In vivo anti-complementary activities of the cobra venom factors from Naja naja and Naja haje

Journal of Immunological Methods, 136 (1991) 287 - 294

287

© 1991 Elsevier Science Publishers B.V. 0022-1759/91/$03.50 ADONIS 002217599100085Q

JIM05853

In vivo anti-complementary activities of the cobra venom factors from Naja naja and Naja haje C a r m e n W. Van den Berg, Piet C. Aerts and H a n s Van Dijk Eijkman- Winkler Laboratory of Medical Microbiology, Department of Experimental Microbiology, Facuhv of Medicine University of Utrecht, Utrecht, The Netherlands (Received 20 September 1990, revised received 5 November 1990, accepted 12 November 19903

The kinetics of complement (C) depletion and recovery of C levels upon injection of B A L B / c mice with cobra venom factors (CVF), from N. naja (C3- and C5-depleting) and N. haje (selectively C3-depleting) were studied. The animals received i.p. or i.v. injections of either of the two preparations. CH50 and hemolytic C3 and C5 levels were followed as parameters of residual complement activity. N. naja CVF turned out to be as efficient in depleting total complement activity as N. haje CVF. Decreased CH50 values could largely be ascribed to C3 depletion. Complement consumption after N. naja CVF, however, lasted longer than after N. haje CVF administration. Estimated functional half-lives of N. naja and N. haje CVF were 11.5 and 4.5 h, respectively. Inhibition ELISAs showed that, after in vivo administration of either of the two CVF preparations, antigenic C3 and C5 kept circulating for days. Key words. Cobra venom factor; C3; C5

Introduction

Since the turn of the century, it has been known that cobra venoms possess strong anti-complementary activity (Ewing, 1894; Flexner and Noguchi, 1903). Cobra venoms contain at least three complement-reactive factors with cobra venom factor (CVF) as the most commonly known (Ballow and Cochrane, 1969; Von Zabern et al., 1981). CVF has been used as a very efficient and selective tool for in vivo complement (C) depletion. It shares structural and functional properties

Correspondence to." C.W. Van den Berg, Eijkman-Winkler Laboratory of Medical Microbiology, University Hospital HP G04.614, P.O. Box 85500, NL-3508 GA Utrecht, The Netherlands. Abbreviations: AP, alternative pathway; BSA, bovine serum albumin; CVF, cobra venom factor; C, complement; GAM, goat anti-mouse; GAR, goat anti-rabbit; TMB, tetramethylbenzidine.

with complement component C3. Antigenic cross-reactivities between C3 and CVF have been reported (Alper and Balavitch, 1976; Minta and Man, 1980; Eggertsen et al., 1983; Vogel et al., 1984; Grier et al., 1987). CVF has also C3b-like activity in forming a Mg2+-dependent complex with factor B which is subsequently activated by factor D to the CVF-dependent C3 convertase, CVF,Bb (MSller-Eberhard, 1967; GStze and MiJller-Eberhard, 1971; Cooper, 1973; Hunsiker et al., 1973; Hensley et al., 1986). This CVF-dependent C3 convertase is extremely stable when compared to the alternative pathway-dependent C3 convertase, C3b, Bb (t,2 in vitro with purified C components at 37°C: 7 h versus 1.5 min) (Vogel and Miiller-Eberhard, 1982; Medicus et al., 1976), explaining its C3-depleting effect. The differences between the in vitro half-lives are explained by the resistance of the CVF,Bb complex to inactivation by factors H and I (Lachmann and Halbwachs, 1975; Nagaki et al., 1978). Apart from the stable

0022-1759/91/$03.50 ~:'~1991 Elsevier Science Publishers B.V. (Biomedical Division)

