Reactions of Aggregated Mercaptoethanol Treated Gamma Globulin with Rheumatoid Factor Precipitin and Complement Fixation Studies.код для вставкиСкачать
Arthritis and Rheumatism O&ial Journal 4 the gmerican Rheumatismassociation VOL. XI, NO. 4 AUGUST, 1968 Reactions of Aggregated Mercaptoethanol Treated Gamma Globulin with Rheumatoid Factor Precipitin and Complement Fixation Studies - By NATHAN J. ZVAIFLER AND PETER SCHUR Reduction and alkylation of human gamma globulin prior to aggregation results in a loss of complement-fixing ability, and retention of the capacity to react with rheumatoid factor. The addition of aggregated, reduced gamma globulin to serum containing high titers of rheumatoid factor produces a typical bell-shaped precipitin curve with an equivalence point and little precipitate in the region of antigen excess. Since aggregated, reduced gamma globulin fixes little complement by itself, it was possible to determine whether its reaction with rheumatoid factor fixes com- I plement. Under appropriate conditions rheumatoid factor can fix complement. When aggregated, reduced gamma globulin was added to heat-inactivated sera with varying titers of rheumatoid f a e tor, no fixation of guinea pig or human complement was observed. But preformed complexes of rheumatoid factor and gamma globulin will fix human complement at 37 C. Intradermal injection of 40 pg N of aggregated, r e d u d gamma globulin into control subjects or patients with rheumatoid arthritis produced no immediate or delayed skin reactions. of the structure of gamma globulin have defined 2 distinct portions of the molecule: one that contains the antibody combining site, and the other responsible for a variety of biologic properties. These properties include the ability to fix to skin and give passive cutaneous anaphylaxis, the capacity to cross the placental membrane, reactivity with rheumatoid factors, and complement fixation.1 The struc- tures essential for the induction of biologic activities also appear to be dependent on disulfide linkages, since treatment with sulfhydryl reducing agents followed by alkylation results in a loss of many of these properties. Wiedermann, Miescher and Franklin observed that when 7s human antibodies or heat-aggregated human gamma globulin were reduced with 2-mercaptoethanol, they lost the ability to fix com- Supported in part (N.J.Z.) b y grants from the Public Health Service (AM45042 and AM-05140). NATHANJ. ZVAIEZER,M.D.: Associate Professor of Medicine, Director, Division of Rheumatic Diseases, Georgetown Unioersity Medical Center, Washington, D . C . 20007.PETERH. SCWR, M.D.: Associate in Medicine, Harvard Medical School, Robert B . Brigham Hospital, Boston, Massochusetts 02120. NVESTIGATIONS ARTHRITISAND ~ E U M A T I S M , VOL.11, No. 4 (AUGUST1968) 523 524 ZVAIFLER AND S C H U R plement while retaining the capacity to react with rheumatoid factors., This dissociation of these 2 biologic properties is of interest, for a variety of studies have examined the interrelationships of rheumatoid factor, antigen antibody complexes (or aggregated gamma globulin) and complement. Most investigations have shown that rheumatoid factor inhibits complement fixation by antigen antibody comple~es.~-7 It has also been inferred that the reaction of rheumatoid factor with aggregated gamma globulin is not complement fixing, but direct testing of this hypothesis has not been possible because aggregated gamma globulin is, of itself, too anticomplementary. The availability of a reagent, aggregated mercaptoethanol treated gamma globulin, which does not fix complement, but does react with rheumatoid factor, provides an opportunity to study this question. MATERIALSAND METHODS Human gamma globulin-Cohn fraction I1 was obtained from the American Red Cross. 2-mercaptoethanol (2ME) in a final concentration of 0.1 M was added to a solution containing 10 mgm./ ml. of human fraction I1 in phosphate buffered saline (0.15 M, pH 7.3). After incubation for 3 hours at room temperature, the reduced gamma globulin solution was dialyzed for 4 hours against large volumes of 0.02 M iodoacetamide, and then overnight in the cold against several changes of buffered saline. In some experiments sulfhydryl reduction was obtained by dialyzing the gamma globulin preparation against an excess of 0.1 M 2-mercaptoethanol for 3 hours at room temperature. The subsequent steps were the same. Aggregation of the gamma globulin preparations was produced by heating at 63 C for 15 to 20 minutes. The SS, and SS, precipitates were prepared according to the method of Christian.8 Fifty ml. of a 1 per cent solution of human gamma globulin was reduced with 0.1 M 2-mercaptoethanol and then heat-aggregated. Ten ml. of 2.18 