Bookmark and Share

BCA Crystallography for Two Seasons: Fall and Spring

The British Crystallographic Association (BCA) has 327 members and includes an Industrial Group, a Physical Crystallography Group, a Chemical Crystallography Group, and a Biological Structure Group. These groups organize sessions for the joint Spring meeting and schedule other meetings during the year, either independently, cooperatively, or in conjunction with other scientific meetings. Accounts of the Fall 1992 and Spring 1993 meeting of the BCA, published in the BCA Newsletter, exemplify the breadth and depth of crystallographic investigation and application currently ongoing in the UK.

Glycosylated and Sugar-Binding Proteins

The functional significance of protein glycosylation is not well understood, and detailed information on the conformation of bound carbohydrates is relatively scarce. The contribution of macromolecular crystallography and molecular modelling to a better understanding of this extremely complex field was the subject of the meeting of the Biological Structure Group in Dec. 1992 at Keele University.

R. Dwek (Oxford, UK) presented an overview of the principles of glycobiology. He described technical advances that allow the separation of different glycoforms of an enzyme and the finding that the carbohydrate can effectively "fine tune" the catalytic activity of an enzyme, with rates. of catalysis that vary by a factor of up to 10. Dwek described evidence for a correlation between the rigidity of the protein and the degree of glycosylation, with important consequences for the susceptibility to cleavage by proteases. He described how the three carbohydrate substituents on tissue plasminogen activator (TPA) modulate three distinct activities of the enzyme, the fibrin binding activity, the plasmin binding activity, and the protease activity. Structural roles of glycans include stabilization of local structure and global structure, anchoring of proteins to the cell membrane, and participation in protein recognition. Dwek went on to describe how the relative abundance of different glycoforms can be used as clinical markers for conditions as diverse as pregnancy and rheumatoid arthritis, and concluded with an exciting development of a drug which inhibits glucose removal and is effective against HIV infection.

The glycan moieties in transferrins, an important class of plasma iron transport proteins, suggest that the carbohydrate occupies the gap between the two lobes of the protein, and may thereby influence both iron uptake and receptor binding (P. Lindley, Daresbury Lab., UK).

The role of the carbohydrate in haemagglutinin, one of the major surface glycoproteins of the influenza virus, was discussed by I. Wilson (San Diego, CA, USA). The virus invades target cells by first binding to a sialic acid receptor located in a dimple on the protein surface. The failure of the immune system to target the sialic acid binding site may be due to the carbohydrate, or to amino acid mutations surrounding the binding site. Desialization by neuraminidase reduces charge and mass heterogeneity, and can have dramatic effects on crystal quality. However, in some cases, such as IL2, the glycosylated protein yields the best crystals. G. Taylor (Bath, UK) presented the structure of a neuraminidase that is believed to be responsible for the breakdown of carbohydrates during bacterial infection. In spite of very low amino acid homology with viral enzymes (15%), the bacterial structure shows an identical topology of six 4-stranded β-sheets.

Observed structure of two domains (O3 and O4) of the CD4 receptor protein on the surface of T-lymphocytes that are the binding site for the AIDS virus. (From Body et al., Science 260, 979-983 (1993).

CD4 is the receptor protein of the AIDS virus and MHC class II antigens, and is located at the surface of T cells. The human protein has two glycosylation sites that are reported to be important for transport and correct folding of the protein. G. Lange (York, UK) described efforts to crystallize CD4 by expressing the protein without the glycosylation sites or with a simplified glycosylation pattern, or mimicking in vivo expression with removal of part of the carbohydrate by neuraminidase treatment. The crystal structure of domains 3 and 4 of the rat protein has been published by the York group (R.L. Brody et al., Science 260, 979-983, 1993). The CD2 cell surface protein that is involved in cell-cell interactions has four glycosylation sites, and a hydrophobic tail which serves as a membrane anchor. The protein was expressed in a lectin resistant cell line, so that the carbohydrate only extended to a mannose residue. Treatment with endoglycosidase-H removed all but the first carbohydrate residue, and this protein yielded crystals that diffracted to 2.8 Å (Y. Jones, Oxford, UK). The structure is made up of an N-terminal V-type immunoglobulin fold connected by a linker region to a C-terminal C-type domain. A flat "platform" on the "top" of the structure may form the binding sites for the LFA3 protein. It is postulated that the conserved carbohydrate may be involved in activation of signalling, while one of the sugars on the C-terminal domain may modulate the "on rate" of the CD2-LFA3 interaction.

