BackgroundGenetic studies have pointed out that CD226 variants, encoding DNAM-1, could be associated with susceptibility to rheumatoid arthritis. Therefore, we aimed to determine the influence of DNAM-1 on the development of arthritis using the collagen-induced arthritis (CIA) mouse model.MethodsCIA was induced in mice on a DBA/1 background, treated in parallel with a DNAM-1 neutralizing monoclonal antibody, a control IgG and PBS, respectively. CIA was also induced in mice deficient for DNAM-1(dnam1−/−) and control dnam-1+/+ mice on a C57/BL6 background. Mice were monitored for clinical and ultrasound signs of arthritis. Histological analysis was performed to search for inflammatory infiltrates and erosions. The Mann–Whitney U test for non-related samples was used for statistical analysis.ResultsThere was a non-significant trend for a less arthritic phenotype in mice receiving anti-DNAM-1 mAb at both clinical, ultrasound and histological assessments. But, we did not observe any difference between dnam1+/+ and dnam1−/− mice for incidence nor severity of clinical arthritis. Histological analysis revealed inflammatory scores similar in both groups, without evidence of erosion. Collagen antibodies levels were similar in all mice, confirming immunization with collagen.ConclusionDespite some clues suggesting a role of DNAM-1 in arthritis, these complementary approaches demonstrate no contribution of CD226/DNAM-1 in the arthritic phenotype. These results contrast with previous studies showing a role in vivo of DNAM-1 in some autoimmune disorders.Electronic supplementary materialThe online version of this article (doi:10.1186/s12950-015-0056-5) contains supplementary material, which is available to authorized users.