Journal of Immunological Methods 337 (2008) 49–54

Contents lists available at ScienceDirect

Journal of Immunological Methods

j ourna l homepage: www.e lsev ie r.com/ locate / j im

Research paper

Enzyme-independent, orientation-selective conjugation of whole humancomplement C3 to protein surfaces

Daniel A. Mitchell a,⁎, Rebecca Ilyas a, Alister W. Dodds b, Robert B. Sim b

a Clinical Sciences Research Institute, Warwick Medical School, University of Warwick, Coventry CV2 2DX, UKb MRC Immunochemistry Unit, Oxford OX1 3QU, UK

a r t i c l e i n f o

Abbreviation: Sulfo EMCS, sulfo-ε-maleimidocapro⁎ Corresponding author. Clinical Sciences Researc

Medical School, Coventry CV2 2DX, UK. Tel.: +44 2476968652.

E-mail address: D.Mitchell@warwick.ac.uk (D.A. M

0022-1759/$ – see front matter © 2008 Elsevier B.V.doi:10.1016/j.jim.2008.05.011

a b s t r a c t

Article history:Received 29 March 2008Accepted 14 May 2008Available online 9 June 2008

Complement C3 is a central component of the humoral immune system. Upon triggering of thecomplement cascade, proteolytic fragments of C3 mediate important processes such asopsonization and lymphocyte activation. C3 possesses an internal thioester that mediatescovalent attachment of proteolytically activated C3 to target surfaces. Treatment of native C3with methylamine cleaves the thioester bond and exposes a free sulfhydryl group at the target-binding face of the protein. Through the use of sulfhydryl-reactive heterobifunctional cross-linking and biotinylation reagents, we demonstrate the capacity to form stable, multimericwhole human C3-protein conjugates in a fashion reflecting the orientation of physiologically-activated C3. We speculate that this C3 conjugation strategy presents a route for targetingdendritic cells andmacrophages. In addition, manipulation of the thioester bond could enhancethe study of biological roles of C3 and related proteins such as C4, and also of transmissibleagents that exploit complement function such as prions.

© 2008 Elsevier B.V. All rights reserved.

Keywords:Complement, ThioesterAdjuvantChemical biology

1. Introduction

The complement system occupies a pivotal space withinthe mammalian immune system, participating within a verybroad range of immunological and inflammatory processes(Walport, 2001a,b). In addition, it forms multiple bridgesbetween innate and adaptive immunity. Of the manycirculating and membrane-bound proteins that comprisethe complement cascade and execute complement functions,C3 represents one of the most significant. A fundamentalbiochemical feature of C3 is the presence of an internalthioester bond, formed post-translationally, that becomesexposed and highly reactive to target surfaces when C3 iscleaved and activated (Gadjeva et al., 1998). Fragments of C3,generated during complement activation, directly mediatekey phenomena including opsonization, phagocytosis, che-moattraction, immune complex ligation, and lymphocyteactivation. Therefore, selective manipulation of C3 carries

yloxysuccinimide.h Institute, Warwick958695; fax: +44 2476

itchell).

All rights reserved.

profound therapeutic and research potential. Seminal workby Dempsey et al. (1996) demonstrated that conjugation ofthe C3 fragment C3d to antigen dramatically lowers thethreshold for cognate B-lymphocyte activation(Dempseyet al., 1996). Themechanism that underpins this phenomenoninvolves the interaction between C3d and complementreceptor 2 (CR2; CD21) in conjunction with the B cell antigenreceptor complex and CD19. This combination of receptorengagement drives powerful co-stimulation and synergisticsignalling motif clustering within the B-lymphocyte. At thebiotechnological level, this approach to immunomodulationutilizes fusion protein constructs incorporating repeatingnucleotide sequences encoding multiple C3d units. Whilstthese advances illustrate a fundamental role for complementin key biological events, and offer attractive strategies invaccine development, there are limitations regarding thescopewithinwhich this specific approach can be applied. Thisrelates to the fact that CR2 expression within the immunesystem is largely restricted to B lymphocytes and folliculardendritic cells (Zabel and Weis, 2001). Also, many keyantigens, such as lipopolysaccharides, are non-protein incomposition and thus not amenable to conventional recom-binant fusion technology.

