DSpace at VNU: Coarse grained simulation reveals antifreeze properties of hyperactive antifreeze protein from Antarctic bacterium Colwellia sp.

DSpace at VNU: Coarse grained simulation reveals antifreeze properties of hyperactive antifreeze protein from Antarctic...

Accepted Manuscript Title: Coarse grained simulation reveals antifreeze properties of hyperactive antifreeze protein from Antarctic bacterium Colwellia sp Author: Hung Nguyen Thanh Dac Van Ly Le PII: DOI: Reference: S0009-2614(15)00639-9 http://dx.doi.org/doi:10.1016/j.cplett.2015.08.042 CPLETT 33239 To appear in: Received date: Revised date: Accepted date: 10-4-2015 13-8-2015 16-8-2015 Please cite this article as: H Nguyen, T.D Van, L Le, Coarse grained simulation reveals antifreeze properties of hyperactive antifreeze protein from Antarctic bacterium Colwellia sp., Chem Phys Lett (2015), http://dx.doi.org/10.1016/j.cplett.2015.08.042 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain *Highlights (for review) (1) We used coarse grained simulation to study the mechanism of Colwellia antifreeze protein (ColAFP) (2) We found the distribution and conformation of ice crystal network surrounding ColAFP at low temperature ip t (3) We identified phase space and located active conformation of water surrounding Ac ce p te d M an us cr ColAFP during the freezing process Page of 22 an us cr ip t *Graphical Abstract (pictogram) (for review) d M K K 285 263 Ac ce p g w-w(r) te 0.5 1.5 r (nm) 2.5 Page of 223.5 *The Manuscript Coarse grained simulation reveals antifreeze properties of hyperactive Hung Nguyen,1 Thanh Dac Van,1,2 and Ly Le1,2,* ip t antifreeze protein from Antarctic bacterium Colwellia sp Life Science Laboratory of Institute for Computational Science and Technology at Ho Chi Minh City, Vietnam School of Biotechnology of Ho Chi Minh International University, Vietnam National University, Vietnam us cr an ABSTRACT The novel hyperactive antifreeze protein (AFP) of Antarctic sea ice bacterium Colwellia sp provides a target for studying the protection of psychrophilic microgoranisms against freezing environment Interestingly, the M Colwellia sp hyperactive antifreeze protein (ColAFP) was crystallized without the structural dynamic characteristics Here, the result indicated, through coarse grained simulation of ColAFP under various subfreezing temperature, that ColAFP remains active at temperature of equal and greater than 275 K (~2 0C) Extensive d simulation analyses also revealed the adaptive mechanism of ColAFP in subfreezing environment Our result provides a structural dynamic understanding of the ColAFP I INTRODUCTION te Keywords: Antifreeze protein, ColAFP, coarse grained simulation, subfreezing temperature Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Ice binding proteins or commonly known as antifreeze proteins (AFPs) are specialized glycoproteins for inhibiting ice nucleation, growth and recrystallization.1 AFPs were found to be expressed in fishes,2-4 insects,5-7 plants8,9 and bacteria10-12 living in extreme cold environments AFPs exhibit ice inhibition mechanism by interacting through the gaps of ice cascade and preventing the ice crystals from linking to each other.13, 14 AFPs are classified based on their thermal hysteresis (TH) which defines their antifreeze activity under subfreezing temperature.15-17 There are commonly two types of AFPs: moderate AFPs and hyperactive AFPs Moderate AFPs are found in Antarctic fish which has TH of approximately 0.50C to 10C with AFP concentration at 10-3 mole level while hyperactive AFPs found in meal worm Tenebrio molitor could exhibit TH level up to 60C with 10-3 mole concentration of AFPs.18 Particularly, hyperactive AFPs found in Antarctic bacterium Marinomonas primoryensis could exhibit a TH of 20C at 2.9 x10-6 moles.19 Despite their variation in TH property, AFPs share similarity in molecular properties such as protein identity, structural arrangement.20 Currently, understanding on the ice inhibition process of various AFPs is still limited due to the lack of information on the molecular recognition and dynamics.21, Page of 22 22 Previous study was able to revealed the key structural component of AFP that involved in its ice binding property.23 Yuichi Hanada et al previously reported that an AFP identified from Antarctic bacterium Colwellia sp strain SLW05 (ColAFP) was homologous to AFPs from a wide variety of psychrophiles They analyzed antifreeze ip t