Structural Studies of a Lipid-Binding Peptide from Tunicate Hemocytes with Anti-Biofilm Activity

Abstract

Clavanins is a class of peptides (23aa) histidine-rich, free of post-translational modifications. Clavanins have been studied largely for their ability to disrupt bacterial membranes. In the present study, the interaction of clavanin A with membranes was assessed by dynamic light scattering, zeta potential and permeabilization assays. We observed through those assays that clavanin A lysis bacterial cells at concentrations corresponding to its MIC. Further, the structure and function of clavanin A was investigated. To better understand how clavanin interacted with bacteria, its NMR structure was elucidated. The solution state NMR structure of clavanin A in the presence of TFE-d₃ indicated an α-helical conformation. Secondary structures, based on circular dichroism measurements in anionic sodium dodecyl sulfate (SDS) and TFE (2,2,2-trifluorethanol), in silico lipid-peptide docking and molecular simulations with lipids DPPC and DOPC revealed that clavanin A can adopt a variety of folds, possibly influencing its different functions. Microcalorimetry assays revealed that clavanin A was capable of discriminating between different lipids. Finally, clavanin A was found to eradicate bacterial biofilms representing a previously unrecognized function.

Structural Studies of a Lipid-Binding Peptide from Tunicate Activity

OPEN

R

A P

Osmar N. Silva,, Eliane S. F. Alves, Csar de la Fuente-Nez, Suzana M. Ribeiro, Santi M. Mandal, Diana Gaspar, Ana S.Veiga, Miguel A. R. B. Castanho,Cesar A. S. Andrade, Jessica M. Nascimento, Isabel C. M. Fensterseifer, William F. Porto, Jose R. Correa, Robert. E.W. Hancock, Suresh Korpole, Aline L. Oliveira,, Luciano M. Liao & Octavio L. Franco,,

have been studied largely for their ability to disrupt bacterial membranes. In the present study, the interaction of clavanin A with membranes was assessed by dynamic light scattering, zeta potential and permeabilization assays. We observed through those assays that clavanin A lysis bacterial cellsat concentrations corresponding to its MIC. Further, the structure and function of clavanin A was investigated. To better understand how clavanin interacted with bacteria, its NMR structure was elucidated. The solution state NMR structure of clavanin A in the presence of TFE-d indicated an -helical conformation. Secondary structures, based on circular dichroism measurements in anionic in silico lipid-peptide docking and molecular simulations with lipids DPPC and DOPC revealed that clavanin A can adopt a variety of

The larvae of protochordates, such as the solitary tunicate Styela clava, display an assemblage of phenotypes that reveal their relationship with vertebrates1, including the presence of a post-anal tail, pharyngeal gill slits, a dorsal tubular nerve cord and a notochord. Furthermore, the body cavity and tissues contain phagocytic cells named hemocytes, which resemble vertebrate granulocytes and macrophages2. In S. clava, hemocytes synthesize two families of antimicrobial peptides, namely styelins and clavanins3. While styelins are larger peptides (~3.6kDa) that contain hydroxylysines and other modied residues4, clavanins are the smallest known histidine-rich peptides, composed of only 23-amino-acid residues in length and having no post-translational modications5.

Departamento Biologia, Instituto de Cincias Biolgicas, Programa de ps-graduao em Gentica e SInova, Ps-graduao Laboratrio de RMN, Instituto Department of Microbiology Central Research Facility,

Programa de Ps-Graduao em Inovao Teraputica, Universidade Federal de Pernambuco, Department of Cellular Biology, Institute of Biological Sciences, University of Brasilia, Programa de Ps-Graduao em Cincias Genmicas e Biotecnologia, Centro de Anlises Protemicas

SCIENTIFIC REPORTS

1

www.nature.com/scientificreports/

Sample Peptide -potential (mV)

POPC None 1.530.20 POPC Clavanin 0.650.43 POPC:POPG None 19.800.56 POPC:POPG Clavanin 19.401.76

Table 1. -potential values for POPC and POPC:POPG at 100M vesicles before and aer clavanin A exposition.

Figure 1. Determination of the diameter of unilamellar lipid vesicles (ULV). The hydrodynamic diameter (Dh) of lipid vesicles in the absence (solid squares) and presence of clavanin A (empty squares) as assessedby dynamic light scattering. In the absence of peptide, both POPC (red) and POPC:POPG (blue) vesicles hada narrow distribution of size centred at 110nm. When the peptide was present, there was evidence of higher association with the anionic POPC:POPG vesicles (less residual 110nm liposomes) as well as a large amount of aggregation and multimodal distributions, likely due to fusion, that were detected.

In recent years, clavanins have been studied with particular attention to their ability to form pores in and disrupt model membranes68. Indeed, some studies suggest that clavanin A (VFQFLGKIIHHVGNFVHGFSHVF-NH2)

is efficiently inserted into dierent phospholipid monolayers in a pH-dependent manner, mainly as a result of hydrophobic interactions. For example, van Kan et al.7 observed that, at neutral pH, wild-type clavanin efficiently released uorophores from unilamellar vesicles, thus indicating a nonspecic permeabilization mechanism. However, clavanin A was unable to permeabilize model bilayers at low pH, suggesting that its activity could be modulated under dierent conditions, as observed for certain broad-spectrum antimicrobial peptides from plants9. Nevertheless, it has been suggested that all cationic amphipathic peptides disrupt membranes at sufficiently high concentrations10,11. Therefore, it was important to determine how peptide fold patterns varied in dierent environments and how this inuenced their functions. Despite previous eorts, very little is known about the structure of clavanin A and its functions other than its antimicrobial property.

The data presented in this study represent the rst description of the structure of clavanin A, its interaction with lipids, and present a novel activity for this peptide: its ability to eradicate biolms in a broad-spectrum manner. These studies included interaction assays of clavanin A with liposomes and lipids, elucidation of the NMR structure of clavanin A, in silico docking and molecular dynamics, and using a ow cell chamber method to characterize the anti-biolm activity of the peptide.

Results

Evaluation of clavanin A and lipid membrane interactions. Membrane interactions are considered to be important components of the action of amphipathic antimicrobial peptides on bacterial cells12. Mammalian membranes are predominantly composed of neutral lipids, while bacterial membranes contain a net negative charge due to a substantial proportion of negatively charged lipids. The interaction of clavanin A at 10M with neutral (POPC) and negatively charged (POPC:POPG, 8:2) LUVs was studied using assessment of -potential (Table1) and dynamic light scattering (DLS). -potential measurements indicated that there was a small neutralization of surface negative charge on liposomes (Table1). In the absence of peptide, dynamic light scattering measurements (Fig.1) showed that the LUVs had a mean diameter of 100 nm. Addition of clavanin A, however, at a peptide:lipid ratio of 1:100, led to larger hydrodynamic ratios (mean diameter of 1900 nm and 1550nm for POPC and POPC:POPG, respectively) (Table2). The peptide clearly bound better to the anionic liposomes as revealed by the smaller residual 110nm peak. Overall these results demonstrated the direct interaction of clavanin A with liposomes leading to liposome fusion.

