| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Centre Hospitalier Universitaire Reims, Laboratoire d’Immunologie, Unité de Formation et de Recherche Médecine Université de Reims Champugne Ardennes, Pôle Biomolécules IFR53 Reims, France;
Centre National de Référence des Groupes Sanguins, Institute National de Transfusion Sanguine, Paris, France; and
Laboratoire de Biologie et Thérapeutique des Pathologies Immunitaires, Equipe de Recherche Associée 7087, Université Pierre et Marie Curie, Centre National de la Recherche Scientifique, Centre l’etude et de Recherche en Virologie et Immunologie, Paris, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain of the C4 binding protein (C4bp) was used. We first
produced a multimeric soluble CR1 that proved to be a better inhibitor
of in vitro C activation than the monomeric form of CR1, then a
heteromultimeric molecule made of CR1 and single-chain Fv
anti-Rh(D) valences able to attach E and providing E with as much
as a 10-fold increase in CR1 density with the same CR1 distribution
pattern as native E. CR1/single-chain Fv anti-Rh(D)-treated E were
able in vitro to attach as many opsonized IC as native E. These data
open the way for future use of multimeric and heteromultimeric forms of
soluble recombinant CR1 as therapy of IC
diseases. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In primates, including man, CR1 on E (CR1/E) transports C3b opsonized IC from the circulation to liver and spleen (3). Immune adherence of IC to E depends on the number of molecules of CR1/E (4). CR1 density on E is controlled by a codominant autosomal gene related to a HindIII polymorphism located in an intron of the CR1 gene (5). CR1 is expressed on E at lower density (200–1000 sites) compared with leukocytes (20,000 sites). However, the organization in clusters of CR1 on E membrane (see the 1988 "cover of the year" of the Complement and Inflammation Journal (6)) favors high avidity multimeric ligand-receptor interactions. Thus, despite the low number of CR1 molecules expressed on each E, CR1 enables E to be the main agent of IC clearance. CR1 also serves as a cofactor of factor I for C3b and C4b cleavage and inactivation (7). Physiologically CR1 density decreases during aging on normal E (8). E from patients suffering from systemic lupus erythematosus (SLE) (9, 10), some hemolytic anemia (11), and HIV infection share an acquired reduction of the expression of CR1 (12, 13, 14). This low density of CR1 on E contributes to impairing IC transport mechanisms in patients with SLE, resulting in deposition of IC outside liver or spleen, which may induce tissue damage in susceptible organs, particularly kidney (15, 16).
In addition, inappropriate addressing of Ag-containing IC leads to an impaired Ab response to the Ag entrapped in IC. Complement consumption on vasculitis induced by IC precipitation leads to a down-spiraling process regarding clearance of IC; IC are poorly opsonized due to C deficiency, poorly transported due to CR1 deficiency on E, and only induce a poor Ab response to the Ag contained in IC, unable to clear them from the circulation.
Restoring the number of CR1/E may be an important new way for therapy
of human IC-mediated diseases (14). In a monkey model, the
CR1 density on E has been enhanced by bleeding these animals to
increase the proportion of young E (17). However, possibly
due to the practical limitation of this model, no significant effect on
experimental IC nephritis was observed. Another system has been
proposed to enhance the IC capture ability of E using bispecific Abs,
one valence being directed against CR1, and the other against the Ag of
interest. In experimental models in monkeys, capture of bacteriophage
X174 containing complexes has been demonstrated (18, 19). In addition to its technical complexity this model suffers
from two major limitations: 1) bispecific Abs cannot be used when CR1
density is already decreased on E; and 2) the precise Ag to be cleared
must be known, and the corresponding specificity should be introduced
into the bispecific Ab construction.
A more direct approach to restore the IC transportation capacity of E would be to attach recombinant CR1 directly on E. A monomeric soluble CR1 recombinant molecule has been produced (sCR1) (20). In vitro, sCR1 is a powerful inhibitor of human complement activation. In vivo, sCR1 is able to reduce the consequences of inflammation in some experimental models, such as postischemic myocardial inflammation, and in organ transplantation (21, 22, 23).Unfortunately, due to its weak affinity for C3b and short in vivo half-life, sCR1 has not demonstrated a beneficial effect in IC-mediated diseases.
Direct chemical linkage of monomeric CR1 randomly cross-links anywhere throughout the CR1 molecule to other E surface molecules and is unlikely to reproduce the tridimensional cluster long chain structure of CR1 on E demonstrated by the typical patchy staining of E for CR1 using anti-CR1 mAbs. Requirements for realistic future immunointervention restoring a physiological or a supraphysiologic density of functional CR1 on E are to obtain multimeric clusters of CR1 and to attach them easily on E by using an Ag-Ab ligand receptor interaction. The design of a recombinant bispecific anti-E/CR1 molecule was undertaken for that purpose.
Because existing multimerizing systems have not been found satisfactory, a new multimerizing system for recombinant protein has been developed to fulfil the required criteria for future in vivo use of soluble bifunctional multimeric molecules.
The multimerizing system should originate from a normal plasma molecule to avoid immunization, should lack any biological function, particularly complement activation or cell surface receptor interactions, and should confer at least five valences or more.
