Cloning, expression, and characterization of tomato (Lycopersicon esculentum) aminopeptidase P.

A cDNA (LeAPP2) was cloned from tomato coding for a 654 amino acid protein of 72.7 kDa. The deduced amino acid sequence was >40% identical with that of mammalian aminopeptidase P, a metalloexopeptidase. All amino acids reported to be important for binding of the active site metals and catalytic activity, respectively, were conserved between LeAPP2 and its mammalian homologues. LeAPP2 was expressed in Escherichia coli in N-terminal fusion with glutathione S-transferase and was purified from bacterial extracts. LeAPP2 was verified as an aminopeptidase P, hydrolyzing the amino-terminal Xaa-Pro bonds of bradykinin and substance P. LeAPP2 also exhibited endoproteolytic activity cleaving, albeit at a reduced rate, the internal -Phe-Gly bond of substance P. Apparent K(m) (15.2 +/- 2.4 microm) and K(m)/k(cat) (0.94 +/- 0.11 mm(-1) x s(-1)) values were obtained for H-Lys(Abz)-Pro-Pro-pNA as the substrate. LeAPP2 activity was maximally stimulated by addition of 4 mm MnCl(2) and to some extent also by Mg(2+), Ca(2+), and Co(2+), whereas other divalent metal ions (Cu(2+), Zn(2+)) were inhibitory. Chelating agents and thiol-modifying reagents inhibited the enzyme. The data are consistent with LeAPP2 being a Mn(II)-dependent metalloprotease. This is the first characterization of a plant aminopeptidase P.

Proline is unique among the proteinogenic amino acids in that its side chain is bonded to both the ␣-carbon and the amino group. The resulting cyclic structure imposes conformational restraints on proline-containing peptides relevant for structure and function of many physiologically important biomolecules. A key role for proline residues is the protection against nonspecific proteolytic degradation. Hence proline is frequently found and conserved in peptide hormones, neuropeptides, and growth factors (1)(2)(3). Many bioactive polypeptides share a Xaa-Pro motif at their N termini shielding them against nonspecific N-terminal degradation. The degradation of these peptides requires proteases with specificity for the Xaa-Pro motif including proline-selective dipeptidases (dipeptidyl peptidases II and IV, cleaving the post-Pro bond) and aminopeptidase P (Xaa-Pro aminopeptidase, cleaving the pre-Pro bond). Cleavage of the Xaa-Pro motif by either one of these peptidases may initiate the proteolytic degradation/inactivation of the peptide or may re-sult in an altered bioactivity (2)(3)(4).
Aminopeptidase P (APP, 1 EC 3.4.11.9) was first isolated from Escherichia coli (5) and has subsequently been characterized from many microbial and mammalian sources (reviewed in Ref. 4). Mammalian APPs are now known to comprise at least two distinct forms, a cytosolic form and a membrane-bound form attached to the plasma membrane by a glycosylphosphatidylinositol (GPI) anchor (6 -14). APPs hydrolyze the peptide bond between any amino acid and a penultimate proline residue at the N termini of oligopeptide and protein substrates. A free amino group is required at the N terminus and the scissile bond must be in the trans configuration (15). The hydrolysis of dipeptides is very slow compared with the hydrolysis of longer chains, indicating the existence of a third subsite for substrate binding, which was confirmed for E. coli and mammalian APPs (10,15). Likely physiological substrates of APP include bradykinin, substance P, and peptide-YY (15)(16)(17)(18), and APP has been implicated in the regulation of cardiovascular and pulmonary functions in vivo (19 -21).
In higher plants, only very few peptides with hormone-like functions are presently known (22,23), but a more general role for peptides as signal molecules in the regulation of plant defense, growth, and development is anticipated (24). Likewise, the proteases involved in the maturation and degradation of plant peptide hormones are still elusive. In the present work, we used a partial cDNA as a probe to isolate the cDNAs of two APPs from tomato. One of the enzymes (LeAPP2) was functionally expressed in E. coli, purified from bacterial extracts, and characterized biochemically. This is the first characterization of an APP from any plant source.

