Plant Physiol, December 2001, Vol. 127, pp. 1414-1417

SCIENTIFIC CORRESPONDENCE

Genetic Analysis of Wound Signaling in Tomato. Evidence for a Dual Role of Jasmonic Acid in Defense and Female Fertility

Department of Energy-Plant Research Laboratory (L.L., C.L., G.A.H.) and Department of Biochemistry and Molecular Biology (G.A.H.), Michigan State University, East Lansing, Michigan 48824

	INTRODUCTION

TOP INTRODUCTION GENETIC ANALYSIS OF WOUND... A ROLE FOR JA... LITERATURE CITED

Genetic analysis of the wound response pathway in tomato (Lycopersicon esculentum) indicates that prosystemin and systemin are upstream components of a defensive signaling cascade that involves complex regulation of jasmonic acid (JA) biosynthesis and the ability of cells to perceive and respond to JA. Recent identification of JA response mutants provides evidence for the hypothesis that the JA transduction pathway also plays an important role in female reproductive development.

Many plants respond to insect attack and wounding by synthesizing an array of phytochemicals that decrease the ability of herbivores to colonize, feed, or reproduce on the plant (Green and Ryan, 1972; Karban and Baldwin, 1997). Wound-inducible proteinase inhibitors (PIs) provide an attractive model system in which to study the signal transduction pathways that regulate this form of defense. In tomato, damage to a single leaflet by mechanical wounding or herbivory results in localized and systemic expression of two Ser PI-encoding genes (Inh-I and Inh-II) within about 2 h (Ryan, 2000; Howe et al., 2000). These proteins can accumulate to high levels in leaves of the damaged plant, where they play a defensive role by inhibiting digestive proteases of some lepidopteran insects. In their pioneering study of wound-inducible PIs 30 years ago, Green and Ryan (1972) proposed that chemical signals generated at the site of wounding traverse the vascular system to activate the systemic expression of PIs. Although many of the signals involved in this response have been identified, relatively little is known about the mechanisms by which they are produced and transported between cells.

A unique component of the wound response pathway in tomato is the peptide signal systemin and the precursor protein prosystemin, from which it is derived (Pearce et al., 1991; McGurl et al., 1992). Tomato prosystemin is encoded by a single gene whose primary transcript is alternatively spliced to generate two active forms of the protein (Li and Howe, 2001). Several lines of genetic evidence indicate that prosystemin is essential for wound-induced expression of PI and other defense-related genes. First, transgenic plants expressing an antisense prosystemin cDNA are deficient in wound-induced systemic expression of PI genes (McGurl et al., 1992). Second, overexpression of prosystemin from a 35S::prosystemin transgene constitutively activates PI expression in unwounded plants (McGurl et al., 1994). Third, mutations that suppress 35S::prosystemin-mediated signaling block wound induction of PIs (Howe and Ryan, 1999). It has been proposed that systemin functions as a mobile wound signal following its proteolytic release from prosystemin (McGurl et al., 1992). Expression of PI genes in tomato leaves in response to wounding and systemin is mediated by JA, a terminal product of the octadecanoid pathway (Farmer and Ryan, 1992; Creelman and Mullet, 1997). This model has been refined to reflect the fact that wound- and systemin-induced expression of PIs involves synergism between JA and ethylene (O'Donnell et al., 1996). Recent studies provide evidence that reactive oxygen species function downstream of JA to amplify wound- and systemin-induced responses (Orozco-Cárdenas et al., 2001). Due to space limitations, the reader is referred to recent reviews for a detailed discussion of the wound-signaling pathway (Bowles, 1998; Ryan, 2000; Walling, 2000; León et al., 2001).

