Scientific Articles
Complement Fixation by Rh Blood Group Antibodies C. HIDALGO, E. L. ROMANO, J. LINARES A N D G. SUAREZ From the lnsiituto Venezolano de Investigaciones CientiJicas, the Servicio de Hematologia y Banco de Sangre, und the Maternidad Concepcibn Palacios, Caracas, Venezuela
Rh blood group antibodies normally do not fix complement. Rh. positive intact red blood cells treated with papain do not lyse when incubated with corresponding antibody and complement. This study was done to determine if complement fixation occurs when antibodies were combined with Rh positive red blood cell ghosts untreated or treated with papain. Complement fixation was observed with IgG anti-D, antiDC and anti-c when papain treated ghosts were used. No complement fixation or a smaller degree of it was observed in the case of untreated Rh positive red blood cell ghosts when incubated with similar anti-Rh antibodies. It is concluded that the papain treatment of Rh positive red blood cell ghosts, possibly by inducing aggregation of antigen sites, allowed complement binding by Rh antibodies.
MOSTof the Rh antibodies, either of the IgG or the IgM class, fail to bind complement although a few examples of complement-binding Rh antibodies are cited in the l i t e r a t ~ r e . ~The * ' ~ majority of the Rh antibodies are IgGl and IgG3 immunoglobulins' although some are of the IgM class.5 IgGl, IgG3 and IgM are known to be complement-binding antibodies in other antigenantibody systems and the reason for the failure of Rh antibodies to bind complement seems to be related to properties of the Rh antigens. A reasonable explanation for the failure of IgG anti-Rh to bind complement is that Rh antigen sites are too far apart on the red blood cell surface so that two IgG molecules cannot be in molecular proximity; that is, within the distance required to bind the first component of complement A change in the surface distribution of Received for publication January 1, 1978: accepted June 21, 1978.
D antigens after protease treatment has been r e p ~ r t e d * .and ~ ~ ,a~ recent ~ paper of Masouredis et af .4 deals extensively with the ' ~ that followsubject. Romano er ~ 1 . found ing papainization the distribution of D antigen sites on the erythrocyte surface could be changed from a dispersed, nonaggregated one to an aggregated distribution. D sites were combined with IgG anti-D and visualized in the electron microscope by a gold-labeled antiglobulin reagent. Theoretically, an aggregated distribution of Rh antigens could make it possible for IgG Rh antibodies to be close enough to bind Clq. Based on this theoretic consideration, it was decided to study whether or not Rh antibodies bind complement when combined with Rh positive red blood cells previously treated with papain. However, when Rh positive intact red bloods cells were treated with papain, washed, sensitized with IgG anti-D and then incubated in the presence of an excess of fresh serum as a source of complement, no lysis was observed. Attempts were made to demonstrate the presence of bound C3K4 in the Rh positive sensitized cells incubated with fresh serum by means of an anti-C3 and anti-C4 antisera. Results were not conclusive because of agglutination of the cells in the absence of anti-complement antisera (unpublished observations). Thus, it was investigated whether or not red blood cell (RBC) ghosts treated with papain would fix complement when combined with specific antibody. For this purpose, RBC ghosts untreated or treated with papain,
0041-1 132/79/0500/0250$00.75 0 J. B. Lippincott Co. Transfusion May-June 1979
250
Volume I Y
Number 3
Volume 19 Number 3
COMPLEMENT AND Rh ANTIBODIES
were combined with Rh antibodies of corresponding specificities, and then complement binding by the Rh-anti-Rh system was assessed by a microcomplement fixation assay. It is demonstrated in this study that the Rh-anti-Rh system is capable of fixing complement under the experimental conditions tested. Materials and Methods Anti-D, anti-c, anti-E and anti-e were commercial anti-Rh typing sera.* All these antisera had a titer of 128 to 256 when tested by an indirect antiglobulin test with red blood cells of corresponding specificities. IgG anti-D was from commercial sources (RhoGAM)* and had an indirect antiglobulin titer of 2048. Anti-DC was obtained from the serum of a woman who gave birth to a child with newborn hemolytic disease due to Rh incompatibility. This antiserum reacted with C positive D negative cells as well as with D positive C negative cells; no attempt was made to further characterize it. This antiserum had an indirect antiglobulin titer of 512 when tested with red blood cells of phenotype DCCee. The IgG fraction of this antiserum was isolated by conventional ion-exchange chromatography using DE-52+ and tris-CI, 0.02 M, pH 8.2 as eluent. IgG anti-A was isolated by ion-exchange chromatography from the serum of a blood group 0 human volunteer. The IgG anti-A preparation had a titer of 512 when tested with adult A cells by an indirect antiglobulin test. All antisera were spun at 120,000 x g for one hour before use in the complement fixation assay to minimize anticomplementarity due to antibody aggregation. A pool of sera from 18 healthy guinea pigs three to six months old was used as source of complement. The guinea pigs were bled by cardiac puncture. The fresh pooled sera was separated in small aliquots and kept frozen at -70 C. Glycerinated sheep cell hemolysin* was used to sensitize indicator red blood cells. It was kept diluted 150 in small volumes frozen at -70 C. The hemolysin was titrated according to the method of Kabat and Mayer.3 As a result, a hemolysin dilution of 1:lOOO in complement
* Ortho Diagnostic Inc., Raritan. NJ. +
*
Whatman Biochemicals Ltd. Hyland. Division of Travenol Laboratories Inc.