Despite their ability to induce multimerization, previous systems based
on Fc
(24), Fcµ (25), leucine zipper
(26), chemical polyethylene glycol linkage
(27), diabodies (28), streptavidin
(29), and protein A (30) failed to pass these
criteria. Some multimerizing proteins such as Fc or diabodies are
direct activators of the complement cascade. Leucine zipper-based
proteins are likely to be immunogenic and poorly soluble as originating
from an intracellular protein as well as poorly stable and difficult to
purify since they are noncovalently linked.
A new system fulfilled the criteria for a fair multimerizing system.
The C4bp molecule is a spider-like structure made of seven -chains
and one ß-chain (31). The basic repetitive structure of
both chains is termed short consensus repeat (SCR). Each SCR of about
60 aa includes two intrachain disulfide bridges. Binding sites for C4b
are located on the four N-terminal SCR of the -chains, whereas the
S-protein binding site is located on the two N-terminal SCR of the
ß-chain (32). Minor forms made from only seven
-chains or five /1 ß-chain molecules have also been described
and found at a lower concentration in normal human plasma
(32, 33).
The C-terminal part of the complement binding protein C4bp lacks
biological function, but is responsible for the polymerization of the
molecule in the cytoplasm of C4bp-producing cells (32).
Due to this property and the lack of biological functions as well as
the probable nonimmunogenicity of a normal human plasma protein, the
C-terminal part of the -chain of C4bp was chosen to set up
homomultimeric and heteromultimeric proteins.
cDNA containing the C-terminal part of the C4bp coding sequence can be fused in 3' with the sequence coding for the protein of interest, then transfected in cells as a single-phase construct. The chimera protein spontaneously multimerizes in the cytoplasm of transfected cells that secrete covalently linked multimeres (34). Cotransfection of cells by two different vectors containing sequences coding for two distinct multimeric molecules was believed to lead to the production of a heteromultimeric protein with valences from both molecules covalently linked together by disulfide bridges.
A C4bp C-terminal fragment-based multimeric soluble CR1 recombinant molecule was initially developed, then an heteromultimeric molecule composed of anti-rhesus(D) (anti-Rh(D)) scFv and CR1 valences was prepared and tested for fixation on E and ability to restore or to enhance IC capture by treated E.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A lymphoblastoid cell line, H2D5D2F5 (referred to as DF5),
derived from the peripheral blood lymphocytes of a hyperimmunized donor
(35) and producing a human monoclonal IgG1
, specific
for erythrocyte Rh(D) Ag, was used as source of monoclonal cells to
rescue VH and VL regions
via PCR amplification.
The single-chain Fv (scFv) was constructed according to the method
described by Marks et al. (36). Briefly, total RNA was
extracted from the lymphoblastoid cells, cDNA was obtained using the
first-strand cDNA synthesis kit (Pharmacia, Uppsala, Sweden) and pd(N)6
random primers. Amplification of VH and
VL coding fragments was effected with the
specific V gene family primers described by Marks et al.
(37): GTG CAG HuVH1aBACK (5'-CAG GTG CAG CTG TCT GG-3')
and HuJH4–5FOR (5'-ACC TAA AAC GGT GAG CTG GGT CCC-3') for the heavy
chain; and HuV1BACK (5'-CAG TCT GTG TTG ACG CAG CCG CC-3') and
HuJ2–3FOR (5'-ACC TAG GAC GGT CAG CTT GGT CCC-3') for the light
chain.
PCR assembly of the paired VH and VL with a (Gly4-Ser)3 linker and SfiI and NotI restriction site addition was performed as described by Marks et al. (36, 37). The assembled structures were cloned into pHEN1 (38), (vector provided by G. Winter, Medical Research Council Laboratory of Molecular Biology, Cambridge, U.K.). The ligation product was introduced into Escherichia coli HB2151 by electroporation, and insert-containing clones were tested for the expression of soluble scFv following isopropyl ß-D-thiogalactoside (IPTG) induction (36). The scFv construct from a clone with high specific agglutinating reactivity to anti-Rh(D) erythrocytes was selected for construction of the chimeric protein.
Primers used to amplify the C-terminal part of the C4bp molecule
A 169, or 202 when including an Myc tag, base pair C-terminal
C4bp fragment for scFv anti-Rh(D)/C4bp was
amplified using the following primers, including, or not, a segment
coding for a Myc tag inserted between anti-Rh(D) scFv and C4bp
fragments: 5' primer with Myc tag,
5'-AGTGCGGCCGCAGAACAAAAACTCATCTCAGAAGAGGATCTGAATGAGACCCCCGAAGGCTGTGA-3';
5' primer without Myc tag, 5'-
AGTGCGGCCGCAGAGACCCCCGAAGGCTGTGA-3'; and 3' primer,
5'-CTCGCGGCCGCCTCGAGTTATAGTTCTTTATCCAAAGTGG-3'. Underlined
sequences represent restriction endonuclease sites. The 5' and 3'
primers contain NotI and XhoI sites,
respectively. The sequence depicted in bold characters codes for a
EQKLISEEDLN Myc tag peptide. The amplified sequence contained at its 3'
end a stop codon and a restriction site for XhoI as well as
at its 5' end a restriction site for NotI.
