The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2007, 179: 455-462.
Copyright © 2007 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Girija, U. V.
Right arrow Articles by Wallis, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Girija, U. V.
Right arrow Articles by Wallis, R.

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 Ficolins1

Umakhanth Venkatraman Girija*, Alister W. Dodds*, Silke Roscher{dagger}, Kenneth B. M. Reid* and Russell Wallis2,*,{dagger},{ddagger}

* Department of Biochemistry, Medical Research Council Immunochemistry Unit, University of Oxford, Oxford, United Kingdom; and {dagger} Department of Infection, Immunity and Inflammation and {ddagger} Department of Biochemistry, University of Leicester, Leicester, United Kingdom


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Ficolins and mannose-binding lectins (MBLs) are the first components of the lectin branch of the complement system. They comprise N-terminal collagen-like domains and C-terminal pathogen-recognition domains (fibrinogen-like domains in ficolins and C-type carbohydrate-recognition domains in MBLs), which target surface-exposed N-acetyl groups or mannose-like sugars on microbial cell walls. Binding leads to activation of MBL-associated serine protease-2 (MASP-2) to initiate complement activation and pathogen neutralization. Recent studies have shown that MASP-2 binds to a short segment of the collagen-like domain of MBL. However, the interaction between ficolins and MASP-2 is relatively poorly understood. In this study, we show that the MASP-2 binding site on rat ficolin-A is also located within the collagen-like domain and encompasses a conserved motif that is present in both MBLs and ficolins. Characterization of this motif using site-directed mutagenesis reveals that a lysine residue in the X position of the Gly-X-Y collagen repeat, Lys56 in ficolin-A, which is present in all ficolins and MBLs known to activate complement, is essential for MASP-2 binding. Adjacent residues also make important contributions to binding as well as to MASP activation probably by stabilizing the local collagen helix. Equivalent binding sites and comparable activation kinetics of MASP-2 suggest that complement activation by ficolins and MBLs proceeds by analogous mechanisms.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The lectin pathway of complement provides a front-line defense within the immune system by neutralizing pathogens via an Ab-independent mechanism (1). The recognition components of this pathway, serum mannose-binding lectins (MBLs)3 and ficolins bind directly to surface-exposed carbohydrates and N-acetyl groups on pathogens and activate the complement cascade via MBL-associated serine protease-2 (MASP-2). Complement activation leads to host-mediated lysis and phagocytosis of pathogens. It also stimulates inflammatory and adaptive immune responses via arrays of complement receptors on host cells (2).

The lectin pathway is associated with health and disease on a number of levels. Point mutations in the human MBL gene lead to a common immunodeficiency, which is especially important when the adaptive immune system is immature or compromised, for example, during HIV infection or following chemotherapy (3, 4). The lectin pathway is also important in the pathogenesis of inflammatory disorders, such as cystic fibrosis and rheumatoid arthritis, where variant MBLs are associated with more severe disease (5). Although normally protective, inappropriate lectin pathway activation is sometimes associated with host damage. For example, activation upon reperfusion of tissues following ischemia causes considerable tissue damage (6, 7, 8), thereby, exacerbating conditions where blood supply is suboptimal or temporarily blocked.

MBLs and ficolins both consist of oligomers of trimeric subunits, each composed of a collagen-like domain linked to a cluster of three C-terminal pathogen-recognition domains (9, 10): C-type carbohydrate-recognition domains in MBLs and fibrinogen-like domains in ficolins. The carbohydrate-recognition domains of MBLs are joined to the collagen-like domain via an {alpha}-helical coiled coil neck which is absent in ficolins. Individual subunits in both families are linked together at the N-terminal ends of polypeptides via disulfide bonds and splay apart either at a break in the Gly-X-Y consensus repeat (called the kink), present in all mammalian MBLs and some ficolins, or at the junction between the short N-terminal domain and the collagen-like domain. Humans produce two serum ficolins designated L- and H-ficolins, together with a leukocyte-associated protein called M-ficolin, whereas rats, mice, and some other mammals produce only two ficolins, ficolins-A and -B, orthologs of human L- and M-ficolin, respectively.

Three MASPs (-1, -2, and -3) circulate with MBL and ficolins (11, 12, 13). However, MASP-2 alone is sufficient to activate the lectin pathway of complement. Once activated, it cleaves C4 and then C4b2 to generate the C3 convertase (C4b2a), the enzyme responsible for activating C3, which is the next component of the pathway (14). MASPs are homologs of C1r and C1s of the classical pathway of complement and comprise two CUB domains (domain found in complement component Clr/Cls, Uegf, and bone morphogenic protein 1) separated by a Ca2+-binding epidermal growth factor-like domain, followed by two complement-control protein modules and a serine protease domain. MASPs are homodimers that circulate as zymogens and only become activated once the lectin component binds to a target surface, such as a bacterial cell (15). Activation leads to cleavage of the MASP polypeptide chain, within a linker, at the N-terminal end of the serine protease domain. MASPs-1 and -2 autoactivate, whereas MASP-3 is activated by an unknown serum protease (16). The roles of MASP-1 and MASP-3 are not known. MASP-1 cleaves C2 but not C4, thus, it might enhance complement activation triggered by lectin.MASP-2 complexes, but cannot initiate activation itself (17). It also possesses thrombin-like activity; its major role might be within the clotting cascade (18).

