Although globular proteins are endowed with well defined three-dimensional structures, they exhibit substantial mobility within the framework of the given three-dimensional structure. The different types of mobility found in proteins by and large correspond to the different levels of organisational hierarchy in protein architecture. They are of considerable structural and functional significance, and can be broadly classified into (a) thermal and conformational fluctuations, (b) segmental mobility, (c) interdomain mobility and (d) intersubunit mobility. Protein crystallographic studies has provided a wealth of information on all of them. The temperature factors derived from X-ray diffraction studies provide a measure of atomic displacements caused by thermal and conformational fluctuations. The variation of displacement along the polypeptide chain have provided functionally significant information on the flexibility of different regions of the molecule in proteins such as myoglobin, lysozyme and prealbumin. Segmental mobility often involves the movement of a region or a segment of a molecule with respect to the rest, as in the transition between the apo and the holo structures of lactate dehydrogenase. It may also involve rigidification of a disordered region of the molecule as in the activation of the zymogens of serine proteases. Transitions between the apo and the holo structures of alcohol dehydrogenase, and between the free and the sugar bound forms of hexokinase, are good examples of interdomain mobility caused by hinge-bending. The capability of different domains to move semi-independently contributes greatly to the versatility of immunoglobulin molecules. Interdomain mobility in citrate synthase appears to be more complex and its study has led to an alternative description of domain closure. The classical and the most thoroughly studied case of intersubunit mobility is that in haemoglobin. The stereochemical mechanism of the action of this allosteric protein clearly brings out the functional subtilities that could be achieved through intersubunit movements. In addition to ligand binding and activation, environmental changes also often cause structural transformations. The reversible transformation between 2 Zn insulin and 4 Zn insulin is caused by changes in the ionic strength of the medium. Adenylate Kinase provides a good example for functionally significant reversible conformational transitions induced by variation in pH. Available evidences indicate that reversible structural transformations in proteins could also be caused by changes in the aqueous environment, including those in the amount of water surrounding protein molecules.

Recent experiments in this laboratory on structural transformations caused by controlled dehydration of protein crystals have been reviewed. X-ray diffraction patterns of the following crystals have been examined under varying conditions of environmental humidity in the relative humidity range of 100-75%: a new crystal form of bovine pancreatic ribonuclease A grown from acetone solution in tris buffer (I), the well-known monoclinic form of the protein grown from aqueous ethanol (II), the same form grown from a solution of 2-methyl pentan-2,4-diol in phosphate buffer (III), tetragonal (IV), orthorhombic (V), monoclinic (VI) and triclinic (VII) hen egg white lysozyme, porcine 2 Zn insulin (VIII), porcine 4 Zn insulin (IX) and the crystals of concanavalin A(X). I, II, IV, V and VI undergo one or more transformations as evidenced by discontinuous changes in the unit cell dimensions, the diffraction pattern and the solvent content. Such water-mediated transformations do not appear to occur in the remaining crystals in the relative humidity range explored. The relative humidity at which the transformation occurs is reduced when 2-methyl pentan-2,4-diol is present in the mother liquor. The transformations are affected by the crystal structure but not by the amount of solvent in the crystals. The X-ray investigations reviewed here and other related investigations emphasize the probable importance of water-mediated transformations in exploring hydration of proteins and conformational transitions in them.

A new form of L-histidine L-aspartate monohydrate crystallizes in space group P2₂ witha = 5.131(1),b = 6.881(1),c= 18.277(2) Å,β= 97.26(1)° and Z = 2. The structure has been solved by the direct methods and refined to anR value of 0.044 for 1377 observed reflections. Both the amino acid molecules in the complex assume the energetically least favourable allowed conformation with the side chains staggered between the α-amino and α-scarboxylate groups. This results in characteristic distortions in some bond angles. The unlike molecules aggregate into alternating double layers with water molecules sandwiched between the two layers in the aspartate double layer. The molecules in each layer are arranged in a head-to-tail fashion. The aggregation pattern in the complex is fundamentally similar to that in other binary complexes involving commonly occurring L amino acids, although the molecules aggregate into single layers in them. The distribution of crystallographic (and local) symmetry elements in the old form of the complex is very different from that in the new form. So is the conformation of half the histidine molecules. Yet, the basic features of molecular aggregation, particularly the nature and the orientation of head-to-tail sequences, remain the same in both the forms. This supports the thesis that the characteristic aggregation patterns observed in crystal structures represent an intrinsic property of amino acid aggregation.

