"Glutamine synthetases are one of the most ancient functioning enzymes in existence and these large oligomeric complexes are found in all extant forms of life where they play a critical role in nitrogen metabolism. Over the past five decades, extensive biochemical studies together with structural investigations have helped build a picture of the mechanism of functioning and regulation in the GSI and GSII families. The most divergent GSIII family, however, is poorly characterized and has only recently been recognized. Structural studies, using both cryo-EM and X-ray crystallography, were undertaken on the type III GS, GlnN, from the opportunistic human pathogen, Bacteroides fragilis, with a view to better understanding the GSIII family in the light of the known structure functionrelationships of the other GS enzymes, and to investigate the potential for the design of selective inhibitors against the divergent family. A low-resolution (16 Ã) reconstruction of GlnN was first determined by single particle cryo-EM and image processing. This structure revealed that GlnN was a double-ringed dodecamer with D6 symmetry and the arrangement of active sites within the hexameric rings closely matched the GSI structure. Following the design of a rapid purification protocol and improvements to the stability and solubility of GlnN, conditions were discovered for the production of diffraction quality inhibitor-bound crystals. A second better diffracting crystal form was also produced following proteolytic processing. The crystal structure of GlnN was solved to near atomic resolution (3.0 Ã) following phase extension of low-resolution SAD phases, taking into account the cryo-EM structure. The higher resolution of the crystal structure revealed that, surprisingly, the orientation of the hexameric rings in GlnN is inverted in comparison to other families. These results have raised interesting questions surrounding the mechanism and driving forces responsible for the evolution of quaternary structure in the GS enzymes and have suggested that the GSI and GSII structure arose following truncation of a large GSIII-like ancestor. Despite the differences in higher order assembly, the GlnN monomer displayed a high degree of similarity with the GSI and GSII structures in the core active site region, thus, suggesting a conservation of reaction mechanism. Structure-based multiple sequence alignment showed that the residues forming the nucleotide binding pocket are the least conserved in the GS superfamily, and several residue positions, which represent altered modes of ligand binding, were suggested as potential avenues for the design of selective inhibitors against GlnN."