Fig. 1. SDS-PAGE andWestern blot analysis of C3-BSA conjugates. All sampleswere run under non-reducing conditions. Prominent high molecular weightspecies visualised by Coomassie staining (Panel A) are observed only insamples within which active sulfo-EMCS cross-linker was used in theconjugation protocol (Indicated as BSA XL C3). These species do not exist inan equivalent mixture of C3 and BSA incubated with inactive sulfo-EMCS(indicated as BSA+C3). Western blotting (Panel B) clearly shows higher orderC3 immunoreactive species greater than 180 kDa in the C3-BSA mixtureexposed to the active sulfo-EMCS protocol.

50 D.A. Mitchell et al. / Journal of Immunological Methods 337 (2008) 49–54

In addition to CR2, other receptors for C3 fragments existsuch as complement receptor 1 (CR1; CD35) that interactswith C3b, and complement receptors 3 (CR3; CD11b/CD18),and 4 (CR4; CD11c/CD18), that interact with iC3b (Micklemand Sim, 1985; Seya et al., 1985). These receptors areexpressed on significant cell populations including macro-phages, neutrophils, monocytes and dendritic cells.

In contrast to CR2, the receptors CR1, CR3 and CR4 do notbind to C3d, which is a considerably smaller C3 fragment(30kDa) than C3b and iC3b (175–180kDa). Formation of iC3boccurs following proteolytic cleavage of active C3 firstly by aC3 convertase to form C3b, and later cleavage of C3b to iC3bby the serine protease Factor I in conjunction with a co-factorsuch as Factor H or, indeed, CR1. The more complex cleavagepatterns of C3b and iC3b, and the requirement for extensivepost-translational processing of the C3 precursor to form thesecreted protein, makes recombinant C3b or iC3b fusionconstruct formation very difficult.

Exploitation of the reactive thioester in active C3 presentsa potential strategy for conjugating larger C3 fragments tocounterpart molecules of interest such as proteins andpolysaccharides. This would provide a means by which toopsonize synthetically materials of this kind, making themsuitable for uptake and processing by leucocytes such asdendritic cells and macrophages. Directing opsonized mate-rial to appropriate populations of these cells could elicit verybroad and articulate immune responses, potentially incorpor-ating multiple effector functions beyond B lymphocytes andimmunoglobulins.

Innovative studies by Villiers et al. (1999) utilized excesspurified C3 in conjunction with trypsin in order to opsonizeovalbumin and investigate its uptake and processing within Blymphocytes(Villiers et al., 1999). Furthermore, similar treat-ment of tetanus toxoid demonstrated the ability of conjugatedC3b to broaden the range of peptide antigen processing forpresentation to T lymphocytes (Cretin et al., 2007). Whilstachieving successful C3b attachment, the efficiency of thismethod is relatively low. In addition, for a broader strategy,the requirement for trypsin early in the protocol places boththe C3 and protein antigen at risk of excessive degradation,resulting in potential loss of function. Furthermore, theprecise location of the conjugation sites on the proteinantigen is very difficult to define.

It has been shown previously that treatment of active C3withmethylamine leads to nucleophilic attack of the thioesterbond, resulting in the formation of an amide and a freesulfhydryl group (Howard, 1980; Isenman et al., 1981). Inaddition, the cleavage of the thioester bond induces sub-stantial conformational changes across the C3 molecule,exposing novel binding sites for receptor binding andproteolysis sites for specific enzymes (Parkes et al., 1981;Seya et al., 1985). Through knowledge of this, we describeeffective and robust methods of whole C3 conjugation thatcircumvent the initial need for enzymes through the use ofchemical cross-linkers. The fulcrum for these strategies lies inthe exposure of the free sulfhydryl group at the reactive faceof C3 that can be targeted with high selectivity by maleimide-containing compounds. As such, covalent linkage of C3 toproteins modified with maleimide groups is possible, inaddition to the formation of multimeric C3 complexes withstreptavidin via site-selective labelling of methylamine-

treated C3 with water-soluble maleimide-biotinylationreagents. We further demonstrate herein the feasibility ofthis conjugation approach and indicate the simple andattractive means by which C3 and its biologically activefragments could comfortably participate within the emergingfield of chemical biology.