25 1
diluent was used to sensitize indicator sheep red blood cells,. . Complement titration was performed by incubating serial dilutions of the guinea pig serum with 2 x lo7 optimally sensitized sheep red blood cells, in a total volume of 1.4 ml. The dilution that gave 90 per cent hemolysis was used, in this case being 1:140. A buffer containing 0.15 M NaCI, 5 mM barbitaUNa-barbital pH 7.4, 0.5 mM MgCI2, 0.15 mM CaCl, and one mu1 of heat-inactivated normal rabbit serum was used. Complement, antibody and RBC ghosts were diluted in this buffer. Sheep red blood cells were obtained locally from the animal house. Blood was obtained from the jugular vein and it was mixed v/v with Alsever's solution. It was kept sterile at 4 C and each lot lasted approximately five weeks. Ghosts from red blood cells of group 0 donors of known Rh phenotypes were obtained by the method of Dodge et a/.* For quantitative purposes, the ghost concentration was expressed as dry weight related to the volume of the suspension. It was assumed that 50 per cent of the dry weight of the RBC ghost is made -up by proteins." Two microliters of twice crystallized papain (BDH, 66.6 mg/ml, four EU/mg protein) were mixed with two ml 0.04 M EDTA and four mlO.02 M 2-mercaptoethanol. This solution was added to 0.9 mg protein RBC ghost and suspended in two ml of 20 mM phosphate buffer pH 8.0. This amount of papain was chosen because it was the least amount which could render Rh positive red blood cells agglutinable by IgG anti-D. The suspension was incubated I5 minutes at 37 C. The ghosts were then separated by centrifugation ( 10,OOO x g ) for 20 minutes, washed three times with 20 mM phosphate buffer pH 8.0, resuspended in two ml of phosphate buffer and kept at 4 C until use. RBC ghosts were used within 24 hours. For control experiments, similar treatment omitting the papain was carried out on ghosts of the same batch. The quantitative microcomplement fixation technique of Wasserman and L e ~ i n e ' was ~ used. The test was performed in plastic disposable tubes. Complement, antibody and RBC ghosts, diluted in the complement diluent buffer were incubated in a water bath at 37 C for one hour in a volume of 1.2 ml. Then, 2 x lo7 sensitized sheep red blood cells. in a volume of 0.2 ml were added and the mixture was further incubated for one hour at 37 C. The degree of hemolysis was assessed by reading the OD at 412 nm of the supernatant fluid. Complement fixation in the experimental tubes
HIDALGO ET AL.
252
Table 1. Complement Fixation by IgG Anti-D and RBC Ghosts Phenotype DCcEe (R, R 2 ) Per Cent Complement Fixation Anti body Dilution
Papain-t reated Ghosts
Untreated Ghosts
1 :75
93 76 68 30 6 0
45 44 44
1:lOO 1:125 1:150 1 :300 1 :600
0 0 0
was determined expressing the difference in OD at 412 nm between the antibody control tube and the experimental tube as a percentage of the OD of the antibody control tube. Antibody control tube refers to those controls in which the RBC ghosts are omitted. The calculations were done in this way because in some experiments, some dilutions of the antibody alone were anticomplementary. RBC ghosts were not found to be anticomplementary.
