Immunoregulatory functions of complement: structural and functional studies of complement receptor type 1 (CR1; CD35) and type 2 (CR2; CD21).
Explore the source record for details and available documents.
SEARCH · Search PubMed
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The interaction of the Epstein-Barr virus/45-kDa proteolytic fragment of C3 (C3dg) receptor (CR2) with its viral ligand, the Epstein-Barr virus glycoprotein gp350/220, initiates the sequence of events leading to virus internalization and B lymphocyte transformation. Soluble recombinant receptor (rCR2) and gp350/220 as well as the natural ligand, C3dg, were subjected to a number of analytical techniques including gel permeation chromatography, density gradient ultracentrifugation, circular dichroism, and electron microscopy in order to determine their hydrodynamic, structural, and binding properties. Both rCR2 and gp350/220 were found to be highly extended proteins (f/fo = 2.1 and 2.4/2.2, respectively). C3dg, in contrast to the viral ligand, is only somewhat elongated (f/fo = 1.5). Soluble rCR2, visualized by high resolution electron microscopy, was shown to be an extended, highly flexible molecule comprised of ringlet domains, each approximately 24.1 A in length, which likely correspond to the short consensus repeat motif deduced from the CR2 cDNA nucleotide sequence. Ligand-binding studies carried out under physiological conditions indicated that gp350/220 binding to rCR2 was saturable and univalent, with a dissociation constant of 3.2 nM. In contrast, monomeric C3dg did not bind to rCR2 under physiological conditions; however, at reduced ionic strength, monomeric C3dg binding could be measured. These studies indicate that the affinity of the C3dg monomer for rCR2 under physiologic conditions is approximately 10(4)-fold less than that of the viral ligand. The molecular properties of rCR2 revealed in these studies provide essential information for future studies of the biologic functions of the Epstein-Barr virus/C3dg receptor.
The human C3d/Epstein-Barr virus receptor (CR2/CD21) is a 145-kDa protein primarily expressed on mature B lymphocytes. CR2 is a member of the regulators of complement activation (RCA) gene family found on band q32 of chromosome 1. The RCA proteins are characterized by the presence of 60-70 amino acid short consensus repeats (SCR). A full length CR2 cDNA was cloned and used to identify overlapping cosmid genomic clones. Analysis of CR2 exon-intron junctions revealed the presence of three types of exons in the short consensus repeat region of CR2. First, four exons each of which encodes two SCR are present. Five exons encode a single SCR. Six exons encode SCRs which are split in identical positions. The order of these types of exons is in a repeated array of four SCRs, indicating that the contemporary CR2 gene likely evolved from a more primitive gene containing four SCRs. The CR2 full length cDNA clone was used to find restriction fragment length polymorphisms (RFLPs). Restriction enzyme TaqI generated 2.55- and 2.10-kilobase (kb) polymorphic bands. This RFLP was mapped near the exon containing the first two SCRs. HaeIII digestion generated polymorphic bands of 1.45, 1.55, and 1.75 kb. The HaeIII 1.45-kb RFLP band maps near the exon containing the 15th SCR. The TaqI and HaeIII RFLPs will provide tools for the genetic analysis of CR2. The organization of the CR2 gene provides insights into the evolution of human CR2 and the RCA gene family.
The gene encoding human complement receptor 2 (CR2) lies within a cluster of genes on human chromosome 1. Polymorphisms have been described for several of these genes or gene products. To examine any possible polymorphisms in the CR2 gene product we have analyzed the CR2 transcriptional products for variations via alternative exon usage and for the presence of different CR2 alleles within the human population. This analysis has suggested that an exon encoding a 60 amino acid short consensus repeat can be alternatively spliced. Evidence for two transcriptional products has been found in a variety of transformed human B cells and two human tonsils. Interestingly the ratio of the two forms of mRNA, that which includes the alternative exon and that which does not, is not constant when examining the RNA from these sources. In addition to the variation introduced by the alternatively spliced product, there appears to exist in the human population at least three different alleles for CR2. These alleles have been defined by multiple nucleotide changes which are not silent but which alter the amino acid encoded at that site. However, we have not identified allelic variants of CR2 which would produce a product varying in the number of long homologous repeats, as has been identified for the various alleles of the closely related protein, human complement receptor 1.
