Appl Biochem Biotechnol

DOI 10.1007/s12010-015-1966-7

Exploring the Effects of Subfreezing Temperature and Salt

Concentration on Ice Growth Inhibition of Antarctic
Gram-Negative Bacterium Marinomonas Primoryensis
Using Coarse-Grained Simulation

Hung Nguyen 1 & Thanh Dac Van 1,2 & Nhut Tran 1 &
Ly Le 1,2

Received: 16 November 2015 / Accepted: 21 December 2015

# Springer Science+Business Media New York 2016

Abstract The aim of this work is to study the freezing process of water molecules surround-
ing Antarctic Gram-negative bacterium Marinomonas primoryensis antifreeze protein
(MpAFP) and the MpAFP interactions to the surface of ice crystals under various marine
environments (at different NaCl concentrations of 0.3, 0.6, and 0.8 mol/l). Our result indicates
that activating temperature region of MpAFPs reduced as NaCl concentration increased.
Specifically, MpAFP was activated and functioned at 0.6 mol/l with temperatures equal or
larger 278 K, and at 0.8 mol/l with temperatures equal or larger 270 K. Additionally, MpAFP
was inhibited by ice crystal network from 268 to 274 K and solid–liquid hybrid from 276 to
282 K at 0.3 mol/l concentration. Our results shed lights on structural dynamics of MpAFP
among different marine environments.

Keywords MARTINI force field . MpAFPs . Coarse-grained simulation . Free energy landscape

Introduction

Organisms living in subfreezing regions have acquired the ability to resist extreme cold
conditions by producing antifreeze proteins (AFPs) which can inhibit ice crystal formation

* Hung Nguyen
[email protected]
* Ly Le
[email protected]

1
Life Science Laboratory of Institute for Computational Science and Technology, Ho Chi Minh City,
Vietnam
2
School of Biotechnology of Ho Chi Minh International University, Vietnam National University, Ho
Chi Minh City, Vietnam
 Appl Biochem Biotechnol

process. AFPs are composed of only protein or in combination with glycan via covalent bonds,
also called antifreeze glycoprotein (AFGP) [1–3]. AFPs protect their host by limiting the
growth of ice crystal inside and outside body fluids [4, 5]. AFPs and AFGPs have been found
in various Antarctic organism including fish [6–8], plants [9–11], insects [12–15], fungi [16],
and bacteria [3, 17–19]. The antifreeze mechanism of AFPs was determined through micro-
scopic analysis which showed that AFPs bind to ice surface and prevent ice crystallization
process. The difference between AFPs’ melting point and freezing point is known as thermal
hysteresis (TH), an important antifreeze indicator of different AFPs. In addition, AFPs at
interface between solid ice and liquid water inhibit the thermodynamically favored growth of
the ice crystal. Previous study showed that ice growth is kinetically inhibited by AFPs
covering the water accessible surfaces of ice [20]. As TH can be measured in laboratory
condition with a nanoliter osmometer [21], TH value of various AFPs was determined such as
fish (0.7–1.5 °C), plants (0.2–0.5 °C), insects (3–6 °C), and bacteria (less than 0.1 °C) [22].
Bacterial AFPs were found to have the lowest TH value and first demonstrated by Duman and
Olsen [3, 23]. AFPs are particularly sensitive to cooling condition as their molecular adaptation
to ice crystal is induced through the gradual formation of ice cascade surrounding AFP which
can also be interpreted as AFPs can be inactivated if subfreezing temperature suddenly drops
[24].
The binding site of AFPs has been found to be relatively hydrophobic and also contained
many potential hydrogen bond donors/acceptors. The idea of hydrogen bonds and the hydro-
phobic effects contribute to ice binding has been under investigation in recent decades [25, 26].
Hydrogen bonds were originally proposed to be the major interaction between AFPs and ice
crystal [27]. However, this hypothesis failed to provide a rational explanation on how an AFP
would preferably bind ice crystal. On the other hand, recent studies proposed that the
hydrophobic effect could have played an important function in the ice inhibition process and
provided with evidence that clathrate-like water on the hydrophobic ice-binding site (IBS) was
found to be released into the solvent upon ice binding, resulting in a gain in entropy [28, 29].
In addition, molecular dynamics simulation studies indicated that the relatively hydrophobic
IBS of AFPs is capable of reordering water molecules into an ice-like lattice and, instead of
shedding bound water molecules upon ice binding, the reordered water molecules might
facilitate the AFP’s interaction with ice by matching certain ice geometrical planes [30].
Although intriguing as it seems, these simulations failed to describe the molecular mechanism
explaining how an AFP acquires the ability to reorder water molecules in which binding
specificity and affinity can assist in the irreversible absorption of AFP into forming ice lattices.
These problems have been elucidated by the ice-binding mechanisms and the first crystal
structure of Antarctic bacterium AFPs. The largest AFP experimental structures demonstrated
that folds or Ca2+-bound parallel beta-helices with an extensive array of ice-like surface waters
were directly anchored via hydrogen bonds to the polypeptide backbone and the adjacent side
chains [26, 30–36].
In recent study, Shuaiqi Guo and colleagues found a Ca2+-dependent protein with molecular
density of 1.5-MDa has particular ice-binding property, called Marinomonas primoryensis
antifreeze protein (MpAFP). This protein has five non-ice-binding regions, in which on ice-
binding region II or RII constitutes approximately 90 % of its content and includes around 120
tandem repeats of 104-residue sequence. Furthermore, Ca2+ ion was found to be essential for
RII monomer folding with Ca2+/protein molar ratio of 10:1. Phase information of RII crystal
structure monomer was determined with a resolution of 1.35 Å by single-wavelength anom-
alous dispersion and molecular replacement methods with Ca2+ as the heavy atom. The RII
Appl Biochem Biotechnol

