Polymorphisms in Mannan-Binding (MBL)-Associated 2 Affect Stability, Binding to MBL, and Enzymatic Activity This information is current as of September 26, 2021. Steffen Thiel, Martin Kolev, Søren Degn, Rudi Steffensen, Annette G. Hansen, Marieta Ruseva and Jens C. Jensenius J Immunol 2009; 182:2939-2947; ; doi: 10.4049/jimmunol.0802053 http://www.jimmunol.org/content/182/5/2939 Downloaded from

References This article cites 31 articles, 14 of which you can access for free at:

http://www.jimmunol.org/content/182/5/2939.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 26, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Polymorphisms in Mannan-Binding Lectin (MBL)-Associated Serine Protease 2 Affect Stability, Binding to MBL, and Enzymatic Activity1

Steffen Thiel,2* Martin Kolev,* Søren Degn,* Rudi Steffensen,† Annette G. Hansen,* Marieta Ruseva,* and Jens C. Jensenius*

Mannan-binding lectin-associated serine protease 2 (MASP-2) is an of the innate . MASP-2 forms com- plexes with the pattern recognition molecules mannan-binding lectin (MBL), H-ficolin, L-ficolin, or M-ficolin, and is activated when one of these proteins recognizes microorganisms and subsequently cleaves complement factors C4 and C2, thus initiating the activation of the . Missense polymorphisms of MASP-2 exist in different ethnic populations. To further char- acterize the nature of these, we have produced and characterized rMASP-2s representing the following naturally occurring Downloaded from polymorphisms: R99Q, D120G, P126L, H155R, 156_159dupCHNH (CHNHdup), V377A, and R439H. Only very low levels of CHNHdup were secreted from the cells, whereas quantities similar to wild-type MASP-2 were found intracellularly, indicating that this mutation results in a misfolded protein. We found that D120G and CHNHdup could not associate with MBL, whereas R99Q, P126L, H155R, V377A, R439H, and wild-type MASP-2 bound equally well to MBL. Accordingly, when D120G and CHNHdup were mixed with MBL, no activation of complement factor C4 was observed, whereas R99Q, P126L, and V377A cleaved C4 with an activity comparable to wild-type MASP-2 and H155R slightly better. In contrast, the R439H variant was http://www.jimmunol.org/ deficient in this process despite its normal binding to MBL. This variant was also not able to autoactivate in the presence of MBL and mannan. We find the R439H variant is common in Sub-Saharan Africans with a frequency of 10%. Our results indicate that individuals with different types of MASP-2 defects may be identified through genotyping. The Journal of Immunology, 2009, 182: 2939–2947.

nderstanding the workings of the complement system is pathway are all found in complexes with the same set of proen- important for unraveling the innate immune defense and zymes, the MBL-associated serine proteases (MASPs): MASP-1, its interactions with the adaptive immune response (1, MASP-2, and MASP-3, as well as MBL-associated protein of 19 U by guest on September 26, 2021 2). The complement system includes more than 30 soluble and kDa, MAp19 (3–5). MASP-1, MASP-2, and MASP-3 all present membrane-bound proteins. A number of these proteins are serine domain structures identical with those of the C1 proteases, C1r proteases found as zymogens in the circulation. To activate these and C1s, and all five proenzymes are activated through the proenzymes, three pathways have evolved over time, as follows: 1) cleavage of the polypeptide chain between the A segment and the classical pathway is activated when the recognizes the B segment, with the latter presenting the protease domain patterns of Fc regions from Ig molecules; 2) the is (exemplified by the structure of MASP-2 in Fig. 1). MAp19 3 activated when mannan-binding lectin (MBL) or one of the three presents the two N-terminal domains (of six domains) of ficolins (H-ficolin, L-ficolin, or M-ficolin) recognizes a pattern of MASP-2, and in addition, four unique amino acids, and is de- carbohydrates or acetylated molecules; and 3) the alternative path- void of enzymatic activity (6). When MBL or ficolin in complex way is activated when the balance between inhibition and activa- with MASP recognizes a ligand, the MASPs are activated. The tion of C3 activation is shifted and is also important for amplifying physiological relevant substrates for activated MASP-1 and the first two pathways. The recognition molecules of the lectin MASP-3 are still under debate, although MASP-1 has been shown to cleave C2 and to possess some -like activity (7, 8). The activity of MASP-2, in contrast, is clearer. Activated *Department of Medical Microbiology and Immunology, University of Aarhus, Aar- MASP-2 very efficiently cleaves the complement factors C4 and hus, Denmark; and †Regional Centre for Blood Transfusion and Clinical Immunol- ogy, Aalborg Hospital, Aalborg, Denmark C2 to the fragments C4b and , and C2b and C2a, respec- Received for publication June 25, 2008. Accepted for publication December 17, 2008. tively, and C4b and C2b join to form a C3 convertase (9, 10). MASP-2 is thus a central molecule in the activation of the com- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance plement system as it translates the recognition of microorgan- with 18 U.S.C. Section 1734 solely to indicate this fact. isms (or altered self) by four pattern recognition molecules, i.e., 1 This work was partly supported by the Danish Research Council and Novo Nordic MBL, H-ficolin, L-ficolin, and M-ficolin, into initiation of the Foundation. enzymatic cascades of the complement system. 2 Address correspondence and reprint requests to Dr. Steffen Thiel, Department of Medical Microbiology and Immunology, University of Aarhus, Aarhus, Denmark. Uncovering mutations resulting in selective deficiencies in the E-mail address: [email protected] immune system may not only further our understanding of basic 3 Abbreviations used in this paper: MBL, mannan-binding lectin; EGF, epidermal mechanisms, but may also have significant clinical implications. growth factor; MASP, MBL-associated serine protease; wt, wild type; SNP, single We and others have defined a number of nonsynonymous poly- nucleotide polymorphism. morphisms in the gene encoding MASP-2. We have previously Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 identified and characterized the distribution of naturally occurring www.jimmunol.org/cgi/doi/10.4049/jimmunol.0802053 2940 ACTIVITIES OF MASP-2 VARIANTS

Table I. The primers used for site-directed mutagenesis

Mutants Primer

R99Qa Forward: 5Ј-GCACAGACACGGAGCAGGCCCCTGGCAAGGAC-3Ј Reverse: 5Ј-GTCCTTGCCAGGGGCCTGCTCCGTGTCTGTGC-3Ј P126L Forward: 5Ј-GACTACTCCAACGAGAAGCTGTTCACGGGGTTCGAG-3Ј Reverse: 5Ј-CTCGAACCCCGTGAACAGCTTCTCGTTGGAGTAGTC-3Ј H155R Forward: 5Ј-CCCACCTGCGACCACCGCTGCCACAACCACCTG-3Ј Reverse: 5Ј-CAGGTGGTTGTGGCAGCGGTGGTCGCAGGTGGG-3Ј CHNHdup Forward: 5Ј-GAAACCGCCCAGGTGGTTGTGGCAGTGGTTGTGGCAG-3Ј Reverse: 5Ј-CTGCCACAACCACTGCCACAACCACCTGGGCGGTTTC-3Ј V377A Forward: 5Ј-CTACCCAGTGGCCGAGCGGAGTACATCACAGGTC-3Ј Reverse: 5Ј-GACCTGTGATGTACTCCGCTCGGCCACTGGGTAG-3Ј R439H Forward: 5Ј-CTACCCAGTGGCCGAGCGGAGTACATCACAGGTC-3Ј Reverse: 5Ј-GACCTGTGATGTACTCCGCTCGGCCACTGGGTAG-3Ј

a Variant of rMASP-2.

