| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
* Laboratoire d’Enzymologie Moléculaire, Institut de Biologie Structurale Jean-Pierre Ebel (Commissariat à l’Energie Atomique-Centre National de la Recherche Scientifique-Université Joseph Fourier), Grenoble, France;
Department of Infection, Immunity, and Inflammation, University of Leicester, Leicester, England;
Department of Medical Microbiology and Immunology, University of Aarhus, Aarhus, Denmark;
Department of Applied Biochemisty and Institute of Glycobiology, Tokai University, Hiratsuka, Japan; and
¶ Department of Biochemistry, Fukushima Medical University School of Medicine, Fukushima, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ala in MASP-1 and Ser618Ala in MASP-2 stabilized the latter in their proenzyme form. Likewise, the MASP-1 and MASP-2 mutants were each activated by their active counterparts, but MASP-3 S645A was not. Activated MASP-3 did not react with C1 inhibitor; had no activity on complement proteins C2, C4, and C3; and only cleaved the N-carboxybenzyloxyglycine-L-arginine thiobenzyl ester substrate to a significant extent. Based on these observations, it is postulated that MASP-3 activation and control involve mechanisms that are different from those of MASP-1 and -2. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, we describe the production at a high yield of recombinant human MASP-3 using a baculovirus-insect cells expression system. Physicochemical and functional characterization of the recombinant protein shows that it shares some of the properties already described for MASP-1 and -2, including the ability to form Ca2+-dependent dimers and to associate with MBL and L-ficolin. In contrast, MASP-3 does not undergo spontaneous self-activation, as now demonstrated in the case of MASP-1 and -2, and does not share the ability of these proteases to react with C1 inhibitor in the active state. Based on these findings, we suggest that activation and control of MASP-3 involve mechanisms different from those of MASP-1 and -2.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pEAK8/pMASP-3 plasmid containing the full-length MASP-3 cDNA was obtained as described previously (6). Oligonucleotides were obtained from Oligoexpress (Paris, France). VentR polymerase was from New England Biolabs (Beverly, MA). Di-isopropyl phosphorofluoridate was from Acros Organics (Noisy-le-Grand, France). trans-Epoxy-succinyl-L-leucyl-amido(4-guanidino)butane (E-64), iodoacetamide, leupeptin, pepstatin A, 1,10-phenanthroline, N-acetylglycine-L-lysine methyl ester (Ac-Gly-Lys-OMe), N
-benzoyl-L-arginine ethyl ester (Bz-Arg-OEt), and p-tosyl-L-arginine methyl ester (Tos-Arg-OMe) were obtained from Sigma-Aldrich (Saint Quentin Fallavier, France). N-Carboxybenzyloxyglycine-L-arginine thiobenzyl ester (Z-Gly-Arg-S-Bzl) was from Enzyme Systems Products (Livermore, CA). Heparin (from porcine intestinal mucosa) and dextran sulfate 5000 were obtained from Sigma-Aldrich. Sepharose-MBL was prepared by coupling purified recombinant human MBL produced according to Vorup-Jensen et al. (19) to CNBr-activated Sepharose 4B (0.2 mg of protein per milliliter of resin; Amersham Pharmacia Biotech, Piscataway, NJ).
Proteins
Activated C1s, C1 inhibitor, and complement proteins C2, C3, and C4 were purified from human plasma according to published procedures (12, 20, 21, 22, 23). MBL was purified from human serum as described by Dumestre-Perard et al. (24), with the following modifications. The gel filtration step was replaced by cation exchange chromatography on a Resource S column (Amersham Pharmacia Biotech) conducted as described by Thielens et al. (17). Anion exchange chromatography on a Mono-Q column (Amersham Pharmacia Biotech) was performed under conditions allowing separation of the MBL-I and MBL-II species, as described by Dahl et al. (6). L-ficolin/P35 was isolated from human plasma as described by Cseh et al. (9). Recombinant MASP-1 and MASP-2 were expressed in a baculovirus/insect cells system and purified as described previously (9, 17). The concentration of purified recombinant wild-type or mutated MASP-3 was determined using an absorption coefficient (A1%, 1 cm at 280 nm) of 12.9 calculated by the method of Edelhoch (25) and a molecular mass of 87,600 Da, as determined by mass spectrometry. Due to the low amount of material recovered, estimation of the concentration of recombinant MASP-1 and MASP-2 was based on Coomassie blue staining after SDS-PAGE analysis using appropriate internal standards and molecular masses of 82,000 and 75,100 Da, respectively.