288 C3 convertase activity, the CVF-dependent C3 convertase from N. naja has also C5-activating capacity, whereas the AP-dependent C3 convertase and the CVF-dependent C3 convertase from N. haje CVF do not (Miyama et al., 1975; Daha et al., 1976; Bauman, 1978; Von Zabern et al., 1980). In vitro activities of N. naja and N. haje CVF vary with the source of complement they are tested with (Von Zabern, 1980; Van den Berg et al., 1990). The CVF species are equally effective in inactivating total C and C3 activity in mouse serum, but differ greatly with respect to their in vitro C5-depleting effect (Van den Berg et al., 1990). N. naja CVF,Bb complexes activate C5 in addition to C3, whereas N. haje CVF,Bb complexes do not. This means that consumption of mouse C by N. haje CVF fully depends on C3 depletion. However, also exhaustion of mouse C by normal doses of N. naja CVF is largely based on C3 inactivation. Although both N. naja and N. haje CVF have frequently been used for in vivo C depletion in mice as well as in other animal species, the in vivo activities of the preparations have never been compared. Half-lives of 125I-labeled N. naja CVF has been estimated in rabbits (t,2 = 32 h) (Cochrane et al., 1970) and mice (t,2= 24 h) (Pepys, 1975). Functional half-lives in vivo have not been published so far. The elimination of CVF in vivo is accelerated by the generation of neutralizing antibodies to the highly immunogenic CVF from about 5 days after CVF injection (Pepys, 1975). The inactivation and recovery of mouse total complement activity and functional C3 and C5 after in vivo N. naja and N. haje CVF application is the subject of this paper. In order to explain the differential C-consuming effects of the two CVF preparations, the half-lives of the anti-complementary components were estimated by a functional assay.

Materials and methods

Cobra venom factors Cobra factors from the Egyptian cobra N. haje (V 4254, Sigma, St. Louis, MO) and the common

Indian cobra N. naja (no. V 4378, also from Sigma) were prepared by Mono Q anion exchange chromatography as described by Beukelman et al. (1987) for N. naja. CVF activities were assayed as described by Van den Berg et al. (1990), using undiluted mouse serum as target for CVF action. Protein contents of the CVF preparations were determined by the Bradford method using bovine serum albumin as a reference (Bradford, 1976).

Mice Male B A L B / c mice, obtained from Iffa Credo (l'Arbresle, France), were used at an age of 10-15 weeks (about 20 g). At different times after CVF or saline injection, the mice were bled by orbital puncture. After clotting of the blood at 20°C for 1.5 h, serum was separated by centrifugation and subsequently stored at - 7 0 ° C until use. C VF injections CVF was injected intraperitoneally (i.p.) in a volume of 0.5 ml of saline or intravenously (i.v.) in a volume of 0.2 ml of saline once or twice with intervals of 8 h. Buffers Veronal-buffered (25 raM) saline (750 raM), p H 7.35 _+ 0.05 (VSB, 5 × ; Van Dijk et al., 1980; 1985), served as a five times concentrated stock solution for the preparation of VSB 2+ (containing 0.15 mM CaC12 and 0.5 mM MgC12), EGTA-VB (containing 2.5 mM MgC12 and 8 m M EGTA), and EDTA-VB (containing 10 mM EDTA). EDTA-Tween 20 (2 mM EDTA, 150 mM NaCI, and 0.03% Tween 20, p H 7.35) was used as ELISA washing buffer. Phosphate-buffered saline (PBS: 20 mM sodium phosphate, 150 m M NaC1, p H 7.4) served as diluent for serum samples. Erythrocytes Rabbit blood ( 1 / 2 diluted in Alsever's old solution) was obtained from bioTrading (Wilnis, the Netherlands) and used as a source of erythrocytes (RaE). Before use, the erythrocytes were washed three times with isotonic N a I to elute possibly adsorbed serum proteins. Antisera Anti-C3 antiserum was obtained by immunizing a rabbit with purified mouse C3 (Van den

289 Berg et al., 1989), and anti-C5 antiserum was obtained by immunizing C5-deficient D B A / 2 mice with C5-sufficient B A L B / c serum plus FCA. In double immunodiffusion and immunoblotting antisera reacted monospecifically with C3 and C5 in mouse serum. CVF preparations did not interfere with the antisera in inhibition ELISAs.

