Tospoviruses : Proceedings of the 7 Th International Symposium on Thysanoptera

Tospoviruses are considered to be dependent on Thysanoptera for their existence in nature, but scarcely 0.15% of Thysanoptera species (9 of 5500) are known to be vectors. Moreover, these vector species are not closely related to each other, suggesting either that many thrips species have lost an association with tospoviruses or that each Tospovirus species has evolved an independent relationship with a thrips species. The lack of evidence for a long evolutionary relationship between thrips and tospoviruses, together with the lack of any evidence that tospoviruses are associated with any native Australian plant species nor any native Australian thrips species, despite that continent being where TSWV was first observed, raise questions concerning where, when, how, and how often the thrips/ tospovirus associations originated. Thysanoptera origin and radiation Thysanoptera are presumed to have evolved from fungus-feeding detritus-living ancestors, a habit that is retained in the members of the Merothripidae, and also in the Psocoptera that presumably shared a common ancestor with thrips (Mound, Heming & Palmer, 1980). Presumably there was an early radiation onto floral structures, and thrips may even have been involved as pollinators during the early radiation of flowering plants (Terry, 2002, this volume), one basal clade thrips genus retaining an association with the reproductive structures of a basal clade plant group, the cycad genus Macrozamia (Terry, 2001). Subsequently three major food sources were adopted by thrips – fungal hyphae and spores, green leaves, and flowers with or without leaves as well. A few species are also predators, and a very few feed only on mosses (Mound & Marullo, 1996). Currently, the 5000 recognised species of thrips are arranged into nine families, the largest of which is the Phlaeothripidae, the sole family in the suborder Tubulifera. The 3200 species in this family exhibit a wide range of biologies, but the 700 members of the subfamily Idolothripinae all feed by imbibing whole fungal spores, a few species being known to have a special sporecrushing apparatus in the fore gut (Mound & Palmer, 1983). In contrast, about half of the 2500 16 THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 17 species in the subfamily Phlaeothripinae feed on fungal hyphae rather than spores, with most of the remaining species feeding on flower or leaf tissues, including a few on mosses, and a very few species predatory on other small arthropods. Some of the leaf-feeding Phlaeothripinae induce galls, particularly in tropical countries, but very few species in this subfamily attack any crops, and none are known to be associated with any tospoviruses. The other eight families in the Thysanoptera are all included in the suborder Terebrantia (Table 1) (Moritz, Morris & Mound, 2001). Members of the Merothripidae and Uzelothripidae are all very small thrips associated with fungal hyphae in warm countries. In contrast, members of the Melanthripidae are usually large and robust, and they all breed in flowers in temperate areas. The Aeolothripidae is a rather larger family, but whereas the species of the two largest genera, Aeolothrips and Desmothrips, are commonly phytophagous in flowers or non-obligate predators on other arthropods, the species in the many small genera from the tropics are all obligate predators. The species of the next two families are rather poorly known, although all five species in one genus of Adiheterothripidae are known only from the flowers of date palms, Phoenix dactylifera. The seventh family, Heterothripidae, is found only in the New World and, with one exception, all the species are flowerliving. The exception is of considerable biological interest, because it has recently been shown to be ectoparasitic on a species of Homoptera, a way of life that is unique amongst Thysanoptera (Izzo et al. 2001; Pinent et al, 2002, this volume). The eighth family of Terebrantia is, with 1700 known species, by far the largest. The Thripidae is found worldwide, from Greenland to the southern oceans, and includes almost all of the pest species of thrips. The species of many genera are associated only with grasses, whereas others are associated only with dicotyledonous plants, some in flowers but others only on leaves. Pest species are commonly more adaptable in their habits, and many of them feed and breed both on leaves and in flowers. Thrips associated with tospoviruses For Tospovirus workers, the important statistic in Table 1 is that only 9 of the 5500 known species of thrips have been shown to be associated with any of these diseases. Thus less than 0.2% of Thysanoptera species are known to be associated with tospoviruses, and there is no evidence that many more thrips species are likely to be implicated. This immediately suggests that the probability of a long evolutionary history between these organisms is unlikely. More significantly, the few thrips species that are associated with tospoviruses are not closely related to each other (Table 2). Only five of the 160 described species of Frankliniella are known to be vectors of tospoviruses, only three of the 280 species of the genus Thrips, and just one of the 90 species of Scirtothrips. Phylogenetically, the third of these thrips genera is widely distant from the first two (Mound & Marullo, 1996). Moreover, Thrips and Frankliniella are members of generic groups that cannot be considered closely related in view of their structural differences. Frankliniella genus-group is probably Gondwanan in origin, whereas Thrips genus-group is a relatively recent lineage, apparently evolving subsequent to the separation of the African and South American continents (Mound, 2002, this volume). Even within the two genera, Thrips and Frankliniella, the vector species are not closely related, neither phylogenetically nor geographically (Table 3). The Eurasian species, F. intonsa, is similar in structure to the western USA species, F. occidentalis, but the species most closely similar to the western flower thrips is F. panamensis (Moritz, Morris & Mound, 2001). F. fusca and F. zucchini both differ considerably in structure from these species, and F. schultzei seems even more distantly related (Mound, 2002, this volume). Similarly, the structural differences between the three species of genus Thrips that are known as vectors are so great that these three cannot be considered as being closely related within this large and diverse genus. Origin of the thrips/Tospovirus relationship Since the known vector species are not closely related to each other, two alternative evolutionary scenarios must be considered. Assuming that the thrips/Tospovirus associations had a single evolutionary origin, then this must have been early in the evolution of the Thripinae. This assumption involves the conclusion that the vast majority of thrips species subsequently SO MANY THRIPS – SO FEW TOSPOVIRUSES? 16 THRIPS AND TOSPOVIRUSES: PROCEEDINGS OF THE 7TH INTERNATIONAL SYMPOSIUM ON THYSANOPTERA 17 FAMILIES SUB-FAMILIES Genera currently valid Species currently valid Species vectoring Tospovirus Phlaeothripidae Phlaeothripinae 35