Cucurbit aphid‐borne yellows virus (CABYV) modifies the alighting,settling and probing behaviour of its vector Aphis gossypii favouring its own spread

Plant pathogens are able to influence the behaviour and fitness of their vectors in such a way that changes in plant–pathogen–vector interactions can affect their transmission. Such influence can be direct or indirect, depending on whether it is mediated by the presence of the pathogen in the vector's body or by host changes as a consequence of pathogen infection. We report the effect that the persistently aphid‐transmitted Cucurbit aphid‐borne yellows virus (CABYV, Polerovirus) can induce on the alighting, settling and probing behaviour activities of its vector, the cotton aphid Aphis gossypii. Only minor direct changes on aphid feeding behaviour were observed when viruliferous aphids fed on non‐infected plants. However, the feeding behaviour of non‐viruliferous aphids was very different on CABYV‐infected than on non‐infected plants. Non‐viruliferous aphids spent longer time feeding from the phloem in CABYV‐infected plants compared to non‐infected plants, suggesting that CABYV indirectly manipulates aphid feeding behaviour through its shared host plant in order to favour viral acquisition. Viruliferous aphids showed a clear preference for non‐infected over CABYV‐infected plants at short and long time, while such behaviour was not observed for non‐viruliferous aphids. Overall, our results indicate that CABYV induces changes in its host plant that modifies aphid feeding behaviour in a way that virus acquisition from infected plants is enhanced. Once the aphids become viruliferous they prefer to settle on healthy plants, leading to optimise the transmission and spread of this phloem‐limited virus.     

Effects of tomato spotted wilt virus isolates, host plants, and temperature on survival, size, and development time of Frankliniella occidentalis

Tomato spotted wilt virus (TSWV) replicates in both its plant hosts and its thrips vectors. Replication of TSWV within thrips suggests the potential for pathological effects that could affect the fitness of its vectors directly, whereas infection of the plant may alter its suitability as a host for thrips development. This study was undertaken to examine the influence of TSWV isolate, host plant, and temperature on potential direct and host-mediated effects of virus infection of the thrips and the plant on Frankliniella occidentalis (Pergande) (Thysanoptera: Thripidae), an important vector of TSWV. Neonate F. occidentalis were reared to adult eclosion on excised foliage of Datura stramonium (L.) (Solanaceae) or Emilia sonchifolia (L.) (Compositae) infected with either the CFL or RG2 isolate of TSWV, or not infected. Effects of the TSWV isolates and host plants on thrips were measured at 18.3, 23.9, and 29.4 °C. Results demonstrate significantly improved survival and a small but significant decrease in development time of F. occidentalis on TSWV-infected plants. These effects resulted from the combined influence of the direct effects of the virus on infected thrips and plant-mediated effects resulting from virus infection of the thrips’ host plant. Our results extend previous findings and help to explain inconsistencies among previously published reports by demonstrating that the manifestation and magnitude of effects of TSWV on F. occidentalis are dependent on host plant, virus isolate, and temperature.     

Vector within‐host feeding preference mediates transmission of a heterogeneously distributed pathogen

1. Ecological theory predicts that vector preference for certain host species or discrimination between infected versus uninfected hosts impacts disease incidence. However, little information exists on the extent to which vector within‐host feeding preference mediates transmission. This may be particularly important for plant pathogens, such as sharpshooter transmission of the bacterium Xylella fastidiosa, which are distributed irregularly throughout hosts. 2. We documented the within‐host distribution of two vector species that differ in transmission efficiency, the leafhoppers Draeculacephala minerva and Graphocephala atropunctata, and which are free to move throughout entirely caged alfalfa plants. The more efficient vector D. minerva fed preferentially at the base of the plant near the soil surface, whereas the less efficient G. atropunctata preferred overwhelming the top of the plant. 3. Next we documented X. fastidiosa heterogeneity in mechanically inoculated plants. Infection rates were up to 50% higher and mean bacterial population densities were 100‐fold higher near the plant base than at the top or in the taproot. 4. Finally, we estimated transmission efficiency of the two leafhoppers when they were confined at either the base or top of inoculated alfalfa plants. Both vectors were inefficient when confined at the top of infected plants and were 20–60% more efficient when confined at the plant base. 5. These results show that vector transmission efficiency is determined by the interaction between leafhopper within‐plant feeding behaviour and pathogen within‐plant distribution. Fine‐scale vector and pathogen overlap is likely to be a requirement generally for efficient transmission of vector‐borne pathogens.     