A 167-bp C-terminal C4bp fragment for CR1/C4bp was amplified using
the following primers: 5' primer, 5'-CCGAGACCCCCGAAGGCTGTGA-3'; and 3'
primer, 5'-CTCGCGGCCGCCTCGAGTTATAGTTCTTTATCCAAAGTGG-3'. The
underlined sequence represents restriction endonuclease site. The 3'
primer contained a NotI site.
PCR amplification
Genomic DNA was used as a template for PCR amplification. A 100-µl reaction mixture was prepared, containing 500 ng of DNA, 500 ng of 5' and 3' primers, 16 nM dNTP, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, and 2 U of AmpliTaq DNA polymerase (Perkin-Elmer, Roissy, France), then subjected to 30 cycles of amplification using a GenAmp PCR System 9600 (Perkin-Elmer). Cycles were 1 min at 94°C, 1 min at 56°C, and 2 min at 72°C. PCR products were analyzed by electrophoresis on a 2% agarose gel. Following that control, DNA was extracted once with phenol/chloroform and once with chloroform, ethanol precipitated, and resuspended in 100 µl of water before digestion with NotI and XhoI for scFv anti-Rh(D) or with NotI for CR1.
Cloning of the CR1/C4bp (multiCR1) and anti-Rh(D) scFv/C4bp (multiRH1) constructs
Restriction enzymes and alkaline phosphatase used for cloning
were purchased from Roche (Meylan, France). The T4 DNA ligation kit
used was from Ozyme (Montigny-Le-Bretonneux, France). CR1 cDNA coding
for the soluble extramembraneous part of CR1 (given by W. Wong Harvard
Medical School, Boston, MA) was inserted in pCDM8 vector (Invitrogen,
Leek, The Netherlands). The cDNA coding for CR1 was digested by
XhoI and BalI. The matching amplified C4bp
fragment was digested by NotI, leading to 5' blunt and 3'
cohesive end ligations. The DNA construct coding for the anti-Rh(D)
scFv was digested by HindIII and NotI. The
matching amplified C4bp fragment was digested by NotI and
XhoI. For each construct, two DNA fragments were linked in a
dephosphorylated pKC3 vector (Fig. 1
).
|
Transfection of pKC3 CR1/C4bp construct was performed using a
calcium phosphate transfection kit (5 Prime-3 Prime, Tebu, Le Perray en
Yvelines, France), then selected for multiCR1 expression by limiting
dilution and screening for CR1 secretion by ELISA. Plasmid pKC3 scFv
anti-Rh(D)/C4bp was cotransfected with the dihydrofolate reductase
(DHFR)-selective plasmid ST4 in a multiCR1 secreting clone. After
limiting dilution multiCR1/scFv anti-Rh(D)-secreting clones were
screened by flow cytometry for their ability to bind papain-treated E
and to attach CR1 onto them.
Cell cultures
CHO DHFR-deficient (DHFR-; ATCC CRL-9096, American Type Culture
Collection, Manassas, VA) were cultured in Ham’s F-12 (Life
Technologies, Cergy Pontoise, France) supplemented with glutamine (2
mM), 10% FCS (Life Technologies), penicillin/streptomycin/fungizone
(1000 U/ml; 1000 µg/ml; 2.5 µg/ml). CHO DHFR- transfected by pKC3
CR1/C4bp and sT4-DHFR were cultured in Ham’s medium without
thymidine and hypoxanthine (Biochrome, l’Aigle, France) supplemented
with 10% heat-inactivated dialyzed FCS (Life Technologies),
penicillin/streptomycin/fungizone (1000 U/ml; 1000 µg/ml; 2.5
µg/ml), and glutamine (2 mM).Cells transfected by pKC3 CR1/C4bp,
pKC3 scFv anti-Rh(D)/C4bp, sT4-DHFR, and pMAMNeo were cultured
in Ham’s medium without thymidine and hypoxanthine supplemented with
10% heat-inactivated dialyzed FCS, penicillin/streptomycin/fungizone
(1000 U/ml; 1000 µg/ml; 2.5 µg/ml), glutamine (2 mM), and G418 (700
µg/ml; Sigma, St. Louis, MO).
Antibodies
mAb anti-CR1 (J3D3 and J3B11) and mAb against Myc Tag (9E10, ATCC CRL-1729) were purified from mouse ascites using the octanoic acid contraprecipitation method (39). Hybridoma cells were cultured in RPMI 1640 medium (Life Technologies) supplemented with 10% heat-inactivated FCS, glutamine (2 mM; Life Technologies), penicillin/streptomycin/fungizone (1000 U/ml; 1000 µg/ml; 2.5 µg/ml; Sigma). J3D3 and J3B11 cells lines were a gift from E. Fischer (Institut National de la Santé et Recherche Medicale, Unit 430, Paris, France). Rabbit anti-human E polyclonal Abs were purchased from Cappel Laboratories (Cochranville, PA). Rabbit anti-BSA polyclonal Abs were purchased from Sigma. Biotinylated goat anti-streptavidin Abs were purchased from Biosys (Compiegne, France).