monomer folding is similar to Ca2+-bound immunoglobulin-like β-sandwich. Interestingly, the

Ca2+ ions are coordinated at the interface between each RII monomer and its symmetry-related
molecules, suggesting that these ions may involve in the stabilization of RII. The proposed role
of RII is to help the aerobic bacterium bind to surface ice in an Antarctic lake in order to
improve oxygen and nutrients accessibility [37].
The concentration of NaCl greatly affects the crystallization process of sea water. Since
NaCl dissociate the ice crystal by positively charged Na+ ion and negatively charged Cl− ion,
the NaCl concentration then will keep the water under liquid form when temperature reaches
lower than freezing point of °C. And NaCl concentration may vary dependent on chronolog-
ical and geographical condition and distribution. From the thermodynamic point of view, the
increase in NaCl concentration and anti-freezing mechanism of AFPs reduce ice crystallization
process, respectively, as such was found in AFPs. This leads to the assumption that the change
in NaCl concentration may have a certain degree of impact on the behavior and activity of
MpAFP. Therefore, in order to capture the full picture of MpAFP anti-freezing mechanism, our
objectives of this study were set to characterize the effects and the inhibiting ability of MpAFP
under freezing process in which different marine environments were incorporated by using
different NaCl concentrations of 0.3, 0.6, and 0.8 mol/l using coarse-grained simulation.

Materials and Methods

Simulated System Preparation The 3D structure of MpAFP was taken from Protein Data
Bank (PDB) with PDB entry 4KDV [37]. We used Visual Molecular Dynamics (VMD)
software [38] to show the structure of atomistic structure and then MARTINIZE version 2.2
to convert atomistic structure existing in PDB format into coarse-grained beads [39, 40]. Here,
each amino acid was modeled by one or two beads according to their specific sizes, which has
been classified into two broad categories: backbone and side chain beads.

Simulation Methods The GROMACS 4.5.5 package [41] with MARTINI force field for
coarse-grained model were used to run MD simulation. The periodic boundary conditions
were used throughout the simulation process; the electrostatic potential was shifted from 0.0 to
1.2 nm and the Lennard Jones (LJ) potential was shifted from 0.9 to 1.2 nm on all three axes
(x, y, and z) [42]. The complexes were positioned inside a cubic box at a distance of 1.2 nm
from the solute to the box surface and water model was used specifically in coarse-grained
water model [43]:

– The melting point of water models used in MD simulation was not near to real water molecules
[44, 45]. Such as the melting temperature of ice I(h) for several commonly used models of
water (SPC, SPC/E, TIP3P, TIP4P, TIP4P/Ew, and TIP5P) obtained from computer simula-
tions at p = 1 bar. Since the melting temperature of ice I(h) for the TIP4P model is now known
[46], it is possible to use the Gibbs–Duhem methodology [47] to evaluate the melting
temperature of ice I(h) for other potential models of water. Previous studies found that the
melting temperatures of ice I(h) for SPC, SPC/E, TIP3P, TIP4P, TIP4P/Ew, and TIP5P models
were 190, 215, 146, 232, 245, and 274 K, respectively. In this research, the coarse-grained
water model is represented as one of the major simplification of the solvent, which can be
implicitly or explicitly modeled as a van der Waals particle. In this water model, the effect of
polarization and the proper screening of interactions depending on the local environment is
 Appl Biochem Biotechnol

