Search

The brain is largely protected from damage due to infection, trauma, and aberrant processes by the innate immune system. These studies were undertaken to determine whether neurons in normal brains constitutively express complement components. In situ hybridization and immunohistochemical studies with specific riboprobes and antibodies, respectively, revealed that most hippocampal neurons, many pyramidal cortical neurons and cerebellar Purkinje neurons in normal murine brains constitutively express C3, C5 and C6. The constitutive expression by neuronal subsets of components of the complement activation and membrane attack pathways suggests that the complement system represents a "first line" of host defense in the brain.

Large numbers of neuritic plaques (NP), largely composed of a fibrillar insoluble form of the beta-amyloid peptide (Abeta), are found in the hippocampus and neocortex of Alzheimer's disease (AD) patients in association with damaged neuronal processes, increased numbers of activated astrocytes and microglia, and several proteins including the components of the proinflammatory complement system. These studies address the hypothesis that the activated complement system mediates the cellular changes that surround fibrillar Abeta deposits in NP. We report that Abeta peptides directly and independently activate the alternative complement pathway as well as the classical complement pathway; trigger the formation of covalent, ester-linked complexes of Abeta with activation products of the third complement component (C3); generate the cytokine-like C5a complement-activation fragment; and mediate formation of the proinflammatory C5b-9 membrane attack complex, in functionally active form able to insert into and permeabilize the membrane of neuronal precursor cells. These findings provide inflammation-based mechanisms to account for the presence of complement components in NP in association with damaged neurons and increased numbers of activated glial cells, and they have potential implications for the therapy of AD.

To identify some of the structural features determining specific protease recognition of complement components C3 and C4, we used site-specific mutagenesis to construct mutants of murine C3 that are cleaved by the C4-specific C1-s protease. Insertion of three amino acid residues corresponding to residues at the C1-s cleavage site of human C4 into murine C3 at the analogous C3 convertase cleavage site was adequate to render the mutant protein susceptible to C1-s cleavage. In addition, insertion of C3-specific residues at the same site or introduction of the C4-specific residues as substitutions rather than as an insertion also rendered the site susceptible to cleavage, but with 10- to 50-fold lower efficiencies, and insertion of even a single amino acid residue affected recognition by C1-s. Finally, insertion of amino acid residues into mC3 partially inhibited cleavage by the alternative-pathway C3 convertase, with insertion of C3- or C4-specific residues giving about the same level of inhibition. A simple interpretation of these data is that C1-s cleavage is dependent primarily on steric accessibility and on recognition of specific amino acid residues at the cleavage site, whereas C3 convertase cleavage is dependent primarily on specific interactions distal to the cleavage site, with only relatively weak, non-C3-specific interactions at the cleavage site itself.

Inactivation of C3 by enzymatic cleavage, nucleophilic addition, or slow freezing and thawing resulted in the acquisition of similar end-state conformations as judged by near-UV circular dichroism. Although inactivation by the two nonenzymatic processes involves no peptide bond scission, the inactivated C3 resembled C3b in that it possessed a free sulfhydryl group not present in the native protein and an increased surface hydrophobicity as evidenced by enhanced binding of the fluorophore 8-anilino-1-naphthalensulfonate (ANS). The C3b-like functional properties of modified C3 [Pangburn, M. K., & Müller-Eberhard, H. J. (1980) J. Exp. Med. 152, 1102-1114] may thus be understood in terms of the similarity of its conformation to that of C3b. The rate of the conformational change following proteolytic cleavage was fast and appeared to be limited by the rate of the enzymatic reaction. In contrast, the rate of conformational change following addition of methylamine was slow and rate limited by the conformational rearrangement itself, not by the chemical modification. A kinetic analysis of the changes in circular dichroism and ANS fluorescence enhancement suggested that the nucleophilic addition was spectroscopically undetectable and was followed by a minimally biphasic, spectroscopically demonstrable conformational rearrangement. The appearance of C3b-like functional activity in nucleophile-modified C3 largely parallels the time course of the spectroscopically detectable conformational change but is distinctly slower than the rate at which hemolytic activity is lost. While fully transconformed methylamine-inactivated C3 can bind factor B and is susceptible to cleavage by C3b inactivator and its cofactor beta 1H, this cleavage occurs at a substantially slower rate than the equivalent process in C3b. The implications of these findings in terms of the mechanism through which the alterative pathway of complement is initiated are discussed.

