UC Davis

Thiamine is an essential vitamin. Once in the cells, thiamine can be converted to thiamine diphosphate (TDP), thiamine monophosphate (TMP), thiamine triphosphate and other rarer derivatives. TDP is a cofactor for six metabolic enzymes in sugar, fat, and amino acid metabolism, supporting energy production. Thiamine deficiency leads to neurologic and cardiovascular damage resulting in beriberi and Wernicke’s encephalopathy. As ionic molecules, thiamine and its mono- and diphosphate derivatives (TMP, TDP) need transporters to cross plasma membranes. Various families of transporters including SLC22A1-2 (OCT1-2), SLC19A1-3 (RFC1, THTR1-2), SLC47A2 (MATE2/2K), SLC35F3, and SLC44A4 perform this task. In the past decade, various marketed drugs including trimethoprim have been found to inhibit the thiamine-specific transporter, THTR2. Currently, metabolome-wide impacts of thiamine deficiency and the tissue-specific importance of THTR2 for thiamine status have not been investigated. In addition, impacts of drug-thiamine interactions on thiamine pharmacokinetics and vitamer status in tissues are lacking. To explore these knowledge gaps, thiamine and its vitamers (TMP, TDP) were quantified using LC-MS/MS and metabolomics were performed with samples from thiamine deficiency, THTR2 deletion, and thiamine-trimethoprim interaction studies.

For the first chapter, thiamine deficiency was induced in mice via thiamine deficient (TD) diet. Tissue thiamine vitamers and metabolome were compared to thiamine sufficient mice. Within three weeks, TD had reduced thiamine vitamers in the brain, heart, liver, kidney, gonadal fat, RBCs, and plasma. While a handful of metabolites, including as 2-propanamidoacetic acid, N-ε-acetyllysine, N-methylvaline, a-aminoadipate, and 2-hydroxybutanoate, were consistently increased in at least four of the tested tissues, a large portion of the changes in the TD metabolome were tissue-specific. In addition, five plasma biomarkers for deficiency were identified.

For the second chapter, a transgenic mouse model was created by inserting the human THTR2 in the gut of the whole-body THTR2 knockout (KO) mice. When comparing the transgenic and the KO mice with the wildtype mice, these models had minor effects on thiamine status in the heart, liver, gonadal fat and kidney. However, these models showed significantly reduced brain thiamine. The metabolomic portion of the study showed reduced cerebral purines and hydroxyglutarates in mice lacking brain THTR2.

For the last chapter, a single dose of trimethoprim with thiamine was orally given to humans and mice in comparison to controls that received only thiamine. Thiamine exposure was hypothesized to decrease in the drug arm due to inhibition of THTR2-dependent intestinal absorption. Time-series plasma and urine were collected from the clinical trials while brain, heart, intestines and time-series plasma were collected from mice. Unexpectedly, both humans and mice showed increases in thiamine AUC in the trimethoprim arm. Metabolomic evaluations showed plasma biomarkers of the organic cation transporter 1 (OCT1) activity and urine biomarkers for OCT2/MATE2 (i.e., the multidrug and toxin extrusion transporter) activities were reduced. Because trimethoprim also inhibited these other transporters, it was unclear whether trimethoprim inhibited thiamine absorption in the gut.

Together, these studies have expanded our view of the metabolic impacts of thiamine deficiency, enhanced our understanding of the tissue specific roles of THTR2, and demonstrated clinically relevant drug-nutrient interactions.