Complement assays CH50, C3, and C5 activities were determined by colorimetric microtiter assays (Klerx et al., 1983; Van den Berg et al., 1989). In the C3 assay methylamine-treated (C3-/C4-depleted) mouse serum was used as C3 reagent (Jessen et al., 1983). C5-deficient D B A / 2 serum was used as reagent in the C5 assay. Functional C parameters in CVFtreated animals were expressed as relative activities (percentages) of those in saline-treated animals. CVF activities were measured by a microtiter assay as described (Van den Berg et al., 1990). In brief: 25/~1 of CVF dilutions were incubated with 110 ttl of undiluted mouse serum at 30°C for 30 min. Residual C activities were estimated by titration of the CVF mouse serum mixture. Activities were defined as the amount of CVF giving a 50% reduction in activity of mouse serum incubated with saline. Inhibition ELISAs Antigenic C3 and C5 after in vivo CVF treatment were determined by specific inhibition ELISAs: soft polyvinyl chloride microtiter plates (Titertek activated, Flow Laboratories, Zwanenburg, The Netherlands) were coated with 100/~1 of mouse C3 or C5 (partially purified as described by Van den Berg et al. (1989) by 4-10% PEG precipitation, Mono Q anion exchange and protein A affinity chromatography) in PBS for 2 h at 37°C. Free places in the wells were blocked with 4% BSA in PBS. Serum samples were diluted with PBS (50 /~1) in the plates and a constant amount of anti-C3 or anti-C5 (50 ~1) was added. The plates were subsequently incubated at 20°C for 1 h and were washed with EDTA-Tween 20. 100 /~1 of G A R / I g G - or GAM/IgG-peroxidase-labeled conjugate (Nordic, Tilburg, The Netherlands), 1/10,000 diluted in PBS/0.05% Tween 20, was added to each well. After 1 h incubation at room

temperature, the plates were washed with EDTATween 20 and developed with the T M B / H 2 0 2 substrate (0.1 M N a A c / c i t r a t e buffer pH 5.5 containing 0.0025% H202 and 1.6% of a solution of tetramethylbenzidine (TMB; Sigma, St.Louis, U.S.A.) in dimethyl sulfoxide (6 mg/ml)). ELISA titers were read at dilutions of the sera giving rise to 50% inhibition of absorbance (450 nm) relative to the absorbance of the antiserum only. Antigenic serum C3 and C5 levels after CVF administration were expressed as percentages of corresponding levels in saline-treated animals.

Statistics Figures represent the arithmetic mean (M) of n independent determinations ___the standard error of the mean (SEM).

Results

According to the schedule described by Beukelman et al. (1987), mice received two i.p. injections of CVF of either N. naja or N. haje (0.8/tg/injection) with an interval of 8 h. 1-5 days later, the mice were bled and serum CH50 values as well as hemolytic and antigenic C3 and C5 were determined (Fig. 1). Depletion by N. naja CVF as determined by CH50 lasted longer than that induced by the equivalent amount of N. haje CVF (Fig. 1A). Measurement of C3 inactivation by N. naja CVF during the first 2 days was complicated by bystander activation of C5 in the C3 reagent of CVF,Bb complexes present in the sera tested (Fig. 1B) (Van den Berg et al., 1990). Bystander activation was confirmed by the hemolysis observed under similar conditions in EDTA-containing buffer. Antigenic C3, however, showed already a drop on the first day (Fig. 1D). Hemolytic C5 turned earlier back to normal than hemolytic C3 and CH50 values. CVF from N. haje, which does not activate mouse C5 in vitro (Van den Berg et al., 1990), gave rise to an apparent inactivation of C5 in vivo o n the first day (Fig. 1C). In the inhibition ELISA (Fig. 1E), however, no depletion of C5 was observed. The early kinetics of C inactivation were studied after a single injection of a lower dose of CVF. Mice received an i.p. or an i.v. injection of a lower

290 "50

A

'se

a

~125 >

25

125

UlO 0 .

00

i j,oo ©

o

75

~ 75

50

~ 50I 25-

25 ~

C

O 0

1

2 Time

3

4

3

5

!