Antibody-independent activation of the alternative C pathway by human lymphoblastoid cell lines latently infected with EBV has been recognized for some time, although the mechanisms involved and the specific cell surface molecule(s) recognized by the C system have not been identified. The present studies, carried out with the purified proteins of the alternative pathway have addressed these questions. Activation of the purified proteins of the alternative pathway by Raji lymphoblastoid cells was found to be antibody independent, confirming earlier findings with serum. Surprisingly, activation was highly dependent on properdin. In other models properdin has been found to augment alternative pathway activation and to be required for lysis of virus infected cells. Molecules which activate the alternative pathway provide binding sites on which C3 breakdown by regulatory proteins is impeded; therefore intact C3b accumulates on the activator. Immunoprecipitation studies with either anti-CR2 or anti-C3 have identified CR2, the R for C3d,g and EBV, as a major covalent and noncovalent binding site for C3 deposition on Raji cells during alternative pathway activation. Covalently bound C3b was dissociated from CR2 by hydroxylamine, indicating attachment via an ester bond. C3b binding after activation was not reduced by an anti-CR2 mAb which blocks CR2 R function, indicating that it was probably not mediated by C3d,g R epitopes on CR2. Direct confirmation of the ability of CR2 to trigger the alternative pathway came from studies with purified CR2 which was found to activate the alternative C pathway in serum or in mixtures of the purified proteins of the pathway. This work provides conclusive evidence that CR2 is a C activator and functions in this capacity on Raji cells.

Reduction and alkylation of C3 in nondenaturing buffers by 2 mM DTT (1 hr, 23 degrees C) (C3 R/A) results in a > 99% loss of C3 hemolytic activity. In contrast, sham alkylated protein, or protein allowed to reoxidize from the mildly reduced state, showed full activity. The loss in lytic activity correlated with the alkylation of approximately 6 liberated sulfhydryls, all of which arose from intrachain disulfides in the alpha polypeptide chain of C3. Data from CD experiments indicate that these bonds are essential for maintaining the native conformation of the protein. The loss of hemolytic activity of C3 R/A was neither due to its inability to be cleaved by C3 convertases nor to a dysfunction of the cleavage-induced labile membrane binding site as cells bearing C3 convertases firmly bound C3 R/A to an extent 80% that of native C3. The hemolytic defect in C3 R/A was found to result from its inability, once deposited on C42-bearing cells, to bind C5 and thereby participate in the formation of a C5 convertase. Unlike the effect on hemolytic activity, C3 R/A showed only a partial loss (3-fold) in its ability to bind to C3 receptors on monocytes. Taken together, these studies differentiate in molecular terms several of the functional sites in C3 and stress the significance of intra-alpha-chain disulfides in maintaining the conformation in activated C3 required for binding of C5.

Serum incubated with purified EBV was found to contain C3 cleavage fragments characteristic of C3c. Since the cofactors necessary for such cleavage of C3b by factor I are not normally present in serum, EBV was tested for factor I cofactor activity. Purified EBV from both human and marmoset EBV-producing cell lines was found to act as a cofactor for the factor I-mediated breakdown C3b to iC3b and iC3b to C3c and C3dg. EBV also acted as a cofactor for the factor I-mediated cleavage of C4b to iC4b and iC4b to C4c and C4d. EBV from both the human and marmoset cell lines accelerated the decay of the alternative pathway C3 convertase. The classical pathway C3 convertase was unaffected. Multiple lines of evidence eliminated the possibility that marmoset or human CR1 was responsible for the functional activities of EBV preparations. The spectrum of activities was different from CR1 in that EBV and EBV-expressing cell lines failed to rosette with C3b or particles bearing C3b, the primary functional assay for CR1, and EBV did not accelerate classical pathway C3 convertase decay, another property of CR1. In addition, CR1 could not be detected immunologically on marmoset or human EBV-expressing cells and mAbs to CR1 failed to alter EBV-produced decay acceleration and factor I cofactor activities, although the antibodies blocked the same CR1-dependent functional activities. The multiple complement regulatory activities exhibited by purified EBV derived from human and marmoset cells differ from those of any of the known C3 or C4 regulatory proteins. These various activities would be anticipated to provide survival value for the virus by subverting complement- and cell-dependent host defense mechanisms.

