| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Pathology, Case Western Reserve University, Cleveland, OH 44106; and
Division of Rheumatology, Department of Medicine, Washington University School of Medicine, St. Louis, MO 63110
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
60-aa-long consensus sequences termed complement control
protein repeats (CCPs). The four CCPs that comprise the
functional portion of the protein are suspended on an
oligosaccharide-rich cushion that in turn is linked to the plasma
membrane by a GPI anchor. Initial mapping of CCP domains of DAF in
cytotoxicity assays with whole serum showed that its activity resides
within CCP2–4 (1). More recent work with assays using
purified complement components has shown that CCPs 2 and 3 are required
for its ability to decay accelerate both the classical and alternative
pathway C3 convertases, C4b2a, and C3bBb
whereas CCP4 is additionally necessary for its alternative pathway
function (2). We previously built a three-dimensional structure of the CCP1–4 of DAF (3) based on a nuclear magnetic resonance (NMR)-generated solution structure of CCPs 15 and 16 of serum C3 convertase regulator factor H (4). A positively charged surface area on CCP2 extending into CCP3 (including positively charged R100, R101, K125, K126, and K127), and exposed hydrophobic moieties in CCPs 2 and 3 (including V121 and F123 in CCP2, and L147, F148, and L171 in CCP3) were proposed as being the primary C3 convertase interactive sites of DAF. In accordance with this, we showed in an earlier study that mutation of the three tandem lysines KKK125–127 in the CCP2–3 linker to three tandem threonines severely impaired the alternative pathway function of DAF, whereas mutation of L147F148 to two consecutive serines impaired the activity of DAF in both pathways. Through mutation of N61 to Q, eliminating the N-linked glycan between CCPs 1 and 2, we also established that in the intact protein this glycan plays no role in the regulatory activity of DAF (5).
Other recent studies have supported the functional importance of the above cited residues. Decay-accelerating activity (DAA) of complement receptor 1 (CR1, the C3b receptor) is abrogated in both pathways by substituting V for F82 (6), the equivalent alignment position to F148 in DAF. Substitution of Q for R39 and K63 in C4 binding protein (C4BP), the equivalent alignment positions to R101 and K127 in DAF, respectively, resulted in a decrease in DAA on the classical pathway C3 convertase (7).
To more precisely characterize the active sites of DAF, we prepared 31
primarily alanine substitution mutants based on 1) the original factor
H-based molecular model, 2) homologies among primate DAF proteins
(8, 9), and 3) sequence comparisons with other human C3
convertase regulators and the vaccinia virus complement-control protein
(VCP), all of which are CCP-based (10, 11, 12, 13, 14, 15, 16, 17, 18, 19). As new
structural information on CCP modules has become available since our
original modeling study, we placed our CCP2–3 functional data on the
crystallographic structure of CCP1–2 of membrane cofactor protein
(MCP) (20), another intrinsic regulator that has a
different inter-CCP orientation than factor H CCP15–16. In this
superimposition the functional residues aligned in a contiguous 26
Å long region of 10 aa, predicting that this region comprises part
of the active sites of DAF where the positively charged area of CCP2
and the hydrophobic area of CCP3 are joined by the positively charged
inter-CCP sequence that could contribute to their spatial
orientation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The DAF cDNA derivatives HuDAF-6H, HuDAF(N61Q), and RhDAF-6H (defined below) encoding the soluble DAF CCP1–4 products were prepared using 1) human DAF 13:2 cDNA in pBT-KS(+) (21), 2) a mutant form (N61Q) of (1) that eliminated the N-linked glycan between CCPs 1 and 2 (5), and 3) the cDNA encoding Rhesus macaque DAF from the cell line NCTC clone 3526 (8).
HuDAF-6H, encoding CCP1–4, N61, and a C-terminal 6XHis tag, was prepared by PCR of (1) using the 5' primer DSIG: 5'-ATA TAC GAA TTC ATG ACC GTC GCG CGG CCG AGC GTG-3' and the 3' primer DCP4RH: 5'-ACA GTG AGA TCT TTA GTG ATG GTG ATG GTG ATG TCC TCT GCA TTC AGG TGG TGG GCC-3'. HuDAF(N61Q), encoding CCP1–4, N61Q, without a C-terminal 6XHis tag, was prepared by PCR of (2) using the 5' primer DSIG and the 3' primer DCP4R: 5'-ACA GTG AGA TCT TTA TCC TCT GCA TTC AGG TGG TGG GCC-3'. RhDAF-6H, encoding CCP1–4 of Rhesus macaque DAF, N61, and a C-terminal 6XHis tag, was prepared by PCR of (3) using the 5' primer DSIGM: 5'-ATA TAC GAA TTC ATG ACT GTC GCG CGG CCG AGC GTG-3' and 3' primer CP4RHM: 5'-ACA GTG AGA TCT TTA GTG ATG GTG ATG GTG ATG TGC TCT ACA TGC AGG TGG TGG GCC-3'. Each of these PCR fragments was subcloned into the EcoRI-BglII sites of pSG5 (Stratagene, La Jolla, CA) and sequenced in full.
