| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Rheumatology, Washington University School of Medicine, St. Louis, MO 63110
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most cells express MCP as a family of four isoforms generated by alternative splicing at two sites (Refs. 15 and 16; reviewed in 1 . The amino terminus is identical for all isoforms, consisting of four of the ligand-interacting modules termed complement control protein (CCP) repeats. Functional sites for C3b and C4b interactions have been mapped to CCP-2, -3, and -4 (17, 18). N-linked glycosylation occurs in CCP-1, -2, and -4 (19). Following the repeats is the alternatively spliced O-glycosylated region that is enriched in serines, threonines, and prolines (STP domain). A juxtamembranous segment of 12 amino acids of unknown function terminates the extracellular portion of MCP. Hydrophobic transmembrane and alternatively spliced cytoplasmic domains (Cyt-1 or Cyt-2) complete the carboxyl terminus. The four isoforms are termed BC1, BC2, C1, and C2 to denote their STP (i.e., BC or C) and cytoplasmic tail (one or two) content. Rarer isoforms have been described (16).
Although the binding site(s) for group A Streptococcus and Neisseria have not been mapped, CCP-1 and -2 are critical for MV binding and infection (18, 20, 21, 22, 23). Additionally, the N-glycan of CCP-2 is essential for MV binding and infection, whereas those of CCP-1 and -4 are of minor importance (21, 24, 25, 26). It has been suggested that, rather than contributing to specific binding, the N-glycan of CCP-2 maintains the conformational angle or reduces flexibility between CCP-1 and -2, thereby providing a more rigid docking site for MV hemagglutinin (21). Because of the above findings and evidence that isoforms with a smaller STP domain have a diminished ability to protect against complement attack (18, 27), we sought to examine the impact of these carbohydrate modifications on complement regulatory function by creating, expressing, and evaluating mutants deleted of these sites.
In the present study, we prepared stably transfected glycan deletion mutants in Chinese hamster ovary (CHO) cells. Initially, we utilized cell lysates of these mutants to assess C3b and C4b cofactor activity and ligand binding. Next, employing clones bearing equivalent expression levels of control or mutant MCP, we evaluated cytoprotection and C4b fragment cleavage on the cell surface. We determined that the N-glycans of CCP-2 and -4 and the STP domain were involved in MCP-mediated cytoprotection, a finding consistent with the evolutionary conservation of these carbohydrate sites.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Four mutants of the BC1 isoform were constructed. For three
constructs a glutamine (Q) replaced the asparagine (N) in CCP-1, -2,
and -4 and were termed NQ1, NQ2, and NQ4 (25). The fourth construct,
termed 
STP, deleted the STP domain by ligating two PCR-derived
segments of BC1 in the expression plasmid pHßApr1.neo (27). These two
segments coded for the signal peptide through the end of CCP-4
(nucleotide residues 42–1000) and the region coding for the
"undefined" segment through the end of the cytoplasmic tail
(nucleotides 1088–1518) (see GenBank MCP-BC1 cDNA, accession no.
X59405). All cDNA clones were sequenced in their entirety to verify
fidelity.
Transfection was performed per manufacturer’s directions using Lipofectin (Life Technologies, Gaithersburg, MD) in CHO K1 cells. Transfected cells were selected and maintained in Ham’s F-12 medium supplemented with 10% FCS and 0.5 mg/ml geneticin. CHO mutant clones were obtained by limiting dilution and were selected for expression levels comparable to wild-type clone BC1 (designated as 23-9, as described in 27 .
Flow cytometry
Flow cytometric analysis was performed as described (27). MCP mAbs utilized were as follows: TRA-2-10 (28), which binds to an epitope in CCP-1 (20); M75, which binds to an epitope in CCP-2 (provided by Tsukasa Seya, Osaka, Japan) (18); and GB24 (29), which binds to a complement functional epitope in CCP-3/4 (17). A rabbit polyclonal Ab to MCP was kindly provided by CytoMed (Cambridge, MA) (27).
