PMC

Production of AVR2 and Inhibition in AF.
(A) Recombinant AVR2 triggers hypersensitive cell death in tomato leaves of MM-Cf2 plants. Recombinant AVR2 was injected into leaves of MM-Cf2 and MM-Cf2/rcr3-3 plants. Picture was taken 4 d after injection.
(B) Inhibition of PLCPs in AFs by AVR2. AF of BTH-treated tomato plants were preincubated with 66 nM AVR2 before adding DCG-04 to label the remaining noninhibited proteases. Biotinylated proteins were detected using streptavidin-HRP. A representative of three independent experiments is shown.
(C) Fluorescent protease activity profiling of PLCPs in AFs in presence or absence of AVR2. AFs of BTH-treated tomato plants were preincubated with 83 nM AVR2 before adding TMR-DCG-04 to label the remaining noninhibited proteases. Proteins were separated by 2D gel electrophoresis, and fluorescent proteins were visualized. A representative of three independent experiments is shown.

Identification of BTH-Induced Tomato Apoplastic Proteases.
(A) Identification of in vitro–labeled tomato PLCPs from AFs. AF of BTH-treated leaves was labeled with DCG-04. After labeling, biotinylated proteins were purified and separated on a 2D gel. Proteins were isolated from Coomassie blue–stained spots/bands and analyzed by tandem MS.
(B) Identification of in vivo–labeled tomato PLCPs from AFs. Tomato leaflets (183) were vacuum infiltrated with 2 μM DCG-04 and labeled in situ. AF was isolated, and biotinylated proteins were purified and separated on protein gel. A small portion of the sample was analyzed on protein blots with streptavidin-HRP to visualize the biotinylated proteins (left panel). Proteins were isolated from Coomassie blue–stained spots/bands and analyzed by tandem MS.
(C) RCR3 is not a major protease in AFs of BTH-treated tomato plants. Cf2/Rcr3 and Cf2/rcr3-3 plants were treated with BTH for 5 d, and AF was isolated. AF was labeled with DCG-04, and biotinylated proteins were displayed on a protein blot using streptavidin-HRP. A representative of three independent experiments is shown.

The Naturally Occurring N194D Mutation in RCR3 Affects Inhibition by AVR2.
(A) Position of variant residues in a structural model of RCR3. Differences between the RCR3^lyc and RCR3^pim alleles (orange) were not affecting AVR2 inhibition (). Four variant residues (red) were selected for targeted mutagenesis.
(B) The RCR3^N194D mutant is less sensitive for AVR2 inhibition. Extracts from agroinfiltrated leaves expressing (mutant) RCR3 proteins were preincubated at pH 5.0 with or without 65 nM AVR2 (A) or 20 μM E-64 (E). DCG-04 was added to label the remaining noninhibited proteases, and biotinylated proteins were detected on protein blots using streptavidin-HRP. The experimental differences in accumulation of RCR3 is shown using RCR3 antibodies. A representative of four independent assays is shown.

Inhibition of Agroinfiltrated Tomato Proteases by AVR2.
(A) Extracts from agroinfiltrated N. benthamiana leaves overexpressing different tomato proteases (indicated on the left) were preincubated for 30 min with 66 nM AVR2. DCG-04 was added after preincubation to label the noninhibited proteases. Biotinylated proteins were visualized on protein blots using streptavidin-HRP. Representatives of at least three independent experiments are shown.
(B) Concentration dependency of inhibition by AVR2. Protease-containing extracts were incubated with different AVR2 concentrations for 30 min. DCG-04 was added after preincubation to label the noninhibited proteases. Biotinylated proteins were visualized on protein blots using streptavidin-HRP. A representative of three independent experiments is shown.
(C) pH dependency of inhibition by AVR2. Extracts from agroinfiltrated leaves overexpressing the proteases were preincubated for 30 min at different pH with or without 66 nM AVR2. DCG-04 was added after preincubation to label the noninhibited proteases. Biotinylated proteins were visualized on protein blots using streptavidin-HRP. A representative of three independent experiments is shown.
(D) AVR2 physically interacts with PIP1. Extracts from agroinfiltrated leaves expressing His-tagged PIP1 were preincubated with or without an excess E-64 and incubated with FLAG-AVR2 at pH 5. Protein complexes containing PIP1-His were purified using Ni-NTA columns and analyzed using anti-PIP and anti-FLAG antibodies. A representative of three independent experiments is shown.

