Diabetic cardiomyopathy (DCM) is a common diabetic complication characterized by diastolic relaxation abnormalities, myocardial fibrosis and chronic heart failure. Although TGF-β/Smad3 signalling has been shown to play a critical role in chronic heart disease, the role and mechanisms of Smad3 in DCM remain unclear. We reported here the potential role of Smad3 in the development of DCM by genetically deleting the Smad3 gene from db/db mice. At the age of 32 weeks, Smad3WT-db/db mice developed moderate to severe DCM as demonstrated by a marked increase in the left ventricular (LV) mass, a significant fall in the LV ejection fraction (EF) and LV fractional shortening (FS), and progressive myocardial fibrosis and inflammation. In contrast, db/db mice lacking Smad3 (Smad3KO-db/db) were protected against the development of DCM with normal cardiac function and undetectable myocardial inflammation and fibrosis. Interestingly, db/db mice with deleting one copy of Smad3 (Smad3 ± db/db) did not show any cardioprotective effects. Mechanistically, we found that deletion of Smad3 from db/db mice largely protected cardiac Smad7 from Smurf2-mediated ubiquitin proteasome degradation, thereby inducing IBα to suppress NF-kB-driven cardiac inflammation. In addition, deletion of Smad3 also altered Smad3-dependent miRNAs by up-regulating cardiac miR-29b while suppressing miR-21 to exhibit the cardioprotective effect on Smad3KO-db/db mice. In conclusion, results from this study reveal that Smad3 is a key mediator in the pathogenesis of DCM. Targeting Smad3 may be a novel therapy for DCM. © 2021 The Authors. Journal of Cellular and Molecular Medicine published by Foundation for Cellular and Molecular Medicine and John Wiley & Sons Ltd.