missense polymorphisms of MASP-2 in different ethnic popula- change R444Q (Fig. 1). This is a change at the P1 position of the cleavage tions, i.e., Caucasians, Hong Kong Chinese, Sub-Saharan Africans, between the A chain and the B chain. This mutant cannot be cleaved and Downloaded from South-African Indians, and Inuits, and have observed associations is thus not activated. with circulating levels of MASP-2 (11). In the present study, we Production of rMASP-2 characterize the functional effects of the resulting naturally occur- Plasmids containing the MASP-2 cDNA inserts were used for the trans- ring variants of the MASP-2 protein through the production of the fection of HEK293 cells (Freestyle 293-F cells; Invitrogen). Briefly, plas- recombinant forms of the different MASP-2 allotypes. Most have mids (1 ␮g/ml) were mixed with lipofectamine 2000 (Invitrogen) and Opti- no effect on the function of the MBL pathway of complement MEM (Invitrogen), according to the manufacturer’s instructions, and used

6 http://www.jimmunol.org/ activation, but one blocks, not the synthesis, but the secretion of for transfection of early passage 293-F cells (30 ml of 10 cells/ml). Cells were cultured for 4 days in Freestyle expression medium (Invitrogen), and the variant, another binds to MBL as potently as the wild type (wt), supernatants were collected by centrifugation and stored at 4°C in the pres- yet fails to activate complement, and yet another shows higher ence of 0.01% sodium merthiolate. complement-activating capacity than the wt. The concentration of MASP-2 in the culture supernatants or in plasma samples (see below) was measured by a sandwich immunoas- say, as described in detail previously (13). In brief, the assay is based Materials and Methods on microtiter wells coated with anti-MASP-2 mAb. Samples diluted Generation of mutations in rMASP-2 10-fold in buffer releasing MASP-2 from complexes with MBL or fi- colins (MASP-2 assay buffer: 10 mM Tris, 1 M NaCl, 10 mM EDTA, We generated expression vectors encoding MASP-2 with the amino acid 7.5 mM NaN , and 0.05% Tween 20 (pH 7.4)) were incubated in wells changes R99Q, D120G, P126L, H155R, 156_159dupCHNH (in the fol- 3 by guest on September 26, 2021 coated with mAb 8B5 (anti-B chain), and after washing, a second, bi- lowing termed CHNHdup), V377A, and R439H. The positions of these otin-labeled anti-MASP-2 mAb (mAb 6G12, anti-N-terminal domains) amino acids are indicated in Fig. 1. was used as detecting Ab, followed by europium-labeled streptavidin We have previously described the construction of a pCI neo plasmid and reading by time-resolved fluorometry. Concentrations are read on with an insert of wt MASP-2 (12). Stratagene’s QuickChange II XL site- standard curves made from in-house standard plasma and calibrated directed mutagenesis kit (catalog 200521-5) was used for site-directed mu- against purified wt rMASP-2, and the test results are verified by in- tagenesis of the cDNA in this vector. PCR were performed with a mixture cluding three internal controls. containing 5 ␮lof10ϫ reaction buffer, 50 ng of template (wt MASP-2 in pCI neo plasmid), 125 ng of forward and reverse primers (Table I), 1 ␮lof Characterization of the rMASP-2s dNTP mix, 3 ␮l of Quick Solution reagent, 1 ␮l of PFU Ultra HF DNA polymerase, and water added to 50 ␮l. The conditions for the PCR were The culture supernatants were subjected to size exclusion chromatography first 95°C for 1 min, and then 32 cycles of the following: 95°C, 50 s; 60°C, ona10mmϫ 30-cm Superose 6 HR column (GE Healthcare). The run- ␮ 50 s; and 68°C, 8 min, ending with 68°C for 7 min. A total of 10 lofthe ning buffer was TBS (10 mM Tris, 145 mM NaCl, and 7.5 mM NaN3 (pH resulting product was loaded on a 1% agarose gel to verify the length of the 7.4)) with 1 mM CaCl2 and 0.01% (v/v) Emulphogen. The column was product. The rest of the product was digested for 1 h with 1 ␮lofDpnIat loaded with 200-␮l samples of supernatants, which had been concentrated 37°C. XL10 Gold cells (Stratagene) were transformed with the DpnI- 10-fold on Centricon concentration units (Amicon, Millipore). Fractions of treated DNA, following the manufacturer’s instructions, and 200 ␮lofthe 0.25 ml were collected in polystyrene microtiter plates (Nunc) previously mixture was plated on Luria-Bertani plates with 100 ␮g of ampicillin/ml. blocked by incubation with TBS containing 0.05% Tween 20 and washed Single colonies were picked and used for the preparation of plasmid using with water. MASP-2 in fractions was quantified after 10-fold (for wt, Miniprep (Qiagen). The primers for the site-directed mutagenesis were R99Q, D120G, P126L, V377H, and R439H) or 2-fold (for H155R and designed using the program primer X (http://bioinformatics.org/primerx/) CNHNdup) dilution in MASP-2 assay buffer. and the sequence of the MASP-2 gene (Pubmed NM 006610, GI 5729914). Samples of rMASP-2 culture supernatants were analyzed by Western To confirm the mutations, the inserts were sequenced. The R99Q, blotting. The electrophoresis was under reducing or nonreducing con- P126L, H155R, and CHNHdup mutations were identified by sequencing ditions, using Bis-Tris gels containing a 4–12% polyacrylamide gradi- with the primer MASP-2 MMR (GTACGACTTCGTCAAGCTGAG), ent (XT Criterion; Bio-Rad) run in XT-MOPS buffer. Proteins were whereas the V377A and R439H mutations were identified by sequenc- transferred onto a nitrocellulose membrane, and MASP-2 bands were ing with the primer Seq2 MASP-2 (GAGCTTCTGCAAGGT). Positive detected with the mouse anti-human MASP-2 mAb 1.3B7 (14), fol- clones were then further sequenced with the following primers for dou- lowed by HRP-conjugated polyclonal rabbit anti-mouse IgG Ab ble determination to verify the integrity of the sequence: pCI MASP-2 (PO260; DakoCytomation) as primary and secondary Abs, respectively. 7249 Fw (TAG AAG CTT TAT TGC GGT AGT TTA TCA), MASP-2 The blot was developed by ECL (SuperSignal West Dura Extended Rev (CTGGGCGGTTTC), Seq CUB-2 upper (GTTCAGTGTCATTCT Duration Substrate, 34075; Pierce), and the signal was detected by a GGACT), Seq CCP-2 lower (ACCTACAAAGCTGTGATTC), MASP-2 cooled charge-coupled device camera (Kodak Image station 1000). Due SP (GCCTGGTCTGAAGCTGTT), Seq SPD lower (TTCTAGATAGT to the low concentration of MASP-2 in the supernatants from H155R GAAACAGAG), and pCI MASP-2 2188 Rev (AAC TCA TCA ATG and CHNHdup, the proteins in these were concentrated by incubation TAT CTT ATC ATG TCT GCT C). for 15 min at room temperature of 400 ␮l of supernatant with 20 ␮lof We have previously made an expression plasmid containing the D120G PX5 beads (protein-binding beads from PATEOF), followed by cen- form of MASP-2 (12). As a negative control for activation of MASP-2, we trifugation and addition of SDS-PAGE sample buffer to the beads and used the product of a previously made plasmid encoding the amino acid loading of this mixture. Relative molecular sizes were interpolated from The Journal of Immunology 2941

