Systemic resistance in citrus to Tetranychus urticae induced by conspecifics is transmitted by grafting and mediated by mobile amino acids

Abstract

Recent research suggests that systemic signalling and communication between roots and leaves plays an important role in plant defence against herbivores. In the present study, we show that the oviposition of the two-spotted spider mite Tetranychus urticae in the systemic leaves of citrus rootstock Citrus aurantium (sour orange) was reduced by 50% when a lower leaf was previously infested with conspecifics. Metabolomic and gene expression analysis of the root efflux revealed a strong accumulation of glutamic acid (Glu) that triggered the expression of the citrus putative glutamate receptor (GRL) in the shoots. Additionally, uninfested sour orange systemic leaves showed increased expression of glutamate receptors and higher amounts of jasmonic acid (JA) and 12-oxo-phytodienoic acid in plants that were previously infested. Glu perception in the shoots induced the JA pathway, which primed LOX-2 gene expression when citrus plants were exposed to a second infestation. The spider mite-susceptible citrus rootstock Cleopatra mandarin (C. unshiu) also expressed systemic resistance, although the resistance was less effective than the resistance in sour orange. Surprisingly, the mobile signal in Cleopatra mandarin was not Glu, which suggests a strong genotype-dependency for systemic signalling in citrus. When the cultivar Clemenules (C. clementina) was grafted onto sour orange, there was a reduction in symptomatic leaves and T. urticae populations compared to the same cultivar grafted onto Cleopatra mandarin. Thus, systemic resistance is transmitted from the roots to the shoots in citrus and is dependent on rootstock resistance.

Keywords: Tetranychus urticae.; Citrus; glutamate-receptor like; glutamic acid; grafting; systemic resistance.

Fig. 1.

Effect of SR treatments of Cleopatra mandarin and sour orange on spider mite oviposition rates. Half of the plants were infested with 10 T. urticae adult females (SR treatments). Three days later, all of the plants were infested with 2-day-old T. urticae adult females on distal clean leaves. Egg number was determined 3 days after the second infestation. Different letters indicate significant differences between the treatments (ANOVA; LSD test, P < 0.05). The figure shows the average results of three biological replicates (n = 9). Cleo con, Cleopatra mandarin uninfested; Cleo SR, Cleopatra mandarin previously infested; SO con, sour orange uninfested; SO SR, sour orange previously infested.

Fig. 2.

PCA of the metabolomic fingerprint of sour orange and Cleopatra mandarin leaf efflux following SR treatments and T. urticae infestation. Non-supervised PCA representation of major sources of variability of ESI⁻ (A) and ESI⁺ (B) signals obtained from a non-targeted analysis by UPLC-Q-TOF-MS to monitor metabolomic changes during spider mite infestation. Four different sets of samples were tested: sour orange uninfested (SO con), sour orange previously infested (SO SR), Cleopatra mandarin uninfested (Cleo con), and Cleopatra mandarin previously infested (Cleo SR). Twelve-week-old plants were infested with 10 mites per plant. Three days later, infested leaves were cut, and the petiole was submerged in an EDTA solution for 8h to collect the leaf efflux. Three independent biological and two technical replicates were randomly injected and analysed (n = 6). This figure is available in colour at JXB online.

Fig. 3.

Signals accumulated in the leaf efflux of SR-treated sour orange. Citric acid, Glu, 2-hydroxyglutarate, and two oxylipins either fully or tentatively identified by exact mass as: 282.25 (hexadecanoic acid) and 256.24 (octadecanoic acid). Boxplot analysis of the relative abundance of the five compounds in sour orange (SO) and Cleopatra mandarin (Cleo) either in the absence (con) or presence of spider mites (SR). Different letters indicate significant differences between the treatments (ANOVA; LSD test, P < 0.05; n = 6).

Fig. 4.

