- Email: [email protected]
anti-DNA complexes. IV. complement fixation
Journal of Immunological Methods, 38 (1980) 269--280 © Elsevier/North-Holland Biomedical Press
269
S T A B I L I T Y O F D N A / A N T I - D N A C O M P L E X E S . IV. C O M P L E M E N T FIXATION
STEEN E. PEDERSEN, RONALD P. TAYLOR, KATHI W. MORLEY and ELEANOR L. WRIGHT
Department of Biochemistry, University of Virginia School of Medicine, Charlottesville, VA 22908, U.S.A. (Received 22 April 1980, accepted 25 June 1980)
We describe an in vitro assay to study complement fixation of antibody/dsDNA immune complexes formed from SLE sera and radiolabeled dsDNA. The method measures the amount of radiolabeled dsDNA which is part of an immune complex bound to red blood cells via the C3b complement component receptor. The assay is dependent upon active complement, red blood cells, and anti-dsDNA antibodies, but it is independent of the red blood cell donor (type O) and the age of the red blood cells (up to 10 days). The method has been compared in some detail with the Farr assay with respect to the antibody/dsDNA ratio in the immune complexes and the relative stability of the complexes as measured by their resistance to dissociation by excess unlabeled dsDNA. Our results indicate that multiple binding of antibodies to dsDNA is required for complement fixation, and that a significant percentage of those antibodies which fix complement are of high avidity. Finally, a double label assay using both [3H]- and [14C]dsDNA indicates that the complement fixing potential of the anti-dsDNA antibodies in an SLE serum is strongly influenced by the order of mixing of the isotopes with the serum. The DNA isotope which is added to the serum first is considerably more effective at fixing complement than the isotope which is added 1 h later. The implications of these results with respect to the pathogenesis of SLE are discussed.
INTRODUCTION T h e r e c o g n i t i o n o f d s D N A b y specific a n t i - d s D N A a n t i b o d i e s in the sera o f patients with systemic lupus e r y t h e m a t o s u s (SLE) is believed to be a crucial step in the p a t h o g e n e s i s o f t h e disease (Tan et al., 1 9 6 6 ; Koffler, 1 9 7 4 ; Stollar, 1 9 7 5 ; Winfield et al., 1977). E x p e r i m e n t a l evidence f r o m a n u m b e r o f l a b o r a t o r i e s n o w suggests t h a t o n e o f t h e p o t e n t i a l c o n s e q u e n c e s o f this b i n d i n g step, c o m p l e m e n t fixation, is a l m o s t invariably associated clinically with g l o m e r u l o n e p h r i t i s in SLE ( R o t h f i e l d and Stollar, 1 9 6 7 ; Sonth e i m e r a n d Gilliam, 1 9 7 8 ; Ballou and K u s h n e r , 1 9 7 9 ; Beaulieu et al., 1 9 7 9 ; Minter et al., 1 9 7 9 ) . T h u s it appears to be p a r t i c u l a r l y i m p o r t a n t t h a t t h e f a c t o r s w h i c h i n f l u e n c e t h e c o m p l e m e n t f i x a t i o n r e a c t i o n ( a n t i b o d y avidity, c o m p l e x c o m p o s i t i o n , etc.) be u n d e r s t o o d in detail.
270 Studies from this laboratory on the rate of dissociation of antibody/ dsDNA complexes indicate that a significant fraction (typically 50% or more) of the anti-dsDNA antibodies in SLE sera bind with such high avidity to dsDNA that the half-life for dissociation of the immune complexes can be m a n y hours at 37°C (Taylor et al., 1979; Riley et al., 1980). This result led us to the observation that the size of the antibody/dsDNA complexes formed in vitro can be controlled kinetically (Taylor et al., 1980). That is, the size of the complexes depends upon the order of addition of a given a m o u n t of dsDNA to an SLE serum. It would certainly seem reasonable, therefore, that the immunological properties of these immune complexes, and in particular their ability to fix complement, would also be governed by similar considerations. In this paper we investigate this possibility. We report the development of a quantitative method to measure the a m o u n t of radiolabeled dsDNA which is part of complement fixing antibody/dsDNA immune complexes prepared in vitro. The method facilitates the detection of these complexes by making use of the presence of C3b receptors on human erythrocytes (RBCs). Our results indicate that the potential complement fixing properties of anti-dsDNA antibodies in SLE sera depend upon the antibody/dsDNA ratio in the immune complexes which t h e y can form. This stoichiometry in turn is to a great extent determined by the sequence of mixing of antib o d y with varying quantities of DNA. The results of a double-label study provide clear proof that in fact the order of mixing does play a direct role in the complement-fixing reaction in the antibody/dsDNA system. The possible importance of this finding with respect to the pathogenesis of SLE is briefly considered. MATERIALS AND METHODS Patients, sera and D N A Sera were obtained from patients with SLE receiving medical care at the University of Virginia Hospital. Sera were heated at 56°C for 25 min to inactivate complement (HI), briefly centrifuged at 3000 X g to remove any aggregated materials, and then stored at --20°C in small aliquots until used. The DNA used in the experiments was either [3H]- or [14C]dsPM2 DNA as described previously (Riley et al., 1980). In a few pilot experiments sonicated dsDNA (MW ca. 5 × 105) was used in place of the dsPM2 DNA. I m m u n e comlexes of this DNA with antibody also fixed complement and were bound to the RBCs but in analogy to the results often seen in the Farr assay, the sensitivity of the assay decreased. Measurement of D N A binding by SLE s e r a - complement fixation and the Farr assay The procedure described by Tsuda et al. (1979) to measure immune
271 complexes has been specifically adapted to study radiolabeled dsDNA in an in vitro prepared immune complex. RBCs are obtained from normal healthy volunteers (blood group O, Rh positive) and are stored and washed following the procedure of Tsuda et al. Unless otherwise stated, the dilution buffer for dsDNA, sera, RBCs, and complement is GVB 2÷ (see Tsuda et al., 1979). Our standard protocol to measure the binding to RBCs of radiolabeled dsDNA/ a n t i b o d y complexes which fix complement is as follows: 125 pl of diluted serum is added to a 1.4 ml polyethylene disposable tube to which had previously been added 375 gl of a solution containing a total of between 20 and 400 ng of [3H]dsPM2 DNA. After the mixture is incubated for 1 h at 37°C, 4 aliquots of 100 pl each are withdrawn and transferred to polyethylene disposable tubes (400 pl m a x i m u m volume). 50 pl of the remaining mixture are removed and transferred to a scintillation vial along with 1.4 ml of water and 10 ml of Beckman Ready-Solve EP in order to obtain a 'total counts' in the experiment. Two of the tubes are examined for complement fixation: 100 pl of a washed 25% suspension of RBCs are added along with 50 pl of a solution of guinea pig complement (29 mg/ml, prepared from lyophilized guinea pig complement, Miles Laboratories). After the tubes are incubated at 37°C for 20 min t h e y are spun at 8000 × g for 5 min at 4 °C. 125 gl of the supernatant are counted along with 1.4 ml of water and 10 ml of Beckman ReadySolve EP. The a m o u n t of [3H]dsDNA bound is calculated by standard procedures appropriate to supernatant counting in radioimmunoassays (Winfield et al., 1977; Taylor et al., 1979). The reproducibility in duplicate points averaged -+5%. The background binding by normal human serum in these experiments was always less than 10% and averaged about 5--7%. The other two tubes are examined for DNA binding in the Farr assay. 25 pl of GVB 2÷ and 125 pl of saturated a m m o n i u m sulfate (SAS) are added and after centrifugation the a m o u n t of [3H]dsDNA bound is determined by counting 125 #l of the supernatant. In a few pilot studies the initial incubation of antibody, DNA and buffer (at 37°C in a final volume of 500 pl) was decreased to just 15 min, and identical results were obtained (compared to an initial 1 h incubation) in both the Farr and complement-fixing assay. We have also noted that these assay results are not particularly sensitive to the incubation time of the antibody/dsDNA complexes with RBCs and complement. If the incubation is prolonged to 1 h, for example, there is a drop in binding to 70% from a control value of 90%, but this level remains constant over m a n y hours of further incubation at 37°C. Possibly some C3b is converted to C3b' and further breakdown products (Porter and Reid, 1979) during the extended incubations, but this does n o t appear to be a serious problem. An incubation time of 20 min at 37°C was found to be the o p t i m u m time for the assay. In a few instances this last incubation was conducted at 4°C instead of 37°C and the a m o u n t of dsDNA bound was usually lower than that observed at 37 ° C. In addition those SLE sera which did not fix complement with dsDNA