Specific inhibitors of lectin pathway activation have important therapeutic potential for treatment of disorders involving transient ischemia, such as heart and kidney disease. Understanding the interactions of components of this pathway at the molecular level is therefore of great interest not only from a biochemical perspective but also as a basis for rational drug design. Our previous studies have shown that the MASP-binding site in rat MBL-A is located entirely within a portion of the collagenous domain, C-terminal to the kink (19) (Fig. 1). Specifically, one or both of Lys46 and Val47 are essential for MASP-binding and complement activation. These residues form part of a short sequence: OGKXGP (where O is hydroxyproline, and X an aliphatic residue other than glycine, or a methionine residue) that is present in almost all MBLs, and that we have termed the putative MASP-binding motif.


Figure 1
View larger version (79K):
[in this window]
[in a new window]

  		FIGURE 1. Aligned sequences of the collagenous domains of MBLs and ficolins. A, Domain organization of MBLs (top) and ficolins (bottom). The carbohydrate-recognition domain of MBL is labeled as CRD. All mammalian MBLs, but only certain ficolins have an interruption or kink within the Gly-X-Y repeat of the collagen-like domain. B, Collagen helical wheel of the putative MASP-binding motif. C, Aligned sequences of the collagen-like domains of MBLs and ficolins. Numbering of triplets is based on the sequence of rat MBL-A. The putative MASP-binding motif is shaded. Lysine residues within the Y position of the Gly-X-Y repeat are underlined. All such residues are at least partially hydroxylated and glycosylated in rat MBL-A and MBL-C (22 23 ). Most of the proline residues in the Y positions are represented as hydroxyproline (O) based on the sequences of rat MBL-A and MBL-C, in which all such residues are at least partially derivatized, except for the proline residue immediately preceding the kink, which is unmodified in each case.

 
Compared with MBLs, relatively little is known about the interactions between ficolins and MASPs. Alignment of the collagen-like domains reveals that sequences akin to the MASP-binding motif of MBLs are also present in the collagenous domains of most ficolins (Fig. 1). In this study, we demonstrate that a lysine residue (Lys56) within the conserved motif is essential for MASP-2 binding and probably mediates the specificity of the ficolin-MASP interaction. Mutations to adjacent residues within the motif also disrupt binding to different extents, indicating that these residues also play important roles in complement activation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Materials

DNA restriction and modifying enzymes were obtained from New England Biolabs or Roche. Tissue culture medium and protein molecular mass markers for SDS-PAGE were purchased from Invitrogen Life Technologies. Sepharose-6B, methotrexate, and N-acetyl glucosamine were obtained from Sigma-Aldrich. Oligonucleotide primers were purchased from Sigma-Aldrich and Eurogentech. Nickel-Sepharose 6 fast flow was purchased from GE Healthcare.

Cloning of rat ficolin-A

The cDNA of rat ficolin-A was amplified from a rat cDNA library using PCR (20). Forward and reverse oligonucleotide primers were: GAACTCGAGGCCACCATGTGGTGGCCTATGCTG and AGAGAATTCTTATGATGCCCGGATTTTCAT. Start and stop codons are underlined. The cDNA was cloned into the polylinker of mammalian expression plasmid pED-4 (21), which contains the dihydrofolate reductase gene as a selectable marker. The resulting plasmid was used to transfect the dihydrofolate reductase deficient, Chinese hamster ovary cell line DXB11, using the calcium phosphate precipitation method (22). We have shown previously that this expression system faithfully carries out essential posttranslational modifications of polypeptides during biosynthesis, including disulfide-bond formation, hydroxylation of proline, and hydroxylation and glycosylation of lysine residues within the collagen-like domain (22, 23). Protein expression was amplified using increasing concentrations of methotrexate (up to 0.5 µM), an inhibitor of dihydrofolate reductase, as described previously (22). Mutant forms of ficolin-A were created by introducing changes to the cDNA using a PCR-based site-directed mutagenesis strategy (20). Mutant cDNAs were cloned into pED-4 in the same way as for the wild-type cDNA.

Protein production and purification

Recombinant proteins were produced from Chinese hamster ovary cells in serum-free medium as described previously (23), and were purified by affinity chromatography on GlcNAc-Sepharose columns. In brief, culture medium (300 ml) containing ficolin-A was diluted with an equal volume of high-salt loading buffer (50 mM Tris-HCl (pH 7.5) containing 25 mM CaCl2 and 1.25 M NaCl) and loaded onto a GlcNAc-Sepharose column (1 ml) pre-equilibrated with high-salt loading buffer. After washes with of high- and low-salt loading buffer (10 ml each; 50 mM Tris-HCl (pH 7.5) containing 25 mM CaCl2 and 150 mM NaCl), protein was eluted with 50 mM Tris-HCl (pH 7.5) containing 150 mM NaCl and 300 mM GlcNAc. Before subsequent characterization, GlcNAc was removed by dialysis.

Catalytically active and inactive forms of rat MASP-2, called MASP-2K and MASP-2A, respectively, were produced by expression in Chinese hamster ovary cells and purified from serum-free medium as described (17, 24). MASP-2A is full-length MASP-2, in which the active site serine residue at position 613 is changed to an alanine. MASP-2A cannot autoactivate, so it is secreted in the zymogen form. MASP-2K is full-length MASP-2 in which the arginine residue at the cleavage site for zymogen activation (Arg424) is replaced by a lysine residue (17). This change reduces the rate of autocatalysis and thereby prevents activation of the zymogen during biosynthesis, secretion, and purification, allowing preparation of pure zymogen. Lectin-MASP-2K complexes autoactivate more slowly than wild-type complexes when bound to a carbohydrate-coated surface to generate catalytically active MASP.

