C3 Convertase of the Alternative Complement ...

THEJOURNALOF BIOLOGICAL CHEMISTRY Vnl 258 No. 12 Issue of June 25, pp. 7411-7415,1983

printed L u . s . ~ .

C3 Convertase of the Alternative Complement Pathway DEMONSTRATION OF AN ACTIVE, STABLE C3b,Bb(Ni) COMPLEX* (Received for publication, January 31,1983)

Zvi Fishelson$,Michael K. Pangburna,andHans J. Muller-EberhardlI From the Department of Immumbgy, Research Institute of Scripps Clinic, La JoUa, California 92037

was then added, and the components were gel-filtered through a Sephadex G-200 column (2.5 X 90 cm) (Pharmacia Fine Chemicals, Piscataway, NJ). Trace amounts of 1261-labeledC3 and '311-labeled C3b,Bb is the C3 convertase (EC 3.4.21.47) of the alterna- Factor B were included in the reaction mixture to allow detection of tive pathway of complement.' The enzyme is responsible for C3b and Bb. The proteins were pure 88 judged by SDS-PAGE (Fig. 1). Bovine thyroglobulin was purchased from Sigma. Proteins were amplification of pathway activation and for deposition on radiolabeled with '%I or I3'I (Amersham Corp.) by the Iodogen techtarget cells of C3b and the membrane attack complex (1-6). nique (Pierce Chemical Co.) (23). '961-labeledClq was kindly provided The enzyme is controlled by the serum proteins Factor H, by Dr. Andrea Tenner of this department. The labeled proteins had Factor I (EC 3.4.21.45), and properdin (1, 7-14). The forma- a specific activity of 0.7-1.8 mCi/mg. RadioactivityAssay of a7Ni-a7Ni2+ was obtained from New England tion of the enzyme requires C3b, Factor B, Factor D (EC 3.4.21.46), and Mg2+ (2-6, 15). After formation of the revers- Nuclear at a concentration of 0.28 M (11.277 mCi/mg) in 0.5 M HCl solution. A 0.01 M stock solution was made in distilled water. Addition ible, bimolecular complex C3b,B(Mg),Factor D cleaves Factor of B3Nito reaction mixtures did not lower the pH below 7.0. "Ni B, releases the activation fragment Ba, and generates the radioactivity (@-emission)was measured using polypropylene liquid active enzymeC3b,Bb,which has a calculated molecular scintillation vials (Kimble, Toledo, OH), ACS-I1counting scintillant weight of 239,000 (C3b, M,= 176,000; Bb, M,= 63,000). The (Amersham Corp.), and a Beckman LS8OOO liquid scintillation counenzyme is a serine protease (16-18) whose catalytic site resides ter. Since '%I and '''I also activate the scintillant, samples containing in the Bb subunit. The enzyme is inherently labile, and the =Ni and radioiodinated proteins were fvst analyzed for ?-radiation spontaneous dissociation of its subunitsresults in irreversible 90 that appropriate corrections for iodine contribution could be made. Electrophoresis-Slab gel electrophoresis in presence of SDS was loss of enzymatic activity. Due to this lability, previous at- performed using the Canalco PAGE system from MiIes Laboratories, Inc., Elkhart, IN. The 5-12% gradient polyacrylamide gel contained * This work was supported by United States Public Health Service 0.01% SDS. Grants AI 17354 and HL 16411. This is Publication 2900 from the Sucrose Density Gradient Ultracentri,bgatiun-Ultracentrifugation Research Institute of Scripps Clinic. The costs of publication of this was performed in 5-20% sucrose (Fisher) density gradients in VBS article were defrayed in part by the payment of page charges. This containing 6 mM EDTA. Five-ml gradients were formed in Ultraarticle must therefore be hereby marked "advertisement" in accord- Clear tubes (% X 2 inch) (Beckman) using a Buchler Auto-Densiance with 18 U.S.C. Section 1734 solely to indicate this fact. Flow I1 C. The tubes were subjected to centrifugation in an SW 50.1 $ Supported by a Chaim Weizmann postdoctoral fellowship. 3 Recipient of American Heart Association Established Investigatorship 81-225. The abbreviations used are: VBS, veronal (5 mM)-buffered sallCecil H. and Ida M. Green Investigator in Medical Research, line (0.15 M), pH 7.2, containing 0.02% (w/v) sodium azide; SDSResearch Institute of Scripps Clinic. PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; * For nomenclature of the alternative pathway of complement, see CBb,Bb(Mg)and CBb,Bb(Ni),the C3 convertase formed with Mg and (1981) J. Zmmunol. 127,1261. Ni, respectively.

