MOLECULAR DYNAMICS SIMULATION FOR SEVEN STRUCTURES OF VISCOTOXIN
ALINA BUŢUa, STELIANA RODINOb,c, MARIANA FERDESa, MARIAN BUŢU b*
aUniversity POLITEHNICA of Bucharest, Splaiul Independentei 313, Sector 6, Bucharest, Romania
bNational Institute of Research and Development for Biological Sciences, 0630031, Splaiul Independentei 296, Bucharest, Romania
cUniversity of Agronomic Sciences and Veterinary Medicine, Mărăşti Blvd. 59, 011464, Bucharest, Romania
It had been proved by scientific research that in the case of a pathogen attack the antimicrobial peptides are key players of the innate immune system. The antimicrobial peptides have been found in all organisms from plants to humans, and even in the microorganisms. The viscotoxins belong to thionine class of antimicrobial peptides and are produced by leaves and stems of Viscum album. In this paper was analyzed the dynamics stability in molecular simulation experiments for seven viscotoxin structures. The conformational structure of the peptide sequence is linked to biological activity and dynamic parameters analysis led to the identification of amino acid residues which show the most important flexibility.
(Received November 26, 2012; Accepted December 18, 2012)
Keywords: Plant antimicrobial peptide, Viscotoxin, “in silico” methods, Molecular dynamics, simulation
1. Introduction
The antimicrobial peptides represent a topic of research increasingly addressed because of the biotechnological potential they possess. The research is targeted both in the direction of identification and characterization of antimicrobial peptides and achieving practical application of these peptides. For selective identification of antimicrobial peptides interesting from therapeutic point of view, various strategies have been applied, with varying results. AMPs are generally defined as sequences of less than 100 amino acid residues with a molecular weight less than 10,000 Da, with a total positive electrical charge (usually between +2 and +9), which present as a particularity the presence of multiple lysine and arginine residues and a substantial part (30% or more) of hydrophobic residues. [1].
Antimicrobial peptides have diverse structures and functions and interact with cell membranes of invading cells by disrupting the membrane integrity. This action leads to cell lysis and, later, to their death [2]. Microbes are the cause of many infectious diseases. The increasing microbial resistance to common antibiotics has become a serious threat in maintaining human health and extensive research is conducted in order to find practical solutions to this issue. Due to their characteristics, antimicrobial peptides have become attractive and safe subjects for researchers who are intensively searching for solutions regarding the resistance to antibiotics [3-6].
The knowledge of the way how a peptide sets its conformation represents the first step in determining the mechanisms underlying its activity (eg antimicrobial activity) and in designing of rational treatments. Another important direction of research is the study of antimicrobial peptides
*Corresponding author: [email protected]
as pote disease peptide resourc reducin are pr describ identif the vis analyz viscoto are dif and for
aminoa structu
(figure beta-st residue strengt
A1 3C8P A2 1JMN A3 1OKH A3 1ED0 B 1JMP
B2 2V9B C1 1ORL
ential bioma es [7-10].
The therm es are issues ces allows u ng of the cos The viscot roduced by bed in the pr fy the influen
scotoxin on zed 7 native oxin sequenc fferent amino r the position
Tabl
A detailed acids sequen ure) is shown
Fig. 1. Se
The 7 stru e 2. a). One tand, and thi es CYS 3 - thens the pep
1 2 3 4 5 6 7
P K SCC P S T N K SCC P N T H K SCC P N T 0 K SCC P N T P K SCC P N T B K SCC P N T L K SCC P N T
arkers of sev odynamic an s that requir us to obtain
sts and of the toxins are an
leaves and resent study nce correspon
the “in silic structures o ces are show oacids corres
ns 6.
le 1. Sequence
d view of t nce (primary n in figure 1.
