The tick-over theory revisited: formation and regulation of the soluble alternative complement C3 convertase (C3(H2O)Bb)

Abstract

The molecular interactions between the components of the C3 convertase of the alternative pathway (AP) of complement and its regulators, in both surface-bound and fluid-phase form, are still incompletely understood. The fact that the AP convertase is labile makes studies difficult to perform. According to the so called tick-over theory, hydrolyzed C3, called C3(H(2)O), forms the initial convertase in fluid phase together with factor B. In the present study, we have applied western blot analysis and ELISA together with fluorescence resonance energy transfer (FRET) to study the formation of the fluid-phase AP convertases C3(H(2)O)Bb and C3bBb and their regulation by factor H and factor I at specific time points and, with FRET, in real time. In our hands, factor B showed a higher affinity for C3(H(2)O) than for C3b, although in both cases it was readily activated to Bb. However, the convertase activity of C3bBb was approximately twice that of C3(H(2)O)Bb, as monitored by the generation of C3a. But in contrast, the C3(H(2)O)Bb convertase was more resistant to inactivation by factor H and factor I than was the C3bBb convertase. Under conditions that totally inactivated C3bBb, C3(H(2)O)Bb still retained approximately 25% of its initial activity.

FIGURE 1

(A) Non-reducing SDS-PAGE gel of C3b (left lane), C3(H₂O) (right lane). (B) Western blot analysis under reducing conditions of (from left to right) C3(H₂O), C3b, EDTA-plasma, and serum using polyclonal anti C3c-antibodies (C) Western blot analysis of Alexa-labeled proteins (reducing SDS-PAGE) under UV illumination. In the sample in the far left lane, factor B (fB Act) had been incubated with factor D and C3b for 60 min to generate Ba and Bb. The other lanes show uncleaved factor B (fB), C3(H₂O), and C3b, respectively. The position of uncleaved factor B (fB (whole)), Bb and Ba, and the alpha and beta chains of C3, and the molecular weights of the markers are indicated.

FIGURE 2

Binding of C3(H₂O) and C3b to factor B in solution. Normalized emission spectra of C3(H₂O)-Alexa 594 (A), C3b-Alexa 594 (B), and C3c-Alexa 594 (C) at 2 nM (squares) and 4 nM (diamonds) with 2 nM factor B-Alexa 546 when excited at 553 nm (the excitation peak for A546). The decrease at 571 nm (A546) and increase at 611 nm (A594) (arrows) indicates an energy transfer that occurs when factor B binds C3(H₂O) or C3b in solution (n=3).

FIGURE 3

Generation of AP convertases, as detected by FRET. Normalized 611 nm peak emission of labeled C3b/C3(H₂O) over time, after excitation at 553 nm (the peak excitation for Alexa 546). After 60 s, labeled factor B with or without unlabeled factor D (fD) was added to C3(H₂O)-Alexa 594 (top) or C3b-Alexa 594 (bottom) in VBS-Mg²⁺. The immediate increase in emission at 60 s is indicative of an interaction between C3b/C3(H₂O) and factor B, which is dampened by the presence of factor D. Similar results were seen in VBS- Ni²⁺ (not shown).

FIGURE 4

Regulation of AP convertases by factor I and factor H. Normalized 611 nm peak emission of labeled C3b/C3(H₂O) over time, measured as described in Fig. 3. At 60 s, labeled factor B was added to C3(H₂O)-Alexa 594 (top) or C3b-Alexa 594 (bottom) in VBS-Mg²⁺. Factor H (FH) and factor I (FI) were added sequentially at 60 s (together with factor B) and at 180 s as indicated, one at each time point. The black curves show factor H added first, followed by factor I, and the grey curves show the reverse. The dotted curve shows the result for addition of buffer alone, to indicate the extent of the dilution effect.

FIGURE 5

Regulation of AP convertases by factor I and factor H, added simultaneously. Normalized emission of C3(H₂O)–Alexa 594 (above) and C3b-Alexa 594 (below), measured as described in Fig. 3. Factor B-Alexa 546 was added at 60 s with or without unlabeled factor H and factor I (fH/I) in VBS-Mg²⁺. If not added at 60 s, fH/I was added at 180 s.

FIGURE 6

Generation of C3a (9 kDa) by soluble AP convertases and regulation by factor I and factor H. The figure shows western blots with densitometric measurements of the C3a fragments generated from C3(H₂O)-Alexa 594 (top) or C3b-Alexa 594 (bottom) incubated with factor B-Alexa 546, unlabeled factor D, and unlabeled substrate C3 for up to 30 min at 37°C. Tests were performed without (diamonds) or with (squares) preincubation with factor H and factor I for 1 minute. The densitometric measurements were normalized to results from corresponding samples in the C3a ELISA (displayed in inserts as mean ± SD). Lane 1: MW markers as in Fig 1 (lowest band 10 kDa, followed by 15, 20, and 25 kDa); lanes 2, 3: without factor I and factor H, 0 and 30 min; lanes 4, 5: with factor I and factor H, 0 and 30 min.

FIGURE 7

Model describing the reactions that occur when C3b/C3(H₂O) is incubated with factor B in the presence of factor D. C3b/C3(H₂O) continually associates and dissociates with uncleaved factor B (Ba + Bb, arrow 1), producing a FRET effect and increased emission. Upon addition of factor D, factor B is cleaved, and the Ba portion dissociates from the complex (arrow 2). The C3b/C3(H₂O)/Bb complex rapidly decays to C3b/C3(H₂O) alone (arrow 3). The free Ba fragments in solution, however, can reassociate with C3b/C3(H₂O) (arrows 4, 5), hindering the binding of intact factor B and thus producing a lower FRET efficiency.