Unexpected Rise of Glass Transition Temperature of Ice Crystallized

Letter Cite This: J. Phys. Chem. Lett. 2018, 9, 4512−4515

pubs.acs.org/JPCL

Unexpected Rise of Glass Transition Temperature of Ice Crystallized from Antifreeze Protein Solution Nobuaki Azuma,† Yuji Miyazaki,*,† Motohiro Nakano,† and Sakae Tsuda*,‡,§ †

Research Center for Structural Thermodynamics, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan Graduate School of Life Science, Hokkaido University, Sapporo 060-0810, Japan § Bioproduction Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Sapporo 062-8517, Japan Downloaded via UPPSALA UNIV on July 31, 2018 at 02:35:44 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Antifreeze protein (AFP) is known to bind to a single ice crystal composed of hexagonally arranged waters, hexagonal ice. To investigate the effect of the AFP binding to a general ice block that is an assembly of numerous hexagonal ice crystals, thermodynamic properties, dynamics, and the crystal structure of the ice block were examined in the presence of type I AFP (AFP-I). Previously, it was found that hexagonal ice has a glass transition based on the proton ordering in the ice lattice at low temperature. Measurements of heat capacity under adiabatic conditions, dielectric permittivity, and powder X-ray diffraction revealed that the glass transition occurs around 140 K in the ice containing 0.01− 1% (w/w) of the AFP-I, which is greater than the value for the pure hexagonal ice (ca. 110 K). These data imply that AFP affects the glass transition kinetics, i.e., the slowness of the proton migration in the ice block. Hence, adsorption of AFP molecules to each hexagonal ice is thought to change the physicochemical properties of the bulk ice.

T

should be an assembly of hexagonal ice crystals in the AFP-Ibound state. Heat capacities Cp per gram of the 1%, 0.1%, 0.01%, and 0.001% (w/w) AFP-I solutions are shown in Figure 1, where the heat capacities are divided by temperature T to see Cp anomalies more clearly. For comparison, the data of the 1% non-AFP globular enzyme lysozyme solution are also plotted. Cp steps were observed for all the solutions which were cooled to subzero temperature at an average rate of 1−2 K min−1. Hexagonal ice has a phase transition from proton disordering to proton ordering at Ttrs = 72 K.11−15 However, because the rate of the proton ordering becomes slower and slower as the temperature of the ice approaches the Ttrs, the proton ordering freezes above the Ttrs before ordering completely; this nonordered ice is called “glassy crystal”.16 When the ice that was cooled fast is warmed slowly, a heat capacity step due to glass transition can be observed, which is attributed to onset of the proton disordering during warming from the glassy state (exothermic) and back to the disorder as the temperature further rises (endothermic). Thus, around the Cp step regions, the change from exothermic to endothermic temperature drifts is regarded as the glass transition temperature Tg (see Figure S2). Because such an aqueous solution gives rise to phase-separated hexagonal ice when it is cooled to subzero temperature, the Cp steps can be

he glass transition of hexagonal ice occurs around 110 K as immobilization of the protons in constituent H2O molecules with a hydrogen-bond network structure, which are disordered in position by rotation of the H2O molecules and/or by hopping of the protons through the crystal lattice.1−3 Bjerrum pointed out that the lattice defects in hexagonal ice called Bjerrum faults are closely related with the proton migration.4 Antifreeze protein (AFP) was first discovered in some fishes living in the Antarctic Ocean.5 AFP has a unique binding ability to specific atoms of hexagonally arranged waters, or hexagonal ice, and inhibits further crystal growth of the ice.6,7 Many kinds of AFPs have been identified and characterized by their binding property to such a single ice crystal.8 However, the influence of AFP-binding on the bulk ice crystal, for example, the glass transition temperature of ice, has not been clarified yet. In this Letter, we report calorimetric and dielectric approaches for the first time to find the influence of AFP on the bulk ice crystal using type I AFP (AFP-I). The AFP-I used here originates from barfin plaice (Liposetta pinnifasciata).9 It consists of 7−10 isoforms with a single amphipathic α-helix structure, in which the amino acid sequence of the main isoform is DTASDAAAAAAATAAAAAAAAAATAKAAAEAAAATAAAAR. It was revealed that the AFP-I binds to the (202̅1) pyramidal plane of a single hexagonal ice crystal to suppress its further crystal growth.10 The measurements of heat capacity and dielectric permittivity of the AFP-I aqueous solutions are expected to clarify the influence of the AFP-I on the bulk ice that © XXXX American Chemical Society

