Molecular Pathogenesis
Defect of a high-affinity thiamine transporter, SLC19A2, causes TRMA; however, it is still unclear how the absence of SLC19A2 expression results in the seemingly divergent disorders of megaloblastic anemia, diabetes mellitus, and deafness. Biochemical studies on fibroblasts or erythrocytes from individuals with TRMA showed that absence of the high-affinity component of thiamine transport results in low intracellular thiamine concentrations [Rindi et al 1992, Stagg et al 1999]. Defective RNA ribose synthesis caused by intracellular thiamine deficiency is thought to be the cause of megaloblastic changes in TRMA [Boros et al 2003]. Slc19a2 knockout mouse models have been developed [Oishi et al 2002, Fleming et al 2003]; the animal models manifest megaloblastic changes, diabetes mellitus, and sensorineural deafness (the main features of TRMA) when dietary thiamine levels are decreased [Oishi et al 2002]. While the mechanism of megaloblastic changes is still to be elucidated, these models showed defects in insulin secretion and selective loss of inner hair cells in cochlea [Oishi et al 2002, Liberman et al 2006].
Questions regarding TRMA disease pathogenesis that still require explanation include why individuals with TRMA do not have manifestations seen in dietary thiamine deficiency [Mandel et al 1984, Poggi et al 1984, Abboud et al 1985] and why the findings in TRMA are organ specific.
Studies showed that a second high-affinity thiamine transporter, encoded by SLC19A3, has major roles in intestinal thiamine uptake in mouse, accounting for the absence of overt thiamine deficiency in persons with TRMA [Reidling et al 2010]. In support of this, two Japanese brothers with a Wernicke's-like encephalopathy were reported to have compound heterozygous pathogenic variants in SLC19A3 [Kono et al 2009]. In addition, the difference in distribution of expression of the two thiamine transporters is critical in TRMA: in pancreatic endocrine cells, the expression of SLC19A2 is much higher than that of SLC19A3 and TRMA-associated SLC19A2 mutated alleles disrupt thiamine uptake significantly [Mee et al 2009]. Similarly, it is hypothesized that in TRMA, the other affected tissues (namely, bone marrow and cochlea) do not express or minimally express SLC19A3 [Eudy et al 2000, Rajgopal et al 2001].
Gene structure.
SLC19A2 spans approximately 22.5 kb and has six exons with a 237-bp 5' UTR and a 1,620-bp 3' UTR [Diaz et al 1999, Dutta et al 1999]. For a detailed summary of gene and protein information, see Table A, Gene.
Pathogenic variants.
SLC19A2 pathogenic variants are distributed throughout the gene with no apparent clustering or mutation hot spots. The majority of SLC19A2 pathogenic variants known to date are predicted to be null for protein because of pathogenic nonsense or frameshift variants and missense variants which would likely severely disrupt the folding and membrane targeting of the transporter. Additionally, some splicing variants and small deletion/duplication variants have been reported. Consistently, Balamurugan & Said [2002] showed that introducing several of these pathogenic variants into transfected HeLa cells resulted in impaired thiamine uptake.
Normal gene product. The 497-amino acid protein, the high-affinity thiamine transporter 1, is predicted to have 12 transmembrane domains.
Abnormal gene product. Pathogenic variants result in either a truncated protein from a premature stop codon or aberrantly folded protein caused by missense variants in transmembrane domains.