- Email: [email protected]
Human complement component C2: Production and characterization of polyclonal and monoclonal antibodies against C2
Molecular Immunology. Vol. 23, No. 8, pp. 879-886, 1986 hinted in Great Britain.
0161-5890/86 $3.00 + 0.00 Pergamon Journals Ltd
HUMAN COMPLEMENT COMPONENT C2: PRODUCTION AND CHARACTERIZATION OF POLYCLONAL AND MONOCLONAL ANTIBODIES AGAINST C2 EVA I. STENBAEK,* CLAUS KOCH,? VIBEKE BARKHOLT* and KAREN G. WELINDER* *Institute of Biochemical Genetics, University of Copenhagen, Oster Farimagsgade 2A, DK-1353
Copenhagen
K, Denmark and tThe Vaccine Department, DK-2300 Copenhagen S, Denmark
Statens
Seruminstitut,
(First accepted 28 October 1985; accepted in revised form 20 March 1986)
Abstract-Rabbit antibodies to human complement component C2 were produced by immunization of rabbits with precipitates from line immunoelectrophoresis, and the antibodies were used to monitor a classical chromatographic purification of C2 and for affinity purification of C2. Twelve monoclonal antibodies with specificity for human complement component C2 were produced by fusion of myeloma cells with spleen cells from mice immunized with the affinity purified C2. The specificity of the monoclonal antibodies was confirmed by their reaction with antigen-antibody precipitates where C2 was the antigen, and by their specific reaction with C2 after separation in SDS-PAGE followed by immunoblotting. The affinity of the monoclonal antibodies varied as demonstrated by the titration curves in ELISA. The antibodies will be of importance for immunospecific purification of human C2 and C2 fragments, for specific depletion of C2 from human serum, and for quantification of C2 for clinical purposes.
INTRODUCTION
inactivation. It is present in allelic forms with a common allele, C2C which has a gene frequency of 95%, with a basic variant, C2B which has a gene frequency of 4-5%, and with two rare acidic variants, C2Al and C2A2, which have gene frequencies of less than 1% (Woods et al., 1984). C2-deficiency is the most common of the complement deficiencies [O.Ol % of the population (Udea et al., 1983)]. The structural gene for C2 is linked to the MHC of all hitherto investigated mammals (Porter, 1984), and there is evidence that C2 from the classical complement pathway and factor B from the alternative pathway have evolved by gene duplication from a common ancestral gene. Functionally, C2 is engaged in the formation of the C3 and C5 convertases of the classical complement pathway. The availability of monoclonal antibodies to human C2 is important from several points of view. It will simplify the purification of substantial amounts of human C2 which in turn will facilitate further structural and functional analyses of human C2. The monoclonal antibodies may also be useful for the study of the interaction of C2 with other complement components during the formation of C3 and C5 convertases.
The complement component C2, which is found in human plasma in low quantities (15-20 mg/l), is involved in activation of the complement system via the classical complement pathway (Porter and Reid, 1979). This activation involves the binding of Cl to the Fc-part of antibodies in antigen-antibody complexes, leading to the formation of a Cl-esterase which as its substrates has C4 and C2. Following C4 cleavage, C2 (M, 100,000) is cleaved into two fragments, C2b (M, 30,000) and C2a (M, 70,000). C2a forms a Mg’+-dependent complex with C4b, and this complex, C4b2a, is the C3 convertase of the classical complement pathway. C2a carries the proteolytic site of the C3 convertase which causes activation of C3 by producing the fragments C3a and C3b. C3b binds to C4b2a, thereby forming the classical C5 convertase, C4b2a3b. The proteolytic site of this convertase is also located on C2a. C2 therefore plays a major role in the activation of the classical, antibody dependent complement pathway. C2 exists as a single chain glycoprotein and is very labile with regard to proteolytic cleavage and to heat Abbreviations: EDTA, ethylenediaminetetraacetate; ELISA, enzyme-linked immunosorbent assay; SDS-PAGE, polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate; PBS, phosphate-buffered saline, PH 7.3; DMEM, Dulbecco’s modified Eagle’s medium.
MATERIALS AND
METHODS
Antibodies
Antibodies to human serum proteins (A206) and peroxidase conjugated rabbit antibodies to mouse 879
EVA I. STENBAEK et
880
immunoglobulins (P260) were obtained from Dakopatts. Antibodies for typing and subtyping the immunoglobulin classes of the monoclonal antibodies, were purchased from Miles. Monospecific, polyclonal antibodies to human C2 were raised in rabbits after S.C. immunization with precipitates obtained from line immunoelectrophoresis, essentially as described by Harboe and Closs (1983) initially using C2 purified by classical chromatographic methods (Kerr, 1981) and antihuman C2 kindly donated by Dr M. Kerr, Dundee. Cell fusion protocol for the production of monoclonal antibodies to human C2 Balb/c mice were immunized i.p. with human C2, purified on an affinity column. The amount of antigen for each immunization was 30 p g/mouse, and the antigen was adsorbed onto Al(OH), (Hem and White, 1984). After two immunizations with an interval of 2 weeks the mice were bled and tested for anti-C2 reactivity by ELISA. Mice with high titres were selected for fusion, and their spleens were removed 3 days after a third immunization. Cell fusion was performed with the myeloma cell line X-63 Ag8 6.5.3 essentially as described by Kiihler and Milstein (1975). Fourteen days after fusion the cell culture supernatants were tested for specific reactivity against human C2, and antibody producing hybridomas were cloned until monoclonality was obtained by means of a limiting dilution method. The monoclonal antibodies were produced as culture supernatants in DMEM supplemented with 10% foetal calf serum. Purification of antibodies The IgG fraction of polyclonal rabbit antisera and the monoclonal antibodies from culture supernatants were purified by affinity chromatography on Protein A-Sepharose CL4B (Hjelm et al., 1972). Samples were adjusted to pH 8.0 before application to the column, and the bound antibodies were eluted with 0.1 M glycine-HCl, 0.5 M NaCl, pH 2.95. ELISA
described by Laemmli (1970) and the proteins were blotted onto nitrocellulose paper (Schleicher and Schuell) in 25 mM Tris-HCl, pH 9.0, 192 mM glytine, 20% (v/v) methanol for 5 hr at 10 V/cm at 8°C (Towbin et al., 1979). The nitrocellulose paper was washed and incubated with purified hybridoma antibodies and then incubated with peroxidase conjugated rabbit anti-mouse immunoglobulins. The peroxidase activity was assayed with H,Oz and 3,5,3’5’tetramethylbenzidine (Merck 8622) supplemented with dioctyl sodium sulphasuccinate (Sigma D-0885). Binding of monoclonal antibodies to precipitates in crossed immunoelectrophoresis Crossed immunoelectrophoresis was performed with human EDTA plasma supplemented with purified human C2 as antigen, against polyclonal anti-human serum, supplemented with rabbit monospecific polyclonal antibody to human C2 (Laurel& 1965). After electrophoresis the plates were washed and incubated first with purified hybridoma antibody and subsequently with peroxidase conjugated rabbit anti-mouse immunoglobulins. Binding of hybridoma antibodies to specific precipitates was detected by staining with H,O, and 3-amino,9-ethylcarbazole (Skjodt et al., 1984). Characterization of Ig class and subclass of monoclonal antibodies Class and subclass of the monoclonal antibodies were determined by Ouchterlony double immunodiffusion in agarose using monospecific rabbit antibodies against mouse IgGl, IgG2a, IgG2b, IgG3, IgA, IgM, kappa light chains and lambda light chains (Miles) (Nilsson, 1983). Immunosorbent column Monoclonal IgG (18 mg) from ascites fluid purified on Protein A-Sepharose 4B was coupled to CNBractivated Sepharose (8 g) yielding 28 ml of immunosorbent gel. RESULTS
Microtiter plates (Technunc) were coated overnight with purified human C2 (0.1 pgg/well) in PBS for direct ELISA, or with the IgG fraction of polyclonal anti-human C2 for indirect ELISA. The anti-C2 coated microtiter plates were incubated with purified C2 (0.1 pg/well). Microtiter plates coated with C2-deficient human serum were used as negative controls. The ELISA was performed essentially as described by Engvall and Perlmann (1972). Ig binding was detected using o -phenylenediamine (OPD) and the reactivity was measured at 492 nm by means of an MR 580 microelisa autoreader (Dynatech). SDS-PAGE
al.
