Research Article Received: 5 March 2014
Revised: 23 May 2014
Accepted: 13 June 2014
Published online in Wiley Online Library: 17 July 2014
(wileyonlinelibrary.com) DOI 10.1002/psc.2673
Micelle bound structure and DNA interaction of brevinin-2-related peptide, an antimicrobial peptide derived from frog skin Susmita Bandyopadhyay,a Boon Yee Ng,a Charmaine Chong,a Ming Zhen Lim,a Sonia Kiran Gill,a Ke Hui Lee,a J Sivaramanb and Chiradip Chatterjeea* Brevinin-2-related peptide (BR-II), a novel antimicrobial peptide isolated from the skin of frog, Rana septentrionalis, shows a broad spectrum of antimicrobial activity with low haemolytic activity. It has also been shown to have antiviral activity, specifically to protect cells from infection by HIV-1. To understand the active conformation of the BR-II peptide in membranes, we have investigated the interaction of BR-II with the prokaryotic and eukaryotic membrane-mimetic micelles such as sodium dodecylsulfate (SDS) and dodecylphosphocholine (DPC), respectively. The interactions were studied using fluorescence and circular dichroism (CD) spectroscopy. Fluorescence experiments revealed that the N-terminus tryptophan residue of BR-II interacts with the hydrophobic core of the membrane mimicking micelles. The CD results suggest that interactions with membrane-mimetic micelles induce an α-helix conformation in BR-II. We have also determined the solution structures of BR-II in DPC and SDS micelles using NMR spectroscopy. The structural comparison of BR-II in the presence of SDS and DPC micelles showed significant conformational changes in the residues connecting the N-terminus and C-terminus helices. The ability of BR-II to bind DNA was elucidated by agarose gel retardation and fluorescence experiments. The structural differences of BR-II in zwitterionic versus anionic membrane mimics and the DNA binding ability of BR-II collectively contribute to the general understanding of the pharmacological specificity of this peptide towards prokaryotic and eukaryotic membranes and provide insights into its overall antimicrobial mechanism. Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd. Additional supporting information may be found in the online version of this article at the publisher’s web site. Keywords: antimicrobial peptide; brevinin-2-related peptide; sodium dodecylsulfate; n-dodecylphosphocholine; two-dimensional NMR; pUC19 DNA
Introduction
J. Pept. Sci. 2014; 20: 811–821
* Correspondence to: Chiradip Chatterjee, School of Applied Science, Republic Polytechnic, Singapore. E-mail: [email protected] a School of Applied Science, Republic Polytechnic, Singapore b Department of Biological Sciences, National University of Singapore, Singapore
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
811
The development of microbial resistance against antibiotics has become a serious problem because of the inappropriate or the constant use of antibiotics and the mutation of the microbes [1]. The search for new antibiotics to treat infections caused by multidrug-resistant microorganisms has thus become urgent and has stimulated interest in the development of antimicrobial peptides (AMPs) as human therapeutics [2,3]. AMPs are present in all kingdoms of life as a first line of defense against deadly pathogens. They demonstrate high potential as novel therapeutic agents as most of the AMPs eradicate both Gram-positive and Gram-negative bacteria [4–6]. In addition, several AMPs have been found to have anti-HIV properties, which have stimulated recent research in this area [7–12]. Amphibians that live in both land and water force themselves to adapt and survive in a variety of conditions. Amphibian skin secretes potent antimicrobial agents to protect them against infections by molds, bacteria, and protozoa and has proven to be valuable source of AMPs [13–16]. A 21-amino-acid peptide, named Brevinin-2related peptide (BR-II), isolated from the skin secretions of the mink frog (Rana septentrionalis) shows broad-spectrum antimicrobial
activity and low hemolytic activity [17]. Recently, it has also been reported that BR-II can protect cells from HIV-1 infection [18]. This peptide shows some sequence similarity to the N-terminus region of other peptides of the brevinin-2 family isolated from several species of Eurasian frog skin [17]. Despite lacking a C-terminus cyclic heptapeptide region (Cys-Lys-Xaa4-Cys-, a common feature of all antimicrobial brevinin-2 peptides known to date), BR-II shows relatively high potency against Gram-positive and Gram-negative bacteria as well as opportunistic yeast pathogens. This demonstrates that the cyclic heptapeptide region in brevinin-2 is not necessary for antimicrobial activity [17]. Thus, BR-II represents a novel therapeutic agent, and understanding its mode of antimicrobial and anti-HIV activities is essential to the development of the compound for medical use.
BANDYOPADHYAY ET AL. The majority of AMPs are believed to act by disrupting bacterial cell membranes; therefore, an important aspect of determining the mode of action employed by an AMP is to study its ability to interact with membranes and the structural folding of AMP in membranes [19–21]. Because of the complexity of the bacterial membranes, such peptide structures were mainly determined by solution NMR using membrane-mimetic models such as organic solvents or detergent micelles. Both deuterated sodium dodecylsulfate (SDS) and n-dodecylphosphocholine (DPC) micelles are widely employed in determining the solution structure of AMP bound to lipid micelles [6,4,5,22]. This can provide valuable insight on a number of physicochemical properties of AMP such as the distribution of the charges on the surface of the peptide, amphipathicity and structural folding, which contribute to the mechanism by which an AMP executes its function. In this paper, we report the interaction of BR-II with SDS and DPC studied by fluorescence and circular dichroism (CD) spectroscopy. In addition, the three-dimensional (3D) solution structures of BR-II bound to DPC and SDS micelles have been determined by NMR spectroscopy for the first time. The comparison between the structures of BR-II in SDS and DPC micelles will lead to a better understanding of the different bioactive conformations of BR-II peptide in negative and zwitterionic membrane environments, respectively. The studies on the interactions between AMPs and DNA are of interest to determine if the interaction of peptide with intercellular target such as DNA takes place after penetration of the peptide to the inner structure of the cell by peptide– membrane interaction and thus, plays a part in overall antimicrobial activity of the peptide [23]. Here, we also report the DNA binding ability of BR-II using agarose gel retardation assay and fluorescence experiments.
