The role of the kidney in diabetic thiamine deficiency.
Diabetes is a chronic epidemic compounded by a burden of vascular complications including diabetic nephropathy. Diabetic nephropathy affects ~40% of patients and is characterised by increased urinary albumin excretion and decreased glomerular filtration rate. Diabetic patients exhibit ~75% decreased plasma thiamine concentration, linked to increased renal thiamine clearance. In streptozotocin-induced diabetic rats, decreased plasma thiamine concentration was also associated with a reduction in expression and activity of transketolase. Transketolase is a thiamine pyrophosphate-dependent enzyme and a critical component of the reductive pentose phosphate pathway, a metabolic pathway leading from glycolysis involved with the synthesis of ribose sugars. It is proposed that increasing the relative flux of glucose through the pentose phosphate pathway can ameliorate hyperglycaemic damage. This thesis investigates mechanisms mediating the increased renal thiamine clearance and the effects of thiamine therapy on type 2 diabetic patients with nephropathy. The hypothesis that hyperglycaemia increases flux through the hexosamine pathway, leading to increased O-glycosylation of the transcription factor Sp1 and subsequent decreased expression of thiamine transporters is investigated. Thiamine transporters in normal human kidney sections were found to be localised to the proximal tubule. Investigations in primary cultures of human proximal tubule epithelial cells and the HK-2 cell line have shown that there is a decreased expression (-48 to -80%) and abundance (-52 to -77%) of thiamine transporters in cells cultured in high glucose concentrations (26mM) with respect to low glucose concentrations (5 mM). There is only limited evidence supporting the involvement of the hexosamine pathway in these decreases. A double-blind, placebo-controlled study investigated the effect of thiamine supplementation on type 2 diabetic patients with microalbuminuria. Thiamine therapy restored plasma thiamine concentrations from 11nM to 98nM, exceeding the published median concentration observed in normal patients (64nM). After three months, thiamine therapy, but not placebo, caused a decrease in the urinary albumin excretion rate relative to baseline (-18 mg day-1). These results show promise for thiamine as a therapy for diabetic nephropathy.