Variants of the human C1 inhibitor serpin domain containing three N-linked carbohydrates at positions 216, 231, and 330 (C1inhΔ97), a single carbohydrate at position 330 (C1inhΔ97DM), or no carbohydrate were produced in a baculovirus/insect cells system. An N-terminally His-tagged C1inhΔ97 variant was also produced. Removal of the oligosaccharide at position 330 dramatically decreased expression, precluding further analysis. All other variants were characterized chemically and shown to inhibit C1s activity and C1 activation in the same way as native C1 inhibitor. Likewise, they formed covalent complexes with C1s as shown by SDS-PAGE analysis. C1 inhibitor and its variants inhibited the ability of C1r-like protease to activate C1s, but did not form covalent complexes with this protease. The interaction of C1 inhibitor and its variants with heparin was investigated by surface plasmon resonance, yielding KD values of 16.7 × 10−8 M (C1 inhibitor), 2.3 × 10−8 M (C1inhΔ97), and 3.6 × 10−8 M (C1inhΔ97DM). C1s also bound to heparin, with lower affinity (KD = 108 × 10−8 M). Using the same technique, 50% inhibition of the binding of C1 inhibitor and C1s to heparin was achieved using heparin oligomers containing eight and six saccharide units, respectively. These values roughly correlate with the size of 10 saccharide units yielding half-maximal potentiation of the inhibition of C1s activity by C1 inhibitor, consistent with a “sandwich” mechanism. Using a thermal shift assay, heparin was shown to interact with the C1s serine protease domain and the C1 inhibitor serpin domain, increasing and decreasing their thermal stability, respectively.

The C1 inhibitor is the only regulator of the early proteases of the classical and lectin pathways of complement (C1r and C1s and mannan-binding lectin-associated serine protease [SP]-2, respectively) (13). Other physiologically relevant target proteases of C1 inhibitor include mannan-binding lectin-associated SP-1 (2, 4), plasma kallikrein and activated factor XII of the contact system (5, 6), and activated factor XI of the intrinsic coagulation system (7). More recently, reactivity toward the C1r-like protease (C1r-LP) has been reported (8). Over the past few years, several potential anti-inflammatory effects of C1 inhibitor, independent of its protease inhibitory activity, and involving noncovalent interactions with various proteins, cell surfaces, or lipids have been revealed (9).

C1 inhibitor is a heavily glycosylated single-chain glycoprotein of 478 aa belonging to the SP inhibitor (serpin) superfamily (10). It has a two-domain structure, with a classical C-terminal serpin domain and a nonconserved, exceptionally large (∼100 aa) N-terminal extension. Of the 13–20 potential O- and N-glycosylation sites identified in C1 inhibitor, 10–17 lie within this N-terminal domain, three N-linked oligosaccharides being attached to the serpin domain through residues Asn216, Asn231, and Asn330 (10) (Fig. 1). The N-terminal extension of C1 inhibitor does not seem to affect protease inhibition (11, 12), and its functional role remains elusive. However, interaction of C1 inhibitor with selectins on endothelial cells is known to be mediated by tetrasaccharides located in this area (9). Like other serpins, C1 inhibitor neutralizes its target proteases by presenting them a scissile peptide bond that matches their substrate specificity. This bond, identified as Arg444-Thr445, lies in the reactive center loop, an exposed segment with a stressed conformation located near the C-terminal end of the serpin domain. Reaction with C1 inhibitor results in the formation of a 1:1 covalent complex between the protease and the bulk of C1 inhibitor, observable by SDS-PAGE analysis (13). As for other serpins, the inhibitory activity of C1 inhibitor toward proteases, such as C1s or activated factor XI, can be greatly enhanced by heparin and other glycosaminoglycans (14), and the precise mechanism of this potentiation remains to be established.

FIGURE 1.

Schematic representation of the structure of human C1 inhibitor and of the C1inhΔ97 variants expressed in this study. The Cys101–Cys406 and Cys108–Cys183 disulfide bridges are represented. *,O-linked carbohydrates; , N-linked carbohydrates.

FIGURE 1.

Schematic representation of the structure of human C1 inhibitor and of the C1inhΔ97 variants expressed in this study. The Cys101–Cys406 and Cys108–Cys183 disulfide bridges are represented. *,O-linked carbohydrates; , N-linked carbohydrates.

Close modal

Several groups have attempted to produce rC1 inhibitor or its serpin domain using various heterologous expression systems. Functional full-length C1 inhibitor was produced by transient expression in COS-1 cells (15) and using a baculovirus/insect cell system (16). Active C1 inhibitor was also expressed at high yield in Pichia pastoris and purified (17). Various constructs corresponding to the serpin domain of C1 inhibitor were also produced in COS-1 cells (10), Escherichia coli (18), and recently in P. pastoris (19). Using the latter recombinant protein, the crystal structure of the C1 inhibitor serpin domain has been solved, revealing a latent conformation with the uncleaved reactive center loop inserted into a seven-stranded antiparallel β-sheet (19). Based on this structure, a model for interaction with heparin has been proposed.

In the current study, we used a baculovirus/insect cells system to express and purify different variants of the serpin domain of C1 inhibitor, with the aim to further investigate its reactivity with proteases, the implication of N-linked carbohydrates, and the interaction with heparin.

Diisopropyl phosphorofluoridate and N-acetylglycine-l-lysine methyl ester were obtained from Sigma-Aldrich (Saint-Quentin-Fallavier, France). Oligonucleotides were purchased from Eurogentec (Liège, Belgium). Restriction enzymes were from New England Biolabs (Beverly, MA). The pcDNA3C1inh plasmid containing the full-length human C1-inhibitor cDNA (Met458 allele) was provided by Véronique Frémeaux-Bacchi (Hôpital Européen Georges Pompidou). C1 inhibitor and activated C1s were purified from human plasma, according to published procedures (2, 20). The C1q recognition subunit of C1 and the proenzyme form of the C1s-C1r-C1r-C1s tetramer were purified from human plasma, as described previously (20, 21). Human recombinant proenzyme C1s and its complement control protein (CCP)2-SP segment were expressed in a baculovirus/insect cells system (22, 23). Human C1r-LP was expressed in CHO-K1 cells and purified as described previously (8). The concentration of purified proteins was determined using the absorption coefficient (A1%, 1cm at 280 nm) and m.w.: activated C1s, 14.5 and 79,800 (24, 25); proenzyme C1s and C1s Ser617Ala, 14.5 and 77,400 (22); C1s CCP2-SP segment, 16.4 and 36,700 (23); C1 inhibitor, 4.5 and 104,000 (10, 26); C1q, 6.8 and 459,300 (20); and C1s-C1r-C1r-C1s, 13.5 and 330,000 (21). For the rC1 inhibitor variants, absorption coefficients were calculated by the method of Gill and von Hippel (27), and m.w. was determined by mass spectrometry: C1inhΔ97, 6.6 and 46,228; C1inhΔ97-ht, 6.6 and 46,740; and C1inhΔ97DM, 6.2 and 43,824.

