Papers in Physics, vol. 12, art. 120004 (2020)

Received: 13 August 2019, Accepted: 20 March 2020
Edited by: M. C. Barbosa
Licence: Creative Commons Attribution 4.0
DOI: https://doi.org/10.4279/PIP.120004

www.papersinphysics.org

ISSN 1852-4249

Flavodoxin in a binary surfactant system consisting of the nonionic
1-decanoyl-rac-glycerol and the zwitterionic lauryldimethylamine-N-oxide:

Molecular dynamics simulation approach

Behnaz Bazaziyan1, Mohammad Reza Bozorgmehr1*, Mohammad Momen-Heravi1,
S. Ali Beyramabadi1

Due to the short time constant of the spin-spin relaxation process, there is a limitation
in the preparation of NMR sample solution for large proteins. To overcome this prob-
lem, reverse micelle systems are used. Here, molecular dynamics simulation was used
to study the structure of flavodoxin in a quaternary mixture of 1-decanoyl-rac-glycerol,
lauryldimethylamine-N-oxide, pentane and hexanol. Hexanol was used as co-solvent. Sim-
ulations were performed at three different co-solvent concentrations. The proportion of
components in the mixture was selected according to experimental conditions. For com-
parison, simulation of flavodoxin in water was also performed. The simulation results show
that the Cα-RMSD for the protein in water is less than for the surfactant mixture. Also,
the radius of gyration of flavodoxin increased in the presence of surfactants. The distance
between the two residues trp-57 and phe-94, as a measure of protein activity, was obtained
from the simulations. The results showed that in the surfactant mixtures this distance
increases. Analysis of the secondary structure of the protein shows that the N-terminal
part of the flavodoxin is more affected by surfactants. The flavodoxin diffusion coefficient
in the surfactant mixture decreased in relation to its diffusion coefficient in water.

I. Introduction

Nuclear magnetic resonance (NMR) spectroscopy
is a powerful technique for studying the structure
of proteins under different conditions [1]. This
method is used for high-purity protein samples in
an aqueous phase. In the case of spin-spin relax-
ation, the transverse component of the magnetiza-
tion vector decays exponentially towards its equi-
librium value in NMR. The time constant describ-
ing this process is known as T2. This name is based
on T1, which is the time constant of the spin-lattice

* bozorgmehr@mshdiau.ac.ir

1 Department of Chemistry, Mashhad Branch, Islamic
Azad University, Mashhad, Iran.

relaxation process. The short T2 time causes pulse
loss during pulse sequences. On the other hand,
with increasing protein size, this time is reduced. It
is therefore difficult to prepare a solution for large
proteins, so using this method for them is limited.
By encapsulating the protein in the aqueous solu-
tion formed in the cavity of the reverse micelles,
and then placing the complex in a non-polar sol-
vent, this problem is largely resolved [2]. In this
case, the viscosity of the non-polar solvent must be
much less than water [3]. Short chain alkanes are a
good candidate for this purpose [4]. The nature of
surfactants, the ability to dissolve surfactant in a
non-polar solvent, and the ratio of water to surfac-
tant are among the important parameters for the
success of this method [5].

