Regulated expression, processing, and secretion of dog mast cell dipeptidyl peptidase I.

Dipeptidyl peptidase I (DPPI) is a cysteine protease found predominantly in myelomonocytic cells, cytotoxic T-cells, and mast cells. Recent studies identify an intracellular role for mast cell-DPPI (MC-DPPI) by activating prochymase and protryptase to their mature forms. To better define MC-DPPI and to explore the possibility of extracellular roles, we purified MC-DPPI from mastocytoma cells. We found the dog C2 mastocytoma cell line to be the richest source yet described for DPPI, purifying up to 200 microg of enzyme per g of cells. Dog MC-DPPI has an Mr of approximately 175,000 and consists of four subunits, each composed of a propeptide, light chain, and heavy chain. The heavy chain is N-glycosylated and is heterogeneously processed to three different forms. NH2-terminal sequences of the heavy chain and propeptide are identical to those predicted from a cDNA clone we sequenced from a mastocytoma cDNA library. The dog cDNA-derived sequence is 86% identical to that of human DPPI. Dog mastocytoma cells incubated with 12-O-tetradecanoylphorbol-13-acetate increase expression of MC-DPPI mRNA. MC-DPPI maintains its activity for dipeptide substrates at a neutral to alkaline pH. Cells stimulated with ionophore or substance P secrete MC-DPPI in parallel with the granule-associated mediators tryptase and histamine. Thus, dog mastocytoma cells secrete DPPI that is active at the pH of extracellular fluids, suggesting that MC-DPPI may act outside the cell.

The C1 family of cysteine proteases includes the lysosomal proteases cathepsins B, C, H, S, and L. Initially, it was believed that these proteases function primarily or exclusively in intracellular protein degradation and turnover (1). This perspective has expanded as members of this group were found to play a role in antigen presentation (2,3) and protease activation (4 -8). In addition, cathepsins B, L, and S can be secreted into the extracellular compartment and may play a part in extracellular matrix degradation and tumor metastasis (9 -11). These observations suggest more diverse roles for this group of proteases.
Dipeptidyl peptidase I (DPPI), 1 also known as cathepsin C, is a unique member of the lysosomal group of cysteine proteases.
DPPI differs from the other members in structure and substrate preferences. By unknown mechanisms, human DPPI is processed into a mature, proteolytically active enzyme consisting of the M r 23,000 heavy chain (amino acids 231-394), the M r 7000 light chain (amino acids 395-463), and an M r 16,000 propeptide that remains associated with the active enzyme. In contrast to the endoproteolytic activity of cathepsins B, S, and L, DPPI has relatively promiscuous exopeptidase activity, hydrolyzing most NH 2 -terminal dipeptides except those with P1 proline or P2 basic residues (12)(13)(14)(15). By removing NH 2 -terminal dipeptides intracellularly, DPPI activates a number of granule-associated serine proteases, including lymphocyte granzymes, neutrophil elastase, and cathepsin G (5). In addition, DPPI may play a role in cell growth (16,17) and in the activation of platelet factor XIII (18) and neuraminidase (19).
Mast cells have granules that contain histamine, heparin, and serine proteases, such as tryptase and chymase (20). Tryptase and chymase are synthesized as inactive zymogens and then are activated by removal of NH 2 -terminal dipeptides by DPPI (7,8). These activated proteases are released by mast cells when stimulated to degranulate by various physiological stimuli (21). Mast cell granules also contain and secrete lysosomal hydrolases, such as ␤-hexosaminidase (22). Therefore, mast cell granules contain a diverse group of molecules, some of which reside in lysosomes in other types of cells, which can be secreted by exocytosis.
We have initiated studies designed to improve our understanding of the intracellular and possible extracellular roles of mast cell DPPI (MC-DPPI). This work reports the purification and biochemical properties of MC-DPPI. We show that dog mast cells produce large amounts of MC-DPPI, which is active over a broad pH range. Unlike other reported DPPIs, MC-DPPI exhibits heterogeneous processing of its heavy chain, and a large fraction of the enzyme can be released from intracellular storage sites. We also found that the levels of MC-DPPI mRNA are subject to modulation. These studies of MC-DPPI show possibly novel processing, regulated expression, and extracellular activity.