The C1s mutant Ser617Ala was expressed using a baculovirus/insect cells system, essentially as described previously for other C1s variants (22). Briefly, a DNA fragment encoding the human C1s signal peptide plus the mature protein was amplified by PCR and cloned into the pFastBac1 vector (Invitrogen, San Diego, CA). The expression plasmid coding for the Ser617Ala mutant was generated using the QuickChange XL site-directed mutagenesis kit (Stratagene, La Jolla, CA). The mutagenic oligonucleotides were designed according to the manufacturer’s recommendations, and the pFastBac1/C1s expression plasmid coding for wild-type C1s was used as a template. The recombinant baculovirus was generated using the Bac-to-Bac system (Invitrogen) and used to infect High Five cells. C1s Ser617Ala was purified from the cell culture supernatant using a single-step affinity chromatography on a Sepharose-C1s column, as described earlier (22).

The C1inhΔ97 construct encoding residues 98–478 of human C1 inhibitor was amplified by PCR using the VentR polymerase and the full-length pcDNA3C1inh plasmid. The sequence of the sense primer (5′-GGGCCCAGCCGGATCCCGGGTCCTTCTGCCCA-3′) introduced a BamHI restriction site (underlined) at the 5′ end of the PCR product and allowed in-frame cloning with the melittin signal peptide of the pNT-Bac expression vector (23). The antisense primer (5′-GACCCCAGGGCCTGAGAAGCTTAGGATCAGGT-3′) introduced a stop codon (bold type) followed by a HindIII site (underlined) at the 3′ end. The amplified PCR product was purified and cloned into the BamHI and HindIII restriction sites of pNT-Bac.

Plasmids coding for the double mutant C1inhΔ97DM lacking glycosylation at Asn216 and Asn231 and the triple mutant C1inhΔ97TM lacking glycosylation at Asn216, Asn231, and Asn330 were generated, using the QuickChange multisite-directed mutagenesis kit (Stratagene), by mutating the corresponding asparagine residues to glutamine.

A 6-His tag was inserted by site-directed mutagenesis in the pNT-BacC1inhΔ97 plasmid between the coding regions of the melittin signal peptide and of residue Gly98. Two complementary 86-bp primers (5′-GCCCTTGTTTTTATGGTCGTGTACATTTCTTACAT-3′ and 5′-CTATG-CGCATCACCATCACCATCACGGGTCCTTCTGCCCAGGACCTGTTAC-3′) containing the 6-His coding region (underlined) were used for this purpose. The QuickChange XL site-directed mutagenesis kit (Stratagene) was used, with a two-step amplification as recommended for large insertions.

The insect cells Spodoptera frugiperda (Sf21) and Trichoplusia ni (High Five) were routinely grown and maintained as described previously (28), and recombinant baculoviruses were generated using the Bac-to-Bac system (Invitrogen). The bacmid DNA was purified using the Qiagen midiprep purification system (Qiagen, Courtaboeuf, France) and used to transfect Sf21 insect cells with cellfectin in Sf900 II SFM medium (Invitrogen), as recommended by the manufacturer. For all C1 inhibitor variants, recombinant virus particles were collected 4 d later and amplified as described by King and Possee (29). High Five cells (1.75 × 107 cells/175 cm2 tissue culture flask) were infected with the recombinant viruses at a multiplicity of infection of 2 in Sf900 II SFM medium at 28°C for 72 h. The culture supernatants containing the C1 inhibitor constructs were collected by centrifugation, supplemented with 1 mM diisopropyl phosphorofluoridate, and stored frozen at −20°C until use.

The same procedure was used for all variants, except C1inhΔ97-ht. Culture supernatants (0.5 l) were dialyzed against 50 mM triethanolamine-HCl (pH 8.6) and loaded onto a Q-Sepharose Fast Flow column (2.8 × 10 cm) (GE Healthcare, Orsay, France) equilibrated in the same buffer. Elution was carried out by applying a 1.2-l linear gradient from 0–350 mM NaCl. Fractions containing the recombinant material were identified by Western blot analysis, dialyzed against 1.5 M (NH4)2SO4, 0.1 M Na2HPO4 (pH 7.4), and further purified by high-pressure hydrophobic interaction chromatography on a TSK-Phenyl 5PW column (Beckman Coulter, Fullerton, CA) equilibrated in the same buffer. Elution was carried out by decreasing the (NH4)2SO4 concentration from 1.5 M to 0 in 30 min. The purified proteins were dialyzed against 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4), concentrated to ∼1 mg/ml by ultrafiltration, and stored at −20°C.

The culture supernatant containing C1inhΔ97-ht was dialyzed against 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4) and loaded onto a 2-ml His-select HF-Nickel affinity gel column (Sigma-Aldrich). Elution was performed by stepwise addition of 6-ml buffer fractions containing increasing imidazole concentrations of 10, 30, 50, and 250 mM. Fractions containing C1inhΔ97-ht were identified by SDS-PAGE analysis and Coomassie blue staining, dialyzed against 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4), and stored at −20°C.

Activated C1s (40 pmol) and C1r-LP (73 pmol) were incubated with equimolar amounts of native C1 inhibitor and different truncated variants, as indicated, for 90 min at 37°C in 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4). Samples were analyzed by SDS-PAGE on 10% acrylamide gels under nonreducing conditions.

Inhibition of C1s esterolytic activity was measured by preincubating activated C1s (0.5 μM) with increasing concentrations of C1 inhibitor or its variants (0.1–1 μM) for 90 min at 37°C in 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4) and then measuring residual C1s activity using N-acetylglycine-l-lysine methyl ester, as described previously (21).

The association rate constant of the inhibition of C1s esterolytic activity was determined as follows. Plasma C1s (6 nM) was diluted in 150 mM NaCl, 5 mM Na phosphate (pH 7.4) containing 1 mg/ml OVA and incubated for various periods at 20°C with excess C1 inhibitor or C1inhΔ97 (120 nM) in the presence of a large excess of each heparin oligomer (26 μg/ml). The residual C1s esterolytic activity [E] was measured at 30°C, as described above. The pseudo first-order rate constant, kass, was calculated according to the equation kass = kobs/[I]0, where [I]0 is the initial inhibitor concentration and kobs is determined from the plot ln [E] = −kobs ⋅ t (30).

Inhibition of C1 activation was assessed by incubating the C1 complex (0.25 μM) reconstituted from equimolar amounts of C1q and the proenzyme C1s-C1r-C1r-C1s tetramer in the presence of increasing concentrations of C1 inhibitor or its variants (0.5–1.5 μM) for 90 min at 37°C in 145 mM NaCl, 50 mM triethanolamine-HCl, 1 mM CaCl2 (pH 7.4). The reaction mixtures were analyzed by SDS-PAGE under reducing conditions, and C1s activation was measured by Western blot using a polyclonal anti-C1s Ab, as described previously (31).

Inhibition of C1r-LP–mediated C1s activation was measured by preincubating C1r-LP (1 μM) with equimolar amounts of C1 inhibitor or C1inhΔ97 for 30 min at 37°C in 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4). Samples were then incubated with proenzyme C1s (2 μM) for 30 min at 30°C in the same buffer, and the extent of C1s activation was assessed by SDS-PAGE under reducing conditions.