2. Materials and methods

2.1. Reagents

All reagents were purchased from Sigma Chemical Com-pany, except for sulfo-ε-maleimidocaproyloxysuccinimide(sulfo-EMCS) and EZ-Link PEO4-Maleimide Biotin, both fromPierce Chemical Company. SDS-PAGE and electroblottingwereperformed using the NuPAGE suite of gels, buffers, PVDF filtersand protein stains (Invitrogen). Polyclonal goat anti-humanC3c antibody was from Dako; horseradish peroxidase-con-jugated anti-goat IgG antibody was from Sigma.

2.2. Purification of human C3

Active human C3 was purified from fresh EDTA-plasma asdescribed previously (Dodds, 1993). The preparation wasdetermined as N 95% pure by SDS-PAGE and the presence ofthe intact thioester was confirmed by observation of autolyticcleavage of the C3 alpha chain obtained via heat denaturationprior to protein reduction with DTT (Sim and Sim, 1981).

2.3. Conjugation of whole human C3 to BSA via manipulation ofthe thioester group

BSAwas prepared as a 1mg/ml solution in a 500µl volumeof 25mM HEPES pH 7.4, 150mM NaCl, 5mM EDTA and treated

51D.A. Mitchell et al. / Journal of Immunological Methods 337 (2008) 49–54

with 50mM iodoacetamide in order to alkylate the free thiolgroup exposed in this protein. The preparation was dialyzedovernight against the above buffer and then treated withsulfo-EMCS, added to a final concentration of 1mM, incubatedfor 4h at 4°C before addition of ethanolamine hydrochloridepH 8.5 to a final concentration of 100mM. The sample wasincubated for 1h at 4°C and dialyzed as before. The dialyzedmaterial was added to 500µl of a 5mg/ml solution of purifiedhuman C3 in 25mM Tris pH 7.5, 150mM NaCl, 5mM EDTA,200mM methylamine hydrochloride pH 7.5. The mixture wasincubated for 24h at 4°C before addition of 5mM L-cysteine,incubation for 1h and further dialysis into 25mM HEPES pH7.4, 150mM NaCl, 5mM EDTA. Downstream analysis via SDS-PAGE and Western blotting involved the NuPAGE MES buffersystem (Fig. 1).

2.4. Thioester-site selective biotinylation of whole C3 andformation of C3 oligomers

Purified C3 was prepared as a 2mg/ml solution in a 1mlvolume of 25mM Tris pH 7.2, 150mM NaCl, 5mM EDTA andwas treated with methylamine hydrochloride pH 7.5 to a finalconcentration of 200mM. PEO4-maleimide Biotin was addedto a final concentration of 1mM. This was incubated for 18h at4°C, after which L-cysteine was added to a final concentrationof 5mM. The mixture was subsequently run on a Superose-6column equilibrated in 25mM HEPES pH 7.4, 150mM NaCl,5mM EDTA. Eluted C3 material was pooled and then added toa solution of FITC-streptavidin at a molar ratio of 4 parts C3 toone part FITC-streptavidin and incubated for 18h at 4°C in thedark. A volume of 1ml of this latter mixture was run later as asample on the same Superose-6 column as used above.

2.5. Protein chromatography and Western blotting

Size exclusion chromatography was performed usingSuperose-6 and Superose-12 columns on AKTA-purifier liquidchromatography apparatus (GE Healthcare) at a flow rate of0.5ml/min with absorbance detection at 280nm and 494nm.Western blotting was performed using 25mM Tris pH 7.4,

Fig. 2. Size exclusion chromatography of C3-BSA conjugates. Panel A shows the eluprotein detection via absorbance at 280 nm. Arrows indicate relative elution peaks forder C3-BSA conjugates. Panel B shows a Coomassie-stained gel of protein containnon-reducing conditions.