272 at 37 ° C (see below) also failed to do so at 4 ° C. In our preliminary studies normal human serum which had n o t been heat inactivated was used as a source o f c o m p l e m e n t . Overall the results were similar to those in which guinea pig c o m p l e m e n t was used, but at times the a m o u n t of DNA b o u n d was somewhat lower (by ca. 25%). In general the reproducibility of our studies was b e t t e r using guinea pig serum instead o f normal hum an serum as a source of c o m p l e m e n t , and this m ay be related to differences in levels of c o m p l e m e n t for different donors, problems with processing and storing of sera, etc. The 'background binding' o f HI-SLE sera in which no additional source of c o m p l e m e n t was added was quite low (see below) and so we are c o n f i d e n t t hat there was no significant c o n t r i b u t i o n to the binding from endogenous c o m p l e m e n t in HI-SLE sera. In addition, we have f o u n d t hat if the RBCs are stored in a sterile Alsever's solution (1 part red blood cells, 2 parts Alsever's) t h e y are quite stable and the results o f cont r o l experiments c o n d u c t e d 10 days apart with the same red blood cells were in excellent agreement. In addition, under the typical experimental conditions we have described, the c o m p l e m e n t - f i x a t i o n reaction led to no visible hemolysis of the RBCs. Finally, we not e t hat the assay we describe might more accurately be considered one of c o m p l e m e n t activation, rather than c o m p l e m e n t fixation. We study activation of c o m p l e m e n t by a n t i b o d y / D N A complexes, and the a m o u n t of c o m p l e m e n t actually consumed is n o t measured. However, in r ecen t years m os t reactions of i m mune complexes with c o m p l e m e n t have generally been described as involving c o m p l e m e n t fixation, w het her b o u n d C3 was detected, i m m une hemolysis measured, etc, and so we will adhere to this convention. Dissociation studies The above protocols were modified slightly to examine the stability of a n t i b o d y / d s D N A complexes which could fix com pl em ent . In our 'inhibition' studies an excess of unlabeled salmon roe DNA (ca. 60 pg) is first mixed with the radiolabeled dsDNA, and t hen a n t i b o d y is added. In the reversal studies the antibody/radiolabeled dsDNA complexes are prepared in the usual manner. After 1 h of incubation at 37°C excess unlabeled salmon roe DNA is added, and the system is incubated for one more h o u r at 37°C b ef o r e it is examined via the Farr and red blood cell com pl em ent -fi xat i on (RBC-CF) assays. Double-label studies are p e r f o r m e d in a similar manner. First [3H]dsPM2 DNA is incubated with the SLE serum for 1 h at 37°C, t hen [14C]dsDNA is added for a second 1 h incubation at 37°C, and t hen the sample is examined in the Farr and c o m p l e m e n t - f i x a t i o n assay. Controls included reversing the order of addition of the isotopes, as well as mixing t hem t o g eth er before adding the antibody.
273 RESULTS
Extensive control experiments indicate that RBCs do indeed bind dsDNA in soluble a n t i b o d y / d s D N A complexes which have fixed complement (Table 1). For example, if either the source of complement or the RBCs are omitted, the binding values are reduced to background levels. It is clear that active complement is essential for the assay, because when it is heat inactivated or absorbed with heat aggregated human IgG (known to fix complement), binding in the assay is eliminated. However, we do not propose to use the assay to quantitate endogenous anti-complementary immune complexes, and in fact we perform the experiment under conditions of RBC and complement excess, so it is possible to definitely measure complement activation by the a n t i b o d y / [ 3 H ] D N A complexes which we prepare. The results do not depend particularly on the source of the RBCs. Four different donors were used for a particular positive control experiment and the range of binding varied only a b o u t 10% for all four (between 70 and 80%). At high levels of binding the assay may be slightly limited by RBCs (but not complement, Fig. 1). However, it can be seen that most of the DNA is bound for sufficiently high antibody input (Fig. 2). We also note that considerably more antibody is needed for complement fixation than for just DNA binding as in the Farr assay {Fig. 2). In addition, at sufficiently high ratios of DNA to antibody the a m o u n t of DNA b o u n d in the complement assay levels off, or even decreases to background values (Fig. 3).
TABLE 1 Binding to RBC o f c o m p l e m e n t - f i x i n g a n t i b o d y / d s D N A c o m p l e x e s p r e p a r e d f r o m SLE plasma m u (6-fold dilution) a n d a total o f ca. 65 ng [ 3 H ] D N A . Evaluation o f the m e t h o d . Protocol
% DNA bound
Protocol
% DNA bound
S t a n d a r d assay (see experimental section) Omit complement O m i t red b l o o d cells B u f f e r c o n t a i n s 0.01 M E D T A G u i n e a pig c o m p l e m e n t was previously h e a t inactivated ( 5 6 ° C for 25 m i n ) Mu, 7S s u b f r a c t i o n e
80
S t a n d a r d assay, NHS a
< 10
<5 <5 <5 <5
S t a n d a r d assay, SLE s e r u m Ta a,b Mu and NHS c Mu and T a d B u f f e r c o n t a i n s 0.4 m g / m l h e a t aggregated h u m a n IgG as well as m u
<: 10 70 75 < 15
80
a N o r m a l h u m a n s e r u m , used u n d i l u t e d . b SLE s e r u m Ta b o u n d D N A in t h e Farr assay, b u t n o t in t h e RBC-CF assay. c T h e i n c u b a t i o n m i x t u r e c o n t a i n e d a 6-fold d i l u t i o n o f m u and u n d i l u t e d NHS. d T h e i n c u b a t i o n m i x t u r e c o n t a i n e d a 6-fold d i l u t i o n o f m u and u n d i l u t e d Ta. e See Waller et al. ( 1 9 8 0 ) for a m o r e c o m p l e t e s t u d y o f c o m p l e m e n t activation b y 7S subf r a c t i o n s o f SLE p l a s m a a n d sera.
274
IO0
75 nl Z £b p -r
50
25
I 0 25
I 5
I IO
I 20
i 40
Final RBC Concentration (Percent)
Fig. 1. [3H]DNA binding in the RBC-CF assay for plasma mu as a function of the final red blood cell concentration for different conditions. The final RBC concentration for the standard protocol was 10%. Experimental points were all taken for plasma mu at different serum dilutions and complement concentrations, as follows: 0, 3-fold dilution, twice the standard complement concentration; ©, 3-fold dilution, standard complement; A, 6-fold dilution, twice standard complement; A, 6-fold dilution, standard complement; G, 10-fold dilution, standard c o m p l e m e n t ; . , 20-fold dilution, standard complement; 0, 50-fold dilution, standard complement. In each experiment ca. 65 ng of [3H]dsPM2 DNA was used per assay (i.e., per 500 pl of incubation mixture).
