Clinical significance of a subset of anti-CENP-A antibodies cross-reacting with FOXE3 in systemic sclerosis

BACKGROUND. Auto-Abs to the histone H3-like centromeric protein-A (CENP-A) are frequently detected in systemic sclerosis (SSc) with limited cutaneous involvement. The long-range goal of this study is to define the origin of CENP-A Abs through the identification of proteins other than CENP-A, which can be their target or induce them. We found that Abs to an immunodominant region of the CENP-A (spanning amino acids 17 to 30; Ap17-30) recognized two distinct, though overlapping, epitopes. One of these is defined by the motif PTPxxGPXXR.OBJECTIVE: 1) to identify proteins expressing the anti-CENP-A-specific motif, 2) to analyze their homology to CENP-A; 3) to screen anti-CENP-A Ab positive sera against peptide-bearing the motif.METHODS: A Swiss-Prot database search was performed to identify proteins expressing anti-CENP-A-specific motifs. Anti-Ap17-30 IgGs were purified from sera by affinity chromatography on an insolubilized peptide column. ELISA was used to test the specificity of serum or Abs against peptides in binding and inhibition assays. Microscopy immunofluorescence assay was performed on fixed-permeabilized HeLa cells, using purified IgG anti-Ap17-30 cells and FITC-conjugated anti-human IgG as detecting reagents. Immunization of BALB/c mice with peptides was performed by intra-peritoneal injections of 10 µg-peptide on day 0, 7, 14 and 70. Sera were drawn weekly until the 13th week and stored at -80°C until used.RESULTS: A database search identified 4 proteins expressing the motif PTPxxGPxxR. Of these, the forkhead box E3 (FOXE3)-derived peptide FOXE3p displayed the highest homology with CENP-A Ap17-30 (identity: 80%), while condensing-2 complex subunit D3-derived peptide (CD3p), protein Jumonji-derived peptide (JMJp) and microtubule-associated protein 1A-derived peptide (MAP1p) displayed 60% homology to Ap17-30. These peptides' antigenic homology with Ap17-30 was evaluated, using IgG anti-Ap17-30 affinity-purified from the sera of 8 anti-CENP-A+ SSc patients (IgG pt1 to pt8). As compared to Ap17-30, Pt1 to pt8 IgG displayed the highest reactivity with FOXE3p, while a lower or no reactivity was observed with the remaining peptides. Furthermore, mouse anti-Ap17-30 and anti-FOXE3 Abs cross-reacted to each other, while Abs elicited with MAP1A partially reacted with CD3. Neither CD3 nor JMJ were found to be immunogenic. Kinetic studies of the reactivity of anti-FOXE3 Abs with Ap17-30 indicated that FOXE3 booster injection on the day 70 markedly increased their relative avidity to Ap17-30. Furthermore mouse anti-FOXE3 sera recognize CENP-A and FOXE3 in western blot, while mouse anti-Ap17-30 recognize CENP-A only. The testing of 69 human anti-CENP-A/Ap17-30+ sera with FOXE3p identified its reactivity with 50.7 % of them.CONCLUSION: 1) FOXE3 displays an antigenic and immunogenic homology with CENP-A-derived Ap17-30; 2) FOXE3p can identify a subset of anti-CENP-A Ab-positive patients. Ongoing studies are expected to establish whether any correlation between FOXE3 reactive Ab-patients and a clinical profile exists.