Contradictory Functions (Activation/Termination) of Neutrophil Proteinase 3 Enzyme (PR3) in Interleukin-33 Biological Activity*

Background: The maturation process of IL-33 (IL-1F11) remains unclear. Results: IL-33 ligand affinity column isolates neutrophil proteinase 3. Conclusion: PR3 is an IL-33-processing enzyme. Significance: PR3 has a dual function in IL-33 biological activity. IL-1 family ligand does not possess a typical hydrophobic signal peptide and needs a processing enzyme for maturation. The maturation process of IL-33 (IL-1F11), a new member of the IL-1 family ligand, remains unclear. Precursor IL-33 ligand affinity column isolates neutrophil proteinase 3 (PR3) from human urinary proteins. PR3 is a known IL-1 family ligand-processing enzyme for IL-1β (IL-1F2) and IL-18 (IL-1F4), including other inflammatory cytokines. We investigated PR3 in the maturation process of precursor IL-33 because we isolated urinary PR3 by using the precursor IL-33 ligand affinity column. PR3 converted inactive human and mouse precursor IL-33 proteins to biological active forms; however, the increase of PR3 incubation time abrogated IL-33 activities. Unlike caspase-1-cleaved precursor IL-18, PR3 cut precursor IL-33 and IL-18 at various sites and yielded multibands. The increased incubation period of PR3 abated mature IL-33 in a time-dependent manner. The result is consistent with the decreased bioactivity of IL-33 along with the increased PR3 incubation time. Six different human and mouse recombinant IL-33 proteins were expressed by the predicted consensus amino acid sequence of PR3 cleavage sites and tested for bioactivities. The human IL-33/p1 was highly active, but human IL-33/p2 and p3 proteins were inactive. Our results suggest the dual functions (activation/termination) of PR3 in IL-33 biological activity.


IL-1 family ligand does not possess a typical hydrophobic signal peptide and needs a processing enzyme for maturation. The maturation process of IL-33 (IL-1F11), a new member of the IL-1 family ligand, remains unclear. Precursor IL-33 ligand affinity column isolates neutrophil proteinase 3 (PR3) from human urinary proteins. PR3 is a known IL-1 family ligand-processing enzyme for IL-1␤ (IL-1F2) and IL-18 (IL-1F4), including other inflammatory cytokines.
We investigated PR3 in the maturation process of precursor IL-33 because we isolated urinary PR3 by using the precursor IL-33 ligand affinity column. PR3 converted inactive human and mouse precursor IL-33 proteins to biological active forms; however, the increase of PR3 incubation time abrogated IL-33 activities. Unlike caspase-1-cleaved precursor IL-18, PR3 cut precursor IL-33 and IL-18 at various sites and yielded multibands. The increased incubation period of PR3 abated mature IL-33 in a time-dependent manner. The result is consistent with the decreased bioactivity of IL-33 along with the increased PR3 incubation time. Six different human and mouse recombinant IL-33 proteins were expressed by the predicted consensus amino acid sequence of PR3 cleavage sites and tested for bioactivities. The human IL-33/p1 was highly active, but human IL-33/p2 and p3 proteins were inactive. Our results suggest the dual functions (activation/termination) of PR3 in IL-33 biological activity.
IL-33, similarly to IL-1␤ and IL-18, is produced as a precursor IL-33 molecule. This precursor IL-33 does not have a signal peptide to be secreted; it is released extracellularly as a mature protein after cleavage (6). It is known that inflammatory caspases are necessary for the cleavage of IL-1␤ and IL-18 (32,33). Although it has been suggested that precursor IL-33 is processed by caspase-1 (6), the exact role of caspases in IL-33 biology still remains unclear (34,35).
