Functional Characterization of the Complement Control Protein Homolog of Herpesvirus Saimiri

Herpesvirus saimiri (HVS) is a lymphotropic virus that causes T-cell lymphomas in New World primates. It encodes a structural homolog of complement control proteins named complement control protein homolog (CCPH). Previously, CCPH has been shown to inhibit C3d deposition on target cells exposed to complement. Here we have studied the mechanism by which it inactivates complement. We have expressed the soluble form of CCPH in Escherichia coli, purified to homogeneity and compared its activity to vaccinia virus complement control protein (VCP) and human complement regulators factor H and soluble complement receptor 1. The expressed soluble form of CCPH bound to C3b (KD = 19.2 μm) as well as to C4b (KD = 0.8 μm) and accelerated the decay of the classical/lectin as well as alternative pathway C3-convertases. In addition, it also served as factor I cofactor and supported factor I-mediated inactivation of both C3b and C4b. Time course analysis indicated that although its rate of inactivation of C4b is comparable with VCP, it is 14-fold more potent than VCP in inactivating C3b. Site-directed mutagenesis revealed that Arg-118, which corresponds to Lys-120 of variola virus complement regulator SPICE (a residue critical for its enhanced C3b cofactor activity), contributes significantly in enhancing this activity. Thus, our data indicate that HVS encodes a potent complement inhibitor that allows HVS to evade the host complement attack.

The complement system is an integral participant in the innate mechanisms of immunity and, thus, has a burden of performing surveillance in the host and protecting it from all the pathogens including viruses (1,2). Earlier studies have decisively demonstrated that both acute and latent viruses are susceptible to complement-mediated neutralization (3,4). Thus, complement exerts a strong selective pressure on viruses during infection. These data suggest that for their successful survival, viruses must have developed mechanisms to subvert this system. Consistent with this premise, genome sequencing of poxviruses and herpesviruses have shown that members of these families encode for structural homologs of human regulators of the complement activation (RCA) 3 family (5)(6)(7)(8)(9).
Sequence comparison of the viral homologs of RCA (vCCPs) show that the sequence similarity between the poxvirus homologs exceeds 91%, whereas that among the herpesvirus homologs varies from 43 to 89%. These data suggest that the herpesvirus homologs are more diverse in structure compared with the poxvirus homologs. Whether this structural diversity in herpesvirus homologs is also reflected in their function is not clear, as among the herpesvirus homologs, detailed functional characterization has been performed only for the Kaposi's sarcoma-associated herpesvirus homolog (Kaposica/KCP) (19 -22).
Herpesvirus saimiri (HVS), the prototype of rhadinoviruses, is regularly found in its natural host, the squirrel monkey. Although it does not cause any disease in its natural host, infection in other New World primates such as tamarins, common marmosets, and owl monkey causes acute peripheral T cell lymphoma within less than 2 months (23,24). In addition, the virus is also capable of transforming simian and human T cells in vitro (25,26). Interestingly, unlike any other viruses, the HVS harbors two homologs of complement regulatory proteins, (i) a homolog of RCA encoded by ORF4 and (ii) a homolog of the terminal complement inhibitor CD59 encoded by ORF15 (27,28). The ORF4 was predicted to encode a protein containing four CCP modules followed by a transmembrane domain. Analysis of posttranscriptional processing indicated that ORF4 transcript occurs as unspliced as well as single-spliced mRNA. The unspliced mRNA codes for a membrane-bound glycoprotein containing four extracellular CCPs along with a transmembrane region, whereas the spliced mRNA codes for a soluble protein that lacks transmembrane region (29).
Initial characterization of the RCA homolog of HVS (named complement control protein homolog, CCPH) showed that expression of the membrane form of this protein on BALB/3T3 cells inhibited C3d deposition on these cells when they were incubated with whole human serum (30). Although this study demonstrated the complement inhibiting activity of this protein, the mechanism by which it inactivates complement activation was not elucidated. In the present study we describe the mechanism of complement regulation by the RCA homolog of HVS. Our results show that the soluble form of the RCA homolog (sCCPH; CCP1-4 without the transmembrane domain) interacts with complement proteins C3b as well as C4b and accelerates decay of the classical/lectin and alternative pathway C3-convertases. In addition, the protein also has the ability to serve as factor I cofactor and support factor I-mediated inactivation of C3b and C4b. Importantly, we show that its factor I cofactor activity for C3b is 14-fold higher in comparison to VCP, the most completely characterized vCCP, and that Arg-118 plays a critical role in enhancing this activity.