EXPERIMENTAL PROCEDURES
Materials-All chemicals are from Sigma unless otherwise specified. Cell Culture-Dog BR and C2 mastocytoma cells were a gift of S. Lazarus and were cultured in Dulbecco's modified Eagle's medium-H16 medium containing 2% supplemented calf serum as described previously (23). Cells were maintained at 37°C in 5% CO 2 and 95% air and were cultivated to a density of 1 ϫ 10 6 cells/ml. In experiments requiring stimulation with phorbol, cells were incubated with 25 ng/ml of 12-O-tetradecanoylphorbol-13-acetate for 18 h and then harvested by centrifugation.
Purification of DPPI-C2 cell pellets were resuspended in 100 mM sodium acetate, 300 mM NaCl, 1 mM EDTA (pH 6.0) (Buffer A), sonicated, and then centrifuged at 10,000 ϫ g for 15 min. The supernatant was adjusted to pH 4.2, heated at 37°C for 2 h, and then centrifuged at 14,000 ϫ g for 15 min. The resulting supernatant was dialyzed overnight in 20 mM bis [ Deglycosylation-Purified MC-DPPI was deglycosylated with N-glycosidase F (PNGase F; New England Biolabs, Beverly, MA). 5 g of DPPI were denatured in a 100-l solution of 0.5% SDS and 1% 2-mercaptoethanol at 100°C for 10 min. 2 units of PNGase F were added, and the mixture incubated in 1% nonylphenoxypolyethoxyethanol (Nonidet P-40) and 0.05 M Na 2 HPO 4 (pH 7.5) at 37°C for 2 h. Proteins were precipitated with quinine sulfate prior to electrophoresis. NH 2 -terminal Sequence Analysis-5 g of purified and deglycosylated MC-DPPI were subjected to SDS-PAGE, transferred to polyvinylidine difluoride membrane (Bio-Rad) in 10 mM CAPS buffer containing 10% methanol for 1 h at 4°C, and then stained with Coomassie Blue. Bands at M r 16,000, 19,000, 20,000, and 22,000 were cut from the membrane and subjected individually to Edman degradation using a 470A gas phase sequencer (Applied Biosystems, Foster City, CA) with an on-line 120A PTH analyzer (Applied Biosystems). Amino acid sequence analysis was performed by the Biomolecular Resource Center at the University of California, San Francisco.
cDNA Cloning--MC-DPPI cDNA was obtained by PCR amplification of cDNAs from a previously prepared BR mastocytoma library (24). The cDNA was amplified by PCR using Taq polymerase and the following primers, which were designed to include conserved regions of human and rat DPPIs (25,26): 5Ј-CGACACACCTGCCAACTGC-3Ј and 5Ј-CCTCTGCGGATCCGGAAG-3Ј. A second set of primers corresponding to nucleotides 1139 -1163 of dog MC-DPPI (5Ј-GCTGGTGGGCTAT-GGTACTGACTC) and a portion of gt10 (5Ј-GAGGTGGCTTATGAG-TATTTCTTCCAGG-3Ј) near the insertion site were used to amplify the 3Ј-end of MC-DPPI, which was subcloned into pCR-Script (Stratagene, La Jolla, CA) and sequenced in both directions. Amino acid sequence was deduced from the cDNA sequence and aligned to published sequences from other species using Gene Works software (Oxford Molecular Group, Campbell, CA).
pH Profile of DPPI Activity-The pH dependence of L-Ala-Ala-pnitroanilide hydrolysis was determined at intervals of 0.5 pH units from pH 3.0 to 8.0 in citrate-phosphate buffers as described previously (27).
Mast Cell Degranulation-Mast cell exocytosis was stimulated as described previously (21). Briefly, mastocytoma cells were harvested by centrifugation, washed twice in Ca 2ϩ -and Mg 2ϩ -free phosphate-buffered saline, and then resuspended in serum-free Dulbecco's modified Eagle's-H16 medium to a final concentration of 15 ϫ 10 6 cells/ml. Cells were then incubated alone, with 1 M ionophore A23187, or with 100 M substance P for up to 60 min. Aliquots were removed at time intervals and centrifuged immediately. Degranulation supernatants were separated from cell pellets, which were resuspended in the same volume as the supernatant and lysed on ice by sonication. Cellular debris were pelleted by centrifuging at 15,000 ϫ g. Recovered supernatants were assayed for tryptase and DPPI activity, and the net percentage of release was calculated as described previously (22). The In Vitro Toxicology Assay kit (Sigma) was used to measure LDH released in degranulation supernatants and lysed cell pellets as a measure of ionophoreinduced toxicity.