120004-1



Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

Flavodoxin is an electron transfer protein that
plays a role in photosynthetic reactions [6]. In
flavodoxins, a Flavin mono nucleotide molecule is
attached to the protein in a non-covalent man-
ner, which causes the redox property of proteins
[7]. From a structural standpoint, flavdoxin is an
α/β protein in which the β sheets are surrounded
by α helices [8]. Since flavodoxin is a simple one-
domain protein, it is a suitable model for protein
studies such as stability and folding. Cheung et al.
have investigated the dynamics of flavodoxin pro-
tein in crowded environments by molecular dynam-
ics simulation [9]. They showed that protein stabil-
ity increases in in silico crowding conditions, and
folding routes experiencing topological frustrations
can be either relieved or enhanced upon manipula-
tion of the crowding conditions. Studies have also
shown that by removing the Flavin mononucleotide
molecule from flavodoxin and creating the Apopro-
tein form, the binding position does not change
greatly; only the aromatic rings of tyrosine-94 and
tryptophan-57 are closer to each other [6]. Gen-
zor et al. studied the stability of apo-flavodoxin
conformation in urea solution in different acidic
and ionic strengths. They found that at a pH of
between 6 and 11 apo-flavodoxin conformation is
stable, whereas at a lower pH, between 2 and 3.5,
proteins lose their stability [10]. The structure of
flavodoxin in the presence of a cross-linked, spheri-
cal, sucrose-based polymer called Ficoll 70 has been
studied experimentally and theoretically [11, 12].
This circular dichroism and in silico study showed
that the helical content of folded apoflavodoxin was
increased by additions of Ficoll 70. Other experi-
mental evidence indicated that correct contact for-
mation around the third β-strand of flavodoxin in
the central sheet is crucial to continued folding
to the native state in crowded media. Research
has also shown that flavodoxin in the presence of
Dextran retains about 70% of its secondary struc-
ture [13].

In this manuscript, a simulation of molecular dy-
namics was used to study the flavdoxin protein in
a mixture of the surfactants 1-decanoyl-rac-glycerol
and lauryldimethylamine-N-oxide. Pentane solvent
was used for solute solvation. For comparison,
molecular dynamics simulation of protein in water
was also performed.

II. Computational details

For the molecular dynamics simulation, four sim-
ulation boxes were designed. In the first box,
the box dimensions were considered according to
the size of the flavodoxin, 5.95 × 5.95 × 5.95 nm3,
and the protein was placed in the center of the
box. Then the box was filled with water solvent
(named as System A). In the second box, the pro-
tein was placed in the center of the box, with
dimensions 6.75 × 6.75 × 6.75 nm3, then 130 1-
decanoyl-rac-glycerol molecules and 70 molecules of
lauryldimethylamine-N-oxide were randomly added
to the box. Also, 50 molecules of water were added
to this box, and finally, the box was filled with 1000
pentane molecules (named as System B). The spec-
ification of the third box was like the second box,
only differing in that 5 hexanol molecules were also
randomly placed in the box (named as System C).
The fourth box was similar to the second box, but
in this box, 10 hexanol molecules were randomly
placed in it (named as System D). The components
used in the various systems are summarized in Ta-
ble 1.

Since 1-decanoyl-rac-glycerol is non-ionic, hex-
anol solvent is used as auxiliary solvent. These
molecular ratios are selected based on the re-
ported optimal ratios in Dodevski et al.’s experi-
mental work [14]. The structure of flavodoxin was
obtained from the protein data bank with code
1obv [15]. To perform molecular dynamics sim-
ulation, the Gromacs software version 5.2.1 was
used [16]. Charm27 all-atom force field was used
for calculations, since the force field parameters
for 1-decanoyl-rac-glycerol, lauryldimethylamine-
N-oxide, hexanol and pentane are not present in
the default version of the Gromacs software. The
structure of these compounds was optimized using
the B3LYP density functional method with the ba-
sis function of 6-31G*. To control the optimization,
frequency calculations were performed and virtual
frequencies were not observed. The charge of each
atom was calculated by the CHELPG method. All
ab initio calculations were done by GAMESS soft-
ware package [17]. The optimized structure of the
compounds is shown in Fig. 1.
The SwissParam web server [18] was used to de-
termine the force field parameters of the studied
compounds. Water model TIP3P [19] was used for
water in simulation boxes. In order to neutralize

120004-2



Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

Table 1: Outline of designed systems.