Biosynthetic cell labeling and immunoprecipitation
Cells were cultured for 1 night, in RPMI 1640 without cysteine and methionine (ICN, Costa Mesa, CA) supplemented with 10% heat-inactivated FCS, glutamine (2 mM), penicillin/streptomycin/fungizone (1000 U/ml; 1000 µg/ml; 2.5 µg/ml), and 50 µCi of [35S]methionine cysteine (Amersham, Orsay, France). Fifty microliters of goat anti-mouse IgG-coated magnetic beads (Dynal, Oslo, Norway) were washed three times in PBS (Biomérieux, Marcy l’Étoile, France), then incubated with 15 µg of J3D3 mAb for 1 night at 4°C. Beads were washed three times in PBS. Transfected cell culture supernatants were incubated with beads for 1 night at 4°C. Washed beads were then resuspended in SDS-PAGE sample buffer for electrophoresis. Reduced and unreduced immunoprecipitates were subjected to electrophoresis in a 3% SDS agarose or acrylamide gel.
Quantitation of soluble CR1 by ELISA (CR1-ELISA)
J3D3 and J3B11 anti-CR1 mAbs recognizing different epitopes of the molecule were used to quantify soluble CR1 antigenic sites in CR1 ELISA. One hundred microliters of J3B11 mAb (1.5 µg/well in PBS, pH 7.9) was coated on polystyrene microwell plates (Nunc Maxisorp immunoplates, PolyLabo/Paul Block, Strasbourg, France) overnight at 4°C. Wells were washed three times in PBS containing 0.05% Tween-20, then saturated for 1 h at 37°C using 200 µl of PBS containing 2% BSA (Sigma) and 0.05% Tween 20. The samples containing CR1 to be quantified were diluted in PBS/0.05% Tween 20 and incubated for 2 h at 37°C. Wells were washed three times in PBS/0.05% Tween 20, then incubated with 100 µl of biotinylated J3D3 (0.05 µg/well) for 1 h at 37°C. Wells were washed again three times in PBS/0.05% BSA, then incubated with streptavidin (0.05 µg/well; Sigma) for 30 min at 37°C. After three washes, wells were incubated with 0.05 µg/well of biotinylated phophatase alkaline (Sigma) for 30 min at 37°C. After three washes, 150 µl of 4 mg/ml paranitrophenylphosphate substrate diluted in alkaline buffer (Sigma) was incubated for 30 min at 37°C. Absorbance was measured at 405 nm in each well on a microplate reader (SLT reader 340 ATCC, Salzburg, Austria); sCR1 was used as standard (a gift from E. Fischer).
Size determination of mCR1 by gel filtration
Sephacryl S-400HR (Pharmacia) was packed in a 1.6 x 70-cm column. The gel was equilibrated in 150 mM phosphate buffer, pH 7.4, supplemented with 1 M NaCl. Chromatography was performed at 4°C at a stable flow rate of 0.113 ml/min. One milliliter of culture supernatant was injected. Fractions were collected every 9 min and analyzed by ELISA for CR1 after their collection. Calibration curve of the column was established using aldolase (158,000 Da), catalase (232,000 Da), ferritin (440,000 Da), and thyroglobulin (669,000 Da). The predicted apparent m.w. for totally excluded molecules (when the elution volume was equal to the dead volume) was in accordance with manufacturer’s gel specifications.
CR1 inhibition of complement-dependant lysis of E
Human E were sensitized by incubation with a rabbit polyclonal Ab directed against human E for 30 min at 37°C. One hundred million AE were incubated for 30 min at 37°C with 100 µl of C5-deficient human serum at a 300-fold dilution in VBS containing 0.15 mM Ca2+, 0.5 mM Mg2+, and various concentrations of sCR1 or multiCR1. After two washes in VBS containing 0.15 mM Ca2+ and 0.5 mM Mg2+, AE were incubated with 100 µl of EGTA-normal human serum (NHS; diluted 1/10) for 10 min each time at 37°C. After centrifugation for 5 min at 700 x g, the 405-nm absorbance of the supernatant was determined using a microplate reader (SLT Reader 340; ATCC, Salzburg, Austria).
Molecular interaction analysis
Molecular interactions were analyzed using a surface plasmon resonance device (Fisons, Cambridge, U.K.). One hundred micrograms of goat anti-mouse Abs were immobilized on aminosilane cuvettes using an ethyl dimebhylaminopropyl carbodiimide/N-hydroxysuccinimide coupling kit (Fisons). Residual NHS groups were blocked with 1 M ethanolamine, pH 8.5, for 2 min. The cuvette was then washed in PBS buffer. One hundred micrograms of 9E10 mAb was added. After 4 min the coated surface was washed four times with PBS. Crude culture supernatant of multiCR1/scFv anti-Rh(D)-secreting cells was added. After 3 min the solid phase was washed four times in PBS. Finally, 20 ng of J3D3 mAb was added for 3 min. The solid phase was then washed three times in PBS. After washing, changes in refractive indexes, expressed as arcs per second, were correlated with the amount of protein bound to the cuvette following each incubation step.
Papain treatment of E
Papain used in this assay was obtained from Merck (Nogent-sur-Marne, France). One hundred microliters of Rh(D)-positive E was washed twice in PBS. Thirty-five microliters of a 0.1% papain solution was added to the pellet. After a 7-min incubation at 37°C, E were washed twice and resuspended in 4 ml of PBS.
Preparation of CR1 reconstituted E
Twenty microliters of a 2.5% solution of papain-treated E
was incubated for 30 min at 37°C with 50 µl of pure or diluted
culture supernatants of cells cotransfected by pKC3 CR1/C4bp and
pKC3 scFv anti-Rh(D)/C4bp.
