Atomistic-Level Investigation of a LL37 Conjugated ... - ACS Publications

Subscriber access provided by University of South Dakota

B: Biophysics; Physical Chemistry of Biological Systems and Biomolecules

Atomistic-Level Investigation of a LL37 Conjugated Gold Nanoparticle by Well-Tempered Metadynamics André Filipe Ferreira, Michela Comune, Akhilesh Rai, Lino Ferreira, and Pedro Nuno Simões J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.8b05717 • Publication Date (Web): 14 Aug 2018 Downloaded from http://pubs.acs.org on August 15, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 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

The Journal of Physical Chemistry

Atomistic-Level Investigation of a LL37 Conjugated Gold Nanoparticle By Well-Tempered Metadynamics †

André F. Ferreira,

‡

Michela Comune,

‡

Akhilesh Rai,

Simões

Lino Ferreira,

‡

and Pedro N.

∗,†

CIEPQPF, Department of Chemical Engineering, University of Coimbra, 3030-790 Coimbra, Portugal, CNC-Center for Neurosciences and Cell Biology, University of Coimbra, 3004-517, Coimbra, Portugal, and Biocant, Biotechnology Innovation Center, 3060-197, Cantanhede, Portugal

E-mail: [email protected]

Phone: +351 239 798 703. Fax: +351 239 798 703

∗

To whom correspondence should be addressed

†

CIEPQPF, Department of Chemical Engineering, University of Coimbra, 3030-790 Coimbra, Portugal

‡

CNC-Center for Neurosciences and Cell Biology, University of Coimbra, 3004-517, Coimbra, Portugal

¶

Biocant, Biotechnology Innovation Center, 3060-197, Cantanhede, Portugal

1

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 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

Abstract LL37 is a cathelicidin-derived AMP with a broad spectrum of antimicrobial activity and wound healing potential. The enhancement of these characteristics were recently demonstrated for a cysteine (CYS) modied cathelicidin-derived (LL37-SH) conjugated with gold nanoparticles (AuNP). Considering the potential of this peptide, we hereby report a computational study in which well-tempered metadynamics was applied to unveil the interaction of LL37-SH and LL37 with a AuNP with atomistic detail. A structural analysis combined with the free energy surface (FES) characterization allowed to assess the role of CYS residue during the formation of the conjugate, as well as to understand how the AuNP improves the antimicrobial activity of the peptide. It was found that CYS promotes a lower conformational entropy (before and after adsorption onto the AuNP) and a faster adsorption process when compared to the LL37 without CYS. The FES for LL37-SH is characterized by one global minimum, while for LL37 a potential metastable state was found. The presence of the AuNP leads to an elongation of the peptides along with the adsorption, which translates into the increase of the solvent accessible surface area. This elongation, combined with the greater availability of positively charged residues upon adsorption rationalize the observed enhancement of the activity of LL37-SH/AuNP conjugate.

Introduction The dramatic appearance of multi-drug resistant pathogenic organisms during the last decades has demanded new classes of antibiotics to be considered. Antimicrobials peptides (AMPs) emerged as an alternative to conventional therapies, such as antibiotics. With nearly 3000 AMPs registered in the antimicrobial peptide database 1 , their therapeutic potential is tremendous 2 . Besides being very eective against several strains of bacteria, fungi and viruses, AMPs can also modulate host physiological functions such as inammation, angiogenesis, and wound healing 35 .

2


Page 2 of 24

Page 3 of 24 1 2 3 4 5 6 7 8 9 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


AMPs conjugated with multiple composition nanoparticles have been used for various applications. The end conjugate (AMP/AuNP) retains, or even improves, the peptide activity, while beneting of AuNPs targeting, thermal and bioimaging characteristics. During the past decade, several studies emerged reporting the synthesis, characterization and activity of these materials 59 . Rai

. 5 and Comune

et al

. 9 have recently reported two distinct

et al

AMPs that shown an improvement of their bioactivity after being immobilized onto a AuNP. Nonetheless, several questions are still unanswered concerning the mechanism that governs the adsorption process of AMPs onto a AuNP. In this context, biomolecular computer simulations became a crucial tool in biochemistry and molecular biology research 10 . Molecular dynamics simulations can capture the behavior of biological macromolecules in full atomic detail. This technique enables the replication of key biochemical processes such as protein folding, drug binding, membrane transport, and the other conformational changes critical to protein function

in silico

10

. Several studies addressing adsorption process of peptides onto

gold surfaces have been reported 1114 . However, the geometry of the surface is at in most of the cases, along with the target AMPs being relatively small in size (less than 20 residues). LL37 is the only cathelicidin-derived antimicrobial peptide found in humans. It is formed by 37 residues, has amphiphilic behavior, in some cases displays α-helical structure, and, most importantly, has been shown to exhibit a broad spectrum of antimicrobial activity 15,16 . LL37 acts as rst line of defense against bacteria, virus and fungi 17 . The wound healing potential of free and immobilized (LL37-AuNP) LL37 has been investigated recently 9 . The results showed that LL37/AuNP conjugate have enhanced pro-regenerative properties when compared to the soluble LL37. The dynamics behind the formation and stability of this conjugate, however, is yet to be further explored. Details on how the peptide is adsorbed onto the surface and if the nal structure is stable are issues not well known. Large peptides and,

