KIM-1, a transmembrane tubular protein with unknown function, is undetectable in normal kidneys, but is markedly induced in experimental renal injury. The KIM-1 ectodomain is cleaved, detectable in urine, and reflects renal damage. KIM-1 expression in human renal biopsies and its correlation with urinary KIM-1 (uKIM-1) is unknown. In biopsies from various renal diseases (n = 102) and controls (n = 7), the fraction of KIM-1 positive tubules and different renal damage parameters were scored. Double labelling was performed for KIM-1 with macrophages (MØ), alpha-smooth muscle actin (alpha-SMA), proximal (aquaporin-1) and distal (E-cadherin) tubular markers and a dedifferentiation marker (vimentin). uKIM-1 at the time of biopsy (n = 53) was measured by ELISA. Renal KIM-1 was significantly increased in all diseases versus controls (p < 0.05), except minimal change. KIM-1 was primarily expressed at the luminal side of dedifferentiated proximal tubules, in areas with fibrosis (alpha-SMA) and inflammation (MØ). Independent of the disease, renal KIM-1 correlated positively with renal damage, negatively with renal function, but not with proteinuria. uKIM-1 was increased in renal patients versus controls (p < 0.001), including minimal change, and correlated positively with tissue KIM-1 and MØ, negatively with renal function, but not with proteinuria. In conclusion, KIM-1 is upregulated in renal disease and is associated with renal fibrosis and inflammation. uKIM-1 is also associated with inflammation and renal function, and reflects tissue KIM-1, indicating that it can be used as a non-invasive biomarker in renal disease.

Albumin induces oxidative stress and cytokine production in proximal tubular cells (PTECs). Albumin-bound fatty acids (FAs) enhance tubulopathic effects of albumin in vivo. We proposed that FA aggravation of albumin-induced oxidative stress in PTECs might be involved. We hypothesized that mitochondria could be a source of such stress. Using a fluorescent probe, we compared reactive oxygen species (ROS) production after exposure of PTECs to bovine serum albumin (BSA) alone or loaded with oleic acid (OA-BSA) (3-30 g/l for 2 h). There was no difference in cellular albumin uptake, but OA-BSA dose-dependently induced more ROS than BSA alone (P<0.001). OA-BSA-induced ROS was significantly alleviated by mitochondrial inhibition, but not by inhibitors of nicotinamide adenine dinucleotide phosphate hydrogenase (NADPH) oxidase, xanthine oxidase, or nitric oxide synthase. Gene expression analysis showed that neither the NADPH oxidase component p22phox nor xanthine oxidase was induced by BSA or OA-BSA. OA-BSA, in contrast to BSA, failed to induce mitochondrial manganese superoxide dismutase 2 (SOD2) expression. OA-BSA showed a greater capacity than BSA to downregulate heme oxygenase-1 mRNA expression and accentuate inflammatory cytokine mRNA and protein. Supplementation of SOD activity with EUK-8 reduced ROS, and interleukin-6 protein expression was suppressed by both mitochondrial inhibition and SOD augmentation. Thus, in PTECs, FAs accentuate albumin-induced oxidative stress and inflammatory cytokine expression via increased mitochondrial ROS, while frustrating protective antioxidant responses.