Immunogold labeling of calcium ionophore-treated E using J3D3 anti-CR1 mAb
Two microliters of calcium ionophore-treated pellet containing E was washed twice in ice-cold PBS including 1% BSA (5 min, 780 x g). E were incubated in 50 µl of PBS/1% BSA containing 2 µg of J3D3 mouse anti-CR1 mAb for 45 min at 4°C. E were washed twice in ice-cold PBS-1% BSA, then incubated for 45 min at 4°C in 50 µl of PBS/1% BSA containing 10 nm gold-labeled goat anti-mouse Ab (British Biocell International, Cardiff, U.K.; and Tebu; 10-fold diluted). E were then washed twice in ice-cold 1% BSA/PBS. As controls, treated E and untreated E were incubated by omitting J3D3 mouse anti-CR1 mAb.
Electron microscopy
J3D3 mouse anti-CR1 mAb and 10 nm gold-labeled goat anti-mouse Ab-immunolabeled E were fixed in PBS containing 1% glutaraldehyde and 2% formaldehyde for 2 h at 4°C, then washed twice in ice-cold PBS. E were postfixed in PBS including 1% OsO4 (Merck, Rahway, NJ), then washed twice in ice-cold PBS. E were pre-embedded in 20 µl of PBS containing 5.6% BSA and 6.2% glutaraldehyde. Resuspension of pellets was carefully avoided. Pre-embedded pellets were collected; dehydrated in 70, 80, 90, and 95°C and absolute ethanol successively; then embedded in Epon (Merck) at 60°C for 48 h. Ultrathin sections (60–80 nm) were collected over copper grids. Grids were osmium tetraoxide counterstained and examined in a JEM 100C JEOL electron microscope (JEOL, Tokyo, Japan).
Indirect immunofluorescence and flow cytometry
E were analyzed for CR1 antigenic site density using indirect immunofluorescence and flow cytometry as previously described (40). Two million washed E were incubated for 30 min at 4°C with 2 µg of biotinylated J3D3 in PBS containing 1% BSA and washed three times in the same buffer. One and a half micrograms of streptavidin-PE (Tebu) or streptavidin-Alexa 568 (Molecular Probes, Eugene, OR) in 50 µl of 1% BSA/PBS, 1 µg of biotinylated anti-streptavidin Ab (Biosys), and 1.5 µg of streptavidin-PE in 50 µL of 1% BSA/PBS was then sequentially added and incubated with the cells for 30 min at 4°C. Stained E were fixed using 0.37% formaldehyde (Merck) in the washing buffer.
The kinetics of the binding of multiCR1/scFv anti-Rh(D) on E were also analyzed after 2-, 5-, 10-, 20-, and 30-min incubation in 10% IC containing serum from an SLE patient in acute flare or 10% human AB serum from healthy individuals.
Flow cytometry of stained cells was performed on a FACStarPlus apparatus (Becton Dickinson, Mountain View, CA). At least 10,000 events for each sample were collected. Mean fluorescence channel was used to quantify the staining of each sample.
CR1 distribution pattern using confocal microscopy
Labeled cells were examined by confocal laser scanning microscopy using a Bio-Rad (Hercules, CA) MRC-1024 equipped with an argon/krypton laser and fitted with the appropriate filter block for detection of Alexa 568 fluorescence. Images were taken using simultaneous dual channel scanning and were transformed into projection views by using sets of 30 consecutive single optical sections. Immunofluorescence and phase contrast images were merged. On black and white prints, immunofluorescence staining looks like dark spots.
BSA iodination
BSA was iodinated following the chloramine T method of Hunter and Greenwood (41). Briefly, 0.5 mCi of 125I (Amersham) was added to 10 µg of BSA in 20 µl of phosphate buffer (0.05 M) and 15 µg of chloramine T (Sigma). The solution was then mixed by vortexing for 30 s. Twenty micrograms of metabisulfite (Merck) was added. After mixing for 30 s, [125I]BSA was separated from free 125I and reducing agent by gel filtration chromatography (Quick Spin columns Sephadex G-50; Roche, Indianapolis, IN).
IC binding assay
Iodinated BSA was incubated in RPMI for 30 min at 37°C with rabbit anti-BSA Ab. IC were formed at a 4-fold Ab excess relative to the equivalence point.
Complement fragment incorporation (opsonization) of IC was conducted in normal human serum. Eighty million erythrocytes were incubated with 250 µl of NHS and 25 µl of IC. After incubation at 37°C for different times, cells were washed three times in ice-cold RPMI. The radioactivity of the pellet was measured in a gamma counter (Packard 1900 TR, Meriden, PA). Papain-treated E were used as a negative control.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CHO eukaryotic cells were transfected by multiCR1-containing
plasmid pKC3 using the calcium phosphate method (42),
batch-cultured in selective culture medium, checked for CR1 expression
using anti-CR1 immunofluorescence detection, then cloned by
limiting dilution. CR1 expression in cloned cells transfected by
multiCR1 pKC3 plasmid was checked by ELISA after limiting dilution
cloning. Selected clones were then subjected to increasing levels of
methotrexate up to 1.26 µg/ml to improve multiCR1 secretion by gene
duplication. Analysis of multiCR1 secreted by selected clones,
performed on agarose electrophoresis after immunoprecipitation,
revealed that in nonreducing conditions, multiCR1 was secreted as a
unique multimeric form, probably made of seven CR1 valences, as
indicated by the seven-band pattern observed in partially reducing
conditions (Fig. 2). Gel permeation
analysis of culture supernatant containing multiCR1 was performed under
nonreducing conditions showing a major homogenous peak of 1.650 kDa
(Fig. 3). Degradation products with an
apparent Mr lower than 50 kDa were
also observed. The amount of degradation products was about 15% of the
native multiCR1 peak.