Patterns of virulence and benevolence in insect‐borne pathogens of plants

Direct and indirect interactions between insect‐borne pathogens and their host plants are reviewed in the context of theoretical analyses of the evolution of virulence. Unlike earlier theories, which maintained that parasites should evolve to be harmless or even beneficial to their hosts, recent models predict that coevolution between pathogen and host may lead to virulence or avirulence, depending on the pathogen transmission system. The studies reviewed here support the hypothesis that virulence can be advantageous for insect‐borne pathogens of plants. Virulent pathogens may be transmitted more readily by vector insects and are likely to induce stronger disease symptoms, thereby potentially making the plant more attractive to vectors. In contrast, the transmission advantage of virulence for seed‐transmitted pathogens is lower and the costs of virulence are high. Pathogens may sometimes benefit plants via indirect interactions that arise through relationships with other organisms. Evidence for the effects of insect‐borne pathogens on plant competition, herbivory, and parasitism also is reviewed, but few studies have measured the outcome of both direct and indirect interactions. Benefits of pathogen infection that accrue to plants from indirect interactions may sometimes outweigh the direct detrimental effects of virulence.     

Herbivore benefits from vectoring plant virus through reduction of period of vulnerability to predation

Herbivores can profit from vectoring plant pathogens because the induced defence of plants against pathogens sometimes interferes with the induced defence of plants against herbivores. Plants can also defend themselves indirectly by the action of the natural enemies of the herbivores. It is unknown whether the defence against pathogens induced in the plant also interferes with the indirect defence against herbivores mediated via the third trophic level. We previously showed that infection of plants with Tomato spotted wilt virus (TSWV) increased the developmental rate of and juvenile survival of its vector, the thrips Frankliniella occidentalis. Here, we present the results of a study on the effects of TSWV infections of plants on the effectiveness of three species of natural enemies of F. occidentalis: the predatory mites Neoseiulus cucumeris and Iphiseius degenerans, and the predatory bug Orius laevigatus. The growth rate of thrips larvae was positively affected by the presence of virus in the host plant. Because large larvae are invulnerable to predation by the two species of predatory mites, this resulted in a shorter period of vulnerability to predation for thrips that developed on plants with virus than thrips developing on uninfected plants (4.4 vs. 7.9 days, respectively). Because large thrips larvae are not invulnerable to predation by the predatory bug Orius laevigatus, infection of the plant did not affect the predation risk of thrips larvae from this predator. This is the first demonstration of a negative effect of a plant pathogen on the predation risk of its vector.     

Pathogen effects on vegetative and floral odours mediate vector attraction and host exposure in a complex pathosystem

Pathogens can alter host phenotypes in ways that influence interactions between hosts and other organisms, including insect disease vectors. Such effects have implications for pathogen transmission, as well as host exposure to secondary pathogens, but are not well studied in natural systems, particularly for plant pathogens. Here, we report that the beetle‐transmitted bacterial pathogen Erwinia tracheiphila – which causes a fatal wilt disease – alters the foliar and floral volatile emissions of its host (wild gourd, Cucurbita pepo ssp. texana) in ways that enhance both vector recruitment to infected plants and subsequent dispersal to healthy plants. Moreover, infection by Zucchini yellow mosaic virus (ZYMV), which also occurs at our study sites, reduces floral volatile emissions in a manner that discourages beetle recruitment and therefore likely reduces the exposure of virus‐infected plants to the lethal bacterial pathogen – a finding consistent with our previous observation of dramatically reduced wilt disease incidence in ZYMV‐infected plants.     