, proteins, are characterized by multiple local minimum which are separated

a fortiori

by high-energy barriers which results in a coarse energy landscape 1821 . Advanced sampling methods, such as metadynamics 22 , are required to properly address the conformation 3



variability inherent to these molecular systems. Metadynamics is a powerful technique for enhancing sampling simulations and reconstructing the free-energy surface (FES) as a function of few selected degrees of freedom, often referred to as collective variables (CVs). The algorithm discourages previously visited sates being resampled, allowing to allocate computacional resources to explore a broader section of the FES 22 . Considering the potential of the LL37/AuNP conjugate, we hereby present a thorough computational study on the interaction of a LL37 AMP with a AuNP, in which the investigated AMP is modied with cysteine (CYS) at the C-terminal to favor its interaction with the Np surface. In looking towards a complete picture of the dynamics of the target AMP, four scenarios were tested: the interaction of i) plain (LL37) and ii) CYS-modied (LL37-SH), AMPs with a 4.2 nm face-centered-cubic (fcc) lattice AuNP; reference systems consisting in iii) LL37-SH and iv) LL37, both alone in water. A set of all-atom well-tempered metadynamics simulations were performed for the four independent systems in order to sample the free energy barriers associated with adsorption process of the LL37-SH onto the AuNP. This study also provided original quantitative information, in terms of free-energy, on the immobilization of the AMP (LL37-SH) onto a AuNP. These computational results contribute to rationalize some of the experimental ndings reported recently 9 .

Methods and computational details MD simulations were performed using GROMACS 2016.1 code 2326 with the PLUMED2 27 plug-in. A 4.2 nm diameter AuNP 28 in water was modeled based on the force eld parameters developed by Heinz

et al

(Figure 1(a)). 29 . The LL37 (LLGDFFRKSKEKIGKEFKRIVQRI-

KDFLRNLVPRTES) initial structure was obtained from the Protein Data Bank 30 (PUB ID: 10.2210/pdb2K6O/pdb 31 ). This structure was modied with a C-terminal cysteine to create the LL37-SH (LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTESC). The relaxed congurations of the AMPs can be found in Figure 1(b) and 1(c). The force elds 99SB 32

4


Page 4 of 24

Page 5 of 24 1 2 3 4 5 6 7 8 9 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


(a)

(b)

(c)

Figure 1: (a)AuNP model formed by 1985 Au atoms with face-centered-cubic lattice, dominated by {111} and {100} facets. Initial conformation of the (b) LL37-SH and (c) LL37 AMPs. These conformations are the outcome of the relaxation run. and TIP3P 33 were used to model the peptides and the water molecules, respectively. Several well-tempered metadynamics 22,34 runs under aqueous conditions were performed: two runs in the presence of a AuNP with LL37-SH and LL37 ; and two runs without AuNP with the LL37-SH and LL37 to be used as reference. The AMPs were rstly relaxed in aqueous solution for 50 ns at 323.15 K and then at 298.15 K for further 50 ns. The resulting conformations were then used as the initial structures for the subsequent simulations. All systems were pre-equilibrated using energy minimization simulations. Box volume was relaxed for 100 ps in NpT ensemble under periodic boundary conditions (PBC). A box of 1500 nm2 (10 nm in xy and 15 nm in z) was prepared with one LL37-SH and a 4.2 nm diameter AuNP. The well-tempered metadynamics was performed for 500 ns. The radius of gyration (Rg ), and distance of the center of mass (COM) of the peptide to the COM of the AuNP (dCOM ) were chosen as CVs. For the system without AuNP, the Rg was considered as CV. The widths of the Gaussian function were set to 0.1 nm for the (Rg ), and 0.2 nm for dCOM . The height of the Gaussian functions was set to 1 kJ mol−1 , and the bias potential was updated every 5 ps throughout the simulation. The temperature was set to 298.15 K with a bias factor of 10 K. The simulations were carried out with a 2 fs time step, in NVT ensemble under PBC. The temperature was xed at 298.15 K using the Nose-Hoover thermostat 35 . The electrostatic 5