Complement Proteins and Their Proteolytically Activated Products-The human complement protein C3 was purified according to Hammer et al. (31) with minor modifications as previously described (17), and native C3 was separated from C3 (H 2 O) by running on a Mono S column (32). The complement factors H and I were kindly provided by Prof. Michael K. Pangburn (University of Texas Health Centre, Tyler, TX.). Human factor B was purified as follows. Human plasma was subjected to a stepwise precipitation with 11 and 26% polyethylene glycol. The 26% polyethylene glycol precipitate was dissolved in 10 mM sodium phosphate, pH 7.4, run on Source Q column (Amersham Biosciences) in the same buffer, and eluted with a linear gradient of 0 -0.5 M NaCl. Fractions containing factor B were identified by Ouchterlony analysis, pooled, and loaded onto a Mono S 5/5 column (Amersham Biosciences) in 50 mM sodium phosphate, pH 6.0. Bound proteins were eluted with a linear salt gradient of 0 -0.5 M NaCl and analyzed by SDS-PAGE. Homo-geneous factor B fractions were pooled and concentrated. The recombinant human soluble form of complement receptor type 1 (sCR1) was a generous gift from Dr. Henry Marsh (AVANT Immunotherapeutics, Inc., Needham, MA.). C3b, the proteolytically activated form of C3, was generated by limited tryptic cleavage of C3 and purified on a Mono Q 5/5 (Amersham Biosciences) column as previously described (16). C4b, the proteolytically activated form of C4, was purchased from Calbiochem. Purity of all the proteins exceeded 95%, as judged by SDS-PAGE analysis.
Cloning, Expression, Purification, and Refolding of the Soluble Form of Herpesvirus Saimiri Complement Control Protein Homolog (sCCPH) and the R118A Mutant-The herpesvirus saimiri CCPH gene (CCP domains 1-4) was PCR-amplified from the CCPH clone pCEX-1 (a kind gift of Drs. John Lambris, Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA and Jens-Christian Albrecht, Institut für Klinische und Molekulare Virologie, Erlangen, Germany) with specific primers 5Ј-GGAATTCAGCTGTCCTA-CACGTAACCAG-3Ј (the EcoRI site is underlined) and 5Ј-CCGCTCGAGCATACATTCAGGAATAGCTGG-3Ј (the XhoI site is underlined) and cloned into the bacterial expression vector pET29 at the EcoRI and XhoI sites. The R118A mutant was constructed from this clone by using the QuikChange II site-directed mutagenesis kit (Stratagene, La Jolla, CA). After verifying the fidelity of both the clones by DNA sequencing, they were transformed into Escherichia coli BL21 cells for expression.
Expression of sCCPH and R118A mutant (numbering according to the mature protein sequence (29)) was performed as described below. A single colony of the bacterial clone expressing sCCPH or the mutant protein was inoculated into 5 ml of LB-kanamycin media (LB media containing 30 g/ml kanamycin) and grown overnight at 37°C, and 2 ml of this culture was transferred into 100 ml of LB-kanamycin. The culture was grown for 2 h at 37°C, and thereafter 10 ml of the culture was transferred to 600 ml of LB-kanamycin and grown at 37°C until the optical density reached 0.6 at A 600 . Protein expression was induced by the addition of 1 mM isopropyl 1-thio-␤-D-galactopyranoside, and the induced culture was further grown for 4 h. The cells were harvested by centrifugation at 8000 rpm at 4°C.
For purification of the expressed proteins, frozen cell pellets (ϳ16 g) were gently resuspended in 48 ml of 50 mM Tris, pH 8.0, 1 mM EDTA, 100 mM NaCl, and 0.1 mM phenylmethylsulfonyl fluoride. The cell suspension was then treated with lysozyme (0.3 mg/ml), stirred for 20 min, mixed with deoxycholic acid (1.3 mg/g), and stored at 37°C for 30 min. After incubation, the lysate was sonicated with 15 pulses of 15 s each and centrifuged at 10,000 ϫ g for 20 min at 4°C. The pellet containing the inclusion bodies was washed twice with 50 mM Tris, pH 8.0, 10 mM EDTA, 100 mM NaCl, and 0.5% Triton X-100 and solubilized in 50 mM Tris, pH 8.0, 1 mM EDTA, 100 mM NaCl, and 8 M urea. The sample was then centrifuged at 4°C for 30 min at 11,000 rpm, and the supernatant obtained was loaded onto a nickel nitrilotriacetic acid-agarose column (Qiagen, Hilden, Germany) pre-equilibrated with 100 mM NaH 2 PO 4 , 10 mM Tris, 8 M urea, pH 8.0. The column was washed with the binding buffer containing 10 mM imidazole, and the bound protein was eluted with 400 mM imidazole.