Received: May 11, 2018 Accepted: July 26, 2018 Published: July 26, 2018 4512

DOI: 10.1021/acs.jpclett.8b01492 J. Phys. Chem. Lett. 2018, 9, 4512−4515

Letter

The Journal of Physical Chemistry Letters

Figure 2. Powder X-ray diffraction patterns of the ice generated from 1% AFP-I aqueous solution and pure hexagonal ice at 250 K together with literature values of hexagonal ice20 and cubic ice.19

Figure 1. Heat capacities Cp of 1% (red), 0.1% (blue), 0.01% (green), and 0.001% (purple) AFP-I and 1% lysozyme (gray) aqueous solutions. The Cp’s are divided by temperature T to clarify the Cp steps. Broken curves indicate the Cp of pure hexagonal ice.17 Arrows show the glass transition temperature Tg of each solution. Determination of the Tg was assisted by the temperature drift measurements shown in Figure S2. For the sake of clarity, the plots of the 0.001%, 0.01%, 0.1%, and 1% AFP-I solutions are shifted upward by 0.3, 0.6, 0.9, and 1.2 mJ K−2 (g solution)−1, respectively.

interaction between the AFP-I molecules and the ice surface strongly affects the glass transition of bulk hexagonal ice. To investigate the dynamical properties of the ice at temperatures below and above the glass transitions observed in the AFP-I bound hexagonal ice polycrystals, we measured enthalpy relaxation by adiabatic calorimetry and dielectric relaxation by dielectric permittivity measurements (see the Supporting Information for details). The relaxation times determined from the enthalpy and dielectric relaxation measurements are summarized in Figure 3. In the 1%, 0.1%, and 0.01% AFP-I solutions, the derived relaxation times show almost the same temperature dependence with the activation energies of 30−40 kJ mol−1. On the other hand, the temperature dependence of the relaxation times of the 0.001% AFP-I solution is rather close to that of pure hexagonal ice17 with the activation energy of 20 kJ mol−1 below 150 K.

regarded as glass transition of the ice.17 Surprisingly, the Tg’s of the 1%, 0.1%, and 0.01% AFP-I solutions were around 140 K, which is about 30 K higher than that of hexagonal ice.17 However, the Tg’s of the 0.001% AFP and 1% lysozyme solutions were 114 and 97 K, which are close to that of hexagonal ice. Because cubic ice, which is one of polymorphs of ice and a metastable phase at ambient pressure, also has a glass transition around 140 K,18 we measured powder X-ray diffraction patterns of the 1% AFP-I solution at subzero temperatures. For comparison, we also took powder X-ray diffraction patterns of pure hexagonal ice. It was confirmed that the ice polycrystals obtained from the 1% AFP-I solution indicate the diffraction patterns of not cubic ice19 but hexagonal ice20 with the same lattice parameters (see Figure 2). The difference between the hexagonal ice generated in AFP aqueous solutions and those done in most of other solutions is the shape of the embryonic ice crystal during crystallization. The ice crystallite derived from AFP solutions forms a bipyramidal shape, while the ice grown from solutions without AFPs forms a disk shape.21 Mahatabuddin et al. revealed from the concentration dependence of the ice crystallite shape for AFP-I aqueous solutions from microscope observations21 that the ice crystallites formed during the crystallization of the AFP-I dilute solutions alter at the concentration between 0.004% and 0.005%. Disklike ice crystallites were observed at concentrations lower than 0.004%, whereas bipyramidal ice crystallites were formed at concentrations higher than 0.005%. This means that AFPs have a critical ice-shaping concentration (CISC), which is the minimal concentration of AFP necessary to modify the ice morphology to a specific shape, i.e., bipyramidal shape (in the case of AFP-I, its CISC is around 0.0045%). Because above the CISC of the AFP-I the increment of the Tg of the hexagonal ices for the AFP-I solutions was found, it is suggested that the