|
|
The abilities of multiCR1 and sCR1 to inhibit in vitro lysis of AE
by complement were compared. A two-step complement activation system
was used to obtain quantitative data comparing mono- and multimeric CR1
for their complement inhibition properties. E were sensitized using
rabbit anti-E Ig and incubated with diluted C5-deficient serum to
moderately opsonize E through C3b deposition in the presence of either
monomeric or multimeric soluble CR1 as inhibitor. Finally, the final
membrane attack pathway of complement was activated to obtain lysis of
E, if opsonized. Both molecules showed an inhibitor activity, multiCR1
exhibited a higher inhibitory effect than sCR1, because lower
concentrations were required for both 50% and total complement
inhibition. MultiCR1 appeared to be a 5 times more potent complement
inhibitor than sCR1 when considering a 50% inhibition level and a 10
times more potent complement inhibitor than sCR1 when considering total
complement activation inhibition (Fig. 4a). MultiCR1 blocked
complement activation exogenously, as established by the lack of
complement inhibition as well as the lack of multiCR1 binding on E when
multiCR1 was incubated with E then washed before the introduction of
complement (Fig. 4b). Binding of multiCR1 on opsonized E was
also verified using flow cytometry (Fig. 4c). This binding
is in accordance with a mechanism of inhibition of hemolysis by
multiCR1 through binding to C3b and exhibits a factor I cofactor
activity.
|
Cells secreting multiCR1 were transfected by a C-terminal,
C4bp-fused, scFv anti-Rh(D)-containing plasmid (Fig. 1
). Clones
expressing the multiCR1/scFv anti-Rh(D) chimeric molecule were
selected after limiting dilution using flow cytometry and CR1
immunostaining on papain-treated E incubated with culture supernatants.
Rh(D) Ag is known to be resistant to papain treatment as well as more
accessible to Ab binding on papain-treated E. Analysis of multiCR1/scFv
anti-Rh(D) chimeric molecule by SDS-PAGE, after immunoprecipitation
using J3D3 anti-CR1 mAb, showed that this molecule is made from
both scFv anti-Rh(D) and CR1 valences (Fig. 5). Surface plasmon resonance study using
a solid phase coated with 9E10 anti-Myc tag mAb and sequential
binding of the chimeric molecule followed by J3D3 anti-CR1 mAb also
demonstrated the presence of both anti-Rh(D) and CR1 moieties in
the same chimeric molecules (Fig. 6).
|
|
Immunofluorescence staining for CR1 of treated or untreated E
revealed a cluster distribution of multiCR1/scFv anti-Rh(D) at the
surface of treated E. Although the sizes of native and recombinant CR1
clusters were similar, the concentrations of CR1 cluster on treated E
were greater than those observed on native E. Control
immunofluorescence staining for Rh(D) Ag of papain-treated E (Fig. 7a) showed homogeneous
staining, demonstrating that no cluster effect was related to Rh(D)
distribution.
|
, e
and f). CR1 antigenic sites enumeration on native and multiCR1/scFv anti-Rh(D)-treated E using flow cytometry
Quantitative flow cytometry was used to assess CR1 antigenic sites
on E. CR1 antigenic sites were quantified on native E exhibiting
various CR1 densities, papain-treated E, and native or papain-treated E
coated with multiCR1/scFv anti-Rh(D). On papain-treated E, CR1 was
totally removed, while rhesus Ags were resistant to papain treatment
and more accessible at the E surface. MultiCR1/scFv anti-Rh(D) not
only enhanced CR1 density on native E of low CR1 density from 180-2000
CR1/E, but also induced largely supraphysiologic levels of CR1 on
papain-treated E (Fig. 8).
|
IC binding study
Binding kinetics of opsonized
125I-iodinated IC on 601 CR1/E, 800 CR1/E or
papain-treated then multiCR1/scFv anti-Rh(D)-reconstituted E
were analyzed. The ability of papain-treated, multiCR1/scFv
anti-Rh(D)-coated E to bind IC was assessed using
125I-labeled IC, prepared using
[125I]BSA and rabbit anti-BSA Abs.