absent. In addition, the polarizable coarse-grained water molecules are represented by three
particles instead of one as in the standard Martini force field. The central particle W is neutral
and interacts with other particles in the system under the influence of the LJ potentials which
depicts standard water molecules. The additional particles WP and WM are bound to the
central particle and carry a positive and negative charge of + q and − q, respectively. They
interact with other particles via Coulomb function, and lack of LJ interactions. The bonds W-
WP and W-WM are constrained to a particular distance and are symbolized as character l. In
addition, the interactions between WP and WM particles inside the same coarse-grained water
bead are excluded making these particles Btransparent^ to each other. In turn, the charged
particles can rotate around the W particle. The dipole momentum of the water bead depends on
the position of the charged particles and can vary from zero (charged particles coincide) to 2lq
(charged particles at the maximal distance). A harmonic angle potential with equilibrium angle
θ and force constant Kθ was added to control the rotation of WP and WM particles which in
turn adjust the distribution of the dipole momentum. The average dipole momentum of the
water bead is dependent on the charge distribution and was set to be on average zero in an
apolar environment, such as the interior of the lipid bilayer. In contrast, some of non-zero
average dipole were observed in bulk water or in other polar environments. The mass of the
charged particles as well as the mass of the central particle was set to 24 amu and summed up
to 72 amu (the mass of four real water molecules) [43].
– The properties of water model in MD simulation were carefully considered since water
molecules play an important function as ubiquitous solvent in biological systems. In this
research, the parameterization of polarizable coarse-grained water model was used in
combination with the coarse-grained MARTINI force field. The three bead model
representing four water molecules was used to show the effective accountability of real
water-oriented polarity. Consequently, the dielectric screening of bulk water was
reproduced. At the same time, the parameterization was used as water model with bulk
water density and oil/water partitioning data which reproduced a similar level of accuracy
as in the standard MARTINI force field.
The steepest descent simulation for minimization was performed over 50,000 steps.
The equilibration was performed for 3 ns at temperature of 303 K using the Berendsen
algorithm [48] and 1 bar constant pressure with the damping coefficient of 0.1 ps of the
Parrinello–Rahman pressure [49]. The structures were generated as the configurations for
our MD simulation with eight temperatures including 268, 270, 272, 274, 276, 278, 280,
and 282 K. The final MD simulation allowed us to integrate the equations of motion with
a time step of 2 fs and our simulation run for totally 100 ns in the leap-frog algorithm [50].

Free Energy Landscape Calculations Free energy landscape is a method used to point out
possible conformations of a molecular entity, or the spatial positions of interacting molecules
in a system, and their corresponding energy levels. In this research, local minimum of an
energy landscape corresponding to metastable low temperature states of a thermodynamic
system was used to determine representative trajectories by identifying the local minimum
energy regions of free energy landscapes. The local minimum was defined by n-dimensional
reaction coordinate V = (V1,…,Vn) and was calculated using the function: ΔG(V) = −
kBT[lnP(V)–lnPmax], where P(V) represents the probability distribution and Pmax represents
the maximum of the distribution, which was subtracted to ΔG = 0 for lowest free energy
minimum [51, 52].
Appl Biochem Biotechnol