SCIENTIFIC REPORTS

2

www.nature.com/scientificreports/

Permeabilization of Liposomes (%) POPC POPC:POPG

3 98.3 92.3 13 98.8 98.7 27 99.4 100.0

Table 2. Permeabilization of liposomes and exposure to dierent concentrations of clavanin A.

Clavanin concentration (g.ml1)

Figure 2. ITC binding experiments of clavanin A with lipid membrane components such as DOPC and DPPC. Displayed are raw heat-associated thermograms (upper panel) and isotherms (lower panel) for each individual set of interactions were displayed. The panels represent the interactions of clavanin A with DOPC(a) and DPPC (b). These data were corrected by subtraction of appropriate blank experiments and then t using nonlinear regression. The conclusions derived from these analyses are presented in Table2.

At higher concentrations, experiments assessing membrane leakage induced by clavanin A, based on detection of increased uorescence due to dye release, were performed according to the method described by Chen and Chen13.

To further study the affinity of clavanin A towards phospolipids, microcalorimetry analyses were performed (Fig.2), showing concentration-dependent binding and endothermic processes. Clavanin A had a greater eect at lower concentrations to vesicle of POPC:POPG mixture at 8:2 molar compared to only POPC. During interaction, it was observed that the G value was negative, indicating that complex formation was energetically favourable and the process was entropy-driven for neutral lipid, POPC and negatively charged vesicles (POPC:POPG), which suggested that the hydrated clavanin A released bound water molecules during its interaction with lipid molecules. The positive values of H and S indicated that the interaction was driven by hydrophobic forces between the peptide and phospholipids. Higher S value for vesicle (POPC:POPG) was observed compared to POPC and the peptide showed better binding affinity to negatively charged vesicles than POPC (Table3).

Clavanin A structure. Far-UV CD spectra of clavanin A (Fig.3A) showed that the peptide had a random conformation in water, which was implied by the presence of a negative band at 200nm. The addition of 1 mM SDS micelles (an anionic lipid mimic) (Fig.3B) or 20% the solvent TFE (Fig.3C) induced a conformational transition in the peptide to an alpha-helical structure, as shown by an intense maximum at 194nm and two minima at 208 and 222nm. Variation of the pH in the environment did not signicantly change the conformation (data not shown). The thermal denaturation curves analyzed from CD measurements at 222nm and pH 5 strongly indicated that clavanin A was a thermally unstable peptide, since spectra consistent with peptide denaturation could be observed (data not shown).

The structure of clavanin A was investigated by solution-phase NMR spectroscopy in 35% TFE, since CD studies indicated an elevated -helical content under these conditions and similar spectra were observed in the anionic lipid mimic SDS. There are reports in the literature that the 2,2,2-triuoroethanol is a solvent that

SCIENTIFIC REPORTS

3

www.nature.com/scientificreports/

ThermaltransitionDOPC (a) DPPC (b) Ergosterol (c) K 1.31105M1 1.56103M1 No interaction H 6.86 E6cal/mol 5.93 E5cal/mol

S 2.08 E4cal/mol/deg 1.76 E3cal/mol/deg

Table 3. ITC binding experiments between clavanin A and DOPC, DPPC and ergosterol. Thermal transition data derived from raw heat associated isotherm using nonlinear regression.

Parameter

Figure 3. Circular dichroism spectra of 41M clavanin A, pH 5.0, at 25C. Spectra are in aqueous solution (A), and in dierent concentrations of SDS (B) and TFE (C).

promotes the formation of hydrogen bonds in regions of the peptide -helical where there is tendency, inducing structure only where there is already a tendency of forming the same14,15. Therefore, due to the limited resolution obtained in experiments using SDS-d25 TFE-d3was used as environment structuring inductor. The complete 1H resonance assignments of the backbone and side-chains of clavanin A were obtained using standard sequential assignment procedures. The elements of secondary structure of the peptide were predicted by the chemical shi index (CSI)16, which showed the deviation of the H chemical shis from the corresponding amino acid embedded in a random coil structure. The presence of at least four consecutive 1 values indicated the presence of a helical conformation. Figure4A indicates the presence of an-helical conformation between residues

SCIENTIFIC REPORTS

4

www.nature.com/scientificreports/

Figure 4. Clavanin A NMR analyses. (A) Chemical Shi Index. Shown are dierences between the observed chemical shis of protons (H) and unstructured value patterns (42). (B) Summary of NOEs used for structure calculation. The thickness of the bars indicates the relative intensities of the NOEs (strong, medium and weak). (C) Front view of N-terminal (le) and C-terminal (right) 3D structure. (D) Lowest energy structure (le) and backbone superimposition of 10 lowest energy structures (right). Color indicates: pink - polar, green- cationic and hydrophobic, blue - residues. (E) Surface structure. Cationic residues are indicated in green. The structure on the right side shows the molecule rotated 180 around the vertical axis.

Gln3-Phe23. The patterns of the NOEs (Fig.4B) also provide a qualitative indication of their secondary structure. The data showed a pattern characteristic of an-helix between residues Phe2-His11 and Val16-His21. Only one restriction of H(i)-NH(i+ 4), between residues Gly13-His17, was observed and no H(i)-NH(i+ 3) spin coupling (NOEs) was found between residues His11-Val16. The NMR structure of clavanin A was calculated using an average of 12.3 restraints for each amino acid residue, including 42 dihedral angles obtained from 13C and 15N chemical shis and was rened to a nal RMSD of 1.05 0.34 (backbone). Statistical analyses are described in Table4. The NMR structure revealed an amphipathic -helical conformation spanning residues Phe2 to Val22 (Fig.4CE). The 3D structure of the peptide in TFE-d3 conformed with a well-dened -helical segment, which was mildly distorted by a slight curvature in the central region between residues Gly13 and His17. These results were consistent with the CSI and NOE data.

The Ramachandran plot indicated the good quality of the NMR models, highlighted by the fact that 100% of and dihedral angles were in the favored allowed regions of -helix17. These results were validated by QUEEN analyses, which indicated both the degree of agreement of the model structures with the experimental data, as well as the quality of their geometrical properties. The rened nal structures in 35% TFE-d3 showed no errors and were well supported by the dataset (data not shown).