activity and solved crystal structure of the Antarctic bacterium Colwellia sp and demonstrated ColAFP TH activity was approximately 40C at concentration of 0.14 mM and able to prevent hexagonal crystallization These results indicated ColAFP has the properties of hyperactive AFP Yuichi’s group further revealed the high resolution cr structure of hyperactive ColAFP (PDBID 3WP9) which has an irregular β-helical structure They observed that ColAFP has an ice-binding site which differs from the established ice-binding sites of other hyperactive AFPs us However, little is known upon the acting mechanism of this novel ice binding site.1 Recently, Rachana and colleagues used the correspondence analysis as an ordination technique to reveal that evolution of genetic factors is a selective and continuous process The method doesn’t make assumption about the data falling into discrete clusters an to allow accurate representation of continuous variation.24, 25 When studying on ColAFP protein sequence, Rachana et al found that bacterial and archaeal ice binding proteins (IBPs) have relatively higher average hydrophobicity than the eukaryotic IBPs Furthermore, bacterial IBPs and archaeal IBPs contain comparatively more strands, and M assumed to undergo higher selection pressure Additionally, molecular docking studies found that the ice crystals form more stable complex with the bacterial and archaeal AFPs than their eukaryotic orthologs Analysis of the docked structures showed that the ice-binding sites (IBS) in all of the orthologs facilitate ice-binding activity even d after mutation using the comprehensive IBS model of Typhula ishikariensis Notably, all of the IBS mutations have te been found to prefer polar and hydrophilic amino acids which assist in the ice-binding using “anchored clathrate mechanism” Horizontal gene transfer studies indicated the strong selection pressure favoring independent evolution of the IBPs in some Antarctic microorganisms facilitates their adaptation and survivability to the adversities in their niche.26 Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 In this study, we investigate the molecular recognition mechanism and interactions of ColAFP in freezing and subfreezing marine water using coarse grained molecular dynamics simulation method II MATERIALS AND METHODS Preparing the protein structure: The 3D structure of Antarctic sea ice bacterium Colwellia sp AFP was collected from Protein Data Bank (PDB code 3WP9) We used Visual Molecular Dynamics (VMD) software27 to analyze and visualize the atomistic structure of ColAFP And then, we used MARTINIZE version 2.2 to transform ColAFP atomistic structure to coarse grained model.28,29 In the coarse grained model, each amino acid represents by one or two beads according to their specific sizes, which has been classified into two broad categories: backbone and side chain beads Page of 22 Simulation Methods: The GROMACS 4.5.5 package30 with MARTINI force field for coarse grained model were used to generate MD trajectory The periodic boundary conditions were used throughout the simulation process; the electrostatic potential was shifted from 0.0 nm to 1.2 nm and the Lennard Jones (LJ) potential was shifted from 0.9 nm to 1.2 nm on all three axes (x, y and z).31 The complexes of ColAFP were positioned inside a cubic box at a distance of 1.2 nm from the solute and the box boundaries Specialized water model was used for coarse grained 32 and coarse grained model of sodium chloride was also included in the water box ip t model (MARTINI force field), with a concentration of 0.6 mol/L to mimic the condition of sea water The coarse grained water model is described  cr in the detail as follow: The coarse grained water model is a representation for one of the major simplification of the solvent, which us is either implicitly or explicitly modeled as a van der Waals particle The effect of polarization and the proper screening of interactions depending on the local environment is absent The polarizable coarse grained water molecules are represented by three particles instead of one as in the standard Martini force an field The central particle W is neutral and interacts with other particles in the system by means of the LJ interactions which is similar to 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 M with other particles via Coulomb function, and lack any LJ interactions The bonds W-WP and W-WM are constrained to a