Tissue-type specific systemin perception and the elusive systemin receptor

Systemin is a wound signaling peptide from tomato that is important for plant defenses against herbivory. The systemin receptor was initially identified as the tomato homolog of the brassinosteroid receptor BRI1, but genetic evidence argued against this finding. However, we found that BRI1 may function as an inappropriate systemin binding protein that does not activate the systemin signaling pathway. Here we provide evidence that systemin perception is localized in a tissue-type specific manner. Mesophyll protoplasts were not sensitive to systemin, while they responded to other elicitors. We hypothesize that the elusive systemin receptor is a protein with high similarity to BRI1 which is specifically localized in vascular tissue like the systemin precursor prosystemin. Binding of systemin to BRI1 may be an artifact of transgenic BRI1-overexpressing plants, but does not take place in wild type tomato cells.Key words: systemin, systemin receptor, brassinosteroids, BRI1, BRL, protoplastsSystemin is thought to be processed from its precursor prosystemin upon insect attack and wounding of tomato leaves. Strong evidence has been gathered for an important role of (pro-)systemin in the activation of defenses against insects, and the underlying signaling pathway has been studied in detail.¹ However, the perception of systemin is controversial. Meindl et al.² and Scheer and Ryan³ identified high affinity, saturable, reversible and specific cell surface binding sites on Solanum peruvianum suspension-cultured cells which are known to be highly sensitive to systemin.⁴ A purification approach using a photoaffinity systemin analog identified a 160 kDa protein as the systemin receptor (SR160).⁵ Follow-up studies showed that overexpression of tomato 35S::SR160 in systemin-insensitive tobacco plants conferred systemin sensitivity to tobacco.⁶ Surprisingly, SR160 turned out to be the tomato homolog of the brassinosteroid receptor BRI1,⁷ which raised many questions as to the functionality of a receptor for two structurally and functionally diverse ligands. It was then shown in two independent papers that a null mutant for tomato BRI1, cu-3, exhibited a normal response to systemin.^8,9 This was strong evidence that SR160/BRI1 does not represent the functional systemin receptor. Our recent data added a peculiar twist to this story. We found that overexpression of tomato BRI1 in tobacco suspension-cultured cells resulted in binding of a fluorescently labeled systemin to the plasma membranes of the transgenic tobacco cells, but not to wild type cells. Surprisingly, this did not result in BRI1-dependent signal transduction and activation of a defense response, although we detected weak BRI1-independent signaling responses to systemin.¹⁰ Together with the identification of BRI1 as the systemin receptor by Scheer and Ryan,⁵ the simplest explanation for this phenomenon is that BRI1 is a systemin binding protein, but not the physiological systemin receptor.Therefore and for other reasons, we suggested that the true systemin receptor may be a protein with very similar properties as BRI1, e.g., a homolog of the BRI1-like (BRL) proteins. The purification strategy employed by Scheer and Ryan5 may have resulted in binding of a photoaffinity-systemin derivative to BRI1 and one or more BRL proteins. Since BRLs and BRI1 have a very similar MW, multiple bands on a SDS-PAGE would not be detectable.Here, we would like to add another aspect of systemin perception. We provide evidence for tissue-specific systemin sensitivity and discuss how this may affect systemin binding to BRI1 and the elusive systemin receptor. Prosystemin is only present in phloem parenchyma cells.¹¹ It can be surmised that the systemin receptor is located close to these cells. Systemin perception results in JA synthesis in companion cells of vascular bundles.¹² Since JA or a JA derivative is the most likely phloem-mobile candidate for a systemic long-distance wound signal, it is thought that JA is moving from companion cells into sieve cells to reach distant parts of the plant for upregulation of wound response genes in leaf cells, including mesophyll cells.^13–15Here, we tested the hypothesis that mesophyll cells lack systemin perception. We generated mesophyll protoplasts from tomato leaf material as well as protoplasts from S. peruvianum suspension-cultured cells, the same cell line that had been used for the purification of SR160/BRI1 and is known to be highly sensitive to systemin. Mesophyll protoplasts showed increased phosphorylation of MAP kinases (MPKs) in response to the elicitors flg22 and chitosan, bacterial and fungal MAMPs, respectively. However, they did not respond to systemin. In contrast, the S. peruvianum protoplasts did respond to systemin and to flg22, demonstrating that the protoplasting procedure did not compromise the systemin perception mechanism (Fig. 1). Immunocomplex kinase assays with specific antibodies against tomato MPK2 produced similar results (data not shown). Since flg22, chitosan and systemin activated the same MPKs (Fig. 1), our data indicate that systemin perception is absent in mesophyll protoplasts. Our leaf protoplasting protocol is a modification of the protocol by Yoo et al. which results in the generation of mesophyll protoplasts.¹⁶ In contrast, suspension-cultured cells do not normally represent specific cell types and it is not known why the S. peruvianum cells are highly sensitive to systemin.Open in a separate windowFigure 1Absence of systemin-induced MPK phosphorylation in mesophyll cells. Protoplasts were generated (protocol available upon request) from S. peruvianum suspension-cultured cells and from S. lycopersicum cv. MicroTom leaves. After a 1.5 hour recovery phase on ice, protoplasts were resuspended in WI medium (0.5 M mannitol, 5 mM ME S pH 5.7, 20 mM KCl), recovered for 1 hour in non-stick tubes with constant rotation on a rotary shaker at room temperature, and then treated with either water (con), 10 nM systemin (sys), 100 nM flg22, or 2.5 µg/ml chitosan (from crab shells—chi) for 10 min at room temperature. Protoplasts were analyzed for MPK phosphorylation by immunoblotting using an anti-phospho-ER K antibody (phospho-p44/42 MA PK (Erk1/2) (Thr202/Tyr204); D13.14.4E; Cell Signaling Technology) at a dilution of 1:2,000. This antibody recognizes MPKs that are phosphorylated on either the Thr and Tyr or on only the Thr within the TE Y phosphorylation motif which is conserved among plant and metazoan MPKs. It is known to recognize the tobacco MPKs SIPK and WIPK²¹ and Arabidopsis MPK6 and MPK3,²² the orthologs of tomato MPK1/2 and MPK3.²³ Bands were visualized as described.¹⁰ Proteins on membranes were stained with Ponceau S to demonstrate equal loading.Intriguingly, BRL1, BRL2 and BRL3 are expressed in the vasculature and function in vascular pattern formation in Arabidopsis, while BRI1 is ubiquitously expressed in dividing and elongating cells. BRL3 is even specifically expressed in phloem cells.¹⁷ This matches the highly specific localization of prosystemin in the phloem parenchyma cells.^11,18 The highest BRI1 expression is found in growing parts of young leaves^17,19 while prosystemin is specifically present in the phloem parenchyma cells throughout all developmental stages.¹¹ In this context, it is also interesting to note that application of systemin to tomato plants via the cut stem results in rapid and strong MPK activation. In this assay, systemin is delivered to leaf cells via the transpiration stream and therefore present in vascular tissue.²⁰Based on the combined evidence, we propose that the true systemin receptor is a BRL or similar protein which is expressed in phloem cells in the vicinity of the parenchyma cells that express prosystemin, but not in mesophyll cells. Because of the similarity between BRLs and BRI1, BRI1 was erroneously identified as the systemin receptor. Inappropriate binding of systemin to BRI1 is consistent with the high similarity between BRI1 and BRLs. However, because of the tissue-specificity of the systemin signaling pathway, inappropriate binding of systemin to BRI1 may rarely occur in wild type plants and may not pose an interference problem for either systemin or brassinosteroid signaling.