Construction of the MASP-3-containing expression plasmid
A DNA fragment encoding the MASP-3 signal peptide plus the mature protein (amino acid residues 1–709) was amplified by PCR using VentR polymerase and pEAK8/pMASP-3 as a template, according to established procedures. The sequence of the sense primer (5'-GAAGATCTATGAGGTGGCTGCTTCT-3') introduced a BglII restriction site (underlined) at the 5' end of the PCR product and that of the antisense primer (5'-CGGAATTCTCACCGTTCCACCTGGGG-3') introduced a stop codon (bold) followed by an EcoRI site (underlined) at the 3' end. The PCR fragment was digested with BglII and EcoRI, purified, and cloned into the BamHI/EcoRI sites of the pFastBac1 baculovirus transfer vector (Invitrogen, San Diego, CA). The final construct was analyzed by dsDNA sequencing (Genome Express, Grenoble, France).
Site-directed mutagenesis
The expression plasmids coding for the MASP-1 S627A, MASP-2 S618A, and MASP-3 S645A mutants were generated using the QuickChange XL site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer’s protocol. The expression plasmids coding for wild-type MASP-1 (17), MASP-2 (9), and MASP-3 (see above) were used as templates. Mutagenic oligonucleotide primers and their complementary antisense primers were designed to substitute alanine (bold codon) for the active-site serine and to create a NaeI restriction site (underlined) to identify the mutants: MASP-1 S627A, sense primer 5'-CCTGTGCGGGTGACGCCGGCGGCCCCATGGTGACC-3'; MASP-2 S618A, sense primer 5'-GCTGCAGAGGTGACGCCGGCGGGGCACTGGTGTTTG-3'; and MASP-3 S645A, sense primer 5'-CGTGCCTTGGAGATGCCGGCGGGGCCTTTGTCATC-3'. The sequences of all mutants were confirmed by dsDNA sequencing.
Cells and viruses
The Spodoptera frugiperda (Ready-Plaque Sf9 cells; Novagen, Madison, WI) and Trichoplusia ni (High Five cells; Invitrogen) insect cells were routinely grown and maintained as described previously (17). The recombinant baculoviruses were generated using the Bac-to-Bac system (Invitrogen), amplified, and titrated as described previously (17).
Protein production and purification
High Five cells (1.75 x 107 cells per 175-cm2 tissue culture flask) were infected with the recombinant viruses at a multiplicity of infection of 2 in Sf900 II serum-free medium (Invitrogen) at 28°C for 72 h. Culture supernatants were collected by centrifugation.
The MASP-1 S627A and MASP-2 S618A mutated proteins were purified according to the protocol used for their wild-type counterparts (9, 17).
The culture supernatant containing wild-type MASP-3 or the S645A mutant was dialyzed against 50 mM NaCl and 50 mM triethanolamine hydrochloride (pH 7.0), and loaded at 1.5 ml/min onto a Q-Sepharose Fast Flow column (2.8 x 10 cm; Amersham Pharmacia Biotech) equilibrated in the same buffer. Elution was conducted by applying a 800-ml linear gradient to 500 mM NaCl in the same buffer. Fractions containing the recombinant protein were identified by SDS-PAGE analysis, and dialyzed against 145 mM NaCl, 1 mM CaCl2, and 50 mM triethanolamine hydrochloride (pH 7.4). Further purification was achieved by affinity chromatography on a Sepharose-MBL column equilibrated in the same buffer. Elution was realized by applying the same buffer containing either 3 M MgCl2 or 5 mM EDTA instead of CaCl2. The purified protein was dialyzed against 145 mM NaCl and 50 mM triethanolamine hydrochloride (pH 7.4), and concentrated by ultrafiltration to 0.3 mg/ml.
Chemical characterization of the recombinant proteins
N-terminal sequence analyses were performed after SDS-PAGE and electrotransfer, using an Applied Biosystems (Foster City, CA) model 477 A protein sequencer as described previously (26). Mass spectrometry analyses were performed using the matrix-assisted laser desorption ionization technique on a Voyager Elite XL instrument (PerSeptive Biosystems, Cambridge, MA), under conditions described previously (27).
Gel electrophoresis and immunoblotting
SDS-PAGE analysis was performed as described previously (28). Western blot analysis and immunodetection of the recombinant proteins were conducted as described by Rossi et al. (29), using the mouse monoclonal anti-MASP-2 Ab 1.3B7 (30) or rabbit anti-peptide Abs directed against either the serine protease domain of MASP-1 or the N-terminal end of MASP-1/3 (5).