However, under these conditions, d e t e c t e d in t h e F a r r assay.
s i g n i f i c a n t D N A b i n d i n g a c t i v i t y is still
S i m i l a r c u r v e s s u c h as t h o s e s e e n in F i g . 2 h a v e also b e e n o b t a i n e d f o r t h e I g G ( 7 S ) s u b f r a c t i o n o f p l a s m a m u , a n d a n u m b e r o f o t h e r S L E 7S sub-
I00
~
o
2A
2B
o
O0
o
,~ 7~
o~
•
•
°°
75
o
Z
9 ~ 5c
50
a~ 25
25 o
% 8
52
128
512
2o4B
Reciprocal Serum Dilution
819~
2
8
32
128
512
Reciprocal Serum Dilution
Fig. 2. A: [3H]DNA binding in the RBC-CF (o) assay and the Farr assay (o) as a function of reciprocal dilution for plasma mu in the standard assay for a total input of ca. 100 ng of [3H]dsPM2 DNA. The results were collected over 1 month in 6 separate experiments using 4 different RBC preparations from 3 donors. B: same as A for serum Ka. A total [3H]dsPM2 DNA input of ca. 65 ng was used. The results were collected over 2 days for a single RBC preparation.
275 40
s@S
50
Z
,,"
20
//
Z
/
2/
i
1ZS
I@
O0
I0
25
50
75
nq ~'H-DNA Total Fig. 3. [ 3 H ] D N A b i n d i n g in t h e R B C - C F assay as a f u n c t i o n o f [ 3 H ] D N A c o n c e n t r a t i o n for s e r u m Ka (e, 2-fold d i l u t i o n ) , m u (©, 8-fold d i l u t i o n ) a n d m u (A, 32-fold d i l u t i o n ) . T h e a m o u n t o f D N A b o u n d in t h e Farr assay for these sera at t h e h i g h e s t D N A i n p u t ( 8 0 ng) was 79 ng, 77 ng a n d 68 ng, respectively. T h e a m o u n t o f D N A i n p u t in t h e assay a n d b o u n d has b e e n c a l c u l a t e d p e r 1 0 0 pl o f t h e orginal i n c u b a t i o n m i x t u r e . T h u s t h e t o t a l a m o u n t o f D N A used in t h e original 5 0 0 pl o f i n c u b a t i o n m i x t u r e (see experim e n t a l s e c t i o n for details) was 4 0 0 ng for t h e highest D N A i n p u t .
fractions as well (Waller et al., 1980). The relationship between immune complex size and complement activation in these systems will be reported separately (Waller et al., 1 9 8 0 ) . Only a b o u t 40% of the sera which bound dsDNA at high titer in the Farr assay were also positive in the standard complement fixation assay. TABLE 2 C o m p a r i s o n o f D N A b i n d i n g c a p a c i t y o f selected SLE sera a. Sera w h i c h fix c o m p l e m e n t w i t h d s D N A Patient
Mu To Ka Su c
Sera w h i c h do n o t fix complement with dsDNA d
Reciprocal titer ( R B C - C F assay) b
Reciprocal titer ( F a r r assay)
Patient
Reciprocal titer ( F a r r assay)
16 6 12 <1
256 50 128 10
F1 Cs Ta Ch
50 12 12 32
a T h e reciprocal titers r e p r e s e n t d i l u t i o n s n e c e s s a r y t o achieve 50% b i n d i n g o f D N A in t h e assay used. Mu is a p l a s m a sample, all o t h e r s are sera. b R e d b l o o d c e l l - c o m p l e m e n t f i x a t i o n assay. c U n d i l u t e d s e r u m Su b o u n d 30% of t h e i n p u t D N A in t h e R B C - C F assay. d T h e sera d o n o t b i n d D N A in t h e R B C - C F assay, even w h e n t h e y are u n d i l u t e d .
276 TABLE 3 D i s s o c i a t i o n at 3 7 ° C for 1 h o f a n t i b o d y / d s D N A c o m p l e x e s w h i c h fix c o m p l e m e n t . C o m p a r i s o n o f t h e R B C - C F assay a n d t h e F a r r assay a % D N A b o u n d , R B C - C F assay
% D N A b o u n d , F a r r assay
Patient
Binding
Inhibition
Reversal
Binding
Inhibition
Reversal
Mu b Mu c Ka d
91 62 86
0 0 0
68 42 48
98 98 99
9 0 0
98 97 97
a See e x p e r i m e n t a l s e c t i o n for details. b 10-fold dilution. c 20-fold dilution. d 8-fold d i l u t i o n ; a f t e r 5 h o f reversal t h e % D N A b o u n d in t h e R B C - C F a s s a y was 38%, a n d it w a s 35% a f t e r 24 h.
TABLE 4 Double-label experiment. Effects of the order of mixing on the generation of a n t i b o d y / d s D N A c o m p l e x e s w h i c h fix c o m p l e m e n t a D N A b o u n d (ng) Plasma mu
3H b i n d i n g b 3 H first, t h e n 14C c 14C b i n d i n g b 14C first, t h e n 3H c Both isotopes mixed Total [3H]dsPM2 DNA d (ng) T o t a l [ 14C ] d s P M 2 D N A (ng)
Serum To
Serum Ka
3H
a4 C
3H
14 C
3H
14 C
bound
bound
bound
bound
bound
bound
-1.1 8.3 7.6 2.9
16.9 14.3 -2.2 9.2
-0 23.4 21.6 6.5
18 13 -0 3
-1.8 12 12 4.5
7.6 7 -1.3 2.1 23
23
23
35
35
35
a T h e s e r u m d i l u t i o n s u s e d w e r e as f o l l o w s : M u , 16; T o , 1.5; Ka, 6. b T h e a m o u n t o f D N A b o u n d in a s t a n d a r d R B C - C F a s s a y was m e a s u r e d . c A f t e r t h e first D N A i s o t o p e h a d i n c u b a t e d w i t h t h e s e r u m f o r 1 h at 3 7 ° C , t h e s e c o n d D N A i s o t o p e was a d d e d , a n d a f t e r o n e m o r e h o u r o f i n c u b a t i o n at 3 7 ° C b i n d i n g o f e a c h i s o t o p e was d e t e r m i n e d in t h e R B C - C F assay. d All r e s u l t s are c a l c u l a t e d per 1 0 0 pl o f t h e i n c u b a t i o n m i x t u r e . e In all cases at least 80% o f b o t h i s o t o p e s w e r e b o u n d , w h e n t h e m i x t u r e s w e r e e x a m i n e d in t h e F a r r assay.
277 Those sera which had high titers in the Farr assay were more likely to form complement-fixing complexes with dsDNA (Table 2), but we feel we have not yet surveyed a sufficient number of sera to establish a trend. The inability of certain SLE sera to form complement-fixing complexes with dsDNA is apparently n o t due to the fact that they are intrinsically anti-complementary (presumably due to the presence of immune complexes). Even when very small amounts of DNA of high specific activity were used, m a n y of these sera still did n o t bind DNA in the RBC-CF assay (data not shown). Such sera could not inhibit the complement fixation reaction in sera which were positive in the assay (Table 1). However, in one case an SLE serum bound more dsDNA in the RBC-CF assay at a 2-fold dilution than when it was undiluted. This result may indicate that a sufficient number of immune complexes was present in this particular serum to partially inhibit the RBC-CF assay at high serum levels. Finally, when a number of SLE sera which were positive in the Farr assay but negative in the complem e n t assay were mixed, the mixtures remained negative in the complement assay, although they still bound DNA in the Farr assay (results not shown). Our results indicate that a significant fraction of those prepared complexes which do fix complement are rather stable to dissociation by excess cold DNA at 37°C (Table 3). Our double-label complement binding studies provide further evidence along these lines. It is clear that the a m o u n t of dsDNA which becomes part of a complex which actually fixes complement is critically dependent upon the order of mixing of antibody and dsDNA (Table 4). It should also be noted that in all cases both isotopes were almost completely bound in the Farr assay. DISCUSSION
We have developed a simple quantitative m e t h o d to measure accurately the a m o u n t of antigen (in this case DNA) which is part of an immune complex which actually fixes complement. The complement-fixing potential of the immune complex clearly depends upon the ratio of antibody to DNA. Under conditions of large DNA excess (i,e., just one or two antib o d y molecules per DNA; Aarden et al., 1976) the complexes do not fix complement, although DNA binding can easily be demonstrated in the Farr assay. For example, this is readily seen in Fig. 2B, at serum dilutions between 1/32 and 1/1024 where only Farr binding activity is detectable. In addition, Fig. 3 (the lower two curves) demonstrates that the complementfixing potential of a given serum dilution can be saturated. At a sufficiently high input of DNA the complexes which are formed are apparently so small that they no longer fix complement, although there is still binding in the Farr assay. Our results suggest that ca. 8--20 times as much serum is needed to achieve comparable DNA binding in the RBC-CF assay as in the Farr assay.
