New Phytologist Research Molecular and chemical mechanisms involved in aphid resistance in cultivated tomato Maria Cristina Digilio1*, Giandomenico Corrado2*, Raffaele Sasso3, Valentina Coppola2, Luigi Iodice3, Marianna Pasquariello2, Simone Bossi4, Massimo E. Maffei4, Mariangela Coppola2, Francesco Pennacchio1, Rosa Rao2 and Emilio Guerrieri3 1 Dipartimento di Entomologia e Zoologia agraria ‘Filippo Silvestri’, Università di Napoli ‘Federico II’, Via Università 100, 80055 Portici (NA), Italy; 2 Dipartimento di Scienze del Suolo della Pianta e dell’Ambiente, Università di Napoli ‘Federico II’, Via Università 100, 80055 Portici (NA), Italy; 3Istituto per la Protezione delle Piante, Consiglio Nazionale delle Ricerche, Via Università 133, 80055 Portici (NA), Italy; 4Dipartimento di Biologia Vegetale, Unità di Fisiologia Vegetale, Università di Torino – Centro della Innovazione, Via Quarello 11 ⁄ A, 10135 Torino, Italy Summary Authors for correspondence: Emilio Guerrieri Tel: +39 081 7753658 ext 11 Email: guerrieri@ipp.cnr.it Rosa Rao Tel: +39 081 2539204 Email: rao@unina.it Received: 11 March 2010 Accepted: 26 April 2010 New Phytologist (2010) 187: 1089–1101 doi: 10.1111/j.1469-8137.2010.03314.x Key words: Aphidius ervi, aphid parasitoid, gene expression, Macrosiphum euphorbiae, Solanum lycopersicum, volatile organic compounds. • An integrated approach has been used to obtain an understanding of the molecular and chemical mechanisms underlying resistance to aphids in cherry-like tomato (Solanum lycopersicum) landraces from the Campania region (southern Italy). The aphid–parasitoid system Macrosiphum euphorbiae–Aphidius ervi was used to describe the levels of resistance against aphids in two tomato accessions (AN5, AN7) exhibiting high yield and quality traits and lacking the tomato Mi gene. • Aphid development and reproduction, flight response by the aphid parasitoid A. ervi, gas chromatography-mass spectrometry headspace analysis of plant volatile organic compounds and transcriptional analysis of aphid responsive genes were performed on selected tomato accessions and on a susceptible commercial variety (M82). • When compared with the cultivated variety, M82, AN5 and AN7 showed a significant reduction of M. euphorbiae fitness, the release of larger amounts of specific volatile organic compounds that are attractive to the aphid parasitoid A. ervi, a constitutively higher level of expression of plant defence genes and differential enhancement of plant indirect resistance induced by aphid feeding. • These results provide new insights on how local selection can offer the possibility of the development of innovative genetic strategies to increase tomato resistance against aphids. Introduction The molecular and chemical mechanisms regulating plant resistance to insects have been studied extensively in a number of systems. The continuously increasing information in this field offers new tools and opportunities for the development of sustainable pest control technologies in agriculture. Two main categories of plant defence mechanism against insects have been proposed. The first, referred to as direct resistance, involves morphological (e.g. thorns, trichomes) and chemical (e.g. antibiotics) features which *These authors contributed equally to this work.  The Authors (2010) Journal compilation  New Phytologist Trust (2010) hamper directly the colonization of the plant by the invading insect. The second, referred to as indirect resistance, involves the production of extra floral nectar (Wäckers & Bonifay, 2004) or the release of volatile organic compounds (VOCs) by the infested plant which become attractive for the natural enemies of insect pests (Agrawal et al., 1999). A remarkable background of basic information has been produced on tomato–pest–natural enemy interactions which is unique for crop plants (Kennedy, 2003). The potato aphid, Macrosiphum euphorbiae (Hemiptera: Aphididae), and other aphid species may cause severe losses to tomato plants by their feeding activity and by the transmission of New Phytologist (2010) 187: 1089–1101 1089 www.newphytologist.com New Phytologist 1090 Research phytopathogenic viruses (Lange & Bronson, 1981; Walgenbrach, 1997). The control of aphids is largely achieved by insecticides, and the possible use of more sustainable control measures is highly desirable. Among these, strategies based on insect-resistant germplasm and biological control agents appear to be particularly promising (Guerrieri & Digilio, 2008). The mechanisms of direct resistance to aphids in tomato plants have been investigated extensively. Indeed, the most studied resistance gene to animal pests is Mi 1.2, isolated from Solanum habrochaites S. Knapp & D.M. Spooner, which confers resistance to the aphid M. euphorbiae, to the whitefly Bemisia tabaci (Gennadius), to the tomato psyllid Bactericerca cockerelli (Sulc) and to three species of nematode, including Meloidogyne incognita (Kofoid & White) (Kaloshian et al., 1997; Rossi et al., 1998; Vos et al., 1998; Nombela et al., 2003; Casteel et al., 2006). The Mi locus has been transferred to several cultivated varieties, but the level of resistance of Mi varieties seems to be limited to some biotypes of M. euphorbiae (Goggin et al., 2001), and varies with plant age (Hebert et al., 2007). The Mi mechanism is a typical gene-for-gene interaction, mediated by recognition processes of aphid-derived elicitors by plant resistance effectors. In this case, a plant resistant to aphids is characterized by the presence of a single resistance (R ) gene, which is usually inherited as a dominant trait (Smith & Boyko, 2007). The mechanisms regulating tomato attractiveness towards natural enemies of aphids have been investigated in recent years (Corrado et al., 2007; Sasso et al., 2007, 2009). The main compounds eliciting a flight response by the aphid parasitoid Aphidius ervi (Hymenoptera: Braconidae), the most effective natural enemy of M. euphorbiae, have been identified by combining behavioural observations, tomato VOC analysis and parasitoid antennal response (Sasso et al., 2007, 2009). In the multifaceted plant response, the greater production of methyl salicylate and terpenes in response to aphid infestation (Sasso et al., 2007) implies the activation of the salicylic acid (SA) and octadecanoid (jasmonic acid, JA) pathways, and indicates the possible occurrence of complex cross-talking interactions between these metabolic pathways, which are mainly involved in the response against pathogens and pests, respectively. In this study, we analyse both the direct and indirect defence mechanisms induced by aphids in tomato plants by performing a biological, behavioural, chemical and genetic characterization of resistant genotypes, selected among landraces and locally adapted accessions. Using the tritrophic system tomato (Solanum lycopersicum)–M. euphorbiae–A. ervi, we show that aphid resistance (direct and indirect) in these genotypes is associated with a constitutively higher level of expression of defence genes that respond to aphids. The gathered data may allow a more New Phytologist (2010) 187: 1089–1101 www.newphytologist.com effective manipulation and exploitation of genetic traits of tomato resistance against aphids. Materials and Methods Plants and insects The tomato (Solanum lycopersicum L.) genotypes used were the cultivated variety M82 (susceptible to aphids) and two accessions (named here AN5 and AN7) cultivated in a geographical area of the Campania region of Italy, Agro Nocerino-Sarnese, where aphid-borne viral diseases are one of the most serious problems. These two accessions belong to a collection of Corbarino cherry-like tomato landraces, partially characterized by Giordano et al. (2000), and were selected in this study for their field performances in terms of insect and virus resistance (Andreakis et al., 2004). For the genotypic characterization of plant material, we also analysed a common cultivar resistant to Verticillium wilt, Fusarium wilt and nematodes (VFN) that contains the Mi gene (Roma). Macrosiphum euphorbiae (Thomas) has been continuously reared on