High mobility group box 1 (HMGB1), an important inflammatory mediator, is actively secreted by immune cells and some non-immune cells or passively released by necrotic cells. HMGB1 has been implicated in many inflammatory diseases. Our previous published data demonstrated that HMGB1 was up-regulated in heart tissue or serum in experimental autoimmune myocarditis (EAM); HMGB1 blockade could ameliorate cardiac fibrosis at the last stage of EAM. And yet, until now, no data directly showed that HMGB1 was associated with cardiac fibrosis. Therefore, the aims of the present work were to assess whether (1) up-regulated HMGB1 could directly lead to cardiac fibrosis in EAM; (2) cardiac fibroblast/myofibroblasts could secrete HMGB1 as another source of high-level HMGB1 in EAM; and (3) HMGB1 blockade could effectively prevent cardiac fibrosis at the last stage of EAM. Our results clearly demonstrated that HMGB1 could directly lead to cardiac collagen deposition, which was associated with PKCβ/Erk1/2 signalling pathway; furthermore, cardiac fibroblast/myofibroblasts could actively secrete HMGB1 under external stress; and HMGB1 secreted by cardiac fibroblasts/myofibroblasts led to cardiac fibrosis via PKCβ activation by autocrine means; HMGB1 blockade could efficiently ameliorate cardiac fibrosis in EAM mice.