Young-Ju Jang

The structural basis for DNA binding by an anti-DNA autoantibody.

We have used single and multiple site-directed mutagenesis, and molecular modeling, to identify critical residues in the DNA binding site of MAb 2C10, an IgG anti-dsDNA autoantibody from an MRL/lpr lupus mouse. Simultaneous replacement of four Arg residues in the CDR3H abolished binding activity. With one exception, replacement of any one of these Arg residues reduced the activity to 20-50% of the unmutated scFv activity. Arg to Asp replacements had a slightly greater effect than Arg to Ala replacements. In the one exceptional case, replacement of Arg99 with Ala actually increased DNA binding five-fold and replacement by Asp had little effect. Mutation of Phe32 and Asn35 to A1a in CDRIH decreased DNA binding, whereas replacement of Arg31 with A1a had negligible effect. Ala substitution of any one of a cluster of Asp residues in CDR1L increased DNA binding three to six-fold, confirming previous findings that the L-chain of MAb 2C10 is not favorable for DNA binding. The L-chain does participate in shaping the selectivity of antigen binding, and mutation of CDR3L residue Asp92 or Asn93 caused a decrease in DNA binding activity. Directed mutagenesis, consistent with a molecular model, indicates that: several CDR amino acids contribute to DNA binding, without one residue dominating; both VH and VL CDR3 domains contribute to specificity of binding whereas the CDR1L hinders DNA binding. The results suggest a significant role for electrostatics in the interaction of DNA with MAb 2C10.

Anti-DNA antibodies: aspects of structure and pathogenicity

Abstract. Anti-DNA antibodies contribute to the pathology of systemic lupus erythematosus. Their depositon in tissue lesions could result from localization of preformed immune complexes of antibodies with DNA or nucleosomes, or from cross-reaction of anti-DNA antibodies directly with tissue proteins. Structural analyses contribute to understanding their pathogenic potential. Primary structures of lupus immunoglobulin G double-stranded DNA-binding autoantibodies are determined by immunoglobulin genes with mutated variable region segments, indicative of selection by immunizing antigen. Arginine, lysine and asparagine residues in complementarity-determining region favor DNA binding. Heavy-chain variable regions make major contributions to DNA binding; affinity and specificity of binding are modulated or can be abrogated by the light-chain variable domain. Crytallographic structure is known for a few antibody-DNA complexes and several ligand-free Fab fragments. Computer modeling supplements this limited information. Structural information of lupus antibody interactions with both DNA and cross-reacting molecules will support use of ligands to inhibit tissue deposition of the antibodies and prevent lesion formation in lupus.

Analysis of the phenotypes of Jurkat clones with different TRAIL-sensitivities.

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) has been shown to exert potent cytotoxic activity against many tumor cells but not normal cells. However, some tumor cells are resistant to TRAIL, and it has not been determined how this occurs. In the present study, we obtained three subgroups of Jurkat clones with TRAIL-sensitive, -partial resistant and -resistant phenotypes. We found that most TRAIL-resistant and -partial resistant clones expressed low levels of DR5, whereas most TRAIL-sensitive clones expressed high levels of Death Receptor (DR5). However, there were clones with a range of different TRAIL-sensitivities that had similar levels of DR5 expression. The expression levels of DR4 and the decoy receptors, DcR1 and DcR2, did not correlate with TRAIL sensitivities. We also compared the subgroups in terms of the expression of Fas-associated death domain protein (FADD), the levels of activation of Receptor Interacting Protein (RIP) and caspases, and cleavage of Poly (ADP-Ribose)Polymerase (PARP). Basal expression levels of FADD were not significantly different among the subgroups. After treatment with TRAIL, both TRAIL-sensitive and partial resistant clones showed high levels of activation of caspase-3, caspase-8, RIP and PARP. Relative basal level and induced level of Phosphoprotein over Expressed in Diabetes/Phosphoprotein Enriched in Astrocytes (PED/PEA-15) after TRAIL treatment were compared in the clones. Basal levels of PED/PEA-15 expression were similar among sensitive, partial resistant and resistant clones. TRAIL did not change the PED/PEA-15 level in the clones. In addition, transduction and expression of the dominant negative form of the I-kBalpha gene did not change TRAIL-sensitivities. Our results showed that the expression levels of DR5, the activation levels of caspase-8, -3 and RIP were critical factors in determining TRAIL-sensitivities in Jurkat cells. The results of our study also suggest that cells with different TRAIL-sensitivities arise through multiple mechanisms even within a single cell line.