The new form of L-arginine D-glutamate is monoclinic, P2₁, witha = 9.941(1),b = 4.668(2),c = 17.307(1) Å,β = 95.27(1)°, and Z = 2. In terms of composition, the new form differs from the old form in that the former is a monohydrate whereas the latter is a trihydrate. The structure has been solved by the direct methods and refined to R = 0.085 for 1012 observed reflections. The conformation of the arginine molecule is the same in both the forms whereas that of the glutamate ion is different. The change in the conformation of the glutamate ion is such that it facilitates extensive pseudosymmetry in the crystals. The molecules arrange themselves in double-layers stabilised by head-to-tail sequences involving main chains, in both the forms. However, considerable differences exist between the two forms in the interface, consisting of side chains and water molecules, between double-layers. A comparative study of the relationship between the crystal structures of L and DL amino acids on the one hand and that between the structures of LL and LD amino acid-amino acid complexes on the other, provides interesting insights into amino acid aggregation and the effect of chirality on it. The crystal structures of most hydrophobic amino acids are made up of double-layers and those of most hydrophilic amino acids contain single layers, irrespective of the chiralities of the amino acids involved. In most cases, the molecules tend to appropriately rearrange themselves to preserve the broad features of aggregation patterns when the chirality of half the molecules is reversed as in the structures of DL amino acids. The basic elements of aggregation in the LL and the LD complexes, are similar to those found in the crystals of L and DL amino acids. However, the differences between the LL and the LD complexes in the distribution of these elements are more pronounced than those between the distributions in the structures of L and DL amino acids.

The crystal structures of the complexes of L and DL histidine with formic acid have been determined as part of an effort to define biologically and evolutionarily important interactions and aggregation patterns. In terms of ionization state and stoichiometry they may be described as L-histidine formate formic acid and DL-histidine formate monohydrate respectively. In the L-histidine complex, amino acid molecules arranged in head-to-tail sequences centred around 2₁ screw axes are interconnected by formic acid molecules and formate ions. Histidine-formate interactions in the structure gives rise to a characteristic interaction pattern involving a linear array of alternating imidazole groups and formate ions. In DL-histidine formale monohydrate, head-to-tail sequences involving glide related molecules are interconnected through main chain-side chain interactions leading to amino acid layers. The layers are held together by formate ions and water molecules arranged in strings along which the ion and the molecule alternate. The patterns of amino acid aggregation in histidine complexes exhibit considerably higher variability than those in complexes involving arginine and lysine do.

The 𝛽-prism II fold lectins of known structure, all from monocots, invariably have three carbohydrate-binding sites in each subunit/domain. Until recently, 𝛽-prism I fold lectins of known structure were all from dicots and they exhibited one carbohydrate-binding site per subunit/domain. However, the recently determined structure of the 𝛽-prism fold I lectin from banana, a monocot, has two very similar carbohydrate-binding sites. This prompted a detailed analysis of all the sequences appropriate for two-lectin folds and which carry one or more relevant carbohydrate-binding motifs. The very recent observation of a 𝛽-prism I fold lectin, griffithsin, with three binding sites in each domain further confirmed the need for such an analysis. The analysis demonstrates substantial diversity in the number of binding sites unrelated to the taxonomical position of the plant source. However, the number of binding sites and the symmetry within the sequence exhibit reasonable correlation. The distribution of the two families of 𝛽-prism fold lectins among plants and the number of binding sites in them, appear to suggest that both of them arose through successive gene duplication, fusion and divergent evolution of the same primitive carbohydrate-binding motif involving a Greek key. Analysis with sequences in individual Greek keys as independent units lends further support to this conclusion. It would seem that the preponderance of three carbohydrate-binding sites per domain in monocot lectins, particularly those with the 𝛽-prism II fold, is related to the role of plant lectins in defence.