Results and Discussions

As shown in Fig. 1, the total energies (Etot is included the kinetic and potential terms of energy) of
MpAFPs system were represented as a function of time. We used the variation in total energy of
MpAFPs systems at subfreezing temperatures to determine their dependence on temperature and
NaCl concentration. This result indicates physical state transformation from liquid to solid of water
molecules in the simulated complexes when subfreezing temperatures were set from 268 to 282 K
and under different NaCl concentrations (NaCl concentrations were set to 0.3, 0.6, and 0.8 mol/l).
At 268 K, water molecules surrounding MpAFP were observed to transform from liquid solution to
completely frozen in 0.3, 0.6, and 0.8 mol/l. From 270 to 274 K, we found that water molecules
surrounding MpAFP were frozen in both 0.3 and 0.6 mol/l concentrations; liquid form was
observed at this temperature range in the case where NaCl concentration was 0.8 mol/l. At
276 K, water molecules surrounding MpAFP became ice-liquid hybrid in 0.3 and 0.6 mol/l and
liquid form in 0.8 mol/l. From 278 to 282 K, water molecules surrounding MpAFP was ice-liquid
hybrid in 0.3 mol/l while it remained in liquid form for both 0.6 and 0.8 mol/l. In addition, we also
found that Etot values of the complexes changed as NaCl concentrations changed. This resulted in
physical state transformation (ice crystal or ice-liquid hybrid or liquid solution). In addition, Etot
value has a tendency to increase as NaCl concentration increased because of the effect of higher
temperature conditions and ions are contributed to the change of kinetic energy lead to Etot value is
increased when NaCl concentration increased. Specifically, at 268 K, the ice state of treated
complexes with different NaCl concentrations of 0.3, 0.6, and 0.8 mol/l were found to be different
in term of Etot. The total energy fluctuated around −62e + 03 (kJ/mol) in 0.3 and 0.6 mol/l, and
around −59e + 03 (kJ/mol) in 0.8 mol/l; or along liquid form but the Etot of complex simulated at
278 K fluctuated around −55e + 03 (kJ/mol) in 0.6 mol/l and −53e + 03 (kJ/mol) in 0.8 mol/l. Thus,
the freezing process and energy states of the complexes were dependent upon NaCl concentration
in marine environment. This means water solvent treated with low NaCl concentration environment
can be frozen at higher temperature comparing to solvent treated with higher NaCl concentrations.
Specifically, water molecules surrounding MpAFPs were completely frozen from 268 to 274 K and
ice-liquid hybrid from 276 to 282 K at 0.3 mol/l. On the other hand, water molecules surrounding

Fig. 1 The total energy Etot(kJ/mol) of the MpAFP systems at different NaCl concentrations of 0.3, 0.6, and
0.8 mol/l and different temperatures from 268 to 282 K
 Appl Biochem Biotechnol

MpAFPs at 0.6 mol/l were frozen from 268 to 274 K, ice-liquid at 276 K and existed in liquid state
at temperatures of equal and greater than 278 K. Finally, for the simulated complexes with 0.8 mol/
l, only the complex at 268 K was completely frozen, while the remaining complexes existed in
liquid state. In sum, these results described the behaviors of water molecules surrounding MpAFP
at subfreezing temperature in which the formation and expansion of ice crystal led to the decrease of
thermodynamic properties of the system and total energy of MpAFPs systems.
As seen from Fig. 2, the radial distribution function was further used to describe the behavior of
water molecules surrounding MpAFP under subfreezing condition. The radial distribution func-
tions were measured to analyze the distances of water molecules that come into contact with
MpAFP. The measurement was represented by the gww(r) values or distances between water
molecules surrounding MpAFP. The gww(r) tended to increase in peak number and height when
water molecules surrounding MpAFP formed ice crystal or ice-liquid hybrid. The gww(r) was used
to identify temperature regions of water molecules surrounding MpAFP at different states (ice
crystal or ice-liquid hybrid or liquid) and subsequently its activity under subfreezing condition.
Figure 2 showed the gww(r) of water molecules surrounding MpAFP at different temperatures in
different marine environments (with NaCl concentrations of 0.3, 0.6, and 0.8 mol/l). At 268 K, the
gww(r) fluctuated around 5 (r) for first peak, 2.3 (r) for second peak, 1.85 (r) for third peak, and 1(r)
for fourth peak for all of three NaCl concentrations. This means water molecules surrounding
MpAFP were completely frozen for all three NaCl concentrations and that MpAFP was not able to
inhibit ice formation process. From 270 to 274 K, the gww(r) value of water molecules surrounding
MpAFP in 0.8 mol/l fluctuated around the first peak (3.9 (r)), the second peak (1.5 (r)), and the third
peak (1.3 (r)) while the gww(r) values of water molecules surrounding MpAFP in 0.3 and 0.6 mol/l
were not changed comparing to the complex simulated at 268 K. At these temperatures, the water