Fifteen-kDa porcine mucosal heparin (Sigma-Aldrich) (10 g) was digested with heparinase I (Grampian Enzymes, Orkney, U.K.) (8 mU/ml) in 150 ml 0.1 mg/ml BSA, 2 mM CaCl2, 50 mM NaCl, and 5 mM Tris-HCl (pH 7.5) for 54 h at 25°C. The reaction was stopped by heating at 100°C for 5 min, and the digest was fractionated on a Bio-Gel P-10 column (4.4 × 150 cm) (Bio-Rad, Hercules, CA) equilibrated with 0.25 M NaCl and run at 1 ml/min. The eluted material, detected by absorbance at 232 nm, consisted of a graded series of oligomers, ranging from 2–18 saccharide units. Only the top fractions of each peak were pooled, and each pool was resubmitted to gel filtration. Samples were dialyzed against water and quantified.

Six-kDa heparin (Sigma-Aldrich) at 1 mM in PBS was reacted for 24 h at room temperature with 10 mM biotin-LC-hydrazine (Pierce, Rockford, IL), as described by Vivès et al. (32). The mixture was extensively dialyzed against water to remove excess biotin and freeze dried.

Analyses were performed on an IQ5 96-well real-time PCR instrument (Bio-Rad), as described previously (33). Proteins (23 μl, 7–26 μM) in 145 mM NaCl, 50 mM triethanolamine-HCl (pH 7.4), in the presence or absence of a 2:1 molar excess of 15-kDa porcine mucosal heparin, were mixed with 2 μl of 5000× SYPRO Orange (Molecular Probes, Eugene, OR) diluted 1:100 in water. Samples were heat-denatured from 20°C to 95°C at 1°C/min. Protein unfolding was monitored by recording changes in the fluorescence of SYPRO Orange. The inflection point of the curves was determined by plotting the first derivative, and the melting temperatures were assessed from the minima. The fluorescence of buffers was checked as a control.

Heparin-binding analyses were performed on a BIAcore 3000 instrument (GE Healthcare). Streptavidin (200 μg/ml in 10 mM Na acetate [pH 4.2]) was immobilized on a CM4 sensor chip (GE Healthcare), using the amine-coupling chemistry, in 10 mM HEPES 150 mM NaCl (pH 7.4) until a coupling level of 2500 resonance units (RUs) was reached. Biotinylated 6-kDa heparin was then captured on the streptavidin surface in 10 mM HEPES, 300 mM NaCl, 0.005% surfactant P20 (pH 7.4) to reach a final coupling value of 30–50 RUs. A flow cell with 2500 RU streptavidin was used as a negative control.

Heparin-binding analyses.

Native C1 inhibitor, C1inhΔ97, C1inhΔ97DM, and activated C1s (15 μl) were injected over the heparin surface at 10 μl/min in 50 mM triethanolamine-HCl, 145 mM NaCl, 0.005% surfactant P20 (pH 7.4) at six concentrations, ranging from 0.05–1 μM. Surfaces were regenerated with 15 μl 2 M NaCl, with additional injections of 5 μl 0.05% SDS and 10 μl 2 M NaCl for the C1inhΔ97 variants. Data were analyzed by global fitting to a 1:1 Langmuir binding model of the association and dissociation phases simultaneously, using BIAevaluation 3.1 software (GE Healthcare). The apparent equilibrium dissociation constants (KD) were calculated from the koff/kon ratios of the dissociation and association constants.

Competition by heparin oligomers.

C1 inhibitor or C1s (each 0.53 μM) was injected on the heparin sensor chip at 50 μl/min in 50 mM triethanolamine-HCl, 145 mM NaCl, 0.005% surfactant P20 (pH 7.4) in the presence of various heparin oligomers (2–20 saccharide units) at concentrations of 12 mg/l for C1 inhibitor and 30 mg/l for C1s. The number of RUs determined in the presence of each oligomer was compared with the value measured in the absence of heparin.

The interaction between C1inhΔ97 and C1s Ser617Ala was monitored using a BIAcore X instrument (GE Healthcare). C1s was immobilized on the surface of a CM5 sensor chip (GE Healthcare) using the amine-coupling chemistry. Binding of C1inhΔ97 was measured over 10,000 RUs immobilized C1s Ser617Ala, at a flow rate of 20 μl/min in 10 mM HEPES, 145 mM NaCl (pH 7.4), containing 0.005% surfactant P20. Each sample was injected over a surface with immobilized BSA for subtraction of the bulk refractive index background.

SDS-PAGE, N-terminal sequence, and MALDI mass spectrometry analyses were carried out as described previously (23, 24, 34). Western blot analysis was performed according to published procedures (23) using a polyclonal anti-C1 inhibitor antiserum.

The objective of this study was to further functionally characterize the C-terminal serpin domain of human C1 inhibitor. For this purpose, the C1inhΔ97 DNA construct coding for residues 98–478 of human C1 inhibitor and two variants lacking N-linked carbohydrates at positions 216, 231, and 330 (C1inhΔ97TM) or solely containing the glycosylation site at Asn330 (C1inhΔ97DM) were expressed using a baculovirus/insect cells system (Fig. 1). All C1 inhibitor variants were secreted in the culture supernatants, at concentrations ∼10 mg/l (C1inhΔ97 and C1inhΔ97-ht), 3 mg/l (C1inhΔ97DM), and 0.2 mg/l (C1inhΔ97TM). Thus, removal of the oligosaccharide linked to Asn330 resulted in a dramatic decrease of the expression yield, precluding purification of material in sufficient amounts to perform further analysis. N-terminal sequence analysis of the purified C1inhΔ97 variants yielded a major sequence Asp-Pro-Gly-Ser-Phe-(Cys)-Pro-Gly-Pro-Val… in all instances, corresponding to the segment Gly98–Val105 of human C1 inhibitor preceded by the two residues Asp-Pro expected to be added at the N terminus, due to insertion of the BamHI restriction site at the cDNA 5′ end (see 1Materials and Methods). Occasionally, a minor (<10%) sequence Gly-Ser-Phe-(Cys)-Pro-Gly-Pro-Val… was also observed, indicating some heterogeneity in the maturation process. Sequence analysis of the C1inhΔ97-ht variant yielded the expected six-His cluster followed by the Gly98-Ser-Phe-(Cys)-Pro-Gly-Pro-Val… sequence. Analysis of C1inhΔ97 and C1inhΔ97-ht by MALDI mass spectrometry consistently yielded rather homogeneous peaks centered on mass values of 46,228 ± 50 Da and 46,741 ± 50 Da, respectively, accounting for the polypeptide chains plus three high-mannose N-linked oligosaccharides with average deduced masses of 1,085–1,117 ± 50 Da. Similarly, analysis of the C1inhΔ97DM variant yielded a mean mass value of 43,824 ± 50 Da, accounting for the polypeptide chain plus the remaining oligosaccharide linked to Asn330 (deduced mass = 920 ± 50 Da).

Analysis of the rC1 inhibitor variants by SDS-PAGE revealed major bands with apparent m.w. ∼49,500 (C1inhΔ97-ht), 44,000 (C1inhΔ97), and 39,500 (C1inhΔ97DM) (Fig. 2). C1inhΔ97-ht and C1inhΔ97 were essentially pure as judged from SDS-PAGE analysis, whereas minor contaminants were still present in the purified C1inhΔ97DM preparation (Fig. 2A, lane4). As observed for full-length C1 inhibitor (Fig. 2A, lane5), all truncated variants reacted with active C1s to form SDS- and urea-stable complexes of apparent m.w. ∼120,000 (C1inhΔ97), 116,000 (C1inhΔ97DM), and 120,000 (C1inhΔ97-ht). Covalent complex formation was almost complete when native C1 inhibitor was allowed to react with an equimolar amount of C1s for 90 min at 37°C, but it was only partial for the recombinant variants under the same conditions, suggesting a decreased efficiency of the truncated forms. Nevertheless, as illustrated in the case of C1inhΔ97-ht, increasing the C1 inhibitor variant/C1s molar ratio to 2:1 readily increased formation of the complex in all instances (Fig. 2B, lane4).

