Specialist v/s Generalist

Amongst the 4,000 aphid species that have been described, approximately 250 are considered a pest (Dixon, 1998; Blackman and Eastop, 2000). Based on their host range, aphids are classified as specialist or generalists. Specialist aphids feed only on a restricted set of related plant species (Lankau, 2007). For instance, mustard aphid (Lipaphis erysimi) and cabbage aphid (Brevicoryne brassicae) feed only on cruciferous plants (Blackman and Eastop, 2000). On the other hand, a generalist aphid, such as the green peach aphid (GPA; Myzus persicae) (Figure 1c and 1d) feeds on a wide array of plant species and is considered polyphagous (Lankau, 2007; Blackman and Eastop, 2000). It has been suggested that specialist aphids have the ability to locate their host plant based on the presence of unique secondary metabolite(s) that act as cues for host recognition, feeding and oviposition (Raybould and Moyes, 2001; Macel and Vrieling, 2003). By comparison, it has been suggested that generalist aphids cue in on a combination of plant primary and secondary metabolites to make their host selection (Powell et al., 2006).

Life Cycle

Aphids are capable of sexual and asexual reproduction. They have a heteroecious holocyclic life cycle, involving alternate hosts. For instance, peach (Prunus persicae) is the primary host for GPA. Other plants that can serve as primary hosts for GPA include black cherry, (P. serotina), Canadian plum (P. nigra), dwarf Russian almond (P. tenella), and peach-almond hybrids (Blackman and Eastop, 2000). Secondary hosts for GPA include more than 50 families of plants that comprise, amongst others, vegetable crops like squash, cabbage, radish, mustard, celery, lettuce, tomato, potato and eggplant (Blackman and Eastop, 2000). Sexual reproduction, which occurs only during a portion of their life cycle, results in the production of female and male sexual morphs. The females lay their eggs on the primary host plant, where they can overwinter. In places where a primary host plant is absent or not available, and if climate permits the asexual stages to survive winter, GPA exhibits an anholocyclic life cycle and reproduces parthenogenically (offsprings produced without fertilization). In the lab, GPA exhibits an anholocylic life cycle under typical Arabidopsis growth conditions. However, in many GPA lineages, sexual reproduction can be induced in the laboratory by maintaining colonies at cool temperatures under short-day conditions (Blackman, 1974).

Feeding Patterns

Aphids display a variety of feeding patterns on plants. The electrical penetration graph (EPG) has provided a powerful technique to characterize the feeding patterns of phloem-feeding insects (Tjallingii, 1990; Tjallingii and Esch, 1993; Reese et al., 2000; Walker, 2000). This technique has been extensively utilized to investigate the details of plant resistance to aphids, whiteflies and leafhoppers (Diaz-Montano et al., 2007; Pegadaraju et al., 2007; Jiang et al., 2000; Stafford and Walker, 2009). In an EPG set up (Figure 2a), the insect on the plant is part of a low-voltage circuit. The patterns of electrical waveforms generated provide information on the feeding activity of the insect (van Helden and Tjallingii, 2000). In case of aphids, the three different phases that constitute the EPG waveforms include pathway phase, sieve element or phloem phase, and xylem phase (Reese et al., 2000; Tjallingii, 2006). During the pathway phase, the stylet is inserted in the tissue but outside the phloem and xylem. The pathway phase includes both the intercellular and/or intracellular stylet insertion or withdrawal during the brief sampling of cells (Tjallingii and Esch, 1993). The sieve element or phloem phase occurs when the stylet tips are in a phloem sieve element. EPG studies have indicated that during the sieve element phase, when the insect is ingesting phloem sap, it periodically resorts to intermittent salivation into the sieve elements. Xylem phase occurs when the insect is ingesting the xylem sap and is often thought to be related to water uptake (Spiller et al., 1990). Water uptake has been suggested as a means for the insect to dilute out the gut contents, and thus minimize dehydration (Pompon et al., 2010). A representative EPG waveform of a GPA feeding on wild type Arabidopsis accession Columbia plant is shown in Figure 2b. EPG has been used to study the impact of mutations in Arabidopsis genes on aphid feeding behavior and thus the contribution of individual genes and mechanisms to different aspects of Arabidopsis defense and susceptibility to aphids (Pegadaraju et al., 2007; Louis et al., 2010a, 2010b, 2012; Singh et al., 2011; Nalam et al., 2012).

Figure 2.

Electrical penetration graph (EPG) setup used for characterizing the feeding behavior of aphids on its host plant.

(A) Diagram depicting the components of EPG system (Illustration by Nick Sloff). The EPG set-up has two components -the insect and plant electrodes. A very thin, gold wire is glued to the dorsum of the aphid using a conductive silver paint. This thin wire (insect electrode), which helps in the free movement of the aphid on the plant surface is connected to the EPG probe. A stiff copper wire is inserted into the soil of the pot, in which the plant is rooted (plant electrode). It electrifies the plant with a low-voltage, low amperage current. As the stylet of the aphid comes into contact with the electrified plant, the circuit is closed and current flows through the insect and the different signals (waveforms) are thus produced. Inset shows an aphid wired with the thin gold wire using silver conductive paint (Picture provided by Gregory Zolnerowich, Kansas State University).