Fig. 2 The radial distribution function (RDF) gww(r) of water–water at NaCl concentrations of 0.3, 0.6, and
0.8 mol/l at different temperatures from 268 to 282 K. Here r is the distance between water–water molecules
Appl Biochem Biotechnol

molecules surrounding MpAFP existed in solid state at 0.3 and 0.6 mol/l and existed in liquid state
at 0.8 mol/l. At 276 K, the peak height of the water gww(r) surrounding MpAFP at 0.3 and 0.6 mol/l
were higher than at 0.8 mol/l. Such phenomenon occurred when the water molecules surrounding
MpAFP at 0.3 and 0.6 mol/l existed in ice-liquid hybrid form while in 0.8 mol/l existed in liquid
form. From 278 to 282 K, water molecules surrounding MpAFP were found to form ice-liquid
hybrid at 0.3 mol/l and liquid solution at 0.6 and 0.8 mol/l.
The bond-oriented order parameters was also used as a method to determine the crystalline
structure of liquid compound when which is changed from liquid to solid transition. The idea
of the bond order parameters is to capture the symmetry of bond orientations regardless of the
bond lengths. The bond was defined as the vector joining a pair of neighboring atoms. And the
neighboring atoms of a given atom as those atoms which have an interatomic distance less than
a cutoff radius of 3.5 Å, equal to the distance to the minimum between the first and the second
peaks of the pair correlation function [53]. Based on the bond-oriented order parameter
profiles, the results in Fig. 3 show that:

(a) For the complexes in 0.3 mol/l, the Fig. 3a showed that water molecules surrounding MpAFP
was completely frozen at 268 to 274 K and in ice-liquid hybrid form from 276 to 282 K.
(b) For the complexes in 0.6 mol/l, the Fig. 3b showed that water molecules surrounding
MpAFPs was completely frozen from 268 to 274 K and ice-liquid hybrid at 276 K. With
the temperatures of equal and greater than 278 K, the water molecules surrounding
MpAFP was in liquid solution.
(c) For the complexes in 0.8 mol/l, the Fig. 3c showed that water molecules surrounding
MpAFP was mostly frozen at 268 K. It was in liquid format temperatures of equal and
greater than 270 K.

Fig. 3 The profile of the globally averaged bond-orientation order parameter vs the temperatures. Here, the Q4,
Q6, Ŵ4, Ŵ6 were corresponding to the four bond order parameters to identify different crystal structures
 Appl Biochem Biotechnol

Figures 4 and 5 showed the free energy landscape of all complexes, which allowed us to
establish fundamental understanding on the interaction between water molecules and ice-binding
site of MpAFP at subfreezing temperatures using local minimum energy as indicator (in which x axis
represents the radius of gyration (Rg) and y axis represents the root mean-square deviation

Fig. 4 Free energy landscapes of MpAFP complexes from 268 to 274 K under different NaCl concentrations of
0.3, 0.6, and 0.8 mol/l, plotted as a function of root mean square deviation (RMSD) and radius of gyration (Rg)
Appl Biochem Biotechnol

Fig. 5 Free energy landscapes of MpAFP complexes from 276 to 282 K under different NaCl concentrations of
0.3, 0.6 and 0.8 mol/l), plotted as a function of root mean square deviation (RMSD) and radius of gyration (Rg)

(RMSD)). Based on free energy landscape values, we were able to select representative trajectories
by identifying the local minimum energy regions of free energy landscapes and also could find out
temperature regions that MpAFP remained active. When MpAFP was in ice or ice-liquid hybrid
environment, the residues of MpAFP would be inhibited by ice crystal which led to the formation of
 Appl Biochem Biotechnol

limited stable configurations. Therefore, local minimum energy regions are expected to be smaller.
In contrast, when MpAFP is in liquid environment, the residues of MpAFP form active configura-
tions which lead to expansion of local minimum energy regions as well as number local minimum
energy regions. In detail, Fig. 4 shows free energy landscape of MpAFP in marine environment with
various NaCl concentrations of 0.3, 0.6, and 0.8 mol/l from 268 to 274 K. At 268 K, local minimum
energy region of MpAFP was discrete. In this case, water molecules surrounding MpAFP were
completely frozen; the ice crystal inhibited and paralyzed the MpAFP activity. From 270 to 274 K,
the local minimum energy regions of MpAFP were found to be particularly discrete in 0.3 and
0.6 mol/l while that was extended in 0.8 mol/l. Under these temperatures, water molecules formed
ice crystal network in both of 0.3 and 0.6 mol/l; and liquid solution in 0.8 mol/l.
The free energy landscapes of MpAFPs under NaCl concentrations of 0.3, 0.6, and 0.8 mol/
l from 276 to 282 K were shown in the Fig. 5. At 276 K, with the result of free energy
landscape, we found that water molecules surrounding MpAFPs formed ice-liquid hybrid in
0.3 and 0.6 mol/l and liquid solution in 0.8 mol/l; which correspond to the expansion of the
local minimum energy regions. From 278 to 282 K, based on local minimum regions, we were
able to identify that water molecules surrounding MpAFP formed ice-liquid hybrid in 0.3 mol/l
and liquid solution at 0.6 and 0.8 mol/l concentrations.