Inhibitor Profiles-1 g of purified MC-DPPI or 30 l of degranulation supernatant were preincubated for 5 min in 500 l of DPPI activation buffer. Inhibitor was then added at the following concentrations: bestatin, 500 g/ml; phenylmethylsulfonyl fluoride, 1 mM; aprotinin, 100 M; cystatin, 50 g/ml; Gly-Phe diazomethyl ketone (Enzyme Systems Products, Dublin, CA), 20 M; E-64, 10 M; or iodoacetic acid, 1 mM. After a 15-min preincubation with inhibitor, 400 l of substrate buffer with the corresponding inhibitor were added, and L-Ala-Ala-pnitroanilide hydrolytic activity measured as above. Activity with inhibitor was compared with that without inhibitor.
Antibody Production and Immunoblotting-100 g of purified dog MC-DPPI plus Freund's adjuvant were injected into two rabbits. At 2-week intervals, rabbits were injected with 100 g of dog MC-DPPI. At week 12, the rabbits were bled. Serum was collected and analyzed by ELISA for MC-DPPI immunoreactivity. Antibodies were produced by Immuno-Dynamics, (La Jolla, CA). For immunoblot analysis, samples were separated by SDS-PAGE under reducing conditions and then transferred to polyvinylidine difluoride membrane (Bio-Rad) in 10 mM CAPS containing 10% methanol for 1 h at 4°C. The membrane was washed with 50 mM Tris-HCl, 0.5 M NaCl, 0.01% Tween-20 (TBS) (pH 7.5), incubated with a 1:1000 dilution of MC-DPPI antiserum in TBS for 1 h, washed with TBS, incubated with a 1:2000 dilution of goat antirabbit antiserum (Sigma) in TBS for 1 h, and washed again. Immunoreactivity was detected using the Fast Red TR/naphthol AS-MX detection system (Sigma).
RNA Blotting-Poly(A)ϩ RNA was isolated from mastocytoma cells, incubated alone or in the presence of phorbol ester, c-kit ligand (a gift from Amgen, Thousand Oaks, CA), tumor necrosis factor-␣ (R & D Systems, Minneapolis, MN), interleukin-1␤ (R & D Systems), using the Micro-fast track kit (Invitrogen, Carlsbad, CA). Denatured poly(A)ϩ RNA was size-fractionated by agarose gel electrophoresis and transferred to Nytran Plus nylon membrane (Schleicher and Schuell) (28). Vacuum-baked membranes were prehybridized at 42°C for 2 h and hybridized with an [␣-32 P]dCTP (Amersham Pharmacia Biotech) random prime-labeled dog MC-DPPI probe at 42°C overnight. After two washes at room temperature and two at 37°C, the membrane was exposed to film for 19 -48 h and then developed. Following removal of previously bound probe, the membrane was hybridized with a ␥-actinlabeled probe to account for possible variations in signal intensity due to differing amounts of mRNA loaded per lane.