Page 6 of 24

interactions were treated with the particle mesh Ewald (PME) method 36,37 and a real space cuto of 1.0 nm. A cuto of 1.2 nm was applied to Lennard-Jones interactions. Hydrogen bonds were constrained with the LINCS algorithm 38,39 . The peptide adsorption behavior was performed in water-based solvents under neutral pH. For the trajectory analysis the following GROMACS and PLUMED tools were used: gmx covar

and

for conformational entropy calculations;

gmx anaeig

vent accessible surface area calculations; ing the simulation in the HILLS le; VERT_TO_FES

sum_hills

gmx sasa

for sol-

to sum the Gaussians deposited dur-

REWEIGHT_BIAS, HISTOGRAM

and

CON-

algorithms 40 were used to reweight the FE energy proles (similar ap-

proach is detailed by Klug

. 41 ).

et al

Results and Discussion The LL37-SH/AuNP conjugate is a system whose therapeutic activity for the enhanced wound healing was recently reported by Comune

. 9 . The authors demonstrated the

et al

improved activity of LL37-SH/AuNP conjugate compared to that of the soluble LL37-SH peptide, in consequence of a prolonged activation of EGFR in keratinocytes. An increase of the skin wound healing activity in an acute wound model was also reported for the LL37SH/AuNP conjugate. For these reasons, the LL37-SH/AuNP conjugate is the main focus of the current discussion. Nevertheless, from a fundamental point of view, a complete analysis should also include the relevant parts of the system. Thus, results of the LL37/AuNP conjugate and LL37-SH and LL37 alone, will also be mentioned and discussed when appropriate throughout the manuscript. The analysis is based on four selected metrics: minimum distance between the centers of mass of peptides and AuNP, dCOM ; radius of gyration Rg ; the end-to-end distance, dEnd-End ; and RMSD of the peptide backbones (the initial structure was used as reference). The discussion is organized as follows: 1) structural-related analysis; 2) free energy and entropy analysis; 3) LL37-SH bioactivity assessment.

6


Page 7 of 24 1 2 3 4 5 6 7 8 9 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


Structural analysis

The proles in Figure 2(a) show that both LL37-SH and LL37 peptides migrate to the gold surface within the rst 20 ns. Overall, the peptides exhibit a typical three-stage behavior already reported in other studies 1214 . In the case of LL37-SH, the rst stage, corresponding to the diusion of the peptide, is characterized by the approximation of the the peptide to the surface within the rst 5 ns, after which the anchoring and step-wise lock-down to the surface occur (along the 5 ns to 100 ns interval). At the end of these two stages, the peptide is considered fully adsorbed to the surface, with limited rearrangement. Notably, the stepwise lock-down for LL37 takes longer times when compared to the LL37-SH (see Figure 2(a) and 2(c)). The time evolution of the distance of each residue to the surface is shown in Figure 2(c). Globally, both peptides display a similar behavior, with all the residues fastly moving towards the AuNP within the rst few nanoseconds of the simulation. However, there are some distinct features near the C-terminal. These dierences are most likely caused by the presence of the CYS residue, whose strong interaction with the Au surface ends up conditioning the adjacent residues, and, consequently, their adsorption proles (compared to those of LL37). In fact, the CYS residue itself adsorbs and locks to the AuNP at the very beginning of simulation time. The information provided by dEnd-End , RMSD and Rg (Figure 3) data, which not only give a clearer picture on the peptide evolution throughout the adsorption process, but also provide additional information regarding the conformational trend and arrangement of the LL37-SH, i.e. whether it remains elongated or folded. Again, the plateauing eect is observed in all metrics plotted in Figure 3, which demonstrates the convergence of the system to a stationary state. On the other hand, while the dEnd-End converges to ca. 3.40 nm in the LL37-SH/AuNP conjugate, it it arrives at ca. 1.84 nm in the case of the LL37-SH/AuNP. This is a clear indication that the adsorption of the peptide on the AuNP surface promotes its elongation, an eect which, however, is quite more pronounced for the LL37-SH when compared to LL37. 7



(a)

Page 8 of 24

(b)

(c)

Figure 2: Time evolution of (a) distance between the centers of mass of peptides and AuNP (dCOM ); (b) snapshot of the conjugates (LL37-SH left and LL37 right) for 50 ns of simulation; (c) distance of each residue to the AuNP for each peptide. (black - LL37-SH, gray - LL37) Free energy and entropy analysis

To check if the FES of the system was properly scanned, its value was determined as a function of the simulation time in terms of the selected CVs. At convergence, the reconstructed 8 ACS Paragon Plus Environment

Page 9 of 24 1 2 3 4 5 6 7 8 9 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