Gel permeation chromatography
Recombinant MASP-3 was analyzed by high-pressure gel permeation chromatography on a TSK G3000 SWG column (7.5 x 600 mm; Tosoh Biosep, Montgomeryville, PA) equilibrated in 145 mM NaCl and 50 mM triethanolamine hydrochloride (pH 7.4), containing either 1 mM EDTA or 1 mM CaCl2 and run at 1 ml/min. Proteins were detected from their absorbance at 280 nm.
Analytical ultracentrifugation
Sedimentation velocity analysis was performed using a Beckman XL-I analytical ultracentrifuge and an AN-60 TI rotor (Beckman Instruments, Fullerton, CA). Experiments were conducted at 20°C in 145 mM NaCl and 50 mM triethanolamine hydrochloride (pH 7.4), containing either 1 mM CaCl2 or 1 mM EDTA. Samples of 400 µl of MASP-3 (0.30 mg/ml) or C1s (0.55 mg/ml) were loaded into 12-mm path-length double-sector cells and centrifuged at 42,000 rpm. Absorbance of the samples was recorded at 275 nm. The sedimentation coefficients (s) were obtained by fitting the sedimentation velocity profiles to the noninteracting single-component model using the SEDFIT program (www.biochem.uthscsa.edu/auc/software), and the continuous distribution of sedimentation coefficients was obtained considering globular proteins. Solvent density was calculated at 1.0057 g/ml, and partial specific volumes were estimated from the amino acid composition at 0.7169 ml/g (MASP-3) and 0.7204 ml/g (C1s), using the SEDNTERP program (www.bbri.org/RASMB/rasmb.html), which was also used for calculation of the corrected s20,w values.
Proteolytic assays and reactivity toward C1 inhibitor
Activated MASP-3 was obtained by incubating proenzyme MASP-3 at 4°C for 8 wk. The proteolytic activity of activated MASP-3 toward C2, C3, and C4 was tested by overnight incubation at 37°C at a 1:10 enzyme:substrate molar ratio, followed by SDS-PAGE analysis under reducing conditions.
The reactivity of MASP-3 toward C1 inhibitor was determined by incubation of the protease with a 4-fold molar excess of inhibitor for 4 h at 37°C, either in the presence or absence of heparin (0.35 mg/ml) or dextran sulfate 5000 (0.1 mg/ml), followed by SDS-PAGE analysis under reducing conditions. Heparin and dextran sulfate were both preincubated with C1 inhibitor for 10 min at 37°C before addition of the protease.
Effect of protease inhibitors on activation of the MASP-3 S645A mutant
Recombinant MASP-3 S645A was incubated in the presence of various protease inhibitors for 4 wk at 4°C. Activation was assessed by SDS-PAGE analysis under reducing conditions and subsequent quantification by gel scanning of the A and B chains characteristic of activated MASP-3.
Real-time surface plasmon resonance spectroscopy and data evaluation
Surface plasmon resonance measurements were performed using a Biacore 3000 instrument (Biacore, Uppsala, Sweden). The running buffer for protein immobilization was 145 mM NaCl, 5 mM EDTA, and 10 mM HEPES (pH 7.4). Protein ligands were diluted to 30 µg/ml in 10 mM acetate (pH 4.0 (MBL) or pH 5.0 (L-ficolin/P35)) and immobilized onto the carboxymethylated dextran surface of a CM5 sensor chip (Biacore) using the amine coupling chemistry (Biacore amine coupling kit) according to the manufacturer’s instructions. Binding of the wild-type and mutant MASP-1, MASP-2, and MASP-3 species was measured over 16,000 resonance units (RU) of immobilized L-ficolin/P35 or 3,000–4,000 RU of immobilized MBL, at a flow rate of 20 µl/min in 145 mM NaCl, 1 mM CaCl2, and 50 mM triethanolamine hydrochloride (pH 7.4), containing 0.005% surfactant P20 (Biacore). Equivalent volumes of each protein sample were injected over a surface with immobilized BSA to serve as blank sensorgrams for subtraction of the bulk refractive index background. Regeneration of the surfaces was achieved by injection of 10 µl of 5 mM EDTA in the case of MBL, and of 10 µl of 1M NaCl/20 mM EDTA in the case of L-ficolin/P35.