7411

Downloaded from www.jbc.org by guest, on April 15, 2012

The purposes ofthis study were to demonstrate the tempts to directly demonstrate the enzyme by ultracentrifuC3 convertase complex, C3b,Bb(EC 3.4.21.47), of the gation, electrophoresis, or molecularsieve chromatography alternative pathway of complement by ultracentrifu- have failed. Since the spontaneous decay dissociation of gation and to determine whether the metal ion required C3b,Bb is not accelerated by EDTA, it was uncertain whether for enzyme formation is present in the active enzyme the metal ion required for enzyme formation remained assocomplex. It has been shown previously that Clb,Bb ciated with the active enzyme. We have recently found that formed with Nia+rather than M e + exhibits enhanced Ni2+can replace Mg in C3b,Bb formation and thatthe enzyme stability. Using sucrose density gradient ultracentri- formed with Ni (CSb,Bb(Ni)) is 6 to 10 times more stable fugation, an enzymatically active C3b,Bb(Ni) complex than that formed with Mg (C3b,Bb(Mg)) (19). Utilizing Ni could be demonstrated which has a sedimentationcoef- and itsradioisotope 63Ni,it was possible in the present study ficient of 10.7 S and which is stable in 10 mM EDTA. to demonstrate the active C3b,Bb complex by ultracentrifuUpon formation of the enzyme with the radioisotope gation and to detect the metal ion in the complex. esNia+,the ultracentrifugal distribution of the metal correlated with that of the enzyme complex. The molar MATERIALS ANDMETHODS ratio of Ni to C3b,Bb was 1:l. Displacement of Ni by Buffers and Reagents-The following buffers wereused: VBS? Mg during formation of the enzyme indicated that both VBS containing 0.1%gelatin (w/v), and VBS containing 0.1% gelatin metals may bind to the samesite in the enzyme. Binding (w/v) and 20 mM EDTA (Sigma). MgC12 and NiClz of analytical of "Ni to the catalytic site bearing fragment Bb was grade were purchased from Fisher. significantly stronger thanits binding to C3b or to the Preparation and Radiolabeling of Purified Complement Proteinszymogen, Factor B. It is proposed that there is one C3 (20), Factor B (21), and Factor D (22) were isolated from human metal-binding site in the Clb,Bb enzyme which is not serum as previously described. C3b and Bb were generated by incususceptible to chelation byEDTA and which is located bating 10 mg of C3, 1 mg of Factor B, and 50 pg of Factor D in 2 ml of VBS containing 4 mM Mg for 60 min a t 37 "C. EDTA (7.5 mM) in the Bb subunit.