tomato, since 1998, in an environmental chamber at 20 ± 1C, 65 ± 5% relative humidity, 18 h light (L) : 6 h dark (D) photoperiod, starting from a colony collected on tomato plants in Scafati (Salerno, Italy). Aphidius ervi Haliday was continuously reared in an environmental chamber at 20 ± 1C, 18 h L : 6 h D photoperiod and 60 ± 5% relative humidity, on its natural host, the pea aphid Acyrthosiphon pisum (Harris), maintained on potted broad bean plants (Vicia faba L., cv Aquadulce), as described previously (Guerrieri et al., 2002). Parasitoids for flight behaviour bioassays were reared as synchronized cohorts, which were standardized as reported previously (Guerrieri et al., 2002). In brief, broad bean plants infested with A. pisum were exposed for 24 h to mated females of A. ervi. The resulting mummies were isolated and, at their emergence, adult parasitoids were sexed and placed in a box with honey at a sex ratio of 1 : 1. Female parasitoids were used for wind tunnel bioassay between 24 and 48 h after their emergence, and had no prior contact with tomato plants and ⁄ or M. euphorbiae (naive). Direct resistance against aphids Five weeks after sowing, 40 plants for each genotype were infested with a newly born first instar nymph of M. euphorbiae. Assays were carried out at 20 ± 1C, 65 ± 5% relative humidity, 18 h L : 6 h D photoperiod. The presence of aphid, presence of exuviae (i.e. occurrence of moulting) and the number of newly laid nymphs (i.e. beginning of reproduction) were monitored daily. Constitutive direct resistance against aphids was evaluated by calculating the maximum intrinsic rate of population  The Authors (2010) Journal compilation  New Phytologist Trust (2010) New Phytologist increase (rm) for each genotype by an iterative solution of the approximation of the Euler equation, rm = loge R0 =T (Birch, 1948), where R0 = X lx m x T= X xlx mx =lx mx with lx and mx representing the age-specific adult survival and the reproduction rates of female offspring at age x (expressed in days), respectively, assigned by taking the mean development time in days + 0.5 as the starting point. The accurate value of rm was then calculated by solving the equation X e rm x Research described, and tested again in a wind tunnel bioassay to assess the attractiveness towards A. ervi after aphid feeding induction. The parameters of the bioassay were set as follows: temperature, 20 ± 1C; relative humidity, 65 ± 5%; wind speed, 25 ± 5 cm s)1; distance between releasing vial and target, 50 cm; light intensity at releasing point, 3600 lx. For each experimental combination, 10 plants were used. One hundred and fifty parasitoid females were tested singly for each target in no-choice experiments, and flight behaviour data were recorded and analysed with the aid of event-recording software (the Observer; Noldus Information Technology, Wageningen, the Netherlands). The percentage of response (oriented flights, landings on the target) to each target was calculated. The number of parasitoids responding to each target in any experiment was compared by a G test for independence, with William’s correction (Sokal & Rohlf, 1981). The resulting values of G were compared with the critical values of v2 (Rohlf & Sokal, 1995). lx mx = 1 Longevity and progeny data were compared by one-way ANOVA, and the mean values were compared by Fisher’s least significant difference (LSD) test. The numbers of reproducing aphids were compared by v2 test. Induced plant response to aphid feeding To obtain information on the induction of the defence response following aphid feeding, tomato plants were infested with M. euphorbiae and leaves were collected after 7 d, after removing the experimental aphids. Approximately 200 mixed instars of potato aphids were gently transferred onto AN5, AN7 and M82 plants, at the five- to six-leaf stage, in order to obtain complete coverage of the lower side of the leaves. This infestation load was restored every other day by adding new aphids to infested plants, as required. Rearing conditions were as already described. Control, uninfested plants were grown under the same conditions. This infestation protocol was adopted to produce the plant material for the assessment of the aphidinduced plant response, both in terms of attractiveness towards parasitoids and of the expression pattern of defence genes. Flight behaviour bioassay Constitutive and induced indirect resistances to M. euphorbiae were measured as the relative attractiveness towards the parasitoid A. ervi in a wind tunnel bioassay (Guerrieri et al., 2002). For each genotype, the constitutive attractiveness was measured by testing 10 uninfested plants, 4 wk old, over 10 consecutive days. The same plants were then infested with M. euphorbiae for 1 wk, as already  The Authors (2010) Journal compilation  New Phytologist Trust (2010) Headspace volatile collection and analyses Volatiles from 10 uninfested plants per genotype were collected for 24 h by an air-tight entrainment system, using an adsorbent trap made of Tenax TA 60 ⁄ 80 (Sasso et al., 2007), soon after plants had been tested in the wind tunnel bioassay. Known amounts of standard molecules (see later) were used to calculate the recovery of the adsorption ⁄ desorption method and for calibration. The traps were desorbed by pipetting 2 ml of redistilled hexane. The eluted extract was concentrated under a nitrogen flux to a volume of 500 ll. The concentrated extracts and standard solutions were injected (3 ll volume) in the split ⁄ splitless injection port (injector) of a 6890N gas chromatograph (Agilent Technologies, Inc., Santa Clara, CA, USA), coupled with an EI-quadrupole 5973 mass spectrometer (Agilent Technologies). Chromatographic separation was carried out with a ZB-5MS column (Zebron Phenomenex, Inc., Torrance, CA, USA). The experimental conditions were as follows: injector, 260C; transfer line to Mass Spectrometer Detector (MSD), 280C; oven temperature: start, 60C; hold, 5 min; programmed from 60C to 87C at 1C min)1; a second temperature ramp was programmed from 87C to 150C at 5C min)1; a third temperature ramp was programmed from 150C to 300C at 20C min)1; hold, 2 min; flow rate of carrier gas (helium), 1 ml min)1. Mass spectral data acquisition was performed in SIM mode and in SCAN mode. The SIM mode parameters set for the mass filter (quadrupole) were organized for each run in four groups of specified masses, according to the retention times and spectra previously found in test runs using standard molecules: first group, 77 m ⁄ z and 99 m ⁄ z; second group, 79 m ⁄ z, 91 m ⁄ z and 93 m ⁄ z; third group, 119 m ⁄ z, 92 m ⁄ z and 120 m ⁄ z; fourth group, 91 m ⁄ z, New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 1091 New Phytologist 1092 Research 93 m ⁄ z, 133 m ⁄ z and 161 m ⁄ z. The SCAN mode parameters were 50 m ⁄ z as the lower mass and 220 m ⁄ z as the higher mass. Both acquisition settings were carried out with an ionization energy of 70 eV. Volatile compounds were identified by comparison of their mass spectra and retention indices (Kováts indices) with those of reference substances, where possible, and by comparison with the National Institute of Standards and Technology (NIST) mass spectral search software v 2.0, using the libraries NIST 98 and Adams (2001). The mean area of each identified peak and the total emission of volatiles were analysed by one-way t-paired test (Genstat 11th, VSN International Ltd Hemel Hempstead UK). The following commercial standards were used for the identification of volatiles collected by air entrainment of the headspace from tomato plants, and were for gas chromatography unless specified (purity level in parentheses): anisolep-allyl (‡ 98.5%), camphor (‡ 95%), d-2-carene (‡ 96%), (E)-b-caryophyllene (‡ 98.5%), chlorobenzene (‡ 99%), a-copaene (‡ 90%), p-cymene (‡ 99.5%), decane (‡ 99.8%), p-dichlorobenzene (‡ 98%), dodecene (‡ 99%), 1,8-cineol (‡ 99%), eugenol (‡ 99%), (E)-b-farnesene (‡ 90%), a-gurjunene (‡ 97%), hexanal (‡ 97%), (Z)-3-hexen-1-ol (‡ 98%), humulene (= a-caryophyllene) (‡ 98%), (Z)-jasmone (‡ 