System box volume
(
nm3

)
1-decanoyl-rac-glycerol lauryldimethylamine-N-oxide pentane hexanol

A 5.95 × 5.95 × 5.95 - - - -
B 6.75 × 6.75 × 6.75 130 70 1000 -
C 6.75 × 6.75 × 6.75 130 70 1000 5
D 6.75 × 6.75 × 6.75 130 70 1000 10

Figure 1: Optimized geometries of studied molecules,
with labeling.

the systems, an appropriate number of Na+ ions
were added to each box. To fix a constant tem-
perature and pressure during the simulations, sys-
tem components were coupled with V-rescale and
Nose-Hoover thermostat [20] respectively, in each
of the equilibration steps and molecular dynamics
simulations. The temperature and pressure used
in the calculations are 298 Kelvin and 1 bar, re-

spectively. LINCS algorithm [21] was employed to
fix the chemical bonds between the atoms of the
non-water molecules and SETTLE algorithm [22]
was used in the case of water molecules. The PME
algorithm was used to calculate electrostatic inter-
actions [23]. Energy minimization was done using
the steepest descent method [24] to eliminate the
primary kinetic energy and to eliminate inappropri-
ate contacts between the atoms. Each simulation
box achieved a two-stage equilibrium in NVT and
NPT ensemble. At this stage, the time of equili-
bration was considered 10 ns with a time step of
2fs. Finally, molecular dynamics was performed by
solving the second Newton equation for 140 ns with
the 2fs time step. Each simulation was repeated
four times under the different initial conditions, to
avoid any dependency on the initial conditions and
to increase the accuracy of the simulations.

III. Results and discussion

An important point when using surfactants in en-
capsulating proteins in NMR spectroscopy is to
maintain the native protein structure and reduce
its reorientation time. Therefore, the structural
and dynamic parameters of flavodoxin in the de-
signed systems were calculated for analysis. One
common method of displaying the stability of a
protein’s structure is to calculate the RMSD val-
ues during the simulation. The root mean square
deviation (RMSD) is defined as:

RMSD(t1, t2) =

[
1

M

N∑
i=1

mi|ri(t1) −ri(t2)|

]1
2

.

(1)

Where ri is the atomic position at time t and M =∑N
i=1 mi. Cα-RMSD variations of flavodoxin for

the studied systems are shown in Fig. 2.

120004-3



Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

0 20000 40000 60000 80000 100000 120000 140000

 time [ps]

0.25

 0.2

0.15

 0.1

0.05

 0

R
M

S
D

 [
n
m

]

Figure 2: Cα-RMSD values (nm) versus time (ps) for
simulated systems defined as A, B, C and D.

According to the figure, protein stability in sur-
factant systems is observed to be somewhat reduced
compared to system A. This decrease in stability
for the C system is greater than for B and D. How-
ever, the difference in Cα-RMSD values in different
systems is not high. On the other hand, given the
fact that flavodoxin has 169 residues, this differ-
ence in stability in various systems is not signifi-
cant. Another factor that can determine the ac-
tivity and stability of the protein is compression of
the conformation [25], in which the radius of gyra-
tion (Rg) indicates the compression of the protein
structure. The Rg of a protein is calculated using
the following equation:

Rg =

(∑
i |ri|

2mi∑
i mi

)1
2

(2)

Where mi is the atomic mass of i and ri the atomic
position of i relative to the center of mass of the
molecule [26]. The Rg of the studied systems was
calculated and the results are shown in Fig. 3.

The Rg values of the proteins vary from 1.48 nm
to 1.63 nm in different systems. Flavodoxin in sys-
tem A has the smallest Rg and in system C the
highest Rg. Since the compactness of the struc-
ture of the proteins is attributed to their activity
[27], according to 3 the presence of flavodoxin in the
surfactant solution decreases its activity. The root
mean square fluctuation (RMSF) values of flavo-
doxin sequences are calculated in different systems
and shown in Fig. 4.

Except for the residues valine-18, isoleucine-21,

0 20000 40000 60000 80000 100000 120000 140000

 time [ps]

R
g
 [

n
m

]

1.69

1.64

1.59

1.54

1.49

1.44

Figure 3: Rg values (nm) versus time (ps) for simulated
systems defined as A, B, C and D.