The 145-kDa molecule that has been identified as the C3d receptor CR2 was isolated from lysates of Raji cells by affinity chromatography by using the monoclonal antibody (MoAb)HB-5. The purified protein was incorporated into 14C-phosphatidylcholine liposomes by deoxycholate dialysis followed by flotation on discontinuous sucrose gradients. Incorporation of the receptor was verified by testing the gradient fractions for CR2 by an enzyme-linked immunosorbent assay. Liposomes were shown to be unilamellar vesicles ranging in diameter from 25 to 100 nm by electron microscopy. The external orientation of CR2 in the membranes was demonstrated by immunoelectron microscopy. The functional activities of liposomes containing CR2 and liposomes without protein were compared. CR2 liposomes bound to EC3d, but not to E, and this binding was inhibited by the anti-CR2 MoAb OKB7 and by a MoAb specific for C3d. Control liposomes failed to bind to either E or EC3d. The ability of CR2 to function as a receptor for Epstein Barr virus (EBV) was tested in two ways. First, CR2 liposomes bound to B95-8, a cell line expressing EBV membrane antigens, but not to B95-8 cells treated with the viral DNA polymerase inhibitor phosphonoformic acid. Second, liposomes containing CR2 were shown by ultracentrifugal analyses to bind directly to purified EBV, and this binding was also inhibited by OKB7. Control liposomes did not bind to B95-8 cells or to EBV. These findings show that CR2 purified from detergent extracts of Raji cells can be reconstituted into lipid membranes with maintenance of its dual functions as a receptor for C3d and EBV.

A considerable body of evidence from this and other laboratories indicates that complement receptor type 2 (CR2) modulates B cell activation and growth. In the present studies we have examined the effects of three different types of CR2 ligands, i.e., monomeric, aggregated, and latex-bound C3dg; mAb to different CR2 epitopes; and UV-inactivated, non-transforming EBV (EBVUV) for their actions on highly purified, high density resting tonsil B cells. Although none of these ligands induced B cells to enter the cell cycle or synergized with either anti-mu or low m.w. B cell growth factor in triggering B cell mitogenesis, aggregated C3dg, latex-bound C3dg, the OKB7 anti-CR2 mAb, and EBVUV-enhanced thymidine incorporation by phorbol ester-activated tonsil B cells. Such enhancement was not T cell or monocyte dependent. The major action of the CR2 ligands thus seems to be to enhance the transition of B cells activated by certain stimuli from the G1 to the S phase of the cell cycle. In contrast to the action of aggregated and latex-bound C3dg, monomeric C3dg was inhibitory for phorbol ester and aggregated C3dg-induced B cell activation. The HB-5 anti-CR2 mAb, which reacts with a different epitope on CR2 from that of OKB7, did not synergize with PMA in B cell activation. These data provide additional evidence for a role for the CR2 in the control of B cell growth and provide a useful model for studying the CR2-mediated signals that affect the growth of B cells.

In pursuing studies on the early events in the infection of human B cells by Epstein-Barr virus (EBV), we examined the host cell attachment phase with a panel of B-cell-specific monoclonal antibodies. One of the monoclonal antibodies, OKB7, directly blocked the attachment of purified EBV to B lymphocytes in the absence of a second anti-immunoglobulin antibody and thereby prevented EBV infection of tonsil and peripheral blood B cells. Although earlier studies have shown a close association of the EBV and complement receptor (CR2), an anti-CR2 monoclonal antibody, anti-B2, did not directly block the binding of EBV to B cells. A comparison of the structures recognized by these monoclonal antibodies on various cell types and their functional and physiochemical properties was undertaken. Flow cytometric analysis revealed that the molecules detected by OKB7 and anti-B2 were coexpressed to the same extent on B cells but were not expressed on T-cell lines. OKB7 and anti-B2 both immunoprecipitated a 145,000-molecular-weight membrane protein with an isoelectric point of 8.2 from membrane extracts of Raji lymphoblastoid cells. OKB7 and, to a lesser extent, anti-B2 directly blocked the attachment of C3d,g-coated fluorescent microspheres and sheep erythrocytes bearing C3d to B cells, indicating that these antibodies also react with CR2. These studies indicate that the EBV-CR2 receptor is a single membrane glycoprotein which possesses multiple antigenic and functional epitopes.