Figure 3: Time evolution of end-to-end distance (dEnd-End ), RMSD and Rg with the simulation time for LL37-SH/AuNP and LL37/AuNP conjugates (top and bottom left), LL37-SH and LL37 peptides (top and bottom right). proles should be similar. The outcome of this analysis is shown in Figure 4, where the proles start to overlap from a simulation time of 400 ns, hence demonstrating that the system was properly sampled for the selected conditions. This, alongside the results previously described (Figs. 2 and 3), conrms that the simulation most likely converged. Figure 5(a) oers an overview of the reweighted FES as function of both CV, while Figs. 5(b) and 5(c) show the normalized Free Energy (FEnorm ) as a function each each CV. A global minimum of -31.4 kJ mol−1 occurs at Rg = 1.62 nm and dCOM = 2.11 nm in the case of LL37-SH/AuNP conjugate (-31.3 kJ mol−1 at 1.55 nm Rg and 2.16 nm dCOM for LL37/AuNP). Figure 5(a) highlights the relation between Rg and dCOM (red curve). When LL37-SH is free the Rg value uctuates between 1.3 to 1.5 nm. However, as it gets closer to the AuNP, this value increases, hitting the lowest free energy at 1.62 nm. The LL37 exhibits a much steadier value when free, around 1.16 nm, eventually hitting 1.55 nm near the surface.

9



Page 10 of 24

The energy proles for both CVs (Figure 5(b) and 5(c)) are mainly characterized by a single basin, meaning that the FES prole is not complex, and that the metadynamics approach was able to easily reach global minimum and properly describe the complete FES. Bellucci

. 42 used a similar methodology (metadynamics) to investigate the interaction

et al

of alanine dipeptide with a gold surface. Some shoulders in the FES prole for dCOM were reported, and the ascribed such a trend to the rst contact of the peptide with the gold surface. In fact, for LL37 (Figure 5(b) gray curve), dCOM starts decreasing steadily from 4.5 nm onwards to the global minimum, but a small shoulder at ca. 2.8 nm is observed. This feature can be related with the rst contact of the peptide with the surface and indicate that a metastable conjugate might occur during the adsorption process of the LL37 to the AuNP (two basins visible in Figure 5(a)). On the other hand, in the case of LL37-SH (Figure 5(b) black curve), dCOM decreases steadily up to 3.5 nm before a steeper descent to the global minimum. These FE proles indicate that the LL37 peptide can take a longer adsorption process, which is in consistent with the precedent analysis. Being CYS the only dierentiating element between the peptides, its presence seems to have a noticeable eect in the formation of the conjugate, and the unveiled smoother FES shows that LL37-SH/AuNP

1 ns 100 ns 200 ns 300 ns 400 ns 500 ns 550 ns 570 ns 580 ns

1 ns 100 ns 200 ns 300 ns 400 ns 500 ns 550 ns 570 ns 580 ns

(a)

(b)

Figure 4: Normalized FE at dierent timestamps for the evaluated CVs and for the LL37SH/AuNP conjugate: (a) dCOM ; (b) Rg .

10


Page 11 of 24 1 2 3 4 5 6 7 8 9 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


(a)

(b)

(c)

Figure 5: (a) Reweighted FES of the system taking into account both CVs for LL37-SH (left) and LL37 (right). Red curve indicate the path of lowest FES for each unique dCOM value. Normalized relative FE (reweighted) of the evaluated CVs: (b) (dCOM ); (c) (Rg ). (black LL37-SH, gray - LL37) conjugate is more easily achieved. A similar behavior, i.e. enhancement of the interaction of a AMP with a gold surface due to the presence of a CYS residue, was reported by Rai

et

. 5.

al

When comparing the Rg for both peptides, remarkable dierences are found depending whether AMPs are in the presence of the AuNP or alone in water (Figure 5(c)). As the FE 11



Page 12 of 24

reaches the minimum, Rg tends to ca. 1.6 nm for both LL37-SH/AuNP and LL37/AuNP conjugates, whereas for the peptides alone that characteristic value decreases to ca. 1.0 nm. Hence, the AMPs in the presence of the AuNP are preferably elongated instead of folded, i.e. the AuNP has a great impact over the AMP structure, somehow compensating the preferable energy state (characterized by lower Rg ) of the LL37-SH or LL37 specimens alone. This result, combined with the data shown in Figure 3, demonstrate that LL37-SH tends to be elongated in the AuNP surface. In the light of these results, we postulate that the elongation of the peptide might lead to an increased availability of the more active functional groups compared to the free peptide, thus enhancing its activity. This inference is consistent with, and support, experimental evidences on these systems recently reported 9 . With FES properly described for both conjugates, the conformational entropy of each peptide was determined. This metric is commonly used to assess a protein stability. 43 It can be easily determined by evaluating the covariance matrix of Cartesian positional coordinates obtainable by computer simulation, namely from the Schlitter's equation (Eq. 1), which uses to the quasi-harmonic quantum mechanical formulation (S0 ): 44