99%), (E)-jasmone, (R)-(+)-limonene, (S)-())-limonene, linalool, (+)-longifolene, menthol, 6-methyl-5-hepten-2-one, methyl salicylate, b-myrcene, (Z)-nerolidol, ocimene, (R)-())-a-phellandrene, a-pinene, skatol, a-terpinene, c-terpinene, a-terpineol, iso-terpinolene. Isolation of genomic DNA from plants and molecular fingerprint Leaves (0.5–2 g) were ground in liquid nitrogen and resuspended in 15 ml of extraction buffer (0.1 M TrisHCl pH 8.0, 0.05 M EDTA, 0.5 M NaCl, 0.01 M b-mercaptoethanol). Membranes were lysed by adding 2 ml of 10% w ⁄ v SDS and incubating at 65C for 20 min. After the addition of 5 ml of 5 M potassium acetate and incubation on ice for 15 min, the extract was centrifuged at 20 800 g for 15 min at 4C. The supernatant was added with 0.6 vol of isopropanol and the pellet of nucleic acids Longevity (d) (mean ± SD) Fertile aphids (%) Progeny, $$ ⁄ $ (mean ± SD) rm k was resuspended in 400 ll of TE buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA), and incubated with 10 ll of 10 mg ml)1 RNase A at 37C for 30 min. The solution was then transferred into a 2 ml Eppendorf tube, incubated with 400 ll cetyltrimethylammonium bromide (CTAB) buffer at 65C for 15 min, and a subsequent extraction with 1 vol of chloroform was performed. After centrifugation for 10 min at 20 800 g at 4C, the aqueous phase was transferred to a new tube, the DNA was precipitated by the addition of 1 vol of isopropanol and centrifuged at 20 800 g (4C) for 10 min. The pellet was washed with 70% ethanol, centrifuged at 20 800 g (4C) for 5 min, air-dried and resuspended in 100 ll TE buffer. The DNA fingerprint with the (GATA)4 oligonucleotide was performed as described previously (Rao et al., 2006). For the analysis of cleaved amplified polymorphic sequences (CAPS) of resistance genes, DNA was amplified using the following conditions: reaction volume of 50 ll containing 1 · Taq polymerase buffer (Promega), 1.5 mM MgCl2, 0.1 mM of each deoxynucleoside triphosphate (dNTP), 0.2–0.4 lM of each of the two oligonucleotide primers, 1 u Taq Polymerase (Promega) and 100 ng of genomic DNA. PCRs started with an initial denaturation step at 94C for 4 min, followed by 30 cycles of amplification that included a denaturation step of 1 min at 94C, an annealing time of 45 s, at the temperatures indicated in Supporting Information Tables S1 and S2, and an extension period of 60 s kb)1 of amplified target at 72C. A final step of 9 min at 72C concluded the PCRs. Cycles were performed using a Mastercycler Gradient PCR machine (Eppendorf, Milan, Italy). Amplicons were cleaved using restriction endonucleases according to the manufacturer’s recommendations (Promega), employing a three-fold excess of enzyme in a final volume reaction of 100 ll. The enzymes and primers employed and the expected fragments are indicated in Supporting Information Table S1. Isolation of RNA from plants and cDNA synthesis To analyse gene expression in plant material standardized as for VOC analysis, total RNA was isolated from 4-wk-old uninfested plants and from plants infested for 7 d with M82 AN5 AN7 16.65 ± 5.64 a 73.08 a 13.58 ± 15.56 a +0.39 1.48 11.08 ± 3.94 b 5.00 b 0.18 ± 0.96 b –0.09 0.91 14.43 ± 2.93 c 20.00 c 0.98 ± 2.42 b +0.05 1.05 Table 1 Demographic analysis of Macrosiphum euphorbiae on the tomato genotypes M82, AN5 and AN7 (rm, maximum rate of increase; k, finite rate of increase) Longevity and progeny data were compared by one-way ANOVA (F-value = 16.873 and 27.238, respectively, n = 40, P < 0.01), and mean values were compared by Fisher’s least significant difference (LSD) test; longevity, t = 1.79, P < 0.01; progeny, t = 3.77, P < 0.01. The number of fertile aphids was compared by v2 test. For both types of test, statistically significantly different values (P < 0.01) are denoted with different letters. New Phytologist (2010) 187: 1089–1101 www.newphytologist.com  The Authors (2010) Journal compilation  New Phytologist Trust (2010) New Phytologist Research Table 2 Relative amounts of the 14 volatile compounds identified in tomato susceptible variety M82 and in the resistant genotypes AN5 and AN7 Kováts index (Z)-3-Hexen-1-ol a-Pinene Myrcene d-2-Carene a-Phellandrene p-Cymene Limonene (Z)-Ocimene c-Terpinene Iso-terpinolene Methyl salicylate (Z)-Jasmone Longifolene (E)-b-Caryophyllene 859 939 991 1002 1003 1025 1029 1037 1060 1089 1192 1393 1408 1419 M82 5141 80 102 27 000 8255 58 307 13 788 129 484 6330 33 166 60 902 13 005 23 800 27 378 1 410 835 ± ± ± ± ± ± ± ± ± ± ± ± ± ± AN5 0.448 16.143 9.983 4.183 24.658 9.629 27.150 3.836 13.025 64.017 0.258 10.711 1.704 135.211 5645 ± 154 148* ± 56 792** ± 13 424 ± 38 915 ± 105 031**± 22 703 931** ± 6718 ± 33 604 ± 1 236 445** ± 20 882* ± 41 042** ± 62 287** ± 3 593 807**± AN7 1.418 61.134 7.774 4.302 12.076 30.502 11 411.537 3.160 15.703 137.059 2.900 18.114 20.117 1361.842 5633 ± 177 975*± 1541** ± 583 458** ± 89 534 ± 34 621 ± 2 354 701*± 4715 ± 30 680 ± 411 059** ± 11 564 ± 21 077 ± 36 331 ± 1 676 884 ± 1.434 57.941 0.843 480.060 55.488 13.226 1317.174 1.287 20.850 145.540 1.533 5.117 7.286 313.327 Values, expressed in nanograms, are the mean ± SD. For each compound, significant differences between the mean values observed for each accession and that of M82 are denoted with asterisks (*, P £ 0.05; **, P £ 0.01). hundreds of M. euphorbiae, as described previously. Aphids were removed manually from infested plants, and the leaves of the third branch were collected and immediately frozen in liquid nitrogen. Approximately 0.5 g of leaf tissue was ground in liquid nitrogen and total RNA was isolated according to already published procedures (Corrado et al., 2008). Ten micrograms of total RNA were treated with 7.5 u of RNase-free DNase I (Pharmacia, Milan, Italy) and first-strand cDNA synthesis was carried out as described previously (Corrado et al., 2005). The PCR amplification of the cDNA coding for the Elongation Factor 1-a (EF1-a) gene, a ubiquitously expressed gene, served as a control for cDNA synthesis and PCR efficiency in the different samples. The sequences annealed by the two primers (EF fw and EF rv; Table S2) are localized in exons I and II of the EF1-a gene, respectively, for the detection of possible contaminant DNA in the PCR amplifications (Corrado et al., 2007). Real-time PCR Real-time PCRs were performed using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems, Milan, Italy). Reactions (total volume, 25 ll) were prepared with 12.5 ll of the 2 · SYBR Green PCR Master Kit (Applied Biosystems), 0.3 pmol of a primer pair and 0.2 ll of cDNA template. For each target, reactions were performed in triplicate and experiments were carried out on three replicates per genotype. The thermal cycling programme started with steps of 2 min at 50C and 10 min at 95C, followed by 40 cycles of a 15 s step at 95C, followed by 1 min at Ta indicated in Table S1. To check the  The Authors (2010) Journal compilation  New Phytologist Trust (2010) specificity of the amplification products, a dissociation kinetics analysis was performed after each assay. The reaction products were also resolved onto an agarose gel to verify amplicon size. The primer pairs used and the size of the expected amplicons are shown in Table S2. Quantification of gene expression was carried out using the 2)DDCt method (Livak & Schmittgen, 2001). We used the housekeeping EF1-a gene as an endogenous reference gene (Nicot et al., 2005) for the normalization of the expression levels of the target genes. The statistical significance of the results was performed by evaluating whether the average 2)DDCt values of the resistant accession were significantly different from those of the calibrator genotype (Student’s t-test). Results Genotypic characterization Tomato accessions that are locally cultivated can be characterized by the presence of intravarietal genetic variability (Ruiz et al., 2005; Rao et al., 2006). Therefore, six plants of the accessions AN5 and AN7 were fingerprinted using a highly polymorphic DNA marker (Vosman et al., 1992; Rao et al., 2006; Caramante et al., 2009). The analysis with the (GATA)4 oligonucleotide indicated that the two local accessions were genetically