2

(days)

3

5

4

,a

1

(days)

Time

2

3

"lme

D

150 ~25

4

5

(days)

150

E

125

© o 75

~ 75.

g ~ 50" 25

25

J

,72 :,

1

2

T,me

3

4

C

1

(days)

2

3

4

J ,me ( d a y s )

Fig. 1. Mice (n - 1 0 per group) were twice injected i.p with 0.8 p.g of N. naja (cLosed symbols) or N. htye (open symbols) CVF with an interval of 8 h. 1-5 days after CVF administration, the mice were bled and blood of two mice was pooled. After clotting and serum preparation, total complement (A) and functional as well as antigenic C3 (B and D respectively) and C5 (C and E) were determined. Functional C3 after N. naja CVF administration was estimated in the presence (circles) and absence (triangles) of Mg 2+. Values are expressed as percentages ( _+ SEM) of corresponding levels in saline-treated animals.

dose (0.3/~g) of N. naja or N. haje CVF. i.v. CVF injection resulted in a very rapid C depletion in serum, whereas i.p. application did not. As in the former experiment, C depletion by N. naja CVF

A

153

lasted longer than by N. haje CVF (Figs. 2A-C). At this lower CVF concentration no significant bystander lysis was observed during the first hours (Fig. 2B). C5 levels increased from 6 to 12 h after

,5c

,~

5:

125

C

125

-- tO0

-- IO0

E ~ 75

i 75

~ 50-

5C

50

25-

25

25

"5 C6

12

24 Time

(hours)

48

a

6

12

24

Tkme ( h o u r s )

46

5

12

24

Time

48

(hours)

Fig. 2. Mic e (n = 1 0 per group) were injected i.p. (circles) or i.v. (squares) with 0.3/tg of N. naja (closed symbols) or N. haje (open symbols) CVF. After various time intervals, the mice were bled and blood of two mice was pooled. After clotting and serum preparation, total complement ( A ) and functional C3 ( B ) and C5 ( C ) levels were measured. Values are exl~essed as percentages (_+_ SEM) of corresponding levels in saline-treated animals.

291 150

150

150

B

125

~125

~

~100

125

I00 o

o_ 75

~ ~°

~ ~0

o

I

I 25

~

0

25

0

o

0.3 1 3 Dose of CVF (pg/mouse)

0

0.3 1 Dose of CVF (~g/mouse}

3

Dose of CVF (~g/mouse)

Fig. 3. Mice (n = 10 per group) were i.p. (circles) or i.v. (squares) injected. Graded doses of N. naja (closed symbols) or N. haje (open symbols) CVF were given. After 3 days, mice were bled and blood of two mice was pooled. After clotting and serum preparation, total complement (A) as well as functional C3 (B), and C5 (C) levels were estimated. Values are expressed as percentages (+ SEM) of corresponding levels in saline-treated animals.

N. naja C V F injection (Fig. 2 C ) . N o (i.p.) or o n l y a slight, t e m p o r a r y (i.v.) effect of N. haje C V F on h e m o l y t i c C5 was observed. T h e effect of the C V F dose on c o m p l e m e n t levels 3 d a y s after a single or d o u b l e i.p. o r i.v. injection was studied. A s shown in Fig. 3, some t e n - f o l d m o r e N. haje C V F t h a n N. naja C V F was n e e d e d to keep mice d e p l e t e d for 3 d a y s of total a n d h e m o l y t i c C3 (Figs. 3A a n d 3B). As also shown in this Fig. (3C), C5 i n a c t i v a t i o n on d a y 3 r e q u i r e d at least a ten-fold higher dose than imp l i c a t e d for effective C3 c o n s u m p t i o n . N o great differences b e t w e e n the effects o f i.p. a n d i.v. (Fig. 3) a n d single or d o u b l e i.p. C V F injection (not shown) were observed. T h e differences in in vivo C i n a c t i v a t i o n b y the C V F p r e p a r a t i o n s o b s e r v e d

"~ 100 -

~

=

75

11.5 h

.,~ 50

0

tl/z

i

,

,

4.5 h

=

i

3

,

,

i

6

,

,

i

,

9

,

1

12

,

,

i

15

Time (hours) Fig. 4. Mice (n = 5 per group) were injected i.v. with 1 #g of N. naja (closed squares) or N. haje (open squares) CVF. After

different times mice were bled and serum CVF activities were determined. The broken line indicates the detection limit of the assay.

so far were likely to be due to distinct e l i m i n a t i o n patterns. T o study C V F e l i m i n a t i o n , mice received single i.v. injections of C V F and, after different time intervals, the a n i m a l s were b l e d for serum. R e s i d u a l C V F in the sera was q u a n t i t a t e d b y a d d i n g an excess of fresh m o u s e s e r u m a n d det e r m i n i n g C H 5 0 values after i n c u b a t i o n at 30°C for 30 min (Fig. 4). T h e f u n c t i o n a l half-lives of N. naja a n d N. haje C V F were a p p r o x i m a t e l y 11.5 h a n d 4.5 h, respectively.

