Enzyme catalyzed phosphate transfer is a part of almost all metabolic processes. Such reactions are of central importance for the energy balance in all organisms and play important roles in cellular control at all levels. Mutases transfer a phosphoryl group while nucleases cleave the phosphodiester linkages between two nucleotides. The subject of our present study is the Lactococcus lactis β-phosphoglucomutase (β-PGM), which effectively catalyzes the interconversion of β-D-glucose-1-phosphate (β-G1P) to β-D-glucose-6-phosphate (β-G6P) and vice versa via stabile intermediate β-D-glucose-1,6-(bis)phosphate (β-G1,6diP) in the presence of Mg(2+). In this paper we revisited the reaction mechanism of the phosphoryl transfer starting from the bisphosphate β-G1,6diP in both directions (toward β-G1P and β-G6P) combining docking techniques and QM/MM theoretical method at the DFT/PBE0 level of theory. In addition we performed NEB (nudged elastic band) and free energy calculations to optimize the path and to identify the transition states and the energies involved in the catalytic cycle. Our calculations reveal that both steps proceed via dissociative pentacoordinated phosphorane, which is not a stabile intermediate but rather a transition state. In addition to the Mg(2+) ion, Ser114 and Lys145 also play important roles in stabilizing the large negative charge on the phosphate through strong coordination with the phosphate oxygens and guiding the phosphate group throughout the catalytic process. The calculated energy barrier of the reaction for the β-G1P to β-G1,6diP step is only slightly higher than for the β-G1,6diP to β-G6P step (16.10 kcal mol(-1) versus 15.10 kcal mol(-1)) and is in excellent agreement with experimental findings (14.65 kcal mol(-1)).