Results IgG anti-D incubated with RBC ghosts from a subject of phenotype DCcEe resulted in a clear complement fixation. No complement fixation was observed with similar RBC ghosts in the absence of IgG anti-D or with D negative RBC ghosts. The results of this experiment are shown in Table 1. Greater complement fixation occurred with papain-treated RBC ghosts in comparison with untreated ghosts. When red blood cells of Rh phenotype Dccee, which are very likely D-heterozygous, were used very little complement fixation was obtained with papain-treated ghosts, and no fixation was obtained with untreated ghosts. With several other examples of anti-D sera, including commercial anti-Rho (D) typing serum, Table 2. Complement Fixation by Anti-c and RBC Ghosts Phenotype ccddee ( r r ) Per Cent Complement Fixation Antibody Dilution
Papai n-treated Ghosts
Untreated Ghosts
1 :2 1 :3 1 :4 1 :5 1 :6
46 76 47 24 0
0 0 0 12 0
Transfusion May-June 1979
no complement fixation was obtained. These latter results may be due to the relatively low concentration of IgG anti-D contained in the typing sera in comparison with the high concentration present in the RhoGAM preparation. As we did not have any potent antisera with anti-C, anti-E and anti-e specificities, commercial typing sera were used. The gammaglobulin fractions were separated from other components ( i . e . , albumin, sucrose, added by the manufacturers) by conventional precipitations with ammonium sulfate followed by DEAE ion exchange chromatography to isolate and concentrate the IgG fraction. However, at all the dilutions assayed, even at 1:2 of the original concentrations, these commercial antisera did not fix complement with either papaintreated or untreated RBC ghosts of the corresponding specificities. Papain-treated and untreated RBC ghosts from ddccee group donors were incubated with dilutions of commercial anti-c starting at 1:2 of the original concentration. The results are presented in Table 2. Complement fixation was obtained with anti-c and papain-treated ghosts at dilutions ranging from 1:2 to 1:5. At similar dilutions of the antiserum, no significant complement fixation was obtained with untreated ghosts. With anti-DC complement fixation was obtained when papain-treated ghosts from red blood cells of phenotype DCCee were used. However at dilutions of 1 : l O or less, the antiserum alone was strongly anticomplementary and it was difficult to assess if complement were fixed or not. No complement fixation was obtained when untreated ghosts were used. The results of this experiment are presented graphically in Figure 1. Complement fixation was assayed using simultaneously anti-D (RhoGAM) and anti-c combined with RBC ghosts of phenotype DccEe untreated and treated with papain. Tubes in which anti-D and anti-c were tested individually for complement fixation using similar RBC ghosts were also set up in the experiment. The results showed complement fixation with papain-treated RBC ghosts for both antisera tested simultaneously in the same proportion as for each antiserum tested individually, that is, no potentiation or additive effect was observed. It is known that IgG anti-A fixes complement. In this experiment, it was investigated if there were any difference in the binding of complement when papain-treated and untreated RBC ghosts from human blood group A donors were compared. The results of this experiment are presented in Table 3. IgG anti-A, when
Volume 19 Number 3
COMPLEMENT AND Rh ANTIBODIES
combined with papain-treated ghosts, fixed complement at a higher proportion than when combined with untreated ghosts. This result demonstrates again that the protease treatment induced changes in the surface properties of A antigen sites.
253
l00r
Discussion
The results presented here clearly indicate that anti-D, anti-c and anti-DC fix complement when incubated with papain-treated RBC ghosts of corresponding specificities. Thus, with IgG anti-D (RhoGAM), complement fixation was obtained up to a dilution of 1:150 when the antibody was incubated with papain-treated DCcEe ghosts. When untreated ghosts were used little or no complement fixation was obtained. For example, while anti-DC fixed complement up to a dilution of 1:60 in the case of papain-treated ghosts, no complement fixation could be obtained with untreated ghosts at the dilution assayed. To explain the appearance of complement binding ability by anti-Rh antibody when papain-treated ghosts are used, an effect of the papain treatment on the RBC surface antigens sites must be invoked. A possibility is that the protease treatment somehow permits certain changes so as to allow aggregation of antigen sites and consequently the binding of the first component of complement by two or more IgG anti-Rh molecules in molecular proximity. This possibility is also supported by the previous findings of NicoIson,H*Y Romano et a1,12 Victoria et al.,l3 and Masouredis et ~ 1 These . ~ authors showed changes in the distribution of surface antigens induced by proteolysis. Complement fixation by anti-Rh antibodies after protease treatment in the present studies was only obtained when RBC ghosts were used as source of antigens. In preliminary experiments, DCcEe (R1R2) red blood cells treated with papain were incubated with IgG anti-D and then tested for lysis with fresh human serum as a source of complement. No lysis was obtained, In
I ANTIBODY DILUTION FIG. 1.
