Under these conditions, CCCP triggers AThTP production presumably by collapsing Δp. This is observed at 37°C as well as at 25°C. At 37°C, CCCP does not substantially affect the energy charge. Therefore, our results with the CV2 strain strongly suggest that Δp is more important than the energy charge as a factor controlling AThTP production. Further investigations showed, however, that factors other than Δp are also important for the control of intracellular AThTP levels. Indeed, when AThTP accumulates under carbon starvation, this accumulation is not accelerated by CCCP. Actually, we consistently found that under these conditions CCCP had

a negative effect on AThTP accumulation (Figure 7A). However, CCCP induced a greater

accumulation of AThTP in the presence of glucose (Figure 7B). Figure 7 Effect of CCCP on the AThTP content of BL21 cells in minimal M9 medium. The bacteria were MLN0128 supplier grown overnight in LB medium and then transferred to M9 minimal medium at 37°C in the absence of substrate (A) or in the presence of 10 mM D-glucose (B), L-malate (C) or in LB medium (D) with (●) or without (○) CCCP (50 μM). In B, iodoacetate was present at 1 (▲) and 5 (▼) mM final concentration. (Means ± SD, n = 3) Furthermore, Dabrafenib supplier the activating effect of glucose was counteracted by iodoacetate, suggesting that the activation is induced by a degradation product rather than by glucose itself. On the other hand, we found that L-malate was much less effective than glucose as an activator of AThTP production in the presence of CCCP (Figure 7C). A good effect of CCCP else was also obtained in LB medium (Figure 7D), probably because of the presence of amino acids entering the glycolytic pathway. This suggests that the unidentified activator can be produced by glucose but not by malate oxidation. It is interesting to point out that the enzyme catalyzing AThTP synthesis in vitro is also activated by an unidentified heat-stable factor [4]. ThTP inhibits the accumulation of AThTP As ThTP and AThTP

accumulate under different conditions and AThTP is never observed in the presence of ThTP, we wondered whether ThTP might inhibit the accumulation of AThTP. In order to check this possibility, we used BL21 strains overexpressing either E. coli AK or GST-hThTPase (a highly specific recombinant human ThTPase). When highly overexpressed in BL21 cells, bacterial AK catalyzes ThTP synthesis [21], leading to an accumulation of high amounts of ThTP (about 10 – 15% of total thiamine), whatever the composition of the medium (presence of glucose or not). Overexpression of AK leads to approximately a 1000-fold increase in AK protein compared to endogenous AK. GST-hThTPase is a highly specific and efficient enzyme that hydrolyzes all intracellular ThTP and when it is overexpressed, the cells are unable to accumulate significant amounts of ThTP [5].