Cell lysates and quantification
Cells were lysed (2 x 107 cells/ml) in 1% Nonidet P-40, 0.05% SDS in TBS (10 mM Tris, pH 7.2, 150 mM sodium chloride) with 2 mM PMSF for 15 min at 4°C, followed by centrifugation in a microcentrifuge at 12,000 x g for 10 min. Supernatants were collected and MCP quantified by ELISA as previously described (17), except as noted below. Briefly, MCP mAb TRA-2-10 was coated at 5 µg/ml in TBS overnight at 4°C in microtiter wells (Nunc modules, Fisher Scientific, St. Louis, MO). Wells were blocked for 1 h at 37°C with 1% BSA and 0.1% Tween-20 in TBS and rinsed in wash buffer (TBS with 0.05% Tween-20). Cell lysates were prepared over nearly a log of dilutions in dilution buffer (wash buffer with 4% BSA and 0.25% Nonidet P-40), applied to wells along with an MCP standard, incubated for 1 h at 37°C, and rinsed for 2 min three times with wash buffer. Next, rabbit anti-MCP antiserum was diluted 1:7000 in dilution buffer, applied for 1 h at 37°C, and similarly washed. Horseradish peroxidase-coupled donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) was incubated for 1 h at 37°C followed by similar washing. Detection was made using the substrate 3,3',5,5'-tetramethylbenzidine (TMB) (Pierce, Rockford, IL). Optical density (630 nm) was measured in an ELISA reader.
Western blot analysis
For Western blot analysis, cell lysates of CHO transfectants (200,000 cell equivalents/lane) were analyzed on a 10% SDS-PAGE and transferred to nitrocellulose as previously described (27).
Confocal microscopy
Cells were treated identically as described for flow cytometry
(27) with the exceptions noted below. Briefly, cells were grown to
70% confluency, collected by brief trypsinization, washed in PBS,
resuspended at 5 x 106 cells/ml in 1% FCS in TBS at
4°C, and 100 µl was placed in wells of a V-bottom microtiter plate.
MCP mAb TRA-2-10 IgG was added at 5 µg/ml for 30 min at 4°C. Cells
washed in 1% FCS, and FITC-conjugated F(ab')2 donkey
anti-mouse IgG (Jackson ImmunoResearch) was added for 30 min at
4°C. Following washing, cells were fixed in 0.5% paraformaldehyde,
collected by cytospin onto microscope slides, and mounted. Fluorescence
images were collected with a x63 oil objective (NA 1.4) using a
Bio-Rad (Hercules, CA) MRC 1024 confocal laser-scanning adaptor
attached to a Zeiss Axoplan upright microscope. The 488-nm lines of an
argon-krypton laser were used for excitation, and the emitted light was
filtered through a 522-nm long-pass filter cube and detected by
respective photomultipliers. Images, each of which is 512 x 512
pixels, were recorded with Bio-Rad Lasersharp software. An image was
averaged by two successive frames, and a stack of 8 to 12 such images
was obtained for each cell.
Cofactor assay
This assay utilizes biotinylated ligands (C3b and C4b). The fragments of the cleavage reaction, resulting from addition of factor I with MCP, were assessed by Western blot analysis following electrophoresis (10% reducing SDS-PAGE). C3b and C4b (Advanced Research Technologies, San Diego, CA) were biotinylated utilizing a 50x molar excess of EZ-Link Sulfo NHS-LC-biotin (Pierce) per the manufacturer’s directions. The sample was dialyzed with a Microcon 30 unit (Amicon, Beverly, MA), and aliquots were stored at -70°C. To optimize the signal, serial dilutions of the reduced biotinylated ligand (25 mM DTT) were evaluated on a 10% SDS-PAGE gel and then transferred to nitrocellulose for Western blot analysis (27). Membranes were rinsed in TTBS (TBS with 0.05% Tween-20) for 2 min and probed with 1:1500 dilution of ExtrAvidin-horseradish peroxidase (Sigma, St. Louis, MO) for 1 h at 37°C with rotation. Blots were washed three times with TTBS for 5 min each with rotation. Using a clean dish, chemiluminescent substrate (SuperSignal; Pierce) was incubated for 1 min with the blot. The blot was wrapped in plastic wrap and x-ray film exposed for varying amounts of time followed by film development. The dilution that produced the optimal signal to noise ratio was utilized.