The effect of ligand interactions with the C3d/C3dg complement receptor (CR2) on proliferation of human B lymphoblastoid cells was investigated by using cell cultures performed at low density (1 to 1.5 x 10(3) cells/ml) in a serum-free defined medium to which only transferrin had been added. This medium does not allow proliferation of Raji cells which die within 48 hr with formation of polykaryons. Addition of purified human C3 to the cultures resulted in a dose-dependent proliferation of the cells. A steady growth of Raji cells with a doubling time of 36 hr was observed in cultures containing 10 micrograms/ml of C3. A growth rate similar to that observed in the presence of native C3 was found in the presence of equimolar concentrations of purified C3dg but not of C3c. F(ab')2 anti-C3d but not F(ab')2 anti-C3c antibodies inhibited the mitogenic effect of C3. Preincubation of Raji cells with monoclonal antibody OKB7 which directly inhibits the binding of C3dg to CR2, totally suppressed C3-induced growth of the cells. C3 did not enhance growth of the T lymphoma-derived cell line JM and monocytic cell line U937 which do not express CR2. These results provide direct evidence that the interaction between CR2 and C3 fragments stimulates proliferation of human cells of the B lineage. Because CR2 also acts as a receptor for Epstein-Barr virus on B cells, our results may pertain to the B cell mitogenic properties of the virus.
Several mouse monoclonal IgG antibodies (AB1, AB2, AB3, and AB5) were developed that reacted with a 140,000 mol wt glycoprotein on the surface of cultured RAJI B lymphoid cells. The antibodies reacted with purified normal human peripheral blood B cells and CLL Ig+ B cells and showed specific germinal center and mantle zone staining in tissue sections of secondary lymphoid organs. Immunodepletion studies using 125I surface-labeled Raji cell membrane antigens demonstrated that the antigen identified by AB5 is the same 140,000 mol wt glycoprotein detected by anti-B2 that has recently been shown to react with the C3d fragment or CR2 receptor. (Iida et al: J Exp Med 158:1021, 1983). Addition of the AB series and anti-B2 monoclonal antibodies to cultures of purified human peripheral blood B cells resulted in the uptake of 3H-thymidine at two to six times background control levels provided that irradiated autologous T cells were added to the culture. Stimulation was not evoked by other monoclonal antibodies to B cell surface molecules (ie, B1, BA-1, BA-2, and HLA-DR). Pepsin-generated F(ab')2 fragments of anti-CR2 antibodies were essentially as effective as the intact IgG molecule in stimulating B cells. Induction of B cell proliferation by antibody binding to CR2 suggests that the C3d receptor may have an integral role in regulation of humoral immune response.
In the majority of 188 non-Hodgkin lymphomas (NHL) investigated in this study, we found a simultaneous expression of the receptors for c3b and c3d (CR1 & CR2; cccorr = 0.69, P less than 0.0005). An analysis of the different histological entities of the Kiel classification revealed that this coexpression was most pronounced for germinal centre-derived (cccorr = 0.63, P = 0.0004) and immunocytic NHL (cccorr = 0.86, P = 0.0024), whereas in chronic lymphocytic leukaemias there tended to be a more heterogeneous pattern of complement receptor (CR) expression (cccorr = 0.29, P greater than 0.10). In contrast to these NHL of mid B cell stage, most of the NHL of early (i.e. acute lymphocytic leukaemia, lymphoblastic NHL) and late B cell stage (i.e. hairy cell leukaemia, immunoblastic NHL, multiple myeloma) did not express either of these receptors. CR positive NHL often showed a follicular arrangement of the neoplastic cells and had higher numbers of T helper/inducer (T4) lymphocytes (PCR1 less than 0.00005, PCR2 less than 0.05). Some cases of mid B cell NHL and all cases of hairy cell leukaemia reacted with antibodies against CR3 (i.e. the ic3b receptor).
Neutrophils have been shown to express a receptor for C3dg that is distinct from CR2 and is termed complement receptor type 4 (CR4). In the present study, other peripheral blood cell types were examined by indirect immunofluorescence and flow cytometry for the presence of C3dg binding activity. Specific uptake of C3dg occurred with neutrophils, platelets, and B lymphocytes, but not with eosinophils or T lymphocytes. Monocytes, contained within a mixed cell population of peripheral blood mononuclear cells and platelets, also bound C3dg, whereas purified monocytes did not. Binding of 125I-labeled glutaraldehyde-cross-linked C3dg to platelets was saturable, with an average of 1940 C3dg molecules bound per platelet at saturation (n = 8), ranging in number from 660 to 3930 molecules bound. Activation of platelets with thrombin did not consistently cause an increase in the expression of CR4 sites. 125I-C3dg binding to platelets was competitively inhibited equally well by unlabeled C3dg and iC3b, and approximately fourfold less well by C3b. The addition of platelets to elutriated monocytes generated C3dg binding activity on these cells by the formation of platelet-monocyte complexes. Thus, the CR4 on platelets accounted for the C3dg binding activity initially observed with partially purified monocytes. The adherent property of platelets may enable them to confer on certain other cell types the ability to localize C3dg-coated immune complexes or particles.