 

Anomalous Dispersion

A symposium on Anomalous Dispersion was jointly sponsored by the Physical Crystallography and the Biological Structures groups at the 1993 BCA Spring Meeting. In a keynote lecture A.J.C. Wilson (IUCr) discussed the manifestation. of dispersion effects in the weighted reciprocal lattice and in the symmetry of Patterson maps, the existence of 90 dichromatic 3D geometrical crystal classes and 1651 dichromatic space groups, and the influence of dispersion on intensity statistics. The full text of his presentation appears in the June issue of the BCA Newsletter.

The use of synchrotron radiation to investigate the elastic tensor in GaP was described by C. Cousins (Exeter U.). Anomalous dispersion was used to reduce the scattering of Ga at the Ga K-edge. In crossing the absorption edge, the 006 rocking curve displayed inexplicable features such as broadening, double peaks, and asymmetry. After careful and extensive modelling, accounting for source and monochromator characteristics, and dispersion in the sample, the features could be replicated.

A.J.C. Wilson (left), keynote lecturer at the BCA Spring Meeting Symposium on Anomalous Dispersion shown here speaking with Theo Hahn, past president of the IUCr at the XV Congress and General Assembly In Bordeaux, France, July 1990. Photo courtesy of W.L. Duax.

The use of rotating anode and sealed tube sources to collect diffractometer data on NiAlPO21 was described by M. Helliwell (Manchester U.). Mo and CuKα radiation provided sufficient contrast to enhance Ni significantly in the difference Fourier map and locate bound Ni atoms in the framework of the zeolite-like structure. M. Hart (Manchester U.) discussed the principles of anomalous dispersion using instructive comparison with visible light, stressing the fact that f' and f'' are in fact tensors and should not be changed or refined independently, and described recent work on X-ray polarization and the stringent demands it makes upon optics.

Crystal Engineering

A Crystal Engineering session organized by S. Maginn of the Chemical Crystallography Group, Spring 1993, concerned the chemical modification of molecules to control solid state structure and properties and crystal growth. G. Desiraju (Hyderabad, India) discussed the critical role of weak intermolecular forces in controlling solid state structure, the directional character of interactions, and patterns revealed by systematic database analyses.

B. Jones (Cambridge) described controlled synthesis of crystalline polymers such as distyryl pyrizines by changing the size of substituents, the dynamics of single crystal to single crystal transformations, analysis of interactions in host-guest complexes, and solvent-less solid state reactions induced by grinding crystalline mixtures.

K. Harris (St. Andrews) described the preparation of acid amide co-crystals engineered by positioning potential hydrogen bond donors and acceptors on specific sites on the molecular building blocks. He examined the structural and dynamic properties of the designer crystals by solid state NMR and X-ray diffraction analysis.

Powder Pattern Interpretation

The Autumn meeting of the Industrial Group was held at the Health and Safety Executive Lab., Sheffield on Nov. 5, 1992. A. Jones, Dir. of Occupational Medicine and Hygiene Lab. welcomed participants. The theme of the meeting, "Applications of Profile Fitting and Simulation" was introduced by S. Fletcher.