Complement fixation by anti-DC and DCCee RBC ghosts treated with papain.
a variant of this experiment, when papaintreated red blood cells were combined with anti-D, and incubated with fresh human serum, no C3 o r C4 could be detected by the use of an anti-C3/C4 antiserum (unpublished results). As it is difficult to remove hemoglobin completely from red blood cells and "spectrin" is believed to stabilize the RBC membrane,9J0 it is tempting to postulate that the membrane of the red blood cells might be stabilized by the interaction of spectrin with hemoglobin on the inner side of the red blood Table 3. Complement Fixation by IgG Anti-A and RBC Ghosts Phenotype A , Per Cent Complement Fixation Antibody Dilution 1 :8
1:lO 1:12.5 1:16 1 :25 1 :50
Papain-treated Ghosts 100 88 79 42 8
0
Untreated Ghosts 68 52 10 0 0 0
Transfusion May-June 1979
HIDALGO ET AL.
254
cell membrane. This hypothesis could explain the difficulty of inducing surface changes in intact red blood cells. On the other hand, an alternative explanation for the difference observed in complement binding by intact red blood cells and red blood cell ghosts, could be that some spectrin may be lost in the preparation of the RBC ghosts. This possibility may explain the unexpected complement fixation which was obtained when untreated RBC ghosts were used (Table 1). This is further supported by the observations of Masouredis et al.,4 who found that some degree of antigen clustering may be seen in RBC ghosts in the absence of protease treatment. However, this observation does not invalidate our results because, as pointed out, the degree of antigen redistribution observed on ghosts of enzymemodified cells was significantly greater than on ghosts of untreated cell^.^ The results presented here demonstrate that anti-Rh antibodies are structurally capable of binding complement. These results also suggest that the protease treatment somehow modified the surface properties of the Rh antigens on RBC ghosts so as to allow complement binding in the presence of specific antibodies. References I . Ayland, J., M. A. Horton, P. Tippett, and A. H. Waters: Complement binding anti-D made in a D" variant woman. Vox Sang. 34:40, 1978. 2. Dodge, J. T., C. Mitchell, and D. J. Hanahan: The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Arch. Biochem. Biophys. 100:119, 1%3. 3. Kabat, E. A., and M. M. Mayer: Experimental Immunochemistry. Springfield, C. C Thomas, 1961, p. 151. 4. Masouredis, S. P., E. J. Sudora, and E. J. Victoria: Immunological and electron microscopic analysis of IgG anti-D saline hemag-
glutination of neuraminidase- and proteasemodified red cells. J. Lab. Clin. Med. 90: 929, 1977. 5. Mollison, P. L.: Blood Transfusion in Clinical Medicine. London, Blackwell Scientific, 1972. p. 279. 6. Ibid, p. 282. 7. Natvig, J. B.,and H. G. Kunkel: Genetic markers
of human immunoglobulin. The Gm and Inv systems. Sem. Haematol. 1:66, 1968. 8. Nicolson, G . L.: Topography of membrane Concanavalin A sites modified by proteolysis. Nature 239:193, 1972. 9. : Anionic sites of human erythrocyte membranes. I. Effects of trypsin, phospholipase C and pH on the topography of bound positively charged colloidal particles. J. Cell. Biol. 57:373, 1973. 10.
11. 12.
13.
14.
15.
-,
and R. G. Painter: Anionic sites of human erythrocyte membranes. 11. Antispectnn-induced transmembrane aggregation of the binding sites for positively charged colloidal particles. J. Cell. Biol. 59:395, 1973. Reynolds, J. A.: Red cell membranes: Facts and fancy. Fed. Proc. 322034, 1973. Romano, E. L., C. Stolinski, and N. C. HughesJones: Distribution and mobility of the A, D and c antigens of human red cell membranes: Studies with a gold labeled antiglobulin reagent. Br. J. Haematol. 30507, 1975. Victoria, E. J., E. A. Muchmore, E. J. Sudora, and S. P. Masouredis: The role of antigen mobility in anti-Rho (D)-induced agglutination. J. Clin. Invest. 56292, 1975. Waller, M., and S. D. Lawler: A study of the properties of the Rhesus antibody (Ri) diagnostic for the rheumatoid factor and its application to Gm grouping. Vox Sang. 7595, 1962. Wasserman, E., and L. Levine: Quantitative microcomplement fixation and its use in the study of the antigenic structure by specific antigenantibody inhibition. J. Immunol. 87:290, 1961.
Carlos Hidalgo, B.Sc., The Instituto Venezolano de Investigaciones Cientificas, Fisiopatologia, Apartado 1827, Caracas 101, Venezuela. Egidio L. Romano, M.D., Ph.D., The Instituto Venezolando de Investigaciones Cientificas (correspondence). JesOs Linares, Jefe Banco de Sangre, Maternidad Concepcion Palacios, Caracas, Venezuela. Guillermo Suiirez, B.Sc., The Institute Venezolano de Investigaciones Cientificas.
