Nymphs of the peach-potato aphid (Myzus persicae) were obtained from the James Hutton Institute, Invergowrie, Dundee, Scotland. The aphids were asexual clones of a wild population isolated in Scotland (Kasprowicz et al., 2008). Aphid colonies were maintained on potato plants (S. tuberosum L. cv. Désirée) inside a mesh cage in a containment glasshouse at 20–22°C under a 16 h/8 h light/dark cycle. Only apterous (wingless) aphids were used and transferred to experimental plants using a fine paintbrush.

Cysts of G. pallida were extracted from infected soil stocks using the Fenwick can method (Fenwick, 1940). Infective second-stage juveniles (J2s) were hatched from the cysts following treatment with 1% sodium hypochlorite aqueous solution (Heungens et al., 1996). J2 nematodes were stored in autoclaved tap water at 10°C and their viability was checked prior to use by observation using a stereo binocular microscope.

Potato tuber cuttings (S. tuberosum L. cv. Désirée) were planted in 18 cm pots containing pesticide-free compost. Growth took place in a glasshouse at 20–22°C under a 16 h/8 h light/dark cycle for a period of 3 weeks. For potato plants infected with nematodes only, 10,000 J2 nematodes suspended in 6 mL of autoclaved tap water were introduced into the compost around the roots of each potato plant. Uninfected potato plants used as a control were mock-inoculated with autoclaved tap water. At 14 days post inoculation (dpi), a fully expanded terminal leaf from the top of each plant was excised using fine tweezers, divided into three samples for RNA, SA, and JA extractions and immediately snap frozen in liquid nitrogen. Five-week-old potato plants were used for infection with aphids alone so ensuring each set of experimental plants were the same age. Twenty apterous aphids of various life-stages were transferred to the second fully expanded leaf with a fine paintbrush and confined to the abaxial surface of the leaf in a 2.5 cm diameter clip-cage. Aphid-free clip-cages were used in control experiments. After 48 h, aphids were carefully removed and the leaf was excised and sampled as previously described. Co-infected potato plants were initially inoculated with ten thousand J2 nematodes, then 14 days later 20 apterous aphids were applied to either infected or control plants for 48 h as previously described. Co-infected samples were collected 48 h post infection (hpi) with aphids.

Total RNA was prepared from frozen leaf tissue of control and infected potato plants using the RNeasy^® Plant Mini Kit (Qiagen, Inc., Valencia, CA, United States). First-strand cDNA was synthesized from 1000 ng RNA using SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA, United States) and oligo(dT)₁₇ primer (500 μg/ml) following the manufacturer’s instructions. Quantitative reverse transcriptase (qRT)-PCR was carried out on the resulting cDNA using Brilliant III Ultra-Fast SYBR^® Green Master Mix and a Mx3005P (v. 4.10) instrument (Agilent Technologies, La Jolla, CA, United States). Genes for expression analysis were selected according to their previously recorded involvement in biotic stress responses (Kombrink et al., 1988; Matton and Brisson, 1989; Fidantsef et al., 1999; Reiss and Horstmann, 2001; Wang et al., 2005) (see Results section for further details). Potato ELONGATION FACTOR 1-α was used to normalize the results (Nicot et al., 2005). Sequences of primers used for amplification of each gene are detailed in Supplementary Data Table 1. Sequences for the chosen genes were found on the National Center for Biotechnology Information website¹ and primers were designed using the online Primer 3 software². Controls for qRT-PCR included reactions containing no template. All primer pairs had an amplification efficiency of 93–101% and R² correlation coefficients for standard curves ranged between 0.94 and 0.99. qRT-PCR was performed on five biological replicates for control and infected samples and each reaction was carried out in triplicate. Ct values were determined using the MxPro software. Relative expression between control and infected samples was determined using the 2^-ΔΔC_t method (Livak and Schmittgen, 2001).