Table II. Production of rMASP-2s IgG). After incubation for 2 h, wells were washed and developed with europium-labeled streptavidin, as above. Results are expressed relative to Mutant MASP-2 (ng/ml)a a standard curve obtained by applying dilutions of a standard serum, as previously described (15). R99Qb 8,721 P126L 730 D120G 10,383 Autoactivation of MASP-2 variants upon incubation with MBL H155R 75 on mannan surface CHNHdup 9 V377A 868 MASP-2 is activated by the cleavage of the 76-kDa polypeptide chain into R439H 707 a 52-kDa A chain and a 31-kDa B chain (9). This activation is largely Wild type 8,459 dependent on the presence of MBL or ficolins. The ability of the various a The concentrations of MASP-2 in the culture supernatants are given. The data MASP-2s to be activated was analyzed by Western blotting after activation are representative of two experiments. on a mannan surface. Microtiter plates were coated with 1 ␮g of mannan b Variant of rMASP-2. in 100 ␮l of coating buffer overnight at 4°C and blocked by incubation with human serum albumin at 1 mg/ml for 1 h, followed by wash in TBS con- taining 0.05% Tween 20. The MASP-2-containing culture supernatants curves constructed on the basis of marker proteins (Precision Plus All were diluted in barbital buffer to reach equal MASP-2 concentrations, and Blue; Bio-Rad). were further mixed with rMBL to approximate final concentrations of 0.5 To estimate the amount of intracellular MASP-2 cells transfected with ␮g of rMBL/ml and 0.35 ␮g of MASP-2/ml (except for the CHNHdup the vector containing CHNHdup or wt MASP-2, cDNA inserts were col- mutant, in which only very low amounts of MASP-2 were present in the

lected by centrifugation and washed twice with PBS. The cells were sub- supernatant), and added to the mannan-coated wells. A total of 12 wells Downloaded from sequently lysed by addition of PBS containing 1% Triton X-100 and pro- (100 ␮l/well) was used for each MASP-2 mutant. After incubation at 37°C tease inhibitor mixture (CompleteMini; Roche) to the cells, followed by for 2 h, the MBL/MASP complexes bound in the wells were collected, incubation for 5 min at 4°C. The lysates were centrifuged for 30 min, 4°C, as follows: the first well of the 12 wells representing each mutant was emp- at 10,000 ϫ g, and the supernatants were collected. The cell culture su- tied, and the bound protein was eluted by adding 120 ␮l of 24 mM Tris- pernatants and the detergent extracts were analyzed by Western blotting, as HCl, 4 M urea, 5% (v/v) glycerol, and 1.5% (w/v) SDS (pH 6.7). After described above. 10-min incubation, the content of the first well was transferred to the next,

Assay for the ability of MASP-2 variants to associate with MBL just emptied well and incubated for 10 min. This was repeated for the http://www.jimmunol.org/ remaining wells. The eluates were reduced with 0.06 M DTT and analyzed rMASP-2 variants were diluted serially in 10 mM Tris-HCl, 1 M NaCl, 5 by SDS-PAGE and Western blotting. The blot was incubated with anti- ␮ mM CaCl2, 100 g of human serum albumin/ml, and 0.05% Triton X-100 MASP-2/MAp19 Ab (mAb 1.3B7), followed by HRP-labeled rabbit anti- (pH 7.4), and added to an equal volume of rMBL (20 ng/ml of the same mouse Ig, as described above. buffer). The mixtures thus all contained 10 ng of MBL/ml, but contained We also attempted to activate the MASP-2 by incubating supernatants varying concentrations of MASP-2 supernatants. Duplicate samples of 100 with MBL and mannose-coated Toyopearl-Hw75 beads (onto which MBL ␮l were then transferred to mannan-coated microtiter wells and incubated binds), but the procedure described above was found much more efficient. overnight at 4°C. Wells were washed with TBS/Tween 20 containing 5

mM CaCl2, and incubated for 2 h with biotin-labeled anti-MASP-2 mAb (mAb 6G12). The wells were washed, and europium-labeled streptavidin, MASP-2 genotyping by guest on September 26, 2021 diluted in 10 mM Tris-HCl, 145 mM NaCl, and 25 ␮M EDTA (pH 7.4), was added. After incubation for 1 h, wells were washed and enhancement DNA was extracted from peripheral blood cells using the QIAamp DNA buffer was added, and the amount of europium in the wells was measured blood mini kit (Qiagen). A real-time TaqMan PCR technique using by time-resolved fluorometry. minor-groove-binder probes was used for screening the MASP-2 gene for the single nucleotide polymorphism (SNP), p.R439H, with a pre- Assay for the ability of MASP-2 variants to activate C4 upon designed/validated TaqMan genotyping assay (C_22273114_20; Ap- incubation with MBL plied Biosystems). DNA amplification was conducted in 25 ␮lofPCR Supernatants containing the rMASP-2 variants were diluted, mixed with containing 20 ng of DNA, 900 nM primers, 200 nM probes, and Taq- rMBL, and incubated in mannan-coated microtiter wells, as described Man Universal PCR Master Mix (Applied Biosystems) on a real-time above. After wash, the wells were incubated for 90 min at 37°C with 100 PCR instrument (ABI Prism 7000). The PCR profile was 2 min at 50°C, ␮l of complement component C4 at 2 ␮g/ml barbital buffer (4 mM sodium 10 min at 95°C, followed by 40 cycles of 15 s at 92°C and 1 min at

barbital, 0.14 M NaCl, 2 mM CaCl2, 1 mM MgCl2, and 7.5 mM NaN3 (pH 60°C. To determine genotypes, endpoint reading of the fluorescence 7.4)). The wells were washed, and a mixture of two biotin-labeled anti- generated during PCR amplification was done on the ABI Prism 7000 human C4 mAbs (162.2 and 162.1 from Bioporto) was added (the Abs using Sequence Detection System software version 2.3 (Applied were biotinylated in-house with 126 ␮g of biotin-normal human serum/mg Biosystems).