Complement Fixation by Rh Blood Group Antibodies C. HIDALGO, E. L. ROMANO, J. LINARES A N D G. SUAREZ From the lnsiituto Venezolano de Investigaciones CientiJicas, the Servicio de Hematologia y Banco de Sangre, und the Maternidad Concepcibn Palacios, Caracas, Venezuela
Rh blood group antibodies normally do not fix complement. Rh. positive intact red blood cells treated with papain do not lyse when incubated with corresponding antibody and complement. This study was done to determine if complement fixation occurs when antibodies were combined with Rh positive red blood cell ghosts untreated or treated with papain. Complement fixation was observed with IgG anti-D, antiDC and anti-c when papain treated ghosts were used. No complement fixation or a smaller degree of it was observed in the case of untreated Rh positive red blood cell ghosts when incubated with similar anti-Rh antibodies. It is concluded that the papain treatment of Rh positive red blood cell ghosts, possibly by inducing aggregation of antigen sites, allowed complement binding by Rh antibodies.
MOSTof the Rh antibodies, either of the IgG or the IgM class, fail to bind complement although a few examples of complement-binding Rh antibodies are cited in the l i t e r a t ~ r e . ~The * ' ~ majority of the Rh antibodies are IgGl and IgG3 immunoglobulins' although some are of the IgM class.5 IgGl, IgG3 and IgM are known to be complement-binding antibodies in other antigenantibody systems and the reason for the failure of Rh antibodies to bind complement seems to be related to properties of the Rh antigens. A reasonable explanation for the failure of IgG anti-Rh to bind complement is that Rh antigen sites are too far apart on the red blood cell surface so that two IgG molecules cannot be in molecular proximity; that is, within the distance required to bind the first component of complement A change in the surface distribution of Received for publication January 1, 1978: accepted June 21, 1978.
D antigens after protease treatment has been r e p ~ r t e d * .and ~ ~ ,a~ recent ~ paper of Masouredis et af .4 deals extensively with the ' ~ that followsubject. Romano er ~ 1 . found ing papainization the distribution of D antigen sites on the erythrocyte surface could be changed from a dispersed, nonaggregated one to an aggregated distribution. D sites were combined with IgG anti-D and visualized in the electron microscope by a gold-labeled antiglobulin reagent. Theoretically, an aggregated distribution of Rh antigens could make it possible for IgG Rh antibodies to be close enough to bind Clq. Based on this theoretic consideration, it was decided to study whether or not Rh antibodies bind complement when combined with Rh positive red blood cells previously treated with papain. However, when Rh positive intact red bloods cells were treated with papain, washed, sensitized with IgG anti-D and then incubated in the presence of an excess of fresh serum as a source of complement, no lysis was observed. Attempts were made to demonstrate the presence of bound C3K4 in the Rh positive sensitized cells incubated with fresh serum by means of an anti-C3 and anti-C4 antisera. Results were not conclusive because of agglutination of the cells in the absence of anti-complement antisera (unpublished observations). Thus, it was investigated whether or not red blood cell (RBC) ghosts treated with papain would fix complement when combined with specific antibody. For this purpose, RBC ghosts untreated or treated with papain,
0041-1 132/79/0500/0250$00.75 0 J. B. Lippincott Co. Transfusion May-June 1979
250
Volume I Y
Number 3
Volume 19 Number 3
COMPLEMENT AND Rh ANTIBODIES
were combined with Rh antibodies of corresponding specificities, and then complement binding by the Rh-anti-Rh system was assessed by a microcomplement fixation assay. It is demonstrated in this study that the Rh-anti-Rh system is capable of fixing complement under the experimental conditions tested. Materials and Methods Anti-D, anti-c, anti-E and anti-e were commercial anti-Rh typing sera.* All these antisera had a titer of 128 to 256 when tested by an indirect antiglobulin test with red blood cells of corresponding specificities. IgG anti-D was from commercial sources (RhoGAM)* and had an indirect antiglobulin titer of 2048. Anti-DC was obtained from the serum of a woman who gave birth to a child with newborn hemolytic disease due to Rh incompatibility. This antiserum reacted with C positive D negative cells as well as with D positive C negative cells; no attempt was made to further characterize it. This antiserum had an indirect antiglobulin titer of 512 when tested with red blood cells of phenotype DCCee. The IgG fraction of this antiserum was isolated by conventional ion-exchange chromatography using DE-52+ and tris-CI, 0.02 M, pH 8.2 as eluent. IgG anti-A was isolated by ion-exchange chromatography from the serum of a blood group 0 human volunteer. The IgG anti-A preparation had a titer of 512 when tested with adult A cells by an indirect antiglobulin test. All antisera were spun at 120,000 x g for one hour before use in the complement fixation assay to minimize anticomplementarity due to antibody aggregation. A pool of sera from 18 healthy guinea pigs three to six months old was used as source of complement. The guinea pigs were bled by cardiac puncture. The fresh pooled sera was separated in small aliquots and kept frozen at -70 C. Glycerinated sheep cell hemolysin* was used to sensitize indicator red blood cells. It was kept diluted 150 in small volumes frozen at -70 C. The hemolysin was titrated according to the method of Kabat and Mayer.3 As a result, a hemolysin dilution of 1:lOOO in complement
* Ortho Diagnostic Inc., Raritan. NJ. +
*
Whatman Biochemicals Ltd. Hyland. Division of Travenol Laboratories Inc.