Salicylic acid extraction was performed on leaf tissue that had been treated with aphids and nematodes both singly and in combination using a modified protocol derived from Raskin et al. (1989). One milliliter of methanol (90%) was added to ground, frozen leaf tissue, and the resulting mixture was vortexed for 1 min followed by sonication in a bath for 5 min. After centrifugation for 5 min at 14,104 × g, the supernatant was collected and the pellet was re-extracted with 500 μl methanol (100%), vortexed for 1 min, re-sonicated for 5 min, and re-centrifuged at 14,104 × g for a further 5 min. Both supernatants were combined and dried using a GeneVac (EZ-2 series). For free SA quantification the dried samples were re-suspended in 250 μl of 5% trichloroacetic acid (TCA) and vortexed. The sample was extracted twice in cyclohexane and ethyl acetate (1:1), vortexed vigorously and centrifuged at 14,104 × g for 1 min. The top organic phase was removed and dried using a GeneVac (EZ-2 series). The remaining phase was subjected to acid hydrolysis using 8M HCl and incubated at 80°C for 1 h to quantify sugar-conjugated (or stored) SA. The sugar-conjugated (or stored) SA sample was extracted twice in cyclohexane and ethyl acetate (1:1), vortexed vigorously and centrifuged at 14,104 × g for 1 min. The top organic phase was removed and dried using a GeneVac. The pooled stored SA extract was re-suspended in 600 μl of water and acetonitrile (95:5) and quantified by high-pressure liquid chromatography (HPLC). Analysis was performed using a Supelcosil^TM LC-18 column (250 mm × 4.6 mm, 5 μm). An injection volume of 20 μl was separated under isocratic conditions using a mobile phase of water, acetonitrile (HPLC grade) and formic acid (60:40:0.1) at a flow rate of 1 ml/min. SA was detected using a Dionex RF 2000 Fluorescence Detector operated at an emission wavelength of 400 nm and an excitation wavelength of 303 nm, respectively. SA was determined and quantified by comparing peaks of recovered SA using calibration standards. Total SA was calculated as the amount of free SA in plant samples plus the amount of sugar-conjugated (or stored) SA in plant samples. The efficiency of SA recovery was calculated by using a deuterium-labeled internal standard of SA-d₆. Twelve biological replicates were used for each condition analyzed.

Leaf tissue was harvested as previously described. The samples were ground into a powder in a Tissue Lyser LT (Qiagen, Hilden, Germany) and 1 ml extraction solvent (methanol/H₂O/formic acid; 80:19:1, v/v/v) was added and mixed. Samples were sonicated at 4°C for 5 min, agitated for 30 min at 4°C and centrifuged at 12,000 × g for 10 min at 4°C. The extraction procedure was repeated with 500 μl solvent and the supernatants were combined. JA was analyzed on a UPLC AxION 2 TOF MS system coupled with an Altus SQ Detector (Perkin Elmer, United Kingdom). For the chromatographic separation the solvents were 0.1% HCO₂H in ultrapure water (A) and 0.1% HCO₂H in methanol (B), the column was a C18 100 X 1.2 mm (Perkin Elmer, United Kingdom) and the flow rate was set at 0.35 ml min^-1. The binary analytical gradient used was as follows: 0 min, 1% B; 20 min, 100% B; 22 min, 100% B; 25 min, 1% B. The compound quantification was assured by calibration curve standards in the range of 5–50 ng/ml. The data analysis was performed using Empower 3 software (Waters, United Kingdom).

To test the effect of G. pallida infection on aphids, 10 apterous adults were placed in a 2.5 cm diameter clip cage on a fully expanded, terminal leaf second from the top of a potato plant pre-infected with 10,000 J2 nematodes 14 days previously or mock-inoculated with water. After 24 h all aphids except for five nymphs were removed. The five nymphs were allowed to develop and the number of aphids inside the clip-cage were counted for 8 days to determine the abundance of aphids on nematode-infested plants and non-infected control plants. Five biological replicates for each condition were used in the experiment.