FIGURE 2.

SDS-PAGE analysis of the rC1 inhibitor constructs and their ability to form covalent complexes with C1s and C1r-LP. A, Lane1, activated C1s; lane2, native C1 inhibitor; lane3, C1inhΔ97; lane4, C1inhΔ97DM; lanes57, native C1 inhibitor, C1inhΔ97, and C1inhΔ97DM incubated for 90 min at 37°C with equimolar amounts of activated C1s. B, Lane1, C1inhΔ97-ht; lane2, purified C1s; lanes3, 4, C1inhΔ97-ht incubated for 90 min at 37°C with activated C1s in a 1:1 and 2:1 molar ratio, respectively. C, Lane1, C1r-LP; lane2, native C1 inhibitor; lane3, C1inhΔ97; lanes4, 5, native C1 inhibitor and C1inhΔ97 incubated for 90 min at 37°C with equimolar amounts of C1r-LP. All samples were analyzed under nonreducing conditions. The molecular masses of standard proteins (expressed in kDa) are shown. Covalent complexes between C1s and the C1 inhibitor variants are indicated by arrows.

FIGURE 2.

SDS-PAGE analysis of the rC1 inhibitor constructs and their ability to form covalent complexes with C1s and C1r-LP. A, Lane1, activated C1s; lane2, native C1 inhibitor; lane3, C1inhΔ97; lane4, C1inhΔ97DM; lanes57, native C1 inhibitor, C1inhΔ97, and C1inhΔ97DM incubated for 90 min at 37°C with equimolar amounts of activated C1s. B, Lane1, C1inhΔ97-ht; lane2, purified C1s; lanes3, 4, C1inhΔ97-ht incubated for 90 min at 37°C with activated C1s in a 1:1 and 2:1 molar ratio, respectively. C, Lane1, C1r-LP; lane2, native C1 inhibitor; lane3, C1inhΔ97; lanes4, 5, native C1 inhibitor and C1inhΔ97 incubated for 90 min at 37°C with equimolar amounts of C1r-LP. All samples were analyzed under nonreducing conditions. The molecular masses of standard proteins (expressed in kDa) are shown. Covalent complexes between C1s and the C1 inhibitor variants are indicated by arrows.

Close modal

Using C1r-LP as a second target protease, we checked its ability to form stable complexes with native C1 inhibitor and C1inhΔ97. As illustrated in Fig. 2C, C1r-LP alone migrated as a diffuse band of apparent m.w. 73,000. No evidence for a band of higher m.w. was obtained after incubation of C1r-LP for 90 min at 37°C in the presence of equimolar amounts of full-length C1 inhibitor or the C1inhΔ97 variant, indicating that a stable complex was not formed in either case.

We next used surface plasmon resonance spectroscopy to test the ability of C1inhΔ97 to interact with C1s, using a C1s mutant in which the active site Ser617 was mutated to Ala to prevent covalent reaction with the inhibitor. As shown in Fig. 3, C1inhΔ97 bound to immobilized C1s Ser617Ala to form a relatively stable complex. Interaction was clearly dose dependent, but the data could not be fitted properly, precluding accurate determination of the kinetic parameters.

FIGURE 3.

Surface plasmon resonance analysis of the interaction of C1inhΔ97 with the C1s Ser617Ala mutant. C1s Ser617Ala was activated by C1r, immobilized on a CM5 sensor chip, and allowed to interact with soluble C1inhΔ97. The interaction curves shown correspond to C1inhΔ97 concentrations of 0.4, 0.5, 0.75, and 1 μM (from bottom to top).

FIGURE 3.

Surface plasmon resonance analysis of the interaction of C1inhΔ97 with the C1s Ser617Ala mutant. C1s Ser617Ala was activated by C1r, immobilized on a CM5 sensor chip, and allowed to interact with soluble C1inhΔ97. The interaction curves shown correspond to C1inhΔ97 concentrations of 0.4, 0.5, 0.75, and 1 μM (from bottom to top).

Close modal

The ability of C1 inhibitor variants to inhibit C1s enzymatic activity was next investigated using N-acetylglycine-l-lysine methyl ester as a synthetic substrate. As depicted in Fig. 4A, native C1 inhibitor and C1inhΔ97 each gradually titrated C1s esterolytic activity, and complete inhibition was reached in each instance at a inhibitor/C1s molar ratio of ∼1:1. Similar inhibition curves were obtained in the case of C1inhΔ97DM and C1inhΔ97-ht (data not shown).

FIGURE 4.

Inhibition of C1s and C1r-LP enzymatic activities. A, Inhibition of C1s esterolytic activity. Activated C1s was preincubated for 90 min at 37°C with increasing amounts of native C1 inhibitor (♦) or C1inhΔ97 (□), and its residual esterolytic activity was measured using N-acetylglycine-l-lysine methyl ester. B, Inhibition of C1r-LP–mediated C1s activation. C1r-LP was preincubated for 30 min at 37°C with equimolar amounts of native C1 inhibitor or C1inhΔ97, and its residual ability to activate proenzyme C1s was measured by SDS-PAGE, as described in 1Materials and Methods. C, SDS-PAGE analysis corresponding to the experiment shown in B. Lane 1, C1r-LP + C1s; lane2, C1r-LP + C1s + C1 inhibitor; lane3, C1r-LP + C1s + C1inhΔ97; and lane4–7, C1r-LP, proenzyme C1s, C1 inhibitor, and C1inhΔ97, respectively. All samples were analyzed under reducing conditions. The molecular masses of standard proteins (expressed in kDa) are shown.

FIGURE 4.

Inhibition of C1s and C1r-LP enzymatic activities. A, Inhibition of C1s esterolytic activity. Activated C1s was preincubated for 90 min at 37°C with increasing amounts of native C1 inhibitor (♦) or C1inhΔ97 (□), and its residual esterolytic activity was measured using N-acetylglycine-l-lysine methyl ester. B, Inhibition of C1r-LP–mediated C1s activation. C1r-LP was preincubated for 30 min at 37°C with equimolar amounts of native C1 inhibitor or C1inhΔ97, and its residual ability to activate proenzyme C1s was measured by SDS-PAGE, as described in 1Materials and Methods. C, SDS-PAGE analysis corresponding to the experiment shown in B. Lane 1, C1r-LP + C1s; lane2, C1r-LP + C1s + C1 inhibitor; lane3, C1r-LP + C1s + C1inhΔ97; and lane4–7, C1r-LP, proenzyme C1s, C1 inhibitor, and C1inhΔ97, respectively. All samples were analyzed under reducing conditions. The molecular masses of standard proteins (expressed in kDa) are shown.