Figure 3. Arrhenius plots of relaxation times for 1%, 0.1%, 0.01%, and 0.001% AFP-I aqueous solutions and pure hexagonal ice.17,27 4513

DOI: 10.1021/acs.jpclett.8b01492 J. Phys. Chem. Lett. 2018, 9, 4512−4515

Letter

The Journal of Physical Chemistry Letters

Figure 4. Schematic diagrams of (a) pure hexagonal ice polycrystals and (b) AFP-I bound hexagonal ice polycrystals.

the lattice defects in the ice. Because hydrogen bonding of regularly spaced array of polar amino acids such as Thr, Asn, and Asp to the ice surface is proposed,7 the proton mobility in the ice would be suppressed as if the protons were anchored to the amino acids inside the ice throughout (see Figure 4). In summary, we have measured heat capacities and dielectric permittivities of the AFP-I aqueous solutions for the first time to reveal unexpected increment of the Tg of the hexagonal ice polycrystals that have AFP-I molecules bound to their surfaces. Adsorption of the AFP-I molecules to the ice leads to increase in the activation energy for proton migration in the ice. This effect is contrary to dopants in hexagonal ice such as HF and alkali hydroxides, which decrease the activation energy of the proton transfer in the ice and thus decrease the Tg. The slowness of the proton migration would be strongly correlated with suppression of the crystal growth of the ice to which AFPs bind. Recently, dielectric measurements of gelatin solutions27,28 indicated that there is a relaxation mode with similar Tg and activation energy

If impurities such as HF are doped in hexagonal ice, the mobility of the protons is accelerated to decrease the glass transition temperature.22 Particularly, when KOH is doped in hexagonal ice, the rearrangement motion of the protons is greatly activated to undergo a first-order phase transition from paraelectric phase (Ih phase) to probably ferroelectric phase (XI phase) at 72 K.11−15 However, the ices derived from most biopolymer aqueous solutions show almost the same Tg as pure hexagonal ice.23−25 As seen in Figure 3, there are two processes for the proton transfer in pure hexagonal ice:4,26 process I above 230 K as generation of lattice defects as a result of the proton transfer and process II below 230 K as migration of the lattice defects. The relaxation times of the hexagonal ices crystallized from the 1%, 0.1%, and 0.01% AFP-I solutions have activation energies (30− 40 kJ mol−1) that are greater than that of pure hexagonal ice in the temperature region of process II. This suggests that adsorption of the AFP-I to the ice surface inhibits migration of 4514