EXPERIMENTAL PROCEDURES
Cloning of LeAPP2-All basic molecular techniques were adapted from published protocols (25,26). A cDNA library from tomato shoot tissue in ZAP (Stratagene, La Jolla, CA) was used (27) and 9.5 ϫ 10 6 plaque-forming units were screened on nitrocellulose membranes. A partial tomato cDNA (LeAPP1) 2 with sequence similarity to human aminopeptidase P was used as a probe. Hybridization with the radiolabeled cDNA ('Prime-It' system, Stratagene) was performed at 42°C in 50% formamide, 5ϫ SSC (1ϫ SSC is 0.15 M NaCl, 0.015 M sodium citrate), 0.5% SDS, 2ϫ Denhardt's solution (1ϫ Denhardt's solution: 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 0.02% bovine serum albumin), 50 mM potassium phosphate buffer (pH 7.0), and 200 g/ml of denatured salmon sperm DNA. Filters were washed in 0.5ϫ SSC, 0.5% SDS at 60°C and were subsequently exposed to x-ray film (Kodak X-Omat AR) using an intensifying screen. Individual positive phage clones were identified in two consecutive rounds of screening. Recombinant pBluescript cDNA phagemids were excised in vivo using the ExAssist helper-phage according to the recommended procedure (Stratagene). RACE-PCR was performed to obtain full-length cDNAs using the SMART RACE cDNA amplification kit (CLONTECH, Palo Alto, CA) according to the manufacturer's instructions. In a first step, singlestranded cDNA was synthesized from total RNA of tomato leaves using M-MLV reverse transcriptase (Promega, Madison, WI) and oligo(dT) as the primer. Subsequently, the full-length LeAPP1 and LeAPP2 cDNAs were amplified using gene-specific primers (5Ј-CCTTGCAAAAGCTCT-GAAGAACCCTGTT-3Ј and 5Ј-AATGAGGCATGTCAAGCTGC-3Ј (Microsynth, Balgach, Switzerland) and the universal primer provided with the kit. The PCR products were gel-purified and cloned into pCR-Script (Stratagene). The identity of all PCR-generated clones was confirmed by sequence analysis of at least three independent PCR products using fluorescent dideoxy chain terminators in the cycle sequencing reaction (PerkinElmer Life Sciences) and the Applied Biosystems model 373A DNA sequencer.
Northern and Southern Blot Analysis-RNA was isolated from different tissues of tomato plants using a phenol-based extraction procedure (26). Total RNA (5 g) was separated on formaldehyde/agarose gels and transferred to nitrocellulose membranes according to standard protocols (26). For Southern blot analysis, genomic DNA was extracted from tomato leaf tissue using the Nucleon Phytopure DNA extraction kit (Amersham Pharmacia Biotech). Ten g of DNA were restricted using the enzymes indicated in the legend to Fig. 2 and were subjected to DNA gel blot analysis using the radiolabeled 3Ј-untranslated regions of the LeAPP cDNAs as probes. Hybridization, washing, and evaluation of blots were done as described (28).
Expression of LeAPP2-The open reading frame of LeAPP2 was amplified by PCR using Pfu Turbo DNA polymerase (Stratagene) and synthetic oligonucleotide primers (forward primer: 5Ј-ATGGCGGATA-CACTCGCAGC-3Ј; reverse primer: 5Ј-GGGGTACCTAAGAGCACT-GAACATCTCA-3Ј (Microsynth, Balgach Switzerland)). The PCR product was cloned into the StuI/KpnI sites of pGEX-G (29) a derivative of pGEX-3x (Amersham Pharmacia Biotech), to yield pGEX-APP2. This construct allows expression of LeAPP2 in N-terminal fusion with glutathione S-transferase (GST) in E. coli under control of the IPTGinducible tac promoter. The expression construct was verified by sequence analysis and was transformed into E. coli BL21 codon plus (DE3)-RIL (Stratagene). Cultures (500 ml) were grown at 37°C to an A 600 of 1.0, IPTG was added to a final concentration of 1 mM, and cultures were grown for another 2 h at room temperature. The cells were harvested by centrifugation (2500 ϫ g, 15 min, 4°C) and stored at Ϫ80°C.