Close modal

We next tested the ability of C1 inhibitor and its C1inhΔ97 variant to inhibit the proteolytic activity of C1r-LP toward proenzyme C1s. In line with previous findings (8), preincubation of C1r-LP with an equimolar amount of native C1 inhibitor nearly totally prevented its ability to activate proenzyme C1s (Fig. 4B, 4C). A similar result was obtained using C1inhΔ97 under the same conditions. Thus, despite the fact that C1 inhibitor or C1inhΔ97 did not form stable complexes with C1r-LP as revealed by SDS-PAGE analysis, both proteins readily inhibited C1r-LP proteolytic activity toward C1s. Whether C1inhΔ97DM also inhibited C1r-LP without forming a stable complex could not be determined because of the low amount of C1r-LP available.

To further functionally characterize the truncated C1 inhibitor variants, we checked their ability to prevent spontaneous C1 activation. In keeping with previous results (31), native C1 inhibitor inhibited C1 activation in a dose-dependent fashion, and maximal inhibitory effect was reached at a C1 inhibitor/C1 molar ratio of ∼2:1 (Fig. 5). Similar inhibition curves were obtained using C1inhΔ97 or C1inhΔ97DM, demonstrating that the serpin domain of C1 inhibitor fully retained the ability to prevent C1 activation and that removal of the carbohydrates at positions 216 and 231 had no impact on this property.

FIGURE 5.

Inhibition of C1 activation. The C1 complex reconstituted from equimolar amounts of C1q and the proenzyme C1s-C1r-C1r-C1s tetramer was incubated for 90 min at 37°C in the presence of increasing amounts of native C1 inhibitor (♦), C1inhΔ97 (□), or C1inhΔ97DM (○). The extent of C1s activation was measured by SDS-PAGE and Western blot analysis, as described in 1Materials and Methods.

FIGURE 5.

Inhibition of C1 activation. The C1 complex reconstituted from equimolar amounts of C1q and the proenzyme C1s-C1r-C1r-C1s tetramer was incubated for 90 min at 37°C in the presence of increasing amounts of native C1 inhibitor (♦), C1inhΔ97 (□), or C1inhΔ97DM (○). The extent of C1s activation was measured by SDS-PAGE and Western blot analysis, as described in 1Materials and Methods.

Close modal

The ability of C1 inhibitor and its variants to interact with heparin was initially assessed by surface plasmon resonance spectroscopy, using the avidin-biotin system to immobilize 6-kDa heparin onto the biosensor surface through its reducing end. As illustrated in Fig. 6A, native C1 inhibitor readily bound to immobilized heparin, with a KD of 16.7 × 10−8 M, a value in the same range as that determined previously by Caldwell et al. (35) under comparable experimental conditions. The truncated proteins C1inhΔ97 and C1inhΔ97DM each retained the ability to bind heparin, with even significantly lower KD values resulting, for the most part, from decreased koff values (Table I). Considering that proteins expressed in insect cells lack the terminal sialic acid characteristic of serum glycoproteins, this is expected to decrease their negative charge and may explain why C1inhΔ97 and C1inhΔ97DM each form heparin–inhibitor complexes with increased stability compared with purified C1 inhibitor. These latter data provided a clear indication that the serpin domain of C1 inhibitor contains the structural elements required for interaction with heparin. In comparative binding experiments, activated C1s also bound to immobilized heparin, although with a significantly higher KD (108 × 10−8 M), resulting essentially from an increased koff value, indicative of a much lower stability of the heparin–C1s complex.

FIGURE 6.

Interaction of C1 inhibitor and C1s with immobilized heparin. A, Surface plasmon resonance analysis of the interaction between native C1 inhibitor and immobilized heparin. Six-kDa biotinylated heparin was immobilized on the sensor chip, as described in 1Materials and Methods. Native C1 inhibitor was injected at six concentrations, ranging from 0.05–1 μM (from bottom to top). B, Inhibition of C1 inhibitor–heparin interaction by heparin oligomers. Native C1 inhibitor was allowed to bind to immobilized heparin in the presence of heparin oligomers of varying sizes, as indicated. The number of RUs determined in each case is expressed relative to the value measured in the absence of heparin. C, Inhibition of C1s–heparin interaction by heparin oligomers. Active C1s was allowed to bind to immobilized heparin in the presence of heparin oligomers of varying sizes, as indicated, and inhibition was determined as in B. D, Potentiation by heparin oligomers of the inhibition of C1s activity by native C1 inhibitor. The association rate constant of the inhibition of C1s esterolytic activity by full-length C1 inhibitor was measured in the presence of the different heparin oligomers, as described in 1Materials and Methods.

FIGURE 6.

Interaction of C1 inhibitor and C1s with immobilized heparin. A, Surface plasmon resonance analysis of the interaction between native C1 inhibitor and immobilized heparin. Six-kDa biotinylated heparin was immobilized on the sensor chip, as described in 1Materials and Methods. Native C1 inhibitor was injected at six concentrations, ranging from 0.05–1 μM (from bottom to top). B, Inhibition of C1 inhibitor–heparin interaction by heparin oligomers. Native C1 inhibitor was allowed to bind to immobilized heparin in the presence of heparin oligomers of varying sizes, as indicated. The number of RUs determined in each case is expressed relative to the value measured in the absence of heparin. C, Inhibition of C1s–heparin interaction by heparin oligomers. Active C1s was allowed to bind to immobilized heparin in the presence of heparin oligomers of varying sizes, as indicated, and inhibition was determined as in B. D, Potentiation by heparin oligomers of the inhibition of C1s activity by native C1 inhibitor. The association rate constant of the inhibition of C1s esterolytic activity by full-length C1 inhibitor was measured in the presence of the different heparin oligomers, as described in 1Materials and Methods.

Close modal
Table I.
Kinetic and dissociation constants for the interaction of C1 inhibitor, C1 inhibitor variants, and C1s with immobilized heparin
Proteinskon (M−1.s−1)koff (s−1)KD (M)a
C1 inhibitor 1.94 × 104 3.32 × 10−3 16.7 ± 2.7 × 10−8 
C1inhΔ97 3.75 × 104 0.87 × 10−3 2.31 ± 0.3 × 10−8 
C1inhΔ97DM 1.83 × 104 0.66 × 10−3 3.61 ± 0.4 × 10−8 
C1s 1.01 × 104 11.0 × 10−3 108 ± 10 × 10−8 
Proteinskon (M−1.s−1)koff (s−1)KD (M)a
C1 inhibitor 1.94 × 104 3.32 × 10−3 16.7 ± 2.7 × 10−8 
C1inhΔ97 3.75 × 104 0.87 × 10−3 2.31 ± 0.3 × 10−8 
C1inhΔ97DM 1.83 × 104 0.66 × 10−3 3.61 ± 0.4 × 10−8 
C1s 1.01 × 104 11.0 × 10−3 108 ± 10 × 10−8 
a

Mean values determined from three separate experiments.

To determine whether the size of heparin influenced its ability to interact with C1 inhibitor, native C1 inhibitor was allowed to bind to immobilized 6-kDa heparin in the presence of heparin oligomers of varying sizes. As illustrated in Fig. 6B, increasing the size of the heparin oligomers from 2 to 20 saccharide units progressively decreased binding of native C1 inhibitor, with 50% inhibition being achieved by oligomers containing eight saccharide units. A similar competition experiment performed using C1s yielded a comparable inhibition curve (Fig. 6C). However, in this case, half-maximal inhibition of the interaction could be achieved using heparin oligomers containing only six saccharide units.