MASP-2 activation kinetics

MASP-2K was incubated alone, with wild-type or mutant ficolin-A in the presence and absence of GlcNAc-Sepharose (5 µl of a 1:1 v/v suspension in a total volume of 30 µl), in 50 mM Tris-HCl (pH 7.5) containing 150 mM NaCl and 5 mM CaCl2, at 37°C with mixing. A 1.2-fold molar excess of ficolin was used to ensure that MASP-2K was fully bound. At various times, aliquots of the suspension were removed from the reaction mix, and proteins were separated by SDS-PAGE under reducing conditions. The amount of MASP activation was quantified from the amount of MASP-2 cleavage by scanning gels using a ChemiGenius (Syngene).

Gel filtration chromatography

Gel filtration chromatography was conducted on a Superdex 200 column (10 mm x 30 cm; GE Healthcare) equilibrated in 50 mM Tris-HCl (pH 7.5) containing 150 mM NaCl at a flow rate of 0.5 ml/min, at room temperature. Bovine thyroglobulin (8 nm), horse spleen apoferritin (5.9 nm), sweet potato beta-amylase (4.15 nm), yeast alcohol dehydrogenase (3.72 nm), BSA (3.52 nm), and bovine carbonic anhydrase (2.39 nm) were used as standards. Values in brackets are the Stoke’s radii of protein standards. The Stoke’s radii of ficolin-A oligomers were determined from a graph of [–log(Kav)]1/2 against the Stokes radius of protein standards, where Kav is the partition coefficient, Ve is the elution volume of the protein, and V0 and Vt are the void and total volumes of the gel filtration column, respectively.

Formula

Analytical ultracentrifugation

All experiments were conducted in a Beckman XL-A analytical ultracentrifuge equipped with absorbance optics using an An60Ti rotor. Before setting up the cells, ficolin-A was dialyzed overnight against 50 mM Tris-HCl (pH 7.5) containing 150 mM NaCl. Experiments were conducted at 40,000 rpm and at 20°C using aluminum centerpieces. Scans were collected at 2-min intervals at 230 nm. Data were analyzed by calculating the sedimentation coefficient distribution c(s), using the software package SEDFIT (25). Sedimentation coefficients are expressed as s20,w by correcting for the effects of buffers (26).

Determination of molecular mass of ficolin-A

The molecular masses (M) of ficolin-A oligomers were calculated from their sedimentation coefficients and Stoke’s radii (RS), where N is Avogadro’s number, {eta} and {rho} are the solvent viscosity and density and Formula is the partial specific volume of the protein.

Formula
The partial specific volume of ficolin-A (0.721 ml/g) was calculated from its amino acid composition as described (26).

Surface plasmon resonance

Measurements were performed using a BIAcore 2000 instrument (GE Healthcare). Protein ligands were diluted to 25 µg/ml in 10 mM sodium acetate (pH 4.0) and immobilized (4000–5000 response units) onto the carboxymethylated dextran surface of a CM5 sensor chip (GE Healthcare) using amine coupling chemistry (BIAcore amine coupling kit; GE Healthcare). Binding was measured in 10 mM HEPES (pH 7.4) containing 150 mM NaCl and 5 mM CaCl2 at a flow rate of 5 µl/min and at 25°C. After injection of 20 µl of ligand, the protein surface was regenerated by injection of 10 µl of 10 mM HEPES buffer (pH 7.4) containing 1 M NaCl and 5 mM EDTA. Data were analyzed by fitting association and dissociation curves to Langmuir binding models for several protein concentrations simultaneously, using BIAevaluation 4.1 software (GE Healthcare). Increasingly complex models were tested until a satisfactory fit to the data was achieved. Apparent equilibrium dissociation constants (KD) were calculated from the ratio of the dissociation and association rate constants (koff:kon).

Ficolin-MASP complex formation on GlcNAc-Sepharose

MASP-2A was incubated alone, with wild-type or mutant ficolin-A in the presence of GlcNAc-Sepharose (5 µl of a 1:1 v/v suspension in a total volume of 30 µl) in 10 mM HEPES (pH 7.4) containing 150 mM NaCl, 5 mM CaCl2, and 0.005% (v/v) surfactant P20. After incubation at 25°C for 15 min, the GlcNAc-Sepharose was pelleted and was washed five times using 300 µl of buffer prior to resuspension in gel loading buffer. Proteins were separated by SDS-PAGE. The amount of MASP and ficolin retained with the GlcNAc-Sepharose pellet was quantified by scanning their band intensities using a ChemiGenius.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Production and characterization of rat serum ficolin