The purified proteins were refolded by using a rapid dilution method (1:50) described previously (33). In brief, the purified protein was added dropwise with continuous stirring into a refolding buffer containing 0.02 M ethanolamine, 1 mM EDTA, 0.5 M L-arginine, 1 mM reduced glutathione, and 1 mM oxidized glutathione, pH 11.0. The sample was then left static for 36 h. The refolded sample was concentrated, dialyzed against phosphate-buffered saline, and subjected to SDS-PAGE, circular dichroism, and sequencing and mass analysis by mass spectrometry (17).
Measurement of Factor I Cofactor Activity-Analysis of factor I cofactor activities of sCCPH and the mutant was essentially performed as described (34). These assays were performed in physiological ionic strength buffer (phosphate-buffered saline).
Measurement of Decay-accelerating Activity-The classical pathway (CP) decay-accelerating activity of sCCPH and the R118A mutant were determined by forming EAC142 (35), and the alternative pathway (AP) C3-convertase decay-accelerating activity was measured by forming C3b,Bb on sheep (E S ) as well as rabbit (E R ) erythrocytes. The details of these methods have been described previously (17,36).
Circular Dichroism (CD)-The sCCPH and its mutant R118A were subjected to CD spectra in the far UV region (190 -360 nm) using a Jasco J18 spectropolarimeter with a cylindrical quartz cell with a path length of 0.01 cm. The resolution was 1 nm, the sensitivity was 20 millidegrees, and the speed was 10 nm/min. Each presented spectrum is the measure of eight measurements. The concentration of both the proteins was 200 g/ml in 10 mM phosphate containing 145 mM NaCl, pH 7.4. All the data were subtracted against the background data using the spectral analysis software.
Flow Cytometry for Measurement of Inhibition of C3b Deposition-Inhibition of the classical and alternative pathwaymediated C3b deposition on erythrocytes by sCCPH and VCP was measured by flow cytometry (19). For measurement of the classical pathway-mediated C3b deposition, 5 l of EA (10 9 /ml in GVB 2ϩ ) was mixed with 2 l of C8-deficient human serum (Calbiochem) and 2 M sCCPH or VCP in a total volume of 44 l and incubated for 30 min at 37°C. The cells were washed with GVB 2ϩ , centrifuged, mixed with 100 l of 1 ⁄ 100-diluted fluorescein isothiocyanate-conjugated F(abЈ)2 anti-C3 goat IgG (Cappel Laboratories, Warrington, PA), and further incubated on ice for 1 h. After incubation, the cells were washed twice with 400 l of GVB, resuspended in 1.0 ml of the same buffer, and analyzed on a FACS Vantage (BD Biosciences). For measurement of alternative pathway-mediated deposition of C3b, 5 l of rabbit erythrocytes (10 9 /ml in GVB) was mixed with 2 l of 0.1 M MgEGTA, 3 l of C8-deficient human serum (Calbiochem), and 30 l of GVB or GVB containing 2 M sCCPH or VCP and incubated for 30 min at 37°C. The cells were washed with GVB, and deposition of C3b was detected as described above. Results are expressed as mean channel fluorescence of 10,000 cells.
Surface Plasmon Resonance Measurements-The kinetics of sCCPH and the R118A mutant binding to C3b and C4b was determined on the surface plasmon resonance-based biosensor BIACORE 2000 (Biacore AB, Uppsala, Sweden). The experiments were performed in phosphate-buffered saline-Tween (10 mM sodium phosphate, 145 mM NaCl, pH 7.4, containing 0.05% Tween 20) at 25°C. For proper orientation of these proteins, the free SH groups of both C3b and C4b were biotinylated and then immobilized on the streptavidin chip (Sensor Chip SA, Biacore AB) (34). FC-2 was immobilized with C3b (1592 RU), FC-3 was immobilized with C4b (1197 RUs), and FC-1 (blank flow cell) served as the control flow cell. Because sCCPH showed very little binding to C3b, more C3b molecules were deposited onto FC-2 by forming C3-convertase (37,38). In brief, ϳ6000 RUs of C3b were deposited using three cycles of C3b deposition. In each cycle, the C3-convertase was formed by injecting a mixture of factors B and D (5 g of B and 0.35 g of D) and then 45 g of native C3 was injected using the co-inject option. Deposition of C3b onto the chip was performed in veronal-buffered saline containing 1 mM NiCl 2 . For measurement of binding of sCCPH and the mutant protein to C3b and C4b, various concentrations of these proteins were injected for 120 s at 50 l/min. Dissociation was measured for 180 s. The sensor chips were regenerated with 30-s pulses of 0.2 M sodium carbonate, pH 9.5. Sensograms obtained for the control flow cell (FC-1) were subtracted from the data for the flow cell immobilized with C3b or C4b, and the surface plasmon resonance data obtained were evaluated by BIAevaluation software version 4.1 using global fitting.

