Proteinase inhibitor-inducing activity of the prohormone prosystemin resides exclusively in the C-terminal systemin domain

Abstract

Prosystemin is the 200-amino acid precursor of the 18-amino acid polypeptide defense hormone, systemin. Herein, we report that prosystemin was found to be as biologically active as systemin when assayed for proteinase inhibitor induction in young tomato plants and nearly as active in the alkalinization response in Lycopersicon esculentum suspension-cultured cells. Similar to many animal prohormones that harbor multiple signals, the systemin precursor contains five imperfect repetitive domains N-terminal to a single systemin domain. Whether the five repetitive domains contain defense signals has not been established. N-terminal deletions of prosystemin had little effect on its activity in tomato plants or suspension-cultured cells. Deletion of the C-terminal region of prosystemin containing the 18-amino acid systemin domain completely abolished its proteinase inhibitor induction and alkalinization activities. The apoplastic fluid from tomato leaves and the medium of cultured cells were analyzed for proteolytic activity that could process prosystemin to systemin. These experiments showed that proteolytic enzymes present in the apoplasm and medium could cleave prosystemin into large fragments, but the enzymes did not produce detectable levels of systemin. Additionally, inhibitors of these proteolytic enzymes did not affect the biological activity of prosystemin. The cumulative data indicated that prosystemin and/or large fragments of prosystemin can be active inducers of defense responses in both tomato leaves and suspension-cultured cells and that the only region of prosystemin that is responsible for activating the defense response resides in the systemin domain.

Polypeptide hormones are important signaling molecules throughout the animal kingdom and in yeast and have been found in plants only recently (1). Only three plant polypeptide signals are known: an 18-amino acid polypeptide called systemin (SYS) that is a defense signal in Solanaceae family members (2–4); a 5-amino acid sulfated polypeptide, called phytosulfokine-α, that causes proliferation of plant cell cultures (5); and oligopeptides of about 10 amino acids, collectively called ENOD40, that can regulate cell division and that seem to be involved in nodule formation in legumes (6).

SYS induces the systemic expression of over 20 defense-related genes in leaves of tomato plants in response to herbivory and mechanical damage (7, 8). The binding of SYS to a cell-surface receptor (9, 10) activates numerous cellular enzymes leading to plant defense gene transcription via the octadecanoid pathway (8, 11–22).

Most animal and yeast polypeptide hormones are synthesized through the secretory pathway as prohormones, processed in secretory vesicles, and secreted in response to appropriate stimuli (23–25). Membrane-anchored polypeptide factors, such as tumor necrosis factor-α, other cytokines, and growth factors, are also synthesized through the secretory system but are not processed to the active form until they are anchored in the membranes where specific proteinases process them in response to external cues (26). SYS is similar to the animal polypeptide hormones and factors in that it is processed from a larger precursor by proteolytic cleavage (27). In plants, prosystemin (PS) has been shown to be associated with the vascular tissue, placing the synthesis of the SYS precursor in the tissue necessary for systemic signaling (28). The release of SYS in response to wounding is analogous to the release of the animal cytokine tumor necrosis factor-α from macrophages into the bloodstream in response to trauma to activate the inflammatory response (29–31).

PS is a hydrophilic protein in which 44% of its amino acids are charged (27). It is a unique protein that shows no homology to any other sequence in the protein database. Although the initial P₁–P₁′ cleavage sites in PS have not been identified, it is clear that they do not involve dibasic cleavage sites that are found in most animal prohormones (32–34). Only one dibasic site is present in PS, and it is located in the middle of the SYS sequence, where its cleavage has been shown to reduce SYS activity severely (35–37). Another difference between the animal prohormones and PS is the absence of a signal peptide for targeting to the secretory pathway (38). Additionally, PS does not contain prenylation sites, N-linked glycosylation sites, or transmembrane domains, suggesting that it is synthesized and stored in the cytoplasm. Many prohormones in animals harbor several different polypeptide hormones, such as proopiomelanocortin, whereas others contain redundant copies of a single hormone, such as preproenkephalin (39). The PS sequence contains only one copy of SYS, near the C terminus, but also contains five imperfect repeated sequences between the SYS sequence and the N terminus of the protein (27). Whether PS harbors more signaling peptides that can serve as defense signals has been a long-standing question.