To produce pure ficolin-A that was not contaminated by ficolin-B or other serum lectins, protein was produced using a well-characterized mammalian expression system and was purified by affinity chromatography on immobilized GlcNAc columns. Rat ficolin-A has not been characterized before, so to begin with we examined its structure and biochemical properties. Four separate species were detected using gel filtration chromatography (Fig. 2), which eluted from the column as one distinct peak and three partially overlapping peaks, implying that ficolin-A exists as multiple oligomeric forms. Four species were also detected by velocity sedimentation ultracentrifugation, enabling measurement of their sedimentation coefficients (Fig. 3). By comparing the shape-independent molecular mass of each oligomer, calculated from its Stokes radius (from gel filtration) and sedimentation coefficient (see equation in Materials and Methods), with the predicted masses of oligomers based on the ficolin-A amino acid sequence (calculated mass ~34 kDa), we concluded that ficolin-A consists of monomers, dimers, trimers, and tetramers of subunits (with three, six, nine, and twelve polypeptide chains, respectively). The proportion of oligomers in fresh protein preparations is shown in Table I. When reapplied to the gel filtration column, individual gel-filtration fractions eluted at their previous positions, revealing that oligomers are stable and do not dissociate into smaller forms upon dilution (data not shown). Further analysis of gel filtration fractions showed that most polypeptides in oligomers are linked together by disulfide bonds, but that the bonding pattern is heterogeneous (Fig. 2B). For example, fraction 1 (Fig. 2A) comprises mainly tetramers of subunits together with smaller amounts of trimers of subunits. However, a ladder of covalently linked polypeptides is observed on SDS-polyacrylamide gels under nonreducing conditions (Fig. 2B), ranging from 1 to 12 polypeptides, implying that although some tetramers are assembled from 12 disulfide-linked polypeptides, others must be formed from combinations of smaller covalently linked structures, associated through noncovalent interactions. Similar heterogeneity has previously been described in both native and recombinant ficolin preparations from a variety of species (27, 28), as well as in some MBLs, including rat MBL-A and -C (22, 23).


Figure 2
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 2. Gel filtration and SDS-PAGE of recombinant ficolin-A. A, Gel filtration elution profile of purified rat ficolin-A. Data were fitted to Gaussian curves, using the software Origin (MicroCal). The best fit was achieved with four curves, which are shown by dotted lines. Addition of extra parameters did not improve the fits significantly. B, SDS-polyacrylamide gel (4–12% linear gradient gel) under nonreducing conditions. Proteins were stained with Coomassie blue. Numbers above the gel correspond to elution fractions from the gel filtration column (A).

 

Figure 3
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 3. Velocity sedimentation ultracentrifugation of ficolin-A. A, Distribution of sedimenting species c(s) analyzed using the software package SEDFIT. Minor differences in the proportions of ficolin oligomers from those detected by gel filtration (in Fig. 2) are probably due to differences in the extinction coefficients of oligomers at 280 nm (gel filtration) and 230 nm (sedimentation). B, Example of data from five absorbance scans. Solid lines show the global fit to the data.

 

View this table:
[in this window]
[in a new window]

  		Table I. Biophysical properties of ficolin-A oligomers measured by gel filtration chromatography and velocity analytical ultracentrifugation

 
To measure the interaction between ficolin-A and MASP-2, ficolin was immobilized on a sensor chip and binding by soluble MASP was monitored, using surface plasmon resonance (Fig. 4). The binding kinetics were complex, probably reflecting the heterogeneous nature of ficolin-A. Nevertheless, reasonable fits were achieved using a two-complex, parallel reaction binding model in which apparent dissociation constants KD1 and KD2 were 5.0 ± 0.5 and 110 ± 10 nM, with association rate constants of 6.0 ± 0.2 and 0.35 ± 0.02 x 105 M–1 s–1 and dissociation rate constants of 3.0 ± 0.2 and 3.8 ± 0.1 x 10–3 s–1. As expected, binding between ficolin-A and MASP-2 was Ca2+-dependent with no binding detected in the presence of EDTA.


Figure 4
View larger version (13K):
[in this window]
[in a new window]

  		FIGURE 4. Binding of MASP-2 to immobilized ficolin-A analyzed by surface plasmon resonance. Comparable amounts of wild-type (A) and K56P, M57O ficolin-A (B) (4489 and 4331 response units, respectively) were immobilized on separate channels of the same sensor chip, and MASP-2A was injected at 0.04, 0.08, 0.16, and 0.33 µM. In A, dotted lines show the fit to a two-complex, parallel reaction-binding model and dashed lines show the global fit to a single component binding model. Residuals to the fits are shown in the panels below. The symbol to the left of the residual plots corresponds to 20 response units. The {chi}2 values were 2.54 and 22.1, respectively. Because ficolin-A consists of four different oligomers, more than two different ficolin-MASP complexes are probably formed on the sensor chip. Nevertheless, reasonable fits were achieved using the relatively simple parallel reaction-binding model, implying either that certain ficolin oligomers bind to MASP-2 only weakly or that different oligomeric forms bind to the MASP with comparable kinetics. It has been shown previously that trimers and tetramers of MBL subunits bind to MASP-2 with similar affinities, dimers bind ~2-fold more weakly whereas single subunits have very low affinities (24 ). By analogy, we suggest that the observed interactions probably reflect MASP-2 binding to immobilized ficolin dimers, trimers, and tetramers.

 
Although ficolins are known to activate MASPs when they bind to a suitable substrate, such as a bacterial cell wall or on immobilized GlcNAc, the mechanism of activation is poorly understood. To characterize this process in more detail, the kinetics of ficolin-MASP activation was measured by following autolysis of the MASP-2 polypeptide. As shown in Fig. 5, in the absence of an activating substrate, ficolin-MASP-2K complexes activate no faster than zymogen MASP-2K alone. However, upon addition of GlcNAc-Sepharose, the rate of MASP-2 cleavage increases markedly. Thus, ficolin-A must trigger complement activation by increasing the rate of MASP autocatalysis, but only when complexes bind to a substrate.