kB T e2 1 2 mhx ic . S = kB ln 1 + 2 }2 0

(1)

where kB is Boltzmann's constant, T is the temperature, e is Euler's Number, } = h/2π (here

h is Planck's constant), m is the mass and hx2 ic classical variance. By using the covariance matrix Cartesian positional coordinates, the method circumvents the need to express the entropy in internal coordinates, although at expenses of a somewhat lower accuracy 45 . We believe that the fact of being a quite approximative approach is not a critical issue when used for comparative purposes, as the case. In fact, this approach has being extensively Free used on peptides 4650 . The conformational entropies (Sconf ) for both peptides free (Sconf ) Ads and interacting (adsorbed) with the AuNP (Sconf ) were computed for the last 5 ns. The

results are summarized in Table 1. The values for the free peptides reveal a signicant

12


Page 13 of 24 1 2 3 4 5 6 7 8 9 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


loss in entropy. Being the incorporation of CYS (i.e., on going from LL37 to LL37-SH) the only dierence between the peptides, this loss is probably consequence of an additional intramolecular stabilization potentiated by that residue. The loss in Sconf observed after the adsorption (both peptides) is explained by the decrease of exibility, i.e. internal degrees of freedom, caused by the structural lock of the peptides into the AuNP. When confronting the two entries of Table 1, it appears that the adsorption process has a higher impact in the LL37-SH conformation exibility, probably due to the strong interaction between cysteine and the AuNP. A detailed analysis of this interaction, however, requires to consider not only entropic eects but also enthalpic issues, notably a presumed covalent binding involving the CYS and the metal surface. The later topic can not be aorded by classical molecular dynamics and therefore is out of the scope of this study. Table 1: Conformational entropy of the peptide backbone using Schlitter's quasiharmonic approximation.

LL37-SH LL37

Free Sconf (J mol−1 K−1 ) 6275.48 6616.61

Ads Sconf (J mol−1 K−1 ) 5154.60 5592.82

LL37-SH bioactivity assessment

Aiming at understanding the reasoning behind the higher antimicrobial activity of the LL37SH adsorbed to the AuNP than its free counterpart recently reported 9 , the solvent accessible surface area (SASA) and the distance of the positively charged functional groups to the peptide backbone was computed. The SASA values refer the total solvent accessible surface area if a molecule was in a bulk solvent alone. This means that this value is not aected when LL37-SH adsorbs to the AuNP surface. Therefore, a decrease of the SASA would represent a compaction of the peptide structure with a decreasing in its surface area, rather than a contact between the peptide and surface with the subsequent exclusion of water 51,52 . However, as it can be seen in Figure 6(a), the LL37-SH/AuNP conjugate has a substantially 13



(a)

Page 14 of 24

(b)

Figure 6: (a) Solvent accessible surface area and (b) distance of the positively charged functional groups to the peptide backbone for the LL37-SH with the AuNP and alone. higher SASA, which indicates that the LL37-SH peptide alone tends to fold. Thus, the AuNP induces the LL37-SH to adopt a more elongated conformation, which is expected to promote an increased availability of the functional groups to interact with cells. This was further explored by assessing the location of the positively charged functional groups relatively to the peptide backbone. In fact, considering that positively charged residues play an important role in the AMP activity 53,54 , the understanding of its location is crucial. The LL37-SH AMP, among other residues, has multiple arginine and a single lysine. These two residues are characterized by being positively charged. At the atomistic level, the functional groups responsible for this charge are guanidino group of arginine, and a protonated amine in lysine. Thus, the distance between these groups, 11 in total, to the peptide backbone was computed for the last 10 ns of simulation. The average values for each system are resumed in Figure 6(b) (error bars are not shown do to their low values, around 0.01% for the highest value). Overall, for LL37-SH/AuNP conjugate 7 of the 11 functional groups are further away from the AMP backbone when compared to LL37-SH alone. This clearly suggests that the LL37-SH adsorbed to the AuNP tends to have the positive side chains more available to interact with biological systems. Thus, these two factors i) the tendency to be elongated and; ii) having the positively charged side chains more available form a quite plausible 14


Page 15 of 24 1 2 3 4 5 6 7 8 9 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


explanation for the enhanced activity of LL37-SH when adsorbed to a AuNP.

