Introduction
Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) biotype B is a widely distributed pest of agriculture. It originated in the Middle East/Asia Minor region (De Barro et al., Reference De Barro, Trueman and Frohlich2005) and has since spread widely through the trade in ornamental plants (Cheek & Macdonald, Reference Cheek and Macdonald1994; Dalton, Reference Dalton2006). One of the striking features of this particular genetic group of B. tabaci is its capacity to induce a range of physiological changes in a range of host plants; these include irregular ripening in tomato, stem blanching in broccoli, uneven colouration of some coloured cauliflower heads, pod blanching in soybean and green bean, uneven ripening in jarradale pumpkin, fruit discolouration in squash, zucchini and jarradale pumpkin and silverleafing in squash, pumpkin and zucchini (Burger et al., Reference Burger, Schwartz and Paris1988; Yokomi et al., Reference Yokomi, Hoelmer and Osborne1990; Schuster et al., Reference Schuster, Mueller, Kring and Price1990; Cohen et al., Reference Cohen, Duffus, Liu and Perry1991; Costa et al., Reference Costa, Ullman, Johnson and Tabashnik1993a,Reference Costa, Ullman, Johnson and Tabashnikb; Chen et al., Reference Chen, McAuslane, Carle and Webb2004; De Barro, unpublished). The symptoms were first observed in Israel (Burger et al., Reference Burger, Schwartz and Paris1988) but were not associated with the feeding by B. tabaci until the outbreak by the B biotype in the USA (Maynard & Cantliffe, Reference Maynard and Cantliffe1989; Yokomi et al., Reference Yokomi, Hoelmer and Osborne1990). The symptoms are due to physiological changes in the plant induced through feeding by the B biotype and not to the transmission of any plant virus (Yokomi et al., Reference Yokomi, Hoelmer and Osborne1990; Cohen et al., Reference Cohen, Duffus and Liu1992; Jimenez et al., Reference Jimenez, Yokomi, Mayer and Shapiro1995). The changes are irreversible; but once whiteflies are removed, the subsequent leaves and fruit develop normally (Costa et al., Reference Costa, Ullman, Johnson and Tabashnik1993b). Since the discovery of the capacity of the B biotype to induce silverleafing in squash, two additional genetic groups of B. tabaci have been found to induce these symptoms, the MS biotype (Delatte et al., Reference Delatte, Reynaud, Granier, Thornary, Lett, Goldbach and Peterschmitt2005) and the Ug6 group of B. tabaci in Uganda (Sseruwagi et al., Reference Sseruwagi, Legg, Maruthi, Colvin, Rey and Brown2005). All three genetic groups belong to the cluster of B. tabaci that originated in Africa, the Mediterranean Basin, Middle East and Asia Minor, the only exception being the genetic group to which biotype Q belongs which is unable to produce the symptoms (Sseruwagi et al., Reference Sseruwagi, Legg, Maruthi, Colvin, Rey and Brown2005).
The capacity of B biotype to induce squash silverleafing was first described by Yokomi et al. (Reference Yokomi, Hoelmer and Osborne1990) and became one of the means by which this biotype was characterized (Costa & Brown, Reference Costa and Brown1991; Perring et al., Reference Perring, Cooper, Rodriguez, Farrar and Bellows1993; Bedford et al., Reference Bedford, Briddon, Brown, Rosell and Markham1994). The symptom is caused by the presence of large intercellular air spaces between deformed mesophyll palisade cells and the adaxial epidermis (Jimenez et al., Reference Jimenez, Yokomi, Mayer and Shapiro1995). As few as three nymphs are able to induce the change which is systemic with symptom severity proportional to the number of nymphs feeding (Schuster et al., Reference Schuster, Mueller, Kring and Price1990; Yokomi et al., Reference Yokomi, Hoelmer and Osborne1990; Hoelmer et al., Reference Hoelmer, Osborne and Yokomi1991; Costa et al., Reference Costa, Ullman, Johnson and Tabashnik1993a). Costa et al. (Reference Costa, Ullman, Johnson and Tabashnik1993a) and van de Ven et al. (Reference van de Ven, LeVesque, Perring and Walling2000) found that adults feeding for up to 10 days on squash seedlings were unable to induce the silverleaf symptom. Furthermore, van de Ven et al. (Reference van de Ven, LeVesque, Perring and Walling2000) found that over a 5-day period genes associated with B biotype feeding on squash were only induced in response to the feeding by nymphs and not by adults. One study, Bharathan et al. (Reference Bharathan, Narayanan and McMillan1992), reported silverleafing as a consequence of 48 h of adult feeding. However, the experimental design is such that this may well have been due to nymphs and, as such, the results are inconclusive.