Substitution mutants
The QuikChange site-directed mutagenesis kit (Stratagene) was
used to generate the substitution mutants in HuDAF(N61Q) or
HuDAF-6H. Most substitution mutants were made in
HuDAF(N61Q). Those that were made in HuDAF-6H are
distinguished by containing a "-6H" after the designated mutation.
Sense primers are listed in Table I
. All
mutations were confirmed by sequencing. Additionally, CCP1 through CCP4
was sequenced in full for the following mutants: R69A, R96A, R100A,
K126A, K127A, F148A, F169A, and L171A.
|
Six to 18 µg of DNA, preincubated for 20–30 min at 20°C with 54 µl of lipofectamine (Life Technologies, Gaithersburg, MD) in 1.8 ml of OptiMEM I (Life Technologies), was mixed with 7.2 ml of OptiMEM I and added to COS1 cells grown to near confluence in Falcon 3084 flasks (BD Biosciences, San Jose, CA). After 5 h at 37°C, 30 ml of DMEM (10% FBS, 1% penicillin/streptomycin) was added, and the cells were incubated overnight. The following day, the cells were washed twice with PBS (Dulbecco’s), and 30 ml OptiMEM I was added. The OptiMEM I supernatant containing the recombinant protein was harvested 2 days later.
Supernatants were concentrated and buffer exchanged into Veronal-buffered saline (145 mM NaCl, 3.12 mM barbital, and 1.82 mM sodium barbital, pH 7.3–7.4) using a Millipore (Bedford, MA) ultrafree-15 centrifugal filter device, biomax-5K or -10K. A two-site immunoradiometric assay (22) and densitometry (OS-Scan Image Analysis System, serial no. 0306-371-10001, release no. 4.20; Oberlin Scientific, Oberlin, OH; Sci Scan 5000 densitometer; U.S. Biochemical, Cleveland, OH) were used to quantitate DAF protein concentrations.
The two-site immunoradiometric assay primarily used CCP1-specific mAb IA10 at 5 µg/ml as capture mAb and 125I-labeled CCP4-specific mAb 2H6 for detection. For certain CCP4 mutants, BRIC216 (BioSource International, Camarillo, CA) against CCP3 was used as capture mAb with 125I-labeled IA10 for detection.
As a second method for quantitating mutants without a 6H tag, HRP-conjugated IA10 was prepared using EZ-link plus activated peroxidase and kit (Pierce, Rockford, IL) and used in densitometry. For CCP4 mutants and Rhesus macaque DAF with 6H tags, an HRP-conjugated anti-6H (C-terminal) mAb (mouse, clone 3D5, IgG2b; Invitrogen, Carlsbad, CA) was used in conjunction with densitometry.
Hemolytic assay buffers
Isotonic Veronal-buffered saline (DGVB2+) contained 72.7 mM NaCl, 1.56 mM barbital (Fisher Scientific, Fair Lawn, NJ), 0.91 mM sodium barbital (Fisher Scientific), l mM MgCl2, 0.15 mM CaCl2, and 2.5% (w/v) dextrose (pH 7.3–7.4) to which 0.1% gelatin was added. Metal-chelating Dextrose gelatin Veronal buffer with EDTA (DGVB-E) substituted 10 mM EDTA for MgCl2 and CaCl2 in DGVB2+. Isoionic Veronal buffer (GVB2+) consisted of 145 mM NaCl, 3.12 mM barbital, 1.82 mM sodium barbital, l mM MgCl2, and 0.15 mM CaCl2 (pH 7.3–7.4) to which 0.1% gelatin was added. Metal-chelating Veronal buffer (GVB-E) substituted 10 mM EDTA for MgCl2 and CaCl2 of GVB2+.
DAA assays
Alternative pathway C3 convertase DAA was determined by ELISA (23). Microtiter wells were coated overnight at 4°C with 1 µg/ml C3b and blocked for 1 h at 37°C with PBS supplemented with 1% BSA and 0.1% Tween 20. Plates were stored at 4°C until needed. Microtiter wells were then incubated at 37°C for 2 h with 400 ng/ml factor B, 25 ng/ml factor D, 2 mM NiCl2, 25 mM NaCl in 10 mM phosphate buffer pH 7.4 supplemented with 4% BSA, and 0.1% Tween 20. The wells were washed extensively, and then the plate-bound C3bBb(Ni2+) complexes were incubated for 15 additional min at 37°C with varying concentrations of mutant or control DAF proteins in 25 mM NaCl-supplemented phosphate buffer, or with 25 mM NaCl-supplemented phosphate buffer alone. Remaining plate-bound C3bBb(Ni2+) complexes were detected with goat anti-factor B polyclonal Ab followed by peroxidase-conjugated rabbit anti-goat polyclonal Ab (23). In each experiment, regression analysis was used to form dose-response curves for mutant and control DAF protein; from the curves mutant activity was calculated as a percentage of the activity of the wild-type protein.
