AIMS/HYPOTHESIS: The AGEs and the receptor for AGEs (RAGE) are known contributors to diabetic complications. RAGE also has a physiological role in innate and adaptive immunity and is expressed on immune cells. The aim of this study was to determine whether deletion of RAGE from bone-marrow-derived cells influences the pathogenesis of experimental diabetic nephropathy. METHODS: Groups (n = 8/group) of lethally irradiated 8 week old wild-type (WT) mice were reconstituted with bone marrow from WT (WT --> WT) or RAGE-deficient (RG) mice (RG --> WT). Diabetes was induced using multiple low doses of streptozotocin after 8 weeks of bone marrow reconstitution and mice were followed for a further 24 weeks. RESULTS: Compared with diabetic WT mice reconstituted with WT bone marrow, diabetic WT mice reconstituted with RG bone marrow had lower urinary albumin excretion and podocyte loss, more normal creatinine clearance and less tubulo-interstitial injury and fibrosis. However, glomerular collagen IV deposition, glomerulosclerosis and cortical levels of TGF-beta were not different among diabetic mouse groups. The renal tubulo-interstitium of diabetic RG --> WT mice also contained fewer infiltrating CD68(+) macrophages that were activated. Diabetic RG --> WT mice had lower renal cortical concentrations of CC chemokine ligand 2 (CCL2), macrophage inhibitory factor (MIF) and IL-6 than diabetic WT --> WT mice. Renal cortical RAGE ligands S100 calgranulin (S100A)8/9 and AGEs, but not high mobility box protein B-1 (HMGB-1) were also decreased in diabetic RG --> WT compared with diabetic WT --> WT mice. In vitro, bone-marrow-derived macrophages from WT but not RG mice stimulated collagen IV production in cultured proximal tubule cells. CONCLUSIONS/INTERPRETATION: These studies suggest that RAGE expression on haemopoietically derived immune cells contributes to the functional changes seen in diabetic nephropathy by promoting macrophage infiltration and renal tubulo-interstitial damage.