Aphid behavioral responses to virus‐infected plants are similar despite divergent fitness effects

We compared the settling preferences and reproductive potential of an oligophagous herbivore, the pea aphid, Acyrthosiphon pisum Harris (Hemiptera: Aphididae), in response to pea plants, Pisum sativum L. cv. ‘Aragorn’ (Fabaceae), infected with two persistently transmitted viruses, Pea enation mosaic virus (PEMV) and Bean leaf roll virus (BLRV), that differ in their distribution within an infected plant. Aphids preferentially oriented toward and settled on plants infected with PEMV or BLRV in comparison with sham‐inoculated plants (plants exposed to herbivory by uninfected aphids), but aphids did not discriminate between plants infected with the two viruses. Analysis of plant volatiles indicated that plants inoculated with either virus had significantly higher green leaf volatile‐to‐monoterpene ratios. Time until reproductive maturity was marginally influenced by plant infection status, with a trend toward earlier nymph production on infected plants. There were consistent age‐specific effects of plant infection status on aphid fecundity: reproduction was significantly enhanced for aphids on BLRV‐infected plants across most time intervals, though mean aphid fecundity did not differ between sham and PEMV‐infected plants. There was no clear pattern of age‐specific survivorship; however, mean aphid lifespan was reduced on plants infected with PEMV. Our results are consistent with predictions of the host manipulation hypothesis, extended to include plant viruses: non‐viruliferous A. pisum preferentially orient to virus‐infected host plants, potentially facilitating pathogen transmission. These studies extend the scope of the host manipulation hypothesis by demonstrating that divergent fitness effects on vectors arise relative to the mode of virus transmission.     

Effects of tomato spotted wilt virus (TSWV) isolates, host plants, and temperature on survival, size, and development time of Frankliniella fusca

The effects of different isolates of the tomato spotted wilt tospovirus (TSWV), host plants, and temperatures on Frankliniella fusca (Hinds) (Thysanoptera: Thripidae), the most important vector of TSWV in North Carolina, were measured in the laboratory. Thrips were reared at either 18.3, 23.9, or 29.4 °C until adult eclosion on excised leaves of Datura stramonium L. or Emilia sonchifolia (L.). Plants were either infected with the TSWV isolates CFL or RG2, or left uninfected (control). The results revealed a positive relationship between larval survival and temperature, regardless of host plant or TSWV isolate. Both survival to adult and percentage transmission of TSWV by F. fusca were significantly affected by the interaction between host plant and TSWV isolate. The consequence of this interaction was that the cohort‐based percentage transmission from infected E. sonchifolia plants for CFL was 1.3‐fold greater than that of RG2, whereas the percentage transmission from infected D. stramonium plants for RG2 was twice that of CFL. Both host plant and TSWV isolates showed significant effects on thrips development time to adult and head capsule width of adult thrips, as well as on the incidence of thrips infection with TSWV. The infection status of these thrips was determined by ELISA for the NSs viral protein. Infected thrips reared on infected host foliage took longer to develop to adult and were smaller than non‐infected thrips which had also been reared on infected host foliage, demonstrating a direct effect of the TSWV on thrips. However, non‐infected thrips reared on non‐infected leaves took longer to develop than non‐infected thrips reared on infected leaves, suggesting an effect of the plant tissue on thrips. In addition, adult thrips reared on TSWV‐infected D. stramonium at 29.4 °C developed smaller head capsules than thrips developing on infected foliage at lower temperatures and on non‐infected leaves of D. stramonium or E. sonchifolia. Both TSWV isolates and host plants differentially affected females more than males. In conclusion, both the infection of thrips by TSWV and TSWV‐mediated changes in host plant quality were found to have significant biological effects on F. fusca.     