Classical pathway C3 convertase DAA was determined using a modification
of a hemolytic C4b2a decay assay (24). In this
modified assay, sensitized sheep erythrocytes (2.3 x
107 in 100 µl) were incubated at 30°C for 15
min, 20 min, and 5 min sequentially with 15 site
forming units (SFU) of guinea pig C1, 5 SFU before decay of human C4
(Quidel, Mountain View, CA), and sufficient human C2 (Advanced Research
Technologies, San Diego, CA) to yield 1 C4b2a site after
decay in DGVB2+ for 15 min at 30°C. Wild-type
DAF, mutated DAF, or DGVB2+ control was added to
the cells during this decay step. Guinea pig serum (C3-9) in GVB-E was
added for l h at 37°C to develop lysis. After pelleting of unlysed
cells and measuring the OD412 of the supernatant,
residual C4b2a sites were calculated.
Classical pathway C5 convertase DAA was determined using a hemolytic
C4b2a3b decay assay (24). For this assay,
sensitized sheep erythrocytes (1 x 107 in
100 µl) were incubated at 30°C for 15 min and 20 min sequentially
with 30 SFU of guinea pig C1 (24) and excess human C4
(determined by titration) (Quidel). Excess human C2 (determined by
titration) (Advanced Research Technologies) and sufficient human C3
(25) were added concurrently for 5 min at 30°C to yield
1 C4b2a3b site after decay in
DGVB2+ for 15 min at 30°C. DAF, mutated DAF, or
DGVB2+ control was added to the cells during this
decay step. Subsequently, a 1/100 dilution of human C5 (1 mg/ml;
Quidel) in DGVB2+ and a 1/150 dilution of C6-9
reagent (guinea pig serum treated with potassium thiocyanate and
hydrazine) (25) in DGVB-E were added for 5 min at 30°C
and l h at 37°C, respectively, to develop lysis. Hemoglobin color was
quantitated, and residual C4b2a3b sites were calculated as
in the classical pathway C3 convertase assay.
Calculations and standards
In each assay, a dose-response curve of percent activity vs
concentration was established for mutants and controls, and the
concentration of DAF protein required for 50% activity was determined.
The concentrations obtained for mutants were compared with the
concentrations of their corresponding controls, and percent activity
was calculated, the control being 100%. Mutants derived directly from
HuDAF(N61Q) were compared to that parental control.
Mutants derived from HuDAF-6H were compared to that
6H-tagged DAF parental protein. Two different lots of HuDAF(N61Q)
standard varied in activity by 10%, and the activity level of the
HuDAF-6H standard was 85% of the HuDAF(N61Q) standards.
For classical pathway C3 convertase assays, 2–3 independent experiments were performed. For alternative pathway C3 convertase assays, 2–5 assays were performed. For classical C5 convertase hemolytic assays, a minimum of two independent experiments were performed, except for mutants K126A, F169A, and R212A-6H, which were tested once relative to HuDAF(N61Q) or HuDAF-6H.
In the case of the double mutant F123A, R101I, there could be some overstatement of DAA in the classical pathway C3 convertase assay because the production of mutant protein was very low and large quantities, i.e., a lesser supernatant dilution, had to be used. However, this mutant exhibited reduced function (25%) relative to the control HuDAF(N61Q) construct. Therefore, this possible effect would not change the conclusions in any considerable way.
Modeling
The molecular modeling of DAF CCP1–4 based on the NMR-determined solution structure of factor H CCP15–16 (4) was conducted in an earlier paper (3). The modeling involved the use of program O (26) for residue changes, residue insertions, and connection of CCPs, whereas the program LSQKAB within the CCP4 suite of crystallographic programs (27) was used for superpositioning of DAF CCPs onto the factor H structure. The coordinate file generated in that study was read by RasMol V2.7.1 (Bernstein + Sons, Bellport, NY) for visual display here.