Materials and Methods Materials Brevinin-2 peptide (98% pure, GIWDTIKSMGKVFAGKTLQNL) was purchased from Nanyang Technological University Biosciences Laboratory (Singapore) and used without further purification. SDS and DPC were obtained from Avanti Polar Lipids (USA) in the highest available purity. Deuterated SDS-d25 and DPC-d38 were purchased from Cambridge Isotope Laboratories (CIL Inc, USA). pUC19 DNA was purchased from Invitrogen (Singapore) and used without further purification. Analytical grade acrylamide and ethidium bromide were obtained from Sigma Aldrich (USA). All other reagents and solvents were of reagent grade and used without further purification unless otherwise specified. Fluorescence Spectroscopy
812
Intrinsic fluorescence measurements were recorded using Perkin Elmer LS50B fluorescence spectrophotometer. Peptide fluorescence was measured by exciting peptide samples at 280 nm and scanning emission between 300 and 460 nm using a slit of 3 nm. A peptide concentration of 2 μM was used for all the fluorescence experiments. After addition of the sample, the system was equilibrated for 5 min before fluorescence was measured. All measurements were made in a semi-micro quartz cuvette (path length 1 cm) at room temperature (~25 °C). All samples were prepared in Millipore water (pH 4.3, adjusted with HCl) for recording the fluorescence spectra in SDS (50 mM) and DPC
wileyonlinelibrary.com/journal/jpepsci
(50 mM) membrane-mimetic micellar solutions, with the baseline corrected using the corresponding control solutions. The changes in intensity and blue shifts in the emission spectra of BR-II in the presence of different model membranes were monitored. Acrylamide quenching was performed on free and micelles bound peptide mixtures to calculate Stern–Volmer constants (KSV) following published protocols [24–26]. The KSV for each combination of the peptide and micelles was calculated using the following equation: F 0 =F c ¼ 1 þ K SV ½Q where F0 is the initial fluorescence of the peptide and Fc is the fluorescence intensity following the addition of soluble quencher, Q. Peptide-DNA fluorescence was measured following the same protocol described earlier. A fixed amount (8.8 μgm) of peptide was treated with increasing amounts of pUC19 DNA (DNA amount: 0, 0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75 μgm). Fluorescence spectra were measured from 320 to 500 nm. After addition of each sample, the system was equilibrated for 5 min before fluorescence was measured. All measurements were made in a semi-micro quartz cuvette at room temperature (~25 °C). All samples were prepared in Millipore water (pH 4.3, adjusted with HCl) for recording the fluorescence spectra with the baseline corrected using the corresponding control solutions. All fluorescence experiments were carried out in triplicate to check for the reproducibility, and they were found consistent. CD Spectroscopy The secondary structures of BR-II in different micelle solutions were recorded in a Jasco J810 spectropolarimeter (Jasco Corp., Tokyo, Japan) at room temperature (~25 °C) using a 0.1 cm path length cuvette and with the baseline corrected using control solutions. All samples were prepared in millipore water (pH 4.3, adjusted with HCl) for recording the CD spectra in SDS (50 mM) and DPC (50 mM) micelles. The peptide concentration was maintained at 20 μM in all cases. NMR Spectroscopy NMR measurements were performed on a Bruker Avance DRX-800 operating at 800 MHz with cryoprobe and Bruker Avance DRX-400 operating at 400 MHz. The sample temperature was maintained at 298 K. The samples were prepared with 2 mM peptide, 100 mM SDS-d25 and another sample with 100 mM DPC-d38 (CIL, Cambridge, MA, USA) and 10% D2O in millipore water (pH 4.3 adjusted using diluted HCL). Two-dimensional (2D) TOCSY spectra (spinlock time of 100 ms) were acquired using the MLEV-17 pulse sequence [27]. 1 H 2D NOESY spectra were acquired using a 150 ms mixing time [28]. Water suppression was achieved using the WATERGATE technique [29]. The NMR spectra were collected with 4096 and 512 data points in the direct and indirect dimensions, respectively, with 16 transients. The raw data were processed using Topspin 2.0 (Bruker Inc), and resonance assignments were made using the Sparky [30]. Structure Calculation The 3D structures of BR-II bound to SDS and DPC micelles were calculated on the basis NOE-derived distance constraints from 2D NOESY spectra. Assignment of the proton spectra of BR-II in the presence of membrane-mimetic solvents (SDS and DPC)
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2014; 20: 811–821
MEMBRANE INTERACTION AND DNA BINDING OF ANTIMICROBIAL BREVININ-2-RELATED PEPTIDE was accomplished using the technique of sequence-specific resonance assignments developed by Wuthrich [31]. On the basis of cross-peak intensities, NOEs were qualitatively categorized to strong, medium, and weak and then translated to upper bound distance limits to 2.5, 3.5, and 5.5 Å, respectively. The lower bounds for all NOE restraints were set to 2.0 Å. Initial structures were generated by backbone torsion angle restraints and distance geometry using CYANA-2.1 [31,32]. For structure calculations, the dihedral angle restraints were defined for all the backbone angles of the peptide, except for Gly. The dihedral angle restraints were referenced on the basis of CD structure, chemical shift index (CSI) values, and NOE patterns. A total of 100 structures were calculated, and 20 lowest-energy structures with no distance violations greater than 0.25 Å or dihedral angle violations larger than 2° were selected. All the structures were analysed using Pymol (http://pymol.sourceforge.net) and