R
M

S
F

 [
n
m

]

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0 20 40 60 80 100 120 140 160

 Residue Number

Figure 4: Average RMSF values for each residue of
Flavodoxin during simulation in A, B, C and D sys-
tems.

tryptophan-66 and glycine-68 in the flavodoxin of
system C, the flexibility of other residues in system
A is more than in the other systems. Another note-
worthy point is that the C-terminal of the protein
is less affected by the non-aqueous environment.
In fact, the greatest difference in the RMSF val-
ues in the various systems related to amino acids
is 1 to 79. In this region, the helices and sheets in
the secondary structure of flavodoxin have smaller
lengths than the helices and sheets in the c-terminal
of the protein. If two faces are considered for pro-
teins, the face of the N-terminal is more exposed
to the environment. In flavodoxins, the distance
between two aromatic amino acids in the ligand
binding site is considered as a measure of protein
activity [8]. The distance between the tryptophan-
57 and phenylalanine-94 residues of flavodoxin in

120004-4



Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

Figure 5: The distance between Trp-57 and Phe-94
(system A).

system A is shown in Fig. 5.

According to the figure, this distance is 7.61
angstroms. The distance between these two amino
acids in systems B, C and D was 9.48, 10.05 and
10.42 angstroms, respectively. According to the
values obtained, flavdoxin activity seems to de-
crease in surfactant solutions. Sampling of the suit-
able structure for analysis in molecular dynamics
trajectories has always been one of the interesting
issues. Averaging the coordinates of the molecular
coordinates in the last few nanoseconds of the sim-
ulation is one of the methods used in this regard
[28]. The sampling of coordinates at a time when
the system is in equilibrium is another method used
for this [29]. In these methods, variations of a
quantity such as Cα-RMSD in the area are used.
However, there is no physical concept in the average
of atomic coordinates [30], so for sampling of the
simulationsthe free energy landscape (FEL) analy-
sis method is used [31]. There are three main stages
of FEL analysis: calculation of the Cα-RMSD and
the flavodoxin radius of gyration (Rg), obtaining
the possibility of the presence of flavodoxin pro-
tein conformation in each corresponding value of
Cα-RMSD and Rg, and calculation of the free en-
ergy of configurations based on the probability val-
ues of presence. The results of the FEL analysis
are shown in 3D diagrams in Fig. 6 for A, B, C
and D systems, respectively.

In these figures, the minimum regions of free en-
ergy are shown in blue. According to these figures,
it can be seen that all systems have a local mini-
mum with the least amount of free energy. Regard-
ing the values of Cα-RMSD and Rg corresponding

Figure 6: Free Energy Landscape of the A, B, C and
D simulated systems.

to the lowest free energy, the structure of flavodoxin
was sampled from the molecular trajectories in the
simulations. Then, using the software polyview 2D
[32], the secondary structure of the flavodoxin ob-
tained in the previous step was analyzed. The re-
sult is shown in Fig. 7 for different systems.

As the color darkens, it indicates that the level
of solvent-accessible surface area for the residue
decreases. With respect to the figures, it is ob-
served that in the B system, the helices between
aspartic-35 and aspartic-46 are eliminated. The
beta sheet between the Valin-31 and Aspartic-35
sequences has also been shortened. The downsiz-
ing of this beta sheet in C and D systems is less
than in the B system. In these areas, by chang-
ing the structure of the helices and sheets, the
level of solvent-accessible surface area of residues
has increased somewhat. The region of change for
the C system is related to the helices between the
tryptophan-57 and the aspartic-75 residues. In gen-
eral, the secondary structure of the flavodoxin in
the c-terminal region has changed little. These re-
sults are consistent with the results of the RMSF
values.