particular distance and is 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 d become “transparent” toward each other As a result, the charged particles can rotate around the W particle te 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 are at the maximal distance) A harmonic angle potential with equilibrium angle θ and force constant Kθ is added to control the rotation of WP and WM Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 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 is expected 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 are observed in bulk water or in some other polar environment The mass of the charged particles as well as of the central particle is set to 24 amu and totaling of 72 amu (the mass of four real water molecules)  Water molecules play an important function as ubiquitous solvent in biological systems, its treatment is crucial to the properties derived from MD simulation 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 that the oriented polarity of real water can be effectively accounted Consequently, the dielectric screening of bulk water is reproduced At the same time, the parameterization was used as a 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 Page of 22 The minimization was converged when the maximum force less than 0.01 kJ/mol/nm; the steepest descent minimization was performed over 50000 steps The equilibration was performed for ns at temperature of 303 K using Berendsen algorithm33 at atmospheric pressure with the damping coefficient of 0.1 ps of Parrinello-Rahman pressure.34 The structures were generated as the configurations for our MD simulation with 12 temperatures including 263 K, 265 K, 267 K, 269 K, 271 K, 273 K, 275 K, 277 K, 279 K, 281 K, 283 K and 285 K In addition, ip t we also performed MD simulation for compounds (including waters and Na+ and Cl- ions with concentration of 0.6 mol/l) at 271 K, 273 K, 275 K, 277 K, 279 K, 281 K and 303 K The final MD simulation allowed us to integrate the cr equations of motion with a time step of fs and our simulation run for totally 100 ns in the leap-frog algorithm.35 Free energy landscape: The free energy landscape along n-dimensional reaction coordinated V = (V1,…,Vn) was us shown by ΔG(V) = -kBT[ln P(V) – ln Pmax], with P(V) was the probability distribution and P max was the maximum of the distribution, which was subtracted to ΔG = for lowest free energy minimum.36, 37 an III RESULTS AND DISCUSSIONS The Fig 1a showed hydrophilic interaction of ColAFP and surrounding water molecules, it was found to M differ with varying temperatures The hydrophilic residues were exposed to liquid water, the interaction between hydrophilic residues and liquid water generates enthalpy lead to the formation of hydrogen bonds and subsequent cage-like structure; meanwhile hydrophobic residues interact with ice crystal to produce entropy As enthalpy and d entropy change, ColAFP ice-binding site switches it interaction to adapt with variation in temperature as was found in experimental conformation change12 Specifically, the results indicated that the hydrophilic interaction between te ColAFP and water molecules is subjected to reduce as ice crystal expanding and vice versa in order to interact with ice crystal Evidently, our simulation results were able to reveal this interaction mechanism in term of hydrophilic interaction between ColAFP and various phases of subfreezing water (ice crystal, ice-liquid hybrid, and liquid Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 water) The results of hydrophilic interaction demonstrate this adaptation mechanism of ColAFP in subfreezing water environment It shows that the hydrophilic interaction energy of ColAFPs complexes has a tendency to increase as temperature increases And when temperature reaches above 275 K, the hydrophilic interaction increases to maximum, this was found to be caused by conformational expansion of ColAFP which subsequently affects the hydrophilic interaction of the system Moreover, we found that the hydrophilic interaction between ColAFP and its surrounding water molecules reached highest with 287.19 (nm2) at 283 K With these findings, it is reasonable to assume that ColAFP can deploy it ability in prohibiting ice formation by adapting its ice binding site conformation to specific subfreezing temperature condition The radial distribution function (gw-w(r)) was used to describe the behavior of water