RESULTS
Expression and Characterization of sCCPH-Because a large quantity of protein was required for conducting multiple assays, we chose to express HVS sCCPH in E. coli using the pET expression system. The soluble form of HVS CCPH was amplified from the HVS clone pCEX-1 and cloned into the expression vector pET29. The expressed protein was purified to homogeneity using histidine affinity (Fig. 1), and the identity of the expressed protein was confirmed by sequencing using mass spectrometry. The amino acid sequence of the expressed protein was consistent with the predicted sequence confirming the identity; the sequence coverage obtained was 86%. The expressed protein was refolded according to the method described by R. A. Smith and co-workers (33). This method typically provided a 20% yield. Analysis of the refolded protein by circular dichroism yielded a peak around 230 nm, which is a characteristic of CCP domains (39) (Fig. 1). These data confirmed proper folding of the protein. The expressed protein was Ͼ95% pure as judged by SDS-PAGE analysis, and it migrated as a single band of 32,000 Da on the gel. Further mass analysis by mass spectrometry confirmed that its molecular mass was similar to its calculated mass (within error Ͻ1%) (Fig. 1).
Previously it has been shown that the membrane form of CCPH inhibits C3d deposition on the target cells (30). To verify if the refolded protein is biologically active, we tested its ability to inhibit C3b deposition on erythrocytes during complement activation. As depicted in Fig. 2, sCCPH inhibited both the classical as well as alternative pathway-mediated deposition of C3b on erythrocytes. Importantly, the data indicated that sCCPH was more active than VCP in inhibiting the alternative pathway-mediated deposition of C3b.
Kinetic Analysis of Interaction of sCCPH with Complement Proteins C3b and C4b-The human complement control proteins inactivate complement by targeting C3b and/or C4b. Because sCCPH inhibited C3b deposition mediated by both the classical and alternative pathways, we sought to analyze its interaction with C3b and C4b by surface plasmon resonance technology. In this assay, C3b and C4b were immobilized in their physiological orientation on a streptavidin chip by labeling their free SH groups with biotin (34), and sCCPH was injected over the chip to measure binding. The sCCPH showed good binding to C4b but very weak binding to C3b (Fig. 3, upper left panel). Binding data obtained by injecting various concentrations of sCCPH fitted well to 1:1 binding model (k a ϭ 158; k d ϭ 5.32 ϫ 10 Ϫ3 ; K D ϭ 3.35 ϫ 10 Ϫ5 ; 2 ϭ 0.273). Because the binding response was very low, we further deposited C3b on the sensor chip to increase the response and reevaluate affinity. More C3b was deposited by forming AP C3-convertase on the chip and flowing native C3   (34)). Top right, deposition of C3b by forming AP C3-convertase on the chip (37,38). Factors B and D mix and C3 (in veronal-buffered saline containing 1.0 mM NiCl 2 ) were repeatedly injected over the chip using the co-inject option of Biacore 2000. Bottom left, sensogram overlay for the interaction between sCCPH and C3b deposited using the AP C3-convertase. The solid lines represent the global fitting of the data to a 1:1 Langmuir binding model (A ϩ B 7 AB; BIAevaluation 4.1). The concentration of sCCPH injected is indicated at the right of the sensograms. Bottom right, sensogram overlay for the interaction between sCCPH and C4b. The arrow indicates the time point used for evaluating the steady-state affinity data. The concentration of sCCPH injected is indicated at the right of the sensograms. (37,38). Three cycles of AP amplification resulted in deposition of ϳ6000 RUs of C3b (Fig. 3, upper right panel); non-covalently associated Bb and C3b were removed by injection of brief pulses of 0.2 M sodium carbonate, pH 9.5. As expected, deposition of C3b using this approach resulted in a decaying surface, suggesting that C3b was attached to the surface by forming ester linkages (40). When sCCPH was flown over the enzyme-coupled C3b, it showed good binding response (Fig. 3, lower left panel).