The discrepancy between the X-ray and NMR structures of Mycobacterium tuberculosis peptidyl-tRNA hydrolase in relation to the functionally important plasticity of the molecule led to molecular dynamics simulations. The X-ray and the NMR studies along with the simulations indicated an inverse correlation between crowding and molecular volume. A detailed comparison of proteins for which X-ray and the NMR structures appears to confirm this correlation. In consonance with the reported results of the investigations in cellular compartments and aqueous solution, the comparison indicates that the crowding results in compaction of the molecule as well as change in its shape, which could specifically involve regions of the molecule important in function. Crowding could thus influence the action of proteins through modulation of the functionally important plasticity of the molecule.

Internal mobility of the two domain molecule of ribosome recycling factor (RRF) is known to be important for its action. Mycobacterium tuberculosis RRF does not complement E. coli for its deficiency of RRF (in the presence of E. coli EF-G alone). Crystal structure had revealed higher rigidity of the M. tuberculosis RRF due to the presence of additional salt bridges between domains. Two inter-domain salt bridges and one between the linker region and the domain containing C-terminal residues were disrupted by appropriate mutations. Except for a C-terminal deletion mutant, all mutants showed RRF activity in E. coli when M. tuberculosis EF-G was also co-expressed. The crystal structures of the point mutants, that of the C-terminal deletion mutant and that of the protein grown in the presence of a detergent, were determined. The increased mobility resulting from the disruption of the salt bridge involving the hinge region allows the appropriate mutant to weakly complement E. coli for its deficiency of RRF even in the absence of simultaneous expression of the mycobacterial EF-G. The loss of activity of the C-terminal deletion mutant appears to be partly due to the rigidification of the molecule consequent to changes in the hinge region.

Structures of crystals of Mycobacterium tuberculosis RecA, grown and analysed under different conditions, provide insights into hitherto underappreciated details of molecular structure and plasticity. In particular, they yield information on the invariant and variable features of the geometry of the P-loop, whose binding to ATP is central for all the biochemical activities of RecA. The strengths of interaction of the ligands with the P-loop reveal significant differences. This in turn affects the magnitude of the motion of the `switch’ residue, Gln195 in M. tuberculosis RecA, which triggers the transmission of ATP-mediated allosteric information to the DNA binding region. M. tuberculosis RecA is substantially rigid compared with its counterparts from M. smegmatis and E. coli, which exhibit concerted internal molecular mobility. The interspecies variability in the plasticity of the two mycobacterial proteins is particularly surprising as they have similar sequence and 3D structure. Details of the interactions of ligands with the protein, characterized in the structures reported here, could be useful for design of inhibitors against M. tuberculosis RecA.

The structures of nine independent crystals of bitter gourd seed lectin (BGSL), a non-toxic homologue of type II RIPS, and its sugar complexes have been determined. The four-chain, two-fold symmetric, protein is made up of two identical two-chain modules, each consisting of a catalytic chain and a lectin chain, connected by a disulphide bridge. The lectin chain is made up of two domains. Each domain carries a carbohydrate binding site in type II RIPS of known structure. BGSL has a sugar binding site only on one domain, thus impairing its interaction at the cell surface. The adenine binding site in the catalytic chain is defective. Thus, defects in sugar binding as well as adenine binding appear to contribute to the non-toxicity of the lectin. The plasticity of the molecule is mainly caused by the presence of two possible well defined conformations of a surface loop in the lectin chain. One of them is chosen in the sugar complexes, in a case of conformational selection, as the chosen conformation facilitates an additional interaction with the sugar, involving an arginyl residue in the loop. The 𝑁-glycosylation of the lectin involves a plant-specific glycan while that in toxic type H RIPS of known structure involves a glycan which is animal as well as plant specific.