FIGURE 1. The polypeptide chain of MASP-2 with the variants produced in the present study indicated. MASP-2 is composed of an N-terminal CUB domain, followed by an EGF domain, a second CUB domain, two CCP domains (complement-control protein domains), an activation peptide, and a serine protease domain. The mature polypeptide chain of human MASP-2 is composed of 686 aa residues, including a 15-aa signal peptide. The numbering of the amino acids on the figure is referring to the protein, including the signal peptide. The amino acid changes introduced in the present study are indicated. When MASP-2 is activated, the polypeptide chain is cleaved at an arginine-isoleucine peptide bond (R444-I445) between the activation peptide and the remainder of the serine protease domain. The resulting two fragments, the A chain and B chain (the serine protease domain), are held together by a disulfide bond (between C434 in the activation peptide section and C552 in the serine protease domain, indicated on the figure by the line connecting the two dots). Some of the polymorphisms are in the SNP databases: P126L, rs56392418; H155R, rs2273343; R439H, rs12085877; V377A, rs2273346. The R99Q, D120G, and CHNHdup polymorphisms have been described in Lozano et al. (26), Stengaard-Petersen et al. (12), and Thiel et al. (11), respectively. The frequencies of the polymorphisms in various populations are described in Thiel et al. (11). 2942 ACTIVITIES OF MASP-2 VARIANTS

MASP-2 activity in plasma The ability of plasma to activate complement factor C4 via the MBL path- way was tested, as previously described (16). In brief, diluted serum sam- ples are incubated in mannan-coated microtiter wells to allow for the bind- ing of MBL-MASP complexes. This is performed in a buffer containing 1 M NaCl, allowing for the binding of MBL/MASP without activating the MASPs, and at the same time dissociating the C1 complex. After wash, purified human complement factor C4 is incubated in the wells at 37°C, allowing for activation of C4 and the covalent binding of C4b to the man- nan surface. The amount of bound C4b is detected with anti-C4 Abs, as described above. Results Production of rMASP-2s FIGURE 2. MASP-2 in culture supernatants and intracellularly. Re- To analyze the consequences of naturally occurring amino acid combinant wt MASP-2 and CHNHdup MASP-2 were expressed in substitutions in MASP-2, we produced mutated rMASP-2s repre- 293-F cells. The cells were spun down and lysed by Triton X-100. The culture supernatants (Supernatant) and the lysates (Lysate) were ana- senting such mutations. After transfection of human 293F cells, we lyzed by Western blotting, developing the blots with anti-MASP-2 Ab. measured the concentration of MASP-2 in the culture supernatants and found that the R99Q and the D120G mutants and the wt- ␮ MASP-2 were produced in similar (8–10 g/ml) amounts, Downloaded from http://www.jimmunol.org/

FIGURE 3. Gel permeation chromatog- by guest on September 26, 2021 raphy. Fractionation of 200 ␮l of 10-fold concentrated supernatant was performed on a Superose 6 column. The fractions were tested for MASP-2 content by time-resolved immunofluorometric assay. On the left y-axis is the concentration of MASP-2 in the fractions, whereas the right y-axis in A gives the absorption at 280 nm for the wt super- natant. The upper graph (A) shows the re- sults when analyzing supernatants contain- ing wt, R99Q, or D120G variants, and the lower graph (B) shows the results from su- pernatants containing P126L, CHNHdup, H155R, V377H, and R439H variants. Note the 10-fold difference in y-axis scale. The elution volume is given on the x-axis, and the elution positions of trypan blue (2000 kDa), thyroglobulin (669 kDa), ferritin (440 kDa), and catalase (232 kDa) are marked in A. The Journal of Immunology 2943 Downloaded from

FIGURE 4. Binding of MASP-2 variants to MBL. MBL and rMASP-2s were mixed and added to microtiter wells coated with mannan. The bound MASP-2 was detected by incubation with biotinylated anti-MASP-2 Ab, FIGURE 6. Activation of rMASP-2s analyzed by Western blotting. A, http://www.jimmunol.org/ followed by europium-labeled streptavidin. The concentration of MASP-2 Activation of rMASP-2 after various time periods. Recombinant wt or in the dilutions of the supernatants is given on the x-axis, and the signal R439H MASP-2 was mixed with rMBL, added to mannan-coated micro- (counts per second) on the y-axis. titer wells, and left at 37°C for the times indicated below the figure. The lane marked “Start” is the preparation before addition to the wells. The wells were subsequently washed, and MBL and MASP-2 were eluted from whereas the P126L, R439H, and V377A mutants were produced at the wells with SDS-PAGE sample buffer. The samples were then analyzed ϳ10-fold lower levels, and the H155R MASP-2 was ϳ100-fold by SDS-PAGE at reducing conditions, followed by Western blotting de- lower than that of wtMASP-2 (Table II). The level of the veloping with anti-MASP-2 Ab. On this and the following graph, the mi- CNHNdup mutant was close to the detection limit of the assay. gration of the molecular mass markers is given on the right side of the

figure in kDa. B, The various rMASP-2 variants (indicated above the lanes) by guest on September 26, 2021 were mixed with rMBL, added to mannan-coated microtiter wells, and left at 37°C. The upper part of the figure, marked “Start,” represents the prep- aration before addition to the wells, whereas the lower part, marked “Ac- tivated,” represents the mixtures left at 37°C. The wells were subsequently washed, and MBL and MASP-2 were eluted from the wells with SDS- PAGE sample buffer. The samples were then analyzed by SDS-PAGE Western blotting developing with anti-MASP-2 Ab.

The catching Ab in the assay used reacts with the serine protease domain and the developing Ab with the N-terminal domains of MASP-2; both regions are not directly influenced by the mutations introduced. The overall difference in the MASP-2 levels in the culture supernatants was also seen when analyzing by Western blotting (data not shown), indicating that the assay estimates all of the mutants correctly. The culture supernatants containing the mu- tant MASP-2s were analyzed by SDS-PAGE Western blotting, and all were found to contain MASP-2 running as a 76-kDa band both in the nonreduced state and under reducing conditions. Because the Ab used recognizes the A chain of MASP-2, this indicates that the MASP-2s are present in a nonactivated state, i.e., as pro-MASP-2. If the rMASP-2 were activated, we would have observed a 52-kDa band in reducing conditions due to cleavage into disulfide-linked A (52 kDa) and B (31 kDa) chains (Fig. 1). FIGURE 5. Deposition of C4 fragments by complexes of MASP-2 vari- ants with MBL. MBL and rMASP-2s were mixed and added to microtiter Intracellular localization of CHNHdup MASP-2 wells coated with mannan. After wash, the wells were incubated at 37°C with purified complement factor C4, and the amount of C4 fragments de- We found only very low amounts of MASP-2 in the culture posited was subsequently measured by incubation with biotinylated anti- supernatants from cells transfected with plasmid encoding C4 Abs, followed by europium-labeled streptavidin. The concentration of CHNHdup (see above). To study whether MASP-2 was present MASP-2 in dilutions of the supernatants is given on the x-axis, and the inside the cells, we lysed cells producing wt or CHNHdup signal (counts per second) on the y-axis. rMASP-2 and analyzed for the presence of MASP-2 in the lysate. 2944 ACTIVITIES OF MASP-2 VARIANTS