M sodium sulfate was added. The precipitate which formed (SS,) was removed by centrifugation. An additional 10 ml. of 2.18 M sodium sulfate was added to the supernatant and a second precipitate formed (SS,). The two precipitates (SS, and SS,) were combined and dissolved in 10 ml. of buffer saline. Precipitin reactions were performed as described by Vaughan.9 Increasing concentrations of gamma globulin solution were introduced into ice-cold phosphate buffered saline (0.15 M, pH 7.2) and then 0.5 ml. of a serum with a high titer of rheumatoid factor was added. The final reaction volume was 5.0 ml. The reaction was allowed to proceed at room temperature for 3 hours, and then at 4 C for 7 days. The tubes were spun at 2,000 RPM for 60 minutes in PR2 International refrigerated centrifuge. The supernatant was decanted and analyzed for rheumatoid factor. The precipitates were washed twice with ice cold buffered saline, packed by centrifugation and analyzed for their nitrogen content by the micro Kjeldahl method.10 Rheumatoid factor was measured by the bentonite flocculation test.11 A single serum containing a high titer (1:4096) of rheumatoid factor was utilized for the precipitin studies. Precipitates of rheumatoid factor and reduced aggregated gamma globulin were made as follows: a pool of serum containing high titers of rheumatoid factor was inactivated at 56 C for 30 minutes, centrifuged at 2,000 RPM in a refrigerated centrifuge, and then 7.5 ml. was added to 27.5 ml. of iced buffered saline containing 0.01 M EDTA. Subsequently 15 ml. of reduced, aggregated FII (10 mgm./ml.) was added dropwise. These proportions were chosen to be at or near equivalence. The mixture was kept at room temperature for one hour and then at 4 C for 72 hours. The precipitates which formed were washed 3 times with phosphate saline and then suspended at the desired protein concentration. These are referred to in the text as rheumatoid factor-reduced gamma globulin complexes or preformed precipitates. Quantitative complement fixation was performed by the method of Barbaro and Becker.12 Hemolysis was measured spectrophotometrically. Guinea pig complement containing approximately 100 C'H,, units per 0.1 ml. was added to iced test tubes containing 0.25 ml. of heat-aggregated mercaptoethanol gamma globulin (10 mgm./ml.) and 0.25 ml. of heat-inactivated human serum containing rheumatoid factor. Sufficient triethanol amine buffered saline ( T B S ) containing 0.1 per cent gelatin was added to bring the total volume to 1.0 ml. The tubes were incubated at 37 C for 80 minutes, then 19 ml. of ice cold TBS was added, and they were centrifuged in the cold at 2,000 RPM for 15 minutes. Residual complement was determined in the supernatants. Controls of REACXIONS OF 525 TREATED GAMMA GLOBULW WITH RF *...............a A YC t R.F. Serum A ZMEYC t R.F. Serum Total pgN ppt10.5 ml I I I I I I I 1 2 3 4 5 6 7 MJ YGlobulin N added10.5 ml Fig. 1.-The globulin (a2ME effect of increasing concentrations of aggregated human gamma (A 6) or +) aggregated mercaptoethanol-treated human gamma globulin on precipitable N from 0.5 ml. of serum containing rheumatoid factor. complement alone, complement and rheumatoid factor serum, and complement and aggregated mercaptoethanol gamma globulin were included with each experiment. In experiments with preformed precipitates 0.25 ml. of the rheumatoid factor complexes was incubated with 1.75 ml. of fresh human serum at 37 C for 2 hours or overnight at 4 C. Controls included 0.25 ml. of reduced, aggregated FII of an equal protein concentration or 0.25 ml. of saline handled in identical fashion, In some experiments 1.75 ml. of guinea pig serum was substituted as a complement source. Following incubation the mixtures were centrifuged in the cold and the supernatants analyzed for residual total hemolytic complement and residual activity of the second and third components of complement. C'2hu was measured by the method of Austenl3 and C'3 by immune adherence14 and conversion of B,C to B,A globulin by micro-immunoelectrophoresisla against an anti human B,C/BIA globulin antiserum prepared in rabbits by the zymosan adsorption method of Mardiney and Muller-Eberhard.16 The per cent complement fixed (%C fixed) was calculated as: C'H,, units fixed x loo total CH,, 526 ZVAIFLW AND S C R V R 0.- ........+ A7C + - m... ........m *......... d A R.F. Serum rC + A R J . Serum AtC + R.F. Serum + EDTA A 2ME% + R.F. Serum r2NYrC + AR.F. Serum A~METG+ R.F. Serum + EDTA 175. 150, Tdal pgN ppU0.5 ml 125 la, 15 50 25 I 1 I I 3 2 I 4 I 5 I 6 1 7 Mg rClabulin N ddedn.5 ml Fig. 2.