More severe OA pathology in a humanized mouse model for APOE-ε4 as compared to APOE-ε3: APOE-isoforms as possible risk factor for inflammatory osteoarthritis development?

Abstract

Purpose: Apolipoprotein E (APOE) is a protein that is closely involved in lipid metabolism. Absence of APOE results in increased serum levels of low-density lipoprotein (LDL) cholesterol, which we previously described to be associated with the development of osteoarthritis (OA) pathology. However, next to its function in lipid metabolism, APOE has been shown to dampen the innate immune system that is strongly involved in the development of OA pathology. In humans, three isoforms of the APOE gene exist (APOE-ε2, APOE-ε3 and APOE-ε4), with different efficacies in suppressing innate immunity (ε2>ε3>ε4). Therefore not surprisingly, the APOE-ε4 isotype has been associated with various inflammation-mediated diseases, including atherosclerosis, Alzheimer’s disease, and multiple sclerosis. However, whether the various APOE isoforms also affect the development of OA remains unknown. Therefore, in the present study we investigated the development of joint pathology after induction of collagenase-induced OA (CIOA) in APOE-ε3 and APOE-ε4 targeted replacement mice. Methods: Wild type, Apoe-/-, and humanized APOE-ε3 and APOE-ε4 targeted replacement mice in which the murine Apoe gene was replaced by the human APOE-ε3 and APOE-ε4 gene, all received a normal diet. CIOA was induced by intra-articular injection of collagenase and mice were sacrificed at day 42 after induction of the model. Serum cholesterol levels were determined using a colorimetric assay. Cartilage degeneration and osteophyte/ectopic bone formation were scored using Safranin O & Fast Green stained sections, representing the entire depth of the joint. Hematoxylin & eosin-stained sections were used to quantify synovial inflammation and chondrocyte death as the percentage of cartilage containing empty lacunae compared to the total cartilage area. Results: Apoe-/-mice showed markedly increased serum LDL levels as compared to their wild type controls (8.90 mmol/L and 0.40 mmol/L, respectively; p<0.001). Moreover, Apoe-/- mice showed significantly increased thickening of the synovium at end-stage OA. Whereas we did not observe differences in cartilage degeneration, we found a strongly increased ectopic bone formation, mainly in the medial collateral ligament. Next, we determined the effects of various human APOE isotypes on the development of joint pathology after induction of CIOA using humanized APOE-ε3 and APOE-ε4 targeted replacement mice. Both mouse strains showed comparable serum LDL levels, which were not raised as compared to wild type mice. Contralateral control joints showed comparable but low synovial inflammation, cartilage degeneration and osteophyte/ectopic bone formation. Interestingly, after induction of CIOA, APOE-ε4 showed increased synovial inflammation as compared to APOE-ε3 mice. In agreement with this finding, we observed significantly increased degeneration of the articular cartilage, which was associated with an increased chondrocyte death. Finally, we found a significantly increased surface area of ectopic bone and osteophyte formation. Conclusions: High LDL cholesterol levels by Apoe deficiency in mice results in increased synovial thickening and ectopic bone formation in end-stage OA. Mice with the human APOE-ε4 gene develop more severe OA pathology as compared to the APOE-ε3 isotype, which is in line with the decreased potency of APOE-ε4 to dampen innate immune responses. This indicates a possible role for the human APOE isotypes as risk factor for the development of OA that needs further investigation.