C3 Convertase of the Alternative Complement Pathway

7412

it was not completely separated from free C3b. These results indicate that the fast sedimenting peak of radioactivity and enzyme activity represents the C3b,Bb(Ni) complex. To determine the sedimentation coefficient of C3b,Bb, its velocity was compared with that of Factor B (5.9 S ) , C3 (9.5 S ) , Clq 200k (11S ) , and thyroglobulin (12and 19 S ) . Relative to these four proteins, C3b,Bb(Ni) had an apparent sedimentation coefficient of 10.7 S (Fig. 3). Demonstration of -Ni in the Enzyme Complex-The effi92k ciency of =Ni (atomic weight, 58.7)and itsradioactive isotope =Ni in forming C3b,Bb was compared and found identical. When the enzyme was formed with C3b, Factors B and D, and 63Niand thensubjected to sucrose density gradient ultra45k centrifugation, the large excess of free 63Niinterfered with the detection of the protein-bound -Ni. Therefore, free -Ni 21kwas removed before ultracentrifugation by filtration of the FIG. 1. Demonstration by SDS-polyacrylamide eleetropho- reaction mixture through aBio-Gel P-6 column.Residual resia of the proteinswed for C3 convertase formation. Elec- proenzyme complex,CSb,B(Ni),which had not been activated trophoresis of 7 pg of C3b, Factor B, or Bb or 10 pg of Factor D was by Factor D was dissociated by addition of 10 mM EDTA to performed in 5-12% gradient gels (see "Materials and Methods"). the reaction mixture. As shown in Fig. 4, the distribution of

C3b B

Bb D

-

-

-

200k represents M, = 200,000, for example.

I

RESULTS

Demonstration of C3b,Bb(Ni) by Sucrose Density Gradient Ultracentrifugation-The C3b,Bb enzyme was formed with purified C3b, Factor B, and Factor D. An SDS-PAGEanalysis of the proteins employed in this study is presented in Fig. 1. To facilitate detection of the C3b,Bb complex, 1261-labeled Factor B was included in the reaction mixture whichwas incubated for 5 min at 24 "C prior to ultracentrifugation. When the enzyme was formed with Mg, only a slowly sedimenting peak of radioactivity was observed, indicating complete decay dissociation of C3b,Bb(Mg)during sedimentation. However, when the enzyme was formed with Ni (Fig. 2), an additional faster sedimenting peak was observed which contained C3 convertase activity. Fifteen-microliter aliquots of the ultracentrifugal fractions were mixed with 5 pl of 1.4 mg/ ml of C3 and incubated for 30 min at 37"C, after which residual C3 hemolytic activity was determined. The distribution of C3 convertase activity correlated closely with that of the faster sedimenting peak of radioactivity (Fig. 2). When differentially radiolabeled C3b and Factor B were employed, C3b wasalso detected in the faster sedimentingpeak, although

I

A

12004

1

-80 -60

' -L .-0 . d

CT

-40

52 3 c3

-20 d

20

10

30

Fraction Number FIG. 2. Demonstrationof the CSb,Bb(Ni) enzyme sby ucm density gradient ultracentrifugation. 81 pg of C3b, 38 pg of '=Ilabeled Factor B, and 3 pg of Factor D were incubated in 0.14 ml of VBS containing 0.1 m M NiClz for 5 min at 24 "C. 10 mM EDTA was added, and ultracentrifugation was performed in a 5-20% sucrose density gradient for 12 h at 43,000 rpm and 2 'C. To assay for C3 convertase activity 15-p1aliquots of selected fractions were incubated with 5 pl of C3 (1.4 mg/ml) for 30 min at 37 'C. 180 pl of cold VBS containing 0.1% gelatin (w/v) and 20 mM EDTA was then added to stop the reaction. Percentage of C3 consumed was determined by effective molecule titration. Fraction 1is at the top of the gradient.

12-

E a

1110-

.-s 9q 8c

._ E 7-

I b 5-

Ib

i2

1'4

16

18

20

Fraction Numb

FIG. 3. Determination of the sedimentation rate of C3b,Bb(Ni). '%labeled Factor B (5.9 S). '261-labeled C3 (9.5 S), '"I-labeled Clq (11 S), and '"I-labeled thyroglobulin ( T c ) (12 S) were used as reference substances and submitted to 5-20% sucrose density gradient ultracentrifugation together with C3b,Bb(Ni) as described under "Materials and Methods." The arrow indicates the position of Clb,'=I-Bb(Ni) in the gradient.