DOI: 10.1021/acs.jpclett.8b01492 J. Phys. Chem. Lett. 2018, 9, 4512−4515

Letter

The Journal of Physical Chemistry Letters

(7) Wen, D.; Laursen, R. A. A Model for Binding of an Antifreeze Polypeptide to Ice. Biophys. J. 1992, 63, 1659−1662. (8) Yeh, Y.; Feeney, R. E. Antifreeze Proteins: Structures and Mechanisms of Function. Chem. Rev. 1996, 96, 601−618. (9) Kamijima, T.; Sakashita, M.; Miura, A.; Nishimiya, Y.; Tsuda, S. Antifreeze Protein Prolongs the Life-Time of Insulinoma Cells during Hypothermic Preservation. PLoS One 2013, 8, e73643. (10) Mahatabuddin, S.; Hanada, Y.; Nishimiya, Y.; Miura, A.; Kondo, H.; Davies, P. L.; Tsuda, S. Concentration-Dependent Oligomerization of an Alpha-Helical Antifreeze Polypeptide Makes it Hyperactive. Sci. Rep. 2017, 7, 42501. (11) Tajima, Y.; Matsuo, T.; Suga, H. Phase Transition in KOHDoped Hexagonal Ice. Nature 1982, 299, 810−812. (12) Tajima, Y.; Matsuo, T.; Suga, H. Calorimetric Study of Phase Transition in Hexagonal Ice Doped with Alkali Hydroxides. J. Phys. Chem. Solids 1984, 45, 1135−1144. (13) Kawada, S.; Dohata, H. Dielectric Properties on 72 K Phase Transition of KOH-Doped Ice. J. Phys. Soc. Jpn. 1985, 54, 477−479. (14) Kawada, S.; Takei, I.; Abe, H. Development of a New Relaxational Process Having Shortened Relaxation Time and Phase Transition in KOH-Doped Ice Single Crystal. J. Phys. Soc. Jpn. 1989, 58, 54−57. (15) Kawada, S. Acceleration of Dielectric Relaxation by KOHDoping and Phase Transition in Ice Ih. J. Phys. Chem. Solids 1989, 50, 1177−1184. (16) Suga, H.; Seki, S. Thermodynamic Investigation on Glassy States of Pure Simple Compounds. J. Non-Cryst. Solids 1974, 16, 171−194. (17) Haida, O.; Matsuo, T.; Suga, H.; Seki, S. Calorimetric Study of the Glassy State X. Enthalpy Relaxation at the Glass-Transition Temperature of Hexagonal Ice. J. Chem. Thermodyn. 1974, 6, 815−825. (18) Yamamuro, O.; Oguni, M.; Matsuo, T.; Suga, H. Heat Capacity and Glass Transition of Pure and Doped Cubic Ices. J. Phys. Chem. Solids 1987, 48, 935−942. (19) Shallcross, F. V.; Carpenter, G. B. X-Ray Diffraction Study of the Cubic Phase of Ice. J. Chem. Phys. 1957, 26, 782−784. (20) Goto, A.; Hondoh, T.; Mae, S. The Electron Density Distribution in Ice Ih Determined by Single-Crystal X-Ray Diffractometry. J. Chem. Phys. 1990, 93, 1412−1417. (21) Mahatabuddin, S.; Nishimiya, Y.; Miura, A.; Kondo, H.; Tsuda, S. Critical Ice Shaping Concentration (CISC): A New Parameter to Evaluate the Activity of Antifreeze Proteins. Cryobiol. Cryotech. 2016, 62, 95−103. (22) Ueda, M.; Matsuo, T.; Suga, H. Calorimetric Study of Proton Ordering in Hexagonal Ice Catalyzed by Hydrogen Fluoride. J. Phys. Chem. Solids 1982, 43, 1165−1172. (23) Itou, T.; Teramoto, A.; Matsuo, T.; Suga, H. Ordered Structure in Aqueous Polysaccharide. 5. Cooperative Order-Disorder Transition in Aqueous Schizophyllan. Macromolecules 1986, 19, 1234−1240. (24) Kawai, K.; Suzuki, T.; Oguni, M. Finding of an Unexpected Thermal Anomaly at Very Low Temperatures Due to Water Confined within a Globular Protein, Bovine Serum Albumin. Thermochim. Acta 2005, 431, 4−8. (25) Kawai, K.; Suzuki, T.; Oguni, M. Low-Temperature Glass Transitions of Quenched and Annealed Bovine Serum Albumin Aqueous Solutions. Biophys. J. 2006, 90, 3732−3738. (26) Sasaki, K.; Kita, R.; Shinyashiki, N.; Yagihara, S. Dielectric Relaxation Time of Ice-Ih with Different Preparation. J. Phys. Chem. B 2016, 120, 3950−3953. (27) Sasaki, K.; Panagopoulou, A.; Kita, R.; Shinyashiki, N.; Yagihara, S.; Kyritsis, A.; Pissis, P. Dynamics of Uncrystallized Water, Ice, and Hydrated Protein in Partially Crystallized Gelatin−Water Mixtures Studied by Broadband Dielectric Spectroscopy. J. Phys. Chem. B 2017, 121, 265−272. (28) Yasuda, T.; Sasaki, K.; Kita, R.; Shinyashiki, N.; Yagihara, S. Dielectric Relaxation of Ice in Gelatin−Water Mixtures. J. Phys. Chem. B 2017, 121, 2896−2901. (29) Kume, Y.; Miyazaki, Y.; Matsuo, T.; Suga, H. Low Temperature Heat Capacities of Ammonium Hexachlorotellurate and its Deuterated Analogue. J. Phys. Chem. Solids 1992, 53, 1297−1304.

