«Can aphid-induced plant signals be transmitted aerially and through the rhizosphere? Keith Chamberlaina, Emilio Guerrierib, Francesco Pennacchioc, ...»
Biochemical Systematics and Ecology 29 (2001) 1063–1074
Can aphid-induced plant signals be transmitted
aerially and through the rhizosphere?
Keith Chamberlaina, Emilio Guerrierib, Francesco Pennacchioc,
Jan Petterssond, John A. Picketta,*, Guy M. Poppya,
Wilf Powella, Lester J. Wadhamsa, Christine M. Woodcocka
Department of Biological and Ecological Chemistry, IACR-Rothamsted, Harpenden, Herts., AL5 2JQ, UK
Centro di Studio CNR, Tecniche di Lotta biologica, Via Universita, 133, 80055 - Portici (NA), Italy " c Dipartimento di Biologia, Difesa e Biotecnologie Agro-Forestali, Universita della Basilicata, Macchia Romana, 85100 Potenza, Italy d Department of Entomology, Swedish University of Agricultural Sciences, P.O. Box 7044, S-750 07, Uppsala, Sweden Received 9 March 2001; accepted 19 April 2001 Abstract Aphids, through their close association with plants, cause systemic release of semiochemicals. These may have negative eﬀects on subsequent aphid colonisation and can also have positive roles with insects that are antagonistic to aphid development, for example parasitoids.
One of the semiochemicals involved in host selection by aphids is methyl salicylate, and since this compound was shown to have a role as a plant stress signal, the hypothesis that aphids might facilitate identiﬁcation of new plant signals was examined. Conﬁrmation was obtained during an investigation of avoidance of unsuitable hosts by the lettuce aphid, Nasonovia ribisnigri. ðZÞ-Jasmone was identiﬁed as a plant-derived semiochemical acting negatively for a number of aphid species, and positively for insect antagonists such as parasitoids and predators. However, when the compound was employed at 0.1 ppm in air above intact plants, these plants then attracted aphid parasitoids long after the ðZÞ-jasmone itself was no longer detectable. A speciﬁc interaction was proposed, since the ðZÞ-jasmone appeared to be selectively taken up by the plants. Aerial interactions between intact barley plants from diﬀerent cultivars, which may be diﬀerentially releasing stress associated signals, can also inﬂuence acceptability to aphids. Furthermore, it has been shown that exudates from the roots *Corresponding author. Tel.: 01582-763133 ext2321; fax: 01582-762595.
E-mail address: firstname.lastname@example.org (J.A. Pickett).
0305-1978/01/$ - see front matter r 2001 Published by Elsevier Science Ltd.
PII: S 0 3 0 5 - 1 9 7 8 ( 0 1 ) 0 0 0 5 0 - 3 1064 K. Chamberlain et al. / Biochemical Systematics and Ecology 29 (2001) 1063–1074 of aphid-infested plants, grown hydroponically or in soil, cause intact plants to become more attractive to parasitoids. r 2001 Published by Elsevier Science Ltd.
Keywords: Induced defence; Aphids; Signals; Rhizosphere; Phytopheromones
1. Introduction In exploring the title question, this study sets out the facts of aphid-related chemical ecology by which the hypothesis regarding associated plant/plant interactions can be tested and critically reviewed. Aphids (Homoptera: Aphididae) have particularly close interactions with plants, probably because of their highly adapted feeding mechanism which allows direct ingestion of phloem sap (Minks and
Harrewijn, 1987). Although likely to be true for most aphids, those in the subfamily:
Aphidinae, which comprise many pest species, are known to employ plant-derived signals, or semiochemicals, in interactions involved in the selection of prospective hosts (Pickett et al., 1992, 1995) and in acceptance after landing and during initial feeding behaviour (Tjallingii, 1990; Powell et al., 1999). Many species in the Aphidinae alternate between host plants, a woody species as their primary or winter host and an herbaceous annual as the secondary or summer host. Alternation can be determined by the generation of seasonal morphs and by the physiological state; for example, landing on a secondary host can be associated with the depletion of lipid reserves from migratory ﬂight. Again, semiochemicals play an important role in selection or avoidance of the respective hosts (Pettersson et al., 1994; Birkett et al.,
2000) and can also modify response to sex pheromones during the migration of sexual morphs to the primary host (Hardie et al., 1992; Guldemond et al., 1993).