RESULTS
Purification of MC-DPPI-DPPI enzymatic activity was readily measured in dog BR and C2 mastocytoma cell sonicates using Gly-L-Phe-p-nitroanilide and L-Ala-Ala-p-nitroanilide. L-Ala-Ala-p-nitroanilide was used in all subsequent experiments because the specific activity was higher for this substrate. Hydrolysis of L-Ala-Ala-p-nitroanilide was 7-fold greater in C2 than in BR cell extracts; therefore, C2 cells were used as the source for MC-DPPI. Hydrolysis of L-Ala-Ala-p-nitroanilide was completely inhibited by Gly-L-Phe-diazomethylketone, a specific DPPI inhibitor (5), indicating that the activity is entirely due to MC-DPPI. Repeated iterations of the purification protocol yielded 90 -200 g of pure MC-DPPI per g of cells. The 6-fold enrichment provided by lentil lectin chromatography was the single most effective step of the purification protocol (Table I). The Mono-Q chromatographic step, although yielding only a small boost in specific activity, was used because it separated MC-DPPI from a few proteins of high M r . Using a specific activity determined by hydrolyzing L-Ala-Ala-p-nitroanilide with a known quantity of purified MC-DPPI and back-calculating, we estimate that C2 cells contain 1-2 pg of MC-DPPI/cell. SDS-PAGE migrates as a single band with an M r of ϳ175,000 in the absence of reducing agents (Fig. 1A). This is approximately the size one expects for a tetrameric assembly of enzymatic units, with each unit consisting of a heavy chain, a light chain, and a portion of propeptide. The faint band of M r 45,000 may be a small amount of single enzymatic units. Under reducing conditions (Fig. 1B), MC-DPPI migrates as three bands: a broad band with M r of 22,000 -25,000, a closely-spaced doublet of M r ϳ16,000, and a band that migrates with the dye front. After N-deglycosylation, the broad, high M r band resolves into three bands with estimated M r values of 19,000, 20,000, and 22,000 (Fig. 1B), suggesting that the heavy chain is heterogeneously processed to three different major species. The M r 16,000 double band resolves into a single band of greater intensity after treatment with PNGaseF. Protein Sequencing-Edman degradation of the three highest molecular weight bands was performed to determine whether variable NH 2 -terminal cleavage explains the heterogeneity observed on SDS-PAGE. Major and minor sequences were obtained for bands with the following M r values: 19,000, LPTSW (major) and RLPTS (minor); 20,000, ISRLP (major) and RLPTS, EE, and LPTSW (minor); 21,000, ISRLP (major) and RLPTS (minor). Alignment of these sequences to those of human DPPI (25,26) and of the cDNA-derived dog MC-DPPI sequence (see below) confirms that they correspond to sequences in the NH 2 -terminal region of the heavy chain, with the exception that there are multiple NH 2 termini extended by 1, 3, and 5 amino acids compared with the human enzyme. NH 2 -terminal sequencing of the M r 16,000 band reveals two sequences, the predominant one, DTPADXTH, corresponding to the pro-region of human DPPI. The minor sequence, DPFNPFEL, represents a small amount of light chain. The propeptide M r of 16,000, as determined by SDS-PAGE, is less than the M r of 23,000 predicted by nucleic acid sequence (see below) and suggests that the propeptide is truncated at its COOH terminus by proteolysis.
MC-DPPI cDNA-To achieve a better understanding of the nucleotide and predicted amino acid sequence of dog MC-DPPI, we cloned a DPPI cDNA fragment from a dog mastocytoma cDNA library using reverse transcription-PCR (Fig. 2). The cloned 1539-base pair cDNA encodes the propeptide, heavy chain, and light chain. A 224-base pair 3Ј-untranslated region ends in a poly(A) tail. A consensus polyadenylation signal (AATAAA) resides 19 base pairs upstream of the poly(A) tail. NH 2 -terminal amino acid sequences obtained by Edman degradation agree with those predicted by the cDNA, suggesting that the cDNA in fact corresponds to the purified enzyme. The nucleotide sequence and predicted amino acid sequence are 86 and 84% identical, respectively, to the corresponding sequences of human DPPI (26). There are four consensus sites for Nglycosylation in the propeptide and one in the heavy chain. All cysteines are conserved among the dog, human, mouse, and rat species (25,26,29), as are the putative catalytic amino acid residues Cys 234 and His 381 (Cys 25 and His 159 by papain numbering (30)). Two other amino acids believed to be important for catalysis by cysteine proteases, Gln 228 and Trp 399 (31), (Gln 19 and Trp 177 by papain numbering), are also conserved. There is strict conservation of 60% of the propeptide (residues 1-203) and 80% of both heavy chain (residues 204 -370) and light chain (residues 371-413) among the four species of DPPI analyzed (Fig. 3).
pH Profile of MC-DPPI Activity-The pH dependence of MC-DPPI activity for hydrolysis of L-Ala-Ala-p-nitroanilide (Fig. 4) reveals that MC-DPPI has activity over a broad pH range and that its optimum activity is at pH 6.5. MC-DPPI retains Ͼ80% of its activity at pH 7.5, a pH similar to that in the extracellular compartment.