in the relaxation modes of the ice crystallized in these solutions. To elucidate this interesting behavior of the AFP-I solutions further, calorimetric and dielectric studies for other AFPs, such as type III AFP (AFP-III), are in progress.

■

EXPERIMENTAL METHODS AFP-I derived from barfin plaice (Liposetta pinnifasciata) was provided by Nichirei Co. Four concentrations of the AFP-I solution (1%, 0.1%, 0.01%, and 0.001% (w/w)) were prepared by dissolving the desalinated and lyophilized AFP-I into Milli-Q water. For comparison, the 1% solution of hen egg-white lysozyme, which was purchased from Sigma-Aldrich, desalinated, and lyophilized, was also prepared. The heat capacity measurements were performed with a laboratory-made adiabatic calorimeter.29 The dielectric permittivity measurements were carried out with a Solatron impedance/gain-phase analyzer 1260 and dielectric interface 1296. The powder X-ray diffraction patterns were collected with a Rigaku X-ray diffractometer VariMax RAPID II. Details of the heat capacity, dielectric permittivity, and powder X-ray diffraction measurements are given in the Supporting Information.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.8b01492. Detailed heat capacity, dielectric permittivity, and powder X-ray diffraction measurements (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*Phone: +81-6-6850-5525. Fax: +81-6-6850-5526. E-mail: [email protected]. *Phone: +81-11-857-8912. Fax: +81-11-857-8983. E-mail: s. [email protected]. ORCID

Yuji Miyazaki: 0000-0001-9093-8868 Motohiro Nakano: 0000-0002-2599-5740 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Nichirei Co. for the gift of the AFP-I. This paper is Contribution No. 59 from the Research Center for Structural Thermodynamics.

■

REFERENCES

(1) Bernal, J. D.; Fowler, R. H. A Theory of Water and Ionic Solution, with Particular Reference to Hydrogen and Hydroxyl Ions. J. Chem. Phys. 1933, 1, 515−548. (2) Pauling, L. The Structure and Entropy of Ice and of Other Crystals with Some Randomness of Atomic Arrangement. J. Am. Chem. Soc. 1935, 57, 2680−2684. (3) Giauque, W. F.; Stout, J. W. The Entropy of Water and the Third Law of Thermodynamics. The Heat Capacity of Ice from 15 to 273 K. J. Am. Chem. Soc. 1936, 58, 1144−1150. (4) Bjerrum, N. Structure and Properties of Ice. Science 1952, 115, 385−390. (5) DeVries, A. L.; Wohlschlag, D. E. Freezing Resistance in Some Antarctic Fishes. Science 1969, 163, 1073−1075. (6) Knight, C. A.; Cheng, C. C.; DeVries, A. L. Adsorption of α-Helical Antifreeze Peptides on Specific Ice Crystal Surface Planes. Biophys. J. 1991, 59, 409−418. 4515

DOI: 10.1021/acs.jpclett.8b01492 J. Phys. Chem. Lett. 2018, 9, 4512−4515