We have investigated the regulation of gene expression by the plant hormone ethylene by cloning mRNAs that accumulate in unripe tomato fruit (Lycopersicon esculentum) exposed to exogenous ethylene. The response to exogenous ethylene is rapid; within 30-120 min we detect an increase in the cloned mRNA concentrations. DNA sequence analysis indicates that one of the ethylene-inducible genes is related to a gene encoding wound-inducible proteinase inhibitor I. We have measured ethylene production during fruit development and detect low basal levels in unripe fruit and much higher levels in ripening fruit. Blot hybridization experiments show that expression of the cloned genes is develop-

A method for isolating and cloning mRNA populations from individual cells in living, intact plant tissues is described. The contents of individual cells were aspirated into micropipette tips filled with RNA extraction buffer. The mRNA from these cells was purified by binding to oligo(dT)-linked magnetic beads and amplified on the beads using reverse transcription and PCR. The cell-specific nature of the isolated mRNA was verified by creating cDNA libraries from individual tomato leaf epidermal and guard cell mRNA preparations. In testing the reproducibility of the method, we discovered an inherent limitation of PCR amplification from small amounts of any complex template. This phenomenon, which we have termed the "Monte Carlo" effect, is created by small and random differences in amplification efficiency between individual templates in an amplifying cDNA population. The Monte Carlo effect is dependent upon template concentration: the lower the abundance of any template, the less likely its true abundance will be reflected in the amplified library. Quantitative assessment of the Monte Carlo effect revealed that only rare mRNAs (<0.04% of polyadenylylated mRNA) exhibited significant variation in amplification at the single-cell level. The cDNA cloning approach we describe should be useful for a broad range of cell-specific biological applications.

The sphinganine analog mycotoxin, AAL-toxin, induces a death process in plant and animal cells that shows apoptotic morphology. In nature, the AAL-toxin is the primary determinant of the Alternaria stem canker disease of tomato, thus linking apoptosis to this disease caused by Alternaria alternata f. sp. lycopersici. The product of the baculovirus p35 gene is a specific inhibitor of a class of cysteine proteases termed caspases, and naturally functions in infected insects. Transgenic tomato plants bearing the p35 gene were protected against AAL-toxin-induced death and pathogen infection. Resistance to the toxin and pathogen co-segregated with the expression of the p35 gene through the T3 generation, as did resistance to A. alternata, Colletotrichum coccodes, and Pseudomonas syringae pv. tomato. The p35 gene, stably transformed into tomato roots by Agrobacterium rhizogenes, protected roots against a 30-fold greater concentration of AAL-toxin than control roots tolerated. Transgenic expression of a p35 binding site mutant (DQMD to DRIL), inactive against animal caspases-3, did not protect against AAL-toxin. These results indicate that plants possess a protease with substrate-site specificity that is functionally equivalent to certain animal caspases. A biological conclusion is that diverse plant pathogens co-opt apoptosis during infection, and that transgenic modification of pathways regulating programmed cell death in plants is a potential strategy for engineering broadspectrum disease resistance in plants.

We have investigated the mechanism of action of the plant hormone ethylene by analyzing the expression of ethylene-inducible genes isolated from tomato (Lycopersicon esculentum). We have found that the expression of each cloned gene is regulated by ethylene in a unique manner. That is, for certain genes ethylene affects transcriptional processes, while for another gene it affects both transcriptional and post-transcriptional processes. Furthermore, induction of gene transcription by ethylene is organ specific for one gene, while for others it is not. In addition, we have measured gene expression as a function of ethylene concentration and have found that each gene displays a unique ethylene dose-response curve. Our results suggest that ethylene modulates gene expression by a variety of mechanisms.