The MCP-based three-dimensional model of DAF CCP2–3 was constructed using a commercially available protein structure software package (InsightII graphics with Homology and Modeler modules; Molecular Simulations, San Diego, CA) installed on an Octane workstation (Silicon Graphics, Mountain View, CA). The DAF CCP2–3 and MCP CCP1–2 primary sequences were aligned by the Homology module, with adjustments for the most uniformly conserved CCP residues (28). The alignment was used by the Modeler module, an automated modeling program of Sali and Blundell (29), to construct a three-dimensional model of DAF CCP2–3 based on the three-dimensional coordinates of MCP CCP1–2 (20).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As indicated in the introduction, a panel of DAF mutants was
generated to map the sites in DAF important for its activity based in
part on the previously derived factor H-based DAF model. Table II lists by CCP the soluble (CCP1–4) DAF
mutants studied and their efficiencies relative to the soluble DAF
standards, HuDAF(N61Q) and HuDAF-6H, in accelerating the decay of the
alternative and classical pathway C3 convertases. Fig. 1A shows the positions of the
mutated residues on the factor H-based DAF model (3). In
CCP2, one of the eight mutations, R96A, abolished DAA (<1%) and two
others, R69A and R100A, markedly reduced DAA (
12%) in the
alternative pathway. Two of these three basic amino acid substitutions,
R96A and R69A, equivalently reduced the function of DAF in the
classical pathway, whereas the third, R100A, had a 2-fold less
negative effect (DAA 25%). Of the other mutations, F123A did not
significantly affect (or only slightly affected) the function of DAF in
both pathways (DAA 82 and 60%), whereas the double substitution mutant
F123A, R101I had a much greater negative effect (DAA 16 and 25%). Two
other substitutions, L70A and N71K, similarly had negative effects on
DAF’s function in the alternative pathway (DAA 25 and 29%,
respectively) while not affecting its classical pathway C3 convertase
function (DAA 98 and 83%, respectively).
|
|
14%) DAF’s alternative pathway C3 convertase
function. K127A had a similar effect on DAF’s classical pathway C3
convertase function, whereas surprisingly, K126A had no effect (DAA
122%). In contrast, K125A, although resembling K126A in its
differential effect on the two pathways, had only a moderately negative
effect (DAA 32%) on DAF’s activity in the alternative pathway while
not substantially altering its classical pathway function (62%).
In CCP3, elimination of any one of the three hydrophobic residues,
F148A, F169A, and L171A, localized near the N-terminal side of CCP3
according to the factor H model, totally or nearly abolished the
alternative pathway DAA of DAF (4%). F148A had a similar markedly
negative impact (DAA 4%) against the classical pathway C3 convertase,
and L171A greatly diminished function (DAA 16%). In contrast, F169A
did not substantially alter function (DAA 64%). Y160A, localized near
the C-terminal side of CCP3 according to the model, greatly reduced
DAF’s alternative pathway function (DAA 20%), whereas it did not
affect classical pathway function (DAA 126%). Mutations L147A, F154A,
S165A, S176A, and V177A had varying, sometimes moderately negative
effects on DAF’s function on one or the other C3 convertase enzymes.
In contrast, mutations S155A and F163A had either no or a slightly
positive impact on DAF’s function in both pathways.
In the linker between CCPs 3 and 4 and in CCP4 itself, three of seven mutations had negative effects on DAF’s alternative pathway function. Of particular note, R206A and R212A greatly reduced DAF’s alternative pathway activity (DAA 25%). In marked contrast, these and the other CCP4 mutations had no greatly adverse effect on DAF’s classical pathway activity (DAA ranged from 56 to 128%).
Critical amino acids for DAF’s activity toward the classical pathway C5 convertase
One way in which the classical and alternative pathway C3
convertases, C4b2a and C3bBb, differ is that
C3b is present in the latter but not the former. Thus it is possible
that DAF sites that are involved in decay acceleration of
C3bBb but not C4b2a interact with C3b. Because
C3b is present in the classical pathway C5 convertase
C4b2a3b, which is susceptible to DAF decay, those DAF
mutants whose alternative pathway function was markedly impaired
relative to classical pathway function (Table II
), were tested for
activity against this enzyme (Table III).
The question at hand was whether the addition of C3b, an alternative
pathway component, to C4b2a would uncover functional
differences at this later step in the complement cascade. The results
were clear; mutations K126A, F169A, and R212A had the same effects on
C4b2a3b decay as those observed on C4b2a decay.
Likewise, the substitutions N71K and R206A showed no significant
differences.
|
Fig. 1B shows the positions on the factor H-based model
of the 28-aa differences between human and Rhesus macaque DAF CCP2–4.
Twenty-three of these differences do not coincide with any of the above
mutated residues, and 27 do not correspond to the residues found to be
functionally important (DAA 16% in either human pathway) (Fig. 2B). RhDAF-6H exhibited
nearly full function against human C4b2a (DAA 92%) but
showed decreased function against human C3bBb (DAA 28%).