We next investigated the effect of heparin size on the potentiation of the inhibition of C1s activity by C1 inhibitor. For this purpose, the association rate constant of the inhibition of C1s esterolytic activity by full-length C1 inhibitor and C1inhΔ97 was measured in the presence of the different heparin oligomers, as described in 1Materials and Methods. As illustrated for full-length C1 inhibitor, increasing the heparin oligomer size progressively increased the kass value and 50% potentiation was achieved using the oligomer containing 10 saccharide units (Fig. 6D). This value is slightly higher but comparable to those measured for the interaction of heparin with C1 inhibitor and C1s. Therefore, this result seems to be consistent with a “sandwich” mechanism rather than with a “bridging” mechanism (19); in the latter case, the oligomer size necessary to achieve half-maximal potentiation would be expected to be much higher. Similar potentiation curves were ob-tained using C1inhΔ97 instead of full-length C1 inhibitor (data not shown).

A thermal shift assay based on the use of a fluorescent probe was used to measure the thermal stability of C1 inhibitor and its truncated form C1inhΔ97-ht. As shown in Fig. 7A, native C1 inhibitor displayed a melting temperature of 51.5°C, and this value remained unchanged when analysis was performed in the presence of a 2:1 molar excess of heparin. C1inhΔ97-ht exhibited a similar melting temperature at ∼50.5°C (Fig. 7B). However, in contrast to native C1 inhibitor, this value was shifted to 47°C in the presence of heparin, indicating that interaction with the isolated serpin domain of C1 inhibitor significantly decreased its thermal stability. Interestingly, a minor negative peak was also observed at 66°C, possibly reflecting the presence of low amounts of a latent form of the serpin domain (19).

FIGURE 7.

Thermal stability of C1 inhibitor and C1s and the effect of interaction with heparin. A, Native C1 inhibitor. B, C1inhΔ97-ht. C, Active C1s. D, C1s CCP2-SP segment. Curves corresponding to analyses conducted in the absence of heparin and in the presence of a 2:1 molar excess of heparin are shown in gray and black, respectively.

FIGURE 7.

Thermal stability of C1 inhibitor and C1s and the effect of interaction with heparin. A, Native C1 inhibitor. B, C1inhΔ97-ht. C, Active C1s. D, C1s CCP2-SP segment. Curves corresponding to analyses conducted in the absence of heparin and in the presence of a 2:1 molar excess of heparin are shown in gray and black, respectively.

Close modal

Analysis of active C1s by the same method yielded two minima at 39°C and 49°C, which are in good agreement with the melting temperatures of 37°C and 49°C previously determined by differential scanning calorimetry (36), corresponding to the CUB1-EGF-CUB2 segment and the SP domain, respectively (Fig. 7C). The third thermal transition at 60°C, corresponding to the CCP1-CCP2 segment (36), could not be detected by the method used. The transition at 49°C was shifted to 52°C in the presence of heparin, indicating that heparin bound to the SP domain and increased its thermal stability. Consistent with this observation, the CCP2–SP segment of C1s only exhibited a transition at 47°C corresponding to the SP domain, and this value was shifted to 51°C in the presence of heparin (Fig. 7D). Using heparin oligomers, it was determined that full thermal stabilization of the SP domain could be achieved using oligomers containing 10 saccharide units, in agreement with the data obtained by surface plasmon resonance spectroscopy.

Different variants of the C1inhΔ97 construct corresponding to the C-terminal serpin domain of human C1 inhibitor were expressed using a baculovirus/insect cells system and purified to homogeneity. The expression yields ranged from 3–10 mg/l and, therefore, were comparable to that reported previously for the production of full-length C1 inhibitor using the same expression system (16). In keeping with the report by Wolff et al. (16) and in contrast with the highly heterogeneous material produced in P. pastoris (12, 19), the C1 inhibitor serpin domain expressed in the current study was homogeneous in terms of size, as shown by SDS-PAGE and MALDI mass spectrometry analyses, indicating relative homogeneity of the N-linked oligosaccharides. In this regard, the mass spectrometry analyses performed on the recombinant proteins are consistent with the occurrence of short, oligomannose carbohydrates comprising two N-acetylglucosamine residues and three to four mannose residues (calculated mass = 893–1055), in keeping with previous analyses (16). For the first time, we have used site-directed mutagenesis to selectively prevent N-linked glycosylation at Asn residues at positions 216, 231, and 330. This strategy reveals that, although mutation of the former two residues has little effect on protein production, removal of the oligosaccharide linked to Asn330 results in a dramatic decrease in the expression yield, suggesting that this particular carbohydrate may play a role in the folding and/or the stability of the serpin domain. Previous attempts to prevent overall glycosylation of rC1 inhibitor or its serpin domain were made using expression in a bacterial system (18) or in the presence of tunicamycin (11, 15). Although these preparations were found to retain protease inhibitory activity, inhibition of glycosylation was reported to significantly decrease the expression yield (15) or to yield a major fraction of the recombinant material in insoluble form (18). In contrast, it is noteworthy that a latent form of the C1 inhibitor serpin domain was crystallized after enzymatic removal of N-linked carbohydrates by Endoglycosidase H (19). Taken together, these observations suggest that glycosylation at Asn330 may be crucial for proper folding of the serpin domain, but it is not required to stabilize the folded conformation.

Using active C1s as a target protease, it was found that formation of a covalent protease–inhibitor complex was comparatively less efficient for the truncated variants than for the full-length protein. Nevertheless, the truncated form C1inhΔ97 was shown to inhibit C1s esterolytic activity as efficiently as native C1 inhibitor and with the expected 1:1 stoichiometry. We also provide evidence that, as demonstrated previously for native C1 inhibitor (8), C1inhΔ97 efficiently inhibits the C1s-cleaving activity of C1r-LP. In contrast, neither of these molecules had the ability to form a stable complex with this protease, as judged from SDS-PAGE analysis. The example of C1r-LP shows that efficient inhibition of a protease activity by C1 inhibitor or its serpin domain does not necessarily require formation of a covalent protease–inhibitor linkage (i.e., cleavage of the reactive Arg-Thr bond), but that formation of a transient, noncovalent complex may, in certain cases, be sufficient to fulfill this function. The data obtained using active C1s as a target are more difficult to interpret, considering that small, but significant, amounts of the covalent complex were formed upon reaction with C1inhΔ97. In contrast, our finding that C1inhΔ97 forms a transient complex with the inactive C1s Ser617Ala mutant does not imply that the reaction does not proceed to a covalent complex when active C1s is the target.

In keeping with previous functional investigations (11, 18), the experiments carried out in the current study using C1s and C1r-LP as target proteases provide further evidence that the serpin domain of C1 inhibitor contains the structural determinants required for mediating protease inhibitory activity and that the N-terminal extension of C1 inhibitor is not implicated in this function.

Among the potential roles considered for the N-terminal extension of C1 inhibitor was the possibility that it participates in the control of C1 activation (11). In this respect, again, our data demonstrate that the C1inhΔ97 and C1inhΔ97DM variants share the ability of native C1 inhibitor to prevent C1 activation, demonstrating that the serpin domain alone is sufficient to fulfill this function. As observed in the case of C1r-LP, it appears likely that the ability of C1 inhibitor to prevent C1 activation involves formation of a reversible complex between its serpin domain and the proenzyme catalytic site of C1r, thereby inhibiting spontaneous C1r activation in the absence of activator or in the presence of mild activators (37).