Purification of Recombinant LeAPP2-E. coli cells were resuspended in 20 ml of buffer A (50 mM Tris-HCL, pH 8.0, 10 mM NaCl, 1 mM EDTA) containing 0.1 mg/ml DNaseI and 1 mg/ml lysozyme. After 20 min at room temperature, cells were lysed by sonication. The cell debris was removed by centrifugation (35000 ϫ g, 15 min, 4°C), and the supernatant was subjected to affinity chromatography on a glutathione-Sepharose 4B column (Amersham Pharmacia Biotech; 1-ml bed volume) equilibrated in buffer A. After extensive washing with buffer A and buffer B (50 mM Tris/HCl, pH 8.0), the fusion protein (GST⅐LeAPP2) was eluted with buffer B containing 10 mM reduced glutathione. The progress of protein purification was monitored by SDS-polyacrylamide gel electrophoresis performed on 10% polyacrylamide gels using the buffer system described by Laemmli (30). Gels were stained for protein detection using Coomassie Brilliant Blue R250.
Steady State Kinetic Analyses-For continuous assay of LeAPP2 activity, H-Lys(Abz)-Pro-Pro-pNA (Bachem; Bubendorf, Switzerland) was used as an internally quenched fluorogenic substrate (31). Unless stated otherwise, the assays contained 125 ng of the GST⅐APP2 fusion protein and 50 M of the substrate in a total volume of 500 l of 0.1 M Tris-HCl pH 7.5, 4 mM MnCl 2 , and the fluorescence was recorded in a Kontron SFM 25 fluorimeter ( ex : 320 nm, em : 411 nm). To analyze the effect of different divalent metal ions on the reaction rate, the enzyme was incubated with the respective salt for 5 min before the reaction was started by addition of the substrate. The inhibitory effects of N-ethylmaleimide, 1, 10-phenanthroline, 2-mercaptoethanol, diethylpyrocarbonate, and EDTA were assayed after a preincubation period of 30 min.
MALDI-TOF/MS Assay of LeAPP2 Specificity-The degradation of bradykinin, substance P (Sigma), and systemin (Enzyme System Products, Livermore, CA; 50 M each) by the GST⅐APP2 fusion protein (250 nM) was analyzed using MALDI-TOF mass spectrometry. The reaction was performed at room temperature in a total volume of 50 l of 50 mM Tris-HCl pH 7.5, 4 mM MnCl 2 . 1-l aliquots were taken at intervals and mixed on the MALDI-TOF/MS sample plate with an equal volume of the crystallization matrix (2 parts of saturated 2,5-dihydroxybenzoic acid in acetone, 1 part 0.1% (v/v) trifluoroacetic acid). Crystals were washed repeatedly with deionized water and air-dried before recording the mass spectra with a Voyage Elite mass spectrometer using the reflectron mode for increased mass accuracy. Peptide masses were further analyzed using the program PAWS v. 8.1.1 (freeware edition for MacOS 7.5, copyright ProteoMetrics, 1997).

