A sensitive and accurate method for quantifying endogenous systemin levels and verifying natural occurrence of Leu-Systemin

Dear Editor,

Tomato systemin (TomSys), the first discovered plant peptide hormone, can initiate systemic mobile signals to stimulate an immune response in both wounded and intact leaves (Pearce et al., 1991). For decades, biologists have been striving to uncover the mechanisms of TomSys function, but detailed information on the TomSys-regulated systemic defense process remains limited to date. The specific roles of three potential transmissible signaling components, ProSys mRNA, TomSys, and jasmonic acid (JA), in activation of systemic defense responses still need to be clarified (Zhang et al., 2020).

Because the function of TomSys in plant defense responses is closely associated with changes in its endogenous levels, it is important to accurately monitor the spatiotemporal dynamics of endogenous TomSys levels. However, limited by the technical bottleneck of TomSys quantitative analysis, researchers have had to roughly estimate TomSys levels indirectly through transcriptomic analysis. Nonetheless, changes in transcript level are not always consistent with TomSys variation (Pastor et al., 2018), and it is therefore critical to construct an accurate and efficient quantification method for endogenous TomSys.

Liquid chromatography coupled to mass spectrometry (LC–MS) has become the mainstream technique for quantitative analysis of ultra-trace endogenous TomSys because of its high sensitivity and high specificity (Du et al., 2010; Pastor et al., 2018). Sample pretreatment prior to LC–MS analysis is critical for improving the performance of the method. TomSys was first discovered in 60 pounds of tomato leaves by sequential C18 and cation-exchange LC fraction collection and bioactivity assays (Pearce et al., 1991). Although time consuming and laborious, this method has occasionally been used with minor modifications (Pearce et al., 2001; Pearce and Ryan, 2003). Recently developed methods for improving TomSys analysis, including high-performance LC fractioning-immunoaffinity column purification (Du et al., 2010), multi-step solid phase extraction (SPE) (Bai et al., 2014), and organic solvent precipitation (Pastor et al., 2018), have rarely been used by other research groups.

From a chemical standpoint, two important structural features can be used as global considerations for the design and development of a high-efficiency sample pretreatment method for TomSys. First, TomSys contains four basic and only two acidic amino acid residues (Supplemental Table 1), resulting in an isoelectric point of about 10.3. In an acidic solvent, each TomSys molecule will carry three to four net positive charges. As a result, the electrostatic interaction between positively charged TomSys molecules and negatively charged SPE sorbents is stronger than that of most other impurities. On this basis, we propose the use of mixed-mode weak cation exchange (WCX) SPE to enrich and purify endogenous TomSys. Second, the presence of a methionine residue in the TomSys structure makes it easily oxidized (Drazic and Winter, 2014). Under normal sample pretreatment conditions, it is difficult to control the level of exposure to oxygen in the air, and then TomSys in the sample matrix is very likely to be oxidized uncontrollably; this will negatively affect the sensitivity and precision of the method but has not been considered in previously developed methods. Eliminating the influence of oxygen and controlling the extent of TomSys oxidation with an appropriate oxidation system is therefore another key to developing high-performance analysis methods.

Based on these considerations, our method development focused on optimizing the WCX SPE-based enrichment and purification process in combination with controlled quantitative oxidation of TomSys under specific conditions (Figure 1A).

Figure 1.

Development of a high-performance method for monitoring dynamic changes in endogenous TomSys levels and verifying the natural occurrence of Leu-Sys in tomato defense responses.

(A) Schematic design of the high-performance method for endogenous TomSys analysis.

(B) Effect of H₂O₂ concentration on the yield of TomSys-O. The oxidation of TomSys was performed at 25°C. Error bars represent means ± standard deviation (SD) (n = 3); statistical differences are indicated with lowercase letters (P < 0.05, one-way ANOVA).

(C) Influence of loading pH on TomSys recovery. Error bars represent means ± SD (n = 3); statistical differences are indicated with lowercase letters (P < 0.05, one-way ANOVA).

(D) Influence of FA content in the washing solvent on the loss rate of TomSys. Error bars represent means ± SD (n = 3); asterisks indicate significant differences as determined by Student’s t-test, α = 0.05. ND, not detected.

(E) Influence of TFA content in the elution solvent on TomSys recovery. Error bars represent means ± SD (n = 3); statistical differences are indicated with lowercase letters (P < 0.05, one-way ANOVA).

(F) Illustration of the wounding treatment of tomato plants. Wounding was performed by nipping the midvein of lower leaves with a hemostat. Wounded leaves and distal intact leaves from wounded plants were collected separately 15 min after wounding, and lower and upper leaves from unwounded plants were collected as a control.

(G) Changes in endogenous levels of TomSys and JA in wild-type tomato leaves in response to wounding. Values are expressed as means ± SD (n = 3); statistical differences are indicated with lowercase letters (P < 0.05, one-way ANOVA).

(H–J) Extracted ion chromatogram (H), isotopic distribution pattern (I), and high-resolution MS/MS spectrum (J) of oxidized endogenous Leu-Sys in tomato leaves and comparison with the synthetic reference standard.

In the preliminary experiment, we observed that O₂ exposure level of sample during different sample concentrating processes likely determined the TomSys recovery rate (Supplemental Figure 1). LC–tandem MS (MS/MS) analysis confirmed that the sulfoxide form TomSys-O (the oxidized form of TomSys) was found after concentrating (Supplemental Figure 2). To eliminate the negative effect of uncontrollable oxidation by atmospheric O₂, we used hydrogen peroxide (H₂O₂) to actively oxidize TomSys to TomSys-O because it has an appropriate oxidation capacity and its reduced product (H_{2O) has} no adverse effects. When the H₂O₂ concentration was 0.1%, the yield of TomSys-O reached about 99.8%, which was significantly higher than that obtained under other conditions (Figure 1B), implying that TomSys can be completely converted to TomSys-O in a controlled and quantitative manner under very mild conditions.

