortiz, Rull, Sinclair and Neven2014). It was reported for the first time in Europe in 1983 in Switzerland, even though it was initially reported as R. indifferens. In Germany, the first specimens were caught in 1999 in Rheinland-Pfalz. Since 2004, the number of insects caught in cherry-growing areas increased considerably and progressively and frequently been detected in other parts of the country. Although it is widely distributed in the Netherlands and Hungary, its distribution remains relatively restricted in Austria, Belgium, Croatia, Poland and Slovenia (fig. 1) (Egartner et al., Reference Egartner, Zeisner, Hausdorf and Blümel2010; EFSA PHL Panel, 2014). Rhagoletis cingulata attacks all cultivated and wild cherries but is particularly damaging to late-maturing varieties, especially sour cherries of the widely planted ‘Schattenmorellen’ variety in Germany (EFSA PHL Panel, 2014). It can also attack and complete its life cycle in P. serotina (black cherry), P. mahaleb (St. Lucie cherry) and P. virginiana (choke cherry), which are considered secondary hosts (Glasgow, Reference Glasgow1933; Johannesen et al., Reference Johannesen, Keyghobadi, Schuler, Stauffer and Vogt2013). However, P. serotina is the main host of R. cingulata in the Netherlands (EFSA PHL Panel, 2014). Additionally to cherries, P. salinica (Japanese plum) and Pyrus communis (European pear) are major hosts of R. cingulata.
Finally, R. completa (that has recently invaded Europe), is also considered native to North America (Bush, Reference Bush1966; Smith & Bush, Reference Smith and Bush1997). Its current distribution includes North-eastern Mexico and South-western USA (Texas) up to Kansas (Central USA) (Bush & Smith, Reference Bush and Smith1998; Rull et al., Reference Rull, Tadeo, Aluja, Guillén, Egan and Feder2012; Yee et al., Reference Yee, Hernández-Ortiz, Rull, Sinclair and Neven2014). Regarding Europe, it was initially identified in Switzerland (1988) and Italy (1991), from where it spread to eight additional countries: Bosnia and Herzegovina and Croatia where it is regularly observed as well as in Austria, France, Germany, Hungary, Spain and Slovenia with restricted or occasional occurrence (fig. 1) (Duso & Dal Lago, Reference Duso and Dal Lago2006; Ostojic et al., Reference Ostojic, Zovko and Petrovic2014; Verheggen et al., Reference Verheggen, Verhaeghe, Giordanengo, Tassus and Escobar-Gutiérrez2017). In 2018, it was detected in Slovakia (Kozánek et al., Reference Kozánek, Semelbauer and Bartoš2018). According to Aluja et al. (Reference Aluja, Guillén, Rull, Höhn, Frey, Graf and Samietz2011), R. completa can be considered as an example of an alien species that settled first in the Mediterranean Basin and then invaded several central European countries by crossing the Alps. Rhagoletis completa attacks several species of walnuts (Juglans spp.) such as Juglans nigra, J. hindsii and J. californica and J. regia in North America. Juglans regia is the only economically significant host in Europe, particularly across cultivars with large and heavy fruits (Bush, Reference Bush1966; Guillén et al., Reference Guillén, Aluja, Rull, Höhn, Schwizer and Samietz2011). In unmanaged orchards, 100% of walnut trees can be infested causing losses in walnut yields of up to 80% (Verheggen et al., Reference Verheggen, Verhaeghe, Giordanengo, Tassus and Escobar-Gutiérrez2017).