	GENETIC ANALYSIS OF WOUND SIGNALING

TOP INTRODUCTION GENETIC ANALYSIS OF WOUND... A ROLE FOR JA... LITERATURE CITED

We are using tomato as a model system for genetic dissection of signaling pathways that regulate wound responses and, more broadly, defense against herbivores. To further define the function of prosystemin and systemin in the wound response, we conducted a screen for mutations that suppress 35S::prosystemin-mediated expression of downstream target genes (Howe and Ryan, 1999). We identified 13 independent mutants, designated spr (suppressed in prosystemin-mediated responses). Eight mutants define four genetic complementation groups called Spr-1, -2, -3, and -4. Two mutants define new alleles of def-1, a JA-deficient mutant that is compromised in wound-inducible PI expression and resistance to Manduca sexta larvae (Lightner et al., 1993; Howe et al., 1996). The three remaining mutants were sterile and thus were not further characterized in the initial study. Mutations in Def-1, Spr-1, and Spr-2, in addition to suppressing the action of 35S::prosystemin, impair wound- and systemin-induced PI expression. This finding provides strong genetic evidence that prosystemin is an essential upstream component in the wound response pathway. The ability of def-1, spr-1, and spr-2 plants to respond to exogenous JA suggests that these mutations affect processes required for JA biosynthesis or accumulation (Fig. 1). Support for this interpretation comes from the finding that def-1 plants are deficient in JA accumulation in response to wounding and systemin (Howe et al., 1996).

View larger version (8K): [in this window] [in a new window]   Figure 1.   Proposed action of mutations in the wound response pathway. The signaling pathway depicted is consistent with the model proposed by Farmer and Ryan (1992). All mutants listed are deficient in wound-inducible systemic expression of PIs and also lack PI expression in response to systemin and 35S::prosystemin. def-1, spr-1, spr-2, and spr-5 plants are responsive to applied MeJA and JA, whereas jai-1 plants are insensitive to these signals. See text for details.

Identification of mutants that are impaired in JA perception would provide a valuable tool to further elucidate the mechanism of wound signaling and its relationship to induced defense. Toward this goal, we screened a fast-neutron-mutagenized population of tomato (cv Micro-Tom) for plants that fail to express polyphenol oxidase and Inh-II upon exposure to gaseous methyl-JA (MeJA; Li et al., 2001). One mutant, designated jasmonic acid-insensitive-1 (jai-1), completely lacked these proteins upon treatment with MeJA or wounding. jai-1 plants displayed normal vegetative growth but produced fruit that lacked mature seed. In over 1,000 jai-1 fruit examined, only two viable seeds were recovered. Reciprocal crosses to wild-type showed that jai-1 is female-sterile; it failed to set seed following pollination with wild-type pollen but readily pollinated and fertilized wild-type pistils. F₁ plants derived from this cross were fully responsive to JA/MeJA and fertile. In a segregating population (108 F₂ plants), the JA-insensitive and sterile phenotypes always co-segregated as if conditioned by a single recessive mutation.

The female-sterile phenotype of jai-1 plants prompted us to investigate the sterile spr lines that were previously generated by ethyl methane sulfonate-mutagenesis (see above; Howe and Ryan, 1999). Attention was focused on two lines, 124A and 436G, that developed flowers but produced either no fruit or fruit containing no viable seed. Reciprocal backcrosses to wild-type indicated that both lines were female-sterile. Analysis of F₂ populations derived from these crosses showed that one-quarter of the progeny lacked wound-inducible Inh-II expression both in the wounded leaf and the undamaged systemic leaf (Table I). Wound-insensitive 436G plants accumulated normal levels of Inh-II in response to exogenous MeJA (Table I), similar to the phenotype of def-1, spr-1, and spr-2 plants (Fig. 1). Complementation tests showed that the recessive mutation harbored by 436G defines a novel locus, designated Spr-5. Backcrossed lines that are homozygous for spr-5 produced viable seed, albeit at reduced levels relative to wild type. This finding indicates that the sterile phenotype of 436G can be attributed in part to a mutation other than spr-5.

View this table: [in this window] [in a new window]   Table I.   Proteinase inhibitor II accumulation in response to wounding and MeJA Values indicate the mean ± SD of Inh-II levels (µg ml1 leaf juice) in leaf tissue of wild-type and two mutant lines that are suppressed in 35S::prosystemin-mediated signaling.