I. Langford (U. of Birmingham, UK) discussed direction dependent diffraction effects in line profile analysis. Indexed Williamson-Hall plots could usually be used to identify the sources of "anisotropic" line profiles. These might include the shape and orientation of diffracting domains, structural "mistakes", nonstoichiometry or anisotropy of elastic constants. Examples of direction-dependent diffraction effects were shown by stacking faults in zinc oxide prepared from the oxalate, although the oxide prepared from the hydroxide showed none. J. Van Berkum (U. of Tech., Delft, The Netherlands) described a method of improving profile analysis using line profile simulations and showed that line profiles computer-simulated for atomic-scale models of strain could be compared with measured line profiles. The method was illustrated for Al-alloys containing misfitting Si precipitates. The use of profile fitting of diffraction data to characterize polymers despite limited numbers of peaks was demonstrated by J. Jutson (BICC Cables, Ltd.). Physical and mechanical properties of ethylene vinyl acetate (EVA) polymers, such as flexibility, stress crack resistance, and impact strength are in part related to their crystallinity which depends in turn on vinyl acetate (VA) present in the copolymer. The use of profile fitting to determine crystallinity, "crystallite size" and unit cell dimensions was demonstrated.

With audience participation W. David (Rutherford Appleton Lab., UK) described a logical probabalistic approach to determining the optimum number of Bragg peaks based on Bayesian analysis. The majority of the audience agreed with the optimum number of peaks for only one of the samples. C. Adam (Staffordshire U.) showed how a multiphase pattern fitting method allows not only the identification, but quantification of phases. The method developed is a cross between independent profile fitting and full-fledged Rietveld refinement. H. Singh Ubhi (Defense Research Agency, Farnborough, UK) discussed strain measurements in cold rolled and annealed titanium and titanium alloys. Changes in line broadening were followed as a function of thermo-mechanical history as an indirect measure of the energy stored during cold working and released during annealing. Peak broadening was found to decrease with annealing. S. Norval (ICI Wilton Research Centre, UK) described the use of simulated line profiles in the interpretation of experimentally observed line broadening from nickel particles in a catalyst system. The results obtained indicated a bimodal distribution of crystallite sizes where microstrain varied with size.

At the Spring Meeting of the Industrial Group the focus remained on powder diffraction analysis, and included a two-day training course conducted by R. Jenkins of the Int'l Center for Diffraction Data (ICDD). At the first formal session J. Kempston reviewed the early development of the powder method at UMIST. With the encouragement of W.L. Bragg, and the leadership of A.J. Bradley Manchester became the world center for powder diffraction and several industrial labs were created nearby.

Papers by manufacturing representatives described the Siemens D5000 modular system, Stoe position sensitive one-dimensional detectors and two-dimensional image plates, the Philips instrument line ranging from 'Compact' table top to the high precision 'MRD', the PDS120 position sensitive detector from Enraf-Nonius which collects the entire diffraction pattern with no movement of the 2θ arm, economical upgrading of customers existing equipment by Hiltonbrooks, and the Cryostream Cooler from Oxford Cryosystems.

R. Jenkins quoted an ICDD study that found users of the 20,000 active powder diffractometers are aware of the need for correct alignment but that few achieve it. He cautioned that for computer search match programs data needs to be correct to better than 1 part in 1,000 in "d". In a subsequent talk on State-of-the-Art Search-Match Software Jenkins demonstrated several PC based programs which permit rapid execution of traditional searches using the CD-ROM version of the database.

A. van den Boom of Philips described their "fuzzy-sets" approach to coping with complex mixtures and texture and sample related effects. P.U. Pennartz (Enraf-Nonius) described the "Windows" style software package. J. Nusinovici (Siemens/Socabim) extolled the virtues of matching the "full trace scan," rather than the usual edited d/l values. J. Shackleton solved a difficult metallurgical problem using a combination of search-match and lattice parameter refining programs from the Philips PC software suite. B. Baumgartner demonstrated the ease of use of the Stoe PC-based search-match software and the effect of data quality on effectiveness of its algorithm via analysis of prepared mixtures.

From the meeting it would appear that the ongoing battle of the computers has been won by the PC and that although Powder Diffraction found its roots back in the 1920s it is still a flourishing and developing technique.

The Industrial Group of the BCA has its own newsletter edited by J. Anwar and T. Carter and distributed by Philips.

Condensed from reports in the BCA Newsletter by A.G.W. Leslie, G. Bushnell-Wye, J. Jutson, T. Carter, and an anonymous author.