In line with previous analyses using the same technique (35), our surface plasmon resonance spectroscopy data confirm that C1 inhibitor binds to heparin with high affinity and provide evidence that interaction is mediated by the serpin domain. C1s was also shown to bind heparin through its SP domain, but with a substantially lower affinity, reflecting the formation of a less stable complex. Consistent with this finding, full inhibition of the interaction by heparin oligomers was achieved more readily in the case of C1s than in the case of C1 inhibitor. An additional and intriguing difference between C1s and C1 inhibitor is that interaction with heparin seems to increase and decrease their thermal stability, respectively. Although these latter observations remain to be explained in molecular terms, our data provide further support to the hypothesis that heparin provides a link between C1 inhibitor and certain of its target proteases. With respect to the mechanism of this interaction, our observation that the heparin oligomer size required to achieve half-maximal potentiation of the inhibition of C1s activity by C1 inhibitor is on the same order as those necessary to inhibit binding of C1 inhibitor and C1s to 6-kDa heparin (Fig. 6B–D) seems to favor a “sandwich,” rather than a “bridging,” model. This conclusion is in line with the mechanism proposed by Beinrohr et al. (19) and seems to be fully consistent with the known characteristics of the potentiation of C1 inhibitor by polyanions, as reviewed by those investigators (19). Attempts to identify the C1 inhibitor residues involved in heparin interaction, by performing proteolytic digestion of C1 inhibitor–heparin cross-linked complexes followed by N-terminal sequence analysis (38), were unsuccessful, suggesting implication of residues located in various areas of the C1 inhibitor three-dimensional structure.

In summary, the current study provides further evidence that the serpin domain of C1 inhibitor is self-sufficient in terms of protease inhibitory activity and demonstrates that this also applies to the regulation of C1 activation and to heparin binding. The N-terminal extension of C1 inhibitor has no implication in these activities, and its precise role remains to be investigated further.

We thank Elodie Loizel for technical assistance with the thermal shift assay, Jean-Pierre Andrieu for performing N-terminal sequence analyses, and Izabel Bérard for mass spectrometry measurements.

Disclosures The authors have no financial conflicts of interest.

This work was supported by the Commissariat à l'Energie Atomique, the Centre National de la Recherche Scientifique, and the Université Joseph Fourier (Grenoble, France).

Abbreviations used in this paper:

CCP

complement control protein

C1r-LP

C1r-like protease

RU

resonance unit

SP

serine protease.

