Triglycerides in the human kidney cortex: relationship with body size

Abstract

Obesity is associated with increased risk for kidney disease and uric acid nephrolithiasis, but the pathophysiological mechanisms underpinning these associations are incompletely understood. Animal experiments have suggested that renal lipid accumulation and lipotoxicity may play a role, but whether lipid accumulation occurs in humans with increasing body mass index (BMI) is unknown. The association between obesity and abnormal triglyceride accumulation in non-adipose tissues (steatosis) has been described in the liver, heart, skeletal muscle and pancreas, but not in the human kidney. We used a quantitative biochemical assay to quantify triglyceride in normal kidney cortex samples from 54 patients undergoing nephrectomy for localized renal cell carcinoma. In subsets of the study population we evaluated the localization of lipid droplets by Oil Red O staining and measured 16 common ceramide species by mass spectrometry. There was a positive correlation between kidney cortex trigyceride content and BMI (Spearman R = 0.27, P = 0.04). Lipid droplets detectable by optical microscopy had a sporadic distribution but were generally more prevalent in individuals with higher BMI, with predominant localization in proximal tubule cells and to a lesser extent in glomeruli. Total ceramide content was inversely correlated with triglycerides. We postulate that obesity is associated with abnormal triglyceride accumulation (steatosis) in the human kidney. In turn, steatosis and lipotoxicity may contribute to the pathogenesis of obesity-associated kidney disease and nephrolithiasis.

Figure 1. Kidney cortex triglyceride content and relationship with body mass index (BMI).

Nephrectomy cortex samples were obtained from 54 patients with serum creatinine <1.5 mg/dL and no proteinuria. All surgical specimens were dissected by experienced clinical pathologists and confirmed to be healthy, away from the tumor. R_S, Spearman rank-order correlation coefficient.

Figure 2. Lipid staining.

Kidney cortex sections were stained with Oil Red O and standard hematoxylin to visualize lipid localization within renal structures. After image acquisition, a computer-based color deconvolution and thresholding algorithm was used to separately visualize and quantify Oil Red O staining, with the original image serving as reference for the manual assignment of lipid staining to discrete structures (i.e. tubule cells, glomeruli, interstitium, other). A. Oil Red O staining was noted predominantly in proximal tubule cells, and generally increased with increasing body mass index (BMI). B. Examples of the localization of lipids within tubule cells, glomeruli and interstitium. C. Relative distribution of lipids within kidney structures based on Oil Red O staining quantified by color deconvolution in 29 study participants with available fixed tissue. Whiskers represent 95% confidence intervals.

Figure 3. Kidney cortex ceramide content and relationship with triglycerides.

Sixteen common ceramide species (Cer, ceramide; GlucCer, glucosylceramide; LacCer, lactosylceramide; DHCer, dihydroceramide) were measured by high performance liquid chromatography-electrospray ionization-tandem mass spectrometry (HPLC-ESI-MS/MS) in kidney cortex samples previously frozen at −80°C. A. The abundance of each ceramide species in human kidney cortex samples with increasing triglyceride content. B. Relationship between total ceramide content (molar sum of all measured ceramides) and triglyceride content in human kidney cortex samples. R_S, Spearman rank-order correlation coefficient. C. Total ceramide content in kidney cortex samples from Zucker diabetic fatty (ZDF) and lean control rats.