MultiCR1-reconstituted E bound IC better than native E, whereas E
depleted of CR1 by papain treatment were no longer able to bind IC
(Fig. 9).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Nevertheless, restoring CR1 density on E might be an attractive way to
interrupt the self-aggravating process of poor Ab response, complement
consumption, and IC precipitation that leads to vasculitis and
glomerulonephritis in SLE patients. Protein substitutive therapy must
be directly linked to E, because sCR1, although able to bind C3b or C4b
and help factor I for inactivation of C3b, cannot serve as an IC
transport and addressing system or protect E from C3b deposition when
complement activation is not fully inhibited. However, soluble CR1 is
fairly active in various inflammation animal models, such as
experimental myocardial infarction (20), acute respiratory
distress, or xenogenic heart transplantation (22), despite
the fact that rather high concentrations of sCR1 are needed and that
the sCR1 half-life in vivo is quite short
(t1/2 = 
100 min)
(52). Due to the shape of CR1, a very long single-chain
multiple SCR molecule, direct chemical cross-linking of CR1 on E was
likely to lead to a very poor yield of functional molecules. A method
to attach CR1 on E in the same three-dimensional distribution pattern
as native CR1 was sought. We considered a method to produce multimeric
CR1 and to attach these molecules to E that could prove a realistic
means of future in vivo therapy. MultiCR1 fixation through an Ag-Ab
reaction was chosen. We previously engineered an scFv anti-Rh(D) to
make it into a multimeric recombinant molecule. We chose to use it to
attach clusters of CR1 on E in the same multimeric molecule
(34).
To multimerize CR1 or scFv anti-Rh(D), a minimal length -chain
C-terminal C4bp multimerizing fragment was used that fulfilled criteria
for a multimerizing protein adapted for future in vivo use: originating
from a normal human plasma protein, devoid of any biological function,
secreted, soluble, covalently linked, and stable (34). In
particular, unlike many other multimerizing systems, the C4bp-based
system is not a complement activator and thus could be used
not only to target cells for destruction but also for substitutive
therapy, for imaging, or to provide anti-inflammatory molecules
locally. A multimeric soluble recombinant CR1 molecule was first
produced. The C-terminal part of the -chain of the C4bp is efficient
to induce polymerization during protein synthesis, and although the
expression vector codes only for a monomer, multimers are assembled in
the cell without the necessity of secondary modifications, resulting in
the secretion of a unique covalently linked soluble molecule. After
five rounds of methotrexate amplification, the multiCR1
concentration was around 2 µg/ml of culture supernatant, a commonly
observed production yield in mammalian cells. The secretion of a very
high Mr molecule (probably >1400 kDa)
suggested that there is no size limitation using the C4bp-based
multimerizing system. The perfect solubility and in vitro stability in
culture supernatant of secreted multiCR1 were noteworthy, as also
illustrated by a gel permeation curve (Fig. 3) showing a symmetrical
peak without any retardation through protein/matrix interaction in the
column. A longer lifespan of multiCR1 could also be expected, as it was
a high Mr molecule, because
albumin-receptor fused sCR1 has already demonstrated an in vivo
half-life of 297 min (52), a 3-fold increase, compared
with that of sCR1. The very high Mr of
multiCR1 precluded an accurate size determination. In 4% PAGE-SDS,
multiCR1 was detected as a single molecular species. Using agarose
electrophoresis and partial reduction, a seven-band pattern was
observed, suggesting a heptamer structure reminiscent of one of the
physiological forms of C4bp in human plasma (53). An
apparent Mr of 1650 kDa was obtained
upon gel permeation chromatography that is also in accordance with the
predicted Mr of a heptamer structure.
We are presently unable to obtain definite structure imaging by
electron microscopy due to the flexibility of the molecule and the
various conformations of multiCR1. MultiCR1 retained its cofactor
activity by enhancing C3b cleavage by factor I. In an Ig-sensitized E
model of complement activation by Ig bound to E, the
anti-inflammatory potency of multiCR1 as a complement activation
inhibitor seemed to be better than that of sCR1, suggesting a better
co-operative effect of C3b/C4b binding sites from different valences in
multiCR1. The mechanism of complement inhibition by multiCR1 was
further analyzed. Preincubation of EA with multiCR1 followed by washing
did not lead to any deposition of multiCR1 on E, demonstrating again
the lack of nonspecific interaction between multiCR1 and E (Fig. 4b). Flow cytometric analysis also showed the binding of
multiCR1 on opsonized EA, illustrating the probable mechanism of
complement inhibition by multiCR1 serving as a cofactor of factor I for
C3b inactivation.
Considering that multimeric CR1 produced using the C-terminal C4bp
multimerizing system was soluble homogenous and functional, we
addressed in a second step the anchoring of multimeric CR1 on E using
an Ag-Ab reaction. For that purpose, a heteromultimeric molecule made
of CR1 and scFv anti-Rh(D) valences was designed and produced. This
was performed by transfecting cells producing multiCR1 by a plasmid
containing the scFv anti-Rh(D) C4bp DNA construct. After
cloning and limiting dilution, clones secreting heterofunctional
multiCR1/scFv anti-Rh(D) were selected using immunofluorescent CR1
detection by flow cytometry on CR1-depleted, multiCR1/scFv
anti-Rh(D)-reconstituted E. Heteromultimeric multiCR1/scFv
anti-Rh(D) chimera retained the functional properties of both CR1
and scFv anti-Rh(D); they were able to bind E and to provide them
with the ability to capture more opsonized IC than normal E used to do.
