STUDIES ON THE HUMAN COMPLEMENT PROTEIN Clq Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSiTY CARLOS R. SLEDGE 1972 Date 0-7639 (at «x Hula} I I, "3 R A R Y Michigan mute University This is to certify that the thesis entitled STUDIES ON THE HUMAN COMPLEMENT PROTEIN Clq presented by Carlos Renaldo Sledge has been accepted towards fulfillment of the requirements for Ph . D . Je in Microbiology M150: prof file/7,1 800K BlNDERY INC. UBRARY BIHD'E RS seamseear Hartman NIL: -..'_ -~ "'2. I * — amomc BY Ej‘q ' "MG & SUNS' i """l A $1 Clq, was The chron Sepharose 0f IgM it obtained in sucros iZed imrnu mobility material ‘ Clr and C The was quant 19”: and 3“ IgM-se were dete and 196 = incteased ABSTRACT STUDIES ON THE HUMAN COMPLEMENT PROTEIN Clq BY Carlos R. Sledge A subunit of the first component of human complement, Clq, was purified by the technique of affinity chromatography. The chromatographic resin was cyanogen bromide activated Sepharose covalently linked to human IgG. To remove traces of IgM it was found necessary to further subject the Clq obtained from the chromatographic step to ultracentrifugation in sucrose gradients. The highly purified Clq was character- ized immunochemically and according to its electrophoretic mobility in various polyacrylamide gel systems. The purified material was capable of combining with a reagent containing Clr and C13 to reconstitute fully active macromolecular C1. The interaction between human Clq and immunoglobulins was quantitatively measured by determining the ability of IgG, 125 IgM, and (Fc)5u to inhibit the binding of I-labeled Clq to an IgM-sepharose complex. The following inhibition constants were determined: IgM = 6.42 x 10"6 M; (Fc)5u = 4.35 x 10-6 M; and IgG = 1.10 x 10-4 M. The heat aggregation of IgM and IgG increased the ability of these proteins to bind 125I-labeled Carlos R. Sledge Clq but had no significant effect on the binding properties 125I-labeled Clq and the of (Fc)5u. The binding between IgM-sepharose complex was inhibitable with various small molecular weight diamino compounds. The most potent inhibitor studied was 2,5-diaminotoluene. STUDIES ON THE HUMAN COMPLEMENT PROTEIN Clq BY Carlos R;’Sledge A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1972 I a Public H while I I a Dr. Davj Plied (it thank t1 PIOVidec Kathy M: Mernitz‘ My Support Valuab 1 ACKNOWLEDGEMENTS I am grateful to the Department of Microbiology and Public Health for providing me with financial assistance while I pursued the studies for my degree. I am extremely thankful to my major professor, Dr. David Bing, for the assistance and encouragement he sup- plied during the period of my research. I would also like to thank the following peOple for the valuable assistance they provided in reading the thesis and discussing the data: Kathy Morris, Dr. Harold Miller, Sandra Spurlock, Janice Mernitz, and Dr. Kristine Knudson. My father and mother, Mr. and Mrs. Edward Sledge, Sr., have provided me a continual source of moral and financial support and I am extremely grateful to them for their in- valuable efforts. ii TABLE OF CONTENTS INTRODUCTION. 0 O O O O O 6 0 3 O 0 O O O O O C O O O 0 LITERATURE REVIEW 0 O O O O O O O O 0 O O O O O O 0 O 0 PART I. The Complement System. . . . . . . . . . . . . Introduction. . . . . . . . . . . . . . . . Nomenclature of Complement. . . o . . . . . The Complement Sequence . . . . . . . . . . Alternate Pathways into the Compl ment System . . . . . . . . . . . . . . . . Biologically Active By—products o the Com- plement System . . . . . . . . . . . . . II. Purification of the First Component of Comple- ment. 0 o o o o o o o o o o a o o o o o o 0 III. Purification and Properties of Clq . . . . . . IV. The Interaction Between Cl and Immunoglobulins REFERENCES. . . . . . . . . . . . . . . o . . . o . o . ARTICLE 1 Purification of the Human Complement Protein Clq by Affinity Chromatography. C. R. Sledge and D. H. Bing. Manuscript submitted to Immunochemistry . . . . . . . . . . . . . . . . ARTICLE 2 Binding PrOperties of the Human Complement Protein Clq. C. R. Sledge and D. H. Bing. Manuscript to be submitted to Journal of Biological Chemistry . O O O O O O O O 0 O O O O O O O O O CONCLUDING REMARKS O O O C 3 O O O O O O I 0 O O O O O 0 iii Page manta-p oo 12 l3 17 19 22 30 56 89 TABLE LIST OF TABLES Page ARTICLE 1 Purification of Clq on IgG-Sepharose. . . . . . 4l Ability of Purified Clq to form Macromolecular Cl. 0 O I O O O O O 0 9 O 0 O C O O O O O O O O 43 ARTICLE 2 The Inhibition Constants (Ki) and Relative Free Energy Values (AF') for IgG, IgM, and (Fc)5u° . 72 iv FIGURE Re ca Ac Te ce Irr FIGURE LIST OF FIGURES LITERATURE REVIEW The complement sequence. . . . . . . . . . . . . Relationship of alternate pathways with the com- plement sequence . . . . . . . . . . . . . . . . ARTICLE 1 Chromatography of the euglobulin fraction of germ 0n EGG-sepharOse o o o o o o o o o o o o o Acrylamide gel electrophoresis analysis of Clq . Ten to forty percent sucrose gradient ultra- centrifugation of Pool q . . . . . . . . . . . . Immunodiffusion analysis . . . . . . . . . . . . ARTICLE 2 125 The binding of I-labeled Clq to the IgM- sepharose complex. . . . . . . . . . . . . . . . Sucrose gradient (10—40%) ultracentrifugation of IgM, (Fc)5u, and IgG . . . . . . . . . . . . . . The inhibition of the binding of 125I-labeled Clq to the IgM-sepharose complex by IgM, (Fc)5u, IgG, and Bovine Serum Albumin. . . . . . . . . . The inhibition of the binding of 125I-labeled Clq to the IgM-sepharose complex by monomeric and heat treated IgM, (Fc)5u, and IgG. . . . . . The binding of l-[4-14C]-diaminobutane to Clq. . 4O 45 48 50 66 69 71 75 LIST OF FIGURES--Continued FIGURE 6. Page The inhibition of the binding of 125I—labeled Clq to the IgM—sepharose complex by diamino alkyl compounds. . . . . . . . . . . . . . . . . 80 The inhibition of the binding of 125I-labeled Clq to the IgM-sepharose complex by diamino aromatic compounds . . . . . . . . . . . . . . . 82 vi INTRODUCTION The complement system consists of nine components and eleven proteins which interact sequentially with each other to mediate cellular injury and promote the inflammatory response. This system is activated by interaction of the first component (Cl) with antigen-antibody complexes. This interaction is mediated by a sub-component of Cl, Clq, which binds to the Fc portion of the immunoglobulin. Due to the binding affinity of Clq for immunoglobulins many investigators believe this molecule to be an antigamma globulin. Clq also binds C15 and Clr in the presence of Ca++ to form the Cl macro- molecular complex. Thus this molecule is of utmost importance since it represents a link between immunoglobulins and the complement effector system. The immunoglobulins possess the capacity to Specifically recognize and bind a foreign cell through sites resident in the Fab portion of the molecule. The Fc region of the immunoglobulin is thereby positioned so it can efficiently bind complement. This event results in the activa- tion of the complement effector system followed by enhanced destruction of the cell. Previous studies performed on the Clq immunoglobulin interaction were focused on the complement binding sites residen ployed based 0 only tw However meric f if this the mol on the ' aggrega study wa and immu of the 5' To necessar‘ large qu. tOgraphy of b10108 erties. COValem: be extrac fied resi tOgraphy ni Clue fOr Obtained :0u1d Sub resident in the immunoglobulin. Most of these studies em- ployed functionally purified macromolecular C1, and were based on complement fixation assays. IgG and IgM were the only two immunoglobulins shown capable of binding Cl. However, there are conflicting reports on whether the mono- meric forms of these immunoglobulins are able to bind C1 or if this property is only induced by physical aggregation of the molecule. Relatively few studies have been performed on the interaction of purified Clq with monomeric and heat aggregated immunoglobulins. The purpose of the present study was to quantitatively analyze the binding between Clq and immunoglobulins and to obtain information on the nature of the site on Clq responsible for this interaction. To accomplish the objectives of the study it was first necessary to develop a procedure which could be used to obtain large quantities of highly purified Clq. Affinity chroma- tography has been shown useful in the purification of a number of biologically active substances with distinct binding prop- erties. This technique employs a specific binding ligand covalently attached to a resin. The protein of interest can be extracted from crude fractions by adsorption to the modi- fied resin. The present study shows that affinity chroma- tography using an IgG-sepharose resin is an efficient tech- nique for the purification of Clq. The Clq preparations obtained by this procedure were found to contain IgM which could subsequently be removed by ultracentrifucation. The binding {i The assay posed of molecularl to bind 1 sepharose formed to immunoglc compounds This is a lite complemen 910bulins SiSt of t the PUrif munog lo The Clq obtained by this procedure was used to study the binding properties of this molecule for immunoglobulins. The assay system developed to study this interaction was com— posed of iodinated Clq, an IgM—sepharose complex, and known molecular forms of IgG and IgM. The ability of IgG and IgM to bind lzsI-labeled Clq and inhibit its