Mutant T683W, like L1181F, also exhibited a lower yield and the CD signal for this mutant also showed a lower ellipticity than the wild-type protein indicating disturbance in the secondary structure of the protein. Thermal denaturation profile was also similar to a destabilized version of the protein with the unfolding temperature being 12uC below the second unfolding transition of the wild-type protein. This mutation also caused 30% loss in activity of the protein. The entrance of the Xe2 cavity consisted of two flaps formed by a shorter stretch of residues 685�C691 and a long loop spanning residues numbered 643�C680. The short loop connecting the strands b25 and b26 of A1 domain 
was quite surface exposed and the residues within were not conserved in the LPurL family. Unlike the surface exposed nature of the short loop, the forty amino acid long loop runs across the length of the FGAM synthetase domain and connects the A1 domain with the gene duplicated B2 domain and was found to be highly conserved. Cavity 3 at the interface of the FGAM synthetase and linker domains was fairly large and also quite close to the surface of the protein. Due to this reason, in silico analysis showed that it was difficult to fill this cavity as many possible alternate conformations for most residues occupied solvent exposed regions. However, F209W mutation that was expected to block the cavity at the expense of some clashes with the adjacent Leu182 and Glu186 residues was made. Surprisingly, the F209W mutation was very successful and had expression and purification profile at par with the wild-type. Both the secondary structure content and activity of the mutant enzyme was very similar to the wild-type protein. Thermal denaturation profile was also similar to the wild-type except that the second unfolding transition of the mutant was 5uC below that of the wild-type protein. The crystal structure of this protein was solved by molecular replacement at 1.5 A ? resolution. The mFo-DFc map depicted clear density for the tryptophan residue that had adjusted into the structure by adopting a similar conformation to that of the phenylalanine residue it replaced. The Ca and Cb atoms of both the amino acids in both structures superposed exactly. Per residue rmsd of the region around the Xe3 binding cavity was calculated between the F209W mutant and the StPurLXenon complex and it was found that mutagenesis at position 209 resulted in a local readjustment of the structure. To accommodate the larger tryptophan residue, side chains of amino acids in close proximity of the original phenylalanine residue like Glu186 and Leu182 residing on helix a7 readjusted adopting alternate rotamers. SCA was performed using programs made available by Ranganathan and coworkers. The PurL MSA was created using both sequence and structural alignment data as described in the material and method section. SCA results R428 yielded four independent sectors out of which structural and functional significance could be attributed to two sectors labeled as green and blue sectors. The green sector included multiple residues from each of the xenon binding cavities and this sector included residues from all four domains. While only 48% of the residues of the entire protein were found buried in the crystal structure, over 69% of the residues were buried in the green sector. Conservation scores of the residues were also analyzed using relative entropy metric and it was found that the positions of the green sector were highly conserved. High fraction of buried residues and high conservation scores suggest that this sector could represent the hydrophobic core of the protein and may have a role in the stability of the protein. The second sector PF-04217903 956905-27-4 termed as the blue sector constituted about 65% of the residues from the glutaminase domain and it included catalytically important amino acids.