150mM NaCl, 1mM EDTA, 0.2% Tween-20, 2% w/v desiccatedmilk as blocking and antibody diluent buffer. Incubationperiods for primary and secondary antibodies were 2h and 1hrespectively. Membrane wash buffer was as above minus thedesiccated milk.

3. Results

3.1. Formation and confirmation of covalent C3-BSA complexes

BSA was treated with sulfo-EMCS and methylamine-C3,as described in Materials and methods, with appropriateamine and sulphydryl reactivity blocking steps and dialysisprocedures in order to eliminate artefactual side reactions.Examination of the sulfo-EMCS mediated conjugationreactions via SDS-PAGE revealed the presence of proteinspecies with molecular weights greater (N 180kDa) thanwhole C3 alone (Fig. 2A). These higher molecular weightprotein species were neither observed in samples of BSAthat had been treated with sulfo-EMCS and then L-cysteine,nor in samples of methylamine-C3 that had been treatedwith ethanolamine-blocked sulfo-EMCS. Similarly, a mix-ture of these two latter samples incubated for 1h and thenrun within the same gel track did not reveal any of thehigher molecular weight species. Western blotting withanti-C3c polyclonal antibody clearly showed that themultiple high molecular weight species in the sample thathad undergone sulfo-EMCS mediated cross-linking containC3 (Fig. 2B).

The cross-linked samplewas run on a Superose-12 columnin order to examine its behaviour on size-exclusion chroma-tography. The elution profile showed the emergence ofprotein shortly after the void volume in the elution, indicatingthe presence of high molecular weight material (Fig. 3A).Compared with the calibrated elution volumes for C3 and forBSA, this material eluted much earlier, indicating that itconsisted of particles of significantly larger radii. Examinationof key fractions from the elution by SDS-PAGE confirmed thepresence of these high molecular weight species eluting early(Fig. 3B).

tion profile of the cross-linking reaction run on a Superose-12 column withor monomeric C3 and BSA. Material left of the dashed line represents highering Fractions (F1–F4) indicated on the elution profile. All samples run under

Fig. 3. Size exclusion chromatography of C3-streptavidin conjugates. Protein detection was performed via absorbance at 280 nm (thin lines) with 494 nm (boldline) to allow concomitant detection of fluorescein. Panel A shows overlaid chromatograms from successive runs of i) biotinylated C3 (dashed line) and ii) FITC-streptavidin on the same Superose-6 column. In the biotinylated C3 elution, the unconjugated biotinylation reagent absorbs strongly at 280 nm and emerges late,close to the total column volume (Vt), whilst C3 has an elution volume (Ve) of 15.5 ml. The Ve for FITC-streptavidin is observed at 17.5 ml. Panel B shows a singlechromatogram of the biotinylated C3 FITC-streptavidin mixture. A high molecular weight peak with A280 and A494 absorbance properties is observed with a Ve of11.5 ml. A peak with a Ve of 15.5 ml, corresponding to C3, is also seen but this does not absorb at 494 nm.

52 D.A. Mitchell et al. / Journal of Immunological Methods 337 (2008) 49–54

3.2. Formation and confirmation of oligomeric C3-streptavidincomplexes

Methylamine-C3 was treated with PEO4-maleimide biotinand incubated with FITC-streptavidin as described above. Themixture was run on Superose-6 with dual absorbancedetection at 280nm for protein and 494nm for fluorescein.As seen in Fig. 3, a large peak is seen early in the elution profilewith absorbance characteristics at both 280 nm and 494 nm.By comparison, PEO4-maleimide biotin-treatedmethlyamine-C3 alone emerges significantly later in the elution profile.Similarly FITC-streptavidin elutes considerably later whenrun alone. The detection of a highmolecular weight peakwithdistinctive A280/494 nm absorbance characteristics clearlyindicates the formation of a multimeric C3 complex formedwith FITC-streptavidin. The elution volume of this complexwas determined as 11.5 ml, equivalent to that of N750 kDaspecies run on this Superose-6 column. This molecular weightis consistent with the formation of a “tetrameric” C3 complexformed with a C3: FITC-streptavidin ratio of 4:1. Similarcomplexes are well-documented, especially through the useof MHC-streptavidin probes (Altman et al., 1996).