equence detail
uctures have of the sciss s is making i
CYS 40, CY ptide structur
7 8 9 10 11 12 13
T T G R N I Y T T G R N I Y T T G R N I Y T T G R N I Y T T G R N I Y T T G R D I Y T T G R N I Y
eral diseases nd interaction
e a big volu new data on e time needed ntimicrobial p stems of V consist from nding to the co” molecul of viscotoxin wn in table 1 a sponding to t
e alignment of
the structure y structure)
ls of Viscotoxi
a very simi ors blades is it to be a sol YS 4 - CYS re and gives
14 1516 17 18 19
N T C R L T N T C R F G N A C R L T N A C R L T N T C R L G N T C R L G N T C R F A
s such as var n parameters ume of resea n molecular
d for researc peptides from Viscum album m 46 residues
modification lar dynamics n. The differ and are label the positions
f the primary
e of viscoto and the alp
in A1 from Vis
ilar spatial c s formed by lid structure.
S 32, CYS 1 a compact fo
20 21 22 23 24 25
G S S R E T G G S R Q V G A P R P T G A P R P T G G S R E R G G S R E R G G S R E R
rious forms s, the structu arch. Exploit dynamics st ch [11, 12].
m plants belo m. The visc s of aminoac n of a residu s and on its rences from led in green s 18, 19, 12,
structure of v
oxin 3C8P pha-helix, be
scum album L
conformation two alpha-h The disulfid 6 – CYS 26 orm to the m
2627 28 29 30 31
C A K L S G C A S L S G C A K L S G C A K L S G C A S L S G C A S L S G C A K L S G
of cancer, A re and stabil ing the adva tudy of pept onging to the cotoxins from cids. In this p ue from the p
s structural s primary str colour. To b 24, 25, 28 an
iscotoxin pept
sequence, i eta stand an
L., 3C8P from
n presenting helix and the de bridges are 6 (figure 2).
molecular surf
3233 34 35 36 37
C K I I S A C K I I S A C K I I S G C K I I S G C K I I S A C K I I S A C K I I S A
AIDS, inflam lity of antimi ances in com tides, involv e thionine cl m the exper paper it is ai primary sequ stability. Th ructure of al be noticed th
nd 37 and fo
tides.
including bo nd turns (sec
PDB [13]
a “scissors”
e other one re disposed b This arran face. (figure
38 3940 41 42 43
S T C P S N S T C P S D S T C P S D S T C P S D S T C P S D S T C P S D S T C P S D
mmatory icrobial mputing ving the ass and riments imed to ence of here are ll these at there or 3C8P
oth the condary
” shape by two between gement e 2.c,d)
44 45 46
Y P K Y P K Y P K Y P K Y P K Y P K Y P K
1JMP ensemb solvate starting proper molecu atom Extend 5000 s The nu simula
constra equilib and an visuali progra 3.2 G
a) Fig. 2. Ne
each resi wi
2. Mater The viscot (B), 2V9B (B
The viscot ble and per ed in water.
g from the rties: RMS,
ular dynamic force field ded) method steps conjuga umber of ato ated systems Table 2. Me
ID 1ED0 1JMN 1JMP 1OKH 1ORL 2V9B 3C9P Dynamics aining the lin bration and w nother 50ps i
ization for p am [22, 23].
The simula GHz. The su
ewcartoon rep due and CPK ithout and d) w
rials and m toxin structu B2), 1ORL ( toxins had b riodic bound
It was prod structures ta accessibility cs simulation Amber99sb . The minim ate gradient m
ms of viscot are presente ethod of struct
Method dete
0 SOLU
N SOLU
P SOLU
H X
DIFF
L SOLU
B X
DIFF
P X
DIFF of heating fr nks containi was perform n NPT ensem primary veri ation was pe upport for p
b) presentation of K representatio with molecula
methods ures used we
(C1) were tak een analyzed dary conditio duced a traje
aken from P y surface ar n it was use [20]. Wate mization was
method. The oxins from s d in Table 2.