To calculate affinities for C3b and C4b, various concentrations of sCCPH were injected over the chip. The sCCPH bound to both C3b and C4b in a dose-dependent manner (Fig. 3). Global fitting analysis of the sensograms showed a good fit of sCCPH-C3b data to the 1:1 binding model with a drifting base line ( 2 ϭ 1.83), but sCCPH-C4b data could not be fitted to 1:1 model, and therefore, it was evaluated by steady-state analysis. These data indicated that sCCPH-C3b interaction follows a simple 1:1 binding model, whereas sCCPH-C4b interaction is complex. A comparison of the affinities showed that sCCPH has a 24-fold higher affinity for C4b than C3b (Table 1).
sCCPH Acts as a Cofactor in Factor I-mediated Cleavage of C3b and C4b-The data presented above indicated that sCCPH interacted with C3b as well as C4b (Fig. 3). We, therefore, analyzed the ability of sCCPH to serve as a cofactor in factor I-mediated cleavage of C3b and C4b. A fluid phase assay was utilized for determining the factor I cofactor activity wherein C3b or C4b was incubated with factor I and sCCPH or the control proteins (VCP, factor H, or sCR1), and cleavages of the ␣Ј chain were assessed by running the samples on SDS-PAGE gels. During factor I-mediated cleavage of C3b, the ␣Ј chain of C3b was cleaved at three distinct sites depending on the cofactor involved. In the assay utilized, the appearance of 68-and 46-kDa fragments indicated cleavage at site 1 and generation of iC3b 1 , the appearance of 68-and 43-kDa fragments indicated cleavages at sites 1 and 2 and generation of iC3b 2 , and the appearance of 43-kDa, C3dg, and 25-kDa fragments indicated the cleavage at all the three sites and generation of C3c and C3dg. In the case of C4b, factor I is known to cleave at two sites. The appearance of C4d and 25-and 16-kDa fragments indicated both these cleavages and generation of C4c and C4d.
It is clear from the data presented in Fig. 4 that sCCPH acted as a factor I cofactor in mediating cleavages of C3b as well as C4b. The data further showed that its cofactor activity for C3b differed from VCP in two ways; (i) it was much more efficient than VCP in supporting C3b cleavages, and (ii) unlike VCP, which primarily supported factor I cofactor activity for the first site, it displayed efficient cofactor activity for both sites 1 and 2, leading to generation of iC3b 2 . In comparison to human complement regulators, its cleavage pattern was similar to factor H, but it differed from sCR1 in that it did not support the third cleavage. The cofactor activity of sCCPH for C4b was similar to VCP and sCR1, and like these proteins it also supported the cleavages at both the sites and led to generation of C4c and C4d.
Because sCCPH showed greater factor I cofactor activity for C3b than VCP (Fig. 4), we studied C3b inactivation as a function of time using equimolar concentrations of different cofactors (Fig. 5). The data indicated that the time required for 50% cleavage of the ␣Ј chain of C3b for sCCPH was 2.5 min, whereas that for VCP was 34 min, indicating that sCCPH was about a 14-fold more efficient cofactor than VCP. The sCCPH, however, was less efficient compared with human complement regulators factor H and sCR1; the time required for 50% cleavage of the ␣Ј chain of C3b in the presence of factor H and sCR1 was 1.25 and 1.0 min, respectively. A similar time course experiment performed for C4b cleavage showed that sCCPH was ϳ2-fold less efficient than VCP (Fig. 6). The time required for 50% cleavage of the ␣Ј chain of C4b for sCCPH, VCP, and sCR1 was 27, 14.5, and 3.5 min, respectively. sCCPH Accelerates the Decay of the Classical and Alternative Pathway C3-convertases-The decay-acceleration activity of sCCPH for the classical and alternative pathway C3-convertases was measured by hemolytic assays. For measuring the CP decayaccelerating activity, sensitized sheep erythrocytes coated with the C3-convertase enzyme (C4b,2a) was incubated with sCCPH or the control pro-   AUGUST 11, 2006 • VOLUME 281 • NUMBER 32 teins, and the enzyme decay was quantitated by measuring hemolysis after adding EDTA sera. sCCPH showed good decayaccelerating activity and decayed 83% enzyme activity at 1 M, but on a molar basis it was 5.8-fold less active than VCP and 18-fold less active than sCR1 (Fig. 7A).