MolMol [33] and PROCHECK-NMR program [34]. Translational Diffusion Experiments Diffusion-ordered spectroscopy (DOSY) experiments were carried out using a 400 MHz Bruker Avance DRX-400 spectrometer with a quad (1H, 13C, 19F, and 31P) probe and z-gradient. The gradient coil constant was found to be 50 G/cm at 100% gradient strength by carrying out the BPLED pulse sequence with water suppression using pre-saturation. The diffusion time Δ was 120 ms, and the gradient duration δ was 2 ms. The gradient strength was increased from 2% to 95% of Gmax in 32 scans. The individual diffusion coefficients were calculated from the DOSY spectra by decay of the peak intensities as a function of gradient strength and evaluated using the Bruker Topspin 2.0 T1/T2 analysis package Topspin 2.0 package (Bruker Inc) using aliphatic Leu-Hδ1/ Hδ2 chemical shifts (about 0.78 ppm) for the peptide. Diffusion constants were derived from the following equation [35]: I ¼ I0 eDγ
g δ ðΔ3δÞ
2 2 2
(1)
where I is the observed intensity, I0 the unattenuated signal intensity, D the diffusion coefficient, γ the gyromagnetic ratio of 1H, g the gradient strength, δ the gradient length, and Δ is the diffusion time. The molar fraction of peptide in micelles (FL) is calculated with Eqn (2) [36–38] DS ¼ DL F L þ DF ð1 F L Þ
Results Interaction of BR-II with Micelles Tryptophan fluorescence The change in fluorescence intensity and the shift in maximum emission wavelength of the tryptophan (Trp) fluorophore of BR-II in the presence of SDS and DPC micelles was examined (Figure 1). The maximum emission wavelength of Trp fluorophore of BR-II in aqueous solution was 363 nm. A maximum blue shift of approximately 5 and 22 nm was observed in the presence of SDS and DPC micelles, respectively (Figure 1). In addition, a change in the intensity of the florescence indicated that the environment of the fluorophore in the presence of micelles had changed compared with the peptide in buffer alone (Figure 1 and its inset). This suggests that the Trp3, which is located at the N-terminus of the BR-II, interacts with the micelles and forms hydrophobic clusters similar to other AMPs [40]. Fluorescence quenching The Stern–Volmer constants (KSV), which determines the extent to which the Trp3 residue of the peptide is buried into the hydrophobic core of the micelles, was calculated from the acrylamide quenching experiment. When the side chain of Trp3 is buried into the micelle phase and becomes less exposed to the surrounding solvent, the fluorescence contribution of Trp3 is less susceptible to the effects of the neutral soluble quencher acrylamide. The lower the KSV value, the greater is the degree of interaction of the Trp fluorophore with the hydrophobic core of the micelles. The highest optimum concentration of lipid was determined by tryptophan fluorescence (Figure 1), in which no further blue shift in fluorescence emission spectra was observed. This optimum concentration of lipid was used for all fluorescence quenching experiments to ensure the ideal complex formation between the peptide and the micelles [25]. The KSV for BR-II in aqueous solution was found to be ~105 M1. The KSV values decreased significantly in the presence of SDS and DPC micelles (Figure 2) indicating that the Trp side chain interacts with the micelles in these two membrane-mimetic environments. This supports the blue shifts observed for the BR-II peptide in the presence of these micelles environment (Figure 1).
(2)
where DS is the measured diffusion coefficients of the peptide in the presence of micelles and DL and DF are the diffusion coefficients of micelles and free peptides in aqueous solution, respectively. The experimental values of diffusion coefficient of micelles (DL) for SDS [21] and DPC are obtained from the literature [39]. DNA Gel Retardation Assay
J. Pept. Sci. 2014; 20: 811–821
Figure 1. Tryptophan (Trp) fluorescence emission spectra of BR-II (2 μM) (A) in buffer, (B) in 50 mM SDS, and (C) in 50 mM DPC. Inset: Bar diagram showing maximum blue shift of fluorophore in presence of different micellar environments (SDS, DPC) compared with BR-II in buffer.
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci
813
The pUC19 DNA (Invitrogen) was mixed with increasing amounts of peptides in 15 μl of buffer (10 mM Tris (pH 7.6)/50 mM NaCl/ 1 mM EDTA). The mixtures were incubated at room temperature for 5 min and then subjected to electrophoresis on a 1% agarose gel in the 1× TAE (40 mM Tris, 20 mM acetic acid, and 1 mM EDTA) buffer solution.
BANDYOPADHYAY ET AL.
Figure 2. Stern–Volmer constants calculated from acrylamide quenching experiments for free BR-II in buffer and in the presence of various micellar environments of SDS and DPC, respectively.
The complete chemical shift assignments of BR-II in the presence of SDS and DPC micelles at 298 K is given in Table 1 in the Supporting information. We assigned ~100% NOE peaks in the 2D NOESY spectrum. The representative NOE cross-peak pattern between the backbone amides and R-protons for BR-II in SDS and DPC is presented in Figure 4A and B, respectively. Nearly all proton resonances including side chains were unambiguously assigned. An ensemble of 100 structures was calculated using 325 (85 intra-residue and 240 inter-residue) NOEs for SDS micelles and 312 (74 intra-residue and 238 inter-residue) for DPC micelles from the analysis of the 2D 1H–1H NOESY spectrum using Cyana-2.1. The best 20 lowest energy structures were retained for analysis. Several long distance NOEs are observed for BR-II in SDS micelles including NOEs between Ser8 (Hα)–Phe13 (HN), Asn20 (Hα)–Ile2 (HN), and Thr5 (Hα)–Asn20 (HN) (Figures 4A and S2).