Downloaded from www.jbc.org by guest, on April 15, 2012

rotor and a Beckman W-65 Ultracentrifuge for 12 h at 43,000 rpm and 2 "C. Eight-drop fractionswere collected from the topof the tube using a Gilson Microfractionator (Gilson Medical Electronics, Inc., Middleton. WI). Separation of Free and Protein-bound aNi onBio-Gel P-6-Plastic tubes (10 X 75 mm; Falcon, Oxnard, CA) werepierced at their bottom with a needle, plugged with scrubbed nylon fibers (Fenwal Labs, Deerfield, IL), and packed with Bio-Gel P-6 (100-200 mesh; BioRad) in VBS. A small piece of Parafilm was wrapped around the middle of each tube which was then placed inside a plastic tube (12 X 75 mm; Falcon). To remove void volume buffer, these small Bio-Gel P-6 columns were centrifuged for 2 min at 1000 rpm and 4 'C in a CRU5000 centrifuge (IEC, Needham Heights, MA), and the inside tubes were transferred to clean tubes (12 X 75 mm).The samples containing =Ni (100-150 pl) were then layered on top of the Bio-Gel P-6, and the tubes were centrifuged at once as above. Only protein-bound "Ni was eluted from the column by this procedure and was collected at the bottom of the tubes (12 X 75 mm). When 10' cpm of =Ni in absence of protein was applied to such a Bio-Gel P-6 column, lese than 50 cpm were eluted. The association constant of a Ni-protein complex was derived from the slope of a Scatchard plot(24) in which bound/free Ni/mol of protein (I"') was plotted uersus bound Ni (mol/mol of protein). Hemolytic Assays-Sheep erythrocytes bearing rabbit hemolysin antibodies and human C1, C4b, and C2a were prepared as described (25,26). Titration of effective C3 molecules was performed as previously described (26).

7413

C3 Convertase of the Alternative Complement Pathway

-

5.9s

9.5s

12s

n

- 1 6000 -12000

-8000

1

-c

1200-

f

800-

E

-4000

10

20

1

-800 A. Wlthout 0

- 600 - 400

30

Fractlon Number

FIG. 4. Demonstration of "Ni intheCBb,Bb(Ni)enzyme using sucrose density gradient ultracentrifugation. 73 pgof C3b, 37 pgof '=I-labeled Factor B, and 2 ggof Factor D were incubated in 0.15 ml of VBS containing 0.1 mM B3Nifor 5 min at 24 "C. 10 mM EDTA was then added, and the reaction mixture was filtered through a Bio-Gel P-6 column. The reaction mixture was subjected to 5-20% sucrose density gradient ultracentrifugation for 12 h at 43,000 rpm and 2 "C. The fractions were analyzed for '=I and for "Ni. Arrows point at the position of the markers: '=I-labeled Factor B (5.9 S), '"I-labeled C3 (9.5 S), and "'I-labeled thyroglobulin (12 S).

ExperiFraction ment No." No.

1'0

1 20

21 22

2 3 4 5

21 1.42 18 19 18 19 1.34 16 17

"Nib

'?61"abe'ed Factor B"'

Ni/CBb,Bb

20

30

40

w

w

mol/ml

0.12 0.14 0.31 0.25 0.083 0.068 0.62 0.58 0.043 0.045

171 158 311 259 127 111 653 639 61 56

1.04 1.31 1.47

0.96 0.90 1.40

1.04 1.19 fi = 1.21 0.21 aB3Niwas quantitated in two fractions of the 10.7 S peak after

*

sucrose density gradient ultracentrifugation. *Total mass (nanogram)/fraction was determined from the total counts/min and thespecific activity of =Ni or '2sI-labeledFactor B. Total '=I-labeled Factor B mass was corrected after determining that theBb fragment contains 62% of '%I-labeled Factor B counts.

Fractlon Numbel

FIG. 5. Lack of C3b,Bb("Ni) complex formation in absence of Factor Das detected by sucrose density gradient ultracentrifugation. 36 pg of C3b and 18 pg of I3'I-labeled Factor B were incubated for 5 min at 24 "C in 0.1 ml of VBS containing 0.1 mM "Ni in presence or absence of 1pg of Factor D. 10 mM EDTA was added, and thereaction mixtures were filtered through Bio-Gel P-6 columns and subjected to sucrose density gradient ultracentrifugation as described under "Materials and Methods." Fraction l is at the top of the gradient.