molecules at subfreezing environments This function measures the distance of surrounding water molecules At subfreezing temperatures, the gw-w(r) exhibits long range correlation due to the reduction in water mobility The gw-w(r) shows a tendency to increase peak number and peak height when water is converted from liquid to crystal Fig 1b illustrates that the gww(r) of water-water in waters-ions complexes simulated at temperatures: 271 K, 273 K, 275 K, 277 K, 279 K, 281 K Page of 22 and 303 K It reveals that the water molecules formed ice crystal from 271 K to 279 K, ice-liquid hybrid at 281 K and liquid solution at 303 K These results indicate fundament states of subfreezing water: ice crystal, ice-liquid te d M an us cr ip t hybrid, and liquid Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Page of 22 ip t cr us an M d te Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Page of 22 Fig The hydrophilic interaction of ColAFP complexes (Fig 1a) and the radial distribution function of waterwater in waters-ions compound (waters were colored by cyan; Na+ and Cl- ions were colored by orange) simulated at 271 K, 273 K, 275 K, 277 K, 279 K, 281 K and 303 K (Fig 1b) The snapshots of waters-ions complexes simulated at 271 K, 281 K and 303 K indicates different water states in the form of ice crystal, ice-liquid hybrid, and liquid ip t accordingly were shown in Fig 1b Based on previous findings from Fig 1b, we were able to fundamentally establish temperature regions in which the surrounding water molecules of ColAFP can be classified (ice crystal, ice-liquid hybrid and liquid) and cr temperature regions that ColAFP is most likely to maintain its active conformation Fig 2a shows the radial distribution function of water - water in ColAFP complexes From initial observation of the radial distribution, we us divided our analysis based on the behavior of subfreezing water into three cases: ice crystal, ice-liquid hybrid, and liquid In the first case, our observation, from four temperature dependent systems of 263 K, 265 K, 267 K and 269 K, showed that the gw-w(r) values decreased when simulation temperatures increased In detail, the gw-w(r) values an were fluctuated around 4.75 for the first peak, 2.5 for the second peak, 1.75 for the third peak and 1.5 for the fourth peak accordingly It means that the water molecules surrounding ColAFP were found to exhibit ice crystal structure at subfreezing temperatures of lower than 269 K In the second case, temperature dependent systems of 271 K and M 273 K, the gw-w(r) values are fluctuated around 4.5 for first peak, 1.8 for second peak and 1.5 for third peak Here, we found that the water molecules exhibited ice-liquid hybrid form and that the water molecules near antifreeze protein was not forming crystal but distant water molecules were found to form crystal cage At 271 K, the ice part d in complex was found to be dominant compared with liquid part Interestingly, this phenomenon was reversed when te the system was at 273 K which revealing antifreeze property of ColAFP In the case of liquid water, the last six remaining systems (from 275 K to 285 K), the gw-w(r) values indicated a fluctuation around for the first peak, 1.4 for the second peak and 1.25 for the final peak In this case, the water molecules were entirely in liquid state Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Additionally, the radial distribution function of water - protein (gw-p(r)) in ColAFP complexes were also shown in Fig 2b, which describes the distance between water and protein changed according to simulated temperatures Like the gw-w(r) values, the states of water molecules around protein such as ice crystal, ice-liquid hybrid, and liquid were also determined using the gw-p(r) Subsequently, we investigate the temperature regions at which active ColAFP could be found with the ice crystal formation surrounding ColAFP We identified that ColAFP inhibits the ice crystal formation at temperatures of greater and equal 275 K (20C) In previous report, Yuichi’s group reported ColAFP was homologous to AFPs from a wide variety of psychrophiles They analyzed antifreeze activity and solved crystal structure of the ColAFP and demonstrated ColAFP TH activity was approximately 40C at concentration of 0.14 mM and able to prevent hexagonal crystallization Comparatively, our simulation results were strongly agreement with previous reports of Yuichi’s group This finding is described