- Top of page
- 1 INTRODUCTION
- 2 METHODS
- 3 RESULTS
- 4 DISCUSSION
The recently discovered FRC strain of the peach potato aphid Myzus persicae (Sulzer) exhibits a resistance factor (RF) of 1679 to the neonicotinoid imidacloprid and 225 to thiamethoxam in topical bioassays. Resistance in this population appears to be due to a combination of two separate mechanisms, a target-site mutation within the nicotinic acetylcholine receptor (nAChR) and enhanced detoxification of the insecticide by cytochrome P450 monooxygenases. While a mixed population and not a clone, all FRC individuals carry the same mutation within the nAChR and exhibit the same levels of neonicotinoid resistance, hence for the purposes of this study they will be treated as a clone. Neonicotinoid resistance is also achieved by the upregulation of P450 genes in the Greek M. persicae clone 5191A. 5191A, however, exhibits a 60-fold RF to imidacloprid in topical bioassays hence this clone's resistance, unlike that of FRC, is only detectable within the laboratory. Myzus persicae is already resistant to three other classes of insecticide by four other distinct mutations. On the basis that resistance to one neonicotinoid chemical generally confers resistance across a whole insecticide class and that the major method of control of this pest is by the application of insecticides the discovery of the FRC resistant strain presents a serious agricultural threat.
Resistance to neonicotinoids in insect pests is commonly achieved by upregulation of metabolic pathways. It has been shown that, compared with susceptible strains, neonicotinoid resistant strains of Bemisia tabaci (Genn.) produce larger amounts of the primary metabolite 5-hydroxy-imidacloprid, which has a lower binding affinity for the nAChR than imidacloprid and would be subjected to secondary metabolism through inactivating conjugation reactions. This upregulated metabolism appears to be due to increased expression of P450s, with monooxygenase activity correlating strongly with neonicotinoid resistance in this species. Other species which show upregulation of P450s correlated with resistance to neonicotinoids include Nilaparvarta lugens (Stal), Drosophila melanogaster (Meigen) and Musca domestica (L.). In terms of neonicotinoid resistance, metabolic changes occur more frequently than target site mutations. Alteration of the nAChR target site has occurred in N. lugens but in this case it arose in an artificially selected laboratory strain. It has also been suggested that target site resistance may occur in the Colorado potato beetle Leptinotarsa decemlineata (Say), but the mutation remains unproven. Other than the M. persicae FRC population there is therefore little evidence of target site based neonicotinoid resistance arising in the field.
Although less commonly cited, behavioural changes may also contribute to insecticide resistance. For example, an imidacloprid resistant strain of Myzus nicotianae Blackman only exhibited such tolerance when the insecticide was applied in an oral ingestion assay rather than topically. This increased tolerance appeared to be linked to an alteration in feeding behaviour shown by the resistant strain compared with a susceptible strain. A resistant strain of the German cockroach Blattella germanica (L.) showed increased aversion to agar containing fructose, glucose, maltose and sucrose, all ingredients commonly contained within commercial gel baits, than a susceptible strain. Topical application of abamectin or fipronil reduced the difference in mortality between resistant and susceptible strains, indicating that this change in feeding behaviour contributed significantly to the resistance.
In theory, dispersal behaviour might also contribute to resistance. An individual which shows an increased tendency to quickly disperse from an insecticide-treated to an untreated plant may ingest a lower dose of insecticide and so be more likely to survive. In this way a resistant strain could potentially arise. This has happened in the Colorado potato beetle Leptinotarsa decemlineata (Say), where a Bt resistant strain shows increased flight activity compared with a susceptible strain upon ingestion of Bacillus thuringiensis (Bt) toxin. Larvae from Bt resistant strains have also been shown to be more behaviourally responsive to the presence of Bt than susceptible larvae, moving away from foliage treated with the toxin. While previous work has linked enhanced dispersal behaviour to increased levels of pyrethroid resistance in two resistant M. persicae clones, this is the first study to investigate the possibility of a behavioural component to neonicotinoid resistance in aphids by comparing the dispersal behaviours of neonicotinoid resistant and susceptible M. persicae clones. This study reports the novel finding of a behavioural component to the high levels of neonicotinoid-resistance exhibited by the M. persicae FRC population.