In contrast to spr-5, MeJA-treated 124A plants expressed no detectable Inh-II (Table I) or polyphenol oxidase (data not shown). This indicates that 124A, like jai-1, is blocked in JA perception or the ability to respond appropriately to the hormone. Complementation tests between 124A and jai-1 were performed using heterozygous maternal parents, and the results showed that the two mutants define the same locus (Table II). We henceforth refer to the fast-neutron allele of jai-1 as jai-1¹ and the EMS allele of 124A as jai-1². Plants homozygous for jai-1² were also unresponsive to relatively high concentrations of JA (50 nmol plant¹) and systemin (5 pmol plant¹) supplied through the cut stem (Fig. 2). Recovery of a JA-insensitive mutant in a screen for suppressors of 35S::prosystemin is significant because it demonstrates that 35S::prosystemin-mediated signaling requires a functional JA response pathway. The unresponsiveness of jai-1 plants to wounding and applied systemin likewise indicates that JA action is essential for wound- and systemin-induced PI expression (Fig. 1).

View larger version (20K): [in this window] [in a new window]   Figure 2.   Proteinase inhibitor II levels in wild-type and jai-12 plants in response to JA and systemin. Fifteen-day-old seedlings were supplied with buffer (B; 15 mM sodium phosphate, pH 7.0), JA (J; 50 nmol plant1), or systemin (S; 5 pmol plant1) through the cut stem. Inh-II levels were measured in the leaves 24 h later. Plants homozygous for jai-12 were selected as described in Table I. Values represent the mean ± SD of at least six plants.

	A ROLE FOR JA IN FEMALE REPRODUCTIVE DEVELOPMENT?

TOP INTRODUCTION GENETIC ANALYSIS OF WOUND... A ROLE FOR JA... LITERATURE CITED

The JA insensitivity and female sterility of independent mutants (jai-1¹ and jai-1²) argues against the possibility that the two phenotypes result from mutations in different genes. Rather, our results support the hypothesis that a single gene, Jai-1, is required both for wound-induced expression of defensive genes and female reproductive development. Conclusive proof will require cloning of Jai-1 and functional complementation of the mutant. The sterility of jai-1 plants could result from a defect in ovule development, embryogenesis, or another maternal process required for seed production. A dysfunction in embryogenesis would be consistent with previous studies implicating JA as an endogenous regulator of embryo development in oilseeds (Wilen et al., 1991). A similar situation could exist in tomato, where JA, its precursor 12-oxo-phytodienoic acid, and various amino acid conjugates of JA are abundant in female organs of the flower (Hause et al., 2000).

A proposed role for JA in female reproductive development in tomato stands in contrast to well-documented studies in Arabidopsis where JA biosynthesis and perception are essential for male, but not female, gametophyte development (McConn and Browse, 1996; Feys et al., 1994; Sanders et al., 2000; Stintzi and Browse, 2000). Although we cannot exclude a role for JA in male gametophyte development in tomato, the ability of jai-1 pollen to induce normal seed set when crossed to a wild-type pistillate parent indicates that JA perception and downstream signaling events are not essential for the production of viable pollen. How might such species-specific differences in jasmonate function be explained? One possibility is that oxylipins such as JA first evolved as low-abundance signaling molecules for the regulation of stress responses (e.g. defense) in vegetative tissues and, subsequently, these compounds were recruited to perform other physiological functions (e.g. reproduction) in specific plant lineages. This scenario was previously suggested to account for species-specific differences in the requirement for flavonoids in male fertility (Burbulis et al., 1996). The range of physiological processes controlled by JA may ultimately reflect the function of specific genes whose expression is regulated by the hormone in a tissue- or cell type-specific manner. The wound response mutants described herein should provide useful tools to investigate the molecular mechanisms by which jasmonates regulate diverse physiological processes.

	FOOTNOTES

Received August 9, 2001; accepted September 2, 2001.

^* Corresponding author; e-mail howeg{at}msu.edu; fax 517-353-9168.