Mouse monoclonal autoantibodies penetrate mouse macrophage cells and stimulate NF-κB activation and TNF-α release

Autoantibodies against double-stranded DNA (dsDNA) are found in the serum of systemic lupus erythematosus (SLE) patients. However, the mechanism by which anti-dsDNA antibodies (Abs) contribute to the pathogenesis of SLE is not yet fully understood. In this study, we investigated four anti-dsDNA mouse monoclonal autoantibodies that share positively charged amino acids (including arginines) in their complementarity determining regions for their ability to penetrate RAW264.7 macrophage cells, activate NF-kappaB and stimulate TNF-alpha production. All four antibodies penetrated into macrophage cells and increased the level of extracellular TNF-alpha; two also activated NF-kappaB. The fact that two of four cell-penetrating anti-dsDNA mAbs induced both NF-kappaB activation and TNF-alpha production in macrophages suggests that at least some autoantibodies against dsDNA may play a role in the pathogenesis of SLE by penetrating into macrophage cells and nuclei, and subsequently inducing the pro-inflammatory cytokine, TNF-alpha, by binding to the NF-kappaB gene and stimulating its transcriptional activity.

Aspirin enhances TRAIL-induced apoptosis via regulation of ERK1/2 activation in human cervical cancer cells

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) triggers tumor-specific apoptosis. However, some tumors and cancer cell lines are resistant to TRAIL. Here, the effect of the non-steroidal anti-inflammatory drug aspirin on sensitization of human cervical cancer cells to TRAIL and the underlying mechanism(s) of the effect were explored. Combination treatment with aspirin and TRAIL markedly enhanced apoptotic cell death, as assessed by lactate dehydrogenase (LDH) assay and analysis of cell cycle sub-G1 phase. The two agents together activated the several caspases and mitochondrial signaling pathway. Whereas Mcl-1 protein level was increased and extracellular signal-related kinase (ERK)1/2 was activated in cells treated with TRAIL alone, combination treatment dramatically inhibited ERK1/2 activation and down-regulated Mcl-1 protein level. An inhibitor of ERK1/2 activation, PD98059, also augmented TRAIL-induced apoptosis. Combination treatment with PD98059 and TRAIL showed the activation of caspases and mitochondrial pathway, and the down-regulation of Mcl-1 level. These results suggest that cancer cells can be sensitized to TRAIL-induced apoptosis by pre-treatment with aspirin via suppression of ERK1/2 activation. These findings provide a basis for further exploring the potential applications of this combination approach for the treatment of cancer, including cervical cancer.

Specific modulation of the anti-DNA autoantibody-nucleic acids interaction by the high affinity RNA aptamer.

Anti-DNA autoantibodies are one of the frequently found autoantibodies in systemic lupus erythematosus patient sera. RNA aptamers for the monoclonal G6-9 anti-DNA autoantibody were selected from a random pool of RNA library. Binding affinity of the best aptamer is around 2nM, which is at least 100-fold higher than that of cognate DNA antigen to the autoantibody. Aptamer binds specifically to the G6-9 autoantibody but not to other similar autoantibodies. Minimal binding motif of the aptamer was mapped, providing a hint for a natural epitope of the autoantibody. DNA binding to the G6-9 autoantibody is shown to be efficiently inhibited by the aptamer. Such binding property of the RNA aptamer could be used not only as a modulator for the pathogenic anti-DNA autoantibody, but also as a useful biochemical reagent for elucidating a fine specificity of the autoantibody-nucleic acid interaction.

Murine Model of Buckwheat Allergy by Intragastric Sensitization with Fresh Buckwheat Flour Extract

Food allergies affect about 4% of the Korean population, and buckwheat allergy is one of the most severe food allergies in Korea. The purpose of the present study was to develop a murine model of IgE-mediated buckwheat hypersensitivity induced by intragastric sensitization. Young female C3H/HeJ mice were sensitized and challenged intragastricly with fresh buckwheat flour (1, 5, 25 mg/dose of proteins) mixed in cholera toxin, followed by intragastric challenge. Anaphylactic reactions, antigen-specific antibodies, splenocytes proliferation assays and cytokine productions were evaluated. Oral buckwheat challenges of sensitized mice provoked anaphylactic reactions such as severe scratch, perioral/periorbital swellings, or decreased activity. Reactions were associated with elevated levels of buckwheat-specific IgE antibodies. Splenocytes from buckwheat allergic mice exhibited significantly greater proliferative responses to buckwheat than non-allergic mice. Buckwheat-stimulated IL-4, IL-5, and INF-γ productions were associated with elevated levels of buckwheat-specific IgE in sensitized mice. In this model, 1 mg and 5 mg dose of sensitization produced almost the same degree of Th2-directed immune response, however, a 25 mg dose showed blunted antibody responses. In conclusion, we developed IgE-mediated buckwheat allergy by intragastric sensitization and challenge, and this model could provide a good tool for future studies.