Discussion This p a p e r deals with a c o m p a r a t i v e s t u d y of the in vivo effects of N. naja a n d N. haje C V F in mice. A l t h o u g h the p r e p a r a t i o n s have f r e q u e n t l y been used to investigate the effects of C d e p l e t i o n in vivo, no d a t a on their relative efficiencies in vivo are available. It was o b s e r v e d that the C V F p r e p a r a t i o n s differ greatly with respect to their C - d e p l e t i n g activities. I n j e c t i o n of mice with equivalent a m o u n t s of N. naja a n d N. haje C V F resulted in longer total C a n d C3 d e p l e t i o n for the former than when N. haje C V F was used (Figs. 1 a n d 2). To o b t a i n similar effects on d a y 3 after injection a b o u t ten-fold higher doses of N. haje C V F than of N. naja C V F w e r e r e q u i r e d (Fig. 3). These results are at v a r i a n c e with results r e c e n t l y o b t a i n e d by us with regard to the in vitro inactivation of m o u s e C b y the s a m e C V F p r e p a r a t i o n s (Van den Berg et al., 1990). Differences in e l i m i n a t i o n rate o f the C V F pre-

292

parations could largely account for the discrepancy between in vitro and in vivo results. By monitoring the disappearance of 1251-labelled CVF from the circulation, Cochrane et al. (1970) and Pepys (1975) estimated the half-lives of N. naja CVFinrabbits(t~= 3 2 h ) and m i c e ( t . = 24h), respectively. Those data, however, did not concern functional half-lives. A functional half-life of 7 h at 37°C was found by Vogel and Miiller-Eberhard (1982) for the N. naja derived CVF,Bb complex in vitro using purified human complement components. In our study functional half-lives of the CVF preparations in vivo were estimated by preincubating small volumes of serum from CVF-injected mice, obtained at different times after CVF application, with an excess of fresh mouse serum as target of complement inactivation. After incubation, residual complement activity was estimated by dilution of the mixture (Fig. 4). The shorter half-life of N. haje CVF observed in this way may be explained by a faster elimination or a functional inactivation in the circulation of N. haje CVF as such or as CVF,Bb complex. Since it has been reported that N. haje CVF is about ten times more susceptible to proteolysis by trypsin than N. naja CVF (Von Zabern et al., 1982), a difference in enzymatic inactivation of the CVF preparations may be a likely explanation for the difference in functional elimination. A higher immunogenicity of N. haje CVF as explanation of its shorter half-life is unlikely, since it takes days before significant antibody formation will occur (Pepys, 1975). In vivo inactivation of C3 and C5 by the CVF preparations was estimated by functional as well as immunochemical assays. The assay of functional C3 in sera of mice prior injected with a high dose of N. naja CVF was complicated by CVF, Bb-induced bystander lysis. This was very similar to the effect of N. naja CVF on mouse serum in vitro (Van den Berg et al., 1990). The fact that antigenic C3 was not completely reduced to zero by CVF application in vivo can be explained by the recognition of CVF,Bb-induced C3 split products by the anti-C3 antiserum used in the ELISA. Moreover, C3 may be constantly replenished from extravascular stores and de novo synthesis. Thus, as long as CVF keeps circulating,