For the cofactor assay, all dilutions were prepared in low salt buffer
(10 mM Tris, pH 7.2, with 25 mM sodium chloride, 1% Nonidet P-40, and
freshly added 2 mM PMSF). C3b or C4b were diluted as determined above.
Factor I (Advanced Research Technologies) was added (100 ng), and cell
lysates (varying from 2.5 to 5 x 108 MCP/assay) were
added in a total of 15 µl. Following an incubation at 37°C for
1.5 h, an equal volume of 2x reducing buffer was added, the
samples heated to 95°C for 3 min, and SDS-PAGE/Western blotting
performed as described above. X-ray films were scanned by using a laser
densitometer (Pharmacia LKB Biotechnology, Piscataway, NJ) and analyzed
with GELscan software (Pharmacia). The generation of the C4d fragment
was monitored and adjusted relative to the input as measured by the
-chair fragment of C4b. Mutants were compared with wild-type MCP.
Ligand binding ELISA
C3b and C4b (Advanced Research Technologies) were coated at 5 µg/ml in TBS overnight at 4°C in microtiter wells (Nunc modules; Fisher Scientific) as described (27).
Briefly, coated wells were blocked 1 h at 37°C with 1% BSA and 0.1% Tween 20 in TBS and washed in low salt ELISA buffer (LSEB) consisting of 10 mM Tris (pH 7.2), 25 mM sodium chloride, and 0.05% Tween 20. Lysates, quantified for MCP by ELISA (see above), were diluted to equivalent quantities over nearly a log in LSEB, and incubated 2 h at 37°C. Following washing in LSEB, rabbit antiserum to MCP was diluted 1:2500 in LSEB and 4% BSA/0.25% Nonidet-P40, and incubated for 1 h 37°C. After washing, horseradish peroxidase-coupled donkey anti-rabbit IgG was added and incubated 1 h 37°C, and the wells were washed. Detection was made utilizing TMB, and optical density (630 nm) was measured in an ELISA reader.
Cytoprotection assays
Clones obtained by limiting dilution (see above) possessing equivalent levels of MCP mutants were isolated and employed in cytoprotection assays as described previously (27). Briefly, 10,000 cells/well were plated overnight in 96-well microtiter plates. The medium was removed and rabbit anti-hamster Ab (IgG fraction; Sigma) was added (13 mg/ml) for 30 min at 4°C. Ab was removed and serum (diluted in gelatin veronal buffer) was added for 1 h at 37°C. Wells were washed twice in Dulbecco’s PBS, and normal medium was added. Cells were grown for 48 h and assessed with the CellTiter 96 kit per manufacturer’s directions (Promega, Madison, WI). Assays were performed in quadruplicate on at least three separate occasions. Untreated, Ab-only, and serum-only controls were performed with each assay. Normal human serum (NHS) and C6-depleted serum (utilized in C4 deposition studies) were obtained from Quidel (San Diego, CA).
C4b fragment deposition
These studies were performed as previously described (27). Briefly, cells were challenged identically as for the cytoprotection assays except that C6-depleted serum was utilized to prevent cytolysis. In addition, for kinetic analysis, the C6-depleted human serum was incubated for the time points indicated. mAbs to C4c and C4d (20 µg/ml) were incubated with these sensitized cells for 30 min at 4°C. Cells were washed and incubated with FITC-conjugated goat anti-mouse IgG for 30 min at 4°C followed by washing. Cells were fixed in 0.5% paraformaldehyde, and FACS analysis was performed.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To analyze the importance of N- and O-linked
glycosylation on MCP complement regulatory function, mutants were
constructed by using site-directed mutagenesis to delete each of these
sites (Fig. 1
). The motif for
N-linked glycosylation (N-X-S/T, with X being any amino acid
except proline (30)), was mutated such that a glutamine (Q) was
substituted for the asparagine (N) in CCP-1, -2, and -4 and termed NQ1,
NQ2, and NQ4, respectively. To assess the role of O-linked
sugars, a mutant deleting the STP domain was created (termed STP).