Conclusions The interaction of LL37-SH (a cathelicidin derived AMP) with a AuNP and the activity of the subsequently formed conjugate was investigated from atomistic-scale simulations by using well-tempered metadynamics. The temporal analysis of dCOM , dEnd-End , RMSD and Rg data showed that the AMP with CYS tends to adsorb and stabilize onto the surface of AuNP faster than LL37. The Sconf analysis showed that the AuNP has a substantial eect on the conformational entropy of both peptides, whose values decreased when adsorbed. Some differences between LL37-SH and LL37 were also found, with the CYS residue reducing the Sconf of the peptide both free and adsorbed. The FES in terms of dCOM and Rg as CVs evidenced a comparable landscape for both peptides, although a secondary low energy site has been identied in the case of LL37, indicating a possible occurrence of metastable LL37/AuNP. Again, the incorporation of CYS in the base peptide appeared to have a noticeable eect in the formation of the LL37-SH/AuNP conjugate, as suggested by a smoother FES. The comparative simulation of the LL37-SH free in water helped in unveiling the eect of the AuNP on this peptide. The results showed a dramatic decreasing in Rg (ca. 50 %.) in the absence of the AuNP, meaning that the peptide is preferably elongated after being adsorbed to the surface of AuNP. This was conrmed by the comparative analysis based on the solvent accessible surface area for the free LL37-SH and in the presence of the AuNP. Together, the revealed elongation and the identied readiness of biological functional groups in the LL37-SH/AuNP conjugate explain the reported enhanced activity of the LL37-SH/AuNP conjugate compared to the LL37-SH peptide

.

per se

15



Page 16 of 24

References

(1) Wang, G.; Li, X.; Wang, Z. APD3: the antimicrobial peptide database as a tool for research and education.

,

, D1087D1093.

Nucleic Acids Res. 2016 44

(2) Li, Y.; Xiang, Q.; Zhang, Q.; Huang, Y.; Su, Z. Overview on the recent study of antimicrobial peptides: origins, functions, relative mechanisms and application. 2012, 37,

Peptides

20715.

(3) Hancock, R. E. W.; Sahl, H.-G. Antimicrobial and host-defense peptides as new antiinfective therapeutic strategies.

,

, 15511557.

Nat. Biotechnol. 2006 24

(4) Nakatsuji, T.; Gallo, R. L. Antimicrobial Peptides: Old Molecules with New Ideas. ,

J.

, 887895.

Invest. Dermatol. 2012 132

(5) Rai, A.; Pinto, S.; Velho, T. R.; Ferreira, A. F.; Moita, C.; Trivedi, U.; Evangelista, M.; Comune, M.; Rumbaugh, K. P.; Simões, P. N. et al. One-step synthesis of high-density peptide-conjugated gold nanoparticles with antimicrobial ecacy in a systemic infection model.

,

, 99110.

Biomaterials 2016 85

(6) Peng, L. H.; Huang, Y. F.; Zhang, C. Z.; Niu, J.; Chen, Y.; Chu, Y.; Jiang, Z. H.; Gao, J. Q.; Mao, Z. W. Integration of antimicrobial peptides with gold nanoparticles as unique non-viral vectors for gene delivery to mesenchymal stem cells with antibacterial activity.

,

, 137149.

Biomaterials 2016 103

(7) Wadhwani, P.; Heidenreich, N.; Podeyn, B.; Bürck, J.; Ulrich, A. S. Antibiotic gold: tethering of antimicrobial peptides to gold nanoparticles maintains conformational exibility of peptides and improves trypsin susceptibility. Biomater.

, , 817828.

Sci. 2017 5

(8) Casciaro, B.; Moros, M.; Rivera-Fernández, S.; Bellelli, A.; de la Fuente, J. M.; Mangoni, M. L. Gold-nanoparticles coated with the antimicrobial peptide esculentin-1a(1-

16


Page 17 of 24 1 2 3 4 5 6 7 8 9 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


21)NH2 as a reliable strategy for antipseudomonal drugs.

,

Acta Biomater. 2017

,

47

170181. (9) Comune, M.; Rai, A.; Chereddy, K. K.; Pinto, S.; Aday, S.; Ferreira, A. F.; Zonari, A.; Blersch, J.; Cunha, R.; Rodrigues, R. et al. Antimicrobial peptide-gold nanoscale therapeutic formulation with high skin regenerative potential.

,

J. Control. Release 2017

(10) Dror, R. O.; Dirks, R. M.; Grossman, J.; Xu, H.; Shaw, D. E. Biomolecular Simulation: A Computational Microscope for Molecular Biology.

,

Annu. Rev. Biophys. 2012

,

41

429452. (11) Heinz, H.; Farmer, B. L.; Pandey, R. B.; Slocik, J. M.; Patnaik, S. S.; Pachter, R.; Naik, R. R. Nature of molecular interactions of peptides with gold, palladium, and Pd-Au bimetal surfaces in aqueous solution.

,

, 970414.

J. Am. Chem. Soc. 2009 131

(12) Yu, J.; Becker, M. L.; Carri, G. A. The inuence of amino acid sequence and functionality on the binding process of peptides onto gold surfaces.

