|Home | About | Journals | Submit | Contact Us | Français|
VOLUME 291 (2016) PAGES 3668–3681
Based on the reprocessed diffraction data for Rv2837c crystal, a better model and electron density was obtained at a little higher resolution of 2.3 Å (Table 1). The new electron density shows that the two protein molecules in the asymmetric unit are in different states. In molecule A, the Fo − Fc map clearly indicates the existence of a catalytic water molecule between the two Mn2+ ions and pApA-like molecule. In molecule B, water molecule is not observed, and the Fo − Fc map suggests two AMPs (Fig. 10). As a precaution, both pApA and AMP were fitted in the active site of each protein molecule and refined. The result clearly shows that two AMPs should be predominant in molecule B and that a pApA fits into the electron density better than two AMPs in molecule A. It also indicates that AMPs are not compatible with the catalytic water molecule. Consequently, we hypothesize that even though molecule A contains a mixture of pApA and AMP, pApA still outweighs AMP (Fig. 11). Thus, in the final model, a pApA is placed in the active site of molecule A, and two AMPs are placed in the active site of molecule B. That means our structure not only catches a snapshot in the middle of catalysis, but it also shows the last moment of the catalysis before the product leaves the active site. It is also confirmed that the new structure is still supportive to the mechanism previously proposed in the paper.
The atomic coordinates and structure factors (code 5JJU; Rv2837c-pApA/AMP) have been deposited in the Protein Data Bank (http://wwpdb.org/).