Background/Purpose: Citrullination is a post-translational protein modification, where arginine (Arg) is converted to citrulline (Cit) by peptidyl arginine deiminases (PADI). The presence of anti-citrullinated protein antibodies (ACPA) in serum is highly specific for rheumatoid arthritis (RA). Some of these ACPA may recognize citrullinated cartilage components such as type II collagen (CII) or the proteoglycan (PG) aggrecan in the RA joints afflicted with inflammation. The major goals of this study were (1) to determine whether a citrullinated form of cartilage PG is present in normal or diseased human cartilage, and (2) to identify citrullinated PG (CitPG)-specific serum antibodies (Abs) in ACPA⁺ RA patients.

Methods: PG from cartilage of normal human joints was extracted with 4M GuHCl and purified by CsCl gradient ultracentrifugation. Crude extracts were also prepared from osteoarthritic (OA) and RA cartilage specimens obtained from joint replacement surgery. Serum ACPA concentrations in RA patients were measured using anti-MCV (Orgentec) and anti-CCP3 (Inova) kits. RA serum samples with high ACPA titers were reacted with both purified PG and crude cartilage extracts in dot blots and Western blots. The PG specificity of reactions was verified using monoclonal Ab G18 to the G1 domain of human PG. Recombinant G1 domain of PG and purified human CII, both citrullinated in vitro with PADI4, were used for the measurement of CitPG- and CitCII-specific serum Abs, respectively, by ELISA. CitPG was visualized in frozen sections of human cartilage specimens by immunohistochemistry.

Results: While there was a strong correlation between anti-MCV and anti-CCP3 titers, the correlation was weaker between anti-CitPG and anti-CCP3 Ab levels in the 68 ACPA⁺ RA serum samples tested. ELISA assays, using CitPG and CitCII as a capture antigens, showed that 66% of the ACPA⁺ sera reacted with CitPG, 46% with CitCII, and 41% with both CitPG and CitCII. When we compared sera from healthy subjects and OA patients with the ‘calibrator’ serum standards supplied with the anti-CCP3 kits, only the anti-CCP3⁺ calibrator sera gave a positive reaction with CitPG. Comparison of matched cartilage and serum samples from OA patients demonstrated the presence of CitPG in ~60% of OA cartilage extracts, while CCP3- or CitPG-specific Abs could not be detected in any of their serum samples. On the other hand, all cartilage specimens from RA patients contained CitPG, and 92% of the serum samples from the same RA donors were positive for both anti-CCP3 and anti-CitPG Abs. Western blot of fragmented cartilage PG showed that the G1 domain contained an abundance of citrullinated epitopes. In cartilage tissue sections, CitPG was predominantly localized intracellularly and in the pericellular matrix of chondrocytes.

Conclusion: PG (especially the Arg-rich G1 domain) in both normal and diseased cartilage is subjected to citrullination in vivo, likely by PADI in chondrocytes. While a significant proportion of normal, OA, and RA cartilage specimens contains CitPG, only sera from RA patients seem to recognize it. There is a significant overlap between anti-CitPG Abs and other ACPA, including anti-CCP3, anti-MCV, and anti-CitCII Abs in RA.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/citrullinated-proteoglycan-aggrecan-present-in-human-cartilage-is-recognized-by-serum-antibodies-from-rheumatoid-arthritis-patients/