Data were analyzed by global fitting to a 1:1 Langmuir binding model of both the association and dissociation phases for several concentrations simultaneously, using the BIAevaluation 3.1 software (Biacore). The apparent equilibrium dissociation constants (KD) were calculated from the ratio of the dissociation and association rate constants (koff/kon).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The recombinant baculovirus for expression of wild-type MASP-3 was generated as described in Materials and Methods and used to infect High Five insect cells for 72 h at 28°C. The amount of recombinant material recovered in the culture medium was estimated by SDS-PAGE and Western blot analysis at 
8 µg/ml. Recombinant MASP-3 was purified to homogeneity as described in Materials and Methods. SDS-PAGE analysis of the purified material showed that the protein migrated as a single band of 94 kDa both under nonreducing and reducing conditions, indicating that the protease had retained a single-chain, proenzyme structure (Fig. 1A). Consistent with this observation, the protein yielded a single N-terminal sequence (His-Thr-Val-Glu-Leu-Asn-Asn-Met-Phe-Gly...) corresponding to that of mature MASP-3. Analysis by mass spectrometry yielded a heterogeneous peak centered on a mass value of 87,581 ± 44 Da. Given the predicted mass of the polypeptide moiety of the protein (79,655 Da), this yields a deduced mass value of 7,926 ± 44 Da for the carbohydrate moiety, consistent with the occurrence of seven oligosaccharide chains comprising two N-acetylglucosamine and four or five mannose residues (calculated masses, 1,055–1,217 Da).
|
Analysis of rMASP-3 by sedimentation velocity was performed as described in Materials and Methods, yielding sedimentation coefficients (s20,w) of 6.2 ± 0.1 and 4.6 ± 0.1 S in the presence of Ca2+ and EDTA, respectively (Fig. 2, A and B). Comparative analysis of human C1s under the same conditions yielded s20,w values of 5.6 ± 0.1 S in the presence of Ca2+ (Fig. 2C) and of 4.0 ± 0.1 S in the presence of EDTA (D), in agreement with the values obtained previously for the dimeric and monomeric forms of C1s (31, 32). These results are consistent with a dimerization of MASP-3 in the presence of Ca2+. Analysis of the recombinant protein by size-exclusion chromatography indicated that EDTA significantly increased its elution position from 13.2 min in the presence of Ca2+ ions to 15.8 min, further supporting the hypothesis that MASP-3 forms Ca2+-dependent and EDTA-sensitive homodimers.
|
The ability of MASP-3 to associate with MBL and L-ficolin/P35 was studied using surface plasmon resonance spectroscopy. As shown in Fig. 3, MASP-3 bound to immobilized MBL and L-ficolin/P35 in the presence of 1 mM CaCl2. When EDTA was substituted for Ca2+ in the running buffer, no significant binding was detectable in the case of MBL (Fig. 3A), whereas a residual binding of 10% was still observed in the case of L-ficolin/P35 (B). The kinetic parameters of both interactions were determined by recording sensorgrams at various protein concentrations (10–100 nM) in the presence of 1 mM CaCl2 and evaluating the data by global fitting as described in Materials and Methods. The association (kon) and dissociation (koff) rate constants and the resulting apparent KD values are listed in Table I. In both cases, the KD values are in the nM range (2.6 and 7.2 nM for binding to MBL and L-ficolin/P35, respectively), the slightly higher value for L-ficolin/P35 mainly resulting from a higher dissociation rate constant (Table I). As observed previously in the case of MASP-1 and MASP-2 (9), preincubation of MASP-3 with soluble MBL inhibited subsequent binding to immobilized L-ficolin/P35 and conversely (Fig. 3), indicating that MBL and L-ficolin compete with each other for binding to MASP-3. It has been shown previously that serum MBL consists of a mixture of oligomers of different sizes with two major forms, MBL-I and MBL-II, and that serum MASP-3 preferentially associates with the latter species (6). MBL-I and MBL-II were isolated as described in Materials and Methods, and the ability of MASP-3 to associate with both MBL forms was tested. The recombinant protease was found to bind to immobilized MBL-I and MBL-II with kon values and koff values of the same order (2.4 and 2.1 x 105 M-1·s-1; 6.2 and 5.3 x 104 s-1, respectively), yielding dissociation constants of 2.6 and 2.5 nM respectively, which were similar to the value obtained using the whole MBL fraction (Table I).