ture determina
d of structur ermination UTION NMR UTION NMR UTION NMR
X-RAY FRACTION UTION NMR
X-RAY FRACTION
X-RAY FRACTION
rom 0 to 310 ing hydrogen med in 50ps, mble. Dynam ification of
rformed in p parallel sim
of all seven vis on of the disulf ar surfaces SU
ere 3C8P (A ken from PD d using mole ons. For the ectory with a PDB. There
rea, dihedra ed the GROM
er was mod realized usin water box h simulation ex
.
ation and num re peptid R 667 R 656 R 661 667 R 676 659 672
0K was achie n with LINC in NVE en mics of produ
the systems parallel on an mulation was
c) scotoxin aligne lfide bridges: a URF/wirefram
A1), 1JMN ( DB [14-18].
ecular dynam e simulation a length of 1 were analyz ls, distances MACS packa deled with
ng 5000 step had a truncate xperiments a
mber of atoms f
de water 4704 4719 4275 4518 5172 4542 4698
eved by resca CS [21]. The
semble, with uction has be s in dynami
n HP comput s LAM/MP
ed by the mas a)-b) different
e representati
A2), 1OKH mics simulati n, the peptid 100ns for ea zed the struc s between C
age, version SPC-E (Sim ps steepest de ed dodecahed nd total num
for the simula
ions 6 5 5 6 6 4 6
aling the vel e step used w
h periodic bo een achieved cs simulatio ter with dual Program (l
d) ss center of
t views; c) ion
H (A3), 1ED0 ion method i de sequence ach of the pe
ctural and d Cα atoms. F n 4.5.3 [19],
mple Point descent metho
dron form.
mber of atom
ated systems.
total 5377 5380 4941 5191 5854 5205 5376
locity for 200 was of 2fs d
oundary con d for 100ns.
on was used l Xeon quad large–scale
0 (A3), in NPT es were eptides,
ynamic For the
and all Charge od, and s of the
0ps and ynamic nditions For the d VMD
core at atomic
molecular massively parallel) [24, 25]. Simulation analysis was performed with GROMACS and VMD programs.
3. Results and discussion
The trajectories from molecular dynamics simulation experiments for seven native structures of viscotoxin were analyzed comparatively.
Analyzing the dynamics evolution of RMSD for all protein atoms of viscotoxin can be observed that the peptides 1JMN and 1JMP show some variations of approx. 0.4 nm being stable along the entire trajectory, and 1ORL presents a variation of almost 0.2 nm on intervals of 5-10 ns. The rest of the variations are around 0.1 nm, and the other 4 structures show variations below the value of 0.1 nm along the entire dynamics recorded.(figure 3).
Fig. 3. RMSD graphic for all protein atoms
Depending on the comparative evolution of the RMSD of the seven structures, the residues have been classified in four categories as follows:
- very stable residues in all 7 structures – residues 9, 16, 20, 26, 31 present a variation of the RMSD below 0.03 nm;
- stable residues in all 7 structures - residues 2, 5, 7, 8, 12-14, 18, 21, 22, 27, 30, 32, 36-38, 40-42 show a variation of the RMSD between 0.03 nm and 0.1 nm;
- stable residues in some of the structures and unstable in other structures - residues 3, 4, 6, 10, 11, 15, 17, 18, 19, 24, 44 (figure 4);
- slightly unstable residues in some structures and unstable in others - residues 23, 25, 28, 29, 33, 35, 43 (figure 5);
- unstable residues in all 7 structures - residues 34, 39 (figure 6).