-The effect on the amount of nitrogen precipitated from rheumatoid serum produced by preheating rheumatoid serum at 56 C. for 30 minutes or the addition of 0.01M EDTA to rheumatoid factor serum prior to reaction with aggregated gamma globulin (A yC) or aggregated mercaitoethanol treated gamma globulin. (A 2ME +). RESULTS Precipitin reaction. The precipitin curve obtained when increasing concentrations of aggregated gamma globulin were added to a serum containing a high titer of rheumatoid factor is shown by the dotted lines in Fig. 1. There is a slow rise in precipitable nitrogen, followed by a long flat zone which lacks a clear-cut equivalence or solu- bilization of the precipitate in the region of antigen excess. The curve obtained with aggregated mercaptoethanol gamma globulin differs in several particulars. The total nitrogen precipitated is less, but there are well-defined zones of equivalence and antigen excess. Tests for rheumatoid factor in the supernatants showed activity up to the points designated by the arrows; be- 527 REACJXONS OF TREATED G A M M A GLOBULIN WlTH RF -A 200 2ME7G f R.F. Serum - - 4 A 2ME SS, - SS2 + R.F. Serum 175 150 125 I I Total I WN ppt10.5 ml n t ' ' 100 15 50 25 I 1 1 2 I 3 1 4 I 5 i ; Mg7Globulin N added10.5 mi Fig. 3.-Comparison of the amount of N precipitated from 0.5 ml. of serum containing rheumatoid factor by aggregated mercaptoethanol treated human gamma globulin (A 2ME yC) and by sodium sulfate fractions from aggregated mercaptoethanol treated human gamma globulin (A 2ME SS1 - SS2). yond this there was complete neutralization of rheumatoid agglutinating activity in both systems. An analysis of the contribution of complement or components of complement to the precipitable nitrogen is shown in Fig. 2. Precipitin reactions were performed as described, except that in one experiment the serum containing rheumatoid factor was heated at 56 C. for 30 minutes prior to reacting with the aggregated gamma globulin or to the mercaptoethanol treated gamma globulin. In another experiment the reaction was performed in the pres- ence of 0.01 M EDTA. Heat inactivation and EDTA treatment results in a decrease in the amount of nitrogen precipitated in the reaction with aggregated gamma globulin, but there is a similar decrease with the aggregated mercaptoethanol-treated gamma globulin. In spite of these decreases there is no significant change in the overall codguration of the 2 precipitin curves. The precipitin reaction of aggregated mercaptoethanol treated SS1 and S S 2 fractions of human gamma globulin with rheumatoid factor serum is shown in Fig. 3. The shape of the precipitin curve obtained 528 ZVAIFLER AND S c E U R 90 - 100 80 70 o x L - 60- 50 - U P 40- 30 20 - 10 I resembles that with the aggregated mercaptoethanol gamma globulin, and is not like the broad flat curve seen in the reactions with aggregated gamma globulin. There is an abrupt rise in precipitable nitrogen, a narrow zone of equivalence and a well-defined solubilization of antigen in the region of antigen excess. Complement Fixation A comparison of the complement fixing ability of aggregated gamma globulin and aggregated mercaptoethanol-treated gamma globulin is shown in Fig. 4. As little as 100 pgm. of aggregated gamma globulin fixes 50 per cent of the available guinea pig complement, while 10 mgm. of aggregated mercaptoethanol gamma globulin is unable to give a simliar degree of complement fixation. The results were similar with fresh human serum as the complement source. The results of 8 experiments analyzing complement fixation by the reaction of rheumatoid factor with aggregated reduced gamma globulin are summarized in Fig. 5. 1 Each experiment was performed with a sera of a different rheumatoid factor titer. Column I, in each instance, shows the residual complement after 60 minutes of incubation with only TBS. Column I1 is the complement left after the reaction of serum containing rheumatoid factor and aggregated mercaptoethanol gamma globulin. Column I11 shows the residual complement with rheumatoid factor serum alone, and IV the residual complement after reaction with only aggregated mercaptoethanol gamma globulin. The titer of rheumatoid factor for the individual sera is recorded beneath. There is no significant difference in the amount of residual complement in Column 111-rheumatoid factor serum plus complement, when compared with Column 11-rheumatoid factor serum and aggregated mercaptoethanol gamma globulin plus complement. Altogether 18 sera with varying titers of rheumatoid factor were analyzed; no evidence for complement fixation could be demonstrated. An additional 6 sera were examined comparing human complement (approximately 50 C’HSo REACITONS OF 529 TREATED G A M M A GLOBULIN M"H RF 64 256 4096 4096 C'H% 32 512 1024 4096 Fig. 5.