The effects of the treatments on gene expression and the levels of endogenous phytohormones JA and SA were determined using a Mann–Whitney U test. A Mann–Whitney U test was also carried out to compare the abundance of aphids on nematode infected plants against non-infected control plants.

There was a significant increase in endogenous SA in the leaves of potato plants 14 days after infection with G. pallida. The level of free SA was significantly greater in nematode-infected plants compared to non-infected control plants (mean ± standard error), 571.33 ± 70.09 ng/g FW for infected plants and 231.20 ± 27.21 ng/g FW for control plants (Mann–Whitney U = 497.5, P = 0.001, sig ≤ 0.05, 2-tailed) (Figure 1A). The presence of nematodes also significantly increased total levels of SA in leaves of potato plants, (4541.42 ± 268.2 ng/g FW for nematode-infected plants and 2132.77 ± 758.57 ng/g FW for control plants, P ≤ 0.01) (Figure 1A). These results suggest an activation of the SAR pathway in the leaves of potato plants, which is mediated by SA (Gaffney et al., 1993).

An elevated level of the endogenous phytohormone SA is known to lead to the expression of pathogen-related (PR) genes, some of which are commonly used molecular markers of SAR (Uknes et al., 1993; Bowling et al., 1994; Cao et al., 1994). We therefore measured the expression of PR-1, PR-2, and PR-5, all of which are coordinately regulated by SA (Cao et al., 1994), in nematode-infected plants 14 dpi. Transcripts of all three PR-genes were detected in leaf tissue from both infected and non-infected potato plants. However, only the expression of PR-5 was significantly induced in nematode infected plants (Mann–Whitney U = 1.000, P = 0.027) (Figure 1C). Transcripts of PR-5, which encodes a thaumatin-like protein, were approximately three-fold higher in nematode-infested plants relative to control plants (Figure 1C).

Five-week-old potato plants infected with aphids were analyzed for endogenous SA and the expression of SA-mediated defense genes. There was a significant increase in free (686 ± 76 ng/g FW, P ≤ 0.001), stored (7010 ± 547 ng/g FW, P ≤ 0.001) and total (8046 ± 555 ng/g FW, P ≤ 0.001) SA in the leaves of potato plants infected with aphids compared to control plants (Free: 276 ± 32 ng/g FW; Stored: 3581 ± 392 ng/g FW; Total: 4055 ± 396 ng/g FW) (Figure 2A). The expression of SA-mediated genes PR-1 (P ≤ 0.001) and PR-5 (P ≤ 0.001) was also significantly elevated. There was no significant increase in PR-2 expression (Figure 2C).

In addition to SA-mediated effects, it is well established that JA has an important role in the plant defense pathway. Hence, we also measured endogenous levels of JA as well as transcript levels of JAZ-1, which is a nuclear-localized protein involved in jasmonate signaling in addition to PR-3. There was a significant increase in endogenous JA in the leaves of plants infected with aphids (729 ± 22 ng/g FW) compared to control plants (356 ± 88 ng/g FW) (P ≤ 0.025) (Figure 2B). In addition there was a significant increase in transcript levels of PR-3 (P ≤ 0.001) and JAZ-1 (P ≤ 0.001) (Figure 2C). However, there was no significant increase in endogenous levels of the phytohormone JA in nematode-infected plants 14 dpi (Mann–Whitney U = 66.000, P = 0.76, sig ≤ 0.05, 2-tailed) (Figure 1B) or in the expression of genes involved in the signaling of JA, PR-3 (P ≤ 0.11) or JAZ-1 (P ≤ 0.286) (Figure 1C) suggesting that nematode infection does not elicit a systemic JA defense response in the leaves of potato plants.