CD spectroscopy CD experiments were performed to study the effect of different membrane environments on the secondary structure of BR-II. In millipore water (pH 4.3, adjusted with HCl), BR-II exhibits spectral characteristics of an unstructured peptide with a broad minimum at 198 nm (Figure 3). Double minimum bands at 225 and 208 nm and a positive CD band at 195 nm in the spectra (Figure 3 and Figure S1 in the Supporting information) were obtained for BR-II in the presence of SDS and DPC micelles, indicating the presence of a predominantly α-helical structure, compared with the random coil structure observed for BR-II in buffer solution. These data clearly indicate that BR-II interacts with all micelles and undergoes significant conformational changes from a random coil state to α-helical state. NMR spectroscopy Two-dimensional NMR spectra of BR-II were obtained at 298 K and pH 4.3 in aqueous solution and in the presence of SDS and DPC micelles. The proton resonance assignments were made using the procedures of Wuthrich [31]. To assign the backbone, side chain resonances and sequential assignments, a combination of 2D [1H, 1H] TOCSY and 2D NOESY spectra, were used. The 1D proton and 2D TOCSY and NOESY spectra for the peptide in water displayed peaks with little dispersion in the HN or Hα region, which is indicative of a random coil conformation for the peptide.
814
Figure 3. Circular dichroism (CD) spectra of the BR-II (A) in buffer, (B) in the presence of SDS, and (C) in the presence of DPC micelles.
wileyonlinelibrary.com/journal/jpepsci
Figure 4. Fingerprint region of the NOESY spectrum for 2 mM BR-II in 9:1 H2O/D2O and (A) 100 mM SDS-d25 and (B) 100 mM DPC-d38 at pH 4.3 and 298 K with a mixing time of 150 ms. All of the amide resonances are labelled. (C) Hα chemical shift index (CSI) for 2 mM BR-II in SDS (blue) and DPC (red) micelles at pH 4.3 and 298 K (peptide/lipid 1:50 mol/mol). The CSI was calculated by subtracting the values measured for the peptide from the random coil shifts reported in the literature [41].
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2014; 20: 811–821
MEMBRANE INTERACTION AND DNA BINDING OF ANTIMICROBIAL BREVININ-2-RELATED PEPTIDE In the presence of DPC micelles, NOE contacts between Met9 (Hα)–Ala14 (HN), Thr5 (QG)–Asn20 (HN), Thr5 (HG2)–Leu21 (HN), Thr5 (Hα)–Gln19 (HN), Thr5 (Hα)–Asn20 (HN), and Thr5 (Hα)– Leu21 (HN) are observed in the NOESY spectra of BR-II, which played a key role in determining the structure of the peptide (Figures 4B and S3). The secondary structure assignment is supported by the CSI values (Figure 4C), the difference between the Hα proton chemical shifts of BR-II in micelles environment, and the chemical shifts in a random coil. An uninterrupted segment of four or more residues experiencing an upfield chemical shift of more than 0.1 ppm indicates an α-helical structure of BR-II in the micelles environment. The CSI values mostly support the 3D structures of BR-II bound to SDS and DPC micelles calculated on the basis of NOE-derived distance constraints from 2D NOESY spectra. There is a turn region formed in between residues Gly10 and Val12 in SDS bound structure of BR-II as evident form 3D structure (Figure 6), although the CSI of these residues show α-helical values (Figure 4C). The difference in secondary structure assignment between CSI method and from 3D structure is possibly due to the oversimplified protocol used by the CSI method, which assigns the secondary structure type to each residue by simply comparing the observed chemical shift with the reference random coil chemical shift values [42]. It is also evident from the literature that the choice of reference random coil chemical shift values can significantly alter the outcome of secondary structure estimation from CSI [43]. Figure 5A and B shows a summary of the backbone NOEs for the secondary structure assignment. Characteristic (i,i + 2; i,i + 3; i,i + 4) medium range NOE correlations diagnostic of α-helical conformation were detected throughout the entire peptide backbone. The structural statistics based on the 20 lowest energy structures for BR-II in SDS and DPC are shown in Table 1. The calculated structures are in good agreement with experimental restraints as shown by structural statistics. Neither distance violations over 0.25 Å nor dihedral angel violations over 2° were found for the structures of the peptide in various media. The backbone dihedral angles (ϕ,ψ) of all residues are well defined as accessed by the PROCHECK-NMR program [34]. The Ramachandran plot analysis using PROCHECK-NMR showed that 94.1% and 94.3 % of the residues were favoured or allowed regions in the presence of SDS and DPC, respectively, and no residues of BR-II were found in the disallowed region in the presence of both the micelles (Table 1). The root-mean-square deviation (RMSD) values were calculated for BR-II using MolMol program [33]. The backbone and heavy atom RMSD for the structurally defined region of BR-II bound to SDS micelles is 0.11 and 0.59 Å, respectively, whereas for BR-II bound to DPC, the values are 0.07 and 0.60 Å, respectively (Table 1).