-The residual guinea pig C'H,, in experiments with 8 different human sera containing rheumatoid factor: I-Complement alone; II-cmmplement, serum and aggregated mercaptoethanol treated gamma globulin; III-complement and serum; IV-complement and mercaptoethanol treated gamma globulin. The number beneath each experiment is the rheumatoid factor titer of the serum tested. units) with guinea pig complement; no complement fixation was noted with either. In all of the experiments performed the residual complement in the presence of heat inactivated rheumatoid factor serum (Column 111) was equal to, or greater than, the complement remaining after incubation with only TBS gelatin. The reason for the apparent protective effect of rheumatoid serum is analyzed in Fig. 6. I n 3 separate experiments, complement was incubated at 37 C. with TBS-gelatin, or in the presence of 6 normal sera or one serum with a high titer of rheumatoid factor. The residual complement was measured at 0, 30 and 60 minutes (only 0 and 60 minutes are shown). The inactivation of complement in TBS-gelatin was 13 per cent of the initial value: 6 per cent in the presence of normal serum and 7 per cent for the serum containing rheumatoid fac- tor. Thus, the protection noted at the end of 60 minutes of incubation reflects the ability of serum to prevent the deterioration of complement, and is not unique to serum containing rheumatoid factor. Complexes of rheumatoid factor and aggregated, reduced gamma globulin were incubated with fresh human serum at 37 C. for 2 hours, or overnight at 4 C., and the residual complement determined. This was compared with an equal concentration of aggregated, reduced gamma globulin alone. The results are shown in Table 1. At 4 C. there is very little complement fixed by either the preformed complexes or the aggregated, reduced gamma globulin; but at 37 C. the complex fkes 43 per cent of the available complement. The complement fixation must be due to the complex, and not the reduced gamma globulin in the complex, for if all the protein 530 ZVAIFLER AND SCHUR - 0 c 0- C +Serum .-e _-___.. A C R.F. (40961Serum loo Q + \ \ 95 \ % \ \- '\ 90 CH50 85 m rs MINUTES effect of heat inactivated normal human serum (C' + serum), heat inactivated human serum containing rheumatoid factor, (C' + R.F. 4096 serum) or triethanohmine buffered saline with 0.1%gelatin (C') on the loss of guinea pig complement activity during incubation at 37 C. for 60 minutes. Fig. &-The in the complex were derived from aggrea otnrl 6 " C V U ' rJta.-J 6 nnmmn """"" A V U U - U rrlnha-lin ; t xV w w~lA &L V U U l U 6 A " " U U only fix 3 per cent of the available complement. Neither the aggregated, reduced gamma globulin, nor the complexes fix significant amounts of guinea pig complement whether tested at 4 C. or at 37 C. (Table 2). The effect of a change in concentration of the complexes on the amount of complement fixed is shown in Fig. 7. Increasing the concentration of complex 75 " minmnrsmr 6'""'" tn c u R R mnm n F nrntnin 6- u- r--results in increasing complement fixation. 2 mgm. of preformed complex fixes about 50 per cent of the available complement and 8 mgm. fixes more than 90 per cent. Analysis of the second and third components of complement was performed to see if the rheumatoid factor-reduced gamma globulin complexes fix complement in the Frnm llUlll I A l l A U l U V." 1-1 531 REACTIONS OF TREA'ITD GAMMA GLOBULIN Wl'lX RF 10D-1 - 1 1000 100 10 10.000 vgms Protein Fig. 7.-The effect of increasing concentrations of rheumatoid factor-reduced gamma globulin complexes on ftration of human complement. Preformed complexes were incubated with fresh human serum at 37 C. for 120 minutes. Table l.-lncubatwn of Aggregated, Reduced Gamma Globulin and RF with Human Serum Reagent Residual C'&o % C' Fired Human C'-S7"C for 2 hours: RF AME FII* 55 AME FII* 94 Saline 97 + Human Ct-4'for 18 hours: RF AME FII* 107 AME FII* 95 + Saline 104 *3.0 mgm. protein/ml. 43 3 0 8 - @loConversion +0 0 0 0 0 tration of aggregated, reduced gamma globulin. Skin Tests The intravenous injection of aggregated gamma globulin into experimental animals has been shown to produce profound effects, kcluding lowering of serum complement, and the induction of vascular and renal lesi0n.l7-~~Intradermal injection of less than 1.0 microgram N of aggregated human gamma globulin results in a marked infiammatory response characterized by usual manner. The results are shown in Table 3. At a protein concentration of 75 Table 2.