PURIFICATION OF FILAGGRIN FROM HUMAN EPIDERMIS AND MEASUREMENT OF ANTIFILAGGRIN AUTOANTIBODIES IN SERA FROM PATIENTS WITH RHEUMATOID-ARTHRITIS BY AN ENZYME-LINKED-IMMUNOSORBENT-ASSAY

Background: The so-called antikeratin antibody (AKA) and the antiperin uclear factor (APF) that recognize proteins related to human epidermal filaggrin belong to the most specific serological markers of rheumato id arthritis (RA). However, assays for the detection of AKA and APF ar e currently based on immunofluorescence, a method that is subject to a rbitrary interpretation and inadequate standardization of the substrat es. Methods: Proteins extracted from human epidermis were separated by reversed-phase high-performance liquid chromatography (HPLC), Filaggr in-containing fractions, identified in immunoblotting by monoclonal an tifilaggrin antibodies, were then subjected to gel filtration HPLC and , finally, to a second reversed-phase HPLC step. Tryptic digestion, am ino acid sequencing and mass spectrometry were used to cornfirm the id entity of the purified protein. Filaggrin was used as antigen in enzym e-linked immunosorbent assay (ELISA) to measure IgG class antifilaggri n antibodies. Results: The filaggrin preparation obtained gave a singl e band in sodium dodecyl sulfate-polyacrylamide gel electrophoresis, b inding monoclonal antifilaggrin in immunoblotting. Amino acid sequence s of all 10 tryptic peptides analyzed were shown to originate from hum an filaggrin. Antifilaggrin antibody levels exceeded the 99th percenti le level of 100 middle-aged blood donors in 26/55 (47%) RA sera. At a similar cutoff level 28/55 (51%) of the RA sera were positive in the A KA test. Of the 26 antifilaggrin-positive sera, 21 were also AKA-posit ive. Conclusion: Human filaggrin can be purified by standard biochemic al techniques, despite the heterogeneity of the protein, and used in E LISA for testing autoantibodies to filaggrin. The sensitivity of the a ssay equals that of the AKA test.