Virus infection alters the predatory behavior of an omnivorous vector

The interactions between parasites and their hosts can cause profound changes in host behavior, including changes that can alter other trophic interactions. The western flower thrips Frankliniella occidentalis is an important omnivorous insect vector of Tomato spotted wilt virus (TSWV), which infects crops worldwide and also infects its thrips vector. Here, we show that tospovirus‐infected female thrips become more predaceous, illustrating how the functional role of omnivores may change in response to pathogen infection. Our findings support the hypothesis that increased predation among virus‐infected female thrips compensates for the detrimental effects of virus infection. Because predatory behavior is unlikely to increase virus transmission to plants, it is doubtful that this shift in feeding behavior is due to an adaptive parasite manipulation of vector behavior. In this study, increases in predatory behavior were observed in female thrips, but not in male thrips. This sexually dimorphic compensatory response indicates that male and female thrips utilize different feeding strategies to compensate for parasite infection, the expression of which is constrained by resource availability. Our findings demonstrate a novel, but potentially common pathway by which viruses can influence the structure of trophic interactions in food webs.     

Effect of Watermelon Silver Mottle Virus on the Life History and Feeding Preference of Thrips palmi

Thrips-borne tospoviruses cause numerous plant diseases that produce severe economic losses worldwide. In the disease system, thrips not only damage plants through feeding but also transmit causative agents of epidemics. In addition, thrips are infected with tospoviruses in the course of virus transmission. Most studies on the effect of tospoviruses on vector thrips have focused on the Tomato spotted wilt virus–Frankliniella occidentalis system. Thus, we focused on another thrips-borne tospovirus, Watermelon silver mottle virus (WSMoV), to examine the effect of virus infection on its vector, Thrips palmi. In this study, the direct and indirect effects of WSMoV on the life history traits and feeding preference of T. palmi were examined. The survival rate and developmental time of the WSMoV-infected larval thrips did not differ significantly from those of the virus-free thrips. Comparing the developmental time of larval thrips fed on the healthy plants, thrips-damaged plants, and thrips-inoculated plants (the WSMoV-infected plants caused by thrips feeding), feeding on the thrips-damaged plants reduced the developmental time, and the WSMoV infection in host plants partially canceled the effect of thrips damage on the developmental time. In addition, no significant variations between the virus-free and WSMoV-infected adult thrips regarding longevity and fecundity were observed. These results implied that WSMoV did not directly affect the life history traits of T. palmi, but the WSMoV infection indirectly affected the development of T. palmi through the virus-infected plants. Furthermore, feeding preference tests indicated that T. palmi preferred feeding on either the thrips-damaged plants or the thrips-inoculated plants to the healthy plants. The effect of tospoviruses on the life history and feeding preference of vector thrips might vary among host plants, virus species, vector species, and environmental factors.     

Vector and virus induce plant responses that benefit a non-vector herbivore

The negative cross-talk between induced plant defences against pathogens and arthropod herbivores is exploited by vectors of plant pathogens: a plant challenged by pathogens reduces investment in defences that would otherwise be elicited by herbivores. This negative cross-talk may also be exploited by non-vector herbivores which elicit similar anti-herbivore defences in the plant. We studied how damage by the thrips Frankliniella occidentalis and/or infection with Tomato spotted wilt virus (TSWV) affect the performance of a non-vector arthropod: the two-spotted spider mite Tetranychus urticae, a parenchym feeder just like F. occidentalis. Juvenile survival of spider mites on plants inoculated with TSWV by thrips was higher than on control and on thrips-damaged plants. However, thrips damage did not reduce spider-mite survival as compared to the control, suggesting that the positive effect of TSWV on spider-mite survival is independent of anti-thrips defence. Developmental and oviposition rates were enhanced on plants inoculated with TSWV by thrips and on plants with thrips damage. Therefore, spider mites benefit from TSWV-infection of pepper plants, but also from the response of plants to thrips damage. We suggest that the positive effects of TSWV on this non-vector species cannot be explained exclusively by cross-talk between anti-herbivore and anti-pathogen plant defences.     