DOPC and DPPC lipids stabilized the clavanin A structure in a water environment. Molecular dynamics simulations were performed to assess the interactions between clavanin A and lipids DOPC and DPPC. It was observed that when the -helical clavanin A was added to water, the peptide unfolded, with a clear transition from an -helix to a coiled structure (Fig.5). Nevertheless, when added to DOPC and DPPC, clavanin A was stabilized in a -helix structure (4.1 residues per turn) with exible termini (Fig.5). This dierence between the simulations could also be observed through the evolution of RMSD along the simulations, where the free peptide assumed a root mean square deviation (RMSD) of around 8 , while the peptide in complex with DOPC and DPPC had tighter RMSD values of around 5 and 4, respectively (data not shown). Despite the observation that

SCIENTIFIC REPORTS

5

www.nature.com/scientificreports/

NOE RESTRAINTS

Total number of distance restraints 283

Number of intra-residue restraints 138 Number of sequential restraints (i, i+1) 81 Number of medium range restraints (i, i+j)j=2,3,4 22

Number of long range restraints (|ij| >5) 0

HydrogenBonds 0 Dihedralangles 42 RMSDa

All heavy atoms 1.990.41

Allbackboneatoms 1.050.34

Table 4. Statistics for the 10 lowest energy structures of clavanin A. aDatafromPyMol.

Figure 5. Standardized secondary structure assignment (DSSP analysis) based on molecular dynamics simulations aer 50ns of simulation starting with the NMR structure. The free clavanin A underwent an unfolding process, while in complex with lipids the -helix is stabilized.

clavanin A underwent an unfolding process in water, the region comprised of residues 9IHHV12 remained in an -helical conformation (Fig.6). In interacting with lipids, the helical structured regions (considering both - and -helix) were larger, comprising the residues 6GKIIHHVGNFVHGFS20 and 4FLGKIIHHVGNFVHGF18, for the DPPC and DOPC complexes, respectively (Fig.6). In addition, both lipids bound preferably to the C-terminal region of the peptide. The binding to DOPC was stabilized by residues Ile8, Val12, Gly13, Asn14, His17, Gly18 and Phe19 (Fig.6), while for DPPC, the residues that stabilized the interactions were Phe4, Gly13, Asn14, His17, Gly18, Phe19, Ser20 and Phe23 (Fig. 6).

In vitro Clavanin A was chemically synthesized and further evaluated with regards to its primary structure by MALDI-ToF analyses, revealing a major peak with a molecular mass of 2,665.43Da (data not shown). In our MIC assays using BM2 glucose minimal medium, clavanin A exhibited modest direct antimicrobial activity with MIC values of >64g.ml1 against E. coli 0157, K. pneumoniae 1825971 (KPC-KP, isolate) and the Gram-positive organism methicillin-resistant S. aureus (Fig.7). We also evaluated the anti-biolm activity of Clavanin A. We observed through microplate assays that Clavanin A inhibited the biolm formation of MRSA (8 g.mL1, MBIC 100% and E. coli O157 (16 g.mL1 MBIC 100%) at concentrations lower than its MIC (>64 g.mL1). Using ow cell method, we observed that clavanin A, at 16 g.mL1, successfully perturbed pre-formed (2 days old) biolms formedof K. pneumoniae 1825971 (KPC-KP isolate) and MRSA (Fig.8). On the other hand, treatment with the peptide only reduced the overall biolm thickness of E. coli 0157 pre-formed biolms (Fig.7). While performing these experiments, we observed that clavanin A at times induced lamentation in K. pneumoniae 1825971 (KPC-KP isolate) (Fig.9). Experiments using transmission electron microscopy also revealed lamentation in cells of S. aureus and E. coli with one half of the peptide MIC. Fig.8). Although dead cells were observed, around 50% of all E. coli cells appeared to be undergoing cell division (Fig.9B).

SCIENTIFIC REPORTS

6

www.nature.com/scientificreports/

Figure 6. Predicted conformation of clavanin A interacting with DOPC (A) and DPPC (B). This shows the residues of clavanin A that interact with the lipids. The dark green regions indicate the residues with close contacts to lipids (less than 3.7).

Figure 7. Evaluation of the antimicrobial and antibiolm activity of Clavanina A against MRSA, K. pneumoniae 1825971 (KpC isolate) and E. coli (O157). MIC, minimal inhibitory concentration; MBIC, minimal biolm inhibitory concentration; (C) control growth (bacteria withouth any treatment). Clavanin exhibit a dose-dependent anti-biolm activity against E.coli and MRSA.

SCIENTIFIC REPORTS

7

www.nature.com/scientificreports/

Figure 8. Anti-biolm evaluation of clavanin (A). Flow cell assays were performed by using peptide concentration at a standard concentration of 16g.ml1. Le and right panels correspond to untreated and peptide-treated Escherichia coli, Klebsiella pneumoniae and Staphylococcus aureus. Clavanin A (16g.mL1)

was able to disturb pre-formed (2 days old) biolms of MRSA and KPC-K. pneumoniae 1825971 in ow cell experiments, eliminating biolm formation completely for the MRSA strain and strongly reducing the thickness of K. pneumoniae.

Experiments using transmission electron microscopy showed that untreated cells of E. coli (Fig.9A) and S. aureus (Fig.9C) showed a normal cellular shape with an undamaged structure and slightly waved outer membrane. Aer incubation with a sub-MIC clavanin A concentration, it was observed the formation of lamentation in cells of E. coli (Fig.9B) and S. aureus (Fig.9D).

Discussion

Clavanins are part of the defense system of sea tunicate S. clava against pathogens, and their antimicrobial and membrane disruptive activities have been extensively studied8,19. However, comprehensive structural studies on these peptides have not been carried out. Here, we focused on peptide clavanin A and performed an array of dierent assays aimed at providing insights about the structure and interactions of this peptide. Since lipids are the classical target of clavanin A7, ITC analyses were performed that clearly showed the ability of clavanin to discriminate between dierent lipids (Fig.2). Moreover, in order to shed some light over the possible mechanism of action of clavanin A, dynamic light scattering and -potential were performed to evaluate the interaction of the peptide with membranes. Addition of clavanin A led to the formation of large structures indicating its interaction with lipid vesicles to cause fusion. This eect was even more pronounced with POPC compared to POPC:POPG, despite apparently lower interaction with the uncharged POPC liposomes (Figs1 and 2). Clavanin A is a peptide rich in lipophilic amino acid residues, favoring its interaction with lipid bilayers by means of hydrophobic interactions. This was also suggested by micro calorimetric titration. Diverse AMPs are capable of triggering