A protein contact map is a convenient way to de-
termine changes in the tertiary structure of proteins
in different conditions. The flavodoxin contact map
was obtained in various simulations and the results
are shown in Fig. 8. In these figures, the flavo-
doxin structure of system A was considered as a
wild structure, and the remaining structures were
compared to system A. In these figures, the lower
half is the contact map of flavodoxin, while the up-
per half is the protein difference contact map; they

120004-5



Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

Helix, β strand, coil.
Dark: completely buried, bright: completely exposed

Figure 7: The secondary structure of the Flavodoxin, along with the solvent-accessible surface area for its residues
in systems A (top-left), B (top-right), C (bottom-left) and D (bottom-right).

Figure 8: (Left) The contact map of Flavodoxin in B (left-lower triangle) and different contact map of Flavodoxin
in B and Flavodoxin in A (right-upper triangle). (Center) The contact map of Flavodoxin in C (left-lower
triangle) and different contact map of Flavodoxin in C and Flavodoxin in A (right-upper triangle). (Right)
The contact map of Flavodoxin in D (left-lower triangle) and different contact map of Flavodoxin in D and
Flavodoxin in A (right-upper triangle). The colors used in the figure are described in the main text.

are shown in different systems relative to system A.
In the lower part of the figure, the black color rep-
resents the common contacts and the green color
indicates the contacts that are in the wild struc-
ture of system A, but not in the structure of the
secondary system; the red color represents the con-
tacts that are in the secondary structure but not
in the structure of system A. In the upper part,

the blue color indicates the regions with the small-
est differences and the red color indicates the areas
with the largest differences. The greater the differ-
ence in red color between Figs. 8 (center) and 8
(right) in comparison with Fig. 8 (left), the more
the tertiary structure of flavodoxin in systems C
and D has been affected. Also, areas of tertiary
structure change in these systems are more focused

120004-6



Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

than in system B. Finally, to determine the dy-
namics of flavodoxin, its diffusion coefficient was
calculated. Prediction of the diffusion coefficients
of molecules in solvent is of theoretical importance.
As the diffusion coefficient value increases, the mo-
bility of the molecule increases [33]. The Einstein
relation was used to calculate the diffusion coeffi-
cient of flavodoxin in all simulated systems:

D =
1

6
lim
t→∞

d

dt
〈|ri(t) −ri(0)|2〉. (3)

Where ri is the atom coordinate vector and the
term inside the angled brackets is the mean square
displacement (MSD). In this approach, the self-
diffusion coefficient is proportional to the slope of
the MSD as a function of time in the diffusional
regime [34]. The diffusion coefficient values of flavo-
doxin are listed in Table 2.

Table 2: Diffusion coefficient values of Flavodoxin

System D × 10−5cm2/s
A 0.086
B 0.004
C 0.004
D 0.016

With regard to the diffusion coefficient, the flavo-
doxin dynamics are slower in systems B, C and D
than in system A. Thus, the time of reorientation
of protein conformation in systems B, C and D is
less than that of system A. In other words, using
this surfactant mixture can lead to a better NMR
spectrum. This result is consistent with the results
in the experimental work of Dodevski et al [14].

IV. Conclusions

By encapsulating the protein in the aqueous solu-
tion formed in the cavity of the reverse micelles,
and then placing the complex in a non-polar sol-
vent, the problem of slow tumbling will be greatly
resolved in NMR spectroscopy. In this work, a mix-
ture of the surfactants 1-decanoyl-rac-glycerol and
lauryldimethylamine-N-oxide were used, together
with the auxiliary solvent hexanol. Changes in
the structure and dynamics of flavodoxin protein
were studied in this mixture. Pentane solvent was

used as a low viscosity solvent. Compared with
the structure of flavodoxin in water, the simula-
tion results indicate that the protein’s secondary
and tertiary structures in the mixture of surfac-
tants are altered. However, its dynamics decrease.
It was observed that in the absence of hexanol the
helical and sheet content of the protein decreased
in different regions. Also, in the presence of hex-
anol the ordered secondary structure of the protein
decreased relative to two component systems con-
sisting of protein and water. However, the level
of reduction of the regular secondary structure was
lower than in the absence of hexanol. Decreasing
the areas of the regular secondary structure made
the sequences more accessible to solvent. In gen-
eral, the secondary structure of the flavodoxin in
the c-terminal region changed little. The different
data obtained from the simulation are consistent
with each other. The results are also in agreement
with previous work in this area [35].