In this study, we identified and characterized a processing enzyme of precursor IL-33. PR3 converts inactive precursor IL-33 to active form, whereas a longer incubation period of PR3 abolishes IL-33 activity. This result suggests the contrasting functions (activation/termination) of PR3 in precursor IL-33 bioactivity. Cloning for Escherichia coli Expression Vectors-Total RNA was isolated with TRI Reagent  (Sigma) from human PC-3 cells and mouse lung tissue. A pair of  human IL-33 sense primers 5Ј-ACAGAATACTGAAAAATG-AAGCC-3Ј and reverse primer 5Ј-CTTCTCCAGTGGTAGC-ATTTG-3Ј, mouse IL-33 sense primers 5Ј-ATGAGACCTAG-AATGAAGTATTC-3Ј and reverse primer 5Ј-GCACAGGCG-TTTTACTGCATT-3Ј, and ␤-actin sense primers 5Ј-ACCAA-CTGGGACGACATGGA-3Ј and reverse primer 5Ј-GTGATG-ACCTGGCCGTCAGG-3Ј was used for the RT-PCR. Moloney murine leukemia virus-RT (Beams Bio, Korea) was used for converting 2 g of total RNA to first strand cDNA, and then PCR was performed at 94°C for 45 s, 70°C for 2 min, and 59°C for 1 min for 30 cycles.

RT-PCR and Molecular
For E. coli expression vector, the PCR product of human and mouse precursor IL-33 cDNA was ligated into T&A cloning vector (RBC, Taiwan). We designed the sense primer with an EcoRI (GAATTC) site and the reverse primer with a KpnI (GGTACC) site to transfer the amplified PCR product into pProEX/HTa (Invitrogen).
For mouse precursor IL-33 and IL-33/p1, the same reverse primer (5Ј-TATTGGTACCTTAGATTTTCGAGAGCTT-3Ј) and the sense primer of mouse precursor IL-33 (5Ј-TAAAG-AATTCATGAGACCTAGAATGAAG-3Ј) and IL-33/p1 (5Ј-TAAAGAATTCTCACTTTTAACACAGTCT-3Ј) were used to amplify cDNA inserts. The PCR product of a single band from each construct was digested with EcoRI and KpnI and then transferred into pProEX/HTa (Invitrogen) for recombinant protein expression in E. coli.
Isolation of Urinary IL-33 Interacting Molecule from Concentrated Human Urine-Human precursor IL-33 (3 mg) was immobilized by coupling to Affi-Gel-15 beads according to the manufacturer's instructions (Bio-Rad). Batches of 300 ml of crude urinary proteins concentrated 500-fold were passed over the precursor IL-33-bound beads at 4°C. The column was washed with 500 ml of phosphate buffer containing 0.5 M sodium chloride, pH 8. Bound proteins were eluted in a 1-ml volume of solution containing 25 mM citric acid, pH 2.2, and then the eluted fraction was immediately neutralized with 2 M glycine.
Generation of Recombinant Protein-Six recombinant (human and mouse precursor IL-33, human and mouse IL-33/p1, and human IL-33/p2, IL-33/p3) proteins were expressed in E. coli Rosetta cells (Novagen, Madison, WI) and purified by a TALON affinity column (Invitrogen). The TALON affinity-purified recombinant proteins were subjected to high perform-ance liquid chromatography (GE Healthcare) with C4 column (Grace Vydac, Hesperia, CA). The two-step purified recombinant proteins were tested for endotoxin level (below 0.25 EU per g of recombinant protein) by using LAL method (Cape Cod, East Falmouth, MA) according to the manufacturer's instructions and then used for experiments.
For PR3 cleavage test with different IL-1 family ligands, commercial PR3 (100 ng from Athens Research & Technology) and the recombinant proteins of precursor human IL-33, mouse IL-33, and human IL-18 (500 ng) were incubated in 20 l of a reaction buffer (50 mM Tris, pH 8) at room temperature. The PR3-cleaved samples were subjected to 10% SDS-PAGE for silver staining.
Western Blots-For the detection of PR3-cleaved human rIL 3 -33, the samples were loaded on 10% SDS-PAGE. Anti-human IL-33 polyclonal antibody was developed by immunizing three BALB/c mice (6 weeks old) with mature human IL-33 (Ser 112 -Thr 270 ) protein (data not shown). Human IL-33 polyclonal antibody was used for detecting a mature size of IL-33, which was cleaved by PR3. In addition, an anti-human PR3 monoclonal antibody (36) was obtained from YbdYbiotech (Seoul, Korea) and used for detecting urinary PR3. Peroxidaseconjugated goat anti-mouse IgG secondary antibody (Jackson ImmunoResearch) was used to develop the blots by using Supex (Neuronex, Korea) and LAS-4000 imaging device (Fujifilm, Japan).