Functional Characterization of HVS CCPH
The AP C3-convertase decay-accelerating activity was quantitated by forming the enzyme (C3b,Bb) on sheep or rabbit erythrocytes and incubating it with sCCPH or the control protein. In comparison to VCP, sCCPH showed 8-fold more decay activity for the enzyme present on sheep cells (Fig. 7B). It was, however, much less efficient compared with human complement regulators factor H and sCR1. Previously it has been shown that VCP does not decay the AP C3-convertase present on rabbit cells (17). To see if this also holds true for sCCPH, we tested its decay activity for the enzyme present on rabbit erythrocytes. Our results showed that unlike VCP it also decays the enzyme present on rabbit erythrocytes. The concentration required for 50% decay was 12.5 M, which was 1.6-fold more compared with the enzyme present on sheep cells.
Arginine 118 of sCCPH Is Crucial for Its Cofactor Activities but Not Decay-accelerating Activities-Because sCCPH showed a robust factor I cofactor activity for C3b cleavage, we sought to identify the determinant that is responsible for its enhanced activity. Recent studies on SPICE has shown that it possesses ϳ100-fold more factor I cofactor activity for C3b than VCP (18), and residues Lys-108 and -120 are primarily responsible for this enhanced cofactor activity (41). Sequence alignment of sCCPH with SPICE showed that sCCPH has Leu (Leu-106) and Arg (Arg-118) in the corresponding positions. Because earlier studies have shown that ionic interactions play a critical role in CCP-C3b/C4b interactions (34,(42)(43)(44) we suspected that Arg-118 might be responsible for the enhanced C3b cofactor activity. To further probe this possibility, we built a three-dimensional model of sCCPH by homology modeling using the crystal structure of VCP (45) as the template. The model structure showed that Arg-118 is exposed to solvent (Fig.  8), which further supported the possibility that this residue could be involved in the cofactor activity.
To study the involvement of Arg-118 in the functional activities of sCCPH, we mutated the Arg-118 to Ala and expressed the R118A mutant in E. coli using the pET expression system. Purification and refolding procedures used for the mutant were essentially similar to that of sCCPH. Sequencing (sequence coverage obtained was 91%) and circular dichroism analysis confirmed the identity and correct folding of the mutant (Fig. 1), respectively. Functional analysis of the R118A mutant showed that its factor I cofactor activity for C3b was drastically decreased compared with sCCPH; it was 50-fold less active compared with sCCPH (Fig. 9). Interestingly, the mutant also showed a 12-fold decrease in factor I cofactor activity for C4b (Fig. 9).
Next, we analyzed if this mutation also affects the decayaccelerating activities of sCCPH. The data showed that the mutant was as active as sCCPH in accelerating the decay of the classical as well as alternative pathway C3-convertases (Fig. 10). Together these data indicated that Arg-118 plays an important role in enhancing the cofactor activities but not the decay-accelerating activities of sCCPH.
Binding of R118A Mutant to C3b and C4b-Binding of the complement control proteins to C3b and C4b is a prerequisite for imparting factor I cofactor activities and decay-accelerating activities; however, a significant body of literature suggests that binding does not always correlate well with these activities (14,46). To determine whether R118A mutation affected binding, we measured binding of this mutant to C3b and C4b using the surface plasmon resonance assay. The R118A mutant showed about an 11-fold decrease in affinity for C3b compared with sCCPH (Table 1), which was a result of a 7.3-fold decrease in the on-rate and a 1.5-fold increase in the off-rate. These data are consistent with the substantial decrease in C3b cofactor activity. There was, however, no decrease in the affinity of C4b (Table 1), although the mutant showed a 12-fold decrease in C4b cofactor activity. Furthermore, it is clear from the data provided in Fig. 10 that R118A FIGURE 5. Time course of factor I cofactor activity of sCCPH, VCP, factor H (fH), and sCR1 for complement protein C3b. Cofactor activity was measured by incubating 3 g of C3b with 0.05 g of factor I and 0.5 M concentrations of the cofactors (as indicated in each gel) at 37°C for the indicated time period. The reactions were stopped by adding the sample buffer containing dithiothreitol, and the cleavage products were visualized by separating the samples on 9.5% SDS-PAGE gel and staining with Coomassie Blue. The intensities of the ␣Ј chain were determined by densitometric analysis and represented graphically (lower panel). mutation had no effect on the decay-accelerating activities; hence, the decrease in K D value for C3b did not correlate with the AP C3-convertase activity.