Conclusions

In this work, we used coarse-grained simulation to study the active MpAFP at low temperatures in
combination with different NaCl concentrations. We have found basic behaviors of MpAFP in
inhibiting the freezing process of water molecules at low temperatures under various NaCl concen-
trations. We explored three important findings as follows: (1) the MpAFP could not be active under
low NaCl concentration (in 0.3 mol/l) and low temperatures. Under 0.6 mol/l NaCl concentration
and the temperatures of equal or greater than 278 K, the MpAFP obtained its active conformation at
these subfreezing temperatures. Under 0.8 mol/l NaCl concentration, MpAFP was activated at the
temperatures much lower than comparing to the MpAFP in 0.6 mol/l NaCl concentration (the
temperatures equal or larger 270 K). Therefore, active temperature region of MpAFP was extended
to lower temperatures when NaCl concentration increased. (2) In three cases, the phase transforma-
tion of water was found to be dependent on NaCl concentrations, the MpAFP systems allowed us to
search for phase space and to locate active conformation in consistent direction than standard single
temperature setup. (3) With coarse-grained simulation, the computing expenses were reduced a
significant way. However, we also failed to capture some detail information about MpAFP structure;
therefore atomistic MD simulation is strongly recommended for further studies.

Acknowledgments The work was funded by the Department of the Navy, Office of Naval Research under grant
number N62909-14-1-N234. The computing resources and support provided by Institute for Computational
Science and Technology, Ho Chi Minh City, Vietnam are gratefully acknowledged.

References

1. Nguyen, D. T., Colvin, M. E., Yeh, Y., Feeney, R. E., & Fink, W. H. (2002). The dynamics, structure, and
conformational free energy of proline-containing antifreeze glycoprotein. Biophysical Journal, 82, 2892–
2905.
Appl Biochem Biotechnol