For detecting the phosphorylation of signaling molecules (IRAK1, NF-B, p38 MAPK, p44/42 MAPK, and JNK), HMC-1 and Raw 264.7 cells were stimulated with precursor and mature IL-33 at various time points. Cells were lysed with kinase lysis buffer (Cell Signaling Technology, Beverly, MA) and then subjected to 10% SDS-PAGE. The samples were transferred to nitrocellulose membranes. The membranes were blocked in 3% BSA/TBST (Santa Cruz Biotechnology, Santa Cruz, CA). The membranes were probed first with rabbit anti-phospho-IRAK1, mouse anti-phospho-NF-B, rabbit anti-phospho-p38 MAPK, mouse anti-phospho-p44/42 MAPK, or rabbit anti-phospho-JNK (Cell Signaling Technology). The membranes were reprobed with rabbit anti-NF-B, p38 MAPK (Santa Cruz Biotechnology), p44/42 MAPK, JNK, or IRAK1 (Cell Signaling Technology) for normalization of each protein and then normalized finally with goat anti-actin (Santa Cruz Biotechnology).
Cell Culture and Bioassay-Mouse macrophage Raw 264.7 cell line was obtained from American Type Culture Collection (ATCC) and maintained according to the instructions. HMC-1 cells were cultured in Iscove's modified Dulbecco's medium enriched with 10% FBS. For a blockade of ST2, a fresh medium (0.2 ml) containing various concentrations of rIL-33 and 0.5 g of anti-ST2 (R&D Systems) was used. Human HMC-1 and mouse Raw 264.7 cells were treated with different stimuli, and the cell culture supernatant was harvested the next day for cytokine assay.
Measurement of Cytokine Level-Human IL-8, mouse TNF␣, mouse MIP-2, and mouse IL-6 ELISA kits were obtained from R&D Systems. Cytokine levels were measured in culture supernatants by sandwich ELISA according to manufacturer's instructions.
Statistical Analysis-The data are expressed as means Ϯ S.E. Statistical significance of differences were analyzed by unpaired, two-tailed Student's t test. Values of p Ͻ 0.05 were considered statistically significant.

Isolation of an IL-33 Interacting Molecule from Urinary
Proteins-In an attempt to isolate a soluble IL-33 interacting molecule, concentrated urine was applied to a column composed of precursor IL-33 covalently immobilized to agarose beads. After extensive washing with PBS, interacting proteins were eluted at low pH citric acid solution. Aliquots of the various fractions were resolved by 10% SDS-PAGE, and the protein bands were visualized by silver staining. The broad bands corresponding to specific IL-33 interacting proteins, ϳ56and 28-kDa bands including the minor 36-kDa band, were detected mainly in fractions 2 and 3 (Fig.  1A). The 56-and 28-kDa protein bands in elution fraction 3 were excised and analyzed by mass spectrometer analysis. The result of mass spectrometer revealed that the bands were neutrophil proteinase 3 (PR3). We verified urinary PR3 in the eluted fractions with a monoclonal antibody specific to human PR3. The anti-human PR3 monoclonal antibody recognized the 56-, 36-, and 28-kDa urinary PR3 and commercial PR3 (Fig. 1B). The results suggest that 28 and 56 kDa are the monomer and dimer forms of mature PR3, respectively. The additional minor 36-kDa band could be the precursor PR3 because the anti-PR3 monoclonal antibody recognized the precursor PR3 in human monocyte cells (THP-1 and U937), and the molecular size of precursor PR3 is similar to the 36-kDa band (data not shown).