Herein, we report the synthesis and biological activities of several recombinant truncated and mutated PS proteins, providing evidence that the only region of the PS prohormone that can activate the early steps of signaling resides exclusively in the SYS domain. The data, however, do not eliminate the possibility that the repeated sequences, when processed in the plant, may be involved in signaling processes.

Materials and Methods

Plant Material.

Wild-type (Lycopersicon esculentum cv. Castlemart and cv. Better Boy) and transgenic tomato plants overexpressing the PS antisense cDNA (cv. Better Boy) were grown for 14–15 days for 17 h at 28°C under >300 μE m⁻²⋅s⁻¹ light followed by a 7-h, 17°C dark period. Plants at this stage of development had two expanded leaves and a small apical leaf.

Cell-Suspension Cultures.

A Lycopersicon peruvianum suspension-culture cell line was maintained and grown under constant light as described (10).

Recombinant PS Production.

All standard recombinant DNA procedures used in this research were performed as described (40). Production of recombinant PS proteins was performed as described (41), with the alterations and additions described below.

PS A195.

A substitution of alanine for threonine at amino acid position 195 was introduced by a point mutation in the DNA sequence encoding PS. This mutation was engineered by PCR mutagenesis with a 5′ end primer (5′-GCGGAGCTCAAGCTTAAACTAAGAAAACCATGGG-3′) and a 3′ end primer (5′-GTTTCTAGAGTTTATTATTGTCTGCTTGCATT-3′). The altered sequence was excised by using BglII and XhoI and ligated into the similarly digested pBluescript SK(−) pPS plasmid (27). The clone was then restricted with NdeI and BamHI endonucleases. The resulting DNA fragment was then purified and ligated back into a similarly digested and purified pET 11d PS (HP) A15 vector (41). This new pET 11d PS (HP) A15 A195 expression vector was then transformed into the Escherichia coli strain BL21 [DE3]. All PCR-derived modifications of the PS cDNA sequence were verified by DNA sequencing.

PS ΔSYS.

The pPS cDNA was digested with BglII and XhoI endonucleases, purified, and ligated to a synthetic double-stranded DNA adapter containing stop codons in all possible frames. The adapter was generated from the following sequences: 5′-GATCTTTAACTAGCTGAGGATCCCTGAC-3′ and 5′-TCGAGTCAGGGATCCTCAGCTAGTTAAA-3′. The resulting clone was then restricted and ligated as described above. The following HPLC columns were used for PS purification as described (41): 1-cm × 10-cm quaternary amine strong anion-exchange (300VHP575; 5-μm particle size; Vydac, Hesperia, CA) and semipreparative C4-RP HPLC column (214TP510; 5-μm particle size; Vydac).

The primary structures of recombinant PS proteins were confirmed by mass spectroscopic analyses, N-terminal sequencing, SDS/PAGE, and immunoblot analyses with either SYS or PS antisera. N-terminal amino acid analysis and mass spectroscopic analysis were performed by the Laboratory for Biotechnology and BioAnalysis at Washington State University.

Wounding, Elicitors, and Bioassays.

Wound and elicitation experiments were performed on 14- to 15-day-old tomato seedlings. PS and SYS derivatives were supplied through their cut stems. Plants were excised at their bases and immediately placed into 1.5-ml Eppendorf tubes containing 1 ml of the elicitor or water and then incubated for 1 h as described (42). The uptake rate was ≈90 μl/h. After 1 h in water, half of the control plants were wounded one or two times on both leaves with a hemostat perpendicular to the midvein. After 24 h, proteinase inhibitor I and II concentrations were assayed from the expressed leaf juice of treated plants by radial immunodiffusion as described (43).

Alkalinization assays of cell culture medium by addition of SYS, SYS Ala-17, and various PS proteins were performed as described (10, 15). A 1–2× concentration of Complete Protease inhibitor mixture (Roche Molecular Biochemicals) or an EDTA-free version of the mixture with 1 μg/ml pepstatin was supplied to either tomato seedlings or suspension cell cultures as indicated.

Apoplastic Fluid.