Table III. Frequency of R439H in native Zambians supernatants with rMASP-2s were mixed with a fixed amount of rMBL and incubated in mannan-coated microtiter wells, and Frequency bound MBL/MASP-2 complexes were detected by reaction with n ϭ 194 an anti-MASP-2 Ab. As seen in Fig. 4, increasing amounts of MASP-2 resulted in increasing MASP-2 signal. No major differ- Phenotype RR:a 82% (n ϭ 159) RH: 18% (n ϭ 35) ence in complex formation was seen between MBL and the dif- HH: (n ϭ 0) ferent MASP-2 mutants, except for the D120G and the CHNHdup Allele frequency R: 91% (353) mutants, which showed no binding to MBL. Due to the very low H: 9% (35) amounts of MASP-2 in the culture supernatant from the CHNHdup a R439 (R), H439 (H). expression, it was only possible to analyze the binding of this variant in the lower concentrations. MASP-2 was by SDS-PAGE Western blotting found to be present Effect of MASP-2 mutations on the ability to activate inside the cells expressing CHNHdup in amounts similar to what complement factor C4 was found inside cells transfected with the plasmid encoding wt When pro-MASP-2 is activated in a MBL/MASP-2 complex, MASP-2 (Fig. 2). The molecular mass of the rCHNHdup MASP-2 MASP-2 will efficiently cleave complement factor C4, generating was similar to the wt MASP-2 (Fig. 2), and the molecular mass of C4b and C4a. C4b binds covalently to nearby amino or hydroxyl the MASP-2 found in the supernatant of the wt MASP-2 was sim- groups, in this case to the mannan coated in the wells (19). The ilar to the MASP-2 found inside the cells. various rMASP-2s were mixed with MBL and added to mannan- Downloaded from coated microtiter wells. Complement factor C4 was subsequently Effect of mutations on size estimated by gel permeation added, and the ability of the various MASP-2s to induce deposition chromatography of C4 fragments on the surface was detected with anti-C4 Abs. The The proteins in the various culture supernatants were fractionated D120G and the CHNHdup variants were found not to mediate C4 according to size on a Superose 6 column with an isotonic column deposition in this setup in agreement with their apparent failure of buffer containing Ca2ϩ, followed by estimation of MASP-2 in the binding to MBL (see above) (Fig. 5). Except for R439H, all of the http://www.jimmunol.org/ fractions. All of the rMASP-2 variants and the wt rMASP-2 eluted MASP-2s that could bind to MBL were also found to be able to at the same volume (Fig. 3); however, the CNHNdup variant was activate C4 (Fig. 5). The H155R mutant seemed to be more effi- present in too low a concentration to give a signal in the analysis cient than the others in this capacity. This mutant was found to be of the fractions from the gel permeation chromatography (Fig. 3B). only marginally better than the others in binding to MBL (Fig. 4). In the calcium-containing buffer, the MASP-2s eluted at a position corresponding to a molecular mass of ϳ500 kDa. This seems to Activation of rMASP-2s indicate that the MASPs at this condition form rather larger com- When a MBL/MASP complex binds to an activating surface, pro- plexes than the reported homodimers (17, 18), but the rather elon- MASP-2 is known to autoactivate (20–22). When pro-MASP-2 is gated structure of the dimer suggested in those papers may influ- activated, the polypeptide chain is cleaved into two chains (52-kDa by guest on September 26, 2021 ence the correct estimation of size of the MASP-2s by this A and 31-kDa B chain) held together by a disulfide bond. To technique. analyze the ability to autoactivate, we mixed MBL with the rMASP-2 variants, incubated the mixture in mannan-coated wells, Effects of MASP-2 mutations on interaction with MBL and analyzed the bound material by SDS-PAGE under reducing Based on the similarity with wt MASP-2 on Western blot analysis conditions, followed by Western blotting. Initially, we tested the and on gel permeation chromatography, we assumed that the var- activation of MASP-2 after incubation for various lengths of time. ious rMASP-2s were present in a native conformation and we thus In Fig. 6A, one can see that some of the wt rMASP-2 is activated continued to study the functions of these. To investigate the ability after 40 min at 37°C, and most of the MASP-2 is activated after of the mutated MASP-2s to bind to MBL, dilutions of culture 120 min. In comparison, the R439H mutant, which binds to MBL,