J. Pept. Sci. 2014; 20: 811–821
Table 1. NMR structural statistics for the 20 lowest energy structures of BR-II bound to SDS and DPC micelles SDS NOE distance constraints Intra-residue (i j) = 0 Sequential (|i j| = 1) Medium range (2 ≤ |i j| ≥ 4) Long range Angle constraints Φ Ψ Distance restraints violations No. of violations (≥0.25 Å) Distance restraints violations No. of violations (≥2 Å) Deviation from mean structure Backbone RMSD (N, Cα, C′) (Å) Heavy atoms (Å) Ramachandran plot analysis (%) Most favoured and allowed region Generously allowed region Disallowed region
DPC
325 85 121 116 3 32 16 16
312 74 126 105 7 32 16 16
0
0
0
0
0.11 ± 0.07 0.59± 0.13 94.1 5.9 0
0.07 ± 0.05 0.60 ± 0.11 94.3 5.7 0
bonds (
Revised: 23 May 2014
Accepted: 13 June 2014
Published online in Wiley Online Library: 17 July 2014
(wileyonlinelibrary.com) DOI 10.1002/psc.2673
Micelle bound structure and DNA interaction of brevinin-2-related peptide, an antimicrobial peptide derived from frog skin Susmita Bandyopadhyay,a Boon Yee Ng,a Charmaine Chong,a Ming Zhen Lim,a Sonia Kiran Gill,a Ke Hui Lee,a J Sivaramanb and Chiradip Chatterjeea* Brevinin-2-related peptide (BR-II), a novel antimicrobial peptide isolated from the skin of frog, Rana septentrionalis, shows a broad spectrum of antimicrobial activity with low haemolytic activity. It has also been shown to have antiviral activity, specifically to protect cells from infection by HIV-1. To understand the active conformation of the BR-II peptide in membranes, we have investigated the interaction of BR-II with the prokaryotic and eukaryotic membrane-mimetic micelles such as sodium dodecylsulfate (SDS) and dodecylphosphocholine (DPC), respectively. The interactions were studied using fluorescence and circular dichroism (CD) spectroscopy. Fluorescence experiments revealed that the N-terminus tryptophan residue of BR-II interacts with the hydrophobic core of the membrane mimicking micelles. The CD results suggest that interactions with membrane-mimetic micelles induce an α-helix conformation in BR-II. We have also determined the solution structures of BR-II in DPC and SDS micelles using NMR spectroscopy. The structural comparison of BR-II in the presence of SDS and DPC micelles showed significant conformational changes in the residues connecting the N-terminus and C-terminus helices. The ability of BR-II to bind DNA was elucidated by agarose gel retardation and fluorescence experiments. The structural differences of BR-II in zwitterionic versus anionic membrane mimics and the DNA binding ability of BR-II collectively contribute to the general understanding of the pharmacological specificity of this peptide towards prokaryotic and eukaryotic membranes and provide insights into its overall antimicrobial mechanism. Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd. Additional supporting information may be found in the online version of this article at the publisher’s web site. Keywords: antimicrobial peptide; brevinin-2-related peptide; sodium dodecylsulfate; n-dodecylphosphocholine; two-dimensional NMR; pUC19 DNA
Introduction
J. Pept. Sci. 2014; 20: 811–821
* Correspondence to: Chiradip Chatterjee, School of Applied Science, Republic Polytechnic, Singapore. E-mail: [email protected] a School of Applied Science, Republic Polytechnic, Singapore b Department of Biological Sciences, National University of Singapore, Singapore
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
811
The development of microbial resistance against antibiotics has become a serious problem because of the inappropriate or the constant use of antibiotics and the mutation of the microbes [1]. The search for new antibiotics to treat infections caused by multidrug-resistant microorganisms has thus become urgent and has stimulated interest in the development of antimicrobial peptides (AMPs) as human therapeutics [2,3]. AMPs are present in all kingdoms of life as a first line of defense against deadly pathogens. They demonstrate high potential as novel therapeutic agents as most of the AMPs eradicate both Gram-positive and Gram-negative bacteria [4–6]. In addition, several AMPs have been found to have anti-HIV properties, which have stimulated recent research in this area [7–12]. Amphibians that live in both land and water force themselves to adapt and survive in a variety of conditions. Amphibian skin secretes potent antimicrobial agents to protect them against infections by molds, bacteria, and protozoa and has proven to be valuable source of AMPs [13–16]. A 21-amino-acid peptide, named Brevinin-2related peptide (BR-II), isolated from the skin secretions of the mink frog (Rana septentrionalis) shows broad-spectrum antimicrobial
activity and low hemolytic activity [17]. Recently, it has also been reported that BR-II can protect cells from HIV-1 infection [18]. This peptide shows some sequence similarity to the N-terminus region of other peptides of the brevinin-2 family isolated from several species of Eurasian frog skin [17]. Despite lacking a C-terminus cyclic heptapeptide region (Cys-Lys-Xaa4-Cys-, a common feature of all antimicrobial brevinin-2 peptides known to date), BR-II shows relatively high potency against Gram-positive and Gram-negative bacteria as well as opportunistic yeast pathogens. This demonstrates that the cyclic heptapeptide region in brevinin-2 is not necessary for antimicrobial activity [17]. Thus, BR-II represents a novel therapeutic agent, and understanding its mode of antimicrobial and anti-HIV activities is essential to the development of the compound for medical use.
BANDYOPADHYAY ET AL. The majority of AMPs are believed to act by disrupting bacterial cell membranes; therefore, an important aspect of determining the mode of action employed by an AMP is to study its ability to interact with membranes and the structural folding of AMP in membranes [19–21]. Because of the complexity of the bacterial membranes, such peptide structures were mainly determined by solution NMR using membrane-mimetic models such as organic solvents or detergent micelles. Both deuterated sodium dodecylsulfate (SDS) and n-dodecylphosphocholine (DPC) micelles are widely employed in determining the solution structure of AMP bound to lipid micelles [6,4,5,22]. This can provide valuable insight on a number of physicochemical properties of AMP such as the distribution of the charges on the surface of the peptide, amphipathicity and structural folding, which contribute to the mechanism by which an AMP executes its function. In this paper, we report the interaction of BR-II with SDS and DPC studied by fluorescence and circular dichroism (CD) spectroscopy. In addition, the three-dimensional (3D) solution structures of BR-II bound to DPC and SDS micelles have been determined by NMR spectroscopy for the first time. The comparison between the structures of BR-II in SDS and DPC micelles will lead to a better understanding of the different bioactive conformations of BR-II peptide in negative and zwitterionic membrane environments, respectively. The studies on the interactions between AMPs and DNA are of interest to determine if the interaction of peptide with intercellular target such as DNA takes place after penetration of the peptide to the inner structure of the cell by peptide– membrane interaction and thus, plays a part in overall antimicrobial activity of the peptide [23]. Here, we also report the DNA binding ability of BR-II using agarose gel retardation assay and fluorescence experiments.