Intercellular wash fluid (ICWF) was obtained from 15- to 17-day-old tomato seedlings (cv. Castlemart). Leaves were cut at the petioles with a razor and placed in 4°C water. Then, 10–20 g of collected leaf tissue was transferred to 1 liter of 25 mM Mes [2-(N-morpholino)ethanesulfonic acid]/150 mM NaCl, pH 6.2 and was subjected to vacuum by a water aspirator for 15 min, breaking the vacuum every 3 min. After infiltration, leaves were gently filtered, washed with cold water, and blotted dry. The leaves were gently transferred into 20-ml syringe barrels and placed in 50-ml conical tubes. The ICWF was collected by centrifugation at 1,800 rpm for 10 min at 4°C in a Sorvall RT6000B centrifuge with a H1000B rotor. The resulting infiltrate solution was clarified further by centrifugation at 2,500 rpm for 3 min. The ICWF was then divided into aliquots and stored at −80°C. Cell medium (CM) was collected from 1-week-old Lycopersicon peruvianum suspension cell cultures by removing the cells by centrifugation.

PS-Processing Assay.

PS (1.5 μg) was added to 150 μl of ICWF/CM/buffer, mixed, and incubated at room temperature. At the times indicated, 15-μl aliquots were removed, combined with 2× Laemmli loading buffer, and boiled for 3–5 min. As indicated in Results and Discussion, Complete Protease inhibitor mixture at 1–2× concentration with 1 μg/ml pepstatin was added to ICWF/CM/buffer 5 min before addition of PS. Similar assays were carried out by adding 10–25 μg of casein (Sigma), cytochrome c (Sigma), BSA (Sigma), histone type III ss (Sigma), or α-amylase (Sigma) to 150 μl of ICWF/CM/buffer and treated as described above. ICWF was also subjected to a boiling water bath for 3 min, cooled, and clarified in a microfuge for 30 s. Separation and analysis of protein samples were performed by denaturing SDS/PAGE, and samples were visualized by Coomassie blue staining solution or by immunoblot analysis as described (41). A SYS-specific antibody was used to monitor the proteolytic degradation of PS. The antibody recognizes only those polypeptides that contain the SYS sequence and was unable to detect PS ΔSYS in an immunoblot analysis (G.P., J.E.D., and C.A.R., unpublished data).

Results and Discussion

PS was produced in Escherichia coli and purified by HPLC (41). The 18-amino acid polypeptide SYS is derived from a single copy domain that is contained in the C-terminal region of the 200-amino acid PS (Fig. 1A). To determine whether PS might be processed in leaves to produce SYS, PS was supplied to young tomato plants through their cut stems, and the synthesis of proteinase inhibitors was assayed. PS was found to be as active in inducing the synthesis of proteinase inhibitors as SYS, with activity in the range of femtomoles per plant (Fig. 2).

Figure 1.

Summary of the various recombinant PS constructs and synthetic SYS peptides tested for bioactivity. The wild-type PS consists of 200 amino acids. The numbers above the boxes denote the amino acid number in the wild-type sequence. PS, PS A195, and PS ΔSYS start with the amino acid glycine at position 2 (41). The cross-hatched boxes represent the location of imperfect repetitive elements within the PS sequence. The shaded boxes denote the location of the SYS peptide in the various PS proteins. The dark lines in PS A195 and A17 (SYS–Ala-17) denote alanine for threonine substitutions in the SYS peptide.

Figure 2.

Accumulation of proteinase inhibitor I in response to wounding (W), SYS, and PS. Tomato seedlings were supplied with the indicated amounts of SYS or PS in water through their cut stems (pM = picomoles per plant; fM = femtomoles per plant). Proteinase inhibitor I accumulation in leaves was measured 24 h after treatment. Control plants were supplied with water alone (H₂O) or incubated with water for 1 h and wounded on the leaves (H₂O-W). The data are representative of a typical experiment. Similar results were obtained in at least 15 independent experiments (six plants per experiment) assaying for the accumulation of both proteinase inhibitor I and II.

In addition to the assay in plants, SYS can be assayed conveniently by using Lycopersicon peruvianum suspension-cultured tomato cells by monitoring the change in pH of the extracellular medium (10, 15). As shown in Fig. 3A, when PS was added to suspension-cultured tomato cells, it caused a concentration-dependent alkalinization of culture medium similar to that caused by SYS (Fig. 3B). However, a comparison of the alkalinization curves produced by SYS and PS in Fig. 3 revealed that the effective dose of PS was at a concentration ≈100-fold higher than that of SYS.

Figure 3.

Alkalinization of suspension-culture medium in response to PS and SYS. Lycopersicon peruvianum suspension-cultured cells were exposed to different concentrations of PS (A) or SYS (B), and the change in pH of the medium was monitored at 5-min intervals. (A) PS concentrations: (♦) 280 nM, (■) 28 nM, (▴) 2.8 nM, (○) untreated control. (B) SYS concentrations: (■) 28 nM, (▴) 2.8 nM, (●) 280 pM, (×) 28 pM, (○) untreated control. For each treatment, the data represent a minimum of five independent experiments showing similar results.