25 1
diluent was used to sensitize indicator sheep red blood cells,. . Complement titration was performed by incubating serial dilutions of the guinea pig serum with 2 x lo7 optimally sensitized sheep red blood cells, in a total volume of 1.4 ml. The dilution that gave 90 per cent hemolysis was used, in this case being 1:140. A buffer containing 0.15 M NaCI, 5 mM barbitaUNa-barbital pH 7.4, 0.5 mM MgCI2, 0.15 mM CaCl, and one mu1 of heat-inactivated normal rabbit serum was used. Complement, antibody and RBC ghosts were diluted in this buffer. Sheep red blood cells were obtained locally from the animal house. Blood was obtained from the jugular vein and it was mixed v/v with Alsever's solution. It was kept sterile at 4 C and each lot lasted approximately five weeks. Ghosts from red blood cells of group 0 donors of known Rh phenotypes were obtained by the method of Dodge et a/.* For quantitative purposes, the ghost concentration was expressed as dry weight related to the volume of the suspension. It was assumed that 50 per cent of the dry weight of the RBC ghost is made -up by proteins." Two microliters of twice crystallized papain (BDH, 66.6 mg/ml, four EU/mg protein) were mixed with two ml 0.04 M EDTA and four mlO.02 M 2-mercaptoethanol. This solution was added to 0.9 mg protein RBC ghost and suspended in two ml of 20 mM phosphate buffer pH 8.0. This amount of papain was chosen because it was the least amount which could render Rh positive red blood cells agglutinable by IgG anti-D. The suspension was incubated I5 minutes at 37 C. The ghosts were then separated by centrifugation ( 10,OOO x g ) for 20 minutes, washed three times with 20 mM phosphate buffer pH 8.0, resuspended in two ml of phosphate buffer and kept at 4 C until use. RBC ghosts were used within 24 hours. For control experiments, similar treatment omitting the papain was carried out on ghosts of the same batch. The quantitative microcomplement fixation technique of Wasserman and L e ~ i n e ' was ~ used. The test was performed in plastic disposable tubes. Complement, antibody and RBC ghosts, diluted in the complement diluent buffer were incubated in a water bath at 37 C for one hour in a volume of 1.2 ml. Then, 2 x lo7 sensitized sheep red blood cells. in a volume of 0.2 ml were added and the mixture was further incubated for one hour at 37 C. The degree of hemolysis was assessed by reading the OD at 412 nm of the supernatant fluid. Complement fixation in the experimental tubes
HIDALGO ET AL.
252
Table 1. Complement Fixation by IgG Anti-D and RBC Ghosts Phenotype DCcEe (R, R 2 ) Per Cent Complement Fixation Anti body Dilution
Papain-t reated Ghosts
Untreated Ghosts
1 :75
93 76 68 30 6 0
45 44 44
1:lOO 1:125 1:150 1 :300 1 :600
0 0 0
was determined expressing the difference in OD at 412 nm between the antibody control tube and the experimental tube as a percentage of the OD of the antibody control tube. Antibody control tube refers to those controls in which the RBC ghosts are omitted. The calculations were done in this way because in some experiments, some dilutions of the antibody alone were anticomplementary. RBC ghosts were not found to be anticomplementary.