Similar to the above mutant proteins, comparable effects were observed
on C4b2a and C4b2a3b decay.
|
In view of the recent availability of the crystal structure of MCP
CCPs 1 and 2 (20), as an alternative to the factor H-based
model, a molecular model of DAF CCP2–3 was produced based on this
structure. Fig. 3, A and
B, places the measured alternative and classical pathway
functional effects of the mutations on the MCP-based structure. When
portrayed in this way, components of DAF’s active sites encompass
10 residues that span 26 Å. In contrast to results obtained when
the data are modeled on factor H CCPs 15 and 16 (Fig. 4, A and B) where
residues found to be most essential, e.g., R96, do not
localize near each other (see Discussion), all of the
residues found to be functionally important primarily align onto a
contiguous region on the same side of the protein. A second important
observation is that the active sites against the alternative pathway
convertase cover a larger surface area than those against the classical
pathway convertase.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our first two structure-function studies of DAF, we prepared recombinant GPI-anchored DAF mutant proteins and exploited their abilities to incorporate into defined hemolytic intermediates so as to dissect the roles of specific amino acids on function against the classical and alternative convertases. Studies with CCP deletion mutants using this system first refined the initial observation that the activity of DAF resides in CCP2–4 (1) by showing that the sites that provide for classical and alternative pathway function differ. Subsequent studies using multiple substitution mutagenesis revealed that KKK125–127 and L147F148 were important for DAA of DAF in both pathways although differential effects were observed (5). This methodology was initially adopted because DAF functions on autologous convertases assembled on the same cell membrane, and its in situ physiological relationship to its ligands could in principle play a role in its function. Early studies had shown that a spacer, although nonspecific, was required between the CCP domains and the plasma membrane for function (1). The finding that fluid phase assays performed as part of our initial investigations yielded equivalent results excluded this possibility and thus constituted an important byproduct of the work.
In this report, we used a fluid phase system primarily using alanine scanning mutagenesis to determine the sites in DAF that contribute to its activity on the two C3 convertases. The premise behind alanine substitution is that it could allow determination of the contribution of side chains, although it cannot answer which property of the side chain is functionally important and whether the properties of the side chain contribute to ligand recognition or to molecular stability (30). However, data from 31 substitution mutants and the "natural variant" Rhesus macaque DAF, which has 28 aa differences from human DAF in CCPs 2, 3, and 4, provide a preliminary picture.
Our original molecular model based on factor H CCP15–16 proposed that a positively charged surface area on CCP2 extending into CCP3 is a primary recognition site in DAF for the convertases. This seems to be confirmed by our data. Mutation of positively charged R69, R96, and R100 in CCP2 and K127 in the linker between CCPs 2 and 3 to A markedly impaired the ability of DAF to decay-accelerate the C3 convertases of both pathways. Interestingly, an analysis of alanine mutagenesis studies has concluded that protein-protein binding sites are enriched with W, R, and Y (31). Another study has also implicated R as one of several polar residues that are conserved at such sites (32). It has been noted that, due to their size and composition, R and K have both charged and hydrophobic properties (33). In the case of these residues, both properties may contribute to the fit of the convertase interactive sites.
From the above, some speculations are possible concerning our data. In our experiments, R69 was found to be critically important to DAF’s activity on both C3 convertase enzymes. R69 sits next to L70, which was found to be important for DAF’s alternative pathway function. This proximity raises the possibility that the hydrophobic components of the two side chains interact to augment DAF’s function on the alternative pathway convertase. In another example, R100 sits next to R101, both of which were found to be important for DAF’s function in both pathways. As mutation of R101 to hydrophobic I (in the double mutant F123A, R101I) significantly impaired function, the charged property of R101 may be its more important feature in the ability of DAF to decay accelerate. As alternative explanations for the results exist, additional studies will be required to confirm these proposals.
The findings that K125 is important and K126
and K127 are critical for DAF’s alternative pathway
function whereas K125 and K126 are less
important for DAF’s classical pathway function are consistent with
previous mutagenesis experiments that showed that mutation of
KKK125–127
TTT (three tandem threonines) abolished
DAF’s alternative pathway function and inhibited, but did not abolish,
DAF’s classical pathway function (5).
The factor H-based molecular model also proposed that hydrophobic moieties in CCPs 2 and 3 contributed to DAF’s convertase recognition site. This was borne out in the present study as mutation of F148 and L171 in CCP3 severely impaired the ability of DAF to decay both C3 convertases.
However, the above formulation provides only a partial explanation in that although we proposed that the positive surface in CCP2 is critical to DAF’s function in both pathways, in the case of alternative pathway function, other residues with different properties were found to be important. Changes of L70A and N71K adversely affected alternative pathway function (25 and 29% active, respectively). It is possible that these changes could cause neighboring residues to orient differently than they would in the native protein. The alteration of L70A could create more space or take away a hydrophobic interaction, whereas the change of N71K could cause excessive repulsion among nearby positive charges. It is noteworthy that the N71K substitution changes human DAF residue 71 to its Rhesus macaque counterpart, and that Rhesus macaque DAF is 28% functional against the human alternative pathway C3 convertase.