and immunoblotting
Purified human C2 was analyzed by SDS-PAGE (5-15% polyacrylamide gradient gel) essentially as
Characterization of monospecific, polyclonal bodies to human C2
anti-
Two antibodies, C2-721 and C2-779, prepared as described in Materials and Methods were selected. C2-721 formed a sharp precipitate with C2 when analyzed in crossed immunoelectrophoresis, indicating a relatively high avidity for C2, whereas C2-779 showed a more diffuse precipitate with C2. Both antibodies were monospecific for human C2 when tested in crossed immunoelectrophoresis with human serum as antigen. When C2-721 was tested against human serum in which C2 had been activated by the classical complement pathway with heat aggregated human immunoglobulin, the position of the precipitate was shifted towards the anode as com-
Polyclonal and monoclonal antibodies against human complement component C2
881
Fig. 1. Crossed immunoelectrophoresis of human serum run against polyclonal, monospecific anti-human C2-721. (a) Human serum diluted 2:l in PBS. (b) Human serum activated with antigen-antibody complexes for 120 min, at 37°C in PBS (2: 1). After electrophoresis in the first dimension for 1 hr at 10 V/cm, antigens were run overnight in the second dimension at 2 V/cm. Both plates were stained with Coomassie Blue R-250.
pared to non-activated C2, indicating the presence of C2a, the proteolytic subunit of C2 (Nagasawa and Stroud, 1977) (Fig. 1). A$inity chromatography by polyclonaI antibody The low affinity antiserum C2-779 was used for affinity purification of human C2. The antibodies were purified on a Protein A-Sepharose column, and coupled to CNBr-activated Sepharose. A 1.5 x 10 cm column containing approx. 30 mg of coupled IgG was prepared. EDTA plasma (25 mi)was applied to the column and after thorough washing with phosphate buffer, pH 8.0, the column was eluted with 0.1 M glycineHC1, pH 2.8. Fractions were immediately neutralized and analyzed by immunoelectrophoresis against an antiserum to human serum proteins and an anti-human C2. The C2 preparation showed a minor contamination of human immunoglobulin. The eluate was analyzed for human complement factor B, but there was no reaction with rabbit anti-human factor B. The purified human C2 gave a sharp precipitate in crossed immunoelectrophoresis with anti-human C2 obtained from Dr M. Kerr, and with C2-721. When tested in SDS-PAGE, the C2 preparation showed a major band with M, of 100,000 (not shown). Production and characterization of monoclonal antibodies Mice were immunized with the immunosorbent purified human C2. We performed two separate cell fusions, and from the first fusion one positive clone was selected and from the second fusion 11 clones were selected for further cloning.
After at least two consecutive phases of cloning the produced monoclonal antibodies (designated C2-Sl through C2-S12) were characterized with respect to specificity in direct and in indirect ELISA, and to reactivity with C2/anti-C2 precipitates developed in crossed immunoelectrophoresis. Finally the reactivity of the monoclonal antibodies was tested by immunoblotting of SDS-gels of C2 preparations. Fig. 2(a) illustrates the SDS-PAGE analyses of preparations of C2 purified by affinity chromatography on the C2-Sl column. Fig. 2(b) demonstrates the reactivity and fragment specificity of the monoclonal antibodies. Two monoclonal antibodies C2-S3 and C2-S9, did not react with human C2 nor C2 fragments after C2 had been exposed for 5 min of boiling in SDS. This indicates that the antibodies might require native C2 to be reactive. Antibodies C2-S2, C2-S4, C2S5, C2-S7, C2-S8, C2-SlO, C2-Sll and C2-S12 reacted with C2 and with a split product which we assume to be C2b (M, 30,000). Antibody C2-SlO reacted weakly with C2 and with C2b, and is therefore not visible in Fig. 2(b), lane 10. The antibodies C2-Sl and C2-S6 reacted with C2 and a fragment which we assume to be C2a (M, 70,000), although C2-Sl reacts reluctantly with C2 and is therefore not seen in Fig. 2(b). C2S6 reacted notably stronger with C2a than with the intact C2. This suggests that C2-S6 reacts with an epitope which is more exposed or more stable in C2a than in C2. It is, furthermore, noteworthy that our rabbit anti-C2 sera, C2-721 from a late bleeding is almost exclusively directed against C2a (Fig. 2(b), lane 14). Figure 3 illustrates the reactivity of one of the
zooI Iis92-
E;fr45
31” 2114-
x 10-3
c2 Gh-
&2b-
-
94
-
67
-
43
-
30
-
20 14
Fig. 2. SDS-PAGE analysis of affinity purified hur~~ar~C2. (a) Different C2 preparations in a S-IS% gel stained by Coomassie Blue R-250, Protein markers are seen in the left lane. (b) Immunoblotting of B SDS-PAiSE gel with Ed& amounts of a partially degraded C2 sample in lanes 1-14. Nitrocelluiose prints of the individurtt &es were incubated with the different rno#~~o~a~ a&bodies agaiztst CZ _Antibo&es CZSI through CZ-S12 (lane l-12), monocIona1 mouse anti-& as a negative control (tane $31, and polyclonal anti-C2 (C2~721), as a positive control (lane 14). Protein markers stained with Amido Black are seen in the lane ‘M.
883
Polyclonal and monoclonal antibodies against human complement component C2
Fig. 3. (a) Cmssed imrnunoe~e~tro~~o~s~sofhmnan EDTA-plasma enriched with purified C2 as antigen and anti-human serum @?06, Dakopatts) supplemented with C&721 as antibody. The first dimension was run for 1hr at 10 V/cm, and the second dlmensmn was run overnight at 2 V/cm. After ekctrophoresia the gel was first incubated with monoclonai anti-C2 (C2-S4) and then with peroxidase-conjugated rabbit anti-mouse imm~~og~ob~i~. The C&m&C2 precipifafe stained by f&s procedure same CfE plate is stained with Coomassie Bhze R-250.
Three antibodies, C2-S2, C2-S9 and C2-S12, did not bind to CZ/anti-C2 precipitates in crossed imm~n~l~tro~hore~s~ Low aBnity or c~rn~ti~o~ with polyslnal antibodies for the same epitopes may explain this lack of reactivity. Antibody C2-S9 reacu ted neither with the C2/anti-C2 precipitate nor with the SIX-PAGE treated C2, The fact that this antibody is an IgM antibdy may explain the lack of reactivity. Table 1 summarizes the data for the 12 monoclonal antibodies. We found that the mono&onal antj~d~es show&
ELBA
Direct
cl-s1
0.340 0.410 a.120 a.290 0.620 O.lOU 0. I24 0.1 IO O.f4O 0.170 1.500 0.270
~,_,~~_“*” CZ-S2 C&S3
c2-s4 CZ-ST c2-scr c2-s7 CZ-SS C2-s CTSIO c2-SI 1 c2-s12
Indirect
1”1
Immunablotting
-“v-f f -
“Arbitrary concns of the 12 monoclon& ** + + imtkate3 the highs affinity. ND = Noi determined.
antibodies,
Fragment
C&W, subclass, type .,_
~,
CIE
Affinity’
specificity
Heavy chain
+ _
++ i++ -I+ i -6 +++ c ++ it+ -i+* +4-+ +++
C2a C2b ND C2b cztr c2tl C2b C2b ND C2b c2b C2b
IgOi ND ??a IgG2a ND IgQl IgOl I@I I$# rao1 IgGI ND
+ f f f + -
-f f f + f + -
+ -I+
-t + -
reacting
C2. (bj Tote
&ferent patterns of reactions when te&& in direct ELISA (C2 attached directly onto the microtiter plates) as compared to indirect ELISA (C2 was attached to a primary layer of rabbit polyclonaf a&X2>. The in~~d~~ values listed in Table I, shows that the ratio direct ELISAjindirect ELISA varies significantly, from > 1 (C2-S2, C2-S5, C2-S7, C2-Sll) through w 1 (C2-S3, C2-S4, C2-S6, C2-S9 and C2-SK!) to
SDS-PAGE/
0.920 0.030 8,160 0.370 0.260 0.119 0.019 1.500 A160 i.5M3 0.102 0.4ya
is mafked
..“”
with a given amount
of puritied
C2 in direct
and indkct
Light
chain
Ww ND ND Wv ND kw’a kww kwpa WV& Wpa kwpa ND ELBA.
884
EVA I. STENBAEKet al.
n
”
Polyclonal and monoclonal antibodies against human complement component C2 seem to be a correlation between the ability of the monoclonal antibodies to bind to precipitates of Cyanti-C2 in crossed immunoelectrophoresis and the ability to bind to C2/anti-C2 complexes on microtiterplates. Table 1 shows the results of antibody class and Ouchterlony immunosubclass typing by precipitation. All antibodies had kappa light chains and all but two antibodies were of the IgGl subclass. Four antibodies have not yet been typed. AJinity
of monoclonal
antibodies
The relative affinities of the monoclonal antibodies were estimated from the slopes of the curves in direct ELISA from a dilution series of each antibody (Fig. 4). Figure 4 and Table 1 show that the affinity varies greatly, a fact which is of importance for further use of the antibodies. Aj%ity
chromatography
by monoclonal
antibody
The low affinity antibody C2-Sl was selected for use in affinity purification of human C2. C2 was first purified partially by euglobulin precipitation and CM-Sephadex C-50 ion-exchange chromatography in the presence of benzamidine (Kerr, 1981) and then passed through the immunosorbent column. The column was washed thoroughly with PBS and C2 was eluted with 0.1 M propionic acid, 0.15 M NaCl, pH = 3.6. The C2 fractions were immediately neutralized (pH = 6.0) with 0.25 M NaOH. C2 was analyzed in SDS-PAGE as shown in Fig. 2(a), where the major component with M, 100,000 is the C2 band. The overall yield was estimated to 40% by amino acid analysis of the purified C2, assuming that normal human plasma contains 20mg/l of C2. DISCUSSION
The human complement component C2 has been purified previously by classical methods as described by Udea et al. (1983), Ngan and Minta (1981), Kerr (1981) and Nagasawa and Stroud (1977). C2 was purified by Udea et al. (1983) using affinity chromatography (polyclonal anti-C2 coupled to CNBractivated Sepharose 4B). Kerr (1981), Ngan and Minta (1981) and Nagasawa and Stroud (1977) obtained highly purified C2 by standard procedures, such as salt precipitation, ion-exchange chromatography and gel filtration. Low yields (1630%) are obtained with these procedures, which are very tedious and time consuming. The low yields and the gradual loss of biological activity during the laborious procedures made it tempting to produce monoclonal antibodies against C2 as a tool for purification and characterization of C2. We have purified C2 from plasma directly on an affinity column with polyclonal, low-affinity rabbit antibody C2-779. Relatively pure C2 which was useful for the immunization of Balb/c mice before cell fusion and also for the initial screening of the positive
885
culture supernatants for anti-C2 producing clones was easily obtained in this way. On the other hand, the limited yield and the presence of impurities in this C2 preparation showed no superiority when compared to classical fractionation procedures. Preliminary experiments with a monoclonal immunosorbent lived up to our expectation that monoclonal antibodies are extremely valuable for affinity chromatographic purification of C2. One of the monoclonal antibodies with low affinity for C2 (C2-Sl), has been used with success for the preparation of biologically active, highly purified C2 in 40% yield. This purification procedure is presently being optimized. In order to confirm the specificity of the monoclonal antibodies the following assays were applied: (1) immunoblotting of C2 separated by SDS-PAGE; (2) postincubation of CIE plates of human serum supplemented with purified C2 run against antihuman serum enriched with polyclonal anti-C2. As seen in Table 1, these assays confirmed the specificity of the monoclonal antibodies for C2 and also gave indications of variations in reactivity. Of the 12 clones producing monoclonal antibodies one produced an IgM antibody (C2-S9), probably of limited value. Ten of the antibodies reacted with C2 partially denaturated after SDS-treatment. These antibodies will be valuable for quantification and detection of C2 split products. The C2-S3 antibody, which does not react with SDS-treated C2, may be useful for the detection of native, biologically active c2. Affinity is an important qualitative feature of a monoclonal antibody. In general low affinity monoclonal antibodies are needed for affinity-purification of the specific antigen, whereas high affinity antibodies are superior with respect to specific removal or immunospecific detection of an antigen. Both types-high affinity and low affinity antibodies as judged from ELISA titration curves-have been found in our panel of monoclonal antibodies to human C2 (Fig. 4). For clinical purposes the antibodies may be employed in methods aimed at quantifying the plasma level of C2 and/or C2 split products and also for the detection of the cellular distribution of C2. It has been described previously that in systemic lupus erythematosus (SLE), heriditary angioedema (HAE) and acute post streptococcal glomerulonephritis (AGN), there is often a significant change in the amount of circulating C2 and C2 split products (Udea et al., 1983; Sjoholm and Sturfelt, 1984). As previously mentioned C2, which is an MHC class III antigen, occurs in allelic forms (Bentley and Porter, 1984; Carol], 1984). There are at present four known allelic genes coding for C2 that give rise to polymorphic forms of C2. It has been suggested that the polymorphism of the class III proteins may be a factor in susceptibility to certain diseases, in particumanifestations lar diseases with auto-immune
EVA I. STENBAEK et al.
886
(Doutre, 1983). The series of monoclonal anti-C2 antibodies which we have produced and characterized may be used to detect clinical C2-deficiency and also to analyse the genetic polymorphism of C2. We therefore believe that the monoclonal antibodies aeainst human comolement comuonent C2 may be a useful tool in further structural and functional analyses of C2 and also with regard to clinical evaluation of the involvement of complement component C2 in certain diseases. 1
1
Henning Sorensen kindly donated the outdated plasma for the preliminary purifications of C2. Dr Bendt Mansa performed the class and subclass typings of the monoclonal antibodies.
Acknowledgements-Dr
Kiihler G. and Milstein C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature,
Nature,
REFERENCES
Acad. Sci. U.S.A.
81, 1212-1215.
Carol1 M. C., Campbell R. D., Bentley D. R. and Porter R. R. (1984) A molecular map of the human major histocompatibility complex class III region linking comnlement eenes C4. C2 and factor B. Nature. Lond. 307. i37-241.Doutre M. S. (1983) Facteurs B, C2 et C4 du complement, systeme HLA et maladies. Path. Biol. 31, 622630. Engvall E. and Perlmann P. (1972) Enzyme-linked immunosorbent assay, ELISA. J. Immun. 109, 129-135. Harboe M. and Closs 0. (1983) Immunization with precipitates obtained by crossed immunelectrophoresis. Stand. J. Immun. 17, 353-359. Hem S. L. and White J. L. (1984) Characterization of aluminium hydroxide for use as an adjuvant in parenteral vaccines. J. Parenf. Sci. Technol. 38, 2-10. Hielm H., Hielm K. and Sieouist J. (1972) Protein A from -Staphyioc&cus aureus. -Its isolation by affinity chromatoeraohv and its use as an immunosorbent for isolatioc of: immunoglobulins. FEBS Left. 28, 73-76. Kerr M. A. (1981) The second component of human complement. Methods in Enzymology 80, (Edited by Lorand L.), p. 54. Academic Press, New York.
Lond. 221, 68G685.
Laurel1 C.-B. (1965) Antigen-antibody crossed immunoelectrophoresis. Analyt. Biochem. 10, 358-361. Nagasawa S. and Stroud R. M. (1977) Cleavage of C2 by Cls into the antigenically distinct fragments C2a and C2b: Demonstration of bindine of C2b to C4b. Proc. natn. Acad. Sci. U.S.A.
14, 299&3001.
Ngan B. Y. and Minta J. 0. (1981) Proteolysis of C2 and factor B: analysis of cleavage products by one- and two-dimensional peptide mapping. Molec. Immun. 18, 1035-1047. Nilsson L.-A. (1983) Double diffusion-in-gel. Stand. J. Immun.
Bentley D. R. and Porter R. R. (1984) Isolation of cDNA clones for human complement component C2. Proc. natn.
Land. 256, 495497.
Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
17, 57-68.
Porter R. R. (1984) The complement components of the major histocompatibility lo&s. CRC Crkal Rev. Biothem. 16. l-19. Porter R. R. and Reid K. B. M. (1979) Activation of the complement system by antibody-antigen complexes: the classical pathway., In Advances in Protein Chemistry 33 (Edited by Anfinsen C. B., Edsall J. T. and Richard F. M.), p. 1. Academic Press, New York. Sjiiholm A. G. and Sturfelt G. (1984) Cleavage of C2 in pathological serum and plasma studied by crossed immunoelectrophoresis. Acid path. microbial. immunol. stand. 92, 265-269.
Skjerdt K., Schou C. and Koch C. (1984) Assay for the specificity of monoclonal antibodies in crossed immunoelectrophoresis. J. Immun. Meth. 12, 243-249. Towbin H., Staehelin T. and Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. natn. Acad. Sci. U.S.A.
16, 435&4354.
Udea A., Kusaba T. and Yanase T. (1983) Determination of the second component of complement (C2) by electroimmunoassay in sera from patients with systemic lupus erythematosus. Clin. exp. Iimun. 53, 217-224. Woods D. E.. Edge M. D. and Colten H. R. (1984) Isolation of a complem&tary DNA clone for the human complement protein C2 and its use in the identification of a restriction fragment length polymorphism. J. clin. Invest. 74, 634638.
0161-5890/86 $3.00 + 0.00 Pergamon Journals Ltd
HUMAN COMPLEMENT COMPONENT C2: PRODUCTION AND CHARACTERIZATION OF POLYCLONAL AND MONOCLONAL ANTIBODIES AGAINST C2 EVA I. STENBAEK,* CLAUS KOCH,? VIBEKE BARKHOLT* and KAREN G. WELINDER* *Institute of Biochemical Genetics, University of Copenhagen, Oster Farimagsgade 2A, DK-1353
Copenhagen
K, Denmark and tThe Vaccine Department, DK-2300 Copenhagen S, Denmark
Statens
Seruminstitut,
(First accepted 28 October 1985; accepted in revised form 20 March 1986)
Abstract-Rabbit antibodies to human complement component C2 were produced by immunization of rabbits with precipitates from line immunoelectrophoresis, and the antibodies were used to monitor a classical chromatographic purification of C2 and for affinity purification of C2. Twelve monoclonal antibodies with specificity for human complement component C2 were produced by fusion of myeloma cells with spleen cells from mice immunized with the affinity purified C2. The specificity of the monoclonal antibodies was confirmed by their reaction with antigen-antibody precipitates where C2 was the antigen, and by their specific reaction with C2 after separation in SDS-PAGE followed by immunoblotting. The affinity of the monoclonal antibodies varied as demonstrated by the titration curves in ELISA. The antibodies will be of importance for immunospecific purification of human C2 and C2 fragments, for specific depletion of C2 from human serum, and for quantification of C2 for clinical purposes.
INTRODUCTION
inactivation. It is present in allelic forms with a common allele, C2C which has a gene frequency of 95%, with a basic variant, C2B which has a gene frequency of 4-5%, and with two rare acidic variants, C2Al and C2A2, which have gene frequencies of less than 1% (Woods et al., 1984). C2-deficiency is the most common of the complement deficiencies [O.Ol % of the population (Udea et al., 1983)]. The structural gene for C2 is linked to the MHC of all hitherto investigated mammals (Porter, 1984), and there is evidence that C2 from the classical complement pathway and factor B from the alternative pathway have evolved by gene duplication from a common ancestral gene. Functionally, C2 is engaged in the formation of the C3 and C5 convertases of the classical complement pathway. The availability of monoclonal antibodies to human C2 is important from several points of view. It will simplify the purification of substantial amounts of human C2 which in turn will facilitate further structural and functional analyses of human C2. The monoclonal antibodies may also be useful for the study of the interaction of C2 with other complement components during the formation of C3 and C5 convertases.
The complement component C2, which is found in human plasma in low quantities (15-20 mg/l), is involved in activation of the complement system via the classical complement pathway (Porter and Reid, 1979). This activation involves the binding of Cl to the Fc-part of antibodies in antigen-antibody complexes, leading to the formation of a Cl-esterase which as its substrates has C4 and C2. Following C4 cleavage, C2 (M, 100,000) is cleaved into two fragments, C2b (M, 30,000) and C2a (M, 70,000). C2a forms a Mg’+-dependent complex with C4b, and this complex, C4b2a, is the C3 convertase of the classical complement pathway. C2a carries the proteolytic site of the C3 convertase which causes activation of C3 by producing the fragments C3a and C3b. C3b binds to C4b2a, thereby forming the classical C5 convertase, C4b2a3b. The proteolytic site of this convertase is also located on C2a. C2 therefore plays a major role in the activation of the classical, antibody dependent complement pathway. C2 exists as a single chain glycoprotein and is very labile with regard to proteolytic cleavage and to heat Abbreviations: EDTA, ethylenediaminetetraacetate; ELISA, enzyme-linked immunosorbent assay; SDS-PAGE, polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate; PBS, phosphate-buffered saline, PH 7.3; DMEM, Dulbecco’s modified Eagle’s medium.
MATERIALS AND
METHODS
Antibodies
Antibodies to human serum proteins (A206) and peroxidase conjugated rabbit antibodies to mouse 879
EVA I. STENBAEK et
880
immunoglobulins (P260) were obtained from Dakopatts. Antibodies for typing and subtyping the immunoglobulin classes of the monoclonal antibodies, were purchased from Miles. Monospecific, polyclonal antibodies to human C2 were raised in rabbits after S.C. immunization with precipitates obtained from line immunoelectrophoresis, essentially as described by Harboe and Closs (1983) initially using C2 purified by classical chromatographic methods (Kerr, 1981) and antihuman C2 kindly donated by Dr M. Kerr, Dundee. Cell fusion protocol for the production of monoclonal antibodies to human C2 Balb/c mice were immunized i.p. with human C2, purified on an affinity column. The amount of antigen for each immunization was 30 p g/mouse, and the antigen was adsorbed onto Al(OH), (Hem and White, 1984). After two immunizations with an interval of 2 weeks the mice were bled and tested for anti-C2 reactivity by ELISA. Mice with high titres were selected for fusion, and their spleens were removed 3 days after a third immunization. Cell fusion was performed with the myeloma cell line X-63 Ag8 6.5.3 essentially as described by Kiihler and Milstein (1975). Fourteen days after fusion the cell culture supernatants were tested for specific reactivity against human C2, and antibody producing hybridomas were cloned until monoclonality was obtained by means of a limiting dilution method. The monoclonal antibodies were produced as culture supernatants in DMEM supplemented with 10% foetal calf serum. Purification of antibodies The IgG fraction of polyclonal rabbit antisera and the monoclonal antibodies from culture supernatants were purified by affinity chromatography on Protein A-Sepharose CL4B (Hjelm et al., 1972). Samples were adjusted to pH 8.0 before application to the column, and the bound antibodies were eluted with 0.1 M glycine-HCl, 0.5 M NaCl, pH 2.95. ELISA
described by Laemmli (1970) and the proteins were blotted onto nitrocellulose paper (Schleicher and Schuell) in 25 mM Tris-HCl, pH 9.0, 192 mM glytine, 20% (v/v) methanol for 5 hr at 10 V/cm at 8°C (Towbin et al., 1979). The nitrocellulose paper was washed and incubated with purified hybridoma antibodies and then incubated with peroxidase conjugated rabbit anti-mouse immunoglobulins. The peroxidase activity was assayed with H,Oz and 3,5,3’5’tetramethylbenzidine (Merck 8622) supplemented with dioctyl sodium sulphasuccinate (Sigma D-0885). Binding of monoclonal antibodies to precipitates in crossed immunoelectrophoresis Crossed immunoelectrophoresis was performed with human EDTA plasma supplemented with purified human C2 as antigen, against polyclonal anti-human serum, supplemented with rabbit monospecific polyclonal antibody to human C2 (Laurel& 1965). After electrophoresis the plates were washed and incubated first with purified hybridoma antibody and subsequently with peroxidase conjugated rabbit anti-mouse immunoglobulins. Binding of hybridoma antibodies to specific precipitates was detected by staining with H,O, and 3-amino,9-ethylcarbazole (Skjodt et al., 1984). Characterization of Ig class and subclass of monoclonal antibodies Class and subclass of the monoclonal antibodies were determined by Ouchterlony double immunodiffusion in agarose using monospecific rabbit antibodies against mouse IgGl, IgG2a, IgG2b, IgG3, IgA, IgM, kappa light chains and lambda light chains (Miles) (Nilsson, 1983). Immunosorbent column Monoclonal IgG (18 mg) from ascites fluid purified on Protein A-Sepharose 4B was coupled to CNBractivated Sepharose (8 g) yielding 28 ml of immunosorbent gel. RESULTS
Microtiter plates (Technunc) were coated overnight with purified human C2 (0.1 pgg/well) in PBS for direct ELISA, or with the IgG fraction of polyclonal anti-human C2 for indirect ELISA. The anti-C2 coated microtiter plates were incubated with purified C2 (0.1 pg/well). Microtiter plates coated with C2-deficient human serum were used as negative controls. The ELISA was performed essentially as described by Engvall and Perlmann (1972). Ig binding was detected using o -phenylenediamine (OPD) and the reactivity was measured at 492 nm by means of an MR 580 microelisa autoreader (Dynatech). SDS-PAGE
al.
and immunoblotting
Purified human C2 was analyzed by SDS-PAGE (5-15% polyacrylamide gradient gel) essentially as
Characterization of monospecific, polyclonal bodies to human C2
anti-
Two antibodies, C2-721 and C2-779, prepared as described in Materials and Methods were selected. C2-721 formed a sharp precipitate with C2 when analyzed in crossed immunoelectrophoresis, indicating a relatively high avidity for C2, whereas C2-779 showed a more diffuse precipitate with C2. Both antibodies were monospecific for human C2 when tested in crossed immunoelectrophoresis with human serum as antigen. When C2-721 was tested against human serum in which C2 had been activated by the classical complement pathway with heat aggregated human immunoglobulin, the position of the precipitate was shifted towards the anode as com-
Polyclonal and monoclonal antibodies against human complement component C2
881
Fig. 1. Crossed immunoelectrophoresis of human serum run against polyclonal, monospecific anti-human C2-721. (a) Human serum diluted 2:l in PBS. (b) Human serum activated with antigen-antibody complexes for 120 min, at 37°C in PBS (2: 1). After electrophoresis in the first dimension for 1 hr at 10 V/cm, antigens were run overnight in the second dimension at 2 V/cm. Both plates were stained with Coomassie Blue R-250.
pared to non-activated C2, indicating the presence of C2a, the proteolytic subunit of C2 (Nagasawa and Stroud, 1977) (Fig. 1). A$inity chromatography by polyclonaI antibody The low affinity antiserum C2-779 was used for affinity purification of human C2. The antibodies were purified on a Protein A-Sepharose column, and coupled to CNBr-activated Sepharose. A 1.5 x 10 cm column containing approx. 30 mg of coupled IgG was prepared. EDTA plasma (25 mi)was applied to the column and after thorough washing with phosphate buffer, pH 8.0, the column was eluted with 0.1 M glycineHC1, pH 2.8. Fractions were immediately neutralized and analyzed by immunoelectrophoresis against an antiserum to human serum proteins and an anti-human C2. The C2 preparation showed a minor contamination of human immunoglobulin. The eluate was analyzed for human complement factor B, but there was no reaction with rabbit anti-human factor B. The purified human C2 gave a sharp precipitate in crossed immunoelectrophoresis with anti-human C2 obtained from Dr M. Kerr, and with C2-721. When tested in SDS-PAGE, the C2 preparation showed a major band with M, of 100,000 (not shown). Production and characterization of monoclonal antibodies Mice were immunized with the immunosorbent purified human C2. We performed two separate cell fusions, and from the first fusion one positive clone was selected and from the second fusion 11 clones were selected for further cloning.
After at least two consecutive phases of cloning the produced monoclonal antibodies (designated C2-Sl through C2-S12) were characterized with respect to specificity in direct and in indirect ELISA, and to reactivity with C2/anti-C2 precipitates developed in crossed immunoelectrophoresis. Finally the reactivity of the monoclonal antibodies was tested by immunoblotting of SDS-gels of C2 preparations. Fig. 2(a) illustrates the SDS-PAGE analyses of preparations of C2 purified by affinity chromatography on the C2-Sl column. Fig. 2(b) demonstrates the reactivity and fragment specificity of the monoclonal antibodies. Two monoclonal antibodies C2-S3 and C2-S9, did not react with human C2 nor C2 fragments after C2 had been exposed for 5 min of boiling in SDS. This indicates that the antibodies might require native C2 to be reactive. Antibodies C2-S2, C2-S4, C2S5, C2-S7, C2-S8, C2-SlO, C2-Sll and C2-S12 reacted with C2 and with a split product which we assume to be C2b (M, 30,000). Antibody C2-SlO reacted weakly with C2 and with C2b, and is therefore not visible in Fig. 2(b), lane 10. The antibodies C2-Sl and C2-S6 reacted with C2 and a fragment which we assume to be C2a (M, 70,000), although C2-Sl reacts reluctantly with C2 and is therefore not seen in Fig. 2(b). C2S6 reacted notably stronger with C2a than with the intact C2. This suggests that C2-S6 reacts with an epitope which is more exposed or more stable in C2a than in C2. It is, furthermore, noteworthy that our rabbit anti-C2 sera, C2-721 from a late bleeding is almost exclusively directed against C2a (Fig. 2(b), lane 14). Figure 3 illustrates the reactivity of one of the
zooI Iis92-
E;fr45
31” 2114-
x 10-3
c2 Gh-
&2b-
-
94
-
67
-
43
-
30
-
20 14
Fig. 2. SDS-PAGE analysis of affinity purified hur~~ar~C2. (a) Different C2 preparations in a S-IS% gel stained by Coomassie Blue R-250, Protein markers are seen in the left lane. (b) Immunoblotting of B SDS-PAiSE gel with Ed& amounts of a partially degraded C2 sample in lanes 1-14. Nitrocelluiose prints of the individurtt &es were incubated with the different rno#~~o~a~ a&bodies agaiztst CZ _Antibo&es CZSI through CZ-S12 (lane l-12), monocIona1 mouse anti-& as a negative control (tane $31, and polyclonal anti-C2 (C2~721), as a positive control (lane 14). Protein markers stained with Amido Black are seen in the lane ‘M.
883
Polyclonal and monoclonal antibodies against human complement component C2
Fig. 3. (a) Cmssed imrnunoe~e~tro~~o~s~sofhmnan EDTA-plasma enriched with purified C2 as antigen and anti-human serum @?06, Dakopatts) supplemented with C&721 as antibody. The first dimension was run for 1hr at 10 V/cm, and the second dlmensmn was run overnight at 2 V/cm. After ekctrophoresia the gel was first incubated with monoclonai anti-C2 (C2-S4) and then with peroxidase-conjugated rabbit anti-mouse imm~~og~ob~i~. The C&m&C2 precipifafe stained by f&s procedure same CfE plate is stained with Coomassie Bhze R-250.
Three antibodies, C2-S2, C2-S9 and C2-S12, did not bind to CZ/anti-C2 precipitates in crossed imm~n~l~tro~hore~s~ Low aBnity or c~rn~ti~o~ with polyslnal antibodies for the same epitopes may explain this lack of reactivity. Antibody C2-S9 reacu ted neither with the C2/anti-C2 precipitate nor with the SIX-PAGE treated C2, The fact that this antibody is an IgM antibdy may explain the lack of reactivity. Table 1 summarizes the data for the 12 monoclonal antibodies. We found that the mono&onal antj~d~es show&
ELBA
Direct
cl-s1
0.340 0.410 a.120 a.290 0.620 O.lOU 0. I24 0.1 IO O.f4O 0.170 1.500 0.270
~,_,~~_“*” CZ-S2 C&S3
c2-s4 CZ-ST c2-scr c2-s7 CZ-SS C2-s CTSIO c2-SI 1 c2-s12
Indirect
1”1
Immunablotting
-“v-f f -
“Arbitrary concns of the 12 monoclon& ** + + imtkate3 the highs affinity. ND = Noi determined.
antibodies,
Fragment
C&W, subclass, type .,_
~,
CIE
Affinity’
specificity
Heavy chain
+ _
++ i++ -I+ i -6 +++ c ++ it+ -i+* +4-+ +++
C2a C2b ND C2b cztr c2tl C2b C2b ND C2b c2b C2b
IgOi ND ??a IgG2a ND IgQl IgOl I@I I$# rao1 IgGI ND
+ f f f + -
-f f f + f + -
+ -I+
-t + -
reacting
C2. (bj Tote
&ferent patterns of reactions when te&& in direct ELISA (C2 attached directly onto the microtiter plates) as compared to indirect ELISA (C2 was attached to a primary layer of rabbit polyclonaf a&X2>. The in~~d~~ values listed in Table I, shows that the ratio direct ELISAjindirect ELISA varies significantly, from > 1 (C2-S2, C2-S5, C2-S7, C2-Sll) through w 1 (C2-S3, C2-S4, C2-S6, C2-S9 and C2-SK!) to
SDS-PAGE/
0.920 0.030 8,160 0.370 0.260 0.119 0.019 1.500 A160 i.5M3 0.102 0.4ya
is mafked
..“”
with a given amount
of puritied
C2 in direct
and indkct
Light
chain
Ww ND ND Wv ND kw’a kww kwpa WV& Wpa kwpa ND ELBA.
884
EVA I. STENBAEKet al.
n
”
Polyclonal and monoclonal antibodies against human complement component C2 seem to be a correlation between the ability of the monoclonal antibodies to bind to precipitates of Cyanti-C2 in crossed immunoelectrophoresis and the ability to bind to C2/anti-C2 complexes on microtiterplates. Table 1 shows the results of antibody class and Ouchterlony immunosubclass typing by precipitation. All antibodies had kappa light chains and all but two antibodies were of the IgGl subclass. Four antibodies have not yet been typed. AJinity
of monoclonal
antibodies
The relative affinities of the monoclonal antibodies were estimated from the slopes of the curves in direct ELISA from a dilution series of each antibody (Fig. 4). Figure 4 and Table 1 show that the affinity varies greatly, a fact which is of importance for further use of the antibodies. Aj%ity
chromatography
by monoclonal
antibody
The low affinity antibody C2-Sl was selected for use in affinity purification of human C2. C2 was first purified partially by euglobulin precipitation and CM-Sephadex C-50 ion-exchange chromatography in the presence of benzamidine (Kerr, 1981) and then passed through the immunosorbent column. The column was washed thoroughly with PBS and C2 was eluted with 0.1 M propionic acid, 0.15 M NaCl, pH = 3.6. The C2 fractions were immediately neutralized (pH = 6.0) with 0.25 M NaOH. C2 was analyzed in SDS-PAGE as shown in Fig. 2(a), where the major component with M, 100,000 is the C2 band. The overall yield was estimated to 40% by amino acid analysis of the purified C2, assuming that normal human plasma contains 20mg/l of C2. DISCUSSION
The human complement component C2 has been purified previously by classical methods as described by Udea et al. (1983), Ngan and Minta (1981), Kerr (1981) and Nagasawa and Stroud (1977). C2 was purified by Udea et al. (1983) using affinity chromatography (polyclonal anti-C2 coupled to CNBractivated Sepharose 4B). Kerr (1981), Ngan and Minta (1981) and Nagasawa and Stroud (1977) obtained highly purified C2 by standard procedures, such as salt precipitation, ion-exchange chromatography and gel filtration. Low yields (1630%) are obtained with these procedures, which are very tedious and time consuming. The low yields and the gradual loss of biological activity during the laborious procedures made it tempting to produce monoclonal antibodies against C2 as a tool for purification and characterization of C2. We have purified C2 from plasma directly on an affinity column with polyclonal, low-affinity rabbit antibody C2-779. Relatively pure C2 which was useful for the immunization of Balb/c mice before cell fusion and also for the initial screening of the positive
885
culture supernatants for anti-C2 producing clones was easily obtained in this way. On the other hand, the limited yield and the presence of impurities in this C2 preparation showed no superiority when compared to classical fractionation procedures. Preliminary experiments with a monoclonal immunosorbent lived up to our expectation that monoclonal antibodies are extremely valuable for affinity chromatographic purification of C2. One of the monoclonal antibodies with low affinity for C2 (C2-Sl), has been used with success for the preparation of biologically active, highly purified C2 in 40% yield. This purification procedure is presently being optimized. In order to confirm the specificity of the monoclonal antibodies the following assays were applied: (1) immunoblotting of C2 separated by SDS-PAGE; (2) postincubation of CIE plates of human serum supplemented with purified C2 run against antihuman serum enriched with polyclonal anti-C2. As seen in Table 1, these assays confirmed the specificity of the monoclonal antibodies for C2 and also gave indications of variations in reactivity. Of the 12 clones producing monoclonal antibodies one produced an IgM antibody (C2-S9), probably of limited value. Ten of the antibodies reacted with C2 partially denaturated after SDS-treatment. These antibodies will be valuable for quantification and detection of C2 split products. The C2-S3 antibody, which does not react with SDS-treated C2, may be useful for the detection of native, biologically active c2. Affinity is an important qualitative feature of a monoclonal antibody. In general low affinity monoclonal antibodies are needed for affinity-purification of the specific antigen, whereas high affinity antibodies are superior with respect to specific removal or immunospecific detection of an antigen. Both types-high affinity and low affinity antibodies as judged from ELISA titration curves-have been found in our panel of monoclonal antibodies to human C2 (Fig. 4). For clinical purposes the antibodies may be employed in methods aimed at quantifying the plasma level of C2 and/or C2 split products and also for the detection of the cellular distribution of C2. It has been described previously that in systemic lupus erythematosus (SLE), heriditary angioedema (HAE) and acute post streptococcal glomerulonephritis (AGN), there is often a significant change in the amount of circulating C2 and C2 split products (Udea et al., 1983; Sjoholm and Sturfelt, 1984). As previously mentioned C2, which is an MHC class III antigen, occurs in allelic forms (Bentley and Porter, 1984; Carol], 1984). There are at present four known allelic genes coding for C2 that give rise to polymorphic forms of C2. It has been suggested that the polymorphism of the class III proteins may be a factor in susceptibility to certain diseases, in particumanifestations lar diseases with auto-immune
EVA I. STENBAEK et al.
886
(Doutre, 1983). The series of monoclonal anti-C2 antibodies which we have produced and characterized may be used to detect clinical C2-deficiency and also to analyse the genetic polymorphism of C2. We therefore believe that the monoclonal antibodies aeainst human comolement comuonent C2 may be a useful tool in further structural and functional analyses of C2 and also with regard to clinical evaluation of the involvement of complement component C2 in certain diseases. 1
1
Henning Sorensen kindly donated the outdated plasma for the preliminary purifications of C2. Dr Bendt Mansa performed the class and subclass typings of the monoclonal antibodies.
Acknowledgements-Dr
Kiihler G. and Milstein C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature,
Nature,
REFERENCES
Acad. Sci. U.S.A.
81, 1212-1215.
Carol1 M. C., Campbell R. D., Bentley D. R. and Porter R. R. (1984) A molecular map of the human major histocompatibility complex class III region linking comnlement eenes C4. C2 and factor B. Nature. Lond. 307. i37-241.Doutre M. S. (1983) Facteurs B, C2 et C4 du complement, systeme HLA et maladies. Path. Biol. 31, 622630. Engvall E. and Perlmann P. (1972) Enzyme-linked immunosorbent assay, ELISA. J. Immun. 109, 129-135. Harboe M. and Closs 0. (1983) Immunization with precipitates obtained by crossed immunelectrophoresis. Stand. J. Immun. 17, 353-359. Hem S. L. and White J. L. (1984) Characterization of aluminium hydroxide for use as an adjuvant in parenteral vaccines. J. Parenf. Sci. Technol. 38, 2-10. Hielm H., Hielm K. and Sieouist J. (1972) Protein A from -Staphyioc&cus aureus. -Its isolation by affinity chromatoeraohv and its use as an immunosorbent for isolatioc of: immunoglobulins. FEBS Left. 28, 73-76. Kerr M. A. (1981) The second component of human complement. Methods in Enzymology 80, (Edited by Lorand L.), p. 54. Academic Press, New York.
Lond. 221, 68G685.
Laurel1 C.-B. (1965) Antigen-antibody crossed immunoelectrophoresis. Analyt. Biochem. 10, 358-361. Nagasawa S. and Stroud R. M. (1977) Cleavage of C2 by Cls into the antigenically distinct fragments C2a and C2b: Demonstration of bindine of C2b to C4b. Proc. natn. Acad. Sci. U.S.A.
14, 299&3001.
Ngan B. Y. and Minta J. 0. (1981) Proteolysis of C2 and factor B: analysis of cleavage products by one- and two-dimensional peptide mapping. Molec. Immun. 18, 1035-1047. Nilsson L.-A. (1983) Double diffusion-in-gel. Stand. J. Immun.
Bentley D. R. and Porter R. R. (1984) Isolation of cDNA clones for human complement component C2. Proc. natn.
Land. 256, 495497.
Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
17, 57-68.
Porter R. R. (1984) The complement components of the major histocompatibility lo&s. CRC Crkal Rev. Biothem. 16. l-19. Porter R. R. and Reid K. B. M. (1979) Activation of the complement system by antibody-antigen complexes: the classical pathway., In Advances in Protein Chemistry 33 (Edited by Anfinsen C. B., Edsall J. T. and Richard F. M.), p. 1. Academic Press, New York. Sjiiholm A. G. and Sturfelt G. (1984) Cleavage of C2 in pathological serum and plasma studied by crossed immunoelectrophoresis. Acid path. microbial. immunol. stand. 92, 265-269.
Skjerdt K., Schou C. and Koch C. (1984) Assay for the specificity of monoclonal antibodies in crossed immunoelectrophoresis. J. Immun. Meth. 12, 243-249. Towbin H., Staehelin T. and Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. natn. Acad. Sci. U.S.A.
16, 435&4354.
Udea A., Kusaba T. and Yanase T. (1983) Determination of the second component of complement (C2) by electroimmunoassay in sera from patients with systemic lupus erythematosus. Clin. exp. Iimun. 53, 217-224. Woods D. E.. Edge M. D. and Colten H. R. (1984) Isolation of a complem&tary DNA clone for the human complement protein C2 and its use in the identification of a restriction fragment length polymorphism. J. clin. Invest. 74, 634638.