binding to the IgM- sepharose complex was determined. Experiments were also per- formed to define the nature of the Clq binding site for immunoglobulins according to the ability of various diamine compounds to inhibit the Clq—immunoglobulin interaction. This thesis is organized into four sections. The first is a literature review in which pertinent information on the complement system, the techniques used for purification of C1, the properties of Clq, and the interaction of Cl with immuno- globulins is presented. The second and third sections con- sist of two manuscripts submitted for publication and concern the purification of Clq and the binding properties of Clq for immunoglobulins. The fourth section, the concluding remarks, is a brief discussion of the Clq molecule and the implications of the present study. LITERATURE REVIEW Part I The Complement System Introduction. The term complement refers to a series of serum proteins which play a vital role in the host response to foreign substances. Interaction with antigen-antibody complexes causes the activation of this series of enzymes into an active form. The occurrence of these events on the surface of cells results in the production of ultrastructural lesions in the cell membrane and the eventual lysis of the cell (1). In addition to the cytolytic effects the complement proteins also promote other events in the inflammatory response such as histamine release, chemotaxis, contraction of smooth muscle, enhanced vascular permeability and increased phagocytosis (2). The following review will be concerned with the nomenclature of the complement system, the reaction sequence of the comple- ment proteins, techniques which have been employed for the purification of C1, and the interaction between Clq and immunoglobulins. Nomenclature of Complement. The complement system con- sists of nine components and eleven distinct proteins. The terminology used for the complement system is that suggested by the 'v refers treated the addi (C) (4,5 The C9. C1 Intermed. designate which hax refers to and c4. ated by j Ponents' rule Pla< that the Fragment1 ment enzj C3b are 1 POnent O‘ by the World Health Organization (3). This nomenclature refers to a model system consisting of sheep erythrocytes (E) treated with rabbit antibody (A) which are then lysed upon the addition of fresh normal serum as the source of complement (C) (4,5). The nine complement components are numbered Cl through C9. C1 consists of three distinct proteins, Clq, Clr, and C18. Intermediate reaction products formed by E, A, and C are designated EAC followed by the number of those components which have interacted. For instance, the EACl,4 notation refers to cell antibody complexes which have reacted with Cl and C4. The symbols for intermediate complexes may be abbrevi- ated by indicating only the first and the last reacting com- ponents, e.g., EAC1,4,2,3 may be written EACl-3. A bar or rule placed over a complement component is used to indicate that the component is in an active enzymatic state, e.g., CI. Fragments of components resulting from cleavage by other comple- ment enzymes are denoted with a small letter, e.g., C3a and C3b are two fragments of C3. The loss by a complement com- ponent of a defined activity is denoted by the suffix "i". The Complement Sequence. A diagram of the complement sequence is shown in Figure l. The first component of comple- ment, Cl, is activated by interaction with an antigen-antibody complex. The three proteins, Clq, Clr, and C13, which comprise Cl require'Ca++ to function as a unit (6). Clq contains the .oocoswom ucoEonEoo one .H ousmflm saw I m.m.n.m.nm.mm n~.ne.moam + mo.mu + n.o.nm.mh.n .na.moflu060Hm m0 .oocosvom usoanmEoo osu QDHB mmmzaumm oumcuouam mo mflzmcofiumHom .m musmflm 23 \. mmHUH wo.AIIiHo NO + v0 e Ammmuuo>coo muv m.v0 .1 mu nmu.W mmu fluo< uoum>fluo¢ mo xoamEoo moo I mmmu e. ommuuo>coo Aooo.ova u 32v oemucooumlllYfloum>fluomoum mQIV. .+lamouv Houomm Eoco> munoo Aooo.om u 32v “ammoc noum>fiuomonm mo 10 The experiments with cobra venom factor were important in elucidating the serum protein C3 proactivator; however, they yielded no immediate clue to the physiological signifi- cance of this protein. Later studies showed that when serum was treated with naturally occurring plant or bacterial poly- saccharides the C3 proactivator was cleaved into two frag- ments with molecular weights of 60,000 and 20,000. The larger fragment had the ability to cleave C3 into C3a and C3b and was termed C3 activator. It was postulated that an unidenti- fied serum enzyme, C3 proactivator convertase, was responsi- ble for cleaving C3 proactivator (23,28). A recent study has demonstrated that a 95 hydrazine sensitive factor isolated from human serum interacts with a serum 35 alpha globulin and this complex is capable of generating C3 activator from puri- fied C3 proactivator (29). It was proposed that the 3s alpha globulin was in fact the C3 proactivator convertase. In 1954 Pillemer.and associates described the properdin system as a group of normal serum factors which interacted with zymosan and other polysaccharides and effected the inactivation of C3 (30,31). The properdin system was shown to play a role in several properties of normal serum such as the killing of certain Gram-negative bacteria, neutralization of certain viruses, and lysis of erythrocytes from patients with paroxysmal nocturnal hemoglobinuria (32,33,34). The formation of a zymosan complex capable of inactivating C3 required, in addition to prOperdin and Mg++, two serum prote: facto: labile comple labile S din wa; from t] proper: properd thereby PIOperd Chemica indicat. The serum p] These ir biologic of 5-28, electr0: preparat 19A, or i Re \— iiu. A 11 proteins designated factor A and factor B. It was shown that factor A was hydrazine sensitive and that factor B was heat labile. These factors were shown to be distinct from the complement components C4 (hydrazine sensitive) and C2 (heat labile) (30,35,36,37). Several investigators challenged the concept that proper- din was a distinct serum protein with properties different from the immunoglobulins (38,39). Nelson (39) proposed that properdin was a natural antibody to zymosan, and that the properdin-zymosan complex first activated C1, C4, and C2 thereby effecting the inactivation of C3. At that time properdin was not available in a homogeneous form, and physio- chemical characterization of a preparation containing properdin indicated that it was a 19S gamma globulin. The physiochemical proof that properdin was a unique serum protein was obtained in 1968 by Pensky et 31. (34). These investigators showed that a homogeneous preparation of biologically active properdin had a sedimentation coefficient of 5.28, a molecular weight of 223,000 and exhibited the electrophoretic mobility of a beta globulin. In addition, the preparation contained no antigenic determinants to IgM, IgG, 19A, or any of the complement components. Recent studies have also shown factor B of the properdin system to be functionally identical to the C3 proactivator (41). Additional studies are required to show the identity of other the Pr‘ g; During activat enzymat cases, leading Hm sma related fragmen cally e. Permeab: generate larger 3 Phenome] CORQIUt: Cl: c frame!“ 12 other factors of the C3 proactivator system with factors of the prOperdin system. Biologically_Active By-products of the Complement System. During the complement sequence many of the components become activated from a proenzyme state to an active enzyme by an enzymatic cleavage which produces two fragments. In most cases, the larger fragment is utilized in the reaction sequence leading to cellular injury mediated by antibody. In some cases the smaller fragment possesses biological activity which is related to the inflammatory response. Thus these smaller fragments may possess anaphylatoxin activity which is classi- cally evidenced by smooth muscle contraction, increased vascular permeability, and the release of histamine. They may also generate chemotaxis for polymorphonuclear leukocytes. The larger fragment may also participate in other biological phenomena such as immune adherence, enhanced phagocytosis, conglutination, and immunoconglutination (2,4,7). Cleavage of C3 by CIT??? results in the production of two fragments, C3a and C3b. The C3b fragment (180,000 molecular weight) when bound to the cell appears to be responsible for immune adherence and enhanced phagocytosis (42,43). The other fragment C3a (6,800 molecular weight) has been shown to possess anaphylatoxin activity (44,45), chemotactic activity, and cause the degranulation of mast cells (2,42). The cleavage of C5 by either C5 peptidase (CI747273) or trypsin results in the formation of two fragments. The smaller frag (46) dist have will CSa mast Cl it and i gator; Prote: PIOYeC Only 1 tion 0 13 fragment, C5a, has been shown to have anaphylatoxin activity (46), and to be chemotactic for leukocytes (2). The CSa is distinct from C3a not only in molecular weight but they also have distinguishable biological prOperties. C3a but not CSa will release histamine from rat peritoneal mast cells, while CSa is much more highly active in degranulating guinea pig mast cells (42,45). The C37677 trimolecular complex has been shown to possess neutrophil chemotactic activity (47). The interaction of the CS76T7 chemotactic factor with the neutrophil results in the activation of a proesterase in the membrane of neutrophils and this enzyme appears to be essential for cell migration (48) 0 Part II Purification of the First Component of Complement To gain a clearer understanding of the mode of action of C1 it has been necessary to purify the macromolecular complex and its subcomponents, Clq, Clr, and C15. A number of investi- gators have taken upon themselves the task of purifying these proteins and a variety of technical procedures have been em- ployed for this purpose. These efforts, however, have met with only limited success, and in no case has a homogeneous prepara- tion of these proteins been obtained. Some of the difficulties these investigators have encountered in the purification of Cl and gen ati age hax be of 001 sex rel lar. dis: rep} con; tion englc Socia l4 and its components are: l) The intimate association between gamma globulins and C1 in the blood serum (44), 2) The dissoci- ation of the Cl macromolecular complex by the use of chelating agents does not appear to be complete and harsher conditions have to be employed to attain complete dissociation (50), 3) The components are very unstable at room temperature, dur- ing extended storage periods, and at ionic strengths above 0.15 (49,50,51,52). Borsos and Rapp (53) used ion exchange chromatography on diethylaminoethyl cellulose columns to prepare functionally pure guinea pig CI. This material contained 50-100% of the CI activity present in whole serum, however, it was found to be contaminated with C3. Nelson 35 El° (54) outlined a series of procedures used in the purification of the nine complement components of guinea pig serum which included the isolation of CI by low ionic strength, neutral pH precipitation of the serum. This procedure apparently yielded CI preparations relatively free of other complement components but containing large amounts of other gamma globulin proteins. It was later discovered that this CI could be further purified by either reprecipitation or by absorbing the CI with an antigen-antibody complex (55). Colten 35 a1. (56) purified CI by zonal ultracentrifuga- tion. This procedure involved the centrifugation of the euglobulin fraction of serum at a high ionic strength to dis- sociate the CI molecule and thereby separate it from heavier 15 contaminating proteins. The CI was then recentrifuged under low ionic strength conditions which allowed the CI to sediment as a 198 molecule which could then be separated from lighter material in the preparations. Linscott (57) tried using Sephadex G-200 for the purification of CI, but found the yields of protein too low from this technique for it to be a useful tool. He was, however, successful in purifying CI by chromatography on Bio-Gel P-200. Hoffman (58) took advantage of the insolubility of CI at low ionic strengths and purified the molecule by solubility chromatography on Bio-Gel P-lO resins. Analysis of these preparations by polyacrylamide gel electrophoresis and immunelectrOphoresis showed that the CI was not homogeneous. The disc gel electrophoresis was per- formed in 7.5% polyacrylamide gels which are not penetrable by CI, consequently they are not sufficient for adequate analysis of this molecule. The specific binding of CI to gamma globulin suggested to Bing (59) that the technique of affinity chromatography would be useful for the purification of CI. Using gamma globulin linked sepharose resins it was possible to absorb CI from the euglobulin fraction of human serum. The absorbed CI could be eluted from the resins in very high yields using 1,4-diaminobutane. The CI purified by this procedure was ten times more active than that prepared by any previously reported procedure. The purification of CI by affinity chromatography takes advantage of a property that is not only n: 16 unique but is also intimately associated with the biological activity of CI. Other procedures employed for CI purifica- tion rely on physico-chemical properties of the molecule (e.g., mol. wt. and charge characteristics) which may overlap those possessed by other functionally unrelated molecules. The CI macromolecule sediments as a 188 molecule, and its three subcomponents Clq, Clr, and C13 sediment as 118, 78, and 4S molecules, respectively (60). The resolution of human Cl into three distinct proteins was first accomplished by Lepow st 31. (6). The components were eluted from a diethylamino- ethyl cellulose ion exchange resin in the order, Clq, Clr, and C15. The develOpment of purification procedures and the description of biochemical characteristics has proceeded more rapidly for C13 and Clq than for Clr. The discovery that human and guinea pig Cl contained a proesterase (Cls) was reported by Lepow 23 31. (61,62) and Becker (63). Haines and Lepow (64,65,66) later described in detail the purification and properties of Cls. The C1; was isolated by a combination of diethylaminoethyl and triethylaminoethyl ion exchange chroma- tography, and was purified 2400 fold with respect to serum. Cls has also been isolated by a combination of carboxymethyl cellulose and diethylaminoethyl cellulose chromatography and zonal electrOphoresis (67). The Specific activity and percent recoveries of the C13 were not reported so it is difficult to determine how valuable this technique is for the isolation of large quantities of Cls. The preparation was, however, judged 5e; elu pur fic: Alth oroc as? bet 17 to be highly purified using mobility in polyacrylamide gels as a criteria of purity. Cls has also been purified by affin- ity chromatography (68). The resin used for absorption was sepharose coupled to meta-aminobenzamidine. The enzyme was eluted using both propionic and acetic acids. The Cls purified by this procedure was comparable in terms of speci- ficity to the C15 purified by ion exchange chromatography (64). Recently Clr has been obtained in a highly purified state (52). It was isolated from human serum by ion exchange chromatography on diethylaminoethyl- and carboxymethyl-cellulose and a final preparative polyacrylamide electrophoresis step. Although the recovery of Clr activity was low using these procedures, it was possible to determine its molecular weight as 168,000 and its behavior on electrOphoresis was that of a beta globulin. Part III Purification and Properties of Clq The first successful isolation of Clq was achieved by absorbing the molecule out of serum using heat aggregated gamma globulin (69). The SZO,w of this molecule was determined to be 11.18 and its activity was labile at 56 C for 30 min. This technique for purifying Clq has found only limited applicability in various laboratories (4) due to the difficulty in preparing soluble gamma globulin aggregates. An alternative method has be 992 co: 01') YO tr. we} tat bui 18 been elaborated which utilizes ion exchange chromatography, gel filtration, and electrophoresis (70). The first step consists of chromatography of the serum euglobulin fraction on carboxymethyl cellulose followed by filtration on Sephadex G-200 and the final step involved separation of the molecule by Pevikon block electrOphoresis. A much simpler technique for the purification of Clq has recently been reported by Yonemasu and Stroud (49). In this procedure various concen- trations of EDTA and EGTA (range 0.026 M-0.06 M) were used to precipitate Clq from whole human serum. These operations were conducted at low ionic strength and the Clq was reprecipi- tated three times and resolubilized in a high ionic strength buffer. The ultrastructural characterization by electron micros- copy of the Clq prepared by the procedure of Yonemasu and Stroud (49) showed the molecule to consist of six terminal subunits attached to a central unit by flexible connecting strands (71). Ultrastructural analysis of Clq prepared by the procedure of Calcott and Muller-Eberhard (70) showed the molecule to be composed of five pentagonal subunits non- covalently connected to a central pentameric subunit (1,72). The marked difference in the shape of the molecule as reported in these communications may have been caused by breakdown of the molecule during preparation for microscopy. The amino acid analysis of Clq showed it to contain a large amount of hydroxyproline, hydroxylysine and glycine. The The to Cl; no] no] 42, the is men 19 The total carbohydrate content of the molecule was 7.7%. The data indicated that Clq has a structural content similar to basement membrane proteins rather than serum proteins (73). Clq is composed of two non-covalently linked subunits with molecular weights of 60,000 (I) and 42,000 (II). The intact molecule contains 6 of the 60,000 m. w. units and 2 of the 42,000 m. w. subunits (74). Muller-Eberhard (75) has reported that Clq has 5 or 6 binding sites for immunoglobulins so it is possible that the 6 subunits are distinct binding sites for immunoglobulins and the 2 smaller subunits are sites of attach- ment for Clr and C15. Part IV The Interaction Between Cl and Immunoglobulins The binding characteristics of Clq are interesting for not only is it capable of binding immunoglobulins (75,76) but it also interacts with C15 and Clr (9) polynucleotides (77,78, 79) and sulphated polysaccharides (80,81). Studies performed on the binding of Clq by immunoglobulins have served to eluci- date the combining regions on the immunoglobulins. The comple- ment binding site was initially localized to the Fc fragment (82). However, Amiraian and Leikhim (83) showed that the F(ab)2 fragment was also capable of binding complement. Schur and Becker (84) showed that SS antibody-antigen complexes were capable of fixing 40% of guinea pig complement. The remaining 20 60% could be absorbed by 78 antibody-antigen complexes but resisted fixation by a fresh preparation of 5S antibody. This finding suggested the occurrence of two distinct sites for C1 in the gamma globulin molecule--one located in the Fc region and the other in the Fab portion. Griffin gt 31. (85) found that labeling of a few trypto- phan residues in rabbit gamma globulin with 2-hydroxy-5-nitro- benzylbromide decreased the complement binding ability of this molecule while having no effect on the antibody combining site. Cohen and Becker (86,87) extended these studies and demon— strated that the sequential amidination and benzylation of gamma globulin significantly altered the combining sites for complement. It was proposed that these treatments altered the lysine and tryptophan residues located in the complement bind- ing site. It has also been shown that the binding of C1 to immunoglobulins can be inhibited by diaminoalkyl compounds (88). IgM and IgG are the only classes of immunoglobulins which have been shown to possess complement binding sites. IgA, IgE, and 19D do not have this ability (89,90,91). The subclasses of IgG exhibit differences in their capacity to bind Clq, with 19GB being the most active, followed in order by IgGl and IgG2. The IgG4 subclass is incapable of binding Clq (89). A differ- ence in complement binding efficiency has also been observed within the IgM immunoglobulin class. Linscott and Hansen found an increase in the number of non-complement fixing guinea pig IgM antibodies with time after immunization (92). 21 The human monoclonal IgM proteins have also been grouped into two subclasses. One group interacts with Clq and the second group fails to interact with this factor (93,94). Various studies performed on the binding of complement by immunoglobulins have yielded conflicting data regarding the capacity of monomeric immunoglobulins to bind complement. It has been found that single molecules of IgG in contact with a red cell surface are incapable of binding C1. At least two molecules of IgG in close proximity on the cell surface are necessary to bind one molecule of Cl (95,96). Ishizaka at 31. (97) demonstrated that monomeric IgG and IgM were not capable of binding Cl. Aggregation of these immunoglobulins with bis- diazotized benzidine induced complement binding properties. Hyslop £3 31. (98) used IgG immunoglobulins aggregated by reaction with a divalent hapten to determine what molecular size of immunoglobulin was necessary to bind complement. These studies revealed that monomers, dimers, and trimers of IgG were incapable of binding C1. However tetramers and higher polymers had the ability to bind Cl. It was suggested that aggregation of immunoglobulins causes a quartenary structural change in the molecule which exposed the complement binding site. Augener gt a1. (89) extended these studies and found that heat aggregation of immunoglobulins did enhance their complement binding ability, however, monomeric IgG and IgM were also capable of binding Cl. Recently, Plaut, Cohen, and Tomasi (99) found that both monomeric IgM and (Fc)5u are capable of binding C1. 10. REFERENCES POLLEY, M. J. (1971) Progress in Immunology (Amos, B., Ed.), Academic Press, New York, pp. 597-608. LEPOW, I. H. (1971) Progress in Immunology (Amos, B., Ed.), Academic Press, New York, pp. 579-595. AUSTEN, K. F., BECKER, E. L., BORSOS, T., LACHMANN, P. J., AND LEPOW, I. H. (1968) Bull. WHO 39, 935-938. MULLER-EBBRHARD, a. J. (1968) Adv. Immunol. 8, 1-80. MAYER, M. M. (1965) CIBA Foundation Symposium on Comple- E222 (Wolstenholm, G. E. W., and Knight, J., Eds.), Churchhill Ltd., London, pp. 4-57. LEPOW, I. H., NAFF, G. B., TODD, E. W., PENSKY, J., AND HINZ, C. F. (1961) J. Exp. Med. 117, 183-1008. MULLER-EBBRHARD, H. J. (1971) Progress in Immunology (Amos, B., Ed.), Academic Press, New York, pp. 553-565. NAFF, G. B., AND RATNOFF, 0. D. (1968) J. Exp. Med. 128, 571-593. RATNOFF, O. D., AND NAFF, G. B. (1969) J. Lab. Clin. Med. 74, 380-387. PATRICK, R. A., TAUBMAN, S. B., AND LEPOW, I. H. (1970) Immunochemistry 7, 217-225. 22 ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 23 MULLER-EBBRHARD, H. J., AND LEPOW, I. H. (1965) J. Exp. Med. 121, 819-833. BECKER, E. L. (1960) J. Immunol. 84, 299-308. SITOMER, G., STROUD, R. M., AND MAYER, M. M. (1966) Immunochemistry 3, 57-69. GIGLI, 1., AND AUSTEN, K. F. (1969) J. Exp. Med. 129, 679-696. MULLER-EBERHARD, H. J., POLLEY, M. J., AND CALCOTT, M. A. (1967) J. Exp. Med. 125, 359-380. MULLER-EBERHARD, H. J. DALMASSO, A. P., AND CALCOTT, M. A. (1966) Jo EXE. Med. 132' 33-54. COOPER, N. R., AND MULLER-EBERHARD, H. J. (1970) J. Exp; Med. 132, 775-793. SHIN, H. S., PICKERING, R. J., AND MAYER, M. M. (1971) STOLZ, R. (1968) J. Immunol. 100, 46-54. PICKERING, R. J., WOLFSON, M. R., GOOD, R. A., AND GERWURZ, H. (1969) Proc. Natl. Acad. Sci. 62, 521-527. SHIN, H. S., SNYDERMAN, R., FREEDMAN, E., AND MERGENHAGEN, S. E. (1969) Fed. Proc. 28, 485. MARCUS, R. L., SHIN, H. 8., AND MAYER, M. M. (1971) Proc. Natl. Acad. Sci. 68, 1351-1354. GOTZE, 0., AND MDLLER-EBERHARD, H. J. (1971) J. Exp. Med. 134, 903-1083. SANDBERG, A. L., OSLER, H. G., SHIN, H. 5., AND OLIVERIRA, B. (1970) J. Immunol. 104, 329-334. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 24 COCHRANE, C. G., MULLER-EBBRHARD, H. J., AND FJELLSTROM, K. (1968) J. Clin. Invest. 47, 21a. SHIN, H. S., GEWURZ, H., AND SNYDERMAN, R. (1969) Proc. Soc. Exp, Biol. Med. 131, 203-206. PICKERING, R. J., WOLFSON, M. R., GOOD, R. A. AND GEWURZ, H. (1969) Fed. Proc. 28, 818. GOTZE, 0., AND MULLER-EBHRHARD, H. J. (1971) J. Immunol. GOTZE, 0., AND MDLLHR-EBERHARD, H. J. (1972) Fed. Proc. 31, 787. PILLEMER, L., BLUM, L., LEPOW, I. H., ROSS, O. A., TODD, E. W., AND WARDLAW, A. c. (1954) Science 120, 279-285. PILLEMER, L., BLUM, L., LEPOW, I. H., WURZ, L., AND TODD, E. W. (1956) Jo EXE. Med. 103' 1.13. WARDLAW, A. D., AND PILLEMER, L. (1956) J. Exp. Med. 103, 553-575. WEDGEWOOD, R. J., GINSBERG, H. S. AND PILLEMER, L. (1956) J. Exp. Med. 104, 707-725. PENSKY, J., HINZ, c. I., JR., TODD, R. J., WEDGEWOOD, R. J., BOYER, J. T. AND LEPOW, I. H. (1968) J. Immunol. 100, 142-152. PENSKY, J., WURZ, L., PILLEMER, L. AND LEPOW, I. H. (1959) Z. Immunitatsforsch. 118, 329-348. BLUM, L., PILLEMER, L., AND LEPOW, I. H. (1959) Z. Immunitatsforsch. 118, 349-357. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 25 PILLEMER, L., LEPOW, I. H., AND BLUM, L. (1953) NELSON, R. A. (1958) J. Exp. Med. 108, 514-535. KABAT, E. A., AND MAYER, M. M. (Editors) (1967) Experimental Immunochemistry, Ed. 2, Charles C. Thomas, Springfield, pp. 231-233. LEPOW, I. H., PILLEMER, L., SCHOENBERG, M., TODD, E. W. AND WEDGEWOOD, R. J. (1959) J. Immunol. 83, 428-436. GOODKOFSKY, I., AND LEPOW, I. H. (1971) J. Immunol. 107, 1200-1204. WARD, P. A. (1971) Inflammation, Immunity and Hyper- sensitivity (Movat, H. Z., Ed.), Harper and Row, New NELSON, D. S. (1963) Adv. Immunol. 3, 131-180. DIAS DA SILVA, W., EISELE, J. W., AND LEPOW, I. H. (1967) J. Exp, Med. 126, 1027-1048. LEPOW, I. H., DIAS DA SILVA, W., AND EISELE, J. W. (1968) Biochemistry of the Acute Allergic Reactions (Austen, K. F. and Becker, E. L., Eds.), Blackwell Scientific Publications, Oxford, pp. 265-282. COCHRANE, C. G., AND MDLLER—EBERHARD, H. J. (1968) Jo EXP. MBd. 127' 371-3860 WARD, P. A., COCHRANE, c. G., AND MOLLER-EBERHARD, H. J. (1965) J. Exp. Med. 122, 327-347. WARD, P. A., AND BECKER, E. L. (1967) J. Exp. Med. 125, 1001-1020. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 26 YONEMASU, K., AND STROUD, R. M. (1971) J. Immunol. 106, 304-313. BING, D. H. (1972) personal communication. COLTEN, H. R., BORSOS, T., AND RAPP, H. J. (1968) J. Immunol. 100, 799-807. DE BRACCO, M. M. E., AND STROUD, R. M. (1971) J. Clin. Invest. 50, 838-848. BORSOS, T., AND RAPP, H. (1963) J. Immunol. 91, 85-858. NELSON, R. A., JR., JENSEN, J., GIGLI, I., AND TAMURA, N. (1966) Immunochemistry 3, 111-135. TAMURA, N., AND NELSON, R. A., JR. (1968) J. Immunol. 101, 1333-1345. COLTEN, H. R., BOND, H. E. BORSOS, T., AND RAPP, H. R. (1969) J. Immunol. 103, 862-865. LINSCOTT, W. D. (1968) Immunochemistry, 5, 314-318. HOFFMAN, L. G. (1969) J. Chromatqg. 40, 39-52. BING, D. H. (1971) J. Immunol. 107, 1243-1249. NAFF, G. B., PENSKY, J., AND LEPOW, I. H. (1964) Jo EXBO Med. 119, 593-6130 LEPOW, I. H., RATNOFF, O. D., ROSEN, F. 5., AND PILLEMER, L. (1956) Proc. Soc. Exp. Biol. Med. 92, 32-37. LEPOW, I. H., RATNOFF, O. D., AND PILLEMER, L. (1956) BECKER, E. L. (1956) J. Immunol. 77, 469-478. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 27 HAINES, A. L., AND LEPOW, I. H. (1964) J. Immunol. 92, 456-467. HAINES, A. L., AND LEPOW, I. H. (1964) J. Immunol. 92, 468-478. HAINES, A. L., AND LEPOW, I. H. (1964) J. Immunol. 92, 479-490. NAGAKI, K., AND STROUD, R. M. (1969) J. Immunol. 102, 421-430. BING, D. H. (1971) Immunochemistry 8, 539-550. MULLER-EBERHARD, H. J., AND KEINKEL, H. G. (1961) Proc. Soc. Exp. Biol. Med. 106, 291-294. CALCOTT, M. A., AND MOLLER-EBERHARD, H. J. (1972) Biochemistry in press. SHELTON, E., YONEMASU, K., AND STROUD, R. M. (1972) Proc. Natl. Acad. Sci. 69, 65-68. SVEHAG, S. E., BLOTH, B. (1970) Acta Path. Microbiol. YONEMASU, K., STROUD, R. M., NIEDERMEIER, W., AND BUTLER, W. T. (1971) Biochem. BiOphys. Res. Comm. 43, 1388-1394. YONEMASU, K., AND STROUD, R. M. (1972) Immunochemistry 9, 545-554. MfiLLER—EBERHARD, H. J. (1969) Ann. Rev. Biochem. 38, 389-414. MHLLER-EBERHARD, H. J. (1968) Textbook of Immuno- pathology (Meischer, E., and Mfiller-Eberhard, H. J., Eds Eds.), Grune and Stratton, New York, pp. 33-47. 77. 78. 79. 80. 81. 82. 83. ) 84. 85. ( 86, 87. ( 88. , 39, Z 9.. .1 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 28 NATALI, P. G., AND TAN, E. (1972) J. Immunol. 108, 318-324. YACHNIN, S., ROSENBLUM, D., AND LATMAN, D. C. (1964) J. Immunol. 93, 540-548. AGNELLO, V., WINCHESTER, R. J., AND KUNKEL, H. G. (1971) J. Immunol. 107, 309. DAVIES, G. E. (1963) Immunology 6, 561-568. BORSOS, T., RAPP, H. J., AND CRISLER, C. (1965) ISHIZAKA, K., ISHIZAKA, T., AND SUGAHARA, T. (1962) AMIRAIAN, K., AND LEIKHIM, E. J. (1961) Proc. Soc. Exp. SCHUR, P. H., AND BECKER, E. L. (1963) J. Exp. Med. GRIFFIN, D., TACHIBANA, D. K., NELSON, B., AND ROSENBERG, L. T. (1966) Immunochemistry_ 4, 23-30. COHEN, 8., AND BECKER, E. L. (1968) J. Immunol. 100, COHEN, 8., AND BECKER, E. L. (1968) J. Immunol. 100, 403-406. WIRTZ, G. H. (1965) Immunochemistry 2, 95-102. AUGENER, W., GREY, H. M., COPPER, N. R., AND MHLLER- EBERHARD, H. J. (1971) Immunochemistry 8, 1011-1020. MHLLER-EBERHARD, H. J. AND CALCOTT, M. A. (1966) Immunochemistry 3, 500. 91. 92. 93. 94. 95. 96. 97. 98. 99. 91. 92. 93. 94. 95. 96. 97. 98. 99. 29 ISHIZAKA, T., ISHIZAKA, K., BORSOS, T., AND RAPP, H. (1966) J. Immunol. 97, 716-727. LINSCOTT, W. D., AND HANSEN, S. S. (1969) J. Immunol. 103, 423-428. MACKENZIE, M. R., WARNER, N. L., LINSCOTT, W. D., AND FUDENBERG, H. H. (1969) J. Immunol. 103, 607-612. MACKENZIE, M. R., CREEVY, N., AND HEH, M. (1971) J. Immunol. 106, 65-68. COHEN, S. (1968) J. Immunol. 100, 407-413. RAPP, H. J., AND BORSOS, T. (Editors) (1970) Molecular Basis of Complement Action, Meredith Corporation, New York, pp. 69-74. ISHIZAKA, T., ISHIZAKA, K., SALMON, 8., AND FUDENBERS, H. HYSLOP, N. E., JR., DOURMASHKIN, R. R., GREEN, N. M., AND PORTER, R. R. (1970) J. Exp. Med. 131, 783-802. PLAUT, A. G., COHEN, 5., AND TOMASI, T. B. (1972) Science 176, 55-56. ARTICLE 1 Purification of the Human Complement Protein Clq by Affinity Chromatography BY C. R. Sledge and D. H. Bing (Manuscript submitted to Immunochemistry) 30 ABSTRACT A subunit of the first component of human complement, C1q,* was purified by the technique of affinity chromatography. The chromatographic resin was cyanogen bromide activated Sepharose covalently linked to human IgG. To remove traces of IgM is was found necessary to further subject the Clq obtained from the chromatographic s-ep to ultracentrifugation in sucrose gradients. The highly purified Clq was character- ized immunochemically and according to its electrophoretic mobility in various polyacrylamide gel systems. The purified material was capable of combining with a reagent containing ‘ Clr and C13 to reconstitute fully active macromolecular Cl. *The terminology used for the complement proteins is that suggested in the Bull. Wld. Hlth. Org. "Nomenclature of Comple- ment," Immunochemistry, 1:137, 1970. Complement components are designated numerically C1, C2, C3, C4, C5, C6, C7, C8 and C9; the subunits of C1 are designated Clq, Clr and C13; acti- vated components are designated by placing a rule over the numeral which refers to the component or subunit. Cellular intermediates carrying complement components are designated EAC, follgwed by the numeral designating the components carried, e.g., EAC1,4. 31 three prese acti‘ of 8 even natt ent wil sit and low com; INTRODUCTION The first component of human complement is composed of three distinct proteins, Clq, Clr and C13 which require the presence of calcium ion to physically associate and form fully active macromolecular CI. Clq binds to the antibody molecule of an antigen-antibody complex, and initiates the sequence of events involved in complement mediated immunecytolysis. The nature of the interaction of Clq with immunoglobulins appar- ently involves the following parameters: 1) only IgG and IgM will combine with Clq (Augener pp 31., 1971), 2) the binding site for Clq is in the Fc region (Mfiller-Eberhard, 1968), and 3) the Clq-immunoglobulin complex can be dissociated by low pH high ionic strength salt solutions and diaminoalkyl compounds (Mfiller-Eberhard, 1968; Wirtz, 1965). The preceding information suggested an affinity chromato- graphic procedure for Clq could be develOped based on absorp- tion to a resin covalently linked to IgG and elution with a suitable diaminoalkyl compound. This report presents evidence that this can be accomplished and that the technique is a rapid reproducible method for obtaining milligram quantities of Clq from a few hundred milliliters of whole serum. 32 MATERIALS AND METHODS Chemicals and Reagents. All chemicals and solvents were reagent grade. Triple distilled water was used for all buf- fers. 1,4-Diaminobutane was obtained as the free base from Aldrich Chemical Co. (Milwaukee, WI). Hemolysin was obtained from Behring Diagnostics (Woodbury, NY). Sheep blood, from a single male sheep, was collected into Alsever's solution. Guinea pig blood and pooled human serum were donated by the Michigan State Public Health Laboratories (Lansing, MI). Chemical Procedures. The procedure of Bing (1971) was used to covalently link human IgG to Sepharose. Forty milli- liters of settled Sepharose 68 were mixed with 4 g of CNBr in 40 m1 of H20 and the pH maintained at 11 by addition of 4 N NaOH with stirring until the pH remained constant. The Sepharose was washed and filtered with 800 ml of ice cold 0.1 M NaHCO3, pH 9.0, resuspended in 40 ml of the same buffer, and 40 m1 of IgG solution (10 mg/ml) added. Coupling to IgG was allowed to proceed with stirring for 24 hours at 4 C. The resin was washed with 0.1 M NaHCO3, pH 9.0, and the ab- sorbancy at 280 nm of the wash determined. Assuming an ex- tinction of 1.5 for a 1 mg/ml solution of IgG (McClure and Edelman, 1966), it was determined that 370.0 mg of protein had 33 beex tiV( 8.1. 581'). huma and cont phorl serur fract serun yiel (Lepc M NaC for 1 Tris- 30 mi to ti and d Free; 4 C.| 0.07. Judgx can 34 been bound to the resin. The resin was then washed exhaus- tively with 0.075 ionic strength Tris-HCl + 0.01 M EDTA, pH 8.1, and stored at 4 C with 0.005 M sodium azide as a pre- servative. Protein Preparations. Human IgG was isolated from pooled human serum by chromatography on DEAE cellulose (Fahey, 1967), and by precipitation with (NH4)ZSO at 40% saturation. It 4 contained only IgG immunoglobulin according to immundelectro- phoretic analysis with a rabbit anti whole human serum anti- serum. The low ionic strength acid precipitate (euglobulin fraction) of serum was prepared by precipitation of human serum with 0.02 ionic strength acetate buffer (pH 5.5) to yield a final serum ionic strength of 0.03 and a pH of 6.4 (Lepow pp 31., 1963). The precipitate was redissolved in 0.3 M NaCl to one-tenth the original serum volume and dialyzed for 18 hours against two 1 L changes of 0.1 ionic strength Tris-HCl + 0.01 M EDTA, pH 8.1. The protein was centrifuged 30 minutes at 10,000 rev/min (12,100 x g) before application to the column. A reagent deficient in Clq (RC1q), but containing Clr and C13, was prepared by incubation of 5 m1 of euglobulin precipitate with 50 ml of IgG Sepharose resin for 15 hours at 4 C. The resin was then poured into a column and eluted with 0.075 ionic strength Tris-HCl + 0.01 M EDTA. The eluate was judged to be depleted of Clq by its inability to form EACI,4 cells (Borsos and Rapp, 1963). 35 Assays. EACI,4 cells were prepared by the method of Mayer (1961). EAC4 cells were obtained by incubation overnight at 4 C in triethanolamine buffered saline containing 0.01 M EDTA (Mayer, 1961). The hemolytic activity of Clq was deter- mined by its ability to form macromolecular Cl when varying concentrations of Clq were added to a constant amount of the RC1q reagent in the presence of 0.02 M CaCl EAC4 cells were 2. added and the resulting EACI,4 intermediate was washed three times with sucrose triethanolamine buffered saline, trans- ferred to a clean tube, C2 and CEDTA were added, and the number of effective CI molecules formed was calculated by the pro- cedure of Borsos and Rapp (1963). Clq activity was also de- tected by the slide agglutination test of Ewald and Schubert (1966) using fraction II gamma globulin coated latex particles. Affinity Chromatography Using IgG-Sepharose. A 2.5 x 9.0 cm column of IgG-Sepharose was poured and equilibrated at 4 C with 0.075 ionic strength Tris-HCl + 0.01 M EDTA, pH 8.1. Six milliliters of euglobulin in 0.1 ionic strength Tris-HCl + 0.01 M EDTA, pH 8.1, were applied to the column and 4-m1 fractions collected. The column was washed with equilibration buffer until the extinction at 280 nm of the effluent was less than 0.05. The column was then eluted with 0.4 M NaCl + 0.01 M EDTA until the extinction of the effluent at 280 nm was less than 0.05. Finally the column was eluted with 0.2 M 1,4- diaminobutane. Tubes containing protein were pooled and dialyzed against 0.15 ionic strength Tris-HCl + 0.01 M EDTA, pH 8.1. 36 Ultracentrifugation. A Beckman Model L2-65B ultra- centrifuge and an SW27 rotor were used for the sucrose gradient ultracentrifugation. The sample was layered on a linear 10-40% (w/v) sucrose gradient buffered with 0.5 ionic strength acetate, pH 5.0. Centrifugation was conducted for 15 hours at 27,000 rev/min, 4 C. Fifty-drop fractions were collected. Moving boundary sedimentation velocity experiments were carried out in the Spinco Model B analytical ultracentrifuge at 3-5 C and a rotor speed of 56,000 rev/min using a AN-D 2350 rotor. Pool q preparations were analyzed at a concentration of 5 mg/ml in both 0.15 ionic strength Tris-HCl, pH 8.1 and 0.5 ionic strength acetate, pH 5.0 buffers. Apparent sedimen- tation coefficients were corrected to 20 C. Immunodiffusion Analysis. Ouchterlony double diffusion was carried out in 0.5% agarose in 0.15 M NaCl containing 0.01 M EDTA and 0.1% sodium azide. Radial immunodiffusion for the assay of human IgM and IgG was conducted in agar immunoplates containing anti-human IgM and anti-IgG )Hyland Laboratories, Los Angeles, CA). The sensitivity of the immunoplates was 220 u9/ml for IgM and 2 mg/ml for IgG. The pool q preparation was made 10 mg/ml and sucrose gradient purified Clq was made 5 mg/ml for analysis in these plates. Acrylamide Gel Analysis. Acid and base acrylamide gel electrophoresis were conducted in 3% spacer and 4% running gels at 5 ma/gel according to the procedure of Maizel (1969). 37 Sodium dodecyl sulfate (SDS) electrophoresis was performed in 4% acrylamide gels containing 0.1% SDS. In all cases 150 ug of protein was applied to each gel and staining was done with 0.25% Coomassie Blue. RESULTS Chromatography on IgG-Sepharose. The elution profile of the euglobulin fraction chromatographed on the IgG-Sepharose column is illustrated in Figure 1. The protein which had no binding affinity for the resin comprised Pool A. Pool U represented nonspecifically bound protein eluted with 0.4 M NaCl + 0.01 M EDTA. The Clq was eluted in Pool q with 0.2 M 1,4-diaminobutane. Table 1 summarizes the results of a typical experiment. The majority of the protein applied to the resin was not absorbed and Pool A represented 85.6% of the applied material. Pool q contained 2.7% of the protein in the euglobu- lin fraction. The ability of the various column fractions to agglutinate gamma globulin coated latex particles was deter- mined. Pool A yielded a weakly positive reaction; Pool U was negative; and Pool q strongly agglutinated the latex particles. The Pool q preparations did not contain any detectable Cls and Clr when tested by specific esterolytic assays with synthetic substrates (Bing, 1969; Haff and Ratnoff, 1968). Hemolytic Activity of the Isolated Clq. Various dilutions of Clq were incubated with a constant amount of the RC1q reagent (see Materials and Methods) in the presence of 0.02 M CaClZ. The ability of this mixture of lyse EAC4 upon the addition of the remaining complement components was determined 38 39 .oemusooceaneoue.fl z ~.o one «Ham 2 Ho.o + Honz z «.0 mo coaufloom mo mucflom one opmowpcfl HBOHHH age .omoumnmomlowm co Eduom mo coauomum cflaonoamso map mo mnmmumoumEousu .H onomflm 40 m2fluom OHDMHOEos one “Ollllo . ommd .q Hoom GofluomDMHHucoomuuHs ucofiomum omOHOSH unwound muHOm 0» cos .m ousmflm 48 H SISA'IOWSH 96 81 m ousmflm mum—232 20_ho 300C. The 2,5-diaminotoluene dihydrochloride mp 300C, and the 3,5-diamino-l,2,4-triazole mp 209C [litmp 206C(10)] were recrystallized 5 times from ethanol. Thin layer chromatog- raphy (butanolzacetic acid:water, 120:30:50 v/v/v) was per- formed on all compounds. The compounds were prepared at a 2M in 0.075 ionic strength Tris- final concentration of 10- HCl, pH 8.1. 1-[4-14C]-diaminobutane dihydrochloride was obtained from Amersham/Searle Corp. This was diluted with unlabeled 1,4-diaminobutane to yield a specific activity of 50 uc/mole. Preparation of IgM Sepharose Resin. The coupling of IgM to sepharose was performed according to the procedure of Cuatrecasas (ll). Fifteen milliliters of settled sepharose 63 were mixed with 4.0 g cyanogen bromide in 15 ml of H20 and the pH maintained at 11 by addition of 4 N NaOH for 15 min or until the pH remained constant. The cyanogen bromide acti- vated sepharose was then reacted with 168 mg of IgM for 36 hrs at 4 C in pH 7.0 0.1 M sodium phosphate buffer. 61 The macroglobulin-sepharose resin was washed extensively with 0.075 ionic strength Tris-HCl, pH 8.1. The absorbance 0.1% at 280 nm of the wash was determined, and using a Elcm of 1.2 for IgM the amount of IgM bound to the resin was calculated. A more analytical determination of the amount of IgM bound to the resin involved the analysis of 1 m1 of 1yophi1ized resin for protein using a modified ninhydrin procedure (12). Iodination of Clq. Clq was iodinated with NalZSI (New England Nuclear) according to the method of Helmkamp pp 31. (13), and had a specific activity of 1,125,000 cpm/mg. f 125 Binding Assays. The binding 0 I-labeled Clq to the macroglobulin-sepharose resin was assayed using the fol- lowing procedure: 0.2 m1 of macroglobulin resin was added 125I-Iabeied Clq to 0.1 m1 of an appropriate dilution of and the mixture incubated at 37 C for 30 min. The sample was then filtered on Whatman No. 1 filter paper and the resin washed with 3 m1 of buffer. The resin and filter paper were counted in a Packard Gamma Scintillation Counter. A control containing an equal amount of cyanogen bromide activated sepharose was included as a blank. All assays were done in triplicate and the values averaged. The variation between identically treated samples ranged from 4.6-6.3%. The buffer used for the assay was 0.075 ionic strength Tris-HCl, pH 8.1 containing 5 mg/ml of Bovine Serum Albumin to minimize non- 125 specific binding. The amount of I-labeled Clq bound was an! 62 expressed as the difference between the blank and the experi- mental values. The following equation describes the inter- 125 action between I-labeled Clq and macroglobulin: K 1 _ 1 d 1 [CIqu IC1q]f x [Igmls + [IgM]s where I”. [Gig]b = bound Clq, ( [Clq]f = free Clq, f, [IgM]S = total number of IgM sites available for binding, and Kd = dissociation constant of the complex. -1 vs (free Clq)-1 were linear and the l Plots of (bound Clq) intercepts at the abscissa and ordinate represent [Rd]. and [total IgM sites]-l respectively. Inhibition of the binding of 125I-labeled Clq to the macroglobulin-sepharose resin by IgG, IgM, (Fc)5u and the heat aggregated forms of thses proteins was studied by incu- bating various concentrations of the proteins with a constant amount of lZSI-labeled Clq for 30 min at 37 C. IgM-sepharose resin (0.2 ml) was added to the mixture and incubation pro- ceeded at 37 C for 30 min. The sample was then filtered and counted. The following two tubes were included as controls: 125I-labeled Clq + 0.2 ml cyanogen bromide activated 125 (1) sepharose (blank), and (2) I—labeled Clq + 0.2 m1 IgM- sepharose. The amount of 125I Clq bound was obtained by 63 subtracting the cpm in the blank tube (1) from the cpm in the control (2) and the experimental tubes. Assigning a 100% bound value to control (2) the percent inhibition values were calculated according to the equation: cpm Experimental Tubes Percent inhibition = (l - cpm Control Tubes ) x 100. 59 L Inhibition studies with the various small molecular weight ( amino compounds were conducted according to the same procedure i.- that was employed for the proteins. Binding of l-[4-14C] Diaminobutane to Clq. Various dilu- l4-l,4-diaminobutane were added to 0.2 m1 6 tions (0.2 m1) of C 6 M Clq or 2.0 x 10' M IgM and the mixture of either 2.0 x 10‘ incubated at 37 C for 30 min. Control tubes containing only 1,4-diaminobutane were included in all assays. Following incubation the samples were treated with 50% saturated (NH 804 to precipitate the protein, and centrifuged at 2000 4)2 rev/min for 30 min. The supernant fractions were removed and 0.2 m1 placed on Whatman filter paper and allowed to dry in a hot air oven. The scintillation fluid was composed of 4 g PPO and 50 mg of POPOP in 1 liter of toluene. Ten m1 of scin- tillation fluid was added to each sample and the vials counted in a Packard Tricarb Liquid Scintillation Counter. RESULTS 125 Binding of I-labeled Clq to the IgM-Sepharose Resin. The amount of macroglobulin bound to the sepharose resin was determined by two separate procedures and the values agreed quite well. Based on the amount of protein present in the wash from the resin it was calculated that 5.2 mg of IgM had been coupled per m1 of sepharose. Analysis of the resin using the modified Ninhydrin procedure yielded a value of 5.9 mg of IgM per ml of resin. The results of two separate experiments on the interaction 125 of I-labeled Clq with the IgM resin are illustrated in Figure l. The agreement between the two sets of data are good and dissociation constants were determined to be 8.69 x 10-7M and 7.04 x 10-7M. 125 Inhibition of the Binding of I-labeled Clq to the IgM-Sepharose Resin binmmunoglobulins. Due to the fact that the IgM-sepharose resin was not characterizable in terms of the molecular form of IgM present on the resin, it was decided not to assume that the binding properties of the macroglobulin- sepharose complex were wholly characteristic of IgM. Instead, the ability of known molecular forms of IgM, (Fc)5u, and IgG 125 to inhibit the binding of I-labeled Clq to the IgM-sepharose complex was determined and used as a quantitative measure of 64 Figure 1. 65 The binding of 125 sepharose complex. I-labeled Clq to the IgM- Assays were carried out as described in "Experimental Procedures." The two plots represent separate experiments done under identical conditions. The following dissoci- ation constants (Kd) were determined: 8.69 x 10'7 M, I. Experiment A, Kd Experiment B, K 7.04 x 10_7 M, . . d 66 Poona C14" coucemmou. pa [FREE cud" CONCENTRATION, pa Figure 1 ‘nl‘ a)... ‘i 67 the interaction between Clq and immunoglobulins. The sedi- mentation properties of the monomeric proteins used in these experiments were determined in a sucrose density gradient and IgM, (Fc)5H, and IgG exhibited sedimentation coefficients of 198, 118, and 78, respectively, when compared to known standards (Figure 2). The ability of IgM, (Fc)5u, and IgG to inhibit the 125I-labeled Clq to the IgM-sepharose complex is binding of shown in Figure 3. Both IgM and (Fc)5u were quite effective in binding the radiolabeled Clq and thereby inhibiting its binding to the IgM resin. The (Fc)5u, however, was slightly more effective on a molar basis as a binding agent than the IgM. IgG was capable of interacting with the 125I-labeled Clq, but it was a much poorer binding agent than either IgM or (Fc)5u. When various concentrations of Bovine Serum Albumin were substituted for the immunoglobulins in the in- hibition assay, no inhibition was detected. This strongly suggested that the inhibition exhibited by the immunoglobulins was not due to nonspecific protein-protein interactions. Inhibition constants (Ki) were determined according to Dixon and Webb (14) using a previously determined TiafiTg. The AF' was calculated using the relationship AF' = -RT 1n Ki' Inhibition constants were calculated for IgG, IgM, and (Fc)5u, respectively (Table l). The AF' values ranged from 5.60 kcal/mole for IgG to 7.60 kcal/mole for (Fc)5u. WT‘W‘ fl 68 E: omm .ouocfiono on» so we m can .mmmwomoo onv so oouuoam ma Hones: :ofluooum .EG own um oocmoHOHQH Hfioau How uouoeouonmouuoomm Hoa cmEoHoolwsoouHm M GA oouwaoco ouoz AHE o.Hv mGOHuooum one .zmH mma can .me me mcflopoum Roxane opp mo HOH>ocoo cofluoucoefloow onu ounceocfl mesons/w . . .UmH A I .1305 u ‘JamH ...mo..3ooooum HoucoEHHomxm= E.” ooQflHomoo mcoflufioooo ozu noon: oouosocoo mos coauomDMHHucoomuuHD .omH can .amxomv .zmH mo coeunmomenooooduuas Abovnoac Momentum omonosm mills... .. .W ‘ .N onomflm 69 N ousmflm OW 0.0 ON Op (v.0 V 7a 8 (ad 0 U c w IN... as. «2. do» + + 2053 I ma h p h F 70 .o .nHEan4 Esuom onH>om “‘vaH A .5on5 u I .ZmH .mnfionfln mo nonufln lance unoonom mo nowponHEHouoo one one momma onu How poms oHo3 =mousooooum Hounoefluomxmg cw oonflHOmoo moonuoE one .nHEJQHo Esuom ona>on onm .me .nmnomv .zmH mn onmEoo omouonmomuzmH onu Ou wau ooaonoHIHmmH mo mnflonfln on» NO coauanflnnfl one on .m ouomflm 71 m ousmam 5 .20....(Chzm0200 N niild 0-9 . o o v _ pt»— - p ‘_711- pthhp b 14 p t — Tow Op ON On O? on ONIONIB :10 NOLLIBIHNI “'30 33d 72 Table l. The inhibition constants (K1) and relative free energy values (AF') for IgG, IgM, and (Fc)5u. Protein Ki(M)* ‘ AF'* (kcal/mole) IgG 1.10 x 10‘4 5.60 IgM 6.42 x 10’6 7.35 (Fc)5u 4.35 x 10"6 7.60 *The K- and AF' values were calculated according to the proce ures described in "Experimental Procedures." 73 Heat aggregated IgM, (Fc)5n, and IgG were also tested for their ability to bind the radiolabeled Clq (Figure 4). The heat aggregated IgG was a considerably more effective binding agent than the monomeric protein. Heat aggregation of IgM also increased its ability to bind Clq, however, heat treatment of (Fc)5u did not considerably alter its binding affinity for Clq. 125 Inhibition of the Binding of I-labeled Clq to the IgM-8gpharose Resin py Small Molecular Weight Amino Compounds. Previous studies had demonstrated that l,4-diaminobutane was capable of inhibiting the binding of both macromolecular Cl and Clq (5,15) to gamma globulin. To investigate the nature of this inhibition more carefully it was decided to determine with which of the two proteins the l,4-diaminobutane was inter- acting. A series of experiments were conducted in which l-[4-14C] diaminobutane was allowed to interact directly with either Clq or IgM. The amount of labeled l,4-diaminobutane bound to these proteins was determined by precipitating the complex with 50% saturated (NH4)ZSO The results (Figure 5) 4. showed that Clq was capable of binding the l-[4-14C] diamino- butane, whereas the same molar concentration of IgM was unable to bind the amine. However, the binding of the amine to Clq could be inhibited by the preincubation of IgM (2.0 x 10-6M) with Clq (2.0 x 10‘6M). 74 .HE\mE N mo coauouunoonoo o um nofluanflnnw Now non Hono ouos mnflououm Had .m enmnomv ooumouu uoon “M .1 Aomv .zmH condone oboe no .zmH Am .omH odononu Home Ag .omH =.mousooooum HounoEwHomxm= 2H oonfiuomoo mo uso oofluuoo ouos whommo onu ono mnwououm onu mo #noEuooHu uoon one Q .omH pen .amxomv .ZmH oouoouu noon ono OHHoEonOE an onmEOo omoumnmoMIZmH one 0» 6H0 boHoonHuHmNH no maHoEHo one no coeueoenne one .v ousmflm U Figure 5. 76 The binding of l—[4-14CJ-diaminobutane to Clq. Assays were carried out as described in "Experimental Procedures." Clq, A ; IgM, I - IgM + Clq, Q . The concentration of IgM and Clq was 2.0 x 10'6 M. CPM X 10‘3 20— 16- 77 -o— *0 10'5 10"4 l 10'3 10'2 1-[4-‘4 C]DIAMINOBUTANE CONCENTRATION, M Figure 5 78 A set of experiments were performed to measure the in- hibitory effect of various diaminoalkyl compounds on the 125 binding of I-labeled Clq to the IgM sepharose resin. Of the diaminoalkyl compounds tested those having eight, ten, and twelve carbons are the most effective inhibitors of 2 binding (Figure 6). At a concentration of 10- M their in- hibiting capacities are approximately equal, however, the 1,12-diaminododecane was a much better inhibitor at a concen- tration of 10.3 M. The inhibition observed with l,4-diamino- 2 M was similar, while butane and 1,7-diaminoheptane at 10- ethylenediamine was the poorest of the alkyl amines in in- hibiting the binding. The relative abilities of 3,5-diamino-1,2,4-triazole, 2,5-diaminotoluene and l,4-diaminopiperazine to inhibit the 125 binding of I-labeled Clq to the IgM resin as shown in Figure 7. The most effective inhibitor compound tested was 2,5-diaminotoluene. At a concentration of 10-2 M it in- hibited 75% of the binding, while 3,5-diamino-l,2,4-triazole and 1,4-diaminopiperazine inhibited the binding by 41% and 30.5%, respectively. Figure 6. 79 ' i The inhibition of the binding of 125I-labeled Clq to the IgM-sepharose complex by diaminoalkyl compounds. Assays were carried out as described in "Experimental Procedures." Ethylenediamine,£>.; l,4-diaminobutane, O ; 1,7-diaminoheptane,l] ; 1,8-diaminooctane, ‘ ; 1,10-diaminodecane, Q ; 1,12-diaminododecane, O. PER CENT INHIBITION 4O 3O 20 1O 80 I F I I / /' / 1 l 1 l _ 10-5 10'4 10-3 10'2 CONCENTRATION . M Figure 6 81 Figure 7. The inhibition of the binding of 125I-labeled Clq to the IgM-sepharose Complex by diamino aromatic compounds. Assays were carried out as described in "Experimental Procedures." 2,5-diaminotoluene, O i 3 , 5-diamino-l , 2 , 4-triazole, I ; l , 4-diamino- piperiazine, Os . PER CENT INHIBITION 80 7O 60 50 4O 3O 20 10 ‘ I I l 1 l 10-5 10-4 10-3 CONCENTRATION, M Figure 7 10'2 DISCUSSION The purpose of the present study has been to obtain quantitative data on the interaction between immunoglobulins and Clq, and to analyze the ability of various diamines to inhibit this interaction.' The IgM-sepharose complex has pro- vided a useful tool to accomplish these objectives. Due to lzsI-labeled Clq it has the high affinity of this complex for been possible to conduct inhibition assays with IgC, IgM, and (Fc)5u. The inhibition constants which have been determined are a direct measure of the affinity of these monomeric immunoglobulins for Clq. The inhibition observed with the diamines has yielded original data on a series of inhibitors of immunoglobulin-Clq interaction, and has provided the first observations on the nature of the Clq binding site. The chemical fixation of IgM to sepharose apparently does block many of the complement binding sites on the mole- cule. It should be recalled that cyanogen bromide coupling to sepharose is via free amino groups on the molecule being attached. To avoid preferential coupling via the epsilon amino groups on IgM, the coupling reaction was done at pH 7.0 (11). Other studies have indicated that free epsilon amino groups on the immunoglobulin molecule are important in the interaction 83 84 with Cl. In spite of these precautions, we have calculated that approximately 10% of the IgM attached to the resin is available for binding (Figure 1, Experiment A). However this calculation is a lower estimate since it is based on a Clq:IgM molar ratio of 1:1 at saturating conditions of Clq, and one other worker has determined that for some macro- globulins the ratio is 0.3:1 (2). On the other hand, it can be speculated that the molecular nature of the resin may resemble an antigen-antibody complex and thus provides a high- “M _-_ A 133': 7:711 ly efficient Clq binding site. The important factor is that the complex exhibits a specific affinity for Clq and this has been used to determine the relative capacity of Clq to bind to known molecular forms of immunoglobulins. The inhibition constants determined for IgM (Ki = 6.42 x 10-6) and for IgG (1.10 x 10-4) clearly demonstrate that the IgM molecule is more effective in binding Clq. This differ- ence is a direct reflection of the number of sites on the molecules capable of binding Clq. Since Clq binding to immunoglobulins is via the Fc portion of the molecule (3), the arrangement of the five Fc regions within IgM must provide a more suitable binding site for Clq than the one Fc position contained in IgG. The (Fc)5u (Ki = 4.35 x 10-6) was observed to be slightly more effective in binding Clq than the parent IgM molecule. This finding agrees with a recent report by Plaut, Cohen, and Tomasi (l6), and suggests that the Clq bind- ing sites in the native IgM molecule are not totally exposed. 