278 The relative sensitivity of the m e t h o d thus compares favorably with other studies in which the serum concentration necessary for binding of DNA is compared to the serum concentration needed for effective complement fixation. For example, Ballou and Kushner (1979) used the Cruthidia luciliae immunofluorescence test to measure anti-dsDNA antibodies and they also found that in general ca. 4--80 times as much serum was needed to achieve complement fixation (using fiuorescein-labeled antibodies to C-3) as opposed to just DNA binding by antibodies. Our results definitely implicate t h e high avidity anti-dsDNA antibodies (Taylor et al., 1979; Riley et al., 1980) in complement fixation. Our studies demonstrate t h a t once multiple binding of antibody to dsDNA occurs (a necessary condition for complement fixation), a significant fraction of the bound antibodies must be high avidity, because the complement-fixing potential of the complex remains high (Table 3), even after an excess of DNA is used in an a t t e m p t to dissociate the complex. This implies that the ability of an antibody/dsDNA complex to fix complement can be controlled quantitatively by the a m o u n t of DNA added as well as when it is added. For example, consider Fig. 3, the middle curve. If 16 ng of dsDNA are added to plasma mu, immune complexes which fix complement are formed, but if 80 ng are added all at once, no complement fixation is observed. One would therefore predict that if 80 ng were added sequentially to the plasma in an aliquot of 16 ng followed by 64 ng 1 h later, the immune complexes formed in the first hour would continue to fix complement, although the equilibrium property of the system would obviously suggest no complement fixation. We note that considerably less 'reversal' is seen in the Farr assay than in the RBC-CF assay (Tables 3 and 4). Under the conditions of the experiment, antibody was in considerable excess over DNA with respect to the Farr assay, and as just one antibody per DNA is apparently needed to 'register' in the Farr assay (Aarden et al., 1976), it is not surprising that little dissociation could be detected by this assay. The importance of the order of mixing of antibody and DNA is further seen in the double-label studies. The dsDNA which is added first fixes complement considerably more efficiently than the second DNA added. That is, the dsDNA which is added first binds most of the high avidity antibodies in large complexes which fix complement. Most of these antibodies are n o t 'available' when the second dsDNA is added, and thus this DNA fixes complement weakly, at best. However, there are sufficient antib o d y molecules available to at least bind the second DNA isotope and register in the Farr assay. It is likely that this is due to re-equilibration of lower avidity antibodies (Riley et al., 1980). To summarize, it seems clear that the potential of a given SLE serum to form antibody/dsDNA complexes which fix complement depends critically upon the antibody/dsDNA ratio in the complexes. This stoichiometry in turn is strongly influenced by the specific sequence of addition of DNA
279 to antibody because the complexes are rather stable and dissociate slowly. All other things equal, the complex sizes can clearly be manipulated such that for a given final concentration of dsDNA, the level of complement binding can be insignificant or very high. The relevance of this discussion in terms of lupus pathogenesis is as follows. It is believed that complement fixation by anti-dsDNA antibodies is a crucial factor in tissue destruction. Whether or not the antibodies which circulate in the serum of a given patient actually do fix complement will clearly depend upon the stoichiometry of the complexes they form with dsDNA. This stoichiometry in turn can be governed b y those processes that lead to release of dsDNA into the blood stream. It is therefore possible that if large quantities of dsDNA are rapidly released into the blood, small complexes which will n o t fix complement will be formed, and perhaps they will be cleared without tissue destruction. Obviously the situation may be considerably more complicated, b u t it does raise the possibility that patients (or more appropriately, NZB/W mice) (Lambert and Dixon, 1968) might be titrated therapeutically with appropriate amounts of dsDNA of a given molecular weight. Finally, we note that the technique we have described is considerably different quantitatively from most c o m m o n l y used protocols (Tung et al., 1978; Zubler and Lambert, 1978) which determine the amount of antibody (usually in equivalents of aggregated IgG) in an immune complex. It is well known that aggregated IgG is used simply as a standard in such experiments and that they convey no information with respect to immune complex size and composition. It is possible that, under a given set of experimental conditions, there could be quite a few IgG molecules b o u n d to DNA, but only two of these could actually have fixed complement. Thus, the quantitative meaning of a measurement which determines total IgG in an immune complex could be misleading. We recognize that the exact theoretical relationship between the density of antibody molecules on a given DNA and their complement-fixing potential remains to be established, b u t it is clear that the methods we have used along with procedures for the measurement of total human IgG in a system (Waller et al., 1979) have the potential of generating the data which is needed to examine this interesting question in detail. ACKNOWLEDGEMENTS It is a pleasure to thank Ms. Naomi W. Taylor for her useful experimental advice. S.E.P. thanks M 2. This work was supported by a grant from the W. Alton Jones Foundation, and b y NIH Grants AM 24083 and AM 11766.
280 REFERENCES Aarden, L.A., F. Lakmaker and R.R. DeGroot, 1976, J. Immunol. Methods 11, 153. Ballou, S.P. and I. Kushner, 1979, Clin. Exp. Immunol. 37, 58. Beaulieu, A., F.P. Quismorio, R.C. Kitridou and G.J. Friou, 1979, J. Rheum. 6,389. Koffler, D.,. 1974, Ann. Rev. Med. 25,149. Lambert, P.H. and F.J. Dixon, 1968, J. Exp. Med. 127, 507. Minter, M.H., B.D. Stollar and V. Agnello, 1979, Arthr. Rheum. 22, 959. Porter, R.R. and K.B.M. Reid, 1979, Adv. Prot. Chem. 33, 1. Rapp, H.J. and T. Borsos, 1970, Molecular Basis of Complement Action (Meredith Corporation, New York). Riley, R.L., D.J. Addis and R.P. Taylor, 1980, J. Immunol. 124, 1. Rothfield, N.F. and B.D. Stollar, 1967, J. Clin. Invest. 46, 1785. Sontheimer, R.D. and J.N. Gilliam, 1978, Clin. Immunol. Immunopathol. 10, 459. Stollar, B.D., 1975, Crit. Rev. Biochem. 3, 45. Tan, E.M., P.H. Schur, R.I. Carr and H.G. Kunkel, 1966, J. Clin. Invest. 45, 1732. Taylor, R.P., D. Weber, A.V. Broccoli and J.B. Winfield, 1979, J. Immunol. 122,115. Taylor, R.P., S.J. Waller, C. Haden and D.J. Addis, 1980, J. Immunol. 124, 2571. Tsuda, F., Y. Miyakawa and M. Mayumi, 1979, Immunology 37,681. Tung, K.S.K., A.J. Woodroffe, T.D. Ahlin, R.C. Williams and C.B. Wilson, 1978, J. Clin. Invest. 62, 61. Waller, S.J., R.P. Taylor and B.S. Andrews, 1979, J. Immunol. Methods 30,171. Waller, S.J., R.P. Taylor, E.L. Wright and M. Johns, 1980, submitted to Arthr. Rheum. Winfield, J.B., I. Faiferman and D. Koffler, 1977, J. Clin. Invest. 59, 90. Zubler, R.H. and P.H. Lambert, 1978, Prog. Allergy 24, 1.