Sequence determination of the VCP has shown that it contains an
eight-amino-acid segment that is repeated in its CCPs 2 and 3
(18). This repetitive segment exhibits variable homology
with sequences in other C3/C4 regulators (Fig. 2A). Although
we did not systematically mutate the parallel sequences in DAF, some
insight concerning DAF’s functional sites may be gained from our data.
Alanine substitutions in CCP3 of DAF within and near the homologous
region to the VCP repeat from F154 to K161 were
found to have little, if any, negative effect on function in either
pathway with the exception of Y160A, which impaired DAF’s alternative
pathway function. The alanine substitution, N220A, in CCP4 of DAF
within the region homologous to the VCP repeat had a minimal effect on
classical pathway function but moderately impaired function in the
alternative pathway (DAA 32%). Human and Rhesus DAF share the same
eight amino acids in CCP3 in the region homologous to the VCP repeat.
However, Rhesus DAF differs from human DAF at two positions (A218L and
K221R) of the eight amino acids in CCP4 in the region homologous to the
VCP repeat (Fig. 2B). It appears from these findings that if
the amino acids in DAF corresponding to those within the VCP repeat
regions are important in DAF’s functionality against the C3
convertases, their effects are greater on the alternative pathway
enzyme C3bBb. Alternatively, it is possible that the VCP
repeat contributes to some other function not shared with DAF, such as
cofactor activity for factor I cleavage of C4b and C3b.
Some insights also may be gained with respect to turns and bends
predicted by the factor H-based model, as intra-CCP structure should be
fairly independent of inter-CCP orientation (34) (Fig. 1A). F148, Y160, V177,
and R212 occur in bends and S72 and
D238 occur in hydrogen-bonded turns (data not shown).
Mutation F148A essentially abolished DAF’s DAA in both pathways (DAA
4%), and mutations Y160A and R212A markedly impaired function in
the alternative pathway (DAA 20 and 25%, respectively). The
substitutions of V177A and D238A moderately impaired function in both
pathways. These findings indicate that the decrease in DAA could
reflect changes in bends or turns in addition to properties of the
residues themselves.
Our data showed that the recognition site for C3bBb extends beyond that for C4b2a. Mutation of F169 in CCP3 impaired DAF’s alternative pathway function while having little or no negative effect on the classical pathway function of DAF. Also striking are R206A and R212A in CCP4, each of which reduced alternative pathway function to 25% while leaving classical pathway function unaffected. As indicated above, the fact that DAF’s classical pathway function was not appreciably affected by these one-amino-acid substitutions gives a degree of confidence about the structural integrity of these altered proteins. The difference in the breadth of the recognition sites on human DAF for the alternative pathway and classical pathway convertases seen in these mutants may relate to recent studies of mouse, rat, and pig DAF. The DAFs of these species are active against classical pathway human complement but not against alternative pathway human complement (35, 36). Our data on recombinant Rhesus DAF are similar. Although Rhesus DAF was almost fully functional (92%) against the human classical pathway C3 convertase, it was 28% functional against the human alternative pathway C3 convertase.
DAA of DAF against the classical pathway C5 convertase
The classical pathway C5 convertase, in addition to C4b2a, contains the alternative pathway C3 convertase component C3b. It has been shown that in the C5 convertase C4b2a3b, C3b is covalently bound to C4b (37). Several of the DAF mutants exhibited marked differences in their ability to accelerate decay of C3bBb and C4b2a; generally, DAA against the alternative pathway C3 convertase suffered more than DAA against the classical pathway C3 convertase. Additionally, the recognition site for C3bBb was found to be broader than for C4b2a. As it is possible that the alternative pathway-specific sites of DAF might interact with C3b and also be able to interact with the C3b subunit of the classical pathway C5 convertase, we compared mutants that differentially affected C3bBb and C4b2a decay for their effects on the classical pathway C5 convertase. Mutations K126A, F169A, and R212A showed no differences in their effects on C4b2a and C4b2a3b regulation relative to the DAF control. Likewise, mutations N71K and R206A showed no significant differences. These data indicate that DAF interacts similarly with the classical pathway C3 and C5 convertases. By comparison, in CR1, CCP1–3 in long homologous repeat (LHR) 1 (i.e., CCP1–7) provides for decay acceleration of the C3 convertases, whereas the first three CCPs in LHR2 (CCP8–14) or LHR3 (CCP15–21) are also required for decay acceleration of the C5 convertases, pointing to involvement of an additional contact site on the (second) C3b component of the trimolecular C5 convertase (6).