Materials and Methods Materials Brevinin-2 peptide (98% pure, GIWDTIKSMGKVFAGKTLQNL) was purchased from Nanyang Technological University Biosciences Laboratory (Singapore) and used without further purification. SDS and DPC were obtained from Avanti Polar Lipids (USA) in the highest available purity. Deuterated SDS-d25 and DPC-d38 were purchased from Cambridge Isotope Laboratories (CIL Inc, USA). pUC19 DNA was purchased from Invitrogen (Singapore) and used without further purification. Analytical grade acrylamide and ethidium bromide were obtained from Sigma Aldrich (USA). All other reagents and solvents were of reagent grade and used without further purification unless otherwise specified. Fluorescence Spectroscopy
812
Intrinsic fluorescence measurements were recorded using Perkin Elmer LS50B fluorescence spectrophotometer. Peptide fluorescence was measured by exciting peptide samples at 280 nm and scanning emission between 300 and 460 nm using a slit of 3 nm. A peptide concentration of 2 μM was used for all the fluorescence experiments. After addition of the sample, the system was equilibrated for 5 min before fluorescence was measured. All measurements were made in a semi-micro quartz cuvette (path length 1 cm) at room temperature (~25 °C). All samples were prepared in Millipore water (pH 4.3, adjusted with HCl) for recording the fluorescence spectra in SDS (50 mM) and DPC
wileyonlinelibrary.com/journal/jpepsci
(50 mM) membrane-mimetic micellar solutions, with the baseline corrected using the corresponding control solutions. The changes in intensity and blue shifts in the emission spectra of BR-II in the presence of different model membranes were monitored. Acrylamide quenching was performed on free and micelles bound peptide mixtures to calculate Stern–Volmer constants (KSV) following published protocols [24–26]. The KSV for each combination of the peptide and micelles was calculated using the following equation: F 0 =F c ¼ 1 þ K SV ½Q where F0 is the initial fluorescence of the peptide and Fc is the fluorescence intensity following the addition of soluble quencher, Q. Peptide-DNA fluorescence was measured following the same protocol described earlier. A fixed amount (8.8 μgm) of peptide was treated with increasing amounts of pUC19 DNA (DNA amount: 0, 0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75 μgm). Fluorescence spectra were measured from 320 to 500 nm. After addition of each sample, the system was equilibrated for 5 min before fluorescence was measured. All measurements were made in a semi-micro quartz cuvette at room temperature (~25 °C). All samples were prepared in Millipore water (pH 4.3, adjusted with HCl) for recording the fluorescence spectra with the baseline corrected using the corresponding control solutions. All fluorescence experiments were carried out in triplicate to check for the reproducibility, and they were found consistent. CD Spectroscopy The secondary structures of BR-II in different micelle solutions were recorded in a Jasco J810 spectropolarimeter (Jasco Corp., Tokyo, Japan) at room temperature (~25 °C) using a 0.1 cm path length cuvette and with the baseline corrected using control solutions. All samples were prepared in millipore water (pH 4.3, adjusted with HCl) for recording the CD spectra in SDS (50 mM) and DPC (50 mM) micelles. The peptide concentration was maintained at 20 μM in all cases. NMR Spectroscopy NMR measurements were performed on a Bruker Avance DRX-800 operating at 800 MHz with cryoprobe and Bruker Avance DRX-400 operating at 400 MHz. The sample temperature was maintained at 298 K. The samples were prepared with 2 mM peptide, 100 mM SDS-d25 and another sample with 100 mM DPC-d38 (CIL, Cambridge, MA, USA) and 10% D2O in millipore water (pH 4.3 adjusted using diluted HCL). Two-dimensional (2D) TOCSY spectra (spinlock time of 100 ms) were acquired using the MLEV-17 pulse sequence [27]. 1 H 2D NOESY spectra were acquired using a 150 ms mixing time [28]. Water suppression was achieved using the WATERGATE technique [29]. The NMR spectra were collected with 4096 and 512 data points in the direct and indirect dimensions, respectively, with 16 transients. The raw data were processed using Topspin 2.0 (Bruker Inc), and resonance assignments were made using the Sparky [30]. Structure Calculation The 3D structures of BR-II bound to SDS and DPC micelles were calculated on the basis NOE-derived distance constraints from 2D NOESY spectra. Assignment of the proton spectra of BR-II in the presence of membrane-mimetic solvents (SDS and DPC)
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2014; 20: 811–821
MEMBRANE INTERACTION AND DNA BINDING OF ANTIMICROBIAL BREVININ-2-RELATED PEPTIDE was accomplished using the technique of sequence-specific resonance assignments developed by Wuthrich [31]. On the basis of cross-peak intensities, NOEs were qualitatively categorized to strong, medium, and weak and then translated to upper bound distance limits to 2.5, 3.5, and 5.5 Å, respectively. The lower bounds for all NOE restraints were set to 2.0 Å. Initial structures were generated by backbone torsion angle restraints and distance geometry using CYANA-2.1 [31,32]. For structure calculations, the dihedral angle restraints were defined for all the backbone angles of the peptide, except for Gly. The dihedral angle restraints were referenced on the basis of CD structure, chemical shift index (CSI) values, and NOE patterns. A total of 100 structures were calculated, and 20 lowest-energy structures with no distance violations greater than 0.25 Å or dihedral angle violations larger than 2° were selected. All the structures were analysed using Pymol (http://pymol.sourceforge.net) and