SCIENTIFIC REPORTS

8

www.nature.com/scientificreports/

Figure 9. Transmission electronic microscopy of bacterial cells challenged by clavanin. (A) Escherichia coli was evaluated in the absence (A) and in the presence (B) of 32g.ml1 clavanin A. Staphylococcus aureus was evaluated in the absence (C) and in the presence (D) of 64g.ml1 clavanin A. Both bacteria tended to lament upon clavanin A exposure. Black bars correspond to 0.2m.

aggregation of lipid vesicles since they interact with neutral lipids via hydrophobic interactions and neutralize liposome charges1921. For example, peptides analogous to trichogin interact with neutral phospholipids and cause vesicle aggregation due to hydrophobic interactions21. Moreover, clavanin A was also shown here to promote membrane leakage consistent with the results of Van Kan et al.6. Clavanin A showed the ability to permeabilize the membranes of vesicles even at modest concentrations. The release of encapsulated solutes induced by AMPs occurs either through transient channels formed due to membrane perforation22. However, the mechanism of action of clavanin is likely to be multifunctional8, consistent with our results presented in this study and the observation that sub-MIC levels of the peptide that do not cause membrane disruption, induce cell lamentation (consistent with cell division inhibition) in some cases (Fig.8). This eect on bacterial morphology may be due to interference of clavanin A with the cell division machinery, which will be the focus of future investigations. Interestingly, an analogous observation was made by Rosenberger et al.23as they reported that in vitro incubation of Salmonella enterica serovar Typhimurium with sub-lethal levels of mouse cathelicidin-related antimicrobial peptide (CRAMP) triggered cell lamentation with arrested septum formation; while Friedrich et al.24 showed similar observations with other peptides.

In the present study, we have shown the interaction of clavanin A with dierent lipids. But how exactly does clavanin A bind to lipids and what are the forces involved in these interactions? To clarify this issue, CD and NMR analyses were performed solving the structure of clavanin A in solution. In the absence of hydrophobic conditions, random conformations were observed for the peptide which changed in the amphipathic environment where adopted a helical structure. These data shed some light over previous CD experiments, where an increase in the molar ellipticity at 222 nm of clavanin A occur in the presence of small unilamellar vesicles of DOPC (Kan et al.7), and also -helical formation of clavanin A in membranes of DOPG:DOPC (1:9) (Kan et al.6). Other AMPs have been shown to undergo similar structural transitions. For example, the polar sh multifunctional peptide Pa-MAP125,26, a conformational transition was detected in which a random fold in water and a full -helical core was observed in SDS and TFE environments.

Structural calculations showed that the lowest energy structures obtained for clavanin A had a break helix in Gly13 (data not shown). Thus, to obtain better resolved structures 13C HSQC and 15N HMBC experiments were performed, through which it was possible to determine the and torsion angle of the main chain, using TALOS+. Aer dihedral angles addition, the peptide presented a curved structure centered in this residue (Gly13). Magainin2, is other antibiotic peptide that also contains 23 amino acid residues and curved helical structure centred on residues Phe12 and Gly1327.

Pukala and collaborators28 showed the importance of structural exibility in the activity of caerin 1.1, a potent broad-spectrum antibiotic peptide. In the membrane mimetic environment, this peptide adopts two -helices structures separated by a exible hinge region delimited by Pro15 and Pro19. The importance of the two Pro residues was assessed by replacement for Ala or Gly. The resulting structures indicate that the central angle of fold and its activity decreased signicantly aer replacing the Pro residue by Gly and more drastic reduction was observed when Pro was replaced by Ala. Thus, highlighting the need for structural curvature in its biological function.

SCIENTIFIC REPORTS

9

www.nature.com/scientificreports/

These relationships indicate that the exible hinge allows an optimal orientation in any N- and C-terminal region of the peptide when it interacts with the membranes of the bacterial cells29,30.

Van Kan et al. observed that substituting Gly13 and Gly18 or only Gly18 with Ala generated more potent peptides against Micrococcus avus. On the other hand, substituting only Gly13 led to slightly less potent peptides, while substitutions in Gly6 almost completely abolished peptide activity6. Therefore, Gly6 seems to be important for peptide exibility, Gly18 for lipid interaction and Gly13 for both. In summary, it seems that are a variety of conditions including the type of lipid, pH6, and concentration that could directly aect the structural and functional properties of clavanin A.

In this context, by NMR, our work demonstrate that clavanin A adopts a curved -helical conformation due to exibility provided by residues of Gly, specially due to Gly-13, endorsing the importance of structural curvature demonstrated by the above authors.

Clavanins show some structural and functional homology to magainin 1, a well-characterized antimicrobial peptide from the skin of Xenopuslaevis that also shows such structural transitions31. Despite the fact that magainins are lysine-rich and clavanins histidine-rich peptides, 15 amino acid residues of clavanin A (located between Leu5 and Phe19) show 12 similarities (6 identical residues and 6 conservative substitutions) with residues present between His2 and Phe16 in magainin 15. According to the results of molecular dynamics simulations this region of clavanin A with high similarity to magainin 1 was the major contributor to interaction of this peptide with lipids DOPC and DPPC (Fig.6).

In simulations, residues Gly13, Asn14, His17, Gly18 and Phe19 had close contacts with lipids. These data clarify the results obtained by van Kan et al.6 in experiments replacing Gly for Ala variants, as described above. Indeed, Franco9 proposed that multifunctional and promiscuous peptides are not only inuenced by their structures but also to micro-environmental conditions that could drive peptide folding and interaction with bacteria, and therefore its activity. This concept is consistent with the data reported here that suggest that the activity of clavanin A may depend on which lipid(s) with which it interacts, consistent with previous studies focusing on other linear AMPs such as magainin32 and LL-3733 that indicate that their activity can be governed by micro-environmental properties.

Finally, ow cell experiments revealed that treatment with 16g.mL1 of clavanin A (i.e., one quarter its MIC) eectively eradicated pre-established biolms formed by KPC-K. pneumoniae 1825971 and MRSA, and reduced the thickness of E. coli 0157 biolms. These results identify a novel, previously unrecognized role for clavanin A as an anti-biolm peptide. The amino acid sequence of clavanin A provides a good template for anti-biolm activity that will serve as the basis for subsequent structure-activity relationship studies aimed at producing synthetic variants with improved activity against biolms.

In conclusion, we have provided novel structural insights of a naturally occurring peptide isolated from the immune system of tunicates and have shown that this peptide exhibits a novel biological activity as an inhibitor of bacterial biolms, which may have implications in enabling the sea tunicate S. clava to combat biolm-related infections in its natural environment.