4. Discussion

The ability to modify protein surfaces site-selectively withnovel, bioactive components represents a very powerful andtimely tool for the generation of highly-defined, tailored andarticulate multifunctional agents. Such agents would becapable of delivering new therapeutic solutions and biological

probes. Conjugation strategies of this kind are particularlyvaluable for immune-related endeavours wherein the target-ing and enhancement of materials towards immune cells candrive powerful physiological responses, especially the uptakeand processing of vaccines and immunomodulators.

The method of C3 conjugation described here provides achemical-based solution that retains the biologically activeorientation of C3 without derivatizing key surface residuesrequired for complement receptor binding or downstream C3proteolytic processing. Through targeting a free sulfhydrylexclusive to the former thioester site, the method employsrobust and reliable maleimide chemistry that has beenreproducibly used throughout the field of protein chemistry.Furthermore, the conjugation method can still be applied tostored C3 that has undergone spontaneous hydrolysis, sincethe free sulfhdryl is generated under these circumstancesalso. Therefore risks to efficiency within a project, broughtabout by the long-term instability of purified C3, areminimized. In addition, our data indicate the formation ofoligomeric complexes of C3, especially in the biotin-strepta-vidin system. This is important, since multiple interactionsbetween C3 fragments and their receptors are required forefficient transduction of cellular signals and mediation ofphagocytosis (Arnaout et al., 1983; Ueda et al., 1994; Dempseyet al., 1996).

Conjugating whole C3 from the start creates a variety ofopportunities. It has been shown previously that thioester-cleaved, or “dead”, C3 resembles C3b and can interact withCR1, thus untreated conjugates could feasibly be usedstraightforwardly in targeting this receptor (Seya et al.,

53D.A. Mitchell et al. / Journal of Immunological Methods 337 (2008) 49–54

1985). Careful, sequential treatment of conjugates withtrypsin and Factor I/Factor H (Parkes et al., 1981) wouldyield an iC3b-like form and later C3d, allowing for targeting toCR2, CR3 and CR4. Of interest is that dermal dendritic cellsexpress CR3 (Rozis et al., 2005), making delivery of iC3b-conjugatedmaterial via the skin a feasible step. Dendritic cellsare arguably the most potent modulators of adaptive immuneresponses. It has been shown recently that stimulation of theCR3 receptor complex on DC populations induces immuno-suppressive effects (Behrens et al., 2007). This opens up theextremely attractive strategy via which, under certaincircumstances, C3-conjugated materials could be targeted tospecific DC populations in order to induce antigen uptake andtolerization. The possibility of generating novel chemical C3dconjugates via sulfhydryl coupling could easily incorporatenon-protein structures such as lipopolysaccharides, peptido-glycan constituents and capsid carbohydrates. Conjugatessuch as these could provide support to existing strategiesusing recombinant C3d fusion proteins.

Study of the complement system remains firmly at theforefront of immunological research, driven especially by therecent solution of C3 structures by X-ray crystallography(Janssen et al., 2005; Janssen et al., 2006). In addition, anumber of gene-targeted mice have been generated withcomplement deficiencies (Holers, 2000; Pickering et al.,2002). Through the development of C3 conjugates incorpor-ating a variety of cross-linking reagents (Flinn et al., 2004),bioactive counterparts and detection/recovery vehicles suchas biotin-streptavidin systems, fluorochromes and magneticparticles, it will be possible to embark upon new, integratedcomplement research. By extrapolation, we expect otherthioester-containing molecules such as C4 and alpha-2-macroglobulin also to be open to this approach. Successfulexpression of recombinant histidine-tagged rat C4, shown tohave an active thioester bond (Chen and Wallis, 2004),provides evidence for rapid thioester protein productionand recovery that will be of great benefit in augmenting anddriving this work.