Preference of the vector thrips Frankliniella occidentalis for plants infected with thrips‐non‐transmissible Tomato spotted wilt virus

The effect of a thrips‐non‐transmissible Tomato spotted wilt virus (TSWV) on insect–host interactions between thrips and Arabidopsis thaliana was analysed. A wild‐type TSWV virulent isolate and a TSWV isolate that induces mild symptoms on inoculated plants (TSWV‐Mo) were used in this study, and TSWV‐Mo isolate was obtained by single local lesion isolation using Petunia x hybrid after several passages on Nicotiana rustica plants. In transmission test, although wild‐type TSWV (TSWV‐wt) was transmitted by two thrips species (transmission ratio; Frankliniella occidentalis, 25%; Thrips tabaci, 10%; and T. palmi, 0%), none of the thrips transmitted TSWV‐Mo. Feeding damage by F. occidentalis in A. thaliana plants was more extensive on TSWV‐wt‐infected plants than on TSWV‐Mo‐infected plants, despite comparable preference. Among the markers of plant defences, salicylic acid‐regulated genes were upregulated threefold to sixfold by TSWV‐wt or TSWV‐Mo infection. In contrast, jasmonate‐regulated genes and jasmonate/ethylene‐regulated genes were not affected by the infections. Pull assays showed that adjacent TSWV‐Mo‐infected plants were preferred over uninfected plants. In conclusion, our results showed that the transmissibility by thrips of TSWV is not related to preference of vector thrips and suggested that TSWV‐Mo‐infected plants may be used as attractants for behaviour control of thrips.     

A negative effect of a pathogen on its vector? A plant pathogen increases the vulnerability of its vector to attack by natural enemies

Plant pathogens that are dependent on arthropod vectors for transmission from host to host may enhance their own success by promoting vector survival and/or performance. The effect of pathogens on vectors may be direct or indirect, with indirect effects mediated by increases in host quality or reductions in the vulnerability of vectors to natural enemies. We investigated whether the bird cherry-oat aphid Rhopalosiphum padi, a vector of cereal yellow dwarf virus (CYDV) in wheat, experiences a reduction in rates of attack by the parasitoid wasp Aphidius colemani when actively harboring the plant pathogen. We manipulated the vector status of aphids (virus carrying or virus free) and evaluated the impact on the rate of attack by wasps. We found that vector status did not influence the survival or fecundity of aphids in the absence of parasitoids. However, virus-carrying aphids experienced higher rates of parasitism and greater overall population suppression by parasitoid wasps than virus-free aphids. Moreover, virus-carrying aphids were accepted as hosts by wasps more often than virus-free aphids, with a greater number of wasps stinging virus-carrying aphids following assessment by antennal palpations than virus-free aphids. Therefore, counter to the prevailing idea that persistent vector-borne pathogens enhance the performance of their vectors, we found that infectious aphids actively carrying a plant pathogen experience greater vulnerability to natural enemies. Our results suggest that parasitoids may contribute to the successful biological control of CYDV by disproportionately impacting virus-carrying vectors, and thus reducing the proportion of vectors in the population that are infectious.     

Relationship between onion thrips (Thrips tabaci) and Stemphylium vesicarium in the development of Stemphylium leaf blight in onion

Stemphylium leaf blight caused by Stemphylium vesicarium and onion thrips (Thrips tabaci) are two common causes of leaf damage in onion production. Onion thrips is known to interact synergistically with pathogens to exacerbate plant disease. However, the potential relationship between onion thrips and Stemphylium leaf blight is unknown. In a series of controlled laboratory and field trials, the relationship between thrips feeding and movement on the development and severity of Stemphylium leaf blight were examined. In laboratory assays, onions (“Avalon” and “Ailsa Craig”) with varying levels of thrips feeding damage were inoculated with S. vesicarium. Pathogen colonisation and leaf dieback were measured after 2 weeks. In pathogen transfer assays, thrips were exposed to S. vesicarium conidia, transferred to onion and leaf disease development was monitored. In field trials, insecticide use was examined as a potential indirect means to reduce Stemphylium leaf blight disease and pathogen colonisation by reducing thrips damage. Results from laboratory trials revealed that a reduction in thrips feeding decreased S. vesicarium colonisation of onion leaves by 2.3–2.9 times, and decreased leaf dieback by 40–50%. Additionally, onion thrips were capable of transferring S. vesicarium conidia to onion plants (albeit at a low frequency of 2–14% of plants inoculated). In field trials, the symptoms and colonisation of Stemphylium leaf blight were reduced by 27 and 17%, respectively with the use of insecticide to control thrips. These results suggest that onion thrips may play a significant role in the development of Stemphylium leaf blight, and thrips control may reduce disease in commercial onion fields.     