Genetic relationships among the three Rhagoletis species
Rhagoletis cingulata belongs to the cingulata group, R. completa to the suavis group and R. cerasi to the Palearctic cerasi group (Bush, Reference Bush1966; Smith & Bush, Reference Smith and Bush1997, Reference Smith, Bush, Aluja and Norrbom2000; Smith et al., Reference Smith, Jaycox, Smith-Caldas and Bush2005). Studies based on genetic markers such as mitochondrial sequences (mainly COI and COII; cytochrome oxidase I and II genes) and morphological characters suggest the proximity of suavis and cingulata groups while the cerasi group is distantly related (Smith & Bush, Reference Smith and Bush1997, Reference Smith, Bush, Aluja and Norrbom2000; Smith et al., Reference Smith, Jaycox, Smith-Caldas and Bush2005; Ramírez et al., Reference Ramírez, Salazar, Palma, Cordero and Meza-Basso2008; Rull et al., Reference Rull, Tadeo, Aluja, Guillén, Egan and Feder2012; Frey et al., Reference Frey, Guillén, Frey, Samietz, Rull and Aluja2013). A recent comparison of mtDNA divergence among the R. suavis, R. pomonella and R. cingulata species groups suggested that these taxa do not share a common biogeographic history, diverging in different regions at different times in the past, despite current similarities in geographic distributions in the United States and Mexico (Glover et al., Reference Glover, Egan, Hood, Rull, Aluja and Feder2018).
Microsatellite markers that amplify across taxa can be a very powerful tool for the resolution of closely related species (cryptic or under incipient speciation) since they are quite stable and conserved in close taxa. In fact, microsatellite markers have been used for the resolution of species complexes in Tephritidae, such as that of Bactrocera correcta (Bezzi) (Qin et al., Reference Qin, Buahom, Krosch, Du, Wu, Malacrida, Deng, Liu, Jiang and Li2016), B. dorsalis (Hendel) (Krosch et al., Reference Krosch, Schutze, Armstrong, Boontop, Bonkin, Chapman, Englezou, Cameron and Clarke2013), B. musae (Tryon) (Drew et al., Reference Drew, Ma, Smith and Hughes2011) and the Ceratitis FAR cryptic species complex (De Meyer et al., Reference De Meyer, Delatte, Ekesi, Jordaens, Kalinova, Manrakhan, Mwatawala, Steck, Van Cann, Vaníčková, Břízová and Virgilio2015). In Rhagoletis, microsatellite markers have been developed for R. cerasi (Arthofer et al., Reference Arthofer, Riegler, Schneider, Krammer, Miller and Stauffer2009), R. completa (Chen et al., Reference Chen, Opp, Berlocher and Roderick2006), R. indifferens (Maxwell et al., Reference Maxwell, Rasic and Keyghobadi2009) and R. pomonella (Velez et al., Reference Velez, Taylor, Noor, Lobo and Feder2006) (table 1). Even though microsatellite markers have not been de novo developed for R. cingulata, Maxwell and colleagues (Reference Maxwell, Rasic and Keyghobadi2009) showed that all 16 microsatellite markers developed in their study for R. indifferens produced the expected amplicon in R. cingulata (16/16 amplified, 16/16 polymorphic). This result was expected, since the above two taxa are considered sister species, as verified in other recent independent studies (Drosopoulou et al., Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011; Johannesen et al., Reference Johannesen, Keyghobadi, Schuler, Stauffer and Vogt2013). Most of the R. indifferens markers amplify also in R. completa (13/16 amplified, 10/16 polymorphic) and to a lesser extent in R. cerasi (9/16 amplified, 6/16 polymorphic) (Augustinos et al., Reference Augustinos, Asimakopoulou, Papadopoulos and Bourtzis2011, Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014). Recently, Johannesen and colleagues (Reference Johannesen, Keyghobadi, Schuler, Stauffer and Vogt2013) managed to cross-amplify more R. indifferens microsatellite markers to R. cerasi (13/14 functional, 12/14 polymorphic). Nonetheless, as it is reported therein, the above loci were significantly less polymorphic in R. cerasi samples than samples derived from R. cingulata and R. indifferens, which is frequently reported when microsatellite markers are transferred from one species to another and becomes more evident at genetically distant species. The distant relationship between R. cerasi and the other two species is further supported by the reduced transferability of microsatellite markers developed for R. cerasi to the other two species since only 3/13 produced the expected amplicon in each species (Arthofer et al., Reference Arthofer, Riegler, Schneider, Krammer, Miller and Stauffer2009). Moreover, a subset of the R. pomonella microsatellite markers developed by Velez et al. (Reference Velez, Taylor, Noor, Lobo and Feder2006) were employed for the genotyping of R. cerasi (Augustinos et al., Reference Augustinos, Asimakopoulou, Papadopoulos and Bourtzis2011, Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014) and R. cingulata populations (Drosopoulou et al., Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011). Consistent with the above results and the established taxonomy, many genetic markers performed well in R. cingulata and fewer in R. cerasi. Although not providing more insight into the genetic relationships of the different Rhagoletis species under study, Chen et al. (Reference Chen, Opp, Berlocher and Roderick2006, Reference Chen, Berlocher, Opp and Roderick2010) successfully genotyped R. completa samples using three of the R. pomonella microsatellite markers, thus further enriching the microsatellite marker ‘toolkit’ for Rhagoletis species. Table 1 summarizes efforts to cross-amplify microsatellite markers in different Rhagoletis species. Markers that are polymorphic, nuclear and can be in situ localized on chromosomes can support progress in different fields of Rhagoletis research. Ongoing and future whole genome sequencing efforts will provide further markers useful for genetic and genomic studies.
Microsatellite markers developed for different Rhagoletis species and their cross-amplification in other Rhagoletis species.
Results in brackets refer to the original studies. Subsequent studies dealing with cross-species amplification are indicated with superscript Latin numbers, under the findings of the original study: IAugustinos et al. (Reference Augustinos, Asimakopoulou, Papadopoulos and Bourtzis2011); IIDrosopoulou et al. (Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011); IIIChen et al. (Reference Chen, Opp, Berlocher and Roderick2006)
Rcer: R. cerasi; Rpom: R. pomonella; Rind: R. indifference; Rcom: R. completa.
Column A: Polymerase chain reaction (PCR) amplification and agarose gel electrophoresis; column B: polymorphism analysis through genotyping of a varying number of individuals; ±: polymorphism of cross-amplified markers was tested through PCR, cloning and measure of size of recovered clones; +: amplified; X: not amplified (or poor resolution during genotyping); nt: not tested; –: not done, due to PCR failure; P: polymorphic; M: monomorphic; *for R. ribicola: only two individuals tested; Rpomv: developed for R. pomonella by Velez et al. (Reference Velez, Taylor, Noor, Lobo and Feder2006), but without reference of testing by them in other species.
Cytogenetic studies and polytene chromosome maps
In general, Rhagoletis species consist of six pairs of chromosomes (Bush, Reference Bush1966; Bush & Boller, Reference Bush and Boller1977; Procunier & Smith, Reference Procunier and Smith1993). Specifically, early cytological analyses of R. cerasi demonstrated its karyotype to consist of six pairs of chromosomes, including one pair of sex chromosomes (Bush, Reference Bush1966). More recent cytological analyses have further confirmed the previous karyotype analysis of Kounatidis et al. (Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008). The same number of chromosomes (2n = 12) was also found in the metaphase mitotic complements of R. completa (Drosopoulou et al., Reference Drosopoulou, Koeppler, Kounatidis, Nakou, Papadopoulos, Bourtzis and Mavragani-Tsipidou2010) and R. cingulata (Drosopoulou et al., Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011). On the contrary, R. meigenii was found to consist of four pairs of autosomes with the male and female karyotype having X0 and XX, respectively (Bush & Boller, Reference Bush and Boller1977).
In most genera of tephritids, including Anastrepha, Bactrocera and Ceratitis, the sex chromosomes are heteromorphic and easily identified, with the Y chromosome being smaller and dot-like (Bedo, Reference Bedo1986; Zacharopoulou, Reference Zacharopoulou1987; Mavragani-Tsipidou et al., Reference Mavragani-Tsipidou, Karamanlidou, Zacharopoulou, Koliais and Kastritsis1992; Cevallos & Nation, Reference Cevallos and Nation2004; Garcia-Martinez et al., Reference Garcia-Martinez, Hernandez-Ortiz, Zepeta-Cisneros, Robinson, Zacharopoulou and Franz2009; Zacharopoulou et al., Reference Zacharopoulou, Augustinos, Sayed, Robinson and Franz2011a, Reference Zacharopoulou, Sayed, Augustinos, Yesmin, Robinson and Franzb). However, the X and Y chromosomes of R. cerasi, R. cingulata and R. completa are very similar in length. Sex chromosomes are long in the case of R. cerasi, and very small, dot-like, in the case of R. cingulata and R. completa (Procunier & Smith, Reference Procunier and Smith1993; Kounatidis et al., Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008; Drosopoulou et al., Reference Drosopoulou, Koeppler, Kounatidis, Nakou, Papadopoulos, Bourtzis and Mavragani-Tsipidou2010, Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011). This observation is consistent with the closer phylogenetic relationship of the latter two species and is in accordance with morphological and genetic data, as well as with the established taxonomy (Smith and Bush, Reference Smith and Bush1997, Reference Smith, Bush, Aluja and Norrbom2000).
Comparative analysis of the polytene chromosomes banding pattern of R. cerasi, R. cingulata and R. completa showed extensive homology of certain polytene regions among these species, with the most extensive chromosome banding pattern conservation recorded between R. cingulata and R. completa (Drosopoulou et al., Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011). This homology, together with the similarity found in the mitotic karyotype of the two species, is indicative of their close phylogenetic relationship. Based on the polytene and mitotic karyotype analysis, R. cerasi seems to be more distantly related to the above species. This observation is also in agreement with the phylogenetic relationships accepted for the species of the Rhagoletis genus (Smith and Bush, Reference Smith and Bush1997, Reference Smith, Bush, Aluja and Norrbom2000).
Another finding, that raises a series of interesting questions, is the presence of multiple asynaptic phenomena both in R. cerasi and R. cingulata, but not in R. completa (Kounatidis et al., Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008; Drosopoulou et al., Reference Drosopoulou, Koeppler, Kounatidis, Nakou, Papadopoulos, Bourtzis and Mavragani-Tsipidou2010, Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011). It has been proposed that such asynaptic phenomena may be linked to the presence of endosymbiotic bacteria, such as Wolbachia, and evolutionary relationship events of integration of their DNA (in part or whole) into the chromosomal DNA (Kounatidis et al., Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008; Drosopoulou et al., Reference Drosopoulou, Koeppler, Kounatidis, Nakou, Papadopoulos, Bourtzis and Mavragani-Tsipidou2010, Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011).
Genetic structuring and population dynamics
Despite the progress in genetic markers development, little is known about these three species regarding the genetic structure and dynamics of their natural populations. In R. cerasi, Schwarz et al. (Reference Schwarz, McPheron, Hartl, Boller and Hoffmeister2003) found little evidence for host race formation, using allozyme markers for the analysis of natural populations derived from Switzerland and Germany and two different host plants, L. xylosteum L. and P. avium L. Recently, Augustinos and his colleagues (Reference Augustinos, Asimakopoulou, Papadopoulos and Bourtzis2011, Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014) used some of the microsatellite markers developed for R. cerasi (Arthofer et al., Reference Arthofer, Riegler, Schneider, Krammer, Miller and Stauffer2009), in addition to several more cross-amplified ones from other Rhagoletis species (R. pomonella and R. indifferens), to perform a population analysis targeting mainly R. cerasi populations in Greece, Northern Europe (Germany) and Russia. This analysis revealed at least three to four genetic groups clustering samples from: (a) Germany, (b) mainland Greece and some Greek islands, (c) Eastern Aegean islands and (d) Russia (just one sample).
Regarding R. cingulata, recent attempts using microsatellite markers failed to determine any genetic differentiation correlated with the host preference or geographic origin of populations derived from the native area of Michigan (Smith et al., Reference Smith, Powell, Teixeira, Armstrong, McClowry, Isaacs, Hood, Feder and Gut2014). On the other hand, Johannesen et al. (Reference Johannesen, Keyghobadi, Schuler, Stauffer and Vogt2013), utilizing also microsatellite markers, clearly demonstrated that R. cingulata samples from the recently invaded regions of Germany and Hungary are genetically different. Based on these findings, this study raised the possibility of independent invasions of this pest in Europe. The same study also detected signals of hybridization events between R. cerasi and R. cingulata. This hybridization ‘sets the alarm on’ regarding the management of cherry pests in the area since it may generate new more aggressive biotypes of pests, such as the ‘Lonicera fly’ which has emerged from the hybridization of R. mendax and R. zephyria (Schwarz et al., Reference Schwarz, McPheron, Hartl, Boller and Hoffmeister2003).
Based on allozyme markers, Berlocher (Reference Berlocher1984) revealed high polymorphism during the colonization process of R. completa in California considering native and introduced populations. Chen and colleagues (Reference Chen, Opp, Berlocher and Roderick2006, Reference Chen, Berlocher, Opp and Roderick2010) used a set of five microsatellite markers (along with allozymes) to analyse natural populations based on new and historical collections. In the first study, they did not observe significant bottlenecks; in fact, in some cases the introduced populations seemed to be more polymorphic than native ones (Chen et al., Reference Chen, Opp, Berlocher and Roderick2006). Since the introduced populations harboured alleles not sampled in the native ones, the need for more extensive sampling in native areas, and the hypothesis of multiple introduction events, were discussed and further addressed in the second study (Chen et al., Reference Chen, Berlocher, Opp and Roderick2010). This study gave insight into R. completa natural population genetic structuring and provided important findings, such as: (a) the need for good sampling across the distribution area of the species; (b) the decline of structuring in introduced and native populations over time and (c) greater fluctuation of genetic variability of introduced populations and increase in their genetic variability over time.
Rhagoletis–Wolbachia associations: dynamic and still unresolved
Recently, much interest has been shown on studies regarding the intracellular bacterium Wolbachia that is found in more than 40% of the terrestrial arthropod species (Werren et al., Reference Werren, Baldo and Clark2008; Zug & Hammerstein, Reference Zug and Hammerstein2012). The presence of Wolbachia has been associated with the induction of a variety of host reproductive phenotypes such as feminization, parthenogenesis, male killing and cytoplasmic incompatibility (CI) (for review see Werren et al., Reference Werren, Baldo and Clark2008; Saridaki & Bourtzis, Reference Saridaki and Bourtzis2010; Mateos et al., Reference Mateos, Martinez, Lanzavecchia, Conte, Guillén, Morán-Aceves, Toledo, Liedo, Asimakis, Doudoumis, Kyritsis, Papadopoulos, Avgoustinos, Segura, Tsiamis and Bourtzis2019). For these reasons, Wolbachia is considered a key player affecting biological and evolutionary processes as well as a tool for the control of agricultural pests and disease vectors (Werren et al., Reference Werren, Baldo and Clark2008; Saridaki & Bourtzis, Reference Saridaki and Bourtzis2010; Mateos et al., Reference Mateos, Martinez, Lanzavecchia, Conte, Guillén, Morán-Aceves, Toledo, Liedo, Asimakis, Doudoumis, Kyritsis, Papadopoulos, Avgoustinos, Segura, Tsiamis and Bourtzis2019).
In the late 1970s, researchers observed the existence of reproductive incompatibility between European populations of R. cerasi derived from north and south Europe (Matolin, Reference Matolin1976; Boller, Reference Boller, Robinson and Hooper1989). Riegler & Stauffer (Reference Riegler and Stauffer2002) attributed this incompatibility to the presence of different Wolbachia strains. Despite the univoltine life cycle of R. cerasi populations and the limited dispersal abilities of adults (Daniel & Grunder, Reference Daniel and Grunder2012), it has been shown that Wolbachia spread in natural host populations is rapid (Bakovic et al., Reference Bakovic, Schebeck, Telschow, Stauffer and Schuler2018). In fact, all natural populations of R. cerasi studied so far have been found to be 100% Wolbachia-infected (Riegler & Stauffer, Reference Riegler and Stauffer2002; Kounatidis et al., Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008; Arthofer et al., Reference Arthofer, Riegler, Schneider, Krammer, Miller and Stauffer2009, Reference Arthofer, Riegler, Schuler, Schneider, Moder, Miller and Stauffer2011; Augustinos et al., Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014; Karimi & Darsouei, Reference Karimi and Darsouei2014). Hosts can be infected by one or more distinct strains of Wolbachia (table 2), with the European populations of R. cerasi were found to be infected by five different Wolbachia strains, named wCer1 to wCer5 (Riegler & Stauffer, Reference Riegler and Stauffer2002; Kounatidis et al., Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008; Arthofer et al., Reference Arthofer, Riegler, Schneider, Krammer, Miller and Stauffer2009, Reference Arthofer, Riegler, Schuler, Schneider, Moder, Miller and Stauffer2011; Augustinos et al., Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014); and R. cerasi samples from Iran were found to be infected with a single new strain of Wolbachia, named wCer6 (Karimi & Darsouei, Reference Karimi and Darsouei2014). But, the R. cerasi–Wolbachia interactions seem to be even more complicated than this. Augustinos and colleagues (Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014) pointed to the presence of additional uncharacterized Wolbachia strains on R. cerasi populations from Greece, and raised concerns about the suitability of the currently used diagnostic markers for genotyping. Considering the information mentioned earlier, along with the fact that Wolbachia cannot be cultivated in the laboratory, the presence of low-titre infections and/or multiple infections in the same individual, as well as the highly recombinant nature of the bacterial chromosome, make Wolbachia characterization in R. cerasi samples a very difficult task (Zabalou et al., Reference Zabalou, Riegler, Theodorakopoulou, Stauffer, Savakis and Bourtzis2004, Reference Zabalou, Apostolaki, Livadaras, Franz, Robinson, Savakis and Bourtzis2009; Schneider et al., Reference Schneider, Riegler, Arthofer, Merçot, Egan and Feder2013; Mateos et al., Reference Mateos, Martinez, Lanzavecchia, Conte, Guillén, Morán-Aceves, Toledo, Liedo, Asimakis, Doudoumis, Kyritsis, Papadopoulos, Avgoustinos, Segura, Tsiamis and Bourtzis2019). Allelic intersection analysis provided an additional tool for the characterization of multiple Wolbachia infections in R. cerasi (Arthofer et al., Reference Arthofer, Riegler, Schuler, Schneider, Moder, Miller and Stauffer2011); however, this analysis requires a priori knowledge on the Wolbachia strains harboured by a species and can be quite laborious. The rapid technological advances in the field of sequencing technologies (such as different types of next generation sequencing and single cell genomics) are expected to shed light on the Wolbachia status of Rhagoletis species and gradually overcome difficulties related to the presence of multiple and/or low-titre Wolbachia infections.
Wolbachia status in Rhagoletis species.
Based on multilocus sequence typing (MLST) and wsp gene analysis, all populations of R. cingulata studied so far have been 100% infected with the Wolbachia wCin2 strain (table 2), which is identical to wCer2 of R. cerasi (Schuler et al., Reference Schuler, Arthofer, Krumböck, Köppler, Vogt, Teixeira, Riegler and Stauffer2009, Reference Schuler, Bertheau, Egan, Feder, Riegler, Schlick-Steiner, Steiner, Johannesen, Kern, Tuba, Lakatos, Köppler, Arthofer and Stauffer2013; Drosopoulou et al., Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011). In addition, Schuler et al. (Reference Schuler, Bertheau, Egan, Feder, Riegler, Schlick-Steiner, Steiner, Johannesen, Kern, Tuba, Lakatos, Köppler, Arthofer and Stauffer2013) demonstrated that several European populations of R. cingulata are infected also with the wCin1 strain (identical to wCer1 of R. cerasi, again based on MLST and wsp genes analysis), but not in any individual studied from the USA (table 2). They have also suggested a possible horizontal transmission of Wolbachia from R. cerasi to R. cingulata during the invasion route of this species in Europe. Screening of one wild population of R. completa (Drosopoulou et al., Reference Drosopoulou, Koeppler, Kounatidis, Nakou, Papadopoulos, Bourtzis and Mavragani-Tsipidou2010), originating from Southern Germany, gave no evidence of Wolbachia infection. Nonetheless, a more recent study on a population collected from Austria (Schuler et al., Reference Schuler, Arthofer, Krumbock, Bertheau and Stauffer2012) showed the presence of low-titre Wolbachia infections with strains similar to wCer1, wCer2 and wCer5 (table 2). The authors suggested that such infections are unlikely to induce reproductive phenotypes such as CI and therefore their possible impact in natural populations is unclear. Regarding R. pomonella, Schuler et al. (Reference Schuler, Arthofer, Riegler, Bertheau, Krumböck, Köppler, Vogt, Teixeira and Stauffer2011) screened both host races (apple and hawthorn-infesting species) for Wolbachia infections and detected the presence of wPom1 and at least three more uncharacterized strains. Specifically, wPom1 strain found to occur in both host races, whereas different sequence types were found at low frequencies only in apple-infesting R. pomonella.
Implementation of SIT and IIT methods
The SIT is a species-specific method of pest population suppression that relies on repetitive releases of mass-produced sterile insects, ideally only males. Sterile males are expected to compete with wild males and mate with wild females, inducing sterility in the wild females (Dyck et al., Reference Dyck, Hendrichs and Robinson2005). For more than 30 years, cytogenetics of tephritids have played a catalytic role in the development of key tools in support of SIT, including the characterization of tephritid genetic sexing strains and the support of integrative taxonomic studies for clarifying relationships between closely related species and/or incipient speciation phenomena (Zacharopoulou et al., Reference Zacharopoulou, Augustinos, Drosopoulou, Tsoumani, Gariou-Papalexiou, Franz, Mathiopoulos, Bourtzis and Mavragani-Tsipidou2017). In fact, polytene chromosomes are an important tool for understanding organization and evolution of chromosomes, mapping of traits of interest, and analysing chromosomal rearrangements (Robinson et al., Reference Robinson, Franz and Fisher1999; Zacharopoulou et al., Reference Zacharopoulou, Augustinos, Drosopoulou, Tsoumani, Gariou-Papalexiou, Franz, Mathiopoulos, Bourtzis and Mavragani-Tsipidou2017). Therefore, the development of polytene chromosome maps for R. cerasi, R. cingulata and R. completa (Kounatidis et al., Reference Kounatidis, Papadopoulos, Bourtzis and Mavragani-Tsipidou2008; Drosopoulou et al., Reference Drosopoulou, Koeppler, Kounatidis, Nakou, Papadopoulos, Bourtzis and Mavragani-Tsipidou2010, Reference Drosopoulou, Augustinos, Nakou, Koeppler, Kounatidis, Vogt, Papadopoulos, Bourtzis and Mavragani-Tsipidou2011) can support SIT approaches in Rhagoletis species in general. Apart from cytogenetics tools, microsatellite markers used for identifying the genetic structure and population dynamics of the three Rhagoletis species could provide valuable insights into colonization patterns and phylogenetic relationships of this species and into ecological strategies in the field.
The mechanism of Wolbachia-induced CI is receiving much attention as a tool for pest and disease control (Mateos et al., Reference Mateos, Martinez, Lanzavecchia, Conte, Guillén, Morán-Aceves, Toledo, Liedo, Asimakis, Doudoumis, Kyritsis, Papadopoulos, Avgoustinos, Segura, Tsiamis and Bourtzis2019; Nikolouli et al., Reference Nikolouli, Colinet, Renault, Enriquez, Mouton, Gibert, Sassu, Cáceres, Stauffer, Pereira and Bourtzis2018). It is worth noting that the IIT, which is based on the Wolbachia-induced CI, had been applied for the population control of R. cerasi even before the aetiological agent (Wolbachia infections) of the reproductive incompatibilities observed between northern and southern European populations was discovered (Boller et al., Reference Boller, Russ, Vallo and Bush1976; Matolin, Reference Matolin1976; Boller, Reference Boller, Robinson and Hooper1989; Riegler & Stauffer, Reference Riegler and Stauffer2002). In the same context, infected R. cerasi populations were used as Wolbachia donors to species that do not harbour this bacterium, such as the Mediterranean fruit fly (medfly), Ceratitis capitata (Wiedemann) (Zabalou et al., Reference Zabalou, Riegler, Theodorakopoulou, Stauffer, Savakis and Bourtzis2004, Reference Zabalou, Apostolaki, Livadaras, Franz, Robinson, Savakis and Bourtzis2009) and the olive fly, B. oleae (Rossi) (Apostolaki et al., Reference Apostolaki, Livadaras, Saridaki, Chrysargyris, Savakis and Bourtzis2011) for the development of IIT-based population control. In diplo-diploid species, CI is commonly expressed as embryonic lethality in crosses between Wolbachia-infected males with females, which are either uninfected or they carry a different Wolbachia strain.
In principle, the IIT technique based on Wolbachia could be used in a way analogous to the SIT; mass-released Wolbachia-infected males are expected to suppress a target pest population. However, the application of IIT strictly depends on the availability of an efficient and robust sexing system, otherwise a population suppression strategy may result in population replacement. In the absence of a perfect sexing system, the combination of SIT and IIT has been proposed via the application of low doses of irradiation which can sterilize any accidentally released females while males would be rendered sterile by both Wolbachia and irradiation (Bourtzis & Robinson, Reference Bourtzis, Robinson, Bourtzis and Miller2006; Brelsfoard et al., Reference Brelsfoard, Clair and Dobson2009; Bourtzis et al., Reference Augustinos, Asimakopoulou, Moraiti, Mavragani-Tsipidou, Papadopoulos and Bourtzis2014, Reference Bourtzis, Lees, Hendrichs and Vreysen2016; Zhang et al., Reference Zhang, Lees, Xi, Gilles and Bourtzis2015a, Reference Zhang, Zheng, Xi, Bourtzis and Gillesb, Reference Zhang, Lees, Xi, Bourtzis and Gilles2016; Mateos et al., Reference Mateos, Martinez, Lanzavecchia, Conte, Guillén, Morán-Aceves, Toledo, Liedo, Asimakis, Doudoumis, Kyritsis, Papadopoulos, Avgoustinos, Segura, Tsiamis and Bourtzis2019). In any case, the development of a population suppression strategy against Rhagoletis species via SIT, IIT or a combination of irradiation and Wolbachia-based approaches would require a mass rearing system; unfortunately, the fact that these species are univoltine and an artificial diet has yet to be developed hinder, for the time being, any plans towards this direction.
Conclusions
In recent years, there is major concern regarding invasive insect species including their interactions with native fauna. Depending on their genetic and behavioural proximity, alien and native species could interact either through mating or through exchanging genetic material (e.g. symbionts) due to common plant hosts and/or predators and parasitoids. These interactions could lead to new pest dynamics through the development of hybrid species or altered behaviour of existing species, phenomena that Europe has experienced during the last decades following the invasion of new members of the Rhagoletis genus. The prediction of these changes and therefore the control of these pests certainly becomes more difficult. It is clear that the clarification of the species identity by using molecular and (cyto)genetic approaches, population analysis and characterization of the symbionts associated with the target pest populations can facilitate control strategies such as SIT, IIT and other genetic methods, which are environment-friendly and species specific. In the current review, the currently available knowledge on the cytogenetics, the population genetic structure and Wolbachia infections of Rhagoletis pest species in Europe was summarized, highlighting the importance of such knowledge as an essential prerequisite of any population control programme.
Acknowledgements
We are grateful to the FAO/IAEA Coordinated Research Program ‘Use of symbiotic bacteria to reduce mass-rearing costs and increase mating success in selected fruit pests in support of SIT application’ for the overall support of this study. This study was also partially (a) supported by IKYDA grant awarded to NTP, and (b) co-funded by the European Social and Natural Resources – EPEAEK II – Pythagoras and the Greek Ministry of Education.