uniform and different, as all plants analysed generated different patterns fully consistent within each genotype (Fig. 1a). Furthermore, we wanted to exclude the presence of a few well-known resistance genes, frequently present in cultivated varieties, which could directly or indirectly influence plant performance in the New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 1093 New Phytologist 1094 Research (a) (kb) 1 2 3 (b) (c) 10.0 (bp) 650 5.0 M82 AN5 AN7 M82 AN5 AN7 R REX-1 (Mi) TG101 (Frl) 350 570 420 3.0 SCN13 (Tm2) 360 170 220 0.5 600 R110 (Pto) 400 bioassays. This analysis was also essential to select an appropriate susceptible control, because many commercial varieties have introgressed major resistance genes. We analysed four CAPS markers, tightly linked to genes conferring resistance to aphids and other biotic stresses. The markers were REX-1, linked to the Mi 1.2 gene (Williamson et al., 1994), TG101, linked to the Frl gene, which confers resistance to Fusarium oxysporum f.sp. radicis-lycopersici (Sacc.) W.C. Snyder and H.N. Hans (Fazio et al., 1999), SCN13, linked to the tm-2 gene (resistance to tobacco mosaic virus) (Sobir et al., 2000), and R110, linked to the Pto gene, involved in the resistance to Pseudomonas syringae pv tomato (Young dye and Wilkie) (Martin et al., 1991) and to other bacterial and fungal plant pathogens (Mysore et al., 2003). The CAPS assay did not show any polymorphism in the analysed samples (Fig. 1b,c), indicating that the resistance alleles of the analysed genes are absent in the AN5 and AN7 accessions, as well as in the control cultivar M82. (rm = )0.09), and a finite rate of increase (k) below unity, which is the value registered in the case of a stable population (Table 1). A positive value of the intrinsic rate of increase was recorded on AN7, although it was close to zero (rm = 0.05), owing to poor reproduction, with only 0.98 ± 2.4 nymphs (mean ± SD) per aphid. On this genotype, k was 1.05, indicating the presence of a very small increase in the aphid population in the ideal conditions of the assay (best climatic conditions and absence of intra- and interspecific competition). The results of the wind tunnel bioassay with uninfested plants are reported in Fig. 2. Both AN5 and AN7 showed a significantly higher level of constitutive attractiveness towards A. ervi with respect to M82 (oriented flights, 32.6%). The uninfested AN5 showed the strongest response Constitutive direct and indirect resistance to M. euphorbiae A. ervi response (%) Fig. 1 Genetic analysis of the tomato accessions AN5 and AN7. (a) (GATA)4 fingerprint of the AN5 (1) and AN7 (2) accessions and of a Macrosiphum euphorbiae-resistant VFN (resistant to Verticillium wilt, Fusarium wilt and nematodes) tomato variety (Roma) (3). (b) Cleaved amplified polymorphic sequence (CAPS) analysis of the REX-1 marker, tightly associated with the Mi aphid resistance gene. A digestion control of a VFN variety, Roma (R), is shown for the REX-1 marker as, in this case, the susceptible allele is not digested by the Taq I enzyme. (c) CAPS analysis of genes that confer resistance to biotic stress. The three panels show the digestion profile of the markers (indicated in capital letters on the left) associated with resistance genes (in parentheses) that were selected as very common in commercial tomato cultivars. 80 b The biological performance of aphids on the three genotypes is reported in Table 1. As indicated by the rm values, M82 control plants were susceptible, allowing the experimental aphids to survive and reproduce much better than on the two germplasm accessions considered. Conversely, both the AN5 and AN7 genotypes displayed a significant level of resistance, with aphids showing impaired growth, which frequently prevented them from reaching the critical size for moulting. On AN5, the aphid population disappeared in 11 ± 3.9 d (mean ± SD), and only 5% of the experimental aphids underwent reproduction. This resulted in a negative value for the intrinsic rate of increase New Phytologist (2010) 187: 1089–1101 www.newphytologist.com b 60 c 40 a b a 20 0 Oriented flights Landings on the target Fig. 2 Percentage of response (oriented flights, landings on the target) of the aphid parasitoid Aphidius ervi towards uninfested plants of the tomato genotypes M82 (white columns), AN5 (black columns) and AN7 (grey columns). Mean values within each response denoted with different letters are significantly different (P < 0.01).  The Authors (2010) Journal compilation  New Phytologist Trust (2010) New Phytologist Research (64.29%), which was significantly higher (P £ 0.01) than that registered for AN7 (49.35%). The percentages of landings on the target showed the same pattern (Fig. 2), as a similar and very high percentage of oriented flights resulted in a landing on the target for all three genotypes (93.1% in M82, 84.4% in AN5 and 89.5% in AN7). Headspace VOC analyses Of the selected standards (see previously), only 14 compounds were detected in the collected VOCs, and, in many cases, significant quantitative differences were observed between genotypes (Table 2). Compared with the M82 control, the AN5 genotype released significantly greater amounts of the following compounds: a-pinene, myrcene, p-cymene, limonene, iso-terpinolene, methyl salicylate, (Z)jasmone, longifolene and (E)-b-caryophyllene. AN7 emissions were significantly higher than those of M82 for a-pinene, myrcene, limonene and iso-terpinolene. Only for one compound (d-2-carene) was the emission of genotype AN7 significantly higher than that of the other two genotypes. The relative proportion of identified compounds, within each genotype considered, and the relative ratio for each compound between the two accessions and the susceptible cultivated variety M82 are reported in Table 3. AN5 releases quantities of limonene and iso-terpinolene 175-fold and 20-fold higher, respectively, than M82, whereas, in AN7, these higher release rates decrease to 18-fold and seven-fold, respectively. Moreover, d-2-carene is released by AN7 in quantities 70-fold higher than by M82. The AN5 genotype released a substantially greater amount of the identified volatiles (sum of the average area peak of each identified compound, 28 223 432.5), followed by AN7 (3 821 604.6) and M82 (703 255.9), and this pattern is consistent with the results of the constitutive attractiveness towards parasitoids of uninfested plants. Table 3 Relative proportion of identified compounds within each tomato genotype, and between the accessions AN5 and AN7 with respect to the cultivated variety M82 (Z)-3-Hexen-1-ol a-Pinene Myrcene d-2-Carene a-Phellandrene p-Cymene Limonene (Z)-Ocimene c-Terpinene Iso-terpinolene Methyl salicylate (Z)-Jasmone Longifolene (E)-b-Caryophyllene  The Authors (2010) Journal compilation  New Phytologist Trust (2010) Constitutive expression of the aphid-induced genes is higher in resistant genotypes The changes in plant gene expression induced by aphids and, in general, by phloem-feeding insects are complex and involve different signalling pathways (Thompson & Goggin, 2006; Girling et al., 2008). Our first aim was to identify the genes belonging to different plant response pathways, which are activated by aphid feeding in tomato. This analysis was carried out on the susceptible cultivar M82, 7 d post-infestation, and the results are shown in Fig. 3. Among the genes of the octadecanoid pathway analysed, only one (LoxC) of the two plastidial tomato Lox isoforms that participate in the synthesis of JA, and the hydroperoxide lyase gene (HPL), producing stress-inducible compounds such as Green Leaf Volatile (GLV), were overexpressed. HPL-derived metabolites are also strictly linked to direct resistance against aphids in potato (Vancanneyt et al., 2001). A significant induction of expression was also observed for the germacrene C synthase (GCS) gene, which encodes for proteins involved in the biosynthesis of terpenoids (a major class of VOC in plants), and for the P4 gene, coding a pathogenesis-related (PR) protein, with unknown functions, that is induced by aphids and