Secretion of MC-DPPI-Incubation of C2 cells with 1 M ionophore stimulates time-dependent release of MC-DPPI and tryptase activities (Fig. 5A). After a 15-min incubation with ionophore, 35% of total active MC-DPPI, 20% of total active tryptase, and 35% of total histamine (data not shown) is released by mastocytoma cells. Similarly, a significant amount of DPPI and tryptase is released by C2 cells after a 60-min incubation with 100 M substance P (Fig. 5B). Control, ionophore, and substance P-stimulated cells are Ͼ90% viable after 15 min as determined by trypan blue exclusion and LDH release, indicating that enzyme release is not due to cell death and lysis.
Inhibition of MC-DPPI-Inhibitor profiles for hydrolysis of L-Ala-Ala-p-nitroanilide by purified MC-DPPI and degranulation supernatants were determined to confirm that activity measured in degranulation supernatants was from MC-DPPI and not from another protease. The relative percentage of inhibition by specific inhibitors is similar for purified MC-DPPI and degranulation supernatants (Table II), confirming that activity measured in the degranulation supernatants is entirely due to MC-DPPI.
Immunoblotting-Presence of MC-DPPI immunoreactivity in the supernatants from control (unstimulated) and degranulated mast cells is detected using a polyclonal antibody generated against MC-DPPI (Fig. 6). More immunoreactive material is present in supernatants generated by ionophore-stimulated cells than in those generated by unstimulated cells. This material co-migrates with purified MC-DPPI, confirming its identity and its secretion under the influence of ionophore. The visualized band corresponds to the heavy chain of MC-DPPI, suggesting that epitopes recognized by the antibody are exclusively on the heavy chain.
DPPI mRNA Regulation-Due to the greater amounts of MC-DPPI protein in C2 cells compared with BR cells, we sought to determine whether this is, in part, due to a difference in mRNA levels. Fig. 7 shows that the basal level of MC-DPPI mRNA is greater in C2 cells than BR cells. Phorbol increases by 6-fold the steady state MC-DPPI mRNA levels in BR cells. C2 cells, however, have a high baseline level of MC-DPPI that is not increased by phorbol stimulation. These results indicate that DPPI mRNA levels can be regulated and that the high levels of mRNA in C2 cells may explain the high-level expression of active enzyme. Levels of MC-DPPI mRNA in BR cells are not changed by incubation of BR cells with 100 ng/ml of c-kit ligand, 5 ng/ml of tumor necrosis factor-␣, or 10 Ϫ10 M interleukin-1␤. DISCUSSION This report describes the purification and characterization of DPPI from dog mastocytoma cells, which are the richest source of the enzyme yet described. Dog MC-DPPI is similar to human DPPI in oligomeric structure, activity at acidic pH, inhibition by inactivators of cysteine proteinases, and inducibility of mRNA expression (15,27,32). Possibly unique features are variable processing of its heavy chain, preservation of activity at a neutral to alkaline pH, and capacity for exocytic secretion. These features raise the possibility that MC-DPPI levels vary when mast cells are exposed to different external stimuli and that MC-DPPI enacts extracellular roles in addition to its previously described intracellular actions.
The strategy we used to purify MC-DPPI is a modification of the techniques used to purify DPPI from other sources (15,27). The purification protocol takes advantage of the stability of MC-DPPI at acidic pH, N-glycosylation (resulting in binding to lentil lectin), and large size relative to other mast cell proteins. Using our purification scheme, we purified up to 200 g of MC-DPPI per g of mastocytoma cells, as compared with 2 (27) and 1 (15) g/g from human kidney and spleen, respectively. These results suggest the possibility that mast cells, because of their wide distribution, are the major source of DPPI in many tissues. From the specific activity determined for purified MC-DPPI, we estimate that C2 mastocytoma cells contain 1-2 pg of MC-DPPI per cell, an amount similar to that of chymase (23), another major mast cell protease.