¹ This research was supported by the National Institutes of Health (grant no. GM57795) and the U.S. Department of Energy (grant no. DE-FG02-91ER20021).

www.plantphysiol.org/cgi/doi/10.1104/pp.010705.

	LITERATURE CITED

TOP INTRODUCTION GENETIC ANALYSIS OF WOUND... A ROLE FOR JA... LITERATURE CITED

• Bowles D (1998) Phil Trans R Soc Lond B 353: 1495-1510[CrossRef][ISI][Medline]
• Burbulis IE, Iacobucci M, Shirley BW (1996) Plant Cell 8: 1013-1025[Abstract]
• Creelman RA, Mullet JE (1997) Annu Rev Plant Physiol Plant Mol Biol 48: 355-381[CrossRef][ISI][Medline]
• Farmer EE, Ryan CA (1992) Plant Cell 4: 129-134[Abstract/Free Full Text]
• Feys BJF, Benedetti CE, Penfold CN, Turner JG (1994) Plant Cell 6: 751-759[Abstract/Free Full Text]
• Green TR, Ryan CA (1972) Science 175: 776-777[Abstract/Free Full Text]
• Hause B, Stenzel I, Miersch O, Maucher H, Kramell R, Ziegler J, Wasternack C (2000) Plant J 24: 113-126[CrossRef][ISI][Medline]
• Howe GA, Lee GI, Itoh A, Li L, DeRocher A (2000) Plant Physiol 123: 711-724[Abstract/Free Full Text]
• Howe GA, Lightner J, Browse J, Ryan CA (1996) Plant Cell 8: 2067-2077[Abstract]
• Howe GA, Ryan CA (1999) Genetics 153: 1411-1421[Abstract/Free Full Text]
• Karban R, Baldwin IT (1997) Induced Responses to Herbivory. Chicago University Press, Chicago
• León J, Rojo E, Sánchez-Serrano JJ (2001) J Exp Bot 52: 1-9[Abstract/Free Full Text]
• Li L, Howe GA (2001) Plant Mol Biol 46: 409-419[CrossRef][ISI][Medline]
• Li L, Shaffer D, Howe GA (2001) Annual Meeting of the American Society of Plant Physiologists, July 2001, Providence, Rhode Island(Abstract no. 128)
• Lightner J, Pearce G, Ryan CA, Browse J (1993) Mol Gen Genet 241: 595-601[CrossRef][ISI][Medline]
• McConn M, Browse J (1996) Plant Cell 8: 403-416[Abstract]
• McGurl B, Orozco-Cárdenas M, Pearce G, Ryan CA (1994) Proc Natl Acad Sci USA 91: 9799-9802[Abstract/Free Full Text]
• McGurl B, Pearce G, Orozco-Cárdenas M, Ryan C (1992) Science 255: 1570-1573[Abstract/Free Full Text]
• O'Donnell PJ, Calvert C, Atzorn R, Wasternack C, Leyser HMO, Bowles DJ (1996) Science 274: 1914-1917[Abstract/Free Full Text]
• Orozco-Cárdenas M, Narváez-Vásquez J, Ryan CA (2001) Plant Cell 13: 179-191[Abstract/Free Full Text]
• Pearce G, Strydom D, Johnson S, Ryan CA (1991) Science 253: 895-989[Abstract/Free Full Text]
• Ryan CA (2000) Biochim Biophys Acta 1477: 112-121[CrossRef][Medline]
• Sanders PM, Lee PY, Biesgen C, Boone JD, Beals TP, Weiler EW, Goldberg RB (2000) Plant Cell 12: 1041-61[Abstract/Free Full Text]
• Stintzi A, Browse J (2000) Proc Natl Acad Sci USA 97: 10625-10630[Abstract/Free Full Text]
• Walling LL (2000) J Plant Growth Regul 19: 195-216[Medline]
• Wilen RW, van Rooijen GJH, Pearce DW, Pharis RP, Holbrook LA, Moloney MM (1991) Plant Physiol 95: 399-405[Abstract/Free Full Text]