Dietary trans-10,cis-12 (t10c12) conjugated linoleic acid (CLA) has been shown to reduce inflammation in a murine collagen-induced arthritis (CA) model. To understand the anti-inflammatory potential of t10c12-CLA in the diet, the minimum dose of pure dietary t10c12-CLA capable of reducing CA was investigated. Because plasma inflammatory cytokines often do not reflect the progression of late-stage arthritis, inflamed tissue cytokine concentrations were also investigated in relation to increasing dietary t10c12-CLA amounts. Mice were randomly assigned to the following dietary treatments upon the establishment of arthritis: corn oil (CO) or 0.125%, 0.25%, 0.375%, or 0.5% t10c12-CLA (wt:wt) for 84 d. Sham mice (no arthritis) were fed CO and served as controls. Arthritic paw score, based on subjective assessment of arthritic severity, and paw thickness decreased linearly overall [16-65% (P < 0.001) and 0.5-12% (P < 0.001), respectively] as dietary t10c12-CLA increased (P < 0.001, R(2) < 0.81). Increasing dietary t10c12-CLA was associated with a decrease in plasma interleukin (IL)-1β at days 21 and 42 compared with CO-fed arthritic mice, such that mice fed ≥0.25% t10c12-CLA had IL-1β concentrations that were similar to sham mice. Plasma cytokines returned to sham mice concentrations by day 63 regardless of treatment; however, an arthritis-induced elevation in paw IL-1β decreased linearly as dietary t10c12-CLA concentrations increased at day 84 (P = 0.007, R(2) = 0.92). Similarly, increasing dietary t10c12-CLA linearly decreased paw tumor necrosis factor (TNF)-α (P = 0.05, R(2) = 0.70). In conclusion, ≥0.125% t10c12-CLA dose-dependently reduced inflammation in a murine CA model.