(B) A representative EPG waveform of GPA feeding on Arabidopsis wild type accession Columbia plant. The different waveform patterns shown here represent different phases of aphid probing of the plant. In the pathway phase (PP) the aphid stylet has penetrated the tissue but is outside the sieve elements and xylem. The pathway phase includes both the intercellular and/or intracellular stylet insertion or withdrawal during the brief sampling of cells. The non-probing (NP) phase represents a state when there is relatively no stylet movement (NP). The sieve element phase (SEP) and the xylem phase (XP) represent the stages when the stylet is in a sieve element or a xylem tissue, respectively.

Saliva

Salivary secretions of aphids have a vital role in plant-aphid interactions (Mutti et al., 2008; De Vos and Jander, 2009). The aphid stylet (Figure 1b) punctures the plant tissue and injects aphid salivary secretions into the host (Tjallingii, 1990). Two types of saliva are injected by aphids into the host plant - a gelling saliva and a watery saliva (Miles, 1999). The proteinaceous gelling saliva, which is secreted as the stylet penetrates the plant tissue, forms a supportive sheath around the stylet and a tight seal around the sites on the cell surface that may have been punctured by the stylet, thus further limiting wounding damage to the plant. Intracellular puncture by the stylet is followed by a rapid switch to secretion of non-gelling (watery) saliva and only this watery saliva is delivered into the penetrated cell or sieve elements (Cherqui and Tjallingii, 2000; Powell, 2005). The watery saliva is suggested to contain various enzymes, including pectinases, cellulases, polyphenol oxidases, peroxidases, and lipases (Campbell and Dreyer 1990; Miles, 1990, 1999). These enzymes may perform vital roles in insect feeding, including lubrication of the stylets, maintaining favorable oxidative-reduction (redox) conditions and detoxification of phenolics and other chemical factors (Miles and Oertli, 1993). Furthermore, once the aphid reaches the sieve elements of the plant, salivary secretions help to limit phloem sealing and callose deposition, which enables the aphid to feed continuously for many hours or even days from a single sieve element (Will and van Bel, 2006; Will et al., 2007, 2009). As discussed below, salivary components could potentially also function as effectors that modulate plant physiology, including defense responses.

Carbohydrate Metabolism

Aphid infestation alters source-sink patterns in the infested plant, leading to the generation of a strong sink in the aphid-infested organ and thus increasing flow of nutrients to the infested organ at the cost of nutrient flow to the natural sink tissues (Mittler and Sylvester, 1961; Dixon, 1998; Girousse et al., 2005). Indeed, expression of genes involved in sugar transport and metabolism were altered in GPA infested Arabidopsis (Moran and Thompson, 2001; Moran et al., 2002; Pegadaraju, 2005). The concentration of sucrose and starch increased in GPA-infested leaves of Arabidopsis (Singh et al., 2011). Since sucrose has a large osmolyte effect, at first thought it would seem that this increase in sucrose and the corresponding increase in osmolahty of the phloem sap would be detrimental to the aphid, which expends lot of energy to counter the high osmolarity of the diet it consumes. However, Singh et al. (2011) demonstrated that GPA numbers were higher on the Arabidopsis tps11 (At2g18700) mutant plant, which accumulates higher levels of sucrose and lower amounts of starch in response to GPA infestation, than in the wild type plant, suggesting that increased sucrose levels are unlikely to be a mechanism utilized by plant to control GPA infestation. Instead, Singh et al. (2011) proposed that the accumulation of starch, at the expense of sucrose, is a mechanism utilized by plants to counter the insect. They demonstrated that in comparison to the wild type plant, insect numbers were higher on the Arabidopsis pgm1 (At5g51820) mutant plant, which is unable to synthesize starch. Similarly, in comparison to the wild type plant, the increment in starch content was lower in the tps11 mutant on which the insect population was larger. It has been suggested that starch accumulation at the expense of sucrose results in a secondary intracellular sink that counters the ability of the insect to alter host physiology to make the plant more suitable for the insect (Singh et al., 2011).

Aphid infestation also impacts nitrogen metabolism in the plants. Pea aphid (Acyrthosiphon pisum) infestation on alfalfa (Medicago sativa) resulted in an evident shift from nitrogen sinks to nitrogen sources (Girousse et al., 2005). Furthermore, aphid settlement on the actively growing zones of stem resulted in a systemic reduction of C and N fluxes, especially at the apical zones of the stem (Girousse et al., 2005). Molecular studies have also indicated changes in expression and/or activity of genes/proteins involved in N metabolism. For example, GPA infestation resulted in increased nitrate reductase activity in cabbage (Wilson et al., 2011). Voelckel et al. (2004) showed that tobacco aphid (Myzus nicotianae) infestation resulted in the induction of glutamate synthase expression. Similarly expression of glutamine synthase was upregulated in GPA infested celery (Divol et al., 2005).

Premature Senescence

GPA infestation resulted in the up-regulation of a subset of SENESCENCE ASSOCIATED GENES (SAG) (Pegadaraju et al., 2005; De Vos et al., 2005), and the activation of cell death and senescence in Arabidopsis (Pegadaraju et al., 2005). Some of these SAG genes were also induced in leaves infiltrated with synthetic diet containing aphid saliva (De Vos and Jander, 2009), suggesting that aphid salivary components elicit cell death in Arabidopsis. Indeed, the putative salivary protein Mp10 from GPA when transiently overexpressed in Nicotiana benthamiana promoted chlorosis (Bos et al., 2010). When compared to wild type Arabidopsis, the hyper-senescent ssi2 (At2g43710) and cpr5 (At5g64930) mutants exhibited enhanced resistance to GPA, and insect population size was larger on the pad4 (At3g52430) mutant, which exhibited delayed onset of senescence and expression of SAG genes in response to GPA infestation. Hence, it was suggested that premature senescence is a defense mechanism employed by the host to control GPA infestation (Pegadaraju et al., 2005). Senescence likely counters the efforts of the insect to increase sink nature of the infested tissues.

Plant Hormones

Plants produce a variety of hormones including auxins, gibberellins, abscisic acid, cytokinins, salicylic acid, jasmonic acid, ethylene, brassinosteroids and peptide hormones. Of these, salicylic acid (SA) and jasmonic acid (JA) are more widely studied for their signaling role in plant defense against various biotic stresses (Shah, 2003; Lorenzo and Solano, 2005; Broekaert et al., 2006; Bari and Jones, 2009; Pieterse et al., 2009). Several studies have shown that aphid feeding on host plant activates SA and JA-mediated defense pathways (Moran and Thompson, 2001; Mewis et al., 2005). GPA feeding on Arabidopsis activates genes that are involved in SA biosynthesis and signaling (Moran and Thompson, 2001; Pegadaraju, 2005). However, GPA population size on the SA biosynthesis mutant sid2 (salicylic acid-induction deficient2; At1g74710), the SA insensitive mutant npr1 (non-expressor of PR-1; At1g64280) and a SA deficient NahG transgenic plant that expresses a SA-degrading salicylate hydroxylase, were comparable to that on wild type plants, suggesting that SA is not a key player in Arabidopsis defense against GPA (Moran and Thompson, 2001; Pegadaraju et al., 2005). Quite to the contrary, Mewis et al., (2005) showed that both the GPA and cabbage aphid populations on the npr1 mutant and the NahG plant were smaller as compared to the wild type plant, suggesting that lack of NPR1 function and SA accumulation resulted in increased resistance. However, the role of SA may differ depending on the plant and insect involved. In contrast to the studies in Arabidopsis, in tomato, it was shown that hyperaccumulation of SA due to knock-down of FAD7 activity resulted in enhanced resistance to potato aphid (Macrosiphum euphorbiae) (Avila et al., 2012).

Since SA signaling is known to attenuate the activation of JA signaling, Mewis et al., (2005) suggested that the higher level of resistance against GPA observed in the npr1 mutant and NahG plants could possibly result from the stronger activation of the JA signaling pathway in these plants. Similarly, in the case of another sap-sucking insect, silverleaf whitefly (Bemisia tabaci), larval growth was reduced on npr1 and NahG plants, whereas the growth was higher on SA-overexpressing cim10 plants as compared to the wild type plants (Zarate et al., 2007). Zarate et al., (2007) suggested a ‘decoy’ hypothesis in which the insect evades host defenses by tricking the host plant into inappropriately activating SA signaling and thus depressing the activation of JA signaling. Indeed, other studies also have shown that resistance against aphids is mediated through the JA pathway (Ellis et al., 2002; Zhu-Salzman et al., 2004; Gao et al., 2007). For example, the GPA population was smaller on the Arabidopsis cev1 (At5g05170) mutant, which has high JA content (Ellis et al., 2002). In contrast, GPA numbers were higher on JA-insensitive mutant coi1 (At2g39940) mutant compared to the wild type plant (Ellis et al., 2002). JA has also been suggested to have an important role in defense against aphids in other plant species. For example, methyl jasmonate (MeJA) treatment of sorghum (Sorghum bicolor) seedlings resulted in fewer numbers of greenbug aphids (Schizaphis graminum) as compared to the untreated control plants (Zhu-Salzman et al., 2004). However, JA-responsive genes were only weakly induced in greenbug-infested sorghum leaves. By comparison, SA responsive genes were strongly induced in greenbug-infested sorghum (Zhu-Salzman et al., 2004). It is likely that, as in the case of whitefly infestation of Arabidopsis, greenbug tricks sorghum plants to induce SA signaling to prevent the timely activation of JA signaling. Similarly, the JA pathway is also required for AKR (Acyrthosiphon kondoi Resistance) mediated resistance against the blue green aphid (Acyrthosiphon kondoi) in Medicago truncatula (Gao et al., 2007).

JA also contributes to basal resistance against cabbage aphid in Arabidopsis. Cabbage aphid populations were smaller on the Arabidopsis cev1 and fou2 (Arabidopsis fatty acid oxygenation up-regulated2; At4g03560) mutants, respectively, both of which have high JA content (Kuśnierczyk et al., 2011). However, EPG studies revealed that the fou2 allele does not limit cabbage aphid feeding from sieve elements, thus suggesting that mechanisms other than feeding deterrence contribute to fou2 and JA-determined resistance against cabbage aphid. In addition to elevated levels of JA, Arabidopsis fou2 mutants also had higher levels of the JA precursor, oxo-phytodienoic acid (OPDA) and the related dinor oxo-phytodienoic acid (dnOPDA) (Bonaventure et al., 2007; Kuśnierczyk et al., 2011). Hence, further experiments are required to tease apart the contributions of JA and/or OPDA to fou2-determined heightened resistance against cabbage aphid. Besides JA, OPDA is also known to function as a signal molecule in Arabidopsis (Taki et al., 2005).

Effectors

Oral secretions from caterpillars contain elicitors of plant defenses (e.g. volicitin, caeliferins and inceptins) (Alborn et al., 1997, 2007; Schmelz et al., 2006). Aphids also intermittently inject into sieve elements a watery saliva that contains proteins and other metabolites. One or more of these salivary components could elicit and/or modulate plant responses. For example, aphid saliva contains oxidases (e.g. glucose oxidase) (Harmel et al., 2008) that could potentially produce hydrogen peroxide (H₂O₂), a key modulator of plant defense (Bede et al., 2006; Maffei et al., 2006), similar to that observed with glucose oxidase from corn earworm (Helicoverpa zea) (Musser et al., 2002). Although the exact function of glucose oxidase in aphid saliva is not known, recently it was shown that the Arabidopsis RBOHD (RESPIRATORY BURST OXIDASE HOMOLOG D; At5g47910) gene, which is responsible for accumulation of reactive oxygen species (ROS), including local accumulation of H₂O₂ at the wound/injury site, modulates defense against GPA (Miller et al., 2009). GPA proliferation was higher on rbohD mutant plants than the wild type plant. These results suggest that any changes in the ROS status in the sieve elements could influence plant resistance/susceptibility to aphids. Further, evidence indicating a potential role for aphid salivary components in modulating plant response against aphids was provided by experiments with Arabidopsis leaves infiltrated with an artificial diet on which GPA had fed and thus contained aphid salivary secretions. Infiltration of diet containing aphid saliva resulted in differential expression of Arabidopsis genes related to signal transduction, response to stress and secondary metabolite biosynthesis (e.g. glucosinolate), and genes encoding various stress associated transcription factors (De Vos and Jander, 2009).

RNAi experiments with pea aphid provided the first evidence of importance of salivary components in facilitating aphid infestation on plants. Silencing expression of the pea aphid gene encoding the C002 protein, which is normally expressed in the salivary glands, adversely impacted the ability of the C002-silenced aphids to feed from the host plant (Mutti et al., 2008), thus suggesting that C002 is critical for aphids to continuously feed from the host plant (Mutti et al., 2008). Similarly, silencing expression of the C002 homolog of GPA by expression of dsRNA in transgenic Arabidopsis resulted in the production of less progeny by the C002-silenced GPA (Pitino et al., 2011). By contrast, agro-infiltration mediated overexpression of the GPA C002 protein in Nicotiana benthamiana enhanced GPA fecundity as compared to the vector control (Bos et al., 2010). These results suggest that C002 facilitates aphid infestation, most likely due to its action in the host plant. By contrast to C002, the Mp10 and MP42 proteins from GPA, which are encoded by genes expressed in the salivary glands of the aphid, when overexpressed in N. benthamiana resulted in reduced fecundity of GPA. Mp10 overexpression in N. benthamiana also resulted in chlorosis and attenuated ROS production in response to treatment with flg22, a bacterial effector peptide (Bos et al., 2010), suggesting that Mp10 impacts plant responses. Although, whether plants perceive the presence of aphids by recognizing these effectors is not known, the above studies suggest that aphid salivary components can elicit plant responses that likely impact the interaction between the host plant and aphid.

Resistance genes

The Arabidopsis-GPA and Arabidopsis-cabbage aphid systems have been successfully utilized by several groups to identify a number of plant genes (Table 2 and 3) and mechanisms that contribute to plant defense against aphids. However, to date, R gene type resistance against aphids in Arabidopsis has not been reported. Only very few examples of gene-for-gene resistance in plant-aphid interactions are known. For example, the Mi-1.2 gene in tomato (Solanum lycopersicum) and the Vat gene in melon are involved in providing protection against certain isolates of potato aphid and melon aphid (Aphis gossypii), respectively (Rossi et al., 1998; Dogimont et al., 2007). Mi-1.2 was the first aphid resistance gene to be cloned. Both Mi-1.2 and Vat genes are members of plant R genes consisting of nucleotide binding site (NBS) and leucine-hch-repeat (LRR) motifs (Milligan et al., 1998, Dogimont et al., 2007). In addition to potato aphids, Mi-1.2 also confers resistance against certain isolates of root-knot nematodes and biotypes of whitefly (B. tabaci), and the tomato psyllid (Bactericera cockerelli) (Nombela et al., 2003; Casteel et al., 2006). Mi-1.2 mediated resistance against root-knot nematodes functions through the activation of hypersensitive response (HR), which is associated with cell death at the feeding site, which results in the blockage of nematode feeding (Dropkin et al., 1969; Roberts and Thomason, 1986). In contrast, the Mi-1.2-mediated resistance to potato aphid is independent of HR (de Illarduya et al., 2003). Presence of Mi-1.2 confers resistance to potato aphid by limiting the insect's ability to feed continuously from the phloem sap (Kaloshian et al., 2000; Palliparambil et al., 2010). The bluegreen aphid resistance gene, AKR, identified in M. truncatula, also maps to a region containing NBS-LRR class of genes (Klingler et al., 2005). In addition, Nr in lettuce, Sd1 in apple, RAG1 and RAG2 in soybean, and RAP1 in M. truncatula have also been reported in conferring resistance against lettuce aphid, rosy-leaf curling aphid, soybean aphid and pea aphid, respectively (Helden et al., 1993; Roche et al., 1997; Hill et al., 2006; Stewart et al., 2009, Kim et al., 2010). Several loci conferring biotype-specific resistance, including Aph in maize, and Gby and Dn in wheat, have also been reported in providing resistance against corn leaf aphid, greenbug and Russian wheat aphid, respectively (Marais and Du Toit, 1993; Boyko et al., 2004; Castro et al., 2005; So et al., 2010).

Table 2.

Arabidopsis mutants and transgenic plants exhibiting altered resistance to green peach aphid

Glucosinolates—a Brassicaceae-specific Chemical Defense

Plants in the Brassicaceae family, which includes Arabidopsis thaliana, accumulate glucosinolates, a family of secondary metabolites that are sources of thioacyanates and other breakdown products that are toxic to some aphids (Rask et al., 2000; Halkier and Gershenzon, 2006). Levy et al., (2005) showed that the expression level of the IQD1 (At3g09710) gene, which encodes a transcription factor that is responsible for glucosinolate accumulation, impacts host plant choice by GPA. Another study showed that the fecundity of both generalist (GPA) and specialist aphid (cabbage aphid) were higher on Arabidopsis coi1 mutant that contained lower amount of glucosinolates than the wild type plant (Mewis et al., 2005). Differences have been observed in the profile of glucosinolates in aphid-infested compared to uninfested Arabidopsis. For example, although the total glucosinolate content does not change in GPA-infested (Kim and Jander, 2007, Louis et al., 2010a), compared to uninfested plants, the GPA-infested plants contain higher levels of indole glucosinolates. Kim et al., (2008) reported that GPA population size was smaller on the Arabidopsis atr1D (At5g60890) mutant plant, which accumulates elevated levels of indole glusocinolates, thus suggesting that indole glucosinolates are detrimental to GPA (Kim et al., 2008).

Glucosinolates themselves are not insecticidal. However, when acted upon by myrosinases, glucosinolates produce toxic thiocyanates and other breakdown products that act as defensive compounds against insects (Chew, 1988; Louda and Mole 1991; Rask et al., 2000). In Arabidopsis, β-thioglucoside glucohydrolases encoded by TGG1 and TGG2 (At5g25980 and At5g26000) contribute to the majority of the myrosinase activity (Barth and Jander, 2006). However, both GPA and cabbage aphid populations were unaffected on Arabidopsis tgg1 and tgg2 single and the ttg1 ttg2 double mutant plants, compared to the wild type plant, suggesting that aphids evade the production of toxic thiocyanates or modulate the activity of enzymes that synthesize these thiocyanates (Barth and Jander, 2006). Instead, Kim et al., (2008) showed that the adverse effect of glucosinolates on aphid performance correlated with the accumulation of indole glucosinolate breakdown products in the insect body, indicating that aphids consume glucosinolates produced by the host plant. In particular, diindolylmethylcysteines and other amino acid conjugates, which form after indole glucosinolate breakdown during aphid feeding from Arabidopsis, reduce aphid reproduction on artificial diets (Kim et al., 2008). Further studies identified the indol-3-ylmethyl glucosinolate -derived 4-methoxyindol-3-ylmethyl glucosinolate and 1-methoxyindol-3-yl-methyl glucosinolate as strong deterrents of GPA proliferation (Kim and Jander, 2007; Pfalz et al., 2009). GPA reproduction was improved on cyp81F2 (At5g57220) mutants, which are defective in the production of 4-methoxyindol-3-ylmethyl glucosinolate (Pfalz et al., 2009; De Vos and Jander, 2009). Unlike in the generalist GPA, the effect of glucosinolate breakdown products against the specialist cabbage aphids have not been reported. It is possible that the specialist aphids are able to evade, suppress and/or adapt to these glucosinolate breakdown products as opposed to the generalist aphids. Readers are directed to some excellent reviews that have summarized glucosinolate biosynthesis and metabolism, and its role in plant defense (Bednarek et al., 2009; Wittstock and Burow, 2010).

Non-protein Amino Acids

Plants produce several non-protein amino acids that serve as intermediates in the synthesis of primary metabolites. In addition, these compounds are also reported to have defense related functions against insects (Rosenthal, 1991). For instance, L-canavanine, an L-arginine analog, is a major storage compound in legumes and also has insecticidal allelochemical properties (Rosenthal, 2001). -acetylornithine has been identified as a new class of defense related compound in Arabidopsis (Adio et al., 2011). This non-protein amino acid has been identified in phloem sap collected from methyl jasmonate-treated Arabidopsis. The Arabidopsis NATA1 gene (At2g39030), which encodes a protein with N-acetyltransferase activity, is involved in the biosynthesis of -acetylornithine and is expressed in the phloemassociated tissues. Expression of the NATA1 gene was induced and -acetylornithine content was higher in GPA infested Arabidopsis compared to uninfested plants. Furthermore, GPA population on aphid diet containing -acetylornithine was significantly reduced, suggesting that this compound has a direct toxic and/or deterrent effect on GPA (Adio et al., 2011). Transient expression of NATA1 in tobacco significantly reduced GPA population size as compared to the vector control plants (Adio et al., 2011). Resistance in these experiments correlated with the level of -acetylornithine. Whether -acetylornithine has any effect on specialist cabbage aphids, is not known. However, by contrast to aphids, although NATA1 expression and -acetylornithine accumulation were also induced in response to infestation by chewing herbivores, the nata1-1 mutation did not affect Pieris rapae (white cabbage butterfly) and Plutella xylostella (diamondback moth) caterpillar growth, suggesting that -acetylornithine accumulation is either not important or not sufficient to deter chewing insects. A recent review by Huang et al. (2011) summarizes the role of non-protein amino acids in plant defense against various insect pests.

9-LOX-derived oxylipins

Recently, it was shown that 9-LOX-derived oxylipins contribute to host susceptibility to aphids (Nalam et al., 2012). In Arabidopsis, GPA infestation resulted in an increase in the level of 9-LOX-derived oxylipins in the petiole exudates (Nalam et al., 2012). Genetic and physiological evidence indicates that GPA cues in on 9-LOX pathway derived oxylipins in Arabidopsis to facilitate infestation (Nalam et al., 2012). GPA population was significantly reduced on Arabidopsis lox5 (At3g22400) mutants, in which the accumulation of 9-LOX-derived oxylipins is attenuated, as compared to the wild type plants. EPG studies indicated that insects on the lox5 mutant had difficulty feeding from sieve elements and xylem tissues. This reduction in feeding activity of GPA on the lox5 mutant also resulted in a reduction in water content in the aphids. Application of the 9-LOX-derived 9-hydroxyoctadecadienoic acid (9-HOD) restored water content and insect population size on the lox5 mutant, thus confirming that 9-LOX products have an important contribution in facilitating insect feeding. 9-HOD and 9-hydroperoxyoctadecadienoic acid (9-HPOD) when added to an artificial diet enhanced GPA population size, suggesting that 9-LOX-derived oxylipins are susceptibility factors that facilitate colonization of GPA on Arabidopsis. Micro-grafting experiments (wild type LOX scions and lox5 mutant rootstock, and vice-versa) demonstrated that the oxylipins that promote GPA susceptibility in Arabidopsis leaves are root-derived (Nalam et al., 2012). GPA infestation of Arabidopsis foliage was found to induce expression of LOX5 and promote accumulation of 9-LOX products in roots from where they are likely transported to shoots via the vasculature. Experiments in potato plants (Solanum tuberosum) have also indicated that accumulation of 9-LOX products is increased in GPA infested plants (Gosset et al., 2009).

MPL1

In Arabidopsis a lipase encoded by the MPL1 (MYZUS PERSICAE INDUCED LIPASE1; At5g14180) gene, is required for the accumulation of an activity in petiole exudates that is detrimental to GPA (Louis et al., 2010a). Loss of this antibiosis activity in the petiole exudates of the mpl1 mutant was accompanied by larger population size of GPA on the mpl1 mutant compared to wild type Arabidopsis plants. Loss of MPL1 function in the mpl1 mutant did not impact insect feeding behavior. MPL1 expression is induced in GPA-infested plants and is constitutively elevated in the ssi2 mutant, which exhibits enhanced resistance to GPA (Louis et al., 2010a, 2010b). Indeed, MPL1 function was required for the ssi2-determined resistance against GPA and accumulation of antibiosis activity in petiole exudates. Furthermore, constitutive overexpression of MPL1 from the Cauliflower mosaic virus 35S promoter resulted in enhanced resistance against GPA, suggesting that the induction of MPL1 expression is important for controlling GPA infestation (Louis et al., 2010a). Constitutive overexpression of PAD4 (described later) and MPL1 in mpl1 and pad4 plants, respectively, rescued the antibiosis deficiency of the mpl1 and pad4 mutants, suggesting that MPL1 and PAD4 contribute to two parallel antibiosis mechanisms and the elevated levels of one component/mechanism, can overcome the deficiency of the other.

Fatty Acid Desaturases

The stearoyl-ACP desaturase activity encoded by the Arabidopsis SSI2 gene catalyzes the desaturation of stearic acid to oleic acid (Shah et al., 2001; Kachroo et al., 2001) and thus contributes to Arabidopsis membrane lipid composition (Nandi et al., 2003). The ssi2 mutant plants constitutively accumulate high levels of SA, which is responsible for the heightened resistance of the ssi2 mutant to some pathogens (Shah et al., 2001; Kachroo et al., 2001). In comparison to the wild type plant, GPA reproduction was lower on the Arabidopsis ssi2 mutant plant as compared to its wild type plant (Pegadaraju et al., 2005). However, GPA growth was comparable on the ssi2 single mutant and the ssi2 nahG (SA-deficient) plant, suggesting that the high levels of SA present in the ssi2 plants do not contribute to the ssi2-conferred resistance to GPA. In addition, GPA counts on the suppressor of npr1-1, constitutive 1 (snc1; At4g16890) mutant, which accumulates high levels of SA (Zhang et al., 2003), was comparable to those on the wild type plant (Pegadaraju et al., 2005). These results suggested that some factor(s) other than SA contributes to the ssi2-conferred hyper-resistance against GPA. Petiole exudates collected from uninfested ssi2 mutant contain elevated levels of antibiosis activity against GPA, than similar exudates collected from the wild type plant. Whether this antibiosis factor is a lipid is not known. However, the ssi2-determined enhanced antibiosis and resistance to GPA was attenuated in the absence of MPL1 activity in the ssi2 mpl1 double mutant (Louis et al., 2010b). Phloem sap contains lipids (Madey et al., 2002; Nalam et al., 2012), and the salivary glands of phloem-feeding insects contain putative lipases (Shukle et al., 2009), thus suggesting that these insects likely encounter lipids in their diet, one or more of which could be detrimental to GPA.

More recently, loss of FAD7 and FAD8 (At3g11170 and At5g05580)-encoded fatty acid desaturases in Arabidopsis and the LeFAD7 activity in tomato (Solanum lycopersicum) were also shown to result in enhanced resistance against GPA and potato aphid (Macrosiphum euphorbiae), respectively (Avila et al., 2012). In Arabidopsis, FAD7 and FAD8 catalyze the desaturation of dienoic acyl chains in plastid galactolipids to yield lipids with thenoic fatty acyl chains. FAD7/FAD8 synthesized thenoic acyl chains are substrates for the synthesis of signaling oxylipids like OPDA and JA. The tomato fad7 plants are deficient in JA, yet they are more resistant to potato aphid. Similarly, the Arabidopsis fad7 fad8 mutant was more resistant to GPA. These results suggest that the impact of fad7 on aphid colonization is independent of fad7's effect on JA. Indeed, potato aphid numbers were comparable on wild type plants and in tomato mutants impaired in JA synthesis (acx1) or perception (jai1-1) (Avila et al., 2012), thus further supporting the suggestion that the impact of fad7 mutation on aphids is unrelated to FAD7's role in JA accumulation. In tomato, loss of FAD7 function resulted in increased antibiotic and antixenotic resistance to potato aphids, which was dependent on the accumulation of elevated SA levels in the fad7 tomato plants as evident from experiments with fad7 NahG plants in which attenuation of SA accumulation by expression of the NahG-encoded salicylate hydroxylase, resulted in suppression of fad7-conferred enhanced resistance to potato aphid. Whether, fad7 fad8 determined enhanced resistance against GPA in Arabidopsis is similarly dependent on SA remains to be determined, although as mentioned above, in Arabidopsis SA is not required for basal resistance against GPA (Moran and Thompson, 2001; Pegadaraju et al., 2005).

PAD4

In Arabidopsis, the PAD4 (PHYTOALEXIN DEFICIENT4) gene modulates antibiotic and antixenotic defense against GPA (Pegadaraju et al., 2005, 2007). In addition, PAD4 promotes premature leaf senescence and SAG13, SAG21 and SAG27 (At2g29350, At4g02380, and At4g44300) expression in GPA-infested plants (Pegadaraju et al., 2005). PAD4 was required for deterring insect settling on plants (Pegadaraju et al., 2007) and the accumulation of a factor in petiole exudates of uninfested plants that was detrimental to GPA (Louis et al., 2010b, 2012). In addition, EPG analysis indicated that PAD4 controls insect feeding from sieve elements (Pegadaraju et al., 2007). PAD4 is also involved in plant defense against a subset of pathogens, where it modulates the synthesis of SA and the phytoalexin, camalexin (Glazebrook et al., 1997; Zhou et al., 1998; Jirage et al., 1999). However, genetic studies have suggested that the involvement of PAD4 in plant defense against GPA is not due to its involvement in SA and camalexin metabolism (Pegadaraju et al., 2005). The N-terminal half of PAD4 shares homology with α/β-fold acyl hydrolases that include lipases and esterases (Zhou et al., 1998; Jirage et al., 1999). Although, the biochemical function of PAD4 is not known, at the molecular level PAD4 protein interacts with its signaling partner EDS1 (ENHANCED DISEASE SUSCEPTIBILITY1; At3g48090) thereby promoting defense against pathogens (Falk et al., 1999; Feys et al., 2005). EDS1, like PAD4, has homology to lipases/acylhydrolases, and like PAD4, EDS1 expression is also induced in GPA-infested leaves (Feys et al., 2005; Pegadaraju et al., 2007). However, PAD4-mediated defense against GPA does not require EDS1 (Pegadaraju et al., 2007; Louis et al., 2012), suggesting that PAD4s involvement in Arabidopsis defense against aphids is distinct from its involvement in defense against pathogens.