-lncubation of Aggregated, Reduced pgm. the rheumatoid factor-reduced gamma Gamma Globulin and RF with Guinea Pig Serum globulin complex fixed 35 per cent of the Residual C'Hro 46 C' Fixed @lC Conversion Reagent total hemolytic complement and 33 per cent Guinea pig C'37"C for 2 hours: of the available C'2, and reduced the im- RF + AME FII* 219 14 mune adherence titer by 50 per cent. When AMEFII* 234 8 Saline 255 the protein concentration was increased to 6.1 mgm. the preformed complexes &ed 95 Guinea pi0 C'-4"C for 18 hours: per cent of the total complement, 97 per RF + AME FII* 212 9 cent of the available C'2 and reduced the AME FII* 203 13 Saline 234 immune adherence titer to 13 per cent of the value obtained with a similar concen*3.0 mgm. protein/ml. 532 ZVAIFLER AND S C H U R Table 3.-Fixation of Human Conlplement Components (C2 and C 3 ) by Different Concentrations of Rheumatoid FactorAggregated, Reduced Gamma Globulin Complexes Reagent AMEFII RF - AMEFII RF - AME FII AME FII Protein (mgm.) 0.75 0.75 6.1 6.1 C't C't Total C'Hro 160 108 300 8 1600 800 1600 200 67 100 5 (34 erythema, edema and a wheal and flare reaction which begins in several minutes and reaches a peak intensity in 15 to 20 minutes.20~21 To assess the phlogistic properties of heat aggregated 2-mercaptoethan01 gamma globulin the following studies were performed: human gamma globulin (FII)was reduced and alkylated and then passed through a 0.22 p millipore filter prior to heat aggregation. The final protein concentration was adjusted to contain 15 to 40 micrograms N in 0.1 ml. This volume was injected intradermally into 7 normal subjects or patients with non-rheumatic diseases. No inflammatory response was observed immediately or at 24 or 48 hours. The same test was performed in 15 subjects with definite or classical rheumatoid arthritis, 12 of whom had rheumatoid factor in their serum. No inflammatory response was noted in any of them. Gamma globulin was isolated by DEAE cellulose chromatography from the serum of 5 patients with seropositive rheumatoid arthritis. This was reduced and alkylated, millipore filtered and then heat aggregated at 63 C. for 15 minutes. The separated gamma globulin was introduced intradermally into the subjects who supplied the serum; this autologous mercaptoethanoltreated gamma globulin produced no immediate or delayed reactions. DISCUSSION The findings reported in this paper are an extension of the studies of Wiedermann. Miescher and Franklin who showed that when 7s human antibodies or heat aggregated gamma globulin were reduced with 2-mercaptoethanol they lost their ability to fix complement while retaining the capacity to react with rheumatoid factor. Their observations on the reaction of mercaptoethanol treated heat aggregated fraction I1 of human plasma with rheumatoid factor serum were made utilizing a semiquantitative capillary tube precipitation method. As can be seen in Fig. 1, when this reaction is analyzed by the classical precipitin reaction there are both quantitative and qualitative differences from the results obtained with non-reduced gamma globulin. The point of maximum precipitation on the precipitin curve with non-reduced gamma globulin shows no constant relationship to the point of complete neutralization of agglutination, and there is not a well-defined solubilization of the precipitates in the region of antigen The reaction of mercaptoethanol-treated, heataggregated gamma globulin is remarkable because it does have the configuration of a classical antigen-antibody reaction, with a well-defined zone of equivalence and solubilization of the precipitate in the region of antigen excess. Examination of the supernatants for residual rheumatoid agglutinating activity did not demonstrate this activity beyond the point of equivalence. Several explanations for the differences in the precipitin reaction were considered. Since mercaptoethanol treatment interferes with complement fixation, the role of complement was studied. Precipitin curves performed in the presence of 0.01 M EDTA showed a reduction of precipitate nitrogen in the reactions with both reduced and nonreduced aggregated gamma globulin. Heating the serum containing rheumatoid factor to 56 C. prior to the addition of gamma globulin resulted in a definite lowering of REAClTONS OF TREATED GAMMA GLOBULIN WITH RF precipitate nitrogen with both preparations. The reduction of precipitable nitrogen by both of these procedures suggests some complement fixation in the reaction of rheumatoid factor with gamma globulin, but cannot explain the differences in the curves obtained with reduced and non-reduced gamma globulin, because the overall configuration of the 2 precipitin curves remained the same. Another possible effect of the mercaptoethanol treatment might be to make gamma globulin less readily aggregatable and thus provide less “reactant” gamma globulin to combine with rheumatoid factor. This possibility was excluded by the studies in which sodium sulfate fractions were prepared from mercaptoethanol treated gamma globulin. Christian has shown that these fractions (SS1 SS2) contain the majority of the reactant material capable of combining with rheumatoid factor.