Glassy water occurs in at least two broad categories: low-density amorphous (LDA) and highdensity amorphous (HDA) solid water. We perform out-of-equilibrium molecular dynamics simulations to study the transformations of glassy water using the ST2 model. Specifically, we study the known (i) compression-induced LDA-to-HDA, (ii) decompression-induced HDA-to-LDA, and (iii) compression-induced hexagonal ice-to-HDA transformations. We study each transformation for a broad range of compression/decompression temperatures, enabling us to construct a "P-T phase diagram" for glassy water. The resulting phase diagram shows the same qualitative features reported from experiments. While many simulations have probed the liquid-state phase behavior, comparatively little work has examined the transitions of glassy water. We examine how the glass transformations relate to the (first-order) liquid-liquid phase transition previously reported for this model. Specifically, our results support the hypothesis that the liquid-liquid spinodal lines, between a lowdensity and high-density liquid, are extensions of the LDA-HDA transformation lines in the limit of slow compression. Extending decompression runs to negative pressures, we locate the sublimation lines for both LDA and hyperquenched glassy water (HGW), and find that HGW is relatively more stable to the vapor. Additionally, we observe spontaneous crystallization of HDA at high pressure to ice VII. Experiments have also seen crystallization of HDA, but to ice XII. Finally, we contrast the structure of LDA and HDA for the ST2 model with experiments. We find that while the radial distribution functions (RDFs) of LDA are similar to those observed in experiments, considerable differences exist between the HDA RDFs of ST2 water and experiment. The differences in HDA structure, as well as the formation of ice VII (a tetrahedral crystal), are a consequence of ST2 overemphasizing the tetrahedral character of water. © 2013 AIP Publishing LLC.

Water exists in at least two families of glassy states, broadly categorized as the low-density (LDA) and high-density amorphous ice (HDA). Remarkably, LDA and HDA can be reversibly interconverted via appropriate thermodynamic paths, such as isothermal compression and isobaric heating, exhibiting first-order-like phase transitions. We perform out-of-equilibrium molecular dynamics simulations of glassy water using the ST2 model to study the evolution of LDA and HDA upon isobaric heating. Depending on pressure, glass-to-glass, glass-to-crystal, glass-to-vapor, as well as glass-to-liquid transformations are found. Specifically, heating LDA results in the following transformations, with increasing heating pressures: (i) LDA-to-vapor (sublimation), (ii) LDA-to-liquid (glass transition), (iii) LDA-to-HDA-to-liquid, (iv) LDA-to-HDA-to-liquid-to-crystal, and (v) LDAto-HDA-to-crystal. Similarly, heating HDA results in the following transformations, with decreasing heating pressures: (a) HDA-to-crystal, (b) HDA-to-liquid-to-crystal, (c) HDA-to-liquid (glass transition), (d) HDA-to-LDA-to-liquid, and (e) HDA-to-LDA-to-vapor. A more complex sequence may be possible using lower heating rates. For each of these transformations, we determine the corresponding transformation temperature as function of pressure, and provide a P-T "phase diagram" for glassy water based on isobaric heating. Our results for isobaric heating dovetail with the LDA-HDA transformations reported for ST2 glassy water based on isothermal compression/decompression processes [Chiu et al., J. Chem. Phys. 139, 184504 (2013)]. The resulting phase diagram is consistent with the liquid-liquid phase transition hypothesis. At the same time, the glass phase diagram is sensitive to sample preparation, such as heating or compression rates. Interestingly, at least for the rates explored, our results suggest that the LDA-to-liquid (HDA-to-liquid) and LDA-to-HDA (HDA-to-LDA) transformation lines on heating are related, both being associated with the limit of kinetic stability of LDA (HDA).

Trichosanthes Kirilowii lectin(TKL) is a new protein purified from a Chinese herb medicine, the tuber ofT1ichosanthes Kililowii maxim. It consists of two peptide chains, each with approximately 30kD molecule weight. TKL has diverse biochemistry, physiology and toxicology activities and binds strongly with galactose and lactose. It shows immunological cross-reactions with both ricin contained in seeds of Ricinus communis and t:1ichosanthin. another interesting protein from Tlichosanthes Kirilowii maxim with anti-AIDS effects. There is high sn·uctural similarity between the Achain of 1icin and trichosanthin. It is important to determine TKL structure and to compar·e the structural aspects of TKL, 1icin and trichosanthin in elucidating the structure-function relationships of these proteins at molecular level.After screening of crystallization conditions with the conventional hanging-drop method, better TKL crystals appear·ed under the following conditions: a drop prepar·ed by mixing 2~11 sample solution with concentration of 8.3mg/ml TKL and 2~1 reservoir solution, equilibrated again'st 500~1 reservoir solution, containin£ 0.5M LioS0.1 and 15% PEG-8000. The crystals belong to an orth~gonal sp~ce group with unit cell par·ameters of a=44.7 A. b=69.5 A and c=180.9 A. and there is one molecule in the asymmetric unit. 3 A diffraction data were collected at room temperature, using Mar Research Image Plate System in our laboratory. R. Kolatkm ar1d William I. Weis, Stanford University, Dept. Structural Biology, Stanford, CAGalactose-binding C-type lectins [·unction in semm glycoprotein clearar1ce, tumor cell recognition, ar1d organization of the extracellular mat:1ix. The crystal structure of a galactose-binding mutant of a Ctype m1imallectin has been solved unliganded and in complex with galactose and N-acetylgalactosar11ine (GalNAc). Three amino acid substitutions and insertion of a glycine-rich loop in wild-type mannose-binding protein A (MBP-A) gives a mutant (QPDWG) that exhibits specificity and affinity for galactose similar to naturallyoccuning galactose-binding C-type lectins. The 3-and 4-OH groups of galactose coordinate the Ca2+ at site 2 and form hydrogen bonds with amino acid residues that also coordinate the Ca2+. Galactose specificity is confened by a glycine-rich loop which holds Trpl89 in a position optrmal for packing against the apolar· face of the galactose ling, and which prevents mannose binding by steric exclusion. The st:11.1cture of the N-acetylgalactosamine/QPDWG complex shows that the 2-acetar11ido group of Gal.l\fAc is mien ted such that it could interact with the amino acid positions identified by site-directed mutagenesis (IobsL S.T. & D1ickamer, K., 1. Bioi. Chem., 271, 1996, in press) as being impo1iant in GalNAc-specific C-type lectin binding sites. An additional mutation of Thr202 -> His in QPDWG (to produce QPDWGH) exhibits an 8-fold increase in GalNAc specificity over s:alactose. The GalNAc/QPDWGH stmcture is currently being ;efined, and the preliminar-y results indicate that His202 is t...