in greater detail in later sections Page of 22 ip t cr us an M d te Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Page 10 of 22 Fig The radial distribution function of water-water (Fig 2a) and water-protein (Fig 2b) in ColAFPs complexes The values reveal the activity of ColAFPs under temperature variations The mean square displacement (MSD) is a method used to describe fundamental dynamic quality, while the temperature dependence of average MSD is used to characterize the internal flexibility of water molecules As seen ip t from Fig 3a, the MSD values were increasing in accordance with their corresponding temperatures The MSDs reached highest at 285 K and lowest at 263 K Indicatively, MSDs can be used to establish subfreezing temperature te d M an us cr regions that water molecules exist in the form of ice crystal, ice-liquid hybrid, and liquid Fig Behavior of water at subfreezing temperatures reveals various phases that lead to ice crystal formation The Fig 3a and Fig 3b illustrate mean square displacement (MSD) and totally energy results of system as a function of Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 100 ns simulation time accordingly As shown in Fig 3b, the total energies (Etot) of ColAFP system were shown as function of time We used the variation in total energy of ColAFP systems at subfreezing temperatures to determine their dependence on temperature factor When subfreezing temperatures were set from 263 K to 273 K, the Etot of ColAFP systems reached stability in accordance with temperature setup as simulation time increased Specifically, the Etot of those systems was found to reach stability at ns for 263 K and 265 K, ns for 267 K, 20 ns for 269 K, 73 ns for 271 K and 98 ns for 273 K From this result, we found that the changes in total energy of ColAFP systems indicate physical state transformation from liquid to solid Especially, at 265 K, the existence of three energy states: the liquid state which ranged from ns to ns, the first solid state from ns to 40 ns and the second solid state from 40 ns to 100 ns When subfreezing temperatures of complexes reached from 275 K and over, the Etot was not significantly changed for all of 100 ns simulation This means the Etot was found to change for ColAFP complex at these subfreezing temperatures but not for the remaining temperatures Collectively, these results describe the behavior of water molecules surrounding ColAFPs at subfreezing temperature in which the formation and expansion of ice Page 11 of 22 crystal lead to decrease of thermodynamic properties of the system and total energy of ColAFP systems as an us cr ip t represented by the gww(r) and MSD (r2) Fig The root-mean-square deviation (RMSD) and average RMSD values of protein were illustrated as a function M of simulation time (Fig 4a) and temperature (Fig 4b) To describe fluctuation and stability of ColAFP in simulated complexes, we used the root-mean-square deviation (RMSD) as a function of simulation time (Fig 4a) and temperatures (Fig 4b) As seen from Fig 4, the d RMSD values have tended to increase when temperature increased, and which allowed us to divide into two separate te states: from 263 K to 271 K, the water molecules surrounding ColAFP were transformed from liquid to ice crystal lead to the RMSDs weakly fluctuated due to the inhibition of ColAFP by ice crystal; from 273 K to 285 K, the ColAFP’s RMSD fluctuated intensively because the water molecules surrounding ColAFP existed in the form of Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 liquid water Based on these results, we were able to determine temperature region that the ColAFP was energetically comparable and exhibited antifreeze function at subfreezing temperature This finding is described with details in free energy landscape section Page 12 of 22 ip t cr us an M d te Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Page 13 of 22 Fig The free energy landscapes of 263 K to 285 K reveals fundamental properties ColAFP at subfreezing temperatures The construction of free energy landscape of all complexes allows us to establish fundamental understanding on the interaction between water molecules and ice-binding site of ColAFPs at subfreezing ip t temperatures using local minimum energy as indication (Fig 5) (in which x axis was the radius of gyration (Rg) and y axis was the root-mean-square deviation (RMSD)) Using these analyses, we were able to select representative trajectories