In addition to advancing vaccine technology and under-standing of the immune system, defined C3 conjugationstrategies could broaden our understanding of transmissibledisease agents that exploit complement function in order todisseminate within the body of the host. A topical example ofthis can be seen within diseases such as the transmissiblespongiform encephalopathies, believed by many to requireprion proteins for transmission and pathology. It has beenshown that the prion protein PrP can bind to complement C1qand subsequently activate the complement cascade in theabsence of antibody (Blanquet-Grossard et al., 2005;Mitchell etal., 2007). There is strongevidence fromanalysis of complementdeficient mice to suggest that this phenomenon is essential fordisease-associated transmission of scrapie PrP material fromperipheral regions to the brain (Klein et al., 2001;Mabbott et al.,2001). Through the use of synthetic PrP-C3 conjugates, we aimto generate novel probes with which to study this.

Acknowledgements

This work is supported by the European Union, ProjectNumber FOOD-CT-2006-023144; and the University of War-wick Research Development Fund. D.A.M. is a Research

Councils UK Academic Fellow. The authors would like tothank fellow Participants within the Immunoprion ProjectConsortium. D.A.M would like to thank Russell Wallis,University of Leicester, for helpful discussion.

References

Altman, J.D., Moss, P.A., Goulder, P.J., Barouch, D.H., McHeyzer-Williams, M.G.,Bell, J.I., McMichael, A.J., Davis, M.M., 1996. Phenotypic analysis ofantigen-specific T lymphocytes. Science 274, 94.

Arnaout, M.A., Dana, N., Melamed, J., Medicus, R., Colten, H.R., 1983. Low ionicstrength or chemical cross-linking of monomeric C3b increases itsbinding affinity to the human complement C3b receptor. Immunology48, 229.

Behrens, E.M., Sriram, U., Shivers, D.K., Gallucci, M., Ma, Z., Finkel, T.H.,Gallucci, S., 2007. Complement receptor 3 ligation of dendritic cellssuppresses their stimulatory capacity. J. Immunol. 178, 6268.

Blanquet-Grossard, F., Thielens, N.M., Vendrely, C., Jamin, M., Arlaud, G.J.,2005. Complement protein C1q recognizes a conformationally modifiedform of the prion protein. Biochemistry 44, 4349.

Chen, C.B., Wallis, R., 2004. Two mechanisms for mannose-binding proteinmodulation of the activity of its associated serine proteases. J. Biol. Chem.279, 26058.

Cretin, F.C., Serra, V.A., Villiers, M.B., Laharie, A.M., Marche, P.N., Gabert, F.M.,2007. C3b complexation diversifies naturally processed T cell epitopes.Mol. Immunol. 44, 2893.

Dempsey, P.W., Allison, M.E., Akkaraju, S., Goodnow, C.C., Fearon, D.T., 1996.C3d of complement as a molecular adjuvant: bridging innate andacquired immunity. Science 271, 348.

Dodds, A.W., 1993. Small-scale preparation of complement components C3and C4. Methods Enzymol. 223, 46.

Flinn, N.S., Quibell, M., Turnell, W.G., Monk, T.P., Ramjee, M.K., 2004. Novelchemoselective linkage chemistry toward controlled loading of ligandsto proteins through in situ real-time quantification of conjugateformation. Bioconjug. Chem. 15, 1010.

Gadjeva, M., Dodds, A.W., Taniguchi-Sidle, A., Willis, A.C., Isenman, D.E.,Law, S.K., 1998. The covalent binding reaction of complementcomponent C3. J. Immunol. 161, 985.

Holers, V.M., 2000. Phenotypes of complement knockouts. Immunopharma-cology 49, 125.

Howard, J.B., 1980. Methylamine reaction and denaturation-dependentfragmentation of complement component 3. Comparison with alpha2-macroglobulin. J. Biol. Chem. 255, 7082.

Isenman, D.E., Kells, D.I., Cooper, N.R., Muller-Eberhard, H.J., Pangburn, M.K.,1981. Nucleophilic modification of human complement protein C3:correlation of conformational changes with acquisition of C3b-likefunctional properties. Biochemistry 20, 4458.