Mutational analysis of PAD4 indicated that Ser118, which is a catalytic residue in α/β fold acyl hydrolases, although not required for PAD4's involvement in defense against pathogens, is required for a subset of PAD4 activities in defense against GPA (Louis et al., 2012). It was essential for limiting insect feeding from sieve elements and for accumulation of the antibiosis factor in petiole exudates of uninfested plants (Louis et al., 2012). However, Ser118 is not required for controlling insect settling and premature senescence in GPA-infested plants. These results suggest that distinct molecular activities of PAD4 modulate different functions in Arabidopsis defense against GPA.

In Arabidopsis, PAD4 also modulates accumulation of camalexin (Tsuji et al., 1992; Rogers et al., 1996). Camalexin levels increase in GPA-infested Arabidopsis leaves. However, there was no significant difference in camalexin content in GPA infested leaves of pad4 mutant and wild type plants (J Louis, J Keerantweep and J Shah, unpublished data), suggesting that PAD4 is not required for camalexin accumulation in GPA-infested Arabidopsis. Camalexin synthesis also requires the PAD3 (At3g26830) gene, which encodes an enzyme that catalyzes the terminal step in the synthesis of camalexin (Schuhegger et al., 2006). PAD3 expression is induced in GPA-infested plants (Pegadaraju et al., 2005). However, GPA population size on the pad3 mutant was comparable to that on the wild type plant (Pegadaraju et al., 2005), thus indicating that PAD3 and camalexin are not essential for controlling GPA infestation on Arabidopsis. Cabbage aphid infestation also results in an increase in camalexin accumulation (Kuśnierczyk et al., 2008). PAD3 expression was induced in response to cabbage aphid infestation (Kuśnierczyk et al., 2008). Furthermore, fecundity of cabbage aphid was significantly higher on the pad3 mutant than on the wild type plant (Kuśnierczyk et al., 2008), thus indicating that unlike GPA infestation, camalexin accumulation has a role in controlling cabbage aphid infestation on Arabidopsis. Comparative analysis of EST sequences from the GPA with the whole genome sequence of the specialist pea aphid revealed that a generalist aphid like GPA has more detoxification enzymes, including cytochrome P450 monooxygenases, glutathione S-transferases, and carboxy/cholinesterases, likely because GPA encounters more diverse host plant species (Ramsey et al., 2010). Presence of these detoxification mechanisms could explain the lack of any obvious effect of camalexin on GPA.

TPS11 and Trehalose

Trehalose is a non-reducing disacchahde composed of two molecules of glucose. It is present in a wide spectrum of living organisms, as varying as bacteria, fungi, insects and angiosperms. Singh and co-workers (2011) recently demonstrated that transient accumulation of trehalose, dependent on the TPS11-encoded trehalose-6-phosphate synthase, modulates Arabidopsis defense against GPA. TPS11 was required for the timely induction of PAD4 in GPA-infested leaves. In addition, TPS11 was required for promoting starch accumulation at the expense of sucrose in GPA-infested leaves. As mentioned earlier, genetic studies have indicated that starch accumulation in Arabidopsis contributes to controlling GPA infestation. Petiole exudates of the tps11 mutant lacked the antibiosis activity that was present in petiole exudates of wild type plants. In addition, EPG analysis indicated that GPA spent more time in sieve element phase on the tps11 mutant, and given a choice the insects preferred to settle on the tps11 mutant than the wild type plant. Trehalose application restored wild type level of resistance against GPA in the tps11 mutant, indicating that the tps11 mutant phenotypes are due to its inability to accumulate elevated levels of trehalose in response to GPA infestation. The role of trehalose in defense against GPA is further evident from experiments with the trehalose hyper-accumulating tre1 (At4g24040) mutant, which contains a T-DNA insertion in the trehalase encoding TRE1 gene. GPA population was smaller on the tre1 mutant, compared to the wild type plant. Similarly, GPA population was also smaller on transgenic plants expressing the bacterial otsB gene, which encodes a trehalose-6-phosphate phosphatase involved in trehalose synthesis. Singh et al. (2011) suggested that TPS11 provides a threshold level of ‘signaling’ trehalose that is essential for regulating Arabidopsis defense against GPA, including the up-regulation of PAD4 expression and full extent of starch accumulation. Trehalose application promotes starch accumulation in Arabidopsis leaves (Wingler et al., 2000; Kolbe et al., 2005; Singh et al., 2011). As shown in other studies, the ability of trehalose to promote starch accumulation in GPA-infested plants could be due to the inhibition of starch turnover (Ramon et al., 2007) and/or the redox activation of AGPase, a key enzyme in starch synthesis, by trehalose-6-phosphate (Paul et al., 2008).