by identifying the local minimum energy regions of free energy landscapes And based on free energy cr landscape, we also was able to identify temperature regions that functional ColAFP remains active When ColAFP was in ice or ice-liquid hybrid environment, the residues of ColAFP were inhibited by ice crystal lead to formation us of limited stable configurations Therefore, local minimum energy regions were smaller In contrast, when ColAFP was in liquid environment, the residues of ColAFP formed functionally active configurations which lead to expansion of local minimum energy regions as well as number local minimum energy regions As shown in Fig 5, an the free energy landscapes of ColAFP were significantly expanded as simulation temperatures increased The free energy landscape of ColAFP showed 12 subfreezing systems (from 263 K to 285 K) indicates bound conformation between ColAFP and ice crystals or ice-liquid hybrid or liquid At these subfreezing temperatures, the activity of M ColAFP was suppressed by ice cage which leads to reduction in the energy expansion and quantity of protein complex configuration On the other hand, free energy landscape of ColAFP at subfreezing temperatures equal and greater than 275 K showed that ColAFP expanded its configuration which indicating highly flexible formation of d protein water complex Consequently, as the ColAFP obtained its flexibility in subfreezing environment, the ice te crystal expansion was successfully prevented and that water at these subfreezing temperature remains in liquid form These results revealed that, at subfreezing temperature of equal and greater than 275 K, the ColAFP can be functional in preventing ice crystal expansion and subsequent ice formation process In more detail, we divided the Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 free energy landscape of ColAFP under the effect of subfreezing temperature into three cases based on the local minimum energy Firstly, from our experimental results, we found that at subfreezing temperatures of less than or equal to 269 K, the local minimum energy regions were discrete because the water molecules surrounding ColAFP was able to form ice cage and subsequently preventing all movement of the system In the second case, at subfreezing temperatures of 271 K and 273 K, the water molecules surrounding ColAFP were predominantly found in the form of ice-liquid hybrid Due to this hybridization in ice formation, the local minimum energy regions were found to be considerably more flexible compared to the previous instance However, when subfreezing temperature was set at 275 K and higher, the local minimum energy regions of the temperatures started to expand This is an indication that the water molecules at these subfreezing temperatures was able to escape the ice formation phase and that active ColAFP was involved to this effect IV CONCLUSIONS We have successfully established structural dynamics understanding of ColAFP using coarse grained simulation under various subfreezing temperatures and found the results to be consistent with experimental Page 14 of 22 observations Our simulation analyses revealed the properties of ColAFP and surrounding water at subfreezing temperatures, and the active conformations that were energetically comparable and exhibiting antifreeze function at subfreezing temperature of equal and greater 275 K, and these results were found to be in agreement with experimental results Our study provided insights into current understanding that: (1) identification of active ColAFP conformation at specific subfreezing temperature in an unbiased way which allows ColAFP to adapt the ip t configuration of its ice binding site in accordance with subfreezing temperature; (2) rather than maximization of contact area between ColAFP and surrounding subfreezing water molecules, our results reveal ingenious way of ColAFP adaptation in conformation which specifically interacts and inhibit water crystal expansion; and (3) in three cr cases examined for the phase transformation of water, the twelve simulations systems allows us to search for phase us space and locate active conformation in consistent direction than standard single temperature setup AUTHOR INFORMATIONS Email: ly.le@hcmiu.edu.vn an * M ACKNOWLEDGMENT The work was funded by the Department of the Navy, Office of Naval Research under grant number N62909-14-1N234 The computing resources and support provided