Further on, the discussion involves the residues that differ in primary structures of the seven viscotoxins. SER 6 from the structure 3C8P structure shows a high stability, with a variation of the RMSD below 0.02nm, with an evolution different from ASN6 from the 6 other structures that have variations between 0.1 nm and 0.15 nm. The residue ASP11 from the structure 2V9B had a similar behavior with ASN11 from structures 3C8P, 1ED0, 1JMP, 1OKH, with RMSD value of 0.7 nm. ASN11 residues from the structures 1JMN and 1ORL were unstable, with RMSD variations of 0.11nm. PHE18 from the structures 1JMN and 1ORL do not behave differently from LEU18 from the other five structures. Both residues LEU18 and PHE18 showed a high instability and had the value of RMSD variation between 0.1 nm and 0.17 nm. On the position 19 in three of the structures was THR (3C8P, 1OKH and 1ED0), and in the other three structures was GLY (1JMN, 1JMP and 2V9B) and in the structure 1ORL can be found ALA. GLY19 and ALA19 are stable residues with RMSD variations of 0.3 nm, respectively 0.4 for ALA19. THR19 exhibits instability in all structures and presents a variation of RMSD of 0.13 nm. Residue 21 had a high stability in all 7 structures, residue SER21 from the structure 3C8P had a RMSD variation value of 0.07 nm, variation of RMSD of ALA21 from the structures 1OKH and 1ED0 is 0.03 nm, and
GKY21 showed the lowest RMSD variation, namely 0.01 nm. The residue PRO24 from the structures 1OKH and 1ED0 was stable and has a variation of the RMSD value of 0.035 nm. The residues GLN24 from the structures 1JMN and GLU24 from the other four structures behave similarly and showed a variation of RMSD between 0.1nm and 0.16 nm.
re
s 3C8P (A1) 1JMN (A2) 1OKH (A3) 1ED0 (A3) 1JMP (B) 2V9B (B2) 1ORL (C1) 3
4
6
11
10
15
17
18
19
24
44
Fig. 4. The dynamics evolution of RMSD for the residues stable in some structures and unstable in other structures
The residue THR25 from the structures 3C8P, 1OKH and 1ED0 was stable with a value of RMSD below 0.05 nm. The residues VAL25 and ARG25 were unstable, but ARG25 showed higher instability during the entire dynamics recorded with a RMSD value of 0.2 nm, while VAL25 oscillated between two stable states, and had a variation value of 0.12nm. The residue SER28 from 1JMN, 1JMP and 2V9B was a stable residue, showing an identical behavior in all three structures and had a RMSD value of 0.07 nm. LYS28 from the other four structures was
unstable with variations between 0.1 nm and 0.16 nm. The residues GLU37 and ALA37 were stable in all seven structures and have RMSD values of 0.3nm.
re
s 3C8P (A1) 1JMN (A2) 1OKH (A3) 1ED0 (A3) 1JMP (B) 2V9B (B2) 1ORL (C1) 23
25
28
29
33
35
43
Fig. 5. The dynamics evolution of RMSD for the residues slightly unstable in some structures and unstable in others
The residues ILE34 and THR39 from all seven viscotoxin structures showed high instability. ILE34 belongs to the second beta-stand and had RMSD values between 0.14 nm and 0.17 nm. The flexibility of this residue is very important in the movement of the "scissors". This residue is involved in the approach of "scissors blades". THR39 is positioned in the neighborhood of the disulfide bridge CYS3-CYS40, being in the same plane with the residue ILE34, but on the opposite side of the structure (Figure 7). The RMSD value of THR39 for all structures was 0.12 nm. The distance Cα - Cα between the two residues did not present significant variations, which leads to the idea that these two have a coordinated movement.
re
s 3C8P (A1) 1JMN (A2) 1OKH (A3) 1ED0 (A3) 1JMP (B) 2V9B (B2) 1ORL (C1) 34
39
Fig. 6. The dynamics evolution of RMSD for the unstable residues in all 7 structures
Fig. 7. CPK representation of residue 34 and
39 Fig. 8. Distribution of dihedral angles with respect to the secondary structure elements
In Figure 9 are compared the variations in distribution of dihedral angle values reported to secondary structure elements as they are described by Ramachandram plot (Figure 8.). The residues that are not found in Figure 9 do not show variations in dynamics nor any differences between the structures. It can be observed that the most important changes were in beta-sheet type structures: residues 2, 3, 4, from the structure 1JMN and 1JMP, residues 40-44 from the structure 1JMN, residues 39, 40, 43 and 45 from the structure 1JMP. Residues 34, 36, 37 and 38 show instability in all structures, in some cases being clearly defined two areas of distribution - residue 36 from the structures 1OKH, 1EDO and the residue 37 in all structures.