Rh(D) Ag is an Ag of a relatively low density at E surface
(5,000–10,000 Rh(D)/E), which is poorly accessible as made of 12
transmembrane spanning with short extracellular segments
(54). However, for an experimental model we took
opportunity of the dramatic increase in Rh(D) Ag presentation by papain
treatment together with the removal of most other protein surface
molecules, including CR1, to demonstrate that E totally depleted of CR1
can be reconstituted to physiological and even supraphysiologic CR1
density following incubation with heterofunctional multiCR1/scFv
anti-Rh(D) molecules. It should be noted that the CR1 density
reached through multiCR1/scFv anti-Rh(D) treatment is not dependant
upon the natural CR1 density of a given E but, rather, upon the rhesus
phenotype of that individual; the rhesus phenotype also leads to
differences in Rh(D) density.
Electron microscopy and confocal analysis of immunostaining for CR1 clearly illustrated the cluster nature of the chimera molecule bound at the E surface, mimicking the natural conformation of CR1 at the E surface.
This heteromultimeric multiCR1/scFv anti-Rh(D) molecule was a
preliminary model designed to explore the potential of the C4bp-based
heteromultimeric molecules for therapy. It is clearly not a definite
drug for human in vivo use. Moreover, even promising data from in vitro
experiments can only be considered as a first step to the final goal of
future in vivo therapy. In addition, Rh(D) Ag is not present in every
individual and should be replaced during future development by a more
common, more expressed, and more accessible Ag, such as the H Ag of the
ABH system or another public E surface determinant expressed at a high
level on E and restricted to these cells. Furthermore, the ratio of the
two moieties of the molecule is not controlled in an / C4bp-based
heterochimera. In the future, developments will include use of the
C-terminal part of both and ß C4bp chains or use of a modified
-chain to control the ratio of two components in heterochimeric
molecules and to modulate the polymerization process. Production
system, vectors, leader sequences, and junction area at the 5' end of
the C4bp multimerizing fragment should also be optimized to meet the
needs of high level production.
Nevertheless, the present research model established the feasibility of heteromultimeric molecules even of high Mr that remain soluble, functional, and able to bind specifically to an Ag-displaying cell surface. Improved anti-E/multiCR1 molecules could be used in the future for in vivo therapy in SLE or HIV-infected patients. Finally, this concept and the C4bp-based multimerizing system could be used in many domains when targeting of a specific molecule on a given cell type is needed, particularly when damages to the targeted cells are to be avoided.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Prof. J. H. M. Cohen, Hopital R Debré, Laboratoire d’Immunologie, 51100 Reims Cedex, France. E-mail address:
3 Abbreviations used in this paper: E, erythrocyte; anti-Rh(D), anti-rhesus(D); DHFR, dihydrofolate reductase; AE, activated erythrocyte; C4bp, C4 binding protein; IC, immune complexes; multiCR1, multimeric CR1; multiRH1, multimeric scFv anti-Rh(D); NHS, normal human serum; scFv, single chain Fv; sCR1, soluble CR1; SCR, short consensus repeat; SLE, systemic lupus erythematosus; VBS, veronal-buffered saline; MFC, mean fluorescence channel.
Received for publication March 29, 1999. Accepted for publication November 17, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. A. Ogun, L. Dumon-Seignovert, J.-B. Marchand, A. A. Holder, and F. Hill The Oligomerization Domain of C4-Binding Protein (C4bp) Acts as an Adjuvant, and the Fusion Protein Comprised of the 19-Kilodalton Merozoite Surface Protein 1 Fused with the Murine C4bp Domain Protects Mice against Malaria Infect. Immun., August 1, 2008; 76(8): 3817 - 3823. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Whitsett, C. J. Bachurski, K. C. Barnes, P. A. Bunn Jr., L. M. Case, D. N. Cook, D. Crooks, M. W. Duncan, L. Dwyer-Nield, R. C. Elston, et al. Functional Genomics of Lung Disease Am. J. Respir. Cell Mol. Biol., August 1, 2004; 31(2/S1): S1 - S81. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zorzetto, I. Ferrarotti, R. Trisolini, L. L. Agli, R. Scabini, M. Novo, A. De Silvestri, M. Patelli, M. Martinetti, M. Cuccia, et al. Complement Receptor 1 Gene Polymorphisms Are Associated with Idiopathic Pulmonary Fibrosis Am. J. Respir. Crit. Care Med., August 1, 2003; 168(3): 330 - 334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zorzetto, C. Bombieri, I. Ferrarotti, S. Medaglia, C. Agostini, C. Tinelli, G. Malerba, N. Carrabino, A. Beretta, L. Casali, et al. Complement Receptor 1 Gene Polymorphisms in Sarcoidosis Am. J. Respir. Cell Mol. Biol., July 1, 2002; 27(1): 17 - 23. [Abstract] [Full Text] [PDF]
|
|
)
and multiCR1 (•). EA were incubated 30 min at 37°C with
C5-deficient human serum at a 300-fold dilution and with different
concentrations of sCR1 and multiCR1. After two washings, EA were
incubated with EGTA-NHS for 10 min at 37°C. E lysis was assessed by
spectrophotometry at 405 nm. Flow cytometric analysis of multiCR1
binding on EA. b, Control staining omitting the
opsonization step. c, Binding of multiCR1 on opsonized
EA.
) or 800 CR1/E (
), multiCR1/scFv anti-Rh(D)
reconstituted E (
), and papain-treated control E (•) were
incubated for various periods of time with IC and NHS at 37°C.