Figure 5
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 5. Kinetics of ficolin.MASP-2 activation analyzed by SDS-PAGE. A, SDS-polyacrylamide gel (4–12% linear gradient gel) of ficolin-MASP-2K complexes, incubated with GlcNAc-Sepharose. Proteins were separated under reducing conditions and were stained with Coomassie blue. The N-terminal fragment of MASP-2K runs as a double band due to differential glycosylation. Activation was measured by quantifying cleavage of the MASP polypeptide. B, Comparison of MASP-2K activation in wild-type and ficolin-A K56P, M57O complexes. There was no detectable difference in activation of zymogen MASP-2K in the presence or absence of GlcNAc-Sepharose.

 
Characterization of the MASP-binding site in ficolin-A

Although serum MBLs and ficolins are functionally analogous in several respects, the domains are different, with the exception of the collagen-like domain (Fig. 1). Because the C-terminal portion of the collagen-like domain of MBL mediates binding to MASP-2, we therefore focused on the equivalent portion of ficolin-A, which contains the motif: OGKXGP that is present in all MBLs and ficolins known to activate complement (Fig. 1). To test whether this sequence encompasses the MASP-binding site in ficolin-A, we produced a recombinant protein containing the double substitution K56P and M57O and compared its properties with those of the wild-type protein. The double mutant failed to bind to MASP-2A under any of the conditions examined (Fig. 4B). As expected from the binding data, it also failed to activate MASP-2 in the presence of GlcNAc-Sepharose (Fig. 5B). Loss of activity was not due to disruption of the collagen-like domain during biosynthesis, because the mutant resembled wild-type ficolin-A, both in terms of its oligomeric and its covalent structure (Fig. 6). Nor could it be explained by loss of carbohydrate binding, because the mutant still bound to GlcNAc-Sepharose, enabling purification by affinity chromatography using the same protocol as for the wild-type protein. Rather, the data suggest that Lys56 and/or Met57 bind to MASP-2, and that these interactions are essential for efficient MASP activation. In rat MBL, equivalent mutations (K46P and V47O) also abolished MASP-2 binding and complement activation (19), so the MASP recognizes comparable sequences within the collagen-like domains of rat ficolin-A and MBL-A.


Figure 6
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 6. Biophysical properties of ficolin-A mutants. A, Gel filtration of wild-type and mutant ficolins. B, Covalent structure of ficolin-A mutants by SDS-PAGE, under nonreducing conditions.

 
Detailed characterization of the putative MASP-binding motif has not been conducted previously in either MBLs or ficolins. To determine which residues contribute to MASP-2 binding, an additional set of site-directed mutants was created, in which individual residues were replaced throughout the conserved motif. Sequence alignment shows that Hyp54, Lys56, and Pro59 are strictly conserved in MBLs and in ficolins, therefore, each residue was changed to an alanine residue. With the exception of mouse ficolin-B and porcine ficolin-B, position 58 is occupied either by an aliphatic side chain or by a methionine residue. To examine the effects of replacing the hydrophobic side chain, typically found in this position with a polar or an acidic side chain, we changed Met58 to a serine residue or a glutamic acid residue. Interestingly, recombinant mouse ficolin-B, which naturally has a glutamic acid residue at the position equivalent to Met57 in rat ficolin-A, neither binds to MASP-2 nor activates complement (29).

The mutant ficolins were initially tested for MASP-2-binding (Table II). The K56A mutant failed to bind to the MASP when either the MASP or the ficolin was immobilized onto the sensor chip (Fig. 7 and Table II), despite having comparable oligomeric and covalent structures as wild-type ficolin-A (Fig. 6). As expected, it also failed to activate MASP-2 on the GlcNAc-Sepharose substrate (Fig. 8). Thus, Lys56 must be essential for efficient MASP-2 binding and complement activation, and the basic side chain of the lysine residue probably forms direct contacts with the MASP-2 polypeptide. All of the other ficolin mutants bound to MASP-2 to some extent (Fig. 7 and Table II) and the kinetic data were analyzed using the two complex, parallel reaction model that was used for analysis of wild-type complexes. MASP-2 binding was most impaired in the M57E mutant, in which the KD values were reduced by ~10- and ~20-fold. Interestingly, much of the differences in binding were due to slower association rates (5- and 8-fold slower compared with the wild-type interactions), implying that electrostatic interactions are important for MASP-2 binding by ficolin-A and that introduction of an acidic side chain near the binding site impairs binding significantly. In contrast, the M57S mutant bound to MASP-2 with similar affinity as wild-type ficolin.


View this table:
[in this window]
[in a new window]

  		Table II. Kinetic properties of MASP-2 interactions and MASP-2 activation by wild-type and mutant ficolins

 

Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 7. Binding of wild-type and mutant ficolins to MASP-2. A, Surface plasmon resonance of wild-type and mutant ficolins binding to immobilized MASP-2A. Ficolins (at 0.05 mg/ml) were injected sequentially on to the same sensor chip. B, Ficolin-MASP-2A complexes were pelleted on GlcNAc-Sepharose and separated by SDS-PAGE (4–12% linear gradient gel), under reducing conditions. Relative amounts of MASP-2 bound to the ficolin (compared with wild-type complexes) were: K56P, M57O—0.00; O54A—0.69; K56A—0.00; M57E—0.11; M57S—1.09; P59A—00.38, from two separate experiments. MASP marker is the starting amount of MASP-2A used in each experiment. MASP-no ficolin shows the amount of MASP-2A associated with GlcNAc-Sepharose in the absence of ficolin, demonstrating that MASP-2 only binds to GlcNAc-Sepharose through its interaction with ficolin-A.

 

Figure 8
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 8. MASP-2 activation by wild-type and mutant ficolins. Proteins were incubated with GlcNAc-Sepharose and were separated by gel electrophoresis.