Molecular Cloning of LeAPP1 and -2-
A partial cDNA with sequence similarity to human APP was serendipitously isolated 3 in a genetic screen for the identification of proteases involved in either the maturation or the degradation/inactivation of systemin an oligopeptide signal molecule involved in wound signal transduction (32). The screen had been devised for the identification of endopeptidases with known specificity (33). Hence, the isolated cDNA, which encodes an APP, i.e. an exopeptidase, likely represents a false positive. The sequence was completed at its 5Ј-end by RACE-PCR and termed LeAPP1 (GenBank TM accession number AJ308541). Using the LeAPP1 cDNA as a probe, a tomato cDNA library was screened at moderate stringency resulting in the isolation of seven clones, four of which corresponded to LeAPP1 although three appeared to encode a second aminopeptidase P, i.e. LeAPP2. As indicated by a sequence comparison with the LeAPP1 cDNA, all three clones were incomplete at their 5Ј-ends, and RACE-PCR was performed to obtain the full-length sequence (GenBank TM accession number AJ310676). The LeAPP2 cDNA encompasses an open reading frame of 1962 base pairs coding for a protein of 654 amino acids with a calculated M r 72,682. The deduced amino acid sequence was compared with those of LeAPP1 and the cytosolic and membrane-bound forms of human aminopeptidase P (82, 46, and 41% sequence identity, respectively; Fig.  1). All amino acid residues implicated in the binding of active site metals (Asp-451, Asp-462, His-525, Glu-561, Glu-574; Ref. 34) and two further histidines (His-432 and His-535, numbers refer to the LeAPP2 sequence) proposed to play a role during catalysis in proton shuttling from the dinuclear metal center to the solvent (35) are conserved between the four sequences. There are no obvious sequence elements for targeting to subcellular compartments, and therefore, LeAPP2 is likely to be a cytosolic enzyme. This conclusion is in agreement with the higher overall similarity of LeAPP2 with the cytosolic form of human APP as compared with the membrane-bound form, which contains N-and C-terminal signal sequences (12). Whereas the sequence similarity between the tomato and human enzymes is high in the N-and C-terminal regions, a stretch of low similarity extends from amino acids 237 to 393 of LeAPP2 where there are two insertions in the plant sequences (amino acids 237-252 and 373-392 of LeAPP2) lacking a counterpart in the human proteins.
The 3Ј-untranslated regions of the LeAPP1 and 2 cDNAs were used as probes on gel blots of tomato genomic DNA (Fig.  2, A and B). The hybridization pattern obtained with the probe derived from the LeAPP1 cDNA may indicate the existence of a third, closely related gene in the haploid tomato genome (Fig.  2A). The LeAPP2 probe, on the other hand, was found to be gene-specific and was thus suitable for the analysis of LeAPP2 expression on RNA gel blots (Fig. 2C). A single class of LeAPP2 transcripts was detected in all tissues analyzed, being more prevalent in tomato roots and cultured cells than in flowers, leaves, and cotyledons (Fig. 2C).
Expression, Purification, and Catalytic Activity of LeAPP2-The open reading frame of the LeAPP2 cDNA was cloned into the expression vector pGEX-G and expressed in E. coli in N-terminal fusion with GST. From 1 liter of E. coli culture, 4 mg of soluble GST⅐LeAPP2 were purified by affinity chroma-tography (Fig. 3). The apparent molecular mass of the purified protein of 105 kDa is consistent with the mass expected for the GST (26 kDa)-LeAPP2 (72.7 kDa) fusion protein. A few minor contaminants with estimated masses of 70, 43, and 35 kDa, respectively, co-purified during affinity chromatography and likely represent degradation products of GST⅐LeAPP2 as indicated by N-terminal amino acid sequence analysis (data not shown). Efficient cleavage of the GST moiety of the fusion protein by factor Xa treatment proved to be impossible. Hence, the GST⅐LeAPP2 fusion protein was used in all further studies. GST⅐LeAPP2 could be stored at Ϫ20°C or, in concentrated solution (0.5 mg/ml), at 4°C for several weeks without a significant loss of activity.