Induction of PIP1 and RCR3 upon BTH Treatment by Fluorescent Protease Activity Profiling and Real-Time RT-PCR.
(A) Fluorescent protease activity profiling of AFs. Plants were treated with BTH for 5 d, and AF was isolated. Equal volumes of AF were used for protease activity profiling with TMR-DCG-04. Proteins were separated by 1D (top) and 2D (bottom) gel electrophoresis, and fluorescent proteins were detected by fluorescent protease activity profiling. Please note that only signals coinciding with PIP1 are significantly stronger from BTH-treated plants. A representative of three independent experiments is shown.
(B) Induction of transcript levels by BTH treatment. Tomato leaves were harvested at 5 d after water (H) or BTH (B) treatment. Quantitative real-time RT-PCR was performed using gene-specific primers. The difference in threshold cycles (dCt) between the protease transcript and ubiquitin transcripts was calculated from three independent samples. Error bars represent sd. A representative of five independent biological experiments is shown.

BTH-Induced Protease Activities in the Tomato Apoplast.
(A) Apoplastic PR accumulation in BTH-treated tomato plants. Five-week-old tomato plants were treated with water or BTH. AFs were isolated after 5 d, and equal volumes were separated on a 17% protein gel to visualize PR protein accumulation in the AF.
(B) Protease activity profiling on equal volumes of total extract and AF of combined leaves from water- or BTH-treated tomato plants. Extracts were labeled at pH 5.0 with DCG-04 in the absence or presence of an excess of E-64. Proteins were separated on 12% protein gels, and biotinylated proteins were detected on a protein blot with streptavidin-HRP. Total extracts corresponds to ∼0.17 cm² leaf area/lane; AF corresponds to ∼7.4 cm² leaf area/lane. A representative of seven independent experiments is shown. An enlarged Coomassie blue–stained gel (cbb) shows equal loading and the accumulation of PR3 proteins.
(C) Protease activity profiling of AFs from young leaves of water- and BTH-treated tomato plants. A Coomassie blue–stained gel (cbb) shows equal loading and the accumulation of PR3 proteins.
(D) Protease activity profiling on AFs isolated from leaves of different age of the same water-treated plant (indicated in illustration on the right). E-64 competition was incomplete, resulting in some remaining 25-kD signal.

Sequence Analyses of Proteases from Tomato Relatives.
(A) Summary of amino acids encoded by variant codons in the protease domains of C14, PIP1, RCR3, CYP3, ALP, CatB1, and CatB2 alleles sequenced from various wild tomato relatives (indicated top right). Amino acids encoded by the variant codons are summarized by leaving out the amino acids of nonvariant codons from the protein alignment. Amino acids encoded by codons different from the S. lycopersicum (lyc) allele are indicated with gray, blue, and red residues if they are identical, similar, or nonsimilar, respectively, compared with the lyc sequence. Dashes indicate missing sequence information. RCR3 of S. cheesmanniae is not shown since it contained a premature stop codon and could be amplified from genomic DNA and not from cDNA (see Supplemental Alignment 2 online).
(B) Number of single nucleotide (nt) polymorphisms per protease.
(C) Ratio of nonsimilar/similar amino acid (aa) substitutions calculated from (A). PIP1 and RCR3 are under diversifying selection; the other proteases are under conservative selection.
(D) Position of variant residues in structural models of PIP1 and RCR3. Positions with nonsimilar variance and similar variance are indicated in red and blue, respectively.

In Silico Analysis of Identified Secreted Tomato Proteases.
(A) Domains encoded by the open reading frames (ORFs) of the identified secreted tomato proteases. The peptides that are reproducibly found in MS analysis are indicated with black bars. sp, signal peptide; pro-, autoinhibitory prodomain; Cys, catalytic Cys residue; protease, mature protease domain; p, Pro-rich domain; granulin, granulin-like domain; *, vacuolar targeting signal NPIR; pI/MW, isoelectric point and molecular weight of the protease after removal of the signal peptide and prodomain. The color coding represents the similarity-colored subfamilies in (B).
(B) Unrooted phylogenetic tree of 145 plant papain-like Cys proteases showing the diversity of the identified tomato proteases (colored names). The classes are subdivided into eight subfamilies (). Accession codes of other plant PLCPs (numbered) are in Supplemental Data Set 1 online. S24988 is a tomato protease of group 6 that was not identified in AFs in this study. Bootstrap values for critical nodes are indicated in gray bold italics.
(C) Conservation of proteases and control proteins within solanaceous species. tBLASTp searches of the tomato proteins to translated EST libraries of tobacco, pepper, N. benthamiana, potato, and petunia resulted in the identification of close homologs from each of these solanaceous species. The amino acid identity over the protease domain was scored and combined for the different species. Resistance proteins, PR proteins, and various household proteins were used as controls (gray; see Methods). Asterisk indicates mature proteins used for tBLASTp analysis. Error bars represent sd.