SA. However, differences were not observed for another pathogeninducible gene, Pti4, which encodes an ethylene-responsive transcription factor which is important for the activation of GCC-box PR genes in tomato (Chakravarthy et al., 2003). The specificity of the plant response to aphids was also indicated by the lack of activation of the prosystemin gene, a primary signal for the systemic transmission of the defence signal induced by herbivore chewers in tomato plants. These data suggest that mechanical damage plays little role in the elicitation of plant response when aphids have established their feeding site. Subsequently, we compared the constitutive expression level in the resistant and susceptible genotypes. As shown in Relative proportion M82 Relative proportion AN5 Relative proportion AN7 Relative proportion AN5 ⁄ M82 Relative proportion AN7 ⁄ M82 0.0027 0.0422 0.0142 0.0044 0.0307 0.0073 0.0682 0.0033 0.0175 0.0321 0.0069 0.0125 0.0144 0.7435 0.0002 0.0055 0.0020 0.0005 0.0014 0.0037 0.8088 0.0002 0.0012 0.0440 0.0007 0.0015 0.0022 0.1280 0.0010 0.0327 0.0003 0.1073 0.0165 0.0064 0.4329 0.0009 0.0056 0.0756 0.0021 0.0039 0.0067 0.3083 1.0982 1.9244 2.1034 1.6262 0.6674 7.6178 175.3418 1.0613 1.0132 20.3021 1.6056 1.7245 2.2750 2.5473 1.0957 2.2218 0.0571 70.6814 1.5356 2.5110 18.1853 0.7449 0.9250 6.7495 0.8892 0.8856 1.3270 1.1886 New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 1095 New Phytologist 1096 Research Response to aphids of the susceptible genotype 80 10 ** b ** ** 8 7 RQ 6 5 4 ** 3 A. ervi response (%) 9 b b b 60 a a 40 20 2 1 0 0 GCS HPL P4 Prosys Pti4 TomLoxC TomLoxD Gene Fig. 3 Relative gene expression analysis in tomato 7 d following a compatible aphid infestation. The graph displays the relative quantity (RQ) for each target gene, shown on a linear scale relative to the calibrator uninfested M82 (open columns) and 7 d after aphid attack (closed columns). Asterisks indicate that the 2)DDCt values were significantly different from the calibrator (P < 0.01; Student’s t-test). Oriented flights Landings on the target Fig. 5 Percentage of response (oriented flights, landings on the target) of the aphid parasitoid Aphidius ervi towards the tomato genotypes (M82, white columns; AN5, black columns; AN7, grey columns) after aphid infestation. Different letters within each response indicate a significant difference (P < 0.01). (a) Response to aphids of the AN5 genotype 4 3.5 RQ 2.5 12 ** 2 1.5 RQ 10 1 0.5 ** 8 2 0 ** 6 4 ** 3 Constitutive expression of defence genes 14 ** ** ** GCS HPL P4 Prosys Gene Pti4 TomLoxC TomLoxD ** (b) Response to aphids of the AN7 genotype 4 ** ** 3.5 3 0 GCS P4 Prosys Pti4 Fig. 4 Relative expression level of genes involved in response to aphids in tomato aphid-susceptible (M82) and aphid-resistant (AN5 and AN7) genotypes. The graph displays the relative quantity (RQ), shown on a linear scale relative to the calibrator M82, of each target in M82 (white columns), AN5 (grey columns) and AN7 (black columns). Asterisks indicate that the 2)DDCt values were significantly different from the calibrator (P < 0.01; Student’s t-test). Fig. 4, both AN5 and AN7 express at significantly higher levels the four genes induced by aphid feeding: HPL, TomLoxC, GCS and P4. Induced plant response to aphid feeding Following aphid infestation, the attractiveness of AN7 and M82 increased significantly, and reached a level similar to that of the uninfested AN5 plants, which, by contrast, did not show any enhancement of their attractiveness (compare Figs 2, 5) (G test showed that uninfested AN5 and all infested genotypes formed a homogeneous group: P = 1.1, not shown). Similar patterns of response were observed by analysing the percentages of landings on the target (Fig. 5): New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 2.5 TomLoxC TomLoxD Gene RQ HPL 2 1.5 1 0.5 0 GCS HPL P4 Prosys Gene Pti4 TomLoxC TomLoxD Fig. 6 Relative gene expression analysis 7 d following aphid infestation of the tomato aphid-resistant genotypes (a, AN5; b, AN7). The graph displays the relative quantity (RQ) for each target gene, shown on a linear scale relative to the uninfested calibrator (open columns) and 7 d after aphid attack (closed columns). Asterisks indicate that the 2)DDCt values were significantly different from the calibrator (P < 0.01; Student’s t-test). the percentage with respect to oriented flights remained consistently high in all three genotypes examined (91.9% in M82, 87.2% in AN5 and 92.6% in AN7). We also monitored the variation in gene expression level in the two resistant genotypes, 7 d following infestation (Fig. 6). The results indicated that only one gene in each genotype (P4 for the AN5 genotype and HPL for the AN7 genotype) was expressed at a higher level relative to uninfested plants. Interestingly, the induced higher expression of  The Authors (2010) Journal compilation  New Phytologist Trust (2010) New Phytologist HPL in AN7 was found to be consistent with the observed enhanced attractiveness. Discussion The biodiversity of local germplasm accessions provides an interesting reservoir of plant material, which may show natural defences against biotic stress agents, associated with high-quality standards. Tomato plants have developed an efficient surveillance system that allows rapid reactions against biotic stress, which complement efficient constitutive defences, such as glandular trichomes, acyl sugar and toxic phenolic compounds. It has been demonstrated that R proteins detect, directly or indirectly, the presence not only of pathogens, but also of pest effectors, as in the case of the Mi 1.2 gene (Rossi et al., 1998). In addition to these genefor-gene interactions, exerting a strong selection pressure on the natural populations of pest insects, tomato plants can mount an effective response by activating metabolic pathways that modulate multifactorial resistance mechanisms, targeting herbivores both directly and indirectly. We have identified two tomato genotypes, which lack the Mi 1.2 gene, with a significantly high level of direct resistance against M. euphorbiae, coupled with attractiveness towards aphid parasitoids. The molecular fingerprint indicated that these genotypes were genetically different and, correspondingly, their performances were also different. For the AN5 genotype, only a few M. euphorbiae reached the adult stage and none reproduced. The resulting negative value of rm indicates a very high level of resistance, which would certainly prevent a colonizing population of aphids from becoming established on this plant material. In the case of AN7, the rm value was just above zero, indicating that this genotype does not allow diffuse aphid colonization. For both germplasm accessions, direct resistance was associated with an enhanced constitutive attractiveness towards the parasitoid A. ervi. The attractiveness of AN5 in the absence of aphid infestation was high and comparable with that recorded for the same parasitoid in response to broad bean plants infested by Acyrthosiphon pisum (Guerrieri et al., 1993; Du et al., 1996), or to tomato plants infested by M. euphorbiae (Sasso et al., 2007). VOC analysis indicated the existence of a positive correlation between the amount of volatiles released and attractiveness towards parasitoids, which could be reinforced by the different relative ratios of specific compounds shared by the considered genotypes. A thorough analysis of the absolute and relative releases of the identified compounds from the three uninfested genotypes, characterized by different degrees of attractiveness towards A. ervi females, allowed additional light to be shed on the mechanisms regulating the interactions between this generalist wasp and plants. The high attractiveness towards uninfested AN5 plants is associated with a high constitutive level of different compounds, with at least three  The Authors (2010) Journal compilation  New Phytologist Trust (2010) Research ((Z)-jasmone, methyl salicylate and (E)-b-caryophyllene)) showing upregulation in tomato plants attacked by aphids (Sasso et al., 2007), and eliciting positive electrophysiological