,

Langmuir 2012

,

28

140817. (13) Penna, M. J.; Mijajlovic, M.; Biggs, M. J. Molecular-level understanding of protein adsorption at the interface between water and a strongly interacting uncharged solid surface.

,

, 53235331.

J. Am. Chem. Soc. 2014 136

(14) Ferreira, A. F.; Rai, A.; Ferreira, L.; Simões, P. N. Findings on the interaction of the antimicrobial peptide cecropin-melittin with a gold surface from molecular dynamics studies.

,

Eur. Biophys. J. 2016

(15) Dürr, U. H.; Sudheendra, U.; Ramamoorthy, A. LL-37, the only human member of the cathelicidin family of antimicrobial peptides. 2006, 1758,

Biochim. Biophys. Acta - Biomembr.

14081425.

17



Page 18 of 24

(16) Wiesner, J.; Vilcinskas, A. Antimicrobial peptides: The ancient arm of the human immune system.

, , 440464.

Virulence 2010 1

(17) Vandamme, D.; Landuyt, B.; Luyten, W.; Schoofs, L. A comprehensive summary of LL-37, the factotum human cathelicidin peptide.

,

, 2235.

Cell. Immunol. 2012 280

(18) Onuchic, J. N.; Luthey-Schulten, Z.; Wolynes, P. G. THEORY OF PROTEIN FOLDING: The Energy Landscape Perspective.

,

, 545600.

Annu. Rev. Phys. Chem. 1997 48

(19) Bernardi, R. C.; Melo, M. C.; Schulten, K. Enhanced sampling techniques in molecular dynamics simulations of biological systems. Biochim.

,

Biophys. Acta - Gen. Subj. 2015

, 872877.

1850

(20) Spiwok, V.; Sucur, Z.; Hosek, P. Enhanced sampling techniques in biomolecular simulations.

,

, 113040.

Biotechnol. Adv. 2015 33

(21) Galvelis, R.; Sugita, Y. Replica state exchange metadynamics for improving the convergence of free energy estimates.

,

, 14461455.

J. Comput. Chem. 2015 36

(22) Barducci, A.; Bonomi, M.; Parrinello, M. Metadynamics. Wiley

Interdiscip. Rev. Com-

, , 826843.

put. Mol. Sci. 2011 1

(23) Berendsen, H. GROMACS: A message-passing parallel molecular dynamics implementation.

,

, 4356.

Comput. Phys. Commun. 1995 91

(24) Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. C. GROMACS: fast, exible, and free.

,

, 170118.


(25) Páll, S.; Abraham, M. J.; Kutzner, C.; Hess, B.; Lindahl, E. Tackling Challenges in Molecular Dynamics Simulations with GROMACS

Exascale Software

; 2015; pp 327.

(26) Abraham, M. J.; Murtola, T.; Schulz, R.; Páll, S.; Smith, J. C.; Hess, B.; Lindahl, E. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers.

,

, 1925.

SoftwareX 2015 1-2

18


Page 19 of 24 1 2 3 4 5 6 7 8 9 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


(27) Tribello, G. A.; Bonomi, M.; Branduardi, D.; Camilloni, C.; Bussi, G. PLUMED 2: New feathers for an old bird.

,

, 604613.

Comput. Phys. Commun. 2014 185

(28) Yu, J.; Becker, M. L.; Carri, G. a. A molecular dynamics simulation of the stabilitylimited growth mechanism of peptide-mediated gold-nanoparticle synthesis. 2010, 6,

Small

22425.

(29) Heinz, H.; Vaia, R. A.; Farmer, B. L.; Naik, R. R. Accurate Simulation of Surfaces and Interfaces of Face-Centered Cubic Metals Using 12-6 and 9-6 Lennard-Jones Potentials. ,

, 1728117290.

J. Phys. Chem. C 2008 112

(30) Berman, H. M. The Protein Data Bank.

,

, 235242.

Nucleic Acids Res. 2000 28

(31) Wang, G. Structures of Human Host Defense Cathelicidin LL-37 and Its Smallest Antimicrobial Peptide KR-12 in Lipid Micelles.

,

, 3263732643.

J. Biol. Chem. 2008 283

(32) Hornak, V.; Abel, R.; Okur, A.; Strockbine, B.; Roitberg, A.; Simmerling, C. Comparison of multiple Amber force elds and development of improved protein backbone parameters.

,

, 71225.

Proteins 2006 65

(33) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. Comparison of simple potential functions for simulating liquid water. 1983, 79,

J. Chem. Phys.

926.

(34) Barducci, A.; Bussi, G.; Parrinello, M. Well-tempered metadynamics: A smoothly converging and tunable free-energy method.

,

, 14.

Phys. Rev. Lett. 2008 100

(35) Evans, D. J.; Holian, B. L. The Nose-Hoover thermostat.