1
Sim
R. B.
,
Reboul
A.
,
Arlaud
G. J.
,
Villiers
C. L.
,
Colomb
M. G.
.
1979
.
Interaction of 125I-labelled complement subcomponents C-1r and C-1s with protease inhibitors in plasma.
FEBS Lett.
97
:
111
115
.
2
Rossi
V.
,
Cseh
S.
,
Bally
I.
,
Thielens
N. M.
,
Jensenius
J. C.
,
Arlaud
G. J.
.
2001
.
Substrate specificities of recombinant mannan-binding lectin-associated serine proteases-1 and -2.
J. Biol. Chem.
276
:
40880
40887
.
3
Davis
A. E.
 3rd
,
Mejia
P.
,
Lu
F.
.
2008
.
Biological activities of C1 inhibitor.
Mol. Immunol.
45
:
4057
4063
.
4
Ambrus
G.
,
Gál
P.
,
Kojima
M.
,
Szilágyi
K.
,
Balczer
J.
,
Antal
J.
,
Gráf
L.
,
Laich
A.
,
Moffatt
B. E.
,
Schwaeble
W.
, et al
.
2003
.
Natural substrates and inhibitors of mannan-binding lectin-associated serine protease-1 and -2: a study on recombinant catalytic fragments.
J. Immunol.
170
:
1374
1382
.
5
Schapira
M.
,
Scott
C. F.
,
Colman
R. W.
.
1982
.
Contribution of plasma protease inhibitors to the inactivation of kallikrein in plasma.
J. Clin. Invest.
69
:
462
468
.
6
Pixley
R. A.
,
Schapira
M.
,
Colman
R. W.
.
1985
.
The regulation of human factor XIIa by plasma proteinase inhibitors.
J. Biol. Chem.
260
:
1723
1729
.
7
Wuillemin
W. A.
,
Minnema
M.
,
Meijers
J. C.
,
Roem
D.
,
Eerenberg
A. J.
,
Nuijens
J. H.
,
ten Cate
H.
,
Hack
C. E.
.
1995
.
Inactivation of factor XIa in human plasma assessed by measuring factor XIa-protease inhibitor complexes: major role for C1-inhibitor.
Blood
85
:
1517
1526
.
8
Ligoudistianou
C.
,
Xu
Y.
,
Garnier
G.
,
Circolo
A.
,
Volanakis
J. E.
.
2005
.
A novel human complement-related protein, C1r-like protease (C1r-LP), specifically cleaves pro-C1s.
Biochem. J.
387
:
165
173
.
9
Davis
A. E.
 3rd
,
Cai
S.
,
Liu
D.
.
2007
.
C1 inhibitor: biologic activities that are independent of protease inhibition.
Immunobiology
212
:
313
323
.
10
Bock
S. C.
,
Skriver
K.
,
Nielsen
E.
,
Thøgersen
H.-C.
,
Wiman
B.
,
Donaldson
V. H.
,
Eddy
R. L.
,
Marrinan
J.
,
Radziejewska
E.
,
Huber
R.
, et al
.
1986
.
Human C1 inhibitor: primary structure, cDNA cloning, and chromosomal localization.
Biochemistry
25
:
4292
4301
.
11
Coutinho
M.
,
Aulak
K. S.
,
Davis
A. E.
 3rd
.
1994
.
Functional analysis of the serpin domain of C1 inhibitor.
J. Immunol.
153
:
3648
3654
.
12
Bos
I. G.
,
Lubbers
Y. T.
,
Roem
D.
,
Abrahams
J. P.
,
Hack
C. E.
,
Eldering
E.
.
2003
.
The functional integrity of the serpin domain of C1-inhibitor depends on the unique N-terminal domain, as revealed by a pathological mutant.
J. Biol. Chem.
278
:
29463
29470
.
13
Salvesen
G. S.
,
Catanese
J. J.
,
Kress
L. F.
,
Travis
J.
.
1985
.
Primary structure of the reactive site of human C1-inhibitor.
J. Biol. Chem.
260
:
2432
2436
.
14
Wuillemin
W. A.
,
te Velthuis
H.
,
Lubbers
Y. T.
,
de Ruig
C. P.
,
Eldering
E.
,
Hack
C. E.
.
1997
.
Potentiation of C1 inhibitor by glycosaminoglycans: dextran sulfate species are effective inhibitors of in vitro complement activation in plasma.
J. Immunol.
159
:
1953
1960
.
15
Eldering
E.
,
Nuijens
J. H.
,
Hack
C. E.
.
1988
.
Expression of functional human C1 inhibitor in COS cells.
J. Biol. Chem.
263
:
11776
11779
.
16
Wolff
M. W.
,
Zhang
F.
,
Roberg
J. J.
,
Caldwell
E. E.
,
Kaul
P. R.
,
Serrahn
J. N.
,
Murhammer
D. W.
,
Linhardt
R. J.
,
Weiler
J. M.
.
2001
.
Expression of C1 esterase inhibitor by the baculovirus expression vector system: preparation, purification, and characterization.
Protein Expr. Purif.
22
:
414
421
.
17
Bos
I. G. A.
,
de Bruin
E. C.
,
Karuntu
Y. A.
,
Modderman
P. W.
,
Eldering
E.
,
Hack
C. E.
.
2003
.
Recombinant human C1-inhibitor produced in Pichia pastoris has the same inhibitory capacity as plasma C1-inhibitor.
Biochim. Biophys. Acta
1648
:
75
83
.
18
Lamark
T.
,
Ingebrigtsen
M.
,
Bjørnstad
C.
,
Melkko
T.
,
Mollnes
T. E.
,
Nielsen
E. W.
.
2001
.
Expression of active human C1 inhibitor serpin domain in Escherichia coli.
Protein Expr. Purif.
22
:
349
358
.
19
Beinrohr
L.
,
Harmat
V.
,
Dobó
J.
,
Lörincz
Z.
,
Gál
P.
,
Závodszky
P.
.
2007
.
C1 inhibitor serpin domain structure reveals the likely mechanism of heparin potentiation and conformational disease.
J. Biol. Chem.
282
:
21100
21109
.
20
Arlaud
G. J.
,
Sim
R. B.
,
Duplaa
A.-M.
,
Colomb
M. G.
.
1979
.
Differential elution of Clq, Clr and Cls from human Cl bound to immune aggregates. Use in the rapid purification of Cl subcomponents.
Mol. Immunol.
16
:
445
450
.
21
Arlaud
G. J.
,
Thielens
N. M.
.
1993
.
Human complement serine proteases C1r and C1s and their proenzymes.
Methods Enzymol.
223
:
61
82
.
22
Bally
I.
,
Rossi
V.
,
Thielens
N. M.
,
Gaboriaud
C.
,
Arlaud
G. J.
.
2005
.
Functional role of the linker between the complement control protein modules of complement protease C1s.
J. Immunol.
175
:
4536
4542
.
23
Rossi
V.
,
Bally
I.
,
Thielens
N. M.
,
Esser
A. F.
,
Arlaud
G. J.
.
1998
.
Baculovirus-mediated expression of truncated modular fragments from the catalytic region of human complement serine protease C1s. Evidence for the involvement of both complement control protein modules in the recognition of the C4 protein substrate.
J. Biol. Chem.
273
:
1232
1239
.
24
Thielens
N. M.
,
Aude
C. A.
,
Lacroix
M. B.
,
Gagnon
J.
,
Arlaud
G. J.
.
1990
.
Ca2+ binding properties and Ca2(+)-dependent interactions of the isolated NH2-terminal alpha fragments of human complement proteases C1-r and C1-s.
J. Biol. Chem.
265
:
14469
14475
.
25
Pétillot
Y.
,
Thibault
P.
,
Thielens
N. M.
,
Rossi
V.
,
Lacroix
M.
,
Coddeville
B.
,
Spik
G.
,
Schumaker
V. N.
,
Gagnon
J.
,
Arlaud
G. J.
.
1995
.
Analysis of the N-linked oligosaccharides of human C1s using electrospray ionisation mass spectrometry.
FEBS Lett.
358
:
323
328
.
26
Harpel
P. C.
1976
.
C1 inactivator.
Methods Enzymol.
45
:
751
750
.
27
Gill
S. C.
,
von Hippel
P. H.
.
1989
.
Calculation of protein extinction coefficients from amino acid sequence data.
Anal. Biochem.
182
:
319
326
.
28
Thielens
N. M.
,
Cseh
S.
,
Thiel
S.
,
Vorup-Jensen
T.
,
Rossi
V.
,
Jensenius
J. C.
,
Arlaud
G. J.
.
2001
.
Interaction properties of human mannan-binding lectin (MBL)-associated serine proteases-1 and -2, MBL-associated protein 19, and MBL.
J. Immunol.
166
:
5068
5077
.
29
King
L. A.
,
Possee
R. D.
.
1992
.
The Baculovirus Expression System: A Laboratory Guide.
Chapman and Hall, Ltd
,
London.
30
Nagase
H.
,
Salvesen
G.
.
1989
.
Inhibition of proteolytic enzymes
. In
Proteolytic Enzymes, a Practical Approach.
Beynon
R. J.
,
Bond
J. S.
, eds.
IRL Press
,
Oxford, U.K.
, p.
144
.
31
Tacnet-Delorme
P.
,
Chevallier
S.
,
Arlaud
G. J.
.
2001
.
Beta-amyloid fibrils activate the C1 complex of complement under physiological conditions: evidence for a binding site for A beta on the C1q globular regions.
J. Immunol.
167
:
6374
6381
.
32
Vivès
R. R.
,
Sadir
R.
,
Imberty
A.
,
Rencurosi
A.
,
Lortat-Jacob
H.
.
2002
.
A kinetics and modeling study of RANTES(9-68) binding to heparin reveals a mechanism of cooperative oligomerization.
Biochemistry
41
:
14779
14789
.
33
Attali
C.
,
Frolet
C.
,
Durmort
C.
,
Offant
J.
,
Vernet
T.
,
Di Guilmi
A.-M.
.
2008
.
Streptococcus pneumoniae choline-binding protein E interaction with plasminogen/plasmin stimulates migration across the extracellular matrix.
Infect. Immun.
76
:
466
476
.
34
Rossi
V.
,
Gaboriaud
C.
,
Lacroix
M.
,
Ulrich
J.
,
Fontecilla-Camps
J. C.
,
Gagnon
J.
,
Arlaud
G. J.
.
1995
.
Structure of the catalytic region of human complement protease C1s: study by chemical cross-linking and three-dimensional homology modeling.
Biochemistry
34
:
7311
7321
.
35
Caldwell
E. E.
,
Andreasen
A. M.
,
Blietz
M. A.
,
Serrahn
J. N.
,
VanderNoot
V.
,
Park
Y.
,
Yu
G.
,
Linhardt
R. J.
,
Weiler
J. M.
.
1999
.
Heparin binding and augmentation of C1 inhibitor activity.
Arch. Biochem. Biophys.
361
:
215
222
.
36
Medved
L. V.
,
Busby
T. F.
,
Ingham
K. C.
.
1989
.
Calorimetric investigation of the domain structure of human complement Cl-s: reversible unfolding of the short consensus repeat units.
Biochemistry
28
:
5408
5414
.
37
Ziccardi
R. J.
1982
.
A new role for C1-inhibitor in homeostasis: control of activation of the first complement of human complement.
J. Immunol.
128
:
2505
2508
.
38
Vivès
R. R.
,
Crublet
E.
,
Andrieu
J. P.
,
Gagnon
J.
,
Rousselle
P.
,
Lortat-Jacob
H.
.
2004
.
A novel strategy for defining critical amino acid residues involved in protein/glycosaminoglycan interactions.
J. Biol. Chem.
279
:
54327
54333
.