Relevance of Glu and glutamate receptor-like genes (GRL) in SR against T. urticae in citrus rootstocks. Cleopatra mandarin (Cleo) and sour orange (SO) plants were previously infested with 10 T. urticae adult females (SR). (A) Three days later, distal uninfested leaves of control and SR plants were collected for RT-PCR analysis. Data are presented as a mean of three independent analyses of transcript expression relative to the housekeeping gene plants ± SD (n = 3). Different letters indicate significant differences (one-way ANOVA, P < 0.05; LSD) between treatments with Ct values as described by Yuan et al. (2006). (B) Spider mite oviposition in sour orange plants treated with 100mM of Glu (SO Glu) compared with control plants (SO con). Three days after Glu treatment, the plants were infested with 2-day-old T. urticae adult females. Three days later, the number of eggs was assessed. The asterisk indicates a significant difference between different treatments (t-test; P < 0.05). (C) GRL and (D) LOX2 expression levels in sour orange plants treated with 100mM of Glu (SO Glu). Three days later, the plants were infested with 2-day-old T. urticae adult females and 3 days after that the leaves were collected for RT-PCR analysis. Data are presented as a mean of three independent analyses of transcript expression relative to the housekeeping gene plants ± SD (n = 3). Different letters indicate significant differences (one-way ANOVA, P < 0.05; LSD) between treatments with Ct values as described by Yuan et al. (2006).

Fig. 5.

Hormone levels in sour orange (SO) and Cleopatra (Cleo) mandarin after SR treatment. Citrus rootstocks were infested with 10 T. urticae adult females. Three days later, uninfested distal leaves of these plants were collected to measure hormone levels. JA, OPDA, SA, and ABA levels were determined in freeze-dried material with targeted HPLC-MS. The results shown are the mean of hormone levels from three independent biological replicates ± SD (n = 3). Different letters indicate significant differences (one-way ANOVA, P < 0.05; LSD) between treatments. Cleo con, Cleopatra mandarin uninfested; Cleo SR, Cleopatra mandarin previously infested; SO con, sour orange uninfested; SO SR, sour orange previously infested.

Fig. 6.

(A) Symptomatic leaves and (B) spider mites in the Clemenules variety grafted onto sour orange or Cleopatra mandarin rootstocks. 2-year-old plants were infested with 20 T. urticae adult females. The samples were collected after the emergence of the first chlorotic leaves until the first symptoms of defoliation. The statistical analyses were conducted with a generalized linear mixed model. This figure is available in colour at JXB online.

Fig. 7.

Amino acid profile in the root efflux from rootstocks following spider mite infestation. Clemenules variety grafted onto sour orange (SO) and Clemenules variety grafted onto Cleopatra mandarin (Cleo) plants were either uninfested or infested (mite cartoon). 2-year-old grafted plants were infested with 20 mites per plant. Three days later, the stem was cut and the root efflux was collected using a Scholander pressure chamber. The samples were quantified by HPLC-Q-TOF-MS and processed using an amino acid library. Boxplots represent the average of three independent experiments with two technical replicates (n = 6). Different letters indicate significant differences (one-way ANOVA, P < 0.05; LSD) between treatments.

Fig. 8.

GRL expression in grafted plants affected by spider mites. Clemenules variety grafted onto sour orange (SO) and Clemenules variety grafted onto Cleopatra mandarin (Cleo) plants were either uninfested or infested (mite cartoon). 2-year-old grafted plants were infested with 20 mites per plant. Three days later, leaves were collected for mRNA analysis. Data are presented as a mean of three independent analyses of transcript expression relative to the housekeeping gene plants ± SD (n = 3). Different letters indicate significant differences (one-way ANOVA, P < 0.05; LSD) between treatments with Ct values as described by Yuan et al. (2006).

Fig. 9.

Hormonal content of grafted plants affected by the spider mites. Clemenules variety grafted onto sour orange (SO) and Clemenules variety grafted onto Cleopatra mandarin (Cleo) plants were either uninfested or infested (mite cartoon). The plants were infested with 20 mites per plant. Three days later, JA, OPDA, SA, and ABA levels were determined in freeze-dried material by HPLC-MS. The results shown are mean hormone levels of three independent analyses ± SD (n = 3). Different letters indicate significant differences (one-way ANOVA, P < 0.05; LSD) between treatments.

Fig. 10.

Model for SR against T. urticae in citrus. Spider mite attack rapidly induces changes in the leaf efflux. The two rootstocks respond differently to mite infestation. Leaf efflux from Cleopatra mandarin contains high amounts of myo-inositol, whereas sour orange also releases citric acid, Glu, and two fatty acids. These compounds can move to distal leaves or to the root. Once the roots detect the signals from the infested leaves, the resistant rootstock, sour orange, increases the transport of Glu to the shoot. The distal leaves receive the signals from the roots and/or from the infested leaves and respond to a future attack. Consequently, sour orange increases the expression of GRL that activates the JA pathway (high levels of OPDA and JA) and reduces the oviposition of T. urticae by 50%. The reduction of T. urticae oviposition in Cleopatra mandarin is 30%, possibly due to an increase in ABA levels. (This figure is available in colour at JXB online).