by Institute for Computational Science and Technology, Ho te REFERENCES d Chi Minh City, Vietnam are gratefully acknowledged Y Hanada, Y Nishimiya, A Miura, S Tsuda, and H Kondo Hyperactive antifreeze protein from an Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Atarctic sea ice bacterium Colwellia sp has a compound ice-binding site without repetitive sequences J FEBS, 281, 3576-3590 (2014) A L DeVries, S K Komatsu, and R E Feeney Chemical and physical properties of freezing point depressing glycoproteins from Antarctic fishes J Biol Chem., 245, 2901-2908 (1970) P L Davies, C L Hew, and G L Fletcher Fish antifreeze proteins: physiology and evolutionary biology Can J Zool., 66, 2611-2617 (1980) C B Marshall, G L Fletcher, and P L Davies Hyperactive antifreeze protein in a fish Nature, 429, 153 (2004) C L Hew, M H Kao, Y.-P So, and K.-P Lim Presence of cystine-containing antifreeze proteins in the spruce budworm, Choristoneura fumiferana Can J Zool., 61, 2324-2328 (1983) R Schneppenheim, and H Theede Isolation and characterization of freezing-point depressing peptides from larvae of Tenebrio molitor Comp Biochem Phys B Biochem Mol Biol., 67, 561-568 (1980) J G Duman, V Bennett, T Sformo, R Hochstrasser, and B M Barnes Anti-freeze proteins in alaskan insects and spiders Insect Physiol., 50, 259-266 (2004) O Atici, and B Nalbantoglu Antifreeze proteins in higher plants Phytochemistry 64, 1187-1196 (2003) Page 15 of 22 M Griffith, and M W F Yaish Antifreeze proteins in overwintering plants: a tale of two activities Trends Plant Sci., 9, 399-405 (2004) 10 N Muryoi, M Sato, S Kaneko, H Kawahara, H Obata, M W F Yaish, Y Yeh, and R E Feeney Antifreeze proteins: structures and mechanisms of function Chem Rev., 96, 601-618 (1996) 11 K Meister, S Ebbinghaus, Y Xu, G D John, A DeVries, M Gruebele, M L David, and M Havenith ip t Long-range protein-water dynamics in hyperactive insect antifreeze proteins Nat Acad of Sci., 110, 16171622 (2012) 12 H Nguyen, L Le, and T B Ho Computational study on ice growth inhibition of Antarctic bacterium cr antifreeze protein using coarse grained simulation J Chem Phys., 140, 225101 (2014) growth inhibition Pol., 44, 690-698 (2012) us 13 H Nada, and Y Furukawa Antifreeze proteins: computer simulation studies on the mechanism of ice 14 J A Raymond, and A L DeVries Adsorption inhibition as a mechanism of freezing resistance in polar fish Proc Natl Acad Sci USA, 74, 2589-2593 (1977) Biophys Chem., 109, 137-148 (2004) an 15 C Yang, and A S Kim The mechanism of the type III antifreeze protein action: a computational study 16 P G Christopher, L C Robert, L D Peter Anchored clathrate water bind antifreeze proteins to ice M National Acad of Sci., 108, 7363-7367 (2011) 17 P G Christopher, A G Jack, P H Christopher, L C Robert, L.-P Johanna, and L D Peter A Ca2+dependent bacterial antifreeze protein domain has a novel β-helical ice-binding fold Biochem., 411, 171- d 180 (2008) te 18 L A Graham, Y C Liou, V K Walker, and P L Davies Hyperactive antifreeze protein from beetles Nature, 388, 727-728 (1997) 19 C P Garnham, J A Gilbert, C P Hartman, R L Campbell, J.Laybourn-Parry, and P L Davies A Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Ca2+dependent bacterial antifreeze protein domain has a novel b-helical ice-binding fold J Biochem., 411, 171-180(2008) 20 Z Jia, and P L Davies Antifreeze proteins: anunusual receptor-ligand interaction Trends Biochem Sci., 27, 101-110 (2002) 21 A J Scotter, C B Marshall, L A Graham, J A Gilbert,C P Granham, and P L Davies The basis for hyperactivity of antifreeze proteins Cryobiology, 53,229-239 (2006) 22 N Pertaya, C Marshall, Y Celik, P L Daives, and I Braslavsky Direct visualization of spruce budworm antifreeze protein interacting with ice crystals:basal plane affinity confers hyperactivity J Biophys., 95,333-341 (2008) 23 H Kondo, Y Hanada, H Sugimoto, T Hoshino, C P Garnham, P L Davies, and S Tsuda Ice-binding site of snow mold fungus antifreeze protein deviates from structural regularity and high conservation PNAS, 109, 9360-9365 (2012) 24 R Banerjee, A Roy, F.Ahmad, S Das, and S Basak Evolutionary patterning of hemagglutinin gene sequence of 2009 H1N1 pandemic J Biomol Struct Dyn., 29, 733-742 (2012) Page 16 of 22 25 S Basak, T Banerjee, S K Gupta, and T C Ghosh Investigation on the causes of codon and amino acid usages variation between thermophilic Aquifex aeolicus and mesophilic Bacillus subtilis J Biomol Struct