DISCUSSION
The regulators of human complement belonging to the RCA family contain 4 -59 copies of CCP domains. It is believed that the sequence variations imposed on the CCP domain fold and the interdomain dynamics determine the differences in functionality of the complement regulators (8,11,(47)(48)(49). In viruses, homologs of complement regulators have been described in members of Herpesviridae and Poxviridae. The sequence similarity among the poxviral complement regulators exceeds 91%, whereas that among the herpesviral complement regulators ranges between 43 and 89%. Therefore, functional characterization of various herpesviral complement regulators would help in determining whether structural diversity in these regulators has led to any change in their functional diversity. Functional complement regulators in Herpesviridae family have been described in herpesvirus saimiri (30), ␥-herpesvirus 68 (50), and Kaposi's sarcoma-associated herpesvirus (KSHV) (19,20), but detailed functional analysis for decay-accelerating activities, factor I cofactor activities, and binding to C3b and C4b have been performed only for the KSHV complement regulator (Kaposica/KCP) (19 -21). In the present study we have analyzed the functional activities of HVS sCCPH to get insight into the functional diversity of sCCPH against the complement system.
A previous study had shown that the membrane form of HVS CCPH inhibits the classical pathway-mediated deposition of C3d onto the target cells (30). Our data on inhibition of C3b deposition onto the target cells by sCCPH show that it inhibits both the classical as well as alternative pathway-mediated deposition of C3b (Fig. 2). These results are consistent with the previous data on herpesviral (␥-HV68 and Kaposica) and poxviral (VCP and SPICE) complement regulators (16,18 -20, 50), which showed inhibitory activities against both the pathways. Earlier, using hemolytic assays it has been shown that vCCPs are efficient in inactivating the classical pathway (IC 50 ϭ 0.1-0.2 M) (16,19); these values are considered significant because the local concentration of these proteins at the site of infection is expected to be very high (2,20,51). Measurement of inhibition of the CP-mediated lysis of sheep erythrocytes showed that like other vCCPs, sCCPH is also an effective inhibitor of the classical pathway (IC 50 ϭ 0.27 M). Further analysis of the CP C3-convertase regulatory activities demonstrated that it contains both effective factor I cofactor activity for C4b (Fig. 6) as well as CP decay-accelerating activity (Fig. 7A). Thus, like other vCCPs, the effective classical pathway inhibitory activity is also conserved in HVS CCPH.
It is clear from the data presented in Fig. 2 that sCCPH is significantly more active than VCP in inhibiting the alternative pathway-mediated deposition of C3b onto erythrocytes. To define the mechanism responsible for this increased activity, we characterized its factor I cofactor activity for C3b and decayaccelerating activity for the AP C3-convertase. The data revealed that sCCPH possesses 14-fold more C3b cofactor activity compared with VCP (Fig. 5). In fact, the cofactor activity was only about 2.5-fold less compared with human complement regulators factor H and sCR1 (Fig. 5). We would like to point out here that the observed difference in the cofactor activ- ity of sCCPH compared with VCP was not due to the difference in affinity for C3b as sCCPH showed lower affinity for C3b compared with VCP (Table 1). This, however, is not surprising as previously it has been shown that CD46, which has a much lower affinity for C3b than CR1, has a higher cofactor activity than CR1 (52). It is likely that the increased cofactor activity of sCCPH is a result of its better interaction with factor I. Analysis of AP C3-convertase decay-accelerating activity showed that sCCPH is a poor decay accelerator of AP C3-convertase (Fig. 7B). Although its activity was 8-fold better compared with VCP, it was Ͼ2000-fold less active compared with factor H and sCR1 (Fig.  7B). Together these data suggest that the increased alternative pathway inhibitory activity of sCCPH was primarily due to its increased factor I cofactor activity for C3b. Factor I is known to cleave C3b at three different positions depending on the cofactors involved: the first between 1281-1282, which generates iC3b 1 ; the second between 1298 -1299, which generates iC3b 2 ; the third between 932-933, which generates C3c and C3dg. Whether vCCPs support the cleavage of C3b to C3c and C3d/C3dg was under debate until recently (18,20), but it is now clear that viral regulators primarily support the cleavage of C3b to iC3b 1 (e.g. VCP (9, 16)) or iC3b 2 (e.g. Kaposica (19,22), SPICE (9,41), and MOPICE (9)) and not C3c and C3dg. Like most of other viral regulators, the sCCPH also supported the generation of C3b to iC3b 2 (Figs. 4 and 5). Because generation of iC3b 1 itself is sufficient to inactivate C3b (16), it is not clear whether inactivation of C3b to iC3b 2 as opposed to iC3b 1 provides any functional advantage to viruses.