responses, even at very low doses (Sasso et al., 2009; E. Guerrieri & C. M. Woodcock, unpublished). However, this change is associated with very high levels of limonene (175 times higher than in M82, and representing 81% and 43% of the total volatiles released by AN5 and AN7, respectively), which has no direct effect on the modulation of A. ervi flight behaviour and its electrophysiological response (Sasso et al., 2009). We may reasonably speculate that limonene may play a role, in combination with other compounds, by enhancing the overall detectability of the blend in which it is present in specific relative amounts. The biological importance of the ratio among compounds in the studied experimental system is also corroborated by the attraction of A. ervi towards the AN7 genotype, which shares with AN5 only some of the volatiles more significantly released by the latter, with the relative proportions of all the others being markedly different. It is also worth noting that the three main aphid-induced compounds mentioned above, involved in A. ervi attractiveness, are extremely common in the volatile blends of different plant species. This further reinforces the important role of the relative ratios in the elicitation of specific biological responses, even though a complex mixture of compounds, including the major components of volatile blends in equal concentrations, proved to be attractive for a large number of parasitoids (James, 2005; James & Grasswitz, 2005). The importance of the ratio among different compounds in a complex blend is certainly one of the major parameters controlling insect–plant interactions (de Bruyne & Baker, 2008). In phytophagous insects, this parameter strongly influences the selection of plants on which insects feed (Visser, 1986; Fraser et al., 2003), even though some groups can exploit very specific chemical cues, uniquely associated with a given group of plants (e.g. isothiocyanates for cruciferous plants) (Blight et al., 1995). There could be an apparent ecological contradiction associated to the concurrent presence of high levels of direct and indirect defences. Plants, as sessile organisms, must protect themselves against biotic stress agents occurring in the environment with powerful weapons, which very often are synthesized ‘on demand’ by activating specific metabolic pathways (Gatehouse, 2002; Howe & Jander, 2008). A number of inducible defence responses in plants occur with profound metabolic changes, some of which are not always involved in the direct defence reaction, but can be profitably exploited by natural antagonists of plant feeders ⁄ pathogens. Therefore, the selective advantage for the plant of getting rid of insect pests can lead to a selection process which stabilizes genetic traits controlling the indirect defences mediated by carnivores, and promotes an active ‘cry for help’ strategy (Dicke, 2009). However, a series of adaptive plant responses New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 1097 1098 Research seems to be more passive in an evolutionary sense (i.e. exploitation by natural enemies of metabolic changes not related to protection), and the proposed adaptiveness of ‘crying for help’ has been challenged in some cases (Van der Meijden & Klinkhamer, 2000; Gatehouse, 2002). These scenarios are not mutually exclusive and could coexist in the same evolutionary context in response to multiple challenges, promoting the development of plant defence traits that contribute significantly to the structure of natural insect communities (Poelman et al., 2008). The experimental data presented here provide an interesting opportunity to analyse this microevolutionary pathway. Aphid fitness is at the lowest on the tomato genotypes considered. T associated higher attractiveness towards aphid parasitoids seems to be a consequence of the increased constitutive expression of aphid-inducible genes, which regulate both direct and indirect resistance. Although the benefit of aphid suppression is obvious, it is more difficult to determine, at this stage, the benefit associated with the higher attractiveness of parasitoids towards poor or very low host populations on fairly resistant plants. In other words, the attraction of parasitoids by uninfested plants can be viewed as a kind of collateral product of a selection process in favour of effective direct defences, and the absence of host aphids on these attractive plants is expected to enhance the rapid dispersal of parasitoids. This hypothesis needs to be tested under field conditions in order to assess the relative contributions of direct and indirect defences to the overall resistance of the tomato genotypes considered here. These accessions offer new tools to analyse the dynamics of the microevolutionary processes driving the selection of alleles conferring plant resistance against insects. Nonetheless, some of the volatiles emitted at higher rates by AN5 and AN7 may also have some direct effects on aphids. The possible direct defensive role of (E)-b-caryophyllene remains to be assessed, whereas it has been reported previously for both methyl salicylate and (Z)-jasmone. For example, barley plants exposed to methyl salicylate were less well accepted and less well colonized by the aphid Rhopalosiphum padi (Glinwood et al., 2007). Similarly, (Z)jasmone has been demonstrated to be repellent for the aphid Nasonovia ribis-nigri (Birkett et al., 2000; Bruce et al., 2008). Conversely, (E)-b-caryophyllene has been more often related to indirect defence only, above ground and below ground, against different pests in different crops (Colazza et al., 2004; Rassman et al., 2005). Therefore, we hypothesize that some of the volatile compounds identified may play a double role. Studies addressing this issue in the experimental system used are certainly worthy of further research efforts. We are currently assessing the possible role in direct defence against aphids of the other volatile compounds released at a higher rate by the two accessions considered. This will shed light on the biological role of these chemicals, and may provide new tools for aphid control. New Phytologist (2010) 187: 1089–1101 www.newphytologist.com New Phytologist The bioassays performed with plant material indicated that, in the tomato genotypes used, the resistance traits are expressed mainly (AN7), if not entirely (AN5), in a constitutive manner. As plant resistance to biotic stress also includes the activation of gene expression, we wanted to test whether aphid responsive genes were constitutively expressed at higher levels in the aphid-resistant genotypes. For this purpose, we first studied plant response after a prolonged aphid attack, by analysing gene expression after 7 d of infestation. Key genes were selected, considering their role in the signalling pathways that, in tomato, are elicited by phytophagous pests. Selected genes belong to the JA and SA pathways and are also involved in the production of bioactive VOCs. As expected, the data indicated that the tomato response to aphids involves different signalling pathways. For instance, it has been shown that tomato lipoxygenases are involved in plant-induced defences towards pests through the production of toxic and antifeedant compounds (Fidantsef et al., 1999). The gene expression analysis performed indicated that, among the tomato Lox isoforms that are targeted to chloroplasts, only TomLoxC is involved in aphid response. This is consistent with the previous evidence which indicated that TomLoxC, as a member of the lipoxygenase gene family, is involved in the response against herbivores (Corrado et al., 2007). More interestingly, the results indicated that, among the genes analysed, overexpression in the resistant genotypes was observed only for genes activated by aphid infestation. Considering that, of the resistant genotypes, AN5 showed a constitutively higher level of expression for HPL, GCS and TomLoxC, it is tempting to speculate that the higher resistance level of this genotype is correlated with these quantitative differences. However, the higher expression of HPL in AN7, induced