269
S T A B I L I T Y O F D N A / A N T I - D N A C O M P L E X E S . IV. C O M P L E M E N T FIXATION
STEEN E. PEDERSEN, RONALD P. TAYLOR, KATHI W. MORLEY and ELEANOR L. WRIGHT
Department of Biochemistry, University of Virginia School of Medicine, Charlottesville, VA 22908, U.S.A. (Received 22 April 1980, accepted 25 June 1980)
We describe an in vitro assay to study complement fixation of antibody/dsDNA immune complexes formed from SLE sera and radiolabeled dsDNA. The method measures the amount of radiolabeled dsDNA which is part of an immune complex bound to red blood cells via the C3b complement component receptor. The assay is dependent upon active complement, red blood cells, and anti-dsDNA antibodies, but it is independent of the red blood cell donor (type O) and the age of the red blood cells (up to 10 days). The method has been compared in some detail with the Farr assay with respect to the antibody/dsDNA ratio in the immune complexes and the relative stability of the complexes as measured by their resistance to dissociation by excess unlabeled dsDNA. Our results indicate that multiple binding of antibodies to dsDNA is required for complement fixation, and that a significant percentage of those antibodies which fix complement are of high avidity. Finally, a double label assay using both [3H]- and [14C]dsDNA indicates that the complement fixing potential of the anti-dsDNA antibodies in an SLE serum is strongly influenced by the order of mixing of the isotopes with the serum. The DNA isotope which is added to the serum first is considerably more effective at fixing complement than the isotope which is added 1 h later. The implications of these results with respect to the pathogenesis of SLE are discussed.
INTRODUCTION T h e r e c o g n i t i o n o f d s D N A b y specific a n t i - d s D N A a n t i b o d i e s in the sera o f patients with systemic lupus e r y t h e m a t o s u s (SLE) is believed to be a crucial step in the p a t h o g e n e s i s o f t h e disease (Tan et al., 1 9 6 6 ; Koffler, 1 9 7 4 ; Stollar, 1 9 7 5 ; Winfield et al., 1977). E x p e r i m e n t a l evidence f r o m a n u m b e r o f l a b o r a t o r i e s n o w suggests t h a t o n e o f t h e p o t e n t i a l c o n s e q u e n c e s o f this b i n d i n g step, c o m p l e m e n t fixation, is a l m o s t invariably associated clinically with g l o m e r u l o n e p h r i t i s in SLE ( R o t h f i e l d and Stollar, 1 9 6 7 ; Sonth e i m e r a n d Gilliam, 1 9 7 8 ; Ballou and K u s h n e r , 1 9 7 9 ; Beaulieu et al., 1 9 7 9 ; Minter et al., 1 9 7 9 ) . T h u s it appears to be p a r t i c u l a r l y i m p o r t a n t t h a t t h e f a c t o r s w h i c h i n f l u e n c e t h e c o m p l e m e n t f i x a t i o n r e a c t i o n ( a n t i b o d y avidity, c o m p l e x c o m p o s i t i o n , etc.) be u n d e r s t o o d in detail.
270 Studies from this laboratory on the rate of dissociation of antibody/ dsDNA complexes indicate that a significant fraction (typically 50% or more) of the anti-dsDNA antibodies in SLE sera bind with such high avidity to dsDNA that the half-life for dissociation of the immune complexes can be m a n y hours at 37°C (Taylor et al., 1979; Riley et al., 1980). This result led us to the observation that the size of the antibody/dsDNA complexes formed in vitro can be controlled kinetically (Taylor et al., 1980). That is, the size of the complexes depends upon the order of addition of a given a m o u n t of dsDNA to an SLE serum. It would certainly seem reasonable, therefore, that the immunological properties of these immune complexes, and in particular their ability to fix complement, would also be governed by similar considerations. In this paper we investigate this possibility. We report the development of a quantitative method to measure the a m o u n t of radiolabeled dsDNA which is part of complement fixing antibody/dsDNA immune complexes prepared in vitro. The method facilitates the detection of these complexes by making use of the presence of C3b receptors on human erythrocytes (RBCs). Our results indicate that the potential complement fixing properties of anti-dsDNA antibodies in SLE sera depend upon the antibody/dsDNA ratio in the immune complexes which t h e y can form. This stoichiometry in turn is to a great extent determined by the sequence of mixing of antib o d y with varying quantities of DNA. The results of a double-label study provide clear proof that in fact the order of mixing does play a direct role in the complement-fixing reaction in the antibody/dsDNA system. The possible importance of this finding with respect to the pathogenesis of SLE is briefly considered. MATERIALS AND METHODS Patients, sera and D N A Sera were obtained from patients with SLE receiving medical care at the University of Virginia Hospital. Sera were heated at 56°C for 25 min to inactivate complement (HI), briefly centrifuged at 3000 X g to remove any aggregated materials, and then stored at --20°C in small aliquots until used. The DNA used in the experiments was either [3H]- or [14C]dsPM2 DNA as described previously (Riley et al., 1980). In a few pilot experiments sonicated dsDNA (MW ca. 5 × 105) was used in place of the dsPM2 DNA. I m m u n e comlexes of this DNA with antibody also fixed complement and were bound to the RBCs but in analogy to the results often seen in the Farr assay, the sensitivity of the assay decreased. Measurement of D N A binding by SLE s e r a - complement fixation and the Farr assay The procedure described by Tsuda et al. (1979) to measure immune
271 complexes has been specifically adapted to study radiolabeled dsDNA in an in vitro prepared immune complex. RBCs are obtained from normal healthy volunteers (blood group O, Rh positive) and are stored and washed following the procedure of Tsuda et al. Unless otherwise stated, the dilution buffer for dsDNA, sera, RBCs, and complement is GVB 2÷ (see Tsuda et al., 1979). Our standard protocol to measure the binding to RBCs of radiolabeled dsDNA/ a n t i b o d y complexes which fix complement is as follows: 125 pl of diluted serum is added to a 1.4 ml polyethylene disposable tube to which had previously been added 375 gl of a solution containing a total of between 20 and 400 ng of [3H]dsPM2 DNA. After the mixture is incubated for 1 h at 37°C, 4 aliquots of 100 pl each are withdrawn and transferred to polyethylene disposable tubes (400 pl m a x i m u m volume). 50 pl of the remaining mixture are removed and transferred to a scintillation vial along with 1.4 ml of water and 10 ml of Beckman Ready-Solve EP in order to obtain a 'total counts' in the experiment. Two of the tubes are examined for complement fixation: 100 pl of a washed 25% suspension of RBCs are added along with 50 pl of a solution of guinea pig complement (29 mg/ml, prepared from lyophilized guinea pig complement, Miles Laboratories). After the tubes are incubated at 37°C for 20 min t h e y are spun at 8000 × g for 5 min at 4 °C. 125 gl of the supernatant are counted along with 1.4 ml of water and 10 ml of Beckman ReadySolve EP. The a m o u n t of [3H]dsDNA bound is calculated by standard procedures appropriate to supernatant counting in radioimmunoassays (Winfield et al., 1977; Taylor et al., 1979). The reproducibility in duplicate points averaged -+5%. The background binding by normal human serum in these experiments was always less than 10% and averaged about 5--7%. The other two tubes are examined for DNA binding in the Farr assay. 25 pl of GVB 2÷ and 125 pl of saturated a m m o n i u m sulfate (SAS) are added and after centrifugation the a m o u n t of [3H]dsDNA bound is determined by counting 125 #l of the supernatant. In a few pilot studies the initial incubation of antibody, DNA and buffer (at 37°C in a final volume of 500 pl) was decreased to just 15 min, and identical results were obtained (compared to an initial 1 h incubation) in both the Farr and complement-fixing assay. We have also noted that these assay results are not particularly sensitive to the incubation time of the antibody/dsDNA complexes with RBCs and complement. If the incubation is prolonged to 1 h, for example, there is a drop in binding to 70% from a control value of 90%, but this level remains constant over m a n y hours of further incubation at 37°C. Possibly some C3b is converted to C3b' and further breakdown products (Porter and Reid, 1979) during the extended incubations, but this does n o t appear to be a serious problem. An incubation time of 20 min at 37°C was found to be the o p t i m u m time for the assay. In a few instances this last incubation was conducted at 4°C instead of 37°C and the a m o u n t of dsDNA bound was usually lower than that observed at 37 ° C. In addition those SLE sera which did not fix complement with dsDNA