Spatial orientation of CCPs
As indicated, since the report of our DAF model based on CCP15–16 of factor H, two modules each of VCP and MCP have been structurally determined by solution NMR or crystallography. An important finding arising from the analyses of these modules is that the orientation between the two CCPs in each is different.
The factor H-based model of DAF predicted that F123 would
be involved in function. However, the substitution of F123A did not
markedly affect function in either pathway. Moreover, some of the
residues we found to be the most functionally important in DAF align on
different sides of the factor H-based model. F148,
F169, and L171 are found on one side, whereas
R96 is found on the other side (Fig. 4, A and
B). When the active residues are spatially oriented on
DAF’s CCPs 2 and 3 according to the angles of MCP’s CCPs 1 and 2 as
determined by its crystal structure (20), they form a
contiguous 26 Å long region of 10 aa (Fig. 3, A and
B). Moreover, the crystal structure of MCP CCP1–2 showed
interdomain contacts between residues that correspond to DAF residues
L147-K126-F148-I172 and
E102-K125, raising the possibility that alanine
substitution of K126, F148, or K125
could perturb to a degree the spatial orientation between CCP modules.
However, as noted earlier, K126A had no effect, and K125A and L147A had
only a small negative effect on DAF’s DAA against the classical
pathway C3 convertase. A recent phylogenetic tree of regulators of
complement activation proteins suggests that DAF is more closely
related to MCP than to factor H (38). Moreover, a common
origin of DAF CCP3 with MCP CCP2 is supported by the fact that both
have an unusual "split" exon structure (39, 40).
Additionally, sequence comparison of DAF and MCP (Fig. 2A)
indicates that DAF’s CCP3 may resemble MCP’s CCP2 in the loop region
between 
strands D and E. In MCP, this region leads in part to the
opposite direction bend between MCP’s CCPs 1 and 2 (20)
than is seen for factor H’s CCPs 15 and 16. Although definitive
assignment of the spatial relationship between CCPs 2 and 3 of DAF must
await its structural analysis, the functional data argue that it may be
closer to that of MCP CCPs 1 and 2 and that this spatial relationship
could be an important element in the functional sites of DAF.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. M. Edward Medof, Institute of Pathology, Case Western Reserve University School of Medicine, 2085 Adelbert Road, Room 301, Cleveland, OH 44106. E-mail address: mxm16{at}po.cwru.edu
3 Abbreviations used in this paper: DAF, decay-accelerating factor; CCPs, complement control protein repeats; CR1, complement receptor 1; C4BP, C4 binding protein; MCP, membrane cofactor protein; DGVB2+, isotonic Veronal buffered saline; HuDAF(N61Q), human DAF CCP1–4 with N61Q; HuDAF-6H, human DAF CCP1–4 with C-terminal 6XHis tag; DAA, decay-accelerating activity; RhDAF-6H, Rhesus macaque DAF CCP1–4 with C-terminal 6XHis tag; VCP, vaccinia virus complement control protein; LHR, long homologous repeat; NMR, nuclear magnetic resonance; SFU, site forming units; GVB2+, isoionic Veronal buffer.
Received for publication March 8, 2001. Accepted for publication June 1, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain CCP1 and CCP2 of C4BP are required for regulation of the classical C3-convertase. Mol. Immunol. 37:445.[Medline]
2-glycoprotein I, and the Ba fragment of B2. J. Immunol. 136:3407.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  H. G. Hocking, A. P. Herbert, D. Kavanagh, D. C. Soares, V. P. Ferreira, M. K. Pangburn, D. Uhrin, and P. N. Barlow Structure of the N-terminal Region of Complement Factor H and Conformational Implications of Disease-linked Sequence Variations J. Biol. Chem., April 4, 2008; 283(14): 9475 - 9487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hafenstein, V. D. Bowman, P. R. Chipman, C. M. B. Kelly, F. Lin, M. E. Medof, and M. G. Rossmann Interaction of Decay-Accelerating Factor with Coxsackievirus B3 J. Virol., December 1, 2007; 81(23): 12927 - 12935. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Kuttner-Kondo, D. E. Hourcade, V. E. Anderson, N. Muqim, L. Mitchell, D. C. Soares, P. N. Barlow, and M. E. Medof Structure-based Mapping of DAF Active Site Residues That Accelerate the Decay of C3 Convertases J. Biol. Chem., June 22, 2007; 282(25): 18552 - 18562. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Harris, D. M. Pettigrew, S. M. Lea, and B. P. Morgan Decay-Accelerating Factor Must Bind Both Components of the Complement Alternative Pathway C3 Convertase to Mediate Efficient Decay J. Immunol., January 1, 2007; 178(1): 352 - 359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Korotkova, I. Le Trong, R. Samudrala, K. Korotkov, C. P. Van Loy, A.