* With this more concentrated reactant gamma globulin, the precipitin curve showed a better defined zone of equivalence and solubilization of the precipitate in antigen excess. It is apparent that none of the studies performed have elucidated the reason for the differences between the reaction of reduced and non-reduced gamma globulin with rheumatoid factor. It is possible that all gamma globulin molecules are not reduced to a similar degree by mercaptoethanol treatment and that the observed precipitin reaction is between rheumatoid factor and a species of mercaptoethanolresistant gamma globulin. A more likeIy explanation is that the mercaptoethanoltreated gamma globulin complexes with rheumatoid factor when aggregated, but when non-aggregated is less able to solubilize these complexes in the zone of antigen excess. A number of studies have analyzed the influence of rheumatoid factor on comple- + 533 ment fixation. Heimer and his associates described an unusual macroglobulin in rheumatoid serum which inhibited fixation of guinea pig complement by rabbit antigen antibody complexes, but had no effect on complement fixation by latex particles coated with human FII. Isolated, purified rheumatoid factor had the opposite effect, that is, it inhibited complement fixation to latex particles coated with FII, but not with antigen antibody c o m p l e ~ e s . When ~ * ~ latex particles coated with FII were incubated with human or guinea pig serum prior to mixing with dilutions of rheumatoid factor, then a 3- to 4-tube decrease in agglutination titer was ob~erved.~ Zvaifler and Block showed that serum containing rheumatoid factor interfered with complement fixation by both human and rabbit antigen-antibody systems.6 Euglobulin preparations from rheumatoid serum will protect a complement-sensitive enzyme system. Davis and Bollet found that the degree of protection paralleled the latex titer and was removed by prior incubation of the euglobulin preparation with sensitized sheep cells.’ An interesting extension of these studies to an in vivo situation showed that rheumatoid factor could prevent the lowering of serum complement that accompanies the intraperitoneal injection of antigen antibody complexes. This is presumably accomplished by trapping the complexes in the extravascular space.23Thus, in all the studies reported, isolated rheumatoid factor, rheumatoid euglobulin preparations, or heat-inactivated serum containing rheumatoid factor has been shown to compete with complement for binding sites on gamma globulin. The ability of mercaptoethanol reduction to impair complement fixation by gamma globulin without inhibiting its combination with rheumatoid factor suggests that the sites are separate but possibly close together, and that rheumatoid factor interferes by steric hinderance or by pro- 534 ducing conformational changes in the gamma globulin? From all of these studies it has been inferred that the precipitation reaction between rheumatoid factor and aggregated gamma globulin does not fix complement. However, this has not been shown heretofore, because aggregated gamma globulin, by itself, is so markedly anticomplementary. The use of mercaptoethanol-treated gamma globulin makes such an analysis possible. The results obtained show that under appropriate conditions rheumatoid factor can fix complement. When aggregated, reduced gamma globulin is added to heat-inactivated human serum containing rheumatoid factor, there is no fixation of either guinea pig or human complement. But preformed complexes of rheumatoid factor and reduced gamma globulin will fix human complement at 37 C. Several features of this reaction are worthy of comment: 1) the greater efficiency of preformed complexes; 2) the need for complement fixation to occur in the warm; and 3) a requirement for human complement. There is a tendency to assume that complement fixation is more efficient when complement is present during the reaction of antigen and antibody than when it is added to preformed precipitate^.