Abstract: In studying transketolase (TK) from Saccharomyces cerevisiae, the majority of researchers use as cofactors Mg(2+) and thiamine diphosphate (ThDP) (by analogy with other ThDP-dependent enzymes), whereas the active site of native holoTK is known to contain only Ca(2+). Experiments in which Mg(2+) was substituted for Ca(2+) demonstrated that the kinetic properties of TK varied with the bivalent cation cofactor. This led to the assumption that TK species obtained by reconstitution from apoTK and ThDP in the presence of Ca(2+) or Mg(2+), respectively, adopt different conformations [Selivanov, V.A., Kovina, M.V., Kochevova, N.V., Meshalkina, L.E., Kochetov, G.A., 2004. Kinetic study of the H103A mutant yeast transketolase. FEBS Letters 567, 270-274]. Analysis of far-UV circular dichroism (CD) spectra and of data, obtained using methods of thermal denaturing, differential scanning calorimetry (DSC) and tryptophan fluorescence spectroscopy, corroborated this assumption. Indeed, the ratios of secondary structure elements in the molecule of apoTK, recorded in the presence of Ca(2+) or Mg(2+), respectively, turned out to be different. The two forms of the holoenzyme, obtained by reconstitution from apoTK and ThDP in the presence of Ca(2+) or Mg(2+), respectively, also differed in stability: the holoenzyme was more stable in the presence of Ca(2+) than Mg(2+).