It has now become accepted that feeding by nymphs is the only means to induce this symptom. However, observations of squash, zucchini and pumpkin plants growing in Bundaberg, Queensland, Australia revealed silverleafing symptoms in the youngest leaves when large numbers of adults, but no nymphs, were found on the plants. While it was possible that some nymphs may have been missed or that they had since been dislodged or consumed by predators, it was equally possible that adult feeding was also able to induce silverleafing, and the following study was undertaken to determine whether this was possible.
Material and methods
Costa et al. (Reference Costa, Ullman, Johnson and Tabashnik1993a) concluded that adults were unable to induce silverleafing, based on a 10-day period of exposure, while van de Ven et al. (Reference van de Ven, LeVesque, Perring and Walling2000) observed no symptoms after five days of adult feeding. Given our field observations of silverleafing symptoms in the presence of large numbers of adults, but apparent absence of nymphs, it is possible that either the exposure periods of these two studies was too short or the numbers of insects used too few to assess accurately the relationship between adult feeding and silverleaf induction. It was, therefore, decided to expose plants to adult feeding for up to 15 days using different numbers of adult insects. However, as B. tabaci eggs take eight days to hatch at temperatures ranging from 24 to 30°C, it would not be possible to exclude the presence of nymphs for a 15-day period if mixed populations of adults were used. It was, therefore, decided to use adult males only.
Bemisia tabaci biotype B, obtained from a range of different crops and supplemented with new field-collected material every year since 1995, was maintained on Hibiscus rosa-sinesis L. (var. Mrs Gordon) in a glasshouse (24–30°C) under natural light. The species is haplo-diploid; so, to obtain male progeny, newly emerged unmated females were obtained by taking hibiscus leaves with 300–400 red-eyed pupae (late 4th instar) and ensuring all adults were removed before placing them in 3 L clear plastic containers at 08:30. Leaves were then examined at 09:30, and newly emerged adults were removed individually using a pooter and sexed with the aid of a microscope. Males were discarded and the females placed onto whitefly-free hibiscus plants and caged securely to exclude contamination. Cages were left undisturbed for 25 days, at which time leaves containing red-eyed pupae were removed and placed into 3 L containers for adults to emerge. A sub-sample of 50 adults was first taken and examined using a stereomicroscope; all were male. The capacity of the B biotype to induce silverleafing is not dependant on prior host experience, but rather is innate to this genetic group (Costa & Brown, Reference Costa and Brown1991; Li & Hu, Reference Li and Hu2005); and, as such, it is not necessary to maintain insects on squash plants prior to experimentation.
The test plants were squash seedlings (Cucurbita pepo L. var. Sunburst) at the one-leaf stage grown in commercial potting mix and fertilized using the slow release fertiliser Osmocote™. A total of 30 adult males were placed into each of six cages, 60 each into a further six cages and 90 and 120 males into a further six cages each; each cage contained a single squash seedling. An additional six cages contained squash plants but no whiteflies as the untreated control and six cages with a mix of 20 male and 100 female adults as the positive control. All cages were maintained in the glasshouse under natural light at 24–30°C. Plants were observed for silverleafing symptoms prior to introducing adults and then daily until day 15, at which time each plant was carefully searched for the presence of eggs or nymphs. Diagnosis of symptoms was made using the scale described in Hoelmer et al. (Reference Hoelmer, Osborne and Yokomi1991). The relationship between the number of males and the onset of symptoms was calculated using linear regression.