Janssen, B.J., Christodoulidou, A., McCarthy, A., Lambris, J.D., Gros, P., 2006.Structure of C3b reveals conformational changes that underlie comple-ment activity. Nature 444, 213.

Janssen, B.J., Huizinga, E.G., Raaijmakers, H.C., Roos, A., Daha, M.R., Nilsson-Ekdahl, K., Nilsson, B., Gros, P., 2005. Structures of complementcomponent C3 provide insights into the function and evolution ofimmunity. Nature 437, 505.

Klein, M.A., Kaeser, P.S., Schwarz, P., Weyd, H., Xenarios, I., Zinkernagel, R.M.,Carroll, M.C., Verbeek, J.S., Botto, M., Walport, M.J., Molina, H., Kalinke, U.,Acha-Orbea, H., Aguzzi, A., 2001. Complement facilitates early prionpathogenesis. Nat. Med. 7, 488.

Mabbott, N.A., Bruce, M.E., Botto, M., Walport, M.J., Pepys, M.B., 2001.Temporary depletion of complement component C3 or genetic defi-ciency of C1q significantly delays onset of scrapie. Nat. Med. 7, 485.

Micklem, K.J., Sim, R.B., 1985. Isolation of complement-fragment-iC3b-binding proteins by affinity chromatography. The identification ofp150,95 as an iC3b-binding protein. Biochem. J. 231, 233.

Mitchell, D.A., Kirby, L., Paulin, S.M., Villiers, C.L., Sim, R.B., 2007. Prion proteinactivates and fixes complement directly via the classical pathway:implications for the mechanism of scrapie agent propagation inlymphoid tissue. Mol. Immunol. 44, 2997.

Parkes, C., DiScipio, R.G., Kerr, M.A., Prohaska, R., 1981. The separation offunctionally distinct forms of the third component of human comple-ment (C3). Biochem. J. 193, 963.

Pickering, M.C., Cook, H.T., Warren, J., Bygrave, A.E., Moss, J., Walport, M.J.,Botto, M., 2002. Uncontrolled C3 activation causes membranoprolifera-tive glomerulonephritis in mice deficient in complement factor H. Nat.Genet. 31, 424.

Rozis, G., de Silva, S., Benlahrech, A., Papagatsias, T., Harris, J., Gotch, F.,Dickson, G., Patterson, S., 2005. Langerhans cells are more efficiently

54 D.A. Mitchell et al. / Journal of Immunological Methods 337 (2008) 49–54

transduced than dermal dendritic cells by adenovirus vectors expressingeither group C or group B fibre protein: implications for mucosalvaccines. Eur. J. Immunol. 35, 2617.

Seya, T., Holers, V.M., Atkinson, J.P., 1985. Purification and functional analysisof the polymorphic variants of the C3b/C4b receptor (CR1) andcomparison with H, C4b-binding protein (C4bp), and decay acceleratingfactor (DAF). J. Immunol. 135, 2661.

Sim, R.B., Sim, E., 1981. Autolytic fragmentation of complement componentsC3 and C4 under denaturing conditions, a property shared with alpha 2-macroglobulin. Biochem. J. 193, 129.

Ueda, T., Rieu, P., Brayer, J., Arnaout, M.A., 1994. Identification of thecomplement iC3b binding site in the beta 2 integrin CR3 (CD11b/CD18). Proc. Natl. Acad. Sci. U. S. A. 91, 10680.

Villiers, M.B., Villiers, C.L., Laharie, A.M., Marche, P.N., 1999. Amplification ofthe antibody response by C3b complexed to antigen through an esterlink. J. Immunol. 162, 3647.

Walport, M.J., 2001a. Complement. First of two parts. N. Engl. J. Med. 344,1058.

Walport, M.J., 2001b. Complement. Second of two parts. N. Engl. J. Med. 344,1140.

Zabel, M.D., Weis, J.H., 2001. Cell-specific regulation of the CD21 gene. Int.Immunopharmacol. 1, 483.