 
Although ficolins and MASP complexes normally circulate as complexes, the key step in the activation process is when complexes bind to an activating surface. To reproduce this process, we examined ficolin-MASP complex formation on GlcNAc-Sepharose. As seen in Fig. 7B, MASP-2A only bound to GlcNAc-Sepharose through its interactions with ficolin-A. Moreover, the amount of MASP associated with each mutant ficolin correlated well with the affinity of the interactions measured by surface plasmon resonance (Table II), confirming that the kinetic measurements are a reliable measure of the strength of the interactions.

MASP-2 activation by the ficolin-A mutants was generally in line with their MASP-2 binding properties, suggesting that defective activation was a result of reduced binding. Thus, activation by the M57E mutant was reduced by ~8-fold, whereas P59A and O54A mutants were 3- and 2-fold lower than wild-type protein, respectively. These effects cannot be attributed to large conformational changes, because the gel filtration profiles and covalent structures of the proteins were normal (Fig. 6). Nevertheless, a Pro -> Ala substitution at the X position, or a Hyp -> Ala change at the Y position, typically lowers the stability of collagen peptides (30), so the phenotypes of the P59A and O54A mutants might be caused by local destabilization of the collagen helix.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Structures of ficolins

The biophysical data show that rat ficolin-A consists of mixtures of oligomers ranging from monomers to tetramers of subunits, each assembled from three identical polypeptide chains. The oligomeric structure is very similar to rat MBL-A, in which dimers, trimers, and tetramers are all capable of activating complement, but monomers have little or no activity (23). In MBL, the relationship between structure and activity can be explained because there are binding sites for two MBL subunits on each MASP-2 dimer (31), so monomers bind to MASPs only weakly and consequently are unlikely to undergo the conformational changes that lead to activation. Given the similarities between the structures of ficolins and MBLs and their MASP-binding sites, described here, it is likely that ficolin oligomers have similar properties to those of MBL oligomers.

Although human L-ficolin is usually described as a tetramer of subunits, smaller oligomers are also detected in serum preparations (32). It is worth noting that dimers, trimers, and tetramers of ficolin subunits have similar Stoke’s radii, due to their asymmetry (Table I), so are likely to coelute on gel filtration columns of low resolution. Furthermore, their sedimentation coefficients are also similar and are unlikely to be resolved using techniques such as gradient ultracentrifugation. Indeed, in this study, the different ficolin oligomers could only be identified from calculations of the Lamm equation solutions c(s), and not by other approaches for determining distributions, including g(s*) or ls-g*(s). For these reasons, it is likely that ficolin preparations previously described as pure tetramers might also contain some other oligomeric forms.

Equivalent binding sites within the collagenous domains of MBLs and ficolins

The data show that MASPs bind to equivalent sites within the central portions of the collagenous domains of ficolins and MBLs. In particular, a conserved lysine residue (Lys56 in ficolin-A) in the X position of the Gly-X-Y collagen repeat is critical for binding and probably forms key contacts with the MASP. Although other adjacent residues might also bind to MASP-2 through main chain contacts, few additional interactions are likely to mediate binding specificity, because elsewhere in the collagen-like domains the sequences are either different in MBLs and ficolins or have been shown to be unimportant for MASP binding. For example, residues N-terminal to the kink or in triplets 6 and 7 of rat MBL-A (Fig. 1), play only minor roles in MASP-2 binding (19). In addition, both the N- and C-terminal portions of the collagenous domains of human MBL and rat MBL-A and -C contain glycosylated-hydroxylysine residues that are likely to block protein-collagen interactions (19). Based on their sequences, derivatized lysine residues are likely to play a similar role in other MBLs and ficolins (Fig. 1).

In theory, Lys56 residues in more than one chain of the collagen triple helix could interact with the MASP. Such a binding mechanism is seen in the structure of a triple helical integrin-binding collagen peptide in complex with the integrin {alpha}2 I domain, in which phenylalanine and arginine residues in two separate chains of the helix within the sequence GFOGER, bind to the integrin (33). The three GFOGER strands adopt a unique orientation in the crystal structure of the complex, probably because of the one-residue stagger of polypeptides in the collagen helix. Interactions with residues in more than one chain, as seen in the integrin-peptide complex, would increase the binding affinity of the interaction, which involves a relatively short segment of collagen helix.

It is of interest to note that C1qs also contain sequences similar to the MASP-binding motif, at equivalent positions of all three polypeptide chains (19). For example, the A, B, and C chains of human C1q contain the sequences: OGKVGY, OGKVGP, and OGKNGP, respectively. These sites probably form part of the binding sites for the MASP homologs, C1r and C1s. In vivo, there is no apparent cross-reactivity between C1q and MASPs or between MBLs/ficolins and C1r/C1s, although interactions between MBL and C1r/C1s have been described in vitro (34). Differences in sequences and in protein architectures might both contribute to binding specificities of the recognition and activation components of lectin and classical pathways.