^^ While this is true for rabbit antibody, it does not hold for all antisera.12 Washed preformed precipitates made with sheep or horse antibodies & complement better than those aggregates that combine in the presence of complement. In an investigation of complement hation by 19s antibodies it has been shown that there are certain requiiements for the antigen. These include a large molecular volume and critically spaced repeating determinants on the antigenic molecule.25 It may be that in the rheumatoid factor precipitates, the reduced gamma globulin assumes a more ap- ZVAIFLER AND SCHUR propriate form for initiating complement hation. The greater efficiency of complement fixation at 4 C. is well recognized, but there are exceptions. Stollar has shown that rabbit and human 19s antibodies are actually more efficient at 37 C. than at 4 C., which was never the case with 7s antibodies.25 In studies with human complement this daerence might be exaggerated, since fixation of the first component of human complement to antigen-antibody complexes does not proceed well in the cold.26 The finding that the rheumatoid factor-reduced gamma globulin complex fixes human complement in the warm but not in the cold is consistent with these observations. The absolute requirement for human complement, and the inability to fm guinea pig complement, is most surprising, but not without some precedent. In a study of the ability of naturally occurring anti A and anti B antibodies to fix complement, Ostgard found a sigdcantly higher frequency of positive results with human complement (43 of 54 samples) than with guinea pig complement (6 of 54). The titers were also higher using human complement?’ The immunoglobulin class of the anti A and B antibodies was not investigated, but the “naturally” occurring form is usually a macroglobulin.28 The relevance of these findings to patients with rheumatoid arthritis can only be speculative. The rheumatoid factorreduced gamma globulin complexes are not a very efficient complement fixing system, since approximately 2 mgm. of preformed complexes are required to fix 50 per cent of the available human complement, whereas only 100 pgm. of aggregated human gamma globulin will fix a similar amount. One clinical situation has been described where there are mixed cryoglobulin complexes consisting of Ig M im- 535 m<;TIONS OF TREATED GAMMA GLOBULIN WlTH RF munoglobulins with rheumatoid factor ac- tivity and Ig G molecules; these patients usually have depressed serum complement levels.29 Small amounts of aggregated gamma globulin produce skin reactions, and treatment with urea removes this property.20.21 Mercaptoethanol treatment of rabbit 7s antibody interferes with its ability to fix to skin and give PCA reactions.s0 Therefore, the finding that intradermal injection of aggregated mercaptoethanol treated FII into normal subjects produced no reaction, even in amounts as great as 40 micrograms N, was not surprising. That it produced no reactions in patients with classicd rheumatoid arthritis might be taken as supporting evidence for the inability of rheumatoid suMMA€uo IN factor-gamma globulin complexes to fix complement; for the two properties, complement fixation and phlogistic response, are parallel phenomena. Such an explanation requires that the conditions for antigen-antibody combination are similar in vivo as in uitro, and that rheumatoid factor can get from the intravascular space to an extravascular space before the reduced gamma globulin diffuses away from the injection site. ACKNOWLEDGMENTS The authors wish to thank Mrs. Jennifer Robiuson and Mrs. Annelore Serenyi for excellent technical assistance. Part of these studies were performed in the laboratories of Dr. Elmer Becker, whose interest and encouragement is gratefully acknowledged. hTElUI"TERLINUA Le reduction e alcoylation de globulina gamma human ante le aggregation resulta in un perdita da capacitate pro le fixation de complemento sed in le retention del capacitate pro reager con factor rheumatoide. Le addition de aggregate, reducite glob& gamma a sero que contine alte titros de factor rheumatoide resulta in le production de un typicamente campaniliforme curva de precipitina con un puncto de equivalentia e pauc precipitato in le region de excess0 de antigeno. Viste que aggregate, reducite globulina gamma per se fixa pauc complemento, il esseva possibile determinar si su reaction con factor rheumatoide resulta in le fixation de complemento. Le constatation esseva: Sub appropriate conditiones factor rheumatoide pote fixar complemento. N d e fixation de complemento de porco de India o de complemento human esseva observate quando aggregate, reducite globulina gamma esseva addite a thermoinactivate seros con varie titros de factor rheumatoide, sed preformate complexos de factor rheumatoide e globulina gamma fixa complemento human a 37 C. Le injection intradennal de 40 pg N de aggregate, reducite globulina gamma ad in subjectos de control0 o ad in patientes con arthritis rheumatoide produce nulle immediate o tardive reactions cutanee. REFERENCES 1. Franklin, E. C.: The immune globulin+ their structure and function and some techniques for their isolation. In: Progress in Allergy. New York, S. Karger, 1964, p. 58. 