MolMol [33] and PROCHECK-NMR program [34]. Translational Diffusion Experiments Diffusion-ordered spectroscopy (DOSY) experiments were carried out using a 400 MHz Bruker Avance DRX-400 spectrometer with a quad (1H, 13C, 19F, and 31P) probe and z-gradient. The gradient coil constant was found to be 50 G/cm at 100% gradient strength by carrying out the BPLED pulse sequence with water suppression using pre-saturation. The diffusion time Δ was 120 ms, and the gradient duration δ was 2 ms. The gradient strength was increased from 2% to 95% of Gmax in 32 scans. The individual diffusion coefficients were calculated from the DOSY spectra by decay of the peak intensities as a function of gradient strength and evaluated using the Bruker Topspin 2.0 T1/T2 analysis package Topspin 2.0 package (Bruker Inc) using aliphatic Leu-Hδ1/ Hδ2 chemical shifts (about 0.78 ppm) for the peptide. Diffusion constants were derived from the following equation [35]: I ¼ I0 eDγ
g δ ðΔ3δÞ
2 2 2
(1)
where I is the observed intensity, I0 the unattenuated signal intensity, D the diffusion coefficient, γ the gyromagnetic ratio of 1H, g the gradient strength, δ the gradient length, and Δ is the diffusion time. The molar fraction of peptide in micelles (FL) is calculated with Eqn (2) [36–38] DS ¼ DL F L þ DF ð1 F L Þ
Results Interaction of BR-II with Micelles Tryptophan fluorescence The change in fluorescence intensity and the shift in maximum emission wavelength of the tryptophan (Trp) fluorophore of BR-II in the presence of SDS and DPC micelles was examined (Figure 1). The maximum emission wavelength of Trp fluorophore of BR-II in aqueous solution was 363 nm. A maximum blue shift of approximately 5 and 22 nm was observed in the presence of SDS and DPC micelles, respectively (Figure 1). In addition, a change in the intensity of the florescence indicated that the environment of the fluorophore in the presence of micelles had changed compared with the peptide in buffer alone (Figure 1 and its inset). This suggests that the Trp3, which is located at the N-terminus of the BR-II, interacts with the micelles and forms hydrophobic clusters similar to other AMPs [40]. Fluorescence quenching The Stern–Volmer constants (KSV), which determines the extent to which the Trp3 residue of the peptide is buried into the hydrophobic core of the micelles, was calculated from the acrylamide quenching experiment. When the side chain of Trp3 is buried into the micelle phase and becomes less exposed to the surrounding solvent, the fluorescence contribution of Trp3 is less susceptible to the effects of the neutral soluble quencher acrylamide. The lower the KSV value, the greater is the degree of interaction of the Trp fluorophore with the hydrophobic core of the micelles. The highest optimum concentration of lipid was determined by tryptophan fluorescence (Figure 1), in which no further blue shift in fluorescence emission spectra was observed. This optimum concentration of lipid was used for all fluorescence quenching experiments to ensure the ideal complex formation between the peptide and the micelles [25]. The KSV for BR-II in aqueous solution was found to be ~105 M1. The KSV values decreased significantly in the presence of SDS and DPC micelles (Figure 2) indicating that the Trp side chain interacts with the micelles in these two membrane-mimetic environments. This supports the blue shifts observed for the BR-II peptide in the presence of these micelles environment (Figure 1).
(2)
where DS is the measured diffusion coefficients of the peptide in the presence of micelles and DL and DF are the diffusion coefficients of micelles and free peptides in aqueous solution, respectively. The experimental values of diffusion coefficient of micelles (DL) for SDS [21] and DPC are obtained from the literature [39]. DNA Gel Retardation Assay
J. Pept. Sci. 2014; 20: 811–821
Figure 1. Tryptophan (Trp) fluorescence emission spectra of BR-II (2 μM) (A) in buffer, (B) in 50 mM SDS, and (C) in 50 mM DPC. Inset: Bar diagram showing maximum blue shift of fluorophore in presence of different micellar environments (SDS, DPC) compared with BR-II in buffer.
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci
813
The pUC19 DNA (Invitrogen) was mixed with increasing amounts of peptides in 15 μl of buffer (10 mM Tris (pH 7.6)/50 mM NaCl/ 1 mM EDTA). The mixtures were incubated at room temperature for 5 min and then subjected to electrophoresis on a 1% agarose gel in the 1× TAE (40 mM Tris, 20 mM acetic acid, and 1 mM EDTA) buffer solution.
BANDYOPADHYAY ET AL.
Figure 2. Stern–Volmer constants calculated from acrylamide quenching experiments for free BR-II in buffer and in the presence of various micellar environments of SDS and DPC, respectively.
The complete chemical shift assignments of BR-II in the presence of SDS and DPC micelles at 298 K is given in Table 1 in the Supporting information. We assigned ~100% NOE peaks in the 2D NOESY spectrum. The representative NOE cross-peak pattern between the backbone amides and R-protons for BR-II in SDS and DPC is presented in Figure 4A and B, respectively. Nearly all proton resonances including side chains were unambiguously assigned. An ensemble of 100 structures was calculated using 325 (85 intra-residue and 240 inter-residue) NOEs for SDS micelles and 312 (74 intra-residue and 238 inter-residue) for DPC micelles from the analysis of the 2D 1H–1H NOESY spectrum using Cyana-2.1. The best 20 lowest energy structures were retained for analysis. Several long distance NOEs are observed for BR-II in SDS micelles including NOEs between Ser8 (Hα)–Phe13 (HN), Asn20 (Hα)–Ile2 (HN), and Thr5 (Hα)–Asn20 (HN) (Figures 4A and S2).