Experimental procedures. Bacterial strains. Strains used included clinical isolates Escherichia coli 0157, Klebsiella pneumonia carabapenemase producing K. pneumoniae 1825971 (KPC-Kp) and methicillin resistant Staphylococcus aureus (MRSA) #SAP0017, as well as reference strains E. coli ATCC8739, K. pneumoniae ATCC13883 and S. aureus ATCC2921.

Peptide synthesis and mass spectrometry analyses. Clavanin A was synthesized by Peptides 2.0 (Chantilly, VA, USA) using N-9-uorenylmethyloxycarbonyl (Fmoc) solid-phase synthesis, and puried by high-performance liquid chromatography (HPLC). The sequence and degree of purity (>95%) was conrmed by MALDI-ToF analyses26.

Bactericidal microdilution assays. Clavanin A MICs against E. coli ATCC8739 and S. aureus ATCC29213 were determined using a standardized microdilution method in polypropylene microtitre plates (TPP, Switzerland) according to CLSI guidelines, with an inoculum of 1 105 cells per mL34. MICs were determined as the lowest concentration tested that led to complete inhibition (100%) aer incubation at 37 C for 24h, in comparison with the untreated negative control group8,35.

Flow cell assays. Biolms were grown in BM2 minimal medium (62mM potassium phosphate buer, pH 7.0, 7mM [(NH4)2SO4, 2mM MgSO4, 10M FeSO4] containing 0.4% (wt/vol) glucose as a carbon source, for 72h, at 37 C in ow cell chambers with channel dimensions of 1 4 40 mm, as previously described36. For the treatment of pre-formed biolms, bacteria were allowed to develop structured 2-day-old biolms prior to treatment with clavanin A for the following 24 h. Biolm cells were then stained using the Live/Dead BacLight bacterial viability kit (Molecular Probes, Eugene, OR) and subsequently examined using a confocal laser scanning microscope (Olympus, Fluoview FV1000); three-dimensional reconstructions were generated using the Imaris soware package (Bitplane AG).

Preparation of cells for transmission electron microscopy (TEM). E. coli ATCC8739 and S. aureus ATCC29213 (1105) for TEM were grown and incubated with or without clavanin A (32 g.mL1 for E. coli and 64g.mL1 for S. aureus) as described above. Cell pellets were obtained from 10mL of each control untreated or treated cell suspension and xed for 1h with 2.5% glutaraldehyde in 0.1M cacodylate buer, pH 7.2. The samples were rinsed 2 10 minutes in 0.1 M cacodylate buer, pH 7.2 and then stained for 30 min with 1% osmium tetroxide/0.8% potassium ferricyanide/5mM CaCl2 in 0.1M cacodylate buer, and rinsed 2 10min in 0.1M cacodylate buer.

The samples were dehydrated in increasing concentrations of acetone 50%, 70%, 90% and 100% for 10min each.

SCIENTIFIC REPORTS

10

www.nature.com/scientificreports/

Aer the addition of 70% acetone, the pellets were carefully removed from the bottom of the tubes with the aid of a wooden toothpick, placed in a glass Petri dish containing 70% acetone, minced into smaller pieces and transferred to glass vials containing 90% acetone and 100% acetone. The samples were inltrated overnight with an Epon plus acetone solution (1:2); followed by 4 hr in pure Epon and embedding in fresh Epon to enable polymerization for 48h at 60C. Ultra-thin sections (~70nm) were obtained and stained for 30min with 2% aqueous uranyl acetate and for 5min with lead citrate. The samples were observed and the images were acquired in a JEOL JEM1011 transmission electron microscope operating at 80KV37.

Preparation of lipid vesicles. Large unilamellar vesicles (LUV) were obtained by the extrusion method37.

Phospholipids 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-rac-(1-glycerol) (POPG) were separately dissolved in chloroform under vigorous agitation at 23 C. Subsequently, the solutions of both pure and mixed lipid (POPC:POPG 8:2 molar) were prepared from organic stock solutions (1 mg mL1) of each lipid. Next, to remove solvents the organic solution was evaporated under reduced pressure (25min at 40 1C) and agitation at 120rpm. The remaining lm was le under reduced pressure overnight. The dried lipid lm was then hydrated by the addition of sodium HEPES buer (pH 7.4) containing 150 mMNaCl, leading to multilamellar vesicles (MLV) formation. Aer that, the MLV suspension was submitted to extrusion using Hamilton 1001RN syringes (Hamilton, USA) tted with polycarbonate membrane lters (= 10mm) with a pore diameter of 100nm (Nucleopore, United Kingdom) to obtain the LUV.

Permeability assays. Permeability was studied by detecting the clavanin A-induced leakage of 100 mM 5(6)-carboxyuorescein (5(6)-CF) from the lipid vesicles (100M). The non-encapsulated 5(6)-CF was removed using size-exclusion chromatography (Econo-Pac 10 DG, Bio-Rad, USA). Then fluorescence spectra were obtained using a Varian Cary Eclipse uorimeter (Mulgrave, Australia) in samples with 0.10mM lipid. The measurements were recorded for 25min at an excitation wavelength of 492nm and emission wavelength of 517nm38.

The maximal release of 5(6)-CF was determined aer permeabilization with 1%Triton X-100. The extent of leakage of 5(6)-CF was calculated by the equation: Leakage (%) = (F-F5)/(F100%-Ff), where F was the uorescence intensity measured 15 min aer peptide addition, and F5 and F100% were the uorescence intensities measured aer 5 min (to allow time for the system to stabilize) in the absence of peptide or aer Triton X-100 addition, respectively. The uorescence intensities were corrected for the dilution introduced by the addition of peptide and Triton X-100.

Evaluation of peptide membrane interaction by dynamic light scattering and zeta potential. The dimensions and zeta () potential of lipid (0.2 mM) vesicles were calculated using a ZetasizerNano ZS90 (Malvern, United Kingdom) with a laser He-Ne operating at a wavelength of 632.8 nm. The mean hydrodynamic diameter was determined by dynamic light scattering detected at an angle of 173. The CONTIN analysis of the scattering correlation function was used to calculate the diusion coefficient, and the Stokes-Einstein equation was used to convert the diusion coefficient into the hydrodynamic diameter. Zeta ()-potential (basically surface charge) was measured by an electrophoretic technique using the Smoluchowski approximation, U = /, where U was the mobility, was the viscosity and was the dielectric constant for pure water. All reported results corresponded to the average of three independent measurements of 10 runs each carried out at 25C. All samples were ltered through Millipore lters with a pore diameter of 0.45m39.