|
|
Production and characterization of MASP-1 S627A, MASP-2 S618A, and MASP-3 S645A
The baculoviruses used for expression of the mutated MASPs in which Ala was substituted for the catalytic Ser residue of each protease were obtained and used to infect High Five cells as described in Materials and Methods. The secretion yields of the mutants were estimated by SDS-PAGE and Western blot analysis of the culture supernatants at 3, 0.15, and 8 µg/ml for the MASP-1, MASP-2, and MASP-3 mutants, respectively. Each of the three mutants migrated under reducing conditions as a single band at 90, 78, and 94 kDa, respectively (Fig. 4), characteristic of the proenzyme form of the proteases. In contrast, the supernatant containing wild-type MASP-1 yielded an additional band at 30 kDa, reactive to Abs directed to the C-terminal end of the molecule and corresponding to the serine protease domain (Fig. 4A), confirming previous observations that the protease undergoes partial activation during the synthesis process (17). Wild-type MASP-2 was also recovered in a partially activated form, as shown by the presence of a 46-kDa band reactive with Abs directed against the N-terminal end of the molecule and corresponding to the A chain (Fig. 4B). In addition, the S618A mutation of MASP-2 abolished production of the N-terminal truncated fragment observed with the wild-type species (9, 12). In the case of MASP-3, no difference was observed between the S645A mutant and its wild-type counterpart (Fig. 4C).
|
|
As shown by SDS-PAGE analysis under reducing conditions, the three mutants retained a single-chain structure after overnight incubation at 37°C (Fig. 6, A–C, lane 2). However, and somewhat surprisingly, MASP-3 S645A was found to undergo slow activation upon prolonged storage at 4°C, with 70% activation after 3 wk and complete cleavage after 1 mo, as revealed by the occurrence of two bands at 56 and 36 kDa, similar to those observed upon storage of wild-type MASP-3 (see Fig. 1B). N-terminal sequence analysis of the 36-kDa fragment confirmed that cleavage had occurred at the susceptible Arg430-Ile431 bond of MASP-3. Because the mutated protease is expected to lack intrinsic proteolytic activity, we checked for the presence in the purified material of possible contaminant proteases from the culture supernatant. MASP-3 S645A was thus incubated for 4 wk at 4°C in the presence of protease inhibitors of various specificities, and their effect on activation was tested as described in Materials and Methods. As summarized in Table II, only the cysteine protease inhibitor iodoacetamide and the metalloprotease inhibitors 1,10-phenanthroline and EDTA had a significant inhibitory effect on MASP-3 activation, to relative extents of 75, 95, and 35%, respectively. Consistently, incubation of the mutant protease in the presence of both iodoacetamide and 1,10-phenanthroline totally blocked the activation process. When the above experiments were performed using wild-type MASP-3, similar effects were observed, with again complete inhibition of activation using a mixture of iodoacetamide and 1,10-phenanthroline. These results provide unambiguous evidence that the observed activation of wild-type MASP-3 does not result from self-activation, but from cleavage by a contaminating protease from the cell culture supernatant. The observation that proenzyme MASP-3 was sensitive to activation at 4°C but was stable upon overnight incubation at 37°C suggested that the contaminating proteolytic enzyme was temperature sensitive. To test this hypothesis, MASP-3 samples were incubated for 4 h at 37°C, followed by prolonged storage at 4°C. This treatment did not abolish subsequent cleavage of MASP-3, indicating that incubation of the extrinsic protease at 37°C did not induce irreversible inactivation of this enzyme.
|
|
Interaction of the mutant MASPs with MBL and L-ficolin/P35
The ability of the SerAla MASP mutants to associate with MBL and L-ficolin/P35 was studied using surface plasmon resonance spectroscopy. As previously observed in the case of the wild-type proteases, the mutant MASPs bound to immobilized MBL and L-ficolin/P35 in the presence of Ca2+ (Fig. 7), and binding of the proteins was inhibited when EDTA was substituted for Ca2+ in the running buffer. Whereas interaction with MBL was totally prevented in the presence of EDTA in each case (Fig. 7, A–C), residual binding of the mutated MASPs to L-ficolin/P35 was observed, to relative extents from 25% (MASP-2) to <10% (MASP-1 and MASP-3) (D–F), as observed previously with the wild-type proteases (9).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As shown by mass spectrometry analysis, rMASP-3 was produced in a glycosylated form, the experimental mass (87,581 ± 44 Da) being consistent with occupancy of the seven predicted glycosylation sites (6) by short high-mannose oligosaccharides. Sedimentation velocity and gel filtration analyses both indicate that MASP-3 forms homodimers in the presence of Ca2+ ions, a feature shared by MASP-1 and MASP-2 and resulting from the ability of their N-terminal CUB1-EGF-CUB2 segments to associate (17, 18, 37). However, there is a notable difference between full-length MASP-3 and its N-terminal CUB1-EGF and CUB1-EGF-CUB2 fragments, because EDTA induces dissociation of the former, but has no significant effect on the sedimentation behavior of the latter (17, 37). Although this clearly indicates that Ca2+ ions take part in the dimerization process, as now established by x-ray crystallography in the case of human C1s and rat MASP-2 (38, 39), this suggests that Ca2+ could become less accessible to EDTA in the CUB1-EGF and CUB1-EGF-CUB2 fragments than in the whole MASP-3 protein, suggesting that the latter undergo some conformational rearrangement.