newly supplied C3 will constantly be inactivated. CVF is known to be cross-reactive with anti-C3 antibodies (Alper and Balavitch, 1976; Minta and Man, 1980; Eggertsen et al., 1983; Vogel et al., 1984; Grier et al., 1987), but CVF concentrations used in this study did not interfere with the C3EL1SA. In vivo C5 inactivation by N. naja and N. haje CVF was also followed by hemolytic and antigenic (inhibition ELISA) assays. At higher doses (Fig. 3C) and shortly after injection (Fig. 1C) N. haje CVF apparently inactivated C5. This observation is in agreement with in vitro results, but must be ascribed to a technical artifact (Van den Berg et al.; 1990): In the functional C5 assay, CVF,Bb complexes inactivate the C3 of the C5 reagent which may falsely be interpreted as C5 depletion. The results on antigenic C5, however, sustain that no C5 depletion takes place (Fig. 1E). Although the effects of the CVF preparations in vivo and in vitro differed quantitatively, they were comparable in qualitative sense: both in vivo and in vitro N. haje CVF failed to inactivate C5. Total C depletion by the CVF preparations was in both situations based on C3 depletion, while only high doses of N. naja CVF give rise to C5 inactivation. With respect to the doses of CVF preparations used in this study, it may be concluded that these were in the low range. In earlier times, doses of 10 25/~g of N. naja CVF per mouse divided over four injections were used (Cochrane et al., 1970: Pepys, 1975), whereas in recent reports one or two injections of 1 4 / ~ g / m o u s e were applied (Beukelman et al., 1987; Vignali et al., 1988). As shown in Fig. IA, a total dose of 1.6 /~g of N. naja CVF is sufficient for 5 days complement depletion. Higher doses, however, are probably not more efficient, since within 1 week CVF-neutralizing antibodies appear in the serum (unpublished). Detailed information on N. haje CVF was so far not available. In conclusion, our results clearly show that CVF from N. naja is far more efficient in depleting total C activity in mice than N. haje CVF. This difference appears to be mainly based on a faster elimination of N. haje CVF. N. haje CVF, however, remains the agent of choice to bring about a short-term, selective depletion of C3.

293

Acknowledgement W e t h a n k Prof. Dr. J . M . N . Willcrs for valuable discussion and comments on the manuscript.

References Alper, C.A. and Balavitch, D. (1976) Cobra venom factor: evidence for its being altered cobra C3 (the third component of complement). Science 191, 1275. Ballow, M. and Cochrane, C.G. (1969) Two anti-complementary factors in cobra venom: hemolysis of guinea pig erythrocyte by one of them. J. Immunol. 103, 944. Bauman, N. (1978) Lack of C5 convertase-generating activity in Naja haje cobra factor. J. Immunol. 120, 120. Beukelman, C.J., Aerts, P.C., Van Dijk, H. and Willers, J,M.N. (1987) A one step isolation procedure for phospholipase A/-free cobra venom factor by fast protein liquid chromatography. J. Immunol. Methods 92, 119. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248. Cochrane, C.G., Mi~ller-Eberhard, H.J. and Aikin, B.S. (1970) Depletion of plasma complement in vivo by a protein of cobra venom: its effect on various immunologic reactions. J. Immunol. 105, 55. Cooper, N.R. (1973) Formation and function of a complex of the C3 proactivator with a protein from cobra venom. J. Exp. Med. 137, 451. Daha, M.R., Fearon, D.T. and Austen, K.F. (1976) C3 requirement for formation of alternative pathway C5 convertase. J. Immunol. 117, 630. Eggertsen, G., Lundwall, A., Hellman, U. and Sj6quist. (1983) Antigenic relationship between human and cobra complement factors C3 and cobra factor (CVF) from the indian cobra (Naja naja). J. Immunol. 131, 1920. Ewing, C.B. (1894) The action of rattle snake venom upon the bacterial power of blood serum. Lancet 72, 1236. Flexner, S. and Noguchi, H. (1903) Snake venom in relation to haemolysis, bacteriolysis, and toxicity. J. Exp. Med. 6, 277. G~Stze, O. and Mi~ller-Eberhard, H.J. (1971) The C3-activator system: an alternative pathway of complement activation. J. Exp. Med. 134, 90s. Grier, A.H., Schultz, M. and Vogel, C.-W. (1987) Cobra venom factor and human C3 share carbohydrate antigenic determinants. 3. Immunol. 139, 1245. Hensley, P. O'Keefe, M.C., Spangler, C.J., Osborne, J.C. and Vogel, C.-W. (1986) The effect of metal ions and temperature on the interaction of cobra venom factor and human complement factor B. J. Biol. Chem. 261, 11038. Hunsiker, L.G., Ruddy, S. and Austen, K.F. (1973) Alternate complement pathway: factors involved in cobra venom factor (CoVF) activation of the third component of complement (C3). J. Immunol. 110, 128.