- Top of page
- 1 INTRODUCTION
- 2 METHODS
- 3 RESULTS
- 4 DISCUSSION
Studies into the behavioural aspects of insecticide resistance in hemiptera have typically focused on differences in feeding behaviour[14, 18, 22, 23] rather than dispersal. In terms of differences in dispersal between resistant and susceptible clones of the same species no studies have yet been published using neonicotinoids. Differences in dispersal behaviour have, however, been shown to exist between M. persicae clones resistant to other insecticides. Clones carrying carboxylesterase resistance exhibit a reduced tendency to disperse from wilting leaves as compared with susceptible clones. Differences in dispersal behaviour in response to parasitoid attack have also been shown between susceptible M. persicae clones and clones carrying either carboxylesterase or kdr resistance, with resistant clones less likely to disperse from leaves after exposure to parasitoids and subsequently more likely to be parasitised.[25, 26] Finally, clones carrying kdr and carboxylesterase resistance have been shown to exhibit a reduced dispersal response to aphid alarm pheromone compared with susceptible clones.[27-29] MACE resistant M. persicae clones, however, appear more responsive to aphid alarm pheromone than non-MACE forms and suffer lower levels of parasitism. MACE resistant and sensitive forms also distribute differently upon the host plant, with larger numbers of MACE forms being found on the growing points of pepper plants. Pyrethroid-resistant M. persicae have been shown to walk off deltamethrin-treated excised potato leaves in greater numbers than a susceptible clone but aphids were only observed for 10 min and were never offered a mix of treated and untreated leaf tissue. Najar-Rodríguez et al. showed differences in landing and settling behaviour of two separate lineages of Aphis gossypii Glover, but the clones differed in preferred host plant species rather than insecticide resistance. Studies showing differences in dispersal between aphid species are more common, but do not link such behaviours with insecticide resistance.[32-34]
Two different dispersal behaviours were noted in 5191A and FRC, the two resistant clones or strains within this study. While 5191A showed an increased tendency to disperse away from all plant tissue regardless of insecticide treatment, the FRC population exhibited an increased ability to locate itself to areas of untreated plant tissue in an environment consisting of mixed insecticide treatment. While these changes in dispersal behaviour should increase the fitness of both clones under thiamethoxam-treated conditions and so contribute significantly to their neonicotinoid resistance, the enhanced ability of the FRC population to locate untreated plant tissue and therefore target its dispersal may be more relevant given the high level of resistance already exhibited by this strain.
The differential dispersal behaviour exhibited by the neonicotinoid resistant FRC population in this study should not be confused with the general repellency effect of some insecticides, which may also cause changes in feeding and dispersal. These effects are often seen and considered beneficial in insect control.[35-38] For example, the large pine weevil Hylobius abietis (L.) will avoid feeding on Scots pine twigs treated with lambda-cyhalothrin but will feed from untreated areas on the same twig[39, 40] while exposure to imidacloprid-infused leaves has been shown to increase the general motility of M. persicae. The study reported here, however, reports the discovery of a dispersal behaviour performed in response to TMX which differs between a resistant and susceptible clone of the same species, thus potentially linking this behaviour to the presence of neonicotinoid resistance in M. persicae.
The dispersal behaviour of the FRC population has significant implications for agricultural practises, particularly when spraying infested crops with neonicotinoids, as it is likely that the FRC, if present, will relocate to any untreated areas within a field. This finding therefore emphasises the importance of ensuring that an insecticide is applied evenly throughout a crop. The modified dispersal behaviour exhibited by the FRC population may also possibly be of benefit in the case of an unsprayed trap crop as there is potential for the aphid to move into this untreated region for destruction by other means.
Alterations in dispersal behaviour could lead to changes in the transmission of aphid-vectored plant viruses. Transmission of plant viruses is the major cause of yield loss and economic damage to crop systems by aphids. Myzus persicae is a known vector of at least 100 plant viruses. Aphids acquire plant viruses during feeding (for a review see Fereres and Moreno). Virions enter and adhere to the aphid stylets or midgut during either short probes of the mouthparts into the plant tissue in the case of non-persistent viruses, or longer sustained phloem feeding in the case of persistent viruses.[48, 49] Transmission occurs when a viruliferous aphid disperses from a viral host plant and relocates to another, healthy, host plant. The inoculation of non-persistent viruses is associated with salivation during brief cell punctures by the stylets[50, 51] while that of persistent viruses is thought to occur during phloem salivation. Previous studies have often shown the application of insecticides to be beneficial in reducing the spread of plant viruses.[20, 52-55] From this study it appears that M. persicae disperses at the same rate regardless of the presence or absence of neonicotinoid treatment, hence in theory, the application of neonicotinoids should exert either a beneficial (due to increased vector mortality) or else no effect on the rates at which plant viruses spread within a crop, supporting such findings.
Of more relevance from a control perspective, the FRC may be more difficult to remove from the field than other M. persicae clones due to this combination of behavioural, metabolic and target site resistance. Treating plants with neonicotinoids would probably cause the FRC to relocate to any untreated or partially sprayed plants. In the case of a persistent virus where feeding must occur for several hours in order for the virus to successfully infect a new host plant, this could limit or slow the spread of plant viruses to untreated plants, with treated plants gaining fuller protection from plant viruses. A plant virus, however, may be more difficult to remove from a crop when resistant aphids are present due to the persistence of the vector.
Different aphid species have been shown to disperse at different rates[33, 34] as do alate individuals of the same clone that differ in ovariole number.[57, 58] This is the first study, however, to show differences in dispersal rates between different aphid clones of the same species with regard to apterous individuals. The greater rate of movement of 5191A compared with the other clones or strains, even under completely control-treated conditions, suggests that different clones may disperse across a field at different rates. It is therefore also possible that the rate of viral spread within a field may vary according to the different aphid clones present. In particular, the presence of 5191A and potentially other resistant clones may lead to the spread of crop disease at a greater rate, although the lack of an enhanced dispersal rate exhibited by the resistant FRC population should be noted. The effect of the enhanced dispersal of resistant M. persicae on the spread of plant viruses however remains theoretical and further work is urgently required to both determine whether this enhanced dispersal occurs under field conditions, and to link such changes in behaviour to any possible effects on vector ability.