Isothermal titration calorimetry assays. To measure the binding affinity of clavanin A towards phospholipids 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and liposome of POPC mixture with 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-rac-(1-glycerol) (POPG) isothermal titration calorimetry (ITC) was performed using the iTC200 System (GE Healthcare, USA) employing non-reactive Hastelloy cells40. The lipid vesicle (POPC:POPG 8:2 molar) was prepared as described above in this section. The dried lipid vesicle was hydrated with sodium HEPES buer (pH 7.4) containing 50mM NaCl and degassed prior to titration at 25C. Isothermal interactions between clavanin A and only lipid and their vesicle were measured by titrating more than 20 injections using 40L of peptide solution (1mM) and lipid molecules in a sample cell at a concentration of 100M in a 200L suspension. Experiments were repeated three times.

Circular dichroism analyses. Circular dichroism (CD) spectra were obtained on a Jasco J-815 instrument (Jasco Co, Tokyo, Japan) using 1mm path quartz cells. Spectra were collected from 190 to 260nm at a speed of 50 nm. min1. CD spectra were obtained with 41mM peptide in the presence of dierent concentrations of SDS (sodium dodecyl sulphate) and TFE (2,2,2-triuorethanol). The eect of variation of pH was analyzed using 5mM sodium acetate buer at pH 3, 4 and 5, and 5mMTris-HCl buer at pHs 7 and 9. Spectra were corrected by subtracting the appropriate blank control and converting to mean residue molar ellipticity. Analyses of percentage of -helicity were performed according to the method by Chen et al.41.

Nuclear Magnetic Resonance. The NMR sample was prepared by dissolving clavanin A to a nal concentration of 2 mM in 500 L of 35% (v/v) TFE-d3 solution, 10% of D2O and pH 4.3. Spectra were acquired at 25 C on a

Bruker Avance III 500 spectrometer operating at 11.75 T. All acquired NMR spectra were referenced using TMSP-2,2,3,3-d4 (sodium-3-trimethylsilylpropionate). 1H1H Total Correlation Spectroscopy (TOCSY) spectra were recorded with 176 transients of 4096 data points, 256 t1 increments and a spinlock mixing time of 80 ms, using the dipsi2esgpph42 pulse sequence. The 1H1H Nuclear Overhauser Eect Spectroscopy (NOESY) was recorded with 64 transients of 4096 data points, 256 t1 increments, mixing time of 250 ms, using the noesygpphw543 pulse sequence. 1H15N So-fast Heteronuclear Multiple Quantum Coherence (sf-HMQC; sfhmqcf3gpph pulse sequence)44 and 1H13C Heteronuclear Single Quantum Coherence spectra (HSQC; hsqcetgp pulse sequence)

SCIENTIFIC REPORTS

11

www.nature.com/scientificreports/

were recorded with 168 transients of 1024 data points and 256 t1 increments. 1H1H TOCSY, NOESY and 1H15N sf-HMQC spectra were acquired using the States-TPPI mode45 and 1H13C HSQC spectra were acquired with the Echo-antiecho46 mode of quadrature detection. All NMR data were processed using nmrPIPE47 and nmrVIEW48 soware. The complete assignment of the backbone and side-chain 1H resonances, to accessible protons, of clavanin A was performed using standard sequential assignment procedures, according to the methodology developed by Wthrich49. The chemical shis were deposited in the BioMagResBank (www.bmrb.wisc.edu) under accession number 25262. Clavanin structure was also deposited in Protein Databank (Code: 2MVE) and validation was described in Supplementary Material 1.

Molecular Modelling. The volumes of NOE correlations from NOESY were converted into distance using XPLOR-NIH (version 2.28) soware. The restraints were classied into short (2.8 ), medium (3.4 ) and long (5.0 ) distances and analyzed using QUEEN (Quantitative Evaluation of Experimental NMR Restraints) so-ware50. Dihedral angles were predicted using the soware TALOS+, where the chemical shi constraints of 13C,

15Ne 1H were used as input12. Structures were calculated using the XPLOR-NIH and the simulated annealing (SA) algorithm51. The structure calculation started with an extended model, with 18,000 steps at high temperature, and 9,000 steps of cooling. The ensemble of 20 lower energy structures (from a total of 200 calculated structures) was then submitted to a water renement protocol. The ensemble of 10 lowest energy structures was chosen to represent the peptide solution 3D structure. The generated structures were validated using Procheck online soware (mordred.bioc.cam.ac.uk/~rapper/rampage.php). The quality of the family of structures was analyzed by standard measures used to quantify dierences between three-dimensional structures, namely the root-mean-square deviation (RMSD), provided by the MOLMOL program52. The NMR structures were submitted to PDB and were assigned with PDB code 2MVE.

Molecular docking and dynamics. The complexes between clavanin A and lipids DOPC or DPPC were constructed using the Hex 6.153. The NMR structure was used for docking; and the coordinates of DOPC and DPPC were obtained from the model CHARMM-GUI server54. Docking experiments were performed with consideration to the shape and electrostatics of each molecule without any docking post processing. The resulting complexes were clustered, using a root mean square (RMS) cut-o of 3 . The cluster with highest affinity was selected as the preferable binding mode. For molecular dynamics (MD), the structures of DOPC and DPPC were parameterized using the PRODRG server55. The GROMACS package (version 4.5)56 was used for performing the MD simulations. The structures in complex with, or without, lipid were immersed in a cubic water box, with a distance of 0.7nm from the edges of the box. Water molecules were represented using the single point charge water model57. Chlorine ions were added to the system in order to neutralize the systems charge. Fiy thousand steps of steepest descent were performed to minimize the system. Next, the molecular dynamics integrator was used for pressure and temperature normalization (100ps each) by using the velocity rescaling thermostat (NVT ensemble) and the Parrinello-Rahman barostat (NPT ensemble), respectively. Then, the system was simulated for 50 ns with minimized energy and normalized pressure (1 bar) and temperature (300K). The geometry of water molecules was constrained by using the SETTLE algorithm58 and all atom bond lengths were linked by using the LINCS algorithm59. The electrostatic corrections were made according to the Particle Mesh Ewald algorithm60, with a cut-o radius of 1.4 nm; the same cut-o radius was also used for van der Waals interactions. The MD simulations were analyzed by means of root mean square deviation (RMSD) and standardize secondary structure assignment (DSSP).

References

1. Sommer, F. et al. Blood system formation in the urochordate Ciona intestinalis requires the variable receptor vCRL1. Mol Biol Evol

29, 30813093 (2012).

2. Colucci-Guyon, E., Tinevez, J. Y., Renshaw, S. A. & Herbomel, P. Strategies of professional phagocytes in vivo: unlike macrophages, neutrophils engulf only surface-associated microbes. J Cell Sci 124, 30533059 (2011).

3. Lehrer, R. I., Andrew Tincu, J., Taylor, S. W., Menzel, L. P. & Waring, A. J. Natural Peptide antibiotics from tunicates: structures, functions and potential uses. Integr Comp Biol 43, 313322 (2003).