The SA-mediated defence pathway was investigated in the leaves of potato plants that had been infected with both G. pallida and M. persicae. There was a significant increase in the levels of stored (9943 ± 1522 ng) and total SA (10750 ± 1557 ng) in the leaves of dual infected plants compared to the controls (Stored: 4665 ± 906 ng; Total: 5409 ± 930 ng; P ≤ 0.012) (Figure 3A). There was no significant difference in the levels of free SA in the leaves of plants that were co-infected (691 ± 45 ng) compared to the controls (743 ± 146 ng) (Figure 3A). There was no significant increase in transcript levels of SA-mediated defense genes (Figure 3A). The significant increase in the levels of stored SA indicates that the SA-mediated defense pathway is up-regulated in the leaves of potato plants; however, it has not been converted into free SA.

There was no significant change in the level of endogenous JA in plants that had been co-infected with both pests (372 ± 73 ng) compared to the controls (392 ± 64, P ≤ 0.855) (Figure 3B). Similarly, when the expression of genes involved in the JA signaling pathway were analyzed, there were no significant differences between the leaves of co-infected plants and control plants (Figure 3C). Due to a significant increase in endogenous levels of JA and the expression of SA-mediated defenses in the leaves of plants infected with aphids only, the reduction of JA in co-infected plants may indicate an antagonistic suppression of JA by the additive increase in SA caused by both nematode and aphid infection together.

There was a significant increase in the abundance of aphids reared on potato plants pre-infected with nematodes for 14 days compared with aphids reared on non-infected control plants (Mann–Whitney U = 3.000, P = 0.011, sig ≤ 0.05, 2-tailed) (Figure 4).

Previous studies have revealed that defense signaling pathways are involved in compatible interactions of plants with cyst nematodes (Heterodera and Globodera spp.) (Jammes et al., 2005; Ithal et al., 2007; Wubben et al., 2008). Similarly, it is well known that many plant defense signaling pathways are up-regulated in response to aphid feeding (De Vos et al., 2005; Kuśnierczyk et al., 2008; Broekgaarden et al., 2011). Our analysis has shown that expression of PR-5, a molecular marker commonly used to indicate activation of SAR (Uknes et al., 1993; Bowling et al., 1994), was significantly increased in leaves of potato plants following infection with G. pallida for 14 days and also in the leaves of 5-week-old plants infected with M. persicae for 48 h. This correlates with the significant increase in free and total SA in leaves of potato plants: the accumulation of the phytohormone SA is required for the activation of SAR in distal tissues of the infected plant (Gaffney et al., 1993). Taken together these results indicate activation of a SAR-induced potato defense pathway following parasitism by G. pallida and infection with M. persicae singly. There was no significant increase in the expression PR-1 or PR-2 in the leaves of nematode-infected potato plants at the time-point examined. Expression of the orthologous genes was reported to increase in the leaves of Arabidopsis thaliana in response to cyst nematode infection, however, this increase was transient and varied considerably between investigations (Wubben et al., 2008; Hamamouch et al., 2011). The length of time post-infection, together with the initial nematode burden, may be critical in determining if PR-gene induction is observed. It is well documented that there is mutual antagonism between SA and JA signaling pathways (Pieterse et al., 2012), therefore the phytohormone JA and the expression levels of the JA-dependent associated genes PR-3 and JAZ-1, a nuclear-localized protein involved in jasmonate signaling (Thines et al., 2007) were quantified. No significant differences were found between nematode-infected plants and control plants in either the amount of JA or the expression of PR-3 and JAZ-1, suggesting that infection with the potato cyst nematode does not alter the JA signaling pathway in the potato plant at 14 dpi. Alternatively, this could indicate antagonistic cross-talk between the SA and JA pathways following infection with G. pallida, as both endogenous SA and the expression of PR-5 was significantly up-regulated. In contrast, it was found that aphid infection induced the JA signaling pathway in the leaves of potato plants as both JA and the expression of PR-3 and JAZ-1 were significantly up-regulated compared to control plants.

