We recently reported that genetic variability in the TKT gene encoding transketolase, a key enzyme in the pentose phosphate pathway, is associated with measures of diabetic sensorimotor polyneuropathy (DSPN) in recent-onset diabetes. Here, we aimed to substantiate these findings in a population-based KORA F4 study. In this cross-sectional study, we assessed seven single nucleotide polymorphisms (SNPs) in the transketolase gene in 952 participants from the KORA F4 study with normal glucose tolerance (NGT; n=394), prediabetes (n=411), and type 2 diabetes (n=147). DSPN was defined by the examination partof the Michigan Neuropathy Screening Instrument (MNSI) using the original MNSI > 2 cut-off and two alternative versions extended by touch/pressure perception (TPP) (MNSI>3) and by TPP plus cold perception (MNSI > 4). After adjustment for sex, age, BMI, and HbA1c, in type 2 diabetes participants, four out of seven transketolase SNPs were associated with DSPN for all three MNSI versions (all p≤0.004). The odds ratios of these associations increased with extending the MNSI score, for example, OR (95% CI) for SNP rs62255988 with MNSI>2: 1.99 (1.16-3.41), MNSI>3: 2.27 (1.26-4.09), and MNSI>4: 4.78 (2.22-10.26); SNP rs9284890 with MNSI>2: 2.43 (1.42-4.16), MNSI>3: 3.46 (1.82-6.59), and MNSI>4: 4.75 (2.15-10.51). In contrast, no associations were found between transketolase SNPs and the three MNSI versions in the NGT and prediabetes groups. The link of genetic variation in transketolase enzyme to diabetic polyneuropathy corroborated at the population level strengthens the concept suggesting an important role of pathways metabolising glycolytic intermediates in the evolution of diabetic polyneuropathy.
Read full abstract