MODELING THE STRUCTURE OF MACROMOLECULAR MATTER

J. Hasek

Institute of Macromolecular Chemistry AS CR, 162 06 Praha 6, Czech Republic

A. Source of experimental data

Elucidation of a function of macromolecular systems heavily needs any experimentally verified knowledge of their structure and the nature of intermolecular interactions. The most comprehensive experimental information about the structure and a direct observation of the effect of these interactions is provided by diffraction methods. Their results are collected in several well developed structure databases containing nowadays more than 200 000 3-D structures offering thus huge number of experimental data on intermolecular interactions. Database containing macromolecular structures are: The database of organic polymers is available since 1995. The PolyBase - release 1.1 contains information about crystalline structure for more than 1.000 polymers and the full 3D structure is described for 261 of them. About 10 % of items belongs to polyamides, 18 % to polydienes, 16 % to polyesters, 12 % to polyhalides, 14 % to polyolefines, 13 % to polyoxides, 3 % to polysulfides, 3 % to polyurethanes and 13 % to other organic polymers.

Because of low ordering in polymeric materials, the quality of refined coordinates is lower in comparison with low molecular and biological samples. Therefore the next version will be added by re-refined coordinates under standard conditions to ensure higher reliability as a starting point for modeling of a real solid phase structure.

B. Theoretical approach

Computer systems (e.g.[7-10]) offer advanced methods for modeling the macromolecular assemblies and their properties. The user is offered by monomer template libraries which allow him to construct a spectrum of macromolecules of any required distribution of ideal conformations. However, to build a realistic model of macromolecular interactions in atomic resolution is highly problematic step without a deeper experimentally based knowledge of the structure behaviour of the studied fragments of molecules.

The software for real polymer structure modeling has to take into account formation of the microcrystalline regions by constraints fixing the desired mode of intermolecular bonding, the anisotropic distribution of the ``ordered dimensions'', the degree of polymerization, the description of a statistical disorder on the short distance level and the multimodal distribution of selected torsion angles.

Real structure modeling necessitates other experimental information on morphology of the material (lamelas, spherulites) and polymorphism. A special care deserves a formation of low ordered structure regions. Electron diffraction is of top interest because it concernes the ordered regions under 100 nm.

C. Joining the experimental and theoretical approaches

The energy minimizing gives usually good results as far as the local atomic arrangements. Contrary to that, the structure information obtained on crystalline polymer samples from diffraction experiment gives good results as far as the global geometry (small relative errors of the long interatomic distances).

Therefore, the optimal results are expected from molecular anealing based on suitable force fields [7] under the constraints hindering the atoms to move from ``too far'' their experimental positions (relatively to the estimated standard deviations resulting from the diffraction experiment). The atomic coordinates refined by classical crystallography ([12],[14], etc.) are re-refined by Discover[7] or by X-Plor [13]. The structures are drawn by O program[11].

It is believed that the next version of PolyBase including (in addition to the experimental data published in original papers) the energy minimized "experimental" 3-D view on the local arrangement of macromolecules in the microcrystalline regions will find direct utilization averywhere, where the molecular ordering has some role in the desired properties of matter. Our interest concerns the polymer body implants, polymer transported drugs, polymers possessing.

References :

  1. J. Hasek : Index of Polymers. Release 1.1. Available from CSCA (hasekj@imc.cas.cz), Prague 1994.
  2. J. Hasek : Database of Experimentally Determined Structures of Organic Polymers - PolyBase. Available from CSCA (hasekj@imc.cas.cz), Prague 1994.
  3. T.F. Koetzle, E.E. Abola, F.C. Bernstein, J.A. Callaway, J.J. Christian, B.R. Deroski, P.A. Esposito, A. Forman, P.A. Langdon, J.E. McCarthy, N.E. Oeder, R.K. Shea, J.G. Skora, K.E. Smith, D.R. Stampf: PDB : Protein Data Bank, (http://www.pdb.bnl.gov/). Chemistry Dept., Brookhaven Nat.Laboratory, NY 11973 Upton, 1993.}
  4. H.M. Berman, W.K. Olson, J. Westbrook, A. Gelbin, T. Demeny, S. Hsieh, D.Beveridge : NDB : Database of Nucleic Acid Structures. Rutgers University, New Brunswick, 1991. Nowadays available also via PDB.
  5. O. Kennard, F.H. Allen, O. Johnson, C.F. Macrae, J.M. Smith, W.D.S. Motherwell, J.J. Galloy, D.G. Watson, R.S. Rowland, S.E. Garner, J.E. Davies, G.F. Mitchell: CSDS : Cambridge Structural Database System. Vols.1,2,3,4., CCDC, 12 Union Road, CB2 1EZ Cambridge, 1993.
  6. ICSD : Inorganic Crystal Structure Database. (vg@fiz-karlsruhe.de). Gmelin--Institute f\"{u}r Anorganische Chemie und Fachinformationszentrum FIZ, Karlsruhe, Germany, 1993.
  7. BIOSYM : BIOSYM Technologies, Inc. -- San Diego, U.S.A. (http://www.biosym.com/)
  8. CHEM-X : Chemical Design Ltd, Oxon, UK, (chemx@applelink.apple.com).
  9. CERIUS2 : Computational chemistry. Molecular Simulations, Ltd.,\\ Cambridge 1995, (svalentine@msicam.co.uk).
  10. SYBYL : TRIPOS Associates, Inc., St.Louis, 1995
  11. T.A. Jones, M. Kjeldgaard: Manual for O, version 5.9., Uppsala University 1993.
  12. G. Sheldrick: SHELXL-93 - Program for structure refinement. Gottingen 1993
  13. A.T. Brunger : X-PLOR - System for Crystallography and NMR, Yale University 1990.
  14. CCP4 Suite, version 2.10 : Computer programs for protein crystallography, Daresbury Laboratory, 1994