res 3C8P (A1) 1JMN (A2) 1OKH (A3) 1ED0 (A3) 1JMP (B) 2V9B (B2) 1ORL (C1) 2
3
4
34
36
37
38
39
40
42
43
44
45
res 3C8P (A1) 1JMN (A2) 1OKH (A3) 1ED0 (A3) 1JMP (B) 2V9B (B2) 1ORL (C1) Fig. 9. Ramachandran plot for viscotoxins
In Fig. 10 are represented the dynamic structures of the seven viscotoxins on every 10 ns aligned by backbone. It is noted that it started from a compact structure and the area from the beginning of the "scissors" remained compact along the entire dynamics, for all structures. The structure is changed in the “scissors" peak and expanded its volume. The structures 1JMP and 1JMP showed the most visible changes and highest volume.
0 ns 10 ns 20 ns 30 ns
40 ns 50 ns 60ns 70 ns
1ED0
1JMN
1JMP
1OKH
1ORL
2V9B
3C8P
80 ns 90 ns 100 ns
Fig. 10. Frames from molecular dynamics simulation trajectories of viscotoxins
4. Conclusions
From the analysis of the molecular dynamics simulation of the seven structures of vicotoxins it was concluded that the sequences are preserving the key elements of secondary structure, but the structure volume increases. The largest variations of dynamic parameters appeared in the structures 1JMN and 1JMP. Regarding the differences linked to the amino acids recorded in the primary structure, they do not significantly influence the molecular dynamics.
Acknowledgement
This work has been funded by the Sectorial Operational Programme Human Resources Development 2007-2013 of the Romanian Ministry of Labour, Family and Social Protection through the Financial Agreement POSDRU/89/1.5/S/52432.
References
[1] A. Giuliani, G. Pirri, S. F. Nicoletto, Central European Journal of Biology 2(1), 1 (2007) [2] C. Aisenbrey, P. Bertani, P. Henklein, B. Bechinger, Eur Biophys J., 36(4-5), 451 ( 2007) [3] H. Duclohier, Curr Pharm Des., 16(28), 3212 (2010)
[4] J.G. Routsias et all, Peptides., 31(9), 1654 (2010)
[5] S.E. Blondelle, K. Lohner, Curr Pharm Des., 16(28), 3204 (2010)
[6] B. Findlay, G.G. Zhanel, F. Schweizer, Antimicrob Agents Chemother., 54(10), 4049 (2010) [7] M.R. Craddock, J.T. Huang, E. Jackson, N. Harris, E.F. Torrey, M. Herberth, S. Bahn, Molecular & Cellular Proteomics, 7, 1204-1213 (2008)
[8] Y. Mohri et all, Br J Cancer., 101(2), 295-302 (2009)
[9] M. Pascariu, A. Nevoie Anghelache, D. Constantinescu, D. Jitaru, E. Carasevici, T. Luchian, Digest Journal of Nanomaterials and Biostructures 7(1), 79 (2011)
[10] Patrick A. M. Jansen, Diana Rodijk-Olthuis, Edward J. Hollox, Marijke Kamsteeg, Geuranne S. Tjabringa, Gys J. de Jongh, Ivonne M. J. J. van Vlijmen-Willems, Judith G. M. Bergboer, Michelle M. van Rossum, Elke M. G. J. de Jong, Martin den Heijer, Andrea W. M. Evers, Mieke Bergers, John A. L. Armour, Patrick L. J. M. Zeeuwen, Joost Schalkwijk, β-Defensin- 2 Protein Is a Serum Biomarker for Disease Activity in Psoriasis and eaches Biologically Relevant Concentrations in Lesional Skin, PLoS ONE, 4(3), e4725 (2009)