As reported for human kidney DPPI (27) 2. Nucleotide and deduced amino acid sequence of dog proDPPI. The nucleotide sequence begins at the start of the propeptide. The propeptide, heavy chain, and light chain correspond to nucleotides 1-594, 595-1098, and 1099 -1305, respectively. The polyadenylation signal is identified by a solid bar. Boldface arrows indicate major NH 2 termini, and smaller arrows indicate minor NH 2 termini identified by protein sequencing. Underlined amino acids are those confirmed by NH 2 -terminal sequencing of the purified enzyme. Additional amino acids from the alternate NH 2 termini of the heavy chain are identified by an asterisk (*). and NH 2 -terminal sequences correspond exactly to those reported for human DPPI (27). Resolution of the propeptide doublet into a single band after N-deglycosylation suggests that it is N-glycosylated. NH 2 -terminal sequencing of deglycosylated material reveals an Asp instead of the predicted Asn at the fifth residue of the propeptide. This is consistent with glycosylation of the first predicted glycosylation site because an Asp is generated by N-deglycosylation. The blank in the sixth cycle of the propeptide sequence results from oxidation of cysteine.
There are several possible explanations for the variable NH 2 termini identified of the MC-DPPI heavy chain. The ISRLP . . . NH 2 terminus may result from processing by a granzyme Blike protease, and the LPTSW . . . NH 2 terminus may result from cleavage by a tryptic-like protease (possibly tryptase). Alternatively, there may be only one major site of endoproteolytic cleavage of the proenzyme, with further exoproteolytic processing of the NH 2 terminus by aminopeptidases or dipeptidyl peptidases, possibly MC-DPPI itself. An initial major endoproteolytic cleavage may generate the EEISRLPT . . . heavy chain isoform, which is then autoprocessed by MC-DPPI itself, yielding the ISRLPT . . . and RLPTS . . . NH 2 termini, or processed by tryptase, yielding the LPTSW . . . NH 2 -terminal isoform identified by Edman degradation. Processive hydrolysis by mast cell monoaminopeptidase cleavage (33) of the EEIS-RLP . . . NH 2 terminus could explain some or all of the subsequent NH 2 termini identified; however, one might have expected to detect EISRL . . . and SRLPT . . . NH 2 termini if aminopeptidase processing had occurred exclusively. The functional significance of variable heavy chain processing is unknown. It is possible that differences in processing affect substrate specificity, pH optimum, and specific activity. Identification of the proteases involved in MC-DPPI processing and the sites of such processing (endoplasmic reticulum, Golgi apparatus, secretory granules, or cell extracts) require future study.
As seen in Fig. 3, much of dog MC-DPPI is conserved in DPPIs of other species. The most conserved region of DPPI is the heavy chain, suggesting that many heavy chain residues, and the heavy chain itself, are critical for enzyme activity and specificity. Conservation of the catalytic amino acids Cys 234 and His 381 , as well as two amino acids, Gln 228 and Trp 399 , which are believed to be important for stabilizing the catalytic dyad and its intermediates (31), suggests that the deduced amino acid sequence is that of catalytically active MC-DPPI. Specific residues critical for determining exopeptidase activity of MC-DPPI have yet to be determined.
The function of the propeptide of MC-DPPI specifically and of DPPIs in general is unknown. The propeptides of cathepsin B and cathepsin L, two of the nearest relatives of DPPI, inhibit enzymatic activity by binding to the substrate binding cleft (34,35). Following a proteolytic event, the propeptide is released, and the enzyme is thereby activated. This scenario probably differs for MC-DPPI because the propeptide is 2-3 times longer than that of cathepsins B and L, and a portion of its propeptide remains associated with the active enzyme. The propeptide of MC-DPPI may initially function similarly to that of cathepsins B and L by occupying the MC-DPPI active site and inhibiting catalytic activity. A COOH-terminal portion of the propeptide may then be removed by proteolysis, leaving a smaller fragment of the propeptide associated with the active enzyme. Alternatively, the propeptide may confer a specific conformation to MC-DPPI, which then changes after removal of the propeptide fragment. The remaining portion of the propeptide is probably disulfide-linked to the mature enzyme, as suggested by the lack of a propeptide fragment on nonreducing SDS-PAGE. Although the function of the portion of propeptide that remains linked with the active enzyme is unknown, it may help to maintain structural   stability or influence the catalytic attributes of the protease. Alternatively, it may be a functionless remnant, as suggested for the portion of the chymotrypsin propeptide that remains disulfide-linked to the mature active enzyme (36).