Human cDNA probes encoding the C3b/C4b complement receptor, CR1, have been used to identify, in the mouse, two new genes which are related to CR1 but which appear to encode a different protein product. These new mouse genes, arbitrarily designated mouse genes X and Y, hybridize specifically to three different cDNA probes derived from human CR1. The degree of hybridization homology between the mouse X and Y genes suggests they are very closely related to one another; however, the chromosomal localization of the mouse X gene to chromosome 8 and the mouse Y gene to chromosome 1 indicates they are distinct gene sequences. The mRNA species detected with the X and/or Y (X/Y) sequences are approximately 2000 bases in length, but vary in both quantity and size depending upon the tissue analyzed. DNA sequence analysis of a cDNA specific for the X and Y sequences indicates the mature protein(s) will contain the 60 amino acid consensus repeat characteristic of a group of other proteins including CR1, the C3d receptor (CR2), H, C4 binding protein (C4bp), the interleukin 2 (Il 2) receptor and others. The identity of the mouse X and Y genes, and the function of the proteins which they encode, is not known; however, the small size of the mRNA and the tissue specific expression suggests they do not encode mouse CR1 or CR2 but instead encode a related protein (or proteins) which is expressed in a wide variety of mouse tissues.
We have investigated the function of C3b receptor (CR1) in the malignant lymphocytes of B-chronic lymphocytic leukemia (B-CLL) mimicking the physiological ligand C3b with the anti-CR1 monoclonal antibody CB04 covalently linked to Sepharose CL-4B (CB04-S). The binding of insolubilized CB04-S to CR1 gave a progression signal to B-CLL cells which became B cell growth factor (BCGF)-responsive. The cells of 13 of 14 cases treated with CB04-S showed an active time-dependent proliferation when BCGF was added to the culture. After 72 hr of exposure to BCGF, the growth fraction evaluated with the Ki67 monoclonal antibody was 23.4 +/- 8.9 and the proportion of cells in S phase assessed by the bromodeoxyuridine incorporation technique was 18.6 +/- 8.5%. The proper sequence of CB04-S followed by BCGF was also important since the proliferation was halved when the sequence was reversed or the two signals were delivered concomitantly. CB04-S and BCGF alone failed to induce any significant proliferation; the percentage of cycling cells was less than 1% overlapping that of control culture cells. On the contrary, the proliferation of normal tonsil B cells was triggered both by CB04-S and by BCGF used as single agents (bromodeoxyuridine+ cells 12.7 +/- 5.1% and 20.0 +/- 7.3, respectively). Together these data indicate that malignant B-CLL cells need a sequential two-step signal based upon CR1 binding in order to be activated in vitro. This is a major difference with normal tonsil B lymphocytes whose proliferation is triggered both by CB04-S and by BCGF used as single agents.
CD22 and CD21 are glycoproteins primarily expressed on normal and neoplastic human B cells. The surface expression of these two molecules parallel each other during normal B cell differentiation, and the reported relative mobilities for CD22 and CD21 are 130/140 kDa and 140 kDa, respectively. Herein we present a detailed analysis of the biosynthesis and structure of CD22 and also compare it directly to CD21. Electrophoresis under reducing and nonreducing conditions suggested that CD22 and CD21 may have similarities in intra-chain disulfide bond formation. Biosynthesis and processing of CD22 and CD21 were very similar with respect to kinetics and post-translational modification, and both could be phosphorylated. However, endoglycosidase digestion (using N-glycanase and endoglycosidase H) and peptide mapping (using V8 protease and N-chlorosuccinimide) strongly suggested that CD22 and CD21 are distinct gene products.