Following the stable transfection of mutant and control cDNAs into CHO
cells, expressed MCP and mutant protein were assessed with polyclonal
and/or monoclonal anti-MCP Abs by FACS and Western blotting.
Subsequently, clones with equivalent levels of expression were isolated
and evaluated.
|
,
the BC1 isoform and glycan deficient mutants displayed the expected
relative mobilities. NQ2 and NQ4 possessed relatively faster mobilities
consistent with the loss of one N-glycan group, whereas the
STP mutant demonstrated the fastest mobility (as expected, since it
lacks all O-glycosylation). The somewhat faster mobility of
NQ1 vs NQ2 and NQ4 was noted previously and likely results from a
conformational effect since the mobility of these forms is equivalent
following reduction (25).
|
). The clones utilized were designated
as 23-9 (BC1), NQ1-8, NQ2-5, NQ4-12 and STP-30. Next, further
comparative evaluations for protein stability and expression patterns
of each of these clones were undertaken. First, epitope sites were
evaluated by FACS with several monoclonal anti-MCP Abs (data not
shown). The clones selected exhibited similar MCP expression levels and
retained epitopes represented by the mAb: 1) TRA-2-10, which binds a
site in CCP-1 and inhibits MV binding (20); 2) M75, which binds to a
complement regulatory and MV functional site in CCP-2 (18, 20), and 3)
GB24, which binds to a complement regulatory site in CCP-3/-4 and
blocks C3b/C4b binding (17). The FACS profiles for the three mAb were
similar, suggesting that these mAb bind to a peptide epitope unaltered
by the lack of glycans (data not shown). Second, clones were exposed to human serum (1:8 dilution) for 1 h at 37°C and then lysed. Western blot analysis with an anti-MCP polyclonal Ab demonstrated no significant differences in pattern of electrophoretic migration or protein quantity of treated and untreated cells (data not shown).
Third, to examine cell surface distribution, confocal microscopy was
performed on clones utilizing an anti-MCP mAb (TRA-2-10) (Fig. 3). This examination revealed a similar
and uniform MCP surface distribution for all clones. In addition,
clones "challenged" by Ab and C6-deficient human serum (see below)
maintained this pattern (data not shown).
|
STP cell lysates exhibit decreased cofactor activity
Initial comparisons of C3b and C4b cofactor activity were
performed utilizing cell lysates of transfectant pools. As indicated in
Figure 4, NQ4 and STP lysates had
reduced C3b and C4b cofactor activity, whereas NQ1 and NQ2 were
equivalent to BC1 (wild type) in C3b activity, but demonstrated a
modest decrease in C4b cofactor activity. These studies suggested that
the N- and O-glycosylation domains may be
important for regulatory function. To further examine this possibility
and to determine whether alterations in ligand binding accounted for
these functional differences, we performed C3b and C4b binding studies.
|
For ligand binding, an ELISA was utilized in which C3b or C4b was
adsorbed to wells and dilutions of quantified transfectant lysates were
evaluated (Fig. 5). All mutants retained
ligand binding capability similar to wild-type BC1. As noted previously
(27), there was a modest, but reproducible, preferential binding by
wild-type BC1 to C4b vs C3b that was maintained by all mutants except
NQ2 (see Fig. 5, A–E). NQ2 also showed a slight
elevation in binding to both ligands, but more so to C3b. There was a
consistent pattern of slightly enhanced binding by NQ2 and slightly
less binding by NQ4 to both C3b and C4b. We concluded that, in this
solid phase assay using solubilized cell lysates, deletion of the
N- or O-linked glycosylation sites had minimal
influence on ligand binding. Retention of ligand binding despite
partial loss of cofactor activity as observed for NQ4 and STP is
consistent with previous findings that binding and cofactor activity
are separable (17, 31).
|
The clones profiled in Figure 2 were utilized in
cytoprotection assays. The cells were treated with a polyclonal Ab to
hamster cell surface Ags followed by incubation with NHS as the source
of complement. As expected and established previously, the classical
pathway mediates cytoinjury to CHO cells in this system (27, 32, 33).