The proteolytic activity of GST⅐LeAPP2 was assayed using an internally quenched, fluorogenic substrate (H-Lys(Abz)-Pro-Pro-pNA, (31)). The release of the N-terminal Lys(Abz) by aminopeptidase P activity was followed spectrofluorometri-cally. The reaction rate was a linear function of the protein concentration over the investigated range of 0 -2 mg/ml. LeAPP2 activity was affected by addition of divalent cations (Fig. 4A). Increasing concentrations of MnCl 2 and MgCl 2 stimulated LeAPP2 activity with MnCl 2 being most effective at a concentration of 4 mM. CaCl 2 and CoCl 2 stimulated LeAPP2 activity at 0.4 mM but were inhibitory at 4 mM. Complete inhibition of LeAPP2 activity was observed after addition of 4 mM ZnCl 2 or CuSO 4 (Fig. 4A). LeAPP2 activity was also inhibited by metal-chelating agents with 1,10-phenanthroline being much more effective than EDTA (half-maximal inhibition at 36 and 340 M, respectively). Furthermore, treatment with the FIG. 1. Sequence alignment of tomato and human APPs. The amino acid sequences deduced from the LeAPP1 (GenBank TM accession number AJ308541) and LeAPP2 (AJ310676) cDNAs are shown and compared with human cytosolic (hcAPP, AF272981) and membranebound (hmAPP, AF195953) APPs. The alignment was done with the program pileup of the University of Wisconsin GCG package and was processed using boxshade at www.ch.embnet.org. Identical residues and conservative replacements are shown with black and gray shading, respectively. Conserved active site residues implicated in metal binding (*) and catalysis (q) are indicated. histidine-alkylating reagent diethylpyrocarbonate or sulfhydryl-modifying reagents (2-mercaptoethanol, N-ethylmaleimide) resulted in half-maximal inhibition of LeAPP2 activity at concentrations of 11, 43, and 79 M, respectively. Apparent K m and k cat values were derived from steady state kinetic analyses in the presence of 4 mM MnCl 2 at pH 7.5 where LeAPP2 exhibited highest activity (Fig. 4, B and C). A K m of 15.2 Ϯ 2.4 M was determined for H-Lys(Abz)-Pro-Pro-pNA, and the catalytic efficiency (K m /k cat ) was found to be 0.94 Ϯ 0.11 mM Ϫ1 ϫ s Ϫ1 .
Substrate Specificity of LeAPP2-For the analysis of LeAPP2 substrate specificity, bradykinin and substance P, two substrates of mammalian APPs, were employed. The proteolytic cleavage of bradykinin and substance P by LeAPP2 was followed by MALDI-TOF/MS. The N-terminal Arg-Pro-bond of bradykinin (Arg-Pro-Pro-Gly-Phe-Ser-Pro-Phe-Arg, M r ϭ 1060.2) was rapidly cleaved by LeAPP2. The subsequent hydrolysis of the Pro-Pro-bond at the newly formed N terminus was much slower (Fig. 5). Hydrolysis of the N-terminal Arg-Pro-bond of substance P (Arg-Pro-Lys-Pro-Gln-Gln-Phe-Phe-Gly-Leu-Met, M r ϭ 1348.6) was considerably slower than the cleavage of the respective bond in bradykinin (Fig. 5) indicating an extended site for substrate recognition in LeAPP2. Surprisingly, prolonged incubation with LeAPP2 resulted in the processing of a second, internal bond in substance P, i.e. the -Phe-Gly-bond (Fig. 5). Whereas MALDI-TOF/MS does not allow a precise quantification of reaction rates, a qualitative estimation of reaction rates is possible by comparing the time required for complete cleavage at a certain position. A preference of LeAPP2 for Arg-Pro-Pro-Ͼ Arg-Pro-Lys-Ͼ Pro-Pro-Gly-Ͼ -Phe-Gly-is evident (Fig. 5). Systemin, a peptide hormone-like signaling molecule from tomato plants (32), was not a substrate of LeAPP2. The observed processing of bradykinin and substance P as well as the hydrolysis of H-Lys(Abz)-Pro-Pro-pNA are clearly the result of aminopeptidase P, i.e. LeAPP2, activity. A protein preparation from E. coli expressing an inactive form of LeAPP2 lacking eleven amino acids from the N terminus was devoid of proteolytic activity indicating the absence of any contaminating E. coli protease from the LeAPP2 preparation (data not shown). DISCUSSION APP, like X-Pro dipeptidase (prolidase) and methionyl aminopeptidases of types I and II, belong to the