A removable virus vector suitable for plant genome editing

Plant genome editing is achieved by the expression of sequence‐specific nucleases (SSNs). RNA virus vector‐mediated expression of SSNs is a promising approach for transgene integration‐free targeted mutagenesis in plants. However, the removal of virus vectors from infected plants is challenging because no antiviral drugs are available against plant viruses. Here, we developed a removable RNA virus vector that carries the target site of tobacco microRNA398 (miR398) whose expression is induced during shoot regeneration. In the inoculated leaves in which expression of miR398 is not induced, insertion of the miR398 target site did not affect the practicability of the virus vector. When shoots were regenerated from the infected leaves, miR398 was expressed and viral RNA was eliminated. The virus vector successfully expressed SSNs in inoculated leaves, from which virus‐free genome‐edited plants were regenerated via tissue culture.     

Influence of a propagative plant virus on the fitness and wing dimorphism of infected and exposed insect vectors

To maximize fitness, plant pathogenic viruses may manipulate their arthropod vectors through direct and indirect (via the host plant) interactions. For many virus-vector-plant associations, insect feeding does not always lead to virus acquisition. In fact, many plant viruses, especially those that propagate into their vectors, are acquired at low rates. Although the majority of insects colonizing an infected plant escape from viral infection, they are still exposed to the indirect effects (i.e. the effect of plant metabolism modification following virus infection). Little information has been reported on the effects of plant viruses on insects that become infected versus those that do not (here referred to as “exposed”). The effect that the Maize mosaic virus (MMV) (Rhabdoviridae) exerts on the fitness and wing dimorphism of the planthopper vector, Peregrinus maidis (Hemiptera, Delphacidae), that developed on leaves from either young or old corn plants was examined. MMV exerted non-consistent to minimal direct effects on developmental time, longevity, nymphal mortality and fecundity. In addition, some small yet significant fitness costs were encountered by exposed planthoppers to escape MMV infection. Furthermore, a significantly higher proportion of macropters over brachypters were produced on MMV-infected old leaves compared with healthy leaves of a similar age. We conclude that the virus influences the dispersal of the vector, promoting a larger production of macropters at the costs of brachypters at a late stage of the plant infection. Because MMV infection in planthoppers did not segregate by wing morphotype, our results indicate that the dispersal of both infected and exposed planthoppers was a likely consequence of the indirect effects of MMV.     

Settling preferences of the whitefly vector Bemisia tabaci on infected plants varies with virus family and transmission mode