Figure 6 compares cytoprotection by
control and mutant clones over a range of serum dilutions. The NQ1
mutant was similar to wild type, suggesting little effect by the
N-glycan in CCP-1, consistent with previous results in which
CCP-1 was deleted without altering ligand binding and cofactor activity
(17, 18). However, mutants NQ2 and NQ4 demonstrated a marked decrease
in their cytoprotective capability with an activity profile very
similar to the RCHO control cells (CHO cells transfected with MCP
construct in the reverse orientation). The mutant deleted of the
STP region exhibited an intermediate capacity for cytoprotection.
|
The differences in the cytoprotective ability of the glycosylation
deletion mutants implied differential potencies of the mutant proteins
to serve as cofactors for the factor I-mediated cleavage of C3b and
C4b. Also, since differences had been noted in cofactor assays
employing cell lysates (see Fig. 4), we assessed cofactor function on
the cell surface. For this investigation, we employed a nonlytic system
in which clones were treated with the same anti-CHO cell Ab, but
C6-deficient human serum was utilized. The cleavage of C4b was
monitored by FACS with mAb against the cleavage fragments of C4b.
Cleavage of C4b by MCP releases a large fragment (C4c), whereas the
smaller fragment (C4d) remains covalently attached to the cell
membrane. As C4b is cleaved, the quantity of C4c is reduced while the
C4d level remains constant. Thus, a comparison of cell surface C4c vs
C4d by FACS permits a quantitative assessment of the kinetics of C4b
cleavage.
Representative histograms comparing C4c vs C4d on the cell surface are
presented in Figure 7. On control cells
that do not express MCP (Fig. 7B), the quantities of C4c and
C4d were coincident, indicating that cell-bound C4b was not cleaved and
establishing that C4b binding protein or other regulatory proteins had
no activity in this regard. All cells expressing MCP possessed C4b
cofactor activity, since there was a reduction of cell-bound C4c
relative to C4d (Fig. 7, C–G). A comparison of
the relative mean fluorescence intensity for these two C4 fragments is
presented in Table I. The relative
intensity values were the lowest for wild-type (BC1) and NQ1
(indicating the highest C4b-cleaving activity) followed by NQ4, NQ2,
and STP. This pattern parallels results obtained by the cytotoxicity
experiments, except that STP had less cofactor activity than
anticipated.
|
|
To further assess these differences in cofactor activity, we
examined the cleavage patterns after 10, 20, and 60 min of complement
exposure (Fig. 8). These data indicate
that MCP-dependent cofactor activity begins concomitant with C4b
deposition. Rapid, incremental cleavage then occurred for at least 20
min for wild type BC1 as well as the mutants. By 20 min, 50% of the
C4b had been cleaved in the case of BC1 and NQ1, whereas less cleavage
took place for NQ4, NQ2, and STP. At the final time point of 60 min,
25 to 35% of the C4b had not been cleaved to C4c and C4d in wild type
and all mutants except STP, in which case 50% remained as C4b.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The goal of the present study was to further examine these glycan
modifications in MCP, and in particular assess the importance of
carbohydrates as they relate to complement regulatory function. The
initial C3b and C4b cofactor assays suggested that glycosylation played
a role in regulation since both NQ4 and the STP mutants demonstrated
reduced activity. However, ligand binding studies indicated that all
mutants bound both C3b and C4b similarly to wild type. To examine these
potentially discrepant observations, we assessed MCP function in a more
biologically relevant assay system.
A significant finding of these subsequent experiments was that the
N-glycans of CCP-2 and -4 were critical for the
cytoprotective activity of MCP. The deletion of either nearly abrogated
this inhibitory capability and, thus, was a much more striking
functional deficiency than would have been predicted from the ligand
binding and cofactor assays employing solubilized cell extracts. This
activity profile was explained, in part, by a decreased ability for C4b
cleavage in the fluid phase (Fig. 4) as well as on the cell surface
(Fig. 7, Table I). These results are consistent with previous data
indicating that CCP-2 and -4 are required for complement regulatory
function (17, 18). The deletion of the N-glycan in CCP-1 did
not diminish the cytoprotective function of MCP, a result also
consistent with previous findings that 1) CCP-1 does not contribute to
complement inhibitory function (17, 18) and 2) it is spliced out in new
world primates (22). Finally, it is worth noting that, in these
cytoprotection experiments, cleavage was strictly MCP dependent with no
detectable activity by C4b-binding protein since non-MCP expressing CHO
cells did not demonstrate cleavage of C4b (see Fig. 7).