-L. Bui, S. L. Moseley, and R. E. Stenkamp Crystal Structure and Mutational Analysis of the DaaE Adhesin of Escherichia coli J. Biol. Chem., August 4, 2006; 281(31): 22367 - 22377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Krych-Goldberg, R. E. Hauhart, T. Porzukowiak, and J. P. Atkinson Synergy between Two Active Sites of Human Complement Receptor Type 1 (CD35) in Complement Regulation: Implications for the Structure of the Classical Pathway C3 Convertase and Generation of More Potent Inhibitors J. Immunol., October 1, 2005; 175(7): 4528 - 4535. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. C. Soares, D. L. Gerloff, N. R. Syme, A. F.W. Coulson, J. Parkinson, and P. N. Barlow Large-scale modelling as a route to multiple surface comparisons of the CCP module family Protein Eng. Des. Sel., August 1, 2005; 18(8): 379 - 388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Servin Pathogenesis of Afa/Dr Diffusely Adhering Escherichia coli Clin. Microbiol. Rev., April 1, 2005; 18(2): 264 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Harris, R. J. M. Abbott, R. A. Smith, B. P. Morgan, and S. M. Lea Molecular Dissection of Interactions between Components of the Alternative Pathway of Complement and Decay Accelerating Factor (CD55) J. Biol. Chem., January 28, 2005; 280(4): 2569 - 2578. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Mark, W. H. Lee, O. B. Spiller, D. Proctor, D. J. Blackbourn, B. O. Villoutreix, and A. M. Blom The Kaposi's Sarcoma-associated Herpesvirus Complement Control Protein Mimics Human Molecular Mechanisms for Inhibition of the Complement System J. Biol. Chem., October 22, 2004; 279(43): 45093 - 45101. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Miyagawa, T. Kubo, K. Matsunami, T. Kusama, K. Beppu, H. Nozaki, T. Moritan, C. Ahn, J. Y. Kim, D. Fukuta, et al. Delta-Short Consensus Repeat 4-Decay Accelerating Factor (DAF: CD55) Inhibits Complement-Mediated Cytolysis but Not NK Cell-Mediated Cytolysis J. Immunol., September 15, 2004; 173(6): 3945 - 3952. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hudault, O. B. Spiller, B. P. Morgan, and A. L. Servin Human Diffusely Adhering Escherichia coli Expressing Afa/Dr Adhesins That Use Human CD55 (Decay-Accelerating Factor) as a Receptor Does Not Bind the Rodent and Pig Analogues of CD55 Infect. Immun., August 1, 2004; 72(8): 4859 - 4863. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Lukacik, P. Roversi, J. White, D. Esser, G. P. Smith, J. Billington, P. A. Williams, P. M. Rudd, M. R. Wormald, D. J. Harvey, et al. Complement regulation at the molecular level: The structure of decay-accelerating factor PNAS, February 3, 2004; 101(5): 1279 - 1284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A. Kuttner-Kondo, M. P. Dybvig, L. M. Mitchell, N. Muqim, J. P. Atkinson, M. E. Medof, and D. E. Hourcade A Corresponding Tyrosine Residue in the C2/Factor B Type A Domain Is a Hot Spot in the Decay Acceleration of the Complement C3 Convertases J. Biol. Chem., December 26, 2003; 278(52): 52386 - 52391. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Uhrinova, F. Lin, G. Ball, K. Bromek, D. Uhrin, M. E. Medof, and P. N. Barlow Solution structure of a functionally active fragment of decay-accelerating factor PNAS, April 15, 2003; 100(8): 4718 - 4723. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Williams, Y. Chaudhry, I. G. Goodfellow, J. Billington, R. Powell, O. B. Spiller, D. J. Evans, and S. Lea Mapping CD55 Function. THE STRUCTURE OF TWO PATHOGEN-BINDING DOMAINS AT 1.7 A J. Biol. Chem., March 14, 2003; 278(12): 10691 - 10696. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Hasan, E. Pawelczyk, P. T. Urvil, M. S. Venkatarajan, P. Goluszko, J. Kur, R. Selvarangan, S. Nowicki, W. A. Braun, and B. J. Nowicki Structure-Function Analysis of Decay-Accelerating Factor: Identification of Residues Important for Binding of the Escherichia coli Dr Adhesin and Complement Regulation Infect. Immun., August 1, 2002; 70(8): 4485 - 4493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Hourcade, L. Mitchell, L. A. Kuttner-Kondo, J. P. Atkinson, and M. E. Medof Decay-accelerating Factor (DAF), Complement Receptor 1 (CR1), and Factor H Dissociate the Complement AP C3 Convertase (C3bBb) via Sites on the Type A Domain of Bb J. Biol. Chem., January 4, 2002; 277(2): 1107 - 1112. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