Mouse ficolin-B naturally contains a glutamic acid residue in the position equivalent to Met57 in rat ficolin-A (Fig. 1), and the recombinant mouse protein does not bind to MASP-2 or activate complement, despite otherwise having the required sequences for binding (29). Ficolin-B is relatively poorly characterized compared with its serum homolog, so absence of complement activity might be caused by structural differences other than those at the MASP-binding site. However, the data presented here suggest that the glutamate side chain in mouse ficolin-B, is likely to disrupt MASP binding and activation. From this perspective, it will be interesting to examine the activity of native rat ficolin-B, which has a methionine in the equivalent position, so it is predicted to bind to MASP-2 based on our proposed model. Most other MBLs and ficolins contain the full MASP-binding motif. Exceptions are porcine MBL-C and ficolin-A, in which the lysine residue is absent in each case. By contrast, porcine MBL-A contains the standard binding motif, whereas ficolin-B contains the similar sequence: VGKAGP. As far as we are aware, complement activation by porcine MBLs and ficolins has not been analyzed. However, based on their sequences, we would predict that these proteins either bind to MASP-2 differently from other members of the MBL or ficolin families or that they have lower complement fixing activities than their other mammalian homologs.


   Acknowledgment
 
We thank Wilhelm Schwaeble for providing the rat liver cDNA library.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
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 with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by Grant 077400 from The Wellcome Trust and Grant G0501425 from the Medical Research Council. The analytical ultracentrifugation facility at Oxford was set up using grants from the Biotechnology and Biological Sciences Research Council and The Wellcome Trust. R.W. is a Research Council U.K. Academic Fellow. Back

2 Address correspondence and reprint requests to Dr. Russell Wallis, Department of Infection, Immunity and Inflammation, Medical Science Building, University of Leicester, University Road, Leicester, U.K. E-mail address: rw73{at}le.ac.uk Back