85 Heat aggregation of IgM and IgG increased the capacity of these immunoglobulins to bind Clq, and the most significant change occurred with the IgG. The heat treatment probably increased the proximity of the Fc regions in the molecule thereby making conditions more favorable for binding Clq. Heat aggregation of immunoglobulins under these conditions has previously been shown to affect only the Fab regions (17). This observation is relevant to the present study since the heat treatment of (Fc)5u did not alter its binding properties. The finding that l-[4-14C] diaminobutane interacts di- rectly with Clq and inhibits its binding to macroglobulin is of considerable interest. This explains the inhibition phenomenon with l,4-diaminobutane observed by other investi- gators (5,15,18), and provides a basis for studying other diamine compounds. The fact that the binding is exponential (Figure 5) indicates there is no discrete binding site on Clq for l,4-diaminobutane. This suggests that its inhibition of Clq-IgM interaction is mediated by the titration of anionic groups on the Clq. When various diaminoalkyl compounds were tested for their ability to inhibit the interaction between 125I-labeled Clq and the IgM-sepharose complex it was observed that 1,12- diaminododecane was the most effective inhibitor (Figure 6). The inhibiting ability of these compounds appeared to be a function of the length of the hydrocarbon chain. 86 Of all the diamino compounds studied 2,5-diaminotoluene was the most potent inhibitor of Clq binding (Figure 7). At 2 M this compound inhibited 75% of the a concentration of 10- binding. These findings suggest that in addition to the anionic groups there are hydrophobic groups on Clq which contribute to the interaction with the immunoglobulins. The nature of this hydrOphobic interaction is dependent on struc- ture as the triazole, a 5 membered ring, is less effective. It is interesting to note that on a molar basis the IgM was only 18 times more effective in binding Clq than 196 (Table 1), while in humans the mean circulating serum concen- tration of IgG is 60 times higher than IgM (21). This sug- gests that the differences observed in the binding . efficiencies of these two molecules may be overcome in the biological system by increasing the concentration of the weaker binding IgG. am“ My 10. REFERENCES AUGENER, W., GREY, H. M., COPPER, N. R., AND MfiLLER— EBERHARD, H. J. (1971) Immunochemistry 8, 1011-1020. MACKENZIE, M. R., CREEVY, N., AND HEH, M. (1971) J. Immunol. 106, 65-68. ISHIZAKA, K., ISHIZAKA, T., AND SUGAHARA, T. (1962) J. Immunol. 88, 690-701. HYSLOP, N. E., DOURMASHKIN, R. R., GREEN, N. M., AND PORTER, R. R. (1970) J. Exp. Med. 131, 783-802. SLEDGE, C. R., AND BING, D. H. (1972) Immunochemistry submitted. FAHEY, J. L. (1967) Methods in Immunology and Immuno- chemistry, Vol. I (Williams, C. A., and Chase, M. W., Eds.), Academic Press, New York, pp. 321-324. FAHEY, J. L., AND HORBETT, A. P. (1959) J. Biol. Chem. 234, 2645-2651. PLAUT, A. G., AND TOMASI, T. B. (1970) Proc. Natl. Acad. Sci. 65, 318-322. LOWRY, D. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem. 193, 265-275. HARRIS, G. (Editor) (1965) Dictionary of Organic Com- pounds, Ed. 2, Oxford University Press, New York. 87 ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. .21. 88 CUATRECASAS, P. (1970) J. Biol. Chem. 245, 3054-3065. SCHIFFMAN, G., KABAT, E. A., AND THOMPSON, W. (1964) Biochemistry 3, 113-117. HELMKAMP, R. W., GOODLAND, R. L., BALE, W. F., SPAR, I. L. AND MUTSCHLER, L. E. (1960) Cancer Research 20, 1495-1500. DIXON, M., AND WEBB, E. C. (Editors) (1958) Enzymes, i Academic Press, New York, p. 25. E- BING, D. H. (1971) J. Immunol. 107, 1243-1244. I PLAUT, A. G., COHEN, 5., AND TOMASI, T. B., JR. (1972) Science 176, 53-56. AUGENER, W., AND GREY, H. M. (1970) J. Immunol. 105, 1024-1030. WIRTZ, G. H. (1965) Immunochemistry 2, 95-102. GRIFFIN, D., TACHIBANA, D. K., NELSON, B., AND ROSENBERG, L. T. (1967) Immunochemistry 4, 23-20. COHEN, 8., AND BECKER, E. L. (1968) J. Immunol. 100, 403-406. DAVIS, B. D., DULBECCO, R., EISEN, H., GINSBERG, H. 5., AND WOOD, W. B., JR. (Editors) (1969) Microbiology, Harper and Row, New York, p. 434. CONCLUDING REMARKS The data presented in this thesis demonstrate: l) Affinity chromatography using IgG-sepharose resins is a useful and efficient technique for the purification of Clq; -‘ n ”LLAJI-L—‘i . 2) The affinity chromatographically purified Clq can be — I further purified by ultracentrifugation in 10-40% sucrose ‘ gradients buffered with a low pH, high ionic strength acetate buffer; 3) The Clq purified by these procedures is capable of binding both monomeric and heat aggregated IgM, IgG, and (Fc)5u; 4) IgM is 18 times more effective in binding Clq than 196, however heat aggregation greatly increases the binding efficiency of both molecules; 5) The Clq-immunoglobulin inter- action is inhibitable by various diamine compounds, and among the compounds studied 2,5-diaminotoluene is the most potent inhibitor; 6) The inhibition pattern exhibited by the diamine compounds suggests that Clq contains an anionic binding site and a hydrophobic binding site for immunoglobulins. The affinity chromatographic technique for the purifica- tion of Clq is valuable because it affords a simple, rapid procedure for isolating this important molecule. The rapidity of the technique is of the utmost importance since Clq is a labile molecule and its biological activity can easily be 89 90 impaired during purification. The presence of the IgM in the Clq preparations suggests that this molecule became bound to Clq during the affinity chromatographic procedure. This is indeed possible since the low ionic strength conditions employed for chromatography were designed to favor Clq- immunoglobulin interaction. Analysis of the Clq preparations by analytical and preparative ultracentrifugation under low and high ionic strength conditions strengthens this interpre- - . yams-2‘ I tation. Gradient ultracentrifugation at high ionic strength 1 resulted in the resolution of at least two distinct peaks with sedimentation values of 10.28 and 18.88, whereas ultracentrifu- gation at low ionic strength resulted in the rapid sedimenta- tion of the proteins to the bottom of the gradient suggesting the presence of a large molecular weight complex. However the suggestion that IgM is bound to the Clq during purification has no implication on the state of association of these mole- cules in normal serum. The interaction observed in the present study is likely an artifact of the purification pro- cedure. The procedure developed for analyzing the interaction of Clq with immunoglobulins is based upon the same principles as affinity chromatography. The use of this approach for the purification of biological molecules has proven to be success- ful. However the present study represents one of the first attempts to use an insolubilized protein for the quantitative study of protein-protein interaction. The advantage of this 91 approach is that the free and bound protein can be easily separated thereby allowing the concentration of one of the two to be determined. It is shown that the IgM-sepharose complex is capable of binding iodinated Clq in a reproducible fashion and that the amount of iodinated Clq bound to the complex can be determined. The inhibition studies with the immunoglobulins lZSI-labeled show theselsubstances capable of binding the Clq and preventing its binding to the complex. This inhibition is judged to be specific since equal molar concentrations of Bovine Serum Albumin have no effect on inhibiting the binding. These results are taken to mean that this system is a valid one for studying the interaction between Clq and immunoglobu- lins. This approach may also prove useful for the study of other interacting protein systems. The reason that IgM is 18 times more efficient than IgG in binding Clq is probably because it contains 5 potential complement regions while IgG only has one. Heat aggregation greatly enhances the ability of these molecules to bind Clq. This enhancement may be due to an increase in the proximity of the complement binding sites and/or the exposure of residues which are hidden in the native molecule. The inhibition of the binding of 125I-1abe1ed Clq to the IgM-sepharose complex by various diaminoalkyl compounds ap- pears to be dependent on the length of the hydrocarbon chain. 92 The 1,12-diaminododecane is the best inhibitor in this group of compounds, and ethylenediamine is the poorest. The 2,5-diaminotoluene is the most potent inhibitor of all the compounds studied. It was also shown that radiolabeled l,4- diaminobutane interacts directly with Clq and inhibits its binding to immunoglobulins. This binding is believed to be via anionic groups on the Clq, some of which are involved in its binding to immunoglobulins. These findings are inter— preted to mean that Clq has an anionic binding site and a hydrophobic site which are responsible for its interaction with immunoglobulins. 7. .' H Mai-r: LJF 1293 03169 4668 3 llHlWlHl)