Loss of efficient cytoprotective capacity by glycosylation mutants could be explained by other reasons such as failure of the mutant protein to be transported to the cell surface or to gross conformational alterations. Multiple lines of evidence suggested that these did not occur. First, we were able to obtain stable clones of mutants and to select for equivalent expression levels utilizing FACS. Second, Western blots bearing similar cell equivalents revealed appropriate m.w. sizes and similar quantities. Third, there were no alterations in the interactions of mutants with peptide-specific mAbs to MCP. Fourth, Western blot profiles of cells exposed to human serum were identical to untreated controls, demonstrating no serum-dependent degradation of MCP. Fifth, confocal microscopy revealed similarly uniform distribution patterns of MCP protein for all clones (also true of cells "challenged" by Ab and complement (C6-deficient serum)). Last, ligand binding assays employing solubilized transfectant lysates demonstrated similar activity of all mutants as compared with wild-type MCP.
The necessity of N-glycans for function is variable among
complement regulatory proteins. For example, the deletion of
N-glycans from C4b-binding protein, decay accelerating
factor (DAF, CD55), and CD59 did not diminish complement inhibitory
activity (38, 39, 40). Consequently, it has been suggested that the
N-glycans of these proteins may serve in other capacities.
Specifically, the single N-site of CD59 is conserved in all
homologous mammalian proteins examined to date and may be a site of
interaction with the T cell accessory molecule CD2 (40). On the other
hand, N-glycans are critical for the function of many other
proteins including the N-linked carbohydrates of Igs that
serve a variety of effector functions including complement activation
and ligand interaction) (reviewed in Refs. 41 and 42). In addition, for
the IFN- receptor, inhibition of N-linked glycosylation
or modulation of carbohydrate processing did not prevent receptor
transport to the cell membrane, yet blocked ligand binding thereby
completely abrogating receptor function (43).
An indication of the importance of N-glycans in MCP is
suggested by their evolutionary conservation across species (Fig. 9). Recently, MCP sequence data have
become available for African green monkey (Vero) (44), guinea pig (45),
pig (46, 47), and several old and new world monkeys (22). The
N-sites in CCP-1, -2, and -4 are conserved in human, Vero,
and all five characterized species of old world monkeys. Interestingly,
MCPs of new world monkeys retain the N-linked sites in CCP-2
and -4, yet delete CCP-1 entirely. Deletion of this domain by new world
monkeys may have afforded protection against a potentially lethal
microorganism such as MV. Three additional N-sites were
possible in Vero, although enzymatic treatment suggested these may not
be utilized (44), which also may be the case for the other old world
monkeys. In 13 of the 14 species examined, the location of the
N-site of CCP-4 was uniformly conserved. The one exception
was guinea pig MCP, which also possessed an N-glycan in
CCP-4, but at a slightly different locale. Such evolutionary
conservation of these posttranslational modifications, in most cases of
the identical sites, strongly suggests they play important biologic
roles.