Results IgG anti-D incubated with RBC ghosts from a subject of phenotype DCcEe resulted in a clear complement fixation. No complement fixation was observed with similar RBC ghosts in the absence of IgG anti-D or with D negative RBC ghosts. The results of this experiment are shown in Table 1. Greater complement fixation occurred with papain-treated RBC ghosts in comparison with untreated ghosts. When red blood cells of Rh phenotype Dccee, which are very likely D-heterozygous, were used very little complement fixation was obtained with papain-treated ghosts, and no fixation was obtained with untreated ghosts. With several other examples of anti-D sera, including commercial anti-Rho (D) typing serum, Table 2. Complement Fixation by Anti-c and RBC Ghosts Phenotype ccddee ( r r ) Per Cent Complement Fixation Antibody Dilution
Papai n-treated Ghosts
Untreated Ghosts
1 :2 1 :3 1 :4 1 :5 1 :6
46 76 47 24 0
0 0 0 12 0
Transfusion May-June 1979
no complement fixation was obtained. These latter results may be due to the relatively low concentration of IgG anti-D contained in the typing sera in comparison with the high concentration present in the RhoGAM preparation. As we did not have any potent antisera with anti-C, anti-E and anti-e specificities, commercial typing sera were used. The gammaglobulin fractions were separated from other components ( i . e . , albumin, sucrose, added by the manufacturers) by conventional precipitations with ammonium sulfate followed by DEAE ion exchange chromatography to isolate and concentrate the IgG fraction. However, at all the dilutions assayed, even at 1:2 of the original concentrations, these commercial antisera did not fix complement with either papaintreated or untreated RBC ghosts of the corresponding specificities. Papain-treated and untreated RBC ghosts from ddccee group donors were incubated with dilutions of commercial anti-c starting at 1:2 of the original concentration. The results are presented in Table 2. Complement fixation was obtained with anti-c and papain-treated ghosts at dilutions ranging from 1:2 to 1:5. At similar dilutions of the antiserum, no significant complement fixation was obtained with untreated ghosts. With anti-DC complement fixation was obtained when papain-treated ghosts from red blood cells of phenotype DCCee were used. However at dilutions of 1 : l O or less, the antiserum alone was strongly anticomplementary and it was difficult to assess if complement were fixed or not. No complement fixation was obtained when untreated ghosts were used. The results of this experiment are presented graphically in Figure 1. Complement fixation was assayed using simultaneously anti-D (RhoGAM) and anti-c combined with RBC ghosts of phenotype DccEe untreated and treated with papain. Tubes in which anti-D and anti-c were tested individually for complement fixation using similar RBC ghosts were also set up in the experiment. The results showed complement fixation with papain-treated RBC ghosts for both antisera tested simultaneously in the same proportion as for each antiserum tested individually, that is, no potentiation or additive effect was observed. It is known that IgG anti-A fixes complement. In this experiment, it was investigated if there were any difference in the binding of complement when papain-treated and untreated RBC ghosts from human blood group A donors were compared. The results of this experiment are presented in Table 3. IgG anti-A, when
Volume 19 Number 3
COMPLEMENT AND Rh ANTIBODIES
combined with papain-treated ghosts, fixed complement at a higher proportion than when combined with untreated ghosts. This result demonstrates again that the protease treatment induced changes in the surface properties of A antigen sites.
253
l00r
Discussion
The results presented here clearly indicate that anti-D, anti-c and anti-DC fix complement when incubated with papain-treated RBC ghosts of corresponding specificities. Thus, with IgG anti-D (RhoGAM), complement fixation was obtained up to a dilution of 1:150 when the antibody was incubated with papain-treated DCcEe ghosts. When untreated ghosts were used little or no complement fixation was obtained. For example, while anti-DC fixed complement up to a dilution of 1:60 in the case of papain-treated ghosts, no complement fixation could be obtained with untreated ghosts at the dilution assayed. To explain the appearance of complement binding ability by anti-Rh antibody when papain-treated ghosts are used, an effect of the papain treatment on the RBC surface antigens sites must be invoked. A possibility is that the protease treatment somehow permits certain changes so as to allow aggregation of antigen sites and consequently the binding of the first component of complement by two or more IgG anti-Rh molecules in molecular proximity. This possibility is also supported by the previous findings of NicoIson,H*Y Romano et a1,12 Victoria et al.,l3 and Masouredis et ~ 1 These . ~ authors showed changes in the distribution of surface antigens induced by proteolysis. Complement fixation by anti-Rh antibodies after protease treatment in the present studies was only obtained when RBC ghosts were used as source of antigens. In preliminary experiments, DCcEe (R1R2) red blood cells treated with papain were incubated with IgG anti-D and then tested for lysis with fresh human serum as a source of complement. No lysis was obtained, In
I ANTIBODY DILUTION FIG. 1.