272 at 37 ° C (see below) also failed to do so at 4 ° C. In our preliminary studies normal human serum which had n o t been heat inactivated was used as a source o f c o m p l e m e n t . Overall the results were similar to those in which guinea pig c o m p l e m e n t was used, but at times the a m o u n t of DNA b o u n d was somewhat lower (by ca. 25%). In general the reproducibility of our studies was b e t t e r using guinea pig serum instead o f normal hum an serum as a source of c o m p l e m e n t , and this m ay be related to differences in levels of c o m p l e m e n t for different donors, problems with processing and storing of sera, etc. The 'background binding' o f HI-SLE sera in which no additional source of c o m p l e m e n t was added was quite low (see below) and so we are c o n f i d e n t t hat there was no significant c o n t r i b u t i o n to the binding from endogenous c o m p l e m e n t in HI-SLE sera. In addition, we have f o u n d t hat if the RBCs are stored in a sterile Alsever's solution (1 part red blood cells, 2 parts Alsever's) t h e y are quite stable and the results o f cont r o l experiments c o n d u c t e d 10 days apart with the same red blood cells were in excellent agreement. In addition, under the typical experimental conditions we have described, the c o m p l e m e n t - f i x a t i o n reaction led to no visible hemolysis of the RBCs. Finally, we not e t hat the assay we describe might more accurately be considered one of c o m p l e m e n t activation, rather than c o m p l e m e n t fixation. We study activation of c o m p l e m e n t by a n t i b o d y / D N A complexes, and the a m o u n t of c o m p l e m e n t actually consumed is n o t measured. However, in r ecen t years m os t reactions of i m mune complexes with c o m p l e m e n t have generally been described as involving c o m p l e m e n t fixation, w het her b o u n d C3 was detected, i m m une hemolysis measured, etc, and so we will adhere to this convention. Dissociation studies The above protocols were modified slightly to examine the stability of a n t i b o d y / d s D N A complexes which could fix com pl em ent . In our 'inhibition' studies an excess of unlabeled salmon roe DNA (ca. 60 pg) is first mixed with the radiolabeled dsDNA, and t hen a n t i b o d y is added. In the reversal studies the antibody/radiolabeled dsDNA complexes are prepared in the usual manner. After 1 h of incubation at 37°C excess unlabeled salmon roe DNA is added, and the system is incubated for one more h o u r at 37°C b ef o r e it is examined via the Farr and red blood cell com pl em ent -fi xat i on (RBC-CF) assays. Double-label studies are p e r f o r m e d in a similar manner. First [3H]dsPM2 DNA is incubated with the SLE serum for 1 h at 37°C, t hen [14C]dsDNA is added for a second 1 h incubation at 37°C, and t hen the sample is examined in the Farr and c o m p l e m e n t - f i x a t i o n assay. Controls included reversing the order of addition of the isotopes, as well as mixing t hem t o g eth er before adding the antibody.
273 RESULTS
Extensive control experiments indicate that RBCs do indeed bind dsDNA in soluble a n t i b o d y / d s D N A complexes which have fixed complement (Table 1). For example, if either the source of complement or the RBCs are omitted, the binding values are reduced to background levels. It is clear that active complement is essential for the assay, because when it is heat inactivated or absorbed with heat aggregated human IgG (known to fix complement), binding in the assay is eliminated. However, we do not propose to use the assay to quantitate endogenous anti-complementary immune complexes, and in fact we perform the experiment under conditions of RBC and complement excess, so it is possible to definitely measure complement activation by the a n t i b o d y / [ 3 H ] D N A complexes which we prepare. The results do not depend particularly on the source of the RBCs. Four different donors were used for a particular positive control experiment and the range of binding varied only a b o u t 10% for all four (between 70 and 80%). At high levels of binding the assay may be slightly limited by RBCs (but not complement, Fig. 1). However, it can be seen that most of the DNA is bound for sufficiently high antibody input (Fig. 2). We also note that considerably more antibody is needed for complement fixation than for just DNA binding as in the Farr assay {Fig. 2). In addition, at sufficiently high ratios of DNA to antibody the a m o u n t of DNA b o u n d in the complement assay levels off, or even decreases to background values (Fig. 3).
TABLE 1 Binding to RBC o f c o m p l e m e n t - f i x i n g a n t i b o d y / d s D N A c o m p l e x e s p r e p a r e d f r o m SLE plasma m u (6-fold dilution) a n d a total o f ca. 65 ng [ 3 H ] D N A . Evaluation o f the m e t h o d . Protocol
% DNA bound
Protocol
% DNA bound
S t a n d a r d assay (see experimental section) Omit complement O m i t red b l o o d cells B u f f e r c o n t a i n s 0.01 M E D T A G u i n e a pig c o m p l e m e n t was previously h e a t inactivated ( 5 6 ° C for 25 m i n ) Mu, 7S s u b f r a c t i o n e
80
S t a n d a r d assay, NHS a
< 10
<5 <5 <5 <5
S t a n d a r d assay, SLE s e r u m Ta a,b Mu and NHS c Mu and T a d B u f f e r c o n t a i n s 0.4 m g / m l h e a t aggregated h u m a n IgG as well as m u
<: 10 70 75 < 15
80
a N o r m a l h u m a n s e r u m , used u n d i l u t e d . b SLE s e r u m Ta b o u n d D N A in t h e Farr assay, b u t n o t in t h e RBC-CF assay. c T h e i n c u b a t i o n m i x t u r e c o n t a i n e d a 6-fold d i l u t i o n o f m u and u n d i l u t e d NHS. d T h e i n c u b a t i o n m i x t u r e c o n t a i n e d a 6-fold d i l u t i o n o f m u and u n d i l u t e d Ta. e See Waller et al. ( 1 9 8 0 ) for a m o r e c o m p l e t e s t u d y o f c o m p l e m e n t activation b y 7S subf r a c t i o n s o f SLE p l a s m a a n d sera.
274
IO0
75 nl Z £b p -r
50
25
I 0 25
I 5
I IO
I 20
i 40
Final RBC Concentration (Percent)
Fig. 1. [3H]DNA binding in the RBC-CF assay for plasma mu as a function of the final red blood cell concentration for different conditions. The final RBC concentration for the standard protocol was 10%. Experimental points were all taken for plasma mu at different serum dilutions and complement concentrations, as follows: 0, 3-fold dilution, twice the standard complement concentration; ©, 3-fold dilution, standard complement; A, 6-fold dilution, twice standard complement; A, 6-fold dilution, standard complement; G, 10-fold dilution, standard c o m p l e m e n t ; . , 20-fold dilution, standard complement; 0, 50-fold dilution, standard complement. In each experiment ca. 65 ng of [3H]dsPM2 DNA was used per assay (i.e., per 500 pl of incubation mixture).