PR3 Converted Precursor IL-33 into Mature IL-33-The biological activity of human and mouse precursor IL-33 was examined in the presence of PR3. The pretreatment of PR3 activated precursor IL-33 and induced IL-8 production by a dose-dependent manner, but precursor IL-33 alone did not induce IL-8 production ( Fig. 2A). The time course of PR3 pretreatment revealed that PR3 induced the highest activity of IL-33 at 5 min, and its activity was decreased along with increasing incubation times (Fig. 2B). PR3 pretreatment for 30 min completely abolished IL-33 activity.
Examination of IL-33 Cleavage by PR3-We tested the cleavage of human and mouse precursor IL-33 proteins with caspase-1 and PR3. Caspase-1 specifically cleaved precursor IL-18 resulting in the production of 18-kDa mature IL-18, which was active in the A549R␤ cell line (data not shown) (37), but precursor IL-33 proteins were not cleaved by caspase-1 (Fig. 3A). Next we examined the cleavage of precursor IL-33 proteins with PR3. As shown in Fig. 3B, PR3 cleaved the precursor IL-33 and produced multibands unlike caspase-1cleaved IL-18 (Fig. 3A). PR3 also cleaved precursor IL-18 as multiband products in silver staining (Fig. 3B, 2nd lane from  right).  Time-dependent Reduction of a Mature Size IL-33 by PR3-IL-33 protein was incubated at different time points to examine whether the multiband products of the PR3 cleaved is a mature IL-33. We used the anti-IL-33 antibody to detect mature IL-33. The result of the Western blot exhibited that PR3-cleaved mature IL-33 was highest at 5 min and decreased gradually along with incubation time (Fig. 4). This result is consistent with bioactivity of the PR3-incubated precursor IL-33 in Fig.  2B. Mature rIL-33 (Ser 112 -Thr 270 ) was completely degraded by PR3 after 30 min of incubation (Fig. 4, far right lane).
Design and Expression of PR3-cleaved rIL-33 Proteins-We aligned the amino acid sequence of human and mouse IL-33 to predict PR3 cleavage sites because we failed to obtain a specific PR3 cleavage site by mass spectrometer analysis (data not shown) due to multiband productions in silver staining.
We predicted multiple PR3 cleavage sites by a previous report using a highly specific enzymatic activity testing tri-peptide library against proteinase 3 (38). A single PR3 cleavage site at the N terminus and two cleavage sites at the C terminus were predicted by the consensus sequence of the PR3 cleavage site. The predicted PR3 cleavage sites are indicated in red as p1, p2, and p3 (Fig. 5A). The previously reported caspase cleavage sites are shown in blue (Fig. 5A). We expressed six rIL-33 proteins as shown at the top of Fig. 5B. The purified rIL-33 proteins were subjected to 10% SDS-PAGE, and the result showed a single band of correct molecular size of each recombinant protein (Fig. 5B). The upper bands of precursor IL-33 and IL-33/p2-3 are dimer forms that were verified by Western blot (data not shown). Additional experiments tested the biological activity of mature human IL-33 (Ser 112 -Thr 270 ) and mouse IL-33 (Ser 109 -Ile 266 ) in the presence or absence of the N-terminal His tag. The N-terminal His tag did not affect the bioactivities of mature human and mouse rIL-33 (data not shown).

Interleukin-33 Processing Enzyme
cells, but no additional cytokines were produced in human HMC-1 cells (data not shown).
Anti-ST2 Specifically Inhibits the Cytokine Productions by IL-33/p1-We examined whether human IL-33 (Ser 117 -Thr 270 ) and mouse IL-33/p1 (Ser 114 -Ile 266 ) induce chemokines and inflammatory cytokines via ST2 on the cell surface of HMC-1 and Raw 264.7 cells. These cells were pretreated with anti-ST2 neutralizing antibody along with a control antibody, and the human and mouse IL-33/p1 proteins were then used to stimulate HMC-1 and Raw 264.7 cells. The anti-ST2 antibody-pretreated HMC-1 cells produced less IL-8 compared with nontreated cells (Fig. 8A), but the control antibody had no effect (data not shown). In mouse Raw 264.7 cells, TNF␣, MIP-2, and IL-6 productions were specif-ically suppressed by the anti-ST2 antibody (Fig. 8B), and the results were statistically significant in both human and mouse cells assays (Fig. 8, A and B).