[11] A. Stavrakoudis, I.G. Tsoulos, Z.O. Shenkarev, T.V. Ovchinnikova, Molecular dynamics simulation of antimicrobial peptide arenicin-2: β-Hairpin stabilization by noncovalent interactions, Peptide Science, 92( 3), 143–155 (2009)
[12] A.M. Namba, M.R. Lourenzoni, L. Degreve, Molecular dynamics study of the differences in the human defensin behaviornear a modelled water/membrane interface, Journal of the Brazilian Chemical Society, 18(3), 611-621/ (2007)
[13] H. M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T. N. Bhat, H. Weissig, I. N. Shindyalov, P. E. Bourne, The Protein Data Bank, Nucleic Acids Research, 28, 235 (2000)
[14] A. Pal, J.E. Debreczeni, M. Sevvana, T. Gruene, B. Kahle, A. Zeeck, G.M. Sheldrick, Structures of viscotoxins A1 and B2 from European mistletoe solved using native data alone, Acta Crystallogr., Sect.D64, 985-992 (2008)
[15] J.E. Debreczeni, B. Girmann, A. Zeeck, R. Kratzner, G.M. Sheldrick, Structure of viscotoxin A3: disulfide location from weak SAD data, Acta Crystallogr., Sect.D59: 2125 (2003)
[16] S. Romagnoli, R. Ugolini, F. Fogolari, G. Schaller, K. Urech, M. Giannattasio, L. Ragona, H.
Molinari, NMR structural determination of viscotoxin A3 from Viscum album L., Biochem.
J. 350, 569-577 (2000)
[17] A. Coulon, A. Mosbah, A. Lopez, A.M. Sautereau, G. Schaller, K. Urech, P. Rouge, H.
Darbon, Comparative membrane interaction study of viscotoxins A3, A2 and B from mistletoe (Viscum album) and connections with their structures, Biochem. J.
374, 71-78 (2003)
[18] S. Romagnoli, F. Fogolari, M. Catalano, L. Zetta, G. Schaller, K. Urech, M. Giannattasio, L. Ragona, H. Molinari, NMR solution structure of viscotoxin C1 from Viscum album species Coloratum ohwi: toward a structure-function analysis of viscotoxins, Biochemistry 42: 12503-12510 (2003)
[19] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A. E. Mark, H. J. Berendsen.
GROMACS: fast, flexible, and free, J Comput Chem 26 (16), 1701 (2005)
[20] V. Hornak, R. Abel, A. Okur, B. Strockbine, A. Roitberg, C. Simmerling, Comparison of multiple AMBER force fields and development of improved protein backbone parameters, Proteins 65, 712-725 (2006)
[21] B. Hess, H. Bekker, H. J. C. Berendsen, J. G. E. M. Fraaije, LINCS: A linear constraint solver for molecular simulations. J. Comp. Chem. 18,1463 (1997)
[22] J. Hsin, A. Arkhipov, Y. Yin, J. E. Stone, K. Schulten, Using VMD: an introductory tutorial, Current Protocol in Bioinformatics, 5, 5.7 (2008)
[23] W. Humphrey, A. Dalke, K. Schulten, VMD - Visual Molecular Dynamics, J. Molec.
Graphics, 14, 33-38 (1996)
[24] S. J. Plimpton, B. A. Hendrickson, A New Parallel Method for Molecular–Dynamics Simulation of Macromolecular Systems, J. Comput. Chem., 17(3), 326–337 (1996)
[25] J. M. Squyres, A. Lumsdaine, A Component Architecture for LAM/MPI, Proceedings, 10th European PVM/MPI Users' Group Meeting, Venice, Italy, Springer-Verlag, series Lecture Notes in Computer Science, 2840, 379-387 (2003)