[1] A D Bax, S Grzesiek, Methodological advances
in protein NMR, Acc. Chem. Res. 26, 131
(1993).

[2] A J Wand, M R Ehrhardt, P F Flynn, High-
resolution NMR of encapsulated proteins dis-
solved in low-viscosity fluids, Proc. Natl. Acad.
Sci. U.S.A. 95, 15299 (1998).

[3] M R Ehrhardt, P F Flynn, A J Wand, Prepa-
ration of encapsulated proteins dissolved in low
viscosity fluids, J. Biomol. NMR 14, 75 (1999).

[4] C R Babu, P F Flynn, A J Wand, Validation
of protein structure from preparations of en-
capsulated proteins dissolved in low viscosity
fluids, J. Am. Chem. Soc. 123, 2691 (2001).

[5] W D Van Horn, M E Ogilvie, P F Flynn, Use
of reverse micelles in membrane protein struc-
tural biology, J. Biomol. NMR 40, 203 (2008).

[6] J Sancho, Flavodoxins: Sequence, folding,
binding, function and beyond, Cell. Mol. Life
Sci. 63, 855 (2006).

[7] K Fukuyama, H Matsubara, L J Rogers, Crys-
tal structure of oxidized flavodoxin from a red
alga Chondrus crispus refined at 1.8 Å resolu-
tion: Description of the flavin mononucleotide
binding site, J. Mol. Biol. 225, 775 (1992).

120004-7

http://doi.org/10.1021/ar00028a001
http://doi.org/10.1021/ar00028a001
http://doi.org/10.1073/pnas.95.26.15299
http://doi.org/10.1073/pnas.95.26.15299
http://doi.org/10.1023/A:1008354507250
http://doi.org/10.1021/ja005766d
http://doi.org/10.1007/s10858-008-9227-5
http://doi.org/10.1007/s00018-005-5514-4
http://doi.org/10.1007/s00018-005-5514-4
http://doi.org/10.1016/0022-2836(92)90400-E


Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

[8] S Maldonado, A Lostao, M P Irún, J
Férnandez-Recio, C G Genzor, E B González,
J A Rubio, A Luquita, F Daoudi, J Sancho,
Apoflavodoxin: Structure, stability, and FMN
binding, Biochimie 80, 813 (1998).

[9] D Homouz, L Stagg, P Wittung-Stafshede,
M S Cheung, Macromolecular crowding mod-
ulates folding mechanism of α/β protein
apoflavodoxin, Biophys. J. 96, 671 (2009).

[10] C G Genzor, C Gómez-Moreno, J Sancho,
A Beldarráın, J L López-Lacomba, M Cor-
tijo, Conformational stability of apoflavodoxin,
Protein Sci. 5, 1376 (1996).

[11] L Stagg et al, Molecular crowding enhances
native structure and stability of α/β protein
flavodoxin, Proc. Natl. Acad. Sci. U.S.A. 104,
18976 (2007).

[12] D Venturoli, B Rippe, Ficoll and dextran vs.
globular proteins as probes for testing glomeru-
lar permselectivity: Effects of molecular size,
shape, charge, and deformability, Am. J. Phys-
iol. Renal Physiol. 288, F605 (2005).

[13] E Steensma, C P van Mierlo, Structural char-
acterisation of apoflavodoxin shows that the lo-
cation of the stable nucleus differs among pro-
teins with a flavodoxin-like topology, J. Mol.
Biol. 282, 653 (1998).