M24 family in the clan MG of metalloproteases (36,37). Whereas the overall sequence similarity between these enzymes is rather low, their C-terminal catalytic domains share a common structural feature called the pita-bread-fold (38). The structures of E. coli methionyl aminopeptidase and APP have been solved and two metal ions were found to be "sandwiched" in the pita-bread domain. The metal ions are liganded by two Asp, one His, and two Glu residues, respectively, which are strictly conserved in this family of proteases (35,39,40). The requirement of these residues for the catalytic activity of porcine APP has been demonstrated by site-directed mutagenesis (34). We report here the cloning and characterization of a related enzyme from tomato called LeAPP2. This is the first characterization of an aminopeptidase P from any plant. LeAPP2 shares considerable sequence similarity with both E. coli and mammalian APPs in both the C-terminal pita-bread-as well as in the N-terminal domains. All the amino acid residues involved in metal binding (34,35) as well as two histidine residues implicated in proton shuttling between the solvent and the dinuclear metal center (35) are conserved in LeAPP2 (Fig. 1).
In addition to the structural similarity, LeAPP2 shares functional characteristics with known APPs. LeAPP2 expressed and purified from E. coli as a GST fusion protein exhibited APP activity, releasing the N-terminal amino acid from peptides with a penultimate proline residue. It was found to process typical substrates of mammalian APPs, i.e. bradykinin and substance P. The catalytic properties as well as structural similarity indicate a closer relationship with the cytosolic as compared with the membrane-bound forms of mammalian APPs. The pH optimum of 7.5 for LeAPP2 activity (Fig. 4B) is consistent with its localization in the cytoplasm. Similar to the cytosolic APP from rat brain (7), LeAPP2 clearly preferred Arg-Pro-Pro-(bradykinin) over Arg-Pro-Lys-(substance P), which indicates an extended binding site for recognition of the P' 2 residue of the substrate as it was reported for E. coli and mammalian APPs (10,15). Like the cytosolic APPs from E. coli, Rattus norvegicus, and Homo sapiens, but unlike the membrane-bound enzymes from R. norvegicus and Bos taurus, the tomato enzyme tolerates a Lys residue in the P' 2 position. LeAPP2 also hydrolyzed the N-terminal Pro-Pro-bond of processed bradykinin, albeit at a slower rate. Cleavage of the Pro-Pro-bond at the N terminus of oligopeptide substrates has also been reported for the cytosolic rat and E. coli APPs (7,41). The rat cytosolic APP functionally expressed in E. coli, however, was found to be unable to hydrolyze the N-terminal Pro-Pro-bond of a synthetic oligopeptide substrate (13). Endopeptidase activity, i.e. the cleavage of the -Phe-Gly-bond in substance P, appears to be a unique feature of LeAPP2. The fact that protein preparations from E. coli cultures carrying the empty expression vector, as well as cultures expressing a truncated, inactive LeAPP2, were devoid of any proteolytic activity, unequivocally shows that the observed endoproteolytic activity is a property of LeAPP2 and not that of a contaminating E. coli enzyme.
There are conflicting reports in the literature with respect to the metal requirement of APPs. Until recently, supported by the crystal structures of E. coli methionyl aminopeptidase and APP, which revealed the presence of dinuclear metal centers in both enzymes (35,39), two manganese (Mn(II)) or zink (Zn(II)) ions per subunit were considered necessary for maximum catalytic activity in cytosolic and membrane-bound APPs, respectively (Ref. 14, and references therein). For methionyl aminopeptidase, on the other hand, two equivalents of cobalt (Co(II)) were proposed to be required based on the reproducible observation of highest activity in vitro in the presence of Co(II). Both the nature and the amount of metal required in vivo have recently been questioned, however.