Time-dependent Activation of Signaling Molecules by IL-33/ p1-Signal transduction of IL-33/p1-activated molecules was also studied. The phosphorylation of IRAK1, NF-B, p38 MAPK, ERK, and JNK was examined with mouse IL-33/p1treated Raw 264.7 cells because mature IL-33 proteins (Ser 109 -Ile 266 from R&D Systems) are known to activate these inflammatory signal pathways. The recombinant precursor IL-33 phosphorylated only ERK1/2 at 60 min and remained for 2 h, but the rest of signaling molecules were not phosphorylated (Fig. 9A). Contrary to precursor IL-33, mature IL-33 stimulation promptly phosphorylated IRAK1 and NF-B in a time-dependent manner, reached the maximal level at 15 min after exposure, and decreased dramatically at 30 min. However, the phosphorylation of I-B was delayed and reached a maximum at 60 min. The phosphorylation of p38 MAPK, ERK1/2, and JNK was similar to that of NF-B; however, its phosphorylation was sustained longer than that of IRAK1 and NF-B (Fig. 9B,  bottom panels). The phosphorylation pattern of NF-B, p38 MAPK, ERK1/2, and JNK was similar to IRAK1, although there was a difference in the density of activation. This result suggested the induction of inflammatory cytokines through the activation of IRAK1, NF-B, p38 MAPK, ERK1/2, and JNK signal pathways.
The role of caspase-1 in the activation of IL-33 was unclear in previous studies. The original study of IL-33 suggested that caspase-1 cleaves the amino acid residue (Asp 110 -Ser 111 ) on human precursor IL-33. This result was generated by the incubation of in vitro-translated IL-33 rather than precursor IL-33 protein (6). Interestingly, our result in Fig. 3A showed that human and mouse precursor IL-33 proteins were not cleaved by caspase-1, whereas precursor IL-18 protein was specifically processed by caspase-1 and became an active mature IL-18 (45).
The precursor IL-33 ligand affinity column isolated urinary PR3 from concentrated human urine (Fig. 1). This result and previous reports of PR3 in the processing of IL-1␤ (IL-1F2) and IL-18 (IL-1F4) led us to investigate whether PR3 is an enzyme for IL-33 processing. In the presence of PR3, precursor IL-33 gained its biological function at 5 min of incubation, and a time-dependent degradation terminated its biological activity after 30 min of incubation (Fig. 2). The rIL-33 and rIL-18 proteins were cleaved resulting in productions of multibands (Fig. 3B) unlike caspase-1-cleaved precursor IL-18 (Fig. 3A).
In addition, PR3 cleaved IL-33 proteins stimulated HMC-1 cells and induced IL-8 production; however, its activity was reduced along with increasing PR3 incubation time. These results suggest that PR3 has dual functions (activation/termination) in IL-33 biology during infection or inflammation. The time-dependent reduction of IL-33 activity (Fig. 2B) was confirmed with an alternative method. We verified reduction of mature size IL-33 by Western blot (Fig. 4).
In contrast to the first report (6), it has been reported that full-length precursor IL-33 is active, and the processing by caspase-1 results in IL-33 inactivation (34). Another report on IL-33 processing by Luthi et al. (35) demonstrated that apoptotic caspases process precursor IL-33 in apoptotic cells resulting in inactivation of precursor IL-33, whereas precursor IL-33 released from necrotic cell is spontaneously active. The result suggested that the cleavage site of caspase-1 does not occur at the site initially proposed (Asp 110 -Ser 111 ) but rather at amino acid residue Asp 178 -Gly 179 , which is the consensus site of cleavage by caspase-3 (34,35,46). Apoptotic caspase-3 and Ϫ7 abolishes the biological activity of precursor IL-33 by cleavage  of the amino acid residue Asp 178 -Gly 179 . However, these studies have been performed in vitro; the biological activity of the precursor IL-33 was shown by only NF-B luciferase assay (35) upon transfection of precursor IL-33 cDNA. Unlike the previous study, the present experiments were performed with PR3cleaved forms of mature human IL-33 (Ser 117 -Thr 270 ) and mouse IL-33 (Ser 114 -Ile 266 ) proteins, inducing inflammatory cytokines and activating the phosphorylation of signaling molecules.