[14] I Dodevski, N V Nucci, K G Valentine, G K
Sidhu, E S O’Brien, A Pardi, A J Wand, Op-
timized reverse micelle surfactant system for
high-resolution NMR spectroscopy of encapsu-
lated proteins and nucleic acids dissolved in
low viscosity fluids, J. Am. Chem. Soc. 136,
3465 (2014).

[15] A Lostao, F Daoudi, M P Irún, Á Ramón,
C Fernández-Cabrera, A Romero, J Sancho,
How FMN Binds to Anabaena Apoflavodoxin:
A hydrophobic encounter at an open binding
site, J. Biol. Chem. 278, 24053 (2003).

[16] D Van Der Spoel, E Lindahl, B Hess, G Groen-
hof, A E Mark, H J Berendsen, GROMACS:
Fast, flexible, and free, J. Comput. Chem. 26,
1701 (2005).

[17] M W Schmidt, K K Baldridge, J A Boatz, S T
Elbert, M S Gordon, J H Jensen, S Koseki,
N Matsunaga, K A Nguyen, S Su, General
atomic and molecular electronic structure sys-
tem, J. Comput. Chem. 14, 1347 (1993).

[18] V Zoete, M A Cuendet, A Grosdidier, O
Michielin, SwissParam: A fast force field gen-
eration tool for small organic molecules, J.
Comput. Chem. 32, 2359 (2011).

[19] D J Price, C L Brooks III, A modified TIP3P
water potential for simulation with Ewald sum-
mation, J. Chem. Phys. 121, 10096 (2004).

[20] G Bussi, D Donadio, M Parrinello, Canonical
sampling through velocity rescaling, J. Chem.
Phys. 126, 014101 (2007).

[21] B Hess, H Bekker, H J Berendsen, J G Fraaije,
LINCS: A linear constraint solver for molecu-
lar simulations, J. Comput. Chem. 18, 1463
(1997).

[22] S Miyamoto, P A Kollman, SETTLE: An an-
alytical version of the SHAKE and RATTLE
algorithm for rigid water models, J. Comput.
Chem. 13, 952 (1992).

[23] U Essmann, L Perera, M L Berkowitz, T Dar-
den, H Lee, L G Pedersen, A smooth particle
mesh Ewald method, J. Chem. Phys. 103, 8577
(1995).

[24] D G Luenberger, Y Ye, Linear and nonlinear
programming, Springer International Publish-
ing, Switzerland (1984).

[25] S Ghaderi, M R Bozorgmehr, A Morsali,
Structure study and predict the function of the
diphtheria toxin in different pH levels (Acidic-
Basic-Natural) using molecular dynamics sim-
ulations, Entomol. Appl. Sci. Lett. 3, 49
(2016).

[26] B Honarparvar, A A Skelton, Molecular dy-
namics simulation and conformational analy-
sis of some catalytically active peptides, J. Mol.
Model. 21, 100 (2015).

[27] H Monhemi, M R Housaindokht, M R Bo-
zorgmehr, M S S Googheri, Enzyme is stabi-
lized by a protection layer of ionic liquids in