|
STP was the least effective cofactor among the mutants. One
explanation may be that these tests assess different parameters in that
cytotoxicity studies measured cell survival, whereas cell lysates and
cell surface C4c content exclusively gauged cofactor activity. Also of
note, the level of cytotoxicity and the C4c/C4d ratio for STP was
comparable to that observed previously for the MCP-C1 isoform (which
has a smaller and less glycosylated STP domain) (27). The
interpretation of this earlier study was that the larger BC segment
augmented the ability of MCP to act as a cofactor for C4b degradation
on the cell surface (27) and is consistent with the present results of
entirely eliminating the STP domain. It should be recalled that more
than 95% of the population express predominantly the higher (BC)
glycosylation forms of MCP, whereas only 5% predominantly express the
C isoforms (48). Taken together, these studies strongly indicate that
the BC isoforms protect better against the classical pathway than C
isoforms. Previous studies also indicated that the STP domain of DAF was functionally important. Deletion of this segment produced a complete loss of decay accelerating activity that was reestablished by re-inserting a similarly sized, but not O-glycosylated, segment from a noncomplement protein (39). These results point out a role for the STP region serving as a nonspecific spacer, perhaps projecting the CCP functional domains away from the plasma membrane. In our investigation, there was a moderate decrease in, but not a complete loss of, function. It is possible that the juxtamembranous segment (the "undefined" domain of 12 amino acids) may account for these differences.
The O-glycosylated region in MCP of the three other species available revealed that all possessed at least one domain for O-glycosylation following the CCP region. Vero MCP had a 14-residue segment identical to the STP-C domain of human MCP, guinea pig a 16-amino acid segment with 73% identity to the human C domain, and pig MCP a 17-amino acid segment with 25% homology to the B domain. Additional cDNA clones and RT-PCR in these species will be required to establish if this segment is alternatively spliced in a similar fashion to humans.
Our study and those of many others indicate that, in a classical
pathway mediated system, large amounts of C4b can be deposited in a few
minutes (49). The kinetic analysis of the cofactor activity of MCP
mutants indicated that: 1) C4b cleavage began on the cell surface
concomitant with deposition; 2) the cleavage pattern was linear for 20
to 30 min, with most of this activity occurring within this time frame;
3) substantial quantities (25%) of C4b remained as such at 60 min,
even with wild-type MCP; 4) surprisingly, there was no cleavage of C4b
on the non-MCP expressing CHO control cells by serum C4b-binding
protein; 5) in this system, the rate of C4b deposition far exceeds the
rate of its cleavage, suggesting limits on the cytoprotective
capabilities of regulatory proteins in Ab-mediated syndromes of
autoimmunity; and 6) the ratio of deposited C4b (6–8 x
105) to MCP (0.5 x 105) is estimated at
greater than 10 to 1; thus, to cleave 75% of the C4b to C4d and C4c
requires an interaction of each MCP with multiple C4bs.
The majority of membrane proteins of eukaryotic organisms are glycosylated. These posttranslational modifications play key roles in biologic functions (30, 50). For MCP, the discovery that the N-glycans of CCP-2 and -4 do not directly participate in ligand binding and only modestly reduce cofactor activity, but substantially influence cytoprotection, provides an instructive example of how these carbohydrate moieties may promote enhanced biologic activity at the cell surface that could not have been appreciated by studying either the isolated protein or solubilized MCP cell extracts. Indeed, the differences in activity point to additional biologic interactions not evaluated in the above experiments. For example, carbohydrates may facilitate the movement or clustering of MCP on a cell membrane that could assist its interactions with ligand, possibly in association with a convertase. Our findings of the necessity of N-glycans for function coupled with their evolutionary conservation across species establishes important biologic roles for these modifications in MCP.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to John P. Atkinson, Division of Rheumatology, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8045, St. Louis, MO 63110.
3 Abbreviations used in this paper: MCP, membrane cofactor protein; MV, measles virus; CCP complement control protein; STP, serine/threonine/proline-enriched segment; CHO, Chinese hamster ovary; TMB, 3,3',5,5'-tetramethylbenzidine; NHS, normal human serum.