Production and characterization of an anti-idiotypic single chain Fv that recognizes an anti-DNA antibody

A well-characterized recombinant anti-idiotype to an anti-DNA antibody can be useful for studies of the regulation of anti-DNA-producing B cells. Using a hybridoma technique, a monoclonal anti-idiotypic antibody, designated O2F3, was obtained, and its scFv gene was constructed. O2F3 single chain Fv (scFv) was produced against an idiotope of a monoclonal anti-DNA antibody, 3D8, that was obtained from an autoimmune-prone mouse, MRL-lpr/lpr. Here we describe the production and in vitro characterization of the O2F3 scFv, and compare it with its parent monoclonal antibody, O2F3 IgM. To characterize O2F3 scFv and O2F3 IgM, we generated recombinant 3D8 fragments, including 3D8 scFv, 3D8 VH, and 3D8 VL, that were used as antigens in several assays. ELISA and Western blot analysis showed that both O2F3 scFv and O2F3 IgM recognized a conformational determinant formed by the association of the variable region heavy and light chains of the 3D8 antibody, suggesting that O2F3 scFv retained a similar binding pattern to its parent O2F3 antibody. The idiotope recognized by O2F3 was shown by competitive ELISA to be outside of the DNA binding site of the 3D8 antibody. This characterized O2F3 scFv could be applied for the regulation of anti-DNA antibody production and the manipulation of recombinant antibody-based proteins to which toxins, enzymes, and chemical agents can be connected.

Cell- and nuclear-penetrating anti-dsDNA autoantibodies have multiple arginines in CDR3 of VH and increase cellular level of pERK and Bcl-2 in mesangial cells

Investigation of characteristics of cell- and nuclear-penetrating anti-double stranded (ds)DNA autoantibodies (autoAbs) is important to understand pathogenesis of lupus nephritis, but has not been clearly explored. The present study reports that three anti-dsDNA monoclonal autoAbs, which contain more than two arginine residues in their CDR3s of variable heavy domain (VH), penetrated into living murine mesangial cells and the cell nuclei. However, an anti-dsDNA monoclonal Ab (mAb) having only one arginine in the CDR3-VH did not penetrate cells. To assess the contribution of antigen-binding sites, especially the VH, in cell- and nuclear-penetration, we evaluated the characteristics of recombinant single chain Fv(scFv), VH, and variable light domain (VL) of a penetrating mAb. The scFv and VH domain, containing arginine in CDR3-VH maintained the ability to penetrate cells and the cell nuclei, whereas the VL domain, having no arginine in CDR3, did not penetrate cells. The penetratingm Abs, scFv, and VH activated ERK and increased cellular protein levels of Bcl-2, whereas the non-penetrating Ab and VL did not. The cell survival was decreased by the penetrating mAbs, scFv and VH, not by the non-penetrating mAb and VL. The data indicate that an antigen-binding site is required for cell-penetration and that positively-charged arginine residues in CDR3-VH contribute to the cell- and nuclear-penetrating ability of a subset of anti-dsDNA autoAbs. Furthermore,the nuclear-penetrating anti-dsDNA autoAbs could possibly function as a pathogenic factor for lupus nephritis by up-regulating ERK activation and Bcl-2 production in mesangial cells. The cell- and nuclear-penetrating VH domain may be exploited as a vehicle for the intra cellular delivery of various useful molecules.

Modulation of fine specificity of anti-DNA antibody by CDR3L residues.

The light chain of 2C10, an anti-double stranded DNA (dsDNA) autoantibody, is not favorable for DNA binding and it was suggested that the light chain might modulate the specificity of the antibody in DNA binding. We studied several mutant scFvs expressing mutated VL and normal VH of 2C10 to explore the role of the light chain in determining the fine specificity of the antibody, which we define as the preferential binding to a specific sequence of bases or a helical conformation compared to dsDNA from calf thymus. The wild-type Fab and scFv of 2C10 bind to poly(dA-dC).(dG-dT) better than to dsDNA. However, in the absence of the light chain domain, the VH domain bound dsDNA better than poly(dA-dC).(dG-dT), indicating the possible involvement of the light chain in determining the fine specificity in DNA binding. The mutations we studied were located in either CDR1L or CDR3L of the antibody. The CDR1 mutants, D28A, D30A, D31A, and D32A have been previously shown to cause an increase in the affinity of 2C10 scFv to DNA. The fine specificity of 2C10 was not affected by the CDR1 mutants which bound to poly(dA-dC).(dG-dT) better than dsDNA. However, CDR3L mutants, D92A and N93A, which had been shown to be involved in direct interaction with DNA, preferred dsDNA to poly(dA-dC).(dG-dT) in their binding. Our results indicate that the fine specificity of 2C10 in DNA binding is modulated primarily by Asp at 92 and Asn at 93 in CDR3L. The effects of CDR1L mutations indicate that this region affects only the affinity but not the fine specificity of 2C10.