3 Abbreviations used in this paper: MBL, mannose-binding lectin; MASP, MBL-associated serine protease. Back

Received for publication March 27, 2007. Accepted for publication April 30, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Turner, M. W.. 1996. Mannose-binding lectin: the pluripotent molecule of the innate immune system. Immunol. Today 17: 532-540. [Medline]
  2. Carroll, M. C.. 2004. The complement system in regulation of adaptive immunity. Nat. Immunol. 5: 981-986. [Medline]
  3. Garred, P., H. O. Madsen, U. Balslev, B. Hofmann, C. Pedersen, J. Gerstoft, A. Svejgaard. 1997. Susceptibility to HIV infection and progression of AIDS in relation to varient alleles of mannose-binding lectin. Lancet 349: 236-240. [Medline]
  4. Neth, O., I. Hann, M. W. Turner, N. J. Klein. 2001. Deficiency of mannose-binding lectin and burden of infection in children with malignancy: a prospective study. Lancet 358: 614-618. [Medline]
  5. Kilpatrick, D. C.. 2002. Mannan-binding lectin: clinical significance and applications. Biochim. Biophys. Acta 1572: 401-413. [Medline]
  6. Hart, M. L., K. A. Ceonzo, L. A. Shaffer, K. Takahashi, R. P. Rother, W. R. Reenstra, J. A. Buras, G. L. Stahl. 2005. Gastrointestinal ischemia-reperfusion injury is lectin complement pathway dependent without involving C1q. J. Immunol. 174: 6373-6380. [Abstract/Free Full Text]
  7. Moller-Kristensen, M., W. Wang, M. Ruseva, S. Thiel, S. Nielsen, K. Takahashi, L. Shi, A. Ezekowitz, J. C. Jensenius, M. Gadjeva. 2005. Mannan-binding lectin recognizes structures on ischaemic reperfused mouse kidneys and is implicated in tissue injury. Scand. J. Immunol. 61: 426-434. [Medline]
  8. Walsh, M. C., T. Bourcier, K. Takahashi, L. Shi, M. N. Busche, R. P. Rother, S. D. Solomon, R. A. Ezekowitz, G. L. Stahl. 2005. Mannose-binding lectin is a regulator of inflammation that accompanies myocardial ischemia and reperfusion injury. J. Immunol. 175: 541-546. [Abstract/Free Full Text]
  9. Fujita, T., M. Matsushita, Y. Endo. 2004. The lectin-complement pathway: its role in innate immunity and evolution. Immunol. Rev. 198: 185-202. [Medline]
  10. Garlatti, V., N. Belloy, L. Martin, M. Lacroix, M. Matsushita, Y. Endo, T. Fujita, J. C. Fontecilla-Camps, G. J. Arlaud, N. M. Thielens, C. Gaboriaud. 2007. Structural insights into the innate immune recognition specificities of L- and H-ficolins. EMBO J. 26: 623-633. [Medline]
  11. Dahl, M. D., S. Thiel, M. Matsushita, T. Fujita, A. C. Willis, T. Christensen, T. Vorup-Jensen, J. C. Jensenius. 2001. MASP-3 and its association with distinct complexes of the mannan-binding lectin complement activation pathway. Immunity 15: 127-135. [Medline]
  12. Sato, T., Y. Endo, M. Matsushita, T. Fujita. 1994. Molecular characterization of a novel serine protease involved in activation of the complement system by mannose-binding protein. Int. Immunol. 6: 665-669. [Abstract/Free Full Text]
  13. Thiel, S., T. Vorup-Jensen, C. M. Stover, W. Schwaeble, S. B. Laursen, K. Poulsen, A. C. Willis, P. Eggleton, S. Hansen, U. Holmskov, et al 1997. A second serine protease associated with mannan-binding lectin that activates complement. Nature 386: 506-510. [Medline]
  14. Wallis, R., A. W. Dodds, D. A. Mitchell, R. B. Sim, K. B. Reid, W. J. Schwaeble. 2007. Molecular interactions between MASP-2, C4, and C2 and their activation fragments leading to complement activation via the lectin pathway. J. Biol. Chem. 282: 7844-7851. [Abstract/Free Full Text]
  15. Wallis, R., R. B. Dodd. 2000. Interaction of mannose-binding protein with associated serine proteases: effects of naturally occurring mutations. J. Biol. Chem. 275: 30962-30969. [Abstract/Free Full Text]
  16. Zundel, S., S. Cseh, M. Lacroix, M. R. Dahl, M. Matsushita, J. P. Andrieu, W. J. Schwaeble, J. C. Jensenius, T. Fujita, G. J. Arlaud, N. M. Thielens. 2004. Characterization of recombinant mannan-binding lectin-associated serine protease (MASP)-3 suggests an activation mechanism different from that of MASP-1 and MASP-2. J. Immunol. 172: 4342-4350. [Abstract/Free Full Text]
  17. Chen, C. B., R. Wallis. 2004. Two mechanisms for mannose-binding protein modulation of the activity of its associated serine proteases. J. Biol. Chem. 279: 26058-26065. [Abstract/Free Full Text]
  18. Presanis, J. S., K. Hajela, G. Ambrus, P. Gal, R. B. Sim. 2004. Differential substrate and inhibitor profiles for human MASP-1 and MASP-2. Mol. Immunol. 40: 921-929. [Medline]
  19. Wallis, R., J. M. Shaw, J. Uitdehaag, C. B. Chen, D. Torgersen, K. Drickamer. 2004. Localization of the serine protease-binding sites in the collagen-like domain of mannose-binding protein: indirect effects of naturally occurring mutations on protease binding and activation. J. Biol. Chem. 279: 14065-14073. [Abstract/Free Full Text]
  20. Sambrook, J., E. F. Fritsch, T. Manatis. 1989. Computer aided interpretation of analytical sedimentation data proteins. Book Molecular Cloning: A Laboratory Manual 8.42-8.45. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
  21. Kaufman, R. J., M. V. Davies, L. C. Wasley, D. Michnick. 1991. Improved vectors for stable expression of foriegn genes in mammalian cells by use of the untranslated leader sequence from EMC virus. Nucleic Acids Res. 19: 4485-4490. [Abstract/Free Full Text]
  22. Wallis, R., K. Drickamer. 1997. Asymmetry adjacent to the collagen-like domain in rat liver mannose-binding protein. Biochem. J. 325: 391-400. [Medline]
  23. Wallis, R., K. Drickamer. 1999. Molecular determinants of oligomer formation and complement fixation in mannose-binding proteins. J. Biol. Chem. 274: 3580-3589. [Abstract/Free Full Text]
  24. Chen, C. B., R. Wallis. 2001. Stoichiometry of complexes between mannose-binding protein and its associated serine proteases: defining functional units for complement activation. J. Biol. Chem. 276: 25894-25902. [Abstract/Free Full Text]
  25. Brown, P. H., P. Schuck. 2006. Macromolecular size-and-shape distributions by sedimentation velocity analytical ultracentrifugation. Biophys. J. 90: 4651-4661. [Medline]
  26. Laue, T. M., D. S. Bhairavi, T. M. Ridgeway, S. L. Pelletier. 1992. S. E. Harding, and A. J. Rowe, and J. C. Horton, eds. Analytical Ultracentrifugation in Biochemistry and Polymers Science 90-125. Royal Society of Chemistry, Cambridge, U.K.
  27. Hummelshoj, T., N. M. Thielens, H. O. Madsen, G. J. Arlaud, R. B. Sim, P. Garred. 2007. Molecular organization of human ficolin-2. Mol. Immunol. 44: 401-411. [Medline]
  28. Ohashi, T., H. P. Erickson. 1998. Oligomeric structure and tissue distribution of ficolins from mouse, pig and human. Arch. Biochem. Biophys. 360: 223-232. [Medline]
  29. Endo, Y., N. Nakazawa, Y. Liu, D. Iwaki, M. Takahashi, T. Fujita, M. Nakata, M. Matsushita. 2005. Carbohydrate-binding specificities of mouse ficolin A, a splicing variant of ficolin A and ficolin B and their complex formation with MASP-2 and sMAP. Immunogenetics 57: 837-844. [Medline]
  30. Persikov, A. V., J. A. M. Ramshaw, A. Kirkpatrick, B. Brodsky. 2000. Amino acid propensities for the collagen triple helix. Biochemistry 39: 14960-14967. [Medline]
  31. Feinberg, H., J. C. Uitdehaag, J. M. Davies, R. Wallis, K. Drickamer, W. I. Weis. 2003. Crystal structure of the CUB1-EGF-CUB2 region of mannose-binding protein associated serine protease-2. EMBO J. 22: 2348-2359. [Medline]
  32. Krarup, A., S. Thiel, A. Hansen, T. Fujita, J. C. Jensenius. 2004. L-ficolin is a pattern recognition molecule specific for acetyl groups. J. Biol. Chem. 279: 47513-47519. [Abstract/Free Full Text]
  33. Emsley, J., C. G. Knight, R. W. Farndale, M. J. Barnes, R. C. Liddington. 2000. Structural basis of collagen recognition by integrin {alpha}2beta1. Cell 101: 47-56. [Medline]
  34. Lu, J., S. Thiel, H. Wiedemann, R. Timpl, K. B. M. Reid. 1990. Binding of the pentamer/hexamer forms of a mannan-binding protein to zymosan activates the proenzyme C1r2C1s2 complex of the classical pathway of complement, without involvement of C1q. J. Immunol. 144: 2287-2294. [Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Girija, U. V.
Right arrow Articles by Wallis, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Girija, U. V.
Right arrow Articles by Wallis, R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS