- Top of page
- Materials and methods
Changes in host acceptance is an important factor in the host specialization of phytophagous insects, and knowledge of the genetic organization of this behaviour is necessary in order to understand how host shifts occur. Here we describe the inheritance of adult host acceptance (oviposition) in three closely related species of Yponomeuta Latreille (Lepidoptera: Yponomeutidae), and their interspecific hybrids. Yponomeuta cagnagellus (Hübner), a specialist on Euonymus europaeus L. (Celastraceae), Y. malinellus Zeller, a specialist on Malus spp. (Rosaceae), and Y. padellus (L.), oligophagous on a number of Rosaceae, were tested for their acceptance of parental hosts in choice tests. Acceptance of E. europaeus is semi-dominant in hybrids of Y. cagnagellus×Y. padellus, and in hybrids of Y. cagnagellus×Y. malinellus. The dominance of this acceptance was confirmed in oviposition tests with backcross hybrids: backcross hybrids F1 × Y. cagnagellus oviposited mainly on E. europaeus and F1 × Y. padellus still deposited more than half of their egg masses on E. europaeus. Reciprocal hybrids did not differ in their host acceptance, indicating that the trait is autosomal. We further studied the effect of larval food on adult host acceptance (‘Hopkins host selection principle’) in split full-sib F1 families. Larval diet influenced oviposition only in one of two hybrid crosses. The F1 hybrid of Y. padellus× Y. cagnagellus, reared on Prunus spinosa L., deposited a significantly lower percentage of egg masses on E. europaeus compared to their full-sib sisters fed with E. europaeus. We did not find this in the reciprocal cross. However, still more than half of the egg masses are deposited on E. europaeus by hybrids that have no experience on this host. We conclude that the semi-dominant character of acceptance of E. europaeus and a tendency of Rosaceae-feeding Yponomeuta to deposit egg masses on this host may have created the opportunity for the host shift of the predecessor of Y. cagnagellus from Rosaceae to the Celastraceae. This shift may have been further facilitated by a weak tendency of adults to oviposit on their larval food source.
- Top of page
- Materials and methods
More than three-quarters of all described species are insects, the majority of which feed on plants. Many of these phytophagous insects are specialists, only feeding on one or a few related plant species (Schoonhoven et al., 1998). The colonization of, and subsequent specialization, on new – sometimes unrelated – host plants occurs frequently within insect species, and this is considered an important factor promoting insect speciation (Ehrlich & Raven, 1964; Mitter & Farrell, 1991; Dres & Mallet, 2002; Funk et al., 2002). There is growing support for the theory that host shifts, and subsequent speciation, can occur in sympatry in initially randomly mating populations (e.g., Bush, 1994; Coyne, 1994; Schluter, 2001; Via, 2001; Via & Hawthorne, 2002). The critical step in this process is the development of a linkage disequilibrium between several genes involved in adaptation to host plants, in spite of the homogenising effects of random mating and recombination. Empirical evidence indicates that ecology plays a key role in this process (Schluter & Nagel, 1995; Orr, 1998; Hatfield & Schluter, 1999; Groman & Pellmyr, 2000; de Mazancourt & Dieckmann, 2004). Mathematical modelling shows the likelihood of such ecological speciation, provided that some assortative mating takes place to counteract recombination between the adaptation genes (Felsenstein, 1981; Johnson et al., 1996; Kondrashov & Shpak, 1998; Dieckmann & Doebeli, 1999; Kondrashov & Kondrashov, 1999; Doebeli & Dieckmann, 2000; Via, 2001; Kirkpatrick & Ravigne, 2002). Although genes for host adaptation are important parameters in such models, detailed experimental data on the actual genetic basis of host adaptation is largely limited to some well-studied species such as those in the genus Drosophila (Jaenike, 1990; Coyne, 1992). Only recently, studies on other species including some Lepidoptera have appeared (Hawthorne & Via, 2001, and references therein). Clearly, we need information on the genetics of host adaptation from a broader set of insects to increase our understanding of the evolution of host associations in insect herbivores and of the effects host-plant shifts can have on speciation.