The amino acid sequence of PS contains five imperfect repetitive elements (27). It had been speculated that these repeated motifs might represent additional biologically active peptides as found in many animal prohormones (39). When 250 pmol of synthetic peptides containing some of these repeated elements were supplied to tomato plants, the peptides failed to induce the accumulation of proteinase inhibitors (G.P. and C.A.R., unpublished data). One potential explanation as to why these peptides did not induce proteinase inhibitors could be that the peptides were not the correct size and/or were not in the proper context for bioactivity. To determine whether the biological activity of PS was due to only the SYS domain, several recombinant PS derivatives with deletions at the N and C termini were produced in Escherichia coli (Fig. 1 B–D and F) to assay for their structure-activity relationships. The activity of PS 185 (Fig. 4), in which the first repetitive element is deleted, was identical to that of PS (Fig. 2). The activity of PS 113, in which the first three N-terminal repetitive elements are deleted, was slightly, but consistently, reduced (Fig. 4) compared with that of PS (Fig. 2). This reduced activity of PS 113 might not be due to the loss of active elements but may be a result of altered folding caused by the large deletion. PS 185 and PS 113 had similar activities, as determined in an assay for the alkalinization response in suspension-cultured cells (data not shown), indicating that the N-terminal 87 amino acids of the PS sequence are not required for induction of proteinase inhibitors in tomato leaves or for the alkalinization response. In contrast, the deletion of the C-terminal 22-amino acid domain of PS (PS ΔSYS, Fig. 1D) that contains SYS completely abolished its proteinase inhibitor-inducing (Fig. 4) and alkalinization activities (data not shown).

Figure 4.

Accumulation of proteinase inhibitor I in response to modified recombinant PS proteins. Tomato seedlings were excised and supplied with the indicated amounts of proteins in water through their cut stems (pM = picomoles per plant; fM = femtomoles per plant). Proteinase inhibitor I accumulation in leaves was measured 24 h after treatment. The data are representative of a typical experiment. Similar results were obtained for PS 185, PS A195, and PS ΔSYS (compare with Fig. 1) in at least 10 independent experiments for each construct and 6 independent experiments for PS 113 (six plants per experiment). Similar results were obtained for Inh II (data not shown).

To investigate further the biological activity of PS in plants and in suspension-cultured cells, competitive inhibition studies were performed with the analog A17-SYS (Fig. 1G) in which the critical amino acid threonine at position 17 was substituted with an alanine. A17-SYS had been shown to be essentially inactive and a potent antagonist of SYS in plants and suspension-cultured cells (15, 35). In tomato plants, A17-SYS is also an antagonist of PS induction of proteinase inhibitor synthesis (Fig. 5), indicating that the SYS domain may be the only region in PS that can activate defense genes. In contrast to plants, only a 10-fold excess of A17-SYS was necessary to inhibit the PS-induced alkalinization competitively (Fig. 6A), whereas achieving a similar inhibition of SYS-induced alkalinization required a 1,000-fold excess of A17-SYS (Fig. 6B).

Figure 5.

Suppression of PS- and SYS-induced proteinase inhibitor I accumulation by A17-SYS (A17). Varying amounts (picomoles per plant) of PS and SYS in 10 mM phosphate buffer (pH 6.5) were supplied to excised tomato seedlings in the presence or absence of 250 pmol of A17. The bars represent the levels of proteinase inhibitor I accumulation in leaves 24 h after treatment. The data represent a minimum of three independent experiments for each treatment (six plants per treatment).

Figure 6.

The effect of A17-SYS and PS A195 on PS- and SYS-induced alkalinization of culture medium. Cells were treated with SYS or PS either separately or in combination with A17 or PS A195. The change in pH of the medium was monitored at the times indicated. (A) Effects of PS A195 and A17 on alkalinization induced by (■) 28 nM PS, (▵) 28 nM PS + 280 nM PS A195, (●) 280 nM A17, (○) 28 nM PS + 280 nM A17, or (▴) 280 nM PS A195. (B) Effects of PS A195 and A17 on alkalinization induced by (■) 280 pM SYS, (▵) 280 pM SYS + 280 nM PS A195, (○) 280 pM SYS + 280 nM A17, (●) 280 nM A17, or (▴) 280 nM PS A195.