Downloaded from www.jbc.org by guest, on April 15, 2012

TABLE I Quantitatwn of "Ni in C3b.Bb(Ni) ~~~~

63Nicorrelated with the 10.7 S peak of 1251-Fa~tor B which contained C3 convertase activity (Fig. 2). Since the sucrose density gradient contained 6 mM EDTA, it is apparent that the 10.7 S enzyme complex retained =Ni that was not susceptible to chelation by EDTA. The stoichiometry of Ni to C3b,Bb was determined from the specific activities of 63Ni and '251-Fa~tor B, and the results show (Table I) that over a Molar Excess of Mg Over Ni wide range of enzyme concentration (56-653 ng of 1251-Fa~tor B/fraction), the molar ratio of 63Nito C3b,Bb varied from 0.9 FIG. 6. Displacement of Ni by Mg during C3b,Bb formation to 1.47. Those results strongly suggest that there is only one as measured by decrease in ultracentrifugally detectable comNi ion/C3b,Bb which is resistant to chelation by EDTA. plex. 18 pg of C3b, 9 pg of I3lI-labeledFactor B, and 1 pg of Factor Further evidence for the identity of the 10.7 S material D were incubated for 5 min at 24 "C in 0.1 ml of VBS containing 0.1 came from two additional experiments. The 10.7 S peak was mMNiC12 and 0,0.2, 0.6,2.0, or 6.0 mMMgC12. 10 mM EDTA was and the reaction mixtures were subjected to sucrose density not observed when Factor D, which is essential for C3b,Bb added, gradient ultracentrifugation as described under "Materials and Methformation, was omitted from the reaction mixture (Fig. 5), ods." Amount of the C3b,'311-Bb(Ni) complex was determined by ''1 and itwas also missing when, prior to ultracentrifugation, the quantitation in the fast sedimenting radioactivity peak. Results are enzyme was incubated for 80 min at 37 "C, resulting in decay expressed as per cent of control (no MgC12). of its activity (not shown). Ni.Protein Interaction-To determine the ability of Mg to measured. At 60-fold molar excess ofMg over Ni (Fig. 6), displace Ni in C3b,Bb formation, a mixture of C3b, '251-Fa~tor most of the Ni was displaced by Mg, as evidenced by the loss B and Factor D was incubated for 5 min at 24 "C with 0.1 mM of the 10.7 S peak (Fig. 2). These results suggest that Ni and Ni andincreasing concentrations of Mg.The reaction mixture Mg compete for the same binding site in C3b,Bb. was then subjected to sucrose density gradient ultracentrifuThe relative affinity of Ni for C3b, Factor B, or Bb was gation, and the radioactivity in the 10.7 S C3b,Bb peak was assessed from the amount of 83Nibound to these proteins at

