Albuminuria is an established marker for the development of diabetic nephropathy. Normal renal handling of albumin involves endocytosis by proximal tubule cells through a lysosomal degradation pathway, which returns small (<15 kDa) albumin fragments to the tubular lumen. These fragments are undetectable by conventional clinical laboratory assays. Albuminuria has been shown in patients with type 1 diabetes to be associated with an impaired degradation pathway¹, evidenced by a shift in the ratio of intact to fragmented urinary albumin. A cross-sectional study demonstrated a decrease in urinary peptides in patients with macroproteinuria compared to normoproteinuric controls². However, urinary peptide excretion studies have not controlled for changes in estimated glomerular filtration rate (eGFR) in patients with diabetes.

Patients with diabetes and eGFR <60ml/min were stratified into normo- (<20mcg/min, n=9), micro- (20-200mcg/min, n=12) or macroalbuminuric (>200mcg/min, n=9) groups. 24 hr urine samples were passed through a 10kDa cut-off protein filter, and the <10kDa (peptide fragments) and >10kDa (intact protein) fractions were assayed separately using the Pierce BCA protein assay to detect peptide bonds based on the Lowry method.

All patients excreted most protein in a fragmented form, with >10 kDa protein contributing from 4.9% to 26.4% of total protein. Macroalbuminuric patients had a reduced proportion of peptide excretion in the form of fragments (<10kDa) compared to micro- and normoalbuminuric patients (81.1% vs 89.0%, p<0.05 and 96.8%, p<0.001 respectively); however there was no difference observed between micro- and normoalbuminuric groups (Figure 1). Absolute urinary fragment concentrations were lower in micro- and macro- groups compared to normoalbuminuric patients, (3389mcg/ml and 3480mcg/ml vs 4396mcg/ml) but these did not reach statistical significance.

In patients with diabetes and reduced renal function, we have demonstrated a reduction in peptide fragment proportional excretion in macroalbuminuric patients after controlling for eGFR. These findings are consistent with the hypothesis that albuminuria is linked to defects in the renal tubule resorption and fragmentation pathways.

Figure 1 Proportion of protein and protein fragments in the urine of patients with diabetes and eGFR <60ml/min. n=9-12, data ±SEM. ** p<0.001, *p<0.05 for fragment% between groups by ANOVA followed by post-hoc test for comparison of means.