Another study reported calpain-dependent processing of precursor IL-33 like IL-1␣, which was exhibited by treatment with calcium ionophore. Calpain processes precursor IL-33, which produces mature IL-33 in human epithelial and endothelial cells in the absence of IL-33 activity (47). Contrary to this result, Ohno et al. (48) reported that caspase-1/8 and calpain are dispensable for IL-33 release from macrophage and mast cells. Endogenous mouse IL-33 was secreted spontaneously. IL-33 secretion from cells was increased by LPS or phorbol 12-myristate 13-acetate plus ionomycin in the peritoneal macrophages of caspase-1-deficient BALB/c mice (48). The discrepancy of these results may be due to species differences. Although there is significant sequence homology between human and mouse IL-33, they share only 55% identity at amino acid level. The experiment of calpain-dependent IL-33 processing was performed with human cell lines, but the opposite report by Ohno et al. (48) was carried out using the peritoneal macrophage of caspase-1-deficient mice.
PR3-cleaved precursor IL-33 sufficiently induces inflammatory cytokines, and the biological activity of mature IL-33/p1, which was produced by PR3 cleavage site at the N terminus, was highly active. Although the predicted caspase-1 cleavage site (Asp 110 -Ser 111 ) is unclear, commercial mature human IL-33 (Ser 112 -Thr 270 ) and mouse IL-33 (Ser 109 -Ile 266 ) are highly active in the production of cytokine. Our mature human IL-33/p1 (Ser 117 -Thr 270 ) and mouse IL-33/p1 (Ser 114 -Ile 266 ) exhibited potent activity, and the induction of cytokine production levels were very similar to that of commercial human and mouse mature IL-33 (Fig. 7A).
We produced human and mouse mature rIL-33 according to the consensus amino acid sequence of PR3 cleavage sites in a previous report (38). Wysocka et al. (38) described the chemical synthesis of the selective chromogenic/fluorogenic substrates and deconvolution of the tri-peptide library against proteinase 3 with the general formula ABZ-X 3 X 2 X 1 -ANB-NH 2 , which yielded the active sequence. The identification of the predicted PR3-cleaved p1, p2, and 3 fragments by enzymatic activity of PR3 was extremely difficult because PR3 cleaves pro-IL-33 transiently and eventually abolishes IL-33 activity by fragmentation (Fig. 3B). As shown in Fig. 4, PR3 digested mature IL-33 completely at 30 min. PR3-cleaved human and mouse p2 are indicated by arrows (2nd and 4th lanes of Fig. 3B), but p1 and p3 were indistinguishable because p1 migrated as PR3 (far right lane of Fig. 3B compared with Western blot in Fig. 4) and p3 migrated as pro-IL-33 (2nd lane of Fig. 5B compared with that of the 4th lane).
Although our prediction of the PR3 cleavage site Ile 116 -Ser 117 of human IL-33 and Thr 113 -Ser 114 of mouse IL-33 is a highly conserved consensus sequence of a potential PR3 cleav-  MARCH 9, 2012 • VOLUME 287 • NUMBER 11 age site, as indicated by a previous report (38), we cannot exclude alternative PR3 cleavage sites of human IL-33 (Ser 111 -Ser 112 ) and of mouse IL-33 (Leu 108 -Ser 109 ). The alternative PR3 cleavage site was used for producing human and mouse mature rIL-33 (R&D Systems).

Interleukin-33 Processing Enzyme
Collectively, we identified PR3 as a processing enzyme of IL-33. In this study, we proposed PR3 cleavage sites and expressed various forms of rIL-33 proteins to verify the contrast functions (activation/termination) of PR3 in the processing of IL-33 during infection or inflammation. Further in vivo studies on PR3-induced IL-33 processing by neutrophils or monocytes could help understand inflammatory disorders of epithelial tissues such as asthma and atopic dermatitis.