CD spectroscopy CD experiments were performed to study the effect of different membrane environments on the secondary structure of BR-II. In millipore water (pH 4.3, adjusted with HCl), BR-II exhibits spectral characteristics of an unstructured peptide with a broad minimum at 198 nm (Figure 3). Double minimum bands at 225 and 208 nm and a positive CD band at 195 nm in the spectra (Figure 3 and Figure S1 in the Supporting information) were obtained for BR-II in the presence of SDS and DPC micelles, indicating the presence of a predominantly α-helical structure, compared with the random coil structure observed for BR-II in buffer solution. These data clearly indicate that BR-II interacts with all micelles and undergoes significant conformational changes from a random coil state to α-helical state. NMR spectroscopy Two-dimensional NMR spectra of BR-II were obtained at 298 K and pH 4.3 in aqueous solution and in the presence of SDS and DPC micelles. The proton resonance assignments were made using the procedures of Wuthrich [31]. To assign the backbone, side chain resonances and sequential assignments, a combination of 2D [1H, 1H] TOCSY and 2D NOESY spectra, were used. The 1D proton and 2D TOCSY and NOESY spectra for the peptide in water displayed peaks with little dispersion in the HN or Hα region, which is indicative of a random coil conformation for the peptide.
814
Figure 3. Circular dichroism (CD) spectra of the BR-II (A) in buffer, (B) in the presence of SDS, and (C) in the presence of DPC micelles.
wileyonlinelibrary.com/journal/jpepsci
Figure 4. Fingerprint region of the NOESY spectrum for 2 mM BR-II in 9:1 H2O/D2O and (A) 100 mM SDS-d25 and (B) 100 mM DPC-d38 at pH 4.3 and 298 K with a mixing time of 150 ms. All of the amide resonances are labelled. (C) Hα chemical shift index (CSI) for 2 mM BR-II in SDS (blue) and DPC (red) micelles at pH 4.3 and 298 K (peptide/lipid 1:50 mol/mol). The CSI was calculated by subtracting the values measured for the peptide from the random coil shifts reported in the literature [41].
Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2014; 20: 811–821
MEMBRANE INTERACTION AND DNA BINDING OF ANTIMICROBIAL BREVININ-2-RELATED PEPTIDE In the presence of DPC micelles, NOE contacts between Met9 (Hα)–Ala14 (HN), Thr5 (QG)–Asn20 (HN), Thr5 (HG2)–Leu21 (HN), Thr5 (Hα)–Gln19 (HN), Thr5 (Hα)–Asn20 (HN), and Thr5 (Hα)– Leu21 (HN) are observed in the NOESY spectra of BR-II, which played a key role in determining the structure of the peptide (Figures 4B and S3). The secondary structure assignment is supported by the CSI values (Figure 4C), the difference between the Hα proton chemical shifts of BR-II in micelles environment, and the chemical shifts in a random coil. An uninterrupted segment of four or more residues experiencing an upfield chemical shift of more than 0.1 ppm indicates an α-helical structure of BR-II in the micelles environment. The CSI values mostly support the 3D structures of BR-II bound to SDS and DPC micelles calculated on the basis of NOE-derived distance constraints from 2D NOESY spectra. There is a turn region formed in between residues Gly10 and Val12 in SDS bound structure of BR-II as evident form 3D structure (Figure 6), although the CSI of these residues show α-helical values (Figure 4C). The difference in secondary structure assignment between CSI method and from 3D structure is possibly due to the oversimplified protocol used by the CSI method, which assigns the secondary structure type to each residue by simply comparing the observed chemical shift with the reference random coil chemical shift values [42]. It is also evident from the literature that the choice of reference random coil chemical shift values can significantly alter the outcome of secondary structure estimation from CSI [43]. Figure 5A and B shows a summary of the backbone NOEs for the secondary structure assignment. Characteristic (i,i + 2; i,i + 3; i,i + 4) medium range NOE correlations diagnostic of α-helical conformation were detected throughout the entire peptide backbone. The structural statistics based on the 20 lowest energy structures for BR-II in SDS and DPC are shown in Table 1. The calculated structures are in good agreement with experimental restraints as shown by structural statistics. Neither distance violations over 0.25 Å nor dihedral angel violations over 2° were found for the structures of the peptide in various media. The backbone dihedral angles (ϕ,ψ) of all residues are well defined as accessed by the PROCHECK-NMR program [34]. The Ramachandran plot analysis using PROCHECK-NMR showed that 94.1% and 94.3 % of the residues were favoured or allowed regions in the presence of SDS and DPC, respectively, and no residues of BR-II were found in the disallowed region in the presence of both the micelles (Table 1). The root-mean-square deviation (RMSD) values were calculated for BR-II using MolMol program [33]. The backbone and heavy atom RMSD for the structurally defined region of BR-II bound to SDS micelles is 0.11 and 0.59 Å, respectively, whereas for BR-II bound to DPC, the values are 0.07 and 0.60 Å, respectively (Table 1).
J. Pept. Sci. 2014; 20: 811–821
Table 1. NMR structural statistics for the 20 lowest energy structures of BR-II bound to SDS and DPC micelles SDS NOE distance constraints Intra-residue (i j) = 0 Sequential (|i j| = 1) Medium range (2 ≤ |i j| ≥ 4) Long range Angle constraints Φ Ψ Distance restraints violations No. of violations (≥0.25 Å) Distance restraints violations No. of violations (≥2 Å) Deviation from mean structure Backbone RMSD (N, Cα, C′) (Å) Heavy atoms (Å) Ramachandran plot analysis (%) Most favoured and allowed region Generously allowed region Disallowed region
DPC
325 85 121 116 3 32 16 16
312 74 126 105 7 32 16 16
0
0
0
0
0.11 ± 0.07 0.59± 0.13 94.1 5.9 0
0.07 ± 0.05 0.60 ± 0.11 94.3 5.7 0
bonds (