Complement fixation by anti-DC and DCCee RBC ghosts treated with papain.
a variant of this experiment, when papaintreated red blood cells were combined with anti-D, and incubated with fresh human serum, no C3 o r C4 could be detected by the use of an anti-C3/C4 antiserum (unpublished results). As it is difficult to remove hemoglobin completely from red blood cells and "spectrin" is believed to stabilize the RBC membrane,9J0 it is tempting to postulate that the membrane of the red blood cells might be stabilized by the interaction of spectrin with hemoglobin on the inner side of the red blood Table 3. Complement Fixation by IgG Anti-A and RBC Ghosts Phenotype A , Per Cent Complement Fixation Antibody Dilution 1 :8
1:lO 1:12.5 1:16 1 :25 1 :50
Papain-treated Ghosts 100 88 79 42 8
0
Untreated Ghosts 68 52 10 0 0 0
Transfusion May-June 1979
HIDALGO ET AL.
254
cell membrane. This hypothesis could explain the difficulty of inducing surface changes in intact red blood cells. On the other hand, an alternative explanation for the difference observed in complement binding by intact red blood cells and red blood cell ghosts, could be that some spectrin may be lost in the preparation of the RBC ghosts. This possibility may explain the unexpected complement fixation which was obtained when untreated RBC ghosts were used (Table 1). This is further supported by the observations of Masouredis et al.,4 who found that some degree of antigen clustering may be seen in RBC ghosts in the absence of protease treatment. However, this observation does not invalidate our results because, as pointed out, the degree of antigen redistribution observed on ghosts of enzymemodified cells was significantly greater than on ghosts of untreated cell^.^ The results presented here demonstrate that anti-Rh antibodies are structurally capable of binding complement. These results also suggest that the protease treatment somehow modified the surface properties of the Rh antigens on RBC ghosts so as to allow complement binding in the presence of specific antibodies. References I . Ayland, J., M. A. Horton, P. Tippett, and A. H. Waters: Complement binding anti-D made in a D" variant woman. Vox Sang. 34:40, 1978. 2. Dodge, J. T., C. Mitchell, and D. J. Hanahan: The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Arch. Biochem. Biophys. 100:119, 1%3. 3. Kabat, E. A., and M. M. Mayer: Experimental Immunochemistry. Springfield, C. C Thomas, 1961, p. 151. 4. Masouredis, S. P., E. J. Sudora, and E. J. Victoria: Immunological and electron microscopic analysis of IgG anti-D saline hemag-
glutination of neuraminidase- and proteasemodified red cells. J. Lab. Clin. Med. 90: 929, 1977. 5. Mollison, P. L.: Blood Transfusion in Clinical Medicine. London, Blackwell Scientific, 1972. p. 279. 6. Ibid, p. 282. 7. Natvig, J. B.,and H. G. Kunkel: Genetic markers
of human immunoglobulin. The Gm and Inv systems. Sem. Haematol. 1:66, 1968. 8. Nicolson, G . L.: Topography of membrane Concanavalin A sites modified by proteolysis. Nature 239:193, 1972. 9. : Anionic sites of human erythrocyte membranes. I. Effects of trypsin, phospholipase C and pH on the topography of bound positively charged colloidal particles. J. Cell. Biol. 57:373, 1973. 10.
11. 12.
13.
14.
15.
-,
and R. G. Painter: Anionic sites of human erythrocyte membranes. 11. Antispectnn-induced transmembrane aggregation of the binding sites for positively charged colloidal particles. J. Cell. Biol. 59:395, 1973. Reynolds, J. A.: Red cell membranes: Facts and fancy. Fed. Proc. 322034, 1973. Romano, E. L., C. Stolinski, and N. C. HughesJones: Distribution and mobility of the A, D and c antigens of human red cell membranes: Studies with a gold labeled antiglobulin reagent. Br. J. Haematol. 30507, 1975. Victoria, E. J., E. A. Muchmore, E. J. Sudora, and S. P. Masouredis: The role of antigen mobility in anti-Rho (D)-induced agglutination. J. Clin. Invest. 56292, 1975. Waller, M., and S. D. Lawler: A study of the properties of the Rhesus antibody (Ri) diagnostic for the rheumatoid factor and its application to Gm grouping. Vox Sang. 7595, 1962. Wasserman, E., and L. Levine: Quantitative microcomplement fixation and its use in the study of the antigenic structure by specific antigenantibody inhibition. J. Immunol. 87:290, 1961.
Carlos Hidalgo, B.Sc., The Instituto Venezolano de Investigaciones Cientificas, Fisiopatologia, Apartado 1827, Caracas 101, Venezuela. Egidio L. Romano, M.D., Ph.D., The Instituto Venezolando de Investigaciones Cientificas (correspondence). JesOs Linares, Jefe Banco de Sangre, Maternidad Concepcion Palacios, Caracas, Venezuela. Guillermo Suiirez, B.Sc., The Institute Venezolano de Investigaciones Cientificas.