However, under these conditions, d e t e c t e d in t h e F a r r assay.
s i g n i f i c a n t D N A b i n d i n g a c t i v i t y is still
S i m i l a r c u r v e s s u c h as t h o s e s e e n in F i g . 2 h a v e also b e e n o b t a i n e d f o r t h e I g G ( 7 S ) s u b f r a c t i o n o f p l a s m a m u , a n d a n u m b e r o f o t h e r S L E 7S sub-
I00
~
o
2A
2B
o
O0
o
,~ 7~
o~
•
•
°°
75
o
Z
9 ~ 5c
50
a~ 25
25 o
% 8
52
128
512
2o4B
Reciprocal Serum Dilution
819~
2
8
32
128
512
Reciprocal Serum Dilution
Fig. 2. A: [3H]DNA binding in the RBC-CF (o) assay and the Farr assay (o) as a function of reciprocal dilution for plasma mu in the standard assay for a total input of ca. 100 ng of [3H]dsPM2 DNA. The results were collected over 1 month in 6 separate experiments using 4 different RBC preparations from 3 donors. B: same as A for serum Ka. A total [3H]dsPM2 DNA input of ca. 65 ng was used. The results were collected over 2 days for a single RBC preparation.
275 40
s@S
50
Z
,,"
20
//
Z
/
2/
i
1ZS
I@
O0
I0
25
50
75
nq ~'H-DNA Total Fig. 3. [ 3 H ] D N A b i n d i n g in t h e R B C - C F assay as a f u n c t i o n o f [ 3 H ] D N A c o n c e n t r a t i o n for s e r u m Ka (e, 2-fold d i l u t i o n ) , m u (©, 8-fold d i l u t i o n ) a n d m u (A, 32-fold d i l u t i o n ) . T h e a m o u n t o f D N A b o u n d in t h e Farr assay for these sera at t h e h i g h e s t D N A i n p u t ( 8 0 ng) was 79 ng, 77 ng a n d 68 ng, respectively. T h e a m o u n t o f D N A i n p u t in t h e assay a n d b o u n d has b e e n c a l c u l a t e d p e r 1 0 0 pl o f t h e orginal i n c u b a t i o n m i x t u r e . T h u s t h e t o t a l a m o u n t o f D N A used in t h e original 5 0 0 pl o f i n c u b a t i o n m i x t u r e (see experim e n t a l s e c t i o n for details) was 4 0 0 ng for t h e highest D N A i n p u t .
fractions as well (Waller et al., 1980). The relationship between immune complex size and complement activation in these systems will be reported separately (Waller et al., 1 9 8 0 ) . Only a b o u t 40% of the sera which bound dsDNA at high titer in the Farr assay were also positive in the standard complement fixation assay. TABLE 2 C o m p a r i s o n o f D N A b i n d i n g c a p a c i t y o f selected SLE sera a. Sera w h i c h fix c o m p l e m e n t w i t h d s D N A Patient
Mu To Ka Su c
Sera w h i c h do n o t fix complement with dsDNA d
Reciprocal titer ( R B C - C F assay) b
Reciprocal titer ( F a r r assay)
Patient
Reciprocal titer ( F a r r assay)
16 6 12 <1
256 50 128 10
F1 Cs Ta Ch
50 12 12 32
a T h e reciprocal titers r e p r e s e n t d i l u t i o n s n e c e s s a r y t o achieve 50% b i n d i n g o f D N A in t h e assay used. Mu is a p l a s m a sample, all o t h e r s are sera. b R e d b l o o d c e l l - c o m p l e m e n t f i x a t i o n assay. c U n d i l u t e d s e r u m Su b o u n d 30% of t h e i n p u t D N A in t h e R B C - C F assay. d T h e sera d o n o t b i n d D N A in t h e R B C - C F assay, even w h e n t h e y are u n d i l u t e d .
276 TABLE 3 D i s s o c i a t i o n at 3 7 ° C for 1 h o f a n t i b o d y / d s D N A c o m p l e x e s w h i c h fix c o m p l e m e n t . C o m p a r i s o n o f t h e R B C - C F assay a n d t h e F a r r assay a % D N A b o u n d , R B C - C F assay
% D N A b o u n d , F a r r assay
Patient
Binding
Inhibition
Reversal
Binding
Inhibition
Reversal
Mu b Mu c Ka d
91 62 86
0 0 0
68 42 48
98 98 99
9 0 0
98 97 97
a See e x p e r i m e n t a l s e c t i o n for details. b 10-fold dilution. c 20-fold dilution. d 8-fold d i l u t i o n ; a f t e r 5 h o f reversal t h e % D N A b o u n d in t h e R B C - C F a s s a y was 38%, a n d it w a s 35% a f t e r 24 h.
TABLE 4 Double-label experiment. Effects of the order of mixing on the generation of a n t i b o d y / d s D N A c o m p l e x e s w h i c h fix c o m p l e m e n t a D N A b o u n d (ng) Plasma mu
3H b i n d i n g b 3 H first, t h e n 14C c 14C b i n d i n g b 14C first, t h e n 3H c Both isotopes mixed Total [3H]dsPM2 DNA d (ng) T o t a l [ 14C ] d s P M 2 D N A (ng)
Serum To
Serum Ka
3H
a4 C
3H
14 C
3H
14 C
bound
bound
bound
bound
bound
bound
-1.1 8.3 7.6 2.9
16.9 14.3 -2.2 9.2
-0 23.4 21.6 6.5
18 13 -0 3
-1.8 12 12 4.5
7.6 7 -1.3 2.1 23
23
23
35
35
35
a T h e s e r u m d i l u t i o n s u s e d w e r e as f o l l o w s : M u , 16; T o , 1.5; Ka, 6. b T h e a m o u n t o f D N A b o u n d in a s t a n d a r d R B C - C F a s s a y was m e a s u r e d . c A f t e r t h e first D N A i s o t o p e h a d i n c u b a t e d w i t h t h e s e r u m f o r 1 h at 3 7 ° C , t h e s e c o n d D N A i s o t o p e was a d d e d , a n d a f t e r o n e m o r e h o u r o f i n c u b a t i o n at 3 7 ° C b i n d i n g o f e a c h i s o t o p e was d e t e r m i n e d in t h e R B C - C F assay. d All r e s u l t s are c a l c u l a t e d per 1 0 0 pl o f t h e i n c u b a t i o n m i x t u r e . e In all cases at least 80% o f b o t h i s o t o p e s w e r e b o u n d , w h e n t h e m i x t u r e s w e r e e x a m i n e d in t h e F a r r assay.
277 Those sera which had high titers in the Farr assay were more likely to form complement-fixing complexes with dsDNA (Table 2), but we feel we have not yet surveyed a sufficient number of sera to establish a trend. The inability of certain SLE sera to form complement-fixing complexes with dsDNA is apparently n o t due to the fact that they are intrinsically anti-complementary (presumably due to the presence of immune complexes). Even when very small amounts of DNA of high specific activity were used, m a n y of these sera still did n o t bind DNA in the RBC-CF assay (data not shown). Such sera could not inhibit the complement fixation reaction in sera which were positive in the assay (Table 1). However, in one case an SLE serum bound more dsDNA in the RBC-CF assay at a 2-fold dilution than when it was undiluted. This result may indicate that a sufficient number of immune complexes was present in this particular serum to partially inhibit the RBC-CF assay at high serum levels. Finally, when a number of SLE sera which were positive in the Farr assay but negative in the complem e n t assay were mixed, the mixtures remained negative in the complement assay, although they still bound DNA in the Farr assay (results not shown). Our results indicate that a significant fraction of those prepared complexes which do fix complement are rather stable to dissociation by excess cold DNA at 37°C (Table 3). Our double-label complement binding studies provide further evidence along these lines. It is clear that the a m o u n t of dsDNA which becomes part of a complex which actually fixes complement is critically dependent upon the order of mixing of antibody and dsDNA (Table 4). It should also be noted that in all cases both isotopes were almost completely bound in the Farr assay. DISCUSSION
We have developed a simple quantitative m e t h o d to measure accurately the a m o u n t of antigen (in this case DNA) which is part of an immune complex which actually fixes complement. The complement-fixing potential of the immune complex clearly depends upon the ratio of antibody to DNA. Under conditions of large DNA excess (i,e., just one or two antib o d y molecules per DNA; Aarden et al., 1976) the complexes do not fix complement, although DNA binding can easily be demonstrated in the Farr assay. For example, this is readily seen in Fig. 2B, at serum dilutions between 1/32 and 1/1024 where only Farr binding activity is detectable. In addition, Fig. 3 (the lower two curves) demonstrates that the complementfixing potential of a given serum dilution can be saturated. At a sufficiently high input of DNA the complexes which are formed are apparently so small that they no longer fix complement, although there is still binding in the Farr assay. Our results suggest that ca. 8--20 times as much serum is needed to achieve comparable DNA binding in the RBC-CF assay as in the Farr assay.