by aphid feeding, is associated with an enhancement of attractiveness, which attains a level similar to that constitutively observed for AN5 (Figs 2, 5). The results strongly suggest that the resistance observed in the landraces AN5 and AN7, although most probably multifactorial, is related to a constitutively higher level of expression of defence genes that respond to aphid attack. In other plant species, genes described as upregulated in aphidresistant genotypes are involved in different physiological processes. They include vacuolar H+-ATPase subunit-like proteins in apple, and cytochrome P450 monoxygenase genes, chlorophyll a ⁄ b-binding protein genes and cellulose synthase genes in sorghum and wheat (Qubbaj et al., 2005; Boyko et al., 2006; Park et al., 2006). It is worth noting that the different biological performances shown by the resistant and susceptible genotypes analysed are associated with the overexpression of genes whose activity is clearly linked to direct and indirect resistance against pests. However, it will be interesting to perform a comprehensive transcriptomic analysis to outline a complete picture of  The Authors (2010) Journal compilation  New Phytologist Trust (2010) New Phytologist tomato response to aphid infestation. Moreover, it will also be of considerable interest to assess the variability of aphid and parasitoid response to the phenotypic traits controlled by these genes, which is likely to occur in different strains, as reported for the Mi gene (Goggin et al., 2001). Apparently, the higher cost associated with plant constitutive defences does not seem to be relevant for AN5 (Zangerl, 2003). Indeed, the resistance traits of the ‘best performer’ AN5 do not influence yield and fruit quality significantly, which are comparable or better than those of other local landraces and commercial hybrids (Giordano et al., 2000; Andreakis et al., 2004). In conclusion, this study has identified aphid-resistant genotypes in cultivated tomato and candidate genes whose expression is associated with aphid resistance. Moreover, the tomato genotypes considered can provide new opportunities for studying the microevolutionary pathways that shape plant resistance. Although further studies are necessary to unravel the complexity of the genetic basis of the observed traits, our results corroborate the hypothesis that traditional germplasm, which has been subjected to very little breeding, may represent an important source of material amenable to low-input farming and ⁄ or organic agriculture. Our data also suggest that the modulation and exploitation of endogenous plant defence could be a valid strategy to improve tomato resistance against aphids, hopefully overcoming some limitations of the single gene resistance approach (Singh & Singh, 2005). Acknowledgements This research was supported by Ministero dell’Istruzione, dell’Universitá e della Ricerca – Laboratorio PubblicoPrivato di Genomica per l’innovazione e la valorizzazione della filiera del pomodoro (GenoPOM), DM17732. References Adams P. 2001. Identification of essential oil components by gas chromatography:quadrupole mass spectroscopy. Carol Stream, IL, USA: Allured Publications. Agrawal AA, Tuzun S, Bent E. 1999. Induced plant defenses against pathogens and herbivores. Biochemistry, ecology, and agriculture. St Paul, MN, USA: APS Press. Andreakis N, Giordano I, Pentangelo A, Fogliano V, Graziani G, Monti LM, Rao R. 2004. DNA fingerprinting and quality traits of Corbarino cherry-like tomato landraces. Journal of Agricultural and Food Chemistry 52: 3366–3371. Birch LC. 1948. The intrinsic rate of natural increase of an insect population. Journal of Animal Ecology 17: 15–26. Birkett MA, Campbell CAM, Chamberlain K, Guerrieri E, Hick AJ, Martin JL, Matthes M, Napier JA, Pettersson J, Pickett JA et al. 2000. New roles for cis-jasmone as an insect semiochemical and in plant defence. Proceedings of the National Academy of Sciences, USA 97: 9329– 9334. Blight MM, Pickett JA, Wadhams LJ, Woodcock C. 1995. Antennal perception of oilseed rape, Brassica napus (Brassicaceae), volatiles by the  The Authors (2010) Journal compilation  New Phytologist Trust (2010) Research cabbage seed weevil Ceutorhynchus assimilis (Coleoptera, Curculionidae). Journal of Chemical Ecology 21: 1649–1664. Boyko EV, Smith CM, Thara VK, Bruno JM, Deng YP, Starkey SR, Klaahsen DL. 2006. Molecular basis of plant gene expression during aphid invasion: wheat Pto- and Pti-likeA sequences are involved in interactions between wheat and Russian wheat aphid (Homoptera: Aphididae). Journal of Economic Entomology 99: 1430– 1445. Bruce TJ, Matthes MC, Chamberlain K, Woodcock CM, Mohib A, Webster B, Smart LE, Birkett MA, Pickett JA, Napier JA. 2008. cis-Jasmone induces Arabidopsis genes that affect the chemical ecology of multitrophic interactions with aphids and their parasitoids. Proceedings of the National Academy of Sciences, USA 105: 4553– 4558. de Bruyne M, Baker TC. 2008. Odor detection in insects: volatile codes. Journal of Chemical Ecology 34: 882–897. Caramante M, Rao R, Monti LM, Corrado G. 2009. Discrimination of ‘San Marzano’ accessions: a comparison of minisatellite, CAPS and SSR markers in relation to morphological traits. Scientia Horticulturae 120: 560–564. Casteel C, Walling LL, Paine T. 2006. Behavior and biology of the tomato psyllid, Bactericerca cockerelli, in response to the Mi-1.2 gene. Entomologia Experimentalis et Applicata 121: 67–72. Chakravarthy S, Tuori RP, D’Ascenzo MD, Fobert PR, Despres C, Martin GB. 2003. The tomato transcription factor Pti4 regulates defense-related gene expression via GCC box and non-GCC box cis elements. Plant Cell 15: 3033–3050. Colazza S, McElfresh JS, Millar JG. 2004. Identification of volatile synomones, induced by Nezara viridula feeding and oviposition on bean spp., that attract the egg parasitoid Trissolcus basalis. Journal of Chemical Ecology 30: 945–964. Corrado G, Arciello S, Fanti P, Fiandra L, Garonna A, Digilio MC, Lorito M, Giordana B, Pennacchio F, Rao R. 2008. The Chitinase A from the baculovirus AcMNPV enhances resistance to both fungi and herbivorous pests in tobacco. Transgenic Research 17: 557–571. Corrado G, Delli Bovi P, Ciliento R, Gaudio L, Di Maro A, Aceto S, Lorito M, Rao R. 2005. Inducible expression of a Phytolacca heterotepala ribosome-inactivating protein leads to enhanced resistance against major fungal pathogens in tobacco. Phytopathology 95: 206–215. Corrado G, Sasso R, Pasquariello M, Iodice L, Carretta A, Cascone P, Ariati L, Digilio MC, Guerrieri E, Rao R. 2007. Systemin regulates both systemic and volatile signalling in tomato plants. Journal of Chemical Ecology 33: 669–681. Dicke M. 2009. Behavioural and community ecology of plants that cry for help. Plant, Cell & Environment 32: 654–665. Du YJ, Poppy GM, Powell W. 1996. Relative importance of semiochemicals from first and second trophic level in host foraging behaviour of Aphidius ervi. Journal of Chemical Ecology 22: 1591– 1606. Fazio G, Stevens MR, Scott JW. 1999. Identification of RAPD markers linked to fusarium crown and root rot resistance (Frl) in tomato. Euphytica 105: 205–210. Fidantsef AL, Stout MJ, Thaler JS, Duffey SS. 1999. Signal interactions in pathogen and insect attack: expression of lipoxygenase, proteinase inhibitor II, and pathogenesis-related protein P4 in the tomato, Lycopersicon esculentum. Physiological and Molecular Plant Pathology 54: 97–114. Fraser AM, Mechaber WL, Hildebrand JG. 2003. Electroantennographic and behavioral responses of the sphinx moth Manduca sexta to host plant headspace volatiles. Journal of Chemical Ecology 29: 1813–1833. Gatehouse JA. 2002. Plant resistance towards insect herbivores: a dynamic interaction. New Phytologist 156: 145–169. Giordano I, Pentangelo A, Villari G, Fasanaro G, Castaldo D. 2000. Caratteristiche agronomiche e idoneità alla trasformazione di pomodori New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 1099 1100 Research dell’ecotipo ‘‘Corbarino’’. Industria Conserve 75: 317–329. (In Italian with English abstract). Girling RD, Madison R, Hassall M, Poppy GM, Turner JG. 