C3 Convertase of the Alternative Complement Pathway

7414

molecule of C3b (9.1 S ) and one molecule of Bb (4.5 S). Each C3b,Bb(Ni) complex contains only one Ni ion which is not susceptible to chelation by EDTA. Since Ni and Mg have similar ionic radii (29) and since Mg can displace Ni in enzyme formation, both metals probably bind to the same site in C3b,Bb. The stoichiometryof C3bBb:metal ion inthe enzyme ," 400 complex is proposedto be 1:l:l. That Ni binds to Bb more strongly than to C3b or Factor 2 .I3 0 0 4 B suggeststhat inC3b,Bb the Ni ion is bound to thecatalytic Bb subunit. The apparent association constant of the Ni .Bb complex at 4 "C, as calculated from the Scatchard plot in Fig. 7, is 8.3 X lo3 M" (log K = 3.9). In contrast, as quoted in Ref. 30, the association constant of the NieEDTAcomplex at 20 "C (log K ) is 18.56. It is probable therefore that in C3b,Bb 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 the metal ion is so tightly bound that all its coordination Bound Ni hole per mole protein) positions are occupied and not available for interaction with FIG. 7. Scatchard plots of Ni binding to Bb, Factor B, and EDTA. The much tighter binding of Ni to the complex C3b. 15 gg of Bb, 18 pg of Factor B, or 36 r g of C3b were incubated compared to its subunits may be a function of a critical for 10 min at 24 "C in 0.1 ml of VBS containing 0.1 DM =Ni and conformational change affecting the metal-binding site in Bb. increasing concentrations of NiClz up to 2 mM Ni. Proteins were Since upon decay dissociation of the enzyme, C3b is capable radiolabeled with '%I. Free Ni was then removed by a rapid gel of new enzyme formation while Bb is not, it is probable that filtration through Bio-Gel P-6, and the protein yield and amount of dissociated Bb loses the critical conformation necessary for Ni bound were determined. high affinity metal binding. In fact, it has been observed in this study that 63Ni is readily released from CBb,Bb(Ni) upon different Ni concentrations. The proteins were incubated at decay of the enzyme. This event may account for lack of a concentration of 2-3 X M in 0.1 mlofVBS for 10 min reassociation of the subunits to an active complex. at 4 "C with 0.1 mM @Niand increasing amounts of NiC12.To remove the free Ni, the reaction mixture was filtered through REFERENCES Bio-Gel P-6 and the amount of protein-bound Ni was deter- 1. Muller-Eberhard, H. J., and Schreiber, R.D. (1980) Adu. Immumined fromthe specific radioactivity. The proteins were used not!. 29, 1-53 in lZ5I-labeled form to allow their quantitation aftergel filtra- 2. Nicholson, A., Brade, V., Schorlemmer, H.-U., Burger, R., Bittertion. Scatchard plots (24) of Ni binding to C3b, Factor B, and Suermann, D., and Hadding, U.(1975)J. Immunol. 115,11081113 Bb (Fig. 7) are based on saturation curves. The association constant of Ni binding to Bb was 8.3 X lo3 M-'. The associa- 3. Medicus, R. G., Gotze, O., and Muller-Eberhard, H. J. (1976) J. Exp. Med. 144,1076-1093 tion constants for Factor B and C3bwere too low to be 4. Daha, M. R., Fearon, D. T., and Austen, K.F. (1976)J. Immunol. determined by the method used. However, it is apparent that 117,630-634 the affinity of =Ni for Bb is markedlyhigher than itsaffinity 5. Vogt, W., Dames, W., Schmidt, G., and Dieminger,L.(1977) Immunochemistry 14,201-205 for either C3b or Factor B. Upon addition of 20 mM EDTA to thereaction mixtures, more than 90% of the bound Ni was 6. Lesavre, P. H., and Muller-Eberhard, H. J. (1978) J. Exp. Med. 148,1498-1509 rapidly removed fromthe proteins (not shown). 