Host adaptation is a complex process, whose components most likely have separate genetic bases. A new host can be incorporated into an insect's diet if adults accept it for oviposition, larvae accept it for feeding, and the larvae are able to complete their life cycle on it. In this study, we use the term ‘host use’ to indicate the suite of adult and larval host acceptance and larval performance. To describe oviposition behaviour, we explicitly use the term ‘host acceptance’ in favour of the more common ‘host preference’ or ‘host choice’. The reason is that ‘acceptance’ simply denotes the observed behaviour, without making inferences about the hypothetical underlying mechanisms (Schoonhoven et al., 1998). Adult host acceptance is the basic constituent of host use in Lepidoptera, as first stadium larvae have little mobility and are generally unable to locate and migrate to a suitable food source if their mother chooses to oviposit on an unsuitable plant (Jermy, 1984; Miller & Strickler, 1984; Thompson & Pellmyr, 1991).
The main hurdle to studying the inheritance of host use is the general lack of suitable biological systems. A suitable system would be one that enables crosses between individuals specialized on different hosts, such as two populations of a species varying in host affiliations, or recently diverged species with differing host associations that have not yet evolved high levels of post-zygotic reproductive isolation (Coyne, 1992; Bush & Smith, 1998; K.H. Hora, F. Marec, P. Roessingh & S.B.J. Menken, unpubl.). Closely related species in the small ermine moth genus Yponomeuta Latreille (Yponomeutidae: Lepidoptera) provide such a system (Menken et al., 1992; Menken & Roessingh, 1998). In this model system we describe here the inheritance of adult host acceptance. We analyse the differences between three related Yponomeuta species feeding on Celastraceae and Rosaceae, which belong to different plant orders within the Eurosids I. Yponomeuta cagnagellus (Hübner) is specialised on Euonymus europaeus L. (Celastraceae), Y. malinellus Zeller is a specialist on Malus spp. (Rosaceae), and Y. padellus (L.) is oligophagous on several rosaceous plants, but is mainly found on Crataegus monogyna Jack. and Prunus spinosa L. (Menken et al., 1992). In the Netherlands, the three species do not hybridise in nature (Menken, 1980), but hybrids can be obtained in the laboratory. F1 hybrids develop well on both parental hosts, and do not show decreased fertility (K.H. Hora, F. Marec, P. Roessingh & S.B.J. Menken, unpubl.). Hybrids of these species can be viewed as artificial intermediate stages of host specialization on the two plant families enabling the study of inheritance of host use loci.
Adult host acceptance in Lepidoptera is controlled by the perception of leaf surface chemicals (Renwick & Chew, 1994). Lepidoptera will in general oviposit on artificial substrates if the appropriate leaf surface compounds are present. In Y. cagnagellus, it has been demonstrated that the surface extracts of host plants are sufficient to stimulate oviposition (Hora & Roessingh, 1999b), indicating that gustation is the principal mode of perception involved in adult host acceptance. In Y. padellus this is not yet known. A phytochemical basis of host acceptance behaviour provides an indication of the nature of the genes involved and suggests an evolutionary mechanism for host shifts through modifications in receptor proteins (Menken & Roessingh, 1998).
Besides being genetically determined, adult host acceptance behaviour could also be influenced by larval experience. A causal link between larval food experience and adult preference would facilitate host shifts (Hopkins, 1917). Such an epigenetic effect of larval food on adult oviposition has only been found in a few studies, given the number of attempts to find it (see Barron, 2001, for review) but it remains an interesting possibility in the context of the evolution of host acceptance.
Using interspecific hybrids, we address the following questions with regard to adult host acceptance:
Is adult host acceptance based on the perception of plant surface compounds?
How is adult host acceptance inherited?
Is there, in addition to genetic effects, also an induced epigenetic effect of larval diet on the adult host acceptance of F1 hybrids?