Diabetic kidney disease (DKD) is the leading cause of end-stage kidney disease, and the current pharmacological treatment for DKD is limited to renin-angiotensin system (RAS) inhibitors. Adenosine is detectable in the kidney and is significantly elevated in response to cellular damage. While all 4 known subtypes of adenosine receptors, namely, A1AR, A2aAR, A2bAR, and A3AR, are expressed in the kidney, our previous study has demonstrated that a novel, orally active, species-independent, and selective A3AR antagonist, LJ-1888, ameliorates unilateral ureteral obstruction-induced tubulointerstitial fibrosis. The present study examined the protective effects of LJ-2698, which has higher affinity and selectivity for A3AR than LJ-1888, on DKD. In experiment I, dose-dependent effects of LJ-2698 were examined by orally administering 1.5, 5, or 10 mg/kg for 12 weeks to 8-week-old db/db mice. In experiment II, the effects of LJ-2698 (10 mg/kg) were compared to those of losartan (1.5 mg/kg), which is a standard treatment for patients with DKD. LJ-2698 effectively prevented kidney injuries such as albuminuria, glomerular hypertrophy, tubular injury, podocyte injury, fibrosis, inflammation, and oxidative stress in diabetic mice as much as losartan. In addition, inhibition of lipid accumulation along with increases in PGC1α, a master regulator of mitochondrial biogenesis, were demonstrated in diabetic mice treated with either LJ-2698 or losartan. These results suggest that LJ-2698, a selective A3AR antagonist, may become a novel therapeutic agent against DKD.