,

J. Chem. Phys. 1985

,

83

4069. (36) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G. A smooth particle mesh Ewald method.

,

, 8577.

J. Chem. Phys. 1995 103

19



Page 20 of 24

(37) Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald: An N.log(N) method for Ewald sums in large systems.

,

, 10089.

J. Chem. Phys. 1993 98

(38) Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M. LINCS: A linear constraint solver for molecular simulations.

,

, 14631472.


(39) Hess, B. P-LINCS: A Parallel Linear Constraint Solver for Molecular Simulation.

J.

, , 116122.

Chem. Theory Comput. 2008 4

(40) Tiwary, P.; Parrinello, M. A Time-Independent Free Energy Estimator for Metadynamics.

,

, 736742.

J. Phys. Chem. B 2015 119

(41) Klug, J.; Triguero, C.; Del Popolo, M.; Tribello, G. A. Using Intrinsic Surfaces to Calculate the Free-Energy Change When Nanoparticles Adsorb on Membranes. J. Phys. , acs.jpcb.8b03661.

Chem. B 2018

(42) Bellucci, L.; Corni, S. Interaction with a Gold Surface Reshapes the Free Energy Landscape of Alanine Dipeptide.

,

, 1135711364.

J. Phys. Chem. C 2014 118

(43) Doig, A. J.; Sternberg, M. J. E. Side-chain conformational entropy in protein folding. , , 22472251.

Protein Sci. 1995 4

(44) Schlitter, J. Estimation of absolute and relative entropies of macromolecules using the covariance matrix.

,

, 617621.

Chem. Phys. Lett. 1993 215

(45) Hikiri, S.; Yoshidome, T.; Ikeguchi, M. Computational Methods for Congurational Entropy Using Internal and Cartesian Coordinates.

,

J. Chem. Theory Comput. 2016

, 59906000.

12

(46) Schäfer, H.; Mark, A. E.; van Gunsteren, W. F. Absolute entropies from molecular dynamics simulation trajectories.

,

, 78097817.

J. Chem. Phys. 2000 113

20


Page 21 of 24 1 2 3 4 5 6 7 8 9 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


(47) Schäfer, H.; Daura, X.; Mark, A. E.; van Gunsteren, W. F. Entropy calculations on a reversibly folding peptide: Changes in solute free energy cannot explain folding behavior. ,

, 4556.

Proteins Struct. Funct. Genet. 2001 43

(48) Allison, J. R.; Müller, M.; van Gunsteren, W. F. A comparison of the dierent helices adopted by α- and β -peptides suggests dierent reasons for their stability. 2010, 19,

Protein Sci.

21862195.

(49) Narzi, D.; Becker, C. M.; Fiorillo, M. T.; Uchanska-Ziegler, B.; Ziegler, A.; Böckmann, R. A. Dynamical Characterization of Two Dierentially Disease Associated MHC Class I Proteins in Complex with Viral and Self-Peptides. J. Mol. Biol. 2012, 415, 429 442. (50) Nayar, D.; Chakravarty, C. Sensitivity of local hydration behaviour and conformational preferences of peptides to choice of water model.

,

Phys. Chem. Chem. Phys. 2014

,

16

1019910213. (51) Eisenhaber, F.; Lijnzaad, P.; Argos, P.; Sander, C.; Scharf, M. The double cubic lattice method: Ecient approaches to numerical integration of surface area and volume and to dot surface contouring of molecular assemblies. J. Comput. Chem. 1995, 16, 273284. (52) Agashe, M.; Raut, V.; Stuart, S. J.; Latour, R. A. Molecular Simulation To Characterize the Adsorption Behavior of a Fibrinogen γ -Chain Fragment. Langmuir

2005, 21,

1103

1117. (53) Jiang, Z.; Vasil, A. I.; Hale, J. D.; Hancock, R. E. W.; Vasil, M. L.; Hodges, R. S. Eects of net charge and the number of positively charged residues on the biological activity of amphipathic α-helical cationic antimicrobial peptides.

,

Biopolymers 2008

, 369383.

90

(54) Jiang, Z.; Mant, C. T.; Vasil, M.; Hodges, R. S. Role of positively charged residues on the polar and non-polar faces of amphipathic α-helical antimicrobial peptides on 21



specicity and selectivity for Gram-negative pathogens. , 7592.

91

22


Page 22 of 24

,

Chem. Biol. Drug Des. 2018

Page 23 of 24 1 2 3 4 5 6 7 8 9 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


Graphical TOC Entry

Schematic representation the FES at dierent stages of simulation: 1) beginning of simulation; 2) peptide near the AuNP surface; 3) conjugate after adsorption.

23


The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32


Page 24 of 24

Atomistic-Level Investigation of a LL37 Conjugated ... - ACS Publications

Recommend Documents