Results
During the 15 days of the study, the squash seedlings grew from the initial one true leaf to four true leaves. The negative whitefly control remained free of silverleaf symptoms throughout the course of the study (fig. 1). The positive whitefly control first showed symptoms at day 10; eggs were observed to have begun to hatch on day 8. None of plants exposed to 30 males showed symptoms while only two of the six plants exposed to 60 males expressed symptoms at day 15 only. Five of the six plants exposed to 90 males first expressed symptoms on day 13 with the sixth becoming symptomatic by day 15. Two of the six plants exposed to 120 males showed symptoms at day 11 and all were expressing symptoms by day 14. None of the plants included in the male-only treatments were found to have eggs or nymphs present at day 15. All six plants exposed to both male and female adults had nymphs (2nd and 3rd instars) present at day 15. Plants exposed to both males and females and those exposed to 120 males produced symptoms that began as veinlets surrounding areoles being slightly lighter than the surrounding tissue, progressing to veinlets becoming distinctly lighter than the surroundings and presenting a characteristic netted appearance, and finally as bleaching of the primary veins. Plants exposed to 90 males progressed only as far as the distinct netted appearance while plants exposed to 60 males showed only the earliest signs of veinlet bleaching. The silverleaf response in plants exposed to males and females was not only faster, but also less gradual as all six plants showed the first symptoms on the same day rather than over several days as in the case of the other treatments. There was also a positive density response (r2=0.60, P=0.0019) in symptom expression, with plants exposed to 120 adult males showing symptoms two days earlier than those exposed to 90 males and four days earlier than those exposed to 60 males, while plants exposed to 30 males showed no symptoms during the course of the experiment.
Fig. 1. The number of squash plants showing silverleaf symptoms over an 18-day period. Eggs began to hatch on day 8 for the treatment 100♀ 20♂. All plants in the remaining five treatments were free of eggs and nymphs at day 15 (—◆—, no adults; —■—, 100♀ 20♂ adults; —▲—, 30♂ adults; —×—, 60♂ adults; —□—, 90♂ adults; —△—, 120♂ adults).
Discussion
Adult, or at least male B. tabaci biotype B, are able to induce the symptoms known as squash silverleaf. The symptoms progressed as far as bleaching of the primary veins before the experiment ended. Symptom expression, excluding the treatment containing both male and female adults, took a minimum of 11 days with 120 males and a maximum of 15 days with 60 males. This would explain why Costa et al. (Reference Costa, Ullman, Johnson and Tabashnik1993a) and van de Ven (Reference van de Ven, LeVesque, Perring and Walling2000) both failed to observe symptoms as they left adults to feed for only 10 and 3–5 days, respectively, with either 20 or 120 adults, respectively. Similarly, it may explain why van de Ven (Reference van de Ven, LeVesque, Perring and Walling2000) observed no activation of the genes SLW1 or SLW3 in response to adult feeding on squash. The treatment exposed to both male and female adults produced symptoms a day earlier than those exposed to 120 males, presumably due to the presence of feeding nymphs.
The time to onset of symptom expression decreased as the numbers of males increased, suggesting the existence of a positive density response in symptom expression. The density dependant nature of the response is similar to that observed with the feeding by nymphs (Schuster et al., Reference Schuster, Mueller, Kring and Price1990; Yokomi et al., Reference Yokomi, Hoelmer and Osborne1990; Hoelmer et al., Reference Hoelmer, Osborne and Yokomi1991; Costa et al., Reference Costa, Ullman, Johnson and Tabashnik1993a). The numbers of adults required to induce silverleafing in squash is many more than the three nymphs per plant reported by Costa et al. (Reference Costa, Ullman, Johnson and Tabashnik1993a) and suggests that there is a fundamental difference in either the way nymphs feed or in the saliva of nymphs verses that of adult whiteflies. The results explain why plants displaying silverleaf symptoms in the youngest leaves can do so without nymphs being present. This study is the first in a series planned to identify the underlying biological differences between different genetic groups of B. tabaci.
Acknowledgements
This study was funded through Horticulture Australia.