Analysis of the interaction of MASP-3 with MBL and L-ficolin/P35 also reveals a behavior similar to that observed for MASP-1 and MASP-2. Thus, the KD values for the binding of MASP-3 to MBL (2.6 nM) and L-ficolin/P35 (7.2 nM) are comparable with those determined previously for MASP-1 (3.2 and 9.2 nM) and MASP-2 (2.6 and 4.6 nM) (9, 17). These results were expected, because the CUB1-EGF-CUB2 modules of MASP-3 are identical with those of MASP-1, which have been shown to mediate binding of the protease to MBL and L-ficolin (9, 37). Binding was clearly Ca2+ dependent, but was only partly sensitive to EDTA in the case of L-ficolin/P35, as previously observed for MASP-1 and MASP-2 (9). Consistent with these observations, the Ca2+ concentrations yielding half-maximal binding of MASP-3 to MBL and L-ficolin/P35 (0.46 and 2.4 µM, respectively) (N. M. Thielens, unpublished observations) were strikingly similar to those obtained previously for MASP-1 (0.47 and 2.7 µM) (9). In the same way, we found that MBL and L-ficolin/P35 were able to compete with each other for binding to MASP-3, as previously observed in the case of MASP-1 and -2 (9). It is also noteworthy that no significant difference was found between the binding constants determined for the SerAla MASP mutants and their wild-type counterparts (see Table I), indicating that the activation state of the proteases has no significant influence on their interaction properties.
It was shown recently that the MASPs in serum preferentially associate with distinct MBL oligomers, MASP-1 interacting with low-size MBL oligomers (MBL-I), and MASP-2 and MASP-3 associating with higher oligomeric species (MBL-II) (6). A nonenzymic protein termed 19-kDa MBL-associated protein (MAp19) or small MBL-associated protein, comprising the CUB1-EGF module pair of MASP-2 prolonged by four specific residues (40, 41), was also found in complex with MBL-I (6). Our analysis of the interaction between rMASP-3 and serum MBL-I and -II shows comparable binding constants for both MBL forms and therefore provides no evidence for a preferential association in vitro. In the same way, no significant difference was detected when comparing the interaction of recombinant MASP-1, MASP-2, and MAp19 with MBL-I and MBL-II (N. M. Thielens, unpublished observations). The reasons for this discrepancy are not clear. It may be emphasized in this respect that MASP-1 and MASP-3 on the one hand, and MASP-2 and MAp19 in contrast, comprise identical N-terminal interaction domains, and therefore are not expected to show differential MBL-binding properties.
A striking difference between MASP-3 and the homologous proteases MASP-1 and -2 was revealed by analysis of the activation state of the recombinant material. Whereas MASP-1 and MASP-2 both undergo partial activation during biosynthesis and secretion, MASP-3 retains a single-chain, proenzyme structure. Mutation of the active serine residue of MASP-1 and MASP-2 prevents their activation and also abolishes proteolytic degradation observed during expression of the wild-type species, yielding N-terminal fragments of 28 and 45 kDa in the cases of MASP-1 and MASP-2, respectively (9, 12, 17). These results, together with the observed ability of activated MASP-1 and MASP-2 to activate their proenzyme SerAla mutant counterparts, provide the first unambiguous experimental evidence of the intrinsic ability of MASP-1 and MASP-2 to undergo activation through autolytic cleavage. Although wild-type MASP-3 was produced in a proenzyme form, the purified protein was found to undergo slow cleavage at the Arg440-Ile441 activation site upon storage at 4°C. This suggested that either MASP-3 underwent self-activation, or that extrinsic activation was mediated by a contaminating protease from the host cell system. In this respect, the presence of cysteine protease activity in the supernatant of baculovirus-infected insect cells has been reported in several studies (42, 43, 44, 45). One of these reports describes a baculovirus cathepsin-like cysteine protease with the ability to cleave peptide substrates with an Arg residue at the P1 position (43), which may account for the cleavage of the Arg440-Ile441 bond of MASP-3. This hypothesis would be consistent with the lack of proteolytic activity observed at 37°C, considering that insect cells are grown at 27–28°C and are not viable at 37°C. The hypothesis of an extrinsic activation of MASP-3 is fully supported by the following observations: 1) substitution of Ala for the active site Ser645 of MASP-3 did not abolish cleavage of the susceptible Arg440-Ile441 bond; 2) activation was prevented in the presence of the cysteine and metalloprotease inhibitors iodoacetamide and 1,10-phenanthroline; and 3) classical serine protease inhibitors such as DFP had no effect. It should also be emphasized that, in contrast to MASP-1 and MASP-2, activated MASP-3 was not able to cleave its proenzyme counterpart, providing further illustration that, under the experimental conditions used in the present study, MASP-3 is not able to self-activate.