FIGURE 7. MASP-2 in R439H heterozy- gotes. A, The concentration of MASP-2 in Zambians. A median of 203 ng/ml (159 tested) and 157 ng/ml (39 tested) is found in wt and R439H heterozygous individuals, re- spectively (given as a line on the figure). Us- ing Mann-Whitney rank sum test (because normality test failed), there is no statistically significant difference. B, The MBL pathway activity of serum from wt and heterozygotes with regard to p.R439H. Deposition of C4b on a mannan surface is given on the y-axis, and the MBL concentration in the sera is given on the x-axis. The Journal of Immunology 2945 but which does not mediate deposition of C4 (see above), does not MASPs are homodimers that circulate as zymogens in complex get activated even after 120 min of incubation (Fig. 6A). It appears with MBL or one of the three ficolins to become activated once the that the reason for the inability to activate C4 is to be found in the pattern recognition molecule binds to a target surface, such as a lack of ability to autoactivate. All of the other rMASP-2 variants bacterial cell. The serine protease domain of MASP-2 has a chy- were seen to be activated after 120 min (Fig. 6B), except for the motrypsin-like structure and has -like substrate specificity, R444Q mutant that we used as negative control. It is not possible cleaving after arginine residues (24). Activation leads to cleavage to see the bands from CHNHdup on the figure due to the very low of the MASP polypeptide chain, between an arginine-isoleucine level of MASP-2 in the supernatants. It is not possible to see a peptide bond at the N-terminal end of the serine protease domain, band from the D120G variant on the lower Western blot on Fig. 6B and thus, the C-terminal end of the activation peptide (Fig. 1). (Activated) because this variant does not bind to the MBL that is Activated MASP-2 is able to cleave C4 with very high efficiency, bound to the mannan-coated wells. Thus, no MASP-2 can be ex- with Km in the nanomolar range (25). pected to be bound and subsequently eluted from the wells. We and others have described a number of amino acid exchange Autoactivation of MASP-2 has previously been reported to be variants of MASP-2 (11, 26). Of the reported variants, we decided not dependent on the presence of MBL (20, 21). We retested this by to produce rMASP-2 representing R118C because we have not found incubating wt rMASP-2 with MBL and mannose-coated beads and this allele in any of the populations that we have studied (11). analyzed the activation of MASP-2 by Western blotting of reduced Other groups have used insect cell lines for the generation of samples. We found that MASP-2 mixed with beads did not lead to recombinant human and mouse MASP-2 (27, 28) or Chinese ham- any activated MASP-2. Mixing MASP-2 with MBL resulted in ster ovary cells for expression of rat MASP-2 (21). We chose to some activation of MASP-2, but the activation of MASP-2 was use the human endothelial kidney cell line 293 to get as close as Downloaded from much more pronounced when MASP-2, MBL, and mannose beads possible to natural production. It would appear even better to use were mixed (data not shown). hepatocytes, which are the primary source of MASP-2 in humans. However, endogenous MASP-2 production from the hepatocytes Frequency of the R439H mutation in Africans might then have distorted the results. To date, the R439H mutant is unique in its ability to bind MBL We found a decent (0.7–10 ␮g/ml) concentration of MASP-2 in without becoming activated after binding of the complex to a man- the culture supernatants of most of the cells transfected with plas- http://www.jimmunol.org/ nan surface (see above). We searched the SNP databases for the mids encoding the naturally occurring MASP-2 variants. For rea- frequency and ethnic distribution of this polymorphism, and found sons unknown to us, the H155R variant was found at a quite low that it was observed among Africans. To extend these data, we concentration in the culture supernatant, but we were able to obtain subsequently tested for the presence of this SNP in 194 Zambian sufficient material for the present studies. The variant CHNHdup Africans. As given in Table III, this SNP is clearly quite common was present in too low an amount in the culture supernatant to with a gene frequency of 9%. Among the 194 individuals, we did allow for many of the experiments, and could only be detected not find any to be homozygous, although it should in principle be after concentration of the supernatant. In contrast, we found that found in 1 of 124 individuals. Clearly, this is not a statistically the amount of MASP-2 inside CHNHdup plasmid-transfected by guest on September 26, 2021 significant observation. 293-F cells was equal to the amount of MASP-2 inside cells trans- fected with the wt plasmid. This indicates that the CHNH variant Concentration and activity of MASP-2 in R439H heterozygotes is misfolded and cannot be correctly processed and exported, but The MASP-2 levels in the 194 Zambian Africans were previ- rather is retained in the endoplasmatic reticulum by quality control ously determined (11). When we divided the individuals in mechanisms, retranslocated to the cytosol, and finally, degraded by wt/wt and wt/R439H individuals, we found no significant dif- the ubiquitin proteasome pathway (29). Hence, the incorrectly ference in MASP-2 levels (Fig. 7A). The activity of the MBL/ folded mutant MASP-2 does not accumulate intracellular, and we MASP complexes in heterozygous individuals was found to be see approximately equal amounts in lysates from cells producing similar to the wt individuals, i.e., in individuals with higher CHNHdup and wt rMASP-2. We have previously found that in- MBL levels (above 1 ␮g/ml), no apparent difference was seen dividuals heterozygous for the CHNHdup variant have lower con- in C4b-depositing activity (Fig. 7B). As seen in the figure, the centrations of MASP-2 in serum than individuals not possessing assay clearly depends on the MBL concentration, i.e., the more this allotype (11). We suggest that this is due to the lack of secre- MBL, the more C4b deposition. tion of the misfolded form of the protein. Apparently, the other normal gene is not able to fully compensate for this. As discussed Discussion below, this variant has also lost the ability to bind to MBL. This Activation of the complement system in response to infections by lack of complex formation would possibly lead to a faster clear- pathogens is an essential component of the immune defense (1, 2). ance from the blood. The stretch of amino acids, H157, N158, However, activation of the complement system is also needed for H159, and L160, was suggested by Gregory et al. (18) to be in- efficient removal of altered-self structures, e.g., dying cells or mu- volved in homodimerization of human MAp19. The variant tated cells (23). Tagging structures with complement factors may 156_159dupCHNH presents a duplication of 4 aa right in this area, occur via several types of recognition. The present study concen- and this could possibly disrupt dimerization of the proteins. We trates on the description of activities of the lectin pathway, which could not study whether this is true because too little material is a relatively newly discovered initiation pathway; however, it is could be collected from the culture supernatants. suggested to be the oldest pathway in evolutionary terms (4). The various other forms of rMASP-2 unexpectedly all eluted The serine protease MASP-2 mediates the activation of down- corresponding to macromolecules of ϳ500 kDa when analyzed by stream complement components through the four known pattern size permeation chromatography in isotonic calcium-containing recognition molecules of the lectin pathway (3). Information on the buffer (Fig. 3). We do not know whether this indicates the forma- concentration and function of MASP-2 may thus be clinically rel- tion of higher oligomeric forms, or whether the MASPs somehow evant. A number of MASP-2 variants are found, and we wished to associate with other proteins in the culture supernatant. The mo- examine the function of these. lecular masses estimated by size-exclusion chromatography rely 2946 ACTIVITIES OF MASP-2 VARIANTS crucially on the relative shapes of the protein under study com- the interaction between MASP-2 polypeptide chains, and it could pared with the standard proteins used for calibration. This may add be that this interaction has increased due to the exchange of amino to this quite high apparent molecular mass we find because the acid residue. dimers of MASP-2 have been reported to be rather elongated (18). The R439H variant behaved strikingly different from the others, In an EDTA-containing buffer, the elution corresponds to the ex- displaying considerably reduced enzymatic activity despite bind- pected size of dimers (data not shown). With regard to the ing to MBL. We and others have previously noted that when pu- polypeptide chain of MASP-2, the CUB1-epidermal growth factor rified MASP-2 is mixed with purified MBL without any ligand (EGF)-CUB2 domains (CUB domain found in complement com- autoactivation of MASP-2 occurs (20, 21). This activation of ponent Clr/Cls, Uegf, and bone morphogenic protein 1) are in- MASP-2 is much more efficient when a ligand is present such as volved in a calcium-dependent homodimer formation (17, 18, 30). provided by a mannan surface, most likely a consequence of con- Together with the amino acids mentioned above, H155 also was formational rearrangements analogous to those of C1 complex ac- suggested to be involved in dimerization (18), but we do not see tivation. Whereas the wt MASP-2 and the other C4-cleaving vari- any effect of the H155R mutation (exchanging a basic residue with ants were capable of autoactivation, the R439H variant was found another basic residue) in the present investigation. incapable of autoactivating, providing a likely explanation for its In plasma, MASP-2 is found in complexes with MBL and with lack of C4-cleaving enzymatic activity (Fig. 6). The R439H mu- the three ficolins. MASP-2 binds to similar sites in the collagenous tation (exchanging a basic residue with another more bulky basic regions of the ficolins and MBL. A conserved lysine residue, i.e., residue) is positioned in the activation peptide, 5 aa N terminally K75 (numbering including signal peptide) in human MBL, K70 in from the R444 cleavage site (Fig. 1). This indicates that a polypep- human H-ficolin, K44 in human L-ficolin, and K73 in human M- tide sequence minimally including these 5 aa is needed for sub- Downloaded from ficolin, within this region is critical for binding and is believed to strate recognition, or alternatively, it may suggest that the folding form contacts with the MASPs (5, 31). Other amino acids nearby of the serine protease domain is influenced by the sequence of are of somewhat lesser importance, although they do influence the amino acids in the activation peptide. The structure of a CCP2- binding patterns of the different MASPs to a varying degree. With serine protease fragment and of a similar fragment in which the regard to the polypeptide chain of MASP-2, the first two domains active serine in the protease domain was mutated has been solved,