7. Weiler, J. M., Daha, M. R., Austen, K. F., and Fearon, D. T. (1976) Proc. Natl. Acad. Sci. U. S. A. 73,3268-3272 DISCUSSION 8. Whaley, K., and Ruddy, S. (1976) J. Exp. Med. 144,1147-1163 Formation of the alternative pathway C3 convertase, 9. Pangburn, M. K., Schreiber, R. D., and Miiller-Eberhard, H.J. (1977) J. Exp. Med. 146,257-270 C3b,Bb,is Mg-dependent (15). Direct ultracentrifugal and electrophoretic demonstration of the inactive but stable 10. Law, S. K., Fearon, D. T., and Levine, R.P. (1979) J . Immunol. 122,759-765 proenzyme complex C3b,B(Mg) has been reported (1, 5). In 11. Gaither, T.A., Hammer, C. H., andFrank, M. M. (1979) J . contrast, C3b,Bb(Mg)is a labile enzyme witha half-life of 1.5 fmmunal. 123. 1195-1204 ~. . min at 37 "C and 3 h at 4 "C, and attempts to demonstrate 12. Fearon, D. T.,and Austen, K. F. (1975) J. Exp. Med. 142,856863 the complex by physical methods were unsuccessful. However, using the autoantibody nephritic factor which increases the 13. Gotze, O., Medicus, R. G., and Muller-Eberhard, H. J. (1977) J. Immunol. 118,525-528 half-life of the enzyme to 40 min at 37 "C, a trimolecular 14.Medicus, R, G., Esser, A. F., Fernandez, H. N., and Mullercomplex of C3b,Bb(Mg)-nephritic factor couldbe demonEberhard, H.J. (1980) J. Immunol. 124,602-606 strated by ultracentrifugation (27, 28). Whereas the proen- 15. Pillemer, L., Blum, L., Pensky, J., and Lepow, I. H.(1953) J. zyme complex C3b,B(Mg) isdissociated by EDTA, the active Immunol. 7 1,331-338 enzyme C3b,Bb(Mg)is refractory to EDTA, which raises the 16.Medicus,R. G., Gotze, O., and Muller-Eberhard, H. J. (1976) S c a d . J . Immunol. 5,1049-1056 question as towhether the metal ion is present in the active J. E., and Niemann, M.A. (1980) J. Biol. Chem. 256, enzyme complex. We have recently found (19) that Ni can 17.Mole, 8472-8476 replace Mg in C3b,Bb formation. Enzyme formation with Ni 18. Reid. K. B. M.. and Porter. R. R. (1981) Annu. Reu. E ~ ~ h e m is more efficient than with Mg, and in addition C3b,Bb(Ni) 56.433-464 . is more stable than C3b,Bb(Mg).In spiteof these differences, 19. Fishelson, Z., and Miiller-Eberhard, H. J. (1982) J. I m n U n o l . 129,2603-2607 C3b,Bb(Ni)and C3b,Bb(Mg) resembleeach other with respect to kinetic characteristics, stabilization by properdin, acceler- 20. Tack, B. F., and Prahl, J. W. (1976) Biochemistry15,4513-4520 0..and Miiller-Eberhard, H. J. (1971) J . Ew. Med. 134, ation of decay by Factor H, and resistance to EDTA (19). A 21. Gotze. 90s~108s stable and active C3b,Bb(Ni) complex with a sedimentation 22. Lesavre, P . H., Hugli, T. E., Esser, A. F., and Muller-Eberhard, coefficient of 10.7 S has now been demonstrated by sucrose H. J. (1979) J. Zmnunol. 123,529-534 density gradient ultracentrifugation. This sedimentation coef- 23. Fraker, P. J., and Speck, J. C., Jr. (1978) Biochem. BWphYs. Res. Comm. 80,849-857 ficientis compatible with a complexcomposition of one

\

Downloaded from www.jbc.org by guest, on April 15, 2012

C3 Convertme of the Alternative Complement Pathway 24. Scatchard,G. (1949) Ann. N. Y. Acad. Sci. 61,660-672 25. Cooper, N.R.,and Mdler-Eberhard, H.J. (1968) Immmclaemistv 5, 155-169 26. Cooper, N. R.,and Miiller-Eberhard, H.J. (1970) J. Exp. Med. 132,775-793 27. Schreiber, R. D.,Gotze, O., and Miiller-Eberhard, H. J. (1976) S c a d . J. Immunol. 6,705-713

7415

28. Daha, M. R., Fearon, D. T., and Austen, K.F. (1976) J. Immuml. 116,l-7 29. Weast, R. C. (1981/82) in CRC Handbook of Chemistry and Physics (Weast, R.C., ed) 62nd Ed., pp. F175, CRC Press, Inc.,

Boca Raton, Florida 30. Kroll, H., and Gordon, M. (1960) Ann. N. Y. Acad. Sci. 88,341352

Downloaded from www.jbc.org by guest, on April 15, 2012