Dyn., 22, 205-214 (2004) 26 R Banerjee, P Chakraborti, R Bhowmick, and S Mukhopadhyay Distinct molecular features facilitating ice-binding mechanisms in hyperactive antifreeze proteins closely related to an Antarctic sea ice bacterium ip t J Biomol Struct Dyn., 5, 1- 18 (2014) 27 W Humphrey, A Dalke, and K Schulten VMD-visual molecular dynamics Molecular Graphics, 14, 3338 (1996) cr 28 J M Siewert, H J Risselada, S Yefimov, D P Tieleman, and H D Alex The MARTINI Force field: Coarge Grained Model for Biomolecular Simulation J Phys Chem B, 111, 7812-7824 (2007) us 29 L Le, and V Molinero Nanophase segregation in supercooled aqueous solutions andtheir glasses driven by the polyamorphism of water J Phys Chem.A, 115, 5900-5907 (2011) 30 B Hess, C Kutzner, D van der Spoel, E Lindahl GROMACS 4: Algorithms for highly efficient, load- an balanced, and scalable molecular simulation Chem Theory Comput., 4, 435-447 (2008) 31 T Darden, D York, and J Pedersen Particle mesh Ewald: An Nlog (N) method for Ewald sums in large systems.J Chem Phys., 98, 10089-10092 (1993) M 32 S O Yesylevskyy, L V Schäfer, D Sengupta, and S J Marrink Polarizable water model for the coarsegrained MARTINI force field PLoS Comput Bio., 6, e1000810 doi:10.1371/journal.pcbi.1000810 (2010) 33 H J C Berendsen, J P M Postma, W F V Gunsteren, A Dinola, and J R Haak Molecular dynamics d with coupling to an external bath J Chem Phys., 81, 3684-3690 (1984) te 34 M Parrinello, A Rahman Polymorphic transitions in single crystals: A new molecular dynamics method Appl Phys., 52, 7182-7190 (1981) 35 R W Hockney, S P Goel, J Eastwood Quit high resolution computer models of plasma Comp Phys., Ac ce p 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 14, 148-158 (1974) 36 Y Mu, P H Nguyen, and G Stock Energy Landscape of a Small Peptide Revealed by Dihedral Angle Principal Component Analysis Proteins, 58, 45-52 (2005) 37 P M Truong, M H Viet, P H Nguyen, C.-K Hu, and M S Li Effect of Taiwan mutation (D7H) on structure of amyloid-β peptides: Replica exchange molecular dynamic study J Phys Chem B, 118, 89728981 (2014) Page 17 of 22 Figure (a) 287.19 Hydrophilic interaction (nm ) 281.66 280.79 285.49 280.99 277.65 259.20 263 265 264.54 269 271 260.20 267 273 an us cr 259.18 264.85 ip t 267.47 275 277 279 281 283 285 M Temperature (K) (b) d 263 Ac ce p g w-w(r) te 271 K 281 K 303 K 0.5 1.5 r (nm) 2.5 3.5 Page 18 of 22 Figure (a) 5.5 4.5 3.5 ip t g w-w(r) cr 2.5 us 1.5 an 0.5 3.0 3.5 2.5 3.0 3.5 19 of 22 Page d Ac ce p 1.25 g w-p(r) 2.5 te (b) 0.75 r (nm) 1.5 1.5 M 0.5 0.5 0.25 0.5 1.5 r (nm) MSD (nm2 ) 400 (a) -0.9e+05 (b) -0.95e+05 E tot (kJ/mol) 500 Ac ce p te d M an us cr ip t Figure 300 200 -1.05e+05 100 -1.0e+05 20 40 60 Time (ns) 80 100 -1.1e+05 20 40 60 Time (ns) Page of 22 80 20100 Ac ce p te d M an us cr ip t Figure (b) (a) Average RMSD (nm) RMSD (nm) 1.5 0.5 Temperature (K) 20 40 60 Time (ns) 80 100 Page 21 of 22 1.4 1.2 1.4 1.2 1.4 1.2 0.8 0.8 0.8 0.6 0.4 0.6 0.4 0.6 0.4 0.2 1.6 263 K 1.65 1.7 1.75 1.8 1.85 1.9 0.2 1.6 0.2 1.6 265 K 1.65 1.7 1.85 1.9 1.75 1.8 0.8 0.6 0.4 0.6 0.4 0.2 1.6 269 K 1.65 1.7 1.75 1.8 1.85 1.9 0.2 1.6 ip t cr 0.8 1.85 1.9 1.4 1.2 us 1.4 1.2 2.5 0.8 1.5 0.6 0.4 an 1.4 1.2 1.75 1.8 271 K 1.65 1.7 1.75 1.8 1.85 1.9 0.2 1.6 273 K 1.65 1.7 0.5 1.75 1.8 1.85 1.9 M RMSD (nm) Rg (nm) 267 K 1.65 1.7 Kcal/mol RMSD (nm) Figure Free energy landscape of Antarctic sea ice bacterium Colwellia sp RMSD (nm) te 1.4 1.2 1.4 1.2 1.4 1.2 0.8 0.8 0.8 0.6 0.4 0.6 0.4 0.6 0.4 0.2 1.6 Ac ce p RMSD (nm) d Rg (nm) 275 K 1.65 1.7 1.75 1.8 1.85 1.9 0.2 1.6 277 K 1.65 1.7 1.75 1.8 1.85 1.9 0.2 1.6 1.4 1.2 1.4 1.2 0.8 0.8 0.8 0.6 0.4 0.6 0.4 0.6 0.4 281 K 1.65 1.7 1.75 1.8 1.85 1.9 0.2 1.6 1.65 1.7 1.75 1.8 1.85 1.9 Rg (nm) 1.4 1.2 0.2 1.6 279 K 283 K 1.65 1.7 1.75 1.8 Rg (nm) 1.85 1.9 0.2 1.6 285 K 1.65 1.7 1.75 1.8 1.85 1.9 Page 22 of 22 ... Page of 223.5 *The Manuscript Coarse grained simulation reveals antifreeze properties of hyperactive Hung Nguyen,1 Thanh Dac Van,1,2 and Ly Le1,2,* ip t antifreeze protein from Antarctic bacterium. .. We used coarse grained simulation to study the mechanism of Colwellia antifreeze protein (ColAFP) (2) We found the distribution and conformation of ice crystal network surrounding ColAFP at low... to the properties derived from MD simulation In this research, the parameterization of polarizable coarse grained water model was used in combination with the coarse grained MARTINI force field