2. Wiedermann, G., Miescher, P. A., and Franklin, E. C.: Effect of mercaptoethanol on complement binding ability of human 7s gamma globulin. Proc. SOC. Exp. Biol. Med. 113:609, 1963. 3. Bernhard, G.G.,Cheng, W., and Talmage, D. W.: The reaction of rheumatoid factor and complement with gamma globulin coated latex. J. Immun. 88:750, 1962. 4. Heimer, R., Levine, F. M., Primach, A., Cor- J. M., and Nosenzo, C.: Inhibition of complement fixation by human serum. J. Immun. 89: 382, 1982. 5. Zvaifier, N. J., and Bloch, K. J.: Rheumatoid factor-an inhibitor of the complement fixation reaction. Arthritis Rheum. 5:127,1962. 6. Heimer, R., Levin, F. M., and Kahn, M. F.: Inhibition of complement fixation by human serum. 11. The activity of a gamma MI globulin and rheumatoid factor in complement fixation reactions. J. Immun. 91:866, 1963. 7. Davis, J. S., and Bollet, A. J.: Protection of a complement sensitive enzyme system by rheumatoid factor. J. Immun. 92:139, 1964. cos, 536 8. Christian, C. L.: Characterization of the “Reactant” (gamma globulin factor) in the FII sheep cell agglutination test. J. Exp. Med. 108: 139, 1958. 9. Vaughan, J. H., Ellis, P. J., and Marshall, H.: Quantitative considerations of the rheumatoid factor. J. Immun. 81:26, 1958. 10. Kabat, E. A., and Mayer, M. M.: Experimental Immunochemistry (ed. 2). Springfield, Ill., Charles C Thomas, 1961. 11. Bloch, K. J., and Bunim, J. J.: Simple, rapid and diagnostic test for rheumatoid arthritisbentonite flocculation test. J.A.M.A. 169:307, 1959. 12. Barbaro, J. F., and Becker, E. L.: Comparison of complement fixing potency of antiserum from various species. J. Infect. Dis. 111:175, 1962. 13. Austen, K. F., and Beer, F.: Measurement of second component of human complement (C’2hu) by its interaction with EAC‘l,gP, 4m cells. J. Immun. 92:946, 1964. 14. Nishioka, K., and Linscott, W. D.: Components of guinea pig complement. I. Separation of a serum fraction essential for immune hemolysis and immune adherence. J. Exp. Med. 118:767, 1963. 15. Scheidigger, J. J.: Une micro-method de 1’ immuno-electrophorese. Arch. Allergy Appl. Immun. 7:103, 1955. 16. Mardiney, M. R., and Muller-Eberhard, H. J.: Mouse B,C globulin: Production of antiserum and characterization in the complement reaction. J. Immun. 949377, 1965. 17. Weigle, W. 0,. and Dixon, F. J.: Relationship of circulating antigen-antibody complexes, antigen elimination and complement fixation in serum sickness. Proc. SOC. Exp. Biol. Med. 99: 226,1958. 18. Rhyne, M. B., and Germuth, F. G., Jr.: The relationship between serum complement activity and the development of allergic lesions in rabbits. J. Exp. Med. 114:633, 1961. 19. McCluskey, R. T., Benacerraf, B., Potter, J. L., and Miller, F.: The pathological effects of intravenously administered soluble antigen-antibody complexes. J. Exp. Med. 111:181, 1960. ZVAIFLER AND -s 20. Christian, C. L.: Studies of aggregated gamma globulin. J. Immun. 84:117, 1960. 21. Ishizaka, K., Ishizaka, T., and Sugakara, T.: Biologic activities of aggregated gamma globulin. 111. Production of Arthus-like reactions. J. Immun. 86:220, 1961. 22. Epstein, W. V., Engleman, E. P., and Ross, M.: Quantitative studies of the precipitation and agglutination reactions between serum of patients with “connective tissue” diseases and a preparation (Cohn Fraction 11) of human gamma globulin. J. Immun. 79441, 1957. 23. Gough, W. W., and Davis, J. S.: Effect of rheumatoid factor on complement levels in vivo. Arthritis Rheum. 9:555, 1966. 24. Hill, B. M., and Osler, A. G.: Kinetic studies of complement fixation. 11. The role of aggregating capacity of antibody and its heterogeneity. J. Immun. 75:146, 1955. 25. Cunniff, R. V., and Stollar, B. D.: Properties of 19s antibodies in complement fixation. I. Temperature dependence and role of antigen structure. J. Immun. 100:7, 1968. 26. Leon, M. A.: Kinetics of human complement. 111. Intermediates in the reaction between human complement and sensitized cells. J. Immun. 79:480, 1957. 27. Ostgard, P., and Orjasaeter, H.: Naturally occurring anti A and anti B and their ability to fix complement. Acta Path. Microbiol. Scand. 56: 426,1962. 28. Kunkel, H. G.: Macroglobulins and high molecular weight antibodies. In F. W. Putnam (ed.): Plasma Proteins. New York, Academic Press, 1960, Vol. 1. pp. 279-307. 29. Meltzer, M., Franklin, E. C., Elias, K., McCluskey, R. T., and Cooper, N.: Cryoglobulinemia-A clinical and laboratory study. 11. Cryoglobulins with rheumatoid factor activity. Amer. J. Med. 40:837, 1966. 30. Wiedermann, G., Ovary, Z., and Miescher, P. A.: Influence of mercaptoethanol treatment of skin sensitivity and complement binding ability of 7s anti dinitrophenol-bovine gamma globulin antibody. Proc. SOC.Exp. Biol. Med. 116:448, 1964.