Received for publication January 13, 1998. Accepted for publication May 28, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
receptor: N-linked carbohydrates contribute to structural heterogeneity and are required for ligand binding. J. Biol. Chem. 265:1710.This article has been cited by other articles:
|
|
 |
|
  M. K. Liszewski, P. Bertram, M. K. Leung, R. Hauhart, L. Zhang, and J. P. Atkinson Smallpox Inhibitor of Complement Enzymes (SPICE): Regulation of Complement Activation on Cells and Mechanism of Its Cellular Attachment J. Immunol., September 15, 2008; 181(6): 4199 - 4207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Fang, V. Fremeaux-Bacchi, M. K. Liszewski, G. Pianetti, M. Noris, T. H. J. Goodship, and J. P. Atkinson Membrane cofactor protein mutations in atypical hemolytic uremic syndrome (aHUS), fatal Stx-HUS, C3 glomerulonephritis, and the HELLP syndrome Blood, January 15, 2008; 111(2): 624 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ram, J. Ngampasutadol, A. D. Cox, A. M. Blom, L. A. Lewis, F. St. Michael, J. Stupak, S. Gulati, and P. A. Rice Heptose I Glycan Substitutions on Neisseria gonorrhoeae Lipooligosaccharide Influence C4b-Binding Protein Binding and Serum Resistance Infect. Immun., August 1, 2007; 75(8): 4071 - 4081. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Servais, V. Fremeaux-Bacchi, M. Lequintrec, R. Salomon, J. Blouin, B. Knebelmann, J.-P. Grunfeld, P. Lesavre, L.-H. Noel, and F. Fakhouri Primary glomerulonephritis with isolated C3 deposits: a new entity which shares common genetic risk factors with haemolytic uraemic syndrome J. Med. Genet., March 1, 2007; 44(3): 193 - 199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Caprioli, M. Noris, S. Brioschi, G. Pianetti, F. Castelletti, P. Bettinaglio, C. Mele, E. Bresin, L. Cassis, S. Gamba, et al. Genetics of HUS: the impact of MCP, CFH, and IF mutations on clinical presentation, response to treatment, and outcome Blood, August 15, 2006; 108(4): 1267 - 1279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Fremeaux-Bacchi, E. A. Moulton, D. Kavanagh, M.-A. Dragon-Durey, J. Blouin, A. Caudy, N. Arzouk, R. Cleper, M. Francois, G. Guest, et al. Genetic and Functional Analyses of Membrane Cofactor Protein (CD46) Mutations in Atypical Hemolytic Uremic Syndrome J. Am. Soc. Nephrol., July 1, 2006; 17(7): 2017 - 2025. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Liszewski, M. K. Leung, R. Hauhart, R. M. L. Buller, P. Bertram, X. Wang, A. M. Rosengard, G. J. Kotwal, and J. P. Atkinson Structure and Regulatory Profile of the Monkeypox Inhibitor of Complement: Comparison to Homologs in Vaccinia and Variola and Evidence for Dimer Formation J. Immunol., March 15, 2006; 176(6): 3725 - 3734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Richards, E. J. Kemp, M. K. Liszewski, J. A. Goodship, A. K. Lampe, R. Decorte, M. H. Muslumanogglu, S. Kavukcu, G. Filler, Y. Pirson, et al. Mutations in human complement regulator, membrane cofactor protein (CD46), predispose to development of familial hemolytic uremic syndrome PNAS, October 28, 2003; 100(22): 12966 - 12971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Barilla-LaBarca, M. K. Liszewski, J. D. Lambris, D. Hourcade, and J. P. Atkinson Role of Membrane Cofactor Protein (CD46) in Regulation of C4b and C3b Deposited on Cells J. Immunol., June 15, 2002; 168(12): 6298 - 6304. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ram, M. Cullinane, A. M. Blom, S. Gulati, D. P. McQuillen, B. G. Monks, C. O'Connell, R. Boden, C. Elkins, M. K. Pangburn, et al. Binding of C4b-binding Protein to Porin: A Molecular Mechanism of Serum Resistance of Neisseria gonorrhoeae J. Exp. Med., January 29, 2001; 193(3): 281 - 296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Wang, M. K. Liszewski, A. C. Chan, and J. P. Atkinson Membrane Cofactor Protein (MCP; CD46): Isoform-Specific Tyrosine Phosphorylation J. Immunol., February 15, 2000; 164(4): 1839 - 1846. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Liszewski, M. Leung, W. Cui, V. B. Subramanian, J. Parkinson, P. N. Barlow, M. Manchester, and J. P. Atkinson Dissecting Sites Important for Complement Regulatory Activity in Membrane Cofactor Protein (MCP; CD46) J. Biol. Chem., November 22, 2000; 275(48): 37692 - 37701. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