120004-8

http://doi.org/10.1016/S0300-9084(00)88876-8
http://doi.org/10.1016/j.bpj.2008.10.014
http://doi.org/10.1002/pro.5560050716
https://doi.org/10.1073/pnas.0705127104
https://doi.org/10.1073/pnas.0705127104
https://doi.org/10.1152/ajprenal.00171.2004
https://doi.org/10.1152/ajprenal.00171.2004
https://doi.org/10.1006/jmbi.1998.2045
https://doi.org/10.1006/jmbi.1998.2045
http://doi.org/10.1021/ja410716w
http://doi.org/10.1021/ja410716w
http://doi.org/10.1074/jbc.M301049200
http://doi.org/10.1002/jcc.20291
http://doi.org/10.1002/jcc.20291
http://doi.org/10.1002/jcc.540141112
http://doi.org/10.1002/jcc.21816
http://doi.org/10.1002/jcc.21816
http://doi.org/10.1063/1.1808117
http://dx.doi.org/10.1063/1.2408420
http://dx.doi.org/10.1063/1.2408420
https://doi.org/10.1002/(SICI)1096-987X(199709)18:123C1463::AID-JCC43E3.0.CO;2-H
https://doi.org/10.1002/(SICI)1096-987X(199709)18:123C1463::AID-JCC43E3.0.CO;2-H
http://doi.org/10.1002/jcc.540130805
http://doi.org/10.1002/jcc.540130805
http://doi.org/10.1063/1.470117
http://doi.org/10.1063/1.470117
http://doi.org/10.1007/978-3-319-18842-3
http://doi.org/10.1007/978-3-319-18842-3
http://doi.org/10.1007/978-3-319-18842-3
https://easletters.com/index.php?journal=journal&page=article&op=view&path5B5D=123
https://easletters.com/index.php?journal=journal&page=article&op=view&path5B5D=123
http://doi.org/10.1007/s00894-015-2645-x
http://doi.org/10.1007/s00894-015-2645-x


Papers in Physics, vol. 12, art. 120004 (2020) / B. Bazaziyan et al.

supercritical CO2: Insights from molecular dy-
namic simulation, J. Supercrit. Fluids 69, 1
(2012).

[28] M R Housaindokht, M R Bozorgmehr, H E
Hosseini, R Jalal, A Asoodeh, M Saberi, Z
Haratipour, H Monhemi, Structural proper-
ties of the truncated and wild types of Taka-
amylase: A molecular dynamics simulation
and docking study, J. Mol. Catal. B: Enzym.
95, 36 (2013).

[29] N Moghtaderi, M R Bozorgmehr, A Morsali,
The study of self-aggregation behavior of the
bilirubin molecules in the presence and absence
of carbon nanotubes: Molecular dynamics sim-
ulation approach, J. Mol. Liq. 208, 342 (2015).

[30] D van der Spoel, E Lindahl, B Hess, A Van
Buuren, E Apol, P Meulenhoff, D Tieleman,
A Sijbers, K Feenstra, R van Drunen, GRO-
MACS user manual version 3.3, (2008).

[31] H Lei, C Wu, H Liu, Y Duan, Folding free-
energy landscape of villin headpiece subdomain
from molecular dynamics simulations, Proc.
Natl. Acad. Sci. U.S.A. 104, 4925 (2007).

[32] A A Porollo, R Adamczak, J Meller,
POLYVIEW: A flexible visualization tool for
structural and functional annotations of pro-
teins, Bioinformatics 20, 2460 (2004).

[33] D Brune, S Kim, Predicting protein diffusion
coefficients, Proc. Natl. Acad. Sci. U.S.A. 90,
3835 (1993).

[34] D Frenkel, B Smit, Understanding molecular
simulation: From algorithms to applications,
Computacional Sciences Series 1, Academic
Press (2002).

[35] B Bazaziyan et al, Reverse micelle surfactant
system comprising the 1-decanoyl-rac-glycerol
and the Lauryldimethylamine-N-oxide: Struc-
ture and dynamics of confined water, Russ. J.
Phys. Chem. A 93, 1122 (2019).

120004-9

http://doi.org/10.1016/j.supflu.2012.04.020
http://doi.org/10.1016/j.supflu.2012.04.020
http://doi.org/10.1016/j.molcatb.2013.05.011
http://doi.org/10.1016/j.molcatb.2013.05.011
http://doi.org/10.1016/j.molliq.2015.04.052
https://doi.org/10.1073/pnas.0608432104
https://doi.org/10.1073/pnas.0608432104
http://doi.org/10.1093/bioinformatics/bth248
http://doi.org/10.1073/pnas.90.9.3835
http://doi.org/10.1073/pnas.90.9.3835
https://doi.org/10.1134/S0036024419060050
https://doi.org/10.1134/S0036024419060050

	Introduction
	Computational details
	Results and discussion
	Conclusions