4. Taylor, S. W., Craig, A. G., Fischer, W. H., Park, M. & Lehrer, R. I. Styelin D, an extensively modied antimicrobial peptide from ascidian hemocytes. J Biol Chem 275, 3841738426 (2000).

5. Lee, I. H. et al. Clavanins, alpha-helical antimicrobial peptides from tunicate hemocytes. FEBS Lett 400, 158162 (1997).6. van Kan, E. J., van der Bent, A., Demel, R. A. & de Kruij, B. Membrane activity of the peptide antibiotic clavanin and the importance of its glycine residues. Biochemistry 40, 63986405 (2001).

7. van Kan, E. J., Demel, R. A., van der Bent, A. & de Kruij, B. The role of the abundant phenylalanines in the mode of action of the antimicrobial peptide clavanin. Biochim Biophys Acta 1615, 8492 (2003).

8. Silva, O. N. et al. Clavanin A improves outcome of complications from dierent bacterial infections. Antimicrob Agents Chemother 59, 16201626 (2015).

9. Franco, O. L. Peptide promiscuity: an evolutionary concept for plant defense. FEBS Lett 585, 9951000 (2011).10. Hancock, R. E. & Rozek, A. Role of membranes in the activities of antimicrobial cationic peptides. FEMS Microbiol Lett 206, 143149 (2002).

11. Wu, M., Maier, E., Benz, R. & Hancock, R. E. Mechanism of interaction of dierent classes of cationic antimicrobial peptides with planar bilayers and with the cytoplasmic membrane of Escherichia coli. Biochemistry 38, 72357242 (1999).

12. Sato, H. & Feix, J. B. Peptide-membrane interactions and mechanisms of membrane destruction by amphipathic alpha-helical antimicrobial peptides. Biochim Biophys Acta 1758, 12451256 (2006).

13. Chen, X. & Chen, Z. SFG studies on interactions between antimicrobial peptides and supported lipid bilayers. Biochim Biophys Acta 1758, 12571273 (2006).

14. Clore, G. M., Martin, S. R. & Gronenborn, A. M. Solution structure of human growth hormone releasing factor. Combined use of circular dichroism and nuclear magnetic resonance spectroscopy. J Mol Biol 191, 553561 (1986).

15. Bach, A. C., 2nd, Selsted, M. E. & Pardi, A. Two-dimensional NMR studies of the antimicrobial peptide NP-5. Biochemistry 26, 43894397 (1987).

SCIENTIFIC REPORTS

12

www.nature.com/scientificreports/

16. Wishart, D. S. & Nip, A. M. Protein chemical shi analysis: a practical guide. Biochem Cell Biol 76, 153163 (1998).17. Ramakrishnan, C. & Ramachandran, G. N. Stereochemical criteria for polypeptide and protein chain conformations. II. Allowed conformations for a pair of peptide units. Biophys J 5, 909933 (1965).

18. Saude, A. C. et al. Clavanin bacterial sepsis control using a novel methacrylate nanocarrier. Int J Nanomedicine 9, 50555069 (2014).19. Matos, P. M., Goncalves, S. & Santos, N. C. Interaction of peptides with biomembranes assessed by potential-sensitive uorescent probes. J Pept Sci 14, 407415 (2008).

20. Abrunhosa, F. et al. Interaction and lipid-induced conformation of two cecropin-melittin hybrid peptides depend on peptide and membrane composition. J Phys Chem B 109, 1731117319 (2005).

21. Epand, R. F. et al. Analogs of the antimicrobial peptide trichogin having opposite membrane properties. Eur J Biochem 268, 703712 (2001).

22. Caaveiro, J. M., Molina, A., Rodriguez-Palenzuela, P., Goni, F. M. & Gonzalez-Manas, J. M. Interaction of wheat alpha-thionin with large unilamellar vesicles. Protein Sci 7, 25672577 (1998).

23. Rosenberger, C. M., Gallo, R. L. & Finlay, B. B. Interplay between antibacterial eectors: a macrophage antimicrobial peptide impairs intracellular Salmonella replication. Proc Natl Acad Sci USA 101, 24222427 (2004).

24. Friedrich, C. L., Moyles, D., Beveridge, T. J. & Hancock, R. E. Antibacterial action of structurally diverse cationic peptides on gram-positive bacteria. Antimicrob Agents Chemother 44, 20862092 (2000).

25. Nascimento, J. M. et al. Elucidation of mechanisms of interaction of a multifunctional peptide Pa-MAP with lipid membranes. Biochim Biophys Acta 1838, 28992909 (2014).

26. Migliolo, L. et al. Structural and functional characterization of a multifunctional alanine-rich peptide analogue from Pleuronectes americanus. PLoS One 7, e47047 (2012).

27. Gesell, J., Zaslo, M. & Opella, S. J. Two-dimensional 1H NMR experiments show that the 23-residue magainin antibiotic peptide is an alpha-helix in dodecylphosphocholine micelles, sodium dodecylsulfate micelles, and triuoroethanol/water solution. J Biomol NMR 9, 127135 (1997).

28. Pukala, T. L., Brinkworth, C. S., Carver, J. A. & Bowie, J. H. Investigating the importance of the exible hinge in caerin 1.1: solution structures and activity of two synthetically modied caerin peptides. Biochemistry 43, 937944 (2004).

29. Boland, M. P. & Separovic, F. Membrane interactions of antimicrobial peptides from Australian tree frogs. Biochim Biophys Acta 1758, 11781183 (2006).

30. Haney, E. F., Hunter, H. N., Matsuzaki, K. & Vogel, H. J. Solution NMR studies of amphibian antimicrobial peptides: linking structure to function? Biochim Biophys Acta 1788, 16391655 (2009).

31. Zaslo, M. Magainins, a class of antimicrobial peptides from Xenopus skin: isolation, characterization of two active forms, and partial cDNA sequence of a precursor. Proc Natl Acad Sci USA 84, 54495453 (1987).

32. Afonin, S. et al. (19)F NMR screening of unrelated antimicrobial peptides shows that membrane interactions are largely governed by lipids. Biochim Biophys Acta 1838, 22602268 (2014).

33. Wang, G., Mishra, B., Epand, R. F. & Epand, R. M. High-quality 3D structures shine light on antibacterial, anti-biolm and antiviral activities of human cathelicidin LL-37 and its fragments. Biochim Biophys Acta 1838, 21602172 (2014).

34. Wiegand, I., Hilpert, K. & Hancock, R. E. Agar and broth dilution methods to determine the minimal inhibitory concentration (MIC) of antimicrobial substances. Nat Protoc 3, 163175 (2008).