It is difficult to elucidate the mechanisms behind the unique dispersal behaviour shown by the FRC in this study, and further work would be required to do so. Potentially it is possible that the FRC population responds to either olfactory cues from TMX-treated host plants, or else chemical signalling molecules ingested in the phloem of TMX-treated plants, in a manner different to that of the neonicotinoid susceptible US1L clone. TMX has been shown to induce upregulation of salicylic acid-associated plant defence and stress responses in treated plants. Myzus persicae appears to use phloem constituents in host discrimination so it is possible that when feeding on treated plant tissue FRC and 5191A detect such stress chemical markers, leading to the induction of the alternative dispersal behaviours shown in this study. US1L may be either incapable of detecting such chemical markers, or else may not respond to them behaviourally. Alternatively the resistant aphids may be responding to olfactory signals produced by the host plant or insecticide. Aphids are capable of detecting plant volatiles and appear to use them when locating host plants. It is the blend of different host volatiles, rather than individual volatiles, which appears attractive to aphids and that certain plant volatiles offered individually may even be repellent. It is therefore possible that treatment of plants with TMX masks or changes such mixes of host plant volatiles and makes them less attractive to resistant M. persicae, while the susceptible clone US1L makes no such distinction. Different lines of the cotton aphid A. gossypii show different behavioural responses to the volatiles of the same host plants; hence it is possible for different clones of M. persicae to also show different behavioural responses to the same volatiles. Further olfactometry work would be needed to determine whether this is truly the case.
Since resistance to one neonicotinoid compound usually confers resistance to all other compounds within that class[63-65] it is possible that the FRC and 5191A M. persicae clones will exhibit similar behavioural differences in response to all neonicotinoid compounds. Although the nature of the relationship between behavioural and physiological insecticide resistance is in debate[66-72] it is still potentially possible that the behavioural resistance exhibited by the FRC clone would occur in the presence of other neonicotinoid compounds, although this would require further work to confirm. Indeed, the resistance factor for the FRC population is substantially higher for imidacloprid than TMX. It should be noted that in this study the FRC aphids were treated with a dose greater than the conventional field dose. TMX is not generally applied to Chinese cabbage, but the maximum recommended application rate on potato is 400 mg L−1 ha−1. Further study is therefore needed to determine whether the FRC would behave in a similar way at conventional field rates. US1L and 5191A would be unable to survive such rates, and so would present no problem under current control regimes. In addition, the dispersal behaviour of the FRC population should also be investigated at 0.1 mg L−1 and 0.9 mg L−1, to verify that any changes in behaviour are due to the resistance of the strain and not due to the increased concentration of TMX applied.
It is difficult to accurately predict the behaviour of insects in the field from laboratory results. Aphids in the field would be exposed to a range of temperature and light conditions, both of which have been shown to affect host-finding ability of the potato aphid Macrosiphum euphorbiae (Thomas) as well as the effects of weather, predation and physical barriers to movement between host plants such as soil.[75, 76] While it has been shown that M. euphorbiae can travel up to 1.8 m over bare soil to successfully locate host plants this experiment is not comparable to such conditions.
Aphids have been shown to disperse upwards from the older leaves at the bottom of the plant to the younger, newer leaves at the top. It is suggested that this is a mechanism for finding suitable feeding spots upon a host as the population rapidly grows. While the leaf tissue in each treatment of this study always originated from two separate plants, it is possible that the dispersal behaviour exhibited by these clones is representative of intra-plant movement on the same host rather than inter-plant dispersal throughout a field. Understanding the differences in intra- and inter-plant dispersal between these clones would require further work in suitable field conditions.
Much dispersal of aphid populations is mediated by winged individuals, known as alates. There is, however, evidence that dispersal of unwinged apterae can also cause substantially large infestations of plants. While revealing several interesting implications, further work on alates is needed before a full understanding of the dispersal differences between these clones can be obtained. In addition, this study only compared dispersal behaviour between single examples of a susceptible, resistant and highly resistant clone or strain. Comparisons between 5191A and FRC and a different susceptible clone, such as 4106A, or further highly neonicotinoid resistant clones which may arise in the future, could reveal differences in behaviour not exhibited by the three clones included in this study. Nevertheless this study reveals significant differences in the dispersal behaviour of M. persicae clones of different neonicotinoid resistance levels, and it would not be unreasonable to suppose that similar behavioural differences might also occur between these clones under field conditions. The lack of literature on the behaviour of neonicotinoid resistant aphid clones makes this an especially interesting study. This may be the first step towards discovering the presence of such behavioural differences in the field between neonicotinoid resistant M. persicae.
In summary, this is the first characterisation of a behavioural resistance in aphids that actively disperse away from neonicotinoid treated leaves to untreated plant material. Together with the intrinsic biochemical and target site resistance it represents new challenges in the field for aphid control and the viruses they spread.