Co-infection of the potato with both G. pallida and M. persicae had a different and unique impact on the levels of endogenous phytohormones and expression of defense-related genes compared to plants that had been infected with each pest singly. An additive effect on SA was observed in co-infected plants, an effect that may be assumed when two pests are applied to a plant. However, a reduced JA effect was noted in dual infected plants even though JA was present in the leaves of plants infected with aphids in isolation. There is literature to suggest that phytohormones do not act independently of one another. The interaction between SA and JA is complex with the main interaction between these two pathways being mutual antagonism (Kunkel and Brooks, 2002). SA has been shown to have an inhibitory effect on JA in tomato (Doherty et al., 1988; Pena-Cortés et al., 1993) and in Arabidopsis (Clarke et al., 2000; Gupta et al., 2000). Therefore, a lack of JA in the leaves of co-infected plants could be construed as antagonistic crosstalk because although infection with plant-parasitic nematodes did not elicit the JA defense pathway in potato plants, infection with aphids alone did.

Plant-mediated interactions between plant parasitic nematodes and aerial pests studied to date have been variable: susceptibility to shoot pathogens and resistance to phloem feeders have been reported with the outcome depending on the parasitic strategy of the nematode involved in the interaction (Biere and Goverse, 2016). To the best of our knowledge there have been no studies of plant-mediated interactions between the potato cyst nematode and specialized above-ground pests or pathogens of potato, however, there have been reports of interactions between G. rostochiensis and below-ground pathogens such as the soil-borne fungus of potato, Rhizoctonia solani (Back et al., 2006). A reduced aphid performance was reported when Plantago lanceolata (Wurst and van der Putten, 2007) was infected with the migratory nematode, Pratylenchus penetrans. Similarly, a decrease in the fecundity of aphids was observed when Agrostis capillaris was infected with a mixture consisting of ectoparasites and migratory endoparasites (Bezemer et al., 2003). Reports using sedentary endoparasites have found negative or neutral impacts on aphids. An infection of H. schachtii on B. oleracea resulted in reduced growth and fecundity of a specialist aphid species, Brevicoryne brassicae as well as a generalist species, M. persicae (Hol et al., 2013). However, in another study using a mix of different parasitic nematode species, no effect on the performance of B. brassicae was found (Kabouw et al., 2011). Our observation that G. pallida, a sedentary endoparasitic nematode, indirectly and positively influences the abundance of M. persicae highlights how aphids may be more damaging to the potato crop in areas where G. pallida is present compared to such areas where there is no infection, however, this requires further investigation. Our study is in contrast to these previous studies and to our knowledge is the first to report the combined molecular and biochemical response of the potato to nematode infection.

Systemic plant resistance to insect herbivores is mediated by the SA and JA wound signaling pathways and the, usually antagonistic, crosstalk between them (Pieterse et al., 2012; Stam et al., 2014). In addition to their role in regulating resistance to biotrophic pathogens, SA-mediated defensive pathways are known to be induced by phloem-feeding insects, and there have also been reports suggesting that SA itself is an effective chemical defense against phloem-sucking herbivory animals (Kaloshian and Walling, 2005; Donovan et al., 2013). As expected, we found induction of the SA pathway in response to nematodes, but any adverse effects of this on the aphids are likely to be negated by the benefits of SA-mediated reductions of the JA-mediated pathway responsible for plant resistance to herbivores (Lazebnik et al., 2014). Indeed, aphids are believed to circumvent the plant’s immune system by eliciting the SA signaling pathway in order to antagonize and suppress the JA one, which is important in mediating resistance to phloem feeders (Ellis et al., 2002; Zhu-Salzman et al., 2004). Thus, our observation of more aphids present on nematode infested plants could reflect circumvention of the SA-mediated defense pathway of the potato plant by M. persicae. Our analysis of the JA-mediated defense pathway in the potato plant showed no up-regulation of endogenous JA or expression of PR-3 or JAZ-1 in leaves of potato plants infected with nematodes when compared to control plants. Aphids could benefit from the situation in which the hormone has not been elicited or even suppressed.