We next tailored the enrichment and purification of endogenous TomSys based on WCX SPE, in which pH has significant effects. The optimized sample loading solvent was formic acid (FA)-modified 20% MeOH (pH 4) (Figure 1C) so that the charge state of TomSys dominated the ionic interaction between TomSys and WCX sorbents, maximizing its retention. In the critical washing step, 90% MeOH with 1% FA was chosen as the most appropriate solvent for removal of co-retained interfering substances with physiochemical properties similar to TomSys, such as neutral or basic peptide molecules with lower isoelectric points, without disrupting the interactions between TomSys and WCX sorbents and causing loss of TomSys molecules (Figure 1D). To elute TomSys from the WCX cartridge with the best recovery, trifluoroacetic acid (TFA) was selected to thoroughly destruct the ionic interactions owing to its stronger acidity than FA. Because a slight residue of TFA after concentrating may result in excessive oxidation of TomSys by subsequent H₂O₂ under acidic conditions, we proposed a stepwise elution process consisting of TFA aqueous solution to break up ionic interactions, followed by MeOH solution as reversed-phase eluent to elute TomSys. Figure 1E and Supplemental Figure 3 show that the combination of 2% aqueous TFA and MeOH produced the best TomSys recovery.

After key parameters addressing the two considerations were fixed, the entire method, including sample pretreatment and LC–MS/MS analysis, was validated (Supplemental Table 2). The recovery was measured to be 99.3% ± 5.2% (n = 3), indicating good accuracy of the method due to almost no loss of target molecules during the multi-step SPE process and nearly 100% conversion to TomSys-O in the oxidation process (Figure 1B and Supplemental Figure 4C). By contrast, when the same method was used without controlled oxidation, both TomSys and its oxidized form, TomSys-O, were detected (Supplemental Figures 4B and 5), and the proportion of TomSys-O varied from 21% to 90% with a high deviation (relative standard deviation = 45%, n = 8), suggesting lower recovery and precision caused by uncontrollable auto-oxidation during sample pretreatment. The matrix effect was determined to be 81.6% ± 11.5% (n = 3), indicating that residual interferences in the purified sample suppressed the LC–MS signal of TomSys by less than 20%. The limit of detection of the proposed method applied to real plant tissue samples was 12.5 pg (Supplemental Table 2), illustrating its higher sensitivity compared with previous methods (Du et al., 2010; Pastor et al., 2018).

Finally, to demonstrate the applicability of the proposed method, it was used to monitor changes in endogenous TomSys levels in response to wounding. As expected, TomSys levels in local wounded leaves were significantly higher than basal levels (Figures 1F and 1G). Correspondingly, a rapid burst of JA was observed in local wounded leaves (Figure 1G). Moreover, for the first time, a significant increase in TomSys was directly observed in unwounded systemic leaves in response to wounding (Figure 1G). This increase in distal TomSys level indicated possible mobility of endogenous TomSys, although this seems unlikely given the spreading velocity reported in the literature (Mucha et al., 2019). Notably, JA levels in systemic leaves after wounding were significantly lower than basal levels (Figure 1G), which may be attributed to JA translocation, as it is thought that JA translocates from wounded to systemic tissues (Stratmann, 2003). It is reasonable to predict that TomSys accumulation in distal intact leaves results from a more complicated mechanism than JA-mediated signaling cascades, which needs to be clarified in further studies. We also performed an ultra-performance LC–high-resolution MS survey to search for Leu-Sys as a putative key precursor of TomSys in TomSys-containing eluents, as it is believed that the proposed high-efficiency sample pretreatment method can cover both Leu-Sys and TomSys simultaneously owing to their similar structures and physicochemical properties (Supplemental Table 1). We found an endogenous Leu-Sys-like compound with the same m/z as Leu-Sys and a retention time (Figure 1H), charge state distribution, isotopic distribution pattern (Figure 1I; Supplemental Table 3), and MS/MS fragmentation pattern (Figure 1J; Supplemental Table 4) identical to those of the synthetic Leu-Sys reference. The occurrence of endogenous Leu-Sys was further confirmed by an increase in the MS signal after spiking with the synthetic standard (Supplemental Figure 6). Consequently, this compound was identified as Leu-Sys, and the occurrence of endogenous Leu-Sys was verified for the first time, providing solid evidence for the predicted Leu-Sys-involved TomSys maturation mechanism (Beloshistov et al., 2018). The identification of natural Leu-Sys suggests that Leu-Sys is a stable TomSys precursor and is maintained at an appropriate basal level, but in sufficient to induce defense responses. When wounding occurs, Leu-Sys accumulation is induced by excessive ProSys synthesis, leading to subsequent TomSys accumulation.

In conclusion, these findings help to elucidate the mechanisms of TomSys maturation and signal transduction in plant defense responses and demonstrate the excellent performance and usefulness of the proposed method. It is conceivable that this method will become a promising and powerful approach to support in-depth TomSys-related studies in the future.

Funding

This work was financially supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (grant no. XDA24040202), the CAS Key Technology Talent Program (2017, Y869041), and the National Natural Science Foundation of China (grant nos. 32270427 and 31800303).

Author contributions

J.C. and P.X. conceived the study; P.X. and J.Y. wrote the paper; and J.Y. and S.C. performed the experiments.

Published: June 10, 2023