2. Yeh, Y., & Feeney, R. E. (1996). Antifreeze protein: structure and mechanisms of function. Chemical
Reviews, 96, 601–618.
3. Nguyen, H., Le, L., & Ho, T. B. (2014). Computational study on ice growth inhibition of Antarctic bacterium
antifreeze protein using coarse grained simulation. The Journal of Chemical Physics, 140, 225101.
4. Raymond, J. A. (2011). Algal ice-binding proteins change the structure of sea ice. Proceedings of the
National Academy of Science of the United State of America, 108, E198.
5. Janech, M. G., Krell, A., Mock, T., Kang, J. S., & Raymond, J. A. (2006). Ice-binding proteins from sea ice
diatoms (Bacillariophyceae). Journal of Phycology, 42, 410–416.
6. DeVries, A. L., Komatsu, S. K., & Feeney, R. E. (1970). Chemical and physical properties of freezing point
depressing glycoproteins from Antarctic fishes. The Journal of Biological Chemistry, 245, 2901–2908.
7. Davies, P. L., Hew, C. L., & Fletcher, G. L. (1980). Fish antifreeze proteins: physiology and evolutionary
biology. Canadian Journal of Zoology, 66, 2611–2617.
8. Marshall, C. B., Fletcher, G. L., & Davies, P. L. (2004). Hyperactive antifreeze protein in a fish. Nature, 429,
153.
9. Worrall, D., Elias, L., Ashford, D., Smallwood, M., Sidebottom, C., Lillford, P., Telford, J., Holt, C., & Bowles,
D. (1998). A carrot leucine-rich-repeat protein that inhibits ice recrystallization. Science, 282, 115–117.
10. Atici, O., & Nalbantoglu, B. (2003). Antifreeze proteins in higher plants. Phytochemistry, 64, 1187–1196.
11. Griffith, M., & Yaish, M. W. F. (2004). Antifreeze proteins in overwintering plants: a tale of two activities.
Trends in Plant Science, 9, 399–405.
12. Tomchaney, A. P., Morris, J. P., Kang, S. H., & Duman, J. G. (1982). Purification, composition, and physical
properties of thermal hysteresis Bantifreeze^ protein from larvae of the beetle, Tenebrio molitor.
Biochemistry, 21, 716–721.
13. Hew, C. L., Kao, M. H., So, Y.-P., & Lim, K.-P. (1983). Presence of cystine-containing antifreeze proteins in
the spruce budworm, Choristoneura fumiferana. Canadian Journal of Zoology, 61, 2324–2328.
14. Schneppenheim, R., & Theede, H. (1980). Isolation and characterization of freezing-point depressing
peptides from larvae of Tenebrio molitor. Comparative Biochemistry and Physiology Part B: Comparative
Biochemistry, 67, 561–568.
15. Duman, J. G., Bennett, V., Sformo, T., Hochstrasser, R., & Barnes, B. M. (2004). Antifreeze proteins in
Alaskan insects and spiders. Journal of Insect Physiology, 50, 259–266.
16. Robinson, C. H. (2001). Cold adaptation in Arctic and Antarctic fungi. New Phytologist, 151, 341–353.
17. Gilbert, J. A., Hill, P. J., Dodd, C. E., & Laybourn-Parry, J. (2004). Demonstration of antifreeze protein
activity in Antarctic lake bacteria. Microbiology, 150, 171–180.
18. Muryoi, N., Sato, M., Kaneko, S., Kawahara, H., Obata, H., Yaish, M. W. F., Yeh, Y., & Feeney, R. E.
(1996). Antifreeze proteins: structures and mechanisms of function. Chemical Reviews, 96, 601–618.
19. Meister, K., Ebbinghaus, S., Xu, Y., John, G. D., DeVries, A., Gruebele, M., David, M. L., & Havenith, M.
(2012). Long-range protein-water dynamics in hyperactive insect antifreeze proteins. Proceedings of the
National Academy of Science of the United State of America, 110, 1617–1622.
20. Jorov, A., Zhorov, B. S., & Yang, D. S. (2004). Theoretical study of interaction of winter flounder antifreeze
protein with ice. Protein Science, 13, 1524–1537.
21. Braslavsky, I., & Drori, R. (2013). LabVIEW-operated novel nanoliter osmometer for ice binding protein
investigations. Journal of Visualized Experiments, 72, e4189.
22. Chattopadhyay, M. K. (2007). Antifreeze proteins of bacteria. Resonance, 12, 25–30.
23. John, G. D., & Olsen, T. M. (1993). Thermal hysteresis protein activity in bacteria, fungi and phylogenet-
ically diverse plants. Cryobiology, 30, 322–328.
24. Fletcher, G. L., Hew, C. L., & Davies, P. L. (2001). Antifreeze proteins of teleost fishes. Annual Review
Physiology, 63, 359–390.
25. Davies, P. L., Baardsnes, J., Kuiper, M. J., & Walker, V. K. (2002). Structure and function of antifreeze
proteins. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 357,
927–935.
26. Christopher, P. G., Robert, L. C., & Peter, L. D. (2011). Anchored clathrate waters bind antifreeze proteins to
ice. Proceedings of the National Academy of Science of the United State of America, 108, 7363–7367.
27. Devries, A. L., & Lin, Y. (1977). Structure of a peptide antifreeze and mechanism of adsorption to ice.
Biochimica et Biophysica Acta, 495, 388–392.
28. Chao, H., Houston, M. E., Jr., Hodges, R. S., Kay, C. M., Sykes, B. D., Loewen, M. C., Davies, P. L., &
Sönnichsen, F. D. (1997). A diminished role for hydrogen bonds in antifreeze protein binding to ice.
Biochemistry, 36, 14652–14660.
29. Baardsnes, J., Kondejewski, L. H., Hodges, R. S., Chao, H., Kay, C., & Davies, P. L. (1999). New ice-
binding face for type I antifreeze protein. FEBS Letters, 463, 87–91.
30. Nutt, D. R., & Smith, J. C. (2008). Dual function of the hydration layer around an antifreeze protein revealed
by atomistic molecular dynamics simulations. Journal of the American Chemical Society, 130, 13066–
13073.
 Appl Biochem Biotechnol