(CUB1-EGF) are involved in the interaction with MBL and fico- and the data indicate flexibility in the activation peptide between http://www.jimmunol.org/ lins, and the third domain (CUB2) has a stabilizing effect on this the CCP2 and the protease domain (22, 24). interaction. The x-ray crystallography studies of the CUB1-EGF- It has been suggested that MASP-2 not only binds to C4 through CUB2 domains from rat MASP-2 and human MAp19 suggest that the active protease domain, but also via parts of the two CCP the interaction between MASP-2 and MBL requires a calcium domains (CCP1-CCP2) next to the protease domain (22, 25), such present in the EGF domain as well as one present in an area often referred to as an exosite. Based on modeling exper- the CUB1 domain. We analyzed the ability of the various variant iments, this exosite has been suggested to include the amino acids MASP-2s to form complexes with MBL. We find that the two R376, E378, E397, and E398 in CCP2 (24). The probably very variants, R99Q and P126L, which are found in the CUB1 domain, small change in functional activity resulting from changing of va- and the variant H155R, which is found in the EGF domain, have line to alanine in position 377 (see above) did not seem to influ- by guest on September 26, 2021 retained the MBL-binding activity, whereas D120G (CUB1 do- ence this C4 binding site. main) and CHNHdup (EGF domain) (see Fig. 1) cannot associate Because it could possibly have a clinical consequence to be with MBL. The V377A variant (CCP2 domain) and the R439H homozygous for the R439H allele, and thus to be functionally de- variant (activation peptide) also bind well to MBL. We have pre- fect in the lectin pathway, we examined for the R439H variant in viously reported on the lack of binding of the D120G variant (12). black Zambians and found 18% to be heterozygous, i.e., a gene This is best explained by the loss of the calcium binding site in the frequency of 9%, similar to the frequency reported on rs12085877 CUB1 domain of MASP-2 by this mutant because D120 is directly in the National Center for Biotechnology Information SNP data- involved in binding of the calcium ion. Mutation of some nearby base for the allele in Sub-Saharan Africans. In the database, the residues also interfere with binding to MBL: when the amino acid allele was reported not to be present in 120 Europeans and 180 residue Y74 or Y121 (numbering including signal peptide) of Asians, whereas it was found in 30 (heterozygotes) (25%) of MAp19 was mutated to alanine, this led to no binding to MBL, and 120 samples from Sub-Saharan Africans, indicating a gene fre- the mutation of E98 or E124 to alanine resulted in very low bind- quency of 12.2% for the allele. This high frequency suggests ing activity, as examined by Gregory et al. (18), who suggested that it may be possible to identify individuals with defects in the that these amino acids are directly involved in the interaction with enzymatic activity of MASP-2. However, we have to date only MBL. No polymorphisms have been seen in these particular amino identified heterozygous individuals among the 194 individuals acids, but apparently the nearby mutation of R99Q (a basic amino tested. These heterozygous individuals show a normally func- acid residue exchanged to an amide) does not interfere with the tioning MBL/MASP-2 pathway (taking into account the differ- interaction of the neighboring E98 with MBL, and the mutation ence in MBL concentration of the different individuals) as mea- P126L (a secondary amine exchanged to an aliphatic group) does sured by the ability of their sera to deposit C4 fragments onto not interfere with the interaction mediated by E124. a mannan surface, mimicking a naturally occurring pathogen- The enzymatic activity of the various mutated MASP-2s was associated molecular pattern (Fig. 7B). Two other polymor- analyzed by studying the ability of the MASPs in complex with phisms seem to be restricted to Sub-Saharan Africans, p.R99Q MBL to induce the deposition of C4 fragments onto a mannan- and p.P126L (11). In the present study, we have had access to coated surface. Consistent with the finding that the MASP-2 mu- too few samples to be able to examine whether certain combi- tants, D120G and CHNHdup, cannot bind to MBL, these mutants nations of the various polymorphisms will influence the lectin failed to induce C4 fragment deposition. In contrast, the variants pathway. The distribution of MASP-2 levels was similar among R99Q, P126L, and V377A cleaved C4 with an activity comparable the heterozygous and the wt individuals (Fig. 7A). to that of the wt MASP-2. The H155R variant had the highest The finding that naturally occurring variant forms of MASP-2 C4-cleaving activity. We do not have an explanation for this, but differ in MBL-binding activity and enzymatic activity might have as mentioned, the amino acid at this position may be involved in implications for the susceptibility to infections of individuals with The Journal of Immunology 2947 the various genotypes. With regard to polymorphisms associated 13. Møller-Kristensen, M., J. C. Jensenius, L. Jensen, N. Thielens, V. Rossi, with deficiency of MASP-2, it is to be expected, as has been ex- G. Arlaud, and S. Thiel. 2003. Levels of mannan-binding lectin-associated serine protease-2 in healthy individuals. J. Immunol. Methods 282: 159–167. perienced with other parts of the immune system, that the conse- 14. Thiel, S., S. V. Petersen, T. Vorup-Jensen, M. Matsushita, T. Fujita, C. M. Stover, quences of impaired MASP-2 function may only become apparent W. Schwaeble, and J. C. Jensenius. 2000. Interaction of C1q and mannan-binding lectin (MBL) with C1r.C1s, MBL-associated serine proteases 1 and 2, and the if the individual encounters pathogens in situations in which other MBL-associated protein MAp19. J. Immunol. 165: 878–887. parts of the antimicrobial defense systems are stressed or lacking. 15. Petersen, S. V., S. Thiel, L. Jensen, R. Steffensen, and J. C. Jensenius. 2001. An The identification of a variant that shows higher complement-activat- assay for the mannan-binding lectin pathway of complement activation. J. Im- munol. Methods 257: 107–116. ing capacity suggests also that a potential dysregulation of comple- 16. Thiel, S., M. Møller-Kristensen, L. Jensen, and J. C. Jensenius. 2002. Assays for ment activation, possibly enhancing the harmful effects of the com- the functional activity of the mannan-binding lectin pathway of complement ac- plement system, may be the result of MASP-2 polymorphisms. tivation. Immunobiology 205: 446–454. 17. Chen, C. B., and J. Wallis. 2001. Stoichiometry of complexes between mannose- We believe that our results will further the understanding of the binding protein and its associated serine proteases: defining functional units for lectin pathway of complement activation and the clinical implica- complement activation. J. Biol. Chem. 276: 25894–25902. tions of deficiencies caused by nonfunctional and gain-of-function 18. Gregory, L. A., N. M. Thielens, M. Matsushita, R. Sorensen, G. J. Arlaud, J. C. Fontecilla-Camps, and C. Gaboriaud. 2004. The x-ray structure of human mutations. mannan-binding lectin-associated protein 19 (MAp19) and its interaction site with mannan-binding lectin and L-ficolin. J. Biol. Chem. 279: 29391–29397. Acknowledgments 19. Møller-Kristensen, M., S. Thiel, A. G. Hansen, and J. C. Jensenius. 2003. On the We are grateful for the expert technical help from Lisbeth Jensen. site of C4 deposition upon complement activation via the mannan-binding lectin pathway or the classical pathway. Scand. J. Immunol. 57: 556–561. Disclosures 20. Vorup-Jensen, T., S. V. Petersen, A. G. Hansen, K. Poulsen, W. Schwaeble, R. B. Sim, K. B. M. Reid, S. J. Davis, S. Thiel, and J. C. Jensenius. 2000. Distinct Downloaded from The authors have no financial conflict of interest. pathways of mannan-binding lectin (MBL)- and C1-complex autoactivation re- vealed by reconstitution of MBL with recombinant MBL-associated serine pro- References tease-2. J. Immunol. 165: 2093–2100. 21. Chen, C. B., and R. Wallis. 2004. Two mechanisms for mannose-binding protein 1. Walport, M. J. 2001. Complement. N. Engl. J. Med. 344: 1058–1066. modulation of the activity of its associated serine proteases. J. Biol. Chem. 279: 2. Markiewski, M. M., and J. D. Lambris. 2007. The role of complement in inflam- 26058–26065. matory diseases from behind the scenes into the spotlight. Am. J. Pathol. 171: 22. Gal, P., V. Harmat, A. Kocsis, T. Bian, L. Barna, G. Ambrus, B. Vegh, J. Balczer, 715–727. R. B. Sim, G. Naray-Szabo, and P. Zavodszky. 2005. A true autoactivating en- 3. Thiel, S. 2007. Complement activating soluble pattern recognition molecules zyme: structural insight into mannose-binding lectin-associated serine protease-2 http://www.jimmunol.org/ with collagen-like regions, mannan-binding lectin, ficolins and associated pro- activations. J. Biol. Chem. 280: 33435–33444. teins. Mol. Immunol. 44: 3875–3888. 23. Trouw, L. A., A. M. Blom, and P. Gasque. 2008. Role of complement and com- 4. Fujita, T. 2002. Evolution of the lectin-complement pathway and its role in innate plement regulators in the removal of apoptotic cells. Mol. Immunol. 45: immunity. Nat. Rev. Immunol. 2: 346–353. 1199–1207. 5. Wallis, R. 2007. Interactions between mannose-binding lectin and MASPs during 24. Harmat, V., P. Gal, J. Kardos, K. Szilagyi, G. Ambrus, B. Vegh, G. Naray-Szabo, complement activation by the lectin pathway. Immunobiology 212: 289–299. and P. Zavodszky. 2004. The structure of MBL-associated serine protease-2 re- 6. Stover, C. M., S. Thiel, M. Thelen, N. J. Lynch, T. Vorup-Jensen, J. C. Jensenius, veals that identical substrate specificities of C1s and MASP-2 are realized and W. J. Schwaeble. 1999. Two constituents of the initiation complex of the through different sets of enzyme-substrate interactions. J. Mol. Biol. 342: mannan-binding lectin activation pathway of complement are encoded by a single 1533–1546. structural gene. J. Immunol. 162: 3481–3490. 7. Matsushita, M., and T. Fujita. 1992. Activation of the classical complement path- 25. Rossi, V., F. Teillet, N. M. Thielens, I. Bally, and G. J. Arlaud. 2005. Functional