A PS analog produced in Escherichia coli having the critical threonine in the SYS domain substituted with alanine (PS A195, Fig. 1F) was found to be inactive in producing the alkalinization response in cell cultures. On the other hand, PS A195 induced a low level of accumulation of proteinase inhibitors in tomato leaves (Fig. 4), but the induction required nearly 100 times more protein than did PS to achieve half-maximal induction of proteinase inhibitor accumulation.

In contrast to A17-SYS, PS A195 did not antagonize the synthesis of proteinase inhibitor I in tomato leaves induced by either PS or SYS (data not shown). PS A195 was only about 10% as effective as A17-SYS in inhibiting either PS or SYS activity in the alkalinization response in suspension-cultured cells (Fig. 6 A and B). These experiments indicate that PS A195 does not have the same capability as A17-SYS to compete with the SYS receptor. This result may reflect a structural change in PS caused by the substitution of threonine with alanine that is not reflected in the smaller A17-SYS. This change may also be reflected in the low-level proteinase inhibitor-inducing activity in PS A195 that was noted above.

Transgenic plants constitutively expressing a PS antisense cDNA have been shown to have a severely impaired systemic wound response (27, 44). The application of PS or SYS to a single wound site on the lower terminal leaflets of 14-day-old PS-antisense tomato plants restored the local (wounded leaf) and systemic (unwounded upper leaf) accumulation of proteinase inhibitors to wild-type levels (data not shown). When PS ΔSYS (see Fig. 1D) was applied to a wound site on PS-antisense tomato plants, no local or systemic increase in proteinase inhibitors was observed, confirming that the SYS domain was essential for systemic signaling (data not shown).

When supplied to plants through cut stems, PS is believed to be transported via the transpiration stream and released into the leaf apoplastic space where it or a processed form interacts with the SYS receptor. Similarly, PS supplied to suspension-cultured cells is thought either to interact directly with the SYS receptor or to be processed by proteolytic enzymes in the cell culture medium to produce an active SYS.

To determine whether putative PS processing enzymes were present in the apoplastic spaces of tomato leaves, PS was incubated with ICWF obtained from leaves of young tomato plants. The reaction was monitored by immunoblot analyses by using rabbit anti-SYS antibodies to visualize SYS-containing polypeptides. The 23-kDa PS migrates anomalously with an apparent molecular mass of 40 kDa (see arrows in Fig. 7). After a 5-min incubation, three or four SYS cross-reactive bands were generated (Fig. 7A). One of the bands migrated with the dye front (1–2 kDa, Fig. 7A, bottom of gel), and on separation by urea/SDS/PAGE (data not shown), this polypeptide was found to migrate slightly faster than the 18-amino acid SYS peptide. This result indicated that this polypeptide was smaller than SYS and still retained SYS antigenic sites. Previous data (35) had shown that sequential deletion of amino acids from the N terminus of SYS caused a precipitous decline in its activity, and deletion from the C terminus totally eliminated activity. Thus, the 1- to 2-kDa antigenic polypeptide produced by ICWF had likely lost some or all of its defense gene-inducing activity. Sequence analysis of the 33-kDa band revealed that the N-terminal 48 amino acids were removed from the recombinant PS. A number of other bands are also produced within 5 min and degraded; however, the 33-kDa band not only persisted but accumulated during the 4-h incubation period. The inefficient conversion of PS to small polypeptides by ICWF suggests that only a very small amount of SYS could have been generated. None of the antigenic bands were produced when only ICWF was incubated at room temperature (data not shown).

Figure 7.

Proteolytic degradation of PS in ICWF from tomato leaves and suspension cell culture medium. PS was added to ICWF, CM, or buffer and incubated at room temperature for the times indicated. Immunoblot analyses of the various treatments with a SYS-specific antibody (1:1,000 dilution) are shown above. The arrows indicate the location of PS. (A) ICWF. (B) CM. (C) ICWF with protease inhibitor cocktail. (D) CM with protease inhibitor cocktail. (E) Heat-treated ICWF with PS. (F) Buffer control (25 mM Mes/150 mM NaCl, pH 6.2 with PS).