31. Gallagher, K. R., & Sharp, K. A. (2003). Analysis of thermal hysteresis protein hydration using the random
network model. Biophysical Chemistry, 105, 195–209.
32. Smolin, N., & Daggett, V. (2008). Formation of ice-like water structure on the surface of an antifreeze
protein. The Journal of Physical Chemistry B, 112, 6193–6202.
33. Wierzbicki, A., Dalal, P., Cheatham, T. E., 3rd, Knickelbein, J. E., Haymet, A. D., & Madura, J. D. (2007).
Antifreeze proteins at the ice/water interface: three calculated discriminating properties for orientation of type
I proteins. Biophysical Journal, 93, 1442–1451.
34. Yang, C., & Sharp, K. A. (2004). The mechanism of the type III antifreeze protein action: a computational
study. Biophysical Chemistry, 109, 137–148.
35. Yang, C., & Sharp, K. A. (2005). Hydrophobic tendency of polar group hydration as a major force in type I
antifreeze protein recognition. Proteins, 59, 266–274.
36. Garnham, C. P., Campbell, R. L., & Davies, P. L. (2011). Anchored clathrate waters bind antifreeze proteins
to ice. Proceedings of the National Academy of Science of the United State of America, 108, 7363–7367.
37. Guo, S., Garnham, C. P., Partha, S. K., Campbell, R. L., Allingham, J. S., & Davies, P. L. (2013). Role of
Ca2+ in folding the tandem β-sandwich extender domains of a bacterial ice-binding adhesion. FEBS Journal,
280, 5919–5932.
38. Humphrey, W., Dalke, A., & Schulten, K. (1996). VMD-visual molecular dynamics. Molecular Graphics,
14, 33–38.
39. Siewert, J. M., Risselada, H. J., Yefimov, S., Tieleman, D. P., & Alex, H. D. (2007). The MARTINI force
field: coarge grained model for biomolecular simulation. The Journal of Physical Chemistry B, 111, 7812–
7824.
40. Le, L., & Molinero, V. (2011). Nanophase segregation in supercooled aqueous solutions and their glasses
driven by the polyamorphism of water. The Journal of Physical Chemistry. A, 115, 5900–5907.
41. Hess, B., Kutzner, C., van der Spoel, D., & Lindahl, E. (2008). GROMACS 4: algorithms for highly
efficient, load-balanced, and scalable molecular simulation. Journal of Chemical Theory and Computation,
4, 435–447.
42. Darden, T., York, D., & Pedersen, J. (1993). Particle mesh Ewald: an Nlog (N) method for Ewald sums in
large systems. The Journal of Chemical Physics, 98, 10089–10092.
43. Yesylevskyy, S. O., Schäfer, L. V., Sengupta, D., & Marrink, S. J. (2010). Polarizable water model for the
coarse-grained MARTINI force field. PLoS Computational Biology, 6, e1000810.
44. Kar, R. K., & Bhunia, A. (2015). Biophysical and biochemical aspects of antifreeze proteins: using
computational tools to extract atomistic information. Progress in Biophysics and Molecular Biology, 119,
194–204.
45. Kar, R. K., & Bhunia, A. (2015). Will it be beneficial to simulate the antifreeze proteins at ice freezing
condition or at lower temperature? The Journal of Physical Chemistry B, 119, 11485–11495.
46. Sanz, E., Vega, C., Abascal, J. L. F., & MacDowell, L. G. (2004). Phase diagram of water from computer
simulation. Physical Review Letters, 92, 255701.
47. Kofke, D. A., & Post, A. J. (1993). Hard particles in narrow pores. Transfer-matrix solution and the periodic
narrow box. The Journal of Chemical Physics, 98, 1331–1336.
48. Berendsen, H. J. C., Postma, J. P. M., Gunsteren, W. F. V., Dinola, A., & Haak, J. R. (1984). Molecular
dynamics with coupling to an external bath. The Journal of Chemical Physics, 81, 3684–3690.
49. Parrinello, M., & Rahman, A. (1981). Polymorphic transitions in single crystals: a new molecular dynamics
method. Journal of Applied Physics, 52, 7182–7190.
50. Hockney, R. W., Goel, S. P., & Eastwood, J. (1974). Quit high resolution computer models of plasma.
Journal of Computational Physics, 14, 148–158.
51. Mu, Y., Nguyen, P. H., & Stock, G. (2005). Energy landscape of a small peptide revealed by dihedral angle
principal component analysis. Proteins, 58, 45–52.
52. Nguyen, H., Van, T. D., & Le, L. (2015). Coarse grained simulation reveals antifreeze properties of
hyperactive antifreeze protein from Antarctic bacterium Colwellia sp. Chemical Physics Letters, 638, 137–
143.
53. Wang, Y., Teitel, S., & Dellago, C. (2005). Melting of icosahedral gold nanoclusters from molecular
dynamics simulations. The Journal of Chemical Physics, 122, 214722.