way by mannose-binding protein in association with a novel C1s-like serine characterization of complement proteases C1s/mannan-binding lectin-associated by guest on September 26, 2021 protease. J. Exp. Med. 176: 1497–1502. serine protease-2 (MASP-2) chimeras reveals the higher C4 recognition efficacy 8. Ambrus, G., P. Gal, M. Kojima, K. Szilagyi, J. Balczer, J. Antal, L. Graf, of the MASP-2 complement control protein modules. J. Biol. Chem. 280: A. Laich, B. E. Moffatt, W. Schwaeble, et al. 2003. Natural substrates and in- 41811–41818. hibitors of mannan-binding lectin-associated serine protease-1 and -2: a study on 26. Lozano, F., B. Sua´rez, A. Mun˜oz, J. C. Jensenius, J. Mensa, J. Vives, and recombinant catalytic fragments. J. Immunol. 170: 1374–1382. J. P. Horcajada. 2005. Novel MASP2 variants detected among North African and 9. Thiel, S., T. Vorup-Jensen, C. M. Stover, W. Schwaeble, S. B. Laursen, Sub-Saharan individuals. Tissue Antigens 66: 131–135. K. Poulsen, A. C. Willis, P. Eggleton, S. Hansen, U. Holmskov, et al. 1997. A 27. Rossi, V., S. Cseh, I. Bally, N. M. Thielens, J. C. Jensenius, and G. J. Arlaud. second serine protease associated with mannan-binding lectin that activates com- 2001. Substrate specificities of recombinant mannan-binding lectin-associated plement. Nature 386: 506–510. serine proteases-1 and -2. J. Biol. Chem. 276: 40880–40887. 10. Matsushita, M., S. Thiel, J. C. Jensenius, I. Terai, and T. Fujita. 2000. Proteolytic 28. Iwaki, D., and T. Fujita. 2005. Production and purification of recombinants of activities of two types of mannose-binding lectin-associated serine protease. mouse MASP-2 and sMAP. J. Endotoxin Res. 11: 47–50. J. Immunol. 165: 2637–2642. 29. Trombetta, E. S., and A. J. Parodi. 2003. Quality control and protein folding in 11. Thiel, S., R. Steffensen, I. J. Christensen, W. K. Ip, Y. L. Lau, I. J. Reason, the secretory pathway. Annu. Rev. Cell. Dev. Biol. 19: 649–676. H. Eiberg, M. Gadjeva, M. Ruseva, and J. C. Jensenius. 2007. Deficiency of 30. Feinberg, H., J. C. Uitdehaag, J. M. Davies, R. Wallis, K. Drickamer, and mannan-binding lectin associated serine protease-2 due to missense polymor- W. I. Weiss. 2003. Crystal structure of the CUB1-EGF-CUB2 region of mannose- phisms. Immun. 8: 154–163. binding protein associated serine protease-2. EMBO J. 22: 2348–2359. 12. Stengaard-Petersen, K., S. Thiel, M. Gadjeva, M. Moller-Kristensen, 31. Teillet, F., M. Lacroix, S. Thiel, D. Weilguny, T. Agger, G. J. Arlaud, and R. Sorensen, L. T. Jensen, A. G. Sjo¨holm, L. Fugger, and J. C. Jensenius. 2003. N. M. Thielens. 2007. Identification of the site of human mannan-binding lectin Inherited deficiency of mannan-binding lectin-associated serine protease 2. N. involved in the interaction with its partner serine proteases: the essential role of Engl. J. Med. 349: 554–560. Lys55. J. Immunol. 178: 5710–5716.