However, it is the ability of aphids to detect stress or induced defence in prospective host plants that provides an opportunity to look further at plant/plant interactions and this phenomenon will be considered in detail, particularly with regard to the production of semiochemicals by the plant as a consequence of aphid feeding. Such semiochemicals, as host or prey-related compounds, present an obvious evolutionary advantage to parasitoids and predators searching for aphids. Indeed, if the argument regarding ‘‘omniscience’’ by plants (Dicke and Bruin, 2001) is considered, then synomones caused to be released by aphid feeding, and employed by insects antagonistic to these herbivores, would also be candidate phytopheromones and would thus provide opportunities for induction of defence against aphid populations developing within the ecosystems.
Mechanisms by which phytopheromones could inﬂuence plant processes, speciﬁcally the inducible secondary metabolism associated with defence, are not known, so direct detection by plants as an identiﬁcation tool is not yet available.
However, use of electrophysiology in the identiﬁcation of semiochemicals related to insect host location (Pickett et al., 1998), and particularly synomones active with the organisms antagonistic to aphids, is explored as an approach to identifying phytopheromones. Also considered are the opportunities oﬀered by the rhizosphere K. Chamberlain et al. / Biochemical Systematics and Ecology 29 (2001) 1063–1074 for transmission, below ground, of signals conveying messages relating to aphid colonisation of nearby plants. Aphids, although comprising important pests, are here providing a model insect very closely associated with plants, largely through their intimate feeding mechanisms. The objective is to use the chemical ecology of these insects, and their parasitoids and predators where possible, to explore plant/ plant interactions by means of phytopheromones. Knowledge of the molecular structures involved could, in the long term, be exploited to devise new approaches to ‘‘switching on’’ inducible defence mechanisms in crop plants prior to attack by pests and possibly pathogens.
2. Induction of plant-derived semiochemicals by aphids
After demonstrating that spring migrants of the bird cherry-oat aphid, Rhopalosiphum padi, were repelled from their primary host, bird cherry, Prunus padus, by volatile semiochemicals acting as non-host cues, active components were identiﬁed by use of electrophysiological recording techniques. Gas chromatography coupled to single cell (neuronal) recordings (GC-SCR) from olfactory organs on the antenna showed that methyl salicylate was a highly active component (Pettersson et al., 1994). This compound signiﬁcantly reduced the attractiveness of secondary host semiochemicals in the laboratory, and also aphid colonisation of crops in the ﬁeld (Pickett et al., 1995, 1997). However, in the ﬁeld trials, it was observed that other cereal aphids which do not colonise primary hosts producing methyl salicylate were also repelled by this compound. Indeed, non-cereal aphids were found to be similarly aﬀected; for example, in laboratory olfactometer assays, the black bean aphid, Aphis fabae, was no longer attracted to volatiles from its host plant when methyl salicylate, at physiologically relevant levels, was added (Hardie et al., 1994).
As work proceeded with a greater range of aphid species, and with a wide range of other insects from ﬁve Orders, neurophysiological responses to this compound were shown to be widespread (Woodcock, unpublished). Methyl salicylate is known to be released from certain plants during herbivory (Bernasconi et al., 1998; Dicke et al.,
1999) and to be directly related to salicylic acid, an internal plant stress signal derived from the inducible phenylalanine-ammonia lyase pathway. Thus, it was proposed that this compound might serve as a signal to aphids that the emitting plant may be undergoing stress through herbivory and may, as a consequence, be unsuitable, through feeding competition or induced chemical defence (Pettersson et al., 1994). At the same time, the suggestion was made that methyl salicylate might act as an external signal inducing defence pathways in intact plants, and subsequently Shulaev et al. (1997), presented evidence in favour of this proposal.
Although the link between plant stress and reduced aphid colonisation by means of methyl salicylate has not conclusively been demonstrated, such an association has been shown for the cereal/R. padi complex (Quiroz et al., 1997). Four compounds, 6-methyl-5-hepten-2-one, (À)- and (+)-6-methyl-5-hepten-2-ol and 2-tridecanone, were shown to be present in volatiles from aphid-infested wheat seedlings but not from intact plants. A mixture of the four compounds in the natural proportions 1066 K. Chamberlain et al. / Biochemical Systematics and Ecology 29 (2001) 1063–1074 counteracted the attractiveness of the volatiles from an intact wheat seedling. So far, we have no evidence that these compounds, at naturally occurring levels, are active as phytopheromones inﬂuencing subsequent colonisation attempts by aphids.
However, it is known that the C6 products of the lipoxygenase, or octadecanoid, pathway such as (E)-2-hexenal can induce expression of defence-related genes in intact plants (Bate and Rothstein, 1998).
Although there are obvious advantages in studying plant-derived semiochemicals arising directly from aphid feeding in the search for putative phytopheromone components, the next development arose again from interactions involved in aphid host alternation. The primary host of the currant-lettuce aphid, Nasonovia ribis-nigri, comprises principally the black currant, Ribes nigrum. Volatiles from intact black currant were investigated, this time using the chromatography coupled-electroantennogram system (GC-EAG). The EAG, although less sensitive than SCR, allowed detection of a range of putative semiochemicals characterising the black currant as a non-host for spring migrants searching for the secondary host, lettuce, Lactuca sativa (Birkett et al., 2000). A mixture of these compounds counteracted the attractiveness of lettuce plants in the laboratory and in the ﬁeld (L.J. Wadhams, unpublished). However, one component, ðZÞ-jasmone (=cis-jasmone), was shown to repel not only spring migrants of N. ribis-nigri but also the hop aphid, Phorodon humuli, and cereal aphids in the ﬁeld (Birkett et al., 2000). As a consequence of this more general activity, and because of its relationship with jasmonic acid, both compounds being produced by the lipoxygenase pathway, further investigations were made for semiochemical activity at higher trophic levels, and then as a putative phytopheromone component.
3. Systemic and species-speciﬁc release of semiochemicals evidenced by interactions with aphid parasitoids In ecosystems containing leguminous (fabaceous) plants, the parasitoid Aphidius ervi (Hymenoptera: Braconidae) uses the pea aphid, Acyrthosiphon pisum, as one of its main hosts, but does not normally attack other legume-feeding aphids such as A.
fabae or the vetch aphid, Megoura viciae (Star", 1973). A. ervi is strongly attracted to y bean plants, Vicia faba, infested with A. pisum, and this response is heightened following a successful oviposition experience on the plant (Du et al., 1997; Guerrieri et al., 1997; Powell et al., 1998). Furthermore, in wind tunnel bioassays, the parasitoids could discriminate between intact bean plants and those upon which the correct host was feeding (Guerrieri et al., 1993), and even between plants infested with A. pisum and those infested with the non-host A. fabae (Du et al., 1996; Powell et al., 1998). Host aphids feeding on a basal leaf of V. faba, which was subsequently removed before testing the plant, also caused increased attractiveness for up to 24 h after the aphids had been removed (Guerrieri et al., 1999). Thus, semiochemicals induced by aphid feeding can be produced systemically, with production continuing after feeding has ceased, and can be so speciﬁc as to denote the presence of a particular aphid species.
K. Chamberlain et al. / Biochemical Systematics and Ecology 29 (2001) 1063–1074 Using GC-EAG with A. ervi, electrophysiologically active components of volatiles entrained from infested bean plants were identiﬁed as comprising ðEÞ-b-ocimene, 6-methyl-5-hepten-2-one, linalool, ðZÞ-3-hexen-1-ol, ðZÞ-3-hexenyl acetate and ðEÞb-farnesene (Du et al., 1998). Some of these were also found in intact plants, but were released at higher levels from plants infested with the various aphid species. In wind tunnel bioassays, all six compounds had a signiﬁcant eﬀect on oriented ﬂight by female A. ervi. However, none of the compounds elicited a response quantitatively comparable to that for the aphid-infested plant, and even a mixture of compounds did not account for its level of attractiveness. For ﬁve of the compounds, parasitoids with oviposition experience exhibited signiﬁcantly stronger ﬂight responses than na.ve females. The exception was ðEÞ-b-farnesene, which although in this context is ı plant-derived, is also generated by aphids as a component of their alarm pheromone (Pickett and Griﬃths, 1980; Hardie et al., 1999). The most active compound in eliciting oriented ﬂight, but also signiﬁcantly more active with experienced parasitoids, was 6-methyl-5-hepten-2-one (Du et al., 1998). It has been proposed that parasitoids ‘‘learn’’ plant-derived chemicals but respond innately to hostderived cues, and the above results provide evidence for this hypothesis (Vet and Dicke, 1992).