A different pattern of PS proteolytic degradation products occurred in cell culture medium compared with the pattern produced with ICWF (Fig. 7B). The 1- to 2-kDa and 23-kDa bands were not detectable, but a 33-kDa band similar to that found in ICWF is generated and accumulated during the 4-h incubation. The absence of small polypeptides with SYS epitopes indicates that the medium may not be capable of producing SYS. Incubations of PS with ICWF also produced species much larger than PS (Fig. 7 A and C). PS alone migrates with a molecular mass much larger (≈40 kDa) than its known size (23 kDa). The reasons for this difference in size are unknown, but they may be manifested in the larger forms that seem to result from proteolysis. We presently have no evidence for covalent bonding of PS to itself or to other components of the ICWF, but it would be interesting to investigate this possibility.

If SYS is not processed by ICWF or by suspension-culture medium, then it is possible that it is generated by membrane-bound enzymes or by enzymes released from cells on wounding. In a previous study, a 50-kDa SYS-binding protein (SBP50) was identified in plasma membranes of tomato (36); this protein shares several characteristics with members of the KEX2 (furin-like) family of proteases. The presence of a KEX2-like processing site within SYS, coupled with the fact that SBP50 is capable of cleaving SYS to a much less active form, suggests that SBP50 may not be involved in processing but in SYS degradation. (35–37).

The addition of a mixture of proteinase inhibitors to both ICWF and CM (Fig. 7 C and D) blocked the degradation of PS into antigenic polypeptides with molecular masses of less than 23 kDa, while increasing the accumulation/stability of the larger antigenic polypeptides. To assess the general proteolytic activity of ICWF, five additional proteins were assayed as substrates. Both casein and histone III proteins were degraded completely within 30 min after addition to ICWF, whereas BSA, cytochrome c, and α-amylase showed no apparent degradation even after 4 h of incubation (data not shown). These results indicated that substantial proteolytic activity is present in the ICWF but that it is highly specific, because PS and its larger fragments, as well as other proteins, are not readily degraded by the proteolytic activity. The entire PS-degrading activity in ICWF could be eliminated with heat treatment (Fig. 6E).

The effects of the protease inhibitor mixture on the biological activity of PS and SYS were analyzed in the two bioassays. Plants supplied simultaneously with PS or SYS and the protease inhibitors showed a reproducible 10–15% reduction in the levels of accumulation of Inh I and II as compared with the plants supplied with PS or SYS alone (data not shown). Similar results were obtained in the Lycopersicon peruvianum suspension-cultured cells, where PS and SYS induced alkalinization was reduced by ≈15% in the presence of protease inhibitors (data not shown). The similar activity of PS in both tomato leaves and suspension-cultured cells in the presence and absence of protease inhibitors supports the earlier data that indicate that PS or large fragments of PS containing the SYS region of the protein can interact with the SYS receptor.

If only the SYS domain of PS contains proteinase inhibitor-inducing activity, then what is the function of its other 182 amino acids? It has been shown that the PS gene is expressed at low levels in the leaves of young tomato plants. Therefore, either PS itself or a processed form must be stored until the plant is wounded. The sequence of PS (27) contains three motifs similar to the KEKE motifs found in mammalian proteins that are believed to be involved in protein–protein interactions (45). These motifs or other sequences may contain information for PS targeting, transport, or storage in the cell. In addition, the unusual hydrophilic nature of this prohormone may allow for its rapid mobilization on wounding, by allowing it to be soluble in a variety of conditions or making it accessible to processing enzymes when released by cellular damage. Although the data indicate that the repetitive elements contained in the PS sequence are not involved directly in wound-induced proteinase inhibitor accumulation induced by endogenously supplied PS, the data do not eliminate the possibility that, in vivo, active polypeptide signals are produced by specific processing enzymes and that the polypeptides might play roles in signaling.

The cumulative evidence presented herein suggests that large fragments of PS containing the SYS domain, or perhaps PS itself, can interact with the SYS receptor and activate the defense response in both tomato leaves and suspension-cultured tomato cells. The structure-activity data indicated further that the C-terminal region of PS containing the SYS peptide is required for activity. Although the data do not eliminate the possible role for the repeated sequences in other signaling or biochemical processes, they do indicate that the activation of the wound response is initiated exclusively by the SYS domain of the precursor. The properties of PS, although similar in some aspects to animal and yeast prohormones as well as membrane-anchored factors, suggest that the pathway for synthesis, compartmentalization, and processing of this plant prohormone is unique.

Abbreviations

PS

prosystemin

SYS

systemin

ICWF

intercellular wash fluid

CM

cell medium