Jessen, T.E., Barkholt, V. and Welinder, K.G. (1983) A simple alternative pathway for hemolytic assay of human complement component C3 using methylamine-treated plasma. J. Immunol. Methods 60, 89. Klerx, J.P.A.M., Beukelman, C.J., Van Dijk, H. and Willers, J.M.N. (1983) Microassay for colorimetric estimation of complement activity in guinea pig, human and mouse serum. J. Immunol. Methods 63, 215. Lachmann, P.J. and Halbwachs, L. (1975) The influence of C3b inactivator (KAF) concentration on the ability of serum to support complement activation. Clin. Exp. [mmunol. 21, 109. Medicus, R.G., G~3tze, O. and Miiller-Eberhard, H.J. (1976) Alternative pathway of complement: recruitment of precursor properdin by the labile C3/C5 convertase and the potentiation of the pathway. J. Exp. Med. 144, 1076. Minta, J.O. and Man, D. (1980) Immunological, structural and functional relationships between an anti-complementary protein from Crotalus atrox venom, cobra venom factor and human C3. Immunology 39, 503. Miyama, A., Kato, T., Horai, S., Yokoo, J. and Kashiba, S. (1975) Trypsin-activated complex of human factor B with cobra venom factor (CVF), cleaving C3 and C5 and generating a lytic factor for unsensitized guinea pig erythrocytes. I. Generation of the activated complex. Biken's J. 18, 193. Miiller-Eberhard, H.J. (1967) Mechanism of inactivation of the third component of human complement (C'3) by cobra venom. Fed. Proc. 26, 744. NagakL K., lida, M. and Inai, S. (1978) Reaction mechanisms of /~-IH globulin. Int. Arch. Allergy Appl. Immunol. 57, 221. Pepys, M.B. (1975) Studies in vivo of cobra factor and murine C3. Immunology 28, 362. Van den Berg, C.W., Van Dijk, H. and Capel, J.P.A. (1989) Rapid isolation and characterization of native mouse complement components C3 and C5. J. ImmunoL Methods 122, 73. Van den Berg, C.W., Aerts, P.C. and Van Dijk, H. (1990) Functional assay of C5-activating and nonactivating cobra venom factor preparations in the mouse system. J. Immunol. Methods 133, 199. Van Dijk, H., Rademaker, P.M. and Willers, J.M.N. (1980) Estimation of classical pathway of mouse complement activity by use of sensitized rabbit erythrocytes. J. lmmunol. Methods 39, 257. Van Dijk, H., Rademaker, P.M., Klerx, J.P.A.M. and Willers, J.M.N. (1985) Study of the optimal reaction conditions for assay of the mouse alternative complement pathway. J. Immun01. Methods 85, 233. Vignali, D.A.A., Bickle, Q.D., Taylor, M.G., Tennent, G. and Pepys, M.P. (1988) Comparison of the role of complement in immunity to Schistosoma mansoni in rats and mice. Immunology 63, 55. Vogel, C.-W. and Mi~ller-Eberhard, H.J. (1982) The cobra venom factor-dependent C3 convertase of human complement. A kinetic and thermodynamic analysis of a protease acting on its natural high molecular weight substrate. J. Biol. Chem. 257, 8292.

294 Vogel, C.-W., Smith, C.A. and Miiller-Eberhard, H.J. (1984) Cobra venom factor: structural homology with the third component of h u m a n complement. J.Immunol. 133, 3235. Von Zabern, I., Hinsch, B., Przyklenk, H., Schmidt, G. and Vogt, W. (1980) Comparison of Naja n. naja and Naja h. haje cobra-venom factors: correlation between binding affinity for the fifth component of complement and mediation of its cleavage. Immunobiology 157, 499.

Von Zabern, I., Przyklenk, H., Damereau, B. and Zimmermann, B. (1981) Isolation and properties of a complement inhibitor from Naja haje venom, distinct from known anti-complementary factors in cobra venom. Scand. J. Immunol. 14, 109. Von Zabern, I. Przyklenk, H. and Vogt, W. (1982) Chain structure of Cobra venom factor from Naja naja and Naja haje venom. Scand. J. Immunol. 15, 357.