278 The relative sensitivity of the m e t h o d thus compares favorably with other studies in which the serum concentration necessary for binding of DNA is compared to the serum concentration needed for effective complement fixation. For example, Ballou and Kushner (1979) used the Cruthidia luciliae immunofluorescence test to measure anti-dsDNA antibodies and they also found that in general ca. 4--80 times as much serum was needed to achieve complement fixation (using fiuorescein-labeled antibodies to C-3) as opposed to just DNA binding by antibodies. Our results definitely implicate t h e high avidity anti-dsDNA antibodies (Taylor et al., 1979; Riley et al., 1980) in complement fixation. Our studies demonstrate t h a t once multiple binding of antibody to dsDNA occurs (a necessary condition for complement fixation), a significant fraction of the bound antibodies must be high avidity, because the complement-fixing potential of the complex remains high (Table 3), even after an excess of DNA is used in an a t t e m p t to dissociate the complex. This implies that the ability of an antibody/dsDNA complex to fix complement can be controlled quantitatively by the a m o u n t of DNA added as well as when it is added. For example, consider Fig. 3, the middle curve. If 16 ng of dsDNA are added to plasma mu, immune complexes which fix complement are formed, but if 80 ng are added all at once, no complement fixation is observed. One would therefore predict that if 80 ng were added sequentially to the plasma in an aliquot of 16 ng followed by 64 ng 1 h later, the immune complexes formed in the first hour would continue to fix complement, although the equilibrium property of the system would obviously suggest no complement fixation. We note that considerably less 'reversal' is seen in the Farr assay than in the RBC-CF assay (Tables 3 and 4). Under the conditions of the experiment, antibody was in considerable excess over DNA with respect to the Farr assay, and as just one antibody per DNA is apparently needed to 'register' in the Farr assay (Aarden et al., 1976), it is not surprising that little dissociation could be detected by this assay. The importance of the order of mixing of antibody and DNA is further seen in the double-label studies. The dsDNA which is added first fixes complement considerably more efficiently than the second DNA added. That is, the dsDNA which is added first binds most of the high avidity antibodies in large complexes which fix complement. Most of these antibodies are n o t 'available' when the second dsDNA is added, and thus this DNA fixes complement weakly, at best. However, there are sufficient antib o d y molecules available to at least bind the second DNA isotope and register in the Farr assay. It is likely that this is due to re-equilibration of lower avidity antibodies (Riley et al., 1980). To summarize, it seems clear that the potential of a given SLE serum to form antibody/dsDNA complexes which fix complement depends critically upon the antibody/dsDNA ratio in the complexes. This stoichiometry in turn is strongly influenced by the specific sequence of addition of DNA
279 to antibody because the complexes are rather stable and dissociate slowly. All other things equal, the complex sizes can clearly be manipulated such that for a given final concentration of dsDNA, the level of complement binding can be insignificant or very high. The relevance of this discussion in terms of lupus pathogenesis is as follows. It is believed that complement fixation by anti-dsDNA antibodies is a crucial factor in tissue destruction. Whether or not the antibodies which circulate in the serum of a given patient actually do fix complement will clearly depend upon the stoichiometry of the complexes they form with dsDNA. This stoichiometry in turn can be governed b y those processes that lead to release of dsDNA into the blood stream. It is therefore possible that if large quantities of dsDNA are rapidly released into the blood, small complexes which will n o t fix complement will be formed, and perhaps they will be cleared without tissue destruction. Obviously the situation may be considerably more complicated, b u t it does raise the possibility that patients (or more appropriately, NZB/W mice) (Lambert and Dixon, 1968) might be titrated therapeutically with appropriate amounts of dsDNA of a given molecular weight. Finally, we note that the technique we have described is considerably different quantitatively from most c o m m o n l y used protocols (Tung et al., 1978; Zubler and Lambert, 1978) which determine the amount of antibody (usually in equivalents of aggregated IgG) in an immune complex. It is well known that aggregated IgG is used simply as a standard in such experiments and that they convey no information with respect to immune complex size and composition. It is possible that, under a given set of experimental conditions, there could be quite a few IgG molecules b o u n d to DNA, but only two of these could actually have fixed complement. Thus, the quantitative meaning of a measurement which determines total IgG in an immune complex could be misleading. We recognize that the exact theoretical relationship between the density of antibody molecules on a given DNA and their complement-fixing potential remains to be established, b u t it is clear that the methods we have used along with procedures for the measurement of total human IgG in a system (Waller et al., 1979) have the potential of generating the data which is needed to examine this interesting question in detail. ACKNOWLEDGEMENTS It is a pleasure to thank Ms. Naomi W. Taylor for her useful experimental advice. S.E.P. thanks M 2. This work was supported by a grant from the W. Alton Jones Foundation, and b y NIH Grants AM 24083 and AM 11766.
280 REFERENCES Aarden, L.A., F. Lakmaker and R.R. DeGroot, 1976, J. Immunol. Methods 11, 153. Ballou, S.P. and I. Kushner, 1979, Clin. Exp. Immunol. 37, 58. Beaulieu, A., F.P. Quismorio, R.C. Kitridou and G.J. Friou, 1979, J. Rheum. 6,389. Koffler, D.,. 1974, Ann. Rev. Med. 25,149. Lambert, P.H. and F.J. Dixon, 1968, J. Exp. Med. 127, 507. Minter, M.H., B.D. Stollar and V. Agnello, 1979, Arthr. Rheum. 22, 959. Porter, R.R. and K.B.M. Reid, 1979, Adv. Prot. Chem. 33, 1. Rapp, H.J. and T. Borsos, 1970, Molecular Basis of Complement Action (Meredith Corporation, New York). Riley, R.L., D.J. Addis and R.P. Taylor, 1980, J. Immunol. 124, 1. Rothfield, N.F. and B.D. Stollar, 1967, J. Clin. Invest. 46, 1785. Sontheimer, R.D. and J.N. Gilliam, 1978, Clin. Immunol. Immunopathol. 10, 459. Stollar, B.D., 1975, Crit. Rev. Biochem. 3, 45. Tan, E.M., P.H. Schur, R.I. Carr and H.G. Kunkel, 1966, J. Clin. Invest. 45, 1732. Taylor, R.P., D. Weber, A.V. Broccoli and J.B. Winfield, 1979, J. Immunol. 122,115. Taylor, R.P., S.J. Waller, C. Haden and D.J. Addis, 1980, J. Immunol. 124, 2571. Tsuda, F., Y. Miyakawa and M. Mayumi, 1979, Immunology 37,681. Tung, K.S.K., A.J. Woodroffe, T.D. Ahlin, R.C. Williams and C.B. Wilson, 1978, J. Clin. Invest. 62, 61. Waller, S.J., R.P. Taylor and B.S. Andrews, 1979, J. Immunol. Methods 30,171. Waller, S.J., R.P. Taylor, E.L. Wright and M. Johns, 1980, submitted to Arthr. Rheum. Winfield, J.B., I. Faiferman and D. Koffler, 1977, J. Clin. Invest. 59, 90. Zubler, R.H. and P.H. Lambert, 1978, Prog. Allergy 24, 1.