Activated MASP-3 exhibited a weak esterolytic activity on the thioester Z-Gly-Arg-S-Bzl, with a kcat/Km ratio of 19.4 x 103 M-1·s-1, comprised between the values obtained previously for MASP-1 and C1r (4 and 47 x 103 M-1·s-1, respectively) (12). None of the other synthetic substrates cleaved by MASP-1 and MASP-2 (12) was cleaved by MASP-3, suggesting a highly restricted specificity. It should be stressed that this activity was not detected using freshly purified proenzyme MASP-3 and blocked in the presence of the serine protease inhibitor DFP, making unlikely a possible contribution of the contaminating protease to the observed esterolytic activity. Activated MASP-3 also exhibited no proteolytic activity toward C2, C3, and C4, thereby confirming that it is not directly involved in the initiation of the lectin pathway of complement.
Another major difference with MASP-1 and MASP-2 lies in the fact that MASP-3 does not react with C1 inhibitor, as indicated by the lack of complex formation with this inhibitor and its absence of effect on the esterolytic activity of MASP-3. Similar properties, i.e., a lack of reactivity toward C2, C3, C4, and C1 inhibitor were observed in the case of rMASP-3 expressed in mammalian cells (Ref. 6 ; M. R. Dahl et al., unpublished data), indicating that these characteristics likely reflect the actual properties of MASP-3 in serum and are not due to a dysfunction related to the particular expression system used.
It has been shown that activated MASP-1 and MASP-2 each react with C1 inhibitor in a 1:1 stoichiometry (12), suggesting that their proteolytic activity in serum is specifically controlled by C1 inhibitor, as well established in the case of C1r and C1s (46). It is also known that C1 inhibitor prevents slow spontaneous activation of the C1 complex in the absence of activator (47) and generally admitted that this negative control is released upon binding of C1 to an activator, allowing activation to proceed. It may thus be proposed that, in a comparable manner, MASP-1 and MASP-2 self-activation is controlled by C1 inhibitor inside MBL- and ficolin-MASP complexes. Binding of these complexes to carbohydrate arrays on pathogenic microorganisms would then overcome the negative control exerted by C1 inhibitor and elicit activation of the MASP.
In contrast to MASP-1 and MASP-2, purified rMASP-3 does not self-activate and shows no reactivity toward C1 inhibitor. Nevertheless, the MASP-3 fraction found in association with MBL or with L- and H-ficolins in serum is recovered in a two-chain, activated form (6, 48). Based on these elements, we postulate that MASP-3 activation proceeds through a mechanism that is different from that of MASP-1 and MASP-2, and possibly requires a yet-unidentified serum cofactor and/or highly restrictive conditions in terms of ligand recognition. The ancient origin of MASP-3, which was traced back to the amphioxus lineage (49), suggests that it may serve (an) important function(s) conserved through evolution. Full understanding of the role of this protease will clearly require further investigation aimed at discovering its physiological substrate(s) as well as elucidating its activation mechanism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Preliminary reports of this study were presented by N.M.T. at the XIX International Complement Workshop in Palermo, Italy, September 22–26, 2002, and at the IX European Meeting on Complement in Human Disease in Trieste, Italy, September 6–9, 2003.
3 Address correspondence and reprint requests to Dr. Nicole M. Thielens, Laboratoire d’Enzymologie Moléculaire, Institut de Biologie Structurale Jean-Pierre Ebel, 41 rue Jules Horowitz, 38027 Grenoble Cedex 1, France. E-mail address: nicole.thielens{at}ibs.fr
4 Abbreviations used in this paper: MBL, mannan-binding lectin; MASP, MBL-associated serine protease; RU, resonance unit; DFP, diisopropyl phosphorofluoridate; CUB module, module originally found in complement proteins C1r/C1s, Uegf, and bone morphogenetic protein-1; MAp19, 19-kDa MBL-associated protein; EGF, epidermal growth factor.