E. coli methionyl aminopeptidase was shown to be maximally activated upon addition of only one Fe(II) ion and iron is likely to be the in vivo ligand. Whereas the first Fe(II) ion is bound with high affinity (K d ϭ 0.3 M), the K d of the second metal binding site was reported to be 2.5 mM, and therefore, this site is likely to be unoccupied in vivo (42,43). Likewise, the two metal binding sites in human cytosolic APP (hcAPP) ap-pear to differ in affinity. Upon expression in E. coli, this enzyme was found to contain only one equivalent of Mn(II), and this was sufficient to support proteolytic activity (14). The hydrolysis of bradykinin and substance P by hcAPP was stimulated 2.7-fold upon further addition of Mn 2ϩ , whereas Mg 2ϩ , Ca 2ϩ , Cu 2ϩ , and Zn 2ϩ were found to be inhibitory (in order of increasing inhibition, Ref. 14). The effects of divalent metal ions on LeAPP2 activity are essentially the same as those observed for hcAPP (Fig. 4A) and, therefore, LeAPP2 is also likely to be a single Mn(II)-dependent enzyme.
The function of the second metal ion binding site remains obscure. Roles in the regulation of proteolytic activity or in positioning the substrate by binding its N-terminal amine group have been proposed (34,43). A competition of substrate and metal ion for the same binding site may explain the earlier observation that the inhibitory and stimulating effects of cations on APP activity can be substrate-dependent (44,45).
The enzymatic properties of LeAPP2 were further characterized using H-Lys(Abz)-Pro-Pro-pNA as the substrate for which an apparent K m of 15.2 Ϯ 2.4 M and a catalytic efficiency (K m /k cat ) of 0.94 Ϯ 0.11 mM Ϫ1 ϫ s Ϫ1 were derived from steadystate kinetic analyses (Fig. 4C). These values are within the range of catalytic constants reported for other APPs (2,7,9,10,14,15,19). Likewise, the inhibitor profile of LeAPP2 is typical for APPs. LeAPP2 was found to be inhibited by chelating agents with 1,10-phenanthroline being much more effective than EDTA. Consistent with the essential role of histidine residues in binding of the active site metal and in catalysis, LeAPP2 was inactivated by a histidine-modifying reagent (diethylpyrocarbonate). Inhibition by 2-mercaptoethanol and Nethylmaleimide may indicate a functionally important cysteine residue. There is, however, no cysteine residue conserved between the two tomato and human enzymes (Fig. 1). Alternatively, thiol reagents may compete with the substrate as a ligand of the active site metal. The inhibiton by both metal Aliquots were taken at the time points indicated on the right and analyzed by MALDI-TOF/MS. The masses observed for bradykinin fragments correspond to peptides lacking one or two amino acids from the N terminus (calculated mass of 904.0 and 806.9, respectively). Two fragments were generated from substance P (right panel). The larger fragment with a mass of 1191.9 was generated by cleavage of the N-terminal Arg residue (calculated mass of 1192.4). Subsequent endoproteolytic cleavage generated a fragment with a mass of 891.4 (calculated mass of 891.0) corresponding to hydrolysis of the -Phe-Gly-bond of substance P. chelators and thiol reagents has been reported widely for other APPs (7,8,10,14,19).
The function of LeAPP2 in planta remains obscure as long as the in vivo substrate(s) are elusive. They will include oligopeptides with an amino-terminal Xaa-Pro motif. Such peptides may arise during protein degradation implying a function for LeAPP2 in protein turnover. Physiological substrates may also include plant peptide hormones implying a function for LeAPP2 in the regulation of hormone stability/activity. Considering the role of mammalian APPs in the degradation of bradykinin and substance P, it is tempting to speculate on such a function for LeAPP2. However, only very few peptide hormonelike signal molecules are known in plants (22), and none of them contains an N-terminal Xaa-Pro motif. They are therefore not likely to be substrates of LeAPPs. Yet peptides are anticipated to play a much broader role in plant signal transduction than presently appreciated (24), and they may require APPs for the regulation of activity.