Because sCCPH showed about a 14-fold higher cofactor activity for C3b, we sought to examine the basis for this increased activity. Earlier, Rosengard et al. (18) demonstrated that SPICE is about 100-fold more potent than VCP in inactivating C3b. Later, using the site-directed mutagenesis approach, it was established that Lys-108 and -120 residues are principally responsible for better functioning of SPICE (41). When we aligned sCCPH sequence with SPICE to determine whether sCCPH contains positively charged residues at the corresponding positions, we found that sCCPH contains Leu (Leu-106) and Arg (Arg-118) at these positions. Based on these, we predicted that Arg-118 might be responsible for the higher cofactor activity of sCCPH. Modeling of the sCCPH structure based on the crystal structure of VCP demonstrated that the side chain of Arg-118 is exposed to solvent (Fig. 8), which further supported this possibility. Thus, we mutated the Arg-118 to Ala and examined its functional activities. We found that removal of charge at this position drastically affected the factor I cofactor activity for C3b (50-fold decrease) and to some extent C4b (12-fold decrease) but had no effect on the decay-accelerating activities (Figs. 9 and 10). These data along with the previous studies on SPICE clearly point out that the presence of a positive charge at this position enhances the C3b cofactor activity in viral homologs. It is interesting to note that despite FIGURE 7. Decay acceleration of the classical and alternative pathway C3-convertase by sCCPH, VCP, factor H, and sCR1. A, the CP C3-convertase (C4b,2a) was formed on sheep erythrocytes and incubated with various amounts of sCCPH or the control proteins (VCP, factor H, and sCR1) for 5 min at 22°C in DGVB 2ϩ . The remaining C3-convertase activity was measured by hemolysis after the addition of 1:100 diluted guinea pig serum containing 20 mM EDTA. B, the AP C3-convertase (C3b,Bb) was formed on sheep erythrocytes and incubated with sCCPH or the control proteins (VCP, factor H, and sCR1) for 10 min at 37°C. The remaining C3-convertase activity was measured by hemolysis after the addition of 1:10 diluted human sera containing 20 mM EDTA. The data were normalized by setting 100% C3-convertase activity to be equal to the average activity in the absence of inhibitor. belonging to two different viral families, substitution of positively charged residue at comparable positions has been seen in SPICE as well as sCCPH (Lys-120 in SPICE and Arg-118 in sCCPH). Although a previous study on SPICE (41) examined the role of Lys-120 in enhancing C3b cofactor activity, it did not look at its role either in C4b cofactor activity or in decay-accelerating activities. Based on our data, we suggest that Lys-120 of SPICE might also play a role in enhancing its cofactor activity for C4b.
In summary, our data clearly show that HVS CCPH possesses all the complement regulatory activities present in Kaposica and other viral regulators. Thus, it seems that despite signifi-cant sequence differences between herpesviral complement regulators, the functional activities have been conserved. These data along with previous observations, therefore, point out that maintenance of various complement regulatory functions must be important to the pox as well as herpesviruses and inhibition of the lectin/classical pathway is crucial to viral survival than the inhibition of alternative pathway. Whether sequence variations in herpesviral complement proteins have resulted in acquisition of any new functions is not clear at present and requires further studies. Previously, it has been demonstrated that CCP homolog of ␥HV-68 plays an important role in complement evasion in vivo (53). Given the fact that sCCPH is an efficient complement inactivator, it is likely that sCCPH may also act as an immune evasion molecule in vivo and protect HVS from the host complement during infection. FIGURE 9. Time course of factor I cofactor activity of R118A mutant for complement proteins C3b and C4b. Cofactor activity was measured by incubating C3b or C4b with factor I and R118A mutant at 37°C for the indicated time period. The reactions were stopped by adding the sample buffer containing dithiothreitol, and the cleavage products were visualized by separating the samples on SDS-PAGE gels and staining with Coomassie Blue. The intensities of the ␣Ј chain were determined by densitometric analysis and represented graphically (lower panel). FIGURE 10. Comparison of the classical and alternative pathway C3-convertase decay-accelerating activity of sCCPH and R118A mutant. Sheep erythrocytes coated with the classical or alternative pathway C3-convertases were incubated with sCCPH or the R118A mutant, and the remaining enzyme activity was measured by incubating the cells with EDTA-sera as described under "Experimental Procedures."