Finding that MC-DPPI has activity over a broad pH range from 6.0 to 7.5 is important for several reasons. MC-DPPI activates the mast cell proteases tryptase and chymase by removing an NH 2 -terminal dipeptide (7,8). This activation likely occurs either in the Golgi apparatus or in secretory granules, which are acidic organelles (pH 5.5-6.5) (37). Therefore, to remove the dipeptide, MC-DPPI must be active at an acidic pH. Proteolytic activity at a neutral to alkaline pH is significant because we showed that MC-DPPI can be released by degranulating mast cells. Activity at neutral to alkaline pH raises the possibility of extracellular roles following release.
In this report, we showed that MC-DPPI is secreted by cultured mast cells in parallel with tryptase and histamine. This observation indicates that MC-DPPI is functionally, and possibly physically, in the same compartment as the classic mast cell secretory products tryptase and histamine. Finding DPPI, a lysosomal protease, in mast cell secretory granules is not surprising because other lysosomal enzymes, such as ␤-hexosaminidase, also reside in these granules (22). The observation that DPPI and tryptase activity coexist in a cytoplasmic granular fraction in extracts of mouse mastocytoma cells further supports the possible colocalization of DPPI and tryptase (5).
Secretion of MC-DPPI by cells stimulated with substance P is significant because it identifies a physiological stimulus whereby MC-DPPI may be released extracellularly. Mast cells have been found in close association with nerve terminals of neurons from the nonadrenergic, noncholinergic nervous system, including those that store substance P (38,39). Depolarization of these neurons may result in release of substance P, which may then act on the nearby mast cell, stimulating secretion of MC-DPPI.
The role of extracellular MC-DPPI is unknown at this time. Individually or in cooperation with other proteases, MC-DPPI may degrade targets such as neuropeptides, cytokines, or extracellular matrix proteins. In doing so, MC-DPPI may influence cellular functions or matrix remodeling. These effects would probably be limited to the local region around the mast cell because circulating protease inhibitors, such as the cystatins (40), should inhibit MC-DPPI as it diffuses away from the cell. However, high concentrations of MC-DPPI in the immediate vicinity of the degranulating mast cell may overwhelm inhibitor defenses temporarily, allowing MC-DPPI to encounter and hydrolyze extracellular substrates.
The polyclonal antibody raised against dog MC-DPPI recognizes predominantly the heavy chain of MC-DPPI, and not the propeptide or light chain, although each of these components is present in the purified material used to immunize rabbits. A possible explanation for this observation is that dog MC-DPPI propeptide and light chain are more homologous to the corresponding fragments of rabbit DPPI than is the heavy chain.
The steady-state quantity of MC-DPPI mRNA and corresponding protein levels rise in cells incubated with phorbol ester. This suggests that MC-DPPI protein expression may be controlled by varying the amount mRNA via transcriptional or posttranscriptional mechanisms. Regulation of MC-DPPI has several implications. First, quantities of enzyme may vary depending on the tissue location of the mast cell. An example of this is the variable expression of tryptase and chymase depending on whether the mast cell is of skin or mucosal origin (20). Second, MC-DPPI may be the rate-limiting step in production of active tryptase and chymase because the amount of active tryptase or chymase will depend on availability of active MC-DPPI. Third, MC-DPPI may be regulated by inflammatory mediators. Mast cells have a role in allergic and inflammatory diseases, such as asthma (41). Inflammatory mediators in these diseases may alter quantities of MC-DPPI and influence its roles in these diseases. The finding that c-kit ligand, tumor necrosis factor-␣, and interleukin-1␤ do not alter MC-DPPI mRNA levels suggests that MC-DPPI may not be subject to regulation by these cytokines.
In conclusion, dog mastocytoma cells produce a DPPI that has structural and enzymatic similarities to and differences from DPPIs purified from other tissues and species. Unique features of MC-DPPI are the heterogeneous processing of its heavy chain, enzymatic activity over a broad pH range, and secretion by mast cells. These findings invoke the possibility that MC-DPPI acts on extracellular targets.