Virus infection may change not only the host‐plant phenotypic (morphological and physiological) characteristics, but can also modify the behavior of their insect vector in a mutualistic or rather antagonistic manner, to promote their spread to new hosts. Viruses differ in their modes of transmission and depend on vector behavior for successful spread. Here, we investigated the effects of the semi‐persistently transmitted Tomato chlorosis virus (ToCV, Crinivirus) and the persistent circulative Tomato severe rugose virus (ToSRV, Begomovirus) on alighting preferences and arrestment behavior of their whitefly vector Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) Middle East Asia Minor 1 (MEAM1) on tomato plants (Solanum lycopersicum L. cv. Santa Clara, Solanaceae). The vector alighting preferences between infected and uninfected plants in choice assays were apparently influenced by the presence of ToCV and ToSRV in the whiteflies or by their previous exposure to infected plants. The observed changes in vector behavior do not seem to benefit the spread of ToCV: non‐viruliferous insects clearly preferred mock‐inoculated plants, whereas ToCV‐viruliferous insects landed on mock‐inoculated and ToCV‐infected plants, indicating a partial change in insect behavior – ToCV was able to directly affect the preference of its vector B. tabaci, but this change in insect behavior did not affect the virus spread because viruliferous insects landed on mock‐inoculated and infected plants indistinctly. In contrast, ToSRV‐viruliferous insects preferred to land on mock‐inoculated plants, a behavior that increases the probability of spread to new host plants. In the arresting behavior assay, the majority of the insects remained on mock‐inoculated plants when released on them. A greater number of insects moved toward mock‐inoculated plants when initially released on ToCV‐ or ToSRV‐infected plants, suggesting that these viruses may repel or reduce the nutritional quality of the host plants for B. tabaci MEAM1.     

The influence of primary infection date and establishment of vector populations on the spread of yellowing viruses in sugar beet

In three field experiments in 1985 and 1986, we studied the effect of the date of primary infection on the spread of beet yellows closterovirus (BYV) and beet mild yellowing luteovirus (BMW) from artificially inoculated sugar beet plants. Laboratory-reared vector aphids, Myzus persicae, were placed on these sources of virus. There was no substantial natural immigration of vectors or viruses. In two experiments, one with BMYV in 1985 and the other in BYV in 1986, populations of vector aphids remained low and there was little virus spread, i.e. c. 50 infected plants from one primarily infected source. The cause of this small amount of spread was the low number of vector aphids. In the third experiment, with BYV in 1986, large populations of M. persicae developed and there was substantial virus spread: c. 2000 infected plants in the plots which were inoculated before canopy closure. In later-inoculated plots in the same experiment, there was much less spread: c. 100 infected plants per virus source plant. Differences between fields in predator impact are implicated as the most probable factor causing differences in vector establishment and virus spread between these three experiments. Virus spread decreased with later inoculation in all three experiments. A mathematical model of virus spread incorporating results from our work has been used to calculate how the initial proportion of infected plants in a crop affects the final virus incidence. This model takes into account the effect of predation on the development of the aphid populations. The processes underlying the spread and its timing are discussed.     

Drought stress affects response of phytopathogen vectors and their parasitoids to infection‐ and damage‐induced plant volatile cues

1. The response of a phytopathogen vector to pathogen‐induced plant volatiles was investigated, as well as the response of the phytopathogen vector's parasitoid to herbivore‐induced plant volatiles released from plants with and without drought stress. 2. These experiments were performed with Asian citrus psyllid (Diaphorina citri), vector of the plant pathogen Candidatus Liberibacter asiaticus (CLas) and its parasitoid Tamarixia radiata as models. Candidatus Liberibacter asiaticus is the presumed causal pathogen of huanglongbing (HLB), also called citrus greening disease. 3. Diaphorina citri vectors were attracted to headspace volatiles of CLas‐infected citrus plants at 95% of their water‐holding capacity (WHC); such attraction to infected plants was much lower under drought stress. Attraction of the vector to infected and non‐stressed plants was correlated with greater release of methyl salicylate (MeSA) as compared with uninfected and non‐stressed control citrus plants. Drought stress decreased MeSA release from CLas‐infected plants as compared with non‐stressed and infected plants. 4. Similarly, T. radiata was attracted to headspace volatiles released from D. citri‐infested citrus plants at 95% of their WHC. However, wasps did not show preference between headspace volatiles of psyllid‐infested and uninfested plants when they were at 35% WHC, suggesting that herbivore‐induced defences did not activate to recruit this natural enemy under drought stress. 5. Our results demonstrate that herbivore‐ and pathogen‐induced responses are environmentally dependent and do not occur systematically following damage. Drought stress affected both pathogen‐ and herbivore‐induced plant volatile release, resulting in concomitant decreases in behavioural response of both the pathogen's vector and the vector's primary parasitoid.