2008. Investigations into plant biochemical wound-response pathways involved in the production of aphid-induced plant volatiles. Journal of Experimental Botany 59: 3077–3085. Glinwood R, Gradin T, Karpinska B, Ahmed E, Jonsson L, Ninkovic V. 2007. Aphid acceptance of barley exposed to volatile phytochemicals differs between plants exposed in daylight and darkness. Plant Signalling & Behavior 2: 321–326. Goggin FL, Williamson VM, Ullman DE. 2001. Variability in the response of Macrosiphum euphorbiae and Myzus persicae (Hemiptera: Aphididae) to the tomato resistance gene Mi. Environmental Entomology 30: 101–106. Guerrieri E, Digilio MC. 2008. Aphid–plant interactions: a review. Journal of Plant Interactions 3: 223–232. Guerrieri E, Pennacchio F, Tremblay E. 1993. Flight behaviour of the aphid parasitoid Aphidius ervi (Hymenoptera: Braconidae) in response to plant and host volatiles. European Journal of Entomology 90: 415– 421. Guerrieri E, Poppy GM, Powell W, Rao R, Pennacchio F. 2002. Plant to plant communication mediating in-flight orientation of Aphidius ervi. Journal of Chemical Ecology 28: 1703–1715. Hebert SL, Jia L, Goggin FL. 2007. Quantitative differences in aphid virulence and foliar symptom development on tomato plants carrying the Mi resistance gene. Environmental Entomology 36: 458–467. Howe GA, Jander G. 2008. Plant immunity to insect herbivores. Annual Review of Plant Biology 59: 41–66. James DG. 2005. Further field evaluation of synthetic herbivore-induced plant volatiles as attractants for beneficial insects. Journal of Chemical Ecology 31: 481–495. James DG, Grasswitz TR. 2005. Synthetic herbivore-induced plant volatiles increase field captures of parasitic wasps. BioControl 50: 871– 880. Kaloshian I, Kinsey MG, Ullman DE, Williamson VM. 1997. The impact of Meu 1-mediated resistance in tomato on longevity, fecundity and behavior of the potato aphid Macrosiphum euphorbiae. Entomologia Experimentalis et Applicata 83: 181–187. Kennedy GG. 2003. Tomato, pests, parasitoids, and predators: tritrophic interactions involving the genus Lycopersicon. Annual Review of Entomology 48: 51–72. Lange WH, Bronson L. 1981. Insect pests of tomato. Annual Review of Entomology 26: 345–371. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(T)()Delta Delta C) method. Methods 25: 402–408. Martin GB, Williams JGK, Tanksley SD. 1991. Rapid identification of markers linked to a Pseudomonas resistance gene in tomato by using random primers and near-isogenic lines. Proceedings of the National Academy of Sciences, USA 88: 2336–2340. Mysore KS, D’Ascenzo MD, He XH, Martin GB. 2003. Overexpression of the disease resistance gene Pto in tomato induces gene expression changes similar to immune responses in human and fruitfly. Plant Physiology 132: 1901–1912. Nicot N, Hausman JF, Hoffmann L, Evers D. 2005. Housekeeping gene selection for real-time RT-PCR normalization in potato during biotic and abiotic stress. Journal of Experimental Botany 56: 2907–2914. Nombela G, Williamson VM, Muniz M. 2003. The root-knot nematode resistance gene Mi-1.2 of tomato is responsible for resistance against the whitefly Bemisia tabaci. Molecular Plant–Microbe Interaction 16: 645– 649. Park SJ, Huang YH, Ayoubi P. 2006. Identification of expression profiles of sorghum genes in response to greenbug phloem-feeding using cDNA subtraction and microarray analysis. Planta 223: 932–947. New Phytologist (2010) 187: 1089–1101 www.newphytologist.com New Phytologist Poelman EH, van Loon JJA, Dicke M. 2008. Consequences of variation in plant defense for biodiversity at higher trophic levels. Trends in Plant Science 13: 534–541. Qubbaj T, Reineke A, Zebitz CPW. 2005. Molecular interactions between rosy apple aphids, Dysaphis plantaginea, and resistant and susceptible cultivars of its primary host Malus domestica. Entomologia Experimentalis et Applicata 115: 145–152. Rao R, Corrado G, Bianchi M, Di Mauro A. 2006. (GATA)(4) DNA fingerprinting identifies morphologically characterized ‘San Marzano’ tomato plants. Plant Breeding 125: 173–176. Rassman S, Köllner TG, Degenhard J, Hiltpold I, Toepfer S, Kuhlmann U, Gershenzon J, Turlings TCJ. 2005. Recruitment of entomopathogenic nematodes by insect-damaged maize roots. Nature 434: 732–737. Rohlf FJ, Sokal RR. 1995. Statistical tables, 3rd edn. New York, NY, USA: WH Freeman and Co. Rossi M, Goggin FL, Milligan SB, Kaloshian I, Ullman DE, Williamson VM. 1998. The nematode resistance gene Mi of tomato confers resistance against the potato aphid. Proceedings of the National Academy of Sciences, USA 95: 9750–9754. Ruiz JJ, Garcia-Martinez S, Pico B, Gao MQ, Quiros CF. 2005. Genetic variability and relationship of closely related Spanish traditional cultivars of tomato as detected by SRAP and SSR markers. Journal of the American Society for Horticultural Science 130: 88–94. Sasso R, Iodice L, Digilio MC, Carretta A, Ariati L, Guerrieri E. 2007. Host-locating response by the aphid parasitoid Aphidius ervi to tomato plant volatiles. Journal of Plant Interactions 2: 175–183. Sasso R, Iodice L, Pickett JA, Woodcock CM, Guerrieri E. 2009. Behavioural and electrophysiological responses of Aphidius ervi (Hymenoptera: Braconidae) to tomato plant volatiles. Chemoecology 19: 195–201. Singh DP, Singh A. 2005. Disease and insect resistance in plants. Enfield, NH, USA: Science Publisher. Smith CM, Boyko EV. 2007. The molecular bases of plant resistance and defense responses to aphid feeding: current status. Entomologia Experimentalis et Applicata 122: 1–16. Sobir TO, Ohmori T, Murata M, Motoyoshi F. 2000. Molecular characterization of the SCAR markers tightly linked to the Tm-2 locus of the genus Lycopersicon. Theoretical and Applied Genetics 101: 64–69. Sokal RR, Rohlf FJ. 1981. Biometry, 2nd edn. New York, NY, USA: WH Freeman and Co. Thompson GA, Goggin FL. 2006. Transcriptomics and functional genomics of plant defence induced by phloem-feeding insects. Journal of Experimental Botany 57: 747–754. Van der Meijden E, Klinkhamer PGI. 2000. Conflicting interests of plants and the natural enemies of herbivores. Oikos 89: 202–208. Vancanneyt G, Sanz C, Farmaki T, Paneque M, Ortego F, Castañera P, Sanchez-Serrano JJ. 2001. Hydroperoxide lyase depletion in transgenic potato plants leads to an increase in aphid performance. Proceedings of the National Academy of Sciences, USA 98: 8139–8144. Visser JH. 1986. Host odour perception in phytophagous insects. Annual Review of Entomology 31: 121–144. Vos P, Simons G, Jesse T, Wijbrandi J, Heinen L, Hogers R, Frijters A, Groeendijk J, Diergaarde P, Reijans M et al. 1998. The tomato Mi-1 gene confers resistance to both root-knot nematodes and potato aphids. Nature Biotechnology 16: 1365–1369. Vosman B, Arens P, Ruskortekaas W, Smulders MJM. 1992. Identification of highly polymorphic DNA regions in tomato. Theoretical and Applied Genetics 85: 239–244. Wäckers FL, Bonifay C. 2004. How to be sweet? Extrafloral nectar allocation by Gossypium hirsutum fits optimal defense theory predictions Ecology 85: 1512–1518.  The Authors (2010) Journal compilation  New Phytologist Trust (2010) New Phytologist Walgenbrach JF. 1997. Effect of potato aphid (Homoptera: Aphididae) on yield, quality, and economics of staked-tomato production. Journal of Economic Entomology 90: 996–1004. Williamson VM, Ho JY, Wu FF, Miller N, Kaloshian I. 1994. A PCRbased marker tightly linked to the nematode resistance gene, Mi, in tomato. Theoretical and Applied Genetics 87: 757–763. Zangerl AR. 2003. Evolution of induced plant responses to herbivores. Basic and Applied Ecology 4: 91–103. Supporting Information Additional supporting information may be found in the online version of this article.  The Authors (2010) Journal compilation  New Phytologist Trust (2010) Research Table S1 Primers for cleaved amplified polymorphic sequence (CAPS) analysis and size of the expected restriction fragments (R, resistance allele; S, susceptible allele). Table S2 Primers employed for the expression study and their main features. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. New Phytologist (2010) 187: 1089–1101 www.newphytologist.com 1101