We sought biochemical evidence for a role of C receptors types 1 (CR1) and 2 (CR2) in B cell activation. A flow cytometer was used to measure the fluorescence of tonsillar cells that had been loaded with the calcium-dependent indicator indo-1, and cells were stimulated by cross-linking cell-bound DA4.4 anti-IgM, Yz-1 anti-CR1 or HB5 anti-CR2 with goat anti-mouse IgG. There was a direct dose-response relationship between the proportion of cells having increased cytoplasmic free calcium concentration (Cai) after addition of second antibody and the amount of cell-bound Fab' DA4.4. In contrast, no rise in Cai was observed after cross-linking bound Yz-1 or HB5. To determine whether CR1 or CR2 could modify the increase in Cai induced by cross-linking membrane IgM, Cai was monitored after addition of second antibody to cells bearing combinations of either Yz-1 or HB5 with a limited amount of DA4.4. The combination of Yz-1 with DA4.4 yielded little or no further increase in the percentage of cells responding to cross-linking with elevated Cai compared with DA4.4 alone. However, the combination of HB5 with limited DA4.4 synergistically enhanced this response, resulting in stimulation that was equivalent to that obtained with optimal concentrations of DA4.4. The synergistic effect of CR2 was also observed with avidin as the cross-linking reagent for bound biotinylated HB5 and DA4.4, occurred in the presence of EGTA, and did not require T cells. Studies of the proliferation of B cell-enriched PBMC demonstrated that, whereas HB5 coupled to Sepharose alone induced little or no DNA synthesis, the combination of HB5 with limited DA4.4 on Sepharose induced a dose-related synergistic increase in the incorporation of [3H]thymidine.
The density of the C3d,g/Epstein-Barr virus receptor (CR2) on human mononuclear cells (MNC) isolated from CPD stabilized blood, was determined by comparing the concentration of CR2 antigen in extracts from Raji cells, with a known receptor density, with extracts of MNC. The concentration of CR2 was determined by a sandwich ELISA using two mutually non-inhibiting monoclonal anti CR2 antibodies (HB5 and OKB7). The percentage of CR2+ MNC was determined by an immunocytochemical staining method. Analysis of MNC from 20 healthy donors, revealed that the density of CR2 molecules varied from 3,600-7,700 on CR2+ cells, and that the percentage of CR2+ MNC had a mean of 13%.
The effect of complement fragments C3b and C4b, on the triggering of antigen-specific human T-cell clones by Epstein-Barr virus-transformed human lymphoblastoid B cells (LCL) when these fragments are covalently coupled to the antigen tetanus toxin (TT) is described. TT was chemically cross-linked to purified C3b [(TT-C3b)n], C4b [(TT-C4b)n] or bovine serum albumin [(TT-BSA)n] as a control. T-cell activation was quantified by tritiated thymidine incorporation and 51Cr release. (TT-C3b)n and (TT-C4b)n induced proliferative responses comparable to (TT-BSA)n but at 18-25 and 4-6 lower concentrations, respectively. This enhancing effect required the covalent cross-linking of the complement fragments to the antigen and involved intracellular processing of the latter by LCL. Antigen presentation was similarly enhanced when measuring the cytotoxic activity of a helper T-cell clone against LCL previously pulsed with (TT-C3b)n or (TT-C4b)n compared with (TT-BSA)n. Binding studies, carried out on LCL using TT radiolabelled with 125I before cross-linking, indicated that (TT-C3b)n and (TT-C4b)n gave three- to four-fold more binding than (TT-BSA)n. Addition of antibodies against CR1 and CR2 or proteolytic removal of these complement receptors with trypsin inhibited by about 60% the enhancing effect of TT-bound C3b and C4b in both binding and functional assays. These results indicate that binding of C3b or C4b to antigen enhances antigen-specific proliferative and cytotoxic responses of T cells by targeting opsonized antigen onto complement receptors CR1 and CR2 of LCL. The putative significance of these findings in terms of regulation of immune responses by complement is discussed.
Complement component C3 and receptors for its activated fragments have been recognized as regulators of humoral immune responses. Cross-linked C3b and C3d will substitute for alpha B cell growth factors (alpha BCGF) produced by macrophages by stimulating activated murine B lymphocytes to enter S phase. In contrast, soluble C3d and synthetic oligopeptides--containing the amino acid sequence of the binding site of C3d to its receptor, the complement receptor type 2--inhibit the entry into S phase of alpha BCGF-stimulated B cells. There is evidence supporting the hypothesis that a portion of the C3 gene at the 3' end, encoding parts of the alpha chain, encodes an alternate C3 product with alpha BCGF activity.