5 The nomenclature of protein modules is that defined by Bork and Bairoch (3 ).
Received for publication October 21, 2003. Accepted for publication January 23, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
fragments of human complement proteases C1r and C1s. J. Biol. Chem. 265:14469.
-hydroxyasparagine, sialic acid, and one of its two carbohydrate chains still reassembles with C1q and C1r to form a functional C1 complex. Biochemistry 31:4254.[Medline]
This article has been cited by other articles:
|
|
 |
|
  I. Bally, V. Rossi, T. Lunardi, N. M. Thielens, C. Gaboriaud, and G. J. Arlaud Identification of the C1q-binding Sites of Human C1r and C1s: A REFINED THREE-DIMENSIONAL MODEL OF THE C1 COMPLEX OF COMPLEMENT J. Biol. Chem., July 17, 2009; 284(29): 19340 - 19348. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lacroix, C. Dumestre-Perard, G. Schoehn, G. Houen, J.-Y. Cesbron, G. J. Arlaud, and N. M. Thielens Residue Lys57 in the Collagen-Like Region of Human L-Ficolin and Its Counterpart Lys47 in H-Ficolin Play a Key Role in the Interaction with the Mannan-Binding Lectin-Associated Serine Proteases and the Collectin Receptor Calreticulin J. Immunol., January 1, 2009; 182(1): 456 - 465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Teillet, C. Gaboriaud, M. Lacroix, L. Martin, G. J. Arlaud, and N. M. Thielens Crystal Structure of the CUB1-EGF-CUB2 Domain of Human MASP-1/3 and Identification of Its Interaction Sites with Mannan-binding Lectin and Ficolins J. Biol. Chem., September 12, 2008; 283(37): 25715 - 25724. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. V. Girija, A. W. Dodds, S. Roscher, K. B. M. Reid, and R. Wallis Localization and Characterization of the Mannose-Binding Lectin (MBL)-Associated-Serine Protease-2 Binding Site in Rat Ficolin-A: Equivalent Binding Sites within the Collagenous Domains of MBLs and Ficolins J. Immunol., July 1, 2007; 179(1): 455 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Teillet, M. Lacroix, S. Thiel, D. Weilguny, T. Agger, G. J. Arlaud, and N. M. Thielens Identification of the Site of Human Mannan-Binding Lectin Involved in the Interaction with Its Partner Serine Proteases: The Essential Role of Lys55 J. Immunol., May 1, 2007; 178(9): 5710 - 5716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Wallis, A. W. Dodds, D. A. Mitchell, R. B. Sim, K. B. M. Reid, and W. J. Schwaeble Molecular Interactions between MASP-2, C4, and C2 and Their Activation Fragments Leading to Complement Activation via the Lectin Pathway J. Biol. Chem., March 16, 2007; 282(11): 7844 - 7851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Moller-Kristensen, S. Thiel, A. Sjoholm, M. Matsushita, and J. C. Jensenius Cooperation between MASP-1 and MASP-2 in the generation of C3 convertase through the MBL pathway Int. Immunol., February 1, 2007; 19(2): 141 - 149. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Iwaki, K. Kanno, M. Takahashi, Y. Endo, N. J. Lynch, W. J. Schwaeble, M. Matsushita, M. Okabe, and T. Fujita Small Mannose-Binding Lectin-Associated Protein Plays a Regulatory Role in the Lectin Complement Pathway J. Immunol., December 15, 2006; 177(12): 8626 - 8632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Rossi, F. Teillet, N. M. Thielens, I. Bally, and G. J. Arlaud Functional Characterization of Complement Proteases C1s/Mannan-binding Lectin-associated Serine Protease-2 (MASP-2) Chimeras Reveals the Higher C4 Recognition Efficacy of the MASP-2 Complement Control Protein Modules J. Biol. Chem., December 23, 2005; 280(51): 41811 - 41818. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Teillet, B. Dublet, J.-P. Andrieu, C. Gaboriaud, G. J. Arlaud, and N. M. Thielens The Two Major Oligomeric Forms of Human Mannan-Binding Lectin: Chemical Characterization, Carbohydrate-Binding Properties, and Interaction with MBL-Associated Serine Proteases J. Immunol., March 1, 2005; 174(5): 2870 - 2877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A. Gregory, N. M. Thielens, M. Matsushita, R. Sorensen, G. J. Arlaud, J. C. Fontecilla-Camps, and C. Gaboriaud The X-ray Structure of Human Mannan-binding Lectin-associated Protein 19 (MAp19) and Its Interaction Site with Mannan-binding Lectin and L-ficolin J. Biol. Chem., July 9, 2004; 279(28): 29391 - 29397. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