35. Fensterseifer, I. C. et al. Eects of cyclotides against cutaneous infections caused by Staphylococcus aureus. Peptides 63, 3842 (2015).

36. de la Fuente-Nunez, C. et al. Inhibition of bacterial biolm formation and swarming motility by a small synthetic cationic peptide. Antimicrob Agents Chemother 56, 26962704 (2012).

37. Hartmann, M. et al. Damage of the bacterial cell envelope by antimicrobial peptides gramicidin S and PGLa as revealed by transmission and scanning electron microscopy. Antimicrob Agents Chemother 54, 31323142 (2010).

38. Domingues, M. M. et al. Biophysical characterization of polymyxin B interaction with LPS aggregates and membrane model systems. Biopolymers 98, 338344 (2012).

39. Domingues, M. M., Castanho, M. A. & Santos, N. C. rBPI(21) promotes lipopolysaccharide aggregation and exerts its antimicrobial eects by (hemi)fusion of PG-containing membranes. PLoS One 4, e8385 (2009).

40. Mandal, S. M. et al. Controlling resistant bacteria with a novel class of beta-lactamase inhibitor peptides: from rational design to in vivo analyses. Sci Rep 4, 6015 (2014).

41. Chen, Y. H., Yang, J. T. & Martinez, H. M. Determination of the secondary structures of proteins by circular dichroism and optical rotatory dispersion. Biochemistry 11, 41204131 (1972).

42. Nguyen, B. D., Meng, X., Donovan, K. J. & Shaka, A. J. SOGGY: solvent-optimized double gradient spectroscopy for water suppression. A comparison with some existing techniques. J Magn Reson 184, 263274 (2007).

43. Grosse, R., Malur, J. & Repke, K. R. [Determination of secondary structures of proteins with infrared spectra]. Acta Biol Med Ger 28, K1117 (1972).

44. Schanda, P., Kupce, E. & Brutscher, B. SOFAST-HMQC experiments for recording two-dimensional heteronuclear correlation spectra of proteins within a few seconds. J Biomol NMR 33, 199211 (2005).

45. Hall, A. S., Collins, A. G., Bryant, D. J., Young, I. R. & Bydder, G. M. Use of solvent suppression technique to enhance changes due to susceptibility variations in magnetic resonance imaging. Magn Reson Med 9, 411418 (1989).

46. Habazettl, J., Schleicher, M., Otlewski, J. & Holak, T. A. Homonuclear three-dimensional NOE-NOE nuclear magnetic resonance/ spectra for structure determination of proteins in solution. J Mol Biol 228, 156169 (1992).

47. Delaglio, F. et al. NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6, 277293 (1995).48. Johnson, B. A. & Blevins, R. A. NMR View: A computer program for the visualization and analysis of NMR data. J Biomol NMR 4, 603614 (1994).

49. Wuthrich, K. The way to NMR structures of proteins. Nat Struct Biol 8, 923925 (2001).50. Nabuurs, S. B. et al. Quantitative evaluation of experimental NMR restraints. J Am Chem Soc 125, 1202612034 (2003).51. Schwieters, C. D. & Clore, G. M. Using small angle solution scattering data in Xplor-NIH structure calculations. Prog Nucl Magn Reson Spectrosc 80, 111 (2014).

52. Koradi, R., Billeter, M. & Wuthrich, K. MOLMOL: a program for display and analysis of macromolecular structures. J Mol Graph 14, 5155, 2932 (1996).

53. Ritchie, D. W. & Venkatraman, V. Ultra-fast FFT protein docking on graphics processors. Bioinformatics 26, 23982405 (2010).54. Jo, S., Lim, J. B., Klauda, J. B. & Im, W. CHARMM-GUI Membrane Builder for mixed bilayers and its application to yeast membranes. Biophys J 97, 5058 (2009).

55. Schuttelkopf, A. W. & van Aalten, D. M. PRODRG: a tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr D Biol Crystallogr 60, 13551363 (2004).

56. Hess, B., Kutzner, C., van der Spoel, D. & Lindahl, E. GROMACS 4: algorithms for highly efficient, load-balanced, and scalable molecular simulation. J Chem Theory Comput 4(3), 435447 (2008)

57. Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F. & Hermans, J. Interaction models for water in relation to protein hydration. Intermolecular Forces pp. 331342 Key: citeulike:2139448 1, 331342 (1981).

58. Miyamoto, S. & Kollman, P. A. Settle: An analytical version of the SHAKE and RATTLE algorithm for rigid water models. Journal of Computational Chemistry 13, 952962 (1992).

SCIENTIFIC REPORTS

13

www.nature.com/scientificreports/

59. Hess, B., Bekker, H., Berendsen, H. J. C. & Fraaije, J. G. E. M. LINCS: A linear constraint solver for molecular simulations. Inorg Chem 18, 14631472 (1997).

60. Cerutti, D. S., Duke, R. E., Darden, T. A. & Lybrand, T. P. Staggered Mesh Ewald: An extension of the Smooth Particle-Mesh Ewald method adding great versatility. J Chem Theory Comput 5, 2322 (2009).

Acknowledgements

We would like to thank CNPq, CAPES (Cincias sem Fronteiras), FAPDF and FUNDECT. D.G. acknowledges Fundao para a Cincia e a Tecnologia - Ministrio da Educao e Cincia (FCT-MEC, Portugal) for fellowship SFRH/BPD/73500/2010 and A.S.V. for funding within the FCT Investigator Programme (IF/00803/2012).

Author Contributions

O.N.S., C.D.L.F.-N., S.M.R., S.M.M., M.A.R.B.C., R.E.W.H. and O.L.F. designed the experiments; A.L.O., E.S.F.A. and L.M.L. performed nuclear magnetic resonance; A.L.O., E.S.F.A., L.M.L. and W.F.P. performed in silico analyses; D.G., A.S.V. and M.A.R.B.C. performed permeability assays; C.A.S.A. and J.M.N. performed dynamic light scattering and zeta potential analyses; O.N.S., I.C.M.F., C.D.L.F.-N., S.M.R. and O.L.F. performed antibacterial analyses; S.M.M. and S.K. performed calorimetry analyses; J.R.C. performed electron microscopy analyses; O.N.S., E.S.F.A., C.D.L.F.-N., S.M.R., A.L.O., L.M.L. and O.L.F. wrote the manuscript.

Additional Information

Competing nancial interests: The authors declare no competing nancial interests.

How to cite this article: Silva, O. N. et al. Structural Studies of a Lipid-Binding Peptide from Tunicate Hemocytes with Anti-Biolm Activity. Sci. Rep. 6, 27128; doi: 10.1038/srep27128 (2016).

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

SCIENTIFIC REPORTS

14